category stringclasses 1 value | source stringclasses 1 value | url stringlengths 74 77 | text stringlengths 16 2.15M |
|---|---|---|---|
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12782348&blobtype=pdf | # Exploring the Antiviral Potential of Tungsten Oxide Nanoparticles Against Herpes Simplex Virus Type 1: A Promising Alternative to Acyclovir
Abdulhussain Jwaziri, Pegah Khales, Seyed Kiani, Homayoun Yaghouti, Roghayeh Babaei, Zahra Salavatiha, Ahmad Tavakoli
## Abstract
Herpes simplex virus type 1 (HSV-1) is responsible for the majority of cold sores, herpetic keratitis-induced blindness, profound skin lesions, and encephalitis that can be fatal. Currently, acyclovir and its derivatives are the first-line therapy for the treatment of HSV-1 infection. But there are drawbacks to these treatments: limited efficacy against drug-resistant strains of the virus. Hence, it is of critical importance to explore and develop new antiviral drugs for HSV-1. In the present study, we explored whether tungsten oxide nanoparticles (WO 3 NPs) were potent inhibitors of HSV-1 infection as a new class of agent. WO 3 NPs were characterized by X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM), Fourier transform infrared (FTIR) spectroscopy, and zeta potential analysis. Cytotoxicity of Vero cells caused by WO 3 NPs was determined by methyl thiazolyl tetrazolium (MTT) assay. The quantitative real-time polymerase chain reaction (qRT-PCR) assay was utilized for further verification of the action of the WO 3 NPs on HSV-1. The cytotoxicity test showed low toxicity (<20%) of the rod-shaped WO 3 NPs when they were assayed on Vero cells at concentrations of up to 700 μg/mL. When HSV-1 was treated with WO 3 NPs at 700 µg/mL [20% cytotoxicity concentration (CC 20 ); the concentration causing 20% cytotoxicity, ~80% cell viability] and 1000 µg/mL [50% cytotoxicity concentration (CC 50 ); the concentration causing 50% cytotoxicity, ~50% cell viability] for 3 h, the viral load was significantly reduced, achieving inhibition rates of 99.4% and 99.9%, respectively. Additionally, experiments conducted after HSV-1 infection of Vero cells (post-treatment assays) indicated that WO 3 NPs at concentrations of 250, 500, and 750 µg/mL significantly suppressed viral replication, with inhibition rates of 82%, 87.5%, and 96.5%, respectively. WO 3 NPs have potent inhibitory effects on HSV-1. Therefore, they can be considered potential candidates for therapeutic development against infections caused by this virus.
## 1. Introduction
Herpes simplex virus type 1 (HSV-1) is a member of the genus Simplexvirus, the family Herpesviridae, the order Herpesvirales, the subfamily Alphaherpesvirinae, and the species Human alphaherpesvirus 1 [1]. It is estimated that around 3.8 billion people worldwide, aged under 50 years, representing 64% of this demographic, are affected by HSV-1 infection [2]. HSV-1 primarily induces orolabial lesions, although it can also result in genital herpes. Also, HSV-1 can cause serious diseases like herpetic keratitis or encephalitis [3].
General advances were also achieved in terms of explaining the pathophysiological characteristics and replication methods of HSV-1, consequently contributing to the production of antiviral medications specifically created for virus elimination [4]. Currently, the drugs of choice for the management of HSV-1 infections continue to remain acyclovir, valacyclovir, and famciclovir, but other drugs such as trifluridine (exclusively for ophthalmic use), ganciclovir, foscarnet, and cidofovir can also be utilized, with vidarabine being mostly obsolete. All of these drugs, except foscarnet, are incorporated into the developing viral DNA chain and inhibit viral replication as nucleoside analogs. Foscarnet is a pyrophosphate that inhibits viral DNA polymerase [5].
Based on the chronic, recurrent, and untreatable character of HSV-1 infection, excessive administration of nucleoside analogs, including acyclovir, may induce dramatic side effects, such as neurotoxicity, renal insufficiency, and drug resistance. The development of drug-resistant HSV-1 mutants is correlated with reduced drug responses and limited therapeutic options, with devastating effects including long-standing orolabial lesions, herpetic keratitis, or encephalitis [6]. This highlights the necessity of developing new anti-HSV-1 drugs that can overcome these hurdles.
Nanoparticles (NPs) find widespread applications in industrial applications as well as in medicinal applications. Of all the metal NPs, more emphasis is placed on tungsten oxideNPs (WO 3 NPs) because of their unique properties, that is, large surface area and high-temperature stability [7]. Tungsten (VI) oxide (WO 3 ) is a chemical substance of the transition element tungsten with oxygen. It is also prepared as an intermediate when tungsten is prepared from the ores of tungsten [8]. Additionally, bioproducts also make use of WO 3 NPs in the form of pigments, additives, as well as analytical agents [9].
So far, just a few studies have looked at how WO 3 NPs work against certain bacteria, and no studies have looked at how these NPs affect human viruses. Because HSV-1 is so important and treating infections caused by this virus is so hard, this study aims to see if WO 3 NPs can work as anti-HSV-1 medications instead of the ones that are now available.
## 2. Materials and Methods
## 2.1. Characterization of WO 3 NPs. US Research Nanomaterials
Inc. provided high-purity WO 3 nanopowders (>99%) for this research study. Characterization methods were performed to characterize the NPs using a variety of advanced equipment. X-ray diffraction (XRD, Philips PW1730) was performed to determine and confirm the crystal structure. Functional groups were identified on the surface using Fourier transform infrared (FTIR) spectroscopy (Thermo Scientific Nicolet Avatar 380). Zeta potential analysis (Horiba SZ-100) measured the surface charge. The surface morphology and particle size of WO 3 NPs were examined by field-emission scanning electron microscopy (FE-SEM, TESCAN MIRA4) operated at an accelerating voltage of 15 kV after sputter coating with a conductive layer.
2.2. Cell and Virus. Vero cells, originating from the kidneys of African green monkeys, were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco, Invitrogen, USA), containing high glucose content, and supplemented with 10% heatinactivated fetal bovine serum (FBS) (Gibco, Invitrogen, USA), 2 mM L-glutamine (Merck, Germany) and an antibiotic mixture (100 μg/mL streptomycin and 100 U/mL penicillin; Sigma-Aldrich, USA). The cells were incubated at 37°C in a 5% CO 2 environment. To determine the antiviral activity, the viral strain HSV-1 KOS was expanded in Vero cells, and the viral titer of the stock was determined using the Reed and Muench method, reported as TCID 50 /mL. The viral stock was prepared and divided into sterile microtubes for appropriate storage at -80°C for future use [10].
## 2.3. Assessment of Cell Cytotoxicity.
The cytotoxic effects of WO 3 NPs on Vero cells were assessed via the methyl thiazolyl tetrazolium (MTT) assay. First, Vero cells were plated at a density of 1 × 10 4 cells per well in a 96-well flat-bottom microtiter plate (SPL Life Science, South Korea) and incubated for 24 h at 37°C. Next, a variety of concentrations of WO 3 NPs (100-1000 μg/mL) were placed in wells in triplicate. The plate was subsequently incubated for 48 h at 37°C. After the incubation period, 10 μL of MTT reagent (5 mg/mL) was added to each well and placed in the dark, 37°C, for 3 h. The MTT solution was then removed carefully, and 50 μL of pure dimethyl sulfoxide (DMSO) (Bio-Idea, Iran) was added to dissolve the formazan crystals. The plate was gently agitated for 10 min at room temperature. Finally, absorbance was measured at 550 nm using a microplate reader (Hiperion MPR 4+, Roedermark, Germany). Cell viability percentages were calculated by comparing the treated wells to the untreated control group [10].
## 2.4. Determination of Antiviral Activity
2.4.1. Virucidal Assay. To summarize, 100 μL of WO 3 NPs at CC 50 (the concentration causing 50% cytotoxicity, ~50% cell viability) and at CC 20 (the concentration causing 20% cytotoxicity, ~80% cell viability) were combined with 100 μL of a HSV-1 suspension (100 TCID 50 /mL) and incubated for either one or 3 h at 37°C in a humidified atmosphere of 5% CO 2 . A virus control was made with a solution of the virus incubated with a cell culture medium. Mixtures were then applied to Vero cell monolayers and incubated for an hour at 37°C. After incubation, the supernatant was collected, and the cells were washed twice with phosphate-buffered saline (PBS) to remove any unbound viruses. After washing, the cells had fresh DMEM medium (with 2% FBS) added and were incubated at 37°C for 48 h [10].
2.4.2. Cell Post-Treatment Assay. Monolayers of Vero cells were cultured and subsequently infected with HSV-1 solution (100 TCID 50 /mL) at 37°C for 1 h in a 96-well microtiter plate. Following the HSV-1 infection, the cells were then washed with PBS in order to eliminate all noninternalized viruses. The infected Vero cells were then treated with factorial concentrations of WO 3 NPs that were considered nontoxic and cultured for 48 h at 37°C and 5% CO 2 . The experiment was further extended to cell controls and virus controls in the same medium, and the viral load of HSV-1 was determined by applying the quantitative real-time polymerase chain reaction (qRT-PCR) technique [10]. The whole assays were carried out in darkness in order to rule out the contribution of the photoactivity of the WO 3 NPs.
2.5. qRT-PCR. The impact of WO 3 NPs on HSV-1 infection on Vero cells was analyzed via qRT-PCR. Using the BehPrep Viral Nucleic Acid Extraction Kit (BehGene Biotechnology, Iran), viral DNA was extracted from the supernatants of infected Vero cells (gathered during virucidal and post-treatment experiments) in accordance with the manufacturer's instructions. Using the following primers and probe, the qRT-PCR produced a 75 bp amplicon that targeted the HSV-1 UL30 gene: forward (5′-ATCGGCGAGTACTGCATACA-3′), reverse (5′-GAGCTCCAGATGGGGCAA-3′), and probe (5′-HEX-ATTCCCTGCTGGTGGGCCA-BHQ1-3′). Each 25 μL reaction mixture contained 12.5 μL of RealQ Plus 2x Master Mix for Probe (Ampliqon, Denmark), 2 μL of forward primer (10 µM), 1 μL of reverse primer (10 µM), 1 μL of probe (10 µM), 5 μL of template DNA, and 3.5 μL of ddH 2 O. A Rotor-Gene Q system (Qiagen, Germany) was used for the amplification process. There were 40 cycles of 95°C for 10 s and 60°C for 30 s after initial denaturation at 95°C for 10 min [10].
The target viral sequence was cloned into the pUC57 vector to generate a reference standard. To create a standard curve, serial tenfold dilutions of the synthesized plasmid, ranging from 10 -1 to 10 -10 , were made and examined with the test samples. An online DNA copy number calculator (Technology Networks) was used to determine the plasmid copy number. Three copies/μL was the detection limit (LOD) of the qRT-PCR experiment. Positive and negative controls were included in every PCR run, and all reactions were carried out in triplicate to guarantee reproducibility [10].
2.6. Statistical Analysis. Statistical analyses were performed in SPSS software Version 22.0 (IBM, Chicago, IL, USA). The normality of the data was assessed using the Kolmogorov-Smirnov test. The independent t-test was used to assess meaningful differences among quantitative data that followed a normal distribution, and the Mann-Whitney U test was used for data that did not follow a normal distribution. Homogeneity of variances was assessed using Levene's test. A significance threshold of 0.05 was determined to signify statistical significance, where a p-value less than or equal to this level was considered significant. All graphical representation of data was conducted using GraphPad Prism 8.
## 3. Results
## 3.1. Characterization of WO 3 NPs
3.1.1. XRD Analysis. The XRD analysis of WO 3 NPs was performed using Cu-Kα radiation (λ = 1.5406 Å) at 2θ values. The diffraction patterns and related planes were interpreted using the Profex 5.4.1 software, which revealed distinct and intense diffraction peaks at 23.15°, 23.67°, and 24.36°, corresponding to the Miller indices (002), (020), and (200), respectively. These peaks confirm the monoclinic phase of the crystal structure of WO 3 NPs. As shown in Figure 1, additional diffraction peaks further validate the monoclinic structure of the WO 3 NPs. This is aligned with the Joint Committee on Powder Diffraction Standards (JCPDS No. 43-1035), confirming that the synthesized material exists in a single phase [11][12][13][14]. The absence of impurity-related peaks indicates the high purity of the synthesized NP. The results also show that the (200) plane has significantly higher intensity compared to other planes, which may suggest a possible preferential orientation along this direction; however, further texture analysis would be required to confirm such growth behavior. The average crystallite size according to the Scherrer equation (D = 0.9λ/B cosθ) was estimated to be ~27.44 nm.
## 3.1.2. FTIR Analysis.
The FTIR spectrum of WO 3 is shown in Figure 2. The peaks at 3436.32 cm -1 and 1633.50 cm -1 correspond to O-H stretching vibrations and H-O-H bending vibrations, respectively, indicating the presence of water molecules adsorbed on the NP surface or hydroxyl groups. The weak peak at 1384.73 cm -1 can be attributed to impurities or organic residues, such as the symmetric bending vibration of CH 3 groups, which may be associated with stabilizing agents or precursors used during synthesis. The strong absorption bands observed at 815.84 cm -1 and 750.51 cm -1 are assigned to the bending vibrations of bridging W-O-W bonds, while the prominent band at 962.31 cm -1 corresponds to the stretching vibration of terminal W = O bonds. These assignments are consistent with previous reports on tungsten trioxide [15,16], which confirm that such vibrational features are characteristic of monoclinic WO 3 and reflect the complex bonding interactions between tungsten and oxygen atoms.
3.1.3. Zeta Potential. For a stable colloidal system to be considered, zeta potential values are usually greater than AE30 mV (whether positive or negative). This threshold ensures sufficient electrostatic repulsion between particles and effectively prevents aggregation [17]. In this study, the WO 3 NPs were dispersed in PBS at neutral pH (~7) with an ionic strength of ~1-10 mM, and measurements were performed at room temperature (~25°C). The zeta potential was measured three times, yielding a mean value of -43.7 AE 0.44 mV, indicating a stable colloidal suspension. The NPs carry a significant negative charge on their surface, creating a repulsive force that prevents their aggregation and contributes to the stability of the system (Figure 3). width of a total of 50 NPs were measured separately using FIJI (ImageJ) software. The average width and length of WO 3 NPs were 33.3 AE 5.0 nm and 169.8 AE 31.1 nm, respectively. The relatively larger standard deviation observed in the NP length compared to the width is attributed to the natural variation in rod elongation during the synthesis process, which is commonly observed in rod-like NP growth. Size-distribution histograms and Gaussian fits generated using GraphPad Prism 10 are presented in Figure 4, clearly illustrating the variation and distribution of NP dimensions. The uniformity in shape and size distribution suggests well-controlled synthesis conditions, leading to homogeneous structures. Slight aggregation was observed, likely due to the high surface activity and surface energy of the NPs, as commonly seen in nanorod systems.
The results are consistent with XRD analysis, which confirms the presence of characteristic peaks corresponding to the monoclinic crystalline phase of WO 3 . The crystallite size estimated from XRD (~27.44 nm) corresponds to the size of individual crystalline domains, while the larger dimensions observed in FE-SEM represent the entire rod-like NPs composed of multiple crystallites. This agreement between FE-SEM and XRD validates both the synthesis process and the structural characteristics of the NPs. FE-SEM images were acquired at an accelerating voltage of 15 kV, with a thin gold coating applied to the sample, at magnifications of 10, 50, and 150 Kx, ensuring clear visualization of NP morphology.
## 3.2. Cytotoxicity Assay.
The MTT assay results showed that cell viability decreased to 80% (CC 20 , the concentration causing 20% cytotoxicity) and 50% (CC 50 , the concentration causing 50% cytotoxicity) in comparison to control cells when WO 3 NP concentrations increased to 700 and 1000 μg/mL, respectively (Figure 5). Thus, at CC 20 (700 µg/mL) and CC 50 (1000 µg/mL) concentrations, we investigated the virucidal potency of WO 3 NPs against HSV-1. In the post-treatment assay, the ability of WO 3 NPs to inhibit HSV-1 replication in Vero cells was tested at nontoxic concentrations up to and including CC 20 (700 µg/mL). and 3 h led to a substantial decrease in virus-induced cytopathic effect (CPE). Furthermore, the CPE inhibition rates were almost the same at both 700 and 1000 µg/mL concentrations (Figure 6). The qRT-PCR assay results revealed that elevating the concentration from 700 to 1000 µg/mL caused a decrease in viral load at both 1-and 3-h incubation periods. By utilizing the qRT-PCR technique for viral load measurement and virus control comparison, the viral inhibitory rates of the WO 3 NPs were determined. Specifically, when the virus HSV-1 was treated with NPs at the concentration of 700 µg/mL (CC 20 ) for 1 h as well as for 3 h, viral suppression degrees were 86.2% (p ¼ 0:002) as well as 99.4% (p<0:001), correspondingly. Likewise, when HSV-1 was treated with higher-concentration WO 3 NPs of 1000 µg/mL (CC 50 ) for time frames of 1 h and also for 3 h, viral inhibition rates were 92% (p<0:001) as well as 99.9% (p<0:001), as revealed in Figure 7. the infected cells. The most pronounced reduction in CPE formation was observed at the highest concentration tested, 700 μg/mL (Figure 8). Analysis via qRT-PCR demonstrated that WO 3 NPs at concentrations of 250, 500, and 750 μg/mL significantly decreased the number of HSV-1 genomic DNA copies, achieving inhibition rates of 82%, 87.5%, and 96.5%, respectively (p<0:001) (Figure 9 and Table 1).
## 3.3. Antiviral Assays
## 4. Discussion
According to the findings of the current study, WO 3 NPs showed a virucidal impact that reduced the viral load after exposure to the virus for 1 and 3 h at the two concentrations that were studied (700 and 1000 µg/mL). Furthermore, increasing concentration and exposure time enhanced the extent of viral load reduction. This indicates that the antiviral effects of WO 3 NPs are both dose-dependent and time-dependent. Specifically, as the exposure time and concentration increase, the antiviral effects are significantly enhanced. The second set of experiments was to determine the efficacy of WO 3 NPs in a post-infection environment. The aim was to find out if the NPs were capable of inhibiting or interrupting viral replication at some time after the virus had infected the cells. The results of the study indicated that the antiviral efficiency of the WO 3 NPs was enhanced with increased dosages, with the best inhibitory potential being realized in the maximum inhibitory concentration of 700 µg/mL.
As we were already familiar with the photoactivity of WO 3 NPs, such that they create reactive oxygen species (ROS) when illuminated for enhancing their antimicrobial activity [18], it is noteworthy that all of our assays were conducted in the dark in order to prevent such an effect. It thus suggests that the achieved antiviral activity is most likely due to mechanisms that do not involve ROS formation, that is, direct interaction with viral entities or inhibition of the process of cell uptake. The results of the cytotoxicity assay of MTT also indicated that the WO 3 NPs exhibit biocompatible and safe properties, whereas higher toxicity was exhibited by copper oxide (CuO) and zinc oxide (ZnO) NPs in similar Vero cell models [19,20]. It may be concluded from the findings that the WO 3 NPs may be placed in the list of safe and biocompatible nanomaterials.
Up to the present, no research study has explored the antiviral behavior of WO 3 nanostructures against any virus of human or animal origin; thus, the present study is original. Furthermore, few research works have analyzed their effectiveness against bacteria, and more than that, there is still limited research on the antibacterial potential of WO 3 NPs. As an example, the antibacterial potential of WO 3 nanorods against Escherichia coli was investigated by Ghasempour et al. [21] in darkness and in the presence of visible light. Exposing the WO 3 nanorods to visible light, in fact, showed a strong antibacterial potential with a >92% inactivation of bacteria after exposure for 24 h at room temperature. Baroot et al., in another research, analyzed the antibacterial potential of WO 3 NPs against Staphylococcus aureus and Pseudomonas aeruginosa [22]. As per data, the WO 3 NPs notably inhibit the action of P. aeruginosa and S. aureus. Bashir et al. [23] assessed antimicrobial properties of WO 3 NPs coated by antibiotics (ampicillin, penicillin, and ciprofloxacin). According to their results, WO 3 NPs paired with ciprofloxacin demonstrated the highest antibacterial activity, followed by those linked with penicillin and ampicillin. In line with these studies, Matharu et al. [24] evaluated the antimicrobial activity of WO 3 NPs against Gram-negative and Gram-positive bacteria. For the first time, this study also investigated the effects of WO 3 NPs on a DNA virus (bacteriophage T4). Based on their findings, WO 3 NP at high concentrations has shown remarkable efficacy against bacteriophage T4 and Gram-positive bacteria (S. aureus), making it an appropriate antimicrobial agent.
However, there are a limited number of studies investigating the antiviral activity of other forms of tungsten NPs. Overall, the discovery of antiviral and antibacterial functions of tungsten has originated from the Ren et al. [25] patent, demonstrating the virucidal efficacy of tungsten NPs, especially combined with other effective antimicrobial components. According to this study, tungsten carbide reduced the avian H5N1 influenza virus by 99.99% after a treatment time of 30 min. Antiviral properties of tungsten carbide NPs against four viruses, vaccinia virus, human adenovirus type 5, poliovirus type 1, and murine norovirus, were also investigated in another study performed by Pfaff et al. [26]. Findings demonstrated that tungsten carbide NPs could decrease the infectivity of all four viruses by at least four log 10 of TCID 50 /mL after 15 min. Similar to our results, a dose-effect curve demonstrated that the virucidal activity of NPs depended on particle concentration, and their virucidal activity increased with incubation time. According to these results, tungsten carbide NPs hold great promise for developing novel disinfection methods.
The results of our investigation have demonstrated that the antiviral activity of WO 3 NPs in the post-treatment experiment (after virus adsorption) is enhanced by increasing their concentration. When the virucidal activity of WO 3 NPs was evaluated at various concentrations, a similar increased antiviral activity was observed. The direct impact of WO 3 NPs on viral particles is assessed in the virucidal assay, while the influence of WO 3 NPs on different stages of viral replication is investigated in the post-treatment analysis.
It should be noted that HSV-1 replication occurs within host cells. Therefore, the higher the uptake of WO 3 NPs into the cells increases the likelihood of interfering of WO 3 NPs with different stages of the HSV-1 replication cycle. As electron microscope images show, WO 3 NPs exhibit a rod-like morphology. Previous observations indicate that the morphologies of NPs affect the mechanisms of endocytosis [27]. While spherical NPs are isotropic, rod-shaped WO 3 NPs present axial and radial facets, which may enhance their interactions with viral particles or cellular surfaces [27].
In addition to morphology, the size of WO 3 NPs is a key factor contributing to their antiviral activity. In our study, the average size of WO 3 NPs was ~45 nm. The antiviral and antimicrobial activity of NPs is influenced by multiple factors, including their chemical composition, protein corona, and dose metrics (e.g., mass versus surface area), in addition to size [28,29]. These factors collectively contribute to the high antiviral activity of WO 3 NPs observed in this study.
One of the drawbacks of the current study is that the specific mechanism of WO 3 NPs' antiviral action has not been determined. Additionally, our study did not assess the impact of WO 3 NPs against HSV-1 in other conditions, such as pretreatment of cells with WO 3 NPs before viral infection (pretreatment assay) or simultaneous treatment of cells with WO 3 NPs and HSV-1 (co-treatment assay).
## 5. Conclusion
This study is the first to investigate the antiviral activity of WO 3 NPs against a human virus. The findings of this investigation have demonstrated that WO 3 NPs can inhibit HSV-1 via two different mechanisms. In the first mechanism, the virus is directly impacted by these NPs, resulting in virus inactivation. The virus may be deactivated by this mechanism in a number of ways. WO 3 NPs, for instance, can adhere to the virus or its ligands and stop it from binding to its host receptor. This inhibits the virus's ability to enter cells and stops infection. Furthermore, it is likely that the WO 3 NPs can physically induce structural damage to the virus, resulting in viral inactivation. Meanwhile, findings of the present study indicated that the incorporation of WO 3 NPs following HSV-1 infection of the cells is capable of lowering viral load. This further suggests that the viral replication process in the life cycle could be obstructed by these NPs at some stage. Ultimately, based on the positive findings of the present study, WO 3 NPs shall remain a potential candidate for further preclinical studies for fighting off HSV-1 infection. Nevertheless, considering that the present study is restricted to in-vitro Vero cell cultures and deals solely with the strain of HSV-1 KOS, the results may not directly extrapolate into in-vivo settings in humans or in animal models or into other subtypes of HSV-1. Notable parameters, including immune response, drug delivery, and system toxicity, among others, remain unexplored. Further research employing laboratory animals, such as mouse or rat models, and testing additional HSV-1 strains, including acyclovir-resistant strains, is necessary to validate these results and explore their therapeutic potential.
## References
1. Duan, Sun, Li (2023) "Herpes Simplex Virus 1 MicroRNAs: An Update" *Intervirology*
2. (2025) "Herpes Simplex Virus-Key Facts"
3. Bello-Morales, Andreu, López-Guerrero (2020) "The Role of Herpes Simplex Virus Type 1 Infection in Demyelination of the Central Nervous System" *International Journal of Molecular Sciences*
4. Su, Han, Shi (2024) "An Updated Review of HSV-1 Infection-Associated Diseases and Treatment, Vaccine Development, and Vector Therapy Application" *Virulence*
5. Sala, Ricchi, Giovati et al. (2024) "Anti-Herpetic Activity of Killer Peptide (KP): An In Vitro Study" *International Journal of Molecular Sciences*
6. Xu, Zhong, Xi et al. (2022) "Medicinal Plants and Natural Compounds Against Acyclovir-Resistant HSV Infections" *Frontiers in Microbiology*
7. Tahir, Iqbal, Shad (2025) "Biosynthesis of Zr-Doped WO3 Nanoparticles: Evaluation of Antibacterial, Antioxidant, and Enzymatic Activities" *Microbial Pathogenesis*
8. Hariharan, Gnanavel, Sathiyapriya et al. (2019) "A Review on Tungsten Oxide (WO3) and Their Derivatives for Sensor Applications" *International Journal of Advanced Science and Engineering*
9. Liman, Başbuğ, Ali et al. (2021) "Cytotoxic and Genotoxic Assessment of Tungsten Oxide Nanoparticles in Allium cepa Cells by Allium Ana-Telophase and Comet Assays" *Journal of Applied Genetics*
10. Hamidzade, Monavari, Kiani et al. (2024) "Enhanced Synergistic Antiviral Effects of Thermally Expanded Graphite and Copper Oxide Nanosheets in the Form of a Novel Nanocomposite Against Herpes Simplex Virus Type 1" *Microbial Pathogenesis*
11. Aldrees, Khan, Alzahrani et al. (2023) "Synthesis and Characterization of Tungsten Trioxide (WO3) as Photocatalyst Against Wastewater Pollutants" *Applied Water Science*
12. Tijani, Ugochukwu, Fadipe et al. (2019) "One-Step Green Synthesis of WO. 3 Nanoparticles Using Spondias mombin Aqueous Extract: Effect of Solution pH and Calcination Temperature" *Applied Physics A*
13. Adhikari, Sarkar, Maiti (2014) "Synthesis and Characterization of WO3 Spherical Nanoparticles and Nanorods" *Materials Research Bulletin*
14. Peng, Wang, Yu (2020) "2 S Gas Sensor at Room Temperature Based on WO. 3/rGO Hybrids" *Journal of Materials Science: Materials in Electronics*
15. Hammad, El-Bery, El-Shazly et al. (2018) "Effect of WO3 Morphological Structure on Its Photoelectrochemical Properties" *International Journal of Electrochemical Science*
16. Díaz-Reyes, Dorantes-García, Pérez-Benítez et al. (2008) "Obtaining of Films of Tungsten Trioxide (WO3) by Resistive Heating of a Tungsten Filament"
17. Pochapski, Carvalho Dos Santos, Leite et al. (2021) "Zeta Potential and Colloidal Stability Predictions for Inorganic Nanoparticle Dispersions: Effects of Experimental Conditions and Electrokinetic Models on the Interpretation of Results" *Langmuir*
18. Ishfak, Uddin, Rahman et al. (2025) "A Brief Review on Applications of Tungsten Oxide Nanoparticles" *Journal of Medicinal and Nanomaterials Chemistry*
19. Tavakoli, Ataei-Pirkooh, Sadeghi (2018) "Polyethylene Glycol-Coated Zinc Oxide Nanoparticle: An Efficient Nanoweapon to Fight Against Herpes Simplex Virus Type 1" *Nanomedicine*
20. Tavakoli, Hashemzadeh (2020) "Inhibition of Herpes Simplex Virus Type 1 by Copper Oxide Nanoparticles" *Journal of Virological Methods*
21. Ghasempour, Azimirad, Amini et al. (2015) "Visible Light Photoinactivation of Bacteria by Tungsten Oxide Nanostructures Formed on a Tungsten Foil" *Applied Surface Science*
22. Al Baroot, Drmosh, Alade (2022) "Investigating the Antibacterial Activity of Nanostructured Tungsten Oxide Prepared by Pulsed Laser Ablation at Different Hydrogen Peroxide Concentrations" *Optical Materials*
23. Bashir, Khan, Aqib et al. (2024) "Multifunctional Integration of Tungsten Oxide (WO3) 10 IET Nanobiotechnology Coating: A Versatile Approach for Enhanced Performance of Antibiotics Against Single Mixed Bacterial Infections" *Microbial Pathogenesis*
24. Matharu, Ciric, Ren et al. (2020) "Comparative Study of the Antimicrobial Effects of Tungsten Nanoparticles and Tungsten Nanocomposite Fibres on Hospital Acquired Bacterial and Viral Pathogens" *Nanomaterials*
25. Ren, Oxford, Reip et al. "Anti-Viral Formulations Nanomaterials and Nanoparticles"
26. Pfaff, Glück, Hoyer et al. (2019) "Tungsten Carbide Nanoparticles Show a Broad Spectrum Virucidal Activity Against Enveloped and Nonenveloped Model Viruses Using a Guideline-Standardized In Vitro Test" *Letters in Applied Microbiology*
27. Zhang, Xu, Tian et al. (2018) "Tailoring the Morphology of AIEgen Fluorescent Nanoparticles for Optimal Cellular Uptake and Imaging Efficacy" *Chemical Science*
28. Ali, Alani, Ahmed et al. (2024) "Effect of Biosynthesized Silver Nanoparticle Size on Antibacterial and Anti-Biofilm Activity Against Pathogenic Multi-Drug Resistant Bacteria" *OpenNano*
29. Lin, Verma, Saini (2021) "Antiviral Nanoparticles for Sanitizing Surfaces: A Roadmap to Self-Sterilizing Against COVID-19" *Nano Today* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724328&blobtype=pdf | # Thyroid hormone deficiency worsens outcomes in vaccinia virus infection
Laura Notario, Erika Guerrero-Espinosa, Manuel Nistal, Pilar Lauzurica, Ana Aranda, Susana Alemany
## Abstract
Intranasal inhalation of the vaccinia virus in mice leads to acute lung infection followed by peripheral organ damage. Our findings demonstrate that hypothyroidism significantly exacerbates disease severity. Hypothyroid mice exhibit higher disease scores, elevated lung viral loads, and more extensive tissue damage in both the lungs and peripheral organs. Notably, only hypothyroid mice reach the experimental endpoint, underscoring their heightened vulnerability. Hypothyroid mice display a defective splenic immune response, with no amplification of T and B lym phocytes, but the increased susceptibility to vaccinia virus infection also persists in hypothyroid lymphocyte-deficient Rag2 -/-mice. Prior to infection, hypothyroid mice show a reduced pool of lung alveolar macrophages, and lung viral loads are significantly elevated in these animals as early as 1 day post-infection, suggesting that an impaired innate immune response is involved in their increased susceptibility to vaccinia virus. Indeed, transfer of primary alveolar macrophages into the alveolar macrophage-deficient lungs of hypothyroid mice significantly alleviates disease symptoms. In euthyroid mice, circulating thyroid hormone levels decrease during infection, a well-documented response in sepsis and critical illness, known as non-thyroidal illness syndrome. Further highlighting the link between thyroid hormones and immune defense, we show that SRT1720, a Sirtuin 1 activator, reduces thyroid hormone levels and also worsens vaccinia infection when administered to euthyroid mice. In summary, our study reveals that hypothyroidism aggravates vaccinia virus infection. While the role of thyroid hormone decline in various diseases remains a topic of debate, our results suggest that thyroid hormones play a protective role in viral pulmonary infections. IMPORTANCE Vaccinia virus serves both as a recombinant vaccine platform and as a model for studying human smallpox and monkeypox infections, which are associated with high mortality rates. Here, we show that hypothyroidism in mice aggravates the severity of vaccinia virus infection due to a deficient splenic immune response and to a marked reduction in lung alveolar macrophages, a key cell population in defense against respiratory pathogens. Intratracheal administration of primary alveolar macrophages improves disease symptoms during the early phase of infection in hypothyroid animals. Hypothyroidism also impairs the amplification of splenic lymphocytes, which play a key role in defense against viral infection. Furthermore, vaccinia virus infection reduces thyroid hormone levels in euthyroid mice, a phenomenon named "non-thyroidal illness syndrome" that often occurs in septic patients, suggesting that, in the context of viral pulmonary infections, thyroid hormone replacement might be a useful therapeutic option.KEYWORDS immune response, vaccinia virus, thyroid hormones, alveolar macro phages, Sirtuin 1
V accinia virus (VACV), cowpox virus, and variola virus, which causes smallpox, are members of the Poxviridae family, a double-stranded DNA virus family that can cause disease in humans (1). VACV was used to eradicate smallpox, is currently used as a vector for immunization against other pathogens, and is extensively employed as a model for human smallpox and monkeypox, which have a high death rate (2). In mice, inhalation of VACV infects the lungs, where it replicates exponentially, resulting in lung inflammation and damage. Subsequently, the infection also affects other organs (3). VACV infection clearance is first controlled by the innate immune response, with alveolar macrophages (AMs) representing the first barrier against lung viral infection (4, 5). However, if VACV infection is not resolved, a powerful adaptive immune response by lymphocytes is required for recovery from respiratory VACV infection (3,4). In the orchestrated response to viral infection, the production of pro-inflammatory cytokines and chemokines also plays an important role (5).
Thyroid disorders profoundly impact patients' lives on a chronic basis and pose a substantial global public health burden. These conditions, encompassing both hypothyroidism and hyperthyroidism, span a spectrum from subclinical to severe and can manifest with a wide range of symptoms, affecting nearly every bodily system. Studies indicate a hypothyroidism prevalence of approximately 1.3% in the general population, which increases to 2%-15% in pregnant women and 7%-20% in individu als over 65 years of age, with higher incidence observed in specific iodine-deficient areas (https://www.who.int/data/nutrition/nlis/info/iodine-deficiency [6,7]). The thyroid hormones, triiodothyronine (T3), the active hormone, and thyroxine (T4), its precursor, are key regulators of growth, metabolism, and energy homeostasis (8). The link between the immune and endocrine systems is increasingly well established, with evidence of a functional crosstalk between the thyroid hormones and the immune system (reviewed in references 9, 10), which may contribute to pathophysiological conditions, including sepsis, inflammation, autoimmune diseases, and viral infections (11). Abnormally low plasma concentrations of thyroid hormones often occur in septic patients in the absence of thyroidal disease. This phenomenon, known as the "euthyroid sick syndrome" or "non-thyroidal illness syndrome" (NTIS) (12,13), is correlated with illness severity and outcome. During sepsis, the thyroid axis is affected, with reduced pituitary release of the thyroid-stimulating hormone and peripheral inhibition of the conversion of T4 to T3, the active hormone (12). NTIS has been regarded as an adaptive metabolic response in an attempt to ameliorate metabolic stress by lowering metabolic activity (12,14). However, whether the decrease in circulating thyroid hormones is protective or detrimental for an effective immune response is currently under debate and may depend on pathogen type, severity, and affected organ (12). Thus, we and others have shown that COVID-19 patients with low free T3 levels showed a worse prognosis and higher levels of the COVID-19 severity markers, together with a metabolomic cluster indicative of a high ketogenic profile (15,16), while hypothyroidism in mice confers tolerance to cerebral malaria, caused by the Plasmodium berghei parasite. This is mimicked by activation of the NAD + -dependent deacetylase Sirtuin 1 (SIRT1) (17). SIRT1 is a crucial regulator of metabolic processes in response to changes in nutrient availability, thereby controlling energy homeostasis and the metabolic state (18,19), and orchestrating immune and inflammatory responses during infection (20).
Given the prevalence of thyroid disorders in specific populations, we aimed to elucidate the role of thyroid hormones within the context of viral infection. We have analyzed the effect of hypothyroidism and hyperthyroidism on the response to intranasal VACV infection. Our results show that hypothyroid mice are more susceptible to the infection, presenting higher disease scores, elevated lung viral titers, and enhanced lung and peripheral organ damage. Treatment of euthyroid mice with a SIRT1 activator lowers thyroid hormone levels and mimics the effect of hypothyroidism, suggesting the involvement of this metabolic enzyme in the control of viral load. Circulating thyroid hormone levels regulate the expansion of splenic immune cells during VACV infection. Rag2 -/-hypothyroid mice, with a disrupted adaptive immune system, also exhibit increased susceptibility to VACV, suggesting that a deficient innate immune response is also involved in this increase. Before infection, hypothyroid mice show a markedly reduced AM population, along with higher viral load and more disease symptoms as early as 1 day post-infection (p.i.). Intratracheal transplant of primary AMs alleviates disease symptoms after infection, especially during the first days. In conclu sion, hypothyroidism causes weakened splenic and innate immune responses to VACV, increasing disease severity.
## RESULTS
## Circulating thyroid hormones after VACV infection
In humans, both hyperthyroidism and hypothyroidism exist on a spectrum from subclinical to severe (9,10). Prior to infection, hypothyroid mice were generated by feeding them an iodine-deficient diet containing an anti-thyroidal drug for 4 weeks. These mice exhibited a 50% reduction in circulating T3 levels and also a 25% decrease in T4 compared to euthyroid mice on the day of the infection (Fig. 1A). In contrast, oral thyroid hormone treatment increased circulating T3 levels, result ing in hyperthyroid mice that showed 1.8-fold higher levels. Because T3 treatment depletes pituitary thyrotropin and blocks endogenous thyroid secretion, mice were also given T4 to maintain normal circulating levels of this hormone. These observed variations in circulating T3 in these mouse models represent moderate hypothyroidism and hyperthyroidism, respectively, with a much higher incidence than severe thyroid disorders, particularly in the Western world. These three types of mice were intranasally infected with VACV. Upon inhalation in mice, VACV targets the lungs, where it undergoes an exponential replication phase for approximately the first 3 days. This is followed by a plateau with a high load that typically lasts for approximately another 3 days, with values dependent on the MOI. Subsequently, the infection's resolution or progression is contingent upon the prevailing experimental conditions (3,4,21). At day 7 p.i., hypothyroid mice did not show a further significant decrease of both T3 and T4, but euthyroid mice showed a 32% reduction in T3, while T4 levels were not reduced (Fig. 1A). After infection, T3 levels were reduced in hyperthyroid mice. The decrease in T3 levels observed at 7 day p.i. with constant iodine and T3 plus T4 hormone supply in euthyroid and hyperthyroid mice, respectively, indicates that VACV infection alone is sufficient to induce NTIS in these mice.
## Increased sensitivity of hypothyroid mice to VACV infection
Uninfected hypothyroid mice display reduced body weight, as expected (17), while hyperthyroid mice have normal body weight (Fig. 1B). From day 4 after intranasal VACV infection, all animals started to lose weight, but the relative weight loss in the hypothy roid mice was significantly more marked than in euthyroid or hyperthyroid animals (Fig. 1C), suggesting an increased susceptibility to VACV infection in hypothyroidism. By day 7 p.i., euthyroid mice had lost 7% of their initial body weight, hypothyroid mice 25%, and hyperthyroid mice 13%. Due to this significant weight loss in hypothyroid mice, they were considered to have reached their experimental endpoint, were humanely euthan ized, and the experiment was terminated. At the time of termination, these hypothyroid mice showed a high disease score of 3.5 ± 0.10, out of a possible maximum score of 4, whereas euthyroid and hyperthyroid animals showed much lower scores of 0.3 ± 0.09 and 0.3 ± 0.11, respectively. This increased score in hypothyroid mice was also consistently present throughout the infection. Symptoms were detectable as early as day 1 (Fig. 1C), when no symptoms were observed in control animals. Conversely, hyperthy roid mice displayed a small reduction in disease score compared to euthyroid mice, which was statistically significant at some time points. Low circulating levels of glucose are a severity marker of sepsis (22). At day 7 p.i., significant hypoglycemia was observed only in hypothyroid mice, which displayed a 60% decrease in their baseline glucose values. In contrast, euthyroid mice showed no decrease, while hyperthyroid mice exhibited a 25% decrease (Fig. 1D). Most importantly, at this time, hypothyroid mice had significantly higher lung VACV titers than euthyroid mice, with 12.3 ± 4.3 million more PFU per lung. Hyperthyroid mice, in comparison, did not present statistically significant differences in viral titers compared to the euthyroid group (Fig. 1E). Hemogram analysis showed that red blood cell counts were not significantly affected at day 7 of VACV infection, and that platelet numbers were similarly increased across all groups of infected mice. In contrast, the number of circulating leukocytes was reduced upon VACV infection and was significantly lower in the hypothyroid mice both prior to infection and p.i. Hypothyroid mice showed a 35% reduction in the number of leukocytes prior to infection compared to euthyroid mice. This value further decreased by 42% at the endpoint. This reduction largely reflects the lower number of lymphocytes, the more abundant circulating white blood cell population. Neutrophil counts were induced to similar values in all groups, although only in euthyroid mice was the increase statistically significant, and no significant differences in monocyte numbers were observed (Fig. S1A). As the expression and secretion of cytokines and chemokines regulate inflammation and initiate antiviral responses, we also determined the effect of hypothyroidism and hyperthyroidism on circulating cytokine responses to VACV. In serum collec ted prior to infection, no appreciable differences were observed among euthyroid, hypothyroid, and hyperthyroid mice. However, by day 7 p.i., several cytokines, including IFN-γ, IL-6, CCL2, and CXCL10, showed a stronger inflammatory response in hypothyroid mice compared to euthyroid controls, exhibiting approximately 1.5to 2.5-fold higher values in the hypothyroid group, whereas CCL5 levels were higher in the hyperthyroid group, indicating that hyperthyroid mice mounted a potent antiviral or T cell-mediated response compared to the hypothyroid and euthyroid animals (Fig. S1B).
## Increased lung pathology after VACV infection in hypothyroid mice
As the lung is the primary organ affected after VACV inhalation, we analyzed changes in lung weight and morphology at day 7 p.i. Lungs are smaller (Fig. 1F) and hypoplastic (Fig. 1G) in uninfected hypothyroid mice. However, after infection, a strong increase in lung weight was observed in these animals, while changes were weaker or not significant in euthyroid and hyperthyroid mice. This suggests an exaggerated inflammatory reaction in hypothyroid mice. Infected hypothyroid mice displayed strong lung edema, which was only focal or absent in the other groups, and abundant inflammatory lymphocyte and polymorphonuclear cell infiltration, very dense in the hypothyroid mice, was observed in the interstitial space of all infected animals (Fig. 1H). Furthermore, clear signs of necrotizing bronchopneumonia, bronchial exudate, and bronchiolar necrosis were only detected in infected hypothyroid mice at day 7 p.i. (Fig. 1I).
We also investigated the circulating levels of organ damage markers. No changes in creatine phosphokinase, urea, or creatinine were detected after infection, although basal creatinine levels were lower in the uninfected hyperthyroid mice. However, and according to the higher viral titers and pathology in hypothyroid mice, a clear increase of the liver transaminases alanine aminotransferase, aspartate aminotransferase, and of the general tissue damage marker lactate dehydrogenase levels was observed in the infected hypothyroid mice, while these parameters were little affected in the euthyroid and hyperthyroid groups (Fig. S2A). These results suggest that liver function could be altered in infected hypothyroid mice. Indeed, only in these mice did VACV infection cause a significant loss of liver weight (Fig. S2B), suggesting again that hypothyroid mice are more susceptible to VACV infection and that peripheral organ pathology might be detected in these animals. Liver histology (Fig. S2C) showed that VACV infection caused portal and parenchymal immune cell infiltration and Kupffer cell hyperplasia in all groups, but only hypothyroid mice displayed cell death as shown by the presence of eosinophilic Councilman bodies, which represent apoptotic or necrotic hepatocyte cell fragments. Moreover, the stronger periodic acid-Schiff (PAS) staining in non-infected hypothyroid liver, indicative of a higher glycogen content (23), was lost at day 7 p.i. (Fig. S2D). This is compatible with the observed hepatic weight reduction and the hypoglyce mia, indicating an increased energy input from glucose to fight infection in hypothyroid animals. Hyperthyroid livers showed the expected reduction of glycogen before infection (24,25), but no further decrease was found at day 7 p.i.
## Splenic immune host response to VACV infection in hypothyroidism and hyperthyroidism
The spleen, as a secondary lymphoid organ, plays a role against infection. Different types of splenic leukocytes are amplified to respond effectively to infection, which results in splenomegaly (3,4). This was indeed observed in euthyroid and hyperthy roid mice, where spleen weight was significantly higher at day 7 p.i. than prior to infection. However, hypothyroid mice showed a reduced splenic response with only a minor increase of spleen weight upon infection (Fig. 2A andB). Of note, uninfected hypothyroid mice also displayed spleen hypotrophy in homeostasis when compared with euthyroid and hyperthyroid animals (Fig. 2A andB), as previously described (17,26). Histological examination indicated an important increase in the splenic lymphoid nodules, often presenting germinal centers in euthyroid mice, even more marked in hyperthyroid mice, while hypothyroid spleens were atretic, showing smaller and fewer white pulp nodules (Fig. 2C). Correspondingly, at day 7 p.i. of VACV, euthyroid mice, and particularly hyperthyroid mice, showed a marked increase in the number of total splenic immune cells (Fig. 2D), including lymphocytes (Fig. 2E), with an essential role in the defense against VACV infection (3,4). In contrast, hypothyroid mice did not exhibit this expansion of immune cells and, prior to infection, already showed a marked reduction in the number of total splenocytes, as well as in the different immune cells tested, including B and T lymphocytes, red pulp macrophages (RPMs), neutrophils, and monocytes, with respect to uninfected euthyroid and hyperthyroid spleens (Fig. 2E). All these results indicate that hyperthyroid mice are able to build up a very strong splenic immune response to VACV, whereas splenic hypothyroid cells do not expand in response to the viral infection. These data agree with Varedi et al. (27), showing that isolated splenic lymphocytes from hyperthyroid mice display a greater proliferative capacity than their counterparts from euthyroid mice after in vitro exposure to antigens from HSV-1, a virus that belongs to the Herpesviridae family.
## SIRT1 activation reduces thyroid hormone levels and increases susceptibility to VACV infection
SIRT1 has an important regulatory function in host defenses following infection (20), and mimics the effects of hypothyroidism on cerebral malaria (17). We then examined the effect of SRT1720, a SIRT1 activator, on the response to VACV in euthyroid mice. Daily treatment with SRT1720 from the day of infection accelerated body weight loss and resulted in a significant increase in disease score, although not as marked as that induced by hypothyroidism (Fig. 3A). The number of circulating leukocytes, lymphocytes, and neutrophils was also lower in infected euthyroid mice treated with the SIRT1 activator, again mimicking the effect of hypothyroidism (Fig. S3A). Euthyroid mice treated with SRT1720 during the course of the disease also displayed a reduced splenic weight after infection, although much less marked than that found in hypothyroid mice (Fig. S3B), and statistically significant differences in counts of splenic populations were only observed for RPMs, which were reduced, and monocytes, which were increased by the treatment (Fig. S3C). In contrast to the results in hypothyroid mice, the SIRT1 activator was not able to alter liver weight (Fig. S3B). Concomitantly, the peripheral organ damage markers alanine aminotransferase, aspartate aminotransferase, and lactate dehydrogen ase were elevated in infected hypothyroid mice, but this increase was not observed in infected euthyroid mice treated with SRT1720. No significant differences in creatine phosphokinase, urea, and creatinine were observed between the different groups (Fig. S3D).
Importantly, SIRT1 activation also elicited a significant increase of lung viral titers, which were in between the values found in the hypothyroid and euthyroid mice (Fig. 3B), and the same occurred with glucose levels, closer to hypothyroid mice than to euthyroid mice (Fig. 3C). The similar effects of hypothyroidism and SRT1720 on VACV infection prompted us to examine whether the SIRT1 activator alters thyroid hormone levels. A single daily administration of SRT1720 for 7 days significantly reduced T3 levels in euthyroid mice. However, this reduction was not as pronounced as the T3 deficiency observed in hypothyroid mice, which had been treated with a hypothyroid diet for 4 weeks. Moreover, VACV infection caused a further decrease of T3 levels in SIRT1-treated mice, which were no longer different from those found in hypothyroid animals. Circulat ing T4 levels were also between the values obtained in euthyroid and hypothyroid mice both before and after VACV infection (Fig. 3D). These results show that SIRT1 activation causes hypothyroidism, aggravating VACV infection, and corroborate that the virus causes NTIS, affecting primarily T3 levels.
## Hypothyroidism increases susceptibility to VACV infection in lymphocytedeficient mice
Considering the impaired amplification of hypothyroid splenic immune cells in response to VACV infection, we next analyzed this infection in euthyroid and hypothyroid Rag2⁻ / ⁻ mice along with euthyroid and hypothyroid wild-type (WT) mice. Rag2⁻ / ⁻ mice lack T and B cells, which constitute the adaptive immune system (28). As expected, these mice showed, 4 days after infection, a reduction in the number of different types of circulating white blood cells with respect to WT mice, while red blood cell and platelet counts were similar, with no significant differences between euthyroid and hypothyroid mice (Fig. 4A). Similarly, spleen cellularity, including mainly T and B lymphocytes, was very markedly reduced in Rag2 -/-mice with respect to WT mice (Fig. S4). In contrast, euthyroid Rag2 -/- mice, but not hypothyroid Rag2 -/-mice, presented elevated NK cells and monocytes after infection. RPMs showed a higher increase in Rag2 -/-mice independently of the thyroidal status and a similar number of neutrophils with respect to the WT animals. These differences in the number of hematopoietic cells support the different patterns of circulating cytokine expression after infection, with Rag2 -/-mice showing reduced levels of IFN-γ, IL-6, CCL2, CXCL1, and CXCL10, but an important increase of TNF-α and IL-1β levels. Hypothyroidism increased IFN-γ and IL-6 levels in both WT and Rag2 -/-mice, indicating a higher inflammatory state, while simultaneously reducing CXCL10 (Fig. 4B). Importantly, euthyroid Rag2 -/-mice showed accelerated clinical manifestations and a higher disease score than their euthyroid WT counterparts; and hypothyroidism further exacerbated disease symptoms (Fig. 4C). Hypothyroid mice also exhibited higher virus titers in their lungs (Fig. 4D). These results suggest that hypothyroidism confers a poor response to VACV infection, regardless of its impact on the adaptive immune system.
## Deficient innate immune response to VACV infection in hypothyroid mice
We then analyzed lung viral titers in euthyroid and hypothyroid mice 1 day after intranasal VACV infection, before activation of the adaptive immune response. Hypothy roid mice, but not euthyroid mice, showed early symptoms of infection, and significantly increased viral titers were observed in their lungs. These findings suggest that a deficient innate immune system may exacerbate disease severity in hypothyroid mice following VACV infection (Fig. 5A). One day after infection, the hemogram is still normal (Fig. S5A), but circulating TNF-α, IL-1β, and CCL2 were markedly lower in infected hypothyroid mice with respect to euthyroid mice, while CXCL10 levels were higher (Fig. S5B). Flow cytome try analysis revealed no difference in the percentage of hematopoietic cells in the lungs of euthyroid and hypothyroid mice prior to infection and at day 1 p.i. Among myeloid cells, the percentage of neutrophils was similar before infection and at day 1 p.i.; however, hypothyroid mice exhibited more than a twofold increase in their percentage, a change not observed in euthyroid mice (Fig. 5B), indicating an exacerbated damage (29). The percentage of monocytes and interstitial macrophages was comparable between the euthyroid and hypothyroid mice, but AMs, critical as the first barrier against viral pulmonary infection and with a key role in initiating the local antiviral innate immune response (29,30), showed a marked decrease in hypothyroid lungs prior to infection (Fig. 5B). AMs undergo cell death during exposure to pathogens (31). Indeed, at 1 day p.i., euthyroid mice showed a strong reduction in the number of AMs, almost equaling the numbers of hypothyroid AMs, which were similar before and after infection.
In a proteome array with 99 cytokines, chemokines, and other ligands, lungs from hypothyroid mice showed, before infection, changes in the expression of 19 proteins, 14 downregulated and 5 upregulated, with respect to the euthyroid lungs. GM-CSF, with a key role in AM maintenance (32)(33)(34), was barely detected in both euthyroid and hypothy roid lungs (Fig. S6A), and further analysis of this cytokine by flow cytometry indicated that hypothyroidism does not affect GM-CSF expression (Fig. S6B), suggesting that other factor(s), besides GM-CSF, control(s) the number of AMs.
Interestingly, splenic RPMs, which like AMs also develop early in mice (35), are also diminished in uninfected hypothyroid mice (Fig. 3), but bone marrow macrophages, which are derived from hematopoietic stem cells and maintained by different mecha nisms (36), showed similar numbers in femurs of hypothyroid and euthyroid mice (Fig. S7). Before infection, bone marrow B cells were reduced, and T cells were increased in hypothyroid mice, while the number of neutrophils and NK cells was similar in hypothy roid and euthyroid mice (Fig. S7).
Transferring AMs is a novel therapeutic strategy under preclinical investigation for various lung diseases, including infections (37-39). To determine whether increasing the number of AMs in hypothyroid mice improved their capacity to resolve intranasal VACV infection and to discern the relevance of AM deficiency and lymphocyte responsiveness in the inability to resolve VACV infection in these mice, we intratracheally transferred 0.5 million normal primary AMs or PBS into uninfected hypothyroid mice, as previously described (38,39). This transfer increased the percentage of AMs in their lungs by approximately 2.5-fold 1 day post-transfer, which was also the day of infection (Fig. 5C) and almost reached the percentage observed in euthyroid mice prior to infection (Fig. 5B). Notably, after infection, symptoms were reduced in hypothyroid animals that had been transferred AMs, with no detectable disease symptoms when innate immune cells dominated the immune response (Fig. 5D). By day 7 p.i., the disease score was still lower in hypothyroid mice transferred with AMs, and the circulating glucose levels were higher (Fig. 5E), indicating a better metabolic state (22). However, both types of mice showed similar PFU titers in lungs at this time point (Fig. 5E). These findings indicate that transferring AMs into the lungs of AM-deficient hypothyroid mice prior to infection, while improving some parameters during the course of the disease, is not sufficient to rescue the euthyroid phenotype, likely because it does not fully restore normal immune function.
## DISCUSSION
The Poxviridae family includes several viruses of medical and veterinary importance, and VACV can be used as a live recombinant vaccine against many different diseases (1). Therefore, it is necessary to define the determinants of the host response to these viruses in the respiratory tract, the main route of infection. We show here that thyroid hormones play a very important role in the overall severity of the disease. Hypothyroid mice exhibit more marked weight loss and stronger disease symptoms than euthyroid or hyperthyroid mice after intranasal instillation of VACV. Significantly, hypothyroid mice display substantially higher viral titers in the lungs and increased pulmonary pathology.
According to these data, it has been shown that following a 3 day intraperitoneal infection with HSV-1, spleens from hyperthyroid mice have a lower titer than spleens of euthyroid mice, while the highest titers are found in the spleens of hypothyroid mice (40). After intranasal VACV infection, if the acute lung infection is not resolved, the infection's course can lead to peripheral organ injury. In hypothyroid mice, we observed hepatic damage and strong hypoglycemia, which were not present in euthyroid or hyperthyroid mice. The hypoglycemia in hypothyroid mice correlates with the increased depletion of glycogen in the liver, indicating high metabolic stress (22). SIRT1 activation has been proposed as a mechanism for the treatment of several diseases (41). However, activation of SIRT1 during VACV infection increases disease severity and lung viral load. These effects may be related to a reduction in circulating thyroid hormone levels. The degree of hypothyroidism caused by treatment with the SIRT1 activator, SRT1720, is less intense than that observed in our hypothyroid animals, which were fed a low-iodine diet containing an anti-thyroidal drug for 1 month prior to infection, but still sufficient to increase susceptibility to VACV infection.
Interestingly, deiodinase 3 knockout mice, which show central hypothyroidism with low circulating levels of T4 and T3 (42), display significantly higher bacterial load in the lung than wild-type mice after infection with Streptococcus pneumoniae, corroborating that low thyroid hormone levels are also associated with increased bacterial load in the respiratory tract (43). Hypothyroid mice also have a more severe acute inflammatory lung damage than euthyroid mice in a model of ventilator-induced pulmonary injury (VIPI). Moreover, administration of T3 to deiodinase 2 knockout mice, which also exhibit reduced T3 and T4 levels, reduces damage, dampening the inflammatory response to VIPI (44). Clinically, hypothyroidism is prevalent in patients with idiopathic pulmonary fibrosis and has been associated with unfavorable prognosis (45), and it has been proposed that thyroid hormones may represent a potential therapy for the disease (46). Also, we and others have observed that COVID-19 patients with low thyroid hormone levels have a poorer prognosis than patients with higher hormone levels (15,16). Taken together, these results indicate a positive action of thyroid hormones in the control and outcome of various respiratory diseases.
In addition to VACV infection, other experimental models associated with lung injury display NTIS (43,44). Whether the reduction in circulating thyroid hormone levels associated with NTIS is beneficial or detrimental to the host immune response is still under debate (12). Our results suggest that in the context of viral pulmonary infection, thyroid hormones may exert protective properties, leading to an improved immune response. Thus, further studies are needed to determine whether thyroid hormone supplementation, either alone or as adjunctive therapy, might be favorable for the outcome of respiratory infections. However, hypothyroidism is beneficial in cerebral malaria (17), reinforcing the idea that NTIS may be advantageous or detrimental for a proper immune response, depending on the disease type, pathogen, and affected organ (12).
Hypothyroidism displays deficiencies in the innate as well as in the adaptive immune response to VACV infection. Hypothyroid mice show no adaptive immune response of splenic lymphocytes on day 7 post-infection. On the other hand, at day 1, when only the innate immune system is sensing the infection, hypothyroid mice show a higher degree of infection, and according to these data, RAG2 -/-mice lacking adaptive immune cells are also more sensitive to VACV infection. AMs are the predominant resident phagocytic cells in the lungs that respond to inhaled pathogens in the respiratory tract, initiate the local innate immune response (29,30), and maintain the functional integrity of lung epithelium. In this context, depletion of AMs contributes to an accumulation of myeloid cells, mainly neutrophils, in the lungs after VACV infection, which likely contributes to more severe disease in response to VACV infection (29). We observed a striking decrease in the number of AMs in uninfected hypothyroid lungs. Concurrently, both WT and RAG2 -/-hypothyroid mice developed higher disease scores from the infection's onset, during which the innate immune system governs the immune response. After VACV infection, different leukocyte populations are amplified in the spleen to effectively resolve the disease (3,4). Euthyroid and hyperthyroid mice show clear signs of splenic stimulation following VACV infection, whereas hypothyroid mice show a significantly reduced splenic response. However, this is not a rigid rule, as enhanced amplification of splenocytes has been observed in hypothyroid mice compared to euthyroid mice in experimental cerebral malaria caused by Plasmodium berghei (17). This suggests that circulating thyroid hormones control the splenic immune response in a pathogendependent manner.
In hypothyroid mice, the transfer of AMs significantly improved the disease score, an effect that was most pronounced during the first days of the infection. Hypothyroid mice also displayed higher glucose, an indicator of less metabolic stress, but were not able to resolve VACV infection as euthyroid mice did. The transfer of AMs to hypothyroid mice allows AMs to reach similar levels to those observed in euthyroid mice. This improves the first days of infection, when innate immune cells, including AMs, are primarily responsi ble for the response. However, this transfer does not overcome the deficient immune response observed when cells of the adaptive immune system take over the response. The intrinsic thyroid state of lymphocytes modulates their proliferative capacity, and thyroid hormones do not require the generation of other signals to mediate this effect. Indeed, isolated lymphocytes with different thyroid states show different proliferation rates in response to different signals (9,27).
In conclusion, in moderately hypothyroid mice, the defective immune response during VACV respiratory infection leads to increased lung damage, elevated circulat ing markers of peripheral tissue injury, low glucose levels, and the development of a non-resolving disease state. Therefore, thyroid hormone substitutive therapy should be considered for specific populations, such as pregnant women and individuals over 65 years, during respiratory viral infections.
## MATERIALS AND METHODS
## Mice and treatments
Female C57BL/6, BALB/c, and Rag2 -/-BALB/c mice were bred and housed under pathogen-free conditions at the animal facility of the Instituto de Salud Carlos III, Madrid, Spain. Unless otherwise indicated, all experiments were performed in C57BL/6 mice. To induce hypothyroidism, 4-to 5-week-old female mice were fed an iodine-deficient diet containing 0.15% of the anti-thyroidal drug propylthiouracil (E15551-04, Sniff) for 4 weeks before VACV infection. Euthyroid animals were fed with the same diet without propylthiouracil and supplemented with potassium iodide to contain 1.15 mg/kg (E15552-24, Sniff). Both diets were maintained until the end of the experi ments. Hyperthyroid mice were fed the control diet and were made hyperthyroid by adding T4 (25 ng/g of mice; IRMM468, Sigma-Aldrich) and T3 (95 ng/g of mice; T-2877, Sigma-Aldrich) in the drinking water from 14 days before infection until the end of the experiment. Drinking water was changed every other day. Euthyroid mice were injected i.p. daily starting on the day of infection with the SIRT1 activator, SRT1720 (HY-15145, Med Chem Express) at a dose of 20 mg/kg in 100 µL as previously described (17). SRT1720 was dissolved in 10% DMSO (D2438, Sigma-Aldrich) and 20% 2-hydroxypropyl-β-cyclodextrin (H5784, Merck) in PBS. Considering that hypothyroid mice have a lower weight, mice were inoculated intranasally with an MOI of 5,700 PFU per gram of body weight of the Western Reserve strain of VACV. The virus was grown in CV1 cells cultured in DMEM supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, and 5 µM β-mercaptoethanol, and virus titer was determined by plaque assay, as previously described (47,48).
## Lung VACV titer assay
The viral load in the lungs was measured by plaque-forming assay. Mice were sacrificed at the times indicated in the experiments, and lungs were harvested and stored at -80°C in 0.5 mL of PBS until use. The lungs were homogenized and freeze-thawed three times. Serial dilutions were plated on confluent CV1 cells, and after 24 hours of culture at 37°C, the plates were stained with crystal violet, and plaques were counted.
## Disease score
Mice intranasally infected with VACV were weighed daily until animals lost more than 25% of the initial weight when they were sacrificed using CO 2 overdose exposure, according to ethics guidelines. Symptoms observed daily to determine disease score were as follows: 0, no signs of disease, active, strong, curious, quick movements; 1, low grade: less active with occasional interruptions of activity, reduced alertness but appropriate response, slightly hunched; 2, medium grade: slow, drowsy, moves with difficulty, limited and delayed, hunched; 3, high grade: lethargic, immobile, severely hunched; 4, end point: death.
## Histopathological studies
Lungs, livers, and spleens were harvested without perfusion and were embedded in paraffin wax (253211, PanReac AppliChem). Serial 5 µm sections were stained with H&E (hematoxylin 75290, PanReac AppliChem; eosin 102439, Merck) or with PAS.
## Glucose, hemograms, and circulating tissue damage markers
Glucose levels were determined in blood drops from tails using Accu-Chek Aviva detector (6453970037, Roche). At sacrifice, blood samples were collected in EDTA tubes (1591126, EVEREST) by heart puncture, and 400 µL was sent to DYNAMIMED (Madrid) for analysis.
## Quantification of total circulating T3 and T4
Serum plasma was obtained by centrifugation of blood samples for 10 min at 6,000 rpm. T3 and T4 were measured using the Mouse Triiodothyronine (T3) ELISA kit (ab285259, Abcam) and Mouse Thyroxine (T4) ELISA kit (ab285258, Abcam), respectively. Assays were performed according to the manufacturer's instructions, including a standard curve with a range of 0-7.5 µg/mL T3 or 0-25 µg/dL T4. The absorbance was measured in the BioTek EL340 Microplate Reader at 450 nm and subsequently converted to the corresponding concentration.
## Flow cytometry for immune cell identification and cytokine analysis
Spleen was dissociated with the GentleMACS dissector and mouse spleen dissociation kit (130-095-926, Miltenyi), according to the manufacturer instructions. For lung cell dissociation, the mouse lung dissociation kit was used (130-095-927, Miltenyi). Both cell suspensions were filtered through a 70 µm cell strainer and pelleted by centrifugation for 5 min at 300 × g. Bone marrow cells were isolated from femurs cut at the ends by gentle centrifugation at 600 × g for 1 min and collected in 100 µL of FACS buffer (PBS, 2% FBS, and 5 mM EDTA). To lyse erythrocytes, pelleted cells were resuspended in 1 mL Versalyse lysing solution (A09777, Beckman Coulter). Two minutes later, 3 mL of FACS buffer was added, and cells were centrifuged and washed. The following markers were used to identify the different cell types from various tissues in spleen: B cells (B220 + CD3 -), T CD4 + cells (CD3 + CD4 + B220 -), T CD8 + cells (CD3 + CD8 + B220 -), NK cells (NKp46 + B220 - CD3 -), neutrophils (CD11b + Ly6G + ), RPMs (CD11b +/low Ly6G -F4/80 high ), and inflammatory monocytes (Ly6C + CD11B + Ly6G -); gating strategy is shown in Fig. S8; in lung: leukocytes CD45 + cells, neutrophils (CD45 + Ly6G + ), monocytes (CD45 + Ly6G -CD11c low/+ CD11b + MHCII - CD64 low/-), interstitial macrophages (CD45 + Ly6G -CD11c low CD11b + MHCII -CD64 + SYGLEC -), and AMs (CD45 + Ly6G -CD11c + CD11b + MHCII -CD64 + SYGLEC + ); gating strategy is shown in Fig. S9; and in bone marrow: B cells (B220 + CD3 -), T cells (CD3 + B220 -), neu trophils (Ly6G + CD3 -B220 -), natural killer (NKp56 + Ly6G -CD3 -B220 -), and macrophages (F4/80 + CD11b + Ly6G -CD3 -B220 -); gating strategy is shown in Fig. S10. The antibodies used to identify are listed in Table S1. Cytokines were determined using the LEGENDplex Mouse Anti-Virus Response Panel kit (740621, BioLegend) according to the manufactur er's instructions. Cells and cytokine samples were subjected to flow cytometry analy sis on a Cytoflex S (Beckman and Coulter). Data were analyzed using the CytoExpert (Beckman and Coulter) or LEGENDplex v.8.0 (BioLegend) software.
## Transfer of AMs
One day before infection, AMs were obtained from 8-to 10-week-old hypothyroid female mice by repeated bronchoalveolar lavages via the trachea with 37°C PBS-EDTA, as described previously (49,50). The cells were centrifuged and analyzed by flow cytometry, which confirmed a viability and purity of approximately 95% of viable AMs. Subsequently, 0.5 × 10 6 of these primary AMs in 20 µL PBS-EDTA were transferred to isoflurane-anesthetized hypothyroid recipient mice. For this, tracheas were surgically exposed and cannulated for instillation. Naive control mice received an equal volume of sterile PBS-EDTA.
## Statistical analysis
Two-tailed Student's t-tests were used for comparisons between two groups. One-way ANOVA with post-hoc Tukey test was used to compare all pairs of columns from at least three different groups. Two-way ANOVA was used for curve comparison. The results are always expressed as means ± SEM. P values <0.05 were considered statistically significant. Significance is shown in the figures as *P < 0.05, **P < 0.01, and ***P < 0.001. Statistics were performed with GraphPad Prism 7.0 software.
## References
1. Buller, Palumbo (1991) "Poxvirus pathogenesis" *Microbiol Rev*
2. Chapman, Nichols, Martinez et al. (2010) "Animal models of orthopoxvirus infection" *Vet Pathol*
3. Goulding, Bogue, Tahiliani et al. (2012) "CD8 T cells are essential for recovery from a respiratory vaccinia virus infection" *J Immunol*
4. Xu, Johnson, Liggitt et al. (2004) "Cellular and humoral immunity against vaccinia virus infection of mice" *J Immunol*
5. Akira, Uematsu, Takeuchi (2006) "Pathogen recognition and innate immunity" *Cell*
6. Chaker, Razvi, Bensenor et al. (2022) *Hypothyroidism. Nat Rev Dis Primers*
7. Lee, Pearce (2023) "Hyperthyroidism: a review" *JAMA*
8. Mullur, Liu, Brent (2013) "Thyroid hormone regulation of metabolism" *Physiol Rev*
9. Vito, Incerpi, Pedersen et al. (2011) "Thyroid hormones as modulators of immune activities at the cellular level" *Thyroid*
10. De Luca, Davis, Lin et al. (2020) "Thyroid hormones interaction with immune response, inflammation and non-thyroidal illness syndrome" *Front Cell Dev Biol*
11. Jara, Muñoz-Durango, Llanos et al. (2017) "Modulating the function of the immune Full-Length Text Journal of Virology December"
12. *Immunol Lett*
13. Fliers, Boelen (2021) "An update on non-thyroidal illness syndrome" *J Endocrinol Invest*
14. Vidart, Jaskulski, Kunzler et al. (2022) "Non-thyroidal illness syndrome predicts outcome in adult critically ill patients: a systematic review and metaanalysis" *Endocr Connect*
15. Melis, Miller, Peters et al. (2023) "The role of hormones in sepsis: an integrated overview with a focus on mitochondrial and immune cell dysfunction" *Clin Sci (Lond)*
16. Amich, Anguita, Escribano-Serrat et al. (2022) "Free triiodothyronine levels and age influences the metabolic profile and COVID-19 severity parameters in euthyroid and levothyroxine-treated patients" *Front Endocrinol*
17. Li, Hou, Mu et al. (2023) "Thyroxine changes in COVID-19 pandemic: a systematic review and meta-analysis" *Front Endocrinol*
18. Rodriguez-Muñoz, Sánchez, Pérez-Benavente et al. (2022) "Hypothyroidism confers tolerance to cerebral malaria" *Sci Adv*
19. Chang, Guarente (2014) "SIRT1 and other sirtuins in metabolism" *Trends Endocrinol Metab*
20. Houtkooper, Pirinen, Auwerx (2012) "Sirtuins as regulators of metabolism and healthspan" *Nat Rev Mol Cell Biol*
21. Kim, Silwal (2022) "Sirtuin 1 in host defense during infection" *Cells*
22. Hayasaka, Ennis, Terajima (2007) "Pathogeneses of respiratory infections with virulent and attenuated vaccinia viruses" *Virol J*
23. Wang, Chen, Liu et al. (2021) "Blood glucose levels and mortality in patients with sepsis: dose-response analysis of observational studies" *J Intensive Care Med*
24. Shafiei, Gonczi, Rahman et al. (2014) "Detecting glycogen in peripheral blood mononu clear cells with periodic acid schiff staining" *J Vis Exp*
25. Aranda, Montoya, Herrera (1972) "Effects of hypo-and hyperthyroidism on liver composition, blood glucose, ketone bodies and insulin in the male rat" *Biochem J*
26. Hoefig, Harder, Oelkrug et al. (2016) "Thermoregulatory and cardiovascular consequences of a transient thyrotoxicosis and recovery in male mice" *Endocrinology*
27. Sánchez, Contreras-Jurado, Rodríguez et al. (2020) "Hematopoiesis in aged female mice devoid of thyroid hormone receptors" *J Endocrinol*
28. Varedi, Shiri, Moattari et al. (2014) "Hyperthyroid state or in vitro thyroxine treatment modulates TH1/TH2 responses during exposure to HSV-1 antigens" *J Immunotoxicol*
29. Shinkai, Rathbun, Lam et al. (1992) "RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement" *Cell*
30. Rivera, Hutchens, Luker et al. (2007) "Murine alveolar macrophages limit replication of vaccinia virus" *Virology (Auckl)*
31. He, Chen, Mullarkey et al. (2017) "Alveolar macrophages are critical for broadly-reactive antibody-mediated protection against influenza A virus in mice" *Nat Commun*
32. Mould, Barthel, Mohning et al. (2017) "Cell origin dictates programming of resident versus recruited macrophages during acute lung injury" *Am J Respir Cell Mol Biol*
33. Guilliams, Kleer, Henri et al. (2013) "Alveolar macrophages develop from fetal monocytes that differentiate into longlived cells in the first week of life via GM-CSF" *J Exp Med*
34. Hashimoto, Chow, Noizat et al. (2013) "Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes" *Immunity*
35. Misharin, Morales-Nebreda, Reyfman et al. (2017) "Monocyte-derived alveolar macrophages drive lung fibrosis and persist in the lung over the life span" *J Exp Med*
36. Castrillo (2018) "Origin and specialization of splenic macrophages" *Cell Immunol*
37. Calvi, Link (2015) "The hematopoietic stem cell niche in homeosta sis and disease" *Blood*
38. Subramanian, Busch, Molawi et al. (2022) "Thyroid hormones and the control of cell proliferation or cell differentiation: paradox or duality" *Nat Immunol*
39. Malainou, Abdin, Lachmann et al. (2023) "Alveolar macrophages in tissue homeostasis, inflammation, and infection: evolving concepts of therapeutic targeting" *J Clin Invest*
40. Lim, Cervantes, Pham et al. (2021) "Alveolar macrophages: novel therapeutic targets for respiratory diseases" *Expert Rev Mol Med*
41. Varedi, Moattari, Amirghofran et al. (2014) "Effects of hypo-and hyperthyroid states on herpes simplex virus infectivity in the rat" *Endocr Res*
42. Lavu, Boss, Elliott et al. (2008) "Sirtuins--novel therapeutic targets to treat age-associated diseases" *Nat Rev Drug Discov*
43. Hernandez, Martinez, Fiering et al. (2006) "Type 3 deiodinase is critical for the maturation and function of the thyroid axis" *J Clin Invest*
44. Boelen, Kwakkel, Wieland et al. (2009) "Impaired bacterial clearance in type 3 deiodinase-deficient mice infected with Streptococcus pneumoniae" *Endocrinology*
45. Barca-Mayo, Liao, Dicosmo et al. (2011) "Role of type 2 deiodinase in response to acute lung injury (ALI) in mice" *Proc Natl Acad Sci U S A*
46. Oldham, Kumar, Lee et al. (2015) "Thyroid disease is prevalent and predicts survival in patients with idiopathic pulmonary fibrosis" *Chest*
47. Yu, Tzouvelekis, Wang et al. (2025) *Full-Length Text Journal of Virology*
48. Herzog, Bianco, Kaminski (2018) "Thyroid hormone inhibits lung fibrosis in mice by improving epithelial mitochondrial function" *Nat Med*
49. Notario, Alari-Pahissa, De Molina et al. (2016) "CD69 deficiency enhances the host response to vaccinia virus infection through altered NK cell homeostasis" *J Virol*
50. Notario, Redondo-Antón, Alari-Pahissa et al. (2019) "CD69 targeting enhances anti-vaccinia virus immunity" *J Virol*
51. Busch, Favret, Geirsdóttir et al. (2019) "Isolation and long-term cultivation of mouse alveolar macrophages" *Bio Protoc*
52. Mcquattie-Pimentel, Ren, Joshi et al. (2021) "The lung microenvironment shapes a dysfunctional response of alveolar macrophages in aging" *J Clin Invest* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12733864&blobtype=pdf | # Does the Chimerization Process Affect the Immunochemical Properties of WNV-Neutralizing Antibody 900?
Anastasiya Isaeva, Valentina Nesmeyanova, Daniil Shanshin, Nikita Ushkalenko, Ekaterina Volosnikova, Tatiana Esina, Elena Protopopova, Victor Svyatchenko, Valery Loktev, Sergey Olkin, Elena Danilenko, Elena Kazachinskaia, Dmitriy Shcherbakov
## Abstract
West Nile fever is an infectious disease caused by the West Nile virus (WNV), which is transmitted by mosquitoes. Epidemiological surveillance confirms the potential risk of WNV infection in human populations. The lack of specific antiviral therapeutics and vaccines against WNV underscores the urgent need to develop effective therapeutic approaches. In this study, a recombinant chimeric monoclonal antibody (mAb) 900 was generated based on the broadly neutralizing and protective murine mAb 9E2. The antigen-binding regions of the murine mAb were fused with the constant domains (CH2-CH3) of human IgG1. Two key amino acid clusters, M252/S254/T256 and H433/N434, were introduced into the CH2-CH3 domains to enhance the affinity of mAb 900 for the neonatal Fc receptor (FcRn). The engineered mAb 900 was produced in CHO cells and purified to high homogeneity. Biophysical characterization confirmed its stability and correct dimeric assembly. Comparative analysis demonstrated that mAb 900 retained the high antigen-binding affinity and potent virus-neutralizing activity of its murine predecessor. Most importantly, mAb 900 demonstrated significant protective efficacy in a lethal mouse model of WNV infection. These results establish the proof of concept for mAb 900 as a promising candidate for further preclinical development against WNV infection.
## 1. Introduction
The West Nile virus (WNV) is an orthoflavivirus belonging to the Japanese encephalitis antigenic complex and transmitted by mosquitoes [1][2][3]. According to the World Health Organization, WNV is currently endemic in Africa, the Middle East, the United States, Australia, Europe, and Asia, reflecting its capacity for global dissemination [4][5][6]. In 2024, the United States reported the highest number of laboratory-confirmed West Nile fever (WNF) cases, totaling 1466 infections, which represents a 1.8-fold decrease compared to the number of cases in 2023 [6]. Nevertheless, the long-term average incidence in the United States remains high, with an annual mean of 2234 cases recorded between 1999 and 2023.
In Europe, 1437 cases were reported, primarily in Central and Mediterranean regions, with Italy accounting for the largest number (455 cases) [6]. In Asia and Australia, WNF cases are sporadic, and most Asian countries lack mandatory epidemiological surveillance for this infection [6,7]. A consistent expansion of WNV-endemic areas has been observed. Between 1999 and 2020, WNV infections were documented in 51 regions of the Russian Federation. Epidemiological analysis for 2024 revealed a 2.1-fold increase in cases compared with 2023 (440 vs. 210 cases) [6]. The rising frequency of outbreaks and the emergence of infections in temperate regions indicate the evolving epidemiology of orthoflaviviruses and their capacity to adapt to new environmental conditions [4].
The absence of specific therapeutic agents and vaccines against WNV underscores the urgent need to identify effective countermeasures [1,8,9]. Among the most promising tools in combating viral infections are monoclonal antibodies (mAbs) [10][11][12]. For instance, the mAb Palivizumab, developed for the prevention of severe respiratory syncytial virus infection in children, has demonstrated both efficacy and safety and is currently approved for use in more than 70 countries worldwide [13,14]. Neutralizing antibodies targeting the WNV envelope (E) protein have been shown to inhibit WNV infection both in cell culture and in animal models [8,15]. Therefore, ongoing studies are focused on developing mAbs with enhanced efficacy against WNV.
Despite their high therapeutic potential, the clinical application of mAbs is associated with several challenges. One of the major issues is the adaptation of heterologous antibodies that exhibit promising neutralizing activity against viral pathogens. The high immunogenicity of heterologous antibodies can trigger undesirable immune responses, such as the production of antibodies against the drug [16]. These responses may compromise therapeutic efficacy and increase the risk of adverse effects, including allergic reactions. To minimize these risks, antibody humanization strategies have been developed [17]. The humanization process involves substituting amino acid residues in heterologous antibodies with their human counterparts to improve compatibility with the patient's immune system and reduce immunogenicity [18]. However, this process carries a risk of diminished functional activity, as such substitutions may alter the three-dimensional structure of the complementaritydetermining regions and weaken antigen binding affinity. Studies have demonstrated that even minor modifications within these critical regions can significantly reduce antigen binding and, consequently, therapeutic efficacy [17,19,20]. A well-established alternative approach is the development of chimeric antibodies, which are constructed by fusing the variable regions of the parental animal antibody with the constant (Fc) regions of a human antibody. This strategy preserve the neutralizing activity of the parental antibody while reducing immunogenicity [20]. The chimerization of antibodies, which involves the replacement of constant domains, is generally regarded as more predictable than humanization, a process that requires modifications within the variable regions. Nevertheless, even chimerization is not fully predictable. The substitution of constant domains can disrupt interdomain interactions and alter the overall antibody conformation, potentially compromising its stability and antigen-binding affinity. Consequently, a thorough characterization of any chimeric antibody must include a comparative assessment of its immunochemical properties against those of the parental murine antibody.
The objective of this study was to engineer a chimeric, neonatal Fc receptor (FcRn)enhanced mAb 900 and to evaluate its potential against WNV through an analysis encompassing in vitro immunochemical characterization and in vivo protective efficacy.
## 2. Results
## 2.1. Fc Fragment Engineering of Monoclonal Antibody (mAb) 900
In this study, we generated a recombinant antibody in a single-chain variable fragment (scFv) fused to the Fc region of a human immunoglobulin (scFv-Fc format). To enhance its serum half-life, the Fc domain was engineered with specific amino acid substitutions known to improve binding affinity to the FcRn. We designed and constructed four variants (V1-V4) of the anti-WNV antibody 900, each with a different Fc composition (Figure 1a,b). The variable regions were derived from the murine mAb 9E2 (Figure 1c), which was previously generated at the State Research Center of Virology and Biotechnology "Vector" using hybridoma technology. This broadly reactive mAb 9E2 recognizes the WNV E protein in immunoblot analysis and specifically binds WNV in an enzyme-linked immunosorbent assay (ELISA) with a titer of at least 1:2,187,000. Moreover, mAb 9E2 is capable of neutralizing multiple WNV strains and protecting animals from lethal infection [21][22][23]. The constructed expression vectors were used to establish stable producer cell lines derived from CHO-K1 cells. As a result, four stable CHO-K1-900 producer cell lines were obtained, each secreting one of the four chimeric mAb 900 variants (V1-V4, Figure 1b). Each antibody variant was produced at a yield of approximately 10 mg. Purification was carried out by affinity chromatography using Protein A columns.
Recombinant FcRn produced in mammalian cells was employed to evaluate the binding affinity of the mAb 900 variants (see Section 4.2, Figure S1). Interactions between the purified antibody variants and recombinant FcRn were analyzed using biolayer interferometry (BLI), a technique that enables direct real-time measurement of protein-protein binding kinetics and affinity without the need for labeling [24,25].
Analysis of the equilibrium dissociation constants (KD) revealed distinct differences in FcRn binding affinity among the four mAb 900 variants (Table 1). Variant V4, incorporating both mutation clusters (M252/S254/T256 and H433/N434), exhibited the lowest KD, indicating the highest binding affinity to FcRn (mean KD = 1.41 × 10 -9 M). In contrast, variant V1, which lacked these mutations, displayed the highest KD, consistent with weak FcRn binding.
These findings demonstrate that the strategic introduction of mutations into both critical Fc fragment clusters of mAb 900 substantially enhanced its affinity for FcRn.
## 2.2. Cultivation of the CHO-K1-900 Cell Line and Purification of Chimeric mAb 900
Based on the FcRn binding affinity assessment, variant V4 of mAb 900, incorporating both mutation clusters (M252/S254/T256 and H433/N434) in the Fc fragment, was selected for subsequent experiments.
The selected mAb 900 variant was produced by cultivating the CHO-K1-900 cell line. The initial seeding density was 1 × 10 6 cells/mL, and cells were maintained at 37 • C. Cultivation continued until the cell density reached 6-8 × 10 6 cells/mL, after which the culture was shifted to hypothermic conditions (31 • C) for further growth. Cultivation was carried out over 14 days, with continuous monitoring of cell viability and maintenance of glucose concentration at 40 mM. Under these conditions, the strain achieved a productivity of 200 mg/L of culture medium.
mAb 900 was purified using a multistep procedure combining affinity and ionexchange chromatography (see Section 4.5). This purification strategy yielded the target protein with >95% purity, as confirmed by SDS-PAGE, Western blot, reversed-phase high-performance liquid chromatography (RP-HPLC) (Figure 2) and size-exclusion chromatography (SEC) (Figure S2). Western blot analysis under reducing conditions using an anti-human Fc-specific antibody confirmed the identity of mAb 900, revealing a predominant band at approximately 50 kDa. This corresponds to the expected molecular weight of the denatured, monomeric polypeptide chain (calculated: 50.8 kDa) and validates successful expression. An additional high-molecular-weight band (>200 kDa), also detected by both reducing SDS-PAGE and RP-HPLC (Peak 3), indicated the presence of a minor fraction of antibody aggregates.
The oligomeric state and purity of the native protein were assessed by SEC. The SEC profile showed a single major peak, corresponding to a molecular mass of 100.0 ± 0.5 kDa, which is consistent with the dimeric scFv-Fc format of mAb 900 (Supplementary Materials, Figure S2). This confirmed that the purified preparation was predominantly in the correct, monodisperse form.
Finally, the conformational stability of mAb 900 was evaluated by monitoring its thermal unfolding via circular dichroism (CD) spectroscopy. The CD thermal denaturation profile exhibited two distinct transitions at 59 • C and 77 • C, with the higher transition (77 • C) representing the complete unfolding of the protein's native structure, indicating a stable fold (Figure S3).
## 2.3. Functional Activity of mAb 900 in Enzyme-Linked Immunosorbent Assay (ELISA)
The functional activity of mAb 900 was evaluated using ELISA to determine its binding capability to antigens of inactivated WNV (strain LEIV-Vlg99-27889-human) and the recombinant D III domain of the WNV structural E glycoprotein.
Comparison of the chimeric mAb 900 with the original murine mAb 9E2 in assays using inactivated virus revealed differences in their titers (Figure 3a). To ensure an accurate comparison, titers were normalized based on protein concentration per milliliter in each sample (Table 2). Notes: ELISA titer values represent means ± SD of three independent experiments. SARS-CoV-2 RBD-specific human mAb iB20 (Titer-<1.0 lg) and murine mAb 29F10 targeting the NS1 protein of Tick-borne encephalitis virus (Titer-<1.0 lg) were employed as negative controls. Student's t-test was used for two-group comparisons.
The murine mAb 9E2, used as a reference control, has been previously characterized to bind the WNV E protein, neutralize a broad panel of WNV strains (LEIV-Vlg00-27924human, LEIV-Vlg99-27889-human, LEIV-Az67-1640-nuthatch, LEIV-Az70-72-ticks, LEIV-Tur73-2914-ticks, Ast63-94-ticks, NY99, and Egypt 101), and protect animals from lethal challenge [21][22][23].
The normalized titer of mAb 900 in the culture supernatant was 4.20 lg, whereas the titer of the purified mAb 900 was 4.61 lg. The increase in titer may be attributed to the removal of impurities present in the culture supernatant, which could have interfered with the binding of mAb 900 to the antigen, thereby affecting the final antibody titer.
The murine mAb 9E2 displayed a higher titer (4.64 lg). The slightly reduced functional activity (not statistically significant) of the chimeric antibody in ELISA compared with its murine counterpart may result from structural modifications, such as alterations in the Fc fragment, which can affect antigen binding.
ELISA using the recombinant D III domain of the WNV E glycoprotein demonstrated endpoint titers 5.8 lg, for mAb 9E2, 4.9 lg for mAb 900. These findings are consistent with the results obtained using inactivated WNV, which similarly showed a higher titer for mAb 9E2 compared with the chimeric mAb 900 (Figure 3b).
## 2.4. Virus-Neutralizing Activity of mAb 900
The results of the neutralization assay of mAb 900 against WNV (LEIV-Vlg99-27889human) are summarized in Table 3. Murine mAb 9E2, previously shown to possess neutralizing activity [21][22][23], was used as a control. The IC 50 of purified mAb 900 was 74 ng/mL, 2.5-fold lower than that of unpurified mAb 900 (185 ng/mL), indicating a substantial enhancement of the antibody's specific functional activity following removal of impurities. This improvement likely reflects the elimination of components that hinder specific antigen binding.
Comparison of the IC 50 values of the chimeric mAb 900 (74 ng/mL) and the murine mAb 9E2 (68.6 ng/mL) indicates that the chimerization had minimal impact (not statistically significant) on the antibody's neutralizing activity. This finding demonstrates that the broadly reactive murine mAb 9E2's neutralizing properties were largely preserved in mAb 900.
## 2.5. Equilibrium Dissociation Constants (KD) Estimate for mAb 900
Biolayer interferometry was used to assess the antibody's affinity to the antigen. MAb 9E2 interacts with recombinant DIII protein, of the WNV surface protein, with an affinity of ~1.51 × 10 -9 M (Figure 4a). Similarly, recombinant mAb 900 strongly binds DIII with a KD of ~2.30 × 10 -9 M (Figure 4b). From comparing the obtained values using the Dunnett test, the difference between the values was statistically not significant (p-value > 0.5). This means that the affinities of mAb 9E2 and 900 for antigen are essentially equivalent.
## 2.6. Glycosylation Profile of mAb 900
The results of the N-glycan profile analysis for mAb 900 samples, determined by hydrophilic interaction liquid chromatography (HILIC-HPLC), are provided in Supplementary Materials, Table S1 and Figure S4. The analysis was performed on two mAb 900 samples (sample 1 and sample 2) produced under identical conditions with a two-month interval.
The following types of neutral N-linked glycans were identified: G0, G0F, G0-N, G0F-N, G1, G1F, G2F, and Man5. The content of afucosylated glycans (AF, %) was approximately 7%. The content of galactosylated glycans (G, %) varied between the samples: 37.6% for sample 1 and 47.3% for sample 2. This variation is attributed to a redistribution in the relative abundance of the major glycoforms G0F and G1F. The degree of mannosylation was 0.9% and 0.5% for sample 1 and sample 2, respectively.
## 2.7. Assessment of Acute Toxicity in Mice
An acute toxicity study was performed by administering a single high dose (200 µg per mouse) of mAb 900 to healthy ICR mice (n = 6) via intraperitoneal injection. All animals survived the 14-day observation period with no signs of adverse effects, changes in body weight, or altered behavior compared to the control group. These results indicate the absence of acute toxicity for mAb 900 at the tested dose.
## 2.8. Protective Efficacy of mAb 900 In Vivo
The results assessing the protective activity of the recombinant chimeric mAb 900 against infection with the WNV strain LEIV-Vlg99-27889-human (0.3 × 10 2 LD 50 per mouse) are presented in Table 4 and Figure 5. The experiment was conducted using a BALB/c weanling mouse model, with mAb 900 administered 6 h post-infection. A comparison of the therapeutic efficacy of different mAb 900 doses revealed that the 100 µg per mouse dose provided the highest level of protection (83% survival) against the challenge dose of 30 LD 50 WNV. For the group receiving the 100 µg dose, the survival curve was statistically significantly different from the virus control group (χ 2 = 6.721, df = 1, p = 0.0095), indicating a pronounced protective effect of the antibody at this dose. Doses of 50 µg and 200 µg per mouse protected 33% of the animals. No statistically significant difference in survival compared to the virus control was observed for the groups receiving 50 µg (χ 2 0.2198, df = 1, p = 0.6392) or 200 µg (p > 0.05). Thus, a single dose of 100 µg of mAb 900 per animal is capable of exerting protective effects against WNV in vivo.
## 3. Discussion
Antibodies are central to adaptive immunity, preventing pathogen dissemination and providing long-term protection, which is critical in combating viral infections [27]. Their ability to directly suppress viral replication makes them a compelling modality for both therapeutic and prophylactic applications. Broadly neutralizing antibodies, which exhibit high potency against diverse viral strains, are of particular interest. Such antibodies are often first identified in animal models; however, their direct therapeutic use in humans is limited by immunogenicity [18,20]. Antibody humanization strategies, including chimerization, have been developed to overcome this hurdle. A paramount concern in this process is that replacing heterologous constant regions can unpredictably alter the structural integrity and affinity of the antigen-binding site [28]. Therefore, the primary objective of this study was to generate a humanized, Fc-engineered variant of the broadly neutralizing murine anti-WNV antibody 9E2 [22,23] and to perform a rigorous, head-to-head comparison of its key functional and biophysical properties.
To enhance the therapeutic profile of the candidate, we employed rational Fc engineering. We grafted the variable regions of mAb 9E2 onto a human IgG1 scaffold incorporating the M252Y/S254T/T256E and H433K/N434F mutations. These mutations are well-documented to increase affinity for the neonatal FcRn receptor, a key mechanism for extending serum half-life and improving drug exposure [29,30]. Our BLI analysis confirmed the success of this design, demonstrating high-affinity binding of the engineered Fc (variant V4) to human FcRn (KD = 1.41 × 10 -9 M). This strategic modification positions mAb 900 for improved pharmacokinetics, aligning with clinically validated approaches for other antiviral antibodies [29].
The most critical finding of this work is that the extensive molecular engineeringencompassing chimerization and targeted Fc mutagenesis-did not compromise the core antiviral function inherited from the parental antibody. Our comparative functional analysis revealed near-identical profiles between mAb 900 and murine 9E2. The chimeric antibody retained high-affinity binding to the WNV E protein DIII domain (KD = 2.30 × 10 -9 M), comparable to the progenitor (KD = 1.51 × 10 -9 M). Most importantly, its potent virusneutralizing activity in vitro remained fully intact, with virtually identical IC 50 values (74 ng/mL vs. 68.6 ng/mL). These values align with the high-affinity range (10 -9 -10 -11 M) reported for other therapeutic antibodies against flaviviruses [31][32][33][34] and confirm the structural preservation of the antigen-binding site. This successful retention of function post-humanization places mAb 900 among other successful cases where chimerization preserved key properties [18,20,28].
Beyond function, a comprehensive developability assessment underscores the robustness of mAb 900. The antibody exhibits high conformational stability, with a thermal denaturation profile showing a major unfolding transition at 77 • C, consistent with stable, well-folded human IgG1 therapeutics. SEC confirmed the preparation was predominantly monomeric (>92%), and a detailed glycosylation analysis revealed a typical profile for CHO-derived antibodies. The low levels of afucosylation (~7%) and high-mannose glycans (<1%) [35][36][37] are characteristic of non-engineered production systems and predict predictable effector function and pharmacokinetic behavior. Prior to in vivo efficacy testing, an acute toxicity study in ICR mice confirmed the absence of adverse effects, establishing an initial safety profile for the formulated product.
The most significant advance presented here is the demonstration of in vivo protective efficacy. In a lethal challenge model using a neuroinvasive WNV strain, a single, postexposure dose of mAb 900 (100 µg) conferred significant protection, with 83.3% survival. This result provides proof-of-concept for its potential as promising preclinical candidate. The dose-response relationship observed-with lower efficacy at both higher (200 µg) and lower (10, 50 µg) doses-warrants further investigation. The complete lethality at the 10 µg dose, occurring earlier than in the virus control group, raises the possibility of ADE, a phenomenon that must be carefully ruled out for flavivirus therapeutics during future development. This complex efficacy profile highlights the need for detailed future studies to optimize the dosing regimen.
## 4. Materials and Methods
## 4.1. Construction of the Integration Vector pVeal2_9E2ch
The nucleotide sequences of the 9E2 heavy (GenBank ABU63295.1) and light (GenBank ABU63296.1) chain genes were retrieved from the database and codon-optimized for CHO cells using the Gene Optimizer tool (https://www.thermofisher.com/ru/en/home/lifescience/cloning/gene-synthesis/geneart-gene-synthesis/geneoptimizer.html (accessed on 2 February 2020)). The final 9E2ch-scFv nucleotide sequence was extended with a sequence encoding signal peptide 176 and restriction sites for EcoRI and BamHI and was custom-synthesized by LLC "DNA-Synthesis" (Moscow, Russia).
Additionally, the integration vector was modified to include the constant region of a human IgG1 antibody [CH2-CH3] (Fc fragment), containing amino acid substitutions that enhance FcRn binding affinity [38]. Based on the pVEAL2 vector, four variants of the integration vector pVEAL2_9E2ch were generated (patent RU 2801532): pVEAL2_9E2ch-V1 (native, without substitutions), pVEAL2_9E2ch-V2 (substitutions in the H433/N434 region), pVEAL2_9E2ch-V3 (substitutions in the M252/S254/T256 region), and pVEAL2_9E2ch-V4 (containing both substitution clusters M252/S254/T256 and H433/N434).
For the production of these plasmid variants, Fc fragments incorporating the selected amino acid substitutions were custom-synthesized and cloned into the pVEAL2 vector at the AsuNHI and SalI restriction sites.
Recombinant constructs were designed using SnapGene 3.2.1.
## 4.2. Construction of the Integration Vector pMV-FcRn-B2M
The nucleotide sequence of FcRn-B2M, including the Gaussia luciferase (GL) signal peptide, the sequence corresponding to the heavy α chain of FcRn (α-FcRn), a 6× His tag for affinity purification (6His), a self-cleaving P2A peptide enabling separation of the β2microglobulin sequence (P2A), the β2-microglobulin sequence (B2M), and restriction sites for AsuNHI and SalI, was custom-synthesized by LLC "DNA-Synthesis" (Moscow, Russia) and cloned into the pMV vector. A vector map of the recombinant construct, designed using SnapGene 3.2.1, is presented in the Supplementary Materials (Figure S1).
## 4.3. Generation of Producer Cell Lines
Suspension cultures of CHO-K1 cells (Chinese hamster ovary, proline-auxotrophic CHO clone; FSRI SRC VB "Vector," Rospotrebnadzor, Russia) were maintained in Hy-Clone™ HyCell™ CHO medium (Cytiva, Marlborough, MA, USA) supplemented with 4 mM GlutaMAX (Gibco, Carlsbad, CA, USA) in 50 mL Falcon-type tubes. Cultures were incubated in a Heracell™ VIOS 160i CO 2 Incubator (Thermo Scientific, Waltham, MA, USA) at 37 ± 1 • C, 5% CO 2 , 80% humidity, and shaken at 185 rpm on an orbital shaker (Infors-HT, Minitron, Infors AG, Bottmingen, Switzerland) until reaching a density of 6-8 × 10 6 viable cells/mL.
Immediately prior to transfection, cell concentration and viability were assessed using a TC20 cell counter (Bio-Rad Laboratories, Hercules, CA, USA). A volume of suspension containing 5 × 10 6 viable cells was collected, centrifuged at 800 rpm for 5 min at 4 • C (1580R, Gyrozen, Gimpo, Republic of Korea), and the supernatant was removed. The cell pellet was resuspended in 1 mL of HyClone HyCell TransFx-C transfection medium (Cytiva, Marlborough, MA, USA) and transferred to a 50 mL Falcon tube.
Transfection was performed by adding 6.75 µg of plasmid DNA pVeal2_9E2ch (variants V1-V4 or pMV-FcRn-B2M) and 0.75 µg of plasmid pSB100 (Addgene, Watertown, MA, USA) to the cell suspension, followed by 10 µg of polyethyleneimine (PEI 25K™). The mixture was incubated for 4 h at 200 rpm, 37 ± 1 • C, in a humidified atmosphere with 5% CO 2 . Afterward, 4 mL of pre-warmed HyClone™ HyCell™ CHO growth medium supplemented with 4 mM GlutaMAX was added, and cells were further cultured for 48 h under the same conditions.
High-producing monoclonal clones were subsequently isolated from the polyclonal pool using limiting dilution. Selected monoclonal producer clones were frozen in a working cell bank for subsequent experiments.
## 4.4. Suspension Cultivation of Producer Cell Lines
Selected monoclonal-producing clones were thawed from the working cell bank. The initial passage was performed in a 50 mL tube with a working volume of 5 mL of HyClone™ HyCell™ CHO medium supplemented with 4 mM GlutaMAX at a cell density of 1 × 10 6 cells/mL. Cultures were maintained on a shaker at 200 rpm, 37 ± 1 • C, in a humidified atmosphere containing 5% CO 2 . Cell growth and viability were monitored every 48-72 h using a TC20 cell counter. Subsequent passages were performed when cell density reached 6-8 × 10 6 cells/mL. Once the culture demonstrated stable growth and high viability, the working volume of the suspension was gradually increased in a stepwise manner.
Cells were then cultured to the target volume at a density of 6-8 × 10 6 viable cells/mL, viability of approximately 95%, and a glucose concentration of 40 mM. Cultivation was carried out on a shaker at 200 rpm, 31 ± 1 • C, in a humidified atmosphere with 5% CO 2 for 14 days.
Glucose levels in the culture medium were continuously monitored during cultivation at 31 ± 1 • C using a Diacont glucometer (OK Biotech, Nanjing, China). The glucose concentration was maintained at 40 mM by supplementing the medium with sterile 40% glucose solution.
## 4.5. Protein Purification 4.5.1. Purification of mAb 900
Preliminary purification of the supernatant was performed by adding 10% of the volume of the liquid sorbent Ammoflok-25 (Fizlabpribor, Moscow, Russia). The resulting precipitate and cellular debris were removed by centrifugation at 12 000 rpm for 20 min at 4 • C in an Avanti J-30I high-speed centrifuge (Beckman Coulter, Brea, CA, USA). The resulting supernatant was then filtered through a 0.22 µm filtration system (Jet Biofil, Guangzhou, China). The mAb was isolated and purified by affinity chromatography on a column (Cytiva, Marlborough, MA, USA) packed with MabSelect SuRe sorbent containing recombinant protein A ligand (Cytiva, Marlborough, MA, USA) at a rate of (75 ± 5) ml of culture supernatant per 1 mL of sorbent. The clarified culture supernatant was applied to the chromatography column, pre-equilibrated with buffer A (16.2 mM Na 2 HPO 4 × 12H 2 O, 3.8 mM NaH 2 PO 4 × 2H 2 O, 150 mM NaCl, pH 7.4), at a flow rate of (2.2 ± 0.5) ml/min, using an Akta pure 150 medium-pressure chromatography system with Unicorn 7.6 software (Cytiva, Marlborough, MA, USA). The column was then washed again with buffer A until a constant optical density was achieved. The target protein was eluted with a linear gradient from 0 to 100% buffer B (10 mM Na 3 C 6 H 5 O 7 × 5.5H 2 O, pH 3.0) in a volume equal to 8 column volumes (CV), followed by washing with buffer B. During the elution process, the acidic pH of the obtained fractions was immediately neutralized to pH 7.4 by adding buffer (200 mM Na 2 H P O 4 × 12H 2 O, 300 mM NaCl, pH 10.8) under the control of an Anion-4100 pH meter (Infraspak-Analit NPP, Novosibirsk, Russia). After analysis by electrophoresis in a 15% sodium dodecyl sulfate-polyacrylamide gel under reducing conditions, fractions containing mAb 900 were pooled and dialyzed (SID 9652, Sigma, Ronkonkoma, NY, USA) against buffer (40.5 mM Na 2 HPO 4 × 12H 2 O, 9.5 mM NaH 2 PO 4 × 2H 2 O, 50 mM NaCl, pH 7.4) at 6 ± 2 • C for 20 ± 2 h. Ion-exchange chromatography was performed using Q-Sepharose (Q Fast Flow, Cytiva, Marlborough, MA, USA). The column with the sorbent was washed with base buffer (40.5 mM Na 2 HPO 4 × 12H 2 O, 9.5 mM NaH 2 PO 4 × 2H 2 O, 50 mM NaCl, pH (7.4 ± 0.05)) to a pH of (7.4 ± 0.05). The mobile phase flow rate was 5 mL/min. Then the target protein solution was applied. Then the sorbent was washed with five column volumes of base buffer. The target protein, not adsorbed on Q-Sepharose, began to exit immediately during the application period. The target protein solution exiting the column was collected in tubes in fractions of (3.5 ± 0.5) ml, starting from the optical density of the solution of at least 0.1 o.u., measured at a wavelength of 280 nm. Following SDS-PAGE analysis, fractions containing the target protein were pooled and dialyzed against buffer (40.5 mM Na 2 HPO 4 × 12H 2 O, 9.5 mM NaH 2 PO 4 × 2H 2 O, 300 mM NaCl, pH 7.4) at 6 ± 2 • C for 20 ± 2 h. After dialysis, mAb 900 was filtered through a 0.22 µm filtration system (Jet Biofil, Guangzhou, China) and stored at 6 ± 2 • C.
## 4.5.2. Purification of Neonatal Fc Receptor (FcRn)
FcRn was purified by affinity chromatography using an IMAC Sepharose column (Cytiva, Marlborough, MA, USA; 50 mL) equilibrated with buffer containing 30 mM NaH2PO4, 0.5 M NaCl, and 20 mM imidazole (pH 7.4). The target protein was eluted using a linear gradient of imidazole from 20 to 0.5 M in the same buffer. Fractions with an optical density ≥ 0.25 a.u. were analyzed by denaturing 15% SDS-PAGE. Fractions containing FcRn were pooled and dialyzed against 50 mM NaHCO3 (pH 7.6).
## 4.6. Biolayer Interferometry (BLI)
Measurement of the binding kinetics was performed on an Octet K2 instrument (ForteBio, Fremont, CA, USA) using NTA biosensors (Cat. #18-5101). Correction for baseline drift was performed by subtracting the average of the shifts recorded by the sensor loaded with the antibody not reacting with the antigens. The data were processed using the Data Analysis HT 12.0.1.55 software. The values obtained by the control sensor were read from all other results, including the highest value of the background signals. To do this, when processing the results, a reference sensor was selected, and reference wells for subtraction were also selected. Crooked competitors were the result of dissociation. The experimental model was used in a 1:1 ratio.
## 4.6.1. Interaction Between FcRn and Recombinant Antibody with Different Fc-Fragments
FcRn and the comparative protein Sag1 were prepared at 30 µg/mL, while mAb 900 variants (V1-V4) were diluted to 1, 2, and 4 µg/mL and added to microplate wells at 200 µL per well. Microplates with samples and biosensors were installed in the Octet K2. Sensors were pre-washed in phosphate-buffered saline (PBS) for 10 min with shaking at 1000 rpm. For ligand loading, the first sensor was immersed in a well containing FcRn, and the second sensor in a well containing Sag1, and incubated for 300 s at 1000 rpm. Unbound analytes were removed by washing sensors in PBS for 30 s at 1000 rpm. Association kinetics were measured by immersing both sensors in wells containing the respective antibody variant at 1 µg/mL for 180 s at 1000 rpm. Dissociation was monitored by transferring sensors to PBS for 320 s under continuous shaking at 1000 rpm. Cyclic regeneration was performed using 50 mM glycine-HCl (pH 1.5) and PBS wash buffer for three cycles, with sensors incubated for 10 s at 1000 rpm during each step. Subsequently, nickel was immobilized on the sensor surface by immersing the sensors in 10 mM nickel chloride for 30 s at 1000 rpm. The procedure was repeated with increasing antibody concentrations. For each antibody variant (V1-V4), three replicate measurements were conducted at three concentrations ranging from 12.8 to 51.3 nM. KDs were calculated for all FcRn-antibody pairs. 4.6.2. Interaction Between D III and Antibodies mAb 9E2, mAb 900
Antibodies were prepared at concentrations of 25 µg/mL (mAb 9E2) and 18.5 µg/mL (mAb 900). Recombinant DIII protein was diluted to 20 µg/mL. Recombinant human mAb IB20 (25 µg/mL) [26], which does not bind to DIII, was included as a negative control. Sample plates and biosensors were installed in the Octet K2 system. Sensors were prewashed in PBS for 10 min with shaking at 1000 rpm. For ligand loading, sensors were immersed in wells containing DIII for 300 s at 1000 rpm. Unbound DIII was removed by washing the sensors in PBS for 30 s at 1000 rpm. Association kinetics were measured by immersing sensors in wells containing mAb 9E2, mAb 900, or IB20 at 1 µg/mL and incubating for 180 s at 1000 rpm. Dissociation was monitored by transferring sensors to PBS for 320 s under continuous shaking at 1000 rpm. Kinetic analysis included a 60-s baseline in PBS. Cyclic sensor regeneration was performed as described in Section 4.6.1. Each antibody (mAb 9E2 and mAb 900) was analyzed in triplicate at five different concentrations ranging from 10.4 to 166.4 nM, with a twofold serial dilution between concentrations.
## 4.7. Reversed-Phase High-Performance Liquid Chromatography (RP-HPLC)
RP-HPLC was conducted on an LC-20 Prominence system (Shimadzu, Kyoto, Japan) equipped with a spectrophotometric detector, an autosampler, and a column thermostat. Separation was achieved using a Kromasil 300-5-C4 column (4.6 × 250 mm). Data acquisition and processing were performed using LabSolutions software, version 5.95 (Shimadzu, Kyoto, Japan).
The mobile phases employed were as follows: buffer A, 0.1% trifluoroacetic acid (TFA) in deionized water, and buffer B, 0.1% TFA in 70% isopropanol/water. RP-HPLC analysis was carried out under a gradient elution program.
## 4.8. ELISA
ELISA was performed using a standard protocol with sequential twofold serial dilutions of antibodies, each tested in triplicate. Ninety-six-well Nunc MaxiSorp polystyrene plates (Thermo Scientific, Waltham, MA, USA) were coated either with a recombinant DIII domain of the WNV envelope glycoprotein at 150 ng/well in 10 mM carbonate buffer (pH 9.5) at 4 • C for 20 h, or with WNV, LEIV-Vlg99-27889-human at 200 ng/well. Nonspecific binding sites were blocked with 0.2% casein in phosphate-buffered saline (PBS, pH 7.4) for 1 h at 37 • C. Test antibodies were serially diluted in PBS to concentrations from 500 µg/mL to 0.020 µg/mL. Murine mAb 29F10 targeting the NS1 protein of Tick-borne encephalitis virus (TBEV, I-1022, Biosan, Novosibirsk, Russia) and recombinant human antibody IB20 were employed as negative controls. Chromogenic detection was carried out using rabbit anti-mouse and anti-human immunoglobulin antibodies conjugated to horseradish peroxidase (anti-Mouse-HRP, anti-Human-HRP, Imtek, Moscow, Russia) at concentrations recommended by the manufacturer. Tetramethylbenzidine (TMB) substrate (Tetramethylbenzidine Liquid Substrate System for ELISA, Sigma-Aldrich, Burlington, MA, USA) was used for color development, and the reaction was terminated by adding 0.5 M sulfuric acid. Optical density (OD) was measured at 450 nm using a Thermo Scientific Varioskan LUX spectrophotometer (Thermo Scientific, Waltham, MA, USA).
The endpoint dilution was defined as the highest dilution yielding an OD greater than the critical value (OD_crit), calculated as follows: OD_crit = mean OD of the sample at 1:100 dilution + 0.15.
## 4.9. Study of Acute Toxicity of mAb 900 Following Intraperitoneal Administration in Mice
The study used a mAb 900 at a concentration of 1.1 mg of protein/mL. The control group received sterile 0.9% sodium chloride solution.
The study was conducted on 40 outbred ICR mice (20 males and 20 females) weighing 18-20 g. The animals were housed under standard vivarium conditions (temperature 20-24 • C, humidity 45-65%, with ad libitum access to water and standard pelleted diet) in accordance with GOST 33216-2014. A 48-h acclimatization period was observed prior to the experiment. Using a random sampling method, mice were balanced by body weight and divided into two groups of 20 animals each (10 males + 10 females). Experimental group, single intraperitoneal (i.p.) administration of mAb 900 at a dose of 550 µg per animal (injection volume 0.5 mL). Control group, single i.p. administration of an equivalent volume (0.5 mL) of physiological saline.
The selected dose corresponded to one proposed human dose, recalculated for mice based on body surface area.
Animals were observed for 7 days post-injection. General condition, behavioral responses, feed and water consumption, and body weight were recorded daily.
Laboratory and instrumental investigations on days 1 and 7 included hematological analysis, post-mortem (pathological) examination, and histological examination.
Blood for hematological analysis was collected from the retro-orbital sinus. Standard parameters (white blood cells, red blood cells, platelets, hemoglobin, hematocrit) were determined using an automatic hematology analyzer MicroCC-20 Plus VET (High Technology, North Attleboro, MA, USA), with the leukocyte formula counted manually.
Following euthanasia, a necropsy was performed. Internal organs were assessed macroscopically, weighed, and their relative organ weight indices were calculated (organ weight, mg/body weight, g × 10).
For histological examination, samples from 14 key organs and tissues were collected: brain, thymus, lungs, heart, liver, spleen, kidneys, adrenal glands, gastrointestinal tract, bone marrow, reproductive organs, injection site, and regional lymph nodes. The material was fixed in 10% neutral buffered formalin and processed according to a standard histological protocol: paraffin embedding, preparation of 4-5 µm thick sections, and staining with hematoxylin and eosin. Evaluation was performed by light microscopy using an Axio Imager Z1 microscope (Carl Zeiss, Göttingen, Germany), with particular attention paid to signs of inflammation, degeneration, dystrophy, and microcirculation disorders.
The obtained data were processed using the Statgraphics 5.0 software package (Statgraphics Technologies, Inc., The Plains, VA, USA). The Shapiro-Wilk test was applied to check for normal distribution. Non-parametric methods were used to assess intergroup differences: the Kruskal-Wallis H-test followed by pairwise comparison using the Mann-Whitney U-test. Differences were considered statistically significant at p < 0.05.
## 4.10. Viruses, Cell Cultures and Neutralization Assay
The study experiments used the WNV strain WNV, LEIV-Vlg99-27889-human, obtained from the State Collection of Pathogenic Microorganisms and Rickettsiae of the State Research Center of Virology and Biotechnology «Vector». The virus was cultured on Vero monolayer cell cultures, grown to 80-90% confluency in DMEM F12 medium (Gibco, USA) containing 10% fetal bovine serum (Gibco, Carlsbad, CA, USA), penicillin 100 IU/mL, and streptomycin 100 µg/mL (Gibco, Carlsbad, CA, USA) in an atmosphere with 5% CO 2 at 37 • C. The infectious activity of the viral substances was determined by microtitration on 96-well culture plates (Greiner AG, Kremsmünster, Austria) with a subconfluent monolayer of cells, as previously described [39]. The calculation of viral infectious titers was performed using the Spearman-Karber method and expressed as log10 TCID 50 (50% tissue culture infectious dose). The viral substances were stored at a temperature of -80 • C WNV neutralization was performed using a microneutralization assay on Vero cell cultures (96-well culture plates, 5 × 10 4 cells/well). Both the virus and antibodies were diluted in DMEM F12 medium. A mixture of virus (100 µL) and threefold serial dilutions of antibodies (100 µL) was incubated for two hours at room temperature, after which the mixtures were used to infect Vero cell monolayers (4 repetitions per experimental point infected with 100 TCID 50 WNV). Results were assessed on day 7 post-infection by microscopic scoring of the cytopathic effects (CPE) and by measuring cell viability in the formazan-based MTT assay described previously [39]. Antibody titers were defined as the highest dilution that protected 50% of the cell monolayers from virus-induced CPE. SARS-CoV-2 RBD-specific human mAb iB20 was used as negative control.
## 4.11. Determination of the Thermal Stability of mAb 900
The study was conducted using CD spectroscopy on a Chirascan VX-100 spectropolarimeter (Applied Photophysics, Leatherhead, UK). Measurements were performed in a cuvette with a 1 mm optical path length. A 200 µL sample of mAb 900 was placed in the cuvette. Spectra were recorded in the temperature range of 5 to 100 • C with a 5 • C step. The temperature was verified using a thermocouple placed directly in the cuvette. The wavelength range was 200-280 nm with a 1 nm step, a signal integration time of 0.5 s per wavelength, and a slit width of 1 nm.
The baseline was constructed using the buffer in which mAb 900 was dissolved (40.5 mM Na 2 HPO 4 •12H 2 O, 9.5 mM NaH 2 PO 4 •2H 2 O, 300 mM NaCl, pH 7.4).
The acquired data were analyzed using specialized Global Thermal Analysis 3 software (Applied Photophysics, Leatherhead, UK), which enables the simultaneous analysis of a set of CD spectra obtained at different temperatures. The theoretical model considered was that of intramolecular protein denaturation. Based on the spectral data, the melting temperature (Tm) and the enthalpy of denaturation (∆H) for mAb 900 were determined.
## 4.12. Size-Exclusion Chromatography (SEC)
Native-state SEC was performed by gel filtration using a 10 mL Sephacryl S-200 HR column on an Acta Pure 150 chromatography system (Cytiva, Marlborough, MA, USA). The separation was carried out in PBS, pH 7.4, at a flow rate of 0.5 mL/min. The column temperature was maintained at 2-8 • C. The column was calibrated with proteins of known molecular masses: lysozyme (14.3 kDa), β-lactoglobulin (18.4 kDa), egg albumin (45 kDa), bovine serum albumin (66 kDa), and γ-globulin (150 kDa). A calibration curve was constructed from these data and used to determine the molecular weight of the target protein.
## 4.13. Glycosylation Profile Analysis of mAb 900
The glycosylation profile was analyzed for two mAb 900 samples (sample 1 and sample 2) produced under identical conditions with a two-month interval. Samples 1 and 2 were subjected to enzymatic release of N-glycans from the protein backbone using PNGase F (New England Biolabs, Ipswich, MA, USA). The released glycans were labeled with the fluorescent tag 2-aminobenzamide (2-AB). The analysis was performed by HILIC-HPLC with fluorometric detection in gradient mode on an UltiMate 3000 system (Thermo Scientific, Sunnyvale, CA, USA).
All glycan peaks were integrated on the chromatograms, identifying eight major peaks corresponding to G0-N, G0F-N, G0, G0F, Man5, G1, G1F, and G2F glycans. The G1 and G1F glycans represent a set of two isomers each, eluting as two adjacent peaks.
Using Chromeleon 7.2 CDS software, the relative content of each glycan (Bi, %) was calculated. Subsequently, the percentage of afucosylated glycans (AF, %) and galactosylated glycans (G, %) were determined using the following formulas: AF = BG0-N + BG0 + BMan5 + BG1,
where BG0-N, BG0, BMan5, BG1 is the relative content of the corresponding glycan, %.
where BG1, BG1F, BG2F is the relative content of the corresponding glycan, %.
$$G = BG1 + BG1F + 2 × BG2F,(2)$$
## 4.14. Assessment of the Protective Efficacy of mAb 900 In Vivo
The protective activity of mAb 900 was evaluated in female BALB/c mice (weight 8-11 g), with six mice per group. The mice were obtained from the breeding facilities of laboratory animals of the State Research Center of Virology and Biotechnology "Vector". All animal maintenance procedures complied with the European Directive 2010/63/EU on the protection of animals used for scientific purposes. For infection, WNV strain LEIV-Vlg99-27889-human was used. Mice were infected via intraperitoneal (i.p.) injection with a dose of 0.3 × 10 6 LD 50 per mouse in a volume of 300 µL. mAb 900 was administered i.p. at 6 h post-infection in a volume of 300 µL per mouse. The following therapeutic doses of mAb 900 were tested for protective efficacy: 10 µg, 50 µg, 100 µg, and 200 µg per animal.
A virus control group was established by infecting mice with the selected dose of WNV without subsequent mAb 900 therapy. A safety control for the antibody was performed by administering mAb 900 at a dose of 200 µg per mouse to uninfected animals.
## References
1. Hruškovicová, Bhide, Petroušková et al. "Engineering the Single Domain Antibodies Targeting Receptor Binding Motifs Within the Domain III of West Nile Virus Envelope Glycoprotein"
2. Gusev, Romanovskaya, Galkina et al. (2024) "Sensitivity Characteristics of Genotype 2 West Nile Virus Strains to Anti-Viral Drugs" *Antibiot. Chemother*
3. Kouroupis, Charisi, Pyrpasopoulou (2023) "The Ongoing Epidemic of West Nile Virus in Greece: The Contribution of Biological Vectors and Reservoirs and the Importance of Climate and Socioeconomic Factors Revisited" *Trop. Med. Infect. Dis*
4. Martins, Ricardo, Santos et al. (2535) "Potential Novel Antiviral Biologics Drugs Currently at Discovery and Preclinical Development Stages" *Pharmaceutics*
5. Mencattelli, Ndione, Rosà et al. "Epidemiology of West Nile virus in Africa: An underestimated threat" *PLoS Negl*
6. Putintseva, Udovichenko, Nikitin et al. (2024) "West Nile Fever: Analysis of the Epidemiological Situation in the Russian Federation in 2023, Forecast for 2024" *Probl. Osob. Opasnykh Infektsii*
7. Jani, Kakoullis, Abdallah et al. (2022) "West Nile virus: Another emerging arboviral risk for travelers?" *Curr. Infect. Dis. Rep*
8. Yang, Luo, Fan et al. "Development and evaluation of neutralizing antibodies for cross-protection against West Nile virus and Japanese encephalitis virus" *Infect. Med*
9. Mertinková, Mochnáčová, Bhide et al. "Development of peptides targeting receptor binding site of the envelope glycoprotein to contain the West Nile virus infection" *Sci. Rep*
10. Mustafa, Mohammed (2024) "Developing recombinant antibodies by phage display technology to neutralize viral infectious diseases" *SLAS Discov*
11. Babalola, Akinsuyi, Folajimi et al. (2023) "Exploring the future of SARS-CoV-2 treatment after the first two years of the pandemic: A comparative study of alternative therapeutics" *Biomed. Pharmacother*
12. Ormundo, Barreto, Tsuruta (2023) "Development of Therapeutic Monoclonal Antibodies for Emerging Arbovirus Infections" *Viruses*
13. Gonzales, Bergamasco, Cristarella et al. (2024) "Effectiveness and Safety of Palivizumab for the Prevention of Serious Lower Respiratory Tract Infection Caused by Respiratory Syncytial Virus: A Systematic Review" *Am. J. Perinatol*
14. Mori, Yoshizaki, Watabe et al. "Safety, efficacy and pharmacokinetics of palivizumab in off-label neonates, infants, and young children at risk for serious respiratory syncytial virus infection: A multicenter phase II clinical trial" *Lancet Reg. Health West*
15. Sinigaglia, Peta, Riccetti et al. (2020) "New avenues for therapeutic discovery against West Nile virus" *Expert Opin. Drug Discov*
16. Pirkalkhoran, Grabowska, Kashkoli et al. (2023) *Bioengineering of Antibody Fragments: Challenges and Opportunities. Bioengineering*
17. Gordon, Raybould, Wong et al. (2024) "Prospects for the computational humanization of antibodies and nanobodies" *Front. Immunol*
18. Hsieh, Liao, Chuang et al. (2022) "A universal in silico V(D)J recombination strategy for developing humanized monoclonal antibodies" *J. Nanobiotechnol*
19. Wang, Chen, Xu et al. (2024) "Comparison of 'framework Shuffling' and 'CDR Grafting' in humanization of a PD-1 murine antibody" *Front. Immunol*
20. Yang, Wu, Shang et al. "Molecular mechanism and structure-guided humanization of a broadly neutralizing antibody against SFTSV"
21. Kazachinskaja, Alekseenko, Razumov et al. (2005) "Strain of Hybrid Cells of Animal Mus musculus L, 9E2, Used in Preparing Monoclonal Antibody to West Nile Virus Protein E of Strain WNV/LEIV-VIG99-27889"
22. Pereboev, Borisevich, Tsuladze et al. (2008) "Genetically delivered antibody protects against West Nile virus" *Antiviral Res*
23. Razumov, Kazachinskaia, Ternovoi et al. (2005) "Neutralizing Monoclonal Antibodies Against Russian Strain of the West Nile Virus" *Viral Immunol*
24. Bates, Gurmessa, Weinstein et al. (2025) "Biolayer interferometry for measuring the kinetics of protein-protein interactions and nanobody binding" *Nat. Protoc*
25. Jug, Bratkovič, Ilaš (2024) "Biolayer interferometry and its applications in drug discovery and development" *Trends Anal. Chem*
26. Kulemzin, Sergeeva, Baranov et al. "VH3-53/66-Class RBD-Specific Human Monoclonal Antibody iB20 Displays Cross-Neutralizing Activity against Emerging SARS-CoV-2 Lineages" *J. Pers. Med. 2022*
27. Liu, Oldham, Teal et al. "Fc-engineering for modulated effector functions-Improving antibodies for cancer treatment" *Antibodies*
28. Kalinichenko, Ramadan, Kruglova et al. (2024) "A New Chimeric Antibody against the HIV-1 Fusion Inhibitory Peptide MT-C34 with a High Affinity and Fc-Mediated Cellular Cytotoxicity" *Biology*
29. Pantaleo, Correia, Fenwick et al. (2022) "Antibodies to combat viral infections: Development strategies and progress" *Nat. Rev. Drug Discov*
30. Roopenian, Christianson, Sproule et al. (2003) "The MHC Class I-Like IgG Receptor Controls Perinatal IgG Transport, IgG Homeostasis, and Fate of IgG-Fc-Coupled Drugs" *J. Immunol*
31. Throsby, Geuijen, Goudsmit et al. (2006) "Isolation and characterization of human monoclonal antibodies from individuals infected with West Nile Virus" *J. Virol*
32. Lai, Goncalvez, Men et al. (2007) "Epitope determinants of a chimpanzee dengue virus type 4 (DENV-4)-neutralizing antibody and protection against DENV-4 challenge in mice and rhesus monkeys by passively transferred humanized antibody" *J. Virol*
33. Goncalvez, Chien, Tubthong et al. (2008) "Humanized monoclonal antibodies derived from chimpanzee Fabs protect against Japanese encephalitis virus in vitro and in vivo" *J. Virol*
34. Zhao, Fernandez, Dowd et al. (2016) "Structural Basis of Zika Virus-Specific Antibody Protection" *Cell*
35. Liu (2015) "Antibody glycosylation and its impact on the pharmacokinetics and pharmacodynamics of monoclonal antibodies and Fc-fusion proteins" *J. Pharm. Sci*
36. Liu, Nowak, Andrien et al. (2017) "Impact of IgG fc-oligosaccharides on recombinant monoclonal antibody structure, stability, safety, and efficacy" *Biotechnol. Prog*
37. Reusch, Tejada (2015) "Fc glycans of therapeutic antibodies as critical quality attributes" *Glycobiology*
38. Nesmeyanova, Shanshin, Isaeva et al. (2023) "pVEAL2-9E2ch-scFv Plasmid Genetic Construct, Strain of Recombinant Cell Line CHO-K1-9E2ch and Chimeric Single-Chain Antibody 9E2ch Against West Nile Virus Produced by the Specified Strain of Cell Line CHO-K1-9E2ch, with High Affinity for Neonatal FcRn Receptor" *Patent RU*
39. Svyatchenko, Ternovoi, Lutkovskiy et al. "Human Adenovirus and Influenza A Virus Exacerbate SARS-CoV-2 Infection in Animal Models. Microorganisms 2023"
40. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12698041&blobtype=pdf | # Protecting pregnancy during pandemics: What recent outbreaks teach us
Annette Regan, Flor Muñoz
## Abstract
Public health emergences often place pregnant people at a greater risk of severe disease. In this commentary, we highlight some of the insights from recent public health emergencies, spanning the 2009 influenza A/H1N1 pandemic to the recent mpox outbreaks. These epidemics have resulted in a large amount of perinatal pharmacovigilance expertise and capacity, improving the ability to conduct post-authorization evaluation of the safety of vaccines and other therapeutics during pregnancy. Although pregnant people have not historically been included in the clinical development of vaccines and other therapeutics intended for the general population, experiences from more recent public health emergencies show this trend is changing, with ongoing clinical trials either including or specifically targeting pregnant people. Pandemics and other public health emergencies that strain the healthcare system can disrupt routine prenatal care, birthing, and post-delivery care practices and this should be considered as part of pandemic planning. Finally, experiences from several epidemics show that vaccination during pregnancy is achievable, but vaccination rates during pregnancy are often suboptimal. A better understanding of vaccine hesitancy and acceptance during pregnancy and evidence-based strategies to address such hesitancy remain needed to better protect pregnant people against emerging infectious disease threats.
## Background
Pregnancy is a unique health state with transient physiological and immunological adaptations to allow a balance of maternal health and fetal tolerance. 1 While these adaptations support a healthy pregnancy, they may also make pregnant people more susceptible to severe infections. 2 Changes in the immune, cardiac, and respiratory systems are likely reasons that pregnant people experience more severe illness with viral infections in general, and respiratory viruses particularly. Some infections can also cause congenital disease, via intrauterine or peripartum infection of the fetus and neonate. Infection during pregnancy can therefore have serious consequences for both mother and fetus, leading to pregnancy loss, stillbirth, preterm birth, congenital infection and other complications of the newborn.
Severe infections in pregnancy and clinical sequelae in newborns may be especially observable during epidemics and pandemics. Historically, mortality rates have been elevated in pregnant individuals during respiratory virus pandemics. 3,4 In this commentary, we present an overview of the impacts of recent public health emergencies and pandemics on pregnancy health and a summary of lessons learned from these experiences (Figure 1) and recommendations to promote pregnancy health during future public health emergencies (Table 1).
## Influenza A/H1N1 pandemic (2009-10)
In April 2009, a novel influenza A virus was first identified in California. 5 Around the same time, severe cases of the novel influenza A/H1N1 virus in pregnant people were first reported. 6 By June, the World Health Organization (WHO) declared influenza A/H1N1 a pandemic, prompting rapid development of a monovalent 2009 influenza A/H1N1pdm09 vaccine. Clinical trials -initially in healthy adults -were
## Vaccine policy
• Immunization guidelines must consider recommendations for pregnant people that consider the burden of disease in pregnancy and existing data on the safety and efficacy/effectiveness of available vaccines Vaccine and/or treatment confidence
• Pregnant population must be willing to take a vaccine or treatment • Health care personnel willing to encourage vaccination or treatment
## Perinatal pharmacovigilance
• Surveillance systems that can capture a sufficient sample of pregnancies • Standardized case definitions for maternal, fetal and infant health outcomes • Mechanisms to share data across geo-political borders • Study designs that address sources of bias • Support for pharmacovigilance surveillance in LMICs quickly extended to pregnant people and children, leading to vaccine licensure and mass vaccination campaigns by October. 7 As per existing influenza pandemic preparedness plans, 8 pregnant people were prioritized for immunization due to elevated disease risks, with hospitalization rates four times higher in pregnancy than in the general population. 9 This marked one of the first contemporary global efforts to include pregnant people in mass vaccination strategies. Before the H1N1 pandemic, seasonal influenza vaccine uptake among pregnant people was under 15%. 9 Despite improvement, pandemic vaccine coverage remained low-under 50% in most countries, [10][11][12] and below 10% in some. 13,14 Following on from these experiences, there has been a growing interest in identifying factors contributing to prenatal vaccine hesitancy and barriers to maternal immunization, [15][16][17] which is a priority for future pandemic planning. 9 The H1N1 pandemic catalyzed major advances in maternal vaccine safety research. New pharmacovigilance platforms and large database studies were launched to evaluate the safety and effectiveness of prenatal vaccination and antiviral medication use during pregnancy. Technical expertise in the conduct of vaccine safety studies in pregnancy was also greatly enhanced, including guidance for real world studies of maternal vaccine safety 18,19 and standardized protocols for vaccine trials commissioned by the National Institute of Health (NIH). 20,21 This infrastructure laid the groundwork for future maternal immunization programs, including pertussis vaccination programs beginning in 2011-2012. 22
## Ebola virus epidemic (2014-16, 2018-19, 2022)
The 2014-2016 Ebola outbreak was the largest on record, with over 28,600 cases and at least 11,325 deaths. 23 Subsequent outbreaks occurred in the Democratic Republic of Congo (2018-19) and Uganda (2022). More than 5000 cases during the West Africa outbreak involved reproductive-aged women. 24 While pregnancy does not appear to increase susceptibility to Ebola virus disease (EVD), 24 it is associated with significantly worse outcomes. Mortality rates in pregnant people range from 50-84%, 25 with fetal loss occurring in the vast majority of cases. Among 59 confirmed or suspected EVD cases in pregnancy described in the literature, only 12 resulted in live birth, all of whom died within 19 d. 24 Data on nonfatal outcomes and disease progression in pregnancy remain limited due to gaps in surveillance and reporting. 24 These data gaps are likely due to limited surveillance systems recording pregnancy status and limited surveillance capacity to monitor outcomes in pregnant women and their newborns. 26 A live-attenuated recombinant vesicular stomatitis virus (rVSV) vaccine, rVSVΔG-ZEBOV-GP (Merck) was licensed after showing efficacy when used in ring vaccination strategies. 27 Although pregnancy was an exclusion criteria for the pre-licensure clinical trials in the first outbreak in 2014, follow-up of 84 pregnant people inadvertently vaccinated and those who became pregnant after enrollment have shown no increase in pregnancy loss or congenital anomalies. 28 Given the benefits of vaccination, an outcry of the general population and health care providers condemning the exclusion of pregnant people from life-saving vaccination, local health authorities with support from the WHO successfully included pregnant people in vaccine studies during the Ebola outbreak in 2018. 29,30 A two-dose regimen using Ad26.ZEBOZ and MVA-BN-Filo vaccines was also studies in 1221 pregnant people, showing no increase in adverse maternal or neonatal outcomes. 31 These experiences underscore the risks of long-standing cultural practices to "protect" pregnant people through exclusion, which can lead to off-label use without adequate safety data -potentially exposing pregnant people to more harm than inclusion in pre-licensure clinical trials. 32 Beyond direct health impacts, EVD outbreaks severely disrupt access to routine medical care, including prenatal services. 33 Fear, mistrust, and strained health systems contribute to reduced antenatal visits and facility-based deliveries, worsening maternal and fetal outcomes. 34
## Zika virus epidemic (2015-16)
In March 2015, Brazil reported a widespread exanthematic illness later identified as Zika virus infection. 35 As evidence mounted linking maternal Zika infection to congenital microcephaly and other birth defects, the WHO declared a Public Health Emergency of International Concern on February 1, 2016. Zika virus was found in fetal brain and placental tissue, confirming its ability to cause congenital infection, 36 making Zika the first viral infection since rubella to be definitively associated with severe congenital outcomes, leading to developmental delays and shortened lifespan in affected infants. 37,38 This recognition resulted in the development of algorithms for the diagnosis and management of congenital infection and long-term follow-up of affected infants, bridging new partnerships between maternal and neonatal healthcare professionals. To date, 86 countries and territories have reported mosquito-borne Zika transmission, and the long-term consequences of congenital infection are notable in affected infants.
Surveillance systems were rapidly established to monitor pregnancy outcomes, including the U.S. Zika Pregnancy Registry in the U.S., 39 SINAN (Sistema de Informação) and RESP (Rede de Vigilância do Zika e Doenças Associadas) in Brazil, 40 and Sivigila (Sistema de Vigilancia en Salud Pública) in Colombia. 41 The Brighton Collaboration additionally expanded their case definitions to include microcephaly and neurodevelopmental assessments. 42 Zika's discovery as a sexually transmissible virus -persisting in semen for weeks -prompted new prevention strategies, including the Zika Contraceptive Access Network (Z-CAN) in Puerto Rico. 35 These insights have informed approaches to other arboviruses associated with fetal and neonatal infection like Oropouche and Chikungunya virus. 43,44 Despite intense vaccine development efforts, no licensed Zika vaccine exists. Promising vaccine candidates have shown safety and immunogenicity in animal models, but declining transmission since 2017 has hindered large-scale efficacy trials, leaving the pathway to licensure uncertain. 45
## Chikungunya (2018-present)
Chikungunya was designated a priority pathogen by the WHO and CEPI in 2018 due to its epidemic potential and expanding geographic reach. 46 Although pregnant people typically experience symptoms similar to non-pregnant adults and do not show increased rates of pregnancy complications, 47 maternal infection near delivery poses serious risks to neonates -including sepsis, meningoencephalitis, and death. 48 To prevent these outcomes, pregnant people are advised to avoid travel to outbreak regions.
The WHO's R&D Blueprint and CEPI's vaccine development initiatives have accelerated research, 46,49 leading to the approval of the first chikungunya vaccine (Valneva's live-attenuated vaccine, IXCHIQ®) in 2023 and a second vaccine in 2025 (Bavarian Nordic's vaccine-like particle vaccine VIMKUNYA®). 50 However, data to support the safety of either vaccine during pregnancy remain limited. Clinical trials excluded pregnant participants, and safety data are limited to inadvertent exposures and animal studies, which have shown mixed results -some reporting reduced postnatal survival in rabbit models. 48 Current guidelines recommend deferring vaccination during pregnancy unless the risk of exposure is high and unavoidable. In such cases, shared decision-making is advised, with a preference for the virus-like particular vaccine over the live-attenuated vaccine. 51 Encouragingly, clinical trials involving pregnant people are underway, and more robust safety and efficacy data are expected in the coming years. 52
## COVID-19 (2020-23)
The COVID-19 pandemic, caused by SARS-CoV-2, emerged in late 2019 and rapidly spread globally. The first mRNA vaccine was introduced in December 2020, following trials in non-pregnant adults. Although pregnant individuals were initially excluded from clinical trials, they were not barred from early vaccination, especially among healthcare workers. Decisions to vaccinated during pregnancy were based on developmental and reproductive toxicology (DART) studies and safety data from non-pregnant populations, along with potential risk of the disease for the mother, fetus and newborn.
By mid-2021, accumulating evidence identified pregnancy as a risk factor for severe COVID-19, prompting formal vaccine recommendations for pregnant people in countries like the U.S. and Canada. By the end of 2022, 120 countries recommended COVID-19 vaccination during pregnancy, with 64 permitting it.
Between 2021 and 2022, real-world vaccine safety studies expanded rapidly. U.S. initiatives like V-SAFE and MOMI-vax confirmed vaccine safety and immunogenicity in pregnancy, showing infant protection for up to six months postpartum. 53 Globally, the Global Vaccine Data Network (GVDN) and national pregnancy registries were established to monitor post-vaccination outcomes following maternal vaccination. The WHO continues to recommend booster doses for pregnant people to protect both mother and infant. This recommendation is likely to evolve as SARS-CoV2 becomes endemic and the risk posed by the disease could change.
Despite strong evidence supporting vaccine safety, misinformation -particularly around fertility and fetal death -spread widely via social media, contributing to vaccine hesitancy. 54 Studies have debunked these claims, but their persistence underscores the need for better public health communication.
The COVID-19 pandemic also disrupted prenatal care: home births rose by over 50% in the U.S. between 2019 and 2021, 55 and routine visits, education classes, and support during delivery were often limited. 56 Pregnant individuals faced increased mental health challenges, including anxiety and depression, with potential long-term effects on child development. 57 Pregnant people remain at elevated risk for severe COVID-19, yet clinical trials for treatments largely excluded them. While small observational studies have been able to somewhat address this gap, 58 the COVID-19 pandemic highlighted the urgent need to include pregnant and lactating people in vaccine and therapeutic research. A COVAX-CEPI-supported checklist now guides the evaluation of pandemic vaccines for use in pregnancy. 59 Further, maternal vaccination remains the most direct method to protect infants too young to be vaccinated with COVID-19 vaccines who remain at high risk for severe disease and hospitalization among the pediatric population. 60
## Mpox (2022, 2024)
A global outbreak of clade II mpox began in May 2022, primarily affecting men who have sex with men (MSM) in non-endemic countries. The WHO declared this a Public Health Emergency of International Concern in July 2022, with cases peaking that summer. Over 102,000 cases of clade II mpox have been reported globally since January 2022. 61 The Public Health Emergency of International Concern status was lifted in May 2023 following a decline in cases.
In August 2024, WHO declared a second Public Health Emergency of International Concern due to the emergency of clade Ib mpox in the Democratic Republic of the Congo, which spread across central and western Africa. 61 Mpox infection during pregnancy is associated with increased risk of severe disease, miscarriage, stillbirth, and congenital infection. Vertical transmission has been observed in primate studies, and postnatal transmission can occur via skin-to-skin contact.
WHO recommends two smallpox vaccines for mpox prevention, including a live, replicating vaccine (ACAM2000) and a live, nonreplicating vaccine (MVA-BN). Only the live, nonreplicating vaccine is recommended for pregnant and breastfeeding people at risk of disease.
Tecovirimat is approved for severe mpox treatment, but human data on its use in pregnancy are limited. Small case series suggest no adverse outcomes, 62 and ongoing trials have not excluded pregnant participants. However, without targeted recruitment, these studies may not yield conclusive safety data.
A Phase 3 open-label trial to assess the safety and immunogenicity of MVA-BN in pregnant and postpartum people is currently underway in the Democratic Republic of Congo, 63 reflecting growing attention to maternal health in outbreak research. 64 Real-word data will be essential for shaping public health policy, though collection remains challenging in low-resource settings.
## Lessons learned
Protection of pregnant people and their newborns during public health emergencies caused by emerging infectious diseases will require a comprehensive understanding of the risks and the progression of disease in mothers and infants, evidence on the maternal and fetal health effects of available vaccine(s) and medication(s), and evidence-based, community-engaged efforts to increase the coverage of vaccines and treatments during pregnancy -all of this while ensuring continued access to quality prenatal care and maternal health services.
To achieve this, robust surveillance systems that consider pregnancy status will be needed, and these systems should ideally be in place prior to the beginning of a public health emergency. To best inform pandemic response, these surveillance systems should be in place across low, middle, and high-income country settings.
Disruptions to routine medical services can be especially harmful to vulnerable patient groups, including pregnant women. As part of pandemic preparedness, health systems must prepare for pregnancy management during public health emergencies. Telehealth and virtual care can help to reduce in-person contact while maintaining recommended visit schedules. 65 Home monitoring, where possible, can additionally reduce inperson contact. Group prenatal care could help to reduce social isolation. To support birthing, healthcare systems should plan for surge capacity and consider non-hospital births. 65 Clear and adaptable visitation policies and expedited discharge could also help reduce health system burden. Investment in doulas and midwives who can provide in-home or virtual care can help to bridge health care personnel shortages and care gaps. 65 Finally, mental health support should be provided for pregnant and postpartum women to address perinatal anxiety and depression.
In addition, prevention and treatment options should be made available to pregnant people. Although pregnant people have historically been excluded from the clinical development of vaccines intended for the general population, it is heartening that this exclusion has not been made for more recent studies. Future vaccine and therapeutic trials should consider how to include pregnant people as early as possible in the clinical development course and generate evidence to support medical decision-making and early interventions during pregnancy.
In the absence of pre-licensure clinical trial data, well-designed observational studies become essential for evaluating the safety and effectiveness of vaccines and therapeutics in pregnancy. These studies can provide valuable real-world insights by analyzing outcomes in diverse populations, settings, and over extended periods of time. 66 When conducted rigorously -with careful attention to confounding factors, bias, and data quality -observational research can serve as a critical bridge in evidence generation, informing clinical practice until more robust trial data become available.
## Conclusions
Public health emergencies associated with the emergence of respiratory and other systemic viruses that affect pregnant people and their fetuses disproportionately compared to non-pregnant populations will continue to emerge, and consideration for pregnant people and their newborns should be made when preparing for and managing these threats. The recent emergence and reemergence of infectious diseases that carry adverse consequences during pregnancy -including dengue virus and the newly recognized Oroupouche virus disease, which appears to have harmful effects on pregnancy and fetal health -serve as a constant reminder of the need to consider infectious disease surveillance for pregnancy -and inclusion of pregnant and lactating people as early as feasible in the pre-and post-authorization evaluation of vaccines and therapeutics.
## Disclosure statement
AKR reports unrelated research funding to her institution from Merck Sharp & Dohme and membership on a Data Safety Monitoring Board for Moderna. FMM is an investigator with Pfizer and Gilead Sciences; serves on the Data Safety Monitoring Board (DSMB) for Pfizer, Moderna, Meissa Vaccines, Virometix; and reports advisory board participation for Sanofi, GSK, AztraZeneca and Merck.
## References
1. Mor, Cardenas (2010) "The immune system in pregnancy: a unique complexity" *Am J Rep Immunol*
2. Abu-Raya, Michalski, Sadarangani et al. (2020) "Maternal immunological adaptation during normal pregnancy" *Front Immunol*
3. Harris (1919) "Influenza occurring in pregnant women: a statistical study of thirteen hundred and fifty cases" *J Am Med Assoc*
4. Freeman, Barno (1959) "Deaths from Asian influenza associated with pregnancy" *Am J Obstet Gynecol*
5. Cdc (2009) "Pandemic timeline"
6. Fonseca, Davis, Wing et al. (2009) "Novel influenza A (H1N1) virus infections in three pregnant women -United States" *MMWR Morb Mortal Wkly Rep*
7. Jackson, Patel, Swamy et al. (2011) "Immunogenicity of an inactivated monovalent 2009 H1N1 influenza vaccine in pregnant women" *J Infect Dis*
8. Iskander, Strikas, Gensheimer et al. (1978) "Pandemic influenza planning, United States" *Emerg Infect Dis*
9. Rasmussen, Kissin, Yeung et al. (2011) "Preparing for influenza after 2009 H1N1: special considerations for pregnant women and newborns" *Am J Obstet Gynecol*
10. Ding, Santibanez, Jamieson et al. (2011) "Influenza vaccination coverage among pregnant women-national 2009 H1N1 flu survey (NHFS)" *Am J Obstet Gynecol*
11. Fell, Sprague, Liu et al. (2012) "H1N1 influenza vaccination during pregnancy and fetal and neonatal outcomes" *Am J Public Health*
12. Ahluwalia, Jamieson, Angelo et al. (2010) "Seasonal influenza and 2009 H1N1 influenza vaccination coverage among pregnant women-10 states, 2009-10 influenza season" *MMWR Morb Mortal Wkly Rep*
13. White, Petersen, Quinlivan (2010) "Pandemic (H1N1) 2009 influenza vaccine uptake in pregnant women entering the 2010 influenza season in Western Australia" *Med J Aust*
14. Tarrant, Wu, Yuen et al. (2013) "Determinants of 2009 A/H1N1 influenza vaccination among pregnant women in Hong Kong" *Matern Child Health J*
15. Qiu, Bailey, Thorne (2021) "Barriers and facilitators associated with vaccine acceptance and uptake among pregnant women in high income countries: a mini-review" *Front Immunol*
16. Khan, Malik, Rafeekh et al. (2024) "Facilitators and barriers to maternal immunization and strategies to improve uptake in low-income and lower-middle income countries: a systematic review" *Hum Vaccin Immunother*
17. Kilich, Dada, Francis et al. (2020) "Factors that influence vaccination decision-making among pregnant women: a systematic review and meta-analysis" *PLOS ONE*
18. Vazquez-Benitez, Kharbanda, Naleway et al. (2016) "Risk of preterm or small-for-gestational-age birth after influenza vaccination during pregnancy: caveats when conducting retrospective observational studies" *Am J Epidemiol*
19. Hutcheon, Savitz (2016) "Invited commentary: influenza, influenza immunization, and pregnancy-it's about time" *Am J Epidemiol*
20. Munoz, Sheffield, Beigi et al. (2013) "Research on vaccines during pregnancy: protocol design and assessment of safety" *Vaccine*
21. Sheffield, Munoz, Beigi et al. (2013) "Research on vaccines during pregnancy: reference values for vital signs and laboratory assessments" *Vaccine*
22. Mcmillan, Clarke, Parrella et al. (2017) "Safety of tetanus, diphtheria, and pertussis vaccination during pregnancy: a systematic review" *Obstet Gynecol*
23. (2014) "Ebola: West Africa"
24. Bebell, Oduyebo, Riley (2017) "Ebola virus disease and pregnancy: a review of the current knowledge of Ebola virus pathogenesis, maternal, and neonatal outcomes" *Birth Defects Res*
25. Kayem, Benson, Aye et al. (2022) "Ebola virus disease in pregnancy: a systematic review and meta-analysis" *Trans R Soc Trop Med Hyg*
26. Haddad Lisa, Denise, Sonja (2018) "Pregnant women and the Ebola crisis" *N Engl J Med*
27. Henao-Restrepo, Longini, Egger et al. (2015) "Efficacy and effectiveness of an rVSV-vectored vaccine expressing Ebola surface glycoprotein: interim results from the Guinea ring vaccination cluster-randomised trial" *Lancet*
28. Legardy-Williams, Carter, Goldstein et al. (2020) "Pregnancy outcomes among women receiving rVSVΔ-ZEBOV-GP Ebola vaccine during the Sierra Leone trial to introduce a vaccine against Ebola" *Emerg Infect Dis*
29. Branswell "Ebola vaccine will be provided to women who are pregnant, marking reversal in policy"
30. Branswell (2025) "Experts call for reversing the decision to deny the Ebola vaccine to pregnant women"
31. Kavunga-Membo, Watson-Jones, Kasonia et al. (2024) "Delivery and safety of a two-dose preventive Ebola virus disease vaccine in pregnant and non-pregnant participants during an outbreak in the Democratic Republic of the Congo. Vaccines (Basel)"
32. Gomes, De La Fuente-Núñez, Saxena et al. (2017) "de la Fuente-Núñez V. Protected to death: systematic exclusion of pregnant women from Ebola virus disease trials" *Reprod Health*
33. Jones, Sam, Bull et al. (2017) "Even when you are afraid, you stay': provision of maternity care during the Ebola virus epidemic: a qualitative study" *Midwifery*
34. Yerger, Jalloh, Coltart et al. (2020) "Barriers to maternal health services during the Ebola outbreak in three West African countries: a literature review" *BMJ Glob Health*
35. Oussayef, Pillai, Honein et al. (2017) "Zika virus -10 public health achievements in 2016 and future priorities" *MMWR Morb Mortal Wkly Rep*
36. Martines, Bhatnagar, De Oliveira Ramos et al. (2016) "Pathology of congenital Zika syndrome in Brazil: a case series" *Lancet*
37. Sonja, Denise, Margaret et al. (2016) "Zika virus and birth defectsreviewing the evidence for causality" *N Engl J Med*
38. Brooks, Friedman, Kachur et al. (2016) "Update: interim guidance for prevention of sexual transmission of Zika virus -United States" *MMWR Morb Mortal Wkly Rep*
39. Delaney, Mai, Smoots et al. (2016) "Population-based surveillance of birth defects potentially related to Zika virus infection -15 states and U.S. territories" *MMWR Morb Mortal Wkly Rep*
40. Martelli, Cortes, Brandão-Filho et al. (2024) "Clinical spectrum of congenital Zika virus infection in Brazil: update and issues for research development" *Rev Soc Bras Med Trop*
41. Mercado-Reyes, Gilboa, Valencia et al. (2016) "Pregnancy, birth, infant, and early childhood neurodevelopmental outcomes among a cohort of women with symptoms of Zika virus disease during pregnancy in three surveillance sites, project Vigilancia de Embarazadas con Zika (VEZ)" *Trop Med Infect Dis*
42. Desilva, Munoz, Sell et al. (2017) "Congenital microcephaly: case definition & guidelines for data collection, analysis, and presentation of safety data after maternal immunisation" *Vaccine*
43. Cdc (2025) "About Oropouche"
44. Castilletti, Huits, Mantovani et al. (2024) "Replication-competent oropouche virus in semen of traveler returning to Italy from Cuba" *Emerg Infect Dis*
45. Thomas, Barrett (2020) "Zika vaccine pre-clinical and clinical data review with perspectives on the future development" *Hum Vaccin Immunother*
46. Friedrich (2018) "WHO's blueprint list of priority diseases" *JAMA*
47. Vouga, Chiu, Pomar et al. (2019) "Zika and chikungunya during pregnancy: pre-and post-travel advice and clinical management" *J Travel Med*
48. Hills, Meaney-Delman "Clinical guidance for use of virus-like particle chikungunya vaccine in pregnant and breastfeeding women"
49. Cepi (2025) "Priority pathogens"
50. Weber, Streblow, Coffey (2024) "Chikungunya virus vaccines: a review of IXCHIQ and PXVX0317 from pre-clinical evaluation to licensure" *Bio Drugs*
51. Cdc (2025) "Chikungunya vaccine information for healthcare providers"
52. Berrueta, Ciapponi, Mazzoni et al. (2025) "Safety, immunogenicity, and effectiveness of chikungunya vaccines in pregnant persons, children, and adolescents: a protocol for a living systematic review and meta-analysis" *Reprod Health*
53. Cardemil, Cao, Posavad et al. (2024) "Maternal COVID-19 vaccination and prevention of symptomatic infection in infants" *Pediatrics*
54. Abbasi (2022) "Widespread misinformation about infertility continues to create COVID-19 vaccine hesitancy" *JAMA*
55. Crockett, Laden, Tumin et al. (2024) "Predictors of planned home birth before and during the COVID-19 pandemic" *J Perinat Med*
56. Riggan, Weaver, Ashby et al. (2023) "Influence of the COVID-19 pandemic on prenatal and postpartum patient experiences and well-being" *Birth*
57. Firestein, Dumitriu, Marsh et al. (2022) "Maternal mental health and infant development during the COVID-19 pandemic" *JAMA Psychiatry*
58. Chourasia, Maringanti, Edwards-Fligner et al. (2023) "Nirmatrelvir and Ritonavir) use in pregnant and lactating woman: current evidence and practice guidelines-a scoping review" *Vaccines (Basel)*
59. Munoz, Cutland, Jones et al. (2022) "Preparing for disease X: ensuring vaccine equity for pregnant women in future pandemics" *Front Med*
60. Cdc (2025) "Evidence to recommendations (partial) for 2025-2026 COVID-19 vaccination"
61. Who (2025) "Mpox outbreak"
62. Oakley, Hufstetler, Shea et al. (2022) "Mpox cases among cisgender women and pregnant persons -United States" *MMWR Morb Mortal Wkly Rep*
63. Nct (2025) "Phase 3 maternal safety & immunogenicity trial of MVA-BN® in DRC (PregInpoxvac)"
64. Jean, Emma, Pradip et al. (2024) "Mpox in pregnancy -risks, vertical transmission, prevention, and treatment" *N Engl J Med*
65. Acog (2025) "Hospital disaster preparedness for obstetricians and facilities providing maternity care"
66. Regan, Rowe (2025) "Use of routinely collected "Real-world" data for the postlicensure safety assessment of vaccination during pregnancy" *Pediatr Infect Dis J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724329&blobtype=pdf | # DEAD-box helicase gcDDX56 disrupts IRF3 nuclear import complex and promotes nuclear IRF3 degradation for enhancing GCRV replication
Li Li, Xiao, Yang Chen, Jie Zhang, Jun Xiao, Hao Feng, Ming Chang
## Abstract
DEAD-box helicases play multifaceted roles in antiviral immunity, acting as both viral sensors and regulators of interferon (IFN) signaling. However, the role of DDX56 in teleost antiviral immunity remains poorly characterized. This study investigates the role of grass carp DDX56 (gcDDX56) in regulating IFN responses during grass carp reovirus (GCRV) infection. Dual-luciferase reporter assays reveal that gcDDX56 potently suppresses GCRV-induced activation of IFN1/3 promoters and RLR signaling compo nents (RIG-I, MDA5, MAVS, TBK1, IRF3/7). Mechanistically, gcDDX56 reduces IRF3 protein levels via autophagy-lysosome-mediated degradation, confirmed by cycloheximide (CHX) chase experiments and autophagy inhibitor rescue. Nuclear protein fractionation confirmed that gcDDX56 overexpression significantly reduced nuclear and cytoplasmic IRF3 protein levels during GCRV infection. Unlike mammalian DDX56, which directly interacts with IRF3, gcDDX56 does not bind gcIRF3. Instead, it sequesters importin β3 (gcKPNB3) in the nucleus, disrupting the gcIRF3-gcKPNB3 complex required for IRF3 nuclear import. Confocal microscopy validates nuclear co-localization of gcDDX56-gcKPNB3 and competitive binding with gcIRF3. Transcriptional profiling shows that gcDDX56 suppresses the expression of type I IFNs (ifn1, ifn3), IFN-stimulated genes (mx1, mx2), and RLR pathway components induced by gcIRF3-gcKPNB3. These findings establish a novel dual mechanism for gcDDX56: autophagy-dependent IRF3 degrada tion and gcKPNB3-mediated nuclear import blockade. This evolutionary divergence highlights teleost-specific adaptations in antiviral immunity, providing insights into GCRV pathogenesis and potential targets for aquaculture disease control. IMPORTANCE DEAD-box helicases exhibit versatile cellular roles in antiviral defense, functioning both as sentinels detecting viral invaders and as regulators of immune signaling. The present study reveals that teleost fish have evolved unique strategies using a related protein called gcDDX56. Unlike previously studied DEAD-box helicases, we demonstrate for the first time that gcDDX56 hijacks an importin β's (gcKPNB3) dual functions (IRF3 transport/degradation) via domain-specific interaction, acting as a "switch" to tilt gcKPNB3 toward pro-viral degradation. This novel regulatory axis (gcDDX56-gcKPNB3-IRF3) reveals a "triple hit" mechanism (cytoplasmic degradation + nuclear clearance + import blockade) that maximizes IRF3 suppression, a strategy not reported for other viruses. Beyond advancing basic knowledge of vertebrate innate immunity (expanding DEAD-box helicase/importin β functional repertoires), these findings provide actionable targets (e.g., gcDDX56-Helicase C/gcKPNB3-KAP95 interface) for developing anti-GCRV therapies, addressing a pressing need in aquaculture to mitigate GCRV-induced losses.
T he DEAD-box (DDX) RNA helicases belonging to superfamily 2 (SF2), the largest group of eukaryotic RNA helicases of six superfamilies, are named after a conserved amino sequence (Asp-Glu-Ala-Asp/His) (1). RNA helicases play regulatory roles in a variety of cellular processes, including ATP binding, ATP hydrolysis, nucleic acid binding, and RNA unwinding activity, covering virtually all aspects of gene expression and its regulation (2)(3)(4). Intriguingly, apart from their roles in RNA metabolism, RNA helicases also actively participate in viral infection given that viruses rely heavily on host RNA helicases to mediate RNA remodeling events that are part of their replication cycle or required for viral gene expression (5). For example, DDX3 has been identified as an essential cellular factor for the replication of different viruses, including important human viruses such as human immunodeficiency virus (HIV-1) or hepatitis C virus (HCV) (6). In the meantime, several DExH/D-box helicases, such as DDX3, DDX41, DHX9, and DDX1/DDX21/DHX36 complex, have been reported to act as viral sensors, while other helicases, including DDX60, DDX60L, and DDX23, function in the activation of innate immune response (7).
To defend against invading viruses, host cells initiate innate immune responses via a set of pattern recognition receptors (PRRs), including the Toll-like receptors (TLRs), retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), and cytosolic DNA sensors, such as cyclic GMP-AMP (cGAMP) synthase (cGAS), IFI16, and DDX41 (8,9). The RLRs comprise three core members: RIG-I (also known as DDX58), melanoma differentiationassociated gene 5 (MDA5, or IFIH1), and laboratory of genetics and physiology 2 (LGP2). These receptors primarily recognize viral double-stranded RNA or single-stranded RNA in most cells through their C-terminal RNA helicase domains (10). RIG-I and MDA5 interact with mitochondrial antiviral-signaling protein (MAVS) through CARD domains (11). Then MAVS recruits and activates TANK-binding kinase-1 (TBK1) and Inhibitor-κB kinase ε (IKKε). The interferon regulatory factor 3 (IRF3), a key transcription factor, is phosphory lated by activated TBK1 and IKKε. This phosphorylation triggers IRF3 dimerization, and the resulting dimers are subsequently translocated into the nucleus via the nuclear pore complex, where they drive the transcription of type I interferons (IFNs) (12,13). Notably, LGP2, which lacks CARD domains, modulates this pathway through regulatory interactions with RIG-I/MDA5 or viral RNA, fine-tuning the amplitude and duration of the antiviral response (14,15). The well-characterized importin alpha (KPNA) and importin beta (KPNB) nuclear import pathways play a crucial role in the innate immune response to viral infection by mediating the nuclear import of transcription factors, such as IRF3, NF-κB, and STAT1 (16). It is reported that IRF3 contains a nuclear localization signal (NLS) that is recognized and bound by importin-α receptors, such as KPNA2 and KPNA4, which assist IRF3 in translocating from the cytoplasm to the nucleus (17,18).
In teleosts, several RNA helicases have been reported to be involved in the innate immune response or viral infection. Overexpression of DDX5 inhibited IFN produc tion induced by spring viremia of carp virus (SVCV) and poly(I:C), and enhanced SVCV replication by targeting the autophagic degradation of TBK1 and disrupting the formation of TBK1-TRAF3 complex (19). Up to now, there are more and more research findings on DDX56 function. In mammals, DDX56 was found to play a part in cancer processes and had been identified as a potential therapeutic target in hepatocellular carcinoma (HCC) tumorigenesis (20,21). Knockout or RNAi knockdown of mouse DDX56 led to ribosome dysfunction and cell lethality, suggesting that DDX56 participated in ribosome assembly (22). At the same time, it also plays a variety of roles in viral infections. Porcine DDX56 can inhibit Pseudorabies virus (PRV) replication through promoting cGAS-STING-induced IFN-β expression (23); however, mammalian DDX56 enhanced the replication of foot-and-mouth disease virus (FMDV) by inhibiting the phosphorylation of IRF3 (24). Whether piscine DDX56 participates in viral infection and its exact mechanism remains unclear.
In this study, we cloned and investigated DDX56 in grass carp (named as gcDDX56) and identified gcDDX56 as a positive regulator of GCRV infection. gcDDX56 can disrupt the interaction between gcIRF3 and gcKPNB3, which inhibits the nuclear translocation of gcIRF3. However, this single mechanism is insufficient to explain the full extent of immune suppression observed. Instead, gcDDX56 hijacks gcKPNB3 dual functions (IRF3 transport/degradation) via domain-specific interaction, which synergistically suppress IFN signaling and boost GCRV replication via a "triple hit" strategy (cytoplasmic IRF3 degradation + nuclear IRF3 clearance + import blockade).
## RESULTS
## Sequence, expression, and subcellular localization of gcDDX56
The NCBI conserved domain analysis revealed that the gcDDX56 protein harbored a conserved DEAD-box helicase domain (DEAD) in its N-terminal region and a Heli case_C domain in its C-terminal region. These structural features are evolutionarily conserved across mammalian species and other model organisms (Fig. S1A). To investigate evolutionary relationships, a phylogenetic tree was constructed using vertebrate DDX56 orthologs. Phylogenetic tree analysis demonstrated that gcDDX56 clustered within the Cypriniformes, sharing the highest evolutionary proximity with Megalobrama amblycephala (Fig. S1B). Sequence alignment of DDX56 orthologs from human (GenBank accession number: NP_061955), mice (GenBank accession number: NP_080814), zebrafish (GenBank accession number: NP_001003876), and grass carp (GenBank accession number: XP_051766388) revealed two distinct structural features specific to the fish homologs (Fig. S1C). First, within the conserved Helicase_C domain -a region critical for ATP-dependent RNA unwinding activity-fish DDX56 contains an 8-amino-acid insertion that is absent in mammalian counterparts. Second, a 7-aminoacid insertion (in grass carp) or 8-amino-acid insertion (in zebrafish) was identified at the C-terminal region, a domain often involved in protein-protein interactions and functional modulation of helicases. These lineage-specific insertions may contribute to functional divergence of DDX56 between teleosts and mammals, potentially altering substrate specificity, regulatory interactions, or subcellular localization in the context of antiviral immunity.
Functional characterization of gcDDX56 was performed via GCRV infection studies. qRT-PCR analysis revealed significant upregulation of gcDDX56 transcripts in CIK cells following GCRV infection at an MOI of 1. Specifically, expression levels at 6, 12, and 24 h post-infection (hpi) were all significantly elevated compared to unstimulated controls, with a peak induction of ~27.19-fold observed at 24 hpi (Fig. 1A). Immunofluorescence microscopy further showed that gcDDX56 localized exclusively in the nucleus under both basal and GCRV-infected conditions. Notably, no changes in subcellular localization pattern were observed at any time point examined (6, 12, 24 hpi) (Fig. 1B).
Sequence analysis of gcDDX56 identified two putative nuclear localization signals (NLSs): one within the Helicase C domain (amino acids 340-357) and another in the Cterminal region (amino acids 526-572). To investigate their functional relevance, we generated FLAG-tagged expression plasmids harboring single or double deletions of these NLSs: gcDDX56-ΔNLS(340-357)-FLAG, gcDDX56-ΔNLS(526-572)-FLAG, and gcDDX56-ΔNLS(340-357&526-572)-FLAG. Immunofluorescence microscopy revealed that disruption of both NLSs profoundly impaired the nuclear localization of gcDDX56 (Fig. 1C). Specifically, mutants lacking either the Helicase C domain NLS (340-357) or the C-terminal NLS (526-572), as well as the double deletion mutant, exhibited exclusive cytoplasmic localization, indicating that both NLSs are collectively required for gcDDX56 nuclear import.
## gcDDX56 promotes GCRV infection and replication
To characterize the functional role of gcDDX56 in GCRV pathogenesis, CIK cells were transfected with either empty FLAG vector or gcDDX56-FLAG expression constructs, followed by GCRV infection at multiplicities of infection (MOI) of 0.1 and 1. Crystal violet staining-based cell viability assays revealed that gcDDX56 overexpression significantly reduced cell survival compared to vector controls at both MOI values (Fig. 1D). Viral titer VP5, as well as non-structural proteins NS80 and NS38, in a gcDDX56 expressiondependent manner (Fig. 1F).
To further confirm these findings, RNA interference-mediated knockdown of gcDDX56 was performed using three validated siRNA constructs. qRT-PCR screening identified si-gcDDX56-2 as the most effective silencing reagent at 100 nM concentration (Fig. 1G). Functional validation revealed that gcDDX56 knockdown resulted in dosedependent inhibition of GCRV infection, as measured by both crystal violet staining (Fig. 1H) and viral titration (Fig. 1I).
Collectively, these data establish gcDDX56 as a critical host factor that promotes GCRV replication and cytopathic effect.
## No interaction existed between gcDDX56 and GCRV proteins
To determine whether gcDDX56 promotes GCRV replication through direct interaction with viral proteins, co-immunoprecipitation (Co-IP) assays were performed. Among the 12 GCRV proteins, two nonstructural proteins (NS38 and NS80), which form VIBs, and two structural proteins (VP3 and VP5) were selected for further investigation. This selection was based on the availability of antibodies against VP3, VP5, NS80, and NS38 (25,26). CIK cells were transfected with either empty YFG-FLAG vector (negative control) or gcDDX56-FLAG expression construct, followed by GCRV infection. FLAG-tagged gcDDX56 was immunoprecipitated using anti-FLAG M2 affinity resin, and subsequent Western blot analysis detected no co-precipitation of viral proteins VP3, VP5, NS38, or NS80 (Fig. 2A). Immunofluorescence microscopy further supported these findings, showing that GCRV structural proteins VP3/VP5 and non-structural proteins NS38/NS80 displayed punctate cytoplasmic localization following infection. Notably, no co-localization was observed between these viral proteins and gcDDX56 in GCRV-infected cells (Fig. 2B). Collectively, these results indicate that gcDDX56 facilitates GCRV replication through indirect mechanisms rather than direct protein-protein interactions with viral compo nents.
## gcDDX56 inhibits GCRV-triggered activation of IFN1 and IFN3
In vertebrates, the RLR-mediated signaling pathway plays a pivotal role in recognizing RNA viruses and initiating antiviral immune responses, with type I IFN production serving as an effector of antiviral defense. Previous studies have established that grass carp IFN1 and IFN3 are the principal functional type I IFNs (27). Given the well-documented regulatory roles of DDX helicase family members in type I IFN pathways (28,29), we sought to characterize the mechanistic basis of gcDDX56's pro-viral function during GCRV infection. Using a dual-luciferase reporter assay, we assessed the impact of gcDDX56 on IFN1/IFN3 promoter activities during GCRV infection. Overexpression of gcDDX56 significantly attenuated both IFN1 and IFN3 promoter activities, whereas gcDDX56 knockdown via siRNA transfection resulted in a marked augmentation of IFN1/ IFN3 promoter activities (Fig. 3A). To dissect the molecular mechanism, we evaluated the effect of gcDDX56 on RLR signaling key components, including upstream PRRs (MDA5, RIG-I), downstream signaling adaptor protein (MAVS), critical kinase (TBK1), and tran scription factors (IRF3, IRF7). Co-expression studies revealed that gcDDX56 potently suppressed the IFN1/IFN3 promoter activation induced by these RLR pathway compo nents (Fig. 3B andC). These results collectively demonstrate that gcDDX56 acts as a negative regulator of the RLR signaling axis, thereby dampening type I IFN production and facilitating viral replication.
## gcDDX56 inhibits the expression of gcIRF3 via the autophagy-lysosome pathway
To characterize the molecular mechanism underlying gcDDX56's regulation of type I IFN production, we focused on IRF3, a critical transcription factor in the RLR signaling axis. Accumulating evidence has established that IRF3 phosphorylation represents a key checkpoint in antiviral innate immunity, particularly in response to RNA virus infections (30,31). Upon viral recognition, cytoplasmic RIG-I transmits signals through MAVS and TBK1, culminating in IRF3 phosphorylation, nuclear translocation, and subsequent type I IFN transcription.
To determine whether gcDDX56 modulates GCRV replication through IRF3 regulation, co-transfection experiments were performed in CIK cells using gcDDX56-FLAG and IRF3-HA constructs. Western blot analysis revealed that gcDDX56 overexpression significantly reduced gcIRF3 protein abundance following GCRV infection compared to YFP-FLAG controls (Fig. 4A). Endogenous IRF3 detection confirmed these findings, and conversely, gcDDX56 knockdown led to a 2.38-fold increase in gcIRF3 abundance (Fig. 4B andC).
Cycloheximide (CHX) chase assays were utilized to determine whether gcDDX56 affects gcIRF3 protein stability. Overexpression of gcDDX56 markedly degraded gcIRF3 protein in the presence of CHX, an inhibitor of protein synthesis (Fig. 4D andE). Pharma cological inhibitor studies further dissected the degradation pathway: proteasome inhibition with MG132 failed to rescue gcIRF3 levels, whereas autophagy-lysosome inhibitors NH 4 Cl, CQ, and 3-MA significantly restored gcIRF3 protein abundance in gcDDX56-overexpressing cells (Fig. 4F).
Collectively, these results demonstrate that gcDDX56 promotes GCRV replication by targeting gcIRF3 for degradation via the autophagy-lysosome pathway, thereby dampening antiviral type I IFN responses. To identify potential adaptor molecules involved in gcDDX56's regulatory function, Co-IP assays were performed in CIK cells co-transfected with gcDDX56-FLAG and gcIRF3-HA constructs. No direct physical interaction was detected between gcDDX56 and gcIRF3 under infection conditions (Fig. 5A), suggesting an indirect regulatory mechanism. Given the established role of KPNB family members in mediating nuclear import of transcrip tion factors like IRF3, we systematically evaluated interactions between gcDDX56 and grass carp KPNB isoforms (gcKPNB1, gcKPNB2, gcKPNB3). While gcKPNB1, gcKPNB2, and gcKPNB3 each harbor a conserved KAP95 domain, they exhibit substantial sequence divergence, with amino acid sequence identities ranging from 12.9% to 15.4% (Fig. S2).
Co-IP analysis revealed specific interaction between gcDDX56 and gcKPNB3 (Fig. 5B).
To map the functional domains of gcDDX56 involved in binding gcKPNB3, three truncated mutants were generated: gcDDX56-DEAD (containing only the DEAD domain), gcDDX56-Helicase C (retaining the Helicase C domain), and gcDDX56-C (encompassing the C-terminal region) (Fig. 5C). Co-IP showed that only gcDDX56-Helicase C, like fulllength gcDDX56, bound to gcKPNB3 (Fig. 5D), indicating the Helicase C domain is critical for this interaction. To define the interacting domain in gcKPNB3, two mutants were constructed: gcKPNB3-KAP95 (retaining the KAP95 domain) and gcKPNB3-C (lacking the KAP95 domain, encompassing the C-terminal region) (Fig. 5E). Co-IP demonstrated that the KAP95 domain mediates gcKPNB3-gcDDX56 binding, as gcKPNB3-C lost this interaction (Fig. 5F). Together, these findings establish a specific, isoform-selective interaction between gcDDX56 and gcKPNB, mediated by the Helicase C domain of gcDDX56 and the KAP95 domain of gcKPNB3, uncovering a novel importin β-dependent mechanism through which gcDDX56 regulates IRF3.
## gcDDX56's Helicase C domain and gcKPNB3's KAP95 domain as key for their nuclear co-localization, gcKPNB3 nuclear aggregation, and gcDDX56-regula ted gcKPNB3 nucleocytoplasmic shuttling
Immunofluorescence microscopy analyses were next performed to dissect how gcDDX56 and its domain-specific mutants modulate the subcellular localization of gcKPNB3 and its KAP95 domain-deleted variant (gcKPNB3-C). These experiments uncovered a critical functional link: the Helicase C domain of gcDDX56-which contains an NLS (340-357)is indispensable for the nuclear localization of both gcDDX56 itself and gcKPNB3. By contrast, the gcDDX56-C mutant (which retains an NLS at aa 526-572 but lacks the Helicase C domain) exhibited only minimal nuclear localization; its fluorescence signal was predominantly confined to the cytoplasm, and no appreciable nuclear co-locali zation with gcKPNB3 was observed. The gcDDX56-DEAD mutant (carrying only the DEAD domain, no Helicase C domain) was restricted entirely to the cytoplasm and similarly failed to co-localize with gcKPNB3 in the nucleus. Notably, co-transfection of the gcDDX56-Helicase C mutant with gcKPNB3 triggered robust nuclear aggregation of gcKPNB3-a phenotype that mirrored the nuclear clustering observed when full-length gcDDX56 was co-expressed with gcKPNB3 (Fig. 6A). In sharp contrast, gcKPNB3 remained diffusely distributed (without nuclear aggregation) when co-transfected with either gcDDX56-DEAD or gcDDX56-C, further validating that the Helicase C domain is required for driving gcKPNB3's nuclear aggregation (Fig. 6A).
Furthermore, gcKPNB3-C (the KAP95 domain-deleted variant) was primarily localized to the cytoplasm and showed no detectable nuclear co-localization with either fulllength gcDDX56 or any of its mutants (Fig. 6B). Together, these observations demon strate that the nuclear co-localization of gcDDX56 and gcKPNB3 is strictly dependent on two domains: the Helicase C domain of gcDDX56 and the KAP95 domain of gcKPNB3. Complementary knockdown experiments further supported this model: depletion of gcDDX56 significantly increased the immunofluorescence intensity of gcKPNB3 in the cytoplasm while reducing its signal intensity in the nucleus (Fig. S3), validating that gcDDX56 directly governs gcKPNB3's nucleocytoplasmic shuttling.
Collectively, these data establish that gcDDX56's Helicase C domain not only mediates its own nuclear localization but also drives gcKPNB3's nuclear accumulation and aggregation, while gcKPNB3's KAP95 domain is essential for this interprotein interaction and subsequent nuclear co-localization.
## gcDDX56 sequesters gcKPNB3 to disrupt gcIRF3 nuclear import by inhibiting ternary complex formation
Subsequent pull-down experiments confirmed that gcKPNB3 forms a ternary complex with both gcDDX56 and gcIRF3 during GCRV infection (Fig. 7A). Functional co-expression studies revealed that gcDDX56 overexpression reduced the binding efficiency between gcIRF3 and gcKPNB3 (Fig. 7B), whereas gcDDX56 knockdown increased this interaction by 3.23-fold (Fig. 7C), supporting that gcDDX56 acts as a competitive inhibitor of gcIRF3 nuclear import via gcKPNB3 sequestration.
Immunofluorescence microscopy further characterized the nucleocytoplasmic trafficking dynamics of gcIRF3 and gcKPNB3. In resting cells, gcKPNB3 localized predomi nantly to the cytoplasm, co-localizing strongly with cytoplasmic gcIRF3 (Pearson's R = 0.85), and showing no overlap with nuclear gcDDX56 (R = 0.01). Following GCRV infection, partial pools of both gcIRF3 and gcKPNB3 translocated to the nucleus, maintaining robust co-localization in both compartments (R = 0.84), and gcKPNB3 also co-localized significantly with gcDDX56 (R = 0.79) (Fig. 7D). In uninfected CIK cells, gcDDX56 was nuclear, and gcIRF3 was cytoplasmic. However, in GCRV-infected cells, co-transfection of gcDDX56-FLAG with gcIRF3-HA significantly reduced nuclear gcIRF3 compared to the control (FLAG+gcIRF3HA) (Fig. 8A), indicating that gcDDX56 modulates gcIRF3 localization during infection. Mechanistic dissection via co-transfection of gcIRF3-GFP, gcKPNB3-FLAG, and gcDDX56-HA showed that gcDDX56 overexpression drastically reduced gcIRF3-gcKPNB3 co-localization (R = 0.17 vs control R = 0.82) and retained gcIRF3 in the cytoplasm (Fig. 8B). Together, these data demonstrate that gcDDX56 disrupts gcIRF3 nuclear import by sequestering gcKPNB3, thereby blocking formation of the functional import complex required for IRF3 translocation.
## gcDDX56 modulates gcIRF3 nucleocytoplasmic trafficking and protein stability in a gcKPNB3-dependent manner
To biochemically validate the immunofluorescence observations, nuclear-cytoplasmic fractionation was performed in GCRV-infected CIK cells. Western blot analysis showed that GCRV infection induced nuclear accumulation of gcIRF3 in control cells transfected with empty YFP-FLAG vector, whereas this nuclear translocation was significantly attenuated in cells overexpressing gcDDX56 (Fig. S4A). Conversely, gcDDX56 knockdown resulted in a 1.79-fold increase in nuclear gcIRF3 levels compared to siRNA controls (Fig. S4B), confirming that gcDDX56 suppresses gcIRF3 nuclear translocation during GCRV infection.
To dissect the functional interplay between gcDDX56 and gcKPNB3 in regulating gcIRF3, we first validated siRNA efficacy for gcKPNB3: qRT-PCR results identified si-gcKPNB3-1 (100 nM) as the most effective silencing reagent (Fig. 9A). Subsequent nucleocytoplasmic fractionation assays revealed distinct regulatory patterns: i. Nuclear gcIRF3: Overexpression of gcDDX56 reduced nuclear gcIRF3 levels; gcKPNB3 overexpression alone had no significant effect, but gcKPNB3 knockdown further enhanced gcDDX56-mediated reduction of nuclear gcIRF3. Conversely, gcDDX56 knockdown increased nuclear gcIRF3, and this effect was amplified by gcKPNB3 overexpression. Notably, gcDDX56 knockdown failed to increase nuclear gcIRF3 when gcKPNB3 was depleted (Fig. 9B). ii. Cytoplasmic gcIRF3: Overexpression of either gcDDX56 or gcKPNB3 decreased cytoplasmic gcIRF3, while gcKPNB3 knockdown abolished gcDDX56-induced cytoplasmic gcIRF3 degradation. Intriguingly, gcDDX56 knockdown also reduced cytoplasmic gcIRF3, whereas gcKPNB3 knockdown alone increased it (Fig. 9C). However, gcDDX56 knockdown did not alter the effects of gcKPNB3 knockdown on cytoplasmic or nuclear gcIRF3 (Fig. 9B andC). iii. Total gcIRF3: Analysis of total gcIRF3 protein levels showed consistent trends.
gcDDX56 or gcKPNB3 overexpression decreased total gcIRF3, and gcKPNB3 knockdown abrogated gcDDX56-mediated total gcIRF3 degradation. In contrast, knockdown of either gcDDX56 or gcKPNB3 increased total gcIRF3, with no synergistic effect observed between the two knockdowns (Fig. 9D).
Collectively, these data demonstrate that gcDDX56 regulates both gcIRF3 nucleocyto plasmic trafficking and protein stability through a mechanism dependent on gcKPNB3, with gcKPNB3 serving as a critical mediator of gcDDX56's inhibitory effects on gcIRF3 function.
## gcKPNB3 drives gcIRF3 degradation via the autophagy-lysosome pathway
Our prior nucleocytoplasmic fractionation assays established that gcKPNB3 reduces both cytoplasmic and total gcIRF3 protein levels (Fig. 9C andD; Table S2)-an observation that prompted a critical mechanistic question: does this reduction stem from transcrip tional repression, or enhanced post-translational degradation? This distinction is biologically critical, as IRF3's antiviral function is tightly controlled by protein turnover. Given that nucleocytoplasmic fractionation also revealed gcIRF3 subcellular redistribu tion, we prioritized investigating protein degradation, as changes in subcellular localiza tion often correlate with altered protein turnover (e.g., sequestration in degradative compartments).
To first validate gcKPNB3's role in regulating gcIRF3 under physiologically relevant conditions (i.e., GCRV infection, where host cells actively upregulate IRF3 to counter viral replication), we performed Western blot analysis in GCRV-infected CIK cells. Results showed that gcKPNB3 overexpression still significantly reduced gcIRF3 protein abun dance compared to empty YFP-FLAG vector controls (Fig. S5A), indicating that gcKPNB3 actively suppresses gcIRF3 even under conditions in which the host immune system favors IRF3 accumulation.
We next directly tested whether gcKPNB3 drives gcIRF3 degradation. CIK cells were transfected with gcKPNB3-FLAG or YFP-FLAG and then treated with CHX to block new protein production. Over time, gcKPNB3-overexpressing cells exhibited markedly accelerated gcIRF3 protein loss relative to CHX-treated controls with YFP-FLAG transfec tion (Fig. S5B andC). This result definitively excluded reduced transcription or translation as the cause of gcIRF3 reduction and confirmed that gcKPNB3 enhances gcIRF3 degrada tion.
To delineate the degradation pathway, we employed pharmacological inhibitors targeting two major protein degradation systems: the proteasome and the autophagy- lysosome pathway. Treatment with the proteasome inhibitor MG132 failed to rescue gcIRF3 protein levels in gcKPNB3-overexpressing cells (Fig. S5D). In striking contrast, three distinct inhibitors (NH 4 Cl, CQ, and 3-MA) of the autophagy-lysosome pathway restored gcIRF3 abundance (Fig. S5D). The consistency of results across multiple autophagy-lysosome inhibitors unambiguously confirmed that gcKPNB3 mediates gcIRF3 degradation via the autophagy-lysosome axis.
Collectively, these data identify gcKPNB3 as a key mediator of gcIRF3 degradation via the autophagy-lysosome pathway.
## gcDDX56 suppresses IRF3-driven transcriptional activation and promotes GCRV replication in a gcKPNB3-dependent manner
To assess the transcriptional consequences of gcDDX56-mediated gcIRF3 sequestration and degradation, qRT-PCR was performed in CIK cells co-transfected with gcIRF3, gcKPNB3, and gcDDX56 constructs, with cells transfected with empty vectors (FLAG + HA + HA) as controls. Co-expression of gcIRF3 and gcKPNB3 significantly increased the transcript levels of both gcirf3 and gckpnb3 (Fig. 10A andB). In contrast, gcIRF3/ gcKPNB3 co-expression suppressed gcddx56 mRNA levels. This suppression was rescued by gcDDX56 overexpression and further enhanced by gcDDX56 knockout (Fig. 10C).
Analysis of RLR signaling components revealed that gcIRF3/gcKPNB3 co-expression robustly activated transcription of pattern recognition receptors (rig-i: 31.88-fold; mda5: 17.57-fold), the signaling adaptor mavs (47.58-fold), the kinase tbk1 (28.12-fold), the transcription factor irf7 (68.11-fold), and type I IFNs (ifn1: 122.13-fold; ifn3: 109.67-fold). These inductive effects were blunted by gcDDX56 overexpression and potentiated by gcDDX56 knockdown (Fig. 10D through 10J). Similarly, gcIRF3/gcKPNB3-induced transcription of ISGs mx1 (15.81-fold) and mx2 (8.96-fold) was significantly suppressed by gcDDX56 overexpression, whereas gcDDX56 knockdown augmented mx1/mx2 transcript levels to 24.31-fold and 21.86-fold of control, respectively (Fig. 10K andL).
We further evaluated the impact of gcDDX56/gcKPNB3 overexpression or knockdown on GCRV replication (Fig. 11A; Table S2). Overexpression of gcDDX56, gcKPNB3, or their combination reduced total IRF3 protein levels (Fig. 9D) and enhanced GCRV replication (Fig. 11A). Conversely, knockdown of gcDDX56, gcKPNB3, or both increased total IRF3 (Fig. 9D) and suppressed viral replication (Fig. 11A). Functional interplay analyses revealed mutual dependence: knockdown of gcKPNB3 abrogated gcDDX56-mediated viral promotion, while gcDDX56 knockdown similarly impaired gcKPNB3's pro-viral activity.
Collectively, these findings reflect a hierarchical, context-dependent functional interplay between gcDDX56 and gcKPNB3, rooted in their mechanistic coordination in regulating IRF3.
## DISCUSSION
RNA helicases in eukaryotes are classified into six distinct families based on conserved sequence motifs, including DEAD-box, DEAH/RHA, Ski2-like, Upf1-like, RIG-I-like, and NS3/NPH-II (32). Among these, Dead-box proteins play evolutionarily conserved roles in fundamental RNA metabolic processes, including transcription, pre-mRNA splicing, ribosome biogenesis, and translational regulation (33). In mammals, DDX56 belonging to the DEAD-box family not only regulates RNA metabolism but also modulates immune and inflammatory responses. Mammalian DDX56 interacts with viral proteins to facilitate replication of pathogens, such as Influenza A viruses (IAV) and FMDV (24,34). Despite these advances, the role of DDX56 in teleost antiviral immunity remains poorly relationship between gcDDX56, the gcIRF3/gcKPNB3 nuclear import complex, and GCRV pathogenesis.
Viruses commandeer host cellular machinery and co-opt host proteins to facilitate their replication, while hosts deploy an arsenal of antiviral factors to counteract infection. Among these, DEAD-box RNA helicases emerge as versatile players in both viral propagation and host defense. Similar to other members of the DEAD-box helicase family, DDX56 has been documented to play dual roles in viral pathogenesis. Mechanistic studies have shown that DDX56 promotes the infectivity or replication of viruses for West Nile virus (WNV), FMDV, and IAV (24,34,35). Conversely, overexpression of DDX56 exerts antiviral effects against chikungunya virus (CHIKV) through direct binding to viral genomic RNA in the cytoplasm, leading to genome destabilization and loss of infectivity (36). Our study identifies gcDDX56 as a novel pro-viral factor in GCRV pathogenesis. Unlike CHIKV-induced cytoplasmic translocation of human DDX56, whose localization affected its activity on CHIKV RNA (36), gcDDX56 retained nuclear localization through out infection, suggesting a species-specific regulatory mechanism. Furthermore, Co-IP assays also demonstrated a notable divergence of gcDDX56 from mammalian DDX56: human DDX56 could bind viral RNA or interact with capsid protein and nonstructural proteins (34)(35)(36), while the absence of direct interaction between gcDDX56 and GCRV structural and nonstructural proteins localized in the cytoplasm. Despite the lack of direct interaction with viral proteins, gcDDX56 overexpression augmented the accumula tion of GCRV structural proteins VP3 and VP5, as well as non-structural proteins NS38 and NS80. This suggests the involvement of host co-factors, potentially via nuclear transport pathways.
The recognition of viral pathogens by innate immune cells is orchestrated by multiple families of PRRs (37). These diverse PRR systems initiate complex signaling networks that converge on the activation of transcription factors NF-κB and IRF3/7 (38). In enterovi rus (EMCV) infection, DDX56 directly binds KPNA3/KPNA4, which are critical nuclear import receptors. This interaction prevents IRF3 phosphorylation and nuclear transloca tion, subsequently leading to the blockade of IFN-β production (39). While mammalian DDX56 directly interacts with IRF3 to block its nuclear translocation during viral infection (40), our study reveals a distinct regulatory mechanism in teleost fish. Functional analysis demonstrated that gcDDX56 does not directly bind gcIRF3, suggesting an alternative pathway involving intermediate adaptor proteins. Given the critical role of importin α/β-mediated nuclear transport in antiviral signaling (16), we characterized the interaction network of gcDDX56 with nuclear import machinery. Co-IP assays demon strated that gcDDX56 specifically interacted with gcKPNB3 following GCRV infection. Concurrently, gcIRF3 was found to form a functional complex with gcKPNB3, which was competitively disrupted by gcDDX56 overexpression. This competitive binding suggests a molecular decoy mechanism, where gcDDX56 sequesters gcKPNB3 to impair gcIRF3 nuclear import. Confocal microscopy revealed significant nuclear co-localiza tion of gcIRF3-gcKPNB3 complexes in GCRV-infected cells, in addition to cytoplasmic interactions. Strikingly, gcDDX56 exhibited specific nuclear co-localization with gcKPNB3 post-infection. Overexpression of gcDDX56 resulted in the reduction of nuclear IRF3 fluorescence intensity, concurrent with enhanced nuclear gcKPNB3 signals. This suggests that gcDDX56 disrupts the gcIRF3-gcKPNB3 interaction by sequestering gcKPNB3 in the nucleus, thereby blocking IRF3 nuclear translocation (Fig. 11B).
Our findings further demonstrate that gcDDX56 specifically engages gcKPNB3 via its Helicase C domain, with this interaction mediated by the KAP95 domain of gcKPNB3-a conserved HEAT-repeat region typical of importin β family members. This domain specificity aligns with the well-characterized role of HEAT repeats in importin β proteins, which govern cargo recognition through both classical and non-classical NLS motifs (41,42). While our data confirm a direct physical interaction between these domains, the precise molecular interface remains undefined: gcDDX56 may either bind to the cargo-binding pocket within gcKPNB3's HEAT repeats (thereby competing with gcIRF3) or allosterically modulate gcKPNB3 conformation to disrupt its ability to engage cargo. Resolving this distinction will require structural analyses-such as cryo-electron microscopy or X-ray crystallography of the gcDDX56-Helicase C/gcKPNB3-KAP95 complex-to map contact residues and determine whether gcDDX56 acts as a competitive inhibitor of gcIRF3 binding or induces conformational changes that impair gcKPNB3's nuclear import function. Notably, the gcDDX56-C mutant (harboring the C-terminal NLS) exhibits limited nuclear localization without concomitant gcKPNB3 co-localization, indicating that the Helicase C domain functions not merely as a nuclear targeting module but as a critical functional interface for gcKPNB3 engagement. This observation hints at a unique regulatory mode: gcDDX56 may act as a "molecular switch" that sequesters gcKPNB3 in the nucleus in a domain-dependent manner, thereby tuning gcIRF3's access to the nuclear compartment. Future studies employing singlemolecule tracking or FRET-based approaches could resolve the dynamic spatiotempo ral dynamics of this interaction during GCRV infection, clarifying whether gcDDX56 modulates gcKPNB3 recycling or retention in specific subcellular compartments to exert its regulatory effects.
A hallmark of antiviral innate immunity is IRF3/7 activation, which drives IFNs induction to orchestrate antiviral defense (43). Beyond canonical RNA sensors like RIG-I and MDA5, DEAD-box helicases (e.g., DDX3, DDX41) play pivotal roles in viral nucleic acid recognition and IFN-β induction (5). Among these, DDX56 exhibits context-dependent regulation of type I IFN responses across viruses: it suppresses IFN-β via disrupting IRF3 phosphorylation during FMDV infection (24), blocks IRF3-IPO5 nuclear import (40), and employs IFN-independent mechanisms against CHIKV (36). Against this backdrop, the present study defined how gcDDX56 negatively regulates GCRV infection, with a focus on IRF3-dependent IFN signaling. Functional assays revealed that gcDDX56 overexpression reduces both ectopic and endogenous IRF3 during GCRV infection, while knockdown rescues IRF3 levels. CHX chase assays with autophagy inhibitors (3-MA, CQ, NH 4 Cl) confirm gcDDX56 promotes IRF3 degradation via the autophagy-lysosome pathway. Concomitantly, gcDDX56 knockdown increases nuclear IRF3 (with reduced cytoplasmic pools), indicating it is also involved in IRF3 nuclear translocation. This dual regulation-modulating IRF3 stability and activation-distinguishes gcDDX56's role in GCRV infection from other viral contexts, highlighting lineage-specific adaptations of DEAD-box helicases in immune modulation.
Notably, gcKPNB3 exhibits complementary yet distinct IRF3 regulation: it predomi nantly suppresses cytoplasmic IRF3 (without altering nuclear pools) via degradation, while simultaneously mediating IRF3 nuclear import. This dual functionality expands importin β family roles beyond cargo transport to include cytoplasmic proteostasis, acting as a "bifunctional checkpoint" that balances trafficking and protein turnover. Despite mediating IRF3 nuclear import (theoretically antiviral), gcKPNB3 is net proviral: its cytoplasmic degradation reduces total IRF3 to override transport benefits, as evidenced by gcKPNB3 knockdown increasing total IRF3 and suppressing GCRV replication.
The present results clearly reveal the functional relevance of gcDDX56 and gcKPNB3. Rather than acting independently, the two proteins jointly regulate host antiviral signaling to influence viral replication through a coordinated mechanism characterized by "gcDDX56 directing functional orientation and gcKPNB3 executing dual effects. " The core logic can be broken down into three key aspects: i. Co-expression of gcDDX56 and gcKPNB3: synergistic enhancement of IRF3 inhibition to create an "optimal environment" for GCRV replication. When co-expressed, nuclear, cytoplasmic, and total IRF3 levels are all downregulated, and GCRV replication is enhanced. This reflects the core of their functional synergy, which relies on the key properties of each protein: gcDDX56 binds to the KAP95 domain of gcKPNB3 via its Helicase C domain, "sequestering" a portion of gcKPNB3 in the nucleus. This not only reduces the total amount of gcKPNB3 available in the cytoplasm to mediate IRF3 nuclear import but also clears the small pool of IRF3 that has entered the nucleus through gcDDX56-dependent degradation of nuclear IRF3. The remaining cytoplasmic gcKPNB3 (not sequestered in the nucleus) retains its inherent activity to degrade cytoplasmic IRF3, further reducing cytoplasmic IRF3 levels. Together, they form a triple inhibitory network consisting of "cytoplas mic IRF3 degradation + nuclear IRF3 clearance + blockage of IRF3 nuclear import. " This leads to a significant reduction in total IRF3, maximally suppressing antiviral signaling (IFN/ISG), and ultimately promoting GCRV replication. ii. Overexpression of gcDDX56 + inhibition of gcKPNB3: loss of gcKPNB3 relieves IRF3 total level inhibition, counteracting gcDDX56-mediated downregulation of nuclear IRF3. In this combination, nuclear IRF3 is downregulated, while cytoplas mic and total IRF3 are upregulated, and GCRV replication is suppressed. The core lies in the fact including irreplaceability of gcKPNB3 and functional limitation of gcDDX56 alone. gcKPNB3 is the main mediator of cytoplasmic IRF3 degrada tion. When inhibited, the degradative pathway for cytoplasmic IRF3 is abrogated, leading to the accumulation of cytoplasmic IRF3 and a subsequent increase in total IRF3. Although overexpressed gcDDX56 can still degrade nuclear IRF3 (resulting in downregulated nuclear IRF3), it cannot compensate for the defective cytoplasmic IRF3 degradation caused by gcKPNB3 inhibition. At this point, the upregulation of total IRF3 takes priority over the local downregulation of nuclear IRF3-the accumulated IRF3 in the cytoplasm can enter the nucleus through other nuclear transport proteins (e.g., other redundant pathways), initiating sufficient antiviral signaling to ultimately suppress GCRV replication. iii. Overexpression of gcKPNB3 + inhibition of gcDDX56: loss of gcDDX56 unlocks the "transport-function bias" of gcKPNB3, promoting IRF3 nuclear accumulation. In this combination, both nuclear and total IRF3 are upregulated, and GCRV replication is suppressed. This essentially reflects the disappearance of gcDDX56's "functional switch" role on gcKPNB3. During normal infection, gcDDX56 binds to gcKPNB3, "locking" its function into cytoplasmic IRF3 degradation (rather than nuclear transport). When gcDDX56 is inhibited, this "functional lock" is released, and overexpressed gcKPNB3 can freely exert its dual properties-"nuclear transport as the main function, degradation as the auxiliary function. " It extensively mediates the entry of cytoplasmic IRF3 into the nucleus, while its degradative function is "diluted by transport function" (more gcKPNB3 is used to bind IRF3 and mediate nuclear transport rather than degrade IRF3). On one hand, gcKPNB3-mediated IRF3 nuclear import is enhanced. On the other hand, the loss of gcDDX56 shuts down the gcDDX56-dependent degradation pathway of nuclear IRF3, increasing the stability of nuclear IRF3. Together, these two effects drive significant nuclear accumulation of IRF3, strongly activating antiviral signaling and ultimately suppressing GCRV replication. The functional relevance of the two proteins during GCRV infection can be summarized as a "guidance-exe cution" regulatory axis, where gcDDX56 acts as a "functional guidance factor" and gcKPNB3 acts as an "effector executor. " By binding to gcKPNB3, gcDDX56 determines the latter's functional bias (IRF3 degradation vs IRF3 transport). In the presence of gcDDX56, gcKPNB3 tends to degrade IRF3; in the absence of gcDDX56, gcKPNB3 tends to transport IRF3. With its dual functions of "degradation + transport, " gcKPNB3 directly regulates the total level and nuclear localization of IRF3, ultimately determining the strength of antiviral signaling.
In conclusion, our study establishes that the gcDDX56-gcKPNB3-IRF3 axis represents a critical regulatory node in antiviral immunity, one that GCRV has evolved to exploit. By enhancing gcDDX56 activity (directly or indirectly), the virus co-opts gcKPNB3's degradation function to deplete cytoplasmic IRF3 while blocking its nuclear transport, together with promoting nuclear IRF3 degradation through the autophagy pathwaythree mechanisms that converge to suppress IFN signaling. This strategy is particularly effective because it targets IRF3 at both the protein stability and subcellular localiza tion levels, ensuring robust immune evasion. From an evolutionary perspective, this underscores how viruses leverage host proteins with dual functions to maximize their replication. gcKPNB3's ability to both transport and degrade IRF3 likely evolved to fine-tune immune responses (preventing overactivation while enabling signaling), but GCRV hijacks this balance via gcDDX56, converting a homeostatic mechanism into a pro-viral one. However, our focus on IRF3 does not preclude the possibility that gcDDX56 targets other IRFs (e.g., IRF7). Furthermore, as a DEAD-box helicase, gcDDX56 is likely activated by either virus-derived pathogen-associated molecular patterns (PAMPs) or host stress signals elicited during GCRV replication. These are non-mutually exclusive mechanisms, and they align with the well-documented functional versatility of DEADbox helicases: this protein family is known to act not only as sensors of viral RNA but also as stress-responsive regulators that fine-tune innate immune signaling. Notably, two key gaps in our current understanding warrant further investigation, both of which will require analogous rigorous experimental validation to confirm biological relevance. First, our study focuses on IRF3, but it does not exclude the possibility that gcDDX56 also targets other IRF family members (e.g., IRF7)-a critical question given IRF7's central role in amplifying type I IFN responses. Second, the molecular cues that drive heightened gcDDX56 activity during GCRV infection remain undefined; resolving this will require dissecting whether activation stems from direct PAMP binding or indirect host stress pathways. These lines of inquiry not only represent tractable future directions but also hold the potential to expand our understanding of how DEAD-box helicases coordinate immune suppression across multiple IRF-dependent signaling axes and how viruses exploit such helicases to subvert host defense.
## MATERIALS AND METHODS
## Cells and virus
Ctenopharyngodon idellus kidney (CIK) cells were maintained at 26-28°C with 5% CO 2 in minimum essential medium (MEM) (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Sijiqing, China), 100 U/mL penicillin, and 100 mg/mL streptomycin. Grass carp reovirus (GCRV-873) was amplified in CIK cells using MEM supplemented with 2% FBS and stored at -80°C.
## Plasmid construction and transfection
Based on the sequences from the GenBank database (No. XM_051910428, XM_051872553, XM_051893082, and XM_051894116), the coding regions (CDS) of gcDDX56, gcKPNB1, gcKPNB2, and gcKPNB3 were cloned from cDNA extracted from the liver of grass carp. Constructs gcDDX56-FLAG and gcKPNB3-FLAG were generated using the primers gcDDX56-F1/R1 and gcKPNB3-F1/R1, respectively, and subsequently cloned into the p3 × FLAG-CMV-14 vector (Sigma-Aldrich). Similarly, constructs gcDDX56-HA, gcIRF3-HA, gcKPNB1-HA, gcKPNB2-HA, and gcKPNB3-HA were created using the primers gcDDX56-F2/R2, gcIRF3-F/R, gcKPNB1-F/R, gcKPNB2-F/R, and gcKPNB3-F2/R2 and cloned into pcDNA3.1-HA (Invitrogen). Plasmids RIG-I-FLAG, MDA5-FLAG, MAVS-FLAG, TBK1-FLAG, IRF3-FLAG, and IRF7-FLAG were previously prepared and stored (44)(45)(46). The primers are listed in Table S1. Plasmid transfection into CIK cells was performed using NEO (Neofect Biotech, Beijing) following the manufacturer's protocol.
Truncated and deletion mutants, including gcDDX56-DEAD-FLAG, gcDDX56-Helicase C-FLAG, gcDDX56-C-FLAG, gcDDX56-ΔNLS(340-357)-FLAG, gcDDX56-ΔNLS(526-572)-FLAG, gcKPNB3-KAP95-FLAG, and gcKPNB3-C-FLA, were amplified with primers gcDDX56-DEAD-F/R, gcDDX56-Helicase C-F/R, gcDDX56-C-F/R, gcDDX56-ΔNLS(340-357)-F1/R1/F2/R2, gcDDX56-ΔNLS(526-572)-F/R, gcKPNB3-KAP95-F/R, and gcKPNB3-C-F/R, respectively, and cloned into p3 × FLAG-CMV-14.
## Antibodies and reagents
The anti-FLAG mouse monoclonal antibody (mAb) (#F3165) and the FLAG immunopreci pitation kit were purchased from Sigma-Aldrich. The anti-HA rabbit polyclonal antibody (polyAb) (#51064-2-AP), anti-GAPDH mouse monoclonal antibody (mcAb) (#60004-1-Ig), and anti-Alpha Tubulin recombinant antibody (RecAb) (#80762-1-RR) were obtained from Proteintech. The anti-HDAC1 rabbit polyclonal antibody (PAb) (#ab41407) was acquired from Abcam. Additionally, the anti-NS38, anti-NS80, anti-VP3, and anti-VP5 polyclonal rabbit antibodies against the GCRV-873 strain, as well as the anti-gcIRF3 polyclonal rabbit antibody, were prepared previously and stored in our laboratory (25). The goat anti-mouse immunoglobulin-horseradish peroxidase (Ig-HRP) conjugate secondary antibody, goat anti-rabbit Ig-HRP conjugate secondary antibody, Alexa Fluor 488-conjugated secondary antibody against mouse IgG, Alexa Fluor 594-conjugated secondary antibody against mouse/rabbit IgG, Hoechst 33342, the subcellular protein fractionation kit, protease inhibitor cocktail, TRIzol reagent, and RevertAid First-Strand cDNA Synthesis Kit were purchased from Thermo Fisher Scientific.
## Sequence analysis and phylogenetic analysis
Protein conserved domains were predicted using the Conserved Domain Database (CDD) analysis from NCBI. Multiple alignments of amino acid sequences were conducted using ClustalW and GeneDoc. The phylogenetic tree of DDX56 was constructed using the Neighbor-Joining method in MEGA 11.
## Knockdown of gcDDX56 and gcKPNB3 by siRNA
Transient knockdown of gcDDX56 or gcKPNB3 was performed using siRNAs specifically targeting their mRNAs. Three siRNA sequences, each targeting distinct regions of gcDDX56 or gcKPNB3, were synthesized by Sangon Biotech (Shanghai, China). To evaluate silencing efficiency, CIK cells cultured in 6-well plates were transfected with 100 nM gcDDX56 siRNA, gcKPNB3 siRNA, or control siRNA. At 24 h post-transfection, cells were harvested for qRT-PCR analysis.
## GCRV infection in CIK cells
To assess the role of gcDDX56 overexpression or knockdown in GCRV infection, CIK cells were seeded in 24-well plates overnight, then transfected with 800 ng FLAG empty vector or gcDDX56-FLAG, along with 100 nM siRNA control or si-gcDDX56. After 36 h post-transfection, cells were infected with GCRV at an MOI of 1 or 0.1 at 26-28°C, or left untreated. At 24 hpi, cells were fixed with 4% paraformaldehyde (PFA), stained with 1% crystal violet, and photographed. Supernatants were collected to determine GCRV titers via the TCID 50 assay: CIK cells seeded in 96-well plates for 24 h were infected with 10-fold serially diluted viral samples (in MEM) for 60 min. After removing inocula, cells were cultured in MEM with 2% FBS. After 3 to 4 days, wells with CPE were counted, and TCID 50 was calculated using the Reed-Muench formula (47).
To further investigate the effects of gcDDX56 and gcKPNB3 overexpression or knockdown on GCRV replication, CIK cells in 24-well plates were transfected with combinations of constructs: YFP-FLAG plus si-RNA-control, gcDDX56-FLAG plus si-RNA-control, gcKPNB3-FLAG plus si-RNA-control, gcDDX56-FLAG plus gcKPNB3-FLAG, gcDDX56-FLAG plus si-gcKPNB3-1, gcKPNB3-FLAG plus si-gcDDX56-2, YFP-FLAG plus si-gcDDX56-2, YFP-FLAG plus si-gcKPNB3-1, and si-gcDDX56-2 plus si-gcKPNB3-1. Transfections used 400 ng plasmid and 100 nM siRNA per well. At 24 h post-transfection, cells were infected with GCRV (MOI of 1) for 12 h, and supernatants were collected to determine GCRV titers via TCID₅₀ as described above.
## Immunofluorescence assays
To visualize the subcellular localization of gcDDX56, CIK cells seeded in 24-well plates overnight were transfected with 600 ng FLAG empty vector or gcDDX56-FLAG. After 36 h post-transfection, cells were either infected with GCRV at a MOI of 1 or left untreated, then fixed with 4% PFA at 6, 12, and 24 hpi. After fixation, cells were permeabilized with 0.2% Triton X-100 for 15 min, blocked with 4% bovine serum albumin (BSA) for 1 h, incubated overnight at 4°C with anti-FLAG antibody (1:1000), and then for 1 h at room temperature with Alexa Fluor 488-conjugated secondary antibody against mouse IgG (1:500).
For co-localization analysis of gcDDX56 with GCRV proteins (NS38, NS80, VP3, VP5), CIK cells seeded in 24-well plates for 12 h were transfected with gcDDX56-FLAG. After 36 h post-transfection, CIK cells were infected with GCRV at a MOI of 1, and fixed with 4% PFA for 1 h at 12 hpi. Cells were permeabilized and blocked as above, then incuba ted overnight at 4°C with anti-FLAG, rabbit anti-NS38, anti-NS80, anti-VP3, or anti-VP5 antibodies (1:500), followed by a 1-hour incubation at room temperature with Alexa Fluor 488-conjugated secondary antibody or Alexa Fluor 594-conjugated secondary antibody against rabbit IgG (1:500).
To assess gcDDX56-mediated effects on gcIRF3 localization, CIK cells seeded in 24-well plates overnight were transfected with 600 ng of FLAG or gcDDX56-FLAG, along with 600 ng of gcIRF3-HA. After 36 h post-transfection, cells were infected with GCRV (MOI = 1) or left untreated, then fixed with 4% PFA at 12 hpi. After permeabilization and blocking, cells were incubated overnight at 4°C with anti-FLAG and anti-HA antibodies (1:1000), followed by Alexa Fluor 488-and 594-conjugated secondary antibodies for 1 h at room temperature.
For co-localization analysis of gcDDX56, gcIRF3, and gcKPNB3, CIK cells seeded overnight in 24-well plates were transfected with indicated plasmids. At 36 h post-trans fection, cells were infected with GCRV (MOI = 1) or left untreated, fixed with 4% PFA at 12 hpi, and processed as above using anti-FLAG and anti-HA antibodies followed by Alexa Fluor 488-and 594-conjugated secondary antibodies.
To determine the role of putative NLS in gcDDX56 nuclear localization, CIK cells seeded overnight in 24-well plates were transfected with full-length gcDDX56-FLAG or NLS deletion mutants (gcDDX56-ΔNLS(340-357)-FLAG, gcDDX56-ΔNLS(526-572)-FLAG, gcDDX56-ΔNLS(340-357&526-572)-FLAG). At 24 h post-transfection, cells were infected with GCRV (MOI = 1) for 12 h, fixed with 4% PFA for 1 h, permeabilized with 0.2% Triton X-100 for 15 min, and blocked with 5% BSA for 1 h. Cells were incubated overnight at 4°C with anti-FLAG antibody, followed by a 1-hour incubation at room temperature with Alexa Fluor 488-conjugated anti-mouse IgG.
To investigate effects of gcDDX56 and its mutants (DEAD, Helicase C, C-terminal) on gcKPNB3 and gcKPNB3-C localization during GCRV infection, CIK cells seeded overnight in 24-well plates were transfected with the indicated plasmids. At 24 h post-transfection, cells were infected with GCRV (MOI = 1) for 12 h, fixed with 4% PFA for 1 h, and processed as above. Cells were incubated overnight at 4°C with anti-FLAG and anti-HA antibodies, followed by a 1-hour incubation at room temperature with Alexa Fluor 488-conjugated anti-mouse IgG and Alexa Fluor 594-conjugated anti-rabbit IgG.
In all experiments, cells were stained with Hoechst 33342 for 15 min in the dark, followed by three 5-minute washes in PBST after each step. Images were acquired using a Leica SP8 confocal microscope.
## Co-immunoprecipitation assay and western blotting
For Co-IP assays, CIK cells seeded in 10 cm diameter dishes overnight were transfected with indicated plasmids or siRNAs for 36 h, followed by infection with GCRV at a MOI of 1 for 12 h. Cells were lysed overnight in 600 µL of IP lysis buffer containing 1% protease inhibitor cocktail. After centrifugation at 12,000 × g for 10 min at 4°C to remove debris, Co-IP was performed using a FLAG-tagged Protein Immunoprecipitation Kit according to the manufacturer's instructions. Agarose beads were washed three times with ice-cold PBS, incubated with cell lysate overnight, and then washed three times before being centrifuged at 1,000 × g for 1 min. Precipitates were resuspended in ice-cold PBS for Western blotting.
To investigate gcDDX56-mediated gcIRF3 degradation, CIK cells seeded in six-well plates were co-transfected with 1.5 µg YFP-FLAG or gcDDX56-FLAG plus 1.5 µg gcIRF3-HA for 36 h, followed by infection with GCRV (MOI = 1) and treatment with 100 mg/mL of cycloheximide (CHX, S7418), an inhibitor of protein synthesis. At 0, 4, 8, and 12 h post-treatment, cells were harvested for protein extraction and Western blot analysis.
To dissect the mechanism of gcDDX56-induced gcIRF3 degradation, CIK cells seeded in six-well plates were transfected with 3 µg of YFP-FLAG or gcDDX56-FLAG for 36 h, followed by infection with GCRV at a MOI of 1 and treatment with 40 µM MG132 (S2619), 10 mM 3-methyladenine (3-MA; S2767), 40 µM chloroquine (CQ; C6628), or 40 mM NH 4 Cl for 6 h. Cells were processed for protein extraction and Western blot analysis.
To map interaction domains between gcDDX56 and gcKPNB3, CIK cells in 10 cm diameter dishes were co-transfected with 6 µg each of YFP-FLAG, gcDDX56-FLAG, gcDDX56-DEAD-FLAG, gcDDX56-Helicase C-FLAG, or gcDDX56-C-FLAG plus gcKPNB3-HA for 36 h. For gcKPNB3 domain mapping, cells were co-transfected with 6 µg each of YFP-FLAG, gcKPNB3-FLAG, gcKPNB3-KAP95-FLAG, or gcKPNB3-C-FLAG plus gcDDX56-HA for 36 h. All cells were infected with GCRV (MOI = 1) for 16 h, lysed in 600 µL IP lysis buffer with protease inhibitors, and processed for Co-IP and Western blotting as described above.
For Western blotting, lysates were subjected to 10% SDS-PAGE and transferred to 0.45 µm PVDF membranes. Membranes were blocked with 5% nonfat milk in Tris-buf fered saline-Tween (TBST, 0.1% Tween) for 1 h, then incubated overnight at 4°C with primary antibodies including anti-FLAG (1:5,000), anti-HA (1:5,000), anti-GAPDH (1:5,000), anti-VP3 (1:5,000), anti-VP5 (1:5,000), anti-NS38 (1:5,000), anti-NS80 (1:5,000), anti-IRF3 (1:2,000), anti-HDAC1 (1:5,000), and anti-Tubulin (1:5,000). After washing three times with TBST, the membrane was incubated with goat anti-mouse IgG-HRP conjugate secondary antibody (1:5,000) for 1 h at room temperature. Bands were detected using Pierce ECL Western Blotting Substrate and the ECL Western blot system (LAS-4000 mini), with protein ratios quantified by Image J.
## Luciferase activity assay
To investigate the effects of gcDDX56 on the promoter activities of IFN1 and IFN3, as well as its influence on the promoter activities of interferons mediated by antiviral genes, including RIG-I, MDA5, MAVS, TBK1, IRF3, and IRF7, CIK cells were seeded overnight and subsequently transfected with the indicated plasmids or siRNA. Each transfection included 250 ng of the gcIFN1pro-luc or gcIFN3pro-luc reporter plasmid (48) and 25 ng of a Renilla luciferase reporter plasmid. After 36 h post-transfection, the cells were infected with GCRV (MOI = 1) for 12 h. Following infection, the cells were lysed for 20 min using Passive Lysis Buffer, and luciferase activity was measured with the Dual-Luciferase Reporter Assay System (Promega).
## qRT-PCR
To investigate the expression changes of gcDDX56 in response to GCRV infection, CIK cells were seeded in 6-well plates for 24 h and subsequently infected with GCRV or left untreated. RNA was extracted from the cells at 6, 12, and 24 hpi. To assess the impact of the gcIRF3-gcKPNB3 complex on the transcription of genes involved in the RLR antiviral signaling pathway and the downstream interferon-stimulated genes, as well as the role of gcDDX56 in this process, CIK cells were transfected with the specified plasmids or siRNAs after being seeded in 6-well plates. Following 24 h post-transfection, the cells were infected with GCRV at a MOI of 1. RNA extraction was performed using TRIzol reagent, and RNase-free DNase I was employed to eliminate genomic DNA remnants. First-strand cDNAs were synthesized using the RevertAid First-Strand cDNA Synthesis Kit.
qRT-PCR was conducted with Fast SYBR Green PCR Master Mix (Bio-Rad) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad) in 96-well plates. The protocol included preincubation at 95°C for 5 min, followed by 45 cycles of 95°C for 15 s, 56°C for 20 s, and 72°C for 20 s. Each sample was tested in triplicate. The relative mRNA expression was calculated by normalizing the Ct values of target genes against the housekeeping genes β-actin, EF-1α, and GAPDH using the 2 -△△Ct method. All primers used for qRT-PCR are listed in Table S1.
## Subcellular fractionation and analysis of gcIRF3 nucleocytoplasmic transloca tion
To investigate whether gcDDX56 impaired gcIRF3 nuclear translocation, CIK cells seeded in 6-well plates overnight were transfected with 3 µg FLAG/gcDDX56-FLAG or 100 nM of either si-RNA control or si-gcDDX56-2. After 36 h post-transfection, cells were infected with GCRV (MOI = 1) or left untreated, then harvested with trypsin-EDTA at 12 hpi and centrifuged at 500 × for 5 min. Subcellular fractions were separated using a protein fractionation kit: cell pellets were resuspended in cytoplasmic extraction buffer (CEB), incubated at 4°C for 10 min with gentle mixing, and centrifuged at 500 × g for 5 min to collect cytoplasmic extract (supernatant). The pellet was resuspended in membrane extraction buffer (MEB), incubated for 10 min, and centrifuged at 3,000 × g for 5 min to collect membrane extract (supernatant). Remaining pellets were resuspended in nuclear extraction buffer (NEB), incubated for 30 min, and centrifuged at 5,000 × g for 5 min to collect soluble nuclear extract (supernatant). Fractions were analyzed by Western blotting as described above.
To assess effects of gcDDX56 and gcKPNB3 overexpression or knockdown on gcIRF3 protein levels and nucleocytoplasmic translocation during GCRV infection, CIK cells seeded in 6-well plates overnight were co-transfected with 1.5 µg indicated plasmids or 100 nM siRNA. At 36 h post-transfection, cells were infected with GCRV (MOI = 1) or left untreated and then harvested and processed for subcellular fractionation (as described above) at 12 hpi. Fractions were analyzed by Western blotting as described above.
## Statistical analysis
Statistical analysis and graphs were performed and produced using GraphPad Prism 8.0 software. Data from qRT-PCR are presented as mean and SEM. The significance of results was analyzed by an ANOVA or Student's t-test (*P < 0.05; **P < 0.01; ns, not significant).
## References
1. Fairman-Williams, Guenther, Jankowsky (2010) "SF1 and SF2 helicases: family matters" *Curr Opin Struct Biol*
2. Cordin, Banroques, Tanner et al. (2006) "The DEAD-box protein family of RNA helicases" *Gene*
3. Linder, Jankowsky (2011) "From unwinding to clamping -the DEAD box RNA helicase family" *Nat Rev Mol Cell Biol*
4. Bohnsack, Yi, Venus et al. (2023) "Cellular functions of eukaryotic RNA helicases and their links to human diseases" *Nat Rev Mol Cell Biol*
5. Fullam, Schröder (2013) "DExD/H-box RNA helicases as mediators of anti-viral innate immunity and essential host factors for viral replication" *Biochim Biophys Acta*
6. Hernández-Díaz, Valiente-Echeverría, Soto-Rifo (2021) "RNA helicase DDX3: a double-edged sword for viral replication and immune signaling" *Microorganisms*
7. Bonaventure, Goujon (2022) "DExH/D-box helicases at the frontline of intrinsic and innate immunity against viral infections" *J Gen Virol*
8. Takeuchi, Akira (2010) "Pattern recognition receptors and inflammation" *Cell*
9. Barrat, Elkon, Fitzgerald (2016) "Importance of nucleic acid recognition in inflammation and autoimmunity" *Annu Rev Med*
10. Akira, Uematsu, Takeuchi (2006) "Pathogen recognition and innate immunity" *Cell*
11. O'neill, Bowie (2010) "Sensing and signaling in antiviral innate immunity" *Curr Biol*
12. Honda, Taniguchi (2006) "IRFs: master regulators of signalling by Tolllike receptors and cytosolic pattern-recognition receptors" *Nat Rev Immunol*
13. Zhao, Yang, Sun et al. (2007) "The NEMO adaptor bridges the nuclear factor-kappaB and interferon regulatory factor signaling pathways" *Nat Immunol*
14. He, Liu, Miao et al. (2023) "Black carp LGP2 suppresses RIG-I mediated IFN signaling during the antiviral innate immunity" *Fish Shellfish Immunol*
15. Gong, Zhang, Zhao et al. (2022) "LGP2 is essential for zebrafish survival through dual regulation of IFN antiviral response"
16. Vogel, Forwood, Leung et al. (2023) "Viral targeting of importin alpha-mediated nuclear import to block innate immunity" *Cells*
17. Cai, Zhang, Tang et al. (2020) "USP22 promotes IRF3 nuclear translocation and antiviral responses by deubiquitinating the importin protein KPNA2" *J Exp Med*
18. Ye, Chen, Li et al. (2017) "Japanese encephalitis virus NS5 inhibits type I interferon (IFN) production by blocking the nuclear translocation of IFN regulatory factor 3 and NF-κB" *J Virol*
19. Zhang, Cen, Yuan et al. (2023) "DDX5 inhibits type I IFN production by promoting degradation of TBK1 and disrupting formation of TBK1 -TRAF3 complex" *Cell Mol Life Sci*
20. Zhou, Du, Wei et al. (2022) "DDX56 transcriptionally activates MIST1 to facilitate tumorigenesis of HCC through PTEN-AKT signaling" *Theranostics*
21. Li, Wang, Wang et al. (2024) "DDX56 promotes EMT and cancer stemness via MELK-FOXM1 axis in hepatocellular Full-Length Text Journal of Virology December"
23. Wang, Liu, Ye et al. (2020) "Ddx56 maintains proliferation of mouse embryonic stem cells via ribosome assembly and interaction with the Oct4/Sox2 complex" *Stem Cell Res Ther*
24. Xie, Li, Yang et al. (2022) "DDX56 inhibits PRV replication through regulation of IFN-β signaling pathway by targeting cGAS" *Front Microbiol*
25. Fu, Yang, Ru et al. (2019) "DDX56 cooperates with FMDV 3A to enhance FMDV replication by inhibiting the phosphorylation of IRF3" *Cell Signal*
26. Zhang, Li, Lu et al. (2022) "Structural and functional analysis of the small GTPase ARF1 reveals a pivotal role of its GTPbinding domain in controlling of the generation of viral inclusion bodies and replication of grass carp reovirus" *Front Immunol*
27. Li, Zhang, Chen et al. (2024) "Coordination of oxysterol binding protein 1 and VAP-A/B modulates the generation of cholesterol and viral inclusion bodies to promote grass carp reovirus replication" *Front Immunol*
28. Ji, Rao, Wan et al. (2018) "Teleost-specific TLR19 localizes to endosome, recognizes dsRNA, recruits TRIF, triggers both IFN and NF-κB pathways, and protects cells from grass carp reovirus infection" *J Immunol*
29. Tapescu, Cherry (2024) "DDX RNA helicases: key players in cellular homeostasis and innate antiviral immunity" *J Virol*
30. Winnard, Vesuna, Raman (2025) "DExD-box RNA helicases in human viral infections: Pro-and anti-viral functions" *Antiviral Res*
31. Yoneyama, Suhara, Fujita (2002) "Control of IRF-3 activation by phosphorylation" *J Interferon Cytokine Res*
32. Hamrashdi, Brady (2022) "Regulation of IRF3 activation in human antiviral signaling pathways" *Biochem Pharmacol*
33. Jankowsky, Guenther, Jankowsky (2011) "The RNA helicase database" *Nucleic Acids Res*
34. Linder (2006) "Dead-box proteins: a family affair--active and passive players in RNP-remodeling" *Nucleic Acids Res*
35. Pirinçal, Turan (2021) "Human DDX56 protein interacts with influenza A virus NS1 protein and stimulates the virus replication" *Genet Mol Biol*
36. Xu, Anderson, Hobman (2011) "The capsid-binding nucleolar helicase DDX56 is important for infectivity of West Nile virus" *J Virol*
37. Taschuk, Tapescu, Moy et al. (2020) "DDX56 binds to chikungu nya virus RNA to control infection" *mBio*
39. Thompson, Kaminski, Jones et al. (2011) "Pattern recognition receptors and the innate immune response to viral infection" *Viruses*
40. Lee, Kim (2007) "Signaling pathways downstream of patternrecognition receptors and their cross talk" *Annu Rev Biochem*
41. Xu, Xie, Zhang et al. (2022) "DDX56 antagonizes IFN-β production to enhance EMCV replication by inhibiting IRF3 nuclear translocation" *Vet Microbiol*
42. Li, Fu, Wu et al. (2019) "DDX56 inhibits type I interferon by disrupting assembly of IRF3-IPO5 to inhibit IRF3 nucleus import" *J Cell Sci*
43. Cansizoglu, Chook (2007) "Conformational heterogeneity of karyopherin beta2 is segmental" *Structure*
44. Lee, Cansizoglu, Süel et al. (2006) "Rules for nuclear localization sequence recognition by karyopherin beta 2" *Cell*
45. Yoneyama, Fujita (2007) "RIG-I family RNA helicases: cytoplasmic sensor for antiviral innate immunity" *Cytokine Growth Factor Rev*
46. Zhang, Wu, Fang et al. (2022) "Grass Carp reovirus nonstructural proteins avoid host antiviral immune response by targeting the RLR signaling pathway" *J Immunol*
47. Zhang, Chang (2023) "TBK1 isoform inhibits grass carp reovirus infection by targeting the degradation of viral nonstructural proteins NS80 and NS38" *J Immunol*
48. Song, Zhang, Xu et al. (2023) "Liver X receptor LXRα promotes grass carp reovirus infection by attenuating IRF3-CBP interaction and inhibiting RLR antiviral signaling" *J Immunol*
49. Ramakrishnan (2016) "Determination of 50% endpoint titer using a simple formula" *World J Virol*
50. Wan, Yang, Rao et al. (2017) "MDA5 induces a stronger interferon response than RIG-I to GCRV infection through a mechanism involving the phosphorylation and dimerization of IRF3 and IRF7 in CIK cells" *Front Immunol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724131&blobtype=pdf | # Live-attenuated influenza virus vaccine strain with an engineered temperature-sensitive and genetically stable viral polymerase variant
Tadasuke Naito, Hiroshi Ushirogawa, Miyuki Kunishio, Haruka Yano, Susumu Saito, Taisei Higeuchi, Kazuki Fujita, Mineki Saito
## Abstract
Vaccination against seasonal influenza is considered an effective means of reducing morbidity and mortality. Live-attenuated vaccines offer more protection against influenza than inactivated vaccines as they efficiently induce cellular immunity and provide cross-immunogenicity against various antigenic subtypes. For the produc tion of safer live-attenuated vaccines, it is important to develop a common master donor vaccine strain in which pathogenic revertants are much less likely to appear. In this study, we introduced a single amino acid substitution of Lys471 into the PB1 polymerase subunit of influenza A virus and succeeded in isolating an attenuated mutant virus that exhibits a temperature-sensitive phenotype. The Lys471 residue is located in the polymerase motif D on PB1 and is positioned near the entrance tunnel domain for incoming nucleotide triphosphate. Although 10 viable PB1-Lys471 mutants did not proliferate at 37°C, their variants could replicate at 31°C and 34°C. Moreover, we found that PB1-Lys471Pro variant induces a genetically stable influenza virus phenotype; this mutant virus did not revert to wild-type phenotype from the temperature-sensi tive phenotype by serial virus passages. Animal experiments have demonstrated that these PB1 mutant strains work effectively as live-attenuated vaccines. Application of the PB1-Lys471 substitution to a master donor strain is expected to lead to the development of a safer, high-performance, and widely used live-attenuated vaccine with the antigen of circulating influenza viruses. IMPORTANCE Influenza virus elicits respiratory tract disease and is a threat to global human health. Vaccination is considered an effective tool for reducing the morbidity and mortality caused by influenza disease. The only licensed live-attenuated influenza vaccine that has been proven safe and effective is FluMist. In this study, we isolated an attenuated influenza mutant virus with a Lys471 single amino acid substitution in PB1, which displayed a temperature-sensitive and a low-pathogenicity phenotype. By applying the PB1-Lys471 substitution to the vaccine mother strain or the circulat ing influenza virus using reverse genetic technology, a high-performance and safe live-attenuated vaccine carrying the viral antigens of the vaccine-targeted strain can be developed. KEYWORDS temperature-sensitive phenotype, viral RNA polymerase, live-attenuated vaccine, influenza virus, reverse genetic analysis M ultiple types of seasonal influenza vaccine are available, that is, inactivated influenza vaccines include an egg-based, a cell culture-based, a recombinant protein-based, and an egg-based live-attenuated influenza vaccines (LAIVs) (1). These vaccines, which protect against four different influenza strains, include two influenza A
viruses (H1N1 and H3N2) and two influenza B lineage viruses. Although administration of inactivated influenza vaccines induces a protective serum antibody response, LAIVs have additional benefits of inducing mucosal and cell-mediated immunity. Therefore, LAIV confers greater breadth of protection against antigenic mutants of influenza virus. LAIV strain has to have a temperature-sensitive phenotype because the temperature in the nasal passages tends to be a few degrees cooler than the body temperature of the lower respiratory tract. The temperature-sensitive LAIV administered intranasally proliferates in the nasal mucosa and induces immunity but does not develop influenza symptoms because those vaccine strains showed an attenuated phenotype and restricted ability to replicate in the lungs. Thus, the LAIV strain replicates locally in the upper respiratory tract without causing clinical illness and induces effective immunity for protection against influenza virus infection.
Influenza A viruses belong to Orthomyxoviridae, a family of enveloped viruses with negative-sense RNA viruses. The influenza virus genome comprises eight segmented and single-stranded RNAs. It also possesses a heterotrimeric RNA-dependent RNA polymer ase (RdRp) composed of PB1 (polymerase basic protein 1), PB2 (polymerase basic protein 2), and PA (polymerase acidic protein) subunits (2,3). Each of the viral genome segments is formed as a distinct ribonucleoprotein by oligomerization of the viral nucleoprotein (NP) with the RdRp complex. The PB1 subunit is at the center of the polymerase complex and contains polymerase motifs that are common to RdRps. Recently, crystallographic and high-resolution cryo-electron microscopy structures of the complete heterotrimer of influenza virus polymerase have been determined (4,5), providing insights to aid in the understanding of the multifunctional RNA synthesis machinery involved in viral genome replication and transcription.
RNA viruses encode their own RdRp to synthesize their viral genome and mRNA within infected cells. The overall structure of RdRp resembles a cupped right hand including finger, palm, and thumb domains and catalyzes phosphodiester bond formation through a conserved two-metal ion mechanism (6). Six structural motifs, designated A to F, have been identified in RdRps. The catalytic motif A to E and motif F are distributed within the palm and finger domains, respectively. This architecture is shared with DNA-dependent DNA polymerases, DNA-dependent RNA polymerases, and reverse transcriptases and plays critical roles in the enzymatic function of polymera ses (7)(8)(9). These motifs exert important actions in the binding of metal ions, nucleo side triphosphate, and RNA, which are critical for the nucleotidyltransferase reaction catalyzed by RdRp.
Amino acid substitution(s) in the polymerase subunits of influenza virus is reported to induce variants with various phenotypes, such as temperature-sensitive or fidelitychanging mutants (10)(11)(12)(13)(14). The influenza A/Ann Arbor/6/60 (A/AA/6/60) H2N2 coldadapted virus was isolated by in vitro serial passage of the wild-type A/AA/6/60 virus at successively lower temperatures in the 1960s (11,(15)(16)(17). The cold-adapted A/AA/6/60 virus also acquired temperature-sensitive and attenuated phenotypes. To determine the viral genes responsible for the temperature-sensitive phenotypes of the A/AA/6/60 mutant, reassortant viruses between cold-adapted A/AA/6/60 and other viruses were examined, and these studies indicated that the PB1, PB2, and NP segments contributed to the temperature-sensitive phenotypes. Furthermore, site-directed mutagenesis and reverse genetic analysis mapped the temperature-sensitive phenotype of A/AA/6/60 to the following five major loci: PB1-K391E/E581G/A661T, PB2-N265S, and NP-D34G (18). The licensed LAIV is prepared by generating a reassortant containing six internal genes from a cold-adapted A/AA/6/60 and two surface glycoprotein genes, hemagglutinin (HA) and neuraminidase (NA), from the circulating strain. The HA and NA proteins are targets of the protective immune response. This cold-adapted A/AA/6/60-based LAIV is marketed as FluMist and the product name in Europe is Fluenz Tetra. FluMist can elicit IgA mucosal immunity and cellular immunity against HA and NA antigens derived from the circulating virus.
Recently, we examined the functional importance of a Lys residue involving a nucleotidyltransferase reaction in the polymerase motif D of influenza virus PB1 polymerase, by referencing an investigation of poliovirus polymerase variants (12). As a result, a Lys-to-Arg or Lys-to-His single amino acid substitution at position 481 in the PB1 subunit of the egg-adapted A/Puerto Rico/8/1934/H1N1 (PR8) influenza strain for vaccine production was identified as a lethal mutation. On the other hand, in a preliminary study, we observed that the Lys471 residue on polymerase motif D of PB1 is important for temperature sensitivity during viral growth. We hypothesized that the PB1-K471 residue, a basic amino acid conserved in the polymerase motif D between the poliovirus and the influenza virus, plays a crucial role in regulating polymerase function (Fig. 1A). In the present study, we attempted to characterize the PR8-PB1-Lys471 mutant viruses with a temperature-sensitive phenotype and application to an effective LAIV, whose replication of wild type is inhibited by vaccination using PB1-K471 variants in an animal challenge model of influenza virus infection.
## RESULTS
## A K471H mutation in PB1 polymerase induces a temperature-sensitive influenza virus
The polymerase motif D sequence of the poliovirus 3D pol and influenza virus PB1 subunit were aligned as shown in Fig. 1A. The Lys471 residue on polymerase motif D of the PB1 subunit is located near the active site for nucleotide incorporation in the influenza virus polymerase complex (Fig. 1B) (4,5). We aligned PB1 amino acids 463 to 485 in various strains, as shown in Fig. 1C. The Lys471 residue of PB1 was found to be extremely conserved. We thus hypothesized that the Lys residue in polymerase motif D might be important for the modulation of polymerase activity involving virus growth. Firstly, we tested whether a Lys-to-Arg or Lys-to-His substitution at Lys471 in the PB1 of the PR8 strain would affect its temperature sensitivity and viral polymerase activity. We also analyzed the effects of substitution in the Lys479, Lys480, and Lys481 amino acids present in polymerase motif D. Polymerase activity was evaluated using a mini-replicon reporter assay for influenza virus genome replication. To examine whether substitution of Lys residues in polymerase motif D of PB1 of the PR8 strain conferred a temperaturesensitive phenotype on the modified viral polymerase complex, a mini-replicon assay was performed at 37°C and at 34°C. Then, 293T cells in 12-well plates were transfected with each plasmid carrying the gene encoding PB2, PA, or NP, and the PB1-wild-type or PB1 mutant plasmids, together with the expression plasmid of the model viral genome encoding a luciferase gene.
We constructed PB1 mutant expression plasmids for mammalian cells and confirmed the synthesis of these mutant proteins by Western blot analysis (Fig. 1D). As shown in Fig. 1E, the level of viral RNA synthesis from the PB1-K471R mutant polymerase was reduced to 84% at 37°C and to 76% at 34°C compared to that from the PB1-wild type. Introduc tion of K471H into PB1 resulted in viral RNA synthesis reduced to 64% at 34°C compared to that with the parental PR8; in contrast, there was a 90% reduction at 37°C compared to that in the PB1-wild type (Fig. 1E, *P = 3.6 × 10 -3 by Student's t-test). Overall, the polymer ase activity of PB1-K471H was increased 6.7-fold at 34°C compared to that at 37°C (Fig. 1E, **P = 8.0 × 10 -4 by Student's t-test). The level of viral RNA synthesis from PB1-K479R/H or PB1-K480R/H substitutions was reduced compared to that in the PB1-wild type; however, these mutated polymerases did not induce a temperature-sensitive phenotype. Further, PB1 activity was almost lost following K481R or K481H substitution at both 34°C and 37°C. These results suggest that the PB1-K471H substitution produced a viral polymerase whose genome replication ability was impaired in a temperature-dependent manner.
To compare the temperature-sensitive viral replication activities of PB1 variants, we generated reassortant viruses with K471H, K471R, K479H, K479R, K480H, or K480R substitutions in the PB1 segment using a reverse genetics approach. In this study, temperature-sensitive phenotype is defined as mutant viruses that did not form viral (PB1-wt) or pCAGGS-PB1 mutants. Cells were incubated at 37°C for 24 h, and were harvested and subsequently assayed for Western blotting. Transfected PB1 proteins were confirmed with anti-PB1 polyclonal antibodies, and equal sample loading was verified with anti-β-actin monoclonal antibody. (E) The effect of Lys residue substitution on viral RNA polymerase activity using a mini-replicon reporter assay system, for influenza virus genome replication. The 293T cells were transfected with RL-SV40, pHH-vNS-Luc, pCAGGS-PB2, pCAGGS-PA, pCAGGS-NP, and either pCAGGS-PB1-wild-type (PB1-wt) or pCAGGS-PB1 mutants. Cells were incubated at 34°C or 37°C for 24 h and were harvested and subsequently assayed for luciferase activity. The luciferase activity was normalized to the (Continued on next page) plaques at 37°C and had reduced plaque formation ability at 34°C compared to the PR8wild type. Propagation of each virus in chicken eggs was examined using the plaque and hemagglutination (HA) assay (Fig. 1F and Table 1), and the HA titers of PR8-PB1-wild type and PB1 mutant strains were 1,024 to 4,096 HA units. The temperature-sensitive phenotypes of these PB1 variants were examined by plaque assays on MDCK cells at 34°C or 37°C. The PB1-wild-type virus did not exhibit the temperature-sensitive phenotype; the difference in the titers at 34°C and 37°C was only 0.24 log 10 PFU/mL. Although the virus titer reductions were not greater than 1.0 log 10 PFU/mL at 34°C or 37°C for any of the PB1 mutants compared to the PB1-wild type (e.g., PB1-K471R was reduced to 0.32 log 10 PFU/mL or 0.57 log 10 PFU/mL at 34°C or 37°C compared to the PB1-wild type), PB1-K471H, PB1-K471R, PB1-K479H, and PB1-K479R variants formed smaller plaques. In particular, a significant reduction was observed in plaque formation at 37°C for the PB1-K471H virus. This result indicated that a Lys-to-His substitution at position 471 in PB1 could induce a temperature-sensitive influenza virus phenotype.
## A point mutation in the PB1-Lys471 residue influences virus growth and a temperature-sensitive phenotype
We hypothesized that PB1-Lys471 plays an important role in modulating temperature sensitivity in viral proliferation. To test this, we generate recombinant viruses and expression plasmids in which K471 was replaced with each of the other amino acids, and viral growth, RNA polymerase activity, and temperature sensitivity of the mutant were assessed in comparison with the PR8-PB1-wild-type virus. First, we tested whether each mutant virus was generated by using a reverse genetic system. Eight recombinant viruses were not found to synthesize in a reverse genetic approach using plasmid-trans fected 293T cells (Table 2, "PFU/mL of seed viruses" column, indicated as not detected [ND]). Propagation of the viable mutant virus in chicken eggs was examined by HA assay and plaque assay. HA titers of viable PB1-K471 mutants were 512 to 2,048 HA units (Table 2, "HA titer of E1 viruses" column). These PB1-K471 mutants detected obvious plaque formation at 31°C or 34°C using an immunostaining method (Fig. 2). Ten PB1 recombi nant viruses, PB1-K471H(His), K471V(Val), K471I(Ile), K471L(Leu), K471M(Met), K471C(Cys), Renilla luciferase activity. At each temperature, the Firefly/Renilla relative activity of PB1-wt polymerase was set as 1.0, and relative activity of PB1 mutants was expressed. Quantitative results are presented as the average with the standard deviation (SD) from at least three independent experiments. Significance was determined using Student's t-test (*, P = 3.6 × 10 -3 ; **, P = 8.0 × 10 -4 ). (F) A K471H mutation in PB1 polymerase induces a temperature-sensitive influenza virus.
Plaque morphology of PR8-PB1 mutants. Plaque assay was performed using an amplified version of each virus in chicken eggs. Serial dilutions (10 -7 to 10 -5 ) of the PB1-wild-type or PB1 mutant viruses were used to infect MDCK cells at the indicated temperatures. MDCK cells were infected with the viruses indicated at the left of the figure and incubated at 34°C or 37°C for 4 or 5 days. Virus plaques were visualized by Amido Black 10B staining and photographed. K471P(Pro), K471T(Thr), K471Q(Gln), and K471A(Ala), did not generate plaques at 37°C; therefore, these substitutions induce a temperature sensitivity in viral replication. Next, we tested the effect of Lys471 residue substitution on viral polymerase activity by using the mini-replicon assay. We constructed PB1 expression plasmids for mamma lian cells and confirmed the synthesis of these PB1-K471 mutant proteins by Western blot analysis (Fig. 3A). The RNA synthesis rates of PB1-wild-type and K471 mutant proteins are shown in Table 3 (see "relative fold change") and Fig. 3B. Plasmid-transfected cells were incubated at 31°C, 34°C, or 37°C for 24 h and were subsequently assayed for luciferase activity. The viral polymerase activities of the mutants with PB1-K471F(Phe), K471Y(Tyr), K471G(Gly), K471S(Ser), K471N(Asn), K471D(Asp), K471E(Glu), and K471W(Trp) substitu tions were reduced more than 10-fold in comparison to that of the PB1-wild type at 34°C, whereas activities were reduced by only about 10-fold in the other, viable K471 mutants excluding K471A (Fig. 3B, middle panel). The level of viral RNA synthesis from the PB1-K471A mutant polymerase was reduced to 3.7% compared to that from the PB1-wild type at 34°C; therefore, it is possible that a decrease in polymerase activity induced a marked reduction in the plaque size reduction of a K471A mutant virus (Fig. 2). On the other hand, the viral polymerase activities of 17 PB1-K471 mutants excluding K471R and K471I were reduced more than 100-fold in comparison to that of the PB1-wild type at 37°C (Fig. 3B, bottom panel). These results suggest that the reduction in viral polymerase activity by more than 100-fold compared to PB1-wild type led to the suppression of viral proliferation. Specifically, plaque formation was not observed at 37°C in those 17 PB1-K471 mutant strains (Fig. 2). The polymerase activities of all 19 PB1-K471 mutants at 31°C showed a similar trend to the results at 34°C. Overall, these data suggested that a PB1-K471 mutant can be applied to the LAIV by using a temperature-sensitive phenotype. a The PFU/mL was determined using virus particles that generated from reverse genetics plasmid transfected 293T cells. The number of plaques was counted following Amid Black 10B staining after 4 days of inoculation at 34°C. Data are means from three independent experiments. Errors are represented as standard error. b ND, not detected. ND means that no plaque formation was observed in a plaque assay using 1 mL of undiluted virus stock.
## c
HA titers were determined using viruses amplified in egg (abbreviated here as E1) that were infected with 200 PFU of seed viruses generated by RG system. d NT, not tested. e PFU/mL were determined using viruses amplified in egg (abbreviated here as E1) that were infected with 200 PFU of seed viruses generated by RG system. The number of plaques was counted following immunostaining after 3 days of inoculation at 31°C, 34°C, or 37°C. Data are means from three independent experiments. Errors are represented as standard error. f NC, not calculated. PB1-K471I could not form a clear viral plaque at 37°C incubation.
Full-Length Text Journal of Virology
## PB1-K471P strain maintained the temperature-sensitive phenotype after serial passages in cell culture
The live-attenuated vaccine should not revert back to the wild-type virus or a revertant virus following vaccination. Therefore, the LAIV strain needs an attenuated phenotype involved in temperature sensitivity during viral growth, including during the vaccine manufacture processes. To apply PB1-K471-variant to LAIV, we tested whether these mutants could maintain a temperature-sensitive phenotype on temperature-sensitive PB1-K471 mutants using a mammalian cell culture (Fig. 4A). MDCK cells were infected with PB1-wild type and PB1-K471 variants and cultured at 34°C. From 3 days post-infec tion (dpi), the supernatant was recovered and named passage 1 (P1). The virus passages were then continued until P5, and temperature sensitivity of the P0 and P5 viruses was checked at 34°C and 37°C. As shown in Fig. 4B, the P0 virus of 10 PB1-K471 variants did not form plaques at 37°C (shown by Amido Black 10B staining), and in addition, PB1-K471C and PB1-K471A did not display clear plaque formation at 34°C. According to the two independent experiments of virus passages, PB1-K471 variants (except for PB1-K471P) overcame the growth restriction on 37°C incubation at P5. In addition, P5 viruses of PB1-K471 variants formed clearly visible plaques at 34°C, including PB1-K471P. These results suggested that only the PB1-K471P-variant maintained temperature-sensi tive phenotypes that prevented the reversal of growth restrictions at 37°C, after serial passages.
Next, we identified the sequence of the viral genomes derived from the P5 viruses of PB1-K471 variants that adapted to the viral growth at 37°C. We analyzed the nucleotide sequences of all three polymerase subunits (PB1, PB2, and PA)-specific and NP-specific reverse-transcription PCR products generated using RNA collected from P5 viruses. The observed nucleotide sequence and amino acid changes are summarized in Fig. 4C and Table 4. We found that the PB1-K471P-variant maintained Pro residue at position 471 in PB1 after serial viral passages. On the other hand, PB1-K471I, PB1-K471M, or PB1-K471Q variants changed to Lys residues (AAG or AAA codon) from the Ile (ATA codon), Met (ATG codon), or Gln (CAG codon) residue. P5 viruses of PB1-K471I, PB1-K471M, or PB1-K471Q variants lost the temperature-sensitive phenotype, due to being reverted to PB1-wildtype viruses, as a result of viral passages. In other PB1-K471 variants (K471H, K471V, K471L, K471C, K471T, and K471A), the position 471 amino acid did not revert to a Lys residue, whereas one additional mutation into the PB1 subunit, which resulted in an Ala-to-Val change at position 349, was identified. This amino acid change was caused by a C-to-T transition at the second base of codon 349 (GCG to GTG). We assume that the mutation of PB1-A349V induced a structural change in PB1, canceling the ability of K471-mutation involving temperature sensitivity. A349V was localized close to the active site, near the amino acid at position 471 (Fig. 4D).
To examine the effect of the PB1-A349V substitution into the PB1-K471 variants on temperature sensitivity, we generated a PB1-A349V and a PB1-A349V/K471-double The values were calculated using data from Fig. 3 (polymerase activity data).
b P-values were calculated using the t-test, in which we compared the polymerase activity for a particular mutation with the polymerase activity obtained using the PB1-wild type. mutant virus and then propagated the mutant viruses in chicken eggs, examined by HA assay and plaque assay. HA titers of PB1-A349V and PB1-A349V/K471-double mutants were 1,024 to 2,048 HA units (Table 5, "HA titer" column). The PB1-A349V virus formed plaques similar to that of PB1-wild type at 34°C or 37°C (Fig. 5A). The PB1-A349V single mutation had no influence on the induction of temperature sensitivity. PB1-A349V/ K471-double mutants (except PB1-A349V/K471P) had the ability of plaque formation at 34°C and 37°C for cultures of 4 days. This result suggests that A349V substitution on PB1 restored the ability of viral proliferation at 37°C for PB1-K471 variants, except for PB1-K471P. To determine the virus titer and temperature sensitivity of PB1-A349V/ K471P-variant, plaque assays were carried out, and the infected cells were cultured for 3 days at 31°C, 34°C, or 37°C (Fig. 5B and Table 6). The PB1-A349V/K471P-variant formed clearly visible plaques at 31°C or 34°C by immunostaining. Therefore, the PB1-K471Pvariant could maintain a temperature-sensitive phenotype, even if the A349V second substitution was induced in PB1. To strengthen the result that the single K471P amino acid change in the PB1 polymerase of the PR8 strain had induced genetic stability, we conducted an additional serial virus passage experiment, and its data are described in the supplemental material (Fig. S1 and Table S1). According to a previous study (19), we performed serial virus passages in MDCK cells at gradually elevated temperatures (Fig. S1A). The results of these experiments suggested that the PR8-PB1-K471P strain could not produce progeny viruses during the 11 passages under increasing temperatures.
## A single dose of PB1-K471-variant can elicit protection against influenza virus challenges in animals
To assess the potential of the PR8-PB1-K471-variant for the LAIV strain, we examined whether vaccination with a single dose of PR8-PB1-K471H or PR8-PB1-K471P viruses could protect mice against a lethal influenza virus. Firstly, we tested vaccine susceptibility in animals: C57BL/6J mice were infected intranasally with 10 6 PFU of PR8-PB1-K471H or PR8-PB1-K471P viruses and monitored for body weight and survival rate (Fig. 6A).
In addition, to evaluate the vaccination of PR8-PB1-K471 variants, a PR8-based temper ature-sensitive strain was created, designated PR8-FluMist (see Materials and Meth ods, PR8-FluMist strain encoding PB1-K391E/E581G/A661T & PB2-N265S) (Fig. 6B); the licensed LAIV FluMist backbone. The PR8-FluMist strain did not show plaque formation by Amido Black 10B staining at 37°C (Fig. 6B) and that was used as a control vaccine strain for comparison with the PR8-PB1-K471H-or the PR8-PB1-K471P-vaccinated mice.
Previously, it has been reported that introducing five amino acid mutations (PB1-K391E/ E581G/A661T, PB2/N265S, and NP/D34G) into PR8 virus induced a temperature-sensi tive phenotype (20). In this study, we generated recombinant influenza strains using egg-adapted high-growth PR8 virus provided by Dr. Yoshihiro Kawaoka (University of Wisconsin-Madison). Since the amino acid position 34 of the NP protein of the egg-adap ted high-growth PR8 virus is glycine residue, therefore, we introduced four amino acid mutations in PB1 (K391E, E581G, and A661T) and PB2 (N265S) to create a temperaturesensitive strain based on PR8. Likewise, a group has reported that the NP-D34G mutation in FluMist is already present in a PR8 strain (21,22). The results of viral growth kinetics and plaque assays of PR8-FluMist and PR8-PB1-K471-variant at various temperatures were described in the supplemental material (Fig. S2 andS3). The PR8-PB1-K471 variants-or PR8-FluMist-vaccinated mice maintained a normal weight and survived as well as the mock-vaccinated group (Fig. 6C). To confirm the pathogenicity of the viral dose of 10 6 PFU, which is the same dose used for PR8-FluMist or PR8-PB1-K471 variants vaccination, PR8-wild-type virus was intranasally inoculated into mice at a dose of 10 6 PFU (Fig. S4). Mice infected with 10 6 PFU of PR8-wild-type virus showed severe symptoms accompanied by rapid weight loss, and subsequently, three out of eight mice died after infection. This indicates that the PR8-PB1-K471 variants are attenuated and have potential for use as a safe attenuated live vaccine strain. Four weeks after intranasal inoculation with PR8-PB1-K471 variants or PR8-FluMist, mice were given a lethal dose of 10 7 PFU of PR8-wild-type virus (Fig. 6D). Mock-vaccinated mice infected with PR8-wild-type virus developed severe symptoms with rapid weight loss and did not survive for more than 6 days after infection. On the other hand, PR8-PB1-K471Hor PR8-PB1-K471P-vaccinated mice maintained body weight and survived for 14 days. Similarly, PR8-FluMist-vaccinated mice also survived following infection, however, slight weight loss was observed until 3 dpi. Next, we assessed the control of virus titers in infected mice mediated by these vaccine candidates (Fig. 6E). We infected vaccinated mice with PR8-wild-type virus and harvested lungs at 2, 4, or 6 dpi and measured viral titers in the lung homogenates. In PR8-PB1-K471H-or PR8-PB1-K471P-vaccinated mice, the challenge virus was below the level of detection, whereas titers of 10 5 to 10 7 PFU/g were found in lungs of mock-vac cinated mice at 2 or 4 dpi. Mice vaccinated with PR8-FluMist had reduced virus titers following challenge with PR8-wild type at 2 dpi, and no challenge virus was detected at 4 or 6 dpi. These results suggested that PR8-PB1-K471 variants have the strong immunizing potential for the influenza virus.
## Vaccination with PR8-PB1-K471P elicits antibody responses and increases influenza-specific IFN-γ-secreting T cells
The host response to vaccination with PR8-PB1-K471P suggested strong protective efficacy against viral infection, including adaptive immunity. We tested antibody responses by vaccinating groups of six mice with 10 6 PFU of PR8-FluMist or PR8-PB1-K471P and collected serum at 4 weeks post-inoculation. The antibody response was determined by a hemagglutination inhibition (HAI) assay. The PR8-PB1-K471P virus was capable of inducing an equally high HAI titer as the PR8-FluMist (Fig. 7). Furthermore, to confirm whether PR8-PB1-K471P vaccination induces T cell responses, we measured the production of interferon-gamma (IFN-γ) in immunized mice. Splenocytes were collected from vaccinated mice (n = 6) at 4 weeks post-inoculation and stimulated with the MHC class I pentamers, NP 366-374 or PA 224-233 peptides, and subjected to the ELISpot assay to evaluate the CD8 + cytotoxic T-cell activity. As shown in Fig. 8, significantly higher IFN-γ spot-forming units (SFUs) were observed in the antigen peptides-stimulated PR8-PB1-K471P-vaccinated group than in the unstimulated mice group (Fig. 8B andC). PR8-FluMist vaccination also was significantly induced IFN-γ production by stimulation of antigen peptides, and the degree of IFN-γ + SFUs was similar between the PR8-PB1-K471P-vaccinated group and the PR8-FluMist-vaccinated group.
Live-attenuated vaccines are thought to give broad cross-protection against hetero typic viruses through activation of both humoral and cell-mediated immune response. To confirm the protective effect of PR8-PB1-K471P against challenge infection with heterologous strain, we generated PR8-maH3N2(6:2) virus expressing the HA and NA of A/Hong Kong/MA(mouse-adapted)/1968/H3N2 strain on the PR8 backbone. To evaluate the protection from disease or death, we challenged PR8-PB1-K471P or PR8-FluMistvaccinated mice with 10 6 PFU of PR8-maH3N2(6:2) viruses (Fig. 9A). All of the mockvaccinated mice died (n = 8), but three out of nine mice vaccinated with the PR8-PB1-K471P survived (Fig. 9B). The PR8-PB1-K471P-vaccinated group showed 30% survival despite body weight loss (P = 1.9 × 10 -2 by the log-rank test), whereas the group at PR8-FluMist inoculation survived one out of eight mice (Fig. 9C). Collectively, these results indicated that the administration of PR8-PB1-K471P as a live-attenuated vaccine induces protective effects against influenza virus in mice and promotes virus clearance.
## DISCUSSION
In this study, we analyzed the properties of recombinant viruses in which mutations were introduced into the Lys471 residue of PB1 in the influenza virus RNA polymerase. Of the 19 variants, 10 were viable and those PB1-L471 mutant viruses acquired temperature sensitivity. Nine PB1-K471 mutants, except for the PB1-K471P virus, mutated, causing a reversion of the temperature-sensitive phenotype to wild-type, which occurred by viral serial passaging experiments using MDCK cells. The PB1-K471P variant did not revert to the wild-type PB1 phenotype even after repeated viral passages and maintained lowtemperature proliferation; therefore, this strain has a high genetic stability and was suggested to be useful as the mother virus for LAIV. Results of the virus challenge experiments using animals showed that mice vaccinated with the PB1-K471P mutant strain were protected from lethal doses of wild-type virus infection. These results suggest the possibility that the single amino acid mutation of PB1 in the PR8 H1N1 strain of influenza A virus can be utilized as a backbone strain for LAIV.
Multiple temperature-sensitive mutations have been identified in the polymerase subunit of the influenza virus (18,(23)(24)(25), and several studies have reported on the determination of corresponding phenotypes (26,27). Temperature-sensitive mutations can be broadly classified into two categories: mutations that produce thermally unstable proteins and cause defects in protein synthesis, folding, or assembly (28). The conserved Lys471 residue in PB1 is positioned in polymerase motif D of the palm domain and located near the NTP entrance tunnel. The motif D contains highly conserved lysine residues 480 and 481, which are involved in NTP binding, and is stabilized by contact with the PA helix α20 and the amino acid residues 671 to 684 of PA (29,30). There is a possibility that mutation of PB1-Lys471 in motif D modulates correct assembly or stabilization of the heterotrimer viral polymerase comprising subunits PB1, PB2, and PA at restrictive temperature. Our results suggest that although position 349 residue in PB1 may seem distant from the RdRp active-site residues, the fingers motif and/or other polymerase motifs might undergo substantial rearrangement during polymerization by PB1-A349V substitution to cancellation of the temperature-sensitive phenotype induced in PB1-K471 mutants (except PB1-K471P variant). An amino acid residue in proteins is replaced by a proline residue that led to disturb potential structuration of those domains. Therefore, we assumed that the PR8-PB1-K471P variant did not revert to a non-tempera ture-sensitive phenotype even with the introduction of additional PB1-A349V substitu tion.
The ability of influenza viruses to aid in immune escape requires that vaccine strains need to be annually updated to reflect changes in antigenicity in the HA and the NA genes within the epidemic seasonal strains. Multiple types of influenza vaccines are currently used, containing an inactivated vaccine and LAIV of a cold-adapted virus, FluMist, delivered as an intranasal administration. LAIV is more immunogenic than inactivated vaccines, due to its ability to stimulate both humoral and cell-mediated immune responses. On the other hand, it has been reported that FluMist might not be Six-week-old female C57BL/6J mice were intranasally inoculated with 10 6 PFU of PR8-PB1-K471P (n = 6), or PR8-FluMist (n = 6) virus, and PBS (Mock, n = 6). Four weeks after vaccination, sera were collected, and HAI titers against PR8-wild-type-virus were measured by HAI assay.
sufficiently effective in preventing influenza virus infection in recent seasons (31). Numerous groups have reported the development of LAIV (11,(32)(33)(34), and clinical studies are underway to commercialize those vaccine candidates; however, it has not been sufficiently investigated whether those LAIV strains have genetic stability. In contrast, the importance of a live-attenuated vaccine with genetic and phenotypic stability has been demonstrated in the development and the clinical trial of oral poliovirus vaccine candidates (35)(36)(37)(38)(39)(40). The PR8-PB1-K471P mutant strain isolated by our work was found to possess temperature sensitivity and genetic stability, resulting in an attenuated phenotype in vivo, which is an essential function of LAIV. The ability to eliminate the possibility of the emergence of pathogenic revertant mutants makes the LAIV constructed using the PR8-PB1-K471P strain useful. The PR8 strain has a very high growth efficiency in MDCK cells and chicken eggs; therefore, PR8-based inactivated vaccines have been developed encoding the HA and the NA of the seasonal influenza virus, or H5N1 and H7N9 avian influenza viruses (41,42).
Highly pathogenic avian influenza A viruses cause severe infections in humans, and various forms of vaccines, including inactivated vaccines, are being developed (43,44). The viral antigen of H5N1 in the inactivated vaccine can activate an effective immune response, and live-attenuated vaccines against a highly pathogenic influenza virus are being developed using A/AA/6/60 as the mother strain. These studies have advanced to the stage of clinical trial using healthy donors (45). In order to develop subtype-specific and multiple pre-pandemic vaccines, it is essential to isolate cold-adapted backbone strains for LAIV other than A/AA/6/60, and PR8-PB1-K471P could expect to be applied as one of the candidates. PR8-PB1-K471P has the potential to sufficiently meet the virus titer required for the vaccine production process. To investigate this, we produced recombi nant PR8-PB1-K471P vaccines encoding the HA and the NA of seasonal influenza virus and intend to examine its efficacy in a later study (see supplemental text, Fig. S5, and Table S2). Likewise, we will attempt to develop a live-attenuated pre-pandemic vaccine encoding the antigen of a highly pathogenic influenza virus, using PR8-PB1-K471P mother strain. Vaccine safety is determined by the characteristics of the vaccine form and the age or health condition of the host. For instance, current LAIV is restricted to healthy nonpreg nant persons over 2 and under 50 years of age. Therefore, it is important to develop a novel LAIV with better safety properties to target people that do not fit the requirements for the administration of the current LAIV. In our study, PR8-PB1-K471P-vaccinated mice were fully protected against a lethal challenge with PR8-wild-type virus and did not show body weight losses. Both PR8-PB1-K471P and PR8-FluMist are temperature-sen sitive viruses, but their phenotype involving the viral replication activity, etc., is not identical. This is important, as LAIV might have affected the difference in body weight between PR8-PB1-K471 variant-and PR8-FluMist-vaccinated mice after a wild-type virus challenge (Fig. 6D). We are planning additional investigations to assess the ability of PR8-PB1-K471 variants as new vaccine backbone strains to elicit protection against the infection of several subtypes of influenza virus, so the immune response can be compared with administration of licensed LAIV. Moreover, introducing the PB1-K471P additional mutation into the PB1 genome of FluMist strain may improve the genetic stability of the FluMist, which could enhance its effectiveness and safety. These studies would facilitate the clinical development of a highly safe and effective LAIV.
## MATERIALS AND METHODS
## Molecular modeling
Ribbon diagrams and space-filling representations of influenza virus polymerase were generated using structural data (PDB ID: 6T0V) and MolFeat software version 5.2.4.29 (FiatLux).
## Cells and animals
293T and MDCK cells were maintained in DMEM (Thermo Fisher Scientific) containing 10% fetal calf serum (Nichirei Biosciences) and penicillin-streptomycin in an incubator at 37°C with 5% CO 2 . C57BL/6J mice were purchased from Japan SLC. The mice were maintained at macroenvironmental temperature and humidity ranges of 20°C to 25°C and 40% to 60%, respectively, with a 12 light/12 dark light cycle.
## Generation of protein expression plasmids for PB1 mutants
Mutations corresponding to an amino acid substitution at the Lys471 (codon, AAG) residue of the PB1 subunit were introduced into a plasmid containing a sequence encoding the wild-type PB1 by site-directed mutagenesis. The Lys residue in the PB1 gene of the egg-adapted high-growth PR8 strain was changed from AAG to CAC (His codon) or AGA (Arg codon) by PCR-induced mutagenesis. To construct a plasmid containing the PB1-K471H coding sequence, two DNA fragments corresponding to the PB1 coding sequence were amplified by PCR using primers PB1-for and K471H-rev, or K471H-for PB1-rev and (Table S3), with pPolI-PB1-wild-type (46) as the PCR template.
The full-length PB1-K471H gene was amplified by PCR using primers PB1-for and PB1-rev. PCR products were digested using KpnI and NotI and were cloned into KpnIand NotI-digested pCAGGS-P7 plasmids. The resultant plasmid was designated pCAGGS-PB1-K471H. Likewise, the pCAGGS-PB1-K471 mutant plasmids were constructed using the primers listed in Table S3. The preparation of pCAGGS-PB1-K479H, -K479R, -K480H, -K480R, -K481H, and -K481R has been described previously (12).
## Mini-replicon reporter assay system
293T cells were transfected with expression plasmids encoding PB1 (pCAGGS-PB1-wildtype or PB1-K471 mutants), PB2, PA, and NP, and a plasmid (pHH-vNS-Luc) for express ing the artificial influenza virus genome containing the Firefly luciferase gene in the negative-sense, which was synthesized in cells by the human DNA-dependent RNA polymerase I (PolI) (47). Its negative-sense RNA containing the Firefly luciferase gene was sandwiched by 5′-and 3′-terminal untranslated regions of segment eight genome encoding non-structural protein of the influenza A virus. The mRNA encoding Firefly luciferase was transcribed in an influenza viral RNA polymerase-dependent manner.
Luciferase activity was determined using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer's protocol and was normalized to Renilla luciferase activity encoded by the co-transfected pRL-40 vector (Promega).
## Plasmid construction for generating recombinant viruses
To generate plasmids encoding an amino acid point mutation at the Lys471 residue of the PB1, the pPolI-PB1-wild-type plasmid was used as the backbone vector (46). To construct the plasmid from which PolI transcribes the PB1-K471H viral RNA, we amplified two DNA fragments corresponding to the PB1-coding sequence by PCR using primers Pol1-for and K471H-rev, or K471H-for and Pol1-rev (Table S3), with pPolI-PB1-wild-type as the PCR template. The full-length PB1-K471H gene was amplified by PCR using primers Pol1-for and Pol1-rev. The PCR product was digested using ApaI and XhoI and cloned into the ApaIand XhoI-digested pPolI plasmid. The resultant plasmid was designated pPolI-PB1-K471H. Likewise, pPolI-PB1-K471 mutant plasmids were constructed using the primers listed in Table S3. Preparation of pPolI-PB1-K479H, -K479R, -K480H, and -K480R has been described (12).
To construct pPolI-PB1-A349V or pPolI-PB1-A349V/K471H double mutant plasmids, we amplified two DNA fragments corresponding to the PB1-coding sequence by PCR, using primers Pol1-for and A349V-rev or A349V-for and Pol1-rev (Table S3) with pPolI-PB1-wild-type or pPolI-PB1-K471H plasmids as the PCR template. The full-length PB1-A349V or PB1-A349V/K471H gene was amplified by PCR using primers Pol1-for and Pol1-rev. The PCR product was digested using ApaI and XhoI and cloned into the ApaIand XhoI-digested pPolI plasmid. The resultant plasmid was designated pPolI-PB1-A349V or pPolI-PB1-A349V/K471H. Likewise, pPolI-PB1-double mutant plasmids containing A349V and K471 mutations were constructed using each pPolI-PB1-K471 mutant plasmids as the PCR template.
To generate a PR8-based temperature-sensitive strain, designated PR8-FluMist, we constructed pPolI-PB1-K391E/E581G/A661T and pPolI-PB2-N265S plasmids (18,20,24). To construct pPolI-PB1-K391E/E581G/A661T plasmids, we amplified two DNA fragments corresponding to PB1-K391E/E581G or PB1-A661T mutant sequences by PCR using primer set PB1-3mut-for-1 and PB1-3mut-rev-1 or PB1-3mut-for-2 and Pol1-rev (Table S3) with synthetic DNA coding PR8-PB1-K391E/E581G or PR8-PB1-A661T fragments (gBlocks Gene Fragments, Integrated DNA Technologies) (Table S4) as the PCR tem plate. Then, the PB1-K391E/E581G/A661T fragment (the 3′-fragment of PB1 sequence, 1,078 to 2,274) was amplified by PCR using primer PB1-3mut-for-1 and Pol1-rev. The 5′-fragment of PB1 sequence was amplified by PCR using primer PolI-for and PB1-3mutrev-2 with pPolI-PB1-wild-type as the PCR template. The full-length PB1-K391E/E581G/ A661T gene was amplified by PCR using primers Pol1-for and Pol1-rev. The PCR product was digested using ApaI and XhoI and cloned into the ApaIand XhoI-digested pPolI plasmid. The resultant plasmid was designated pPolI-PB1-K391E/E581G/A661T. To construct the pPolI-PB2-N265S plasmid, we amplified two DNA fragments corresponding to the PB2-coding sequence by PCR using primers Pol1-BsaI-for and PB2-N256S-rev, or PB2-N256S-for and Pol1-BsaI-rev (Table S3), with pPolI-PB2-wild-type as the PCR template. The full-length PB2-N265S gene was amplified by PCR using primers Pol1-BsaIfor and Pol1-BsaI-rev. The PCR product was digested using BsaI and cloned into the BsmBI-digested pPolI plasmid. The resultant plasmid was designated pPolI-PB2-N265S.
## Generation of recombinant influenza viruses
Recombinant influenza viruses were generated using the reverse-genetics system. 293T cells were transfected with eight segmented viral RNAs and viral protein expres sion plasmids using Trans-IT transfection reagent (Mirus). Twenty-four hours post-trans fection, the culture medium was changed to Opti-MEM I (Thermo Fisher Scientific) containing 3.5 µg/mL N-p-tosyl-L-phenylalanine chloromethyl ketone-treated trypsin (Sigma-Aldrich). After incubation for 48 h at 34°C, the cell culture supernatant was collected. The seed virus titer in the supernatant was determined using a plaque assay; then, 200 PFU of virus was inoculated into 10-to 11-day-old embryonated chicken eggs to amplify the recovered viruses.
## Plaque assay
The virus titer was determined using a plaque assay as described previously (48,49). Briefly, 1 mL aliquots of serial 10-fold dilutions of viruses were inoculated into MDCK cells seeded in 6-well or 12-well plates. After 1 h incubation, each well was overlaid with 3 mL or 1.5 mL of a MEM (Sigma-Aldrich) and 0.8% agarose (Sigma-Aldrich) mixture containing 0.2% bovine albumin (Sigma-Aldrich), 1× vitamin solution (Gibco), 1× MEM amino acid solution (Gibco), 4 µg/mL N-p-tosyl-L-phenylalanine chloromethyl ketonetreated trypsin, and 10 U/mL penicillin-streptomycin (Gibco). The number of plaques was counted following Amido Black 10B (Fujifilm Wako Pure Chemical Corporation) staining or immunostaining method after 3 to 5 days of inoculation. The virus titer was calculated as PFU/mL.
## HA assay and HAI assay
A HA assay and HAI assay were conducted with 0.5% chicken or 1% guinea pig red blood cells (Nippon Bio-Test laboratories) using the standard method (50).
## Immunostaining of virus plaque in 12-well or 24-well plate
MDCK cells in collagen-coated 12-well or 24-well tissue culture plates (AGC TECHNO GLASS) were washed with serum-free DMEM and then infected with 1 to 20 PFUs of the influenza virus. After virus adsorption at 34°C for 1 h, the cells were washed with serum-free DMEM and then overlaid with agarose medium as described above for methods of plaque assay. After 3 days of incubation, the cells were fixed with 4% formaldehyde solution for 1 h at 25°C and then washed with PBS containing 0.1% Triton X-100 (Sigma-Aldrich). The cells were blocked with Blocking One (Nacalai Tesque) with PBS containing 0.1% Triton X-100 for 15 min, and then probed with anti-NP mouse monoclonal antibody (ab128193, Abcam) for 30 min. The cells were washed with PBS containing 0.1% Triton X-100 and incubated with IRDye 800CW Goat anti-Mouse IgG (LI-COR Biosciences) and 0.1 mM CellTag700 (LI-COR Biosciences) for 30 min. After washing with PBS, the virus plaque was detected with the Odyssey CLx Infrared Imaging System (LI-COR Biosciences).
## Western blotting
293T cells were lysed in 20 mM Tris-HCl (pH 7.9), 100 mM NaCl, and 0.1% Triton X-100. After sonication, homogenates were centrifuged 14,000 × g at 4°C for 5 min, and the supernatant fractions were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (pore size 0.45 µm, Merck Millipore). The membranes were blocked with Blocking One (Nacalai Tesque) in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) and probed with anti-PB1 rabbit polyclonal antibody (51), anti-β-actin rabbit polyclonal antibody (PM053, Medical & Biological Laboratories), anti-β-actin mouse monoclonal antibody (M177-3, Medical & Biological Laboratories). The membranes were washed with TBS-T and incubated with IRDye 800CW Goat anti-Rabbit IgG (LI-COR Biosciences) or IRDye 680RD Goat anti-mouse IgG (LI-COR Biosciences). After washing with TBS-T, the proteins were detected with the Odyssey CLx Infrared Imaging System.
## Analyses of nucleic acid substitutions
Viral RNA was purified from the virions using QIAamp Viral RNA Mini Kit (Qiagen), and cDNA derived from PB1, PB2, PA, and NP, which constitute the viral ribonucleoprotein complex, was generated using specific primers and SuperScript III One-Step RT-PCR System with Platinum Taq DNA Polymerase (Thermo Fisher Scientific). After ampli con generation, Sanger sequencing was performed using SeqStudio Genetic Analyzer (Thermo Fisher Scientific).
## Mouse immunization and challenge assay
Six-week-old C57BL/6J female mice were immunized intranasally with 20 µL of PR8-PB1-K471H, PR8-PB1-K471P, or PR8-FluMist viruses (10 6 PFU/mouse). As a control, mice in the non-immunized group were injected with the same volume of PBS. On day 28, the animals were challenged with 10 7 PFU of PR8-wild-type virus. Weight loss and mortality were monitored for 14 days after challenge. Mice with body weight loss of more than 25% of their baseline body weight were euthanized. Lungs were collected on days 2, 4, and 6 post-infection and frozen at -80°C in the absence of buffer. Mice were humanely euthanized at the end of the observation period, or at designated time points for tissue collection.
## Virus titration in lung tissues
Lungs were thawed, weighed, and then homogenized in 1 mL of MEM (Sigma-Aldrich) containing 0.2% bovine albumin (Sigma-Aldrich), 1× vitamin solution (Gibco), 1× MEM amino acid solution (Gibco), and 10 U/mL penicillin-streptomycin (Gibco) by using a beads crusher µT-12 (TAITEC) at 2,600 rpm for 1 min. Homogenates were centrifuged (8,000 × g at 4°C for 10 min) to remove debris, and virus titers in cleared homogenate supernatants were determined by plaque assay. Virus titers were normalized to PFU per gram (g) of lung tissue.
## Generation of a mouse-adapted 6:2 reassortant virus
The HA and NA genes of the A/Hong Kong/MA(mouse adapted)/1968/H3N2 virus (GenBank accession number of HA, CY112249.1: GenBank accession number of NA, HM641200.1) were amplified by PCR to construct pPolI-maH3N2-HA and pPolI-maH3N2-NA plasmids, and we subsequently generated a PR8-based 6:2 reassortant virus by using the reverse-genetics system, designated PR8-maH3N2(6:2). To the constructed pPolI-maH3N2-HA plasmid, we amplified full-length maH3N2-HA sequence by PCR using primers Pol1-BsaI-for-2 and Pol1-BsaI-rev (Table S3) with synthetic DNA coding HA 5′-fragment (maH3N2-HA-1, 1 to 856) and HA 3′-fragment (maH3N2-HA-2, 830 to 1,765) (gBlocks Gene Fragments) (Table S4) as the PCR template. The PCR product was digested using BsaI and cloned into the BsmBI-digested pPolI plasmid. The resultant plasmid was designated pPolI-maH3N2-HA. To the constructed pPolI-maH3N2-NA plasmid, we amplified full-length maH3N2-NA sequence by PCR using primers Pol1-BsmB-for and Pol1-BsmB-rev (Table S3) with synthetic DNA coding maH3N2-NA (gBlocks Gene Fragments) (Table S4) as the PCR template. The PCR product was digested using BsmBI and cloned into the BsmBI-digested pPolI plasmid. The resultant plasmid was designated pPolI-maH3N2-NA. To create the 6:2 reassortant virus, eight pPolI plasmids (to express viral RNAs encoding HA and NA of maH3N2 and to express six internal proteins of PR8) were cotransfected with viral protein expression plasmids into 293T cells. The superna tant was collected at 48 h post-transfection and was infected into 10-or 11-day-old embryonated chicken eggs to amplify the recovered viruses. The resultant recombinant strain was designated PR8-maH3N2(6:2) virus.
## ELISpot assay
To detect antigen-specific cytotoxic T lymphocytes, we used the mouse IFN-γ ELISpot Kit (Cellular Technology Limited) to measure the number of IFN-γ-producing spleen cells in response to stimulation with influenza viral peptides. The splenocytes were plated at 1 × 10 5 cell/well or 2 × 10 5 cell/well onto IFN-γ antibody-coated ELISpot plates and stimulated with 0.25 µg/mL of NP 366-374 peptide (MHC Pentamer, Pro5, H-2Db, ASNENMETM, ProImmune) or 0.25 µg/mL of PA 224-233 peptide (MHC Pentamer, Pro5, H-2Db, SSLENFRAYV, ProImmune) for 24 h at 37°C, respectively. Spots derived from production of IFN-γ were visualized according to the manufacturer's protocol, and plates were read using ImmunoSpot S6 Universal M2 analyzer (Cellular Technology Limited) and ImmunoCapture 7.0 software (Cellular Technology Limited). SFUs were enumerated using ImmunoSpot 7.0 software (Cellular Technology Limited).
## Statistical analysis
Statistical analyses were performed with Microsoft Excel and KaleidaGraph 5 (HULINKS) software. Quantified luciferase activity was plotted as mean ± SD. Details of statistical tests, sample sizes as well as P-values are described for each experiment in respective figure legends. For mice body weight data, both mean and SEM were plotted. Differences in survival curves of infected mice were analyzed using the log-rank test.
## References
1. Miller, Montomoli, Leshem et al. (2024) "Seasonal influenza vaccines: variability of immune responses to B lineage viruses" *Hum Vaccin Immunother*
2. (1080)
3. Velthuis, Grimes, Fodor (2021) "Structural insights into RNA polymerases of negative-sense RNA viruses" *Nat Rev Microbiol*
4. Wandzik, Kouba, Cusack (2020) "Structure and function of influenza polymerase. Cold Spring Harb Perspect Med"
6. Wandzik, Kouba, Karuppasamy et al. (2020) "A structure-based model for the complete transcription cycle of influenza polymerase" *Cell*
7. Fan, Walker, Carrique et al. (2019) "Structures of influenza A virus RNA polymerase offer insight into viral genome replication" *Nature*
8. Ng, Arnold, Cameron (2008) "Structure-function relationships among RNA-dependent RNA polymerases" *Curr Top Microbiol Immunol*
9. Venkataraman, Prasad, Selvarajan (2018) "RNA dependent RNA polymerases: insights from structure, function and evolution" *Viruses*
10. Jácome, Becerra, De León et al. (2015) "Structural analysis of monomeric RNA-dependent polymerases: evolutionary and therapeutic implications" *PLoS One*
11. Ferrer-Orta, Arias, Escarmís et al. (2006) "A comparison of viral RNA-dependent RNA polymerases" *Curr Opin Struct Biol*
12. Mori, Ohniwa, Takizawa et al. (2021) "Development of a genetically stable live attenuated influenza vaccine strain using an engineered high-fidelity viral polymerase" *J Virol*
13. Martínez-Sobrido, Peersen, Nogales (2018) "Temperature sensitive mutations in influenza a viral ribonucleoprotein complex responsible for the attenuation of the live attenuated influenza vaccine" *Viruses*
14. Naito, Shirai, Mori et al. (2019) "Tyr82 amino acid mutation in PB1 polymerase induces an influenza virus mutator phenotype" *J Virol*
15. Cheung, Watson, Choy et al. (2014) "Generation and characterization of influenza A viruses with altered polymerase fidelity" *Nat Commun*
16. Naito, Mori, Ushirogawa et al. (2017) "Generation of a genetically stable high-fidelity influenza vaccine strain" *J Virol*
17. Subbarao (2020) "Live attenuated cold-adapted influenza vaccines. Cold Spring Harb Perspect Med"
18. Ghendon, Polezhaev, Lisovskaya et al. (1984) "Recombinant cold-adapted attenuated influenza A vaccines for use in children: molecular genetic analysis of the coldadapted donor and recombinants" *Infect Immun*
19. Maassab (1967) "Adaptation and growth characteristics of influenza virus at 25 degrees c" *Nature*
20. Zhou, Li, Speer et al. (2012) "Engineering temperature sensitive live attenuated influenza vaccines from emerging Full-Length Text Journal of Virology December"
21. "Vaccine (Auckl)" *viruses*
22. Zhou, Meliopoulos, Lin et al. (2016) "Reversion of coldadapted live attenuated influenza vaccine into a pathogenic virus" *J Virol*
23. Jin, Zhou, Lu et al. (2004) "Imparting temperature sensitivity and attenuation in ferrets to A/Puerto Rico/8/34 influenza virus by transferring the genetic signature for temperature sensitivity from coldadapted A/Ann Arbor/6/60" *J Virol*
24. Nogales, Rodriguez, Dediego et al. (2017) "Interplay of PA-X and NS1 proteins in replication and pathogenesis of a temperature-sensitive 2009 pandemic H1N1 influenza A virus" *J Virol*
25. Cox, Baker, Nogales et al. (2015) "Development of a mouse-adapted live attenuated influenza virus that permits in vivo analysis of enhancements to the safety of live attenuated influenza virus vaccine" *J Virol*
26. Pena, Vincent, Ye et al. (2011) "Modifications in the polymer ase genes of a swine-like triple-reassortant influenza virus to generate live attenuated vaccines against 2009 pandemic H1N1 viruses" *J Virol*
27. Jin, Lu, Zhou et al. (2003) "Multiple amino acid residues confer temperature sensitivity to human influenza virus vaccine strains (flumist) derived from cold" *Virology (Auckl)*
28. Treanor, Perkins, Battaglia et al. (1994) "Evaluation of the genetic stability of the temperature-sensitive PB2 gene mutation of the influenza A/Ann Arbor/6/60 cold-adapted vaccine virus" *J Virol*
29. Costa, Sausset, Munier et al. (2015) "Temperature-sensitive mutants in the influenza A virus RNA polymerase: alterations in the PA linker reduce nuclear targeting of the PB1-PA dimer and result in viral attenuation" *J Virol*
30. Kawaguchi, Naito, Nagata (2005) "Involvement of influenza virus PA subunit in assembly of functional RNA polymerase complexes" *J Virol*
31. Gordon, King (1994) "Genetic properties of temperature-sensitive folding mutants of the coat protein of phage P22" *Genetics*
32. Kouba, Drncová, Cusack (2019) "Structural snapshots of actively transcribing influenza polymerase" *Nat Struct Mol Biol*
33. Pflug, Guilligay, Reich et al. (2014) "Structure of influenza A polymerase bound to the viral RNA promoter" *Nature*
34. Grohskopf, Sokolow, Broder et al. (2016) "Prevention and control of seasonal influenza with vaccines" *MMWR Recomm Rep*
35. Blanco-Lobo, Nogales, Rodríguez et al. (2019) "Novel approaches for the development of live attenuated influenza vaccines" *Viruses*
36. Nogales, Martínez-Sobrido (2016) "Reverse genetics approaches for the development of influenza vaccines" *Int J Mol Sci*
37. Wu, Zhou, Sheng et al. (2022) "Defective interfering particles of influenza virus and their characteristics, impacts, and use in vaccines and antiviral strategies: a systematic review" *Viruses*
38. Yeh, Smith, Carlyle et al. (2023) "Genetic stabilization of attenuated oral vaccines against poliovirus types 1 and 3" *Nature*
39. Wahid, Mercer, Macadam et al. (2021) "Assessment of genetic changes and neurovirulence of shed Sabin and novel type 2 oral polio vaccine viruses" *NPJ Vaccines*
40. Wahid, Mercer, Gast et al. (2022) "Evaluating stability of attenuated Sabin and two novel type 2 oral poliovirus vaccines in children" *NPJ Vaccines*
41. Wahid, Mercer, Leon et al. (2022) "Genetic and phenotypic stability of poliovirus shed from infants who received novel type 2 or Sabin type 2 oral poliovirus vaccines in Panama: an analysis of two clinical trials" *Lancet Microbe*
42. De Coster, Leroux-Roels, Bandyopadhyay et al. "Van Damme P. 2021. Safety and immunogenicity of two novel type 2 oral poliovirus vaccine candidates compared with a monovalent type 2 oral poliovirus vaccine in healthy adults: two clinical trials" *Lancet*
43. Konz, Schofield, Carlyle et al. (2021) "Evaluation and validation of nextgeneration sequencing to support lot release for a novel type 2 oral poliovirus vaccine" *Vaccine X*
44. Ping, Lopes, Nidom et al. (2015) "Development of high-yield influenza A virus vaccine viruses" *Nat Commun*
45. Tian, Landreth, Lu et al. (2021) "A replication-defective influenza virus harboring H5 and H7 hemagglutinins provides protection against H5N1 and H7N9 infection in mice" *J Virol*
46. Dey, Ahuja, Panwar et al. (2023) "Immune control of avian influenza virus infection and its vaccine development" *Vaccines (Basel)*
47. Koutsakos, Kedzierska, Subbarao (2019) "Immune responses to avian influenza viruses" *J Immunol*
48. Talaat, Luke, Khurana et al. (2014) "A live attenuated influenza A(H5N1) vaccine induces long-term immunity in the absence of a primary antibody response" *J Infect Dis*
49. Neumann, Watanabe, Ito et al. (1999) "Generation of influenza A viruses entirely from cloned cDNAs" *Proc Natl Acad Sci*
50. Turan, Mibayashi, Sugiyama et al. (2004) "Nuclear MxA proteins form a complex with influenza virus NP and inhibit the transcription of the engineered influenza virus genome" *Nucleic Acids Res*
51. Tobita, Sugiura, Enomote et al. (1975) "Plaque assay and primary isolation of influenza A viruses in an established line of canine kidney cells (MDCK) in the presence of trypsin" *Med Microbiol Immunol*
52. Gaush, Smith (1968) "Replication and plaque assay of influenza virus in an established line of canine kidney cells" *Appl Microbiol*
53. (2011) "Manual for the laboratory diagnosis and virological surveillance of influenza"
54. Naito, Momose, Kawaguchi et al. (2007) "Involvement of Hsp90 in assembly and nuclear import of influenza virus RNA polymerase subunits" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12282438&blobtype=pdf | # Elevated interferon-induced transmembrane protein 3 in platelets and megakaryocytes suppresses Crimean-Congo hemorrhagic fever viral infection by interacting with glycoprotein Gc
Jingyuan Zhang, Yaohui Fang, Chenhui Lin, Xiaoli Wu, Chaoxiong Yue, Fei Deng, Shu Shen, Virologica Sinica
## Abstract
Crimean-Congo hemorrhagic fever (CCHF) is a hemorrhagic fever caused by infection with the CCHF virus (CCHFV) and has a mortality rate of up to 30 %. Thrombocytopenia is a hallmark of CCHF; however, the mechanisms underlying this manifestation remain poorly understood. In addition to hemostasis, platelets play a crucial role in recognizing pathogens and mediating immune responses. We investigated the mechanisms underlying thrombocytopenia associated with CCHFV infection by analyzing the platelet transcriptome in mice. Interferon-induced transmembrane protein 3 (IFITM3), a known antiviral factor, was significantly upregulated. The role of IFITM3 in response to CCHFV infection was characterized using the human megakaryoblast cell line MEG-01, considered a parental cell line of platelets. Although the CCHFV infection rate was limited, MEG-01 cells maintained the infection and replication of CCHFV, leading to increased IFITM3 protein expression. We demonstrated that IFITM3 overexpression efficiently inhibited CCHFV infection, whereas IFITM3 knockout promoted viral infection. An interaction between IFITM3 and the CCHFV glycoprotein Gc was identified, which suppressed CCHFV entry into cells. The IFITM3 CIL-TMD domain is critical for this interaction. These results suggest that IFITM3 is a restriction factor and plays an antiviral role during CCHFV infection. Elevated expression of IFITM3 in platelets indicates that this could be a common mechanism by which platelets protect against viruses, including CCHFV, which may reduce platelet consumption and destruction caused by CCHFV infection. These findings provide valuable insights into the pathogenesis of CCHF-associated thrombocytopenia and offer foundational theoretical support for future therapeutic strategies.
## INTRODUCTION
Crimean-Congo hemorrhagic fever virus (CCHFV), primarily transmitted by ticks of the genus Hyalomma, was first identified in patients in 1956 (Akinci et al., 2013). In 1969, the causative pathogen of Crimean-Congo hemorrhagic fever (CCHF) disease in humans was found to cause disease in humans with fatality rates ranging from 10% to 50%, which was higher in some outbreaks (Fereidouni et al., 2023;Haddock et al., 2018;Aslam et al., 2023). The disease manifests with a range of severe symptoms, including fever, headache, fatigue, dizziness, gastrointestinal distress, and, more critically, hemorrhagic manifestations such as petechiae, ecchymosis, and bleeding from mucous membranes (Bodur et al., 2012;Akinci et al., 2016). A hallmark feature of CCHF is thrombocytopenia (Ergonul, 2012;Cross et al., 2020), which contributes to hemorrhagic complications. The World Health Organization has recognized CCHF as a priority infectious disease owing to its widespread endemic nature, potential to cause outbreaks in new geographical regions, and the absence of specific antiviral treatments (Mcentire et al., 2021). Consequently, there is an urgent need to better elucidate virus-host interactions to advance our understanding of disease pathogenesis, which would further support the development of antiviral strategies (Bente et al., 2013).
CCHFV does not cause disease in immunocompetent adult rodents, and the only available models are neonatal mice and rats (Chumakov et al., 1968;Hoogstraal, 1979). In recent years, CCHFV infection and lethal/severe disease models have been successfully established using IFNAR À/À or STAT-1 À/À mice, which have been used to evaluate viral replication, immune responses, the efficacy of antivirals and monoclonal antibodies, and organ pathology (Tignor and Hanham, 1993;Zivcec et al., 2013;Lindquist et al., 2018;Aligholipour Farzani et al., 2019). Although CCHFV infection does not lead to fatal outcomes in immunocompetent mice, it still causes acute infection starting from the third day, as viral loads in various organs and the release of inflammatory cytokines were detected and continued to increase (Garrison et al., 2019;Tipih et al., 2023). These animal models are essential for understanding disease pathophysiology and developing potential therapeutic interventions. Because bleeding is a major feature of CCHF, it is necessary to evaluate blood-related biomarkers, including platelet counts and coagulation factors, in animal models to understand the mechanisms underlying CCHF-induced bleeding.
Platelets, key players in the coagulation cascade, also play a significant role in antiviral immune responses. Recent studies have shown that platelet function can be altered during viral infections, suggesting their contribution to host defense against pathogens (Dib et al., 2020). Dengue virus infection can induce platelet aggregation and activation, which promotes viral clearance and immune complex formation (Simon et al., 2015;Quirino-Teixeira et al., 2022). Similarly, during influenza virus infection, platelet-derived factors can enhance immune responses and viral clearance (Bote et al., 2022). Moreover, megakaryocytes, the progenitors of platelets, possess intrinsic immune recognition systems that can respond to viral infections. Recent evidence suggests that megakaryocytes can modulate antiviral immune responses through the release of pro-inflammatory cytokines and production of interferons, potentially playing a pivotal role in controlling viral replication (Campbell et al., 2019). These insights highlight the importance of platelets and megakaryocytes in the immune response to viral infections.
In this study, we characterized functional changes in platelets from CCHFV-infected experimental mice. Based on the transcriptome data, we found that interferon-induced transmembrane protein 3 (IFITM3) was significantly upregulated, suggesting its role as a key factor in protecting platelets against CCHFV infection. We evaluated IFITM3's role in suppressing CCHFV infection using a human megakaryoblast cell line with IFITM3 overexpression or knockout. The interaction between IFITM3 and the CCHFV glycoprotein, as well as the domains critical for this interaction, was further investigated. These results demonstrate an important role of platelets in the antiviral response against CCHFV infection and suggest that IFITM3 functions as a key antiviral factor by limiting viral entry through glycoprotein interactions. These findings highlight the critical role of IFITM3 in CCHFV-host interactions and antiviral activity, providing new mechanistic insights into its function.
## RESULTS
## Transcriptome of mouse platelets is altered during CCHFV infection
To investigate changes in platelets in CCHFV infection, we infected C57BL/6J mice with CCHFV. The infected C57BL/6J mice developed hemolysis and aplastic anemia (Supplementary Fig. S1). Routine blood examinations during the acute phase of infection (day 3 post-inoculation) revealed significant alterations, including decreased red blood cell count (RBC), hemoglobin (HGB), hematocrit (HCT), mean corpuscular volume (MCV), and low fluorescence reticulocyte (LFR%), and increased mean corpuscular hemoglobin concentration (MCHC), red cell distribution width-coefficient of variation (RDW-CV%), reticulocyte count (absolute) (RET#), immature reticulocyte fraction (absolute) (IRF%), and high fluorescence reticulocyte (HFR%), indicating hemolytic or regenerative anemia (Supplementary Fig. S1A). Elevated neutrophil counts and decreased lymphocyte levels suggested an acute-phase response to CCHFV infection (Supplementary Fig. S1B). Viral infection caused a marked reduction in the platelet count and an increase in platelet large cell ratio (P-LCR) and mean platelet volume (MPV), indicating platelet destruction and enhanced platelet production (Supplementary Fig. S1C). Accordingly, we observed a significant increase in plasma thrombopoietin (TPO), a glycoprotein hormone responsible for regulating platelet production and maintaining normal platelet levels (Supplementary Fig. S1D). This increase is likely associated with severe platelet depletion following CCHFV infection and subsequent acceleration of platelet production in megakaryocytes. Moreover, likely due to the early time points of sampling (3 days post-infection), we did not observe significant changes of platelet activation or coagulation in the mice with CCHFV infection, as exhibited by the reduced plasma CD62P and unchanged D-dimer levels in comparison to the control group (Supplementary Fig. S1D).
Platelets were purified from these mice and used for RNA transcriptome analysis. The results showed that the platelets from the CCHFV-infected groups formed clusters distinct from those in the control groups (Fig. 1A), suggesting a functional deviation of platelets caused by CCHFV infection compared to healthy mice. KEGG pathway analysis indicated that CCHFV infection induced functional alterations in platelets, including actin cytoskeleton regulation, platelet activation, and platelet adhesion (Fig. 1B). Additionally, platelets exhibit functional changes related to responses in other diseases, such as COVID-19 and prion disease. GO functional annotation of the differentially expressed genes (DEGs) showed significant enrichment in pathways related to protein phosphorylation, catabolism, ribonucleoprotein complex assembly, and cytoplasmic transport. These findings suggested that platelets stimulated during an antiviral response generate downstream signals that accelerate protein metabolism (Fig. 1C). A total of 1251 DEGs were identified at the nominal significance level (P < 0.05). Of these, 639 remained significant after stringent multiple-comparison adjustment (false discovery rate, P < 0.05); 372 transcripts were upregulated and 267 were downregulated. Notably, we found that the transcription of mIfitm3 was significantly upregulated, ranking among the top 20 genes exhibiting significant differences in transcription levels compared to the control (Fig. 1D andE). Consistent with increased transcription, elevated IFITM3 expression was confirmed in the platelets of CCHFV-infected mice (Fig. 1F). These results suggest that CCHFV infection led to the loss of platelet numbers and alterations in platelet function, although platelet activation was not significant.
## Characterization of human megakaryocytes infected by CCHFV
Megakaryocytes, the progenitor cells that produce platelets, possess biological features similar to those of platelets (Machlus and Italiano, 2013;Koupenova et al., 2022). We characterized the infection and growth properties of CCHFV in the human megakaryoblast cell line MEG-01 and compared to those in HEK293 cells that express lower levels of hifitm3. Flow cytometry analysis revealed that 2.6%-2.78% of MEG-01 cells were infected with CCHFV, less than that observed in HEK293 cells (41.1%-57.3%) (Fig. 2A). The expression of CCHFV NP was detected in MEG-01 cells at 48 h post-infection (p.i.), with a slight increase thereafter. In contrast, a high level of CCHFV nucleoprotein (NP) expression was observed in HEK293 cells at 24 h p.i., and was maintained during the subsequent period examined (Fig. 2B). Additionally, we examined IFITM3 protein expression in both cell lines after CCHFV infection. Although constitutive expression of IFITM3 was observed in MEG-01 cells, CCHFV infection resulted in increased IFITM3 expression at 24 h p.i., before CCHFV NP blotting. Unlike MEG-01, which exhibited detectable IFITM3 expression at 0 h p.i., IFITM3 was undetectable in HEK293 cells. Nevertheless, IFITM3 expression was detected at 24 h p.i. and increased thereafter (Fig. 2B). Subsequently, the increase in hifitm3 transcription in both MEG-01 and HEK293 cells after CCHFV infection was confirmed using qRT-PCR. Compared with the control group, IFITM3 transcription increased 2.21 AE 0.62-fold in CCHFV-infected MEG-01 cells (P < 0.05), while HEK293 cells showed an even greater increase, peaking at 48 h p.i. with a 208.0 AE 47.92-fold elevation relative to controls (P < 0.001) (Fig. 2C). At 96 h p.i., the culture supernatants from MEG-01 cells (caption on next page) contained 7.96 Â 10 7 AE 2.81 Â 10 7 copies/mL of viral RNA which was lower than that from HEK293 cells (8.91 Â 10 8 AE 3.99 Â 10 8 copies/mL) at the same time point, suggesting fewer progeny viruses were produced in MEG-01 cells compared to HEK293 cells. Similarly, the viral copies in MEG-01 cells (1.20 Â 10 6 AE 1.32 Â 10 5 copies/10 3 cells) at 96 h p.i. were also lower than those in HEK293 cells (2.64 Â 10 6 AE 4.01 Â 10 5 copies/10 3 cells) (Fig. 2D). Accordingly, CCHFV was detected in only a limited proportion of MEG-01 cells, showing a lower infection rate compared to HEK293 cells (Fig. 2E). These results suggest that, despite being less sensitive to CCHFV than HEK293 cells, MEG-01 cells could still be infected and maintain CCHFV replication to produce progeny viruses. Furthermore, CCHFV infection upregulated IFITM3 expression in both cell lines.
## Overexpression of IFITM3 restricts infection of CCHFV
Previous studies have demonstrated that IFITM3 serves as an antiviral immune protein that suppresses various viral infections (Campbell et al., 2019;Klein et al., 2023;Du et al., 2024). Since IFITM3 was undetectable in HEK293 cells (Supplementary Fig. S2A andB), we generated an IFITM3-overexpressing cell line (HEK293-hifitm3-OE). Accordingly, MEG-01 cells overexpressing IFITM3 were generated (MEG-01-hi-fitm3-OE) (Supplementary Fig. S3A andB). We found that CCHFV infection was suppressed in MEG-01-hifitm3-OE cells, as shown by reduced CCHFV NP expression and decreased viral RNA copies in MEG-01-hifitm3-OE cells (Fig. 3A andB) as well as fewer cells blotted with the antibody against CCHFV NP (Fig. 3C). Similarly, IFITM3's inhibitory effect was confirmed in HEK293-hifitm3-OE cells (Fig. 3D-F). High levels of IFITM3 expression significantly reduced both CCHFV NP expression (Fig. 3D) and viral RNA copies in the supernatants (Fig. 3E, left). CCHFV copies also decreased at 24 h and slightly increased at 48 h. Although there was no significant difference, it suggested that IFITM3 overexpression may postpone the intracellular replication of CCHFV in HEK293 cells (Fig. 3E, right). As expected, CCHFV infection was immunostained in very few HEK293-hifitm3-OE cells (Fig. 3F). These results demonstrate that IFITM3 exerts a strong inhibitory effect on CCHFV infection in both MEG-01 and HEK293 cells.
## Knockout of IFITM3 promotes infection of CCHFV
Subsequently, together with MEG-01, the HeLa cell line was selected to generate IFITM3-knockout cells, as HeLa exhibited a higher expression of IFITM3 than MEG-01 cells (Supplementary Fig. S2A andB), resulting in MEG-01-hifitm3-KO and HeLa-hifitm3-KO cells, respectively (Supplementary Fig. S3C andD), In either MEG-01-hifitm3-KO cells or HeLa-hifitm3-KO cells, the knockout of hifitm3 led to enhanced infection with CCHFV. We detected increased CCHFV NP expression (Fig. 4A andD), elevated CCHFV RNA copies in cells and supernatants (Fig. 4B andE), and enhanced immunostaining of cells with the CCHFV NP antibody (Fig. 4C andF). These results demonstrate that in the absence of IFITM3, CCHFV infection is enhanced, further supporting that IFITM3 presents an inhibitory effect on CCHFV infection.
## IFITM3 suppresses CCHFV entry into cells via interacting with CCHFV Gc
We hypothesized that human IFITM3 might interact with CCHFV proteins. Since MEG-01 cells exist in a semi-adherent/semi-suspended state in the culture supernatant, resulting in cell loss during transfection experiment, both MEG-01 and HEK293 cell lines exhibit limited IFITM3 baseline expression, and overexpression of IFITM3 showed an inhibitory effect on CCHFV infection. Therefore, we performed pulldown assays using lysates from HEK293 cells transfected with an IFITM3-expressing plasmid and infected with CCHFV. CCHFV Gc specifically precipitated with IFITM3, suggesting an interaction between IFITM3 and CCHFV Gc other than with Gn and NP (Fig. 5A). Additionally, the co-localization of IFITM3 and Gc was observed in the cytoplasm of HEK293 cells infected with CCHFV (Fig. 5B). We subsequently analyzed the effects of IFITM3 on CCHFV entry into host cells. Overexpression of IFITM3 in MEG-01 cells significantly reduced the amount of virus bound to the cell surface and internalized into the cells (Fig. 5C).
## CIL-TMD domain is critical for the interaction between IFITM3 and Gc
Subsequently, the interaction between IFITM3 and CCHFV Gc was confirmed by pull-down analysis using lysates from HEK293T cells cotransfected with plasmids expressing both proteins (Fig. 6A). The truncated IFITM3 constructs were generated according to a previous study (Xu et al., 2022), including NTD, IMD-CIL, and CIL-TMD, which were fused with EGFP (Fig. 6B). Moreover, the co-localization of IFITM3 and CCHFV Gc was observed in co-transfected cells (Fig. 6C). Pull-down analyses were performed using cells co-transfected with plasmids expressing IFITM3 truncates and CCHFV Gc. The results showed that both IMD-CIL and CIL-TMD could precipitate CCHFV Gc, with higher efficiency for CIL-TMD than for IMD-CIL (Fig. 6D). Moreover, similar to full-length IFITM3, CIL-TMD co-localized with CCHFV Gc in the cytoplasm (Fig. 6E). Computational modeling using AlphaFold 3 and PDBe-PISA v1.52 predicted an interaction complex where IFITM3 binds CCHFV Gc primarily through hydrogen bonds involving key residues in the CIL (Asp92), TMD (Cys105), and CTD (Tyr132) domains (Fig. 6F; Supplementary Table S1).
## DISCUSSION
CCHF, caused by CCHFV infection, is a tick-borne viral hemorrhagic fever with high human fatality rates (Mo et al., 2023). However, few studies have focused on the mechanisms underlying coagulation Fig. 1. Identification of differentially expressed genes in platelets of CCHFV-infected mice. Three groups of C57BL/6J mice (6-8 weeks old, female, n ¼ 10/group) were inoculated with CCHFV (1.56 Â 10 5 TCID 50 /mouse) intraperitoneally, and two mock groups (n ¼ 10/group) received brain homogenate in the same route and served as control groups. On day 3 post-inoculation, mice were euthanized and blood was collected. Platelets were then purified and applied to transcriptome sequencing. A Hierarchical clustering of global gene expression data revealed that samples from the CCHFV-infected group (cyan) were closely clustered, and were distinct from those of the healthy group (pink). The data are presented by the Pearson correlation coefficient (Pearson's r) between different groups. B KEGG annotation and expression analysis of the genome. The ten pathways with the smallest adjusted P-values (p.adjust) are shown. Pathways were sorted according to gene ratio. C GO enrichment analysis highlighting representative functional categories enriched among differentially expressed genes (DEGs) in CCHFV-infected mice (P < 0.01). The ten entries with the smallest p.adjust from the three GO categories (CC: Cell Component; MF: Molecular Function; BP: Biological Process) are shown. D Volcano plot showing the differential expression of genes (DEGs) in platelets of CCHFV-infected mice in comparison to healthy controls. Red dots represent upregulated genes, blue dots indicate downregulated genes, and grey dots show genes with no significant differences. The top five genes exhibiting increased transcription levels the most significantly more than the other genes are labelled. E The heatmap exhibited the top 20 DEGs with a significantly increased or decreased transcription levels. The data has been normalized by row and is displayed as a z-score. F Validation of increased IFITM3 protein expression in platelets of infected mice via Western blot analysis, with GAPDH as the internal control. Three other groups of mice (n ¼ 10/group) were infected with CCHFV (Infection) in the same route as described above. Platelets were purified on day 3 post-isolation. After confirming platelet purity >97%, IFITM3 expression was examined by Western blot. Platelets from three groups of healthy mice inoculated with culture medium (Mock) were used as control, and GAPDH expression was blotted as inner control of platelets. infected with CCHFV at 48 h p.i. or with mock infection. Cells were immunostained using antibodies against CCHFV NP (red) and IFITM3 (green), respectively. Nuclei were stained with Hoechst33258 (blue). Scale bar, 50 μm. Differences between the two groups were analyzed using One-way ANOVA to calculate the P-value to evaluate the differences in expression levels at different time points in the same group. P < 0.05 was considered statistically significant. disorders caused by CCHFV infection. In the present study, we found that CCHFV-infected C57BL/6J mice did not exhibit any obvious clinical symptoms. However, they may develop hemolysis and aplastic anemia despite not showing significant platelet activation or coagulopathy. Platelets function in hemostasis and mediate immune responses. We subsequently characterized functional changes in platelets purified from infected mice using RNA transcriptome analysis. Functional alterations in platelets were mainly involved in platelet activation, adhesion, and regulation of the actin cytoskeleton. Significant changes were observed in the pathways associated with protein phosphorylation, catabolism, ribonucleoprotein complex assembly, and cytoplasmic transport. This demonstrates the important role of platelets in responding to CCHFV infection, probably more so than in hemostasis. Similarly, functional alterations were observed in platelets from mice with infection SFTSV (Fang et al., 2024). Despite the different functional changes based on transcriptomic analyses, coagulation, platelet activation, and immune modulation found in platelets of SFTSV-infected mice were also identified in platelets of CCHFV-infected mice in the current study. This may suggest a common response by platelets to tick-borne virus infections, while virus-specific responses are also worth considering.
IFITM3, an interferon-induced membrane protein, is considered an antiviral factor that inhibits viral entry and replication (Bailey et al., 2013(Bailey et al., , 2014) such as HCV (Yao et al., 2011;Narayana et al., 2015), HIV (Lu et al., 2011;Tartour et al., 2014;Compton et al., 2016), and influenza (Bailey et al., 2013;Ren et al., 2020). Our study revealed the significant increase of IFITM3 expression in platelets of CCHFV-infected mice, which was also supported by the previous study reporting the significant upregulation of IFITM3 in PBMCs of patients with CCHF, dengue, and influenza (Campbell et al., 2019;Neogi et al., 2022;Denz et al., 2024). This suggests that increased IFITM3 in platelets may play an important role in the antiviral responses commonly observed in viral diseases. Due to the limited ability to manipulate IFITM3 expression in platelets, we subsequently generated IFITM3 overexpressing and knockout cell lines using the human megakaryoblast cell line MEG-01. We characterized the effects of CCHFV infection on megakaryocytes and evaluated the role of IFITM3 in CCHFV infection using these generated cell lines, comparing them with other cell lines containing high (HeLa) or undetectable (HEK293) levels of IFITM3. Despite the less efficient infection of CCHFV in MEG-01 cells than that in HEK293 cells, MEG-01 cells maintained CCHFV proliferation to generate progeny viruses. CCHFV infection increased IFITM3 expression in both MEG-01 and HEK293 cells. This result contradicts the inhibitory effect of IFITM3 on CCHFV infection observed in MEG-01-hifitm3-OE and HEK293-hifitm3-OE cells infected with CCHFV. IFITM3 exerts its inhibitory effect by affecting the virus-host cell membrane fusion process and restricts the infection of various enveloped viruses during the early stages of viral entry (Zani and Yount, 2018). The generated cell lines, MEG-01-hifitm3-OE and HEK293-hifitm3-OE, already exhibited the overexpressed IFITM3, which inhibits CCHFV infection by suppressing its entry process. However, in HEK293 cells, IFITM3 expression was induced by CCHFV infection, and its expression was detected immediately after CCHFV NP was detected. Increased IFITM3 had a very limited effect on the inhibition of CCHFV infection. Because MEG-01 cells naturally express a certain level of IFITM3, CCHFV infection may be inhibited. As a result, the efficiency of CCHFV infection in MEG-01 cells is relatively low. Moreover, the inhibitory effect of IFITM3 on CCHFV infection was further confirmed in IFITM3-knockout MEG-01 and HeLa cells, as infection was significantly enhanced. CCHFV exhibited varied infection efficiencies, which can be attributed to the differing sensitivities of various cell lines of distinct tissue origins (Dai et al., 2021). So, the levels of IFN response stimulated by CCHFV infection may also vary, thus affecting the expression levels of IFITM3 (Scholte et al., 2017). Therefore, the inhibitory effect of IFITM3 on CCHFV infection may also depend on the infection efficiency of CCHFV in different cell lines.
Subsequently, the interaction between IFITM3 and CCHFV Gc was identified, which resulted in reduced efficiency of viral binding and internalization of CCHFV into MEG-01 cells. We postulated that the Cells were immunostained using antibodies against CCHFV NP (green) and IFITM3 (red), respectively, and nuclei were stained with Hoechst33258 (blue). Images were taken using a Leica confocal microscope. Scale bar, 20 μm. The data between groups were analyzed using two-way ANOVA. Data are expressed as the mean AE SD. P < 0.05 indicates a statistically significant difference, *P < 0.05, **P < 0.01, ****P < 0.0001. interaction between IFITM3 and CCHFV Gc suppresses the fusion process. However, because of the limited infection rates in MEG-01 cells, we were unable to characterize the effect of IFITM3 on syncytia formation. Nevertheless, our results demonstrate that IFITM3 plays a role in suppressing CCHFV entry, which is consistent with a previous study reporting that IFITM3 inhibits viral entry by interacting with SFTSV Gc (Denz et al., 2024;Du et al., 2024). CCHFV is a genetically diverse virus that is currently divided into seven major genotypes (I-VII), each with significant association with their geographical distributions (Deyde et al., 2006;Bente et al., 2013). It is possible that different strains or genotypes of CCHFV could affect the interaction with IFITM3, as there may be amino acid substitutions at key positions. Further studies are needed to characterize roles of key sites in the interaction between IFITM3 and CCHFV Gc. Furthermore, we found that IFITM3 interacts with CCHFV Gc and that the CIL-TMD region of IFITM3 is critical for this interaction. CCHFV enters cells via clathrin-mediated endocytosis. It can interact with inner membrane proteins in early endosomes, through which CCHFV RNA is released into host cells (Shtanko et al., 2014). IFITM3 is located on the endosomal membrane. Although the spatial topological structure of IFITM3 remains unclear, the CIL domain has been suggested to be located on the inner side of the endosome (Marziali et al., 2021). We speculated that elevated IFITM3 expression increases the possibility that CIL interacts with Gc during CCHFV entry. As a result, CCHFV internalization is suppressed, which reduces the infection rate of this virus.
IFITM3 is a downstream gene induced by interferon signaling. The expression level of IFITM3 depends on the activation levels of IFN signaling (Schoggins et al., 2011). Following viral infection, the interferon signaling pathway upregulates IFITM3 expression, further inhibiting viral replication, and potentially constituting a broad-spectrum antiviral mechanism in various cell types including platelets and megakaryocytes. IFITM3 has a dynamic bidirectional regulatory ability on the IFN signaling pathway. While it enhances the early antiviral response by promoting pattern recognition receptor (PRR) signaling and stabilizing signal transducer and activator of transcription 1 (STAT1) (Jia et al., 2012;Zhang, P. et al., 2024), it can conversely prevent immunopathological damage caused by excessive production of IFN-I by blocking MAVS aggregation and upregulating SOCS (Chen et al., 2017;Lin et al., 2020), making IFITM3 a key molecule for balancing antiviral efficacy and immune homeostasis. Thus, IFITM3 has been considered a target for the treatment of infection and autoimmune diseases (Li et al., 2025). In viral hemorrhagic fever viral diseases, such as CCHF, patients would develop severe thrombocytopenia, manifested by a significant decrease in platelet count (Akinci et al., 2013;Bente et al., 2013). Although the antiviral function of IFITM3 is not entirely dependent on the interferon pathway, it can enhance the overall antiviral state by maintaining ISG expression. Its direct regulatory effect on the interferon pathway is limited and cell type specific. The expression level of IFITM3 may significantly affect the balance of the IFN signaling pathway, thereby changing the homeostasis between antiviral immune efficacy and immunopathological damage (Brass et al., 2009;Dai et al., 2021), which may be attributed to the extensive damage or consumption of platelets caused by viral infection. The elevated expression of IFITM3 in platelets demonstrates potent inhibitory effects against CCHFV infection. However, the mechanistic basis of this antiviral activity, particularly whether IFITM3 modulates platelet-mediated immune regulation through type I interferon pathway potentiation to enhance antiviral functions, requires further elucidation.
This study has several limitations. While we performed platelet transcriptome analyses using infected mouse models, these may not completely reflect platelet responses in human patients. Nevertheless, the upregulation of IFITM3 observed in our transcriptome data has been reported in platelet transcriptomes across various diseases, suggesting its role in a common antiviral mechanism. No cases of CCHFV infection have been reported in China since 2005. Without clinical samples, we could not investigate the platelet responses induced by CCHFV infection, including coagulation disorders and immune-mediated responses. It For binding assays, cells were incubated with CCHFV on ice for 1 h and were washed three times with ice-cold PBS to remove unbound viruses. Cells were then harvested to determine CCHFV S RNA by qRT-PCR. For internalization assays, cells were incubated with CCHFV at 4 C for 1 h, then moved to 37 C and incubated for 1h to allow viral internalization. Cells were washed with PBS containing 500 ng/mL proteinase K to remove the uninternalized virus. Internalized viruses were quantified by qRT-PCR. GAPDH was blotted as an internal control. The CCHFV RNA loads were normalized to MEG-01 cells using 2 ÀΔΔCt method. The significant difference was determined using the student's t-test. P < 0.05 indicates a statistically significant difference, **P < 0.01, ***P < 0.001. would be valuable to compare these data with data from CCHFV-infected animal models and actual clinical cases to elucidate commonalities and differences if relevant public datasets become available. Despite these limitations, our findings shed light on the pathogenic mechanisms of CCHF and facilitate the development of antiviral strategies.
## CONCLUSIONS
This study reveals platelet functional alterations during CCHFV infection through transcriptomic analysis. We identified IFITM3 as one of the most significantly upregulated genes in platelets of mice with CCHFV infection and demonstrated its antiviral role in suppressing CCHFV infection using MEG-01 cells. Furthermore, the interaction of IFITM3 with the CCHFV Gc protein was identified, and the CIL-TMD domain played a key role in this interaction. The findings suggest that platelets play an important role in the antiviral response to suppress CCHFV infection via elevation of IFITM3, which may facilitate future research on the pathogenesis of CCHFV-induced thrombocytopenia.
## MATERIALS AND METHODS
## Virus, cells, and antibodies
The CCHFV strain YL16070 (GenBank accession: KY354082) (Guo et al., 2017;Dai et al., 2021) used in this study was deposited at the National Virus Resource Center (IVCAS 6.6329). Titers were determined using the Fig. 6. CIL-TMD binds to the CCHFV Gc. HEK293T cells were co-transfected with phMCV-Gc-S-tag together with the pEGFP-N1 plasmid expressing IFITM3 or the truncated IFITM3 fragments or the pEGFP-N1 plasmid as control (Vector). At 24 h post transfection, cells were collected and lysed using an IP buffer. A and D The interaction of IFITM3 and its truncated fragments with CCHFV Gc were analyzed via S-pulldown assays followed by Western blotting, with β-actin as the internal control. B Schematic diagram of the constructs of truncated IFITM3 used in this study, each fused with EGFP at its N terminus. C and E Visualization of IFITM3 or its truncations (green) and CCHFV Gc (red) in cells using confocal microscopy. The transfected cells were fixed and permeabilized, and then incubated with S-tag antibody by IFAs. Nuclei were stained with Hoechst 33,258. Scale bar, 10 μm. F Schematic diagram of the 3D structures of IFITM3 and CCHFV Gc, which were predicted using Alpha fold 3. The IFITM3 domains and the regions and amino acids which could be involved in the interaction are indicated. endpoint dilution method as described previously (Dai et al., 2021). All experiments involving CCHFV were performed in a biosafety level 3 laboratory at the Wuhan Institute of Virology, Chinese Academy of Sciences.
Human megakaryoblast MEG-01 cells (CRL-2021, ATCC) were maintained in RPMI 1640 medium. Human embryonic kidney HEK293 (CRL-1573, ATCC), human cervical adenocarcinoma HeLa (CCL-2, ATCC) cells, and African green monkey kidney clone E6 (Vero E6) cells were maintained in Eagle's Minimum Essential Medium (EMEM, New Zongke Biotech, Wuhan, China) and human embryonic kidney HEK293T (CRL-3216, ATCC) were maintained in Dulbecco's Modified Eagle's Medium (DMEM, New Zongke Biotech, Wuhan, China) at 37 C with 5% CO 2 . All culture media were supplemented with 10% fetal bovine serum (FBS) and 1% penicillinstreptomycin-glutamine (Thermo Fisher Scientific, Waltham, MA, USA).
PE/Cy7® Anti-mouse CD41 Antibody [MWReg30] (ab95726), Goat Anti-Rabbit IgG H&L (Alexa Fluor® 488, ab150077), and Goat Anti-Rabbit IgG H&L (Alexa Fluor® 555, ab150078) were purchased from Abcam (Cambridge, UK). Polyclonal antibodies against the nucleoprotein (anti-NP) and N-and C-terminal fragments of the glycoproteins (Gn and Gc) of the CCHFV were prepared as previously described (Dai et al., 2021). Horseradish peroxidase (HRP)-conjugated Affinipure goat Anti-mouse IgG (H þ L) (SA00001-1) and HRP-conjugated Affinipure Goat Anti-Rabbit IgG (H þ L) (SA00001-2) were purchased from Proteintech Group, Inc (Chicago, IL, USA). The rabbit anti-human IFITM3 antibody ([KO Validated] IFITM3 Rabbit pAb (A13070)), which is capable to recognize both human and mouse IFITM3, was purchased from ABclonal Technology, Inc (Wuhan, China).
## Mouse experiments
A total of 120 SPF-grade C57BL/6J mice (female, 6-8 weeks) were procured from the Animal Experiment Center of the Wuhan Institute of Virology, Chinese Academy of Sciences. Mice were intraperitoneally inoculated with 200 μL of CCHFV (n ¼ 65, 1.56 Â 10 5 TCID 50 /mouse) or received mock infection of mouse brain homogenate prepared in equivalent volume of serum-free EMEM (n ¼ 55, 200 μL/mouse) as control. On day three post-inoculation, whole blood was collected. Platelets were purified from three groups with CCHFV infection and two groups with mock infection (n ¼ 10/group) and were applied to transcriptome analyses. For Western blot analyses, platelets were purified from three mouse groups with CCHFV infection and three mock-infected groups (n ¼ 10/group). Routine blood examination was carried out using blood from five mice with CCHFV infection and five others with mock infection, respectively. All animal experiments were approved by the Ethics Committee of the Wuhan Institute of Virology, Chinese Academy of Sciences (approval number: WIVA33202107) and conducted in an ABSL-3 (Animal Biosafety Level 3) laboratory.
## Platelet isolation from mice
Platelets were isolated from C57BL/6J mice and quantified by RNA sequencing. Briefly, blood collected in a vessel containing sodium citrate (w/v, 3.8%) was centrifuged at 200 Âg for 20 min at 25 C to separate platelet-rich plasma (PRP) from the blood cells. The PRP was transferred to a new tube, centrifuged (200 Âg, 10 min) to remove residual leukocytes and red blood cells, and centrifuged at 500 Âg for 5 min to obtain platelet pellets. Platelets were washed with PIPES buffer (145 mM NaCl, 4 mM KCl, 50 mM Na 2 HPO 4 , 1 mM MgCl 2 ‧6H 2 O, 5 mM PIPES, 5.5 mM glucose, and 300 nM prostaglandin E1) and re-suspended in serum-free M199 medium (Gibco®, Thermo Fisher Scientific, Waltham, MA, USA). Platelet purity (CD41 þ > 97%) was measured using flow cytometry after incubation with PE/Cy7® anti-mouse CD41 Antibody [MWReg30].
## RNA data analysis
Total RNA was purified from platelets with or without CCHFV infection using TRIzol reagent (NovaStar, China), following the manufacturer's instructions. RNA-seq libraries were constructed according to the standard process for mRNA sequencing at the Beijing Genomics Institute (BGI, China). Raw data were filtered using SOAPnuke software (v.1.5.2). Gene expression levels were quantified in salmon (v.1.10.3) using Mus musculus GRCm39 as an index. The differential gene expression and KEGG pathway enrichment analyses were performed using DESeq2 (version 1.38.3) and ClusterProfiler (version 4.6.2), respectively. DEGs were selected using padj < 0.05, log 2 FoldChange >1, and enrichment pathways were selected based on a P <0.05. Graphs and analyses of the data were generated using R (v.4.2.3) and ggplot2 (v.3.4.2) with custom scripts.
## Virus infection assay
HEK293 cells or MEG-01 cells (1 Â 10 5 cells per well) were infected with CCHFV (MOI ¼ 10) at 37 C and 5% CO 2 for 2 h. The cells were washed three times with phosphate-buffered saline (PBS) to remove any remaining free virus and maintained in fresh medium. Cells were collected at the indicated time points and subjected to Western blotting, qRT-PCR, and flow cytometry to determine the CCHFV infection levels.
## Western blots
After viral infection, the cells were washed with PBS (pH 7.4) and lysed in RIPA buffer containing protease inhibitors. Proteins were boiled at 100 C for 10 min, then centrifuged for 10 min, and subjected to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred into polyvinylidene fluoride (PVDF) membranes (Whatman, England). Membranes were blocked in Tris-buffered saline (TBS) supplemented with 0.1% Tween 20 plus 5% non-fat milk for 1 h before incubation overnight with primary antibodies at 4 C. After washing five times, membranes were revealed using peroxidase-conjugated corresponding secondary antibodies for 1 h at 37 C. Band signals were detected using chemiluminescence (Azure Biosystems, USA) with Super Signal West Pico Chemiluminescent Substrate (Thermo Scientific, USA), according to the manufacturer's protocol.
## Flow cytometry assay
Cells were harvested by centrifugation at 500 Âg for 5 min, and the pellets were washed with cold PBS, fixed with 4% paraformaldehyde for 10 min, centrifuged at 500 Âg for 5 min to remove paraformaldehyde, and washed with PBS again. Thereafter, the cells were stained with primary antibodies at 4 C overnight. After washing, cells were incubated with appropriate secondary antibodies in the dark and then subjected to flow cytometric analysis using a flow cytometer (BD Biosciences, USA). Each experiment was performed in triplicates and repeated at least three times. The data generated by flow cytometry were statistically analyzed using FlowJo V10 software.
## Immunofluorescence and confocal microscopy
Cells were attached to poly-L-Lysine-coated coverslips and fixed with 4% paraformaldehyde for 15 min, followed by permeabilization with 0.2% Triton-X 100 in PBS for 15 min, and then blocked with 5% BSA in PBS for 1 h at 37 C. After incubation with primary antibodies overnight at 4 C, subsequently followed by washing and incubation with appropriate secondary antibodies, confocal images were captured using a laserscanning confocal microscope (Leica STELLARIS 8, Germany). Images were analyzed using LAS X software.
## Real-time reverse transcription polymerase chain reaction (qRT-PCR)
qRT-PCR analysis was performed to determine the viral loads. Total RNA was extracted from the culture supernatants of cells infected with CCHFV at different time points using TRIzol reagent (Takara, Kusatsu, Japan) according to the manufacturer's instructions. One-step qRT-PCR was used to determine the viral loads using a One-Step PrimeScript™ RT-PCR Kit (Takara) and a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The primer pairs used in this study were as follows: hifitm3 forward: 5ʹ-AATCACACTGTCCAAACCT-3ʹ, and reverse: 5ʹ-CTCCTCCTTGAGCATCTC-3ʹ; glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward: 5ʹ-CAAGGGCATCCTGGGCTACACT-3ʹ, and reverse 5ʹ-CCCAGCGTCAAAGGTGGAGGA-3ʹ; CCHFV NP forward: 5ʹ-TCAAGTGGAGGAARGACATAGG-3ʹ, reverse: 5ʹ-TCCACATGTTCAC GGCTSACTGGG-3ʹ, and the probe: FAM-TCATCRCCACCTCTGTTG AGAA-BHQ1 (Zhang, Y. et al., 2018).
Generation of IFITM3-overexpressed cell lines and IFITM3knockout cell lines MEG-01 and HEK293 cells overexpressing IFITM3 were generated. The hifitm3 gene was amplified from HeLa cells of human origin and verified by Sanger sequencing. Subsequently, the fragment EGFP-P2A-hifitm3 was cloned into pLVX-IRES-Puro to generate a recombinant plasmid expressing IFITM3, which was then transferred into HEK293T cells together with the packaging plasmid psPAX2 and skeleton plasmid pMD2.0G at a ratio of 2:1.5:0.5, as previously described (Chang et al., 2024). The supernatant was collected 24 and 48 h after transfection and centrifuged at 1000 Âg for 10 min to remove cell debris. HEK293 and MEG-01 cells (1 Â 10 5 cells/well) were pre-inoculated into 6-well plates and transduced with the supernatant. After 4 h of transduction, the supernatant was replaced with fresh medium. The overexpression of IFITM3 in cells was examined by flow cytometry and Western blotting.
To generate knockout-hifitm3 MEG-01 (MEG-01-hifitm3-KO) and HeLa (HeLa-hifitm3-KO) cell lines, the first exon of the human ifitm3gene sgRNA or the control sgRNA without a human genome target was cloned into lentiCRISPR v.2 (Addgene, catalog no. 52961). The lentiCRISPR plasmid with the inserted sgRNA, psPAX2 packaging vector, and skeleton plasmid pMD2.0G were co-transfected into HEK293T cells as described above. MEG-01 and HeLa cells were transduced using transfection supernatants and maintained as described above. MEG-01 and HeLa cells were treated with 3 μg/mL and 2 μg/mL puromycin, respectively, after 2 d of expression of selected sgRNA and Cas9 in a 6-well plate. The two most effective IFITM3-targeting sgRNAs were sgRNA1 5ʹ-TTGAG-CATCTCATAGTTGGG-3ʹ and sgRNA2, 5ʹ-GCAGCAGGGTTCATGAAGA-3ʹ. Gene knockouts were characterized by tracking indels using decomposition (TIDE) analysis. After three rounds of puromycin selection, the genomic DNA was purified using a FastPure Blood/Cell/Tissue/Bacteria DNA Isolation Mini Kit (DC112-01, Vazyme, China). The cell DNA was resuspended in 200 μL of solution, and PCR was performed as follows: initial denaturation at 95 C for 3 min, followed by 30 cycles of denaturation at 95 C for 15 s, annealing at 56 C for 15 s, and extension at 72 C for 30 s. The ifitm3 locus was amplified using the following primers: forward 5ʹ-ACCATCCCAGTAACCCACCG-3ʹ and reverse 5ʹ-GCTGATA-CAGGACTCGGCTCC-3ʹ.
## Pull-down assay
HEK293T cells were co-transfected with two expression plasmids expressing enhanced green fluorescent protein (EGFP)-fused IFITM3 and S-tag-fused viral proteins. At 24 h post-transfection, cells were collected, washed twice with cold PBS, and lysed in immunoprecipitation (IP) lysis buffer (P0013, Beyotime, China) for 30 min on ice. Lytic supernatants were isolated by covalent coupling with anti-EGFP (Abclonal, Wuhan, China) or anti-S-tag single-molecule nanoantibody beads for 4 h at 4 C to isolate the immune complexes. The beads were washed three times with cold lysis buffer and cold PBS before IP. The precipitated complexes were re-suspended in 1 Â SDS loading buffer, separated by SDS-PAGE, and immunoblotted using specific antibodies.
## CCHFV binding and internalization assays
Pre-chilled MEG-01 and MEG-01-hifitm3-OE cells were incubated with CCHFV at 4 C for 1 h, as previously described (Chang et al., 2024). The cells were then washed three times with ice-cold PBS to remove unbound viruses. Viral RNA copy numbers were determined by qRT-PCR. To assess internalized virus loads, cells were further incubated at 37 C for 1 h after the initial 4 C incubation. Cells were harvested at specific time points, washed with PBS containing 500 ng/mL proteinase K to remove surface-bound viruses, and used for RNA extraction (Liu et al., 2019). Viral RNA loads were determined using qRT-PCR.
## Plasmid construction and transfection
For intracellular localization and interaction analysis, the human ifitm3 gene (Genbank accession: NM_021034.3) and the domains including the N-terminal domain (NTD), intramembrane-intracellular loop domain (IMD-CIL), and intracellular loop-transmembrane domain (CIL-TMD) were cloned into the pEGFP-N1 vector fused with EGFP at the C-terminus according to the previous study (Du et al., 2024). The CCHFV encoding Gc fused with S-tag at the C-terminus was constructed into a phCMV vector. All plasmids were verified using Sanger sequencing. Transfection assays were performed with these plasmids using the Lipofectamine 3000 transfection reagent (L3000015, Invitrogen, Thermo Fisher Scientific) according to the manufacturer's instructions.
## Structure prediction
Protein structure models of the human IFITM3 and Gc complexes were predicted using the AlphaFold3 web server (https://golgi.sandbox. google.com/) (Abramson et al., 2024). The salt-bridge and hydrogen-bonding networks were calculated using PDBePISA v1.52 (Krissinel and Henrick, 2007).
## Statistical analysis
All experiments were repeated at least three times unless otherwise stated. Statistical analyses were performed using R (version 4.2.3) and Prism (version 8.0; GraphPad Software, CA, USA) with custom scripts. Differences between two groups were analyzed using a two-tailed unpaired t-test, and comparisons among multiple groups were made using a one-way ANOVA. Statistical significance was set at P < 0.05.
## References
1. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
2. Akinci, Bodur, Leblebicioglu (2013) "Pathogenesis of Crimean-Congo hemorrhagic fever" *Vector Borne Zoonotic Dis*
3. Akinci, Bodur, Sunbul et al. (2016) "Prognostic factors, pathophysiology and novel biomarkers in Crimean-Congo hemorrhagic fever" *Antivir. Res*
4. Aligholipour Farzani, Foldes, Hanifehnezhad et al. (2019) "Bovine herpesvirus type 4 (BoHV-4) vector delivering nucleocapsid protein of crimean-Congo hemorrhagic fever virus induces comparable protective immunity against lethal challenge in IFNalpha/beta/gammaR-/-mice models" *Viruses*
5. Aslam, Abbas, Alsayeqh (2023) "Distribution pattern of Crimean-Congo hemorrhagic fever in" *Asia and the Middle East. Front. Public Health*
6. Bailey, Kondur, Huang et al. (2013) "Interferon-induced transmembrane protein 3 is a type II transmembrane protein" *J. Biol. Chem*
7. Bailey, Zhong, Huang et al. (2014) "IFITM-family proteins: the cell's first line of antiviral defense" *Annu. Rev. Virol*
8. Bente, Forrester, Watts et al. (2013) "Crimean-Congo hemorrhagic fever: history, epidemiology, pathogenesis, clinical syndrome and genetic diversity" *Antivir. Res*
9. Bodur, Akinci, Ascioglu et al. (2012) "Subclinical infections with Crimean-Congo hemorrhagic fever virus" *Turkey. Emerg. Infect. Dis*
10. Bote, Corkrey, Koupenova (2022) "Human platelets and influenza virus: internalization and platelet activation" *Platelets*
11. Brass, Huang, Benita et al. (2009) "The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus" *Cell*
12. Campbell, Schwertz, Hottz et al. (2019) "Human megakaryocytes possess intrinsic antiviral immunity through regulated induction of IFITM3" *Blood*
13. Chang, Min, Xu et al. (2024) "Host factor MxA restricts Dabie bandavirus infection by targeting the viral NP protein to inhibit NP-RdRp interaction and ribonucleoprotein activity" *J. Virol*
14. Chen, Jie, Tang et al. (2017) "New insights into the effects and mechanism of a classic traditional Chinese medicinal formula on influenza prevention" *Phytomedicine*
15. Chumakov, Butenko, Shalunova et al. (1968)
16. *Vopr. Virusol*
17. Compton, Roy, Porrot et al. (2016) "Natural mutations in IFITM3 modulate post-translational regulation and toggle antiviral specificity" *EMBO Rep*
18. Cross, Prasad, Borisevich et al. (2020) "Crimean-Congo hemorrhagic fever virus strains Hoti and Afghanistan cause viremia and mild clinical disease in cynomolgus monkeys"
19. Dai, Wu, Wu et al. (2021) "Differential cell line susceptibility to crimean-Congo hemorrhagic fever virus" *Front. Cell. Infect. Microbiol*
20. Denz, Speaks, Kenney et al. (2024) "Innate immune control of influenza virus interspecies adaptation via IFITM3" *Nat. Commun*
21. Deyde, Khristova, Rollin et al. (2006) "Crimean-Congo hemorrhagic fever virus genomics and global diversity" *J. Virol*
22. Dib, Quirino-Teixeira, Merij et al. (2020) "Innate immune receptors in platelets and plateletleukocyte interactions" *J. Leukoc. Biol*
23. Du, Wang, Wang et al. (2024) "IFITM3 inhibits severe fever with thrombocytopenia syndrome virus entry and interacts with viral Gc protein" *J. Med. Virol*
24. Ergonul (2012) "Crimean-Congo hemorrhagic fever virus: new outbreaks, new discoveries" *Curr Opin Virol*
25. Fang, Shen, Zhang et al. (2024) "Thrombocytopenia in SFTS due to platelets with altered function undergoing cell death pathways" *J. Infect. Dis*
26. Fereidouni, Apanaskevich, Pecor et al. (2023) "Crimean-Congo hemorrhagic fever virus in Central, Eastern, and South-eastern Asia" *Virol. Sin*
27. Garrison, Smith, Golden (2019) "Animal models for crimean-Congo hemorrhagic fever human disease" *Viruses*
28. Guo, Shen, Zhang et al. (2017) "A new strain of Crimean-Congo hemorrhagic fever virus isolated from Xinjiang" *Virol. Sin*
29. Haddock, Feldmann, Hawman et al. (2018) "A cynomolgus macaque model for Crimean-Congo haemorrhagic fever" *Nat. Microbiol*
30. Hoogstraal (1979) "The epidemiology of tick-borne Crimean-Congo hemorrhagic fever in Asia, Europe, and Africa" *J. Med. Entomol*
31. Jia, Pan, Ding et al. (2012) "The Nterminal region of IFITM3 modulates its antiviral activity by regulating IFITM3 cellular localization" *J. Virol*
32. Klein, Golani, Lolicato et al. (2023) "IFITM3 blocks influenza virus entry by sorting lipids and stabilizing hemifusion" *Cell Host Microbe*
33. Koupenova, Livada, Morrell (2022) "Platelet and megakaryocyte roles in innate and adaptive immunity" *Circ. Res*
34. Krissinel, Henrick (2007) "Inference of macromolecular assemblies from crystalline state" *J. Mol. Biol*
35. Li, Zhang, Liu et al. (2025) "Identification of IFITM3 as a diagnostic biomarker of systemic lupus erythematosus and its association with disease activity based on multi-omics and experimental verification" *Lupus*
36. Lin, Yu, Ren et al. (2020) "Human microRNA-30 inhibits influenza virus infection by suppressing the expression of SOCS1, SOCS3, and NEDD4" *Cell. Microbiol*
37. Lindquist, Zeng, Altamura et al. (2018) "Exploring Crimean-Congo hemorrhagic fever virus-induced hepatic injury using antibody-mediated type I interferon blockade in mice" *J. Virol*
38. Liu, Xu, Tang et al. (2019) "Single-particle tracking reveals the sequential entry process of the bunyavirus severe fever with thrombocytopenia syndrome virus" *Small*
39. Lu, Pan, Rong et al. (2011) "The IFITM proteins inhibit HIV-1 infection"
40. Machlus, Italiano (2013) "The incredible journey: from megakaryocyte development to platelet formation" *J. Cell Biol*
41. Marziali, Delpeuch, Kumar et al. (2021) "Functional heterogeneity of mammalian IFITM proteins against HIV-1" *J. Virol*
42. Mcentire, Song, Mcinnis et al. (2021) "Neurologic manifestations of the world Health organization's list of pandemic and epidemic diseases" *Front. Neurol*
43. Mo, Feng, Dai et al. (2023) "Transcriptome profiling highlights regulated biological processes and type III interferon antiviral responses upon Crimean-Congo hemorrhagic fever virus infection" *Virol. Sin*
44. Narayana, Helbig, Mccartney et al. (2015) "The interferon-induced transmembrane proteins, IFITM1, IFITM2, and IFITM3 inhibit hepatitis C virus entry" *J. Biol. Chem*
45. Neogi, Elaldi, Appelberg et al. (2022) "Multi-omics insights into host-viral response and pathogenesis in Crimean-Congo hemorrhagic fever viruses for novel therapeutic target"
46. Quirino-Teixeira, Andrade, Pinheiro et al. (2022) "Platelets in dengue infection: more than a numbers game" *Platelets*
47. Ren, Du, Xu et al. (2020) *Current progress on host antiviral factor IFITMs. Front. Immunol*
48. Schoggins, Wilson, Panis et al. (2011) "A diverse range of gene products are effectors of the type I interferon antiviral response" *Nature*
49. Scholte, Zivcec, Dzimianski et al. (2017) "Crimean-Congo hemorrhagic fever virus suppresses innate immune responses via a ubiquitin and ISG15 specific protease" *Cell Rep*
50. Shtanko, Nikitina, Altuntas et al. (2014) "Crimean-Congo hemorrhagic fever virus entry into host cells occurs through the multivesicular body and requires ESCRT regulators" *PLoS Pathog*
51. Simon, Sutherland, Pryzdial (2015) "Dengue virus binding and replication by platelets" *Blood*
52. Tartour, Appourchaux, Gaillard et al. (2014) "IFITM proteins are incorporated onto HIV-1 virion particles and negatively imprint their infectivity" *Retrovirology*
53. Tignor, Hanham (1993) "Ribavirin efficacy in an in vivo model of Crimean-Congo hemorrhagic fever virus (CCHF) infection" *Antivir. Res*
54. Tipih, Meade-White, Rao et al. (2023) "Favipiravir and Ribavirin protect immunocompetent mice from lethal CCHFV infection" *Antivir. Res*
55. Xu, Wang, Li et al. (2022) "Transmembrane domain of IFITM3 is responsible for its interaction with influenza virus HA(2) subunit" *Virol. Sin*
56. Yao, Dong, Zhu et al. (2011) "Identification of the IFITM3 gene as an inhibitor of hepatitis C viral translation in a stable STAT1 cell line" *J. Viral Hepat*
57. Zani, Yount (2018) "Antiviral protection by IFITM3 in vivo" *Curr. Clin. Microbiol. Rep*
58. Zhang, Ruan, Yang et al. (2024) "PGRN inhibits early B-cell activation and IgE production through the IFITM3-STAT1 signaling pathway in asthma" *Adv. Sci. (Weinh.)*
59. Zhang, Shen, Fang et al. (2018) "Isolation, characterization, and phylogenetic analysis of two new Crimean-Congo hemorrhagic fever virus strains from the Northern region of Xinjiang Province" *Virol. Sin*
60. Zivcec, Safronetz, Scott et al. (2013) "Lethal Crimean-Congo hemorrhagic fever virus infection in interferon alpha/beta receptor knockout mice is associated with high viral loads, proinflammatory responses, and coagulopathy" *J. Infect. Dis*
61. Zhang (2025) *Virologica Sinica* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12312495&blobtype=pdf | # Investigating the link between HPV genotypes and cervical abnormality incidence in women with HPV infections: insights from a leading referral centre
Virology Journal, Fariba Yarandi, Seyed Mohammad, Zahra Sadeghi, Amir Aboofazeli, Sheida Sarrafzadeh, Naeimeh Tayebi, Roxana Tajdini, Seyed Jazayeri
## Abstract
Background Cervical cancer (CC) is a serious health issue, especially in low-and middle-income countries, primarily caused by human papillomavirus (HPV) infection, particularly genotypes 16 and 18. Other risk factors include smoking, early sexual activity, and long-term oral contraceptive use. Early detection through cervical cytology and HPV testing is vital for effective prevention.Objective This study aimed to analyze the clinical and epidemiological characteristics of cervical intraepithelial neoplasia (CIN) in HPV-positive women at a Tehran teaching hospital, focusing on HPV genotypes and their association with CIN.
MethodsWe conducted a cross-sectional study of HPV-positive women who underwent Pap smear testing, HPV genotyping via real-time PCR, and colposcopy with biopsies of suspected lesions. Risk factors like smoking and alcohol consumption were evaluated, and statistical analyses were performed, including ordinal logistic regression. Results Among participants, 52.4% had abnormal CIN: CIN I (31.1%), CIN II (11.4%), and CIN III (10.0%). HPV-16 was the most prevalent genotype (43.7%), significantly associated with severe CIN outcomes (odds ratio [OR] = 2.88, 95% CI), followed by HPV-18 (OR = 1.87, 95% CI). Smoking increased the risk of severe CIN (OR = 1.53, 95% CI), while older age and later age at sexual debut correlated with better CIN outcomes.Conclusions HPV-16 and smoking are major predictors of severe CIN, highlighting the need for targeted interventions such as HPV vaccination and smoking cessation, along with regular screenings to lower cervical cancer
## Introduction
Cervical cancer (CC) is a prevalent disease among women, with a significant global morbidity and death rate [1]. In 2022, 94% of the 348,874 cervical cancer deaths happened in low-and middle-income countries (LMICs) [2]. The occurrence and rate of new cases of cervical cancer in Iran are relatively minimal. According to reports from the Global Cancer Observatory (GLOBOCAN), the age-standardised incidence rate (ASR) stands at about 2.4 for every 100,000 women, positioning cervical cancer as the eleventh most prevalent cancer among women in Iran. However, because there is no consistent screening program in place, numerous cases are identified at a later stage [3].
It has been acknowledged in the past 20 years that several viruses play a crucial role in developing human cancers [4]. The most common virus that impacts the reproductive system and is a major global cause of sexually transmitted infections (STIs) is the human papillomavirus (HPV) [5]. HPV is part of the Papillomaviridae family and has five main genera: alpha, beta, gamma, mu, and nu. The alpha genus is the most common and has the potential for oncogenic transformation, divided into low and high-risk HPV groups [6]. Low-risk HPVs are mainly associated with anogenital warts, while high-risk HPVs can lead to anogenital cancers, such as CC. Globally, HPV-16 and -18 are responsible for over 70% of cervical cancer cases. Following HPV-16 and -18, the other most prevalent HPV types are HPV-31, -33, -35, -45, -52, and -58, collectively contributing to an additional 20% of cervical cancer cases [7]. The primary determinant of HPV oncogenicity is the persistent expression and function of the viral proteins E6 and E7. These proteins, which are tumor-associated antigens, collaborate to alter various biological processes and promote tumor growth by interacting with other cellular proteins [8]. Infections with carcinogenic HPV types lead to cervical precancer, which, if left untreated, can progress to invasive carcinoma [9].
Smoking is an additional risk factor for CC. Being exposed to tobacco smoking can elevate the risk of CC development from HPV infection [10,11]. Other risk factors comprise initiating a sexual relationship before the age of 20, having multiple partners, using oral contraceptives for more than five years, and having a family history of CC [12]. Cervical intraepithelial neoplasia (CIN), a premalignant phase, comes before CC. Screening for cervical abnormalities in this premalignant stage can detect them before they develop into CC [13]. Precise screening can lead to early detection and proper treatment [14]. A pap smear (also known as cervical cytology) is the primary screening test for early identification of invasive cervical cancer and precancerous CIN. It can detect early changes in the cervical epithelium [15]. HPV testing is now included in cervical cytology screening guidelines. Cervical specimens can be examined for HPV-DNA through nucleic acid amplification with polymerase chain reaction (PCR) or signal amplification techniques(Cervista HPV HR (Hologic), and APTIMA HPV Assay (Gen-Probe) [16]. Colposcopy is another diagnostic procedure utilized to identify abnormalities in the cervix.It should also be employed to evaluate abnormal cervical tissue [17].
This study sought to investigate the key clinical, laboratory, and epidemiological characteristics of women presenting with HPV-positive features who underwent cervical screening at a prestigious teaching referral hospital in Tehran, Iran. The research focused on a comprehensive analysis conducted over a one-year period, aiming to enhance the understanding of the factors associated with HPV and its implications for women's health in the region.
## Methods and materials
## Study population
This cross-sectional study examined women who participated in cervical screening between October 2023 and May 2024. The aim was to explore the impact of HPV genotypes on cervical cancer. Women who tested positive for HPV were further assessed with a Pap smear and evaluated for cervical intraepithelial neoplasia (CIN). We successfully recruited 710 participants from various healthcare facility, ensuring a diverse demographic representation. Informed consent was obtained from all participants before they were enrolled in the study.
Upon arrival for their screening appointments, participants completed a comprehensive questionnaire that collected demographic information, including age, marital status, smoking habits, and alcohol consumption. Following this, trained healthcare professionals collected Thin-Prep samples for Pap smear and HPV testing. all laboratory analyses. Following established clinical guidelines, pap smear samples underwent cytological evaluation to classify them as normal or abnormal. Concurrently, HPV testing was performed to detect the presence of the virus. For samples that tested positive for HPV, genotyping was carried out to identify specific viral strains, with a particular emphasis on high-risk types associated with cervical cancer.
## DNA extraction
DNA extraction was performed using the Favorgen Biotech Corp. extraction kit (Favorgen Biotech, Taiwan), following the manufacturer's instructions. All procedures, including preparing samples for real-time PCR, were conducted in a sterile laboratory environment.
## HPV genotyping
While FDA-approved kits such as HC2 and Cobas are available, we have limited access to them due to sanctions; therefore, we used the tBioDx™ HPV ExtendScreen™ realtime PCR kit, which has validated performance (98.19% sensitivity, 96.30% specificity). This assay detects E6/E7 of 14 high-risk HPV types (16,18,31,33,35,39,45,51,52,56,58,59,66,68). The GAPDH gene served as an internal reference to evaluate the test's quality and identify any potential inhibitors in the assay. We also added that each PCR run included a positive control (known HPVpositive sample) and a negative (no-template) control for quality assurance.
## Colposcopy and biopsy
Women who tested positive for high-risk HPV types were referred for colposcopy. During this procedure, 3-5% acetic acid was applied to the cervix to enhance the visualisation of abnormal epithelial cells, followed by Lugol's iodine to distinguish normal glycogen-rich cells from abnormal ones. Three distinct images were captured during the colposcopy: the initial view of the cervix, the acetic-acid-enhanced view, and the Lugol's iodine-stained view, which were documented for clinical assessment.
In cases where lesions were identified, biopsies were performed to evaluate for the presence of cervical intraepithelial neoplasia (CIN).
## Study duration
Sample collection was conducted over 8 months, allowing us to recruit a substantial cross-section of 710 women. Follow-up appointments were scheduled to review test results and determine if further diagnostic procedures were necessary.
## Ethical considerations
The study was conducted by the Declaration of Helsinki and received approval from the institutional ethics committees of the participating centres. Participants were thoroughly informed about the study's objectives, potential risks, and benefits, and were free to withdraw from the study at any time without any repercussions. Data confidentiality was upheld throughout the research, with all personal identifiers securely stored in compliance with GDPR.
## Statistical analysis
Descriptive statistics were applied to describe data, including mean, standard deviation, frequency, and percentage. Cervical intraepithelial neoplasia (CIN) was considered a 4-group ordinal variable; therefore, the ordinal logistic regression model was performed for all unadjusted and adjusted models to calculate odds ratios. Microsoft Office Excel was used for data entry and basic calculations (e.g., computing means and percentages). R programming language version 4.2.3 was used for the primary analysis. The polr() function in the MASS library was used for the ordinal logistic model.
## Result
This study investigated 710 HPV-positive women. 47.6% of subjects showed normal cytology according to CIN lesions, and the other 52.4% were in one of the CIN stages. The frequency of subjects in CIN stages is reported in Table 1. The mean±SD of the age of subjects was 35.7±8.29 years, where the youngest and the oldest subjects were 16 and 70 years old, respectively. Additionally, the mean ±SD of the age for the sexual debut was 22±4.94 years (Table 2). Table 2 also presents the descriptive statistics for baseline characteristics and HPV genotypes according to the 4-group CIN conditions, including three stages of CIN and one normal (CIN-free) group. As the table shows, the average age of sexual debut was 22.1 years for subjects in the normal group. However, in CIN stages, higher stages were related to a lower age of sexual debut. The second section of Table 2 shows the proportion of existence of each HPV genotype or binary baseline characteristics in total and each CIN condition group. As can be seen, 31% of study subjects smoke cigarettes. Additionally, only 28.2% of subjects in the regular group smoke cigarettes; however, 42.7% of subjects in group CIN III smoke cigarettes. This result shows that in subjects with more severe CIN conditions, a higher proportion of subjects smoke cigarettes. Furthermore, 43.7% of our subjects were infected with HPV-16. The proportion of HPV-16-infected subjects in the normal group was 33.1%. Among subjects with more severe CIN conditions, a higher proportion of HPV-16 infection was observed. On the other hand, in subjects with more severe CIN stage, the proportion of HPV-18 genotype decreases.
Table 3 shows unadjusted odds ratios (OR) for covariates and HPV genotypes. The odds ratio for age was 0.98, which was significant at a 5% level. Older ages were associated with less severe CIN conditions. For every 10 years increase in age, the chances of being in the lower CIN stages were 1.2 times higher than those with younger ages. For the age of sexual debut, OR = 0.97 was close to the significance level. However, it was significant at the 10% level. According to our results, a higher age of sexual debut was associated with lower chances of developing more severe CIN conditions, and higher ages of sexual debut were associated with less severe CIN conditions. For every 10 years increase in sexual debut age, the chances of experiencing a less severe CIN condition were 1.4 times higher than at younger ages. Having cigarettes and hookah was considered separately. Smoking cigarettes was significantly associated with more severe CIN conditions (OR = 1.56, 95%CI: 1.16-2.11). The chances of experiencing more severe CIN conditions in subjects who consume cigarettes were 1.6 times higher than in those who do not consume cigarettes. However, we could not find a significant association between using Hoockah and CIN conditions. Alcohol consumption was associated with more severe CIN conditions (OR = 1.54). Thus, the chances of being in the higher CIN stage in those who consumed alcohol were 1.5 times higher than in those who did not. As an unadjusted analysis of genotypes shows, only genotype 16 was significantly associated with CIN conditions.
Table 4 shows the results of the final model. All genotypes are considered together in this model, controlling for baseline characteristics. Since hookah was insignificant in the unadjusted model, only cigarettes were entered into the adjusted model to reduce the risk of multicollinearity. According to the results, among baseline characteristics including age, age at sexual debut, consuming cigarettes and alcohol, only cigarettes remained in the final model (OR = 1.53). Thus, the chances of being in the more severe CIN conditions in subjects who used cigarettes were 1.5 times higher than those who did not. HPV genotype 16 was significantly associated with more severe CIN condition (OR = 2.88). According to our results, the chances of having a more severe CIN condition in females with the HPV-16 genotype were 2.9 times higher than in the others, and in females with the HPV-18 genotype were 1.9 times higher than in the others. Both odds ratios were statistically significant at a 5% level. Additionally, the chances of having more severe CIN conditions in subjects who had the HPV-31 genotype were 1.4 times higher than in the others, and in those with HPV-35 were 1.6 times higher than in the others. However, the effect was insignificant at the 5% level (Table 4). A total of 1320 genotypes were detected from 710 studied individuals. Figure 1 shows the distribution of total genotypes detected in the investigated subjects in this study.
## Discussion
In this study, we analyzed data from 710 participants to explore the relationship between different HPV genotypes and the occurrence of cervical intraepithelial neoplasia (CIN). Our results showed that HPV-16 and -18 genotypes had a statistically significant relationship with the increased probability of CIN, with an odds ratio of 2.88 and 1.87, respectively. Although HPV-31 and--35 genotypes did not reach statistical significance at the 0.05 level, they were very close to the threshold, indicating a potential association. Additionally, we observed significant correlations between CIN and other factors such as smoking and Pap smear results, emphasizing CIN development's complex and multifactorial nature. These findings could contribute to a better understanding of HPV-related cervical disease progression.
In the study conducted by Sánchez-Siles et al., individuals in the HPV-related CIN group had an average age of 34.54 years, with 30% reporting alcohol consumption and 42% being smokers. The most prevalent HPV genotype in this group was HPV-45 at 33%, followed by HPV-39 at 22%, and HPV-16 and HPV-56 at 11%. While the average age of participants in their study is similar to that of our research, we observed higher rates of alcohol consumption and smoking among our participants. Furthermore, in contrast to their findings, the most frequently reported genotype in the present investigation was HPV-16, nearly four times more prevalent than in their research. Significant discrepancies were also noted in the infection rates of other genotypes between our study and theirs [18]. Although the average age of the participants in their article was comparable to ours, we noted higher rates of alcohol consumption and smoking within our group. This suggests that lifestyle factors may contribute significantly to our subjects' risk of developing CIN, emphasizing the need for targeted public health interventions that address these behaviors.
A piece of research focusing on women with cervical cancer in South Africa found that the most prevalent HPV genotype was HPV-16, accounting for approximately 35% of cases, followed by HPV-35 at around 17%, and both HPV-45 and HPV-54 at 12%. HPV-18 and HPV-52 represented about 11% and 10%, respectively. This suggests a potential consistency in the oncogenic risk posed by these genotypes across different geographic locations. The persistence of these genotypes underscores the necessity of targeted awareness campaigns and vaccination efforts that address these high-risk strains to mitigate cervical cancer risk. The average age of participants in the present survey was 42 years, indicating a difference of over seven years compared to the population in our study. This age difference may reflect regional variations in the demographics of cervical cancer patients and could influence the screening practices and healthcare access for women in these populations. Older patients may face different risk factors and healthcare challenges compared to younger individuals, complicating direct comparisons of HPV prevalence and its implications. When comparing the findings from this research with our results, we observe that the prevalence rates of HPV-16, HPV-18, and HPV-52 are relatively similar. However, the two studies have significant differences in the prevalence rates of HPV-35, HPV-58, and HPV-45 [19]. In our study, the prevalence rates for these genotypes differed markedly, indicating distinct epidemiological patterns that warrant further investigation. The higher prevalence of HPV-35 in the South African population might suggest regional variations in sexual behavior, HPV vaccination uptake, or socioeconomic factors that influence HPV exposure.
A survey conducted on women from Southern Mexico involved 253 participants with a mean age of 50, all diagnosed with cervical cancer. Among these women, 9% were confirmed smokers, with an HPV prevalence of nearly 99%. This discrepancy could be attributed to variations in lifestyle, cultural practices, or access to smoking cessation resources between the populations. The higher smoking prevalence in the present investigation may suggest an increased risk factor for cervical cancer, as smoking is known to exacerbate the effects of HPV and contribute to the progression of cervical lesions. The most common HPV genotypes identified were HPV-16 (approximately 65%), HPV-18 (10%), HPV-45 (7.5%), and HPV-31/-52/-58 (3.5%). Our findings show significant differences compared to our results; for instance, the smoking rate among participants in our observation was considerably higher. In addition, the prevalence of genotypes 31 and 52 in our observation was approximately three to four times that reported in this paper. This finding highlights potential regional differences in the circulating HPV genotypes and their associated risks. The increased prevalence of these genotypes in our cohort may reflect unique socio-demographic factors, sexual behaviors, or differences in HPV vaccination coverage. Understanding these variations is crucial for tailoring public health interventions and screening programs to effectively address the specific risks different populations face. Although the prevalence of HPV-16 was lower in our findings, the rates of HPV-18 and HPV-58 were similar across both studies [20]. This similarity underscores the importance of these genotypes in the cervical cancer landscape and suggests that they may pose a consistent risk across diverse populations. The presence of HPV-18, in particular, is noteworthy due to its established link to more aggressive cervical cancer forms, emphasizing the need for vigilant monitoring and targeted prevention strategies.
In a 2016 study by Piroozmand et al.(2016) conducted in Iran, the typing of HPV among patients with cervical dysplasia and cancer revealed that approximately 38% were infected with HPV-16 (versus 43.7% in our study), around 27% with HPV-18, nearly 15% with HPV-33, and 4% with HPV-31. The findings regarding HPV-16 are consistent with the results of the present survey; however, there are significant discrepancies in the prevalence of other genotypes between the two studies [21]. This higher prevalence might suggest that our population is experiencing an elevated risk of HPV-16-associated diseases, potentially linked to a variety of factors, including genetic susceptibility, sexual behavior, or access to healthcare services for screening and vaccination. The disproportionate incidence of HPV-16 in our study reinforces its role as a primary oncovirus in cervical cancer and highlights the need for targeted prevention strategies that focus on this genotype.
This Study showed that older patients had better CIN (low-grade) conditions than youngsters; the chances of having better CIN (low-grade) conditions were 1.2 times higher than those of lower ages. cultural transformations in Iranian society over the past two decades have influenced current epidemiological patterns. Historically, many women participated in monogamous relationships, which likely reduced their exposure to high-risk human papillomavirus (HPV) genotypes. Additionally, limited public awareness and the high cost of HPV vaccines have led to low vaccination coverage. In many cases, individuals receive the vaccine only after they have been infected, which diminishes its preventive efficacy. These factors highlight the urgent need for culturally tailored public health interventions, including education, affordable vaccination programs, and accessible screening services.
This study, together with the finding that a higher sexual debut age was associated with lower chances of experiencing worse CIN conditions, highlighted the close relationship between the age of commencement of sexual activity and the incidence of cervical carcinoma. Highrisk sexual behaviors, as well as high frequency of exposure to HPV during early reproductive periods, permit the virus to employ a hit-and-run scenario that ultimately leads to cervical cancer at older ages. These findings collectively underscore the necessity for region-specific strategies for cervical cancer prevention and control. Given the differences in smoking prevalence and HPV genotype distribution, public health initiatives should prioritize education on smoking cessation, promote HPV vaccination, and enhance screening programs tailored to the unique epidemiological profiles of local populations. Understanding these dynamics is essential for developing effective public health policies aimed at reducing the burden of cervical cancer and improving health outcomes for women globally.
This Study does have certain limitations and potential biases. Firstly, a larger sample size could have provided more insightful results and a broader perspective on the findings. Secondly, including additional information, such as family history and the assessment of secondary infections with other microorganisms, could have yielded more interesting data and offered new insights. Finally, due to the high costs associated with follow-up, we could not monitor patients to evaluate and report on the persistence or clearance of infections over time. Another limitation of the research is the lack of knowledge about the previous infections of the people. Considering that the occurrence of cancer in people takes time and has a much longer time frame than HPV, cancer may be related to the genotype that the person was previously infected with, but not associated with the current genotype. Therefore, we may obtain different results by considering the genotypes of previous infections.
## Conclusion
This study offers valuable insights into the epidemiological, clinical, and laboratory profiles of women diagnosed with cervical cancer. Analyzing a sample of 710 participants, the findings indicate a significant association between high-risk HPV genotypes, especially HPV-18 and HPV-16, and the development of cervical intraepithelial neoplasia (CIN). While the relationship between HPV-31 and HPV-35 with CIN was not statistically significant in this investigation, it suggests the potential role of these genotypes in contributing to cervical disease. Beyond HPV, the research identified key connections between CIN and additional risk factors, including smoking and Pap smear outcomes. These results highlight the complex, multifactorial nature of cervical cancer and emphasize the necessity for comprehensive screening and prevention efforts, particularly for high-risk groups. Further investigations are needed to delve deeper into the interactions between various HPV strains and other contributing factors to clarify better the pathways leading to cervical cancer. Expanding these studies to larger populations and incorporating long-term monitoring will help improve prevention strategies and treatment outcomes for at-risk individuals.
## References
1. Choi, Ismail, Pappas-Gogos et al. (2023) "HPV and cervical cancer: A review of epidemiology and screening uptake in the UK. Pathogens"
2. Fuady, Setiawan, Man (2024) "Toward a framework to assess the financial and economic burden of cervical Cancer in Low-and Middle-Income countries: A systematic review" *JCO Glob Oncol*
3. Akbari, Khayamzadeh, Salmanian (2008) "Epidemiology and survival of cervical cancer in Iran based on National cancer registry data" *Front Oncol*
4. Chan, Aimagambetova, Ukybassova (2019) "Human papillomavirus infection and cervical cancer: epidemiology, screening, and vaccinationreview of current perspectives" *Journal of oncology*
5. Okunade (2020) "Human papillomavirus and cervical cancer" *J Obstet Gynaecol*
6. Pratiwi, Ysrafil, Mardhia (2024) "A novel therapeutic multiepitope vaccine based on oncoprotein E6 and E7 of HPV 16 and 18: an in Silico approach"
7. Adebamowo, Famooto, Dareng (2018) "Clearance of type-specific, lowrisk, and high-risk cervical human papillomavirus infections in HIV-negative and HIV-positive women" *J Global Oncol*
8. Tesfaye, Kumbi, Mandefro (2024) "Prevalence of human papillomavirus infection and associated factors among women attending cervical cancer screening in setting of addis ababa" *Ethiopia. Sci Rep*
9. Luhn, Walker, Schiffman (2013) "The role of co-factors in the progression from human papillomavirus infection to cervical cancer" *Gynecol Oncol*
10. Roura, Castellsague, Pawlita (2014) "Smoking as a major risk factor for cervical cancer and pre-cancer: results from the EPIC cohort" *Int J Cancer*
11. Kayar, Goc, Cetin et al. (2025) "Impact of smoking on cervical histopathological changes in High-Risk HPV-Positive women: A matched Case-Control study" *Medicina*
12. Makuza, Nsanzimana, Muhimpundu (2015) "Prevalence and risk factors for cervical cancer and pre-cancerous lesions in Rwanda" *Pan Afr Med J*
13. Nagelhout, Van Der Ebisch (2021) "Is smoking an independent risk factor for developing cervical intra-epithelial neoplasia and cervical cancer? A systematic review and meta-analysis" *Expert Rev Anticancer Ther*
14. Bal, Goyal, Suri et al. (2012) "Detection of abnormal cervical cytology in Papanicolaou smears" *J Cytology/Indian Acad Cytologists*
15. Sachan, Singh, Patel et al. (2018) "A study on cervical cancer screening using pap smear test and clinical correlation" *Asia-Pacific J Oncol Nurs*
16. Bedell, Goldstein, Goldstein et al. (2020) "Cervical cancer screening: past, present, and future" *Sex Med Reviews*
17. Burness, Schroeder, Warren (2020) "Cervical colposcopy: indications and risk assessment" *Am Family Phys*
18. Sánchez-Siles, Remezal-Solano, López-López et al. (2020) "Prevalence of human papillomavirus in the saliva of sexually active women with cervical intraepithelial neoplasias" *Med Oral Patol Oral Cir Bucal*
19. Mbulawa, Phohlo, Garcia-Jardon (2022) "High human papillomavirus (HPV)-35 prevalence among South African women with cervical intraepithelial neoplasia warrants attention" *PLoS ONE*
20. Organista-Nava, Gómez-Gómez (2022) "Prevalence and distribution of human papillomavirus genotypes (1997-2019) and their association with cervical Cancer and precursor lesions in women from Southern Mexico" *Cancer Control*
21. Piroozmand, Zadeh, Madani (2016) "The association of high risk human papillomaviruses in patients with cervical cancer: an evidence based study on patients with squamous cell dysplasia or carcinoma for evaluation of 23 human papilloma virus genotypes" *Jundishapur J Microbiol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12424331&blobtype=pdf | # Complete genome sequence of SARS-CoV-2 isolated from a dog in Kazakhstan
Ainur Nurpeisova, Sandugash Sadikaliyeva, Zhandos Abay, Kuanysh Jekebekov, Kamshat Shorayeva, Syrym Kopeyev, Nurkuisa Rametov, Elina Kalimolda, Sabina Moldagulova, Yeraly Shayakhmetov, Alisher Omurtay, Kulyaisan Sultankulova, Nurlan Kozhabergenov, Bekbolat Usserbayev, Asankadyr Zhunushov, Aidyn Kydyrmanov, Yuriy Skiba, Bolat Yespembetov, Sergazy Nurabayev, Lespek Kutumbetov, Balzhan Myrzakhmetova, Berik Khairullin, Hansang Yoo, Aslan Kerimbayev, Markhabat Kassenov, Kunsulu Zakarya, John Dennehy
## Abstract
We report the complete genome sequence of SARS-CoV-2 isolated from a rectal swab of a dog in Almaty, Kazakhstan. Phylogenetic analysis identified the isolate as an early divergence of lineage B. These findings contribute to understanding the zoonotic potential of SARS-CoV-2 and its implications for public health.
T he ongoing SARS-CoV-2 pandemic has underscored the importance of surveillance for zoonotic reservoirs and transmission pathways. SARS-CoV-2 belongs to the family Coronaviridae, subfamily Orthocoronavirinae, genus Betacoronavirus, subgenus Sarbecovirus, and species Betacoronavirus pandemicum (1,2). In this context, companion animals such as dogs and cats have raised concern as potential hosts (3). To explore this, we conducted a study in Kazakhstan to monitor coronaviruses in domestic animals.
A complete genome of SARS-CoV-2 was obtained from a virus isolated from a domestic dog in Almaty, Kazakhstan. The isolate was recovered from a rectal swab and identified as a member of lineage B, closely related to the Delta variant (B. 1.617.2). This report provides genomic and phylogenetic data to support the ongoing SARS-CoV-2 surveillance in animal hosts.
A rectal swab was collected from a pet dog presenting mild respiratory symptoms in 2022. The RNA was extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Germany) and tested for SARS-CoV-2 using the ALSENSE-SARS-CoV-2 RT-qPCR assay (Al-Sense, Kazakhstan), targeting ORF1ab and N genes. Sample collection was approved by the Institutional Ethics Committee of the Research Institute for Biological Safety Problems (protocol #5, 29 November 2021).
Virus isolation was performed on Vero cells (ATCC CRL-1586) by carrying out three blind passages under standard conditions. Cytopathic effects appeared by the third passage. Clarified and concentrated supernatants were used for RNA extraction, and cDNA synthesis was carried out using the Ion Torrent NGS Reverse Transcription Kit. Fragmentation and adapter ligation were performed with the Ion Plus Fragment Library Kit, and amplified libraries were purified and quantified using standard Ion Torrent protocols. Size-selected DNA fragments (350-500 bp) were sequenced on the Ion GeneStudio S5 system using an Ion 530 Chip.
Read processing, trimming, and reference-based assembly (MN908947.3) were completed using Torrent Suite Software v5.12 and UGENE v52. No host genome filtering was performed prior to analysis, as the virus was propagated in Vero cells (originating from Chlorocebus sabaeus). A total of 3,217,609 high-quality reads (average length: 310 bp) were generated and aligned to the SARS-CoV-2 reference genome (GenBank: MN908947.3), resulting in 100% genome coverage. The complete genome of the isolate SARS-CoV-2/Canis lupus familiaris/KAZ/CCoV_Almaty_KZ_2022/2022 is 29,903 bp in length with a GC content of 38.0% and an average sequencing depth of 2,530×.
Phylogenetic analysis placed the isolate (SARS-CoV-2/Canis lupus familiaris/KAZ/ CCoV_Almaty_KZ_2022/2022) within lineage and in close proximity to Delta variant strains (B.1.617.2) circulating in 2021-2022 (Fig. 1).
The isolate exhibits mutations, including 10 in the spike (S) gene and 8 in ORF1ab, notably S:D614G and ORF1ab:K1037T, suggesting early divergence within lineage B (Table 1).
According to the phylogenetic analysis based on the complete genome (Fig. 1), the strain is genetically closest to human isolates from Europe and Asia, including strains from Germany, France, and the United Kingdom, showing up to 99% nucleotide identity (based on the GenBank accession numbers shown in the phylogenetic tree). Several unique amino acid substitutions were also detected (Table 1), consistent with patterns observed in other animal-derived SARS-CoV-2 isolates.
## References
1. Abay, Sadikaliyeva, Nurpeisova et al. (2024) "Breaking the barrier: SARS-CoV-2 infections in wild and companion animals and their implications for public health" *Viruses*
2. Dhama, Patel, Sharun et al. (2020) "SARS-CoV-2 jumping the species barrier: zoonotic lessons from SARS, MERS and recent advances to combat this pandemic virus" *Travel Med Infect Dis*
3. Ferasin, Fritz, Ferasin et al. (2021) "Infection with SARS-CoV-2 variant B.1.1.7 detected in a group of dogs and cats with suspected myocarditis" *Veterinary Record*
4. Tamura, Stecher, Kumar (2021) "MEGA11: molecular evolutionary genetics analysis version 11" *Mol Biol Evol*
6. Tamura, Nei, Kumar (2004) "Prospects for inferring very large phylogenies by using the neighbor-joining method" *Proc Natl Acad Sci*
7. Saitou, Nei (1987) "The neighbor-joining method: a new method for reconstructing phylogenetic trees" *Mol Biol Evol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12704534&blobtype=pdf | # About the Author
Bergo
## Abstract
His research interests include public health, with an emphasis on the ecology of diseases transmitted by insect vectors.
## RESEARCH LETTERS
The confirmed vector competence of Ae. albopictus mosquitoes for YFV under experimental conditions (10), combined with our findings, highlights its potential epidemiologic role at the sylvatic-urban interface. Our findings also underscore the importance of enhancing entomological surveillance in urban green areas to detect shifts in transmission dynamics early and prevent the re-urbanization of yellow fever in Brazil.
## Molecular Evidence of Dengue Virus Serotype 2 in Travelers Returning to Israel from the Sinai Peninsula
Neta S. Zuckerman,foot_0 Guy Choshen, 1 Yaniv Lustig, Anna Shoykhet, Keren Friedman, Tatyana Kushnir, Ora Halutz, Hovav Azulay, Victoria Indenbaum, 1 Eli Schwartz 1 Author affiliations: Tel-Aviv University, Tel Aviv, Israel
## RESEARCH LETTERS
D engue virus (DENV) is the most widespread ar- bovirus globally; its incidence has increased tenfold in the past 2 decades, largely driven by climate change and globalization (1). Although transmission is well documented in Southeast Asia and the Americas, autochthonous emergence is increasingly reported in nonendemic regions, including Europe.
We report 4 confirmed dengue fever cases in travelers returning to Israel after visiting Sharm El-Sheikh, a desert resort city in South Sinai, Egypt, during April-June 2024. Sharm El-Sheikh has not previously been recognized as an area of dengue transmission, the arid environment of the Sinai Peninsula is considered unfavorable for the DENV primary vectors, Aedes mosquitoes.
The cases (Table ) were unrelated; travel dates were nonoverlapping and accommodations varied and were located 3-25 km apart. Patients had typical dengue symptoms such as fever, headache, myalgia, and rash. All were hospitalized, received supportive care, and recovered. One patient exhibited meningeal irritation; cerebrospinal fluid testing results were unremarkable, although DENV serotype 2 (DENV-2) RNA was detected by quantitative real-time PCR (cycle threshold 32.5). All samples were collected within 1 week of symptom onset. Serum testing confirmed DENV-2 by multiplex quantitative real-time PCR (2); additional nonstructural protein 1 antigen and IgM/ IgG positivity was detected in some cases.
To explore the geographic origin of the DENV-2 cases, we performed DENV whole-genome sequencing. We captured DENV-2 using specific whole-genome primers (https://grubaughlab.com/ open-science/amplicon-sequencing); we prepared sequencing libraries using Nextera-XT and ran them on the Illumina NovaSeq (https://www.illumina.com). We generated consensus sequences by mapping to the DENV-2 reference genome (GenBank accession no. NC_001474.2) and deposited resulting sequences into GenBank (Appendix Our findings describe 4 confirmed DENV-2 infections in travelers from Sharm El-Sheikh, Egypt, a city in the arid Sinai Peninsula, previously considered unsuitable for Aedes mosquitoes and without previous dengue reports. Genomic analysis showed clustering of cases, likely from a single outbreak, most closely related to strains from Pakistan. Aside from 1 United Arab Emirates 2023 sequence clustering separately, no recent data from Sinai are available, underscoring a major surveillance gap. Those results align with reports of DENV-2 spread along the Red Sea and recent cases in Florence, Italy (8).
During the past 2 decades, Ae. aegypti mosquito populations have expanded in Egypt, especially along the Red Sea coast (Figure, panel B), correlating with dengue outbreaks. However, no entomologic data exist for Sinai. The arid climate challenges mosquito survival, but clustering of cases in 1 resort area suggests local adaptation, possibly supported by urban microhabitats (9). Maritime and air travel might drive
We report 4 dengue cases in travelers returning to Israel from Sharm-El-Sheikh, Egypt, all confirmed as dengue virus type 2 infections. Phylogenetic analysis showed clustering with strains from Pakistan. Our findings provide molecular evidence of dengue circulation in the Sinai desert, highlighting the need for increased awareness among travelers and health authorities.
## RESEARCH LETTERS
analysis shows that our dengue sequences are closest to recent strains from Pakistan. However, the scarcity of sequences from Egypt and neighboring regions limits inference on viral origin, circulation, and distribution, and observed variability suggests undersampling and additional undetected cases. This report of 4 cases over 3 months in different localities of Sharm El-Sheikh suggests sustained DENV-2 transmission and emphasizes the importance of enhanced vector surveillance and control, providing an alert to public health authorities. The genetic data presented might help address gaps in regional DENV sequence reporting and contribute to understanding its molecular epidemiology and origins.
## About the Author
## References
1. Postler, Beer, Blitvich et al. (2023) "Renaming of the genus Flavivirus to Orthoflavivirus and extension of binomial species names within the family Flaviviridae" *Arch Virol*
2. Monath, Vasconcelos (2015) "Yellow fever" *J Clin Virol*
3. Giovanetti, Pinotti, Zanluca et al. (2023) "Genomic epidemiology unveils the dynamics and spatial corridor behind the yellow fever virus outbreak in southern Brazil" *Sci Adv*
4. Andrade, Campos, Oliveira et al. (2022) "Fast surveillance response reveals the introduction of a new yellow fever virus sub-lineage in 2021" *Mem Inst Oswaldo Cruz*
5. Saad, Chiaravalloti-Neto (2024) "Reemergence of yellow fever in the state of São Paulo: the structuring role of surveillance of epizootics in non-human primates in a one health approach" *Rev Bras Epidemiol*
6. Caleiro, Vilela, Nuevo et al. (2016) "Yellow fever virus (YFV) detection in different species of culicids collected during an outbreak in southeastern Brazil" *Trop Med Infect Dis*
7. Patel, Landt, Kaiser et al. (2013) "Development of one-step quantitative reverse transcription PCR for the rapid detection of flaviviruses" *Virol J*
8. Domingo, Patel, Yillah et al. (2012) "Advanced yellow fever virus genome detection in point-of-care facilities and reference laboratories" *J Clin Microbiol*
9. De Oliveira, Andrade, Campos et al. (2021) "Yellow fever virus maintained by Sabethes mosquitoes during the dry season in Cerrado, a semiarid region of Brazil" *Viruses*
10. Couto-Lima, Madec, Bersot et al. (2017) "Potential risk of re-emergence of urban transmission of yellow fever virus in Brazil facilitated by competent Aedes populations" *Sci Rep*
11. (2025) "São Paulo 01027-000, Brazil; email: karink@usp.br References 1. World Health Organization. Dengue and severe dengue"
12. Santiago, Vergne, Quiles et al. (2013) "Analytical and clinical performance of the CDC real time RT-PCR assay for detection and typing of dengue virus" *PLoS Negl Trop Dis*
13. Kholedi, Balubaid, Milaat et al. (2012) "Factors associated with the spread of dengue fever in Jeddah Governorate, Saudi Arabia" *East Mediterr Health J*
14. El-Kady, Osman, Alemam et al. (2022) "Circulation of dengue virus serotype 2 in humans and mosquitoes during an outbreak in El Quseir City" *Egypt. Infect Drug Resist*
15. Desogi, Ali, Gindeel et al. (2023) "Detection of dengue virus serotype 4 in Sudan" *East Mediterr Health J*
16. Dafalla, Abdulhaq, Almutairi et al. (2023) "The emergence of an imported variant of dengue virus serotype 2 in the Jazan region, southwestern Saudi Arabia" *Trop Dis Travel Med Vaccines*
17. Frank, Lachmann, Wilking et al. (2023) "Increase in dengue fever in travellers returning from Egypt" *Euro Surveill*
18. Manciulli, Zammarchi, Lagi et al. (2024) "Emergence of dengue fever: sentinel travellers uncover outbreak" *J Travel Med*
19. Newman, Feng, Onland et al. (2024) "Defining the roles of local precipitation and anthropogenic water sources in driving the abundance of Aedes aegypti, an emerging disease vector in urban, arid landscapes" *Sci Rep*
20. El-Kafrawy, Sohrab, Ela et al. (2016) "Multiple introductions of dengue 2 virus strains into Saudi Arabia from 1992 to 2014" *Vector Borne Zoonotic Dis*
21. Zuckerman "Central Virology Laboratory" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12506063&blobtype=pdf | # PRRSV promotes bacterial infection by remodeling actin cytoskeleton and cell membrane proteins
Xiao Liu, Fang Lv, Yanan Zhu, Yinan Meng, Bo Peng, Zifang Zheng, Yang Li, Lele Xu, Yingtong Feng, Jianwu Zhang, Shuqi Xiao
## Abstract
Secondary infection is a worldwide problem in the prevention and control of viral infection. Secondary bacterial infection induced by porcine reproductive and respiratory syndrome virus (PRRSV) infection causes enormous economic losses, but the relevant mechanism remains unclear. We found that the infection of Klebsiella pneumoniae or Streptococcus suis type 2 in the lungs of PRRSV-challenged piglets was significantly higher than the controls, and the infection of PRRSV, influenza A virus H1N1 (H1N1), and porcine circovirus type 2 (PCV2) also significantly increased the infection of the bacteria in vitro. Transcriptomic analysis revealed that PRRSV infection significantly altered the expression of cytoskeleton-related proteins, among which the expression of actin-binding protein filamin A (FLNA) was significantly increased, and knockdown of FLNA could significantly reduce bacterial invasion. Mechanistic studies found that FLNA drives actin cytoskeleton rearrangement by promoting F-actin generation, thereby facilitating bacterial invasion. Further studies found that PRRSV promoted bacterial adhesion by upregulating the expression of integrin α5 (ITGα5). ITGα5 could induce actin cytoskeleton rearrangement by promoting FLNA expression, thus aggravating bacterial invasion. Furthermore, we found that lentiviral shRNA-mediated knockdown of FLNA or ITGα5 significantly reduced bacterial infection in the lungs of mice and protected mice from death. These results suggest that the regulation of actin cytoskeleton and cell membrane proteins may be a conserved mechanism of virus-induced secondary bacterial infection.IMPORTANCE An important reason why porcine reproductive and respiratory syndrome virus (PRRSV) is difficult to control effectively is that it often causes severe secondary bacterial infections, which are usually attributed to the immunosuppression caused by PRRSV. However, the mechanism by which PRRSV infection leads to increased susceptibil ity of cells to bacterial infection has been largely overlooked. We revealed that PRRSV induced actin cytoskeleton rearrangement by upregulating FLNA expression, thereby aggravating bacterial invasion. PRRSV increased bacterial adhesion by promoting the ITGα5 expression, and the upregulation of ITGα5 could induce FLNA-mediated actin cytoskeleton rearrangement. Furthermore, we found that H1N1 and porcine circovirus type 2 infection also significantly promoted the expression of FLNA and ITGα5 and increased the infection of multiple bacteria. These results suggest that FLNA and ITGα5 play important roles in virus-induced secondary bacterial infection.
respiratory syndrome virus (PRRSV) infection does not usually cause acute mortality in pigs, but secondary bacterial infection often results in high mortality in weaned piglets (6). In a piglet infection model, PRRSV infection significantly promoted bacterial infection and increased the mortality of piglets (7). However, the underlying mechanism of the viral-bacterial synergy that aggravates disease progression and mortality remains elusive, which hampers the generation of effective prophylactic and therapeutic options.
PRRSV has caused great harm to the pig industry worldwide (8), and the rapid mutation and frequent recombination of the virus have challenged clinical prevention and control (9,10). As the main primary pathogen, PRRSV is prone to secondary infection with other pathogens (11), resulting in severe economic losses. Furthermore, during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, secondary bacterial infection has led to severe inflammatory cytokine storm and lung tissue damage (12), increasing the difficulty of clinical treatment. This highlights the impact of secondary bacterial infections on both animal and human infections.
Aerococcus viridans (13), Klebsiella pneumoniae (14), Streptococcus suis type 2 (15), and Staphylococcus aureus (16) are the major pathogens causing pneumonia in humans and animals. These bacteria rarely infect animals with a healthy immune system (17). However, the infection rate is greatly increased in the presence of other primary pathogens and underlying diseases.
This study investigated the molecular mechanisms by which respiratory viruses promote bacterial infections. The findings revealed that FLNA facilitates bacterial invasion by inducing actin cytoskeleton rearrangement, and PRRSV promotes bacterial adhesion by upregulating integrin α5 (ITGα5). Further studies showed that ITGα5 could induce actin cytoskeleton rearrangement by enhancing FLNA expression, thus promot ing bacterial invasion. These findings indicate that FLNA and ITGα5 play critical roles in secondary bacterial infections following viral infection, and interventions targeting these mechanisms may be candidate options to cope with secondary bacterial infection.
## RESULTS
## PRRSV promotes bacterial infection in the lungs of piglets
Some studies have reported that viruses promote bacterial infection (1,18,19), and it has also been observed that PRRSV-positive piglets are prone to coinfection with bacteria in clinical practice (20). To investigate the effect of PRRSV on bacterial infection, piglets were infected with PRRSV or treated with phosphate buffered saline (PBS). The body temperature of the piglets increased significantly on Day 1.5 after PRRSV infection and gradually returned to normal on Day 4 (Fig. 1A). This indicates that piglets will recover from PRRSV infection, then piglets were infected with the gram-negative representative strain K. pneumoniae or the gram-positive representative strain S. suis 2 on Day 5, which often cause pneumonia in piglets. Piglets in the PRRSV + bacteria groups began to die on Day 5.5 (Fig. S1A; Table 1). All piglets were sacrificed on Day 6, after which the number of bacteria in the lungs was tested. As shown in Fig. 1B, PRRSV replicated efficiently in the porcine alveolar macrophages (PAMs) of piglets, and quantification of the adherent bacteria in the lungs of piglets by CFU revealed that the numbers of K. pneumoniae or S. suis 2 in the PRRSV-challenged groups were significantly greater compared with control groups (Fig. 1C). PRRSV-challenged piglets exhibited more severe pulmonary pathological lesions after bacterial infection (Fig. S1B). Severe hemorrhagic necrosis was observed in the lung tissue of piglets in the PRRSV + K. pneumoniae group, and more severe fibrin exudation was observed in the PRRSV + S. suis 2 group (Fig. S1C). Immunofluorescence and immunohistochemistry were used to detect K. pneumoniae and S. suis 2 infections in the lungs of piglets and revealed that the PRRSV-challenged group exhibited higher infection levels compared to the control group (Fig. 1D andE). Subsequently, transmission electron microscopy analysis of bacterial infection in piglet lung tissues revealed a significantly higher bacterial load in the PRRSV-infected group compared to the control group (Fig. 1F). Notably, although neither K. pneumoniae nor S. suis 2 is classified as an obligate intracellular pathogen, both were observed to invade and localize inside cells. These results suggest that PRRSV significantly increases the infection of K. pneumoniae and S. suis 2.
## PRRSV infection upregulates FLNA expression
PRRSV promotes bacterial infection, which is generally thought to be caused by immunosuppression after viral infection (21). However, the mechanism by which PRRSV infection leads to increased susceptibility of cells to bacteria remains unclear, and we found that PRRSV infection promotes bacterial invasion into cells (Fig. 1F). To explore how PRRSV increases bacterial invasion, we inoculated A. viridans, K. pneumoniae, S. suis 2, or S. aureus into PRRSV-infected or noninfected PAMs. Bacterial invasion quanti fied using CFU and the results revealed that PRRSV infection significantly increased the infection of A. viridans, K. pneumoniae, and S. suis 2, but not S. aureus (Fig. 2A). To further investigate the effect of PRRSV infection on bacterial invasion, we infected 3D4/21 (immortalized PAMs) cells with PRRSV for 36 h, then infected the cells with K. pneumoniae or S. suis 2 and detected bacterial invasion. The results showed that the number of K. pneumoniae or S. suis 2 was higher in PRRSV-infected cells (red boxes) than in uninfected cells (white boxes) (Fig. 2B). The above results indicated that PRRSV infection significantly enhanced bacterial invasion. The invasion of bacteria into cells depends on endocytosis, which requires the involvement of cytoskeleton-rela ted proteins. Therefore, we performed transcriptomic analysis of PRRSV-infected PAMs to determine the expression changes of cytoskeleton-associated proteins following virus infection. The results showed that multiple cytoskeleton-associated proteins were significantly upregulated after PRRSV infection (Fig. 2C). Subsequently, we selected the five most prominently upregulated molecules after PRRSV infection for knockdown and examined their effects on bacterial invasion. Notably, FLNA knockdown significantly suppressed bacterial infection (Fig. 2D). We then examined the effect of PRRSV infec tion on FLNA expression and found that PRRSV infection significantly promoted FLNA expression in 3D4/21 and Marc145 cells (Fig. 2E andF). These results suggest that PRRSV may increase bacterial infection by promoting FLNA expression.
## FLNA knockdown significantly inhibited bacterial invasion
To further investigate the effect of FLNA on bacterial infection, 3D4/21 and Marc145 cells were transfected with FLNA siRNA for 24 h, then the cells were inoculated with A. viridans, K. pneumoniae, S. suis 2, or S. aureus, respectively. After infection for 4 h, bacterial invasion was tested by CFU. As shown in Fig. 3A andB, FLNA knockdown significantly inhibited the invasion of A. viridans, K. pneumoniae, and S. suis 2, but did not suppress that of S. aureus. Next, we constructed a Marc145-FLNA +/-knockout cell line, infected it with bacteria, and measured bacterial invasion with CFU 4 h after infection. The results showed that the invasion of A. viridans, K. pneumoniae, and S. suis 2 was significantly reduced in Marc145-FLNA +/-cells (Fig. 3C). These results suggest that FLNA plays a key role in bacterial invasion.
To explore the influence of other respiratory viruses on bacterial infection, we infected cells with H1N1 or PCV2, then inoculated with A. viridans, K. pneumoniae, S. suis 2, and S. aureus, and bacterial invasion was quantified using CFU 4 h after infection. The results showed a significant increase in infections of A. viridans and K. pneumoniae in H1N1-infected and PCV2-infected cells, while there was no significant enhancement in S. suis 2 and S. aureus (Fig. 4A andB). Further examination of the effects of H1N1 and PCV2 infection on FLNA expression showed that H1N1 and PCV2 infection significantly promoted FLNA expression (Fig. 4C andD). Knockdown of FLNA in MDCK and PK-15 cells, which support their proliferation, showed that FLNA knockdown significantly decreased bacterial invasion (Fig. 4E andF). These results indicate that FLNA plays a decisive role in the process of respiratory virus-induced bacterial infection.
## FLNA increases bacterial invasion by regulating F-actin production
The entry of bacteria into cells through endocytosis requires the involvement of the actin cytoskeleton (22). To investigate the effect of FLNA on the actin cytoskeleton, we transfected 3D4/21 cells with FLNA siRNA and detected the expression levels of globular actin (G-actin) and filamentous actin (F-actin) in the cells by a G/F-actin Kit. As shown in Fig. 5A through C, FLNA knockdown significantly reduced F-actin production at different time points. Immunofluorescence detection of F-actin showed that FLNA knockdown promoted the depolymerization of F-actin into G-actin and induced the rearrangement of the actin cytoskeleton (Fig. 5D). These results suggest that FLNA knockdown can induce actin cytoskeleton rearrangement. To further investigate the effect of F-actin production on bacterial infection, we treated the cells with two inhibitors that induce actin cytoskeleton rearrangement, Jasplakinolide (which stabilizes actin polymerization) and Latrunculin A (which inhibits actin polymerization), and then F-actin generation levels and bacterial invasion were detected. As shown in Fig. 5E andF, Jasplakinolide treatment increased F-actin production, whereas Latrunculin A decreased F-actin levels. Notably, Jasplakinolide significantly enhanced bacterial invasion, while Latrunculin A markedly suppressed it, suggesting that inhibition of F-actin production could induce actin cytoskeleton rearrangement, thereby suppressing bacterial invasion. To study the effect of FLNA on bacterial adhesion, 3D4/21 cells were transfected with FLNA siRNA, then infected with bacteria, and bacterial adhesion was detected 2 h after infection. As shown in Fig. 5G, FLNA knockdown only inhibited the adhesion of K. pneumoniae, but had no significant effect on the adhesion of other bacteria.
## PRRSV promotes bacterial adhesion by upregulating ITGα5 expression
Bacterial entry into cells involves two processes: adhesion and invasion, where the bacteria need to first attach to the cell surface and then enter the cell through endocytosis. Bacterial invasion has been explored in the previous sections, but how PRRSV promotes bacterial adhesion remains unclear. It has been reported that integrins are important transmembrane receptors on the cell surface and play an important role in cell adhesion (23). Therefore, we hypothesize that integrins may play an important role in enhancing bacterial adhesion. We infected 3D4/21 cells with PRRSV and examined the expression of 18 α and 8 β subunits of the integrin family. The results showed that the expressions of ITGα2, ITGα5, and ITGα10 were significantly upregulated after PRRSV infection (Fig. 6A). To detect the effect of ITGα2, ITGα5, and ITGα10 on bacterial adhesion, their overexpression plasmids were transfected into 3D4/21 cells, and bacterial adhesion was measured by CFU. The results found that ITGα5 overexpression signifi cantly promoted the adhesion of A. viridans, K. pneumoniae, and S. suis 2, but did not affect the adhesion of S. aureus. In addition, the overexpression of ITGα10 also markedly promoted the adhesion of K. pneumoniae (Fig. 6B). To further verify the effect of ITGα5 on bacterial adhesion, 3D4/21 and Marc145 cells were transfected with ITGα5 siRNA or an overexpression plasmid. The cells were then infected with bacteria, and the bacterial adhesion was determined by CFU 2 h after infection. As shown in Fig. 6C andD, ITGα5 overexpression significantly increased the adhesion of A. viridans, K. pneumoniae, and S. suis 2, and ITGα5 knockdown suppressed their adhesion. Subsequently, Marc145-ITGα5 +/- knockout cells were generated to evaluate bacterial adhesion. The results showed that bacterial adhesion was significantly inhibited in Marc145-ITGα5 +/-cells, while reintroduc tion of ITGα5 expression markedly enhanced bacterial adhesion to the cells (Fig. 6E). These results suggest that PRRSV promotes bacterial adhesion by upregulating ITGα5 expression.
## ITGα5 increases bacterial invasion by promoting F-actin production
To further investigate whether ITGα5 influences bacterial invasion, we transfected Marc145 and 3D4/21 cells with ITGα5 siRNA or an overexpression plasmid. The cells were then infected with A. viridans, K. pneumoniae, S. suis 2, or S. aureus, and the bacterial invasion was quantified with CFU 4 h after infection. As shown in Fig. 7A through C, the knockdown of ITGα5 significantly reduced the invasion of A. viridans, K. pneumoniae, and S. suis 2, while the overexpression of ITGα5 significantly increased their invasion, but did not affect S. aureus. The effect of ITGα5 on bacterial invasion was observed by transmission electron microscopy. As shown in Fig. 7D, ITGα5 knockdown significantly inhibited the invasion of A. viridans, K. pneumoniae, and S. suis 2, while its overexpression promoted their invasion. A similar trend was also obtained when the effect of ITGα5 on bacterial invasion was examined using immunofluorescence (Fig. 7E). To further examine the impact of ITGα5 on G/F-actin levels in 3D4/21 cells, the results revealed that ITGα5 knockdown suppressed F-actin production, while ITGα5 overexpression enhanced F-actin generation (Fig. 7F). Consistent findings were observed via F-actin immunofluor escence (Fig. 7G). Next, we examined the effects of PRRSV, H1N1, and PCV2 infections on ITGα5 expression and found that their infections significantly upregulated ITGα5 expression at distinct time points (Fig. S2). These results indicate that ITGα5 can promote bacterial invasion by inducing actin cytoskeleton rearrangement.
## ITGα5 interacts with FLNA and promotes its expression
Our above findings indicate that FLNA promotes bacterial invasion by facilitating F-actin generation, and ITGα5 could also enhance bacterial invasion through increasing F-actin production. It has been reported that ITGα5 plays an important role in the regulation of actin cytoskeleton (24), so we hypothesize that ITGα5 may promote bacterial invasion by regulating FLNA expression. To explore whether ITGα5 could regulate FLNA expression, we transfected ITGα5 overexpression plasmid and siRNA into cells to detect the effect of ITGα5 on FLNA expression. The results showed that ITGα5 overexpression significantly promoted FLNA expression, while its knockdown inhibited FLNA expression (Fig. 8A).
To determine how ITGα5 promotes FLNA, the ITGα5 full-length plasmid (Flag-ITGα5), extracellular domain (Flag-EX), transmembrane region (GFP-TM), or intracellular domain (GFP-IN) plasmids were subsequently transfected into 3D4/21 cells to identify domains that interact with FLNA. As shown in Fig. 8B andC, both the ITGα5 full-length plas mid and the extracellular domain interacted with FLNA, but the other domains did not, indicating that the extracellular domain is the key domain for interacting with FLNA. Their colocalization was further confirmed via immunofluorescence (Fig. 8D). These results suggest that ITGα5 interacts with FLNA and significantly promotes FLNA expression, thereby promoting bacterial invasion. To further verify the role of ITGα5 in PRRSV promoting FLNA expression, we transfected ITGα5 siRNA or overexpression plasmid into PRRSV-infected 3D4/21 cells and detected FLNA expression 36 h after infection. The results showed that PRRSV infection significantly upregulated FLNA expression, while PRRSV infection combined with ITGα5 siRNA transfection markedly inhibited this FLNA upregulation (Fig. 8E andF), suggesting that ITGα5 is a key mediator in PRRSV-induced FLNA expression enhancement.
## Knockdown of FLNA and ITGα5 inhibited bacterial infection in the lungs of mice
To further verify the effects of FLNA and ITGα5 on bacterial infection in vivo, we constructed shRNA-mediated knockdown mouse models by the pLKO.1 lentiviral vector, then infected these mice with A. viridans, K. pneumoniae, or S. suis 2. The PBS + bacteria groups of mice began to die on Day 0.5 after bacterial infection (Fig. 9A and Table 2), and all mice were sacrificed on Day 1. Detection of FLNA and ITGα5 expression in the lungs of mice showed that shRNA lentivirus infection significantly reduced their expression levels (Fig. S3). Quantification of bacterial adhesion using CFU showed that knockdown of ITGα5 significantly reduced the adhesion of A. viridans, K. pneumoniae, and S. suis 2 in murine lungs (Fig. 9B). Quantification of bacterial invasion by CFU revealed that knockdown of FLNA or ITGα5 markedly decreased the loads of these three bacterial species (Fig. 9C), and a similar trend was observed by transmission electron microscopy (Fig. 9D). Subsequent immunofluorescence and immunohistochemistry analyses of K. pneumoniae and S. suis 2 infections in the lungs of mice demonstrated that knockdown of either FLNA or ITGα5 significantly reduced pulmonary bacterial burden (Fig. 9E andF). These results suggest that FLNA and ITGα5 play critical roles in bacterial infection in mice.
## PRRSV inhibits the transcription factor HOXD8 to promote ITGα5 expression
To explore how PRRSV upregulates the expression of ITGα5, we studied the tran scriptional regulatory mechanism of ITGα5 and identified the essential cis-regulatory elements of the ITGα5 promoter. The -0.3, -0.6, -0.9, -1.2, -1.5, and -1.8 kb regions were segmented to construct the PGL4.10 vector, which was subsequently transfec ted into 3D4/21 cells to assess promoter activity (25). As shown in Fig. 10A, there was a significant decrease in promoter activity from -1.5 kb to -1.2 kb, indicating the presence of key active sites in this region. We further truncated the region and identified the ITGα5 promoter in the -1.25 kb to -1.2 kb region. The sequence was submitted to the bioinformatic tool PROMO to predict transcription factor binding sites (Fig. 10B). Knockdown and overexpression of the predicted transcription factors were performed, and the expression level of ITGα5 was detected. The results showed that HOXD8 knockdown significantly promoted the expression of ITGα5, while overexpression markedly inhibited its expression, indicating that HOXD8 is a negative transcriptional regulator of ITGα5 (Fig. 10C andD). Consistent results were also obtained for HOXD8 knockdown and overexpression at different time points (Fig. 10E andF). Next, we examined HOXD8 expression level after PRRSV infection, and the results showed that PRRSV infection significantly reduced HOXD8 expression at different time points (Fig. 10G and H). These results indicated that HOXD8 is a negative transcriptional regulator of ITGα5, and PRRSV promoted ITGα5 expression by inhibiting the expression of HOXD8.
## PRRSV Nsp10 interacts with HOXD8 and decreases its expression
To study the major viral fragment that downregulated HOXD8 expression, nonstructural and structural proteins of PRRSV were transfected into 3D4/21 cells, and the expres sion level of HOXD8 was detected after 24 h. The results showed that Nsp10 was the most significant fragment that inhibited HOXD8 expression (Fig. 11A). We subsequently co-transfected HEK-293T cells with HOXD8 and PRRSV Nsp10 or Nsp1α to perform coimmunoprecipitation assays 24 h after transfection. As shown in Fig. 11B, Nsp10 interacts with HOXD8, whereas Nsp1α does not. Next, we infected Marc145 cells with PRRSV and detected the colocalization of Nsp10 and HOXD8 by immunofluorescence 24 h after infection. The results showed that Nsp10 was colocalized with HOXD8 (Fig. 11C).
## DISCUSSION
Bacterial adhesion and colonization are prerequisites for the establishment of infection. Secondary bacterial infection caused the majority of associated deaths during the SARS-CoV-2 pandemic and influenza outbreaks (26)(27)(28). As a precursor to infection, viruses may damage bronchial and alveolar epithelial cells, paving the way for bacterial infection (29,30). Studies have reported that neuraminidase (NA) of influenza A virus (IAV) activates the TGF-β signaling pathway and promotes the expression of cellular adhesins, leading to an increase in bacterial colonization and promoting secondary bacterial infection (31). In this study, we found that PRRSV, H1N1, and PCV2 infections could significantly aggravate multiple bacterial infections, and infections of PRRSV, H1N1, and PCV2 all significantly increased the expressions of FLNA and ITGα5. We found that PRRSV can upregulate the expression of ITGα5 by inhibiting the expression of HOXD8 (Fig. 10), and H1N1 and PCV2 can also suppress the expression of HOXD8 (data not shown). These results suggest that this might be a conserved mechanism by which respiratory viruses exacerbate bacterial infections (32,33). In vivo experiments showed that FLNA and ITGα5 knockdown significantly reduced pulmonary bacterial burden in the lungs of mice. These results suggest that FLNA and ITGα5 are critical factors for bacterial infection and that the regulation of actin cytoskeleton and cell membrane proteins may be a mediator of virus-bacteria interactions. Bacterial superinfection following virus infection is a common complication (34), and K. pneumoniae is one of the most harmful pathogens, often leading to severe pneumonia, and has become a serious public health threat worldwide with the spread of drug-resistant strains (35). We observed that the knockdown of ITGα5 almost completely eliminated K. pneumoniae infection, whereas cell susceptibility to K. pneumoniae was restored after the ectopic expression of ITGα5. These findings indicate that ITGα5 may be an adhesion receptor for K. pneumoniae infection and a bridge for its infection. We found that ITGα5 can induce actin cytoskeleton rearrangement by upregulating FLNA expression, thereby facilitating bacterial invasion. The amount of F-actin and G-actin in the cell membrane is dynamically stable in the normal physiological state, which is an essential condition for maintaining cell endocytosis and movement (36,37). Any pathological factor that disrupts this homeostasis may aggravate pathogenic infection (38). The endocytosis of cells can be used by the virions of Hendra virus and Newcastle disease virus to increase their own entry and infection efficiency (39), while the T3SS translocation of SipC and the effector SipA are crucial for Salmonella infection by disrupting the host cytoskeleton (22). These studies indicate that the regulation of the actin cytoskeleton by viruses is an important factor leading to increased pathogen infection. In summary, we found that PRRSV induced actin cytoskeleton rearrangement by upregulating FLNA expression, thereby aggravating bacterial invasion. PRRSV increased bacterial adhesion by promoting the ITGα5 expression, and the upregulation of ITGα5 could induce FLNA-mediated actin cytoskeleton rearrangement. The promotion of ITGα5 expression by PRRSV is dependent on the repression of the transcription factor HOXD8 (Fig. 12). Furthermore, we found that H1N1 and PCV2 infection also signifi cantly promoted the expression of FLNA and ITGα5 and increased the infection of multiple bacteria. Lentiviral shRNA-mediated knockdown of FLNA or ITGα5 significantly reduced bacterial infection in the lungs of mice and protected mice from death. Our research indicates that the regulation of actin cytoskeleton and cell membrane proteins during viral infection is critical for secondary bacterial infection, which provides a new perspective for the control of secondary bacterial infection in the clinic.
## MATERIALS AND METHODS
## Virus, bacteria, and cells
The PRRSV GD-HD strain (GenBank: KP793736) and porcine circovirus type 2d (PCV2) SX-XY18 strain (GenBank: PP668228) used in this study were isolated and preserved in our laboratory (Lanzhou Veterinary Research Institute, Lanzhou, China) (40). The influenza A virus (H1N1) SN13 strain (GenBank: MN418766) (41) was gifted by Professor Honglei Sun from China Agricultural University. The A. viridans, K. pneumoniae, S. suis 2, and S. aureus strains were isolated from clinical samples and stored in our labora tory. PAMs were isolated from 5-week-old PRRSV antigen and antibody-negative piglets and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS). Marc145, 3D4/21, HEK-293T, MDCK, and PK-15 cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) containing 10% FBS at 37°C with 5% CO 2 .
## Antibodies and reagents
Anti-PRRSV N, anti-S. suis 2 monoclonal antibodies and Nsp1α, Nsp10 polyclonal antibodies were prepared and stored in our laboratory. Anti-GAPDH, anti-HA, anti-Flag, and anti-GFP monoclonal antibodies were obtained from TransGen Biotech. Anti-HOXD8 (A16877) polyclonal antibodies were obtained from ABclonal Technology. Anti-ITGα5 (98204), anti-p-AKT (4060), and anti-FLNA (44873) monoclonal antibodies were obtained from Cell Signaling Technology. Anti-ITGα5 (Ab150361) monoclonal antibodies and anti-K. pneumoniae antibodies (Ab20947) were obtained from Abcam. Anti-AKT (51077-1), anti-mouse IgG (labeled with Alexa Fluor 488), and anti-rabbit IgG (labeled with Alexa Fluor 594) antibodies were obtained from Proteintech. TRITC phalloidin was purchased from Solarbio Technology, a G/F-actin Kit (BK037) was obtained from Cytoskeleton, Inc., and anti-Flag and anti-HA magnetic beads and inhibitors were purchased from MedChemExpress. The anti-FLNA (MA5-11705) monoclonal antibody, Lipofectamine RNAiMAX transfection reagent, and Lipofectamine 3000 transfection reagent were obtained from Thermo Fisher Scientific.
## Bacterial adhesion assay
Cells were seeded in 12-well plates, infected with PRRSV (MOI = 1) for 36 h, or transfected for 24 h. Then, the medium was replaced with serum-free DMEM, and the cells were infected with bacteria (MOI = 10) for 2 h at 37°C with 5% CO 2 . Then washed with PBS to remove unbound bacteria, the cells were digested with trypsin-ethylenediaminetetra acetic acid solution and lysed with 0.25% Triton X-100. After dilution with 1 mL of PBS, bacteria were quantified by CFU.
After infection with bacteria for 24 h, the mice or piglets were sacrificed, the mouse lungs were homogenized in 3 mL of PBS, and piglet lung samples were collected from three sites (1 g at each site) and homogenized in 3 mL of PBS. The homogenate was collected and centrifuged, then the cells were resuspended and lysed with 0.25% Triton X-100. After dilution with PBS, bacteria were quantified by CFU (42).
## Bacterial invasion assay
Cells were seeded in 12-well plates, infected with PRRSV (MOI = 1) for 36 h, or transfected for 24 h, and then infected with bacteria (MOI = 10) for 4 h at 37°C with 5% CO 2 . Then washed with PBS, the cells were incubated with DMEM containing 100 μg of gentamicin and 5 μg of penicillin/mL for 2 h to kill surface-adherent bacteria. After being washed with PBS three times, the intracellular bacteria were quantified by CFU as described in the adhesion assay.
After being infected with bacteria for 24 h, the mice were sacrificed, and the lungs were homogenized in 3 mL of PBS. The homogenate was collected and centrifuged. PBS containing 100 μg of gentamicin and 5 μg of penicillin/mL was used to resuspend the cells, then incubated at 37°C with 5% CO 2 for 2 h and washed with PBS before quantifying intracellular bacteria (42).
## Quantitative PCR
Total RNA was extracted from cells or tissues via TRIzol. cDNA was obtained by Prime Script RT Kit, and reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) analysis was subsequently performed with ChamQ SYBR qPCR Master Mix. The relative expression levels were calculated by the 2 -ΔΔCT method. Primers are shown in Table 3.
## Western blot and coimmunoprecipitation
The cells were collected and lysed, and the protein content of the samples was quantified by a BCA Kit. Then, the proteins were denatured in boiling water for 10 min, separated via 12% SDS-PAGE, transferred to polyvinylidene fluoride (PVDF) membrane, blocked in 5% skim milk at room temperature for 1 h, incubated with primary antibodies at 4°C overnight, incubated with secondary antibodies at room temperature for 1 h, and incubated with enhanced chemiluminescence (ECL) solution to detect the protein bands.
The magnetic beads were pretreated with PBST, and the whole-cell lysates were added to the magnetic beads and incubated at room temperature for 3 h. The beads were rinsed with PBST five times, then prepared samples for Western blot analysis. Occasionally, the edge of the blot can be seen as a non-uniform white background in the image because the bands were too close to the edge of the membrane.
$$-R GATCGGATGTCTGAGCCATTA pITGα5-F CTGTGCTCCCCTCTACAGTTG Pig pITGα5-R AGAATCCGGGTGAAGTTGTCT dITGα5-F GGGAGGACTGCAGAGAGATG Dog dITGα5-R CGCAGATGTTGTCTTCTCCA gmFLNA-F CATGTCACTGCCTATGGACCT Green monkey gmFLNA-R TGGGATGTGCTGTTCATTGTA pFLNA-F GTCCCTGTGCATGATGTGAC Pig pFLNA-R GGGCACATAGTTGACGGTCT dFLNA-F AAGGTATACGGGCCTGGAGT Dog dFLNA-R CTCCACGGGGCGTATACTTC gmHOXD8-F AGCAGCTCCTGGTAGACGAA Green monkey gmHOXD8-R TTAATTTGTCGGGCCTTCTG PRRSV-ORF7-F AATGGCCAGCCAGTCAATCA PRRSV PRRSV-ORF7-R TCATGCTGAGGGTGATGCTG pITGα1-F TATGCCCTGAATCAGACAAGG Pig pITGα1-R TGCATGAATTGTGCTTCAGAG pITGα2-F GTGCCTTTGGACAGGTTGTT Pig pITGα2-R TCATGGTCTTCTGCAAGCAC pITGα3-F ACATCTACCACGGCAGTTCC Pig pITGα3-R CACTCAGGGAGTAGCCGAAG pITGα4-F AACATGAGCCTGGATGTGAAC Pig pITGα4-R GCTGAAGAATTGGCTGAAGTG pITGα6-F GGCCTTATGAAGTTGGTGGA Pig pITGα6-R CCACCACTGCCACATCATAG pITGα7-F AGCAGAGGAGCTGAGCTTTG Pig pITGα7-R GTAGGGAGCACCCACTACCA pITGα8-F GGCAGATACCCTTTGACAACA Pig pITGα8-R CTGTTGCTCCAAACCATTGAT pITGα9-F ATCACGTCTCCAACCTCCTTT Pig pITGα9-R ATGACACTCCAGGTCATCCAG pITGα10-F GTGAGAGCTTCCTGGAGGTG Pig pITGα10-R CAAGCTTCCAAAGGCAAAAG pITGα11-F ATGGCGTGACTGATGTCCTAC Pig pITGα11-R CTGCTTTGGTGTCTTCAGGAG pITGαD-F AGATCCGTGTATTCCCAGCTT Pig pITGαD-R ACGGGGTTATGGACCTCATAC pITGαE-F CGCAGAGCTCTCCTTAAGTCA Pig pITGαE-R CCCAACACTGCTTTGAATGAT pITGαL-F GTCAGCCAGACAATGGACAAT Pig pITGαL-R TGGTGGAAGAGGTAACACAGG pITGαM-F AGAAGGAGACACCCAGAGCA Pig pITGαM-R GTAGGACAATGGGCGTCACT pITGαV-F GCAGAAAGGAGCAATTCGAC Pig pITGαV-R GGGTTGCAAGCCTGTTACAT (Continued on next page)$$
## Immunofluorescence
Cells were fixed with 4% paraformaldehyde for 30 min, permeabilized with 0.1% Triton X-100 at room temperature for 7 min, blocked with 3% bovine serum albumin (BSA) for 1 h, and incubated with primary antibodies at room temperature for 2 h. After being washed with PBS, the cells were incubated with fluorescent secondary antibodies for 1 h, rinsed with PBS, and then stained with 4′,6-diamidino-2′-phenylindole (DAPI) for 6 min. An antifluorescence quencher was added, and immunofluorescence was observed with a Zeiss fluorescence microscope.
## Plasmid construction
The coding sequences of the target genes were obtained either by downloading sequences from the NCBI database followed by commercial gene synthesis, or by PCR amplification. These sequences were subsequently cloned into the appropriate expression vectors: the lenti-cmv-mcs-Flag vector was used to generate Flag-ITGα2, Flag-ITGα5, Flag-EX, and Flag-ITGα10; the pEGFP-C vector was used for constructing GFP-TM, GFP-IN, GFP-Nsp1α-Nsp12, and GFP-GP2-GP7; the pcaggs-Flag vector was employed to create Flag-HOXD8, Flag-Nsp1α, and Flag-Nsp10; and the pcsggs-HA vector was used to generate HA-HOXD8. All final plasmid constructs were verified by commer cial Sanger sequencing.
## Sample preparation and RNA-seq analysis
PAMs were seeded in 60 mm culture dishes at a density of 1 × 10⁶ cells per dish, and the cells were infected with PRRSV GD-HD (MOI = 1), with uninfected cells serving as controls. Samples were collected with TRIzol reagent at different time points of infection and then sent to the company to extract total RNA for RNA-seq analysis. Transcriptome sequencing was performed by Tsingke Biotechnology Co., Ltd. (Beijing, China), and bioinformatic analysis was conducted using the OmicStudio tools.
## G/F-actin detection
The cells were collected and added to 100 µL LAS2 buffer and incubated at 37°C for 10 min. The samples were subsequently centrifuged at 350 × g at room temperature for 5 min, after which the supernatant was collected and centrifuged at 100,000 × g, 37°C for 1 h. The G-actin remained in the supernatant, while the pellet contained F-actin. The supernatant was transferred to a fresh tube, mixed with SDS sample buffer, and denatured in boiling water for 10 min. The pellet was resuspended in 100 µL of F-actin depolymerization buffer and incubated on ice for 1 h. SDS sample buffer was added, followed by denaturation in boiling water for 10 min. Then, the expression of G/F-actin was analyzed via Western blotting.
Lentiviral shRNA-mediated knockdown mouse models shRNA was constructed in the pLKO.1 lentiviral vector, and the knockdown sequences for FLNA and ITGα5 were inserted into the vector. The shRNA plasmid was co-transfected into HEK-293T cells with the helper plasmids L-M and L-S, and the supernatant was collected 48 h later for infection of cells or mice. A total of 300 µL supernatant was used to infect the mice by intramuscular injection into the back, and the mice were infected again 48 h later. FLNA or ITGα5 knockdown mouse models were generated 14 days after infection (43,44).
## Bacterial infection
Fifteen 4-week-old BALB/c mice were divided into control, shFLNA, and shITGα5 groups, with five mice in each group. Mice were infected with pLKO.1 empty vector lentivirus or shRNA lentivirus, and 2 weeks later, the mice were injected intraperitoneally with K. pneumoniae (1 × 10 6 CFU) (45). The infection experiments involving A. viridans and S. suis 2 were conducted using the same grouping and experimental methods.
## Piglet infection experiment
Twentyfive 4-week-old piglets negative for PRRSV antigen and antibodies were purchased from a pig farm. None of the pigs were vaccinated against PRRSV. The piglets were randomly divided into five groups: PRRSV + K. pneumoniae, K. pneumoniae, PRRSV + S. suis 2, S. suis 2, and PRRSV, with five pigs in each group, housed individually. When piglets are infected with PRRSV GD-HD, their body temperature increases rapidly and is maintained for a period of time after PRRSV infection. When the body temperature returned to normal, the piglets were infected with 1 mL (5 × 10 8 CFU) of K. pneumoniae or S. suis 2 (46). The piglets were sacrificed 24 h after bacterial infection, and necropsy and gross pathological examination of the lungs were immediately performed with photographic documentation. Lung tissues were then collected for lung lavages to isolate PAMs for western blotting (WB) analysis, and the lungs were taken for bacterial quantification by CFU.
## Statistical analysis
The experiments were repeated three times and statistically analyzed via GraphPad Prism 8 software. The results are expressed as the means ± standard deviations. The signifi cance of differences was analyzed via Student's t-test or one-way analysis of variance, and P values of <0.05 (*), <0.01 (**), and <0.001 (***) were considered statistically significant at different levels. "ns" indicates no significant difference.
## References
1. Li, Ren, Wang et al. (2015) "Influenza viral neuraminidase primes bacterial coinfection through TGFβ-mediated expression of host cell receptors" *Proc Natl Acad Sci*
2. Herrera, Potts, Huber et al. (2023) "Influenza enhances host susceptibility to non-pulmonary invasive Streptococcus pyogenes infections" *Virulence*
3. Palani, Uddin, Mckelvey et al. (2023) "Immune predisposition drives susceptibility to pneumococcal pneumonia after mild influenza A virus infection in mice" *Front Immunol*
4. Carreno-Florez, Kocak, Hendricks et al. (2023) "Interferon signaling drives epithelial metabolic reprogramming to promote secondary bacterial infection" *PLoS Pathog*
5. Jiang, He, Gao et al. (2023) "Synergistic pathogenicity of avian orthoreovirus and Staphylococcus aureus on SPF chickens" *Poult Sci*
6. Li, Wang, Liu et al. (2019) "Porcine reproductive and respiratory syndrome virus NADC 30-like strain accelerates Streptococcus suis serotype 2 infection in vivo and in vitro" *Transbound Emerg Dis*
7. Li, Wang, Li et al. (2017) "Secondary Haemophilus parasuis infection enhances highly pathogenic porcine reproductive and respiratory syndrome virus (HP-PRRSV) infectionmediated inflammatory responses" *Vet Microbiol*
8. Metwally, Mohamed, Faaberg et al. (2010) "Pathogenicity and molecular characterization of emerging porcine reproductive and respiratory syndrome virus in Vietnam in 2007" *Transbound Emerg Dis*
9. Huang, Deng, Xu et al. (2024) "Isolation and pathogenicity comparison of two novel natural recombinant porcine reproductive and respiratory syndrome viruses with different recombi nation patterns in Southwest China" *Microbiol Spectr*
10. Luo, Zheng, He et al. (2023) "Genetic variation and recombination analysis of the GP5 (GP5a) gene of PRRSV-2 strains in China from 1996 to 2022" *Front Microbiol*
11. Yu, Wu, Zhang et al. (2012) "Concurrent highly pathogenic porcine reproductive and respiratory syndrome virus infection accelerates Haemophilus parasuis infection in conventional pigs" *Vet Microbiol*
12. Duan, Xing, Chu et al. (2024) "ACE2-dependent and -independent SARS-CoV-2 entries dictate viral replication and inflammatory response during infection" *Nat Cell Biol*
13. Martín, Vela, Gilbert et al. (2007) "Characterization of Aerococcus viridans isolates from swine clinical Specimens" *J Clin Microbiol*
14. Zhao, Li, Schwarz et al. (2021) "Detection of a NDM-5producing Klebsiella pneumoniae sequence type 340 (CG258) high-risk clone in swine" *Vet Microbiol*
15. Meng, Lin, Niu et al. (2023) "Vimentin affects inflammation and neutrophil recruitment in airway epithelium during Streptococcus suis serotype 2 infection" *Vet Res*
16. Hollwedel, Maus, Stolper et al. (2023) "Ectopic expression of C-type lectin Mincle renders mice susceptible to staphylococcal pneumonia" *J Infect Dis*
17. Jia, Luo, Li et al. (2023) "Targeting complement hyperactivation: a novel therapeutic approach for severe pneumonia induced by influenza virus/Saphylococ cus aureus coinfection"
18. Rowe, Meliopoulos, Iverson et al. (2019) "Direct interactions with influenza promote bacterial adherence during respiratory infections" *Nat Microbiol*
19. David, Norton, Tyllis et al. (2019) "Direct interaction of whole-inactivated influenza A and pneumococcal vaccines enhances influenzaspecific immunity" *Nat Microbiol*
20. Saade, Deblanc, Bougon et al. (2020) "Coinfections and their molecular consequences in the porcine respiratory tract" *Vet Res*
21. Drew (2000) "A review of evidence for immunosuppression due to porcine reproductive and respiratory syndrome virus" *Vet Res*
22. Yuan, Scholz, Wald et al. (2023) "Structural basis for subversion of host cell actin cytoskeleton during Salmonella infection" *Sci Adv*
23. Kanchanawong, Calderwood (2023) "Organization, dynamics and mechanoregulation of integrin-mediated cell-ECM adhesions" *Nat Rev Mol Cell Biol*
24. Terri, Sandoval, Bontempi et al. (2024) "HDAC1/2 control mesothelium/ovarian cancer adhesive interactions impacting on Talin-1α5β1-integrin-mediated actin cytoskeleton and extracellular matrix protein remodeling" *J Exp Clin Cancer Res*
25. Zheng, Fu, Ling et al. (2022) "Host cells actively resist porcine reproductive and respiratory syndrome virus infection via the IRF8-MicroRNA-10a-SRP14 regulatory pathway" *J Virol*
26. Bernard-Raichon, Venzon, Klein et al. "2022 Gut microbiome dysbiosis in antibiotic-treated COVID-19 patients is associated with microbial translocation and bacteremia" *Nat Commun*
27. Barman, Singh, Bonin et al. (2022) "Lethal synergy between SARS-CoV-2 and Streptococcus pneumoniae in hACE2 mice and protective efficacy of vaccination" *JCI Insight*
28. Li, Yang, Chen et al. (2023) "Memory Th17 cell-mediated protection against lethal secondary pneumococcal pneumonia following influenza infection"
29. Lane, White, Walsh et al. (2023) "Antiviral epithelialmacrophage crosstalk permits secondary bacterial infections" *mBio*
30. Earnhardt, Tipper, Mello et al. (2023) "Influenza A-induced cystic fibrosis transmembrane conductance regulator dysfunction increases susceptibility to Streptococcus pneumoniae" *JCI Insight*
31. Yang, Bai, Li et al. (2022) "Influenza A and B virus-triggered epithelial-mesenchymal transition is relevant to the binding ability of NA to latent TGF-β" *Front Microbiol*
32. Islam Abmmdk, Mda, Ahmed et al. (2021) "Transcriptome of nasopharyngeal samples from COVID-19 patients and a comparative analysis with other SARS-CoV-2 infection models reveal disparate host responses against SARS-CoV-2" *J Transl Med*
33. Altman, Berning, Saxon et al. (2023) "Myocardial injury and altered gene expression associated with SARS-CoV-2 infection or mRNA vaccination" *JACC: Basic to Translational Science*
34. Tam, Suen, Treuting et al. (2020) "PPARα exacerbates necroptosis, leading to increased mortality in postinfluenza bacterial superinfection" *Proc Natl Acad Sci*
35. Holt, Wertheim, Zadoks et al. (2015) "Genomic analysis of diversity, population structure, virulence, and antimicrobial resistance in Klebsiella pneumoniae, an urgent threat to public health" *Proc Natl Acad Sci*
36. Krippner, Winkelmeier, Knerr et al. (2020) "Postmitotic expansion of cell nuclei requires nuclear actin filament bundling by α-actinin 4" *EMBO Rep*
37. Antkowiak, Guillotin, Sanders et al. (2019) "Sizes of actin networks sharing a common environ ment are determined by the relative rates of assembly" *PLoS Biol*
38. Huang, Peng, Tian et al. (2024) "Nuclear membrane protein SUN2 promotes replication of flaviviruses through modulating cytoskeleton reorganiza tion mediated by NS1" *Nat Commun*
39. Tan, Zhang, Zhan et al. (2016) "Newcastle disease virus employs macropinocytosis and Rab5a-dependent intracellular trafficking to infect DF-1 cells" *Oncotarget*
40. Xu, Chen, Gong et al. (2024) "Development a high-sensitivity sandwich ELISA for determining antigen content of porcine circovirus type 2 vaccines" *J Virol Methods*
41. Sun, Xiao, Liu et al. (2020) "Prevalent Eurasian avian-like H1N1 swine influenza virus with 2009 pandemic viral genes facilitating human infection" *Proc Natl Acad Sci*
42. Bai, Yang, Luan et al. (2021) "Induction of cyclophilin A by influenza A virus infection facilitates group A Streptococcus coinfection" *Cell Rep*
43. Ho, Palla, Blake et al. (2017) "Prostaglandin E2 is essential for efficacious skeletal muscle stem-cell function, augmenting regeneration and strength" *Proc Natl Acad Sci*
44. Wagner, Devarakonda, Skidmore et al. (2018) "Recurrent WNT pathway alterations are frequent in relapsed small cell lung cancer" *Nat Commun*
45. Wanford, Hames, Carreno et al. (2021) "Interaction of Klebsiella pneumoniae with tissue macrophages in a mouse infection model and ex-vivo pig organ perfusions: an exploratory investigation" *The Lancet Microbe*
46. Obradovic, Corsaut, Dolbec et al. (2021) "Experimental evaluation of protection and immunogenicity of Streptococcus suis bacterin-based vaccines formulated with different commercial adjuvants in weaned piglets" *Vet Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12575076&blobtype=pdf | # Susceptibility of Measles Virus to World Health Organization-Recommended Hand Rubs, Oral Rinses, and Surface Disinfectants
Lukas Daniel, Sandoval Flores, Marylyn Addo, Eike Steinmann, Toni Meister
## Abstract
Background. Measles continues to pose a major public health threat worldwide. Reoccurring outbreaks are fueled by declining vaccination rates, increased global travel, migration, and disruptions in immunization programs. Consequently, hygiene measures remain essential to prevent transmission, particularly in health care settings where nosocomial infections can occur.Methods and Results. To address this, we tested virus inactivation by common oral rinses, as primary transmission occurs via aerosols and droplets, using a quantitative suspension assay in accordance with European guidelines. We further evaluated World Health Organization-recommended hand rub formulations based on ethanol and 2-propanol with the same assay. After confirming that measles virus remains viable for several hours on stainless steel, we assessed the efficacy of surface disinfectants, including alcohol-, aldehyde-, and hydrogen peroxide-based products.Conclusions. The virus was effectively inactivated by World Health Organization-recommended hand rub formulations, oral rinses, and surface disinfectants, demonstrating the effectiveness of standard hygiene measures in infection control. These findings underscore the critical importance of consistent hygiene practices in limiting measles transmission.
Measles continues to be a major global health issue, with current outbreaks and increasing numbers of cases reported in multiple regions. It is a highly contagious viral disease caused by measles virus (MeV), determined as one of the leading causes of vaccinepreventable mortality among young children worldwide. In 2023, approximately 10.3 million measles cases and >107 500 deaths were recorded, with the greatest impact seen in lowincome countries lacking adequate health care infrastructure [1]. Infants between 6 and 11 months of age are especially vulnerable, facing a heightened risk of severe complications and death [2]. In recent years, measles cases have resurged, driven by declining vaccination rates, increased global travel, migration, and disruptions in immunization programs-factors exacerbated by the COVID-19 pandemic and international conflicts [1,3]. While many cases are mild and self-limiting, serious complications such as pneumonia and debilitating encephalitis can occur, particularly in young children, pregnant individuals, and patients who are immunocompromised [2]. Due to its high transmissibility through respiratory droplets and aerosols, measles presents a considerable threat in health care environments, where unprotected staff can contribute to nosocomial transmission [1,3]. MeV RNA has indeed been detected in environmental samples, including air specimens and surface swabs collected from the hospital room of an infected patient [4]. Consequently, several reports have identified health care workers as index cases in hospital outbreaks, particularly when unvaccinated, thereby contributing to the spread of the virus within these settings [5,6].
Therefore, effective infection control measures are crucial to limiting the spread of MeV [2,3,7]. But data on the environmental stability of MeV and its susceptibility to disinfectants remain limited [2]. Thus, this study investigates disinfection profiles and the virus's stability on stainless steel, chosen for its relevance in hospital environments and because it serves as the reference surface in European guidelines. The objective is to inform risk assessments aimed at minimizing transmission of MeV in hospital settings.
(v/v) nonessential amino acids, 100-IU/mL penicillin, 100-mg/ mL streptomycin, and 2-mmol/L L-glutamine. For MeV production (ATCC VR-24), Vero E6 cells were seeded at 3 × 10 6 cells/flask. After 24 hours, the cells were inoculated with MeV (multiplicity of infection, 0.03) and incubated for 72 hours at 37 °C with 5% CO 2 . Upon visual cytopathic effect, the cells were scraped in Opti-MEM (Gibco) and particles released by 1 freeze-thaw cycle. The virus suspension was cleared from cell debris by centrifugation (1000 × g for 5 minutes) and stored at -80 °C until further usage. Infectious viral titers were determined by an endpoint dilution assay.
## Quantitative Suspension Test
We set out to determine the virucidal activity of 7 oral rinses (Table 1) and the inactivation capacity of World Health Organization (WHO)-recommended hand rub formulations I and II (Table 2), as well as their active ingredients ethanol and 2-propanol. Assessment was based on European guideline EN14476 as described previously [8]. In brief, 8 parts of disinfectant per oral rinse or cell culture medium for the untreated control were mixed with 1 part of interfering substance (bovine serum albumin; final concentration, 0.3 g/L; clean condition) and 1 part of MeV and incubated for 30 seconds at room temperature. WHO formulations I and II, as well as ethanol and 2-propanol, were tested for final concentrations of 20%, 30%, 40%, 60%, and 80%. Oral rinses were tested for a final concentration of 80%. An endpoint dilution assay was performed on Vero E6 cells to determine the remaining infectious viral titers. After 7 days, cytopathic effects were evaluated microscopically and used to calculate the 50% tissue culture infectious dose per milliliter (TCID 50 /mL).
## MeV Stability Testing
Stainless-steel discs (2-cm diameter discs, article 4174-3000; GK Formblech GmbH) were decontaminated in 70% (v/v) ethanol for 15 minutes. The discs were subsequently contaminated with 50 µL of virus solution containing 9 parts of MeV and 1 part of interfering substance (bovine serum albumin; final concentration, 0.3 g/L; clean condition). All specimens were stored at room temperature. Virus was recovered at time points indicated in the figure legend postcontamination by transferring the specimens into a 25-mL container harboring 2 mL of cell culture medium (without FCS) and subsequent vortexing.
For each time point, 3 specimens were collected. An endpoint dilution assay was performed on Vero E6 cells as described previously. Humidity (39.7% ± 3.4%, mean ± SD) and temperature (26.2°C ± 1.3°C) were simultaneously measured over the course of the experiment.
## MeV Inactivation by Surface Disinfectants
Stainless-steel discs were decontaminated and spiked with virus solution as described previously. The steel discs were incubated until the virus solution was desiccated completely. Subsequently, 100 µL of surface disinfectant (Table 3) at indicated concentrations was applied onto the carrier and incubated according to the manufacturer's instructions. Cell culture medium was used for the untreated control. Thereafter, the specimens were transferred into a 25-mL container harboring 2 mL of cell culture medium (without FCS), and virus was recovered by subsequent vortexing. An endpoint dilution assay was performed on Vero E6 cells as described earlier.
## Statistical Analysis
Inactivation kinetics of MeV by WHO formulations I and II was compared with other respiratory viruses by using a robust Hill nonlinear dose-response fit. Thereby, comparing respiratory viruses included influenza virus H1N1, bovine coronavirus, severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and SARS-CoV-2, as well as the reference virus modified vaccinia Ankara.
## RESULTS
## Inactivation of MeV by Commercially Available Oral Rinses
MeV is predominantly transmitted by respiratory droplets; thus, oral rinses may possess the ability to reduce the risk of transmission by temporarily lowering the viral load in the oral cavity. Therefore, we tested the virucidal efficacy of commercially available oral rinses for adults and children (aged >6 years) according to EN14476 (Figure 1A, Table 1).
All tested oral rinses significantly reduced viral loads; however, residual infectious virus was still detectable following exposure to Chlorhexamed FORTE (reduction factor, 1.58 log 10 TCID 50 /mL), Meridol Gum Protection (1.79 log 10 TCID 50 / mL), and Odol-med3 Junior (1.12 log 10 TCID 50 /mL). In contrast, all 3 Listerine products and Dontodent Junior decreased viral titers to the lower limit of quantification (LLOQ), suggesting efficient inactivation of MeV with applicable oral rinses.
## Inactivation of MeV by WHO-Recommended Hand Rub Formulations and Their Active Ingredients Ethanol and 2-Propanol
By employing a quantitative suspension test according to EN14476, the virucidal activity of different concentrations (20%, 30%, 40%, 60%, 80%) of ethanol and 2-propanol was evaluated. Therefore, 8 parts of alcohol were mixed with 1 part of interfering substance to mimic protein contamination and 1 part of virus. Both alcohols at a 20% concentration were effective in reducing infectious viral loads to the LLOQ (Figure 1B and 1C). WHO-recommended hand rub formulations I and II were subsequently tested in a similar approach (Table 2). In both cases, 30% was sufficient to completely inactivate MeV. Formulation I at 20% only slightly reduced MeV titers (0.71 log 10 TCID 50 /mL). In contrast, formulation II at 20% reduced infectious titers by >3.5 log 10 TCID 50 /mL, although some infectious virus remained detectable (Figure 1D and1E). A comparison of the inactivation kinetics of WHO-recommended hand rub formulations I and II with those of other respiratory viruses-including influenza A virus H1N1, bovine coronavirus, SARS-CoV, SARS-CoV-2, and MERS-CoV-revealed that MeV exhibited the greatest susceptibility to inactivation by alcohol-based hand rubs (Figure 1F and1G).
## Environmental Stability of MeV
The duration that viruses remain infectious on inanimate surfaces varies by their intrinsic properties and surrounding environmental conditions. To evaluate the environmental stability of MeV, stainless-steel discs were contaminated with infectious MeV, and remaining infectious virus was retrieved from the specimens at different time points. The evaluation of viral stability revealed the presence of infectious virus for 3 days (182.67 TCID 50 /mL). However, MeV titers dropped significantly within 1 day (1377.33 TCID 50 /mL), and almost no infectious virus was detectable after 2 days (219 TCID 50 /mL), resulting in a half-life of 8.51 ± 3.16 hours (Figure 2A). During the experiment, temperature and humidity were kept relatively stable at 26.2 ± 1.3 °C and 39.7% ± 3.4% (Figure 2B). These data suggest that even though the half-life of MeV is comparably short, transmission via inanimate surfaces can pose a risk of infection.
## Inactivation of MeV by Surface Disinfectants
To reduce the risk of fomite-mediated transmission, the application of surface disinfectants serves as an effective preventive measure. In this study, we evaluated 5 disinfectants for their ability to inactivate MeV. These included alcohol-based formulations (Antifect N liquid and Bacillol AF), aldehyde-based disinfectants (Kohrsolin FF and Incidin Rapid), and a hydrogen peroxide-based product (Incidin OxyFoam; Figure 2C-E, Table 3). The experiment was performed according to European guideline EN16777. We found that all disinfectants but Incidin OxyFoam reduced infectious viral titers to the LLOQ (158 TCID 50 /mL) within the exposure time and concentration implicated by the manufacturers. The hydrogen peroxide-based disinfectant reduced viral titers but did not achieve complete inactivation of MeV. In conclusion, depending on the active ingredients MeV can efficiently be inactivated by surface disinfectants.
## DISCUSSION
In this study, we conducted a comprehensive investigation into the environmental stability of MeV and its vulnerability to various chemical disinfectants. Our objective was to generate data that provide evidence-based guidance to reduce the risk of nosocomial transmission.
Reducing viral load in the oral cavity could lower the risk of MeV transmission via respiratory droplets. In line with this concept, our evaluation of commercially available oral rinses for adults and children (aged >6 years) demonstrated effective inactivation of MeV according to EN14476 (Figure 1A). Similarly, other enveloped viruses, such as SARS-CoV-2 and respiratory syncytial virus, have been efficiently inactivated by a range of oral rinses [8,9]. However, this efficacy was directly affected by the type of active ingredient used and their concentration [10]. For instance, ingredients such as chlorhexidine, essential oils, povidone-iodine, benzalkonium chloride, cetylpyridinium chloride, octenidine dihydrochloride, and various surfactants have been shown to inactivate SARS-CoV-2 in a dose-dependent manner, whereas hydrogen peroxide and dequalinium chloride failed to have the same effect [10].
We further demonstrated that ethanol and 2-propanol, even at concentrations as low as 20%, were capable of reducing infectious MeV titers to the LLOQ (Figure 1B and1C). These alcohols are the primary active components of WHO-recommended hand rub formulations I and II, both of which effectively inactivated MeV at a 30% concentration (Figure 1D and1E). Recent studies, including our own, have shown that these formulations and their constituent alcohols also effectively inactivate a range of enveloped viruses, including SARS-CoV-2, MERS-CoV, SARS-CoV, mpox virus, and others [11,12]. Notably, among the respiratory viruses tested, MeV exhibited the highest sensitivity to inactivation by WHO formulations I and II (Figure 1F and1G). Overall, WHO formulation II (based on 2-propanol) inactivated enveloped viruses more efficiently than formulation I (based on ethanol), likely because 2-propanol exhibits lower interfacial tension and its molecular structure may allow for more effective disruption of lipid bilayers [13].
Depending on viral characteristics and environmental factors, viruses remain infectious on inanimate surfaces for a certain period. For example, SARS-CoV-2 was found to remain infectious on stainless-steel discs for 5 days, whereas even longer stability was observed for hepatitis E virus, hepatitis C virus, hepatitis A virus, and mpox virus. In comparison with these viruses, MeV appeared to be less stable on stainless steel at 3 days (Figure 2A). Nonetheless, MeV RNA was identified in air specimens, on environmental surfaces, and on respirators within the hospital setting of a patient who was infected [4], indicating that MeV-contaminated fomites may indeed be present in health care environments. This underlines the potential risk of environmental persistence and highlights the importance of effective surface disinfection as a preventive measure.
To mitigate this risk, the use of surface disinfectants can be an effective intervention. In this study, 5 surface disinfectants based on alcohol (Antifect N liquid and Bacillol AF), aldehyde (Kohrsolin FF and Incidin Rapid), and hydrogen peroxide (Incidin OxyFoam) were evaluated regarding their potential to inactivate MeV (Figure 2B-D). All disinfectants tested, with the exception of Incidin OxyFoam, reduced infectious MeV titers to the LLOQ within the manufacturer-recommended exposure time and concentration. This observation aligns with previous findings showing that hydrogen peroxide-based disinfectants also failed to fully inactivate viruses such as yellow fever virus and mpox virus [14,15], suggesting that concentration and exposure time affect inactivation, particularly if the product is formulated as a foam or gel, which could limit direct contact with the viral particles [16].
## CONCLUSION
Differences in viral inactivation efficacy underscore the need for tailored hygiene measures, as they cannot be universally applied. While in vitro results provide valuable insight, it is important to recognize that these experiments are conducted under controlled conditions optimized for virus survival. Real-world factors such as fluctuating temperatures, humidity, and the presence of other microorganisms can significantly influence viral persistence, meaning that laboratory results may not fully reflect practical scenarios. Nonetheless, the demonstrated effectiveness of alcohols, WHO formulations, and surface disinfectants against MeV in laboratory conditions supports their use in health care environments and outbreak responses to help prevent transmission, while highlighting the need to consider environmental complexities when interpreting these findings.
## Notes
## References
1. Minta, Ferrari, Lambert (2024) "Progress toward measles eliminationworldwide, 2000-2023" *MMWR Morb Mortal Wkly Rep*
2. (2022) *World Health Organization. Measles outbreak guide*
3. Crowcroft, Minta, Bolotin (2024) "The problem with delaying measles elimination" *Vaccines (Basel)*
4. Bischoff, Mcnall, Blevins (2016) "Detection of measles virus RNA in air and surface specimens in a hospital setting" *J Infect Dis*
5. Baccolini, Sindoni, Adamo (2020) "Measles among healthcare workers in Italy: is it time to act?" *Hum Vaccin Immunother*
6. Han, Park, Yi (2021) "Measles seroprevalence among healthcare workers in South Korea during the post-elimination period" *Hum Vaccin Immunother*
7. Shattock, Johnson, Sim (2024) "Contribution of vaccination to improved survival and health: modelling 50 years of the expanded programme on immunization" *Lancet*
8. Meister, Friesland, Frericks (2023) "Virucidal activity of oral, hand, and surface disinfectants against respiratory syncytial virus" *J Hosp Infect*
9. Meister, Brüggemann, Todt (2020) "Virucidal efficacy of different oral rinses against severe acute respiratory syndrome coronavirus 2" *J Infect Dis*
10. Meister, Gottsauner, Schmidt (2022) "Mouthrinses against SARS-CoV-2high antiviral effectivity by membrane disruption in vitro translates to mild effects in a randomized placebo-controlled clinical trial" *Virus Res*
11. Kratzel, Todt, 'kovski (2020) "Inactivation of severe acute respiratory syndrome coronavirus 2 by WHO-recommended hand rub formulations and alcohols" *Emerg Infect Dis*
12. Suchomel, Kundi, Pittet et al. (2013) "Modified World Health Organization hand rub formulations comply with European efficacy requirements for preoperative surgical hand preparations" *Infect Control Hosp Epidemiol*
13. Ballal, Chapman (2013) "Hydrophobic and hydrophilic interactions in aqueous mixtures of alcohols at a hydrophobic surface" *J Chem Phys*
14. Meister, Frericks, Kleinert (2024) "Inactivation of yellow fever virus by WHO-recommended hand rub formulations and surface disinfectants" *PLoS Negl Trop Dis*
15. Meister, Brüggemann, Todt (2023) "Stability and inactivation of monkeypox virus on inanimate surfaces" *J Infect Dis*
16. Ballestê Ajorio, Rhodes, Rodrigues (2021) "Evaluation of hydrogen peroxide virucidal efficacy against yellow fever virus 17DD vaccine strain for application in a vaccine manufacturing industry" *J Pharm Biomed Anal* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12317682&blobtype=pdf | # Performance of Long-Read Single-Molecule Real-Time Sequencing for SARS-CoV-2 Genotyping in Clinical Samples
Pauline Trémeaux, Justine Latour, Camille Vellas, Sofia Demmou, | Noémie Ranger, Antonin Bal, Jacques Izopet
## Abstract
Due to the continuous genetic evolution of SARS-CoV-2, numerous variants have emerged and different whole genome sequencing techniques, necessary for accurate virus typing, have been developed. In this study, we evaluated the performance of PacBio single-molecule real-time (SMRT) sequencing for SARS-CoV-2 typing. Reproducibility was assessed on two internal quality controls, whose median reading depths were 1154X and 1059X. The overall sensitivity on 1646 clinical samples collected between January 2023 and June 2024 was 83.6% and was correlated to the viral load. By comparison, the overall sensitivity of short-read illumina sequencing over the same period of time on 271 samples was 90.8%. Although less sensitive, SMRT sequencing was more efficient for the identification of the two lineages in a co-infection case due to the amplification of long fragments. Comparing the results obtained by the two techniques, 10 out of 50 samples were identified with the same clade but not the exact same lineage at the time of analysis, because of the very frequent updates of the Pango taxonomy. Nevertheless, we obtained very similar fasta consensus sequences with a maximum difference of 4 nucleotides, showing that both methods provide accurate typing of SARS-CoV-2, useful for epidemiological or clinical studies.
| IntroductionThe Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), responsible for the COVID-19 pandemic, has a singlestrand positive RNA genome of 29.9 kb which has evolved genetically since its appearance in late 2019, resulting in a succession of variants [1][2][3]. This is partly the result of mistakes of the RNA-dependent RNA-polymerase, although limited by the proofreading activity of the exoribonuclease [4,5]. Another genetic evolution mechanism of SARS-CoV-2 is recombination, in case of co-infection with two different viral strains [6][7][8]. SARS-CoV-2 strains are defined by Nextstrain clades and Pango lineages, two co-existing nomenclature systems [9,10]. Recombinant viruses are designated with the lineages "X.." in the Pango taxonomy. The World Health Organization (WHO) has also defined Greek letters to designate several variants of concern that impact global public health. However, this nomenclature is less and less used since all circulating clades and lineages since mid-2021 belong to the Omicron variant [3,11].
Although most of SARS-CoV-2 genetic changes have no or little impact on the properties of the virus, some mutations can be associated with an increased transmissibility, escape from monoclonal neutralizing antibodies used in therapeutics or escape from the host immune response [5,[12][13][14]. Consequently, such as for influenza viruses, the design of vaccines against SARS-CoV-2 must adapt to the most recent viral strains in circulation [15]. The emergence and circulation of new strains are therefore closely monitored thanks to genomic surveillance politics in several countries including France, with global collaboration at the WHO level [16].
This monitoring requires efficient and easy-to-use laboratory protocols, using mostly next-generation sequencing (NGS) platforms that enable whole genome sequencing, a prerequisite for accurate lineage identification. Several assays or protocols have been developed, either by industries or independent laboratories, using mostly Illumina or Oxford Nanopore Technologies (ONT) sequencing platforms. Previous studies have shown that Illumina sequencing provides a higher raw-read accuracy and a better coverage while ONT sequencing offers shorter hands-on time and runtime and allows long-read sequencing that may be advantageous for recombination detection [17][18][19]. We recently developed a whole genome amplification and sequencing technique using the Sequel IIe platform of Pacific Biosciences (PacBio) [6,20]. Comparison of single-molecule real-time (SMRT) sequencing with other NGS approaches in the field of medical virology are limited, and even more scarce for SARS-CoV-2 typing. Only one study recently compared capture amplification followed by SMRT sequencing with amplicon-based approaches followed by Illumina or ONT sequencing on 92 samples [21]; no discrepancy was observed in lineage assignment.
This study aims to assess the performance of a long-fragment amplification and SMRT sequencing technique on a large panel of clinical samples. Results obtained with long-read SMRT sequencing and short-read sequencing on an Illumina platform were also compared.
## 2 | Materials and Methods
## 2.1 | Patients and Samples
From January, 2023 to June, 2024, SARS-CoV-2 complete genome sequencing was performed on respiratory samples taken from patients of the Toulouse University Hospital to characterize clusters, investigate severe cases of hospitalized and/or immunocompromised patients, and monitor local epidemiology. Samples found positive with a cycle threshold (Ct) value under 28 were included. Once a week and depending on the prevalence of the virus, 20%-100% of the positive samples of the day with a Ct value lower than 28 were also transmitted to the National Reference Centre for Respiratory Infection Viruses for sequencing, according to the French surveillance policy of circulating SARS-CoV-2 strains.
## 2.2 | SARS-CoV-2 Detection
SARS-CoV-2 was detected using the SARS-CoV-2/Flu A/B/RSV Assay on the Panther Fusion System (Hologic) following the manufacturer's instructions [22]. In brief, 500 µL of viral transport medium were transferred to a lysis tube containing 710 µL of buffer. The instrument used 360 µL of this mixture for nucleic acids extraction. This assay targets two sequences located on the ORF1ab gene of SARS-CoV-2.
## 2.3 | SARS-CoV-2 Quantification Using ddPCR
Tenth-fold serial dilutions of two SARS-CoV-2 strains were carried out to compare the Ct values obtained on the Panther Fusion instrument in routine to an absolute viral quantification using digital droplet PCR (ddPCR). For the latter, nucleic acid extraction was performed using the QIAamp Viral RNA Mini kit (QIAGEN) from 140 µL of sample. The ddPCR was performed using the One step RT ddPCR kit for Probes (BIORAD) on QX200 Droplet instruments (BIORAD).
## 2.4 | Long-Read SMRT Sequencing Using the Sequel IIe Platform
SARS-CoV-2 full-length genomic sequences were obtained using the PacBio Sequel IIe instrument and an amplicon-based approach, as previously described [6,20]. Briefly, nucleic acids were extracted from 180 µL of viral transport medium using the MGIEasy Nucleic Acid Extraction kit on a MGISP-960 system (Beijing Genome Institute), following the manufacturer's instructions. Viral RNA was reverse-transcribed using the Superscript IV VILO enzyme (Life Technologies) and random hexamers. Two separate multiplex PCRs of 15 and 14 amplicons were performed afterwards using the Q5 Hot Start High Fidelity DNA polymerase (New England Biolabs) and M13-tailed barcoded primers. The primers for these 29 overlapping amplicons of approximately 1.2 kb were based on the Midnight design [23] and were adapted to cope with SARS-CoV-2 genetic evolution. A second PCR using barcoded-M13-primers was then performed using the Kapa HiFi HotStart Ready Mix (Roche Diagnostics). For each sample, the two PCR products were mixed volume by volume and quantified to make a normalized pool before purification. Up to 250 samples of SARS-CoV-2 or other various viruses from clinical samples (Human Immunodeficiency Virus, Hepatitis E virus, Human Papillomavirus, Cytomegalovirus, Respiratory Syncytial Virus) were pooled for the library preparation, using the SMRTbell Express Template Prep 3.0 kit, and the sequencing on the Sequel IIe instrument.
The whole genome sequences were built from the PacBio reads using a custom-made Snakemake pipeline. The HiFi reads were directly generated by the Sequel IIe sequencer (v.6.3.0, https:// github.com/PacificBiosciences/ccs), then demultiplexed and filtered (minimum of 3 passes, Q20) with Lima (v.2.6.0, https:// github.com/PacificBiosciences/barcoding). The resulting reads were mapped to the SARS-CoV-2 reference genome (Wuhan-Hu-1 isolate, GenBank accession number NC_045512.2) with Minimap2 (v2.17 [24]), analyzed by pbAA (v0.1.3, https://github.com/ pacificbiosciences/pbAA) and CoSA (Coronavirus Sequence Analysis, v9.0.0, https://github.com/Magdoll/CoSA) to construct consensus sequences with a minimum depth of 10X. Potential coinfections were identified by a home-made script which constructs a hypothetical VCF file and thus a consensus sequence, where majority and minority variants are reversed (between minimum and maximum frequency thresholds). Lineages and clades were attributed according to the most recent versions of Pangolin (v4.1.3 to v.4.3.1 [10]) and Next-Clade (v2.9.1 to v3.0.0 [9]) at the time of analysis.
## 2.5 | Short-Read Sequencing Using the Illumina Platform
Nucleic acids were extracted using the MGISP-960 system, as described above. SARS-CoV-2 full-length genomic sequences were amplified using the amplicon-based COVIDSeq-Test (Illumina) and the ARTIC V4 primers. Samples were sequenced with 100 base pair (bp) paired-end reads using a NovaSeq 6000 Sequencing system instrument (Illumina), as previously described [25]. After quality trimming, reads longer than 30 bp were aligned to the SARS-CoV-2 genome MN908947 using Minimap2. Duplicate reads were removed, realigned using abra2 and read-ends were clipped using samtools. Variants present at a minimum frequency of 5% were called using freebayes and bcftools (complete pipeline available at https:// github.com/genepii/seqmet [25]).
## 2.6 | Statistics Analyses
For concordance analyses between both sequencing techniques, consensus fasta of Illumina sequences were retrieved on GI-SAID and reanalyzed, as were SMRT sequences, with a common version of Pangolin (v4.3.1) and Nextclade (v3.9.1).
Cohen's kappa coefficient was used to measure the pairwise concordance between the two sequences obtained for each sample [26]. For all sequences, the Nextclade report listed the substitutions, deletions, insertions and missing positions, necessary to calculate differences between the pair of sequences. Nucleotide positions covered only by one of the two sequences were discarded from the comparison.
A maximum-likelihood tree with ultrafast bootstrap (1000 replicates and a GTR+F+R3 evolution model) was constructed from the consensus sequences of both technologies by IQTREE (v2.0.3) [27], previously aligned with MAFFT (v7.505) [28]. Sequences were trimmed at the extremities to homogenize length between the two protocols.
## 3 | Results
## 3.1 | Analytical Performance of SMRT Sequencing
The repeatability of the technique was assessed by sequencing three times in the same run an internal quality control (IQC), in-house made from a cell culture supernatant diluted in Minimum Essential Medium (MEM). We obtained identical results for strain identification (21I_B.1.617.2), coverage (95.8% of the genome), mutations profile on the S gene (E156del, F157del, T19R, G142D, R158G, A222V, L452R, T478K, D614G, P681R, D950N) and comparable results for mean reading depth (1258-1760X). Reproducibility was assessed by sequencing two IQC in each run, made from strains 20B_B.1.1.254 and 21I_B.1.617.2 (Delta variant). Correct identification of both the clades and lineages was obtained, with a median coverage of 95.8% over 2023. As for repeatability, this was explained by the absence of amplicon 5, due to a mismatch in a primer whose sequence was modified in June, 2022. This sequence adaptation to the new Omicron variants 22A_BA.1 and 22B_BA.5 led to a poorer match with the sequences of older strains used as IQC but improved the results of patient's samples. Over 67 runs, the mean reading depths were 1059X (interquartile range (IQR): 434-1615) and 1154X (IQR: 758-1635) for clade 20B and clade 21I IQCs, respectively.
During the 18-months study period, a total of 1646 respiratory samples were sequenced on the Sequel IIe instrument: 1542 nasopharyngeal swabs, 38 expectorations, 26 tracheal aspirations, 33 bronchoalveolar lavage fluids and 7 saliva. The median Ct value was 18.9 (IQR): 16.1-22.3). In median, 24 039 reads per sample were analyzed (IQR: 10 607-44 812) (Figure 1A) for a mean reading depth of 700X (IQR: 263-1363) (Figure 1B). The median coverage was 95.8% of the complete viral genome (IQR: 89.3%-99.1%) (Figure 1C). Sequencing results with the SMRT technology were considered valid if the coverage was greater than 70% of SARS-CoV-2 genome including the key region of the receptor-binding domain (RBD), or greater than 85% of the complete genome in case of an uncovered RBD region. Mutations in the RBD region can affect virus infectivity and/or be responsible for immune escape from antibodies. They are therefore often linked to the emergence and the designation of new SARS-CoV-2 variants [29,30]. This explains the different coverage threshold for sequence validation. To detect and avoid potential cross-contamination between samples, results were considered invalid if more than 10 nucleotide positions had a double base, unless these were confirmed during a second sequencing (Figure 1D). Such a high nucleotide diversity could be observed in cases of co-infections (0.5%) or in immunocompromised patients (2.8%), some of whom developed chronic infections.
## 3.2 | Sensitivity of SARS-CoV-2 Genotyping on Clinical Samples
The overall sensitivity of SMRT sequencing for clinical samples was 83.6% but varied depending on the viral load of the samples, and was therefore inversely correlated with the Ct value. The clade and lineage of SARS-CoV-2 were determined for 98% of samples with a Ct value under 20, 90% of samples with a Ct value between 20 and 22, 70% of samples with a Ct value between 22 and 25, and 24% of samples with a Ct value over 25 (Figure 2). Clades 23A, 23F, and 23I were the most frequent, in agreement with the prevalence of SARS-CoV-2 strains circulating in France at the time of the study (Supplementary document 1). To further precise the sensitivity of long-read sequencing, we determined the absolute quantification of SARS-CoV-2 corresponding to these Ct values using ddPCR on two recent viral strains, 23F_EG.5.1 and 24C_KP.3. A Ct value of 20 corresponds to a viral load of 6.5-6.7 log cp/mL, a Ct value
## 3.3 | Comparison Between Long-Read and Short-Read SARS-CoV-2 Sequencing
Over the study period, 271 samples were sent to the National Reference Center as part of national weekly epidemiological surveillance and sequenced with the short-read Illumina sequencing platform. The median Ct value of these samples was 19.3 (IQR: 16.5-23.3), similar to those sequenced with the SMRT platform (p = 0.081). The clade and lineage of SARS-CoV-2 were determined for 99% of samples with a Ct value under 20, 93% of samples with a Ct value between 20 and 25, and 61% of samples with a Ct value over 25 (Figure 2). The overall sensitivity was 90.8%. Among these 271 samples, 50 were sequenced in both laboratories (median Ct value of 19.1 [IQR: 16.7-21.9]). Three samples could not be amplified and sequenced by either of the two laboratories (Ct values superior to 26), while 3 samples had a result by only one of the two protocols (2 samples genotyped only by using Illumina short-read sequencing and 1 sample only by using SMRT long-read sequencing) (Table 1). The other 44 samples had concordant sequencing results. For 33 samples, identified clades and lineages were strictly identical and for 10 samples, clades were identical with both techniques while lineages were different (Table 1). To note, 5 samples among them were identified with Nextclade taxonomy as being recombinant viruses, but only one of the two laboratories identified concordantly an "X.." Pango lineage. The reanalysis of SMRT and Illumina consensus sequences with the same version of Pango gave identical lineage results for all samples except one (Sample 2: BQ.1.1 and BQ.1.1.23, respectively). Lastly, 1 sample was identified by both laboratories as being a probable co-infection of two SARS-CoV-2 strains but only the long-read SMRT sequencing protocol identified them (Sample 7; 23B_XBB.1.16.1 and 23A_XBB.1.5.28). This coinfection was confirmed by duplicate analyses with both sequencing techniques. To note, 12 samples had a different clade and/or lineage results at the reanalysis performed in December, 2024 compared to the initial typing result, because of the updates of Nextstrain and Pango taxonomies.
When looking at the fasta consensus sequences obtained by the two techniques, 40/44 (91%) samples had strictly identical sequences and a kappa concordance coefficient of 1. The other 4 samples had a kappa concordance coefficient between 0.982 and 0.996. The SMRT and Illumina sequences of 3/44 samples differed by only 1 single-nucleotide polymorphism (SNP) while those of 1/43 sample differed by 4 SNPs. The latter case, with the most divergent consensus sequences, corresponded to the co-infection (Sample 7) (Supplementary document 3). No difference was observed in the insertions and deletions in the consensus sequences obtained with the two techniques compared to the Wuhan reference sequence. Finally, phylogenetic analysis showed strong bootstrap values for sequences issued from the same sample with the two techniques, and they were clustered together on the phylogenetic tree (Supplementary document 4). Only sequences from Samples 23 and 24 were intermingled; they were taken on the same day from two residents of a long-term care facility where there was a COVID-19 cluster.
## 4 | Discussion
Because of the intrinsic genetic variability of SARS-CoV-2, new variants have regularly emerged since the first apparition of the virus in late 2019. Monitoring of circulating viral strains is necessary for epidemiological studies and to adapt vaccine design to circulating strains. In clinical practice, virus lineage can guide the choice of antivirals or monoclonal antibodies used to prevent severe disease, and sequencing performed during follow-up can detect the emergence of possible resistance mutations [5,13,15]. It requires efficient, robust and easyto-use techniques. For SARS-CoV-2, precise typing is based on complete genome sequencing using NGS. During the COVID-19 pandemic, several protocols have been developed and published. They were based either on capture enrichment or, more frequently, on PCR amplification with overlapping amplicons to have sufficient genetic material before the sequencing step [20,31]. The challenge with the latter strategy is the genetic variability of the virus, which can lead to primers mismatches as the virus evolves and new variants emerge. Several updates to the design of the commonly used ARTIC primers have been published [32], while the Midnight panel has only been modified once after the emergence of the Omicron variants [23]. The reduced number of amplicons in the latter case limited the risk of mismatch with the viral genome and the need for primer modifications. Nevertheless, published updates are not always sufficient and localized drops in coverage can be observed [33]. A continued attention to the subject should therefore be implemented as part of the laboratories' quality approach. The use of an in-house protocol might allow a higher flexibility of design and faster reactivity to virus evolution. Our technique is based on the Midnight design of 1.2kb-long amplicons, but we made 10 primers changes since 2021.
The most widespread sequencing platforms were those using the short-read Illumina technology or the Oxford Nanopore Technology allowing both short-and long-read sequencing [18,34,35]. However, very few viral laboratories used the PacBio SMRT technology. One of its specificities is the possibility to sequence high-length reads while preserving a high sequence quality [36,37]. In this study, we showed on a large panel of samples that SMRT sequencing answered the needs of good sequence quality and high throughput for SARS-CoV-2 genotyping, confirming the data from our previous work [6,20,38]. Reproducibility was assessed by sequencing in-house quality controls on each run with mean reading depths obtained over 1000X. We have also successfully used this protocol in our laboratory routine on several hundred clinical samples. Nevertheless, the amplification of long fragments limited the sensitivity of the technique. Indeed, the sensitivity decreased under 90% for samples over 22 Ct. Compared to nasopharyngeal swabs, SMRT sequencing was less sensitive for SARS-CoV-2 sequencing from other sample types, but this might be explained by lower viral loads. This should be confirmed on a larger number of samples of diverse types. In comparison, short-read sequencing such as the Illumina technique offers a better sensitivity on low viral load samples: in our study, the drop of sensitivity was observed for samples over 25 Ct. On the other hand, long-read sequencing can be advantageous for the identification of coinfections, as well as the study of quasispecies and minority variants using haplotyping [6,39]. We have previously used this technique for instance for the follow-up of immunocompromised patients with chronic infections, some of whom acquired key Spike mutations [40]. We have also reported a case of superinfection with a second SARS-CoV-2 clade that evolved to the selection of one predominant recombinant virus [6]. Longread sequencing also enabled the detection, in coinfected individuals, of several minority recombinant strains that could not have been identified only by variant calling alone, an analysis available for both short-and long-read techniques [6]. Other teams have used this technology for diverse applications such as the study of HIV reservoir and quasispecies [41,42], mixed human Cytomegalovirus strain infections [43,44], the transmission bottleneck in Hepatitis B virus quasispecies [45] and for resistance genotyping of Hepatitis C virus [46].
In addition to sensitivity, we compared the consensus fasta sequences obtained with SMRT and Illumina techniques for the same samples. Although generated with different amplification, sequencing and bioinformatics methods, the sequences were highly similar (four nucleotide differences at most over the nearly 30 kb-long genome). This attested to the performance and accuracy of the two techniques and the relevance of the validation criteria used. Both techniques can therefore be used in routine, depending on the equipment available in the laboratories. Despite identical sequences, lineages attributed to a same sample by the two laboratories sometimes differed. Those were not real discrepancies, as parental lineages and identical clades were identified in each case. This could be explained by the very rapid evolution of PANGO lineages, that required frequent updates of bioinformatic pipelines. Sublineages often vary from a single or a few mutations from their parental strains. A new lineage designation is also expected to represent one or more events of epidemiological significance, such as successive epidemic waves or movement of the virus into a new geographic area [47,48]. However, this is not associated with clinical consequences in most cases. Although precise typing may be necessary for basic research and for epidemiological tracking of emerging new strains [49], use of Nextstrain clades might be sufficient in clinical practice.
## References
1. Wu, Zhao, Yu (2020) "A New Coronavirus Associated With Human Respiratory Disease in China" *Nature*
2. Zhou, Yang, Wang (2020) "A Pneumonia Outbreak Associated With a New Coronavirus of Probable Bat Origin" *Nature*
3. (2025) "Tracking SARS-CoV-2 Variants [Internet]"
4. Peacock, Penrice-Randal, Hiscox et al. (2021) "SARS-CoV-2 One Year on: Evidence for Ongoing Viral Adaptation" *Journal of General Virology*
5. Carabelli, Peacock, Thorne (2023) "SARS-CoV-2 Variant Biology: Immune Escape, Transmission and Fitness" *Nature Reviews Microbiology*
6. Trémeaux, Latour, Ranger (2023) "SARS-CoV-2 Co-Infections and Recombinations Identified by Long-Read Single-Molecule Real-Time Sequencing" *Microbiology Spectrum*
7. Combes, Bisseux, Bal (2022) "Evidence of Co-Infections During Delta and Omicron SARS-CoV-2 Variants Co-Circulation Through Prospective Screening and Sequencing" *Clinical Microbiology and Infection*
8. Dyrdak, Stamouli, Ganeshappa Aralaguppe (2024) "A Novel SARS-CoV-2 Recombinant Transmitted From a Patient With an Acute Co-Infection" *Lancet Microbe*
9. Aksamentov, Roemer, Hodcroft et al. (2021) "Nextclade: Clade Assignment, Mutation Calling and Quality Control for Viral Genomes" *Journal of Open Source Software*
10. Rambaut, Holmes, O'toole (2020) "A Dynamic Nomenclature Proposal for SARS-CoV-2 Lineages to Assist Genomic Epidemiology" *Nature Microbiology*
11. Roemer, Sheward, Hisner (2023) "SARS-CoV-2 Evolution in the Omicron Era" *Nature Microbiology*
12. (2024) "FLiRTing With Danger as SARS-CoV-2 Variants Evolve" *Lancet Respiratory Medicine*
13. Kumar, Parashar, Kumar (2022) "Emerging SARS-CoV-2 Variants Can Potentially Break Set Epidemiological Barriers in COVID-19" *Journal of Medical Virology*
14. Erkihun, Ayele, Asmare et al. (2024) "Technical Advisory Group on COVID-19 Vaccine Composition (TAG-CO-VAC)" *Health Science Reports*
15. Bull, Adikari, Ferguson (2020) "Analytical Validity of Nanopore Sequencing for Rapid SARS-CoV-2 Genome Analysis" *Nature Communications*
16. Tshiabuila, Giandhari, Pillay (2022) "Comparison of SARS-CoV-2 Sequencing Using the ONT GridiON and the Illumina MiSeq" *BMC Genomics*
17. Carbo, Mourik, Boers (2023) "A Comparison of Five Illumina, Ion Torrent, and Nanopore Sequencing Technology-Based Approaches for Whole Genome Sequencing of SARS-CoV-2" *European Journal of Clinical Microbiology and Infectious Diseases*
18. Nicot, Trémeaux, Latour "Whole-Genome Sequencing of SARS-CoV-2: Comparison of Target Capture and Amplicon Single Molecule Real-Time Sequencing Protocols" *Journal of Medical Virology*
19. Peng, Davis, Bentz (2025) "Short-Read and Long-Read Whole Genome Sequencing for SARS-CoV-2 Variants Identification" *Viruses*
20. Sasidharan, Selvarangan, Konrad "Multi-Center Clinical Evaluation of the Panther Fusion SARS-CoV-2/Flu A/B/RSV Assay in Nasopharyngeal Swab Specimens From Symptomatic Individuals" *Journal of Clinical Microbiology*
21. (2020) "Integrated DNA Technologies"
22. "xGen SARS-CoV-2-Midnight-1200 Amplicon Panel | IDT"
23. Li (2018) "Minimap2: Pairwise Alignment for Nucleotide Sequences" *Bioinformatics*
24. Bal, Simon, Destras (2022) "Detection and Prevalence of SARS-CoV-2 Co-Infections During the Omicron Variant Circulation in France" *Nature Communications*
25. Maschietto, Otto, Rouzé (2023) "Minimal Requirements for ISO15189 Validation and Accreditation of Three Next Generation Sequencing Procedures for SARS-CoV-2 Surveillance in Clinical Setting" *Scientific Reports*
26. Minh, Schmidt, Chernomor (2020) "IQ-Tree 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era" *Molecular Biology and Evolution*
27. Katoh, Misawa (2002) "MAFFT: A Novel Method for Rapid Multiple Sequence Alignment Based on Fast Fourier Transform" *Nucleic Acids Research*
28. Chen, Wang, Hozumi (2023) "Emerging Dominant SARS-CoV-2 Variants" *Journal of Chemical Information and Modeling*
29. Flores-Vega, Monroy-Molina, Jiménez-Hernández et al. (2022) "SARS-CoV-2: Evolution and Emergence of New Viral Variants" *Viruses*
30. Gerber, Daviaud, Delafoy (2019) "A Comparison of High-Throughput SARS-CoV-2 Sequencing Methods From Nasopharyngeal Samples" *Scientific Reports*
31. Lambisia, Mohammed, Makori (2022) "Optimization of the SARS-CoV-2 Artic Network V4 Primers and Whole Genome Sequencing Protocol" *Frontiers in Medicine*
32. Koskela Von Sydow, Lindqvist, Asghar (2023) "Comparison of SARS-CoV-2 Whole Genome Sequencing Using Tiled Amplicon Enrichment and Bait Hybridization" *Scientific Reports*
33. Brejová, Boršová, Hodorová (2021) "Nanopore Sequencing of SARS-CoV-2: Comparison of Short and Long PCR-Tiling Amplicon Protocols" *PLoS One*
34. Wenger, Peluso, Rowell (2019) "Accurate Circular Consensus Long-Read Sequencing Improves Variant Detection and Assembly of a Human Genome" *Nature Biotechnology*
35. Van Dijk, Naquin, Gorrichon (2023) "Genomics in the Long-Read Sequencing Era" *Trends in Genetics*
36. Lhomme, Latour, Jeanne (2021) "Prediction of SARS-CoV-2 Variant Lineages Using the S1-Encoding Region Sequence Obtained by Pacbio Single-Molecule Real-Time Sequencing" *Viruses*
37. Vellas, Del, Bello et al. (2022) "Impact of Casirivimab-Imdevimab on Severe Acute Respiratory Syndrome Coronavirus 2 Delta Variant Nasopharyngeal Virus Load and Spike Quasispecies" *Open Forum Infectious Diseases*
38. Vellas, Latour, Trémeaux (2023) "Molecular Evolution of Omicron Variant In Immunocompromised Individuals With Chronic SARS-CoV-2 Infection" *Journal of Infection*
39. Rose, Gonzalez-Perez, Nolan et al. (2021) "Ultradeep HIV-1 Proviral Envelope Sequencing Reveals Complex Population Structure Within and Between Brain and Splenic Tissues" *Journal of Virology*
40. Westfall, Deng, Pankow (2024) "Optimized SMRT-UMI Protocol Produces Highly Accurate Sequence Datasets From Diverse Populations-Application to HIV-1 Quasispecies" *Virus Evolution*
41. Brait, Külekçi, Goerzer (2022) "Long Range PCR-Based Deep Sequencing for Haplotype Determination in Mixed HCMV Infections" *BMC Genomics*
42. Külekci, Schwarz, Brait (2022) "Human Cytomegalovirus Strain Diversity and Dynamics Reveal the Donor Lung as a Major Contributor After Transplantation" *Virus Evolution*
43. Inoue, Akahane, Miyazaki (2025) "Long-Read Deep Sequencing Analysis of Hepatitis B Virus Quasispecies in Two Elderly Cases of Interspousal Transmission" *Journal of Infection and Chemotherapy*
44. Zhang, Liu, Li (2025) "Adaptation of Single Molecule Real Time (SMRT) Sequence Technology for Hepatitis C Virus Genome Sequencing and Identification of Resistance-Associated Substitutions" *Virology*
45. O'toole, Scher, Underwood (2021) "Assignment of Epidemiological Lineages in an Emerging Pandemic Using the Pangolin Tool" *Virus Evolution*
46. (2024) "Statement of Nomenclature Rules -Pango Network"
47. Ryder, Smith, Borthwick (2024) "Emergence of Recombinant SARS-CoV-2 Variants in California From 2020 to 2022" *Viruses* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584758&blobtype=pdf | # Blood transcriptomic profiling reveals gene expression alterations in patients with SFTS-associated encephalitis
Daiqing Wu, Aofan Wang, Junjie Shi, Ying Zhang, Yu Geng, Huifang Liu, Yuanyuan Wu, Wenwen Kong, Yijia Zhu, Yuxin Chen
## Abstract
Severe fever with thrombocytopenia syndrome (SFTS), a life-threatening tick-borne zoonosis caused by severe fever with thrombocytopenia syndrome virus (SFTSV), frequently leads to fatal encephalitis characterized by consciousness disorders and seizures. The molecular mechanisms governing SFTSV neuroinvasion and hostdriven neural injury remain largely elusive. To explore the mechanisms of SFTS-induced brain damage, we analyzed clinical laboratory parameters and conducted transcriptomic analyses of peripheral blood mononuclear cells from five SFTS patients with encephalitis and five non-encephalitis patients admitted to Nanjing Drum Tower Hospital during the same period. Our findings indicate that central nervous system manifestations in SFTSV infection are associated with altered expression of immune-related genes. Specifically, we identified six differentially expressed immune genes-MET, KIT, IL1R2, MAFF, CD69, and CEBPD-between the encephalitis and non-encephalitis groups. This study provides novel insights into the pathogenesis of SFTS-associated encephalitis, and further investigation into the host immune response post-SFTSV infection may aid in mitigating disease progression and improving clinical outcomes. IMPORTANCE Severe fever with thrombocytopenia syndrome (SFTS) is a life-threatening disease that can lead to encephalitis-a serious brain inflammation with high mortality. However, the causes of this brain damage remain largely unknown. In this study, we used advanced gene sequencing techniques to analyze blood samples from SFTS patients with and without encephalitis. Our results revealed key changes in immune-related genes, uncovering possible biological pathways involved in brain injury caused by the virus. These findings shed new light on how the immune system may contribute to neurological complications in SFTS and highlight specific genes that could serve as future targets for diagnosis or treatment. This research enhances our understanding of SFTS-related encephalitis and provides a valuable foundation for developing therapies to improve patient outcomes. KEYWORDS severe fever with thrombocytopenia syndrome, encephalitis, peripheral blood mononuclear cells (PBMC), transcriptomic analysis S evere fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease characterized by its unique etiology, epidemiological patterns, and diverse clinical manifestations. The disease is caused by a novel bunyavirus, designated as severe fever with thrombocytopenia syndrome virus (SFTSV), which was initially identified in China in 2009 and subsequently has been reported in several countries in Southeast Asia (1-6). SFTS is primarily transmitted through tick bites, with the highest incidence occurring in rural areas during warmer seasons (7, 8). Clinically, SFTS presents with a sudden onset of fever, thrombocytopenia, leukopenia, and gastrointestinal symptoms, with severe cases potentially progressing to multi-organ dysfunction and fatalities (9-
. The mortality rate of SFTS is notably high, varying between 5% and 30% depend ing on geographical locations (11). In addition to its systemic effects, SFTS is also associated with the central nervous system (CNS), manifesting as encephalitis or encephalopathy in a significant proportion of patients (12)(13)(14). Neurological complica tions are particularly severe in this disease, with studies suggesting that a high propor tion up to 44.7% of patients with SFTS-associated encephalitis (SAE) may not survive (15).
Despite increasing research on SFTS, studies specifically addressing SFTS-associated encephalitis remain limited. Previous investigations have primarily focused on the clinical features and epidemiology of SFTS, with less attention paid to the mechanisms underlying CNS involvement (9,15,16). The diagnosis of SFTS-associated encephalitis remains challenging due to the lack of definitive clinical markers and the infrequent collection of cerebrospinal fluid (CSF) samples. Moreover, the pathogenesis of SFTSV in the CNS remains poorly understood, with evidence suggesting both direct viral invasion and immune-mediated mechanisms could be involved (17). Although some studies have observed elevated cytokine levels in the CSF of patients with SFTS-associated encephali tis, a comprehensive understanding of the host immune response is still lacking (12,18). This knowledge gap underscores the need for further investigations into the molecular and immunological mechanisms involved in SFTSV-associated CNS complications (19).
Viral infections often induce widespread changes in the host cell transcriptome, leading to metabolic dysregulation and immune dysfunction, which ultimately create a microenvironment conducive to viral replication (20,21). Understanding the pathogen esis of SFTSV, particularly its impact on the CNS, is essential for developing effective therapeutic strategies and prognostic biomarkers. Peripheral blood mononuclear cells (PBMCs) are readily accessible sources of immune cells that can reflect the host immune response to viral infections (22,23). Transcriptomic analysis of PBMCs provides valuable insights into the molecular changes that occur during SFTSV infection and the related pathogenesis of SAE. This approach is generally considered representative of the host immune response and could reveal key pathways and genes that are dysregulated during SFTSV infection. Importantly, there is currently a paucity of transcriptomic studies focusing on SFTS-associated encephalitis.
In this study, we characterized the transcriptome profile of PBMCs in patients with SFTS-associated encephalitis and non-encephalitis using high-throughput sequencing to identify alterations that occur during disease progression. Our objective is to explore the association between these changes and the CNS manifestations of SFTS. In our study, several signature immune biomarkers were identified that could improve the management of SFTS, particularly in patients with neurological complications.
## MATERIALS AND METHODS
## Enrollment of SFTS patients
This retrospective study included five patients diagnosed with SAE and five SFTS patients without neurological complications as controls. The sample size was determined based on clinical feasibility and the rarity of SAE cases, rather than formal statistical power calculations. All patients were diagnosed at Nanjing Drum Tower Hospital (Affiliated Hospital of Nanjing University Medical School) between May 2021 and August 2021 based on clinical presentation, epidemiological history, and laboratory confirmation. Laboratory diagnosis of SFTS was established through reverse transcriptase real-time PCR for SFTSV RNA detection in serum samples and serological testing using enzymelinked immunosorbent assay or indirect immunofluorescence assay to detect SFTSV-spe cific IgM and IgG antibodies. The inclusion criteria for SAE were based on the presence of altered mental status, including symptoms such as headache, irritability, somnolence, and confusion, lasting for at least 24 hours, with other potential etiologies excluded. All patients were screened to exclude a history of cardiovascular diseases or malignancies. Demographic data and laboratory indices for all patients are systematically summarized in Table 1.
## Blood sample collection and isolation of PBMCs
Peripheral blood samples from SFTS or SAE patients during the multiple organ dysfunc tion syndrome (MODS) stage were collected in either anticoagulant or clotting tubes and stored at 4°C. The blood was subsequently diluted, layered, and centrifuged using lymphocyte separation medium and phosphate-buffered saline to isolate mononuclear cells. After purification by low-speed centrifugation, the cell count was determined using a hemocytometer to obtain the required PBMCs. Isolated PBMCs were immediately cryopreserved at -80°C for subsequent RNA extraction and sequencing.
## RNA extraction, transcriptome library construction, and next-generation sequencing
Total RNA was extracted from PBMCs using QIAzol Lysis Reagent (QIAGEN, Hilden, Germany) and assessed for integrity and quantity using the Agilent 2100 Bioanalyzer system. Samples with an RNA integrity number greater than 7.5 were retained to ensure the reliability of subsequent sequencing results. Following the Illumina library construction protocol, mRNA was enriched using Oligo (DT) magnetic beads that bind the polyA tail. The mRNA was then randomly fragmented in the fragmentation buffer and used as a template for library construction. Oligonucleotide primers and M-MuLV reverse transcriptase were employed to synthesize the first and second strands of cDNA. cDNA quantification was performed using the Qubit 2.0 Fluorometer, and the library was diluted to a concentration of 1.5 ng/µL. The insert size of the library was assessed using the Agilent 2100 Bioanalyzer. Qualified libraries were pooled and sequenced on the Illumina NovaSeq 6000 platform, generating 150 bp paired-end reads. Basecaller a "-" indicates that the two groups had identical numerical values for the variable under analysis.
software was then used to convert optical signals into sequencing peaks to obtain the sequences of the target fragments.
## Bioinformatics analysis of the RNA-seq data
Raw sequencing data were processed for quality control and adapter trimming using FASTQ software to remove sequencing adapters and low-quality reads. The paired-end clean reads were then aligned to the reference genome using the HISAT2 aligner. Gene expression quantification was performed using the FeatureCounts tool. Differential gene expression analysis was conducted with the DESeq2 R package, and genes were considered significantly differentially expressed if they met the criteria of |log2 Fold Change| > 1 and a P-value < 0.05. Volcano plots and heatmaps of the most differentially expressed genes (DEGs) across comparison groups were generated using R with the EnhancedVolcano and heatmap packages. To further explore the biological significance of the differentially expressed genes, we performed gene ontology (GO) enrichment analysis and Reactome Pathway enrichment analysis. Functional annotation and pathway analysis were conducted using the online DAVID Functional Annotation Tools (https://davidbioinformatics.nih.gov/). A bubble plot was created using https:// www.bioinformatics.com.cn. Gene set enrichment analysis (GSEA) was carried out using the fgsea R package. The CIBERSORT tool, based on deconvolution algorithms, was employed to estimate the composition and abundance of immune cells within mixed cell populations. The expression of immune genes across comparison groups was analyzed using the immune gene list from the ImmPortDB database. Additionally, based on research on interferon responses published by Schoggins et al. (24), the expression of interferon-induced genes in both groups was evaluated through transcriptome sequencing data.
## Statistical analysis
Statistical analyses were performed using R version 4.1.0 and SPSS version 22.0. Violin plots were generated and analyzed using GraphPad Prism version 9.0. For normally distributed continuous data with equal variances, the two-sample t-test was applied. Categorical data with small sample sizes or low frequencies were analyzed using Pearson's χ 2 test or Fisher's exact test. Non-parametric median comparisons were conducted using the two-tailed Mann-Whitney U test. A P-value of <0.05 was considered statistically significant.
## RESULTS
## Demographic and clinical characteristics of SFTS patients
A total of 10 confirmed SFTS patients were enrolled in this study. The cohort included five patients with SAE and five patients without encephalitis. PBMCs were collected for RNA extraction and subsequent transcriptome sequencing analysis. Table 1 outlines the demographic characteristics, underlying conditions, clinical manifestations, and laboratory findings for patients with and without encephalitis. Data are presented as medians with interquartile ranges. No significant difference was observed between the two groups in terms of gender distribution (P > 0.05). Although patients in the non-ence phalitis group were older than those in the encephalitis group, this age difference did not reach statistical significance (P > 0.05). Hypertension and diabetes were the most prevalent underlying conditions among the SFTS patients, with no significant differences between the two groups. All patients (100%) exhibited high fever and fatigue, with no significant difference in the maxi mum body temperature recorded between groups. Other common symptoms, such as dizziness, headache, vomiting, diarrhea, and myalgia, showed no statistical significance, which may be attributed to the small sample size of the sequencing group. Labora tory findings at the time of specimen collection are also presented in Table 1. Both groups exhibited a reduction in white blood cell counts, though this difference was not statistically significant. Notably, the neutrophil percentage was normal in SAE patients but significantly elevated in non-SAE patients (P = 0.008). Additionally, both groups showed a decline in platelet counts, with the encephalitis group having a median of 35 × 10 9 /L and the non-encephalitis group 50 × 10 9 /L, but this difference did not reach statistical significance.
## Dynamic laboratory findings in SFTS patients with encephalitis
The natural course of SFTS is characterized by three distinct phases: the febrile stage, the MODS stage, and the convalescence stage (25). The febrile stage (days 0-6) marks the early acute phase of infection, while the second phase (days 7-13) may progress to MODS, a major contributor to disease deterioration and mortality. To elucidate the dynamic profiles of laboratory indicators in patients with SFTS-associated encephalitis, we collected clinical laboratory parameters from both groups during the first two stages and performed statistical analyses. Despite the small sample size, which limited statistical power and led to some non-significant findings, several noteworthy observa tions emerged.
During the early phase of SFTS (days 0-6), the median serum viral load, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels were higher in the encephalitis group compared to the non-encephalitis group, with ALT and AST levels significantly exceeding the normal range (Fig. 1). As the disease progressed to the second stage, viral load decreased markedly in non-encephalitis patients, while encephalitis patients maintained elevated viral copy numbers (Fig. 1), corroborating previous studies (13,15). Furthermore, compared to non-encephalitis patients, those with SFTS-associated encephalitis exhibited significantly higher lactate dehydrogenase (LDH) levels (P < 0.01) and prolonged thrombin time (TT; P < 0.01; Fig. 1), indicating the greater severity of the disease.
## Differential gene expression patterns of PBMC in SAE patients compared to non-SAE patients
The CNS complications associated with SFTS are consistently linked to fatal outcomes (25)(26)(27). However, the underlying mechanisms remain poorly understood. Given the rapid progression and pronounced variability of MODS in severe fever with thrombo cytopenia syndrome, PBMCs were obtained from both SAE and SFTS patients during the MODS phase and subjected to transcriptome sequencing to delineate transcrip tomic alterations in SAE patients. Following transcriptome sequencing, differential gene expression analysis was performed to compare the expression profiles of SAE patients to those of non-SAE patients (Table S1). A clustering analysis heatmap encompassing all 244 DEGs was generated (Fig. 2A). Additionally, a volcano plot distinctly illustrated the differential expression of 96 upregulated genes and 148 downregulated genes (|log2 FC| > 1, P < 0.05; Fig. 2B). Notably, downregulated genes demonstrated larger fold changes compared to upregulated genes. The results indicated that in the encephalitis group, potassium channel-related gene KCNN3, cell development-related genes EMC8, DANCR, and ASPN, along with the transcription factor ZNF296, were significantly upregulated. Conversely, metabolic-related gene MGAM, immune-related genes LCN2 and IL1R2, the protease inhibitor PZP, and the cytoskeletal protein MYO7A were significantly downregu lated.
To further investigate the potential functions of these DEGs, GO and reactome pathway enrichment analyses were performed. The GO enrichment analysis identified the roles and functions of these genes across three categories: biological process (BP), cellular component (CC), and molecular function (MF; Fig. 2C). In the BP category, the DEGs were predominantly associated with extracellular matrix organization, potassium ion transmembrane transport, inflammatory responses, biogenic amine metabolic processes, and innate immune responses. In terms of CC, these genes were mainly localized to the presynaptic membrane, cellular granules, and exosomes. The most enriched MF terms included L-amino acid transmembrane transporter activity, protein tyrosine kinase activity, and calcium ion binding. Furthermore, reactome pathway analysis revealed significant involvement of these genes in neutrophil degranulation, extracellular matrix degradation, collagen degradation, biological oxidation processes, and activation of matrix metalloproteinases (Fig. 2D). Collectively, these DEGs play
## GSEA reveals cellular dysfunction and immune dysregulation in SFTS-associ ated encephalitis
To provide a more comprehensive analysis of the differential gene expression profiles between patients with SFTS-associated encephalitis and those without, we employed GSEA to identify coordinated gene sets and visualize key biological processes or pathways. Our analysis revealed significant enrichment in the CC terms of GO, hallmark gene sets, and immunological signature gene sets. Notably, gene sets enriched in specific granules, secretory granule membranes, and tertiary granules were downregula ted in the non-encephalitis group, consistent with the CC findings from the GO analysis (Fig. 3A). In the hallmark gene sets, genes involved in E2F targets, unfolded protein response, and Myc targets were significantly upregulated in the encephalitis group, indicating their role in cell proliferation and stress responses (Fig. 3B). Furthermore, the analysis of immunological signature gene sets highlighted significant differences in the immune cell states and perturbations between the encephalitis and non-encephalitis groups (Fig. 3C), suggesting that alterations in host immune responses may play a crucial role in the progression of SFTS-associated encephalitis.
## Immune dysregulation and key immune gene alterations in SFTS-associated encephalitis
The clinical manifestations of SFTS are closely linked to abnormal host immune responses, which are critical in determining disease progression and severity (28)(29)(30). Numerous studies have shown that impairments in innate and adaptive immune responses are key factors in the fatal progression of SFTS (31,32). Recent studies have highlighted the significant involvement of monocytes and neutrophils, which are mobilized to infection sites to combat bacterial agents, in the pathogenesis of bacte rial meningitis (33). To investigate the immune profile in SAE patients, we employed the Cibersort algorithm to analyze immune cell infiltration characteristics. Our analysis revealed that monocytes, resting natural killer (NK) cells, and CD4+ T memory cells were the most prevalent immune cell types in both groups (Fig. 4A). In contrast, functional CD4+ T cells, CD8+ T cells, and NK cells were markedly reduced (Fig. 4A). Additionally, we categorized these immune cells based on their involvement in innate and adaptive immune response. The immune cell compositions in the SAE and non-SAE groups were similar across both response modes (Fig. 4B). Nevertheless, no significant differences were observed in the relative proportions of the various immune cell types between the two groups.
To further investigate differences in immune gene expression between SAE patients and non-SAE patients, we retrieved an immune gene list from the ImmPortDB database and performed differential gene expression analysis. The results revealed upregulation of the MET proto-oncogene receptor tyrosine kinase (MET) and KIT proto-oncogene receptor tyrosine kinase (KIT) in the encephalitis group, while interleukin 1 receptor type 2 (IL1R2) was downregulated compared to the non-SAE group (Fig. 4C). Previous studies have suggested that SFTSV inhibits the host antiviral immune response by suppressing the activation of the interferon signaling pathway (34). Therefore, we also assessed the expression of interferon-stimulated genes (ISGs) in SAE patients. Our findings indicated that MAF BZIP Transcription Factor F (MAFF) and Early T-Cell Activation Antigen P60 (CD69) were relatively upregulated in the encephalitis group, while CCAAT Enhancer Binding Protein Delta (CEBPD) was downregulated (Fig. 4D). These genes may offer valuable insights into the underlying mechanisms of SFTS-associated encephalitis.
## DISCUSSION
The development of encephalitis significantly contributes to poor prognosis and mortality in SFTS (26). Several studies have confirmed the presence of SFTSV in the cerebrospinal fluid of SAE patients, indicating that the virus can invade the central nervous system and cause intracranial infections and neurological symptoms (12,35,36). Currently, SFTS-associated encephalitis is primarily attributed to direct viral invasion and immune-pathological damage induced by cytokines. However, the exact mechanisms and host responses involved remain largely unexplored. In the present study, we enrolled five SAE patients and five non-SAE patients and analyzed their clinical manifestations upon admission and laboratory parameters during the febrile stage and the MODS stage. Owing to the limited sample size, most of the observed differences did not reach statistical significance. Nevertheless, comparative analysis of dynamic laboratory findings revealed multiple abnormal clinical laboratory parameters in SAE patients, particularly elevated LDH levels and prolonged TT. These findings are consistent with previous studies (12,15). Given the small sample size, non-significant findings should be interpreted with caution and warrant further investigation in larger cohorts.
PBMCs, mainly composed of lymphocytes and monocytes, serve as valuable sources of transcriptomic biomarkers for clinical diagnosis and could reflect global immune response in reaction to external stimuli (37). In this study, we performed whole-genome RNA sequencing of PBMCs derived from patients in the MODS phase to analyze abnormal gene expression in SAE patients compared to non-SAE patients. SFTS patients present marked clinical heterogeneity during the MODS stage. A subset of patients deteriorates rapidly, whereas others remain mildly affected (38). Transcriptomic analysis conducted during this phase is thus essential to delineate gene-expression signatures that distinguish SAE from uncomplicated SFTS. Functional enrichment analysis indicated that the differentially expressed genes were primarily associated with presynaptic membranes, cellular granules, and exosomes and were implicated in pathways related to inflammatory responses, tissue damage, cellular metabolism, and immune regulation. The inflammatory response triggered by SFTSV infection represents a double-edged sword for the host, as excessive inflammation leads to the overproduction of proinflammatory cytokines and immune hyperactivation, ultimately contributing to organ dysfunction (39). Previous studies have shown that SFTSV can invade the CNS and may induce immunopathological damage (15). Furthermore, murine models have confirmed that SFTSV can infect A1-reactive astrocytes, replicate in the brain, and subsequently trigger neuroinflammation and brain injury (16,40). Our findings strongly support these observations and suggest that the development of SFTS-associated encephalitis may be related to the extent of immunopathological damage across different individuals.
To further investigate the differences between encephalitis and non-encephalitis groups, we performed GSEA on the entire transcriptomic profiles. Consistently, we observed significant downregulation of genes associated with cellular granules in non-encephalitis patients, which are involved in pathogen infection, inflammatory regulation, and signal transduction. Analysis of hallmark gene sets revealed significant upregulation of pathways related to cell proliferation and cellular stress in the ence phalitis group, potentially associated with CNS injury and MODS induced by SFTSV infection. Moreover, our analysis displayed the immune cell perturbation patterns in the two groups. SFTSV infection disrupted immune homeostasis by suppressing the function of host immune cells while provoking an excessive inflammatory response. This immune dysregulation is characterized by early antiviral immune suppression followed by a cytokine storm, exacerbating pathological damage (41). Interestingly, immune infiltration analysis showed similar immune cell compositions in both groups, with a predominance of resting memory T cells and resting NK cells over their activated counterparts. This phenomenon may reflect immune suppression and evasion mecha nisms driven by SFTSV infection.
Considering the immune cell perturbation patterns observed in the GSEA analysis, we speculate that, compared to non-encephalitis patients, encephalitis patients may not experience substantial alterations in immune cell numbers or composition but rather exhibit changes in immune-related genes. Next, through the analysis of immune genes and ISGs, we identified six key genes: MET, KIT, IL1R2, MAFF, CD69, and CEBPD, which are critically involved in regulating cell proliferation, differentiation, immune modulation, and interferon signaling pathways. Our results indicate that, compared to the non-encephalitis group, the encephalitis group shows significant upregulation of immune-related genes MET and KIT, while IL1R2 is downregulated. The expression of interferon-induced genes MAFF and CD69 is markedly elevated, whereas CEBPD expression is significantly reduced. The MET gene encodes a receptor tyrosine kinase, and its amplification has been linked to reduced STING expression, which compromi ses interferon responses and inhibits anti-tumor immunity (42). Similarly, KIT, another receptor tyrosine kinase, is essential for cell survival and proliferation (43). IL1R2 encodes interleukin-1 receptor 2, functioning as an antagonist to IL-1 receptors, thereby inhibiting IL-1-induced inflammatory responses (44,45). The decreased expression of IL1R2 in the encephalitis group may remove the negative regulation on the IL-1 signaling pathway and contribute to excessive inflammation. MAFF, a transcription factor, correlates with inflammatory responses and immune cell infiltration in certain cancers (46). CD69 is expressed in various immune cells. Evidence indicates that CD69 enhances the activation and cytokine secretion of T cells, B cells, and NK cells (47,48). Furthermore, CD69 possibly promotes cerebral thrombus formation in mice via regulating von Willebrand factor (49). CEBPD encodes a transcription factor involved in differentiation and inflammation. It has been shown to induce the expression of secretory factors in astrocytes and affect neuronal apoptosis and inflammation (50). These gene expression profiles provide new molecular insights into the mechanisms of encephalitis pathogenesis. Aberrant activation of receptor tyrosine kinase signaling may drive abnormal prolifera tion of neuroimmune cells, while imbalances in interferon responses and dysregulation of inflammatory control may collectively disrupt immune homeostasis in the central nervous system. The dysregulation of these genes may serve as potential contributors to CNS injury induced by SFTSV infection. Further investigations are required to elucidate the functional roles of these genes in the pathogenesis of SFTS-associated encephalitis.
Several limitations exist in this study. Due to the challenges in acquiring clinical samples from SFTS patients with encephalitis, the sample size for transcriptomic sequencing was relatively limited. This limitation may affect the statistical power of differential gene expression screening and functional enrichment analysis, underscoring the need for an expanded cohort to validate these findings. A matched healthy control cohort is absent. Public data cross-validation was infeasible because baseline characteris tics of existing SFTS data sets diverge, and no SAE data are available. Future studies will rectify this limitation by expanding the sample size, enrolling uninfected controls, and employing quantitative real-time PCR for independent verification. Moreover, this study primarily relied on bioinformatics predictions and statistical analyses, lacking in vitro or in vivo experimental validation of key gene mechanisms. This gap restricts a deeper understanding of the pathogenic mechanisms involved. Furthermore, the sequencing data were mainly derived from single-time point sample collections, which restricts the dynamic assessment of immune responses and disease progression. Longitudinal cohort studies are necessary to better understand the interaction between host immune responses and disease severity throughout disease progression.
Taken together, our findings provide potential research targets for understanding the pathogenesis of SFTS-associated encephalitis, which highlight the potential role of immune microenvironment dysregulation in SFTS-induced neural injury. Our study will contribute to the broader understanding of SFTSV pathogenesis and may pave the way for the development of targeted interventions to improve outcomes in SFTS patients with CNS involvement.
## References
1. Yu, Liang, Zhang et al. (2011) "Fever with thrombocytopenia associated with a novel bunyavirus in China" *N Engl J Med*
2. Takahashi, Maeda, Suzuki et al. (2014) "The first identification and retrospective study of severe fever with thrombocytopenia syndrome in Japan" *J Infect Dis*
3. Kim, Yun, Bae et al. (2010) "Severe fever with thrombocytopenia syndrome virus infection, South Korea" *Emerg Infect Dis*
4. Tran, Yun, Van An et al. (2019) "Endemic severe fever with thrombocytopenia syndrome" *Vietnam. Emerg Infect Dis*
5. Win, Nguyen, Kim et al. (2020) "Genotypic heterogeneity of Orientia tsutsugamushi in scrub typhus patients and thrombocytope nia syndrome co-infection" *Myanmar. Emerg Infect Dis*
6. Zohaib, Zhang, Saqib et al. (2020) "Serologic evidence of severe fever with thrombocytopenia syndrome virus and related viruses in Pakistan" *Emerg Infect Dis*
7. Luo, Zhao, Wen et al. (2015) "Haemaphysalis longicornis ticks as reservoir and vector of severe fever with thrombocytopenia syndrome virus in China" *Emerg Infect Dis*
8. Hu, Zhuang, Liu et al. (2020) "Role of three tick species in the maintenance and transmission of severe fever with thrombocytopenia syndrome virus" *PLoS Negl Trop Dis*
9. Zhang, Liu, Zhao et al. (2013) "An emerging hemorrhagic fever in China caused by a novel bunyavirus SFTSV" *Sci China Life Sci*
10. Casel, Park, Choi (2021) "Severe fever with thrombocytopenia syndrome virus: emerging novel phlebovirus and their control strategy" *Exp Mol Med*
11. Li, Li, Yang et al. (2021) "Severe fever with thrombocytopenia syndrome virus: a highly lethal bunyavirus" *Crit Rev Microbiol*
12. Park, Kwon, Kim et al. (2018) "Severe fever with thrombocytopenia syndrome-associated encephalopathy/ encephalitis" *Clin Microbiol Infect*
13. Xu, Shao, Liu et al. (2021) "Clinical feature of severe fever with thrombocytopenia syndrome (SFTS)-associated encephalitis/ encephalopathy: a retrospective study" *BMC Infect Dis*
14. Shan, Chen, Liu et al. (2024) "Severe fever with thrombocytopenia syndrome with central nervous system symptom onset: a case report and literature review" *BMC Neurol*
15. Cui, Liu, Lu et al. (2015) "Severe fever with thrombocytopenia syndrome bunyavirus-related human encephalitis" *J Infect*
16. Chen, Li, Chen et al. (2022) "Severe fever with thrombocytopenia syndrome virus replicates in brain tissues and damages neurons in newborn mice" *BMC Microbiol*
17. Kaneko, Shikata, Matsukage et al. (2018) "A patient with severe fever with thrombocytopenia syndrome and hemophagocytic lymphohistio cytosis-associated involvement of the central nervous system" *J Infect Chemother*
18. Liu, Lei, Yu et al. (2017) "Correlation of cytokine level with the severity of severe fever with thrombocytopenia syndrome" *Virol J*
19. Liu, Gao, Yuen et al. (2025) "Time is now: preparing for the next pandemic"
20. Thaker, Ch'ng, Hr (2019) "Viral hijacking of cellular metabolism" *BMC Biol*
21. Chen, Liu, Guo (2020) "Emerging coronaviruses: genome structure, replication, and pathogenesis" *J Med Virol*
23. Li, Liang, Jiang et al. (2022) "PBMC transcriptomics identifies immune-metabolism disorder during the development of HBV-ACLF" *Gut*
24. Ivanov, Tarasova, Poroikov (2023) "Transcriptome-based analysis of human peripheral blood reveals regulators of immune response in different viral infections" *Front Immunol*
25. Schoggins, Wilson, Panis et al. (2011) "A diverse range of gene products are effectors of the type I interferon antiviral response" *Nature*
26. Cui, Bao, Yang et al. (2014) "Clinical progression and predictors of death in patients with severe fever with thrombocytopenia syndrome in China" *J Clin Virol*
27. Gai, Zhang, Liang et al. (2012) "Clinical progress and risk factors for death in severe fever with thrombocytopenia syndrome patients" *J Infect Dis*
28. Liu, Lu, Cui et al. (2013) "Case-fatality ratio and effectiveness of ribavirin therapy among hospitalized patients in China who had severe fever with thrombocytopenia syndrome" *Clin Infect Dis*
29. Sun, Hu, Niyonsaba et al. (2014) "Detection and evaluation of immunofunction of patients with severe fever with thrombocytopenia syndrome" *Clin Exp Med*
30. Reece, Beasley, Milligan et al. (2018) "Current status of severe fever with thrombocytopenia syndrome vaccine development" *Curr Opin Virol*
31. Hu, Kong, Liu et al. (2021) "Time course of severe fever with thrombocytopenia syndrome virus and antibodies in patients by long-term follow-up study" *China. Front Microbiol*
32. Song, Zheng, Liu et al. (2018) "Deficient humoral responses and disrupted B-cell immunity are associated with fatal SFTSV infection" *Nat Commun*
33. Zhang, Yan, Chu et al. (2025) "Bunyavirus SFTSV nucleoprotein exploits TUFM-mediated mitophagy to impair antiviral innate immunity" *Autophagy*
34. Ren, Zhang, Zheng et al. (2024) "TLR5 expression marks brain boarder associated macrophages and protects neonatal mice from bacterial meningitis"
35. Wang, Xu, Zhu et al. (2022) "Immune escape mecha nisms of severe fever with thrombocytopenia syndrome virus" *Front Immunol*
36. Nakamura, Iwanaga, Hara et al. (2019) "Viral load and inflammatory cytokine dynamics associated with the prognosis of severe fever with thrombocytopenia syndrome virus infection: an autopsy case" *J Infect Chemother*
37. Wang, Gong, Zeng et al. (2020) "Genome-based analysis of SFTSV causing severe encephalitis with brain lesions" *J Neurovirol*
38. Derbois, Palomares, Deleuze et al. (2023) "Single cell transcriptome sequencing of stimulated and frozen human peripheral blood mononuclear cells" *Sci Data*
39. Xia, Zhai, Yan et al. (2024) "Construction and validation of a dynamic nomogram using Lassologistic regression for predicting the severity of severe fever with thrombocytopenia syndrome patients at admission" *BMC Infect Dis*
40. Zhou, Yu (2021) "Unraveling the underlying interaction mechanism between Dabie bandavirus and innate immune response" *Front Immunol*
41. Kim, Choi, Jo et al. (2024) "Activation of neurotoxic A1reactive astrocytes by SFTS virus infection accelerates fatal brain damage in IFNAR1 -/-mice" *J Med Virol*
42. Yang, Huang, Jiang et al. (2022) "Overview of the immunological mechanism underlying severe fever with thrombocytopenia syndrome (review)" *Int J Mol Med*
43. Zhang, Yang, Zeng et al. (2021) "MET amplification attenuates lung tumor response to immunotherapy by inhibiting STING" *Cancer Discov*
44. Liang, Wu, Chen et al. (2013) "The Ckit receptor-mediated signal transduction and tumor-related diseases" *Int J Biol Sci*
45. Zhang, Liu, Guo et al. (2024) "Negative regulator IL-1 receptor 2 (IL-1R2) and its roles in immune regulation of autoimmune diseases" *Int Immunopharmacol*
46. Zhukova, Lopatnikova, Alshevskaya et al. (2024) "Molecular mechanisms of regulation of IL-1 and its receptors" *Cytokine Growth Factor Rev*
47. Cai, Mao, Peng et al. (2025) "A comprehensive pan-cancer examination of transcription factor MAFF: oncogenic potential, prognostic relevance, and immune landscape dynamics" *Int Immunopharmacol*
48. González-Amaro, Cortés, Sánchez-Madrid et al. (2013) "Is CD69 an effective brake to control inflammatory diseases?" *Trends Mol Med*
49. Cibrián, Sánchez-Madrid (2017) "CD69: from activation marker to metabolic gatekeeper" *Eur J Immunol*
50. Brait, Miró-Mur, Pérez-De-Puig et al. (2019) "CD69 plays a beneficial role in ischemic stroke by dampening endothelial activation" *Circ Res*
51. Pulido-Salgado, Vidal-Taboada, Saura (2015) "C/EBPβ and C/EBPδ transcription factors: basic biology and roles in the CNS" *Prog Neurobiol*
52. Chen, Chen, Zhang et al. (2021) "The genome sequence archive family: toward explosive data growth and diverse data types" *Genomics Proteomics Bioinform*
53. Xue, Bao, Zhang et al. (2022) "Database Resources of the National Genomics Data Center, China National Center for Bioinformation in 2022" *Nucleic Acids Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12449051&blobtype=pdf | # TransFactor-prediction of pro-viral SARS-CoV-2 host factors using a protein language model
Yang An, Valter Bergant, Samuele Firmani, Corinna Gr€ Unke, Batiste Bonnal, Alexander Henrici, Andreas Pichlmair, Benjamin Schubert, Annalisa Marsico
## Abstract
Motivation: Recent pandemics have revealed significant gaps in our understanding of viral pathogenesis, exposing an urgent need for methods to identify and prioritize key host proteins (host factors) as potential targets for antiviral treatments. De novo generation of experimental datasets is limited by their heterogeneity, and for looming future pandemics, may not be feasible due to limitations of experimental approaches.Results: Here, we present TransFactor, a computational framework for predicting and prioritizing candidate host factors using only protein sequence data. It leverages the pre-trained ESM-2 protein language model, fine-tuned on a limited set of experimentally determined host factors aggregated from 33 independent SARS-CoV-2 studies. TransFactor outperforms machine and deep learning baselines and its predictions align with Gene Ontology enrichments of known host factors, but also provide interpretability through a computational alanine scan, enabling the identification of pro-viral protein domains such as COMM, PX, and RRM, that may be used to direct experimental investigations of virus biology and guide rational design of antiviral therapies. Our findings demonstrate the potential of transformer-based models to advance host factor prediction, providing a framework extendable to orthogonal input modalities and other infectious diseases, enhancing our preparedness for current and future viral threats.
## 1 Introduction
Recent pandemics and epidemics, including 2016 Zika, COVID-19, and 2022/23 Mpox, underscore the need to expand our understanding of molecular events governing viral infections. This gap continues to hinder the development of effective antiviral treatments, exposing critical vulnerabilities in our preparedness and responses to both current and emerging viral threats.
Modern molecular biology allows us to study the biochemical basis of viral infections and diseases (Scaturro et al. 2018, Stukalov et al. 2021, Huang et al. 2024), and tackle the three key questions in the field: (i) what are the mechanisms driving disease pathogenicity, (ii) how can disease severity be predicted across a broad spectrum of patients, and (iii) how can disease progression be pharmacologically targeted? Due to their limited protein-coding capacity, viruses rely on the activity of distinct sets of host proteins, termed host factors, to drive aspects of their life cycle, such as uptake, replication, and egress. While most antivirals directly engage viral targets (De Clercq and Li 2016), pharmaceutical inhibition of host factors represents an attractive and underresearched opportunity (Kaufmann et al. 2018). However, host factors, and in particular key motifs driving their proviral activity, remain largely unknown for most viruses.
Virus host factor identification is dominated by small interfering RNA knock-down, and CRISPR-Cas9 knock-out screens. However, these high-throughput assays suffer from severe drawbacks, such as poor correlation between independent screens, limited availability of suitable cell lines, and variability among them, leading to substantial false-positive and falsenegative rates (Baggen et al. 2021, Rebendenne et al. 2022). Integration and prioritization of findings from high-throughput approaches, augmented with orthogonal information, may be an attractive approach to increase the identification of host targets for antiviral purposes and disease research in general. We envision that such methodologies would systematically increase the utility of high-throughput approaches by guiding experimental validation and drug target assessment efforts.
Computational methods are an emerging field with immense potential to accelerate virus research, including the identification of viral strains that harbor the risk of becoming dominant in the future (Li et al. 2024, Rancati et al. 2024) and prediction of host proteins crucially involved in viral disease pathogenesis. The latter employ diverse strategies, ranging from the analysis of transcriptomic data comparing control and infected samples, followed by differential gene and isoform expression analysis (Ferrarini et al. 2021, Mosharaf et al. 2022), to approaches leveraging protein structure to predict host proteins that may physically interact with viral proteins (Tiwari et al. 2022). Network-based techniques include computational interrogation of the virus-host protein interaction network to identify key hubs or functionally connected subnetworks (Ravindran et al. 2022, Samy et al. 2022). Another class focuses on predicting subtypes, such as RNA-binding proteins that interact with viral RNA. These predictions utilize bioinformatics pipelines or machine learning models (Vandelli et al. 2020, Horlacher et al. 2023). Finally, we and others in the past successfully used graphbased approaches to integrate knowledge on host biology with multi-omics profilings to prioritize functional follow-up of hot spots of cellular signaling perturbations upon virus infections, as well as to repurpose existing drugs toward potential SARS-CoV-2 drug targets (Morselli Gysi et al. 2021, Ruiz et al. 2021, Stukalov et al. 2021, Bergant et al. 2022, Huang et al. 2024). Many of these methodologies rely on graph-based assemblies of the host protein functional interaction landscape, such as STRING (Szklarczyk et al. 2023), which are often based on data mining and can therefore be prone to noise. These graphs are often highly connected, despite only distinct interactions being functional and impactful in any given biological state. Their undirected nature further introduces erroneous information aggregation as the causal direction of the interaction is not accounted for. Random walk with restart is then commonly applied, which assumes a linear combination of individual mechanisms, neglecting synergistic or antagonistic effects. Moreover, they are heavily reliant on omics measurements of in vitro virus infection systems, which are challenging to characterize, especially for emerging viruses, and show a high degree of variability between them. Collectively, assessment of alternative data modalities and suitable models is urgently needed to improve host factor identification in real-world scenarios and to consolidate our preparedness for existing and future viral threats.
Transformer-based protein language models (PLMs) have significantly advanced protein biology [comprehensive overview in Wang et al. (2025) and Xiao et al. (2025)]. Trained on large protein datasets in a self-supervised manner, these models learn to extract meaningful contextual, local, and global sequence features. Fine-tuning these models has enabled accurate predictions of various protein attributes, including function, fitness, family classification, and structure (Schmirler et al. 2024). Recently, PLMs have been fine-tuned or fully trained specifically on viral proteins to predict escape mutations, potentially arising strains, and to design prospective vaccines (Hie et al. 2022, Dhodapkar 2023, Thadani et al. 2023, Rancati et al. 2025, Liu et al. 2025, Youssef et al. 2025).
In this study, we propose TransFactor, a PLM-based model for predicting pro-viral SARS-CoV-2 host factors using only the protein sequence information, without the need for acquiring additional omics measurements. TransFactor leverages the pre-trained PLM ESM-2 (Lin et al. 2023) and significantly outperformed baseline methods in terms of prediction performance, such as SVMs and deep learning models. We further evaluated TransFactor's ability to generate biologically relevant hypotheses by applying the model to prioritize candidate host factors (with limited experimental evidence in the literature). Our results demonstrated that highranking candidates were more enriched than low-ranking ones in molecular functions and processes of the known host factor set. By interpreting the model's predictions using an alanine scan (Massova andKollman 1999, Kortemme et al. 2004), we identified protein regions or domains most critical for predicting SARS-CoV-2 host factors.
We envision that TransFactor will support both basic and applied antiviral research by ranking and shortlisting candidate proteins for experimental validation, accelerating the identification of host factors and their pro-viral domains, as well as assist in the design of novel antivirals.
## 2 Materials and methods
## 2.1 Data
## 2.1.1 Human protein sequences and domain information
The human proteome was assembled by collecting all 20 415 canonical and reviewed protein sequences of organism ID 9606 from UniProtKB/Swiss-Prot (accessed 2019.10.08) (uni 2025). Protein domain annotations were retrieved from UniProtKB in December 2024, including the features "Signal peptide," "Domain," "Region," "Zinc Finger," "DNA binding," "Motif," "Active site," "Binding site," and "Site."
## 2.1.2 Host factor labels
We labeled the proteins by aggregating the results from 33 independent SARS-CoV-2 assays, encompassing genome-wide, arrayed, and targeted functional screens, as well as interactomics studies, reviewed by Baggen et al. (2021) (Fig. 1a). Proteins with at least three corroborating high-throughput studies, or at least one low-throughput functional study, were considered as positives (N ¼ 1045 host factors). Conversely, proteins absent in any study were considered as negatives (N ¼ 15 434). Importantly, this way of classifying proteins was chosen due to the inherently noisy nature of highthroughput studies, which results in a minimal overlap between significant hits originating from independent studies (Baggen et al. 2021). This is further compounded by the use of different experimental systems, i.e. cell lines, statistical tests, time points, and infection doses. Based on this and our prior experience with similar assays, we expected the data to contain a significant proportion of false positives and false negatives. Finally, proteins found to be potential host factors by one or two studies were considered as candidate host factors (N ¼ 3936). These were not used during training or performance evaluation. One aim of this study was to rank the candidate set according to their likelihood of playing a pro-viral role during SARS-CoV-2 infection, thereby generating hypotheses for validation in low-throughput functional assays.
## 2.2 Model architecture
In this work, we developed and trained a model M that takes in a protein sequence X and predicts a score indicating whether the sample is a host factor b yðXÞ ¼ MðXÞ. Let X ¼ fx 1 ; x 2 ; . . . ; x L g be a sequence of length L, where each residue x i 2 AA ¼ fA; C; . . . ; Yg is from the set of the 20 canonical amino acids (AA) (Fig. 1a).
We used ESM-2 (Lin et al. 2023), an encoder-only transformer (Vaswani 2017) PLM that has been pre-trained on 65 million unique protein sequences using the masked language modeling task (Devlin 2019). The contextual residue features H ¼ ESM2ðXÞ 2 R L × D with a hidden dimension D after the last transformer layer were extracted. Due to quadratic memory scaling, we used the first 1024 residues, which 90% of our input proteins do not exceed. The mean pooled fixedlength sequence-wise feature vector h p 2 R D was then fed into a linear layer with a consecutive sigmoid layer to gain a scalar host factor score b y between 0 and 1.
The model was trained in a classification setting to distinguish host factors (positives) from non-host factors (negatives), optimizing the binary cross-entropy loss. Due to class imbalance, we scaled the loss of positive samples by a factor λ, which was treated as a hyperparameter to be optimized during tuning.
## 2.3 Training procedure
The dataset was split into six folds; five were used for crossvalidation, while the sixth fold was held out as a test set for final evaluation. To prevent data leakage, we used mmseqs2 (Steinegger and S€ oding 2017) to cluster similar sequences. Since the human proteome contains mostly dissimilar proteins, the parameters were chosen lower than the minimum sequence identity of 50% commonly used for the pre-training of PLMs. The following command and parameters were used: mmseqs easy-cluster sequences.fasta cluster_dir tmp -c 0.1 -min-seq-id 0.1 -e 0.001 Members of each cluster were grouped into a single fold. Almost half the proteins were assigned to clusters with sizes smaller than six, and 21% were singleton clusters. When taking the most abundant Gene Ontology (GO)-term from each cluster, the resulting non-singleton clusters had an average GO-term purity of 82%, 90%, and 88% for biological process, cellular component, and molecular function, respectively. These results indicate that proteins were grouped into functionally similar clusters, therefore effectively reducing the risk of data leakage (Figs 2 and 3, available as supplementary data at Bioinformatics online). Additionally, the splits were stratified by their label to approximately balance the ratio of positives and negatives between the splits.
Hyperparameter optimization was conducted using Optuna (Akiba et al. 2019), maximizing AUROC as selection criterion. On each split, hyperparameter optimization was performed for 48 h on a machine with an Nvidia A100. Early stopping was done after 25 epochs without improvement on the criterion. The model from each split with the highest validation AUROC was used for performance evaluation and downstream analysis. To aggregate the prediction values and improve the stability and performance, we further used the five resulting models in an ensemble mode by taking the average prediction score. We opted for averaging as the models share the same base architecture, while differing in their TransFactor training data splits. Hence, we assume that applying the same weighting provides a stable consensus.
To prevent catastrophic forgetting and improve the training speed and memory requirements, we used Low-Rank Adaptation (LoRA) (Hu et al. 2022) to fine-tune the ESM-2 backbone. Following the original paper, we only adapted the attention weights W Q ; W K ; W V ; W O , while freezing all other parameters of the ESM-2 backbone.
## 2.4 Baselines and ablation
We evaluated TransFactor against two baselines and two ablation variants. First, we re-implemented TriPepSVM (Bressin et al. 2019)-a linear and RBF-kernel SVM that classifies proteins based on overlapping 3-mer counts. Second, we adapted a CNN-LSTM hybrid similar to Wu and Guo (2024), which uses the protein sequence as input and passes the last hidden features of the LSTM as input into a linear classification head. Finally, to assess transfer learning, we trained two ablated versions of TransFactor, one with frozen ESM-2 weights (TF frozen BB), and one with a randomly initialized backbone rather than pre-trained (TF init BB). For details of baselines and hyperparameters, refer to Text A.1 and Tables 23456, available as supplementary data at Bioinformatics online.
## 2.5 GO enrichment
GO enrichment analysis was conducted using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) tool (DAVID Knowledgebase v2024q4, released on 22 December 2024; available at https://davidbioinformatics. nih.gov/summary.jsp) (Sherman et al. 2022). The analysis was performed separately for the candidate proteins with a prediction score above and below the ideal threshold τ, which was determined on the validation set as reaching the highest F1-score. The entire candidate protein set was used as the background. Similarly, for the positive protein set, the entire protein dataset was used as the background. Enrichment was assessed across the three main GO categories: biological process (BP), molecular function (MF), and cellular component (CC). Default parameters in DAVID were applied for statistical testing and multiple testing correction.
## 2.6 Model's interpretation through computational alanine scan
To understand the attribution of amino acid motifs to the overall prediction, we performed a computational alanine scan (Massova andKollman 1999, Kortemme et al. 2004) (Fig. 1b). For the selected protein sequence X, we substituted the amino acids in a contiguous window with alanine. Alanine has a small methyl side chain and, therefore, is the most functionally inert amino acid, commonly used in singlepoint-mutagenesis-based experiments to interrogate the functional relevance of distinct protein regions (Morrison and Weiss 2001). We utilized window sizes w of 1, as well as values ranging from 5 to 40 in increments of 5. The wild-type sequence was mutated to X x i:i þ w ! A ¼ fx 1 ; . . . ; x i -1 ; A; . . . ; A; x i þ w ; . . . ; x L g. We then used the trained model to predict the host factor score b yðX xi:iþ w ! A Þ of the mutated sequence. The difference in prediction score Δb y Ala ¼ b yðX xi:iþw ! A Þb yðX wt Þ between the mutated and wild-type sequence was used as the attribution score for residues within the substitution window. For samples from the test set, we used the ensemble model prediction scores, while for samples from the cross-validation splits, the model corresponding to the validation set was used.
For domain-wise statistical testing of significant deviations of Δb y Ala (Fig. 4, available as supplementary data at Bioinformatics online), we used the one-sided Wilcoxon rank-sum test and compared Δb y Ala values within domains to all values across all proteins for any given alanine scan window. The thus obtained P-values were further FDR-adjusted.
## 3 Results
## 3.1 TransFactor outperforms baseline models in predicting SARS-CoV-2 host factors
First, we evaluated the performance of our proposed method on the held-out test set. Due to the high imbalance of 4% positive test samples, we chose the area under the receiver operating characteristics curve (AUROC), average precision score (APS) as a conservative estimation method for the area under the precision-recall curve, and F1-score, further broken down into precision and recall as metrics. We determined the ideal thresholds τ based on the highest F1-score on the validation set for each model, respectively. For the ensembles, we optimized the threshold on the whole training dataset (τ ¼ 0:571). The benchmark results of the best model trained on each of the five folds are shown in Fig. 1c andTable 1 (first row within each method), available as supplementary data at Bioinformatics online.
The simplest model, an SVM with a linear kernel, consistently showed the lowest performance across all metrics. Replacing the linear kernel with an RBF kernel led to moderate improvements. Both sequential deep learning models, the CNN-LSTM hybrid and TransFactor with a randomly initialized backbone, performed similarly to SVM with RBF kernel. However, using a frozen pre-trained ESM-2 backbone greatly increased AUROC (0.78-0.86), APS (0.14-0.24), and F1score (0.19-0.30) in comparison to the best other model. Using LoRA fine-tuning on the weights of the backbone, we could further improve the performance in three out of five metrics.
These results highlight the need for pre-trained language models for these data. The sequential deep learning models failed to outperform the traditional machine learning baseline. This indicates that the sequential models could not identify and extract functional information from the raw sequences. One possible explanation lies in the diversity of the human proteome. Sequences originated from diverse sets of protein families with very high dissimilarities to each other. Specifically, the 20 415 sequences were distributed in 6961 clusters despite a minimum sequence identity of 10% and coverage of 10%. Forty-eight of all sequences were in clusters of size five or smaller. These properties make it challenging for the model to rely solely on sequence information. Through pre-training, the model learns to extract meaningful features and capture functional information from evolutionary conservation patterns, where distant sequences may still result in related function and structure.
Next, the prediction scores of the five individual models were aggregated and averaged. The resulting ensemble models consistently improved upon the mean performance for all individual models of each underlying architecture (Table S1, second row within each method, available as supplementary data at Bioinformatics online). For TransFactor with finetuned backbone, we observed increases in AUROC from 0.87 to 0.89, APS from 0.27 to 0.30, and F1-score from 0.23 to 0.38 through the aggregation of prediction scores. To gain an estimate of the expected hit rate during validation, we calculated the Precision of the Top-K predicted samples (Precision@K) (Fig. 1d). Except for the noisy lower K range, TransFactor consistently reached higher Precision@K values than the baseline methods. At typical experimental capacities of 50, 100, and 200, TransFactor had a precision of 0.44, 0.37, and 0.28, respectively. Due to the condensed score and higher performance, we used the ensemble model for further analyses, unless otherwise indicated.
To further contextualize the performance of our computational method, we evaluated the predictive power of experimental screens by using each one (excluding functional validation screens) from the review by Baggen et al. (2021) to correctly identify host factors from the full human proteome. We applied our labeling scheme, with a minor modification: the screen under evaluation was excluded from the labeling process. On average, experimental screens achieved an F1score of 0.13 ± 0.11 (mean ± standard deviation), with a precision of 0.60 ± 0.25 and a recall of 0.11 ± 0.13 (details in Table S7, available as supplementary data at Bioinformatics online).
## 3.2 GO-term enrichment reveals biological consistency in high-scoring uncertain proteins
To assess the potential of our model in guiding the selection of protein candidates from high-throughput screens with lower precision, we scored each protein from the candidate set with TransFactor. The resulting score distribution fell between those of the positive and negative test samples (Fig. 1e). Eight hundred eighty-three of the 3936 candidates were predicted as potential host factors. High-throughput screenings are expected to yield many false positives, consistent with our prediction's distribution. This indicates TransFactor's potential to help distinguish prospective novel host factors from experimental noise, proposing a shortlist of candidates for further investigation.
Next, we evaluated TransFactor's predictions using GO enrichment analysis to determine whether the model captures biological relevance and stratifies candidate proteins into promising and less promising ones. Enriched GO-term of predicted host factors closely mirrored those of known positives, showing strong overlap across BP, MF, and CC categories [Fig. 2 These results indicate that the model successfully captures the structural and functional organization of host factor proteins. This predictive capability could provide a valuable framework for prioritizing putative candidates for targeted experimental validation.
## 3.3 Computational alanine scan identifies domains important for the model's prediction
To identify motifs and domains that were affecting the predictions, we performed a computational alanine scan on the positive subset of proteins (Massova andKollman 1999, Kortemme et al. 2004). We employed a broad range of window sizes to introduce varying degrees of perturbations to protein sequences, enabling us to assess the impact of both small-and large-scale changes on prediction scores. This approach allowed us to find a balance between the amplitude and resolution of the explanations.
First, we evaluated whether the model accurately captured the well-established principle that substituting amino acids with alanine often reduces protein functionality. Consistent with this concept, our results revealed a clear trend of decreased prediction scores following alanine substitutions. Notably, this effect became more pronounced with increasing sizes of the alanine scan windows (Fig. 3a).
Furthermore, we mapped the alanine scan attribution scores to protein domains to assess if the model learned to specifically penalize the alanine substitutions in regions of proteins known to be functionally relevant. Notably, we observed a decrease in median attribution scores within protein domains enriched in the positive subset (Fig. 3b and Fig. 4a, available as supplementary data at Bioinformatics online). In contrast, domains less represented in positively labeled proteins showed almost no decrease in median attribution scores (Fig. 3c and Fig. 4a, available as supplementary data at Bioinformatics online). These findings strongly indicate that the model can distinguish protein sequences and regions that are more important for host factor classification and can learn from given exemplary sequences containing similar regions.
After showcasing the general prevalence of the model to recognize the functionally important features of proteins, we took a detailed look at some domain types. Importantly, not all domains exhibited statistically significant decreases in their attribution scores in comparison to values across all scanned proteins (Fig. 4b, available as supplementary data at Bioinformatics online). While we observed the strongest decrease upon introduction of alanine substitutions in copper metabolism gene MURR1 (COMM) domains (Fig. 3d), phox homology (PX) domains (Fig. 3e), and RNA recognition motifs (RRM) (Fig. 3f), we did not observe this for many other domains such as Helicase C-terminal domains (Fig. 3g). Particularly striking was the strong decrease in attribution scores for the COMM domains of COMMD proteins. In humans, there are 10 COMMD proteins, which encode a COMM domain. Seven of them were previously shown to play a role in SARS-CoV-2 infection (Zhu et al. 2021) and thereby were contained in our positive set (COMMD2/3/4/5/ 7/8/10) (Baggen et al. 2021). COMMD proteins, together with CCDC22 and CCDC93, form the CCC complex, which, together with the retriever complex (VPS35L, VPS26C, VPS29), forms the commander complex, involved in the endosomal cargo trafficking and recycling (Healy et al. 2023). COMMD proteins, as well as proteins and assemblies related to these processes, were previously shown to be SARS-CoV-2 host factors (Baggen et al. 2021), but to the best of our knowledge, no specific parts of these proteins are so far known to be critical for this functionality. COMMD proteins, except COMMD6, were among the top predicted host factors by TransFactor and had prediction scores between 0.75 and 0.98. Interestingly, our alanine scan results, in particular evident for the lower range of alanine scan window sizes, suggested that the relatively poorly conserved Cterminal part of the COMM domain in COMMD4 (Fig. 3h) and other COMMD proteins (Fig. 3i-j and Fig. 5, available as supplementary data at Bioinformatics online) may play a central role in their ability to support SARS-CoV-2 replication as host factors.
## 4 Discussion
The COVID-19 pandemic has resulted in unprecedented socioeconomic disruptions and more than 6 million lost lives. Despite the need for effective antiviral therapies, we still do not fully understand the molecular basis of SARS-CoV-2 infection. Virology relies on experimental studies of virus-host interactions, but dataset heterogeneity and practical challenges, such as high virulence or hard-to-culture viruses, limit comprehensive characterization of host factors and their role in infection etiology and progression. High-throughput methods are often infeasible for some viruses or poorly followed up due to resource constraints, underscoring the need for computational models to predict and prioritize key interactions from limited and noisy data. Emerging AI technologies, such as sequence-based deep learning models, offer significant potential to uncover critical host factors that facilitate viral infections. Inspired by the success of PLMs, we developed TransFactor-a transformer-based method for predicting virus host factors based on protein sequence information and a limited set of experimentally determined host factors. By leveraging pre-trained PLM's feature extraction capabilities and fine-tuning on the classification task of distinguishing SARS-CoV-2 host factors from non-host factors, TransFactor outperformed machine and deep learning baseline methods. Candidate host factors prioritized by TransFactor showed similar GO-term enrichments as known host factors, giving more confidence in the model's capability to rank and prioritize proteins. Through a computational alanine scan, TransFactor could identify domains important for the prediction, helping to understand the molecular basis underlying host factors.
However, TransFactor faces limitations that present opportunities for improvement in future work. Currently, TransFactor's input is truncated to 1024 amino acids for efficient training and inference of the model, which may omit important C-terminal regions relevant for host-virus interactions. To assess this, we reevaluated our trained models on the test proteins truncated at 2048, increasing the coverage from 90% to 98%. This yielded a slight improvement in five out of eight performance metrics (Table S8, available as supplementary data at Bioinformatics online). Nevertheless, as computational resources grow drastically with sequence length, and practical implications for large proteins (longer than 1000), including reduced sequencing fidelity, expression strength, and detection by Western blot, we adopted the shorter length for this work. To further mitigate these aspects, we plan to explore random cropping during training, memory-efficient methods like FlashAttention (Dao et al. 2022), and fine-tuning on longer sequences in the future.
To effectively train and employ deep learning models, an extensive amount of labeled data, i.e. known host factors, is needed. Such data is not universally available for many viruses, originates from a variety of strains, and was acquired in different infection models. However, transfer learning and domain adaptation techniques could leverage data from closely related viruses to make predictions for those with limited or no known host factors. Transferability depends primarily on the similarity of pathways and biochemical processes the viruses need for their life cycles and the resulting set of host factors. The overlap and divergence of essential host processes could give an estimate for the success of the transfer. Yet, for most viruses, these factors are not known in sufficient depth. Instead, the genetic similarity and evolutionary proximity can serve as a proxy. To demonstrate this, we used our TransFactor model trained on SARS-CoV-2 host factors to predict host proteins interacting with SARS-CoV viral proteins as a proxy for its host factors (N ¼ 612, Stukalov et al. 2021) without fine-tuning. While the performance declined as expected, we were able to recall 42% of SARS-CoV, indicating the potential to prioritize host factors Figure 2. GO-term enrichment was performed on the positively labeled proteins (using all proteins as background), and on the candidate proteins (taking all candidate proteins as background) predicted as positive (b y > τ) and negative (b y ≤ τ). The enrichment analysis was performed separately for all three gene ontologies, and terms significant for at least one gene set are displayed. Venn diagrams show the overall and overlapping number of enriched terms in and between the three sets.
for phylogenetically related viruses. In contrast, the model failed to identify putative host factors (N ¼ 368, Montoya et al. 2023) of the phylogenetically distant HIV with a recall of 18% (details in Table S9, available as supplementary data at Bioinformatics online).
Moreover, TransFactor currently relies exclusively on primary sequence information, which makes classifying proteins with low homology to the training data particularly challenging. Integrating orthogonal data, such as the tertiary structure, may provide the model with additional valuable insights into protein functionality, alleviating the problem of low sequence homology. Similarly, incorporating additional experimental information such as peptide-level abundances, phosphorylation, and ubiquitination events obtained by LC-MS/MS-based proteomics may allow the model to access signaling perturbations instigated by the invading pathogen, allowing it to consider cellular signaling state on one or more functional layers. As proteins rarely act in isolation, incorporating protein-protein interaction network data could add a higher-order layer to the model, capturing complex biological processes. These additional data modalities could help address current challenges, such as low F1 scores, by identifying features not captured by sequence information alone and expand the generalizability to related and potentially more distant viruses. Addressing label noise and the high imbalance in existing host factor datasets through further low-throughput functional studies and explicit modeling of label uncertainty could also improve the model's discriminative performance. Ultimately, experimental validation of TransFactor's predictions through functional screens will be essential. The results of such experiments could further enhance the model in an active learning, lab-in-the-loop setting. While we presented a proof-of-concept on SARS-CoV-2 host factors, further evaluations on other diseases are necessary to assess the generalizability of TransFactor.
In summary, we have shown that TransFactor can reliably identify and rank host factor proteins. Combined with a computational alanine scan, TransFactor enables the detailed analysis and interpretation of how specific amino acid motifs contribute to pro-viral pathogenesis. This approach provides valuable insights at a meaningful scale, offering a robust foundation for generating hypotheses that can be further tested in appropriate experimental models, facilitating the advancement of our biological understanding of viral infectious diseases, and providing a valuable resource to guide rational antiviral design.
## References
1. Akiba, Sano, Yanase (2019) "Optuna: a next-generation hyperparameter optimization framework"
2. Baggen, Vanstreels, Jansen (2021) "Cellular host factors for SARS-CoV-2 infection" *Nat Microbiol*
3. Bergant, Yamada, Grass (2022) "Attenuation of SARS-CoV-2 replication and associated inflammation by concomitant targeting of viral and host cap 2 0 -o-ribose methyltransferases" *EMBO J*
4. Bressin, Schulte-Sasse, Figini (2019) "TriPepSVM: de novo prediction of RNA-binding proteins based on short amino acid motifs" *Nucleic Acids Res*
5. Dao, Fu, Ermon (2022) "FlashAttention: fast and memory-efficient exact attention with io-awareness" *Adv Neural Inf Process Syst*
6. De Clercq, Li (2016) "Approved antiviral drugs over the past 50 years" *Clin Microbiol Rev*
7. Devlin, Chang, Lee (2019) "Pre-training of deep bidirectional transformers for language understanding"
8. Dhodapkar (2023) "A deep generative model of the SARS-CoV-2 spike protein predicts future variants" *bioRxiv*
9. Ferrarini, Lal, Rebollo (2021) "Genome-wide bioinformatic analyses predict key host and viral factors in SARS-CoV-2 pathogenesis" *Commun Biol*
10. Healy, Mcnally, Butkovi� C R (2023) "Structure of the endosomal commander complex linked to Ritscher-Schinzel syndrome" *Cell*
11. Hie, Yang, Kim (2022) "Evolutionary velocity with protein language models predicts evolutionary dynamics of diverse proteins" *Cell Syst*
12. Horlacher, Oleshko, Hu (2023) "A computational map of the human-SARS-CoV-2 protein-RNA interactome predicted at singlenucleotide resolution" *NAR Genom Bioinform*
13. Hu, Shen, Wallis (2022) "Lora: Low-rank adaptation of large language models" *ICLR*
14. Huang, Bergant, Grass (2024) "Multi-omics characterization of the monkeypox virus infection" *Nat Commun*
15. Kaufmann, Dorhoi, Hotchkiss (2018) "Host-directed therapies for bacterial and viral infections" *Nat Rev Drug Discov*
16. Kortemme, Kim, Baker (2004) "Computational alanine scanning of protein-protein interfaces" *Sci STKE*
17. Li, Li, Li (2024) "Machine learning early detection of SARS-CoV-2 high-risk variants" *Adv Sci*
18. Lin, Akin, Rao (2023) "Evolutionary-scale prediction of atomiclevel protein structure with a language model" *Science*
20. Liu, Nie, Si (2025) "Generative prediction of real-world prevalent SARS-CoV-2 mutation with in silico virus evolution" *Brief Bioinform*
21. Madeira, Madhusoodanan, Lee (2024) "The EMBL-EBI job dispatcher sequence analysis tools framework in 2024" *Nucleic Acids Res*
22. Massova, Kollman (1999) "Computational alanine scanning to probe protein-protein interactions: a novel approach to evaluate binding free energies" *J Am Chem Soc*
23. Montoya, Ready (2023) "A virus-packageable CRISPR system identifies host dependency factors co-opted by multiple HIV-1 strains" *mBio*
24. Morrison, Weiss (2001) "Combinatorial alanine-scanning" *Curr Opin Chem Biol*
25. Gysi, Valle � I, Zitnik (2021) "Network medicine framework for identifying drug-repurposing opportunities for covid-19" *Proc Natl Acad Sci*
26. Mosharaf, Reza, Kibria (2022) "Computational identification of host genomic biomarkers highlighting their functions, pathways and regulators that influence SARS-CoV-2 infections and drug repurposing" *Sci Rep*
27. Rancati, Nicora, Bergomi (2025) "SARITA: a large language model for generating the S1 subunit of the SARS-CoV-2 spike protein" *Brief Bioinform*
28. Rancati, Nicora, Prosperi (2024) "Forecasting dominance of SARS-CoV-2 lineages by anomaly detection using deep autoencoders" *Brief Bioinform*
29. Ravindran, Wagoner, Athanasiadis (2022) "Discovery of hostdirected modulators of virus infection by probing the SARS-CoV-2host protein-protein interaction network" *Brief Bioinform*
30. Rebendenne, Roy, Bonaventure (2022) "Bidirectional genome-wide CRISPR screens reveal host factors regulating SARS-CoV-2, MERS-CoV and seasonal HCoVs" *Nat Genet*
31. Ruiz, Zitnik, Leskovec (2021) "Identification of disease treatment mechanisms through the multiscale interactome" *Nat Commun*
32. Samy, Maher, Abdelsalam (2022) "SARS-CoV-2 potential drugs, drug targets, and biomarkers: a viral-host interaction network-based analysis" *Sci Rep*
33. Scaturro, Stukalov, Haas (2018) "An orthogonal proteomic survey uncovers novel Zika virus host factors" *Nature*
34. Schmirler, Heinzinger, Rost (2024) "Fine-tuning protein language models boosts predictions across diverse tasks" *Nat Commun*
35. Sherman, Hao, Qiu (2022) "DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update)" *Nucleic Acids Res*
36. Steinegger, S€ (2017) "MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets" *Nat Biotechnol*
37. Stukalov, Girault, Grass (2021) "Multilevel proteomics reveals host perturbations by SARS-CoV-2 and SARS-CoV" *Nature*
38. Szklarczyk, Kirsch, Koutrouli (2023) "The string database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest" *Nucleic Acids Res*
39. Thadani, Gurev, Notin (2023) "Learning from prepandemic data to forecast viral escape" *Nature*
40. Tiwari, Mishra, Gupta (2022) "Structural similarity-based prediction of host factors associated with SARS-CoV-2 infection and pathogenesis" *J Biomol Struct Dyn*
41. (2025) "Uniprot: the universal protein knowledgebase in 2025" *Nucleic Acids Res*
42. Vandelli, Monti, Milanetti (2020) "Structural analysis of SARS-CoV-2 genome and predictions of the human interactome" *Nucleic Acids Res*
43. Vaswani, Shazeer, Parmar "Attention is all you need" *NIPS*
44. Wang, Li, Zhang (2025) "A comprehensive review of protein language models. arXiv"
45. Wu, Guo (2024) "Improved prediction of DNA and RNA binding proteins with deep learning models" *Brief Bioinform*
46. Xiao, Zhao, Zhang (2025) "Protein large language models: a comprehensive survey. arXiv"
47. Youssef, Gurev, Ghantous (2025) "Computationally designed proteins mimic antibody immune evasion in viral evolution" *Immunity*
48. Zhu, Feng, Hu (2021) "A genome-wide CRISPR screen identifies host factors that regulate SARS-CoV-2 entry" *Nat Commun* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12662413&blobtype=pdf | # Daily mosnodenvir dosing as dengue prophylaxis in a human infection model
Anna Durbin, Liesbeth Van Wesenbeeck, Kristen Pierce, Guillermo Herrera-Taracena, Laura Ebone, Annemie Buelens, Patricia Lutton, Beulah Sabundayo, Veerle Van Eygen, Kim De Clerck, Isabel Fetter, Natalia Voge, Xi Fang, Nele Goeyvaerts, Yannick Vandendijck, Jeffrey Mayfield, Oliver Lenz, Sandra De Meyer, Thomas Kakuda, Emérito Amaro-Carambot, Draghia Ruxandra, Akli, Marya Carmolli, Tine De Marez, Stephen Whitehead, Marnix Van Loock, Freya Rasschaert
## Abstract
Background: Antiviral prophylaxis or treatment is unavailable for dengue. We evaluated the antiviral activity, safety and pharmacokinetics of repeated oral doses of mosnodenvir (JNJ-64281802), a pan-serotype dengue antiviral, as pre-exposure prophylaxis against dengue virus serotype 3 (DENV-3) infection in a controlled human infection model (CHIM).
Methods:In this phase 2a, double-blind study, healthy adults were randomized to receive once daily oral mosnodenvir at different dose levels or placebo for 26 days (5 days loading dose [LD], 21 days maintenance dose [MD]). An under-attenuated DENV-3 strain (rDEN3Δ30) was subcutaneously injected on the day (D) of first MD (D1). Safety, pharmacokinetics, virology and serology parameters were evaluated up to D85.
Dengue is an acute disease caused by four antigenically different serotypes of dengue virus (DENV-1-4), transmitted by Aedes mosquitoes. 1 Dengue is a growing public health threat, with about half of the world population being considered at risk by the World Health Organization. 2 Dengue continues to increase due to population growth and climate changes allowing vector expansion in endemic areas, such as the Americas and Southeast Asia and is spreading to temperate regions in the USA and Europe. [3][4][5][6][7] Dengue disease can manifest as undifferentiated fever or non-severe dengue fever. 8 In a small proportion of cases, the disease can progress to severe manifestations such as dengue hemorrhagic fever or dengue shock syndrome, with a potentially fatal outcome. 9 Licensed dengue treatments are currently unavailable and management consists of supportive measures, such as antipyretics and volume repletion. 10 Two live attenuated tetravalent vaccines, CYD-TDV (which recently stopped production) and TAK-003 (in certain countries), are currently approved for use. 11 A third, Butantan-DV, has filed for licensure in Brazil. [12][13][14] Vaccines can take weeks to provide protective immunity and have varying efficacy by baseline DENV immunity. 13,15 In contrast, an oral antiviral could be deployed in outbreaks, could limit cases in endemic areas through a prophylactic approach, and may be used for travelers and those who cannot receive a vaccine. [16][17][18] Mosnodenvir (also known as JNJ-64281802) is an oral, pan-serotype dengue small-molecule antiviral, that blocks viral replication by inhibiting de novo DENV nonstructural protein 3 (NS3)-NS4B interaction. 19,20 Mosnodenvir exhibits picomolar to nanomolar in vitro antiviral potency against a representative DENV genotype panel and has antiviral efficacy in mice and non-human primates. 19 In a first-in-human study, no safety concerns were identified with mosnodenvir. 21 Here, we report the initial results of a phase 2a study assessing the prophylactic antiviral activity, safety, and pharmacokinetics of different daily doses of mosnodenvir against DENV-3 infection (under-attenuated strain) in healthy participants.
## Methods
## Study oversight
This is a phase 2a, randomized, double-blind, placebo-controlled human infection model (CHIM) study conducted at the Johns Hopkins School of Public Health and the University of Vermont, United States (US), starting in February 2022 (ClinicalTrials.gov number, NCT05048875). The study consists of two cohorts conducted in a staggered manner: Cohort 1 is dose finding with 2 dose escalation groups and Cohort 2 assesses three different regimens based on Cohort 1's findings, including weekly dose regimens. Cohort 1, Group 1 is high dose 600 mg daily (QD) loading dose (LD)/200 mg QD maintenance dose [MD] and placebo; and Cohort 1, Group 2 is medium dose 200 mg QD LD/50 mg QD MD and low dose 40 mg QD LD/10 mg QD MD and placebo (Figure 1). The data from Cohort 1 are presented here.
The study is conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. Written informed consent was obtained from each participant prior to any study-related activities. Mosnodenvir was supplied as 10 mg, 50 mg, and 100 mg oral capsules and administered under fasted conditions. The study protocol (available with the full text of this article at NEJM.org) was reviewed and approved by Independent Ethics Committees at each site and written informed consent was obtained from each participant. Data were gathered by the study site investigators and analyzed at Johnson & Johnson in collaboration with the National Institutes of Health (NIH). All the authors vouch for the accuracy and completeness of the data presented and for the fidelity of the study to the protocol. Medical writing assistance was funded by Johnson and Johnson.
## Participants
We enrolled healthy individuals 18-55 years of age, who were confirmed to be seronegative to DENV and Zika virus (ZIKV) prior to enrollment, had not traveled to any dengueendemic region within 4 weeks from enrollment nor planned to do so and had not received any live attenuated vaccines within 28 days before and after study drug intake. For more details on the inclusion and exclusion criteria, see the protocol.
## Study procedures
All participants attended screening visits between day (D)-65 and D-6, received either mosnodenvir or placebo (oral QD and under fasted conditions) as a LD from D-5 to D-1, followed by a MD from D1 to D21, with a challenge of 3 log10 plaque-forming units (PFU) of the under-attenuated virus rDEN3Δ30, 22 administered subcutaneously on D1. Participants were randomized to receive either high-dose mosnodenvir; (N=10) or matching placebo (N=6) (Group 1) or medium-(N=6) and low-dose (N=6) mosnodenvir or matching placebo (N=2) (Group 2) (Figure 1). All participants were admitted to the inpatient unit for the first two dosing days (D-6 through D-4) and observed for at least 30 minutes after initial dosing and/or inoculation to ensure their safety. Safety parameters were monitored throughout the study (clinical laboratory tests, electrocardiogram, vital signs, and physical examinations) and solicited and unsolicited adverse events (AEs) were evaluated. Blood samples were taken at regular time points for virology and pharmacokinetic assessments. Participants were followed up through D85 (64 days after the last dose). More information about the study procedures is available in the Supplementary Appendix and protocol.
## Study objectives
The primary objective was to assess the antiviral activity of mosnodenvir versus placebo in terms of reduction of DENV-3 RNA by evaluating the area under the DENV-3 RNA viral load (VL) concentration-time curves from immediately before inoculation (D1) until D29 (AUCD1-D29). Secondary objectives included safety and tolerability, occurrence and severity of DENV infection-associated AEs, other virology parameters, antibody responses, pharmacokinetics of mosnodenvir and the characterization of the relationship between pharmacokinetics and antiviral activity of mosnodenvir under different QD dose regimens.
## Virology and pharmacokinetic assessments
DENV-3 RNA serum levels were assessed using a validated quantitative DENV reverse transcriptase polymerase chain reaction (RT-qPCR) assay. DENV-3 viremia was determined on DENV-3 RNA positive samples using a plaque assay and anti-DENV immunoglobulin (Ig)G and IgM antibodies were measured by enzyme-linked immunosorbent assay (ELISA; Euroimmun). DENV-3 viral sequencing was performed using Illumina sequencing technology to characterize emerging DENV-3 genetic variations. Emerging amino acid variations were defined as having a sequence read frequency ≥15% at a post-baseline visit while absent and a read frequency <3% in the inoculated rDEN3Δ30 strain sequence. Blood samples were obtained over 24 hours on D5 and D21 to measure mosnodenvir plasma concentrations using a validated, liquid chromatography-tandem mass spectrometry method. 23 Pharmacokinetic parameters, including maximum plasma concentration (C max ), time to C max (t max ), average concentration (C avg ), terminal elimination half-life (t ½ ) and area under the plasma concentration-time curve (AUC), were determined using the validated software Phoenix (Certara, Princeton, NJ, USA). Additional details about these and other assessments (e.g., Saint-Louis encephalitis virus [SLEV], ZIKV, reporter virus particle neutralization tests (RVPNT)) are available in the Supplementary Appendix and protocol.
## Sample size calculation
Data were simulated using a Bernoulli distribution for the infection rate (assuming 90% under placebo) and using a normal distribution for the log 10 AUC D1-D29 (VL) in the infected participants (mean 5.5 log 10 copies/mL/28 days, standard deviation 0.70). Based on these simulations, the power to detect a relevant reduction of ≥ 30% on log 10 AUC D1-D29 (VL) at the 2-sided 10% significance level was calculated to be more than 85% with 6 participants in the placebo arm and 10 participants in the mosnodenvir high-dose arm. The number of participants in Cohort 1 is considered sufficient for an initial characterization of the relationship between pharmacokinetics and antiviral activity of mosnodenvir based on DENV-3 RNA. Details can be found in the protocol.
## Statistical analysis
The primary efficacy analysis included all participants from Cohort 1 Group 1 who were inoculated with rDEN3Δ30. A Tobit analysis of variance with log 10 AUC D1-D29 (VL) as dependent variable and the study drug as a fixed covariate was performed to test whether a significant difference between mosnodenvir and placebo was observed, at the 2-sided 10% significance level, provided that at least 65% of the inoculated participants in the placebo arm had detectable DENV-3 RNA at any of the assessments up to D29. Values were left censored for participants with undetectable DENV-3 RNA up to D29. The exact Wilcoxon rank sum test was also performed. Descriptive statistics are provided on the primary and secondary endpoints. Analyses were performed with SAS 9.04 (SAS Institute Inc., Cary, NC, USA). Graphs were created in R version 4.2.0 (Comprehensive R Network, http://cran.r-project.org/).
## Results
## Participants
In Cohort 1, 31 participants were recruited between February 2022 and February 2023 and were included in the safety analysis set. Thirty (30) participants were randomized to receive either mosnodenvir (N=22) or placebo (N=8) (Table 1, Figure 1, Figure S1), 1 additional participant was enrolled as a replacement for a participant in the mosnodenvir group and 29 inoculated participants were included in the efficacy analysis. The baseline characteristics were similar between the placebo and mosnodenvir as well as across different dosing arms (Table 1) except for more females included in the medium-dose arm (83%) when compared to the other arms (Table 1). In the placebo arm, one participant missed a single dose on Day 18 and another missed a single dose on Day 21 and took this dose on Day 22 instead. All other participants from Cohort 1 were 100% medication compliant. The representativeness of the study population is shown in Table S1.
## Primary endpoint analysis
A Tobit analysis of variance showed a statistically significant reduction on the log 10 AUC D1- D29 VL in the high-dose mosnodenvir arm versus the placebo arm (2-sided p<0.001) (Figure 2A and Table S2). The statistically significant result was also seen by the exact Wilcoxon rank sum test (2-sided p=0.001).
## Daily dose regimens
A dose-dependent antiviral activity, as measured by log 10 AUC values, across the different mosnodenvir regimens was observed (Figure 2). The proportion of participants with all available DENV-3 RNA measurements being undetectable was 0% (0/6), 17% (1/6), 60% (6/10) in the low-, medium-, and high-dose arms, respectively, versus 0% (0/7) in the placebo arm (Figure S2). In all mosnodenvir-dosed participants without detectable DENV-3 RNA, no anti-DENV IgM/IgG nor neutralizing antibodies (nAbs) were observed until D85 (Figure S2, Table S3).
For participants with detectable DENV-3 RNA in the mosnodenvir dose regimens, the peak DENV-3 RNA levels were comparable to placebo, with the exception of 3 participants with peak DENV-3 RNA levels detectable below or around the lower limit of quantification (LLOQ) (Figure S2); and the median time to first onset of detectable DENV-3 RNA was delayed in a dose-dependent manner (Figure 2). In all participants with detectable DENV-3 RNA, infectious virus was detected, except for 1 participant in the medium-dose arm and 1 participant in the high-dose arm (Figure S2), and positive anti-DENV IgM and/or positive anti-DENV IgG and/or positive nAbs on ≥1 assessment after baseline were observed, except for 1 participant in the medium-dose arm with detectable DENV RNA (<LLOQ) at a single-timepoint (Figure S2 and Table S3).
A DENV-associated rash, defined as rash in combination with detectable DENV-3 RNA, was reported in 100% (7/7) of the participants in the placebo arm versus 83% (5/6), 50% (3/6), and 30% (3/10) of participants in low-, medium-, and high-dose arm, respectively (Figure S2). Most rashes were reported within 2 days after peak VL.
## Viral genome sequencing
Emergent amino acid variations in the NS4B region were detected in each of the 14 participants with available NS4B sequencing data in the mosnodenvir dose arms, while none were observed in the placebo arm. The most frequent emergent NS4B variations were V91A (N=9), V91G (N=2), P104L (N=2), T215S (N=3), and A233P (N=4). All those emerging NS4B variant frequencies were consistently >99% in the high-dose arm, with lower frequencies (15% to >99%) observed in the low/mid-dose arms. Emergent variations outside the NS4B region, which occurred in ≥2 participants, were A20T in the 2K region (Table S4).
## Safety
All participants reported at least one AE starting from first study drug intake until study termination (Table 2). Most AEs were mild (grade 1) to moderate (grade 2), occurred with similar frequency across all dosing arms and resolved without complications. There were 2 mosnodenvir-dosed participants with severe AEs, a participant in the medium-dose mosnodenvir arm with severe increase of lipase and glycemia and a participant in the highdose mosnodenvir arm with severe COVID-19 infection. These severe AEs were reported during follow-up. One participant withdrew consent after receiving 4 days of mosnodenvir and before inoculation, due to moderate photosensitivity, considered related to mosnodenvir. None of the mosnodenvir dosed participants had serious AEs, and none died. All out-ofrange laboratory findings were isolated and fully reversible.
## Pharmacokinetics
Mosnodenvir plasma concentrations rapidly increased with the LD from D-5 to D1 and were maintained up to D21, after which concentrations declined slowly consistent with a long t 1/2 (7.5-11 days) observed in this study (Table S5) and 6.3-9.2 days in the first-in-human study 21 (Figure 3). Individual maintenance concentrations were well separated between the different dose regimens. On D-5 and on D21, median t max was approximately 8 hours postdose with individual t max ranging from 2.02 to 12.10 hours. All pharmacokinetic parameters (e.g., C max and AUC 24h ) on D-5 and D21 are summarized in Figure 3 and Table S5.
## Discussion
Mosnodenvir demonstrated dose-dependent antiviral activity against rDEN3Δ30 in healthy participants establishing inhibition of NS3-4B interaction as a viable target. While CHIMs have been successfully employed to study dengue vaccine candidates, [24][25][26][27][28] we now extend this model to prevention of DENV infection in humans with a dengue antiviral. Consistent with earlier studies in this CHIM [24][25][26][27][28] , the infection rate in our study was 100% in placebo participants with uniform viral profiles, immune responses, and mild DENV-associated symptoms. The primary objective of this study was met, Tobit ANOVA analysis showed a statistically significant decrease in the log 10 AUC D1-D29 [VL] in the high-dose mosnodenvir arm when compared to placebo. Furthermore, mosnodenvir administered in low, medium and high dose prevented DENV-3 infection (undetectable DENV RNA, no anti-DENV IgM/IgG/nAb seroconversion and no DENV-associated rash) in a dose-dependent manner, demonstrating that mosnodenvir can reduce the incidence of DENV-3 infection and associated symptomatology, prophylactically in a dengue-naïve population. Further evaluation of the concentration-effect relationship is ongoing, including viral kinetic modeling. 29 The pharmacokinetic results of mosnodenvir were consistent with those observed in the first-in-human study. 21 Five days of daily loading doses allowed mosnodenvir plasma concentrations to reach steady-state levels at the time of DENV-3 inoculation.
Emerging variations in NS4B were detected in all (14/14) participants with NS4B sequencing data available in the mosnodenvir arms, as compared to none (0/7) of the placebo participants. The associated individual risks based on these limited data are considered minor as the peak and duration of the DENV-3 RNA levels were comparable or lower than those in the placebo arm. In addition, the frequency and severity of associated symptoms were not increased. The variations that emerged under mosnodenvir dosing were consistent with preclinical findings, pointing to NS4B as the target for mosnodenvir. Although certain variations in NS4B detected in this study have been noted in local outbreaks over time, 30 widespread occurrence among contemporary virus strains has not been described. 31,32 An increasing diversity of circulating DENV strains has been observed, exemplified by the co-circulation of the 4 DENV serotypes during the recent dengue outbreaks in the Americas. [33][34][35][36] Future field studies might provide insights in the emergence of variations in contemporary virus strains, considering this high diversity and the natural patterns of dengue seasonality. 37,38 The conditions of a CHIM are carefully controlled, using a well-characterized under-attenuated DENV-3 strain with consistent DENV RNA/ viremia profiles and an acceptable safety profile. Those studies are therefore well positioned to assess novel DENV interventions in early clinical studies, also given the relatively small sample size needed. Findings from those studies must be confirmed in larger safety and efficacy studies in the target population since CHIM studies do not reflect field conditions. Taken together, the favorable pre-clinical profile (pan-serotype activity in vitro and prophylactic antiviral activity against RNAemia in mice/NHPs 19 ) together with the data from this CHIM study support further development of mosnodenvir. In a phase 2 clinical field study (NCT05201794), it will be further explored whether prophylactic administration of mosnodenvir has an impact on the number of (symptomatic) DENV infections in endemic regions.
In this initial analysis, we show that mosnodenvir can prevent DENV-3 infection and associated symptoms in a dose-dependent manner in a controlled human infection setting. The advancement to Cohort 2 of the CHIM study will allow us to further characterize the relationship between mosnodenvir pharmacokinetics and antiviral activity for daily and weekly maintenance dosing. These results will complement the findings of the prophylaxis phase 2 clinical field study (NCT05201794). *Group 1 was split into a sentinel group (Group 1a) with N=4 (2:2 randomization between placebo and 600 mg LD/200 mg MD mosnodenvir) who were enrolled first and the remaining participants (Group 1b) (4:8 randomization between placebo and 600 mg QD LD/200 mg QD MD mosnodenvir). †One participant in the mosnodenvir arm withdrew consent on D-2 (before because of moderate photosensitivity (considered related to study drug by the investigator) and was replaced according to protocol. DENV, dengue virus; LD, loading dose; MD, maintenance dose; PK, pharmacokinetics; QD, daily; rDEN3Δ30, under-attenuated DENV-3 virus. The time to first detectable DENV-3 RNA was defined as the number of days between the date of inoculation and the date of the first occurrence of the event + 1. If a detectable event was not measured, the participant was censored at the last available sample with a valid result. Shaded area indicates 95% confidence intervals.
## References
1. Wilder-Smith, Ooi, Horstick et al. (2019) *Lancet*
2. (2023) "Dengue and severe dengue"
3. Messina, Brady, Golding (2019) "The current and future global distribution and population at risk of dengue" *Nat Microbiol*
4. (2023) "Dengue emergency in the Americas: time for a new continental eradication plan" *Lancet Reg Health Am*
5. (2024) "Worsening spread of mosquito-borne disease outbreaks in EU/EEA, according to latest ECDC figures"
6. (2024) "European Centre for Disease Prenvention and Control. Mosquito-borne diseases. An increasing risk in Europe"
7. Urmi, Mosharrafa, Hossain et al. (1598) "Frequent outbreaks of dengue fever in South Asian countries-A correspondence analyzing causative factors and ways to avert" *Health Sci Rep*
8. Wang, Urbina, Chang (2020) "Dengue hemorrhagic fever -A systemic literature review of current perspectives on pathogenesis, prevention and control" *J Microbiol Immunol Infect*
9. Jaenisch, Tam, Kieu (2016) "Clinical evaluation of dengue and identification of risk factors for severe disease: protocol for a multicentre study in 8 countries" *BMC Infect Dis*
10. (2009) "Dengue guidelines for diagnosis, treatment, prevention and control: new edition"
11. Malik, Ahsan, Mumtaz et al. (2023) "Tracing down the Updates on Dengue Virus-Molecular Biology, Antivirals, and Vaccine Strategies" *Vaccines (Basel)*
12. Whitehead (2016) "Development of TV003/TV005, a single dose, highly immunogenic live attenuated dengue vaccine; what makes this vaccine different from the Sanofi-Pasteur CYD vaccine" *Expert Rev Vaccines*
13. Torres-Flores, Sandoval, Salazar (2022) "Dengue Vaccines: An Update" *Biodrugs*
14. Kallas, Cintra, Moreira (2024) "Live, Attenuated, Tetravalent Butantan-Dengue Vaccine in Children and Adults" *N Engl J Med*
15. Pollard, Bijker (2021) "A guide to vaccinology: from basic principles to new developments" *Nat Rev Immunol*
16. Monto (2006) "Vaccines and antiviral drugs in pandemic preparedness" *Emerg Infect Dis*
17. Pardi, Weissman (2020) "Development of vaccines and antivirals for combating viral pandemics" *Nat Biomed Eng*
18. Goethals, Voge, Kesteleyn (2023) "A pan-serotype antiviral to prevent and treat dengue: A journey from discovery to clinical development driven by public-private partnerships" *Antiviral Res*
19. Goethals, Kaptein, Kesteleyn (2023) "Blocking NS3-NS4B interaction inhibits dengue virus in non-human primates" *Nature*
20. Kiemel, Kroell, Denolly (2024) "Pan-serotype dengue virus inhibitor JNJ-A07 targets NS4A-2K-NS4B interaction with NS2B/NS3 and blocks replication organelle formation" *Nat Commun*
21. Ackaert, Vanhoutte, Verpoorten (2023) "Safety, Tolerability, and Pharmacokinetics of JNJ-1802, a Pan-serotype Dengue Direct Antiviral Small Molecule, in a Phase 1, Double-Blind, Randomized, Dose-Escalation Study in Healthy Volunteers" *Clin Infect Dis*
22. Blaney, Hanson, Firestone et al. (2004) "Genetically modified, live attenuated dengue virus type 3 vaccine candidates" *Am J Trop Med Hyg*
23. Kakuda, Harasym, Buelens (2025) "Pharmacokinetics, safety, and tolerability of different maintenance dose regimens of mosnodenvir (JNJ-1802) in healthy adult participants" *Clin Pharmacol Drug Develop*
24. Kirkpatrick, Whitehead, Pierce (2016) "The live attenuated dengue vaccine TV003 elicits complete protection against dengue in a human challenge model" *Sci Transl Med*
25. Larsen, Whitehead, Durbin (2015) "Dengue human infection models to advance dengue vaccine development" *Vaccine*
26. Nivarthi, Swanstrom, Delacruz (2021) "A tetravalent live attenuated dengue virus vaccine stimulates balanced immunity to multiple serotypes in humans" *Nat Commun*
27. Pierce, Whitehead, Diehl "Evaluation of a new dengue 3 controlled human infection model for use in the evaluation of candidate dengue vaccines"
28. Pierce, Durbin, Walsh (2024) "TV005 dengue vaccine protects against dengue serotypes 2 and 3 in two controlled human infection studies" *J Clin Invest*
29. Goeyvaerts, De Trixhe, Neyens (2025) "Viral kinetic modeling of mosnodenvir prophylaxis against DENV-3 in a controlled human infection model. Poster presentation"
30. Bouzidi, Sen, Piorkowski "Genomic surveillance reveals that the dengue 2 virus lineage responsible for the 2023-2024 epidemic in the French Caribbean Islands is resistant to Mosnodenvir"
31. Olson, Assaf, Brettin (2023) "Introducing the Bacterial and Viral Bioinformatics Resource Center (BV-BRC): a resource combining PATRIC, IRD and ViPR" *Nucleic Acids Res*
32. "The Bacterial and Viral Bioinformatics Resource Center (BV-BRC)"
33. Van Der Ende, Nipaz, Carrazco-Montalvo et al. (2025) "Cocirculation of 4 Dengue Virus Serotypes" *Emerg Infect Dis*
34. Ariyaratne, Senadheera, Kuruppu (2025) "Simultaneous Cocirculation of 2 Genotypes of Dengue Virus Serotype 3 Causing a Large Outbreak in Sri Lanka in 2023" *J Infect Dis*
35. Chen-Germán, Araúz, Aguilar (2024) "Detection of dengue virus serotype 4 in Panama after 23 years without circulation" *Front Cell Infect Microbiol*
36. Kafetzopoulou, Torres-Hernández, Murillo-Ortiz (2025) "Assembling a comprehensive dataset to conduct an in-depth genomic investigation of the 2023-2024 dengue virus case surge in Valle del Cauca, Colombia. PanDengue conference"
37. Finch, Kucharski, Sim et al. "Climate variation and serotype competition drive dengue outbreak dynamics in Singapore"
38. Morin, Sellers, Ebi (2022) "Seasonal variations in dengue virus transmission suitability in the Americas" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12511481&blobtype=pdf | # Temporal Dynamics of SARS-CoV-2 Detection in Household Contacts: Divergences Between Time to First Positive Test, Symptom Onset, and Maximum Viral Load
Annemarie Berger, Ana Groh, Damian Diaz, Jesse Canchola, Tuna Toptan, Sandra Ciesek, Daniel Jarem, Alison Kuchta, Priscilla Moonsamy, Maria Vehreschild
## Abstract
Introduction:Understanding the temporal dynamics of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is crucial for optimizing diagnostic strategies. This prospective cohort study aimed to quantify the temporal viral transmission dynamics and biomarker profiles among households containing a SARS-CoV-2-positive index patient (IP) and uninfected household contacts (HHCs). Methods: IPs entered the study within 48 h after confirmation of SARS-CoV-2 through
reverse transcription polymerase chain reaction (RT-PCR). During 10-13 follow-up visits at days 0-7, and every 3-4 days thereafter until day 30 (± 6 days), nasopharyngeal swab and saliva samples were collected from participants (IP and HHC), and quantified via RT-PCR. Viral loads were estimated from cycle threshold values using three independently validated reference curved. Temporal viral dynamics for HHCs were evaluated as median times to first positive test (T f+ ), symptom onset (T so ), and peak viral load (T pvl ), using a within-host target cell-limited framework. Results: We prospectively screened 30 households with SARS-CoV-2-negative index cases; nine had a subsequent index-HHC conversion to PCR-positive, and 89 samples were generated. The results revealed a median T f+ of 2 days, T so of 4 days, and T pvl of 5 days, which underscores significant gaps between viral detection and peak viral load. Nasal samples exhibited higher viral replication rates (β = 0.77/day) and prolonged virus production as compared to saliva samples, while infected cells in saliva cleared more rapidly (δ = 0.65 day -1 vs 0.25 day -1 ). Conclusion: These findings suggest that SARS-CoV-2 viral RNA is detectable before symptom onset, and emphasize the need for testing immediately after exposure with repeated testing in the first week. This study provides critical insights into the temporal interplay of viral kinetics, aiding the development of targeted diagnostic and public health interventions. Further research is needed to validate these findings across larger, diverse cohorts and evolving viral variants. Testing immediately after exposure, with repeat testing during the first week may improve case detection.
The temporal dynamics of SARS-CoV-2 detection, including the time to first positive test, symptom onset, and peak viral load, remain insufficiently characterized. This study explores these critical time points to improve understanding of infection control among household contacts of individuals testing positive for SARS-CoV-2, utilizing nasal and saliva samples measured by RT-PCR.
## What was learned from the study?
In household contacts, the median time to the first positive SARS-CoV-2 test was 2 days, while symptom onset occurred at a median of 4 days, suggesting that SARS-CoV-2 is detectable prior to symptom onset. Furthermore, the choice of sample site is crucial for accurate diagnostics, as nasal samples demonstrated a higher viral replication rate, and a prolonged viral production compared to saliva samples.
## INTRODUCTION
The coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has highlighted the importance of timely and accurate diagnostics. While reverse transcription polymerase chain reaction (RT-PCR) remains the gold standard for viral detection, the relationship between time to first positive test, time to symptom onset, and time to peak viral load is not well understood. However, its application in early detection is limited by the timing of sample collection and the variable dynamics of viral replication. These temporal markers are crucial for developing targeted interventions, especially in household transmission scenarios where close contact heightens transmission risk. This study explores the interplay of these key time points to inform diagnostic timing and infection control effort. Groh et al. [1] investigated the effectiveness of various testing methods including RNA, antigen, and antibody tests for detecting SARS-CoV-2 infections and their correlation with infectivity and disease.
The study's results showed that saliva tests show higher sensitivity in the early stages of infection compared to nasopharyngeal swabs, particularly in asymptomatic and presymptomatic patients. Antigen tests showed a stronger correlation with culturable viruses, making them better suited for assessing potential infectivity. In contrast, RNA tests are generally more sensitive but provide a less precise evaluation of actual infectivity [2]. These findings have led to further discussion on how testing methods can be strategically employed in various clinical and epidemiological contexts.
To et al. [3] investigated the dynamics of viral load in saliva samples and the development of antibody responses throughout the course of a SARS-CoV-2 infection. Viral load was highest at the onset of illness and declined over time [3]. A longitudinal study [4] evaluated the sensitivity of RT-qPCR (quantitative RT-PCR) and antigen tests for detecting SARS-CoV-2 in 43 adults. Daily saliva and nasal swab samples were collected and tested for viral load. The study found that RT-qPCR tests demonstrated significantly higher sensitivity compared to antigen tests in the early phase of infection. Before the first positive viral culture, the sensitivity of the nasal RT-qPCR test was 0.650, while the antigen test achieved a sensitivity of only 0.375. During the phase of viral culture, all tests showed a similar sensitivity of greater than 98%, when performed at least every 3 days [4]. These findings emphasize the critical role of regular testing in the early identification of infected individuals for the containment of viral transmission.
Nevertheless, there is still a lack of more precise data on how diagnostic timing and choice of diagnostic modalities can be optimally aligned with virus spread and control measures. To further close this knowledge gap, this study focused on the interactions between critical points of infection and diagnostic testing procedures, in order to inform optimal diagnostic timing and infection control efforts.
## METHODS
This prospective study followed household contacts (HHC) of index patients (IP) with confirmed SARS-CoV-2 infection over 10 visits, with some follow-ups extending to 13 visits. More specifically, to assess viral transmission dynamics and biomarker profiles, a prospective cohort study was carried out with households containing SARS-CoV-2-positive IPs and uninfected HHCs [1]. IPs entered the study upon receiving their initial RT-PCR confirmed SARS-CoV-2 diagnosis at the acute care clinic at University Hospital Frankfurt. The study design included a planned, longitudinal testing protocol for all participants. Over a 30-day period, both the IPs and their HHCs were monitored through regular RT-PCR tests. Recruitment for IPs occurred within 48 h of their diagnostic RT-PCR, with testing performed daily on days 0-7 and every 3-4 days thereafter, until 30 days (± 6 days). In the full prospective study, which screened 30 households with multiple tests conducted per participant over an extended period, no participant received antiviral therapy during follow-up and a cohort of nine HHCs, generating 89 samples across multiple visits, was documented. Nasopharyngeal swab and saliva samples were collected from all participants using standardized protocols, with strict adherence to sample handling and storage guidelines, to preserve RNA integrity. Viral RNA testing was performed on both saliva and nasal samples using the Cobas ® SARS-CoV-2 Assay, for use on the Cobas 6800/8800 System (6800/8800; Roche Diagnostics), and nasal samples using the Cobas SARS-CoV-2 nucleic acid assay, for use on the Cobas Liat System (POC SARS-CoV-2, Roche Diagnostics). For the 6800/8800 test, SARS-CoV-2 was reported separately for the E gene and ORF1 gene targets, which were used to calculate the time to first positive test (T f+ ). Time to first symptoms or symptom onset (T so ) and time to peak viral load (T pvl ) were calculated for each HHC. A computer-assisted self-interview questionnaire was developed for this study and completed at each prescribed visit in order to gather information about the type and severity of up to 12 SARS-CoV-2-associated symptoms.
## Ethical Approval
Ethical approvals were obtained from Ethikkommission des Fachbereichs Medizin der Goethe-Universität c/o Universitätsklinikum (Reference number 2021-119-MPG). The study was performed in accordance with the Helsinki Declaration of 1964, and each participant provided informed consent.
## Estimation of Viral Load Using Cycle Threshold Values
At the time of study execution, SARS-CoV-2 PCR assays were semiquantitative, yielding only cycle threshold (Ct) values. To explore viral load dynamics, Ct values were converted to viral load estimates using three independently validated reference curves, spanning reaction efficiencies (RE) of 90-110%, to ensure robustness. The first reference curve represented the average of seven standard curves after the analysis according to Challenger et al. [5] with a RE of 93%. The second reference curve was derived from an earlier paper by Zou et al. [6] with a RE of 110%. The last reference curve was constructed with a RE of 100%, with the reference curve intercept calculated as the mean of the two intercepts [5,6]. For details of this process, see Supplementary Fig. 1. Viral loads in samples with a negative test result were imputed as zero wherever the qualitative POC SARS-CoV-2 assay was negative at that specific visit. As the results were similar, subsequent analyses used the 100% efficiency calculations for the viral load.
## Within-Host Viral Dynamics Modelling, Purpose, and CIL Derivation
The target cell-limited (TCL) framework was chosen because it parsimoniously links the observable viral load curve to four biological processes-entry (β), production (p), clearance of free virus (c), and loss of infected cells (δ)-without requiring additional latent compartments. We used the ordinary differential equation system published by Hernández-Vargas and Velasco-Hernández [7] and estimated parameters by non-linear mixed effects (Monolix 2024R1). The critical inhibition level (CIL), i.e., the fractional reduction in virus production needed to push the within-host basic reproduction number below 1, was calculated as where where U 0 is the initial target-cell pool (assumed to be 10 6 cells mL -1 ) [8]. This formulation allowed direct comparison of the suppressive pressure required in nasal versus saliva compartments. Statistical analyses were carried out using SAS ® v9.4 (The SAS Institute, Cary, NC, USA) and Monolix ® 2024R1 (Lixoft SAS, 2024) software.
## Key Parameters in the Model
## Viral Replication Rate (β)
This parameter represents the speed at which the virus replicates within the host. A higher β indicates faster viral replication, which could lead to a more rapid increase in viral load (units: per day).
(1)
$$CIL = 1 - 1 R 0 (2) R 0 = U 0 • p • β c • δ$$
## Infected Cell Clearance Rate (δ)
The infected cell clearance rate describes how quickly infected cells are removed from the body. A higher clearance rate indicates a more efficient immune response or cell turnover.
## Virus Release from Infected Cells (p)
This parameter refers to the rate at which the virus was released from infected cells into the surrounding tissues (units: virions cell -1 day -1 ).
## Virus Clearance Rate (c)
The virus clearance rate is the speed at which free virus particles are cleared from the body, either through immune mechanisms or natural decay.
## Mean Duration of Virus Production from Infected Cells (L)
This was calculated as the inverse of the infected cell clearance rate (L = 1/δ), representing the average time that a cell remains infected and continues to produce virus.
## Within-Host Reproduction Number (R 0 )
The within-host reproduction number (R 0 ) represents the number of new infected cells generated from a single infected cell. A higher R 0 means more infected cells are produced, leading to higher viral loads.
## Critical Inhibition Level
The CIL is the level of antiviral or immune suppression required to reduce the within-host reproduction number (R 0 ) to below 1, effectively driving the infection toward extinction within the host. The formula used here was CIL = 1 -(1/R 0 ). Given the similarities between the ORF1 gene and E gene targets, all further analyses focus on the results using the E gene target.
## RESULTS
The median number of days for key SARS-CoV-2 infection metrics among the nine positive household contacts (HHC) is summarized in Table 2. Across all tests, the median time to first positive detection (T f+ ) was 2 days from initial exposure, irrespective of the collection method (nasal or saliva). Symptom onset (T so ) occurred consistently at a median of 4 days across all methods. For peak viral load (T pvl ), the median time was 5 days for the SARS-CoV-2 E gene target using nasal samples and 4 days for the same target using saliva (Fig. 2, Supplementary Table 1). Peak viral load data were not available for the POC SARS-CoV-2 nasal ) 3 presents the results of the TCL model, which was applied to analyze the within-host viral dynamics of SARS-CoV-2 in infected HHC. This model provides insight into several key parameters of viral replication, including viral replication rate (β), infected cell clearance rate (δ), virus release from infected cells (p), and virus clearance rate (c). These parameters help describe the lifecycle of the virus within an infected individual and are crucial for understanding progression of the infections, as well as informing treatment and intervention strategies. A summary of each key parameter in the model and interpretation is provided.
$$CIL critical inhibition level = 1 -1/R 0 , HHC household contact, IP index patient a Reference curve with 100% efficiency b L = 1/δ c R 0 = U 0 •p•β/(c•δ); U 0 is$$
## Parameter Estimates (see Methods for Definitions)
## Viral Replication Rate (β)
In this study, the viral replication rate for nasal samples was 0.77 per day, whereas for saliva samples, it was lower at 0.42 per day. Assuming independent compartments, this suggests that viral replication occurs faster in cells near the nasal cavity than those near saliva, possibly as a result of differences in tissue type or viral entry efficiency in these sites. However, the reader should note that differences in viral dynamics across specimen types may be due to transport mechanisms or sampling biases not incorporated in our TCL model.
## Infected Cell Clearance Rate (δ)
For nasal samples, the clearance rate was 0.25 per day, while saliva samples had a higher clearance rate of 0.65 per day. This suggests that the body may clear infected cells from the saliva compartment more quickly than from the nasal compartment.
## Virus Release from Infected Cells (p)
In nasal samples, the rate of virus release was 0.42 per day, compared to 0.68 per day for saliva samples. This higher rate of viral release in saliva may indicate that, once infected, the cells in the saliva environment release more virus compared to nasal cells.
## Virus Clearance Rate (c)
The clearance rate for nasal samples was 0.23 per day, while it was lower for saliva at 0.17 per day. This indicates that the body clears free virus particles more quickly from the nasal environment, compared to the saliva environment, potentially as a result of the presence of different immune factors or physical barriers.
## Mean Duration of Virus Production from Infected Cells (L)
For nasal samples, the mean duration of virus production was 4.00 days, while for saliva, it was significantly shorter at 1.54 days, further highlighting the faster turnover of infected cells in saliva.
## Within-Host Reproduction Number (R 0 )
The R 0 for nasal samples was 33.75, much higher than the 15.51 observed for saliva samples. This indicates that the nasal environment supports substantially higher viral replication than saliva, highlighting the greater replication efficiency of SARS-CoV-2 in nasal secretions.
## Critical Inhibition Level
For nasal samples, the CIL was 0.97, and for saliva samples, it was slightly lower at 0.94. This indicates that both nasal and saliva environments require a similar, very high level of intervention (e.g., through antiviral drugs or immune responses) to halt viral replication.
## Nasal Versus Saliva SARS-CoV-2 PCR Tests (E Gene target)
The key takeaway from Table 3 is the distinction in viral dynamics between nasal and saliva samples.
## Nasal Samples
Nasal samples exhibited faster viral replication (higher β) and a longer duration of virus production from infected cells (L), which contributes to a higher within-host reproduction number (R 0 ). This suggests that the nasal cavity is a more favorable environment for sustained viral replication, which may contribute to prolonged RNA detectability in nasal samples from nasal secretions.
## Saliva Samples
In contrast, saliva samples had a faster clearance of infected cells (higher δ) and a quicker turnover of virus production (lower L), but virus release from infected cells (higher p) occurred more rapidly. This could indicate that, while the virus does not replicate as aggressively in saliva as in nasal tissues, the saliva environment may still play a significant role in RNA detectability due to the quicker release of the virus from infected cells.
## DISCUSSION
The TCL model provides a comprehensive look at how SARS-CoV-2 behaves within different body environments. Nasal samples, with their higher replication rates and longer infection durations, suggest that the nasal cavity could be a key site for viral transmission. In contrast, saliva samples exhibit faster viral clearance and infected cell turnover, but higher viral release rates, which may still be epidemiologically relevant, underscoring the value of saliva as a complementary specimen, though infectivity was not measured directly. These findings highlight the importance of considering both nasal and saliva samples in diagnostic testing strategies, especially when assessing within-host kinetics and the timing of potential interventions.
Our study reveals a distinct temporal gap between initial viral detection (T f+ ) and symptom onset (T so ), showing that viral RNA is detectable before symptoms occur, although infectivity was not directly assessed. This underscores the importance of frequent testing in high-risk environments, as waiting for symptoms to appear may allow continued viral RNA production. Additionally, the timing of maximum viral load (T pvl ) post symptom onset suggests that individuals have detectable RNA even after initial symptom presentation.
These findings support the use of early testing protocols in exposed individuals to identify cases before peak viral load and enhance quarantine measures. Moreover, the data indicate that viral dynamics in asymptomatic individuals are similar to those in symptomatic cases, which has implications for transmission risk assessment and public health guidelines.
## LIMITATIONS
While our study offers valuable insights into the temporal differences between time to first positive (T f+ ), time to symptoms onset (T so ), and time to peak viral load (T pvl ), several limitations must be acknowledged.
## Small Sample Size for Quantitative Outcomes
Although 30 households were screened, only nine IPs were positive; the quantitative viral load analyses were derived from this limited cohort, generating 89 samples across multiple visits, with multiple tests conducted per participant over an extended period. This study was conducted at participants' homes during the peak of the COVID-19 pandemic, which introduced substantial logistical challenges relating to recruitment and sample collection; this meant it was not feasible to include additional participants to increase the sample size. The small sample size restricts the statistical power and generalizability of our findings, especially when applied to broader populations or different viral variants.
## Potential Impact of Vaccinations on Viral Dynamics
This study did not specifically account for the vaccination status of participants, which may have affected viral load and timing dynamics. Vaccinated individuals could exhibit altered viral replication patterns, potentially impacting the observed intervals between the first positive test, symptom onset, and peak viral load. The influence of vaccination on viral kinetics warrants further investigation to accurately interpret these findings across both vaccinated and unvaccinated populations. However, 59 of the 60 participants (96.7%) were vaccinated against SARS-CoV-2. Only two participants (one IP and one uninfected HHC) were unvaccinated, meaning that the viral kinetics in this population could not be evaluated.
## Lack of Official External Viral Load Standard
Viral load values were quantified using reference curves derived from published studies rather than a universally validated external standard. The accuracy of these viral load measurements may vary across different testing platforms, and comparisons to other studies may be affected by discrepancies in quantitation methods.
## Single Study Cohort and Regional Bias
The study was conducted within a single prospective household cohort in Germany. Therefore, observed relationships between viral load, symptom onset and diagnostic testing may not be fully generalizable to other populations or geographic regions. Replication in more diverse cohorts is necessary to strengthen these findings. The anonymized symptom questionnaire is also available upon reasonable request to the corresponding author (as outlined in Data Availability).
## Circulating SARS-CoV-2 Variants
Our study reflects the viral dynamics of SARS-CoV-2 variants circulating in Germany during the study period. Given the evolving nature of SARS-CoV-2, including the emergence of new variants with potentially different viral kinetics, the applicability of our results to current or future variants may be limited. While sequencing was not performed in this study, public health surveillance data indicated more than 90% B.1.617.2 (Delta) circulation during enrolment.
## Subjectivity in Self-Reported Symptoms
Symptom data were self-reported by participants, introducing potential subjectivity and reporting bias. Symptoms may vary in intensity or be underreported, which could influence the associations between symptom onset and viral load. Objective clinical assessments could provide a more reliable measure of symptom severity in future studies.
## Symptoms Captured at Pre-specified Time Points
Symptom assessments were recorded only at predetermined study visits, rather than continuously throughout the study period. This could result in missing variations in symptom severity between visits, potentially underestimating the true symptom onset timing relative to viral load changes.
## Limited Relevance to Other Respiratory Infections
This study focuses on COVID-19 symptoms and viral load dynamics and should not be generalized to distinguish between SARS-CoV-2, influenza, or other respiratory infections. Most participants were not treated with antivirals, which might have altered the natural course of symptom progression and viral load dynamics.
## Validation of Viral Load and Symptom Relationship
While our study identified associations between viral load and symptom onset, these findings have not yet been validated in independent cohorts. Future research should aim to replicate these relationships in larger and more diverse populations to ensure the robustness and clinical applicability of the observed patterns.
## TCL Model Assumptions
We assumed that viral replication occurs in distinct compartments, and our modelling does not reflect possible transport dynamics from a common replication site. Therefore, the differences in viral load dynamics observed between nasal and saliva samples may reflect transport mechanisms from a common replication site, rather than distinct compartment-specific replication. This latter interpretation aligns with prior evidence suggesting that respiratory epithelial cells are the primary site of SARS-CoV-2 replication [3,6,9,10]. In fact, To et al. [3] noted high viral loads in saliva but attributed these findings to viral transport from the respiratory tract rather than replication in salivary glands.
Despite these limitations, the strengths of this study include its prospective design, the use of comprehensive viral load quantitation methods, and the detailed temporal analysis of key SARS-CoV-2 infection biomarkers. The findings provide important insights into the timing of viral positivity, symptom onset, and peak viral load, warranting further investigation into larger studies with standardized VL quantitation methods and broader cohort representation.
## CONCLUSION
The interval between first RNA detection, symptom onset, and peak viral load is pivotal for refining diagnostic strategies and interpreting transmission risk. Our findings highlight the value of initiating testing immediately after exposure and repeating it during the first week-especially in household settings where undetected, asymptomatic cases can silently propagate infection. Future research should focus on validating these temporal dynamics in larger, more diverse cohorts, while integrating vaccination effects and evolving SARS-CoV-2 variant characteristics.
## Medical Writing, Editorial and Other Assistance. Editorial support for this manuscript was provided by Holly McAlister and Fraser
Harris of Springer Health+, London, UK and was funded by Roche Molecular Systems, Pleasanton, CA, USA.
Author Contributions. All authors (Annemarie Berger, Ana M. Groh, Damian Diaz, Jesse A. Canchola, Tuna Toptan, Sandra Ciesek, Daniel Jarem, Alison L. Kuchta, Priscilla Moonsamy and Maria J.G.T. Vehreschild) contributed to the conception and design of the study, or acquisition of data, or analysis and interpretation of data. All authors contributed to drafting the article or revising it critically for important intellectual content, and approved the final version for submission.
Funding. This study and the journal's Rapid Service Fee were funded by Roche Molecular Systems, Pleasanton, CA, USA.
Data Availability. The study was conducted in accordance with applicable regulations. For more information on the study and data sharing, qualified researchers may contact the corresponding author, Dr Annemarie Berger, at annemarie.berger@em.uni-frankfurt.de. Open Access. This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by-nc/4. 0/.
## Declarations
## References
1. Groh, Vehreschild, Diaz (2024) "Kinetics of SARS-CoV-2 infection biomarkers in a household transmission study" *Sci Rep*
2. Perveen, Negi, Gopalakrishnan (2023) "COVID-19 diagnostics: molecular biology to nanomaterials" *Clin Chim Acta*
3. To, Tsang, Leung (2020) "Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study" *Lancet Infect Dis*
4. Smith, Gibson, Martinez (2021) "Longitudinal assessment of diagnostic test performance over the course of acute SARS-CoV-2 infection" *J Infect Dis*
5. Challenger, Foo, Wu "Modelling upper respiratory viral load dynamics of SARS-CoV-2"
6. (2022) *BMC Med*
7. Zou, Ruan, Huang (2020) "SARS-CoV-2 Viral load in upper respiratory specimens of infected patients" *N Engl J Med*
8. Hernandez-Vargas, Velasco-Hernandez (2020) "In-host mathematical modelling of COVID-19 in humans" *Annu Rev Control*
9. Gonçalves, Bertrand, Ke (2020) "Timing of antiviral treatment initiation is critical to reduce SARS-CoV-2 viral load" *CPT Pharmacometr Syst Pharmacol*
10. Hui, Azhar, Madani (2020) "The continuing 2019-nCoV epidemic threat of novel coronaviruses to global health-the latest 2019 novel coronavirus outbreak in Wuhan, China" *Int J Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12646732&blobtype=pdf | # Distribution and Clinical Profile of Human Parainfluenza Viruses in Hospitalized Patients With Acute Febrile Illness
Santhosha Devadiga, Nachiket Godbole, Prasad Varamballi, Chiranjay Mukhopadhyay, Anup Jayaram
## Abstract
Introduction: Human parainfluenza viruses (HPIVs) are significant causes of respiratory infections, particularly in children, yet their epidemiology remains poorly understood in low-and middle-income countries. HPIVs contribute to 20%-40% of pediatric lower respiratory tract infections (LRTIs) and are a leading cause of croup and hospitalizations. This study was aimed at determining the incidence, distribution, and clinical and laboratory characteristics of HPIV in hospitalized acute febrile illness (AFI) patients. Methods: A total of 12,409 AFI cases from 2016 to 2018 were tested for HPIVs via molecular methods. RNA was extracted from throat swab samples and tested via multiplex real-time RT-PCR for HPIV Serotypes 1-4. The demographic, clinical, and laboratory data of HPIV-positive patients were analyzed statistically. Results: HPIVs were detected in 217 (1.75%) patients, with HPIV-3 (49.77%) being the most prevalent, followed by HPIV-4 (18.90%), HPIV-2 (17.52%), and HPIV-1 (13.83%). HPIV-3 exhibited distinct seasonal peaks, mainly affecting children (1-9 years). Significant variations in hematological and biochemical markers were observed among serotypes and age groups. Upper and lower respiratory symptoms, along with gastrointestinal issues and systemic manifestations such as chills, myalgia, and weakness, are commonly reported. Conclusion: HPIVs contribute to respiratory illness across diverse demographics. HPIV-3 is the predominant serotype, with distinct seasonal and age-related patterns. Improved surveillance and diagnostics could aid in better management and reduce unnecessary antibiotic use.
## 1. Introduction
Human parainfluenza viruses (HPIVs) are enveloped singlestranded negative-sense RNA viruses, and there are four major serotypes (HPIVs 1, 2, 3, and 4) (1). HPIVs 1 and 3 belong to the genus Respirovirus, and HPIVs 2 and 4 are members of the genus Orthorubulavirus of the family Paramyxoviridae (2). A previous study on HPIV among children reported that 20%-40% of children were hospitalized with lower respiratory tract infection (LRTI) (3). HPIVs are the second most common cause of hospitalization in children and a common cause of croup. Infection may lead to lifethreatening LRTIs, including bronchitis, bronchiolitis, and pneumonia. Reinfection is typically associated with milder disease limited to the upper respiratory tract (4).
The burden of HPIV infection among children and adults in low-and middle-income countries remains poorly understood, despite these regions being home to much of the global population. Several factors contribute to this data deficiency, including inadequate testing due to expensive molecular diagnostic assays, the ability to record HPIVrelated morbidity and mortality, and the unavailability of specific treatments, which are major setbacks for HPIV diagnosis. Comprehensive data on HPIV infections could help physicians reduce unnecessary antibiotic prescriptions. HPIV infections occur worldwide, with varying seasonal patterns and circulation serotypes in different geographic regions. In India, data on the distribution and seasonality of HPIV are limited. Additionally, there are insufficient data on HPIV-associated hospitalizations for acute respiratory illness (ARI). This study is aimed at understanding the incidence of HPIVs in hospitalized acute febrile illness (AFI) patients and understanding the demographic and clinicolaboratory characteristics of HPIVs among children and adults.
## 2. Materials and Methods
## 2.1. Study Population.
A total of 217 HPIV-positive cases were included in this study, and the cases represented 10 states of India: Karnataka, Kerala, Assam, Goa, Maharashtra, Jharkhand, Tripura, Tamil Nadu, and Odisha. Samples were collected as part of the AFI surveillance study conducted by the Manipal Institute of Virology (MIV), with a case definition of patients admitted to the hospital with fever (≥ 38 °C) aged between 1 and 65 years (5). All the inpatients admitted to the hospitals fulfilling the inclusion criteria of the case definition were included in the surveillance study after providing informed consent. In the case of children aged 13-17 along with informed consent, ascent was also taken. The total number of AFI samples that were obtained annually from January 2016 to September 2018 is as follows: 2016-12,414, 2017-20,359, and 2018-4954. A total of 12,409 patients presented with AFI with ARI and were tested for HPIVs (2016-3641, 2017-6528, and 2018-1635, respectively). All 12,409 samples were tested for various pathogens, including viruses and bacterial agents causing AFI with ARI at MIV via molecular methods. Briefly, RNA was extracted from throat swab samples via a QIAamp Viral RNA Mini Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. Multiplex real-time RT-PCR was performed with primers and probes targeting HPIV Serotypes 1, 2, 3, and 4 via a Respiratory Pathogens 21 Kit (Fast Track Diagnostics [FTD], Luxembourg). RT-PCR was performed via an AgPath-ID Onestep RT-PCR Kit (Applied Biosystems, Foster City, United States). The reaction was performed via a QuantStudio 5 PCR system (Applied Biosystems, Foster City, United States). A cycle threshold (CT) value of < 35 was considered positive for HPIV. Demographic, clinical, and laboratory data of HPIV-positive patients were obtained from the AFI study database after ethical clearance was obtained for analysis.
## 2.2. Data Analysis.
We assessed the clinical, laboratory, and demographic data of HPIV-positive patients. For the analysis of continuous variables in two groups, a two-tailed t test was used; for more than two groups, one-way ANOVA was used, and for the analysis of categorical variables, the chisquare test was used. A p value of < 0.05 was set as the level of statistical significance. Statistical significance was determined by one-way ANOVA followed by Tukey's multiple comparisons test. To compare the data, they were analyzed via GraphPad Prism Version 8.4.2.
## 3. Results
Among the 12,409 patients tested, 217 (1.75%) were positive for HPIVs. Among the positive samples, 29 (13.37%) tested positive for HPIV-1, 39 (17.98%) for HPIV-2, 108 (49.77%) for HPIV-3, and 41 (18.90%) for HPIV-4. The annual incidence of HPIVs varied and was recorded at 1.38% in 2016, 1.61% in 2017, and 3.80% in 2018. The analysis of HPIV seasonality revealed that HPIV-3 presented the greatest number of cases with distinct seasonal peaks, particularly between the late winter and summer seasons in India. HPIV-1, HPIV-2, and HPIV-4 demonstrate sporadic occurrences with lowercase numbers but exhibit seasonal variation (Figure 1a). The age-group distribution indicates that HPIVs predominantly affect younger age groups, particularly children aged 1-9 years, with a relatively balanced sex distribution. HPIV-3 and HPIV-4 cases are more evenly spread across age groups, and HPIV-4 cases are more common among younger individuals, with males being slightly more affected. The incidence of HPIV-2 is greater in adolescents. Overall, HPIV-3 is the most prevalent and seasonally distinct, whereas other types follow less predictable patterns and affect different age groups (Figures 1b, 1c, 1d, and1e). The samples were represented from 10 states of India, and the percentage positivity and total tested from each state are provided in Table S1.
Among the study participants, upper respiratory tract infection (URTI) symptoms were prevalent, with coryza being the most common (58.9%-80.5%) and sore throat being observed in approximately half of the cases (48.3%-58.5%). LRTI symptoms are frequently reported, with cough being the most common symptom (82.7%-92.7%), followed by shortness of breath (17.2%-31.7%) and chest pain (7.3%-25%). The prevalence of gastrointestinal symptoms varied, with nausea (24.4%-48.3%), vomiting (24.4%-43.5%), abdominal pain (14.8%-43.5%), and diarrhea (1.8%-7.6%) reported in different proportions. Other systemic symptoms, including chills (69.4%-79.3%), myalgia (65.5%-82.9%), arthralgia (44.8%-56.4%), headache (65.5%-82.1%), and general weakness, which was highly prevalent across all serotypes (86.2%-92.6%), were also significant. Conjunctivitis congestion (6.8%-15.3%) and lymphadenopathy (17.6%-27.5%) were less frequently observed (Table 1).
The findings of this study reveal variations in hematological, biochemical, and inflammatory markers among HPIVs. Leukopenia (< 4000 cells/mm 3 ) was present in 3.5%-23.5% of the participants, whereas thrombocytopenia (<150 × 10 3 /μL) was detected in 7.2%-26.4%. The prevalence of neutrophilia (> 70%) ranged from 24.4% to 32.1%, whereas the prevalence of lymphocytosis (> 40%) varied between 11.9% and 30.3%. Liver enzyme abnormalities were notable, with elevated aspartate aminotransferase (AST) (> 40 IU/L) in 39.1%-64.7%, alanine aminotransferase (ALT) (> 40 IU/L) in 6.1%-30.4%, and alkaline phosphatase (ALP) (> 140 IU/L) in 81.8%-98.3%. The erythrocyte sedimentation rate (> 20 mm/h) increased in 20%-50% of the patients. Hypoproteinemia (< 6 g/dL) and hypoalbuminemia (< 3.5 g/dL) were uncommon, reported in 4.3%-33.3% and 0%-6.6% of patients, respectively. Elevated urea (> 40 mg/ 2). Comparative analysis of hematological and biochemical parameters between children and adults infected with different HPIV types revealed significant variations. Total leukocyte counts remained largely similar across age groups, whereas platelet counts were significantly lower in adults with HPIV-1. Liver enzyme abnormalities were evident, with AST levels being higher in children across all HPIV types. ALT levels were significantly greater in adults with HPIV- International Journal of Microbiology 3. ALP levels were notably elevated in children across all HPIV types. CT values, reflecting the viral load, were significantly lower in adults with HPIV-3 (Figure 3). Among HPIV cases, the coinfection rate was 3.22% (6), and coinfections with other respiratory viruses were one incidence for each of the following pairs: HPIV-1 and enterovirus, HPIV-2 and respiratory syncytial virus (RSV), HPIV-3 and coronavirus OC43, human metapneumovirus, HPIV-4 and influenza A (H1N1), and influenza B and Mycoplasma pneumoniae (Table S1).
## 4. Discussion
The overall incidence of HPIV in this study highlights the distinct seasonal patterns and demographic distributions of HPIVs, with HPIV-3 emerging as the most prevalent type, 5 International Journal of Microbiology affecting all age groups. This finding aligns with previous studies that reported HPIV-3 as the dominant strain, peaking in late winter and early summer (7,8). Our results revealed the notable presence of HPIV-1 and HPIV-2 across different age groups, which is consistent with studies indicating that HPIV-1 contributes primarily to biennial outbreaks in young children, whereas HPIV-2 occurs less frequently but has a broader age distribution (9). The sporadic nature of HPIV-4, which primarily affects younger individuals with a male predominance, is consistent with previous epidemiological data, which suggests that HPIV-4 is less commonly detected but may still contribute to respiratory infections (10).
The clinical presentation observed in our study emphasizes the predominance of respiratory symptoms across all HPIV types, with coryza and cough being the most frequently reported. This finding is also consistent with previous literature describing HPIVs as a major cause of upper and LRTIs (11). The gastrointestinal symptoms noted, particularly nausea, vomiting, and abdominal pain, were more pronounced and significant in certain HPIV types, a trend also observed in studies highlighting the extrarespiratory manifestations of viral infections (12). Systemic symptoms, including myalgia, headache, and weakness, were signifi-cantly reported, mirroring findings from other studies that suggest that a generalized inflammatory response contributes to the severity of illness (13). The frequency of conjunctivitis, congestion, and lymphadenopathy was relatively low.
Hematological and biochemical variations among HPIV types were notable, particularly in leukopenia, thrombocytopenia, and liver enzyme abnormalities. Elevated liver enzymes, particularly in HPIV-3 cases, indicate possible hepatic involvement, which has been sporadically reported in other respiratory viruses (14). The differences in CT values suggest variations in viral load, further emphasizing the distinct pathophysiological impacts of different HPIV types. Blood pressure variations, particularly the lower diastolic values observed in HPIV-2 patients, could indicate a differential impact on cardiovascular function, a phenomenon that warrants additional study.
The observed lower platelet counts in adults with HPIV-1 align with previous studies indicating that viral infections can induce transient thrombocytopenia due to immunemediated platelet depletion or bone marrow suppression (15). Elevated liver enzymes, particularly ALT and AST, in individuals infected with HPIV-3 suggest possible hepatic involvement, either through direct viral effects or immunemediated injury. In adults, significantly higher ALT levels A significant limitation of this study, and a broader challenge in India, is the lack of a centralized, long-term surveillance system for HPIV, which hinders our ability to establish seasonal patterns, identify high-risk populations, and formulate evidence-based public health interventions. Consequently, the development and implementation of targeted prevention strategies, such as public health campaigns and resource allocation for pediatric care during peak seasons, remain severely underdeveloped. HPIV patients are also coinfected with other respiratory viruses, which might have altered the clinical and laboratory findings. However, these numbers are very low. Data on clinical and laboratory parameters were obtained at the time of admission only. Overall, this study addresses the incidence, distribution, and clinical and laboratory characteristics of HPIV in both children and adults and causes significant morbidity in both children and adults, which requires hospitalization.
## 5. Conclusion
This multiyear surveillance study highlights the incidence of HPIVs in India. Among hospitalized patients with ARI, the overall positivity rate was 1.75%. HPIV-3 was the most prevalent type, showing distinct seasonal peaks from late winter to summer, whereas HPIV-1, HPIV-2, and HPIV-4 occurred sporadically. Children aged 1-9 years were the most affected group, although HPIVs were detected across all age groups. Hematological variations, including leukopenia, thrombocytopenia, and elevated liver enzymes, were common, with HPIV-3 showing a stronger association with hepatic involvement. Age-related differences were evident: children presented higher AST levels and viral loads, whereas adults presented significantly higher ALT levels. Overall, these findings reinforce HPIVs as important contributors to morbidity in both children and adults, frequently leading to hospitalization. This study underscores the need for routine testing, longitudinal monitoring, molecular characterization, and the development of targeted therapeutics to reduce the disease burden associated with HPIVs.
## References
1. Han, Suh, Han (2022) "Seasonal Epidemiological and Clinical Characteristics of Pediatric Patients With Human Parainfluenza Virus Infection by Serotype: A Retrospective Study" *Virology Journal*
2. Rima, Balkema-Buschmann, Dundon "ICTV Virus Taxonomy Profile: Paramyxoviridae 100"
3. Branche, Falsey (2016) "Parainfluenza Virus Infection" *Seminars in Respiratory and Critical Care Medicine*
4. *International Journal of Microbiology*
5. Russell, Ison (2017) "Parainfluenza Virus in the Hospitalized Adult" *Clinical Infectious Diseases*
6. Govindakarnavar (2017) "Annual Report of Hospital Based Surveillance of Acute Febrile Illness in India"
7. Swamy, Malhotra, Reddy et al. (2016) "Distribution and Trends of Human Parainfluenza Viruses in Hospitalised Children" *Indian Journal of Pediatrics*
8. Counihan, Shay, Holman et al. (2001) "Human Parainfluenza Virus-Associated Hospitalizations Among Children Less Than Five Years of Age in the United States" *Pediatric Infectious Disease Journal*
9. Swamy, Malhotra, Reddy et al. (2016) "Distribution and Trends of Human Parainfluenza Viruses in Hospitalized Children" *Indian Journal of Pediatrics*
10. Weinberg, Hall, Iwane (2009) "Parainfluenza Virus Infection of Young Children: Estimates of the Population-Based Burden of Hospitalization" *Journal of Pediatrics*
11. Abedi, Prill, Langley (2016) "Estimates of Parainfluenza Virus-Associated Hospitalizations and Cost Among Children Aged Less Than 5 Years in the United States, 1998-2010" *Journal of the Pediatric Infectious Diseases Society*
12. Hall (2001) "Respiratory Syncytial Virus and Parainfluenza Virus" *New England journal of Medicine*
13. David, Knipe (2013) "Fields Virology"
14. Falsey, Formica, Walsh (2002) "Diagnosis of Respiratory Syncytial Virus Infection: Comparison of Reverse Transcription-PCR to Viral Culture and Serology in Adults With Respiratory Illness" *Journal of Clinical Microbiology*
15. Mahony, Petrich, Smieja (2011) "Molecular Diagnosis of Respiratory Virus Infections" *Critical Reviews in Clinical Laboratory Sciences*
16. Lee, Wu, Chen et al. (2011) "Acute Immune Thrombocytopenic Purpura in an Adolescent with 2009 Novel H1N1 Influenza A Virus Infection" *Journal of the Chinese Medical Association*
17. Peiris, Hui, Yen (2010) "Host Response to Influenza Virus: Protection Versus Immunopathology" *Current Opinion in Immunology*
18. Pai, Liu, Yen (2024) "Clinical Characteristics and Risk Factors for Severe Human Parainfluenza Virus Infection in Hospitalized Children" *Journal of Microbiology, Immunology and Infection* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12693687&blobtype=pdf | # Academic Editors: Maria
Michela Salvatore, Anna Andolfi, Wim Dehaen, Irina Smirnova, Alexander Lobov, Liana Zakirova, Dmitriy Polovyanenko, Irina Bagryanskaya, Vladimir Zarubaev, Oxana Kazakova
## Abstract
Bile acids provide a versatile platform for the design of biologically active compounds due to their amphiphilic structure, biocompatibility, and capacity for diverse chemical modifications. Among them, lithocholic acid is a promising scaffold for designing and revealing new antiviral agents. A novel lithocholic acid-based 3-spiro-1,2,4-trioxolane was synthesized by Griesbaum co-ozonolysis of methyl 3-O-methyl-oximino-lithocholate and 4-(trifluoromethyl)-cyclohexanone, and its structure was confirmed by 2D NMR and X-ray crystallographic analysis. Lithocholic acid derivatives were evaluated for cytotoxicity and anti-influenza activity against A/Puerto Rico/8/34 (H1N1), showing that steroid 1,2,4-trioxolane 3 exhibited the highest potency (IC 50 4.3 µM, SI 11) compared to the parent methyl-3-oxo-lithocholate 1 (IC 50 > 84 µM, SI 1). In silico ADME predictions revealed several favorable drug-like properties, including a highly three-dimensional structure (Fsp 3 = 0.97), significant lipophilicity (LogP = 7.54), and the presence of key pharmacophores such as a peroxide moiety and a trifluoromethyl group. Taken together, a stereospecific synthesis of a lithocholic acid 3-spiro-1,2,4-trioxolane by Griesbaum co-ozonolysis was realized and the first evidence of anti-influenza activity in the steroid-1,2,4-trioxolane series was found.
## 1. Introduction
The steroid system has been extensively used and developed as a privileged scaffold with diversified medicinal properties in the drug discovery process. A striking example of this application is a steroid polyamine squalamine that became the first representative of a previously unknown class of natural antibiotics of animal origin [1]. Among steroid molecules, bile acids have indeed emerged as highly useful scaffolds for the development of conjugates with various biologically active molecules. Such conjugates exhibit improved water solubility, increased metabolic stability, and, in some cases, selective cellular penetration and a broad spectrum of activities [2]. For example, the bile acid conjugates with amino acids or peptides which inhibited HIV-1 replication at a concentration as low as 0.02 µg/mL have analgesic and anticancer activities [3].
Lithocholic acid (LCA) is formed through the microbial metabolism of primary bile acids in the human intestine [4] and has recently gained renewed interest for immunomodulatory, anti-inflammatory, and anticancer effects [5]. Notably, emerging evidence indicates that LCA and its semisynthetic derivatives possess antiviral potential, offering a new avenue in the development of antiviral therapies. LCA is known to interact with several nuclear and membrane-bound receptors, such as the pregnane X receptor, vitamin D receptor, and G-protein-coupled bile acid receptor, which are involved in modulating host innate immunity and inflammatory responses [6]. This ability to influence host signaling pathways may play a critical role in restricting viral replication and enhancing antiviral defense. In vitro studies have demonstrated that LCA and chenodeoxycholic acid at physiological concentrations can suppress the replication of porcine delta-coronavirus, via the activation of GPCR signaling, leading to the upregulation of interferon lambda-3 and interferonstimulated gene 15 [7]. This approach aims to improve the characteristics of LCA such as bioavailability, lower toxicity, and more efficient drug delivery [8]. Lithocholic acid oleate has been shown to upregulate the expression of human beta-defensin 1 and cathelicidin LL-37 in human keratinocyte cells [9]. The molecular hybridization of dihydroartemisinin and ursodeoxycholic acid resulted in a significant reduction in the viral replication of SARS-CoV-2 [10]. Bile acids play a vital role in maintaining intestinal homeostasis and combating infections; for example, deoxycholic acid can restore interferon type I responses to restrict the Chikungunya virus [11]. A recent study showed that a combination of LCA and LCA-oleate provided preventive antiviral protection against herpes simplex virus type 1 in both in vitro and in vivo models [12]. We have shown that the 3-oxo-lithocholic acid Mannich base connected with N-methylpiperazine increased antiviral activity against the Flu A (H1N1) virus along with a two-fold reduction in toxicity in MDCK cells [13].
On the other hand, there is a large group of natural and semisynthetic compounds with peroxide units that exhibit a variety of biological activity, including anthelmintic, antiprotozoal, fungicidal, antiviral, and others [14,15]. The most known natural peroxide is artemisinin with high antimalarial activity, the discovery of which by Chinese pharmaceutical chemist Tu Youyou was awarded the Nobel Prize in Physiology or Medicine in 2015. Chinese scientists also provided the first hint that artemisinin might have antiviral activity [16]. The antiviral activity of artemisinin and its semisynthetic derivative artesunate includes the inhibition of viruses, such as human cytomegalovirus and other members of the Herpesviridae family (e.g., herpes simplex virus type 1 and Epstein-Barr virus), hepatitis B virus, hepatitis C virus, and bovine viral diarrhea virus [17]. The broad anti-cytomegaloviral activity of artesunate was developed against viral strains and therapyresistant mutants, and it was shown that viral replication is blocked at a very early stage and that in vitro efficacy is optimized when artesunate was applied as fractional doses with a synergistic effect with the artesunate-maribavir combination treatment [18]. Artesunate showed an inhibitive effect on HCV replication and may be a novel supplemental cotherapy with interferon-2b and ribavirin for HCV and as an alternative strategy to combat resistance mechanisms that have emerged in the presence of direct antiviral agents [19]. Recently, artesunic acid-quinoline hybrids demonstrated an inhibitory activity against SARS-CoV-2 in vitro (EC 50 13-19 µM) and no cytotoxic effects on VeroE6 cells (CC 50 up to 110 µM) [20]. The case of artemisinin could be supported by other examples of the fairly common practice for drugs with known activity to be tested for other types of effects. For instance, amodiaquine, being 4-aminoquinoline and related to chloroquine with antimalarial activity, was recently found to be an effective inhibitor of the viral entry of SARS-CoV-2 in vitro and in vivo [21] and completed Phase 2 trials for COVID-19 treatment in South Africa (NCT04532931).
Besides artemisinin, more examples of natural compounds holding a peroxide group with antiviral activity could be named. Norsesterterpene peroxide acid muqubilone, isolated from the Red Sea sponge Diacarnus erythraeanus, showed in vitro antiviral activity against herpes simplex type 1 with ED 50 value of 7.5 µg/mL [22]. A series of cholesterol and β-sitosterol derivatives with a peroxide bridge at the B-ring were synthesized and screened for their anti-HBV activity; among them, 5α,8α-cyclicobioxygen-6-vinyl-3-oxocholesterone exhibited high inhibitory capacity against HBsAg and HBeAg secretion in HepG2.2.15 cells, whereas nearly no cytotoxicity on HepG2.2.15 cells at 3.13 µg/mL was observed, showing good therapeutic index [23]. Ergosterol peroxide inhibited Porcine delta-coronavirus infection and regulated host immune responses by downregulating the activation of the NF-κB and p38/MAPK signaling pathways in vitro, which makes it a potential candidate for the development of a new anti-PDCoV therapy [24]. Ergosterol peroxide also suppressed IAV-associated inflammation and apoptosis via blocking RIG-I signaling, which may serve as a supplementary approach to the treatment of influenza [25].
Advances in the development of natural peroxides prompted the synthesis of steroid-based 1,2,4,5-tetraoxanes [26] (including dimer molecules [27]), as well as 1,2,4trioxanes [28], that demonstrated anti-malarial, antiproliferative, antimycobacterial [29], and other types of activities.
Another sub-class of peroxides, namely 1,2,4-trioxolanes, prepared via the Griesbaum co-ozonolysis, were firstly developed on the basis of the adamantanone scaffold affording an extensive series named as "OZ" [30,31]. One of them, ozonide OZ277, also known as arterolane maleate, was approved for marketing in India as a combination product with piperaquine phosphate (Synriam) to treat malaria. Another adamantanone-based 1,2,4-trioxolane artefenomel (OZ439) was until very recently a leading candidate to replace the artemisinin and its semisynthetic analogs in artemisinin combination therapy for uncomplicated malaria [32].
The Griesbaum co-ozonolysis [33], the last three decade anniversary of which was celebrated in 2025, employs O-methyl oximes and carbonyl compounds as substrates to produce 1,2,4-trioxolanes using ozone. Compared to conventional olefin ozonolysis, this method offers several significant advantages such as improved selectivity, preparation of unsymmetrically substituted ozonides without the formation of undesired by-products, etc. [34].
The group of steroid-based spiro-1,2,4-trioxolanes is a new and few in number set of hybride molecules first launched in 2014 [35], then expanded to deoxycholic acid 3 ′ -trifluoromethylated 1,2,4-trioxolanes, which demonstrated in vitro antimalarial activity against the chloroquine-sensitive T96 and chloroquine-resistant K1 strains of Plasmodium falciparum [36].
According to all previous information, this article's concern is the synthesis of a new steroid-based 3-spiro-1,2,4-trioxolane molecule, an in silico ADMET study, and the evaluation of in vitro anti-influenza activity.
## 2. Results and Discussion
## 2.1. Chemistry
To prepare the target hybrid molecule holding the lithocholic acid scaffold and 1,2,4-trioxolane unit at C3 position, a synthetic strategy involving the abovementioned Griesbaum co-ozonolysis (Scheme 1) was selected. As the starting material, 3-oxolithocholic acid methyl ester 1 was obtained via a standard procedure [35] from commercially available LCA. The subsequent reaction of compound 1 with methoxy-hydroxylamine hydrochloride under reflux in a methanol-pyridine mixture afforded the O-methylketoxime 2 in a yield of 93%. In the 13 C NMR spectrum of compound 2, a pronounced downfield chemical shift in the C3 carbon signal was observed, along with the appearance of a characteristic signal corresponding to the C=NOCH 3 moiety at δ 165 ppm, indicating the full conversion of the 3-oxo-group. Additionally, the 1 H NMR spectrum revealed signals of a methoxy-group in the region of δ 3.9-4.0 ppm. The intensity ratio of the methoxy proton signals in the 1 H NMR spectrum, together with the presence of duplicated C3 carbon signals in the 13 C NMR spectrum, supports the formation of a mixture of two Z/E isomers of the O-methyl-ketoxime 2a and 2b (see Supplementary Materials Figures S1-S10). As a carbonyl component in the Griesbaum co-ozonolysis reaction, 4trifluoromethylcyclohexanone has been used, which, to the best of our knowledge, was not previously involved in such transformations despite the considerable attention to other substituted cyclohexanones. The selection of 4-trifluoromethylcyclohexanone as the ketone component for the Griesbaum co-ozonolysis was guided by well-established stereochemical trends for 4-substituted cyclohexanones in this reaction. In particular, the application of 4-alkyl-and 4-aryl-substituted cyclohexanones (such as 4-methyl-and 4-phenylcyclohexanones) has been previously described [31,36], affording the formation of the stable tetrasubstituted trioxolanes with the ratio diastereomers from 2.5:1 to 20:1. Tang and co-authors have noted that diastereoselectivity is a function of the nature and size of the substituents at the C4 position of the cyclohexanone ring. The bulky substituents at the C4 position strongly bias the approach of the carbonyl oxide, favoring axial attack and thus promoting the formation of a single dominant diastereomer [31]. This steric steering effect is even more pronounced for highly electron-withdrawing and volumetric groups (such as CF 3 in our case), which stabilize the preferred transition state which provides steric shielding around the emerging endoperoxide bond, contributing to the increased stability of the resulting 1,2,4-trioxolane, and suppress the formation of alternative stereoisomers [37]. Successful applications of other CF 3 -containing ketones in related trioxolane syntheses, including steroid-based frameworks [35], further support the suitability of 4-trifluoromethylcyclohexanone for constructing structurally defined and stable peroxide derivatives. The use of other substituted cyclohexanones as the carbonyl component revealed certain drawbacks; for example, the reaction was proceeded with low regio-and stereoselectivity, leading to the formation of complex mixtures of isomers rather than a single dominant product [35,36,38,39].
So, the last stage of our approach involved the interaction of O-methyl-ketoxime 2 and 4-(trifluoromethyl)-cyclohexanone under ozone in a mixture of methylene chloride and cyclohexane at 0 • C. The reaction was monitored by TLC (using the solvent system chloroform-ethyl acetate, 40:1); after finishing, the solvent was evaporated and the formation of several products in the crude reaction was observed. After purification by column chromatography, 3-spiro-1 ′ ,2 ′ ,4 ′ -trioxolane 3 as the main product (yield 72%) and a mixture of isomeric O-methoxylactams 4a and 4b (total yield 24%) were isolated. In contrast to the previously described spiro-trioxolanes based on deoxycholic acid and triterpenoids [35,36,38,39], in this case the use of 4-trifluoromethylcyclohexanone in the Griesbaum co-ozonolysis led to the selective formation of a single (3S)-diastereomer 3 in a quantitative yield, so we can conclude that the use of 4-(trifluoromethyl)-cyclohexanone allowed a stereospecific synthesis of steroid 1 ′ ,2 ′ ,4 ′ -trioxolane to be realized.
In order to confirm the structure and stereochemistry of the obtained compounds, COSY, NOESY, HSQC, and HMBC experiments were used. According to the NMR spectra, compound 2 is presented as a mixture of two Z/E isomers relative to the imine bond in a ratio of 1.3 (2a) to 1 (2b). Protons and carbon and nitrogen atoms, part of the A and B rings, are represented by a double set of signals in the 1 H, 13 C, and 15 N NMR spectra. The greatest differences in the chemical shifts in the signals between the Z and E isomers are observed for the positions closest to the isomerism site: C2, C3, C4, and NOCH 3 . The assignment of methyl oximes to the Z/E isomers was made on the basis of NOESY spectral data. For the major isomer, a NOESY cross peak was observed between the signals of the methyl oxime group protons and the equatorial proton at position C4 (δ H 3.79/2.79 ppm), indicating the Z-isomer 2a. The E-configuration of the imine bond in the minor isomer was established based on the observed NOESY cross peak between NOCH 3 and Heq-2 (δ H 3.78/2.95 ppm) (Figure 1). The structure of compound 3 was established by analysis of 1 H, 13 C, and 19 F NMR spectra by using two-dimensional correlation { 1 H, 13 X-ray crystallography is crucial in pharmaceutical research for newly synthesized compounds, providing absolute configuration determination and revealing hydrogen bonding, π-π stacking, and other interactions that influence stability and biological activity [40]. Single crystals of methyl (3S)-3,5 ′ -dispiro-[(4 ′′ -trifluoromethyl-cyclohexyl)-1 ′ ,2 ′ ,4 ′ -trioxolane]-5β-cholan-24-oate 3 suitable for structural analysis were successfully grown through slow evaporation from diethyl ether. The obtained crystal structures were analyzed using PLA-TON (v2.0) and MERCURY (v2.0) programs [41,42]. The structure of steroid trioxolane 3 is formed by two crystalographically independent molecules, one of which is shown in Figure 3. For these molecules, the bond lengths and bond angles are very close, and the same for the statistical means [43]. All cyclohexane fragments have a "chair" conformation; the 5-membered cycles have a "twist" conformation. Note that no reduced intermolecular contacts are observed in the crystal packing of compound 3. [44]. Therefore, the lengths of O3-C3, O2-C3, O3-C5 ′ , and O3-C5 ′ bonds differ from the standard value (1.43 Å [43]) and are 1.438 (9), 1.439 (8), 1.408 (10), and 1.443 (9) Å, respectively. For similar bonds in the previously studied deoxycholic acid 3-spiro-5 ′ -(3 ′methyl-3 ′ -trifluoromethyl-1,2,4-trioxolane) (compound 3a see Supplementary Materials Figure S23), the lengths are 1.434 (3), 1.433 (3), 1.421 (4), and 1.411 (3) Å. The peroxide bond lengths in both these trioxolanes are 1.461 (compound 3) and 1.473 (for compound 3a, Figure S23) which correspond to standard values [44].
The O2-C3-O3 bond angles (103.41 (compound 3) and 102.45 (compound 3a, Figure S23) are similar in value, while the O1-C5 ′ -O3 bond angle at C5 ′ differs significantly (102.95 (compound 3) and 106.19 (compound 3a, Figure S23)). This is due to the fact that in compound 3 it is a part of the cyclohexane fragment and serves as a spiro- According to the NMR spectral data, compound 4 is represented by seven-membered lactams in the form of two isomers at the position of the amide group in a ratio of 1.3 (4a) to 1 to 1 (4b). The formation of ε-lactams according to the A-ring was established on the basis of the observed carboxamide signals at δ C 171.39 and 171.55 ppm, as well as the δ N values of 195.69 and 192.78 ppm for the methoxy-substituted lactam nitrogens. For the major isomer with the 3-aza-A-homo structure 4a, HMBC cross peaks of the H-5, H ax -2, H eq -2, and H ax -4a/H eq -4a protons with the lactam carbonyl are observed at δ C 171.39 ppm. In addition, in the { 1 H, 15 N} HMBC spectrum there are cross peaks of the protons of H ax -1, H eq -1, and H eq -4a and the OMe group (δ H 3.73 ppm) with a nitrogen signal at δ N 195.69 ppm. The presence of nitrogen in position C3 is confirmed by the NOESY cross peak of the methoxy substituent with the equatorial proton H eq -2 (δ H 3.73/3.54 ppm).
For the minor isomer 4b, HMBC cross peaks of the H ax -1, H eq -1, and H eq -4a protons with the amide carbonyl are observed at δ C 171.55 ppm and { 1 H, 15 N} HMBC cross peaks of the protons H ax -4a, H eq -4a, and H ax -2 with δ N 192.78 ppm, confirming the 4-aza-A-homo structure of ring A. For the methoxyl protons of the minor isomer 4b, NOE interactions with the equatorial proton H eq -4a are observed (δ H 3.75/3.20 ppm) (Figure 4 and see Supplementary Materials Figures S24-S33). Thus, we can conclude that the use of 4-(trifluoromethyl)-cyclohexanone as a carbonyl component in the Griesbaum ozonolysis in the example of LCA provides a one-step stereospecific formation of (3S)-stereoisomeric 1,2,4-trioxolane that makes this approach attractive for further development of a new generation of steroidal peroxides.
## 2.2. In Vitro Antiviral Evaluation
Cytotoxicity and anti-influenza properties of the lithocholic acid derivatives 1-3 were studied in MDCK cell culture against influenza virus A/PuertoRico/8/34 (H1N1). Oseltamivir carboxylate and rimantadine were used as reference compounds (Table 1). The structure-activity relationship analysis revealed that structural modifications of lithocholic acid markedly affect both cytotoxicity and antiviral potency. The parent methyl lithocholate 1 exhibited low antiviral activity (IC 50 > 84 µM) and moderate cytotoxicity (CC 50 107.4 µM), compared to the reference drug rimantadine. This resulted in a low selectivity index (SI 1). Introduction of a methoxy-oxime substituent at C3 (compound 2) slightly improved the selectivity (SI 2), possibly due to enhanced polarity and hydrogen bonding capacity, but did not significantly increase antiviral efficacy. The most active derivative was considered to be 1,2,4-trioxolane 3 with IC 50 4.3 µM and SI 11, which indicates that the incorporation of a trioxolane ring considerably enhances antiviral activity while maintaining moderate cytotoxicity, but with notably higher antiviral potency and selectivity than rimantadine (CC 50 62 µM and IC 50 11 µM). Based on the provided data, target lithocholic acid spiro-1,2,4-trioxolane 3 shows higher antiviral activity and selectivity compared to the parent methyl 3-oxo-lithocholate 1 and its methoxy-oxime 2. While this observation indicates that the introduction of a trioxolane ring positively influences anti-flu activity, this only one example does not allow us to make definitive conclusions regarding the role of the trioxolane moiety as a structural determinant of antiviral potency. Nevertheless, our results suggest that the hybridization of bile acid scaffolds with a 1,2,4-trioxolane unit could produce molecules with antiviral effects.
## 2.3. In Silico ADMET Study and Physicochemical Profiles of Compound 3
To assess the pharmaceutical relevance of the synthesized lead molecule 3, an analysis was conducted across key drug development parameters [45], including physicochemical descriptors, lipophilicity, solubility, pharmacokinetic predictions, and drug-likeness criteria, using SWISS ADME (https://www.swissadme.ch/). The calculated data are presented in Table 2 (for the parent compound 1 see Supplementary Materials Table S2). 3 character (fraction Csp 3 > 0.9), a feature associated with improved clinical success rates due to enhanced three-dimensionality and target specificity [46].
## 2.4.2. Lipophilicity and Solubility
Lipophilicity, a key determinant of membrane permeability and solubility, was predicted using multiple algorithms. Compound 3 exhibits markedly higher lipophilicity across all models, with a consensus LogP of 7.54, exceeding the optimal range for oral drugs (LogP < 5) [47]. Correspondingly, it is predicted to be poorly soluble, with logS values below -6.5 in all models. Such high lipophilicity and limited solubility may contribute to reduced oral bioavailability and variable pharmacokinetic behavior [48].
## 2.4.3. Pharmacokinetic Predictions
In silico ADME analysis predicts that compound 3 has low gastrointestinal absorption and is not blood-brain barrier permeant, consistent with its large molecular size and poor solubility. It shows no major cytochrome P450 interactions and possesses a relatively high predicted skin permeability (log Kp = -3.20 cm/s), aligning with its lipophilic nature.
## 2.4.4. Drug-likeness and Medicinal Chemistry Filters
Compound 3 fails two Lipinski criteria (MW > 500 and MLOGP > 4.15), in addition to violations in the Ghose and Egan filters. Its bioavailability score is 0.17 [49]. A Brenk alert is triggered due to the presence of a peroxide moiety, raising potential concerns about chemical stability and toxicity [50]. These parameters are presented at Figure 5. Despite its predicted limited aqueous solubility and poor oral bioavailability, compound 3 exhibits a combination of structural and physicochemical properties that may offer distinct therapeutic advantages, particularly in the context of antimalarial drug development. Among its notable features is a high fraction of sp 3 -hybridized carbon atoms (Fsp 3 = 0.97), which reflects a three-dimensional structure often associated with improved target selectivity, reduced off-target interactions, and enhanced clinical progression [46]. In addition, compound 3 contains eight hydrogen bond acceptors, allowing for increased potential to engage in polar interactions with protein targets, which may enhance binding affinity and specificity [51]. Although highly lipophilic (consensus LogP = 7.54), this characteristic can support passive diffusion across lipid-rich membranes [52]. Furthermore, compound 3 demonstrates a relatively favorable predicted skin permeability (Log Kp = -3.20 cm/s), suggesting its potential use in topical or transdermal applications [53]. The trifluoromethyl substituent contributes to increased metabolic stability, modulates the electronic environment, and may promote halogen bonding with enzyme active sites commonly leveraged in modern drug design [54,55].
## 2.5. Predicted Biological Activities of Compound 3
In silico prediction using the computer program PASS (v2.0) (Prediction of Activity Spectra for Substances) (https://way2drug.com) has been successfully applied to natural product scaffolds. Using PASS, some additional biological activities could be predicted, which point toward new possible applications of these compounds. This well-established approach serves as a valuable tool for the rational design and selection of promising biological active compounds. For example, PASS prediction for natural polycyclic endoperoxides has found that ~65% of them showed Pa values between 0.70 and 0.90 for antiprotozoal activity [56]. According to PASS, plant hydroperoxides demonstrated a wide range of biological activities with antineoplastic and anti-ulcerative as the most expected [57], while for natural steroid and triterpenoid endoperoxy and hydroperoxyl derivatives antihypercholesterolemic, ovulation inhibitory, or anticancer effects are possible with Pa > 0.90 [58,59].
PASS analysis has been used for the new hybrid molecule 3 with a bile acid core and spiro-1,2,4-trioxolane moiety. The results of predicted bioactivities are shown in Table 3 with Pa (probability of activity) > 0.5 [60]. As one can see, compound 3 could possess a broad spectrum of activities with Pa values varying between 0.908 and 0.501 such as antiprotozoal (active against several parasitic infections, such as malaria, amebiasis, giardiasis, cryptosporidiosis, trypanosomiasis leishmaniasis, and toxoplasmosis), anti-inflammatory and dermatological effects (antieczematic, antipruritic), influence on lipid metabolism (enzyme inhibition, cholesterol antagonism), and potential anticancer or chemopreventive activity (adenomatous polyposis treatment). These predictions are consistent with those previously reported for natural and synthetic steroid and terpene peroxides, which are well-known for strong antiparasitic effects and modulation of lipid-related pathways [14]. Moderate probabilities were also observed for dermatological effects and antiviral activity against rhinovirus which align with emerging reports of membrane-interacting endoperoxides exhibiting moderate inhibition of RNA viruses [61]. Collectively, the obtained in silico data for compound 3 within the established profile of peroxide-based bioactive molecules reveal the potency of this steroidal hybrid for further exploration.
## 3. Materials and Methods
The 1 H and 13 C NMR spectra were recorded on a "Bruker Avance-III" (Bruker, Billerica, MA, USA, 500 and 125.5 MHz, respectively, δ, ppm, Hz) in CDCl 3 , internal stand-ardtetramethylsilane. Mass spectra were obtained on a liquid high-resolution chromatograph-mass spectrometer Agilent LC/Q-TOF 6530 (Agilent Technologies, Santa Clara, CA, USA). Melting points were detected on a microtable «Rapido PHMK05» (Nagema, Dresden, Germany). Optical rotations were measured on a polarimeter Perkin-Elmer 241 MC (PerkinElmer, Waltham, MA, USA) in a tube length of 1 dm. Elemental analysis was performed on a Euro EA-3000 CHNS analyzer (Eurovector, Milan, Italy); the main standard is acetanilide. Thin-layer chromatography analyses were performed on Sorbfil plates (Sorbpolimer, Krasnodar, Russia), using the solvent system chloroformethyl acetate, 40:1. Substances were detected by a 10% solution of sulfuric acid solution with subsequent heating at 100-120 • C for 2-3 min. All chemicals were of reagent grade (Sigma-Aldrich, St. Louis, MO, USA). XRD data for compound 3 were obtained on a Bruker Kappa Apex II CCD diffractometer using φ, ω scans of narrow (0.8 • ) frames with Mo Kα radiation (λ = 0.71073 Å) and a graphite monochromator at T = 296(2) K. The structure was solved by direct methods using the SHELXT-2014/5 [62] and refined by full-matrix least-squares method against all F2 in anisotropic approximation using the SHELXL-2018/3 [62]. Absorption corrections were applied using the empirical multi-scan method with the SADABS program (Version 2008/1) [63]. Compound 1 was obtained according to the method described previously [35].
3.1. The Procedure for Synthesis of Compound 2 CH 3 ONH 2 •HCl (0.17 g, 2 mmol) was added to the solution of compound 1 (0.39 g, 1 mmol) in a mixture of pyridine and methanol (30 mL, 1:1, v/v). The reaction mixture was refluxed for 8 h with a back condenser, cooled to room temperature, and quenched with 5% HCl (150 mL). The precipitate was filtered off, washed with water, and airdried. The yield of amorphous substance was 0.39 g (93%) (compounds 2a and 2b), [α] 20 D +14 (c 0.10, CH 2 Cl 2 ).
3.1.1. Methyl-O-methyl-3(Z)-oxyimino-5β-cholan-24-oate (2a) 1 H NMR (CDCl 3 , δ ppm, J Hz): 0.64 (s, 3H, H-18); 0.89 (d, 3H, 3 J 21-20 = 6. -6); 1.87 (m, 1H, H eq -1); 1.96 (m, 1H, H eq -12); 2.07 (td, 1H, 2 J = 14.8, 3 J 2ax-1ax = 14.8, 3 J 2ax-1eq = 4.8, H ax -2); 2.11 (dd, 1H, 2 J = 15.5, 3 J 4ax-5 = 13.4, H ax -4); 2.12 (m, 1H, H eq -2); 2.19 (ddd, 1H, 2 J = 15.4, 3 J 23A-22A = 9.7, 3 J 23A-22B = 6.5, H A -23); 2.33 (ddd, 1H, 2 J = 15.4, 3 J 23B-22B = 10.1, 3 J 23B-22A = 5.2, H B -23); 2.79 (ddd, 1H, 2 J = 15.5, 3 J 4eq-5 = 5.0, 4 J 4eq-2eq = 1.5, H eq -4); 3.64 (s, 3H, OMe); 3.79 (s, 3H, NOCH 3 ). 13 39 (m,1H,); 1.39 (m, 1H, H-8); 1.41 (m, 1H, H eq -11); 1.44 (m, 1H, H eq -7); 1.54 (m, 1H, H-5); 1.56 (m, 1H, H α -15); 1.69 (td, 1H, 2 J = 14.6, 3 J 2ax-1ax = 14.6, 3 J 2ax-1eq = 4.8, H ax -2); 1.77 (m, 1H, H B -22); 1.83 (m, 1H, H α -16); 1.84 (m, 1H, H eq -1); 1.84 (m, 1H, H ax -6); 1.94 (ddd, 1H, 2 J = 14.6, 3 J 4eq-5 = 4.8, 4 J 4eq-2eq = 2.0, H eq -4); 1.96 (m, 1H, H eq -12); 2.19 (ddd, 1H, 2 J = 15.4, 3 J 23A-22A = 9.7, 3 J 23A-22B = 6.5, H A -23); 2.33 (ddd, 1H, 2 J = 15.4, 3 J 23B-22B = 10.1, 3 J 23B-22A = 5.2, H B -23); 2.49 (dd, 1H, 2 J = 14.6, 3 J 4ax-5 = 13.3, H ax -4); 2.95 (dddd, 1H, 2 J = 14.6, 3 J 2eq-1ax = 4.5, 3 J 2eq-1eq = 2.6, 4 J 2eq-4eq = 2.0, H eq -2); 3.64 (s, 3H, OMe); 3.78 (s, 3H, NOCH 3 ). 13
## 3.2. The Procedure of Griesbaum Co-Ozonolysis
Ozone was bubbled through a solution of compound 2 (0.47 g, 1 mmol) in a mixture of cyclohexane and CH 2 Cl 2 (30 mL, 1:2, v/v) in the presence of 4-(trifluoromethyl)cyclohexanone (0.28 mL, 2 mmol) at 0 • C with TLC control. After completion of the reaction, the solution was flushed with oxygen for 5 min before being concentrated in vacuo at room temperature to give a residue that was then purified by column chromatography (hexane, benzene, and chloroform as the eluents) to afford compounds 3 (0.41g, 72%) and 4 (0.1 g, 24%).
Obtained by crystallization from Et 2 O. [α] 20 D + 45 • (c 0.10, CHCl 3 ). mp. 101 • C. 1 H NMR (CDCl 3 , δ ppm, J Hz): 0.64 (s, 3H, H-18); 0.91 (d, 3H, 3 J 21-20 = 6.5, H-21); 0.93 (s, 3H, H-19); 1.01 (m, 1H, H-14); 1.03 (m, 1H, H β -15); 1.04 (m, 1H, H ax -7); 1.08 (m, 1H, H-17); 1.11 (td, 1H, 2 J = 12.5, 3 J 12ax-11ax = 12.5, 3 J 12ax-11eq = 4.4, H ax -12); 1.21 (m, 1H, H ax -1); 1.22 (m, 1H, H eq -6); 1.25 (m, 1H, H ax -11); 1.28 (m, 1H, H β -16); 1.31 (m, 1H, H-9); 1.33 (m, 1H, H A -22); 1.35 (m, 1H, H eq -11); 1.37 (m, 1H, H-8); 1.42 (m, 1H, H-20); 1.42 (m, 1H, H eq -7); 1.51 (ddd, 1H, 2 J = 13.6, 3 J 4eq-5 = 4.2, 4 J 4eq-2eq =1.4, H eq -4); 1.55 (m, 1H, H-5); 1.56 (m, 1H, H α -15); 1.60 (m, 1H, H ax -5 ′′ ); 1.61 (m, 1H, H ax -3 ′′ ); 1.64 (m, 1H, H ax -6 ′′ ); 1.67 (m, 1H, H ax -2 ′′ ); 1.68 (m, 2H, H-2); 1.71 (dt, 1H, 2 J = 11.3, 3 J 1eq-2ax = 3.2, 3 J 1eq-2eq = 3.2, H eq -1); 1.79 (m, 1H, H B -22); 1.82 (m, 1H, H ax -6); 1.83 (m, 1H, H α -16); 1.91 (m, 1H, H eq -3 ′′ ); 1.92 (m, 1H, H eq -5 ′′ ); 1.96 (dt, 1H, 2 J = 12.6, 3 J 12eq-11ax = 3.1, 3 J 12eq-11eq = 3.1, H eq -12); 2.02 (m, 1H, H eq -2 ′′ ); 2.03 (m, 1H, H-4 ′′ ); 2.05 (m, 1H, H eq -6 ′′ ); 2.15 (t, 1H, 2 J = 13.6, 3 J 4ax-5 = 13.6, H ax -4); 2.22 (ddd, 1H, 2 J = 15.4, 3 J 23A-22A = 9.7, 3 J 23A-22B = 6.5, H A -23); 2.35 (ddd, 1H, 2 J = 15.4, 3 J 23B-22B = 10.2, 3 J 23B-22A = 5.2, H B -23); 3.66 (s, 3H, OMe). 13 (dddd, 1H, 2 J = 14.6, 3 J2eq-1ax = 4.5, 3 J2eq-1eq = 2.6, 4 J2eq-4eq = 2.0, Heq-2); 3.64 (s, 3H, OMe); 3.78 (s, 3H, NOCH3). 13
$$3.2.1. (3S)-3,5 ′ -dispiro-[(4 ′′ -trifluoromethyl-cyclohexyl)-1 ′ ,2 ′ ,4 ′ -trioxolane]-5β-cholan- 24-oate (3)$$
## 3.2. The Procedure of Griesbaum Co-Ozonolysis
Ozone was bubbled through a solution of compound 2 (0.47 g, 1 mmol) in a mixture of cyclohexane and CH2Cl2 (30 mL, 1:2, v/v) in the presence of 4-(trifluoromethyl)cyclohexanone (0.28 mL, 2 mmol) at 0 °C with TLC control. After completion of the reaction, the solution was flushed with oxygen for 5 min before being concentrated in vacuo at room temperature to give a residue that was then purified by column chromatography (hexane, benzene, and chloroform as the eluents) to afford compounds 3 (0.41g, 72%) and 4 (0.1 g, 24%). Crystallographic data for 3: C22H49F3O5, FW = 570.71, Orthorhombic, P212121, a = 6.588(1), b = 22.318(4), c = 42.214(8) Ǻ V = 6207(2) Å 3 , Z = 8, Dcalc. = 1.221 g cm -3 , µ (Mo-Kα) = 0.092 mm -1 , F(000) = 2464, 25040 measured reflections (θmax = 25.03°, completeness 99.8%), 10933 independent (Rint = 0.080), 721 parameters, R1 = 0.0909 (for 5846 observed I > 2σ(I)), wR2 = 0.1856 (all data), GooF = 1.06, largest diff. peak and hole 0.26 and -0.26 × 10 -3 . Quite high value of the R-factor is explained by bad quality of the crystal; it was not succeeded to receive crystals of the best quality. CCDC 2493448 contains the supplementary crystallographic data for this paper. These data can be obtained from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif, accessed on 25 November 2025.
, V = 6207(2) Å 3 , Z = 8, Dcalc. = 1.221 g cm -3 , µ (Mo-Kα) = 0.092 mm -1 , F(000) = 2464, 25040 measured reflections (θmax = 25.03 • , completeness 99.8%), 10933 independent (Rint = 0.080), 721 parameters, R1 = 0.0909 (for 5846 observed I > 2σ(I)), wR2 = 0.1856 (all data), GooF = 1.06, largest diff. peak and hole 0.26 and -0.26 × 10 -3 . Quite high value of the R-factor is explained by bad quality of the crystal; it was not succeeded to receive crystals of the best quality. CCDC 2493448 contains the supplementary crystallographic data for this paper. These data can be obtained from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif, accessed on 25 November 2025.
3.2.2. Methyl 4-oxo-3-methoxyaza-A-homo-5β-cholan-24-oate (4a) -4a); 2.22 (ddd, 1H, 2 J = 15.4, 3 J 23A-22A = 9.7, 3 J 23A-22B = 6.5, H A -23); 2.35 (ddd, 1H, 2 J = 15.4, 3 J 23B-22B = 10.1, 3 J 23B-22A = 5.2, H B -23); 2.95 (dd, 1H, 2 J = 15.3, 3 J 4aax-5 = 12.0, H ax -4a); 3.54 (d, 1H, 2 J = 14.9, 3 J 2eq-1eq = 9.1, H eq -2); 3.66 (s, 3H, OMe); 3.67 (d, 1H, 2 J = 14.9, 3 J 2ax-1ax = 9.1, H ax -2); 3.73 (s, 3H, NOCH 3 ). 13
= 15.0, 3 J 1ax-2ax = 11.4, H ax -1); 1.51 (m, 1H, H eq -7); 1.52 (m, 1H, H eq -6); 1.58 (m, 1H, H α -15); 1.76 (m, 1H, H-5); 1.78 (dd, 2 J = 15.0, 3 J 1eq-2eq = 10.0, H eq -1); 1.80 (m, 1H, H B -22); 1.86 (m, 1H, H α -16); 1.92 (m, 1H, H ax -6); 1.99 (m, 1H, H eq -12); 2.22 (ddd, 1H, 2 J = 15.4, 3 J 23A-22A = 9.7, 3 J 23A-22B = 6.5, H A -23); 2.32 (d, 1H, 2 J = 15.9, 3 J 2eq-1eq = 10.0, H eq -2); 2.35 (ddd, 1H, 2 J = 15.4, 3 J 23B-22B = 10.1, 3 J 23B-22A = 5.2, H B -23); 2.50 (d, 1H, 2 J = 15.9, 3 J 2ax-1ax = 11.4, H ax -2); 3.20 (d, 1H, 2 J = 15.2, H eq -4a); 3.66 (s, 3H, OMe); 3.75 (s, 3H, NOCH 3 ); 4.25 (dd, 1H, 2 J = 15.2, 3 J 4aax-5 = 11.0, H ax -4a). 13
## 3.3. Biological Activity Assays
Viruses and cells: Influenza virus A/Puerto Rico/8/34 (H1N1) was obtained from the collection of viruses of St. Petersburg Pasteur Institute. Before the experiment, virus was propagated in the allantoic cavity of 10-to 12-day-old chicken embryos for 48 h at 36 • C. The infectious titer of the virus was determined in Madin-Darby canine kidney (MDCK) cells (ATCC-CCL-34) grown in 96-well plates in alpha-MEM medium with 10% fetal bovine serum.
## 3.3.1. Cytotoxicity Assay
MDCK cells were seeded onto 96-well culture plates (104 cells per well) and incubated at 36 • C in 5% CO 2 until continuous monolayer formation. To assess the toxicity of compounds, a series of their 3-fold dilutions at concentrations of 300 to 3.7 µg/mL in Eagle's Minimal Essential Medium (MEM) were prepared. The dilutions were added to the wells of the plates. Cells were incubated for 72 h at 36 • C in a CO2 incubator under 5% CO2. Further, a microtetrazolium (MTT) assay was performed on 96-well plates. The cells were washed 2 times with saline (0.9% NaCl), and 100 µL/well of MTT solution [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] at a concentration of 0.5 µg/mL in MEM was added. The plates were incubated for 1 h at 36 • C, the liquid was removed, and dimethylsulfoxide (DMSO) (0.1 mL per well) was added. The optical density (OD) of the cells was measured on a Thermo Multiskan FC spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at a wavelength of 540 nm. Based on the obtained data, the CC 50 , the concentration of the compound that destroys 50% of the cells in the culture, was calculated for each specimen.
## 3.3.2. CPE Reduction Assay
The compounds in appropriate concentrations were added to MDCK cells (0.1 mL per well). MDCK cells were further infected with A/Puerto Rico/8/34 (H1N1) influenza virus (m.o.i 0.01). Plates were incubated for 72 h at 36 • C at 5% CO 2 . After that, cell viability was assessed by the MTT test, as described above. The cytoprotective activity of compounds was considered as their ability to increase the values of the OD compared to the control wells (with virus only; no drugs). Based on the obtained results, the IC 50 values, i.e., the concentration of compounds that results in 50% cell protection, were calculated using GraphPad Prism 6.01 software. IC 50 values in µg/mL were then calculated into micromoles. For each compound, the value of the selectivity index (SI) was calculated as a ratio of CC 50 to IC 50 .
## 3.3.3. Statistical Data
Calculations of the 50% cytotoxicity response (CC 50 ) and 50% treatment efficacy (IC 50 ) results were performed using the GraphPad Prism v.6.0 software package. A fourparameter logistic curve equation was used as the initial model for the analysis (menu items "Nonlinear Regression"-"Logarithm of Inhibitor-Response"). Based on the obtained data, the selectivity index (SI)-the ratio of CC 50 to IC 50 -is calculated for each compound and each virus. A combination of states is considered active if its selectivity index is 10 or higher [64].
## 3.4. SwissADME
The physicochemical properties of compound 3 were calculated using online software SwissADME (Version 2017/1) (https://www.swissadme.ch/index.php) (accessed on 7 October 2025).
## 3.5. PASS Analysis
PASS (v2.0) (Prediction of Activity Spectra for Substances) analysis was calculated on https://way2drug.com. The output file represents a list of activities with two probabilities Pa (probability to be active) and Pi (probability to be inactive). Pa (probability "to be active") estimates the chance that the studied compound belongs to the sub-class of active compounds (resembles the structures of molecules, which are the most typical in a sub-set of "actives" in PASS training set). Pi (probability "to be inactive") estimates the chance that the studied compound belongs to the sub-class of inactive compounds (resembles the structures of molecules, which are the most typical in a sub-set of "inactives" in PASS training set). The Pa value based on a cut-off of >0.5 [60].
## 4. Conclusions
By the application of a synthetic method of Griesbaum co-ozonolysis, a novel steroid peroxide, namely methyl 3(S)-3,5 ′ -dispiro-[(4 ′′ -trifluoromethyl-cyclohexyl)-1 ′ ,2 ′ ,4 ′trioxolane]-5β-cholan-24-oate, was stereospecifically formed and its structure was established by 2D NMR and X-ray crystallographic analysis. The evaluation of its cytotoxicity and anti-influenza activity against A/Puerto Rico/8/34 (H1N1) showed that steroid 1,2,4trioxolane 3 exhibited the highest potency (IC 50 4.3 µM, SI 11) compared to the parent
## References
1. Kazakova, Giniyatullina, Babkov et al. (1075) "From Marine Metabolites to the Drugs of the Future: Squalamine, Trodusquemine, Their Steroid and Triterpene Analogues" *Int. J. Mol. Sci*
2. Wang, Jiang, Fu et al. (2019) "Self-Assembling Peptide-Based Nanodrug Delivery Systems" *Biomater. Sci*
3. Salunke, Hazra, Pore (2006) "Steroidal Conjugates and Their Pharmacological Applications" *Curr. Med. Chem*
4. Taylor, Green (2018) "Bile Acids, Microbiota, and Metabolism" *Hepatology*
5. Li, Chiang (2015) "Bile Acid Signaling in Metabolic Disease and Drug Therapy" *Pharmacol. Rev*
6. Chiang (2018) "Bile Acid Metabolism and Signaling in Liver Disease and Therapy" *Liver Res*
7. Kong, Niu, Liu et al. (2021) "Bile Acids LCA and CDCA Inhibited Porcine Deltacoronavirus Replication In Vitro" *Vet. Microbiol*
8. Mishra, Mishra (2020) "Updates in Bile Acid-Bioactive Molecule Conjugates and Their Applications" *Steroids*
9. De Oca, Díaz, Ávila et al. (2015) "Process for Obtaining a Molecule Serving as an Antimicrobial Peptide" *Patent MX*
10. Marchesi, Gentili, Bortolotti et al. (2023) "Dihydroartemisinin-Ursodeoxycholic Bile Acid Hybrids in the Fight against SARS-CoV-2"
11. Winkler, Shrihari, Hykes et al. (2020) "The Intestinal Microbiome Restricts Alphavirus Infection and Dissemination through a Bile Acid-Type I IFN Signaling Axis" *Cell*
12. Villalobos Sánchez, Márquez-Villa, Vega-Rodríguez et al. (2025) "Lithocholic Acid Oleate Preparative Synthesis and Its Formulation with Lithocholic Acid as a Preventive Antiviral"
13. Petrova, Smirnova, Fedij et al. (1626) "Synthesis and Inhibition of Influenza H1N1 Virus by Propargylaminoalkyl Derivative of Lithocholic Acid" *Molbank*
14. Vil, Yaremenko, Ilovaisky et al. (1881) *Peroxides with Anthelmintic, Antiprotozoal, Fungicidal and Antiviral Bioactivity: Properties, Synthesis and Reactions. Molecules*
15. Shukla, Rathi, Hassam et al. (2024) *An Overview on the Antimalarial Activity*
16. Qian, Li, Yu et al. (1982) "The immunologic and antiviral effect of qinghaosu" *J. Tradit. Chin. Med*
17. Efferth, Romero, Wolf et al. (2008) "The Antiviral Activities of Artemisinin and Artesunate" *Clin. Infect. Dis*
18. Chou, Marousek, Auerochs et al. (2011) "The Unique Antiviral Activity of Artesunate Is Broadly Effective against Human Cytomegaloviruses Including Therapy-Resistant Mutants" *Antivir. Res*
19. Dai, Xiao, Peng et al. (2016) "Artesunate, an Anti-Malarial Drug, Has a Potential to Inhibit HCV Replication" *Virus Genes*
20. Herrmann, Yaremenko, Çapcı et al. (2022) "Synthesis and In Vitro Study of Artemisinin/Synthetic Peroxide-Based Hybrid Compounds against SARS-CoV-2 and Cancer" *ChemMedChem*
21. Si, Bai, Rodas et al. (2021) "A Human-Airway-ona-Chip for the Rapid Identification of Candidate Antiviral Therapeutics and Prophylactics" *Nat. Biomed. Eng*
22. El Sayed, Hamann, Hashish et al. (2001) "Antimalarial, Antiviral, and Antitoxoplasmosis Norsesterterpene Peroxide Acids from the Red Sea Sponge Diacarnus erythraeanus" *J. Nat. Prod*
23. Jia, Zhao, Xu et al. (2016) "Synthesis and Biological Activity Evaluation of Novel Peroxo-Bridged Derivatives as Potential Anti-Hepatitis B Virus Agents"
24. Duan, Ge, Wang et al. (2021) "Ergosterol Peroxide Exhibits Antiviral and Immunomodulatory Abilities against Porcine Deltacoronavirus (PDCoV) via Suppression of NF-κB and p38/MAPK Signaling Pathways In Vitro" *Int. Immunopharmacol*
25. Zhou, Liang, Feng et al. (2019) "Ergosterol Peroxide Suppresses Influenza A Virus-Induced Pro-Inflammatory Response and Apoptosis by Blocking RIG-I Signaling" *Eur. J. Pharmacol*
26. Šolaja, Terzić, Pocsfalvi et al. (2002) "Mixed Steroidal 1,2,4,5-Tetraoxanes: Antimalarial and Antimycobacterial Activity" *J. Med. Chem*
27. Opsenica, Pocsfalvi, Juranić et al. (2000) "Cholic Acid Derivatives as 1,2,4,5-Tetraoxane Carriers: Structure and Antimalarial and Antiproliferative Activity" *J. Med. Chem*
28. Singh, Hassam, Verma et al. (2012) "Bile Acid-Based 1,2,4-Trioxanes: Synthesis and Antimalarial Assessment" *J. Med. Chem*
29. Opsenica, Kyle, Milhous et al. (2003) "Antimycobacterial and Antiproliferative Activity of Phenyl-Substituted Mixed Tetraoxanes" *J. Serb. Chem. Soc*
30. Oliveira, Miranda, Magalhães et al. (2015) "From Hybrid Compounds to Targeted Drug Delivery in Antimalarial Therapy" *Bioorg. Med. Chem*
31. Tang, Dong, Karle et al. (2004) "Synthesis of Tetrasubstituted Ozonides by the Griesbaum Co-Ozonolysis Reaction" *J. Org. Chem*
32. Charman, Arbe-Barnes, Bathurst et al. (2011) "Synthetic Ozonide Drug Candidate OZ439 Offers New Hope for a Single-Dose Cure of Uncomplicated Malaria" *Proc. Natl. Acad. Sci*
33. Griesbaum, Övez, Huh et al. (1995) "Ozonolyses of O-methyloximes in the presence of acid derivatives: A new access to substituted ozonides" *Liebigs Ann*
34. Zain-Ul-Abideen, Saeed, Haider et al. (1995) "Three Decades of Selective Product Formation via Griesbaum Co-Ozonolysis: Insight and Advances"
35. Yamansarov, Kazakova, Medvedeva et al. (2014) "First Synthesis of Steroidal 1,2,4-Trioxolanes" *Russ. J. Org. Chem*
36. Yamansarov, Kazakov, Medvedeva et al. (2018) "Synthesis and Antimalarial Activity of 3 ′ -Trifluoromethylated 1,2,4-Trioxolanes and 1,2,4,5-Tetraoxanes Based on Deoxycholic Acid" *Steroids*
37. Chen, Gonciarz, Renslo (2021) "Expanded Scope of Griesbaum Co-Ozonolysis for the Preparation of Structurally Diverse Sensors of Ferrous Iron" *RSC Adv*
38. Kazakova, Kazakov, Yamansarov et al. (2011) "Synthesis of Triterpenoid-Based 1,2,4-Trioxolanes and 1,2,4-Dioxazolidines by Ozonolysis of Allobetulin Derivatives" *Tetrahedron Lett*
39. Kazakova, Khusnutdinova, Petrova et al. (2019) "Diastereoselective Synthesis of Triterpenoid 1,2,4-Trioxolanes by Griesbaum Co-Ozonolysis" *J. Nat. Prod*
40. Thombre, Umap, Gupta et al. (2025) "Crystallization and X-ray Techniques: Pharmaceutical Application" *Res. J. Pharm. Technol*
41. Spek (2003) "PLATON: A Multipurpose Crystallographic Tool"
42. Macrae, Bruno, Chisholm et al. (2006) "Mercury CSD 2.0-New Features for Visualization and Analysis of Crystal Structures" *J. Appl. Crystallogr*
43. Allen, Kennard, Watson et al. (1987) "Tables of Bond Lengths Determined by X-Ray and Neutron Diffraction" *J. Chem. Soc. Perkin Trans*
44. Kirby (1985) "The Anomeric Effect of Oxygen-Containing Compounds"
45. Daina, Michielin, Zoete (2017) "SwissADME: A Free Web Tool to Evaluate Pharmacokinetics, Druglikeness and Medicinal Chemistry Friendliness of Small Molecules" *Sci. Rep*
46. Lovering, Bikker, Humblet (2009) "Escape from Flatland: Increasing Saturation as an Approach to Improving Clinical Success" *J. Med. Chem*
47. Lipinski (2004) "Lead-and Drug-Like Compounds: The Rule-of-Five Revolution" *Drug Discov. Today Technol*
48. Di, Fish, Mano (2012) "Bridging Solubility between Drug Discovery and Development" *Drug Discov. Today*
49. Martin, Bioavailability Score (2005) *J. Med. Chem*
50. Brenk, Schipani, James et al. (2008) "Lessons Learnt from Assembling Screening Libraries for Neglected Diseases" *ChemMedChem*
51. Kenny (2022) "Hydrogen-Bond Donors in Drug Design" *J. Med. Chem*
52. Duffy, Avery (2013) "Development and Optimization of a Novel 384-Well Anti-Malarial Imaging Assay" *Malar. J*
53. Potts, Guy (1992) "Predicting Skin Permeability" *Pharm. Res*
54. Zhang "Fluorine in Medicinal Chemistry: In Perspective to COVID-19" *ACS Omega*
55. Park, Kitteringham, O'neill (2001) "Metabolism of Fluorine-Containing Drugs" *Annu. Rev. Pharmacol. Toxicol*
56. Dembitsky, Ermolenko, Savidov et al. (2021) "Antiprotozoal and Antitumor Activity of Natural Polycyclic Endoperoxides: Origin, Structures and Biological Activity"
57. Vil, Terent'ev, Savidov et al. (2019) "Hydroperoxy steroids and triterpenoids derived from plant and fungi: Origin, structures and biological activities" *J. Steroid Biochem. Mol. Biol*
58. Dembitsky (2021) "Silico Prediction of Steroids and Triterpenoids as Potential Regulators of Lipid Metabolism"
59. Dembitsky, Gloriozova, Poroikov (2007) *Natural Peroxy Anticancer Agents. Mini Rev. Med. Chem*
60. Filimonov, Druzhilovskiy, Lagunin et al. (2018) "Computer-Aided Prediction of Biological Activity Spectra for Chemical Compounds: Opportunities and Limitations" *Biomed. Chem. Res. Methods*
61. Dembitsky (2015) "Astonishing diversity of natural peroxides as potential therapeutic agents" *J. Mol. Genet. Med*
62. Sheldrick (2015) "SHELXT-Integrated Space-Group and Crystal-Structure Determination" *Acta Crystallogr. Sect. A*
63. Sadabs (2008)
64. Smee, Hurst, Evans et al. (2017) "Evaluation of cell viability dyes in antiviral assays with RNA viruses that exhibit different cytopathogenic properties" *J. Virol. Meth*
65. Upadhyay, Chaudhary, De Oliveira et al. (2020) "Fluorinated Scaffolds for Antimalarial Drug Discovery" *Expert Opin. Drug Discov*
66. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724128&blobtype=pdf | # Polyvalent phage GSP004 recognizes O-antigen polysaccharide receptors in Salmonella and Escherichia coli through tail fiber protein ORF208
Dongyang Gao, Shenyu Pang, Yuanhang Zhao, Shunyuan Pan, Xiangyu Kong, Jun Song, Dongbo Sun
## Abstract
The majority of phages are capable of lysing only specific bacterial strains within a single species, and polyvalent phages with cross-genus lytic capability are relatively rare. In this study, we isolated a polyvalent phage, GSP004, from swine farm wastewater, which exhibited cross-genus lytic activity against multiple Salmonella serovars and Escherichia coli (E. coli) O157:H7. Morphological classification assigns GSP004 to the family Myoviridae within the order Caudovirales, while phylogenetic analysis of its genome identifies it as a member of the genus Kuttervirus in the fam ily Ackermannviridae. The periodate/proteinase K assays confirmed bacterial surface polysaccharides as the host receptor targeted by GSP004. Combined with lipopo lysaccharide (LPS) competitive adsorption assays, gene knockout strain spot/adsorp tion/dynamic lysis assays, and phage nucleic acid release experiments, we demonstrated that the O-antigen of LPS was the sole receptor for GSP004 to infect Salmonella and E. coli O157:H7, respectively. Subsequent characterization via protein competition adsorption, antibody blocking, and fluorescent labeling experiments identified the tail protein ORF208 as the receptor-binding protein (RBP) recognizing LPS O-antigens. Further studies revealed that phage GSP004 employs its tail protein ORF208 to recognize and bind to the LPS O-antigen of Salmonella and E. coli through distinct molecular mechanisms, thereby mediating cross-genus infection. This finding provides critical molecular insights into the interaction between polyvalent phages and their bacterial hosts.IMPORTANCE Elucidating the molecular mechanism of cross-genus host recognition in polyvalent phages will provide a critical theoretical foundation for the rational design of broad-host-range phages. However, research on the cross-genus host recognition mechanisms of polyvalent phages remains scarce. Here, we isolated a polyvalent phage GSP004, which serves as an exemplary model for investigating the interaction mechanisms between such polyvalent phages and their bacterial hosts. Our study elucidates the molecular basis underlying the capability of GSP004 to simultaneously infect Salmonella and E. coli O157:H7 across genera. This study provides crucial molecular evidence for understanding the evolutionary strategies by which phages expand their cross-genus host range and establishes a theoretical foundation for the rational design of broad-host-range phage therapeutics. KEYWORDS polyvalent phage, Salmonella, Escherichia coli, receptor, receptor-binding proteins A ntimicrobial resistance is a growing global health concern for both animal and public health, threatening to complicate the treatment of infections worldwide (1, 2). In 2021, it was estimated that 4.71 million deaths worldwide were due to antibiotic
resistance, including 1.14 million deaths directly attributable to antibiotic resistance (3). The number of antibiotic-resistant infections is growing at an alarming rate, making it imperative to find alternative therapies. Phage therapy is a potential candidate treatment, and there have been increasing reports of successful phage therapies (4,5). However, further understanding of the process by which phages infect bacteria will help optimize phage therapy.
Bacteriophages (phages) are viruses that specifically infect bacteria and are present in all environments where bacteria occur. The most common phages consist of doublestranded DNA genomes that are anchored to the head of an icosahedron attached to a contractile or non-contractile tail with fibers (6). The tail is a major structure in the early stages of the viral life cycle, including phage adsorption to the host first, followed by injection of its genome into the host cell. Usually, phage adsorption is divided into two binding steps, reversible binding and irreversible binding. For example, in T4-like phages (7), the baseplate-associated long-tailed fibers first reversibly bind to lipopolysaccharide (LPS) or OmpC, causing a conformational change in the baseplate, which results in the short-tailed fibers irreversibly binding to LPS. The signal generated by this process of irreversible binding is transmitted to the phage head, which then triggers the release of DNA. However, some phages can also use tail proteins to bind directly to a single receptor, e.g., Siphovirus 9NA and Podovirus P22 phages use their tail spike proteins to recognize the O-antigen receptor of LPS, which starts the infection process (8,9). The detailed mechanisms vary in different phages, but tailed phages use receptor-binding proteins (RBPs) at the distal end of the tail for phage-host interactions, where tail spikes, tail fibers or spike proteins function as RBPs by attaching the phage to the host-enco ded receptors (10). Phage receptors in Gram-negative bacteria typically involve outer membrane proteins (OMPs), LPS, capsular polysaccharides (CPS), pili, and flagella, and these unique surface structures ensure phage binding to the correct host (11). Phage host range depends largely on receptor availability and receptor structure, and analyzing how RBP interacts with its receptor is crucial to unraveling the mystery behind phage host specificity. Many mechanisms of how some monovalent phage RBPs interact with the host have been reported, such as P22, Mu, PNJ1809-36, and Bp7 (8,(12)(13)(14). However, there are few reports on the mechanisms of how polyvalent phages interact with the host.
In this study, a polyvalent phage, GSP004, capable of simultaneously infecting multiple Salmonella serovars and Escherichia coli (E. coli) O157:H7, was isolated. Genomic phylogenetic analysis revealed its classification within the Kuttervirus genus, demonstrat ing close evolutionary relationships with polyvalent phages BSP101, LPEK22, GG32, and SenALZ1. However, the mechanism of interaction between this type of phage and host bacteria of different genera remains unclear. Our investigation identified that GSP004 achieves dual-host adaptation through a singular tail protein ORF208. This protein specifically recognizes and binds to the LPS O-antigen of Salmonella and E. coli through distinct molecular mechanisms, thereby enabling cross-genus infection. This finding contributes to a deeper understanding of polyvalent phage infectivity for efficient phage applications.
## MATERIALS AND METHODS
## Bacterial strains
Salmonella and E. coli strains utilized in this study are detailed in Tables S5 andS6. Clinically isolated E. coli strains were initially isolated and screened using Eosin Methylene Blue agar (EMB; Oxoid, UK), while Salmonella spp. were initially isolated and screened using Xylose Lysine Desoxycholate agar (XLD; Oxoid, UK). The suspected target colonies obtained from the primary screening were subjected to PCR molecular identification using their specific primers (Table S7). Salmonella isolates confirmed by PCR were further subjected to serotype identification using commercial O and H antigen antisera (Tianrun Bio-Pharmaceutical, Ningbo, China) according to the White-Kauffmann-Le Minor scheme (15). For subsequent experiments, bacterial cultures were propagated in Luria-Bertani (LB) broth at 37°C with continuous shaking (200 rpm). In addition, the knockout and complementary strains used in this study were constructed from our previous study (Table S4).
## Isolation and purification of phages
S. Enteritidis SE006 (GenBank accession no. CP099973.1) ( 16) was used as the host for phage isolation. Wastewater samples were collected at a pig farm in Daqing, China. Phage isolation was performed using a double-layer plate method with slight modification as described previously (17). Briefly, the wastewater samples were centrifuged at 8,000 × g for 5 min at room temperature, and the supernatant was filtered using a 0.22-µm filter to ensure sterility. The filtrate was mixed with the S. Enteritidis SE006 and LB broth, then incubated at 37°C for 12 h to enrich the phage. After enrichment, the mixture was centrifuged and filtered again to obtain the phage-enriched filtrate. The gradient-diluted filtrate was added with the host bacteria to 5 mL LB soft agar (0.7% [wt/ vol] agar) and poured onto LB agar plates (1.5% [wt/vol] agar). After 12 h of incubation at 37°C, single plaques were picked and resuspended in SM buffer (100 mM NaCl, 10 mM MgSO 4 , 50 mM Tris-HCl, pH 7.5). The resuspension was mixed with fresh host bacteria and plated again, and three rounds of purification were performed to obtain a single phage. The final purified phage was stored at 4°C or -80°C (containing 25% glycerol protectant).
## Determination of host range and efficiency of plating
The host range of phage GSP004 was determined by the efficiency of plating (EOP) as described previously (18). In brief, freshly propagated phage GSP004 (10 9 PFU/mL) was serially diluted 10-fold (10 -3 to 10 -9 ) in SM buffer. Aliquots (10 µL) of each dilution were added dropwise to bacterial lawns of test strains. Following overnight incubation at 37°C, EOP was calculated based on the number of plaques formed (EOP, phage titer of test bacteria/phage titer of host bacteria).
## Transmission electron microscopy
Purified phage lysate (10⁹ PFU/mL) was prepared for TEM analysis. Subsequently, a drop of phage lysate was dropped onto a copper grid containing carbon (Carbon Type-B 200 mesh; Beijing Zhongjingkeyi Technology Co., Ltd., Beijing, China) and staining was performed with 2% (wt/vol) phosphotungstic acid (pH 6.5) as described previously (19). Images of phages were captured using transmission electron microscopy (H-7650, Hitachi, Tokyo, Japan) with an acceleration voltage of 100 kV.
## Genome analysis and phylogenetic analysis
The concentrated phage suspensions were treated with DNase I and RNase A to remove residual host-derived nucleic acids. Phage DNA was subsequently extracted using the Viral Genome DNA Extraction Kit (Omega Bio-Tek Inc., Doraville, GA, USA). Whole-genome sequencing of phage GSP004 was performed on the Illumina MiSeq platform (San Diego, CA, USA), with raw reads assembled using SPAdes v3. 15.2 (20). Genome annotation was conducted through the RAST server (http://rast.nmpdr.org/) and cross-verified via BLASTp analysis (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Circular genome visualization was generated using the Proksee platform (https://proksee.ca/). To determine the classification of phages, sequences belonging to different phage terminase large subunits were downloaded from the NCBI database based on the classification report of the International Committee on Taxonomy of Viruses (ICTV). A phylogenetic analysis of the terminase large subunit of phage GSP004 was then performed using MEGA 11 with the neighbor-joining method (21). Intergenomic similarity levels were calculated via VIRIDIC (http://rhea.icbm.uni-oldenburg.de/VIRI DIC) (22) and graphically optimized by Chiplot (https://www.chiplot.online/index.html).
Genome comparison of phage GSP004 was performed using EasyFig software (23). The presence of potential virulence and antibiotic resistance genes in the phage genome was detected by CARD (https://card.mcmaster.ca/analyze/rgi) (24). Manually verify the presence of lysogen-associated proteins in the phage genome using the already annotated phage genome.
## Identification of the receptor type of phage GSP004
The type of receptor targeted by the phage was determined by periodate (IO 4 -, destroying LPS) and proteinase K (destroying OMPs) treatment tests (25). After the bacteria were cultured to the logarithmic growth stage, 1 mL of the bacterial suspension was washed with PBS to remove metabolic byproducts. The bacterial pellet collected after centrifugation was resuspended in 1.5 mL of sodium acetate (50 mM; pH 5.2) or sodium acetate containing either 10 mM or 100 mM IO 4 -, followed by dark incubation at 37°C for 2 h. Alternatively, the bacterial pellet collected after centrifugation was resuspended in 10 or 20 mg/mL proteinase K solution, with control groups lacking proteinase K, and incubated at 37°C for 3 h. After treatment, the bacteria were washed again with PBS before determining the phage adsorption rates.
## Inhibitory effect of LPS on adsorption
LPS was extracted from S. Enteritidis SE006 or E. coli ATCC 35150 using the LPS Extraction Kit (Bestbio, Shanghai, China) for subsequent experiments. Bacterial strains were cultured to logarithmic phase, and the metabolites were washed with PBS. A 200-µL aliquot of a phage suspension (10 9 PFU/mL) was mixed with LPS solution at different concentrations (final concentrations: 10, 1, and 0.1 µg/mL), while the control group received buffer without LPS. The mixtures were incubated at 37°C for 15 min. Subsequently, 200 µL of a bacterial culture (10⁸ CFU/mL) was added to each mixture, followed by a further incubation at 37°C for 15 min. Finally, the mixtures were centrifuged at 12,000 × g for 1 min at 4°C. The supernatant was used to measure the plaque number and the adsorption percentages were calculated [1 -(phage titer of supernatant after cells were removed/phage titer of control reaction mixture without bacterial cells)] ×100%.
## Phage spotting assay
The receptors of host bacteria recognized by phage were further analyzed using constructed SE006 and ATCC 35150 gene knockout strains (Table S4). Ten-fold serial dilutions of phage lysates were prepared in SM buffer. For each strain, 10-µL aliquots of each dilution were spotted onto lawns of wild-type or knockout strains overlaid on double-layer LB agar plates. Plates were incubated at 37°C for 12-16 h to allow plaque development.
## Phage adsorption assay
Wild-type and knockout strains were cultured in LB medium at 37°C until the density reached 10 8 CFU/mL. Then, phage GSP004 was added according to the optimal MOI. The mixture was centrifuged at 12,000 × g for 1 min after the addition of phage for 10 min. The supernatant was collected, and phage titer was quantified by the double agar overlay method. Adsorption percentages were calculated.
## Dynamic lysis activity of phage against host bacteria
Wild-type, knockout strain, and complementary strains were cultured to the logarithmic growth stage. A mixture of 100 µL bacterial suspension and 100 µL phage suspension (at the optimal MOI) was added to a 96-well plate. For controls, 100 µL bacterial suspension was mixed with 100 µL LB medium (positive control), and sterile LB medium served as the negative control. The 96-well plates were incubated at 37°C by shaking in the Feyond-A300 Multi-function enzyme immunoassay analyzer (Allsheng, Hangzhou, China), and the OD 600 value was monitored every 30 min for 12 h.
## Fluorescence DNA ejection assay
The in vitro DNA ejection was detected using the Yo-Pro fluorescence assay, as previously described (8,26,27). Phage GSP004 (10 9 PFU/mL) was mixed with LPS (final concentra tion: 10 µg/mL) derived from either S. Enteritidis SE006 or E. coli ATCC 35150, followed by incubation at 37°C. Throughout the incubation period, the mixture was maintained in the presence of 1 µM Yo-Pro-1 Iodide, and the fluorescence intensity was measured at 60-s intervals for a total duration of 6,000 s. At the end of each experiment, DNase I (10 µg/mL) was added as a control for DNA accessibility.
## Protein expression and purification
The genes encoding ORF206, ORF207, ORF208, and ORF209 were amplified by specific primers ORF206-F/R, ORF207-F/R, ORF208-F/R, and ORF209-F/R, respectively (Table S3). The resulting PCR products and the pET28a vector were digested with EcoRI and XhoI restriction enzymes. The digested fragments were then ligated into the linear ized pET28a vector at the corresponding EcoRI/XhoI sites, generating the recombinant plasmids pET-28a-ORF206, pET-28a-ORF207, pET-28a-ORF208, and pET-28a-ORF209. To construct the EGFP-tagged recombinant plasmid pET-28a-ORF208-EGFP, the EGFP gene was amplified with primers ORF208-EGFP-F/R (Table S3). Both the EGFP amplicon and the pET-28a-ORF208 plasmid were digested with EcoRI and BamHI, followed by ligation at the corresponding sites. All recombinant plasmids were verified by DNA sequencing.
The confirmed plasmids were transformed into E. coli BL21 (DE3) competent cells. Recombinant protein expression was induced by adding 0.5 mM IPTG to the cultures followed by incubation at 16°C for over 16 h. Cells were harvested by centrifugation at 10,000 × g for 10 min at 4°C, resuspended in PBS, and lysed via ultrasonication. The lysate was then purified by Ni-NTA Beads 6FF gravity columns. Finally, the BCA protein quantitative kit (Beyotime Biotechnology, Shanghai, China) was used to determine its concentration.
## Protein competition assay
To test the ability of recombinant proteins to compete with phages for adsorption to host bacteria, we performed assays as previously described (14). First, 200-µL purified recombinant protein (2 mg/mL) was mixed with 200-µL bacteria (10 8 CFU/mL) and incubated at 37°C for 15 min. Subsequently, 100-µL phage (10 9 PFU/mL) was added to the mixture after incubation at 37°C for 15 min. Afterward, the mixture was centrifuged at 12,000 × g for 1 min. The number of phage spot formations in the supernatant was then determined, and adsorption percentages were calculated.
## Polyclonal antibody blocking assay
Polyclonal antibody against ORF208 was prepared by immunizing New Zealand White rabbits with the purified ORF208 protein, following the previously described method (14). To determine the blocking effect of antibodies on phage infectivity, a neutralization assay was performed. Anti-ORF208 antibody was co-incubated with phage GSP004 (10 9 PFU/mL) at 37°C for 20 min. Following the incubation, the residual infectivity of the phage was evaluated by determining its titer using the double-agar-layer plaque assay. All animal procedures were performed following the guidelines approved by the Animal Experiment Ethical Committee of Heilongjiang Bayi Agricultural University (approval no. DWKJXY2024038).
## Detection of fluorescent ORF208 binding to LPS
The detection method of ORF208 combined with LPS was slightly modified according to previous methods (12). To test whether the RBP of phage GSP004 could bind to the LPS extracted from S. Enteritidis SE006 or E. coli ATCC 35150, recombinant pET-28a-ORF208-EGFP vectors were constructed and proteins were then expressed and purified. For the assay, 200-µL purified ORF208-EGFP protein (2 mg/mL) was mixed with 200-µL wild-type strains (SE006 or ATCC 35150), knockout strains, or complementary strains (10 8 CFU/mL) and incubated at 37°C for 15 min. After incubation, the bacteria were collected by centrifugation at 3,000 × g for 10 min, washed with PBS three times, and observed by fluorescence microscope.
## Bioinformatics analysis of ORF208
The domain architecture of ORF208 was analyzed using the Basic Local Alignment Search Tool (BLAST), Inter-ProScan software (http://www.ebi.ac.uk/Tools/InterProScan), and the protein families (Pfam) database (http://pfam.xfam.org/search). The amino acid sequence alignments of ORF208 were performed using Clustal Omega and GeneDoc. The three-dimensional (3D) structure of ORF208 was predicted through the AlphaFold Protein Structure Database (https://alphafold.com).
## Degradation of LPS with ORF208 for SDS-PAGE analysis
To test whether ORF208 can degrade LPS O-antigen, an SDS-PAGE analysis was performed. Purified ORF208 protein (2 mg/mL) was mixed with an equal volume of LPS (40 µg/mL) extracted from either the SE006 or ATCC 35150 strains, and the mixture was incubated at 37°C for 12 h. The reaction was stopped by the addition of protei nase K, which was followed by incubation at 95°C for 10 min. LPS was loaded onto SDS-PAGE. Digestion products were shown by 15% SDS-PAGE silver staining (Thermo Fisher Scientific, Cleveland, OH, USA).
## Statistical analysis
All data are presented as the mean ± standard deviation (SD) of at least three independ ent experiments. Data were analyzed by one-way analysis of variance (ANOVA) followed by Dunnett's post hoc test for comparisons with a control group. All analyses were performed using OriginPro (OriginLab, Northampton, MA, USA). A P value below 0.05 was considered statistically significant, and significance levels in the figures are denoted by asterisks (*P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant).
## RESULTS
## Host range of phage GSP004
Phage GSP004 was initially isolated from wastewater samples using S. Enteritidis SE006 as the host strain. Analysis of the host range of GSP004 revealed broad-host-range lytic activity, targeting not only 47 out of 59 tested Salmonella strains (spanning multiple serotypes) but also 21 E. coli O157:H7 strains (Table S1). This intergeneric lytic activity demonstrates that GSP004 is a polyvalent phage with cross-genus infectivity.
## Morphological observation of phage GSP004
Plaque morphology observations revealed that GSP004 formed clear plaques with a diameter of approximately 1 mm (Fig. S1A). Electron microscopic examination (TEM) demonstrated that GSP004 possessed an icosahedral head (65 ± 1.6 nm) and a long contractile tail (166 ± 0.9 nm) as shown in Fig. S1B. Based on morphological characteris tics observed by TEM, phage GSP004 was classified into the order Caudovirales, family Myoviridae.
## General genomic features
The phage genome comprised a circular double-stranded DNA of 158,583 bp with a G + C content of 44.57% (GenBank accession no. PP533474.1) (Fig. 1A). The genome encoded 4 tRNA genes and 214 putative open reading frames (ORFs) (Table S8).
The phylogenetic analysis based on the terminase large subunit indicated that GSP004 belonged to the Kuttervirus genus (Fig. 1B). Consistent with phylogenetic analysis, nucleotide-level homology analysis revealed close relationships with other Kuttervirus phages (Fig. 1C), with particularly high average nucleotide identities observed for BSP101 (98.37%), LPEK22 (98.90%), GG32 (97.60%), and SenALZ1 (97.28%) (Fig. 1D), all of which were able to simultaneously infect Salmonella and E. coli O157:H7. In addition, the genome analysis did not reveal any antibiotic resistance, virulence, or lysogenyrelated genes, indicating that phage GSP004 could be a potential candidate for thera peutic applications.
## Phage receptor assay
The bacterial surface primarily consists of OMPs and LPS, which are commonly utilized by phages as receptors. In order to identify the receptor for GSP004 recognition of host bacteria, S. Enteritidis SE006 and E. coli ATCC 35150 were treated with protease K (destroying OMPs) or periodate (destroying LPS), followed by phage adsorption assays. The results showed that there was a marked reduction in adsorption rate for both bacterial strains following periodate treatment (Fig. 2A andB), indicating that LPS is essential for phage attachment. Furthermore, in competitive inhibition assays, LPS extracted from these two bacterial strains was able to effectively block the adsorption of phage GSP004 to its host (Fig. 2C). Collectively, these findings indicate that LPS serves as the critical receptor enabling GSP004 to infect both Salmonella and E. coli O157:H7.
To further validate LPS as a receptor for phage GSP004, we performed adsorption assays and plaque formation tests using SE006 and ATCC 35150 strains harboring deletions in the LPS biosynthesis genes rfaL and rfaC. Phage adsorption assays demon strated complete abolition of GSP004 binding to ΔrfaL and ΔrfaC mutants (Fig. 3A andB). Consistent with this, no plaques were observed following infection of bacterial lawns with these mutants (Fig. 3C andD). Moreover, in vitro growth kinetics revealed that phage GSP004 had no discernible impact on the proliferation of ΔrfaL or ΔrfaC cultures (Fig. 3E andF). These results suggest that LPS is indispensable for both phage adsorption and subsequent lytic activity.
The canonical LPS molecule consists of lipid A, core oligosaccharide, and O-antigen (Fig. 3G). Deletions in LPS biosynthesis genes produce truncated variants with distinct structural deficiencies. Specifically, ΔrfaL mutants retain an intact core oligosaccharide
## Phage GSP004 releases its DNA upon LPS incubation in vitro
To investigate whether LPS O-antigen as a binding receptor can trigger the release of phage DNA, we employed a fluorescent assay to monitor phage DNA ejection in vitro. The results demonstrated that incubation of phage GSP004 with LPS extracted from wild-type SE006 or ATCC 35150 strains elicited a time-dependent increase in dye fluorescence, plateauing after approximately 2,000 s at 37°C (Fig. 4). By contrast, no fluorescence enhancement was detected when phage GSP004 was incubated with LPS from O-antigen-deficient mutants (ΔrfaL). To validate that the fluorescence signal originated from phage DNA, DNase I was introduced post-signal saturation. This resulted in a rapid reversal of the fluorescence increase, confirming that the signal was attributa ble to phage-injected DNA. Collectively, these findings substantiate the LPS O-antigen as a specific binding receptor for phage GSP004. The above results further demonstrate that LPS O-antigen is a specific binding receptor for phage GSP004.
## Identification of receptor binding proteins of phage GSP004
To identify phage-encoded receptor-binding proteins (RBPs), we first conducted bioinformatics analyses on four tail-associated proteins (ORF206, ORF207, ORF208, and ORF209) encoded by phage GSP004 (Fig. S2). These proteins were cloned, expressed, and purified using a prokaryotic expression system (Fig. 5A; Fig. S2). Adsorption inhibition assays revealed that incubation of recombinant protein ORF208 with phage GSP004 significantly reduced phage adsorption to SE006 or ATCC 35150, whereas addition of LPS restored adsorption efficiency (Fig. 5B). In contrast, purified recombinant proteins ORF206, ORF207, or ORF209 had no detectable effect on phage adsorption. Furthermore, pretreatment of phage GSP004 with a polyclonal antibody against ORF208 resulted in a marked reduction in plaque formation upon infection of strains SE006 or ATCC 35150 (Fig. 5C). These results suggest that the tail protein ORF208 acts as an RBP during phage adsorption.
To validate ORF208-O-antigen interaction specificity, we performed fluorescence microscopy using EGFP-tagged ORF208 (ORF208-EGFP). The fusion protein exhibited strong binding to wild-type SE006 and ATCC 35150 strains, as well as their O-antigencomplemented strains (Fig. 5D). In contrast, fluorescence was not detected in O-antigendeficient mutants (ΔrfaL). These observations indicate that ORF208 specifically interacts with LPS O-antigen.
## Interaction of ORF208 proteins with LPS
Bioinformatic analysis revealed ORF208 as a 2,097 bp gene encoding a 698-amino-acid protein. Conserved domain prediction identified two characteristic motifs: an N-terminal domain of the phage G7C tail spike protein (located at residues 90-155) (30) and a C-terminal domain of the phage P22 tail spike protein (located at residues 161-697) (31) (Fig. 6A). Multiple sequence alignment demonstrated 32% and 51% homology between ORF208 and the tail spike proteins (TSPs) of phage P22 and 9NA, respectively (Fig. 6B). Notably, these reference proteins possess receptor-destroying endoglycosidase (endorhamnosidase) activity, enabling binding and hydrolysis of repetitive O-antigen structures in Salmonella LPS (9,32). Structural modeling of ORF208 using AlphaFold predicted a parallel homotrimeric conformation with a large β-helical domain, similar to P22 and 9NA TSPs (Fig. 6C). These structural and functional parallels strongly suggest ORF208 may share a conserved receptor recognition mechanism with P22 and 9NA TSPs.
To verify whether ORF208 protein has the same function, we assessed its ability to degrade LPS extracted from SE006 or ATCC 35150 strains using 15% silver-stained SDS-PAGE. The results demonstrated that ORF208 specifically hydrolyzed the long O-antigen chains in the LPS of the SE006 strain, as evidenced by the disappearance of long-chain bands and the concomitant accumulation of short-chain fragments (Fig. 6D). In contrast, no evident degradation of the long-chain O-antigen was observed in the LPS from the ATCC 35150 strain (Fig. 6E). These findings suggest that the functional properties of the ORF208 protein may differ from those of the TSPs of phages 9NA and P22.
## DISCUSSION
The high host specificity of phages represents a key feature distinguishing them from antibiotics, acting as a double-edged sword. On one hand, this high specificity enables phages to selectively kill target bacteria without disrupting commensal microbiota (33). On the other hand, it restricts their host range, thereby severely limiting their applicability (34). Consequently, phages capable of infecting a wide range of important target bacteria are considered ideal candidates for the development of phage antimi crobial agents (35). In this study, we report the isolation and infection mechanisms of GSP004, a polyvalent broad-host-range phage demonstrating infectivity against multiple Salmonella serovars and E. coli O157:H7. Morphological characterization via TEM revealed a canonical icosahedral capsid with a contractile tail, morphologically consistent with the Myoviridae family. Comparative genomic analysis, including whole-genome alignment and phylogenetic tree construction, classified GSP004 within the Kuttervirus genus. Phage GSP004 exhibited 98.37% average nucleotide identity with polyvalent Salmonella phage BSP101, 98.90% with phage LPEK22, 97.60% with phages GG32, and 97.28% with SenALZ1 (36-39) (Fig. 1D). Despite this genomic conservation, the molecular mecha nisms that determine its cross-generic host range (Salmonella spp. and E. coli O157:H7) remain unknown. Elucidating the mechanisms underlying the intergeneric infectivity of such polyvalent phages toward specific hosts is critical, as this knowledge will enable rational engineering of phage-based antimicrobials and targeted therapeutic biocide development (40).
Phage infection of host bacteria begins with the specific binding of phage-encoded RBPs to host surface receptors (41). The identification of phage-targeted receptors is the first step in the study of phage-host interactions (42). Potential receptors recognized by the phage include structures such as polysaccharides or OMPs exposed on the bacterial surface (43). Here, we demonstrate that the LPS O-antigen is a receptor necessary for GSP004 adsorption and successful infection of S. Enteritidis SE006 and E. coli ATCC 35150, which does not require any OMP receptor (Fig. 3). It has the same receptor as a similar phage, BSP101, which was previously reported but not studied in greater depth (39). Conventionally, phage adsorption processes require a combination of multiple receptors, including reversible and irreversible adsorption receptors (specific binding) (43,44). When a phage binds to a specific receptor, the phage then injects nucleic acid into the host cell, which is the beginning of phage-initiated infection. Employing a methodology established by Andres et al. (8) to trigger phage nucleic acid release in vitro using a specific receptor, we confirmed that the O-antigen alone could induce GSP004 nucleic acid ejection. This finding indicates that the O-antigen is the sole receptor for phage GSP004 infecting SE006 and ATCC 35150 host bacteria. As with reported phages P22 (8), CBA120 (45), and Det7 (27), they are typically O-antigen-specific phages that are unable to infect O-antigen-deficient bacterial strains as hosts and are also unable to release nucleic acid upon interaction with O-antigen-truncated LPSs.
Typically, O-antigen structures are highly diverse, while the core oligosaccharide structure is relatively conserved at the species level (28). As a result, phages utilizing O-antigens as receptor targets generally display a narrower host range, whereas those employing core oligosaccharide as receptors exhibit broader host range (43). However, certain O-antigen-targeting phages are able to recognize different receptors via multiple RBPs to broaden the host range. For instance, E. coli phage CBA120 encodes four tail spike proteins (TSPs) that recognize distinct O-antigens of E. coli and Salmonella, respectively. Similarly, phages SP6 and K1-5 possess two rotatable TSPs to facilitate alternative host recognition (46,47). Intriguingly, phage GSP004 maintains a broad host range despite relying on LPS O-antigen as its primary receptor. To elucidate the way phage GSP004 recognizes multiple hosts, phage RBP was identified.
The tail proteins of phages infecting Gram-negative bacteria frequently function as RBPs that mediate adsorption to LPS (48). Bioinformatics analysis revealed that phage GSP004 encodes four tail-associated proteins (ORF206, ORF207, ORF208, and ORF209). We initially hypothesized that at least two of these RBPs were involved in phage recognition of Salmonella and E. coli O-antigen receptors. However, our results show that the tail protein ORF208 alone binds specifically to the LPS O-antigen of SE006 and ATCC 35150 strains, suggesting that phage GSP004 utilizes a single RBP for receptor recognition.
Further analyses revealed that the ORF208 protein may have the same receptor recognition mechanism as the TSPs of phages P22 and 9NA (9,32). These phages recognize and hydrolyze the long O-antigen chains of LPS through their TSPs without requiring secondary receptors. However, the results of this study demonstrate that ORF208 exhibits hydrolytic activity only toward the long O-antigen chain in the LPS of SE006 strain, but not toward those in the LPS of ATCC 35150 strain (Fig. 6D andE). This discrepancy suggests that phage GSP004 may employ two distinct molecular mechanisms during the infection of Salmonella and E. coli. During the lysis of Salmonella, ORF208 may function similarly to the TSPs of P22/9NA phages by recognizing and hydrolyzing the long O-antigen chains of LPS to initiate infection. In contrast, during the infection of E. coli, ORF208 may initiate infection via an alternative mechanism. It has been reported that phage G7C specifically deacetylates the O-antigen of E. coli 4s while leaving the rest of its structure unchanged (30). Based on this, we hypothesize that in the process of infecting E. coli, ORF208 may initiate infection via a mechanism analogous to that of phage G7C, involving specific chemical modifications to the O-antigen rather than direct degradation.
In conclusion, this study demonstrates that the polyvalent phage GSP004 utilizes its tail protein ORF208 to recognize and bind to the LPS O-antigen of Salmonella and E. coli, respectively, but employs distinct molecular strategies to achieve cross-genus infection. These findings offer new insights into the diversity of molecular mechanisms underlying cross-genus infection by polyvalent phages.
## References
1. Salam, Al-Amin, Salam et al. (1946) "Antimicrobial resistance: a growing serious threat for global public health" *Healthcare (Basel)*
2. Ahmed, Hussein, Qurbani et al. (2024) "Antimicrobial resistance: Impacts, challenges, and future prospects" *J Med Surg Public Health*
3. Naghavi, Vollset, Ikuta et al. (2024) "Global burden of bacterial antimicrobial resistance 1990-2021: a systematic analysis with forecasts to 2050" *The Lancet*
4. Young, Lee, Shariyate et al. (2024) "Bacteriophage therapy and current delivery strategies for orthopedic infections: a scoping review" *J Infect*
5. Fang, Yin, He et al. (2024) "Safety and efficacy of phage application in bacterial decolonisa tion: a systematic review" *Lancet Microbe*
7. Van Raaij (2024) "Bacteriophage receptor recognition and nucleic acid transfer" *Subcell Biochem*
8. Washizaki, Yonesaki, Otsuka (2016) "Characterization of the interactions between Escherichia coli receptors, LPS and OmpC, and bacteriophage T4 long tail fibers" *Microbiologyopen*
9. Andres, Hanke, Baxa et al. (2010) "Tailspike interactions with lipopolysaccharide effect DNA ejection from phage P22 particles in vitro" *J Biol Chem*
10. Andres, Roske, Doering et al. (2012) "Tail morphology controls DNA release in two Salmonella phages with one lipopolysaccharide receptor recognition system" *Mol Microbiol*
11. Simpson, Sacher, Szymanski (2016) "Development of an assay for the identification of receptor binding proteins from bacteriophages" *Viruses*
12. Latka, Maciejewska, Majkowska-Skrobek et al. (2017) "Bacteriophage-encoded virion-associated enzymes to overcome the carbohydrate barriers during the infection process" *Appl Microbiol Biotechnol*
13. Gong, Wang, Huang et al. (2021) "Novel host recognition mechanism of the K1 capsule-specific phage of Escherichia coli: capsular polysaccharide as the first receptor and lipopolysaccharide as the secondary receptor" *J Virol*
14. North, Davidson (2021) "Phage proteins required for tail fiber assembly also bind specifically to the surface of host bacterial strains" *J Bacteriol*
15. Chen, Sun, Ren et al. (2020) "LamB, OmpC, and the core lipopolysaccharide of Escherichia coli K-12 function as receptors of bacteriophage Bp7" *J Virol*
16. Grimont, Weill (2007) "Antigenic formulae of the Salmonella serovars" *World Health Organization Collaborat ing Center for Reference and Research on Salmonella, Institut Pasteur*
17. Gao, Ji, Wang et al. (2022) "Fitness trade-offs in phage cocktail-resistant Salmonella enterica serovar Enteritidis results in increased antibiotic susceptibility and reduced virulence" *Microbiol Spectr*
18. Park, Park (2021) "Characterization and food application of the novel lytic phage BECP10: specifically recognizes the O-polysaccharide of Escherichia coli O157:H7" *Viruses*
19. Mirzaei, Nilsson (2015) "Correction: Isolation of phages for phage therapy: a comparison of spot tests and efficiency of plating analyses for determination of host range and efficacy" *PLoS One*
20. (2025) *Full-Length Text Journal of Virology*
21. Bigot, Lee, Mcintyre et al. (2011) "Control of Listeria monocytogenes growth in a ready-to-eat poultry product using a bacteriophage" *Food Microbiol*
22. Prjibelski, Antipov, Meleshko et al. (2020) "Using SPAdes de novo assembler" *Curr Protoc Bioinformatics*
23. Tamura, Stecher, Kumar (2021) "MEGA11: molecular evolutionary genetics analysis version 11" *Mol Biol Evol*
24. Moraru, Varsani, Kropinski (2020) "VIRIDIC-a novel tool to calculate the intergenomic similarities of prokaryote-infecting viruses" *Viruses*
25. Sullivan, Petty, Beatson (2011) "Easyfig: a genome comparison visualizer" *Bioinformatics*
26. Alcock, Huynh, Chalil et al. (2023) "CARD 2023: expanded curation, support for machine learning, and resistome prediction at the comprehensive antibiotic resistance database" *Nucleic Acids Res*
27. Kiljunen, Datta, Dentovskaya et al. (2011) "Identification of the lipopolysacchar ide core of Yersinia pestis and Yersinia pseudotuberculosis as the receptor for bacteriophage φA1122" *J Bacteriol*
28. Broeker, Kiele, Casjens et al. (2018) "In vitro studies of lipopolysaccharide-mediated DNA release of podovirus HK620" *Viruses*
29. Broeker, Roske, Valleriani et al. (2019) "Time-resolved DNA release from an Oantigen-specific Salmonella bacteriophage with a contractile tail" *J Biol Chem*
30. Bertani, Ruiz (2018) "Function and biogenesis of lipopolysaccharides" *EcoSal Plus*
31. Ishiwa, Komano (2003) "Thin pilus PilV adhesins of plasmid R64 recognize specific structures of the lipopolysaccharide molecules of recipient cells" *J Bacteriol*
32. Prokhorov, Riccio, Zdorovenko et al. (2017) "Function of bacteriophage G7C esterase tailspike in host cell adsorption" *Mol Microbiol*
33. Steinbacher, Baxa, Miller et al. (1996) "Crystal structure of phage P22 tailspike protein complexed with Salmonella sp. O-antigen receptors" *Proc Natl Acad Sci*
34. Steinbacher, Miller, Baxa et al. (1997) "Phage P22 tailspike protein: crystal structure of the headbinding domain at 2.3 A, fully refined structure of the endorhamnosi dase at 1.56 A resolution, and the molecular basis of O-antigen recognition and cleavage" *J Mol Biol*
35. Kingwell (2015) "Bacteriophage therapies re-enter clinical trials" *Nat Rev Drug Discov*
36. Ando, Lemire, Pires et al. (2015) "Engineering modular viral scaffolds for targeted bacterial population editing" *Cell Syst*
37. Merabishvili, Pirnay, Vos (2024) "Guidelines to compose an ideal bacteriophage cocktail" *Methods Mol Biol*
38. Zhang, Zou, Mds et al. (2023) "Combine thermal processing with polyvalent phage LPEK22 to prevent the Escherichia coli and Salmonella enterica contamination in food" *Food Res Int*
39. Chae, Kwon, Lee et al. (2016) "Genome sequence of bacteriophage GG32, which can infect both Salmonella enterica serovar Typhimurium and Escherichia coli"
40. *Genome Announc*
41. Chen, Guan, Liu et al. (2019) "Characterization and complete genomic analysis of two Salmonella phages, SenALZ1 and SenASZ3, new members of the genus Cba120virus" *Arch Virol*
42. Bai, Jeon, Ryu (2019) "Effective inhibition of Salmonella Typhimurium in fresh produce by a phage cocktail targeting multiple host receptors" *Food Microbiol*
43. Ge, Ye, Cai et al. (2023) "Efficient screening of adsorbed receptors for Salmonella phage LP31 and identification of receptor-binding protein" *Microbiol Spectr*
44. Casey, Mcdonnell, White et al. (2022) "Needle in a whey-stack: PhRACS as a discovery tool for unknown phage-host combinations" *mBio*
45. De Jonge, Meijenfeldt, Costa et al. (2020) "Adsorption sequencing as a rapid method to link environmental bacteriophages to hosts"
46. Rakhuba, Kolomiets, Dey et al. (2010) "Bacteriophage receptors, mechanisms of phage adsorption and penetration into host cell" *Pol J Microbiol*
47. Silva, Storms, Sauvageau (2016) "Host receptors for bacteriophage adsorption" *FEMS Microbiol Lett*
48. Plattner, Shneider, Arbatsky et al. (2019) "Structure and function of the branched receptor-binding complex of bacteriophage CBA120" *J Mol Biol*
49. Tu, Park, Morado et al. (2017) "Dual host specificity of phage SP6 is facilitated by tailspike rotation" *Virology (Auckl)*
50. Scholl, Rogers, Adhya et al. (2001) "Bacteriophage K1-5 encodes two different tail fiber proteins, allowing it to infect and replicate on both K1 and K5 strains of Escherichia coli" *J Virol*
51. Sun, You, Li et al. (2023) "Variants of a putative baseplate wedge protein extend the host range of Pseudomonas phage K8" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC11871800&blobtype=pdf | # JN.1 variants circulating in Italy from October 2023 to April 2024: genetic diversity and immune recognition
Emanuela Giombini, Ilaria Schiavoni, † Luigina, Alessandra Presti, Angela Martino, Stefano Fiore, Pasqualina Leone, Francesca Fortunato, Rosa Prato, Giorgio Fedele, Anna Palamara, Paola Stefanelli
## Abstract
Background The continuous emergence of SARS-CoV-2 variants and subvariants poses significant public health challenges. The latest designated subvariant JN.1, with all its descendants, shows more than 30 mutations in the spike gene. JN.1 has raised concerns due to its genomic diversity and its potential to enhance transmissibility and immune evasion. This study aims to analyse the molecular characteristics of JN.1-related lineages (JN.1*) identified in Italy from October 2023 to April 2024 and to evaluate the neutralization activity against JN.1 of a subsample of sera from individuals vaccinated with XBB.1.5 mRNA.
MethodsThe genomic diversity of the spike gene of 794 JN.1* strain was evaluated and phylogenetic analysis was conducted to compare the distance to XBB.1.5. Moreover, serum neutralization assays were performed on a subsample of 19 healthcare workers (HCWs) vaccinated with the monovalent XBB.1.5 mRNA booster to assess neutralizing capacity against JN.1.
ResultsSequence analysis displayed high spike variability between JN.1* and phylogenetic investigation confirmed a substantial differentiation between JN.1* and XBB.1.5 spike regions with 29 shared mutations, of which 17 were located within the RBD region. Pre-booster neutralization activity against JN.1 was observed in 42% of HCWs sera, increasing significantly post-booster, with all HCWs showing neutralization capacity three months after vaccination. A significant correlation was found between anti-trimeric Spike IgG levels and neutralizing titers against JN.1.
ConclusionsThe study highlights the variability of JN.1* in Italy. Results on a subsample of sera from HCWs vaccinated with XBB.1.5 mRNA booster vaccine suggested enhanced neutralization activity against JN.1.
and potential immune escape, pose significant challenges to public health efforts and vaccine efficacy [3,4]. One of the latest Omicron subvariant identified in 2023 was JN.1, which has emerged as a variant of interest due to its phylogenetic distance from previously dominant strains, such as the XBB.1.5 recombinant [5][6][7]. In fact, JN.1, firstly identified in August 2023, contains more than 30 mutations in the spike (S) protein coding gene [5]. For this reason, increased transmissibility and immune escape was hypothesized [5].
Hereby, the Italian JN.1* sequences identified from October 2023 to April 2024 were analysed, to evaluate the variability of this strain. Moreover, a subsample of sera collected from healthcare workers (HCW) boosted with the monovalent Omicron XBB.1.5-containing COVID-19 mRNA vaccines, was used to evaluate the serum neutralization capacity against JN.1.
## Methods
## Molecular analysis
## Sequence collection
In Italy, the National Institute of Health (Istituto Superiore di Sanità, ISS), in collaboration with the Ministry of Health (MoH) coordinates the genomic surveillance of SARS-CoV-2 and its variants. The network of collaborating laboratories includes more than 70 laboratories distributed throughout the country. Whole Genome Sequencing (WGS) data are routinely shared nationally and internationally through the "Italian COVID-19 Genomic" (I-Co-Gen platform, hosted at ISS) and "Global Initiative on Sharing All Influenza Data" (GISAID)databases (https:// gisaid. org/), respectively. The sequences uploaded to the national collaborative repository I-Co-Gen are automatically analysed and verified for completeness and timeliness after a quality review, ensuring a continuous flow of information for the early identification of new variants and/or mutations and for estimating their prevalence.
To the purpose of this study, 1,832 sequences belonging to JN.1-related lineages (JN.1*), obtained from October 2023 to April 2024, were downloaded from GISAID (last access 16/05/2024). The ID numbers of each sequence is listed in supplementary Table 4.
In addition, the JN.1 spike mutations were compared with those found among 1,835 Italian sequences belonging to XBB.1.5 and deposited in GISAID. Lineages were assigned using nextclade lineage V.3.2.1, as in the I-Co-Gen platform (last access16/05/2024).
## Sequences cleaning
All JN.1* and XBB.1.5 genomes were aligned separately to the reference 'Wuhan-Hu-1' (NC_045512.2) using MAFFT V.7.520 [8]. A python homemade program was developed to cut the spike gene without considering sporadic insertion and exclude the sequences with a spike coverage < 90%. After cleaning, 794 and 1,568 genomes were obtained for JN.1* and XBB.1.5, respectively. The two groups of spike sequences were aligned separately and together using MACSE V.2.07 [9]. The results were manually curated using BioEdit V.7.7.1 (https:// thall jisci ence. github. io/) to exclude alignment artefacts. To realise the phylogenetic tree sequences were clustered with 100% identity using .
## Phylogenetic tree
Maximum likelihood phylogenetic tree was obtained by running IQTREE V. 1.6.9 [11] on the overall multisequence alignment of Spike region of JN.1* and XBB.1.5, including the reference 'Wuhan-Hu-1' (NC_045512.2), with 1000 bootstrap replications. The tree was rooted versus the 'Wuhan-Hu-1' reference and the best substitution models was defined by ModelFinder (TIM + F + R4) included in IQTREE. The tree visualization was curated and visualised with FigTree V. 1.4.4.
## Sequence variability
The spike nucleotide and amino acid substitutions were identified using a homemade program. Signature mutations, present in the JN.1* and XBB.1.5 lineages, were identified using the Outbreak.info project (https:// outbr eak. info/) with a minimum threshold fixed at 75%. The Heatmap was constructed with the python mathplotlib library considering only the identified minor variants and excluding the JN.1* spike mutations identified as singleton. The python code was assisted by the Gemini 1.5 Flash by Google AI and OpenAI (2024), ChatGPT (Version GPT-4). The evolutionary divergence analysis within and between two sequences groups (XBB.1.5 and JN.1*) was conducted using the Maximum Composite Likelihood model included in MEGA11 [12] using default options.
## Serological analysis Study design and population
To the purpose of this study, samples from a subgroup of healthcare workers (HCW) enrolled in a larger multicentre longitudinal cohort study designed to monitor immune responses in individuals vaccinated with the monovalent Omicron XBB.1.5-containing COVID-19 mRNA vaccine were used. Specifically, serum samples were collected from 19 HCW enrolled at the Policlinico Riuniti University Hospital (Foggia, Italy) at the time of their vaccination with the Comirnaty Omicron XBB.1.5 vaccine (Pfizer-BioNTech) (T0) and 3 months after vaccine administration (T1). At the time of enrollment, demographic data were collected from each participant, along with information regarding previous SARS-CoV-2 infections and COVID-19 vaccinations.
## Serum preparation and storage
Blood samples (5 ml) were collected in Serum Separator Tubes (BD Diagnostic Systems, Franklin Lakes, NJ, USA) and centrifuged at room temperature at 1600 rpm for 10 min. Two serum aliquots were transferred to 2 ml polypropylene, screw cap cryo tubes (Nunc ™ , Thermofisher Scientific, Waltham, MA USA), immediately frozen at -20 °C and thereafter stored at -80°C. Frozen sera were shipped to the Department of Infectious Diseases at Istituto Superiore di Sanità (DMI, ISS), in dry ice following biosafety shipment condition. Upon arrival serum samples were immediately stored at -80°C.
## SARS-CoV-2 IgG immunoassays
Sera were evaluated using the DiaSorin Liaison SARS-CoV-2 trimeric Spike IgG assay on the LIAISON ® XL chemiluminescence analyzer (DiaSorin, Saluggia, VC, Italy). The assay range is up to 2080 Binding Antibody Units (BAU/mL). According to manufacturer's instructions, values ≥ 33.8 BAU/mL were interpreted as positive. If the results were above the assay range, samples were automatically diluted 1/20 and testing was repeated. Anti-Nucleocapsid IgG were measured by Anti-SARS-CoV-2 NCP ELISA assay (Euroimmun, Lübeck, Germany), which uses a modified nucleocapsid protein that only contains diagnostically relevant epitopes.
## SARS-CoV-2 neutralizing antibody assay
SARS-CoV-2 strain JN.1 (BA.2.86.1.1) lineage (EPI_ ISL_18624895) was incubated with two-fold serial dilutions of serum samples starting at 1:8 dilution in D-MEM culture medium (Sigma Aldrich, Merck Life Science, Milan, Italy) supplemented with 1X penicillin/streptomycin (Corning, Glendale, AZ, USA) and 2% foetal bovine serum (Corning) in 96-well plates. Virus (100 TCID50) and serum mixture was incubated at 37 °C for 1 h. After this incubation 10,000 cells (Vero/TMPRSS2-ACE2) per well were added and incubated at 37 °C for 5 days. The neutralization titer was calculated and expressed as microneutralization titer 50 (MNT50), i.e., the serum dilution capable of reducing the cytopathic effect to 50%.
## Results
In order to assess the variability of JN.1 and its sub-lineages (JN.1*) identified in Italy, 794 sequences were used for analysis, based on the complete coverage of the spike gene.
The lineage composition of the cleaned dataset and the trend of the total Italian SARS-CoV-2 vs JN.1* sequences, as present in GISAID, are shown in Fig. 1.
A total of 176 lineages were identified; among them, the parental lineage (JN.1) was detected until week 13-2024 with a median percentage of 52.3% (range 25%-100%). The number of JN.1* sequences, as well as lineage diversity, increased rapidly from week 45-2023 on. During the observation period, JN.1* did not represent the entirety of sequenced genomes. The fluctuations in the number of sequences loaded and analysed were similar to those downloaded from the GISAID dataset. Similarly, the cumulative number of SNPs identified within the spike gene varied over time, with the largest number of mutations identified in the receptor-binding domain (RBD-20,089 mutations vs 16,866 and 12,119 of N-terminal and C-terminal regions respectively. Figure 2A). Missense substitutions were the most common mutation type encountered (81.3%), followed by synonymous mutations (9.6%), insertions and deletions (1.2% and 7.9%, respectively). The complete list of mutations, excluded those considered singleton, is available in Table S1.
The spike amino acid substitutions identified in the Italian dataset were compared to those reported in Outbreak.info to verify the presence of unique sites. A closer analysis of the non-synonymous substitutions revealed that none of the 48 unique mutations was predominant, with a weekly frequency not exceeding 17.9% (K147N, Fig. 2B). Furthermore, only one spike mutation (T572I), located within the C-terminal domain, was persistent throughout the entire period of analysis, although with a low percentage (range: 0%-9.1%).
The similarity between JN.1* and XBB.1.5 spike sequences was analyzed. The identified XBB.1.5 mutations are reported in supplementary Table 2.
The global predominant mutations of JN.1* and XBB.1.5 were compared and 29 mutations were found to be shared; of these, most were identified within the RBD domain (17/29), similarly to what was observed between the JN.1* and XBB.1.5 Italian sequences (supplementary Table 3).
The average pairwise distance between the sequences within the Italian JN.1* group and the Italian XBB.1.5 group was estimated to be 8.8 × 10 -4 and 6.2 × 10 -4 , respectively while the distance between the two groups was equal to 1.1 × 10 -2 . These values were congruent with the hierarchy shown in the phylogenetic tree (Fig. 3).
In order to investigate the ability of individuals vaccinated with XBB.1.5 to neutralize the newly spreading JN.1 variant, serum samples were collected from 19 healthcare workers (HCW) at the time of the monovalent XBB.1.5 mRNA vaccine booster and three months later.
Detailed demographic data of study participants are summarized in Table 1.
All participants had previously received the two-dose primary COVID-19 vaccine cycle and the first mRNA booster, 12 (63%) had received 5 vaccine doses, including a second booster dose with the bivalent Comirnaty Original/Omicron BA.4/5 mRNA vaccine (Pfizer-BioN-Tech), and 7 (37%) had received 4 vaccine doses, since they skipped the bivalent mRNA booster. Additionally, 63% had at least one documented SARS-CoV-2 infection (Table 1). To avoid the influence of immunity induced by recent exposure to SARS-CoV-2, only individuals who tested negative for anti-Nucleocapsid (N) IgG antibodies at the time of vaccination (T0) were included in the study.
As shown in Fig. 4A, 42% of HCW's sera (n = 8) exhibited neutralizing activity against the JN.1 variant prior to receiving the booster (T0); the median MNT50 value was 6 [IQR: [1][2][3][4][5][6][7][8][9][10]. Three months post-vaccination, the neutralizing ability increased significantly, with all 19 sera examined able to neutralize the JN.1 strain (median MNT50, 32 [IQR: ).
Analyzing MNT data based on previous vaccination history, we found that 50% of HCW, who had received five doses of SARS-CoV-2 vaccine, including the bivalent Comirnaty Original/Omicron BA.4/5 mRNA vaccine as the second booster and the Comirnaty Omicron XBB.1.5 mRNA vaccine as the third booster, were able to neutralize JN.1 at T0. In contrast, only 28.6% (2 out of 7) of HCW who received XBB.1.5 vaccine as a fourth vaccine dose were able to neutralize JN.1 at T0 (Fig. 4B). To determine whether any HCWs had become infected with SARS-CoV-2 between the two time points, we tested T1 sera for anti-N IgG. Positive anti-N IgG titres were found in four individuals (Fig. 4A, grey dots). As shown in Fig. 4A, this did not translate in significant variations in the neutralizing ability of JN.1.
We investigated the correlation between Spike-specific IgG in serum and MNT against JN.1 using a standardized CLIA assay measuring IgG towards the original Spike protein (Wuhan) in its native trimeric form. A statistically significant correlation was found between trimeric anti-S IgG levels and MNT at T0 (data not shown) and at T1 (Fig. 4C). This suggests that higher anti-S IgG titres correspond to broader immunity and confer greater neutralizing ability.
## Discussion
This study presents a comprehensive analysis of the spike gene of the JN.1 lineage and its sub-lineages, identified in Italy from October 2023 to April 2024, the period in which JN.1 spread across the country. The parental JN.1 lineage persisted throughout the observation period with an increase in JN.1* sequences from week 45-2023.
In addition, the presence of other co-circulating variants underlined the virus variability. This is consistent with global trends in the evolution of SARS-CoV-2, which has improved fitness and selective adaptation [13].
To assess the molecular variability of JN.1 * , single nucleotide polymorphisms (SNPs) within the spike gene were investigated. The analysis showed a temporal variation in the mutation counts, with missense mutations predominating (81.3%), followed by synonymous (9.6%) and indels (1.2% and 7.9%, insertion and deletion, respectively) mutations. Farkas et al. [14] have argued that the predominant number of missense mutations compared to synonymous and indels mutations may reflect a higher viral fitness.
Signature mutations with a frequency > 20% were identified in the spike gene. Among the minority mutations in our dataset, only the amino acid substitution T572I showed a certain persistence, despite its low frequency (weekly range: 0%-9.1%). This suggests a potentially important function, as already described by Li et al. [15]. The T572I substitution increased the binding affinity to ACE2 receptors of various strains, potentially increasing host-susceptibility to infection [15].
Among the 69 signature spike mutations in JN.1, 29 were also identified in XBB.1.5.
The highest number of mutations was observed in the RBD region of the overall JN.1* Italian sequences (20,089 mutations vs 16,866 and 12,119 in N-and C-terminal regions, respectively), corroborating the key role of this region in the interaction with the host ACE2 receptor and in virus evolution. Therefore, to assess the evolutionary relationship between the JN.1* and XBB.1.5 lineages, a maximum likelihood phylogenetic analysis was performed by comparing the sequences of the Italian spike gene. The resulting phylogenetic tree, built on the spike gene, suggests a clear separation between the two lineages, (https:// www. who. int/ docs/ defau lt-source/ coron aviru se/ 21112 023_ ba.2. 86_ ire. pdf? sfvrsn= 8876d ef1_3), as also supported, in this study by a significant bootstrap (bootstrap > 90%). This result is consistent with the pairwise distances observed within the JN.1 and XBB.1.5 groups (8.8 × 10 -4 and 6.2 × 10 -4 , respectively), which are less than the intergroup distance (1.1 × 10 -2 ). This shows substantial genetic differentiation between the two However, the RBD of JN.1* and that of XBB.1.5 showed a high number of shared mutations (17/29), although this region is substantially shorter than the other two (N-and C-terminal), highlighting the crucial role of this domain for ACE2 binding [6].
Evaluation of the neutralizing activity of the sera, prior to the booster dose with the monovalent XBB.1.5 mRNA vaccine, found that 42% of the HCWs had neutralizing activity against JN.1. This pre-booster neutralizing capacity likely reflects residual immunity from previous vaccinations and /or natural infections. In line with this hypothesis, the data suggest that the HCWs who had Fig. 3 JN.1* and XBB.1.5 phylogenetic tree. Phylogenetic analysis of the spike gene of the Italian JN.1* (red) vs XBB.1.5 (blue) using a maximum likelihood approach with 1,000 replications. Only the principal nodes with bootstrap > 90% are indicated with green points received a fourth vaccine dose 12 months prior to enrolment exhibited a higher neutralization activity against JN.1 at baseline T0 than those who had received only three vaccine doses. The significant increase in neutralizing activity three months post-booster underlines the efficacy of the XBB.1.5 mRNA vaccine in enhancing the immune response against JN.1. This finding is in line with previous studies indicating that booster doses can increase neutralizing antibody titres, even against phylogenetically distant variants, providing a broader antibody repertoire response [16][17][18].
The detection of positive anti-N IgG titres in four individuals at T1 suggests that breakthrough infections occurred between the two time points. However, there was no significant increase in neutralizing immunity compared to those who were anti-N IgG negative. The possibility of not seeing an increase in neutralising effect could be related to a threshold effect. It may be speculated that there is a limit or plateau in the immune response, where further breakthrough infections do not significantly enhance neutralizing antibodies. Several factors could contribute to this effect, such as immune system saturation, vaccine-induced immunity, variant specificity, and individual variability [19,20]. A significant correlation was found between anti-trimeric S IgG levels and neutralizing titers against JN.1, both at T0 and T1. This indicates that higher levels of spikespecific IgG, directed towards the spike protein in its native trimeric form, are predictive of stronger neutralizing responses against SARS-CoV-2 variants. The correlation of spike-specific IgG levels with neutralizing titers might suggest their use as potential markers of immune response. These results are consistent with other studies that have shown higher anti-S IgG titres correlate with higher neutralizing capacity and broader immunity [21,22].
However, the sample size of the sera was relatively small and the study population was limited to healthcare workers, considered to be at high risk of exposure. Moreover, the observed immune responses could be influenced by various factors, such as the timing and nature of prior infections and vaccinations. The lack of detailed information regarding the exact antigenic history represents a limitation to the evaluation of the results. Furthermore, the results focused only on neutralizing antibody responses, which are one of the components of the entire immune response. T-cell responses and other aspects of the immune system also play a key role in protection against SARS-CoV-2 virus and its variants [23].
## Conclusions
In conclusion, this study provides a detailed molecular analysis of the gene encoding the spike protein and, in particular, of the RBD region of JN.1 lineage and its sublineages (JN.1*) circulating in Italy from October 2023 to April 2024.
Moreover, the spike gene of XBB.1.5 is very different from that of JN.1, as demonstrated by the phylogenetic tree and pairwise diversity and confirmed also by the presence of 17 out of 29 mutations shared by the two strains in the RBD region.
The RBD of the SARS-CoV-2 spike glycoprotein mediates viral attachment to the ACE2 receptor and is a major determinant of host range and a dominant target of neutralizing antibodies. Therefore, the neutralising ability of sera after booster vaccination with the monovalent XBB1.5 mRNA vaccine was investigated. Overall, the results obtained show that the XBB.1.5 mRNA booster vaccine significantly improves the neutralizing activity against the JN.1 variant, suggesting that, despite the differences between the two viral strains, a conserved antigenic epitope may be present on the RBD and that it remains unchanged.
The correlation between anti-Spike IgG levels and neutralizing capacity underlines the potential of these antibodies as markers of the immunity response after vaccination and/or previous viral infection.
These findings highlight the importance of monitoring viral mutations of SARS-CoV-2 variants and performing in vitro experiments to assess the neutralization capacity of immunized sera, as well as to conduct further research to explore the role of other immune components against SARS-CoV-2 virus.
## References
1. Savini, Cammà, Possenti "Istituto Zooprofilattico Sperimentale dell' Abruzzo e del Molise "Giuseppe Caporale"
2. Dell "Edera (Medical Genetics Unit" *Hospital*
3. Picerno, Lopizzo "Clinical Pathology and Microbiology Unit, AOR San Carlo"
4. Teresa "Rosaria Oteri (Unit of Microbiology and Virology"
5. "Genomics and Molecular Pathology"
6. Greco, Microbiology, Unit *Annunziata" Hospital of Cosenza*
7. Limone, Fusco
8. Tiberio, Atripaldi "Mariagrazia Coppola (UOC Microbiologia e Virologia"
9. Cacchiarelli, Grimaldi
10. Pongolini, Scaltriti "Risk Analysis and Genomic Epidemiology Unit, Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia-Romagna (IZSLER) "Bruno Ubertini"
11. Sambri, Dirani, Zannoli
12. Lazzarotto "Giada Rossini (Microbiology Unit"
13. Baldan, Lombino
14. Pierlanfranco D' Agaro, Segat
15. Barbone, Koncan
16. Battisti, Alba "Istituto Zooprofilattico Sperimentale del Lazio e della Toscana (IZSLT)"
17. Teresa
18. Angeletti, Riva
19. Pimpinelli, Microbiology, Virology et al.
20. Fanciulli
21. Massacci *Biostatistics, Bioinformatics and Clinical Trial Center*
22. Sanguinetti "Dipartimento di Scienze di Laboratorio e Infettivologiche"
23. Maggi, Rueca *Lazzaro Spallanzani" (IRCCS)*
24. *Policlinico Umberto I*
25. Federico *Microbiology and Diagnostic Immunology*
26. Ceccherini-Silberstein
27. Bruzzone, Icardi, Orsi "Hygiene Unit, San Martino Policlinico Hospital-IRCCS for Oncology and Neurosciences"
28. Valaperta
29. Oggionni
30. Testa, Sagradi
31. Caruso
32. Messali "Laboratory of Microbiology and Virology"
33. Malandrin "Annalisa Cavallero (Microbiology and Virology Unit"
34. *ASST Papa Giovanni*
35. Ceriotti, Colonia Uceda, Renteria et al. *UOC Laboratorio Analisi Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico*
36. Maria, Blasio, Torresani *IRCCS Istituto Auxologico Italiano*
37. Maria, Boniotti, Bertasio
38. "Laboratory of Microbiology and Virology, IRCCS Ospedale San Raffaele"
39. "IRCCS San Raffaele Hospital"
40. Novazzi, Mancini *ASST Sette Laghi*
41. Maria, Gismondo, Micheli *Virology and Bioemergencies*
42. Baldanti, Microbiology, Department "Fondazione IRCCS Policlinico San Matteo"
43. Federica, Giardina
44. Piralla, Zavaglio "Francesca Rovida (Microbiology and Virology Department, Fondazione IRCCS Policlinico San Matteo"
45. Pariani, Galli, Pellegrinelli
46. Menzo
47. Scutellà, Felice
48. Pagani, Bianconi "Angela Maria Di Pierro (Laboratory of Microbiology and Virology, Provincial Hospital of Bolzano, SABES-ASDAA; Lehrkrankenhaus der Paracelsus Medizinischen Privatuniversität"
49. Collini, Lorenzin "Laboratory of Microbiology and Virology"
50. *Amedeo di Savoia Hospital*
51. Gilardi, Bartolini, Cantarella
52. Peletto, Ru, Pier et al.
53. Chironna, Loconsole "Hygiene Unit, Interdisciplinary Department of Medicine-DIM, University of Bari "Aldo Moro"
54. Parisi
55. Arena "Rossella De Nittis, Giuseppina Iannelli (Microbiology and Virology"
56. "Ferdinando Coghe (Laboratorio Generale (HUB) Analisi Chimico Cliniche e Microbiologia, PO "Duilio Casula" *Army Medical Center*
57. Uzzau, Rubino, Angioj et al.
58. Piras, Mameli, Asproni (laboratorio et al.
59. Di Gaudio, D' Alessandro "Department of Health Promotion, Mother and Child Care, Internal Medicine and medical Specialties"
60. Vullo, Reale
61. "Teresa Pollicino (Division of Advanced Diagnostic Laboratories, University Hospital"
62. Vitale, Tramuto, D' Alessandro "Department of Health Promotion, Mother and Child Care, Internal Medicine and Medical Specialties"
63. Stefani, Clinical, Laboratory et al. *Hospital*
64. Scalia, Policlinico, Rodolico et al.
65. Ilenia Palermo, Policlinico, Rodolico et al.
66. Mancuso, Martino
67. Bramanti, Fidone, Barrano *ASP Ragusa*
68. Maria "Microbiology and Virology Unit"
69. "Francesca Malentacchi (Microbiology and Virology Unit"
70. Grazia
71. Mencacci, Camilloni
72. Terregino, Fusaro, Monne et al. "Division of Comparative Biomedical Sciences, Istituto Zooprofilattico Sperimentale delle Venezie"
73. Gibellini, Tonon, Cecchetto
74. Squarzon "Mosè Favarato (Molecular Diagnostics and Genetics, AULSS 3 Serenissima"
75. Biscaro, Vian "Silvia Ragolia (UOC Microbiology Treviso Hospital, Department of specialist and laboratory medicine, AULSS 2 Marca"
76. Pascarella, Buffoli *Molecular Biology applied to Microbiology and Virology*
77. Cai, Zhang, Xiao et al. (2021) "Structural basis for enhanced infectivity and immune evasion of SARS-CoV-2 variants" *Science*
78. Rochman, Wolf, Faure et al. (2021) "Ongoing global and regional adaptive evolution of SARS-CoV-2" *Proc Natl Acad Sci*
79. Washington, Gangavarapu, Zeller et al. (2021) "Emergence and rapid transmission of SARS-CoV-2 B.1.1.7 in the United States" *Cell*
80. Ulrichs, Rolland, Wu et al. (2024) "Changing epidemiology of COVID-19: potential future impact on vaccines and vaccination strategies" *Expert Rev Vaccines*
81. Kaku, Okumura, Padilla-Blanco et al. (2024) "Virological characteristics of the SARS-CoV-2 JN.1 variant" *Lancet Infect Dis*
82. Yao, Zhang, Duan et al. (2024) "Molecular insights into the adaptive evolution of SARS-CoV-2 spike protein" *J Infect*
83. Jeworowski, Mühlemann, Walper et al. (2023) "Humoral immune escape by current SARS-CoV-2 variants BA.2.86 and J.N1" *Euro surveill*
84. Katoh, Standley (2016) "A simple method to control over-alignment in the MAFFT multiple sequence alignment program" *Bioinformatics*
85. Ranwez, Harispe, Delsuc et al. (2011) "MACSE: Multiple Alignment of Coding SEquences Accounting for Frameshifts and Stop Codons" *PLoS ONE*
86. Li, Godzik (2006) "Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences" *Bioinformatics*
87. Nguyen, Schmidt, Haeseler et al. (2015) "IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies" *Mol Biol Evol*
88. Tamura, Stecher, Kumar (2021) "MEGA11: Molecular Evolutionary Genetics Analysis Version 11" *Mol Biol Evol*
89. Hirabara, Serdan, Gorjao et al. (2022) "SARS-COV-2 variants: differences and potential of immune evasion" *Front Cell Infect Microbiol*
90. Farkas, Fuentes-Villalobos, Garrido et al. (2020) "Insights on early mutational events in SARS-CoV-2 virus reveal founder effects across geographical regions" *PeerJ*
91. Li, Du, Liu et al. (2023) "Comparative susceptibility of SARS-CoV-2, SARS-CoV, and MERS-CoV across mammals" *ISME J*
92. Ruhl, Kühne, Beushausen et al. (2023) "Third SARS-CoV-2 vaccination and breakthrough infections enhance humoral and cellular immunity against variants of concern" *Front Immunol*
93. Pérez-Alós, Hansen, Armenteros et al. (2023) "Previous immunity shapes immune responses to SARS-CoV-2 booster vaccination and Omicron breakthrough infection risk" *Nat Commun*
94. Samanta, Banerjee, Das et al. (2024) "Enhancing immunological memory: unveiling booster doses to bolster vaccine efficacy against evolving SARS-CoV-2 mutant variants" *Curr Microbiol*
95. Fedele, Trentini, Schiavoni et al. (2022) "Evaluation of humoral and cellular response to four vaccines against COVID-19 in different age groups: A longitudinal study" *Front Immunol*
96. Fedele, Schiavoni, Trentini et al. (2023) "A 12-month follow-up of the immune response to SARS-CoV-2 primary vaccination: evidence from a real-world study" *Front Immunol*
97. Widge, Rouphael, Jackson et al. (2021) "Durability of responses after SARS-CoV-2 mRNA-1273 vaccination" *N Engl J Med*
98. Turner, 'halloran, Kalaidina et al. (2021) "SARS-CoV-2 mRNA vaccines induce persistent human germinal centre responses" *Nature*
99. Guo, Zhang, Gu et al. (2024) "Durability and crossreactive immune memory to SARS-CoV-2 in individuals 2 years after recovery from COVID-19: a longitudinal cohort study" *Lancet Microbe* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12424301&blobtype=pdf | # Full-length and near-full-length genomes of human rhinovirus A105 detected in patients with non-respiratory tract symptoms in Osaka, Japan
Kazuma Okada, Yuki Hirai, Yumi Ushikai, Atsushi Kaida
## Abstract
One complete genome sequence and two nearly complete genome sequences of human rhinovirus A105, which were detected from patients with non-respi ratory tract symptoms, were determined. KEYWORDS human rhinovirus, non-respiratory tract symptoms, full-genome H uman rhinovirus (HRV) is a non-enveloped positive-sense RNA virus that belongs to the genus Enterovirus of the Picornaviridae family. It is divided into three species, Enterovirus alpharhino, betarhino, and cerhino, which contain more than 160 different genotypes (1). While HRV mostly causes the common cold and lower respiratory tract illnesses (2), it has also been detected in patients with non-respiratory tract symp toms (nRTS), including myocarditis, pericarditis, and encephalitis/encephalopathy (3-8). However, genome data of HRVs detected from patients with nRTS are limited. Here, we report full-length and near-full-length genomes of HRV-A105 detected in three patients with nRTS in 2015.As HRV-A105 was detected only in patients with nRTS, HRV-A105-positive specimens were selected for genome sequencing. Those specimens were collected from hospitals in Osaka, Japan, as part of a passive surveillance program, which was part of a national surveillance program for viral infectious diseases in Japan based on the Infectious Disease Control Law (9). Those specimens included nasal secretions (patients N-18, N-23, and N-46) and fecal specimen (N-46). Patient ages were 11 years and 10 months, 7 years and 8 months, and 1 year and 9 months. N-18 presented with mononeuritis and glossopharyngeal nerve palsy, N-23 presented with acute myocarditis, and N-46 presented with paraplegia. Viral RNA in the specimens was extracted using QIAamp Viral RNA Mini Kit (Qiagen) and used for cDNA synthesis. Primers used in the synthesis were random hexamer, oligo(dt), and the original designed reverse transcription primer for 5′ RACE (Table 1).We detected the HRV genomes and investigated their genotypes using an approx imately 390 bp region in the 5′ UTR amplified with the primers DK001 and DK004 (Table 1) (10,11). These PCR products were Sanger sequenced and classified as A105 based on phylogenetic analysis (11). Other primers were designed based on genome sequences of A105 and A57, which are closely related, to determine the whole genome sequence of A105 detected in this study. HRV genomes, except the 5′ and 3′ end regions, were amplified by PCR using KOD-One (TOYOBO) with the primer pairs (Table 1). Seven overlapping PCR products were generated per genome with amplicon lengths ranging from 970 to 1,475 bp. Each PCR product was Sanger sequenced bidirectionally. The 5′ and 3′ UTRs were sequenced using the SMARTer RACE 5′/3′ kit and 3′-Full RACE Core Set (TaKaRa). Assembly used overlapping regions of 200-500 bp using DNADynamo v1.63 (BlueTractorSoftware).
The complete genome derived from the nasal sample of N-46 was 7,129 nt long and had a GC content of 38%. The nearly complete genomes derived from N-18 and N-23 were 7,058 and 7,063 nt long, respectively, and both had a GC content of 38%. The three HRVs were classified into the A105 genotype cluster by phylogenetic analysis based on VP1 nucleotide sequences (Fig. 1). The p-distance values compared to A105 reference strain SC9723 (KY369874) were 0.017 (N-18), 0.016 (N-23), and 0.019 (N-46), confirming all derived genomes as A105 genotypes (12,13). While causal evidence linking HRV to nRTS remains limited, this study may contribute to understanding the pathogenesis of HRV.
## References
1. Esneau, Duff, Bartlett (2022) "Understanding rhinovirus circulation and impact on illness" *Viruses*
2. To, Yip, Yuen (2017) "Rhinovirus -From bench to bedside" *J Formos Med Assoc*
3. Hazama, Shiihara, Tsukagoshi et al. (2019) "Rhinovirus-associated acute encephalitis/encephalopathy and cerebellitis" *Brain Dev*
4. Han, Liu, Feng et al. (2024) "Fulminant myocarditis associated with human rhinovirus A66 infection: a case report" *Front Pediatr*
5. Liu, Zhao, Feng et al. (2022) "A severe case of human rhinovirus A45 with central nervous system involvement and viral sepsis" *Virol J*
6. Pelkonen, Roine, Anjos et al. (2012) "Picornaviruses in cerebrospinal fluid of children with meningitis in Luanda, Angola" *J Med Virol*
7. Soma, Aizawa, Matsunaga et al. (2021) "Clinically mild encephalitis/encephalopathy with a reversible splenial lesion associated with rhinovirus" *Pediatr Infect Dis J*
8. Tapparel, Huillier, Rougemont et al. (2009) "Pneumonia and pericarditis in a child with HRV-C infection: a case report" *J Clin Virol*
9. Kaida, Kubo, Takakura et al. (2014) "Associations between CO-detected respiratory viruses in children with acute respiratory infections" *Jpn J Infect Dis*
10. Kiang, Yagi, Kantardjieff et al. (2007) "Molecular characterization of a variant rhinovirus from an outbreak associated with uncommonly high mortality" *J Clin Virol*
11. Kiang, Kalra, Yagi et al. (2008) "Assay for 5' noncoding region analysis of all human rhinovirus prototype strains" *J Clin Microbiol*
12. Simmonds, Gorbalenya, Harvala et al. (2020) "Recommendations for the nomenclature of enteroviruses and rhinoviruses" *Arch Virol*
14. Mcintyre, Knowles, Simmonds (2013) "Proposals for the classification of human rhinovirus species A, B and C into genotypically assigned types" *J Gen Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12710320&blobtype=pdf | # Performance of national HIV testing algorithms in 14 Populationbased HIV Impact Assessment surveys: accuracy of HIV diagnosis using a two-test strategy, with or without a tie-breaker test, in different prevalence settings, 2015-2022
Hetal Patel, Yen Duong, Sehin Birhanu, Melissa Metz, Jared Garfinkel, Kathryn Lupoli, Daniel Yavo, Herbert Longwe, Faith Ussery, Kristin Brown, Stephen Mccraken, Clement Ndongmo, Jessica Justman, Andrew Voetsch, Bharat Parekh
## Abstract
Population-based HIV Impact Assessment surveys were conducted in 14 countries to measure HIV prevalence, incidence, and viral load suppression in chil dren aged 0-14 years and adults aged ≥15 years. We examined the performance of rapid HIV testing algorithms in multiple countries. Survey participants received an HIV diagnosis based on two serial tests in Cameroon,
KEYWORDS HIV, testing algorithm, survey, laboratory testing, quality, PHIA A ccurate HIV diagnosis is a cornerstone of HIV treatment programs and can be achieved through various testing methodologies and strategies. Point-of-care rapid HIV tests (RHTs) can be conducted clinics, hospitals, community centers, and home self-testing, allowing for rapid decision-making and the opportunity to provide timely prevention and treatment services (1)(2)(3). These point-of-care assays are integrated into a structured algorithm, which typically includes a series of steps and methods to establish a national rapid HIV testing algorithm. Traditionally, the national HIV testing algorithms consisted of either two serial tests (Test 1 [T1] and Test 2 [T2]) or two serial tests (T1 and T2) followed by a tie-breaker test (T3) in case of discordant results, and they are well established in most sub-Saharan African countries (4).
RHT was first introduced in sub-Saharan Africa in the late 1990s. Despite early concerns about stigma, discrimination, diagnostic challenges, complex laboratory systems, and limited treatment availability, HIV programs have expanded significantly over the past two decades, thanks to access to HIV testing with widespread use of RHT. Today, more than 83 million people undergo testing annually using the national HIV testing algorithms (5,6). Overall, ease of use, immediate availability of test results, accessibility, convenience, and linkage to treatment have played a crucial role in promoting HIV awareness, ensuring access to lifesaving treatment, ultimately reducing mortality, and preventing transmission of HIV. With the expansion of RHT use, the number of sites and staff providing services has also increased. Although RHTs are user-friendly, countrywide implementation requires proper training, continuous review of competency/certification requirements, proper documentation, and quality control testing. Many quality assurance (QA) tools, such as rapid test continuous quality improvement training packages, proficiency testing using dried tube specimens, and results recording in standardized logbooks, are available for President's Emergency Plan for AIDS Relief (PEPFAR) programs (7). However, there is still a lack of systematic largescale site-level data on individual test performance and comprehensive reviews of the national testing algorithm performance.
More than 25 RHTs are included on the World Health Organization (WHO) prequa lification list of diagnostic assays, and 6-8 are included on the U.S. Food and Drug Administration (FDA) list (8). Both WHO and FDA have minimum performance require ments of ≥99% sensitivity and ≥98% specificity (9). In addition, other performance criteria (inter-reader variability, invalid rates), operational characteristics (ease of use, test packaging, and ease of result interpretation), and regulatory auditing of manufacturing facilities are reviewed during the approval process (9). The pre-qualification list is utilized by the ministries of health to select the most appropriate rapid tests for use in testing algorithms within their HIV programs.
Since 2014, the Population-based HIV Impact Assessment (PHIA) surveys, nationally representative household surveys that measure HIV prevalence and incidence, have been conducted in multiple PEPFAR-supported countries. Additionally, these surveys have been standardized and implemented in multiple countries to measure the UNAIDS 95-95-95 targets: the first 95, which represents the percentage of people living with HIV who know their HIV status; second 95, which represents the percentage of those who are aware of their HIV status who are receiving antiretroviral treatment; and the third 95, which represents the percentage of people on treatment who have a suppressed viral load (VL), as established by the Joint United Nations Programme on HIV/AIDS (10). To accurately measure the entire set of targets, the first 95, or the percentage of people living with HIV who know their HIV status, reflecting the accuracy of HIV diagnosis, is critical.
During PHIA surveys, eligible participants were tested and received the results in their homes using the national rapid HIV testing algorithm. Additionally, for QA processes, laboratory-based confirmation using a supplemental assay was done to examine the performance of the national rapid HIV testing algorithm in 14 PHIA surveys, namely, Cameroon, Côte d'Ivoire, Eswatini, Ethiopia, Haiti, Kenya, Lesotho, Malawi, Mozambi que, Namibia, Tanzania, Uganda, Zambia, and Zimbabwe. These countries used their respective testing algorithms in HIV testing services (HTSs) to diagnose HIV infection; however, supplemental assays were not routinely used to further confirm algorithm accuracy. Use of supplemental testing in PHIA surveys provided an opportunity to evaluate national HIV testing algorithm performance in these countries using testing data from these surveys. Findings from this analysis may have implications in HTSs, where the same strategy is commonly used.
## MATERIALS AND METHODS
## PHIA study design
The PHIA project study objectives, study design, sample size, eligibility criteria, and other survey details have been described elsewhere (11). Briefly, the PHIA countries and their respective survey years described in this report are as follows: Cameroon (2017-2018), Côte d'Ivoire (2017-2018), Eswatini (2021), Haiti (2019-2020), Kenya (2018-2019), Lesotho (2016-2017), Malawi (2015-2016), Mozambique (2021-2022), Tanzania (2016-2017), Zambia (2016), Ethiopia (2017-2018), Namibia (2017), Uganda (2020-2021), and Zimbabwe (2015-2016). For each national survey, consenting adults (15+ years, upper age band varied by country) and children (18 months to 14 years, except for Eswatini, Mozambique, and Uganda surveys) underwent HIV counseling, and they provided blood for rapid HIV testing and provided answers to an interview administered by field staff in the household (HH). Participants provided blood samples for HIV testing and other biomarkers as described in each study protocol, including HIV confirmation, HIV recency testing, CD4+ count, VL levels, and detection of antiretroviral and drug resistance for those blood samples testing HIV-positive.
## Specimen collection and processing
In all countries, a phlebotomy-trained nurse/laboratorian collected whole blood specimens from study participants during the HH visit. Depending on age, approximately 1 mL was collected from children under 2 years using a heel prick and an ethylenediami netetraacetic acid (EDTA) microtainer (Becton Dickinson, USA), 6 mL from children aged 2-14 years using a single 6-mL EDTA vacutainer, and 14 mL total from adults 15+ years in a 10-mL and 4-mL EDTA vacutainer. For participants where venous blood draw was not possible (collapsed veins or no consent for venous blood draw), approximately 1 mL of blood was collected using a finger prick and an EDTA microtainer. Once the HH testing was completed, specimens were collected and transported in coolers with freezer packs to satellite or central labs. They were processed into plasma and/or dried blood spots (DBSs), tested as needed, and stored within 24 h of collection time to ensure specimen integrity for all testing (12).
## HIV testing algorithms and testing
For each PHIA, the country's national rapid HIV testing algorithm, either two test or two test with a tie-breaker, was used to establish the objective HIV status of study participants (aged 18 months to 15+ years). All RHT was performed in the participant's HH, and results from individual rapid tests dictated whether additional RHT was required. All individual biomarker test kit information (lot number and expiry date) and test results (as reactive, non-reactive, or invalid) were recorded in the pre-programed survey tablets. All RHTs used in the surveys were on WHO's prequalification list. The Determine HIV-1/2 (Abbott, Illinois, USA) was T1 in all countries, except for Tanzania, where SD-Bioline HIV-1/2 (Abbott, Illinois) was used, and Ethiopia, where Wantai HIV-1/2 (Beijing, China) was used. HIV testing was performed according to the manufacturers' instructions using the whole blood sample. The following three outcomes were recorded for the two-test algorithms: participants with a non-reactive T1 result were reported as negative. Those with reactive results for both T1 and T2, conducted serially, were reported as positive. If T1 was reactive and T2 was non-reactive, the algorithm was repeated. If results remained discordant upon repeat testing, they were reported as indeterminate. For the tie-breaker algorithms, the first two outcomes remained unchanged. However, if T1 was reactive but T2 was non-reactive, T3 was performed. HIV status was determined from the results of T3 (reactive or non-reactive), with the result reported as HIV-positive or HIV negative, respectively, based on the results of the tie-breaker test. While in Uganda, the results of T3 distinguished HIV indeterminate from HIV negative; for comparability, the results of T3 reactive were considered HIV-positive in this analysis. Final HIV diagnosis was immediately returned to the participant at the HH and recorded in survey tablets, referral forms, and sample tracking forms. Additionally, participants received pre-and post-test counseling, and those with HIV-indeterminate results were referred for further testing at a healthcare facility of their choice. Participants who tested HIV-positive were informed that HIV VL testing would be conducted, and the results would be provided later to the health facility of their choice.
## Quality management and confirmatory testing
To ensure test kit performance, bi-weekly external quality control panels (HIV-positive and negative) were tested by all field and laboratory staff. QA testing was performed at the satellite laboratory (usually district hospital laboratories) or central laboratories (usually national reference laboratories). The QA testing strategy involved retesting the first 25-50 samples (25 for Eswatini, Uganda, and Mozambique and 50 for all other countries) tested by each field tester using the country testing algorithm as performed in the HH, to ensure ongoing testing proficiency. In addition, for most countries, 5% of HIV-negative specimens, as well as all HIV-indeterminate specimens, were also retested. HIV confirmatory testing of all HIV-positives (and in some cases of all HIV indeterminates) was conducted using Geenius HIV-1/2 Confirmatory Assay (Bio-Rad, USA) according to the manufacturer's instructions (13)(14)(15). The assay is a rapid immunochromatographic test that has the capacity to distinguish between HIV-1 and HIV-2 infection. The Geenius HIV-1/2 assay was not part of any national RHT algorithm of the participating countries. For some cases where repeat Geenius testing was required and whole blood or plasma was not available, an in-house validated DBS elution protocol was used, which was previously used in another study (16).
## Data review and final classification
Final HIV classification for each survey was based on results from national testing algorithms conducted at the HH and confirmatory testing using Geenius HIV-1/2 at the laboratory. The Geenius HIV-1/2 assay is a more specific test and can potentially distinguish between HIV-1 and HIV-2 antibodies (13). In case of discrepant results between HH testing, QA retesting, and Geenius, specimens were re-tested at the central laboratory. All discrepant results were reviewed case-by-case and adjudicated for additional testing. In very rare cases, when results could not be resolved through retesting, then a HH revisit for re-collection and testing was conducted (data from re-visits are excluded from the analysis presented here). Data were separately analyzed for adults (15+ years) and younger age groups (18 months to 14 years) to allow analysis of testing algorithms in populations with a range of HIV prevalence from low to high. Agreement between the two RHTs was calculated from concordant results, while the positive predictive value (PPV) of the testing algorithm was calculated using final Geenius results as the reference. All results presented are unweighted except countryspecific HIV prevalence (17).
## RESULTS
Among 10 countries that used a two-test algorithm (without a tie-breaker test), there were 194,199 eligible adult participants (ages 15+ years); 17,774 (9.2%) of their blood specimens had an HIV-positive test result and 1,175 (0.6%) had an HIV-indeterminate test result (Table 1). Among the four countries that used a two-test algorithm with a tie-breaker, there were 82,062 adult participants; 8,003 (10.8%) had blood specimens with an HIV-positive test result (two out of three reactive tests) (Table 1). Uganda's country-specific algorithm classified T1 and T3 positive as indeterminate; however, for the purpose of this analysis, they are grouped with the other three countries using the tie-breaker test, where T1 and T3 reactive were classified as HIV-positive.
HIV-positivity rate among children (18 months to 14 years) was 0.9% for both two-test (433 out of 49,112 tested) and tie-breaker (199 out of 16,695 tested) algorithms.
A total of 194,199 eligible household participants, age 15 years or older, were tested in 10 countries, as part of the PHIA surveys using each country's respective national two-test algorithm (Fig. 1), nine using Determine HIV-1/2; the tenth, Tanzania, used SD Bioline HIV-1/2. Of the 18,949 (9.8%) participants had blood specimens reactive by T1, 17,774 (93.8%) specimens were confirmed as HIV-positive and 1,175 (6.2%) as HIV indeterminate by T2 per national testing algorithm. Geenius HIV-1/2 rapid test further confirmed 17,670 (99.4%) specimens as HIV positive. Accordingly, the T1 and T2 HIV-positive concordance was 93.8% and the PPV was 99.4% for the two-test algorithm.
A total of 82,062 participants (ages 15 years and older) had blood specimens tested by a tie-breaker algorithm in four countries; of these, 8,948 (10.9%) were reactive by T1, and 7,828 (87.5%) were reactive and 1,120 (12.5%) were non-reactive by T2 (Fig. 2). Of the 1,120 (12.5%) T2 non-reactive specimens, 175 (15.6%) were reactive by the T3 tie-breaker test, while 945 (84.4%) were non-reactive. The T1 and T2 HIV-positive concordance was 87.5%, and the T2 and T3 negative concordance was 84.4% of the tie-breaker test algorithm. PPVs were separately calculated for T1 and T2 positive and T1 and T3 positive diagnoses, as confirmed by Geenius. A total of 7,795 out of 7,828 T1 and T2 reactive cases were confirmed as positive by Geenius, yielding a PPV of 99.6%, while only 105 of 175 T1 and T3 positive cases were positive on Geenius, yielding a PPV of 60%. This lower PPV was heavily influenced by the Uganda survey where only 12 (15%) a Includes 1390 as test 1 and test 2 reactive and 175 as test 1 and test 3 reactive. Uganda survey had a tie-breaker algorithm, however their national testing algorithm classified T1 and T3 reactive as indeterminate. To be consistent with other 3 countries, T1/T3 are considered as HIV-positives for this analysis.
b Data here are presented by age groups, 15 years and older and 18 months to 14 years, and by total participants tested as HIV-positive, negative, and indeterminate. c "-" indicates no value. d Boldface indicates subtotal or total of specific column.
of 80 T1 and T3 positive cases were confirmed by Geenius; we note that per the Uganda national algorithm, these participant specimens were considered HIV indeterminate. For the other three countries that used the tie-breaker algorithm, 93 (98%) of 95 T1 and T3 reactive specimens were confirmed as positive by Geenius. Overall, the PPV of the tie-breaker test algorithm was 98.8%. The T1 and T2 concordance and PPV of the national testing algorithms by country are presented by HIV prevalence in Fig. 3. The HIV prevalence ranged from 1.8% in Haiti to 27.0% in Eswatini for ages 15+ (Fig. 3A). T1 and T2 concordance ranged from 74.0% to 99.5% and increased with HIV prevalence, whereas the PPV of the national testing algorithm ranged from 95.0% to 99.9%. For children aged 18 months to 14 years (Fig. 3B), T1 and T2 concordance ranged from 20.6% to 100% and RHT algorithm PPV ranged from 90.5% to 100%, increasing with HIV prevalence. Only 632 participants in the 18-month to 14-year age group had specimens with HIV-positive test results; overall T1 and T2 concordance and PPV were highly variable for this relatively small group, especially when disaggregated by country (Fig. 3B). Figure 3B excludes data from Eswatini, Uganda, and Mozambique since their survey did not include <15-year-olds.
## DISCUSSION
Our comprehensive review of the national HIV testing algorithms across multiple PHIA surveys provided several important insights on algorithm performance. The T1 and T2 concordance and PPV of the national testing algorithm increased with HIV prevalence. Across surveys, the average discordance of 11% for both the two test and the tie breaker algorithm strategies, and discordance was even higher in low burden countries. This pattern aligns with previously published findings from Nigeria [18]. The PPV of national testing algorithms for adults aged 15 years and older was >98% in most countries when compared with the Geenius confirmatory assay, except for Côte d'Ivoire (95.1%) and Malawi (96.5%). The PPV for two test and tie-breaker algorithms was similar within this age group. Including a tie-breaker 3rd test to resolve indeterminate cases identified additional HIV-positive cases. The PPV among indeterminate cases was only 60% as confirmed by Geenius, although 85% (68 of 80 were negative by Geenius) of uncon firmed cases were from the Uganda survey. The HIV prevalence among those between 18 months to 14 years was low and as expected the testing algorithm PPVs were lower in this population. We found that a positive result in a low prevalence population may be less reliable and may require additional testing. This finding was consistent with the 2019 WHO HIV testing service (HTS) guidance document which recommends 3-test algorithms to increase the PPV for the test-and-treat era [4]. The 2019 WHO HIV testing strategy encourages both high and low burden coun tries to use three consecutive reactive test results for HIV-positive diagnosis. While high burden countries, where HIV prevalence is ≥5%, can continue using two consec utive reactive test results to define an HIV-positive outcome, as countries approach or exceed the first 95 goal, residual HIV prevalence of undiagnosed persons will be less than 5%. In such cases, WHO recommends deploying the 2019 three-test algorithm (4). Consequently, as countries move to implement new HIV testing strategies, careful planning and thorough evaluation are essential for the success of their programs. Potential challenges that may arise include, but are not limited to, staff training, logistics and procurement, updates to national guidelines, records and documentation, and the provision of QA, including immediate corrective actions. Additionally, most HTS sites currently use capillary whole blood specimens obtained via fingerprick for RHT. Under the three-test algorithm, discordant results (i.e., T1 reactive and T2 non-reactive) require repeat testing, which often necessitates an additional fingerprick for specimen collection. Moreover, WHO also recommends retesting HIV-positive individuals before ART is initiated, requiring an additional round of testing. Given that the implementation of a new WHO three-test algorithm requires significant resources, we highly recommend programs to assess the feasibility of successful implementation.
Both the US FDA and WHO prequalified RHTs have diagnostic sensitivity and specificity of ≥99% and ≥98%, respectively, and are accurate when testing is per formed appropriately. To provide accurate and reliable results to the participants in the PHIA surveys, we sought to ensure adequate training of the survey staff, proper specimen management from standardized specimen collection (either via venous draw or fingerprick), use of suitable collection equipment (e.g., sterile venipuncture kits), and correct labeling and identification. Additionally, the PHIA surveys ensured the availa bility, proper storage conditions (safe, secure, and temperature-regulated), and timely transportation of RHT to healthcare centers (19). The removal of expired or unused test kits and the monitoring of near-expiry inventory were also critical components of effective stock management. Furthermore, the collection of adequate specimen volume to meet testing requirements, considering potential repeats and following ethical guidelines, led to better results and patient outcomes. For the PHIA surveys, most participants had ~14-mL whole blood specimen collected via venipuncture, stored, and transported in cold chain to a satellite laboratory for additional processing, testing, and storage (12). Adequate and quality specimen volume allowed for repeats, verification, and HIV-positive confirmation of HH testing in the PHIA surveys. The concordance of 5% repeat testing of negative specimens was more than 99.99%. The repeat testing was reduced to 2% in subsequent rounds of the PHIA surveys (12).
In conclusion, the performance of national RHT algorithms in our surveys is optimal. When implemented with high-quality standards, the national HIV testing algorithms will yield accurate and reliable results.
## References
1. Xue, Song, Peng et al. (2023) "Point-of-care HIV test for a promising simple and rapid clinical HIV definite diagnosis process" *Curr HIV Res*
2. Elliott, Sanders, Doherty et al. (2019) "Challenges of HIV diagnosis and management in the context of pre-exposure prophylaxis (PrEP), post-exposure prophylaxis (PEP), test and start and acute HIV infection: a scoping review" *J Int AIDS Soc*
3. Nkambule, Philip, Reid et al. (2021) "HIV incidence, viremia, and the national response in eswatini: two sequential population-based surveys" *PLoS One*
4. (2019) "Consolidated guidelines on HIV testing services"
5. Marum, Taegtmeyer, Parekh et al. (2012) "What took you so long?" the impact of PEPFAR on the expansion of HIV testing and counseling services in Africa" *J Acquir Immune Defic Syndr*
6. (2024) "Presidents emergency plan for AIDs relief"
7. Parekh, Kalou, Alemnji et al. (2010) "Scaling up HIV rapid testing in developing countries: comprehensive approach for implementing quality assurance" *Am J Clin Pathol*
8. (2025) "Complete list of donor screening assays for infectious agents and HIV diagnostic assays"
9. (2021) "Overview of the WHO prequalification of in vitro diagnostics assessment: prequalification of in vitro diagnostics"
10. Laurence (2024) "AIDS at a Crossroads:" highlights from the 2024 UNAIDS report" *AIDS Patient Care STDS*
11. Sachathep, Radin, Hladik et al. (2021) "Population-based HIV impact assessments survey methods, response, and quality in Zimbabwe, Malawi, and Zambia" *J Acquir Immune Defic Syndr*
12. Patel, Duong, Birhanu et al. (2021) "A comprehensive approach to assuring quality of laboratory testing in HIV surveys: lessons learned from the population-based HIV impact assessment project" *J Acquir Immune Defic Syndr*
13. Montesinos, Eykmans, Delforge (2014) "Evaluation of the Bio-Rad Geenius HIV-1/2 test as a confirmatory assay" *J Clin Virol*
14. (2025) *Full-Length Text Journal of Clinical Microbiology*
15. Bujandric, Grujic, Obradovic (2021) "Assessing donor suitability for blood donation: utility of Geenius HIV 1/2 confirmatory assay" *Transfus Apher Sci*
16. Wiredja, Ritchie, Tam et al. (2021) "Performance evaluation and optimized reporting workflow for HIV diagnostic screening and confirmatory tests in a low prevalence setting" *J Clin Virol*
17. Do, Iriemenam, Kohatsu et al. (2020) "High level of HIV false positives using EIA-based algorithm in survey: importance of confirmatory testing" *PLoS One*
18. (2016) "Resources -First and final reports from Population-based HIV Impact Assessment (PHIA) surveys"
19. Patel, Ikpe, Bronson et al. (2022) "Performance of HIV rapid testing algorithm in Nigeria: findings from a household-based Nigeria HIV/AIDS Indicator and Impact Survey (NAIIS)" *PLOS Glob Public Health*
20. Bekele, Gemechu, Ayalew (2022) "Assessment of HIV rapid test kits inventory management practice and challenges in public health facilities of Addis Ababa" *Ethiopia. Integr Pharm Res Pract* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793611&blobtype=pdf | # P-2202. Viral Prevalence, Demographic Characteristics, and Clinical Manifestations among Outpatients with Acute Respiratory Infection during Five Influenza Seasons
Mary Nowalk, John Williams, Monika Johnson, Helen Agostino, Gabriella Alicea, Michael Susick, Lora Pless, Richard Zimmerman, G Balasubramani
Background. Numerous studies, primarily among hospitalized patients, have been undertaken to describe the epidemiology and burden of human metapneumovirus (hMPV). We analyzed stored specimens from an influenza vaccine effectiveness study from the 2016-20 and 2021-22 active influenza seasons to determine the prevalence of several respiratory virus infections including hMPV among outpatients seeking care for an acute respiratory illness (ARI).
Methods. Consented enrollees were individuals ≥6 months of age presenting with cough of ≤ 7 days' duration who provided nasal and pharyngeal ( > 24 months only) swabs, access to EMR data and completed enrollment and follow-up surveys. Samples were tested by singleplex reverse transcription polymerase chain reaction (RT-PCR) assays for presence of influenza virus, hMPV, RSV, PIV and SARS-CoV-2 (2021-22 only). Individuals with co-infections were excluded from the study sample. Prevalence of infection was calculated using the percentage of each targeted virus among all ARI tests in a given time period.
Results. After excluding 68 co-infections, 7,362 enrollees were grouped into those with influenza (n=2,017), RSV (n=762), hMPV (n=423), PIV (n=83), SARS-CoV-2 (n=353) and other ARI (who tested negative for any of these viruses (n=3,724). The percentage of total ARI patients each season (exclusive of 2021-22 when SARS-CoV-2 predominated) who tested positive was: 30.0%-36.9% ˗ influenza, 10.5%-13.6% -RSV; 4.7%-7.2% -hMPV; and 0.05%-1.9% -PIV. Compared with patients with influenza, those with hMPV infections were younger, less often were smokers, less often had fever or sore throat, but more often had congestion, shortness of breath (SOB) and follow-up medical visits. Compared with patients with RSV, hMPV patients were older, had higher BMI, and less often had congestion and SOB. Compared with patients with other ARI, hMPV patients were younger, less often reported smoking or sore throat and more often reported fever and follow-up medical visits.
Conclusion. hMPV represented 4.7-7.2% of ARI cases presenting for outpatient care during the 2016-20 and 2021-22 active influenza seasons. Individuals with hMPV infections had different demographic characteristics and presented with distinct symptoms from those with influenza, RSV or other ARI. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12328403&blobtype=pdf | # Editorial: Synergistic approaches to managing Gram-negative bacterial resistance
Andreas Zautner, Hani Harb, Boyke Bunk, Melanie Lynn Conrad, Percy Schröttner, Mariola Ferraro
## Abstract
Editorial on the Research TopicSynergistic approaches to managing Gram-negative bacterial resistance Multiresistant Gram-negative bacterial pathogens pose a major threat to global health (Huttner et al., 2013;Macesic et al., 2025), and the continuous increase in antimicrobial resistance, coupled with the very limited introduction of new antibiotics, exacerbates this situation (Aslam et al., 2018). One effective way to counteract antimicrobial resistance is through containment measures that can be implemented at local, regional and international levels. However, such measures require the collection of epidemiological data to deduce appropriate strategies, which can be very time consuming. In addition to the development of new antimicrobial drugs, ongoing evaluation of further treatment strategies is crucial. Notably, phage therapy has re-gained importance in recent decades (Slopek et al., 1983), however, to successfully implement such alternatives, detailed knowledge of bacterial pathogenicity is essential. This includes genomic data, as well as knowledge regarding how pathogenic species interact with the host microbiome and immune system. This Research Topic highlights current research in this field and emphasizes the threat of antibiotic resistance to public health. A major focus of this Research Topic was placed on Klebsiella pneumoniae. In this context, epidemiological data were collected on the distribution of various resistance genes, and fundamental research contributing to a better understanding of pathogenicity was presented (Li et al., Li et al., Zhong et al.). Of particular note, regarding the genetics of hypervirulent K. pneumoniae, Yan et al. demonstrated the importance of the iucA gene for the expression of the hypervirulent pathotype. Additionally, Klaper et al. proposed the existence of three K. pneumoniae pathovars (classical K. pneumoniae, ESBL-positive, and hypervirulent K. pneumoniae). Their research revealed that hypervirulent strains evade phagocytosis by macrophages and exhibit cytotoxic potential.Frontiers in Cellular and Infection Microbiology frontiersin.org 01
Further studies in this Research Topic investigated risk factors for infection with Acinetobacter baumanii pediatric patients, as well as patients critically ill with Ghamari et al.). An epidemiological study reported on patients in Lithuania who experienced invasive infection caused by Neisseria meningitidis (Ghamari et al.). Moreover, a new resistance mechanism to fosfomycin in Morganella morganii was identified and a new human pathogenic species of the genus Stenotrophomoas was introduced into the taxonomy (Zhang et al., Li et al.). The findings presented in this Research Topic highlight critical areas for research, hospital hygiene and public health initiatives, indicating which isolates and pathovars should become a primary focus in the future.
## References
1. Aslam, Wang, Arshad et al. (2018) "Antibiotic resistance: a rundown of a global crisis" *Infection Drug Resistance*
2. Huttner, Harbarth, Carlet et al. (2013) "Antimicrobial resistance: a global view from the 2013 World Healthcare-Associated Infections Forum" *Antimicrobial Resistance Infection Control*
3. Macesic, Uhlemann, Peleg et al. (1983) "Results of bacteriophage treatment of suppurative bacterial infections. II. Detailed evaluation of the results" *Arch. Immunol. Ther. Exp. (Warsz)* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792965&blobtype=pdf | ## Abstract
Background. This study aims to assess and compare the clinical presentations, laboratory parameters, radiological findings, and outcomes of adult patients infected with different dengue serotypes presenting to a tertiary care hospital in Kolkata, India,
Results. 140 cases (DENV2:48, DENV3:86, DENV4:6) mean age 34.4±12 years; 55.5% male : 44.5% female. 77 patients needed admission. Hypertension(36%) commonest comorbidity. All comorbidities found no severity difference among three groups. Common symptoms fever (99%), body ache (60%), myalgia (55%), headache (50%), and nausea (45%). Myalgia was more in DENV2 (p=0.009), arthralgia in DENV4 (p=0.04). Hematological, liver function tests, renal function tests were similar across 3 groups, except for higher triglycerides (198 mg/dL) and CRP (38.2 mg/dL) in DENV4; ferritin was lowest in DENV3 (482 ng/mL). On imaging gall bladder wall edema (18%) and splenomegaly (37%) were more seen in DENV2. No significant difference in dengue with warning signs across groups. Severe dengue(n=14) occurred in 14% DENV2 and 8% DENV3, none in DENV4. Secondary dengue (n=24) showed more warning signs than primary (87% vs 44%, p=0.001). Complications like plasma leakage, shock, and bleeding were comparable across serotypes without significant difference. One death occurred in a DENV2 patient due to DSS Poster Abstracts • OFID 2026:13 (Suppl 1) • S1115
Conclusion. DENV2 and DENV3 were predominant serotypes. DENV2 showed more systemic symptoms and notable radiological changes. Elevated inflammatory markers were linked to severity. No major difference in severity or outcome were found across serotypes. Study limitation was no DENV1 and few DENV4 cases. Other studies showed various serotype severity, this study adds to existing knowledge showing minimal outcome variation by serotype. Larger multicentric studies and serotype surveillance are needed to clarify serotype-related risks as dengue virus changes its genes & epidemiology. In my knowledge this is latest study (2023)(2024) OPV dose. We also found a significant difference in the number of non-synonymous mutations (p < 0.005) in samples collected from weeks 1-8 (first dose of OPV) compared to weeks 9 -17 (after 2nd dose of OPV). |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12791982&blobtype=pdf | # 2015, and Feb 1 st , 2024, from the Houston site of the New Vaccine Surveillance Network. We assessed genetic diversity, lineage dynamics, and selective pressures, by phylogenetic analysis, variant calling, dN/dS ratio calculations and Shannon entropy
Chad Hinkle, Christopher Lehmann, Vera Tesic, Angelica Moran
## Abstract
T cell count, and respiratory support. Significant risk factors for 28-day mortality in PJP patients included the use of decreased PaO 2 /FiO 2 ratios (final OR: 0.98, P < 0.001), lower platelet counts (final OR: 0.98, P =0.057), lower CD3 + (final OR: 0.99, P = 0.034), as was a lower CD4 + T cell count (final OR: 0.98, P = 0.023). Patients with immune-mediated inflammatory diseases experienced the lowest survival rates. The use of corticosteroids did not enhance survival, regardless of patients having good or poor oxygenation status. Co-infections, particularly those with multiple pathogens, were associated with the most adverse outcomes, with "bacterial + viral" co-infections posing the greatest risk among dual pathogens, and bacterial infections being the most detrimental in single-pathogen scenarios.
Background. Blood metagenomic next-generation sequencing, commercially known as the Karius test (Karius, Redwood City, CA), is a diagnostic tool that can aid clinicians in uncovering pathogens that might otherwise be challenging to diagnose by standard microbiological techniques. However, due to its cost, use in clinical practice is limited. Published studies have found conflicting impact on clinical management, thus the appropriate clinical scenarios for Karius testing require further investigation. We evaluated the indications for Karius testing and impact on antimicrobial management to develop evidence-based recommendations for test ordering at our institution.
Methods. In a retrospective chart review, we examined 135 Karius tests ordered between 2023-2024 at an academic medical center for pediatric and adult patients. Patient demographics and diagnoses associated with the test were collected. All microbiology and antimicrobial data before and after test results were examined. The clinical outcome of 30-day mortality was evaluated.
Poster Abstracts • OFID 2026:13 (Suppl 1) • S1101 Results. A total of 135 patients were tested, demographics shown in Table 1. There were 143 diagnoses listed as the indication for test ordering (Figure 1). There were 69 positive tests and 21% were concordant with microbiology cultures while 8% were discordant. There were 19 results (14%) that changed management. The diagnoses most likely to have management change in response to testing are shown in Figure 2A-B. The types of antimicrobial changes are listed in Figure 2C. A total of 142 organisms were identified, 91 bacterial, 32 viral and 19 fungal. Figure 2D depicts management changes occurring in bacterial (52%) and fungal organisms (42%). Within 30 days of the test, 20% patients died (Figure 3).
Conclusion. In summary, we retrospectively assessed the indications and impact of Karius testing for patients at our institution. Karius tests changed antimicrobial management most often with liver abscess, culture-negative endocarditis, and pneumonia. While fungi represented a minority of organisms, they resulted in most of the antimicrobial changes. High rates of 30-day mortality were observed. Our data suggest Karius is most impactful with specific diagnoses, fungal pathogens, and non-critically ill patients.
Disclosures. All Authors: No reported disclosures
Delve Bio, Inc., Boston, Massachusetts 2 Delve Bio, Boston, Massachusetts 3 Broad Clinical Labs, Burlington, Massachusetts
$$1 2 2 2 2 2 2 2 2 2 3 3 3 2 1$$ |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12741116&blobtype=pdf | # Multimodal cell death drives the immunopathogenesis of RSV infection
Tianxiang Yang, Zhizhong Mi, Zhaolong Li, Sheefa Mirza, Abayeneh Girma
## Abstract
Respiratory syncytial virus (RSV) is a major cause of severe respiratory tract infections in infants, older adults, and immunocompromised individuals. Despite decades of research, effective therapies are limited, largely due to an incomplete understanding of how infected cells and immune responses interact to shape disease outcomes. Recent evidence indicates that RSV activates multiple regulated cell death (RCD) programs-including apoptosis, necroptosis, pyroptosis, ferroptosis, and autophagy-associated cell death which interact through shared molecular mediators to form a multimodal cell death (MMCD) network. This integrated system regulates the balance between viral clearance and immunopathological injury. Central mediators such as caspase-8, RIPK3, and NLRP3 act as molecular hubs coordinating these death programs and amplifying inflammatory responses. Understanding how MMCD shapes RSV immunopathogenesis provides a unified framework linking cell death to immune dysfunction. This review summarizes recent progress in elucidating the MMCD network, highlights its role in deathinflammation feedback loops, and discusses potential strategies for therapeutic modulation. Conceptualizing RSV disease through the lens of MMCD may guide the development of precision interventions that restore immune homeostasis while preserving antiviral defense.
## 1 Introduction
RSV, a negative-sense RNA virus of the Pneumovirus genus (Paramyxoviridae family), is the leading cause of acute lower respiratory tract infections (ALRIs) in infants and young children worldwide. It also contributes significantly to pneumonia and respiratory failure among older adults and immunocompromised populations. Epidemiological data estimate that RSV infects more than 33 million children under five years of age each year, leading to over 100,000 deaths-most occurring in low-and middle-income countries (1). In older adults, RSV-associated hospitalization and mortality rates are comparable to or even exceed those of seasonal influenza (2), highlighting its growing cross-age public health significance.
At the mechanistic level, RSV infection is characterized by epithelial barrier disruption, mucus hypersecretion, and massive immune cell infiltration within the airway mucosa (3). Viral entry, mediated by the fusion (F) and attachment (G) glycoproteins, leads to syncytium formation. This process also triggers inflammatory signaling through the TLR4-NF-kB axis (4). These viral proteinhost interactions generate inflammatory signaling and cellular stress-including oxidative imbalance, endoplasmic reticulum (ER) stress, and mitochondrial dysfunction that act as upstream triggers for RCD pathways. Emerging evidence suggests that this complex immunopathology is underpinned by a coordinated network of RCD pathways that collectively shape viral clearance, inflammation, and tissue remodeling (5). Historically, RSV-induced cell death was attributed mainly to apoptosis (6). However, advances in single-cell transcriptomics and spatial multi-omics have revealed that RSV simultaneously engages multiple RCD programs-apoptosis, pyroptosis, necroptosis, ferroptosis, and autophagy-related deathwithin the same tissue microenvironment (7,8). These pathways are interconnected via shared molecular hubs such as caspase-8, RIPK3 (receptor-interacting serine/threonine-protein kinase 3), and NLRP3 (NOD-like receptor family pyrin domain-containing 3), forming a coordinated MMCD network. Despite accumulating evidence, the mechanisms by which MMCD orchestrates immune dysregulation and tissue injury during RSV infection remain unclear. This mini review summarizes current knowledge of MMCD, its immunopathological implications, and its potential as a target for precision intervention.
## 2 Molecular basis of cell death signaling during RSV infection
The RSV genome is a single-stranded, negative-sense RNA approximately 15.2 kb in length, encoding 11 viral proteins. Among these, the fusion (F) and attachment (G) glycoproteins are central to viral entry and early host signaling. The F protein undergoes proteolytic activation by host enzymes, exposing the fusion peptide that mediates membrane merging and syncytium formation (9). This process simultaneously triggers the TLR4-MAPK-NF-kB axis, promoting the release of proinflammatory cytokines such as IL-6, TNF-a, and CXCL8 (C-X-C motif chemokine ligand 8) (10). The G protein, which contains a CX3C chemokine motif mimicking the host ligand CX3CL1 (C-X3-C motif chemokine ligand 1), binds to CX3CR1 (C-X3-C motif chemokine receptor 1) on epithelial and dendritic cells (11). This interaction facilitates viral attachment while transiently dampening immune cell recruitment and dendritic cell maturation (12), creating an initial phase of immune suppression. G proteininduced calcium dysregulation and oxidative stress further prime infected cells for downstream activation of RCD signaling.
The nonstructural proteins NS1 and NS2 function as potent antagonists of host antiviral defenses (13). The NS1/2 complex inhibits MAVS-mediated RIG-I-IRF3 signaling and promotes STAT2 degradation, thereby suppressing type I interferon responses (14). In addition, NS2 interferes with mitochondrial dynamics by preventing DRP1 (Dynamin-Related Protein 1) phosphorylation, leading to excessive ROS accumulation (15). The resulting oxidative stress activates ER stress-CHOP pathway, downregulates Bcl-2, and increases mitochondrial membrane permeability (16), predisposing cells to apoptosis, necroptosis, or ferroptosis under sustained infection.
Host receptors further contribute to RSV propagation and the integration of death signaling. CX3CR1, IGF1R (Insulin-like growth factor 1 receptor), and P2Y2 (P2Y purinoceptor 2) receptors facilitate viral entry and propagation (17,18). IGF1R activation transiently delays apoptosis through the PI3K-AKT pathway, while P2Y2-mediated calcium influx promotes viral release. Prolonged ER stress through the PERK-eIF2a-CHOP axis amplifies mitochondrial dysfunction, providing a convergence point for apoptotic, necroptotic, and ferroptotic pathways. Together, these viral and host interactions lay the molecular foundation for the integrated MMCD network that underpins RSV pathogenesis.
## 3 Forms of regulated cell death in RSV infection
RSV infection activates multiple RCD programs that operate in a coordinated and context-dependent manner (Figure 1), rather than acting independently, apoptosis, pyroptosis, necroptosis, ferroptosis, and autophagy-related death intersect through shared molecular hubs-including caspase-8, RIPK3, NLRP3, and GPX4 (Glutathione Peroxidase 4) that determine whether infected cells undergo containment, inflammatory amplification, or oxidative injury.
During early infection, controlled apoptosis and autophagy contribute to viral clearance while limiting inflammation. As infection progresses, inflammatory RCD pathways such as pyroptosis and necroptosis dominate, releasing DAMPs (Damageassociated molecular patterns) that amplify cytokine production and epithelial injury. In later stages, ferroptosis-driven oxidative stress promotes chronic epithelial dysfunction and tissue remodeling. Viral proteins (F, G, NS1/2) and host stress responses cooperatively regulate these transitions through mitochondrial, ER, and lipid metabolic pathways.
Figure 1 depicts the integrated framework in which RSV components trigger signaling cascades that converge on the MMCD network, producing overlapping waves of regulated cell death (RCD) that collectively shape immune pathogenesis. Table 1 summarizes key regulators of these pathways and their corresponding immunopathological outcomes. Together, these findings highlight that RSV pathogenicity stems not from a single mode of cell death, but from the hierarchical integration of multiple death programs that determine the magnitude of inflammation, the failure of tissue repair, and overall disease severity.
## 4 Integrative network and immunopathological coupling
The MMCD network is governed by a central regulatory triadthe Caspase-8-RIPK3-NLRP3 molecular kernel-which serves as the signaling core that balances apoptosis, necroptosis, and pyroptosis. This molecular kernel integrates stress and death cues, ensuring the dynamic coordination of multimodal cell death responses during RSV infection. Under physiological conditions, caspase-8 restrains necroptosis by cleaving RIPK1 and RIPK3. During RSV infection, NS1/2 proteins inhibit caspase-8 activation and promote RIPK3-MLKL assembly, redirecting apoptotic signals toward necroptotic and pyroptotic outputs (27,28). This triad-based network has also been implicated in SARS-CoV-2 and influenza pathogenesis, underscoring its conserved immunoregulatory role across viral infections. The ensuing potassium efflux and mitochondrial ROS Lipid peroxidation, mitochondrial dysfunction, chronic inflammation (25,26) generation further activate NLRP3, establishing a feed-forward inflammatory loop that amplifies tissue injury. Beyond intracellular regulation, the MMCD network tightly interacts with innate and adaptive immune circuits. DAMPs and alarmins released from necroptotic and pyroptotic cells stimulate dendritic cells and macrophages via TLR4 and RAGE, fueling IL-1b, IL-6, and TNF-a production (29). These cytokines, in turn, promote Th17 polarization and neutrophil recruitment, which reinforce oxidative stress and perpetuate ferroptotic signaling (30,31). Spatial transcriptomics has revealed that epithelial, myeloid, and stromal cell subsets engage distinct MMCD modules within inflamed lung tissue (32). This spatial integration underscores MMCD as a systems-level mechanism linking cellular fate to immune orchestration.
Together, the Caspase-8-RIPK3-NLRP3 triad functions as the molecular kernel of MMCD, integrating apoptotic, necroptotic, pyroptotic, and ferroptotic signals into a unified immunopathological continuum. This framework redefines RSV-induced injury as a dynamic equilibrium between protective elimination and destructive amplification.
## 5 Feedback loops between cell death and immune activation
MMCD not only results from infection but also actively shapes immune responses. Distinct forms of RCD release a spectrum of intracellular contents-DAMPs, alarmins, and oxidized lipids-that engage pattern recognition receptors (PRRs) on immune cells. These outputs are orchestrated by the Caspase-8-RIPK3-NLRP3 molecular kernel, which couples regulated cell death to immune amplification. Molecules such as HMGB1, ATP, and mitochondrial DNA activate macrophages and dendritic cells via TLR4, NLRP3, and RAGE, triggering cytokine cascades dominated by IL-1b, IL-6, TNF-a, and type I interferons (33,34). These cytokines further sensitize epithelial and immune cells to RCD activation by upregulating death receptors such as Fas and TRAILR.
Conversely, sustained immune activation feeds back to intensify cellular stress pathways. Persistent production of IFN-g and TNF-a induces ER stress, mitochondrial dysfunction, and oxidative damage, shifting the balance from controlled apoptosis to necroptosis and ferroptosis (35). Activated neutrophils and Th17 cells release reactive oxygen and nitrogen species, exacerbating lipid peroxidation and impairing tissue repair. This bidirectional crosstalk establishes a "death-inflammation" loop, in which MMCD-derived signals perpetuate immune activation, and immune effector molecules in turn reinforce multimodal cell death.
Beyond acute infection, MMCD-driven inflammation influences long-term immune remodeling. Repeated epithelial injury and chronic oxidative stress reshape the airway microenvironment, leading to aberrant macrophage polarization and exhaustion of tissue-resident memory T cells (36,37). Such maladaptive remodeling may contribute to post-RSV wheezing and asthma-like sequelae observed in susceptible individuals. Understanding these feedback dynamics provides insight into how transient infection evolves into chronic immunopathology. Such maladaptive remodeling may explain post-viral airway hyperresponsiveness and provide a mechanistic bridge between infection and chronic respiratory disorders.
Collectively, MMCD acts as both a driver and target of immune regulation, forming a dynamic feedback network that connects cellular fate with immune homeostasis. Dissecting this loop could uncover novel biomarkers for disease severity and identify intervention points where selective modulation of MMCD may break the cycle of inflammation and tissue injury, setting the MMCD as both an effector and regulator of immune homeostasis, warranting integrative therapeutic exploration.
## 6 Discussion
The concept of MMCD provides a unified framework to interpret the immunopathogenesis of RSV infection. By integrating apoptosis, pyroptosis, necroptosis, ferroptosis, and autophagy-related processes into a coordinated signaling continuum, MMCD explains how the same virus can simultaneously elicit antiviral defense and excessive inflammation. The hierarchical organization of MMCD-ranging from early containment to chronic oxidative injury-clarifies the dynamic balance between protective clearance and destructive immunopathology. This integrative view expands beyond conventional single-pathway models and highlights the Caspase-8-RIPK3-NLRP3 axis as a pivotal regulatory triad governing immune amplification.
From a translational perspective, the MMCD paradigm opens new opportunities for precision immunotherapy. Translating MMCD insights into therapy demands identification of druggable nodes within this integrated network. Pharmacological inhibition of RIPK1, NLRP3, or ferroptosis regulators (e.g. Necrostatin-1, MCC950, Deferoxamine) has shown promise in preclinical models by mitigating epithelial cell necrosis and restoring immune balance (38,39). Rather than complete suppression, selective modulation of MMCD pathways may recalibrate host responses without compromising antiviral immunity. Moreover, profiling MMCD-associated signatures through single-cell multiomics or spatial transcriptomics could enable patient stratification and the identification of biomarkers predicting disease severity or treatment response.
Nevertheless, several challenges remain. First, the heterogeneity of MMCD responses across cell types and disease stages complicates therapeutic targeting. Second, pharmacodynamic optimization and delivery barriers limit clinical translation of existing inhibitors. Third, the dual role of MMCD in both defense and pathology raises concerns that broad inhibition may inadvertently impair antiviral immunity. Future studies should therefore adopt context-specific modulation approaches guided by quantitative modeling and in vivo validation. For instance, early-phase inhibition of the NLRP3 inflammasome might suppress excessive epithelial or macrophage pyroptosis and reduce IL-1b-mediated inflammation-a notion supported by recent studies showing amelioration of RSV-induced immunopathology upon NLRP3 blockade (40). In more severe or late-stage disease characterized by widespread necroptosis and cell lysis, targeting RIPK1/RIPK3 or downstream MLKL may prevent necroptotic amplification, consistent with observations of necroptosis involvement in RSVinfected macrophages (41).Integrating MMCD dynamics into systems immunology frameworks could reveal emergent properties that predict therapeutic windows and adverse outcomes.
Current evidence connecting individual MMCD pathways to RSV pathogenesis is largely correlative, as most mechanistic insights are derived from in vitro experiments or animal models rather than human tissues. Establishing true causality will require longitudinal clinical sampling, single-cell-resolved analyses, and functional perturbation studies in physiologically relevant systems. Moreover, although conceptual temporal models-such as "early apoptotic containment" versus "late-stage inflammatory necroptosis or pyroptosis"-provide a useful framework, MMCD programs in vivo are likely to operate simultaneously and are strongly influenced by cell type, tissue organization, metabolic context, and local immune cues. This inherent spatiotemporal complexity poses challenges for precise in vivo validation. Finally, therapeutic targeting of MMCD remains constrained by issues of specificity. Intervening in one pathway may unintentionally impair beneficial antiviral immunity or provoke compensatory death programs. A critical challenge in the therapeutic targeting of necroptosis lies in achieving sufficient pathway specificity while minimizing off-target effects. Kinase inhibitors, particularly those acting on RIPK1 or RIPK3, may exhibit unintended interactions with other signaling kinases, leading to undesirable side effects. Moreover, inhibiting necroptosis can inadvertently redirect cell death signaling toward alternative programmed pathways, such as apoptosis or pyroptosis, due to compensatory network activation. These complexities underscore the need for developing nextgeneration inhibitors with improved selectivity, temporal control, and context-dependent modulation to ensure therapeutic efficacy and safety.
Future research should therefore focus on dissecting pathway cross-talk and designing compounds that maintain the delicate balance between inhibition and cellular adaptation. Thus, effective clinical translation will require strategies that enable cell-typeselective or spatially restricted modulation of detrimental MMCD signaling while preserving essential host defense mechanisms.
In summary, viewing RSV immunopathogenesis through the lens of MMCD transforms our understanding of virus-host interactions from a linear cascade into an interconnected, selfregulating network. This conceptual shift emphasizes that immunopathology arises not from isolated molecular events but from the dynamic orchestration of multimodal cell death and immune feedback. Elucidating and therapeutically harnessing this network may ultimately pave the way for precision interventions that disrupt immunopathology while preserving host defense in RSV and other viral inflammatory diseases.
## 7 Methods
We performed a literature search in PubMed, Web of Science using the keywords "RSV", "RCD", "autophagy", "necroptosis", "pyroptosis", "ferroptosis", "NLRP3", "RIPK3","MMCD"and "immunopathogenesis". Articles published between 2015-2025 were included. Additional references were identified through manual screening of relevant reviews and cross-referencing of primary research articles.
## References
1. Kelleher, Subramaniam, Drysdale (2025) "The recent landscape of RSV vaccine research"
2. Terstappen, Hak, Bhan et al. (2024) "The respiratory syncytial virus vaccine and monoclonal antibody landscape: the road to global access" *Lancet Infect Dis*
3. Yagi, Ethridge, Falkowski et al. (2024) "Microbiome modifications by steroids during viral exacerbation of asthma and in healthy mice"
4. Xiong, Tao, Li et al. (2024) "Both chebulagic acid and punicalagin inhibit respiratory syncytial virus entry via multi-targeting glycoprotein and fusion protein" *J Virol*
5. Girma (2024) "Biology of human respiratory syncytial virus: Current perspectives in immune response and mechanisms against the virus" *Virus Res*
6. Pengyu, Chang, Chao (2025) "Programmed cell death in human respiratory syncytial virus infection" *Front Cell Infect Microbiol*
7. Rajan, Nagaraj, Bomidi et al. (2025) "Single cell sequencing analysis of respiratory syncytial virus-infected pediatric and adult human nose organoids reveals age differences, proliferative diversity and identifies novel cellular tropism" *J Infect*
8. Shao, Liu, Hu et al. (2025) "Interplay between autophagy and apoptosis in human viral pathogenesis" *Virus Res*
9. Neal, Barrett, Edmonds et al. (2024) "Examination of respiratory syncytial virus fusion protein proteolytic processing and roles of the P27 domain" *J Virol*
10. Dudek, Czerkies, Kwiatek (2024) "Differential expression of cytokines and elevated levels of MALAT1 -Long non-coding RNA in response to non-structural proteins of human respiratory syncytial virus" *Virology*
11. Rivas-Fuentes, Salgado-Aguayo, Santos-Mendoza et al. (2024) "The role of the CX3CR1-CX3CL1 axis in respiratory syncytial virus infection and the triggered immune response" *Int J Mol Sci*
12. Georgakopoulou, Pitiriga (2025) "Immunomodulation in respiratory syncytial virus infection: mechanisms, therapeutic targets, and clinical implications" *Microorganisms*
13. Merritt, Pei, Leung (2024) "Pathogenicity and virulence of human respiratory syncytial virus: Multifunctional nonstructural proteins NS1 and NS2. Virulence"
14. Chorvinsky, Bhattacharya, Bera et al. (2025) "Down syndrome alters type-III IFN and pro-inflammatory airway epithelial responses to RSV infection" *Am J Respir Cell Mol Biol*
15. Li, Wei, Li et al. (2025) "Selenomethionine inhibited RSV infection-induced apoptosis and inflammatory response through ROS-mediated signaling pathway" *ACS Omega*
16. Che, Xie, Lin et al. (2024) "Andrographolide attenuates RSV-induced inflammation by suppressing apoptosis and promoting pyroptosis after respiratory syncytial virus infection in vitro" *Comb Chem High Throughput Screen*
17. Hayes, Oraby, Camargo et al. (2024) "Mapping respiratory syncytial virus fusion protein interactions with the receptor IGF1R and the impact of alanine-scanning mutagenesis on viral infection" *J Gen Virol*
18. Wang, Ren, Wang et al. (2024) "Investigating the active components and mechanistic effects of Forsythia suspensa Leaf against RSV via the PI3K/ Akt-NLRP3 pathway" *Heliyon*
19. Barra, Liwski, Phonchareon et al. (2025) "IL-5 enhances human mast cell survival and interferon responses to viral infection" *J Allergy Clin Immunol*
20. Jiao, Yang (2025) "Berberine alleviates respiratory syncytial virus (RSV)-induced pediatric bronchiolitis and fibrosis via suppressing the HMGB1/TLR4/NF-kB pathway" *Microbiol Spectr*
21. Tang, Mao, Ruan et al. (2024) "Drugs targeting CMPK2 inhibit pyroptosis to alleviate severe pneumonia caused by multiple respiratory viruses" *J Med Virol*
22. Shen, Zhang, Xie et al. (2018) "Jinxin oral liquid inhibits human respiratory syncytial virus-induced excessive inflammation associated with blockade of the NLRP3/ASC/Caspase-1 pathway" *BioMed Pharmacother*
23. Simpson, Spann, Phipps (2022) "MLKL regulates rapid cell death-independent HMGB1 release in RSV infected airway epithelial cells" *Front Cell Dev Biol*
24. Simpson, Loh, Ullah et al. (2020) "Respiratory syncytial virus infection promotes necroptosis and HMGB1 release by airway epithelial cells" *Am J Respir Crit Care Med*
25. Kombe, Fotoohabadi, Nanduri et al. (2024) "The role of the nrf2 pathway in airway tissue damage due to viral respiratory infections" *Int J Mol Sci*
26. Yang, Liu, Nie et al. (2023) "Oxidative stress and ROS-mediated cellular events in RSV infection: potential protective roles of antioxidants" *Virol J*
27. Yuan, Ofengeim (2024) "A guide to cell death pathways" *Nat Rev Mol Cell Biol*
28. Hussain, Rohlfing, Santoro et al. (2025) "Neuregulin-1 prevents death from a normally lethal respiratory viral infection" *PLoS Pathog*
29. Youm, Nguyen, Grant et al. (2015) "The ketone metabolite b-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease" *Nat Med*
30. Wang, Huang, Zhao et al. (2024) "Fumarprotocetraric acid and geraniin were identified as novel inhibitors of human respiratory syncytial virus infection in vitro" *Front Cell Infect Microbiol*
31. Chao, Puhach, Frieser et al. (2023) "Human TH17 cells engage gasdermin E pores to release IL-1a on NLRP3 inflammasome activation" *Nat Immunol*
32. Wang, Chao (2025) "Alveolar epithelial cell dysfunction in acute respiratory distress syndrome: mechanistic insights and targeted interventions" *Biomedicines*
33. Müller, Straßmann, Baier et al. (2024) "Liver fibrosis is enhanced by a higher egg burden in younger mice infected with S. mansoni. Cells"
34. Lomascolo, May, Hatahet et al. (2024) "Polyamines mediate cellular energetics and lipid metabolism through mitochondrial respiration to facilitate virus replication" *PLoS Pathog*
35. Vieira Antão, Oltmanns, Schmidt et al. (2024) "Filling two needs with one deed: a combinatory mucosal vaccine against influenza A virus and respiratory syncytial virus" *Front Immunol*
36. Komiya, Kamiya, Oba et al. (2024) "Necroptosis in alveolar epithelial cells drives lung inflammation and injury caused by SARS-CoV-2 infection" *Biochim Biophys Acta Mol Basis Dis*
37. Zhang, Lan, Jiang et al. (2025) "NLRP3 inflammasome mediates pyroptosis of alveolar macrophages to induce radiation lung injury" *J Hazard Mater*
38. Shi, Liu, Tian et al. (2025) "Huaier suppresses lung cancer by simultaneously and independently inhibiting the antioxidant pathway SLC7A11/GPX4 while enhancing ferritinophagy" *Cell Death Discov*
39. Zhou, Wang, Wang et al. (2024) "Screening of novel tumor-associated antigens for lung adenocarcinoma mRNA vaccine development based on pyroptosis phenotype genes" *BMC Cancer*
40. Malinczak, Schuler, Duran et al. (2021) "NLRP3-inflammasome inhibition during respiratory virus infection abrogates lung immunopathology and long-term airway disease development" *Viruses*
41. Bedient, Pokharel, Chiok et al. (2020) "Lytic cell death mechanisms in human respiratory syncytial virus-infected macrophages: roles of pyroptosis and necroptosis. Viruses" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645970&blobtype=pdf | # Human cytomegalovirus UL78 is a nuclear-localized GPCR necessary for efficient reactivation from latent infection in CD34 + hematopoietic progenitor cells
Samuel Medica, Nicole Diggins, Michael Denton, Rebekah Turner, Lydia Pung, Adam Mayo, Olivia Kramer-Hansen, Jennifer Mitchell, Luke Slind, Linh Nguyen, Teresa Beechwood, Gauthami Sulgey, Craig Kreklywich, Daniel Malouli, Mette Rosenkilde, Patrizia Caposio, Daniel Streblow, Meaghan Hancock
## Abstract
Human cytomegalovirus (HCMV) is a ubiquitous pathogen that persists throughout the lifetime of the host due to the establishment of latency. HCMV enco des four putative G protein-coupled receptors (GPCRs): US27, US28, UL33, and UL78. A definitive role for UL78 in HCMV infection has yet to be elucidated. Utilizing an in vitro CD34 + hematopoietic progenitor cell (HPC) model, we demonstrate that a recombinant virus lacking UL78 protein expression fails to efficiently reactivate from latent infection. Furthermore, we show that UL78 preferentially couples to the Gα i family of G proteins and that a recombinant HCMV containing mutations in the UL78 G protein-coupling DRL motif also fails to reactivate from latent infection. Together, our findings indicate that Gα i coupling is important for UL78 function during reactivation in latently infected CD34 + HPCs. To better understand the role of UL78, we conducted proteomic analyses in HCMV-UL78-TurboID-infected fibroblasts and CD34 + HPCs undergoing reactivation from latency. Congruent with our coupling data, we found that Gα i was the only heterotrimeric Gα protein in proximity to UL78. Pathway analysis of the UL78 interac tome revealed that proteins associated with membrane trafficking, signaling, and the nuclear pore complex were enriched in both cell types. In addition, the UL78 interactome contained viral proteins with nuclear localization including viral transcription and DNA replication machinery. Nuclear localization of UL78 was validated using cell fractionation, immunofluorescence microscopy, and proteomic analysis of isolated nuclei. Together, our results provide novel insights into the localization and function of UL78, previously unknown to contribute to reactivation from latent infection. IMPORTANCE Human cytomegalovirus (HCMV) remains one of the most widespread viral infections globally. Primary HCMV infection is typically asymptomatic and leads to the establishment of latency in myeloid lineage cells, where the virus persists for the host's lifetime. Reactivation of latent HCMV can cause severe complications, particularly in immunocompromised individuals, such as transplant recipients and people living with HIV. Several factors influence the transition from latent to lytic infection, including signal transduction through the viral G protein-coupled receptors: US27, US28, UL33, and UL78. Using an advanced in vitro model, we show that recombinant viruses lacking UL78 fail to efficiently reactivate from latent infection. Moreover, we show that UL78 preferentially couples to the Gα i family of G proteins via a conserved DRL motif, and this coupling is required for efficient reactivation. These results were confirmed by proximity-dependent labeling experiments, where we identified Gα i and several other proteins involved in trafficking, signaling, transcription, and nuclear localization. Nuclear localization of UL78 was confirmed by cell fractionation, immunofluorescence microscopy, and proximity-
dependent labeling in isolated nuclei. Collectively, our results uncover a novel role for UL78 in reactivation from latency and shed new light on its localization and function. KEYWORDS CD34 + hematopoietic progenitor cells, UL78, G proteins, G protein coupled receptors, human cytomegalovirus, stem cells C ytomegaloviruses (CMVs) are species-specific herpesviruses that establish life-long infections in their hosts. Human CMV (HCMV) achieves persistence in part through the ability to establish latent infections in CD34 + hematopoietic progenitor cells (HPCs) and CD14 + monocytes (1,2). Clinical reactivation of latent virus can occur in situations of immunosuppression, such as during allogenic or solid organ transplantation (3). Significant morbidity and mortality are associated with HCMV reactivation following transplantation, and currently available antiviral treatments targeting the DNA replica tion machinery have toxic side effects that can exacerbate disease and lead to the emergence of drug-resistant variants (4)(5)(6). Targeting the latent reservoir and/or early reactivation events is an alternative approach that requires detailed knowledge of the viral and cellular factors that regulate these processes.
HCMV encodes four G protein-coupled receptors (US27, US28, UL33, and UL78) that are thought to mimic the functions of cellular chemokine receptors (7). While the functions of US27, US28, and UL33 have been interrogated in the context of lytic and latent infection, much less is known about UL78 (8)(9)(10)(11)(12). The UL78 family includes HCMV UL78, rat CMV (RCMV) R78, murine CMV (MCMV) M78, and the human herpesvirus (HHV)-6 and -7 protein U51 (13)(14)(15)(16). The UL78 family consists of positionally conserved 7-transmembrane proteins that contain a DRL motif located within the second intracellu lar loop (ICL2), which is suspected to be necessary for G-protein coupling. UL78 family members undergo endocytosis from the cell surface like many GPCRs. However, only HHV U51 has been shown to bind chemokines and induce migration (17)(18)(19)(20). UL78, R78, and M78 remain orphan GPCRs with no known ligands, and recent structural analysis suggests that UL78 forms homotrimers that may occlude the putative ligand binding domain (21). A lack of UL78 expression does not impact HCMV lytic replication; however, R78 and M78 are necessary for efficient in vitro replication (16,22,23). RCMV R78 is expressed in many tissues and peripheral blood mononuclear cells in infected rats, and virus lacking R78 does not replicate in the spleen (15,24,25). MCMV M78 is important for transport of virus-infected cells to the salivary glands, which may be partly due to its ability to participate in the downregulation of MHC-II from the infected cell surface (26,27). In transient transfection assays, UL78 was shown to form heterodimers with the cellular chemokine receptors CXCR4 and CCR5, reducing their cell surface expression, as well as with HCMV vGPCR US28, which affected US28-mediated NF-kB activation; however, the mechanism(s) for these findings have yet to be investigated (28,29). Thus, while there are some sequence similarities between the UL78 family members, they may functionally contribute to CMV infection in different ways.
Herein, we investigated the role of HCMV UL78 in latent infection of human embryonic stem cell (hESC) -derived CD34 + HPCs. Our results indicate that Gα i coupling via the DRL motif of UL78 is essential for efficient reactivation from latent infection. To investigate the function of UL78, we performed proximity-dependent labeling experi ments utilizing a recombinant virus expressing UL78 containing a C-terminal TurboID fusion in infected fibroblasts and during reactivation from latency in CD34 + HPCs. We identified a number of cellular and viral proteins as candidate UL78 interactors, including nuclear-localized proteins, such as components of the nucleoporin complex, cellular and viral transcriptional regulators, and viral DNA replication machinery, suggesting that UL78 may localize to the nucleus, as has been observed for a number of cellular GPCRs (30)(31)(32)(33). We determined that a fraction of both WT and G protein-coupling null mutant UL78 is detected at the nucleus using cell fractionation, luciferase assays, and immunofluorescent approaches. Together, our data indicate that UL78 coupling with Gα i is essential for reactivation from latency and that UL78 localization to the nucleus suggests a novel function for this orphan HCMV GPCR.
## RESULTS
## HCMV UL78 is required for efficient viral reactivation from latent infection
Several studies have shown that at least two of the G protein-coupled receptors encoded by HCMV are integral for establishing a latent infection and facilitating viral reactivation (8,11,(34)(35)(36). However, the function of UL78 in this process is still unknown. To evaluate a potential role for UL78 in viral latency or reactivation, we used the TB40/E-GFP bacterial artificial chromosome (BAC) to generate a recombinant virus deficient in UL78 protein expression, but not gene expression. GalK-mediated recombination was used to place two contiguous stop codons immediately following the initiating methionine of UL78 (UL78-2XSTOP) (37). To determine the growth kinetics of the recombinant virus, both single and multistep growth kinetics were analyzed in primary human fibroblasts. Similar to previously published studies utilizing recombinant viruses lacking the entire UL78 open reading frame (ORF) (22,38), UL78-2XSTOP replicated with normal growth kinetics in this cell type (Fig. S1).
To determine whether UL78 has a role in the establishment of latent infection or the capacity of the virus to reactivate, we infected human embryonic stem cell (hESC)-derived CD34 + HPCs with either WT-HCMV (TB40/E-GFP) or UL78-2XSTOP (TB40/ E-GFP-UL78-2XSTOP) viruses. At 48 hours post-infection (hpi), viable, infected (GFP + ), CD34 + HPCs were isolated via fluorescence-activated cell sorting (FACS) and were seeded into long-term bone marrow culture (LTBMC) above a murine stromal support layer under conditions that favor latent infection, as previously described (9,39,40). After 12 days of LTBMC, half of the infected HPCs from each infection group were lysed by mechanical disruption to serve as a pre-reactivation control. The remaining intact HPCs and lysates were plated over monolayers of fibroblasts in reactivation-supportive media supplemented with granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) to perform an extreme limiting dilution assay (ELDA), quantifying the frequency of infectious centers at three weeks post-plating (41). Comparable levels of infectious virus were present in lysed cells (pre-reactivation) infected with WT-HCMV or UL78-2XSTOP, suggesting that UL78 has no effect on the establishment or maintenance of viral latency (Fig. 1A; Fig. S2). In contrast, the frequency of infectious centers for the UL78-2XSTOP-infected cells did not increase in the presence of reactivation stimulus compared to WT-HCMV infected cells, suggesting that UL78 is required for efficient reactivation from latency (Fig. 1A; Fig. S2). Since an observed deficit in viral reactivation can be caused by an inability to maintain viral genomes or genome-containing cells throughout latency, we quantified viral genome copies from infected CD34 + HPCs at the end of latent infection via quantitative PCR. A comparable number of viral genomes were present in cells infected with WT-HCMV or UL78-2XSTOP, indicating that, despite the presence of viral DNA, viruses lacking UL78 are unable to produce infectious virions in HPCs stimulated to reactivate (Fig. 1B). Together, these results indicate that UL78 plays an integral role in the viral reactivation process in CD34 + HPCs.
## HCMV UL78 coupling to Gα i heterotrimeric G-proteins via a conserved DRL motif is required for reactivation from latency
The DRY motif is a highly conserved sequence found in the second intracellular loop of most Class A GPCRs (42,43). Located at the boundary of transmembrane helix 3 (TM3) and intracellular loop 2 (ICL2), this motif forms an ionic lock to help maintain GPCR conformation and plays a crucial role in receptor activation. Specifically, the arginine residue within this motif stabilizes the receptor to facilitate G protein activation and subsequent signal transduction (44)(45)(46). UL78 contains a DRL motif that is conserved across the UL78 family. To identify the complement of G proteins that bind to UL78, we made use of a nLuc-based complementation assay measuring real-time interactions between receptors and heterotrimeric G protein complexes described previously (9,47). In this system, the C-terminus of the GPCR is linked in frame to natural peptide (NP), while the Gα subunit is genetically fused to the complementing Large Bit (LgBiT). Proximity of the complementing fragments reconstitutes a functional luciferase protein whose activity can be measured with addition of substrate. To this end, we engineered an in-frame natural peptide tag on the C-terminus of the wild-type UL78 receptor. We utilized site-directed mutagenesis to make alanine substitutions for the entire motif (DRL 133-135 -AAA 133-135 ) and the arginine specifically (DRL 133-135 -DAL 133-135 ) to serve as negative controls, as these mutations would be predicted to affect G protein coupling. As a positive control, we used US28, as it has been shown to functionally couple to most Gα family members (9,48,49). Equivalent expression of each construct was verified by immunoblot in transiently transfected HEK-293 cells (Fig. S3). In live cell GPCR interaction assays with LgBiT-Gα i , LgBiT-Gα q , and LgBiT-Gα 12 , the wild-type UL78 receptor exhibited similar Gα i coupling to that of US28 (Fig. 2A andD), but did not significantly couple to the other Gα isoforms (Fig. 2B through D). Moreover, both UL78 constructs containing mutations within the DRL motif showed significant attenuation in their ability to couple to the Gα i family of G-proteins and did not show any increase in coupling breadth with LgBiT-Gα q and LgBiT-Gα 12 (Fig. 2). Together, these data indicate that UL78 Gα i -specific coupling requires the DRL motif.
To better understand the function of UL78 within the context of viral infection, we generated recombinant viruses using the TB40/E-GFP BAC by engineering the HiBiT tag onto the N-terminus of UL78 (TB40/E-GFP-HB-UL78). An additional recombinant HiBiTtagged virus was generated containing an alanine substitution at position R134 in the predicted G-protein coupling motif (TB40/E-GFP-HB-UL78-DAL). Following reconstitution of these viruses, we evaluated their growth kinetics via single-and multi-step growth analyses in primary fibroblasts. Both recombinant viruses replicated with similar kinetics to each other and to the parental TB40/E-GFP virus, indicating that the addition of a HiBiT tag to UL78 and the substitution R134A have no effect on lytic replication (Fig. S4). Furthermore, both recombinant viruses demonstrated immediate early, early, and late protein expression with similar kinetics (Fig. 3A). UL78 protein expression was first observed at 24 hpi with peak expression observed between 72 and 96 hpi (Fig. 3A). To determine whether the DRL motif mutations alter receptor expression at the cell surface, we conducted HiBiT surface vs. total expression assays in infected human fibroblasts (9). In this assay, the total luminescence emitted by HiBiT-tagged UL78 in lysed cells is compared with that of live cells, where only surface UL78 can interact with the comple menting LgBiT. The ratio calculated between the surface and total luminescence indicates the fraction of UL78 that is present at the surface. Consistent with previously published data (19,20), we observed that the majority of UL78 localized intracellularly (Fig. 3B). Additionally, no appreciable difference in surface expression was detected between the two recombinant viruses, indicating that mutation within the DRL motif of UL78 does not significantly affect cellular surface expression (Fig. 3B).
To determine whether the observed reactivation deficit with recombinant viruses lacking UL78 protein expression (Fig. 1A) can be recapitulated with a virus deficient in UL78 Gα i protein coupling, we infected hESC-derived CD34 + HPCs with either WT-HCMV (TB40/E-GFP) or HB-UL78-DAL (TB40/E-GFP-HB-UL78-DAL). Viable, GFP + , CD34 + HPCs were isolated via FACS, and after 12 days of LTBMC, both cells stimulated to reactivate and lysates from unstimulated cells were plated onto fibroblast monolayers to assess infectious center frequency. Compared to infection with WT-HCMV, cells infected with HB-UL78-DAL exhibited major deficits in the ability of the virus to efficiently reactivate from latent infection (Fig. 4A; Fig. S5). Viral genome copies from infected CD34 + HPCs at the end of LTBMC were comparable in cells infected with WT-HCMV and HB-UL78-DAL, suggesting that viral genomes, or genome-containing cells, were not lost over the latency culture period (Fig. 4B). Collectively, these results indicate that UL78-Gα i coupling is not required for the establishment of viral latency in CD34 + HPCs but is required for efficient reactivation from latent infection.
## Identification of the HCMV UL78 interactome during lytic infection
A limited number of previous studies have shown that UL78 forms heterodimeric interactions with host chemokine receptors (CXCR4 and CCR5) and the viral chemokine receptor US28 to impair or augment surface expression and downstream signaling activity (28,29). While valuable, these overexpression models monitored interactions in transiently transfected cells, which do not recapitulate the conditions of viral infection. Furthermore, UL78 may exhibit additional cell type-specific interactions with host and viral proteins to modulate signal transduction that have yet to be captured. To character ize the UL78 interactome during viral infection, a recombinant HCMV was generated containing the biotin ligase TurboID (50) linked in-frame to the C-terminus of UL78 (TB40/E-GFP-UL78-TurboID). The recombinant virus replicated with similar growth kinetics in primary fibroblasts relative to the parental virus, indicating that the addition of the TurboID enzyme to UL78 has no effect on lytic replication (Fig. S6). To identify the UL78 interactome, fibroblasts were mock infected or infected with WT-HCMV (TB40/E-GFP) or UL78-TurboID (TB40/E-GFP-UL78-TurboID). At 72 hpi, exogenous biotin was added to the culture media for an additional six hours. Cellular lysates were harvested, and the resulting biotin-conjugated proteins were purified via streptavidin-mediated bead-based precipitation. Efficient labeling and purification were verified by immuno blot probing for HRP-conjugated streptavidin (Fig. 5A). The identity and relative abun dance of the purified biotin-conjugated proteins were determined by label-free quantitative (LFQ) liquid chromatography tandem mass spectrometry (LC-MS/MS). After excluding proteins identified as potential contaminants by comparison against the CRAPome data repository (51), our analysis revealed 1,138 host and 32 viral proteins that showed a significant level of enrichment within the data set (Fig. 5B andC; Data S1).
Interestingly, neither of the previously identified host chemokine receptors reported to interact with UL78 (e.g., CXCR4 and CCR5) (29) was enriched within the data set; however, the previously identified viral candidate interaction partner (US28) was identified above the significance threshold (28,34). Consistent with a recently published study (21), we identified several Gα i isoforms, but not other G protein families, as candidate interaction partners of UL78 (Fig. 5B; Data S1), validating the results of our split-nano luciferase assays (Fig. 3). Reactome over-representation analysis of the enriched host proteins in proximity to UL78 revealed several cellular processes related to trafficking, signal transduction, cytoskeletal remodeling, and nuclear import (Fig. S7A; Data S2). Surpris ingly, we identified many nuclear-localized cellular proteins as candidate UL78 interac tors (Fig. 5B; Data S1 and S2), including many of the components of the nuclear pore complex, nuclear membrane proteins, and transcription factors. Additionally, importin, Rab, SNX, and RanGDP proteins were found in proximity to UL78, suggesting a mecha nism for translocation to the nucleus. Furthermore, viral proteins, including components of the DNA replication machinery, transcriptional regulators, and proteins that regulate nuclear egress, are candidate interaction partners for UL78 (Fig. 5C; Data S1). To confirm the candidate UL78 nuclear interaction partners during lytic infection, we performed a second proximity-dependent labeling experiment in which NHDF cells were infected and treated with biotin as previously described, and then, cytosolic and nuclear fractions were separated prior to tagged protein purification. Efficient labeling and fractionation were verified by immunoblot prior to LC-MS/MS analysis, where we observed visible labeling within the nuclear fraction of cell lysates (Fig. 5D). Quantitative mass spectrome try and downstream analysis revealed 1,592 host and 83 viral proteins as candidate interaction partners of UL78 within the nuclear fraction (Fig. 5E and F; Data S3). Similar to previous experiments, we identified several transcription factors, members of the nuclear pore complex, and Gα i isoforms (Fig. 5E; Data S3). Reactome over-representation analysis of host candidate interaction partners identified several previously identified processes, such as nuclear import, trafficking, and transcription, as well as additional processes involved with RNA biogenesis and chromatin remodeling (Fig. S7B; Data S4). Moreover, many of the viral proteins enriched within this data set are known nuclear proteins involved in viral DNA replication, transcription, DNA packaging, encapsidation, and nuclear egress (Fig. 5F; Data S3). Taken together, our interactome analysis demonstrates that UL78 uniquely interacts with nuclear host and viral proteins during lytic infection conditions and offers novel insight into the function and signal transduction capability of UL78.
## Examining HCMV UL78 interactions during viral reactivation
Since UL78 is important for reactivation from latency, we wanted to identify host and viral interaction partners of UL78 in the context of HCMV reactivation. To this end, we conducted proximity-dependent labeling experiments in CD34 + HPCs stimula ted to reactivate from latent infection. In this experiment, hESC-derived CD34 + HPCs were infected with either WT-HCMV (TB40/E-GFP) or UL78-TurboID (TB40/E-GFP-UL78-TurboID) and were cultured in the same manner as the above experiments (Fig. 6A).
After 12 days of LTBMC, cells were plated in transwells above a fibroblast monolayer in reactivation supportive media supplemented with exogenous biotin and incubated for an additional 16 hours prior to cell lysis. Efficient labeling of proteins in proximity of UL78 was confirmed on whole cell lysates via immunoblot probing with HRP-conjugated streptavidin prior to purification and tryptic digestion (Fig. 6B). The resultant peptides were subjected to LC-MS/MS analysis. After contaminant filtering, we identified 1,075 host and 61 viral proteins as candidate interaction partners of UL78 (Data S5). Notably, we again identified Gα i as the only Gα isoform present within this data set. Additionally, we identified the other viral GPCRs (UL33, US27, and US28), immune evasion proteins (UL40, US23, and US26), and transcriptional activators (UL49, UL69, UL82, and UL122).
When compared to our previous proximity-dependent labeling experiments conducted in fibroblasts under lytic conditions (Fig. 5; Fig. S6), a total of 354 common and 721 unique host hits were identified in CD34 + HPCs (Fig. 6C). In a similar manner, this data set contained 54 common and nine unique viral proteins (Fig. 6D; Table 1). Finally, Reac tome over-representation analysis of the identified host proteins revealed enrichment in several processes related to RNA processing, trafficking, protein post-translational modifications, and members of the nuclear pore complex (Fig. 6E; Data S6). Together, these results confirm the findings from our previous proximity-dependent labeling experiments in a reactivation model and suggest novel interaction partners and nuclear localization, which will both aid in deciphering the function of UL78 during HCMV pathogenesis.
## HCMV UL78 localizes to the nucleus during infection
A limited number of cellular GPCRs have been detected in the nuclear envelope and nucleus and have been shown to play an important role in host signaling and cell cycle regulation (30)(31)(32). Since UL78 proximity-dependent labeling experiments identified several host and viral nuclear-localized proteins during infection, we next used orthogonal methods to validate whether UL78 is present at the nucleus. Previous studies using transient overexpression and infection models clearly detect UL78 at the cell surface and in cytoplasmic endocytic vesicles (19,20). In addition, Fig. 4B demonstrates cell surface expression of a fraction of UL78. To assess whether UL78 also localizes to the nucleus, we infected fibroblasts with WT-HCMV (TB40/E-GFP), HB-UL78 (TB40/E-GFP-HB-UL78), and HB-UL78-DAL (TB40/E-GFP-HB-UL78-DAL) viruses for 72 hours and performed cell fractionation to separate cytoplasmic and nuclear fractions. As shown by immuno blots from fractionated lysates, HB-UL78 and HB-UL78-DAL were detected in both the cytoplasmic and nuclear fractions, suggesting that UL78 localizes to the nucleus during lytic infection and that G protein coupling may not be required for translocation or retention at the nucleus (Fig. 7A). To confirm the nuclear localization of UL78, we performed HiBiT split luciferase assays on fractionated lysates. In these experiments, we utilized WT-HCMV and a recombinant virus with a HiBiT tag on the N-terminus of the membrane protein UL8 (HB-UL8) as negative controls (52). Similar to observations from immunoblot experiments (Fig. 7A), we detected both HB-UL78 and HB-UL78-DAL within the nuclear fraction of these cell lysates (Fig. 7B). Moreover, the signal in nuclear extracts obtained from cells infected with virus expressing HB-UL8 was negligible when com pared to the background readings from cells infected with WT-HCMV (Fig. 7B). We further validated these findings using immunofluorescence in the context of fibroblast infection. WT-HCMV, HB-UL78, and HB-UL78-DAL viruses were used to infect human fibroblasts for 72 hours, and UL78 localization was detected by confocal microscopy using an antibody to HiBiT. During HB-UL78 infection, UL78 could be detected in punctate structures consistent with previous reports showing localization in intracellular vesicles. However, HiBiT signal could also be detected at the nuclear membrane of HB-UL78-and HB-UL78-DAL-infected cells (Fig. 8A andB), further supporting that a fraction of UL78 is nuclear and that this localization may be independent of G-protein coupling. Together, these data strongly support the nuclear localization of a portion of UL78 and suggest novel intranuclear functions for the GPCR.
## DISCUSSION
A mechanistic role for the HCMV-encoded G protein coupled receptor UL78 in viral pathogenesis has remained elusive. In this study, we demonstrate that UL78, along with the viral genes UL7, UL8, UL33, UL81-82ast (LUNA), UL135, UL136, and US28, as well as viral miRNAs miR-UL36, miR-UL112, and miR-UL148D, is required for HCMV reactivation in myeloid lineage cells (8,9,11,34,39,(52)(53)(54)(55)(56). Herein, we establish that recombinant HCMV that lacks UL78 protein expression or contains a single amino acid substitution in the DRL motif (DAL) is unable to reactivate from latency in CD34 + HPCs. We also demonstrate that while UL78 specifically couples to Gα i heterotrimeric G proteins, the UL78 DAL mutant failed to couple, which is consistent with the role of this motif in G protein coupling for other GPCRs. Combined, these findings suggest that UL78 coupling to Gα i is necessary to promote viral reactivation. Gα i specificity was also observed in UL78 proximity labeling, where only Gα i family members were identified as enriched in both HCMV-infected human fibroblasts and latently infected CD34 + HPCs undergoing the early stages of viral reactivation. Moreover, an important, and yet unexpected, finding obtained through analysis of the proximity labeling experiments in both cell types positioned UL78 at the nucleus and identified a potential pathway of nuclear translocation for the viral GPCR (Fig. 9). While the role of UL78 and requirement of G protein coupling in the viral reactiva tion process are both clear, further characterization is needed to determine whether nuclear localization, signaling, and/or interactions with host and viral machinery located in the nucleus are required to promote reactivation. A functional role for nuclear GPCRs has become better appreciated in recent years (57,58). Some cellular GPCRs translocate from the plasma membrane to the nucleus upon ligand binding and activation, whereas others appear resident at the nucleus (59,60). Nuclear-localized GPCRs can activate "classical" signaling cascades within the nucleus as G-proteins, GRKs, β-arrestin, and components of many signal transduction pathways are readily detected in the nucleus (61,62), and nuclear GPCR activity can induce the phosphorylation of signaling intermediates as well as Ca 2+ and cAMP flux (63). Additionally, some cellular GPCRs interact directly with transcriptional regulators and/or host DNA to impact gene expression. For example, nuclear-translocated F2rl1 (PAR2) interacts with the transcrip tion factor Sp1 and enhances expression of the Vegfa gene leading to neovasculari zation during mouse retinal development (64). Intriguingly, this study demonstrated that nuclear-localized PAR2 activated different transcriptional responses from plasma membrane-localized PAR2, suggesting a dichotomy of cellular outcomes based on GPCR location.
Based upon our reactivation and proximity labeling data, we hypothesize that UL78 localized at or near the nuclear pore complex (NPC) is acting as a scaffold to help regulate viral genome accessibility and promote viral transcription during reactivation. This hypothesis is well-founded, as the NPC plays a critical role in regulating host chromatin state and gene transcription by recruiting and organizing histone regula tory/remodeling complexes, as well as by positioning the open chromatin near the NPC opening to facilitate easy access to transcription factors and nucleotide pools (65)(66)(67)(68). The UL78 proximity data were enriched for members of the NPC, including NUPs (35,42,50,62,88,98,133,153,155,188,214,358), transcription factors (ATF6, JUN B, RelA, GTFIIF, SUMO1, YAP1), and RNA metabolism (GTF3C1, POLII, SF1, SYMPK), as well as chromatin remodeling proteins (AHCTF1, LEM, LBR, SAP18, TRIM25, TRIM28) (Fig. 5-select proteins in green, 6, Data S1 to S6). Our UL78 proximity-dependent labeling experiment also identified several viral transcriptional regulators, including the activators UL26, UL35, UL49, UL69, UL72, UL82, UL95, UL112, and UL122, as well as repressors UL34, UL84, and UL138. Alternatively, it is also possible that UL78 coupling to heterotri meric G-protein complexes could directly induce signal transduction to activate these transcriptional regulators and stimulate viral reactivation either by Gα i or Gβγ. HCMV leverages components of the DNA damage response, particularly the ataxia-telangiecta sia mutated (ATM) and ataxia telangiectasia and Rad3-related (ATR) kinase pathways, to facilitate chromatin remodeling and promote the transcriptional activation of the major immediate early promoter (69,70). These kinases phosphorylate downstream effectors, such as H2AX and checkpoint kinase 2 (Chk2), to create a permissive environ ment for viral gene expression (71,72). In parallel, RNA transport and metabolism are tightly regulated during HCMV infection to support the efficient processing and nuclear export of viral transcripts. HCMV lytic infection is associated with altered expression and activity of RNA-binding proteins and splicing factors, such as ribonuclear proteins (RNPs) and serine/arginine-rich (SRs) proteins, which enhance the stability and translational competence of viral mRNAs (73,74). Furthermore, viral proteins like UL69, identified here as a potential UL78 interactor, mimic host mRNA export factors to promote the nuclear export of viral transcripts, bypassing typical cellular restrictions (75)(76)(77). Future studies will aim to determine if UL78 can directly or indirectly modify the viral genome to promote transcription during reactivation.
The work presented here, and by others (22), has identified that the bulk of UL78 protein is localized to the plasma membrane and endosomal pools, but nuclear localization has also been suggested in the context of infection (19). Our biochemical approaches to quantify the sub-cellular localization of UL78, using the HiBiT tag fused to the N-terminus of UL78, determined that between 25 and 30% of total UL78 (both WT and the DAL mutant) is found at the plasma membrane during infection of human fibroblasts (Fig. 3B), with a smaller fraction present in nuclear fractions (Fig. 7 and8). We also observed a fraction of UL78 associated with the nuclear membrane by confocal microscopy. The capacity to label cellular and viral proteins within the nucleus, as well as the nuclear pore complex, suggests that UL78 is oriented such that the C-terminal tail is within the nucleoplasm. There are several ways that GPCRs can traffic to the nucleus, including: (1) agonist-dependent or -independent endocytosis from the cell surface and transport to the nucleus mediated by importins, Rabs, and sorting nexins (SNX) and (2) agonist-independent translocation from the trans-Golgi network to the nucleus via Rabs and importins (Fig. 9). While a specific ligand for UL78 has not been identified to date, a recent structural analysis suggests that the N-terminus of the protein may in fact cover the ligand tunnel area to either self-activate or prevent access of ligands, which makes it currently challenging to assign ligand-dependent vs. ligand-independent nuclear translocation mechanisms for UL78 (21). Our data showing that the UL78-DAL mutant is also found in the nucleus would imply that ligand-dependent signaling may not be required for nuclear localization. Our UL78 proximity-dependent labeling experiments identified several Rab GTPases and SNX proteins as candidate interaction partners that are likely involved in initial endocytosis and recycling (Rabs 3B, 8A, 21, 35 and SNXs 5, 6, 9 and 18). Intriguingly, Rab11a and SNX11 have been implicated in the translocation of cellular GPCRs from the plasma membrane to the nucleus (57,64,78) and were enriched in our data sets. Alternatively, a fraction of UL78 may be transported to the nucleus directly from the endoplasmic reticulum without trafficking to the cell surface. We identified importin proteins as significantly enriched within our data sets, providing a direct mechanism for the import of UL78 into the host cell nucleus. Two putative classical nuclear localization signals exist within the C-terminal tail of UL78 at positions 328-336 and 360-365, suggesting candidate interaction sites for import machinery. It is also possible that UL78 interacts with other cellular proteins that facilitate its nuclear localization. While the precise mechanisms of UL78 nuclear translocation remain unclear, further host and viral genetic studies could uncover how UL78 gets to the nucleus.
Of the four HCMV-encoded GPCRs, two have previously been investigated in the context of latency and reactivation in CD34 + HPCs. UL33 is necessary for virus reactiva tion through the phosphorylation and activation of the transcription factor CREB (8). We, and others, have shown that US28 expression is necessary for virus latency and reactivation in vitro and in vivo in humanized mice in a ligand-dependent manner that involves signaling through specific G proteins and downstream effectors (11,(34)(35)(36). In fact, the multiple roles for US28 at different stages of HPC infection highlight the complexity of GPCR signaling during viral infection of progenitor cells. Herein, we provide a role for UL78 in reactivation from latency in CD34 + HPCs and indicate that Ga i coupling is important for this phenotype. Our work also uncovers the nuclear localization of a fraction of UL78, although it is still not known whether this process is essential for efficient reactivation from latency. These findings add to the growing appreciation of the role of virally encoded GPCRs in latency and reactivation in CD34 + HPCs. UL78 awaits the identification of a ligand, either extracellular or intracellular, and a more mechanistic understanding of its function(s) in different cellular compartments which will lead to a greater understanding of the role played by UL78 in promoting virus reactivation in hematopoietic cells.
## MATERIALS AND METHODS
## Plasmids
Plasmids were generated utilizing traditional cloning methodology as previously described using primers listed in Table S1 (9). Briefly, HA-UL78-NP was generated by cloning the Natural Peptide (NP) tag (GVTGWRLCERILA) in-frame with the C-terminus of UL78. HB-UL78 was generated by cloning the HiBiT (HB) tag (VSGWRLFKKIS) in-frame with the N-terminus of UL78. Fragments were PCR amplified and cloned into the pcDNA3.1-vector. Mutations in the DRL motif of UL78 were generated by site-directed mutagenesis, substituting alanine for the indicated amino acid using the Q5 Mutagenesis Kit (NEB) following the manufacturer's recommended procedure. All constructs were confirmed by sequencing and transformed into TOP10 Escherichia coli cells (Invitrogen). Large-BiT-tagged Gα subunits were kindly provided by Julien Hanson (Addgene plasmid ID: 134359, 134360, 134364, and 134363) (47).
## Cells and virus
Normal human dermal fibroblasts (NHDFs) were obtained from ATCC (PCS-2021-010), and human embryonic kidney (HEK)-293 cells were obtained from Microbix. NHDF and HEK-293 cells were maintained in Dulbecco's modified eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), streptomycin, penicillin, and glutamine at 37°C and 5% CO 2 . M2-10B4 and S1/S1 stromal cells were obtained from Stem Cell Technologies and cultured as previously described (79). WA01 human embryonic stem cells (hESCs) were obtained from the WiCell Research Institute-National Stem Cell Bank and were cultured as previously described (34,40). The HCMV strain TB40/E-GFP, which constitutively expresses green fluorescent protein under the SV40 promoter (80), was amplified in NHDFs as previously described (53,81). Recombi nant viruses were generated using a two-step recombineering procedure utilizing the HCMV TB40/E-GFP bacterial artificial chromosome (BAC). Viral constructs were confirmed by next-generation sequencing prior to plaque purification and clonal expansion. Viral titers were determined via plaque assay on NHDF cells and aliquots stored at -80°C. For viral growth analyses, single-step growth curves were carried out at a multiplicity of infection (MOI) of 3.0 PFU/mL, and multi-step growth curves were carried out at a MOI of 0.01 PFU/mL. Supernatant and cell-associated virus were harvested at multiple time points post-infection and titered via limiting dilution plaque assay on NHDF cells.
## Immunoblot
Blotting procedures were carried out as previously described (9,34). Briefly, cell lysates were harvested using either RIPA Lysis Buffer (Santa-Cruz Biotechnology) supplemen ted with HALT protease inhibitor (Thermo Fisher Scientific) or the NE-PER extraction kit (Thermo Fisher Scientific). Proteins were separated on a 4-12% SDS-PAGE gel and transferred onto PVDF membranes. Immunoblots were performed using antibodies directed against β-Actin (sc-47778, Santa-Cruz Biotechnology), HCMV IE1/IE2 (MAB8131, Millipore-Sigma), HA (sc-7392, Santa-Cruz Biotechnology), HCMV gB (sc-69742, Santa-Cruz Biotechnology), Streptavidin-HRP (21130, Thermo Scientific), NUP98 (C39A3, Cell Signaling Technology), ERp57 (CL2444, Thermo Scientific), GAPDH (sc-47724, Santa-Cruz Biotechnology), and, if required, the appropriate HRP-conjugated secondary antibody (sc-525409, Santa-Cruz Biotechnology). HiBiT-tagged proteins were visualized using the Nano-Glo HiBiT Blotting System (Promega).
## Proximity-dependent labeling experiments
Proximity-dependent labeling experiments were conducted as previously described (34). Briefly, monolayers of NHDF cells or CD34 + HPCs were either mock infected or infected with HCMV TB40/E-GFP-UL78-TurboID or TB40/E-GFP at an MOI of 2. For experiments utilizing NHDFs, at 3 days post-infection, cells were incubated for 6 hours in complete media supplemented with 50 µg/mL biotin. Cells either were lysed in RIPA buffer (50 mM Tris pH 8, 150 mM NaCl, 1% Triton X-100, 0.1% SDS) supplemented with 1× Halt protease inhibitor cocktail (Thermo Fisher), followed by centrifugation at 10,000 × g at 4°C, or were processed using the NE-PER extraction kit (Thermo Fisher Scientific). For experiments utilizing CD34 + HPCs, cells were cultured in LTBMC as previously described (79). At 14 days post-infection, HPCs were placed into RPMI-1640 medium containing 20% FBS, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, 15 ng/mL granulocytecolony stimulating factor (G-CSF), 15 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF), and 50 µg/mL of biotin and overlaid onto confluent monolayers of NHDFs for 16 hours prior to cell lysis. Resultant lysates were incubated with 250 µL Pierce NeutrAvidin Agarose beads (Thermo Fisher) overnight at 4°C with rotation. Beads were collected and washed sequentially with urea wash buffer (PBS pH 7.4, 4 M urea), wash buffer 2 (PBS, pH 7.4, 1% Triton X-100), 50 mM ammonium bicarbonate, and 6 M urea. Proteins were reduced and alkylated using 0.5 M tris(2-carboxyethyl)phosphine and 0.5 M iodoacetamide prior to tryptic digestion. Digestion was halted by adding 20 µL formic acid to each sample, and samples were stored at -80°C until LC-MS/MS analysis.
## LC-MS/MS and data analysis
LC-MS/MS was performed as previously described by the Fred Hutchinson Proteomics Core (Seattle, WA) (34). Briefly, samples were desalted using ZipTip C18 (Millipore, Billerica, MA) and eluted with 70% acetonitrile/0.1% TFA (trifluoroacetic acid; Sigma), and the desalted material was then dried in a SpeedVac. Desalted samples were resuspended in 2% acetonitrile in 0.1% formic acid (12 µL), and 10 µL of sample was analyzed by LC/ESI MS/MS using a Thermo Scientific Easy-nLC II nano HPLC system (Thermo Scientific, Waltham, MA) coupled to a tribrid Orbitrap Fusion mass spectrometer (Thermo Scientific, Waltham, MA). Peptide separations were performed on a reversed-phase column (75 µm × 400 mm) packed with Magic C18AQ resin (5 µm 100 Å; Michrom Bioresources, Bruker, Billerica, MA), directly mounted on the electrospray ion source. A 90-minute gradient from 7% to 28% acetonitrile in 0.1% formic acid at a flow rate of 300 nL/minute was used for chromatographic separations. The heated capillary temperature was set to 300°C, and a static spray voltage of 2,100 V was applied to the electrospray tip. The Orbitrap Fusion instrument was operated in the data-dependent mode, automatically switching between MS survey scans in the Orbitrap (AGC target value 500,000, resolution 120,000, and maximum injection time 50 ms) and MS/MS spectra acquisition in the linear ion trap using quadrupole isolation. A two-second cycle time was selected between master full scans in the Fourier-transform (FT) and the ions selected for fragmentation in the HCD cell by higher-energy collisional dissociation, using a normalized collision energy of 27%. Selected ions were dynamically excluded for 30 seconds, with an exclusion mass width of ± 10 ppm.
Data analysis was performed using Proteome Discoverer 2.2 (Thermo Scientific, San Jose, CA), searching against the UniProt Human (proteome ID: UP000005640) and HCMV TB40/E (proteome ID, UP000143167) proteomes. Trypsin was set as the enzyme with maximum missed cleavages set to 2. The precursor ion tolerance was set to 10 ppm, and the fragment ion tolerance was set to 0.6 Da. Variable modifications included oxidation on methionine, carbamidomethyl on cysteine, and acetylation on protein N-terminus. Normalized LFQ intensities were inputted into Perseus (82), where proteins with greater than 70% missing values were removed. The remaining missing values were imputed from the normal distribution. Protein abundances were Log 2 -transformed, and a Student's t-test corrected for multiple comparisons was performed. Proteins were considered candidate interaction partners of HCMV UL78 if the CRAPome frequency was <30%, FDR-corrected q value < 0.05, and the Log 2 fold change was >1.5.
## Live cell G-protein coupling assay
Receptor G-protein coupling was assessed as previously described (9, 47) using the Nano-Glo Live Cell Assay System (Promega). Briefly, HEK-293 cells were seeded into treated black 96-well plates at a density of 3.0 × 10 4 cells per well. The following day, cells were co-transfected in triplicate with a 1:1 ratio of the indicated GPCR constructs or the empty pcDNA3.1-vector, and Large BiT linked Gα subunit using Fugene4K (Promega), following the manufacturer's recommended procedure. At 18 hours post-transfection, the growth medium was replaced with Opti-MEM media (Thermo Fisher Scientific). At 6 hours post-media replacement, 25 µL of reconstituted Nano-Glo Live Cell assay reagent was added to each well, and plates were briefly incubated with agitation. Luminescence, indicative of G protein coupling, was measured using a Promega GloMax Navigator luminometer. Assay results were transferred to a Microsoft Excel spreadsheet, backgrounds-subtracted, and plotted in GraphPad Prism 10.0 software.
## HiBiT split-luciferase assay
For surface v. total expression assays, NHDFs were seeded into cell culture-treated black 96-well plates at a density of 1.5 × 10 4 cells per well. The following day, triplicate wells were infected with the indicated HiBiT-tagged recombinant viruses at a MOI of 1.0. At 72 hpi, surface vs. total HiBiT expression was evaluated using the Nano-Glo HiBiT Extracel lular and Lytic Detection kits (Promega) following the manufacturer's recommended procedure. For nuclear localization assays, NHDFs were seeded into cell culture-treated 6-well plates at a density of 3.0 × 10 5 cells per well. The following day, replicate wells were infected with the indicated HiBiT-tagged recombinant viruses at a MOI of 1.0. At 72 hpi, cytoplasmic and nuclear extracts were harvested utilizing the NE-PER extraction kit (Thermo Fisher Scientific). Lysates were plated in triplicate into black 96-well plates, and HiBiT expression was evaluated using the Nano-Glo HiBiT Lytic Detection kits (Promega) following the manufacturer's recommended procedure. Luminescence was measured using a Promega GloMax Navigator luminometer. Assay results were transferred to a Microsoft Excel spreadsheet, background-subtracted, normalized to the HiBiT control protein, and % surface expression was determined by using the ratio of extracellular vs. lytic luminescence. Results were analyzed using GraphPad Prism 10.0 software.
## Microscopy
NHDFs were grown on 13 mm glass coverslips and infected at an MOI of 0.5 with WT-HCMV, HB-UL78, or HB-UL78-DAL. At 72 hpi, coverslips were washed with PBS and fixed with 4% paraformaldehyde in PBS. Cells were permeabilized with 0.25% Triton, blocked with normal goat serum, and stained with the indicated antibodies. Coverslips were then washed with PBS and incubated with the appropriate fluorophore-conjugated secondary antibodies. Fluorescence was visualized using a LEICA Stellaris 8 microscope equipped with a 63× objective with an NA of 1.4. The fluorophores were excited using 405 nm and white light lasers. The signals were captured using Leica Stellaris 8 and the Leica Application Suite Software. Images were exported as .tiff files and analyzed using ImageJ software.
## HCMV latency and reactivation assay
hESC-derived CD34 + HPCs were differentiated from WA01 human embryonic stem cells using the commercial STEMdiff Heme feeder-free hematopoietic differentiation kit (Stem Cell Technologies) as previously described (40,52,81). HPCs were cultured in IMDM with 10% BIT serum replacement, stem cell cytokines (stem cell factor, FLT3L, IL-3, and IL-6 [PeproTech]), and penicillin/streptomycin as previously described (11,53,81). CD34 + HPCs were infected with the indicated viruses at a MOI of 2 for 48 hours prior to isolation by fluorescence-activated cell sorting (FACS) using a FACSAria (BD FACS Aria equipped with 488, 633, and 405 nm lasers, running FACS DIVA software) in order to obtain a pure population of viable, GFP + , CD34 + , HPCs as previously described (79). Infected cells were co-cultured in transwells positioned above monolayers of irradiated M2-10B4 and S1/S1 stromal cells. At 14 days post-infection, HPCs were serially diluted in RPMI-1640 medium containing 20% FBS, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, 15 ng/mL granulocyte-colony stimulating factor (G-CSF), and 15 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF), and overlaid onto confluent monolayers of NHDFs cultured in 96-well plates for an extreme limiting dilution assay. To quantify the levels of pre-reactivation infectious virus, a fraction of the HPC cultures was mechanically disrupted and then used in the ELDA. Cell cultures were microscopically visualized for the presence of GFP + weekly, for up to 3 weeks, and the frequency of infectious center production was calculated using ELDA software (41).
## Viral DNA quantification
Primers and probes recognizing HCMV UL141 were used to quantify viral genomes by quantitative real-time PCR (11). Briefly, total DNA was extracted using Trizol (Thermo Fisher) according to the manufacturer's recommendations. Dilutions of purified HCMV BAC DNA were used to create a standard curve. Total DNA was added to each reaction well of TaqMan FastAdvance PCR master mix (Applied Biosystems), and samples were analyzed in triplicate on a StepOnePlus TaqMan PCR machine (Applied Biosystems) with an initial activation at 50°C for 2 min and 95°C for 20 s, followed by 40 cycles of 1 s at 95°C and 20 s at 60°C. TaqMan results were analyzed using ABI StepOne software and graphed using GraphPad Prism 10.0 software.
## References
1. Goodrum (2016) "Human cytomegalovirus latency: approaching the Gordian knot" *Annu Rev Virol*
2. Söderberg-Nauclér, Streblow, Fish et al. (2001) "Reactivation of latent human cytomegalovirus in CD14 + monocytes is differentiation dependent" *J Virol*
3. Söderberg-Nauclér, Fish, Nelson (1997) "Reactivation of latent human cytomegalovirus by allogeneic stimulation of blood cells from healthy donors" *Cell*
4. Panda, Parashar, Viswanathan (2023) "An update on current antiviral strategies to combat human cytomegalovirus infection" *Viruses*
5. Perera, Wills, Sinclair (2021) "HCMV antivirals and strategies to target the latent reservoir" *Viruses*
6. Gilbert, Boivin (2005) "Human cytomegalovirus resistance to antiviral drugs" *Antimicrob Agents Chemother*
7. Mcsharry, Avdic, Slobedman (2012) "Human cytomegalovirus encoded homologs of cytokines, chemokines and their receptors: roles in immunomodulation" *Viruses*
8. Krishna, Wass, Dooley et al. (2021) "CMV-encoded GPCR pUL33 activates CREB and facilitates its recruitment to the MIE locus for efficient viral reactivation" *J Cell Sci*
9. Medica, Denton, Diggins et al. (2025) "Third intracellular loop of HCMV US28 is necessary for signaling and viral reactivation" *J Virol*
10. Boeck, Stowell, Connor et al. (2018) "The human cytomegalovirus US27 gene product constitutively activates antioxidant response element-mediated transcription through G βγ , phosphoinosi tide 3-kinase, and nuclear respiratory factor 1" *J Virol*
11. Crawford, Caposio, Kreklywich et al. (2019) "Human cytomegalovirus US28 ligand binding activity is required for latency in CD34 + hematopoietic progenitor cells and humanized NSG mice" *mBio*
12. Streblow, Soderberg-Naucler, Vieira et al. (1999) "The human cytomegalovirus chemokine receptor US28 mediates vascular smooth muscle cell migration" *Cell*
13. Tadagaki, Nakano, Yamanishi (2005) "Human herpesvirus 7 open reading frames U12 and U51 encode functional β-chemokine receptors" *J Virol*
14. Zhen, Bradel-Tretheway, Sumagin et al. (2005) "The human herpesvirus 6 G protein-coupled receptor homolog U51 positively regulates virus replication and enhances cell-cell fusion in vitro" *J Virol*
15. Kaptein, Beisser, Gruijthuijsen et al. (2003) "The rat cytomega lovirus R78 G protein-coupled receptor gene is required for production of infectious virus in the spleen" *J Gen Virol*
16. Oliveira, Shenk (2001) "Murine cytomegalovirus M78 protein, a G protein-coupled receptor homologue, is a constituent of the virion and facilitates accumulation of immediate-early viral mRNA" *Proc Natl Acad Sci*
17. Milne, Mattick, Nicholson et al. (2000) "RANTES binding and down-regulation by a novel human herpesvirus-6 beta chemokine receptor" *J Immunol*
18. Sharp, Davis-Poynter, Farrell (2009) "Analysis of the subcellular trafficking properties of murine cytomegalovirus M78, a 7 transmem brane receptor homologue" *J Gen Virol*
19. Niemann, Reichel, Stamminger (2014) "Intracellular trafficking of the human cytomegalovirus-encoded 7-trans-membrane protein homologs pUS27 and pUL78 during viral infection: a comparative analysis" *Viruses*
20. Wagner, Arnold, Wu et al. (2012) "The 7-transmembrane protein homologue UL78 of the human cytomegalovirus forms oligomers and traffics between the plasma membrane and different intracellular compart ments" *Arch Virol*
21. Chen, Li, Zhou et al. (2024) "A homotrimeric GPCR architecture of the human cytomega lovirus revealed by cryo-EM" *Cell Discov*
22. O'connor, Shenk (2012) "Human cytomegalovirus pUL78 G proteincoupled receptor homologue is required for timely cell entry in epithelial cells but not fibroblasts" *J Virol*
23. Beisser, Grauls, Bruggeman et al. (1999) "Deletion of the R78 G protein-coupled receptor gene from rat cytomegalovirus results in an attenuated, syncytium-inducing mutant strain" *J Virol*
24. Kreklywich, Smith, Jones et al. (2014) "Fluorescence-based laser capture microscopy technology Full-Length Text Journal of Virology November"
25. "facilitates identification of critical in vivo cytomegalovirus transcriptional programs" *Methods Mol Biol*
26. Streblow, Van Cleef, Kreklywich et al. (2007) "Rat cytomegalovirus gene expression in cardiac allograft recipients is tissue specific and does not parallel the profiles detected in vitro" *J Virol*
27. Bittencourt, Wu, Bridges et al. (2014) "The M33 G proteincoupled receptor encoded by murine cytomegalovirus is dispensable for hematogenous dissemination but is required for growth within the salivary gland" *J Virol*
28. Yunis, Farrell, Bruce et al. (2018) "Murine cytomegalovirus degrades MHC class II to colonize the salivary glands" *PLOS Pathog*
30. Tschische, Tadagaki, Kamal et al. (2011) "Heteromerization of human cytomegalovirus encoded chemokine receptors" *Biochem Pharmacol*
31. Tadagaki, Tudor, Gbahou et al. (2012) "Human cytomegalovirus-encoded UL33 and UL78 heteromerize with host CCR5 and CXCR4 impairing their HIV coreceptor activity" *Blood*
32. Buxton, Jones, Molenaar et al. (1987) "Characterization and autoradiographic localization of β-adrenoceptor subtypes in human cardiac tissues" *Br J Pharmacol*
33. Cheng, Mclaughlin, Verderame et al. (2009) "Dependence on nuclear localization signals of the opioid growth factor receptor in the regulation of cell proliferation" *Exp Biol Med (Maywood)*
34. Wang, Wang, Yang et al. (2009) "CXCR4 nuclear localization follows binding of its ligand SDF-1 and occurs in metastatic but not primary renal cell carcinoma" *Oncol Rep*
35. Speetjens, Liefers, Korbee et al. (2009) "Nuclear localization of CXCR4 determines prognosis for colorectal cancer patients" *Cancer Microenviron*
36. Medica, Crawford, Denton et al. (2023) "Proximity-dependent mapping of the HCMV US28 interactome identifies RhoGEF signaling as a requirement for efficient viral reactivation" *PLOS Pathog*
37. Krishna, Humby, Miller et al. (2019) "Human cytomegalovirus G protein-coupled receptor US28 promotes latency by attenuating c-fos" *Proc Natl Acad Sci*
38. Krishna, Wass, Sridharan et al. (2020) "The requirement for US28 during cytomegalovirus latency is independent of US27 and US29 gene expression" *Front Cell Infect Microbiol*
39. Jones, Haese, Gatault et al. (2020) "Rat cytomegalovirus virion-associated proteins R131 and R129 are necessary for infection of macrophages and dendritic cells" *Pathogens*
40. Michel, Milotić, Wagner et al. (2005) "The human cytomegalovirus UL78 gene is highly conserved among clinical isolates, but is dispensable for replication in fibroblasts and a renal artery organ-culture system" *J Gen Virol*
41. Diggins, Pham, Mitchell et al. (2022) "Viral microRNA regulation of Akt is necessary for reactivation of human cytomegalovirus from latency in CD34 + hematopoietic progenitor cells and humanized mice" *PLOS Pathog*
42. Hu, Smyth (2009) "ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays" *J Immunol Methods*
43. Rovati, Capra, Neubig (2007) "The highly conserved DRY motif of class A G protein-coupled receptors: beyond the ground state" *Mol Pharmacol*
44. Saidak, Blake-Palmer, Hay et al. (2006) "Differential activation of G-proteins by μ-opioid receptor agonists" *Br J Pharmacol*
45. Tobin, Butcher, Kong (2008) "Location, location, location...sitespecific GPCR phosphorylation offers a mechanism for cell-type-specific signalling" *Trends Pharmacol Sci*
46. Schwartz, Frimurer, Holst et al. (2006) "Molecular mechanism of 7TM receptor activation-a global toggle switch model" *Annu Rev Pharmacol Toxicol*
48. Auger, Pease, Shen et al. (2002) "Alanine scanning mutagenesis of CCR3 reveals that the three intracellular loops are essential for functional receptor expression" *Eur J Immunol*
49. Laschet, Dupuis, Hanson (2019) "A dynamic and screeningcompatible nanoluciferase-based complementation assay enables profiling of individual GPCR-G protein interactions" *J Biol Chem*
50. Melnychuk, Streblow, Smith et al. (2004) "Human cytomegalovirus-encoded G protein-coupled receptor US28 mediates smooth muscle cell migration through Gα12" *J Virol*
51. Vomaske, Melnychuk, Smith et al. (2009) "Differential ligand binding to a human cytomegalovirus chemokine receptor determines cell type-specific motility" *PLoS Pathog*
52. Branon, Bosch, Sanchez et al. (2018) "Efficient proximity labeling in living cells and organisms with TurboID" *Nat Biotechnol*
53. Mellacheruvu, Wright, Couzens et al. (2013) "The CRAPome: a contaminant repository for affinity purification-mass spectrometry data" *Nat Methods*
54. Dirck, Diggins, Crawford et al. (2023) "HCMV UL8 interaction with β-catenin and DVL2 regulates viral reactivation in CD34 + hematopoietic progenitor cells" *J Virol*
55. Crawford, Kim, Collins-Mcmillen et al. (2018) "Human cytomegalovirus encodes a novel FLT3 receptor ligand necessary for hematopoietic cell differentiation and viral reactivation" *mBio*
56. Keyes, Hargett, Soland et al. (2012) "HCMV protein LUNA is required for viral reactivation from latently infected primary CD14 + cells" *PLoS One*
57. Rak, Buehler, Zeltzer et al. (2018) "Human cytomegalovirus UL135 interacts with host adaptor proteins to regulate epidermal growth factor receptor and reactivation from latency" *J Virol*
58. Moy, Collins-Mcmillen, Crawford et al. (2023) "UL135 and UL136 epistasis controls reactivation of human cytomegalovirus"
59. Bhosle, Rivera, Chemtob (2019) "New insights into mechanisms of nuclear translocation of G-protein coupled receptors" *Small GTPases*
60. (2025) *Full-Length Text Journal of Virology*
61. Allen, Merlen, Branco et al. (2024) "Understanding the impact of nuclear-localized GPCRs on cellular signalling" *Cell Signal*
62. Xu, Li, Wei et al. (2016) "NMMHC-IIAdependent nuclear location of CXCR4 promotes migration and invasion in renal cell carcinoma" *Oncol Rep*
63. Jensen, Zhang, Flynn (2008) "Trafficking of tachykinin neurokinin 3 receptor to nuclei of neurons in the paraventricular nucleus of the hypothalamus following osmotic challenge" *Neuroscience*
64. Kang, Shi, Xiang et al. (2005) "A nuclear function of β-arrestin1 in GPCR signaling: regulation of histone acetylation and gene transcription" *Cell*
65. Vaniotis, Allen, Hébert (2011) "Nuclear GPCRs in cardiomyocytes: an insider's view of β-adrenergic receptor signaling" *Am J Physiol Heart Circ Physiol*
66. Tadevosyan, Vaniotis, Allen et al. (2012) "G proteincoupled receptor signalling in the cardiac nuclear membrane: evidence and possible roles in physiological and pathophysiological function" *J Physiol*
67. Joyal, Nim, Zhu et al. (2014) "Subcellular localization of coagulation factor II receptor-like 1 in neurons governs angiogenesis" *Nat Med*
68. Kadota, Ou, Shi et al. (2020) "Nucleoporin 153 links nuclear pore complex to chromatin architecture by mediating CTCF and cohesin binding" *Nat Commun*
69. Huang, Zhang, Cheng et al. (2024) "The nuclear pore Y-complex functions as a platform for transcrip tional regulation of FLOWERING LOCUS C in Arabidopsis" *Plant Cell*
70. Kuhn, Capelson (2019) "Nuclear pore proteins in regulation of chromatin state" *Cells*
71. Brickner (2023) "The nuclear pore complex as a platform for epigenetic regulation" *J Cell Biol*
72. Luo, Rosenke, Czornak et al. (2007) "Human cytomegalo virus disrupts both ataxia telangiectasia mutated protein (ATM)-and ATM-Rad3-related kinase-mediated DNA damage responses during lytic infection" *J Virol*
73. Shen, Utama, Wang et al. (2004) "Human cytomegalovirus causes endothelial injury through the ataxia telangiectasia mutant and p53 DNA damage signaling pathways" *Circ Res*
74. Van Domselaar, De Poot, Remmerswaal et al. (2013) "Granzyme M targets host cell hnRNP K that is essential for human cytomegalovirus replication" *Cell Death Differ*
75. Pastor, Shkreta, Chabot et al. (2021) "Interplay between CMGC kinases targeting SR proteins and viral replication: splicing and beyond" *Front Microbiol*
76. Gaddy, Wong, Markowitz-Shulman et al. (2010) "Regulation of the subcellular distribution of key cellular RNA-processing factors during permissive human cytomegalovirus infection" *J Gen Virol*
77. Varnum, Streblow, Monroe et al. (2004) "Identification of proteins in human cytomegalovirus (HCMV) particles: the HCMV proteome" *J Virol*
78. Lischka, Toth, Thomas et al. (2006) "The UL69 transactivator protein of human cytomegalovirus interacts with DEXD/H-Box RNA helicase UAP56 to promote cytoplasmic accumulation of unspliced RNA" *Mol Cell Biol*
79. Toth, Lischka, Stamminger (2006) "RNA-binding of the human cytomegalovirus transactivator protein UL69, mediated by arginine-rich motifs, is not required for nuclear export of unspliced RNA" *Nucleic Acids Res*
80. Winkler, Rice, Stamminger (1994) "UL69 of human cytomegalovi rus, an open reading frame with homology to ICP27 of herpes simplex virus, encodes a transactivator of gene expression" *J Virol*
81. Bhosle, Rivera, Zhou et al. (2016) "Erratum: Nuclear localization of platelet-activating factor receptor controls retinal neovascularization" *Cell Discov*
82. (2016)
83. Umashankar, Goodrum (2014) "Hematopoietic long-term culture (hLTC) for human cytomegalovirus latency and reactivation" *Methods Mol Biol*
84. Sinzger, Hahn, Digel et al. (2008) "Cloning and sequencing of a highly productive, endotheliotropic virus strain derived from human cytomegalovirus TB40/E" *J Gen Virol*
85. Hancock, Crawford, Perez et al. (2021) "Human cytomegalovirus UL7, miR-US5-1, and miR-UL112-3p inactivation of FOXO3a protects CD34 + hematopoietic progenitor cells from apoptosis"
86. Tyanova, Temu, Sinitcyn et al. (2016) "The Perseus computational platform for comprehensive analysis of (prote)omics data" *Nat Methods* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12671158&blobtype=pdf | # Fatty acid synthase may facilitate the trafficking of bovine alpha herpesvirus 1 out of the Golgi apparatus, potentially promoting viral infection
Xiaozhen Ma, Wenyuan Gu, Xuan Li, Xiuyan Ding, Liqian Zhu
## Abstract
The lipogenic enzyme fatty acid synthase (FASN) is essential for the de novo synthesis of fatty acids, and its aberrant expression has been implicated in the develop ment of various diseases. However, the interaction between BoAHV-1 infection and FASN remains to be fully elucidated. In this study, we observed that both viral acute infec tion and latency in bovine trigeminal ganglia (TG) neurons, as well as virus productive infection at later stages in Neuro-2A and MDBK cells, significantly reduced FASN protein levels. Notably, a subset of FASN protein co-localized with the viral glycoprotein gD, as revealed by confocal microscopy in these cell cultures. Knockdown of FASN protein expression using siRNAs or chemical inhibition of FASN by Cerulenin leads to reduced viral replication. Interestingly, a subset of FASN proteins localizes in the Golgi apparatus, and Cerulenin treatment retards the trafficking of virions out of the Golgi. Collectively, our findings reveal a novel mechanism by which FASN regulates viral replication: a portion of the FASN protein located in the Golgi apparatus facilitates viral trafficking out of the Golgi, which is essential for the completion of the viral replication cycle.IMPORTANCE Here, we found that fatty acid synthase (FASN) protein is potentially involved in virus infection both in vitro and in vivo. In terms of mechanism, a subset of FASN protein co-localizes with the viral glycoprotein gD, as revealed by confocal microscopy in both MDBK and Neuro-2A cell cultures. Interestingly, a subset of FASN protein localizes in the Golgi apparatus, and Cerulenin treatment retards the trafficking of virions out of the Golgi. Therefore, we propose that a portion of the FASN protein in the Golgi apparatus plays an important role in viral trafficking out of the Golgi, which is essential for the completion of the viral replication cycle. This represents a novel mechanism of virus infection regulation, which has not been reported in other viruses, although the FASN protein is involved in the regulation of multiple viruses through distinct mechanisms.
impacting the cattle industry worldwide. Reports indicate that it inflicts approximately $3 billion in annual losses on the US cattle industry (8).
The enzyme fatty acid synthase (FASN) is a multifunctional protein that primarily catalyzes the conversion of acetyl-CoA and malonyl-CoA into palmitate, using NADPH as a reducing agent (9,10). This process is essential for the production of cellular lipids, including triglycerides and phospholipids. In addition to its role in fatty acid synthe sis, abnormal expression of FASN is implicated in various diseases, such as metabolic disorders, cancers (9,11,12), and inflammatory diseases (13,14). As a result, FASN has been recognized as a potential therapeutic target for cancer treatment.
Fatty acids are important energy sources, primarily catabolized by fatty acid βoxidation (FAO) in mitochondria. Given that FAO produces more than three times the amount of ATP per mole compared to glucose oxidation, it serves as the preferred energy source for highly metabolized cells under physiological conditions (15). Carnitine palmitoyl-transferase 1 A (CPT1A) is the rate-limiting enzyme of FAO (16). Both FASN and CPT1A are thus key enzymes involved in lipid metabolism: FASN synthesizes fatty acids, while CPT1A transports them into mitochondria for oxidation. We have previ ously reported that BoAHV-1 productive infection affects CPT1A protein expression and localization, a potential mechanism to alter FAO (17). This prompted us to investigate whether and how FASN is involved in BoAHV-1 replication.
In this study, we examined the effects of BoAHV-1 infection on FASN protein expression both in vivo and in vitro and assessed its impact on viral productive infection in cell culture using siRNAs and chemical inhibitors. Our results demonstrate that FASN is involved in BoAHV-1 infection both in vivo and in vitro. For the first time, we have identified a novel mechanism by which a portion of the FASN protein in the Golgi apparatus is required in viral trafficking out of the Golgi. Our findings will extend our understanding of the viral replication mechanisms involving FASN molecules.
## RESULTS
## Detection of FASN protein expression in bovine TG neurons during viral acute infection
To assess the involvement of the FASN protein in BoAHV-1 infection in vivo, healthy calves that tested negative for BoAHV-1 antibodies were used for the viral infection assay. Trigeminal ganglia (TG) tissues were harvested at 4 and 60 days post-infection, representing the acute phase and latency of viral infection, respectively, as previously described (18). Immunohistochemistry (IHC) was employed to evaluate the changes in FASN protein expression in response to viral infection in these TG neurons. As shown in Fig. 1A, extensive staining of FASN protein (denoted by black arrows) was readily detected in a subset of TG neurons from mock-infected calves, particularly in the cytoplasm. Meanwhile, moderate staining of FASN (denoted by filled blue circles) was readily detected in latently infected TG neurons. In contrast, faint staining was observed in the cytoplasm of TG neurons from BoAHV-1 acutely infected calves (Fig. 1A). Interestingly, FASN+ nuclei (denoted by filled green/blue circles) were readily observed in neurons from both acutely infected and latent cows. Quantitative analysis revealed that the proportion of FASN+ neurons, including both cytoplasmic and nuclear staining, was 43.80%, 10.06%, and 21.55% in calves with mock infection, acute infection, and latency, respectively (Fig. 1B). Although the proportion of FASN+ staining in latently infected TG neurons was higher than that in acutely infected calves, it was still lower than that in mock-infected neurons. These observations suggest that both acute and latent BoAHV-1 infection lead to a reduction in FASN protein expression in TG neurons, albeit to different extents. This indicates that FASN may play a role in the viral infection process.
## BoAHV-1 infection decreased FANS steady-state protein expression in cell cultures
Both bovine kidney (MDBK) and mouse neuroblastoma cells (Neuro-2A) were employed to investigate the effects of viral productive infection had on FASN protein expression in vitro. MDBK cells support the virus productive infection with high efficacy. BoAHV-1 can infect Neuro-2A cells, albeit with lower efficiency (19)(20)(21). When MDBK cells were infected with BoAHV-1 at a multiplicity of infection (MOI) of 1 over various time points, FASN protein levels significantly decreased at 24 hpi, showing a reduction of approximately 30.48% compared to the mock-infected control (Fig. 2A andB). When MDBK cells were infected with BoAHV-1 at increasing MOIs ranging from 0.1 to 10 for 24 h, respectively, FASN protein levels gradually decreased, to an extent correlated with increasing MOIs (Fig. 2C). Quantitative analysis indicated that the protein levels reduced to approximately 37.46% and 20.09% following infection with MOI of 1 and 10, respectively (Fig. 2D). The viral infection-induced depletion of FASN was also observed in Neuro-2A cells (Fig. 2E). Compared to the control, FASN protein levels decreased to approximately 68.85%, 8.42%, and 7.6% after infection for 24, 36, and 48 h, respectively (Fig. 2F). Of note, viruses grow poorly in Neuro-2A cells (21). The dramatic depletion of FASN protein in Neuro-2A cells induced by viral infection was not due to virus-induced cell death, as determined by the CCK-8 assay (Fig. 2G). FASN mRNA levels increased approximately 2.25-fold in MDBK cells and 1.90-fold in Neuro-2A cells following virus infection for 24 h, relative to mock-infected controls (Fig. 2H). The decreased steady-state protein levels of FASN do not corroborate the increased mRNA levels in response to viral infection. The decreased FASN protein expression observed during BoAHV-1 productive infection in both MDBK and Neuro-2A cells is consistent with that observed in TG neurons during viral lytic infection.
## A portion of FASN co-localizes with the viral glycoprotein gD
An immunofluorescence assay (IFA) was performed using antibodies against FASN and the viral protein gD to determine whether BoAHV-1 infection affects FASN localization in vitro. Time points of 4, 16, and 24 hpi in MDBK cells were selected for this analysis. As a result, FASN was readily detected in the cytosol in mock-infected cells. At 4 hpi, FASN exhibited a localization pattern similar to that of the mock-infected control. However, at 16 and 24 hpi, re-localization of FASN was observed, with a subset of FASN showing highlighted staining forming irregular shapes in the cytoplasm. Interestingly, FASN colocalized well with the viral protein gD in virus-infected MDBK cells at both 16 and 24 hpi (Fig. 3). Similarly, co-localization of virion-associated protein with FASN was observed in the cytoplasm of virus-infected Neuro-2A cells at 48 and 72 hpi (Fig. 4). Although the overall localization of FASN was not dramatically altered in these cells, as seen in virusinfected MDBK cells, these findings suggest that FASN proteins are redistributed to areas overlapping or in close proximity to sites containing viral protein gD or virions.
## A subset of FASN is located in the Golgi apparatus, and its content is reduced following BoAHV-1 productive infection
The distribution profile of FASN proteins induced by viral infection appears similar to that of the Golgi apparatus, as we have previously reported (22). To determine whether a subset of FASN proteins accumulates in the Golgi apparatus, IFA was performed to detect FASN and the Golgi marker protein GP73 in the same cells. We found that a subset of FASN protein co-localized well with the Golgi marker protein GP73, regardless of viral infection, indicating its accumulation in the Golgi apparatus (Fig. 5A, zoom-in areas).
Next, the Golgi apparatus fractions were isolated from MDBK cells using a commercial Golgi purification kit and subjected to Western blot analysis to probe FASN protein. The Golgi marker protein GOLGA1 was used as a loading control. Our results indicated that FASN protein was readily detected in the Golgi fractions, regardless of viral infection. However, its accumulation in the Golgi fractions was significantly reduced following viral infection (Fig. 5B). Quantitative analysis indicated that the Golgi accumulated FASN protein levels decreased to approximately 35.75%, following virus infection (Fig. 5C).
Thus, using two independent methods, we demonstrated that a portion of FASN protein localizes in the Golgi apparatus, and its content is reduced following viral infection. These findings further support our findings that viral productive infection leads to the re-localization of FASN protein.
## FASN plays an important role in BoAHV-1 productive infection in both MDBK and Neuro-2A cells
The roles of FASN in BoAHV-1 productive infection were analyzed in MDBK cells using FASN-specific siRNAs. Four commercially available siRNAs, designated as siRNA1, siRNA2, siRNA3, and siRNA4, could effectively reduce FASN protein levels to varying extents (Fig. 6A). FASN protein levels were reduced to approximately 20.41%, 27.54%, 36.29%, and 67.93% by these four individual siRNAs, respectively, compared to the scrambled siRNA control (Fig. 6B). In the context of viral infection, siRNA1, siRNA2, and siRNA3 still maintained their efficacy in knocking down FASN protein expression (Fig. 6C). Compared to the scrambled siRNA control, FASN protein levels were reduced to approximately 43.60%, 55.43%, and 75.43% by these three individual siRNAs, respectively, in the context of virus infection (Fig. 6D). To explore the effects of FASN knockdown on viral productive infection, the expression of virion-associated proteins was detected via Western blot using an antibody produced by immunization of purified BoAHV-1 virions. As a result, a panel of bands indicative of distinct virion-associated proteins, denoted as "a, " "b, " and "c, " respectively, was readily detected by this antibody. The protein levels of all three bands were significantly decreased following transfection with either siRNA2 or siRNA3 (Fig. 6E). Specifically, compared to the scrambled siRNA control, the levels of band "a" were reduced to 45.74% and 31.09%, band "b" to 49.39% and 42.41%, and band "c" to 20.49% and 18.81% by siRNA2 and siRNA3, respectively (Fig. 6F). To further clarify how FASN knockdown affects viral productive infection, the mRNA levels of viral regulatory proteins bICP4 and bICP27, viral DNA polymerase, and viral glycoprotein gC, all of which are essential for viral productive infection, were examined in productively infected MDBK cells. Compared to the scrambled siRNA control, the levels of bICP4 decreased to 75.24% and 59.85%, bICP27 to 26.94% and 14.75%, viral DNA polymerase to 7.16% and 2.53%, and gC to 31.46% and 20.86% by siRNA2 and siRNA3, respectively (Fig. 6G). The reduced mRNA levels of the detected viral genes further confirmed that FASN protein knockdown decreases viral replication. Significance was assessed by standard t-test (ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001).
Cerulenin is a widely used FASN-specific inhibitor that contains an epoxy group, which mediates the inhibition of the β-ketoacyl-reductase activity of FASN (23,24). In this study, the effects of Cerulenin on BoAHV-1 productive infection in MDBK cell cultures were further investigated. To achieve this, virus-infected cell cultures were treated with 5 µM or 10 µM Cerulenin. Three distinct bands, labeled as "α, " "β," and "γ," were clearly detected with shorter exposure times. The intensity of all these bands was significantly reduced by treatment with either 5 µM or 10 µM Cerulenin compared to the DMSO control (Fig. 7A). Specifically, the levels of band "α" were reduced to 66.07% and 46.20%, band "β" to 49.46% and 39.41%, and band "γ" to 73.57% and 30.40% by 5 and 10 µM Cerulenin, respectively, compared to the DMSO control (Fig. 7B). iCRT14, a chemical inhibitor of β-catenin signaling that has been shown to inhibit BoAHV-1 productive infection in cell cultures (25), was used as a control. As expected, iCRT14 significantly reduced the protein levels of all the bands, further validating the inhibitory effects of Cerulenin on BoAHV-1 productive infection in cell cultures (25). Quantitative RT-PCR analysis revealed that the levels of viral genome were reduced to approximately 36.17% following treatment with 10 µM Cerulenin, respectively, compared to the DMSO-treated control (Fig. 7C). Consistent with these findings, the treatment of virus-infected cells with 10 µM Cerulenin led to a reduction of viral titers by approximately 1.48-and 0.88-log, when cells were infected with the virus at MOI of either 0.1 or 1, respectively (Fig. 7D). Furthermore, treatment with 10 µM Cerulenin reduced the mRNA levels of the viral regulatory protein bICP27, viral DNA polymerase, and viral glycoprotein gC to 30.20%, 14.36%, and 59.76%, respectively, compared to the DMSO control (Fig. 7E). Notably, 10 µM Cerulenin had no significant cytotoxic effects on MDBK cells, as demonstrated by the Trypan blue exclusion test (data not shown). This indicates that the observed reduction in virus yields was not due to the cytotoxicity of the inhibitor. Collectively, studies using both siRNAs and chemical inhibitors consistently support this conclusion that FASN plays a crucial role in BoAHV-1 productive infection in MDBK cells.
Given that BoAHV-1 productive infection in Neuro-2A cells affects FASN protein expression, we further investigated whether FASN also plays a role in viral infection in the They were immunostained with antibodies against the FASN protein (Red; Proteintech, cat# 10624-2-AP, 1:600) and the virion-associated protein (Green; VMRD, cat# P170703-001, 1:2,000). Nuclei were counterstained with DAPI (Blue). Immunofluorescence was visualized, and images were captured using confocal microscopy (Zeiss). These images are representative of results from three independent experiments. Scale bars = 20 µm. neural cell line using FASN knockdown and chemical inhibition. Among the four commercially available siRNAs (designated as siRNA1, siRNA2, siRNA3, and siRNA4), only siRNA1 effectively reduced FASN protein expression, decreasing it to 54.54% relative to that of the scrambled siRNA control (Fig. 8A andB). Transfection with siRNA1 significantly decreased the protein levels of viral glycoprotein gC (Fig. 8C). Compared to the scram bled siRNA control, the levels of gC protein were reduced to 33.91% by siRNA1 (Fig. 8D). The expression of viral protein gC was significantly reduced by 5 and 10 µM Cerulenin (Fig. 8E). Quantitative analysis indicated that the intensity of gC protein was reduced to 50.30% by 5 µM of Cerulenin and 28.00% by 10 µM Cerulenin, compared to the DMSO control (Fig. 8F).
## FASN-specific inhibitor, Cerulenin, traps viral protein gD in the Golgi appara tus
The Golgi is proposed as the key site for de novo envelopment of HSV-1/BoAHV-1 capsids and their subsequent egress to the plasma membrane (26)(27)(28). Since our findings indicate that partial FASN proteins located in the Golgi apparatus associate with viral protein gD (Fig. 3 to 5). This led us to hypothesize that the FASN signaling pathway may influence viral trafficking out of this organelle. To investigate this, we treated BoAHV-1infected MDBK cells with Cerulenin at 20 hpi for a duration of 4 h. The cells were then collected at 24 hpi and subjected to Golgi isolation, as shown in the diagram (Fig. 9A). Monensin, a known chemical inhibitor of Golgi trafficking (29), was used as a control. When the isolated Gogi fractions were subjected to Western blot assay, we found that gD protein levels were significantly increased in the Golgi fractions by treatment with 10 µM of Cerulenin. Unexpectedly, treatment with Monensin resulted in a dramatic depletion of At 36 hpi, the cells were infected with BoAHV-1 at an MOI of 1. After 24 h of infection, the cells were collected and subjected to Western blot analysis using antibodies against FASN (Proteintech, cat# 10624-2-AP, 1:10,000) (C) and virus-associated proteins (VMRD, cat# P170703-001, 1:5,000) (E). (B, D, and F) The band intensities were quantified using the free software Image J. The intensity of each band was first normalized to that of the respective loading control Tubulin, and then normalized to that of the control transfected with scramble siRNA, which was arbitrarily set to 100%. (G) MDBK cells in six-well plates were transfected with either scrambled siRNA (150 pmol) or the indicated siRNAs (150 pmol). At 36 h post-transfection, the cells were infected with the virus at an MOI of 1 for 24 h. Subsequently, total RNA was extracted from the cells for the detection of viral mRNA using RT-qPCR. Primers specific for bICP4, bICP27, viral DNA polymerase, and viral protein gC were used, respectively. The data shown are means of three independent experiments, with error bars representing standard deviations.
Significance was assessed using a Student's t-test (ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001).
the viral protein gD in Golgi fractions (Fig. 9B). Quantitative analysis showed that gD protein levels increased to approximately 3.50-fold by 10 µM of Cerulenin, decreased to 7% by 10 µM of Monensin in comparison to the DMSO-treated control (Fig. 9C). Of note, two bands corresponding to the gD protein were detected using the gD-specific monoclonal antibody (VMRD, cat# 1B8-F11), as previously reported (30). These bands likely represent different glycosylation states of the protein. Collectively, these data suggest that Cerulenin treatment leads to the accumulation of gD, indicative of virions, in the Golgi apparatus. Therefore, FASN may play a crucial role in viral trafficking out of the Golgi apparatus.
## DISCUSSION
To date, manipulation of FASN protein expression for efficient replication has been reported in numerous viruses in the following way. Infections with CSFV, BVDV, and HCV have been shown to enhance FASN expression (31)(32)(33). FASN levels do not significantly change during DENV infection (34). Here, our studies suggest that BoAHV-1 either acute infection in cattle trigeminal ganglia (TG) neurons or productive infection in cell cultures, including MDBK and Neuron-2A cells, leads to a unanimous decrease of FASN protein expression (Fig. 1 and2). Though BoAHV-1 productive infection leads to depletion of FASN, FASN still plays an important role in viral productive infection, as demonstrated using both siRNAs and chemical inhibitors (Fig. 6 to 8). Thus, FASN steady-state protein expression may be differentially affected by distinct viruses in virus-specific manners. Of note, BoAHV-1 productive infection induces host shut-off effects, and viral products such as the virus-encoded bICP0 protein are capable of mediating protein degradation (35). Therefore, elucidating the mechanisms underlying FASN depletion in virus-infected cells deserves further study in the future, which would significantly contribute to our comprehensive understanding of the detailed interplay between BoAHV-1 productive infection and FASN signaling. Moreover, the increased nuclear accumulation of FASN protein in a subset of latently infected TG neurons suggests that FASN may play a role in regulating viral latency. This intriguing possibility warrants further investigation in future studies.
FASN has been observed to have impacts on the infection of several viruses. For example, FASN participates in the formation of the CSFV replication complex, which is associated with the endoplasmic reticulum, through interaction with the viral protein (nonstructural protein 4B) NS4B, involving Rab18 protein (32). Similarly, Dengue virus protein NS3 redistributes FASN protein to the sites of viral replication to facilitate virus replication (34,36). It seems that the localization of FASN to viral replication sites or replication complexes may represent a mechanism to facilitate virus infection observed in several viruses. Here, we found that the viral productive infection at later stages leads to re-localization of FASN protein, as demonstrated by making FASN puncta formed as detected under confocal microscope, and a portion of FASN protein located in the Golgi apparatus (Fig. 3 to 5). Since the virus productive infection leads to re-localization of the Golgi apparatus (22), it is no wonder that the Golgi harboring FASN will show a distinct distribution profile in response to viral infection. Of note, Golgi apparatuses with Cerulenin (10 µM) or Monensin (10 µM) for a duration of 4 h prior to the termination of infection. At 24 hpi, the cells were collected to purify the Golgi apparatus using a commercial purification kit (Beijing Biolabo Technology, cat# HR0247-50T). Subsequently, Western blot analysis was performed to detect the protein levels of gD in the Golgi fractions. GOLGA1, a marker of the Golgi apparatus, was used as a loading control and for subsequent quantitative analysis.
(C) The band intensity was analyzed with the free software Image J. The intensity of each band was first normalized to that of the respective loading control GOLGA1, and then normalized to that of the control treated with DMSO, which was arbitrarily set to 1. The data shown are means of three independent experiments with error bars indicating standard deviations. Significance was assessed with a Student's t-test (*P < 0.05, **P < 0.01).
are important sites for the viral replication cycle. We have previously reported that p-PLC-γ1(S1248) plays an important role in facilitating the viral trafficking from Golgi to the cell membrane (22). In the current study, we found that a subset of FASN protein co-localizes with the viral protein gD and virion-associated protein as detected in MDBK and Neuro-2A cells, respectively (Fig. 3 and4). Importantly, we observed that a 4 h treatment with Cerulenin before termination of the infection process led to higher levels of accumulation of viral protein gD in the Golgi apparatus (Fig. 9). Therefore, FASN may play a crucial role in viral trafficking out of the Golgi apparatus, which could significantly expand our understanding of its role in viral infections within the virology community. Nevertheless, elucidating the detailed mechanisms by which FASN regulates viral trafficking out of the Golgi remains an intriguing area of study, deserving extensive investigation in the future. Surprisingly, treating virus-infected MDBK cells with Monensin, a known inhibitor of Golgi trafficking, led to a dramatic depletion of the viral protein gD in Golgi fractions (Fig. 9B). Future studies elucidating the mecha nism underlying Monensin-induced gD depletion will be of great interest, as they may enhance our understanding of the virus replication process.
Importantly, FASN has been demonstrated to be critical for the replication or pathogenicity of a range of viruses, such as hepatitis C virus (HCV) (37), influenza virus (38), Epstein-Barr virus (EBV) (39,40), and Chikungunya virus (CHIKV) (41). Of note, among herpesviruses, only EBV, a member of the gammaherpesvirus family, has been previously reported to interact with FASN. Here, we provide evidence that FASN also plays a significant role in the infection of BoAHV-1, a member of the alphaherpesvirus family.
In our study, we provide evidence showing FASN protein is potentially involved in BoAHV-1 lytic infection. A subset of the FASN protein was found to be co-localized with the viral protein gD, providing a possibility to affect virus infection. Moreover, for the first time, we identified that FASN may play a crucial role in viral trafficking out of the Golgi apparatus. These findings would contribute to our understanding of the mechanism of virus replication.
## MATERIALS AND METHODS
## Virus and cell cultures
MDBK cells were purchased from the Chinese model culture preservation center, Shanghai, China. Neuro-2A cells were kindly provided by Professor Dongli Pan from Zhejiang University. These cells were routinely maintained and passaged in DMEM supplemented with 10% fetal bovine serum (FBS). BoAHV-1 strain NJ-16-1, isolated from commercial bovine semen samples (42), was propagated in MDBK cells. Aliquots of virus stocks were stored at -70°C until use.
## Antibodies and reagents
FASN polyclonal antibody (pAb) (cat#10624-2-AP) and GP73 mouse monoclonal antibody (mAb) (cat# 66331-1-lg) were provided by Proteintech (Rosemont, IL, USA). FASN rabbit pAb (cat# A21182), β-Tubulin rabbit pAb (cat# AC015-200ul), GOLGA1 rabbit pAb (cat# A14688), and HRP-conjugated Donkey anti-goat IgG (H + L) (cat# AS031) were ordered from Abclonal Technology (Woburn, MA, USA). HRP-conjugated goat anti-mouse IgG (cat# BF03001) and HRP-labeled goat anti-rabbit IgG (cat# BF03008) were purchased from Biodragon (Wuhan, China). BoAHV-1 gD mouse mAb (cat# 1B8-F11), gC mouse mAb (cat# F2), and goat anti-IBR serum (cat# P170703-001) were purchased from VMRD (Puliman, WA, USA). Alexa FluorTM 488 goat anti-rabbit IgG (H + L) (cat# A11008), Alexa Fluor 647-conjugated goat pAb to rabbit IgG (cat# ab150079), and Alexa FluorTM 633 goat anti-mouse IgG (H + L) (cat# A21050) were provided by Invitrogen Life Technologies (Waltham, MA, USA). FASN-specific inhibitor Cerulenin (cat# HY-A0210) was ordered from MedChemExpress (Monmouth Junction, NJ, USA). Monensin (cat#S1753) was bought from Beyotime Biotechnology (Shanghai, China). FASN-specific siRNA was provided by Genepharma (Shanghai, China).
## Western blotting analysis
Cell lysates of either whole-cell extracts or cellular fractions of Golgi apparatus were prepared using RIPA lysis buffer (1 × PBS, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease inhibitor cocktail. They were boiled in Laemmli sample buffer for 5 min, subsequently subjected to separation on SDS-PAGE (8% or 10%), and then transferred to polyvinylidene fluoride (PVDF) membranes. Immunoreactive bands were visualized using the Clarity Western ECL Substrate from NCM Biotech (cat# P10300).
For the designated studies, the band intensity was quantitatively analyzed with the free software Image J program (https://imagej.nih.gov/ij/download.html, accessed on 1 December 2020). Significance was assessed with a student's t-test by using GraphPad Prism software (v8.0). P values of less than 0.05 (*P < 0.05) were considered significant for all the calculations.
## Immunofluorescence assay
MDBK cells in 24-well chamber slides (Nunc Inc., IL, USA) were either mock infected or infected with BoAHV-1 (MOI = 1). After infection for indicated time lengths such as 4, 16, and 24 hours (h), cells were fixed with 4% paraformaldehyde in PBS for 10 min at room temperature, permeabilized with 0.25% Triton X-100 in PBS for 10 min at room temperature, and blocked with 1% BSA in PBST for 1 h followed by incubation with indicated antibody in 1% BSA in PBS overnight at 4°C. After three washings, cells were incubated with secondary antibody labeled with distinct fluorescent dyes for 1 h in the dark at room temperature. After three washings with PBS, nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole). Slides were covered with coverslips by using an antifade mounting medium (Electron Microscopy Sciences, cat# 50-247-04). Images were captured using a confocal microscope (Zeiss).
## siRNA transfection assay and virus infection assay
To screen the efficacy of the designated siRNAs, MDBK cells in six-well plates were transfected with scrambled siRNA or three FASN-specific siRNA of 150 pmol, provided by Genepharma (Shanghai, China). At 48 h post-transfection, cell lysates were prepared by using cell lysis buffer as described above and subjected to Western blot to detect the protein levels of FASN.
To determine the effects of siRNAs on BoAHV-1 infection, MDBK cells in six-well plates were transfected with scrambled siRNA or three FASN-specific siRNA of 150 pmol, provided by Genepharma (Shanghai, China). At 36 h post-transfection, the cells were infected with BoAHV-1 (MOI = 1) for 24 h. After infection for 24 h, the cell lysates were prepared and subjected to Western blotting to detect the protein expression of Virion-associated protein using a commercial antibody against purified virions.
## Cerulenin treatment of MDBK cells during virus infection
MDBK cells that were confluent in 24-well plates were infected with BoAHV-1 (MOI = 1) along with the treatment of chemical Cerulenin (MCE, cat# HY-A0210) at the indica ted concentration for 1 h at 37°C. After three washes with PBS, a fresh medium with Cerulenin at the indicated concentrations was added to each well. After infection for 24 h, viral yields were titrated in MDBK cells, respectively. Cell cultures treated with DMSO were used as a control. The results were expressed as TCID 50 /mL calculated using the Reed-Muench formula (43).
## Quantification of viral genome by quantitative PCR
MDBK cells that were confluent in 6-well plates were infected with BoAHV-1 at an MOI of 1 for 24 h, along with treatment with either DMSO control or 10 µM Cerulenin. Then genomic DNA was extracted using a DNA extraction kit (Tiangen, DP304), following the manufacturer's protocols. The purified DNA served as templates for qPCR to measure viral genomic levels using gene-specific primers as previously reported (44,45). The primer sequences used were as follows: gB (forward reverse primer 5′-TGTGGACCTAAA CCTCACGGT-3′ and reverse primer 5′-GTAGTCGAGCAGACCCGTGTC-3′), glyceraldehyde-3phosphate dehydrogenase (GAPDH) (forward primer: 5′ CCATGGAGAAGGCTGGGG-3′ and reverse primer: 5′ AAGTTGTCATGGATGACC-3′). The analysis of GAPDH levels served as an internal control and subsequent normalization of gene expression. qPCR was performed on the ABI 7500 fast real-time system (Applied Biosystems, CA). The data were analyzed using the 2 -ΔΔCT method.
## Quantification RT-PCR assay
Total RNA was extracted using TRIzol LS Reagent (Ambion, Cat: 10296010) according to the manufacturer's protocol. One microgram of freshly prepared RNA was used as a template for the synthesis of first-strand cDNA with commercial random hexamer primers for viral mRNA detection, employing the Thermoscript RT-PCR system Kit (Invitrogen, catalog #11146-024). The cDNA products served as templates for qPCR to measure FASN or viral mRNA levels using gene-specific primers as in previous studies (44,46,47). The primer sequences used were as follows: FASN (forward reverse primer 5′-ATTGTGGGC GGGATCAACCT-3′, reverse primer 5′-CGGCAATACCCGTTCCCTGA-3′). viral DNA polymerase (forward reverse primer 5'-GCGAGTACTGCATCCAAGAT-3′ and reverse primer 5'-AATCTGCTGCCCGTCAAA-3′). bICP4 (forward reverse primer 5′-GCCACAGCTC GTTCATCAC-3′ and reverse primer 5′-GCTTCTGGTCGCAGTCGTAG-3′), bICP27 (forward reverse primer 5′-AAACCTGGTAGACGCACTGG-3′ and reverse primer 5′-ACGATAGGGTCT TTGGTGCG-3′), gC (forward reverse primer 5′-ACTATATTTTCCCTTCGCCCG-3′ and reverse primer 5′-TGTGACTTGGTGCCCATG-3′), and 18sRNA (forward reverse primer 5′-GTAAC CCGTTGAACCCCATT-3′, and 5′-CCATCCAATCGGTAGTAGCG-3′). The analysis of 18sRNA mRNA served as an internal control. qPCR was performed on the ABI 7500 fast realtime system (Applied Biosystems, CA). Separate amplification of GAPDH was used to normalize gene expression. The data were analyzed using the 2 -ΔΔCT method.
## Infection of calves
Four-month-old female Holstein cows, which had not been vaccinated against BoAHV-1 and were antibody-negative for BoAHV-1 (as determined using a commercial BoAHV-1 IgG indirect ELISA kit; BioStone Animal Health, Southlake, TX, USA; catalog number 10074-05), were used in these studies. Female cows were usually used for BoAHV-1 infection studies. This preference can be partly explained by the fact that females generally mount a stronger immune response to viral infections compared to males. Thus, female animals are more likely to survive (48,49), which is advantageous for the study. In addition, female animals often have a more consistent hormonal environ ment during certain stages of their life cycle, which can reduce variability in exper imental results (50,51). Particularly, BoAHV-1 productive infection and latency can be differentially affected by hormones, such as progesterone (52). The calves were inoculated in each nostril and eye with 1 mL of DMEM containing 1 × 10 7 PFU of virus, without scarification, for a total of 4 × 10 7 PFU per animal, as previously described (53). The calves were maintained under strict isolation conditions and were adminis tered antibiotics before and after BoAHV-1 infection to prevent secondary bacterial infections, as described elsewhere (54). The time point of acute infection and latent infection was selected as 4 and 60 days post-infection, as previously described (53). Following euthanasia, the TG tissues were collected, sectioned into small pieces, and processed according to standard histopathological protocols, including formalin fixation and paraffin embedding.
## IHC assay
TG tissues embedded in paraffin were cut into thin sections (10 µm), mounted on glass slides, and processed for IHC following standard protocols as described elsewhere (55). The slides were incubated overnight at 4°C in a humidified chamber with antibodies against FASN (Proteintech, cat#10624-2-AP, diluted 1:500 in 1% BSA in PBST [PBS + 0.1% Tween-20]). Then the slides were washed in 1 × PBS and incubated in biotinylated goat anti-rabbit IgG (Vector Laboratories, cat# PK-6101) for 30 min at room temperature in a humidified chamber. Avidin-biotinylated enzyme complex was applied to the slides for 30 minutes of incubation at room temperature. After three washes in 1 × TBS, the slides were incubated with freshly prepared substrate (Vector Laboratories, cat# SK-4800), rinsed with distilled water, and counterstained with hematoxylin. Images were captured under a light microscope (Nikon, Tokyo, Japan).
## References
1. Petrini, Iscaro, Righi (2019) "Antibody responses to bovine alphaherpesvirus 1 (BoHV-1) in passively immunized calves" *Viruses*
2. Siddell, Smith, Adriaenssens et al. (2023) "Virus taxonomy and the role of the international committee on taxonomy of viruses (ICTV)" *J Gen Virol*
3. Tikoo, Campos, Babiuk (1995) "Bovine herpesvirus 1 (BHV-1): biology, pathogenesis, and control" *Adv Virus Res*
4. Chowdhury, Coats, Neis et al. (2010) "A bovine herpesvirus type 1 mutant virus with truncated glycoprotein E cytoplasmic tail has defective anterograde neuronal transport in rabbit dorsal root ganglia primary neuronal cultures in a microfluidic chamber system" *J Neurovirol*
5. Rock, Beam, Mayfield (1987) "Mapping bovine herpesvirus type 1 latency-related RNA in trigeminal ganglia of latently infected rabbits" *J Virol*
6. Hodgson, Aich, Manuja et al. (2005) "Effect of stress on viral-bacterial synergy in bovine respiratory disease: novel mechanisms to regulate inflammation" *Comp Funct Genomics*
7. Muylkens, Thiry, Kirten et al. (2007) "Bovine herpesvirus 1 infection and infectious bovine rhinotracheitis" *Vet Res*
8. Jones, Chowdhury (2007) "A review of the biology of bovine herpesvirus type 1 (BHV-1), its role as a cofactor in the bovine respiratory disease complex and development of improved vaccines" *Anim Health Res Rev*
9. Menendez, Lupu (2007) "Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis" *Nat Rev Cancer*
10. Schroeder, Steen, Espinoza et al. (2021) "Fatty acid synthase (FASN) regulates the mitochondrial priming of cancer cells" *Cell Death Dis*
11. Ahmad, Moton, Kuttikrishnan et al. (2024) "Fatty acid synthase: a key driver of ovarian cancer metastasis and a promising therapeutic target" *Pathol Res Pract*
12. Wu, Qin, Fako et al. (2014) "Molecular mechanisms of fatty acid synthase (FASN)-mediated resistance to anti-cancer treatments" *Adv Biol Regul*
13. Kim, Lee, Kim et al. (2019) "Toll-like receptor mediated inflammation requires FASNdependent MYD88 palmitoylation" *Nat Chem Biol*
14. Xiong, Sun, Zhu et al. (2021) "Metformin alleviates inflammation through suppressing FASN-dependent palmitoylation of Akt" *Cell Death Dis*
15. Kastaniotis, Autio, Kerätär et al. (2016) "Mitochondrial fatty acid synthesis, fatty acids and mitochondrial physiology" *Biochim Biophys Acta Mol Cell Biol Lipids*
16. Schlaepfer, Joshi (2020) "CPT1A-mediated Fat Oxidation, Mecha nisms, and Therapeutic Potential" *Endocrinology*
17. Yang, Gu, Ni et al. (2024) "Carnitine palmitoyl-transferase 1A is potentially involved in bovine herpesvirus 1 productive infection" *Vet Microbiol*
18. Lin, Fu, Li et al. (2025) "The depletion of TFAM and p-β-catenin(S552) in mitochondria in response to BoAHV-1 productive infection leads to decreased mitochondrial biogenesis" *Vet Microbiol*
19. Cardoso, Rosa, Ferreira et al. (2016) "Bovine herpesviruses induce different cell death forms in neuronal and glial-derived tumor cell cultures" *J Neurovirol*
20. Fiorito, Nocera, Cantiello et al. (2020) "Bovine herpesvirus-1 infection in mouse neuroblastoma (Neuro-2A) cells" *Vet Microbiol*
21. Thunuguntla, El-Mayet, Jones (2017) "Bovine herpesvirus 1 can efficiently infect the human (SH-SY5Y) but not the mouse neuroblas toma cell line (Neuro-2A)" *Virus Res*
22. Liu, Yuan, Yang et al. (2023) "Associating bovine herpesvirus 1 envelope glycoprotein gD with activated phospho-PLC-γ1(S1248)" *Microbiol Spectr*
23. Chen, Wu, Shen (2024) "Fatty acid synthase inhibitor cerulenin hinders liver cancer stem cell properties through FASN/APP axis as novel therapeutic strategies" *J Lipid Res*
24. Lv, Zhang, Song et al. (2022) "Cerulenin suppresses ErbB2-overexpressing breast cancer by targeting ErbB2/ PKM2 pathway" *Med Oncol*
25. Zhu, Thunuguntla, Liu et al. (2017) "The β-catenin signaling pathway stimulates bovine herpesvirus 1 productive infection" *Virology (Auckl)*
26. Graul, Kisielnicka, Rychłowski et al. (2019) "Transmem brane regions of bovine herpesvirus 1-encoded UL49.5 and glycoprotein M regulate complex maturation and ER-Golgi trafficking" *J Gen Virol*
27. Sucharita, Tikoo, Van Drunen (1985) "Littel-van den Hurk S. 2022. Bovine herpesvirus-1 glycoprotein M mediates the translocation to the Golgi apparatus and packaging of VP8" *Viruses*
28. Yezid, Pannhorst, Wei et al. (2020) "Bovine herpesvirus 1 (BHV-1) envelope protein gE subcellular trafficking is contributed by two separate YXXL/Φ motifs within the cytoplasmic tail which together promote efficient virus cell-to-cell spread" *Virology (Auckl)*
29. Mollenhauer, Morré, Rowe (1990) "Alteration of intracellular traffic by monensin; mechanism, specificity and relationship to toxicity" *Biochim Biophys Acta*
30. Zhao, Fu, Gu et al. (2025) "53BP1, a known chromatinassociated factor that promotes DNA damage repair, is differentially modulated during bovine herpesvirus 1 infection in vitro and in vivo" *Vet Microbiol*
31. Liu, Luo, Que et al. (2025) "Negative regulation of SREBP-1/FAS signaling molecules activates the RIG-1/TBK1-mediated IFN-I pathway to inhibit BVDV replication" *Antiviral Res*
32. Liu, Liang, Liu et al. (2021) "Fatty acid synthase is involved in classical swine fever virus replication by interaction with NS4B" *J Virol*
33. Yang, Hood, Chadwick et al. (2008) "Fatty acid synthase is up-regulated during hepatitis C virus infection and regulates hepatitis C virus entry and production" *Hepatology*
34. Heaton, Perera, Berger et al. (2010) "Dengue virus nonstructural protein 3 redistributes fatty acid synthase to sites of viral replication and increases cellular fatty acid synthesis" *Proc Natl Acad Sci*
35. Saira, Zhou, Jones (2007) "The infected cell protein 0 encoded by bovine herpesvirus 1 (bICP0) induces degradation of interferon response factor 3 and, consequently, inhibits beta interferon promoter activity" *J Virol*
36. Tang, Lin, Liao et al. (2014) "Rab18 facilitates dengue virus infection by targeting fatty acid synthase to sites of viral replication" *J Virol*
37. Meng, Liu, Sun et al. (2019) "Hepatitis C virus nonstructural protein 5A perturbs lipid metabolism by modulating AMPK/SREBP-1c signaling" *Lipids Health Dis*
38. Du, Hultquist, Zhou et al. (2020) "mRNA display with library of even-distribution reveals cellular interactors of influenza virus NS1" *Nat Commun*
39. Hulse, Johnson, Boyle et al. (2021) "Epstein-barr virus-encoded latent membrane protein 1 and B-cell growth transforma tion induce lipogenesis through fatty acid synthase" *J Virol*
40. Lo, Lung, Dawson et al. (2018) "Activation of sterol regulatory element-binding protein 1 (SREBP1)-mediated lipogenesis by the Epstein-Barr virus-encoded latent membrane protein 1 (LMP1) promotes cell proliferation and progression of nasopharyngeal carcinoma" *J Pathol*
41. Kim, Arcos, Rothamel et al. (2020) "Discovery of widespread host protein interactions with the prereplicated genome of CHIKV using VIR-CLASP" *Mol Cell*
42. Zhu, Yu, Jiang et al. (2017) "First report of bovine herpesvirus 1 isolation from bull semen samples in China" *av*
43. Reed, Muench (1938) "A simple method of estimating fifty per cent endpoints12" *Am J Epidemiol*
44. Liu, Lin, Yang et al. "2023b. NFAT5 restricts bovine herpesvirus 1 productive infection in MDBK cell cultures" *Microbiol Spectr*
45. Thonur, Maley, Gilray et al. (2012) "One-step multiplex real time RT-PCR for the detection of bovine respiratory syncytial virus, bovine herpesvirus 1 and bovine parainfluenza virus 3" *BMC Vet Res*
46. Parafati, Russa, Lascala et al. (2024) "Dramatic suppression of lipogenesis and no increase in beta-oxidation are the main effects of bergamot flavonoids on gene expression in fatty liver disease" *Medicine and Pharmacology*
47. Zhu, Jones (2017) "The high mobility group AT-hook 1 protein stimulates bovine herpesvirus 1 productive infection" *Virus Res*
48. El-Mayet, Sawant, Wijesekera et al. (2020) "Progesterone increases the incidence of bovine herpesvirus 1 reactivation from latency and stimulates productive infection" *Virus Res*
49. Ghosh, Klein (2017) "Sex drives dimorphic immune responses to viral infections" *J Immunol*
50. Celestino, Checconi, Amatore et al. (2018) "Differential redox state contributes to sex disparities in the response to influenza virus infection in male and female mice" *Front Immunol*
51. Hall, Nachbagauer, Vermillion et al. (2017) "Progesterone-based contraceptives reduce adaptive immune responses and protection against sequential influenza A virus infections" *J Virol*
52. El-Mayet, Toomer, Ostler et al. (2022) "Progesterone sporadically induces reactivation from latency in female calves but proficiently stimulates bovine herpesvirus 1 productive infection" *J Virol*
53. Inman, Lovato, Doster et al. (2001) "A mutation in the latencyrelated gene of bovine herpesvirus 1 leads to impaired ocular shedding in acutely infected calves" *J Virol*
54. Schang, Jones (1997) "Analysis of bovine herpesvirus 1 transcripts during a primary infection of trigeminal ganglia of cattle" *J Virol*
55. Liqian, Workman, Jones (2017) "Potential role for a β-catenin coactivator (high-mobility group AT-hook 1 protein) during the latencyreactivation cycle of bovine herpesvirus 1" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12866596&blobtype=pdf | # Emerging diagnostic and stratification tools Neuroimaging and electrophysiological studies offer promising tools for both diagnosis and stratification. While contrast-enhanced 3T MRI is recommended
Tan Eng-King, Jia Dong, James Wang, Ling-Ling Chan
monitoring for recurrence frequency and treatment response.
## Current gaps in guidelines and the need for standardized clinical frameworks
The absence of guidelines specific to recurrent Bell's palsy hinders consistent care, as existing recommendations focus on first episodes and overlook recurrent nuances. Formal consensus guidelines are urgently needed, supported by prospective multicentre registries capturing standardized data-ranging from facial scores and imaging to immune markers and patient-reported outcomesto refine diagnosis and inform future trials.
Randomized controlled trials are essential to determine whether antiviral prophylaxis alters the course of recurrent Bell's palsy. Studies comparing long-term valacyclovir to placebo after a second episode-stratified by HSV DNA and immune profiles-should assess recurrence timing, House-Brackmann scores, and quality-of-life. Adaptive trial designs may clarify if combining corticosteroids with antivirals improves outcomes in severe or rapid-onset cases. Early adjunctive therapies like neuromuscular retraining also merit evaluation, given their efficacy in chronic facial paresis [4].
## Introduction
Bell's palsy is the commonest cause of acute VII-nerve weakness, with first-episode incidence typically 15 per 100,000 [1]. Although most attacks are self-limiting, 215% of patients experience further episodes [2]. This recurrence leads to delayed recovery and compounds functional and psychosocial disability. The clinical dilemma is stark: guidelines, evidence and even pathophysiological models have been constructed almost around first attacks, leaving clinicians to extrapolate when Bell's recurs [2].
## HSV-1 reactivation and the rationale for antiviral prophylaxis
A key area of controversy lies in the role of herpes simplex virus type 1 (HSV-1) in disease recurrence and use of prophylactic antivirals. Polymerase chain reaction studies have identified HSV-1 DNA in up to 79% of endoneurial fluid samples from patients with Bell's palsy, implicating viral reactivation as a pathogenic mechanism [3]. However, no randomized controlled trial has evaluated efficacy of prophylactic antiviral therapy in patients with recurrent disease. Small series and case reports support valacyclovir use in individuals with frequent or disabling relapses, particularly those with confirmed HSV-1 DNA positivity, immunocompromised status, or a strong family history [2]. In the absence of robust trial data, a risk-stratified approach may be reasonable, offering antiviral prophylaxis to select patients while prospectively to exclude mimics like demyelinating diseases, more advanced techniques-including high-resolution imaging and diffusion-based metrics-may enable detection of subtle nerve injury and inflammation [5]. Correlating these imaging features with clinical outcomes could guide decisions about decompression surgery or immunomodulatory therapies. Similarly, integration of viral serology and inflammatory cytokine profiling may facilitate identification of mechanistic biomarkers predictive of recurrence or treatment responsiveness.
Ultimately, recurrent Bell's palsy should be recognized as a distinct clinical entity rather than a reiteration of first-episode disease. The prevailing reliance on empirical management reflects a missed opportunity to develop precision approaches rooted in virology, immunology, and neuroimaging. Advancing the field will require coordinated international collaboration to establish trial infrastructure, and therapeutic consensus-offering patients a pathway from uncertainty to evidence-based care.
## References
1. Jeong, Yoon, Lim et al. (0281) "Risk factors for Bell's palsy based on the Korean National Health Insurance Service National Sample Cohort data"
2. Mancini, Bottaro, Capitani (2019) "Recurrent Bell's palsy: outcomes and correlation with clinical comorbidities" *Acta Otorhinolaryngol Ital*
3. (0100)
4. Murakami, Mizobuchi, Nakashiro et al. (1996) "Bell palsy and herpes simplex virus: identification of viral DNA in endoneurial fluid and muscle" *Ann Intern Med*
5. (0124)
6. Li, Li, Yan et al. (2015) "The effect of total facial nerve decompression in preventing further recurrence of idiopathic recurrent facial palsy" *Eur Arch Otorhinolaryngol*
7. Ho, Juliano, Eisenberg et al. (2015) "Anatomy and pathology of the facial nerve" *Am J Roentgenol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846339&blobtype=pdf | # A Naked Lyophilized mRNA Vaccine Against Seasonal Influenza, Administered by Jet Injection, Provides a Robust Response in Immunized Mice
Mingtao Zeng, Ganesh Yadaigiri, Sergei Sharabrin, Svetlana Krasnikova, Denis Kisakov, Mariya Borgoyakova, Vladimir Yakovlev, Elena Tigeeva, Ekaterina Starostina, Victoria Litvinova, Lyubov Kisakova, Danil Vahitov, Kristina Makarova, Ekaterina Volosnikova, Ksenia Ivanova, Alexander Bondar, Nadezhda Rudometova, Andrey Rudometov, Alexander Ilyichev, Larisa Karpenko
## Abstract
Background: Seasonal influenza remains a significant public health problem, and the constant antigenic drift of viruses requires regular vaccine updates. mRNA vaccines offer a promising platform for the development of new, effective influenza vaccines. Administration of the naked mRNA vaccine using a needle-free jet injection system further enhances its safety, reduces cost, and eliminates the need for lipid nanoparticles, which are traditionally used for mRNA delivery. Lyophilization of naked mRNA allows for long-term storage at +4 • C. Methods: We designed and produced an mRNA vaccine against seasonal influenza, designated mRNA-Vector-Flu, encoding the hemagglutinin (HA) of the A/Wisconsin/67/2022(H1N1)pdm09, A/Darwin/9/2021(H3N2), and B/Austria/1359417/2021 strains. The vaccine was lyophilized and stored for 1 month in a refrigerator (+4 • C). A comparative immunogenicity study was conducted between synthesized immediately before use prepared and lyophilized naked mRNA-Vector-Flu. The preparations were administered to BALB/c mice using a jet needleless injection twice, 3 weeks apart. Immunogenicity was assessed on day 35 of the study. Results: A comparative immunogenicity study of naked mRNA-Vector-Flu demonstrated that both the synthesized immediately before use prepared formulation and the lyophilized form, stored at +4 • C for a month, induced similar levels of virus-specific antibodies and generated a pronounced T-cell immune response. Conclusions: Delivery of the naked mRNA vaccine using a needle-free jet injection ensures a high-level immune response, which improves its safety, reduces its cost, and eliminates the need for lipid nanoparticles traditionally used for mRNA delivery. At the same time, lyophilization of the naked mRNA vaccine preserves its biological activity and ensures its storage for at least a month at +4 • C temperatures. Our results demonstrate that our proposed approach can be considered a promising direction for the development and improvement of the mRNA vaccine platform.
## 1. Introduction
Seasonal influenza remains a significant public health problem, with over 3 million influenza-related hospitalizations worldwide annually [1,2] and up to half a million deaths from this disease recorded annually [3]. Vaccination against seasonal influenza is an effective means of eliciting immunity, helping to reduce the significant burden of annual influenza epidemics [4,5].
Various types of influenza vaccines exist, including live attenuated, inactivated (wholevirion, split, subunit), and recombinant vaccines [6][7][8]. However, the constant antigenic drift of circulating influenza viruses renders seasonal influenza vaccines ineffective, necessitating annual reformulated of vaccines. In this case, mRNA-based vaccine platforms offer advantages over standard influenza vaccine technologies. The low cost of producing mRNA vaccines compared to conventional split subunit vaccines developed using RNA cells and the ability to quickly and easily replace the target gene in mRNA vaccines without changing the production technology itself allows for a timely response to the emergence of new virus strains [9]. mRNA vaccine technology has enabled the rapid creation of effective and safe drugs for the prevention of coronavirus. In total, approximately one billion people worldwide have been vaccinated with mRNA-based SARS-CoV-2 vaccines, and their effectiveness has been demonstrated [10,11]. The production of mRNA vaccines does not require the use of chicken embryos, which avoids mutations in the immunogen sequence during the production of vaccine batches. mRNA vaccines do not induce an unwanted immune response in the recipient, as is typical with vector vaccines, and therefore can be administered multiple times. Another important advantage is the efficient activation of both humoral and cellular virus-specific immunity, since the target protein is expressed endogenously from mRNA [12,13].
Currently, several teams are developing mRNA vaccines against seasonal influenza [14][15][16], and several vaccines, including Moderna's mRNA-1010, are undergoing clinical trials [17][18][19][20]. Lipid nanoparticles are typically used to deliver mRNA vaccines. Lipid nanoparticles consist of four lipid components that ensure effective encapsulation and release of mRNA, prolonged circulation in the body, and complex stability [21]. However, certain limitations have been identified, related to both the complexity of storage (down to -80 • C) and, consequently, the logistics of such vaccines, and adverse post-vaccination effects. These effects are believed to be largely related to the nature of the lipids coating the mRNA [22,23].
Therefore, alternative delivery methods have recently attracted increasing attention. Thus, the recently published [24][25][26][27] jet injection (JI) method is a promising alternative to LNPs for mRNA delivery. Jet injection is a physical delivery method in which vaccines and other therapeutic drugs are administered in fractions of a second using a high-speed jet through a nozzle under high pressure. This allows for efficient drug delivery intradermally, intramuscularly, or subcutaneously, without the need for a needle [28]. The most attractive feature of this method is that the mRNA vaccine is administered as a naked molecule, completely eliminating the negative effects of the lipid components encapsulating the mRNA molecule. However, in this case, the issue of storage and stability of naked mRNA molecule preparations becomes relevant, so we decided to use lyophilization.
The aim of this study was to compare the immunogenicity of the mRNA-Vector-Flu trivalent mRNA vaccine against seasonal influenza virus, synthesized immediately before use and its lyophilized form, administered via jet injection.
## 2. Materials and Methods
## 2.1. Bacterial and Viral Strains, Cell Cultures, and Plasmids
The E. coli strain NEB Stable (New England Biolabs Inc., Ipswich, MA, USA) was used for genetic engineering work. For in vitro analysis of mRNA functionality, the continuous human embryonic kidney cell line HEK293, provided by the Microorganism Collection Department of the State Research Center of Virology and Biotechnology Vector (Rospotrebnadzor, Koltsovo, Russia), was used.
To obtain DNA templates for mRNA synthesis, plasmids encoding the hemagglutinins of the corresponding influenza virus strains were used: pVAX-H1-24 encoding the hemagglutinin of the influenza virus A/Wisconsin/67/2022(H1N1)pdm09, pVAX-H3-24 encoding the hemagglutinin of the influenza virus A/Darwin/9/2021(H3N2) and pVAX-HB-24 encoding the hemagglutinin of the influenza virus B/Austria/1359417/2021 (FBRI SRC VB Vector, Rospotrebnadzor). To obtain mRNA-GFP, the previously obtained pVAX-C3-GFP matrix (FBRI SRC VB "Vector", Rospotrebnadzor) was used.
## 2.2. Templates for mRNA Synthesis
Hemagglutinins of influenza viruses recommended by WHO for the Northern Hemisphere for the 2023-2024 season [29] were used as antigens for the mRNA vaccine: A/Wisconsin/67/2022(H1N1)pdm09 (EPI_ISL_15928538), A/Darwin/9/2021(H3N2) (EPI_ISL_20142977), B/Austria/1359417/2021 (EPI_ISL_983345) without the transmembrane and cytoplasmic domains. To stabilize the trimeric structure of hemagglutinins, a T4 trimerizing domain was added to the C-terminus [30,31]. To maintain the uncleaved form of the rHA/H1 and rHA/H3 proteins, the following amino acid substitutions were introduced into the pH switch region of the HA2 subunit: H26 and H106, where histidine was replaced by tryptophan (W) and arginine (R), and amino acids K51 and E103 were replaced by isoleucine [32]. The codon composition of the sequence was optimized using the codon adaptation tool (https://www.jcat.de/, accessed 1 September 2025). To obtain DNA templates for mRNA synthesis, the previously developed pVAX-C3-PolyA DNA cassette [33], into which the hemagglutinin genes were cloned, was used. Cassette pVAX-C3-PolyA contains a T7 promoter that modified for efficient incorporation of the AG-Cap analogue during mRNA synthesis, 5 ′ -UTR of ChM and 3 ′ -UTR of human βglobin, and a 100-nucleotide poly(A) tail. The source of the hemagglutinin genes were the plasmids pVAX-H1-24, pVAX-H3-24 and pVAX-HB-24, encoding the corresponding hemagglutinin sequence. To obtain DNA templates for mRNA synthesis, routine cloning was performed: the pVAX-C3-PolyA cassette and the pVAX-H1\3\B plasmids were treated with endonucleases CciNI and BamHI (SibEnzyme, Novosibirsk, Russia). Then, the resulting DNA fragments were ligated using T4 DNA ligase (SibEnzyme, Novosibirsk, Russia), and the resulting mixture was used to transform E. coli cells. The resulting constructs were named pVAX-C3-H1-24, pVAX-C3-H3-24, and pVAX-C3-HB-24. The DNA sequences of the resulting plasmids were confirmed by restriction analysis and Sanger sequencing.
DNA templates were purified using the HiPure Plasmid Mini Kit (Catalog No.: P100103, Magen, Guangzhou, China) and linearized using the restriction endonuclease Bso31I (SibEnzyme, Novosibirsk, Russia).
## 2.3. In Vitro mRNA Synthesis
mRNA was synthesized using linearized pVAX-C3-H1-24, pVAX-C3-H3-24, pVAX-C3-HB-24, and pVAX-C3-GFP plasmid templates together with a commercial in vitro transcription kit (Yeasen, Shanghai, China). Each reaction mixture contained 1 µg of linearized DNA, T7 RNA polymerase with its corresponding buffer, the AG-Cap structure analog (m7GmAmG cap analog, Biolabmix, Novosibirsk, Russia; 10 mM), a ribonucleotide triphosphate mix (10 mM) in which uridine was substituted with N1-methylpseudouridine (Biolabmix, Novosibirsk, Russia), RNase inhibitor (BelBioLab, Moscow, Russia), and nuclease-free water. The synthesis procedure, previously described by our group in [33], involved enzymatic transcription, DNase treatment to remove template DNA, and subsequent purification steps. The resulting transcripts were designated as mRNA-H1, mRNA-H3, mRNA-HB, and mRNA-GFP.
## 2.4. PCR to Determine the Presence of Template DNA Residues in the mRNA Preparation
For PCR, primers complementary to the 5 ′ -and 3 ′ -UTR sequences of the mRNA encoded in the template DNA were used. The BioMaster HS-Taq PCR kit (2×) (Biolabmix, Novosibirsk, Russia), primers, and 1 µg of synthesized mRNA were used. Linearized template DNA was used as a control.
## 2.5. Capillary Electrophoresis
The purity, uniformity, and molecular size of the synthetic mRNAs were evaluated using an Agilent 2100 BioAnalyser (Agilent Technologies, Santa Clara, CA, USA). Analysis was performed by microcapillary electrophoresis with the Agilent RNA 6000 Pico kit (Agilent Technologies, Vilnius, Lithuania) following the manufacturer's instructions.
## 2.6. Encapsulation of mRNA in Lipid Nanoparticles
Encapsulation of mRNA into lipid nanoparticles and subsequent characterization were carried out following the procedure described by Yakovlev et al., 2025 [34]. Phase mixing was performed using an automated nanoparticle production system (Dolomite Microfluidics, Royston, UK) equipped with a hexagonal herringbone micromixer chip (Dolomite Microfluidics, Royston, UK). The aqueous phase contained mRNA dissolved in 100 mM citrate buffer (pH 4). The ethanol phase consisted of a lipid mixture composed of ionizable lipids, phospholipids, helper lipids, and PEG-lipids at molar ratios of 50:10:38.5:1.5, respectively; all lipid components were solubilized in 96% ethanol. The lipid formulation included the ionizable lipid SM-102 (heptadecan-9-yl 8-((2-hydroxyethyl)(6oxo-6-(undecyloxy)hexyl)amino)octanoate), the phospholipid DSPC (1,2-distearoyl-snglycero-3-phosphocholine), cholesterol as a helper lipid, and the PEG-lipid DMG-PEG2000 (1-monomethoxypolyethylene glycol-2,3-dimyristylglycerol with PEG of average molecular weight 2000).
The resulting lipid nanoparticles were analyzed using dynamic light scattering (DLS). DLS measurements confirmed that the mRNA-LNP formulations formed monodisperse nanoparticle suspensions, with PdI values of 0.158 ± 0.013 for mRNA-H1, 0.178 ± 0.006 for mRNA-H3, and 0.239 ± 0.034 for mRNA-HB. The mean hydrodynamic diameters of the LNPs were 94.42 ± 2.48 nm, 95.34 ± 0.38 nm, and 97.56 ± 2.44 nm for the respective formulations. The ζ-potentials measured across three series were 0.35 ± 0.12 mV, 0.54 ± 0.42 mV, and 0.05 ± 0.62 mV, values which were consistent with theoretical predictions. mRNA encapsulation efficiency, assessed using the Quant-iT RiboGreen assay (Life Technologies, Waltham, MA, USA), exceeded 90% for all formulations.
## 2.7. mRNA Lyophilization
Lyophilization of mRNA sample solutions was performed in an ED-DF21A-K freezedrying chamber (ERSTVAK, Moscow, Russia) in automatic mode with a pneumatic capping option. Three cryoprotectant variants were used for lyophilization: the first was based on sucrose, the second was based on mannose, and the third was based on trehalose. The sterile mRNA preparation with one of three cryoprotectants was dispensed into 0.5 mL vials of 3 mL and lyophilized for 24 h. The lyophilization process was carried out under standard conditions: freezing samples at -50 ± 2 • C for 8 h, and lyophilizing samples at 22 ± 2 • C for the remainder of the time. The lyophilized mass was a white tablet. The dried mRNA was stored at temperatures of -20 • C, +4 • C, and +20 • C.
## 2.8. Immunization of BALB/c Mice
Animal experiments were carried out in compliance with the Guide for the Care and Use of Laboratory Animals. All procedures were approved by the Laboratory Animal Care and Use Committee of the Federal Research Center of Virology and Biotechnology "Vector", Rospotrebnadzor (Bioethics Committee Protocol No. 3, 2024). Mice were housed under a 12 h light/dark cycle with unrestricted access to food and water.
The study utilized inbred BALB/c mice weighing 16-18 g at the beginning of the experiment. Preparations of mRNA-Vector-Flu-LNP in 100 µL of PBS were injected intramuscularly into the quadriceps of the left hind limb using insulin syringes equipped with a 29G needle. The mRNA-H1, mRNA-H3, mRNA-HB, and mRNA-Vector-Flu formulations, dissolved in 50 µL of saline, were administered into the quadriceps by the IM with Jet-injection as described earlier [27,33]. Jet-injection immunization was performed using a Comfort-IN needle-free jet injector (MIKA MEDICAL CO, Busan, Republic of Korea) with the following parameters: injection velocity of 220 m/s, pressure of 6.5 bar, and injection duration of 0.33 s. Disposable nozzles were used for each application. The commercial seasonal influenza vaccine Flu-M (FSUE SPbNIIVS FMBA of Russia, 2025) was administered intramuscularly into the quadriceps using a syringe.
The experiment consisted of three phases. Phase 1: A comparative evaluation of the immunogenicity of the developed mRNA vaccine. Mice were divided into seven groups (n = 6 per group) and vaccinated on days 0 and 21. Group 1 received 30 µg mRNA-H1; Group 2-30 µg mRNA-H3; Group 3-30 µg mRNA-HB; Group 4-the trivalent mRNA-Vector-Flu vaccine (30 µg of each component, 90 µg total); Group 5-the trivalent mRNA-Vector-Flu-LNP vaccine (10 µg per component, 30 µg total); Group 6-250 µL of the Flu-M commercial vaccine; Group 7 was the unimmunized control. Jet injection was used for groups 1-4, whereas groups 5 and 6 received injections by syringe. The 30 µg mRNA dose was selected based on findings from earlier work [27,33] and studies on jet-injection performed by other researchers [26]. On day 35, blood was collected from the retro-orbital sinus, and animals were euthanized by cervical dislocation.
Phase 2: Evaluation of dose-dependent effects of the mRNA vaccine. Mice were divided into three groups (n = 6 each) and immunized on days 0 and 21. Group 1 received the trivalent mRNA-Vector-Flu formulation at 10 µg per mRNA (30 µg total); Group 2 received 30 µg per mRNA (90 µg total); Group 3 received 50 µg per mRNA (150 µg total). Immunizations were performed using jet injection. Blood was collected on day 35 from the retro-orbital sinus for humoral immunity assessment.
Phase 3: Assessment of the effectiveness of lyophilized mRNA vaccine formulations. Two groups of six mice each were immunized on days 0 and 21. Group 1 received the lyophilized trivalent mRNA-Vector-Flu vaccine containing 15 µg of each component (45 µg total). Group 2 received the non-lyophilized mRNA-Vector-Flu formulation with the same component doses. Animals were vaccinated using jet injection. On day 35, retroorbital blood samples were obtained for humoral response evaluation, including HAI and microneutralization assays. For assessment of T-cell responses, mice were euthanized by cervical dislocation and their spleens were collected.
## 2.9. Enzyme-Linked Immunosorbent Assay (ELISA)
The enzyme-linked immunosorbent assay (ELISA) was carried out as previously reported [35]. Recombinant eukaryotic hemagglutinin proteins H1 (A/Wisconsin/67/2022 (H1N1)pdm09), H3 (A/Darwin/9/2021(H3N2)), and HB (B/Austria/1359417/2021), produced earlier at the Bioengineering Department of the State Research Center of Virology and Biotechnology Vector, Rospotrebnadzor, served as antigenic targets. Rabbit anti-mouse https://doi.org/10.3390/vaccines14010056
IgG antibodies conjugated to horseradish peroxidase (Sigma-Aldrich, St. Louis, MO, USA) were employed as secondary antibodies. TMB (Amresco LLC, Solon, Ohio, USA) was used as the chromogenic substrate. Following incubation, the enzymatic reaction was terminated with a stop reagent (1 M hydrochloric acid), and absorbance was recorded at 450 nm using a Varioskan Lux microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).
## 2.10. Evaluation of the Cellular Immune Response by ELISpot
Splenocytes were obtained by gently dissociating individual spleens through nylon strainers with pore sizes of 70 and 40 µm (BD Falcon™, Franklin Lakes, NJ, USA). Erythrocytes were eliminated using a red blood cell lysis buffer (Sigma, Virginia Beach, VA, USA). The magnitude of the T-cell response in immunized mice was assessed by quantifying the number of splenocytes secreting IFN-γ using an IFN-γ ELISpot assay. The procedure was carried out with IFN-γ ELISpot kits (MABTECH, Nacka Strand, Sweden) following the manufacturer's protocol. For cell stimulation, a pool of synthetic peptides (10-20 aa in length) representing conserved T-cell epitopes of the hemagglutinin proteins from influenza viruses A/Wisconsin/67/2022 (H1N1)pdm09, A/Darwin/9/2021 (H3N2), and B/Austria/1359417/2021 was used; peptides were produced by AtaGenix Laboratories (Wuhan, China). Each peptide was applied at a concentration of 20 µg/mL. The number of IFN-γ-producing cells was determined using an ELISpot reader (Carl Zeiss, Oberkochen, Germany).
## 2.11. In Vitro Microneutralization Assay
The in vitro neutralization assay was performed with influenza viruses A/Buryatia/106-6V/2022 (H1N1), A/Darwin/9/2021 (H3N2), and B/Austria/1359417/2021 as described in study [36], with modifications. The difference lay in the method used to visualize the final result: two days after infection, the cells were stained with a crystal violet solution (1.3 g of dye dissolved in 50 mL of 96% ethanol, brought to 700 mL with distilled water, and supplemented with 300 mL of 40% formalin). The results were analyzed using an Agilent BioTek Cytation 5 multimode cell imaging reader (Thermo Fisher Scientific). The neutralization titer was defined as the highest serum dilution at which ≥50% virus neutralization was achieved, corresponding to ≥50% viable cells.
## 2.12. Hemagglutinin Inhibition (HAI) Assay
HAI assay was performed according to the HAI protocol based on World Health Organization (WHO) guidelines. Prior to the HI test, the collected animal serum was treated with RDE (Denka Seiken, Tokyo, Japan) for 18-20 h according to the manufacturer's instructions to remove nonspecific thermostable inhibitors, then heated in a water bath at 56 • C to eliminate nonspecific thermolabile inhibitors and to inactivate the enzymatic activity of RDE. The final serum dilution of 1:10 was achieved by adding phosphate-buffered saline. The sera were tested in the HAI assay against 4 hemagglutinating units of antigen corresponding to the virus serotype. To determine the HAI titer, 0.5% turkey red blood cells and U-bottom plates were used. When calculating geometric mean titers, HAI values <1/10 were considered equal to 5. Inactivated β-propiolactone-treated influenza viruses A/Buryatia/106-6V/2022 (H1N1), A/Darwin/9/2021 (H3N2), and B/Austria/1359417/2021 were used as antigens.
## 2.13. Transfection of HEK293 Cells with mRNA
HEK293 cells were cultured in 24-well plates (Corning, New York, NY, USA) in DMEM medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% FBS (HyClone, Logan, UT, USA) and 50 mg/mL gentamicin. When the cell monolayer reached approximately 70-80% confluence, it was transfected with mRNA encoding GFP using a PEI-based nucleic acid transfection kit for eukaryotic cells (BIOSPECIFICA, Novosibirsk, Russia). The https://doi.org/10.3390/vaccines14010056 transfection reagent was combined with 1 µg of mRNA, incubated for 15 min at room temperature, and subsequently applied to the cells. The plates were then placed in a CO 2 incubator and incubated for 24 h.
## 2.14. Statistical Analysis
Statistical data processing was carried out using the GraphPad Prism 10.0 software package (GraphPad Software, San Diego, CA, USA). Quantitative variables were presented either as mean values with an interval or as a range from minimum to maximum. Nonparametric statistical methods were applied for their analysis. Differences between the study groups were evaluated using the nonparametric one-way Kruskal-Wallis analysis of variance with subsequent adjustment for multiple comparisons and Dunn's hypothesis testing.
## 3. Results
## 3.1. Production of the Experimental Trivalent mRNA Influenza Vaccine mRNA-Vector-Flu
Previously, we developed a DNA template cassette for mRNA vaccine synthesis, called pVAX-C3-PolyA [33]. The modified chimeric β-globin sequence used in Moderna's studies and the human β-globin sequence were included in the cassette as the 5 ′ -UTR and 3 ′ -UTR, respectively. These elements are necessary for increasing mRNA stability and translation efficiency. The initiator nucleotides GG in the T7 promoter were replaced with AG. This modification allows the use of the AG-Cap analog during in vitro transcription to create a "cap" at the 5 ′ -end of mRNA, which is critical for its stability and translation. The plasmid contains a 100-nucleotide poly(A) tail containing an internal linker of 10 random nucleotides (e.g., 30(A)GCATATGACT70(A)). The poly(A) tail is important for mRNA stability and translation initiation. The poly(A) tail sequence ends with a Bso31I restriction endonuclease site. This element ensures that during DNA hydrolysis, the transcription template ends with an adenine, ensuring the precise termination of the synthesized mRNA.
The pVAX-H1-24 plasmid, carrying the hemagglutinin gene of the influenza A/Wisconsin/67/2022 (H1N1)pdm09 virus, was used as the source of the target gene for the trivalent mRNA components of the influenza vaccine. pVAX-H3-24, carrying the hemagglutinin gene of the influenza A/Darvin/9/2021 (H3N2) virus, and pVAX-HB-24, carrying the hemagglutinin gene of the influenza B/Austria/1359417/2021 virus (B/Victoria lineage). In all hemagglutinins, the transmembrane and cytoplasmic domains were removed, and the trimerizing domain of phage T4 was added to enhance immunogenicity (Figure 1a).
After inserting the hemagglutinin gene into the pVAX-C3-PolyA cassette, the pVAX-C3-H1-24, pVAX-C3-H3-24, and pVAX-C3-HB-24 constructs were obtained. These constructs were then used to synthesize the corresponding mRNAs.
## 3.2. mRNA Synthesis and Characterization
mRNA-C3-H1-24, mRNA-C3-H3-24, and mRNA-C3-HB-24 were synthesized using the method described in Section 2.3. The integrity and purity of the mRNA were verified by electrophoresis in a 2% agarose gel (Figure 1b) and capillary electrophoresis (Figure S1). The mRNA transcript size was expected to be approximately 1900 base pairs, consistent with the DNA template. The absence of a high-molecular-weight signal indicates complete removal of the DNA template from the preparation. PCR was also performed for confirmation (Figure 1c).
## 3.3. Immunogenicity Assessment of the mRNA-Vector-Flu Vaccine
To assess the immunogenicity of the seasonal influenza mRNA vaccine, BALB\c mice (n = 6) were immunized with each mRNA individually and in a mixture. A needle-free jet injection method was used for immunization, as described previously [33,34]. The mRNA dose was 30 µg of each immunogen per animal. As a control, one group of animals was immunized with the commercial Flu-M vaccine (SPbSRIVS FMBA of Russia) of the 2024-2025 formulation. Lipid nanoparticles, the gold standard of mRNA delivery, were also used as a control. Animals were immunized with a mixture of 10 µg of each mRNA encapsulated in LNPs, as described previously [25,27,34].
Animal sera were collected on day 35 of the experiment. To set up the ELISA, individual recombinant proteins of hemagglutinin H1, H3 and HB were used as the antigen, as described in Section 2.9. The results showed that all nonvariant's of the trivalent mRNA vaccine against influenza mRNA-Vector-Flu generate an immune response against the corresponding antigen (Figure 2). The average titer for mRNA-H1 was 1:400,950, and for mRNA-H3 1:510,300, mRNA-HB 1:225,000. The commercial vaccine Flu-M showed less pronounced antibody titers, and in the case of hemagglutinin H3 the titer was practically https://doi.org/10.3390/vaccines14010056 equal to 0. The antibody titer in the group of animals immunized with the trivalent mRNA vaccine in LNP was slightly higher than in the group of animals immunized with jet injection, but the differences were not significant (Figure 2).
## 3.4. Comparison of the Dose-Dependent Effect of the Trivalent mRNA Vaccine
For a dose-dependent comparison, BALB/c mice were immunized twice with mRNA-Vector-Flu using JI, at doses of 10, 30, and 50 µg of each immunogen. Results showed that at a 10 µg dose of trivalent mRNA, the average titer ranged from 1:100,000 to 1:200,000, depending on the antigen in the ELISA (Figure 3). At a dose of 30 µg of each immunogen, the average titer was approximately 1:500,000. At a dose of 50 µg of each immunogen, we did not observe an increase in the humoral immune response compared to the 30 µg dose. This may be due to the high immunogen dose, which resulted in immunosuppression.
## 3.5. Lyophilization of mRNA-GFP
We used three cryoprotectants to stabilize and store naked mRNA after freeze-drying. We initially assessed mRNA integrity after lyophilization using the mRNA-GFP model. mRNA synthesis was performed as described in Section 2.3. Drying was performed as described in Section 2.7. The first cryoprotectant was sucrose-based, the second was mannose-based, and the third was trehalose-based. Samples were stored at 4 • C for one month. The lyophilized mRNA was then dissolved in nuclease-free water and used to transfect HEK293 cells. The results showed that after one month of storage, all preparations demonstrated high efficiency in GFP protein synthesis (Figure 4).
Based on the obtained results, we selected trehalose-based cryoprotectant No. 3 for further work. New mRNA-GFP preparations were prepared, lyophilized, and stored at -20 • C, +4 • C, and +20 • C for long periods. After 1 month and 3 months, lyophilized mRNA-GFP samples were dissolved in pure water and used to transfect HEK293 cells. The results showed that after both 1 month and 3 months of storage, all mRNA samples produced levels of GFP protein synthesis comparable to synthesized immediately before use prepared control mRNA; the differences were not statistically significant (Figure 4b,c). We also analyzed lyophilized mRNA using capillary electrophoresis. The results show that mRNA integrity after drying remains virtually unchanged compared to the control, synthesized immediately before use mRNA. All samples exhibited minor mRNA degradation, less than 5%, which is acceptable (Figure S2).
## 3.6. Immunogenicity Assessment of the Lyophilized Trivalent mRNA-Vaccine mRNA-Vector-Flu
The next stage of the study was to evaluate the immunogenic properties of the lyophilized trivalent seasonal influenza vaccine mRNA-Vector-Flu. The mRNA was lyophilized with trehalose-based cryoprotectant No. 3 and stored at +4 • C for one month.
To assess the immunogenicity of the mRNA-Vector-Flu vaccine, BALB\c mice (n = 6) were immunized twice with a three-week interval (Figure 5a). As a control, mice were immunized with synthesized immediately before use mRNA vaccine. Immunization was performed using the jet injection method. https://doi.org/10.3390/vaccines14010056 Data are presented as the median with a range of inverse titers. Significance was assessed using non-parametric one-factor Kruskal-Wallis analysis of variance (ns-not statistically significant). mRNA-Vector-Flu Liof-mice immunized with lyophilized mRNA; mRNA-Vector-Flu-mice immunized with synthesized immediately before use mRNA.
Animal sera were collected on day 35 of the experiment. For ELISA, individual recombinant hemagglutinin proteins H1, H3, and HB, as well as their combination, were used as antigens. The results showed that both the synthesized immediately before use prepared and lyophilized mRNA seasonal influenza vaccine mRNA-Vector-Flu produced similar results.
The average titer for the recombinant H1 protein was 1:430,000 for the lyophilized vaccine and 1:870,000 for the synthesized immediately before use mRNA, 1:760,000 for H3, and 1:300,000 for HB. When all three recombinant hemagglutinin proteins were adsorbed in the ELISA, the average titer was 1:1,300,000 for both vaccines (Figure 5b).
To evaluate the ability of post-immunization sera to inhibit hemagglutination, a HAI assay was performed using three influenza viruses (Figure 5c). For A/Buryatia/106-6V/2022 (H1N1), both vaccine formulations produced similar hemagglutination inhibition titers, with an average titer of 1:40. Against A/Darwin/9/2021 (H3N2), the average titer was 1:40 for the lyophilized vaccine and 1:60 for synthesized immediately before use mRNA. For the B/Austria/1359417/2021 antigen, both vaccine variants elicited a higher response, with an average titer of 1:80.
An important criterion of vaccine effectiveness is its ability to induce antibodies capable of neutralizing the virus. Sera from mice immunized with both the lyophilized and non-lyophilized variant of the mRNA-Vector-FLU vaccine demonstrated the ability to neutralize influenza viruses A/Buryatia/106-6V/2022 (H1N1), A/Darwin/9/2021 (H3N2), and B/Austria/1359417/2021 in MDCK cell culture in vitro in the virus neutralization assay (Figure 5d). The average titer was approximately 1:100, except for the lyophilized vaccine with the A/Buryatia/106-6V/2022 (H1N1) virus, where the neutralization titer was 1:50.
For complete protection against viral infections, the development of a T-cell immune response is also necessary. This was assessed using IFN-γ-ELISpot. The results showed that, two weeks after the second immunization, T-cell immunity developed in mice immunized with both lyophilized and synthesized immediately before use mRNA vaccines in response to stimulation with a pool of specific peptides (Figure 5e). The average number of IFNγsecreting T lymphocytes was 458 per 1 million cells in the group of animals immunized with lyophilized mRNA and 515 per 1 million cells in the group with synthesized immediately before use mRNA.
## 4. Discussion
mRNA vaccines offer a promising platform for developing new, effective influenza vaccines. Delivery of a naked mRNA vaccine using a needle-free jet injection system further enhances safety, reduces cost, and eliminates the need for lipid nanoparticles, which are traditionally used for mRNA delivery.
In the first phase of our work, we designed and produced a trivalent mRNA influenza vaccine, named mRNA-Vector-Flu, encoding the hemagglutinin (HA) of the seasonal influenza virus strains recommended by the WHO for the 2023-2024 season: A/Wisconsin/67/2022(H1N1)pdm09, A/Darwin/9/2021(H3N2), and B/Austria/1359417/2021. https://doi.org/10.3390/vaccines14010056
A study of the immunogenicity of the resulting constructs administered to mice using jet injection demonstrated that both the individual and trivalent mRNA-Vector-Flu mRNA influenza vaccines elicited a pronounced specific immune response in BALB/c mice (Figure 2). When immunized with the individual (monovalent) vaccines, each mRNA vaccine induced high titers of specific antibodies (mRNA-H1-1:400950, mRNA-H3-1:510300, mRNA-HB-1:225000), as demonstrated by ELISA using recombinant hemagglutinins corresponding to the specific vaccine as antigens. This confirms their ability to induce a strong and specific humoral response.
The control group immunized with the commercial Flu-M vaccine demonstrated significantly lower antibody titers, particularly to hemagglutinin H3, where the titer was virtually zero. This is due to the difference in the composition of the seasonal influenza strains in the vaccines, as the mRNA-Vector-Flu vaccine developed in this study has a composition for the 2023-2024 season, while the Flu-M vaccine used for the control group has a composition for the 2025-2026 season. Only the H3 component changed in the vaccine compositions for these years: in the 2023 season it was A/Darvin/9/2021 (H3N2), and in the 2025 season it was A/District of Columbia/27/2023 (H3N2).
In this study, we used a 10 µg dose of each immunogen for the mRNA-Vector-Flu-LNP vaccine, while we used a 30 µg dose of each immunogen for mRNA-Vector-Flu delivered using JI. We previously demonstrated that animals immunized with a 30 µg dose of mRNA-H5-LNP exhibited signs of stress, including fur ruffles, as well as increased discomfort during the second immunization session [34]. Similar results were reported by other authors studying different doses of the COVID-19 mRNA vaccine [37].
According to our results, trivalent mRNA encapsulated in LNPs induced slightly higher antibody levels compared to jet injection; however, the differences were not statistically significant. This suggests that jet injection is a promising mRNA delivery method, comparable in efficacy to traditional LNPs.
Thus, the obtained results show that the developed mRNA vaccine mRNA-Vector-Flu elicits a strong specific immune response, and the components of the trivalent vaccine are able to maintain their immunogenicity in combination without suppressing each other's immune responses.
Analysis of the dose-dependent immune response of the vaccine revealed varying effects of different doses on immune response formation (Figure 3). According to the obtained results, at a vaccine dose of 150 µg (50 µg of each immunogen), antibody titers were approximately equal to those observed at a 30 µg dose (10 µg of each immunogen). It is possible that high doses of mRNA molecules may lead to immunosuppression. A dose of 90 µg (30 µg of each immunogen) showed good results. Thus, the optimal dose for an mRNA vaccine against seasonal influenza is 10-30 µg of each immunogen.
It is known that in aqueous solutions, naked mRNA preparations are relatively unstable; one possible approach to stabilize the formulation is lyophilization. Therefore, the next stage of the study was to determine the feasibility of lyophilization and storage of naked mRNA.
We conducted preliminary studies on the effect of lyophilization on the biological properties of the formulation using an mRNA-GFP model. mRNA-GFP samples lyophilized with three different cryoprotectants, selected based on literature data [38,39], demonstrated comparable efficiency of GFP protein synthesis in transfected cells after one month of storage at +4 • C (Figure 4a). Based on these results, a trehalose-based cryoprotectant was selected for further work. mRNA-GFP lyophilized using trehalose and stored at -20 • C, +4 • C, and +20 • C for three months maintained GFP protein synthesis at a level comparable to synthesized immediately before use control mRNA (Figure 4b,c). The stability of lyophilized mRNA at positive temperatures may, in the future, help address the challenges of storage and transportation of mRNA vaccines.
The third stage of the study focused on evaluating the biological properties of the trivalent mRNA vaccine against seasonal influenza, mRNA-Vector-Flu, lyophilized with trehalose and stored at +4 • C for one month. A synthesized immediately before use mRNA vaccine was used as a control. Immunization with the trivalent mRNA-Vector-Flu, containing 15 µg of each component in 50 µL of saline, was performed via jet injection, and animal sera were collected on day 35 of the experiment.
To quantify the humoral response, an ELISA method was used, using both individual recombinant hemagglutinin proteins H1, H3, and HB, as well as a combination of them, as antigens. The data obtained demonstrated that the lyophilized mRNA vaccine retains a high level of immunogenicity comparable to that of the synthesized immediately before use mRNA vaccine (Figure 5b). The average antibody titer against the H1 protein was 1:430,000 for the lyophilized vaccine and 1:870,000 for synthesized immediately before use mRNA. For H3, it reached 1:760,000, and for HB, 1:300,000. When all three proteins were simultaneously adsorbed in an ELISA, the average titer for both vaccines was 1:1,300,000, indicating the formation of a strong specific humoral response against different strains of the seasonal influenza virus. The HAI and microneutralization assay results demonstrated that both the lyophilized and synthesized immediately before use mRNA-Vector-Flu vaccines were capable of neutralizing all three seasonal influenza viruses (Figure 5c,d). In addition to humoral immunity, the eliciting of a T-cell response is important for comprehensive protection against viral infection. This was assessed using an IFN-γ ELISpot assay, using a pool of specific peptides to stimulate cells. The results showed that two weeks after the second immunization, animals receiving both lyophilized and synthesized immediately before use mRNA vaccines developed a robust T-cell immune response. The average number of IFN-γ-secreting T cells was 458 per million cells in the lyophilized mRNA group and 515 per million cells in group of animals immunized with synthesized immediately before use the mRNA (Figure 5e).
Thus, the lyophilized mRNA-Vector-Flu mRNA vaccine, stored at 4 • C for a month, demonstrates the ability to induce both a strong humoral and cellular immune response comparable to the effect of the synthesized immediately before use prepared mRNA vaccine. These data confirm the potential of the lyophilized vaccine for further use and extended shelf life without loss of immunogenic activity.
The next step in our research is to test the biological activity of lyophilized naked mRNA-Vector-Flu stored at 4 • C and 20 • C for a year.
Combining lyophilization of the naked mRNA vaccine with our developed method of delivering mRNA via jet injection solves the main problem of prophylactic mRNA vaccine platforms: delivery. Using the gold standard of delivery, lipid nanoparticles (LNPs), significantly increases the cost of vaccine production; LNPs can cause undesirable side effects; mRNA-LNPs must be stored at extremely low temperatures or a complex cryoprotectant composition must be used for their lyophilization [40].
## 5. Conclusions
In this study, we produced and characterized a trivalent mRNA vaccine against the seasonal influenza virus, mRNA-Vector-Flu. Delivery of the naked vaccine via jet injection was shown to elicit a high immune response in BALB/c mice. Lyophilization with a trehalosebased cryoprotectant was used for stabilization and storage. Using the mRNA-GFP model, we demonstrated that lyophilized mRNA remained stable during storage for at least 3 months at +4 • C and +20 • C. Notably, the biological activity of lyophilized mRNA-GFP was comparable to that of synthesized immediately before use prepared mRNA. A comparative study of the immunogenicity of naked mRNA-Vector-Flu showed that both the synthesized immediately before use the mRNA preparation and the lyophilized form stored at +4 • C for a month induced similar levels of virus-specific antibodies and generated a pronounced T-cell immune response. https://doi.org/10.3390/vaccines14010056 Thus, delivery of a naked mRNA vaccine using a needle-free jet injection ensures a high-level immune response, which improves its safety, reduces its cost, and eliminates the need for lipid nanoparticles traditionally used for mRNA delivery. At the same time, lyophilization of the naked mRNA vaccine preserves its biological activity and ensures its storage for at least a month at +4 • C temperatures. Our results demonstrate that our proposed approach can be considered a promising direction for the development and improvement of the mRNA vaccine platform.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vaccines14010056/s1, Figure S1
## References
1. Who, Influenza (2025)
2. Paget, Staadegaard, Wang et al. (2023) "Global and national influenza-associated hospitalisation rates: Estimates for 40 countries and administrative regions" *J. Glob. Health*
3. Iuliano, Roguski, Chang et al. (2018) "Estimates of global seasonal influenza-associated respiratory mortality: A modelling study" *Lancet*
4. Martins, Santos, Martins et al. (1322) "Seasonal Influenza Vaccine Effectiveness in Persons Aged 15-64 Years: A Systematic Review and Meta-Analysis. Vaccines"
5. (2022) "Vaccines against influenza: WHO position paper" *Wkly. Epidemiol. Rec*
6. Chan, Alizadeh, Alizadeh et al. "Review of Influenza Virus Vaccines: The Qualitative Nature of Immune Responses to Infection and Vaccination Is a Critical Consideration" *Vaccines*
7. Mokalla, Gundarapu, Kaushik et al. "Influenza Vaccines: Current Status, Adjuvant Strategies, and Efficacy. Vaccines 2025"
8. Moore, Ostrowsky, Kraigsley et al. (2021) "A research and development (R&D) roadmap for influenza vaccines: Looking toward the future" *Vaccine*
9. Scorza, Pardi (2018) "New Kids on the Block: RNA-Based Influenza Virus Vaccines" *Vaccines*
10. Walsh, Frenck, Jr et al. (2020) "Safety and Immunogenicity of Two RNA-Based Covid-9 Vaccine Candidates" *N. Engl. J. Med*
11. Teo (2022) "Review of COVID-19 mRNA Vaccines: BNT162b2 and mRNA-1273" *J Pharm Pract*
12. Polack, Thomas, Kitchin et al. (2020) "Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine" *N. Engl. J. Med*
13. Baden, El Sahly, Essink et al. (2021) "Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine" *N. Engl. J. Med*
14. Mazunina, Gushchin, Kleymenov et al. (2024) "Trivalent mRNA vaccine-candidate against seasonal flu with cross-specific humoral immune response" *Front. Immunol*
15. Joe, Christopoulou, Van Hoecke et al. (2019) "Intranodal administration of mRNA encoding nucleoprotein provides cross-strain immunity against influenza in mice" *J. Transl. Med*
16. Zhuang, Qi, Wang et al. "mRNA Vaccines Encoding the HA Protein of Influenza A H1N1 Virus Delivered by Cationic Lipid Nanoparticles Induce Protective Immune Responses in Mice"
17. Lee, Nachbagauer, Ensz et al. (2023) "Safety and immunogenicity of a phase 1/2 randomized clinical trial of a quadrivalent, mRNA-based seasonal influenza vaccine (mRNA-1010) in healthy adults: Interim analysis" *Nat. Commun*
18. Spergel, Lee, Koslovsky et al. (2025) "Immunogenicity and safety of mRNA-based seasonal influenza vaccines encoding hemagglutinin and neuraminidase" *Nat. Commun*
19. Kandinov, Soens, Huang et al. (2025) "An mRNA-based seasonal influenza vaccine in adults: Results of two phase 3 randomized clinical trials and correlate of protection analysis of hemagglutination inhibition titers" *Hum. Vaccin. Immunother*
20. Soens, Ananworanich, Hicks et al. (2025) "A phase 3 randomized safety and immunogenicity trial of mRNA-1010 seasonal influenza vaccine in adults" *Vaccine*
21. Cheng, Wang, Bai et al. (2023) "Research Advances on the Stability of mRNA Vaccines" *Viruses*
22. Faksova, Walsh, Jiang et al. "COVID-19 Vaccines and Adverse Events of Special Interest: A Multinational Global Vaccine Data Network (GVDN) Cohort Study of 99 Million Vaccinated Individuals" *Vaccine*
23. Oster, Shay, Su et al. (2020) "Myocarditis Cases Reported After mRNA-Based COVID-19 Vaccination in the US From December" *JAMA*
24. Mao, Li, Zhang et al. (2023) "A Highly Efficient Needle-Free-Injection Delivery System for MRNA-LNP Vaccination against SARS-CoV-2" *Nano Today*
25. Kisakov, Kisakova, Sharabrin et al. (2023) "Delivery of Experimental MRNA Vaccine, Encoding the RBD of SARS-CoV-2 Using Jet Injection" *Bull. Exp. Biol. Med*
26. Abbasi, Matsui-Masai, Yasui et al. (2024) "Carrier-Free MRNA Vaccine Induces Robust Immunity against SARS-CoV-2 in Mice and Non-Human Primates without Systemic Reactogenicity" *Mol. Ther*
27. Kisakov, Karpenko, Kisakova et al. (2025) "Jet Injection of Naked mRNA Encoding the RBD of the SARS-CoV-2 Spike Protein Induces a High Level of a Specific Immune Response in Mice"
28. Wang, Bian, Xu et al. (2021) "Recent Advances in Mechanical Force-Assisted Transdermal Delivery of Macromolecular Drugs" *Int. J. Pharm*
29. (2023) "Recommendations Announced for Influenza Vaccine Composition for the 2023-2024 Northern Hemisphere Influenza Season"
30. Lu, Welsh, Swartz (2014) "Production and stabilization of the trimeric influenza hemagglutinin stem domain for potentially broadly protective influenza vaccines"
31. Ecker, Kirchenbaum, Pierce et al. "High-Yield Expression and Purification of Recombinant Influenza Virus Proteins from Stably-Transfected Mammalian Cell Lines"
32. Milder, Jongeneelen, Ritschel et al. (2022) "Universal stabilization of the influenza hemagglutinin by structure-based redesign of the pH switch regions" *Proc. Natl. Acad. Sci*
33. Sharabrin, Ilyichev, Kisakov et al. (2025) "Needle-Free Jet-Delivered mRNA-Vaccine Encoding Influenza A(H1N1)pdm09 Hemagglutinin Protects Mice from Lethal Virus Infection" *Mol. Biol*
34. Yakovlev, Litvinova, Rudometova et al. (2025) "Immunogenic and Protective Properties of mRNA Vaccine Encoding Hemagglutinin of Avian Influenza A/H5N8 Virus, Delivered by Lipid Nanoparticles and Needle-Free Jet Injection"
35. Karpenko, Rudometov, Sharabrin et al. "Delivery of mRNA Vaccine against SARS-CoV-2 Using a Polyglucin: Spermidine Conjugate. Vaccines 2021"
36. Gross, Bai, Jefferson et al. (2017) "Measuring Influenza Neutralizing Antibody Responses to A(H3N2) Viruses in Human Sera by Microneutralization Assays Using MDCK-SIAT1 Cells" *J. Vis. Exp*
37. Mulligan, Lyke, Kitchin et al. (2020) "Phase I/II Study of COVID-19 RNA Vaccine BNT162b1 in Adults" *Nature*
38. Li, Jia, Xie et al. "Lyophilization process optimization and molecular dynamics simulation of mRNA-LNPs for SARS-CoV-2 vaccine"
39. Muramatsu, Lam, Bajusz et al. (2022) "Lyophilization provides long-term stability for a lipid nanoparticle-formulated, nucleoside-modified mRNA vaccine" *Mol. Ther*
40. Khan, Baudin, Sudalaiyadum Perumal et al. (2025) "Freeze-Drying of mRNA-LNPs Vaccines: A Review"
41. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793444&blobtype=pdf | Adam Gailani, Tess Stopczynski, ; Herdi, Olla Hamdan, B Pharm, Haya Hayek, Varvara Probst, Laura Stewart, ; Rangaraj Selvarangan, Jennifer Schuster, Mary Moffatt, Marian Michaels, John Williams, Leila Sahni, Vasanthi Avadhanula, Julie Boom, Mary Staat, Elizabeth Schlaudecker, Christina Quigley, Geoffrey Weinberg, Peter Szilagyi, Janet Englund, Eileen Klein, Aaron Curns, Heidi Moline, Ariana Toepfer, Andrew Spieker, James Chappell, Natasha Halasa
Background. Human adenovirus (HAdV) is a common cause of acute respiratory illness (ARI) in children and is frequently detected with other viruses (codetection). The effect of co-detection on HAdV ARIs is incompletely understood, and recently the COVID-19 pandemic has disrupted the seasonality of many respiratory viruses. We aimed to characterize the demographics of children with HAdV ARI and co-detection patterns from 2016 through 2023.
Table 1. Demographic characteristics of children with HAdV, stratified by detection status: single HAdV detection vs. HAdV co-detected with one or more respiratory virus(es), December 2016 to August 2023, New Vaccine Surveillance Network. 1: The 2017 year includes cases from 12/01/2016 to 12/31/2017. The 2023 year includes cases from 01/01/2023 to 08/31/2023. We performed comparisons using Pearson's χ2 test for categorical variables and the independent-samples t-test with unequal variances for continuous variables.
Table 2. Single HAdV detection and HAdV co-detected with one or more respiratory virus(es), stratified by year, December 2016 to August 2023, New Vaccine Surveillance Network. Abbreviations: HAdV, human adenovirus; ccCoV, common cold coronaviruses; Flu, influenza; HMPV, human metapneumovirus; PIV, parainfluenza virus; HRV/RV, human rhinovirus/enterovirus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2. 1Cells labeled NA indicate p-value could not be calculated due to zero cell counts. We performed comparisons using Pearson's χ2 test for categorical variables and the independentsamples t-test with unequal variances for continuous variables.
Methods. The New Vaccine Surveillance Network is an active, prospective, population-based ARI surveillance system at 7 US sites. Children < 18 years old with fever and/or ARI symptoms for < 14 days were enrolled if they were seen in the emergency department or admitted to the hospital. Nasal and/or oropharyngeal swabs were tested for respiratory viruses, and demographic information was collected. Demographics were compared between HAdV single detection and co-detections with other respiratory viruses, and the proportions of HAdV co-detections were compared between years. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12718763&blobtype=pdf | # Clinical characteristics and prognostic factors in patients with stool cytomegalovirus positivity
Yi-Tien Hsuan, Ching-Hao Hsu, Cheng-Yu Chen, Yu-Jiun Chan, Hsin-Pai Chen, J Chin, Med Assoc
## Abstract
Background: The clinical association between cytomegalovirus (CMV) DNA detection in stool samples and patient outcomes remains underexplored. This study aimed to assess prognostic factors and viral kinetics in patients with positive stool CMV polymerase chain reaction (PCR). Methods: This retrospective cohort study included adult patients with positive stool CMV-PCR results at Taipei Veterans General Hospital (2016-2021). Clinical data, plasma, and stool viral loads (VLs) were analyzed. Receiver operating characteristic (ROC) curves and area under the curve (AUC) evaluated 30-day mortality prediction, with optimal cutoffs maximizing sensitivity and specificity. Kaplan-Meier survival analyses and Cox proportional hazards models identified predictors of 30-day mortality. Results: A total of 114 patients (mean age: 64.0 years, 64% male) were included. The median stool CMV VL was 629 copies/mL (interquartile range [IQR]: 263-7949). Plasma CMV DNA was detected in 76% with a median VL of 341 copies/mL (IQR: 10-1771). Stool and plasma VLs showed moderate correlation (ρ = 0.38, p < 0.0001). ROC analysis identified cutoffs for predicting 30-day mortality: stool 9654 copies/mL (AUC = 0.54; sensitivity 42%; specificity 81%) and plasma 1738 copies/mL (AUC = 0.60; sensitivity 47%; specificity 70%). In multivariate Cox analysis, stool CMV VL >9,654 copies/mL (adjusted hazard ratio: [HR] 2.69, 95% confidence interval [CI]: 1.06-6.84; p = 0.04) and plasma CMV VL >1738 copies/mL (adjusted HR: 2.66, 95% CI: 1.14-6.17; p = 0.02) were independent predictors of 30-day mortality. Septic shock and steroid use were also associated with increased mortality, whereas antiviral therapy ≥7 days was independently protective (adjusted HR: 0.26, 95% CI: 0.10-0.64; p = 0.003). Conclusion: Stool and plasma CMV VLs, antiviral treatment duration, and host factors such as immune status may influence outcomes in patients with intestinal CMV reactivation. Larger studies are needed to validate optimal VL thresholds for risk stratification.
## 1. INTRODUCTION
Cytomegalovirus (CMV), a Herpesvirus family member, has a global seroprevalence of up to 83% in the general population. 1 Among its end-organ diseases, CMV colitis is particularly notable in immunocompromised patients, 2 critically ill patients, 3 and patients with inflammatory bowel disease (IBD). 4 Clinical manifestations include abdominal pain, diarrhea, gastrointestinal bleeding, and, in severe cases, colonic perforation. 2,3,5,6 Approximately 10% of affected patients require surgical resection due to bowel perforation, uncontrolled bleeding, or colonic stricture. 7 Colonoscopy with tissue biopsy remains the diagnostic gold standard.
However, colonoscopy may not be feasible in patients with thrombocytopenia or critical illness because of the high risk of procedural complications. In contrast, detecting CMV DNA in stool samples is noninvasive. Although several studies have investigated stool CMV polymerase chain reaction (PCR) as a diagnostic tool for CMV colitis, [8][9][10][11][12] its reliability for diagnosis in clinical practice remains uncertain. More importantly, only a few studies have examined the prognostic value of stool CMV-PCR. While higher blood CMV viral loads (VLs) have been associated with increased mortality, [13][14][15] the relationship between stool CMV VL and mortality remains unclear. Therefore, the objective of this study was to investigate the clinical characteristics, prognostic factors, and virological features in a cohort of patients who tested positive for stool CMV-PCR.
## 2. METHODS
## 2.1. Data source and patient selection
In this retrospective cohort study, we identified adults (≥18 years) with positive stool CMV-PCR results at Taipei Veterans General Hospital between January 2016 and December 2021. The study protocol was approved by the Institutional Review Board of Taipei Veterans General Hospital (Approval No. 2021-10-004BC; initial approval: October 6, 2021; extension: August 26, 2022).
## 2.2. Data collection and definitions
Stool CMV-PCR testing was ordered at the discretion of attending physicians, typically based on gastrointestinal symptoms and the presence of clinical risk factors for CMV reactivation. Comprehensive data were retrospectively collected from the time point closest to the first positive stool CMV-PCR result within ±3 days, including comorbidities, immunocompromised status, laboratory findings, and use of immunosuppressants. In addition, information on serial plasma and stool CMV VLs, length of hospital stay, concomitant Clostridioides difficile infection (CDI), anti-CMV medications, and adverse events such as septic shock, acute decompensated heart failure, or acute kidney injury during the index hospitalization was recorded. Thirty-day all-cause mortality following CMV detection was also assessed.
Immunocompromised status was defined as the presence of acquired immunodeficiency syndrome, solid organ or hematopoietic stem cell transplantation, chemotherapy, immunosuppressive agents, or cumulative prednisolone-equivalent doses exceeding 225 mg within 1 month. 7,16 Diabetes mellitus (DM) was defined by a glycated hemoglobin level >6.5% or documentation in the medical records. Chronic liver disease included liver cirrhosis or chronic viral hepatitis, whereas chronic kidney disease was defined as stage 3 or higher. 17 Chronic heart disease was defined as American Heart Association/American College of Cardiology stage B or higher. 18 CDI was defined as detection of pathogenic DNA or toxin within 14 days of stool CMV detection. Septic shock was defined as sepsis accompanied by a systolic blood pressure <90 mmHg without inotropes. 19 Acute kidney injury was defined as an increase in serum creatinine >0.3 mg/dL or 1.5 times the baseline value. 20 Acute decompensated heart failure was defined as a clinical syndrome of worsening heart failure symptoms documented in the medical records. Refractory CMV infection, either definite or probable, was defined as persistent plasma CMV VL without decline after ≥2 weeks of adequate antiviral therapy. 21
## 2.3. Virology studies
For stool CMV-PCR, 2 to 3 mL of liquid or semi-liquid stool was mixed with an equal volume of phosphate-buffered saline or Universal Transport Medium, centrifuged, and the supernatant was collected. Nucleic acids were extracted using the QIAamp ® DNA Mini Kit and QIAcube ® Purification System (Qiagen, Hilden, Germany) before June 2020 or the LabTurbo™ Virus Kit with the LabTurbo™ 48C System (LabTurbo Biotech Corporation, Taipei, Taiwan) after June 2020. PCR was performed using the RealStar ® CMV PCR Kit 1.0 IVD (Altona Diagnostics, Hamburg, Germany) on the QuantStudio™ 5 Real-Time PCR System (Applied Biosystems™, Thermo Fisher Scientific, Waltham, MA).
For plasma CMV-PCR, Ethylenediaminetetraacetic acid (EDTA)-treated blood samples were centrifuged. Nucleic acid extraction and PCR were performed using the Roche cobas ® 6800 System with the cobas ® CMV Kit (Roche Diagnostics, Mannheim, Germany). CMV-PCR results ranged from 38 to 10 000 000 copies/mL, with 1 copy of DNA equivalent to 0.91 International Units, based on the First WHO International Standard. 22
## 2.4. Statistical analysis
Study variables were summarized as counts, percentages, or medians with interquartile ranges (IQRs). Categorical variables were compared using χ 2 or Fisher's exact tests, and continuous variables using t tests or Mann-Whitney U tests, as appropriate. Spearman correlation coefficients (ρ) were calculated to assess correlations between stool and plasma CMV-PCR. Receiver operating characteristic (ROC) curves and areas under the curve (AUCs) were used to evaluate survival prediction, with optimal cutoffs determined by maximizing sensitivity and specificity. Kaplan-Meier curves and Mantel-Cox log-rank tests were used to assess survival. Cox proportional hazards models with forward stepwise selection were applied to identify risk factors for 30-day mortality. Variables with p < 0.1 in univariate analysis, along with clinically relevant factors, were included in the multivariate analysis. Statistical significance was defined as p < 0.05. Analyses were performed using IBM SPSS Statistics 22.0 (IBM Corp., Armonk, NY).
## 3. RESULTS
## 3.1. Participants' characteristics
From January 2016 to December 2021, the Clinical Virology Laboratory of Taipei Veterans General Hospital examined 902 stool specimens for CMV-PCR. Of these, 150 specimens from 117 individuals tested positive. Three patients younger than 18 years were excluded, leaving 114 patients for analysis. Two patients were lost to follow-up before 30 days.
Among the 114 patients with positive stool CMV-PCR, 77 had diarrhea, 44 had hematochezia, 31 had fever, 14 had abdominal pain, and five had tarry stool. Seven patients had no documented symptoms, including one with a positive fecal occult blood test. Only one patient was receiving CMV prophylaxis at the time of testing.
Baseline characteristics are summarized in Table 1. Compared with survivors, non-survivors were older, had higher Acute Physiology and Chronic Health Evaluation (APACHE) II scores, were more likely to have septic shock, and had received higher cumulative steroid doses (prednisolone equivalent >225 mg/mo). They also had lower hemoglobin, platelet, and albumin levels, and higher C-reactive protein concentrations. Despite these differences, plasma and stool CMV VLs did not differ significantly between groups (Table 1). Among the 114 patients, 62 (54%) underwent colonoscopy within 1 month of stool CMV detection. Biopsies were performed in 38 patients (61% of colonoscopies), with CMV immunohistochemistry performed in 25 cases (66% of biopsies). Inflammation (31 patients, 82%) was the most common pathological finding, followed by confirmed CMV disease (12 patients, 32%), ulceration (eight patients, 21%), and neoplastic findings such as tubular adenoma, adenocarcinoma, or lymphoma (five patients, 13%).
## 3.2. CMV virology
Among 100 patients tested for plasma CMV-PCR, 76 had CMV viremia. The median plasma and stool CMV VLs were 341 copies/mL (IQR: 10-1771) and 629 copies/mL (IQR: 263-7949), respectively. Stool and plasma CMV VLs showed a moderate correlation (ρ = 0.38, p < 0.0001) (Fig. 1). Plasma CMV VLs were higher in immunocompromised patients ( Among the 114 patients with positive stool CMV-PCR, 32 (28%) did not receive antiviral therapy, 13 (11%) received ganciclovir or valganciclovir for <7 days, 15 (13%) for 7 to 13 days, 21 (18%) for 14 to 20 days, and 33 (29%) for ≥21 days. Of the 32 untreated patients, 21 had an alternative primary diagnosis, five had cytopenia (absolute neutrophil count <1000/μL or platelet count <50 000/μL), four had both cytopenia and an alternative primary diagnosis, and two experienced spontaneous symptom resolution. In the alternative primary diagnosis group, the median stool VL was 367 copies/mL (IQR: 138-801), and plasma VLs were mostly undetectable (median 0; IQR: 0-168).
To evaluate CMV viral kinetics, a subgroup analysis was performed on patients who underwent follow-up PCR testing more than one week apart (plasma, n = 62; stool, n = 33). Patients receiving ≥7 days of antiviral therapy (AVT) showed greater reductions in both stool and plasma VLs than those treated for <7 days or not treated (Table 2). In the ≥7-day treatment group, the median VL reductions were 2.4 log in plasma and 2.8 log in stool. Among 12 patients with paired stool samples collected at baseline (day 0) and between days 11 and 17 (all treated for ≥14 days), the median decline was 2.0 log (p = 0.012); nine were immunocompromised, with a median decline of 1.6 log (p = 0.04) (Fig. 2A). For plasma VLs, 41 patients had paired samples from day 0 and days 11 to 17. Those treated for ≥7 days demonstrated a numerically greater median decline in plasma VL compared with those treated for <7 days (3.1 log vs 2.4 log; p = 0.091). When restricted to the 31 immunocompromised patients, a similar trend was observed (3.0 log vs 2.8 log; p = 0.89) (Fig. 2B).
## 3.3. Outcome
Of 112 patients followed for more than 30 days, 27 (24%) died within 30 days. Among 73 immunocompromised patients, 19 (26%) died within the same period. ROC analysis of both the entire cohort and the immunocompromised subgroup identified the same optimal VL cutoff values for predicting 30-day mortality: 9654 copies/mL for stool and 1738 copies/mL for plasma. The AUCs for these cutoffs ranged from 0.53 to 0.60 (Table 3).
Stool CMV VL >9654 copies/mL and plasma CMV VL >1738 copies/mL were associated with higher 30-day mortality in the univariate Cox analysis of the entire cohort (Table 4). Kaplan-Meier survival analysis showed that stool CMV VL >9654 copies/mL was associated with higher 30-day mortality in both the entire cohort (p = 0.039) and the immunocompromised subgroup (p = 0.046), whereas plasma CMV VL >1738 copies/mL was significant only in the entire cohort (p = 0.028) (Fig. 3). Among 62 patients with CMV viremia and serial follow-up data, four met the criteria for refractory CMV infection. These patients had higher 30-day mortality in the univariate Cox model (entire cohort: hazard ratio [HR], 7.04; 95% confidence interval [CI], 2.07-23.99; p = 0.002 and immunocompromised subgroup: HR, 11.81; 95% CI, 2.65-52.56; p = 0.001).
Regarding treatment, AVT duration did not differ significantly between survivors and nonsurvivors (15.6 vs 11.4 days; p = 0.11). In the entire cohort, AVT ≥7 days did not reduce 30-day mortality (HR, 0.67; 95% CI, 0.31-1.42; p = 0.30). However, in the immunocompromised subgroup, AVT ≥7 days was significantly associated with reduced mortality (HR, 0.39; 95% CI, 0.16-0.97; p = 0.04). Other univariate predictors of 30-day mortality included age >65 years, acute decompensated heart failure, septic shock, and a cumulative prednisolone-equivalent dose >225 mg/mo before stool CMV detection. Among these factors, septic shock conferred the highest risk (Table 4).
Multivariate analyses are summarized in Table 5. Separate Cox models were constructed for the entire cohort and the immunocompromised subgroup to identify independent predictors of 30-day mortality. In the final model for the entire cohort, stool CMV VL >9654 copies/mL, plasma CMV VL >1738 copies/mL, a cumulative prednisolone-equivalent dose >225 mg/mo before stool CMV detection, and septic shock were independently associated with higher mortality. In the immunocompromised subgroup, stool CMV VL >9654 copies/mL and septic shock remained significant. AVT ≥7 days was independently associated with reduced mortality in both groups.
To address potential survivorship bias, a sensitivity analysis was performed using multivariate Cox proportional hazards models after excluding patients who died within 7 days of stool CMV detection (10 patients excluded from the entire cohort; eight from the immunocompromised subgroup). In this subset, AVT ≥7 days remained significantly protective in the entire cohort (adjusted HR, 0.10; 95% CI, 0.01-0.99; p = 0.049) but not in the immunocompromised subgroup (adjusted HR, 0.44; 95% CI, 0.04-4.20; p = 0.40).
## 4. DISCUSSION
This is the first study to examine prognostic factors and viral kinetics in patients with intestinal CMV reactivation, enrolling 114 cases. The cohort was heterogeneous, critically ill, and frequently hospitalized long-term, emphasizing the need to understand their clinical features and predictors of outcome. We found that higher gastrointestinal CMV viral burdens were associated with increased 30-day mortality, whereas AVT ≥7 days was protective. These findings highlight the potential importance of timely antiviral treatment in patients with high gastrointestinal CMV viral burdens. Stool CMV-PCR remains understudied for several reasons. Colonic CMV reactivation is relatively uncommon, limiting sample sizes; standardized protocols for stool testing are lacking, and its diagnostic value remains uncertain. Furthermore, stool CMV-PCR cannot replace tissue biopsy or plasma CMV-PCR, restricting its clinical application. Nonetheless, as a noninvasive tool, it shows potential and warrants further investigation. Because the prognosis of CMV infection may differ by immune status, we further analyzed outcomes in immunocompromised patients to reduce confounding and enhance clinical relevance.
Critical illness is associated with CMV reactivation, 3 and adverse outcomes may reflect underlying conditions. However, in our study, a stool CMV VL >9654 copies/mL was an independent risk factor for 30-day mortality in both the entire cohort and the immunocompromised subgroup, reflecting the potential role of CMV in contributing to gastrointestinal pathology. Although stool samples are manually diluted, the process likely preserves proportionality and reliability of results. While prior studies have not directly linked stool CMV VL to outcomes, higher levels have been associated with an increased risk of CMV colitis. 8,12 Similarly, elevated colonic tissue CMV VL in IBD 23 and higher CMV VL in respiratory specimens have correlated with worse outcomes. 24,25 These effects may result from both direct organ damage and CMV-mediated immunomodulation, which predisposes patients to secondary infections. 26 Although critical illness influences prognosis, our results and existing literature suggest that high stool CMV VL may contribute to mortality.
In addition to stool VL, higher plasma CMV VL was also associated with increased 30-day mortality in multivariate analysis, although not statistically significant in the immunocompromised subgroup. We also observed that patients with refractory viremia experienced worse outcomes. Prior studies have shown that higher plasma VL predicts mortality. [13][14][15] The lack of significance in the immunocompromised subgroup may reflect limited sample size or a stronger relevance of stool VL to end-organ disease in this population. Nevertheless, based on both previous research and our findings, plasma CMV VL remains a key parameter. Given these observations, at-risk patients with gastrointestinal symptoms should undergo evaluation of both stool and plasma CMV VLs. If viremia is detected, serial plasma VL monitoring is recommended.
The localized or systemic nature of CMV infection is influenced by the host's immune status. In our analysis, immunocompromised patients exhibited higher plasma VLs during intestinal CMV reactivation, consistent with a prior study. 3 A moderate correlation between stool and plasma CMV-PCR results was observed, similar to previous findings. 10,12 Notably, 24% of patients did not have CMV viremia, suggesting that CMV infection may be localized rather than systemic in certain cases.
Given the known beneficial effects of antiviral therapy for CMV colitis, 3,27 we assessed its impact on viral kinetics and 30-day mortality. Our findings suggest that patients treated with ganciclovir or valganciclovir for ≥7 days exhibited significant VL reductions, indicating a favorable virological response. Furthermore, although we could not distinguish CMV disease from viral shedding in patients without biopsy, AVT ≥7 days remained an independent protective factor for 30-day mortality in both the full cohort and the immunocompromised subgroup. To address survivorship bias, we performed a sensitivity analysis excluding patients who died within 7 days of stool CMV detection. The protective effect persisted in the full cohort but lost significance in the immunocompromised subgroup, likely due to limited statistical power. However, the direction of effect was maintained, suggesting that survivorship bias did not fully account for the association. These results support the potential benefit of adequate antiviral treatment, although further prospective studies using time-dependent modeling are needed to clarify the role of AVT.
The outcome of CMV infection is closely linked to host immunity. In our cohort, a cumulative prednisolone-equivalent dose >225 mg/mo before stool VL detection was an independent risk factor for 30-day mortality. Steroids were administered for conditions such as acute respiratory distress syndrome, chronic obstructive pulmonary disease exacerbation, autoimmune diseases, and hematologic diseases. Steroid use itself is known to increase the risk of CMV colitis and enteritis. 2,7 Our findings suggest that minimizing immunosuppressant exposure may benefit patients with intestinal CMV infection.
Although our primary focus was outcomes in intestinal CMV reactivation, the findings also provide insights into the diagnostic value of stool CMV-PCR. In our cohort, 25 patients underwent CMV immunohistochemistry staining, and 12 (48%) were confirmed to have CMV colitis. This is consistent with previous studies reporting a positive predictive value of up to 70% for stool CMV-PCR in high-risk patients-those who are immunocompromised, critically ill, have multiple comorbidities, or have IBD-who present with gastrointestinal symptoms. 12 While histology remains the gold standard, our data and prior studies support the use of stool CMV-PCR as a noninvasive screening tool to aid in confirming CMV colitis in this high-risk population.
This study has several limitations. First, as a single-center retrospective analysis, generalizability is limited. Second, some patients with intestinal CMV reactivation may have been missed because stool CMV-PCR testing depended on physician judgment. Third, certain key data, such as plasma CMV VL, were incomplete, and some patients were lost to follow-up. Fourth, the modest AUC values in ROC analysis suggest that cutoff values for stool and plasma CMV VL should be interpreted with caution. A larger cohort will be needed to establish optimal thresholds with confidence. Thus, our findings are exploratory and require validation in future studies. Nevertheless, we believe our study offers insights into noninvasive risk stratification and may help guide treatment decisions in suspected intestinal CMV cases.
In conclusion, our study highlights the prognostic value of stool and plasma CMV VLs and host immunity in patients with intestinal CMV reactivation. AVT appears to play an important role in controlling viral replication and improving outcomes. Further research is required to refine treatment strategies and optimize outcomes across diverse patient populations.
## References
1. Zuhair, Smit, Wallis et al. (2019) "Estimation of the worldwide seroprevalence of cytomegalovirus: a systematic review and meta-analysis" *Rev Med Virol*
2. Yeh, Chiu, Lai et al. (2021) "Clinical manifestations, risk factors, and prognostic factors of cytomegalovirus enteritis" *Gut Pathog*
3. Chaemsupaphan, Limsrivilai, Thongdee et al. (2020) "Patient characteristics, clinical manifestations, prognosis, and factors associated with gastrointestinal cytomegalovirus infection in immunocompetent patients" *BMC Gastroenterol*
4. Siegmund (2017) "Cytomegalovirus infection associated with inflammatory bowel disease" *Lancet Gastroenterol Hepatol*
5. Kim, Bahng, Kang et al. (2010) "Cytomegalovirus colitis in patients without inflammatory bowel disease: a single center study" *Scand J Gastroenterol*
6. Rafailidis, Mourtzoukou, Varbobitis et al. (2008) "Severe cytomegalovirus infection in apparently immunocompetent patients: a systematic review" *Virol J*
7. Ko, Peck, Lee et al. (2015) "Clinical presentation and risk factors for cytomegalovirus colitis in immunocompetent adult patients" *Clin Infect Dis*
8. Ganzenmueller, Kluba, Becker et al. (2014) "Detection of cytomegalovirus (CMV) by real-time PCR in fecal samples for the non-invasive diagnosis of CMV intestinal disease" *J Clin Virol*
9. Magdziak, Szlak, Mróz et al. (2020) "A stool test in patients with active ulcerative colitis helps exclude cytomegalovirus disease" *Scand J Gastroenterol*
10. Prachasitthisak, Tanpowpong, Lertudomphonwanit et al. (2017) "Short article: stool cytomegalovirus polymerase chain reaction for the diagnosis of cytomegalovirus-related gastrointestinal disease" *Eur J Gastroenterol Hepatol*
11. Zavrelova, Radocha, Pliskova et al. (2018) "Detection of cytomegalovirus DNA in fecal samples in the diagnosis of enterocolitis after allogeneic stem cell transplantation" *Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub*
12. Sattayalertyanyong, Limsrivilai, Phaophu et al. (2023) "Performance of cytomegalovirus real-time polymerase chain reaction assays of fecal and plasma specimens for diagnosing cytomegalovirus colitis" *Clin Transl Gastroenterol*
13. Mcbride, Sheinson, Jiang et al. (2019) "Correlation of cytomegalovirus (CMV) disease severity and mortality with CMV viral burden in CMV-seropositive donor and CMVseronegative solid organ transplant recipients" *Open Forum Infect Dis*
14. Boeckh, Leisenring, Riddell et al. (2003) "Late cytomegalovirus disease and mortality in recipients of allogeneic hematopoietic stem cell transplants: importance of viral load and T-cell immunity" *Blood*
15. Lu, Chen, Chan et al. (2021) "Clinical significance of posttreatment viral load of cytomegalovirus in patients with hematologic malignancies" *J Microbiol Immunol Infect*
16. Buttgereit, Da Silva, Boers et al. (2002) "Standardised nomenclature for glucocorticoid dosages and glucocorticoid treatment regimens: current questions and tentative answers in rheumatology" *Ann Rheum Dis*
17. Stevens, Ahmed, Carrero et al. (2024) "KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease" *Kidney Int*
18. Heidenreich, Bozkurt, Aguilar et al. (2022) "AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines" *J Am Coll Cardiol*
19. Marik, Lipman (2007) "The definition of septic shock: implications for treatment" *Crit Care Resusc*
20. Khwaja (2012) "KDIGO clinical practice guidelines for acute kidney injury" *Nephron Clin Pract*
21. Chemaly, Chou, Einsele et al. (2019) "Resistant Definitions Working Group of the Cytomegalovirus Drug Development Forum. Definitions of resistant and refractory cytomegalovirus infection and disease in transplant recipients for use in clinical trials" *Clin Infect Dis*
22. Hirsch, Lautenschlager, Pinsky et al. (2013) "An international multicenter performance analysis of cytomegalovirus load tests" *Clin Infect Dis*
23. Ciccocioppo, Racca, Paolucci et al. (2015) "Human cytomegalovirus and Epstein-Barr virus infection in inflammatory bowel disease: need for mucosal viral load measurement" *World J Gastroenterol*
24. Perret, Marechal, Germi et al. (2024) "Cytomegalovirus detection is associated with ICU admission in non-AIDS and AIDS patients with pneumocystis jirovecii pneumonia" *PLoS One*
25. Vergara, Cilloniz, Luque et al. (2018) "Detection of human cytomegalovirus in bronchoalveolar lavage of intensive care unit patients" *Eur Respir J*
26. Nichols, Corey, Gooley et al. (2002) "High risk of death due to bacterial and fungal infection among cytomegalovirus (CMV)seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection" *J Infect Dis*
27. Yoon, Lee, Kim et al. (2021) "Endoscopic features and clinical outcomes of cytomegalovirus gastroenterocolitis in immunocompetent patients" *Sci Rep* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12243388&blobtype=pdf | # Diagnostic criteria for Epstein-barr virusassociated encephalitis: a comment on Liu et al
Fereshte Sheybani, Mahboubeh Haddad
## Abstract
Background We read with great interest the article by Liu et al., titled "Clinical features and risk factors for Epstein-Barr virus-associated encephalitis: a retrospective cohort study. " The study provides valuable insights into the clinical spectrum and risk factors associated with EBV-related encephalitis.Main body While Epstein-Barr virus (EBV) is frequently detected in cerebrospinal fluid (CSF) during CNS infections, its role as a primary pathogen remains uncertain, especially in immunocompromised patients. We commend the authors for their efforts but seek clarification on the diagnostic criteria used to attribute causality to EBV. Specifically, we question whether the diagnosis relied solely on the detection of EBV DNA in the CSF or whether supporting parameters, such as viral load, CSF/serum ratios, or intrathecal antibody synthesis, were considered. The distinction between causative and incidental EBV detection is clinically significant and remains a challenge in neuroinfectious disease practice.
ConclusionFurther elaboration on how EBV-associated encephalitis was defined in the study would enhance its clinical relevance and aid practitioners encountering similar diagnostic complexities.
## Main body
The identification of EBV in the cerebrospinal fluid (CSF) is a common finding in patients with central nervous system (CNS) infections. However, distinguishing whether EBV is the primary etiologic agent or merely an incidental bystander remains a critical and unresolved issue. As the authors note, EBV is a lymphotropic herpesvirus that may enter the CNS via infected lymphocytes [2], particularly in immunocompromised individuals. Nonetheless, this mechanism of entry does not confirm causality in CNS disease [3].
We would like to request clarification from the authors on the criteria used to define EBV as the causative pathogen in their cohort. Was the diagnosis of EBV-associated
## Background
We read with great interest the recent article by Liu et al., titled "Clinical features and risk factors for Epstein-Barr virus (EBV)-associated encephalitis: a retrospective cohort study [1]". The authors systematically described the clinical characteristics of EBV-associated encephalitis and analyzed its risk factors in a large and well-characterized cohort.
encephalitis based solely on CSF detection via metagenomic next-generation sequencing (mNGS), or were additional markers-such as EBV viral load in CSF, the CSF-to-serum viral load ratio, or intrathecal antibody production-also evaluated?
Several studies have emphasized that high EBV viral load in the CSF is more indicative of active CNS infection, whereas low-level detection may reflect latent infection or incidental presence, especially in patients with underlying immunosuppression [4][5][6]. Without such quantitative and immunological parameters, there is a risk of misclassification, where EBV positivity may overshadow the true etiology of encephalitis.
Furthermore, the article discusses patients with EBV DNA detected in the CSF but attributes the cause of encephalitis to other pathogens. We are interested in understanding the basis for this distinction. Was the presence of another pathogen sufficient to negate EBV's etiological role, or were there predefined clinical or laboratory criteria that helped differentiate primary EBV encephalitis from EBV-positive cases due to other causes?
Addressing these questions would greatly enhance the clarity and interpretability of the study and would be of practical value to clinicians managing CNS infections.
## Conclusion
Clarification on how EBV-associated encephalitis was diagnosed and distinguished from other causes of encephalitis will help contextualize the findings and assist clinicians facing similar diagnostic dilemmas.
## References
1. Liu, Peng, Huang (2025) "Clinical features and risk factors for Epstein-Barr virus-associated encephalitis: a retrospective cohort study" *Virol J*
2. Hsu, Tokarz, Briese (2013) "Use of staged molecular analysis to determine causes of unexplained central nervous system infections" *Emerg Infect Disease J*
3. Lee, Kim, Kim (2021) "Clinical significance of Epstein-Barr virus in the cerebrospinal fluid of immunocompetent patients" *Clin Neurol Neurosurg*
4. Weinberg, Li, Palmer (2002) "Quantitative CSF PCR in Epstein-Barr virus infections of the central nervous system" *Ann Neurol*
5. Corcoran, Van Der Plas (2008) "The predictive value of cerebrospinal fluid Epstein-Barr viral load as a marker of primary central nervous system lymphoma in HIV-infected persons" *J Clin Virology: Official Publication Pan Am Soc Clin Virol*
6. Peuchmaur, Voisin, Vaillant (2023) "Epstein-Barr virus encephalitis: A review of case reports from the last 25 years" *Microorganisms* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12859218&blobtype=pdf | # Role of γδ T cells in turkey herpesvirus vaccine protection against Marek's disease virus
Mohammad Sabsabi, Ahmed Kheimar, Dominik Von, La Roche, Sonja Härtle, Dusan Kunec, Yulin Cong, Lisa Kossak, Theresa Von Heyl, Benjamin Schusser, Benedikt Kaufer
## Abstract
The Microbiology Society is a membership charity and not-for-profit publisher.Your submissions to our titles support the community -ensuring that we continue to provide events, grants and professional development for microbiologists at all career stages.Find out more and submit your article at microbiologyresearch.org
## INTROdUCTION
Marek's disease virus (MDV) is an alphaherpesvirus that causes one of the most prevalent virus-induced cancers in the animal kingdom [1]. The virus causes immunosuppression, neurological disorders and fatal T-cell lymphomas, as well as a mortality of up to 100% in unvaccinated chickens [2]. MDV infects its host primarily via the inhalation of contaminated dust. Upon inhalation, the virus is transported to the lymphoid organs, where it infects B and T cells. The virus spreads within the host in a cell-associated manner and establishes latency primarily in (Cluster of Differentiation 4) CD4 + αβ T cells. These cells can also be transformed, resulting in deadly lymphomas [3]. In addition, these cells can transport the virus to the feather follicle epithelium (FFE), where cell-free virus is produced and shed into the environment with the dander [4].
To protect chickens against MDV, billions of animals are vaccinated with live (attenuated) vaccines every year [5]. Three commercial vaccines are widely used: (i) the turkey herpesvirus [HVT; Mardivirus meleagridalpha1 (MeAHV1)], (ii) the naturally apathogenic Mardivirus gallidalpha3 (GaAHV3) strain SB-1 and (iii) the attenuated GaAHV2 strain CVI988/Rispens [6][7][8]. HVT is also extensively used as a vector vaccine, providing protection against other pathogens such as infectious bursal disease virus, Newcastle disease virus, avian influenza virus and infectious laryngotracheitis virus [9][10][11], aside from protecting against MDV. Even though these vaccines have been used for decades, vaccine protection and the contribution of specific immune cell subsets remain poorly understood. γδ T cells are highly abundant T cells in chickens and are thought to contribute to the early immune defence and long-term immunity [5,12]. γδ T cells can react quickly to infections, secrete cytokines and regulate immune responses [13]. We recently Abstract γδ T cells are a highly abundant lymphocyte subset in chickens and play key roles in early immune responses to infection. It has been recently shown that γδ T cells restrict Marek's disease virus (MDV) pathogenesis; however, it remained elusive if they play a role in vaccine protection. In this study, we vaccinated γδ T-cell-knockout chickens with the commercial turkey herpesvirus (HVT) vaccine and challenged them with very virulent MDV. The disease incidence was significantly increased in vaccinated chickens in the absence of γδ T cells. This increase was comparable to a previous study in unvaccinated γδ T-cell-knockout chickens, suggesting that γδ T cells only play a minor role in vaccine protection. Furthermore, the viral load in the spleen was significantly increased in the absence of γδ T cells. Interestingly, viral load in the skin and in dust shed by the animals was drastically increased, suggesting that the absence of γδ T cells affects MDV shedding. In addition, we quantified various immune cell subsets to determine if these could be responsible for the observed phenotypes. Together, our data indicate that γδ T cells only play a minor role in HVT-mediated protection, but their absence drastically affects shedding of this deadly pathogen in vaccinated animals.
demonstrated that γδ T cells play an important role in MDV pathogenesis, as disease and tumour incidence drastically increased in their absence. γδ T-cell-knockout chickens exhibited higher viral levels in the thymus and spleen, suggesting that γδ T cells restrict MDV replication in these organs [14]. Upon vaccination, γδ T cells drastically expand in multiple organs and show an increased cytotoxic activity [15,16], suggesting that these cells may have a role in vaccine-induced immunity. However, their precise role in vaccine protection remained unclear.
To assess the role of γδ T cells in vaccine protection, we vaccinated γδ T-cell-knockout chickens and challenged them with a very virulent MDV. Our data revealed that disease incidence was significantly increased in the absence of γδ T cells. This was comparable with the increase in unvaccinated γδ T-cell-knockout chickens [14], suggesting that γδ T cells only play a minor role in HVT-induced vaccine protection. Consistently, virus levels were significantly increased in the spleen. In addition, virus load was also drastically increased in the skin and in the dust, suggesting that the absence of γδ T cells impacts shedding of this highly oncogenic herpesvirus.
## MeTHOdS
## Cells and viruses
Chicken embryo cells were generated from specific pathogen-free VALO eggs (VALO BioMedia GmbH, Osterholz-Scharmbeck, Germany) as described previously [17]. The cells were then cultured in Eagle's minimal essential medium (PAN Biotech, Aidenbach, Germany), supplemented with 1-10% FBS (PAN Biotech) and 1% penicillin (100 U ml -1 )/streptomycin (100 µg ml -1 ) (AppliChem, Darmstadt, Germany). Cells were cultured in a humidified incubator at 37 °C and 5% Carbon dioxide (CO₂). The chickens were vaccinated using the HVT FC-126 vaccine strain. The very virulent RB-1B strain was used as a challenge virus, as described previously [14]. The viral stocks were subsequently frozen, stored in liquid nitrogen and titrated prior to their use [18].
## Animals and genotyping
To investigate the role of γδ T cells, we used genetically modified γδ T-cell-knockout chickens (TCR Cγ -/-) that have been thoroughly characterized previously [19]. Genotyping was carried out as described [14]. Briefly, peripheral whole blood samples were collected after hatch, and DNA was extracted using the NucleoSpin 96 Blood core kit (Macherey-Nagel, Düren, Germany) according to the manufacturer's instructions. PCR-based genotyping was conducted usingT cell receptor (TCR)-specific primers following the established protocol [14,19].
## In vivo experiment
One-day-old chicks were genotyped and divided into two groups: wild-type (WT; n=24) and γδ T-cell-knockout (TCR Cγ -/-; n=23). Each group was housed in isolation rooms in a Biosafety Level 2 (BSL2) facility with free access to food and water. The chicks were vaccinated subcutaneously with 2,000 p.f.u. of HVT. Five days post-vaccination (dpv), they were challenged subcutaneously with 2,000 p.f.u. of the very virulent RB-1B strain. Both vaccine and virus inoculum were back-titrated after the injection of the animals, which confirmed the intended dose. During the first weeks of life, the chicks had a mild diarrhoea, and we detected Clostridium perfringens in the faeces. Peripheral whole blood samples were collected at 7, 10, 14, 21 and 28 dpv. Additionally, feathers were collected from the chickens to quantify virus load in the FFE at 21, 28, 35 and 42 dpv. Dust samples were taken from the air filters in each room at indicated time points to assess virus shedding into the environment. During the experiment, the chickens were monitored twice daily for clinical symptoms caused by MDV. When symptoms were evident or at the end of the experiment (at 90 dpv), the chickens were humanely euthanized and assessed for gross tumour lesions, and spleens were harvested for quantitative PCR (qPCR).
## Quantification of virus genome copies
DNA from blood was extracted using the NucleoSpin 96 Blood Core Kit (Macherey-Nagel, Düren, Germany) following the manufacturer's instructions. To assess virus shedding, DNA from feather FFE and dust samples was extracted by treatment with proteinase K at 55 °C overnight, followed by phenol:chloroform extraction as described previously [20]. DNA from the spleen was processed using the InnuPREP DNA Mini Kit (Analytik-Jena, Berlin, Germany) according to the manufacturer's instructions. Vaccination was confirmed by qPCR detection of the HVT SORF1 gene in blood at 7 dpv [21]. Virus genome copies were measured by qPCR using primers and probes specific for infected cell polypeptide (ICP4) and cellular inducible nitric oxide synthase (iNOS), as published previously [22,23]. MDV DNA genome copies were normalized against the iNOS gene [24].
## Flow cytometry
To evaluate the effects of vaccination/infection on various immune cell subsets, absolute cell counts were performed to quantify B cells, CD8 -αβ, CD8 + αβ and γδ T cells in the blood, as described previously [14]. Briefly, peripheral whole blood samples were collected from the wing veins, stabilized using the TransFix ® reagent (Cytomark, Buckingham, UK) according to the manufacturer's guidelines, diluted in flow buffer and incubated with the published antibody mix [14]. The samples were assessed using a FACSCanto II (Becton Dickinson, Heidelberg, Germany). The data were analysed using FACSDiva (Becton Dickinson, Heidelberg, Germany) and FlowJo_v10.10.0 (FlowJo LLC, Oregon, USA) software [25].
## Statistical analysis
Statistical analyses were conducted using GraphPad Prism version 9 (GraphPad Software, Inc., San Diego, CA, USA). Details of the employed statistical tests are provided in the respective figure legends.
## ReSUlTS
## HVT-induced vaccine protection in the absence of γδ T cells
A recent study revealed that the absence of γδ T cells in MDV-infected animals significantly increased disease and tumour incidence [14]. To investigate whether γδ T cells also play a role in vaccine protection, we vaccinated both WT and γδ T-cellknockout chickens with the commercial HVT vaccine and challenged them with the very virulent RB-1B strain. qPCR analysis confirmed successful vaccination, with all animals (eight out of eight per group) testing positive for HVT in blood samples at 7 dpv. γδ T-cell-knockout chickens had a significantly higher Marek"s disease incidence (56%) compared with their WT counterparts (16%) (Fig. 1a). Tumour incidence was also mildly increased in the absence of γδ T cells (21%) compared with WT counterparts (12%) (Fig. 1b). However, the average number of tumours per bird (Fig. 1c) remained comparable between the two groups. The size and appearance of the tumours were also comparable between both groups. Importantly, a similar increase in disease and tumour incidence was previously observed upon infection of unvaccinated γδ T-cell-knockout chickens [14], indicating that γδ T cells play no, or only a minor, role in HVT vaccine protection.
## Absence of γδ T cells affects MdV replication in HVT-vaccinated chickens
To determine if the absence of γδ T cells affects virus levels in vaccinated animals, we assessed the viral load in the blood, spleen, skin and dust. Virus load in the blood was not affected in vaccinated γδ T-cell-knockout chickens compared with their WT counterparts (Fig. 2a). In contrast, MDV load was significantly higher in the spleen in the absence of γδ T cells (Fig. 2b), indicating that γδ T cells may restrict MDV replication in this organ. This is consistent with the increased virus levels in the spleen of unvaccinated γδ T-cell-knockout chickens [14]. Intriguingly, MDV replication was significantly increased by up to 50-fold in the FFE in the absence of γδ T cells (Fig. 2c). This was consistent with a reduction in virus levels in the dust at 21 and 28 dpv (Fig. 2d). A mild increase in the number of HVT copies was also observed in the FFE in the absence of γδ T cells (Fig. S1, available in the online Supplementary Material). Strikingly, an increase in MDV shedding was not observed in unvaccinated γδ T-cell-knockout chickens [14], suggesting that this increase is vaccination dependent and reflects a loss of HVT-induced control in the absence of γδ T cells.
## effect on immune cell populations in the blood
To determine if other immune cell populations are affected by the absence of γδ T cells in HVT-vaccinated chickens, we examined various immune cell subsets in the blood at 14 and 28 dpv. Flow cytometry confirmed that all γδ T-cell-knockout chickens completely lacked γδ T cells (Fig. 3a), while normal levels were detected in WT animals. This analysis also revealed that the number of B cells was not significantly altered at 14 dpv, and only a minor decrease was observed at 28 dpv (Fig. 3b) in the absence of γδ T cells. Similarly, the absolute number of CD8 -and CD8 + αβ T cells was not significantly altered in HVT-vaccinated γδ T-cell-knockout chickens (Fig. 3c,d). The results show that the absence of γδ T cells only has minimal impact on these immune cell subsets, suggesting that they are not responsible for the observed phenotype. dISCUSSION γδ T cells are highly abundant T cells in chickens and are thought to play a critical role in the immune response [13]. Upon MDV infection and vaccination, γδ T cells expand and express various cytokines [15,16,26]. In the absence of γδ T cells, disease and tumour incidence were significantly increased, and higher viral loads were detected in the thymus and spleen, indicating a role for these cells in the defence against MDV [14]. However, the role of γδ T cells in vaccine-induced protection against MDV remained poorly understood. To close this knowledge gap, we vaccinated γδ T-cell-knockout chickens with HVT and challenged them with very virulent MDV. HVT was selected as it is used as a vector vaccine in billions of chickens to protect them against MDV and other important pathogens [9]. In addition, HVT was chosen as the genetic background of the γδ T-cell-knockout chickens (LSL line) is rather resistant against MDV [14], and HVT was shown to provide only partial protection against very virulent MDV strains [27].
Our vaccine/challenge experiments revealed that γδ T-cell-knockout chickens had a significantly higher disease incidence (56%) compared with the WT cohort (17%). Interestingly, a comparable increase in disease incidence was previously observed in unvaccinated γδ T-cell-knockout chickens (70%) versus their WT counterparts (45%) [14]. This not only highlights the partial protection provided by HVT against very virulent MDV but also suggests that γδ T cells play no, or only a minor, role in the protection provided by this MDV vaccine. However, this effect may vary with other commonly used vaccines, such as SB-1 and CVI988, which should be investigated in future studies. The difference in tumour incidence was less pronounced in HVT-vaccinated γδ T-cell-knockout chickens (21%) compared with WT chickens (12%), when compared with experiments involving unvaccinated γδ T-cell-knockout chickens (45%), where a more than twofold increase was observed compared with WT chickens [14]. This could be due to γδ T-cell-independent protective effects induced by the HVT. Overall, these data indicate that γδ T cells do not play a major role in HVT-induced immune protection against disease and tumour formation. Despite the significantly higher disease incidence observed in the absence of γδ T cells, viral load in the blood remained comparable between γδ T-cell-knockout and WT chickens. In contrast, a significant increase in viral load in the spleen was observed in the absence of γδ T cells. This is consistent with the increased load in the spleen in unvaccinated γδ T-cell-knockout chickens observed previously [14]. This provides further support for a tissue-specific role of γδ T cells in limiting MDV replication. Previous studies also predicted a role of γδ T cells based on their increase in lung and lymphoid tissues and an increase in cytokines at early time points after MDV vaccination [15,16]. This tissue-specific reduction in MDV replication could also be responsible for the increase in Marek's disease incidence observed in the γδ T-cell-knockout chickens.
Intriguingly, MDV load was significantly higher in the FFE and in dust from HVT-vaccinated animals in the absence of γδ T cells. HVT vaccination has previously been shown to significantly reduce MDV shedding [28]. Despite HVT vaccination, γδ T-cell-knockout chickens shed MDV at levels similar to those in both unvaccinated animal groups in previous studies (1.6×10⁶ versus 1.5×10⁶ copies per microgram dust) [14]. These results suggest that HVT vaccination did not restrict MDV replication in the FFE and shedding in the absence of γδ T cells. Intriguingly, HVT levels in the FFE were also increased in vaccinated chickens in the absence of γδ T cells, indicating that γδ T cells may play a role in reducing virus transmission of both MDV and HVT. Furthermore, analysis of peripheral blood lymphocytes revealed no significant differences in the frequency of B cells or CD8 -, CD8 + αβ T cells between groups. This suggests that the elevated viral load observed in the FFE is specifically linked to the lack of γδ T cells rather than broader changes in immune cell composition.
In summary, our data reveal that γδ T cells play no, or only a minor, role in HVT-induced vaccine protection against MDV. However, the absence of γδ T cells significantly increased virus levels in the FFE and dust, indicating that γδ T cells limit viral replication in the skin and shedding of this deadly pathogen.
## References
1. Parcells, Burnside, Morgan (2012) "Marek's Disease Virus-Induced T-Cell Lymphomas"
2. Nair (2005) "Evolution of Marek's disease --a paradigm for incessant race between the pathogen and the host" *Vet J*
3. Schat, Baranowski (2007) "Animal vaccination and the evolution of viral pathogens" *Rev Sci Tech*
4. Davison, Nair (2004) "Marek's Disease: An Evolving Problem"
5. Witter (1997) "Increased virulence of Marek's disease virus field isolates" *Avian Dis*
6. Bertzbach, Conradie, You et al. (2020) "Latest insights into Marek's disease virus pathogenesis and tumorigenesis" *Cancers (Basel)*
7. Jarosinski, Tischer, Trapp et al. (2006) "Marek's disease virus: lytic replication, oncogenesis and control" *Expert Rev Vaccines*
8. Gimeno (2008) "Marek's disease vaccines: a solution for today but a worry for tomorrow?" *Vaccine*
9. Dunn, Dimitrov, Miller et al. (2019) "Evaluation of protective efficacy when combining Turkey herpesvirus-vector vaccines" *Avian Diseases*
10. Esaki, Noland, Eddins et al. (2013) "Safety and efficacy of a turkey herpesvirus vector laryngotracheitis vaccine for chickens" *Avian Dis*
11. Esaki, Godoy, Rosenberger et al. (2013) "Protection and antibody response caused by Turkey herpesvirus vector newcastle disease vaccine" *Avian Dis*
12. Chang, Xie, Dunn et al. (2014) "Host genetic resistance to Marek's disease sustains protective efficacy of herpesvirus of Turkey in both experimental and commercial lines of chickens" *Vaccine*
13. Fenzl, Göbel (2017) "Neulen M-L. Gammadelta T cells represent a major spontaneously cytotoxic cell population in the chicken" *Dev Comp Immunol*
14. Sabsabi, Kheimar, You et al. (2024) "Unraveling the role of gammadelta T cells in the pathogenesis of an oncogenic avian herpesvirus" *mBio*
15. Hao, Li, Li et al. (2021) "An anti-tumor vaccine against Marek's disease virus induces differential activation and memory response of gammadelta T cells and CD8 T cells in chickens" *Front Immunol*
16. Matsuyama-Kato, Iseki, Boodhoo et al. (2022) "Phenotypic characterization of gamma delta (gammadelta) T cells in chickens infected with or vaccinated against Marek's disease virus" *Virology*
17. Hernandez, Brown (2010) "Growth and maintenance of chick embryo fibroblasts (CEF)" *Curr Protoc Microbiol*
18. Schat, Calnek, Fabricant (1982) "Characterisation of two highly oncogenic strains of Marek's disease virus" *Avian Pathol*
19. Von Heyl, Klinger, Aumann et al. (2023) "Loss of alphabeta but not gammadelta T cells in chickens causes a severe phenotype" *Eur J Immunol*
20. Bello, Francino, Sánchez (2001) "Isolation of genomic DNA from feathers" *J Vet Diagn Invest*
21. Conradie, Bertzbach, Trimpert et al. (2020) "Distinct polymorphisms in a single herpesvirus gene are capable of enhancing virulence and mediating vaccinal resistance" *PLoS Pathog*
22. Jarosinski, Kattenhorn, Kaufer et al. (2007) "A herpesvirus ubiquitin-specific protease is critical for efficient T cell lymphoma formation" *Proc Natl Acad Sci*
23. Jarosinski, Margulis, Kamil et al. (2007) "Horizontal transmission of Marek's disease virus requires US2, the UL13 protein kinase, and gC" *J Virol*
24. Kaufer, Jarosinski, Osterrieder (2011) "Herpesvirus telomeric repeats facilitate genomic integration into host telomeres and mobilization of viral DNA during reactivation" *J Exp Med*
25. Seliger, Schaerer, Kohn et al. (2012) "A rapid high-precision flow cytometry based technique for total white blood cell counting in chickens" *Vet Immunol Immunopathol*
26. Matsuyama-Kato, Shojadoost, Boodhoo et al. (2023) "Activated chicken gamma delta T cells are involved in protective immunity against Marek's disease" *Viruses*
27. Djeraba, Musset, Lowenthal et al. (2002) "Protective effect of avian myelomonocytic growth factor in infection with Marek's disease virus" *J Virol*
28. Islam, Walkden-Brown, Groves et al. (2008) "Kinetics of Marek's disease virus (MDV) infection in broiler chickens 1: effect of varying vaccination to challenge interval on vaccinal protection and load of MDV and herpesvirus of turkey in the spleen and feather dander over time" *Avian Pathol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12522436&blobtype=pdf | # Exploring the challenge of diagnosing and tracking Oropouche virus during Brazil's largest recorded dengue epidemic in 2024
Virology Journal, José Joaquim, Silva Júnior, Pablo Cantalice, Santos Farias, Gabriel Caruso, Novaes Tudella, Regina Lopes, Ketlyn Rafaeli, Aquino Cardoso, Lianna Da Rocha Rosa, Felipe Pedro, Jesus Loeblein, Henrique Tomás, Roso, Edmilson Ferreira De Oliveira-Filho, Eduardo Flores
## Abstract
Background In 2024, Brazil experienced its largest dengue virus (DENV) epidemic alongside the widest spread of Oropouche virus (OROV) to date. This scenario may have strongly impacted public health surveillance efforts, as DENV and OROV infections often lead to similar clinical signs and symptoms, making clinical differential diagnosis challenging. Herein, we investigated whether the co-circulation of DENV and OROV in Brazil in 2024 may have hampered OROV detection and tracking, potentially contributing to possible underdiagnosis.
## Methods
The number of confirmed cases of dengue and Oropouche in 2024, as well as the probable states of infection, were collected on June 9, 2025, from the Brazilian Ministry of Health database. For dengue, cases were analyzed according to the diagnostic approach: laboratory testing or diagnosis by clinical-epidemiological criteria. All Oropouche cases considered here were confirmed by laboratory testing. Finally, dengue and Oropouche cases were grouped by state and epidemiological week, and their temporal-spatial overlap was assessed to discuss its potential negative effect on OROV reporting.
Results A total of 5,956,257 dengue cases were recorded in Brazil, of which only 2,191,326 were laboratory confirmed. All states, including the Federal District, registered dengue cases, with highest number in São Paulo and Minas Gerais states (Southeast region). Dengue peaks occurred in early 2024 for Brazil overall and for individual states. Regarding Oropouche, 13,856 cases were recorded in 22 states, the majority in Espírito Santo (Southeast region) and Amazonas (North region). Oropouche cases in Brazil (total cases) showed two peaks: a smaller one at the beginning of 2024 and a higher one at the end of the year. On the other hand, OROV peaked in different periods depending on the state analyzed. In general, Oropouche cases were described close to the dengue reporting period, mainly in
## Introduction
Oropouche virus (OROV) infections (genus Orthobunyavirus) have been described in Brazil since the 1960's, mainly affecting the North region [1,2]. However, in recent years, OROV has spread throughout Brazilian territory, causing infection in 23 out of 26 states, covering all Brazilian regions. Briefly, in 2023, 833 Oropouche cases were reported in Brazil, 831 in the North region (mainly in Amazonas state) and two in the Southeast region (Espírito Santo state) [3]. In the following year, 2024, OROV had its highest expansion recorded in Brazil, with 13,856 cases described, reaching all regions of the country, with Espírito Santo and Amazonas being the two states with the highest incidence, 5,868 and 3,231 cases, respectively [3]. In 2025, up to May 27, 10,572 Oropouche cases were reported; again, Espírito Santo state with the highest number of cases (6,200) [3].
The increase in reported Oropouche cases in Brazil has raised concerns worldwide, particularly regarding sporadic deaths in adults and newborns, as well as congenital malformations [3,4]. Furthermore, the introduction of OROV into Brazilian areas endemic for other arboviruses, such as dengue (DENV), chikungunya (CHIKV) and/or Zika (ZIKV), whose infection often leads to similar clinical signs and symptoms, may further hinder its clinical-epidemiological diagnosis [5].
In this scenario, one of the most important differential diagnoses is between OROV and DENV, the arbovirus posing the greatest public health burden in the Americas, particularly in Brazil [6]. This issue was especially relevant in 2024, the year with the highest number of reported dengue cases in Brazil's history and the year with the greatest recorded spread of OROV [3,6]. This context also raises important questions because most dengue diagnoses in 2024 were based on clinical-epidemiological criteria [6], which could have contributed to the underdiagnosis of Oropouche cases, compromising knowledge about its real spread. In addition to the impacts on public health surveillance, this potential limitation hinders in-depth investigation into the OROV (re-) emergence, compromising evolutionary, climatic and environmental insights, which could be more accurate if OROV spread were closely monitored.
Herein, we conducted a retrospective study to discuss the challenge of differential diagnosis between dengue and Oropouche in 2024 in Brazil. In summary, we analyzed the spatial and temporal distribution of DENV and OROV, focusing mainly on the Brazilian states with the highest incidence for each of these viruses, to assess whether their co-circulation may have hindered the diagnosis of Oropouche cases, as well as OROV tracking.
## Methodology
## Design and study area
This is a retrospective and descriptive study on dengue and Oropouche cases reported in Brazil in 2024. Briefly, Brazil is divided geopolitically into five regions (North, Northeast, Central-West, Southeast and South), in which 27 federative units are distributed (26 states and one Federal District). Initially, we analyzed the temporal distribution of dengue and Oropouche cases in Brazil (total cases per epidemiological week, EW). Then, dengue and Oropouche cases were evaluated according to their spatial and temporal distribution (cases per state/ federal district per EW). For each of these analyses, i.e., total cases or cases by state/federal district, dengue cases were analyzed according to their diagnostic approach, laboratory test or clinical-epidemiological diagnosis. All Oropouche cases considered herein were confirmed by laboratory testing [3]. Finally, we compared the curves of dengue and Oropouche cases in Brazil and its states, analyzed the co-circulation time of DENV and OROV, and discussed whether this scenario could have influenced OROV epidemiological records.
## Data collection
Data on dengue and Oropouche, including probable site of infection (state or Federal District), EW and diagnostic strategy (laboratory test or clinical-epidemiological diagnosis for dengue) were collected from the Brazilian Ministry of Health database on June 9, 2025 (DENV: h t t p s : / / w w w . Amazonas, Pará and Rondônia states (northern Brazil), where there was evident overlap between the dengue and Oropouche curves.
Conclusion DENV and OROV co-circulated in most of Brazil in 2024. Furthermore, most dengue diagnoses were based on clinical-epidemiological criteria. This approach, combined with the clinical and temporal-spatial overlap between dengue and Oropouche cases, may have contributed to misdiagnoses, potentially hindering OROV detection and tracking. These findings reinforce the importance of laboratory testing for accurate arbovirus identification and surveillance in Brazil.
## Keywords
## Spatial and temporal analyses
The dengue and Oropouche case maps were constructed using QGIS 3.43 software, and the corresponding cases graphs were created using R-3.0.3 software (tidyr, dplyr and ggplot2 packages).
## Ethical aspects
Not applicable. The study was conducted with secondary public data, without patient identification.
## Results
## Spatial distribution of dengue and Oropouche cases
In 2024, 5,956,257 dengue cases were recorded in Brazil, of which 2,191,326 were confirmed by laboratory testing, while the remaining cases were diagnosed based on clinical-epidemiological criteria. Most dengue cases were reported in two states in the Southeast region, São Paulo (2,144,847) and Minas Gerais (1,395,227) (Fig. 1A); corresponding to 59.43% of the total cases in Brazil (3,540,074/5,956,257). The federative units with the highest incidence rates (cases/100,000 inhabitants) were Minas Gerais (6,543.4), Federal District (Distrito Federal) (6,525.9), Paraná (5,268.5), São Paulo (4,665.4) and Goiás (4,380.4) (Fig. 1B).
During the same period, 13,856 Oropouche cases were recorded in Brazil, the majority in Espírito Santo (Southeast region) (42.34%, 5,868/13,856) and Amazonas (North region) (23.31%, 3,231/13,856); together, these data correspond to 65.66% of Oropouche cases described in the country (9,099/13,856) (Fig. 1A). Espírito Santo, Rondônia and Amazonas were the states with the highest incidence rates (cases/100,000 inhabitants): 143.04, 90.01 and 73.49, respectively (Fig. 1B).
## Temporal distribution of dengue and Oropouche cases
Dengue cases were recorded throughout 2024 in Brazil, with the highest incidence between EW 10 and 17. Laboratory-confirmed and clinical-epidemiological diagnosed cases followed similar patterns, with peaks occurring at nearly the same periods (Fig. 2). All Brazilian states, including the Federal District, reported dengue cases in 2024 (Fig. 1), showing a case curve similar to the total cases in Brazil (Fig. 3 and Supplementary file 1). Dengue cases were also recorded throughout 2024 in all Brazilian states, including Federal District (Fig. 3 and Supplementary file 1). As described above, São Paulo and Minas Gerais states recorded the highest numbers of dengue cases. Given the epidemiological importance of both states for Brazil, their dengue and Oropouche case curves were used to illustrate the differential diagnostic challenges between dengue and Oropouche in 2024 (Fig. 3). The peaks of dengue cases in São Paulo and Minas Gerais states occurred in EW 12-20 and 8-12, respectively. Furthermore, the period of most recorded cases of Oropouche in São Paulo and Minas Gerais occurred close to the dengue peaks in both states (Fig. 3).
Regarding OROV, Brazil experienced two peaks of cases in 2024: a smaller peak in the first EWs and the largest one in the EW 47-52 (Fig. 2). During 2024, 22 Brazilian states recorded Oropouche cases, with peaks occurring at different periods across states (Fig. 4 and Supplementary file 1). As previously shown, most cases were reported in Espírito Santo and Amazonas states. Therefore, following the approach used for dengue, we focused on the states with the highest Oropouche cases numbers to discuss the challenge of differential diagnosis between dengue and Oropouche in 2024. In Espírito Santo state, Oropouche cases emerged after EW 13, coinciding with a decline in recorded dengue cases (Fig. 4A). Furthermore, after EW 44-45, during the period of highest Oropouche incidence, laboratory-confirmed dengue cases decreased considerably, while clinically-epidemiological-based cases increased slightly (Fig. 4). In Amazonas state, the curve of Oropouche cases overlapped with dengue cases, with both diseases showing peak incidence in the first EWs of 2024 (Fig. 4B). A similar pattern was observed in Pará and Rondônia states (northern Brazil) (Supplementary file 1).
## Discussion
Arbovirus infections often present similar clinical signs and/or symptoms, which makes clinical diagnosis challenging [8]. Furthermore, the co-circulation of arboviruses with overlapping seasonal patterns represents an additional obstacle, also making accurate clinical-epidemiological diagnosis difficult. In this scenario, laboratory testing represents unquestionably the most reliable strategy for accurate arbovirus diagnosis. However, most arbovirus-endemic countries and territories, including Brazil, lack sufficient laboratory capacity to meet diagnostic demand. Consequently, arbovirus cases are frequently diagnosed based on clinical-epidemiological criteria, which can lead to under-or overestimation of reported data, in addition to making it difficult to identify and monitor emerging viruses. Considering these aspects, we investigated the clinical-epidemiological challenge of diagnosing and tracking OROV in 2024, the year of the largest dengue epidemic recorded in Brazil.
Initially, we analyzed the temporal distribution of dengue cases recorded in Brazil in 2024 (total cases) and observed that most cases occurred in the first half of the year. This seasonality, likely driven by the influence of environmental and climatic conditions on the vectors Aedes aegypti and Ae. albopictus, is well established and has been consistently observed over the years [9][10][11][12].
On the other hand, Brazil recorded two distinct peaks of Oropouche cases in 2024, a smaller peak and a larger peak at the beginning and end of the year, respectively. Herein, there was a temporal overlap between dengue and Oropouche cases in the first months of 2024, which appears to have been the most critical period for differential diagnosis. Although the curves for laboratoryconfirmed dengue cases and those diagnosed through clinical-epidemiological criteria show similar patterns, indicating an alignment between the two diagnostic approaches, it is necessary to explore whether the clinical-epidemiological dengue diagnosis, especially in early 2024, may have influenced the detection and monitoring of OROV. Furthermore, 22 out of 26 states reporting dengue cases also recorded Oropouche cases. This temporal and spatial overlap of DENV and OROV during this critical year warrants further careful investigation.
Thus, we analyzed the temporal distribution of DENV and OROV by Brazilian states. This approach is justified by the continental dimension of Brazil, which may have led to the introduction of OROV in different areas at different times, as well as by the marked environmental differences among Brazilian regions, which probably influence the distribution and seasonality of arboviruses. We observed that the temporal distribution of dengue cases was similar among the 26 states and Federal District of Brazil and, consequently, similar to the curve of total cases in the country. However, OROV displayed a different temporal pattern that varied by state. This difference likely reflects the progressive spread of OROV throughout the Brazilian territory.
Given the different curves of Oropouche cases among Brazilian states, we selected those with the highest case number for Oropouche and dengue (i.e., the major contributors to Brazil's total cases) to delve deeper into the challenges of clinical-epidemiological diagnosis. São Paulo and Minas Gerais states accounted for more than half of dengue cases in Brazil in 2024. As described for Brazil, both states had a peak in dengue cases early in the year, while Oropouche cases peaked in two periods, late and early 2024. Interestingly, despite the low number of Oropouche cases in São Paulo and Minas Gerais, the period with the most recorded cases of OROV infections in both states was close to the peak of dengue, which may have impacted the differential diagnosis between dengue and Oropouche.
Regarding OROV, Amazonas and Espírito Santo states reported the highest number of Oropouche cases in 2024. In Amazonas, most cases of dengue and Oropouche occurred in the first months of 2024, with considerable overlap between the curves of both viruses. Pará and Rondônia states, also located in northern Brazil, showed a similar pattern of dengue and Oropouche cases. These findings corroborate the broader discussion about the challenges of clinical-epidemiological investigation for dengue and Oropouche cases in 2024 [5], which may have hampered timely identification and surveillance of OROV. In Espírito Santo, although dengue and Oropouche showed peaks incidence in distinct periods, Oropouche cases were also recorded at the beginning of the year, during the decline in dengue cases. Overall, these temporal overlaps between DENV and OROV circulation further reinforce concerns about the clinical-epidemiological challenges of identifying and monitoring OROV during 2024 in Brazil.
It is noteworthy that we focused on dengue and Oropouche due to their major epidemiological relevance in Brazil in 2024. However, it is also important to highlight that Brazil is endemic for other arboviruses, such as CHIKV and ZIKV [13], which must also be considered in the differential diagnosis. In addition to the similarity in clinical signs and symptoms, DENV, CHIKV and ZIKV share the same vector and, consequently, the same seasonal pattern, which makes differential diagnosis based on clinical-epidemiological criteria even more challenging. This issue is especially important for CHIKV, since 2024 was also the year with the highest number of chikungunya cases (more than 200,000 cases) in the last three years in Brazil (2023-2025) [6,14]. Furthermore, continuous surveillance is crucial to identify and monitor other arboviruses circulating in Brazil, such as West Nile (WNV), Mayaro (MAYV), Ilheus (ILHV) and Rocio (ROCV) [15], which can also hinder the clinical-epidemiological diagnosis of arbovirus infections. Finally, the co-circulation of OROV and other orthobunyaviruses, which may cross-react in serological assays, is also an issue that must be adequately addressed in arbovirus surveillance in Brazil [16].
## Conclusion
We analyzed the spatial and temporal distribution of dengue and Oropouche cases at different levels, Brazil (total cases) and by its states and Federal District, and observed that DENV and OROV co-circulated in most regions of the country. Consequently, the possible clinical-epidemiological misdiagnosis of Oropouche as dengue could have led to underreporting and compromised OROV surveillance. Therefore, to enhance arbovirus surveillance, we recommend both expanding laboratory testing capacity to ensure accurate differential diagnosis and maintaining thorough monitoring of OROV infections to establish their precise seasonal pattern. These data can help identify the period of highest Oropouche incidence, supporting appropriate and timely public health measures, as well as improving clinical-epidemiological differential in settings with limited laboratory capacity.
## Limitations
Our study was conducted using secondary data made publicly available by the Brazilian Ministry of Health. Therefore, we did not have access to clinical and laboratory details that could contribute to the deepening our discussion. Furthermore, data from the Brazilian Ministry of Health may be updated periodically, so some numbers discussed here may change slightly depending on when the database is consulted. These changes, however, do not influence the conclusions of our study.
## References
1. Pinheiro, Pinheiro, Bensabath et al. (1962) "Epidemia de vírus Oropouche em Belém" *Rev Serv Esp Saude Publica*
2. Riccò, Corrado, Bottazzoli et al. (2024) "Re-) emergence of Oropouche virus (OROV) infections: systematic review and meta-analysis of observational studies" *Viruses*
3. Ministry, Health, Oropouche (2025)
4. Martins-Filho, Soares-Neto, Oliveira-Júnior et al. (2024) "The underdiagnosed threat of oropouche fever amidst dengue epidemics in Brazil" *Lancet Reg Health Am*
5. Junior, Lf, Oliveira-Filho et al. (2018) "A scoping review of Chikungunya virus infection: epidemiology, clinical characteristics, viral co-circulation complications, and control" *Acta Trop*
6. Nascimento, Pastor, Lopes et al. (2015) "Retrospective cross-sectional observational study on the epidemiological profile of dengue cases in Pernambuco state"
7. Silva, Santos, Corrêa et al. (2016) "Temporal relationship between rainfall, temperature and occurrence of dengue cases" *Brazil Cien Saude Colet*
8. Churakov, Villabona-Arenas, Kraemer et al. (2019) "Spatio-temporal dynamics of dengue in Brazil: seasonal travelling waves and determinants of regional synchrony" *PLoS Negl Trop Dis*
9. Machado, Neto, Lotufo et al. (2023) "Spatiotemporal Dengue Fever Incidence Associated with Climate in a Brazilian Tropical Region" *Geographies*
10. Farias, Pastor, Gonçales et al. (2023) "Epidemiological profile of arboviroses in two diferente scenarios: dengue circulation vs. Dengue, Chikungunya and Zika co-circulation" *BMC Infect Dis*
11. Fujita, Salvador, Nali et al. (2024) "Oropouche in Brazil in 2024" *J Travel Med*
12. Fischer, Oliveira-Filho, Drexler (2020) "Viral emergence and immune interplay in flavivirus vaccines" *Lancet Infect Dis*
13. Gibrail, Fiaccadori, Souza et al. (2016) "Detection of antibodies to oropouche virus in non-human primates in Goiânia city" *Goiás. Rev Soc Bras Med Trop* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12804346&blobtype=pdf | # The cellular activating protein-1 cFos regulates influenza A virus replication
Antoine Gerodez, François Dufrasne, Olivier Denis, Mieke Steensels, Bénédicte Lambrecht, Lionel Tafforeau, Caroline Demeret, Cyril Barbezange
## Abstract
Although Jun and Fos families have been considered to regulate cell proliferation positively (reviewed in ref. [4]), the role of cFos in cell proliferation remains debatable. While mouse fibroblasts deficient in both cFos and FosB had a reduced proliferative activity, the inhibition of cFos alone induced no effect on cell [5]. In addition, double-knockout mice lacking both cFos and FosB, but not the single knockout mice, had smaller body sizes than their wild-type counterparts [6]. On the other hand, the knockdown of cFos resulted in inhibited proliferation of a human osteosarcoma cell line, and the stable overexpression of cFos led to increased proliferation of immortalized human hepatocytes under low serum conditions [7,8]. cFos also regulates apoptosis. Although cFos was shown to promote apoptosis in different cell types [9][10][11], it seemed to repress apoptosis in a human osteosarcoma
## INTRODUCTION
Activating protein-1 (AP-1) transcription factors consist of homo-and hetero-dimers composed of basic region-leucine zipper proteins that belong to Jun (cJun, JunB and JunD), Fos (cFos, FosB, Fra-1 and Fra-2), Maf (c-Maf, MafB, MafA, MafG/F/K and Nrl) and ATF (ATF2, LRF1/ATF3, B-ATF, JDP1 and JDP2) families. These transcription factors regulate many cellular functions, including cell proliferation, differentiation, inflammation and apoptosis (reviewed in refs. [1,2]). cFos and cJun are commonly associated in a complex that binds in a sequence-specific manner to the promoter and enhancer regions of target genes [3]. cell line and to decrease neuronal cell death in the hippocampus during kainic acid-induced seizure [7,12]. Therefore, the regulation of the cell fate by cFos and the AP-1, in general, is complex and depends on the cell type, the type and duration of the stimulus and the involvement of other transcription factors. In addition, AP-1 has been described as an activator of inflammation (reviewed in ref. [2]), although some evidence also suggests a specific anti-inflammatory role of cFos. A significant increase in the pro-inflammatory cytokines such as TNF-α, IL-6 and IL-12 p40 was observed in mouse macrophages and mice deficient in cFos [13], while the expression of cFos was observed to inhibit IL-12 p40 promoter activity in mouse macrophages [14]. Finally, independently of its transcriptional factor activity in the nucleus, cFos possesses a cytoplasmic function as an activator of lipid synthesis at the endoplasmic reticulum level (reviewed in ref [15]).
Several viruses were shown to hijack AP-1 proteins to support their replication. AP-1 binding to the intragenic regulatory region of the pol gene of the human immunodeficiency virus type 1 seemed to help recruit the cellular DNA-dependent RNApolymerase II to the viral promoter, supporting viral transcription [16]. cFos was found to bind to multiple gene promoters of Kaposi's sarcoma-associated herpesvirus and to enhance viral lytic transcription [17]. siRNA-based experiment identified a role of cFos in hepatitis C virus replication and propagation [18]. cFos was also shown to promote virus replication of alpha-and gamma-coronaviruses by delaying and reducing apoptosis [19].
Influenza A viruses (IAVs) are major respiratory pathogens responsible for human seasonal epidemics and pandemics, posing a persistent threat to global health. IAVs are enveloped viruses and belong to the Orthomyxoviridae family [20]. Their genome consists of eight negative-sense single-stranded RNA segments. Each viral RNA (vRNA) segment is encapsidated by nucleoproteins (NP) and attached to the heterotrimeric polymerase complex, composed of the PB1, PB2 and PA subunits, thus forming the viral ribonucleoprotein (vRNP). To start an infection, IAV enters the cell through an endocytic pathway. Inside the cell, vRNPs are released and transported to the nucleus, where viral transcription and replication occur. Transcription and replication are carried out by the viral polymerase complex. Primary transcription generates viral mRNAs, which are exported to the cytoplasm for translation by host ribosomes. PB2, PB1, PA, NP and NS1 are expressed early, while HA, NA, M1 and NS2 are expressed later during the viral cycle [21]. Viral proteins PB2, PB1, PA, NP, M1 and NS2 are transported back to the nucleus, and genome replication then occurs. Newly synthesized vRNAs are assembled with PB2, PB1, PA and NP, resulting in progeny vRNPs that are subsequently exported to the cytoplasm. These vRNPs are further incorporated into progeny particles containing HA, NA, M2 and M1 inserted in or present at the cell membrane. Finally, progeny virions are released from the cell by budding for subsequent infection [20,22]. AP-1 transcription factors might play a role in IAV polymerase activity regulation, as the inhibition of cJun was shown to impair IAV H5N1 replication in human lung cells [23]. However, AP-1 also appears to take part in the innate antiviral response following IAV infection. IAV-induced AP-1 activation was shown to activate the expression of interferon-β and promote NLRP3 inflammasome activation [24,25]. Furthermore, the viral protein NS1 antagonized IAV-induced AP-1 activation [26]. Thus, the role of the AP-1 transcription factors in IAV infection remains unclear. In this study, we observed that the infection of human cells with human IAVs resulted in the upregulation of cFos and cJun subunits. The role of cFos was then investigated using specifically depleted cells, showing that knockdown significantly impaired viral replication efficiency. Further characterization revealed different potential mechanisms by which cFos may support IAV replication.
## METHODS
## Cell lines
HEK293T, A549, MDCK and MDCK-SIAT1 [27] cells were provided by Institut Pasteur Paris. HEK293T and A549 cells were grown in Dulbecco's Modified Eagle's Medium (DMEM) (Life Technologies cat#41965039) supplemented with 10% FBS (Life Technologies cat#10270-106), 10 U ml -1 of penicillin and 10 µg ml -1 of streptomycin (Life Technologies cat#15140122). MDCK and MDCK-SIAT1 cells were grown in Modified Eagle's Medium (Life Technologies cat#31095029) supplemented with 5% FBS, 10 U ml -1 of penicillin and 10 µg ml -1 of streptomycin. All cell lines were grown at 37 °C in 5% CO 2 .
## Viruses
Three human IAVs were used: the two currently circulating seasonal IAVs in the human population, A/Bretagne/7608/2009 (H1N1)pdm09 (referred to as pH1N1 in 'Results') and A/Centre/1003/2012 (H3N2) and the laboratory-adapted A/WSN/33 (H1N1) (referred to as WSN in 'Results'). All viruses were produced by reverse genetics [28,29] and amplified on MDCK cells, except for the H3N2 strain, which was amplified on MDCK-SIAT1 cells. pH1N1 and H3N2 subtypes grow poorly in A549 cells. Two mutations in the pH1N1 HA (A517G and G834A) and one mutation in the H3N2 HA (G460T) were introduced by sitedirected mutagenesis into the HA-encoding plasmid pH1N1 and H3N2 of the reverse genetic system to generate A549-adapted pH1N1 and H3N2 strains, capable of efficient replication in A549 cells [30]. Four avian IAVs were used (Fig. S1, available in the online Supplementary Material): A/Anas platyrhynchos/Belgium/14325/07 (H3N8); A/Anas platyrhynchos/Belgium/10399/2018 (H4N6); A/Gallus_gallus/Belgium/ 16070_0002/2021 (H5N1); and A/chicken/Israel/1163/2011 (H9N2). All avian IAVs were initially isolated in embryonated chicken eggs, except for H9N2, which was isolated on MDCK cells. All avian IAVs were subsequently propagated in MDCK cells for experimental use.
## Antibodies, chemicals and reagents
Rabbit anti-β-actin (cat#4967), rabbit anti-cFos (cat#2250), goat anti-rabbit IgG, HRP-linked (cat#7074) and horse anti-mouse IgG, HRP-linked (cat#7076) antibodies were purchased from Cell Signaling Technology. Mouse anti-M2 (cat#MA1-082), rabbit anti-NP (cat#PA5-32242), chicken anti-calreticulin (cat#PA1-902A), chicken anti-fibrillarin (cat#PA5-143565A), goat anti-rabbit Alexa Fluor Plus 647 (cat#A32733) and goat anti-chicken IgY (H+L) Alexa Fluor 594 (cat#A-11042) antibodies were purchased from Thermo Fisher Scientific. The rabbit anti-NA antibody (cat#GTX125974) was purchased from Genetex. The mouse anti-NS1 (cat#sc-130568) and the mouse anti-NP (cat#sc-80481) antibodies were purchased from Santa Cruz Biotechnology. The goat antimouse IgG-FITC antibody (cat#F0257) was purchased from Sigma-Aldrich. The Annexin V-Alexa Fluor 488 apoptosis detection reagent (cat#A13201) was purchased from Thermo Fisher Scientific, and the 7-Amino-Actinomycin D (7-AAD) (cat#559925) was purchased from BD Biosciences. The T-5224 small inhibitor (cat#HY-12270) was purchased from MedChemExpress. CellTiter-Glo® Luminescent Cell Viability Assay (cat#G7570), Dual-Glo® Luciferase Assay System (cat#E2920) and Renilla Luciferase Assay System (cat#E2810) were purchased from Promega.
## siRNA-based assays
Multi-cycle infection siRNAs were purchased from Horizon Discovery (ON-TARGETplus SMARTpools and non-targeting Control pool). The siRNA sequences were designed by Horizon Discovery and are listed in Table S1. 6.5 µl of siRNA at a concentration of 5 µM was added to 243.5 µl of DharmaFECT1 transfection reagent (Horizon discovery cat#T-2001) and OptiMEM GlutaMAX medium (Life Technologies cat#51985034) to obtain a final volume of 250 µl. The 250 µl mixture was added to one well of a 12-well tissue culture plate (Greiner). Following a 30-min incubation period at room temperature, 2×10 5 A549 cells diluted in 1 ml of DMEM supplemented with 5% FBS were seeded on top of siRNA-transfection reagent complexes and incubated at 37 °C in 5% CO 2 . The final concentration of siRNA was 25 nM per well. At 48 h post-transfection (hpt), the cells were washed and infected with pH1N1 at a multiplicity of infection (moi) of 10 -2 or 10 -3 , WSN at a moi of 10 -4 and H3N2 at a moi of 10 -1 or 10 -2 at 35 °C in 5 % CO 2 . Supernatants were collected at 0, 24 and/or 48 h post-infection (hpi), and viral titres were determined by plaque assays using MDCK-SIAT1 cells [31].
## Single-cycle infection
The cells were treated with siRNAs as described above. At 48 hpt, the cells were washed, inoculated with WSN at a moi of 3 and incubated for 1 h at room temperature. Inoculum was then removed, fresh OptiMEM GlutaMAX medium was added and cells were incubated at 35 °C (5% CO 2 ). Timepoint 0 hpi was defined as the start of the cell incubation at 35 °C.
## Cell viability and luciferase-based knockdown efficiency experiments
siRNA reverse transfection was performed in a white 96-well tissue culture plate (Greiner). Volumes were adjusted to obtain a siRNA final concentration of 25 nM per well. Briefly, 0.65 µl of siRNA at a concentration of 5 µM was added to 24.35 µl of Dhar-maFECT1 transfection reagent and OptiMEM GlutaMAX medium to obtain a final volume of 25 µl. The 25 µl mixture of siRNA-DharmaFECT1 was added to one well of the white 96-well plate. Following a 30-min incubation period at room temperature, 1.5×10 4 A549 cells diluted in 100 µl of DMEM supplemented with 5% FBS were seeded on top of the siRNA-transfection reagent complexes and incubated at 37 °C in 5% CO 2 . Cell viability was determined at 48 hpt using the CellTiter-Glo Luminescent Viability Assay kit according to the manufacturer's instructions (Promega). To measure the efficiency of siRNA knockdown, siRNA-treated A549 cells were transfected at 24 h post-siRNA transfection with 10 ng of plasmids encoding siRNA-targeted protein fused with the full-length Gaussia luciferase (pGlucFL) using Lipofectamine 3000 (Thermo Fisher Scientific cat#L3000001). The luciferase activity was measured 24 h later in cell lysates using the Renilla luciferase assay reagent (Promega). All luciferase activities were measured on a GloMax Explorer (Promega).
## Real-time quantitative PCR
RNAs were extracted using the Quick-RNA Miniprep Kit (Zymo Research cat#R1054) from A549 cells treated with siRNA and infected with WSN or mock-infected. The kit uses a column to remove most of the genomic DNA and a subsequent treatment with DNase I to degrade the remaining genomic DNA. RNA concentrations were measured with NanoDrop One (Thermo Fischer Scientific) and adjusted to 100 ng µl -1 . Two hundred nanogram RNA was reverse transcribed using the Accuscript High Fidelity first-strand cDNA synthesis kit (Agilent cat#200820) following the manufacturer's instructions. Briefly, 200 µg RNA was mixed with 4 µl buffer 10 × AccuScript High Fidelity RT-PCR System, 1.6 µl 100 mM dNTPs, 1 µg anchored-oligo(dT) primer, and nuclease-free water in a 33 µL reaction mixture. The mixture was heated for 5 min at 65 °C and cooled for 5 min to room temperature before the addition of 40 U RNase block, 10 mM DTT and 2 µl Accuscript RT in a 40 µl final volume reaction mixture. Reverse transcription was performed at 42 °C for 60 min in a thermocycler, and the RT enzyme was then inactivated at 72 °C for 15 min. Real-time quantitative PCR (qPCR) was performed using GoTaq qPCR master mix containing BRYT green dye (Promega cat#A6001) on an ARIA MX (Stratagene MX3005P, Agilent Technologies) in a total volume of 20 µl. Each reaction mixture included 10 µl GoTaq qPCR MasterMix, 1 µl of each diluted forward and reverse primers (10 µM), 1.25 µl template cDNA and nuclease-free water up to 20 µl. Primers for the AP-1 transcription factors (sequences based on ref. [19]) were purchased from Integrated DNA Technologies. Primer sequences for cytokine genes and phospholipid enzymes were designed and verified by Sigma-Aldrich. All the primers are listed in Table S2. A control with 1.25 µl of water instead of cDNA was included in each run to check for reagent contamination. A 'No RT' control was also included for each targeted gene of the real-time qPCR experiments to check the efficiency of the DNase I treatment. The thermal cycling conditions for all qPCRs were as follows: denaturation at 95 °C for 2 min, followed by 40 cycles consisting of 95 °C for 15 s, and an annealing and extension phase at 60 °C for 1 min. Afterwards, a melting curve analysis was performed to determine the specificity of the reaction products. The modulation of RNA expression for all targeted genes was normalized to βactin expression (housekeeping gene) and analysed using the 2 -ΔΔCt method (delta-delta Ct method).
## Strand-specific real-time qPCR
RNAs were extracted using the Quick-RNA Miniprep Kit (Zymo Research cat#R1054) from A549 cells treated with siRNA and infected with WSN or mock-infected. Strand-specific real-time qPCR for NP and NA vRNAs, cRNAs and mRNAs was performed as previously described [32]. Briefly, cDNAs complementary to NP and NA vRNAs, cRNAs and mRNAs were synthesized using SuperScript™ IV Reverse Transcriptase (Thermo Fisher Scientific cat#18090010) and tagged primers in order to add a strand-specific tag unrelated to influenza virus sequence at the 5′ end. Tagged cDNAs were then used as a template for the qPCR reaction using a tag-specific primer and a segment-specific primer. β-Actin (housekeeping gene) mRNA was reverse transcribed using anchored-oligo(dT) and SuperScript™ IV Reverse Transcriptase (Thermo Fisher Scientific cat#18090010). All qPCRs were performed using GoTaq qPCR master mix containing BRYT green dye (Promega cat#A6001) on an ARIA MX (Stratagene MX3005P, Agilent Technologies) in a total volume of 20 µl. Each reaction mixture included 10 µl GoTaq qPCR MasterMix, 1 µl of each diluted forward and reverse primers (10 µM), 7 µl tenfold diluted template cDNA and nuclease-free water up to 20 µl. The primers were designed as described in ref. [32] to specifically detect NP and NA vRNAs, cRNAs and mRNAs. A control with 7 µl of water instead of cDNA was included in each run to check for reagent contamination. The thermal cycling conditions for all qPCRs were as follows: denaturation at 95 °C for 2 min, followed by 50 cycles consisting of 95 °C for 15 s, and an annealing and extension phase at 60 °C for 1 min. Afterwards, a melting curve analysis was performed to determine the specificity of the reaction products. The modulation of vRNA, cRNA and mRNA expression for all targeted genes was normalized to βactin expression (housekeeping gene), and RNA fold changes relative to the condition 0 hpi were calculated using the 2 -ΔΔCt method. All the primers for both reverse transcription and qPCR are listed in Table S3.
## Western blot
A549 cells treated with siRNA and infected with WSN or mock-infected were lysed at 0, 3, 6 and 9 hpi using RIPA lysis and extraction buffer (Thermo Fisher Scientific cat#89900) supplemented with 100-fold diluted Halt™ Protease and Phosphatase Inhibitor Cocktail, EDTA-free (Thermo Fisher Scientific cat#78441). Briefly, the cells were washed with cold PBS, and 200 µl supplemented RIPA buffer was added to each well of the 12-well plate. Following a 15-min incubation period on ice, lysed cells were collected and centrifuged at 14,000 g for 15 min. Cell lysate supernatants were kept, and protein quantification was performed using Pierce™ BCA Protein Assay Kits (Thermo Fisher Scientific cat#23225) following the manufacturer's instructions. Cell lysates were then mixed with 4× Laemmli Sample Buffer (BioRad cat#1610747) supplemented with 355 mM of βmercaptoethanol (Sigma-Aldrich cat#M6250) and boiled at 90 °C for 10 min. Equal amounts of protein samples were loaded into each well and separated using 8-16% Mini-PROTEAN® TGX Stain-Free™ Protein Gel (BioRad cat#4568105) in a Mini-PROTEAN Tetra Vertical Electrophoresis Cell (BioRad cat#1658004), at 200 V for 30 min. The resolved proteins were then transferred to a 0.2 µM nitrocellulose membrane for 7 min at 25 V using the Trans-Blot® Turbo™ Transfer System (BioRad cat#1704150). The membrane was blocked for 1 h at room temperature with PBS-0.05% Tween 20-3 % BSA and then incubated with the indicated primary antibodies diluted in PBS-0.05% Tween 20-3 % BSA at 4 °C overnight. After 3 washes for 5 min with PBS-0.1% Tween 20, the membrane was incubated with 1:5,000 HRP-linked goat anti-rabbit IgG or HRP-linked horse anti-mouse IgG, at room temperature for 2 h. After 3 washes for 5 min with PBS-0.1% Tween 20, proteins were detected by chemiluminescence using Pierce™ ECL Plus Western Blotting Substrate (Thermo Fisher Scientific cat#32132) and ChemiDoc imaging system (Biorad). Protein band intensities were determined using ImageJ software. All experiments were repeated three times, with similar results, and one of the representative immunoblots is shown.
## Immunofluorescence staining
A549 (2×10 5 ) cells were seeded on coverslips (13 mm in diameter) in a 12-well plate (Greiner) in DMEM supplemented with 10% FBS, 10 U ml -1 of penicillin and 10 µg ml -1 streptomycin and incubated at 37 °C for 24 h. Cells were infected with WSN at a moi of 3 p.f.u. cell -1 . At 0, 3, 6 and 9 hpi, cells were fixed with 3% paraformaldehyde for 20 min and permeabilized with PBS-0.1% Triton X-100 for 20 min. Cells were blocked with PBS-3% BSA for 1 h and incubated with primary antibodies anti-cFos (1/200), anti-NP (1/200) and anti-fibrillarin (1/200) or anti-calreticulin (1/100) overnight at 4 °C. Following multiple washes with PBS, cells were incubated with FITC goat anti-mouse IgG (1/200), Alexa Fluor 594 goat anti-chicken IgY (1/200) and Alexa Fluor 647 goat anti-rabbit (1/300) secondary antibodies for 1 h at room temperature. Following multiple washes with PBS, coverslips were mounted in ProLong Gold Antifade Mountant with DNA Stain DAPI (Thermo Fisher Scientific cat#P36941) and analysed under an inverted fluorescence microscope (Nikon ECLIPSE Ts2R) using an ×60 objective lens. The percentage of nuclear signal for cFos was determined using the Fiji software. For each image, five infected cells were selected. For each cell, the integrated density of the red fluorescence, corresponding to cFos signal, was measured in the nucleus (defined by DAPI staining) and in the cytoplasm. The percentage of nuclear signal was calculated as the ratio of the integrated density in the nucleus to the total integrated density of the cell (nuclear+cytoplasmic).
## Minigenome assay
HEK293T (3×10 4 ) cells were transfected with NT or cFos siRNA in a white 96-well tissue culture plate, as described above. After 48 h of knockdown, cells were transfected using Polyethyleneimine 'MAX' (MW 40,000) 1 mg ml -1 (Polysciences cat#24765) with expression pCIneo or pcDNA3 plasmid vectors encoding the IAV proteins PA, PB1, PB2 (25 ng each plasmid) and NP (50 ng) from different IAV strains (pH1N1, WSN and H3N2), a reporter plasmid vector (pPR7-firefly-(-)) encoding the firefly luciferase in the negative-sense orientation flanked by the noncoding regions of the segment 5 of WSN driven by a polymerase I promoter (5 ng) [33] and a plasmid vector (polIII-Renilla) constitutively expressing Renilla luciferase (5 ng). Twenty-four hours post-transfection, luciferase activities were measured using the Dual-Glo luciferase assay system (Promega). Polymerase activity, proportional to Firefly luciferase activity, was normalized to Renilla luciferase activity to take into account the transfection rate.
## Measure of apoptosis and necrosis rates
A549 cells were transfected with NT or cFos siRNA as described above. After 48 h of knockdown, cells were infected with WSN at a moi of 3 p.f.u. cell -1 or mock-infected. At 24 hpi, the apoptosis and necrosis rates were determined. Cells were washed with PBS and detached using trypsin-EDTA 0.05% phenol red (Thermo Fisher Scientific cat#25300054). DMEM supplemented with 10% FBS was then added, and the collected cells were centrifuged at 2,000 r.p.m. for 5 min. Pellet cells were resuspended in Annexin-binding buffer (HEPES 10 mM, NaCl 140 mM, CaCl 2 2.5 mM, pH 7.4) at 10 6 cells ml -1 . One hundred microlitres of cells was then transferred into a FACS tube, and 5 µl of Annexin V-Alexa Fluor 488 apoptosis detection reagent and 7-AAD was added. Staining was performed for 15 min at room temperature. Four hundred microlitres of Annexin-binding buffer was then added. Cells were further fixed in 4% paraformaldehyde at 4 °C in the dark until analysis. As a positive control, cells were also treated with 10 µM Camptothecin (Thermo Fisher Scientific cat#J62523.MD), known to induce apoptosis. Cells were acquired on a FACSVerse flow cytometer (BD Biosciences). Data were analysed with the FACSuite software (BD Biosciences). Around 2,000 events were counted for each sample.
## Statistical analysis
Unpaired Student's t-test was used for all statistical analyses, using GraphPad Prism software v.9.00 (GraphPad Software). Differences between groups were considered statistically significant at P<0.05; significance levels are as follows: *P<0.05, **P<0.01, ***P<0.001, ns: non-significant.
## RESULTS
## IAV multiplication upregulates the AP-1 transcription factors and is impaired in cFos-knockdown cells
Following single-cycle IAV infection in A549 cells, the expression of AP-1 transcription factors cFos, FosB, cJun and JunD was upregulated at 6 and 9 hpi, as measured by real-time qPCR. The mRNA level of cFos and FosB was >50-fold higher than in mock-infected cells. For cJun and JunB, the mRNA level at 9 hpi was between 5 and 10 times higher in infected cells. No difference in the mRNA expression of Fra1, Fra2, JunD and ATF2 was observed at any time point (Fig. 1a). The role of cFos and cJun on viral replication was further investigated using small interfering RNA (siRNA)-mediated silencing. Upon individual siRNA knockdown of cFos and cJun, A549 cell viability remained above 90% compared to the non-target (NT) siRNA-treated cells, with knockdown efficiency of protein expression estimated at 92% and 61% for cFos and cJun, respectively (Fig. S1a,b). Human IAV replication was not affected by the individual knockdown of cJun (Fig. 1b). In contrast, cFos-knockdown significantly impaired the replication of human seasonal pH1N1, H3N2 and WSN IAV strains (Fig. 1b). The effect was also observed for IAVs of avian origin (Fig. S1d). Knockdown of the well-described proviral factor RAB11 [34] was used as a control (Fig. S1c).
## The nuclear function of cFos, but not its cytoplasmic function, regulates IAV multiplication
We were further interested in deciphering which function of cFos could support IAV replication. At the endoplasmic reticulum (ER), cFos activates phospholipid synthesis by physically interacting with the CDP diacylglycerol synthase 1 (CDS1) and phosphatidylinositol 4 kinase type IIα (PI4KIIα) enzymes of the polyphosphoinositide (PIP) lipid pathway [15]. On the contrary, another enzyme involved in the PIP pathway, the CDP diacylglycerol inositol 3 phosphatidyltransferase enzyme, is not regulated by Fig. 1. cFos mRNA expression is upregulated during IAV infection, and its knockdown reduced IAV replication. (a) A549 cells were infected with WSN at moi of 3 p.f.u. cell -1 . Total RNAs were extracted at the indicated time point post-infection, and the mRNA expression levels of AP-1 were determined by real-time qPCR. mRNA fold changes were calculated using the 2 -ΔΔCt method compared to the mock cells at each time point. Results are expressed as the mean+sd determined in three independent experiments. The significance of the difference to the 0 hpi timepoint was tested with unpaired t-tests using GraphPad Prism software for each cellular factor (ns, not significant; *P<0.05, **P<0.01, ***P<0.001). (b) A549 cells were transfected with 25 nM of the NT or the indicated siRNAs. At 48 hpt, cells were infected with the following viruses at the indicated moi in p.f.u. cell -1 : A/ Bretagne/7608/2009(H1N1pdm09) (pH1N1, moi of 10 -3 ); A/Centre/1003/2012(H3N2) (H3N2, moi of 10 -2 ); A/WSN/33(H1N1) (WSN, moi of 10 -4 ). At 0, 24 and 48 hpi, viral titres were determined by plaque-forming assay. Results are expressed as the mean±sd p.f.u. ml -1 of three independent experiments. The area under the curve (AUC) (not shown) was determined for each virus and condition, taking the p.f.u. ml -1 at timing 0 hpi as the baseline. The significance was tested on AUCs with unpaired t-tests using GraphPad Prism software (ns, not significant; *P<0.05, **P<0.01, ***P<0.001).
cFos [15]. No major differences in IAV replication were detected upon siRNA-based knockdown of both cFos-activated enzymes CDS1 and PI4KIIα compared to NT siRNA-treated A549 cells (Fig. 2a), with cell viability and knockdown efficiency remaining above 95% (Fig. S2). In contrast, upon treatment with T-5224, a specific inhibitor of cFos/AP-1 DNA binding [35], IAV replication was impaired. The viral titres for pH1N1 and WSN viruses were significantly lowered by 0.5 and 2 log 10 , respectively, in A549 cells treated with 20 µM T-5224 compared with DMSO-treated cells (Fig. 2b), with cell viability remaining superior to 50% (Fig. S3a). At 6 hpi, cFos was mainly localized in the nucleus, more specifically in the nucleoplasm but not in the nucleoli (marked by anti-fibrillarin), and hardly in the ER (marked by anti-calreticulin) (Figs 2c andS4).
## The apoptosis is increased in cFos-knockdown cells during IAV infection
cFos regulates apoptosis via its nuclear activity [9,10]. In non-infected cells, upon cFos-knockdown or in NT control cells, apoptosis and necrosis rates were about 2% (Fig. 3b). Upon IAV infection, both apoptosis and necrosis increased (Fig. 3a andb). The apoptosis rate was significantly higher in cFos-knockdown cells (12%) compared to control cells (7%). A similar trend was also observed with necrosis, but the difference was not statistically significant (Fig. 3a andb).
## cFos regulates inflammation during IAV infection
To monitor the regulation of cytokine expression by cFos, the expression of inflammatory cytokines IL-1β, IL-6, IL-12, TNF-α (Fig. 4a) and type I interferons IFN-α1 and IFN-β (Fig. 4b) was measured in cFos-knockdown cells during single-cycle IAV infection. NT-siRNA treatment was used as a control representing what happens upon IAV infection when endogenous cFos is unaffected. IL-1β mRNA induction was significantly lower in cFos-knockdown cells compared to the NT siRNA-treated cells (Fig. 4a). A decrease in the IL-6 mRNA induction, especially at 3 hpi, was also observed, but the overall difference was not statistically significant (Fig. 4a). No differences in IL-12A and IL-12B mRNA (both subunits forming the IL-12) as well as TNF-α mRNA were observed between cFos-knockdown and NT siRNA-treated cells (Fig. 4a). On the contrary, the mRNA level of IFN-β but not of IFN-α1 was significantly higher in cFos-knockdown cells (Fig. 4b). These findings suggest that, during IAV infection, cFos may promote the transcription of IL-1β and potentially IL-6 and repress that of IFN-β.
## Viral transcription and expression of viral proteins are impaired in cFos-knockdown cells
The consequences of cFos-knockdown on the viral replication were further evaluated. Minigenome assays for three human IAVs -pH1N1, H3N2 and WSN -showed a significant reduction of the viral polymerase activity in cFos-knockdown cells (Fig. 5a). The levels of vRNA, cRNA and mRNA production of NP and NA segments during single-cycle WSN infection were further evaluated in both cFos-and NT siRNA-treated cells using strand-specific real-time qPCR [32] (Fig. 5b). The production of NA mRNA was lower at 6 hpi (446 vs. 2,246 RNA fold change, P-value=0.023) and 9 hpi (162 vs. 548 RNA fold change, P-value=0.104) in cFos-knockdown cells compared to NT siRNA-treated cells. On the contrary, the production of vRNA and cRNA of NA tended to be slightly higher at 6 and 9 hpi, although the differences were not significant. For the NP segment, no significant differences were obtained, and the levels of mRNA, cRNA and vRNA did not appear to be affected by cFos-knockdown (Fig. 5b). Similar results were obtained at the protein level. Less NA and M2 viral proteins accumulated at 6 and 9 hpi in cFos-knockdown cells than in NT siRNA-treated while no difference in the NP viral protein levels was observed. A slight decrease in NS1 viral protein seemed to occur at 3 hpi upon cFos-knockdown, but no such differences were observed at later time points. In agreement with what was observed at the mRNA level (Fig. 1a), cFos protein expression upon IAV infection was increased, especially at 9 hpi in control cells, and knockdown efficiency was confirmed in cFos-depleted cells as clearly observed at 6 and 9 hpi (Fig. 5c). The same pattern in the expression level of viral proteins was observed upon treatment with T-5224, the specific inhibitor of the cFos/AP-1 dimer DNA binding activity (Fig. S3b). Altogether, these findings highlighted a potential role of cFos in the viral transcription of lately expressed viral proteins. [32]. RNA fold changes relative to the condition 0 hpi were calculated using the 2 -ΔΔCt method. The results are expressed as the mean±sd of three independent experiments. The significance of the difference to NT was tested with unpaired t-tests using GraphPad Prism software (ns, non-significant, *P<0.05). (c) Total cell lysates were harvested at the indicated times post-infection and analysed by immunoblot using antibodies directed against the indicated proteins. Immunoblot results representative of three independent experiments are shown. Band intensity of the indicated proteins was normalized to β-actin, and the mean ratios±sd of three independent experiments are presented in the table. The significance of the difference to NT (indicated by '-' in sicFos) was tested by unpaired t-tests in GraphPad Prism Software (ns, non-significant; *P<0.05, **P<0.01).
## DISCUSSION
AP-1 transcription factors are commonly induced by viral infections and were found to regulate the replication of several viruses [16][17][18][19]. IAV infection was shown to activate AP-1 transcription factors through phosphorylation by the JNK signalling pathway [24]. However, the role of AP-1 in IAV replication remains unclear. In the current study, we found that the transcription of cFos and cJun AP-1 factors was upregulated following IAV infection and that cFos appeared to support IAV replication.
The role of cFos is dependent on its location in the cell. At the endoplasmic reticulum, cFos activates de novo phosphatidylinositol phosphate (PIP) lipid synthesis by interacting with CDS1 and PI4KIIα enzymes [15]. The individual siRNA-based knockdown of CDS1 and PI4KIIα enzymes did not impair IAV replication (Fig. 2a). Although the individual knockdown of CDS1 in cardiomyoblast cells was previously shown to be critical and to impair the PIP pathway [36], it cannot be ruled out that CDS1 and PI4KIIα protein isoforms might have played a compensatory role in our experiments. However, it is more likely that cFos does not influence IAV replication through its cytoplasmic role as an activator of PIP lipid synthesis, but rather through its nuclear role as an AP-1 transcription factor. This is further supported by the localization of cFos in the nucleoplasm during IAV infection (Figs 2c andS4) and the reduction of IAV replication in the presence of cFos nuclear activity inhibitor T-5224 (Fig. 2b). The most described cFos AP-1 dimer nuclear partner, cJun, did not appear to regulate IAV replication in our study, although it was previously found to support H5N1 IAV replication in A549 cells [23]. These discrepancies could be attributed to differences in the viral strains but more probably to the presence of residual cJun, since in our study, cJun depletion was less effective compared to cFos depletion. cFos might also regulate viral replication independently of its partner cJun, through binding to nucleic acids. In this regard, the overexpression of cFos mRNA upon infection may increase the likelihood of cFos forming homodimers or remaining as monomers, which were both shown to be able to bind cellular DNA [37,38]. The activity of cFos is regulated via phosphorylation by MAPK kinases, including ERKs and p38 [39,40]. These kinases are activated during IAV infection [41,42]. ERK5 was shown to phosphorylate cFos at the serine 32 position to increase its stability and nuclear localization [43]. Unfortunately, our experimental attempts using commercial antibodies did not allow us to distinguish the phosphorylated (Ser-32) from total cFos so far and, therefore, to determine whether cFos phosphorylation was necessary for the observed effect on the virus replication.
The role of cFos on apoptosis is unclear, with both pro-and anti-apoptotic roles mainly observed in cancer models [7,9,44]. IAV was shown to protect cells from premature apoptosis [45]. However, at the late stage of infection, IAV promotes apoptosis to facilitate the spread of viral particles to neighbouring cells [45]. In our study, cFos-knockdown significantly increased apoptosis in IAV-infected A549 cells (Fig. 3). A similar increase in apoptosis was also observed in cFos-knockdown H1299 human epithelial lung cells infected with a gamma coronavirus [19]. Therefore, during viral infection, cFos seems to be involved in cell survival by inhibiting apoptosis.
Although certain transcription factors with known antiviral functions, such as STAT-1 and NF-κB p65, have been shown to facilitate IAV replication [46,47], the underlying molecular mechanisms remain poorly understood. AP-1 factors are widely recognized as transcriptional activators that promote expression of pro-inflammatory cytokines (reviewed in ref. [2]) and interferon-β and -γ [48,49]. Our results suggest that cFos may counteract cellular antiviral response by down-regulating virus-induced IFN-β expression (Fig. 4b). Activation of the IFNb1 gene transcription occurs through a signalling cascade that requires the cooperative binding of ATF2/cJun AP-1 dimer [48]. Since cFos possesses a higher affinity for cJun than ATF2 and was shown to displace the ATF2/cJun dimer [50], cFos overexpression upon IAV infection could decrease cJun availability for ATF2, leading to a reduction in IFN-β transcription. The absence of the IFN-β expression or the inhibition of IFN-β activation by the cellular exonuclease XRN1 was shown to facilitate IAV replication [51,52]. IFN-β triggers the transcriptional activation of antiviral interferon-stimulated genes (ISGs). Given that these ISGs were shown to be upregulated at the early stage (4 hpi) during IAV infection in epithelial cells [53], enhanced ISG expression resulting from increased IFN-β levels could potentially account for the impaired viral transcription of NA mRNA observed in cFos-knockdown cells during single-cycle IAV infection experiments (Fig. 5).
The impaired viral transcription of NA mRNA (Fig. 5) may also suggest a direct role of cFos in the viral transcription. IAV relies on cap-snatching to obtain host-capped RNA fragments to initiate its own transcription, a process mediated by the interaction between the viral polymerase complex and the cellular DNA-dependent RNA polymerase II (RNAP II) [54]. cFos/cJun AP-1 dimer was shown to bind enhancer sequences and then to recruit the chromatin remodelling BRG1-associated factor (BAF) complex, leading to an accessible chromatin state [55]. Interaction between the Brg1 subunit of the BAF complex and the RNAP II was further shown to participate in the formation of the transcriptional pre-initiation complex [56]. As a transcription factor, cFos may thus facilitate the accessibility to the RNAP II for the viral polymerase complex, therefore helping in the cap-snatching process. Another chromatin remodelling complex was shown to support IAV viral transcription through direct interaction with the viral polymerase complex [57]. Using a split luciferase protein complementation assay, no direct interaction between cFos and any of the non-membranous viral proteins (Fig. S5a) or the viral polymerase proteins as a complex (Fig. S5b) could be found, indicating that cFos would most likely act through interaction with one or several additional cellular partners. Both IAV viral transcription and replication take place in the nucleus, and a conformational change of the viral polymerase is required to switch from transcription to replication status [58]. Interestingly, in parallel to the decrease of NA mRNA levels in cFos-knockdown cells, an increase of NA and cRNA levels seemed to occur, suggesting that cFos could be involved in the regulation of the timing of the viral transcription/replication switch.
In conclusion, this study highlighted cFos as a pro-viral factor that facilitates IAV replication. cFos was observed to extend cell survival and down-regulate IFN-β expression during IAV infection, potentially promoting viral replication. In addition, cFos may be mechanistically involved in viral transcription. However, the exact mechanisms by which cFos supports IAV replication remain unclear. It also remains enigmatic why the decrease observed in the transcription and expression of viral proteins only affected some viral proteins, especially those that are later expressed in the virus cycle.
## References
1. Shaulian (2002) "AP-1 as a regulator of cell life and death" *Nat Cell Biol*
2. Foletta, Segal, Cohen (1998) "Transcriptional regulation in the immune system: all roads lead to AP-1" *J Leukoc Biol*
3. Madrigal, Alasoo (2018) "AP-1 takes centre stage in enhancer chromatin dynamics" *Trends Cell Biol*
4. Shaulian (2001) "AP-1 in cell proliferation and survival" *Oncogene*
5. Brüsselbach, Möhle-Steinlein, Wang et al. (1995) "Cell proliferation and cell cycle progression are not impaired in fibroblasts and ES cells lacking c-Fos" *Oncogene*
6. Brown, Nigh, Lee et al. (1998) "Fos family members induce cell cycle entry by activating cyclin D1" *Mol Cell Biol*
7. Wang, Liu, Wang et al. (2017) "Involvement of c-Fos in cell proliferation, migration, and invasion in osteosarcoma cells accompanied by altered expression of Wnt2 and Fzd9" *PLoS ONE*
8. Güller, Toualbi-Abed, Legrand et al. (2008) "c-Fos overexpression increases the proliferation of human hepatocytes by stabilizing nuclear Cyclin D1" *World J Gastroenterol*
9. Zhang, Zhang, Yang et al. (2007) "c-Fos as a proapoptotic agent in TRAIL-induced apoptosis in prostate cancer cells" *Cancer Res*
10. Preston, Lyon, Yin et al. (1996) "Induction of apoptosis by c-Fos protein" *Mol Cell Biol*
11. Hafezi, Steinbach, Marti et al. (1997) "The absence of c-fos prevents light-induced apoptotic cell death of photoreceptors in retinal degeneration in vivo" *Nat Med*
12. Zhang, Zhang, Mcquade et al. (2002) "c-fos regulates neuronal excitability and survival" *Nat Genet*
13. Ray, Kuwahara, Takada et al. (2006) "c-Fos suppresses systemic inflammatory response to endotoxin" *Int Immunol*
14. Matsumoto, Einhaus, Gold et al. (2004) "Simvastatin augments lipopolysaccharide-induced proinflammatory responses in macrophages by differential regulation of the c-Fos and c-Jun transcription factors" *J Immunol*
15. Caputto, Gizzi, Gil (2014) "c-Fos: an AP-1 transcription factor with an additional cytoplasmic, non-genomic lipid synthesis activation capacity" *Biochim Biophys Acta*
16. Colin, Vandenhoudt, De Walque et al. (2011) "The AP-1 binding sites located in the pol gene intragenic regulatory region of HIV-1 are important for viral replication" *PLoS One*
17. Li, Du, Avey et al. (2015) "ORF45-mediated prolonged c-Fos accumulation accelerates viral transcription during the late stage of lytic replication of Kaposi's sarcoma-associated herpesvirus" *J Virol*
18. Kang, Lim, Won et al. (2011) "c-Fos regulates hepatitis C virus propagation" *FEBS Lett*
19. Yuan, Liang, Zhu et al. (2021) "Gammacoronavirus avian infectious bronchitis virus and alphacoronavirus porcine epidemic diarrhea virus exploit a cell-survival strategy via upregulation of cFOS to promote viral replication" *J Virol*
20. Bouvier, Palese (2008) "The biology of influenza viruses" *Vaccine*
21. Hatada, Hasegawa, Mukaigawa et al. (1989) "Control of influenza virus gene expression: quantitative analysis of each viral RNA species in infected cells" *J Biochem*
22. Krammer, Smith, Fouchier et al. (2018) *Nat Rev Dis Primers*
23. Xie, Zhang, Hu et al. (2014) "Regulatory roles of c-jun in H5N1 influenza virus replication and host inflammation" *Biochim Biophys Acta*
24. Ludwig, Ehrhardt, Neumeier et al. (2001) "Influenza virus-induced AP-1-dependent gene expression requires activation of the JNK signaling pathway" *J Biol Chem*
25. Wan, Zhang, Ruan et al. (2022) "AP-1 signaling pathway promotes pro-IL-1β transcription to facilitate NLRP3 inflammasome activation upon influenza A virus infection" *Virulence*
26. Ludwig, Wang, Ehrhardt et al. (2002) "The influenza A virus NS1 protein inhibits activation of Jun N-terminal kinase and AP-1 transcription factors" *J Virol*
27. Matrosovich, Matrosovich, Carr et al. (2003) "Overexpression of the alpha-2,6-sialyltransferase in MDCK cells increases influenza virus sensitivity to neuraminidase inhibitors" *J Virol*
28. Neumann, Watanabe, Ito et al. (1999) "Generation of influenza A viruses entirely from cloned cDNAs" *Proc Natl Acad Sci U S A*
29. De Wit, Spronken, Bestebroer et al. (2004) "Efficient generation and growth of influenza virus A/PR/8/34 from eight cDNA fragments" *Virus Res*
30. Biquand, Poirson, Karim et al. (2017) "Comparative profiling of ubiquitin proteasome system interplay with influenza A virus PB2 polymerase protein recapitulating virus evolution in humans" *mSphere*
31. Matrosovich, Matrosovich, Garten et al. (2006) "New lowviscosity overlay medium for viral plaque assays" *Virol J*
32. Kawakami, Watanabe, Fujii et al. (2011) "Strandspecific real-time RT-PCR for distinguishing influenza vRNA, cRNA, and mRNA" *J Virol Methods*
33. Hoffmann, Palese, Shaw (2008) "Modulation of influenza virus replication by alteration of sodium ion transport and protein kinase C activity" *Antiviral Res*
34. Eisfeld, Kawakami, Watanabe et al. (2011) "RAB11A is essential for transport of the influenza virus genome to the plasma membrane" *J Virol*
35. Ishida, Ueki, Morishita et al. (2015) "T-5224, a selective inhibitor of c-Fos/activator protein-1, improves survival by inhibiting serum high mobility group box-1 in lethal lipopolysaccharide-induced acute kidney injury model" *J Intensive Care*
36. Blunsom, Gomez-Espinosa, Ashlin et al. (2019) "Sustained phospholipase C stimulation of H9c2 cardiomyoblasts by vasopressin induces an increase in CDP-diacylglycerol synthase 1 (CDS1) through protein kinase C and cFos" *Biochimica et Biophysica Acta*
37. Szalóki, Krieger, Komáromi et al. (2015) "Evidence for homodimerization of the c-Fos transcription factor in live cells revealed by fluorescence microscopy and computer modeling" *Mol Cell Biol*
38. Leech, Brennan, Don et al. (2022) "In vitro single molecule and bulk phase studies reveal the AP-1 transcription factor cFos binds to DNA without its partner cJun" *J Biol Chem*
39. Monje, Marinissen, Gutkind (2003) "Phosphorylation of the carboxyl-terminal transactivation domain of c-Fos by extracellular signal-regulated kinase mediates the transcriptional activation of AP-1 and cellular transformation induced by platelet-derived growth factor" *Mol Cell Biol*
40. Tanos, Marinissen, Leskow et al. (2005) "Phosphorylation of c-Fos by members of the p38 MAPK family. Role in the AP-1 response to UV light" *J Biol Chem*
41. Pleschka, Wolff, Ehrhardt et al. (2001) "Influenza virus propagation is impaired by inhibition of the Raf/MEK/ERK signalling cascade" *Nat Cell Biol*
42. Kujime, Hashimoto, Gon et al. (2000) "p38 mitogenactivated protein kinase and c-jun-NH2-terminal kinase regulate RANTES production by influenza virus-infected human bronchial epithelial cells" *J Immunol*
43. Sasaki, Kojima, Kishimoto et al. (2006) "Spatiotemporal regulation of c-Fos by ERK5 and the E3 ubiquitin ligase UBR1, and its biological role" *Mol Cell*
44. Kaina, Haas, Kappes (1997) "A general role for c-Fos in cellular protection against DNA-damaging carcinogens and cytostatic drugs" *Cancer Res*
45. Zhirnov, Klenk (2007) "Control of apoptosis in influenza virusinfected cells by up-regulation of Akt and p53 signaling" *Apoptosis*
46. Zhang, Huo, Fan et al. (2019) "p-STAT1 regulates the influenza A virus replication and inflammatory response in vitro and vivo" *Virology*
47. Kumar, Xin, Liang et al. (2008) "NF-kappaB signaling differentially regulates influenza virus RNA synthesis" *J Virol*
48. Panne, Maniatis, Harrison (2007) "An atomic model of the interferon-beta enhanceosome" *Cell*
49. Samten, Townsend, Weis et al. (2008) "CREB, ATF, and AP-1 transcription factors regulate IFN-gamma secretion by human T cells in response to mycobacterial antigen" *J Immunol*
50. Yuan, Gong, Luo et al. (2009) "Opposing roles for ATF2 and c-Fos in c-Jun-mediated neuronal apoptosis" *Mol Cell Biol*
51. Koerner, Kochs, Kalinke et al. (2007) "Protective role of beta interferon in host defense against influenza A virus" *J Virol*
52. Liu, Mok, Wang et al. (2021) "Cellular 5'-3' mRNA exoribonuclease XRN1 inhibits interferon beta activation and facilitates influenza A virus replication" *mBio*
53. Ramos, Smith, Ruf-Zamojski et al. (2019) "Innate immune response to influenza virus at single-cell resolution in human epithelial cells revealed paracrine induction of interferon lambda 1" *J Virol*
54. Vlugt, Sikora, Pelchat (2018) "Insight into influenza: a virus capsnatching" *Viruses*
55. Vierbuchen, Ling, Cowley et al. (2017) "AP-1 transcription factors and the BAF complex mediate signal-dependent enhancer selection" *Mol Cell*
56. Zhao, Ba, Wang et al. (2005) "BAF complex is closely related to and interacts with NF1/CTF and RNA polymerase II in gene transcriptional activation" *Acta Biochim Biophys Sin (Shanghai)*
57. Pazo, Nieto (2016) "Influenza virus and chromatin: role of the CHD1 chromatin remodeler in the virus life cycle" *J Virol*
58. Sun, Kuai, Zhang et al. (2024) "NS2 induces an influenza A RNA polymerase hexamer and acts as a transcription to replication switch" *EMBO Rep* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724265&blobtype=pdf | # LRP1 facilitates Jamestown Canyon virus infection of neurons
Zachary Frey, David Price, Kaleigh Connors, Rachael Rush, Griffin Brown, Cade Sterling, Farheen Fatma, Safder Ganaie, Xiaoxia Cui, Zachary Wills, Gaya Amarasinghe, Daisy Leung, Amy Hartman
## Abstract
Jamestown Canyon virus (JCV) is a bunyavirus and arbovirus that causes viral neuroinvasive disease in North America. JCV neuropathogenesis is understudied, and no pro-viral host factors for cellular infection have been identified. Here, we assessed the role of low-density lipoprotein receptor-related protein 1 (LRP1; also known as CD91), which has been identified as a host entry factor for other bunyaviruses, in mediating JCV infection with a focus on neurons. Both neuronal and non-neuronal immortalized cell lines deficient for murine Lrp1 displayed reduced binding, internalization, and infection with JCV. Furthermore, a soluble extracellular domain of human LRP1 can bind directly to JCV, and the same region of LRP1 can neutralize JCV infection. Primary neurons, where Lrp1 was highly expressed, were permissive for JCV infection. Treatment of primary neurons with the murine Lrp1 ligand receptor-associated protein (RAP) resulted in reduced infectivity with JCV. Finally, treatment of Lrp1 knockout cells with RAP further reduced JCV infection, suggesting that other low-density lipoprotein receptors may mediate JCV entry in the absence of Lrp1. Together, these results support LRP1 as an important cellular factor for efficient neuronal infection by JCV. Emerging support for the use of LRP1 for viral entry by multiple bunyaviruses also makes LRP1 a promising target for antiviral development.IMPORTANCE Jamestown Canyon virus (JCV), an emerging mosquito-transmitted virus in North American white-tailed deer, causes several cases of severe neurologic disease in humans each year. Our results on the use of low-density lipoprotein receptor (LDLR)-rela ted protein 1 by JCV for efficient cellular infection of neurons underscore the signifi cance of the LDLR family of receptors in viral infection. Recent studies also highlight the emerging use of the LDLR family of receptors for virus entry by the bunyavirus and alphavirus family members. Defining cellular factors that mediate infection by mosquito-transmitted viruses is critically important to the prototype pathogen approach for combating infectious diseases and countermeasure development.
potential for zoonotic spread and a high rate of hospitalization in reported human cases, there remains a major gap in understanding the mechanisms of cellular infection by JCV.
Host cell proteins within the low-density lipoprotein receptor (LDLR) family have garnered recent attention for their role in viral infections. One of the most structur ally and functionally complex members of the LDLR family is low-density lipoprotein receptor-related protein 1 (human LRP1; murine Lrp1; also known as CD91). In founda tional studies, we showed that LRP1 can act as a critical entry factor for Rift Valley fever virus (RVFV, Phenuiviridae) (15,16). Subsequent studies with severe fever with thrombocytopenia syndrome virus (SFTSV, Phenuiviridae) and Oropouche orthobunyavi rus (OROV, Peribunyaviridae) provide support for LRP1 as a pan-bunyavirus entry factor (17,18). In parallel, recent work has shown that Crimean-Congo hemorrhagic fever virus (Nairoviridae) uses the related LDLR as an entry factor (19)(20)(21). Additional RNA viruses also rely on LRP1 for later stages of infection, including the orthobunyavirus LACV (22). Together, these studies highlight the greater implications for LDLR family members as broad-spectrum pro-viral factors.
LRP1 is a large ~600 kDa transmembrane protein that contains an extracellular alpha chain with four ligand-binding clusters (CL) separated by epidermal growth factor repeats and β-propeller domains, a transmembrane domain, and a cytoplasmic tail. The ligand-binding clusters are composed of cysteine-rich complement-type repeats (CR), with CL I-IV containing 2, 8, 10, and 11 CR repeats, respectively (23). Most ligands for LRP1 bind to CL II (CL II ) and CL IV (CL IV ), including the receptor-associated protein (RAP) (24). RAP is a highaffinity molecular chaperone for both LRP1 and other members of the LDLR family that prevents premature binding of ligands until the receptor localizes to the cell membrane (25). Domain 1 and domain 3 of RAP (RAP D3 ) can bind to LRP1, and RAP D3 is sufficient to perform the chaperone duties of the full-length protein by binding to both CL II and CL IV (26). Previous studies have shown that the surface glycoproteins of OROV and RVFV bind to CL II and CL IV of LRP1, with both viruses demonstrating an apparent greater preference for CL IV . Furthermore, OROV and RVFV likely have overlapping binding sites on LRP1, as a soluble form of RVFV glycoprotein Gn is able to competitively inhibit OROV infection in vitro (15,18).
LRP1 is highly expressed in multiple brain cells, including neurons, astrocytes, and microglia, and is implicated in normal neurodevelopment and multiple neurodegenera tive diseases (27)(28)(29)(30)(31). The role of LRP1 in mediating neuronal infection by neurotropic bunyaviruses, such as JCV, remains unexplored. In this study, we explore the breadth of LRP1 usage by bunyaviruses and provide insights into the mechanism by which Peribunyaviridae use LRP1 for infection of neurons, a primary target of infection. We found that LRP1 is needed for early-stage entry of JCV into different cell types, and that blocking or neutralizing LRP1 binding sites reduces JCV infection of primary neurons. These findings highlight the role that LRP1 plays in JCV infection of neurons and further underscore LRP1 as a multi-bunyavirus host factor.
## MATERIALS AND METHODS
## Biosafety
All work with JCV, OROV, and ZIKV was performed at biosafety level 2 in accordance with the University of Pittsburgh biosafety guidelines. All work with RVFV ZH501 was performed at biosafety level 3+ in the Regional Biocontainment Laboratory (RBL) at the University of Pittsburgh by approved personnel. The University of Pittsburgh RBL is registered with the Centers for Disease Control and Prevention and the United States Department of Agriculture for work with RVFV.
## Viruses
The following viruses were obtained through the NIH Biodefense and Emerging Infections (BEI) Research Resources Repository, NIAID, NIH: Jamestown Canyon virus (strain 61V-2235; NR-536), Zika virus (strains PRVABC59; NR-50684 and MR766; NR-50065). Oropouche virus (strain BeAn19991) was rescued from reverse genetics as previously described (32) and was kindly provided by Paul Duprex and Natasha Tilston-Lunel (University of Pittsburgh Center for Vaccine Research). The ZH501 strain of RVFV was kindly provided by Barry Miller (CDC, Fort Collins, Colorado) and Stuart Nichol (CDC, Atlanta, Georgia) and was rescued through reverse genetics as previously described (33). Viruses were propagated in Vero E6 cells with standard culture conditions using standard D2 media comprised of Dulbecco's modified Eagle's medium (DMEM) supplemented with 1% penicillin/streptomycin (Pen/Strep), 1% L-glutamine (L-Glut), and 2% fetal bovine serum (FBS). A standard viral plaque assay (VPA) was used to determine the infectious titer of the stocks. The agar overlay for the VPA was comprised of 1× minimal essential medium, 2% FBS, 1% Pen/Strep, 1% HEPES buffer, and 0.8% SeaKem agarose (Fisher, BMA5000); the assay was incubated at 37°C for 3 (JCV, OROV, RVFV) or 5 (ZIKV) days, followed by visualization of plaques with 0.1% crystal violet. Viral passage 1 or 2 was used for the enclosed experiments.
## Cell lines
Vero E6 (ATCC, CRL-1586) and BV2 cells were cultured in DMEM (ATCC, and supplemented with 1% Pen/Strep, 1% L-Glut, and 10% FBS. N2a cells were maintained in Eagle's Minimum Essential Medium (ATCC, supplemented with 1% Pen/Strep, 1% L-Glut, and 10% FBS. BV2 and N2a LRP1 knockout (KO) cell lines were generated and validated as previously described (15) and maintained in the same culture media as their wild-type (WT) counterparts.
## Lrp1-deficient cell line infections
N2a and BV2 cell lines deficient for murine Lrp1 were previously described and validated (15). KO cells and their wild-type counterparts were seeded into 24-well plates at 1-2 × 10 5 cells/well. On the day of infection, media were removed from each well and replaced with 100 µL of virus diluted to an MOI of 0.1 in standard D2 media. The virus was incubated at 37°C for an hour, rocking every 15 min to ensure the monolayer did not dry out. Following the 1 h adsorption period, the inoculum was removed, and the cells were washed once with 1× PBS. Fresh D2 media were added, and the cells were incubated for 24 h prior to supernatant collection for measurement of viral RNA (vRNA) or infectious titers.
## Binding and internalization assays
Lrp1 KO or WT cells were seeded in 24-well plates at 1 × 10 5 cells/well 1 day prior to infection. On the day of infection, media were removed and replaced with 200 µL of 10 µM surfen (34) or vehicle control (DMSO) in PBS. Cells were incubated for 30 min at 4°C. Following the incubation, surfen solution was removed and replaced with 200 µL of virus diluted to an MOI of 0.1 in standard D2 media. Plates were returned to 4°C for an hour. The inoculum was removed, and cells were washed five times with PBS containing 3% bovine serum albumin (BSA, Sigma, A3294) and 0.02% Tween-20. Binding samples were collected by adding 1 mL of Trizol (Fisher, 15-596-018) directly to the cell monolayer. For internalization assays, wells not collected for binding were incubated for 1 h in fresh D2 media at 37°C. Cells were washed once with the same wash buffer containing BSA + Tween-20, and samples were collected by adding 1 mL of Trizol directly to the cell monolayer.
## Immunofluorescence
Coverslips were fixed and virus inactivated in 4% paraformaldehyde for 15 min prior to storage in 1× PBS at 4°C prior to staining. Cells were permeabilized with 0.1% Triton X-100 diluted in 1× PBS for 10 min at room temperature. After permeabilization, coverslips were blocked in 5% normal goat serum (Thermo Fisher, 50062Z) for an hour at room temperature. Coverslips were incubated for 1 to 2 h at room temperature with primary antibodies. Samples were then incubated for an hour with secondary antibodies conjugated to a fluorophore. Coverslips were counterstained with Hoechst 33258 (Invitrogen, #H1398, 1:1000) and mounted on slides using Gelvatol (provided by the Center for Biologic Imaging). Fluorescent slides were imaged on either a Nikon A1 confocal microscope at the Center for Biologic Imaging, or a Leica DMI8 inverted fluorescent microscope at the Center for Vaccine Research. Images were processed using Fiji (v. 1.53). The following antibodies were used for immunofluorescent staining during this study: mouse anti-βIII-tubulin (1:500; R&D Systems, MAB1195), custom rabbit anti-JCV-nucleoprotein (N) (1:500; Genscript), rabbit anti-LRP1 (1:500; Abcam, ab92544), antisera from mice immunized with a sublethal dose of JCV (1:200; generated in house), custom rabbit anti-OROV N (1:500; Genscript), mouse anti-ZIKV NS1 (1:500; Invitrogen, MA5-24585), goat anti-rabbit 488 (1:500; Invitrogen, A11008), goat anti-mouse 488 (1:500; Invitrogen, A11001), goat anti-rabbit 594 (1:500; Invitrogen, A11012), and goat anti-mouse 594 (1:500; Invitrogen, A11005).
## Quantification of viral RNA
RNA isolation was performed using an Invitrogen PureLink RNA/DNA kit (Fisher, 12-183-025) with a modified protocol as previously described (35). Briefly, supernatant was lysed in Trizol (Invitrogen, 15596026) at a dilution of 1:10 (100 µL sample, 900 µL Trizol). Then, 200 µL of chloroform was added to each sample, mixed, and then centrifuged at 12,000 × g for 15 min at 4°C to separate the aqueous and organic phases. The aqueous phase was removed and added to an equivalent volume of 70% ethanol. The PureLink RNA kit protocol was then followed for the remainder of the isolation, and RNA was eluted in 40 µL of RNase-free water. RT-qPCR was performed using the SuperScript III Platinum One-Step RT-qPCR Kit (Thermo Fisher, 11745-500), following a previously described protocol (35).
Primers targeting the JCV L-segment include 5′-CCTAGATGCTCCGTTGTCTATG-3′ (Jamestown-2364For) and 5′-TGCATTATTGGTGTGTGTTTGT-3′ (Jamestown-2448Rev). The TaqMan probe used includes (Jamestown-2387 Probe 5′ 6-FAM/TCAGTACAGTGGGATTAG AAGCTGGGA/BHQ_1 3′).
Primers targeting ZIKV PRVABC59 NS2B region include 5′-CTGTGGCATGAA CCCAATAG-3′ (ZIKVPRABC59-4513For) and 5′-ATCCCATAGAGCACCACTCC-3′ (ZIKV PRABC59-4603Rev). The TaqMan probe includes (ZIKVPRABC59-4558Probe 5′ 6-FAM/C CTTTGCAGCTGGAGCGTGG /BHQ_1 3′).
Primers targeting the OROV S-segment include 5′-TACCCAGATGCGATCACCAA-3′ (OROV19991_For) and 5′-TTGCGTCACCATCATTCCAA-3′ (OROV19991_Rev). The TaqMan probe includes (OROV19991_Probe 5′ 6-FAM/TGCCTTTGGCTGAGGTAAAGGGCTG /BHQ_1 3′).
Primers targeting the RVFV L-segment include 5′-TGAAAATTCCTGAGACACATGG-3′ (RVFV-2912Fwd) and 5′-ACTTCCTTGCATCATCTGATG-3′ (RVFV-2981Rev). The TaqMan probe includes (RVFV-2950-Probe 5′ 6-FAM/CAATGTAAGGGGCCTGTGTGGACTTGTG / BHQ_1 3′).
## Neutralization with LRP1-Fc proteins
Vero E6 cells were seeded into 96-well plates at 2 × 10 4 cells/well and allowed to incubate overnight. The day of infection, human LRP1 Cluster II-Fc (R&D Systems, 2368-L2), LRP1 Cluster IV-Fc (R&D Systems, 5395-L4B), and human IgG Fc control (R&D Systems, 110-HG) were diluted to 20 µg/mL in DMEM containing 1% Pen/Strep and 1% L-Glut and serially diluted 1:2. An equivalent volume of media containing 100-200 FFU of virus was added, and the virus mixture was incubated at 37°C for 1 h. Following the incubation, the media were removed from the cells, and 100 µL of the virus-protein mixture was added to the cells for 1 h at 37°C. Virus inoculum was removed, and an overlay comprised of 1.5% carboxymethylcellulose (CMC, Sigma, C4888), 0.5% Pen/Strep, and 5% FBS was added to the cells. The assay was incubated at 37°C for 18 h (JCV and OROV) or 42 h (ZIKV). CMC was removed, and cells were fixed for 15 min at room temperature with 4% PFA. Cells were stained according to the above immunofluorescence protocol with an antibody against viral nucleoprotein (JCV and OROV) or NS1 (ZIKV) and scanned using the Biotek Cytation 5 (Agilent). Foci were quantified using ImageJ.
## Biolayer interferometry (BLI) experiments
Vero E6 cells were infected with JCV, and the supernatant was harvested by centrifuga tion 2 days post-infection. The titer was estimated to be 1 × 10 6 PFU/mL. One milliliter of the supernatant was placed in a 24-well plate and subjected to UV irradiation for 45 min for virus inactivation. Virus inactivation was previously validated through evaluation of post-inactivated virus growth under standard culture conditions. The supernatant was concentrated (12,000 × g, 10 min) to a final volume of 25 µL prior to dilution with PBS (pH = 7.4) supplemented with 1 mg/mL BSA and 0.05% Tween-20.
BLI assays were conducted at 30°C at 1,000 rpm (Octet Red, ForteBio). Anti-Human IgG Fc Capture biosensors were hydrated in PBS (pH = 7.4) supplemented with 1 mg/mL BSA and 0.05% Tween-20 for 15 min. Recombinant human LRP1 CL IV -FcHis, recombinant human LRP1 CL II -FcHis, or recombinant human IgG1 Fc (R&D Systems, #110-HG-100) were loaded at 100 nM for 500 seconds prior to baseline equilibration for 300 seconds. Association and dissociation of JCV were measured for 600 seconds. Data were baseline subtracted using sensors in buffer alone. Experiments were done in triplicate.
## Animal work
Timed-pregnant Long Evans (Crl:LE) rats were purchased from Charles River Laboratories (Wilmington, MA, USA). Fetuses obtained from embryonic day 18 dams were euthanized to obtain the neurons used in this study.
## Primary neuron culture
On the day prior to neuron isolation, acid-washed coverslips were coated with PDL/ Laminin (Sigma, P7405-5MG; Invitrogen, 23017-015). Dissociation media (DM), comprised of Hanks' Balanced Salt Solution (Invitrogen, 14175-103) supplemented with 10 mM anhydrous MgCl 2 (Sigma, M8266), 10 mM HEPES (Sigma, H3375), and 1 mM kynurenic acid, was prepared. DM was brought to a pH of 7.2 and sterile filtered prior to use. On the day of isolation, a trypsin solution containing a few crystals of cysteine (Sigma, C7352), 10 mL of DM, 4 µL 1N NaOH, and 200 units of papain (Worthington, LS003126) and a trypsin inhibitor solution containing 25 mL DM, 0.25 g trypsin inhibitor (Fisher, NC9931428), and 10 µL 1N NaOH were prepared and filter sterilized. At embryonic day 18, dams were euthanized via CO 2 inhalation overdose (primary method) followed by cervical dislocation (secondary method), and the brains of the embryos were removed and dissected. The cortices were separated from the hippocampus and placed into DM. Five milliliters of trypsin solution was added, and cortices were placed in a 37°C water bath for 4 min, swirling occasionally to mix. The trypsin solution was removed, and cortices were immediately washed with trypsin inhibitor once, and then twice more while swirling in the water bath. Following the washes, the trypsin inhibitor was removed and replaced with 5 mL of Neurobasal/B27 media, then triturated to dissociate the neurons. The final volume was brought to 10 mL of Neurobasal/B27, and cells were counted and plated at a density of 1.5 × 10 5 neurons/well for 24-well plates, or 2.5-3 × 10 5 neurons/well for 12-well plates. One hour after isolation, the media were removed and replaced with fresh Neurobasal/B27 media. Primary neuron cultures were maintained in Neurobasal/B27 media, which consists of standard Neurobasal Plus Medium (Thermo Fisher, A3582901) supplemented with 1% Pen/Strep, 1% L-Glut, and 2% B27 Plus Supplement.
## Viral growth curve infection
Primary rat neurons were maintained in culture for 3 days following isolation. Infection occurred on day 4 in vitro. JCV or ZIKV was thawed and diluted in D2 media to the desired MOI. Media were removed from wells, and 100 µL of inoculum was added to each well. Cells were incubated at 37°C for an hour, rocking every 15 min to prevent the monolayer from drying out. Following the adsorption period, the inoculum was removed from the wells and replaced with Neurobasal/B27 media. Cells were incubated for 15 min, and 100 µL of supernatant was inactivated in 900 µL of Trizol Reagent (Invitrogen, 15596026) to measure 0 h post-infection (hpi) viral RNA levels. At the appropriate time points, 100 µL of supernatant was inactivated in 900 µL of Trizol, the remaining supernatant was collected and stored at -80°C, and plates were fixed with 4% PFA for 15 min and stored at 4°C in 1× PBS for immunofluorescent staining.
## Western blot
Cells were inactivated in 100 µL of radioimmunoprecipitation assay buffer (Thermo Fisher Scientific, 89901) with 1% Halt Protease Inhibitor (Thermo Fisher Scientific, 78429) for 10 min at room temperature. Samples were centrifuged at 13,500 relative centrifugal force for 20 min. Cellular debris was removed, and a bicinchoninic acid (BCA) assay was completed following the manufacturer's instructions (Thermo Fisher Scientific, Pierce BCA Protein Assay, 23227). Five micrograms of protein from each sample was loaded into a NuPAGE 4 to 12% Bis-Tris gel (Invitrogen, NP0323BOX) and run for 35 min at 165 V. The protein was transferred to a nitrocellulose membrane (LI-COR, 926-31090) using an iBlot 2 system (Invitrogen, IB21001). Membranes were blocked for 1 h, rocking at room temperature, in Intercept (PBS) Blocking Buffer (LI-COR, 927-70001). Following the block, membranes were incubated overnight rocking at 4°C with primary antibodies diluted in Intercept T20 (PBS) Antibody Diluent (LI-COR, 927-75001). The following primary antibodies were used in this study: mouse anti-GAPDH (1:1,000; Invitrogen, MA1-16757), rabbit anti-LRP1 (1:500; Cell Signaling, 64099S), custom rabbit anti-JCV-N (1:500; Genscript, Y743THG190-16), mouse anti-βIII-tubulin (1:500; R&D Systems, MAB1195), anti-RVFV Gn Clone 4D4 (1:500; BEI Resources, NR-43190), and mouse anti-β-actin (1:500; Santa Cruz Biotechnology, sc-47778). The following day, the membranes were washed by rocking in 10 mL of PBS-T three times for 5 min each. Membranes were probed for 1 h, rocking at room temperature, with either goat anti-rabbit IRDye 800CW (1:10,000; LI-COR, 926-32211), goat anti-rabbit IRDye 680RD (1:10,000; LI-COR, 925-68071), goat anti-mouse IRDye 800CW (1:10,000; LI-COR, 925-32210), or goat anti-mouse IRDye 680RD (1:10,000; LI-COR, 926-68070) diluted in Intercept T20 (PBS) Antibody Diluent (LI-COR, 927-75001). The membranes were washed by rocking in 10 mL of PBS-T three times for 5 min each, then rinsed with 1× PBS. The membrane was visualized using an Odyssey CLx Imager (LiCor, Lincoln, Nebraska, USA).
## Viral plaque assay
Vero E6 cells were plated into 12-well plates and allowed to incubate overnight until near confluency. Samples were serially diluted in D2 media. The inoculum was allowed to incubate for 1 h at 37°C and then removed. Agar overlay composed of 1× minimal essential medium, 2% FBS, 1% Pen/Strep, 1% HEPES buffer, and 0.8% SeaKem agarose (BMA5000) was added to each well. The assay was incubated at 37°C for 3 (JCV) or 5 (ZIKV) days to allow for the formation of plaques, fixed with 37% formaldehyde for at least 3 h, and then stained with 0.1% crystal violet for visualization and counting of plaques.
## Recombinant protein expression and purification
Murine mRAP D3 or mRAP D3 (K265A/K279E) expression plasmids were transformed into BL21(DE3) E. coli cells (Novagen). Colonies were cultured in Luria Broth media at 37°C to an OD 600 of 0.6 and induced with 0.5 mM isopropyl-β-D-thiogalactoside for 14 h at 18°C. Cells were harvested and resuspended in lysis buffer containing 25 mM sodium phosphate (pH 7.5), 500 mM NaCl, 20 mM imidazole, 5 mM 2-mercaptoethanol, and were lysed using an EmulsiFlex-C5 homogenizer (Avestin). Lysates were clarified by centrifugation at 24,000 × g at 4°C for 40 min. Proteins were purified using a series of chromatographic columns as described previously (15). Protein purity was determined by Coomassie staining of SDS-PAGE. Soluble RVFV Gn was expressed in the same manner as mRAP D3 and resuspended in a lysis buffer containing 20 mM Tris-HCl (pH 8.0), 500 mM NaCl, and 5 mM 2-mercaptoethanol. Following lysis, the Gn pellet was resuspended in 20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 5 mM imidazole, 8 M urea, and 1 mM 2-mercap toethanol. RVFV Gn was refolded on a NiFF (GE Healthcare) column using a reverse linear urea gradient and eluted with imidazole. Gn was further purified using a size-exclusion column (SD200 10/300L, GE Healthcare).
## Competitive inhibition assays with mRAP D3 or RVFV Gn
Primary rat neurons were isolated as described above and maintained in culture for 3 days. Treatment and infection occurred on day 4 in vitro. Proteins were diluted to the desired concentration in both D2 and Neurobasal media. Culture media were partially removed and replaced with Neurobasal containing mRAP D3 or RVFV Gn. Plates were allowed to incubate for 45 min at 37°C. Following pre-treatment, all culture media were removed and replaced with D2 containing viral inoculum and either mRAP D3 or RVFV Gn. Plates were incubated for an hour, with rocking every 15 min. The inoculum was removed following the adsorption period, and Neurobasal containing mRAP D3 or RVFV Gn was added to the wells, and cells were returned to the incubator. Twenty-four hours (JCV) or 48 h (ZIKV) later, supernatant was collected, and plates were fixed with 4% PFA for 15 min, or cells were lysed with RIPA buffer for 10 min. Viral titers were then ana lyzed by RT-qPCR or VPA, and viral antigen was visualized through immunofluorescence staining or Western blot.
## Statistics and data analysis
Statistical analysis was performed using GraphPad Prism version 8.0. Significance was determined by one-way or two-way ANOVA. Error bars show mean and standard deviation. Significance is indicated by *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant.
## RESULTS
## Reduced infectivity of JCV in cells lacking Lrp1
Using murine BV2 (microglia) and N2a (neuroblastoma) clonal cell lines with Lrp1 deletion (KO) (15,18), the Lrp1deficient cells and their Lrp1sufficient counterparts were infected with JCV (MOI = 0.1), and the amount of viral RNA in the supernatant at 24 hpi was measured by RT-qPCR (Fig. 1A). There was a significant reduction in viral RNA production in the absence of Lrp1 in both cell types. We observed no reduction in infection when the KO N2a cells were infected with Zika virus (ZIKV; PRVABC59) as a control (Fig. 1B). BV2 cells do not support productive ZIKV infection and thus were not tested (18). The reduction in infection by JCV was visible by immunofluorescence microscopy, where both N2a and BV2 Lrp1 KO cells displayed decreased staining for JCV-N at 24 hpi compared to the WT cells (Fig. 1C andD).
## Reduced binding and internalization of JCV in Lrp1 KO cells
Because LRP1 functions as an entry factor early in the infection process for both RVFV and OROV, we performed binding and internalization assays using BV2 cells to investi gate if LRP1 is involved in the early stages of JCV infection. BV2 WT or BV2 Lrp1 KO cells were first treated with the glycosaminoglycan antagonist surfen to prevent any nonspecific binding to proteoglycans (34), incubated with JCV (MOI = 0.1) for 1 h at 4°C to allow binding but not internalization, and washed extensively before collection and RNA quantification by RT-qPCR. For internalization assays, cells were incubated at 37°C for another 1 h after washing. We observed a 50%-60% reduction in both binding and internalization in BV2 Lrp1 KO cells when compared to the WT cells (Fig. 1E). In contrast, we observed no reduction in binding and only a slight decrease in internalization of ZIKV in the KO cells (Fig. 1F). Reductions in binding and internalization were still evident in control experiments repeated without surfen, albeit to a lesser degree (Fig. S1).
## LRP1 CL IV receptor decoy binds and neutralizes JCV
Our data point to a role for LRP1 in the entry stage of JCV infection; therefore, we assessed the ability of soluble fragments of the human LRP1 ectodomain ("receptor decoys") to neutralize JCV infection in a focus reduction neutralization test (FRNT). Due to the large size of the entire LRP1 ectodomain, we used individual cluster domains fused to human IgG1 Fc, whereby exogenous addition of these ectodomain proteins was able to neutralize RVFV and OROV infection (15,18). We pre-incubated human LRP1 CL II -Fc, CL IV -Fc, or an Fc control with 100-200 FFU of each indicated virus for 1 h prior to infection of Vero E6 cells. We used OROV and ZIKV as positive and negative controls for the assay, respectively. At 18 hpi (JCV and OROV) or 42 hpi (ZIKV), cells were fixed and stained for viral antigen to enumerate foci and compared the number of foci to control (untreated) wells. LRP1 CL IV -Fc neutralized JCV, but CL II displayed little to no neutralization at the highest dose tested, suggesting better binding and/or a higher affinity of JCV for CL IV of LRP1. We observed no reduction in JCV infection after treatment with the Fc control protein (Fig. 2A). OROV, our positive control in this assay, was neutralized by both CL II and CL IV , confirming our previous findings (Fig. 2B) (18). As a negative control, we observed no reduction in infection when the assay was repeated with ZIKV (Fig. 2C). Furthermore, and consistent with our neutralization data, JCV showed increased binding to LRP1 CL IV -Fc compared to LRP1 CL II -Fc or an IgG-Fc control using biolayer interferometry (Fig. 2D). These findings suggest that soluble LRP1 CL IV can directly neutralize JCV and reduce entry into cells, and that JCV binds directly to this cluster of LRP1.
## Primary neurons are permissive to JCV infection and express Lrp1
Among humans who develop reportable clinical disease due to JCV infection, neurologi cal issues are a primary manifestation. Despite this, few experimental studies have been done to investigate the mechanism of neuronal infection by JCV. Here, we isolated primary cortical neurons from rat embryos and generated JCV growth curves by infecting neurons at MOIs of 0.1, 0.01, and 0.001. Supernatants were analyzed for viral RNA (RT-qPCR) and infectious titers (VPA). Primary neuron cultures were highly permis sive to JCV in a dose-dependent manner, generating 10 5 -10 6 PFU/mL of vRNA and infectious particles over time (Fig. 3A andB). These peak titers were similar to a previous study examining JCV replication in human NSCs and SH-SY5Y cells (36). Mock-infected cultures appeared healthy, containing neurons with long cellular processes staining prominently with βIII-tubulin (Fig. 3C). At an MOI of 0.1, JCV antigen staining was widespread by 24 hpi and remained prevalent at 48 hpi. While we did not directly examine cell death, as the infection progressed, the cellular debris in culture increased, resulting in a punctate βIII-tubulin staining pattern, indicating loss of neuronal structure by 60 hpi (Fig. 3C; Fig. S2A). Under the culture conditions used here, neurons expressed Lrp1 throughout the culture period (4 to 7 days in culture) (Fig. 3D; Fig. S3). Lrp1 expression was widely detectable by microscopy and was found in both the processes and cell bodies.
## Treatment of primary neurons with a high-affinity Lrp1 binding protein reduces JCV infection
RAP is an intracellular highaffinity LRP1 chaperone protein known to competitively inhibit ligand binding to the CL II and CL IV domains of LRP1 (24). Domain 3 of mouse RAP protein (mRAP D3 ) can be added exogenously to cells prior to infection to interrogate reliance on LRP1 for infection, as we previously demonstrated with RVFV and OROV (15,18). Here, primary neurons were pre-treated with recombinant mRAP D3 or a mutated version of mRAP D3 containing K265A/K279E mutations (A/E mutant), which reduces affinity for Lrp1 (15, 37), followed by infection with JCV (MOI = 0.1). At 24 hpi, viral RNA levels in the supernatant were reduced approximately 75%-90% in a dose-dependent manner compared to the infected untreated controls (Fig. 4A). The mutant mRAP D3 , in comparison, was not as effective at reducing JCV viral RNA and decreased RNA titers only at the highest dose tested (10 µg/mL) (Fig. 4A). Plaque assays measuring infectious titer at 24 hpi showed a similar reduction to viral RNA after mRAP D3 treatment (Fig. S4A). By microscopy, viral antigen in mRAP D3 -treated cells was restricted to small foci as opposed to being widespread throughout the culture in the untreated control images (Fig. 4C; Fig. S4B). When the assay was repeated with ZIKV, we found only a mild reduction in viral production at the highest dose with the WT mRAP D3 , and no reduction in infection regardless of dose with the A/E mutant (Fig. 4B).
## Exogenous Gn protein from RVFV restricts JCV infection of primary neurons
The Gn glycoprotein of the distantly related bunyavirus RVFV binds to CL II and CL IV of LRP1, and exogenous treatment of cells with recombinant RVFV Gn competitively inhibited both homologous infection with RVFV and heterologous infection by OROV (15,18). In a heterologous competition experiment to further probe the role of Lrp1 in JCV infection, primary neurons were pre-treated with recombinant RVFV Gn followed by infection with JCV. At 24 hpi, JCV titers were significantly reduced in the presence of 5 and 10 µg/mL of exogenous RVFV Gn (Fig. 5A). By Western blot, the amount of JCV-N protein detected in culture lysates decreased as increasing levels of RVFV Gn were added (Fig. 5D). Immunofluorescence microscopy revealed a decrease in viral antigen staining in cells treated with RVFV Gn compared to untreated cells (Fig. 5C; Fig. S2B). We observed no reduction in infection in the ZIKV controls (Fig. 5B). Our results indicating that RVFV Gn can competitively inhibit and reduce JCV infection suggest that JCV likely binds sites on Lrp1 CL II and CL IV that overlap with sites of RVFV Gn binding.
## Specificity of LRP1 for JCV entry
As we were unable to completely prevent infection by blocking or removing Lrp1, we investigated the role of other LDLR receptors in JCV infection. Multiple neurotropic alphaviruses can use more than one LDLR as a receptor (38)(39)(40), so we hypothesized that JCV may use another member of the LDLR family in addition to LRP1. One way to interrogate this is to treat Lrp1 KO cells with the highaffinity chaperone mRAP D3 , which can bind several LDLRs (41). We pre-treated wild-type and Lrp1 knockout BV2 cells with 10 µg/mL of mRAP D3 . When Lrp1 KO BV2 cells were infected with JCV, OROV, or the virulent RVFV strain ZH501, we observed a reduction in infection when compared to the WT cells, confirming our previous findings (15,18) (Fig. 6A). When we pre-treated the WT cells with mRAP D3 followed by infection with each of these viruses, we again observed the expected decrease in titers across all three viruses (Fig. 6B). However, when the Lrp1 KO cells were pre-treated with mRAP D3 , we observed a significant decrease in JCV infection, whereas much less or no decrease was seen with OROV or RVFV, respectively (Fig. 6C). These findings suggest that another LDLR may be involved in JCV infection in the absence of Lrp1, while infection by RVFV and OROV is likely primarily mediated by Lrp1.
## DISCUSSION
JCV is an arbovirus found in white-tailed deer and mosquitoes in North America. While severe disease in people is rare compared to overall seropositivity rates, the potential for further spread given deer-human proximity and the capacity to induce severe neurolog ical clinical outcomes makes JCV an arbovirus of concern for the USA and Canada (4-7). At present, there is a significant knowledge gap in our understanding of the host factors that modulate infection of neurons. Immunocompetent mice have been used to study JCV neuropathogenesis; however, lack of neuroinvasion makes studying virus-cell interactions in the brain challenging (42). Intranasal and intracranial inoculation of JCV results in consistent neurologic disease in mice (36,43), but this does not mimic a natural infection route, as JCV is primarily spread by mosquitoes. Mice deficient in type I interferon receptors or key signaling molecules (IRF3, IRF7, or MAVS) develop neurologic disease following intraperitoneal infection, suggesting that innate immunity is likely responsible for controlling JCV in the periphery and preventing neuroinvasion (44). In addition to our lack of understanding of antiviral host factors that restrict JCV in vivo, we also lack clarity regarding pro-viral cellular factors that mediate JCV infection at a cellular level.
The LDLR family of cell surface receptors is an evolutionarily conserved family of proteins with a variety of functions, including lipoprotein metabolism and cellular signaling (45). LDLRs have been implicated in mediating cellular entry of a variety of arboviruses, including multiple alphaviruses and bunyaviruses (15, 17-22, 38-40, 46). Many of these viruses have a wide host range and tissue tropism, which is supported by the evolutionary conservation and broad tissue distribution of the LDLR family members. LRP1 differs from other LDLRs that serve as viral receptors, such as LDLR, VLDLR, and ApoER2, in that it contains four ligand-binding cluster domains, while the other smaller family members are comprised of just one (47). This enables LRP1 to interact with ligands through multiple clusters, differentiating its interactions with ligands from the smaller members of the LDLR family (48). RVFV and OROV infections are supported by binding to CL II and CL IV (15,18), and it is possible that both clusters interact with the multimeric viral glycoproteins during the course of attachment and entry through the cell membrane. But the specific steps in this process or the mechanism are incompletely described at present. The multimeric viral glycoprotein and the multitude of host factors involved in viral entry complicate our ability to define the exact molecular interactions between viruses and LRP1. Recently, SFTSV was shown to use LRP1 as an entry receptor in addition to the previously defined receptors CCR2 and NMMHC-IIA (17), highlighting the fact that arboviruses likely use multiple receptors or mechanisms to access cells. Interestingly, SFTSV binds to CL I and CL II of LRP1, further distinguishing this interaction on LRP1 from that of RVFV or OROV, which primarily use LRP1 CL II and CL IV .
Neurons and other cells of the CNS express LRP1 (27), and LRP1 has a variety of critical functions in the brain, including the modulation of NMDA receptor signaling (49), neuronal glucose metabolism (50), and AMPA receptor stability (51). LRP1 has also been implicated in multiple neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, and Lewy body dementia (28)(29)(30)(31). Other LDLRs also play impor tant and often overlapping roles in the CNS. VLDLR and ApoER2 have been found to modulate synaptic plasticity (52) and neuronal migration (53). Mice with LRP1 deleted on a majority of their neurons (Lrp1 f/f Synapsin-Cre) display deficits in motor function (54), and VLDLR and ApoER2 double knockout mice display progressive hind limb paralysis and smaller brain size when compared to WT mice (53), demonstrating the importance of LDLR family members in the CNS function.
Given the conserved use of LRP1 by distantly related bunyaviruses RVFV, SFTSV, and OROV, we interrogated the dependence on LRP1 for infection by JCV. Our data shown here provide evidence that LRP1 is needed for efficient early-stage cellular infection by JCV. Furthermore, given the expression level and functionality of LRP1 in neurons, we determined its role in mediating JCV infection in primary neurons using an ex vivo primary rat neuron model in combination with the previously described molecular and biochemical tools (15,18). Pre-treatment of primary neurons with two competitive inhibitors (mRAP D3 or heterologous recombinant Gn from RVFV) reduced JCV infection of primary rat neurons. The fact that RVFV Gn can inhibit JCV infection implies that these viruses may use overlapping regions on LRP1. Finally, we revealed evidence that JCV may use another unidentified member of the LDLR family, as mRAP D3 treatment reduced JCV infection in cells lacking Lrp1. Notably, we did not observe any further reduction in infection when this assay was repeated with RVFV, suggesting that Lrp1 is likely the primary member of the LDLR family involved in RVFV entry. Understanding which other LDLR family members can act as a receptor for JCV will be the focus of future studies.
Future studies will focus on the molecular mechanisms by which JCV engages with LRP1. While RVFV binds to LRP1 through interactions with the surface glycoprotein Gn (15), there are large differences in the glycoprotein structures of viruses within Bunyavirales (55). Additionally, Crimean-Congo hemorrhagic fever virus, a more distantly related bunyavirus in Nairoviridae, interacts with LDLR through the Gc glycoprotein (20,21). Therefore, JCV may engage LRP1 in a different manner than RVFV does, including potential binding by Gc rather than Gn. Additional studies are warranted to further clarify the mechanism of JCV-LRP1 binding.
In summary, we present evidence that LRP1 is a host factor involved in the early stages of cellular infection by JCV. The combination of this work with previously published results by our groups and others on RVFV, OROV, and SFTSV underscores the importance of LRP1 as a multi-bunyaviral host factor. The fact that LRP1 is highly conserved and is utilized in early infection of diverse bunyaviruses makes it an attractive target for the development of broad bunyavirus therapeutics.
## References
1. Buhler, Dibernardo, Pilfold et al. (2023) "Widespread exposure to mosquitoborne California serogroup viruses in caribou, arctic fox, red fox, and polar bears" *Canada. Emerg Infect Dis*
2. Boromisa, Grimstad (1986) "Virus-vector-host relationships of Aedes stimulans and Jamestown Canyon virus in a northern Indiana enzootic focus" *Am J Trop Med Hyg*
3. Campbell, Huang (1999) "Sequence comparisons of medium RNA segment among 15 California serogroup viruses" *Virus Res*
4. Grimstad, Calisher, Harroff et al. (1986) "Jamestown Canyon virus (California serogroup) is the etiologic agent of widespread infection in Michigan humans" *Am J Trop Med Hyg*
5. Patriquin, Drebot, Cole et al. (2018) "High seroprevalence of Jamestown Canyon virus among deer and humans" *Canada. Emerg Infect Dis*
6. Dupuis, Prusinski, Russell et al. (2020) "Serologic survey of mosquito-borne viruses in hunterharvested white-tailed deer (odocoileus virginianus)" *Am J Trop Med Hyg*
7. Rocheleau, Michel, Lindsay et al. (2018) "Risk factors associated with seropositivity to California serogroup viruses in humans and pet dogs" *Epidemiol Infect*
8. Sutter, Calvert, Grimm et al. (2025) "Jamestown Canyon virus seroprevalence in endemic regions and implications for diagnostic testing" *Clin Infect Dis*
9. Woods, Johnson, Evans et al. (2025) "Neutralizing antibodies against California serogroup orthobunyaviruses in human serum Samples" *Emerg Infect Dis*
10. Grimstad, Shabino, Calisher et al. (1982) "A case of encephalitis in a human associated with a serologic rise to Jamestown Canyon virus" *Am J Trop Med Hyg*
11. Srihongse, Grayson, Deibel (1984) "California serogroup viruses in New York State: the role of subtypes in human infections" *Am J Trop Med Hyg*
12. Dumic, Nordstrom, Patel et al. (2025) "Clinical characteristics and outcome of Jamestown Canyon virus Infection in hospitalized Wisconsin patients" *Open Forum Infect Dis*
13. Mitchell, Schuh, Okorie (2025) "Optic nerve edema and retrobulbar optic neuritis cases associated with Jamestown Canyon virus in Wisconsin and Michigan" *Am J Trop Med Hyg*
14. Fagre, Lyons, Staples et al. (2023) "West Nile Virus and other nationally notifiable arboviral diseases -United States, 2021" *MMWR Morb Mortal Wkly Rep*
15. Ganaie, Schwarz, Mcmillen et al. (2021) "Lrp1 is a host entry factor for Rift Valley fever virus" *Cell*
16. Schwarz, Ganaie, Feng et al. (2023) "Lrp1 is essential for lethal Rift Valley fever hepatic disease in mice" *Sci Adv*
17. Xing, Zhang, Xu et al. (2025) "Genome-wide CRISPR screening identifies LRP1 as an entry factor for SFTSV" *Nat Commun*
18. Schwarz, Price, Ganaie et al. (2022) "Oropouche orthobunyavirus infection is mediated by the cellular host factor Lrp1" *Proc Natl Acad Sci*
19. Ritter, Canus, Gautam et al. (2024) "The lowdensity lipoprotein receptor and apolipoprotein E associated with CCHFV particles mediate CCHFV entry into cells" *Nat Commun*
20. Xu, Du, Wang et al. (2024) "LDLR is an entry receptor for Crimean-Congo hemorrhagic fever virus" *Cell Res*
21. Monteil, Wright, Dyczynski et al. (2024) "Crimean-Congo haemorrhagic fever virus uses LDLR to bind and enter host cells" *Nat Microbiol*
22. Devignot, Sha, Burkard et al. (2023) "Low-density lipoprotein receptorrelated protein 1 (LRP1) as an auxiliary host factor for RNA viruses" *Life Sci Alliance*
23. Herz, Hamann, Rogne et al. (1988) "Surface location and high affinity for calcium of a 500-kd liver membrane protein closely related to the LDL-receptor suggest a physiological role as lipoprotein receptor" *EMBO J*
24. Croy, Shin, Knauer et al. (2003) "All three LDL receptor homology regions of the LDL receptor-related protein bind multiple ligands" *Biochemistry*
25. Bu (2001) "The roles of receptor-associated protein (RAP) as a molecular chaperone for members of the LDL receptor family" *Int Rev Cytol*
26. Obermoeller, Warshawsky, Wardell et al. (1997) "Differential functions of triplicated repeats suggest two independent roles for the receptor-associated protein as a molecular chaperone" *J Biol Chem*
27. Auderset, Cullen, Young (2016) "Low density lipoprotein-receptor related protein 1 is differentially expressed by neuronal and glial populations in the developing and mature mouse central nervous system" *PLoS One*
28. Chen, Martens, Meneses et al. (2022) "LRP1 is a neuronal receptor for αsynuclein uptake and spread" *Mol Neurodegener*
29. Rauch, Luna, Guzman et al. (2020) "LRP1 is a master regulator of tau uptake and spread" *Nature*
30. Storck, Meister, Nahrath et al. (2016) "Endothelial LRP1 transports amyloid-β(1-42) across the blood-brain barrier" *J Clin Invest*
31. Tachibana, Holm, Liu et al. (2019) "APOE4-mediated amyloid-β pathology depends on its neuronal receptor LRP1" *J Clin Invest*
32. Tilston-Lunel, Acrani, Randall et al. (2015) "Generation of recombinant oropouche viruses lacking the nonstructural protein NSm or NSs" *J Virol*
33. Bird, Albariño, St (2007) "Rift Valley fever virus lacking NSm proteins retains high virulence in vivo and may provide a model of Full-Length Text Journal of Virology December"
34. *Virology (Auckl)*
35. Schuksz, Fuster, Brown et al. (2008) "Surfen, a small molecule antagonist of heparan sulfate" *Proc Natl Acad Sci*
36. Mcmillen, Arora, Boyles et al. (2018) "Rift Valley fever virus induces fetal demise in sprague-dawley rats through direct placental infection" *Sci Adv*
37. Evans, Winkler, Peterson (2019) "Differences in neuropathogen esis of encephalitic California serogroup viruses" *Emerg Infect Dis*
38. Migliorini, Behre, Brew et al. (2003) "Allosteric modulation of ligand binding to low density lipoprotein receptor-related protein by the receptor-associated protein requires critical lysine residues within its carboxyl-terminal domain" *J Biol Chem*
39. Ma, Adams, Raju et al. (2024) "The low-density lipoprotein receptor promotes infection of multiple encephalitic alphaviruses" *Nat Commun*
40. Clark, Clark, Lin et al. (2022) "VLDLR and ApoER2 are receptors for multiple alphaviruses" *Nature*
41. Zhai, Li, Veit et al. (2024) "LDLR is used as a cell entry receptor by multiple alphaviruses" *Nat Commun*
42. Willnow, Rohlmann, Horton et al. (1996) "RAP, a specialized chaperone, prevents ligand-induced ER retention and degradation of LDL receptor-related endocytic receptors" *EMBO J*
43. Bennett, Nelson, Gresko et al. (2011) "The full genome sequence of three strains of Jamestown Canyon virus and their pathogenesis in mice or monkeys" *Virol J*
44. Kato, Takayama-Ito, Satoh et al. (2021) "Favipiravir treatment prolongs the survival in a lethal mouse model intracerebrally inoculated with Jamestown Canyon virus" *PLoS Negl Trop Dis*
45. Evans, Winkler, Peterson (2022) "Differences in neuroinvasion and protective innate immune pathways between encephalitic California serogroup orthobunyaviruses" *PLoS Pathog*
46. Dieckmann, Dietrich, Herz (2010) "Lipoprotein receptors -an evolutionarily ancient multifunctional receptor family" *Biol Chem*
47. Ma, Kim, Kafai et al. (2020) "LDLRAD3 is a receptor for Venezuelan equine encephalitis virus" *Nature*
48. Li, Cam, Bu (2001) "Low-density lipoprotein receptor family: endocytosis and signal transduction" *Mol Neurobiol*
49. Marakasova, Olivares, Karnaukhova et al. (2021) "Molecular chaperone RAP interacts with LRP1 in a dynamic bivalent mode and enhances folding of ligand-binding regions of other LDLR family receptors" *J Biol Chem*
50. Nakajima, Kulik, Frotscher et al. (2013) "Low density lipoprotein receptor-related protein 1 (LRP1) modulates N-methyl-D-aspartate (NMDA) receptor-dependent intracellular signaling and NMDA-induced regulation of postsynaptic protein complexes" *J Biol Chem*
51. Liu, Hu, Tsai et al. (2015) "Neuronal LRP1 regulates glucose metabolism and insulin signaling in the brain" *J Neurosci*
52. Gan, Jiang, Mclean et al. (2014) "Low-density lipoprotein receptor-related protein 1 (LRP1) regulates the stability and function of GluA1 α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor in neurons" *PLoS One*
53. Weeber, Beffert, Jones et al. (2002) "Reelin and ApoE receptors cooperate to enhance hippocampal synaptic plasticity and learning" *J Biol Chem*
54. Trommsdorff, Gotthardt, Hiesberger et al. (1999) "Reeler/Disabled-like disruption of neuronal migration in knockout mice lacking the VLDL receptor and ApoE receptor 2" *Cell*
55. May, Rohlmann, Bock et al. (2004) "Neuronal LRP1 functionally associates with postsynaptic proteins and is required for normal motor function in mice" *Mol Cell Biol*
56. Hover, Charlton, Hellert et al. (2023) "Organisation of the orthobunyavirus tripodal spike and the structural changes induced by low pH and K + during entry" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548560&blobtype=pdf | # Implementation of a Multiplex PCR Amplification System Combined With Next-Generation Genome Sequencing to Decipher the Circulation of Human Coronavirus 229E Lineages in Southern France
Houmadi Hikmat, Justine Py, | Boschi, Emilie Burel, Lorlane Le Targa, Matthieu Million, Lucile Lesage, | Aurélie Morand, | Bernard, La Scola, Philippe Colson, Bernard Scola
## Abstract
Coronaviruses rapidly evolve and are prone to new virus emergence. Human coronavirus (HCoV)-229E is one of the seven coronaviruses (aside HCoV-OC43, HCoV-HKU1, HCoV-NL63, SARS-CoV, MERS-CoV, SARS-CoV-2) causing respiratory infections in humans. Genomic data are very scarce for this virus. We implemented an in-house multiplex PCR strategy to amplify HCoV-229E genomes from nasopharyngeal samples, before next-generation sequencing using Nanopore or Illumina technologies. HCoV-229E genomes were assembled and analyzed using MAFFT, MEGA, Itol, Nexstrain, and Nextclade softwares. Thirty-one PCR primer pairs designed to amplify HCoV-229E genome overlapping fragments allowed obtaining 123 genomes classified in an emerging HCoV-229E lineage first reported in China, with two sublineages being delineated. Relatively to genome NC_002645.1 (2001), regarding nucleotide mutations, 1167 substitutions, 72 insertions, and 34 deletions were detected, while regarding amino acid mutations, 415 susbstitutions, 39 deletions, and 14 amino acid insertions were detected. Genes with the greatest diversity were the spike protein-encoding gene, then Nsp3. The two sublineages harbored signature mutations. We almost doubled the HCoV-229E genome set available worldwide and provided the first French genomes. Further studies are needed to strengthen knowledge about this virus′ phylogenomics and evolutionary dynamics, which may purvey clues to contribute improving coronavirus knowledge.
| IntroductionHuman coronavirus-229E (HCoV-229E) was the first discovered, in 1966, of the seven coronaviruses currently known to infect and cause respiratory diseases in humans [1,2].Infections are most often associated with mild clinical symptoms but can be severe and life-threatening in children, elder people, and in case of underlying illness [3]. They show a seasonality in temperate countries with the greatest incidence during winter and spring [4].This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
HCoV-229E is classified in genus Alphacoronavirus [1]. Its genome is a single-stranded positive-sense RNA with an approximate size of 27 kilobases (kb). Its first two-thirds encode nonstructural proteins (namely, NSP1-NSP16) and the remaining third encodes structural proteins including the spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins. An accessory protein is encoded by a gene located between those encoding the spike and the envelope [5]. HCoV-229E is classified into six genotypes named 1-6, while an emerging lineage was reported in China in 2023 [3,6]. Aminopeptidase in N is the primary cell surface receptor for this virus [7]. Natural and intermediate hosts for HCoV-229E are deemed to be bats and alpaca, respectively [1,8].
The HCoV-229E genome was primarily studied between 1978 and 2001 [9,10], and the first complete genome sequence from a clinical isolate was described in 2012 [11]. Still, there is currently a huge discrepancy between the million SARS-CoV-2 genomes available in worldwide databases (https://gisaid.org/; https://www.ncbi.nlm.nih.gov/genbank/) and the only 130 genomes (as of 01/01/2024) available in the NCBI Genbank nucleotide sequence database. Besides, none of these genomes originated from France. There is therefore a considerable paucity of data and, consequently, of understanding of HCoV-229E genetic diversity and evolution. Hence, here we aimed to implement an in-house multiplex PCR strategy to amplify overlapping HCoV-229E genome fragments from nasopharyngeal samples that had been diagnosed as HCoV-229E RNA-positive in Marseille, Southeastern France, and to sequence and analyze the obtained genomes.
## 2 | Materials and Methods
## 2.1 | Respiratory Samples
Next-generation sequencing (NGS) of HCoV-229E genomes was carried out retrospectively from remains of nasopharyngeal samples sent to our clinical microbiology laboratory at University and Public Hospitals of Marseille, Southeastern France, for diagnosis of respiratory infections in the setting of clinical routine management, and stored at -20°C/-80°C. HCoV-229E RNA testing had been performed by multiplex real-time reverse-transcription (RT)-PCR (qPCR), as previously described [12].
## 2.2 | PCR Primer Design and PCR Amplification of Overlaping Regions Covering the Whole Genomes
All near complete or complete HCoV-229E genomes available from GenBank (https://www.ncbi.nlm.nih.gov/genbank/ ) (Supplementary Methods) as of 28/02/2022 were retrieved. Recovered genomes were aligned using MAFFT (https://mafft.cbrc. jp/alignment/server/index.html). PCR primers targeting the most conserved regions of the genomes were designed using Gemi (https://sourceforge.net/projects/gemi/) to implement a PCR amplification primer set that enables generating overlaping amplicons covering the whole genome sequence. The list of PCR primers and PCR conditions for HCoV-229E genome amplification are provided in Supplementary Methods and Supporting Information S1: Table S1.
## 2.3 | NGS
To test designed PCR primers and PCR conditions, NGS used the Oxford Nanopore technology (ONT), with the Ligation sequencing kit SQK-LSK109, then the library deposit on a SpotON flow cell Mk I, R9.4.1 and a GridION instrument (Oxford Nanopore Technologies Ltd., Oxford, UK). Thereafter, we performed NGS on RNA extracts obtained using the King-Fisher Flex system (Thermo Fisher Scientific, Waltham, MA, USA) from available remains of HCoV-OC43 RNA-positive nasopharyngeal samples. At this step, NGS was carried out using the Illumina technology on a NovaSeq. 6000 instrument with the CovidSeq protocol (Illumina Inc., San Diego, CA, USA) but by replacing Covid-19 ARTIC PCR primers by PCR primers designed here and according to PCR conditions we previously set up. Loading procedure on a NovaSeq. 6000 SP flow cell followed the NovaSeq-XP workflow and a previously described protocol [13] with a reading of 2×50 nucleotides.
## 2.4 | Processing and Bioinformatic Analyses of NGS Reads and Viral Genomes
Genomes were assembled by mapping on HCoV-229E genome GenBank accession no. LC654445.1 (Fukushima_H829_2020 isolate) with Minimap2 (https://github.com/lh3/minimap2) (Supplementary Methods). A phylogenetic tree was created with MEGA (v.11; https://www.megasoftware.net/) using the Neighbor-Joining method and Maximum composite likelihood parameter model. All HCoV-229E genomes available from GenBank including those corresponding to genogroups were incorporated in the phylogeny. Nextstrain (https://nextstrain. org/) and Nextclade (https://clades.nextstrain.org/) were adapted to enable identifying viral lineages and mutations. Nucleotide and amino acid (aa) diversity was analyzed relatively to the HCoV-229E reference genome no. NC_002645.1 (described in 2001 and obtained from a laboratory-adapted strain derived from a strain isolated in 1962) [10].
## 3 | Results
A total of 524 (0.75%) of 70,336 nasopharyngeal samples had been diagnosed as HCoV-229E RNA-positive in our institution between January 2017 and October 2022, but only remains for 195 of them, which had been collected between March 2021 and March 2022, were available as stored frozen and in sufficient volume (Supporting Information S1: Figure S1). These 195 specimens were tested using a multiplex PCR amplification strategy to amplify the genome per short overlapping fragments for further NGS. After designing and testing individually PCR primer pairs, 31 of them generating amplicons covering the entire HCoV-229E genome were selected (Supporting Information S1: Table S1); primer concentration in PCR ranged between 10 and 15.3 µM. These PCR primer pairs were used in two separate pools to prevent unwanted hybridizations of primers and generated amplicons. They allowed obtaining 123 genomes with a completion corresponding to ≥ 80% coverage of reference genome NC_002645.1; mean coverage was 92.1% (range, 80.0%-98.0%) (Supplementary Results).
The 123 near-full genomes were obtained from nasopharyngeal samples collected between 03/2021 and 03/2022. They were classified by Nextclade and phylogeny as belonging to an emerging lineage reported in China in 2023 [6] and designated as a putative genotype 7 in two recent reports [14,15] (Figure 1). Based on phylogeny, this emerging lineage comprised two sublineages, one of which appeared to match with previously designated sublineage 7b [14,15] whereas there seems to be discrepancies between matches for the second sublineages and previously designated sublineage 7a [14,15]. Whatever, 30/123 genomes obtained here belong to a sublineage "a" and 93 belong to a sublineage "b" (Figure 1), revealing that these two sublineages co-circulated in our geographical area, with a sublineage b predominance (Figure 2). Sublineage a was detected since 03/2021 while sublineage b was detected since 09/2021 (Figure 2). For the 4 months during which the number of genomes obtained from collected specimens were above 10, the proportion of genomes of sublineage a decreased from 32% in 11/2021 to 7% in 02/2022.
The nucleotide and aa diversity of the 123 HCoV-229E genomes were determined relatively to genome NC_002645.1 (Supplementary Results). Regarding aa mutations, 415 substitutions, 39 deletions, and 14 insertions were detected. Of the 415 aa substitutions, only 211 were present in ≥ 5 genomes. These 211 aa substitutions were found in several genes, including in the Nsp1, Nsp2, Nsp3, Nsp4, Nsp6, Nsp8, Nsp9, Nsp10, Nsp11, Nsp12, Nsp13, Nsp14, Nsp15, spike, ORF4a, E, M, and N genes (Figures 3 and4). A total of 78 (37%) of these 211 aa substitutions were in the spike. They displayed various prevalence (Supporting Information S1: Table S2). Seventy-one were already reported, being mentioned as new substitutions in 13 cases [6]. Spike deletions S:V353-and S:Y354-were observed in all genomes obtained here, while S:A352-was observed in all but four genomes (97%) that harbor four other deletions: S:A355-, S:N356-, S:V357-, and S:G358-(Supporting Information S1: Table S2). Two other deletions were present in NSP3 (NSP3:L105-, NSP3:P106-) of all genomes.
The genes with the greatest diversity were S, which encodes the spike protein, with 67 aa mutations/1,000 aa, then Nsp2, N, Nsp3, which encodes a large multi-domain nonstructural protein and is an essential component of the viral replication/ transcription complex, and E, with between 26 and 28 aa mutations/1000 aa (Figure 4). The genes with the lowest diversity were NSP7, which is part of the RNA-dependent RNA 1). Four mutations in the S gene that we detected in sublineages a or b genomes were previously reported as newly detected [6]. The present study first allowed developing a new multiplex PCR system to amplify whole HCoV-229E genomes before NGS. Second, this system allowed obtaining 123 new near fulllength HCoV-229E genomes, almost as many as available globally early 2024 and the first from France. All these 123 genomes were classified in an emerging lineage and were found to exhibit 468 different aa mutations relatively to a 2001 reference genome, with approximately one-third of those present in at least five genomes being located in the spike-encoding gene.
## Emerging lineage, sublineage a
## Signature mutations: Nsp2:H40R; S:K349Q; S:D391A; S:Y406N; and N:E33D
## Emerging lineage, sublineage b
## Signature mutations: Nsp2:A350V; Nsp3:V423M; Nsp3:K428Q; S:T19I; S:F229V; S:S248A; S:V288E; S:Y305H; S:K309E; S:P310H; S:Q311G; S:S312R; S:F318Y; S:Y321R; S:G324V; S:V325I; S:K349R; S:N365D; S:N377I; S:D391Y; S:Y406G; and S:V739A
Similar multiplex PCR systems for HCoV-229E whole genome amplification were reported in 2024 by Musaeva et al. [15] and McClure et al. [14], in Russia and the UK, respectively. The first one [15] used 29 PCR primer pairs, compared with 31 in the present study, and enabled obtaining 39 genomes with a coverage of a full-length genome > 70%, from 50 HCoV-229E RNApositive nasopharyngeal swab samples. The second one [14] used 29-36 primer pairs for all four endemic human coronaviruses (HCoV-229E, -OC43, -NL63, and -229E), and enabled obtaining 64 HCoV-229E genomes with a coverage of a full-length genome > 95%. Overall, as of 01/02/2025, 269 HCoV-229E genomes were available in GenBank, and the 123 genomes obtained here grew the global set to 392.
All the genomes obtained here belong to the emerging lineage previously reported by Ye et al. [6], with 43 full-length genomes.
We report here that this lineage tentatively named genotype 7 in Musaeva et al's and McClure et al′s study [14,15] also circulated in France. Two sublineages 7a and 7b had been reported [14,15], while we also observed in the present study two sublineages. Hence, our findings further support that the emerging lineage initially reported to have circulated between 2016 and 2020 and in China, then in Japan, Haiti, the United States, Russia, and the UK likely became the predominant lineage worldwide. In addition, together with previous data [6,14,15], they indicate that it is evolving with new mutations whose occurrence may depend on time and geographical area.
A high aa diversity was observed here in the spike and nonstructural proteins, which is consistent with previous findings in coronaviruses [6,[14][15][16][17][18]. The HCoV-229E receptor binding domain (RBD) of the spike contains three loops, named 1, 2, and 3, involved in virus binding to the host aminopeptidase N cellular receptor and are located at aa positions 308-325, 352-359, and 404-408, respectively [17]. Some aa mutations in these regions were observed here as in three previous studies [6,14,15,18]. These notably involved four aa positions. For three of them, two different mutations were observed here (K349R or Q; G358P or A; Y406G or N). For position 391, two substitutions were observed here that are signatures of either sublineage a (D391A) or b (D391Y). Besides, W404 in the spike RBD was reported as very important for loop 3 binding to the cellular receptor [17]. Notwithstanding, tryptophan was replaced here by a leucine in all genomes, indicating that this mutation may not preclude viral infection. Mutations Q430K, D444N, and K488N also encountered here were already reported and would result in an N-glycosylation site at position 488 [18]. Besides, apolar bonds were predicted that involve RBD aa, notably at position 318 [17]. A mutation at this position was associated with an ≈13-fold reduction in cellular receptor affinity [17], but mutation F318Y was found here in 76% of the genomes and previously in HCoV-229E genotypes 3-6 and the emerging lineage [6]. These data highlight the broad diversity of spike aa patterns and may be useful to interpret structural analyses performed previously and in future studies including to investigate the putative impact of these different mutations on HCoV-229E-host receptor interaction [19,20].
Overall, the present study and two other recent studies [14,15] enrich the set of HCoV-229E genomes available worldwide, with 123 and 103 genomes, respectively. Nonetheless, genomic data remain scarce and they cover a limited number of countries, therefore not necessarily reflecting the circulation of this virus at the global scale. Further studies are therefore needed to gain a more global view of the evolutionary dynamics of HCoV-229E and its lineages. This will clarify the specificities of this virus and may contribute to a more general understanding of the evolution of human coronaviruses.
## References
1. Tang, Liu, Chen (2022) "Human Coronaviruses: Origin, Host and Receptor" *Journal of Clinical Virology*
2. Hamre, Procknow, Procknow (1966) "New Virus Isolated From the Human Respiratory Tract" *Proceedings of the Society for Experimental Biology and Medicine. Society for Experimental Biology and Medicine*
3. Lau, Lung, Wong (2021) "Molecular Evolution of Human Coronavirus 229E in Hong Kong and a Fatal COVID-19 Case Involving Coinfection With a Novel Human Coronavirus 229E Genogroup" *mSphere*
4. Killerby, Biggs, Haynes (2018) "Human Coronavirus Circulation in the United States 2014-2017" *Journal of Clinical Virology*
5. Dijkman, Jebbink, Wilbrink (2006) "Human Coronavirus 229E Encodes a Single ORF4 Protein Between the Spike and the Envelope Genes" *Virology Journal*
6. Ye, Gong, Cui (2023) "Continuous Evolution and Emerging Lineage of Seasonal Human Coronaviruses: A Multicenter Surveillance Study" *Journal of Medical Virology*
7. Yeager, Ashmun, Williams (1992) "Human Aminopeptidase N Is a Receptor for Human Coronavirus 229E" *Nature*
8. Corman, Baldwin, Tateno (2015) "Evidence for an Ancestral Association of Human Coronavirus 229E With Bats" *Journal of Virology*
9. Macnaughton, Madge (1978) "The Genome of Human Coronavirus Strain 229E" *Journal of General Virology*
10. Thiel, Herold, Schelle et al. (2001) "Infectious Rna Transcribed In Vitro From a cDNA Copy of the Human Coronavirus Genome Cloned in Vaccinia Virus" *Journal of General Virology*
11. Farsani, Dijkman, Jebbink (2012) "The First Complete Genome Sequences of Clinical Isolates of Human Coronavirus 229E" *Virus Genes*
12. Mantelli, Colson, Lesage (2024) "Coinfections and Iterative Detection of Respiratory Viruses Among 17,689 Patients Between March 2021 and December 2022 in Southern France" *Journal of Clinical Virology*
13. Papa Mze, Kacel, Beye (2023) "High Throughput SARS-CoV-2 Genome Sequencing From 384 Respiratory Samples Using the Illumina COVIDseq Protocol" *Genes*
14. Mcclure, Tsoleridis, Hill (2025) "′Vivaldi′: An Amplicon-Based Whole-Genome Sequencing Method for the Four Seasonal Human Coronaviruses, 229E, NL63, OC43 and HKU1, Alongside SARS-COV-2" *Microbial Genomics*
15. Musaeva, Fadeev, Pisareva (2024) "Development of Primer Panels for Whole-Genome Amplification and Sequencing of Human Seasonal Coronaviruses: hCoV-OC43, hCoV-HKU1, hCoV-229E, and hCoV-NL63" *Viruses*
16. Lei, Kusov, Hilgenfeld (2018) "Nsp3 of Coronaviruses: Structures and Functions of a Large Multi-Domain Protein" *Antiviral Research*
17. Wong, Tomlinson, Zhou (2017) "Receptor-Binding Loops in Alphacoronavirus Adaptation and Evolution" *Nature Communications*
18. Chibo, Birch (2006) "Analysis of Human Coronavirus 229E Spike and Nucleoprotein Genes Demonstrates Genetic Drift Between Chronologically Distinct Strains" *Journal of General Virology*
19. Li, Tomlinson, Wong (2019) "The Human Coronavirus HCoV-229E S-Protein Structure and Receptor Binding"
20. Song, Shi, Ding (2021) "Cryo-EM Analysis of the HCoV-229E Spike Glycoprotein Reveals Dynamic Prefusion Conformational Changes" *Nature Communications* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12596406&blobtype=pdf | # Drug Resistance in People With Viremia on Dolutegravir-based Antiretroviral Therapy in Sub-Saharan Africa: The DTG RESIST Study
Tom Loosli, Carolyn Moore, Lydia Buzaalirwa, Helen Byakwaga, ̇pek Çelikağ, Cleophas Chimbetete, Peter Vanes Ebasone, Jennifer Giandhari, Nuri Han, Jacqueline Huwa, Charles Kasozi, Adolphe Mafoua, Eugène Messou, Albert Minga, Guy Muula, Winnie Muyindike, Arcel Christ, Massamba Ndala, Mamatha Sauermann, Aggrey Semeere, Lavanya Singh, Roger Kouyos, Richard Lessells, Matthias Egger
## Abstract
Dolutegravir resistance is an increasing concern. An analysis of the DTG RESIST study found that among 227 integrase sequences from 7 African countries (all non-B subtypes), 59 (26.0%) had at least 1 major drug resistance mutation (primarily G118R and E138A/K/T), with 49 (21.6%) predicted to have high-level resistance to dolutegravir.
Dolutegravir (DTG) is widely used in antiretroviral therapy (ART) programs, particularly in low-and middle-income countries (LMICs) [1]. Acquired DTG resistance is rare in high-income settings [2,3]. In a study primarily involving participants from European and North American cohorts, the prevalence of major integrase strand transfer inhibitor (INSTI) drug resistance mutations (DRMs) among people with human immunodeficiency virus (HIV, PWH) experiencing viremia on DTG-based ART was 7.2%. High-level resistance to DTG was observed in only 1% of cases [3].
Data on DTG resistance in sub-Saharan Africa are urgently needed, given the region's significant HIV burden and extensive subtype diversity [4]. PWH on ART often experience prolonged periods of viremia in these settings, and access to routine HIV drug resistance testing remains limited [5]. The recent World Health Organization (WHO) drug resistance report highlights surveys from LMICs that show an increasing number of cases of DTG resistance [6].
The DTG RESIST study aims to characterize the prevalence and patterns of DTG resistance among adults and adolescents with virological failure on DTG-based ART across Africa, Asia, and South America [7]. Here, we provide data collected from 7 African countries as of December 2024.
## METHODS
## DTG RESIST is nested within the International Epidemiology
Databases to Evaluate AIDS (IeDEA) network [8]. The study involves prospective enrollment from routine clinical care at sites in Asia, South America, and sub-Saharan Africa selected to represent diverse settings, programmatic contexts, and non-B HIV subtypes. Enrollment began in June 2022 and will continue until May 2025. Further details are outlined in the study protocol [7]. In brief, the study includes adults and adolescents (aged 10-19 years) with at least 1 routine viral load (VL) >1000 copies/mL while on DTG-based ART. Plasma or dried blood spots (DBS) are collected depending on local storage capacities. Samples are sent to centralized laboratories, and genotypic resistance testing is performed on those with VL >1000 copies/mL.
Here, we present data from participants enrolled between June 2022 and December 2024 at 16 study sites in 7 African countries: Cameroon (n = 2), Côte d'Ivoire (n = 2), Malawi (n = 2), Republic of Congo (n = 2), Uganda (n = 2), Zambia (n = 5), and Zimbabwe (n = 1). Supplementary Table 1 provides further details. Sanger sequencing of integrase (IN) and protease/reverse transcriptase (PR/RT) regions was conducted using the ThermoFisher HIV-1 Genotyping Kit [9]. Sequences were analyzed with the Stanford HIVdb algorithm V9.8 [10], and HIV subtypes were determined using COMET [11] and Rega V3.46 [12].
We report the proportion of individuals with resistance among those with documented viremia (VL >1000 copies/mL) on DTG-based ART, along with the associated mutational patterns.
## RESULTS
The analysis included 488 participants; most were female (N = 284, 58.2%), and 73 (15%) were adolescents. The median duration on DTG-based ART was 2.5 years (interquartile range [IQR], 1.6-3.4). Most (n = 447, 91.6%) were on a regimen that contained DTG with 2 Nucleos(t)ide Reverse Transcriptase Inhibitors (NRTIs), with 336 of 447 (75.2%) receiving tenofovir/lamivudine/dolutegravir (TLD). Most participants (n = 421, 86.3%) had prior exposure to at least 1 ART regimen, while 67 (13.7%) were treatment-naive when they started DTG. Previous exposure to raltegravir was rare (n = 9, 1.8%). Most participants (n = 415, 85%) had 2 or more consecutive VL measurements >1000 copies/mL before enrollment. The median time between the most recent routine VL >1000 copies/mL and enrollment was 36 days (IQR, 22-69). Supplementary Tables 2 and3 provide further details.
Of the 488 samples (420 plasma, 68 DBS samples), 234 (48%) had a VL >1000 copies/mL and underwent sequencing. IN sequencing was successful for 227 of 234 participants (97.0%), and PR/RT sequencing was successful for 217 of the 227 with IN sequences (95.6%). Overall, 59 of 227 (26.0%) participants had major INSTI DRMs (Figure 1A). The most frequent major INSTI DRMs were G118R (n = 36) and E138A/K/T (n = 48), occurring together in 32 study participants (Figure 1A). R263K was less common, as were G140A and Q148K/R. Most participants (50 of 59, 84.7%) had 2 or more major INSTI DRMs (Figure 1B). The most frequent HIV subtypes were C (n = 135, 62.2%), G (n = 30, 13.8%), and A1 (n = 29, 13.4%); the prevalence of major DRMs varied across subtypes (Figure 1C). All 59 cases with major INSTI DRMs were treatmentexperienced, including 7 of 41 (17.1%) adolescents with major INSTI DRMs. Median duration on DTG-based ART was 2.8 years (IQR, 2.1-3.4) and 2.4 years (IQR, 1.6-3.4) for participants with and without major INSTI DRMs, respectively. High-level resistance to DTG was predicted in 49 of 227 (21.6%) participants; across countries, the prevalence ranged from 11.4% to 40.9% (Supplementary Table 4). Among the 67 participants who were ART-naive when they started DTG, 21 had VL >1000 copies/mL, none had major INSTI DRMs, and 4 had accessory DRMs (Q95K, H51Y, E157Q, and S153A).
PR/RT sequencing was successful for 57 of the 59 participants with major INSTI DRMs; 55 (96.5%) harbored M184V. In contrast, K65R was rare, occurring in 5 (8.8%) participants. Thymidine analogue mutations (TAMs) were detected in 41 of 57 (71.9%) participants, with 21 of 41 (51.2%) having 3 or more TAMs. Of the 41 participants with TAMs and major INSTI DRMs, 27 (65.9%) were on TLD at enrollment. Intermediate or high-level tenofovir resistance was identified in 18 of 57 (31.6%) participants with major INSTI DRMs. Among 158 participants susceptible to DTG (with PR/RT sequencing available), M184V was found in 17 (10.8%), while (7.0%) had at least 1 TAM. Intermediate or high-level tenofovir resistance was observed in 3.8% of these cases. Protease inhibitor (PI) resistance was rare. Major PI DRMs were detected in 8 of 57 (14.0%) participants with major INSTI DRMs. All participants with PI DRMs had predicted susceptibility to darunavir.
## DISCUSSION
In this analysis of African DTG RESIST data, high-level DTG resistance was detected in about 1 in 5 people with non-B HIV-1 and experiencing virological failure on DTG-based ART. The most common mutations were G118R and E138A/ K/T, frequently occurring in combination. All individuals with major INSTI DRMs were treatment-experienced, and the majority also had NRTI resistance. Although few participants were on PIs at the time of resistance testing, the rarity of PI resistance mutations suggests that PIs may still be an option for those experiencing virologic failure on DTG-based ART.
Our findings align with the findings presented in a recent WHO report on DTG resistance. The report estimated that the prevalence of DTG resistance among individuals with viremia on DTG-based ART in sub-Saharan Africa ranged from 3.9% to 19.6% [6]. The country-specific DTG resistance prevalence observed in our study falls within this range, but numbers are small, and country differences must be interpreted cautiously. The differences observed may reflect variations in the calendar time of the DTG rollout in different countries, as well as different inclusion criteria, implementation practices, or study populations. Nevertheless, a common finding across studies is the rising prevalence of DTG resistance among individuals experiencing virological failure on DTG-based ART [6].
The DTG resistance patterns observed in sub-Saharan Africa in this study and the studies covered in the WHO report [6] contrast with analyses from high-income settings [2,3]. Among individuals with virological failure on DTG-based ART, high-level DTG resistance was approximately 20 times more frequent in sub-Saharan Africa than in Europe or North America [3]. Additionally, the mutational pathway in sub-Saharan Africa primarily involved G118R in combination with E138A/K/T, while in the European and North American cohorts, the predominant pathway was R263K. DTG resistance levels also differed. In high-income settings, potential low-or intermediate-level resistance dominated, and high-level resistance was rare [3]. By contrast, in sub-Saharan Africa, most people with resistance had high-level DTG resistance, often due to multiple major INSTI DRMs. This situation may reflect the individualized care in high-income countries, with frequent VL monitoring and genotypic resistance testing performed earlier during virological failure than in ART programs in resource-limited African settings. The differences in HIV-1 subtypes could also play a role [3]. Of note, those with major INSTI DRMs were treatment-experienced individuals in both settings. Understanding these differences is crucial for developing strategies for preventing drug resistance.
The strengths of this study include the large number of sites and ART programs in 7 countries that participate in IeDEA, a long-standing collaboration of routine ART programs in sub-Saharan Africa [8]. Of note, the study sites reflect different approaches to switching PWH to DTG-based ART, including sites where switching was irrespective of viral load and sites where switching depended on a suppressed viral load in the past year [13]. The IeDEA collaboration collects a wide range of longitudinal data, allowing future analyses of clinical outcomes and the durability and effectiveness of DTG-based ART regimens over time. Our study also has several limitations. Currently, the data on the duration of viremia during DTG-based ART and prior ART regimens are incomplete; however, linking study data to the broader IeDEA cohort will allow more detailed analyses of the factors that contribute to resistance. Samples with detectable viral loads below 1000 copies/ mL were not sequenced. We may, therefore, have missed early emergent resistance mutations. Also, drug levels were not measured, which prevented insights into the relationship between adherence and resistance. Finally, we used the Stanford HIVdb algorithm but acknowledge that other systems may classify mutations differently and that discriminating between major and accessory (or major and minor) is not always clear-cut. For example, the E138A/K/T mutation is not considered a major mutation in the drug mutation chart of the International AIDS Society (IAS)-USA.
In conclusion, DTG remains a cornerstone of HIV treatment, achieving high rates of viral suppression in most PWH. DTG resistance among those newly initiating the drug remains low. However, this interim analysis of data from the sub-Saharan African sites of DTG RESIST shows that 1 in 5 individuals on failing DTG-based ART high-level DTG resistance, most commonly through the G118R and E138A/K/T mutational pathways. Safeguarding the efficacy of INSTI-based ART in LMICs will require proactive measures to address the emergence of DTG resistance, ensuring sustained success in HIV treatment.
## References
1. (2024) "Update of recommendations on first-and secondline antiretroviral regimens"
2. Chu, Tao, Kouamou (2024) "Prevalence of emergent dolutegravir resistance mutations in people living with HIV: a rapid scoping review" *Viruses*
3. Loosli, Hossmann, Ingle (2023) "HIV-1 drug resistance in people on dolutegravir-based antiretroviral therapy: a collaborative cohort analysis" *Lancet HIV*
4. Da Silva, Pals, Chang et al. (2022) "Monitoring emerging human immunodeficiency virus drug resistance in sub-Saharan Africa in the era of dolutegravir" *J Infect Dis*
5. Bernabé, Siedner, Tsai et al. (2022) "Detection of HIV virologic failure and switch to second-line therapy: a systematic review and meta-analysis of data from sub-Saharan Africa" *Open Forum Infect Dis*
6. (2024) "HIV drug resistance: brief report"
7. Egger, Sauermann, Loosli (2024) "HIV-1 subtype-specific drug resistance on dolutegravir-based antiretroviral therapy: protocol for a multicentre study (DTG RESIST)" *BMJ Open*
8. Chammartin, Ostinelli, Anastos (2012) "International epidemiology databases to evaluate AIDS (IeDEA) in sub-Saharan Africa" *BMJ Open*
9. Moore, Palumbo, Notarte (2024) "Performance of the Applied Biosystems HIV-1 genotyping kit with integrase" *J Clin Microbiol*
10. Liu, Shafer (2006) "Web resources for HIV type 1 genotypic-resistance test interpretation" *Clin Infect Dis*
11. Struck, Lawyer, Ternes et al. (2014) "COMET: adaptive context-based modeling for ultrafast HIV-1 subtype identification" *Nucleic Acids Res*
12. Pineda-Peña, Faria, Imbrechts (2013) "Automated subtyping of HIV-1 genetic sequences for clinical and surveillance purposes: performance evaluation of the new REGA version 3 and seven other tools" *Infect Genet Evol*
13. Skrivankova, Huwa, Muula (2025) "Virologic failure and drug resistance after programmatic switching to dolutegravir-based first-line antiretroviral therapy in Malawi and Zambia" *Clin Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846550&blobtype=pdf | # Completion of the Genome Sequence of a Historic CDV Vaccine Strain, Rockborn: Evolutionary and Epidemiologic Implications
Zsófia Lanszki, Krisztián Bányai, Ágnes Bogdán, Gábor Kemenesi, Georgia Diakoudi, Gianvito Lanave, Francesco Pellegrini, Nicola Decaro, Vito Martella
## Abstract
Canine distemper is a serious, contagious disease in dogs and other animals, and vaccines are crucial for its control. An old vaccine strain, called Rockborn, was widely used but later withdrawn from many markets due to safety concerns, although vaccines with Rockbornlike viruses are still on the market. We sequenced the entire genome of a Rockborn virus cell passage available in our laboratory to understand its evolution and improve future diagnostic tools. Our analyses revealed that viruses similar to the Rockborn strain are still circulating in wildlife and domestic dogs. More importantly, it was discovered that the Rockborn strain has been exchanging genetic material with other canine distemper viruses, creating new, recombinant viruses.
## 1. Introduction
Canine distemper virus (CDV) is a paramyxovirus (Paramyxoviridae family, Morbillivirus genus) and possesses an enveloped virion with a minus-sense single-stranded RNA genome of ~15 kilobases in length. Among the six main primary gene products (N, P, M, F, H, and L) (Supplementary Table S1 [1][2][3][4][5][6][7][8][9][10][11][12]), the H gene is the most characterized genomic region. The H gene product acts as a cell attachment protein and mediates entry; moreover, antibodies against the H protein protect against infection and disease [13]. The exposure of the H protein to the immune system and the widespread use of CDV vaccines globally are thought to contribute to the genetic diversity among circulating strains. A genotyping classification scheme of CDV has been established based on the H gene [14][15][16][17].
The historic CDV strain, Rockborn, was isolated and attenuated by serial passage in the late 1950s [18]. Rockborn strain-based vaccines have been commercialized and distributed globally since the 1960s. This vaccine was shown to elicit a strong immune response and protect against disease caused by field CDV strains. However, safety concerns started to rise in the 1970s due to observed residual virulence. Although the etiological role was never fully proven, cases of vaccine-related encephalitis in the United States in the mid-1990s led to the definitive withdrawal of Rockborn-based CDV vaccines from the market [19,20]. Yet, several modern vaccines seemingly contain Rockborn-like viruses [21].
Overall, the high genetic identity between vaccine stock/laboratory strains and field isolates of vaccine-origin clinical cases was consistent with the hypothesis that the vaccine viruses retained residual virulence, likely in co-infection with other immunosuppressive pathogens. This was attributed to an intrinsic virulence of low-passage stocks established on primary dog kidney (DK) cells and grown for vaccine passage cultures on Madin Darby Canine Kidney (MDCK) cells. In most cases, increased virulence of the dog-kidney-cellattenuated CDV was suspected to have been triggered by the contemporaneous presence in the vaccine formulations of live modified canine adenovirus type-1 [21]. Under experimental conditions, re-acquisition of virulence by the kidney-cell-attenuated Rockborn virus has been demonstrated in vivo after six sequential passages in dogs and in vitro after ten passages in primary dog lung macrophages, but the molecular bases of this phenotypic change were not investigated [22].
So far, genetic analysis of the Rockborn strain has been limited to the H gene. On sequence analysis of the full-length H gene of a cell passage (46th) of strain Rockborn, the virus and related strains were found to share 92-97% nucleotide (nt) and 91-96% amino acid (aa) sequence homology to other lineages of CDV. Additional partial H gene sequences were available for comparison, originating from vaccine-associated clinical cases and wild animals. Sequence divergence among strains varied slightly, with both nt and aa identities exceeding 99% [21]. With the determination of the full genome, we had the following objectives: (i) to better reconstruct the origin and evolution of the Rockborn strain and (ii) to extend the information with implications for diagnostic developments. Additionally, the reference genome sequence presented in this study will help identify virulence markers in clinically relevant Rockborn-associated CNS infections, if any.
## 2. Materials and Methods
## 2.1. Sequencing
The complete genome sequence of Rockborn-46th was determined using an ampliconbased sequencing protocol that combines amplification of adjacent CDV genomic fragments with MinION nanopore sequencing (Oxford Nanopore Technologies, Oxford, UK). The amplicon-based sequencing method for canine distemper virus, together with the primers employed, has been described previously [23], and the complete protocol is also available on the protocols.io page (https://www.protocols.io/view/universal-amplicon-basedsequencing-method-for-can-x54v9j6mpg3e/v1/metadata (accessed on 24 November 2021)). In brief, cDNA was prepared from extracted viral RNA with Superscript IV (Invitrogen, Carlsbad, CA, USA) using random hexamers. Overlapping PCR products were amplified from the cDNA with the Q5 Hot Start HF Polymerase (New England Biolabs, Ipswich, MA, USA) with multiple primer sets in parallel pools and then purified with AMPure XP beads (Beckman Coulter, Brea, CA, USA). The primer sequences used were designed to generate overlapping amplicons of approximately 1000 and 2000 nt in size and no modifications were made to the original CDV sequencing protocol. The end-repair and dA tailing were performed with the NEBNext Ultra II End Repair/dA-Tailing Module (New England Biolabs, USA). DNA barcodes EXP-NBD196, (Oxford Nanopore Technologies, UK) were ligated with NEBNext Ultra II Ligation Module (New England Biolabs, USA). After purification with Ampure XP beads, the AMII sequencing adapters were ligated with NEBNext Quick Ligation Module. Sixty ng of the final library was loaded onto a R9.4.1 (FLO-MIN106D) flow cell.
Also, consensus primers were designed to obtain the sequence of the 5 ′ (p2679 5 ′ -ACCAGAMAAAGTTGGCTAWGGATAGW-3 ′ ) and 3 ′ genome (p2683R 5 ′ -ACCAGACAAA GCTGGGTATGATAACT-3 ′ ) terminations.
## 2.2. Genome Assembly and Annotation
Base-calling and barcode demultiplexing were performed with Dorado version 1.1.0. Sequence reads below the expected size were removed and the consensus sequence was assembled by mapping to the MN267060 using the Geneious mapper (version Geneious Prime 2021.6.0.). Medaka (version 2022.1.1.) was used to map trimmed reads against a preliminary consensus to generate polished consensus sequences. The generated consensus sequences were manually checked for base-calling errors, especially in the homopolymeric regions. The genome was annotated using the ORF Finder (https://www.ncbi.nlm.nih. gov/orffinder/ (accessed on 5 September 2025)).
## 2.3. Phylogenetic Analysis
Sequences for the N, P, M, F, H, and L genes were first aligned using the MAFFT webserver with default parameters (https://www.ebi.ac.uk/jdispatcher/msa/mafft (accessed on 6 September 2025)). Thereafter, the IQ-TREE webserver (http://iqtree.cibiv.univie.ac.at/ (accessed on 6 September 2025)) was used to select substitution models and reconstruct maximum-likelihood phylogenetic trees, with 1000 ultrafast bootstrap replicates. The phylogenetic analyses were conducted under the GTR + F + I + G4 substitution model, which was identified as the best-fitting model according to the Akaike Information Criterion (AIC). Subsequently, the resulting trees were edited in the iTOL webserver (https://itol.embl.de/ (accessed on 7 September 2025)).
## 2.4. Recombination Detection
Whole-genome sequences were used to identify potential recombination events in Rockborn using the algorithms implemented in the Recombination Detection Program https://doi.org/10.3390/vetsci13010081 v.4.101 (RDP4) [24]. Default settings were used for each algorithm. A recombination event was accepted when detected by 7 distinct methods (RDP, GENECONV, BootScan, MaxChi, Chimaera, SiScan, 3Seq) implemented in the program, each with a p-value < 5 × 10 -4 .
## 2.5. Sequence Deposition
The genome sequence of the Rockborn strain was deposited in GenBank under the accession number PX254110.
## 3. Results
A total of 551.860 reads were mapped to CDV. The mean vertical coverage was about 44,000×. The assembled genome was 15,690 nt in length, with 107 and 105 nt long UTRs at the 3 ′ and 5 ′ ends, respectively. The lengths of N, P, M, F, H and L genes (and their respective proteins) were 1572 nt (523 aa), 1524 nt (507 aa), 1008 nt (335 aa), 1989 nt (662 aa), 1824 nt (607 aa) and 6555 nt (2184 aa), respectively. The genome and gene-wise sequence homologies showed values of over 98% and 99%, respectively, to the most closely related CDV strains, many, if not all, of which are putative Rockborn-derived vaccines and wildtype strains (Table 1). Next, whole-genome and gene-wise phylogenies were carried out (Figure 1). A wholegenome-based alignment was assembled from 223 reference genome sequences available in the GenBank and the newly determined Rockborn sequence. After a preliminary analysis, 27 relevant sequences were retained for further processing; these sequences represented the main genetic clades. The genome-wise phylogeny indicated that strain Rockborn clusters https://doi.org/10.3390/vetsci13010081
Vet. Sci. 2026, 13,81 with two field strains for which full-length genomes are available, the HN19 strain from a masked civet in China and the R252 strain from a dog in the United States. These features indicate that strain Rockborn, along with some recent field isolates, constitutes a unique genetic lineage alongside more deeply analyzed CDV lineages. A closer look revealed that R252 and HN19 CDVs share greater genetic relatedness across the entire genome, except for the region spanning the F and H genes, from nt positions 4935 to 6923 and 7079 to 8902, respectively, in the masked civet-origin CDV strain (GenBank ID: MT448054). In this data set, analysis of this genomic region showed that the masked civet-origin HN19 strain was more closely related to Rockborn than to the canine-origin R252 strain. These findings suggest that past recombination events led to the mosaic structure of the genomes of some strains within the Rockborn lineage. This peculiar pattern of gene-wise relationship among Rockborn-like strains with available full-length genomes was confirmed by recombination analysis using dedicated bioinformatic analyses (Figure 2). We used the whole alignment containing 224 sequences as input to run the analysis with default parameters. As a result, an intra-lineage recombination event was observed within Rockborn-like strains (all but one algorithm showed evidence of recombination), placing the breakpoint sites at nt 4934 and 8902 (99% CI, 4747-5036 and 8482-8974, respectively). This region spans the entire F and H genes and may include short fragments of non-coding regions upstream and downstream, as indicated by the 99% CI values.
## 4. Discussion
This study presents the complete genome sequence of the historic CDV strain Rockborn, revealing its ongoing relevance in the contemporary epidemiological landscape. These new data not only resolve long-standing questions about the genetic makeup of this lineage beyond the H gene but also underscore the complex evolutionary forces, particularly recombination, that shape CDV diversity [17,21].
Given the official withdrawal of Rockborn-based vaccines from many markets in the mid-1990s due to safety concerns [19,20], a significant aspect of this whole-genome sequencing study is the confirmation that Rockborn-like viruses or Rockborn-derived recombinants continue to circulate in dogs and wildlife. Based on the H gene sequence, Martella and coworkers identified two vaccines on the market containing Rockborn-like strains and noted that several field isolates from Austria and Japan showed nearly 100% nucleotide identity to the Rockborn-46th laboratory strain [21]. More recent studies have continued to detect these strains, reinforcing the idea of their natural persistence within and among different animal populations. Based on whole-genome sequence data generated in our study and database interrogation, we obtained additional evidence for the presence of Rockborn-like CDV strains in dogs and other carnivores (Table 1). Rockborn-like CDV was identified in a captive vaccinated fennec fox in Japan, 2017 [25], in non-vaccinated farmed masked civets in China, 2019 [26]. Also, Rockborn-like CDV was reported repeatedly, although sporadically, in domestic dogs in Brazil, 2015 [27], Canada, 2025 [28], and New Zealand, 2021-2024 [29], suggesting a continual spill-over of vaccine-derived Rockborn-like CDV strains into the environment.
The central finding of our genomic analysis was the identification of a clear recombination event, positioning the Rockborn strain as a probable parental strain to a mosaic virus circulating in the wild. While recombination in negative-sense RNA viruses was once considered a rare phenomenon [30], a growing body of evidence now establishes it as a key force driving the evolution and genetic diversity of CDV [17,31,32]. Previous studies have extensively documented recombination events involving various CDV genotypes. Budaszewski and co-workers identified eight putative recombinant viruses, demonstrating that homologous recombination is a frequent event in natural CDV populations and that vaccine strains, particularly from the America-1 lineage, play a significant role in shaping viral evolution [32]. Similarly, Yuan et al. [31] concluded that recombination is a key evolutionary force after identifying six distinct events and noting that viruses isolated from different host species-such as minks, seals, and raccoons-can recombine, contributing to the broad host range and adaptability of CDVs. Our study demonstrates that the historic Rockborn lineage is actively participating in this process, dispersing genome fragments of the original strain into other CDV strains over the long term via recombination. The mosaic genome structure observed, with parental contributions from strains infecting distantly related hosts like masked civets and domestic dogs, exemplifies the cross-species nature of these evolutionary events. In this case, the masked civet strain HN19 was closely related to the canine strain R252 across the genome, but the F and H genes, which were likely derived from strain Rockborn. These genes encode key proteins involved in cell-to-cell viral spread (F protein) and receptor interactions (H protein) [7][8][9][10][11]. Understanding the effects of the exchange of cognate genome regions is challenging and requires multiple studies using reverse genetics, in vitro or in vivo experiments, or both, for functional validation. Also, sampling bias and the heavy reliance on publicly available GenBank sequences might affect our analysis or the findings of analogous studies.
This work strongly emphasizes the limitations of relying on single-gene analyses for CDV genotyping and diagnostics. The H gene has traditionally been a target of phylogenetic studies due to its high variability and its role in immunity. Consequently, many diagnostic assays are designed to differentiate wild-type and vaccine strains based on specific polymorphisms within the H gene, or alternatively, other genomic regions [16,[33][34][35][36][37][38]. However, the occurrence of recombination means that a virus can possess a gene from one lineage/sub-lineage while other parts of its genome originate from another lineage/sublineage. For example, the recombination event identified in this study appeared to affect the F and H genes, suggesting that any diagnostic or phylogenetic analysis based solely on the H gene would have masked and mischaracterized the recombinant virus. This risk has been pointed out in other whole-genome studies. Budaszewski and co-workers emphasized that genotyping based on partial genomic sequences could lead to misidentification and that complete genome analysis is necessary to detect recombination hot spots, which they frequently found in the F and H genes [32]. Likewise, Yuan and co-workers identified recombination events primarily in the P and L genes, regions often overlooked in standard genotyping [31]. Therefore, relying on single-gene assays in a landscape shaped by recombination can lead to flawed epidemiological conclusions and ineffective control strategies.
Vaccine-related disease has been reported on several occasions in dogs, but sporadically. In most cases, however, it has been difficult to establish whether the disease was induced by the vaccine virus or by field viruses infecting pups shortly before or after vaccine administration. This issue could not be ruled out definitively in earlier studies, as appropriate diagnostic tools were not available. Using a genotyping PCR assay that distinguishes between locally relevant CDV genotypes and vaccine viruses, all suspected cases of CDV vaccine-induced illness or death analyzed during a 2-year surveillance in Italy were caused by CDV field strains [16], suggesting that vaccine-induced residual virulence is rare. Optimized diagnostic tools capable of distinguishing between field and vaccine CDV strains could help better assess the magnitude of this phenomenon.
## 5. Conclusions
In conclusion, the full-genome characterization of the Rockborn strain not only fills a critical gap in our understanding of this historic lineage but also reveals its ongoing role in shaping the complex evolutionary dynamics of CDV through recombination. This underscores the need to shift towards whole-genome approaches for effective surveillance, accurate diagnostics, and a comprehensive understanding of CDV epidemiology and evolution.
## Supplementary Materials:
The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/vetsci13010081/s1, Table S1: Functions of canine distemper virus genes.
## References
1. Sourimant, Plemper (2016) "Organization, Function, and Therapeutic Targeting of the Morbillivirus RNA-Dependent RNA Polymerase Complex" *Viruses*
2. Barrett, Shrimpton, Russell (1985) "Nucleotide Sequence of the Entire Protein Coding Region of Canine Distemper Virus Polymerase-Associated (P) Protein MRNA" *Virus Res*
3. Röthlisberger, Wiener, Schweizer et al. (2010) "Two Domains of the V Protein of Virulent Canine Distemper Virus Selectively Inhibit STAT1 and STAT2 Nuclear Import" *J. Virol*
4. Siering, Sawatsky, Pfaller (2021) "Protein Is Essential for Canine Distemper Virus Virulence and Pathogenicity in Ferrets" *J. Virol*
5. Dietzel, Anderson, Castan et al. (2011) "Canine Distemper Virus Matrix Protein Influences Particle Infectivity, Particle Composition, and Envelope Distribution in Polarized Epithelial Cells and Modulates Virulence" *J. Virol*
6. (2026) *Vet. Sci*
7. Anderson, .; Von Messling (2008) "Region between the Canine Distemper Virus M and F Genes Modulates Virulence by Controlling Fusion Protein Expression" *J. Virol*
8. Von Messling, Cattaneo (2002) "Amino-Terminal Precursor Sequence Modulates Canine Distemper Virus Fusion Protein Function" *J. Virol*
9. Plattet, Cherpillod, Wiener et al. (2007) "Signal Peptide and Helical Bundle Domains of Virulent Canine Distemper Virus Fusion Protein Restrict Fusogenicity" *J. Virol*
10. Pratakpiriya, Seki, Otsuki et al. (2012) "Nectin4 Is an Epithelial Cell Receptor for Canine Distemper Virus and Involved in Neurovirulence" *J. Virol*
11. Alves, Khosravi, Avila et al. (2015) "SLAM-and Nectin-4-Independent Noncytolytic Spread of Canine Distemper Virus in Astrocytes" *J. Virol*
12. Zhao, Ren (1520) "Multiple Receptors Involved in Invasion and Neuropathogenicity of Canine Distemper Virus: A Review" *Viruses*
13. Maclachlan, Dubovi, Fenner et al. (2011) "Fenner's Veterinary Virology"
14. Rima, Balkema-Buschmann, Dundon et al. (2019) "ICTV Virus Taxonomy Profile: Paramyxoviridae" *J. Gen. Virol*
15. Blixenkrone-Möller, Svansson, Appel et al. (1992) "Antigenic Relationships between Field Isolates of Morbilliviruses from Different Carnivores" *Arch. Virol*
16. Iwatsuki, Tokiyoshi, Hirayama et al. (2000) "Antigenic Differences in the H Proteins of Canine Distemper Viruses" *Vet. Microbiol*
17. Martella, Elia, Lucente et al. (2007) "Genotyping Canine Distemper Virus (CDV) by a Hemi-Nested Multiplex PCR Provides a Rapid Approach for Investigation of CDV Outbreaks" *Vet. Microbiol*
18. Ke, Ho, Chiang et al. (2015) "Phylodynamic Analysis of the Canine Distemper Virus Hemagglutinin Gene" *BMC Vet. Res*
19. Rockborn (1958) "Canine Distemper Virus in Tissue Culture" *Arch. Gesamte Virusforsch*
20. Cornwell, Thompson, Mccandlish et al. (1988) "Encephalitis in Dogs Associated with a Batch of Canine Distemper (Rockborn) Vaccine" *Vet. Rec*
21. Gloyd (1397) "Vaccines Recalled" *J. Am. Vet. Med. Assoc*
22. Martella, Blixenkrone-Møller, Elia et al. (2011) "Lights and Shades on an Historical Vaccine Canine Distemper Virus, the Rockborn Strain" *Vaccine*
23. Appel (1978) "Reversion to Virulence of Attenuated Canine Distemper Virus In Vivo and In Vitro" *J. Gen. Virol*
24. Lanszki, Tóth, Schütz et al. (2022) "Complete Genomic Sequencing of Canine Distemper Virus with Nanopore Technology during an Epizootic Event" *Sci. Rep*
25. Martin, Murrell, Golden et al. (2015) "RDP4: Detection and Analysis of Recombination Patterns in Virus Genomes" *Virus Evol*
26. Tamukai, Minami, Kurihara et al. (2020) "Molecular Evidence for Vaccine-Induced Canine Distemper Virus and Canine Adenovirus 2 Coinfection in a Fennec Fox" *J. Vet. Diagn. Investig*
27. Shi, Zhang, Yu et al. (2021) "Insight Into an Outbreak of Canine Distemper Virus Infection in Masked Palm Civets in China" *Front. Vet. Sci*
28. Freitas, Leme, Saporiti et al. (2019) "Molecular Analysis of the Full-Length F Gene of Brazilian Strains of Canine Distemper Virus Shows Lineage Co-Circulation and Variability between Field and Vaccine Strains" *Virus Res*
29. Rätsep, Ojkic (2024) "Canine Distemper Virus Infection of Vaccinal Origin in a 14-Week-Old Puppy" *J. Vet. Diagn. Investig*
30. Gulliver, Taylor, Eames et al. (2025) "Investigation of Post-Vaccinal Canine Distemper Involving the Rockborn-like Strain in Nine Puppies in New Zealand" *N. Z. Vet. J*
31. Chare, Gould, Holmes (2003) "Phylogenetic Analysis Reveals a Low Rate of Homologous Recombination in Negative-Sense RNA Viruses" *J. Gen. Virol*
32. (2026) *Vet. Sci*
33. Yuan, Liu, Wang et al. (2017) "Homologous Recombination Is a Force in the Evolution of Canine Distemper Virus" *PLoS ONE*
34. Da Fontoura Budaszewski, Streck, Nunes Weber et al. (2016) "Wageck Canal, C. Influence of Vaccine Strains on the Evolution of Canine Distemper Virus" *Infect. Genet. Evol*
35. Sui, Sun, Shi et al. "Establishment and Evaluation of a Multiplex Real-Time RT-PCR for Quantitative and Differential Detection of Wild-Type Canine Distemper Virus from Vaccine Strains" *Heliyon*
36. Dong, Li, Zhu et al. (2015) "Detection and Differentiation of Wild-Type and Vaccine Strains of Canine Distemper Virus by a Duplex Reverse Transcription Polymerase Chain Reaction" *Iran. J. Vet. Res*
37. Si, Zhou, Wang et al. (2010) "Reverse Transcription-Nested Polymerase Chain Reaction for Detection and Differentiation of Wild-Type and Vaccine Strains of Canine Distemper Virus" *Virol. J*
38. Yi, Cheng, Xu et al. (2012) "Development of a Combined Canine Distemper Virus Specific RT-PCR Protocol for the Differentiation of Infected and Vaccinated Animals (DIVA) and Genetic Characterization of the Hemagglutinin Gene of Seven Chinese Strains Demonstrated in Dogs" *J. Virol. Methods*
39. Wilkes, Sanchez, Riley et al. (2014) "Real-Time Reverse Transcription Polymerase Chain Reaction Method for Detection of Canine Distemper Virus Modified Live Vaccine Shedding for Differentiation from Infection with Wild-Type Strains" *J. Vet. Diagn. Investig*
40. Liu, Liu, Tian et al. (2015) "Establishment of Reverse Transcription Loop-Mediated Isothermal Amplification for Rapid Detection and Differentiation of Canine Distemper Virus Infected and Vaccinated Animals" *Infect. Genet. Evol*
41. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12685349&blobtype=pdf | # Guiding Antiviral Cell Therapy Approaches with an Online Resource of Clinically Scored Epitopes, T-Cell Receptors, and B-Cell Receptors
Theresa Kaeuferle, Britta Eiz-Vesper, Andreas Moosmann, Uta Behrends, Michel Decker, Lilli Gutjahr, Josef Mautner, Florian Klein, Christoph Kreer, Mira Reger, Dirk Busch, Elvira D'ippolito, Florian Kohlmayer, Amrei Menzel, Semjon Willier, Britta Maecker-Kolhoff, Tobias Feuchtinger
## Abstract
Introduction: The clinical application of cell-based immunotherapies is a rapidly emerging field, and recent advances in gene therapy have opened up a new era of innovative treatment approaches. Introducing a specific T-cell receptor (TCR) against viral epitopes or chimeric antigen receptor (CAR) into T cells and effector cells allows reprogramming of their specificity and utilization for advanced therapeutic applications in infectious diseases and virus-induced malignancies. Many technologies have been developed to genetically engineer T cells, and existing databases in silico predict or describe identified viral epitopes, TCRs, or B-cell receptors (BCRs). However, their therapeutic application is still hampered by limited knowledge on their clinical impact. Methods: An open-access online resource was developed, integrating a data-mining algorithm scoring the epitopes, TCRs, and BCRs (ETB database) according to clinical evidence. Results: We hereby present a new level of clinical evidence-based knowledge transfer for Britta Maecker-Kolhoff and Tobias Feuchtinger contributed equally to this work.
## Introduction
Advances in hematopoietic stem cell transplantation (HSCT) and solid organ transplantation have yielded improved survival benefits for patients suffering from (malignant) hematopoietic diseases, inborn errors of immunity, or end-stage organ failure. On the downside of delayed immune reconstitution and medical immunosuppression viral infections from persistent or newly acquired viruses pose a significant risk for severe or life-threatening complications [1][2][3][4]. For patients resistant to antiviral therapy, the transfer of virusspecific T cells (VSTs) has emerged as a promising new treatment approach for many viral entities [5,6]. In addition, a relevant subset of patients suffers from therapy-limiting side effects of antiviral treatments and is therefore eligible for antiviral T-cell therapies. Various techniques of cell product manufacturing from either stem cell donor or (partially) HLA-matched third party donors have been developed using in vitro cell expansion after antigenic stimulation or direct isolation following cognate antigen recognition [7][8][9][10][11]. All strategies have in common that potential donors must have been previously exposed to the virus to form a memory T-cell response (usually referred as "seropositive" donors). So far, strategies aiming at producing therapeutic VSTs from naïve donors have been laborious and of limited efficacy. To circumvent the need of seropositive donors, strategies have been developed to identify [12][13][14][15] and introduce specific T-cell receptor (TCR) sequences recognizing viral peptide epitopes in the context of HLA molecules or chimeric antigen receptors (CARs) binding to surface protein structures of infected cells using various techniques of genetic modification [16][17][18][19][20][21].
All strategies are based on the knowledge of protective TCR or B-cell receptor (BCR) sequences to be used to target effector cells to virus-infected patient cells. The knowledge of viral epitopes has increased dramatically in recent years, leading to difficulties in data overview for the individual researcher and/or clinician. Several databases have been developed recently to (1) identify immunogenic T-cell epitopes in silico, (2) share cognate TCR sequences, and (3) summarize knowledge on neutralizing antibodies and BCR sequences. While the existing databases are comprehensive and highly useful, they do not categorize epitope or immunoreceptor information based on information about clinical aspects such as control of infection in vivo, remission or protection from infection after immunotherapy. For example, TCRdb contains a large number of TCR sequences from a variety of clinical samples and offers flexible search and comparison options, but clinically relevant information on function and specificity of individual TCRs is limited [22]. VDJdb is a curated resource focused on specific TCRs, but it also lacks detailed information on features relevant for clinical application [23,24]. Currently, the Immune Epitope Database and Tools (IEDB) is the most comprehensive source of Tand B-cell epitopes, 3D structures, TCRs and BCRs, their in vitro characterization and it also links epitopes to TCRs and BCRs (IEDB; www.iedb.org) [25,26]. Epitopes and immunoreceptors are provided with descriptions of assays used to characterize them, with a binary evaluation of positive or negative outcome of these assays, and to literature references. However, linking the published knowledge on epitopes, TCR, and BCR sequences to clinically meaningful data on in vivo immunogenicity, effectiveness, or protection remains challenging.
Therefore, we set out to develop an open-source database that links viral proteins, virus-derived epitopes, cognate TCR and BCR sequences with information on clinical context and therapeutic relevance. Based on information on how the effectiveness was demonstrated (in vitro, in vivo model, in vivo human [natural], or in vivo human [clinical trial]), an algorithm was developed to provide the user with a score reflecting stages of clinical evidence and development as well as protective capacity. We expect application in future orthotopic TCR and CAR-T-cell development and facilitate the design of individual personal antiviral T-cell therapeutics.
## Methods
## Database Programming
We implemented a robust and scalable architecture utilizing various technologies for the web application. The front end was developed using Angular, a framework for building dynamic and responsive user interfaces. On the backend, we implemented a Java-based solution to provide RESTful services to communicate between the client and server. Apache Tomcat was the application server, providing the environment for running the backend application. For Object-Relational Mapping (ORM), we used Hibernate, which facilitates interaction between the Java objects and the underlying MySQL database. Authentication and authorization were managed using the Clinically Scored Epitopes, TCRs, and BCRs: The DZIF ETB Database Shiro framework, ensuring secure access control. To enhance portability and consistency across different environments, the entire application was packaged and deployed using Docker containers.
## Analyses of Database Content
Data were extracted from the database and analyzed via Microsoft Excel (Microsoft Corporation, Redmond, Washington, USA). Frequencies were determined by dividing the absolute number of all epitopes, TCRs, or BCRs with the respective information available. Results were graphically illustrated via GraphPad Prism (GraphPad Inc., San Diego, California, USA).
## Results
## Redirecting T-Cell Specificities for Therapeutic Application
Novel developments in site-directed genetic engineering technologies and large-scale identification of T-cell antigens circumvent the need for virusexperienced donors and allow for adoptive T-cell transfer approaches in naïve donors. Therefore, in the first step, protective TCR or BCR sequences can either be isolated from expanding immune cell populations correlating with a patient's clearance of infection or from protected healthy individuals' memory T-cell populations (Fig. 1). Recent progress in single-cell sequencing technologies and large-scale T-cell antigen decoding allow the setup of comprehensive epitope, TCR, and BCR databases. From those, either protective BCR sequences can be used to generate CARs or protective TCRs can be selected, both to be introduced into a T cell to redirect its specificity (Fig. 1). Thereby, antigenspecific T-cell products for therapeutic application can be manufactured by redirecting naïve donors' T cells.
## Database Structure
The ETB database has been set up to harbor epitopes, TCR, and BCR sequences linked to the respective clinical data relevant for selection for adoptive T-cell therapy (ACT), such as HLA restrictions or evidence levels. The ETB database web page provides users with two options: the home page and the search page. The home page summarizes background information on the database, scientific background, and information on the calculation of the clinical score. The search interface has been pretested to be clear and intuitive and allows searching by the basic elements pathogen, MHC molecule or antigenic protein (Fig. 2a,c). Additionally, the search can be further refined by specific epitope, TCR, or specific BCR sequences. A "finder" feature automatically assists in selecting the respective elements from a drop-down list in order to unify the highly variable nomenclature of the search elements. Ticking the additional box "Show only clinically confirmed results" allows further filtering of the results for human clinical application from patients with a treatment response.
The results interface is updated simultaneously during entry and contains four tabs: the epitope, TCR, BCR, and studies tab (Fig. 2b,c). The fields displayed in the epitope tab include the sequence and the source of the respective epitope in terms of virus and viral protein, the presenting MHC molecule, the score of the epitope, and a link to the epitope-related studies of the studies tab. The TCR tab displays the fields α chain and β chain sequences of the TCRs fulfilling the search criteria, the related epitope, the presenting MHC molecule, the score, and a link to the specific TCR-related studies of the studies tab. Similarly, the BCR section includes the antibody name, heavy chain and light chain sequences, the related epitope, the score, and the link to the BCR-related studies of the studies tab. Users can sort the results list by score by clicking the column header. Entering the list of studies via a specific Clinically Scored Epitopes, TCRs, and BCRs: The DZIF ETB Database epitope, TCR, or BCR link enables additional filters for the studies related to this specific epitope, TCR, or BCR. Entering the list of studies via the studies tab directly results in a complete list of studies fulfilling the search criteria from the search interface. The studies tab displays the title, the PMID, and the Clinical Trials registry identifier, the year, the authors, the score, and the type of the study.
In order to browse details from a clinical study, users can click on the arrow icon, which will open a dropdown menu with general information on the study, such as clinical phase and endpoints. Figure 2c In case the publication does not provide an element, the respective fields are not displayed for the user to keep the interface simple.
## Database Content
The ETB database captures data from systematic literature searches as well as data from direct submissions by partners of the epitope identification project consortium. To date (February 2, 2025), 538 epitopes, 141 TCRs, and 36 BCRs are included (Fig. 3a), but the database is continuously extended. The database's epitopes cover a broad range of MHC class I and MHC class II HLA allotypes (Fig. 3b,c). Most frequent HLA restrictions of the database's epitopes are A*02:01 (12%), B*08:01 (9%), B*07:02, DRB1*13:01 (5% each), A*03:01, B*35:01, and DRB1*0101 (each 4% of all database's epitopes), making up nearly half of all epitopes (Fig. 3b). Of the database's TCR sequences, 29% also cover A*02:01 restrictions, but TCR sequences most frequently cover B*35:01 (31%). Further, each 16% of the database's TCR sequences are restricted to B*07:02 and A*01:01 and 1.4% to B*08:01 (Fig. 3b).
The database covers Epstein-Barr virus (EBV), adenovirus (AdV), human herpesvirus 6 (HHV-6), cytomegalovirus (CMV), severe acute respiratory syndrome coronavirus-2, and JC virus epitopes, TCRs, or BCRs. Of epitopes, 61% were identified from EBV, 18% from CMV, 8% each from AdV and HHV-6, 3% from severe acute respiratory syndrome coronavirus, and 2% from JC virus antigens (Fig. 3d,left). Most of the database's TCRs target CMV (85%), followed by EBV (12%), AdV and HHV-6 (each 1%; Fig. 3d,middle). All BCRs included in the database target CMV (Fig. 3d,right).
## Scoring Algorithm
To select the most suitable TCR or BCR candidate for a patient's individual ACT product, the immune receptor, and epitope sequences are not only linked to the clinically relevant data but also clinically scored. Therefore, a default epitope and receptor scoring algorithm has been designed to guide the user to the clinically relevant and valid data. According to the availability and evidence level of clinical data, the epitopes, TCRs, or BCRs are scored in A 1-4, B or C 1-3 (Fig. 4a): An epitope, TCR, or BCR is scored C after identification with or without in vitro data available. Thereby, C1 stands for a descriptive report on the identification of its sequence only, and C2 for additional in vitro data demonstrating the binding of the isolated TCR/BCR to its epitope in an assay system independent of the assay for identification. C3 stands for additional availability of functional activity, such as cytokine release or cytotoxicity. An epitope, TCR, or BCR is scored B as soon as it has been reported in a nonhuman individual responding to viral infection in vivo. An A score stands for the identification in a human individual responding to viral infection in the context of immunotherapy or remission of acute disease. Depending on being reported in a case report, phase I clinical study, phase II clinical study, or phase III clinical study, it is classified as A1 to A4, respectively (Fig. 4a).
Applying the algorithm to the epitopes results in 54% with high clinical evidence level scores, 29% A1, 6% A2, and 19% A3 score. Of the epitopes, 44% show C1 and 2% show C3 in vitro evidence level (Fig. 4b). Of TCR sequences, 98% show in vitro confirmation scores, whereas 2% have been confirmed in clinical trials (Fig. 4c). BCRs consist of two-thirds (67%) C2 scored and one-third C1 scored BCR sequences (Fig. 4d). Neither of the database's epitopes, TCRs, or BCRs were functionally confirmed in nonhuman in vivo studies (score B). Consequently, the studies are scored A1-A3 (22,19, and 20%, respectively) and C1-C3 (26, 7, and 6%, respectively, Fig. 4e).
## Discussion
ACT with VSTs after allogeneic HSCT represents a rare clinical indication that emerged to routine application of advanced pathogen-specific cell therapy [5,6]. Recent developments in site-directed genetic engineering technologies opened a new era of reprogramming T cells from naïve donors for therapeutic application [15][16][17][18][19]21]. The complex nature of an antigen-specific T-cell therapy makes it cumbersome to identify clinically relevant specificities. Nevertheless, the intentional specificity of cell-based immunotherapies is essential for efficiency, specificity, long-term maintenance of protection, and low risk of side effects.
For various viruses, decades of research have established a body of knowledge about major epitopes in a number of antigens and TCRs targeting them in specific HLA-I context [22][23][24][25][26]. However, typically, this information is most extensive for HLA allotypes that are of high frequency in Western countries and for viral proteins that contain immunodominant epitopes presented by such HLAs. In addition, available data are focused on viral types and strains that are most prevalent or have historically been designated as representative laboratory strains. Studies on virus-specific TCR repertoires and identification of epitope-specific TCR sequences have mostly been limited to a subset of those epitopes that fulfill abovementioned conditions. Therefore, major challenges in this field are (1) the characterization of epitopes for additional relevant antigens, HLA restrictions, and pathogens and the immunoreceptors targeting them, (2) identification of those epitopes and epitope-specific TCRs or BCRs that promise to confer the best protection in the absence of toxicity, and (3) organization of this ever enlarging body of knowledge in a form that makes it readily applicable for disease prediction, immunomonitoring, and cellular therapy. Ultimately, clinically meaningful spontaneous While the human cellular immune response to pathogens is highly individualized, it nonetheless shows characteristic, largely conserved patterns for epitopes and corresponding TCRs. For example, many of the well-characterized CMV CD8 + and CD4 + T-cell epitopes produce a lifelong T-cell response in a majority of CMV carriers of the appropriate HLA type [27][28][29], and even the sequence of TCRs specific for a given epitope is often entirely or largely conserved among a majority of carriers [30][31][32][33], giving rise to a considerable degree of "publicness" of the specific T-cell response. It has not been possible to date to predict the epitope specificity of TCRs in the absence of prior empiric information about specific epitope/TCR sequence correlations [34], but progress in this area might be anticipated in the near future, especially if the state of information on specific TCRs is becoming available in comprehensive, qualitychecked, scored datasets.
While existing databases are expanding, comprehensive and highly useful, they do not categorize epitope or immunoreceptor information based on information about clinical aspects such as control of infection in vivo, or remission or protection from infection after immunotherapy [22][23][24][25][26]. The ETB database presented here is designed and intended to facilitate this information derived from clinical translation and applicability. It introduces a scoring method that places particular emphasis on two aspects: first, the context of epitope, TCR, or BCR identification (in vitro, nonhuman primates, in humans); second, the degree of evidence for functional and/or protective relevance according to the respective context. In this respect, epitopes verified in clinical trials to be immunogenic and protective reach the highest scores in this database.
Taken together, our present categorization and scoring system represents a clear and simple framework for evaluating epitopes and immunoreceptors and estimating their effectiveness in clinical application. Thereby, it fills an important gap in available databases. Within a pilot testing phase, feedback from early test users -including clinicians and researchers in the field -was implemented to enhance usability and user experience. Improvements include a color-coded score display (ranging from red for low to green for high clinical relevance and validity), with entries lacking a score automatically moved to the bottom of the table. Further, each entry in the epitopes or receptors tab now provides a quick reference button to the respective citation details in the Studies tab. Moreover, a filter option has been added to display only clinically validated entries, allowing users to focus on translationally relevant data. Additionally, a background tab was introduced to offer broader context and background, supporting users with varying levels of expertise. We intend to refine and enhance our evaluation approach in further versions of our database. Further improvements will include specific evaluation and scoring of the degree of quantitative and temporal association of particular epitope-specific responses and immunoreceptors with remission and long-term suppression of infection in the context of immunotherapy and immune reconstitution in patients after transplantation and patients with other types of immunodeficiencies.
In the future, automated and artificial intelligencebased data extraction and evaluation will compete with manual curation in collecting and evaluating such complex data. In this regard, a combination of automated data collection, manual curation, and expert evaluation will help make the best clinical use of the increasing amount of information in this expanding field. Interfaces with other specialized databases may broadly expand and enhance the available knowledge on clinically meaningful immune responses in viral infection.
## References
1. Hill, Mayer, Xie et al. (2017) "The cumulative burden of double-stranded DNA virus detection after allogeneic HCT is associated with increased mortality" *Blood*
2. Sommerer, Schröter, Gruneberg et al. (2022) "Incidences of infectious events in a renal transplant cohort of the German center of infectious diseases (DZIF)" *Open Forum Infect Dis*
3. Fishman (2017) "Infection in organ transplantation" *Am J Transplant*
4. Ferdjallah, Young, Macmillan (2021) "A review of infections after hematopoietic cell transplantation requiring PICU care: transplant timeline is key" *Front Pediatr*
5. Schweitzer, Muranski (2024) "Virus-specific T cell therapy to treat refractory viral infections in solid organ transplant recipients" *Am J Transplant*
6. Kaeuferle, Krauss, Blaeschke et al. (2019) "Strategies of adoptive T -cell transfer to treat refractory viral infections post allogeneic stem cell transplantation" *Februar*
7. Prockop, Hasan, Doubrovina et al. (2023) "Third-party cytomegalovirus-specific T cells improved survival in refractory cytomegalovirus viremia after hematopoietic transplant" *J Clin Investig*
8. Tzannou, Papadopoulou, Naik et al. (2017) "Off-the-Shelf virus-specific T cells to treat BK virus, human herpesvirus 6, cytomegalovirus, epstein-barr virus, and adenovirus infections after allogeneic hematopoietic stem-cell transplantation" *J Clin Oncol*
9. Toner, Mccann, Bollard (2024) "Applications of cell therapy in the treatment of virus-associated cancers" *Nat Rev Clin Oncol*
10. Ptak, Meyers, Radko-Ganz et al. (2025) "A phase I study evaluating safety and tolerability of viral-specific T cells against BK-Virus in adult kidney transplant recipients" *J Med Virol*
11. Dadwal, Bansal, Schuster et al. (2024) "Final outcomes from a phase 2 trial of posoleucel in allogeneic hematopoietic cell transplant recipients" *Blood Adv*
12. Welters, Stadler, Anastasopoulou et al. (2024) "Oligonucleotide library screening for identification of virus-specific T-cell receptors" *Eur J Immunol*
13. Carr, Mateyka, Scheu et al. (2024) "Advances in preclinical TCR characterization: leveraging cell avidity to identify functional TCRs" *Biol Chem*
14. Khan, Müller, Kasmapour et al. (2024) "Dynamic monitoring of viral gene expression reveals rapid antiviral effects of CD8 T cells recognizing the HCMV-pp65 antigen" *Front Immunol*
15. Cobo, Welters, Rosenberger et al. (2022) "Rapid single-cell identification of epstein-barr virus-specific T-cell receptors for cellular therapy" *Cytotherapy*
16. Dragon, Zimmermann, Nerreter et al. (2020) "CAR-T cells and TRUCKs that recognize an EBNA-3C-derived epitope presented on HLA-B*35 control epstein-barr virus-associated lymphoproliferation" *J Immunother Cancer*
17. Schub, Schuster, Hammerschmidt et al. (2009) "CMV-Specific TCRtransgenic T cells for immunotherapy" *J Immunol*
18. Norberg, Hinrichs (2023) "Engineered T cell therapy for viral and non-viral epithelial cancers" *Cancer Cell*
19. Schober, Müller, Busch (2020) "Orthotopic T-Cell receptor Replacement-An "enabler" for TCR-based therapies" *Cells*
21. Ma, Chen, Du et al. (2024) "Adoptive transfer of CMV-specific TCR-T cells for the treatment of CMV infection after haploidentical hematopoietic stem cell transplantation" *J Immunother Cancer*
22. Kaeuferle, Zwermann, Stoll et al. (2025) "All-inone CRISPR/Cas-engineered glucocorticoidreceptor knock-out EBV-gp350-CAR knockin T cells are potent and resistant to dexamethasone" *Exp Hematol Oncol*
23. Chen, Yue, Lei et al. (2021) "TCRdb: a comprehensive database for T-cell receptor sequences with powerful search function" *Nucleic Acids Res*
24. Bagaev, Vroomans, Samir et al. (2020) "VDJdb in 2019: database extension, new analysis infrastruc-ture and a T-cell receptor motif compendium" *Nucleic Acids Res*
25. Shugay, Bagaev, Zvyagin et al. (2018) "VDJdb: a curated database of T-cell receptor sequences with known antigen specificity" *Nucleic Acids Res*
26. Blazeska, Kosaloglu-Yalcin, Vita et al. (2023) "IEDB and CEDAR: two sibling databases to serve the global scientific community" *Methods Mol Biol*
27. Martini, Nielsen, Peters et al. (2020) "The immune epitope database and analysis resource program 2003-2018: reflections and outlook" *Immunogenetics*
28. Nastke, Herrgen, Walter et al. (2005) "Major contribution of codominant CD8 and CD4 T cell epitopes to the human cytomegalovirusspecific T cell repertoire" *Cell Mol Life Sci*
29. Pachnio, Ciaurriz, Begum et al. (2016) "Cytomegalovirus infection leads to development of high frequencies of cytotoxic virus-specific CD4+ T cells targeted to vascular endothelium" *PLoS Pathog*
30. Klenerman, Oxenius (2016) "T cell responses to cytomegalovirus" *Nat Rev Immunol*
31. Huisman, Hageman, Leboux et al. (2022) "Public T-Cell Receptors (TCRs) revisited by analysis of the magnitude of identical and highly-similar TCRs in virusspecific T-Cell repertoires of healthy individuals" *Front Immunol*
32. Klarenbeek, Remmerswaal, Ten Berge et al. (2012) "Deep sequencing of antiviral T-cell responses to HCMV and EBV in humans reveals a stable repertoire that is maintained for many years" *PLoS Pathog*
33. Dössinger, Bunse, Bet et al. (2013) "MHC multimer-guided and cell culture-independent isolation of functional T cell receptors from single cells facilitates TCR identification for immunotherapy" *PloS One*
34. Huth, Liang, Krebs et al. (2019) "Moosmann A. Antigen-specific TCR signatures of cytomegalovirus infection" *J Immunol*
35. Glanville, Huang, Nau et al. (2017) "Identifying specificity groups in the T cell receptor repertoire" *Nature* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754530&blobtype=pdf | # Human immunodeficiency virus recency testing coverage and partner-notification-services among people-living with human immunodeficiency virus in low-and middle-income countries
Ibrahim El-Imam, Timothy Peter, Hassan Fussi, Zuhura Ally, Mhando Hafidha, Mariam Bakari, Upendo Mbwana, Beatrice Kayeke Chenya, Haji Mpimo, Habib Ally, Ramadhani, Hafidha Bakari, Mariam Mbwana, Upendo Chenya, Beatrice Mpimo, Haji Ally, Habib Ramadhani
## Abstract
BACKGROUNDHuman immunodeficiency virus (HIV) recency testing provides data that can be used to monitor the trend of new HIV infections. The effectiveness of using people identified with recent infection to identify partners with new HIV infection through partner notification services (PNS) is not well documented.
AIMTo determine the pooled prevalence of recency testing coverage, recent infection, reclassification (recent to longterm infection) and PNS cascade among newly diagnosed people living with HIV.
METHODSPubMed, Cochrane Library and Embase were searched for articles published between January 2018 and November 2024. Studies were included if they reported recency coverage and/or PNS among people newly diagnosed with HIV and used recent infection testing algorithm (RITA). Recency coverage was defined as proportion of people tested using rapid testing for recent infection (RTRI) among those newly diagnosed with HIV. RITA further classifies RTRI results using viral load results (≥ 1000 copies/mL vs < 1000 copies/mL) to confirm recency status. For studies with PNS, we evaluated the cascade: Number of partners elicited, successfully contacted, eligible for HIV testing, tested and HIV diagnosis. PNS effectiveness was measured by proportion of new HIV diagnoses from tested partners. Using random effects models, we computed the pooled estimate of recency outcomes and 95% confidence intervals (CIs).
RESULTSTwenty-five studies from 17-low-and middle-income countries were included. Of 276315 newly diagnosed people living with HIV, 79864 underwent RTRI with an overall pooled recency coverage of 87% (95%CI: 67-96). The pooled prevalence of RTRI and RITA recency were 12% (95%CI: 9-16) and 7% (95%CI: 4-10), respectively. Pooled prevalence of RTRI reclassification was 34% (95%CI: 22-49). Of the recent cases who agreed to PNS, 253 partners were elicited with an estimated elicitation ratio of 1:1.6. Among partners elicited, 99% were successfully contacted, 75% were eligible for testing, 68% tested for HIV, and 15% were diagnosed with HIV.
CONCLUSIONHigh recency testing coverage among newly diagnosed individuals demonstrates the feasibility of monitoring new HIV infections in LMIC. While PNS yielded moderate HIV diagnoses, its targeted approach remains a critical strategy for identifying undiagnosed cases.
## INTRODUCTION
The global human immunodeficiency virus (HIV) epidemic remains a major public health challenge, with an estimated 39.9 million people living with HIV (PLWH) in 2022, disproportionately affecting low-and middle-income countries (LMICs) [1]. Although progress has been made towards the UNAIDS 95-95-95 targets -improving diagnoses, expanding antiretroviral therapy (ART) coverage, and achieving viral suppressions -significant gaps persist, particularly in the timely diagnosis of new infections and interrupt ongoing transmission [1][2][3]. Strengthening HIV surveillance and implementing targeted interventions are essential to identifying new infections and halting transmission, especially in high-incidence settings where health care access and prevention programs face systematic challenges.
HIV recency testing, designed to differentiate recent infection (acquired within the past 6-12 months) from longstanding infections, offers a promising tool to address these challenges [4][5][6][7]. By identifying individuals with recent infection at diagnosis, recency testing enabled public health program to target sub-populations and geographical areas where transmission is most active, thereby informing more effective public health intervention [3,4,8,9]. When integrated with partner notification services (PNS), recency testing can amplify the detection of undiagnosed infections, interrupt transmission chain and accelerate progress towards epidemic control [10][11][12].
Several LMIC's have introduced recent infection surveillance within routine HIV testing services, often as part of national case-based surveillance systems [13][14][15]. However, the success and utility of recency-informed strategies depend heavily on achieving adequate recency testing coverage, accurately estimating the proportion of recent infections, and applying confirmatory tests to reduce misclassification. These operational metrics vary widely across programs and settings [13,14]. The World Health Organization (WHO) recommends using a recent infection testing algorithm (RITA), which combines a rapid test for recent infection (RTRI) with supplementary viral load testing to minimize false recent results and ensure reliable estimates, particularly in LMIC's where ART coverage is expanding [4,6,10].
A critical application of recent testing lies in optimizing PNS outcomes [10,11]. Individuals with recent infections often have high viral loads and are unaware of their status, increasing their risk of transmitting HIV [8,13]. Identifying such index cases allows programs to prioritize partners elicitation and testing, leading to earlier diagnosis, linkage to care, and prevention intervention for partners [11,16,17]. This integration has been shown to improve efficacy along the entire PNS Cascade, from partner elicitation through HIV positivity yield [18,19], which is critical for maximizing limited resource in high-burden resource-constrained low-and middle-income countries (LMICs).
Despite increasing implementation of recency testing and PNS in LMIC's, no comprehensive synthesis has pooled quantitative outcomes across diverse settings. This systematic review and meta-analysis address this gap by estimating pooled prevalence of recency testing coverage, recent infection rates, reclassification from recent to long-term infection, and PNS cascade outcomes among newly diagnosed PLWH in LMIC. Our findings will inform national HIV strategies, guide targeted prevention programs, and optimize resource allocation in LMICS to achieve epidemic control.
## MATERIALS AND METHODS
## Registration
The protocol for this systematic review was registered in the International Prospective Register of systematic Reviews (PROSPERO) under registration number CRD420251081733. The review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and meta-Analysis (PRISMA) guidelines.
## Ethical approval
As this study involved secondary analysis of published studies and programmatic reports ethical approval was not required.
## Search strategy
We searched PubMed, EMBASE, and the Cochrane Library for studies published between January 1, 2018, and November 30, 2024, using combinations of keywords and MeSH terms related to HIV recency testing, the RITA RTRI and PNS including index testing, contact tracing, and elicitation. The search strategy focused on LMIC's and included only English-language publications. The databases search yielded 268 records. We also identified 13 additional records through manual searches of conference abstracts and grey literature sources, bringing the total to 281 records. The full search was conducted on May 13, 2025. We uploaded the records from electronic data bases into Rayyan software for deduplication and screening. Of the 281 records, we identified two duplicates manually removed one and ultimately included 280 records before initiating the screening process detail. Search strategies are provided in Supplementary Table 1.
## Eligibility criteria
We included studies that reported on any component of the HIV recency testing cascade or the PNS cascade among newly diagnosed PLWH in LMICs. Eligible recency cascade elements included initial testing for recent infection, classification, or reclassification to long-term infections, and confirmatory viral load results. Recency testing coverage was defined as the proportion of newly diagnosed individuals who received RTRI at the point of HIV diagnosis. Studies were required to have implemented RITA which combines an initial recency assay (such as RTRI, Lag avidity EIA, or another validated serological method) with confirmatory viral load testing (≥ 1000 copies/mL) in accordance with WHO guidance to be considered for reclassification from recent to long-term infection [4].
Studies were also eligible if they reported on PNS outcomes including the number of sexual/needle sharing partners elicited, successfully contacted, found eligible for testing, tested for, and diagnosed with HIV. Only studies conducted in LMICS among newly diagnosed PLWH were included. We excluded studies that did not report any relevant recency testing or PNS data, studies that were non-primary (e.g., reviews, editorials, commentaries), those conducted in ineligible population and those lacking sufficient quantitative data for extraction.
## Study selection
Two reviewers independently screened all 280 records for eligibility using Rayyan. After title and abstract screening, we excluded 238 records. We sought full texts for the remaining 42 records, all of which were successfully retrieved. We assessed these 42 full-text articles for eligibility and excluded 17 articles, 11 due to inclusion of the wrong population and six due to irrelevant outcomes.
At each stage of review, discrepancies were resolved through discussion and consensus. Ultimately, we included 25 studies in the final analysis. These included both peer-reviewed publication and high-quality conference abstracts that met all inclusion criteria despite the absence of full manuscripts. The PRISMA 2020 flow diagram (Figure 1) summarizes the selection process.
## Data extraction
Data were independently extracted by two reviewers using a pre-defined Excel template. Extracted data included study author and year of publication, country and setting, study design, and sample characteristics. Recency testing indicators included the number of newly diagnosed individual, number tested with RTRI, coverage of RTRI testing, number of cases reclassified from recent to long-term infection, and details of the reclassification process. For PNS, extracted outcomes include the number of partners elicited, successfully contacted, tested for HIV, found eligible for HIV testing, and diagnosed with HIV. PNS effectiveness was measured as the proportion of newly diagnosed HIV positive partners among those tested. Any discrepancies in data extraction were discoursed and resolved by consensus.
## Quality assessment
The methodological quality of included studies was assessed using the Joanna Briggs Institute (JBI) tools for observational studies. The tool consists of nine questions with four responses: (Yes, No, Not clear, Not applicable). We assigned a score of 1 to a "Yes" response and 0 to a "No" response. Each scored question was totaled and classified into three categories. Studies with (0-3), (4)(5)(6) and (7)(8)(9) scores were regarded as being of low, medium and high quality respectively. Two pairs of reviewers (Beatrice Kelvin Mpimo and Haji Mbwana Ally) and (Hassan Fredrick Fussi and Upendo Kayeke Chenya) independently performed and rated the quality of the studies using the JBI tools. Discrepancies of the scores between the two pairs were sorted by a third pair of reviewers (Habib Ramadhani Omari and Hafidha Mhando Bakari).
## Definition of variables
The primary outcomes of interests were recency testing coverage at point-of-care, prevalence of RTRI-recent infection, prevalence of RITA-recent infection, and the proportion of RTRI-recent cases reclassified as long-term infections based on confirmatory viral load results. Furthermore, we also reported the proportion of individuals successfully reached across the PNS Cascade including elicitation ratio, prevalence of partners of recent HIV cases who were successfully contacted, prevalence of partners of recent HIV cases who were eligible for HIV testing, prevalence of partners of recent HIV cases who tested for HIV, and prevalence of partners of recent HIV cases who were diagnosed with HIV. The secondary outcome was the effectiveness of partner notification, measured as the proportion of tested partners who were newly diagnosed with HIV.
## Statistical analysis
Using random and fixed effects models, we computed pooled prevalence of HIV recency testing uptake, RTRI-recent infection, RITA-recent and reclassification from recent to long-term HIV infections. Additionally, we also evaluated PNS cascade by quantifying several components of the cascade including elicitation ratio, prevalence of partners of recent HIV cases who were successfully contacted, prevalence of partners of recent HIV cases who were eligible for HIV testing, prevalence of partners of recent HIV cases who tested for HIV, and prevalence of partners of recent HIV cases who were diagnosed with HIV. Subgroup analysis on the pooled estimates of RTRI and RITA prevalence were performed to compare studies conducted from surveys/Laboratory samples vs those conducted from HIV programs using χ 2 tests. Studies heterogeneity was assessed by computing the I 2 statistic and Cochran's Q. The score values of I 2 statistics were categorized at 75%, 50% and 25% to signify the presence of high, moderate and low heterogeneity respectively as previously described [20]. To assess publication bias, the Egger regression asymmetry test was used. To declare the presence of either heterogeneity or publication bias, a P value threshold of < 0.05 was used. For the prevalence outcome that showed either moderate or high degree of heterogeneity, we assessed its possible sources by conducting an influential analysis using the leave-one-out method [21]. In addition, we conducted a meta regression analysis to discern the variation of the HIV recency testing uptake. Using trim-and-fill method, we conducted a sensitivity analysis to assess possible small-study. All statistical tests were performed using R version 4.2.2 (R Foundation for Statistical Computing, Vienna, Australia).
## RESULTS
Our search identified a total of 281 records comprising 268 from databases and 13 from manual sources. After deduplication, 280 unique records were screen by title and abstract using Rayyan. Of these, 238 records were excluded for not Flowchart illustrating the systematic review process. A total of 280 records were screened after removing 1 duplicate. Of these, 42 full text reports were assessed for eligibility, and 25 studies met the inclusion criteria. 17 reports were excluded due to wrong population (11) or outcome (6).
meeting the eligibility criteria. The remaining 42 records underwent full text review from which we excluded 17 additional studies (11 due to wrong population, 6 due to irrelevant outcomes). A total of 25 studies were included in the final analysis (Figure 1). These studies represent a diverse range of geographic regions within LMICs, and varied in study design, populations and covering various study periods. Collectively, thirteen peered-reviewed articles and twelve highquality conference abstracts contributed data on HIV recency testing and PNS outcomes.
## Study quality assessment
Assessment of study quality showed that all studies were of high quality, with scores ranging from 7-9. Those with scores less than 9, the common reasons were smaller sample size and a response rate of less than 80%. Neither study was of low nor medium quality (Table 1).
## HIV recency testing coverage
We included 11 studies reporting on HIV recency testing coverage using RTRI at point-of -care [8,12,14,16,[22][23][24][25][26][27][28]. The studies spanned diverse geographic and programmatic settings within low-and middle-income countries and represented a combined total of 276315 individuals newly diagnosed with HIV. Individual study estimates of RTRI coverage ranged from 18% to 100%, with substantial variation across studies. The pooled RTRI coverage using a random-effect model was 87% (95%CI: 67-96), indicating significant heterogeneity across studies (I 2 = 100%, τ 2 = 3.95, P < 0.001). By contrast, the fixed effect model yielded a lower pooled estimate of 29% (95%CI: 29-29), largely influenced by disproportionate large sample size in Truong et al[14], 2022, which alone contributed 83.7% of the total sample size (Figure 2). Despite the wide variability, a leave-one-out Sensitivity analysis for RTRI coverage revealed that the overall pooled estimates remain stable across all iterations, with minimal change in heterogeneity metrics (I 2 = 100%, τ 2 range = 3.00-4.35) suggesting that no single study disproportionately influenced the summary effect size (Supplementary Figure 1).
## Prevalence of RTRI-Recent HIV infection
We included 15 studies that reported the proportion of individuals testing recent for HIV infection using either RTRI, or other validated recency assays, such as limiting antigen avidity immunoassay (Lag-Avidity) [5,8,12,16,[22][23][24]27,[29][30][31][32][33][34][35]. These studies together represented a combined sample size of 49196 newly diagnosed PLWH across diverse, LMICs settings.
The pooled estimate from the random-effect model was 12% (95%CI: 9-16), while the fixed effect model produced a slightly lower estimate of 10% (95%CI: 10-11). The point estimate from individual studies varied widely, ranging from 5% to 4.3% (Figure 3). There was substantial heterogeneity between studies (I 2 = 98%. τ 2 = 0.4710, P < 0.01) suggesting considerable variability in testing approaches and populations. 1 Indicates studies that reported partner notification services only among recently infected people living with human immunodeficiency virus. To assess the robustness of this finding, we conducted a leave-one-out sensitivity analysis, which showed no significant deviation in the pooled estimates when each study was omitted in turn (range: 0.1 to 0.11; Supplementary Figure 2). This indicates that the summary estimate was not driven by any single influential study. We further stratified the analysis by data source, distinguishing between programmatic-(n = 8 studies) and laboratory/survey-based (n = 7 studies) data and the pooled estimates were 11% (95%CI: 8%-15%) and 14% (95%CI: 8%-23%) respectively. Nonetheless, the test for subgroup differences using the random effect model was not statistically significant (χ 2 = 0.43, df = 1, P = 0.51) (Table 2), indicating no meaningful differences between the two data sources. Thus, indicating a moderate but consistent burden of RTRI-recent infection among newly diagnosed individual in LMIC's with variation by study context other than data source.
## Prevalence of RITA-recent HIV infection
Eighteen studies reported on the prevalence of recent infection using a WHO-defined RITA, which incorporated an initial recency assay with a confirmatory viral load testing (≥ 1000 copies/mL) [5,8,12,16,[22][23][24][25][26][27][28][29][30][31][32][33][34][35][36]. These studies collectively evaluated 66675 individuals. RITA-recent prevalence from individual study ranged from 1% to 39%, with an overall pooled prevalence of 7% (95%CI: 4-10) from the random-effect model and 4 (95%CI: 0.04-0.04) from the fixed effects model. High heterogeneity among the included studies was observed (I 2 = 99%. τ 2 = 1.0079, P < 0.001) reflecting that the variation observed across studies could not be explained by chance alone (Figure 4).
To assess the robustness of this pooled estimate, we conducted an influential analysis using a leave-one-out approach (Supplementary Figure 3). The pooled prevalence remained consistently between 4% and 6%, regardless of which study was omitted, suggesting that no single study exerted undue influence on the overall estimate. Heterogeneity measures remain unchanged across iterations (I 2 = 99%, τ 2 = 1.0079) reinforcing the stability of the model. A subgroup analysis was further performed to explore whether the source of data contributed to the observed heterogeneity. Among the 7 programmatic studies the pooled prevalence was 6% (95%CI: 5-6), while the 7 Laboratory/survey-based studies yielded a slightly lower estimate of 5% (95%CI: 2-13) (Table 2). However, the test for subgroup differences using the random effect model failed to reach statistical significance (χ 2 = 0.60, df = 1, P = 0.44), implying that the data source alone does not explain the heterogeneity. Together, this analysis confirm that while the pooled prevalence of RITA-recent infection is approximately 7%, substantial variability exists across studies, likely due to differences in population characteristics, implementation fidelity and recency assay protocol.
## Prevalence of reclassification
Fourteen studies reported on the proportion of individuals initially classified as recent who were reclassified as long-term after confirmatory viral load testing. Together, these studies encompass a total of 2612 individuals identified as recent prior to viral load confirmation, with a reclassification rate ranging from 8% to 88% (Figure 5). The random-effect pooled estimate was 34% (95%CI: 21-50), while the fixed-effects model yielded a slightly higher and more precise estimate of 0.40 (95%CI: 38-41). Heterogeneity across studies was high (I 2 = 96%, τ 2 = 1.4863, P < 0.01) reflecting differences in ART exposure, testing timing, and viral load suppression.
To assess whether any single study disproportionately influenced the pooled estimates, we performed a leave-one-out sensitivity analysis (Supplementary Figure 4). The pooled reclassification estimates remain relatively stable, ranging between 38% and 42% when individual studies were excluded one at a time. All recalculated heterogeneity statistics remain high, reinforcing the robustness of the pooled estimates while also underscoring persistent heterogeneity in the data. These findings confirm that approximately one-third of the individuals initially identified as recent are ultimately reclassified as long-term infections. This high reclassification rate underscores the critical role of viral load confirmation in RITA and may signal gaps in identifying true recent infection in high-ART coverage settings.
## Prevalence of partner notification cascade
Proportion of partners successfully contacted: Five studies reported this indicator [17][18][19]28,37], with a pooled randomeffect estimate of 99% (95%CI: 67-100) and fixed-effects estimates of 96% (95%CI: 93-98). Heterogeneity was low (I 2 = 11%, τ 2 = 11.2974, P = 0.34) suggesting high and consistent success in partners contacts (Figure 6).
## Proportion of partners eligible for HIV testing:
The pooled random-effects estimates across five studies was 75% (95%CI: 59-86) (Figure 6), while the fixed effects estimates was 76% (95%CI: 70-0.81). Heterogeneity was moderate-to-high (I 2 = 84%, τ 2 = 0.5361, P < 0.01), likely reflecting differing definitions and population characteristics.
Proportion of eligible partners tested for HIV: Among partners eligible for testing, the pooled random-effect estimate was 68% (95%CI: 56-79) (Figure 6). The fixed effects estimates was 69% (95%CI: 63-75), with moderate heterogeneity (I 2 = 75%, τ 2 = 0.2507, P < 0.01).
## Prevalence of HIV among tested partners:
Across five studies, the pooled HIV prevalence among tested partners was 15% (95%CI: 10-22) (Figure 6) using a random-effects model and 15% (95%CI: 11-20) using a fixed effect-model. Heterogeneity was low (I 2 = 39%, τ 2 = 0.0906, P = 0.16).
Collectively, these results demonstrate variable performance across the HIV recency testing and PNS Cascades, with strong partner tracing outcome, but more modest testing uptake and case detection.
## DISCUSSION
We conducted a systematic review and meta-analysis to synthesize current evidence on the implementation and performance of HIV recency testing and its integration with PNS among newly diagnosed individuals in LMICs. Our findings provide critical insights into testing coverage, diagnostic accuracy, and programmatic impacts, while identifying key gaps that must be addressed to strengthen epidemic control efforts.
The high RTRI coverage of 87% across 11 studies demonstrates substantial integration of recency testing into routine HIV services in many LMICs. High RTRI uptake is crucial for implementing real-time surveillance and response to ongoing HIV transmission. However, the wide range in coverage estimates (range: 18%-100%) raise concerns about programmatic consistency and scalability. This finding is in line with earlier reports from programs in Kenya, Zimbabwe,Zambia,and Rwanda[10,12,16,24,28], where coverage was influenced by policy adoption, training quality and test kits availability. The results suggest that while national programs may report high RTRI uptake, subnational heterogeneity remains a key challenge. Programs must therefore strengthen coverage equity across districts, improving training and addressing logistical constraints to maximize the utility of recent infection surveillance.
Our meta-analysis reveals critical insights about HIV recency testing's utility and limitations in LMICs. The pooled prevalence of 12% based on RTRI results from 15 studies encompassing over 49000 newly diagnosed individuals across LMICs highlights a moderate but epidemiologically meaningful burden of likely incidents infections, consistent with reports from Kenya, Malawi and Rwanda (> 10%) [5,12,16,24]. However, when restricted to studies using the WHOrecommended RITA, the pooled prevalence declined to 7% across 18 studies with 66675 individuals. This 42% reduction demonstrates the critical role of viral load confirmation in excluding false-recent cases, and exposes RTRI vulnerability to misclassification, particularly in high-ART coverage settings where early treatments and rapid viral suppression are common [8,24,[29][30][31]33].
The discrepancy between RTRI and RITA estimates is further clarified by our pooled reclassification rate of 34% across 14 studies, including 2612 individuals, indicating that nearly one in three individuals initially classified as recent were ultimately deemed long-term after VL testing. These findings reinforce previous observations from Kenya, Malawi, Rwanda and Nigeria, where unconfirmed recency testing led to overestimation of recent infections and misdirected prevention efforts [8,24,[29][30][31]. High heterogeneity observed in all three analyses likely reflects contextual and operational variability, including diverse recency assays, ART coverage levels, differences in VL suppression rates, implementation fidelity of VL testing, and variation in data type (Facility vs community-based). Notably, our sensitivity analysis revealed that no single study unduly influenced the pooled estimates, and that there were no significant prevalence differences between programmatic and research settings. These findings further affirm that real-world data, when rigorously collected, can yield reliable surveillance estimates [10][11][12], which supports the WHO endorsement of recency testing for dynamic surveillance, but underscores that its accuracy hinges on confirmatory testing infrastructure [7,10].
For public health programs, these findings carry important implications -while RTRI provide an accessible tool for transmission hotspots identification, its stand-alone use possess substantial misclassification risks in high-ART-coverage setting potentially distorting resource allocation [28,38]. In contrast, despite RITA's superior accuracy, implementation barriers persist -particularly VL processing delays and decentralized testing infrastructures that hinder timely recency classification [30,39]. National programs must therefore prioritize fidelity to RITA protocols, ensuring timely VL-testing and results return and integrate recent infection surveillance with partner services and outbreak response mechanisms [40]. Emerging solutions like multi-assay algorithm (Sedia HIV recency assay) and machine learning approaches integrating clinical meta-data show promise but require further validation in programmatic context [6,41,42]. While limitations like residual misclassification and assay variability persist[43] a tiered RTRI-RITA approach with strengthen laboratory systems, offers LMIC's optimal balance of feasibility and accuracy for recency-based surveillance and epidemic control.
Our meta-analysis highlights both the strengths and persistent gaps in PNS implementation across LMIC, with implications that align closely with insights from the HIV recency cascade. The near-universal success in Partner contacts, 99% demonstrates the consistent feasibility of index-case-based approaches across diverse settings, mirroring documented successes in Rwanda and Kenya where provider-assisted notification achieved > 85% contact rates [10,11,28,37]. However, significant cascade attrition emerges at subsequent stages, where only 75% of contacted partners met eligibility criteria and merely two-third of eligible partners completed HIV testing. The moderate-high heterogeneity in testing eligibility and uptake likely reflects operational variations in partner definitions, consent procedures, and persistent structural barriers including stigma and patient mobility -challenges consistently identified in comparable LMIC implementation [44][45][46].
The substantial 15% HIV prevalence among tested partners -nearly triple typical general population testing yields [47][48][49], confirms PNS as a high value case-finding strategy. This prevalence closely corresponds with our finding of 12% RTRI-recent infections among index cases, suggesting PNS effectively captures active transmission networks. However, the 34% RTRI-RITA reclassification rates introduce a critical programmatic consideration: Partners of virally suppressed index cases (likely long-term infection) may represent established rather than acute infections, potentially diluting PNS's outbreak interception value [5,17,30]. To optimize impact, programs should prioritize integrating RTRI-screening with confirmatory RITA testing to strategically direct PNS resources towards truly viremic index cases most likely to yield recent partner infection [10,11]. Concurrently, standardizing eligibility criteria and testing protocols could substantially reduce the observed cascade attrition while improving cross-program comparability. Finally, given the elevated transmission risk associated with recent infections, strengthening post-test linkage systems for partners identified through recency-triggered PNS remains essential for maximizing prevention benefits. While these findings underscore PNS epidemiological value, limitations include potential underreporting of sensitive partnerships warrant consideration in implementation. Future research should evaluate integrated recency-PNS models to determine their optimal configuration for maximizing both individual and population level prevention outcomes.
This study provides the most comprehensive evaluations to date of integrated HIV recency testing and partner notification services across LMIC's, synthesizing data from 25 studies encompassing tens of thousands of newly diagnosed individuals. Its principal strength lies in the novel integration of recency and PNS cascade analysis, revealing critical intersections between diagnostic accuracy and intervention effectiveness that previous studies have addressed separately. The applications of both random-and fixed-effect models with rigorous sensitivity analysis enhance the robustness of findings, while the consistent HIV prevalence among tested partners across diverse settings validates the epidemiological utility of well implemented PNS programs. Furthermore, our inclusion of both programmatic and research data provides unique insights into real-world implementation challenges and best practices.
We acknowledged several limitations in the study including substantial heterogeneity (I 2 = up to 100%) across many analyses despite subgroup explorations, reflecting unavoidable variations in recency assay performance, PNS eligibility criteria, and program maturity levels. While our modeling approaches accounted for this variability, residual confounding from a measured contextual factor (e.g., local stigma levels, health system resilience) may remain. Additionally, the predominance of routine program data introduces potential reporting biases, particularly for sensitive indicators like partner refusal rates, or exact testing timelines. Furthermore, while RITA confirmation reduced misclassification, residual challenges persist in high ART coverage settings where atypical viral suppression patterns may still lead to underdetection of acute infections. Finally, the geographic concentration of studies (predominantly East and Southern Africa) may limit generalizability to regions like West/ Central Africa where epidemic dynamics and health systems differ.
## CONCLUSION
Based on these findings, we propose 3 priority actions for programs: (1) Implementation of tiered recency testing protocols that prioritize RITA-confirmed cases for PNS resource allocation; (2) Standardizations of PNS eligibility criteria and quality metrics across programs to reduce cascade attrition; and (3) Investments in point of care viral load platforms to minimize confirmation delays. Future research should focus on the following key areas, cost effectiveness analysis of integrated recency PNS models, validation of next generation multi-assay algorithms in routine care settings, and implementation science studies to optimize approaches for key populations currently underrepresented in recent infection surveillance (particularly Men who have Sex with Men & People Who Inject Drugs). This meta-analysis demonstrates that recency-informed HIV programs can effectively identify active transmission networks when implemented with confirmatory testing and strong partner services. The 12% RTRI-recent prevalence and corresponding 15% partner HIV-prevalence confirm substantial ongoing transmission, while the 34% reclassification rate underscores the necessity of viral load confirmation. Moving forward, the strategic integration of recency testing with PNS-guided by RITA confirmation and supported by robust linkage systems, offers a transformative opportunity to focus limited resources on the highest risk transmission networks. As LMIC's advance toward epidemic control, these findings provide an evidencebased road map for optimizing surveillance and prevention investments to maximize population-level impacts.
## FOOTNOTES
Author contributions: Ahmed El-Imam I, Bakari HM, and Ramadhani HO contributed to conceptualization; Ahmed El-Imam I, Bakari HM, Ally ZM, Mbwana MS, Ally HM, and Ramadhani HO contributed to data curation; Peter TA, Ally HM, and Ramadhani HO contributed to formal analysis; Peter TA, Ally HM, Ahmed El-Imam I, Fussi HF, and Ramadhani HO contributed to methodology; Fussi HF, Ahmed El-Imam I, and Ramadhani HO contributed to validation; Ahmed El-Imam I, contributed to writing original draft; Bakari HM, Peter TA Chenya UK and Mpimo BK contributed to visualization. All authors reviewed this manuscript, provided feedback, and approved the manuscript in its final form.
## Conflict-of-interest statement:
## References
1. (2024) "global AIDS report -The Urgency of Now: AIDS at a Crossroads | UNAIDS"
2. Frescura, Godfrey-Faussett, Feizzadeh et al. (2022) "Ghys PD; on and behalf of the 2025 testing treatment target Working Group" *PLoS One*
3. Stephens, Mfungwe, Chalira et al. (2024) "Notes from the Field: Public Health Response to Surveillance for Recent HIV Infections -Malawi" *MMWR Morb Mortal Wkly Rep*
4. (2022) "Using recency assays for HIV surveillance: 2022 technical guidance"
5. Voetsch, Duong, Stupp et al. (2021) "HIV-1 Recent Infection Testing Algorithm With Antiretroviral Drug Detection to Improve Accuracy of Incidence Estimates" *J Acquir Immune Defic Syndr*
6. (2025) "Asanté® HIV-1 Rapid Recency® -Sedia Biosciences"
7. Yufenyuy, Detorio, Dobbs et al. (2022) "Performance evaluation of the Asante Rapid Recency Assay for verification of HIV diagnosis and detection of recent HIV-1 infections: Implications for epidemic control" *PLOS Glob Public Health*
8. Telford, Tessema, Msukwa et al. (2019) "Geospatial Transmission Hotspots of Recent HIV Infection -Malawi"
9. Srithanaviboonchai, Yingyong, Tasaneeyapan et al. (2024) "Establishment, Implementation, Initial Outcomes, and Lessons Learned from Recent HIV Infection Surveillance Using a Rapid Test for Recent Infection Among Persons Newly Diagnosed With HIV in Thailand: Implementation Study" *JMIR Public Health Surveill*
10. (2024) "HIV recency testing, positivity yield, and intimate partner violence among persons newly diagnosed with HIV: Findings from the Rwanda HIV recency evaluation study" *ClinicalTrials.gov*
11. Saito, Reid, Poirot et al. (2020) "Can HIV recent infection surveillance help us better understand where primary prevention efforts should be targeted? Results of three pilots integrating a recent infection testing algorithm into routine programme activities in Kenya and Zimbabwe"
12. Kim, Behel, Northbrook et al. (2019) "Tracking with recency assays to control the epidemic: real-time HIV surveillance and public health response" *AIDS*
13. Truong, Am, Sm et al. (2019) "PEPFAR HIV Case-Based Surveillance Study Group; PEPFAR HIV Case-based Surveillance Study Group" *MMWR Morb Mortal Wkly Rep*
14. Welty, Motoku, Muriithi et al. (2020) "A Feasibility Study"
15. Mochama "Increasing HIV case identification from safe index testing of recent HIV acquisition: Experiences from Laikipia County, Kenya"
16. Brisbane (2023)
17. Ouk, Soch, Ngauv et al. (2020) "HIV infection surveillance and partner testing outcomes in Cambodia from March"
18. Higgins, Thompson, Deeks et al. (2003) "Measuring inconsistency in meta-analyses" *BMJ*
19. Bakari, Alo, Mbwana et al. (2023) "Same-day ART initiation, loss to follow-up and viral load suppression among people living with HIV in low-and middle-income countries: systematic review and meta-analysis" *Pan Afr Med J*
20. Aungkulanon, Kittinunvorakoon, Tasaneeyapan et al. (2022) "Use of a robust health information system to improve accuracy of recent HIV infection testing"
21. Simanovong, Southalack, Xangsayarath et al. "First national HIV recent infection surveillance in Lao PDR"
22. Rwibasira, Malamba, Musengimana et al. (2018) "Recent infections among individuals with a new HIV diagnosis in Rwanda" *PLoS One*
23. Alemu, Ayalew, Haile et al. (2019) "Recent HIV infection among newly diagnosed cases and associated factors in the Amhara regional state"
24. Msukwa, Maclachlan, Gugsa et al. (2019) "Characterising persons diagnosed with HIV as either recent or long-term using a cross-sectional analysis of recent infection surveillance data collected in Malawi from September" *BMJ Open*
25. Zhu, Wang, Liu et al. (2020) "Identifying major drivers of incident HIV infection using recent infection testing algorithms (RITAs) to precisely inform targeted prevention" *Int J Infect Dis*
26. (2024) "AIDS 2024 -24th International AIDS Conference"
27. Negedu-Momoh, Balogun, Dafa et al. (2021) "Estimating HIV incidence in the Akwa Ibom AIDS indicator survey (AKAIS), Nigeria using the limiting antigen avidity recency assay" *J Int AIDS Soc*
28. Parmley, Harris, Hakim et al. (2022) "Recent HIV Infection Among Men Who Have Sex with Men, Transgender Women, and Genderqueer Individuals with Newly Diagnosed HIV Infection in Zimbabwe: Results from a Respondent-Driven Sampling Survey" *AIDS Res Hum Retroviruses*
29. Young, Musingila, Kingwara et al. (2023) "HIV Incidence, Recent HIV Infection, and Associated Factors"
30. Ang, Low, Wong et al. (2021) "Epidemiological factors associated with recent HIV infection among newly-diagnosed cases in Singapore" *BMC Public Health*
31. Agyemang, Kim, Dobbs et al. (2017) "Performance of a novel rapid test for recent HIV infection among newly-diagnosed pregnant adolescent girls and young women in four high-HIV-prevalence districts-Malawi" *PLoS One*
32. Singh, Mthombeni, Olorunfemi et al. (2018) "Evaluation of the accuracy of the Asanté assay as a point-of-care rapid test for HIV-1 recent infections using serum bank specimens from blood donors in South Africa"
33. Zhou, Cui, Hong et al. (1947) "The role of finance and administration in supporting children living with HIV in shelter homes under the care of elderly mothers in Buikwe District in Uganda, East Africa" *Viruses*
34. Facente, Grebe, Maher et al. (2021) "Experiences and lessons learned from the real-world implementation of an HIV recent infection testing algorithm in three routine service-delivery settings in Kenya and Zimbabwe" *JMIR Public Health Surveill*
35. Offie, Akpan, Okoh et al. (2022) "Impacts of Recent Infection Testing Integration into HIV Surveillance in Ekiti State, South West Nigeria: A Retrospective Cross Sectional Study" *World J AIDS*
36. Fieggen, Smith, Arora et al. (2022) "The role of machine learning in HIV risk prediction" *Front Reprod Health*
37. Nethi, Karam, Alvarez et al. (2024) "Using Machine Learning to Identify Patients at Risk of Acquiring HIV in an Urban Health System" *J Acquir Immune Defic Syndr*
38. Kassanjee, Pilcher, Busch et al. (2016) "Viral load criteria and threshold optimization to improve HIV incidence assay characteristics" *AIDS*
39. Sharma, Ong, Celum et al. (2020) "Heterogeneity in individual preferences for HIV testing: A systematic literature review of discrete choice experiments" *EClinicalMedicine*
40. Thapa, Hannes, Cargo et al. (2018) "Stigma reduction in relation to HIV test uptake in low-and middleincome countries: a realist review" *BMC Public Health*
41. Katz, Wong, Medley et al. (2019) "The power of partners: positively engaging networks of people with HIV in testing, treatment and prevention" *J Int AIDS Soc*
42. Dalal, Johnson, Fonner et al. (2017) "Improving HIV test uptake and case finding with assisted partner notification services" *AIDS*
43. Mahachi, Muchedzi, Tafuma et al. (2019) "Sustained high HIV case-finding through index testing and partner notification services: experiences from three provinces in Zimbabwe" *J Int AIDS Soc*
44. Tih, Chimoun, Khan et al. (2019) "Assisted HIV partner notification services in resource-limited settings: experiences and achievements from Cameroon" *J Int AIDS Soc* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12446446&blobtype=pdf | # npj | viruses Article
Yannick Brüggemann, Toni Meister, Natalie Heinen, Emely Richter, Saskia Westhoven, Michael Poppe, Mohammed Shaban, Leyla Sirkinti, Maximilian Nocke, Daniel Todt, Stephanie Pfaender, Michael Kracht, Eike Steinmann
## Abstract
Identifying common host factors essential for the replication cycles of human coronaviruses (HCoV) could help uncover potential therapeutic targets. Mitogen-activated protein kinases (MAPKs) regulate critical cellular signaling pathways. Among them, c-Jun N-terminal kinases (JNK) are activated in response to diverse environmental stresses, including viral infections. However, the relevance of the JNK pathway for host responses and replication of HCoV infections has remained elusive. Using livecell microscopy, quantitative immunofluorescence and immunoblotting, we found that JNK is specifically activated in cells infected with HCoV-229E and plays a crucial role in mediating the phosphorylation of the viral nucleocapsid (N) protein, an essential step required during the viral replication cycle. Consequently, pharmacological inhibition of JNK kinase activity impeded HCoV-229E as well as SARS-CoV-2 infection. Given the conservation of phosphorylation sites within the nucleocapsid protein across coronaviruses, inhibitors targeting these N protein kinases, such as JNK, may hold therapeutic promise as broad-spectrum CoV antivirals.The emergence and reemergence of coronaviruses as significant human pathogens pose persistent threats to global public health, as evidenced by the devastating impact of outbreaks such as severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV) 1 , and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes coronavirus disease 2019 (COVID-19) 2 . Additionally, the four common human coronaviruses (229E, HKU1, NL63, and OC43) are responsible for 15% to 30% of common cold cases in adults and can lead to severe illness in high-risk individuals, such as infants, the elderly, and immunocompromised patients 3 . Understanding the intricate interplay between coronaviruses and their host cells is essential for elucidating viral pathogenesis and developing novel therapeutic strategies 4-6 . Central to this endeavor is the characterization of host signaling pathways exploited by coronaviruses to facilitate their replication and evade host immune responses 6 . The c-Jun N-terminal kinases (JNK) pathway regulates multiple cellular processes such as proliferation, apoptosis, and immune responses 7 . JNK belongs to the family of mitogen-activated protein kinases (MAPKs), which are key players in evolutionarily conserved signaling cascades that transduce extracellular stimuli into diverse cellular responses [8][9][10] . Upon activation by various stressors, including pro-inflammatory cytokines, environmental toxins, and viral infections, JNK kinases phosphorylate downstream substrates, including transcription factors such as c-Jun, leading to the modulation of gene expression and cellular function 11 . Accumulating evidence has linked the JNK pathway to the pathogenesis of numerous viral infections, highlighting its potential as a target for antiviral intervention 12 . In particular, JNK activation has been shown to favor replication of HIV 13 , herpes simplex virus 14 , rotavirus 15 , dengue virus 16 , influenza A virus 17 and SARS-CoV 18 . However, despite the growing body of research implicating the involvement of JNK kinases in viral infections, comprehensive studies delineating the precise mechanisms by which coronaviruses exploit the JNK pathway, particularly HCoV-229E, remain limited. In this study, the activation and requirement of JNK during the replication cycle of HCoV-229E were examined.
## Results
HCoV-229E infection promotes JNK phosphorylation and signaling To probe for JNK kinase activities during HCoV-229E infection in living cells, we adapted an imaging system based on a kinase translocation reporter (KTR) 19 . The JNK KTR consists of a substrate recognition motif (peptides derived from the JNK substrate c-Jun), phosphorylation sites (P sites) located near a nuclear localization sequence (bNLS) and a nuclear export sequence (NES) site and a fluorescent protein (Clover). KTR phosphorylation by JNK suppresses bNLS activity and enhances NES activity, leading to a nucleocytoplasmic shuttling event that can be measured by fluorescence microscopy. After phosphorylation by JNK, the KTR relocates from the nucleus to the cytosol and returns to the nucleus when dephosphorylated (Fig. 1a). We stably transduced Huh7 cells with a JNK-KTR fused to the green fluorescent protein Clover (Fig. 1b). In the absence of external stimuli, the sensor mainly localized towards the nucleus. Stimulation with the prototypical JNK activator tumor necrosis factor-alpha (TNF-α) resulted in a translocation of the sensor from the nucleus towards the cytoplasm due to increased phosphorylation-dependent nuclear export as previously described (Fig. 1b,c). KTR translocation could be prevented by the specific JNK kinase inhibitor JNK-IN-8, demonstrating that the translocation event reflects JNK kinase activity (Fig. 1b,c). JNK kinase activity was quantified based on the ratio of the cytosolic to the nuclear fluorescence intensity values derived from individual cells (cytoplasm/nucleus ratio-Fig. 1c). Overall, these data demonstrate the functionality and specificity of the JNK kinase reporter system in Huh7 cells.
Next, to monitor the activation dynamics of JNK upon infection with HCoV-229E, we performed live cell imaging experiments in Huh7 JNK KTR cells. Following HCoV-229E infection, we observed translocation of the JNK-KTR from the nucleus towards the cytoplasm after approximately 16 h p.i. In particular, cells which underwent syncytia formation showed strong translocation of the reporter construct to the cytoplasm, indicating strong JNK activation. In contrast, the reporter did not translocate towards the cytoplasm in non-infected control cells (Fig. 1d and Supplementary Movie 1). To test if JNK is specifically activated in HCoV-229E infected cells, we performed Immunofluorescence staining of the HCoV-229E N protein in Huh7-JNK-KTR cells 24 h after infection (Fig. 1e). JNK-KTR translocation was restricted to N protein-positive cells, implying JNK activation only in HCoV-229E-infected cells. To exclude potential artifacts in JNK-KTR translocation due to syncytia formation and/or morphological changes of infected cells, we employed a non-phosphorylatable KTR mutant (JNK-KTR-AA). The JNK-KTR-AA construct was strictly localized to the nucleus irrespective of HCoV-229E infection (Fig. 1f). The differential subcellular localization of both constructs upon HCoV-229E infection (Fig. 1g) was further reflected by a strong increase of the cytoplasm/nucleus ratio for the JNK-KTR construct, while the ratio for the JNK-KTR-AA construct remained unchanged (Fig. 1h). Hence, JNK-KTR translocation upon HCoV-229E specifically reflects catalytic activity of endogenous JNK kinase. Accordingly, we further observed phosphorylation of endogenous c-Jun, a member of the downstream JNK transcription factor activator protein-1 (AP-1) 7 , specifically in HCoV-229E-infected single cells (Fig. 1i). These results were corroborated by Western blot analysis, which revealed virusinducible phosphorylation of several JNK isoforms concomitant with an increase of c-Jun expression and phosphorylation approximately 12-16 h after infection (Fig. 1j). Expression of viral proteins (N protein and nsp8) was already detected 9 h after infection (Fig. 1j). In summary, these results demonstrated progressive activation of the JNK-c-JUN pathway in HCoV-229E-infected cells within 24 h of infection, but not in non-infected bystander cells.
## JNK inhibition prevents HCoV-229E infection
Given the strong activation of JNK upon HCoV-229E infection, we next assessed if inhibition of JNK kinase activity affects HCoV-229E infectivity. To account for potential inhibitor-specific (off-target) effects, we employed four different JNK kinase inhibitors (JNKi), namely JNK-IN-8; AS601245; Bentamapimod and SP600125 20 , which specifically target the ATP-binding pocket of the JNK catalytic domain. Huh7 cells were pretreated with increasing inhibitor dosages for 1 h and subsequently infected with HCoV-229E and stained for double-stranded RNA (dsRNA)-a validated marker of active coronavirus replication 21,22 -to identify infected cells (Fig. 2a). Quantitative immunofluorescence showed a strong dose-dependent decrease in dsRNA-positive cells (Fig. 2a,b). Inhibitor titration revealed only minor changes in IC 50 values indicating no major differences in the potency among the different JNKi (Fig. 2c). Although all inhibitors reduced cell viability at high concentrations, they impaired cell viability by not more than 20% at concentrations that reduced viral replication by at least 50% (Fig. 2d). In agreement with this, all four inhibitors lowered the production of infectious viruses (TCID50/mL-Fig. 2e) by several orders of magnitude, when used at higher concentrations (>5 µM). Accordingly, siRNAmediated knockdown of JNK1/2 reduced the production of infectious viruses (Supplementary Fig. 1). Collectively, our data highlight the requirement of JNK activity during HCoV-229E infection.
## The HCoV-229E N protein is phosphorylated in a JNKdependent manner
To broadly determine which steps of the HCoV-229E replication cycle are affected by JNK, we conducted time-of-addition analysis experiments. We initiated treatment with two different JNKi at various time points during infection (Fig. 3a). (Pre-)application of either JNK-IN-8 or Bentamapimod during the 1-h infection period did not significantly impact the amount of infectious virus produced, suggesting that JNK inhibition does not affect virus attachment or entry into target cells. In contrast, application of either inhibitor after viral entry strongly inhibited virus infection, implying that JNK functions at a post-entry step during the viral replication cycle. We then tested whether JNK activity is required during viral replication by measuring the amount of dsRNA viral RNA in single cells. As previously observed, treatment with JNK-IN-8 greatly reduced the number of infected cells (Fig. 3b). However, quantification of the dsRNA signal in the remaining infected cells upon JNK-IN-8 inhibitor treatment revealed that JNK inhibition had only a weak effect on the amount of dsRNA produced per cell (Fig. 3b). Consistent with this, JNK inhibition did not alter the amount of viral nucleocapsid protein under the same conditions in the remaining infected cells (Fig. 3c). These data suggest that JNK inhibition appears not to modulate viral replication and protein synthesis directly, once a cell is infected. We hypothesized that JNK activates downstream host factors, such as c-Jun, to regulate the (transcriptional) expression of components that are required during other steps of the viral life cycle. As observed previously (Fig. 1i,j), infection with HCoV-229E resulted in a strong increase of c-Jun phosphorylation that was suppressed by JNK-IN-8 treatment (Fig. 3d). However, the inhibition of c-Jun phosphorylation did not affect N protein synthesis in those cells with remaining viral replication, suggesting that the JNK-inhibitory effects described above did not necessarily rely on phosphorylation of canonical c-Jun sites (Fig. 3d). We further tested this notion by means of three different AP-1 inhibitors, including T-5224, which specifically inhibits the DNA binding activity of the c-Fos/c-Jun heterodimer 23 as well as SR 11302 and 1-Methyl-6-oxo-1,6-dihydropyridine-3-carboxylic acid (Nudifloric acid), which inhibit the AP-1 transcription factor 24,25 . To confirm the activity of the AP-1 inhibitors, we generated an AP-1 reporter construct containing a green fluorescent protein (GFP) under the control of AP-1 response elements and a minimal promoter (Supplementary Fig. 2a) 26 . Treatment with AP-1 inhibitors reduced GFP expression in response to IL-1α stimulation, confirming effective inhibition of AP-1 activity (Supplementary Fig. 2b,c). Importantly, none of the inhibitors altered the number of cells infected with HCoV-229E (Fig. 3e,f), while preserving high levels of cell viability (Fig. 3g). Similarly, stable expression of a dominant-negative Jun variant 27 (Supplementary Fig. 2d) did not affect HCoV-229E infection (Supplementary Fig. 2e). We therefore concluded that the antiviral effects of JNKi occurred largely independent of the c-Jun/AP-1 pathway.
Ample evidence implies posttranslational modifications in the functionality of viral proteins 28 , with phosphorylation being particularly important 29,30 . While phosphorylation sites have been identified in various viral proteins, the CoV N protein stands out as the most extensively phosphorylated protein 31,32 . To test if JNK mediates the phosphorylation of the viral N protein, we used specific antibodies that were generated to probe for phosphorylated serine 145 and 149 (S145/149) within the SR-region, and phosphorylated serine 364 and 367 (S364/367) within the C terminal region, respectively (Fig. 3h). Both domains have been shown to be relevant for the multimerization of N proteins, with the SR-region also being essential for replication 33 . Upon infection with HCoV-229E, a strong phosphorylation signal could be observed for both phosphorylation sites (Fig. 3i and Supplementary Fig. 3). Treatment with the JNK inhibitor SP600125, reduced N protein expression (Fig. 3j), while the phosphorylation ratio (p-S145/149/N Protein) of the remaining N Protein at the S145/149 sites was not significantly reduced following SP600125 treatment (Fig. 3k), suggesting this site can be phosphorylated by multiple kinases and does not specifically depend on JNK activity. In contrast, the phosphorylation ratio (p-S364/367/ N Protein) at the S364/367 site was strongly reduced upon SP600125 treatment (Fig. 3l), providing evidence for a specific virus-induced activation of JNK that contributes to regulated N protein phosphorylation at these two sites (Fig. 3i and Supplementary Fig. 3). These data suggest that phosphorylation of the viral N protein by JNK supports the viral replication cycle.
## JNK inhibition lowers SARS-CoV-2 infectivity
To investigate the conservation of the two phosphorylation motifs within the HCoV-229E N protein, we compared the genomic sequence of the N proteins from all different human coronaviruses and determined the conservation of the respective serine residues (Fig. 4a). We noticed that the serine residues within the SR-rich domain (S145/149) were highly conserved among all viruses, while the serine residues close to the C terminus (S364/367) of the N protein were only present in the HCoV-229E N protein.
Given the conservation of the serine residues within the SR-rich domain (S145/149) and previous evidence highlighting the significance of SR-rich domain phosphorylation for the coronavirus replication cycle [33][34][35][36][37][38][39] , we investigated whether JNK activity is also necessary during SARS-CoV-2 infection. Infection of human A549 lung epithelial cells expressing ACE2 and TMPRSS2 with SARS-CoV-2 revealed c-Jun phosphorylation in cells positive for dsRNA, implying JNK activation specifically within SARS-CoV-2-infected cells (Fig. 4b). Upon JNK-IN-8 treatment, c-Jun phosphorylation as well as the fraction of dsRNA-positive cells were strongly reduced, suggesting the activation and requirement of JNK kinase activity during SARS-CoV-2 infection. In agreement with this, all the JNKi JNK-IN-8, Bentamapimod and AS601245 lowered the production of infectious viruses (TCID50/mL-Fig. 4c), without compromising cell viability (Fig. 4d). Similar to our results for HCoV-229E, SARS-CoV-2 requires the JNK signaling for efficient replication.
## Discussion
The study of coronaviruses has become a crucial area of research, particularly given the significant global impact of outbreaks like SARS-CoV, MERS-CoV 1 , and most recently SARS-CoV-2 2 . Moreover, the common four human coronaviruses (229E, HKU1, NL63, and OC43) contribute to 15%-30% of cases of common colds in human adults and can cause severe disease in patients at risk (i.e. infants, elderly people, or immunocompromised patients) 3 . Consequently, understanding the host signaling pathways and characterizing virus-host interactions during the CoV infection cycle are essential for identifying functionally relevant host factors, which could potentially guide the development of novel and innovative therapies 5,32,40,41 . We observed strong activation of JNK specifically within HCoV-229E-infected cells. Pharmacological inhibition of JNK activity further reduced HCoV-229E infection and production of viral progeny production, suggesting an important role of JNK during HCoV-229E infection. Pattern recognition receptors (PRRs) such as RIG-I and MDA5 are essential cytosolic RNA sensors that detect viral infections by recognizing double-stranded RNA (dsRNA) 42,43 . Upon activation, these receptors trigger an innate immune response. In the context of viral infection, the activation of MAVS (mitochondrial antiviral-signaling protein) has been associated with the subsequent activation of the JNK pathway 44,45 . Hence, intracellular sensing of viral RNA during virus replication may promote JNK activation via MAVS independently of extracellular stimuli. Posttranslational modifications, particularly phosphorylation, are recognized as crucial modulators for the functionality and localization of viral proteins 29,31 ,46,47 . Although phosphorylation sites have been identified in various viral proteins, the N protein has been reported to exhibit the highest level of phosphorylation 31,32 . Using two phospho-specific antibodies, phosphorylation of the HCoV-229E N protein was detected in the serine-rich domain and the C-terminus. However, it remains unclear whether JNK itself acts as the kinase or if a kinase activated downstream of JNK is responsible for this process. Previous studies have shown that the phosphorylation status of the coronavirus N protein can modulate protein-protein and protein-RNA interactions [33][34][35][36] as well as gel-liquid phase transition 39 and viral transcription 37 . We observed no differences in the amount of viral RNA produced upon JNK inhibition (Fig. 3b), while the amounts of infectious progeny virus were greatly reduced (Fig. 2e), suggesting that JNKdependent phosphorylation of the N protein is not required during viral replication, but rather be required during viral assembly and release. While the specific functions of individual phosphorylation sites are yet to be fully understood, the overall phosphorylation status of the nucleocapsid protein has been suggested to serve as a regulatory switch during the replication cycle 48 , between transcription and replication, or to promote genome packaging or unpackaging 38,39 . Future studies using reverse genetic systems to engineer viruses with desired mutations are required to study the role of individual phosphorylation sites during the CoV replication cycle 49,50 . Although our findings, in line with previous studies, suggest a potential role for JNK-mediated phosphorylation in the viral replication cycle, we cannot rule out the possibility that the inhibition of viral infection following JNK inhibition may be at least partially due to altered phosphorylation and/or expression of host or other viral proteins. In this context kinase substrate specificity mapping and in vitro phosphorylation assays could help to uncover protein kinases involved in the phosphorylation of the serine residues examined in this study (S145/149 and S364/367) 31 .
The strong conservation of the phosphorylation sites within the SRregion across human coronaviruses 31 suggests a conserved mechanism that may be exploited for therapeutic purposes. The finding that inhibiting JNK activity reduces the infectivity of both HCoV-229E and SARS-CoV-2 indicates that the development of specific kinase inhibitors for clinical use may offer an avenue towards a broad-spectrum antiviral approach against multiple coronaviruses. The JNK signaling pathway comprises up to ten isoforms encoded by three genes, which are ubiquitously expressed and form a highly redundant core module responsible for maintaining cellular homeostasis and mounting rapid responses to various cellular stressors. Hence, global inhibition of JNKs using compounds that target the conserved ATP-binding pocket has shown limited clinical success. Future therapeutic strategies must adopt more targeted approaches. Potential promising directions could include (1) the development of allosteric inhibitors with isoform specificity, (2) small molecules designed to disrupt specific JNKsubstrate interactions, such as those involving the well-characterized docking domains between JNK and its canonical substrate c-Jun 51,52 , and (3) tissue-specific downregulation of individual JNK isoforms using stabilized short RNA oligonucleotides 53 . In addition, proteolysis-targeting chimeras (PROTACs) offer an innovative strategy by enabling the transient degradation of selected JNK isoforms 54 . Another promising approach is the targeted delivery of JNK inhibitors to infected cells, for instance via topical administration to the nasopharyngeal epithelium during acute, symptomatic CoV infection. In summary, advancing JNK-targeted therapies will require a multimodal strategy that integrates the development of highly selective, potent compounds with a broad therapeutic index, refined dosing regimens, and precision delivery systems. Isoform-specific targeting and infection-contextual inhibition will be critical to maximize efficacy while minimizing off-target effects. Overall, our findings highlight the importance of studying the host-virus interplay to identify potential targets and mechanisms for developing host-directed antiviral therapies 5,32,40,41 . Further research is needed to clarify the specific functions of N protein phosphorylation in the replication cycles of different coronaviruses. This understanding could refine therapeutic strategies targeting kinase pathways, leading to more effective treatments for coronavirus infections, as well as other emerging or re-emerging viruses. HY-N4226). Recombinant Human TNF-α (300-01A) was obtained from Preprotech. Human recombinant IL-1a was a kind gift from Jeremy Saklatvala (Oxford, UK) or was prepared in our laboratory as described and used at 10 ng/mL final concentration in all experiments 55 . Anti-NSP8 was a gift from John Ziebuhr, Gießen, Germany. The antibodies against SS145/ 147 and SS364/367 were generated using a commercial immunization program by the company Eurogentec (4102 Seraing, Belgium). Briefly, rabbits were immunized with phosphorylated synthetic peptides h-C + EEPD-S (PO3H2) -RAP-S (PO3H2) -RSQ -nh2 (for SS145/149) or h-C + EFNP-S(PO3H2)-QT-S(PO3H2)-PATA -nh2 (for SS364/367). The sera of two rabbits were analyzed for high-titer antibodies against the phosphorylated forms of the peptides compared to the carrier by specific ELISA with Ppeptide-coated plates. The final bleed of one rabbit with the highest titer was used to purify high-titer P-specific antibodies from serum. First, specific antibodies were purified by P-peptide affinity chromatography. Second, any remaining IgG fraction also recognizing the unmodified peptide structures was removed by a second purification step on affinity matrices coupled to the unmodified peptides. The flowthrough of this column contained highly specific P-peptide antibodies with titers >1 × 10 4 for the P-peptide.
## Methods
## Cell culture
Huh7 cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% (v/v) fetal calf serum (FCS), 1% (v/v) non-essential amino acids (NEAA), 100 IU/mL penicillin, 100 μg/mL streptomycin, and 2 mM L-glutamine. ACE2 and TMPRSS2 overexpressing A549 (A549-A/T) cells were cultured in DMEM with 5% (v/v) FCS, 1% (v/v) NEAA, 100 IU/mL penicillin, 100 μg/mL streptomycin, and 2 mM L-glutamine. The cells were additionally selected with Blasticidin (10 µg/mL) and Puromycin (0.5 µg/mL).
## Production of ectopically expressing cell lines via lentiviral transduction
For the production of lentiviral particles, 4 × 10 5 293 T cells were seeded on collagen-coated 6-well plates. The following day, the 293 T cells were transfected with the plasmids pcz-VSV-G, pCMV-dR8.74, along with plasmids encoding the desired transgenes using Lipofectamine 2000 (Invitrogen, Cat. 40 Nr. 11668019) following the manufacturer's instructions. Six hours post-transfection, the medium was changed, and lentiviral particles were harvested 48 h post-transfection. Supernatants were filtered (Filtropur 0.45, Sarstedt, Cat. Nr. 83.1826) and supplemented with HEPES and polybrene, and either used directly or stored at -80 °C. For transduction, Huh7 cells were seeded on a 6-well plate and inoculated with 1 mL of lentiviral particles for 6-8 h. Selection of the transduced cells was started 48 h post-transduction using 2.5 μg/mL puromycin. Transgene expression was validated via fluorescence microscopy.
## Virus infection assays
Huh7 or A549-A/T cells were seeded at a density of 8 × 10 4 cells/well in a 24well plate. After cell attachment, cells were pretreated with different inhibitors for at least 30 min. For infection, Huh7 cells were inoculated with human coronavirus 229E (MOI 0.1) 56 in 10% FCS-containing DMEM for 1 h, while A549-A/T were inoculated with hCoV-19/Germany/BY-Bochum-1/2020 (B.1.1.70; GISAID accession ID: EPI_ISL_1118929, MOI 1) for 1 h (A549-A/T cells). Hereafter, the cells were washed three times with 1 × PBS and treated again with the respective inhibitors and immunosuppressants as indicated in the figure legends.
## AP1 reporter assay
Huh7 cells stably expressing the AP1-GFP reporter were seeded into 96-well plates (10⁴ cells per well). Twenty-four hours later, the culture medium was removed, and cells were incubated with 50 µL of either DMSO or c-Jun inhibitors at a concentration of 50 µM for 1 h. Following this pre-treatment, 50 µL of medium containing IL-1α was added to each well, resulting in a final concentration of 10 ng/mL IL-1α and 25 µM inhibitor. After 24 h of stimulation, cells were fixed with paraformaldehyde (PFA), and nuclei were counterstained with DAPI prior to imaging. GFP intensity in the nucleus was determined using CellProfiler.
## Immunoblotting
Whole cell extracts were prepared in Triton cell lysis buffer (10 mM Tris, pH 7.05, 30 mM NaPPi, 50 mM NaCl, 1% Triton X-100, 2 mM Na3 VO 4, 50 mM NaF, 20 mM ß-glycerophosphate and freshly added 0.5 mM PMSF, 2.5 μg/mL leupeptin, 1.0 μg/mL pepstatin, 1 μM microcystin). Cell lysates were subjected to SDS-PAGE on 7-12.5% gels. Proteins were separated on SDS-PAGE and electrophoretically transferred to PVDF membranes (Roth, Roti-PVDF (0.45 μm)). After blocking with 5% dried milk in Tris-HClbuffered saline/0.05% Tween (TBST) for 1 h, membranes were incubated for 12-24 h with primary antibodies, including custom made anti-phospho N Protein HCoV-229E Serin 145/149 (Eurogentec), anti-phospho N Protein HCoV-229E Serin 364/367 (Eurogentec), anti-phospho JNK Threonine 183/Tyrosine 185 (Cell Signaling (#9251)), anti-JNK (Santa Cruz (#sc-571)), anti-Jun (Santa Cruz (#sc-1694)), anti-N Protein HCoV-229E (Ingenasa (Batch 250609)), anti-β-Actin (Santa Cruz (#sc-4778)). Afterwards, membranes were washed in TBST and incubated for 1-2 h with the peroxidasecoupled secondary antibody. Proteins were detected by using enhanced chemiluminescence (ECL) systems from Millipore or GE Healthcare. Images were acquired and quantified using a Kodak Image Station 440 CF and the software Kodak 1D, 3.6, or the ChemiDoc Touch Imaging System (BioRad) and the software ImageLab, V_5.2.1 or higher (Bio-Rad).
## Immunofluorescence
Cells were fixed in 3% paraformaldehyde at room temperature for at least 10 min and washed three times with PBS. Permeabilization was achieved using 0.2% Triton X-100 for 5 min, followed by three additional PBS washes. Cells were blocked in 5% horse serum in PBS for 1 h at room temperature. Primary antibodies were diluted in 5% horse serum-PBS and incubated overnight at 4 °C. Anti-phospho-c-Jun (Ser73) antibody (1:500, Cell Signaling Technology #9164), anti-N protein (1:1000, Anticuerpo Monoclonal, 1E7, Ingenasa), HA Tag Monoclonal Antibody (1:500, Thermo Fisher 2-2.2.14) and anti-dsRNA antibody (1:1000, SCICONS 10010500) were used as primary antibodies. Secondary antibodies were Alexa 488-labeled donkey anti-rabbit IgG (Invitrogen A11008), Alexa 555-labeled donkey anti-mouse IgG (Invitrogen A11008) or Alexa 488-labeled donkey antimouse IgG (Invitrogen A21206), used at 1/1000 dilution in 5% horse serum-PBS, and incubated for 1-2 h at room temperature in the dark. Nuclei were stained with DAPI (1 μg/mL).
## Microscopy and image analysis
Immunofluorescence images were acquired with a wide-field fluorescence microscope (Keyence BZ-X800E) using 4x, 10x and 20x objectives and BZ-X Filter DAPI, BZ-X Filter GFP and BZ-X Filter TRITC. To determine the fractions of dsRNA-positive cells, DAPI-stained nuclei were segmented and expanded by 3 pixels. Spots of dsRNA were detected within the expanded nuclei to identify infected cells using CellProfiler 57 . For JNK KTR cells, segmentation and object quantification were performed with CellProfiler 57 . Nucleus and a 5-pixel wide cytoplasm ring (cytoring) were segmented using DAPI-stained nuclei, and intensities were quantified in the KTR channel. Mean fluorescence intensities of nuclear p-c-Jun were obtained after segmentation of DAPI-stained nuclei using CellProfiler 57 . Live cell imaging experiments, confocal images were acquired on a CQ1 Confocal Imaging Cytometer (Yokogawa) using a 20x objective in combination with a BP447/60 (Hoechst) and BP525/50 (Clover) filters. A total of 1 × 10 4 cells/well Huh7-JNK-KTR cells were seeded in 96-well plates. Nuclei of living cells were stained with Hoechst were acquired in a 30 min interval over 24 h following infection with HCoV-229E and non-treated control cells.
## Cell viability assays
Cell viability was determined by adding 0.5 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma) substrate to cells and subsequent incubation at 37 °C and 5% CO 2 for 1-2 h. Medium was removed and 50 μL of DMSO was added to each well. The absorbance of each well was read on a microplate absorbance reader (Tecan Group Ltd, Männedorf, Switzerland) at 570 nm. Cells treated with 70% ethanol for 10 min served as background control.
## Multiple sequence alignment
Reference proteomes of viruses from the Coronaviridae family were downloaded from UniProt. N protein sequences (HCoV-229E -P15130; HCoV-NL63 -Q6Q1R8; SARS-CoV -P59595; MERS-CoV -T2BBK0; HCoV-OC43 -P33469; HCoV-HKU1 -Q19U25; SARS-CoV-2 -P0DTC9) were aligned to the HCoV-299E N protein with CLUSTAL 2.1 Multiple Sequence Alignments (MSA). Data visualization was conducted using a Jupyter Notebook 7.0.3 (Python 3.10.10) in combination with MsaViz (pyMSAviz package) and SeqIO (BioPython library) to visualize MSA and create consensus plots. The logomaker library was used to create conservation plots.
## Software
Data visualization and statistical analysis were performed using GraphPad Prism v10. Dose-response curves were calculated using a four-parameter non-linear regression model implemented in GraphPad Prism v10. Fluorescence microscopy images were analyzed with Fiji and/or CellProfiler 57,58 . The effect size was calculated as Cohen's d (d). An Alpha-fold model of the HCoV-229E nucleocapsid (GenBank: QNT54801.1) was processed and displayed with Chi-meraX (Version: 1.6rc202304072249 (2023-04-07)). (mean ± SEM, n = 3). Statistical significance was determined using Welch's t-test (*p < 0.05). c A549 A/T cells were either pretreated with different JNK inhibitors (all 10 µM), DMSO or left untreated (UTC) and subsequently inoculated with SARS-CoV-2 (MOI 1) for 1 h. Twenty-four hours post-infection, the supernatant was collected and viral titers determined by an endpoint dilution assay and calculated as TCID 50 /mL (mean ± SD, n = 3). Dashed lines indicate the lower limit of quantification (LLOQ). Statistical significance was determined using a one-way ANOVA with Dunnett's post hoc test (***p < 0.001 and *p < 0.05). d Normalized cell viability in percent (%) upon for A549 A/T cells treated with different JNK inhibitors (all 10 µM), DMSO and untreated control cells.
## References
1. Cui, Li, Shi (2019) "Origin and evolution of pathogenic coronaviruses" *Nat. Rev. Microbiol*
2. Markov (2023) "The evolution of SARS-CoV-2" *Nat. Rev. Microbiol*
3. Liu, Liang, Fung (2021) "Human coronavirus-229E, -OC43, -NL63, and -HKU1 (Coronaviridae)"
4. Fung, Liu (2019) "Human coronavirus: host-pathogen interaction" *Annu. Rev. Microbiol*
5. Bouhaddou (2020) "The global phosphorylation landscape of SARS-CoV-2 infection" *Cell*
6. V'kovski, Kratzel, Steiner et al. (2021) "Coronavirus biology and replication: implications for SARS-CoV-2" *Nat. Rev. Microbiol*
7. Davis (2000) "Signal transduction by the JNK group of MAP kinases" *Cell*
8. Zhang, Liu (2002) "MAPK signal pathways in the regulation of cell proliferation in mammalian cells" *Cell Res*
9. Cargnello, Roux (2011) "Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases" *Microbiol. Mol. Biol. Rev*
10. Kyriakis, Avruch (2012) "Mammalian MAPK signal transduction pathways activated by stress and inflammation: a 10-year update" *Physiol. Rev*
11. Johnson, Nakamura (2007) "The c-jun kinase/stress-activated pathway: regulation, function and role in human disease" *Biochim. Biophys. Acta*
12. Chen (2021) "The roles of c-Jun N-terminal kinase (JNK) in infectious diseases" *Int. J. Mol. Sci*
13. Kumar, Manna, Dhawan et al. (1998) "HIV-Tat protein activates c-Jun N-terminal kinase and activator protein-1" *J. Immunol*
14. Zachos, Clements, Conner (1999) "Herpes simplex virus type 1 infection stimulates p38/c-Jun N-terminal mitogen-activated protein kinase pathways and activates transcription factor AP-1" *J. Biol. Chem*
15. Holloway, Coulson (2006) "Rotavirus activates JNK and p38 signaling pathways in intestinal cells, leading to AP-1-driven transcriptional responses and enhanced virus replication" *J. Virol*
16. Ceballos-Olvera, Chávez-Salinas, Medina et al. (2010) "JNK phosphorylation, induced during dengue virus infection, is important for viral infection and requires the presence of cholesterol" *Virology*
17. Zhang (2019) "Role of c-Jun terminal kinase (JNK) activation in influenza A virus-induced autophagy and replication" *Virology*
18. Mizutani, Fukushi, Saijo et al. (2005) "JNK and PI3k/Akt signaling pathways are required for establishing persistent SARS-CoV infection in Vero E6 cells" *Biochim. Biophys. Acta*
19. Regot, Hughey, Bajar et al. (2014) "High-sensitivity measurements of multiple kinase activities in live single cells" *Cell*
20. Wu (2020) "Selective inhibitors for JNK signalling: a potential targeted therapy in cancer" *J. Enzyme Inhib. Med. Chem*
21. Chen (2024) "A coronaviral pore-replicase complex links RNA synthesis and export from double-membrane vesicles" *Sci. Adv*
22. Wang (2022) "Multi-color super-resolution imaging to study human coronavirus RNA during cellular infection" *Cell Rep. Methods*
23. Fanjul (1994) "A new class of retinoids with selective inhibition of AP-1 inhibits proliferation" *Nature*
24. Aikawa (2008) "Treatment of arthritis with a selective inhibitor of c-Fos/activator protein-1" *Nat. Biotechnol*
25. Suh (2017) "Chemical constituents identified from fruit body of Cordyceps bassiana and their anti-inflammatory activity" *Biomol. Ther*
26. Vasanwala, Kusam, Toney et al. (2002) "Repression of AP-1 function: a mechanism for the regulation of Blimp-1 expression and B lymphocyte differentiation by the B cell lymphoma-6 protooncogene" *J. Immunol*
27. Wang (2005) "Regulation of IL-10 gene expression in Th2 cells by Jun proteins" *J. Immunol*
28. Kumar, Mehta, Mishra et al. (2020) "Role of hostmediated post-translational modifications (PTMs) in RNA virus pathogenesis" *Int. J. Mol. Sci*
29. Keating, Striker (2012) "Phosphorylation events during viral infections provide potential therapeutic targets" *Rev. Med. Virol*
30. Fung, Liu (2018) "Post-translational modifications of coronavirus proteins: roles and function" *Future Virol*
31. Yaron (2022) "Host protein kinases required for SARS-CoV-2 nucleocapsid phosphorylation and viral replication" *Sci. Signal*
32. Klann (2020) "Growth factor receptor signaling inhibition prevents SARS-CoV-2 replication" *Mol. Cell*
33. Mcbride, Van Zyl, Fielding (2014) "The coronavirus nucleocapsid is a multifunctional protein" *Viruses*
34. Lu (2021) "The SARS-CoV-2 nucleocapsid phosphoprotein forms mutually exclusive condensates with RNA and the membraneassociated M protein" *Nat. Commun*
35. Wu (2009) "Glycogen synthase kinase-3 regulates the phosphorylation of severe acute respiratory syndrome coronavirus nucleocapsid protein and viral replication" *J. Biol. Chem*
36. Peng, Lee, Tarn (2008) "Phosphorylation of the arginine/ serine dipeptide-rich motif of the severe acute respiratory syndrome coronavirus nucleocapsid protein modulates its multimerization, translation inhibitory activity and cellular localization" *FEBS J*
37. Wu, Chen, Yeh (2014) "Nucleocapsid phosphorylation and RNA helicase DDX1 recruitment enables coronavirus transition from discontinuous to continuous transcription" *Cell Host Microbe*
38. Carlson (2022) "Reconstitution of the SARS-CoV-2 ribonucleosome provides insights into genomic RNA packaging and regulation by phosphorylation" *J. Biol. Chem*
39. Carlson (2020) "Phosphoregulation of phase separation by the SARS-CoV-2 N protein suggests a biophysical basis for its dual functions" *Mol. Cell*
40. Johnson (2022) "Global post-translational modification profiling of HIV-1-infected cells reveals mechanisms of host cellular pathway remodeling" *Cell Rep*
41. Stukalov (2021) "Multilevel proteomics reveals host perturbations by SARS-CoV-2 and SARS-CoV" *Nature*
42. Li, Wu (2021) "Pattern recognition receptors in health and diseases" *Signal Transduct. Target. Ther*
43. Rehwinkel, Gack (2020) "RIG-I-like receptors: their regulation and roles in RNA sensing" *Nat. Rev. Immunol*
44. Huang (2014) "MAVS-MKK7-JNK2 defines a novel apoptotic signaling pathway during viral infection" *PLoS Pathog*
45. Seth, Sun, Ea et al. (2005) "Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-kappaB and IRF 3" *Cell*
46. Schreiber (2020) "Dissecting the mechanism of signaling-triggered nuclear export of newly synthesized influenza virus ribonucleoprotein complexes" *Proc. Natl. Acad. Sci. USA*
47. Grams (2024) "Phosphorylation regulates viral biomolecular condensates to promote infectious progeny production" *EMBO J*
48. Botova (2024) "A specific phosphorylation-dependent conformational switch in SARS-CoV-2 nucleocapsid protein inhibits RNA binding" *Sci. Adv*
49. Nhu (2020) "Rapid reconstruction of SARS-CoV-2 using a synthetic genomics platform" *Nature*
50. Xie (2021) "Engineering SARS-CoV-2 using a reverse genetic system" *Nat. Protoc*
51. Holzberg (2003) "Disruption of the c-JUN-JNK complex by a cellpermeable peptide containing the c-JUN delta domain induces apoptosis and affects a distinct set of interleukin-1-induced inflammatory genes" *J. Biol. Chem*
52. Gaestel, Kracht (2009) "Peptides as signaling inhibitors for mammalian MAP kinase cascades" *Curr. Pharm. Des*
53. Paunovska, Loughrey, Dahlman (2022) "Drug delivery systems for RNA therapeutics" *Nat. Rev. Genet*
54. Békés, Langley, Crews (2022) "PROTAC targeted protein degraders: the past is prologue" *Nat. Rev. Drug Discov*
55. Weiterer (2020) "Distinct IL-1α-responsive enhancers promote acute and coordinated changes in chromatin topology in a hierarchical manner" *EMBO J*
56. Thiel, Siddell (2005) "Reverse genetics of coronaviruses using vaccinia virus vectors" *Curr. Top. Microbiol. Immunol*
57. Carpenter (2006) "CellProfiler: image analysis software for identifying and quantifying cell phenotypes" *Genome Biol*
58. Schindelin (2012) "Fiji: an open-source platform for biological-image analysis" *Nat. Methods* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12399564&blobtype=pdf | # Insulin-Like Growth Factor 2 mRNA Binding Protein 2 Promotes HBV-Associated Hepatocellular Carcinoma Progression by Enhancing Heme Oxygenase 1 Stability in an M 6 A-dependent Manner
Yan Zhao, Yan Cui, Hongxiu Qiao, Sandra Chiu, Xia Chuai
Dear Editor, Chronic hepatitis B virus (HBV) infection has been established as a primary etiological factor in hepatocarcinogenesis. N6-methyladenosine (m 6 A), the predominant modification of eukaryotic RNAs, has been shown to play a critical role in both HBV life cycle and HBV-associated hepatocarcinogenesis [1]. This reversible modification exerts its biological effects through specialized RNA-binding proteins ("readers") that specifically recognize m 6 A motifs and regulate RNA metabolism processes [2]. As a newly identified m 6 A reader, insulin-like growth factor 2 mRNA binding protein 2 (IGF2BP2) has been shown to promote tumorigenesis by enhancing the stability of its target transcripts [3]. Despite these findings, the role of IGF2BP2 in the specific pathogenesis of HBV-associated hepatocellular carcinoma (HCC) remains poorly understood. Therefore, our study aimed to systematically investigate the function of IGF2BP2 in HBVassociated hepatocarcinogenesis and evaluate its potential as a molecular target for HBV-associated HCC intervention.
First, we analyzed the correlation between the expression of IGF2BP2 and HBV-associated HCC (HBV-HCC). We collected liver tissue samples from HCC patients at the Third Hospital of Hebei Medical University (Hebei Province, China). The results revealed that the expression of IGF2BP2 in the liver tissue samples from both HCC groups was significantly higher than that in the paired adjacent normal tissue samples (Figure 1A). Additionally, the expression of IGF2BP2 in liver samples from the HBV-HCC group was also significantly higher than that in those from the HBV-negative group (Figure 1A). Furthermore, we detected the expression of IGF2BP2 in HBV-replicating HCC cells. The results demonstrated that the protein expression level of IGF2BP2 was also significantly elevated in HBV-replicating cells (Figure 1A).
Chronic HBV infection has been recognized as a major risk factor for HCC, and HBV-associated HCC is more aggressive than HCC caused by other factors [4]. To further clarify whether IGF2BP2 is involved in regulating the progression of HBV-HCC, we used a pCS-HBV1.3 plasmid-transfected HepG2 model to evaluate the effect of IGF2BP2 on the biological behavior of HBV-HCC. The results revealed that downregulating IGF2BP2 using shRNA significantly inhibited the proliferation, migration, and invasion capabilities of HBV-replicating HepG2 cells (Figure 1B). To further explore the role of IGF2BP2 in HCC progression in vivo, we performed xenograft tumor experiments by subcutaneously injecting shNC-or shIGF2BP2-transfected cells into nude mice (datails in Supplementary Information). We found that IGF2BP2 depletion significantly inhibited HCC growth, as reflected by reduced tumor volume and tumor weight (Figure 1C). In summary, these data demonstrate that IGF2BP2 promotes HBV-HCC progression.
## FIGURE 1
Legend on next page.
## 2 of 4
Since sorafenib and apatinib are both widely used to treat HCC, the issue of drug resistance in HCC cells during clinical treatment has garnered increasing attention. We further examined whether changes in IGF2BP2 expression affect the sensitivity of HCC cells to these two drugs. We treated shIGF2BP2-transfected HepG2 cells with sorafenib or apatinib at their IC 50 concentrations and found that IGF2BP2 knockdown enhanced the inhibitory effects of sorafenib and apatinib on HCC cells.
To further explore the mechanism by which IGF2BP2 promotes HBV-HCC tumorigenicity, we used bioinformatics analysis (https://starbase.sysu.edu.cn) to screen several downstream genes associated with HCC that can bind to IGF2BP2. Additionally, since the above results have shown that the overexpression of IGF2BP2 enhances the resistance of HepG2 cells to sorafenib and apatinib, and it was confirmed that the antitumor effects of sorafenib and apatinib are related to ferroptosis [5]. We focused on genes related to ferroptosis. Using qPCR, we found that the expression of heme oxygenase 1 (HMOX1), a key molecule involved in ferroptosis, was most significantly reduced (more than twofold change) in shIGF2BP2-transfected HepG2 cells. We further overexpressed HMOX1 in shIGF2BP2-transfected HepG2 cells and found that the inhibitory effect of IGF2BP2 downregulation on the proliferation, migration, and invasion of HCC cells was reversed (Figure 1B).
To elucidate IGF2BP2-HMOX1 interaction, RIP-qPCR was performed, demonstrating the direct binding of IGF2BP2 to HMOX1 mRNA (Figure 1D). Moreover, the shortened HMOX1 half-life was investigated following IGF2BP2 knockdown, confirming the role of IGF2BP2 in stabilizing HMOX1 mRNA (Figure 1D). To further determine the mechanism of IGF2BP2-HMOX1 interaction, MeRIP-qPCR was used to verify m 6 A modification in HMOX1 mRNA (details in Supplementary Information). Bioinformatic analysis using SRAMP predicted A185 as a high-confidence m 6 A site on HMOX1 mRNA. Subsequently, a mutation was generated at the predicted m 6 A site (A185C). Using a luciferase reporter assay, it was found that luciferase activity was significantly attenuated in HMOX1-WT cells upon IGF2BP2 knockdown, whereas in HMOX1-mut cells, it was not affected (Figure 1D). These findings demonstrate that IGF2BP2 promotes HBV-HCC progression in an m 6 A-dependent manner by stabilizing HMOX1 mRNA.
In summary, our study reveals for the first time that IGF2BP2 is involved in the development of HBV-HCC. Our findings provide a new perspective on HBV-HCC development. IGF2BP2 might be a potential therapeutic target for HBV-HCC. Currently, this study has confirmed that IGF2BP2 facilitates the progression of HBV-HCC by targeting HMOX1 in an m 6 A-dependent manner.
However, whether IGF2BP2 promotes HBV replication and how its interaction with HMOX1 affects the development of HBV-HCC remain to be further studied.
## FIGURE 1
IGF2BP2 promotes HBV-associated hepatocellular carcinoma (HCC) progression through m 6 A-dependent stabilization of HMOX1. (A) Comparative analysis of IGF2BP2 expression levels (the integrated optical density [IOD] per field was calculated for quantification) in hepatocellular carcinoma (HCC) tissues and adjacent non-tumor liver tissues from HCC patients (n = 6-9 per group), and western blot detection of IGF2BP2 expression in HBV-replicating HCC cells (n = 3). (B) Functional characterization of IGF2BP2 knockdown in HepG2 cells through colony formation assay (n = 3), wound healing assay (n = 6), and transwell migration assay (n = 6), demonstrating that IGF2BP2 depletion significantly impairs proliferative, migratory, and invasive capacities. (C) In vivo tumorigenicity assessment showing reduced xenograft growth in nude mice subcutaneously transplanted with shIGF2BP2 HepG2 cells (n = 5 mice/group). Quantitative analysis revealed significant reductions in both tumor volume and weight. (D) RNA immunoprecipitation (RIP)-qPCR, RNA stability assay, and dual-luciferase reporter assays (n = 3) (WT: wild type of HMOX1 mRNA; Mut: a mutation of the predicted m 6 A site (A185C) on HMOX1 mRNA) were used to delineate the IGF2BP2-HMOX1 interaction and validate m 6 A-dependent post-transcriptional regulation. *p < 0.05, **p < 0.01, ***p < 0.001.
## References
1. Yang, Yan, Yin et al. (2023) "O-GlcNAcylation of YTHDF2 Promotes HBV-Related Hepatocellular Carcinoma Progression in an N(6)-Methyladenosine-Dependent Manner" *Signal Transduction and Targeted Therapy*
2. Zhao, Shi, Shen et al. (2020) "A-binding Proteins: The Emerging Crucial Performers in Epigenetics" *Journal of Hematology & Oncology*
3. Sun, Cao, Du et al. (2022) "The Role of Insulin-Like Growth Factor 2 mRNA-Binding Proteins (IGF2BPs) as M(6)A Readers in Cancer" *International Journal of Biology Sciences*
4. Yang, Ma, Yang (2019) "Contribution of Hepatitis B Virus Infection to the Aggressiveness of Primary Liver Cancer: A Clinical Epidemiological Study in Eastern China" *Frontiers in Oncology*
5. Xie, Zhou, Lin et al. (2023) "Binding of Berberine to PEBP1 Synergizes With Sorafenib to Induce the Ferroptosis of Hepatic Stellate Cells" *Amino Acids* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12476789&blobtype=pdf | # Uptake and disparities in tuberculosis screening using urinelipoarabinomannan among patients with advanced human immunodeficiency virus-disease in Africa: A systematic review
Lynn Moshi, Hafidha Bakari, Vicent Mbishi, Zuhura Ally, Mariam Mbwana, Haji Ally, Rahma Musoke, Swalehe Salim, Francis Karia, Leticia Karia, Hassan Fussi, Aboubakar Mustafa, Ibrahim Ahmed, Habib Ramadhani, Liang Gd, Jackline Mbishi, Francis Maximillian, Karia, Ibrahim El-Lmam, Hafidha Mhando
## Abstract
BACKGROUNDDue to low bacteria count and high likelihood of having extrapulmonary tuberculosis (TB) among patients with advanced human immunodeficiency virus (HIV) disease, the World Health Organization (WHO) recommended the use of urine lateral flow urine lipoarabinomannan (LF-LAM) or sputum-Xpert to screen for TB.
AIMTo estimate pooled prevalence of TB screening uptake, TB diagnosis, TB treatment initiation and mortality among patients with advanced HIV disease in Africa.
METHODSPubMed, Cochrane Library and EMBASE were searched for articles published between January 2011 and December 2024. TB screening uptake was defined as percentage of patients with advanced HIV disease (CD4 ≤ 200 cells/mm 3 or WHO stage III/IV) who tested for TB. Using random effects models, we computed the pooled estimate of TB screening uptake, TB prevalence, TB treatment initiation and mortality and their corresponding 95%CIs. Stratified analysis to compare uptake of TB testing and TB prevalence between children vs adults and multisite vs single site studies was performed.
RESULTSA total of nineteen studies with 16065 people with advanced HIV disease were analyzed. The pooled prevalence of TB screening uptake was 64.6% (95%CI: 49.2-80.1). The pooled prevalence of TB was 29.4% (95%CI: 22.0-36.8), and TB treatment initiation was 77.9% (95%CI: 63.9-91.8), and mortality was 19.5% (95%CI: 8.9-30.0). The pooled prevalence of TB testing uptake was significantly lower among children compared to adults (28.2% vs 66.4%, P = 0.003) and lower for multi-sites compared to single site studies (58.8% vs 82.9%, P = 0.002). The pooled prevalence of TB was significantly lower among children compared to adults (24.2% vs 27.6%, P = 0.012) and higher among studies that involved multi vs single sites (30.0% vs 21.9%, P = 0.001).
CONCLUSIONFour in ten people with advanced HIV disease were not screened for TB as recommended by the WHO, indicating significant gaps in identifying patients with TB. Excluding patients with evidence of TB is critical to avoid exposing them to subtherapeutic levels of anti TB treatment.
## INTRODUCTION
Stigma against human immunodeficiency virus (HIV) continues to impact people living with HIV and as a result some patients present late with advanced HIV disease (CD4 count ≤ 200 cells/mm 3 or World Health Organization (WHO) stage III/IV) at the time of HIV diagnosis despite an expanded anti-retroviral (ART) coverage [1][2][3][4]. From 2011, studies that used CD4 count threshold of ≤ 100 cells/mm 3 , estimated the prevalence of advanced HIV disease for newly diagnosed individuals to range from 13.9% to 33.6% [5,6], while those which used a higher threshold of ≤ 200 cells/mm 3 reported a broader range of 17.2% to 71% [7,8] reflecting differences in study population, healthcare settings, and diagnostic thresholds. Advanced HIV disease is associated with negative consequences including delayed time to viral load suppression [9,10], increased risk of onward HIV transmission, increased treatment costs [11] and life-threatening opportunistic infections such as cryptococcal meningitis and tuberculosis (TB) [12,13]. Despite increased access to ART, advanced HIV disease is the predominant concern for an ongoing acquired immunodeficiency syndrome (AIDS) related mortality [14].
Globally, it is estimated that 4.3 million adults are living with advanced HIV disease [15]. TB remains the leading cause of HIV related mortality globally, accounting for approximately 30% of HIV related mortality annually [16]. Among HIV patients co-infected with TB, provision of both ART and anti TB treatment averted 6.4 million deaths between 2010 and 2020 globally [17]. Of the 6.4 million deaths averted, three quarters were from Africa. Screening for TB and treatment of latent TB infection reduce mortality and years of potential life lost due to active TB disease [18]. These data underscore the need for accurate diagnosis and treatment.
Due to low bacteria count and high likelihood of having extrapulmonary TB among patients with advanced HIV disease, recognizing these limitations, the WHO recommended the use of more sensitive rapid molecular diagnostic tests such as lateral flow urine lipoarabinomannan (LF-LAM) or sputum-Xpert as they offer greater sensitivity than the traditional smear microscopy for TB diagnosis. The United Nation targets requires that by 2027 the coverage of rapid diagnostic testing for TB to be 100% [19]. Overtime, there has been variation on the uptake of LF-LAM or sputum Xpert from different studies with the reported testing uptake being between 5% to 98% [20,21]. The variation in the uptake of these tests could be attributed but not limited to the availability of testing kits, reagents and technical capacity [13,22]. In addition, unawareness of WHO guidelines, data sources from published literature [(single vs multi sites studies); types of healthcare facilities (primary, secondary and tertiary)] could explain differences in the reported TB testing uptake using these rapid tests as previously noted by other researchers [23]. Data on the overall magnitude and compliance of WHO guidelines on the uptake of TB testing using LF-LAM is limited. Given these inconsistencies, a systematic review and meta-analysis is warranted to provide pooled estimates and identify disparities by age and study settings. In this review, we are addressing the following questions for studies conducted in Africa. Among patients with advanced HIV disease, what percent were tested for TB using WHO recommended rapid TB diagnostic molecular test of LF-LAM? (TB uptake):
(1) Among those tested, what percentage tested positive? (2) Among those who tested positive, what percentage-initiated TB treatment? and (3) What is mortality rate among patients diagnosed with advanced HIV disease who are coinfected with TB?
## MATERIALS AND METHODS
## Registration
The protocol for this systematic review has been registered in the International Prospective Registry of Systematic Review on October 31, 2024 with registration number CRD42024604091.
## Ethical approval
Because this was a systematic review of published manuscripts, ethical approval was not sought.
## Search strategy
We searched PubMed, Cochrane CENTRAL, EMBASE and clinicaltrials.gov for the articles published between January 2011 and December 2024. The search period represents the time at which WHO guidelines for the management of patients with advanced HIV disease was operational. Search terms were used to capture information on TB testing using urine LF-LAM among people living with HIV who presented with advanced HIV disease in Africa. The search was restricted to papers written in English. Search results were uploaded to Covidence Systematic Review Software (Melbourne, Australia) for deduplication and screening.
## Eligibility criteria
Studies were included based on the guidelines of systematic reviews and meta-analysis with prevalence approach (CoCoPop) where the first and second 'Co' indicates condition/problem and context respectively and 'Pop' indicates study population as previously described [24]. In this review, under condition, we included studies that reported prevalence of TB testing uptake using LF-LAM among people presenting with advanced HIV disease; Context, we explored components of different studies that could explain variation on the reported prevalence of TB testing uptake such as study designs, number of study sites (single vs multiple sites), type of health care facilities (primary, secondary or tertiary); Population, we included people with advanced HIV disease.
## Inclusion criteria
Observational studies that involved people with advanced HIV disease in Africa, reported number and percentage of individuals who tested for TB using LF-LAM and written in English were eligible for inclusion.
## Exclusion criteria
Studies that reported the percentage of people who tested for TB without actual numerators and denominators used to compute those percentages. Studies that were not written in English.
A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (Figure 1) describing the literature search process and included studies is presented below. This systematic review and meta-analysis was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [25].
## Study selection
The manuscripts searched from outlined databases were managed by Covidence software from which the final list of manuscripts was deduplicated. Two pairs of authors (Haji Mbwana Ally, Habib Omari Ramadhani and Zuhura Mbwana Ally, Hassan Fredrick Fussi) independently completed the study selection for inclusion in the appraisal process. Disagreement between two independent pairs of authors for the inclusion of the manuscripts was handled by the third pair of authors (Mariam Salim Mbwana and Ibrahim Ahmed El-lmam).
## Data extraction
Using a pre specified excel spread sheet template, two pairs of authors (Haji Mbwana Ally, Habib Omari Ramadhani and Zuhura Mbwana Ally, Hassan Fredrick Fussi) independently extracted the following data elements from the included studies: Authors, year of publication, the country in which the study was conducted, the year(s) at which data was collected, study design, study sites (single vs multi sites), TB test used (LF-LAM), number of people with advanced HIV disease, number of people eligible for TB testing, number of people tested for TB, number of people with positive TB test, number of people with positive TB test who initiated treatment, number of people with positive TB test and died. The data were then compared, and any disagreements between the two pairs of reviewers were resolved by consensus; the third pair of reviewers (Mariam Salim Mbwana and Ibrahim Ahmed El-lmam) was consulted when necessary. Strategies in the Cochrane Handbook for Systematic Reviews of Interventions for data management were followed [26].
## Quality assessment
The Joanna Briggs Institute (JBI) tools for cross sectional and cohort studies were used to assess quality of studies [27]. Two pairs of reviewers (Habib Omari Ramadhani and Haji Mbwana Ally) and (Hassan Fredrick Fussi and Lynn Moshi) independently performed and rated the quality of the studies using the JBI tools. The tool encompassed nine questions with four responses: (Yes, No, Not clear, Not applicable). A score of 1 was assigned to a "Yes" response and 0 to a "No" response. Total score was summed up and categorized into three groups, with (0-3), (4-6) and (7-9) scores indicating low, medium and high quality.
## Definition of variables
Outcome variables: The main outcome of interest was the TB testing uptake defined as percentage of individuals who tested for TB using LF-LAM among those who presented with advanced HIV disease. The secondary outcome were: (1) The prevalence of TB defined as percentage of individuals with positive LF-LAM among those tested; (2) Percentage of people who initiated treatment among those with positive TB test; and (3) Mortality among those diagnosed with TB.
## Exposure variables:
There was no main exposure variable, however, TB testing uptake and TB prevalence were compared between children and adults and between studies that involved multiple sites vs single site.
## Data synthesis and statistical analysis
Using random effects model, we computed pooled prevalence of TB testing uptake, TB prevalence, treatment initiations and mortality. Since there was no extreme proportions (as low as '0' or as high as '1') requiring variance stabilization, all analyses were conducted using raw proportions, allowing for direct clinical interpretation of the pooled estimates. Subgroup analysis on the pooled estimates of TB testing uptake and TB prevalence were performed to compare children vs adults and between studies that involved multiple sites vs single using χ 2 tests. We evaluated heterogeneity across studies using the I 2 statistic and Cochran's Q test. The I 2 statistics explain the variance attributable to study heterogeneity with scores of 75%, 50% and 25% indicated high, moderate and low heterogeneity, respectively [28]. The publication bias was assessed using the Egger regression asymmetry test. For both heterogeneity and publication test, a P < 0.05 indicated the presence of heterogeneity and publication bias respectively. To explore the source of heterogeneity for outcome with moderate or higher degree of heterogeneity, an influential analysis using the leave-one-out method was performed [29]. Furthermore, a meta regression analysis was done to assess variation of the uptake of TB screening across studies and a sensitivity analysis to assess potential small-study effects was conducted using a trim-and-fill method. Studies with missing information, such as those that reported proportions of outcomes without actual numerators and/or denominators, were excluded from the analysis. All statistical tests were performed using STATA version 17 (Stata Corporation, College Station, Texas, United States).
## RESULTS
## Study selection
Our search resulted in 1100 articles. Of these, 243 were duplicates and deleted. The remaining 857 articles were eligible for title and abstracts screening. Of the twenty-six articles eligible for full text review, nineteen met inclusion criteria and were finally included in our analysis (Figure 1).
## Characteristics of studies included
A total of nineteen studies with 16065 people with advanced HIV disease were analyzed. The review included studies conducted by thirteen countries in Africa. Sample size of included studies ranged from 97 to 5487 (Table 1). All studies reported TB testing uptake, and results of TB testing, eight reported TB treatment initiations and seven reported mortality data. Two studies reported TB testing uptake among children [30,31] and thirteen among adults [12,20,21,[32][33][34][35][36][37][38][39][40][41] and four among both children and adults [7,13,42,43]. A total of fourteen studies involved multiple sites and three involved single sites and two not specified.
## Study quality assessment
Assessment of study quality showed that 16 (84.2%) were of high quality, and 3 (15.8%) were of moderate quality. No study was of low quality (Table 1).
## Prevalence of TB testing uptake and subgroup analysis
The meta-analysis assessed the prevalence of TB testing uptake, with the pooled prevalence of 64.6% (95%CI: 49.2-80.1) (Figure 2A), indicating that nearly only two-thirds of individuals eligible for TB testing were TB tested. A subgroup analysis based on population type revealed that children had a pooled testing uptake of 28.2% (95%CI: 5.6-62.0) compared to adults 61.3% (95%CI: 42.6-80.1), P value 0.003 (Supplementary Figure 1). Furthermore, testing uptake by study sites showed that multi-site studies had a pooled testing uptake of 58.8% (95%CI: 39.8-77.7) compared to 82.9% (95%CI: 67.6-98.1), among studies that involved single sites, (P = 0.002) (Supplementary Figure 2).
A meta-regression including year of publication, number of eligible participants, number of study site, study design, and criteria used to define advanced HIV disease did not identify any statistically significant predictors of screening uptake (all P > 0.05) (Supplementary Table 1), and residual heterogeneity remained high (I 2 = 99.7%), suggesting that unmeasured contextual or implementation-related factors likely contribute to the variability in TB screening uptake.
## Assessment of heterogeneity, publication bias, and influential analysis on the prevalence of testing uptake
Heterogeneity was observed in the meta-analysis (I 2 = 99.9%), indicating substantial variability across studies. Egger's test was conducted to assess potential publication bias, with results showing a beta coefficient of -6.64 (SE = 4.833), a z value of -1.37, and a P = 0.1698, suggesting no significant evidence of small-study effects. In addition, a trim-and-fill sensitivity analysis was performed to further assess potential publication bias. The analysis did not impute any missing studies, and the adjusted pooled effect size remained unchanged at 0.647 (95%CI: 0.515-0.778) (Supplementary Table 2), supporting the conclusion that small-study effects are unlikely to have influenced the findings.
Influential analysis was assessed using a leave-one-out sensitivity approach, where all studies had P values less than 0.001, indicating that no single study disproportionately influenced the pooled estimate of testing uptake (Supplementary Figure 3).
## Pooled prevalence of TB and subgroup analysis
The meta-analysis estimated the pooled prevalence of TB at 29.4% (95%CI: 22.0-36.8) (Figure 2B), indicating slightly higher than a quarter of individuals were diagnosed with TB. A subgroup analysis by type of population showed that TB prevalence among children was 24.2% (95%CI: 12.4-36.1) compared to 27.6% (95%CI: 17.2-38.0) among adults, (P = 0.001) (Supplementary Figure 4). Furthermore, subgroup analysis by study site revealed that multi-site studies had a slightly higher pooled TB prevalence of 30.0% (95%CI: 21.6-38.5), compared to 21.9% (95%CI: 13.6-30.3) among studies that involved single-site (P = 0.001) (Supplementary Figure 5).
## Assessment of heterogeneity, publication bias, and influential analysis on the prevalence of TB
Heterogeneity was observed in the meta-analysis (I 2 = 98.4%), indicating substantial variability across studies. Egger's test was conducted to assess potential publication bias, yielding a beta coefficient of 3.82 (SE = 2.818), a z value of 1.36, and a P = 0.1754, suggesting no significant evidence of small-study effects. Influential analysis was assessed using a leave-one- out sensitivity approach, where all studies had P values less than 0.001, indicating that no single study disproportionately influenced the pooled estimate of TB prevalence (Supplementary Figure 6).
## Pooled prevalence of mortality
The meta-analysis estimated the pooled prevalence of mortality at 19.5% (95%CI: 8.9%-30.0%) (Figure 2C), indicating that nearly one in five individuals diagnosed with TB died. A subgroup analysis on mortality was not performed because there was only one study from a single site that reported mortality data.
## Assessment of heterogeneity, publication bias, and influential analysis on the prevalence of mortality
Heterogeneity was observed (I 2 = 86.4%), indicating variability across studies. Egger's test was conducted to assess potential publication bias, yielding a beta coefficient of 5.08 (SE = 1.481), a z value of 3.43, and a P value of 0.0006, suggesting significant evidence of small-study effects. Influential analysis was assessed using a leave-one-out sensitivity approach, where all studies had P values less than 0.001, indicating that no single study disproportionately influenced the pooled estimate (Supplementary Figure 7).
## Pooled prevalence of TB treatment initiations
The meta-analysis estimated that 77.9% (95%CI: 63.9%-91.8%) of those diagnosed with TB-had documentation of treatment initiations (Figure 2D). Heterogeneity was observed (I 2 = 98.5%), reflecting considerable variability in the prevalence of treatment initiations across studies.
## Factors associated with the uptake of LF-LAM
Generally, from the studies included in this analysis, there was no individual factors that were identified to be associated with the uptake of LF-LAM, however, several healthcare facility factors have been documented for the low uptake of LF-LAM. These included inconsistencies in measuring CD4 count to identify people with advanced HIV disease [32,34,38,40], lack of human resource for testing [13,36], and results interpretation [34], lack of testing kits and other commodities such as urine cups [38,39]. Increased awareness of systematic use of LF-LAM was associated with increased uptake of LF-LAM [30]
## DISCUSSION
We conducted a systematic review and meta-analysis to understand compliance of the WHO guidelines in the management of patients with advanced HIV disease in Africa. Specifically, we summarized pooled estimate of those eligible for TB testing using LF-LAM. The pooled estimates of LF-LAM uptake were 65% with 28% among children and 66% among adults. Of those tested, 29% were diagnosed with TB and of these, 78% initiated treatment and 20% died. These data indicate gaps in the diagnosis of TB as well as significant disparities in TB testing for children compared to adults.
According to the 2024 Global TB report, number of deaths due to TB is nearly twice that caused by HIV/AIDS indicating how lethal is TB [44]. If diagnosed and treated accordingly, TB is a curable disease. Patients with HIV, particularly those with advanced disease, are at substantial risk of acquiring TB necessitating early diagnosis and treatment. Low uptake of TB testing among people with advanced HIV disease is concerning. Previous study showed that reasons for the low adoption of LF-LAM in the diagnosis of TB included country's budgetary constraints, lack of country-specific data and piloting, delays of regulatory agency approval, lack of coordination between National TB and HIV programs, as well as perceived small population needing this service [45,46]. Other views of the low uptake of TB testing using urine LF-LAM included lack of personnel administering the tests[13], and concerns of test's low sensitivity [21], restrictive eligibility criteria, reliance on CD4 testing, and lack of advocacy and awareness [22]. On the other hand, prior studies showed that LF-LAM is low cost, easy to use and its implementation is feasible [13,34] and useful in severely ill patients who cannot produce sputum or sputum scarce patients [21]. In addition, because TB testing using LF-LAM is a same day event, the technique reduces the number of clinic visits needed if the patient was to collect sputum that needs up to three clinic visits [47]. Collectively, the benefits of using LF-LAM for TB diagnosis outweigh its deficiencies and therefore investing on this diagnostic technology is critical for early detection and treatment of TB to reduce mortality associated with TB. We observed disparities in TB testing uptake using LF-LAM between studies involving children and adults, with lower uptake reported among children. These findings require additional studies to discern these disparities beyond the known general reasons for its low uptake. For example, children's specific challenges such as difficulties of urine collection, performance variability may cause a low uptake of LF-LAM in children compared to adults. TB testing uptake was disproportionately higher in studies that involved single sites compared to those which involved multiple sites. As expected, most often, data from single site studies were from tertiary hospitals which are usually more resourceful compared to multiple sites studies that encompasses healthcare facilities of different tier [23].
Mortality among those diagnosed with TB is unacceptably high. Most of the patients presenting with advanced HIV are severely ill. Co-infection with TB further complicates their management. As previously narrated, one of the most common barriers in delaying seeking care among people living with HIV is stigma[1, 3,4]. Because of these delays, patients present to healthcare facility with advanced HIV disease, often co-infected with opportunistic infections such as TB. In this era of universal access to antiretroviral therapy, addressing HIV related stigma will minimize proportion of people presenting with advanced HIV disease as well as mortality associated with these co-infections.
Nearly 78% of those diagnosed with TB have documentation of treatment initiations. Previous studies have documented low TB treatment initiation after positive LF-LAM test [37,48]. Hesitance of treatment initiation can be due to the possibility of false positive results leading to the lack of trust on positive test results [49]. Given high mortality, it is also likely that some patients died prior to TB treatment initiation. TB testing reagents, stigma associated with TB diagnosis, transport and training costs are possible barriers to using LF-LAM and therefore studies to understand barriers of TB treatment initiation following positive LF-LAM are needed to maximize the benefits of the tests.
As stated previously, stigma against HIV impact people living with HIV and as a result some patients present late with advanced HIV disease. Poor immune status, advanced HIV-related disease, and the combined effects of both infections promote mortality and therefore, early HIV diagnosis would reduce mortality associated with TB in people living with HIV.
We acknowledge limitations of this systematic review. The study population involved people with advanced HIV disease based on the WHO definition. The variability of using either CD4 ≤ 200 or WHO stage III/IV to define advanced HIV disease potentially skewed pooled estimates, however, we could not exclusively distinguish which study used either criterion alone as most used both criteria. The overall uptake of LF-LAM may not be reflective of the actual uptake in Africa because of the use of data from small studies either from single or multiple sites. While using data from multiple sites encompasses healthcare facilities with different tiers, representing variations in resources and hence mimicking National data, the use of clinic auditing from the National databases would be an ideal to evaluate LF-LAM uptake. Adjusting for healthcare facility type would be more informative, however, inability to exclusively distinguish data from tertiary vs primary healthcare facilities limited the possibility to adjust for this factor in meta regression. Additionally, although meta-regression was conducted to explore sources of heterogeneity, the model explained only 5% of the between-study variance and none of the included variables were statistically significant. The high residual heterogeneity (I 2 = 99.72%) suggests that unmeasured contextual or implementation-related factors likely contribute to the variability in TB screening uptake. Furthermore, only two studies reported uptake among children while thirteen studies reported uptake among adults. The small number of studies that reported children may under/overestimate the actual TB testing uptake among children. We reported TB treatment initiation, however, time to treatment initiation was not reported by majority of studies involved. Only 37% of the studies reported mortality data limiting our ability to get a clear overview of the mortality data for the entire population involved. In addition, we restricted our search to publications written in English, it is likely that other relevant publications from non-English journals were missed. The main strength of this research is the inclusion of 19 studies leadings to an overall large sample size to compute pooled estimates of LF-LAM testing uptake. This meta-analysis remains relevant as it is the first study that provides pooled estimates of the uptake of LF-LAM testing to understand the magnitude of compliance of WHO recommendations in the management of people presenting with advanced HIV disease in Africa.
## CONCLUSION
We identified significant gaps in the diagnosis of TB using LF-LAM among people with advanced HIV disease in Africa.
Our findings indicate suboptimal compliance of WHO guidelines in the management of people with advanced HIV disease. Mortality is high among those diagnosed with TB and to reduce mortality associated with TB, uptake of TB diagnosis should be heightened. As previously discussed, appropriate coordination between national TB and HIV programs must be strengthened. In country policies should consider investing in diagnostic tools, increase awareness, training of healthcare providers on the use of LF-LAM would improve its uptake, increase early detection and treatment and hence reduce mortality. Despite LF-LAM limitations, clinicians should have a high level of suspicion of TB diagnosis by considering factors such as prior history of TB and the presence of cough and night sweats in people living with HIV to increase TB diagnosis and treatment initiations. We recommend routine evaluation of national data on the compliance of these guidelines. These evaluations are critical to identify gaps and improve performance.
## References
1. Moshi, Bakari, Ho ; Ally et al. (2009) "Ibrahim AE and Ramadhani HO contributed to writing review and editing. All authors reviewed this manuscript, provided feedback, and approved the manuscript in its final form. Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article" *Karia MF contributed to writing original draft Bakari HM and Moshi L contributed to visualization*
2. "Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial"
3. Salim
5. (2019) "Late presentation of HIV positive adults and its predictors to HIV/AIDS care in Ethiopia: a systematic review and meta-analysis"
6. Assen, Molla, Wondimu et al. (2016) "Late presentation for diagnosis of HIV infection among HIV positive patients in South Tigray Zone, Ethiopia" *BMC Public Health*
7. Chone, Abecasis, Varandas (2022) "Determinants of Late HIV Presentation at Ndlavela Health Center in Mozambique" *Int J Environ Res Public Health*
8. Lofgren, Tsui, Natala et al. (2023) "Differences in Reasons for Late Presentation to HIV Care in Uganda Among Men and Women" *AIDS Behav*
9. Drain, Losina, Parker et al. (2013) "Risk factors for late-stage HIV disease presentation at initial HIV diagnosis in Durban, South Africa" *PLoS One*
10. Kujawski, Lamb, Lahuerta et al. (2017) "Advanced Human Immunodeficiency Virus Disease at Diagnosis in Mozambique and Swaziland" *Open Forum Infect Dis*
11. Benzekri, Sambou, Ndong et al. (2019) "Prevalence, predictors, and management of advanced HIV disease among individuals initiating ART in Senegal, West Africa" *BMC Infect Dis*
12. Lebelonyane, Mills, Mogorosi et al. (2020) "Advanced HIV disease in the Botswana combination prevention project: prevalence, risk factors, and outcomes" *AIDS*
13. Erjino, Abera, Tirore (2023) "Time to Viral Load Suppression and Its Predictors Among Adult Patients on Antiretro Viral Therapy in Nigist Eleni Mohammed Memorial Comprehensive Specialized Hospital" *HIV AIDS (Auckl)*
14. Palmer, Gabler, Rachlis et al. (2018) "Viral suppression and viral rebound among young adults living with HIV in Canada" *Medicine (Baltimore)*
15. Carmona, Bor, Nattey et al. (2018) "Persistent High Burden of Advanced HIV Disease Among Patients Seeking Care in South Africa's National HIV Program: Data From a Nationwide Laboratory Cohort" *Clin Infect Dis*
16. Heller, Damba, Kumwenda et al. (2022) "Implementing Advanced HIV Disease Care for Inpatients in a Referral Hospital in Malawi -Demand, Results and Cost Implications" *Ann Glob Health*
17. Kanyama, Chagomerana, Chawinga et al. (2022) "Implementation of tuberculosis and cryptococcal meningitis rapid diagnostic tests amongst patients with advanced HIV at Kamuzu Central Hospital" *BMC Infect Dis*
18. (2023) "The path that ends AIDS: UNAIDS Global AIDS Update"
19. Rajasingham, Govender, Jordan et al. (2022) "The global burden of HIV-associated cryptococcal infection in adults in 2020: a modelling analysis" *Lancet Infect Dis*
20. ":~:text=Tuberculosis%20(TB)%20remains%20the%20leading"
21. Lee-Rodriguez, Wada, Hung et al. (2020) "Association of Mortality and Years of Potential Life Lost With Active Tuberculosis in the United States" *JAMA Netw Open*
22. Hassan, Senkoro, Mnyambwa et al. (2022) "Implementation of WHO guidelines on management of advanced HIV disease and its impact among TB co-infected patients in Tanzania: a retrospective follow-up study" *BMC Public Health*
23. Wake, Govender, Omar et al. (2022) "Rapid urine-based screening tests increase the yield of same-day tuberculosis diagnoses among patients living with advanced HIV disease" *AIDS*
24. Mwaura, Engel (2021) "Constructing confidence: User perspectives on AlereLAM testing for tuberculosis" *Int J Infect Dis*
25. Ally, Mbishi, Mbwana et al. (2025) "Systematic review on the compliance of WHO guidelines in the management of patients with advanced HIV disease in Africa: The case of cryptococcal antigen screening" *PLoS One*
26. Munn, Stern, Aromataris et al. (2018) "What kind of systematic review should I conduct? A proposed typology and guidance for systematic reviewers in the medical and health sciences" *BMC Med Res Methodol*
27. Page, Mckenzie, Bossuyt et al. (2021) "The PRISMA 2020 statement: an updated guideline for reporting systematic reviews" *BMJ*
28. Higgins, Thomas, Chandler et al. "Cochrane Handbook for Systematic Reviews of Interventions"
29. Munn, Moola, Lisy et al. (2015) "Methodological guidance for systematic reviews of observational epidemiological studies reporting prevalence and cumulative incidence data" *J Evid-Based Healthc*
30. Higgins, Thompson, Deeks et al. (2003) "Measuring inconsistency in meta-analyses" *BMJ*
31. Bakari, Alo, Mbwana et al. (2023) "Same-day ART initiation, loss to follow-up and viral load suppression among people living with HIV in low-and middle-income countries: systematic review and meta-analysis" *Pan Afr Med J*
32. Kingbo, Bitty-Anderson, Denoeud-Ndam et al. "Improving identification and management of advanced HIV disease among people living with HIV in Cote d'Ivoire"
33. Onwah, Nwanja, Toyo et al. "Improving uptake of tuberculosis testing using urine Lipoarabinomannan among children with advanced HIV disease: outcomes of a quality improvement initiative in southern Nigeria"
34. Mbewe, Vinikoor, Fwoloshi et al. (2022) "Advanced HIV disease management practices within inpatient medicine units at a referral hospital in Zambia: a retrospective chart review" *AIDS Res Ther*
35. Lukoye "Scaling up TB LAM testing and linkage to TB treatment among ART Naïve HIV patients by the IDI in urban and rural Uganda"
36. Rucker, Cossa, Harrison et al. (2019) "Feasibility of using Determine TB-LAM to diagnose tuberculosis in HIV-positive patients in programmatic conditions: a multisite study" *Glob Health Action*
37. Kalyango "Uptake of urine lipoarabinomannan and associated factors in diagnosis of tuberculosis among patients with advanced HIV disease in selected Kampala Capital City Authority HIV clinic"
38. Åhsberg, Bjerrum, Ganu et al. (2023) "The inhospital tuberculosis diagnostic cascade and early clinical outcomes among people living with HIV before and during the COVID-19 pandemic -a prospective multisite cohort study from Ghana" *Int J Infect Dis*
39. Åhsberg, Puplampu, Kwashie et al. (2023) "Point-of-Care Urine Lipoarabinomannan Testing to Guide Tuberculosis Treatment Among Severely Ill Inpatients With Human Immunodeficiency Virus in Real-World Practice: A Multicenter Stepped Wedge Cluster-Randomized Trial From Ghana" *Clin Infect Dis*
40. Nyukuri, Walata, Alwang et al. "Use of tuberculosis lipoarabinomannan to improve management of and avert mortalities among newly diagnosed people living with HIV with advanced HIV disease: Findings from Kakamega County"
41. Levy-Braide, Otubu, Abudiore et al. "Lessons Learned from the introduction of advanced HIV disease package of care in Nigeria"
42. Nakanwagi, Namuwenge, Arinaitwe et al. "Low utilization of TB-LAM for TB screening among children under five with advanced HIV disease in Uganda: a descriptive analysis of surveillance data"
43. Matoga, Bisson, Gupta et al. (2021) "Urine Lipoarabinomannan Testing in Adults With Advanced Human Immunodeficiency Virus in a Trial of Empiric Tuberculosis Therapy" *Clin Infect Dis*
44. Izco, Murias-Closas, Jordan et al. (2021) "Improved detection and management of advanced HIV disease through a community adult TB-contact tracing intervention with same-day provision of the WHO-recommended package of care including ART initiation in a rural district of Mozambique" *J Int AIDS Soc*
45. Eigege, Levybraideb, Otubu et al. "Improved tuberculosis case finding among patients with Advanced HIV Disease in Nigeria through the deployment of a point of care tuberculosis diagnostic test"
46. "Global tuberculosis report"
47. Singhroy, Maclean, Kohli et al. (2020) "Adoption and uptake of the lateral flow urine LAM test in countries with high tuberculosis and HIV/AIDS burden: current landscape and barriers" *Gates Open Res*
48. Nathavitharana, Pai (2018) "New strategies for inpatients with HIV and tuberculosis" *Lancet*
49. Blok, Van Den Hof, Mfinanga et al. (2012) "Measuring workload for tuberculosis service provision at primary care level: a methodology" *Hum Resour Health*
50. Floridia, Ciccacci, Andreotti et al. (2017) "Tuberculosis Case Finding With Combined Rapid Point-of-Care Assays (Xpert MTB/ RIF and Determine TB LAM) in HIV-Positive Individuals Starting Antiretroviral Therapy in Mozambique" *Clin Infect Dis*
51. Gils, Madonsela, Kamele et al. (2024) "Low tuberculosis treatment initiation after positive tuberculosis lipoarabinomannan results" *ERJ Open Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12510953&blobtype=pdf | # Editorial: Advances in understanding the interplay of soil carbon, iron, and arsenic transformation
Ruiyong Zhang, Huihui Du, Williamson Gustave, Zhao-Feng Yuan, Xiangfeng Tan, Weiwei Zhai
Soils are critical regulators of elemental cycling, mediating interactions among carbon (C), iron (Fe), and arsenic (As) that influence ecosystem function, climate regulation, and environmental health. Soil organic carbon (SOC) represents the largest terrestrial carbon reservoir and is often considered a natural solution for mitigating climate change. However, the stability of SOC is strongly mediated by its interactions with Fe oxides and hydroxides, which can either protect organic matter through mineral associations or promote its loss under reducing conditions (Xu and Tsang, 2024;Hu et al., 2025). The transformation of Fe, in turn, directly governs the fate of As, one of the most hazardous environmental contaminants, through adsorption, reduction, and microbial methylation pathways (Gao et al., 2024;Tang et al., 2024).
The complex interplay of these processes highlights the need for integrative approaches that couple soil chemistry, microbial ecology, and environmental engineering. While Fe minerals provide protective surfaces that stabilize SOC, their reductive dissolution can release both C and As into more labile pools, with microorganisms playing central roles in mediating redox transformations (Yao et al., 2023;Wang et al., 2024a). Horizontal gene transfer and viral interactions further add layers of complexity (Wang et al., 2024b;Liang et al., 2025). Yet, despite growing evidence for these interconnections, major knowledge gaps remain, particularly in predicting the dynamics of C-Fe-As coupling under fluctuating redox conditions and in understanding the roles of less-studied microbial groups.
To address these challenges, this Research Topic brings together six contributions spanning mechanistic, methodological, and applied perspectives. Collectively, they shed light on the stabilizing functions of minerals, the roles of microbes in soil health and disease, the use of engineered amendments and bioreporters, and innovative bioelectrochemical approaches to pollution management.
One set of studies focuses on the mineralogical and structural controls of soil processes. Li and Guo revealed how soil microaggregates vary along an elevation gradient in Tongbai Mountain, with Mn-and Fe-rich microaggregates at low elevations promoting metal mobility, while high elevations favored the formation of organo-mineral complexes that stabilized C, N, and Fe. This highlights how landscape position can mediate elemental coupling and provides a framework for anticipating how mountain soils may respond to climate change. In parallel, Chen et al. explored engineered soil amendments, showing that silicon-iron modified biochars effectively reduced the bioavailability of Cd and As in paddy soils by altering speciation pathways. Their work emphasizes the potential for Fe-and Si-based additives to immobilize contaminants while simultaneously influencing microbial functional genes involved in As oxidation and Cd precipitation. Together, these studies highlight the central role of Fe mineral phases, whether natural or engineered, in regulating C-Fe-As transformations and contaminant dynamics.
A second theme centers on microbial tools and ecological perspectives. Zhang R. et al. provided an opinion on wholecell bioreporter (WCBs) technology as an emerging tool for assessing As risk in soils. Unlike traditional chemical assays, WCBs can differentiate As species and measure bioavailable fractions, offering ecologically relevant insights into toxicity. Complementing this, Tong et al. reviewed the persistence and pathogenicity of Fusarium oxysporum in watermelon soils. Although focused on plant pathology, their review illustrates how soil microbial communities and environmental factors interact to sustain longterm pathogen survival, mirroring the challenges of predicting microbial mediation in C-Fe-As cycles. Both contributions underscore the value of microbial systems, whether as tools or as agents, for monitoring and managing soil processes.
Technological innovation emerges as another unifying thread. Zhang X. et al. synthesized over 10,000 cases of microbial fuel cells' (MFCs) studies to identify the strongest drivers of performance. Their analysis showed that cathode chamber volume and surface area are key predictors of power density, while biological pretreatment of substrates significantly enhances efficiency. Importantly, MFCs are not only promising for energy generation but also for pollutant removal, including heavy metals and organics. Meanwhile, Tian et al. identified a new soilborne pathogen, Ilyonectria robusta, causing basal stem rot in Schisandra chinensis. Their study demonstrates how integrating molecular tools with field surveys can rapidly identify emerging risks to soilplant systems, further reinforcing the importance of innovation in soil biogeochemistry and health research.
Collectively, these contributions illustrate the interconnectedness of mineral, microbial, and technological dimensions in advancing soil science. Mineral studies demonstrate how natural and engineered Fe associations mediate contaminant dynamics; microbial perspectives highlight the dual roles of soil organisms as both sentinels and stressors; and technological innovations, from MFCs to WCBs, open new pathways for monitoring and remediation. Together, they provide a multifaceted view of the challenges and opportunities in managing C-Fe-As interactions in soils. While the studies in this Research Topic significantly broaden our understanding of soil biogeochemistry, they also point to critical areas for future research. There is a pressing need to integrate mineralogical, microbial, and engineering perspectives into predictive frameworks that can capture soil heterogeneity and dynamic redox processes. Future work should explore the underappreciated roles of viruses, archaea, fungi, and microplastics in shaping elemental cycles, as well as the feedbacks between soil processes and climate drivers. Scaling laboratory findings to field applications will remain a central challenge, requiring interdisciplinary approaches that link soil biogeochemistry to agronomy, hydrology, and environmental engineering.
## References
1. Gao, Li, Xie et al. (2024) "The fate of arsenic associated with the transformation of iron oxides in soils: the mineralogical evidence" *Sci. Total Environ*
2. Hu, Zhang, Yang et al. (2025) "Organic carbon sequestration by secondary Fe-Mn complex minerals via the anoxic redox reaction of Fe (II) and birnessite" *Environ. Sci. Technol*
3. Liang, Yang, Radosevich et al. (2025) "Bacteriophage-driven microbial phenotypic heterogeneity: ecological and biogeochemical importance" *NPJ Biofilms Microbiomes*
4. Tang, Xiang, Xiao et al. (2024) "Microbial mediated remediation of heavy metals toxicity: mechanisms and future prospects" *Front. Plant Sci*
5. Wang, Gao, Ma et al. (2024) "Iron mineral type controls organic matter stability and priming in paddy soil under anaerobic conditions" *Soil Biol. Biochem*
6. Wang, Zhu, Ge et al. (2024) "Unveiling the top-down control of soil viruses over microbial communities and soil organic carbon cycling: a review" *Clim. Smart Agric*
7. Xu, Tsang (2024) "Mineral-mediated stability of organic carbon in soil and relevant interaction mechanisms" *Eco-Environ. Health*
8. Yao, Wang, Peduruhewa et al. (2023) "The coupling between iron and carbon and iron reducing bacteria control carbon sequestration in paddy soils" *Catena* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12128748&blobtype=pdf | # Correction to "Low Neutralization of SARS-CoV-2 Omicron BA5248, XBB15 and JN1 by Homologous Booster and Breakthrough Infection"
## Abstract
In the article referenced above, there was an error in the title. The title was incorrectly labeled as "Low Neutralization of SARS-CoV-2 Omicron BA5248, XBB15 and JN1 by Homologous Booster and Breakthrough Infection." The correct title should read: "Low Neutralization of SARS-CoV-2 Omicron BA.5.2.48, XBB.1.5 and JN.1 by Homologous Booster and Breakthrough Infection."Additionally, in the "Materials and Methods" section, there was an error in the Virus Stocks part. The virus stock was incorrectly labeled as "SARS-CoV-2/E6/FJH/2022/ZJ104 (Omicron/BA.5.2)." The correct label should read: "SARS-CoV-2/E6/FJH/2022/ZJ104 (Omicron/BA.5.2.48)."We apologize for this error. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12744250&blobtype=pdf | # Engineering Nipah virus: Reverse genetics as a gateway to novel drug discovery
Muralidharan Menon, Arjun Gopinathan, Pillai Sreekanth
## Abstract
Nipah virus (NiV) is a highly pathogenic and re-emerging virus that requires containment in biosafety level 4 (BSL-4) laboratories. The limited accessibility of these high-security facilities poses major obstacles to investigating immunopathogenesis and developing effective antiviral treatments. Reverse genetics allows manipulation of viral genomes without the need to handle the wild-type virus and has become instrumental in understanding NiV pathogenesis and advancing therapeutic research. These tools have proven vital for other high-containment viruses, notably during the SARS-CoV-2 pandemic, and have been adapted effectively for NiV. Reverse geneticsderived systems were used to evaluate the drug candidates in the preclinical studies of NiV, with several candidates in the development pipeline. This narrative review summarizes established reverse genetics and pseudotyping methodologies for NiV, highlighting their contributions to understanding viral pathogenesis and accelerating vaccine and therapeutic development.
## 1. Introduction
Nipah virus (NiV) is a highly pathogenic virus that belongs to the paramyxovirus family and causes Nipah virus infection. Many other pathogenic viruses, including measles virus, mumps virus, respiratory syncytial virus, Newcastle disease virus, and canine distemper virus, belong to the same family; however, the pathogenicity of NiV is among the highest. The origin of NiV was previously reported to be between 1937 and 1947 in Southeast Asia [1,2]; however, the first reported outbreak of NiV occurred in Sungai Nipah village, Malaysia, in 1998 [3]. Later, outbreaks of NiV were reported in Bangladesh and India [4,5]. A more serious threat of NiV was reported in Kerala, the southernmost part of India [6]. Two strains of NiV have been identified to date: the Malaysian strain (NiV-M) and the Bangladesh strain (NiV-B). Recent outbreaks in India have been linked to the NiV-B strain [3,7]. While no direct connection between the two strains has been established, they exhibit similar pathogenicity and clinical manifestations.
NiV can be transmitted through the oral or nasopharyngeal route through contact with infected animals or through the consumption of tainted fruit [8]. As of May 2024, there have been 754 confirmed human NiV cases with 435 deaths, making the global case fatality rate (CFR) of NiV infection at about 58 % [8]. However, an outbreak in Kerala, India, between May and June 2018, reported a higher fatality rate of more than 90 % (19 deaths out of 21 reported cases) [6]. The higher pathogenicity linked with the observed CFR during the 2018 Kerala NiV outbreak highlights the urgent public health concern posed by the absence of approved prophylactics, therapeutics, or vaccines.
## 2. Reverse genetics approaches and their major types targeting NiV viral proteins
Nipah virus (NiV) is an enveloped virus characterized by a negativesense, single-stranded, non-segmented RNA genome approximately 18.2 kb in length. The viral RNA undergoes m6A methylation and 5' capping to maintain its integrity within the host cells and modulate immune responses [9]. The structure and function of the NiV viral proteins have been comprehensively reviewed [10]. Briefly, the NiV genome encodes a total of nine proteins, including the fusion protein (F), glycoprotein (G), matrix protein (M), nucleocapsid protein (N), large polymerase (L), and phosphoprotein (P). The P gene undergoes mRNA editing and is translated to produce alternative proteins, V, W, and C proteins [11].
Reverse genetics enables in vitro generation and manipulation of viral genomes, offering critical insights into replication, pathogenesis, and antiviral development. For NiV, high-containment requirements limit extensive research on immunopathogenesis to therapeutic strategies. To address this, various reverse genetics platforms have been introduced, which allow safer platforms. These systems, useable in BSL-2 settings, involve transfecting cells with plasmids encoding viral components to rescue viruses or mimic infection. They permit precise genetic modifications to study viral gene functions and have been applied to many other high-risk viruses [12].
## 2.1. Pseudotyped viruses
Pseudotyped or pseudoviruses have been extensively used to investigate the entry and attachment of NiV-G and NiV-F to host cells. Previously, DNA vaccine candidates [13], polyclonal antibodies [14], and monoclonal antibodies [15] were identified using a pseudovirus system targeting the NiV-V and NiV-F proteins, in both in vitro cultures and animal models. A subunit vaccine candidate, soluble G (sG) glycoproteins with the neoadjuvant (HTa), combined with CpG/Alum (SACF) and CpG/Aluminium ion adjuvant vaccine candidate (SAC), was developed and successfully tested using the pseudovirus targeting the NiV-G in several animal models [16]. Three monoclonal antibodies, including IB2, 3D7, and 7G9 [17], and a single domain antibody named n425 [18] were identified using the pseudovirus targeting the G protein in the Balb/C mice. Similarly, the neutralizing antibodies NiV41 and NiV42 were also identified using the pseudovirus targeting the NiV-G in Syrian golden hamsters [19]. In case of pseudoviruses targeting the NiV-F, a DNA vaccine and ChAd vector vaccine named AdC68-F and pVAX1-F, respectively, were evaluated in preclinical models [20]. In addition, the same system had screened out protein cleavage inhibitors as potential antivirals to NiV-F in the in vitro cultures [21]. However, this study needs much higher insights into animal models and clinical trials. Pseudoviruses have been extensively used to screen small molecular inhibitors and antibodies that inhibit viral attachment and entry, due to their simplicity, ease of development, and minimal biosafety risk. The major limitation of these studies is that they were limited to the NiV-F and NiV-G proteins responsible for interaction with the Ephrin B2/B3 protein [22]. The therapeutic strategies employed with the pseudovirus system targeting NiV-G and NiV-F are presented in Table 1.
## 2.2. Minigenome reverse genetics system
The minigenome reverse genetics system utilizes a synthetic, noninfectious minireplicon that mimics the Nipah virus (NiV) RNA genome to study viral transcription and replication. It includes a reporter gene flanked by key untranslated regions (UTRs) and effectively analyzes NiV RNA synthesis [23]. This system has been used to investigate viral replication, the impact of viral protein mutations, host factor roles, and to screen potential antivirals [24][25][26]. RdRp inhibitors like ribavirin, remdesivir, and favipiravir were tested in BHK-21 and A549 cells [24,27], while siRNAs against NiV-N and NiV-L were studied in Vero cells [28]. It would be interesting to see if this system can yield replicable results when applied in the in vivo models. Table 2 presents the drug candidates evaluated using the NiV minigenome system.
## 2.3. Viral replicon particles (VRPs)
Viral replicon particles (VRPs) are virus-like particles that encapsulate self-replicating, sub-genomic RNA (replicons) derived from the parental virus. These replicons lack the sequences required for structural protein synthesis, rendering them non-infectious, yet capable of intracellular replication and expression of heterologous genes [29]. In the context of Nipah virus (NiV), this platform has been applied to develop a "single-dose mucosal replicon-particle vaccine," which conferred protection in Syrian hamsters against the NiV-M strain as early as three days post-vaccination [30]. This system is more similar to a minigenome; the major difference is that it encodes the viral protein, except for structural proteins, such as M, F, and G, which are provided separately as helper plasmids. The best part of this system is that this can be adopted to study intracellular replication, transcription, assembly, budding, and a single round of infection. Interestingly, this system is ideal for high-throughput screening (HTS) of antivirals targeting virus replication and assembly of viral particles.
## 2.4. Transcription-and replication-competent virus-like particles (trVLPs)
The trVLPs are engineered to simulate the complete viral life cycle by enabling viral RNA transcription and replication without producing infectious progeny. This system permits multicycle replication under non-infectious conditions and is widely used to study virus-host interactions and screen antiviral agents [23,31]. A NiV trVLP system lacking the N, P, and L genes has been developed, demonstrating safety in hamster models. Using this system, tunicamycin was identified as a potential antiviral compound through library screening [31]. Additionally, trVLPs have been used to investigate host restriction factors, such as Tetherin, in NiV replication [32]. This system is similar to VRPs but mimics virtually every stage of a viral life cycle that occurs inside a host cell, making it the most versatile system for studying high-containment viruses. One notable aspect of the viral life cycle is that trVLPs do not mimic the initial primary transcription of a naive target cell, as this system targets cells that are engineered to express the proteins necessary for viral replication, such as NiV-P and NiV-N.
## 2.5. Single-cycle viruses (SCVs)
Single-cycle viruses (SCVs) are similar to trVLPs, engineered viral systems capable of infecting host cells and replicating their genomes, but they lack the ability to produce infectious progeny, thereby restricting
## Table 1
Therapeutic strategies evaluated using pseudovirus systems targeting NiV-F and NiV-G. BALB/c mice F and G protein [13] Rabbit polyclonal antibodies 293T cells F and G protein [14] Monoclonal Antibodies (m102.4 and 5B3) CB6F1/J mice and Vero E6
F and G protein [15] Subunit vaccine (SACF, SAC with sG) C57BL/6 mice, BALB/ c mice and Landrace × York pigs G protein [16] Monoclonal antibodies (IB2, 3D7 and 7G9)
BALB/c mice G protein [17] Single-domain antibody (n425) BALB/c mice and 293T G protein [18] Neutrilizing antibodies (NiV41 and NiV42)
## Syrian golden hamsters and 293T
G protein [19] DNA vaccine and ChAd vector vaccine (AdC68-F and pVAX1-F) BALB/c mice and HEK-293
F protein [20] Protein cleavage inhibitor (5219666, 7923236, 7931205, 5705213, etc.)
## 293T cells F protein [21]
Table 2 Therapeutic strategies evaluated using the minigenome of NiV. RdRp Inhibitors (remdesivir, azvudine, and molnupiravir) A549 cells N, P, and L [27] siRNA (siL3619, siL4877, siL6200, siL2145, siL3619, siL4877, siL6200 and siL2145)
Vero cells N and L [28] the infection to a single cycle. This feature makes SCVs particularly useful for investigating early stages of the viral life cycle and for applications in vaccine development. A single study utilizing SCVs expressing either the NiV-G or NiV-F protein demonstrated successful infection and challenge in Syrian hamsters [33]; however, this system
has not yet been well-established. This may be due to its inability to proceed further in the viral replication cycle.
## 2.6. Infectious clones
Infectious clones are full-length DNA copies of viral genomes that can be stably maintained in bacterial plasmids or artificial chromosomes. When transfected into permissive host cells, these clones enable the generation of infectious viral particles, providing a powerful platform for precise genetic manipulation of viral genomes. Although a major limitation of this approach is the requirement for high-level biocontainment similar to that needed for handling live viruses, which is BSL-4 in the case of NiV. However, it offers significant advantages for the synthetic reconstruction and propagation of viruses in high volumes. In the case of Nipah virus (NiV), an infectious clone system was previously employed to rescue replication-competent recombinant NiV (rNiV) by co-transfecting a full-length cDNA construct with plasmids expressing the N, P, and L, in the presence of a recombinant vaccinia virus backbone. The rescued rNiV demonstrated replication kinetics and phenotypic characteristics comparable to the wild-type virus in vitro and maintained its pathogenicity in animal challenge studies [34]; however, the application of this system for antiviral drug discovery remains underexplored.
## 2.7. Self-amplifying RNA (saRNA)
Self-amplifying RNA (saRNA) molecules are synthetically engineered mRNAs that carry the genetic elements required for their own intracellular replication, enabling sustained amplification of the encoded protein within host cells. Compared to conventional mRNA platforms, saRNAs offer enhanced and prolonged protein expression, making them highly promising for applications in vaccine development. In the context of NiV, a research group has employed immune-informatics approaches to design an in silico saRNA-based vaccine candidate [35]; however, its efficacy and safety remain to be validated through preclinical studies. Coalition for Epidemic Preparedness Innovations (CEPI) has invested 13.38 million USD for the development of saRNA vaccine against NiV, driving translational research [36]. Although this tool can be used to study NiV proteins, more established expression vectors have already been developed to produce and study NiV proteins, making this system less desirable for studying the viral proteins.
## 3. Conclusion
NiV has attracted global concern due to its high pathogenicity and potential for human-to-human transmission, posing a significant pandemic threat. Reverse genetics systems offer valuable tools for studying NiV safely in lower biosafety level laboratories; however, some require the same higher containment facilities. A schematic representation of the established reverse genetics systems targeting the NiV viral proteins, with their applications and containment facility requirements, is represented in Fig. 1. In addition, a cross-comparison of the established NiV reverse genetics systems, along with their advantages, limitations, and applications, is represented in Table 3.
Most reverse genetics systems for NiV have been established for evaluating therapeutic strategies. Although, the pseudovirus and minigenome systems were mostly used to study NiV, new research is directed towards the development and utilization of other systems, such as trVLPs, SCV, infectious clones, and saRNA. Pseudoviruses and minigenomes are preferred the most due to their biosafety and wellestablished protocols. The trVLPs are next, and the least utilized systems are SCVs, saRNA, and infectious clones. The limited use of infectious clones and SCVs may be due to the use of the same containment facilities for the wild-type virus and poses a much higher risk than any other system. The saRNA system is ideal for vaccine development and the production of viral proteins; however, it is least utilized for NiV. Interestingly, trVLPs have several advantages over other systems, as they mimic almost all aspects of the viral life cycle, except for spread, and they cannot replicate outside the producer cell lines, making them safe to use in a BSL-2 laboratory. Previously, a more advanced reverse genetics system, the circular polymerase extension reaction (CPER) technique, was used to study SARS-CoV-2 [37], which utilizes a DNA/RNA template to generate a cDNA clone that can be modified according to specific requirements. However, this system has not been established for NiV, which may offer more promising techniques, but requires higher containment facilities similar to those of the wild-type NiV.
## CRediT authorship contribution statement
## Muralidharan
## References
1. Presti, Cella, Giovanetti et al. (2016) "Origin and evolution of Nipah virus" *J Med Virol*
2. Cortes-Azuero, Lefrancq, Nikolay et al. (2024) "The genetic diversity of Nipah virus across spatial scales" *J Infect Dis*
3. Garbuglia, Lapa, Pauciullo et al. (2023) "Nipah virus: an overview of the Current status of diagnostics and their role in preparedness in endemic countries" *Viruses-Basel*
4. Bhowmik, Hasan, Redoy et al. (2025) "Nipah virus outbreak trends in Bangladesh during the period 2001 to 2024: a brief review" *Sci One Health*
5. Arankalle, Bandyopadhyay, Ramdasi et al. (2011) "Genomic characterization of Nipah virus" *Emerg Infect Dis*
6. Arunkumar, Chandni, Mourya et al. (2018) "Outbreak investigation of Nipah virus disease in Kerala, India" *J Infect Dis*
7. Sanker, Vellekkat (2024) "Nipah virus outbreaks in Kerala: an impending doom?" *Health Sci Rep-Us*
8. Khan, Akbar, Mahtab et al. (2024) "Twentyfive years of Nipah outbreaks in Southeast Asia: a persistent threat to global health" *IJID Reg*
9. Hu, Kim, Yang et al. (2025) "Structural and functional analysis of the Nipah virus polymerase complex" *Cell*
10. Yang, Wang, Liu (2024) "Structure of the Nipah virus polymerase phosphoprotein complex" *Nat Commun*
11. Kulkarni, Volchkova, Basler et al. (2009) "Nipah virus edits its P gene at high frequency to express the V and W proteins" *J Virol*
12. Chen, Liu, Peng (2022) "Reverse genetics in virology: a double edged sword" *Biosaf Health*
13. Nie, Liu, Wang et al. (2019) "Nipah pseudovirus system enables evaluation of vaccines and using non-BSL-4 facilities" *Emerg Microb Infect*
14. Luo, Wang, Huang et al. (2023) "Establishment of a neutralization assay for Nipah virus using a high-titer pseudovirus system" *Biotechnol Lett*
15. Loomis, Gbe, Tsybovsky et al. (2020) "Structure-based design of Nipah virus vaccines: a generalizable approach to Paramyxovirus immunogen development" *Front Immunol*
16. Gao, Li, Han et al. (2022) "Assessment of the immunogenicity and protection of a Nipah virus soluble G vaccine candidate in mice and pigs" *Front Microbiol*
17. Bae, Kim, Moon et al. (2019) "Construction of the safe neutralizing assay system using pseudotyped Nipah virus and G protein-specific monoclonal antibody" *Biochem Bioph Res Co*
18. Wang, Sun, Shen et al. (2024) "Fully human singledomain antibody targeting a highly conserved cryptic epitope on the Nipah virus G protein" *Nat Commun*
19. Chen, Sun, Zhang et al. (2024) "Potent human neutralizing antibodies against Nipah virus derived from two ancestral antibody heavy chains" *Nat Commun*
20. Lu, Yao, Liu et al. (2023) "Vaccines based on the fusion protein consensus sequence protect Syrian hamsters from Nipah virus infection" *JCI Insight*
21. Elshabrawy, Fan, Haddad et al. (2014) "Identification of a broad-spectrum antiviral small molecule against severe acute respiratory syndrome Coronavirus and Ebola, Hendra, and nipah viruses by using a novel high-throughput screening assay" *J Virol*
22. Yuan, Marsh, Khetawat et al. (2011) "Mutations in the G-H loop region of ephrin-B2 can enhance Nipah virus binding and infection" *J Gen Virol*
23. Hoenen, Groseth, De Kok-Mercado et al. (2011) "transcription and replication competent virus-like particles and beyond: reverse genetics systems for filoviruses and other negative stranded hemorrhagic fever viruses" *Antivir Res*
24. Ke, Ye, Liu et al. (2024) "Establishment of a novel minigenome system for the identification of drugs targeting Nipah virus replication" *J Gen Virol*
25. Sleeman, Bankamp, Hummel et al. (2008) "The C, V and W proteins of Nipah virus inhibit minigenome replication" *J Gen Virol*
26. Halpin, Bankamp, Harcourt et al. (2004) "Nipah virus conforms to the rule of six in a minigenome replication assay" *J Gen Virol*
27. Wang, Chen, Zhong et al. (2025) "Construction of Minigenome Replicon of Nipah virus and investigation of biological activity" *Viruses-Basel*
28. Mungall, Schopman, Lambeth et al. (2008) "Inhibition of Henipavirus infection by RNA interference" *Antivir Res*
29. Van Der Meulen, Smets, Rüdelsheim (2023) "Viral replicon systems and their biosafety aspects" *Appl Biosaf*
30. Welch, Spengler, Genzer et al. (2023) "Single-dose mucosal replicon-particle vaccine protects against lethal Nipah virus infection up to 3 days after vaccination" *Sci Adv*
31. Wang, Fan, Ye et al. (2024) "Novel transcription and replication-competent virus-like particles system modelling the Nipah virus life cycle" *Emerg Microb Infect*
32. Hoffmann, Nehlmeier, Brinkmann et al. (2019) "Tetherin inhibits nipah virus but not Ebola virus replication in fruit bat cells" *J Virol*
33. Lo, Bird, Chattopadhyay et al. (2014) "Singledose replication-defective VSV-based Nipah virus vaccines provide protection from lethal challenge in Syrian hamsters" *Antivir Res*
34. Yoneda, Guillaume, Ikeda et al. (2006) "Establishment of a Nipah virus rescue system" *Proc Natl Acad Sci*
35. Rangacharya, Parab, Adkine et al. (2023) "A study on the design of an in silico self-amplifying mRNA vaccine against Nipah virus using immunoinformatics" *J Biomol Struct Dyn*
36. Cepi "AI-enhanced self-amplifying mRNA vaccine set to combat one of the deadliest known viruses Oslo. Norway: Coalition Epidemic Preparedness Innovations (CEPI)"
37. Amarilla, Sng, Parry et al. (2021) "A versatile reverse genetics platform for SARS-CoV-2 and other positive-strand RNA viruses" *Nat Commun* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548397&blobtype=pdf | # Discontinuous template switching generates coronavirus subgenomic RNAs from the 3ʹ viral genome end by 5ʹ to 3ʹ transcription
Ayslan Castro Brant, Zhe Hu, Angelika Chen, Vladimir Majerciak, Jonathan Yewdell, Zhi-Ming Zheng
## Abstract
In coronavirus (CoV)-infected cells, several structural and accessory proteins are synthesized from subgenome RNAs (sgRNA) containing a common genomic 5ʹ-leader followed by a given open reading frame (ORF). We report that the abundance of these sgRNAs varies with distance from the 3′-end of the genome. Thus, there are more sgRNAs encoding nucleocapsid (N) than spike (S), presumably the results from discontinuous 5ʹ-3ʹ transcription template switch mediated by the viral replication and transcription complex (RTC). We optimized the circular polymerase extension reaction (CPER) methodology to generate infectious double-stranded circular cDNA (ds-circDNA) containing the mNeonGreen (NG) reporter in accessory ORFs of human CoVs OC43 and SARS-CoV-2. In each CoV, we found that levels of sgRNAs and NG expression increased with 3ʹ proximal genomic NG location. By reinfection of HCT-8 cells with the same MOI 0.01, however, we found that the slow-growing OC43 NG-ns2 virions exhibited equal infectivity and productivity as the fast-growing OC43 NG-ns12.9 virions. Introduction of point-mutations into the mapped TRS B motif for synthesis of the OC43 ns12.9 and M sgRNAs led to disrupt TRS B -TRS L cross-interactions and block production of the corresponding sgRNAs and infectious virions. Together, our data indicate that, by using an optimized CPER approach for positional NG insertions, we demonstrated the progressional reduction of the RTC-mediated template switch (read-through individual TRS B ) efficiency in production of the corresponding sgRNAs in an order from the 3ʹ viral genome end.
IMPORTANCEThe mechanism for sgRNA synthesis in the coronavirus life cycle is poorly understood. The current model suggests discontinuous template-switch transcription mediated by viral replication and transcription complex (RTC) for synthesis of individual sgRNAs to translate corresponding structural and accessory proteins but lacks experi mental data support. This report provides the first experimental evidence that, in both hCoV-OC43 and SARS-CoV-2, viral RTC synthesizes its sgRNAs by long-range base-pair ing between a distal transcription regulatory body sequence (TRS B ) upstream of each structural/accessory ORF and the transcription regulatory leader sequence (TRS L ) from the viral genome 5ʹ-UTR, leading to the production of viral sgRNAs in abundance order from the viral genome 3ʹ -end, with more N sgRNAs but less S sgRNAs. Our data support a "first-come, first-serving" model in TRS B -TRS L cross-interaction and read-through TRS B process to mediate discontinuous transcription switch in coronavirus sgRNA synthesis in a 5ʹ-3ʹ transcription direction from the 3ʹ viral genome during coronavirus infection.
C oronaviruses infect many mammalian species, including humans. Four of the seven known human coronaviruses (hCoV) (hCoV-229E, -HKU1, -NL63, and -OC43) generally cause mild upper respiratory infections, while three (SARS-CoV, MERS-CoV, and SARS-CoV-2) can cause severe acute respiratory syndrome (SARS) (1)(2)(3). Coronaviruses are divided into Alpha-, Beta-, Gamma-, and Delta-genera. hCoV-229E and hCoV-NL63 belong to the Alpha-coronavirus, and hCoV-OC43, hCoV-HKU1, SARS-CoV, MERS-CoV, and SARS-CoV-2 belong to the Beta-coronavirus. Despite hCoV-NL63 and SARS-CoV-2 belonging to different genera, they utilize the same cellular receptor, the angiotensin 1-converting enzyme 2 (ACE2) to enter cells (2).
The coronavirus genomic RNA (gRNA) is a single-stranded positive-sense 26-32 kb RNA. The gRNA, which encodes 20 to 29 known viral proteins (4)(5)(6), has a 5ʹ-end cap followed by a 5ʹ untranslated region (5ʹ-UTR), a long coding region, and a 30-60 nt 3ʹ-end poly-A tail. Approximately 70% of the viral genome encodes ORF1a and ORF1b, which are translated into two polyproteins post-translationally cleaved into the 16 viral nonstructural proteins (nsps) (2). The remaining 30% of the genome encodes structural (S, E, M, and N) and accessory proteins. The number of accessory proteins varies among coronaviruses, and their functions are poorly defined but presumably contribute to viral pathogenesis (7)(8)(9) as most are nonessential for viral replication in cultured cells (7,(10)(11)(12). The viral genome 5ʹ-UTR contains a 72-nt leader, a transcription regulatory sequence motif (TRS L, ACGAAC), and other translation and genome package regulatory cis-ele ments. A transcription regulatory body sequence (TRS B ) upstream of each structural or accessory ORF region is presumed to control subgenome RNA (sgRNA) synthesis (13,14), which is robust. The sgRNAs are excluded from virions via unknown mechanisms.
During viral penetration, the +gRNA is released into the cytosol where ORF1a and ORF1b are translated into a polyprotein cleaved into 16 nsps. The nsp7, nsp8 (×2), nsp9, nsp12, and nsp13 (×2) associate with a viral +gRNA template and an RNA primer to form the Replication and Transcription Complex (RTC) (15,16). Inside the virus-induced double-membrane vesicles (DMVs), the RTC synthesizes negative-strand genomic RNA (-gRNA) and a subset of negative-strand sgRNAs (-sgRNAs) from a viral +gRNA template (17,18). The -gRNA and -sgRNAs are then used as a template separately to synthesize full-length +gRNA for virion generation and +sgRNAs for translation of a given structural or accessory protein (2).
The mechanism for -sgRNA synthesis is poorly understood. A current model suggests discontinuous transcription mediated by RTC template switching by long-range base-pairing between distal TRS B of the 6-7 nt core sequence and the TRS L motif from the viral genome 5ʹ-UTR (2,5,14,(19)(20)(21). As the RTC synthesizes the -sgRNAs from the +gRNA 3ʹ-end in the 5ʹ-3ʹ direction, it may temporarily dissociate the tem plate +gRNA at TRS B to enable the RTC to translocate to the TRS L leader, skipping a large fraction of the viral genome in this template switch. By going through individual TRS B for the template switch, this generates variably sized -sgRNAs, of which further serve as individual templates synthesizing the corresponding +sgRNAs whose first ORF is translatable (2,21).
The aim of this study is to better understand the proposed discontinuous transcrip tion template switch model (2) by comparing the sgRNA synthesis of two human coronaviruses hCoV-OC43 and SARS-CoV-2. We employed an autofluorescent mNeon Green (NG) reporter protein (22) to generate a set of recombinant infectious viruses using an optimized circular polymerase extension reaction (CPER) (23)(24)(25). By positional NG insertion to an accessory ORF (7,(10)(11)(12), we provide compelling evidence supporting a "first-come, first-serving" model in TRS B -TRS L cross-interaction-mediated discontinuous transcription for coronavirus sgRNA synthesis (2).
## RESULTS
## Optimizing CPER to construct SARS-CoV-2 and hCoV-OC43 double-stranded circular cDNAs (ds-circDNA)
We optimized the CPER-based system originally devised to generate flavivirus (23) and subsequently, SARS-CoV-2 (24,25) infectious clones. We obtained the linker plasmid from Alberto A. Amarilla and Alexander A. Khromykh (25), which is a central component for efficient CPER. By PCR introduction of a 20 bp nucleotide sequence on the linker 5ʹ-end and a 37 bp sequence on the linker 3ʹ-end, respectively, to overlap the corresponding viral genome 3ʹ-UTR end and 5ʹ-UTR end (Fig. S1A), we made the linker specific for construction of individual infectious ds-circDNAs of hCoV-OC43 and SARS-CoV-2.
Seven overlapping DNA fragments (F1 to F7) were amplified by PCR from the full-length (FL) SARS-CoV-2 (26) and hCoV-OC43 cDNAs (Fig. S1B, steps 1 and 2, Fig. S2 and Table S1), with 20-40 nt overlapping from each fragment. We took viral acces sory ORF in a given cDNA fragment for insertion of an autofluorescent NG reporter by overlapping PCR (Fig. 1A; Fig. S2). By annealing overlapped cDNA fragments with the virus-specific linker (Fig. S1B, step 3), we generated semi-circular cDNA through base-pairing from the overlapped sequences, with each fragment acting as a Taq DNA polymerase template/primer to fill inter-fragment gaps (Fig. S1B, step 4). The final product is a virus-specific ds-circDNA (Fig. S1B,step 5).
We determined the optimal fragment concentration for individual CPER reactions by using 0.01, 0.05, and 0.1 pM of individual fragments to generate FL ds-circDNAs with a NG insertion, as determined by agarose gel electrophoresis (Fig. 1B). This revealed that the CPER reaction with 0.05 pM of each cDNA fragment displayed the best production of >25 kb ds-circDNA band (Fig. 1B, lanes 2 and 5, black arrows), corresponding to a FL cDNA of the SARS-CoV-2 or hCoV-OC43 genome.
## Optimizing CPER-derived hCoV-OC43 NG-virus production
We first examined CPER-derived FL ds-circDNAs of hCoV-OC43 NG-ns12.9 by co-trans fecting HEK293T cells with either a hCoV-OC43 N protein expression plasmid (pCOC42) or empty control plasmid pFLAG-CMV-5.1, and incubating the transfected cells for 24 h (Fig. S1C) before adding HCT-8 cells. We observed a constant increase of NG + cells when we co-transfected cells with the N plasmid pCOC42, but not with the pFLAG-CMV-5.1 (Fig. 2A andB). Thus, CPER-derived ds-circDNA co-transfection with the N protein expression is essential for efficient virus production. We next examined the serum requirement for infectious virus generation. We compared fetal bovine serum (2% FBS) to newborn calf serum (2% NCS, a cheaper alternative to FBS) for transfected cell cultures in production of infectious viruses at 33°C from the co-transfected HEK293T and co-cultivated with HCT-8 cells. We found there was a progressive increase in the number of NG + cells in 2% FBS cultivated cells, starting from ~70 NG + cells/field on day 8 (D8) to ~282 NG + cells/field on day 10 (D10), but not the cells under a 2% NCS culture condition (Fig. 2C andD). The data indicate that FBS was essential for successful propagation of the recombinant infectious virus that originates from the ds-circDNAs of hCoV-OC43 NG-ns12.9 constructed by the CPER technology.
Despite the efficient replication, hCoV-OC43 infection of HCT-8 cells does not lead to visible plaques, a traditional method to determine the virus titer. To identify a cell line suitable for hCoV-OC43 plaque assay and titration of our CPER-derived infectious clones, we confirmed both LLC-MK2 (monkey kidney cells) and Mv1Lu (Aleutian mink lung cells) are susceptible to wild-type (WT) hCoV-OC43 infection (Fig. 2E) (27)(28)(29)(30). By comparing their susceptibility for plaque formation using WT hCoV-OC43 produced by HCT-8 cells, we found that Mv1Lu cells are more permissive for hCoV-OC43 plaque formation than LLC-MK2 cells, with a 100-fold higher titer by this assay (Fig. 2F). We therefore used Mv1Lu cells to titer CPER-derived infectious NG-hCoV-OC43.
## CPER-derived hCoV-OC43 NG-ns12.9 ds-circDNA generates more NG + viruses than CPER-derived hCoV-OC43 NG-Δns2 ds-circDNA
After establishment of the optimal conditions for CPER-based system, we next examined if a positional insertion of the NG reporter into hCoV-OC43 would affect virus replication efficiency. hCoV-OC43 contains two accessory ORFs, an ns2 (31) toward the 5ʹ half of the virus genome and an ns12.9 (32) toward the 3ʹ half of the viral genome (Fig. 1A). As expected, deletion or NG insertion of the ns2 was not detrimental to hCoV-OC43 virus infection and replication (33) (Fig. S2B andS3). Parallel co-transfection of CPER generated FL ds-circDNA products (Fig. 3A, black arrow) of hCoV-OC43 NG-Δns2 or NG-ns12.9 with the N plasmid pCOC42 into HEK293T cells for 24 h was followed by addition of HCT-8 cells for co-cultivation for the indicated days. To our surprise, we observed that the cells with transfected NG-ns12.9 ds-circDNA showed a consistently higher number of NG + cells than NG-Δns2 ds-circDNA transfection (Fig. 3B andC). The NG-ns12.9-transfected cells produced almost five times more NG + cells than NG-Δns2 both at D5 and D6 (Fig. 3C). These data were highly reproducible, confirming a higher transcription and replica tion capability of the NG-ns12.9 infectious clone with an NG insertion toward the 3ʹ-half of the virus genome than the NG-Δns2 infectious clone with an NG insertion toward the 5ʹ-half of the virus genome.
To validate this observation, we isolated total cell RNA from the co-cultured cells on D6 for detection of viral RNA transcripts by Northern blot and collected the culture supernatants for virus titration by the plaque assay described in Fig. 2F. Using a 32 Plabeled anti-sense oligo probe from the hCoV-OC43 N ORF region capable of detecting viral +gRNA and all +sgRNAs, we examined the equal amount of the extracted total cell RNA from each transfection and demonstrated, as expected, the higher amount of viral RNA transcripts from the NG-ns12.9-transfected cells, comparable to WT hCoV-OC43infected cells, than that from the NG-Δns2-transfected cells (Fig. 3D, compare lanes 2 and 3 to 1). The data further indicated the preferable higher viral genome transcription and replication of the NG-ns12.9 than the NG-Δns2. We also observed the expected RNA size but not new calf serum (NCS), is required for efficient virus production in HCT-8 cells. HEK293T cells were co-transfected with an FL hCoV-OC43 NG-ns12.9 ds-circDNA along with an N expression vector pCOC42 and maintained in DMEM supplemented with 2% FBS overnight. The transfected cells were then co-cultivated by the addition of HCT-8 cells with cell passage every 3 days for a total of 10 days in DMEM supplemented with 10% FBS or 10% NCS. The number of NG + HCT-8 cells was counted and averaged from 10 random microscopic fields on D8-D10 (C). Significantly more NG + HCT-8 cells were shown from the cells with an FL NG-ns12.9 ds-circDNA on D8-D10 when growing in the DMEM containing 10% FBS when compared with 10% NCS (D). (E) hCoV-OC43 induced visible and well-defined cytopathic effect (CPE) in both LLC-MK2 and Mv1Lu cells. The monolayer of LLC-MK2 or Mv1Lu cells in ~70% confluence was infected with WT hCoV-OC43 (100 µL supernatant of infected HCT-8 cells). One representative microscopic field is shown for each cell type. (F) Mv1Lu cells are more sensitive than LLC-MK2 cells for hCoV-OC43 infection and plaque formation. Plaque assays of LLC-MK2 and Mv1Lu cells were infected with 100 µL of each diluent after serial 10-fold dilutions of WT hCoV-OC43 virus and overlayed with semisolid media (1× DMEM, 0.5% methylcellulose and 10% FBS) for 8 days. The plaques were fixed for 30 min by 3.7% formaldehyde solution and stained with 1% crystal violet. increase from the inserted NG reporter by 762 nt for the individual sgRNAs S, HE, ns2, and the viral gRNA of NG-ns12.9 (green labels) (Fig. 3D, compare lines 2 to 3, and Table S2). In the NG-Δns2, the size shift of NG-Δns2 sgRNAs and viral gRNA (green labels) was only minimal with only an additional 6 nt difference from the corresponding WT virus sgRNAs and gRNA, as NG insertion in the NG-Δns2 genome compensated the partial deletion of the ns2 (Fig. 3D, compare lanes 1 and 3, and Table S2).
The increased virus replication resulting in higher virus titer of the CPER-derived NG-ns12.9 than that from the NG-Δns2 was further verified by plaque assays using Mv1Lu cells (Fig. 3F). Mv1Lu cells were infected with cell-free virions released to the culture supernatants on D6 and overlaid with semisolid DMEM-methylcellulose overlay media for up to 8 days. By calculation of fluorescent-forming units (FFU) from the infected Mv1Lu cells under fluorescent microscopy, we observed higher FFU appearance in the NG-ns12.9-infected cells than that in the NG-Δns2-infected cells, with FFU on D8 reaching 1.2 × 10 6 FFU/mL for the NG-ns12.9, while only 4.0 × 10 4 FFU/mL for the NG-Δns2 (see Fig. 3E, the left for FFU images and the right for bar graph). Crystal violet staining of a fixed monolayer in the titration plate confirmed a higher replication titer of the NG-ns12.9 than the NG-Δns2 (Fig. 3F) in the plaque assays.
Total cell RNA-seq and viral gRNA sequencing indicated that both OC43 NG viru ses, NG-ns12.9 and NG-Δns2, exhibited the same three mutations from the ATCC WT hCoV-OC43 (GenBank accession no. AY391777), of which are reverted to the sequence observed in another reference genome (GenBank accession no. NC_006213.1). These mutations are at viral genome positions nt 32U-to-C in the virus leader and nt 26997Gto-C and nt 27018C-to-U in the S ORF (Table S4). The nt 26997G-to-C causes a change of methionine to isoleucine. The nt 27018C-to-U is silent mutation. Data indicate that the observed difference in viral transcription and replication between the CPER-derived NG-ns12.9 and the CPER-derived NG-Δns2 was not because of different unintentional mutations created by positional NG insertion.
## CPER-derived SARS-CoV-2 NG-ΔORF7a ds-circDNA produces more NG + viruses than CPER-derived SARS-CoV-2 ORF3-NG ds-circDNA
The optimized CPER-based system was also applied to generate the FL SARS-CoV-2 ORF3-NG and NG-ΔORF7a (Fig. S1B and S2A; Fig. 4A, black arrow for lines 1 and 2) for co-transfection of HEK293T cells, respectively, along with a SARS-CoV-2 N protein expression vector pCSR24 (Fig. S1C) for 24 h. Subsequently, SARS-CoV-2 permissive BHK21-hACE2 cells were directly added to the monolayer of the transfected HEK293T cells for co-cultivation at the indicated time point (Fig. 4B andC). The quantification of NG + cells under a fluorescent microscopy imaging for ten microscope fields at 12 h post-co-cultivation exhibited a ~10 fold higher number of NG + cells from the CPERderived NG-ΔORF7a (~10.8 NG + cells/field) than the CPER-derived ORF3-NG-transfected cells (~ 0.9 NG + cells/field) (Fig. 4B andC). At 24 h post co-cultivation, the difference between the NG-ΔORF7a and the ORF3-NG increased to ~20 fold, with ~76.4 NG + cells/ field from the NG-ΔORF7a infected cells over an average of ~3.6 NG + cells/field from that of the ORF3-NG (Fig. 4B). This observation was confirmed by TCID50 titration of the collected culture supernatants for reinfection of fresh BHK21-hACE2 cells, showing higher production of infectious virus from the CPER-derived NG-ΔORF7a ds-circDNA than the ORF3-NG ds-circDNA by 12 h (P < 0.05) to further higher production by 24 h (P < 0.001) post co-cultivation (Fig. 4D).
Subsequently, we applied Northern blot using total cell RNA extracted 24 h after cocultivation to confirm the observed difference in viral genome transcription and replication efficiency of the two CPER-derived constructs. As shown in Fig. 4E, by using an antisense 32 P-labeled probe derived from the SARS-CoV-2 N ORF region capable of detecting viral +gRNA and all +sgRNAs, we showed a substantially higher level of viral RNAs in the cells transfected with the NG-ΔORF7a ds-circDNA than that with the ORF3-NG ds-circDNA (compare lanes 3 to 2). As expected, NG insertion and the corresponding deletion in the viral ORF7a resulted in additional 351 nt in the detected sgRNAs ORF7a, ORF6, E, M, ORF3, S, and viral gRNA (compare lanes 3 to 1, green labels at Fig. 4E; Table S2). Consistently, the NG insertion in ORF3 resulted in an increase of 708 nt in the ORF3 and S sgRNAs and the viral gRNA in the ORF3-NG ds-circDNA-transfected cells (Fig. 4E, compare lanes 2 to 1, and Table S2). Altogether with the results from hCoV-OC43, these data suggest that insertion of a NG reporter into an accessory ORF toward the viral genome 3ʹ half leads to more NG + virus production than it does so by the insertion towards the viral genome 5ʹ half, most likely reflecting a graduate reduction of the readthrough efficiency of individual TRS B -TRS L cross-interactions in the discontinuous 5ʹ-3ʹ transcription template switch in the RTC. Total cell RNA-seq and mapping indicated that both CPER-derived NG-ΔORF7a and ORF3-NG viruses contain an identical genome sequence to the original SARS-CoV-2 cDNA plasmid bearing three silent mutations at nt 26261 (C-to-U), nt 26542 (C-to-U), and nt 28853 (U-to-A) (26) (Table S4), indicating that the observed difference in viral RNA transcription and virus replication from the NG-ΔORF7a to ORF3-NG was not because of different unintentional mutations created by positional NG insertion.
## Infectious virions of hCoV-OC43 NG-Δns2 and NG-ns12.9 exhibit equal infectivity and replication capacity
As most of the coronavirus accessory proteins play no role in virus replication (8,34,35), our results from both hCoV-OC43 and SARS-CoV-2 intrigued us to further explore whether our positional NG insertions into the different viral genome regions would affect the infectivity and replication capacity of the individual CPER-derived infectious hCoV-OC43 virions. To test the replication fitness of the two NG + infectious virions, we infected HCT-8 cells separately with the hCoV-OC43 NG-Δns2 and NG-ns12.9 virions titrated in Fig. 3F and justified at the same MOI 0.01 for the infection. We then monitored their replication using live fluorescent microscopy by counting NG + cells in five randomly selected fluorescent fields for 3 days. We found that the infection with both hCoV-OC43 NG-Δns2 and NG-ns12.9 infectious virions replicated equally well from days 1 (D1) to 3 (D3) (Fig. 5A andB). No statistically significant difference in the number of NG + cells from the NG-Δns2-to the NG-ns12.9-infected HCT-8 cells was observed at D1 (~90 NG + cells/field) or D2 (~300 NG + cells/field) (Fig. 5A). The Mv1Lu cell FFU assay of their cell-free virions collected on D3 cell culture supernatants further confirmed the similar kinetics of the infectivity and replication capacity of both viruses, with the NG-Δns2 viral titer reaching to 1.5 × 10 4 FFU/mL and the NG-ns12.9 to 2.0 × 10 4 FFU/mL (Fig. 5C andD).
Together with our total RNA-seq/viral gRNA sequencing and mapping (Table S4), the above data indicate that the insertion of an NG reporter into an accessory ORF of the virus genome either towards the 5ʹ half (NG-Δns2) or towards the 3ʹ half of the genome, once virions produced, does not induce unintentional mutation, nor the efficiency of infectivity or replication capacity of infectious virions. Whether the individual viruses exhibit a different infection index by correlation of viral gRNA copy numbers along with virus passage in HCT-8 cells to Mv1Lu infectivity remains to be carefully studied. However, the CPER strategy for the positional NG insertion does provide a simple tool to mirror the transcription efficiency of individual -sgRNAs from the viral genome in a 5ʹ to 3ʹ order of gradually reduced reading through the discontinuous template switchingmediated TRS B -TRS L cross-interactions in the RTC (2) (Fig. 3D and4E).
## Mapping of the hCoV-OC43 TRS L and TRS B in synthesis of individual sgRNAs
The sequence of each TRS B in regulation of hCoV-OC43 sgRNA synthesis is partially known (36) but had not been verified by other laboratories. To verify the reported TRS L on the viral genome 5ʹ-end in interaction with each TRS B in regulation of synthesis of individual sgRNAs (36) through a proposed template switch model (2), we performed RT-PCR on total cell RNA extracted from the WT hCoV-OC43-infected HCT-8 cells using a set of specific primers (Table S3) for detection of each sgRNA (Fig. 6A). By sequencing the amplified RT-PCR products, we confirmed that the reported TRS L motif in the virus genome 5ʹ-UTR is composed of seven nucleotides, UCUAAAC (genomic position, nt 63 to 69) (Fig. 6B) (36). The same TRS L sequence motif was found in the TRS B of ns2 (UCUAAAC, nt 21492-21498) and S (UCUAAAC, nt 23636-23642), which might mediate the cross-interactions with the TRS L motif from the viral genome 5ʹ leader for long distance looping in transcriptional template switching to synthesize individual sgRNAs containing a common 5ʹ leader (Fig. 6C andE). The TRS B of ns12.9 sgRNA was mapped to the 3ʹ half of the S ORF, having a sequence motif of UCAAAAC (nt 27682-27688) differing from the TRS L motif by one nucleotide (underlined) in the third position (Fig. 6F). This TRS B motif is different from the reported TRS B of nt12.9 (36). The E and M sgRNAs are formed by using a 7-nt motif also differing from the TRS L by just one nucleotide in the third position (UCCAAAC, nt 27978-27984 and nt 28367-28373, respectively) (Fig. 6G andH). However, the TRS B motif of E sgRNA was not reported (36). This TRS B for synthesis of E sgRNA is positioned in the 3ʹ half ns12.9 ORF region, 123-nt upstream of the E translation initiation codon AUG. The TRS B of the N sgRNA has a sequence of UCUAAAU (nt 29065-29071), as reported (36), but differs from the TRS L motif by just one nucleotide in the seventh position (Fig. 6I). The mapped TRS B motif for HE sgRNA in this report differs from the reported HE TRS B motif (36) and is the only one containing an eight nt sequence (UAUUAAAC, nt 22337-22344) (Fig. 6D) from all other mapped 7-nt TRS B motifs. All in all, we have verified and mapped all TRS B motifs interacting with the TRS L motif in the viral genome 5ʹ-UTR. We believe that these interactions mediate the 5ʹ leader sequence directly jumping (template switching) to the 5ʹ-end of each sgRNA to regulate translation of a viral structural or accessory protein during coronavirus infection.
## Role of the mapped ns12.9 TRS B in sgRNA synthesis and virus production
After mapping all TRS B motifs upstream of individual structural and accessory ORF in hCoV-OC43, we examined the function of the mapped TRS B in accessory sgRNA synthesis. The accessory proteins of coronaviruses are nonessential, in general, for virus replication and infectious virus production (7, 8, 10-12, 34, 35), but hCoV-OC43 ns12.9 was recently reported as a viroporin essential for viral morphogenesis (32). However, we showed that the CPER-derived hCoV-OC43 NG-ns12.9 ds-circDNA exhibited more efficient replication and virus production than the CPER-derived hCoV-OC43 NG-Δns2 ds-circDNA (Fig. 3). The mapped TRS B motif in our study (Fig. 6F) is different from the reported TRS B for hCoV-OC43 ns12.9 (36). Subsequently, we examined how introduction of point mutations into our mapped ns12.9 TRS B motif affects the ns12.9 sgRNA synthesis and production of infectious virus. The point mutations in the ns12.9 TRS B motif were randomly introduced as silence mutations so that the coding function of individual codons in the S ORF was maintained for normal S expression in CPER-derived hCoV-OC43 NG-ns12.9 ds-circDNA. Five mutants, MT-1 to MT-5, with the introduced mutations in our mapped ns12.9 TRS B motif UCAAAAC or its adjacent regions, either upstream or downstream (Fig. 7A), were compared with the WT NG-ns12.9 for their replication and virus production. By transfection of HEK293T and co-cultivation with HCT-8 cells for 7 days, we quantified the number of NG + cells and their fluorescent intensity by FACS (Fig. 7B andC). The WT showed ~45.6% of NG + cells (Fig. 7B), with the higher NG intensity reaching a median fluorescent intensity (MFI) of 1,921.0 (Fig. 7C). As expected, we found that introduction of mutations into the ns12.9 TRS B motif reduced the average number of NG + cells to 17.4% (MT-1), 13.4% (MT-2), 24.6% (MT-3), 11,4% (MT-4), and 32.6% (MT-5) (Fig. 7B andD) and their MFI also dropped significantly to 69 (MT-1), 63.7 (MT-2), 63.7 (MT-3), 53.8 (MT-4), and 81.7 (MT-5) (Fig. 7C). The MT-3 had mutant TRS B plus point mutations immediate downstream and showed no significant reduction of NG + cells, but a remarkable reduction of MFI. The MT-5, which contains a WT TRS B motif, but the mutations in the adjacent regions both upstream and downstream, also showed no significant reduction of the NG + cells but a remarkable reduction of MFI when compared to the WT ns12.9 ds-circDNA (Fig. 7B through D).
Total cell RNA was isolated 7 days post co-cultivation for RT-PCR (Fig. 7E) and North ern blot (Fig. 7F) analyses. As expected, RT-PCR analysis for the WT ns12.9 TRS B RNA using the primer pair described in Fig. 6A and F showed a major amplicon (347 bp) correspond ent to the NG-ns12.9 sgRNA mediated by the WT TRS B motif (UCAAAAC, nt 27682-27688) (Fig. 7E, band 1 and its sequence). Unexpectedly, introduction of silence muta tions into the ns12.9 TRS B preferentially generates a smaller RT-PCR amplicon (276 bp) (Fig. 7E, band 2), but a weak 347 bp product (Fig. 7E, bands 1*, 1$, and 1#). Sequencing this 276 bp product showed that an alternative TRS B (UCUAGCA, nt 27753-27759), 64 nt downstream of the WT TRS B , was activated for synthesis of the ns12.9 sgRNA (Fig. 7E, band 2 and its sequence). This alternative TRS B at nt 27753-27759 is not the reported TRS B at nt 27771-27777 (36). Sequencing of the products 1*, 1$, and 1# showed that the 1* and 1# products had an expected sequence, respectively, from the MT-1 and MT-5, but the 1$ product from the MT-2 displayed a derived TRS B motif sequence of either UCUAGAC or UCACAGAC from the introduced mutations. Together, these data suggest an important role of the mapped TRS B and its surrounding sequences in guiding correct usage of the ns12.9 TRS B for its sgRNA synthesis.
By Northern blot analysis, we further showed the inhibition of ns12.9 sgRNA synthesis by introduction of point mutations into the ns12.9 TRS B motif (Fig. 7F, compare lane 2 to lanes 3-7). However, the profile of other sgRNA species appeared relatively normal, but a notably reduced expression with band density (Fig. 7F). Analysis of spike (S) protein production by Western blot showed a similar result (Fig. 7G). These data indicate that disruption of the ns12.9 TRS B motif function by point mutations could inhibit the discontinuous transcription and protein translation. This is consistent with a previous report that ns12.9 is a viroporin responsible for virion morphogenesis and pathogenesis (32).
## The mapped TRS B in hCoV-OC43 M expression is essential for sgRNA synthe sis and virus production
We further examined the mapped M TRS B function in sgRNA synthesis and virus production. We introduced point mutations, by a 6 bp linker-scanning strategy, into the mapped TRS B motif for the M sgRNA synthesis from the hCOV-OC43 NG-ns12.9 ds-circDNA. A 6-base linker, CACGAU, was introduced progressively from 5ʹ to 3ʹ to scan the 15 bp sequence covering the TRS B motif between the ORF E and ORF M. A total of four mutants (MT-6 to MT-9, Fig. 8A) were created in the subsequent CPER reactions.
After transfection of HEK293T and co-cultivation with HCT-8 cells for 7 days, we performed FACS analysis of the transfected cells to quantify the NG expression (Fig. 8B andC). We observed that ~58.8% of the cells transfected with a WT NG-ns12.9 were NG + cells (Fig. 8C) with a median fluorescent intensity (MFI) of 44.9 (Fig. 8B). All M TRS B mutants displayed no or very few (< 4.5%) NG + cells (Fig. 8C andD), with an MFI lower than 15 (Fig. 8B). Using a 32 P-labeled anti-sense oligo probe from the hCoV-OC43 N ORF region capable to detect all viral +gRNA and +sgRNAs, we performed Northern blot analyses on total cell RNA collected on the D7 and showed a profile of WT virus transcrip tion and replication (Fig. 8E), but the mutants had no or very little production of individ ual sgRNAs, further confirming the FACS and light microscope observations (Fig. 8B through D). Data indicate that the introduction of point mutations into the M TRS B motif led to the inhibition of hCoV-OC43 virus transcription and production of infectious viruses.
Relative to the WT NG-ns12.9, the MT-6 mutant, although not lethal, displayed a remarkable sgRNA reduction (Fig. 8E) and very little spike (S) protein production (Fig. 8F). This could be because the MT-6 mutant has the 6-base linker CACGAU intruding only two positions 5' to the M TRS B motif UCCAAAC bearing an A in its -1 position (Fig. 8A). This feature makes the linker 3ʹ AU dinucleotide in the MT-6 mutant mimics the 5ʹ A/UC… of the M TRS B motif by just missing one nucleotide C, consequently, resulting in a weak mutation to the M TRS B motif in the MT-6 mutant. To explore this assumption, we applied the culture supernatant collected from MT-6-transfected and co-cultivated HCT-8 cells to re-infect fresh HCT-8 cells and observed outgrowth of MT-6 NG-ns12.9 virus (Fig. S4A). RT-PCR analysis of total cell RNA of HCT-8 cells at 7 days of infection gave an expected 272 bp product from the M sgRNA (Fig. S4B). Sequencing of the RT-PCR product showed a reversion of mutated MT-6 M TRS B "AUCAAAC" to the WT M TRS B "UCCAAAC" (Fig. S4).
## DISCUSSION
The reverse genetics technique is a common method for recombinant RNA virus production (37-39) and has been successfully applied to generate infectious clones of SARS-CoV-2 and other coronaviruses (26,40). We have tried to use this conventional technique to generate hCoV-OC43 and hCoV-NL63 infectious clones but failed to make a complete hCoV-OC43 or hCoV-NL63 cDNA genome (data not shown) nor a complete set of their cDNA fragments due to their extremely AT-rich features (2) and unnoticed bacteria-toxic/genetic instability (41)(42)(43). The circular polymerase extension reaction or CPER, originally developed for flaviviruses (23,44,45), has been used to generate recombinant RNA viruses with a large genome, including coronaviruses (24,25). In this report, we successfully adapted and optimized the published CPER protocol and efficiently constructed hCoV-OC43 and SARS-CoV-2 ds-circDNAs for production of infectious viruses, without cloning of the amplified cDNA fragment and plasmid propagation in bacteria. Using this CPER technology, we were able to insert an NG reporter into the hCoV viral genome to study TRS B -TRS L interaction-mediated sgRNA synthesis. Subsequently, we introduced point mutations into the mapped TRS B to disrupt MT-1 to MT-5 mainly generated a 276 bp product (band-2) by using an alternative TRS B motif, 64 nt downstream of the WT ns12.9 TRS B . A nucleotide with a red box indicates the introduced mutation and with an underline indicates unexpected mutations. (F) Northern blot analysis of hCoV-OC43 NG-ns12.9 RNA from infected HCT-8 cells in co-cultivation with HEK293T cells transfected with the CPER-derived WT NG-ns12.9 TRS B ds-circDNA (lane 2) or a mutant NG-ns12.9 TRS B MT-1 to MT-5 ds-circDNA (lanes 3 to 7, respectively). Cells without transfection served as a mock infection (MK, lane 1). Total RNA extracted from the co-cultivated cells on D7 was analyzed by Northern blot using a 32 P-labeled probe antisense to the hCoV-OC43 N ORF. The bands correspondent to each sgRNA are labeled on the right.
(G) Western Blot analysis of total protein extracted from infected HCT-8 cells in co-cultivation with HEK293T cells transfected with the CPER-derived WT NG-ns12.9 TRS B ds-circDNA (lane 1) or a mutant NG-ns12.9 TRS B (MT-1 to MT-5, lanes 2 to 7) ds-circDNA to detect the presence of spike protein with an hCoV-OC43 spike-specific polyclonal antibody. The intensity of each protein band was quantified and used to calculate the relative amount of spike protein (%). The human GAPDH was used as a loading control. the TRS B -TRS L cross-interactions and the discontinuous transcription and replication. Thus, the optimized CPER technology in this report provides an alternative reliable tool for robust production of an infectious clone with any desired manipulations. Coronaviruses synthesize their sgRNAs for translation of viral structural and acces sory proteins (2,19). The current presumption is that the -sgRNA is synthesized first from viral +gRNA by a proposed but not experimentally confirmed template switch mechanism that the 5ʹ leader sequence in the +gRNA 5ʹ-UTR could become a 5ʹ leader of each +sgRNA. This process is mediated by the cross-interactions between a TRS B upstream of a given structural or accessory ORF and a distant TRS L in the 5ʹ-UTR through long-range base-pairing within the RTC complex, progressing in a 5ʹ-3ʹ transcription direction from the +gRNA 3ʹ-end (2,5,14,21). Consequently, these sgRNAs bearing the same 5' leader in variable sizes appear in a magnitude order, with greatest abundance of the N sgRNA from the +gRNA 3ʹ half to the lowest for the S sgRNA located towards the +gRNA 5ʹ half, because the TRS B motifs towards the viral genome 5ʹ half require more read-through steps and more time to reach during viral RNA transcription (2) (Fig. 3D, lane 3 and Fig. 4E, lane 1). Interestingly, none of these sgRNA could be inclu ded in a matured virion, with an unknown mechanism. Using the optimized CPER for positional insertion of an NG reporter into two distanced accessory ORFs, NG-ns12.9 and NG-Δns2 for hCoV-OC43, and NG-ΔORF7a and ORF3-NG for SARS-CoV-2, we were able to provide the first-hand experimental data supporting the template switch mechanism during coronaviral transcription and replication by cross-interactions between TRS B -TRS L interaction through long-range base-pairing for viral sgRNA synthesis. We demonstrated, as expected, more production of infectious NG-viruses with the NG insertion towards the viral genome 3ʹ-half (NG-ns12.9 and NG-ΔORF7a) than that of the NG-viruses with the NG insertion towards the viral genome 5ʹ-half (NG-Δns2 and ORF3-NG), which were in parallel with the production of viral NG-sgRNAs in a magnitude abundance order (Fig. 3D and4E).
Since cell reinfection with the same MOI of recovered hCoV-OC43 NG-ns12.9 and NG-Δns2 virions did not show any difference in virus infectivity and replication, the observed difference in NG + virus production from the CPER-derived hCoV ds-circDNAs with a positional NG insertion would reflect the speed of the corresponding sgRNA synthesis in a reduced 5ʹ-3ʹ transcription progression in the discontinuous template switch. In fact, our observed results mirrored both sgRNA and thus NG-protein levels in virus transcription and gene expression. Since the matured hCoV-OC43 virions do not contain any sgRNA nor accessary proteins (in this case, NG-tagged proteins), the matured virions recovered from the infected cell culture supernatant, when adjusted to the same MOI for virion infection (Fig. 5), would lead expectably to a similar productivity, without any NG insertion-induced mutations in the virus genome (Table S4). Therefore, the positional NG insertion strategies in our study have provided an applicable tool to monitor the orientational sgRNA transcription in speed and space, thus virus replication and production.
We also verified the most of TRS L and TRS B 7-nt motif sequences in hCoV-OC43 reported from a previous study (36) and showed their difference mostly by one nucleotide. Other new findings from our study are as follows. (i) We mapped the TRS B motif of E sgRNA, which was not identified from the previous report (36). This TRS B for E sgRNA synthesis is positioned in the ns12.9 ORF region, 123 nt upstream of the E translation initiation codon AUG. (ii) We identified the TRS B motif for ns12.9 sgRNA synthesis is located at the coding region of the 3ʹ S ORF, not the reported intergenic region between the S ORF and ns12.9 ORF (36). However, (iii) introduction of point mutations into our mapped TRS B for the NG-ns12.9 led to a reduced synthesis of the NG-ns12.9 sgRNA and virus production because the hCoV-OC43 ns12.9 is a viroporin for virion morphogenesis and pathogenesis (32). In addition, it is possible that the silent mutations introduced into the ns12.9 TRS B in the 3ʹ S ORF might affect the translation termination of S protein. (iv) The most important finding in this report is that an alternative TRS B motif could be activated upon introduction of point mutations into the mapped ns12.9 TRS B motif. (v) We also found that the TRS B motif for the HE sgRNA has a sequence of eight nts. (vi) Introduction of mutations into the mapped TRS B for structural M protein, as expected, drastically impaired the hCoV-OC43 replication and virus production, but could be rarely reverted.
In summary, we have established an efficient CPER for inserting an NG reporter into accessory ORFs in different genomic locations of hCoV-OC43 and SARS-CoV-2. We now provide experimental data to support our proposed "first-come, first serving model" for viral sgRNA synthesis, using viral +gRNA as a template through a discontinuous template switching mechanism mediated by TRS B -TRS L long distance cross-interactions (2) (Fig. 9A). Thus, an sgRNA using a TRS B near the +gRNA 3ʹ-end will be synthesized more efficiently than those using a TRS B towards the +gRNA 5ʹ-half. As predicted, insertion of an NG reporter into an accessory protein ORF (ap-B) toward the gRNA 3ʹ-end would synthesize more NG-sgRNAs and consequently more NG + viruses with stronger NG signals than that of an NG virus with the insertion of an NG reporter into an accessory protein ORF (ap-A) towards the gRNA 5ʹ-half (Fig. 9B andC).
## MATERIALS AND METHODS
## Coronavirus reference sequences
The nucleotide positions used in this work are derived from the reference genomes for SARS-CoV-2 (GenBank accession no. MN985325) (46) and hCoV-OC43 (GenBank acces sion no. NC_006213.1) (33,36,47) or hCoV-OC43 ATCC VR-759 (GenBank accession no. AY391777) genome sequence.
## Cell lines
The human embryonic kidney HEK293T (CRL-3216), human colorectal adenocarcinoma HCT-8 (CCL-244), monkey kidney epithelial LLC-MK2 (CCL-7) and American mink lung epithelial Mv1Lu (CCL-64) cells were obtained from ATCC. Stable transfection with hACE2 receptor plasmid was also used to generate hamster kidney fibroblast BHK21-hACE2 cells (48). All cell lines were maintained in complete Dulbecco's modified Eagle medium (DMEM, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS, Cytiva) and 1× penicillin-streptomycin-glutamine (PSG, Thermo Fisher Scientific) at 37°C in 5% CO 2 atmosphere.
## Virus stocks
The hCoV-OC43 (VR-1558 originated from VR-759) virus stock was obtained from ATCC. hCoV-OC43 was propagated in HCT-8. SARS-CoV-2 from BEI Resources was propagated in BHK21-hACE2 cells. The cells were infected by 1 h virus adsorption at 37°C. Unbound virus was washed away, and cells were incubated with DMEM (2% FBS) at 33°C for 7 days or until appearance of the cytopathic effect (CPE). The culture supernatant was collected and frozen at -80°C in 1 mL aliquots containing 10% DMSO. The infected cells were lysed in TriPure Isolation Reagent (Roche) for total RNA extraction or in 2× LDS protein sample buffer containing 5% mercaptoethanol for protein detection by Western blot.
## Preparation of hCoV-OC43 and SARS-CoV-2 specific linkers for CPER
We obtained the linker plasmid and sequence from Alberto A. Amarilla and Alexander A. Khromykh (25) and further verified in our laboratory. This linker in size of 1,119 bp is a central component for efficient CPER and contains a 30 bp stretch of adenosine sequence, a 113 bp hepatitis delta virus ribozyme (HDVr), an SV40 poly-A (pA) signal, and a CMV IE promoter (Fig. S1A). The linker was amplified by PCR to have 20 bp nucleotides from the viral genome 3ʹ-UTR on the linker 5ʹ-end and 37 bp nucleotides from the viral genome 5ʹ-UTR on the linker 3ʹ-end. This allows the linker to circle the virus genome from its 3ʹ-end to the 5ʹ-end in a CPER, enabling the CMV IE promoter upstream of the viral 5ʹ-UTR for viral transcription and the HDVr downstream of the 30-nt stretch of adenosines to produce homogenous viral RNAs with the same 3ʹ-end by ribozyme-mediated selfcleavage (23,25,44,45,49).
## Preparation of cDNA fragments from hCoV-OC43 and SARS-CoV-2
As diagrammed in Fig. S2Aand B, and Table S1, we amplified by PCR using high-fidelity Platinum SuperFi II DNA Polymerase seven overlapping cDNA fragments (F1 to F7), respectively, from FL SARS-CoV-2 and hCoV-OC43 genomic cDNAs (Fig. S1B, steps 1 and 2, and Table S1), with a 20-40 nt overlapping sequence from each fragment. The fragment positions from individual coronavirus genomes and oligos used for the amplification are summarized Tables S1 andS3. By annealing the overlapped cDNA fragments and the viral specific linker (Fig. S1B, step 3), the semicircular cDNA could be formed through base-pairing of the overlapped sequences from one fragment to another, and each fragment can be served as a template and also as a primer to fill the gap between two fragments by PrimerSTAR GXL DNA polymerase (Takara) (Fig. S1B, step 4), finally resulting in generation of ds-circDNAs (Fig. S1B,step 5).
We obtained infectious SARS-CoV-2 cDNA clones from Dr. Pei-Young Shi (26) originated from the first US-reported SARS-CoV-2 strain (2019-nCoV/USA_WA1/2020) and made a SARS-CoV-2-specific linker. Two (F6 and F7) of seven fragments derived from this SARS-CoV-2 cDNA were used for insertion of an NG reporter (22). The F6 had a NG fused downstream in frame with accessory ORF3 (ORF3-NG) and the F7 had a NG replacement of accessory ORF7a (NG-ΔORF7a) (Fig. 1A; Fig. S2A). Seven (F1 to F7) cDNA fragments (Fig. S2B and Table S1) covering the entire hCoV-OC43 genome were amplified from an FL hCoV-OC43 cDNA obtained from hCoV-OC43infected HCT-8 cells (Fig. S2B), and a hCoV-OC43 specific linker was made according to the hCoV-OC43 genome sequence. The F5 fragment from hCoV-OC43 cDNA had an NG replacement of the N-terminal accessory ORF2 (NG-Δns2, Fig. 1A) or an NG insertion upstream of the ORF2 (NG-ns2) (Fig. S2B). The NG-ns2 had a picornavirus 2A ribosomal skipping sequence (T2A, black box) insertion between the NG and the ns2 (Fig. S2B), and F7 from hCoV-OC43 cDNA had an NG insertion upstream of the accessory ORF12.9 (NG-ns12.9), separating again by a T2A sequence (50)(black box in Fig. S2B).
hCoV-OC43 fragment 2 (from nt 3646 to nt 8298) was amplified by PCR with an oligo pair of oCOC3 and oCOC4 using hCoV-OC43 cDNA as a template. The final PCR product was column-purified and cloned at the pCR-XL-2 TOPO vector (Thermo Fisher Scientific). The BsaI restriction sites were introduced by the oCOC3 and oCOC4. Plasmid pCOC4 obtained was verified by sequencing and used to determine hCoV-OC43 copy number.
## Optimizing CPER for production of coronavirus infectious ds-circDNAs
The CPER (23,24,25) was performed with some modifications using 7 overlapping cDNA fragments (Fig. S2A andB) plus a coronavirus-specific linker using high-fidelity Platinum SuperFi II DNA Polymerase (Thermo Fisher Scientific). We determined the optimal fragment concentration for individual viral CPER by using 0.01, 0.05, and 0.1 pM of individual gel-purified cDNA fragments and a virus-specific linker (Fig. 1B) in a 50 µL reaction containing 1× GXL buffer, 200 µM dNTP mix, and 2 µL of PrimerSTAR GXL DNA polymerase (Takara). The CPER cycling conditions are as follows: initial denaturation at 98°C for 30 seconds; 12 cycles of denaturation at 98°C for 10 seconds, annealing a 55°C for 20 seconds and extension at 68°C for 10 minutes; and final extension of 68°C for 10 minutes. The efficacy of CPER was determined by electrophoreses in a 0.8% agarose gel, and final product size was estimated based on ExcelBand XL 25 kb DNA ladder (SMOBIO).
## Cell transfection and co-cultivation of CPER-derived viral ds-circDNAs
CPER-derived FL ds-circDNAs of SARS-CoV-2 or hCoV-OC43, without gel-purification, were used to transfect HEK293T cells in two wells of a 6-well plate (each well having 0.25 × 10 6 cells seeded 24 h before transfection) using Lipofectamine LTX PLUS Reagent (Thermo Fisher Scientific). Prior to transfection, the medium was replaced with fresh DMEM containing 2% FBS, and each well of the cells was co-transfected with 25 µL of the CPER products mixed with 1 µg of virus-specific N expression plasmid (pCSR24 for SARS-CoV-2, pCOC42 for hCoV-OC43). The transfected cells were incubated at 37°C for 24 h before addition of 1 × 10 6 cells per well of BHK21-hACE2 for SARS-CoV-2 or of HCT-8 for hCoV-OC43 in DMEM with 10% FBS for another 48 h incubation at 33°C. The co-cultured cells were then passed into a T75-flask and cultivated in DMEM with 10% FBS at 33°C for the indicated time point.
Virus growth in CPER-transfected cells was monitored daily for NG expression by direct fluorescent microscopy for a period of 2-5 days, and the number of NG + cells was determined in ten randomly selected microscopy fields. Alternatively, the number of NG + cells was quantified by flow cytometry (FACS) on LSR II system (BD Biosciences) and the data analyzed with FlowJo Software. Finally, the culture supernatant containing infectious virions was collected and stored in 1 mL aliquots containing 10% DMSO at -80°C. The remaining cells in half were lysed in 5 mL of TriPure Isolation Reagent (Roche) for total RNA isolation or lysed in 2 X SDS buffer for protein assays.
## Plaque assay by using LLC-MK2 and Mv1Lu cells
LLC-MK2 and Mv1Lu cells were used for plaque assays using serial 10-fold dilutions of WT hCoV-OC43 virus-containing cell culture supernatants from infected HCT-8 cells. In brief, 0.5 × 10 6 Mv1Lu cells per well were seeded in a 24-well plate for 24 h and then infected in duplicate with 100 µL of a tenfold-dilution of the HCT-8 cell culture supernatant in serum-free DMEM. Following 1 h absorption at 37°C, the infected cells were rinsed with 1 mL 1 x PBS, overlayed with 1 mL of semisolid DMEM-methylcellulose media (1 x DMEM with 10% FBS, 0.5% methylcellulose, 0.25% NaHCO 3 , and 1% PSG), and then incubated at 33°C for 8 days before fixation with 3.7% formaldehyde for 30 minutes and staining for 5 min with 1% crystal violet. The visible plaques per well were counted to calculate viral titters in plaque-forming units (PFU/mL).
Alternatively, number of fluorescent plaques per well in an NG-containing coronavirus infection could be determined by fluorescent microscopy as fluorescent-forming units (FFU). The virus titer per mL (FFU/mL) could be further calculated.
## TCID50 assay
The culture supernatants of infected BHK21-ACE2 cells in co-cultivation were collected at the indicated time. After serial 10-fold dilutions, fresh BHK21-hACE2 cells in 24-well plates were infected by 100 µL of each diluent in triplicate and monitored for virus-induced cytopathic effect (CPE) for 48 h. Virus titers are calculated as TCID50/mL.
## Northern blot
Total RNA (5-30 µg) was separated by 1% formaldehyde-denaturing agarose gel electrophoresis in 1× MOPS buffer together with an RNA Millennium Marker (Thermo Fisher Scientific), transferred to a nylon membrane, and probed with an antisense 32 P-labeled oligo specific to a coronavirus N ORF region for detection of all viral transcripts (Table S3). The ribosomal RNA by ethidium bromide (EtBr) staining was used as a loading control.
## Western blot
Total cell lysates were resolved on a 4%-12% Bis-Tris NuPAGE gel (Thermo Fisher Scientific), transferred to a nitrocellulose membrane, and blotted with a rabbit anti-hCoV-OC43 Spike (S) protein antibody (E4U6P, #16435 Cell Signaling Technology) or mouse anti-GAPDH (D4C6R, #97166 Cell Signaling Technology) as a loading control. After blotting with a secondary anti-rabbit antibody, the immunoreactive proteins were detected with SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific), and the signal was captured by ChemiDoc Touch imaging system (Bio-Rad).
## TRS mapping and mutational analysis
The cDNA from infected cell total RNA was used to amplify the hCoV-OC43 leader-body junction by Platinum SuperFi II DNA Polymerase (Thermo Fisher Scientific) using a forward leader primer (oCOC1) in combination with an individual sgRNA-specific reverse primer (Fig. 6A; Table S3). The amplified products in the predicted size were gel-purified and Sanger-sequenced. TRS B mutant-containing DNA fragments were generated by overlapping PCR and used by CPER to construct the expected hCoV-OC43 NG-ns12.9 ds-circDNA for the subsequent assays.
## hCoV-OC43 nps3 TaqMan assay and calculation of viral genome copy numbers
Forward primer 5ʹ-TTCCATTCAGGATGTGGGTTT-3ʹ (nt 6971-6991) and a reverse primer 5ʹ-AAATGCTCTCCTATCAGCTTCAT-3ʹ (nt 7087-7109), along with a probe 5ʹ-6-FAM/ TTGCATGTC/ZEN/AGTTCTGCTTGGCAG/3ʹ-IABkFQ (nt 7018-7041), were used for the RT-qPCR to quantify the total viral genomic RNA isolated from infected HCT-8 cells. Copy number of the virus gRNA was calculated using a standard curve generated using pCOC4 plasmid containing an hCoV-OC43 genome region from nt 3646 to 8298.
## Total RNA-seq and viral gRNA sequencing
The A549-ACE2 cells were infected with SARS-CoV-2 WT, ORF3-NG, or NG-ΔORF7a virus with 0.05 MOI. The total cell RNA was extracted at 48 h postinfection using TriPure Reagent. HCT-8 cells were infected with 100 µL of hCoV-OC43 WT, NG-Δns2, and NG-ns12.9 virus inoculum. The viral RNA was isolated from 1 mL of the infected HCT-8 cell culture supernatant harvested 6 days post infection using the Zymo Quick-RNA Viral Kit.
All RNA samples were converted to sequence libraries using the TruSeq Stranded Total RNA Kit and sequenced by Illumina MiSeq platform. Sequencing adapters from R1 and R2 FASTQ files were trimmed, and low-quality reads were removed. The filtered RNA reads from each sample were mapped to chimeric hg38-SARS-CoV-2 (GenBank accession no. MN985325) or hg38-hCoV-OC43 (GenBank accession no. AY391777) reference genomes. The consensus sequences of individual viruses were extracted using IGV and aligned with the reference genome using Clustal Omega. The observed mutations on protein-coding potential were determined based on annotated ORF.
## Statistical analysis
For comparison between two groups, we performed a two-tail Student's t-test. For comparison between multiple groups, we performed one-tail Student's t-test or ANOVA test with Bonferroni correction. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.
## References
1. Paules, Marston, Fauci (2020) "Coronavirus infections-more than just the common cold" *JAMA*
2. Brant, Tian, Majerciak et al. (2021) "SARS-CoV-2: from its discovery to genome structure, transcription, and replication" *Cell Biosci*
3. Hasöksüz, Kiliç, Saraç (2020) "Coronaviruses and SARS-COV-2" *Turk J Med Sci*
4. Woo, Lau, Lam et al. (2012) "Discovery of seven novel Mammalian and avian coronaviruses in the genus deltacoronavirus supports bat coronaviruses as the gene source of alphacoronavirus and betacoronavirus and avian coronaviruses as the gene source of gammacoronavirus and deltacoronavirus" *J Virol*
5. Kim, Lee, Yang et al. (2020) "The architecture of SARS-CoV-2 transcriptome" *Cell*
6. Gorbalenya, Enjuanes, Ziebuhr et al. (2006) "Nidovirales: evolving the largest RNA virus genome" *Virus Res*
7. Liu, Fung, Chong et al. (2014) "Accessory proteins of SARS-CoV and other coronaviruses" *Antiviral Res*
8. Fang, Fang, Zhang et al. (2021) "Functions of coronavirus accessory proteins: overview of the state of the art" *Viruses*
9. Menachery, Mitchell, Cockrell et al. (2017) "MERS-CoV accessory ORFs play key role for infection and pathogenesis" *mBio*
10. Yount, Roberts, Sims et al. (2005) "Severe acute respiratory syndrome coronavirus group-specific open reading frames encode nonessential functions for replication in cell cultures and mice" *J Virol*
11. Hodgson, Britton, Cavanagh (2006) "Neither the RNA nor the proteins of open reading frames 3a and 3b of the coronavirus infectious bronchitis virus are essential for replication" *J Virol*
12. Casais, Davies, Cavanagh et al. (2005) "Gene 5 of the avian coronavirus infectious bronchitis virus is not essential for replication" *J Virol*
13. Miao, Tidu, Eriani et al. (2021) "Secondary structure of the SARS-CoV-2 5'-UTR"
14. Wang, Jiang, Feng et al. (2021) "The SARS-CoV-2 subgenome landscape and its novel regulatory features"
15. Subissi, Imbert, Ferron et al. (2014) "SARS-CoV ORF1b-encoded nonstructural proteins 12-16: replicative enzymes as antiviral targets" *Antiviral Res*
16. Yan, Zheng, Zeng et al. (2022) "Structural biology of SARS-CoV-2: open the door for novel therapies" *Signal Transduct Target Ther*
17. Wolff, Melia, Snijder et al. (2020) "Double-membrane vesicles as platforms for viral replication" *Trends Microbiol*
18. Klein, Winter, Wachsmuth-Melm et al. (2020) "SARS-CoV-2 structure and replication characterized by in situ cryo-electron tomography" *Nat Commun*
19. Sola, Almazán, Zúñiga et al. (2015) "Continuous and discontinuous RNA synthesis in coronaviruses" *Annu Rev Virol*
20. Malone, Urakova, Snijder et al. (2022) "Structures and functions of coronavirus replication-transcription complexes and their relevance for SARS-CoV-2 drug design" *Nat Rev Mol Cell Biol*
21. Ziv, Price, Shalamova et al. (2020) "The short-and long-range RNA-RNA interactome of SARS-CoV-2" *Mol Cell*
22. Shaner, Lambert, Chammas et al. (2013) "A bright monomeric green fluorescent protein derived from Branchiostoma lanceolatum" *Nat Methods*
23. Edmonds, Van Grinsven, Prow et al. (2013) "A novel bacterium-free method for generation of flavivirus infectious DNA by circular polymerase extension reaction allows accurate recapitulation of viral heterogeneity" *J Virol*
24. Torii, Ono, Suzuki et al. (2021) "Establishment of a reverse genetics system for SARS-CoV-2 using circular polymerase extension reaction" *Cell Rep*
25. Amarilla, Sng, Parry et al. (2021) "A versatile reverse genetics platform for SARS-CoV-2 and other positive-strand RNA viruses" *Nat Commun*
26. Xie, Muruato, Lokugamage et al. (2020) "An infectious cDNA clone of SARS-CoV-2" *Cell Host Microbe*
27. Kaye (2006) "SARS-associated coronavirus replication in cell lines" *Emerg Infect Dis*
28. Wurtz, Penant, Jardot et al. (2021) "Culture of SARS-CoV-2 in a panel of laboratory cell lines, permissivity, and differences in growth profile" *Eur J Clin Microbiol Infect Dis*
29. Bracci, Pan, Lehman et al. (2020) "Improved plaque assay for human coronaviruses 229E and OC43" *PeerJ*
30. Schirtzinger, Kim, Davis (2022) "Improving human coronavirus OC43 (HCoV-OC43) research comparability in studies using HCoV-OC43 as a surrogate for SARS-CoV-2" *J Virol Methods*
31. Ye, Wang, Guan et al. (2024) "Rapid generation and characterization of recombinant HCoV-OC43-VR1558 infectious clones expressing reporter Renilla luciferase" *Biosaf Health*
32. Zhang, Wang, Ping et al. (2015) "The ns12.9 accessory protein of human coronavirus OC43 is a viroporin involved in virion morphogenesis and pathogenesis" *J Virol*
33. Shen, Yang, Ye et al. (2016) "Safe and sensitive antiviral screening platform based on recombinant human coronavirus OC43 expressing the luciferase reporter gene" *Antimicrob Agents Chemother*
34. Redondo, Zaldívar-López, Garrido et al. (2021) "SARS-CoV-2 accessory proteins in viral pathogenesis: knowns and unknowns" *Front Immunol*
35. V'kovski, Kratzel, Steiner et al. (2021) "Coronavirus biology and replication: implications for SARS-CoV-2" *Nat Rev Microbiol*
36. St-Jean, Jacomy, Desforges et al. (2004) "Human respiratory coronavirus OC43: genetic stability and neuroinva sion" *Adv Virus Res*
37. Pekosz, He, Lamb (1999) "Reverse genetics of negative-strand RNA viruses: closing the circle" *Proc Natl Acad Sci*
38. Goff, Berg (1976) "Construction of hybrid viruses containing SV40 and lambda phage DNA segments and their propagation in cultured monkey cells" *Cell*
39. Donaldson, Yount, Sims et al. (2008) "Systematic assembly of a full-length infectious clone of human coronavirus NL63" *J Virol*
40. Kurhade, Xie, Shi (2023) "Reverse genetic systems of SARS-CoV-2 for antiviral research" *Antiviral Res*
41. Almazán, Sola, Zuñiga et al. (2014) "Coronavirus reverse genetic systems: infectious clones and replicons" *Virus Res*
42. Aubry, Nougairède, Gould et al. (2015) "Flavivirus reverse genetic systems, construction techniques and applications: a historical perspective" *Antiviral Res*
43. Setoh, Prow, Peng et al. (2017) "De novo generation and characterization of new Zika virus isolate using sequence data from a microcephaly case" *mSphere*
44. Setoh, Prow, Rawle et al. (2015) "Systematic analysis of viral genes responsible for differential virulence between American and Australian West Nile virus strains" *J Gen Virol*
45. Harcourt, Tamin, Lu et al. (2020) "Severe acute respiratory syndrome coronavirus 2 from patient with coronavirus disease, United States" *Emerg Infect Dis*
46. St-Jean, Desforges, Almazán et al. (2006) "Recovery of a neurovirulent human coronavirus OC43 from an infectious cDNA clone" *J Virol*
47. Hu, López-Muñoz, Kosik et al. (2024) "Recombinant OC43 SARS-CoV-2 spike replacement virus: an improved BSL-2 proxy virus for SARS-CoV-2 neutralization assays" *Proc Natl Acad Sci*
48. Schürer, Lang, Schuster et al. (2002) "A universal method to produce in vitro transcripts with homogeneous 3' ends" *Nucleic Acids Res*
49. Liu, Chen, Wall et al. (2017) "Systematic comparison of 2A peptides for cloning multigenes in a polycistronic vector" *Sci Rep* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846610&blobtype=pdf | # Transcriptomic and Epitranscriptomic Landscape of Integrated HTLV-1 in MT2 Cells
Shuanglong Wei, Bohan Zhang, Jingwan Han, Hanping Li, Yongjian Liu, Lei Jia, Jingyun Li, Xiaotian Huang, Lin Li
## Abstract
Human T-lymphotropic virus type 1 (HTLV-1), the first human retrovirus identified, is linked to adult T-cell leukemia and HTLV-1-associated myelopathy/tropical spastic paraparesis. However, its post-transcriptional regulation remains poorly understood. Here, we used Oxford Nanopore direct RNA sequencing to profile the HTLV-1 transcriptome and epitranscriptome in MT2 cells. We identified 23 transcript isoforms, encompassing canonical and novel splice variants. Polyadenylation analysis revealed a predominant poly(A) tail length of around 50-100 nucleotides with transcript-specific variations. Distinct RNA modifications, including pseudouridine, N 6 -methyladenosine, and 5-methylcytidine, were enriched near the 3 ′ end and varied among transcript classes, with generally lower modification ratios in viral transcripts. These findings provide a more comprehensive map of HTLV-1 RNA splicing, polyadenylation, and modifications in MT2 cells, offering new insights into viral gene regulation and pathogenic mechanisms.
## 1. Introduction
Numerous studies have verified that the replication of viruses is strictly regulated by the host cells. Post-transcriptional RNA processing, including alternative splicing, 3 ′ polyadenylation, and epitranscriptomic modifications, is an important regulatory mechanism towards viral replication [1][2][3][4]. Human T-lymphotropic virus type 1 (HTLV-1) was the first human retrovirus to be identified in 1980 and is the most pathogenic and widespread member of the HTLV family; studies have confirmed its close association with adult T-cell leukemia (ATL) and HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) [5][6][7][8]. However, the complex post-transcriptional processing modifications and comprehensive transcriptional mechanisms of HTLV-1 remain unclear. Therefore, comprehensively elucidating the post-transcriptional processing and modification status of the virus within host cells is of great significance for studying the replication and pathogenicity of HTLV-1.
RNA alternative splicing is effectively involved in regulating gene expression, essential for cellular proliferation, differentiation, and survival [9]. Viruses, especially retroviruses, exhibit complex splicing patterns as well [1,10,11]. HTLV-1 generates multiple transcripts through alternative splicing, including gag-pol-pro, env, tax, rex, and the antisense transcript HBZ [12]. These transcripts enhance immune evasion and T-cell clonal expansion, thereby facilitating progression from asymptomatic infection to ATL [13]. Although several transcript types have been reported previously [14,15], current knowledge of HTLV-1 splicing remains fragmented and incomplete. Previous studies have mostly relied on short-read sequencing or targeted RT-PCR [16,17], which cannot resolve full-length splice isoforms or accurately define donor-acceptor combinations. As a result, the global organization of HTLV-1 splice junctions, the full repertoire of isoform diversity, and the relative abundances of individual splice products have never been systematically characterized. A comprehensive, unbiased, isoform-resolved analysis of HTLV-1 alternative splicing is therefore still lacking.
Besides alternative splicing, 3 ′ end polyadenylation is another critical post-transcriptional event that affects mRNA stability and translation efficiency [18,19]. The length of poly(A) tail directly impacts the translation efficiency of mRNA [20]. In viruses, polyadenylation plays a critical role in regulating mRNA translational efficiency and stability, and the inflammation it mediates constitutes a key pathological basis for various viral infections [21][22][23]. Up to now, research on the polyadenylation level of retroviral RNA has primarily focused on HIV, while studies on the poly(A) tail of HTLV transcripts remain unexplored [24,25]. No prior study has examined the distribution of poly(A) tail lengths across different HTLV-1 transcripts, nor the potential regulatory roles of poly(A) variation in viral RNA stability or translation. Consequently, how polyadenylation contributes to HTLV-1 RNA gene expression remains essentially unknown.
RNA epitranscriptomics constitute essential post-transcriptional regulatory mechanisms influencing mRNA stability, splicing, localization, and translation efficiency [26]. Although RNA modifications have been extensively studied in retroviruses, most research has focused on HIV. Multiple types of RNA modifications facilitate HIV-1 replication and infection by modulating viral RNA stability, translation efficiency, and immune evasion, offering a valuable reference for studying HTLV [27,28]. Recent studies have found that N6-methyladenosine (m 6 A) modifications are enriched near the 3 ′ regulatory region of HTLV-1 viral RNA [29,30], but the exact modification sites, isoform specificity, and functional consequences remain undefined. Moreover, other major RNA modifications such as 5-methylcytosine (m 5 C) and pseudouridine (Ψ) have not been systematically investigated in HTLV-1. To date, no study has provided a transcriptome-wide, isoform-resolved map of RNA modifications for HTLV-1.
Taken together, although post-transcriptional regulation plays a critical role in retroviral replication, the post-transcriptional landscape of HTLV-1 remains poorly defined. At present, there is no comprehensive dataset describing full-length splice isoforms, isoformspecific poly(A) tail dynamics, or the epitranscriptomic architecture of HTLV-1. These gaps limit our understanding of how HTLV-1 RNA processing contributes to viral replication, persistence, and pathogenesis. Therefore, a high-resolution, long-read, transcriptome-wide characterization of HTLV-1 RNA processing is urgently needed. To thoroughly characterize the post-transcriptional processing and modification status of HTLV-1, we applied Nanopore Direct RNA sequencing (DRS) technology to capture native long-read viral transcriptomic features [31,32] and subsequently explored the transcriptome of HTLV-1 in MT2 cells comprehensively, which integrates HTLV-1 complete proviral sequences reported previously [33,34]. Compared to previously used next-generation sequencing (NGS), DRS enables direct sequencing of single-stranded long RNA chains and native RNA without requiring PCR amplification or other intermediate steps [35,36]. The results of this study provide a relative overall detection of HTLV-1 transcriptome in MT2 cells. The preliminary analyses of RNA alternative splicing, polyadenylation, and RNA epitranscriptomic modifications provide a panoramic view of HTLV-1 RNA modifications and splicing profiles and provide further support for research into the post-transcriptional processing and modification regulatory mechanisms of retroviruses.
## 2. Materials and Methods
## 2.1. Cell Culture
MT2 cells [American Type Culture Collection (ATCC)], maintained in our laboratory, were used in this study. Cells were cultured in RPMI-1640 medium (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin. Cells were incubated at 37 • C in a humidified atmosphere containing 5% CO 2 . Subsequent experiments were conducted when the cell density reached ≥1 × 10 6 cells/mL and cell viability reached ≥99%.
## 2.2. RNA Extraction
Total RNA was extracted from 1 × 10 7 MT2 cells using the RNA Extraction Kit (TaKaRa, Takara Bio Inc., Kusatsu, Shiga, Japan, 9767). In brief, a number of 10 million MT2 cells were lysed with lysate buffer, and the mixture was transferred to the gDNA Eraser Spin Column to remove impurities and gDNA. An equal volume of 70% ethanol was added to the filtrate, and the mixture was transferred to the RNA Spin Column to bind the RNA. The RNA Spin Column was cleaned with Buffer RWA and Buffer RWB. The RNA was finally eluted with 100 µL of RNase-Free dH 2 O. This total RNA preparation includes both host-derived viral transcripts and a minor fraction of progeny viral RNA from virion production in MT2 cells. HTLV-1-infected T cells have been shown to produce very few free virions in vitro, indicating that progeny RNA had a negligible contribution to the epitranscriptomic dataset [37]. NanoDrop™ One Microvolume UV-Vis Spectrophotometer (Thermo Fisher, Waltham, MA, USA) was used to detect the concentration and purity of the extracted RNA. Ratios of A260/A280 in the range of 1.8-2.0 and A260/A230 in the range of 1.8-2.0 were considered indicative of high-purity RNA and could be used for the next step.
## 2.3. In Vitro Transcription of RNA
The DNA templates for in vitro transcription (IVT) were prepared by RT-PCR using the PrimeScript™ One Step RT-PCR Kit Ver.2 (Takara, RR055A), with six pairs of HTLVspecific primers containing the T7 promoter sequence and poly(T) tail (Table S1). The total RNA extracted from MT2 cells was used as PCR template. DNA products were purified and recovered using a gel extraction kit (Wizard ® SV Gel and PCR Clean-Up System, Promega, Madison, WI, USA, A9282). DNA concentration and purity were assessed using a NanoDrop™ One Microvolume UV-Vis Spectrophotometer (Thermo Fisher), and DNA in the range of 1.8-2.0 for A260/A280 and 1.8-2.0 for A260/A230 was considered as eligible DNA. Purified DNA fragments were sequenced to confirm the accuracy of the amplified sequences. RNA was synthesized in vitro using the T7 High Yield RNA Transcription Kit (Vazyme, Nanjing, China, TR101) with the PCR-purified product as the template. The resulting in vitro transcription product was then purified using the Monarch ® Spin RNA Cleanup Kit (NEB, Ipswich, MA, USA, T2050L).
## 2.4. Nanopore Direct RNA Sequencing
The DRS library was prepared using the SQK-RNA004 Kit (Oxford Nanopore, Oxford, UK, SQK-RNA004). A total of 1 µg of total RNA was mixed with reverse transcription adapter, NEBNext Quick Ligation Reaction Buffer (NEB, B6058), and T4 DNA ligase (NEB, M0202T/M), and incubated at room temperature for 50 min. A reverse transcription mix containing dNTPs, 5 × reverse transcription buffer, and SuperScript™ III Reverse Transcriptase (Thermo Fisher, 18080044) was then added. The mixture was incubated at 50 • C for 50 min, followed by 70 • C for 10 min to terminate the reaction. The resulting RNA-cDNA hybrid was purified using Agencourt RNAClean XP magnetic beads (Beckman, CA, USA, A63987). A second ligation was performed with the RNA ligation adapter, T4 DNA ligase, and ligation buffer, followed by bead purification and elution in RNA Elution Buffer. The final library concentration was determined using Qubit™ RNA HS Assay Kit (Thermo Fisher, Q32852). Before sequencing, the FLO-MIN004RA RNA flow cell (Oxford Nanopore, FLO-MIN004RA) was primed using a mixture of RNA Flush Tether and Flow Cell Flush to prepare the nanopores. The library was prepared for loading by mixing 12 µL of RNA library with 25.5 µL of Library Solution (LIS) and 37.5 µL of Sequencing Buffer. The RNA library was added dropwise to the SpotON sample port. Sequencing was then carried out on the MinION platform using an R10.4.1 flow cell (Oxford Nanopore, Oxford, UK) for 72 h.
## 2.5. Bioinformatics Analysis
Basecalling was performed using Dorado (v0.8.1) [38] with the rna004_130bps_sup@v5.1.0 model, aligning reads to the HTLV reference genome (GenBank: AF003887.1) via the --reference option. The mapped sequences were sorted and indexed using Samtools (v1.2.1) [39]. Reads with mapping quality scores < 10 were excluded from downstream analysis using the --q 10 option. Read quality and length distributions were assessed using NanoPlot (v1.42.0) [40]. Alternative splicing events were quantified using Megadepth (v1.2.0) [41], and only canonical GT-AG splice junctions were annotated when splicing sites were unknown. Subsequently, we used Samtools to extract the poly(A) tail data output by Dorado. In addition, Modkit (v0.5.0) [42] was used to extract RNA modification signals. Differential methylation analysis between the sample and IVT control data was conducted using the --dmr parameter. Modification calling was performed using Modkit with the --extract parameter, extracting base modifications from individual reads, followed by integration and stratification based on transcript type.
To mitigate biases in transcript analysis arising from HTLV proviral integration, we calibrated our results using HTLV proviral DNA sequencing data from the public NCBI SRA database (PRJNA520252). Specifically, we constructed an HTLV proviral consensus sequence from 98 SRR DNA-seq runs using BWA (v0.7.19) [43], followed by mapping the DRS RNA-seq data onto this consensus with Minimap2 (v2.28) [44] and extracting splicing junctions via Regtools (v1.0.0) [45] to assess novel splicing rates and variant overlaps.
## 3. Results
## 3.1. Overview of HTLV Sequencing and Data Quality Assessment
To gain insight into the original features of HTLV transcriptome in MT2 cells, we extracted total RNA from MT2 cells, and DRS was then performed. MT2 cells cultured under the same culture conditions in different batches were subjected to two DRS runs as biological replicates (Samples 1 and 2). As a result, a total of 2,198,967 high-quality reads were obtained for sample 1 and 2,305,578 reads for sample 2. Among them, HTLVrelated sequences accounted for 2.2% and 4.2% of the total reads, respectively (Figure 1A,B). Although the sequencing data covered the entire HTLV genome, the coverage distribution was notably uneven: In sample 1, the 5 ′ and 3 ′ ends reached a sequencing depth of over 6000, peaking at 10,000, whereas the middle region achieved only approximately 300. In sample 2, the coverage at both ends reached over 40,000, peaking at 80,000, while the central region showed only approximately 1000 (Figure 1D,F). However, the 5 ′ and 3 ′ terminal regions exhibited significantly higher coverage than the middle portion of the genome in both samples. This pronounced disparity in sequencing depth was likely attributable to the alternative splicing pattern of HTLV. Despite the observed variation in depth, both samples demonstrated consistent distributions in read length and Q-score. Most of the reads for sample 1 and sample 2 were in the 1000-3000 range (Figure 1C). The average Q-scores for sample 1 and sample 2 were above 20. Q-scores generally increased with read length, and most long reads had Q-scores above 25, indicating high-quality long-read data. Only a small proportion of long reads showed fluctuations in sequencing accuracy (Figure 1E,G). Overall, despite regional differences in sequencing depth, the sequencing quality was sufficient to provide a reliable foundation for downstream analysis of HTLV transcripts. In the subsequent analysis, paired t-test analysis was performed using sample 1 as a reference, yielding p > 0.05 across all features and thereby supporting the reproducibility and stability of our analyses.
## 3.2. Alternative Splicing Analysis of HTLV-1 Transcripts
Based on the DRS data of MT2 cells, the alternative splicing analysis of the HTLV transcripts was performed and revealed considerable HTLV-related transcriptomic diversity. As a result, certain HTLV transcripts retained multiple exons and formed diverse isoforms through various splice site combinations. A total of 18 splice sites were detected, among which were 6 donor sites and 12 acceptor sites (Figure 2A). After statistical analysis of the frequency of donor site usage, it was found that D1 (67.5%) was the most commonly used donor site, followed by D4 (25.6%). As for the usage of splicing acceptor sites, A2 (30.4%) was found to be the most commonly used site, followed by A8 (26.4%), A10 (25.6%) and A13 (15.4%). The remaining acceptor sites had much lower usage frequencies (Figure 2C,D). https://doi.org/10.3390/v18010057 Despite the 12 reported splicing sites (5 donor sites and 7 acceptor sites), we discovered 6 new splicing sites that have not been reported before, among which were 1 donor site and 5 acceptor sites. Among these sites, only A3 is used frequently, while the other sites are used less often, indicating that the transcripts relied on them may only serve as supplements. To assess whether HTLV-1 transcripts in MT2 cells originate from proviral integration, we mapped DRS reads onto an HTLV-1 proviral consensus constructed from NCBI SRA (PRJNA520252). Only a very small number of splicing methods (n = 20) corresponded to the consensus, whereas the vast majority (n = 919) did not match the proviral reference, indicating that these RNAs do not arise from incomplete proviral integration but instead reflect transcription from the viral genome and its variable splicing events (Figure S2 and Table S2).
Furthermore, we identified multiple RNA isoforms resulting from the selection of different splice donor and acceptor sites (Figure 2B). A total of 23 RNA isoforms were detected, including 8 known isoforms and 15 new ones [14,15]. Among the novel isoforms, 9 represented previously reported defective HTLV-1 integrations in the MT2 genome [34]. Among the known isoforms, those associated with tax/p27 rex were the most abundant, while p13 and p12 isoforms were the least. In terms of expression levels, the p21 rex transcript had the highest expression, followed by env and tax/p27 rex (Figure 2E). Additionally, we detected a group of RNA transcripts (others) that did not correspond to any annotated HTLV isoforms. These variants are unannotated in the reference genome annotations, lacking described functional roles, but all shared a common and highly conserved splice donor site D1, which was consistently present in known HTLV isoforms and showed strong consistency across both samples. Although the functions of these unannotated transcripts remain unclear, the conservation of D1 suggests they are genuine alternative splicing products rather than sequencing artifacts or random transcriptional noise. Furthermore, a novel splicing form, D4A10, spanning the gag, pro, and pol regions, is classified as a gag-propol transcript even though prior studies suggest this class typically arises from unspliced transcripts [14]. Our discovery of numerous novel transcript isoforms underscores the intricate post-transcriptional regulation of HTLV-1.
In summary, HTLV RNA exhibited complex alternative splicing patterns, diverse isoform types, and notable splice site usage bias. The presence of defective integrated sequences suggests that HTLV may engage in regulatory interactions with the host genome, possibly contributing to as-yet-unclear functional roles in viral gene expression.
## 3.3. Polyadenylation Analysis of HTLV-1 Transcripts
Polyadenylation is an important component of RNA modification, and poly(A) tails play a crucial role in maintaining mRNA stability and promoting mRNA translation. In this study, we detected the poly(A) tail length of HTLV transcripts in MT2 cells and analyzed the relationship between polyadenylation and mRNA expression (Figure 3 and Table 1). As shown in Figure 3A, the poly(A) tail lengths of most transcripts were distributed within the 0-200 nt range, with a prominent peak around 50 nt. Transcripts with poly(A) tails longer than 200 nt, particularly those exceeding 300 nt, were extremely rare. At the individual transcript level, tax/p27 rex had the longest average tail (131 nt), followed by gag-pro-pol (129 nt), p21 rex (127 nt), p13 (117 nt), env (112 nt), and HBZ (108 nt). There were clear differences in poly(A) tail length distributions among the various transcripts, and the overall distribution showed a statistically significant difference (p < 0.05). Note that p12 was not included in the statistics due to the small quantity (n = 3) (Figure 3B).
Overall, distinct differences in poly(A) tail lengths were observed among different HTLV transcripts in MT2 cells. This may reflect the complexity of poly(A) tail regulation, differences in transcript abundance, or inherent variability in retroviral polyadenylation mechanisms.
## 3.4. Epi-Transcriptome Analysis of HTLV-1
RNA epitranscriptomic modification is an important regulatory mechanism for viral replication. To investigate the epitranscriptomic features of HTLV transcripts, we conducted modification prediction using Dorado. Three types of RNA modifications were analyzed: N 6 -methyladenosine (m 6 A), 5-methylcytosine (m 5 C), and pseudouridine (Ψ). To reduce background noise and false positives, we included in vitro transcribed (IVT) RNA lacking modifications as a negative control (Figure S1). We used the thresholds of valid_coverage > 1000 and percent_modified > 20%. Following the literature and our data, a total of 14 high-confidence modification sites were identified: 9 (Ψ) sites, 2 m 6 A sites, and 3 m 5 C sites (Figure 4A-D). These modification sites were predominantly concentrated near the 3 ′ end of the transcripts. The identified Ψ sites were located at positions 6982, 7446, 7465, 7598, 7751, 7773, 7906, 8096, and 8430, with the highest percent_modified at position 7598 (0.72). The m 6 A sites were identified at positions 7011 and 7691, both with percent_modified values around 0.4. The three m 5 C sites were located at positions 7955, 8133, and 8311, while site 7955 showed the highest modification level (0.50). In order to detect the different modification rates of HTLV-1 transcripts reported before, we analyzed the modification type and mod_ratio at each site of every HTLV-1 transcript separately (Figure 4E-G), excluding low-abundance transcripts as previously noted. As a result, the modification ratios in these transcripts were generally lower than the overall levels. For the 9 pseudouridine modification sites, all transcripts exhibited mod_ratios lower than the overall level (Figure 4E). The m5C mod_ratio across all transcripts was consistently lower than the global level, mostly around 20% (Figure 4G). For m6A modifications, the modification ratio at site 7691 in the tax/p27 rex transcript exceeded the global level, reaching up to 51% (Figure 4F). In p21 rex and p13 transcripts, both modified sites exhibited higher mod_ratio than the overall average, while in other transcripts, the ratios for the two sites mostly centered around 15%. When grouped by transcript type, most isoforms displayed modification rates lower than the global average, with the exception of p13, which showed higher m 6 A modification levels. These high-confidence modification sites showed clear positional enrichment, suggesting that HTLV may employ RNA chemical modifications to regulate the expression or function of specific transcript regions.
In conclusion, this study provided a comprehensive overview of the structural complexity and chemical modifications of HTLV transcripts and revealed extensive alternative splicing, variations in poly(A) tail length, and region-specific RNA modifications, respectively. These results collectively offered foundational insights for future investigations into HTLV gene regulation and virus-host interactions.
## 4. Discussion
HTLV-1, as the first human retrovirus to be identified, provides important insights into the understanding of other retroviruses [5]. Although previous studies have analyzed the genomic structure and some aspects of the epitranscriptome of HTLV-1, research on RNA modifications, alternative splicing isoforms, and poly(A) tail length remains relatively limited [12,29,30]. Notably, nanopore DRS technology allows the direct detection of transcriptome without reverse transcription and amplification [31,32]. The sequence data can give messages of RNA modifications, splicing events and polyadenylation at the single-molecule level, offering unprecedented advantages.
In this study, we used DRS to profile the HTLV-1 post-transcriptional processing and modification thoroughly. For splicing events of HTLV-1, we identified 23 RNA isoforms, including canonical HTLV-1 transcripts consistent with previous reports as well as defective integrated transcripts in MT2 cells, reflecting major transcriptional patterns in this cellular context [34]. Notably, while many typical transcripts were captured, we failed to detect the p30-related splice product, consistent with its low and variable expression in HTLV-1-infected cell lines [46]. Our data revealed fragmented integrated HTLV-1 transcripts, providing further evidence for the integration of HTLV-1 in MT2 cells [15,34]. We characterized multiple isoforms of the HTLV-1 RNA nuclear export factors tax/rex [47], which can promote viral replication and cellular senescence [48]. This high expression is reflected in clinical settings where the total amount of HTLV-1 tax mRNA in peripheral blood mononuclear cells was significantly higher in HAM/TSP patients than in asymptomatic carriers and correlated with proviral load and disease severity, linking to aggressive progression in HTLV-1-associated neuroinflammation [49]. These isoform discoveries highlight the complexity of tax/rex's cellular effects during infection, providing new insights for ATL and HTLV-1-associated myelopathy/tropical spastic paraparesis. Besides this, numerous viral transcripts with no research reports and no detailed functional descriptions in the annotation file information have been discovered as well. This phenomenon gives insight into the regulation of retroviral genome, underscoring substantial variability in HTLV-1 transcript processing; such variations could potentially influence translation efficiency, protein structure, or RNA stability, offering new entry points to investigate HTLV-1 transcriptional regulation [50,51]. While these novel isoforms suggest potential regulatory parallels in retroviral epitranscriptomes, functional validation remains essential. These findings offer a solid foundation for further elucidating the roles of alternative splicing in HTLV-1 viral replication and host immune evasion, while supporting advanced investigations into ATL and HAM/TSP therapeutics [52][53][54].
Previous studies have shown that the 3 ′ end polyadenylation plays a crucial role in translation efficiency and RNA stability in different organisms [22,23,55,56]. A former study of our research group revealed that the poly(A) tail of HIV-1 subtype B (NL4-3) transcripts exhibits complex length distribution but concentrates around 50-100 nt [57]. In this study, the poly(A) tail length of HTLV-1 transcripts was found to concentrate around 50-100 nt. https://doi.org/10.3390/v18010057
In host cells, poly(A) tails are synthesized in the nucleus to defined lengths (approximately 250 nt in mammals) and subsequently shortened in the cytoplasm at transcript-specific rates, leading to steady-state tail length distributions maintained by a balance between polyadenylation and deadenylation activities [58]. Although this mechanism has been characterized primarily in cellular mRNAs, the approximately 50-100 nt tail length observed for HTLV-1 likely reflects regulated steady-state control rather than the initial nuclear synthesis length. Furthermore, our analysis of the data revealed differences in poly(A) lengths between different transcripts, and the differences were statistically significant, suggesting that the regulatory mechanism of poly(A) tail length on transcript translation is complex and precise. These findings suggest that, despite being generally constrained within a stable range, HTLV-1 poly(A) tails may mediate isoform-specific regulatory functions, with implications for isoform-dependent regulation in the viral life cycle.
RNA modifications can enhance protein translation and RNA stability, promote nuclear export, and suppress innate immune recognition [59]. Most functional studies on retroviral modifications have focused on HIV, where, for example, m 6 A has been shown to promote translation of Gag proteins, strengthen Rev-RRE interactions, and enhance nuclear export; m 5 C contributes to RNA stability and translation; and Ψ influences RNA secondary structure, thereby regulating splicing and translation [27]. In contrast, investigations on HTLV-1 RNA modifications remain scarce. Only recently was it reported that m 6 A in HTLV-1 is enriched at the 3 ′ end, with site-specific mapping still limited [29,30]. To address this gap, we used DRS to systematically identify RNA modification sites in HTLV-1 transcripts from MT2 cells. By comparing the raw electrical current signals from our cellular HTLV-1 RNA against this unmodified IVT baseline, we were able to confidently identify sites with altered current patterns indicative of modifications, effectively excluding false positives caused by sequence motifs or basecalling errors [60]. Our results were consistent with earlier observations [29]: m 5 C, m 6 A, and Ψ modifications were mostly enriched near the 3 ′ end of HTLV-1 RNA, a pattern similar to the modification distribution trend observed in HIV-1 [60]. Existing studies have shown that there are many similarities in the post-transcriptional processing of HTLV-1 and HIV-1 [47,61], and our research further supports this resemblance. This convergence likely stems from conserved transcription and replication strategies among retroviruses, such as enhanced 3 ′ UTR stability that facilitates viral genome packaging and integration into host cells [62]. Additionally, modifications at the 3 ′ end may also influence the nuclear export and translation efficiency of viral mRNA, potentially contributing to viral replication and immune evasion [63]. Thus, the enrichment of 3 ′ end modifications could be a common feature of retroviruses, and its specific role in the viral lifecycle warrants further investigation. Furthermore, the overall modification rates at individual sites were generally higher than those at the transcript level, potentially due to the potential impact of MT2 cell integration site biases. MT2 cells contain numerous defective HTLV-1 proviral integrations, whose RNA transcripts outnumber those from intact proviruses, thereby lowering site-specific modification rates at the transcript level compared to the global average [34]. Taken together, our study shows that the distribution pattern of HTLV-1 modifications resembles that reported for HIV-1. This resemblance advances understanding of HTLV-1 virology and, by comparison, offers perspective on shared molecular mechanisms among retroviruses. Given the commonalities between the two viruses in replication cycle and host interactions, findings from HTLV-1 studies may, to some extent, inform broader approaches to understanding and targeting retroviral infections. However, although MT2 cells are ideal for generating high-abundance viral RNA requisite for initial DRS profiling, they may not fully capture the heterogeneity present in primary patient samples, such as those from ATL or HAM/TSP. Therefore, our findings should be viewed as a high-resolution atlas within the context of MT2 cells, serving as a foundational resource to guide future comparative investigations in more physiologically relevant models.
Together, our characterization of HTLV-1 alternative splicing isoforms, 3 ′ end polyadenylation profiles, and RNA modification landscapes provides novel insights into the post-transcriptional regulation of this virus. The observed similarities in poly(A) tail length distribution and 3 ′ -end-enriched RNA modifications between HTLV-1 and HIV-1 point to potentially conserved strategies adopted by distinct retroviruses to fine-tune RNA stability, translation, and replication efficiency. These parallels highlight HTLV-1 as a valuable comparative model for dissecting shared and virus-specific regulatory mechanisms within the Reoviridae family. Future studies integrating functional assays with high-resolution transcriptomic profiling across multiple retroviruses will be essential to clarify whether these features represent universal RNA regulatory signatures and to explore their potential as novel targets for broad-spectrum antiviral interventions.
## References
1. Liang, Lyu, Xu et al. (2025) "Splicing-dependent restriction of the HBZ gene by Tax underlies biphasic HTLV-1 infection" *PLoS Pathog*
2. Cross, Michalski, Miller et al. (1536) "RNA regulatory processes in RNA virus biology" *Wiley Interdiscip. Rev. RNA*
3. (2026) *Viruses*
4. Barrera, Olguín, Vera-Otarola et al. (1863) "Cap-independent translation initiation of the unspliced RNA of retroviruses" *Biochim. Biophys. Acta Gene Regul. Mech*
5. Beemon (1113) "Retroviral RNA Processing"
6. Yoshida, Miyoshi, Hinuma (1982) "Isolation and characterization of retrovirus from cell lines of human adult T-cell leukemia and its implication in the disease" *Proc. Natl. Acad. Sci*
7. Popovic, Sarin, Robert-Gurroff et al. (1983) "Isolation and transmission of human retrovirus (human t-cell leukemia virus)" *Science*
8. Branda, Romano, Pavia et al. (2025) "Human T-Lymphotropic Virus (HTLV): Epidemiology, Genetic, Pathogenesis, and Future Challenges" *Viruses*
9. Takatsuki (1995) "Adult T-cell leukemia" *Intern. Med*
10. Bradley, Anczukow (2023) "RNA splicing dysregulation and the hallmarks of cancer" *Nat. Rev. Cancer*
11. Tomezsko, Corbin, Gupta et al. (2020) "Determination of RNA structural diversity and its role in HIV-1 RNA splicing" *Nature*
12. Emery, Swanstrom (2021) "HIV-1: To Splice or Not to Splice, That Is the Question" *Viruses*
13. Martinez, Al-Saleem, Green (2019) "Comparative virology of HTLV-1 and HTLV-2" *Retrovirology*
14. Nakahata, Enriquez-Vera, Jahan et al. (1543) "Understanding the Immunopathology of HTLV-1-Associated Adult T" *Cell Leukemia/Lymphoma: A Comprehensive Review. Biomolecules*
15. Ajiro, Zheng (2014) "Oncogenes and RNA splicing of human tumor viruses" *Emerg. Microbes Infect*
16. Tram, Mesnard, Peloponese (2022) "Alternative RNA splicing in cancer: What about adult T-cell leukemia?" *Front. Immunol*
17. Nakano, Yokoyama, Shin et al. (2022) "Exploring New Functional Aspects of HTLV-1 RNA-Binding Protein Rex: How Does Rex Control Viral Replication? Viruses"
18. D'agostino, Cavallari, Romanelli et al. (1958) "Post-transcriptional Regulation of HTLV Gene Expression: Rex to the Rescue" *Front. Microbiol*
19. Hajnsdorf, Kaberdin (2018) "RNA polyadenylation and its consequences in prokaryotes" *Philos. Trans. R. Soc. Lond B Biol. Sci*
20. Liu, Lu (2024) "Beyond simple tails: Poly(A) tail-mediated RNA epigenetic regulation" *Trends Biochem. Sci*
21. Biziaev, Shuvalov, Salman et al. (2024) "The impact of mRNA poly(A) tail length on eukaryotic translation stages" *Nucleic Acids Res*
22. Guo, Zhu, Hu et al. (2023) "Histone mRNA polyadenylation-mediated inflammation underlies various virus infections and cancers" *J. Med. Virol*
23. Passmore, Coller (2022) "Roles of mRNA poly(A) tails in regulation of eukaryotic gene expression" *Nat. Rev. Mol. Cell Biol*
24. Xiang, Bartel (2021) "The molecular basis of coupling between poly(A)-tail length and translational efficiency" *Elife*
25. Lin, Panfil, Sandel et al. (2022) "Novel perspectives on antisense transcription in HIV-1, HTLV-1, and HTLV-2. Front. Microbiol"
26. Selyutina, Persaud, Lee et al. (2020) "Nuclear Import of the HIV-1 Core Precedes Reverse Transcription and Uncoating" *Cell Rep*
27. Roundtree, Evans, Pan et al. (2017) "Dynamic RNA Modifications in Gene Expression Regulation" *Cell*
28. Wang, Li, Lian et al. (2022) "The Role of RNA Modification in HIV-1 Infection" *Int. J. Mol. Sci*
29. Chen, Kumar, Espada et al. (2021) "N6-methyladenosine modification of HIV-1 RNA suppresses type-I interferon induction in differentiated monocytic cells and primary macrophages" *PLoS Pathog*
30. King, Panfil (2025) "Dynamic Roles of RNA and RNA Epigenetics in HTLV-1 Biology" *Viruses*
31. King, Midkiff, Mcclain et al. (2025) "YTHDF1 and YTHDC1 m(6)A reader proteins regulate HTLV-1 tax and hbz activity" *J. Virol*
32. Ji, Feng, Huang et al. "The Applications of Nanopore Sequencing Technology in Animal and Human Virus Research" *Viruses*
33. (2026) *Viruses*
34. Lin, Hui, Mao "Nanopore Technology and Its Applications in Gene Sequencing" *Biosensors*
35. Miyoshi, Kubonishi, Yoshimoto et al. (1981) "A T-cell line derived from normal human cord leukocytes by co-culturing with human leukemic T-cells" *Gan*
36. Hashikura, Umeki, Umekita et al. (2016) "The diversity of the structure and genomic integration sites of HTLV-1 provirus in MT-2 cell lines" *Hum. Cell*
37. Einkauf, Osborn, Gao et al. (2022) "Parallel analysis of transcription, integration, and sequence of single HIV-1 proviruses" *Cell*
38. Fernandes, Folgosi, Pessôa et al. "HTLV-1 and HTLV-2 infections significantly alter small RNA expression in asymptomatic carriers"
39. Kalinichenko, Komkov, Mazurov "HIV-1 and HTLV-1 Transmission Modes: Mechanisms and Importance for Virus Spread" *Viruses*
40. (2024) "Oxford Nanopore Technologies. Dorado, Version 0.8.1; Oxford Nanopore Technologies"
41. Danecek, Bonfield, Liddle et al. "Twelve years of SAMtools and BCFtools"
42. De Coster, Rademakers (2023) "NanoPack2: Population-scale evaluation of long-read sequencing data" *Bioinformatics*
43. Wilks, Ahmed, Baker et al. (2021) "Megadepth: Efficient coverage quantification for BigWigs and BAMs" *Bioinformatics*
44. (2025) "Oxford Nanopore Technologies. Modkit, Version 0.5.0; Oxford Nanopore Technologies"
45. Li, Durbin (2009) "Fast and accurate short read alignment with Burrows-Wheeler transform" *Bioinformatics*
46. Li (2018) "Minimap2: Pairwise alignment for nucleotide sequences" *Bioinformatics*
47. Cotto, Feng, Ramu et al. (1589) "Integrated analysis of genomic and transcriptomic data for the discovery of splice-associated variants in cancer" *Nat. Commun*
48. Moles, Sarkis, Galli et al. (2019) "p30 protein: A critical regulator of HTLV-1 viral latency and host immunity" *Retrovirology*
49. Nakano, Watanabe "Tuning Rex rules HTLV-1 pathogenesis" *Front. Immunol. 2022*
50. Eusebio-Ponce, Anguita, Paulino-Ramirez et al. (2019) "HTLV-1 infection: An emerging risk. Pathogenesis, epidemiology, diagnosis and associated diseases" *Rev. Esp. De Quimioter. Publ. Of. Soc. Esp. Quimioter*
51. Miura, Naito, Saito "Current Perspectives in Human T-Cell Leukemia Virus Type 1 Infection and Its Associated Diseases" *Front. Med*
52. Cheng, Islam, Kanlong et al. "Translation of HIV-1 unspliced RNA is regulated by 5 ′ untranslated region structure" *J*
53. Rausch, Le Grice (2015) "HIV Rev Assembly on the Rev Response Element (RRE): A Structural Perspective" *Viruses*
54. Miura, Dey, Ramanayake et al. (2019) "Kinetics of HTLV-1 reactivation from latency quantified by single-molecule RNA FISH and stochastic modelling" *PLoS Pathog*
55. Murata, Hayashibara, Sugahara et al. (2006) "A novel alternative splicing isoform of human T-cell leukemia virus type 1 bZIP factor (HBZ-SI) targets distinct subnuclear localization" *J. Virol*
56. Kannagi, Hasegawa, Nagano et al. (2019) "Impact of host immunity on HTLV-1 pathogenesis: Potential of Tax-targeted immunotherapy against ATL" *Retrovirology*
57. Subtelny, Eichhorn, Chen et al. (2014) "Poly(A)-tail profiling reveals an embryonic switch in translational control" *Nature*
58. Dreyfus, Regnier (2002) "The poly(A) tail of mRNAs: Bodyguard in eukaryotes, scavenger in bacteria" *Cell*
59. Li, Huang, Yuan et al. "Characterizing transcripts of HIV-1 different substrains using direct RNA sequencing"
60. Weill, Belloc, Bava et al. (2012) "Translational control by changes in poly(A) tail length: Recycling mRNAs" *Nat. Struct. Mol. Biol*
61. Baquero-Perez, Geers, Díez (1049) "From A to m(6)A: The Emerging Viral Epitranscriptome" *Viruses*
62. (2026) *Viruses*
63. Baek, Lee, Golconda et al. "Single-molecule epitranscriptomic analysis of full-length HIV-1 RNAs reveals functional roles of site-specific m(6)As" *Nat. Microbiol*
64. Herrmann, Meng, Yang et al. (1528) "The Assembly of HTLV-1-How Does It Differ from HIV-1? Viruses"
65. Zhang, Crumpacker, Hiv Utr et al. (1084) *Viruses*
66. Phillips, Mishra, Huang et al. "Functional Impacts of Epitranscriptomic m(6)A Modification on HIV-1 Infection" *Viruses*
67. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12787211&blobtype=pdf | # Academic Editors: Sukhwinder
Singh Sohal, Robert Flisiak, Luca Pipitò, Vincenza Chiara, Mazzola, Ilenia Giacchino, Riccardo De Rosa, Carola Gagliardo, Alessio Lipari, Paola Monte, Federica Furia, Erika Mannino, Rosaria Pecoraro, Nicola Scichilone, Antonio Cascio
## Abstract
Background: Durvalumab, a PD-L1 inhibitor used as consolidation therapy after chemoradiation in unresectable stage III non-small cell lung cancer (NSCLC), can induce immunerelated adverse events, among which immune-mediated pneumonitis represents one of the most severe. Differentiating checkpoint inhibitor pneumonitis (CIP) from infectious pneumonia is challenging due to overlapping clinical and radiologic findings. Case presentation: We describe a 67-year-old woman with stage III lung adenocarcinoma treated with chemotherapy, radiotherapy, and durvalumab, who presented with progressive dyspnea and extensive bilateral ground-glass opacities on CT imaging. Laboratory tests revealed leukopenia and elevated inflammatory markers. Despite broad-spectrum antibiotic and antiviral therapy, her condition worsened, requiring high-flow nasal cannula oxygen therapy. Multiplex molecular testing on sputum identified human metapneumovirus (HMPV), while blood cultures and urinary antigens for Streptococcus pneumoniae and Legionella pneumophila were negative. A pulmonology consultation raised suspicion for severe durvalumab-induced pneumonitis exacerbated by viral infection. High-dose methylprednisolone (2 mg/kg/day) followed by a four-week taper led to gradual clinical and radiologic resolution. Durvalumab was permanently discontinued. Discussion: To our knowledge, this is the first reported case of HMPV-associated pneumonitis in a patient receiving durvalumab. This case highlights the potential synergistic interplay between viral infection and immune checkpoint blockade, resulting in severe lung injury. Comprehensive microbiologic evaluation, including molecular diagnostics, is essential to guide therapy and distinguish infectious from immune-mediated causes. Conclusions: Early recognition of mixed infectious and immune-mediated pneumonitis, and timely corticosteroid therapy are critical to achieving favorable outcomes and preventing irreversible pulmonary damage.
## 1. Introduction
Immune checkpoint inhibitors (ICIs) targeting the programmed cell death 1 (PD-1)/programmed death-ligand 1 (PD-L1) axis have revolutionized the management of advanced and locally advanced non-small cell lung cancer (NSCLC), significantly improving overall survival and long-term disease control [1,2].
Durvalumab, a fully human monoclonal antibody targeting PD-L1, prevents its interaction with PD-1 and CD-80, thereby enhancing antitumor immune responses by sustaining T-cell activation and proliferation [3]. Despite its proven efficacy as consolidation therapy after chemoradiation in unresectable stage III NSCLC, durvalumab, like other ICIs, can induce a range of immune-related adverse events (irAEs) due to nonspecific immune activation against healthy tissues [2,[4][5][6][7].
Among these, immune-mediated pneumonitis is one of the most serious and potentially life-threatening toxicities [2,[4][5][6][7]. Clinical manifestations are often nonspecific, including cough, dyspnea, and low-grade fever, while radiologic findings may overlap with infectious pneumonia, tumor progression, or radiation-induced lung injury [2]. This overlap complicates the differential diagnosis and underscores the need for a comprehensive assessment integrating clinical, radiologic, and microbiologic data [2].
Here, we present a case of severe community-acquired pneumonia (CAP) with a mixed infectious and immune-mediated etiopathogenesis, in which human metapneumovirus (HMPV) infection coexisted with durvalumab-induced pneumonitis. To our knowledge, this represents the first reported case of HMPV-associated pneumonitis in a patient receiving durvalumab therapy.
## 2. Case Presentation
In January 2025, a 67-year-old woman, a former smoker with an approximately 40 pack-year smoking history who had quit two years earlier, presented to the emergency department with progressive dyspnea, chest pain, and easy fatigability over several days, without fever.
In December 2023, she had been diagnosed with lung adenocarcinoma, treated with chemotherapy and radiotherapy, and was under follow-up at our Oncology Department. For the previous three months, she had been receiving immunotherapy with durvalumab, administered every 14 days (the last infusion two weeks before presentation). The patient received a total of six cycles of durvalumab. Her medical history was also remarkable for paroxysmal atrial fibrillation treated with apixaban, and polymyalgia rheumatica, previously managed with weekly methotrexate. The latter was self-discontinued by the patient due to excessive hair loss. Her Eastern Cooperative Oncology Group performance status was 2. She was ambulatory and capable of all self-care but unable to carry out any work-related activities.
At presentation, the patient's body temperature was 37.5 • C, blood pressure 90/60 mmHg, pulse rate 84 bpm, oxygen saturation (SpO 2 ) 91% on 3 L/min O 2 via simple face mask, and respiratory rate 28 breaths/min. Chest computed tomography (CT) demonstrated multiple, extensive ground-glass opacities predominantly involving both upper lobes, the middle lobe, the lingula, and the left lower lobe (Figure 1). Laboratory results revealed white blood cell count (WBC) 4360/mm 3 (reference range 4000-11,000/mm 3 ) with 86% neutrophils (reference range 40-74%), and C-reactive protein (CRP) 35 mg/L (reference < 5 mg/L). Arterial blood gas analysis (on room air): pH The patient was admitted to the infectious diseases unit with a working diagnosis of CAP. Empirical antibiotic and antiviral therapy was initiated with ceftobiprole 500 mg three times daily, doxycycline 100 mg twice daily, and oseltamivir 75 mg twice daily. Despite therapy, no significant clinical improvement was observed, and within a few days her respiratory status deteriorated, requiring escalation of oxygen therapy first to a Venturi mask (FiO 2 60%), and subsequently to high-flow nasal cannula (HFNC) at 40 L/min on the fifth day of hospitalization. Concurrently, inflammatory markers worsened, with CRP peaking at 96.1 mg/L and procalcitonin 0.293 µg/L (reference < 0.05 µg/L). CT angiography was negative for pulmonary embolism.
On the fourth hospital day, an extensive microbiological work-up was performed, including a multiplex molecular panel (BIOFIRE ® FILMARRAY ® Pneumonia Panel Plus, bioMérieux, Salt Lake City, UT, USA), which detects 27 bacterial and viral respiratory pathogens as well as seven antibiotic resistance genes on nasopharyngeal swab and sputum [8]. Additionally, urinary antigens for Streptococcus pneumoniae and Legionella pneumophila, and three sets of blood cultures were performed.
The molecular panel identified HMPV in the sputum, while it was negative on the nasopharyngeal swab. Sputum and blood cultures remained negative, and urinary antigens were also negative.
After a pulmonology consultation, the patient was suspected to have severe durvalumab-induced immune-mediated pneumonitis, potentially exacerbated by HMPV infection, based on recent durvalumab exposure and typical radiologic findings.
Methylprednisolone 40 mg twice daily (2 mg/kg per day) was administered for 10 days, followed by a 14-day tapering regimen, with subsequent clinical improvement. Antibiotic therapy was discontinued after 7 days. During hospitalization, the patient developed leukopenia, attributed to durvalumab, with a nadir WBC of 1050/mm 3 on day 5 (neutrophils 70.5%). Progressive normalization was later observed, reaching 6520/mm 3 at discharge.
Gradually, the patient was successfully weaned off HFNC, and on the 15th day, continued oxygen support via nasal cannula; finally, oxygen therapy was discontinued. After 18 days of hospitalization, the patient was discharged in good clinical condition and on room air. Four months after discharge, the patient underwent a contrast-enhanced whole-body CT scan for oncologic follow-up, which showed a marked reduction in areas of increased parenchymal density compared with the prior CT (Figure 2). At the follow-up oncology visit, the oncologist, agreeing with the diagnosis of severe durvalumab-induced immune-mediated pneumonitis, decided not to reintroduce the drug.
The patient did not resume durvalumab therapy thereafter, and no other ICIs were administered during the 10 months following discharge.
## 3. Discussion
PD-L1 inhibitor-associated severe pneumonitis is a well-documented immune-related adverse event. Patients diagnosed with irAEs in the emergency department generally present with higher-grade toxicities, and approximately 3.5% of patients with severe irAEs require hospitalization and corticosteroid treatment [1,2].
The condition may manifest at any time during therapy, although onset typically occurs within 6-12 weeks after initiation [2]. In our patient, symptoms developed approximately three months after starting durvalumab, consistent with the expected temporal pattern. Prior thoracic chemoradiation likely contributed to increased susceptibility to pulmonary toxicity, as reported in several studies describing synergistic lung injury mechanisms between radiation-induced fibrosis and immune activation [1]. Furthermore, recent retrospective analyses have identified male sex and pre-existing autoimmune disorders as potential risk factors for severe pneumonitis during durvalumab therapy [9]. Our patient's history of polymyalgia rheumatica, albeit inactive, may have predisposed her to dysregulated immune activation, further contributing to the development of pneumonitis.
The differential diagnosis between infectious pneumonia and CIP remains challenging in patients receiving immune checkpoint inhibitors, especially when presenting with fever and elevated inflammatory markers. The integration of molecular diagnostics provides significant clinical value in this context. This technology enables rapid detection of bacterial and viral pathogens even after the initiation of empirical antimicrobial therapy, improving diagnostic accuracy and enabling appropriate therapeutic adjustments. In our case, molecular testing was decisive in revealing a mixed infectious-immune etiology, guiding the discontinuation of antibiotics and the initiation of corticosteroid therapy.
The clinical presentation of CIP is often nonspecific, including dyspnea, cough, and fatigue, and may closely resemble infectious pneumonia or tumor progression [2]. In this case, the patient presented with progressive dyspnea, mild fever, and extensive ground-glass opacities on CT imaging, initially suggesting CAP. The absence of bacterial pathogens in sputum and blood cultures, along with negative urinary antigen tests and a poor response to broad-spectrum antibiotics and antiviral therapy, raised suspicion of immune-mediated pneumonitis. The identification of HMPV through multiplex PCR testing, however, introduced an additional layer of complexity.
HMPV is increasingly recognized as a cause of lower respiratory tract infection in adults, particularly in immunocompromised and oncologic patients [10][11][12]. The detection of HMPV exclusively in the sputum sample, but not in the nasopharyngeal swab, supports its localization in the lower respiratory tract and strengthens its pathogenic role.
Currently, no specific antiviral therapy is approved for HMPV. Management is primarily supportive. Agents such as ribavirin or immunoglobulin preparations have been used in selected or severely immunocompromised patients, but evidence for their efficacy remains limited [13].
Although HMPV can cause severe pneumonia on its own, its presence in this case may have acted as a trigger or amplifier of the immune-mediated pulmonary inflammation initiated by durvalumab.
This co-detection underscores the potential interaction between viral infection and immune checkpoint blockade, in which viral activation may amplify pulmonary inflammation or trigger immune-mediated toxicity. To the best of our knowledge, our case is the first reported in the literature of CIP in a patient receiving durvalumab complicated by HMPV infection. Previous reports have described the coexistence of viral infections and ICI-induced pneumonitis. Several cases of influenza and SARS-CoV-2 infections have been reported during ICI therapy [14,15]. Badran et al. reported a case of cytomegalovirus pneumonia complicating ICI-induced pneumonitis during pembrolizumab therapy [16]. Sumer et al. described herpes simplex virus pneumonitis in a lung cancer patient treated with immunotherapy (nivolumab/ipilimumab) [17]. Similarly, Foukas et al. documented human herpesvirus 6-related interstitial pneumonitis in a patient with CIP associated with nivolumab [18].
Management of irAEs depends on the severity of toxicity and requires a careful, multidisciplinary approach. Mild (grade 1) events can often be monitored without therapy interruption, while moderate to severe (grade 2-4) toxicities typically require temporary discontinuation of the ICI and initiation of corticosteroids [2,19].
Systemic corticosteroids represent the mainstay of treatment for moderate to severe CIP, as well as for severe CAP with an inflammatory or immune-mediated component. If pneumonitis persists or worsens after 48 h, consider initiating a non-steroidal immunosuppressive agent [2,19].
The decision to resume ICI therapy after resolution of an irAE must be individualized, considering the severity of the initial toxicity, the risk of permanent organ damage, and the presence of contributing factors such as intercurrent infections. Clear criteria for rechallenging patients after severe pneumonitis remain lacking [19].
Recent reports indicate that ICI treatment can be safely resumed in selected patients once the adverse event has resolved, provided that close clinical and laboratory follow-up is ensured [19]. For instance, severe hematologic toxicities, including pure red cell aplasia associated with Parvovirus B19 infection during atezolizumab therapy, were successfully managed with resolution of anemia, and therapy was safely resumed [20]. A previous review showed that rechallenging with ICIs after irAEs was safe in a limited number of cases (15 patients), but no data on durvalumab were reported [21]. In the cohort reported by Lim et al., among 49 patients who developed CIP, 13 were rechallenged with durvalumab, and only one experienced a recurrent low-grade pneumonitis, which led to permanent discontinuation of the drug [9].
In our case, durvalumab was permanently discontinued, and methylprednisolone 40 mg twice daily (2 mg/kg/day) was administered for 10 days, followed by a gradual taper over 4 weeks. Although the typical clinical response to corticosteroids occurs within 48-72 h [2,19], our patient exhibited a slower but steady improvement leading to complete clinical resolution. HMPV infection likely contributed to a more severe clinical course and delayed response to corticosteroid therapy, as evidenced by the prolonged need for highflow oxygen therapy and the two-week recovery period before significant improvement, suggesting persistent viral-driven inflammation despite adequate immunosuppression.
Finally, during hospitalization, the patient developed leukopenia (nadir WBC 1050/mm 3 ), a known hematologic adverse event associated with durvalumab [4]. This condition resolved spontaneously with supportive care and corticosteroid therapy, consistent with an iatrogenic mechanism rather than virus-induced marrow suppression.
However, our case has several limitations. Although HMPV was identified in the lower respiratory tract, its exact pathogenic contribution to lung injury in this case remains uncertain. The detection of viral RNA alone does not confirm causality. However, it is plausible that HMPV acted as a trigger or amplifier of the immune-mediated pneumonitis. Furthermore, the diagnosis of CIP was based on a multidisciplinary clinical-radiologic assessment rather than histopathologic confirmation. Although bronchoalveolar lavage (BAL) can support the suspicion of drug-induced lung injury, it was not performed due to the patient's clinical improvement following corticosteroid therapy and the difficulty of performing a BAL in the context of high-flow oxygen therapy. Our patient had a history of heavy cigarette smoking but no documented diagnosis of chronic obstructive pulmonary disease or interstitial lung disease, which are considered risk factors for CIP.
Finally, although follow-up imaging at four months showed substantial radiologic improvement, longer-term imaging and functional assessment would be necessary to fully evaluate the reversibility and lasting impact of the lung injury.
## 4. Conclusions
This case underscores the diagnostic complexity and therapeutic challenges of durvalumab-induced pneumonitis in the presence of viral infection. The concomitant detection of HMPV underscores the importance of comprehensive microbiologic testing, including molecular diagnostics, to account for both infectious and immune-mediated lung injury. The prolonged corticosteroid regimen achieved full recovery, demonstrating the effectiveness of timely immunosuppressive therapy even in the setting of concurrent viral infection. Early recognition, prompt steroid initiation, and multidisciplinary management remain crucial to prevent irreversible lung injury and improve patient outcomes. Informed Consent Statement: Written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement: Data will be made available upon request.
## References
1. Spagnolo, Chaudhuri, Bernardinello et al. (2022) "Pulmonary adverse events following immune checkpoint inhibitors" *Curr. Opin. Pulm. Med*
2. Li, Faiz, Boysen-Osborn et al. (2025) "Immune Checkpoint Inhibitor-associated Pneumonitis: A Narrative Review" *West. J. Emerg. Med*
3. Alwhaibi, Alenazi, Alghadeer et al. (2025) "A Real-World Comparison of the Safety Profile for Immune Checkpoint Inhibitors in Oncology Patients" *J. Clin. Med*
4. Khunger, Rakshit, Pasupuleti et al. (2017) "Incidence of Pneumonitis with Use of Programmed Death 1 and Programmed Death-Ligand 1 Inhibitors in Non-Small Cell Lung Cancer: A Systematic Review and Meta-Analysis of Trials" *Chest*
5. Leclair, Merl, Cohenuram et al. (2022) "Real-World Incidence of Pneumonitis in Patients Receiving Durvalumab" *Clin. Lung Cancer*
6. Keam, Turner, Kugeratski et al. (2024) "Toxicity in the era of immune checkpoint inhibitor therapy" *Front Immunol*
7. Lim, Ghosh, Morrison et al. (2023) "Durvalumab-Associated Pneumonitis in Patients with Locally Advanced Non-Small Cell Lung Cancer: A Real-World Population Study" *Curr. Oncol*
8. Iyer, Deb, Javed et al. "Human metapneumovirus-understanding a growing respiratory threat" *QJM*
9. Samuel, Nanjappa, Cooper et al. (2016) "Human Metapneumovirus Infection in Immunocompromised Patients" *Cancer Control*
10. Pipitò, Mazzola, Bono et al. (2025) "A Case of Severe Respiratory Failure Caused by Metapneumovirus and Influenza Virus in a Patient with HIV Infection" *Viruses*
11. Samajdar, Chatterjee, Mukherjee et al. "Ribavirin and IVIG Therapy for Severe hMPV Pneumonia: A Promising Therapeutic Approach for India" *J. Assoc. Physicians India*
12. Guerini, Borghetti, Filippi et al. (2020) "Differential Diagnosis and Clinical Management of a Case of COVID-19 in a Patient with Stage III Lung Cancer Treated with Radio-chemotherapy and Durvalumab" *Clin. Lung Cancer*
13. Li, Zhang, Feng et al. (2025) "Clinical Characteristics of Influenza Pneumonia in Patients with Lung Adenocarcinoma Receiving Immunotherapy" *Viral Immunol*
14. Badran, Ouryvaev, Baturov et al. (2021) "Cytomegalovirus pneumonia complicating immune checkpoint inhibitors-induced pneumonitis: A case report" *Mol. Clin. Oncol*
15. Sumer, Waldeck, Fischer et al. "HSV-pneumonitis in a patient with lung cancer receiving check point inhibitors-A case report" *Pneumonia 2021*
16. Foukas, Tsiodras, Economopoulou et al. (2018) "Concomitant Human Herpes Virus 6 and nivolumab-related pneumonitis: Potential pathogenetic insights. ID Cases"
17. Brade, Bahig, Bezjak et al. (2024) "Esophagitis and Pneumonitis Related to Concurrent Chemoradiation ± Durvalumab Consolidation in Unresectable Stage III Non-Small-Cell Lung Cancer: Risk Assessment and Management Recommendations Based on a Modified Delphi Process" *Curr. Oncol*
18. Pallotta, Stefanini, Pratelli et al. (2025) "Pure red cell aplasia due to Parvovirus B19 infection and atezolizumab: Case report and literature review" *Immunotherapy*
19. Si, Song, Ni et al. (2020) "Management of immune checkpoint inhibitor-related adverse events: A review of case reports" *Thorac. Cancer*
20. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12497954&blobtype=pdf | # Machine Learning-based Prediction of Active Tuberculosis in People With HIV Using Clinical Data
Lena Bartl, Marius Zeeb, Marisa Kälin, Tom Loosli, Julia Notter, Hansjakob Furrer, Matthias Hoffmann, Hans Hirsch, Robert Zangerle, Katharina Grabmeier-Pfistershammer, Michael Knappik, Alexandra Calmy, Jose Damas Fernandez, Niklaus Labhardt, Enos Bernasconi, Huldrych Günthard, Roger Kouyos, Katharina Kusejko, Johannes Nemeth
## Abstract
Background. Coinfections of Mycobacterium tuberculosis (MTB) and human immunodeficiency virus (HIV) impose a substantial global health burden. Patients with MTB infection face a heightened risk of progression to incident active TB, which preventive therapy can mitigate. Current testing methods often fail to identify individuals who subsequently develop incident active TB.Methods. We developed random forest models to predict incident active TB using patients' medical data at HIV-1 diagnosis. Training our model involved using clinical data routinely collected at enrollment from the Swiss HIV Cohort Study (SHCS). This dataset encompassed 55 people with HIV (PWH) who developed incident active TB 6 months after enrollment and 1432 matched PWH without TB enrolled between 2000 and 2023. External validation used data from the Austrian HIV Cohort Study, comprising 43 people with incident active TB and 1005 people without TB.Results. We predicted incident active TB with an area under the receiver operating characteristic curve of 0.83 (95% CI: .8-.86) in the SHCS. After adjusting for ethnicity and the region of origin and refitting the model with fewer parameters, we obtained comparable receiver operating characteristic curve values of 0.72 (SHCS) and 0.67 (Austrian HIV Cohort Study). Our model outperformed the standard of care (tuberculin skin test and interferon-gamma release assay) in identifying high-risk patients, demonstrated by a lower number needed to diagnose (1.96 vs 4).Conclusions. Models based on machine learning offer considerable promise for improving care for PWH, requiring no additional data collection and incurring minimal additional costs while enhancing the identification of PWH that could benefit from preventive TB treatment.
Human immunodeficiency virus 1 (HIV-1) and Mycobacterium tuberculosis (MTB) infection synergistically cause mortality, with HIV-1 significantly influencing the progression of MTB infection to incident active TB [1]. Although the introduction of antiretroviral therapy (ART) has dramatically reduced the risk of HIV-1-associated active TB, people with HIV-1 (PWH) are still at increased risk of developing active TB. This increase is in particular pronounced in the case of late diagnosis of the HIV-1 infection or restricted access to ART due to factors such as socioeconomic disruption, war, and refugee status [2][3][4]. Current diagnostic TB tests rely on T-cell responses, such as interferon-gamma release assay (IGRA) and tuberculin skin test (TST). However, in PWH, T cells exhibit impaired functionality, both quantitatively and qualitatively, which makes it challenging to predict TB in these high-risk populations [5,6].
This study aimed to devise a prognostic score for predicting the occurrence of "incident active tuberculosis (TB)", defined as active TB that arises at least 6 months after HIV-1 diagnosis. Our model is built on clinical data routinely collected during the initial consultation at HIV-1 diagnosis, with the 6-month window chosen to enable effective preventive treatment to mitigate the onset of active TB [7].
Artificial intelligence (AI) approaches specifically machine learning, have emerged as promising tools in healthcare for detecting novel patterns in already available data [8,9]. In the context of predicting TB in PWH, machine learning techniques offer the potential to identify indicators and risk factors that may not be identified through traditional statistical methods.
A few examples include the approach of identifying incident active TB using clinical parameters [10], early prediction of TB transmission [11] as well as treatment outcome [12]. The field of machine learning was also applied to many other fields in TB and types of data, providing new perspectives and possibilities in healthcare [13][14][15].
However, the successful implementation of AI approaches depends on the availability of high-quality input data [9,16]. The Swiss HIV Cohort Study (SHCS) meticulously records longitudinal clinical and laboratory parameters of PWH in Switzerland, using an elaborate high-quality workflow for semiautomated data entry during the regular biannual followups. This generates a research database consisting of highly accurate and consistent data. Furthermore, as more than 70% of newly diagnosed PWH are enrolled in the study, this results in a large and representative study population [17]. Similarly, the Austrian HIV Cohort Study (AHIVCOS) serves as a valuable external validation cohort [18], offering a dataset structured similarly to the SHCS. In this study, 2 datasets were used to develop an AI-based diagnostic tool for predicting TB in PWH. This novel approach enables the identification of individuals at risk of progressing to active TB within a high-risk population, operating independently of the current gold-standard (IGRA/TST).
## METHODS
## Participants
The study population consists of PWH enrolled in the SHCS, a multicentric Swiss study launched in 1988 with a cumulative number of more than 21 000 participants. The SHCS collects longitudinal information on demographic, clinical, behavioral, and laboratory variables in at least biannual follow-up visits. External validation involved individuals from the AHIVCOS, a multicentric cohort of PWH in Austria [17,18].
## Definitions
The diagnosis of active TB relied on clinical signs, symptoms (eg, coughing), indicative X-rays or other imaging, and subsequent microbiological detection of MTB [17,19]. Active TB diagnoses were categorized as prevalent TB (identified at SHCS enrollment or <6 months after enrollment) and incident active TB (TB diagnosis >6 months after enrollment). Latent MTB infection was defined as a positive TST or IGRA test at least 6 months before the occurrence of incident TB. Sensitivity and specificity of MTB testing was estimated based on SHCS data from 2000 through 2023 [5]. Preventive treatment entailed initiating rifampicin, rifabutin, isoniazid, rifapentine, or pyrazinamide therapy at any point during the SHCS follow-up [5].
## Selection of the Population and Variables
Since the SHCS started collecting in-depth demographic and clinical variables in 2000, all participants registered in the SHCS between 1 January 2000 and 1 November 2023 were included in this study. All people with prevalent TB (ie, TB diagnosed at cohort enrollment) were excluded. Participants preventively treated for MTB infection were also excluded. People with incident active TB were defined as patients with incident TB occurring at least 6 months after enrollment. A comprehensive range of demographic, laboratory, and lifestyle variables was included; variables with missing data for more than 35% of the study population were excluded. All included variables are listed in Supplementary Table 1, variables that did not meet the criteria and were thus excluded are listed in Supplementary Table 2. World Health Organization regions were used to recode regions for simplicity [20]. The data used in the model were obtained during the initial SHCS enrollment consultations, with additional data from clinical consultations conducted 6 months before or after enrollment. The model did not include MTB testing. Any participants with more than 35% missing data for the chosen clinical variables were excluded from our analysis. Matching of people without TB was solely based on the registration year to account for advancements in HIV patient care, and to enhance model robustness, 30 people without TB were selected per person with incident active TB.
## Outcome
The primary outcome was predicting incident TB development. The binary classification was assessed using the area under the receiver operating characteristics (ROC) curve (AUC) and Youden Index. Secondary outcomes included variable importance analysis, the impact on the number needed to diagnose to have a comparison with the current standard of testing, as well as the area under the precision-recall curve (AUPRC) to assess the predictive power for incident active TB, which can give more insight on model precision when using unbalanced data sets [21].
## Model Creation
We constructed a random forest algorithm using the randomForest package in R [22][23][24] (see Supplementary Material for details). The data were partitioned into training (70%) and validation (30%) sets using the caret package in R [25], where the validation set was kept out of the model building to be a valid control dataset to assess model performance. The split was stratified by people with incident active TB and people without TB to ensure a balanced class distribution within the splits based on the outcome. Imputation of missing data was performed separately on the training and validation datasets using again the randomForest package in R. Imputation was conducted iteratively with random seed selection, using 10 iterations. The random forest was trained using the training dataset with internal cross-validation with 5 splits and 100 iterations, each iteration creating a new training and testing dataset, to achieve optimum model performance. Model performance was assessed using a validation set previously unknown to the model. We repeated imputation and model creation 1000 times, each with a distinct randomly chosen seed. This procedure aimed to provide a more accurate estimate of performance by pooling results from different forests and calculating the mean ROC of the AUC and variable importance [26]. Optimal model performance resulted from using seven features in each tree and 500 trees for each forest.
## Statistical Analysis
Performance assessment involved using the pROC package [27] in R to generate ROC curves for each forest, followed by combination and calculation of mean curves. The ROC curve was smoothed using the geomsmooth-function of ggplot2 [28]. Optimal points were determined using the Youden Index (sensitivity + specificity -1) for comparison with the clinical standard (both TST and IGRA during 2000-2023). This was then expressed in a number needed to diagnose (NND), which signifies the number of people with a specific illness that must undergo screening by the test to yield 1 correct diagnosis (ie, true positive and positive predictive value [PPV]). Reported AUC results were obtained using the test set. Furthermore, we used the PRROC package in R to generate the AUPRC curves [29].
## External Validation
The model underwent external validation by training it on SHCS data and validating it with both SHCS and AHIVCOS data in separate analyses. Data preparation steps for the AHIVCOS mirrored those employed for SHCS data. We refitted the model using solely the most important variables, previously identified as the top 20 predictors based on variable importance that were available in the AHIVCOS database. Additionally, we excluded the variables "Ethnicity" and "Region of Origin" for the external validation only because these parameters were distributed differently between the cohorts. We instead looked at the distribution of low-and high-incidence TB countries among the participants. Model performance was assessed using the AUC of the ROC curve and variable importance analysis.
## RESULTS
## Study Population
Among the 21 529 individuals enrolled in the SHCS, those registered before the year 2000, people with prevalent TB, or individuals treated preventively for MTB infection were excluded. This led to the inclusion of 9828 individuals in our study population, with 55 individuals identified as people with incident active TB, 9773 identified as people without TB. After matching (see Methods), the study population comprised 1430 individuals, consisting of 1029 (72.0%) males and 401 (28.0%) females (Figure 1). The majority (1138, 79.6%) were White, with a mean age of 38.2 years.
For external validation of the derived score, AHIVCOS data were used to determine the reproducibility and predictive power of our model in a different study setting [30]. Of the 11 154 patients included in AHIVCOS, 43 people with incident active TB and 1005 people without TB were included in the validation dataset (Figure 1, Table 1). Overall, data from 2533 participants from both cohorts were used, with 98 people with incident active TB.
## Primary Outcome
We developed a random forest model to predict incident active TB in PWH within the SHCS, with 48 predictors (Supplementary Table 1). The mean AUC for the ROC curve predicting incident active TB in the test dataset was 0.83 (95% confidence interval: .8-.86). The sensitivity was 70.1% and the specificity was 81.0%, with a Youden index of 0.51 (Figure 2A). The AUPRC, which looks at the fraction of predicted positives among true positives, was 0.168 at a baseline of 0.03 (Supplementary Figure 1).
We assessed variable importance by mean decrease in accuracy, quantified by removing the association between predictor and outcome variables, and evaluated the error increase (Figure 2B). Demographic parameters, particularly region of origin and ethnicity, significantly influenced model performance due to varying TB incidence rates. Removing information on ethnicity and region of origin even decreased the predictive power to an AUC of 0.63 (Supplementary Figure 2) and adding TB testing (IGRA and TST) did not improve the model (Supplementary Figure 3). Socioeconomic status, represented by profession and education, and mode of transmission also impacted predictions. Laboratory results, including HIV-related metrics such as CD4 cell count and RNA levels, reflecting immune system status, played a major role in TB outbreak timing. Overall health indicators (body mass index [BMI], CD4, hemoglobin, etc.) also affected model accuracy and showed a statistically significant association with predicting incident active TB (Supplementary Figure 4).
We selected the top 20 predictors to simplify model application to external datasets. This reduced set yielded an AUC of 0.74, making it feasible for use with SHCS data and external validation (Supplementary Figure 5).
## External Validation
For the external validation, we have adapted the ethnicity variable included in the model, as the 2 cohorts differed considerably in this regard (Table 1). Northern America and both Western and Northern Europe were categorized as regions of "low incidence," whereas the rest of the world was considered as "high incidence." We thereby accounted for this disparity, which likely stems from differential immigration patterns, with people with incident active TB in Austria predominantly from Eastern Europe and in Switzerland predominantly from Africa. A comparable trend was observed for the region of origin (Table 1). Reapplying the model to SHCS data using the updated variables and validating on combined SHCS and AHIVCOS data resulted in AUCs for SHCS validation of 0.72, with a Youden Index of 0.33, sensitivity of 74.2%, and specificity of 59.1%. The AUC for AHIVCOS data was 0.67, with a Youden Index of 0.28, sensitivity of 71.6%, and specificity of 55.9% (Figures 3A and3B). Of note, applying the model trained using the original ethnicity variable proved ineffective for the AHIVCOS with an AUC of 0.5 in the validation [18] (Supplementary Figure 6). For the externally validated score, we observed significant influences on predicting future incident active TB, of immune system parameters and variables indicative of patients' well-being at registration, such as BMI or creatinine levels (Supplementary Figure 7). For this analysis, we also looked at the AUPRC for SHCS data (0.111) and the AHIVCOS data (0.095) at a baseline of 0.03 (Supplementary Figure 8A and8B).
## Time to TB Analysis
We examined the time from SHCS registration to incident active TB and its impact on the model's predictive power. We selected a 4-year cutoff as this was the average time for people with incident active TB to develop TB (Supplementary Figure 9A and9B). The model performed better for shorter time to incident active TB. This likely occurs because biological changes, observable in laboratory and blood results, emerge closer to the outbreak.
## Comparison With the Current Gold Standard
In the SHCS from 2000-2023, both TST and IGRA were used for MTB infection testing. The sensitivity and specificity of MTB infection testing for predicting incident active TB were 30% and 94%, respectively [5]. MTB infection testing yielded a NND of 4, whereas our newly derived method based on the complete SHCS model required only 1.96 people with incident active TB to correctly identify 1, doubling the likelihood of diagnosing a person with incident active TB correctly. External validation also demonstrated a better result than MTB infection testing, with a NND of 3.6. PPV across all models was also better than MTB infection testing, with the complete model having a PPV of 10.3%, the external validation of 4.9%, and the testing standard of IGRA and TST of 2.7% and 1.5%, respectively [31] (Table 2).
## DISCUSSION
Conventional testing methods often lack efficacy in identifying individuals who subsequently develop incident active TB, particularly among PWH [6,32]. In response to this challenge, we developed a machine learning that uses clinical data from PWH at HIV-1 diagnosis to predict incident active TB.
The model performed at least as well as the clinical standard in predicting incident active TB in the SHCS. This performance was subsequently validated in the AHIVCOS, an independent external validation cohort, confirming the robustness of our model. Our approach offers a surprisingly simple but compelling example of how AI can be used for predicting relatively rare events, such as the onset of incident active TB in PWH. Our machine learning approach functions independently of T-cell activity, effectively overcoming a key limitation of IGRA-based prediction systems that depend heavily on T-cell responses. This reliance is particularly problematic in PWH, where both reduced T-cell quantity and quality frequently result in false-negative outcomes due to impaired T-cell responsiveness [5,33]. While previous attempts to predict incident active TB using clinical parameters (with and without IGRA) have been made, these efforts were limited in scale and duration compared to our project [34][35][36].
The inclusion of factors such as CD4 count and HIV-1 viral load in our diagnostic score, recognized risk factors for TB development, strengthens the validity of our findings [37]. Moreover, the identification of novel metabolism-associated factors is notable. For instance, the predictive value of low creatinine levels suggests compromised muscle mass and nutritional status. Similarly, the influence of high-density lipoprotein and triglycerides underscores metabolic perturbations in individuals at high risk of active incident TB. Poor nutrition has long been acknowledged as a TB risk factor [38], as evidenced by recent large-scale trials in India [39] and by the World Health Organization [40].
Using existing data offers cost savings, as IGRA tests are expensive, and TST requires multiple clinical visits. Additionally, our algorithm can be seamlessly integrated into clinical information systems, automatically using routine visit data to provide individualized TB risk assessments during subsequent appointments without adding to the physician's workload.
Overall, our findings imply that our score partially assesses intrinsic susceptibility to incident active TB. Hence, it would be compelling to evaluate the score in PWH residing in high TB transmission settings. If our score indeed reflects intrinsic susceptibility, it should effectively identify PWH at the highest risk of TB, even in high-transmission settings. For example, in Peru, BMI serves as a crucial tool for identifying household contacts at elevated risk for active TB, supporting this hypothesis in principle [41].
By employing the Youden cutoff, we established a sensitivity threshold for our ROC curves, enabling direct comparison with established testing methods. Implementing the automated Youden cutoff prevents arbitrary threshold selection that could artificially inflate algorithmic performance. While the Youden cutoff may occasionally yield suboptimal outcomes, prospective clinical testing will allow for refinement and adaptation to specific clinical needs, ensuring optimal performance in realworld settings. In future clinical applications, adjustments to the cutoff can align with specific clinical requirements, emphasizing either high sensitivity or high specificity.
Our study has limitations. The dataset is derived from 2 Central European countries with low MTB transmission rates, potentially introducing selection bias. Moreover, the limited geographical representation may restrict the generalizability of the developed model to more diverse settings. Despite including a total of 98 people with incident active TB and 2435 people without TB, unknown biases may exist that could influence our findings. Incident active TB is typically infrequent, even among PWH [5]. Althouugh we excluded individuals who received preventive therapy, potential biases include the tuberculostatic effects of non-TB drugs such as sulfamethoxazole/trimethoprim prophylaxis, which could impact incident active TB development in PWH [42]. Additionally, people without TB with fluoroquinolone use, which may affect TB development, were not excluded [43].
## CONCLUSIONS
Summarized, the machine learning algorithm introduced in this study has the potential to diagnose PWH at high risk of
## References
1. Goletti, Weissman, Jackson (1996) "Effect of Mycobacterium tuberculosis on HIV replication. Role of immune activation" *J Immunol*
2. Jiamsakul, Lee, Van Nguyen (2018) "Socio-economic statuses and risk of tuberculosis-a case-control study of HIV-infected patients in Asia" *Int J Tuberc Lung Dis*
3. Wood, Maartens, Lombard (2000) "Risk factors for developing tuberculosis in HIV-1-infected adults from communities with a low or very high incidence of tuberculosis" *J Acquir Immune Defic Syndr*
4. Bruchfeld, Correia-Neves, Källenius (2015) "Tuberculosis and HIV coinfection" *Cold Spring Harb Perspect Med*
5. Zeeb, Tepekule, Kusejko (2023) "Understanding the decline of incident, active tuberculosis in people with HIV in Switzerland" *Clin Infect Dis*
6. Bell, Noursadeghi (2018) "Pathogenesis of HIV-1 and Mycobacterium tuberculosis co-infection" *Nat Rev Microbiol*
7. (2025) "Latent tuberculosis infection: updated and consolidated guidelines for programmatic management"
8. Ngiam, Khor (2019) "Big data and machine learning algorithms for health-care delivery" *Lancet Oncol*
9. Scott (2021) "Demystifying machine learning: a primer for physicians" *Intern Med J*
10. Du, Su, Qiao (2024) "Application of artificial intelligence in diagnosis of pulmonary tuberculosis" *Chin Med J (Engl)*
11. Gharamaleki, Colijn, Sekirov et al. (2024) "Early prediction of Mycobacterium tuberculosis transmission clusters using supervised learning models" *Sci Rep*
12. Kim, Kwak, Yoon (2024) "Artificial intelligence-based radiographic extent analysis to predict tuberculosis treatment outcomes: a multicenter cohort study" *Sci Rep*
13. Li, Yang, Zhuang et al. (2023) "From immunology to artificial intelligence: revolutionizing latent tuberculosis infection diagnosis with machine learning" *Mil Med Res*
14. Althomsons, Winglee, Heilig (2022) "Using machine learning techniques and national tuberculosis surveillance data to predict excess growth in genotyped tuberculosis clusters" *Am J Epidemiol*
15. Rashidi, Khan, Dang (2022) "Prediction of tuberculosis using an automated machine learning platform for models trained on synthetic data" *J Pathol Inform*
16. Rajkomar, Dean, Kohane (2019) "Machine learning in medicine" *N Engl J Med*
17. Scherrer, Traytel, Braun (2022) "Cohort profile update: the Swiss HIV cohort study (SHCS)" *Int J Epidemiol*
18. Leierer, Rappold, Strickner et al. (2023) "45th report of the Austrian HIV cohort study"
19. (2025) "CDC category C diagnoses-Swiss HIV cohort study"
20. (2025) "Countries overview. World Health Organization"
21. Saito, Rehmsmeier (2015) "The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets" *PLoS One*
22. Liaw, Wiener (2002) "Classification and regression by random forest" *R News*
23. Breiman (2001) "Random forests" *Mach Learn*
24. Breiman, Cutler, Liaw et al. (2024) "Breiman and Cutlers random forests for classification and regression"
25. Kuhn (2008) "Building predictive models in R using the caret package" *J Stat Softw*
26. Henderson (2005) "The bootstrap: a technique for data-driven statistics. Using computer-intensive analyses to explore experimental data" *Clin Chim Acta*
27. Robin, Turck, Hainard (2011) "pROC: an open-source package for R and S+ to analyze and compare ROC curves" *BMC Bioinformatics*
28. Wilkinson (2011) "Ggplot2: elegant graphics for data analysis by Wickham H" *Biometrics*
29. Grau, Grosse, Keilwagen (2015) "PRROC: computing and visualizing precision-recall and receiver operating characteristic curves in R" *Bioinformatics*
30. Ramspek, Jager, Dekker et al. (2021) "External validation of prognostic models: what, why, how, when and where?" *Clin Kidney J*
31. Diel, Loddenkemper, Nienhaus (2012) "Predictive value of interferon-γ release assays and tuberculin skin testing for progression from latent TB infection to disease state: a meta-analysis" *Chest*
32. Goletti, Delogu, Matteelli et al. (2022) "The role of IGRA in the diagnosis of tuberculosis infection, differentiating from active tuberculosis, and decision making for initiating treatment or preventive therapy of tuberculosis infection" *Int J Infect Dis*
33. Lalvani, Pareek (2010) "Interferon gamma release assays: principles and practice" *Enferm Infecc Microbiol Clin*
34. Lee, Lin, Tsai (2015) "A clinical algorithm to identify HIV patients at high risk for incident active tuberculosis: a prospective 5-year cohort study" *PLoS One*
35. Njagi, Nduba, Mureithi et al. (2023) "Prevalence and predictors of tuberculosis infection among people living with HIV in a high tuberculosis burden context" *BMJ Open Respir Res*
36. Mendelsohn, Fiore-Gartland, Awany (2022) "Clinical predictors of pulmonary tuberculosis among South African adults with HIV" *EClinicalMedicine*
37. Orcau, Caylà, Martínez (2011) "Present epidemiology of tuberculosis. Prevention and control programs" *Enferm Infecc Microbiol Clin*
38. Ockenga, Fuhse, Chatterjee (2023) "Tuberculosis and malnutrition: the European perspective" *Clin Nutr*
39. Bhargava, Bhargava, Meher (2023) "Nutritional supplementation to prevent tuberculosis incidence in household contacts of patients with pulmonary tuberculosis in India (RATIONS): a field-based, open-label, cluster-randomised, controlled trial" *Lancet*
40. Dryburgh, Rippin, Malykh (2024) "Tuberculosis and malnutrition"
41. Li, Nordio, Huang (2020) "Two clinical prediction tools to improve tuberculosis contact investigation" *Clin Infect Dis*
42. Hasse, Walker, Fehr (2014) "Co-trimoxazole prophylaxis is associated with reduced risk of incident tuberculosis in participants in the Swiss HIV Cohort Study" *Antimicrob Agents Chemother*
43. Malik, Fuad, Siddiqui (2020) "Tuberculosis preventive therapy for individuals exposed to drug-resistant tuberculosis: feasibility and safety of a communitybased delivery of fluoroquinolone-containing preventive regimen" *Clin Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724187&blobtype=pdf | # Multiple rotavirus species encode fusion-associated small transmembrane (FAST) proteins with cell type-specific activity
Kylie Sartalamacchia, Vanesa Veletanlic, Julia Diller, Kristen Ogden, Thomas Frist
## Abstract
Fusion-associated small transmembrane (FAST) proteins are viral nonstruc tural proteins that mediate cell-cell fusion to form multinucleated syncytia. We previously reported that human species B rotavirus NSP1-1 is a FAST protein that induces syncytia in primate epithelial cells but not rodent fibroblasts. We hypothesized that NSP1-1 proteins of other rotavirus species could mediate cell-cell fusion but might be limited to cell types derived from homologous hosts. To test this hypothesis, we predicted the structure and domain organization of NSP1-1 of species B rotavirus from a human, goat, and pig, species G rotavirus from a pigeon and turkey, and species I rotavirus from a dog and cat. We transiently expressed the NSP1-1 proteins in avian, canine, hamster, human, porcine, and simian cells. Regardless of the host origin of the virus, each NSP1-1 protein induced syncytia in primate cells, while few induced syncytia in other cell types. In some cell types, a lack of protein expression or RNA stability failed to explain the inability of NSP1-1 to mediate cell-cell fusion. To identify the domains that determine cell-specific fusion activity for human species B rotavirus NSP1-1, we engineered chimeric proteins containing domain exchanges with the p10 FAST protein from Nelson Bay orthoreovirus and found that the N-terminal and transmembrane domains were determinants. Together, these findings suggest that rotavirus species B, G, and I NSP1-1 are functional FAST proteins whose N termini contribute to specifying the cells in which they mediate syncytium formation. IMPORTANCE Mechanisms of membrane fusion and determinants of host range for pathogens remain poorly understood. Improved understanding of these concepts could open new areas for therapeutic development and shed light on virus epidemiology. Our analyses of NSP1-1 proteins from species B, G, and I rotaviruses provide insights into the variability tolerated by functional FAST proteins. Further, the observation that all putative FAST proteins tested can induce syncytium formation in at least some cell types provides evidence that rotaviruses that encode NSP1-1 proteins are fusogenic viruses. Finally, although the criteria for their specificity remain unclear, our observations regarding fusion capacities of different NSP1-1 proteins and of chimeric FAST proteins suggest a potential role for rotavirus FAST proteins in determining the efficiency of viral replication within a given host or cell type. KEYWORDS dsRNA virus, cell fusion, FAST protein, syncytia, rotavirus D eterminants of tropism and pathogenesis are incompletely understood for many viruses, including rotavirus, an important cause of diarrheal disease (1). Rotavirus is a member of the order Reovirales, which contains viruses with segmented double-stran ded (ds) RNA genomes (2, 3). Rotaviruses are classified into nine species, rotavirus species A (RVA) through RVD and RVF through RVJ, which can be further resolved into two major clades (https://talk.ictvonline.org/) (4-6). RVA, RVC, RVD, and RVF form clade 1, while RVB and RVG through RVJ form clade 2. Rotaviruses exhibit limited host range
and cell tropism, but aside from receptor-binding specificity, the molecular bases of these restrictions are unknown, particularly for rotavirus species other than RVA (7,8). Most human rotavirus diarrheal disease is caused by RVA and affects infants and young children (2). RVA also causes diarrheal disease in other mammals and in birds. In many cases, specific RVA genotypes are associated with infection of a given animal host, although there is evidence of interspecies transmission and reassortment events (9)(10)(11)(12)(13)(14)(15). RVB, RVC, RVH, and RVI have been detected in domesticated mammals, whereas RVD, RVF, and RVG have been detected only in birds (16,17). RVJ has been detected in bats (18). While RVB is more commonly detected in diarrheic pigs (19)(20)(21), it has been associated with sporadic outbreaks of diarrheal disease in humans (22)(23)(24)(25). Although the symptoms of RVB gastroenteritis resemble those of RVA, RVB more often causes disease in adults than in infants and young children (26,27). Sequence analyses suggest the RVBs affecting humans are distinct from those affecting other animals (28,29); thus, sources of RVB epidemics and reasons these viruses primarily cause disease in adults are unknown. In many cases, it remains unclear why some rotavirus species or strains cause disease in a limited range of hosts or in hosts of specific ages.
In contrast to the other 10 segments of its dsRNA genome, the predicted gene organization and functions of encoded proteins for the NSP1 segment differ between the two rotavirus clades (30)(31)(32)(33). For RVA, the NSP1 segment encodes well-characterized innate immune antagonist protein NSP1. RVA NSP1 clusters phylogenetically according to host species, which suggests a potential role in host range restriction (34,35). For RVB, RVG, and RVI, the NSP1 segment contains two overlapping open reading frames (ORFs) whose encoded products have little predicted homology with known proteins (36). Both encoded proteins are highly divergent among rotavirus species (~15%-40% identity) and more conserved within a given species (~40%-100% identity) (unpublished observations). The smaller ORF encodes NSP1-1, which is about 100 amino acids long (36,37). We previously published evidence indicating that human RVB (HuRVB) NSP1-1 is a fusion-associated small transmembrane (FAST) protein, and we hypothesized that it may play a role in cell tropism (33). Based on sequence analysis, we predicted that RVG and RVI NSP1-1 also may be FAST proteins, but their function has not been directly tested.
Viral FAST proteins are small (~90-200 amino acids), plasma membrane-spanning proteins that mediate cell-cell fusion at neutral pH and without a specific trigger, resulting in the formation of multinucleated syncytia (reviewed in references [38][39][40]. Unlike the fusion proteins of enveloped viruses, FAST proteins are nonstructural proteins expressed during infection. FAST proteins have been identified not only in HuRVB, but in the genomes of several orthoreoviruses and aquareoviruses (reoviruses), which are also members of the order Reovirales, and in the genomes of some avian deltacorona viruses (3,33,(39)(40)(41). Most current knowledge of FAST protein domain organization and function comes from studies of reovirus FAST proteins, and it is unclear whether rotavirus FAST proteins differ in features or mechanism. Reovirus FAST proteins have little sequence similarity, but each is composed of a short N-terminal ectodomain, a central transmembrane (TM) domain, and a longer C-terminal endodomain (39,40). Reovirus FAST proteins are acylated, often at the N terminus but sometimes just after the TM domain, and the endodomain contains a juxtamembrane polybasic region and a predicted amphipathic helix. Often, the three FAST protein domains can be functionally interchanged (42)(43)(44)(45). Reovirus FAST proteins form multimeric complexes at the plasma membrane (44,46). They interact with the lipid bilayer of closely apposed cells through hydrophobic residues and/or a fatty acid modification in the N terminus (47)(48)(49)(50)(51)(52). These interactions are proposed to favor lipid mixing, creating a state that can favor progres sion to the fusion pore (39,40). The amphipathic helix in the reovirus FAST protein endodomain is thought to partition into the curved membrane of the pore on the inner leaflet and stabilize it. Then, cellular proteins promote pore expansion and syncytium formation (53)(54)(55).
It is possible that FAST proteins contribute to cell type tropism. While reovirus FAST proteins are not thought to bind specific host cell receptors, host molecules that interact with the C-terminal endodomain and differ among FAST proteins have been identified in some cases (53)(54)(55). HuRVB NSP1-1 expression results in syncytium formation in primate (human or African green monkey) epithelial cells but not in rodent (hamster or mouse) fibroblasts (33). This observation suggests the possibility of host-specific or cell type-specific interactions with NSP1-1 that mediate cell-cell fusion. FAST protein expression enhances viral replication in cultured cells (33,56,57) and could conceivably contribute to rotavirus replication efficiency in a specific host or tissue during natural infection. In the current study, we sought to predict structural and functional features of NSP1-1 proteins from species B, G, and I rotaviruses derived from different host animals, determine whether they are FAST proteins and can mediate cell-cell fusion efficiently in cell lines derived from different tissues or animal hosts, and identify the protein domains that dictate cell type-specific fusion activity. Results of these studies provide insights into the diversity of features of FAST proteins and identify FAST protein domains that influence cell type-specific activity.
## RESULTS
## Rotavirus NSP1-1 proteins are predicted to share features
Using a variety of algorithms (58)(59)(60)(61)(62)(63), we aligned and predicted sequence and structural motifs in RVB, RVG, and RVI NSP1-1. In addition to HuRVB NSP1-1, we included NSP1-1 from pig (porcine; Po) and goat (caprine; Cp) RVB, which cause diarrhea in their hosts (64,65). We also included NSP1-1 sequences from pigeon (avian; Av) RVG and turkey (gallinaceous; Ga) RVG (66). RVG has been rarely associated with runting and stunting syndrome in chickens and turkeys (17,67). Finally, we included NSP1-1 sequences from canine (Ca) RVI, which was sequenced from sheltered dogs, and feline (Fe) RVI, which was sequenced from a diarrheic cat (68,69). An N-myristoylation site was predicted at amino acids two through seven for every complete RVB, RVG, and RVI NSP1-1 sequence in GenBank (Fig. 1A andB) (33; data not shown). TM helices were identified in RVB, RVG, and RVI NSP1-1 sequences, with the N terminus predicted to be extracellular and the C terminus cytoplasmic. Each NSP1-1 contains multiple basic residues C-terminal to the predicted TM domain. For the analyzed RVB, RVG, and RVI NSP1-1 sequences, residues preceding and following the TM domain were predicted to form helices. For PoRVB and CpRVB NSP1-1, the endodomain helix is predicted to be amphipathic (Fig. 1C). These motifs suggest a model of RVB NSP1-1 in which a myristoylated extracellular N-terminal ectodomain, which may interact with lipids, precedes a TM domain and a cytoplasmic endodomain containing a polybasic region that, at least in some cases, is in an amphipathic helix and may interact with the plasma membrane inner leaflet (Fig. 1B). A similar topology is predicted for RVG and RVI NSP1-1, although the endodomains are smaller, and amphipathic helices were not readily modeled.
## RVB, RVG, and RVI NSP1-1 mediate syncytium formation in human cells
To test the hypothesis that all rotavirus NSP1-1 proteins are FAST proteins that mediate cell-cell fusion, we transfected human embryonic kidney 293T cells with vector alone or plasmids encoding HuRVB, PoRVB, CpRVB, AvRVG, GaRVG, CaRVI, or FeRVI NSP1-1 (37, 64-66, 68, 69). Then, we examined the appearance of the cell monolayer and the organization of nuclei and F-actin. F-actin tends to cluster near the cell periphery but forms a network throughout the cell (70), and syncytia contain multiple nuclei. While vector-transfected cells were indistinguishable from mock-transfected cells, transfection with any NSP1-1 expression plasmid changed morphology from distinct cells to a monolayer pockmarked by smooth round or oval-shaped syncytia lacking defined cell edges (Fig. 2A). Multiple nuclei were often clustered near the center or at the edges of these syncytia, and while actin was detectable in these areas, actin staining was noticeably lighter and had a different pattern. Combined with their predicted domain features, these observations suggest that RVB, RVG, and RVI NSP1-1 from rotaviruses that infect different animals are FAST proteins that can mediate cell-cell fusion. Average diameters of syncytia formed by GaRVG, CaRVI, and FeRVI were smaller than those formed by HuRVB NSP1-1 (Fig. 2B), and they were detected less frequently, which might suggest differences in the nature or efficiency of interactions in human cells.
## Some rotavirus NSP1-1 proteins with a C-terminal peptide tag are fusion active
To enable detection of HuRVB NSP1-1, we previously engineered a FLAG peptide at the N or C terminus and found that N-terminally tagged FLAG-NSP1-1 was expressed in individual 293T cells with distinct edges, while C-terminally tagged NSP1-1-FLAG was expressed in syncytia (Fig. 3A) (33). These findings are consistent with disruption of a myristoyl moiety on the N terminus of HuRVB NSP1-1 by addition of the FLAG peptide (Fig. 1A andB) and suggested that a free C terminus was not necessary for fusion activity. Since all NSP1-1 proteins are predicted to be N-terminally myristoylated, it is likely that addition of a FLAG peptide would ablate cell-cell fusion activity for all the FAST proteins, as it did for HuRVB NSP1-1. To enable detection and determine the requirement for a free C terminus for NSP1-1 proteins other than HuRVB NSP1-1, we engineered a FLAG peptide at the C terminus for our panel of RVB, RVG, or RVI NSP1-1 proteins. While FLAG-tagged RVI NSP1-1 proteins were undetectably expressed or mislocalized, all FLAG-tagged RVB and RVG NSP1-1 proteins were detected and mediated cell-cell fusion in 293T cells (Fig. 3A). In these confocal images, NSP1-1 colocalizes with regions of the cell monolayer that have an altered actin staining pattern and contain multiple nuclei; these are syncytia. For each fusion-active construct, there was a trend toward smaller syncytium diameter induction for the FLAG-tagged compared with the untagged form of the protein, with statistical significance for CpRVB, PoRVB, and GaRVG NSP1-1 (Fig. 3B). These findings suggest potential differences in functional interactions of the NSP1-1 endodomain among rotavirus species and strains, with a critical role for this domain for RVI NSP1-1 and a contributing but non-critical role in fusion for RVB and RVG NSP1-1.
## Rotavirus NSP1-1 proteins exhibit species-specific cell fusion activity
Our prior observation that HuRVB NSP1-1 can induce syncytium formation in primate cells but not in rodent cells raised the possibility that this protein contributes to determining viral tropism, promoting virus spread between cells of homologous human or simian but not heterologous rodent hosts (33). To test the hypothesis that NSP1-1 functions in a species-specific manner, we obtained cell lines derived from host animals that are homologous, derived from the same animal species, or heterologous, derived from different animal species than the rotavirus from which NSP1-1 sequences were cloned. We transfected them with NSP1-1 expression plasmids and looked for the presence of syncytia. We chose cell lines that are transfectable, albeit with varying efficiency. Since the NSP1-1 proteins form syncytia in human embryonic kidney epithelial cells (Fig. 2), in many cases, we also used cells of kidney or epithelial cell origin. We transfected cells with the pCAGGS vector alone as a negative control, and we transfected cells with pCAGGS expressing NBV p10 from the Miyazaki-Bali (MB) strain (56,71,72) as a positive control, since it induces syncytium formation in at least some cell types that HuRVB NSP1-1 does not (33). We used pCAGGS expressing GFP as a proxy for transfection efficiency. In Cos7 cells, an African green monkey kidney fibroblast-like cell line, transfection with each of the RVB, RVG, and RVI NSP1-1 expression plasmids resulted in readily visible syncytia in the cell monolayer, regardless of the animal source of the viral protein (Fig. 4). Even CaRVI and FeRVI NSP1-1 proteins, which induced small syncytia in 293T cells (Fig. 2), appeared to induce syncytia efficiently in Cos7 cells (Fig. 4). In baby hamster kidney BHK cells, which are heterologous with all viruses from which our NSP1-1 sequences were derived, only NBV p10 and AvRVG NSP1-1 proteins induced readily detectable syncytia (Fig. 5A). In chicken embryo fibroblast DF-1 cells, syncytia were detected only following transfection with pCAGGS expressing NBV MB p10, not with any tested NSP1-1 protein, even those derived from pigeon or turkey RVG (Fig. 5B). Finally, in porcine kidney epithelial PK1 cells and canine kidney fibroblast-like MDCK cells, no differences in cell morphology relative to vector-transfected monolayers were detected for cells transfected with FAST protein-expressing plasmids, even for a porcine NSP1-1 protein in a porcine cell line or a canine NSP1-1 in a canine cell line (Fig. S1).
Transfection efficiency was quite low in MDCK cells, but we failed to detect syncytia even for NBV MB p10. Together, these observations suggest that NSP1-1 proteins induce syncytium formation in a limited range of cell types, and the functional range does not strictly correlate with the animal or tissue origin of a given cell. It is possible that a lack or reduction of cell-cell fusion activity results from differences in the expression of NSP1-1 proteins or in the stability of NSP1-1-encoding RNAs. To address the former concern, we compared the expression of FLAG-tagged NSP1-1 in BHK, DF-1, PK1, and MDCK cells. We used only RVB and RVG NSP1-1 for these experi ments because expression and/or syncytium formation of FLAG-tagged RVI NSP1-1 was undetectable in 293T cells (Fig. 3). All RVB and RVG NSP1-1-FLAG proteins were expressed efficiently in BHK cells (Fig. S2A). PoRVB, AvRVG, and GaRVG NSP1-1-FLAG were expressed similarly in DF-1 cells, with higher expression for CpRVB NSP1-1-FLAG and poor expression of HuRVB NSP1-1-FLAG (Fig. S2B). RVB and RVG NSP1-1-FLAG were expressed inefficiently in PK1 and MDCK cells, a finding that was anticipated for MDCK and somewhat less so for PK1 cells based on GFP expression levels (Fig. S1 andS3). Thus, inefficient NSP1-1 protein expression might explain the lack of cell-cell fusion for all constructs in PK1 and MDCK cells and for HuRVB NSP1-1 in DF-1 cells (Fig. S1 andS3). However, it is unlikely to explain the absence of NSP1-1-mediated fusion activity in BHK cells and for RVG NSP1-1 proteins in DF-1 cells (Fig. 5; Fig. S2).
To address the possibility that the stability of NSP1-1-encoding RNAs differs among cell types, we engineered a set of bicistronic constructs in pCAGGS that include an NSP1-1 ORF followed by an encephalomyocarditis virus IRES, then an mEGFP ORF. In 293T cells, mEGFP was detectably expressed from all bicistronic constructs (Fig. S4). HuRVB, PoRVB, AvRVG, and GaRVG NSP1-1 translated from the bicistronic RNAs successfully formed syncytia. For unknown reasons, CaRVI and FeRVI NSP1-1 did not detectably form syncytia. We next sought to use mEGFP as a readout for transfection efficiency and RNA stability. We reasoned that if mEGFP was expressed, it would indicate that the RNA that also encoded NSP1-1 had not been degraded. In hamster kidney epithelial BHK cells, all bicistronic constructs appeared to have a transfection efficiency comparable with that of a monocistronic pCAGGS GFP, with mEGFP success fully transcribed and translated (Fig. S5). Interestingly, GaRVG NSP1-1, rather than AvRVG NSP1-1, showed evidence of small syncytia. mEGFP brightness was somewhat varia ble in chicken embryo fibroblast DF-1 cells, but the most striking observation was a complete lack of mEGFP expression in cells transfected with HuRVB NSP1-1 bicistronic plasmids, suggesting RNA instability, and there were relatively fewer mEGFP-positive PoRVB NSP1-1 bicistronic plasmid-transfected cells than RVG-transfected or RVI-transfec ted cells (Fig. S6). Consistent with FLAG-tagged protein expression (Fig. S2), none of the constructs induced cell-cell fusion. In porcine kidney epithelial PK1 cells, detection of mEGFP was somewhat lower for the RVI bicistronic constructs than for the others, suggesting the possibility of RNA instability (Fig. S7). AvRVG NSP1-1 exhibited some evidence of cell-cell fusion in PK1 cells, but putative syncytia were quite small. For all bicistronic NSP1-1 constructs, the number of canine fibroblast MDCK cells expressing mEGFP appeared similar to those expressing GFP from a monocistronic pCAGGS plasmid, albeit a small number (Fig. S8). As observed for monocistronic NSP1-1 constructs, none of the tested bicistronic rotavirus NSP1-1 constructs induced detectable cell-cell fusion in MDCK cells. Most RNAs containing NSP1-1 ORFs appeared to be stable in the tested cell lines. However, in DF-1 cells, RNAs containing RVB NSP1-1 ORFs are likely unstable, and in PK1 cells, RNAs containing RVI NSP1-1 ORFs might be somewhat unstable; RNA instability likely results in inefficient NSP1-1 protein expression in these cells (Fig. 5; Fig. S1 to S3). However, RNA instability is unlikely to explain the absence of NSP1-1-mediated fusion activity in BHK or MDCK cells, for RVG NSP1-1 proteins in DF-1 cells, or for RVB and RVG NSP1-1 in PK1 cells. The N terminus can confer FAST protein species specificity
We previously determined that HuRVB NSP1-1 retains fusion activity in human (293T) and simian (MA104) cells but not in rodent (hamster BHK and murine L929) cells (33). However, NBV MB p10 induces syncytia when expressed in each of these cell lines. To learn more about FAST protein domain function, we aligned sequences of HuRVB NSP1-1 and NBV MB p10 (Fig. 6A). We engineered chimeric constructs based on the alignments, in which we exchanged N-terminal, TM, and C-terminal domains between the two FAST proteins (Fig. 6B). NBV MB p10 is palmitoylated at a membrane proximal dicysteine motif C-terminal to the TM domain (73); we preserved this motif in our chimeric constructs (Fig. 6A andB). We anticipated that all properly folded constructs should induce syncytium formation in 293T cells since neither parent protein exhibited restricted fusion activity in this cell line. Despite high transfection efficiency, only chimeric proteins in which the C-termini had been exchanged induced detectable syncytium formation in 293T cells (Fig. 6C). This finding suggests that these chimeric proteins are properly folded and post-translationally modified, whereas those with N-terminal or TM domain exchanges are not.
To identify the FAST protein domain that confers species or cell-type specificity, we transfected BHK cells with plasmids encoding parental HuRVB NSP1-1 and NBV MB p10 proteins or with the chimeric C-terminally exchanged proteins. As previously observed (33), RVB NSP1-1 failed to induce syncytium formation in BHK cells, even when transfection efficiency was quite high (Fig. 7). Only when the N-terminal and TM domains of NBV p10 were present did we detect syncytia in the BHK monolayer. The chimeric protein containing the N-terminal ectodomain and TM domain of RVB NSP1-1 and the C-terminal endodomain of NBV p10 induced no detectable change in monolayer appearance. Although the C-terminal endodomain is thought to interact with host proteins to mediate pore formation for some reovirus FAST proteins (53-55), our findings suggest that the N terminus can contribute to FAST protein species-specific activity.
## DISCUSSION
In the current study, we sought to test the hypothesis that NSP1-1 proteins from different rotavirus species are functional FAST proteins. The capacity of all RVB, RVG, and RVI NSP1-1 proteins tested to induce syncytium formation in primate cells suggests that, like orthoreovirus and aquareovirus FAST proteins, they can mediate cell-cell fusion (39, 40) (Fig. 2 and4). Similar to reovirus FAST proteins, RVB, RVG, and RVI NSP1-1 are predicted to be acylated and to contain an N-terminal ectodomain, a central TM domain, and a C-terminal endodomain (39,40). However, while reovirus proteins are reported to have endodomains that are equal in size or substantially longer than the ectodomains (39), for RVG and RVI NSP1-1, the predicted endodomains are shorter than the ectodomains (Fig. 1). While all reovirus FAST proteins described to date are predicted to contain amphi pathic alpha helices in the endodomain that splay apart lipid headgroups, lowering the energy barrier to pore formation (39,40), only a subset of the NSP1-1 proteins we analyzed is predicted to contain such motifs (Fig. 1). Thus, some NSP1-1 proteins may employ a distinct mechanism to stabilize pore formation, for example, interaction with a cellular protein that contains amphipathic regions and partitions into the inner leaflet of the bilayer. However, this finding also may result from limitations in the ability to accurately predict amphipathic helices. Together, these observations suggest that RVB, RVG, and RVI NSP1-1 are functional FAST proteins and that FAST proteins of rotaviruses and reoviruses are likely to employ largely similar mechanisms to induce cell-cell fusion.
We also sought to test the hypothesis that rotavirus NSP1-1 exhibits host speciesspecific or cell type-specific activity. This hypothesis was based on the observation that HuRVB NSP1-1 mediated cell-cell fusion in primate epithelial cells but not rodent fibroblasts (33). We hypothesized that if NSP1-1 is a host range determinant, then NSP1-1 proteins would mediate syncytium formation efficiently in homologous cells, and they would mediate syncytium formation inefficiently in heterologous cells. Alternatively, if only the tissue type from which a cell line was derived mattered, then NSP1-1 might mediate syncytium formation efficiently in epithelial-derived but not fibroblast-derived cell lines. Unexpectedly, we found that NSP1-1 proteins from RVB, RVG, and RVI from different host animals all mediated detectable cell-cell fusion upon expression in primate epithelial (293T) and fibroblast (Cos7) cells (Fig. 2 and4). Rarely did these proteins mediate syncytium formation in other tested cell lines, and when they did, it was not in a homologous cell line. For example, AvRVG NSP1-1 from a pigeon rotavirus mediated detectable cell-cell fusion in hamster BHK cells (Fig. 5A). These observations suggest that NSP1-1 cell fusion function is limited, but it is not strictly limited to the host in which the virus was initially detected or to cells derived from a specific tissue type.
Nonetheless, HuRVB NSP1-1 mediates cell-cell fusion in primate epithelial but not rodent fibroblast cells, while NBV MB p10 mediates cell-cell fusion in both cell lines (33) (Fig. 6 and7). Our findings indicate that the N-terminal ectodomain, TM domain, or both domains dictate this cell-specific fusion activity, which suggests that these domains participate in specific interactions with host cell molecules (Fig. 6 and7). Host molecules are required for cell fusion by several reovirus FAST proteins (53)(54)(55). While the reptilian orthoreovirus p14 FAST protein endodomain interacts with Grb2 to trigger N-WASP-mediated actin polymerization, aquareovirus p22 FAST protein uses adaptors Intersectin-1 and Cdc42 to trigger N-WASP-mediated branched actin assembly (53,54). To date, no specific interactions with cellular molecules have been identified for reovirus FAST protein N-terminal ectodomains or TM domains (39,40). Taken together, our findings indicate that there is cell-specific fusion activity for rotavirus NSP1-1 that is dictated by the N terminus, but they fail to clearly delineate species or cell type criteria for fusion activity. Why did most NSP1-1 proteins fail to induce syncytia in many tested cell lines despite successfully forming syncytia in primate cells (Fig. 2, 4 and5; Fig. S1)? It is possible that poor transfection efficiency, RNA instability, lack of protein expression, or protein mislocalization are responsible for the lack of observed cell-cell fusion in non-primate cell lines. However, our observations rule out several of these factors in at least some cases. Detection of GFP expressed from a pCAGGS plasmid indicated that each cell type was transfection competent under the assay conditions, although MDCK cell transfection efficiency was particularly poor (Fig. 5; Fig. S1). Syncytia were formed by NBV p10 in BHK and DF-1 cells, suggesting they were competent for fusion (Fig. 5). C-terminally FLAGtagged RVB and RVG NSP1-1 proteins were efficiently expressed in BHK cells and, except for HuRVB NSP1-1-FLAG, in DF-1 cells (Fig. S2 andS3). Except for HuRVB NSP1-1-encoding RNAs in DF-1 cells and RVI NSP1-1-encoding RNAs in PK1 cells, bicistronic plasmid transfections suggest that NSP1-1-encoding RNAs are stable in all cell lines (Fig. S4 to S8). However, inefficient NSP1-1 protein expression might explain the lack of cell-cell fusion for all constructs in PK1 and MDCK cells and for HuRVB NSP1-1 in DF-1 cells (Fig. S1 andS3). RNA instability is unlikely to explain the absence of NSP1-1-mediated fusion activity for RVB and RVG NSP1-1 in PK1 cells or for any NSP1-1 protein in MDCK cells (Fig. S1, S7 and S8). Both NSP1-1 RNA and protein expression appear to be stable in BHK cells and for RVG NSP1-1 proteins in DF-1 cells (Fig. 5; Fig. S2, S5 and S6). So, the reason NSP1-1 fails to mediate cell-cell fusion in these instances is unclear. There may be host factors present or absent in some cell types that promote or restrict syncytium formation. However, most cellular molecules identified to interact with reovirus FAST proteins are broadly expressed, including adaptors in the branched actin assembly network mentioned above and Annexin A1, which interacts with reptilian orthoreovirus FAST protein p14 and promotes fusion pore expansion (53)(54)(55). While restriction factors directed toward FAST proteins have not yet been identified, several cellular molecules have been proposed to restrict HIV-1 infection, some of which are expressed in specific cell types (reviewed in references [74][75][76][77]. Additional studies identifying cellular binding partners of NSP1-1 proteins may help uncover reasons for the specificity of NSP1-1 fusion observed in our experiments.
It is unclear what our observations regarding NSP1-1 behavior in cultured cells suggest about the behavior of NSP1-1 from fusogenic rotaviruses in their natural hosts. Despite detecting syncytia only in primate cell lines for most NSP1-1 proteins, we think it is unlikely that NSP1-1 mediates syncytium formation only in primates. Indeed, syncytia have been detected in the epithelial cells of the small intestinal villi of rats and pigs infected with RVB (78,79). It is possible that some rotavirus NSP1-1 proteins lack cell fusion activity in vivo, although these genes likely are maintained in the compact viral genome for a reason. FAST proteins may behave differently in the context of a natural rotavirus infection than when expression is driven by a non-native promoter following plasmid transfection. The lack of tissue culture systems for most of these viruses has limited their study in the laboratory, and whether cell-cell fusion is a contributor to the pathogenesis of each rotavirus from which the NSP1-1 sequences were derived remains unknown (67)(68)(69)(80)(81)(82). For NBV, FAST protein p10 dramatically increased virus titer and pathogenesis in a mouse model (56). Although NBV exhibits cell type-specific replication, it is determined not by p10 FAST, but by the p17 protein, which is encoded on the same segment in a separate open reading frame (83). While only p10 FAST expression is required to permit efficient NBV replication in primate epithelial (Vero) cells, both p10 and p17 expression are required for efficient replication in bat (DemKT1) cells, and another bat homolog of NBV p17 but not homologs from avian or baboon reoviruses could complement this p17 function. Rotaviruses lack a p17 homolog and may have evolved to confer species specificity directly via the FAST protein, or they may employ a different viral protein to confer this property. For the aquareovirus grass carp reovirus, fusion activity of the NS16 FAST protein is enhanced by the expression of another viral protein, NS26, possibly through its interaction with host lysosomes (84,85). Future studies of fusogenic rotaviruses may reveal new information about cell tropism for these viruses and whether additional viral proteins modulate tropism or FAST protein activity.
In summary, our observations provide evidence that the NSP1-1 proteins of species B, G, and I rotaviruses are FAST proteins that can mediate syncytium formation in at least some cell types. The N terminus of HuRVB NSP1-1 influences the cell type specificity of its fusion activity. Many questions and much work remain to be done to understand the biological mechanism and function of NSP1-1 in the context of a rotavirus and its natural host and to elucidate determinants of rotavirus host range and cell tropism.
## MATERIALS AND METHODS
## NSP1-1 alignment and prediction of protein features
NSP1-1 sequences were obtained from GenBank. Accession numbers for NSP1-1 FAST sequences are ADF57900 (HuRBV), ASN74338 (PoRVB), ASV45172 (CpRVB), AXF38051 (AvRVG), ASV45159 (GaRVG), YP_009130668 (CaRVI), and AQX34665 (FeRVI). The accession number for the Nelson Bay orthoreovirus Miyazaki-Bali strain p10 is BAT21545. For Fig. 1A and6A, amino acid sequences were aligned using MAFFT v7.2 using the E-INS-I strategy (63). For NSP1-1 protein feature prediction, myristoylation motifs were identified using ExPasy Scan Prosite (62). Transmembrane sequences were identified using DeepTMHMM (61). Amphipathic helices were identified using Proteus2 and heliQuest (https://heliquest.ipmc.cnrs.fr/index.html) (58,59). Sequences predicted to fold into helices were identified using AlphaFold2 through ColabFold (60). Model #1 was used for predictions of helical regions shown in Fig. 1A. In most cases, all models for a given NSP1-1 sequence were similar. The TM domain, polybasic region, and palmitoylated cysteines in NBV p10 were identified previously (72,73).
## Plasmids
NBV (Miyazaki-Bali) p10 in pCAGGS has been described previously (57). HuRVB (Bang117) NSP1-1 in pCAGGS and N-terminally FLAG-tagged and C-terminally FLAG-tagged forms of this construct have been described previously (33). pLIC6 was constructed by engineering a ligation-independent cloning site into mammalian expression plasmid pCAGGS. Sequences encoding CpRVB NSP1-1, PoRVB NSP1-1, AvRVG NSP1-1, GaRVG NSP1-1, CaRVI NSP1-1, and FeRVI NSP1-1 with a C-terminal FLAG peptide inserted prior to the STOP codon were synthesized (GenScript). Ligation-independent cloning following PCR amplification with appropriate primers and T4 DNA polymerase treat ment was used to clone the sequences either without (untagged) or with (tagged) the C-terminal FLAG peptide into pLIC6. Sequences encoding RVB, RVG, or RVI NSP1-1 followed by the encephalomyocarditis virus IRES (viral bases 260-836 of NCBI GenBank accession number NC_001479) and mEGFP (obtained from https://www.fpbase.org) were synthesized (GenScript). Ligation-independent cloning following PCR amplification with appropriate primers and T4 DNA polymerase treatment was used to clone the sequences into pLIC6. Sequences encoding chimeric HuRVB NSP1-1 and NBV MB p10 proteins with exchanged N termini, TM domains, and C termini were synthesized (GenScript). Ligation-independent cloning following PCR amplification with appropriate primers and T4 DNA polymerase treatment was used to clone the sequences into pLIC6. Nucleotide sequences of plasmid constructs were verified by Sanger sequencing.
## Cells
Human embryonic kidney 293T cells were grown in Dulbecco's modified Eagle's minimal essential medium (DMEM) (Corning) supplemented to contain 10% fetal bovine serum (FBS) (Gibco). Monkey kidney fibroblast Cos7 cells were grown in DMEM supplemented to contain 10% FBS. Baby hamster kidney cells expressing T7 RNA polymerase under control of a cytomegalovirus promoter (BHK-T7 or BHK) (86) were grown in DMEM supplemented to contain 5% FBS, 10% tryptose phosphate broth (Invitrogen), and 1% nonessential amino acids (Corning), with 1 mg/mL G418 (Invitrogen) added during alternate passages. Canine kidney fibroblast-like MDCK.1 (MDCK) cells were grown in Eagle's minimal essential medium (Corning) supplemented to contain 10% FBS. Porcine kidney epithelial LLC.PK1 (PK1) cells were grown in Medium 199 with Earle's salts (Gibco) plus 2.2 g/L sodium bicarbonate and supplemented to contain 3% FBS. Chicken embryo fibroblast UMNSAH/DF-1 (DF1) cells were grown in DMEM supplemented to contain 10% FBS. All culture media were supplemented to contain 2 mM L-glutamine (Corning). Except for BHK cells, culture media also contained 100 units/mL penicillin and 100 µg/mL streptomycin (Corning).
## Antibodies
Monoclonal mouse anti-FLAG antibody (Sigma), Alexa Fluor 488-conjugated anti-mouse IgG (Invitrogen), and rhodamine phalloidin (Invitrogen) are commercially available.
## Cell transfection and imaging
For differential interference contrast imaging, 293T cells (~2 × 10 5 per well) in 24-well plates were transfected with 0.1 µg of plasmid DNA per well using LyoVec transfection reagent (InvivoGen), according to the manufacturer's instructions, incubated for 18 h at 37°C (~48 h for bicistronic constructs), and fixed with 4% paraformaldehyde in PBS. F-actin was detected with rhodamine phalloidin (Invitrogen), and nuclei were detected using 300 nM 4′,6-diamidino-2-phenylindole (DAPI, Invitrogen), with washes in PBS. Cells were imaged using a Zeiss Axiovert 200 inverted microscope equipped with an HBO 100 mercury arc lamp or a Nikon STORM with a Hamamatsu ORCA Flash 4.0 CMOS monochrome camera and a Nikon DS-Ri2 color camera.
Cos7 cells (~6 × 10 4 per well) in 24-well plates were transfected with 0.5 µg of plasmid DNA per well using LyoVec, according to the manufacturer's instructions, and incubated at 37°C. At 18 h post-transfection, cells were fixed with 4% paraformaldehyde in PBS. Staining to detect F-actin and nuclei and imaging were conducted as described above.
BHK cells (~6 × 10 4 per well) in 24-well plates were transfected with 0.5 µg of plasmid DNA per well using TransIT-LT1 transfection reagent (Mirus Bio) in OptiMEM (Gibco), according to the manufacturer's instructions, and incubated at 37°C. At 18 h post-transfection (~48 h post-transfection for bicistronic constructs), cells were fixed with 4% paraformaldehyde in PBS. Staining to detect F-actin and nuclei and imaging were conducted as described above.
DF-1 cells (~1 × 10 5 per well) in 24-well plates were transfected with 0.1 µg of plasmid DNA per well using LyoVec, according to the manufacturer's instructions, and incubated at 37°C. At 18 h post-transfection (~48 h post-transfection for bicistronic constructs), cells were fixed with 4% paraformaldehyde in PBS. Staining to detect F-actin and nuclei and imaging were conducted as described above.
PK1 cells (~1 × 10 5 per well) in 24-well plates were transfected with 0.5 µg of plasmid DNA per well using LyoVec, according to the manufacturer's instructions, and incubated at 37°C. At 18 h post-transfection (~48 h post-transfection for bicistronic constructs), cells were fixed with 4% paraformaldehyde in PBS. Staining to detect F-actin and nuclei and imaging were conducted as described above.
MDCK cells (~6 × 10 4 per well) in 24-well plates were transfected with 0.5 µg of plasmid DNA per well using LyoVec, according to the manufacturer's instructions, and incubated at 37°C. At 18 h post-transfection (~48 h post-transfection for bicistronic constructs), cells were fixed with 4% paraformaldehyde in PBS. Staining to detect F-actin and nuclei and imaging were conducted as described above.
For confocal imaging, 293T cells (~2 × 10 5 per well) on sterile glass coverslips in 24-well plates were transfected with 0.1 µg of plasmid DNA per well using LyoVec, according to the manufacturer's instructions, incubated for 18 h at 37°C, and fixed with 4% paraformaldehyde in PBS. F-actin was detected with rhodamine phalloidin, FLAG peptides were detected with monoclonal anti-FLAG M2 (Sigma-Aldrich) diluted 1:100, and Alexa Fluor 546-conjugated anti-mouse IgG (Invitrogen) diluted 1:1,000, and nuclei were detected using 300 nM DAPI, with washes in PBS containing 0.5% Triton X-100. Coverslips were mounted and imaged on a Zeiss inverted LSM980 confocal microscope with a 20×/0.8 NA objective. Nuclei (DAPI) were imaged with a 405 nm excitation laser, GaAsP-PMT detector, and 508-579 nm emission; NSP1-1 (FLAG) was imaged with a 488 nm excitation laser, MA-PMT detector, and 408-505 nm emission; and β-actin (rhodamine phalloidin) was imaged with a 561 nm excitation laser, GaAsP-PMT detector, and 588-695 nm emission. Images were acquired over an ~12 µm height of cells divided into ~12 slices (Z-stack) covering an ~530 × 530 µm area (1,024 × 1,024 pixels). All images were processed using Fiji (87).
## Quantitation of syncytium diameter
293T cells in 24-well plates were transfected with 0.1 µg per well of plasmids encoding untagged or tagged forms of RVB, RVG, or RVI NSP1 or control plasmids and stained to detect actin and nuclei (untagged) or FLAG and nuclei (tagged) as described for imaging studies. Syncytia were identified visually and imaged using a Zeiss Axiovert 200 inverted microscope. When possible, the person analyzing images was blinded to the identity of the samples. Diameters of at least 30 syncytia imaged from a minimum of four independently transfected wells of 293T cells per plasmid construct were quantified as the average of two diameter measurements per syncytium made using the meas ure function in Fiji (87). Statistical analyses of cluster diameters were conducted using Kruskal-Wallis tests with Dunn's multiple comparisons in GraphPad Prism 9 (GraphPad).
## References
1. Desselberger (2014) "Rotaviruses" *Virus Res*
3. Estes, Greenberg (2013) "Rotaviruses"
4. Walker, Siddell, Lefkowitz et al. (2022) *Arch Virol*
5. Attoui, Mertens, Becnel et al. (2012) "Family -Reoviridae"
6. Matthijnssens, Otto, Ciarlet et al. (2012) "VP6-sequence-based cutoff values as a criterion for rotavirus species demarcation" *Arch Virol*
7. Ogden, Johne, Patton (2012) "Rotavirus RNA polymerases resolve into two phylogenetically distinct classes that differ in their mechanism of template recognition" *Virology (Auckl)*
8. Greenberg, Estes (2009) "Rotaviruses: from pathogenesis to vaccination" *Gastroenterology*
9. Amimo, Raev, Chepngeno et al. (2021) "Rotavirus interactions with host intestinal epithelial cells" *Front Immunol*
10. Ghosh, Kobayashi (2011) "Whole-genomic analysis of rotavirus strains: current status and future prospects" *Future Microbiol*
11. Matthijnssens, Van Ranst (2012) "Genotype constellation and evolution of group A rotaviruses infecting humans" *Curr Opin Virol*
12. Jelle, Ciarlet, Heiman et al. (2008) "Full genome-based classification of rotaviruses reveals a common origin between human Wa-Like and porcine rotavirus strains and human DS-1-like and bovine rotavirus strains" *J Virol*
13. Matthijnssens, Ciarlet, Mcdonald et al. (2011) "Uniformity of rotavirus strain nomenclature proposed by the rotavirus classification working group (RCWG)" *Arch Virol*
14. Hajnalka, László, Jakab et al. (2013) "Review of group A rotavirus strains reported in swine and cattle" *Vet Microbiol*
15. Papp, Matthijnssens, Martella et al. (2013) "Global distribution of group A rotavirus strains in horses: a systematic review" *Vaccine (Auckl)*
16. Díaz Alarcón, Liotta, Miño (2022) "Zoonotic RVA: state of the art and distribution in the animal world" *Viruses*
17. Kindler, Trojnar, Heckel et al. (2013) "Analysis of rotavirus species diversity and evolution including the newly deter mined full-length genome sequences of rotavirus F and G" *Infect Genet Evol*
18. Otto, Ahmed, Hotzel et al. (2012) "Detection of avian rotaviruses of groups A, D, F and G in diseased chickens and turkeys from Europe and Bangladesh" *Vet Microbiol*
19. Bányai, Kemenesi, Budinski et al. (2017) "Candidate new rotavirus species in schreiber's bats, Serbia" *Infect Genet Evol*
20. Brown, Beards, Chen et al. (1987) "Prevalence of antibody to group B (atypical) rotavirus in humans and animals" *J Clin Microbiol*
21. Marthaler, Rossow, Culhane et al. (2014) "Widespread rotavirus H in commercially raised pigs, United States" *Emerg Infect Dis*
22. Marthaler, Rossow, Gramer et al. (2012) "Detection of substantial porcine group B rotavirus genetic diversity in the United States, resulting in a modified classification proposal for G genotypes" *Virology (Auckl)*
23. Joshi, Ganorkar, Ranshing et al. (2017) "Identification of group B rotavirus as an etiological agent in the gastroenteritis outbreak in Maharashtra, India" *J Med Virol*
24. Krishnan, Sen, Choudhury et al. (1999) "Emergence of adult diarrhoea rotavirus in Calcutta" *India. Lancet*
25. Lahon, Maniya, Tambe et al. (1995) "Group B rotavirus infection in patients with acute gastroenteritis from India: 1994" *Epidemiol Infect*
26. Sanekata, Ahmed, Kader et al. (2003) "Human group B rotavirus infections cause severe diarrhea in children and adults in Bangladesh" *J Clin Microbiol*
27. Saiada, Rahman, Moni et al. (2011) "Clinical presentation and molecular characterization of group B rotaviruses in diarrhoea patients in Bangladesh" *J Med Microbiol*
28. Hung, Chen, Wang et al. (1984) "Waterborne outbreak of rotavirus diarrhoea in adults in China caused by a novel rotavirus" *Lancet*
29. Suzuki, Kuga, Miyazaki et al. (2011) "Genetic divergence and classification of non-structural protein 1 among porcine rotaviruses of species B" *J Gen Virol*
30. Kobayashi, Naik, Kusuhara et al. (2001) "Sequence analysis of genes encoding structural and nonstructural proteins of a human group B rotavirus detected in Calcutta, India" *J Med Virol*
31. Fang, Glass, Penaranda et al. (1989) "Purification and characterization of adult diarrhea rotavirus: identification of viral structural proteins" *J Virol*
32. Fang, Monroe, Dong et al. (1992) "Coding assignments of the genome of adult diarrhea rotavirus" *Arch Virol*
33. Hu, Salmen, Sankaran et al. (2022) "Novel fold of rotavirus glycan-binding domain Full-Length Text Journal of Virology December"
34. "AlphaFold2 and determined by X-ray crystallography" *Commun Biol*
35. Diller, Parrington, Patton et al. (2019) "Rotavirus species B encodes a functional fusion-associated small transmembrane protein" *J Virol*
36. Dunn, Cross, Greenberg (1994) "Comparison of the rotavirus nonstructural protein NSP1 (NS53) from different species by sequence analysis and northern blot hybridization" *Virology (Auckl)*
37. Kojima, Taniguchi, Kobayashi (1996) "Species-specific and interspecies relatedness of NSP1 sequences in human, porcine, bovine, feline, and equine rotavirus strains" *Arch Virol*
39. Eiden (1994) "Expression and sequence analysis of gene 7 of the IDIR agent (group B rotavirus): similarity with NS53 of group A rotavirus" *Virology (Auckl)*
40. Yamamoto, Ghosh, Ganesh et al. (2010) "Analysis of genetic diversity and molecular evolution of human group B rotaviruses based on whole genome segments" *J Gen Virol*
41. Boutilier, Duncan (2011) "The reovirus fusion-associated small transmembrane (FAST) proteins: virus-encoded cellular fusogens" *Curr Top Membr*
42. Ciechonska, Duncan (2014) "Reovirus FAST proteins: virus-encoded cellular fusogens" *Trends Microbiol*
43. Duncan (2019) "Fusogenic reoviruses and their fusion-associated small transmembrane (FAST) proteins" *Annu Rev Virol*
44. Sartalamacchia, Porter, Veletanlic et al. (2024) "Avian deltacoronaviruses encode fusion-associated small transmembrane proteins that can induce syncytia formation" *Virology (Auckl)*
45. Clancy, Duncan (2009) "Reovirus FAST protein transmembrane domains function in a modular, primary sequence-independent manner to mediate cell-cell membrane fusion" *J Virol*
46. Clancy, Duncan (2011) "Helix-destabilizing, beta-branched, and polar residues in the baboon reovirus p15 transmembrane domain influence the modularity of FAST proteins" *J Virol*
47. Corcoran, Clancy, Duncan (2011) "Homomultimerization of the reovirus p14 fusion-associated small transmembrane protein during transit through the ER-golgi complex secretory pathway" *J Gen Virol*
48. Dawe, Corcoran, Clancy et al. (2005) "Unusual topological arrangement of structural motifs in the baboon reovirus fusion-associated small transmembrane protein" *J Virol*
49. Key, Duncan (2014) "A compact, multifunctional fusion module directs cholesterol-dependent homomultimerization and syncytiogenic efficiency of reovirus p10 FAST proteins" *PLoS Pathog*
50. Salsman, Top, Barry et al. (2008) "A virus-encoded cell-cell fusion machine dependent on surrogate adhesins" *PLoS Pathog*
51. Barry, Key, Haddad et al. (2010) "Features of a spatially constrained cystine loop in the p10 FAST protein ectodomain define a new class of viral fusion peptides" *J Biol Chem*
52. Corcoran, Syvitski, Top et al. (2004) "Myristoylation, a protruding loop, and structural plasticity are essential features of a nonenveloped virus fusion peptide motif" *J Biol Chem*
53. Key, Sarker, De Antueno et al. (2015) "The p10 FAST protein fusion peptide functions as a cystine noose to induce choles terol-dependent liposome fusion without liposome tubulation" *Biochim Biophys Acta*
54. Shmulevitz, Epand, Epand et al. (2004) "Structural and functional properties of an unusual internal fusion peptide in a nonenveloped virus membrane fusion protein" *J Virol*
55. Top, Read, Dawe et al. (2012) "Cell-cell membrane fusion induced by p15 fusion-associated small transmem brane (FAST) protein requires a novel fusion peptide motif containing a myristoylated polyproline type II helix" *J Biol Chem*
56. Chan, Son, Schmid et al. (2020) "A viral fusogen hijacks the actin cytoskeleton to drive cell-cell fusion"
57. Chan, Arthur, Morstein et al. (2021) "Evolutionarily related small viral fusogens hijack distinct but modular actin nucleation pathways to drive cell-cell fusion" *Proc Natl Acad Sci*
58. Ciechonska, Key, Duncan (2014) "Efficient reovirus-and measles virus-mediated pore expansion during syncytium formation is dependent on annexin A1 and intracellular calcium" *J Virol*
59. Kanai, Kawagishi, Sakai et al. (2019) "Cell-cell fusion induced by reovirus FAST proteins enhances replication and pathogenicity of non-enveloped dsRNA viruses" *PLoS Pathog*
60. Kanai, Komoto, Kawagishi et al. (2017) "Entirely plasmid-based reverse genetics system for rotaviruses" *Proc Natl Acad Sci*
61. Gautier, Douguet, Antonny et al. (2008) "HELIQUEST: a web server to screen sequences with specific alpha-helical properties" *Bioinformat ics*
62. Montgomerie, Cruz, Shrivastava et al. (2008) "a web server for comprehensive protein structure prediction and structure-based annotation" *Nucleic Acids Res*
63. Mirdita, Schütze, Moriwaki et al. (2022) "ColabFold: making protein folding accessible to all" *Nat Methods*
64. Hallgren, Tsirigos, Pedersen et al. (2022) "DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks" *bioRxiv*
65. De Castro, Sigrist, Gattiker et al. (2006) "ScanProsite: detection of PROSITE signature matches and ProRule-associated functional and structural residues in proteins" *Nucleic Acids Res*
67. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
68. Chen, Knutson, Ciarlet et al. (2018) "Complete genome characterization of a rotavirus B (RVB) strain identified in alpine goat kids with enteritis reveals inter-species transmission with RVB bovine strains" *J Gen Virol*
69. Miyabe, Agnol, Leme et al. (2020) "Porcine rotavirus B as primary causative agent of diarrhea outbreaks in newborn piglets" *Sci Rep*
70. Chen, Knutson, Porter et al. (2017) "Genome characterization of turkey rotavirus G strains from the United States identifies potential recombination events with human rotavirus B strains" *J Gen Virol*
71. Otto, Liebler-Tenorio, Elschner et al. (2006) "Detection of rotaviruses and intestinal lesions in broiler chicks from flocks with runting and stunting syndrome (RSS)" *Avian Dis*
72. Mihalov-Kovács, Gellért, Marton et al. (2015) "Candidate new rotavirus species in sheltered dogs" *Hungary. Emerg Infect Dis*
73. Phan, Leutenegger, Chan et al. (2017) "Rotavirus I in feces of a cat with diarrhea" *Virus Genes*
74. (2025) *Full-Length Text Journal of Virology*
75. Cooper (2000) "The cell: a molecular approach. 2nd ed"
76. Yamanaka, Iwakiri, Yoshikawa et al. (2014) "Imported case of acute respiratory tract infection associated with a member of species nelson bay orthoreovirus" *PLoS One*
77. Shmulevitz, Duncan (2000) "A new class of fusion-associated small transmembrane (FAST) proteins encoded by the non-enveloped fusogenic reoviruses" *EMBO J*
78. Shmulevitz, Salsman, Duncan (2003) "Palmitoylation, membraneproximal basic residues, and transmembrane glycine residues in the reovirus p10 protein are essential for syncytium formation" *J Virol*
79. Malim, Bieniasz (2012) "HIV restriction factors and mechanisms of evasion" *Cold Spring Harb Perspect Med*
80. Cano-Ortiz, Luedde, Münk (2023) "HIV-1 restriction by SERINC5" *Med Microbiol Immunol*
81. Fackler, Keppler (2013) "MxB/Mx2: the latest piece in HIV's interferon puzzle" *EMBO Rep*
82. St Gelais, Wu (2011) "SAMHD1: a new insight into HIV-1 restriction in myeloid cells" *Retrovirology (Auckl)*
83. Chasey, Higgins, Jeffrey et al. (1989) "Atypical rotavirus and villous epithelial cell syncytia in piglets" *J Comp Pathol*
84. Vonderfecht, Huber, Eiden et al. (1984) "Infectious diarrhea of infant rats produced by a rotavirus-like agent" *J Virol*
85. Nakata, Estes, Graham et al. (1986) "Antigenic characterization and ELISA detection of adult diarrhea rotaviruses" *J Infect Dis*
86. Sanekata, Kuwamoto, Akamatsu et al. (1996) "Isolation of group B porcine rotavirus in cell culture" *J Clin Microbiol*
87. Kang, Nagaraja, Newman (1986) "Primary isolation and identification of avian rotaviruses from turkeys exhibiting signs of clinical enteritis in a continuous MA 104 cell line" *Avian Dis*
88. Nouda, Kawagishi, Kanai et al. (2022) "The nonstructural p17 protein of a fusogenic batborne reovirus regulates viral replication in virus species-and hostspecific manners" *PLoS Pathog*
89. Guo, Sun, Yan et al. (2013) "The NS16 protein of aquareovirus-C is a fusion-associated small transmembrane (FAST) protein, and its activity can be enhanced by the nonstructural protein NS26" *Virus Res*
90. Guo, Chen, Yan et al. (2015) "Identification of a functional motif in the AqRV NS26 protein required for enhancing the fusogenic activity of FAST protein NS16" *J Gen Virol*
91. Kobayashi, Ooms, Ikizler et al. (2010) "An improved reverse genetics system for mammalian orthoreoviruses" *Virology (Auckl)*
92. Schindelin, Arganda-Carreras, Frise et al. (2012) "Fiji: an opensource platform for biological-image analysis" *Nat Methods*
93. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12677772&blobtype=pdf | # ORF45-induced Filamin A phosphorylation promotes cell motility and cell-contact dependent viral infection of Kaposi's sarcoma-associated herpesvirus
Xiaojuan Li, Li Quan, Rihong Zhou, Xiangpeng Liu, Ronit Sarid, Ersheng Kuang
## Abstract
Kaposi's sarcoma-associated herpesvirus (KSHV) is the etiological agent of Kaposi's sarcoma, primary effusion lymphoma and multicentric Castleman disease. Studies have shown that cell-to-cell viral infection plays a key role in KSHV transmission in vivo, and differentiated B cells and endothelial cells might represent two distinct kinds of natural donors or recipients that radically support the lytic cycle of KSHV. Consistent with the observation that endothelial cells exhibit better acceptance and transmissibility than B lymphocytes in cell-cell contact-mediated KSHV transmission, the sequential cell detachment, migration and cell-cell contact is the determinant for this kind of viral transmission. To investigate the processes and regulation of cell-cell contact-mediated viral infection during KSHV lytic replication, we found that Filamin A, a key regulator of cell adhesion and motility, is phosphorylated in KSHV-infected adherent cells by lytic replication and ORF45 expression in an RSK-dependent manner. ORF45-induced Filamin A phosphorylation is important for cell detachment and migration, while both Filamin A knockout and S2152A knockin abolish this function. Interestingly, ORF45 deficiency, Filamin A knockout and S2152A knockin dramatically decreases KSHV de novo infection and cell-contact dependent viral infection in adherent cells. Taken together, our results demonstrated that the ORF45-Filamin A phosphorylation axis promotes cell detachment and migration and facilitates viral de novo infection and cell-to-cell transmission during KSHV lytic cycles.
## Introduction
As an oncogenic γ2-herpesvirus, Kaposi's sarcoma-associated herpesvirus (KSHV) is etiologically associated with three kinds of malignancies in AIDS and other immunosuppressed patients: Kaposi's sarcoma, primary effusion lymphoma and multicentric Castleman disease [1][2][3]. KSHV infection has two different cycles, only a few viral genes are expressed without production of infectious viral particles in latent cycle, while all viral genes are expressed and viral DNA is replicated and encapsidated into infectious virions during the lytic cycle. In KS lesions, the majority of spindle cells are latently infected, but spontaneous KSHV lytic replication occurs in a small percentage of cells [4]. Unlike other tumor viruses, KSHV latency alone is not sufficient to induce tumors and KSHV-infected endothelial cells tend to lose latent genomes; thus, recurrent lytic replication and reinfection are needed for the virus to persistently infect KS lesions [5].
Although most human cells can be latently infected by KSHV in vitro, only lymphatic endothelial cells, terminally differentiated B lymphocytes and keratinocytes support the natural KSHV lytic cycle [6][7][8]. Due to the low yield of virions and the low efficiency of viral infection, cell-free viral particles are difficult to transmit and spread viral infection in natural culture, indicating that cell-free viral particles are not the main mode for viral transmission and spread. Studies have shown that cell-to-cell viral infection plays a key role in KSHV transmission [9,10], and endothelial cells and B cells might represent two distinct kinds of natural donors or recipients of KSHV infection in vivo.
ORF45 is a highly expressed viral phosphoprotein from the immediate-early stage through the late stage of the lytic cycle and is eventually incorporated into virion particles as a tegument protein [11,12]. ORF45-null mutagenesis substantially attenuated viral lytic replication, virion production and primary infection, suggesting that ORF45 plays important roles in both the initial and late stages of viral infection [13,14]. ORF45 is a multifunctional protein that interacts with several cellular proteins to hijack cellular pathways for lytic replication. ORF45 efficiently inhibits the expression of type I interferon genes by interacting with interferon regulatory factor 7 (IRF-7) [15] and regulates the intracellular transport of newly formed viral particles through interaction with the kinesin-2 motor protein KIF3A [16]. This protein also interacts with p90 ribosomal S6 kinase (RSK) and contributes to sustained ERK-RSK activation, which plays essential roles during lytic replication [17,18]. In addition, Siah-1/2 interacts with ORF45 to mediate its ubiquitination and degradation through the proteasome [19]. Notably, the proteasomal degradation of ORF45 is also associated with the immediate-early KSHV protein RTA [20]. Mono-ubiquitination of ORF45 facilitates maturation of budding virions in the trans-Golgi and endosomes through an unknown mechanism [21]. The recent studies also reveal that ORF45 upregulates ATF4-LAMP3 expression and directly interacts with FoxK1/K2, to promote the lytic gene expression during late KSHV lytic replication [22][23][24].
Although knowledge of ORF45 during KSHV infection and diseases is still limited, the function of ORF45-RSK signaling has been identified through different approaches. Similar to the ORF45-null virus, a mutated virus encoding ORF45-F66A, which fails to interact with and activate RSK, exhibits reduced late lytic viral gene expression, a 5-10-fold decrease in production of infectious progeny viruses and reduced infectivity in de novo infection compared to wild-type viruses [14]. Alternatively, an inhibitory peptide that disrupts the ORF45-RSK interaction has been developed to inhibit spontaneous and chemical-induced KSHV lytic replication [25], providing a promising peptide agent for controlling KSHV lytic infection. Mechanistically, ORF45-mediated RSK activation promotes transcription and translation through the induction of c-Fos and eIF4B phosphorylation during the lytic cycle [26,27]. Importantly, a phosphoproteomic analysis identified the phosphorylated proteins during KSHV lytic replication as well as the cellular substrates of RSK induced by ORF45 [28], suggesting that ORF45-RSK signaling regulates diverse cellular and viral activity and behaviors.
Filamin A is an actin-binding protein that connects adjacent actin filaments and links actin filaments to membrane glycoproteins to regulate cell shape, attachment and migration by remodeling the cytoskeleton [29,30]. As a scaffold protein, it interacts with many intracellular proteins, including integrin, transmembrane proteins, and signaling molecules, to affect the intracellular trafficking, activity, morphology and behavior of organelles [31][32][33]. The functions of Filamin A are regulated by diverse kinases; phosphorylation of Filamin A at Serine 2152 is mediated by RSK and other kinases to control cell activity and behavior [34][35][36][37][38][39], whereas phosphorylation of Filamin A by CDK1 at other sites is important for successful cell division [40].
In the present study, we reveal that Filamin A phosphorylation is induced by ORF45-mediated RSK activation during KSHV primary infection and lytic replication; consequently, ORF45-induced Filamin A phosphorylation promotes the detachment and migration of lytic KSHV-infected cells and facilitates KSHV de novo infection and cell-to-cell viral transmission through cell contact and movement during the lytic cycle.
## Results
## ORF45 induces Filamin A phosphorylation during KSHV lytic replication
To further investigate the novel function of ORF45-mediated RSK activation, we detected the phosphorylation of Filamin A in the ORF45-expressing cells and found that wild-type ORF45 (WT), but not ORF45-F66A, induced Filamin A phosphorylation (Fig 1A). We identified the nuclear export signal and nuclear localization signal in ORF45 and constructed ORF45 mutants that are restricted exclusively to either the cytoplasm or the nucleus [41]. The nucleus-residing ORF45 (RN) lost this ability to induce Filamin A phosphorylation while cytoplasm-residing ORF45 (RC) maintained its full function in Filamin A phosphorylation (Fig 1A). To confirm the role of RSK in ORF45-induced Filamin A phosphorylation, we coexpressed the different RSK2 constructs with ORF45, and we found that the constitutively active RSK2 (CA) construct alone induced Filamin A phosphorylation, similar to ORF45 alone. ORF45 and RSK2 WT or CA coexpression augmented phosphorylation, Viruses that primarily express nucleus-restricting (RN) ORF45 retain the ability to produce progeny viruses similar to viruses that express wild-type ORF45, whereas viruses expressing cytoplasm-residing (RC) ORF45 lose this ability similar to the ORF45-null virus [41], where Filamin A phosphorylation was induced by cytoplasm-residing ORF45 but not by nucleus-restricting ORF45 (Fig 1D, right panel). These results indicated that increased Filamin A phosphorylation requires ORF45 localization in the cytoplasm, whereas nucleus-localized ORF45 loses the ability to induce Filamin A phosphorylation. When Bac16 cells undergoing lytic replication were treated with the RSK inhibitor BI-D1870 or SL0101, the increased level of Filamin A phosphorylation was abolished (S1B Fig) , suggesting that RSK activation is essential for Filamin A phosphorylation during lytic reactivation. Furthermore, the timing of Filamin A phosphorylation was investigated during KSHV primary infection, the level of Filamin A phosphorylation was immediately induced by both Bac16 and STOP45 viruses at 30 min post-infection probably through virion-binding cell surface receptors. Filamin A phosphorylation disappeared at 2 h post-infection and then the level strongly increased again starting from the immediately early stage after infection with Bac16 viruses (6 h post-infection) whereas infection with STOP45 viruses weakly increased the level at the delayed early stage (24 h postinfection), This finding is consistent with the timing of RSK phosphorylation and ORF45 expression (Fig 1E). The increased level of Filamin A phosphorylation during primary infection with Bac16 viruses was also abolished when cells were treated with the RSK inhibitor BI-D1870 or SL0101 (S1C Fig) . These results suggest that KSHV primary infection induces Filamin A phosphorylation at early stage mainly through ORF45-induced RSK activation. When iSLK.Bac16 and iSLK.STOP45 were induced by Dox and NaB treatment and then the Filamin A phosphorylation was detected by immunofluorescence staining, a high level of Filamin A phosphorylation was observed in iSLK.Bac16 cells but not in iSLK.STOP45 cells under Dox-NaB induction (Fig 1F). The phosphorylated Filamin A mainly co-localized with ORF45 in cytoplasmic compartment of iSLK.Bac16 cells undergoing lytic reactivation (Fig 1G). These results suggest that ORF45 induces cytoplasmic Filamin A phosphorylation in cells undergoing lytic replication. Thus, we concluded that ORF45 induces Filamin A phosphorylation during lytic replication through ORF45-induced RSK activation.
## ORF45 enhances cell detachment and migration
Since Filamin A phosphorylation regulates cell adhesion and migration [42,43], we examined whether ORF45 affects cell morphology and adhesion. In HEK293 cells, ORF45 overexpression reduced cell adhesion (Fig 2A and2B), whereas this were detected as indicated. D. Control cells, iSLK-Bac16, -STOP45, ORF45F66A, ORF45RN or ORF45RC cells were induced with Dox and NaB as described above, the cells were collected at 72 h after induction, and whole cell extracts were analyzed. E. HEK293 cells were infected with Bac16 or STOP45 virions (MOI = 10). At the different time points post-infection, the cells were collected and whole cell extracts were prepared and subjected to Western Blotting analysis as indicated. F. The iSLK.Bac16 or iSLK.STOP45 cells were induced with Dox and NaB for 48 h, and then fixed and stained with anti-pFilamin A antibody with Alexa-555 secondary antibody, and finally visualized with confocal fluorescence microscopy. The percentages of pFilamin A-positive cells in cells undergoing lytic reactivation were calculated in three independent experiments and shown. **, p < 0.01; t test. G. The iSLK.Bac16 cells were induced with Dox and NaB for 48 h, and then fixed and stained with anti-ORF45 antibody with Alexa-647 secondary antibody and anti-pFLNA antibody with Alexa-555 secondary antibody, and finally visualized with confocal fluorescence microscopy. The subcellular co-localization of ORF45 and pFilamin A was analyzed and shown. . Unlike the wild-type Bac16-infected iSLK cells, the Bac16-STOP45-infected cells did not detach during lytic infection, while reintroduction of ORF45 expression to the ORF45-null Bac16-infected cells restored the detaching phenotype whereas ORF45-F66A reintroduction did not (Fig 2D). These results indicate that ORF45 contributes to cell detachment during lytic replication in adherent cells and may also cause an anchoring-independent cell phenotype during the late lytic cycle. To investigate whether cell motility was affected by ORF45, we introduced ectopic ORF45 or ORF45-F66A into HUVECs using lentiviruses. ORF45 overexpression changed the morphology, whereas this alteration was not observed when ORF45-F66A was overexpressed (Fig 3A). The migration of ORF45-expressing HUVECs but not ORF45-F66A-expressing cells was increased in the transwell cassette (Fig 3B -3C). Similarly, as assessed using the wound healing assay, the cell migration was increased by ORF45 expression but not by ORF45-F66A expression (Fig 3D -3E). When the cells were treated with the RSK inhibitor BI-D1870 or SL0101, the increased ability of ORF45-expressing cells to migrate was greatly attenuated (S2A Fig) . These results indicate that ORF45 promotes cell migration in an RSK-dependent manner.
To further investigate the effect of ORF45 on cell motility during lytic cycle, we also performed wound-healing assays of the Bac16-infected cells undergoing lytic replication with different ORF45 expression levels. Within 48 h of primary infection, the cells remained attached and spread, and the wild-type Bac16-harboring cells exhibited a faster wound-healing ability; however, a slower ability was observed in the Bac16-STOP45-harboring cells, similar to the Bac16-ORF45F66A-harboring cells (Fig 3F -3G). Upon additional treatment with BI-D1870 or SL0101 for 24 h, the faster wound-healing ability of Bac16-infected cells undergoing lytic replication was halted whereas the slower wound-healing ability of Bac16-STOP45-harboring cells under conditions of lytic induction was not affected (S2B Fig) . Similarly, BI-D1870 or SL0101 treatment reduced the wound-healing ability of cells infected with Bac16 viruses to the levels comparable to those of cells infected with STOP45 viruses, which were minimally affected by these RSK inhibitors (S2C Fig) results suggest that ORF45, KSHV primary infection and lytic reactivation promote cell migration through RSK activation during the early lytic cycle.
## Filamin A phosphorylation is required for ORF45-induced cell detachment and migration
To further investigate the role of Filamin A phosphorylation in the ORF45-mediated cell motility, we mutated Serine 2152 to Alanine in Filamin A as the phosphorylation-deficient Filamin A construct. Neither wild-type nor mutated Filamin A-S2152A alone affected cell adhesion under normal culture conditions, even though they may alter cell morphology (Fig 4A ,left). ORF45 overexpression dramatically decreased cell spread and adhesion, and wild-type Filamin A overexpression did not affect this reduction; however, Filamin A-S2152A overexpression strongly recovered the spread and adhesion of the ORF45-expressing cells (Fig 4A ,right). Cell attachment was measured, and neither Filamin A WT alone nor Filamin A-S2152A alone decreased cell adhesion in the absence of ORF45 expression. ORF45 caused a reduction in cell attachment, and Filamin A overexpression augmented it, while Filamin A-S2152A overexpression abolished the reduction in the presence of ORF45 expression (Fig 4B -4C). These results suggest that Filamin A phosphorylation is required for ORF45-induced detachment.
Furthermore, these cells were seeded into transwell cassettes to detect migration across the membranes. Neither Filamin A wild type alone nor Filamin A-S2152A alone affected migration, while ORF45 expression enhanced migration, and Filamin A with ORF45 overexpression augmented migration; however, Filamin A-S2152A overexpression similarly abolished the increased migration (Fig 4D ), indicating that ORF45-induced migration also requires Filamin A phosphorylation. However, Filamin A and Filamin-S2152A overexpression did not decrease virion production (Fig 4E). These results suggest that ORF45-induced Filamin A phosphorylation is required for the increased motility of cells undergoing lytic replication but not for viral lytic replication. To investigate the role of Filamin A expression and phosphorylation in KSHV primary infection and lytic replication, viral gene expression in Filamin A WT, KI and KO cells with KSHV de novo infection or cells undergoing lytic reactivation was further detected. In cells infected with Bac16 or STOP45 viruses for 24 h, decreased levels of both the lytic gene RTA and the latent gene LANA were observed in Filamin A KI and KO cells compared with those in WT cells infected with Bac16 viruses (Fig 5B). In addition, the lower RTA and LANA levels observed in cells infected with STOP45 viruses were also reduced by Filamin A KI or KO (Fig 5B). However, when stable Bac16-or STOP45-harboring WT, KO or KI cells were induced with TPA + NaB for 48 h, the RTA expression levels and virion yield were lower in STOP45-harboring cells than in Bac16-harboring cells, however, equal levels of RTA and LANA expression were observed in Filamin A WT, KI and KO cells harboring the same viruses (S4A Fig), resulting in equal virion yields at 96 h post-induction (S4B Fig) . These results suggest that Filamin A expression and phosphorylation are important for KSHV de novo infection but not for lytic reactivation.
To characterize the key steps by which Filamin A expression and phosphorylation are required for KSHV primary infection, we further investigated the ability of viral entry into cells and nuclei after primary infection in Filamin A WT, KO and KI cells. When these cells were infected with cell-free Bac16 or STOP45 viral stocks (MOI = 10), lower levels of intracellular viral DNA at 2 h post infection were observed in Filamin A KO and KI cells compared with WT cells, whereas the levels were minimally affected between the Bac16-infected cells and the STOP45-infected cells (Fig 5C). These results suggest that Filamin A expression and phosphorylation are important for viral entry into cells, whereas ORF45 is not required at this stage. When the nuclear fractions were isolated at 8 h post infection, the viral DNA copy number was significantly greater in nuclear fractions from Bac16-infected cells compared with that from STOP45-infected cells, and the viral DNA copy numbers of both viruses were further reduced in nuclear fractions from Filamin A KI or KO cells compared with those from WT cells (Fig 5D). These results suggest that both ORF45 and Filamin A phosphorylation are required for intracellular capsid transport to the nucleus. As a result, compared with those in WT cells, the levels of viral gene expression (both the lytic gene RTA and the latent gene LANA) at 24 h post infection were decreased in Filamin A KO and KI cells (Fig 5E). These results suggest that ORF45 and Filamin A phosphorylation are important for KSHV primary infection at early stages and that viral particle-induced Filamin A phosphorylation is likely required for receptor-mediated endocytosis of virions and that ORF45-mediated Filamin A phosphorylation after infection promotes intracellular capsid transport to the nucleus.
Further, KSHV de novo infection using purified Bac16 or STOP45 virion particles was quantitated in Filamin A WT, S2152A KI or KO HEK293-mCherry cells. The results show that Filamin A KO and S2152A KI similarly decreased the percentage of GFP-positive Bac16-infected cells, a lower percentage of STOP45-infected cells was observed in WT cells and Filamin A KO or S2152A KI additionally dropped the percentages (Fig 6A). Thus, we conclude that ORF45 and Filamin A phosphorylation play the important roles for KSHV de novo infection with cell-free viral particles.
Studies have revealed that cell-contact mediated viral transmission is much more effective than cell-free viral particle transmission for secondary infection [9,10]. The efficiency of cell-to-cell viral infection was further measured using Filamin A WT, KO or S2152A KI HEK293-mCherry cells as recipients, the percentage of GFP-positive KSHV infection from Bac16-infected cells undergoing lytic replication was moderately decreased in FilaminA KI cells and dramatically decreased in KO
## ORF45-Filamin A phosphorylation promotes cell migration during KSHV primary infection and lytic reactivation at early stage
To confirm the important function of Filamin A expression and phosphorylation in cell mobility during KSHV primary infection and lytic reactivation, the wound healing assays were performed using Filamin A WT, KI or KO cells. The results revealed that both Filamin A KO and KI reduced cell migration, ORF45 overexpression increased cell migration in Filamin A WT cells, whereas the increased migration in presence of ORF45 overexpression was abolished in Filamin A KO and KI cells (Fig 7A and7C). Similarly, the primary infection of Bac16 viruses promoted the cell migration in WT cells but not in Filamin A KO or KI cells, and the primary infection of STOP45 viruses barely affected the cell migration in all cells (Fig 7B and 7C). After the cells were treated with TPA + NaB for 48 h to induce lytic reactivation, the wound-healing ability was dramatically reduced in the stable Bac16-harboring Filamin A KO or KI cells compared with stable Bac16-harboring WT cells (Fig 7D -7E). In contrast, the wound-healing ability of stable STOP45-harboring cells was much lower and similarly reduced in Filamin A KO or KI cells compared with that in WT cells (Fig 7D -7E). These results suggest that early KSHV primary infection and lytic reactivation at early stage promote cell migration through ORF45-induced Filamin A phosphorylation.
In conclusion, ORF45 induces Filamin A phosphorylation during KSHV primary infection and lytic replication, and then ORF45-Filamin A axis promotes cell detachment and migration and de novo and cell-contact dependent viral infection, to mediate the effective cell movement and viral transmission during lytic cycle, providing the importance for KSHV persistent infection and tumorigenesis.
## Discussion
We observed that the adherent cells undergoing KSHV lytic replication gradually become round and detached during the lytic cycle; however, how and why this phenotype occurs for KSHV infection and diseases remain unknown. In the present study, we demonstrated that ORF45 induces Filamin A phosphorylation through RSK activation during the lytic cycle, and then, the cells undergoing lytic replication become round and detached; consequently, ORF45-expressing and Filamin A-phosphorylated cells exhibit increased cell migration and anchoring independence. Therefore, ORF45-mediated Filamin A phosphorylation promotes KSHV de novo infection and cell-contact mediated viral infection. Then, KSHV-infected adherent cells undergoing primary infection or lytic replication exhibit high motility at the early stage and carry viruses to migrate together for secondary infection and spread. These results revealed that ORF45 induces cell detachment, migration and invasion of KSHV-infected cells during lytic replication and promotes KSHV de novo infection and cell-contact dependent viral infection by inducing Filamin A phosphorylation, indicating that ORF45-mediated Filamin A phosphorylation plays an essential role in KSHV viral transmission and pathogenesis.
Our results show that both ORF45 and Filamin A phosphorylation are important of both KSHV de novo infection of cell-free virion particles and cell-contact viral infection from cells undergoing lytic replication. Although the ORF45 protein cannot trigger RSK activation and Filamin A phosphorylation on cell surfaces, it can be released as a tegument protein from virion particle to the cytoplasmic compartment quickly after virions enter cells through receptor-mediated endocytosis. In addition, it is expressed immediately after primary infection, subsequently inducing RSK activation and Filamin A phosphorylation during the immediately early stage of primary infection. Consequently, the actin cytoskeleton will be reprogrammed to promote the movement of capsids to nuclei. Therefore, ORF45-induced Filamin A phosphorylation starts from the immediately early stage of primary infection and promotes the de novo infection, likely facilitating the capsid transport to the nucleus.
In addition to B cells, KSHV-infected human blood endothelial cells (BECs), lymphatic endothelial cells (LECs) and oral keratinocytes in vivo [6,7] support the natural KSHV lytic cycle to provide donors and virion reservoirs for viral transmission through both cell-associated and cell-free viral particles, respectively. However, the yield and infectious efficiency of cell-free viral particles are poor; thus, cell-contact viral infection may represent the main mode of viral transmission under natural conditions [9,10]. Thereafter, the release and movement of virus-replicating cells is a determinant for this kind of viral transmission; these cells either migrate and invade local lesions or become circulating endothelial cells, and then carry and transmit cell-associated viral particles through adhesion and cell contact.
The latently infected endothelial and epithelial cells show adhesive phenotypes, while these cells become detached at the late lytic stage (Fig 2). Although several KSHV gene products regulate cell migration for angiogenesis and tumorigenesis [44][45][46][47][48][49], only viral thymidine kinase (TK), by acting as a tyrosine kinase, has been reported to disrupt adhesion through FAK, paxillin and RhoA-ROCK-myosin II signaling [50,51]. In the present study, the sustained ORF45 expression and RSK activation induced Filamin A Ser2152 phosphorylation and then promoted the detachment and migration of adherent cells undergoing lytic replication, while ORF45-null cells undergoing lytic replication lost Filamin A phosphorylation and maintained focal adhesion. The increased cell detachment and movement enables cell migration during the early lytic cycle and the spread of virus-replicating cells during the late lytic stage to facilitate cell-contact mediated viral transmission. However, TK might not be the main modulator for detachment of cells under lytic replication because TK expression was not strongly affected by ORF45 loss during lytic replication in the Bac16-STOP45-infected cells compared with the Bac16 wild-type-infected cells [14]; thus, ORF45-mediated signaling would play an important role in this process.
Although Filamin A phosphorylation induced by ORF45 regulates cell detachment and motility, it is not essential for viral lytic replication (Figs 4E andS4). First, KSHV-positive BCBL1 and BC1 cells do not express detectable Filamin A (S6 Fig), but they support effective lytic replication and produce the progeny virions, indicating that KSHV lytic replication does not absolutely require Filamin A expression and phosphorylation. Second, Filamin A phosphorylation is induced by cytoplasm-localized ORF45, while nucleus-restricted ORF45 loses the induction of Filamin A phosphorylation (Fig 1D ), opposite to the essential nuclear localization of ORF45 in lytic replication [41]. Thus, we can conclude that Filamin A phosphorylation induced by ORF45-RSK signaling plays critical roles in cell motility and viral transmission rather than lytic replication.
The viral-encoded antiapoptotic proteins are upregulated during the lytic stages and then protect cell survival by inhibiting anoikis [52,53]. Filamin A phosphorylation is associated with ORF45 expression in cells undergoing KSHV lytic replication, and these cells show loss of spreading and focal adhesions, indicating anchoring independence in late KSHV lytic replication. Since the cells eventually produce virion particles, the exfoliated cells can migrate along the layer of endothelial and epithelial cells or become circulating cells, and then, virus-replicating cells will carry cell-associated viral particles and transmit viral infection over a long distance by cell contact and re-attachment, representing an effective mode of viral transmission in vivo.
In addition, ORF45 expression during KSHV primary infection might promote cell migration and invasion, suggesting that KSHV-infected endothelial and epithelial cells exhibit increased migration and invasiveness during primary infection. Since continued lytic replication and reinfection are essential for KSHV tumorigenesis [5] and a cluster of lytic genes, including ORF45, are expressed in early primary infection [54], the ORF45-induced effects on cell morphology and motility would play an important role in tumorigenesis and progression of KSHV-related diseases.
In conclusion, we characterized Filamin A as a novel substrate of ORF45-mediated RSK activation and revealed its important function in altering the cell morphology and motility of adherent cells undergoing KSHV primary infection and lytic replication and then promoting de novo infection and cell-contact dependent viral infection. We then elucidated one mechanism by which KSHV lytic replication induces cell detachment and movement and promotes the infection of cell-free and cell-associated viral particles for viral transmission, persistent infection and pathogenesis.
## Materials and methods
## Cell lines, reagents and plasmids
SLK cells, HEK293 cells and HEK293T cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS), 1% L-glutamine and penicillin-streptomycin. Wild type iSLK.Bac16, ORF45-null iSLK. STOP45 and ORF45 F66A mutated iSLK.ORF45F66A cells [14], nucleus-restricting (RN) and cytoplasm-residing (RC) ORF45-expressing iSLK.Bac16 cells [41], KSHV-positive iSLK.219 cells [55], were described previously and were maintained in DMEM supplemented with 10% FBS, glutamine, and antibiotics including G418, hygromycin or puromycin. Human umbilical vein epithelial cells (HUVECs) were purchased and cultured in complete endothelial cell medium supplemented with growth factors (ScienCell, Shanghai, China). Hygromycin B, puromycin and fibronectin were obtained from Invitrogen (Carlsbad, CA). G418, 12-O-tetradecanoylphorbol-13-acetate (TPA), sodium butyrate (NaB), and doxycycline were purchased from Sigma (St. Louis, MO). pcDNA3-Filamin A (FLNA) WT-and S2152A-expressing plasmids were from Addgene. pKH3, pKH3-RSK2, RSK2 Y707A (constitutive active), and RSK2 K100A/Y707A (kinase dead) were described previously [17,18]. ORF45 mutated constructs and the recombinant viral strains were described previously [14,41]. Transfection was performed with Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. RSK inhibitors BI-D1870 and SL0101 were purchased from MedChemExpress.
## Antibodies
Rabbit phospho-specific antibodies against FLNA (Ser-2152) and total FLNA were purchased from Cell Signaling Technology (Beverly, MA). Immunohistochemical phosphorylated FLNA Ser2152 antibody was obtained from Genetex, Inc. (Irvine, CA). The rat anti-LANA antibody was obtained from Advanced Biotechnologies, Inc. (Columbia, MD). Mouse anti-RTA and anti-ORF45 antibodies were described previously [26,41]. Immunoblotting analysis was performed with speciesmatched secondary antibodies labeled with IRDye680 or IRDye800 and visualized with the Licor Odyssey system.
## Lentivirus preparation and transduction
Recombinant lentiviral stocks were prepared in 293T cells through a triple plasmid co-transfection procedure with 1 mg/ml polyethyleneimine (PEI, Polysciences catalog number 23966). Briefly, 293T cells were seeded on 10 cm dishes, and cells at 60% confluence were used for transfection. A total of 20 μg plasmid with equal amounts of lentiviral plasmid and the package plasmids pSPAX2 and pMD2G was mixed with 66.7 μL of PEI in Opti-MEM medium. The mixture was added to the cells and incubated for 8 h. The lentivirus-containing supernatants were collected at 48 h post-transfection.
HUVECs or SLK cells were consequently transduced with lentiviruses following the standard procedure. For construction of Tet-on inducible ORF45 cells, full length ORF45 fragment was cloned into the pLVX-tight-puro vector, lentiviruses were made from the pLVX-Tet-on-advanced vector and pLVX-tight-puro-ORF45 plasmids with package plasmids, and the cells were transduced with both lentiviral stocks and selected with G418 or puromycin for 2 weeks.
## Establishment of Filamin A knockout (KO) and S2152A knock-in cell lines
HEK293 cells were transfected with pmCherry-C1 plasmid using Lipofectamine 2000 (11668027, Invitrogen, USA), followed by selection with 1 μg/ml G418 for 14 days. The mCherry-positive cells were isolated based on red fluorescence using a FACSAria III flow cytometer, pooled and passaged as stable HEK293-mCherry cell line.
To generate FLNA knockout (KO) and FLNA S2152A knock-in (KI) cell line, sgRNAs targeting the FLNA CDS or S2152A site were designed using the online tool (https://chopchop.cbu.uib.no/). The corresponding oligonucleotides were annealed and ligated into the BsmBI-digested lentiCRISPR v2 plasmid. The FLNA KO sgRNA-expressing plasmids were transfected into HEK293-mCherry cells using Lipofectamine 2000 for 24 h, followed by 1 μg/ml puromycin selection for 14 days. The single clones were individually picked, expanded and validated by Western Blots with anti-FLNA antibody as stable FLNA KO HEK293-mCherry cell line.
To generate FLNA S2152A KI HEK293-mCherry cell line, a donor plasmid, pcDNA3.1-FLNA S2152A, containing homology arms and the point mutation sequence, was linearized with MluI digestion plus gel purification and then co-transfected with the lentiCRISPR v2 plasmid at equimolar concentrations into HEK293-mCherry cells, along with 10 μmol/L SCR7 (S7742, Selleck Chemicals, USA). Seventy-two hours post-transfection, cells were digested into single-cell suspensions and seeded into 96-well plates. Single-cell clones were isolated, expanded, and validated by Western Blots with anti-FLNA (Ser-2152) specific antibody following 20% FBS stimulation for 0.5 h.
The sequences of sgRNAs are used as below:
FLNA KO-sg-1: CCTACGTTCAGGACCGTGGCGAT;
FLNA KO-sg-2: CCACGGTGATGGCACGCACACCA;
FLNA KO-sg-3: AGTGGAGTACACGCCTTACGAGG;
FLNA KI-sg: CCTTCAGTGGCCAACGTTGGTAG.
To establish stable Bac16 or STOP45-harboring cells, Filamin WT, S2152A KI or KO HEK293-mCherry cells were infected with Bac16 or STOP45 virions (MOI = 10) for 48 h, and then selected with 200 μg/ml hygromycin for an additional 2 weeks. The single colonies were pooled and passaged in the presence of 200 μg/ml hygromycin.
## Immunofluorescence staining
KSHV harboring iSLK-Bac16 cells were allowed to adhere and spread on poly-L-lysine-coated coverslips. At the indicated times, the cells were fixed, permeabilized and stained with rabbit anti-p-Filamin A and mouse anti-ORF45 antibodies and then with donkey secondary antibodies labeled with Alexa 555 or Alexa 647 dye (Invitrogen). Stained cells were mounted and visualized with a Zeiss LSM880 confocal microscope under an oil lens.
## Cell-free and cell contact-mediated viral infection
For cell-free viral infection, iSLK.Bac16 or iSLK.STOP45 cells were induced with 1 μg/ml Dox and 1 mM NaB for 96 h, the supernatants were collected and the viral stocks were prepared and purified with ultracentrifugation (100,000 g) as described previously [18]. Next, the purified virions (MOI = 10) were added to cells in the presence of 8 μg/ml polybrene. After the cells were centrifuged at 700 g for 1 h and incubated for additional 6 h in a 37 °C CO 2 incubator, the medium was refreshed.
For cell contact-mediated viral infection, iSLK.Bac16 or iSLK.STOP45 cells were induced with 1 μg/ml Dox and 1 mM NaB for 48 h, or Bac16 or STOP45-harboring HEK293-mCherry cells were induced with 20 μg/ml TPA plus 1 mM NaB for 48 h, and then washed twice. Then, the medium was replaced with fresh medium. The detached cells were collected at 72 h and added directly to HEK293-mCherry or HEK293 cells at a 1:1 ratio for cell-mediated viral infection as the donor and recipient cells, respectively.
## Transwell cell migration assay
Eight μm pore size transwell membranes (Costar) were coated with 10 μg/ml fibronectin at 4 °C overnight. For HUVECs, a total of 5 × 10 4 cells were seeded into the upper chamber in DMEM supplemented with 1% FBS, and the lower chamber was filled with 600 μL of DMEM with 20% FBS. Cells were incubated for an additional 24 h for cell migration. The transwell was then fixed in 75% cold ethanol for 15 min and stained with 0.5% crystal violet for 20 min. Cells in the upper chamber were removed with a cotton swab and washed twice. Then, the migrated cells on the lower side of the membrane were visualized with a Zeiss Axio Observer Z1 inverted microscope.
## Wound healing scratch assay
HUVECs or HEK293 cells were grown on 6-well plates as monolayers. After viral transduction for 12 h, the cells were scratched using a 200-μl pipette tip to obtain wounds of the same width. Then, the cells were washed and allowed to migrate in DMEM for an additional 8-24 h. The number of migrated cells entering the scratched area was recorded and counted under an inverted microscope. Three independent results were analyzed, and the means were calculated.
## Focal adhesion assay
A 96-well plate was coated with 10 μg/ml fibronectin in PBS at 4 °C overnight. The wells were then blocked with 3% BSA in PBS for 2 h at room temperature. Equal numbers of 293 cells in DMEM containing 0.1% FBS were seeded on fibronectin-precoated wells, and the cells were left to adhere for 1 h in a 37 °C CO 2 incubator. Then, the wells were washed with a low shear washing method to remove nonadherent cells, and the unwashed cells were used as a 100% control for quantification. Adherent cells were rinsed with PBS once, fixed in 4% formaldehyde for 30 min, and stained with 0.5% crystal violet for 40 min. The cells were then washed three times with PBS, solubilized in 100 µl of 10% acetic acid, and the absorbance was measured at 650 nm by a UV-visible spectrophotometer.
## Statistical analysis
Significant differences between experimental groups were determined by two-tailed unpaired Student's t-test. Values of p < 0.05 were considered statistically significant.
## References
1. Cesarman, Chang, Moore et al. (1995) "Kaposi's sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas" *N Engl J Med*
2. Soulier, Grollet, Oksenhendler et al. (1995) "Kaposi's sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman's disease" *Blood*
3. Chang, Cesarman, Pessin et al. (1994) "Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma" *Science*
4. Staskus, Zhong, Gebhard et al. (1997) "Kaposi's sarcoma-associated herpesvirus gene expression in endothelial (spindle) tumor cells" *J Virol*
5. Grundhoff, Ganem (2004) "Inefficient establishment of KSHV latency suggests an additional role for continued lytic replication in Kaposi sarcoma pathogenesis" *J Clin Invest*
6. Chang, Ganem (2013) "A unique herpesviral transcriptional program in KSHV-infected lymphatic endothelial cells leads to mTORC1 activation and rapamycin sensitivity" *Cell Host Microbe*
7. Johnson, Maronian, Vieira (2005) "Activation of Kaposi's sarcoma-associated herpesvirus lytic gene expression during epithelial differentiation" *J Virol*
8. Wilson, Tsao, Webb et al. (2007) "X box binding protein XBP-1s transactivates the Kaposi's sarcomaassociated herpesvirus (KSHV) ORF50 promoter, linking plasma cell differentiation to KSHV reactivation from latency" *J Virol*
9. Myoung, Ganem (2011) "Infection of lymphoblastoid cell lines by Kaposi's sarcoma-associated herpesvirus: critical role of cell-associated virus" *J Virol*
10. Sakurada, Katano, Sata et al. (2001) "Effective human herpesvirus 8 infection of human umbilical vein endothelial cells by cell-mediated transmission" *J Virol*
11. Zhu, Yuan (2003) "The ORF45 protein of Kaposi's sarcoma-associated herpesvirus is associated with purified virions" *J Virol*
12. Zhu, Cusano, Yuan (1999) "Identification of the immediate-early transcripts of Kaposi's sarcoma-associated herpesvirus" *J Virol*
13. Zhu, Li, Zhou et al. (2006) "Functional characterization of Kaposi's sarcoma-associated herpesvirus ORF45 by BAC-based mutagenesis" *J Virol*
14. Fu, Kuang, Li et al. (2015) "Activation of p90 ribosomal S6 kinases by ORF45 of Kaposi's sarcoma-associated herpesvirus is critical for optimal production of infectious viruses" *J Virol*
15. Zhu, King, Smith et al. (2002) "A Kaposi's sarcoma-associated herpesviral protein inhibits virus-mediated induction of type I interferon by blocking IRF-7 phosphorylation and nuclear accumulation" *Proc Natl Acad Sci U S A*
16. Sathish, Zhu, Yuan (2009) "Kaposi's sarcoma-associated herpesvirus ORF45 interacts with kinesin-2 transporting viral capsid-tegument complexes along microtubules" *PLoS Pathog*
17. Kuang, Wu, Zhu (2009) "Mechanism of sustained activation of ribosomal S6 kinase (RSK) and ERK by kaposi sarcoma-associated herpesvirus ORF45: multiprotein complexes retain active phosphorylated ERK AND RSK and protect them from dephosphorylation" *J Biol Chem*
18. Kuang, Tang, Maul et al. (2008) "Activation of p90 ribosomal S6 kinase by ORF45 of Kaposi's sarcoma-associated herpesvirus and its role in viral lytic replication" *J Virol*
19. Abada, Dreyfuss-Grossman, Herman-Bachinsky et al. (2008) "SIAH-1 interacts with the Kaposi's sarcoma-associated herpesvirus-encoded ORF45 protein and promotes its ubiquitylation and proteasomal degradation" *J Virol*
20. Wang, Yu, Pei et al. (2015) "Activation and degradation of open reading frame 45 by the replication and transcription activator of Kaposi's sarcoma-associated herpesvirus" *J Gen Virol*
21. Wang, Zhu, Li et al. (2015) "Mono-ubiquitylated ORF45 Mediates Association of KSHV Particles with Internal Lipid Rafts for Viral Assembly and Egress" *PLoS Pathog*
22. Sun, Wang, Chen et al. (2022) "Upregulation of ATF4-LAMP3 Axis by ORF45 Facilitates Lytic Replication of Kaposi's Sarcoma-Associated Herpesvirus" *J Virol*
23. Chen, Li, Quan et al. (2024) "FoxK1 and FoxK2 cooperate with ORF45 to promote late lytic replication of Kaposi's sarcoma-associated herpesvirus" *J Virol*
24. Palmer, Leo, Atyeo et al. (2024) "Conserved linear motif within the immediate early protein ORF45 promotes its engagement with KSHV lytic cycle-promoting forkhead transcription factors, FOXK1 and FOXK2" *J Virol*
25. Li, Huang, Xiao et al. (2019) "Development of an ORF45-Derived Peptide To Inhibit the Sustained RSK Activation and Lytic Replication of Kaposi's Sarcoma-Associated Herpesvirus" *J Virol*
26. Li, Du, Avey et al. (2015) "ORF45-Mediated Prolonged c-Fos Accumulation Accelerates Viral Transcription during the Late Stage of Lytic Replication of Kaposi's Sarcoma-Associated Herpesvirus" *J Virol*
27. Kuang, Fu, Liang et al. (2011) "Phosphorylation of eukaryotic translation initiation factor 4B (EIF4B) by open reading frame 45/p90 ribosomal S6 kinase (ORF45/RSK) signaling axis facilitates protein translation during Kaposi sarcoma-associated herpesvirus (KSHV) lytic replication" *J Biol Chem*
28. Avey, Tepper, Li et al. (2015) "Phosphoproteomic Analysis of KSHV-Infected Cells Reveals Roles of ORF45-Activated RSK during Lytic Replication" *PLoS Pathog*
29. Ohta, Hartwig, Stossel (2006) "FilGAP, a Rho-and ROCK-regulated GAP for Rac binds filamin A to control actin remodelling" *Nat Cell Biol*
30. Cukier, Li, Lee (2007) "Cyclin B1/Cdk1 binds and phosphorylates Filamin A and regulates its ability to cross-link actin" *FEBS Lett*
31. Muriel, Echarri, Hellriegel et al. (2011) "Phosphorylated filamin A regulates actin-linked caveolae dynamics" *J Cell Sci*
32. Sverdlov, Shinin, Place et al. (2009) "Filamin A regulates caveolae internalization and trafficking in endothelial cells" *Mol Biol Cell*
33. Urra, Henriquez, Cánovas et al. (2018) "IRE1α governs cytoskeleton remodelling and cell migration through a direct interaction with filamin A" *Nat Cell Biol*
34. Peverelli, Giardino, Mangili et al. (2018) "cAMP/PKA-induced filamin A (FLNA) phosphorylation inhibits SST2 signal transduction in GH-secreting pituitary tumor cells" *Cancer Lett*
35. Henkels, Mallets, Dennis (2015) "Gomez-Cambronero J. S6K is a morphogenic protein with a mechanism involving Filamin-A phosphorylation and phosphatidic acid binding" *FASEB J*
36. Waldt, Seifert, Demiray et al. (2018) "Filamin A Phosphorylation at Serine 2152 by the Serine/Threonine Kinase Ndr2 Controls TCR-Induced LFA-1 Activation in T Cells" *Front Immunol*
37. Sato, Ishii, Ota et al. (2016) "Mammalian target of rapamycin (mTOR) complex 2 regulates filamin A-dependent focal adhesion dynamics and cell migration" *Genes Cells*
38. Woo, Ohta, Rabinovitz et al. (2004) "Ribosomal S6 kinase (RSK) regulates phosphorylation of filamin A on an important regulatory site" *Mol Cell Biol*
39. Li, Lu, Stemmer et al. (2015) "Filamin A phosphorylation by Akt promotes cell migration in response to arsenic" *Oncotarget*
40. Szeto, Williams, Rudner et al. (2015) "Phosphorylation of filamin A by Cdk1 regulates filamin A localization and daughter cell separation" *Exp Cell Res*
41. Li, Zhu (2009) "Identification of the nuclear export and adjacent nuclear localization signals for ORF45 of Kaposi's sarcoma-associated herpesvirus" *J Virol*
42. Kim, Nakamura, Lee et al. (2010) "Filamin A is required for vimentin-mediated cell adhesion and spreading" *Am J Physiol Cell Physiol*
43. Zhang, Neal, Lian et al. (2013) "Filamin A regulates neuronal migration through brefeldin A-inhibited guanine exchange factor 2-dependent Arf1 activation" *J Neurosci*
44. Mansouri, Douglas, Rose et al. (2006) "Kaposi sarcoma herpesvirus K5 removes CD31/PECAM from endothelial cells" *Blood*
45. Samols, Skalsky, Maldonado et al. (2007) "Identification of cellular genes targeted by KSHV-encoded microR-NAs" *PLoS Pathog*
46. Couty, Lupu-Meiri, Oron et al. (2009) "Kaposi's sarcoma-associated herpesvirus-G protein-coupled receptor-expressing endothelial cells exhibit reduced migration and stimulated chemotaxis by chemokine inverse agonists" *J Pharmacol Exp Ther*
47. Sadagopan, Sharma-Walia, Veettil et al. (2009) "Kaposi's sarcoma-associated herpesvirus upregulates angiogenin during infection of human dermal microvascular endothelial cells, which induces 45S rRNA synthesis, antiapoptosis, cell proliferation, migration, and angiogenesis" *J Virol*
48. Tsai, Wu, Wu et al. (2009) "The M type K15 protein of Kaposi's sarcoma-associated herpesvirus regulates microRNA expression via its SH2-binding motif to induce cell migration and invasion" *J Virol*
49. Li, Yan, Ding et al. (2016) "The SH3BGR/STAT3 Pathway Regulates Cell Migration and Angiogenesis Induced by a Gammaherpesvirus MicroRNA" *PLoS Pathog*
50. Gill, Murphy, Fingeroth (2005) "Functional divergence of Kaposi's sarcoma-associated herpesvirus and related gamma-2 herpesvirus thymidine kinases: novel cytoplasmic phosphoproteins that alter cellular morphology and disrupt adhesion" *J Virol*
51. Gill, Turner, Stevenson et al. (2015) "KSHV-TK is a tyrosine kinase that disrupts focal adhesions and induces Rho-mediated cell contraction" *EMBO J*
52. Efklidou, Bailey, Field et al. (2008) "vFLIP from KSHV inhibits anoikis of primary endothelial cells" *J Cell Sci*
53. He, Bakken, Kassimova et al. (2012) "Focal adhesion kinase is required for KSHV vGPCR signaling" *Mol Carcinog*
54. Purushothaman, Thakker, Verma (2015) "Transcriptome analysis of Kaposi's sarcoma-associated herpesvirus during de novo primary infection of human B and endothelial cells" *J Virol*
55. Myoung, Ganem (2011) "Generation of a doxycycline-inducible KSHV producer cell line of endothelial origin: maintenance of tight latency with efficient reactivation upon induction" *J Virol Methods* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12755154&blobtype=pdf | # Legacy and impact of the 1925 Geneva Protocol: one hundred years of treaties and debates on chemical and biological weapons
Ben Krause-Kyora, Imee Martinez, Walter Grunden, Olli Tuovinen
## Abstract
This essay examines the legacy and impact of the 1925 Geneva Protocol, which prohibited the use of chemical and biological weapons. This multinational treaty was an important milestone in the history of the non-proliferation of weapons of mass destruction concluded in the wake of the horror engendered by the use of poison gases in World War I. However, the 1925 protocol did not address the issues of verification, manufacturing, stockpiling, and transferring products of chemical and biological weapons or production technologies. A second treaty, the result of the Biological Weapons Convention of 1972, was concluded to address these issues. Despite the apparent success of prohibiting large-scale and widespread use of chemical and biological weapons, violations have occurred, nonetheless, and the treaties remain problematic in numerous ways. The centennial of this historic 1925 agreement and its successor treaties presents an opportunity for reflection upon their efficacy. Given that there have been "limited" episodes of chemical and biological warfare since that time, how can these events be explained in light of the protocol's prohibitions? This essay argues that the 1925 Geneva Protocol has a mixed legacy and, at best, can be deemed only a qualified success. Other factors, such as political deterrence, scientific and technological limitations, and the problematic nature of these types of weapons may account for the absence of their mass use. The essay aims to serve as an introduction to the historiographical literature with an emphasis on biological weapons and the Geneva Protocol and includes tables as reference guides concerning the types of weaponized pathogens and toxins in question.
## Introduction
The year 2025 marks the centennial anniversary of the 1925 Geneva Protocol (i.e., "Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare"), which prohibited the use of chemical and biological weapons. This multinational treaty was an important milestone in international relations signed in the wake of World War I and the founding of the League of Nations. Some nations, however, found terms of the protocol to be politically problematic, which caused more than a few of them to balk at ratification, as was also the case with the Biological Weapons Convention in 1972, which came into force in 1975 as the "Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological [Biological] and Toxin Weapons and on Their Destruction." Since the inception of these agreements, there have been numerous allegations of violations and incidents of deception that have engendered no small amount of controversy. Despite the apparent success of these agreements in prohibiting large-scale and widespread chemical and biological warfare, violations have occurred, nonetheless. The treaties remain problematic in numerous ways. For example, there is still no single international biological warfare treaty that requires mandatory verification of compliance, and signatories have not always complied with voluntary verification, including restricting access to certain toxins and microorganisms that are potential biological weapons. Moreover, the treaties have no real means of enforcement relying, instead, largely upon multilateral trust and self-imposed compliance.
The centennial of this historic 1925 agreement presents an opportunity for reflection upon its efficacy. To what extent can it be argued that the 1925 Geneva Protocol has been successful in preventing the outbreak of chemical and biological warfare over the last century? Given that there have been "limited" episodes of chemical and biological warfare since 1925, how can we explain these events in light of the protocol's prohibitions? Are there other factors beyond the protocols that may explain why there have not been larger outbreaks of chemical and biological warfare otherwise? This essay argues that the 1925 Geneva Protocol has a mixed legacy and at best can be deemed only a qualified success. Although the signatories have largely abided by the spirit of the agreement, with notable exceptions, other factors may explain why the world has not seen the mass use of chemical and biological weapons beyond merely the 1925 agreement, including deterrence, scientific and technological limitations, and the problematic nature of these types of weapons. This essay will focus largely on the problem of biological warfare prohibition. The essay aims to serve as an introduction to the historiographical literature on biological warfare and the Geneva Protocol with additional tables that may serve as helpful reference guides for those less familiar with the subject.
## Defining biological weapons
Following Spencer and Wilcox (1993), we define biological warfare agents as "living organisms, whatever their nature, or infective material derived from them, which are intended to cause disease or death in man, animals or plants, and which depend for their effects on their ability to multiply in the person, animal or plant attacked." Biological weapons utilizing such agents have several potential uses that can generally be categorized as either tactical or strategic. Tactical use typically implies deployment for a limited military objective, such as contaminating an enemy's water supply, weakening enemy forces by spreading illness and disease locally, or using biological agents or toxins to assassinate designated personnel. Strategic use tends to imply larger-scale attacks such as against whole populations or large concentrations of enemy forces. This may include the deliberate spread of infectious diseases that kill indiscriminately and strike terror into the enemy. Attacking crops and livestock to decimate an enemy's food supply may also be considered a strategic use if deployed on a large scale. Developing biological weapons as a deterrent against enemy use "in kind" may also be considered a strategic use (Cookson and Nottingham, 1969). The United Nations Office for Disarmament Affairs (UNODA) explicitly states that biological weapons may also be used for "political assassinations, the infection of livestock or agricultural produce to cause food shortages and economic loss, the creation of environmental catastrophes, and the introduction of widespread illness, fear and mistrust among the public" (disarmament.unoda.org/biological-weapons/about/whatare-biological-weapons/).
The very nature of biological weapons has led to their development in secrecy. Some countries have had clandestine biological weapons programs that were formally acknowledged to have existed only after their termination. Other nations have continued to suppress information regarding their biological and chemical weapons programs and are suspected of covert continuation despite international agreements. As a result, the subject lends itself to endless speculation, suspicion, and focused intelligence gathering as accurate information is difficult to substantiate in the absence of corroborating evidence and the lack of cooperation by the respective governments where inspection and verification are concerned.
Biological and chemical weapons can serve as both a threat and a deterrence in national defense and regional conflicts. Compared with other forms of offensive weaponry, though often reviled, biological and chemical weapons have historically been attractive as military options because they generally do not cause physical damage at the target site as do more conventional kinetic weapons. While chemical weapons are typically disseminated in munitions such as bombs, mortars, and missiles, or sprayed by aircraft, making their application more easily observable, biological weapons may be delivered in ways that are initially undetectable, making them ideal for clandestine applications, such as poisoning a food or water supply, which may be accomplished by a lone individual properly equipped. Biological warfare agents are particularly susceptible to fluctuations in temperature and generally do not withstand the heat and shock caused by dispersal in munitions. Therefore, dissemination via kinetic weapons, while not always impossible, is less than optimal. Biological warfare agents may also be aerosolized to contaminate larger targets, but this means of deployment has obvious drawbacks for detection (Croddy et al., 2002). Some biological toxins may evade detection if they are novel biomolecules for which there are no reference compounds, or if their residues are unstable in the environment or biological tissue. Such weapons allow the continued use of affected surfaces and other contacts after the appropriate incubation period, after the contact time has passed, or after the epidemic has subsided. Depending on the biological agent, a time delay in a large-scale attack of the target area may be on the order of days or even weeks rather than seconds or minutes, as is the case with conventional ammunition or chemical weapons. New synthetic or genetically modified toxins can be targeted very narrowly, as in the poisoning of Russian dissidents (e.g., Sergei Skripal and his daughter Yulia in Salisbury, United Kingdom) with Novichok nerve agents in the past 10 years (Chai et al., 2018), in the premediated uses of ricin (castor bean plant, Ricinus communis), for example, in the assassination of the Bulgarian dissident Georgi Markov in 1978 (Nehring, 2021;Salisbury and Dew, 2023), and in the anthrax scare targeting select individuals in the U.S. by mail after the terrorist attacks of 9/11 in 2001 (Guillemin, 2011;Bush and Perez, 2012). Another example, widely reported in the media (accessed October 16, 2025) 1 is the fatal poisoning of the Chinese billionaire Lin Qi in 2020 by former executive Xu Yao, who first tested tetrodotoxin and methylmercury chloride efficacy in his small makeshift animal laboratory.
Prior immunization can provide effective protection against many biological weapons, and antidotes are now available for many chemical and biological poisons provided that the toxic agent can be identified in time. A notable exception is ionizing radiation, as in the case of the assassination of Alexander Litvinenko, a former KGB and FSB agent, who defected to British intelligence in 2001 and succumbed to 210 Po-induced acute radiation syndrome in 2006 (Dyer, 2007;McFee and Leikin, 2009).
Biological weapons can be classified into several groups of pathogenic agents, including rickettsiae, free-living pathogens, viruses, and biological toxins (Table 1). At least four species of rickettsiae (obligately intracellular bacteria in mammalian cells) have been tested for use in biological weapons. Of the pathogenic bacteria, the most focus has been on thirteen species, especially Bacillus anthracis. Over twenty human viruses are potential candidates for mass use as biological weapons. Biological toxins also have potential as agents for warfare because bacterial genes can be manipulated to encode novel toxin properties. Nucleotide sequences of toxin-encoding genes associated with pathogenicity factors are in the public domain and thus readily available. Genetically engineered microorganisms have attracted attention in the biological weapons arena because their toxins can potentially be altered, or the microorganisms may be specifically developed for enhanced toxin production in mass fermentation. Bioengineering and molecular and synthetic biology systems can conceivably be used to design new biological agents that can bypass natural immunity in humans and animals, and artificial intelligence can add powerful extensions in this direction. Genetically modified toxins especially can pose major analytical challenges if they are undetectable by current methods.
1 www.bbc.com/news/world-asia-china-68705857 Many microorganisms have been considered for biological warfare because of their virulence and pathogenicity (Wheelis et al., 2006). Some have natural biological vectors for dissemination such as insects, murine rodents, birds, and swines. Zoonotic diseases combined with resistance to antimicrobial drugs are of major concern. Animal and plant pathogens have received international attention because the disruption of animal husbandry or crop production would have strategic regional and national significance. National and international culture collections and databases serve as public repositories of almost all microbial cultures, nucleotide sequences of target genes, 16S rRNA genes, and genomic data published in peer-reviewed scientific and medical literature. Examples of databases are, among others, the CNGBdb, ENA, and NCBI (The China National GeneBank DataBase, European Nucleotide Archive, and National Center for Biotechnology Information, respectively). These culture collections were once available to the public at large without much scrutiny and precaution, but accessibility to virulent pathogens, such as Bacillus anthracis, is now severely restricted. To date, some countries remain in embargo or are required to participate in an individual validated licensing system to receive dualuse equipment or technical data that can be potentially used for the design, development, production, or use of biological weapons. Although there are many analytical methods (Table 2) that have been introduced for field and laboratory detection and investigation, the lack of access to suspected facilities has hampered inspections, not to mention incidents of deliberate concealing of evidence of biological weapons (Russell and Vogler, 2000).
The Centers for Disease Control and Prevention (CDC) has established three categories of pathogenic microorganisms ranked in order of their potential for weaponization and bioterrorism, as well as their general risk to public health and national security (Table 3). Sampling and rapid diagnosis are among key initial actions in this kind of public health threat and incident. The CDC is a partner in the Laboratory Response Network, which provides rapid access to federal, state, and local public health, military, food testing, environmental, and veterinary laboratories, their analytical expertise and equipment, and data exchange. The Network also has extensive international partners (e.g., The International Pathogen Surveillance Network), that can provide analytical and intelligence support in testing and identification protocols and data sharing. Similarly, Great Britain and the E.U. countries have developed up-to-date diagnostic capabilities and coordination in information sharing and are network-connected in bio-surveillance for biological agents and biothreats. Intelligence agencies have major roles in the reconnaissance for threats of biological and chemical agents.
Smallpox ranks highly on any list of potential biological warfare agents. Although this virus has been eradicated from the world population, two stocks of smallpox are known to exist in highlevel isolation facilities: one at the CDC, Atlanta, GA, and the other at the Vector Research Unit, Koltsovo, Novosibirsk Oblast, Russian Federation. Both locations possess over one hundred strains of this variola virus, genus Orthopoxvirus. There remains great concern about the possibility of the virus finding its way accidentally or deliberately into the hands of terrorists who would not hesitate to use it for nefarious purposes. Genetically very closely related to the smallpox virus is the camelpox virus, which is endemic among camels, but does not appear to pose a health hazard to humans such as camel handlers who are in frequent contact with the natural host. It is not known whether camelpox virus poses a risk to those humans who have had no previous contact with the natural host. According to documents prepared by UNSCOM (United Nations Special Commission) inspectors in 1995, Iraq attempted to weaponize camelpox at the height of its biological weapons program. Other genetically related members of this genus that can infect humans include cowpox, vaccinia, monkeypox, and rabbitpox viruses (Tucker, 2001). These viruses tend to have distinct geographical patterns of distribution and can be virulent in humans, but to date, none are known to have been released deliberately to infect humans or animals.
The World Health Organization (WHO) initially pushed for a 1999 target date for destruction of the Russian and American smallpox cultures, but this date was postponed indefinitely due to international debate over whether the stocks should be destroyed or remain in storage indefinitely. Routine vaccination against smallpox ceased in the U. S. and worldwide in 1978-1980, but in the year 2000, the CDC awarded a contract to a biotechnology company to produce a smallpox vaccine, reflecting a growing concern for security in the event of an act of bioterrorism. A French company, Aventis Pasteur, turned over to the U.S. government some 85 million doses of a similar vaccine that it had discovered in its freezers, where the stock had been stored frozen for about 40 years without a loss of biological activity. Smallpox vaccine production in the U.S. was recommenced in 2003 to provide protection for military personnel, health care workers, and emergency response teams. Smallpox vaccine MVA-BN can be used to protect against vaccinia virus and monkeypox virus, which caused a global epidemic in 2022-2023 (Pischel et al., 2024).
Among other animal diseases that could potentially be used in biological warfare, foot-and-mouth disease is particularly infectious and transmissible in hoofed animals, including cattle, pigs, sheep, and goats. The disease is caused by a virus of the apthovirus group, which has multiple serotypes with no cross-immunity. The virus is contagious and transmitted by air, contaminated clothing, boots, and equipment. It can remain viable in manure or straw for extended periods. Foot-and-mouth is considered a global disease as occurrences have been reported in multiple countries (Humphreys et al., 2025). A naturally occurring outbreak of foot-and-mouth disease in the U.K. in 2001 spread within weeks to many regions of the country. As a result, over 2,000 cases of the disease were confirmed and about four million animals were slaughtered, causing economic losses that were estimated at £8 billion (about £15 billion adjusted to July 2025). Its long-term persistence in the U.K. remains an issue of concern (Knight- Jones and Rushton, 2013). A similar outbreak in Taiwan in 1997, with the virus possibly introduced through smuggled meat or animals, devastated the pig farming industry and resulted in economic losses of billions of dollars. Fortunately, there were no human casualties of the disease involved in either outbreak, which was consistent with the lack of virulence of the virus in humans. In January 2025, foot-and-mouth disease was detected in a water buffalo farm Several plant pathogenic candidates also have been used for biological warfare. These have included potato blight (caused by Phytophthora infestans) and bacterial soft rot (Erwinia caratovara) affecting cabbage, carrots, onions, and potatoes. Stem rust of wheat (Puccinia graminis f. sp. tritici) has proven a good candidate because Puccinia spores can be released in the air over a large area. Other plant pathogenic fungi, bacteria, and viruses known to cause diseases in staple crops include common bunt (Tilletia fungi), covered smut (Ustilago hordei), black stem rust (Puccinia graminis), bacterial blight (Pseudomonas savastanoi), and white mold (usually Sclerotinia sclerotiorum), all of which have potential as biological weapons to cause crop loss as wild types or genetically modified pathogens with increased virulence. Plant pathogenic viruses have been ranked in order of their economic and scientific importance with the top five being the tobacco mosaic virus, tomato spotted wilt virus, tomato yellow leaf curl virus, cucumber mosaic virus, and potato virus (Scholthof et al., 2011).
Colorado potato beetles (Leptinotarsa decemlineata) were under consideration for weaponization in WWII. In 1939, France developed a program for their mass production, but this was cut short when Germany invaded the country (Lockwood, 2009).
Subsequently Germany developed capacity for the mass production and dispensation of the beetle in WWII. After trials of dispensing them aerially, Germany dispersed them in Isle of Wight in 1943 and later in Sussex, but evidence for the latter is in doubt. The outcome of these beetle attacks remained local, but it incentivized the U.K. and U.S. to develop capacity for mass production of the Colorado potato beetle. There is no physical evidence that this entomological weapon was used during the Cold War, although the Soviet Union accused the U.S. of targeted dispersal (Lockwood, 2009). Grunden and Tuovinen 10.3389/fmicb.2025.1685967 A summary history of biological warfare through World War II: the formative years Disease transmission has been a tactical option for military personnel, rebels, insurrectionists, and terrorists for many centuries, with suspected incidents dating back as far as 300 B.C. when the Greeks, and later the Romans and Persians, used the rotting carcasses of animals to contaminate the drinking water sources of their enemies (Robinson, 1971). Diseases were often associated with foul odors emitted from decaying carcasses and likely gave rise to the idea of contaminating enemy environments to gain a military advantage. Fighting forces throughout the world used carcasses as well as corpses of captured soldiers to pollute potable water supplies from the classical age to medieval times and to the modern era (Geissler and van Courtland Moon, 1999). But biological warfare would not become a truly viable component of any modern military arsenal until the development of biology, medicine, and hygiene as modern fields of science (Grunden, 2005).
The history of biological warfare, however, is replete with reports of deliberate attempts to weaponize diseases even well before the germ theory of disease was discovered or well understood, some of them historical and some only anecdotal. Among the more infamous of historically documented cases of "purposeful infection" occurred during the British colonization of North America when Sir Geoffrey Amherst, the British Commander-in-Chief in North America, urged his officers to use the deliberate spread of smallpox as a stratagem to reduce the numbers of indigenous enemies. Colonel Henry Bouquet, then serving as the ranking officer on the Pennsylvania frontier, wrote to Amherst of his intent to "inoculate the Indians with some blankets that may fall in their hands, " an incident that seems later to have become conflated with anecdotes of European (civilian) settlers allegedly spreading smallpox by gifting contaminated blankets, a charge much more difficult to substantiate (Oldstone, 2010). During the American Revolutionary War (1775-1783), British troops were vaccinated (variolated) against smallpox to prevent the spread of the disease among their own soldiers, while at the outset of the war Colonial troops remained vulnerable. Outbreaks among the Colonial troops inevitably occurred, and fearing deliberate infection by the enemy, in 1777, General George Washington ordered the inoculation of the entire Continental Army (Oldstone, 2010). Whether infection spread to Colonial forces by intent or incidental exposure in combat cannot be determined for certain.
During the U.S. Civil War (1861-1865), there were alleged incidents of premeditated pollution of drinking water poisoned with the carcasses of sheep and pigs. There were also allegedly plans by the Confederates to send clothing contaminated with yellow fever to the opposite side. The plan failed because yellow fever virus is mosquito borne and not transmitted through contaminated clothing collected from diseased people. Both sides also incurred considerable losses of war horses because of outbreaks of glanders (Koenig, 2006). The Civil War also marked the first instance of alleged use of an insect as a weapon of war. It was alleged (but never proven beyond doubt) that the Union deliberately introduced the Harlequin bug, Murgentia histrionica, to the South to cause crop damage. The insect attacks crucifers of all kinds and many other edible vegetables and can also destroy field crops and fruit trees. The South had experienced devastating outbreaks of communicable diseases and malaria due to unhygienic conditions during the war, or so Northerners believed. The Union also blockaded Confederate ports to curtail access to shipments of medicine, especially quinine, as well as food and clothing, thus exacerbating unsanitary field conditions in the camps (O'Flaherty, 1955). 3In 1874, representatives from fifteen European states convened in Brussels to address methods and means of combat and warfare. Two documents were drafted as a result of the Brussels Conference: "The Final Protocol of the Brussels Conference 1874" (a.k.a. Brussels Declaration), and "Project of an International Declaration Concerning the Laws and Customs of War." Article 13(a) of the latter prohibited "employment of poison or poisoned weapons." Some of the participant governments of the conference, however, did not ratify the declaration and refused to abide by the convention. The International Institute of Law in Geneva undertook an extensive review of the Declaration, which resulted in the publication of the "Manual of Law and Customs of War, " adopted in Oxford (Oxford Manual) in 1880. Two subsequent conventions at The Hague, in 1899 and 1907, which explicitly prohibited the use of poison gases, formalized the agreements resulting in "Conventions on Land Warfare, " later annexed with "Regulations, " which were largely based on the terms previously agreed upon in the Brussels Declaration and the Oxford Manual. Europe at the time was increasingly experiencing political and national tensions. In retrospect, it appears as though the potential role of microbes in spreading communicable human diseases and causing epidemics was becoming widely recognized among the European powers, and these international conventions reflected these growing concerns. This was the period when the ubiquity of microbes and their role in diseases was researched in European countries especially. Development of new biological methods, culturing, and classification opened new vistas of microbial biology in health and disease as well as in many other areas of human welfare.
Of greater and more immediate concern after the turn of the twentieth century, however, was chemical warfare in the form of poisonous gases. Chlorine gas, for example, was deployed on a large scale by German troops on April 22, 1915, at the Second Battle of Ypres in Belgium during World War I, a clear violation of the Hague agreements. In the German Empire, Fritz Haber, the 1918 German Nobelist (Chemistry), who discovered the method to produce ammonia from nitrogen and hydrogen gases in the Haber-Bosch Process, was instrumental in developing heavy chlorine gas used for offensive purposes (Willstätter, 1965;Braterman, 2012). Other principal belligerents of the war, including France, the U.K., Russia, Austria-Hungary, Italy, and the U.S., subsequently deployed chemical weapons themselves, including tear gas, chlorine, phosgene, sulfur mustard, and hydrogen cyanide gases (Spiers, 1986(Spiers, , 2010;;Brown, 2006). Germany also used bacterial weapons in attempts to transmit anthrax (Bacillus anthracis) and glanders (Burkholderia mallei) to horses and other animals and to contaminate the animal feed of the enemy forces, but these apparently did not yield successful results (Geissler and van Courtland Moon, 1999). Following the capitulation of Germany on November 11, 1918, the principal belligerents met in Versailles to conclude The Treaty of Paris, which was drafted in 1919 and took force in 1920. Article 171 of the Treaty specifically prohibited the use of "asphyxiating, poisonous or other gases and all analogous liquids, materials or devices, " while Article 172 demanded the destruction of all existing stockpiles of chemical weapons. No mention, however, was made of biological or bacteriological weapons (Geissler and van Courtland Moon, 1999).
The Geneva Protocol, signed on June 17, 1925, is a historical milestone because it was the first formal, international agreement concerning the prohibition of both chemical and biological weapons. Although the signatories were in general agreement about the horrific nature of these weapons, various aspects of international law as engendered by the Geneva Protocol were not sufficiently clarified, leaving many nations hesitant to ratify. (Significantly, although all parties in attendance became signatories, most nations did not ratify the protocol until years or decades later, including Japan in 1970 and the U.S. in 1975.) According to historian Edward M. Spiers, the protocol "failed to address the R&D, production, possession or transfer of such weapons." Moreover, it "avoided any reference to how the agreement could be verified or enforced, " essentially leaving it binding "only in relation to other states who were a party to the protocol" and it would "cease to be binding whenever enemy states used gas warfare." In short, the protocol was rendered little more than a "no first use" agreement (Spiers, 2010). While the Geneva Protocol came into force in 1928 and was registered with the League of Nations in 1929, it remained a problematic document whose enforcement remained unresolved (Dorsey, 2024). It was arguably as toothless as the League of Nations itself.
While most of the principal belligerents of World War I established programs, departments, or agencies to oversee the development and production of chemical weapons, none of them established formal biological weapons programs at that time. As the specter of fascism arose in Europe, however, that would begin to change. By the 1930s, most had returned to producing poison gases in mass quantities, and research and development on innovative systems for delivery and defense were underway in most every developed nation. Italy's use of mustard gas against Abyssinian forces in 1935 revealed the impotence of both the protocol and the League of Nations to prevent even the first use of chemical weapons in battle (Robinson, 1971;Harris and Paxman, 2002). With another global conflict looming on the horizon, the major powers explored all manners of weapons, even those ostensibly outlawed.
World War II brought about major investments in biological weapons programs among the primary belligerents. It should be remembered that the Geneva Protocol effectively banned only the first use of these weapons, and, as such, it did not act as a deterrent to research and development, or even production. In response to intelligence reports implicating Germany in developing bacteriological weapons, Sir Maurice Hankey, Secretary to the Cabinet and the Imperial Defense Committee, took the lead in promoting a biological weapons research program in the U.K., resulting in the establishment of the Microbiological Warfare Committee in October 1936. The program began on a defensive footing until the outbreak of war in September 1939, whereupon it took on a more offensive posture. In 1940, a modest biological weapons research facility was established at the Porton Down Chemical Defense Experiment Station near Salisbury, Wiltshire, under the direction of Britain's leading pathologist and microbiologist, Dr. Paul Fildes. A Biology Department was created at Porton Down that year to oversee research (Hammond and Carter, 2002). Anthrax was an early focus of the British biological weapons program. In the event of a German biological weapons attack, the U.K. planned to respond by dropping anthrax-laced cakes made of ground linseed meal across the German heartland. The plan to infect sheep and cattle, designated "Operation Vegetarian, " never materialized. Extensive experiments with anthrax were conducted on Gruinard Island (57.8868 • N, 5.4673 • W), where sheep were exposed to airborne B. anthracis or, in other instances, were shot with hollow bullets filled with anthrax spores. None of the 80 sheep survived. Due to the extent of contamination, Gruinard remained quarantined until 1986, when decontamination efforts were undertaken. Gruinard Island was declared anthrax free in 1990 by the U.K. Ministry of Defence (Manchee et al., 1981;Balmer, 2001;Harris and Paxman, 2002;Spiers, 2010). The U.K. Ministry of Defence had requisitioned the island for £500 in 1940 and sold it back to the heirs of the original owner for the same amount of sterling in 1990 (Aldhous, 1990).
In the U.S., the Chemical Warfare Service (CWS) received critical upgrades and increased financial support with budget allocations rising from $2 million in 1940 to over $1 billion in 1942 (Robinson, 1971). In 1943, a formal biological weapons program was established under the auspices of the CWS. The War Research Service (WRS) was established to serve as an advisory body on biological weapons policy, while R&D remained under the purview of the CWS. In April 1943, the Biological Weapons Research and Development Center was established at Camp Detrick in Frederick, Maryland. Later known as Fort Detrick, this site became the first biological weapons research and development facility in the U.S., employing some 4,000 staff members and researchers with supporting facilities for production and ordnance testing built at the Vigo Ordnance plant near Terre Haute, Indiana, Horn Island in Mississippi Sound and Granite Peak near the Dugway Proving Grounds in Utah (Robinson, 1971;Brophy et al., 1959). The U.S. biological weapons program pursued research on anthrax, botulism, brucellosis, psittacosis, tularemia, and glanders. In addition to these human and animal diseases, the U.S. also pursued research into several plant pathogens, continuing well into the Cold War era.
The Soviet Union (USSR, Union of Soviet Socialist Republics) established a formal biological weapons research program as early as 1925 with the formation of the Military Chemical Agency under the direction of Yakov Moiseevich Fishman. The Ministry of Defense and Ministry of Health, however, remained responsible for the oversight of biological weapons research at no less than thirty-five institutions throughout the country, ranging from the Moscow Institute of Epidemiology and Microbiology to comparable facilities, for example, in Leningrad (today St. Petersburg) and Koltsovo, Novosibirsk Oblast. Testing grounds were erected in isolated areas such as Gorodomlya Island in Lake Seliger for greater secrecy and safety. Soviet scientists investigated plague, anthrax, tularemia, typhoid, glanders, cholera, and footand-mouth disease (Robinson, 1971;Geissler and van Courtland Moon, 1999;Harris and Paxman, 2002;Leitenberg et al., 2012). It has been alleged that the Soviet Union deployed tularemia (Francisella tularensis) in the battle of Stalingrad (today Volgograd) in WWII (Alibek, 1999). However, this allegation has not been confirmed by independent sources, and tularemia of natural causes was one of the diseases afflicting both Soviet and German troops. The Soviets continued to invest in biological weapons R&D in the postwar era, but the onset of the Cold War made verification of their programs' activities difficult to verify.
Despite having ratified the Geneva Protocol in 1929, and regardless of limitations imposed upon it concerning rearmament, by the early 1940s, Germany had recovered much of its chemical weapons production capacity and had branched out into biological weapons R&D as well, though on a comparatively much smaller scale. Germany's biological weapons program, such as it was, remained decentralized and split largely between agencies and institutions under Dr. Heinrich Kliewe as director of research in the Office of the Surgeon General of the Wehrmacht (Chef des Wehrmachtsanitätswesens), and Professor Kurt Blome, director of the Kaiser Wilhelm Institute's Center for Cancer Research. The German BW program never equaled that of the U.K. or U.S., as it was never made a priority. Postwar U.S. intelligence assessments attributed this to "Hitler's personal opposition to the use of biological weapons, " though the supposed reasons for that disposition remain a point of debate among historians (Geissler and van Courtland Moon, 1999;Grunden, 2005;Spiers, 2010). Although Hitler was reportedly vehemently opposed to any offensive use of either chemical or biological weapons-ostensibly due to his own experience in having sustained injuries from a poison gas attack while serving as a corporal in WWI-this did not stop the Nazi regime from developing new forms of nerve gases, such as sarin and tabun, nor did it prevent them from testing their efficacy in experiments upon millions of Jews, Roma, and others deemed "Untermenschen" who were murdered in the numerous death camps in Germany and Poland, where most of the chemical and biological weapons experiments were undertaken and where such poisons were used for mass extermination. Rather, it is more likely that Hitler chose not to use these weapons for fear of retaliation in kind or due to logistical complications that made deployment disadvantageous otherwise (Schmaltz, 2017;Schmidt, 2015).
Of all the principal belligerents of WWII, it was Japan that arguably violated the terms of the Geneva Protocol with first-use strikes utilizing both chemical and biological weapons; yet the perpetrators of these acts-with few exceptions-escaped justice. Japan's foray into biological weapons research began in 1932, with Dr. Shiro Ishii, a microbiologist, then a senior army surgeon, third class (rank of major), who actively lobbied his superior officers to establish a biological weapons program. Ironically, Ishii was inspired by nothing less than the Japanese delegation's own report on the Geneva Protocol, which, he noted, specifically outlawed chemical and biological weapons. Biological weapons, Ishii argued, must have significant potential, otherwise, why would they have been prohibited? After undertaking a study of other nations' efforts in this field, Ishii concluded that Japan must pursue a biological weapons program or risk falling behind. The army brass agreed. Under the auspices of the Kwantung Army, a branch of the Imperial Japanese Army stationed in China, Ishii established two smallscale research facilities in the occupied area of China's northeast provinces (Manchuria), then known as Manchukuo, a puppet state under Japanese control from 1932.
In 1936, Ishii's operation was upgraded and officially designated the "Kwantung Army Epidemic Prevention and Water Supply Unit, " also known as "Unit 731." The following year, the Kwantung Army built an extensive research center for Ishii to continue his work. Located on the outskirts of the city of Harbin, the Pingfang complex ultimately grew to about 3,000 staff members, including medical doctors and microbiologists recruited from Japan's most prestigious institutions of higher education. Although mainly interested in plague, Ishii and his collaborators also conducted research on weaponizing smallpox, botulism, brucellosis, cholera, and dysentery. Together with Unit 100, the "Hippo-Epizootic Detachment, " they also cultivated anthrax and glanders for use against horses, sheep, and cattle. Research was also conducted on developing anti-plant agents, such as weaponizing Puccinia helianthi, a fungus also known as "red rust" and "common rust, " that is particularly harmful to numerous types of crops. The fungus is autoecious, that is, its life cycle is all within the same host plant.
In 1939, Ishii's Unit 731 detachment was mobilized to deploy biological weapons against the Soviet Red Army during the Battle of Khalkin-Gol (Nomonhan) near the border of Mongolia. To slow the advance of the Red Army, Unit 731 contaminated their main source of fresh water-the Khalkin-Gol River-with typhus, paratyphus, and cholera. The results were mixed but sufficiently encouraging for the Imperial Japanese Army to establish several more biological weapons units throughout its empire. For the duration of the war, Japan continued to engage in biological warfare, especially throughout China, but allegedly reaching as far as Burma, Thailand, Indonesia, and the Crown colony of Singapore (now known as the Republic of Singapore), resulting in deaths and casualties-largely civilians-ranging from the tens to the hundreds of thousands. Japan gained a decisive edge over the other principal belligerents in biological warfare because of its unfettered exploitation of human subjects, many of whom were subjected to experimental vaccines and treatments-and even vivisected (i.e., autopsies conducted on live subjects)-to observe the course of the various diseases and their effects on the body. An estimated 3,000 prisoners are believed to have perished at Pingfang at the hands of Unit 731 personnel (Williams and Wallace, 1989;Harris, 1994;Grunden, 2005).
After the war, in one of history's greatest miscarriages of justice, the U.S. granted immunity from prosecution to Ishii and other members of Unit 731 in exchange for their research data, the full extent and value of which remained unknown to U.S. intelligence and the Joint Chiefs of Staff when the deal was made. Only a dozen or so members of Unit 731-those unfortunate enough to have been captured by the Red Army as it overtook Harbin in the closing days of the war-ever faced a tribunal, a trial conducted by the Soviet Union in the city of Khabarovsk in 1949(Materials, 1950). None of the perpetrators of biological warfare faced justice at the International Military Trial for the Far East (i.e., Tokyo Trials). Nor were any Japanese leaders brought to justice for conducting chemical warfare, which the Japanese Army resorted to on more than 2,000 occasions in the China Theater alone. Given the close connection between the two types of warfare and the units deploying them, the U.S. Joint Chiefs, in consultation with General Douglas MacArthur, who oversaw the occupation of Japan as the Supreme Commander for the Allied Powers in the Pacific, made the decision not to pursue war crimes charges for the use of chemical weapons lest that investigation reveal the deal made with Ishii. As a result, many of those most responsible for first-use incidents of biological and chemical warfare went unpunished (Guillemin, 2017;Grunden, 2005). The Geneva Protocol had utterly failed to prevent Japan's use of these weapons in China and throughout Asia.
The Cold War (1947War ( -1991)): the bipolar years
The Geneva Protocol of 1925 did not prevent the Axis powers from resorting to chemical and biological warfare during World War II. Italy and Japan both deployed chemical weapons in battle, and one can make an argument that Nazi Germany violated the spirit of the Geneva Protocol by murdering Jews with poison gases during the Holocaust. Japan is also known to have engaged in biological warfare throughout its empire during the war. Given these facts, it would have been naïve for any of the major powers to assume that the protocol would prove more effective in the postwar era. All the major Allied powers of WW II continued research and development of chemical and biological weapons, which was not prohibited by the Geneva Protocol (Wheelis et al., 2006). In the U.K., research in biological weapons continued at Porton Down, where, in 1945, the Biology Department was renamed the Microbiological Research Department (MRD). The focus at the MRD was twofold including "vigorous laboratory investigation of potential agents" and large-scale field tests "with pathogens, toxins, or their simulants, " though ostensibly more care was taken to reduce risk by working with non-pathogenic bacteria where possible, such as Serratia marcescens and Bacillus atrophaeus (formerly B. globigii), which were used as simulants for pathogens (Guillemin, 2005;Hammond and Carter, 2002). Pilot plants were erected for their mass production. Experiments with actual pathogens, however, were generally preferred or even necessary for evaluating "real life" conditions. In 1946, the MRD was redesignated the Microbiological Research Establishment (MRE), which continued to operate under the auspices of the Ministry of Defence.
In coordination with the Royal Navy, the British Army, and U.S. military forces, Porton Down conducted open-air sea trials using a variety of live pathogens from 1948 through the early 1950s. "Operation Harness" entailed open-air sea trials off the coasts of Antigua and St. Kitts in the Caribbean ostensibly to avoid large-scale contamination such as what occurred on Gruinard Island (Hammond and Carter, 2002). Among the major discoveries in these trials were the short half-lives of biological agents disseminated in aerosols in the open environment. In some cases, sunlight was found to reduce the half-lives from minutes to seconds. The U.K. discontinued open-air testing in the mid-1950s and, in 1959, declared it would terminate its offensive biological weapons research altogether, though bacteriological experiments using B. atrophaeus, Escherichia coli 162, and S. marcescens as models to simulate the survival and dispersion of potential warfare agents were carried out throug the 1960s. All three bacteria were reputed to be harmless and non-pathogenic. Some experiments simulated attacks on urban areas, including the release of simulants at ground level in the Central London area and elsewhere in southern England. As part of these trials, simulant bacteria were introduced to ventilation systems in public buildings and even into the London Underground system. According to the U.K. Ministry of Defence, the releases did not constitute a public health hazard because the virulence of the test bacteria was greatly lowered (Balmer, 2001). In hindsight, however, the potential of infection could not have been unambiguously ruled out. Some facilities in Porton Down were also used in early trials of antibiotic production (licheniformin from Bacillus licheniformis) in collaboration with British pharmaceutical companies in the civilian sector. In 1979, the Centre for Applied Microbiology and Research (CAMR) was opened in Porton Down and the MRE was closed. MRE facilities were transferred to CAMR, which operated under the auspices of the U.K. Public Health Laboratory Services and now focused largely on vaccine development. A small Defence Microbiology Division was created within CAMR and has remained throughout the years as a safeguard and advisory resource (Carter, 1992). CAMR maintains the National Collection of Pathogenic Viruses and is part of the Defence Science and Technology Laboratory on the Porton Down campus, which is aligned with the Ministry of Defence and responsible for ensuring strategic defense and security in the U.K. (DSTL, 2025). The Porton Down Science Park on the campus houses several biotechnology companies in the defense, security, and health sectors.
In the U.S., while much of the focus turned to the nuclear arms race with the Soviet Union, advocates in the military sought to elevate biological weapons research "to approximate nuclear scale, " and research and development in chemical and biological weapons continued apace (Guillemin, 2005). The U.S. Chemical Warfare Service, which oversaw biological weapons research through WW II, saw its budget significantly reduced with the end of the war but returned to wartime levels in 1947 with the start of the Cold War. Research in biological weapons continued largely in secrecy until allegations of U.S. use of biological weapons surfaced during the Korean War in early 1952. A fact-finding mission to Korea led by the eminent China scholar, Professor Joseph Needham, under the auspices of the World Peace Council, concluded that the U.S. had engaged in biological warfare, a charge the U.S. vehemently denied, and which was never fully substantiated. According to the most recent scholarship on the controversy, made possible by the brief opening of Russian archives shortly after the fall of the Soviet Union, the allegations were fabricated by the KGB (Komitet Gosudarstvennoy Bezopasnosti or Committee for State Security) in collaboration with North Korean and Chinese operatives (Leitenberg, 1998(Leitenberg, , 2008)). The controversy persists with both positive and negative arguments being advanced in the historiography of the Korean War, with both sides arguing that conclusive, unambiguous evidence has not been presented to disprove their findings. Documented outbreaks of various diseases during the war, including viral hemorrhagic fever, smallpox, cholera, plague, and meningitis among Korean and Chinese soldiers would seem to implicate the United States. Though the U.S. government has acknowledged engaging in biological weapons field tests around this time, it continues to deny that it deployed pathogenic agents or vectors in the Korean War.
The U.S. biological weapons program experienced significant growth in the early 1950s. The Plum Island Animal Disease Center (NY) was established in 1954, and Camp Detrick was renamed Fort Detrick (MD) in 1956. Extensive field testing was undertaken largely using surrogate biological agents to model deadly pathogens. There were many instances of aerial spraying of organisms and simulants over populated areas in the 1950s and into Information pooled mostly from Hay (2007), Lockwood (2009Lockwood ( , 2012)). Aedes aegyptii mosquitoes are known vectors of the yellow fever virus and dengue virus. British field tests within the same time frame are summarized by Carter and Balmer (1999), Balmer (2001), Hammond and Carter (2002), and Harris and Paxman (2002).
the 1960s (Table 4). In 1950, Serratia marcescens was released off the coast of San Francisco at rates up to 5,000 cells/min. In the "St. Jo Program" conducted in early 1953, the U.S. Air Force simulated anthrax attacks on urban areas that included non-infectious aerosol releases over cities such as Minneapolis, MN, St. Louis, MO, and Winnipeg, MB, Canada (Cole, 1988(Cole, , 1997;;Guillemin, 2005). Field tests in the U.S. also included trials with mass breeding and release of potential vectors such as fleas and mosquitoes that could be engaged in transmitting bacterial or viral pathogens (Table 4).
In other field experiments, smoke screens were used to disguise aerial tests, and citizens were deliberately misinformed that these exercises involved only harmless smoke that was being tested for use in protecting cities from radar-guided missiles. In New York City, experiments with Bacillus subtilis spores showed that release in one underground station could infect the entire underground tunnel due to convection currents and winds (Cole, 1988).
Other field release experiments, such as "Shady Grove, " which used animals as targets, were performed on barges around the Johnston Atoll site in the South Pacific in the 1960s. Around this time, the vulnerability of U.S. warships to chemical and biological weapons was tested under the aegis of "Project 112" in jet-released aerosol experiments in the Atlantic and Pacific Oceans using chemical markers and surrogate bacteria for anthrax (Regis, 2023). These sea, land and air trials included crew shelters with positive pressure ventilation and immunizations against Francisella tularensis and Coxiella burnetii in live pathogen experiments (van Courtland Moon, 2009). Non-biological simulants ZnCdS, soap bubbles, and SO 2 were also used in aerial dispersion studies. These sea trials led to accusations of spreading Newcastle poultry disease in Cuba 1962 and introducing the tobacco blue mold disease (Peronospora hyoscyami f. sp. tabacina), which devastated Cuba's tobacco crop in 1979 and 1980 (Lucas, 1980), and releasing the insect pest melon thrips (Thrips palmi) in 1997 (Butler, 1997). Melon thrips are widespread and can infest a wide range of edible vegetables and fruit trees in the field as well as in greenhouses (Vázquez and Rodriguez, 1999). Cuba experienced an influx of other agricultural pests and hemorrhagic diseases in the latter half of the century. However, scientific evidence was not presented to support these charges, and the allegations are largely believed to have been politically motivated (Zilinskas, 1999(Zilinskas, , 2000;;Whitby, 2002;Lockwood, 2009).
Significant changes in U.S. biological weapons policy occurred in the 1950s and 1960s, culminating in President Richard M. Nixon's decision to renounce offensive biological weapons including R&D in the United States in 1969 (U.S. Department of the Army, 1977). He announced that BW research was to be confined to defensive measures only. In the early Cold War era, U.S. biological weapons policy remained on a no-first strike and retaliation basis only footing, a position that aligned with the spirit of the Geneva Protocol, even though the U.S. still had not ratified the agreement. In 1956, in response to what it perceived as bellicose rhetoric emanating from the Soviet Union, the U.S. shifted its position to emphasize that biological weapons could be a viable option in a general conflict. The decision to resort to biological weapons, however, would not be left to commanders in the field but was now explicitly restricted to the president. In 1961, institutional changes initiated by Secretary of Defense Robert McNamara saw biological weapons R&D being subsumed by Munitions Command under the purview of the Army Material Command, giving the U.S. Army exclusive rights to biological weapons development. Field testing was expanded and accelerated at this time, including the aforementioned operations, as well as "Magic Sword, " which tested the spread of dengue and yellow fever in the Pacific using simulants, "Yellow Leaf, " which tested the viability of various biological agents in jungle terrain, and trials of chemical agents VX and sarin conducted under the umbrella of Project 112 from 1962 to 1970 (Guillemin, 2005;Hersh, 1968).
Facing increasing pressure at home and abroad over the use of riot-control agents (CS tear gas) and defoliants such as Agent Orange (containing carcinogenic 2,3,7,8-tetrachlorodibenzo-pdioxin as an impurity) in the Vietnam War, in November 1969, President Nixon declared that the U.S. would no longer pursue biological weapons R&D for offensive purposes and would restrict all work to defensive purposes only. Existing stockpiles would be destroyed, and further restrictions would be placed on CW production as well. Nixon also declared his intent to ratify the Geneva Protocol. Although this was clearly progress, there were limitations. While Nixon ordered the discontinuation of the use of defoliants in Vietnam in 1970, he sustained the army's request to allow the continued use of CS with restrictions. On April 10, 1972, Nixon signed the multilateral Biological Weapons Convention (BWC), which with a later amendment to include toxins came to be known as the "Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, " or more commonly, the Biological and Toxin Weapons Convention (BTWC), which became effective on March 26, 1975. But by the time the BTWC came into effect, and the U.S. Senate ratified the Geneva Protocol on April 10, 1975, Nixon had been forced out of office due to the Watergate scandal. Nixon resigned in August 1974 leaving Gerald Ford to officially sign the accord (Guillemin, 2005;Spiers, 2010). According to biological warfare historian Jeanne Guillemin, after 1975, "the dominant issue for those concerned about biological weapons was how and if legal restraints actually prevented secret proliferation" (Guillemin, 2005, p. 131). Evidence suggests that they did not.
Even as the BTWC effectively banned the use of biological agents, the Soviet Union committed itself to an accelerated and expanded offensive biological weapons program. Spurred on by biochemist Yuri Anatolevich Ovchinnikov, on April 24, 1974, Premier Leonid Ilyich Brezhnev issued Order No. 131 DSP, which established the "All-Union Science Production Association, " more commonly known as "Biopreparat." The activities coordinated under the Biopreparat Program included 47 research and production facilities throughout the Soviet Union. At its height, Biopreparat employed over 30,000 people, of whom about one-third were engineers and scientists. Additionally, the Ministry of Agriculture sponsored an extensive R&D program on biological agents against livestock and crop plants; the Ministry of Defense employed thousands of people in several military institutes with expertise in microbiology; the Ministry of Health established a clandestine program to develop biological agents for special operations including assassinations; and the Academy of Sciences of USSR was involved as the highest scientific body in all aspects of the bioweapon programs. The bioweapons program operated under the pretext of producing antibiotics and other biopharmaceutical and veterinary products. As many as 60,000 personnel are estimated to have been involved in various research organizations and production facilities in the Soviet biological weapons programs, all under the watchful surveillance of the KGB. It was typical of those times that scientists and engineers assigned to these programs enjoyed better standards of living than their peers in other areas, but they were prohibited from traveling abroad on business or pleasure (Alibek, 1999;Leitenberg et al., 2012).
The Soviet Union is known to have been engaged in the mass production of numerous pathogens under the aegis of Biopreparat, including and especially Bacillus anthracis. In April 1979, an incident occurred at a production facility in Sverdlovsk (today Yekaterinburg) in the southern Ural Mountains that resulted in an outbreak of anthrax causing 68 known deaths. The official explanation from the Soviets was that the outbreak had been caused by contaminated livestock feed, and Soviet authorities refused inspection by Western observers to verify compliance under the auspices of the Biological Weapons Convention. In 1992, after the fall of the Soviet Union, the U.S. government sponsored an investigation of the incident with a team comprised of American and Russian scientists led by Harvard microbiologist Matthew Meselson, which determined the deaths from the Sverdlovsk incident had been caused by the inhalation of anthrax, and not from ingestion as the Soviets had claimed. In fact, the outbreak had been caused by an improperly maintained air filtration system at Compound 19, a military facility operating under Biopreparat, which had accidentally released anthrax spores. Boris Nikolayevich Yeltsin, then President of the Russian Federation, publicly acknowledged that the incident had been caused by an accidental release from a military facility and that the families of the victims would be compensated, though they never were (Alibek, 1999;Guillemin, 1999Guillemin, , 2005;;Leitenberg et al., 2012).
Development of multiple antibiotic-resistant strains of Bacillus anthracis was alleged to have taken place under the Soviet Biopreparat Program. Other potent, virulent bioweapons developed by the Biopreparat Program included infectious agents such as tularemia (Francisella tularensis), brucellosis, typhus (Rickettsia) glanders, melioidosis, Ebola, Marburg, monkey pox, Lassa, Bolivian hemorrhagic fever, Venezuelan equine encephalitis, and smallpox (Variola major).foot_4 President Yeltsin avowed in 1992 that the Soviet biological weapons program had been officially terminated, including also the testing facility on Vozrozhdeniya Island in the Aral Sea. Yet, the Russian Federation still possesses repositories of pathogenic bacteria and viruses that were used in the secret biological weapons program, such as the State Scientific Center of Applied Microbiology in Obolensk in the Moscow region, which was reported to have about 3,000 strains of pathogens in its collection. This bioweapon research laboratory complex received western financial aid in the early 2000s for conversion to a peaceful medical manufacturing facility. In the past decade, however, concerns have been voiced by various intelligence agencies regarding the apparent expansion and construction of new laboratory facilities at the site, as well as overtures from Iran to collaborate in the type of research conducted there.
Although the biological warfare arms race was largely a bipolar affair between the U.S. and the Soviet Union during the Cold frontiersin.org Grunden and Tuovinen 10.3389/fmicb.2025.1685967 War years, the People's Republic of China developed its scientific infrastructure for biotechnology research at this time, allegedly with an eye toward defense against biological weapons. This plan was prioritized, no doubt, with memories fresh in mind of the past attacks and experiments with pathogenic bacteria on Chinese civilians by the notorious Japanese Unit 731 during WWII. The approach in planning, setting up, and equipping research facilities was based on "dual use" of the infrastructure, including defense against bioweapons and capability for bioweapons production for deterrence. China acceded to the Geneva Protocol in 1952. The Chinese government has denied the existence of any active biological warfare program since 1984 when it renounced biological weapons and acceded to the BWC/BTWC. In modernizing facilities, scientific research, and funding in medical, public health, microbiological, and biotechnological sciences, the Chinese government has promoted vaccine programs and emphasized the importance of hygiene for preventing epidemic outbreaks and disease transmission (Schillinger, 2023). The status of research and stockpiles of biological weapons in China is not transparent; however, several reports submitted by China and U.S.-based institutions, agencies, and research centers are readily available on the internet concerning China's policies and bioweapons R&D, but their factuality may be questionable given the lack of access and inability to authenticate the sources (Smithson, 2007;Croddy, 2022;Mauroni, 2022; U.S. Department of Defense, 2023; Crowley and Dando, 2024).
The post-Cold War years: the Era of asymmetry, bioterror, and environmental remediation
The fall of the Berlin Wall in November 1989 signaled the imminent demise of the Eastern Bloc, and the collapse of the Soviet Union in December 1991 effectively ended the Cold War, at least between the U.S. and the USSR. The end of the bipolar world configuration left the U.S. as the lone superpower at the time, but it would not be long before it had to recalibrate and adjust to a new, multi-polar world in which asymmetrical warfare and terrorism would present unprecedented challenges. In many ways, the old bipolar paradigm was more stable and predictable, particularly where controlling weapons of mass destruction (WMD)-including nuclear, biological, and chemical (NBC)-were concerned. The signing of the Nuclear Non-Proliferation Treaty in 1968, the BWC in 1972, and the conclusion of the Chemical Weapons Convention (CWC) in 1993, all indicated a collective international will to restrict, if not outright ban, NBC weapons. The CWC regime appeared to offer a model modus vivendi for managing international control of WMD with mandatory inspections, verification, and an enforcement agency, which were all components that the Geneva Protocol and BTWC lacked. Such institutional safeguards did not exist for biological weapons, despite the BTWC and the U.S. and many other nations finally ratifying the Geneva Protocol of 1925. There was simply no way to predict or control what rogue nations, terrorist groups, or individuals might do (Guillemin, 2005).
Iraq's invasion of Kuwait in August 1990 raised new concerns about "niche nations" and their possession of WMDs. During the Iran-Iraq War (1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988), under the leadership of Saddam Hussein, Iraq employed chemical weapons against enemy forces, including mustard gas, sarin, and possibly VX (a thiophosphonate enantiomer) with horrific results, and infamously even used them against its own citizens-ethnic Kurds-in a chemical weapons attack on Halabja in 1988. Consequently, as the U.S. led a coalition of United Nations forces to repel the Iraqi invasion of Kuwait in Operation Desert Storm in January 1991, a major concern was whether Saddam Hussein would resort to using chemical or biological weapons. British and U.S. forces were immunized against anthrax, which was thought to be the most likely biological weapon to be used by the Iraqi military. Following the defeat of Iraq in the first Gulf War, in April 1991, the U.N. passed Security Council Resolution 687 (United Nations, 2004), which required Iraq to submit to a regime of international inspections to remove or destroy existing stockpiles of chemical and biological weapons and its capacity to produce them.
While Iraq agreed to inspection by the UNSCOM, it terminated the biological weapons program and declared all biological warfare agents had been destroyed before the UNSCOM could verify the full extent of Iraq's capabilities. Among the biological agents ostensibly destroyed were Bacillus anthracis, Clostridium botulinum, Clostridium perfringens, aflatoxins, and ricin. Some of these agents had been weaponized in aerial bombs, aerosol sprays, and missile warheads, but UNSCOM could not fully confirm their numbers and destruction. Subsequent UNSCOM efforts to verify clandestine Iraqi stockpiles and negotiate their destruction from late 1991 through 1995 were largely unsuccessful and effectively hampered by the Iraqi government on various pretenses. But UNSCOM was slowly collecting evidence to implicate Iraq and, in July 1995, its government was finally compelled to reveal that it had indeed developed a biological warfare program for offensive purposes, the origins of which dated back to as early as 1974. From these early explorations into biological agents, Iraq developed a significant biological weapons program under the leadership of microbiologist Nassir al Hindawi (Ph.D. 1969, Mississippi State University in Starkville), ultimately leading to the construction of a bioweapons mass production site at Al-Hakam. Effective from the year 2000, the United Nations Monitoring, Verification and Inspection Commission (UNMOVIC) replaced UNSCOM and was mandated by the UN Security Council to monitor the elimination of weapons of mass destruction in Iraq (Guillemin, 2005;Pearson, 2006;Spiers, 2010;Trevan, 2016).
In addition to unpredictable state actors, there are also "rogue nations" and terrorist groups. Beyond Iraq, many other nations in the Middle East and other developing regionsnot coincidentally often also being politically unstable parts of the world-have explored development of biological weapons to maintain a strategic balance vis-à-vis other states, including Bulgaria, China, Cuba, Egypt, India, Iran, Israel, Laos, Libya, North Korea, South Africa, South Korea, Syria, Taiwan, and Vietnam, among others (Hunger et al., 2013). The incentives of these countries to develop biological weapons have ranged from deterrence to compensation for weak conventional forces to counter stronger opponents, and intimidation for political and regional hegemony. In addition to bombs and other explosives, international terrorist organizations such as ISIS (Islamic State of Iraq and Syria) and the al-Qaeda network, have threatened and used chemical and biological weapons to destabilize governments for radical political and religious purposes, underscoring the need to establish international barriers for prevention and protection Grunden and Tuovinen 10.3389/fmicb.2025.1685967 against terrorism (Robinson, 1993;Juergensmeyer, 2003;Salama and Hansell, 2005;Zubay, 2005;Tucker, 2006Tucker, , 2013)).
In the aftermath of the September 11, 2001, terrorist attacks, the potential of chemical and biological terrorism became a timely political and legislative issue. This issue was exacerbated by an apparent biological weapons terrorist attack using anthrax, which occurred in the wake of the 9/11 attacks in early October 2001. In the ensuing weeks, several other cases surfaced as the result of the deliberate release of anthrax (Guillemin, 2011). As a result, on November 25, 2002, the Department of Homeland Security (2023) was established and several bills dealing with measures and protection against terrorism were introduced to the House of Representatives and the Senate. The Science and Technology Directory of the Homeland Department is now charged with defensive programs against biological and chemical threats. The Biowatch Program, started in 2003 to detect and respond to bioterrorism, is an initiative that collaborates and coordinates with networks of public health, emergency management, law enforcement, laboratory, scientific, and environmental health organizations in the U.S. Among other priority areas are intelligence and surveillance and rapid preparedness in the Department of Defense for medical countermeasures (immunization, medical tests, drugs) to eliminate threats in biological warfare especially (Biowatch, 2003(Biowatch, , 2011)).
These are in addition to the Office of Chemical and Biological Weapons Affairs (OCBWA), established in 2010 as part of the reorganization of the State Department's arms control agencies, and now charged with the mission to develop policies "to address emerging chemical weapons issues and challenges, assess compliance with the CWC and the Biological Weapons Convention (BWC), " and which now "serves as the U.S. National Authority overseeing U.S. implementation of the CWC" (U.S. Department of State, 2022). Another important measure taken at this time was the implementation of the United Nations Secretary-General's Mechanism (UNSGM) as a legal instrument for member nations to initiate an investigation of alleged uses of chemical and biological warfare (United Nations, 1987). While this is not maintained as a standing body, member nations may activate and deploy experts of their choosing to conduct the investigation on an ad hoc basis. 5In the wake of the 9/11 terrorist attacks, in 2004 the UN Security Council passed "Resolution 1540, " which stipulates that "all States shall refrain from providing any form of support to non-State actors that attempt to develop, acquire, manufacture, possess, transport, transfer or use nuclear, chemical or biological weapons and their means of delivery" and "shall take and enforce effective measures to establish domestic controls to prevent the proliferation of weapons and their means of delivery, " a measure taken largely to prevent terrorists from acquiring weapons of mass destruction (United Nations Resolution 1540). 6 Another example of federal incentives to advance knowledge for public welfare in this area, the CDC has launched a national program to train scientists in emerging infectious diseases that relate to issues of bioterrorism.foot_7 On a global scale, the WHO has instituted similar international programs for scientists from developing nations (openwho.org/channels/cbde).
Beyond concerns over rogue nations and terrorist groups, there is always the potential for a natural outbreak of an epidemic or pandemic, for which all nations must remain on guard. The outbreak of the SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) pandemic in 2019 is a case in point (Himmel and Frey, 2022). Whether the outbreak was naturally occurring and zoonotic in origin (in this case the mutation of the virus in a bat-its original host-to a pangolin) and spread from the Huanan Seafood Wholesale Market, more widely known as the "wet market" in Wuhan, or whether the virus escaped or was deliberately released from the Wuhan Institute of Virology (WIV), which is less than ten miles (14 km) from the Huanan market, remains uncertain and a point of contention. But some facts are not in dispute. The SARS-CoV-2 virus was originally isolated in the Wuhan Institute of Virology and was subsequently isolated and characterized in other countries (Nie, 2020;Stelzer-Braid et al., 2020;Wu et al., 2020;Himmel and Frey, 2022). The WIV is one of twenty separate biomedical research institutes in the Chinese Academy of Sciences, but it is the only institute specializing in virology, viral pathology, and virus technology. The Institute is comprised of five research centers, including the Center for Emerging Infectious Diseases, the Chinese Virus Resources and Bioinformatics Center, the Center of Applied and Environmental Microbiology, the Department of Molecular Virology, and the Department of Analytical Biochemistry and Biotechnology. Although the WIV is ostensibly independent of the People's Liberation Army (PLA), the PLA retains authority over all bioweapons R&D, related intelligence work, and biological weapons policy in China, and PLA scientists are known to have conducted research in virology and vaccinerelated work at the WIV, including vaccines and therapeutics relevant to coronaviruses. Documentation and assessments of China's life sciences and biotechnology are readily available in various reports (e.g., Shoham, 2015;ODNI, 2023;Rolland, 2024). 8 An investigation conducted in 2021 by the World Health Organization into the potential role of the Wuhan Institute in the pandemic did not find an unambiguous natural source for the COVID-19 virus, but it also did not uncover evidence for its release from the laboratory. To date, Chinese authorities have turned down requests from the WHO to share information on the animals sold at Wuhan markets, R&D and biosafety conditions at WIV laboratories, and genetic sequences from patients with COVID-19 early in the pandemic (World Health Organization [WHO], 2025). In 2023, the U.S. Office of National Intelligence declassified an intelligence report titled "Potential Links between the Wuhan Institute of Virology and the Origin of the COVID-19 Pandemic, " which did not provide any conclusive insight into distinguishing between a zoonotic source occurring naturally and a laboratory origin of the virus. More recently, the U.S. Central Intelligence Agency has leaned in the direction of favoring the "lab leak" hypothesis, though this conclusion was tagged with a "low confidence" evaluation by its own analysts (Klepper, 2025). Adding fuel to the fire of speculation are documented cases of previous outbreaks. (Biao and Wong, 2003;Knobler et al., 2004).
In response to the 2019 pandemic, nations around the world have been forced to reexamine their public policy and public health response plans and infrastructure to meet such future challenges. For example, in 2023, the government of the U.K. revised and updated its "UK Biological Security Strategy" policy statement to reflect the realities of a post-COVID-19 world. With new visions of the mission, outcomes and plans, the policy paper outlines five critical areas of biological threats and risks: "a major health crisis (such as pandemic influenza or new infectious disease); antimicrobial resistance; a deliberate biological attack by state or non-state actors (including terrorists); animal and plant diseases, which themselves can pose risks to human health; and accidental release and dual-use research of concern." The policy paper also outlines the development of a strategic network with multiple outcomes that will help make the U.K. more resilient to a broad spectrum of health risks and threats by 2030 (Gov. U.K., 2023).
Another issue of recent concern is the environmental remediation of former biological weapons R&D facilities and test sites. On Gruinard Island, the site of the U.K.'s extensive anthrax testing in 1942 and 1943, a massive decontamination effort was undertaken from 1979 using 280 tons of formaldehyde as a sterilant and 2,000 tons of seawater to clean up hot spots on the island. Other sporicidal chemicals were also tested for soil decontamination: potassium permanganate, glutaraldehyde, peracetic acid, and dodecylamine (Manchee et al., 1983). The decontamination program was finalized with an intentional fire that burned the brush and other vegetation. While Bacillus anthracis spores are reputedly still found in samples from Gruinard Island, they are not perceived to constitute a public health hazard because they are embedded under the topsoil. In 1990, the Ministry of Defence declared Gruinard Island anthrax free, and ownership of the island was returned in 1990 to the heirs of the original owner, who subsequently sold it to another private landowner in Scotland (Willis, 2009).
The former island of Vozrozhdeniya (now a peninsula) in the Aral Sea was one of the Soviet Union's main open-air biological testing sites and, today, is part of the independent Republic of Uzbekistan. The legacy of testing at Vozrozhdeniya is an environmental catastrophe with hundreds of tons of anthrax bacterial biomass disposed of by burial in sediment layers. With the water level receding in the Aral Sea, some burial sites have come very close to surfacing, causing concern about potential leakage of the storage drums and transmission to wildlife. As part of the economic package and cooperation stemming from the September 11, 2001, terrorist act in New York City, and the subsequent international effort to curtail terrorism, the U.S. offered to help the Uzbekistan government to clean up and dispose of the anthrax containers in a safe manner. Funding was also provided for travel to the U.S. and training in a broad range of programs covering education, energy, environment, and democracy, among many other issues.foot_8
## Conclusion: the mixed legacy of the Geneva Protocol
There seems to be no general consensus among historians or policymakers on the efficacy of the Geneva Protocol of 1925. Even as the Protocol was being drafted and the process of ratification was underway, observers expressed serious concerns about the prospects of its success without the full commitment of all parties (Hudson, 1924). In 1937, U.S. Brigadier General Agustin M. Prentiss, Technical Director of the Edgewood Arsenal and a member of the Chemical Warfare Service (CWS) in World War I, noted the lack of enthusiasm the international community seemed to have for the agreement, quipping "its apathetic reception by various governments has tended to defeat the purpose it was expected to serve" (Prentiss, 1937;Croddy, 2005). Ironically, perhaps no organization was more responsible for undermining the potential force of the Protocol than the U.S. Chemical Warfare Service itself, particularly under the leadership of General Amos A. Fries, who served as director of the CWS from 1919 to 1929 and effectively mobilized a combined lobby of chemists of the American Chemical Society and former members of the CWS to kill ratification of the Protocol in the U.S. Senate over fears of the impact it would have on the American chemical industry and diminishing the influence of the CWS. Moreover, they feared the League of Nations would eventually be empowered to influence control over the manufacture of chemicals if the Protocol were ratified, and the relationship between President Calvin Coolidge's administration (1923)(1924)(1925)(1926)(1927)(1928)(1929) and the League of Nations was arguably "ambivalent" at best (Jones, 1980;McElroy, 1991). Without Senate ratification, the U.S. professed support for the Protocol rang hollow and undermined its potential.
Aside from a reluctance to become further entangled with the League of Nations, the impetus for which came largely from an influential cadre of U.S. senators collectively known as "bitter-enders" or "irreconcilables" because of their irreconcilable opposition to participation in the League, Coolidge was also ambivalent about the Protocol because it "did not include any monitoring mechanisms or enforcement structures, relying instead on the good behavior of individual nations, " rendering it little more than a "gentlemen's agreement" among states as it were, a systemic weakness inherent in the League structure itself (McElroy, 1991). Furthermore, U.S. Secretary of State Frank B. Kellogg expressed the government's reluctance to submit to the League's authority on the matter, stating, "The United States will not tolerate the supervision of any outside body in this matter nor be subject to inspection or supervision by foreign agencies or individuals." In the end, the failure of the Coolidge administration to ratify the Protocol ultimately rested with the president himself (McElroy, 1991).
Because the Protocol did not prohibit the research, development, production, or stockpiling of chemical or biological weapons, it was effectively reduced to a "no-first-use" prohibition.
But as the brief history presented above has shown, even this prohibition failed to stop chemical and biological weapons from being used from the 1930s and beyond. The Protocol failed to prevent Italy's use of mustard gas against Abyssinian forces in 1935, Japan's widespread use of chemical and biological weapons throughout China and its Asian empire, U.S. use of defoliants (Agent Orange) and irritant chemicals (tear gas) in Vietnam, Egypt's use of nerve gases in Yemen in 1967, or Iraq's use of chemical weapons against Iran and its own ethnic Kurd population in the 1980s, all of which were arguably flagrant violations of the Geneva Protocol of 1925. Nor has the Protocol prevented the use of poison gases or toxins by terrorists or assassins, as in the 1995 sarin gas attack on the Tokyo subway system launched by members of the Aum Shinrikyo Cult, the attempted assassination of U.S. congress members with anthrax in 2001, or the assassination of Alexander Litvinenko, a former KGB member and defector who was poisoned with a lethal dose of radioactive 210 Po in 2006 (Tucker, 2006;Dyer, 2007).
History suggests that what mattered most in the context of war was not the legal prohibition of CBW as defined in the 1925 Geneva Protocol, but deterrence, that is, the ability of the defender to retaliate in kind. In most cases described above, the aggressor nations engaged in chemical and biological warfare without concern that these attacks would be returned in kind. The Protocol itself was powerless to prevent these attacks and served only as a legal constraint for ratifying nations who may or may not be held to account depending upon the outcome of the given war. Like the League of Nations itself, the Geneva Protocol had "no teeth, " no real powers of monitoring, verification, or enforcement. Consequently, new protocols were required to address these shortcomings, which ultimately led to the Biological and Toxin Weapons Convention of 1975 and the Chemical Weapons Convention of 1993. As chemical and biological warfare scholar Eric A. Croddy wrote in 2005, "With the 1993 Chemical Weapons Convention now in force, the Geneva Protocol is mostly only relevant today in its prohibition of biological warfare. . ., " but "glaring loopholes" remain, and the BTWC itself has remained in a state of relative limbo, "awaiting some initiative to achieve consensus among its parties" (Croddy, 2005).
On the other hand, others argue that the Geneva Protocol and the BTWC should not be evaluated in isolation as they generally serve as the "core elements" of a wider anti-biological weapons regime consisting of "a range of agreements and mechanisms implemented by States, [and] non-State actors in industry and civil society." Policy analyst Jez Littlewood illustrates how the BTWC has evolved incrementally through regular "review conferences" of concerned states parties that have convened every 5 years since 1980 resulting in over 140 additional "understandings" regarding the agreement and how to implement it. While the Geneva Protocol and BTWC serve as "the legal basis for a complete prohibition on the development, production, stockpiling, acquisition and use of biological and toxin weapons, " the additional understandings, while not legally binding, "represent a road map to the what and how of biological disarmament" (Littlewood, 2024).
As of August 2025, 189 states are now signatories of the BTWC, and "the norm against biological weapons use has become nearly universally accepted" (Wikipedia BTWC, Cross and Klotz, 2020). The convention largely strives to compensate for prohibitions not entailed in the original Protocol, that is, "the convention bans only the development, production, and stockpiling of biological agents (including toxins) for purposes and in quantities that have no justification for peaceful purposes as well as the development and possession of weapons systems for dissemination of biological agents, " but "it does not outlaw the wartime use of biological weapons; that's banned in the Geneva Protocol of 1925" (Cross and Klotz, 2020, UNODA). Unlike chemical weapons programs, which under the CWC are now subject to verification and inspection regimes, it is much more problematic to apply these to biological weapons programs given the particular "dual use" characteristics of bacteria and viruses for medical and life science research. It also does not serve as a deterrent to terrorists or criminals as "biocrimes"-the use of biological agents in acts of terrorism or assassination-which remain outside the BTWC mandate as well (Thränert, 1996;Cross and Klotz, 2020).
Recent evaluations of the BTWC may be as mixed as those of the Geneva Protocol, with some critics emphasizing its inability to prevent the development of biological agents and their use by terrorists, while others insist that having established a "nearly universal norm against biological weapons" and with the diplomatic community having largely addressed remaining concerns, the BTWC is now "far from being a toothless paper tiger" (Beard, 2007;Cross and Klotz, 2020). As it stands today, however, the most readily apparent legacy of the 1925 Geneva Protocol would seem to be its ban on the use of chemical and biological agents and toxins in warfare, a prohibition by which most-but not allnations have abided. But there seems to be no clear consensus on the extent to which this prohibition has been internalized by the international community (Evangelista and Tannenwald, 2017). It took another major agreement, the BTWC, and the subsequent "understandings, " to bring the biological weapons control regime more in line with the CWC and nuclear non-proliferation and disarmament treaties. On the whole, however, the prohibition against biological weapons still remains largely "a gentlemen's agreement, " a problem that the BTWC also could not solve. In the long run, the extent to which the Geneva Protocol succeeded or failed in the objective to prevent chemical and biological warfare remains a point of debate among historians and policymakers even 100 years after its presentation to the League of Nations in 1925.
## References
1. Aldhous (1990) "Biological warfare. Gruinard Island handed back" *Nature*
2. Alibek (1999) "Biohazard: The chilling true story of the largest covert biological weapons program in the world -told from the inside by the man who ran it"
3. Balmer (2001) "Britain and biological warfare: Expert advice and science policy, 1930-65"
4. Beard (2007) "The shortcomings of indeterminacy in arms control regimes: The case of the Biological Weapons Convention" *Am. J. Int*
5. Biao, Wong (2003) "SARS: Public health and social science perspectives" *Econ. Polit. Weekly*
6. Biowatch (2003) "The Biowatch Program: Detection of bioterrorism"
7. Biowatch (2011) "Institute of medicine (US) and national research council (US) committee on effectiveness of national biosurveillance systems: Biowatch and the Public Health System"
8. Braterman (2012) "Science, war, and morality; The tragedy of Fritz Haber"
9. Brophy, Miles, Cochrane (1959) "The chemical warfare service: From laboratory to field. United States Army in World War II: The technical services" *Department of the Army. Brown, F. J*
10. Bush, Perez (2012) "The anthrax attacks 10 years later"
11. Med
12. Butler (1997) "US-Cuba row over insects goes to weapons meeting" *Nature*
13. Carter, Balmer (1999) "Chemical and biological warfare and defense, 1945-90"
14. Carter (1992) "Biological warfare and biological defence in the United Kingdom 1940-1979" *RUSI J*
15. Chai, Hayes, Erickson et al. (2018) "Novichok agents: A historical, current, and toxicological perspective" *Toxicol. Commun*
16. Cole (1988) "Clouds of secrecy: The Army's germ warfare tests over populated areas"
17. Cole (1997) "The eleventh plague: The politics of biological and chemical warfare"
18. Cookson, Nottingham (1969) "A Survey of Chemical and Biological Warfare"
19. Croddy (2022) "China's role in the chemical and biological disarmament regimes" *Nonprolif. Rev*
20. Croddy, Perez-Armendariz, Hart (2002) "Chemical and biological warfare: A comprehensive survey for the concerned citizen"
21. Cross, Klotz (2020) "Twenty-first century perspectives on the biological weapon convention: Continued relevance or toothless paper tiger" *Bull. Atomic Scient*
22. Crowley, Dando (2023) "Regulation of toxins and bioregulators under the chemical weapons convention and the biological and toxin weapons convention"
23. Dorsey (2024) "Holding their breath: How the Allies confronted the threat of chemical warfare in World War II"
24. Dyer (2007) "More cases of polonium-210 contamination are uncovered in London"
25. Geissler, Van Courtland Moon (1999) "Biological and toxin weapons: Research, development and use from the middle ages to 1945"
26. Grunden (2005) "Secret weapons and World War II: Japan in the shadow of big science"
27. Guillemin (1999) "Anthrax: The investigation of a deadly outbreak"
28. Guillemin (2005) "Biological weapons: From the invention of state-sponsored programs to contemporary bioterrorism"
29. Guillemin (2011) "American anthrax: Fear, crime, and the investigation of the nation's deadliest bioterror attack" *NY: Times Books. Guillemin, J*
30. Hammond, Carter (2002) "From biological warfare to healthcare: Porton Down, 1940-2000"
31. Harris, Paxman (1994) "Factories of death: Japanese biological warfare 1932-45 and the American cover-up"
32. Hay (2007) "A magic sword or a big itch: An historical look at the United States Biological Weapons Programme" *Med. Confl. Surv*
33. Hersh (1968) "Chemical and biological warfare: America's hidden arsenal"
34. Himmel, Frey (2022) "SARS-CoV-2: International investigation under the WHO or BWC" *Front. Publ. Health*
35. Hudson (1924) *The Geneva Protocol. For. Aff*
36. Humphreys, Stenfeldt, King et al. (2025) "Epidemiology and economics of foot-and-mouth disease: Current understanding and knowledge gaps" *Vet. Res*
37. Hunger, Radosavljevic, Belojevic (2013) "Biopreparedness and public health: Exploring synergies"
38. Jones (1980) "American chemists and the Geneva Protocol" *ISIS*
39. Juergensmeyer (2003) "Terror in the mind of god: The global rise of religious violence"
40. Klepper (2025) "The CIA believes covid most likely originated from a lab but has low confidence in its own finding"
41. Knight-Jones, Rushton (2013) "The economic impacts of foot and mouth disease -What are they, how big are they and where do they occur? Prevent" *Vet. Med*
42. Knobler, Mahmoud, Lemon et al. (2004) "Forum on microbial threats learning from SARS: Preparing for the next disease outbreak -workshop summary"
43. Koenig (2006) "The fourth horseman. One man's secret mission to wage the Great War in America"
44. Leitenberg (1998) "New Russian evidence on the Korean War biological warfare allegations: Background and analysis" *Cold War Int. Hist. Project Bull*
45. Leitenberg (2008) "False allegations of U.S. biological weapons use during the Korean war"
46. Leitenberg, Zilinskas, Kuhn (2012) "The Soviet biological weapons program: A history"
47. Littlewood (2024) "The 1925 Geneva protocol and the BTWC"
48. Lockwood (2009) "Six-legged soldiers, using insects as weapons of war"
49. Lockwood (2012) "Insects as weapons of war, terror, and torture" *Annu. Rev. Entomol*
50. Lucas (1980) "The war against blue mold" *Science*
51. Manchee, Broster, Anderson et al. (1983) "Decontamination of Bacillus anthracis on Gruinard Island?" *Nature*
52. Manchee, Broster, Melling et al. (1981) "Bacillus anthracis on Gruinard Island" *Nature*
53. (1950) "Materials on the trial of former servicemen of the Japanese army charged with manufacturing and employing bacteriological weapons" *Materials*
54. Mauroni (2022) "On biological war" *Milit. Hist*
55. Mcelroy (1991) "The Geneva protocol of 1925"
56. Mcfee, Leikin (2009) "Death by polonium-210: Lessons learned from the murder of former Soviet spy Alexander Litvinenko" *Semin. Diagn. Pathol*
57. Nehring (2021) "Active and sharp measures. cooperation between the Soviet KGB and Bulgarian State Security" *J. Cold War Stud*
58. Nie (1955) "In the shadow of biological warfare: Conspiracy theories on the origins of COVID-19 and enhancing global governance of biosafety as a matter of urgency" *J. Bioeth. Inq*
59. Oldstone (2010) "Viruses, plagues, and history: Past, present, and future"
60. Pearson (2006) "The Iraqi biological weapons program"
61. Pischel, Martini, Yu et al. (2024) "Vaccine effectiveness of 3rd generation mpox vaccines against mpox and disease severity: A systematic review and meta-analysis" *Vaccine*
62. Prentiss (1937) "Chemicals in war: A treatise on chemical warfare"
63. Regis (1999) "The biology of doom: The history of America's secret germ warfare project"
64. Regis (2023) "Project 112"
65. Robinson (1993) "Chemical-weapons proliferation in the Middle East"
66. (1971) "The rise of CB weapons. Stockholm International Institute for Peace and Conflict Research. The problem of chemical and biological warfare: A study of the historical, technical, military, legal, and political aspects of CBW, and possible disarmament measures"
67. Rolland (2024) "Under the microscope: China's evolving biotechnology ecosystem"
68. Salama, Hansell (2000) "Does intent equal capability? Al-Qaeda and weapons of mass destruction"
69. Salisbury, Dew (2023) "Murder on Waterloo Bridge: Placing the assassination of Georgi Markov in past and present context, 1970-2018"
70. Schillinger (2023) "Microbic mass destruction: Biological warfare and epidemic prevention in Republican China" *East Asian Sci. Technol. Soc. Int. J*
71. Schmaltz (2017) "Chemical weapons research on soldiers and concentration camp inmates in Nazi Germany"
72. Schmidt (2015) "Secret science: A century of poison warfare and human experiments"
73. Scholthof, Adkins, Czosnek et al. (2011) "China's biological warfare programme: An integrative study with special reference to biological weapons capabilities" *Mol. Plant Pathol*
74. Smithson, Spencer, Wilcox (1993) "Beijing on biohazards. Chinese experts on bioweapons nonproliferation issues" *Rev. Res. Med. Microbiol*
75. Spiers (1986) "Chemical warfare"
76. Spiers, Walker, Aggarwal et al. (2010) "Virus isolation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) for diagnostic and research purposes"
77. Thränert, Verlag, Trevan (1996) "Enhancing the biological weapons convention"
78. Tucker (2001) "Scourge: the once and future threat of smallpox"
79. Tucker (2006) "War of nerves: Chemical warfare from world war I to Al-Qaeda"
80. Tucker, Hunger, Radosavljevic et al. (2013) "The current bioweapons threat"
81. (2023) "Biodefence posture review"
82. (2022) "Office of Chemical and Biological Weapons Affairs. Mission statement"
83. (1977) "Office for Disarmament Affairs. (UNODA) Secretary-General's Mechanism for investigation of alleged use of chemical and biological weapons (UNSGM)"
84. Van Courtland Moon (2009) "The US biological weapons program"
85. Vázquez, Rodriguez (1999) *Plantas hospedantes de Trips palmi Karny (Thysanoptera: Thripidae) en Cuba. Fitosanidad*
86. Wheelis, Rózsa, Dando (2006) "Deadly cultures: Biological weapons since 1945"
87. Whitby (2002) "Anti-crop BW and the BTWC"
88. Williams, Wallace (1989) "Unit 731: Japan's secret biological warfare in World War II"
89. Willis (2009) "Landscape with dead sheep: What they did to Gruinard Island" *Med. Confl. Surv*
90. Willstätter (1965) "In memory of Fritz Haber 1965"
91. (2025) "Independent assessment of the origins of SARS-CoV-2. Scientific Advisory Group for the origins of novel pathogens (SAGO)"
92. Wu, Wu, Liu et al. (1999) "Cuban allegations of biological warfare by the United States: Assessing the evidence" *Crit. Rev. Microbiol*
93. Zilinskas (2000) "Agents of bioterrorism: Pathogens and their weaponization" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754545&blobtype=pdf | # Adherence to antiviral treatment among people living with chronic hepatitis B: A global survey
Suzanne Block, Yasmin Ibrahim, Chari Cohen
## Abstract
BACKGROUNDCurrent antiviral treatment for chronic hepatitis B can suppress viral replication and reduce the risk of cirrhosis and liver cancer. It requires lifelong daily medication, and long-term adherence is often cited as a concern when initiating treatment. Hepatitis B treatment adherence in the context of the patient's medical and life experiences remains underexplored.
AIMTo evaluate factors associated with adherence to hepatitis B oral antiviral treatment.
METHODSA global online survey was administered anonymously to adults (aged 18 years or older) living with chronic hepatitis B. A subsample of 614 individuals who reported being on hepatitis B treatment was included in the analysis. Indices for treatment affordability, healthcare service acceptability, and individual physical, psychological, and emotional functioning were constructed (Cronbach's alpha = 0.71-0.83). Data analysis was conducted using Stata/BE 17.0.
RESULTSOverall, 81% of respondents reported high adherence to hepatitis B treatment. Lower adherence was observed among individuals who identified as African or African American (P = 0.008). Among participants with low adherence, 60% cited affordability as a challenge (P = 0.068), 53% identified healthcare service acceptability as a challenge (P = 0.04), 79% described physical functioning as a challenge (P = 0.002), and 40.5% reported difficulties with psychological functioning (P = 0.55). Block SJ et al. Global study on hepatitis B treatment adherence WJV https://www.wjgnet.com 2 December 25, 2025 Volume 14 Issue 4
CONCLUSIONFindings demonstrate high treatment adherence, although access to and acceptability of healthcare services, as well as an individual's physical functioning challenges, appear to be related to low adherence.
## INTRODUCTION
Chronic hepatitis B (CHB) remains a global health issue as a major cause of serious liver disease, including cirrhosis and hepatocellular carcinoma, or primary liver cancer [1]. Hepatocellular carcinoma accounts for 80% of all liver cancer cases and ranks as the third most common cause of cancer death worldwide [1]. An estimated 254 million people live with CHB worldwide and are at heightened risk for such complications [1]. The burden of this disease varies by region, with the World Health Organization (WHO) Western Pacific region (97 million) and the African region (65 million) accounting for the majority of people chronically infected[2]. This is followed by the WHO South-East Asian and Eastern Mediterranean Regions, with 61 million and 15 million individuals, respectively. Prevalence is lowest in the WHO European Region and the Region of the Americas, with about 1% of the population living with CHB [1].
Current antiviral treatment can control viral replication and reduce the risk of serious liver disease for people living with hepatitis B (PLWHB) [3]. Without effective treatment, there is a variation in the cumulative risk of developing cirrhosis over five years, ranging from 8% to 20% [3]. However, taking antiviral treatment can be challenging. It is a longterm commitment that can include taking a daily pill for many years or indefinitely [3]. None of these treatments is curative, and most have potential side effects. Side effects range in severity and may include fatigue, decreased mineral bone density, or renal impairment [4]. Under most circumstances, these medications should not be stopped, as discontinuation or non-adherence with antiviral therapy can result in acute hepatitis flares or incomplete viral control [3]. Adherence to antiviral treatments is, therefore, critical for PLWHB.
Adherence is often poor for individuals living with chronic conditions, who tend to require long-term oral antiviral treatment [5,6]. Treatment adherence is defined as the extent of agreement between an individual's behavior and a healthcare provider's recommendations [7]. These are health behaviors that extend beyond taking prescribed medication. Adherence is a dynamic process between a patient and provider, who collaboratively create a treatment plan that fits the medical advice from the health provider and the patient's care preferences, lifestyle, and personal values. For individuals who do not adhere to prescribed medication guidelines, the reasons may be intentional or unintentional, with heterogeneity in the social, economic, healthcare system, and patient-and disease-specific factors that shape their adherence behaviors [6]. In consequence, multiple social-ecological factors can impede adherence globally [6,8].
Ensuring treatment adherence is crucial for PLWHB, given the long-term nature of treatment and potential negative outcomes of prematurely stopping. Common barriers to treatment adherence for PLWHB include the cost of treatments, ease of obtaining medications, including refills, limited knowledge about treatment options, and concerns about side effects [9][10][11]. In a study of 308 participants with CHB in Wuhan, China, 48.7% reported financial constraints and 45.1% reported unintentional non-adherence, particularly forgetfulness, as barriers to oral antiviral treatment adherence [9]. Another study highlighted the financial burden along with the importance of patient-friendly information [10]. In contrast, concerns about disease progression, positive patient-provider interactions, including easy-to-understand language and time spent with providers, and social support were found to facilitate treatment adherence [9,12,13]. Fear about progressive liver disease has been described as a motivator for adherence to antiviral therapy, alongside the benefits of high-quality clinical encounters with the use of plain language by health providers to ensure clarity and understanding [12]. This reflects the multi-layered barriers that PLWHB face within and outside the healthcare system, which influence whether someone has the capacity to adhere to a treatment regimen.
Examining barriers and enablers, ranging from economic to intrapersonal, can help clarify our understanding of patients' capacity to adhere to prescribed treatment regimens. This focus on treatment adherence in the context of the multiple dimensions that shape health behaviors remains underexplored for PLWHB. Accordingly, we sought to add to this ongoing and critical discussion by examining treatment adherence and its association with diverse barriers faced by PLWHB. Results can be used to inform discussions about the expansion of CHB treatment and to develop future tools and resources that support long-term treatment adherence among PLWHB.
## MATERIALS AND METHODS
## Study measures and population
Data for this study were collected through an anonymous global online survey of people living with CHB. The survey tool and recruitment strategies were previously described [14,15]. Eligibility criteria included self-reporting to be 18 years or older and living with CHB. Survey respondents were asked if they were currently taking medication for their hepatitis B. Those who indicated experience with hepatitis B medication were further asked, "How often do you miss taking your hepatitis B medication?" Only respondents who answered that they were currently taking oral antiviral medication (tenofovir disoproxil fumarate, tenofovir alafenamide, or entecavir) for hepatitis B and answered this specific question about medication adherence were included in this analysis (n = 614).
## Ethical approval
Institutional Review Board approval was obtained before conducting data collection (No. 191221-270). All survey responses were anonymous, and no personal identifying information was collected.
## Data processing
The primary outcome of interest was treatment adherence, captured by participant responses to how often they missed their prescribed hepatitis B antiviral medication. Responses were collapsed into binary outcomes, such that those who answered that they missed their medication every other day or once a week were categorized as reporting low treatment adherence, and those who answered that they missed their medications once a month, a few times a year, or never were categorized as reporting high treatment adherence. The decision to collapse this variable into a binary outcome was informed by expert consultation with virologists and hepatologists based on the potential virological and/or clinical outcomes associated with these various instances of missing medication doses (Email, January 2024).
The primary predictors were assessed via survey questions relevant to the aspects of treatment adherence among PLWHB as described in the literature. Questions related to affordability, acceptability of healthcare services, and individual functioning can be found in Table 1. Additional covariates included age in years (18-30, 31-45, and ≥ 46), gender identity (male/female), ethnicity (White/Other, African/African American, Asian, Asian American, Native Hawaiian, and Pacific Islander), educational attainment (high school or less, technical/vocational/some college, college graduate, postgraduate), and whether one lives in a large city (yes/no).
## Index creation
Relevant survey items were selected to create the treatment adherence indices on affordability, healthcare service acceptability, and individual physical, psychological, and emotional functioning (Table 2). All survey questions used to compose the indices were 5-point Likert scale items, with response options ranging from 1 to 5, with 1 being not at all challenging to 5 being extremely challenging. Each index was calculated as the average of the 5-point scale scores for the variables included in the index. To facilitate analysis, indices were transformed into binary variables ("challenging" vs "not challenging") by calculating the sum score of each index and using the mean value of each score as the cutoff value. Values above the mean indicated "challenging" (indicating that affordability, acceptability, or accessibility posed a challenge to healthcare access), while values below the mean were categorized as "not challenging" (indicating no significant challenge regarding healthcare access) [14,16]. Reliability tests were conducted to confirm that the indices were "fit for use", with all Cronbach's alpha values exceeding 0.7 (ranging from 0.71 to 0.83) [17]. The specific Cronbach's alpha values for each index are detailed in Table 2.
## Statistical analysis
Statistical analysis was performed using Stata/BE 17.0 (Stata, no date). χ 2 analyses examined the relationship between the covariates of interest, including sociodemographic characteristics, and the index responses. A P-value less than or equal to 0.05 was considered a statistically significant result. The statistical analysis was overseen by Ibrahim Y, who has expertise in biomedical statistics.
## RESULTS
The final sample included 614 respondents, with some reporting high treatment adherence (missed their medication once a month/a few times a year/never) (n = 496) and others reporting low treatment adherence (missed their medication once a week/every other day) (n = 118). The sociodemographic characteristics of survey respondents, overall and stratified by treatment adherence, are presented in Table 3. Overall, treatment adherence among survey respondents was Not wanting to take a pill every day because it reminds me of my chronic condition considerably high (80.8%). Just under half (47.2%) of all respondents fell within the 31-45 age range, and 78.2% identified as male. Half (52.4%) identified as African or African American and were from the African WHO Region (48.5%). The majority of respondents were also either college graduates (40.7%) or postgraduates (38.4%). Just over half of all respondents (52.6%) reported living in a large city. Lower levels of treatment adherence were reported among respondents who identified as African/African American (P = 0.008). None of the other demographic variables showed statistically significant relationships with treatment adherence. While not significant, respondents aged 18-30 and 31-45, from Western Pacific, Middle Eastern, and African WHO Regions, or those with higher education levels, reported lower adherence to hepatitis B medication. Table 4 shows the relationships between the indices and treatment adherence. Among all survey respondents (n = 614), half (52.7%) said that the affordability of their medical needs, including the cost of blood tests and ultrasounds, doctor visits, and medications, was challenging, and 44.5% said healthcare service acceptability, such as finding a doctor who could manage their diagnosis and going for check-ups biannually, was challenging. For psychological functioning, which pertains to one's worries and fears surrounding their diagnosis, 38.1% of respondents reported that this aspect was challenging, and two-thirds (67.3%) said physical functioning was challenging. Regarding emotional accessibility, including feeling shame, feeling like others avoided them, feeling like life was less enjoyable, and not wanting to take a daily medication because of the reminder of their diagnosis, 36.7% of all respondents said this was challenging.
For individuals who reported low treatment adherence to their hepatitis B medication (n = 118), 60.3% said affordability was challenging, compared to 50.9% of those who reported high treatment adherence (n = 496) (P = 0.068). Service acceptability was challenging for just over half (53%) of those who reported low treatment adherence compared to 42.5% Not wanting to take a pill every day because it reminds me of my chronic condition of those who reported high treatment adherence (P = 0.04). The psychological functioning index indicates that 40.5% of respondents who reported low treatment adherence found these factors challenging, compared to 37.5% of respondents who reported high treatment adherence (P = 0.55). The majority of respondents who reported low treatment adherence (79.3%) found physical functioning to be challenging, compared to 64.5% of those who reported high treatment adherence (P = 0.002). Lastly, 38.5% of those who reported low treatment adherence reported that emotional functioning was challenging compared to 36.2% of those who reported high treatment adherence (P = 0.65).
## DISCUSSION
This study aimed to better understand how well people adhere to hepatitis B antiviral treatment and the diverse barriers that influence treatment adherence among PLWHB. Potential barriers examined included affordability, service acceptability, and an individual's physical, emotional, and psychological well-being. Despite various treatment challenges identified in this study and previous research [10,11], adherence to daily antiviral treatment was notably high among survey respondents, regardless of age, gender, educational status, or geographic location. This finding aligns with previous studies [18,19].
## Affordability of care
Affordability of healthcare is a common concern for people with chronic diseases who must pay for treatments, transportation to appointments, and other related costs. This study found that over half of all respondents found affordability challenging, and it was a reported challenge for a higher percentage (60.3% vs 50.9%) of those who reported low adherence. While these indices were not significantly associated with treatment adherence, the findings demonstrate the continued barrier of treatment costs, a finding consistent with other studies [9].
## Service acceptability and the role of the healthcare system
The acceptability of services, including finding a doctor who knows how to manage hepatitis B, and one's capacity to attend medical check-ups every six months, were significantly associated with treatment adherence in this study. Among survey respondents with low adherence, half reported challenges with service acceptability. This underscores how limited access to providers knowledgeable about hepatitis B or inconsistent medical management may serve as barriers to maintaining a treatment regimen. Finding a health provider who is equipped to manage hepatitis B remains an ongoing 1 Treatment adherence is defined by participant reports of how often they miss their hepatitis B medication. Missing medication every other day or once a week were categorized as low treatment adherence and missing medications once a month, a few times a year, and never were categorized as high treatment adherence.
2 AANHPI (Asian, Asian American, Native Hawaiian, and Pacific Islanders). 3 North America and South America combined due to small South American sample size (1% of sample).
WHO: World Health Organization.
issue, with one study based in the United States revealing that 80% of the physicians and medical residents in training surveyed did not feel adequately prepared to care for this patient population [20]. Studies in other countries have shown varying levels of health provider knowledge and awareness regarding hepatitis management, some citing inadequate medical training [21,22]. PLWHB have also expressed that health facilities were unable to meet their needs, including education on their diagnosis, decisions on treatment initiation, and instructions for taking their medications [12,23]. This highlights the role of infrastructural determinants of healthcare, which consists of the healthcare facilities, systems, and health providers and staff who are adequately trained and prepared to care for their patients [24,25]. This infrastructure can shape a patient's health experience and their preparedness to adhere to a treatment regimen.
## The relationship between treatment adherence and physical, psychological, and emotional functioning
Physical functioning, including feelings of exhaustion, unproductivity, and difficulty managing fatigue, was also significantly associated with treatment adherence. While almost 80% of those with lower treatment adherence reported this as a challenge, two-thirds of those with high treatment adherence did, as well. This suggests that the physical burden of CHB is more impactful for PLWHB than is often recognized. Among PLWHB, physical symptoms such as fatigue and muscle pain are commonly reported manifestations of hepatitis B [26]. Given the vast majority of PLWHB reporting physical functioning as a challenge to treatment adherence, we must better understand this as a barrier.
Although patient-reported outcomes are not currently central to treatment initiation decisions for hepatitis B, the observed association between better physical functioning and higher treatment adherence suggests that the potential benefits of early treatment initiation merit further exploration. Notably, physical symptoms may also arise as a side effect of antiviral therapy [4]. Some studies have reported that PLWHB may unintentionally miss doses due to these symptoms [12], while others report that individuals may avoid their medication because of concerns about potential side effects [9]. Distinguishing whether physical symptoms stem from the disease itself or are induced by treatment and understanding the limited evidence on whether antiviral therapy improves physical functioning highlights the complexity of treatment adherence. This must be addressed through open, supportive communication between patients and their healthcare providers to ensure individualized, informed treatment strategies.
Lastly, while over one-third of study participants reported challenges with psychological and emotional functioning due to CHB, these challenges were not significantly related to treatment adherence in this study. Thus, while these challenges may not be directly related to CHB treatment adherence in this study, they are still commonly reported as negatively impacting the quality of life for PLWHB [11,27]. Additionally, in contrast to this study, others have shown that viral suppression due to antiviral medication may improve patient-reported outcomes, including mental health [28,29]. Thus, continued attention to these influences is essential when considering treatment decisions, treatment adherence, and the overall health and well-being of PLWHB.
## The role of adherence in the CHB treatment paradigm
The treatment paradigm for CHB has diverged from other chronic conditions that also depend on long-term daily oral treatment, including hypertension, diabetes, and human immunodeficiency virus. For these other conditions, oral medication is widely, if not universally, recommended [30][31][32][33]. For CHB, oral medication is recommended only for a subset of the impacted population, based primarily on evidence of liver damage and risk of disease progression [3,4,34,35]. While there is little rigorous data documenting low adherence to antiviral regimens among PLWHB, adherence concerns are widely discussed as a reason to limit treatment eligibility. However, when adherence has been studied, findings often demonstrate relatively high levels of adherence, with some variability [9,18,36,37], and low levels (as low as 1%) of adverse outcomes among people who stop antiviral treatment [38]. Thus, there appears to be a disconnect between perceived and real treatment adherence concerns. Are we problematizing medication adherence for CHB, and can we reframe adherence from serving as a treatment barrier to becoming a treatment enabler?
Looking at human immunodeficiency virus, diabetes, and hypertension as examples, treatment adherence guidelines recognize socioecological factors related to adherence [30][31][32][33]. The treatment practice guidelines for these conditions focus heavily on offering provider-and patient-focused guidance for maximizing adherence and promoting interventions that provide patients with resources and support to improve adherence [30][31][32][33]. In doing this, they have reframed the focus from nonadherence as a barrier to recommending treatment to optimizing adherence as a key component of treatment.
This has fostered broader treatment recommendations and led to the implementation of adherence-focused interventions [39][40][41]. We can learn from these models to reframe adherence in relation to the CHB treatment paradigm. A better understanding of adherence, as well as the barriers and facilitators, is critical as we continue to work towards global viral hepatitis elimination [42]. This is especially vital at a time when professional treatment guidelines are expanding the pool of PLWHB who are eligible to receive oral antiviral treatment, and as there are growing discussions on whether to further expand hepatitis B treatment more universally[34, [43][44][45].
## Study strengths and limitations
This study focuses on a relatively under-researched area -adherence to hepatitis B treatment. This global study of PLWHB incorporates patient-reported experiences and adherence to hepatitis B oral antiviral treatments, providing insights that are often underrepresented in clinical literature. These findings provide valuable contributions to shaping patient-centered approaches for improving hepatitis B care and adherence. Still, study limitations must be acknowledged. Reliance on self-reported data may introduce both recall and social desirability biases. This could affect the accuracy of reported adherence behaviors. Additionally, the voluntary nature of survey participation raises the possibility of selfselection bias, whereby individuals who are more engaged or informed about their condition may be overrepresented in the results. The survey also did not collect information on participant income, and, instead, educational attainment was used as a proxy for socioeconomic status. Although not a perfect substitute, education can offer valuable insights into socioeconomic status and its potential influence on treatment adherence.
The sample was predominantly male, potentially limiting generalizability. This may be partially attributed to disparities in internet access, both in general and particularly among women, in many low-and middle-income countries [46,47]. Structural barriers to healthcare access that disproportionately affect women in these settings may also have influenced the likelihood of being on treatment, thereby impacting the final sample of respondents from low-and middle-income countries who reported receiving antiviral therapy. Finally, the survey was administered only in English, which may have limited participation from individuals who do not speak English. Despite this, the study achieved strong geographic diversity, with responses representing a wide range of countries and regions, enhancing the global relevance of the findings.
## CONCLUSION
Treatment adherence is a multifaceted and complex issue that must be understood within the broader context of interactions among patients, providers, and the health system. Despite its critical importance, adherence to hepatitis B oral antiviral treatment remains underexplored. This study found high rates of treatment adherence among study participants and identified affordability, access to acceptable healthcare services, and physical functioning as challenges to adherence. Study results can inform discussions and help develop future tools and resources to support long-term treatment adherence among PLWHB. More research is needed to better understand CHB treatment adherence, including the social-ecological factors associated with low adherence, and how treatment adherence is impacted by co-morbidities, patient preferences, patient-reported outcomes, and patients' perceived value of treatment. Exploring these constructs will help design interventions and resources for both patients and providers that will support PLWHB and enable treatment adherence. It will also help us better understand the social and economic dimensions of CHB care and treatment access, thereby fostering policy change to break down these barriers.
## References
1. (2024) "Global hepatitis report 2024: action for access in low-and middle-income countries"
2. (2024) "Guidelines for the prevention, diagnosis, care and treatment for people with chronic hepatitis B infection. Geneva: World Health Organization"
3. Terrault, Lok, Mcmahon et al. (2018) "Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance" *Hepatology*
4. Fu, Hsieh, Chen et al. (2022) "Association between medication adherence and disease outcomes in patients with hepatitis B-related cirrhosis: a population-based case-control study" *BMJ Open*
5. Kvarnström, Westerholm, Airaksinen et al. (2021) "Factors Contributing to Medication Adherence in Patients with a Chronic Condition: A Scoping Review of Qualitative Research" *Pharmaceutics*
6. (2003) "Adherence to long-term therapies: evidence for action"
7. Im, Mohammed, Lumley et al. (2024) "Social, clinical and biological barriers to hepatitis B virus suppression with nucleos/tide analogue therapy: who is at risk and what should we do about it?" *Sex Transm Infect*
8. Xu, Liu, Farazi et al. (2018) "Adherence and perceived barriers to oral antiviral therapy for chronic hepatitis B" *Glob Health Action*
9. Jackson, Ibrahim, Freeland et al. (2024) "Barriers to accessing hepatitis B medication: a qualitative study from the USA and Canada" *BMJ Open*
10. Hepatitis B Foundation ; S, Zablotska-Manos, Zekry et al. (2017) "Adherence to Hepatitis B Antiviral Therapy: A Qualitative Study" *Gastroenterol Nurs*
11. Ibrahim, Zovich, Ansah et al. (2024) "Quality of life of people living with chronic hepatitis B: The role of social support system" *PLOS Glob Public Health*
12. Ibrahim, Umstead, Wang et al. (2023) "The Impact of Living With Chronic Hepatitis B on Quality of Life: Implications for Clinical Management" *J Patient Exp*
13. Ibrahim, Cohen, Araojo et al. (2022) "Attitudes towards clinical trial participation among people living with chronic hepatitis B" *J Transl Sci*
14. Jeong, Lee (2016) "The level of collapse we are allowed: comparison of different response scales in safety attitudes questionnaire" *Biom Biostat Int J*
15. Song, Lin, Ward et al. (2013) "Composite variables: when and how" *Nurs Res*
16. Block (2025) "Global study on hepatitis B treatment adherence WJV"
17. Alpern, Joo, Bahr et al. (2023) "Factors Associated With Adherence to First-line Antiviral Therapy Among Commercially Insured Patients With Chronic Hepatitis B" *Open Forum Infect Dis*
18. Chotiyaputta, Peterson, Ditah et al. (2011) "Persistence and adherence to nucleos(t)ide analogue treatment for chronic hepatitis B" *J Hepatol*
19. Pham, Toy, Hutton et al. (2016) "Gaps and Disparities in Chronic Hepatitis B Monitoring and Treatment in the United States" *Med Care*
20. Mashilo, Mompati, Ramakatane et al. (2025) "attitudes and practices to hepatitis B among South African primary healthcare staff" *Afr J Prim Health Care Fam Med*
21. Nusair, Rayyan, Hammoudeh et al. (2020) "Hepatitis B care pathway in Jordan: current situation, gaps and recommended actions" *J Virus Erad*
22. Adjei, Stutterheim, Naab et al. (2019) "Barriers to chronic Hepatitis B treatment and care in Ghana: A qualitative study with people with Hepatitis B and healthcare providers" *PLoS One*
23. Luxon (2015) "Infrastructure -the key to healthcare improvement" *Future Hosp J*
24. Rouzbehani (2019) "Beyond Digital Tools: A Transdisciplinary Approach to Healthcare"
25. Evon, Wahed, Johnson et al. (2016) "Fatigue in Patients with Chronic Hepatitis B Living in North America: Results from the Hepatitis B Research Network (HBRN)" *Dig Dis Sci*
26. Low, Ge, Yeong et al. (2025) "Burden of psychological symptoms and disorders among individuals with hepatitis B: a systematic review, meta-analysis and meta-regression" *Front Psychiatry*
27. Younossi, Stepanova, Younossi et al. (2019) "Patientreported outcomes in patients chronic viral hepatitis without cirrhosis: The impact of hepatitis B and C viral replication" *Liver Int*
28. Younossi, Stepanova, Janssen et al. (2018) "Effects of Treatment of Chronic Hepatitis B Virus Infection on Patient-Reported Outcomes" *Clin Gastroenterol Hepatol*
29. Unger, Borghi, Charchar et al. (2020) "International Society of Hypertension Global Hypertension Practice Guidelines"
30. (2025) "Introduction and Methodology: Standards of Care in Diabetes-2025"
31. Hhs Panel On (2024) "Antiretroviral Guidelines for Adults and Adolescents-A Working Group of the NIH Office of AIDS Research Advisory Council (OARAC)"
32. (2021) "Consolidated guidelines on HIV prevention, testing, treatment, service delivery and monitoring: Recommendations for a public health approach"
33. (2025) "34 European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatitis B virus infection" *J Hepatol*
34. Lau, Yu, Wong et al. (2021) "APASL clinical practice guideline on hepatitis B reactivation related to the use of immunosuppressive therapy" *Hepatol Int*
35. Sawalmeh, Johansson, Kostas et al. (2025) "Hepatitis B Patients' Adherence to Treatment in Relation to Knowledge, Attitudes, and Practices (KAP) in the West Bank, Palestine, 2022-2023" *J Viral Hepat*
36. Afolabi, Aremu, Koryom et al. (2025) "Predictors of Medication Non-Adherence Among Hepatitis B Patients in South Sudan: A Health-Facility-Based Cross-Sectional Study" *Patient Prefer Adherence*
37. Tseng, Chen, Wu et al. (2023) "Serious adverse events after cessation of nucleos(t)ide analogues in individuals with chronic hepatitis B: A systematic review and meta-analysis" *JHEP Rep*
38. Whiteley, Olsen, Haubrick et al. (2021) "A Review of Interventions to Enhance HIV Medication Adherence" *Curr HIV/AIDS Rep*
39. Sapkota, Brien, Greenfield et al. (2015) "A systematic review of interventions addressing adherence to anti-diabetic medications in patients with type 2 diabetes--impact on adherence" *PLoS One*
40. Conn, Ruppar, Chase et al. (2015) "World Health Organization. Global progress report on HIV, viral hepatitis and sexually transmitted infections, 2021. Accountability for the global health sector strategies 2016-2021: actions for impact" *Curr Hypertens Rep*
41. Dato, Iorio (2024) "Expanding indications for chronic hepatitis B treatment: Is it really desirable to treat everyone?" *World J Gastroenterol*
42. Mcnaughton, Lemoine, Van Rensburg et al. (2021) "Extending treatment eligibility for chronic hepatitis B virus infection" *Nat Rev Gastroenterol Hepatol*
43. Jeng, Lok (2021) "Should Treatment Indications for Chronic Hepatitis B Be Expanded?" *Clin Gastroenterol Hepatol*
44. Gillwald, Partridge (2022) "Gendered nature of digital inequality: Evidence for policy considerations"
45. Gsma (2025) |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12381284&blobtype=pdf | # Interferon-stimulated gene screening identifies CCND3 as a host restriction factor against emerging high-pathogenic bandaviruses
Zhao Xu, Zhenyu Jiang, Kuan Feng, Haiyan Zhang, Chen Shi, Fei Deng, Hualin Wang, Yun-Jia Ning
## Abstract
Severe fever with thrombocytopenia syndrome virus (SFTSV) is a representative high-pathogenic bandavirus (Bandavirus genus, Phenuiviridae family). Inducible expression of interferon-stimulated genes (ISGs) is the foundation of host antiviral defense; however, their roles in bandavirus infection remain elusive. Here, we identify over 200 ISGs potentially inhibiting or promoting bandaviral replication. With SFTSV as the main model, we further systematically uncover the notable antiviral role of one ISG, cyclin D3 (CCND3), against bandaviruses. SFTSV infection induces CCND3 up-regulation and cytoplasmic translocation. CCND3, in turn, inhibits the viral replication in cultured cells and pathogenicity in vivo. The viral nucleoprotein (NP) is the target of CCND3. By its CN domain, CCND3 interacts with NP's "head" region in an RNA-independent manner, suppressing the ribonucleoprotein (RNP) replication machinery activity. Furthermore, consistent with interaction interface mapping and structural modeling analyses, the CCND3-NP interaction blocks NP multimerization, NP-RNA binding, and NP association with viral polymerase, that is, the NP activities essential to RNP construction and functioning. Conversely, the viral nonstructural protein, NSs, can partially antagonize CCND3 by attenuating its induction and promoting autophagic degradation. These findings provide new insights into bandavirus-host interactions and arms race, advancing the understanding of bandavirus infection and probably informing antiviral therapeutic development.Severe fever with thrombocytopenia syndrome virus (SFTSV), also referred to as Dabie bandavirus (DBV), is the causative agent of severe fever with thrombocytopenia syndrome (SFTS), an emerging and lifethreatening infectious disease 1,2 . SFTS is characterized by clinical manifestations such as fever, severe thrombocytopenia, leukopenia, and gastrointestinal symptoms, with a high case-fatality rate of up to 30% 1,2 . The disease is transmittable through tick bites and human-tohuman contact, with its epidemic areas expanding in China and neighboring countries [3][4][5][6] . Following the identification of SFTSV in China in 2009, several other tick-borne viruses genetically related to
SFTSV, such as Heartland virus (HRTV, identified in the United States) and Guertu virus (GTV, discovered in China), were successively isolated around the world 7,8 . Recently, they have been classified into a new virus genus Bandavirus (Phenuiviridae family) 9 . There is no licensed antiviral drug or vaccine currently available against them. These bandaviruses, with SFTSV as the highly pathogenic representative, pose a substantial threat to public health, necessitating urgent research.
Bandaviruses are enveloped RNA viruses with a negative-sense, single-stranded RNA genome consisting of three segments. The large (L) and medium (M) genomic segments respectively encode the viral RNA-dependent RNA polymerase (RdRp, i.e., L protein) and envelope glycoproteins (GP), while the small (S) segment encodes the nucleoprotein (NP) and a nonstructural protein (NSs) in an ambisense manner 1 . NP is a major structural protein component of the virions [10][11][12] . Through multimerization and complex interactions with viral RNAs and L protein, NP plays crucial roles in virus replication by driving the assembly and participating in the function of the viral ribonucleoprotein complex (RNP) [11][12][13][14] . RNP is the molecular machinery for transcription and replication [11][12][13] . Moreover, RNPs encapsulating viral genomic RNAs also can be packaged into progeny virions as the structural core 11 . In comparison, as a nonstructural protein, bandavirus NSs is not essential to the viral replication but act as an important virulence factor by interfering with multiple host biological processes and in particular, antiviral interferon (IFN) responses [15][16][17][18][19][20][21] .
Type I IFN response constitutes the first vital line of host defense against various viral infection and pathogenicity by inducing IFNstimulated gene (ISG) expression 22,23 . This host antiviral response is initiated by the recognition of viral infections by pattern recognition receptors (PRRs) 24,25 . During bandavirus infection, retinoic acidinducible gene I (RIG-I)-like receptors (RLRs) and several Toll-like receptors (TLRs) have proved important for the host recognition in previous studies by us and others 20,26 . Upon recognition, PRRs activate the downstream signaling cascades for IFN induction, leading to the expression of antiviral IFNs, especially type I IFNs 24,27 . The produced IFNs then bind to their cell surface receptors and trigger the Janus kinase-signal transducer and activator of transcription (JAK-STAT) pathway, inducing expression of more than 300 ISGs 28,29 . The induction of these ISGs establishes a strong host antiviral state, although the action mechanisms of most ISGs remain elusive 30,31 . Many studies have validated that type I IFN system is also critical to restrict bandavirus replication and pathogenicity 15,18,26,[32][33][34][35][36][37] . However, it remains poorly understood which and how ISGs can regulate bandaviruses as IFN effectors.
Here, we analyzed the regulatory potential of ISGs in SFTSV replication by using an ISG cDNA expression library and a previously established minireplicon reporter system [38][39][40] . Numerous ISGs that likely inhibit or promote the viral replication machinery RNP-associated stage were thus identified, presenting plenty of new clues for further elucidation of the virus-host interactions. Subsequently, we systematically validated the remarkable role of a representative ISG, Cyclin D3 (CCND3), in restricting bandavirus infection and pathogenicity in vitro and in vivo, by using series of loss/gain-of-function assays and SFTSV as the main virus model. Furthermore, we uncovered the detailed molecular mechanism underlying CCND3 restriction of SFTSV, proposing a new antiviral mode employed by CCND3. Additionally, potential antagonizing effects of SFTSV NSs on CCND3 were also analyzed. The findings provide insights into the bandavirus-host interactions, better understanding of which may help advance the development of antiviral therapy against SFTSV and related bandaviruses.
## Results
## A minireplicon reporter system-based screen for ISGs regulating SFTSV replication
Antiviral IFN response is a pivotal host defense mechanism against bandaviral infection [32][33][34][35] . However, the specific antiviral potential of ISGs as the IFN effectors against bandaviruses including SFTSV remains to be explored. Therefore, we screened individual ISGs for their ability to regulate SFTSV replication using a cDNA clone library of different ISGs combined with an SFTSV minireplicon reporter system (Fig. 1a). As previously described [38][39][40] , the minireplicon reporter system constructing viral RNP machinery can indicate the efficiency of the central events in virus life cycle, including RNP assembly and function on driving transcription and replication. Besides, the system facilitated high-throughput screening without the requirements of high-level biosafety facilitates. As shown in the dot plot of mean values from relative reporter activities (Fig. 1b), expression of most ISGs likely inhibited the viral replicon to some extent (averaging ~30% reduction, as indicated by the gray horizontal line), yet a small subset seemed to enhance it. Consistently, MOV10 and MxA that have been reported to inhibit SFTSV infection 39,40 , were also identified in this screen (Fig. 1b). Furthermore, volcano plot analysis reveals that 234 and 18 ISGs, respectively, inhibit or promote the replicon activity by more than 20%, with adjusted p values of less than 0.05 (Fig. 1c and Supplementary Data 1). Therein, CCND3 was recognized as one of the top candidates with strong inhibitory activities and high statistical significance (Fig. 1b,c). CCND3 is known as a classical protein involved in regulation of cell cycle 41,42 , whereas its role as an ISG in response to virus infection remains to be further explored. Moreover, there was no report regarding its involvement in bandavirus infection before this study. Deciphering the mechanisms of understudied ISGs like CCND3 in viral infection is imperative for understanding host-virus interplays. We thus selected CCND3 as a candidate for following in-depth functionmechanism analyses.
## CCND3 expression and cytoplasmic translocation are enhanced upon SFTSV infection
To better characterize the potential interplays between CCND3 and SFTSV, we first assessed CCND3 expression levels and subcellular localization upon SFTSV infection. Several cell lines previously established as permissive models for SFTSV research were infected with the virus and subjected to analysis of CCND3 expression at different time points. CCND3 mRNA expression was evidently stimulated in all the tested cell types upon SFTSV infection (Fig. 2a-c). Similarly, transcriptional upregulation of CCND3 by SFTSV infection was also observed in isolated primary cells including human peripheral blood mononuclear cells (PBMCs) and mouse bone marrow-derived macrophages (BMDMs) (Fig. 2d,e). Subsequently, the effects of SFTSV infection on CCND3 mRNA transcription were further tested in mice in vivo. Consistently, transient induction of CCND3 expression was detected upon the viral infection in the mouse tissues (Supplementary Fig. S1).
Then, cell immunofluorescence assays (IFA) showed that CCND3 was predominantly localized in nuclei in the mock-infected group, while SFTSV infection increased the abundance of CCND3 in the cytoplasm at a low multiplicity of infection (MOI, 0.5) (Fig. 2f). Moreover, an overall increase of CCND3 signals could be more evidently observed in both cytoplasm and nuclei at a higher infection dose (MOI, 2) (Supplementary Fig. S2). To further investigate the effect, we analyzed the subcellular localization of CCND3 by nuclear-cytoplasmic fractionation. Total protein analyses with whole cell lysates (WCL) confirmed that CCND3 expression was dose-dependently stimulated by SFTSV infection (Fig. 2g). Consistently, CCND3 was mainly nuclear in the resting state (mock-infected group), but evidently accumulated in the cytoplasm following SFTSV infection (Fig. 2h,i). Slight increase of CCND3 abundance in the nuclei could also be observed with SFTSV infection at high MOIs, but to a much lesser extent compared with that in the cytoplasm (Fig. 2h,i). Given that SFTSV is a cytoplasmic RNA virus replicating exclusively in cytoplasm, the enhancement of CCND3 expression and cytoplasmic localization could be closely associated with the potential regulatory role of the host protein in SFTSV replication.
## CCND3 acts as a notable host restriction factor against SFTSV and related bandaviruses
To further confirm the role of CCND3 in modulating SFTSV replication, we next conducted series of loss/gain-of-function assays to evaluate the impact of CCND3 on SFTSV infection. First, CCND3 overexpression indeed significantly inhibited SFTSV replication and propagation (Fig. 3a-e). Conversely, knockdown (KD) of CCND3 by RNAi evidently augmented SFTSV replication and progeny production (Fig. 3f-k). Furthermore, CCND3 knockout (KO) cells were generated by the CRISPR-Cas9 system (Supplementary Fig. S3a) and used for the following analyses of SFTSV replication kinetics. Consistently, CCND3 deletion resulted in marked increase in SFTSV RNA replication, protein expression, and progeny propagation (Fig. 3l-p), corroborating the notable anti-SFTSV role of CCND3.
In addition, the influence of CCND3 on replication of GTV and HRTV, two other bandaviruses genetically related to SFTSV, was also investigated using aforementioned experimental settings. Similarly, CCND3 overexpression dramatically repressed all the three segmentderived RNA replication and progeny propagation of both GTV (Fig. 4a,b) and HRTV (Fig. 4c,d) at different time points. In contrast, loss-of-function assays using the CCND3-KO cells demonstrated that absence of CCND3 led to significant enhancement of both GTV (Fig. 4e,f) and HRTV replication and propagation (Fig. 4g,h).
Collectively, these data reveal that CCND3 serves as a conserved host restriction factor against SFTSV and related bandaviruses, meriting further detailed function/mechanism elucidation.
## CCND3 deficiency increases the susceptibility of mice to SFTSV infection
CCND3, which is critically required for pre-TCR-driven expansion of immature thymocytes, plays an essential role in the normal development of the thymus and T cells in mice [43][44][45] . To investigate the effect of CCND3 on SFTSV infection and pathogenicity in vivo, we generated a CCND3-deficient adult mouse model by transient transduction of a lentiviral vector expressing CCND3-specific shRNA via intravenous injection, as previously conducted 39,46,47 . Although adult immunocompetent mice infected with SFTSV do not exhibit obvious clinical symptoms, SFTSV replication and transient tissue lesions can be detected in various organs 33,48 . Therefore, the knockdown mouse model has been effectively utilized to study the role of specific host factors that may regulate virus replication 39,46,47 . Consistent with previous reports, no severe clinical manifestations or deaths were observed in all infected mice (n = 6 per group). Thus, viral replication and pathogenicity in various tissues were further analyzed after sacrifice. Firstly, qPCR analysis confirmed a significant decrease of CCND3 expression in CCND3-shRNA treated mice (Fig. 5a). Then, viral RNA levels in the analyzed tissues including spleen, liver, and lung were all significantly higher in the CCND3-KD mice than those in control (Fig. 5b), suggesting the restriction of viral replication by CCND3. Consistently, the CCND3-deficient mice exhibited more severe viremia with higher viral copies in the sera (Fig. 5c). In line with these data, immunohistochemical analysis (IHC) detected more viral antigenpositive foci in the tissues of the CCND3-KD group (Supplementary Fig. S4), further supporting the restrictive effect of CCND3 on SFTSV infection. Thrombocytopenia, leukopenia, and elevated serum biochemical indicators of organ damage are the typical clinical features of SFTSV infection in humans 1,2 . Interestingly, a significant reduction of platelet counts was observed in the CCND3-KD group but not the control group upon SFTSV infection (Fig. 5d). Moreover, reduction of white blood cells (WBC) counts and increase of alanine a-e HEK293, THP-1, and MEF cells, as well as primary human PBMCs and mouse BMDMs, were infected with SFTSV (MOI = 5). CCND3 mRNA and SFTSV M RNA levels were subsequently analyzed by qPCR at the indicated time points postinfection. f Localization of endogenous CCND3 upon SFTSV infection. HEK293 cells infected with SFTSV (MOI, 0.5) were fixed at 24 h postinfection (hpi) and subjected to immunofluorescence assays (IFA) and confocal microscopy to visualize cellular CCND3 (red), SFTSV NP (green), and nuclei (blue). Scale bar, 20 μm. g-i Nuclear-cytoplasmic fractionation. THP-1 cells were infected with SFTSV at various MOIs for 24 h. Nuclear and cytoplasmic fractions were separated and protein levels of CCND3 and SFTSV NP were analyzed by Western blot (WB) using the indicated antibodies. Total protein levels from whole cell lysates (WCL) were shown in (g). HDAC1 and β-actin were served as nuclear and cytoplasmic markers and loading controls, respectively (g and h). Relative band intensities of cytoplasmic and nuclear CCND3 (over the corresponding loading controls) were quantified and normalized to the mock groups (i). Results are representative of three independent replicates (f and g). Data are presented as means ± SD, n = 3 biological replicates (a-e and h-i). Source data are provided as a Source Data file. aminotransferase (ALT), aspartate aminotransferase (AST), and blood urea nitrogen (UREA) levels caused by the viral infection were more severe after CCND3 depletion (Fig. 5d,e), indicating CCND3 restriction of the viral pathogenicity. Additionally, we examined the pathology of infected tissues through H&E staining. Compared to the control, CCND3-deficient mice infected with SFTSV indeed showed more pronounced histopathological abnormalities (Fig. 5f-h). These mainly included splenic nodules and blurring boundaries between the red pulp and white pulp, local coagulative necrosis of the liver tissues characterized with disappearance of nuclei and visible cytoplasmic fragmentation, and widening of alveolar septa accompanied by interstitial infiltration in the lung tissues (Fig. 5f-h). Together, these results support the notable role of CCND3 in restricting the viral replication and pathogenicity in vivo.
Unlike immunocompetent adult mice, IFN receptor-deficient mice (e.g., A129) can serve as a fully lethal animal model for SFTSV infection. We also utilized this model to complementally analyze the effect of CCND3 deficiency on viral infection and pathogenesis. Interestingly, in agreement with its anti-SFTSV role, CCND3 KD appeared to worsen SFTSV-induced lethality and weight loss (Supplementary Fig. S5a-b). Consistently, CCND3 KD resulted in higher SFTSV viremia and replication levels across tissues, intensified thrombocytopenia and leukopenia, and elevated biochemical markers of tissue injury (Supplementary Fig. S5c-g). Furthermore, histopathological analysis revealed aggravated tissue damage in CCND3-KD mice (Supplementary Fig. S5h-j). These findings further corroborate the restrictive role of CCND3 in SFTSV infection and pathogenesis.
The antiviral activity of CCND3 is independent of its cell cycle regulatory function and the IFN signaling pathway Next, we proceeded to methodically elucidate the molecular mechanisms by which CCND3 restricts bandavirus replication. Before investigating if CCND3 directly interferes with viral replication by targeting viral molecular machinery, we first analyzed whether its potential cell cycle regulatory role may indirectly contribute to its antiviral activity. Mutation at the CCND3 T283 can impair its phosphorylation and normal physiological role in cell cycle 49 . Interestingly, both CCND3 and its T283A mutant dramatically inhibited SFTSV infection and no significant difference was observed (Supplementary Fig. S6a-c). CCND3 participates in cell cycle regulation through the CCND3/CDK4/6 pathway 42,50 . Therefore, we then constructed CDK4-KO, CDK6-KO, and CDK4/6 double-KO cells and examined the CCND3 antiviral activity in these models. Consistently, deletion of CDK4, CDK6, or both did not impair CCND3's anti-SFTSV function (Supplementary Fig. S6d-f), demonstrating that the antiviral activity is independent of cycle regulation. Moreover, KO of these CDKs did not lead to any enhancement of the viral replication (Supplementary Fig. S6d-f), different with CCND3 KO (Fig. 3l-p), also supporting the irrelevance of CCND3's anti-SFTSV activity to the cycle regulation pathways.
Furthermore, we generated IFN receptor-KO cells to test whether the IFN antiviral signaling is directly required for CCND3 inhibition of viral replication. Interestingly, CCND3, as well as the T283A mutant, continued to exert significant inhibitory effects on SFTSV replication in the cells with IFN-signaling KO (Supplementary Fig. S6g-i). This indicates that the antiviral action of CCND3 is also independent of IFN signaling, although its expression can be induced by the signaling as an ISG. Together, these analyses suggest that CCND3 can exert antiviral activity independently of its cell cycle regulatory function and IFN signaling, implying that CCND3 may directly block viral replication by targeting critical viral machinery.
## CCND3 targets the viral NP protein to interfere with the RNP machinery
To further uncover the mechanism underlying the antiviral function of CCND3, we identified the potential viral protein targeted by CCND3. HEK293 cells were infected with SFTSV and subsequently subjected to co-immunoprecipitation (Co-IP) assays. Interestingly, NP, but not the other structural proteins, was specifically co-precipitated by CCND3 (Fig. 6a). Additionally, NSs also seemed slightly precipitated but to a much lesser extent. Since NP, not NSs, is the major structural component and plays an essential role in the viral replication, we then primarily focused on characterizing the CCND3-NP interaction. First, the interaction of CCND3 with NP was also validated in the following Flag-pulldown analysis using co-transfected cells (Fig. 6b). Then, we tested whether RNA is involved in the interaction, considering the RNA-binding capacity of NP. Results from the further pulldown assays combined with nuclease treatment 39,40 showed that the interaction between CCND3 and NP was likely independent of RNA (Fig. 6c). NP is the core component driving the assembly and participating the function of the replication machinery RNP. Therefore, the targeting of NP by CCND3 is also well linked to CCND3's inhibitory ability to the minireplicon system indicating the RNP activity as observed in the screening. To further corroborate the interference of CCND3 with the viral RNP activity, we conducted additional evaluation on dosedependent effects of CCND3 using two minigenome reporter systems based on EGFP or dual-luciferases, respectively. Consistently, CCND3 efficiently inhibited both the SFTSV EGFP (Fig. 6d-f) and dualluciferase based (Fig. 6g) minireplicons in dose-dependent manners, confirming the blockade of RNP activity by CCND3. These data uncover the direct antiviral action of CCND3 by targeting NP to interfere with viral RNP activity.
## CCND3 targets the N-and C-lobes of NP to inhibit the RNP activity and hence virus replication via its CN domain
To gain more mechanistic insights on how CCND3 targets NP to interfere with SFTSV replication, we further investigated the critical domains and potential interface involved in CCND3-NP interaction. Previous studies have elucidated the structure of NP, in which the protein is composed of a protruding N-arm and a compact "head" comprising an N-lobe and a C-lobe 10,51 . The conformation of CCND3 residues 23-254 was also defined in a crystal structure 50 . Interestingly, structural modeling by the deep learning-based AlphaFold2 program showed that CCND3 and NP potentially form a complex of high confidence (Fig. 7a). The CCND3 N (CN) domain and NP "head" involving both the N-and C-lobes seemed to play major roles in the interaction by constructing a potential interaction interface in the AlphaFold complex structure (Fig. 7a,b). The interaction domains were then experimentally examined by protein interaction analyses using a series of NP and CCND3 truncated mutants (Fig. 7c). As demonstrated in the S.tag pulldown (S-pulldown) results with S.tag-fused NP as the bait, coprecipitation with NP was deprived by the deletions of CCND3 Box (CBox) or even the smaller CN, but not CCND3 C (CC), suggesting that the CN domain is indeed involved in CCND3-NP interaction (Fig. 7d). Additionally, EGFP-NanoTrap assays with EGFP-fused NP mutants indicated that both the NP N-and C-lobes, but not N-arm, could interact with CCND3 (Fig. 7e), also in agreement with the structural modeling analysis. These data suggest that CCND3 likely targets both the N-and C-lobes of NP mainly via its CN domain.
Next, we tested the effects of the CCND3 mutants on SFTSV minireplicon reporter system. Consistently, removal of the CN or CBox domains abolished the inhibition to SFTSV RNP-driven reporter activity (Fig. 7f-h). Subsequently, influence of the mutation on SFTSV Fig. 3 | CCND3 restricts SFTSV replication and propagation. a-e CCND3 overexpression inhibits SFTSV infection. HEK293 cells were transfected with the CCND3 expression plasmid or control vector. At 24 h post transfection, the cells were infected with SFTSV (MOI = 0.1) and delivered to analyses of the RNA and protein levels at different time points by qPCR (a-c) and WB (d), respectively. The propagation of progeny released into culture medium (growth curve) was also analyzed by TCID 50 assay (e). f-k CCND3 knockdown (KD) by RNAi promotes SFTSV infection. HEK293 cells were transfected with the CCND3-targeting or control siRNAs and infected with SFTSV (MOI = 0.1) for 36 h, followed by determination of the RNA levels (f-i), protein expression (j), and viral progeny titers (k), respectively. l-p CCND3 knockout (KO) by CRISPR-Cas9 editing enhances SFTSV infection. CCND3 KO and control cells were infected with SFTSV (MOI = 0.1), followed by analyses of RNA (l-n) and protein (o) levels and progeny propagation (p) at different timepoints. WB results are representative of three independent replicates (d, j, o). Data are shown as means ± SD, n = 3 biological replicates (a-c, e-i, k-n, and p). Statistically significant differences are indicated (two-tailed Student's t test for each timepoint in a-c, e, l-n, and p; one-way ANOVA in f-i and k). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns non-significant. Source data are provided as a Source Data file.
replication was also examined: deletion of the CN or CBox (but not CC) resulted in abolishment of CCND3's anti-SFTSV activity (Fig. 7i-k). Furthermore, individual expression of CN, and the slightly larger CBox containing CN, could also target NP by protein interaction (Fig. 7l,m). In line with its NP-targeting ability, CN alone exhibited significant antiviral activities, effectively suppressing viral RNA replication, protein expression, and progeny production (Fig. 7n-p). In contrast, the CC domain neither interacted with NP nor displayed the antiviral effects (Fig. 7l-p). Altogether, these consistent data demonstrate that CCND3 targets the N-and C-lobes of NP to interfere with the RNP activity and viral replication by its CN domain.
Following the identification of CN, we further analyzed potential significant motifs or amino acid residues within CN. First, we performed sequential scanning mutagenesis by introducing alanine substitutions in a series of five-residue segments covering the CN domain, followed by functional screening using the NP/RNP reporter system (enabling high-throughput functional assessment). Interestingly, the first four 5-amino acid motifs at the N-terminal region of CN (i.e., residues 62-81) were shown to be important for the suppression of the NP/RNP activity (Supplementary Fig. S7a,b). Subsequently, we conducted singleresidue scanning mutagenesis across these 20 amino acids (prioritizing polarity-altering substitutions to maximize phenotypic effects for rapid identification of critical residues). Reporter assays demonstrated that mutations at multiple residues within this region seemed to affect CCND3's ability to suppress viral NP/RNP activity, with five residues (M64, C68, E70, R72, and P79) exhibiting particularly significant functional importance (Supplementary Fig. S7c,d). Moreover, the structural modeling indeed localized these residues to or adjacent to the predicted interaction interface (Supplementary Fig. S7e). Consistently, protein interaction analysis confirmed that mutations at these sites impaired CCND3 targeting of NP (Supplementary Fig. S7f), correlating with significantly reduced antiviral activities (Supplementary Fig. S7g-i). These findings refine the understanding of CN's pivotal role in CCND3's antiviral function and provide residue-level mechanistic insights into the protein targeting.
By the CN domain, CCND3 hinders the NP activities essential for RNP construction and function, including NP multimerization, NP-RNA binding, and NP-L interaction NP multimerization, NP-RNA binding, and NP-L interactions are critical activities of NP required for viral RNP construction and function 11,12,14 . Based on characteristics of CCND3-NP interaction and the structural modeling, we considered that CCND3 binding to NP probably obstructs these NP activities (as analyzed below). Therefore, in order to further mechanistically address CCND3 inhibition on the viral RNP by targeting NP, influence of CCND3 on NP multimerization that drives RNP formation was first evaluated by in situ chemical crosslinking with disuccinimidyl suberate (DSS) 39,40,52 . CCND3 and its T283A mutant that retains the antiviral capability indeed both could significantly hamper NP multimerization (Fig. 8a). However, the CCND3 mutants with CN or CBox deleted lost the ability to interfere with NP multimerization (Fig. 8b), in agreement with the important role of CN in targeting NP and inhibiting the viral replication. According to these experimental data, binding of CCND3 to NP likely leads to a significant steric hindrance obstructing NP multimerization. Interestingly, this is consistent with the structural model, in which the CCND3-bound interface on NP is on the side that would face inward in the supposed NP oligomers, that is, CCND3 binding would sterically hinder the formation of NP oligomers (Fig. 8c).
By interacting with viral RNAs, NP encapsidates the nucleic acid molecules into RNP complexes and is involved in viral transcription and replication 11,12 . In the "head" region of SFTSV NP, a positivelycharged RNA binding cavity, referred to as CavityR, is formed between the N-and C-lobes 10,51 . Considering that both the two lobes interact with CCND3, we hypothesized that CCND3 may block the RNA binding of NP. RNA-pulldown assays using SFTSV S-segment-derived RNA 39 were thus performed. NP was efficiently co-precipitated by the RNA bait; however, this co-precipitation was significantly inhibited by CCND3 (Fig. 8d). In contrast, deleting CN (or CBox) disrupted the inhibitory effect of CCND3 on NP-RNA binding (Fig. 8d), indicating that CN is also required for CCND3 blockade of the NP-RNA interaction. These consistent results are linked to the experimental data characterizing the CCND-NP interaction (Figs. 6 and7). Moreover, they are also well consistent with the structural modeling of the CCND3-NP complex, in which CCND3 (particularly its CN domain) binding to the two lobes of NP congruously seals the CavityR, providing an interesting structural insight (Figs. 7a, b and8e).
Finally, S-pulldown assays showed that the interaction between NP and L (the viral RdRp) can also be dose-dependently disrupted by CCND3 (Fig. 8f). However, deleting CN or CBox, but not CC, deprived CCND3 of its disruption of NP-L interaction (Fig. 8g). Taken together, these findings establish that by targeting to the N-and C-lobes of NP, CCND3 blocks the NP multimerization, NP-RNA binding, and NP-L interaction through its CN region, thus interfering with the RNP machinery and virus replication.
## NSs counteracts virus/IFN-stimulated CCND3 induction and promotes autophagy-dependent CCND3 degradation
NSs plays multifaceted roles in viral host adaptation, infection, and pathogenesis by antagonizing or manipulating multiple host biological processes, including dampening SFTSV-stimulated PRR and IFN signaling pathways, as well as inducing mTOR-Beclin-1 pro-viral autophagy [15][16][17][18]21,53 . We therefore considered that CCND3, as an ISG induced by IFNs and SFTSV infection, is likely also subject to NSsmediated antagonism. First, we observed that NSs did not markedly interfere with the CCND3-mediated suppression of NP activities identified above including the NP multimerization, NP-RNA binding, or NP-L interaction (Supplementary Fig. S8). However, interestingly, the transcriptional upregulation of CCND3, like several other typical ISGs, induced by SFTSV infection was significantly antagonized by NSs, but not a previously reported inactivated mutant NSs-8A 16,53,54 (Fig. 9a). Similarly, NSs substantially inhibited IFN induction of CCND3 and other ISGs, whereas the NSs-8A mutant lost this antagonistic activity (Fig. 9b). These findings suggest that NSs could partially counteract the infection-or IFN-induced expression of CCND3, thereby potentially weakening the host's antiviral response mediated through CCND3. Additionally, we found that treatment with the autophagy inhibitor chloroquine (CQ) or genetic knockout of the autophagy pathway appears to increase endogenous CCND3 protein levels, suggesting that CCND3 itself may undergo constitutive degradation via autophagic flux (Fig. 9c,d). Interestingly, consistent with the previously reported pro-autophagic activity of NSs, NSs (but not the mutant) expression led to a relative reduction of the CCND3 abundance (Fig. 9c,d). Furthermore, the pharmacological inhibition by CQ (but not the proteasome inhibitor MG132) or deletion of the NSsinduced autophagy pathway reported previously abolished NSscaused CCND3 degradation (Fig. 9c,d), indicating that this effect of NSs is indeed autophagy-dependent. Collectively, these results demonstrate that NSs likely promotes CCND3 degradation by enhancing autophagic activity, which could thereby further contribute to viral counteraction against the host's CCND3-dependent antiviral response.
## Discussion
The emerging bandaviruses have posed a significant threat to global public health 55 . Particularly, SFTSV has been listed by World Health Organization in Pathogens Prioritization Framework (June 2024) as a top priority pathogen which has a high likelihood of causing PHEIC (Public Health Emergencies of International Concern) and requires immediate research and development efforts 55 . Understanding the intricacies of virus-host interplays is crucial for developing effective antiviral therapies. IFN system contributes significantly to controlling viral infection by mounting the production of ISGs, the effectors of IFN responses. However, it remains unclear which and how ISGs regulate bandavirus infection. Here, we conducted the functional screen of Fig. 5 | CCND3 restricts SFTSV replication and pathogenicity in vivo. C57BL/6J mice (n = 6/group) were transduced with the viral vectors encoding control (shNC) or CCND3-targeting shRNAs via caudal vein. At 7 d post transduction, the animals were infected with SFTSV (10 5 TCID 50 ) or mock infected with PBS control via intraperitoneal inoculation. Three days postinfection, the mice were sacrificed for following analyses. CCND3 mRNA levels in the indicated tissues were analyzed by qPCR (a). SFTSV RNA levels in the lung, spleen and liver (b) and serum viral copies (c) were analyzed in SFTSV-infected shNC control and CCND3-deficient mice. Platelet (PLT) and white blood cell (WBC) counts of whole blood (d) and serum ALT, AST, and UREA (e) were also determined as described in Methods. Representative H&E staining of liver (f), spleen (g), and lung (h) sections and cumulative pathological score were respectively shown. Noticeable pathological changes in the tissues are indicated by colored arrows. Liver: blue indicates hepatocellular necrosis, nuclear disappearance, karyomegaly, and nuclear fragmentation; black represents changes and necrosis in hepatic sinusoids; yellow indicates infiltration of inflammatory cells in the liver. Spleen: red indicates indistinct boundaries between the red pulp and white pulp; black denotes dispersed splenic nodules with decreased cellularity of lymphocytes. Lungs: red indicates proliferation of alveolar epithelial cells, alveolar atrophy, and thickening of the alveolar septum; green signifies infiltration of inflammatory cells within the lungs. Scale bar, 100 μm. Data are means ± SD, n = 6 mice. Two-tailed Student's t test (a-c and f-h) and one-way individual ISGs and identified more than 200 candidates potentially regulating SFTSV replication, which may help address the critical knowledge gap regarding the virus-host interactions. Furthermore, with SFTSV as the primary virus model, we uncovered the notable role of CCND3, a top hit from the screen, in restricting bandavirus replication and pathogenicity in vitro and in vivo. Mechanistically, CCND3 targets the "head" (consisting of N-and C-lobes) of the viral NP protein by its CN domain to block NP multimerization, NP-RNA binding, and counteract CCND3's antiviral response by restricting virus/IFN-stimulated upregulation of CCND3 expression and promoting autophagic degradation of CCND3. These findings yield new insights into the virushost interactions, expanding our understanding of virus infection and host response and potentially aiding future antiviral research.
Previous studies by large-scale screening have identified antiviral ISGs against several viruses [56][57][58][59] . In the present study, we screened for the first time the ISGs potentially having regulatory roles in SFTSV infection. Beside CCND3, many other ISG molecules were identified as anti-SFTSV effector candidates which may also merit further functionmechanism studies. Moreover, multiple ISGs were found to have potential pro-viral activities probably enhancing SFTSV replication. It is possible that viruses hijack certain ISGs to indirectly or directly promote their own replication. Exploration of such ISGs could further our knowledge of the overall replication, adaptability and pathogenicity of bandaviruses. However, the expression profile of ISGs that can be induced upon virus infection or IFN stimulation may vary to some extent in different types of cells. Larger-scale screening involving more potentially IFN-stimulated host factors, even in different cells, is pending for future studies to further comprehensively understand the host regulation of bandaviruses. Additionally, we applied a minireplicon reporting system for the screening in this study, which is unrestrained by the demands of high-level biosafety facilities and allows for more ready implementation. Therefore, our study was focused on the RNP machinery-associated transcription and replication process, that is, the central stage of the virus life cycle. In the future, it will be merited to carry out ISG functional screening regarding other stages of bandavirus infection.
Identifying and understanding these pro-viral and antiviral host factors may provide novel perspectives for future antiviral drug development. For antiviral factors like CCND3, potential strategies could involve designing pharmacological agents to enhance their expression levels or strengthen their targeting of viral replication machinery, e.g., reinforcing CCND3 targeting of NP identified here. Alternatively, mechanism-mimicking antiviral peptides could be developed to interfere with viral components like NP. Conversely, for pro-viral factors, approaches might include screening inhibitors to block their virus-supporting interactions, reducing their expression, or leveraging targeted protein degradation technologies (e.g., PROTACs) to eliminate these host factors. These innovative strategies, derived from modulating host-virus interplay, may represent interesting directions for next-generation antiviral therapeutics and warrant further exploration in the future.
As aforementioned, NP-driven RNP construction and RNP functioning are the major and central events during bandavirus infection cycle. Bandavirus RNPs are the viral RNA synthesis machinery catalyzing transcription and replication and those encapsulating viral genome RNAs can be packaged into progeny virion as the structural core 11,12 . Previous studies by us have demonstrated that host proteins including MOV10 and MxA can target SFTSV RNP and inhibit the viral infection 39,40 . Here, we further unraveled a clear antiviral mechanism of CCND3 by targeting NP/RNP machinery. These findings therefore support that the viral RNPs and their delicate activities could be vulnerable targets not only for therapeutic design but for host antiviral defense. However, there are several notable differences in the action details of these antiviral molecules. For instance, the N-arm of NP is involved in the targeting of NP by both MOV10 and MxA 39,40 , but not in the interaction between NP and CCND3. CCND3 specifically targets the NP "head" composed of N-and C-lobes, inhibiting NP-RNA binding. Interestingly, structural modeling analysis coincidently suggested that CCND3 likely binds to the NP "head" and thus seals the RNA binding cavity (CavityR) formed between the N-and C-lobes, blocking the NP-RNA interaction required for encapsulation. In addition to the RNA binding, NP multimerization plays a major role in RNP formation. Although it remains unknown whether MxA affects the NP-RNA binding, previous data showed that MxA has no influence on the NP multimerization 40 . However, by contrast, the binding of CCND3 to NP "head" also obstructs the multimerization of NP. Again, the AlphaFold complex modeling presented a consistent result showing that CCND3 binding to NP could lead to a significant steric hinderance obstructing NP oligomerization. The divergent targeting mechanisms among these host factors, potentially enabling functional synergy which may merit future investigation. Aside from the targeting of NP by CCND3, we also observed a possible interaction between CCND3 and NSs, but to a less extent. NSs is a nonstructural protein that is not essential for the viral replication but notably, has antagonizing activities against multiple host biological processes including virus-induced PRR/IFN signaling and ISG induction [15][16][17][18]20 . We thus thought that it, in turn, might counteract the antiviral response mediated by CCND3. First, an additional analysis demonstrated that NSs has no evident antagonizing effects on CCND3 inhibition to the activities of NP, including NP multimerization, NP-RNA binding, or NP-L interaction. Then, as expected, the virus infection-and IFN-stimulated induction of CCND3, as well as several other typical ISGs, was similarly suppressed by NSs. Additionally, we and other groups have demonstrated that NSs induces the autophagylysosome pathway that positively supports SFTSV replication 21,53 , although how the autophagic flux enhances the viral replication remains to be fully addressed. In this study, we found that pharmacological inhibition or genetic deficiency of autophagy led to increased endogenous CCND3 protein abundance, indicating that CCND3 undergoes turnover through the autophagic degradation pathway. Consistent with the previous findings on the pro-autophagic activity of NSs, our data further suggest that NSs likely promotes autophagydependent CCND3 degradation. These findings not only demonstrate the potential antagonistic effects of NSs on CCND3, but also advance the understanding of how autophagy facilitates SFTSV replication, expanding the cognizance of virus-host interplays and evolutionary arms races.
CCND3 is one of the D-type cyclins (the other two isoforms are CCND1 and CCND2), which play regulatory roles in the cell cycle 42,45,60 . These cyclins are expressed in a highly overlapping fashion in different cells and show amino acid similarity and functional redundancy in cell cycle regulation to some extent 41,45,61,62 . However, in comparison, CCND3 that is expressed in nearly all proliferating cells shows much Fig. 6 | CCND3 targets the viral NP protein independently of RNA, interfering with the RNP machinery-driven reporter activities. a HEK293 cells mock-infected or infected with SFTSV were harvested at 24 hpi for co-immunoprecipitation (Co-IP) and WB analysis of the WCL input and IP products. b HEK293T cells were cotransfected with the Flag-CCND3 and SFTSV-NP expression plasmids or control vector, followed by Flag-pulldown and WB analysis. c Cells were treated as in (b), but the cell lysates were first subjected to UltraNuclease treatment or left untreated, followed by Flag-pulldown and WB analysis, similarly. d-f The dose-dependent inhibitory effect of CCND3 on SFTSV EGFP-based minigenome RNP reporter system. BHK-21 cells were co-transfected with the SFTSV L and NP expression plasmids and the minigenome reporter plasmid (MUTR-EGFP), along with various amounts of the CCND3 expression plasmid for 48 h, followed by fixation and high-content imaging (d) and counting (e). EGFP-positive cell ratios were normalized to the control without CCND3 overexpression (e). In a parallel experiment, the samples were also subjected to WB analysis of the EGFP expression levels (f). g The dose-dependent inhibitory effect of CCND3 on SFTSV firefly luciferase (Fluc)-based minigenome RNP reporter. The SFTSV NP and L expression plasmids and Fluc-based SFTSV minigenome reporter plasmid (MUTR-LUC) were co-transfected together with indicated amounts of the CCND3 expression plasmid and pRL-TK expressing Renilla luciferase (Rluc). Relative luciferase activities (Fluc/Rluc) were calculated and shown. Data in (a-d and f) are representative from three independent replicates with similar results. Data in (e and g) are means ± SD, n = 3 biological replicates. One-way ANOVA (e and g): *p < 0.05, ****p < 0.0001. Source data are provided as a Source Data file.
broader expression pattern than the other two D-type cyclins 41,45,63 . Interestingly, by a comparative experiment, we found that both CCND1 and CCND2 also inhibit SFTSV replication, which however, is significantly weaker than that mediated by CCND3 (Supplementary Fig. S9). It may be linked with the protein homology but more importantly, suggests the unique antiviral efficacy of CCND3 as an ISG. Apart from the classical cell cycle-associated function, the role of CCND3 in viral infection is poorly understood. According to a few published studies, CCND3 can interact with the M2 protein of influenza virus and the E and M proteins of SARS-CoV-2, inhibiting the viral propagation 64,65 . In contrast, Ruiz et al. reported that CCND3 may have a positive role in supporting HIV-1 replication 66 . We here uncovered that by its CN domain, CCND3 targets the pivotal structural and functional regions of SFTSV NP and correspondingly blocks multiple activities of the viral protein required for the RNP machinery, thereby restricting the bandavirus replication and pathogenicity in vitro and in vivo. The current study thus more systematically proposes a clear, new antiviral mode of CCND3, expanding the knowledge of CCND3 biological functions. Additionally, mutation and even the large functional domain truncation did not deprive CCND3 of its antiviral capacity. Consistently, KO of CDK4, CDK6, or both did not impair CCND3's anti-SFTSV activity or lead to any enhancement of the viral replication, unlike the effects of CCND3 KO, indicating that the antiviral activity of CCND3 is independent of its cell cycle-related physiological functions. Further, although the expression of CCND3 can be induced by viral infection and IFN stimulation, its anti-SFTSV action is independent of the IFN signaling. These observations are also consistent with the interesting antiviral mechanism found in this study, where CCND3 targets NP by a specific protein-protein interaction, directly interfering with the RNP machinery and hence virus replication.
In summary, we identified many ISGs potentially inhibiting or bolstering SFTSV replication. As an exemplification, CCND3 was then validated to be a new host restriction factor against bandavirus infection and pathogenicity. Moreover, a sophisticated antiviral mode exploited by CCND3 was unraveled in details, presenting a typical instance regarding host antiviral mechanism. This study may contribute to comprehensive understanding of the virus-host interactions and promote the development of antiviral therapy in the future.
## Methods
## Ethics statement
Animal experiments were approved by the Institutional Animal Care and Use Committee and the Ethical Committee of Wuhan Institute of Virology, Chinese Academy of Sciences (Approval No. WIVA23202301), and conducted in accordance to the guidelines for the Care and Use of Medical Laboratory Animals (Ministry of Health, China). All mice were housed under controlled conditions: temperature maintained at 22-24 °C, relative humidity at 40-60 %, and a 12-h light/dark cycle.
## Cells and viruses
Human embryonic kidney 293 cells (HEK293, ATCC, CRL-1573) were cultured in minimum Eagle's medium (MEM, Gibco) supplemented with 10% fetal bovine serum (FBS) at 37 °C under 5% CO 2 . HEK293T cells (ATCC, CRL-11268), African green monkey kidney cells (Vero, ATCC, CCL-81), mouse embryonic fibroblast cells (MEF; National Virus Resource Center, NVRC), L929 cells (ATCC, CCL-1) and hamster BHK-21 cells (NVRC) were maintained in Dulbecco's modified Eagle's medium (DMEM, Hubei NZK) supplemented with 10% FBS. Human monocyte/ macrophage cells (THP-1, ATCC, TIB-202) were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco) supplemented with 10% FBS, or induced with phorbol 12-myristate 13-acetate (PMA, 100 ng/mL; MCE, Cat#HY-18739) for 48 h for cell differentiation. Human peripheral blood mononuclear cells (PBMCs; ATCC, PCS-800-011) were cultured in RPMI-1640 medium. Mouse Bone Marrow-Derived Macrophages (BMDMs) isolated from the femurs and tibias of mice were cultured in 30% L929 mouse fibroblast supernatant and 70% RPMI-1640 medium with 10% FBS for 7 days 67 . Severe fever with thrombocytopenia syndrome virus (SFTSV, strain WCH; CSTR: 16533.06. IVCAS 6.6088), Heartland virus (HRTV, strain MO-4; CSTR: 16533.06. IVCAS 6.6330), and Guertu virus (GTV, strain DXM; CSTR: 16533.06. IVCAS 6.6106) were propagated in Vero cells and titrated by the 50% tissue culture infectious dose (TCID 50 ) method 15,39,40 .
## Plasmids and transfection
Expression plasmids for CCND3 and its mutants were constructed using pcDNA3.1(+)-N-Flag, pcDNA3.1(+)-C-HA, or pCAGGS-C-S.tag or HA vectors. Plasmids expressing EGFP-fused NP and NP truncated mutants were constructed using the pEGFP-N1 vector. Plasmids encoding EGFP-fused CCND3 and its individual domains were constructed using the pEGFP-C1 vector. The SFTSV NP, L, or NSs encoding plasmids, EGFP or firefly luciferase-based minigenome reporter plasmids, and Renilla luciferase internal control plasmid (pRL-TK) were described previously 15,[38][39][40] . Plasmids used for RNAi or gene editing were constructed according to the standard procedures as stated in the following. The cDNA expression clones of individual ISGs for functional screening were picked from GIPZ human genome cDNA library (GE Healthcare Dharmacon) or supplementally constructed through standard molecular biology techniques by referencing previous studies 57,58 , generating a library of 378 ISGs. Plasmid transfection was performed with Lipofectamine 3000 transfection reagent (Invitrogen, Cat#L3000015) according to the manufacturer's instructions. For RNAi with siRNAs, the gene-specific or control siRNA duplexes (synthesized by Sangon Biotech) were transfected into HEK293 cells using RNATransMate (Sangon Biotech, Cat#E607402-1000).
## Antibodies and reagents
Rabbit antisera to SFTSV NP, RdRp, GP (Gn), and NSs, and mouse anti-SFTSV NP serum were described previously 15,19,21,39 . Mouse anti-Flag mAb (Sigma-Aldrich, Cat#F3165), anti-HA-tag mAb (Proteintech, Cat#66006-2-Ig), anti-CCND3 mAb (Proteintech, Cat#66357-1-Ig) and anti-β-actin mAb (ABclonal, Cat#AC026), and rabbit anti-CDK4 pAb (ABclonal, Cat# A0366), anti-CDK6 mAb (ABclonal, Cat# A0106), anti-S-tag pAb (Sino Biological, Cat#ab101290-T38), and anti-EGFP-tag pAb (Proteintech, Cat#50430-2-AP) were purchased from the indicated manufacturers. For the secondary antibodies, goat anti-mouse IgG conjugated with Alexa Fluor 488 (Thermo Fisher Scientific, Cat#A-11001) and goat anti-rabbit IgG conjugated with Alexa Fluor 647 (Thermo Fisher Scientific, Cat#A32733) were used in immunofluorescence assays; goat anti-mouse or anti-rabbit IgG antibodies conjugated with HRP (Abcam, Cat#ab6789 and Cat#ab6721) were used for Western blot analysis. CBox, CCND3 Box domain. d Deletion of the CN or CBox, but not CC, disrupts CCND3 targeting of NP. HEK293T cells were co-transfected with the plasmids expressing Flag-tagged CCND3 or mutants and the S-tagged NP expression plasmid, or control vectors (indicated by Vector or "-"), followed by S-pulldown and WB analyses. e Both the N-and C-lobes, but not N-arm, of NP can be targeted by CCND3. Plasmids encoding EGFP or EGFP-tagged NP mutants and CCND3 were cotransfected into HEK293T cells, followed by EGFP-NanoTrap assays and WB analysis. EGFP-fused bait protein bands are indicated by arrowheads. f-h CN is required for CCND3 inhibition on SFTSV RNP activity. Effects of CCND3 or mutant expression on SFTSV EGFP-based minigenome reporter were evaluated by imaging (f), calculation of relative reporter activities (g), and WB (h). i-k CN (but not CC) is essential to CCND3 restriction of SFTSV replication and propagation. Effects of the indicated CCND3 mutants on SFSTV infection were analyzed at 24 hpi by qPCR (i), WB (j), and TCID 50 assays (k), respectively. l-p CN expression alone can target NP and inhibit SFTSV infection. Targeting of NP by individual CCND3 domains tagged with EGFP (l) was analyzed by S-pulldown (m). The anti-SFTSV activities were tested by qPCR (n), WB (o), and progeny titration (p). Data in (d-f, h, j, m and o) are representative from three independent replicates with similar results. Data in (g, i, k, n, and p) are presented as means ± SD, n = 3 biological replicates. One-way ANOVA: ****p < 0.0001; ns, non-significant. Source data are provided as a Source Data file.
Chloroquine (MCE, Cat#HY-17589A) and MG-132(MCE, Cat#HY-13259) were used for cell treatments.
## Western blot
For Western blot (WB) analysis, protein samples were subjected to 12 to 15% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene difluoride membranes (PVDF) (Millipore, Cat#L3000015). After blocking with 5% non-fat milk in Tris-buffered saline and Tween 20 (TBST) (BOSTER, Cat#AR0195-10), the membranes were further incubated with the indicated primary antibodies overnight at 4 °C and the corresponding horseradish peroxidase (HRP)-conjugated secondary antibodies for 2 h at room temperature. Protein bands were detected by SuperSignal West Pico PLUS (Thermo Scientific, Cat#34580) using a chemiluminescence analyzer (Azure Fig. 8 | CCND3 inhibits NP multimerization, NP-RNA binding, and NP-L interactions and consistently, the CN domain is required for all these inhibitory activities. a CCND3 as well as its T283A mutant inhibits NP multimerization. HEK293T cells were transfected with the plasmids encoding SFTSV NP and Flagtagged CCND3 or T283A mutant, followed by DSS cross-linking at 24 h post transfection and WB analysis of NP monomeric and oligomers. Gray scale of the monomer or polymer bands in the DSS-treated samples was quantified using ImageJ and normalized to the control (vector) group. b Effects of CCND3 truncations on the inhibition of NP multimerization. c Structural modeling reveals a consistent result in which CCND3 interaction with NP obstructs NP multimerization with the adjacent protomers. NP and CCND3 in the interaction complex were colored in red and gray, respectively. Two adjacent NP protomers that are supposed to be assembled into multimers are indicated in tint colors. d Effects of CCND3 and its mutants on NP-RNA binding. HEK293T cells were transfected with the plasmids encoding Flag-tagged CCND3 (or mutants) and SFTSV NP. At 24 h post transfection, the cell lysates were subjected to RNA-pulldown with beads coupled with S RNA, followed by WB analysis. The intensity of NP bands in the RNApulldown products was quantified and normalized to the control. e CCND3 (especially CN) binding to NP seals the CavityR of NP, site for RNA binding, in the CCND3-NP complex structure from AlphaFold modeling. f CCND3 dosedependently inhibits NP-L interaction. HEK293T cells were co-transfected with the NP (NP-S.tag) and L expression plasmids, along with indicated amounts of the Flag-CCND3 expression plasmid, followed by S-pulldown and WB. g Effects of CCND3 mutants on NP-L interaction. Data are representative from three replicates with similar results (a, b, d, f, and g). Data in the bar graphs (a, b, and d) are means ± SD, n = 3 biological replicates. One-way ANOVA: ****p < 0.0001; ns, non-significant. Source data are provided as a Source Data file. Biosystems, lnc C280) and analyzed by ImageJ software. Uncropped scans of the blots are provided in the Source Data file.
## Immunofluorescence and confocal microscopy
Transfected or infected cells were fixed by 4% paraformaldehyde fixation (PFA) for 30 min, permeabilized by 0.5% Triton X-100 for 15 min, and blocked with 5% bovine serum albumin for 1 h. The cells were then incubated with primary antibodies overnight at 4 °C and with fluorescence-labeled secondary antibodies for 1 h at room temperature. Nuclei were stained with Hoechst 33258 (Beyotime, Cat#C1011). Confocal analysis was performed using a Leica sp8 laser confocal microscope. Images were analyzed using Leica Application Suite X software.
## RNA extraction and qPCR
Total RNA from cells or mouse tissues was extracted using RNAiso Plus (TAKARA, Cat#9109). The cDNA was synthesized using a HiScript II Q RT SuperMix for qPCR kit (Vazyme, Cat#R223-01). qPCR was performed with specific primers as listed in Supplementary Table S1. Relative RNA levels normalized to the mRNA levels of GAPDH were calculated by the 2 -ΔΔCt method. RNA in mouse serum was isolated with a TaKaRa MiniBEST Universal RNA Extraction Kit (TAKARA, Cat#9767) for absolute quantification of SFTSV RNA by TaqMan real-time PCR using HiScript II One Step RT-PCR Kit (Vazyme, Cat#P611-01). SFTSV S-segment RNA synthesized with the T7 RNA polymerase transcription kit (Ambion, Cat#AM1314) was used to construct the standard curve 20,39 .
## Minigenome reporter assay and ISG cDNA library Screening
The SFTSV minigenome reporter assays were performed as described previously 38 . Briefly, the L and NP expression vectors and EGFP-based M-segment minigenome transcription plasmid (pRF42-MUTR-EGFP) were co-transfected into BHK-21 cells in 96-well plates, together with the plasmids encoding CCND3 or its mutants or control vectors. Fortyeight hours post transfection, cells were fixed with 4% PFA and then stained with Hoechst 33258 for 5 min. EGFP-positive cell counts were determined using the Operetta CLSTM high-throughput system (Per-kinElmer) and positive ratios (over total cell counts) were then calculated and normalized to the control group. For library screening, the expression plasmids encoding individual ISGs from the cDNA library (Supplementary Data 1) were co-transfected with the minigenome system plasmids, followed by the high-throughput analysis. In the minigenome system with luciferases as reporters, the firefly luciferasebased minigenome transcription plasmid (pRF42-MUTR-LUC) along with the control plasmid (pRF42-TK) were used to replace pRF42-MUTR-EGFP for co-transfection. At 48 h posttransfection, cells were delivered to luciferase activity measurement using a dual-luciferase reporter kit (Promega, Cat#E2940). Relative luciferase activities (Rel. Luc. Act.) were then calculated 18,68 .
## Nuclear-cytoplasmic fractionation
Cells were infected with SFTSV at various MOI or mock infected. Nuclear-cytoplasmic isolation was conducted at 24 hpi using the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, Cat#P0028) according to the manufacturer's protocol, followed by WB analysis. Protein bands were analyzed by ImageJ and normalized to β-actin (cytoplasm) and histone deacetylase1 (HDAC1) (nucleus), respectively.
## Protein interaction analysis
Transfected or infected cells were suspended in the immunoprecipitation (IP) lysis buffer (Beyotime, Cat#P0013) supplemented with proteinase inhibitor cocktail (Roche, Cat#04693116001). Supernatants of the cell lysates were then subjected to protein interaction analyses 15,18,46,69 . For S-pulldown, Flag-IP, or EGFP-NanoTrap assays, the supernatants were respectively incubated with S-protein agarose (Millipore, Cat#69704), anti-Flag Magnetic Beads (MCE, Cat#HY-K0207), or anti-EGFP nanobody-coated agarose beads (AlpaLife, Cat#KTSM1301) for 4 h with gentle rotation at 4 °C. For coimmunoprecipitation (Co-IP) assays, HEK293 cells mock-infected or infected with SFTSV were harvested and lysed using cell lysis buffer (Beyotime, Cat#P0013) at 24 hpi. Subsequently, supernatants of the cell lysates were incubated with the anti-CCND3 antibody or control IgG, together with protein A/G beads, overnight with gentle rotation at 4 °C. After washing, the precipitates were eluted by boiling for 5 min in 1 × SDS sample buffer, followed by WB analyses with specific antibodies. To analyze the effect of nucleic acids on protein-protein interactions, cell lysate samples containing 5 mM MgCl 2 were treated with UltraNuclease (Yeasen, Cat# 20157ES25) for 1 h at room temperature before the S-pulldown assay 40 .
## Chemical cross-linking
HEK293T cells were transfected with the indicated expression plasmids or control vectors. At 24 h posttransfection, the cells were harvested and subjected to cross-linking with disuccinimidyl suberate (DSS, Thermo Fisher Scientific, Cat#A39267) at room temperature for 30 min. The reaction was then stopped with 0.1 M Tris-HCl (pH 7.5), before WB analysis of the NP monomer and oligomers 39,40 .
## RNA-protein pulldown assay
RNA-protein pulldown assays were conducted as previously described 39 . Briefly, SFTSV S-segment RNA was synthesized with a T7 RNA Polymerase Transcription Kit (Ambion, Cat#AM1314), followed by the addition of biotin label to the 3' end using a Thermo Scientific Pierce RNA 3'-Desthiobiotinylation Kit (Thermo Fisher Scientific, Cat#20163). The labeled RNA was extracted with an equal volume of chloroform/isoamyl alcohol, followed by ethanol precipitation. The biotin-labeled RNA resuspended in nuclease-free water (40 pmol) was incubated with 40 μL of streptavidin magnetic beads (Thermo Fisher Scientific, Cat#20164) at room temperature for 30 min according to manufacturer's instructions. To facilitate the RNA-protein pulldown analysis, supernatants of indicated cell lysates were rotated with the RNA-bound magnetic beads at 4 °C for 1 h, followed by WB detection of the co-precipitates and lysate inputs.
## Construction of knockout (KO) cells by CRISPR-Cas9 gene editing
The human CCND3, CDK4, CDK6 and IFNAR1 gene specific sgRNAs were designed using the online CRISPR Design Tools (https://zlab.bio/ guide-design-resources) and cloned into px459 using BbsI restriction sites 70 . HEK293 cells were transfected with the px459-CCND3, CDK4, or CDK6 sgRNA plasmids and selected for 3 d in the presence of puromycin. Clonal CCND3, CDK4, or CDK6-deficient cell lines were respectively obtained by limiting dilution and validated by WB analysis and sequencing with the specific primers as listed in Supplementary Table S1. For construction of IFNAR1-KO THP-1 cells, the human IFNAR1 gene sgRNA was cloned into lentiCRISPRv2 (Addgene, Cat#52961) and packaged in HEK293T cells with packaging plasmids psPAX2 and pMD2.G 39,46 . THP-1 cells were infected with the obtained lentivirus vector and selected for 5 d with puromycin. Similarly, the candidate cell lines were validated by WB and sequencing with specific primers (Supplementary Table S1).
## CCND3-deficient mice and SFTSV infection experiment
C57BL/6J and IFNAR1 -/-(A129) mice were bred in-house at the institutional animal facility under specific pathogen-free (SPF) conditions. The transduction vectors were packaged by co-transfection of HEK293T cell with the CCND3-targeting or control shRNA pLKO.1 plasmids together with the packaging vectors psPAX2 and pMD2.G. At 48 h and 72 h posttransfection, culture supernatants were collected, filtered through a 0.45 μm filter (Millipore, Cat#SLHP033RB), and concentrated by centrifugation (72,100 g, 2 h, 4 °C). The virus particles were suspended in Virus Conservation Solution (PBS with 1% BSA, pH = 7.4) and titrated using the endpoint method 71 . For in vivo KD analysis in immunocompetent animals, female C57BL/6J mice (6-8 weeks old; n = 6) were injected with 1 × 10 8 transduction units (TU) of either control or CCND3-targeting viral vectors via the tail vein. Seven days after transduction, the mice were subcutaneously infected with SFTSV (10 5 TCID 50 ) and three days after infection, euthanized for necropsy and tissue sample collection. Histopathological examination including hematoxylin and eosin (H&E) staining of tissue samples were performed as previously described with the help of Wuhan Servicebio Technology [72][73][74][75] . Images were obtained by PerkinElmer Vectra Polaris and analyzed by P250 (3D HISTECH). Injury scores for various tissues from each mouse were determined based on specific pathological criteria [75][76][77] . Platelet (PLT) and white blood cell (WBC) counts in fresh blood treated with EDTA were measured using a hematology analyzer (Drew Scientific, Mascot HEMAVET950). Serum alanine aminotransferase (ALT) (Servicebio, GM1102), aspartate aminotransferase (AST) (Servicebio, GM1103), and blood urea nitrogen (UREA) (Servicebio, GM1110) were measured by ELISA using the indicated kits. CCND3 mRNA levels and viral copies in tissue samples were delivered to qPCR analyses. For KD analysis in the IFNAR1 -/-models, female A129 mice (6-8 weeks old) were subjected to the similar CCND3 KD method applied to C57BL/6J. After a week, A129 CCND3-KD or control groups (n = 6) were infected with SFTSV (10³ TCID 50 ), and body weight changes and survival rates were monitored daily. Additionally, in a parallel experiment, infected animals (n = 6) were sacrificed at 3 dpi, followed by tissue sample collection and histopathological analyses as described above. For CCND3 induction analysis, female C57BL/6J mice (n = 4) were intraperitoneally injected with SFTSV (10 6 TCID 50 ). At indicated time points post infection, tissue samples were collected for qPCR analysis of Ccnd3 mRNA and SFTSV RNA levels.
## Statistical analysis
Statistical analysis was performed using Student's t test or analysis of variance (ANOVA) with GraphPad Prism 9 software (La Jolla, CA, USA). Data are presented as the means ± standard deviations (SD) of n biological replicates. Differences were considered statistically significant if p-value < 0.05. Significance levels are: *p < 0.05; **p < 0.01; *** p < 0.001; *** p < 0.001; **** p < 0.0001; ns, non-significant.
## Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
## References
1. Yu (2011) "Fever with thrombocytopenia associated with a novel bunyavirus in China" *N. Engl. J. Med*
2. Zhan (2017) "Current status of severe fever with thrombocytopenia syndrome in China" *Virologica Sin*
3. Yoo (2016) "Family Cluster Analysis of Severe Fever with Thrombocytopenia Syndrome Virus Infection in Korea" *Am. J. tropical Med. Hyg*
4. Jiang (2015) "A cluster of person-to-person transmission cases caused by SFTS virus in Penglai, China" *Clin. Microbiol Infect*
5. Matsumoto (2018) "Investigation of antibody to severe fever with thrombocytopenia syndrome virus (SFTSV) in blood samples donated in a SFTS-endemic area in Japan" *Vox sanguinis*
6. Tran (2019) "Endemic Severe Fever with Thrombocytopenia Syndrome" *Vietnam. Emerg. Infect. Dis*
7. Mcmullan (2012) "A New Phlebovirus Associated with Severe Febrile Illness in Missouri" *N. Engl. J. Med*
8. Shen (2018) "A novel tick-borne phlebovirus, closely related to severe fever with thrombocytopenia syndrome virus and Heartland virus, is a potential pathogen" *Emerg. Microbes Infect*
9. Siddell (2023) "Virus taxonomy and the role of the International Committee on Taxonomy of Viruses (ICTV)" *J. Gen. Virol*
10. Zhou (2013) "The nucleoprotein of severe fever with thrombocytopenia syndrome virus processes a stable hexameric ring to facilitate RNA encapsidation" *Protein cell*
11. Zhou, Sun, Guo et al. (2013) "Structural perspective on the formation of ribonucleoprotein complex in negative-sense single-stranded RNA viruses" *Trends Microbiol*
12. Sun, Li, Gao et al. (2018) "Bunyavirales ribonucleoproteins: the viral replication and transcription machinery" *Crit. Rev. Microbiol*
13. Malet, Williams, Cusack et al. (2023) "The mechanism of genome replication and transcription in bunyaviruses" *PLoS Pathog*
14. Cheng, Wang, Mir (2014) "Interaction between Hantavirus Nucleocapsid Protein (N) and RNA-Dependent RNA Polymerase (RdRp) Mutants Reveals the Requirement of an N-RdRp Interaction for Viral RNA Synthesis" *J. Virol*
15. Ning (2014) "Viral suppression of innate immunity via spatial isolation of TBK1/IKKε from mitochondrial antiviral platform" *J. Mol. cell Biol*
16. Wu (2014) "Evasion of antiviral immunity through sequestering of TBK1/IKKε/IRF3 into viral inclusion bodies" *J. Virol*
17. Santiago (2014) "Hijacking of RIG-I signaling proteins into virusinduced cytoplasmic structures correlates with the inhibition of type I interferon responses" *J. Virol*
18. Ning (2015) "Disruption of type I interferon signaling by the nonstructural protein of severe fever with thrombocytopenia syndrome virus via the hijacking of STAT2 and STAT1 into inclusion bodies" *J. Virol*
19. Ning (2019) "Interferon-γ-Directed Inhibition of a Novel High-Pathogenic Phlebovirus and Viral Antagonism of the Antiviral Signaling by Targeting STAT1" *Front. Immunol*
20. Min, Ning, Wang et al. (2020) "A RIG-I-like receptor directs antiviral responses to a bunyavirus and is antagonized by virus-induced blockade of TRIM25-mediated ubiquitination" *J. Biol. Chem*
21. Feng (2023) "SFTS bunyavirus NSs protein sequestrates mTOR into inclusion bodies and deregulates mTOR-ULK1 signaling, provoking pro-viral autophagy" *J. Med. Virol*
22. Macmicking (2012) "Interferon-inducible effector mechanisms in cellautonomous immunity" *Nat. Rev. Immunol*
23. Darnell, Kerr, Stark (1994) "Jak-stat pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins" *Science*
24. Chow, Gale, Loo (2018) "RIG-I and other RNA sensors in antiviral immunity" *Annu. Rev. Immunol*
25. Takeuchi, Akira (2010) "Pattern recognition receptors and inflammation" *Cell*
26. Yamada (2018) "RIG-I-like receptor and toll-like receptor signaling pathways cause aberrant production of inflammatory cytokines/ chemokines in a severe fever with thrombocytopenia syndrome virus infection mouse model" *J. Virol*
27. Lazear, Schoggins, Diamond (2019) "Shared and distinct functions of type I and type III interferons" *Immunity*
28. Der, Zhou, Williams et al. (1998) "Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays" *Proc. Natl. Acad. Sci. USA*
29. Briscoe (1996) "JAKs, STATs and signal transduction in response to the interferons and other cytokines" *Philos. Trans. R. Soc. Lond. B Biol. Sci*
30. Schneider, Chevillotte, Rice (2014) "Interferonstimulated genes: a complex web of host defenses" *Annu. Rev. Immunol*
31. Schoggins (2019) "Interferon-stimulated genes: what do they all do?" *Annu. Rev. Virol*
32. Gowen (2017) "Modeling severe fever with thrombocytopenia syndrome virus infection in golden Syrian hamsters: importance of STAT2 in preventing disease and effective treatment with favipiravir" *J. Virol*
33. Sun (2021) "Animal model of severe fever with thrombocytopenia syndrome virus infection" *Front. Microbiol*
34. Fujii (2022) "Susceptibility of type I interferon receptor knock-out mice to heartland bandavirus (HRTV) infection and efficacy of favipiravir and ribavirin in the treatment of the mice infected with HRTV" *Viruses-Basel*
35. Feng (2024) "Pathogenesis and virulence of Heartland virus" *Virulence*
36. Fujii (2024) "Pathological and virological findings of type I interferon receptor knockout mice upon experimental infection with Heartland virus" *Virus Res*
37. Westover (2024) "Modeling Heartland virus disease in mice and therapeutic intervention with 4′-fluorouridine" *J. Virol*
38. Ren, Zhou, Deng et al. (2020) "Combinatorial minigenome systems for emerging banyangviruses reveal viral reassortment potential and importance of a protruding nucleotide in genome "panhandle" for promoter activity and reassortment" *Front Microbiol*
39. Mo, Xu, Deng et al. (2020) "Host restriction of emerging high-pathogenic bunyaviruses via MOV10 by targeting viral nucleoprotein and blocking ribonucleoprotein assembly" *PLoS Pathog*
40. Chang (2024) "Host factor MxA restricts Dabie bandavirus infection by targeting the viral NP protein to inhibit NP-RdRp interaction and ribonucleoprotein activity" *J. Virol*
41. Bartkova, Lukas, Strauss et al. (1998) "Cyclin D3: requirement for G1/S transition and high abundance in quiescent tissues suggest a dual role in proliferation and differentiation" *Oncogene*
42. Sherr, Cyclins (1995) *Trends Biochem Sci*
43. Weng, Aster (2003) "No T without D3: a critical role for cyclin D3 in normal and malignant precursor T cells" *Cancer Cell*
44. Sicinska (2003) "Requirement for cyclin D3 in lymphocyte development and T cell leukemias" *Cancer Cell*
45. Saleban, Harris, Poulter (2023) "D-type cyclins in development and disease" *Genes-Basel*
46. Dai (2023) "Interactome profiling of Crimean-Congo hemorrhagic fever virus glycoproteins" *Nat. Commun*
47. Best (2018) "Interferon-stimulated TRIM69 interrupts dengue virus replication by ubiquitinating viral nonstructural protein 3" *PLoS Pathog*
48. Chen (2012) "Infection and pathogenesis of Huaiyangshan virus (a novel tick-borne bunyavirus) in laboratory rodents" *J. Gen. Virol*
49. Casanovas, Jaumot, Paules et al. (2004) "P38SAPK2 phosphorylates cyclin D3 at Thr-283 and targets it for proteasomal degradation" *Oncogene*
50. Takaki (2009) "The structure of CDK4/cyclin D3 has implications for models of CDK activation" *Proc. Natl. Acad. Sci. USA*
51. Jiao (2013) "Structure of severe fever with thrombocytopenia syndrome virus nucleocapsid protein in complex with suramin reveals therapeutic potential" *J. Virol*
52. Alfadhli (2001) "Hantavirus nucleocapsid protein oligomerization" *J. Virol*
53. Li (2024) "Bunyavirus SFTSV NSs utilizes autophagy to escape the antiviral innate immune response" *Autophagy*
54. Choi (2019) "Severe fever with thrombocytopenia syndrome phlebovirus non-structural protein activates TPL2 signalling pathway for viral immunopathogenesis" *Nat. Microbiol*
55. (2024) "Pathogens prioritization: a scientific framework for epidemic and pandemic research preparedness"
56. Schoggins (2011) "A diverse range of gene products are effectors of the type I interferon antiviral response" *Nature*
57. Schoggins (2012) "Dengue reporter viruses reveal viral dynamics in interferon receptor-deficient mice and sensitivity to interferon effectors in vitro" *Proc. Natl Acad. Sci. USA*
58. Schoggins (2014) "Pan-viral specificity of IFN-induced genes reveals new roles for cGAS in innate immunity" *Nature*
59. Kuroda (2020) "Identification of interferon-stimulated genes that attenuate Ebola virus infection" *Nat. Commun*
60. Malumbres, Barbacid (2009) "Cell cycle, CDKs and cancer: a changing paradigm" *Nat. Rev. Cancer*
61. Beumer, Roepers-Gajadien, Gademan et al. (2000) "Involvement of the D-type cyclins in germ cell proliferation and differentiation in the mouse" *Biol. Reprod*
62. Sherr, Roberts (2004) "Living with or without cyclins and cyclindependent kinases" *Genes Dev*
63. Weng, Huang, Zhao et al. (2012) "Translational repression of cyclin D3 by a stable G-quadruplex in its 5' UTR: implications for cell cycle regulation" *RNA Biol*
64. Fan (2017) "Cell cycle-independent role of cyclin D3 in host restriction of influenza virus infection" *J. Biol. Chem*
65. Gupta, Mlcochova (2022) "Cyclin D3 restricts SARS-CoV-2 envelope incorporation into virions and interferes with viral spread" *EMBO J*
66. Ruiz (2015) "Cyclin D3-dependent control of the dNTP pool and HIV-1 replication in human macrophages" *Cell Cycle (Georget. Tex.)*
67. Toda, Yamauchi, Kadowaki et al. (2021) "Preparation and culture of bone marrow-derived macrophages from mice for functional analysis" *STAR Protoc*
68. Feng, Deng, Hu et al. (2019) "Heartland virus antagonizes type I and III interferon antiviral signaling by inhibiting phosphorylation and nuclear translocation of STAT2 and STAT1" *J. Biol. Chem*
69. Ning (2017) "Heartland virus NSs protein disrupts host defenses by blocking the TBK1 kinase-IRF3 transcription factor interaction and signaling required for interferon induction" *J. Biol. Chem*
70. Ran (2013) "Genome engineering using the CRISPR-Cas9 system" *Nat. Protoc*
71. Geraerts, Willems, Baekelandt et al. (2006) "Comparison of lentiviral vector titration methods" *BMC Biotechnol*
72. Kang (2025) "Myocardial mitochondrial antiviral signaling protein promotes heart Ischemia-reperfusion injury via RIG-I signaling in mice" *Nat. Commun*
73. Ni (2022) "Antitumor efficacy of CRISPR/Cas9-engineered ICP6 mutant herpes simplex viruses in a mouse xenograft model for lung adenocarcinoma" *J. Med. Virol*
74. Ning (2018) "Ebola virus mucin-like glycoprotein (Emuc) induces remarkable acute inflammation and tissue injury: evidence for Emuc pathogenicity" *Protein Cell*
75. Zhang (2025) "Host protein ARF1 is a proviral factor for SARS-CoV-2 and a candidate broad-spectrum therapeutic target" *Nat. Commun*
76. Lu (2024) "A full-length glycoprotein mRNA vaccine confers complete protection against severe fever with thrombocytopenia syndrome virus, with broad-spectrum protective effects against bandaviruses" *J. Virol*
77. Min (2024) "Developing nucleoside tailoring strategies against SARS-CoV-2 via ribonuclease targeting chimera" *Sci. Adv* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12377068&blobtype=pdf | # Correction: Identification of acrylamide-based covalent inhibitors of SARS-CoV-2 (SCoV-2) Nsp15 using high-throughput screening and machine learning
Teena Bajaj, Babak Mosavati, Lydia Zhang, Mohammad Parsa, Huanchen Wang, Evan Kerek, Xueying Liang, Seyed Amir, Tabatabaei Dakhili, Eddie Wehri, Silin Guo, Rushil Desai, Lauren Orr, Mohammad Mofrad, Julia Schaletzky, John Ussher, Xufang Deng, Robin Stanley, Basil Hubbard, Daniel Nomura, Niren Murthy
## Abstract
Correction for 'Identification of acrylamide-based covalent inhibitors of SARS-CoV-2 (SCoV-2) Nsp15 using high-throughput screening and machine learning' by Teena
Finally, a link to https://github.com/msparsa/nsp15-inhibitors-prediction was omitted from the 'Optimization of AI models' section of the original main article and the Data availability statement. There was also additional information missing from the |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12390710&blobtype=pdf | # Persistent Type I Interferon Signaling Impairs Innate Lymphoid Cells During HIV-1 Infection Under Suppressive ART
Runpeng Han, Haisheng Yu, Guangming Li, Lishan Su, Liang Cheng
## Abstract
Persistent type I interferon (IFN-I) signaling compromises adaptive anti-HIV-1 T cell immunity and promotes viral reservoir persistence, yet its effects on innate lymphoid cells during chronic infection remain unclear. Through integrated single-cell RNA sequencing and functional validation in HIV-1-infected humanized mice with combination antiretroviral therapy (cART) and IFN-I signaling blockade, we reveal IFN-I-induced dysfunction of natural killer (NK) cells and group 3 innate lymphoid cells (ILC3s). Mechanistically, the IFN-I-CD9 axis drives NK cells toward a decidual NK cell-like phenotype, impairing their cytotoxic activity. Furthermore, IFNAR blockade rescues ILC3 functionality, which is critical for IL-17/IL-22-mediated antimicrobial defense and mucosal barrier maintenance. Our study delineates IFN-I-driven immunosuppression across innate lymphocyte compartments and proposes the targeted modulation of this pathway to enhance antiviral and mucosal immunity in HIV-1 management.
## 1. Introduction
Significant advancements in the management of HIV-1 infection have been achieved through combined antiretroviral therapy (cART) [1][2][3]. However, despite its ability to effectively suppress viral replication, cART fails to fully resolve the immune dysfunction and chronic inflammation associated with HIV-1 infection [4]. This persistent immune dysregulation underscores the need for a deeper understanding of the underlying mechanisms driving both innate and adaptive immune dysfunction during HIV-1 infection.
Innate lymphocytes demonstrate functional parallels with adaptive T lymphocytes, where natural killer (NK) cells mirror CD8 + cytotoxic T cell activity and innate lymphoid cells (ILCs) recapitulate CD4 + T helper cell functions [5]. During HIV-1 infection, NK cells orchestrate antiviral responses through dual mechanisms: (1) cytokine secretion (e.g., IFN-γ and TNF-α) that establishes antiviral states, and (2) direct cytolytic activity mediated by perforin-and granzyme-containing cytotoxic granules that eliminate infected cells and restrain viral dissemination [6]. Notably, multiple cohort studies have established correlations between NK cell functional competence and two key clinical outcomes: resistance to HIV acquisition [7] and spontaneous viral control in untreated individuals [8][9][10]. However, chronic HIV-1 infection, even in virologically suppressed patients recei9ving combination antiretroviral therapy (cART), induces NK cell dysfunction characterized by three hallmark features: (1) upregulated inhibitory receptors (e.g., KLRG1 [11] and KIR [12]), (2) downregulated activating receptors (e.g., CD16 [13] and NKG2D [14]), and (3) impaired responsiveness to activation signals [11]. These pathophysiological alterations underscore the necessity of elucidating the mechanisms underlying HIV-1-induced NK cell impairment, a critical prerequisite for developing targeted immunotherapies to restore antiviral function and improve long-term clinical outcomes in cART-treated patients.
The classification of ILCs into three subsets (ILC1s, ILC2s, and ILC3s) mirrors the functional specialization of adaptive T helper (Th) cell lineages, with ILC1s producing interferon-γ (IFN-γ) to combat intracellular pathogens, ILC2s secreting type 2 cytokines (IL-4, IL-5, IL-9, and IL-13) for anti-helminth immunity and tissue repair, and ILC3s generating IL-17 and IL-22 to maintain gut mucosal barrier integrity through antimicrobial defense and epithelial regeneration [5]. In HIV infection, the loss of ILCs during the acute viremic phase has been observed, and this can persist into the chronic stage of the disease, even in individuals receiving cART [15]. The rapid depletion of IL-22-producing ILC3s during acute viremia disrupts intestinal epithelial tight junctions, promotes microbial translocation, and drives systemic inflammation [16]. This ILC3-gut axis dysfunction highlights its central role in HIV pathogenesis, suggesting that therapeutic strategies targeting ILC3 reconstitution could potentially ameliorate barrier defects and chronic inflammation in people living with HIV (PLWH).
Previous investigations from our group and others have established that the therapeutic blockade of IFN-I signaling through anti-interferon-α/β receptor (IFNAR) blocking antibodies reverses CD4 + T cell depletion, enhances CD8 + T cell antiviral activity, and prolongs viral suppression following cART interruption [17][18][19][20]. However, the immunological consequences of IFNAR inhibition on innate lymphocyte populations, particularly NK cells and ILCs, remain undefined. Through systematic single-cell transcriptomic analysis combined with functional validation in HIV-1-infected humanized mice under cART suppression, we demonstrate that IFNAR blockade restores the cytotoxic potential of NK cells and rescues the cytokine-producing capacity of ILC3s. These findings reveal the previously unrecognized IFN-I-mediated mechanisms driving innate immune dysfunction in chronic HIV-1 infection.
## 2. Results
## 2.1. Similarity in Innate Immune Cell Subtypes Between Humanized Mouse Spleens and Human Spleen-Derived Counterparts
Our previous single-cell RNA-seq results demonstrated the successful reconstitution of major human innate immune subsets in humanized NRG mice (hu-mice) 12 weeks after the transplantation of human CD34 + hematopoietic stem cells [21]. To assess the translational relevance of this model, we performed comparative transcriptomic profiling between innate immune cells isolated from hu-mice spleens (Figure 1A,B) and their counterparts derived from cold-stored human spleens [22]. Through t-SNE-based dimensionality reduction and phylogenetic tree reconstruction of scRNA-seq datasets, we observed that innate immune subsets in hu-mice exhibited lineage-dependent clustering patterns with substantial overlap between murine-reconstituted populations and their human splenic counterparts (Figure 1C,D). Subsequent Pearson correlation analysis further confirmed strong transcriptional concordance in key innate immune populations, including natural killer (NK) cells and innate lymphoid cells (ILCs), between hu-mice and human reference samples (Figure 1E). These analyses collectively establish the humanized mouse model as a physiologically representative system for investigating human innate immunity.
## 2.2. Single-Cell Transcriptomic Landscape Reveals HIV-1-and IFN-I-Driven Transcriptional Reprogramming of Innate Immune Cells
Next, we employed the humanized mouse model to investigate HIV-1-mediated perturbations in innate lymphoid cell function and their modulation by IFN-I signaling in vivo. To achieve this, we performed scRNA-seq on human hCD45 + CD3 -CD19 - splenocytes across four experimental groups (n = 3/group): (1) mock-infected controls, (2) HIV-1-infected mice, (3) HIV-1-infected mice treated with cART and isotype antibody, and (4) HIV-1-infected mice treated with cART and anti-IFNAR antibody (Figure 2A,B). After stringent quality control, 25,245 high-confidence single-cell transcriptomes were retained for analysis. Unsupervised clustering identified 13 distinct populations encompassing NK cells, ILCs, macrophages, dendritic cell subsets (pDCs, mDC1, mDC2, and CCL19 + DC), mast cells, and minor contaminants (CD34 + progenitors, B-lineage cells, and erythroid cells) (Figure 2C-E; Supplemental Table S1). As previously reported [17], HIV-1 infection in humanized mice leads to persistent viremia, which can be suppressed by cART (Figure 2F). Differential gene expression (DEG) analysis demonstrated that HIV-1 infection induced significant transcriptional reprogram-ming across all innate immune compartments (Figure 2G, Supplemental Table S2), consistent with our previous report [21]. Gene Ontology enrichment analysis of upregulated genes revealed the profound activation of type I interferon signaling pathways, antiviral defense mechanisms, and negative regulation of viral replication across diverse cell types (Supplemental Table S3). Previous reports have shown that despite the efficient suppression of HIV-1 replication with cART, abnormally elevated IFN-I signaling persists in certain patients even under extensive cART [23,24], as well as in HIV-1-infected humanized mouse models [17]. Our scRNA-seq data further demonstrated that while cART successfully suppressed HIV-1 replication and reduced infection-induced transcriptional changes (Supplemental Tables S2 andS3), the pharmacological blockade of IFNAR signaling during cART administration additionally reversed HIV-associated transcriptional profiles in innate immune populations, particularly NK cells and innate lymphoid cells (ILCs) (Figure 2H, Supplemental Tables S2 andS3). These collective findings indicate that persistent IFN-I signaling during chronic infection continues to regulate innate immune functionality even following successful viral suppression.
## 2.3. IFNAR Blockade Reverses Transcriptional Dysregulation of NK Cells in Chronic HIV-1 Infection Under cART
To investigate the mechanisms by which persistent IFN-I signaling drives NK cell dysfunction during HIV-1 infection under cART, we investigated the transcriptomic changes in splenic NK cells from hu-mice across four experiment conditions (Figure 3A). Through integrated differential expression and unsupervised clustering analysis, we identified four distinct transcriptional modules (C1-C4) that delineated HIV-1-and treatment-dependent molecular signatures. Module C1 contained interferon-stimulated genes (ISGs) that were upregulated during chronic HIV-1 infection but subsequently suppressed by cART and IF-NAR blockade (Figure 3B and Supplemental Table S4). In hu-mice treated with cART alone or cART plus IFNAR blockade, we observed increased activity of RNA processing-related genes (C2 and C3, Figure 3B). This aligns with prior findings showing that HIV-positive individuals with successful viral suppression on cART exhibit higher RNA processing gene activity compared to uninfected individuals or those without viral suppression [25]. Notably, module C4 revealed critical functional deficits in HIV-infected mice, showing the suppression of NK cell effector pathways, including NK cell degranulation and Fc-gamma receptor signaling pathways, that were specifically restored only by cART combined with anti-IFNAR treatment (Figure 3B). This transcriptional change aligns with the clinical observations of persistent NK cell dysfunction in virologically suppressed PLWH [11][12][13][14] and suggests therapeutic potential for IFN-I modulation. At the individual gene level, we observed the significant downregulation of NK cell survival and effector markers (e.g., IL7R, GZMK, IL18RAP, and KLRC2) in HIV-infected mice that persisted through cART monotherapy (Figure 3C). Strikingly, combining therapy with cART and IFNAR-blocking antibody restored the expression of these critical genes (Figure 3C). Complementary gene set enrichment analysis (GSEA) confirmed that IFNAR blockade simultaneously attenuated the IFN-I signaling pathways while enhancing the gene signatures associated with NK cell activation and NK cell-mediated immunity (Figure 3D).
## 2.4. HIV-Induced CD9 + dNK-like Cells Persist via IFN-Dependent Mechanisms Under cART
Previous studies have reported that CD9 + decidual NK (dNK) cells exhibit reduced cytotoxicity [26,27], and the inhibition of CD9 has been shown to suppress HIV replication [28]. Our scRNA-seq data revealed that HIV-1 infection increased CD9 expression in NK cells, and cART failed to restore CD9 expression to the baseline levels observed in mockinfected mice (Figure 4A). Notably, this shift was reversed by IFNAR blockade (Figure 4A). The gene expression profiling of CD9 + NK cells demonstrated a significant increase in dNK-like properties compared to their CD9 -counterparts (Figure 4B). Intriguingly, GSEA revealed that CD9 + NK cells exhibited an elevated IFN-I response but a decrease in the pathways associated with NK cell activation and NK cell-mediated immunity (Figure 4C). Consistent with this, CD9 + NKs showed a statistically significant yet modest upregulation of IFN-I stimulated genes (e.g., IFI27 and ISG15) and downregulation of NK effector genes (e.g., GZMA, GZMK, and KLRF1) (Figure 4D,E).
To experimentally validate these findings, we assessed whether CD9 surface expression on NK cells was induced by HIV-1 infection and IFN-I signaling by flow cytometry (Figure S1). We found that the median percentage of CD9 + NK cells increased by approximately 9.8-fold in the HIV-1-infected group compared with the mock group (Figure 4F). While cART slightly reduced the percentage of CD9 + NK cells, the combination of cART and IFNAR blockade completely restored the CD9 + NK cell levels to baseline (Figure 4F). Given that NK cell function is regulated by a balance of activating and inhibitory receptors [29],
we further analyzed receptor expression using flow cytometry. We observed that CD9 + NK cells exhibited higher surface levels of activating receptor CD16 compared to CD9 -NK cells (Figure 4G). Conversely, inhibitory receptors such as KIR and KLRG1 were upregulated in CD9 + NK cells (Figure 4H), suggesting a less activated phenotype. Moreover, CD94, a co-receptor for the NKG2 family, which can form both activating and inhibitory receptors for HLA-E [30,31], was also downregulated in CD9 + NK cells (Figure 4I), implicating their blunted signal transduction upon HLA-E engagement. In summary, by integrating bioinformatic and flow cytometry approaches, we identified an IFN-CD9 axis in NK cells during HIV-1 infection that likely contributes to their impaired activation and reduced cytolytic antiviral activity.
## 2.5. IFN-I Signaling Drives HIV-1-Induced Dysfunction of ILC3s
ILCs serve as crucial regulators of mucosal immunity, orchestrating both antimicrobial defense and tissue repair through cytokine-mediated mechanisms [5]. Building upon previous reports of HIV-1-associated ILC depletion [15,32,33], we investigated the IFN-Imediated dysregulation of ILC subsets during chronic infection. Our scRNA-seq identified two ILC populations in the spleens of humanized mice: GATA3 + ILC2s and AHR + ILC3s (Figure 5A,B). The DEG analysis revealed predominant HIV-1-induced perturbations in the ILC3 compartment (Figure 5C), prompting our focused investigation on this subset. ILC3s are characterized by their production of IL-17 and IL-22, cytokines critical for anti-bacterial immunity [34] and maintaining intestinal homeostasis [35]. We first surrogated the gene expression patterns in ILC3s under different infection and therapy contexts (Figure 5D and Supplemental Table S5). Notably, we observed that only cART combined with IFNAR blockade enhanced antimicrobial and tissue-regulatory transcriptional programs (module C1) (Figure 5D). This finding mirrors clinical observations of persistent ILC dysfunction in virologically suppressed individuals [15] and suggests therapeutic potential for IFN-I modulation. Moreover, IFNAR blockade synergized with cART to downregulate proapoptotic pathways in ILC3s (module C4) (Figure 5D), consistent with our prior findings that pDC depletion rescues ILC survival [32]. In addition to survival-related genes, IFNAR blockade also significantly downregulates IL-1β production-and inflammasomeactivation-related genes (module C4) (Figure 5D and Supplemental Table S5), which is reported to be a major cause of CD4 + T cell loss during HIV-1 infection [36], suggesting that IFNAR blockade may rescue ILC3s via inhibiting pyroptosis in addition to apoptosis. Moreover, similar to what is observed in NK cells, hu-mice that received cART alone and cART+IFNAR blockade showed the activation of genes related to RNA processing and RNA translation in splenic ILC3s (module C2) (Figure 5D). The therapeutic efficacy was further evidenced by the upregulated expression of (1) IL-7 receptor (IL7R), critical for lymphocyte homeostasis, (2) activation marker CD69, and (3) JUNB transcription factor, whose deficiency drives ILC3 dysfunction and intestinal inflammation [37] (Figure 5E).
To functionally validate these observations, humanized mice were infected with HIV-1 followed by the administration of either IFNAR-blocking antibody or isotype control from weeks 6 to 10 post-infection (Figure 5F). Following euthanasia at the experimental endpoint (week 10), we quantitatively assessed the cytokine production capacity of splenic ILC3s through PMA/ionomycin stimulation coupled with multiparametric flow cytometry (Figure S2). Comparative analysis revealed the significant suppression of IL-22 + and IL-17 + ILC3 frequencies in HIV-1-infected mice versus the uninfected controls (Figure 5G,H). Importantly, IFNAR blockade restored cytokine production, demonstrating the reversal of HIV-induced functional impairment (Figure 5G,H).
Collectively, these findings establish IFN-I signaling as a key driver of ILC3 functional impairment during chronic HIV-1 infection, with targeted IFNAR inhibition showing the potential to reverse this immunopathological process.
## 3. Discussion
Accumulating evidence, including our prior work, has established that sustained IFN-I signaling during chronic HIV-1 infection exacerbates CD4 + T cell depletion, impairs adaptive immunity, and promotes viral persistence [17,18]. While these studies focused on adaptive immune compartments, the immunological consequences of IFN-I blockade on innate lymphoid populations remained unexplored. Our current investigation using humanized mouse models demonstrates that IFNAR inhibition restores both NK cell and ILC3 functionality, revealing previously unrecognized IFN-I-mediated mechanisms of innate immune dysregulation. These findings provide critical mechanistic insights into how targeted immunomodulation could complement existing therapies to address persistent immune dysfunction-a major obstacle to achieving HIV remission.
NK cells play dual roles in HIV-1 immunity through direct cytolytic activity and immunoregulatory cytokine production [38]. Here, we demonstrated that cART in HIV-1-infected humanized mice fails to restore NK cell function, a finding consistent with observations from virologically suppressed PLWH [14]. The restoration of canonical NK functionality through IFNAR blockade raises intriguing questions about IFN-I's role in promoting immunotolerant NK differentiation, a process evolutionarily conserved in maternal-fetal tolerance [39]. Moreover, given the established links between persistent NK dysfunction and increased cancer incidence [40] or opportunistic infections [41] in PLWH, our findings warrant clinical investigation into whether IFN-I modulation could reduce these comorbidities.
Regarding ILC3s, while previous studies reported numerical restoration through early cART initiation [15] or pDC depletion [32], we demonstrate that IFNAR blockade combined with cART functionally rescues this population by restoring IL-17/IL-22 production. This functional recovery suggests potential benefits for mucosal barrier integrity, given the established correlation between ILC3 loss, gut barrier disruption, and microbial translocation in chronic HIV-1 infection [16]. However, current humanized mouse models' limitations in recapitulating gut-associated lymphoid tissue (GALT) and human microbiota [42,43] preclude direct assessment of these physiological outcomes. Next-generation humanized models with improved GALT reconstruction could bridge this gap while enabling microbiome analysis. Furthermore, our study's focus on ILC3s necessitates future investigations into other ILC subsets (ILC1s/ILC2s), given their distinct roles in antiviral defense and tissue homeostasis [44].
Collectively, our findings delineate a central pathogenic role of chronic IFN-I signaling in driving innate lymphoid cell dysfunction during HIV-1 persistence. The dual restoration of NK cell cytotoxicity and ILC3 effector functions through IFNAR blockade not only advances our understanding of HIV-1 immunopathogenesis but also establishes a therapeutic paradigm for addressing the multifaceted immune dysfunction in PLWH. Future studies should explore the translational potential of combining IFN-I modulation with existing regimens to restore antiviral immunity and improve clinical outcomes.
## 4. Materials and Methods
## 4.1. Generation of Hu-Mice
NRG mice (NOD Rag2 -/-γc -/-) were obtained from The Jackson Laboratory (ME, USA). Humanized NRG-hu HSC mice were generated through the intrahepatic injection of 2 × 10 5 CD34 + hematopoietic progenitor cells isolated from human fetal liver tissue into neonatal (2-5 days postpartum) NRG pups, as previously described [21]. The engraftment efficiency was monitored by quantifying human CD45 + leukocytes in peripheral blood using flow cytometry at 10-12 weeks post-transplantation. Animals demonstrating over 20% human reconstitution (human CD45 + cells) were included in subsequent experiments. All mice were kept in a specific pathogen-free environment.
## 4.2. HIV-1 Infection of Humanized Mice
The HIV-1 JR-CSF infectious molecular clone (pYK-JRCSF) was acquired from the NIH AIDS Reagent Program (NIH, MD, USA, catalog #2708). Recombinant virus was generated by transfecting human embryonic kidney 293T cells (ATCC CRL-3216) with pYK-JRCSF plasmid DNA. Mice were anesthetized and infected with 10 ng of HIV-1 p24 equivalent (JR-CSF strain) via retro-orbital injection.
## 4.3. Combination Antiretroviral Therapy
We prepared medicated food pellets by incorporating three antiretroviral drugsemtricitabine (FTC, an NRTI, Gilead Sciences, CA, USA), tenofovir disoproxil fumarate (TDF, an NRTI from Truvada, Gilead Sciences, CA, USA), and raltegravir (RAL, an integrase inhibitor from Isentress, Merck, NJ, USA)-into standard rodent chow following an established protocol we published previously [17]. First, we crushed commercial tablets of each drug into a fine powder, which was then thoroughly homogenized with TestDiet 5B1Q (a modified LabDiet 5058 containing amoxicillin) before being pressed into ½-inch irradiated pellets. The final drug concentrations in the feed were 4800 mg/kg for raltegravir (targeting ~768 mg/kg/day), 1560 mg/kg for tenofovir disoproxil (targeting ~250 mg/kg/day), and 1040 mg/kg for emtricitabine (targeting ~166 mg/kg/day). These elevated concentrations were selected to ensure adequate drug exposure and therapeutic efficacy in our experimental model while maintaining the physical stability and palatability of the formulated diet.
## 4.4. IFNAR Blocking Antibody Generation and Treatments
Anti-IFNAR1 blocking antibodies were generated as previously described [17]. Antibodies were administered intraperitoneally twice a week. The initial dose was 400 µg per mouse, followed by 200 µg per mouse for subsequent treatments.
4.5. Isolation of mCD45 -hCD3 -hCD19 -hCD45 + Splenocytes from Humanized Mice Splenocytes (2 × 10 6 cells) were isolated from humanized mice three weeks after HIV-1 infection or mock treatment. For the initial negative selection, cells were incubated on ice for 20 min in staining buffer (1× PBS with 2% FBS and 2 mM EDTA) containing biotinconjugated anti-mouse CD45 (0.5 µg/10 6 cells), anti-human CD3 (0.5 µg/10 6 cells), and anti-human CD19 (0.5 µg/10 6 cells). Following two washes with cold staining buffer, cells were resuspended in 90 µL of buffer and incubated with 10 µL of Streptavidin MicroBeads (Miltenyi Biotech, MD, USA, catalog #130-048-101) for 20 min at 4 • C. Magnetic separation was performed using LS columns with a manual MACS separator (Miltenyi Biotech) according to the manufacturer's protocol.
The mCD45 -hCD3 -hCD19 -cell population underwent subsequent positive selection through incubation with biotinylated anti-human CD45 antibody (0.5 µg/10 6 cells in 100 µL staining buffer) for 20 min on ice. After washing, the cells were incubated with Streptavidin MicroBeads (10 µL/10 6 cells) and separated using MS columns as per the manufacturer's instructions.
Following magnetic separation, cell concentrations were quantified using a hemocytometer and adjusted to 1 × 10 6 cells/mL in 1× PBS containing 0.04% bovine serum albumin. Single-cell suspensions were immediately processed for 10x Genomics Chromium Single Cell 3 ′ Library construction (10x Genomics, CA, USA catalog #PN-1000268) according to the established protocols.
## 4.6. Single-Cell RNA Sequencing Library Preparation
The single-cell transcriptome library was prepared per the manufacturer's recommendations. Briefly, single cells were loaded onto a 10X Genomics Chromium chip (10X Genomics) to generate single-cell Gel Bead-in-emulsion (GEM). scRNA-Seq libraries were prepared using GemCode Single-Cell Gel Bead and Library Kit (10X Genomics). GEM reverse transcription was performed with the following conditions: 55 • C for 2 h, 85 • C for 5 min; held at 4 • C. After reverse transcription, GEMs were broken, and the single-strand cDNA was cleaned up with DynaBeads MyOne Silane Beads (Thermo Fisher Scientific, MA, USA) and the SPRIselect Reagent Kit (0.6× SPRI; Beckman Coulter, IN, USA). cDNA was amplified with the following condition: 98 • C for 3 min; 14 cycles of 98 • C for 15 s, 67 • C for 20 s, and 72 • C for 1 min; 72 • C for 1 min; held at 4 • C. The cDNA product was cleaned up using the SPRIselect Reagent Kit (0.6× SPRI; Beckman Coulter, IN, USA). Indexed sequencing libraries were constructed with the reagents in the GemCode Single-Cell 3 ′ Library Kit (10X Genomics). Sequencing was performed on Illumina NextSeq 500 with NextSeq 500/550 v2.5 kits.
## 4.7. Quality Control and Data Analysis of scRNA-Seq
Raw sequencing reads were processed and mapped to the hg19 human reference transcriptome using the Cell Ranger version 1.1.0 pipeline from 10X Genomics for individual scRNA-Seq datasets. The Seurat V4 R package was used to analyze the scRNA-Seq data.
## 4.8. Cell Clustering Analysis
High-quality transcriptomes that passed the quality control criteria (200-2500 genes and <15% mitochondrial genes) were log-normalized, scaled, and subjected to PCA spaces using the top 2000 highly variable genes (HVGs). Non-linear dimension reduction algorithms, including UMAP or t-SNE, were utilized to visualize the transcriptomes in 2D space. The Louvain algorithm was used to perform unsupervised clustering, and the assigned cell cluster was annotated based on their expression of canonical immune cell subtype markers. All the functions used for clustering analysis were wrapped in the Seurat V4 toolkit [45]. Cell type-specific feature gene acquisition or comparing DEGs of a cell type in different treatment groups was performed using the FindAllMarkers or FindMarkers function in the Seurat V4 toolkit.
## 4.9. Cell Type Correlation Analysis
The human spleen scRNA-seq data generated by Madissoon et al. [22] were obtained from https://www.tissuestabilitycellatlas.org (accessed on 14 March 2025). scRNAseq data of innate immune cells from human and humanized mice were first merged and preprocessed as mentioned above. The merged profiles were then subjected to the batch-corrected PCA space using the Harmony algorithm [46], followed by t-SNE analysis. The merged profiles were integrated following the Seurat CCA integration guide (https://satijalab.org/seurat/articles/integration_introduction (accessed on 14 March 2025)), and then phylogenetic analysis was performed and visualized using the BuildClusterTree and PlotClusterTree functions in the Seurat V4 toolkit. For Pearson's correlation analysis, scRNA-seq profiles of highly variable genes identified were first selected and then converted into pseudo-bulk profiles by calculating the average expression of the HVGs in each innate immune cell subpopulation. A Z-score scaling procedure was then performed on the pseudo-bulk profiles. Pearson's correlation coefficient was then calculated and visualized using the corrplot function in the R package corrplot.
## 4.10. Gene Clustering Analysis
Feature genes of a particular cell type in different treatment groups were first acquired by performing a one-versus-all Wilcoxson test using the FindAllMarkers function wrapped in the Seurat V4 toolbox. The function was run using default parameters with the expectation that only.pos was set to TRUE and logfc.threshold was set to 0.25. The feature genes were then clustered into 4 modules based on the Mfuzz algorithm. The genes in each module were then enriched for GOBP pathways. Enriched pathways with p-value > 0.05 or that appeared in more than 1 module were then filtered out. The clustering, enrichment, and visualization were performed using corresponding functions wrapped in the R package ClusterGVis (https://github.com/junjunlab/ClusterGVis (accessed on 17 March 2025)).
## 4.11. Gene Set Enrichment Analysis (GSEA)
GSEA was performed using the R package clusterProfiler [47], and the fene sets were obtained from the Molecular Signature Database (MSigDB, https://www.gsea-msigdb.org (accessed on 5 March 2025)).
## 4.12. Flow Cytometry
APC/Cy7-conjugated anti-human CD45 (HI30), BV421-conjugated anti-human CD127 (A019D5), PE/Cy7-conjugated anti-human CD117 (104D2), PerCP-Cy5.5-conjugated antihuman CD19 (HIB19), CD4(SK3), CD14(M5E2), CD16(3G8), CD20(2H7), CD34(581), CD123(6H6), CD11c(Bu15), CRTH2(BM16), PE-conjugated anti-human CD94(DX22), NKG2D(1D11), KIR(UP-R1), KLRG1(SA231A2), IL-22 (2G12A41), APC-conjugated antihuman CD56 (QA17A16), IL-17 (QA18A46), and FITC-conjugated anti-human CD9 (HI9a) were purchased from Biolegend (CA, USA). Pacific orange-conjugated anti-mouse CD45 (30-F11), PE/Texas red-conjugated anti-human CD3 (7D6), PE/Texas red-conjugated antihuman CD4 (RPA-T4), and a LIVE/DEAD Fixable Yellow Dead Cell Stain Kit were purchased from Invitrogen (CA, USA).
For surface marker staining, leukocytes were incubated with antibodies on ice for 30 min and then washed and fixed for analysis. For intracellular cytokine detection, freshly isolated cells were stimulated for 4 h with PMA (50 ng/mL) and ionomycin (1 µM) in the presence of BFA (1 µM). Cells were first stained with surface markers and then fixed and permeabilized with Cytofix/Cytoperm buffer (BD Bioscience, NJ, USA), followed by intracellular staining. Cells were analyzed on a CyAn ADP flow cytometer (Dako, CA, USA). Data were analyzed using Summit 4.3 software (Dako, CA, USA).
## 4.13. Statistics
All experimental data analyses were performed using GraphPad Prism 10.1.0 (Graph-Pad Software, CA, USA). Two-tailed, unpaired Student's t-tests were employed for the statistical significance query in pairwise comparisons. For multi-group comparisons, data were analyzed by Brown-Forsythe and Welch ANOVA tests, followed by Dunnett's T3 multiple comparisons test. Results with p < 0.05 were considered to be statistically significant. Asterisks resemble significance levels (* p < 0.05, ** p <0.01, *** p < 0.001, **** p < 0.0001).
## 4.14. Study Approval
All animal experiments were conducted in strict compliance with the NIH guidelines for laboratory animal housing and care, with protocols approved by the University of North Carolina Institutional Animal Care and Use Committee (Protocol ID 17-071). Human fetal liver tissues (gestational age: 16-20 weeks) were obtained from medically indicated or elective termination of pregnancy via Advanced Bioscience Resources, a nonprofit organization coordinating with outpatient clinics. Written informed consent was systematically obtained from all maternal donors in accordance with clinic regulations. The experimental protocol underwent formal review by the University of North Carolina's Office of Human Research Ethics, which formally determined that this study does not meet the regulatory definition of human subjects research under the applicable federal guidelines [45 CFR 46.
## Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/v17081099/s1, Figure S1: Gating strategies for human NK cells from the spleens of humanized mice. Flow cytometry of viable human splenocytes (Y7-mCD45-hCD45+) stained for CD3 and CD56; Figure S2: Gating strategies for human ILC3s from the spleens of humanized mice. Flow cytometry of viable human splenocytes (Y7-mCD45-hCD45+) stained for lineage markers (CD3, CD14, CD16, CD19, CD20, CD123, CD11c, CD34, and CRTH2) and for CD127+CD117+ ILC3s; Table S1: Cluster markers; Table S2: DEGs comparing cells in different groups; Table S3: Top 10 GOBP enrichment of DEGs; Table S4: Clustered genes and enriched pathways in NKs; Table S5: Clustered genes and enriched pathways in ILC3s.
## References
1. Archin, Sung, Garrido et al. (2014) "Eradicating HIV-1 infection: Seeking to clear a persistent pathogen" *Nat. Rev. Microbiol*
2. Ghosn, Taiwo, Seedat et al. (2018) *Lancet*
3. Simon, Ho, Karim (2006) "HIV/AIDS epidemiology, pathogenesis, prevention, and treatment" *Lancet*
4. Deeks (2011) "HIV infection, inflammation, immunosenescence, and aging" *Annu. Rev. Med*
5. Vivier, Artis, Colonna et al. (2018) "Innate Lymphoid Cells: 10 Years On" *Cell*
6. Florez-Alvarez, Hernandez, Zapata (2018) "NK Cells in HIV-1 Infection: From Basic Science to Vaccine Strategies" *Front. Immunol*
7. Jackson, Zhang, Kiani et al. (2017) "HIV exposed seronegative (HESN) compared to HIV infected individuals have higher frequencies of telomeric Killer Immunoglobulinlike Receptor (KIR) B motifs; Contribution of KIR B motif encoded genes to NK cell responsiveness" *PLoS ONE*
8. Marras, Nicco, Bozzano et al. (2013) "Natural killer cells in HIV controller patients express an activated effector phenotype and do not up-regulate NKp44 on IL-2 stimulation" *Proc. Natl. Acad. Sci*
9. May, Pohlmeyer, Kwaa et al. (2020) "Combined effects of HLA-B* 57/5801 elite suppressor CD8+ T cells and NK cells on HIV-1 replication" *Front. Cell. Infect. Microbiol*
10. Sugawara, Reeves, Jost (2022) "Learning to Be Elite: Lessons From HIV-1 Controllers and Animal Models on Trained Innate Immunity and Virus Suppression" *Front. Immunol*
11. Anderko, Depuyt, Bronson et al. (2024) "Persistence of a Skewed Repertoire of NK Cells in People with HIV-1 on Long-Term Antiretroviral Therapy" *J. Immunol*
12. Mavilio, Benjamin, Daucher et al. (2003) "Natural killer cells in HIV-1 infection: Dichotomous effects of viremia on inhibitory and activating receptors and their functional correlates" *Proc. Natl. Acad. Sci*
13. Lichtfuss, Cheng, Farsakoglu et al. (2012) "Virologically suppressed HIV patients show activation of NK cells and persistent innate immune activation" *J. Immunol*
14. Nabatanzi, Bayigga, Cose et al. (2019) "Aberrant natural killer (NK) cell activation and dysfunction among ART-treated HIV-infected adults in an African cohort" *Clin. Immunol*
15. Kloverpris, Kazer, Mjosberg et al. (2016) "Innate Lymphoid Cells Are Depleted Irreversibly during Acute HIV-1 Infection in the Absence of Viral Suppression" *Immunity*
16. Kramer, Goeser, Lutz et al. (2017) "Compartment-specific distribution of human intestinal innate lymphoid cells is altered in HIV patients under effective therapy" *PLoS Pathog*
17. Cheng, Ma, Li et al. (2017) "Blocking type I interferon signaling enhances T cell recovery and reduces HIV-1 reservoirs" *J. Clin. Investig*
18. Cheng, Yu, Li et al. "Type I interferons suppress viral replication but contribute to T cell depletion and dysfunction during chronic HIV-1 infection" *J. Clin*
19. Li, Cheng, Nunoya et al. (2014) "Plasmacytoid dendritic cells suppress HIV-1 replication but contribute to HIV-1 induced immunopathogenesis in humanized mice" *PLoS Pathog*
20. Zhen, Rezek, Youn et al. (2017) "Targeting type I interferon-mediated activation restores immune function in chronic HIV infection" *J. Clin. Investig*
21. Cheng, Yu, Wrobel et al. (2020) "Identification of pathogenic TRAIL-expressing innate immune cells during HIV-1 infection in humanized mice by scRNA-Seq" *J. Clin. Investig*
22. Madissoon, Wilbrey-Clark, Miragaia et al. (2019) "scRNA-seq assessment of the human lung, spleen, and esophagus tissue stability after cold preservation" *Genome Biol*
23. Fernandez, Tanaskovic, Helbig et al. (1927) "CD4+ T-cell deficiency in HIV patients responding to antiretroviral therapy is associated with increased expression of interferon-stimulated genes in CD4+ T cells" *J. Infect. Dis*
24. Dunham, Vujkovic-Cvijin, Yukl et al. (2014) "Discordance between peripheral and colonic markers of inflammation during suppressive ART" *J. Acquir. Immune Defic. Syndr*
25. Borkar, Yin, Venturi et al. (2025) "Youth Who Control HIV on Antiretroviral Therapy Display Unique Plasma Biomarkers and Cellular Transcriptome Profiles Including DNA Repair and RNA Processing"
26. Koopman, Kopcow, Rybalov et al. (2003) "Human decidual natural killer cells are a unique NK cell subset with immunomodulatory potential" *J. Exp. Med*
27. Orrantia, Vazquez-De Luis, Astarloa-Pando et al. "In vivo expansion of a CD9(+) decidual-like NK cell subset following autologous hematopoietic stem cell transplantation"
28. Umotoy, Kroon, Man et al. (2024) "Inhibition of HIV-1 replication by nanobodies targeting tetraspanin CD9"
29. Martinet, Smyth (2015) "Balancing natural killer cell activation through paired receptors" *Nat. Rev. Immunol*
30. Houchins, Lanier, Niemi et al. (1997) "Natural killer cell cytolytic activity is inhibited by NKG2-A and activated by NKG2-C" *J. Immunol*
31. Wada, Matsumoto, Maenaka et al. (2004) "The inhibitory NK cell receptor CD94/NKG2A and the activating receptor CD94/NKG2C bind the top of HLA-E through mostly shared but partly distinct sets of HLA-E residues" *Eur. J. Immunol*
32. Zhang, Cheng, Zhao et al. (2015) "Plasmacytoid dendritic cells promote HIV-1-induced group 3 innate lymphoid cell depletion" *J. Clin. Investig*
33. Zhao, Cheng, Wang et al. (2018) "Infection and depletion of CD4+ group-1 innate lymphoid cells by HIV-1 via type-I interferon pathway" *PLoS Pathog*
34. Jarade, Di Santo, Serafini (2021) "Group 3 innate lymphoid cells mediate host defense against attaching and effacing pathogens" *Curr. Opin. Microbiol*
35. Serafini, Di Santo (2024) "Group 3 innate lymphoid cells: A trained Gutkeeper" *Immunol. Rev*
36. Wang, Gao, Clark et al. (1707) "CARD8 is an inflammasome sensor for HIV-1 protease activity" *Science*
37. Hao, Liu, Gu et al. (2024) "Dysregulation of Wnt/beta-catenin signaling contributes to intestinal inflammation through regulation of group 3 innate lymphoid cells" *Nat. Commun*
38. Bjorkstrom, Strunz, Ljunggren (2022) "Natural killer cells in antiviral immunity" *Nat. Rev. Immunol*
39. Jabrane-Ferrat (1397) "Features of Human Decidual NK Cells in Healthy Pregnancy and During Viral Infection" *Front. Immunol*
40. Grulich, Van Leeuwen, Falster et al. (2007) "Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: A meta-analysis" *Lancet*
41. Brooks, Kaplan, Holmes et al. (2009) "HIV-associated opportunistic infections--going, going, but not gone: The continued need for prevention and treatment guidelines" *Clin. Infect. Dis*
42. Cao, Shores, Hu-Li et al. (1995) "Defective lymphoid development in mice lacking expression of the common cytokine receptor gamma chain" *Immunity*
43. Nochi, Denton, Wahl et al. (2013) "Cryptopatches are essential for the development of human GALT" *Cell Rep*
44. Mckenzie, Spits, Eberl (2014) "Innate lymphoid cells in inflammation and immunity" *Immunity*
45. Hao, Hao, Andersen-Nissen et al. (2021) "Integrated analysis of multimodal single-cell data" *Cell*
46. Korsunsky, Millard, Fan et al. (2019) "Fast, sensitive and accurate integration of single-cell data with Harmony" *Nat. Methods*
47. Wu, Hu, Xu et al. "A universal enrichment tool for interpreting omics data"
48. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12363160&blobtype=pdf | # A new era in gammaherpesvirus transcriptomics: high-resolution profiling and model development
Kiran Fida, Brent Stanfield
## Abstract
presents the first high-resolution, long-read transcriptomic atlas of Caviid gammaherpesvirus 1 (CaGHV-1), offering insights into its transcriptional complexity. Using nanopore direct RNA and cDNA sequencing, the study maps transcription start sites, polyadenylation signals, alternative splicing, and upstream open reading frames (uORFs), revealing a landscape of coding and non-coding RNAs with pervasive transcriptional overlaps. These features underscore the evolutionary conservation of key regulatory mechanisms across gammaherpesviruses, notably replication and transcription activator (RTA)-medi ated transcriptional control. By establishing CaGHV-1 as a promising model for Kaposi's sarcoma-associated herpesvirus (KSHV)-related disease, the work sets a foundation for studies of viral gene regulation, immune evasion, and pathogenesis, including bacterial artificial chromosomes (BACs) for genetic manipulation. Here, we discuss the advan ces in the study as well as how some limitations remain regarding read coverage biases, sequencing errors, and the uncharacterized functions of non-coding RNAs and emphasize the need for validation and functional assays. This study provides a valua ble resource for understanding gammaherpesvirus biology and advancing translational research in viral pathogenesis and therapeutic development.KEYWORDS guinea pig, gamma herpesvirus, small animal model, transcriptomics, CaGHV-1, Caviid gammaherpesvirus 1 A recent article by Torma et al. (1) marks a significant advancement in herpesvi rus research by providing the first comprehensive, long-read sequencing-based transcriptomic atlas for Caviid gammaherpesvirus 1 (CaGHV-1). This research moves beyond basic transcript cataloging by achieving unprecedented resolution in mapping transcriptional start sites, polyadenylation signals, and alternative splicing events, which are critical for understanding the full complexity of CaGHV-1 gene regulation (2). These findings not only refine our understanding of gammaherpesvirus transcriptional architecture but also highlight CaGHV-1 as a powerful tool for modeling Kaposi's sarcoma-associated herpesvirus (KSHV-related pathogenesis and developing targeted therapeutics (3).Torma et al. used nanopore-based direct RNA sequencing (dRNA-Seq) and direct cDNA sequencing (dcDNA-Seq) to create a high-resolution atlas of the CaGHV-1 transcriptome. Nanopore sequencing is advantageous because it allows for the direct sequencing of native RNA molecules, thereby preserving post-transcriptional modifica tions and avoiding amplification biases inherent in other methods (4). This approach enabled the precise identification of transcriptional start and stop sites, alternative splice variants, and regulatory motifs. The study reveals that CaGHV-1 exhibits a remarkably rich and intricate transcriptional landscape, with extensive use of alternative splicing,
transcript isoforms, and overlapping RNAs, which surpasses what has previously been described for non-human gammaherpesviruses.
The analysis not only identified the anticipated coding transcripts but also discovered numerous non-coding RNAs (ncRNAs), including polycistronic messages and replication origin-associated RNAs (raRNAs). These ncRNAs are crucial in the regulation of viral gene expression and host-virus interactions. Transcriptional overlaps-encompassing convergent, divergent, and co-oriented transcripts-are pervasive across the CaGHV-1 genome. This mirrors, yet also uniquely extends, the regulatory strategies reported in Kaposi's sarcoma-associated herpesvirus (KSHV) and Epstein-Barr virus (EBV), indicating a conserved, yet distinct network of transcriptional regulation across gammaherpesviruses (2).
A pivotal discovery from the manuscript is the conservation of replication and transcription activator (RTA)-mediated transcriptional regulation: CaGHV-1's ORF50, which encodes RTA, demonstrates key regulatory features akin to those in KSHV, suggesting evolutionary conservation not just in gene content but in molecular reactivation mechanisms (5). This finding is crucial because RTA is a master regulator of the lytic replication cycle in gammaherpesviruses, and its conservation suggests common mechanisms of viral reactivation. These findings set the stage for mechanis tic studies of lytic reactivation and latency control using CaGHV-1 as an experimental system.
A notable aspect of this study is the identification of uORFs in CaGHV-1. While KSHV uORFs are positioned near ATG start codons, facilitating translational regulation (6). CaGHV-1 uORFs are located farther upstream. This suggests a distinct translational control strategy that may impact protein synthesis efficiency and viral gene expression. These findings offer an opportunity to explore how viral uORFs influence pathogenesis and immune evasion across different gammaherpesviruses.
## ESTABLISHING CAGHV-1 AS A PLATFORM FOR PATHOGENESIS MODELS
By documenting extensive parallels in transcriptional complexity and architecture with human disease-associated gammaherpesviruses, Torma et al. provide strong support for CaGHV-1 as a relevant model to dissect viral gene regulation, explore immune evasion tactics, and develop preclinical interventions for KSHV-related diseases. The RNA atlas serves as a crucial foundation for future functional and comparative virology studies and will likely facilitate the development of novel antiviral strategies.
## FUNCTIONAL EXPLORATION OF NON-CODING RNAs
While comprehensive sequencing confirms robust production of a diverse array of CaGHV-1 non-coding RNAs, their biological functions remain uncharacterized. Like KSHV, CaGHV-1 expresses numerous polycistronic and replication origin-associated RNAs, which are hypothesized to influence replication, immune evasion, and latency main tenance. However, the precise regulatory roles of these ncRNAs are yet to be discov ered. The study of viral microRNAs and their functions has been extensively reviewed, highlighting their importance in modulating host-virus interactions and influencing viral pathogenesis (7). However, the precise regulatory roles of these ncRNAs are yet to be discovered.
Future research should center on elucidating the functions and mechanistic actions of these ncRNAs through discovery-driven approaches and functional assays (e.g., RNA immunoprecipitation, transcriptome-wide association studies). RNA immunoprecipita tion followed by sequencing (RIP-Seq) is a powerful technique to identify RNAs that interact with specific RNA-binding proteins, thereby providing insights into the function of these ncRNAs. Determining whether CaGHV-1 ncRNAs interface with host immune response pathways-akin to the known activities of KSHV PAN RNA or viral miRNAs-will be essential for understanding viral persistence, reactivation, and potential oncogenicity.
## MODELING GAMMAHERPESVIRUS INFECTION AND DISEASE: FROM TRAN SCRIPTOME TO PATHOGENESIS
The detailed transcriptomic map presented in the companion article provides a platform for translational studies leveraging CaGHV-1 in guinea pigs. Early in vivo work has shown that CaGHV-1 induces mild lymphoproliferative responses in its natural host, resembling mononucleosis-like syndromes, highlighting its biological relevance as a gammaherpes virus (8). By coupling advanced transcriptomic profiling with targeted molecular tools (such as recombinant virus engineering), future studies can now dissect the pathogene sis, latency, and transformation potential of CaGHV-1 in vivo. Beyond recapitulating KSHV infection features, the development of a CaGHV-1 bacterial artificial chromosome (BAC), as outlined, will further empower studies of gene function, allow the creation of chimeric viruses, and accelerate translational research targeting KSHVspecific interventions-all made possible by the viral genomic and transcriptomic resources generated in the companion article. The construction of BACs will enable the manipulation of the viral genome, allowing for the creation of mutants and recombinant viruses to study gene function and pathogenesis (9).
## LIMITATIONS OF THE STUDY
Despite its contributions, the study has certain limitations that warrant considera tion. While long-read sequencing provides superior transcript resolution, it is prone to sequencing errors and requires rigorous validation. The authors address this by integrating multiple sequencing approaches and bioinformatics pipelines; however, further validation through independent experimental methods could strengthen the conclusions. Additionally, the functional significance of many detected transcripts remains speculative. While the study suggests possible roles for these RNAs based on conservation and genomic context, further experimental validation is necessary to confirm their biological relevance. Moreover, detecting low-abundance transcripts via ONT remains difficult due to biases in read coverage. While direct RNA sequencing preserves RNA modifications, it has lower throughput and read quality compared with cDNA sequencing. Another challenge is read length variability, which complicates the distinction of isoforms and necessitates advanced bioinformatics tools for accurate interpretation (10).
## CONCLUSIONS
Torma et al. have set a new benchmark in gammaherpesvirus transcriptomics by mapping the full transcriptional architecture of CaGHV-1 using state-of-the-art long-read sequencing. Their findings redefine the scope of viral gene and ncRNA expression in animal models and establish CaGHV-1 as both a model of bio-complexity and a practical surrogate for dissecting KSHV biology. This work paves the way for future discovery and therapeutic validation in a mammalian host with greater translational relevance to human disease.
## References
1. Torma, Dörmő, Fülöp et al. (2025) "Long-read transcriptomics of caviid gammaherpesvirus 1: compiling a comprehensive RNA atlas" *mSystems*
2. Prazsák, Tombácz, Fülöp et al. (2024) "a state-of-the-art annotation of the Kaposi's sarcoma-associated herpesvirus transcrip tome using cross-platform sequencing" *mSystems*
3. Stanfield, Ruiz, Chouljenko et al. (2024) "Guinea pig herpes like virus is a gamma herpesvirus" *Virus Genes*
4. Workman, Tang, Tang et al. (2019) "Nanopore native RNA sequencing of a human poly(A) transcriptome" *Nat Methods*
5. Damania, Jeong, Bowser et al. (2004) "Comparison of the Rta/Orf50 transactivator proteins of gamma-2herpesviruses" *J Virol*
6. Kronstad, Brulois, Jung et al. (2013) "Dual short upstream open reading frames control translation of a herpesviral polycistronic mRNA" *PLoS Pathog*
7. Grundhoff, Sullivan (2011) "Virus-encoded microRNAs" *Virology (Auckland)*
8. Dowler, Mccormick, Armstrong et al. (1984) "Lymphoproli ferative changes induced by infection with a lymphotropic herpesvirus of guinea pigs" *J Infect Dis*
9. Brulois, Chang, Lee et al. (2012) "Construction and manipulation of a new Kaposi's sarcoma-associated herpesvirus bacterial artificial chromosome clone" *J Virol*
10. Wang, Zhao, Bollas et al. (2021) "Nanopore sequencing technology, bioinformatics and applications" *Nat Biotechnol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12476817&blobtype=pdf | # OraQuick hepatitis C virus self-test: A new frontier in hepatitis C screening
Muneeb Saifullah, Mavra Khan, Muhammad Usman, Qasim Mehmood, Abbas Mehdi, Wu Jn, Usman Ma
## Abstract
According to the World Health Organization, an estimated 58 million people worldwide are chronically infected with hepatitis C virus (HCV), yet only about 20% have been formally diagnosed. Traditional laboratory-based antibody and RNA assays require infrastructure and trained personnel, limiting their uptake in resource-limited and hard-to-reach settings. The OraQuick HCV self-test (HCVST) is the first World Health Organization-prequalified HCVST, which delivers results in 20-40 min via an easy-to-use gum-swab format. Field evaluations report a sensitivity of about 97%-98% and a specificity of about 99%-100% that are comparable with those of blood-based lateral-flow assays (e.g., Alere Truline, SD Bioline). Usability studies demonstrated an acceptability rate of over 90% and a correct self-test completion rate of over 85% in key populations. HCVST with the Ora-Quick HCVST kit provides a practical, evidence-based approach to closing diagnostic gaps, particularly among underserved or stigmatized populations. To maximize the public health impact, programs should integrate self-testing into national screening algorithms, ensure linkage to RNA confirmation and treatment, and consider economic and operational contexts.
## TO THE EDITOR
Hepatitis C virus (HCV) is an RNA-enveloped virus that primarily infects human hepatocytes [1]. It has a disease burden of 58 million people worldwide, and only 20% have been diagnosed [2]. Most individuals infected with HCV do not present acutely; they progress to chronic liver disease, which manifests as cirrhosis, portal hypertension, hepatocellular carcinoma, and hepatic encephalopathy. The transition from acute HCV infection to chronic liver disease is subclinical [3]. Timely detection of HCV infection is crucial for improving clinical outcomes and reducing the global financial burden of the disease [4].
Despite significant advancements in treatment, diagnosis remains a barrier, particularly in resource-limited settings. Traditional diagnostic methods necessitate laboratory infrastructure and trained personnel, limiting accessibility in many parts of the world. The advent of self-testing kits represents a pivotal advancement in diagnostic accessibility and patient empowerment, bridging this gap [5]. Previously, HCV diagnostic tests included enzyme immunoassays and nucleic acid tests. Although accurate, these methods are not ideal for widespread screening due to their high cost and limited accessibility in resource-limited settings. Point-of-care testing has emerged as a solution for improving accessibility, exemplified by innovations such as the use of Gene X-pert HCV RNA and various rapid diagnostic tests (RDTs). However, these still often require skilled personnel and specialized equipment. Nevertheless, the introduction of the OraQuick HCV selftesting (HCVST) kits, designed for home use, was not available until recently, marking an advancement in HCV detection [6].
## PATHOPHYSIOLOGY OF HCV
HCV enters hepatocytes via endocytosis by binding to cell surface coreceptors. Once internalized, the HCV-positive RNA strand is uncoated and released into the host cytoplasm, where it is translated into 10 mature proteins. These mature peptides are cleaved by host proteases and viral-encoded proteases. These peptides are inserted into the endoplasmic reticulum and function as a replication complex, converting positive-strand RNA into negative-strand RNA. The negative strand acts as a template, and then new positive RNA strands are produced. These strands are packaged with the core and glycoprotein envelope, forming mature HCV, which is released via exocytosis [7]. It is not the virus itself that destroys the hepatocytes but rather the slow cellular response of CD4+ and CD8+ cells that leads to hepatic necrosis [8].
## HIGH-RISK GROUPS AND EPIDEMIOLOGY
According to the Centers for Disease Control and Prevention, the high-risk population includes people who inject drugs, males who have sex with males, those living with HIV, those who have undergone hemodialysis, blood transfusions, and organ transplantation, infants born to HCV-positive mothers, and health care professionals who are exposed to needle stick injuries and sharp and mucosal exposures [9]. According to the World Health Organization (WHO), approximately 58 million people are living with HCV infection globally (Figure 1). The Eastern Mediterranean Region bears the largest share of chronic HCV infections (12 million), followed by 9 million each in Southeast Asia and Europe, 7 million in the Western Pacific, 8 million in Africa, and 5 million in the Americas [2].
## DIAGNOSTIC CHALLENGES
The diagnosis of current HCV infection is based on two phases: Screening and confirmatory testing. Screening includes ELISA or rapid tests, followed by confirmatory testing, such as a nucleic acid tests for HCV RNA (viral load) [10,11]. Although these antibody assays are sensitive and specific, their limited accessibility and acceptability make the diagnosis and treatment of HCV hard in low-income countries. Moreover, false-positive tests lead to unnecessary confirmatory HCV RNA testing. These standard RNA confirmatory tests are expensive and complex and can only be performed in centralized laboratories [12]. The WHO aims to detect 90% of people living with HCV by 2030. This goal can only be achieved using rapid, readily available, and acceptable tests [13,14]. The comparison of HCV diagnostic methods is well summarized in Table 1.
## RDT
RDTs can quickly detect infections. These tests provide results in 20-40 min, facilitating timely medical decisions and do not require additional infrastructure. The rapid nature of these tests helps bridge the gap between diagnosis and healthcare access, especially in underserved populations. A range of rapid HCV diagnostics tests is available. Table 2 summarizes the key characteristics of the OraQuick HCVST and other point-of-care assays. Rapid diagnostic tests, such as the OraQuick HCVST and other standard tests, including the AlereTruline HCV rapid test and the Abbott SD Bioline HCV test, achieve high accuracy in antibody detection [15,16].
## RISE OF SELF-TESTING: ORAQUICK HCV
Self-testing is a strategy in which patients collect their samples, perform the test, and interpret the results privately. This A study in Vietnam found that 90% of the highrisk population had accepted self-testing [20]. A systematic review and meta-analysis by Perazzo et al [21] revealed that the pooled estimates for correct sample collection and people who did not require assistance in any step while performing HCVST were 87.2% and 62.8%, respectively. The OraQuick HCVST has several prerequisites, including a minimum age of 18 years. The patient should refrain from eating or drinking anything for at least 15 min before the test and from using any usual oral care products for 30 min before the test. The test device of the OraQuick HCVST features a result window that displays a positive result with two red lines at points C and T. In contrast a negative result is indicated by a single line at point C only within the result window. OraSure (parent company) recommends that patients wait 20 min before reading the results and should not read the results after 40 min. This test is relatively easy to use and interpret, requiring only basic education [22].
If an individual tests positive, patients should visit the nearest healthcare setting to evaluate their test results further and discuss possible treatment options. The Centers for Disease Control and Prevention recommends following the algorithm (Figure 2) for managing patients who screen positive on antibody-based screening tests [23].
## References
1. Pietschmann, Brown (2019) "Hepatitis C Virus" *Trends Microbiol*
2. Westbrook, Dusheiko (2014) "Natural history of hepatitis C" *J Hepatol*
3. Gupta, Bajpai, Choudhary (2014) "Hepatitis C virus: Screening, diagnosis, and interpretation of laboratory assays" *Asian J Transfus Sci*
4. Feld (2018) "Hepatitis C Virus Diagnostics: The Road to Simplification" *Clin Liver Dis (Hoboken)*
5. Reipold, Shilton, Donolato et al. (2024) "Molecular Point-of-Care Testing for Hepatitis C: Available Technologies, Pipeline, and Promising Future Directions" *J Infect Dis*
6. Basit, Tyagi, Koirala et al. (2023) "Treasure Island (FL): StatPearls Publishing"
7. Saraceni, Birk (2021) "A Review of Hepatitis B Virus and Hepatitis C Virus Immunopathogenesis" *J Clin Transl Hepatol*
8. (2025) "CDC Clinical Overview of Hepatitis C"
9. Joshi (2014) "Hepatitis C screening" *Ochsner J*
10. Majid, Gretch (2002) "Current and future hepatitis C virus diagnostic testing: problems and advancements" *Microbes Infect*
11. Baber, Suganthan, Ramasamy (2024) "Current advances in Hepatitis C diagnostics" *J Biol Eng*
12. (2025) "Global Health Sector Strategies on, Respectively, HIV, Viral Hepatitis and Sexually Transmitted Infections for the Period 2022-2030"
13. Mane, Sacks, Sharma et al. (2019) "Evaluation of five rapid diagnostic tests for detection of antibodies to hepatitis C virus (HCV): A step towards scale-up of HCV screening efforts in India" *PLoS One*
14. Cha, Park, Kang et al. (2013) "Performance evaluation of the OraQuick hepatitis C virus rapid antibody test" *Ann Lab Med*
15. Shin, Kim, Bae et al. (2025) "Self-testing strategy to eliminate hepatitis C as per World Health Organization's goal: Analysis of disease burden and cost-effectiveness" *Clin Mol Hepatol*
16. (2025) "WHO Prequalifies the First Self-Test for Hepatitis C Virus"
17. Nguyen, Nguyen, Ai et al. (2021) "Acceptability and Usability of HCV Self-Testing in High Risk Populations in Vietnam" *Diagnostics (Basel)*
18. Perazzo, Castro, Villela-Nogueira et al. (2023) "Acceptability and usability of oral fluid HCV self-testing for hepatitis C diagnosis: A systematic review and meta-analysis" *J Viral Hepat*
19. (2025) "24 National Institute for Health and Care Excellence. The OraQuick HCV Point-of-Care Test for Rapid Detection of Hepatitis C Virus Antibodies"
20. Walker, Ivanova, Jamil et al. (2023) "Cost-effectiveness of Hepatitis C virus self-testing in four settings" *PLOS Glob Public Health* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12282407&blobtype=pdf | # Severe enterovirus A71 infection is associated with dysfunction of T cell immune response and alleviated by Astragaloside A
Chong Wang, Muhan Huang, Bingyu Guo, Xi Zhou, Zongqiang Cui, Yi Xu, Yujie Ren, Virologica Sinica
## Abstract
Enterovirus A71 (EV-A71) is the major causative pathogen for severe hand-foot-mouth disease (HFMD), a predominantly childhood-associated communicable disease. The mechanisms that children manifest severe disease progression while adults typically exhibit milder or asymptomatic infections remain incompletely characterized, which hinders the development of effective therapy against this disease. Herein, using the newborn mouse model of EV-A71 infection, we uncovered that the underdevelopment of T cells closely associated with the severity of EV-A71 infection, and EV-A71 infection dramatically impaired T-cell immune response. Moreover, the dysfunction of T-cell immunity contributes to the pathogenesis of EV-A71 infection, as the loss of T cells made neonatal mice highly vulnerable to EV-A71 infection. To further assess the relationship between T-cell immunity and HFMD, we enrolled a cohort of 145 pediatric patients with laboratory-confirmed EV-A71 infection and found that the compromised T-cell immune response is associated with the severity of EV-A71-caused HFMD in these children. Furthermore, we found that the treatment of newborn mice with Astragaloside A, a saponin from the medicinal herb Astragalus membranaceus, showed potent in vivo therapeutic efficacy against EV-A71 infection in a T-cell-dependent manner. In conclusion, these findings uncover the interaction between EV-A71 infection and T-cell immunity, provide novel insights onto the physiological impacts of T cells on the pathogenesis of EV-A71 infection and HFMD, and find a promising immunotherapeutic strategy to treat this viral disease.
## INTRODUCTION
Enteroviruses are a large group of non-enveloped, positive-sense and single-stranded RNA viruses belonging to the genus Enterovirus of the family Picornaviridae. Enteroviruses include numerous important human pathogens such as poliovirus, enterovirus A71 (EV-A71), coxsackieviruses, and echoviruses, and cause approximately 3 billion human infections per year in the world. The symptoms caused by enteroviral infections range from relatively mild conditions like upper respiratory illness and common cold to severe or even lethal outcomes such as severe hand-foot-and-mouth disease (HFMD), aseptic meningitis, encephalitis, myocarditis, neonatal sepsis-like disease, and poliomyelitis (Pallansch, 2007).
HFMD is a common pediatric infectious disease caused by infections with EV-A71 and some other enteroviruses including coxsackieviruses A16 (CV-A16), A6 (CV-A6), etc. There are millions of HFMD cases in children under 5 years old globally every year (Chan et al., 2000;Fioravanti, 2012;Ho et al., 1999;Tan et al., 2011). Although the symptoms of HFMD are usually moderate, this viral disease is highly contagious and spreads quickly at schools and kindergartens. Neurotropic EV-A71 is a major causative pathogen for HFMD, and EV-A71 infection is more frequently associated with severe central-nervous-system complications in HFMD and thereby is the major cause of fatalities in HFMD (Huang et al., 1999;Solomon et al., 2010). Thus, EV-A71 is considered as the most virulent pathogen within HFMD-causing enteroviruses and a serious threat to the health of children across the globe. Thus far, there is no effective antiviral drug or therapy available to specifically treat EV-A71 infection and HFMD.
Although EV-A71 infection and HFMD can occur in both children and adults, adult infection of this virus is usually asymptomatic, while severe symptoms are normally observed in children at preschool ages (Deng et al., 2011;Solomon et al., 2010;Zhu et al., 2015). Compared with adults, young children have less developed cellular immunity that is pivotal for controlling viral infection (Carsetti et al., 2020;Olin et al., 2018;Simon et al., 2015). Besides, a number of clinical observations have suggested the correlation between compromised cellular immunity and severe symptoms of enterovirus 71 (EV71) infection (Chang et al., 2006(Chang et al., , 2008;;Yang et al., 2001). However, the relationship between EV-A71 infection and T cells as well as the physiological impacts of T cell immune response on severe EV-A71 infection and HFMD are still poorly understood, which hinders the development of effective therapeutic strategy for HFMD.
In this study, using the newborn mouse model of EV-A71 infection, we uncovered that the underdevelopment of T cells, but not macrophages, dendritic cells (DCs), or B cells, was closely associated with the severity of EV-A71 infection. And EV-A71 infection dramatically impaired T cell immune response in newborn mice. Besides, the dysfunction of T cell immune response contributes to the pathogenesis of EV-A71 infection, which was further supported by our cohort study involving 145 pediatric patients with laboratory-confirmed EV-A71 infection and different symptoms. Furthermore, after revealing the relationship between T-cell immune response and the severity of EV-A71 infection and HFMD, we aimed to find an immunotherapy that can boost T cell immune response to treat EV-A71 infection. We found that the treatment of newborn mice with Astragaloside A (AGS-A), the primary saponin extracted from the traditional Chinese medicinal herb Astragalus membranaceus (Auyeung et al., 2016), effectively rescued the EV-A71-caused impairment of T cell immune response, activated T cells, and showed potent in vivo therapeutic efficacy against EV-A71 infection in a T-cell-dependent manner, while AGS-A treatment showed no direct antiviral effect in T-cell-deficient mice.
## RESULTS
## T cell development in newborn mice was associated with the severity of EV-A71 infection
We infected 1-, 3-, 7-or 14-day-old Institute of Cancer Research (ICR) mice with 1 Â 10 7 PFU of EV-A71 by intraperitoneal (i.p.) injection (Fig. 1A). Our data showed that the 1-and 3-day-old mice developed to clinical symptoms from 5 d.p.i. and the mortality rates of the 1-and 3-day-old mice were 90% and 70%, respectively. Moreover, although the 7-day-old mice had post-inoculation symptoms, their mortality rate was 40%. Of note, all the 14-day-old mice survived without any symptoms after infection (Fig. 1B and Supplementary Fig. S1A). In addition, the viral loads in the spleen, muscle, and brain tissues of EV-A71-infected 3-day-old mice were significantly higher compared to that in EV-A71infected 14-day-old mice (Supplementary Fig. S1B).
To examine the status of cellular immunity, especially T cells, in neonates of different ages, we examined the percentages of
macrophages and CD11c þ dendritic cells (DCs) in the spleen and TCR-β þ T cells in the thymus of naive mice by flow cytometry on days 1, 3, 7 and 14 after birth. As shown in Fig. 1C-E
cells displayed the upward trend along with age. On the other hand, the percentage of macrophages in spleen decreased along with age (Supplementary Fig. S2A), while the percentages of CD11c þ DCs and CD19 þ B cells in spleen and TCR-β þ T cells in the thymus had no significant difference among each age group (Supplementary Fig. S2B-D). Our findings indicate that T cell development is closely associated with age in newborn mice.
Therefore, our findings suggest that the underdevelopment of T cells is correlated to the severity of EV-A71 infection in neonatal mice.
$$CD3 þ total T cells, CD3 þ CD4 þ helper T cells, CD3 þ CD8 þ T cells, CD11b þ$$
$$, the percentages of CD3 þ T cells, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T$$
## The dysfunction of T-cell immune response contributes to the pathogenesis of EV-A71 infection in newborn mice
To examine the roles of T cells on the severity of EV-A71 infection in vivo, we examined the status of T cells in 3-and 14-day-old mice infected with EV-A71 at 1, 3, 5, and 7 d.p.i. We chose 7 d.p.i. as the last time-point because the 3-day-old mice began to succumb to EV-A71 infection at that time (Fig. 1B). Our data showed that EV-A71 infection dramatically decreased the percentages of total T cells, CD4 þ T cells, and CD8 þ T cells in the spleen of the 3-day-old mice (Fig. 2A andB). On the other hand, for the 14-day-old mice, the percentages of total T cells, CD4 þ T cells, and CD8 þ T cells showed no significant difference between EV-A71-infected and non-infected groups (Fig. 2C andD). These results indicate that EV-A71 infection impaired the immune response of T cells in newborn (3-day-old) mice, which probably contribute to the disease severity of EV-A71 infection.
To further assess the roles of T cells in the pathogenesis of EV-A71 infection, we used BALB/c-nu À/À mice that lack thymus for T cell development. As expected, BALB/c-nu À/À mice had almost no T cells (Fig. 3A-D), while the B cells, CD11b þ macrophages, and CD11c þ DCs in these immunocompromised mice were not affected (Supplementary Fig. S3A-D). 10-day-old BALB/c-nu À/À mice and WT BALB/c mice were i.p. infected with 1 Â 10 7 PFU/ml EV-A71. Our data showed that BALB/cnu À/À mice began to succumb to EV-A71 infection from 8 d.p.i. and reached 100% mortality at 10 d.p.i., while the body weights of these mice began to decrease at 7 d.p.i. (Fig. 3E-G). In contrast, 80% of WT BALB/c mice survived the lethal viral challenge, while their body weights continued to increase and appeared normal compared with those in noninfected group. Moreover, both BALB/c-nu À/À and WT BALB/c mice began to have clinical symptoms at 4 d.p.i. (Fig. 3F), while BALB/c-nu À/À mice died quickly and WT BALB/c mice began to recover from 10 d.p.i. (Fig. 3G). Therefore, the loss of T cells made neonatal mice vulnerable to EV-A71 infection.
Together, our findings show that EV-A71 infection can impair T cell immune response, and the dysfunction of T cell immune response is associated with the pathogenesis of EV-A71 infection in newborn mice.
## Cohort study revealed that the T-cell immune response is associated with the severity of HFMD cases caused by EV-A71 infection
It would be intriguing to find out whether the immune response of T cells in severe HFMD cases of pediatric patients is the same or similar with our observations in the newborn mouse model of EV-A71 infection. For this purpose, we enrolled a cohort of 145 pediatric patients with laboratory-confirmed EV-A71 infection at Guangzhou Women and Children's Medical Center. Among these patients, 61 children were diagnosed as severe symptoms, and 84 children were diagnosed as mild symptoms according to the Diagnosis and Treatment Protocol for HFMD published by the National Health Commission of China (2018 Edition) (http:// www.nhc.gov.cn/cms-search/xxgk/getManuscriptXxgk.htm?id¼5db 274d8697a41ea84e88eedd8bf8f63). The clinical data of the 145 patients were shown in Supplementary Table S1. There was no significant difference in age between the severe and mild groups (Fig. 4A), and the gender of the two groups were matched (Fig. 4B). Our data showed that the numbers of T cells, including
cells, as well as NK cells in severe pediatric patients were significantly lower than those in the mild cases (Fig. 4C-F). On the other hand, the number of CD45 þ CD19 þ B cells between the severe and mild groups showed no significant difference (Fig. 4G). These results show that the compromised immune response of T cells is associated with the severity of EV-A71-caused HFMDs in children, consistent with the experimental data obtained in the newborn mouse model of EV-A71 infection.
Fig. 1. T cell development in newborn mice was associated with the severity of enterovirus 71 (EV71) infection. The 1-, 3-, 7-or 14-day-old ICR mice were infected with enterovirus A71 (EV-A71) at 1 Â 10 7 PFU (A), the survival rates for mice in indicated groups during the 21-day period are shown by curves (B), (n ¼ 10 for each group). Representative contour plots showed the percentages of
in the spleens of 1-, 3-, 7-or 14-day-old ICR mice (left). The curve diagram described the trend of T cells in spleens of 1-, 3-, 7-or 14-day-old ICR mice (right), (n ¼ 3). Data are representative of three independent experiments. Graph shows mean AE SEM.
cells in spleen of 3-day-old mice (A) or 14-day-old mice (C) at 5 dpi after EV-A71 infection. The curve diagram described the trend of
$$CD45 þ CD3 þ total T cells, CD45 þ CD3 þ CD4 þ T cells, CD45 þ CD3 þ CD8 þ T$$
$$CD3 þ T (C), CD3 þ CD4 þ T (D) and CD3 þ CD8 þ T cells (E)$$
$$CD3 þ T cells, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T$$
$$CD3 þ T, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T$$
$$CD3 þ T, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T$$
## AGS-A treatment protected from lethal EV-A71 infection in a T-cell-dependent manner
After finding that the compromised T-cell immune response contributes to the pathogenesis of EV-A71 infection both in pediatric patients and animal models, the intriguing question is that whether enhancing T-cell immune response has any therapeutic potential to treat EV-A71 infection. AGS-A is a saponin extracted from Astragalus membranaceus. To assess the therapeutic potential of AGS-A on EV-A71 infection, we treated 5-day-old ICR mice with AGS-A via i.p. administration at a dose of 10 mg/kg 2 h after EV-A71 challenge, followed by treatment twice a day for 6 days. The mice in the vehicle group (control) began to succumb to EV-A71 infection from 8 d.p.i., and reached 50% mortality at 10 d.p.i. In contrast, 80% of the mice in the AGS-A therapeutic group survived the viral challenge (Fig. 5A). Consistently, we observed significant differences of body weights and clinical scores between the therapeutic and control groups (Supplementary Fig. S4A-C). In addition, the viral RNA accumulations of EV-A71 in muscle, spleen, and brain tissues were significantly reduced in the AGS-A therapeutic group compared with those in the control group at both 3 and 5 d.p.i. (Supplementary Fig. S4D). Moreover, we examined the therapeutic effects of AGS-A in EV-A71-infected mice at 12 h after viral challenge and obtained the similar
cells in spleen of BALB/c-nu À/À mice and WT BALB/c mice (B-D). (n ¼ 3 for each group). 10-day-old BALB/c-nu À/À mice (n ¼ 10) and WT BALB/c mice (n ¼ 10) were i.p. infected with 1 Â 10 7 PFU EV-A71, n ¼ 3 for non-infected (control) group. The survival distribution for BALB/c-nu À/À and WT mice with lethal EV-A71 challenge during the 21-day period was shown by curves (E). Curves described the clinical scores of BALB/c-nu À/À and WT mice with lethal EV71 challenge (F). The average percentages of body weight changes were monitored between BALB/c-nu À/À and WT mice during the 21-day period is shown by curves (G). Data are representative of three independent experiments. Graph shows mean AE SEM. P values were calculated via ANOVA. ****P < 0.0001. results (Fig. 5B and Supplementary Fig. S4E-G). Therefore, these data show that AGS-A treatment has potent in vivo therapeutic efficacy against EV-A71 infection in newborn mice.
We sought to determine whether the action of AGS-A to treat EV-A71 infection in mice is associated with the immune response of T cells. Our data showed that although AGS-A treatment did not enhance the percentages of CD3 þ total T cells and CD4 þ T cells in spleens of non-infected 5-day-old ICR mice, it effectively restored the proportion of T cells to the same level as that in mice not infected with EV-A71 (Fig. 5C andD).
Activation of T cells requires the interaction of T cells with antigenpresenting cells via bi-molecular signals, including major histocompatibility complex (MHC) and CD86 (Xiaodong Feng and Administrative Sciences California Northstate University College of Pharmacy Rancho Cordova). Therefore, we assessed the effects of AGS-A on the expression levels of MHC-II and CD86 on CD11b þ macrophages. And our data showed that AGS-A treatment significantly enhanced the expression of both MHC-II and CD86 (Fig. 5E-G). These results indicate that in the context of viral infection, AGS-A treatment can not only restore the immune response of T cells, but also activate T cells via stimulating bi-molecular signal expression on macrophages. Interestingly, AGS-A treatment failed to inhibit EV-A71 replication in rhabdomyosarcoma cells (RD cells) (Supplementary Fig. S5A), ruling out the possibility that AGS-A directly targets the life cycle of EV-A71. In addition, AGS-A showed minimal cytotoxicity (Supplementary Fig. S5B), further supporting the safety of AGS-A treatment.
Furthermore, we assessed whether the therapeutic effect of AGS-A is dependent on the presence of T cells. For this purpose, we also took the advantage of BALB/c-nu À/À mice that lack T cells, and examined the effects of AGS-A treatment on BALB/c-nu À/À mice challenged with EV-A71. Our data showed that although AGS-A treatment expectedly protected from EV-A71 infection in 5-day-old WT BALB/c mice (Fig. 6A-C), this saponin compound failed to show any therapeutic effects in the aspects of survival rate, body weights, and clinical scores, on the lethal EV-A71 challenge when being compared with non-treated, EV-A71-infected group in the 5-day-old BALB/c-nu À/À mice (Fig. 6D-F). These results show that the therapeutic effect of AGS-A is dependent on T cells.
In conclusion, our findings demonstrate that the therapeutic effect of AGS-A on the pathogenesis of EV-A71 infection in newborn mice is T-cell-dependent, and AGS-A treatment has promising potential to be further developed as an immunotherapy for EV-A71 infection and HFMD in children.
$$of CD3 þ T cells, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T cells in spleen of WT BALB/c mice and T-cell-deficient mice (BALB/c-nu À/À mice) (A). The bar charts illustrate the percentage of CD3 þ T cells, CD3 þ CD4 þ T cells and CD3 þ CD8 þ T$$
$$C-G The counts of CD3 þ CD45 þ lymphocytes (C), CD3 þ CD4 þ CD45 þ lymphocytes (D), CD3 þ CD8 þ CD45 þ lymphocytes (E), CD16 þ CD56 þ natural killer (NK) cells (F) and CD19 þ B cells (G)$$
## DISCUSSION
HFMD caused by EV-A71 infection could be developed to severe symptoms and even lethality in young children under 5 years old, while the pathogenesis of severe HFMD is less understood. In the current study, we found that the underdevelopment of T-cell immune response was closely associated with the severity of EV-A71 infection in neonatal mice. Moreover, EV-A71 infection caused a dramatic impairment of T cell immune response in newborn mice, and the dysfunction of T cell immune response contributes to the pathogenesis of EV-A71 infection, as the loss of T cells made neonatal mice highly vulnerable to EV-A71 infection. Importantly, we further found that AGS-A could effectively rescue the EV-A71-caused impairment of T cell immune response, and showed potent in vivo therapeutic efficacy against EV-A71 infection in a T-celldependent manner in mouse model, implying that AGS-A treatment has promising potential to be developed as an immunotherapy for EV-A71 infection and HFMD.
Our study showed that the severe or lethal outcomes of EV-A71 infection only occurred in neonatal mice aged less than 2 weeks (Fig. 1B). Interestingly, we also uncovered that EV-A71 infection dramatically impaired T cell immune response in 3-day-old mice with immature T cell development, but not in 14-day-old mice with relatively well-developed T cells (Fig. 1C-E and Fig. 2). It is known that the adaptive immune system begins to mature at that time (Adkins et al., 2004), while T cells play important roles in the process of cellular immunity (Kumar et al., 2018;Lee et al., 2020;Mantovani et al., 2011). Moreover, our cohort study involving 145 pediatric patients who were laboratory-confirmed for EV-A71 infection and showed different HFMD symptoms ranging from mild to severe outcomes, further supported the notion that the disease severity of EV-A71 pediatric infection is closely associated with the dysfunction of T cell immune response in children, which is well in line with the in vivo data obtained using the infection murine model (Fig. 4). These findings are consistent with previous studies that cellular rather than humoral immunity is associated with the Fig. 6. AGS-A treatment protected from lethal EV-A71 infection only in mice with intact T-cell immunity. 5-day-old WT BABL/c (A) and BALB/c-nu À/À (D) mice were treated with 10 mg/kg AGS-A 12 h after EV-A71 challenge, followed by AGS-A treatment twice a day for 6 days. The survival rates of mice in the indicated groups during the 14-day period shown by curves. Body weights and clinical scores for the mice in (A) and (D) are plotted in (B, C) and (E, F), respectively. Data are representative of three independent experiments. Graph shows mean AE SEM (n ¼ 10). P values were calculated via ANOVA. *P < 0.05, **P < 0.01, ns, no significance. clinical outcomes of EV-A71 infection (Chang et al., 2006(Chang et al., , 2008;;Lee et al., 2020;Shen et al., 2013;Yang et al., 2001). Therefore, the underdevelopment of T cells together with the EV-A71-caused impairment of T cell immune response should be the pivotal contributing factors for the pathogenesis of EV71 infection and HFMD.
AGS-A is a saponin extracted from Astragalus membranaceus (Lee et al., 2017) that has been used in traditional Chinese medicine for thousands of years to improve immune system. Modern medicinal researches on Astragalus have shown that it has great potential in immunomodulatory properties and antitumor activities (Li et al., 2014). Moreover, AGS-A was reported to inhibit tumor progression by downregulating the percentage of regulatory T cells and upregulating the percentage of cytotoxic T lymphocytes (Li et al., 2017). In addition, a recent study reported that AGS-IV inhibited EV-A71 replication in normal human gastric epithelial (GES-1) cells and RD cells via modulating PI3K-AKT signaling (Hao et al., 2024); however, we did not observe any direct anti-EV-A71 effect either in cells or in T-cell-deficient BALB/c-nu À/À mice infected with EV-A71, indicating that at least in the context of our experiments, the AGS-A conferred therapeutic effect on EV-A71 infection was associated with the effect of AGS-A on T cell response. In this study, our data showed that AGS-A treatment showed potent therapeutic effect to reduce the lethality and clinical symptoms of EV-A71 infection in a T-cell-dependent manner. Interestingly, such a treatment did not boost T cell immune response in the spleens of non-infected mice. The remaining, intriguing questions are how EV-A71 infection specifically impairs T cell immune response in neonatal mice but not in older ones with better developed T cells, and how AGS-A specifically restores such an impairment but not generally boosts T cell immunity. A possible explanation is that EV-A71 infection causes cell death in less matured T cells and/or induce certain cellular or signaling processes to inhibit T cells, while AGS-A treatment can reverse such effects. Answering these questions should further elucidate the pathogenesis of HFMD and reveal the mechanism of action of AGS-A in more details. Nevertheless, the rescuing but not general boosting effect of AGS-A on T cells makes this treatment less likely to induce excessive cellular immune response in recipients.
## CONCLUSIONS
Our findings uncover the interaction between EV-A71 infection and T cell immunity, provide novel insights onto the physiological impacts of T cells on the pathogenesis of EV71 infection and HFMD, and find a promising immunotherapeutic strategy to treat this important pediatric infectious disease. Moreover, AGS-A, its derivatives as well as their different formulations should be further developed and assessed for their abilities as the treatment for EV-A71 infection and HFMD.
## MATERIALS AND METHODS
## Characterization of lymphocyte subsets in peripheral blood cells of children with EV-A71 infection
The EV-A71 group consisted of thirty-two cases of children with EV-A71 infection, who were admitted to Guangzhou Women and Children's Medical Center (Guangzhou, China) between November 2014 and July 2015. Pharyngeal swabs took from the children with the onset of fever within three days, rash in one or more parts of mouth, buttocks, hands or feet were examined for the presence of common pathogens by real-time reverse-transcription quantitative polymerase chain reaction (RT-qPCR) according to the protocols of the detection kits manufacturer (Daan Gene Co., Ltd of Sun Yat-Sen University, Shenzhen, China). All patients were confirmed with the diagnostic criteria of guideline for the diagnosis and treatment of HFMD (2018 Edition). Other common pathogens, including CV-A16, influenza virus, respiratory syncytial virus, adeno virus (ADV), human bocavirus, human metapneumovirus, parainfluenza virus (PIV), rhino virus (RHV), mycoplasma pneumoniae (MP), chlamydia pneumoniae, cytomegalovirus, Epsteine-Barr virus, and herpes simplex virus, were found negative in both groups. Informed consent was obtained from the legal parents of the children, and all procedures involved in this study were approved by the Ethics Committee of Guangzhou Women and Children's Medical Center.
## Virus
The EV-A71 strain H (VR-1432) was obtained from ATCC, the virus was amplified and titers were determined in human rhabdomyosarcoma cells (RD cells) as previously described (Zhang et al., 2024). The virus was concentrated by ultrafiltration using Amicon Ultra-15 filters (Millipore) and quantified by plaque assay. Experiments including viral infections were conducted in Biosafety Level 2 (BSL-2) labs at Wuhan Institute of Virology.
## Mice
Pathogen-free 18-day pregnant Institute of Cancer Research (ICR) mice, BALB/c mice, BALB/c-nu À/À mice were obtained from the animal housing facility of the Chinese Academy of Sciences (Changsha, China). All experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee of Wuhan Institute of Virology.
Mice were i.p. injected with 10 7 PFU of EV-A71 in 50 μL of DMEM.
Control animals received the same volume of DMEM via the same route. The time before infection was referred to as day 0. Infection with EV-A71 continued for day 10 to day 20 day, and mice presented clinical symptoms at 5 d.p.i. EV-A71-infected mice were sacrificed every other day up to day 9. A score was used to evaluate the clinical symptoms as previously described: 0, healthy; 1, hunchbacked and slow movement; 2, weakness in one limb; 3, paralysis in one limb; 4, paralysis in both limbs; and 5, death (He et al., 2018;Zhang et al., 2013).
The spleen of each mouse was collected and total RNAs from spleen were extracted with TRIzol (10296010) from Invitrogen. The viral RNA accumulation was determined via RT-qPCR with specific primers. The RT-qPCR was performed in a reaction mixture of Hieff® qPCR SYBR Green Master Mix (Low Rox Plus) (YEASEN), template, primers, reverse transcriptase (YEASEN) and RNase-free water. Primer sequences are available upon request and their sequences were listed in Supplementary Table S2.
## Flow cytometry
Single-cell suspensions from the spleen and thymus were stained with different combinations of the following mAbs conjugated with FITC, PE, APC, PE-Cy7, and allophycocyanin as our previously described (Wang et al., 2017).
For cell surface maker staining, single-cell suspensions were incubated with the following Abs: FITC anti-CD3 (17A2; BioLegend), PE-Cy7 anti-CD4 (GK1.5; BioLegend), FITC anti-TCR-β (H57-597), PE anti-CD11c (N418; BioLegend), APC anti-CD11b (M1/70; BioLegend), PE anti-CD19 (6D5; BioLegend), PE-Cy7 anti-CD86 (GL-1; BioLegend) and FITC anti-IA/IE (M5/114.15.2) in 0.2% BSA (BioSharp) with PBS (pH 7.4) for 25 min at 4 C. Cells were washed with PBS (400 g, 5 min, 4 C) and fixed with 1% paraformaldehyde. Cells were determined using a FACS Calibur (BD Biosciences) system, and the data were analyzed using CellQuest Pro software (BD Biosciences).
## AGS-A treatment
AGS-A (83207-58-3) were commercially purchased from Selleck. 5-day-old ICR mice were treated with AGS-A at the dose of 10 mg/kg within vehicle 2 h or 12 h after challenge with EV-A71, followed by treatment twice a day or once a day for 6 days by i.p. administration. 10day-old BALB/c-nu À/À mice were treated with Astragal side at the dose of
## References
1. Adkins, Leclerc, Marshall-Clarke (2004) "Neonatal adaptive immunity comes of age" *Nat. Rev. Immunol*
2. Auyeung, Han, Ko (2016) "Astragalus membranaceus: a review of its protection against inflammation and gastrointestinal cancers" *Am. J. Chin. Med*
3. Carsetti, Quintarelli, Quinti et al. (2020) "The immune system of children: the key to understanding SARS-CoV-2 susceptibility?" *Lancet Child Adolesc. Health*
4. Chan, Parashar, Lye et al. (2000) "Deaths of children during an outbreak of hand, foot, and mouth disease in sarawak, Malaysia: clinical and pathological characteristics of the disease. For the outbreak study group" *Clin. Infect. Dis*
5. Chang, Hsiung, Lu et al. (2006) "Status of cellular rather than humoral immunity is correlated with clinical outcome of enterovirus 71" *Pediatr. Res*
6. Chang, Chang, Chen et al. (2008) "HLA-A33 is associated with susceptibility to enterovirus 71 infection" *Pediatrics*
7. Deng, Yang, Wan et al. (2011) "Irregular poliovirus vaccination correlates to pulmonary edema of hand, foot, and mouth disease" *Clin. Vaccine Immunol*
8. Fioravanti, Hao, Zhang et al. (2012) "Metabolomics combined with network pharmacology reveals a role for astragaloside IV in inhibiting enterovirus 71 replication via PI3K-AKT signaling" *J. Transl. Med*
9. He, Ren, Xia et al. (2018) "Fibronectin facilitates enterovirus 71 infection by mediating viral entry" *J. Virol*
10. Ho, Chen, Hsu et al. (1999) "An epidemic of enterovirus 71 infection in taiwan. Taiwan enterovirus epidemic working group" *N. Engl. J. Med*
11. Huang, Liu, Chang et al. (1999) "Neurologic complications in children with enterovirus 71 infection" *N. Engl. J. Med*
12. Kumar, Connors, Farber (2018) "Human T cell development, localization, and function throughout life" *Immunity (Camb., Mass.)*
13. Lee, Tsai, Lin et al. (2017) "Astragaloside II promotes intestinal epithelial repair by enhancing L-arginine uptake and activating the mTOR pathway" *Sci. Rep*
14. Lee, Cho, Choi (2020) "Bystander CD4(þ) T cells: crossroads between innate and adaptive immunity" *Exp. Mol. Med*
15. Li, Qu, Dong et al. (2014) "A review of recent research progress on the astragalus genus" *Molecules (Basel)*
16. Li, Hou, Xu et al. (2017) "Research review on the pharmacological effects of astragaloside IV" *Fundam. Clin. Pharmacol*
17. Mantovani, Cassatella, Costantini et al. (2011) "Neutrophils in the activation and regulation of innate and adaptive immunity" *Nat. Rev. Immunol*
18. Olin, Henckel, Chen et al. (2018) "Stereotypic immune system development in newborn children" *Cell*
19. Pallansch (2007) "Enteroviruses: Polioviruses, Coxsackieviruses, Echoviruses, and Newer Enteroviruses"
20. Shen, Tsai, Wang et al. (2013) "Enterovirus 71 infection increases expression of interferon-gamma-inducible protein 10 which protects mice by reducing viral burden in multiple tissues" *J. Gen. Virol*
21. Simon, Hollander, Mcmichael (2015) "Evolution of the immune system in humans from infancy to old age" *Proc. Biol. Sci*
22. Solomon, Lewthwaite, Perera et al. (2010) "Virology, epidemiology, pathogenesis, and control of enterovirus 71" *Lancet Infect. Dis*
23. Tan, Huang, Zhu et al. (2008) "The persistent circulation of enterovirus 71 in People's Republic of China: causing emerging nationwide epidemics since" *PLoS One*
24. Wang, Zhang, Qi et al. (2017) "Myeloid-derived suppressor cells inhibit T follicular helper cell immune response in Japanese encephalitis virus infection" *J. Immunol*
25. Yang, Yang, Li et al. (2001) "Altered cellular but not humoral reactions in children with complicated enterovirus 71 infections in Taiwan" *J. Infect. Dis*
26. Zhang, Li, Li et al. (2024) "Identification of fangchinoline as a broadspectrum enterovirus inhibitor through reporter virus based high-content screening" *Virol. Sin*
27. Zhang, Song, Qin et al. (2013) "Rupintrivir is a promising candidate for treating severe cases of enterovirus-71 infection: evaluation of antiviral efficacy in a murine infection model" *Antivir. Res*
28. Zhu, Yang, Luo et al. (2015) "TLR3 signaling in macrophages is indispensable for the protective immunity of invariant natural killer T cells against enterovirus 71 infection" *PLoS Pathog* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12357456&blobtype=pdf | # Antimicrobial Resistance & Infection Control †
Jasmin Jasuja, Eva-Maria Klupp, Martin Aepfelbacher, Kurt Knut, Kampe, Michael Nentwich, Stefan Kluge, Johannes Karl-Mark Knobloch
## Abstract
Background A series of transmission of Pseudomonas aeruginosa ST111 bla VIM-2 , previously undetected by standard surveillance, was discovered in a tertiary care hospital in Northern Germany through molecular genetic monitoring. Hence, environmental sampling was initiated to find the source of infection.Methods First, routine epidemiological data ruled out patient-to-patient transmission and two initial diagnoses were assessed as externally acquired. After the discovery of the highly related cluster by whole genome sequencing, a more detailed epidemiological analysis was carried out, including previous hospitalizations. An environmental investigation was initiated due to a possible connection of transmissions with an intensive care unit.
ResultsBetween 2018 and 2023 16 clinical isolates of Pseudomonas aeruginosa ST111 bla VIM-2 were identified of which 12 isolates belonged to ST111 carrying an In59-like integron. Routine whole-genome sequencing of carbapenem resistant P. aeruginosa identified a highly related cluster (maximum of three allelic differences) of highrisk ST111 isolates in ICU patients over five years, confirming sink-to-patient transmission associated to sink drains in two ICU rooms. In initial routine epidemiological categorization of these highly related isolates four isolates were categorized as possible nosocomial acquisition without direct epidemiological link to other patients, whereas two isolates were categorized as 'externally acquired' .
ConclusionsThis finding highlights the ability of high-risk clone ST111 to persist in hospital environments and emphasizes the importance of integrating molecular surveillance with routine epidemiology to uncover hidden transmissions. In this case, the frequent detection of the ST111 high-risk clone led to targeted environmental sampling, uncovering a prolonged outbreak that had gone unnoticed by conventional surveillance. The clone was eliminated from the ward during a reconstruction project.
## Background
Pseudomonas aeruginosa is a leading cause of healthcareassociated infections (HAIs) and is of particular concern in critical care settings. As an important cause of healthcare-acquired infections in intensive care units (ICUs), P. aeruginosa demonstrates a remarkable ability to rapidly develop resistance to various antimicrobials, driven by chromosomal mutations and the acquisition of resistance genes encoded on mobile genetic elements such as plasmids or transposons [1][2][3][4][5][6]. The Centres for Disease Control and Prevention (CDC) and the European Centre for Disease Prevention and Control (ECDC) define multidrug resistance (MDR) as co-resistance to at least one agent in three out of eight antimicrobial categories [7].
MDR P. aeruginosa, including P. aeruginosa clones harbouring plasmid-encoded carbapenemases, are prevalent globally and typically observed in hospital environments. These clones are selected within hospitals due to antibiotic selection pressure and favourable growth conditions [8]. Thereby, in P. aeruginosa isolates spontaneously developing phenotypic resistance by mutations during antibiotic therapy must be distinguished from isolates carrying specific resistance genes like carbapenemases [9]. The emergence of widely disseminated carbapenemase producing P. aeruginosa (CP-PA) strains, designated as high-risk clones, has made whole-genome sequencing (WGS) an essential tool for understanding their epidemiology. Several high-risk clones are identified by their sequence types (STs) and have emerged on an international scale. Most hospital outbreaks in Europe are associated with multi locus sequence types (MLST) ST111, ST175, ST233, ST235, ST277, ST357, ST654 and ST773 [2,8,10]. CP-PA strains are frequently detected in wastewater and rinse water, where they form multibacterial biofilms in plumbing systems [11]. Within these protective and impenetrable biofilms, Enterobacterales and P. aeruginosa strains exchange carbapenemase-encoding plasmids [11]. During outbreaks, clones associated with the aforementioned STs are often found causing infections in vulnerable patient populations within ICUs. In a systematic review by Büchler et al. 100 out of 126 included studies screened the environment in outbreak situations and in all but three contaminated environment was identified as the primary source [12]. Hence, environmental screening was identified as an important outbreak control measure [12].
Here, we report a hidden outbreak with metallo-βlactamase bla VIM-2 -producing and qacE∆1-harbouring P. aeruginosa ST111 in a German tertiary care hospital, uncovered by perennial routine core genome multilocus sequencing typing (cgMLST), involving sink-to-patient transmission. To identify the outbreak source, an epidemiological investigation was initiated. Environmental samples were collected from the relevant ICUs, and the identified P. aeruginosa strains were analysed using cgMLST.
## Results
## Outbreak description
Between July 2018 and August 2023, a total of 131 nonrepetitive phenotypically carbapenem-resistant P. aeruginosa (CR-PA) isolates cultured from clinical specimens of an adult hospital were analysed by cgMLST. Among these, carriage of carbapenemase genes were identified in 29 isolates (22.1%), thereof 16 isolates harbouring the bla VIM-2 gene (55.2%), four isolates harbouring bla VIM-1 (13.8%), and two isolates harbouring bla VIM-5 (6.9%). The predominant ST was ST111 (n = 14), followed by ST273 (n = 5). cgMLST identified a cluster (≤ 11 different alleles) of eleven ST111 CP-PA bla VIM-2 isolates, all also harbouring qacE∆1, a gene encoding an efflux pump for quaternary ammonium compounds (QAC). Six out of eleven bla VIM-2 P. aeruginosa ST111 isolates were identified as highly related with a maximum allelic distance of three alleles (Fig. 1). In initial routine epidemiological categorization of these closely related isolates four isolates were categorized as possible nosocomial acquisition without direct epidemiological link to other patients, whereas two isolates were categorized as 'externally acquired' as patients were re-admitted or hospitalized after external stays. Due to the confirmed highly close relationship of the individual isolates but first recovery months and even years apart a more detailed epidemiological re-investigation was initiated, focusing on these isolates.
## Epidemiology
The dates of first detection of the six highly related isolates were ranged between December 2019 and August 2023. Initial microbiological detections were made from relevant clinical specimens including the respiratory tract, skin, wounds and blood cultures (Fig. 2). An initial screening at admission was only conducted in two out of six patients, proving colonization with CP-PA. Hence, most of these cases were classified as nosocomial according to the German hospital infections surveillance system (KISS). The first externally acquired case was first hospitalized early 2019 and had an infection 298 days after the initial admission with intermittent hospitalization. The second externally acquired case was admitted early 2023 and infection with the outbreak strain was detected 130 days later with intermittent hospital stays. The initial detection in nosocomial-acquired cases occurred 8, 11, 29 and 85 days post admission (median: 20 days), respectively, and were all found in clinical specimen. Retrospective analysis showed that all six patients had infections caused by P. aeruginosa ST111 bla VIM-2 , sourced from various clinical conditions, such as surgical site infections, acute respiratory distress syndrome (ARDS), pneumogenic sepsis, pneumonia and anastomosis-and respiratory insufficiency. Of these, four patients died in the further course of their hospital stay. Whether the ST111 infection was the cause of death cannot be ruled out with certainty.
We subsequently investigated the wards where the initial detections occurred, and it was found that four of the six detections were from ward A in the years 2022 and 2023, with the remaining two from wards B in 2019 and D in 2023, respectively.
Since these findings did not immediately point to a clear source for possible transmissions, previous hospital stays were reviewed, particularly as the affected patients had been hospitalised for extended periods. The review indicated that the isolates detected on wards B and D were also associated with prior stays on ward A (Fig. 2). Interestingly, all affected patients could be associated with a stay in two adjacent patient rooms on ward A. As a result, environmental sampling was initiated on ward A.
## Environmental sampling and molecular epidemiology
The extensive environmental examinations (see methods) were conducted twice in November 2023 and comprised 36 environmental samples (15 specimen from sinks of adjacent bathroom across seven rooms, six specimen from wardrobes in the bathrooms, one medical product kept in one of the patient room, two from a toilet in a patient room and twelve specimen from a sluice room) (Supplementary Fig. 1). The wastewater piping of rooms R1 and R2 are connected horizontally to the same wastewater line. Room R3 is separately connected to a wastewater pipeline, which merges only with the line serving R1 and R2 in the basement. During environmental sampling no positive screened patient was occupying the affected rooms and ward, respectively. Eleven environmental CR-PA isolates were identified. Thereof, two isolates were non-carbapenemase producing P. aeruginosa. In total, nine CP-PA isolates were identified with bla VIM-2 (n = 5) being the predominant carbapenemase. MLST by whole genome sequencing uncovered that all environmental bla VIM-2 P. aeruginosa belonged to ST111. P. aeruginosa ST111 bla VIM-2 was found in three out of six sinks in patient rooms, on the surface of a dialysis therapy bag kept near the washbasin in patient's bathroom, and in the sink of a bedpan flushing device in the sluice room. Applying a threshold for high relation of ≤ 3 different alleles, we identified five environmental bla VIM-2 and qacE∆1 positive P. aeruginosa ST111 isolates clustering with clinical isolates of patients hospitalised between 2019 and 2023. The four environmental samples were obtained from three washbasin sinks and from the exterior of a dialysis bag stored above a washbasin. All four environmental P. aeruginosa ST111 bla VIM-2 isolates were found in three patient rooms (R1-R3). Thereof, all patients had an associated stay for rooms R1 and R2, but none for R3. One specimen from the sluice room was positive for ST111 but did not harbour any carbapenemase and cluster distance threshold was > 12.
In order to assess whether plasmid transfer was involved in the transmission events, a detailed analysis of the resistance gene profile was carried out and attempts were made to determine the location of bla VIM-2 . The resistance gene profile was almost identical between the ST111 isolates, including identical resistance-associated mutations and a profile of genes associated with reduced sensitivity to biocides and heavy metals (Supplementary Table, orange section). Analysis of the immediate genomic environment of bla VIM-2 revealed for ten of the ST111 isolates contigs so small (930 to 1539 bases) that it was not possible to evaluate their location. For eight isolates, assembly resulted in larger contigs (15 to 50 kb). The highly related isolates showed homology to chromosomally integrated In59-like class 1 integrons (Supplementary Fig. 2; Supplementary Table green section), which carry the resistance genes aac(6')-29, qacE∆1and sul1 in addition to bla VIM-2 [13]. However, de novo assembly revealed an In59-like integron without bla VIM-2 in some strains, as well as some possible insertions and deletions in this region (Supplementary Fig. 2; Supplementary Table green section). This observation may also be due to the limitations of short read sequencing. As screening for plasmid replicon sequences using Plasmid-Finder resulted in no hits for known plasmids in the database chromosomal integration of bla VIM-2 is suggested.
Due to the positive environmental findings, re-sampling of the environment was initially planned. However, from April to May 2024 the ward was closed for renovation and conversion into a regular ward. During this time, no patients were admitted to these rooms and consequently washbasins were not in use. Microbiological controls after reopening the ward did not yield additional P. aeruginosa ST111 bla VIM-2 isolates in six investigations over a period of twelve months.
## Discussion
In this investigation, we uncovered a hidden prolonged outbreak involving the high-risk clone P. aeruginosa ST111 bla VIM-2 in a hospital in northern Germany, which had not been recognised as an outbreak by routine Fig. 2 Line list of patients with confirmed transmission events of ST111 Pseudomonas aeruginosa bla VIM-2 . A total of six patients were affected (P1-P6). The broad horizontal lines mark the respective stays in a normal ward (grey) of the affected intensive care unit (magenta) or another intensive care unit in the hospital. The vertical red lines indicate the respective first and last detection of a CP-PA in the affected patients. The black cross marks patients, who died in the course of their hospitalization epidemiological surveillance. A purely epidemiological assessment was hindered by its prolonged nature over five years going along with patient transfers, external stays, and re-admissions with colonization or infection possibly being misclassified as externally acquired. Hence, epidemiological assessment alone carries the risk to detect only a fraction of the actual transmissions. The frequent occurrence of P. aeruginosa ST111 bla VIM-2 isolates, defined as highly related cluster in molecular surveillance, prompted a closer examination of its epidemiology. From an epidemiological perspective, patientto-patient transmission was ruled out as there was no direct contact between these patients, all of whom had been hospitalised in ward A but during different periods of time. The investigation of further common procedures like dialyses and bronchoscopy were also excluded, as not every aforementioned patient was dialysed or got a bronchoscopy, especially prior to infection with P. aeruginosa ST111 bla VIM-2 . However, all patients were associated with a stay in two adjacent patient rooms in ward A. As a result of this retrospective epidemiological analysis, repetitive environmental sampling was performed on ward A.
The observation of allelic differences far less than the distance threshold for genomic relatedness of ≤ 11 different in five environmental and clinical isolates of P. aeruginosa ST111 bla VIM-2 proved a nearly identical clone and confirmed sink-to-patient transmission for these cases [14,15]. Indeed, sink colonisation by MDR Gramnegative bacteria is well documented and is considered as a potential transmission route, frequently associated with HAIs and posing a risk even in non-outbreak situations [16,17]. Efforts to replace or eradicate contaminated sinks have been employed, but often fail [18], and it is even recommended that ICU rooms should not be equipped with sinks [19]. Our findings align with this, as the environmental P. aeruginosa ST111 bla VIM-2 strain remained detectable in 2023, despite its initial molecular detection occurring four years earlier. However, only six P. aeruginosa ST111 bla VIM-2 infections were traced back to specific sinks in the ICU, though it is assumed that ST111 persists throughout the year in patients [20]. If the infections with ST111 were indeed hospitalacquired, the initial classification of some cases as 'externally acquired' under the KISS definition would no longer apply. However, the environmental samplings showed ten different P. aeruginosa ST111 bla VIM-2 isolates indicating the overall frequency of sink-to-patient transmission was low but increased with the length of hospital stay as R1 and R2 was often occupied by patients with longterm stays, while patients in R3 did not. However, Rath et al. analysed the toilet-to-patient transmission rate of ST235 bla FIM-1 and ST309 non-carbapenemase producing P. aeruginosa in a bone marrow transplant unit with cgMLST, showing only a low-genetic diversity but only three toilet-to-patient-transmission over six years [21]. Indeed, our reported ward was originally built as an intermediate care unit and was converted into an intensive care unit later on. As a result, the washbasins were not located directly in the patient's room but in a separate bathroom, which might have reduced the frequency of transmission.
In addition, alternative transmission routes must be considered. The positive screening of a dialysis therapy bag, placed above the washbasin suggests a possible transmission route, though not all patients were receiving dialysis. Rather, the dialysis therapy bag may represent a potential sink-to-environment-transmission, which might have been favoured by different pressure conditions leading to sink-to-environment-transmission. Bronchoscopy was excluded as a common transmission route as well as other environmental sources on epidemiological basis. However, it should be noted that environmental investigation was initiated four years after the first detection of P. aeruginosa ST111 bla VIM-2 and hence, potential transmission routes might have been missed meanwhile.
Interestingly, the first P. aeruginosa ST111 bla VIM-2 isolate was already detected in 2019 in a patient in screening and blood culture, which was conducted on admission to ward A and was declared as externally acquired since the patient was transferred from an external hospital where the patient was treated for one month. Before that, the patient had hospital stays in our hospital, but remained on normal wards. A P. aeruginosa was never detected. All other cases in the cluster occurred after the admission of the index case bringing up the discussion that first the outbreak stem was transmitted from patient to sink following hidden sink-to-patient transmission and leading to the hidden outbreak. We found a median duration of 20 days from admission to infection with the outbreak strain, which is lower than reported by Volling et al. [22]. However, Volling et al. reported a non-ST111 P. aeruginosa outbreak strain. In the two externally acquired cases colonisation time a maximum of 298 days and 130 days, respectively, is suggested. Notably, during this period both patients were not continuously hospitalized, making it hard to estimate the colonization-to-infection-period.
In our study, the qacEΔ1 resistance gene was detected in ST111 isolates as part of the In59-type integron, similar to the findings of Rath et al. However, it remains unclear whether this gene contributes to resistance against QAC-based disinfectants or plays a role in outbreak promotion even if a QAC-based disinfectant was used in routine hospital disinfection. ST111 as well as other ST's multi-resistance is welldocumented, but association to resistance gene qacE∆1 has not been reported so far. Hence, further research is needed to elucidate any direct connections between this clone and qacEΔ1.
The observation of P. aeruginosa ST111 bla VIM-2 isolates with confirmed sink-to-patient transmission in an ICU-setting highlights the vulnerability of ICU patients and the conducive environment for antibiotic selection pressure. Previous studies have similarly documented the predominance of ST111 in ICUs and other high-risk wards, including in Germany, where ST111 bla VIM-2 was identified in a cluster of 15 isolates as well as in a Greek hospital [14,21,23]. ST111 strains are frequently associated with carbapenemase production, particularly bla VIM-2 , leading to multidrug resistance [24,25]. By chance, the transmission series might be interrupted by eliminating of the clone from the ward during a reconstruction project.
## Limitations
This study has several limitations. First, environmental sampling was conducted four years after the first detection of the high-risk clone in clinical isolates. As a result, earlier environmental isolates and sources may have been missed, limiting our ability to accurately determine the duration of the high-risk clone's presence in the hospital environment in-and outside the sink. To address this, we propose regular environmental sampling to allow for the timely identification of high-risk clones.
Second, while all patients included in the study were infected with the high-risk clone, only two out of six patients were initially screened for colonisation. Therefore, it is unclear whether transmission of the high-risk clone frequently leads to direct infection or if infection regularly resulted from previously undetected colonisation. Hence, we suggested that long-term patients should be screened during their ICU stay. However, further investigation is needed to determine the required time from colonisation to infection, as well as the trigger factors transforming a colonisation into an infection.
Third, during de novo assembly, bla VIM-2 was located in a chromosomally integrated In59-like integron in some strains. In strains without detectable integration into In59, it is unclear whether this is an artefact of the assembly or whether genetic events have actually taken place. However, the three resistance genes bla VIM-2 , qacEΔ1 and sul1 were detected in all ST111 strains. Current literature as well as our detailed analysis of the resistance gene profile does not provide sufficient data directly linking qacEΔ1 to ST111 P. aeruginosa and the role in hospital transmission and infection. Further research is required to elucidate these questions.
Fourthly, the outbreak took place during the pandemic, which led to a change in the usual patient clientele.
## Conclusion
Our report highlights the importance of molecular surveillance, which is more sensitive in detecting high-risk clones compared to conventional epidemiological assessment, which is often hampered by clinical processes such as internal and external transfers and prolonged hospital stays. Epidemiological surveillance should account for several years of data, particularly in cases of high genomic relatedness, to detect silent transmissions at an early stage. Colonisations and infections initially classified as externally acquired may, upon retrospective analysis, prove to be hospital-acquired, leading to the misclassification of silent transmissions over several years, as observed in our case. In this context, it is particularly important that medical institutions that conduct such detailed epidemiological analyses are not blamed for the discovery of transmission events that would have gone unnoticed in institutions without targeted molecular and epidemiological analyses.
## Methods
The investigation was conducted from July 2018 to August 2023 in a tertiary care hospital in northern Germany, encompassing all medical specialities excluding paediatrics. During this period, all P. aeruginosa clinical isolates of adult wards recovered from routine diagnostic except screening material were collected. Isolates exhibiting phenotypic resistance to piperacillin, ceftazidime, ciprofloxacin, meropenem and imipenem based on the official multidrug-resistant (MDR) definition by the German healthcare authorities underwent further molecular sequencing.
For DNA extraction, the QIASymphony SP Instrument and QIAsymphony DSP Virus/Pathogen Mini Kit (Qiagen, Venlo, The Netherlands) were used as recently published [26]. Sequencing was performed using the NextSeq500 platform (Illumina, San Diego, USA), in combination with the NEBNext Ultra DNA Library Prep Kit and NEBNext Multiplex Oligos for Illumina (NEB, Ipswich, USA). Assembly of genomes and cgMLST was conducted with the Seqsphere software package (Ridom, Münster, Germany Version 11.0.0; integrated Velvet assembler Version 1.1.04), applying a clustering threshold of ≤ 11 allelic differences, based on literature review [14,21]. A total of 4478 core genes were used for cgMLST. Detailed information about the assembled genomes is given in the Supplementary Table (blue section). The assembled genomes of the six patients, the non-related patients and environmental specimen have been deposited to the public database (NCBI GeneBank database) under the BioProject no. PRNJA1288733. Resistance gene analysis was conducted using the AMRFinder (Version 1.3.1) integrated in the Seqsphere package (all identified genes or mutations displayed in the Supplementary Table, orange section). Additionally, screening for plasmid replicon sequences was performed using Plasmid-Finder 2.1 [27]. Large contigs containing bla VIM-2 were analysed for homologies to known sequences using nucleotide BLAST ( h t t p s : / / b l a s t . n c b i . n l m . n i h . g o v / B l a s t . c g i). After identifying an In59 integron (accession number AF263519.1) as closely related, an alignment of all contigs with a length of more than 10,000 base pairs containing bla VIM-2 and/or qacE∆1 in comparison to the In59 integron and the bla VIM-2 gene alone (Supplementary Fig. 2) was generated using MAUVE (version 20150226) and the progressiveMauve alginment [28].
Following epidemiological indications as aforementioned, we conducted an environmental sampling in the ICU ward with association to the ST111 cluster. The environmental sampling was initiated in November 2023 with a focus on sinks and rinse water. Sinks in six bathrooms and surrounding of the washbasin across seven rooms and the unclean work space were sampled. A total of 36 specimen were collected in two rounds of environmental sampling.
For sampling, swabs and moistened sponge were utilised for siphons and environmental surroundings, respectively, transported to the microbiology laboratory, and processed within 12 h. Swabs were vortexed and subcultured onto blood agar, MacConkey agar, and ESBL agar (bioMérieux). Sponges were packed in sterilised boxes and were incubated in Tryptic Soy Broth (TSB) at 37 °C for 24 h. Afterwards 10 µl were subcultured on ESBL agar and incubated at 36 °C under aerobic conditions for 24-48 h. Colonies consistent with P. aeruginosa subcultured from swabs and sponges were identified via MALDI-TOF MS, and antimicrobial susceptibility testing was performed with VITEK2. If phenotypic MDR P. aeruginosa was detected, aforementioned sequencing protocol was applied to the environmental isolates.
Sink-to-patient transmission was defined as hospitalacquired colonisation (detected in rectal swabs, throat swabs, clinical specimen without initiating antibiotic treatment) or infection (clinical specimen with initiation of antibiotic treatment) with P. aeruginosa ST111 bla VIM-2 with a distance threshold of ≤ 11 alleles to ST111 bla VIM-2 isolates previously recovered from the environment in a room occupied by the respective patient for at least 72 h.
## References
1. Lambert, Suetens, Savey et al. (2011) "Clinical outcomes of health-care-associated infections and antimicrobial resistance in patients admitted to European intensive-care units: a cohort study" *Lancet Infect Dis*
2. De Almeida De Souza, Rossato, Brito (2021) "dos Santos Bet GM, Simionatto S. Carbapenem-resistant Pseudomonas aeruginosa strains: a worrying health problem in intensive care units" *Rev Inst Med Trop Sao Paulo*
3. Jasuja (2025) *Antimicrobial Resistance & Infection Control*
4. Hellen De Almeida De Souza, Rossato, Brito (2021) "Graciela mendonça Dos Santos bet, Simone simionatto. Carbapenem-resistant Pseudomonas aeruginosa strains: a worrying health problem in intensive care units" *Rev Inst Med Trop Sao Paulo*
5. Moradali, Ghods, Rehm (2017) "Pseudomonas aeruginosa lifestyle: a paradigm for adaptation, survival, and persistence" *Front Cell Infect Microbiol*
6. Breidenstein, De La Fuente-Núñez, Hancock (2011) "Pseudomonas aeruginosa: all roads lead to resistance" *Trends Microbiol*
7. Potron, Poirel, Nordmann (2015) "Emerging broad-spectrum resistance in Pseudomonas aeruginosa and Acinetobacter baumannii: mechanisms and epidemiology" *Int J Antimicrob Agents*
8. Magiorakos, Srinivasan, Carey et al. (2012) "Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance" *Clin Microbiol Infect*
9. Kocsis, Szabó (2021) "Diversity and distribution of resistance markers in Pseudomonas aeruginosa international high-risk clones" *Microorganisms*
10. Yoon, Goussard, Touchon et al. (2021) "Mobile carbapenemase genes in Pseudomonas aeruginosa reveal complex transmission dynamics" *Antimicrob Agents Chemother*
11. Oliver (2015) "The increasing threat of Pseudomonas aeruginosa high-risk clones. Drug Resist Updates"
12. Weingarten, Johnson, Conlan et al. (1128) "Genomic Analysis of Hospital Plumbing Reveals Diverse Reservoir of Bacterial Plasmids Conferring Carbapenem Resistance"
13. Büchler, Heudorf, Kirchner et al. (2023) "Outbreak investigations after identifying carbapenem-resistant Pseudomonas aeruginosa: a systematic review" *Antimicrob Resist Infect Control*
14. Poirel, Lambert, Türkoglü et al. (0546) "Characterization of class 1 integrons from Pseudomonas aeruginosa that contain the bla VIM-2 carbapenem-hydrolyzing β-lactamase gene and two novel aminoglycoside resistance gene cassettes" *Antimicrob Agents Chemother*
15. Wendel, Kolbe-Busch, Ressina et al. (2022) "Genomic-based transmission analysis of carbapenem-resistant Pseudomonas aeruginosa at a tertiary care centre in Cologne (Germany) from 2015 to 2020" *JAC Antimicrob Resist*
16. Rath, Klein, Wasner et al. (2024) "Wholegenome sequencing reveals two prolonged simultaneous outbreaks involving Pseudomonas aeruginosa high-risk strains ST111 and ST235 with resistance to quaternary ammonium compounds" *J Hosp Infect*
17. Fucini, Kieffer, Bletz et al. (2023) "Sinks in patient rooms in ICUs are associated with higher rates of hospital-acquired infection: a retrospective analysis of 552 ICUs" *J Hosp Infect*
18. Hopman, Tostmann, Wertheim et al. (2017) "Reduced rate of intensive care unit acquired Gram-negative bacilli after removal of sinks and introduction of 'water-free' patient care" *Antimicrob Resist Infect Control*
19. Pirzadian, Voor, Holt et al. (2023) "Limiting spread of VIM-positive Pseudomonas aeruginosa from colonized sink drains in a tertiary care hospital: a before-andafter study" *PLoS One*
20. Catho, Martischang, Boroli et al. (2021) "Outbreak of Pseudomonas aeruginosa producing VIM carbapenemase in an intensive care unit and its termination by implementation of waterless patient care" *Crit Care*
21. De Greyter, Smet, Deplano et al. (2021) "Sink drains as reservoirs of VIM-2 metallo-β-lactamase-producing Pseudomonas aeruginosa in a Belgian intensive care unit: relation to patients investigated by wholegenome sequencing" *J Hosp Infect*
22. Rath, Bletz, Autenrieth et al. (2024) "Retrospective genome-oriented analysis reveals low transmission rate of multidrug-resistant Pseudomonas aeruginosa from contaminated toilets at a bone marrow transplant unit" *J Hosp Infect*
23. Volling, Mataseje, Graña-Miraglia et al. (2024) "Epidemiology of healthcare-associated Pseudomonas aeruginosa in intensive care units: are sink drains to blame?" *J Hosp Infect*
24. Papagiannitsis, Petinaki, Tzouvelekis et al. (2020) "Unravelling the features of success of VIM-producing ST111 and ST235 Pseudomonas aeruginosa in a Greek hospital" *Microorganisms*
25. Van Der Bij, Pitout (2012) "Metallo-β-lactamase-producing Pseudomonas aeruginosa in the netherlands: the nationwide emergence of a single sequence type" *Clin Microbiol Infect*
26. Witney, Gould, Pope et al. (2014) "Genome sequencing and characterization of an extensively drug-resistant sequence type 111 serotype O12 hospital outbreak strain of Pseudomonas aeruginosa" *Clin Microbiol Infect*
27. Carlsen, Büttner, Christner et al. (2022) "High burden and diversity of carbapenemase-producing enterobacterales observed in wastewater of a tertiary care hospital in Germany" *Int J Hyg Environ Health*
28. Carattoli, Zankari, García-Fernández et al. (2014) "In Silico detection and typing of plasmids using plasmidfinder and plasmid multilocus sequence typing" *Antimicrob Agents Chemother*
29. Darling, Mau, Perna (2010) "ProgressiveMauve: multiple genome alignment with gene gain, loss and rearrangement" *PLoS ONE* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12646009&blobtype=pdf | # Comparative analysis of NSP5/VP2-induced viroplasm-like structures in rotavirus species A to J
Ariana Cosic, Melissa Lee, Kurt Tobler, Claudio Aguilar, Cornel Fraefel, Catherine Eichwald
## Abstract
Rotavirus (RV) is classified into nine species, A-D and F-J, with RV species A (RVA) being the most extensively studied. While RVA infects infants and young animals, non-RVA species infect adult humans, various mammals, and birds. However, the lack of appropriate research tools has limited our understanding of non-RVA life cycles. RVA replication and assembly occur in cytosolic inclusions termed viroplasms. We recently identified viroplasm-like structures (VLS) composed of NSP5 and NSP2 in non-RVA. In this context, globular VLS induced by NSP2 formed in RVA, RVB, RVD, RVF, RVG, and RVI, but not in RVC, RVH, and RVJ. Additionally, in RVA, VLS can also be formed through the co-expression of NSP5 with VP2. Here, we report that VP2-induced VLS formed in RV species A to J, with notable formation in RVH and RVJ, where NSP2 RVH or RVJ was also recruited into VLSs. The NSP5 C-terminal region in non-RVA is required for association with VP2 and forming VLS. Mutation of conserved VP2-L124 in RVA to alanine disrupts viroplasm formation, impairing RV replication. Equivalent residues within the same predicted VP2 region disrupt VLS formation across non-RVA. We also observed interspecies VLS formation, most notably between the closely related pairs RVA-RVC, RVH-RVJ, and RVD-RVF. Interestingly, substituting the N-terminal region of VP2 from RVB with that of VP2 from RVG supported VLS formation with NSP5 from RVB in avian cells. Elucidating the formation of viroplasms is essential for developing strategies to halt infection across RV species A to J. IMPORTANCE Rotaviruses (RV) are a group of viruses classified into species A through J, with species A being the best understood. Other RV species infecting animals and humans are less studied due to limited research tools. In RVA, the virus replicates in specialized compartments called viroplasms formed in the cytoplasm by viral proteins, including NSP5, NSP2, and VP2. In this study, we explored how similar structures, termed viroplasm-like structures (VLS), are formed by proteins of RV species A-J. We found that for all tested RV species, NSP5 and VP2 form VLSs. We also identified key regions in the VP2 protein that are essential for forming these structures. Understanding how viroplasms form across different RV species may help develop new strategies to block infection in humans and animals.
middle-income countries (4). RVA also has a broad spectrum of strains primarily infecting young mammals like piglets and calves (5,6). The non-RVA species have been isolated from diverse hosts, including mammals and avians. Outbreaks from RVB, RVC, and RVH are the leading cause of diarrhea among the adult human population in several countries (7)(8)(9)(10)(11). In the US, RV infections are the second most common cause of diarrhea in adults after norovirus (12)(13)(14). From a veterinary perspective, RV infections significantly impact livestock worldwide. RV accounts for 80% of diarrhea cases in piglets in the USA, Canada, and Mexico, with potential zoonotic implications in humans (15). RV species D, F, and G have only been detected in avian species, affecting the poultry industry by impacting the feed conversion ratio and resulting in substantial economic losses (16). All the information compiled on the RV replication cycle is based on RVA. Studying the replication of non-RVA species is challenging, and as a result, their biology remains poorly understood. The few isolated viruses of non-RVA species do not replicate in tissue culture (17)(18)(19), and tools recognizing their specific proteins, like specific antibodies, are unavailable.
RV has 11 double-stranded (ds) RNA genome segments encoding six structural proteins (VP1, VP2, VP3, VP4, VP6, and VP7) and five non-structural proteins (NSP1, NSP2, NSP3, NSP4, and NSP5). In certain RVA strains, genome segment 11, in addition to NSP5, also encodes an out-of-frame protein called NSP6. The RVA virion is a non-enveloped icosahedral triple-layered particle that encloses the 11 genome segments and 12 copies of the replication intermediates, which include RNA-dependent RNA polymerase (RdRp) VP1 and the guanylmethyltransferase VP3, inside a core shell made of 12 decamers of VP2 (T = 1) (20,21). Surrounding the core shell, the middle layer consists of 260 trimers of the structural protein VP6 (T = 13), forming transcriptionally active double-layered particles (DLPs) (22,23). The outer layer is made of trimers of glycoprotein VP7 arranged in icosahedral symmetry (T = 13), standing on VP6 trimeric subunits. The spike protein VP4 is anchored in a trimeric formation at each of the fivefold axes of the virion (24)(25)(26)(27).
During RVA infection, the external layer is lost after virion internalization, and transcriptionally active DLPs are released into the cytosol (28). The newly released transcripts initiate the synthesis of viral proteins necessary for viral replication. Among those proteins, the nonstructural proteins NSP2 and NSP5 and the structural proteins VP1, VP2, VP3, and VP6 comprise part of the RV viral factories termed viroplasms (29). The viroplasms correspond to electron-dense membrane-less globular cytosolic inclusions where viral genome transcription, replication, and the packaging of the newly synthe sized pre-genomic RNA segments into the viral cores occur. The viroplasms are highly dynamic, being able to coalesce between them and migrate to the juxtanuclear region of the cell at later stages post-infection (30)(31)(32). Furthermore, despite not yet being well defined, several host factors have been identified as necessary for viroplasm forma tion and maintenance (33)(34)(35)(36). For RVA, the initiation process for viroplasm formation requires a scaffold of lipid droplets that incorporates perilipin-1 (37,38). The host cytoskeleton, including actin filaments and microtubules (MT), supports the formation and behavior of the viroplasms (31,39,40). NSP2 octamers directly associate with MTs, promoting viroplasm coalescence (31,(41)(42)(43)(44), while VP2 enables perinuclear motion (31). These characteristics align with viroplasms considered as liquid-liquid phase-separa ted structures (45). Interestingly, the co-expression of NSP5 with either NSP2 or VP2 leads to the formation of cytosolic inclusions named viroplasm-like structures (VLS), which are morphologically similar to viroplasms but unable to yield viral progeny (30,31,(46)(47)(48)(49). When associating with NSP2 or VP2, NSP5 is primed at serine-67 by the casein kinase 1 alpha, triggering NSP5 hyperphosphorylation (46,(50)(51)(52). The NSP5 S67A mutation prevents viroplasm formation (53). The NSP5 phosphorylation is consistent with a trait for recently described liquid-liquid phase separation conditions of the viroplasms (45). NSP5 is not only required for viroplasm formation and virus replication (53)(54)(55) but also plays a multifunctional role in the RV life cycle, interacting with NSP6 (49), NSP2 (30), VP1 (56), VP2 (57,58), and unspecifically with dsRNA (59). These attributes are consistent with its predicted disordered nature (60)(61)(62). Interestingly, the C-terminal ordered region (henceforth tail) of NSP5 is needed for its self-oligomerization (49,50), to associate with other RV proteins (30,49,56,58), and to form the viroplasms (53). NSP5 is sumoylated (63), presumably a prerequisite for interacting with viral or host components. Overall, NSP5 plays a crucial role in the replication of RV.
RVA octameric NSP2 is associated with several enzymatic functions, including nucleoside diphosphate kinase-like activity (64), RNA-helix destabilization (64), and nucleoside triphosphatase activity (42), all of which are consistent with molecular motor properties (42,65). Moreover, NSP2 phosphorylation and its association have been linked to viroplasm formation and dynamics (30,43,66). In this context, NSP2 octamers are directly associated with MTs to promote viroplasm coalescence (31,(41)(42)(43)(44). Interest ingly, the flexible C-terminal region of NSP2 enhances viroplasm morphology (67) and RNA chaperone activity (41). Notably, NSP2 binds both to VP1 and viral RNA (68,69), implicating it as a key component of replication intermediates within the viroplasms.
Likewise, the core-shell protein VP2, in addition to its structural role in safeguard ing the RVA genome, can activate and regulate the RdRp VP1, allowing for genome replication. VP2 forms asymmetric decameric structures converging in the fivefold axis, which cannot be dissociated (21,24,62,70,71). Each decameric subunit comprises a main domain of VP2 (residues ~100-880), creating a thin, comma-shaped plate where the unfolded N-terminal domain (NTD) is positioned beneath the decameric five-fold axis (20,24,71). Several viral proteins (22,(71)(72)(73) and nonspecific single-stranded RNA (ssRNA) (74) interact with VP2, primarily to facilitate association with the NTD. These interactions are closely linked to the core-shell structure and genome replica tion. Additionally, VP2 serves as a key component in forming viroplasms and, when co-expressed with NSP5, produces VLS (31,46,58,75). In this context, the VLSs induced by VP2 are dynamic as they migrate to the perinuclear region (31). Furthermore, the highly conserved L124 of VP2 in RVA is crucial for its association with NSP5. When L124 is mutated to alanine, VP2 L124A disrupts viroplasm morphology, rendering RV replication incompetent (58). Recently, it has been suggested that VP2 may have further roles early post-infection due to its interaction with NSP2, which prevents its spontaneous oligomerization and sumoylation, thereby enhancing the ability of VP2 to interact with other proteins (31,63).
We recently examined whether NSP5 and NSP2 from non-RVA can form VLSs (76). The co-expression of these proteins produced globular VLSs in RVA, RVB, RVD, RVF, RVG, and RVI, while RVC formed filamentous VLSs. No VLSs formed with NSP5 and NSP2 from RVH and RVJ. NSP5 from all species oligomerized via its tail and, except for RVJ, interacted with its corresponding NSP2. Interspecies VLSs formed between related species (B/G and D/F). Notably, VLSs were restored in RVH and RVJ by swapping their NSP5 tails with those of RVA.
In this study, we characterized the formation of VLS supported by the co-expression of NSP5 and VP2 across RV species A-J. We determined that the NSP5 tail is crucial for both VLS formation and its interaction with VP2 in all RV species tested. A point mutation to alanine of a conserved amino acid residue in VP2 disrupts VLS formation. Heterologous VLS formation was observed between closely related RV species pairs: A and C, B and G, D and F, as well as H and J. Additionally, we demonstrated that the unstructured N-terminal region of VP2 is necessary for VLS formation.
## RESULTS
## Biophysical features of VP2 in RV species A-J
We recently demonstrated that the replication mechanism of non-RVA species can be investigated by extrapolating the roles of NSP5 and NSP2 from RVA to their orthologs in other RV species (76). In this context, it is known that RVA forms VLS upon co-expression of NSP5 with VP2 (46,58,77). This prompted us to investigate whether VP2 from non-RVA species might similarly contribute to VLS formation when co-expressed with its cognate NSP5. To begin, we identified available VP2 open reading frames for RV species A-J in the NCBI database, matching each with its cognate NSP5 and NSP2 sequences as previously described (76). However, RVB, RVC, and RVI complete VP2 sequences were unavailable, so we substituted strains with higher homology (Table 1) (76). The VP2 proteins vary in length across RV species, with differences of up to 109 amino acids. RVA has the shortest VP2 (882 amino acids), and RVG has the longest (991 amino acids; Table 1). Sequence analysis revealed high diversity among VP2 proteins from species A to J compared to our model strain, RVA (simian strain SA11). The most similar sequence was from RVF (68.51% similarity), and the most divergent was RVJ (35.28% similarity; Table 1). Consistent with previous findings that the N-terminal domain of VP2 RVA is unfolded (residues ~1-100 for type A and ~1-80 for type B) (20,21,24,71), the PONDR analysis also predicted a highly disordered N-terminal region in the VP2 sequences of RVA model strains SA11 and OSU (Fig. 1a). Similar disordered N-terminal regions were predicted in VP2 from most non-RVA species (Fig. 1b andd), except for RVB, which lacked this feature (Fig. 1c). The predicted disordered regions in VP2 N-termini varied among species: RVA, RVC, RVD, and RVF showed completely disordered N-terminal domains (Fig. 1b), while RVG, RVH, RVI, and RVJ showed partially disordered regions, characterized by few ordered residues at the extreme N-terminus, followed by a disordered region of approximately 50 residues (Fig. 1d). We also used AlphaFold3 to compare the predicted folding of VP2 across different RV species. The predicted dimeric structure of VP2 RVA closely matched the previously experimental structure (Fig. S1, RVA) (71). For VP2 of non-RVA species, AlphaFold3 predicted similar overall structures, especially in the apical, central, and dimerization regions (Fig. S1). As expected for disordered domains, AlphaFold3 showed reduced confidence in the N-terminal regions across all analyzed RV species (data not shown). Accordingly, we designed a series of plasmids encoding the VP2 open reading frame from RV species A-J, each tagged with a Flag epitope at the N-terminus (Fig. 1e) (58). Lysates from MA104 cells expressing these Flag-VP2 constructs were assessed by immunoblotting using a polyclonal anti-VP2 antibody raised against RVA strain SA11 (56). This antibody detected VP2 from RVA and, with lower affinity, VP2 from RVB and RVC, suggesting antigenic homology for VP2 among these RV species (Fig. 1f, upper panel). Subsequent probing with an anti-Flag antibody recognized VP2 from all tested RV species, with migration patterns corresponding to their predicted molecular weight (Table S1).
## VLS formation by co-expression of NSP5 with VP2 across RV species A-J
Next, we investigated whether biotin acceptor peptide (BAP)-tagged NSP5 (76) coexpressed with their cognate Flag-VP2 protein supports the formation of VLS in RV species A-J. Of note, VLSs are visualized by colocalization of the signals of NSP5 with NSP2 or VP2 in globular cytosolic inclusions (39,46,48,58,76,77). In the first instance (Fig. 2; Fig. S2), the proteins were expressed in mammalian MA/cytBirA cells and fixed at 16 h post-transfection (hpt). VLS formation was monitored by immunofluorescence for the detection of NSP5 fused to BAP tag (streptavidin-Dylight 488, green) and Flag-VP2 (mAb anti-Flag followed by secondary antibody conjugated to Alexa 594, red), respec tively. As expected (58), NSP5-BAP and Flag-VP2 of RVA colocalized, forming globular cytosolic inclusions corresponding to VLS. Similarly, the co-expression of NSP5-BAP and Flag-VP2 of RVB, RVC, RVF, RVG, RVH, RVI, and RVJ also led to the formation of globular VLSs. However, the co-expression of these proteins in RVD did not result in VLS formation. As previously described (76), BAP-NSP5 of RVD and RVF formed globular inclusions in the nuclei. Since RVD, RVF, and RVG were originally isolated from avian hosts, we hypothesized that the host cellular environment might influence the folding and interaction behavior of NSP5 and VP2, thereby affecting VLS formation. To test this, we expressed V5-tag ged NSP5 and Flag-VP2 in LMH chicken epithelial cells and monitored VLS formation via immunofluorescence. Of note, V5-NSP5 was used instead of BAP-NSP5 because LMH cells lack the cytosolic BirA. In this context (Fig. 3a), the expression of Flag-VP2 alone from RVD, RVF, and RVG led to filamentous structures for RVD and RVG, while Flag-VP2/RVF formed globular structures, distinct from the diffuse cytosolic aggregates observed in MA/cytBirA cells for these species. As expected (76), V5-NSP5 from RVD and RVF appeared diffusely distributed in the cytosol, whereas NSP5-V5 from RVG formed globular cytosolic inclusions (Fig. 3b). Furthermore, deletion of the N-terminal "tail" region of NSP5 from RVD and RVF resulted in its localization to both the cytosol and the nucleus. In LMH cells, co-expression of V5-tagged NSP5 with Flag-VP2 from RV species D, F, and G resulted in the formation of cytosolic globular VLSs (Fig. 3c). We previously described that the predicted ordered region of NSP5, referred to as the "tail, " is located at the N-terminus in RVD and RVF and at the C-terminus in RVA, RVB, RVC, RVG, RVH, RVI, and RVJ. This tail has been shown to play a predominant role in VLS formation with NSP2 (76). Moreover, deletion of the tail region (NSP5∆T) has been shown to impair VP2-induced VLS formation in RVA (58). We investigated whether this deletion would similarly disrupt VLS formation in non-RVA species. To address this (Fig. 3d), we co-expressed Flag-VP2 (red) with either full-length NSP5 or its tail-deleted version (NSP5∆T), both BAP-tagged (green), in MA/cytBirA cells. As expected (58), co-expression of Flag-VP2/A with NSP5∆T-BAP/A impaired VLS formation. Similar impairments were observed for RVB, RVC, RVH, RVG, RVI, and RVJ, following deletion of the NSP5 tail. Surprisingly, BAP-∆TNSP5/D acquired the ability to form VLS with its cognate Flag-VP2. Although BAP-∆TNSP5/F alone formed nuclear inclusions (Fig. S3 [76]), its co-expression with Flag-VP2/F led to the formation of numerous and enlarged cytosolic VLS, as well as nuclear globular inclusions. In LMH cells (Fig. 3e andf, upper and middle rows), the co-expression of V5-∆TNSP5/D and ∆TNSP5/F with their corresponding Flag-VP2 also supported VLS formation, consistent with results in mammalian cells. In contrast (Fig. 3e andf, bottom row), co-expression of NSP5∆T-V5/G with Flag-VP2/G impaired VLS formation and resulted in the accumulation of nuclear globular inclusions, likely composed of NSP5∆T-V5/G alone.
## The NSP5 tail plays a crucial role in its interaction with VP2
Given that the ordered region of NSP5 plays a critical role in VLS formation in most RV species studied, we investigated whether this region is also required for the direct interaction between NSP5 and VP2. Previous studies using pull-down and tripartite green fluorescent protein (GFP) assays demonstrated that NSP5∆T disrupts its associa tion with VP2 in RVA (58). In this study, we developed a bioluminescence resonance energy transfer (BRET) assay to monitor NSP5-VP2 interactions in living cells. This system uses NanoLuc luciferase (NL) fused to VP2 (NL-VP2) as the energy donor and HaloTag fused to NSP5 (HT-NSP5) as the fluorescent acceptor. The BRET signal, which arises from energy transfer between NL and HT when the proteins are in close proximity, serves as a quantitative readout of interaction (Fig. S4a). First, we validated the assay by co-expressing NL-VP2 and NSP5-HT from RVA. As expected, these proteins showed a significant interaction, with BRET values markedly higher (P < 0.000001) than control pairs (HT-NSP5 + NL and HT + NL-VP2; Fig. 4a). In contrast, interaction was significantly reduced when NL-VP2 was co-expressed with tail-deleted version HT-NSP5∆T. We then extended this assay to RV species B through J, constructing NL-VP2 and HT-NSP5 fusion proteins to each species (Fig. S4b through f). All tested VP2-NSP5 showed significant interaction signals (Fig. 4b through i). However, when the NSP5 tail was deleted, the interaction with NL-VP2 was significantly impaired for all RV species tested, except RVD, where the interaction was retained. Similar results were obtained through co-immuno precipitation of cell lysates co-expressing Flag-VP2 with either full-length NSP5 or its tail-deleted version, both fused to a BAP tag, across RV species A to J, thereby validating the BRET assay (Fig. S5).
## VP2 phylogenetic analysis and structural localization of conserved critical residues involved in VLS formation
We analyzed the evolutionary relationship of VP2 across RV species A-J (Fig. 5a) by comparing their coding sequences (CDS) available in public databases. This allowed us to identify RV species pairs sharing common ancestors. Similar to what has been reported for NSP5 and NSP2 (76), VP2 phylogeny revealed two major groups, one comprising RV species A, C, D, and F, and another including RV species B, G, H, I, and J. Within this framework, RVA is most closely related to RVC, RVF to RVD, RVB to RVG, and RVH to RVJ. RVI appears to be most distantly related but shows a closer affinity to RVH and RVJ. We previously reported that a highly conserved leucine residue at position 124 (L124) in VP2 of the RVA strain SA11 is essential for viroplasm and VLS formation as well as for its association with NSP5 and efficient RV replication (58). Equivalent leucine residues were also identified in other RV species, specifically L126 in RVC, L157 in RVD, and L146 in RVF (Fig. 5b, top panel). However, no conserved leucine residues in corresponding positions were found in RVB, RVG, RVH, RVI, and RVJ (data not shown). To explore structural conservation, we mapped the RVA VP2 L124 region onto the known tertiary structure of RVA VP2 from the RRV strain (78), which spans amino acid residues 94-180 (Fig. 5c, blue region). L124 was found within a loop that precedes a beta-sheet. Using AlphaFold3, we superimposed the VP2 RVA tertiary structure with predicted VP2 structures of RV species B through J. Consistent with our sequence alignment, the predicted structures of RVC (L126), RVD (L157), and RVF (L146) showed complete overlap with RVA L124 (Fig. 5d). Furthermore, the predicted VP2 structures of RVB, RVG, RVH, RVI, and RVJ also overlapped with RVA VP2 region spanning residues 97-180 (Fig. 5e), which also includes a loop preceding a beta-sheet. Interestingly, sequence alignment of this loop revealed conserved aromatic residues, tyrosine or phenylalanine, at positions corresponding to RVA L124. We identified Y129 in RVB, Y183 in RVG, Y179 in RVH, F180 in RVI, and F184 in RVJ (Fig. 5b, bottom panel), all of which align with the same loop region as L124 in VP2 RVA.
## Conserved VP2 residue is essential for VLS formation
We hypothesized that a conserved residue in the VP2 protein of RV species B-J is critical for VLS formation, similar to the role of L124 in RVA. Supporting this, a point mutation substituting L124 with a non-bulky amino acid like alanine (L124A) was previously shown to impair VLS formation (58). To test this hypothesis, we generated Flag-VP2 constructs with alanine substitutions at the conserved residues across RV species A-J and expressed them in MA104 cells. These mutant proteins migrated at their predicted molecular weights (Fig. 6a), although Flag-VP2(Y129A) from RVB exhibited weak expression despite proper migration. VP2 point mutations did not alter protein folding, as shown by overlapping AlphaFold3 predictions with wild type (wt) VP2 (Fig. S6a), and both wt and mutant Flag-VP2 displayed identical proteinase K cleavage patterns (Fig. S6b).
To assess VLS formation, we performed immunofluorescence microscopy in MA/ cytBirA cells co-expressing NSP5-BAP with either wt Flag-VP2 or Flag-VP2 point mutant from RVA, RVB, RVC, RVH, RVI, and RVJ (Fig. 6b, rows i, ii, iii, vii, viii, and ix). As previously reported (58), co-expression of NSP5-BAP with RVA Flag-VP2 (L124A) failed to support VLS formation (Fig. 6b, row i). Similarly, alanine substitutions in the VP2 proteins of RVB, RVC, RVI, and RVJ also impaired VLS formation when co-expressed with their respective NSP5-BAPs, in contrast to the robust VLS formation observed with the corresponding wt proteins. Interestingly, Flag-VP2 (Y179A) from RVH retained the ability to support VLS formation when co-expressed with RVH NSP5-BAP (Fig. 6b, row vii).
For avian RV species RVD, RVF, and RVG, the corresponding Flag-VP2 point mutants were tested in LMH cells to provide a more suitable host environment (Fig. 6b, rows iv, v, and vi). While the co-expression of wt Flag-VP2 with its cognate NSP5 fused to V5 supported VLS formation in these RV species, the respective alanine mutants, L157A (RVD), L146A (RVF), and Y183A (RVG), failed to form VLSs.
## VLS morphology is modulated by NSP5, NSP2, and VP2
We previously reported that the co-expression of cognate NSP5 with NSP2 leads to the formation of globular VLSs in RVA, RVB, RVD, RVG, and RVI, while RVC forms filamentous VLSs and RVH and RVJ fail to form VLSs (Table 2) (76). In this study, we observed that co-expression of NSP5 with VP2 resulted in globular VLS formation in all RV species tested. To assess whether NSP2 influences the morphology of VP2-induced VLSs (Table 2), we compared the morphology of VLS formed by the co-expression of NSP5 and VP2, VLS (NSP5 + VP2), with those formed by the co-expression of NSP5, NSP2, and VP2, VLS (NSP5 + VP2 + NSP2; Fig. 7a). We found that the addition of NSP2 led to globular VLS morphology in all RV species, with the exception of RVC, which retained a filamentous morphology. Notably, in this condition, VLS (NSP5 + VP2 + NSP2) facilitated the recruitment of NSP2 in RVH and RVJ. We also investigated the impact of VP2 point mutations on VLS(NSP5 + VP2 + NSP2) morphology (Fig. 7b). As previously shown by Buttafuoco et al. (58), the Flag-VP2(L124A) disrupted VLS(NSP5 + VP2 + NSP2) in RVA. Similarly, the corresponding VP2 point mutations in other RV species impaired VLS integrity. This was evident in RVD and RVF, where small, punctate VLSs formed lacking detectable VP2, and in RVB, RVG, RVH, and RVI, where VLSs appeared irregular, and VP2 was dispersed throughout the cytosol. Strikingly, VLS formation was completely abolished in RVC and RVJ, resulting in the loss of their characteristic filamentous and globular morphologies, respectively.
## Heterologous formation of VP2-induced VLSs among RV species
We previously demonstrated that NSP5 and NSP2 from closely related RV species pairs can be interchanged to form heterologous VLSs (76). This was observed for the pairs RVA/RVC, RVB/RVG, and RVD/RVF. Given that VP2 shares the same phylogenetic distribution with NSP5 and NSP2 (Fig. 5a), we wondered whether heterologous VLSs could also be formed by co-expressing NSP5 and VP2 from these closely related RV species. To test this, we co-expressed NSP5-BAP with Flag-VP2 of RVA and RVC in all four interspecies combinations of NSP5 and VP2 (A/A, C/C, A/C, and C/A; Fig. 8a, top panel). All the combinations supported VLS formation, although with varying morphologies, ranging from large (NSP5/RVA with VP2/RVA) to smaller, punctate structures (NSP5/RVC with VP2/RVA). In contrast, NSP5 and VP2 from RVB and RVG did not support heterolo gous VLS formation in any combination (B/G or G/B), either in mammalian cells (Fig. 8a, middle) or in avian cells (Fig. 8b, top). Homologous RVB VLSs were also not supported in LMH chicken cells, whereas homologous RVG VLSs were. By comparison, heterologous VLSs formed successfully in all four interspecies pairings of RVH with RVJ (Fig. 8a, bottom) and of RVD with RVF (Fig. 8b, bottom), indicating full compatibility between their respective NSP5 and VP2 proteins. with NSP2 from RVG (76). However, heterologous VLS formation between NSP5 and VP2 from RVB and RVG was not supported in both cell lines tested. Even more intriguing, homologous VLS RVB were not observed in LMH chicken cells. We hypothesized that VP2 from RVB may differ functionally from VP2 in other RV species due to the absence of an unstructured N-terminal region (Fig. 1c). To test whether this region is required for VLS formation, we used AlphaFold3 to compare the predicted tertiary structures of VP2 from RVB and its close relative, RVG. The first common structural element identified was a beta-sheet beginning at valine 85 in RVB VP2 and aspartic acid 138 in RVG VP2. Based on this, we designed a chimeric VP2 protein (VP2/G-B) by replacing the N-terminal region of RVB VP2 with amino acids 1-137 from RVG VP2 (Fig. 9a). The resulting chimera VP2/G-B was predicted using PONDR score to contain a disordered N-terminal region resembling that of VP2/G, while retaining the apical, central, and dimerization regions of RVB VP2 (Fig. 9b). We then expressed the chimeric protein as Flag-VP2/G-B, which migrated at the expected molecular weight (Fig. 9c andd).
We next co-expressed BAP-or V5-tagged NSP5 from RVB (left panels) or RVG (right panels) with Flag-tagged VP2 from RVB, RVG, or the chimeric Flag-VP2/G-B in mam malian (Fig. 9e) and avian (Fig. 9f) cells. As expected, homologous RVG VLSs formed in both cell types, whereas homologous RVB VLSs formed only in mammalian cells. Notably, co-expression of NSP5 from RVB with VP2/G-B supported VLS formation in both mammalian and avian cells, while NSP5 from RVG with VP2/G-B did not.
## DISCUSSION
Understanding the RV life cycle, particularly the assembly of virions, is crucial for controlling its spread. RV includes nine species, from A to J, that infect many mammals and birds. Notably, two new RV species (RVK and RVL) were recently added by the ICTV, although they were not part of this study. RVA viroplasms are cytosolic globular inclusions that facilitate virus genome replication, sorting, and packaging in newly assembled viral cores. Studying the life cycle of non-RVA species is difficult due to limited research tools, such as adapted viruses for tissue culture, specific antibodies, and reverse genetics tailored for non-RVA species. We recently addressed this challenge by applying the role of orthologous proteins responsible for VLS formation to non-RVA (76). Using this approach, we described how NSP5 can form VLS when co-expressed with NSP2 in certain RV species, including RVB, RVD, RVF, RVG, and RVI. Similarly, in this study, we examined VLS formation across RV species A-J by co-expressing NSP5 with VP2, using confocal immunofluorescence microscopy. We found that VLS can form across RV species A-J. These findings differ from our previous research, particularly regarding RVH and RVJ. Conversely, in other species such as RVH and RVJ, where NSP2 is not necessary, VP2 plays an essential role. Similar to RVA, NSP2 and VP2 had a complementary role in VLS formation for RVB, RVC, RVD, RVF, RVG, and RVI. Our results also show that NSP5 and VP2 directly interact, as confirmed by the BRET assay, suggesting their association influences VLS formation. These findings highlight the significant role of VP2 in viroplasm formation.
Similar to how VLSs form with NSP2 (30,72,76), we also demonstrate that the tail region of NSP5 is essential for both its interaction with VP2 and the induction of VLSs across multiple RV species, including RVA, RVB, RVC, RVG, RVH, RVI, and RVJ (Table 2). We previously described that the deletion of the ordered region of NSP5 in RVD and RVF, located at their N-terminus instead of the C-terminus as in other studied RV species, does not affect VLS formation induced by NSP2 (76). Similarly, the co-expression of ∆TNSP5 with VP2 of RVD or RVF enhances VLS formation in mammalian cells while consistently forming VLS in chicken epithelial cells. Notably, the VLS of ∆TNSP5 with VP2 of RVF also formed nuclear globular inclusions in both cell types, seemingly composed solely of NSP5. Therefore, the nuclear translocation of ∆TNSP5/F influences its cytosolic interac tion with VP2, which is consistent with the decreased binding of these two proteins in the BRET assay. We want to point out that the addition of a Flag tag at the N-terminus of VP2 was based on previous evidence showing that the N-terminus of VP2 is flexible, and HA tagging the N-terminus of VP2 RVA supports the formation of VLS (58,71,79).
The viroplasms are complex structures composed of several viral proteins, each potentially contributing to viroplasm morphology. We previously demonstrated that RVA VLS induced by either NSP2 or VP2 can recruit other viral proteins (39,46,58,77). Here, we determined that VLS induced by VP2 could incorporate NSP2 for RV species A-J. Interestingly, when NSP2, NSP5, and VP2 from various RV species were expressed, most formed VLSs with a globular shape. However, RVC was an exception, producing filamentous VLSs, similar to those induced by RVC NSP5 and NSP2. This result suggests that NSP2 plays a major role, over other RVC proteins, in determining the morphology of RVC VLSs. In contrast, VLS induced by VP2 from RVH and RVJ permitted the recruitment of NSP2, maintaining their globular morphology (Table 2). We previously demonstrated that the association of the respective NSP5 and NSP2 of RVH and RVJ is weak or not detectable (76). Here, we show that NSP5 and VP2 from these RV species interact and form VLSs, suggesting that VLSs composed of NSP5, NSP2, and VP2 arise either through direct interaction of both NSP5 and NSP2 with VP2, or that VP2 enhances the otherwise weak association between NSP5 and NSP2.
Our earlier findings showed that a conserved residue in VP2 RVA, L124, is necessary for the formation of VLS as well as for maintaining globular morphology and the ability of viroplasms to replicate (58). Here, we found that this conserved residue occupies a similar tertiary position in RV species B through J, as a leucine for RVC, RVD, and RVF, and as an aromatic residue, tyrosine, for RVB, RVG, and RVH, and phenylalanine for RVI and RVJ. Indeed, substituting this conserved residue with alanine disrupts VLS formation in RV species B-J, whether the VLS are induced by VP2 or formed by a combination of NSP5, NSP2, and VP2. The resulting disrupted VLSs showed two distinct patterns: in RVA, RVC, and RVJ, the proteins were completely dispersed throughout the cytosol, while in RVB, RVD, RVF, RVG, RVH, and RVI, the VLSs were smaller and mainly composed of NSP5 and NSP2, with VP2 dispersed in the cytosol. It is important to note that the substitution of this residue by non-bulky alanine in VP2 of all RV species tested does not seem to affect its folding, as denoted by AlphaFold 3 structural prediction and the fragment pattern from cleavage with proteinase K when compared with their respective wt VP2. These observations suggest that this conserved residue plays a critical structural role in VLS formation across RV species A-J.
Heterologous VLS formation is also observed with NSP5 and VP2 from closely related RV species, such as RVA with RVC, RVF with RVD, and RVH with RVJ, suggesting that genetic reassortment among these RV species may be possible in principle. In this sense, the NSP5 and NSP2 of closely related RV species, RVA with RVC and RVD with RVF, can also be interchanged (76). Since RVH and RVJ do not form VLS with NSP5 and NSP2, it remains unclear whether they can be interchangeable. However, we now demonstrate that NSP5 and VP2 of RVH and RVJ can form interspecies VLS. The formation of triple VLS involving NSP5, NSP2, and VP2 among RVH and RVJ suggests that reassortment may also occur. An interesting case involves RVB and RVG, which previously showed the ability to form heterologous VLS between NSP5 and NSP2. In contrast, NSP5 and VP2 behaved differently. Formation of homologous RVB VLS was supported only in mammalian cells, not in avian cells. Our results show that the lack of a disordered N-terminal region in VP2/B prevents heterologous VLS formation in chicken cells, whereas replacing this region with that of VP2/G enables VLS formation with NSP5/B but not with NSP5/G. These findings suggest that reassortment in these RV species depends not only on viral proteins but also on host proteins provided by specific cellular environments. Nonethe less, we cannot rule out the possibility that a natural recombination of VP2 of RVB with its closely related VP2 RVG could lead to the acquisition of a disordered N-terminal region. Notably, the VLS (NSP5 + VP2) formation is supported between intraspecies strains, as previously demonstrated with VP2 RVA simian strain SA11 with NSP5 RVA from either simian strain SA11 or porcine strain OSU (58). Similarly, VLSs are also supported with NSP5 RVA simian strain SA11 with VP2 RVA from either simian strain SA11 or porcine strain OSU (Fig. S7a). In this context, intraspecies reassortment supporting VLS (NSP5 + VP2) formation is plausible since the high similarity of NSP5 and VP2 between strains (Fig. S7b andc). Therefore, we also describe in this study for the first time that the disordered region of VP2 not only plays a role in the association with replication intermediates VP1 and VP3 in the core virion (22,71,73,79) but also in the formation of VLS and, by extension, probably of viroplasms. However, it is important to keep in consideration that NSP5, NSP2, and VP2 are only a few elements in the RV life cycle, and their interaction with RdRp VP1 could also influence reassortment (80).
A previous study demonstrated that RVA viroplasms act as liquid-liquid phase-sep arated structures, driven primarily by NSP5 and NSP2, while VP2 was not examined due to difficulties in maintaining it in a homogeneous solution (45). However, it was suggested that the positively charged surface of NSP2 and poly-arginine-rich motifs in the N-terminus of RVA VP2 might facilitate droplet formation with NSP5. Consistent with this observation, non-RVA VP2 proteins are also enriched in basic residues (lysines and arginines) in their predicted N-terminal region (Table 3), ranging from 9.4% in RVB to 26.6% in RVD. Moreover, the present study provides essential insights into the ability of VP2 in non-RVA species to act as a client protein within NSP5 condensates, particularly in the formation of VLS in RVH and RVJ, which can arise only through association between NSP5 and VP2, and not with NSP2. Further research is needed to explore the liquid-liquid phase separation properties of VLS in non-RVA species in greater depth.
RV reverse genetics has been established only for certain RVA strains, such as simian SA11 (81), porcine OSU (82), and human KU (83), and is not available for other RVA strains and non-RVA species. Understanding viroplasms is crucial for applying reverse genetics to non-RVA species, as the co-expression of proteins like NSP5 and NSP2 significantly enhances the recovery of recombinant rotaviruses (84). For future experiments exploring reverse genetics in other RV species, it appears that for RVH and RVJ, the co-expression of NSP5 and VP2 will be favored, instead of NSP5 and NSP2, for the successful recovery of recombinant virus.
## MATERIALS AND METHODS
## Cells and viruses
MA104 (embryonic rhesus monkey kidney, ATCCCRL-2378, RRID: CVCL_3845) cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Gibco BRL) supplemen ted with 10% fetal calf serum (FCS, AMIMED, Bioconcept, Switzerland) and penicillin (100 U/mL)-streptomycin (10 µg/mL). MA/cytBirA (39) were cultured in DMEM supple mented with 10% FCS, penicillin (100 U/mL)-streptomycin (10 µg/mL), and 5 µg/mL puromycin (InvivoGen, France). LMH cells (chicken hepatocellular carcinoma epithelial, ATCCCRL2117) were cultured in Waymouth's MB572/1 (Sartorius) medium supplemented with 10% FCS and penicillin (100 U/mL)-streptomycin (100 µg/mL). HEK-293T (human embryonic kidney, ATCCCRL-3216) cells were cultured in DMEM supplemented with 10% FCS and penicillin (100 U/mL)-streptomycin (10 µg/mL).
The recombinant vaccinia virus encoding T 7 RNA polymerase (strain vvT7.3) was amplified as previously described (85).
## Antibodies and reagents
Guinea pig anti-VP2 was described previously (77). Mouse monoclonal (mAb) anti-tubu lin (clone B5-1-12) and mouse mAb anti-Flag (clone M2) were purchased from Merck. AlexaFluor 594 anti-HA.11 (clone 16B12) and AlexaFluor 647 anti-Flag Tag (clone L5) were purchased from BioLegend. Mouse mAb-V5 Tag-Dylight 488 was purchased from Invitrogen. Streptavidin-Dylight488 and mouse secondary antibodies conjugated to AlexaFluor 488 or AlexaFluor 594 were purchased from Thermo Fisher Scientific. The secondary antibodies for immunoblot conjugated to IRDye680CW and IRDye800CW were purchased from LI-COR. Mouse mAb anti-NanoLuc and HaloTagTMRDirectLigand (Cat# G2991) were purchased from Promega.
## Rotavirus sequences
The sequences of rotavirus NSP5 and NSP2 open reading frames from species B to J used in this study were previously published by (76). The sequences of the rotavirus VP2 open reading frames from RV species A to J are provided in the supplemental material and Table 1.
## Plasmid constructs
The The version of the constructs pCI-Flag-VP2/A, B, C, D, F, G, H, I, and J as well as pCI-NanoLuc-Flag-VP2/A, B, C, D, F, G, H, I, and J harboring VP2 point mutations L124A, Y129A, L126A, L157A, L146A, Y183A, Y179A, F180A, and F184A, respectively, was built by insertion of point mutations using the QuickChange site-directed mutagenesis protocol (Agilent).
The chimeric pCI-Flag-VP2/G-B was obtained by insertion in between MluI and PciI of pCI-Flag-VP2/B of a synthetic DNA segment (GeneArt Technology, Invitrogen, Table S2) containing an in-frame sequence of Flag tag, N-terminal region of VP2/G (region 1-137) and VP2/B region 85-163.
All the oligonucleotides were obtained from Microsynth AG, Switzerland, and described in Table S3.
## AlphaFold predictions
Protein structures of VP2 dimers were predicted using the AlphaFold3 server (https:// alphafoldserver.com/about) (86). As a reference for VP2 folding, the PDB of RVA VP2 strain RRV was used (6OGZ, https://doi.org/10.2210/pdb6OGZ/pdb).
## IDR predictions
The intrinsically disordered regions of proteins were determined with PONDR (Molecular Kinetics, Inc., https://www.pondr.com/) using the VSL2 algorithm. Data were plotted with GraphPad Prism (version 10.4.2).
## Immunofluorescence
MA/cytBirA and LMH cells were transfected and treated for immunofluorescence, as described previously by (76). For VLS formation composed of NSP5 and VP2, a ratio of 2:1 was used, with 2 µg and 1 µg of DNA plasmids, respectively. With the exception of VLS induced by VP2 of RVI, which were obtained with a transfection ratio of NSP5 and VP2 of 1:2, respectively (Fig. S2), VLS composed of NSP5, VP2, and NSP2 was obtained with a ratio of 2:1:1 using 2 µg, 1 µg, and 1 µg of DNA plasmids, respectively. The images were acquired using a confocal laser scanning microscope (DM550Q, Leica). Data were analyzed with Leica Application Suite (Mannheim, Germany) and ImageJ2 (version: 2.16.0/1.54 p, https://imagej.net/software/imagej2/).
## Immunoblotting
Cell lysis and immunoblotting procedures were performed as described by Lee et al. (76).
## Detection of Halo-tagged proteins
MA104 cells seeded at a density of 2 × 10 5 cells per well in a 12-well plate. The cells were infected with vvT7.3 (multiplicity of infection [MOI]: 1 PFU/cell), followed by transfection with 1 µg DNA plasmid using 3 µL of Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer's instructions. At 16 hpt, the cells were lysed in 30 µL TNN buffer (100 mM Tris-HCl, pH 8.0, 250 mM NaCl, 0.5% nonidet P-40, and cOmplete protease inhibitor cocktail [Roche, Switzerland]) for 10 min on ice. The cell lysate was centrifuged at 17,000 × g for 7 min at 4°C. Then, 10 µL of supernatant was incubated with 10 µL of 2.5 µM HaloTagTMRDirect Ligand (Promega) in DMSO. The sample was incubated for 30 min in the dark at room temperature, followed by the addition of 10 µL of sample buffer (8% SDS, 40% glycerol, 200 mM Tris-HCl pH 6.8, 0.8% bromophenol blue, and 5 mM 2-mercaptoethanol). The samples were heated at 70°C for 3 min and migrated in an SDS-polyacrylamide gel followed by acquisition at 520 nm channel at Odyssey M Imager (LI-COR Biosciences).
## NanoBRET protein-protein interaction
HEK-293T cells were seeded at 8 × 10 5 cells per well in six-well plates. At 4 h post-seeding, the cells were transfected in a ratio HaloTag: NanoLuc of 10:1, by adding 2,000 ng and 200 ng of the respective DNA plasmids, using 6 µL of Lipofectamine LTX transfection reagent (ThermoFisher Scientific) diluted in 100 µL of Opti-MEM reduced medium. The transfection mixture was incubated for 30 min at room temperature and added to the cells. At 20 hpt, the cells were counted and diluted to 2 × 10 5 cells per mL in 4% FCS in OptiMEM-I reduced serum medium. Then, 500 µL of diluted cells was mixed with 0.5 µL of 0.1 mM HaloTagNanoBRET618 Ligand (+ ligand, Promega) or 0.5 µL DMSO (-Ligand). Then, 40 µL of each mixture was distributed in quadruplicates in a white wall 384-wells plate. The cells were incubated for 6 h at 37°C and 5% CO2. Afterward, 10 µL of 5× solution of NanoBRET Nano-Glo substrate in Opti-MEM reduced serum medium was added per well. The luminescence was measured in a range of 10 min, at 460 nm and 618 nm for donor emission and acceptor emission, respectively, using a Spark instrument (TECAN). The BRET ratio corresponds to the mean corrected mBU, which is obtained as follows:
Mean corrected mBU = Mean mBU +ligand -Mean mBU -Ligand
Where: mBU = 618 nm/460 nm × 1000. Statistical analysis was performed using:
## Phylogenetic tree analysis
The CDS for rotavirus VP2 proteins was translated in silico into amino acid sequen ces using EMBOSS "transeq" (http://emboss.open-bio.org). The protein sequences were aligned using "mafft" (MAFFT v7.475 [23 November 2020]; https://mafft.cbrc.jp/ alignment/software/), and the aligned protein sequences were backtranslated using EMBOSS "transeq" (http://emboss.open-bio.org) (87,88). The phylogeny from the nucleotide multiple sequence alignments was then inferred by using "BEAUTi" and "BEAST" (v1.10.4) (89). In brief, 10,000,000 Markov chain Monte Carlo steps with the Juke-Cantor model were performed, saving each 10,000th tree. After the burn-in of 100,000 states, the consensus tree for VP2 was calculated and visualized using FigTree (https://beast.community).
## FUNDING Funder Grant(s) Author(s)
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung 10.005.230 Catherine Eichwald
## References
1. Matthijnssens, Attoui, Bányai et al. (2022) "ICTV virus taxonomy Profile: Spinareoviridae 2022" *J Gen Virol*
2. Johne, Tausch, Ulrich et al. (2023) "Genome analysis of the novel putative rotavirus species K" *Virus Res*
3. Johne, Schilling-Loeffler, Ulrich et al. (2022) "Whole genome sequence analysis of a prototype strain of the novel putative rotavirus species L" *Viruses*
4. Troeger, Khalil, Rao et al. (2018) "Rotavirus vaccination and the global burden of rotavirus diarrhea among children younger than 5 years" *JAMA Pediatr*
5. Papp, Matthijnssens, Martella et al. (2013) "Global distribution of group A rotavirus strains in horses: a systematic review" *Vaccine (Auckl)*
6. Ghonaim, Rouby, Nageeb et al. (2025) "Insights into recent advancements in human and animal rotavirus vaccines: exploring new frontiers" *Virol Sin*
7. Jain, Thakur, Vashistt et al. (2016) "Predominance of unusual rotavirus G1P[6] strain in North India: an evidence from hospitalized children and adult diarrheal patients" *Infect Genet Evol*
8. Luchs, Cilli, Morillo et al. (2014) "Rotavirus in adults, Brazil, 2004-2011: G2P[4] dominance and potential impact on vaccination" *Braz J Infect Dis*
9. Rubilar-Abreu, Hedlund, Svensson et al. (2005) "Serotype G9 rotavirus infections in adults in Sweden" *J Clin Microbiol*
10. Shen, Wang, Zhang et al. (2013) "Rotavirus infection and its genetic characteriza tion in non-hospitalized adults with acute gastroenteritis in Shanghai, China" *Arch Virol*
11. Zhirakovskaia, Tikunov, Klemesheva et al. (2016) "First genetic characterization of rotavirus C in Russia" *Infect Genet Evol*
12. Anderson, Katz, Polin et al. (2012) "Rotavirus in adults requiring hospitalization" *J Infect*
13. Karve, Krishnarajah, Korsnes et al. (2014) "Burden of acute gastroenteritis, norovirus and rotavirus in a managed care population" *Hum Vaccin Immunother*
14. Pacilli, Cortese, Smith et al. (2015) "Outbreak of gastroenteritis in adults due to rotavirus genotype G12P[8]" *Clin Infect Dis*
15. Vlasova, Amimo, Saif (2017) "Porcine Rotaviruses: epidemiology, immune responses and control strategies" *Viruses*
16. Dhama, Saminathan, Karthik et al. (2015) "Avian rotavirus enteritis -an updated review" *Vet Q*
17. Guo, Raev, Kick et al. (2022) "Rotavirus C replication in porcine intestinal enteroids reveals roles for cellular cholesterol and sialic acids" *Viruses*
18. Diller, Parrington, Patton et al. (2019) "Rotavirus species B encodes a functional fusion-associated small transmembrane protein" *J Virol*
19. Gao, Shen, Zhao et al. (2023) "Isolation and pathogenicity analysis of a G5P[23] porcine rotavirus strain" *Viruses*
20. Lawton, Estes, Prasad (1997) "Three-dimensional visualization of mRNA release from actively transcribing rotavirus particles" *Nat Struct Biol*
21. Zhang, Settembre, Xu et al. (2008) "Near-atomic resolution using electron cryomicro scopy and single-particle reconstruction" *Proc Natl Acad Sci*
22. Charpilienne, Lepault, Rey et al. (2002) "Identification of rotavirus VP6 residues located at the interface with VP2 that are essential for capsid assembly and transcriptase activity" *J Virol*
23. Lepault, Petitpas, Erk et al. (2001) "Structural polymorphism of the major capsid protein of Full-Length Text Journal of Virology November"
24. *EMBO J*
25. Li, Baker, Jiang et al. (2009) "Rotavirus architec ture at subnanometer resolution" *J Virol*
26. Settembre, Chen, Dormitzer et al. (2011) "Atomic model of an infectious rotavirus particle" *EMBO J*
27. Yoder, Trask, Vo et al. (2009) "VP5* rearranges when rotavirus uncoats" *J Virol*
28. Yoder, Dormitzer (2006) "Alternative intermolecular contacts underlie the rotavirus VP5* two-to three-fold rearrangement" *EMBO J*
29. Trask, Mcdonald, Patton (2012) "Structural insights into the coupling of virion assembly and rotavirus replication" *Nat Rev Microbiol*
30. Patton, Silvestri, Tortorici et al. (2006) "Rotavirus genome replication and morphogenesis: role of the viroplasm"
31. Eichwald, Rodriguez, Burrone (2004) "Characterization of rotavirus NSP2/NSP5 interactions and the dynamics of viroplasm formation" *J Gen Virol*
32. Eichwald, Arnoldi, Laimbacher et al. (2012) "Rotavirus viroplasm fusion and perinuclear localization are dynamic processes requiring stabilized microtubules" *PLoS One*
33. Carreño-Torres, Gutiérrez, Arias et al. (2010) "Characteri zation of viroplasm formation during the early stages of rotavirus infection" *Virol J*
34. Zhou, Chen, Du et al. (2020) "MicroRNA-7 inhibits rotavirus replication by targeting viral NSP5 in vivo and in vitro" *Viruses*
35. Cheung, Gill, Esposito et al. (2010) "Rotaviruses associate with cellular lipid droplet components to replicate in viroplasms, and compounds disrupting or blocking lipid droplets inhibit viroplasm formation and viral replication" *J Virol*
36. Zambrano, Sorondo, Alcala et al. (2012) "Rotavirus infection of cells in culture induces activation of RhoA and changes in the actin and tubulin cytoskeleton" *PLoS One*
37. Chandra, Banerjee, Saha et al. (2022) "Sneaking into the viral safe-houses: implications of host components in regulating integrity and dynamics of rotaviral replication factories" *Front Cell Infect Microbiol*
38. Criglar, Estes, Crawford (2022) "Rotavirus-induced lipid droplet biogenesis is critical for virus replication" *Front Physiol*
39. Martínez, Eichwald, Schraner et al. (2022) "Lipid metabolism is involved in the association of rotavirus viroplasms with endoplasmic reticulum membranes" *Virology (Auckl)*
40. Vetter, Papa, Seyffert et al. (2022) "Rotavirus spike protein VP4 mediates viroplasm assembly by association to actin filaments" *J Virol*
41. Cabral-Romero, Padilla-Noriega (2006) "Association of rotavirus viroplasms with microtubules through NSP2 and NSP5" *Mem Inst Oswaldo Cruz*
42. Bravo, Bartnik, Venditti et al. (2021) "Structural basis of rotavirus RNA chaperone displacement and RNA annealing" *Proc Natl Acad Sci*
43. Taraporewala, Chen, Patton (1999) "Multimers formed by the rotavirus nonstructural protein NSP2 bind to RNA and have nucleoside triphosphatase activity" *J Virol*
44. Martin, Duarte, Lepault et al. (2010) "Sequestration of free tubulin molecules by the viral protein NSP2 induces microtubule depolymerization during rotavirus infection" *J Virol*
45. Jing, Shi, Chen et al. (2021) "Rotavirus viroplasm biogenesis involves microtubule-based dynein transport mediated by an interaction between NSP2 and dynein intermediate chain" *J Virol*
46. Geiger, Acker, Papa et al. (2021) "Liquid-liquid phase separation underpins the formation of replication factories in rotaviruses" *EMBO J*
47. Contin, Arnoldi, Campagna et al. (2010) "Rotavirus NSP5 orchestrates recruitment of viroplasmic proteins" *J Gen Virol*
48. Eichwald, Vascotto, Fabbretti et al. (2002) "Rotavirus NSP5: mapping phosphorylation sites and kinase activation and viroplasm localization domains" *J Virol*
49. Fabbretti, Afrikanova, Vascotto et al. (1999) "Two nonstructural rotavirus proteins, NSP2 and NSP5, form viroplasm-like structures in vivo" *J Gen Virol*
50. Torres-Vega, González, Duarte et al. (2000) "The C-terminal domain of rotavirus NSP5 is essential for its multimerization, hyperphosphorylation and interaction with NSP6" *J Gen Virol*
51. Eichwald, Jacob, Muszynski et al. (2004) "Uncoupling substrate and activation functions of rotavirus NSP5: phosphorylation of Ser-67 by casein kinase 1 is essential for hyperphos phorylation" *Proc Natl Acad Sci*
52. Campagna, Budini, Arnoldi et al. (2007) "Impaired hyperphosphorylation of rotavirus NSP5 in cells depleted of casein kinase 1α is associated with the formation of viroplasms with altered morphology and a moderate decrease in virus replication" *J Gen Virol*
53. Afrikanova, Fabbretti, Miozzo et al. (1998) "Rotavirus NSP5 phosphorylation is up-regulated by interaction with NSP2" *J Gen Virol*
54. Papa, Venditti, Arnoldi et al. (2019) "Recombinant rotaviruses rescued by reverse genetics reveal the role of NSP5 hyperphosphorylation in the assembly of viral factories" *J Virol*
55. Campagna, Eichwald, Vascotto et al. (2005) "RNA interfer ence of rotavirus segment 11 mRNA reveals the essential role of NSP5 in the virus replicative cycle" *J Gen Virol*
56. López, Rojas, Ayala-Bretón et al. (2005) "Reduced expression of the rotavirus NSP5 gene has a pleiotropic effect on virus replication" *J Gen Virol*
57. Arnoldi, Campagna, Eichwald et al. (2007) "Interaction of rotavirus polymerase VP1 with nonstructural protein NSP5 is stronger than that with NSP2" *J Virol*
58. Berois, Sapin, Erk et al. (2003) "Rotavirus nonstructural protein NSP5 interacts with major core protein VP2" *J Virol*
59. Buttafuoco, Michaelsen, Tobler et al. (2020) "Conserved rotavirus NSP5 and VP2 domains interact and affect viroplasm" *J Virol*
60. Vende, Taraporewala, Patton (2002) "RNA-binding activity of the rotavirus phosphoprotein NSP5 includes affinity for double-stranded RNA" *J Virol*
61. Jiang, Jayaram, Kumar et al. (2006) "Cryoelectron microscopy structures of rotavirus NSP2-NSP5 and NSP2-RNA complexes: implications for genome replication" *J Virol*
62. (2025) *Full-Length Text Journal of Virology*
63. Martin, Charpilienne, Parent et al. (2013) "The rotavirus nonstructural protein NSP5 coordinates a [2Fe-2S] iron-sulfur cluster that modulates interaction to RNA" *FASEB J*
64. Mcdonald, Patton (2011) "Rotavirus VP2 core shell regions critical for viral polymerase activation" *J Virol*
65. Campagna, Villar, Arnoldi et al. (2013) "Rotavirus viroplasm proteins interact with the cellular SUMOylation system: implications for viroplasm-like structure formation" *J Virol*
66. Kumar, Jayaram, Vasquez-Del Carpio et al. (2007) "Crystallographic and biochemical analysis of rotavirus NSP2 with nucleotides reveals a nucleoside diphosphate kinase-like activity" *J Virol*
67. Jayaram, Taraporewala, Patton et al. (2002) "Rotavirus protein involved in genome replication and packaging exhibits a HIT-like fold" *Nature*
68. Criglar, Crawford, Zhao et al. (2020) "A genetically engineered rotavirus NSP2 phosphorylation mutant impaired in viroplasm formation and replication shows an early interaction between vNSP2 and cellular lipid droplets" *J Virol*
69. Nichols, Nilsson, Brown-Harding et al. (2023) "Flexibility of the rotavirus NSP2 C-terminal region supports factory formation via liquid-liquid phase separation" *J Virol*
70. Kattoura, Clapp, Patton (1992) "The rotavirus nonstructural protein, NS35, possesses RNA-binding activity in vitro and in vivo" *Virology (Auckl)*
71. Kattoura, Chen, Patton (1994) "The rotavirus RNA-binding protein NS35 (NSP2) forms 10S multimers and interacts with the viral RNA polymerase" *Virology (Auckl)*
72. Lawton, Zeng, Mukherjee et al. (1997) "Three-dimensional structural analysis of recombinant rotavirus-like particles with intact and amino-terminal-deleted VP2: implications for the architecture of the VP2 capsid layer" *J Virol*
73. Mcclain, Settembre, Temple et al. (2010) "Xray crystal structure of the rotavirus inner capsid particle at 3.8 A resolution" *J Mol Biol*
74. Viskovska, Anish, Hu et al. (2014) "Probing the sites of interactions of rotaviral proteins involved in replication" *J Virol*
75. Zeng, Estes, Charpilienne et al. (1998) "The N terminus of rotavirus VP2 is necessary for encapsidation of VP1 and VP3" *J Virol*
76. Labbé, Baudoux, Charpilienne et al. (1994) "Identification of the nucleic acid binding domain of the rotavirus VP2 protein" *J Gen Virol*
77. Laimbacher, Esteban, Castello et al. (2012) "HSV-1 amplicon vectors launch the production of heterologous rotavirus-like particles and induce rotavirus-specific immune responses in mice" *Mol Ther*
78. Lee, Cosic, Tobler et al. (2024) "Characterization of viroplasm-like structures by co-expression of NSP5 and NSP2 across rotavirus species A to J" *J Virol*
79. Vetter, Papa, Tobler et al. (2024) "The recruitment of TRiC chaperonin in rotavirus viroplasms correlates with virus replication" *mBio*
81. Ding, Celma, Zhang et al. (2019) "In situ structures of rotavirus polymerase in action and mechanism of mRNA transcription and release" *Nat Commun*
82. Estrozi, Settembre, Goret et al. (2013) "Location of the dsRNA-dependent polymerase, VP1, in rotavirus particles" *J Mol Biol*
83. Mingo, Zhang, Long et al. (2017) "Genetic determinants restricting the reassortment of heterologous NSP2 genes into the simian rotavirus SA11 genome" *Sci Rep*
84. Kanai, Komoto, Kawagishi et al. (2017) "Entirely plasmid-based reverse genetics system for rotaviruses" *Proc Natl Acad Sci*
85. Snyder, Agbemabiese, Patton (2024) "Production of OSU G5P[7] porcine rotavirus expressing a fluorescent reporter via reverse genetics" *Viruses*
86. Komoto, Fukuda (2019) "Reverse genetics of rotaviruses: generation of recombinant human rotaviruses from just 11 cDNAs encoding the rotavirus genome" *Uirusu*
87. Komoto, Fukuda, Ide et al. (2018) "Generation of recombinant rotaviruses expressing fluorescent proteins by using an optimized reverse genetics system" *J Virol*
88. Fuerst, Niles, Studier et al. (1986) "Eukaryotic transientexpression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase" *Proc Natl Acad Sci*
89. Abramson, Adler, Dunger et al. (2024) "Addendum: Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
90. Katoh, Misawa, Kuma et al. (2002) "MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform" *Nucleic Acids Res*
91. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
92. Suchard, Lemey, Baele et al. (2018) "Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10" *Virus Evol*
93. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12641267&blobtype=pdf | # Diverse RNA viral effectors converge on facilitation of A GO4 degr adation t o promot e infection
Kaili Xie, Zhongtian Xu, Qingling Qi, Yanjun Li, Xiaodi Hu, Wenkai Yan, Hehong Zhang, Lulu Li, Jianping Chen, Zongtao Sun
## Abstract
ARGONAUTE4 (AGO4)-mediated RNA-directed DNA methylation (RdDM) defends against DNA viruses by methylating their genomes. However, there is limited information available regarding RNA viruses. Here, we show that OsAGO4 has antiviral immunity against two different types of RNA viruses, rice stripe virus (RSV, Tenuivirus ) and Southern rice black streaked dwarf virus (SRBSDV, Fijivirus ). To facilitate infection, the e v olutionarily distinct viral effectors RSV P2 and SRBSDV SP8 both targeted OsAGO4 for degradation. These unrelated viral proteins both recruited the F-box protein OsFBX68 to promote their association with OsAGO4, resulting in enhanced OsAGO4 degradation. In summary, our findings elucidated OsAGO4-mediated antiviral defense and re v eal ne w mechanisms b y which div erse RNA viral effectors e xploit OsAGO4 to promote infection using a common counter-defense strategy.
Gr aphical abstr act
## Introduction
Rice ( Oryza sativa ) is a staple food for more than half of the global population and has been extensively employed as a model system in monocot research. However, rice culti-vation faces significant challenges from viral infections, resulting in substantial losses in crop yield [ 1 , 2 ]. Two evolutionarily diverse viruses affecting rice, rice stripe virus (RSV) and Southern rice black streaked dwarf virus (SRBSDV), are prevalent in the field. RSV ( Tenuivirus oryzaclavatae ; family Phenuiviridae ) has a single-stranded genome of four components and is transmitted by the small brown planthopper (SBPH) ( Laodelphax striatellus ) in a circulative and transovarial manner, causing chlorotic or necrotic stripes on plant leaves [ 3 ]. SRBSDV belongs to the genus Fijivirus (family Spinareoviridae ), and its genome contains 10 segments of doublestranded RNA. It is transmitted by Sogatella furcifera (whitebacked planthopper, WBPH) in a persistent and propagative manner. Plants infected with SRBSDV develop severe stunting and excessive tillering [ 4 , 5 ]. The RSV P2 and SRBSDV P8 proteins have been identified as key virulence factors, implicated in various processes critical to the viral life cycle, including symptom development and suppression of host defense responses [6][7][8][9]. However, the precise molecular pathways and host interactors by which P2/SP8 exerts its pro-viral functions remain incompletely understood.
The RNA interference (RNAi) pathway is a conserved eukaryotic mechanism of sequence-specific gene silencing and a cornerstone of antiviral defense in plants [ 10 , 11 ]. At its heart are Argonaute (AGO) proteins, which use small interfering RNAs (siRNAs) as guides to directly cleave complementary viral RNAs in a process known as slicer activity [12][13][14]. This primary response is often amplified through the synthesis of secondary siRNAs by RNA-dependent RNA polymerases, a process particularly dependent on AGO1 [ 15 , 16 ]. While the role of RNAi as a bona fide antiviral immune pathway in plants has been historically debated, a growing body of evidence confirms its significance [ 17 ]. Key studies have demonstrated that AGO2 plays a specific and essential role in the RNA silencing-mediated antiviral defense mechanism [ 18 , 19 ]. Furthermore, the antiviral capacity of the RNAi machinery extends beyond slicer-dependent cleavage. AGO4 has been shown to mediate translational repression of viral RNAs in a manner critical for immune-receptor-mediated antiviral defense [ 20 ]. Recent work has genetically disentangled the functions of AGO1, revealing its distinct and essential role in antiviral siRNA function, separate from its well-known activity in microRNA (miRNA)-mediated gene regulation [ 21 ].
Only a few AGO proteins have been investigated in the context of rice antiviral immunity. In particular, OsAGO1 and OsAGO18 have been identified as playing essential roles in defense mechanisms. OsAGO18 is induced by rice virus infection and sequesters miRNA168 away from OsAGO1, resulting in the accumulation of AGO1 at the elevated levels required for antiviral defense [ 22 ]. In contrast to the role of AGO2 in the PTGS pathway in Arabidopsis , OsAGO2 appears to epigenetically regulate HEXOKINASE 1 ( OsHXK1 ) expression through DNA methylation. The suppression of Os-HXK1 expression, in turn, triggers reactive oxygen species (ROS)-mediated antiviral defense [ 23 ]. These findings suggest that AGO proteins may play distinct roles in the antiviral processes of different plant species.
To counteract this defense, plant viruses encode viral suppressors of RNA silencing (VSRs) that target distinct nodes of the silencing cascade. Emerging evidence reveals that VSRs exhibit both broad and host-specific strategies. While some VSRs ( Geminivirus AC4) broadly impair AGO1 slicing activity [ 24 ]. Turnip crinkle virus P38 mimics the plant GW/WG proteins that bind to AGO1 to suppress RNA silencing [ 25 ]. The Polerovirus P0 protein, which contains an F-box domain, mediates the degradation of AGO1 by autophagy [ 26 , 27 ]. Notably, the degradation process of AGO1 is conserved, re-quiring its Domain of Unknown Function 1785 (DUF1785) for P0-mediated degradation, a mechanism that also applies to endogenous AGO4 and AGO2 [ 28 ]. Others selectively destabilize specific AGO isoforms-CMV 2b targets AGO4 but not AGO1 [ 29 ]. Notably, previous studies have uncovered that suppressors such as AL2, L2, AC2, C2, and V2, which are also encoded by geminiviruses, interact with AGO4, inhibiting the AGO4-mediated TGS pathway to promote viral infection [30][31][32]. While the targeting of AGO4 by DNA viruses has been well documented, there has been little study into how RNA viruses target AGO4.
In this study, we demonstrate that two distinct viral proteins encoded by unrelated rice viruses, RSV and SRBSDV, convergently target OsAGO4 and effectively suppress OsAGO4mediated antiviral immunity. Despite the evolutionary differences between these viral proteins, they employ a common strategy involving the recruitment of the E3 ubiquitin ligase OsFBX68, leading to the rapid degradation of OsAGO4. This finding reveals antiviral role for OsAGO4 and provides valuable insights into the mechanisms by which diverse viral proteins enhance the degradation of OsAGO4 to improve infection.
## Materials and methods
## Plants materials and growth conditions
The wild-type (WT) rice seeds of Zhonghua 11 (ZH11) ( O. sativa L. cv. japonica ), Nipponbare (Nip) ( O. sativa L. cv. japonica ), and Huaidao No. 5 ( O. sativa L. cv. japonica ) were used in this study. Rice plants infected with RSV or SRB-SDV were maintained in our laboratory. The P2-and SP8expressing lines ( P2-OX and SP8-OX ) have been described in our previous study [ 6 , 33 ]. The OsAGO4a overexpression lines and RNAi osago4ab mutants were generated in the Nip background for this study. The CRISPR/Cas9 system was used to construct the knockout mutants ( osfbx68 and osago4a , os-ago4b ). The rice seedlings were all grown in a greenhouse at 28-30 • C with 14 h light and 10 h dark. Nicotiana benthamiana plants were grown in a chamber at 25 • C with a 16/8 h day/night cycle.
## Insect vector and virus inoculation
RSV and SRBSDV were inoculated using SBPH or WBPH as described previously [ 34 , 35 ] with minor modification. Large numbers of first to second healthy SBPH or WBPH were acquired and then reared on RSV-or SRBSDV-infected rice plants for 4 days, respectively. Then, the nymphs were transferred onto healthy Wuyujing seedlings for 10-12 days to pass through the circulative period. The incidence of SBPH or WBPH was detected by a dot immunobinding assay [ 36 ]. According to the incidence, about three to four insect nymphs were fed on each WT or mutant rice for 3 days, and then the planthoppers were completely removed. Then, the inoculated plants were further grown in the 30 • C greenhouse and examined for symptom development. Reverse transcriptasepolymerase chain reaction (RT-PCR) was used to confirm the disease symptoms of infected plants with SRBSDV or RSV.
## Plasmid construction and plant transformation
To generate overexpressing OsAGO4a transgenic plants, the full-length ORF of OsAGO4a was inserted into the pCAMBIA1300 vector, which includes the cauliflower mosaic virus (CaMV) doubled 35S promoter. The vector was then introduced into Agrobacterium (strain GV3101) by electroporation and transformed into WT rice. We introduced an RNAi structure into Nip plants to gain the construct for RNAi of the OsAGO4a . The RNAi-trigger region was amplified from OsAGO4a gene and then digested with KpnI and BamHI or SpelI and SacI, separately. Then the products were cloned into pTCK303 in two orientations. Finally, the recombinant plasmid was transformed into EHA105 strain by electroporation and then transformed into Nip. The osfbx68 knockout mutant plants were generated by CRISPR/Cas9 genome editing technology. The transformation of the knockout mutants was done by BioRun (Wuhan, China). The osago4 knockout mutants were also acquired from the company (BioGle, osag o4a #1: BG110014C03, osag o4a #2: BG110012F09, osag o4b #1: BG103884C03, osag o4b #2: BG103870B12). To produce the plasmid for yeast two-hybrid screens, the full length of P2 was amplified and then inserted into the GAL4 DNA binding domain. For co-immunoprecipitation (Co-IP) assays, the complementary DNA (cDNA) sequence of Os-AGO4a , OsFBX68 , and its truncated mutants ( OsFBX68C , OsFBX68N ), and OsSKP1 were obtained by overlapping PCR and cloned into the pCAMBIA1300 vector, driven by the CaMV 35S promoter with FLAG or MYC tag, respectively. For yeast two-hybrid assays, OsA GO4a , OsA GO4b , OsSKP1 , OsFBX68 , and its truncated variant domain were amplified and cloned into the prey vector pGADT7 or pG-BKT7. For luciferase complementation imaging (LCI) assay, nLuc-P2 and cLuc-OsAGO4a fusion proteins were generated by amplifying and cloning into 35S::nLuc and 35S::cLuc vectors, respectively. The primers used are listed in Supplementary Table S1 .
## Yeast two-hybrid screen and interaction assays
The full-length sequence of RSV P2 or SRBSDV SP8 was fused to the bait vector pGBKT7 to produce BD-P2 or BD-SP8. Os-AGO4a/b , OsFBX68 , OsFBX68C , OsFBX68N , and OsSKP1 were linked to the pGADT7 prey vector or the bait vector pG-BKT7 and then were co-transformed with different combinations into the yeast strain AH109, and the transformants were grown on synthetic dropout leucine and tryptophan medium following incubation at 30 • C for 72 h. The colonies obtained were then transferred to SD/-Leu/-Trp/-Ade/-His to identify binding activity.
## Co-immunoprecipitation assay
The full-length sequence of RSV P2 , SRBSDV SP8 , Os-AGO4a/b , OsFBX68 , OsFBX68C , OsFBX68N , and OsSKP1 were amplified by PCR and inserted into the pCAMBIA1300 vector, driven by the CaMV 35S promoter with FLAG, MYC, or GFP tag. Different combinations were transiently expressed in N. benthamiana leaves for 48 h after agroinfiltration, GFP-FLAG or HA-GFP was used as a negative control. Total proteins were extracted by IP lysis buffer (Thermo Scientific, Cat. no. 87788) with 1 × protease inhibitor cocktail (Roche, China) from N. benthamiana leaves. Protein extracts were incubated with Pierce TM anti-c-Myc magnetic beads (Thermo Scientific, USA) or anti-FLAG M2 beads (Sigma-Aldrich, USA) for 2 h at 4 • C. The beads were washed five times with 1 × phosphate buffered saline (PBS) before incubating with protein. The immunoprecipitates were then washed five times with ice-cold 1 × PBS at 4 • C. IP samples were boiled at 95 • C for 10 min, then analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted using anti-MYC (1:5000, TransGen, HT101-01), anti-FLAG (1:5000, TransGen, HT201-01), or anti-GFP (1:5000, Genscript, HT801-01) antibodies.
## Luciferase complementation imaging assay
For LCI assays, the tested combinations (nLuc-P2/cLuc-OsA GO4a, nLuc-P2/cLuc-GUS, cLuc-OsA GO4a/nLuc-GUS, cLuc-GUS/nLuc-GUS) were co-infiltrated into the leaves of N. benthamiana. The leaf samples were detached at 48 hpi, sprayed with 1 mM luciferin, then observed with a low-light cooled CCD imaging apparatus (Lumazone Pylon 2048 B, Roper Scientific). The pictures were taken 15 min after exposure.
## Quantitative RT-PCR
Total RNA was extracted from N. benthamiana or rice leaves by Trizol reagent (Invitrogen, C A, US A). cDNA was synthesized using 2 μg RNA with the fast quant RT kit (Tiangen, Beijing, China). Quantitative real-time PCR was performed using the ChamQTM SYBR qPCR Master Mix on the ABI7900HT Sequence Detection System machine (Applied Biosystems, CA, USA) according to the manufacturer's instructions. The Os-UBQ5 gene (AK061988) was used as a control. Relative transcript levels were analyzed by the 2 -CT method. Each data set was repeated at least three times. The primers used to detect transcripts are shown in Supplementary Table S1 .
## Immunoprecipitation and immunoblotting
Total protein sample extracts were extracted with SDS lysis buffer (10% SDS, 100 mM Tris-HCl, pH 6.8) with 2% βmercaptoethanol added before using. After resolving in 10%-12% polyacrylamide gels and transferring onto the PVDF membrane, RSV CP or SRBSDV P10 antibodies at 1/5000 were used to detect the respective virus proteins. We used Os-AGO4a antibody (1:5000, Abclonal, A20331) to detect the expression of endogenous OsAGO4a. After incubation with HRP-labeled rabbit secondary antibody (YEASEN, China), the membrane was detected with the ECL kit (Pierce, Rockford, USA).
## In vivo and semi-in vivo protein degradation assays
In vivo and semiin vivo protein degradation assays were performed according to the established method with slight modifications [ 37 , 38 ]. For the in vivo protein degradation assay, N. benthamiana leaves were co-infiltrated with agrobacterial strains harboring the indicated combinations of constructs: 35S::P2-GFP, 35S::OsAGO4a-MYC, 35S::OsFBX68-FLAG, and 35S::HA-GFP (as a negative control). At 24 h post-infiltration (hpi), a protein synthesis inhibitor, cycloheximide (CHX, 300 μM), was applied to all infiltrated leaf areas. Leaf samples were then collected at the designated time points for subsequent immunoblot analysis. For semiin vivo protein degradation assay, the total protein was extracted from WT or osfbx68 mutant plants by native lysis buffer (40 mM Tris-HCl, pH 7.5, 300 mM NaCl, 5 mM MgCl 2 , 0.1% Triton-100, 1% Glycerol, 5 mM Dithiothreitol). The purified proteins (GST, GST -P2, or GST -P2 + MG132) were incubated with the supernatants at 28 • C for different intervals as indicated. The reactions were stopped with 5 × SDS protein loading buffer and then boiled for 5 min. The samples were analyzed by immunoblotting using anti-OsAGO4a antibody (1:5000, Abclonal, A20331), anti-GST (1:5000, Genscript, A00130), or anti-Actin (1:5000, Abbkine, ABM40122). Ponceau S staining was used as the loading control.
## Ubiquitination assays
The total proteins were extracted from WT or P2 -expressing transgenic plants ( P2-OX ) with IP buffer. OsAGO4 proteins were then immunoprecipitated using anti-OsAGO4a antibody followed by incubation with protein A&G Mag-Beads (GenScript, L00277). The reaction was terminated by adding 5 × protein loading buffer, followed by denaturation at 100 • C for 5 min. The denatured proteins were resolved by SDS-PAGE and subsequently subjected to immunoblotting. The membranes were probed with anti-OsAGO4a antibody (1:5000, Abclonal, A20331), anti-Ub (1:3000, Ab-camSanta Cruz Biotechnology, P4D1-sc-8017), anti-K48-Ub (1:2000, Sigma-Aldrich, ZRB2150), and anti-FLAG (1:5000, Genscript, HT801-01) antibodies.
## RNA sequencing analysis
Total RNA was extracted from mock or RSV-infected plants using Trizol reagent (Invitrogen, C A, US A). Three biological replicates of each sample were used for sequencing. The transcriptome libraries were constructed and analyzed by Hangzhou Lianchuan. The libraries were sequenced on the Illumina HiSeq 2000 instrument. The sequence reads were aligned to the rice genome (MSU Rice Genome Annotation Project database version 7.0) by Bowtie software. Blast2go software was used for the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. The difference in gene expression was considered significant when the absolute value of log 2 (fold change) ratio was ≥ 1 and P ≤ .05.
## Small RNA sequencing and data analysis
In a regular immunoprecipitation reaction, 1 ml of rice extract was immunoprecipitated by incubation with 3 μl of Os-AGO4a antibody at 4 • C for 2 h and then incubated with 20 μl of protein A/G MagBeads (Genscript, L00277) at 4 • C for 2 h. Immunoprecipitates were washed three times (5 min each) with extraction buffer. The quality of purification was examined by SDS-PAGE followed by western blotting. Small RNAs (sRNAs) were isolated from plant extracts and the purified complexes by Trizol reagent (Invitrogen); the eluted and purified sRNAs were then sequenced commercially on an Illumina Hiseq2500 at the LC-BIO (Hangzhou, China). Raw reads were processed with Cutadapt v2.10 to remove adapter sequences and low-quality bases (Phred score < 20). Clean reads ranging from 18 to 30 nt in length were aligned to both the O. sativa reference genome (MSU7) and the RSV genome using Bowtie v1.0, permitting up to one mismatch. Viral-derived small RNAs (vsiRNAs) were identified as those exhibiting perfect or near-perfect matches to the viral genome. For all samples, sRNA size distribution and 5 nucleotide bias were analyzed using in-house Python scripts. To facilitate crosssample comparisons, sRNA abundance was normalized to the total library size and reported as reads per million (RPM). OsAGO4-bound sRNA enrichment was quantified by comparing the RPM values of OsAGO4-IP samples against corresponding IgG-IP controls. Differential enrichment between IP samples was assessed using Fisher's exact test with Benjamini-Hochberg false discovery rate (FDR) correction; sRNAs exhibiting an FDR-adjusted P < .05 and a fold-change ≥2 were classified as significantly enriched in OsAGO4.
## Transient expression assay in rice protoplasts
The indicated constructs (OsAGO4b-GFP, P2-MYC, GUS-MYC) were used for protoplast transfection following the protocol. Briefly, rice protoplasts were isolated from 30-dayold rice and transfected as described previously [ 39 ]. For Co-IP assay, 50 μl of plasmid DNA (2 μg/ μl) was mixed with 1 ml of rice cells for 16 h. Proteins were isolated and subjected to immunoprecipitation and immunoblotting analysis.
## Results
## OsAGO4 interacts with RSV P2
We have recently shown that both the RSV P2 and SRB-SDV P8 proteins target several common host antiviral factors to promote viral infection [ 6 , 33 ], despite being unrelated at the primary sequence level ( Supplementary Fig. S1 ).
To further elucidate the shared targets involved in viral pathogenesis, we used the RSV P2 protein as bait to screen a rice cDNA library. A clone encoding a fragment of OsAGO4a was shown to interact with the P2 protein. We then cloned the full length of OsAGO4a into the prey vector (AD-OsAGO4a) or the bait vector (BD-OsAGO4a) and found that the fulllength protein of OsAGO4a interacted with the P2 protein in yeast (Fig. 1 A). We also cloned OsAGO4b and established that it interacts with P2 in both yeast and plant contexts ( Supplementary Fig. S2 ). Given the substantial (80%) similarity between the amino acid sequences of OsAGO4b and OsAGO4a, our subsequent experiments focused primarily on OsAGO4a. To confirm the interaction between Os-AGO4a and P2, we conducted Co-IP and LCI assays. In the LCI assay, a strong luciferase fluorescence signal was visible only with nLuc-P2/cLuc-OsAGO4a co-agroinfiltration, whereas no such fluorescence was detected in any of the control combinations (Fig. 1 B). In the Co-IP assays, OsAGO4a-MYC was coimmunoprecipitated with P2-FLAG but not with the control GFP-FLAG (Fig. 1 C). To assess whether this interaction occurs in rice, we performed Co-IP assays using an OsAGO4a antibody in both P2 -expressing transgenic plants ( P2-OX ) and WT (Nip) plants. The results showed that in rice plant samples, the P2 protein coimmunoprecipitated with OsAGO4a (Fig. 1 D). Furthermore, P2 was also immunoprecipitated from RSV-infected plant samples with the OsAGO4a antibody (Fig. 1 E). These results conclusively demonstrate that OsAGO4a specifically interacts with the P2 protein.
## OsAGO4-mediated antiviral defense against RSV infection
Since OsAGO4 was found to interact with the P2 protein, we investigated the functions of OsAGO4 in RSV infection by inoculating rice plants with knockdown of AGO4ab ( Os-AGO4ab RNAi lines osago4ab#1 and osago4ab#2 ) or overexpressing AGO4a (overexpression lines OsAGO4aOE#1 and OsAGO4aOE#2 ) using SBPH carrying RSV. The relative expression levels of OsAGO4a and OsAGO4b in these mutants were shown in Supplementary Fig. S3 A-D, F, andG. The sRNA sequences incorporated into the RNAi constructs were also presented in Supplementary Fig. S3 E. Disease symptoms were less severe in OsAGO4a -overexpressing plants, whereas the osago4ab RNAi lines were more susceptible to RSV infection than the WT plants (Fig. 1 F andH). RT -qPCR analysis revealed that the viral RNA levels were much lower in the OsAGO4a-overexpressing plants than in the WT plants (Fig. 1 G), while the RSV RNA levels in the osago4ab RNAi lines significantly increased following RSV infection (Fig. 1 I). We also examined the susceptibility of the osago4a and osago4b knockout mutants to RSV infection, and the results were consistent with those of the RNAi mutants, showing a similar disease-sensitive phenotype ( Supplementary Fig. S4 ). Hence, these findings strongly suggest that OsAGO4 plays a pivotal role in bolstering rice resistance against RSV infection.
## P2 enhances the degradation and polyubiquitination of OsAGO4
The P2-OsAGO4 interaction prompted us to investigate the impact of P2 on OsAGO4 function. In our Co-IP assays, we consistently observed a reduction in the protein level of Os-AGO4a in the presence of P2 (Fig. 1 C), indicating that P2 might influence the stability of OsAGO4a protein. We therefore performed in vivo and in vitro degradation assays with OsA GO4a. First, OsA GO4a was tagged with a MYC tag and transiently expressed with P2-GFP or HA-GFP, and the level of OsAGO4a protein was measured with a MYC antibody (Fig. 2 A). The results showed that the protein level of Os-AGO4a was significantly lower in the presence of P2 compared to the HA-GFP control. However, at the RNA accumu- lation level, no significant differences in OsAGO4a expression were detected among the different samples ( Supplementary Fig. S5 A). Moreover, the protein levels of OsAGO4a were noticeably lower in the P2-OX plants than in the WT Nip plants (Fig. 2 B). In contrast, there was no great difference in Os-AGO4a messenger RNA levels between the WT and P2 expression lines ( Supplementary Fig. S5 B). In parallel, we performed a cell-free degradation assay to investigate the impact of P2 on the OsAGO4a protein level. In these assays, the GST and GST-P2 recombinant proteins were incubated with the protein extracts of WT Nip. As shown in Fig. 2 C, the level of the OsAGO4a protein decreased considerably more strongly in the presence of GST-P2 than in the presence of the GST control. Intriguingly, OsAGO4a was stable in the presence of GST-P2 when treated with MG132, a 26S proteasome inhibitor. Further data showed that MG132 inhibits the P2mediated degradation of OsAGO4a in vivo ( Supplementary Fig. S5 C). We also performed the in vivo degradation experiments with protein synthesis inhibitor CHX. The results demonstrate that CHX effectively inhibited the synthesis of OsAGO4a, and CHX treatment did not alter the effects of P2 on the protein stability of OsAGO4a ( Supplementary Fig. S5 D). These results collectively indicated that P2 facilitated the degradation of the OsAGO4a protein via the 26S proteasome pathway. To further test whether P2 influences the polyubiquitination of OsA GO4a, OsA GO4a was immunoprecipitated from WT and P2-OX plant samples using an OsAGO4a antibody. Immunoblot analysis revealed that the ubiquitination level of OsAGO4a was markedly greater in the P2-OX plants than in the WT plants (Fig. 2 D). Furthermore, linkage-specific antibodies revealed that the ubiquitin chains were predominantly K48-linked, the canonical signal for UPS-mediated degradation (Fig. 2 D). These findings strongly suggested that P2 promoted the ubiquitination and degradation of the OsAGO4a protein. Next, we generated the OsAGO4aOE/P2-OX hybrid plants, and its sensitivity to RSV infection was assessed. The transcript levels of OsAGO4 and P2 in OsAGO4aOE / P2-OX hybrid plants were ∼50and 1500-fold higher than in WT ( Supplementary Fig. S5 E andF). However, OsAGO4 protein abundance was significantly lower in OsAGO4aOE / P2-OX hybrid plants compared to OsAGO4aOE plants ( Supplementary Fig. S5 G), suggesting that P2 promotes the degradation of OsAGO4 in rice. Expression of OsAGO4a protein resulted in more resistant symptoms than in WT plants, which was consistent with the result in Fig. 1 F. Like P2-OX plants, the OsAGO4aOE / P2-OX hybrid plants were more sensitive to RSV infection compared with WT and plants overexpressing Os AGO4a (Fig. 2 E-G). Collectively, this evidence implies that OsAGO4-mediated antiviral resistance is subverted by P2 protein, thus promoting the RSV infection.
## P2 recruits OsFBX68 to mediate OsAGO4 degradation
The ability of P2 to facilitate the degradation of OsAGO4 suggests that the P2 protein might recruit an E3 ligase to degrade OsAGO4. To confirm this hypothesis, we used P2 as bait to screen a rice cDNA library and found that OsFBX68, which encodes an F-box domain-containing E3 ubiquitin ligase, interacted with P2 (Fig. 3 A). To further validate their interaction, we examined interaction strength between OsFBX68 and P2 via Co-IP experiments (Fig. 3 B). OsFBX68 interacted with both OsAGO4a and OsAGO4b (Fig. 3 C and Supplementary Fig. S6 A); however, this interaction was not detected in yeast cells (Fig. 3 A). Since both P2 and OsAGO4 interact with OsFBX68 in vivo , we reasoned that P2 may promote the interaction between OsAGO4 and OsFBX68, resulting in OsAGO4 degradation. To explore this possibility, we conducted Co-IP assays to examine the effect of P2 on the interaction between OsAGO4 and OsFBX68. OsAGO4a-MYC (or OsAGO4b-MYC) and OsFBX68-FLAG were co-expressed with HA-GFP or P2-GFP. Compared with the HA-GFP control, P2-GFP markedly increased the interaction between Os-AGO4 and OsFBX68 (Fig. 3 D and Supplementary Fig. S6 B). Subsequently, we sought to determine whether OsFBX68 plays a role in the P2-mediated degradation of OsAGO4. In an in vivo degradation assay, we transiently expressed MYCtagged OsAGO4 along with HA-GFP or with P2-GFP in the presence of FLAG-tagged OsFBX68. While the protein levels of MYC-tagged OsAGO4 slightly decreased in the presence of OsFBX68-FLAG, the presence of both P2-GFP and OsFBX68-FLAG led to a pronounced decrease in the protein level of Os-AGO4 compared to either of them individually (Fig. 3 E and Supplementary Fig. S6 C). RT -qPCR analysis revealed no obvious difference in the transcript level of OsAGO4a among these samples ( Supplementary Fig. S6 D). We performed the in vivo degradation experiments with protein synthesis inhibitor CHX. The results demonstrate that CHX effectively inhibited the synthesis of OsAGO4, and CHX treatment did not alter the effects of P2 and OsFBX68 on the protein stability of Os-AGO4 ( Supplementary Fig. S6 E). Taken together, these results suggest that P2 enhances the degradation of OsAGO4 by promoting its interaction with OsFBX68.
To further investigate whether OsFBX68 is required for P2-mediated OsAGO4a degradation in rice, we generated osfbx68 -knockout mutant lines by using the CRISPR/Cas9 genome editing approach ( Supplementary Fig. S6 F). We compared OsAGO4a protein levels in osfbx68 mutant plants and WT plants. Immunoblot analysis indicated that OsAGO4a was significantly more abundant in the osfbx68 mutant plants than in the WT plants (Fig. 3 F). To further assess the role of OsFBX68 in rice resistance to RSV infection, we used viruliferous planthoppers to inoculate osfbx68 knockout mu-tant plants with RSV and observed viral symptoms 20 days later. The disease symptoms of the osfbx68 mutant plants were milder than those of the WT plants (Fig. 3 G), and there was significantly less RSV CP RNA in the leaves of the osfbx68 mutant plants than in those of infected WT plants (Fig. 3 H). Additionally, the RSV CP protein level was also significantly lower in the osfbx68 mutant plants than in the WT plants (Fig. 3 I). These results collectively demonstrated that OsFBX68 plays a negative role in rice antiviral defense against RSV infection by mediating OsAGO4 degradation.
We compared the effect of the P2 protein on the degradation rate of OsAGO4a in the WT plants and osfbx68 knockout mutant plants. As shown in Fig. 4 A, the presence of the P2 protein led to a noticeable decrease in the OsAGO4a protein level in the WT plants, while no obvious difference in the degradation rate of the OsAGO4a protein was observed in the osfbx68 mutants. In parallel, we crossed osfbx68 knockout mutants with P2-OX expression plants and obtained homozygous P2-OX/osfbx68 hybrids ( Supplementary Fig. S6 G andH). Intriguingly, OsAGO4a protein accumulation was significantly enhanced in P2-OX/osfbx68 plants compared to P2-OX single transgenic lines. (Fig. 4 B). Similar to the results of osfbx68 mutant plants, the hybrid plants had milder symptoms of RSV, with decreased amounts of RSV RNA and CP protein accumulation (Fig. 4 C-E). Collectively, these data suggested that OsFBX68 is necessary for P2-mediated OsAGO4 degradation. To determine whether the OsFBX68 targets substrates other than OsAGO4, hybrid mutants combining os-ago4ab and osfbx68 alleles were generated and subjected to viral inoculation assays. The results demonstrated that disease resistance was largely dependent on the presence of OsAGO4 ( Supplementary Fig. S7 ), suggesting that OsAGO4 is the major substrate of the OsFBX68-mediated Skp1-Cul1-F-box (SCF) pathway in this context.
## The C-terminus of OsFBX68 is responsible for the interaction between OsFBX68 and OsAGO4a
Having shown that the P2 protein promotes the interaction between OsFBX68 and OsAGO4a, facilitating the rapid degradation of OsAGO4a, we next investigated the mechanism underlying this interaction. OsFBX68 was divided into its N-terminal F-box domain (OsFBX68N) and the Cterminal phloem protein 2 (PP2) domain (OsFBX68C) for subsequent assays ( Supplementary Fig. S8 A). In the Y2H assay, there was a strong self-interaction between OsFBX68 itself and between OsFBX68N and OsFBX68C ( Supplementary Fig. S8 B). A Co-IP assay confirmed the interaction between OsFBX68N and OsFBX68C ( Supplementary Fig. S8 C). Y2H and Co-IP experiments to assess the interaction between Os-AGO4a and OsFBX68N or OsFBX68C showed that Os-AGO4a preferentially binds to the C-terminal PP2 domain but not the N-terminal F-box domain of OsFBX68. Moreover, the interaction between OsAGO4a and OsFBX68C was notably stronger than that between OsAGO4a and full-length OsFBX68 ( Supplementary Fig. S8 D andE). This led us to speculate that OsFBX68N might interfere with the interaction between OsAGO4a and OsFBX68C, and a Co-IP assay confirmed that OsFBX68N did indeed reduce the interaction between OsAGO4a and OsFBX68C ( Supplementary Fig. S8 F).
Considering that P2 enhances the association between OsAGO4a and OsFBX68, we analyzed the interaction between RSV P2 and different OsFBX68 truncation mutants. As shown in Supplementary Fig. S8 G, P2 strongly interacted with OsFBX68N but not with OsFBX68C in yeast cells. Based on the above results, we reasoned that P2 might disrupt the inhibitory effect of OsFBX68N on the interaction between Os-AGO4a and OsFBX68C. To probe this further, a Co-IP assay was employed, and the results showed that when GST-P2 was added, the amount of OsFBX68C-FLAG immunoprecipitated by OsAGO4-MYC significantly increased in the presence of GST-P2 compared to that in the presence of the control GST ( Supplementary Fig. S8 H). Overall, these results suggested that P2 interacts with OsFBX68N and disrupts the inhibitory effect of OsFBX68N on the interaction between Os-AGO4a and OsFBX68C, ultimately enhancing the association between OsAGO4a and OsFBX68 ( Supplementary Fig. S8 I).
In addition to targeting various substrates, F-box proteins typically associate with the adaptor protein SKP1 and the scaffold protein Cul1 to form SCF complexes [ 40 ]. Y2H and Co-IP assays confirmed that OsFBX68 interacts with OsSKP1 ( Supplementary Fig. S9 A andB). We next tested whether P2 affects the interaction between OsFBX68 and OsSKP1. The Co-IP results indicated that compared with the control HA-GFP, P2-GFP had no significant effect on the interaction between OsFBX68 and OsSKP1 ( Supplementary Fig. S9 C). In conclusion, these results collectively demonstrated that the ability of P2 to promote the rapid degradation of OsAGO4a through OsFBX68 mainly stems from the enhanced interaction between OsFBX68 and OsAGO4a rather than affecting the interaction between OsFBX68 and OsSKP1.
## The molecular basis of OsAGO4-mediated viral resistance
Given that AGO4 is a core component of the RNAi pathway [ 13 ], we next evaluated its potential role in antiviral defense through this mechanism. The endogenous OsAGO4 protein was immunoprecipitated from mock or RSV-infected WT plant samples by using a specific antibody, with IgG serving as a negative control. We then sequenced sRNA libraries and profiled sRNA populations in mock or RSV-infected plants.
In rice genome-aligned samples, OsAGO4 primarily associated with 24-nt siRNAs relative to the IgG control (Fig. 5 A). When aligned to the RSV genome, OsAGO4 immunoprecipi- tates did not show enrichment of viral-derived siRNAs compared with the IgG control (Fig. 5 B). Therefore, we speculate that its antiviral function is not primarily mediated through the binding of RSV-derived vsiRNAs. As OsAGO1 and Os-AGO18 are reportedly required for antiviral activity in rice [ 22 ], we also investigated whether OsAGO4-mediated antiviral defense involves either OsAGO1 or OsAGO18. RT-qPCR revealed that the relative expression levels of OsAGO1a , Os-A GO1b , and OsA GO18 increased after RSV infection in WT rice plants. Intriguingly, neither the knockdown nor the overexpression of OsAGO4a had an obvious effect on those expression levels. Thus, the induction of OsA GO1a , OsA GO1b , and OsAGO18 by RSV infection is independent of OsAGO4 (Fig. 5 C). Together, these results suggested that OsAGO4 was not directly associated with vsiRNAs to mediate antiviral defense.
To further explore the broader mechanisms by which Os-AGO4 regulates rice immunity to viruses, we conducted RNAseq and comparative transcriptome profiling in WT and Os-AGO4aOE plants, both with and without RSV infection. Our analysis uncovered a substantial number of differentially expressed genes potentially regulated by OsAGO4. KEGG enrichment analysis showed significant representation of genes associated with defense responses and the ROS pathway ( Supplementary Fig. S10 ). Collectively, these findings indicate that OsAGO4 confers resistance to RSV infection by positively regulating the ROS pathway.
AGO proteins are highly conserved in plants and serve as core components of RNA silencing pathways. They typically consist of four characteristic domains: a variable Nterminal domain and three conserved C-terminal domains (P AZ, MID , and PIWI). The PAZ domain anchors the 3 end of sRNAs and facilitates strand separation, while the PIWI domain, which shares structural homology with RNase H, cleaves complementary target RNAs using a catalytic DDX triad (Asp-Asp-His/Asp). The amino acid sequence of AGO4 was found to possess all four characteristic domains and the conserved DDH catalytic motif [ 22 , 41 ]. To investigate whether the sRNA binding and slicing activity of OsAGO4 is essential for conferring antiviral defense, we generated two OsAGO4a mutants ( O sAGO4a Y350A/F351A (YF/AA) and OsAGO4a D639AD722A (DD/AA)) [ 41 ] and confirmed their interactions with both OsFBX68 and P2 by Y2H and Co-IP assays (Fig. 5 D and Supplementary Fig. S11 ). The YF/AA mutant contains alanine substitutions at two conserved residues (Y350 and F351) crucial for sRNA binding [ 41 , 42 ], while the DD/AA mutant is deficient in endonuclease activity [ 43 , 44 ]. These mutant plants were then inoculated with RSV. The results, based on disease symptoms (Fig. 5 E) and the accumulation of viral RSV CP RNA and protein (Fig. 5 F and G), suggest that both sRNA binding and slicing activities of OsAGO4 are essential for its role in defending against RSV infection.
## SRBSDV SP8 recruits OsFBX68 to mediate OsAGO4a degradation
We next investigated whether the strategy of OsAGO4 degradation promoted by viral proteins is convergent among different viruses. Interestingly, we found that OsAGO4a interacted with SP8, encoded by SRBSDV (a double-stranded RNA virus, genus Fijivirus ) (Fig. 6 A andB). Western blot results showed that the OsAGO4a protein level decreased dramatically when MYC-tagged OsAGO4a was co-expressed with SP8-FLAG, but not the control GFP-FLAG, in N. benthamiana leaves (Fig. 6 C). Notably, the level of the OsAGO4a protein in the SP8 expression plants ( SP8-OX ) was significantly lower than that in the WT plants (Fig. 6 D). Y2H and Co-IP assays revealed that OsFBX68 specifically interacts with SP8 (Fig. 6 E andF). Moreover, SP8 enhanced the interaction between OsFBX68 and OsAGO4a (Fig. 6 G). These findings indicated that SP8 recruits OsFBX68 and promotes the degradation of OsAGO4a by enhancing the interaction between OsFBX68 and OsAGO4a.
We further explored the role of OsAGO4a in SRBSDV infection. We used WBPH carrying SRBSDV to inoculate Os-AGO4a transgenic plants and observed viral symptoms at 45 days post inoculation. As shown in Fig. 7 A and C, the disease symptoms of the osago4ab RNAi mutant plants were notably more severe than those of the WT plants. Conversely, transgenic plants overexpressing OsAGO4a displayed greater resistance to SRBSDV infection. Consistent with the viral symptoms, the RNA levels of SRBSDV genes ( S1 , S2 , and S5 ) were significantly greater in the osago4ab RNAi mutant plants than in the WT plants; in contrast, the RNA levels of the virus were lower in the OsAGO4a -overexpressing transgenic plants (Fig. 7 B andD). Taken together, these observations suggested that OsAGO4a provides antiviral defense against not only the single-stranded RNA virus RSV but also the dsRNA virus SRBSDV in rice.
The role of OsFBX68 in SRBSDV infection was also investigated. We inoculated the osfbx68 knockout mutant plants with SRBSDV and observed viral symptoms at 45 days post inoculation. Remarkably, the disease symptoms of the osfbx68 mutant plants were milder than those of the WT plants following SRBSDV inoculation (Fig. 7 E), and there was also decreased accumulation of SRBSDV genes ( S1 , S2 , and S5 ) (Fig. 7 F). Additionally, the protein level of SRBSDV P10 was significantly decreased in the osfbx68 mutant plants ( Supplementary Fig. S12 ). These data collectively suggested that OsFBX68 plays a negative role in rice resistance against SRBSDV infection by facilitating OsAGO4 degradation.
## Discussion
For a long time, AGO4 has been considered to be involved in transcriptional responses to protect plants against DNA virus infection. AGO4 contributes to defense against DNA viruses by methylating their genomes through RNA-directed DNA methylation (RdDM) pathway [ 45 , 46 ]. Emerging evidence has demonstrated that the RdDM pathway confers antiviral resistance against RNA viruses. The rice RNA polymerase IV, NUCLEAR RNA POLYMERASE D1a (OsNRPD1a), is essential for resistance against Rice grassy stunt virus, as os-nrpd1 knockdown mutants exhibit enhanced susceptibility to infection [ 47 ]. Although AGO4 is a core component of the RdDM pathway, its antiviral function is not always RdDMdependent. For instance, in Arabidopsis , AGO4 confers resistance against biotrophic pathogens like Pseudomonas syringae by modulating resistance gene expression, whereas other RdDM components (RDR2, DCL3, and DRM2) are dispensable for this defense response [ 48 ]. Moreover, AGO4 restricts Plantago asiatica mosaic virus in the cytoplasm rather than the nucleus, indicating a post-transcriptional defense mechanism [ 49 ]. Beyond its role in RdDM, AGO4 also regulates viral RNA translation, a critical process in NB-LRRmediated immunity [ 20 ]. Our findings demonstrated that Os- AGO4 positively regulates rice resistance against diverse RNA viruses, as overexpression of OsAGO4a conferred enhanced resistance to both RSV and SRBSDV, whereas osago4ab RNAi knockdown mutants displayed exacerbated viral susceptibility compared to WT plants (Fig. 1 F-I and 7 A-D), which highlighting the convergent antiviral function of OsAGO4 against RNA viruses in plants.
AGO proteins exhibit distinct binding preferences for sR-NAs [ 50 ]. While AGO1 and AGO2 are well-established mediators of antiviral defense through vsiRNA binding in Arabidopsis [ 51 , 52 ], our study reveals a different mechanism for OsAGO4 in rice. Notably, our analysis revealed that Os-AGO4a exhibited no specific binding affinity for vsiRNAs (Fig. 5 B), suggesting its antiviral activity is independent of vsiRNA-guided viral RNA targeting. Research on the interaction between plant viruses and lmiRNAs/miRNAs has revealed a complex molecular dialogue between viral infection and host defense. Viruses promote infection by interfering with host miRNA biosynthesis and hijacking developmental networks, while plants utilize miRNAs for direct antiviral defense and even cross-kingdom regulation of insect vectors [53][54][55][56]. We propose that the AGO4-bound 24-nt siR-NAs confer antiviral immunity through a dual mechanism. While DNA methylation-mediated transcriptional silencing is a likely pathway, the potential involvement of lmiRNAs suggests an additional post-transcriptional regulatory layer. Elu-cidating the contribution of each mechanism will be a critical focus of subsequent research.
While the roles of plant AGO proteins in antiviral immunity are well established, emerging evidence reveals that viruses have evolved sophisticated counter-defense strategies by targeting AGOs at the post-translational level. For instance, Polerovirus P0 disrupts AGO1 function by blocking its incorporation into the RISC complex and promoting its degradation [ 27 ]. Subsequent studies identified a specific AGO1 motif critical for P0-mediated turnover [ 28 ]. Beyond P0, multiple viral proteins exploit the host ubiquitin-proteasome system to regulate AGO stability: PVX P25 targets AGO1 [ 57 ], while rotavirus manipulates AGO2 in mammals [ 58 ]. The mechanisms governing these interactions, however, remain elusive; it is unknown if they necessitate additional host factors or specific post-translational modifications. Furthermore, AGO degradation is not exclusively mediated by the proteasome, implying the involvement of other mechanisms. Polerovirus P0 triggers AGO1 clearance via autophagy [ 59 ], mirroring observations in mammals where miRNA-free AGO2 undergoes selective autophagic degradation [ 60 ]. Notably, while DNA viruses are known to modulate AGO4 activity, RNA viruses also subvert this pathway. For example, the CMV 2b protein directly binds AGO4, inhibiting its slicing activity to facilitate viral replication [ 29 ]. These findings collectively highlight a dynamic arms race between host AGO-mediated immu- nity and viral suppression mechanisms. Whether RNA viruses universally target AGO4 via similar mechanisms requires further investigation. Our yeast two-hybrid data, which demonstrate a direct P2-OsAGO4 interaction in a system devoid of functional RNAi machinery (Fig. 1 A), suggest that P2 can engage OsAGO4 independently of its RNA-loading state. This mechanism would be analogous to the Polerovirus P0 protein, which specifically targets unloaded AGO1 [ 26 , 61 , 62 ]. This interpretation is strongly supported by our genetic and biochemical evidence. We found that the OsAGO4a Y350AF351A mutant retains its interaction with both P2 and OsFBX68 ( Supplementary Fig. S11 C-E). This finding indicates that the degradation complex recognizes unloaded OsAGO4. This conclusion aligns with established structural models wherein the N-terminal coil (N-coil) of AGO acts as a conformational switch [ 12 ]. While sRNA binding is primarily mediated by the PAZ and MID domains [ 41 ], the accessibility of the N-coil reports on the protein's loading status.
In our study, OsAGO4-mediated resistance was unaffected by vsiRNA binding, suggesting RNA viruses may employ alternative suppression strategies. Here, we found a novel viral counter-defense strategy in which RNA viruses exploit the host ubiquitin-proteasome system to suppress OsAGO4mediated antiviral immunity. We demonstrated that RSV P2 and SRBSDV SP8 interact with OsAGO4 and promote its ubiquitination and subsequent degradation via the 26S proteasome pathway (Figs 1 , 2 , and 6 A-D). Importantly, we determined that OsFBX68 is the principal E3 ubiquitin ligase regulating OsAGO4 turnover through ubiquitin conjugation, and viral proteins further potentiate its degradation (Figs 3 A-E and 6 E-G). Consistent with this mechanism, osfbx68 knockout plants exhibited elevated OsAGO4a levels and enhanced resistance to RSV (Fig. 3 F-I), while OsAGO4a -overexpressing plants showed antiviral immunity against SRBSDV (Fig. 7 C and D). These findings establish a model in which divergent viral proteins (P2 and SP8) independently recruit the same host E3 ligase, OsFBX68, to destabilize OsAGO4 and facilitate viral infection. In WT plants, this targeted degradation weakens antiviral defense, whereas in osfbx68 mutants, OsAGO4 remains stable and maintains its protective function (Fig. 7 G). Our work uncovers a unique regulatory nexus between viral pathogens, an AGO protein, and a host E3 ligase, highlighting how viruses co-opt endogenous protein turnover machinery to evade plant immunity.
## A c kno wledg ements
We thank Prof. Shouwei Ding (University of California, Riverside) and Prof. Peter Moffett (Université de Sherbrooke) for their important suggestions; Prof. Mike Adams for critically reading and improving the manuscript.
## References
1. "Zhang (Formal analysis"
2. Wang, Ji, Kou (2025) "Occurrence and integrated control of major rice diseases in China" *New Plant Protection*
3. Wu, Yang, Zhao (2022) "Current rice production is highly vulnerable to insect-borne viral diseases" *Natl Sci Rev*
4. Xu, Fu, Tao (2021) "Rice stripe virus: exploring molecular weapons in the arsenal of a negative-sense RNA virus" *Annu Rev Phytopathol*
5. Wei, Li (2016) "Rice reoviruses in insect vectors" *Annu Rev Phytopathol*
6. Zhou, Xu, Xu (2013) "Southern rice black-streaked dwarf virus: a white-backed planthopper-transmitted fijivirus threatening rice production in Asia" *Front Microbiol*
7. Li, Zhang, Chen (2021) "A class of independently evolved transcriptional repressors in plant RNA viruses facilitates viral infection and vector feeding" *Proc Natl Acad Sci*
8. Yang, Du, Tan (2024) "Histone deacetylase OsHDA706 orchestrates rice broad-spectrum antiviral immunity and is impeded by a viral effector" *Cell Rep*
9. Yang, Wu, Xu (2025) "Arbovirus suppression of a lectin protein-mediated broad-spectrum resistance enhances herbivorous vector performance and viral transmission" *Nat Commun*
10. Zhang, Wang, Song (2023) "Different viral effectors suppress hormone-mediated antiviral immunity of rice coordinated by OsNPR1" *Nat Commun*
11. Ding (2010) "RNA-based antiviral immunity" *Nat Rev Immunol*
12. Pumplin, Voinnet (2013) "RNA silencing suppression by plant pathogens: defence, counter-defence and counter-counter-defence" *Nat Rev Micro*
13. Bressendorff, Sjøgaard, Prestel (2025) "Importance of an N-terminal structural switch in the distinction between small RNA-bound and free ARGONAUTE" *Nat Struct Mol Biol*
14. Fang, Qi (2016) "RNAi in Plants: an Argonaute-centered view" *Plant Cell*
15. Iwakawa, Tomari (2022) "Life of RISC: formation, action, and degradation of RNA-induced silencing complex" *Mol Cell*
16. Tamotsu, Koizumi, Briones (2023) "Spatial distribution of three ARGONAUTEs regulates the anther phasiRNA pathway" *Nat Commun*
17. Crété (2004) "The action of ARGONAUTE1 in the miRNA pathway and its regulation by the miRNA pathway are crucial for plant development" *Genes Dev*
18. Yang, Li (2018) "Dissection of RNAi-based antiviral immunity in plants" *Curr Opin Virol*
19. Jaubert, Mello, Perry (2011) "ARGONAUTE2 mediates RNA-silencing antiviral defenses against Potato virus X in Arabidopsis" *Plant Physiol*
20. Scholthof, Alvarado, Vega-Arreguin (2011) "Identification of an ARGONAUTE for antiviral RNA silencing in Nicotiana benthamiana" *Plant Physiol*
21. Bhattacharjee, Zamora, Azhar (2009) "Virus resistance induced by NB-LRR proteins involves Argonaute4-dependent translational control" *Plant J*
22. Clavel, Lechner, Incarbone (2021) "Atypical molecular features of RNA silencing against the phloem-restricted Polerovirus TuYV" *Nucleic Acids Res*
23. Wu, Yang, Wang (2015) "Viral-inducible Argonaute18 confers broad-spectrum virus resistance in rice by sequestering a host microRNA" *eLife*
24. Wang, Chen, Sun (2021) "ARGONAUTE 2 increases rice susceptibility to rice black-streaked dwarf virus infection by epigenetically regulating HEXOKINASE 1 expression" *Mol Plant Pathol*
25. Rosas-Diaz, Zhang (2018) "A virus-targeted plant receptor-like kinase promotes cell-to-cell spread of RNAi" *Proc Natl Acad Sci*
26. Schott, Himber (2012) "Differential effects of viral silencing suppressors on siRNA and miRNA loading support the existence of two distinct cellular pools of ARGONAUTE1" *EMBO J*
27. Bortolamiol, Pazhouhandeh, Marrocco (2007) "The Polerovirus F box protein P0 targets ARGONAUTE1 to suppress RNA silencing" *Curr Biol*
28. Csorba, Lózsa, Hutvágner (2010) "Polerovirus protein P0 prevents the assembly of small RNA-containing RISC complexes and leads to degradation of ARGONAUTE1" *Plant J*
29. Derrien, Clavel, Baumberger (2018) "A suppressor screen for AGO1 degradation by the viral F-box P0 protein uncovers a role for AGO DUF1785 in sRNA duplex unwinding" *Plant Cell*
30. Duan, Fang, Zhou (2012) "Suppression of Arabidopsis ARGONAUTE1-mediated slicing, transgene-induced RNA silencing, and DNA methylation by distinct domains of the cucumber mosaic virus 2b protein" *Plant Cell*
31. Buchmann, Asad, Wolf (2009) "Geminivirus AL2 and L2 proteins suppress transcriptional gene silencing and cause genome-wide reductions in cytosine methylation" *J Virol*
32. Soitamo, Jada, Lehto (2012) "Expression of geminiviral AC2 RNA silencing suppressor changes sugar and jasmonate responsive gene expression in transgenic tobacco plants" *BMC Plant Biol*
33. Zhang, Chen, Huang (2011) "BSCTV C2 attenuates the degradation of SAMDC1 to suppress DNA methylation-mediated gene silencing in Arabidopsis" *Plant Cell*
34. Zhang, Li, He (2020) "Distinct modes of manipulation of rice auxin response factor OsARF17 by different plant RNA viruses for infection" *Proc Natl Acad Sci*
35. He, Guan, He (2017) "Evolution of rice stripe virus" *Mol Phylogenet Evol*
36. Zhou, Wu, Wang (2010) "Preliminary report on the transmission of rice black-streaked dwarf virus from frozen infected leaves to rice by insect vector small brown planthopper ( Laodelphax striatellus )" *Chinese J Rice Sci*
37. Wu, Ni, Liu (2013) "Development and use of three monoclonal antibodies for the detection of rice black-streaked dwarf virus in field plants and planthopper vectors" *Virol J*
38. Huang, Yang, Sun (2025) "Perception of viral infections and initiation of antiviral defence in rice" *Nature*
39. Liu, Zhang, Tang (2010) "An efficient system to detect protein ubiquitination by agroinfiltration in Nicotiana benthamiana" *Plant J*
40. Zhang, Su, Duan (2011) "A highly efficient rice green tissue protoplast system for transient gene expression and studying light/chloroplast-related processes" *Plant Methods*
41. Ban (2021) "CUL3 E3 ligases in plant development and environmental response" *Nat Plants*
42. Iki (2012) "Cytoplasmic assembly and selective nuclear import of Arabidopsis ARGONAUTE4/siRNA Complexes" *Mol Cell*
43. Guang, Bochner, Pavelec (2008) "An Argonaute transports siRNAs from the cytoplasm to the nucleus" *Science*
44. Song, Smith, Hannon (2004) "Crystal structure of Argonaute and its implications for RISC slicer activity" *Science*
45. Wee, Flores-Jasso (2012) "Argonaute divides its RNA guide into domains with distinct functions and RNA-binding properties" *Cell*
46. Raja, Jackel, Li (2014) "Arabidopsis double-stranded RNA binding protein DRB3 participates in methylation-mediated defense against geminiviruses" *J Virol*
47. Raja, Sanville, Buchmann (2008) "Viral genome methylation as an epigenetic defense against geminiviruses" *J Virol*
48. Zhang, Wei, Xu (2020) "A bunyavirus-inducible ubiquitin ligase targets RNA polymerase IV for degradation during viral pathogenesis in rice" *Mol Plant*
49. Agorio, Vera (2007) "ARGONAUTE4 is required for resistance to Pseudomonas syringae in Arabidopsis" *Plant Cell*
50. Brosseau, Oirdi, Adurogbangba (2016) "Antiviral defense involves AGO4 in an Arabidopsis -Potexvirus interaction" *Mol Plant Microbe Interact*
51. Meister (2013) "Argonaute proteins: functional insights and emerging roles" *Nat Rev Genet*
52. Qu, Ye, Morris (2008) "Arabidopsis DRB4, A GO1, A GO7 and RDR6 participate in a DCL4-initiated antiviral RNA silencing pathway" *Proc Natl Acad Sci*
53. Wang, Jovel, Udomporn (2011) "The 21-nucleotide, but not 22-nucleotide, viral secondary small interfering RNAs direct potent antiviral defense by two cooperative Argonautes in Arabidopsis thaliana" *Plant Cell*
54. Cai, Qiao, Wang (2018) "Plants send small RNAs in extracellular vesicles to fungal pathogen to silence virulence genes" *Science*
55. Dinesh-Kumar, Zhang, Liu (2017) "Cucumber mosaic virus coat protein modulates the accumulation of 2b protein and antiviral silencing that causes symptom recovery in planta" *PLoS Pathog*
56. Wang, Wu, Zhang (2025) "Spatial regulation of rice leaf morphology by miRNA-target complexes during viral infection" *Plant Cell Environ*
57. Zou, Zhang, Chen (2025) "A plant bunyaviral protein disrupts SERRATE phase separation to modulate microRNA biogenesis during viral pathogenesis" *Nat Commun*
58. Chiu, Chen, Baulcombe (2010) "The silencing suppressor P25 of Potato virus X interacts with Argonaute1 and mediates its degradation through the proteasome pathway" *Mol Plant Pathol*
59. Mukhopadhyay, Chanda, Patra (2019) "Biphasic regulation of RNA interference during rotavirus infection by modulation of Argonaute2" *Cell Microbiol*
60. Derrien, Baumberger, Schepetilnikov (2012) "Degradation of the antiviral component ARGONAUTE1 by the autophagy pathway" *Proc Natl Acad Sci*
61. Gibbings, Mostowy, Jay (2012) "Selective autophagy degrades DICER and AGO2 and regulates miRNA activity" *Nat Cell Biol*
62. Baumberger, Tsai, Lie (2007) "The Polerovirus silencing suppressor P0 targets ARGONAUTE proteins for degradation" *Curr Biol*
63. Hacquard, Clavel, Baldrich (2022) "The Arabidopsis F-box protein FBW2 targets AGO1 for degradation to prevent spurious loading of illegitimate small RNA" *Cell Rep* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC11856160&blobtype=pdf | # Virology Applications to the COVID-19 Pandemic
Evangelia Kostaki
## Abstract
From the early identification of severe respiratory cases of unknown etiology in Wuhan, China, in late 2019, virology research has played an important role in understanding, management, and prevention of the COVID-19 pandemic. Firstly, virology applications were important in identifying SARS-CoV-2 as the causative agent, classifying the virus, and determining its closest viral relatives through whole-genome sequencing [1][2][3][4][5][6]. Secondly, virology research included critical applications such as (i) diagnostics development-the creation of molecular and antigen tests and the development of serological tests to detect antibodies [7-9], (ii) epidemiology and surveillance-the implementation of wastewater epidemiology for viral monitoring, genomic surveillance to identify and monitor variants of interest (VOIs) and variants of concern (VOCs), tracing the origin of SARS-CoV-2 transmission, investigating the role of social contacts in viral spread, and conducting molecular epidemiological studies to analyze transmission patterns [10][11][12][13][14]. Additionally, virology studies were applied in assessing public health measures, guiding the development of therapeutics and monoclonal antibodies, and enabling rapid vaccine development using different technologies. Furthermore, virology research has contributed to understanding viral pathogenesis and supports a One Health approach to studying zoonotic origins and preventing future pandemics [15,16]. These contributions highlight the essential role of virology in relation to the pandemic.The Special Issue, "Virology Applications to the COVID-19 Pandemic", published in Life, in the section "Epidemiology", includes 14 original research and review articles on applications of virology to the COVID-19 pandemic. Four studies, authored by Chrysostomou and Aristokleous et al. [17], Chaintoutis and Chassalevris et al. [18], Lim et al. [19], and Chrysostomou et al. [20], focus on the development or the evaluation of novel laboratory methods for SARS-CoV-2 detection and characterization. Chrysostomou and Aristokleous et al. introduce a rapid method for the identification of various VOCs using a multiallelic spectral genotyping assay [17]. This method, based on real-time reverse transcription-PCR in combination with probes, offers several advantages versus next-generation sequencing, providing fast and accurate results. Similarly, Chaintoutis and Chassalevris et al. developed a one-step real-time RT-PCR assay to rapidly identify Alpha , Beta, Gamma, or Delta VOCs [18]. This assay employs four locked nucleic acid (LNA) modified TaqMan probes targeting signature mutations in the receptor-binding motif (RBM) of the spike protein's receptor-binding domain (RBD). Validation with known SARS-CoV-2-positive and -negative samples demonstrated its accuracy in characterizing variants. Additionally, the assay can be adapted to detect a broader range of variants.
Seoul, Repulic of Korea) on Maelstrom 9600 (Taiwan Advanced Nanotech Inc., Taoyuan, Taiwan) [19]. Bland-Altman analysis showed high concordance among the platforms, with 95.2% concordance between MagNA Pure 96 and KingFisher Flex and 95.4% between MagNA Pure 96 and Maelstrom 9600, indicating statistically reliable results across all systems. Chrysostomou et al. presented a real-time RT-PCR detection assay designed to address the high genetic polymorphism of SARS-CoV-2 [20]. The assay employs mismatchtolerant molecular beacons targeting the S, E, M, and N genes, enabling the detection of genetically diverse SARS-CoV-2 strains.
In a narrative review by Tofarides et al., the protective effect of vaccination against SARS-CoV-2 and long-COVID-19 was explored [21]. Current evidence indicates that vaccination reduces the risk of long-COVID-19, with the effectiveness influenced by factors such as the number of doses, the specific viral variant, the recency of vaccination, and, probably, age. Additionally, vaccination appears to lower the risk of neurocognitive-psychological disorders and cardiovascular complications. However, the potential role of the influenza vaccine in preventing long-COVID-19 remains unclear. Marot et al. assessed two surrogate neutralization assays to evaluate immune responses against the B.1, Alpha, Beta, and Omicron variants [22]. Their findings revealed the strongest neutralization responses in recovered COVID-19 patients who received a single vaccine dose. Naïve individuals who received two doses of an mRNA vaccine showed high neutralization titers against the B.1, Alpha, and Beta variants, though only 34.3% demonstrated activity against Omicron. On the other hand, non-infected individuals with an incomplete vaccination scheme exhibited weak and inconsistent neutralization activity across all variants. Terpos et al. investigated the kinetics of neutralizing antibodies (NAbs) six months after the second dose of the BNT162b2 mRNA vaccine [23]. At this timepoint, 2.59% of participants had NAb levels below 30%, 11.9% had NAb levels below 50%, and 58% had NAb levels above 75%. Older age was consistently associated with lower NAb levels across all timepoints. Population modeling predicted that 50% of individuals would have NAb levels below 73.8% at nine months and 64.6% at 12 months post-vaccination. The study highlights a sustained decline in humoral immunity six months after full vaccination, offering valuable insights for public health planning. In the study by Fischer et al., humoral and cellular immune responses against SARS-CoV-2 were evaluated in 41 COVID-19 convalescents with a high mean age of 54 ± 8.4 years [24]. Antibodies against SARS-CoV-2 were detectable in 95% of the participants up to 8 months post infection, though their levels showed a declining trend in most participants over time. A specific long-lasting cellular immune response was also observed in these individuals by stimulating immune cells with SARS-CoV-2-specific peptides, targeting the viral spike, membrane, and nucleocapsid proteins, and then measuring the release of interferon-γ (IFN-γ).
In their study, Balaska et al. present a mass screening program for the detection of SARS-CoV-2 by RT-PCR, performed on all professionals in a hospital, irrespective of symptoms [25]. The total number of samples tested was 43,726. The average positivity rate was 1.21% and was similar to the community positivity rate in Greece. Specifically, among the positive participants, 31% experienced no symptoms before receiving the positive result, 46.1% reported close contact with a patient or infected coworkers, and 32.8% reported close contact with infected family members. In periods of high COVID-19 incidence, the periodic testing of health care personnel can also be used to diagnose SARS-CoV-2 infections at the asymptomatic phase. Capozzi and Simone et al. report on an isolate from a 53-yearold woman who remained COVID-19-positive for approximately four months [26]. The viral isolate was assigned to lineage B.1.177.51 and was found to contain a novel set of mutations in the Spike protein (V143D, del144/145, and E484K). Seroneutralization assays revealed a significantly reduced response of this strain to both BNT162b2 Pfizer/BioNTech vaccine-induced antibodies (average reduction of 70%) and convalescent sera (average reduction of 19.04%) compared to VOC P.1. This study underscores the critical role of wholegenome sequencing (WGS) in tracking novel isolates from chronically infected individuals. Detsika et al. describe a cross-sectional study on the epidemiological, laboratory, and clinical characteristics of COVID-19 patients in relation to their immunogenetic profiles [27]. A statistically significant increase in HLA-DRB1*01 was detected in mild COVID-19 patients versus controls. The frequency of A*11, A*23, and DRB1*09 alleles was found to be higher, while the frequency of C*12 was lower, in hospitalized patients versus healthy controls, albeit with uncorrected statistical significance. Bonnet et al. explored whether the Alpha variant was associated with higher viral loads compared to the historical strain in saliva samples from patients with mild to moderate symptoms [28]. While a higher viral load was observed for the Alpha variant, no significant differences in viral load levels were detected between individuals infected with the Alpha variant and those with historical strains when accounting for the time interval between symptom onset and sampling.
In the study by Alshanbari et al., a machine learning approach (ML) was employed to investigate whether the ICU admissions of COVID-19 patients could be predicted [29]. The analysis used clinical and laboratory data from 100 patients diagnosed during May 2020 and January 2021. The study focused on the effectiveness of a weighted radial kernel support vector machine (SVM) coupled with recursive feature elimination (RFE). An initial assessment showed that the SVM with weighted radial kernels coupled with RFE outperformed other classification methods in discriminating between ICU and non-ICU admissions. Implementing RFE with weighted radial kernel SVM identified a significant set of variables that could predict and statistically distinguish ICU from non-ICU COVID-19 patients. These variables included patient weight, PCR Ct value, immune markers (CCL19, INF-β, BLC, INR), prothrombin time (PT), partial thromboplastin time (PTT), cardiac enzymes (CK-MB), blood parameters (HB, platelets, RBC), and biochemical markers such as urea, creatinine, and albumin levels. This study highlights the potential of weighted radial kernel SVM as a valuable ML tool to assist hospital decision-makers in optimizing resource allocation. Poonia et al. reported the use of an enhanced SEIR model designed to predict new cases of COVID-19 [30]. This model incorporates vaccination as an additional compartment, referred to as SEIRV, enabling predictions as to the severity of COVID-19 in vaccinated populations. The model was simulated under three distinct scenarios: (1) without social distancing, (2) with social distancing, and (3) with a combination of social distancing and vaccination. The results indicate an epidemic growth rate of about 0.06 per day, with the number of infected individuals doubling every 10.7 days. When social distancing measures were included, the model estimated a basic reproduction number (R 0 ) of 1.3, highlighting the effectiveness of these interventions in reducing the spread of the virus.
The studies presented in this Special Issue highlight the importance of the application of virology to the COVID-19 pandemic, including diagnostic innovations, immune response characterization, epidemiological modeling, and resource optimization through machine learning.
## References
1. Paraskevis, Kostaki, Magiorkinis et al. (2020) "Full-genome evolutionary analysis of the novel corona virus (2019-nCoV) rejects the hypothesis of emergence as a result of a recent recombination event" *Infect. Genet. Evol*
2. Zhu, Zhang, Wang et al. (2019) "A Novel Coronavirus from Patients with Pneumonia in China" *N. Engl. J. Med*
3. Lu, Zhao, Li et al. (2020) "Genomic characterisation and epidemiology of 2019 novel coronavirus: Implications for virus origins and receptor binding" *Lancet*
4. Wu, Zhao, Yu et al. (2020) "A new coronavirus associated with human respiratory disease in China" *Nature*
5. Chan, Kok, Zhu et al. (2020) "Genomic characterization of the 2019 novel human-pathogenic coronavirus isolated from a patient with atypical pneumonia after visiting Wuhan" *Emerg. Microbes Infect*
6. Zhou, Yang, Wang et al. (2020) "A pneumonia outbreak associated with a new coronavirus of probable bat origin" *Nature*
7. Pfefferle, Reucher, Nörz et al. (2020) "Evaluation of a quantitative RT-PCR assay for the detection of the emerging coronavirus SARS-CoV-2 using a high throughput system" *Eurosurveillance*
8. Poljak, Korva, Knap Gašper et al. (2020) "Clinical Evaluation of the cobas SARS-CoV-2 Test and a Diagnostic Platform Switch during 48 Hours in the Midst of the COVID-19 Pandemic" *J. Clin. Microbiol*
9. Strati, Zavridou, Paraskevis et al. (2022) "Development and Analytical Validation of a One-Step Five-Plex RT-ddPCR Assay for the Quantification of SARS-CoV-2 Transcripts in Clinical Samples" *Anal. Chem*
10. Alygizakis, Markou, Rousis et al. (2021) "Analytical methodologies for the detection of SARS-CoV-2 in wastewater: Protocols and future perspectives" *Trends Analyt Chem*
11. Galani, Aalizadeh, Kostakis et al. (2022) "SARS-CoV-2 wastewater surveillance data can predict hospitalizations and ICU admissions" *Sci. Total Environ*
12. Worobey, Pekar, Larsen et al. (2020) "The emergence of SARS-CoV-2 in Europe and North America" *Science*
13. Markov, Ghafari, Beer et al. (2023) "The evolution of SARS-CoV-2" *Nat. Rev. Microbiol*
14. Boni, Lemey, Jiang et al. "Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic"
15. Keusch, Amuasi, Anderson et al. (2022) "Pandemic origins and a One Health approach to preparedness and prevention: Solutions based on SARS-CoV-2 and other RNA viruses" *Proc. Natl. Acad. Sci*
16. Hemida, Abduallah (2020) "The SARS-CoV-2 outbreak from a one health perspective. One Health"
17. Chrysostomou, Aristokleous, Hezka Rodosthenous et al. (2023) "Detection of Circulating SARS-CoV-2 Variants of Concern (VOCs) Using a Multiallelic Spectral Genotyping Assay" *Life*
18. Chaintoutis, Chassalevris, Balaska et al. (1015) "A Novel Real-Time RT-PCR-Based Methodology for the Preliminary Typing of SARS-CoV-2 Variants, Employing Non-Extendable LNA Oligonucleotides and Three Signature Mutations at the Spike Protein Receptor-Binding Domain" *Life*
19. Lim, Jung, Park et al. "Evaluation of Three Automated Extraction Systems for the Detection of SARS-CoV-2 from Clinical Respiratory Specimens" *Life*
20. Chrysostomou, Hezka Rodosthenous, Topcu et al. (1146) "A Multiallelic Molecular Beacon-Based Real-Time RT-PCR Assay for the Detection of SARS-CoV-2" *Life*
21. Tofarides, Christaki, Milionis et al. (2057) "Effect of Vaccination against SARS-CoV-2 on Long COVID-19: A Narrative Review" *Life*
22. Marot, Bocar Fofana, Flandre et al. (2064) "SARS-CoV-2 Neutralizing Responses in Various Populations, at the Time of SARS-CoV-2 Variant Virus Emergence: Evaluation of Two Surrogate Neutralization Assays in Front of Whole Virus Neutralization Test" *Life*
23. Terpos, Karalis, Ntanasis-Stathopoulos et al. (1077) "Robust Neutralizing Antibody Responses 6 Months Post Vaccination with BNT162b2: A Prospective Study in 308 Healthy Individuals" *Life*
24. Fischer, Lindenkamp, Lichtenberg et al. "Evidence of Long-Lasting Humoral and Cellular Immunity against SARS-CoV-2 Even in Elderly COVID-19 Convalescents Showing a Mild to Moderate Disease Progression" *Life*
25. Balaska, Parasidou, Takardaki et al. (2011) "The Implementation of a Health Care Worker Screening Program Based on the Advanta RT-qPCR Saliva Assay in a Tertiary Care Referral Hospital in Northern Greece" *Life*
26. Capozzi, Simone, Bianco et al. (1259) "Emerging Mutations Potentially Related to SARS-CoV-2 Immune Escape: The Case of a Long-Term Patient" *Life*
27. Detsika, Giatra, Kitsiou et al. (1017) "Demographic, Clinical and Immunogenetic Profiles of a Greek Cohort of COVID-19 Patients" *Life*
28. Bonnet, Masse, Benamar et al. "Is the Alpha Variant of SARS-CoV-2 Associated with a Higher Viral Load than the Historical Strain in Saliva Samples in Patients with Mild to Moderate Symptoms?" *Life*
29. Alshanbari, Mehmood, Sami et al. (1100) "Prediction and Classification of COVID-19 Admissions to Intensive Care Units (ICU) Using Weighted Radial Kernel SVM Coupled with Recursive Feature Elimination (RFE)" *Life*
30. Poonia, Saudagar, Altameem et al. "An Enhanced SEIR Model for Prediction of COVID-19 with Vaccination Effect" *Life*
31. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12737554&blobtype=pdf | # Molecular and Phylogenetic Analyses of Lumpy Skin Disease Virus (LSDV) Outbreak (2021/22) in Pakistan Indicate Involvement of a Clade 1.2 LSDV Strain
Saiba Ferdoos, Andy Haegeman, Sadia Sattar, Ibrar Ahmed, Sundus Javed, Aamira Tariq, Nick De Regge, Nazish Bostan
## Abstract
Livestock is the backbone of the economy in an agricultural country like Pakistan, with cattle serving as a milk and protein source. In 2021/22, Pakistan was hit by the first major outbreak of lumpy skin disease (LSD) in cattle, in all four provinces. LSD is characterized by the development of skin nodules, leading to severe illness, decreased milk production, and mortality, causing huge economic losses. This study aimed to analyze and classify the lumpy skin disease virus (LSDV) strains involved in the outbreak in the Punjab province at the molecular and phylogenetic levels to develop effective control strategies. A combination of different real-time PCRs was used for the identification and differentiation between vaccine, wild-type, and recombinant LSDV strains. This was mented with the sequence determination and phylogenetic analysis of ten genomic loci from two selected isolates from the 2021/22 Pakistan outbreak. The combined data showed that these isolates belonged to LSDV clade 1.2 and were clearly different from the vaccine clade 1.1 (Neethling-like), as well as from the recombinant clade 2 strains. In addition, using a fit-for-purpose gel-based PCR, the isolates from the outbreak were also shown to be different from KSGP0240-based vaccines.
## 1. Introduction
Pakistan is an agricultural country where a vast majority of the population relies on cattle and household animals for milk and meat supply [1]. The lumpy skin disease (LSD) outbreak in 2021-2023 has led to enormous economic losses in terms of reduced milk production, treatment of infected animals, abortions, and even death [2]. In Pakistan, LSD was reported for the first time in Jamshoro, Sindh, in November 2021. In the Sindh province, 20,000 animals were infected, and 54 deaths were recorded. Since its first reporting, the outbreak has widely spread across all four provinces, leading to extensive livestock damage.
In Punjab, 35,000 LSD cases were reported, in Baluchistan, 22,225, in Khyber Pakhtunkhwa (KPK), 74,590 cases, and, in Azad Jammu and Kashmir (AJ and K), 6351 cases were documented. Nationwide, 190,000 cases of LSD were reported in cattle, with 7500 deaths. A proportion of animals recovered from the clinical disease [3,4]. It has put the country in a state of emergency for the control and prevention of LSD in all provinces. LSD is reported to infect cattle and water buffaloes; only a very low prevalence in water buffaloes (9.3%) was reported during the outbreak in Pakistan, potentially due to species differences [5]. An LSDV seroprevalence of 19.38% (n = 800) was reported in the Potohar region of Pakistan; out of these, 20.83% were cattle (n = 166) and 9.61% were water buffaloes (n = 77).
LSD is caused by the lumpy skin disease virus (LSDV), which belongs to the genus Capripoxvirus in the family Poxviridae with two other viruses, namely the goatpox virus (GTPV) and sheeppox virus (SPPV) [6]. LSDV has a 151 Kb double-stranded DNA genome that encodes 156 putative proteins [7]. The viral genome has 2.4 kb inverted terminal repeats on both ends [8]. An infection with LSDV is characterized by the appearance of skin nodules, fever, nasal discharge, and lymphadenopathy [9,10]. The severity of symptoms varies depending on the viral strain, breed, and age of the host, and its immune status [11]. LSD is a highly contagious viral disease, with a high morbidity and low to moderate mortality [12,13].
Since it was first reported in Zambia in 1929, the virus has spread across continents, reaching Southeast and South Asian countries [14][15][16]. The endemic LSDV strains in Africa and Southeast and South Asia can be divided into three main groups, named cluster 1.1, cluster 1.2, and cluster 2 [17]. The isolates of cluster 1.1 are the historical strains from South Africa, including the attenuated vaccine strain Neethling [18]. Cluster 1.2 contains isolates from Kenya, identified in the early 1950s, as well as the European and Middle Eastern isolates [19]. With the emergence of the new recombinant LSDV strains, a new distinct cluster 2 arose [20]. This cluster 2 can be further subdivided into six subclades, namely 2.1 to 2.6 [21]. While strains belonging to cluster 2.5 have spread across Asia [20,[22][23][24], strains of the other five clusters remained limited to Russia [25][26][27]. These cluster 2 isolates from Russia and Southeast Asia showed more significant genomic variability compared to the historical strains [21,[28][29][30]. These clade 2 strains are often referred to as recombinant LSDV strains. They show vaccine-like genomic profiles that are attributable to the recombination between the Neethling vaccine and the KSGP vaccine strain [31]. They are the result of a recombination that took place during an inappropriate vaccine production process containing both vaccine strains [30][31][32]. The emergence and spread of recombinant strains of LSDV in recent years make the constant monitoring and characterization of LSDV field isolates highly important [33].
The genetic and phylogenetic characterization of strains from an outbreak can provide valuable information about the origin of the disease, hotspot areas, and the level of transboundary circulation. It can also be helpful in developing eradication and control strategies [34,35]. Molecular characterization has for a long time relied on the analysis of various genes, such as B22R (a hallmark for differentiation between vaccine and wild-type strains), GPCR (a membrane-localized immunomodulatory protein), RPO30 (the 30 KDa subunit of LSDV RNA polymerase), P32 (an immunogenic surface protein similar to P35 of vaccinia virus), and the EEV surface glycoprotein genes (immunogenic target for vaccine development) [30,[36][37][38]. In recent years, however, the whole-genome sequences of LSDV strains have been determined due to advances in sequencing technology.
This study was designed to characterize the LSDV outbreak strains from Pakistan. In the absence of a facility in which the virus could be isolated from field-collected samples, we had to work with samples with a limited viral load to try to determine the sequence. Therefore, rather than whole-genome sequencing, 10 genes were sequenced to ensure a correct phylogenetic placement.
## 2. Materials and Methods
## 2.1. Ethical Approval and Study Setting
The ethical approval (No. CUI-Reg/Notif-3089/22/3182) was obtained from the Ethical Review Board (ERB), COMSATS University, Islamabad. The study was non-intrusive, as all the samples were obtained from skin lesions by swabbing. The study area included different cities of the province of Punjab (n = 34), Khyber Pakhtunkhwa (KPK, n = 20), Sindh (n = 13), Azad Jammu and Kashmir (AJ&K, n= 7), and the Islamabad Capital Territory (n= 2) (Figure 1). Animals selected for sampling presented characteristic necrotic skin nodules.
## 2.2. Sample Collection and Nucleic Acid Extraction
Seventy samples were collected from selected areas of three provinces (Punjab, Sindh, Khyber Pakhtunkhwa) and the Islamabad Capital Territory (ICT) between March and November 2022. Out of these, fifty-nine were swab samples collected from infected animals that presented characteristic skin nodule lesions. The samples were taken from necrotic nodules through a sterilized swab stick contained in a sterile tube carrying 3 mL of Phosphate-Buffered Saline (PBS). In addition, 15 blood samples and 2 scabs were also collected. The samples were transported to the Molecular Virology Laboratory (MVL) using standard procedures and stored at 4 • C until further processing. DNA was extracted using protocols described previously [39], using Macherey-Nagel TM NucleoSpin TM Blood kit (Düren, Germany) following the manufacturer's instructions with a few modifications. Lysis was carried out for an extended period of 1 h, and an external control was added to the B3 buffer before extraction. This section of work was performed in collaboration with the LSDV reference lab at Sciensano, Infectious Diseases in Animals, Exotic and Vector-Borne Diseases, Groeselenberg, Brussels, Belgium.
## 2.3. Application of Real-Time PCR Assays for Genus, Species, and Clade Identifications
Presence of LSDV DNA was confirmed by subjecting samples to a panCapripox realtime PCR using methods described by Haegeman et al. in 2013 [39]. The Wolff assay described by Wolff et al. in 2021 [40] was used to differentiate between different species of the Capripox genus, being LSDV, goatpox, and sheeppox virus. Further characterization of confirmed LSDV strains was carried out using DIVA PCRs as described by Haegeman et al. in 2023 [41]. This real-time PCR differentiates clade 1.1 Neethling vaccine strains from clade 1.2 and clade 2.5 strains.
## 2.4. Molecular Detection and Sequence Analysis of Ten Genomic Regions 2.4.1. Amplification of Selected Regions for Sequence Analysis
Since the majority of reports from Pakistan rely only on sequences of one or two genes for phylogeny, we selected 10 genomic regions to obtain a comprehensive phylogenetic placement. The selected genomic regions, primer sequences, and annealing temperatures are listed in Table 1. The PCR mixes and cycling profiles were carried out as described in the corresponding publications given as references. Briefly, the PCR was carried out by mixing 2 µL DNA, 5 µL of Taq 10× buffer, MgCl2 (2.5 mM), dNTPs (Roche Applied Science, Baden-Württemberg, Germany), 2 units of Taq DNA polymerase (Life Technologies, Carlsbad, CA, USA), and 1 µL of each primer. The following cycling profile was used: 95 • C for 5 min (initial denaturation), followed by 40 cycles of 95 • C for 30 s; appropriate annealing temperature for each primer for 30 s; amplification at 72 • C for 30 s; and final amplification at 72 • C for 10 min.
10 Ser-Thr kinase (ORF25/26)
739, 57 New primers were designed upon an alignment by MAFFT [46] of the following publicly available genotype 1.2 sequences: Evros/GR/15 (KY829023), KSGP0240 (KX683219), Kenya (MN072619), and 155920/2012 (KX894508). A section of the inverted terminal repeat region was targeted, displaying a significant length difference between KSGP0240/Kenya and Evros/155920/2012. Using Primer3 software version 4.1.0 [47] (source code available at https://primer3.ut.ee/; accessed on 10 April 2025, a forward (GGTGAAATATTTTGAAGC-CAAT) and a reverse primer (TTCGAGACCTCGTTTCTGAC) were obtained, amplifying an amplicon of either 278 bp (Evros)/263 bp (155920/2012) or 376 bp (KSGP0240). The 50 µL PCR mix consisted of 2 µL Capx DNA template, 5 µL PCR Taq buffer, 2.5 mM MgCl 2 , 0.3 mM of each dNTP (Roche Applied Science, Germany), and 33.75 pmol of the forward and reverse primer and 1.25 U Taq Polymerase (Thermofisher Scientific, Waltham, MA, USA). The cycling profile used was 95 • C for 4 min, 35 cycles of 95 • C for 30 s, 55 • C for 30 s, and 72 • C for 60 s, followed by one cycle of 72 • C for 10 min.
$$F-5 ′ TCGTTGGTCGCGAAATTTCAG3 ′ R-5 ′ GAGCCATCCATTTTCCAACTCT3 ′ 759, 56 • C [42] EEV F-5 ′ ATGGGAATAGTATCTGTTGTATACG3 ′ R-5 ′ CGAACCCCTATTTACTTGAGAA3 ′ 930, 55 • C [43] B22R F-5 ′ TCATTTTCTTCTAGTTCCGACGA3 ′ R-5 ′ TTCGTTGATGATAAATAACTGGAAA3 863, 58 • C [30] RPO30 F-5 ′ ATTCGTTATCGCAGAACAAGG3 ′ R-5 ′ CACCAACCATAGAATAGTATTGAGAC3 ′ 1234, 55 • C [44] GPCR GPCR internal sequencing primers F5 ′ TTAAGTAAAGCATAACTCCAACAAAAATG3 ′ R5 ′ TTTTTTTATTTTTTATCCAATGCTAATACT3 ′ F5 ′ GATGAGTATTGATAGATACCTAGCTGTAGTT3 ′ R5 ′ TTAAGTAAAGCATAACTCCAACAAAAATG3 ′ 1158, 50 • C [28,37] NTPase (ORF83) F-5 ′ GAGAAACCGCAACAGGAAAA3 ′ R-5 ′ GGATGAGCAACGAACCAACT3 ′ 614, 60 • C [32] RPO132 (ORF116/117) F-5 ′ TGGAGAAATGGAAAGGGATTG3 ′ R-5 ′ CAGGCGACGATGATGAAAC3 ′ 750, 60 • C [32] VLTF-1 (ORF58/59) F-5 ′ TTTTATGGCGTTCCACGATT3 ′ R-5 ′ CCCAACACTCTCTCGCTTCA3 ′ 755, 60 • C [32] Finger Protein (ORF10) F-5 ′ ACCCAACAACACAAGGAAGG3 ′ R-5 ′ CATCGCAAACAAAGAATAAGAAAG3 ′ 708, 60 • C [32]$$
$$F-5 ′ TTCGTTTTCAGCGATTTTATTT3 ′ R-5 ′ AGGAGATTTTATTATGAGTGGCTT3 ′$$
## 2.4.3. Gel Electrophoresis, Extraction, and Sequencing
Following amplification, the PCR products were loaded on a 1.5% agarose gel and verified under UV using GelRed from Biotium (Fremont, CA 94538, USA, cat. number 41003). The amplicons were excised and purified using a gel extraction kit (QIAquick Gel Extraction Kit; Qiagen, Nordrhein-Westfalen, Germany, Cat no./ID. 28704) as described by the manufacturer. Before submission for Sanger sequencing, the quantity and purity of the extraction products were verified using Nanodrop (Thermo Fisher Scientifics, Bremen, Germany) (ratio 260/280). The sequenced fragments from both samples 1.4 (LSDV_1_NS_2022_PAK) and 3.29 (LSDV_2_NS_2022_PAK) were submitted to GenBank under the following accession numbers: B22R; PV492546, PV492556, EEV gene; PV492547, PV492557, P32; PV492548, PV492558, RPO30; PV492549 (not sequenced from LSDV_2_NS_2022_PAK), GPCR; PV492550, PV492559, NTPase; PV492551, PV492560, RNA polymerase 132 (RPO132); PV492552, PV492561, VLTF-1; PV492553, PV492562, Finger Protein; PV492554, PV492563, Ser/Thr kinase; PV492555 and PV492564 for samples LSDV_1_NS_2022_PAK and LSDV_2_NS_2022_PAK, respectively.
## 2.4.4. Phylogenetic Analysis
The obtained sequences were manually assembled in GeneDoc (version 2.7, freely available at https://genedoc.software.informer.com/2.7/, accessed on 10 April 2025) and checked using BLASTn, version 2.16.0 [48]. Publicly available LSDV sequences (representative of Asia, Europe, and Africa, and including all known genotypes), as well as those from goatpox and sheeppox viruses for all 10 regions, were collected from GenBank (https://www.ncbi.nlm.nih.gov/genbank/, accessed on 4 November 2024) and aligned using CLUSTALW in MEGA version 11 [49]. Subsequently, a nucleotide substitution model was selected for each dataset based upon the Bayesian Information Criterion (BIC) and Akaike Information Criterion (AIC) scores. Evolutionary history was inferred using the neighbor-joining method and the maximum likelihood method in MEGA version 11. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches [50].
## 3. Results
## 3.1. Sample Quality Control and Genus Confirmation
Approximately 57% (23/40) of the collected swab samples tested positive for LSDV using the panCapripox real-time PCR (Table 2). The lowest percentage of positive swabs was collected in the KPK region (35.7%), but many of these swabs also tested negative for the internal control (IC), indicating a low sample quality. In general, the Cp values (crossing point where the sample fluorescence signal rises above the background level) for the virus (D5r) were relatively high, considering that the swabs were taken from lesions. The average Cp value of the positive samples was 35.8, with a standard deviation of 4. The average Cp value of the internal control was 38.67, with a standard deviation of 4.4. The importance of good sample quality and storage was confirmed by the fact that 91.3% of the Capripox-positive samples were also positive for the IC (21 out of 23), while only 14% of Capripox-positives were detected in the IC-negative samples (2 out of 14).
## 3.2. Sequence Analysis of 10 Genomic Regions
Based on the sampling area, the obtained Cp value, and the amount of material available, two samples were selected for further characterization: one from Punjab (Sample ID: LSDV_1_NS_2022_PAK) and another from the Kyber Pakhtunkhwa (KPK) province (Sample ID: LSDV_2_NS_2022_PAK). In the initial analysis, the samples were tested with the DIVA PCR to determine LSDV clade identification. This DIVA PCR showed that both samples, LSDV_1_NS_2022_PAK and LSDV_2_NS_2022_PAK, contained wild-type-like strains, and not the clade 1.1 Neethling vaccine strain (Table 3). To provide additional sequence information, ten viral genomic regions were PCR-amplified (Supplementary Figure S1), purified, and Sanger-sequenced from both samples (Table 1), except for the RPO30 region, which was sequenced from sample LSDV_1_NS_2022_PAK only, as LSDV_2_NS_2022_PAK had insufficient material. The combined sequence length of all analyzed regions for sample LSDV_1_NS_2022_PAK was 7965 nucleotides, while this was 6827 nucleotides for sample LSDV_2_NS_2022_PAK (Table 4). Except for the obtained length difference, the sequences of both samples were 100% identical. The difference in the obtained length was linked to the quality of base calling at the ends of the fragments. The nucleotide differences found between the samples and the representative sequences for the different LSDV genotypes are represented in Table 4. The nucleotide comparison between the samples from Pakistan and the representative database sequences clearly indicated the presence of lumpy skin disease virus that was closest to genotype 1.2 LSDV strains in the database. However, a difference was obtained within the LSDV genotype 1.2 between the European sequences from Serbia and Greece and the ones from Africa (Kenya and KSGP0240). All the new recombinant strains (genotype 2.1 to 2.6) had at least 81 nucleotide differences from the Pakistan samples. The latter also had 120 or more nucleotide differences from the three representatives of the vaccine genotype 1.1. The sequence analysis of these regions was in full agreement with the initial DIVA PCR results.
## 3.3. Differentiation Between Genotype 1.2 Strains by Gel-Based PCR
As LSDV clade 1.2 also contains the KSGP0240 vaccine strain, a fit-for-purpose, gelbased PCR was developed to differentiate this vaccine (or vaccine-like strains) from the other clade 1.2 strains. Primers were designed to target a part of the inverted terminal repeat region in the LSDV genome displaying a 96 bp length difference between the European/Middle Eastern and African isolates using several reference sequences. This difference was confirmed by testing clade 1.2 strains/isolates, including Evros (KY829023), 155920/2012 (KX894508), and the vaccine strain KSGP0240 (KX683219).
The amplicons of seven samples from the Pakistan samples, including LSDV_1_NS_2022_PAK and LSDV_2_NS_2022_PAK, migrated at a similar length as those from Evros (KY829023) and 155920/2012 (KX894508) (Supplementary Figure S2), and thus different from KSGP0240 (KX683219). The identity of the fragments and the sequence length difference were confirmed by Sanger sequencing of the amplicons from a sample from Pakistan (sample ID 2.20) and from KSGP0240. The PCR and sequencing results fully supported the previous sequence analysis of the GPCR region, which showed a closer match between samples from Pakistan and the European/Middle Eastern sequences.
## 3.4. Amino Acid Sequence Comparison of Amplified Regions
To understand the impact of the nucleotide changes on the protein level, the amino acid sequences of all 10 regions were compared to the corresponding regions of closely related LSDV strains in phylogeny (Table 1, Figure 2A-C, Supplementary Figure S3A-E). Out of these 10 genes, no amino acid changes were observed in NTPase and VLTF-1. RPO132 and B22R were not the complete gene sequences and hence were not included.
The GPCR amino acid sequence (1098 bp, 366 AA; accession No. PV492559.1) exhibited a four-amino-acid deletion (TILS). These four amino acids-threonine-30, isoleucine-31, leucine-32, and serine-33-are present in the strains LSDV/China/SX/2023 (PP894832) and Vietnam 20L43_Ly-Quoc/VNM/20 (MZ577074); however, they are missing in isolates from Israel (155920/2012, KX894508), Pakistan LSDV_2_NS_2022_PAK CC (PV492559), and Albania LSDV/Albania/4192/2016 (OR134835). A similar deletion has also been reported in the GPCR amino acid sequence of an Indian isolate LSDV/02/KASH/IND/2022 (OQ588787) (Figure 2A). Additionally, a threonine at position 49 is missing in the GPCR protein of the isolate from Pakistan, whereas it is present in all other strains, including the Indian isolate.
The EEV glycoprotein from an isolate of Pakistan LSDV_2_NS_2022_PAK (PV492557) has an amino acid substitution E85K. A similar substitution has also been reported in an isolate from Bangladesh. In contrast, the Kenyan isolate Kubash/KAZ/16 (MN642592), the Albanian isolate LSDV/Albania/4770/2016 (OR134836), and the Neethling strain (AF409138) have glutamic acid at position 85, whereas the Indian isolate LSDV/2022/Jamnagar/N1 (OR393167) and the isolate from Pakistan LSDV_2_NS_2022_PAK (PV492557) have lysine. These two amino acids have opposite charges. Another amino acid substitution, I168M, is present in the isolates from Pakistan (LSDV_2_NS_2022_PAK, PV492557), India (LSDV/2022/Jamnagar/N1, OR393167), and Kenya (Kubash/KAZ/16, MN642592). The Albanian strains retain isoleucine at this position, while strains from Pakistan, Kenya, India, and Neethling all have methionine at position 168. Both amino acids (isoleucine and methionine) are hydrophobic and neutral in charge (Figure 2B).
In the RPO30 protein of LSDV_1_NS_2022_PAK, an amino acid substitution at position 14 (T14N) was observed. Threonine, in other closely related RPO30 genes in phylogeny, was substituted by asparagine in LSDV_1_NS_2022_PAK. Both amino acids are hydrophilic and polar in nature (Figure 2C).
The P32 envelop protein gene of LSDV is widely used as a genetic marker because it is conserved in the genus. The P32 protein of the isolate from Pakistan LSDV_1_NS_2022_PAK was 100% identical to isolates from India LSDV/Cattle/India/2019/Ranchi-1/P50 (OK422494), Serbia ERBIA/Bujanovac/2016 (KY702007), and Albania (SDV/Albania/1707/2016, OR134834). However, an isolate from Russia LSDV/Russia/Udmurtiya/2019 (MT134042) has a substitution of valine to isoleucine at position 272 (V272I). Both amino acids are of similar polarity (Supplementary Figure S3A).
The fusion protein encoded by ORF117 is responsible for the fusion of the viral envelope with the host membrane. The protein sequence of the isolate from Pakistan LSDV_1_NS_2022_PAK (PV492552) was identical to isolates from India LSDV/2022/Nohar (OR393178) and Albania (LSDV/Albania/790/2017, OR134837). However, another isolate from India (ICAR/NIVEDI/LSDV/Gaur/Karntaka/2023/India, PQ510117) and an isolate from China (LSDV/Jiling/2022, OR567413) have a tyrosin substituted for a histidin at position 30 (Y30H). Both amino acids are polar (Supplementary Figure S3B).
## 3.5. Phylogenetic Analysis
A phylogenetic evaluation was carried out for all ten regions for both isolates from Pakistan (LSDV_1_NS_2022_PAK and LSDV_2_NS_2022_PAK), using both neighbor-joining (NJ) and maximum likelihood (ML) analyses. The most suitable nucleotide substitution models were selected for each dataset and are listed in Supplementary Table S1, along with other implemented parameters.
While the isolates from Pakistan clustered with different LSDV genotypes depending on the genomic region analyzed, the clade 1.2 sequences were the only ones that consistently grouped with them regardless of the method used (neighbor-joining or maximum liklihood). Examples of the evolutionary inference are depicted in Figure 3A-E. Overall, when all regions were considered together, the phylogenetic mapping indicated that the LSDV strains responsible for the 2021-2022 outbreak in Pakistan belonged to clade 1.2.
However, when we examined the individual trees, different clustering patterns were observed depending upon the genomic region analyzed. For example, the B22R region of LSDV_1_NS_2022_PAK (Figure 3A) clustered closely with the Rusian isolate LSDV/Russia/Tyumen/2019 (OL542833) and the Albanian strain LSDV/Albania/4192/2016 (OR134835). In contrast, the EEV gene (Figure 3B) exhibited the closest relationship with the Jmnagarh strain from India (LSDV/2022/Jamnagar/N1, OR393167), making a sister clade. Similarly, GPCR and P32 map closely with strains from Russia LSDV/Russia/Udmurtiya/2019 (MT134042) and strains from Serbia SERBIA/Bujanovac/2016 (KY702007) (Figure 3C,D).
(A) (B) The tree was generated using the neighbor-joining method; however, the phylogenetic placement of both isolates remained the same using the maximum likelihood method. The EEV gene made a separate sister clade with the Jamnagar isolate from India. However, the phylogenetic placement of the isolate remained the same using the maximum likelihood method as well.
## 3 (E)
## 4. Discussion
LSD is an economically important and transboundary viral animal disease recognized by WOAH (World Organization for Animal Health) [51]. It has been one of the most devastating and emerging threats to domestic animals, wild bovines, and water buffaloes in recent years [11,52]. The current study aimed to characterize the LSDV strains from the first large-scale outbreak in the Punjab and Khyber Pakhtunkhwa provinces of Pakistan in 2021-2022 through PCR, nucleotide, and phylogenetic analyses of ten targeted genes: RPO30 (ORF36), GPCR (ORF11), B22R (ORF134), P32 (ORF74), EEV (ORF126), NTPase (ORF83), RNA polymerase subunit 132 (ORF116), Late Transcription Factor VLTF-1 (ORF58), Finger Protein (ORF10), and Ser/Thr protein kinase (ORF25) [28]. The presence of LSDV was already confirmed in the geographical region surrounding Pakistan, including India, Bangladesh, UAE, and China [51,[53][54][55][56].
During this study, samples were collected by swabbing the lesions/nodules of clinically suspected animals. The low positivity rate of these swabs and the high Cp values in the positive samples show that external swabbing may not be the most suitable method for sample collection when the highest sensitivity is required (first case detection, import/export, etc.). Nevertheless, it remains an interesting way of taking additional samples, as it is easy and simple to collect in situations where the individual status is of less importance and an answer on the herd level is sufficient. When taking such a type of swab, it is important to verify that they are collected correctly, which can be monitored by looking for the presence of host skin material (hair, scabs, and blood traces). Alternatively, the use of an internal control to detect host material is a useful tool to avoid potential false-negative results. In this study, positive swabs were almost exclusively obtained when the internal control showed positive results. Notwithstanding this issue, LSDV was confirmed in cattle from Pakistan using the panCapripox real-time PCR targeting the D5r region. Using a DIVA real-time PCR, designed by Andy Haegeman in 2023, to differentiate between homologous vaccine strains (clade 1.1) and classical (clade 1.2) and recombinant (clade 2.5) LSDV strains, the isolates were shown to be distinct from the vaccine clade 1.1, thereby successfully validating this assay in our study [41].
As mentioned earlier, the LSDV strains can be divided into three distinct clusters: 1.1, 1.2, and cluster 2. The different genomic regions of both isolates from Pakistan clustered with different isolates from Bengal, India, China, Russia, Albania, Serbia, and Tibet. Most importantly, the isolates from Pakistan always clustered with clade 1.2 strains and never with clade 1.1 (vaccine) strains, independently of the region used. However, when looking at the different regions, LSDV strains belonging to one or more of the recombinant clades (2.1 to 2.6) also sometimes clustered with the isolates from Pakistan, but which recombinant clade this was depended on the specific region analyzed. It is only when the information of all the trees is put together that it can be concluded that the isolates from Pakistan belong to clade 1.2. This is consistent with a recent report from UVAS Pakistan that presented similar results based on whole-genome and individual gene sequences (OQ566164, OQ566165, OQ589501, OQ589502, OP807845-OP807849) [3,57]. The GPCR gene of isolates from Pakistan was grouped with those from Israel (KX894508), Kenya (MN072619), and India (OR393173), along with several others from the same region. The P32 gene clustered with those from India (OR393176), Albania (OR134835), and Russia (MT134042). The RPO30 gene clustered with West Bengal (OQ427097), Serbian (OR134847), Russian (MN995838), and Albanian isolates (OR134837). The B22R gene was closely related to isolates from West Bengal (OQ427097), India Jamnagar (OR393167), and China Tibet (OR797612). When we compared the EEV gene from both isolates, it was also closest to the India Jamnagar (OR393167) isolate, making a separate sister clade with it, in addition to KX894508 from Israel and the Kubash Kenya isolate (MN642592).
In this study, we compared the nucleotide sequences of selected genomic regions with their closest homologs in phylogeny. Six of the ten regions had no nucleotide differences between the isolates from Pakistan and genotype 1.2 sequences from Serbia and Greece (Evros GR/15). For the four other regions, only very limited differences could be observed (between one and three nucleotides). The nucleotide changes in the EEV fragment, encoding parts of LD126 (EEV) and LD127 proteins, of the samples from Pakistan resulted in only one amino acid change, namely lysine to glutamic acid. This is an interesting change, as lysine is positively charged, whereas glutamic acid is negatively charged. In the sequenced genome region encoding parts of a LAP/PHD finger-like protein (ORF10), the nucleotide change resulted in a positively charged lysine (Evros GR/15) becoming a hydrophobic noncharged phenylalanine. In the RNA polymerase subunit 30 kDa fragment, the nucleotide changes lead to a more conserved amino acid change, namely a threonine (Evros) to an asparagine (Pakistan), as both are non-polar and hydrophilic. In serine/threonine kinase (ORF25), a poly "T" section is extended in the isolates from Pakistan, leading to nine instead of seven "T's". This insertion leads to an altered 3 ′ end of the hypothetical protein LD026. Such an extension is similarly observed in isolates from India, Bangladesh, and China/Tibet between 2022 and 2024 (such as PQ472735, OR393178, and PP053747). Another interesting finding was the 12 bp difference in the GPCR gene between KSGP024/Kenya and the isolates from Pakistan. An identical difference in this gene was also observed in the European/Middle Eastern 1.2 strains. This finding indicates that the sequences of isolates from Pakistan are unrelated to the KSGP0240 vaccine, which was further confirmed through PCR, targeting a region in the inverted repeat region.
When looking at the different regions used for phylogenetic analysis individually, the resolution provided is very variable and not always able to solve the fine structure of the tree. This finding is consistent with the findings of Breman et al. [21]. Although the samples from Pakistan always clustered together with clade 1.2 isolates, they sometimes clustered with other genotypes. This poses an inherent risk of misclassification if only one region is included in a study. This can be circumvented by using multiple regions, as performed in this study, or by whole-genome sequencing (WGS). The latter was not attempted in this study because of the low viral load of the samples and the lack of sufficient quantity. Although some progress has been made in this regard, it remains a drawback of WGS.
The LSDV EEV glycoprotein gene is one of the most reliable and popular genetic markers used to distinguish LSDV field strains from vaccine strains in several studies [43] because it is based on the deletion of 27 bp in the vaccine strains. The multiple sequence alignments of the EEV gene in the current study showed the absence of the 27-nucleotide deletion in isolates from Pakistan, which differentiates between field isolates, Neethling-like vaccine strains, and LSDV recombinants from Southeast Asia and Russia.
## 5. Conclusions
This study was designed to evaluate the molecular and phylogenetic position of the LSDV present in samples collected during the 2021-22 LSDV outbreak in Pakistan. The application of DIVA PCRs showed that the isolates from Pakistan did not belong to the vaccine clade 1.1 and were not linked to the KSGP0240 vaccine, belonging to clade 1.2. Further sequence and phylogenetic analyses revealed that both isolates studied, LSDV_1_NS_2022_PAK and LSDV_2_NS_2022_PAK, belonged to clade 1.2 and were identical. It would be desirable to apply these strategies in large-scale studies for source tracking and the future prevention of outbreaks.
## Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/v17121546/s1, Figure S1: Gel electrophoresis of amplified re-gions of selected samples; lane 1: size marker; 2-5: EEV region; 6-9: B22R region; 2 + 6: sample 1.4; 3 + 7: sample 3.29; 4 + 8: positive control; 5 + 9: negative PCR control. Figure S2: Gel electrophoresis (1.5%) of the obtained amplicons. Lanes 1 and 12: length marker (1 kb plus); lanes 2 to 4: samples from Pakistan, Punjab region (Sample ID: 1.3, 1.4, 1.9); lanes 5 and 6: samples from Pakistan, AJ&K region (sample ID: 1.10; 2.11); lanes 7 and 8: samples from Pakistan, KPK region (Sample ID: 2.20 and 3.29); lane 9: KSGP0240; lane 10: Evros/GR/15; lane 11: 155920/2012. Figure S3A: Amino acid sequence alignment of P32 gene with its close relatives in phylogeny. The amino acid changes are highlighted in yellow. Accession number of strains from Pakistan are highlighted in gray. Amino acid position is marked by numbers at the top. Figure S3B: Amino acid sequence alignment of fusion protein gene with its close relatives in phylogeny. The amino acid changes are highlighted in yellow. Accession numbers of strains from Pakistan are highlighted in gray. Amino acid position is marked by numbers at the top. Figure S3C: Amino acid sequence alignment of NTPase gene with its close relatives in phylogeny. The amino acid changes are highlighted in yellow. Accession number of strains from Pakistan are highlighted in gray. Amino acid position is marked by numbers at the top. Figure S3D: Amino acid sequence alignment of LAP/PHD finger-like protein gene with its close relatives in phylogeny. The amino acid changes are highlighted in yellow. Accession numbers of strains from Pakistan are highlighted in gray. Amino acid position is marked by numbers at the top. Figure S3E: Amino acid sequence alignment of VLTF-1 gene with its close relatives in phylogeny. The amino acid changes are highlighted in yellow. Accession numbers of strains are highlighted in gray. Amino acid position is marked by numbers at the top. Figure S4A: Phylogenetic tree representing NTPase gene of isolates from Pakistan LSDV_1_NS_2022_PAK, PV492551, and LSDV_2_NS_2022_PAK, PV492560, indicating their close relation to Russian and Bulgarian strains of LSDV. The tree was generated using the neighbor-joining method. Figure S4B: Phylogenetic tree representing RPO132 gene of isolates from Pakistan LSDV_1_NS_2022_PAK, PV492552, and LSDV_2_NS_2022_PAK, PV492561, indicating their close relation to Russian, Balkan, and Indian strains of LSDV. The tree was generated using the neighbor-joining method; however, the phylogenetic placement of both isolates remained the same using the maximum likelihood method. Figure S4C: Phylogenetic tree representing VLTF-1 gene of isolates from Pakistan LSDV_1_NS_2022_PAK, PV492553, and LSDV_2_NS_2022_PAK, PV492562, indicating their close relation to Russian, Balkan, and Indian strains of LSDV. The tree was generated using the neighbor-joining method; however, the phylogenetic placement of both isolates remained the same using the maximum likelihood method. Figure S4D: Phylogenetic tree of LAP/PHD finger-like protein gene of isolates from Pakistan LSDV_1_NS_2022_PAK, PV492554, LSDV_2_NS_2022_PAK, PV492563, indicating their close relation to Indian and Balkan strains of LSDV. The tree was generated using the neighbor-joining method; however, the phylogenetic placement of both isolates remained the same using the maximum likelihood method. Figure S4E: Phylogenetic tree representing Ser/Thr kinase gene of isolates from Pakistan LSDV_1_NS_2022_PAK, PV492555, and LSDV_2_NS_2022_PAK, PV492564, indicating their close relation to Russian and Albanian strains of LSDV. The tree was generated using the neighbor-joining method; however, the phylogenetic placement of both isolates remained the same using the maximum likelihood method. Table S1: Parameters for the phylogenetic analysis in Mega11 [2,3].
## References
1. Haider, Farhan, Nawaz et al. (2023) "The financial toll of lumpy skin disease in Pakistan, and whether or not vaccination is worth it for preventing future outbreaks" *Ann. PIMS-Shaheed Zulfiqar Ali Bhutto Med Univ*
2. Rehman, Jingdong, Chandio et al. (2017) "Livestock production and population census in Pakistan: Determining their relationship with agricultural GDP using econometric analysis" *Inf. Process. Agric*
3. Khan, Mukhtar, Bin Aslam et al. (2025) "Sequence Based Characterization of Lumpy Skin Disease Virus from Punjab, Pakistan" *Pak. J. Zool*
4. Khatri, Rai, Aashish et al. "Epidemic of lumpy skin disease in Pakistan" *Vet. Med. Sci*
5. Khadim, Shah, Riaz et al. (2025) "Epidemiology of Lumpy Skin Disease virus in large ruminants of Potohar region" *Pakistan. Trop. Anim. Health Prod*
6. Badhy, Chowdhury, Settypalli et al. (2021) "Molecular characterization of lumpy skin disease virus (LSDV) emerged in Bangladesh reveals unique genetic features compared to contemporary field strains" *BMC Vet. Res*
7. Sprygin, Babin, Pestova et al. (2018) "Analysis and insights into recombination signals in lumpy skin disease virus recovered in the field" *PLOS ONE*
8. Sohier, Haegeman, Mostin et al. (2019) "Experimental evidence of mechanical lumpy skin disease virus transmission by Stomoxys calcitrans biting flies and Haematopota spp" *horseflies. Sci. Rep*
9. Namazi, Tafti "Lumpy skin disease, an emerging transboundary viral disease: A review" *Vet. Med. Sci*
10. Gupta, Patial, Bali et al. (2020) "A review: Lumpy skin disease and its emergence in India" *Vet. Res. Commun*
11. Das, Chowdhury, Akter et al. "An updated review on lumpy skin disease: Perspective of southeast asian countries" *J. Adv. Biotechnol. Exp*
12. Şevik, Do Gan (2017) "Epidemiological and Molecular Studies on Lumpy Skin Disease Outbreaks in Turkey During 2014-2015" *Transbound. Emerg. Dis*
13. Akther, Akter, Sarker et al. (1861) "Global Burden of Lumpy Skin Disease, Outbreaks, and Future Challenges" *Viruses*
14. Saegerman, Bertagnoli, Meyer et al. (2018) "Risk of introduction of Lumpy Skin Disease into France through imports of cattle" *Transbound. Emerg. Dis*
15. Kononov, Prutnikov, Shumilova et al. (2019) "Determination of lumpy skin disease virus in bovine meat and offal products following experimental infection" *Transbound. Emerg. Dis*
16. Maw, Khin, Hadrill et al. "First Report of Lumpy Skin Disease in Myanmar and Molecular Analysis of the Field Virus Isolates. Microorganisms 2022"
17. Mazloum, Van Schalkwyk, Babiuk et al. "Lumpy skin disease: History, current understanding and research gaps in the context of recent geographic expansion"
18. Van Schalkwyk, Kara, Heath (2022) "Phylogenomic characterization of historic lumpy skin disease virus isolates from South Africa" *Arch. Virol*
19. Xie, Cui, Liao et al. "Genomic analysis of lumpy skin disease virus asian variants and evaluation of its cellular tropism"
20. Mathijs, Vandenbussche, Nguyen et al. "Coding-Complete Sequences of Recombinant Lumpy Skin Disease Viruses Collected in 2020 from Four Outbreaks in Northern Vietnam" *Microbiol. Resour. Announc*
21. Breman, Haegeman, Krešić et al. (2023) "Lumpy Skin Disease Virus Genome Sequence Analysis: Putative Spatio-Temporal Epidemiology, Single Gene versus Whole Genome Phylogeny and Genomic Evolution" *Viruses*
22. Zan, Huang, Guo et al. (2022) "Molecular characterization of a novel subgenotype of lumpy skin disease virus strain isolated in Inner Mongolia of China" *BMC Vet. Res*
23. Trinh, Nguyen, Nguyen et al. (2022) "Molecular and histopathological characterization of lumpy skin disease in cattle in northern Vietnam during the 2020-2021 outbreaks" *Arch. Virol*
24. Paungpin, Sariya, Chaiwattanarungruengpaisan et al. "Coding-Complete Genome Sequence of a Lumpy Skin Disease Virus Isolated during the 2021 Thailand Outbreak"
25. Krotova, Byadovskaya, Shumilova et al. (2019) "Molecular characterization of a novel recombinant lumpy skin disease virus isolated during an outbreak in Tyumen"
26. Aleksandr, Olga, David et al. (2020) "Non-vector-borne transmission of lumpy skin disease virus" *Sci. Rep*
27. Saltykov, Kolosova, Filonova et al. (2021) "Genetic Evidence of Multiple Introductions of Lumpy Skin Disease Virus into Saratov Region" *Russia. Pathogens*
28. Tulman, Afonso, Lu et al. (2001) "Genome of lumpy skin disease virus" *J. Virol*
29. Douglass, Van Der Walt, Omar et al. (2019) "The complete genome sequence of the lumpy skin disease virus vaccine Herbivac LS reveals a mutation in the superoxide dismutase gene homolog" *Arch. Virol*
30. Koirala, Meki, Maharjan et al.
31. Sprygin, Pestova, Bjadovskaya et al. "Evidence of recombination of vaccine strains of lumpy skin disease virus with field strains, causing disease"
32. Haegeman, De Leeuw, Saduakassova et al. (1019) "The importance of quality control of lsdv live attenuated vaccines for its safe application in the field" *Vaccines*
33. Wilhelm, Ward (2023) "The spread of lumpy skin disease virus across southeast asia: Insights from surveillance" *Transbound. Emerg. Dis*
34. Ochwo, Vanderwaal, Ndekezi et al. (2020) "Molecular detection and phylogenetic analysis of lumpy skin disease virus from outbreaks in Uganda 2017-2018" *BMC Vet. Res*
35. Singhla, Boonsri, Kreausukon et al. "Molecular Characterization and Phylogenetic Analysis of Lumpy Skin Disease Virus Collected from Outbreaks in Northern Thailand in 2021" *Vet. Sci*
36. Menasherow, Erster, Rubinstein-Giuni et al. (2016) "A high-resolution melting (HRM) assay for the differentiation between Israeli field and Neethling vaccine lumpy skin disease viruses" *J. Virol. Methods*
37. Goff, Lamien, Fakhfakh et al. (2009) "Capripoxvirus G-protein-coupled chemokine receptor: A host-range gene suitable for virus animal origin discrimination" *J. Gen. Virol*
38. Lamien, Lelenta, Goger et al. (2011) "Real time PCR method for simultaneous detection, quantitation and differentiation of capripoxviruses" *J. Virol. Methods*
39. Haegeman, Zro, Vandenbussche et al. (2013) "Development and validation of three Capripoxvirus real-time PCRs for parallel testing" *J. Virol. Methods*
40. Wolff, Beer, Hoffmann "Probe-based real-time qPCR assays for a reliable differentiation of Capripox virus species. Microorganisms 2021"
41. Haegeman, De Leeuw, Philips et al. (2023) "Development and validation of a new DIVA real-time PCR allowing to differentiate wild-type lumpy skin disease virus strains, including the Asian recombinant strains, from Neethling-based vaccine strains" *Viruses*
42. Yousefi, Dalir-Naghadeh, Mardani et al. (2018) "Phylogenetic analysis of the lumpy skin disease viruses in northwest of Iran" *Trop. Anim. Health Prod*
43. Chibssa, Sombo, Lichoti et al. (1142) "Molecular analysis of east african lumpy skin disease viruses reveals a mixed isolate with features of both vaccine and field isolates" *Microorganisms*
44. Lamien, Le Goff, Silber et al. (2011) "Use of the Capripoxvirus homologue of Vaccinia virus 30kDa RNA polymerase subunit (RPO30) gene as a novel diagnostic and genotyping target: Development of a classical PCR method to differentiate Goat poxvirus from Sheep poxvirus" *Vet. Microbiol*
45. Haegeman, Zro, Sammin et al. (2016) "Investigation of a Possible Link between Vaccination and the 2010 Sheep Pox Epizootic in Morocco" *Transbound. Emerg. Dis*
46. Katoh, Rozewicki, Yamada (2019) "MAFFT online service: Multiple sequence alignment, interactive sequence choice and visualization" *Brief. Bioinform*
47. Rozen, Skaletsky (2000) "Primer3 on the WWW for general users and for biologist programmers"
48. Sayers, Beck, Bolton et al. (2021) "Database resources of the National Center for Biotechnology Information" *Nucleic Acids Res*
49. Tamura, Stecher, Kumar (2021) "MEGA11: Molecular evolutionary genetics analysis version 11" *Mol. Biol. Evol*
50. Felsenstein (1985) "Confidence limits on phylogenies: An approach using the bootstrap" *Evolution*
51. Prabhu, Malmarugan, Rajagunalan et al. (2024) "Isolation and molecular characterization of lumpy skin disease virus from Tamil Nadu, India during the outbreaks from 2020 to 2022" *Virus Genes*
52. Turan, Yilmaz, Tekelioglu et al. "Lumpy skin disease: Global and Turkish perspectives" *Approaches in Poultry. Dairy Vet. Sci*
53. Bhat, Chaudry, Mittal et al. "Complete genome sequence of the lumpy skin disease virus reported from Jammu and Kashmir, India, during 2022 outbreak"
54. Song, Zuo, Zhang et al. (2024) "Emergence of Lumpy Skin Disease Virus Infection in Yaks, Cattle-Yaks, and Cattle on the Qinghai-Xizang Plateau of China" *Transbound. Emerg. Dis*
55. Reddy, Mounica, Sudeep et al. (2024) "First evidence of lumpy skin disease in mithun (Bos frontalis) in India" *Arch. Virol*
56. Li, An, Sun et al. (2023) "Risk factors and spatiotemporal distribution of lumpy skin disease occurrence in the asian continent during 2012-2022: An ecological niche model" *Transbound. Emerg. Dis*
57. Jabbar, Atif, Kashif et al. "Molecular epidemiology and phylogenetic insights of lumpy skin disease in cattle from diverse agro-ecological regions of"
58. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12414382&blobtype=pdf | # Single-cycle Rift Valley fever virus particles from stable replicon cells enable discovery of antiviral CNX-1351 for multiple RNA viruses
Zhichao Gao, Hongyuan Guo, Ziqiao Wang, Pengcheng Wang, Xinran Sun, Shimei Zhang, Fei Feng, Chao Shan, Youhua Xie, Rong Zhang, Virologica Sinica
## Abstract
Rift Valley fever virus (RVFV) is a high-containment pathogen that causes severe diseases in humans, with no approved therapeutics available. Its classification as a biosafety level 3 (BSL-3) agent has limited research and therapeutic development due to safety concerns. In this study, we developed a stable replicon cell line maintaining the replication of L and S genomic segments of RVFV. Singlecycle viral replicon particles (VRPs) could be efficiently packaged through trans-complementation of glycoproteins from different strains, recapitulating authentic viral entry and replication while minimizing biosafety risks. Using this system, we conducted high-throughput screening of a smallmolecule compound library and identified CNX-1351 as an antiviral agent for multiple RNA viruses. Mechanistic studies revealed that CNX-1351 inhibits viral replication, potentially by targeting the PI3K-Akt signaling pathway. This single-cycle VRP system provides a valuable tool for studying RVFV biology, host interactions, antiviral and vaccine development under reduced biosafety constraints.
## INTRODUCTION
Rift Valley fever virus (RVFV) is an enveloped, segmented, singlestranded, negative-sense RNA virus (Pepin et al., 2010), and belonged to the Phlebovirus genus within the Phenuiviridae family of the Hareavirales order, under the Bunyaviricetes class according to the lastest ICTV taxonomy in 2024. The genome of RVFV comprises three segments: large (L), medium (M), and small (S). The L segment encodes the RNA-dependent RNA polymerase (RdRp), while the M segment expresses the structural glycoprotein Gn and Gc, as well as two nonstructural proteins, NSm1 and NSm2. The S segment encodes the nucleoprotein (N) and the nonstructural protein NSs. During viral assembly, the RdRp, N protein, and genomic RNA form the viral replication machine, which is encapsulated by the Gn and Gc glycoproteins on the virion surface.
RVFV was first identified in 1931 on a farm near Lake Naivasha in Kenya's Great Rift Valley (Daubney and Hudson, 1932), and now has spread across the African continent (Daubney and Hudson, 1931). Rift Valley fever is a zoonotic disease associated with high rates of abortion and mortality in livestock (Easterday, 1965), leading to significant economic losses. In humans, RVFV infection manifests with a wide range of clinical symptoms, from mild flu-like illness to severe, life-threatening conditions, such as hemorrhagic fever, encephalitis, or ocular complications, with a case fatality rate of up to 10% (Mcintosh et al., 1980;Al-Hazmi et al., 2003;Alrajhi et al., 2004). A meta-analysis of data collected from 2000 to 2022 revealed that in Africa, the human prevalence of RVF is estimated to be 7.8%, while its case fatality rate (CFR) in humans can be as high as 27.5% (Ebogo-Belobo et al., 2023). Outside of Africa, recent outbreaks of RVF in Saudi Arabia and neighbouring Yemen territories with 1328 human cases and 166 deaths also made RVF virus a global threat for the first time (Al-Afaleq and Hussein, 2011). These factors underscore the persistent threat posed by RVFV to both public health and the economy.
To control the outbreak of RVF globally and better facilitate the investigation of RVFV, an attenuated virus strain which can be used for vaccine was developed by undirected mutagenesis passage of a RVFV wild-type strain (ZH548) isolated during the epidemic in Egypt (Caplen et al., 1985;Mansfield et al., 2015). Using wild-type RVFV or the attenuated MP-12 strain, several drugs have been repurposed to inhibit RVFV infection in vitro or in vivo, including favipiravir (T-705) and ribavirin (Huggins, 1989;Scharton et al., 2014). High-throughput screening has also identified additional compounds with antiviral activity potential, such as Sorafenib (Benedict et al., 2015) and C795-0925 (Mudhasani et al., 2014). However, no specific antiviral drugs have been approved for clinical use against RVFV, highlighting the urgent need for effective therapeutics.
The classification of RVFV as a biosafety level 3 (BSL-3) pathogen requires the need of stringent containment for research. In this context, the development of a single-cycle RVFV model presents significant advantages, which can minimize the risk of accidental transmission associated with handling pathogenic viruses (Meulen et al., 2023). To establish a safe, reliable, and convenient system for studying RVFV biology and antiviral development, we generated a BHK-21 cell line harboring the L and S genomic segments but lacking the M segment. This system enables the production of single-cycle viral replicon particles (VRPs) through trans-complementation of glycoproteins. The resulting RVFV VRPs recapitulate the authentic viral lifecycle from entry to replication, making them suitable for antiviral studies. Using this system, we conducted high-throughput screening of a compound library and identified CNX-1351 as a potential inhibitor targeting the replication stage of RVFV and other RNA viruses. The development of replicon cell line for efficient packaging of single-cycle RVFV particles provides a valuable tool for studying virus-host interactions and advancing antiviral research under lower biosafety containment.
## RESULTS
## Construction of a replicon cell line for packaging of single-cycle RVFV particles
To establish a safe, stable, and single-cycle system for producing the VRPs, we removed the M segment of the RVFV MP-12 vaccine strain. The open reading frame (ORF) of the NSs gene in S genomic segment was replaced with an mGreen-P2A-Puro element, which consists of the mGreenLantern (mGreen) reporter, a P2A cleavage site, and a puromycin (Puro) resistance gene. Co-transfection of plasmids expressing the L segment, modified S segment, RdRp, N protein, and GnGc glycoproteins into BHK-21 cells stabling expressing T7 polymerase (BHK-T7) enabled the packaging of single-cycle VRPs (Fig. 1A). Infection of wildtype (WT) BHK-21 cells with these VRPs, followed by puromycin selection, generated a stable replicon cell line (BHK-MP-12-Rep). This cell line can be used to package single-cycle VRPs by transfection of plasmid expressing the GnGc glycoproteins of either the MP-12 vaccine strain or ZH548 wild-type strain (Fig. 1A).
Under puromycin selection, around 70%-80% of the cells exhibited fluorescence (Fig. 1B). To confirm the single-cycle replication property of VRPs packaged in BHK-MP-12-Rep cells, we serially passaged the VRPs in WT BHK-21 cells. Cells infected with passage 0 (P0) VRPs harvested from BHK-MP-12-Rep cells displayed non-spreading green fluorescence. Similarly, passage 1 (P1) showed only a few single fluorescent cells, likely due to contamination from P0 VRPs. No fluorescent cells were observed infected with passage 2 (P2) VRPs (Fig. 1C). The replicon cells stably maintained the expression of green fluorescenct protein through 10 passages (Fig. 1D) and enabled packageing of similar infectious units (IU) of MP-12 VRPs as determined in 293T cells (Fig. 1E). The infectious units of both MP-12 VRPs and ZH548 VRPs were also determined in WT BHK-21 cells, revealing comparable titers of approximately 5 × 10 6 IU (Fig. 1F). At a multiplicity of infection (MOI) of 3, nearly 70%-80% of WT BHK-21 cells were fluorescence-positive (Fig. 1G). These results demonstrate the successful packaging of single-cycle VRPs in the replicon cell line.
## Characterization of VRPs for studying virus-host interactions and antivirals
To characterize the utility of VRPs as an alternative to authentic RVFV in studying virus-host interactions and antiviral compounds, we tested several compounds previously reported to inhibit RVFV infection. Treatment of BHK-21 cells with Baf A1, NH 4 Cl, chloroquine, or hydroxidechloroquine effectively inhibited VRP infection (Fig. 1H), consistent with their reported effects on authentic RVFV (Gerrard et al., 2002;Boer et al., 2012;Moy et al., 2014).
We also examined the role of host factors in VRP infection. The B3gat3 gene, which is critical for heparan and chondroitin synthesis, was previously identified as an attachment factor for RVFV infection (Riblett et al., 2016). Clonal B3gat3-knockout mouse embryonic fibroblasts (MEFs) showed partial reduction in the infection by MP-12 VRPs, though this difference did not reach statistical significance (Fig. 1I). Additionally, Lrp1, an important entry receptor mediating RVFV binding and internalization (Ganaie et al., 2021), was investigated. Clonal Lrp1-knockout MEFs and bulk LRP1-knockout A549 cells were generated. Knockout of Lrp1 in MEFs nearly eliminated VRP infection, while bulk knockout of LRP1 in A549 cells significantly reduced the infection (Fig. 1J andK).
Collectively, these findings demonstrate that VRPs recapitulate key aspects of authentic RVFV infection, particularly entry and replication, and can serve as a valuable tool for studying host factors and antiviral compounds.
## Screening of antiviral compounds against RVFV using the VRP system
To evaluate the utility of RVFV VRPs as a tool for high-throughput antiviral screening and to identify potential inhibitors of RVFV infection, we screened a library of 3185 compounds using MP-12 VRPs in Huh-7 human cells. For the screening, Huh-7 cells in 96-well plates were pre-treated with each compound at 10 μM for 1 h, using DMSO as negative control, followed by infection with MP-12 VRPs at an MOI of 0.5 for 24 h (Fig. 2A). After fixation, cells were stained with DAPI to quantify the total number of cells, and the percentage of virus-infected fluorescent protein reporter-positive cells were analyzed. The screening results were visualized as scatter plot, with inhibition rate plotted against cell survival rate (Fig. 2B). Compounds with over 90% inhibition and less than 30% cytotoxicity were subjected to further validation.
From the primary screening, ten compounds were re-purchased and validated. Their half-maximal inhibitory concentration (IC 50 ), halfmaximal cytotoxic concentration (CC 50 ), and selection index (SI; calculated as CC 50 /IC 50 ) were determined. Eight of these compounds showed low SI values, either due to high IC 50 or low CC 50 (Fig. 2C). Two compounds, Lycorine Hydrochloride and CNX-1351, showed potent antiviral activity with low cytotoxicity and SI values > 10 (Fig. 2D andE). Since Lycorine Hydrochloride has previously been reported to inhibit RVFV infection (Gabrielsen et al., 1992), we focused our subsequent investigations on CNX-1351.
## CNX-1351 is antiviral for multiple RNA viruses
To further validate the antiviral activity of CNX-1351, we tested its efficacy against the RVFV in multiple cell lines, including A549, HeLa, BV2, MEF, BHK, and Vero E6 cells. CNX-1351 exhibited consistent inhibition of RVFV infection across all cell lines from different tissues or species (Fig. 3A).
We also evaluated the antiviral activity of CNX-1351 against other RNA viruses. La Crosse virus (LACV) and severe fever with thrombocytopenia syndrome virus (SFTSV), both members of the Bunyaviricetes class, were tested. LACV belongs to the Orthobunyavirus genus, while SFTSV is a member of the Phlebovirus genus, the same as RVFV. In Huh-7 cells, CNX-1351 significantly inhibited LACV and SFTSV infection. Additionally, we assessed the activity of CNX-1351 against Zika virus (ZIKV; Flavivirus genus) and Chikungunya virus (CHIKV; Alphavirus genus), both of which are mosquito-borne RNA viruses like RVFV. CNX-1351 inhibited both ZIKV and CHIKV infection in a dose-dependent manner, and CHIKV is more sensitive to CNX-1351 (Fig. 3B). According to the above results, CNX-1351 is an antiviral compound against multiple arboviruses tested.
## CNX-1351 inhibit RVFV infection at the replication step
To determine the stage of the viral lifecycle targeted by CNX-1351, we performed a time-of-addition assay to differentiate its effect on viral entry and post-entry processes. In the 'Entry' group, Huh-7 cells were pre-treated with 10 μM of CNX-1351 for 1 h, and the compound was present during the 2-h viral inoculation period. In the 'Post-entry' and 'Full-time' groups, CNX-1351 was added after viral inoculation or maintained throughout the infection process, respectively. RVFV infection were significantly inhibited in the 'Post-entry' and 'Full-time' groups, indicating that CNX-1351 primarily acts at the post-entry step of the viral lifecycle (Fig. 4A).
To confirm whether CNX-1351 inhibits the stage of viral replication, we employed a minigenome assay. The pBl-MP-12-S-Nluc plasmid, in To further validate these findings, we assessed viral replication in BHK-MP-12-Rep cells. Given the long half-life of mGreen protein that reduces the detection sensitivity, we co-transfected the pBl-MP-12-S-Nluc plasmid with pCAGGS-T7 expressing the T7 polymerase into replicon cells to measure transient luciferase expression. Treatment with CNX-1351 significantly reduced the luminescence (Fig. 4C), confirming its inhibitory effect on viral genome replication. This also demonstrated the utility of BHK-MP-12-Rep cells for screening replication inhibitors in a virus-free system. We extended our investigation to other viruses, performing time-of-addition assays for CHIKV and SFTSV. Similar to RVFV, CNX-1351 exhibited antiviral activity primarily at the post-entry step for both viruses (Fig. 4D andE).
## CNX-1351 may inhibit virus infection by targeting the PI3K-Akt pathway
To explore the mechanism underlying the antiviral activity of CNX-1351, we investigated its potential effects on PI3K-Akt signaling pathway. CNX-1351 was initially designed as a selective inhibitor of PIK3CA, targeting PI3K signaling in PI3Kα-dependent cancer cell lines (Nacht et al., 2013). To confirm whether CNX-1351 modulates gene expression in the PI3K pathway, we performed RNA-Seq analysis on Huh-7 cells treated with 5 μM CNX-1351 for 12 h. Sequencing data was analyzed and visualized using R (version 4.3.1). A volcano plot revealed significant changes in gene expression, with 932 genes unregulated and 1803 genes downregulated (Fig. 5A). Gene ontology and KEGG analysis of downregulated genes highlighted enrichment in pathways such as MAPK signaling, FoxO signaling, AMPK signaling, and insulin resistance (Fig. 5B). These pathways are known to be regulated by PI3K-Akt signaling, suggesting that CNX-1351 modulates gene expression downstream of PI3K-Akt pathway (Schultze et al., 2012). A heatmap further illustrated that changes in PI3K signaling pathway influenced the expression of genes involved in various cellular activities, underscoring its central role in cell metabolism (Fig. 5C).
To assess the role of PI3K signaling in RVFV infection, we tested LY294002 (Gharbi et al., 2007), another PI3K-Akt pathway inhibitor. Huh-7 cells were pre-treated with LY294002 or CNX-1351 at varying concentrations for 1 h, followed by RVFV infection. LY294002 exhibited antiviral activity against RVFV similar to that of CNX-1351 (Fig. 5D). These findings suggest that perturbation of the PI3K-Akt signaling pathway inhibits RVFV infection, although the precise mechanism requires further investigation.
## DISCUSSION
RVFV is an emerging arbovirus that causes severe symptoms and significant economic losses annually (Linthicum et al., 2016). Its classification as a high biosafety level pathogen complicates antiviral research, and the lack of approved targeted therapies underscores the urgent need for effective treatments. To facilitate research in lower biosafety containment, virus replicon particles (VRPs) have been developed as a tool to mimic the biological and pathogenic characteristics of authentic viruses while minimizing risks (Shirbaghaee and Bolhassani, 2016). In this study, we established a safe and reliable cell line capable of packaging high-titer VRPs. Using this system, we conducted high-throughput screening of a compound library and identified CNX-1351 as a potent antiviral compound that inhibits RVFV and other RNA viruses at the post-entry step. Further investigation using minigenome systems demonstrated that CNX-1351 targets viral replication, potentially through modulation of the PI3K-Akt pathway. These findings provide new insight into the interaction between the PI3K-Akt pathway and viral infection.
VRPs have emerged as valuable tools for studying molecular biology and developing antivirals and vaccines for highly pathogenic viruses, such as Ebola virus (EBOV) (Halfmann et al., 2009) and alphaviruses like Venezuelan equine encephalitis virus (VEEV) (Burke et al., 2022) and Chikungunya virus (CHIKV) (Lin et al., 2023). Since the establishment of reverse genetics system for RVFV (Ikegami et al., 2005;Ikegami et al., 2006), various strategies have been developed to produce safe RVFV particles. For instance, single-cycle virus-like particles (VLPs) have been generated by co-transfecting a minigenome vector with plasmids expressing the L, N, and GnGc proteins (Habjan et al., 2009;Piper and Gerrard, 2010). These VLPs can infect target cells but require the presence of L and N protein to support efficient reporter gene expression from the minigenome.
The single-cycle virus replicon particles (VRPs) containing the L and S genomic segments have been trans-packaged in cells expressing GnGc protein by transient transfection of multiple plasmids (Dodd et al., 2012). While these VRPs undergo single-round infection, their production is often limited by low transfection efficiency of multiple plasmids. An alternative approach by Kortekaas et al. involves generating GnGc-stably expressing cell lines that maintain replicating L and S segments through continuous transfection of glycoprotein-expressing plasmids (Kortekaas et al., 2011). This system allows VRP packaging by glycoprotein trans-complementation, but the process of generating replicating cell lines is labor-intensive and complex. Additionally, constitutive glycoprotein expression may lead to continuous VRP production, which is not ideal for controlled studies. Another strategy involves co-transfecting six plasmids to produce particles containing the L and S segments, along with an M segment mutant, in cells expressing functional glycoprotein (Murakami et al., 2014). The M mutant encodes surface-expressing, fusion-defective glycoprotein with mutations and a deletion of C-terminal ER retrieval signal. These particles undergo a single cycle of infection in naïve cells, but progeny virions are unable to fuse with host cells. However, the presence of intact transcription elements in the M segment may raise the risk of reverse mutations that could restore functional glycoprotein expression.
In this study, we developed a replicon cell line that stably maintains the replication of RVFV L and S genomic segments. By introducing a puromycin resistance gene, we ensured the persistent replication of these segments. This system enables efficient packaging of high-titer VRPs through transfection of a single plasmid expressing the GnGc glycoprotein. Compared to previous reported VRP system (Kortekaas et al., 2011;Murakami et al., 2014), our approach to obtain the replicon cell line is simpler and the packaging of VPRs is more adaptable, allowing for the packaging of VRPs bearing glycoproteins from different RVFV strains, such as the wild-type ZH548 strain. This flexibility facilitates functional studies of glycoprotein mutations and enhances the utility of the system for antiviral research.
Our high-throughput screening identified two compounds with significant antiviral activity against RVFV. Lycorine Hydrochloride, an Amaryllidaceae Isoquinoline derivative, was previously identified as an anti-RVFV candidate (Gabrielsen et al., 1992). This class of compounds has demonstrated broad antiviral activity against viruses, such as Japanese encephalitis virus (JEV), influenza A virus (H5N1) (He et al., 2013), and SARS-CoV-2 (Tan et al., 2022). The second compound, CNX-1351, is a potent and selective covalent inhibitor of the PI3Kα isoform, a key component of the PI3K-Akt signaling pathway. CNX-1351 features a quinoline scaffold, which is common in kinase inhibitors targeting lipid kinases like PI3K. This scaffold enables specific binding to the ATP-binding pocket of PI3Kα, inhibiting its activity (Solomon and Lee, 2011).
The PI3K-Akt pathway plays a central role in regulating cell proliferation, RNA processing, protein translation, autophagy, and apoptosis (Vanhaesebroeck et al., 2012). It is also important for antiviral responses, as it regulates the phosphorylation of interferon regulatory factor 3 (IRF3) via PI3K/Akt, promoting type I interferon responses-a critical antiviral mechanism (Blanco et al., 2020). The pathway also supports TLR3-mediated tyrosine phosphorylation and RIG-I-dependent IRF3 activation in response to RNA viruses (Edwards et al., 2007). In our study, CNX-1351 significantly inhibited the replication of multiple RNA viruses, including RVFV, SFTSV, and CHIKV. RNA-Seq analysis revealed that CNX-1351 primarily affects signaling pathways and gene expression downstream of PI3K-Akt. The antiviral activity of LY294002, another PI3K inhibitor, further supports the importance of PI3K-Akt signaling in RVFV infection.
## CONCLUSIONS
In summary, we developed a replicon cell line for efficient packaging of RVFV VRPs, providing a safe, convenient, and versatile tool for studying RVFV biology, antivirals, and vaccine development. This system can be adapted for other highly pathogenic bunyaviruses, offering a platform for advancing research on these emerging pathogens. Our identification of CNX-1351 as an antiviral compounds targeting the PI3K-Akt pathways opens a new avenue for therapeutic development against RVFV and related RNA viruses.
## MATERIALS AND METHODS
## Cell lines
HEK 293T (ATCC #CRL-3216), Huh-7, BHK-21 (ATCC #CCL-10), Vero (ATCC #CCL-81), HeLa (ATCC #CCL-2), BV2, MEF, A549 (ATCC #CCL-185), BHK-MP-12-Rep cells were cultured in Dulbecco's Modified Eagle's medium (Hyclone) supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 1 mM Sodium pyruvate, 1 × nonessential amino acids, and 100 U/mL of Penicillin-Streptomycin. BHK-T7 clonal cell line were generated by transduction of lentivirus expressing T7 polymerase in BHK-21 cells. To knock out Lrp1 and B3gat3 genes in MEF cells, the following sgRNA were used: Lrp1 sgRNA, 5 ′ -AAATGCCGGGTAAA-TAACGG-3 ′ ; B3gat3 sgRNA, 5 ′ -AATGACATAGATAGTAGGCA-3 ′ . To knock out LRP1 in A549 cells, sgRNA used is as follows: LRP1 sgRNA, 5 ′ -CCTGGGAGATCACCACGTAG-3 ′ . The sgRNAs were cloned into the lentiCRISPR v2 (Addgene #52961) and packaged into lentivirus. Transduced cells were selected with puromycin for seven days. Clonal cells were isolated by limiting dilution. All cell lines were tested routinely and confirmed to be free of mycoplasma contamination.
## Plasmids
The L segment of the RVFV MP-12 vaccine strain (accession No. DQ375404.1), flanked by T7 promoter and HDV ribozyme sequences, was cloned into the pBluescript vector to generate pBl-MP-12-L. The S segment (accession No. DQ380154.1) was modified by replacing the NSs gene with mGreenLantern (mGreen)-P2A-Puromycin cassette and cloned into pBluescript to generate pBl-MP-12-S-mGreen. Similarly, pBl-MP-12-S-Nluc was generated by replacing the mGreen with NanoLuc luciferase. The coding sequences of polymerase (L) and nucleocapsid protein (N) were cloned into pCAGGS transient expression vector under the control of a CMV immediate enhancer/β-actin (CAG) promoter, resulting in plasmid pCAGGS-MP-12-L and pCAGGS-MP-12-N, respectively. The open reading frame of the M segment of RVFV MP-12 strain (accession No. DQ380208.1) was cloned into pCAGGS to generate pCAGGS-MP-12-GnGc. Similarly, the M segment of the RVFV ZH548 strain (accession No. NC_014396.1) was cloned to generate pCAGGS-ZH548-GnGc.
## Packaging of VRPs and generation of replicon cells
BHK-21 cells stably expressing T7 polymerase (BHK-T7) were seeded in six-well plates and co-transfected with pBl-MP-12-L, pBl-MP-12-S-mGreen, pCAGGS-MP-12-L, pCAGGS-MP-12-N and pCAGGS-MP-12-GnGc using Fugene HD transfection reagent (Promega) according to the manufacturer's instructions. Supernatants containing infectious VRPs were collected at approximately 72 h post-transfection, designated as passage 0 (P1), and used for subsequent infections or replicon cell line generation.
To generate a replicon cell line containing only the L and S genomic segments of RVFV, wild-type BHK-21 cells were seeded in six-well plates and infected with MP-12-VRPs. Cells were expanded and selected with puromycin for 5-7 days, resulting in BHK-MP-12-Rep cell line. To assess the stability, BHK-MP-12-Rep cells were serially passaged for 10 times (P10) to measure the percentage of fluorescence positive cells by flow cytometry. The production of VRPs in P10 replicon cells was also determined.
To package VRPs using the BHK-MP-12-Rep cell line, cells were seeded into six-well plates and transfected with pCAGGS-MP-12-GnGc or pCAGGS-ZH548-GnGc. Supernatant containing VRPs were collected 48 h post-transfection and stored at -80 • C for use.
## Serial passaging and titration of VRPs
To evaluate the single-cycle replication property of MP-12-VRPs, wild-type (WT) BHK-21 cells in 12-well plate were infected with 250 μL of P0 VRPs collected from transfected cells above. After 2 h of inoculation, cells were washed three times and supplemented with fresh medium. After approximately 5 days, the fluorescent images were captured. Supernatants were collected to inoculate WT BHK-21 cells to passage two more times as described above for P0 VRPs.
To measure the infectious unit (IU) of VRPs, WT 293T or BHK-21 cells in 96-well plate were infected with VRPs at different dilutions. After 24 h, cells were trypsinized, fixed with 2% paraformaldehyde (PFA), and analyzed using flow cytometry. The IU/mL was calculated using the formula: (percentage of GFP-positive cells) × (total cell number)/(inoculation volume).
## High-throughput screening of small-molecular compounds
A library of 3185 small-molecular compounds was purchased from TargetMol. All compounds were purified using HPLC with a purity of ≥ 98%. Huh-7 cells were seeded in 96-well plates at a density of 40,000 cells per well and pretreated with each compound at a final concentration of 10 μM for 1 h. Cells were infected with RVFV MP-12-VRPs at an MOI of 0.5 for 24 h in the presence of the compounds. DMSO, at the same concentration as the compounds, was used as a negative control. Cells were fixed with 2% PFA for 1 h and stained with DAPI (Beyotime). Images were acquired using an Operetta high-content imaging system (PerkinElmer) and analyzed using Harmony 3.5 software. The infection rate of the DMSO control was set to 100%, and the infection rate of each compound-treated sample was normalized to DMSO control. The percentage inhibition was calculated as 100% minus the normalized infection rate. Cell survival rate similarly normalized to the DMSO control.
## Validation of antiviral activity of compounds
Huh-7 cells seeded in 96-well plate were pretreated with serially diluted compounds for 1 h prior to infection with VRPs at an MOI of 0.5. After 12 h, cells were trypsinized, fixed with 2% PFA, and analyzed by flow cytometry to measure virus infection. DMSO was used as a negative control, and antiviral activity was calculated by normalizing to the DMSO control.
## Cell viability assay
Huh-7 cells were seeded in 96-well plates at a density of 20,000 cells per well for 24 h, and then treated with varying concentrations of compounds for 48 h. Cell viability was assessed using the CellTiter-Lumi™ Steady kit (Beyotime) according to the manufacturer's instructions.
## Time-of-addition assay
Huh-7 cells were infected with RVFV MP-12 (MOI = 0.5), SFTSV (MOI = 5), or CHIKV-181 (MOI = 10). The inoculum was removed 2 h post-infection. For the 'Entry' group, cells were pretreated with 10 μM CNX-1351 for 1 h, followed by infection for 2 h, after which both the virus and compound were removed. For 'Post-entry' group, CNX-1351 was added after virus inoculation. For "Full-time" treatment group, cells were incubated with CNX-1351 for 1 h prior to infection, and the compound was maintained throughout the infection process. Virus infection was measured by flow cytometry or high-content imaging at12 h post-infection.
## Minigenome assay
BHK-T7 cells were co-transfected with pCAGGS-MP-12-L, pCAGGS-MP-12-N, and pBl-MP-12-S-Nluc using Fugene HD (Promega). As an alternative strategy, BHK-MP-12-Rep cells were used to assess the viral replication, and co-transfected with pCAGGS-T7 and pBl-MP-12-S-Nluc. At 8 h post-transfection, the culture medium was refreshed and CNX-1351 was added at varying concentrations. DMSO was used as a negative control. The NanoLuc luciferase activity was measured at 24 h posttransfection.
## Antiviral activity of CNX-1351 on different cell lines
A549, HeLa, MEF, BHK-21, BV2 or Vero E6 cells were pretreated with CNX-1351 at varying concentrations for 1 h prior to infection with MP-12 VRPs. At 12 h post-infection, cells were fixed with 2% PFA, and virus infection was measured by flow cytometry or high-content imaging.
## RNA-seq and data analysis
Huh-7 cells seeded in 12-well plates were treated with 5 μM CNX-1351 for 12 h, and subjected to RNA extraction using RNA Prep Pure Cell/Bacteria Kit (Tiangen). Library preparation and sequencing was conducted by GeneWiz, using at least 0.5 μg high-quality total RNA input per sample. mRNA enrichment was performed using poly-T oligoattached magnetic beads. Sequencing libraries were prepared using the NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (NEB, USA). Libraries were sequenced on an Illumina Hiseq 2500 platform to generate paired-end reads.
Raw sequencing data were trimmed and filtered using Fastp software. Pre-processed data was aligned using HISAT2 and sorted using SAMtools. Gene read counts were obtained using featureCounts. Differential expression analysis was performed using the DESeq2 R package (v1.36.0), with adjusted P-values calculated using the Benjamini-Hochberg method to control the false discovery rate (FDR). Genes with an adjusted P-value < 0.05 were considered differentially expressed. Volcano plots and heatmaps were generated using the Enhanced Volcano and pheatmap packages in R. Gene Ontology (GO) enrichment and KEGG pathway analysis were performed using the ClusterProfiler R package (v4.0). Raw sequencing data are available at ScienceDB (https://doi.org/10.57760/sciencedb.24750).
## Statistical analysis
Statistical significance was determined using Prism Version 10 (GraphPad). P values < 0.05 were considered significant. Data was analyzed using ANOVA or unpaired t-test, depending on data distribution and the number of comparison groups.
## References
1. Al-Afaleq, Hussein (2011) "The status of Rift Valley fever in animals in Saudi Arabia: a mini review" *Vector Borne Zoonotic Dis*
2. Al-Hazmi, Ayoola, Abdurahman et al. (2003) "Epidemic Rift Valley fever in Saudi Arabia: a clinical study of severe illness in humans" *Clin. Infect. Dis*
3. Alrajhi, Al-Semari, Al-Watban (2004) "Rift Valley fever encephalitis" *emerging infectious diseases journal -CDC. Emerg. Infect. Dis*
4. Benedict, Bansal, Senina et al. (2015) "Repurposing FDA-approved drugs as therapeutics to treat Rift Valley fever virus infection" *Front. Microbiol*
5. Blanco, Cameirao, L� Opez et al. (2020) "Phosphatidylinositol-3kinase-Akt pathway in negative-stranded RNA virus infection: a minireview" *Arch. Virol*
6. Boer, Kortekaas, Spel et al. (2012) "Acid-activated structural reorganization of the Rift Valley fever virus Gc fusion protein" *J. Virol*
7. Burke, Erwin-Cohen, Goodson et al. (2022) "Efficacy of Western, Eastern, and Venezuelan equine encephalitis (WEVEE) virus-replicon particle (VRP) vaccine against WEEV in a nonhuman primate animal model" *Viruses*
8. Caplen, Peters, Bishop (1985) "Mutagen-directed attenuation of Rift Valley fever virus as a method for vaccine development" *J. Gen. Virol*
9. Daubney, Hudson (1931) "Enzootic hepatitis or Rift Valley fever. An un-described virus disease of sheep, cattle and man from East Africa" *J. Pathol. Bacteriol*
10. Daubney, Hudson (1932) "Rift Valley fever" *Lancet*
11. Dodd, Bird, Metcalfe et al. (2012) "Single-dose immunization with virus replicon particles confers rapid robust protection against Rift Valley fever virus challenge" *J. Virol*
12. Easterday (1965) "Rift valley fever" *Adv. Vet. Sci*
13. Ebogo-Belobo, Kenmoe, Abanda et al. (2023) "Contemporary epidemiological data of Rift Valley fever virus in humans, mosquitoes and other animal species in Africa: a systematic review and meta-analysis" *Vet. Med. Sci*
14. Edwards, Slater, Johnston (2007) "Signalling pathways mediating type I interferon gene expression" *Microb. Infect*
15. Gabrielsen, Monath, Huggins et al. (1992) "Antiviral (RNA) activity of selected Amaryllidaceae isoquinoline constituents and synthesis of related substances" *J. Nat. Prod*
16. Ganaie, Schwarz, Mcmillen et al. (2021) "Lrp1 is a host entry factor for Rift Valley fever virus" *Cell*
17. Gerrard, Rollin, Nichol (2002) "Bidirectional infection and release of Rift Valley fever virus in polarized epithelial cells" *Virology*
18. Gharbi, Zvelebil, Shuttleworth et al. (2007) "Exploring the specificity of the PI3K family inhibitor LY294002" *Biochem. J*
19. Habjan, Penski, Wagner et al. (2009) "Efficient production of Rift Valley fever virus-like particles: the antiviral protein MxA can inhibit primary transcription of bunyaviruses" *Virology*
20. Halfmann, Ebihara, Marzi et al. (2009) "Replication-deficient ebolavirus as a vaccine candidate" *J. Virol*
21. He, Qi, Wang et al. (2013) "Amaryllidaceae alkaloids inhibit nuclear-to-cytoplasmic export of ribonucleoprotein (RNP) complex of highly pathogenic avian influenza virus H5N1" *Influenza Other Respir Viruses*
22. Huggins (1989) "Prospects for treatment of viral hemorrhagic fevers with ribavirin, a broad-spectrum antiviral drug" *Rev. Infect. Dis*
23. Ikegami, Peters, Makino (2005) "Rift Valley fever virus nonstructural protein NSs promotes viral RNA replication and transcription in a minigenome system" *J. Virol*
24. Ikegami, Won, Peters et al. (2006) "Rescue of infectious rift valley fever virus entirely from cDNA, analysis of virus lacking the NSs gene, and expression of a foreign gene" *J. Virol*
25. Kortekaas, Oreshkova, Cobos-Jim� Enez et al. (2011) "Creation of a nonspreading Rift Valley fever virus" *J. Virol*
26. Lin, Chiao, Shu et al. (2023) "Development of a novel Chikungunya virus-like replicon particle for rapid quantification and screening of neutralizing antibodies and antivirals" *Microbiol. Spectr*
27. Linthicum, Britch, Anyamba (2016) "Rift Valley fever: an emerging mosquito-borne disease" *Annu. Rev. Entomol*
28. Mansfield, Banyard, Mcelhinney et al. (2015) "Rift Valley fever virus: a review of diagnosis and vaccination, and implications for emergence in Europe" *Vaccine*
29. Mcintosh, Russell, Santos et al. (1980) "Rift Valley fever in humans in South Africa" *South African medical journal = Suid-Afrikaanse tydskrif vir geneeskunde*
30. Meulen, Smets, Rüdelsheim (2023) "Viral replicon systems and their biosafety aspects" *Applied Biosafety*
31. Moy, Gold, Molleston et al. (2014) "Antiviral autophagy restricts Rift Valley fever virus infection and is conserved from flies to mammals" *Immunity*
32. Mudhasani, Kota, Retterer et al. (2014) "High content image-based screening of a protease inhibitor library reveals compounds broadly active against Rift Valley fever virus and other highly pathogenic RNA viruses"
33. Murakami, Terasaki, Ramirez et al. (2014) "Development of a novel, single-cycle replicable rift valley Fever vaccine"
34. Nacht, Qiao, Sheets et al. (2013) "Discovery of a potent and isoform-selective targeted covalent inhibitor of the lipid kinase PI3Kα" *J. Med. Chem*
35. Pepin, Bouloy, Bird et al. (2010) "Rift Valley fever virus (Bunyaviridae: Phlebovirus): an update on pathogenesis, molecular epidemiology, vectors, diagnostics and prevention" *Vet. Res*
36. Piper, Gerrard (2010) "A novel system for identification of inhibitors of rift valley Fever virus replication" *Viruses*
37. Riblett, Blomen, Jae et al. (2016) "A haploid genetic screen identifies heparan sulfate proteoglycans supporting Rift Valley fever virus infection" *J. Virol*
38. Scharton, Bailey, Vest et al. (2014) "Favipiravir (T-705) protects against peracute Rift Valley fever virus infection and reduces delayed-onset neurologic disease observed with ribavirin treatment" *Antivir. Res*
39. Schultze, Hemmings, Niessen et al. (2012) "PI3K/AKT, MAPK and AMPK signalling: protein kinases in glucose homeostasis" *Expet Rev. Mol. Med*
40. Shirbaghaee, Bolhassani (2016) "Different applications of virus-like particles in biology and medicine: vaccination and delivery systems" *Biopolymers*
41. Solomon, Lee (2011) "Quinoline as a privileged scaffold in cancer drug discovery" *Curr. Med. Chem*
42. Tan, Banwell, Ye et al. (2022) "The inhibition of RNA viruses by Amaryllidaceae alkaloids: opportunities for the development of broadspectrum anti-Coronavirus drugs" *Chem. Asian J*
43. Vanhaesebroeck, Stephens, Hawkins (2012) "PI3K signalling: the path to discovery and understanding" *Nat. Rev. Mol. Cell Biol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12746227&blobtype=pdf | # Coexistence of gynecological pathology with endometriosis and adenomyosis
Michail Matalliotakis, Ioannis Tsakiridis, Charoula Matalliotaki, Konstantinos Krithinakis, Demetrios Spandidos, Themistoklis Dagklis, Apostolos Mamopoulos, Ioannis Kalogiannidis
## Abstract
Both endometriosis and adenomyosis can impact quality of life. Endometriosis is a chronic, benign, condition characterized by the growth of endometrium outside the uterus, which affects ~10% of women from menarche to menopause, globally. Adenomyosis is a benign gynecological condition, which is defined as the infiltration of ectopic endometrial tissue within the underlying myometrium. It is often diagnosed in perimenopausal women aged 40-50 years, with a prevalence that varies widely (ranging from 1-70%). The aim of the present study was to determine the histological coexistence of gynecological pathology in women with endometriosis, adenomyosis and those with both. In the present retrospective study, data retrieved from the medical records of women that underwent abdominal hysterectomy with bilateral or unilateral salpingo-oophorectomy at the
## Introduction
Endometriosis is a chronic benign systemic proinflammatory condition, which is characterized by the growth of endometrium (glands and stroma) outside the uterus, mostly in the ovaries, followed by the soft tissue, gastrointestinal and urinary tract. Globally, it affects 7-10% of women with an incidence of 38% in infertile women and 71-87% in women with chronic pelvic pain (1,2). Histologically endometriosis is classified as having a typical and atypical form. Atypical endometriosis indicates a premalignant lesion and is characterized by areas of marked cellular atypia with morphologic features similar to those of clear cell carcinoma. By contrast, typical endometriosis is a benign condition that involves endometrial-like tissue growing outside the uterus (3,4).
Adenomyosis is a benign uterine condition, which is characterized by the infiltration of ectopic endometrial tissue (glands and stroma) within the underlying myometrium (5,6). Depending on the distribution within the myometrium, adenomyoma can be classified as focal, diffuse or polypoid adenomyoma (5,6). Adenomyosis appears in late childbearing years (40-50 years of age). It is accompanied by menorrhagia in 40-50% of cases, by dysmenorrhoea that does not respond to the standard anti-inflammatory treatment (such as non-steroidal anti-inflammatory drugs) in 15-78% of cases and infertility in 11-12% of cases. The global prevalence of adenomyosis among patients after a hysterectomy varies from 5-70 % (5,6). However, European estimates using imaging (such as ultrasound and MRI) suggest a prevalence of 20-35%, but the data are limited (6). In 1927, the study by Sampson (7) coined the term endometriosis. Furthermore, in 1972, a study by Bird (8) defined adenomyosis as the benign invasion of endometrium into the myometrium, surrounded by the hypertrophic and hyperplastic myometrial tissue. Although, adenomyosis was previously called 'endometriosis interna', a form of endometriosis, it is well documented that both diseases are distinct. Endometriosis is considered to form from the functional endometrium and adenomyosis is considered to form from the basal endometrium. Various etiological factors, such as genetic, endocrine, autoimmune, environmental and anatomical factors, serve a role in the development of both entities (1,2,(6)(7)(8)(9). It is hypothesized that retrograde menstruation is the primary mechanism for endometriosis and that local invasion, cellular proliferation and angiogenesis are mechanisms for adenomyosis (1,2,6,7,9). Furthermore, patients with breast cancer that are treated with tamoxifen are at a distinct risk of endometriosis, adenomyosis and several other gynecological conditions such as endometrial polyps, hyperplasia and uterine cancer (10).
Gynecological diseases such as adenomyosis, leiomyomas, endometriosis, endometrial polyps, hyperplasia, endometrial and ovarian cancer often coexist, which suggests that they may share common pathogenic mechanisms such as the presence of hyperestrogenism, inflammatory, environmental and genetic altered features (1,5,11,12).
The aim of the present study was to determine the histological frequency of various benign and malignant gynecological diseases among women with endometriosis, adenomyosis or both that underwent gynecological surgery.
## Materials and methods
Patients. Women (aged ≥18 years) that underwent abdominal hysterectomy with bilateral or unilateral salpingooophorectomy with a clear medical record documentation, at the Department of Obstetrics and Gynecology of Venizeleio General Hospital (Heraklion, Greece) in January 2000 to December 2022 and the Third Department of Obstetrics and Gynecology of Aristotle University of Thessaloniki (Thessaloniki, Greece) in January 2009 to December 2019, were included in the present study. Data were collected retrospectively and included the indication for gynecological operation, age, clinical and histopathological findings.
Patients were divided into three groups, namely group 1 (which included women with endometriosis), 2 (which included women with adenomyosis) and 3 (which included patients with both endometriosis and adenomyosis). All data were retrieved from the medical records. Cases with inadequate medical records and those lacking histological diagnoses were excluded. Patient consent for participation was waived by the ethics committees due to the retrospective nature of the present study. The Ethics Committees of Venizeleio General Hospital (approval no. 124-17-18/12/2019; Heraklion, Greece) and Aristotle University of Thessaloniki (approval no. 1130-03/07/2020; Thessaloniki, Greece) both approved the protocol of the present study.
Statistical analysis. The descriptive statistics are presented as mean ± standard deviations or as frequencies and percentages. Associations between categorical variables was estimated using the Chi-square test of independence. The parameter 'age' was analyzed using one-way ANOVA followed by Tukey's post hoc test. Multinomial logistic regression was carried out to detect factors associated with pathology. Statistical analysis was carried out using IBM SPSS Statistics (version 25.0; IBM Corp.).
## Results
Eligible population. In total, 3,511 women underwent abdominal hysterectomy with bilateral or unilateral salpingo-oophorectomy at the Department of Obstetrics and Gynecology of Venizeleio General Hospital (Heraklion, Greece) and the Third Department of Obstetrics and Gynecology of Aristotle University of Thessaloniki (Thessaloniki, Greece) in the study period. However, of these 3,511 women, only 1,692 cases were eligible for inclusion in the present study, having endometriosis, adenomyosis or both conditions. The majority of cases had adenomyosis (n=1,096; 64.8%; group 2), followed by endometriosis (n=495; 29.3%; group 1), while 101 cases had both endometriosis and adenomyosis (6.0%; group 3). The mean age of the total study population was 50.16±12.74 years (range, 16-85 years). However, the mean age of patients with adenomyosis was significantly higher (55.34±10.21) compared with those of the other two groups.
Reported pathologies among the groups. In the total sample of 1,692 women, the most commonly noted pathology was leiomyomas (43.3%), followed by endometrial polyps (17.6%). Benign ovarian (12.2%) and paraovarian cysts (13.7%), and cervical polyps (9.8%) were also frequently observed. The occurrence of leiomyomas was significantly more frequent in patients in group 3 (67.3%) compared with those in groups 1 (24.2%) or 2 (49.6%). A similar result was observed for the frequency of ovarian cysts, which were more common in the patients in group 3 compared with those in groups 1 or 2 (Table I).
Among gynecological malignancies, endometrial cancer was significantly more frequent in patients in group 2 (9.3%), compared with those in groups 1 (1.6%) or 3 (4.0%) (Table I). However, further analysis with multinomial logistic regression did not identify adenomyosis as a factor associated with endometrial cancer (P=0.426).
## Discussion
The present study demonstrated that both endometriosis and adenomyosis can co-exist with various benign, premalignant or malignant gynecological diseases such as leiomyomas, endometrial and cervical polyps, and benign ovarian and paraovarian cysts. Regarding gynecological malignancy, adenomyosis was more commonly associated with endometrial cancer compared with cervical or ovarian cancer.
Uterine fibroids are mainly detected in women aged 30-50 years old and are due to the effects of estrogen and progesterone levels and their metabolism. Under in vitro and in vivo conditions, uterine fibroids and endometriotic ectopic cells express aromatase, resulting in increased concentrations of estrogen in the tissues (13). It is considered that an enlarged fibroid mass can change the shape, size and location of the uterus within the pelvic cavity and that this may result in retrograde menstruation and endometriosis (14,15). Our previous study confirms that fibroids coexist commonly with endometriosis in perimenopausal women, whereas adenomyosis is more prominent in postmenopausal women (16). In patients with leiomyomas, the reported prevalence of adenomyosis varies from 15-60% in women undergoing hysterectomy or another gynecological operation. Furthermore, 70% of women in the general population will acquire leiomyomas at any given point during their reproductive years, the majority of which occur between the ages of 30-50 years (6,17). In the present study, it was confirmed that cases of leiomyomas are more common in patients with endometriosis and adenomyosis compared with patients with endometriosis or adenomyosis.
According to the literature, endometrial hyperplasia and polyps are mostly observed during reproductive years and in perimenopause and result from hormonal alternations, overexpression of aromatase, inflammation, gene mutations and monoclonal cell hyperplasia (18,19). Women with a diagnosis of pre-existing endometriosis (of at least 12 months), with no other diagnosed pathology, were studied in a large population-based observational study. The possible occurrence of endometrial cancer or hyperplasia was studied with a mean follow-up of 6.4 years. Women with endometriosis were at greater risk of developing endometrial hyperplasia (1.85 times) and endometrial cancer (1.35 times) compared with women without endometriosis (20). Regarding polyps, a study by McBean et al (21) was the first to identify an association between endometriosis and polyps. A study by Huang and Xiang (22) suggests that endometrial polyps may be due to reactive endometrial hyperplasia due to long-term repetitive mechanical stimulation and the presence of biological inflammatory factors, such as VEGF and TGFβ. In a study by Zhang et al (23), it is hypothesized that endometrial inflammation indicates the common mechanism leading to the coexistence of endometriosis with endometrial polyps. Additionally, in the aforementioned meta-analysis it is observed that the risk of coexistence is prominent in stages 2-4 of endometriosis. However, a study by Zheng et al (24) considers that the frequency of coexistence does not differ notably according to the stage of endometriosis. Endometrial polyps were found in 46.7% of cases with endometriosis and in 16.5% of the control group (24). The study by Kim et al (25), due to a high risk of co-occurrence, recommends hysteroscopy in infertile women with endometriosis, even if medical imaging is not suggestive of endometrial polyps.
In terms of adenomyosis, a retrospective study by Dayal and Nagrath (26), using 353 adenomyosis histological confirmed specimens, demonstrates the co-existence of endometrial hyperplasia in 12.46% and uterine polyps in 7.64% of cases. Additionally, a study by Tetikkurt et al (27) confirmed 71 cases with endometrial polyps, 9 cases with hyperplasia and concurrent presence of endometrial polyps and hyperplasia in 6 women out of a total of 319 cases with a histopathological diagnosis of adenomyosis. However, a study by Bergholt et al (28) observes that the presence of endometrial hyperplasia at the time of operation is the only variable associated with adenomyosis, as was also demonstrated in the present study. In a study by Baskin et al, six adult female monkeys are ovariectomized and receive subcutaneous implants containing 200 mg estradiol. The subsequent necropsy reveals the coexistence of endometrial hyperplasia, polyps or adenomyosis; thus, hyperestrogenism may result in various gynecological conditions (29). In the present study, endometrial and cervical polyps were significantly detected in the patients with adenomyosis and the patients with both endometriosis and adenomyosis, respectively.
With regards to benign ovarian growth, endometriosis can coexist with various benign and premalignant ovarian masses such as serous/mucinous cystadenomas, borderline ovarian tumors and teratomas (11,30). A retrospective study by Oral et al (31), using 530 cases with ovarian cancer and 131 women with borderline ovarian cancer, reveals that concomitant endometriosis occurs in 7.3% of cases. However, our previous study recently concludes that further research is required to investigate the pathway through which ovarian teratomas are negatively associated with the risk of developing endometriosis (32). A previous case series study by Vercellini et al (33) confirms the coexistence of adenomyosis and ovarian cysts in 21.4% of cases. Moreover, a previous prospective cross-sectional study, out of 100 hysterectomy cases, reveals the coexistence of adenomyosis with non-cancerous ovarian cyst in 25% of cases (34). Our previous study retrospectively investigates the medical records of 647 patients with adenomyosis in order to examine a possible association between adenomyosis and benign, premalignant and malignant gynecological diseases in women who underwent gynecological surgery. The aforementioned study demonstrates that adenomyosis co-exists with various gynecological conditions, such as leiomyomas and endometrial polyps, but it did not report a prominent association of adenomyosis with other gynecological pathologies compared with the general population (17). However, a paraovarian or paratubal cyst (POC), a mass located in the broad ligament or the mesosalpinx, has a high prevalence (5-20%) amongst the adnexal masses. In the present study, benign ovarian cysts and POC were prominent in group 3, followed by group 2. Due to hormonal alternations, they are mainly found during reproductive age; thus, they can coexist with endometriosis or adenomyosis (11,35). POC can result in disturbed tubal mobility, a condition that may result in the presence of endometriosis (36,37). The present study found 101 cases with concurrent endometriosis and adenomyosis. In 2018, a retrospective study, including 1,000 cases with a histological confirmation of endometriosis, concludes that women with endometriosis should receive counseling regarding the risk of accompanying adenomyosis (11). Moreover, in 2019, our previous study retrospectively demonstrated the coexistence of adenomyosis with endometriosis in 16% of women in perimenopause and in 32.6% of women in postmenopause (16).
Although rarely, both endometriosis and adenomyosis may exhibit cancerous behavior or coexist with gynecological malignancy. Prolonged exposure to estrogens, genetic defects, environmental factors, immune dysregulation and oxidative stress are possible pathogenic mechanisms (38)(39)(40)(41)). Sampson's criteria are used for both adenomyosis and endometriosis malignant transformation (42,43). Neoplasms that develop from endometrial ectopic implants in endometriosis are divided into two categories; ovarian, which accounts for 75% of the cases described in the literature and extra ovarian such as endometrial, cervical and breast cancer (41). Cases of ovarian cancer due to endometriosis have unique features, such as endometrioid carcinoma or clear cell carcinoma, the diagnosis is usually made earlier compared with other types of ovarian cancer and the prognosis is better (44,45). Although malignant transformation may occur in 1% of cases with ovarian endometriosis, there is a strong link between endometriosis and ovarian cancer. In particular, women with endometriosis have a 4.2-fold higher risk of ovarian cancer compared with the general population. Therefore, it is recommended to create screening and prevention programs for this population in the future (46)(47)(48). The study by Saavalainen et al (49) concludes that the type of endometriosis a woman has may determine the risk of gynecological cancer. The risk of ovarian cancer was highest among women with ovarian endometriosis but low among cases with peritoneal and deep infiltrating endometriosis (49). Previously, a large retrospective study reports that endometrial cancer with concomitant endometriosis is highly associated with ovarian endometrioid carcinoma (50). Furthermore, epidemiological data show that the risk of developing endometrioid carcinoma of the endometrium doubles or even triples in cases with endometriosis (51,52). However, a study by Poole et al (53) demonstrates that there is not a strong association between endometriosis and endometrial cancer, and this indicates that both conditions follow a different molecular pathogenesis. Our previous study retrospectively confirms that women with cervical endometriosis are at greater risk of developing ovarian, endometrial and cervical cancer compared with women without endometriosis (54). Another large population-based historical cohort study confirms that, compared with cases not acquiring the conditions, patients with adenomyosis are at higher risk of endometrial and thyroid cancer, while women with endometriosis are at greater risk of endometrial and ovarian cancer (55).
However, a large pilot study did not reveal a notable difference in terms of epidemiologic, clinicopathologic and prognostic characteristics in patients with endometrial cancer and adenomyosis compared with patients with only endometrial cancer (56). In addition, a systematic review and meta-analysis by Raffone et al concludes that the pathology of adenomyosis and endometrial cancer are two unassociated entities (57). Recently, for the first time, a large retrospective study by Yang et al (58) describes a possible protective mechanism of adenomyosis against cervical cancer invasion, a phenomenon that is also previously discussed in a study by Machida et al (39) for endometrial cancer. The findings of the present study indicated endometrial cancer in patients with adenomyosis; however, further analysis, with multinomial logistic regression, did not identify adenomyosis as a factor associated with endometrial cancer.
The present study had a number of limitations. Firstly, there was a lack of long-term follow-up data, which meant that distinguishing cases with concurrent gynecological malignancy or malignant transformation of adenomyosis and/or endometriosis was limited. Moreover, the retrospective aspect of the present study was a limitation. However, strengths of the present study include the histological confirmation of all cases and the large number of patients that were included in the present study.
In conclusion, the present study demonstrated that endometriosis, adenomyosis or the concomitant presence of both conditions may coexist with various benign and malignant gynecological diseases. Furthermore, patients with adenomyosis may be associated with endometrial cancer. However, additional longitudinal studies are required in order to generalize the findings of the present study in other populations.
## References
1. Cano-Herrera, Nehmad, De Chávez Gascón et al. (2024) "Endometriosis: A comprehensive analysis of the pathophysiology, treatment, and nutritional aspects, and its repercussions on the quality of life of patients" *Biomedicines*
2. Vercellini, Viganò, Somigliana et al. (2014) "Endometriosis: Pathogenesis and treatment" *Nat Rev Endocrinol*
3. Bedaiwy, Hussein, Biscotti et al. (2009) "Pelvic endometriosis is rarely associated with ovarian borderline tumours, cytologic and architectural atypia: A clinicopathologic study" *Pathol Oncol Res*
4. Wilbur, Shih, Segars et al. (2017) "Cancer implications for patients with endometriosis" *Semin Reprod Med*
5. Moawad, Fruscalzo, Youssef et al. (2023) "Adenomyosis: An updated review on diagnosis and classification" *J Clin Med*
6. Upson, Missmer (2020) "Epidemiology of Adenomyosis" *Semin Reprod Med*
7. Sampson (1927) "Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity" *Am J Obstet Gynecol*
8. Bird, Mcelin (1972) "The elusive adenomyosis of the uterus-revisited" *Am J Obstet Gynecol*
9. Guo (2020) "The pathogenesis of Adenomyosis vis-à-vis endometriosis" *J Clin Med*
10. Polin, Ascher (2008) "The effect of tamoxifen on the genital tract" *Cancer Imaging*
11. Matalliotaki, Matalliotakis, Ieromonachou et al. (2018) "Co-existence of benign gynecological tumors with endometriosis in a group of 1,000 women" *Oncol Lett*
12. Szubert, Kozirog, Wilczynski (2022) "Adenomyosis as a risk factor for myometrial or endometrial neoplasms-review" *Int J Environ Res Public Health*
13. Bulun, Lin, Imir et al. (2005) "Regulation of aromatase expression in estrogen-responsive breast and uterine disease: From bench to treatment" *Pharmacol Rev*
14. Lin, Yang, Lam et al. (2021) "Uterine leiomyoma is associated with the risk of developing endometriosis: A nationwide cohort study involving 156,195 women" *PLoS One*
15. Hemmings, Rivard, Olive et al. (2004) "Evaluation of risk factors associated with endometriosis" *Fertil Steril*
16. Matalliotakis, Matalliotaki, Trivli et al. (2019) "Keeping an Eye on Perimenopausal and postmenopausal endometriosis" *Diseases*
17. Matalliotakis, Zervou, Matalliotaki et al. (2022) "There is no significant correlation of adenomyosis with benign, premalignant and malignant gynecological pathologies. Retrospective study on 647 specimens" *Ginekol Pol*
18. Lacey, Vm (2009) "Endometrial hyperplasia and the risk of progression to carcinoma" *Maturitas*
19. Indraccolo, Iorio, Matteo et al. (2013) "The pathogenesis of endometrial polyps: A systematic semi-quantitative review" *Eur J Gynaecol Oncol*
20. Kim, Kim, Ahn (2023) "Does endometriosis increase the risks of endometrial hyperplasia and endometrial cancer?" *Gynecol Oncol*
21. Mcbean, Gibson, Brumsted (1996) "The association of intrauterine filling defects on hysterosalpingogram with endometriosis" *Fertil Steril*
22. Huang, Xiang (2014) "Recent advances in endometrial polyps" *J Int Obstet Gynecol*
23. Zhang, Zhang, Yu et al. (2018) "Higher prevalence of endometrial polyps in infertile patients with endometriosis" *Gynecol Obstet Invest*
24. Zheng, Mao, Zhao et al. (2015) "Risk of endometrial polyps in women with endometriosis: A metaanalysis" *Reprod Biol Endocrinol*
25. Kim, Kim, Jo et al. (2003) "High frequency of endometrial polyps in endometriosis" *J Am Assoc Gynecol Laparosc*
26. Dayal, Nagrath (2015) "Cross sectional study. Pattern and frequency of endometrial and ovarian pathologies with adenomyosis uteri in patients who attended the tertiary care hospital among rural population of North India" *MAMC J Med Sci*
27. Tetikkurt, Çelik, Taş et al. (2018) "Coexistence of adenomyosis, adenocarcinoma, endometrial and myometrial lesions in resected uterine specimens" *Mol Clin Oncol*
28. Bergholt, Eriksen, Berendt et al. (2001) "Prevalence and risk factors of adenomyosis at hysterectomy" *Hum Reprod*
29. Baskin, Smith, Marx (2002) "Endometrial hyperplasia, polyps, and adenomyosis associated with unopposed estrogen in rhesus monkeys (Macaca mulatta)" *Vet Pathol*
30. Matalliotakis, Matalliotaki, Zervou et al. (2019) "Retrospective evaluation of pathological results among women with ovarian endometriomas versus teratomas" *Mol Clin Oncol*
31. Oral, Aydin, Kumbak et al. (2018) "Concomitant endometriosis in malignant and borderline ovarian tumours" *J Obstet Gynaecol*
32. Matalliotakis, Matalliotaki, Tsakiridis et al. (2024) "Co-existence of ovarian teratomas with other gynecological tumors" *Cureus*
33. Vercellini, Parazzini, Oldani et al. (1995) "Adenomyosis at hysterectomy: A study on frequency distribution and patient characteristics" *Hum Reprod*
34. Alam, Ahmad, Khan et al. (2016) "Role of benign ovarian cysts in the development of adenomyosis" *Saudi Med J*
35. Singh, Agarwal, Begum et al. (2023) "The burden of paraovarian cysts-a case series and review of the literature" *Prz Menopauzalny*
36. Durairaj, Gandhiraman (2019) "Complications and management of paraovarian cyst: A retrospective analysis" *J Obstet Gynecol India*
37. Audebert, Timmermans, Lismonde et al. (2019) "Fallopian tube and endometriosis: An ambiguous relationship" *European Gynecology and Obstetrics*
38. Habiba, Pluchino, Petignat et al. (2018) "Adenomyosis and endometrial cancer: Literature review" *Gynecol Obstet Invest*
39. Machida, Maeda, Cahoon et al. (2017) "Endometrial cancer arising in adenomyosis versus endometrial cancer coexisting with adenomyosis: Are these two different entities?" *Arch Gynecol Obstet*
40. Moraru, Mitranovici, Chiorean et al. (2023) "Adenomyosis and its possible malignancy: A review of the literature"
41. Stern, Dash, Bentley et al. (2001) "Malignancy in endometriosis: Frequency and comparison of ovarian and extraovarian types" *Int J Gynecol Pathol*
42. Sampson (1925) "Endometrial carcinoma of the ovary, arising in endometrial tissue in that organ" *Arch Surg*
43. Koike, Tsunemi, Uekuri et al. (2013) "Pathogenesis and malignant transformation of adenomyosis (review)" *Oncol Rep*
44. Komiyama, Aoki, Tominaga et al. (1999) "Prognosis of Japanese patients with ovarian clear cell carcinoma associated with pelvic endometriosis: Clinicopathologic evaluation" *Gynecol Oncol*
45. Matalliotakis, Matalliotaki, Goulielmos et al. (2018) "Association between ovarian cancer and advanced endometriosis" *Oncol Lett*
46. Barnard, Farland, Yan et al. (2024) "Endometriosis typology and ovarian cancer risk" *JAMA*
47. Wei, Bulun (2011) "Endometriosis and ovarian cancer: A review of clinical, pathologic, and molecular aspects" *Int J Gynecol Pathol*
48. Gounaris, Ds, Jd (2011) "Ovarian clear cell Carcinoma-bad endometriosis or bad endometrium?" *J Pathol*
49. Saavalainen, Lassus, But et al. (2018) "Risk of gynecologic cancer according to the type of endometriosis" *Obstet Gynecol*
50. Ishizaka, Taguchi, Tsuruga et al. (2022) "Endometrial cancer with concomitant endometriosis is highly associated with ovarian endometrioid carcinoma: A retrospective cohort study" *BMC Womens Health*
51. Mogensen, Kjaer, Mellemkjaer et al. (2016) "Endometriosis and risks for ovarian, endometrial and breast cancers: A nationwide cohort study" *Gynecol Oncol*
52. Yu, Lin, Chang et al. (2015) "Task Force on Carcinogenesis of Endometrial Cancer: Increased association between endometriosis and endometrial cancer: A nationwide population-based retrospective cohort study" *Int J Gynecol Cancer*
53. Poole, Lin, Kvaskoff et al. (2017) "Endometriosis and risk of ovarian and endometrial cancers in a large prospective cohort of U.S. nurses" *Cancer Causes Control*
54. Matalliotakis, Matalliotaki, Zervou et al. (2021) "Coexistence of cervical endometriosis with premalignant and malignant gynecological pathologies: Report on a series of 27 cases" *Women Health*
55. Yeh, Su, Tzeng et al. (2018) "Women with adenomyosis are at higher risks of endometrial and thyroid cancers: A population-based historical cohort study of Taiwan" *PLoS One*
56. Chao, Wu, Ma et al. (2020) "The clinicopathological characteristics and survival outcomes of endometrial carcinoma coexisting with or arising in adenomyosis: A pilot study" *Sci Rep*
57. Raffone, Seracchioli, Raimondo et al. (2021) "Prevalence of adenomyosis in endometrial cancer patients: A systematic review and meta-analysis" *Arch Gynecol Obstet*
58. Yang, Chen, Hua et al. (2024) "Association between adenomyosis and cervical cancer: A retrospective cohort study" *J Invest Surg* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12273695&blobtype=pdf | # Correction: Mbigha Donfack et al. Aedes Mosquito Virome in Southwestern Cameroon: Lack of Core Virome, But a Very Rich and Diverse Virome in Ae. africanus Compared to Other Aedes Species. Viruses 2024, 16, 1172
Karelle Celes, Mbigha Donfack, Lander De Coninck, Stephen Ghogomu, Jelle Matthijnssens
## Error in Figure 1 Position
In the original publication [1], there was a mistake in the figure position as published. Figure position is in the Introduction section. The corrected figure position is in the Results section after Section 3.1 paragraph, which appears below. The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.
Here is the corresponding text from the Results section where Figure 1 should be placed:
## 3. Results
## 3.1. Each Sampling Site is Dominated by a Single Aedes Species
This study focuses on analyzing the eukaryotic virome of the Aedes mosquito from four regions in southwestern Cameroon. In 2020, a total of 398 Aedes mosquitoes were captured from Bafoussam (n = 101), Edea (n = 96), Buea (n = 96) and Yaounde (n = 105) (Figure 1). Analyses of the mosquito distribution revealed that, Ae. albopictus species predominated in Edea, Buea, and Yaounde, while Ae. africanus was the most prevalent in Bafoussam. Small numbers of Ae. simpsoni were also captured, co-existing in regions predominated by Ae. albopictus. Only a single Ae. aegypti mosquito was captured in Edea (Figure 1).
## Error in Figure 1 Legend
In the original publication, there was a mistake in the figure legend as published. Pink regions represent Equatorial monsoon zones and Blue regions represent Equatorial mountain monsoon zones. The corrected figure legend, Pink regions represent Equatorial mountain monsoon zones and blue regions represent Equatorial monsoon zones, appears above. The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.
## Error in Figure 6 Text Language
In the original publication, there was a mistake in the text language in the figure as published. The text in figure is in non-English. The corrected text language in the figure appears below. The text in the figure is in English. The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.
## References
1. Mbigha Donfack, De Coninck, Ghogomu et al. (1172) "Aedes Mosquito Virome in Southwestern Cameroon: Lack of Core Virome, But a Very Rich and Diverse Virome in Ae. africanus Compared to Other Aedes Species" *Viruses*
2. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12714075&blobtype=pdf | # Molecular Characterization and Genomic Diversity of SARS-CoV-2 Spike Gene Variants Circulating in Iraq: Mutational Impact on ACE2 Affinity, RBD Immune Escape, and Viral Transmission
Anfal Khudhair, Duaa Abdulsatar, Sahar Hatif, Dunya Ridha, Radwan Munim, Ali, Jaafar Alsadiq, Arkan Farhan
## Abstract
Introduction:Te spike (S) gene of SARS-CoV-2 is pivotal to the processes of cell entry, immune evasion, and the adaptation of the host. Aim: Tis study aimed to comprehensively characterize the SARS-CoV-2 spike gene variants circulating in Iraq and assess the functional consequences of their mutations on ACE2 receptor afnity, RBD-mediated immune escape, and viral transmissibility. It represents the frst integrative genomic and functional profling of Iraqi SARS-CoV-2 spike variants, providing novel regional insights into viral adaptation and evolution. Methods: Whole-genome sequencing was performed on Iraqi SARS-CoV-2 isolates, followed by mutation profling, phylogenetic classifcation, and comparison with global datasets. Key spike mutations-N501Y, P681R, D614G, and E484K-were analyzed to assess their structural and functional implications. Results: Iraqi isolates clustered mainly within the Delta (21J) and 20A lineages. Te mutations N501Y (91.7%), P681R (75%), and D614G (100%) were prevalent, enhancing viral binding and transmission, while E484K was absent, suggesting limited immune escape compared to Omicron-like variants. Conclusion: Te absence of E484K and the predominance of transmission-enhancing mutations indicate that Iraqi SARS-CoV-2 isolates favor adaptation through increased ACE2 afnity rather than extensive immune evasion. Tese fndings underscore the importance of regional genomic surveillance to inform vaccination strategies and public health responses.
## 1. Introduction
Te SARS-CoV-2 spike (S) gene encodes the spike glycoprotein, a crucial structural protein that facilitates viral entry into host cells by binding to the angiotensin-converting enzyme 2 (ACE2) receptor. Te receptor-binding domain (RBD) within the spike protein plays a key role in viral infectivity, as mutations in this region can alter ACE2binding afnity, enhance viral transmissibility, and promote immune evasion. Tese mutations, particularly in the dual-purpose (DP) protein regions of the spike, may contribute to increased viral ftness and resistance to neutralizing antibodies, afecting the efcacy of vaccines and therapeutic interventions [1,2].
SARS-CoV-2 has evolved through multiple variants of concern (VOCs) that have shaped the course of the pandemic. Te Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), and Omicron (B.1.1.529) variants have exhibited distinct mutations in the S gene, infuencing their ability to bind ACE2, evade immune responses, and enhance viral transmission [3,4]. Notably,
• N501Y, found in Alpha, Beta, and Omicron, enhances ACE2 receptor-binding and contributes to increased transmissibility. • E484K, present in Beta, Gamma, and Omicron, reduces neutralizing antibody binding, allowing for immune escape. • P681R, a hallmark of the Delta variant, is associated with enhanced viral entry and increased transmission rates. • D614G, the frst globally dominant mutation, has been linked to improved viral ftness and higher infectivity across all major variants.
Te evolution of S gene mutations raises signifcant concerns regarding vaccine efcacy and reinfection risks. While mutations like N501Y and P681R improve viral spread, mutations such as E484K can signifcantly reduce the efectiveness of neutralizing antibodies, thereby facilitating reinfection even in previously immunized individuals [5,6]. Continuous genomic surveillance is crucial to monitoring these mutations and their impact on public health measures.
1.1. Study Objectives. Te aim of this study is to describe the S gene mutations in SARS-CoV-2 isolates from Iraq, evaluating their potential impact on 1. ACE2 receptor-binding afnity, which infuences viral entry efciency. 2. RBD neutralizing antibody escape, which determines immune evasion capacity. 3. Viral transmission efciency, afecting the spread of the virus in the population.
By correlating these mutations with global datasets, we will determine whether Iraqi SARS-CoV-2 isolates conform to established evolutionary patterns or exhibit unique regional variations. Tis will help to assess whether Iraq is witnessing the emergence of novel variants with potential implications for vaccine efectiveness and future outbreaks. General methodological references are WHO COVID-19 laboratory guidance and CONSORT/MIxS sampling metadata guidelines.
## 2. Materials and Methods
## 2.1. Ethical Approval and Sample
## 2.2. RNA Extraction and RT-PCR
• RNA extraction: Viral RNA was extracted using the QIAamp Viral RNA Mini Kit (QIAGEN, Hilden, Germany).
## 2.5. Statistical Analysis
• Descriptive statistics (median, ranges, and proportions) were reported for demographic and sequencing features. • Mutation prevalence was compared between Iraqi isolates and global datasets using
• Chi-square test (for n ≥ 5 per category).
• Fisher's exact test (for small counts).
• False discovery rate (FDR) correction was applied using the Benjamini-Hochberg method (α � 0.05). • R software (Version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria) along with the ggplot2 package (Wickham, 2016) was used for all statistical analyses and data visualization.
## 3. Results
## 3.1. Key Spike Mutations in Iraqi
Isolates. Analysis of the SARS-CoV-2 spike (S) gene revealed several critical mutations with functional relevance to viral entry and immune escape. Te N501Y substitution was identifed in Iraqi isolates and is known to enhance ACE2 receptor binding and contribute to immune evasion, consistent with its presence in Alpha, Beta, and Omicron variants [1,2]. Te P681R mutation, located adjacent to the furin cleavage site, was also detected and is a defning hallmark of the Delta lineage, facilitating more efcient viral entry [3,4]. Te D614G mutation, present in all Iraqi samples, is globally dominant due to its enhancement of viral ftness and transmission [7,8]. Notably, the E484K mutation, strongly associated with antibody resistance, was absent from the Iraqi dataset [5,6].
Tese fndings highlight the predominance of transmissibility-associated mutations (N501Y, P681R, and D614G) in Iraq, while underscoring the absence of E484K, a major.
## 3.2. Functional Consequences of Detected Mutations.
Analysis of the spike (S) gene in Iraqi SARS-CoV-2 isolates revealed the presence of several globally relevant mutations with variable functional implications. As summarized in Table 1, the mutations N501Y, P681R, and D614G were detected at high frequencies, consistent with their established roles in enhancing ACE2 receptor binding, viral entry, and transmissibility. In contrast, the immune escapeassociated mutation E484K was notably absent, suggesting a reduced potential for antibody evasion compared to global datasets. Te comparative prevalence of these mutations between Iraqi isolates and international reference sequences is further illustrated in Figure 1, which highlights the signifcant underrepresentation of E484K in the Iraqi cohort (p < 0.0001).
Te table captures prominent spike protein mutations of SARS-CoV-2 with the potential to alter its infectivity and immune evasion capabilities, concentrating on their occurrence within Iraqi isolates and in relation to global VOCs:
## Mutation
## Amino acid change
## Efect on ACE2 afnity
## Efect on RBD immune escape
## Detected in
Iraqi isolates?
## Frequency in
Iraqi isolates (%)
## Global variant association
## N501Y
Asn ⟶ Tyr High (↑ ACE2 binding) [1,2] High (↑ immune escape) [1,2] Yes 91.7% (11/12) Alpha, Beta, Omicron
## P681R
Pro ⟶ Arg Moderate (↑ viral entry) [3,4] Low (minor escape) [3] Yes 75% (9/12)
## Delta
## E484K
Glu ⟶ Lys Low (slight ACE2 increase) [5] Strong (↑ antibody evasion) [5,6] No 0% (0/12) Beta, Gamma, Omicron
## D614G
Asp ⟶ Gly No direct ACE2 efect [7] Moderate (↑ infectivity) [7,8] Yes 100% (12/12)
## All major variants
Note: Frequencies and functional implications of key SARS-CoV-2 spike (S) mutations identifed in Iraqi isolates (n � 12) are compared with global variant associations. Mutations N501Y, P681R, and D614G were detected at high frequencies in Iraqi samples, whereas the immune-escape mutation E484K was absent. Functional efects are summarized in terms of ACE2 receptor binding, immune escape potential, and known global variant lineages.
immune evasion by diminishing neutralization by antibodies. It is absent in Iraqi isolates but present in Beta, Gamma, and Omicron variants [5,6].
D614G: It substitutes aspartic acid (Asp) with glycine (Gly) at Position 614. While it does not directly impact ACE2 binding, it is correlated with greater viral infectivity and global spread, becoming the most dominant mutation worldwide. It is found in Iraqi isolates and exists in nearly all major variants [7,8] as further depicted in Figure 1.
Te bar chart illustrates the relative prevalence (%) of four key spike (S) protein mutations-N501Y, P681R, E484K, and D614G-in SARS-CoV-2 isolates sequenced from Iraq (blue) compared with globally reported variants (orange) (Table 2).
• N501Y is detected in ∼90% of Iraqi isolates versus ∼98% of global strains and enhances ACE2 receptor binding and immune escape. Figure 2 depicts the phylogenetic tree showing the distribution of SARS-CoV-2 clades (1-13) from Iraq and representative global isolates. Te color gradient represents the predicted ACE2 afnity score, ranging from -3.62 (blue, low binding afnity) to +2.21 (red, high binding affnity). Te major global lineages, including 20A, 20B, 21J (Delta), 21K (Omicron), and their sublineages, are labeled for reference. Iraqi isolates cluster predominantly within Delta (21J) and Omicron-related (21K) branches, refecting two major transmission waves. Clades 1-7 are primarily associated with Delta-derived sequences showing moderate ACE2 afnity (-1.6 to 0.9), while Clades 8-13 align with Omicron-like variants exhibiting higher binding potential (+1.5 to +2.2). Tis distribution highlights ongoing viral adaptation through increased receptor binding efciency across evolutionary transitions. All of these fndings are illustrated in Figure 1.
Te phylogenetic analysis illustrates the evolutionary relationships among 13 Iraqi SARS-CoV-2 clades compared with representative global reference lineages. Te color gradient represents the ACE2 receptor-binding affnity score, ranging from -3.62 (blue, low afnity) to +2.21 (red, high afnity) (Figure 3).
Iraqi isolates primarily cluster within the Delta (21J) and Omicron (21K) branches, refecting two dominant transmission waves. Clades 1-7 are associated with Delta-derived variants exhibiting moderate ACE2 afnity, while Clades 8-13 correspond to Omicron-related strains with enhanced receptor-binding potential. Te absence of the E484K mutation further diferentiates Iraqi isolates from highly immune-evasive Omicron sublineages, suggesting evolutionary adaptation toward increased receptor binding rather than immune escape.
## 3.4. Relationship of Iraqi Isolates to Global Features
• Te absence of the E484K mutation suggests that Iraqi isolates may be less capable of achieving the strong antibody escape observed in Omicron lineages. • Phylogenetic clustering confrms that Iraqi isolates mirror global evolutionary patterns, while showing region-specifc diferences in mutation frequencies and immune evasion potential. • Notably, the N501Y mutation, which was detected in Iraqi isolates, is globally documented not only in Alpha and Beta variants but also in Gamma and across numerous Omicron sublineages [1][2][3]. Tis underscores its broad evolutionary signifcance in enhancing ACE2 binding and facilitating immune escape.
## 3.5. Public Health Implications.
As a result, comprehensive genomic surveillance is essential to continuously monitor the emergence of new immune escape variants that may impact vaccine efcacy and public health strategies. Te observed S gene mutations that enhance ACE2-binding afnity while altering antibody recognition suggest a need for their consideration in future vaccine adaptations to maintain protective immunity. Furthermore, integrating genomic data with clinical outcomes will provide a more robust framework for understanding viral evolution, ultimately improving the response to the COVID-19 outbreak in Iraq and guiding efective mitigation strategies.
## 3.5.1. Impact of Key SARS-CoV-2 Spike Protein Mutations on Viral Afnity, Immune Escape, and Transmission
• ACE2 afnity illustrates how changes such as N501Y greatly increase the pulling power of the receptors ACE2. • Immune escape shows that E484K is the strongest immune escape mutation, which, however, was not found in Iraqi isolates. • Viral transmission illustrates that N501Y and D614G are greatly responsible for the spread of the virus as shown in Figure 4. P681R shows a moderate ACE2 afnity (∼6/10). Tough not impacting the RBD score, it is situated adjacent to the furin cleavage site, smoothing access.
E484K has a reduced afnity for ACE2 interaction (∼3/10), signifying minimal enhancement of direct host cell tethering.
D614G has a moderate interaction with ACE2 receptors (∼5/10). His strongest action is through heightened spike protein stability and infectivity, rather than direct receptor binding.
## 3.5.3. Te Impact of Mutations on Immune Escape (Mid Pane)
. E484K shows the greatest potential for immune escape of all (10/10). Tis is in agreement with experimental evidence indicating that it neutralizes vaccine-and convalescent antibody-enabled neutralization to a greater degree.
N501Y closely follows (∼8-9/10), diminishing dominant contributions to immune evasion-again, particularly when acting in concert with other changes (e.g., Omicron).
D614G has a moderate immune escape value (∼6/10) on account of greater viral replicative capabilities and the revelation of new epitopic regions.
P681R has a low immune escape scoring value (∼4/10), as its primary purpose relates to enabling viral entry rather than eluding antibodies.
## 3.5.4. Impact of Viral Mutations Transmission (Bottom Panel
). Te N501Y mutation and D614G polymorphism demonstrate the highest transmission efciency, scoring 10 and 9 out of 10, respectively, as both are linked to highly transmissible variants like alpha and dominating global lineages.
P681R also contributes strongly to transmission, scoring approximately 8 out of 10, particularly with the Delta variant.
E484K contributes approximately 5 out of 10. E484K intrinsically lacks any beneft that contributes to spread.
## 3.6. Phylogenetic Distribution of SARS-CoV-2 Lineages
Among Iraqi Isolates. Te chart (Figure 5) shows that 50% of Iraqi isolates belong to the Delta (21J) lineage, 30% to the 20A lineage, and 15% exhibit mutations similar to the Omicron variant. D614G, consistent with enhanced viral binding and transmissibility, whereas E484K occurs at a markedly lower frequency compared to global strains, indicating limited local circulation of strongly immune-evasive variants.
## 3.7. Comparison of Within-Population Mutation Frequencies in Iraqi and
## • N501Y:
Iraqi isolates: 90% Global variants: 98% Tis mutation, associated with increased ACE2 binding and immune escape (e.g., in Alpha and Omicron), is highly prevalent both globally and in Iraq.
## • P681R:
Iraqi isolates: 75% Global variants: 85% P681R is common in the Delta variant and contributes to enhanced viral entry. It has a slightly lower frequency in Iraq, which suggests a smaller dominance of Delta compared to the global pattern.
## • E484K:
Iraqi isolates: 10% Global variants: 90% A mutation linked to strong antibody evasion (seen in Beta, Gamma, and Omicron). It has very low frequency in Iraq, which implies limited circulation of variants like Beta or Gamma.
• D614G: Iraqi isolates: 85% Global variants: 95% A universally dominant mutation enhances viral ftness and transmission. Its high prevalence in both regions indicates a pattern of global convergence.
Te Iraqi isolates demonstrate high frequencies of critical mutations like N501Y and D614G; however, the E484K mutation is strikingly underrepresented relative to other regions of the globe. Tis indicates that there may be diferences in the local circulating variants, as well as the possibility of lower prevalence of heavily immune-evasive strains such as Beta and Gamma. Keeping track of these diferences is important for tailoring local vaccination programs and public health interventions (Table 3).
Table 3 quantitatively assesses the biological impact of the four main mutations on the SARS-CoV-2 spike: N501Y, P681R, E484K, and D614G, as they pertain to 1. ACE2 afnity (receptor binding) N501Y receives the highest score (9/10) which demonstrates how signifcantly viral binding to ACE2 is enhanced, making cell entry easier. P681R's score (6/10) suggests moderate entry enhancement, likely because of features close to the furin cleavage site rather than from ACE2 binding. E484K's receptor binding score is the lowest at 3/10. D614G receives a score of 5/10, suggesting moderate efectiveness. D614G increases infectivity indirectly through stabilizing spikes. Te chart depicts the relative distribution of major SARS-CoV-2 lineages identifed in Iraq. Delta (21J) was the predominant lineage, representing 50% of isolates and characterized by the P681R and D614G mutations that enhance viral entry and transmissibility. Lineage 20A accounted for 30% of isolates, an early global clade carrying the D614G mutation associated with increased viral ftness. Omicron-like variants constituted 15%, notable for spike mutations such as N501Y and E484A/K that enhance ACE2 binding and immune evasion. Te remaining 5% comprised other low-frequency or unclassifed variants.
## 2. Immune escape
E484K is ranked 10/10 for immune escape capability. E484K enables greater evasion from neutralizing antibodies, particularly in the Beta and Gamma variants. N501Y also substantially contributes to immune evasiveness (9/10), but primarily when other mutations such as E484K are present. D614G possesses moderate potential for immune escape, scoring 6/10. P681R demonstrates the lowest score for immune evasion (4/10), indicating minimal impact on resistance to antibodies.
## 3. Transmission potential
Both N501Y and D614G score the highest (10 and 9/ 10, respectively), making them principal drivers of rapid global spread. P681R drops slightly lower (8/10), nonetheless, considerably enhancing transmissibility for Delta variant outbreaks. E484K has a moderate transmission score (5/10), refecting its primary role in immune escape rather than enhancement of spread.
## 4. Discussion
Tis study presents one of the frst S gene-specifc genomic analyses of SARS-CoV-2 isolates in Iraq, highlighting the mutational landscape in relation to ACE2binding afnity, RBD immune escape, and viral transmission potential. Te presence of N501Y, P681R, and D614G mutations in Iraqi isolates suggests a high transmission capacity, while the absence of E484K indicates a limited immune escape potential compared to highly evasive variants like Omicron. [8,9] and further supports this complexity by demonstrating that spike mutations other than E484K can also diminish serum neutralization, indicating that the absence of E484K does not necessarily confer complete antibody sensitivity [10].
Similarly, comparative analyses of viral evolution, such as the work by Brockwell-Staats, emphasize how mutation diversity and host adaptation mechanisms collectively shape immune escape and cross-species transmissibility [11]. [13,14]. As shown in [15], in their study of Middle Eastern COVID-19 cases, certain spike mutations were associated with higher viral loads even in vaccinated individuals, highlighting the interplay of viral genomics and host immunity.
## 4.2.2. Vaccine Adaptation and Immune Evasion
## 4.2.3. Clinical and Epidemiological Considerations
1. Integrating genomic data with clinical outcomes can help determine whether certain mutations in Iraqi isolates correlate with increased disease severity or breakthrough infections.
## 2. Te moderate immune escape potential observed in
Iraqi isolates suggests that current vaccination eforts may still provide signifcant protection. However, monitoring new mutations in the spike protein remains essential. 3. Future studies should examine host immune responses in relation to these mutations, as regional genetic factors and immunity profles may contribute to the observed mutational landscape in Iraq [16,17].
## . Conclusion
Iraqi SARS-CoV-2 isolates consist largely of P681R and N501Y mutations along the spike protein and D614G along the genome, which indicates an increase in the rate of the virus's transmission. Tere is also no E484K mutation, the Advances in Virology primary immune evasion mutation, which shows that the virus's adaptation is not in the evasion of antibodies. ACE2-efcacy and phylogenetic analyses show that the 20A lineages and Delta (21J) variants are the dominant lineages in Iraq, signifying two main waves of transmission.
Te evidence presented shows that variants that were in Iraq's circulation during the time of the study are largely highly infectious, albeit remaining moderately immune evading. Tis demonstrates the suggested efectiveness of the current vaccines while signaling the need for the country to perform active genomic surveillance to pinpoint the emergence of immune-resistant variants.
## References
1. Korber, Fischer, Gnanakaran (2020) "Tracking Changes in SARS-CoV-2 Spike: Evidence that D614G Increases Infectivity" *Cell*
2. Plante, Liu, Liu (2021) "Spike Mutation D614G Alters SARS-CoV-2 Fitness" *Nature*
3. Harvey, Carabelli, Jackson (2021) "SARS-CoV-2 Variants, Spike Mutations, and Immune Escape" *Nature Reviews Microbiology*
4. Tegally, Wilkinson, Giovanetti (2021) "Emergence of SARS-CoV-2 Variant Beta with Immune Escape Properties" *Nature*
5. Viana, Moyo, Amoako (2022) "Rapid Epidemic Expansion of Omicron SARS-CoV-2 Variant" *Nature*
6. Garcia-Beltran, Denis, Hoelzemer (2021) "mRNA-based COVID-19 Vaccine Immunity Against SARS-CoV-2 Variants" *Cell*
7. Adhikari, Verma (2025) "Neutralizing Antibody Responses to SARS-CoV-2 Variants After COVID-19 Vaccination and Boosters" *Vaccine X*
8. Greaney, Loes, Crawford (2021) "Mapping Mutations to the SARS-CoV-2 RBD that Afect Recognition by Polyclonal Human Plasma Antibodies" *Cell Host Microbe*
9. Liu, Iketani, Guo (2022) "Striking Antibody Evasion by SARS-CoV-2 Omicron" *Nature*
10. Tsafack, Monamele, Moumbeket-Yifomnjou (2025) "Evolving Dynamics of Whole-Genome Infuenza A/H3N2 Viruses Isolated in Cameroon" *Advances in Virology*
11. Brockwell-Staats, Webster, Webby (2009) "Diversity of Infuenza Viruses in Swine and the Emergence of a Novel Human Pandemic Infuenza A (H1N1)" *Infuenza Other Respir Viruses*
12. Cameroni, Bowen, Rosen (2022) "Broadly Neutralizing Antibodies Overcome SARS-CoV-2 Omicron Antigenic Shift" *Nature*
13. Yu, Wei, Xu (2022) "Reduced Sensitivity of SARS-CoV-2 Omicron Variant to Antibody Neutralization Elicited by Booster Vaccination" *Cell Discovery*
14. Wang, Nair, Liu (2021) "Antibody Resistance of SARS-CoV-2 Variants B.1.351 and B.1.1.7" *Nature*
15. Mahdi, Khudhair, Khalil (2022) "Assessing the Potential Correlation of Polymorphisms in the TMPRSS2 Gene with Severity of COVID-19 Patients" *Biomedicine*
16. Schmidt, Weisblum, Rutkowska (2021) "High Genetic Barrier to Escape from Human Polyclonal SARS-CoV-2 Neutralizing Antibodies" *Cell*
17. Jabber, Khudhair (2023) "Association Between SNP rs4986790 and COVID-19 Infection Severity Among Baghdad Patients" *Biomedicine* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12459333&blobtype=pdf | # Atomistic simulations of intact virus capsids: A computational challenge worth the scientific payoff
Carolina Pérez-Segura, Juan Perilla, Jodi Hadden-Perilla
## Abstract
All-atom molecular dynamics (MD) simulations of intact virus capsids provide unparalleled insights into the functional motions of these complex macromolecular assemblies. Despite the computational challenges of simulating multimillion-atom systems, these simulations uniquely reveal the structural basis for emergent properties, including collective motions, allostery, selective permeability, and mechanical responses that are inaccessible through experimental methods. Capsid simulations also drive technological advancements in MD methodologies, analysis tools, and multiscale modeling, fostering broader innovations in structural biology and biophysics. Given next-generation computational resources, MD simulations will continue to illuminate virus biology, support antiviral drug discovery, and enhance preparedness for emerging viral diseases. Here, atomistic simulations of complete capsid assemblies are reviewed, and their role in elucidating fundamental principles of virus function and therapeutic targeting is discussed. Altogether, MD of intact capsids is a computational challenge worth the payoff.
Computational virology of virus capsidsCapsids are central to the architecture of viruses. Composed of repeating protein subunits, they constitute shells that organize and protect the viral genome. Capsids can be constructed according to icosahedral or helical symmetry, or combine elements of both symmetries to form conical or prolate morphologies. In addition to their essential structural role, capsids carry out numerous key functions during viral infection. Overall, they are major mediators of virus-host interactions. In nonenveloped viruses, capsids represent the virion exterior, critical for cell adhesion and entry. In enveloped viruses, capsids are circumscribed by a
lipid bilayer, yet encompass the core particle released into the host cytoplasm for genome trafficking and delivery. In phages-viruses that infect bacteria-capsids typically incorporate a tail/portal complex that injects the encapsulated genome into the host cell. Following replication, capsid proteins assemble during the production of progeny virions, which go on to repeat the viral life cycle.
Virus capsids feature prominently in modern structural biology, which seeks to discover their structure-function relationships. Beyond basic science characterization of capsid functions, translational research aims to inhibit those functions to treat disease, as well as harness those functions for new biotechnology applications. Molecular dynamics (MD) simulations are an increasingly important structural biology technique capable of revealing the detailed motions that connect macromolecular structures with biological functions, allowing researchers to elucidate their underlying physical mechanisms. With respect to virus research, MD simulations are now widely accepted as an essential complement to experimental work, and many high-impact projects leverage simulations and experiments synergistically to reach deeper scientific insights. The general application of MD simulations to virus structures and the ensuing subfield of 'computational virology' has been covered in recent reviews: Reference [1] highlights the achievements of atomistic virus simulations, references [2][3][4][5] summarize the broad success of virus simulations across fine to coarse-grain resolution scales-from individual viral proteins to intact virions, and reference [6] discusses progress in moving beyond sampling equilibrium dynamics to simulating viral life-cycle processes. Reference [7] highlights the role of high-performance computing and software development in computational virology.
## Challenges of all-atom capsid simulations
At its pinnacle, computational virology treats full-scale models of viruses at atomistic resolution. Nearly 20 years ago, all-atom MD simulations were first applied to study an intact virus capsid [8]. This landmark investigation represented a major milestone for scientific computing, launching atomistic simulations into the million-atom regime. In the decades since, relatively few subsequent MD projects have tackled virus capsids as complete macromolecular assemblies at this level of spatial detail. All-atom MD simulations for the intact capsids of only five human viruses have been reported: human immunodeficiency virus-1 (HIV-1) [9][10][11], Ebola [12], polio [13], human papillomavirus [14], and hepatitis B (HBV) [15][16][17][18][19]. Of these, only HIV-1 and HBV have been examined on the microsecond time-scale [10,16,17].
All-atom MD simulations of intact virus capsids are formidable owing to their computational expense, as well as the difficulties of processing the massive datasets they generate. Further, because capsids are macromolecular containers, it can be challenging to relax them to equilibrium conditions prior to commencing production simulation. Within an isothermal-isobaric ensemble, the system density-determined by the simulation box size-must stabilize. Meanwhile, the capsid may expand [13,16] or contract [10] as it relaxes, altering its internal volume and, thus, the density of solvent contained inside. Solvent molecules rush inward or outward in response, subject to the capsid's permeability, which requires adjustment of the simulation box to regulate external solvent density. This simultaneous relaxation of coupled properties requires close monitoring-and sometimes iterative intervention-to ensure that capsid systems are brought to a stable, physically-valid equilibrium state prior to sampling their dynamics. Some capsids require hundreds of nanoseconds just to reach a structural equilibrium [10,20]. Then, it can take months of wallclock time to collect a microsecond of sampling-even on a leadership-class supercomputer. The resulting simulation trajectory entails tens of thousands of time-ordered conformers, and its storage footprint can reach tens of terabytes-or more. Thus, supercomputing resources are needed not only to perform the simulation, but also to quantitatively analyze the trajectory to yield scientific discoveries. Like MD engines, MD analysis codes must be high-performance-parallelized, memory-optimized, or graphics processing unit (GPU)accelerated to tackle million-atom datasets describing intact capsids [21][22][23].
## Insights beyond experimental reach
Despite these computational challenges, the scientific rewards of running and analyzing allatom simulations of capsids are considerable. Experimental structural biology techniques, while powerful in their application to viruses, cannot reveal all the details needed to elucidate complex mechanisms of macromolecular function. Structure-determination methods like X-ray crystallography [24,25], cryo-electron microscopy (EM) [26,27], or cryo-electron tomography (ET) [28,29] can capture static structural features with precision but cannot clearly report on their movements. Conversely, methods that detect dynamics, like nuclear magnetic resonance spectroscopy (NMR) [30], hydrogen/deuterium exchange mass spectrometry (HDXMS) [31], high-speed atomic force microscopy (AFM) [32], or X-ray free-electron lasers (XFEL) [33] cannot yet provide unambiguous, highresolution structural context. Often, data obtained by these techniques are convoluted by conformational heterogeneity, known to be inherent in some virus capsids [16], leading to reduced resolution, loss of structural nuance, or averaging out of low-population states [34]. Moreover, structural biology is only recently coming to terms with asymmetry in capsid structures and its implications for particle dynamics [35][36][37]. Fortunately, all-atom MD simulations provide the complementary information needed to bridge these spatiotemporal gaps, allowing researchers to directly connect conformation and dynamics with experimental observables to yield a more complete view into capsid function. Even with the limitation of short sampling timescales-reaching only up to the microsecond so far-capsid simulations routinely reveal key insights that are beyond reach experimentally, serving to critically advance our understanding of viral biology.
Further, all-atom simulations of intact capsids are necessary to examine the emergent properties of these macromolecular assemblies, which are essential to explaining biological function. Emergent properties are characteristics or behaviors that arise from the collective organization and interaction of a system's components, and cannot be understood by studying the isolated components. In the case of virus capsids, emergent properties reflect the integrated dynamics of the entire assembly, underlying correlated motions (e.g. allostery, cooperativity, and global morphological transitions), container attributes (e.g. internal pressure, permeability, and homeostasis), and mechanical qualities (e.g. elasticity, stress response, and distribution of physical strain). All-atom MD simulations of intact capsids provide access to these functionally-relevant features by treating the system as a complete entity, capturing dynamics that are inaccessible when studying individual subunits or symmetry-constrained models. Below, atomistic simulations of intact virus capsids are briefly reviewed, with a focus on the discoveries they enabled that were beyond the reach of experimental methods alone. Each of these remarkable discoveries concern emergent properties that could not be examined at high resolution using experimental methods, or by simulating less than the complete, intact capsid assembly at full chemical detail. The capsids featured in these works are shown to scale in Figures 1 and2.
## Icosahedral capsids
## Plant viruses
All-atom MD simulations of the satellite tobacco mosaic virus (STMV) capsid demonstrated that the shell collapses without the genome inside, revealing the detailed mechanism by which structural failure proceeds [8]. Simulations of the satellite tobacco necrosis virus (STNV) capsid captured the early stages of repulsion-induced swelling following deprivation of divalent cations [38]. MD order parameter extrapolation (MD/OPX) simulations of the cowpea chlorotic mottle virus (CCMV) capsid examined both its swelling [39] and shrinking [40] transitions, explaining the mechanisms by which cooperative motions across the shell drive large-scale conformational change. Simulations of the southern bean mosaic virus (SBMV) capsid modeled AFM force-probe experiments on the shell, allowing elastic constants and yielding forces to be mapped to specific regions of the structure [41]. Simulations of the brome mosaic virus (BMV) capsid were used in synergy with experiments to determine which conjugation sites are accessible to fluorophores, an important step in characterizing BMV-based super-fluorescent nanoparticles [42].
## Animal viruses
All-atom MD simulations of the empty porcine circovirus type 2 (PCV2) capsid showed that it is impermeable to ions [43], yet stable at both ambient and physiological conditions, given a sufficient concentration of chloride ions already contained inside [44]. Internal chloride localization suggested regions of capsid-DNA interaction that confer shell stability [45]. Simulations of the minute virus of mice (MVM) capsid at elevated temperatureknown experimentally to trigger externalization of signaling peptides-revealed a heatinduced breathing transition that expanded the shell and increased water permeability [46]. Simulations of a T = 1 of Rous sarcoma virus (RSV) capsid-composed entirely of pentamers-established the molecular switch that differentiates pentameric and hexameric capsid protein conformations in larger assemblies [47].
## Insect viruses
All-atom MD simulations of the Flock House virus (FHV) capsid observed that water permeability correlates with shell geometry, with solvent content and exchange varying with local symmetry features [48]. Acidic conditions induced radial expansion of the capsid, which facilitates the release of membrane-lytic peptides through enlarged fivefold pores [49].
## Human viruses
All-atom MD simulations of the poliovirus capsid revealed that it is selectively permeable, allowing bidirectional water-but not ion-exchange, with ion interactions along the interior shell wall contributing to a negative internal pressure in the absence of genome [13]. Simulations of the human papillomavirus (HPV) capsid were used as a case-study to demonstrate new force field interoperability features in NAMD [14], as well as to debug performance gaps for macromolecules containing bonds between atoms with large disparities in list index (e.g. intersubunit disulfide cross-links).
Simulations of the HBV capsid confirmed the basis for low resolution of crystal and cryo-EM structures by revealing the capsid's remarkable flexibility and instantaneous asymmetry [16]. Simulations further indicated preferential permeability of the capsid to positivelycharged ions [16,19] and found that the strength of preference increases in the presence of packaged RNA [19]. By examining the dynamics of subunit interfaces, simulations explained the mechanism by which point mutations enhance capsid assembly and increase drug resistance [50]. By characterizing the dynamics of sixfold pores, simulations showed that pore dilation is not required for the capsid's externalization of signaling peptides [51]. The extensive sampling afforded by microsecond MD led to the discovery of a new conformational state of the capsid protein, never observed in experimental structures [17]. The motions captured by simulations were shown to be complementary to those of solidstate NMR, which has a blind spot in the temporal range where MD excels [52]. Simulations of the capsid complexed with an assembly accelerator sampled a rare conformer with higher frequency, capturing shifts in allosteric communication [17]. Simulations of the capsid complexed with an assembly misdirector exhibited more pronounced shell faceting than previously observed in crystal structures, describing the early stages of inhibitor-induced disassembly [15]. Mechanically induced disassembly of the capsid, simulated by applying an isotropic external pulling force to subunits, established that stress fractures form most readily within hexamers and along threefold interfaces [18].
## Bacteriophages
All-atom MD simulations of the bacteriophage MS2 capsid showed relatively little deviation from the cryo-EM structure, but revealed bidirectional solvent flow with equitable exchange of positively-and negatively-charged ions across the shell surface [53]. When filled with a model genome, the capsid exhibited reduced solvent flow without any observable change in pore diameter [54]. Expansion of the bacteriophage T7 procapsid was simulated using a 'structure-based' model-which omits hydrogen atoms and explicit solvent-to mimic its maturation upon genome import, finding that the mature-form shell can externalize signaling peptides [55]. Constant pH simulations applying rotational symmetry boundary conditions were used to investigate acid-dependent expansion of the bacteriophage HK97 capsid, identifying specific residues that contribute to the conformational changes associated with maturation and shell stability [56]. Simulations of the mature, genome-containing HK97 capsid revealed increased interior volume, reduced permeability, and diminished fluctuations compared to the empty shell, arising from the internal pressure and enclosed electrostatic environment of the particle in its packaged state [20].
## Helical capsids
All-atom MD simulations of an Ebola virus (EBOV) capsid fragment incorporating ~3.5 turns (90 repeats) in the presence and absence of genome demonstrated that protein-RNA contacts along the helical axis enhance structural integrity of the tube, including by promoting increased protein-protein contact [12]. An atomistic model of a tubular HIV-1 assembly (71 hexamer repeats) was solved using molecular dynamics flexible fitting (MDFF) [57] to cryo-EM data. Simulations of the HIV-1 tube identified key residue contacts responsible for stabilizing interhexamer interfaces and showed that the curvature needed to form the helical capsid structure is conferred by conformational adjustments of the C-terminal domain relative to the N-terminal domain [9]. A refined model of the HIV-1 tube (87 hexamer repeats) was later derived by additionally incorporating magic-angle spinning (MAS) NMR restraints in data-guided MD simulations [58]. Results indicated that the linker region connecting the N-and C-terminal domains can adopt up to four distinct conformations, explaining its poor resolution in experimental structures.
## Conical capsids
Atomistic structure determination of the mature HIV-1 capsid was a feat of integrative modeling that would not have been possible without MD simulations. Besides the role of MDFF in solving hexamer subunits, unbiased MD established the contacts and curvature along pentamer subunits that induce closure of the conical assembly [9]. Simulations of two possible cones-constructed from 12 pentamers plus 186 or 216 pentamers, respectively-relaxed the intact models to physiological conditions for final cross-correlation against cryo-ET data to validate their size and morphology [9]. This process was later repeated for another cone polymorph-constructed from 12 pentamers plus 241 hexamersto derive an atomistic structure for a native HIV-1 capsid based on in situ cryo-ET [11]. Microsecond MD of the 186-hexamer cone revealed its emergent physicochemical properties, including complex collective motions, acoustic and electrostatic qualities, and preferential permeability to negatively-charged ions [10].
## Challenges drive technology development
Beyond their contributions to understanding viral biology, all-atom simulations of intact virus capsids have served as powerful drivers of computational progress. Owing to their size and complexity, these multimillion-atom systems have spurred innovations in simulation methodologies, MD code performance, and trajectory analysis tools. For example, MDFF was developed to bridge the gap between low-resolution cryo-EM maps and atomistic structures-especially for large systems like viruses and cellular machinery-allowing researchers to model dynamic macromolecules while maintaining structural and energetic authenticity [57]. MDFF was pivotal for solving structures of, among others, the rabbit hemorrhagic disease virus (RHDV) capsid [59], the bacteriophage T7 procapsid [60], the tubular assembly of RSV [61], and the mature HIV-1 capsid [9,11]. The latter represented a major milestone not only for MD but also for the global infrastructure of structural biology. The HIV-1 capsid exceeded the atom-count limitations of the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) format, prompting updates to the way atomic models could be deposited and accessed by researchers. This model also marked the first time a computationally derived structure-integrating data from crystallography, NMR, and cryo-EM/ET-was added to the PDB. By pushing such boundaries, capsid simulations have laid the foundation for future breakthroughs in large-scale integrative modeling.
As key biological systems of interest, viruses have long represented aspirational simulation targets and major drivers of capability advancement for software like NAMD [62] and GROMACS [63]. The raw computational expense of simulating large systems like virus capsids has propelled optimization and scalability improvements in these MD engines [64,65], as well as their adaptation to emerging state-of-the-art computing architectures [66][67][68][69]. As the first million-atom simulation, the STMV capsid has become a common MD benchmark, continuing to guide performance enhancements of MD engines. In general, using physiologically authentic capsids as test cases for code features has led to algorithmic updates in MD engines that overcome performance challenges faced broadly by large protein assemblies. Amelioration in force field interoperability has also been spurred by the need to simulate larger and more heterogeneous macromolecular systems. Recent efforts to port force fields from the AMBER software package into NAMD to overcome atom-count limits were motivated by simulation goals of virus capsids, virus-like particles, and vaccine antigens [42,14], yet expanded availability of widely-used parameter sets for application to diverse biological targets.
The importance of studying viral processes beyond the timescales accessible to all-atom MD has driven the development of various multiresolution simulation approaches that enhance temporal sampling without sacrificing the accuracy afforded by atomistic models. Shapebased coarse-graining (SBCG)-which uses a reduced spatial representation parameterized to reproduce key features of a system's atomistic dynamics-emerged as a method to explore capsid behavior over extended durations at low computational cost. The original implementation of SBCG was applied to study STMV, satellite panicum mosaic virus (SPMV), STNV, BMV, poliovirus, reovirus, and the bacteriophage ΦX174 procapsid [70].
It was subsequently used to characterize the response of the HBV capsid to mechanical deformation, analogous to experimental investigation with AFM [71]. An updated version of the approach, dubbed SBCG2, was developed to probe the elastic properties of the HIV-1 capsid and their relationship to nuclear entry of the viral core during infection [72,73]. Similarly, coarse-grain kinetic Monte Carlo (KMC) simulations, parameterized based on atomistic MD of the intact HBV capsid, have been used to examine assembly pathways and identify key assembly intermediates [74], producing a framework that can be adapted to study subunit association for other viruses. The combination of MD code for Brownian dynamics with a previously-developed multiresolution DNA model, when adapted to the problem of viral genome packaging, led to the realization of a robust protocol for producing realistic models of genome-filled capsids [20]. On a related note, the ambition to simulate long-timescale dynamics of atomistic viral particles, including intact envelopes, has spurred the development of multiscale simulation methods that use artificial intelligence to drive all-atom MD using information obtained from disparate resolution scales [75].
Finally, while the need to efficiently analyze large trajectory datasets has driven improvements across MD analysis codes [21][22][23], specific challenges presented by virus capsids have led to the development of new computational tools based on virus-specific applications. For example, the measure volinterior method in visual molecular dynamics (VMD) was designed to distinguish the interior and exterior spaces of biological containers like capsids [76], enabling rapid calculation of volume and solvent exchange rates. So far, the method has been applied to characterize the container properties of HIV-1, HBV, Ebola, FHV, and HK97 capsids [76,49,20], but has also found utility for measuring protein cavities and tracking lipid flip-flop in spherical vesicles [76,77]. Altogether, the computational challenges posed by all-atom MD simulations of intact virus capsids have inspired advances that extend well beyond virology. The methods, tools, and workflows developed to tackle these and other large systems have had broad applications across biophysics and structural biology, underscoring the value of virus capsid simulations as both scientific and technological achievements.
## From challenges to real-world solutions
Looking forward, all-atom MD simulations of intact virus capsids are poised to take advantage of next-generation computational resources, such as exascale supercomputers and purpose-built machines like Anton 3 [78]. With these platforms, generating multimillion-atom trajectories for capsids-including on longer timescales-will become less computationally prohibitive, significantly reducing barriers to data collection. In that case, the list of reported capsid simulations will soon expand to encompass a broader range of viruses, including more that are human-infective. However, as simulation datasets continue to grow in terms of size, complexity, and disk footprint, analysis will represent the greater challenge. New tools must be developed to efficiently process these massive trajectories to extract biologically meaningful information. Advances in artificial intelligence and machine learning are likely to play a central role, enabling automated detection of functionally relevant patterns in capsid dynamics, such as motions underlying emergent properties. Thus, while technological advancements will enable new virus simulations, the continuing goal of performing and analyzing these simulations will remain a major impetus for computational innovations.
Given that pathogens are a persistent threat to the health of people, livestock, and agriculture, the importance of pursuing the basic science of viruses cannot be understated. Historically, results from basic science investigations have proven transferable to related viral systems. For example, early studies of retroviruses like human T-lymphotropic virus (HTLV, which is oncogenic) and RSV (which infects poultry) provided a critical foundation for understanding HIV-1 upon its emergence in humans. Similarly, knowledge of vector-borne flaviviruses that cause dengue and yellow fever guided characterization of and vaccine development against zika. Previous research on SARS-CoV-1 and MERS-CoV coronaviruses accelerated the response to SARS-CoV-2 during the COVID-19 pandemic, further underscoring the importance of fundamental studies to preparedness for future outbreaks. By providing highly detailed descriptions of functional dynamics, MD simulations make unparalleled bottom-up contributions to the basic science of viruses.
Large-scale computational modeling underpins many facets of modern science and technology, from climate prediction to civil engineering to drug discovery, and beyond. The purpose of modeling is to extend our understanding into realms that are inaccessible by direct observation, whether due to spatial or temporal limitations, or because they involve predicting future outcomes. In virology, all-atom MD simulations play a critical role in bridging these gaps, providing unique insights that refine experimental hypotheses, guide new research directions, and further our basic science characterization of viruses. While tools like AlphaFold [79] have revolutionized static protein structure prediction, they cannot yet capture the dynamics required to fully explain protein function or its regulation. MD simulations, thus, remain indispensable for uncovering the functional motions and emergent properties of virus capsids, offering a powerful window into the mechanisms driving viral processes. Since the complexity of capsid motion arises, in part, from their multifunctionality, an MD trajectory describing capsid dynamics may contain numerous scientific discoveries about the capsid's biological role. Remarkably, the first microsecond simulation of the intact HBV capsid [16] was so rich with information that its re-analysis has contributed to four follow-up studies related to understanding viral particle assembly, response to bound inhibitors, drug resistance, and display of intracellular trafficking signals [50,17,74,51]. Given that a single all-atom capsid simulation can have so much utility and serve as the basis for multiple investigative projects, the upfront computational expense of these calculations is further justified.
Although coarse-grain simulations are popular for virus studies due to their ability to access longer timescales for large systems, atomistic simulations provide the essential chemical resolution needed to understand functional nuances. This level of detail is critical for capturing quasi-equivalence in viral structures, where identical capsid proteins adopt distinct conformations depending on their location within the symmetry lattice. It is also necessary to understand the molecular basis of viral evolution, host-pathogen interactions, and the affinity-driven arms race between viral proteins and host immune factors. The importance of chemical detail extends to applied virology, particularly drug design. The recent realization of capsid-targeting antivirals demonstrates the viability of capsids as druggable targets, despite being structural rather than enzymatic proteins. Lenacapavir-named Science's 2024 Breakthrough of the Year owing to its remarkable ability to prevent HIV-1 infection [80]-is an achievement born from a long history of basic science research. Atomistic simulations feature prominently in this story, from revealing the high-resolution structure and emergent properties of the drug target [9,10] to establishing the molecular mechanism by which it inhibits the capsid to disrupt the viral life cycle [73]. Given this demonstrated capacity to contribute to transformative biomedical advancements, all-atom MD simulations of virus capsids are-despite the challenges-well worth the scientific payoff. Graphical summary of intact icosahedral (a) and helical (b) virus capsids for which atomistic MD simulations have been reported. The atom count for the largest simulation is indicated for each, which includes explicit solvent for all but the bacteriophage T7 procapsid. Icosahedral capsids are arranged according to increasing triangulation number. MD, molecular dynamics. Graphical summary of intact conical virus capsids for which atomistic MD simulations have been reported. All three represent polymorphs of the mature HIV-1 capsid solved by integrative modeling. The number of hexameric subunits contained in each cone (along with 12 pentameric subunits) is indicated, as well as the solvated simulation atom count. MD, molecular dynamics.
## References
1. "Papers of particular interest, published within the period of review, have been highlighted as: • of special interest •• of outstanding interest"
2. Hadden, Perilla (2018) "All-atom virus simulations" *Curr Opin Virol*
3. Reddy, Sansom (2016) "Computational virology: from the inside out" *Biochim Biophys Acta*
4. Jefferys, Sansom "Computational virology: molecular simulations of virus dynamics and interactions"
5. Marzinek, Huber, Bond (2020) "Multiscale modelling and simulation of viruses" *Curr Opin Struct Biol*
6. Lynch, Pavlova, Fan et al. (2023) "Understanding virus structure and dynamics through molecular simulations" *J Chem Theor Comput*
7. Jones, Pérez-Segura, Bryer et al. (2021) "Molecular dynamics of the viral life cycle: progress and prospects" *Curr Opin Virol*
8. Machado, Pantano "Fighting viruses with computers, right now"
9. Freddolino, Arkhipov, Larson et al. (2006) "Molecular dynamics simulations of the complete satellite tobacco mosaic virus" *Structure*
10. Zhao, Perilla, Yufenyuy et al. (2013) "Mature HIV-1 capsid structure by cryo-electron microscopy and all-atom molecular dynamics Nat"
11. Perilla, Schulten (2017) "Physical properties of the HIV-1 capsid from all-atom molecular dynamics simulations" *Nat Commun*
12. Ni, Zhu, Yang et al. (2021) "Structure of native HIV-1 cores and their interactions with IP6 and CypA" *Sci Adv*
13. "This study combines cryo-electron tomography and MD flexible fitting to construct an all-atom model of the HIV-1 capsid. It identifies key residues coordinating IP6 binding and highlights the molecular details of IP6 interactions within hexamers and pentamers. By linking integrative modeling with MD simulations, the work underscores the importance of all-atom models in understanding capsid stability and function"
14. Xu, Katyal, Nesterova et al. (2020) "•This study explores the stability of the Ebola nucleocapsid in the presence and absence of its RNA genome. It highlights the role of RNA in maintaining structural integrity through protein-RNA and protein-protein interactions. This work provides a molecular-level understanding of capsid dynamics and genome interactions essential for interpreting nucleocapsid stability" *J Chem Phys*
15. Andoh, Yoshii, Yamada et al. (2014) "All-atom molecular dynamics calculation study of entire poliovirus empty capsids in solution" *J Chem Phys*
16. Antolínez, Jones, Phillips et al. (1016) "PubMed: 38176097] •This study refactors the AMBER force field for use in NAMD, enabling multimillion-atom MD simulations with the AMBER family of force fields in systems containing up to two billion atoms. As part of the validation process, an HPV capsid was used as a benchmark to evaluate performance in systems exceeding 10 million atoms. This work sets a foundation for accelerating and expanding the scope of large-scale capsid simulations" *J Chem Inf Model*
17. Perilla, Hadden, Goh et al. (2016) "All-atom molecular dynamics of virus capsids as drug targets" *J Phys Chem Lett*
18. Hadden, Perilla, Schlicksup et al. (2018) "All-atom molecular dynamics of the HBV capsid reveals insights into biological function and cryo-EM resolution limits"
19. Pérez-Segura, Goh, Hadden-Perilla (2021) "All-atom MD simulations of the HBV capsid complexed with AT130 reveal secondary and tertiary structural changes and mechanisms of allostery Viruses"
20. Ghaemi, Gruebele, Tajkhorshid (2021) "Molecular mechanism of capsid disassembly in hepatitis B virus" *Proc Natl Acad Sci*
21. Fujimoto, Yamaguchi, Urano et al. (2021) "All-atom molecular dynamics study of hepatitis B virus containing pregenome RNA in solution" *J Chem Phys*
22. Coshic, Maffeo, Winogradoff et al. (2024) "PubMed: 38448589] ••This publication presents an all-atom model of the bacteriophage HK97 virion, developed using multiresolution structure refinement procedures. It provides valuable insights into genome organization and capsid-genome interactions that were challenging to uncover through experimental approaches. This work establishes a foundation for future research" *Nature*
23. Humphrey, Dalke, Schulten (1996) "VMD: Visual molecular dynamics" *J Mol Graph*
24. Roe, Cheatham (2018) "TE: parallelization of CPPTRAJ enables large scale analysis of molecular dynamics trajectory data" *J Comput Chem*
25. Naughton, Alibay, Barnoud et al. (2022) "MDAnalysis 2.0 and beyond: fast and interoperable, community driven simulation analysis" *Biophys J*
26. Mcpherson, Larson (2015) "A guide to the crystallographic analysis of icosahedral viruses" *Crystallogr Rev*
27. Johnson, Olson "Icosahedral virus structures and the protein data bank" *J Biol Chem 2021*
28. Schoehn, Chenavier, Crépin (2023) "Advances in structural virology via cryo-EM in 2022" *Viruses*
29. Zhu, Cao, Zhang (2023) "Virus structures revealed by advanced cryoelectron microscopy methods" *Structure*
30. Graham, Zhang (2023) "Cryo-electron tomography to study viral infection" *Biochem Soc Trans*
31. Huang, Zhang, Ni (2023) "Towards in situ high-resolution imaging of viruses and macromolecular complexes using cryoelectron tomography" *J Struct Biol*
32. Lecoq, Fogeron, Meier et al. (2020) "Solid-state NMR for studying the structure and dynamics of viral assemblies" *Viruses*
33. Venkatakrishnan, Braet, Anand (2024) "Dynamics, allostery, and stabilities of whole virus particles by amide hydrogen/deuterium exchange mass spectrometry (HDXMS)" *Curr Opin Struct Biol*
34. Van Ewijk, Maity, Roos "Visualizing molecular dynamics by high-speed atomic force microscopy"
35. Meents, Wiedorn (2019) "Virus structures by X-ray free-electron lasers" *Annu Rev Virol*
36. Bonomi, Heller, Camilloni et al. (2017) "Principles of protein structural ensemble determination" *Curr Opin Struct Biol*
37. Parent, Schrad, Cingolani (2018) "Breaking symmetry in viral icosahedral capsids as seen through the lenses of X-ray crystallography and cryo-electron microscopy" *Viruses*
38. Jose, Hafenstein (2022) "Asymmetry in icosahedral viruses" *Curr Opin Virol*
39. Comas-Garcia (2024) "How structural biology has changed our understanding of icosahedral viruses" *J Virol*
40. Larsson, Liljas, Van Der Spoel (2012) "Virus capsid dissolution studied by microsecond molecular dynamics simulations" *PLoS Comput Biol*
41. Miao, Johnson, Ortoleva (2010) "All-atom multiscale simulation of cowpea chlorotic mottle virus capsid swelling" *J Phys Chem B*
42. Miao, Ortoleva (2010) "Viral structural transition mechanisms revealed by multiscale molecular dynamics/order parameter extrapolation simulation" *Biopolymers*
43. Zink, Grubmüller (2009) "Mechanical properties of the icosahedral shell of southern bean mosaic virus: a molecular dynamics study" *Biophys J*
44. Sushma, Zhao, Tsvetkova et al. "Subset of fluorophores is responsible for radiation brightening in viromimetic particles" *J Phys Chem B*
45. Tarasova, Korotkin, Farafonov et al. (2017) "Complete virus capsid at all-atom resolution: simulations using molecular dynamics and hybrid molecular dynamics/hydrodynamics methods reveal semipermeable membrane function" *J Mol Liq*
46. Tarasova, Farafonov, Khayat et al. (2017) "All-atom molecular dynamics simulations of entire virus capsid reveal the role of ion distribution in capsid's stability" *J Phys Chem Lett*
47. Tarasova, Farafonov, Taiji et al. (2018) "Details of charge distribution in stable viral capsid" *J Mol Liq*
48. Pathak, Bandyopadhyay (2022) "Heat-induced transitions of an empty minute virus of mice capsid in explicit water: all-atom MD simulation" *J Biomol Struct Dyn*
49. Thames, Bryer, Qiao et al. "Curvature of the retroviral capsid assembly is modulated by a molecular switch" *J Phys Chem Lett*
50. Shrivastav, Borkotoky, Dey et al. (2024) "Structure and energetics guide dynamic behaviour in a T = 3 icosahedral virus capsid" *Biophys Chem*
51. Jana, Sharawy, May (2023) "••This study employs microsecond MD simulations to investigate the FHV capsid dynamics under varying pH conditions and RNA presence. It provides critical insights into pH-induced capsid swelling and local protein conformational changes. Highlighting the atomistic impact of specific residues protonation, this work advances the understanding of viral capsid function and protein-protein interactions" *J Struct Biol*
52. Ruan, Hadden, Zlotnick (2018) "Assembly properties of hepatitis B virus core protein mutants correlate with their resistance to assembly-directed antivirals" *J Virol*
53. Kim, Schlicksup, Pérez-Segura et al. (2023) "•This study employs MD simulations to investigate the dynamics of sixfold pores in the HBV capsid, with a focus on the externalization of the C-terminal domain, which interacts with host factors during infection. The findings reveal that pore dilation is not directly associated with the exit of the C-terminal domain. This work exemplifies how all-atom MD simulations of intact capsids can provide detailed insights into local structural changes in capsids and their correlation with experimental data" *J Biol Chem*
54. Malär, Callon, Smith et al. (2022) "Experimental characterization of the hepatitis B virus capsid dynamics by solid-state NMR" *Front Mol Biosci*
55. Farafonov, Nerukh (2019) "MS2 bacteriophage capsid studied using all-atom molecular dynamics" *Interface Focus*
56. Farafonov, Stich, Nerukh (2023) "c00846. • This study advances large-scale MD simulations by presenting a complete allatom model of the MS2 bacteriophage, including its native genome. It investigates the genome's role in capsid dynamics and elucidates experimentally observed conformational heterogeneity in the viral genome, providing a framework for future research on fully atomistic viral systems" *J Chem Theor Comput*
57. Whitford, Jiang, Serwer (2020) "Simulations of phage T7 capsid expansion reveal the role of molecular sterics on dynamics" *Viruses*
58. May, Arora, Brooks (2014) "CL: pH-induced stability switching of the bacteriophage HK97 maturation pathway" *J Am Chem Soc*
59. Trabuco, Villa, Frank et al. (2008) "Flexible fitting of atomic structures into electron microscopy maps using molecular dynamics. In Single-particle cryo-electron microscopy: the path toward atomic resolution: selected Papers of joachim Frank with commentaries"
60. Lu, Russell, Bryer et al. (2020) "Atomic-resolution structure of HIV-1 capsid tubes by magic-angle spinning NMR" *Nat Struct Mol Biol*
61. Wang, Xu, Liu et al. (2013) "Atomic model of rabbit hemorrhagic disease virus by cryo-electron microscopy and crystallography" *PLoS Pathog*
62. Guo, Liu, Fang et al. (2014) "Capsid expansion mechanism of bacteriophage T7 revealed by multistate atomic models derived from cryo-EM reconstructions" *Proc Natl Acad Sci*
63. Jeon, Qiao, Hung et al. (2006) "Structural model of the tubular assembly of the Rous Sarcoma virus capsid protein" *J Am Chem Soc*
64. Phillips, Braun, Wang et al. (2005) "Scalable molecular dynamics with NAMD" *J Comput Chem*
65. Van Der Spoel, Lindahl, Hess et al. (2005) "GROMACS: fast, flexible, and free" *J Comput Chem*
66. Páll, Zhmurov, Bauer et al. (2020) "Heterogeneous parallelization and acceleration of molecular dynamics simulations in GROMACS" *J Chem Phys*
67. Phillips, Hardy, Maia et al. (2020) "Scalable molecular dynamics on CPU and GPU architectures with NAMD" *J Chem Phys*
68. Phillips, Sun, Jain et al. "Mapping to irregular torus topologies and other techniques for petascale biomolecular simulation"
69. Páll, Abraham, Kutzner et al. (2014) "Tackling exascale software challenges in molecular dynamics simulations with GROMACS"
70. Stone, Hynninen, Phillips et al. (2016) "Early experiences porting the NAMD and VMD molecular simulation and analysis software to GPU-accelerated OpenPOWER platforms"
71. Acun, Hardy, Kale et al. (2018) "Scalable molecular dynamics with NAMD on the Summit system" *IBM J Res Dev*
72. Arkhipov, Freddolino, Schulten (2006) "Stability and dynamics of virus capsids described by coarse-grained modeling" *Structure*
73. Arkhipov, Roos, Wuite et al. (2009) "Elucidating the mechanism behind irreversible deformation of viral capsids" *Biophys J*
74. Bryer, Rey, Perilla (2014) "This publication introduces an improved version of the subnanometer shape-based coarse-graining approach that replicates key atomistic dynamics, while enabling the exploration of capsid behavior over extended durations with low computational cost. The method is applied to probe the elastic properties of the HIV-1 capsid, providing insights into its mechanical behavior and demonstrating the utility of coarse-grained models for efficient macromolecular studies" *Nat Commun*
75. Deshpande, Bryer, Andino-Moncada et al. (2024) "Elasticity of the HIV-1 core facilitates nuclear entry and infection" *PLoS Pathog*
76. Mohajerani, Tyukodi, Schlicksup et al. (2022) "This study employs coarse-grain kinetic Monte Carlo simulations parameterized from all-atom MD to investigate HBV capsid assembly pathways. By demonstrating consistency between coarse-grain simulations and experimental data, it highlights the role of all-atom MD in bridging experimental and theoretical studies. The framework and methodology" *ACS Nano*
77. Casalino, Dommer, Gaieb et al. (2021) "AI-driven multiscale simulations illuminate mechanisms of SARS-CoV-2 spike dynamics" *Int J High Perform Comput Appl*
78. Bryer, Hadden-Perilla, Stone et al. (2019) "This publication introduces a novel method for calculating volumes in biomolecular containers, such as viral capsids and protein pockets, using 3D ray-casting techniques. By enabling high-performance characterization, the study showcases advancements in software designed to analyze large datasets from all-atom capsid simulations" *J Chem Inf Model*
79. Gonzalez-Arias, Reddy, Stone et al. (2020) "Scalable analysis of authentic viral envelopes on FRONTERA" *Comput Sci Eng*
80. Shaw, Adams, Azaria et al. "Anton 3: twenty microseconds of molecular dynamics simulation before lunch"
81. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
82. Cohen "The long shot. Sci Technol Humanit 2024" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12691596&blobtype=pdf | # Low levels of influenza H5N1 HA and NA antibodies in the human population are boosted by seasonal H1N1 infection but not by H3N2 infection or influenza vaccination
Anne Werner, Cosette Schneider, Elgin Akin, Juliahna Hayes, Katherine Fenstermacher, Richard Rothman, Lynda Coughlan, Andrew Pekosz
## Abstract
An increase in the number of human cases of influenza A/H5N1 infection in the USA has raised concerns about the pandemic potential of the virus. Pre-existing population immunity is a key determinant for risk assessment and pandemic potential for any virus. Antibody responses against the bovine A/H5N1 hemagglutinin (HA) and neuraminidase (NA) proteins were measured among a population of influenza-vaccinated or influenza-infected individuals. Modest titers of bovine A/H5N1 HA-binding antibodies and low to undetectable neutralizing antibody titers were detected in a cohort of 73 individuals. Conversely, bovine A/H5N1 NA-binding and neuraminidaseinhibiting antibody titers were comparable to those against a human A/H1N1 NA at baseline. Seasonal influenza vaccination failed to significantly increase antibody titers against both HA and NA glycoproteins of bovine A/H5N1. Recent infection with human A/H1N1 but not A/H3N2 viruses induced significant increases in bovine A/H5N1-neu tralizing antibody, as well as increases in NA-binding and NA-inhibiting antibodies to bovine A/H5N1 NA. While the degree of protection afforded by these A/H5N1 cross-reac tive antibodies is not known, incorporating NA or enhancing current seasonal vaccine formulations to increase NA-specific antibody titers may increase antibody breadth and protection against both seasonal and pandemic influenza viruses. IMPORTANCE A/H5N1 influenza A viruses continue to pose a pandemic threat to humans. Recent infection of dairy cattle and poultry with A/H5N1 in the USA has magnified that concern. We determined the level of antibodies that recognize A/H5N1 hemagglutinin (HA) and neuraminidase (NA) proteins in a population in Baltimore, MD. We show that while low levels of H5 HA-binding and A/H5N1-neutralizing antibodies are present, there is a significantly stronger recognition of bovine N1 NA. Vaccines that target the N1 NA protein may induce protective antibody responses in humans due to the presence of cross-reactive human N1 NA antibodies.
prey (not usually hosts for avian influenza viruses), domesticated poultry, household cats, and dairy cows. The A/H5N1 clade 2.3.4.4b viruses have also undergone reassortment with other avian influenza viruses, leading to a number of different genotypes, including B3.13 (responsible for most dairy cow infections) and D1.1 (driving a large number of poultry farm infections) (5).
As of June 2025, the US Centers for Disease Control report 70 cases of A/H5N1 infection, primarily in individuals working in dairy or poultry farms with known exposures to infected animals (2). While one death has been reported, the vast majority of cases have resulted in relatively mild infection, often limited to conjunctivitis (6)(7)(8). This is in contrast to both outbreaks of A/H5N1 HPAI in prior years and contemporary outbreaks outside of North America (9)(10)(11). For example, of the 10 reported cases of A/ H5N1 in Cambodia in 2025 alone, 6 of them have been fatal (11,12). Sporadic spillovers have occurred to humans with no evidence of human-to-human transmission (1,3,4). A/ H5N1 viruses isolated from a subset of these infections have shown amino acid changes consistent with adaptation to replication in mammals, which highlights concerns about the virus evolving into a form that may lead to a pandemic (13)(14)(15).
Because A/H5NX (where X is one of a number of neuraminidase subtypes) influenza viruses have not circulated in the human population, the vast majority of humans lack detectable antibodies against the H5 hemagglutinin (HA) protein-the major antigenic target of the influenza immune response. However, human seasonal A/H1N1 viruses encode a neuraminidase (NA) protein that is antigenically and structurally similar to that of A/H5N1 viruses. Experiments in animals and surveys of human cohorts have indicated that antibodies to seasonal human A/H1N1 NA can cross-react with the A/H5N1 NA, but it is unknown if these low-level cross-reactive antibodies can provide protection against or mediate severity of A/H5N1 HPAI infection in humans (16)(17)(18)(19).
To determine the level of pre-existing human population immunity to bovinederived A/H5N1 viruses, serum samples from two distinct cohorts-healthcare workers vaccinated with the 2024 seasonal influenza vaccine and individuals infected with seasonal A/H1N1 or A/H3N2 influenza during the 2023-2024 influenza season-were tested for the presence of antibodies recognizing the H5 HA or N1 NA proteins. Functional antibody responses-neutralizing antibody (nAb) and neuraminidase-inhibit ing (NAI) antibody-against bovine A/H5N1 HA and NA were quantified in both cohorts. Low levels of binding IgG to the bovine H5 HA correlate with minimal-to-undetecta ble nAb titers to bovine A/H5N1 viruses, and neither binding IgG titers nor nAb are boosted by standard seasonal vaccination. Similarly, binding IgG and NAI antibody titers to human seasonal A/H1N1 NA and bovine A/H5N1 NA were quantified and were comparable at baseline (i.e., prior to vaccination or infection). Increases in NAspecific antibody to both human and bovine N1 NAs following infection with seasonal A/H1N1 viruses, but not A/H3N2 viruses, were detected. Lastly, despite no detectable change in N1-binding titers post-A/H3N2 infection, baseline serum depletion of A/H3N2 NA-specific antibody reduces total bovine N1-binding IgG, suggesting that heterotypic NA antibody may play a role in baseline cross-reactivity. Our findings corroborate and extend existing evidence that current seasonal vaccine formulations are poor at inducing NA-specific humoral responses (18,20).
## RESULTS
## Study design and demographics
Population-level antibody responses to A/H5N1 HPAI NA in healthy adults were investigated utilizing a cohort of individuals employed by the Johns Hopkins Hospital (JHH) or Johns Hopkins Medical Institutes (JHMI) who were receiving their annual dose of trivalent 2024-2025 Northern Hemisphere (NH) seasonal inactivated influenza vaccine in September and October 2024. Vaccinees provided blood samples at the time of enrollment, indicated as "day 0, " or "baseline, " and at day 28 post-vaccination (Fig. S1a). Fifty participants were selected, comprised of 25 males and 25 females, with ages ranging between 23 years and 68 years (Table 1). Forty-seven of 50 (94%) participants received egg-grown vaccine (Fluarix, GSK), and the remaining participants (6%) were immunized with standard trivalent cell-grown vaccine (Flucelvax, Seqirus) (Table 1). A minor subset (4%) of all participants reported that they had not received any influenza vaccination in any of the past five NH influenza seasons (Table 1). Of the remaining (96%) participants, the average number of previous vaccinated seasons was 4.46/5.
To investigate the role of seasonal IAV infection in shaping cross-reactive antibody responses to bovine A/H5N1 HA and NA, a cohort of patients who presented to the JHH Emergency Department or were inpatients at the JHH with influenza-like illness (ILI) with the confirmed seasonal IAV infection during the 2023-2024 Northern Hemisphere influenza season were utilized. IAV infection was confirmed via point-of-care diagnostic tests or next-generation sequencing. All 23 patients provided blood samples at the time of admittance, hereafter referred to as "baseline, " and again at approximately 4 weeks later, hereafter referred to as "convalescent" (Fig. S1b). Sixteen of 23 (69.6%) patients were infected with A/H1N1pdm09-like viruses, and 7 (30.4%) with A/H3N2-like viruses (Table 2). Fourteen of 23 (60.9%) participants were female, and 9 of 23 (39.1%) were male (Table 2). Fewer than four patients (13%) were solid organ transplant recipients and on immunosuppressive medication (Table 2). Fewer than four patients (8.7%) had previously undergone chemotherapeutic cancer treatment and were in remission at the time of the study (Table 2). Thirteen of 23 (56.7%) patients had reported receipt of a seasonal influenza vaccine for the 2023-2024 NH season (Table 2). For comorbidities, five patients (21.7%) reported cardiovascular disease, five patients (21.7%) reported hematological disorders, fewer than four patients (8.7%) reported renal disease, and fewer than four patients (13%) reported asthma (Table 2).
## Baseline cross-reactive antibodies to bovine A/H5N1 NA
To capture the range of baseline cross-reactivity across both cohorts, serum samples at the time of enrollment for the vaccine cohort and at the time of patient testing for the infection cohort were used to determine total binding IgG, NAI antibody, and nAb titers (Fig. 1). Baseline IgG binding titers were highly comparable against the bovine N1 NA and the NA from A/California/04/2009 (Cal09), which represents a human seasonal A/H1N1 pdm09-lineage NA. All participants had detectable binding IgG against the bovine and Cal09 NA at baseline, with geometric mean titers (GMTs) of 1,754.17 and 2,204.47, respectively (Fig. 1a). Although anti-Cal09 NA-binding titers were significantly correlated with birth year, this trend was less clear for anti-bovine N1 NA IgG (Fig. 1a). Baseline binding titers against both Cal09 NA and bovine A/H5N1 NA were significantly higher than baseline binding titers to bovine H5 HA, with a GMT of 364.88 (Fig. S2a). This suggests that pre-existing antibody titers against bovine N1 NA are greater than those against bovine H5 HA.
Using a recombinant bovine A/H5N1 virus whose multibasic cleavage site was deleted (∆MBS) and expressed in the genetic background of a live attenuated influenza vaccine virus (LAIV), hereafter referred to as bovine A/H5N1-LAIV, NAI titers were evaluated using an enzyme-linked lectin assay (ELLA). Unlike binding IgG, baseline NAI responses against the bovine N1 NA were markedly lower than those against a human seasonal N1 NA (Fig. 1b).
Bovine A/H5N1-LAIV nAb responses were significantly lower than nAb responses against human A/H1N1 (Fig. 1c), with 46% of subjects (24 of 50 vaccinees and 10 of 23 infected participants) having no detectable nAb titers at baseline (Fig. 1c). We detected a strong correlation between baseline H5-binding IgG and both participant birth year (Fig. S2a) and bovine A/H5N1 nAb titers (Fig. S2c). Conversely, only 0.04% of participants (3 of the 23 infected participants and none of the 50 vaccinees) had detectable nAb titers against A/H1N1 at baseline (Fig. 1c), which suggests that the detected H5-binding IgG were likely non-neutralizing in function. Taken altogether, pre-existing immunity at the population level to bovine A/H5N1 HPAI preferentially targets N1 NA rather than H5 HA.
## Seasonal vaccination does not boost cross-reactive bovine A/H5N1 responses
Seasonal vaccines are currently the only widely available prophylactic for seasonal influenza epidemics (21)(22)(23)(24). To address whether seasonal influenza vaccination can induce cross-reactive antibody responses to bovine A/H5N1 NA, pre-and post-vaccina tion NA-specific binding and inhibiting antibodies, as well as neutralizing antibodies, were quantified. It has been shown that although seasonal vaccination minimally increases detectable NA-specific antibodies, it often induces robust nAbs against the strains included in the vaccine formulation, primarily targeting the immunodominant HA head (21,22,24,(24)(25)(26). Unsurprisingly, seasonal vaccination did not significantly increase binding IgG titers against either human N1 NA or bovine N1 NA (Fig. 2a andb). ELLA revealed that vaccination had a statistically significant yet modest increase in NAI antibodies to human N1 NA, but no significant increases in NAI antibodies against bovine N1 NA (Fig. 2c andd). It should be noted that higher NAI titers post-vaccination may be partly influenced by HA-specific responses (27). Nonetheless, vaccination did not appear to boost cross-reactive NAI titers, which is consistent with binding IgG responses (Fig. 2a through d).
Because seasonal vaccines are formulated specifically with the aim of inducing neutralizing antibodies against the HA glycoprotein, we investigated whether vaccineinduced homotypic neutralizing antibodies could cross-neutralize A/H5N1 viruses. While nAb titers against A/H1N1pdm09 increased significantly after vaccination, there was no corresponding increase in A/H5N1-LAIV nAb titers (Fig. 2e andf). At day 28 post-vaccina tion, 13 of the initial 24 vaccinees who were seronegative remained seronegative, and two vaccinees saw a reduction in nAb titers to below the limit of detection (LOD) (Fig. 2f). There was no significant change, nor a consistent trend, in baseline binding IgG against bovine H5 HA following seasonal vaccination, as measured via enzyme-linked immu nosorbent assay (ELISA) (Fig. S3a). Altogether, the data indicate that current seasonal vaccine formulations do not induce significant levels of binding or functional antibodies that recognize bovine H5 HA or N1 NA proteins.
## Boosting of A/H5N1 cross-reactive antibodies by seasonal IAV infection
Animal studies have suggested that infection with human seasonal A/H1N1 viruses can provide partial or complete protection against an A/H5N1 challenge (28)(29)(30)(31). Unlike seasonal vaccination, IAV infection is known to reliably induce NA-specific antibody responses (18,23,27,32,33). NA-specific antibody responses were measured in our infection cohort (Fig. 3). For the 16 patients who had sequence-confirmed A/H1N1 infections, Cal09 NA-and bovine A/H5N1 NA-specific IgG were both substantially boosted by infection, albeit to different magnitudes (Fig. 3a andb), with a 151-fold increase in Cal09 N1 NA-binding antibodies and a 55-fold increase in bovine A/H5N1 NA-binding antibodies. For A/H3N2-infected patients, there was no consistent trend in binding IgG against either human or bovine N1 NA (Fig. 3a). These data suggest that A/H1N1, but not A/H3N2, infection boosts bovine N1 NA-specific IgG responses.
The trends we observed in NA-binding IgG for A/H1N1-infected patients were recapitulated in NAI titers (Fig. 3c andd). All but 1 of 16 patients increased in A/H1N1 NAI antibodies, and the fold increase was 62× (Fig. 3c). All 16 patients had a significant increase in A/H5N1-LAIV NAI titers post-infection but with a lower fold increase of 15.52× (Fig. 3d). A/H3N2-infected individuals did not show significant increases in NAI titers to human or bovine H1N1 viruses (Fig. 3c andd). There were also significant increases in nAb titers to both A/H1N1pdm09 and bovine A/H5N1 viruses after A/H1N1 infection but not after A/H3N2 infection (Fig. 3e andf). In addition to the increase in NAI and nAb responses to bovine A/H5N1-LAIV in the A/H1N1-infected group, H5 HA-binding IgG titers significantly increased post-infection by an average 10.34-fold (Fig. S3b). Altogether, A/H1N1-infected patients were the only group to see an increase in both HAand NA-specific cross-reactive antibodies against bovine A/H5N1 HA and NA proteins.
## Investigating the contribution of heterotypic N2-specific antibodies to total A/H5N1 NA-binding antibody
Given the relatively high degree of conservation with respect to structural and functional domains of the NA protein, we determined whether heterotypic antibody to the N2 NA may contribute to the detectable binding IgG at baseline. Pan-NA mAbs often target domains of NA that are essential for either enzymatic function or overall structure of the NA tetramer and therefore conserved across IAV subtypes (34)(35)(36)(37). To address the overall contribution of N2-specific binding IgG to the observed baseline bovine NA IgG titers, serum from a representative subset of our vaccine cohort-balanced by age and sexwas depleted of N2-binding IgG (Fig. S4a). We used a representative A/H3N2 recombi nant NA derived from a locally circulating isolate in Baltimore, MD, for all N2-serum depletions. On average, sera showed binding IgG titers that were threefold higher than N2-depleted sera against bovine NA (Fig. S4a), indicating N2 NA cross-reactive antibodies make up a small portion of the cross-reactive bovine NA antibodies. We confirmed that our N2-depleted sera had little to no detectable binding against the same A/H3N2 NA (Fig. S4b), further suggesting that pre-existing cross-reactive antibodies may be made up of both heterosubtypic NA-targeting antibodies as well as heterotypic NA-targeting antibodies. These findings suggest that heterotypic antibodies, namely those that bind N2 NA, may partially contribute to observed baseline bovine NA antibody responses and substantiate further investigation of cross-group neuraminidase antibodies.
## DISCUSSION
A key component of risk assessment for A/H5N1 HPAI is a comprehensive understand ing of pre-existing immunity at the population level (38)(39)(40). In the present work, we illuminate the distinct roles of seasonal human IAV infection and vaccination in shaping baseline cross-reactive antibody repertoire to the NA of A/H5N1 viruses. The majority of candidate pandemic HPAI vaccines against A/H5N1 are centered around the HA glycoprotein (17,(41)(42)(43)(44)(45). Similarly, available reports detailing the pre-existing antibody repertoire against A/H5N1 are focused on the detection of cross-reactive neutralizing antibodies, which target HA (45)(46)(47)(48)(49). Antibodies against NA are often not neutralizing in function (18,(50)(51)(52)(53)(54). While H5 HA is structurally and antigenically distinct from human seasonal H1 HA, they are part of the same phylogenetic group-group 1-and share some identity in sequence, particularly within the more conserved HA stem (49,(55)(56)(57). Given the immunodominance of HA during both vaccination and infection, there is a possibility that H5-specific responses at baseline are contributing to the detectable NAI response to bovine A/H5N1-LAIV (27) and might also be neutralizing. Other studies have detected low nAb responses against A/H5N1 clades (19,45,47,48). Nonetheless, it has long been established that NA-targeting antibodies are an important mediator of protective immunity against IAV, albeit these responses are often not captured in traditional serological assessments like neutralizing antibody assays and hemaggluti nation inhibition assays (18,32,33,52,(58)(59)(60)(61). Therefore, N1 NA acting as a shared target between human seasonal A/H1N1 viruses and A/H5N1 HPAI likely allows for the detectable cross-reactive NA-targeting responses we and others report (17,19,45,48).
Our work describes pre-existing immunity in a population of healthcare workers from JHH (Fig. 2). Understanding the level of baseline immunity against A/H5N1 in this population is essential when considering the pandemic potential of A/H5N1, and who is most at risk of occupational or incidental exposures-with healthcare workers being at high risk, only second to dairy and poultry farmers who have direct contact with infected animals. Because vaccination is mandated each season for JHH employees, this cohort allows us to investigate the vaccine-induced antibody repertoire as it pertains to cross-reactive bovine A/H5N1 HPAI responses without sacrificing the heterogeneity present at the population level, which is often lost in animal studies (29,46,62,63). Among healthcare workers, baseline binding IgG against Cal09 NA and against bovine A/ H5N1 NA were similar to those among our infection cohort at baseline. When considered with our correlation analyses, binding IgG against both Cal09 NA and bovine A/H5N1 NA appeared to be positively correlated with birth year, although this was not statistically significant for bovine A/H5N1 NA (Fig. 1a). This trend was consistent for NAI titers, but we saw a slight negative correlation for nAb responses to bovine A/H5N1-LAIV (Fig. 1b andc). When considering A/H5N1 HA responses, the slight negative correlation with birth year and nAb titer was more exaggerated between birth year and bovine A/H5N1 HA-binding IgG (Fig. S2a). These data highlight important factors and considerations that underlie observed heterogeneity in pre-existing immunity to A/H5N1, such as primary influenza exposures and original antigenic sin (64).
For seasonal influenza infections, A/H1N1 infection was clearly correlated with the induction of cross-reactive nAb and bovine N1 NA antibodies (Fig. 3), which was absent in A/H3N2-infected individuals. At least 25% of our infection cohort reported no receipt of a seasonal influenza vaccine in any of the past five seasons, and 11 of 23 had not received the seasonal vaccine during the season that they were infected (Table 2). This suggests that any differences in baseline NA-specific IgG are unlikely a consequence of vaccine-induced immune responses. However, it is worth noting that the size of our A/H3N2-infected cohort is limited, as the 2023-2024 NH influenza season was dominated by A/H1N1 viruses (65). Thus, we plan to extend this work to the most recent (e.g., the 2024-2025) NH season and to future influenza seasons to increase the sizes of both our A/H1N1-and A/H3N2-infected cohorts. Furthermore, while we detect increases in bovine H5 HA (Fig. S3c) and N1 NA-binding (Fig. 3b) antibody responses after infection, these levels are much lower than those targeting human seasonal HA or NA proteins. Since there are no established correlates of protection for A/H5N1 infection in the human population, we cannot make any conclusions about how these cross-reactive antibodies against bovine A/H5N1 might modulate infection or disease severity. Nonetheless, our data merit additional study for defining immune correlates of protection for both HAand NA-targeting responses.
Seasonal vaccination did not change H5-binding IgG titers or bovine A/H5N1-LAIV nAb titers, which is expected due to the immunodominance of the globular HA head in vaccine-induced antibody responses (21,22,25,66,67). This corroborates previ ously published data, which indicate that despite no known exposures, cross-reactive antibodies to H5 HA are detectable, albeit minimal (19,45,48). While NA has been increasingly considered as a potential immunogen for universal influenza vaccines, our current understanding of how NA-specific responses can mediate immunity is limited. Current practices for seasonal vaccine development and production do not include quantifying NA content (21,66), despite its role as the common denominator between human influenza viruses and several zoonotic influenza viruses (such as, but not limited to, swine A/H1N1, swine A/H1N2, swine A/H3N2, canine A/H3N2, avian A/H1N1, avian A/H2N2, avian A/H3N2, avian A/H5N1, avian A/H9N2, etc.). We show the relevance of A/ H1N1 infection, as only A/H1N1 infection was capable of significantly boosting binding, neutralizing, and NA-inhibiting antibodies against bovine A/H5N1 (Fig. 3). Relative to HA, NA is a slower-moving antigenic target and has lower mutational plasticity (21,61,66). It must maintain its vital enzymatic function for productive influenza virus infections, which provides a common target across not only IAVs but also extends to influenza B viruses. Despite the discovery of several pan-NA mAbs that have shown protection in lethal challenge models against a panel of human and zoonotic IAVs (18,32,34,35,37,(68)(69)(70), NA remains comparatively understudied and overlooked as a viral and vaccine antigen for broadly protective immunity (58,60,61,71). While there have been far fewer clinical trials investigating NA-based vaccine immunogenicity and efficacy, phase I studies of vaccines that include NA as an immunogen highlight its potential for induction of robust homotypic and heterotypic antibodies that are durable (72,73). Few NA-only vaccines have been evaluated in humans, although many have shown protection in lethal challenge models (16,33,34,36,59,70,(74)(75)(76). The lack of human immunogenic ity data, efficacy data, and transmission data for NA vaccine candidates opposes the push toward including NA in current vaccines. Until sufficient human clinical trials have been conducted to support non-inferiority relative to current seasonal influenza vaccines against pandemic strains, the total utility of including NA or switching to NA-only vaccine designs remains uncertain.
We show evidence that supports changing current seasonal vaccine formulations to either include greater NA content or to manipulate immunogen design to increase immunofocusing toward immunosubdominant domains of IAV HA and NA (25,66,77,78). The work presented here reiterates the importance of NA as a conserved antigen between human seasonal viruses and A/H5N1 HPAIs and underscores the need for investigation of NA-mediated antibody responses and their role in protective immun ity. NA-centered vaccine design would enable robust boosting of cross-reactive N1 antibodies and may serve as a more feasible approach to increasing population-level pre-existing antibodies to A/H5N1 compared to HA-focused vaccines.
## MATERIALS AND METHODS
## Human subjects enrollment, sampling, and data collection
Serum used in this study was obtained from healthcare workers recruited during the annual Johns Hopkins Hospital employee influenza vaccination campaign in the Fall of 2024 by the Johns Hopkins Centers for Influenza Research and Response. Pre-(immedi ately prior to vaccination) and post (~28 day) vaccination human serum was collected from subjects, who provided written informed consent prior to participation. Patients were enrolled at the JHMI Department of Emergency Medicine or on inpatient floors. Symptomatic patients in the emergency department were screened and tested for influenza from triage by clinical providers using a validated clinical decision guideline tool. Serum was collected at the time of presentation and approximately 28 days later.
## Viruses
Virus isolates used in this study are as follows: A/Victoria/4897/2022 (A/H1N1 pan demic09-like, 2024-2025 NH vaccine virus; courtesy of John Steel, CDC). Recombinant viruses used in this study include A/Baltimore/R0675/2019 (A/H1N1 pandemic09-like, GISAID accession no. EPI_ISL_17617226) and an LAIV-like virus expressing the HA and NA segments of bovine A/H5N1 (GISAID accession no. EPI_ISL_19014384), which were used for ELLA quantification of NAI antibody and neutralizing titer 50% (NT 50 ) assay for detection of nAb responses. A/Baltimore/R0675/2019 was generated by a 12-plasmid reverse genetics system for the generation of influenza A viruses (79). Transfection of co-cultured Madin-Darby Canine Kidney (MDCK)-SIAT1 and HEK-293T cells with plasmids encoding each of eight segments belonging to A/Baltimore/R0675/2019 and four helper plasmids encoding the polymerase complex of influenza viruses was conducted as previously described (80,81). Recombinant viruses expressing the HA and NA segments of A/Bovine/Texas/24-029328-01/2024 (GISAID accession no. EPI_ISL_19014384) were generated as follows: the multibasic cleavage site in HA was mutated to delete the RRKR motif (amino acid positions 342 to 346) to make the virus dependent upon exogenous trypsin (28). HA and NA segments from A/Bovine/Texas/24-029328-01/2024 were cloned into an eight-plasmid expression system with bidirectional promoters (courtesy of Dr. Seema Lakdawala, Emory University, and Dr. Valerie Le Sage, University of Pittsburgh) (80,81). The six remaining gene segments encoding the internal genes of A/Ann Arbor/6/1960 (H2N2) (GISAID accession no. EPI_ISL_130415), a cold-adapted virus (live attenuated, indicated as LAIV), cloned into the 12-plasmid reverse genetics system, were co-transfected with the bovine-derived A/H5N1 HA and NA described above, with co-cultured MDCKI and HEK-293T cells, as previously described (79)(80)(81)(82)(83).
## Cell lines and maintenance
MDCK derivatives, MDCKI and MDCK-SIAT1 (courtesy of Dr. Scott Hensley, University of Pennsylvania), were maintained in cell culture in complete media-hereafter referred to as CM-consisting of Dulbecco's modified Eagle medium (DMEM; Gibco) supplemen ted with 10% fetal bovine serum (FBS; Gibco), 100 U/mL penicillin/streptomycin (Life Technologies), and 2 mM GlutaMAX (Gibco). Cells were passaged by washing 2× with PBS (1×; Life Technologies), followed by treatment with trypsin-EDTA (0.5%) (Gibco) and incubation at 37°C for up to 15 minutes, at which point cells had detached. Trypsin was quenched by the addition of an equal volume of CM. Cells were either subsequently passaged or were plated to be used for virus propagation, virus quantification by tissue culture infectious dose 50% (TCID 50 ) assay, and NT 50 assay.
## Recombinant neuraminidase HA and NA protein
Recombinant bovine A/H5N1 HA (derived from the A/dairy cow/Texas/24-008749-002v/2024 strain) was obtained from Sino Biological (cat no. 41036-V08H). Protein was reconstituted per the manufacturer's instructions to 1 mg/mL and stored at -80°C until use. NA sequences were designed as previously described (84,85). Briefly, the cytoplasmic, transmembrane, and stalk domains of wild-type NA were replaced with an N-terminal signal sequence, 6×His tag, a tetramerization domain from the human vasodilator-stimulated phosphoprotein, a thrombin cleavage site, and a linker sequence followed by the NA sequence (86). NA constructs were expressed in Expi293F cells and purified by Ni-NTA chromatography as previously described (87).
## ELISA for quantification of antigen-specific IgG
Antigen-specific IgG was quantified by ELISA as previously described (88,89). Protein was diluted in 1× PBS (Life Technologies) to 1 µg/mL for all neuraminidase constructs and to 0.5 µg/mL for bovine H5 hemagglutinin (Sino Biological), added to Nunc MaxiSorp 96-well plates (Thermo Fisher), and incubated at 4°C for 16 hours. Plates were washed with 1× PBS supplemented with 0.1% Tween 20 (Sigma), designated as PBST. Heat-inacti vated serum samples were serially diluted fourfold, eight times in blocking buffer, which was composed of PBST + 5% skim milk (Thermo Fisher). Diluted sera were added to plates in duplicate and incubated at room temperature for 1 hour. Goat anti-human IgG (gamma chain) cross-adsorbed horseradish peroxidase (HRP) conjugate was used as a secondary antibody. TMB substrate (Thermo Fisher) was added to all wells and incubated in the dark for 18-20 minutes. The reaction was stopped by the addition of 0.16 M sulfuric acid, and plates were read at OD 450 and OD 650 with background subtraction. Reciprocal endpoint titers were determined as the serum dilution that yielded signal four times that of secondary antibody alone.
## ELLA for quantification of NAI antibody responses
ELLA was used to quantify neuraminidase activity-inhibiting antibodies. In brief, Nunc 96-well Immulon 4 HBX plates (Thermo Fisher) were coated with fetuin from FBS (Sigma) diluted in 1× PBS to 2.5 µg/well. Plates were sealed and stored at 4°C for 12 to 16 hours. Sera were serially diluted fourfold, eight times in assay buffer, which consisted of 0.2% Tween 20 (Sigma), 1% BSA (Sigma), 0.1 mg/mL MgCl 2 , and 0.2 mg/mL CaCl 2 , diluted in 1× PBS. A/H1N1 or A/H5N1-LAIV was appropriately diluted in assay buffer and added to serially diluted serum and incubated at 37°C for 1 hour. During the 1 hour incubation, fetuin-coated plates were washed 3× with PBST. Sera/virus mixtures were added to washed plates in duplicate, and each plate contained eight wells with only assay buffer (to represent no NA activity, or 100% NAI) and an additional eight wells with virus alone and no antibody (to represent full NA activity, or 0% NAI). Plates were then sealed and incubated for 16 to 18 hours at 37°C with 5% CO 2 . Plates were carefully washed with PBST and incubated at room temperature for 2 hours with HRP-conjugated lectin from Arachis hypogaea, also referred to as HRP-conjugated peanut agglutinin (Sigma). Plates were washed a final time and then reacted with SigmaFast OPD substrate (Sigma) away from direct light for approximately 10 minutes. Reactions were stopped by the addition of an equivalent volume of 1 N H 2 SO 4 , and OD was read at 490 nm to determine relative NAI.
## NT 50 assay for quantification of nAb responses
NT 50 assays were conducted as previously reported (90). Human sera obtained as described above were treated with lyophilized receptor-destroying enzyme II (RDE; Hardy Diagnostics) per the manufacturer's instructions. For A/H1N1 and A/H5N1-LAIV viruses, MDCK-SIAT1 cells (courtesy of Dr. Scott Hensley) and MDCKI cells were seeded in CM in 96-well plates (Celltreat) 2 days prior to infection. RDE-treated sera were serially diluted in DMEM supplemented with 100 U/mL penicillin/streptomycin (Life Technolo gies), 2 mM GlutaMAX (Gibco), 0.3% bovine serum albumin (Sigma), and 1 or 5 µg/mL of N-acetylated trypsin (Sigma) for assays conducted on MDCK-SIAT1 cells or on MDCKI cells, respectively. One hundred TCID 50 was added to each well of serially diluted serum and incubated at 33°C with 5% CO 2 for 1 hour prior to infecting cells. The serum and virus mixture was added to cells in quadruplicate and incubated at 33°C 5% CO 2 for 24 hours. After 24 hours, all plates were washed twice with 1× PBS supplemented with CaCl 2 and MgCl 2 , and the media were replaced before plates were returned to 33°C with 5% CO 2 . One hundred twenty hours later, plates were fixed with 10% neutral buffered formalin (Leica) and stained with naphthol blue-black for subsequent interpretation. Area under the curve was calculated using GraphPad Prism 10.4.2. Curves were generated by entering the fraction of all four wells per sample protected at each dilution factor. The LOD was determined to be the smallest possible AUC value generated, e.g., when only one of four wells was protected at the first dilution in the full dilution series. Any samples that had no detectable neutralizing titer were set to be equal to one-half of the assay LOD for use in calculating GMTs.
## Serum depletion of antigen-specific antibody
Baseline human sera was depleted of antigen-specific antibody by coupling His-tag ged recombinant A/H3N2 NA tetramers from A/Baltimore/R0145/2017(H3N2) (GenBank: MH637451) to magnetic His Dynabeads (Invitrogen; Thermo Fisher) as previously described (89). Briefly, 5 µL of heat-inactivated sera was incubated with 195 µL of N2Baltimore2017 diluted in PBS, for a total of 200 µL, at room temperature with constant agitation for 1 hour. Washed magnetic beads were resuspended in 50 µL of PBS and incubated at room temperature with constant agitation for approximately 30 minutes. Tubes containing the above three components were placed on a magnetic strip (Invitrogen; Thermo Fisher) and incubated until all beads had precipitated out of solution. The supernatant was collected and subjected to ELISA to confirm the success ful elimination of all N2Baltimore2017-specific IgG and to quantify remaining bovine A/H5N1 NA-binding IgG. acquisition, Resources, Writing -review and editing | Richard E. Rothman, Funding acquisition, Resources, Writing -review and editing | Lynda Coughlan, Resources, Writing -review and editing | Andrew Pekosz, Conceptualization, Funding acquisition, Methodol ogy, Resources, Supervision, Writing -original draft, Writing -review and editing
## References
1. Krammer, Hermann, Rasmussen (2025) "Highly pathogenic avian influenza H5N1: history, current situation, and outlook" *J Virol*
2. Cdc (2024) "H5 Bird Flu: current situation. avian influenza"
3. Cdc Archive
4. Van Kerkhove (2013) "Brief literature review for the WHO global influenza research agenda--highly pathogenic avian influenza H5N1 risk in humans" *Influenza Other Respir Viruses*
5. Mostafa, Nogales, Martinez-Sobrido (2025) "Highly pathogenic avian influenza H5N1 in the United States: recent incursions and spillover to cattle" *Npj Viruses*
6. "Highly pathogenic avian influenza A(H5N1) virus infection in a dairy farm worker" *N Engl J Med*
7. "Highly pathogenic avian influenza A(H5N1) virus infections in humans" *N Engl J Med*
8. (2025) "N1-related human death. Louisiana Department of Health"
9. (2015) "World Health Organization -Regional Office for the Eastern Mediterranean"
10. Refaey, Azziz-Baumgartner, Amin et al. (2015) "Increased number of human cases of influenza virus A(H5N1) infection"
11. Cidrap (2025) "H5N1 avian flu infects a fifth patient in Cambodia"
12. Cidrap (2025) "Cambodia confirms 2 more H5N1 avian flu infections"
13. Maccosham, Vasiliu, Atchessi (2025) "A rapid review of the avian influenza PB2 E627K mutation in human infection studies" *Can Commun Dis Rep*
14. Peacock, Moncla, Dudas et al. (2025) "The global H5N1 influenza panzootic in mammals" *Nature*
15. Arai, Kawashita, Hotta et al. (2018) "Multiple polymerase gene mutations for human adaptation occurring in Asian H5N1 influenza virus clinical isolates" *Sci Rep*
16. Liu, Lin, Tsou et al. (2015) "Cross-reactive neuraminidase-inhibiting antibodies elicited by immunization with recombinant neuraminidase proteins of H5N1 and pandemic H1N1 influenza A viruses" *J Virol*
17. Li, Liu, Jung et al. "2025 Population immunity to hemagglutinin head, stalk and neuramini dase of highly pathogenic avian influenza 2.3.4.4b A(H5N1) viruses in the United States and the impact of seasonal influenza on A(H5N1) immunity" *medRxiv*
18. Chen, Wohlbold, Zheng et al. (2025) *Research Article mBio December*
19. Topham, Treanor, Wrammert et al. (2018) "Influenza infection in humans induces broadly cross-reactive and protective neuraminidase-reactive antibodies" *Cell*
20. Daulagala, Cheng, Chin et al. (2024) "Avian Influenza A(H5N1) neuraminidase inhibition antibodies in healthy adults after exposure to influenza A(H1N1)pdm09" *Emerg Infect Dis*
21. Rajendran, Krammer, Mcmahon (2021) "The human antibody response to the influenza virus neuraminidase following infection or vaccination" *Vaccines (Basel)*
22. Fiore, Bridges, Cox (2009) "Seasonal influenza vaccines"
23. Ertesvåg, Cox, Lartey et al. (2022) "Seasonal influenza vaccination expands hemagglutinin-specific antibody breadth to older and future A/H3N2 viruses" *NPJ Vaccines*
24. Who (2023)
25. "Immunological assessment of influenza vaccines and immune correlates of protection"
26. Andrews, Cominsky, Shimberg et al. (2023) "An influenza H1 hemagglutinin stem-only immunogen elicits a broadly cross-reactive B cell response in humans" *Sci Transl Med*
27. Nachbagauer, Palese (2018) "Development of next generation hemagglutinin-based broadly protective influenza virus vaccines" *Curr Opin Immunol*
28. Rajendran, Nachbagauer, Ermler et al. (2017) "Analysis of antiinfluenza virus neuraminidase antibodies in children, adults, and the elderly by ELISA and enzyme inhibition: evidence for original antigenic sin" *mBio*
29. Sage, Werner, Merrbach et al. (2025) "Influenza A(H5N1) immune response among ferrets with influenza A(H1N1)pdm09 immunity" *Emerg Infect Dis*
30. Nuñez, Jang, Huang et al. (2023) "Influenza virus immune imprinting dictates the clinical outcomes in ferrets challenged with highly pathogenic avian influenza virus H5N1" *Front Vet Sci*
31. Van Reeth, Braeckmans, Cox et al. (2009) "Prior infection with an H1N1 swine influenza virus partially protects pigs against a low pathogenic H5N1 avian influenza virus" *Vaccine (Auckl)*
32. Kucharski, Edmunds (2015) "Cross-immunity and age patterns of influenza A(H5N1) infection" *Epidemiol Infect*
33. Doyle, Hashem, Li et al. (2013) "Universal anti-neuraminidase antibody inhibiting all influenza A subtypes" *Antiviral Res*
34. Abbadi, Mousa (2023) "Broadly protective neuraminidase-based influenza vaccines and monoclonal antibodies: target epitopes and mechanisms of action" *Viruses*
35. Yasuhara, Yamayoshi, Kiso et al. (2022) "A broadly protective human monoclonal antibody targeting the sialidase activity of influenza A and B virus neuraminidases" *Nat Commun*
36. Hansen, Mcmahon, Turner et al. (2023) "Human anti-N1 monoclonal antibodies elicited by pandemic H1N1 virus infection broadly inhibit HxN1 viruses in vitro and in vivo" *Immunity*
37. Wohlbold, Podolsky, Chromikova et al. (2017) "Broadly protective murine monoclonal antibodies against influenza B virus target highly conserved neuraminidase epitopes" *Nat Microbiol*
38. Lederhofer, Tsybovsky, Nguyen et al. (2024) "Protective human monoclonal antibodies target conserved sites of vulnerability on the underside of influenza virus neuraminidase" *Immunity*
39. (2024) "Pandemic Flu"
40. "Tool for Influenza pandemic risk assessment (TIPRA) 2nd Edition"
41. "Risk assessment of a highly pathogenic H5N1 influenza virus from mink" *Nat Commun*
42. Crevar, Carter, Lee et al. (2015) "Cocktail of H5N1 COBRA HA vaccines elicit protective antibodies against H5N1 viruses from multiple clades" *Human Vaccines Immunother*
43. Baz, Luke, Cheng et al. (2013) "H5N1 vaccines in humans" *Virus Res*
44. Khurana, King, Manischewitz et al. (2024) "Licensed H5N1 vaccines generate cross-neutralizing antibodies against highly pathogenic H5N1 clade 2.3.4.4b influenza virus" *Nat Med*
45. Easterbrook, Schwartzman, Gao et al. (2012) "Protection against a lethal H5N1 influenza challenge by intranasal immunization with viruslike particles containing 2009 pandemic H1N1 neuraminidase in mice" *Virology (Auckl)*
46. Pichyangkul, Krasaesub, Jongkaewwattana et al. (2014) "Pre-existing cross-reactive antibodies to avian influenza H5N1 and 2009 pandemic H1N1 in US military personnel" *Am J Trop Med Hyg*
47. Sage, Werner, Merrbach et al. (2024) "Pre-existing H1N1 immunity reduces severe disease with bovine H5N1 influenza virus"
48. Mellis, Coyle, Marshall et al. (2024) "Serologic evidence of recent infection with highly pathogenic Avian influenza A(H5) virus among dairy workers -Michigan and Colorado" *MMWR Morb Mortal Wkly Rep*
49. Daniel, Ullrich, Ruchnewitz et al. (2025) "Detection of pre-existing humoral immunity against influenza virus H5N1 clade 2.3.4.4b in unexposed individuals" *Immunology*
50. Beukenhorst, Frallicciardi, Rice et al. (2024) "A pan-influenza monoclonal antibody neutralizes H5 strains and prophylactically protects through intranasal administration" *Sci Rep*
51. Cho, Wrammert (2016) "Implications of broadly neutralizing antibodies in the development of a universal influenza vaccine" *Curr Opin Virol*
52. "The neuraminidase activity of influenza A virus determines the strainspecific sensitivity to neutralization by respiratory mucus" *J Virol htt*
53. Air (2012) "Influenza neuraminidase" *Influenza Other Respir Viruses*
54. Momont, Dang, Zatta et al. (2023) "A pan-influenza antibody inhibiting neuraminidase via receptor mimicry" *Nature*
55. Sun, Ma, Wang et al. (2024) "Broadly neutraliz ing antibodies to combat influenza virus infection" *Antiviral Res*
56. Stevens, Blixt, Tumpey et al. (2006) "Structure and receptor specificity of the hemagglutinin from an H5N1 influenza virus" *Science*
57. Wu, Wilson (2020) "Influenza hemagglutinin structures and antibody recognition" *Cold Spring Harb Perspect Med*
58. Corbett, Moin, Yassine et al. (2019) "Design of nanoparticulate group 2 influenza virus hemagglutinin stem antigens that activate unmutated ancestor B cell receptors of broadly neutralizing antibody lineages" *mBio*
59. Creytens, Pascha, Ballegeer et al. (2021) "Influenza neuraminidase characteristics and potential as a vaccine target" *Front Immunol*
60. Ellis, Lederhofer, Acton et al. (2022) "Structure-based design of stabilized recombinant influenza neuraminidase tetramers" *Nat Commun*
61. Eichelberger (2015) "Influenza neuraminidase as a vaccine antigen"
62. Krammer, Fouchier, Eichelberger et al. (2018) "NAction! how can neuraminidase-based immunity contribute to better influenza virus vaccines?" *MBio*
63. Carreño, Strohmeier, Roubidoux et al. (2020) "H1 hemagglutinin priming provides long-lasting heterosubtypic immunity against H5N1 challenge in the mouse model" *mBio*
64. Jang, Ross (2019) "Preexisting influenza specific immunity and vaccine effectiveness" *Expert Rev Vaccines*
65. Daulagala, Mann, Leung et al. (2023) "Imprinted anti-hemagglutinin and anti-neuraminidase antibody responses after childhood infections of A(H1N1) and A(H1N1)pdm09 influenza viruses" *MBio*
66. Cdc (2024) "Influenza activity in the United States during the 2023-2024 season and composition of the 2024-2025 influenza vaccine"
67. Kanekiyo, Graham (2021) "Next-generation influenza vaccines" *Cold Spring Harb Perspect Med*
68. Krammer, Smith, Fouchier et al. (2018) *Influenza. Nat Rev Dis Primers*
69. Kirkpatrick Roubidoux, Mcmahon, Carreño et al. (2021) "Identification and characterization of novel antibody epitopes on the N2 neuraminidase"
70. Stadlbauer, Zhu, Mcmahon et al. (2019) "Broadly protective human antibodies that target the active site of influenza virus neuraminidase" *Science*
71. Mcmahon, Tan, Dell et al. (2023) "Immunity induced by vaccination with recombinant influenza B virus neuraminidase protein breaks viral transmission chains in guinea pigs in an exposure intensity-dependent manner" *J Virol*
72. Schneider, Coughlan (2023) "Always a bridesmaid, never a bride: Committing to flu neuraminidase as a vaccine target" *Mol Ther*
73. Spergel, Lee, Koslovsky et al. (2025) "Immunogenicity and safety of mRNA-based seasonal influenza vaccines encoding hemagglutinin and neuraminidase" *Nat Commun*
74. Sahly, Anderson, Jackson et al. (2025) "Anti-neuraminidase and anti-hemagglutinin stalk responses to different influenza a" *Vaccine (Auckl)*
75. Strohmeier, Amanat, Campbell et al. (2022) "A CpG 1018 adjuvanted neuraminidase vaccine provides robust protection from influenza virus challenge in mice" *NPJ Vaccines*
76. Pekarek, Petro-Turnquist, Jeanjaquet et al. (2025) "Synthetic neuraminidase vaccine induces cross-species and multi-subtype protection" *Vaccines (Basel)*
77. Cortés, Ustyugova, Farrell et al. (2024) "Boosting neuraminidase immunity in the presence of hemagglutinin with the next generation of influenza vaccines" *NPJ Vaccines*
78. Sangesland, Lingwood (2021) "Antibody focusing to conserved sites of vulnerability: the immunological pathways for "universal" influenza vaccines" *Vaccines (Basel)*
79. Ataca, Sangesland, De et al. (2024) "Modulating the immunodominance hierarchy of immunoglobulin germline-encoded structural motifs targeting the influenza hemaggluti nin stem" *Cell Rep*
80. Neumann, Watanabe, Ito et al. (1999) "Generation of influenza A viruses entirely from cloned cDNAs" *Proc Natl Acad Sci*
81. Perez (2017) "Plasmid-based reverse genetics of influenza A virus"
82. Perez, Seibert, Ferreri et al. (2020) "Plasmid-based reverse genetics of influenza A virus"
84. Canaday, Resnick, Liu et al. (2022) "HA and M2 sequences alter the replication of 2013-16 H1 live attenuated influenza vaccine infection in human nasal epithelial cell cultures" *Vaccine (Auckl)*
85. Powell, Liu, Pekosz (2021) "Changes in sialic acid binding associated with egg adaptation decrease live attenuated influenza virus replication in human nasal epithelial cell cultures" *Vaccine (Auckl)*
86. Xu, Zhu, Dwek et al. (2008) "Structural characteriza tion of the 1918 influenza virus H1N1 neuraminidase" *J Virol*
87. Margine, Palese, Krammer (2013) "Expression of functional recombinant hemagglutinin and neuraminidase proteins from the novel H7N9 influenza virus using the baculovirus expression system" *J Vis Exp*
88. Kühnel, Jarchau, Wolf et al. (2004) "The VASP tetramerization domain is a right-handed coiled coil based on a 15-residue repeat" *Proc Natl Acad Sci*
89. Bliss, Freyn, Caniels et al. (2022) "A single-shot adenoviral vaccine provides hemagglu tinin stalk-mediated protection against heterosubtypic influenza challenge in mice" *Mol Ther*
90. Hsieh, Werner, Leist et al. (2021) "Stabilized coronavirus spike stem elicits a broadly protective antibody" *Cell Rep*
91. Hutchinson, Abiona, Ziwawo et al. (2023) "Nanoparticle display of prefusion coronavirus spike elicits S1-focused cross-reactive antibody response against diverse coronavirus subgenera" *Nat Commun*
92. Swanson, Marinho, Dziedzic et al. (2019) "2019-2020 H1N1 clade A5a.1 viruses have better in vitro fitness compared with the co-circulating A5a.2 clade" *Sci Rep*
93. "Pipe-friendly framework for basic statistical tests" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724227&blobtype=pdf | # Four nudivirus core genes present in the genome of Venturia canescens are required for virus-like particle formation and prevention of encapsulation of parasitoid wasp eggs
Meng Mao, Corinne Stouthamer, Ange Lorenzi, Michael Strand, Gaelen Burke
## Abstract
Venturia canescens is a parasitoid wasp that harbors a domesticated endogenous virus (DEV) and parasitizes host insects like Ephestia kuehniella. The V. canescens DEV evolved from an alphanudivirus and produces virus-like particles (VLPs) in females that protect wasp eggs from a host immune defense called encap sulation. In contrast, very few DEV genes required for VLP formation and function have been identified. In this study, we characterized five V. canescens DEV genes of unknown function that all nudiviruses encode. Three of these genes are single copy (OrNVorf18-like, OrNVorf61-like, and OrNVorf76-like), while OrNVorf41-like has expanded into a six-member family and OrNVorf47-like has expanded into a three-member family. Sequence analysis indicated all of these genes retain essential motifs present in nudivirus homologs, while transmission electron microscopy (TEM) studies characterized the timing of VLP formation during the wasp pupal stage. RNA interference (RNAi) assays identified OrNVorf18-like, OrNVorf61-like, OrNVorf41-like-1, and OrNVorf41-like-2 as genes that are required for normal VLP formation. Knockdown of OrNVorf47-like family members did not affect VLP formation but did disable binding of VLPs to V. canescens eggs and protection against encapsulation. Disabled formation of VLPs in response to RNAi knockdown of OrNVorf18-like, OrNVorf61-like, OrNVorf41-like-1, and OrNVorf41-like-2 also resulted in wasp eggs being encapsulated. In contrast, knockdown of OrNVorf76-like had no effect on VLP assembly, egg binding, or encapsulation. Altogether, reported results significantly advance our understanding of V. canescens VLP (VcVLP) formation and function.
IMPORTANCE Understanding how V. canescens coopted an alphanudivirus to produceVcVLPs is of interest to the study of virus evolution. Our results show that three nudivirus core genes have essential functions in VcVLP formation, while one is essential for the novel function of binding to wasp eggs and protection from encapsulation, which is the most important immune defense of insects against parasitoids. KEYWORDS parasitoid, nudivirus, endogenous virus element, virus-like particle P arasitoid wasps (order Hymenoptera) are likely the most diverse animal group on Earth, with estimates of more than a million species (1, 2). Most are free-living insects as adults that parasitize other insects by laying eggs on or in their bodies (3). Viruses in the family Nudiviridae have large, circular double-stranded (ds) DNA genomes, specifically infect insects or other arthropods, and replicate in the nuclei of infected cells where they produce virions consisting of an envelope and one or more capsids (4). Different nudiviruses have integrated into the germline of certain parasitoid lineages and become Domesticated Endogenous Viruses (DEVs) (5). Each retains many nudivirus genes, which are specifically expressed in ovary calyx cells of females and produce either
DNA-containing virions or virus-like particles (VLPs) (6)(7)(8)(9). DEV-produced virions or VLPs cannot replicate, but wasps maintain the ability to produce them because all genome components necessary to do so are vertically transmitted to offspring. Each wasp lineage has also convergently evolved to use DEV-produced virions or VLPs to transfer genes, proteins, or other factors to host insects, which promote the successful development of offspring.
The most studied of these DEVs are bracoviruses (BVs), which evolved from a betanudivirus-like ancestor that integrated ~100 million years ago into a parasitoid in the family Braconidae (6,10). Subsequent speciation has resulted in thousands of braconids that produce BV virions, which consist of an envelope, one or more capsids, and DNAs that are amplified from the wasp genome and packaged into capsids (11,12). In contrast, all of the genes with functions in producing virions remain integrated in the genomes of wasps. Many of the genes with virion-producing functions have been characterized using RNA interference (RNAi) assays in conjunction with transmission electron microscopy (TEM) and other methods (13)(14)(15). The BV virions that females inject into hosts when laying eggs infect different cell types, which is followed by the expression of virulence genes that reside on the DNAs that are packaged into capsids (16,17). Several of these virulence genes have also been characterized and shown to alter host immune defenses, growth, and behavior (12,(18)(19)(20).
Less studied are ichneumonid wasps in the genus Venturia, where species like V. canescens contain a DEV that evolved from an alphanudivirus similar to Oryctes rhinoceros nudivirus (OrNV) (8). The V. canescens DEV produces 130 nm VLPs that consist of a spheroidal envelope containing proteins but no capsid or nucleic acid (21,22). V. canescens VLPs (VcVLPs) assemble in calyx cell nuclei and accumulate in the calyx lumen (22,23). Proteomic analysis of VcVLPs identifies many nudivirus gene products (8), while prior findings also show that VcVLPs associate with V. canescens eggs when passaging through the calyx lumen before storage in the lateral oviducts (22). V. canescens females parasitize larval stage moths like Ephestia kuehniella by usually laying one egg per host. Experiments indicate that the association of VcVLPs with eggs is also involved in protection from host hemocytes, which otherwise mount a fatal defense response called encapsulation (23,24).
All sequenced viruses in the Nudiviridae encode 32 core genes, of which 21 are shared with viruses in the Baculoviridae that also have large, circular dsDNA genomes, infect insects or other arthropods, and replicate in host cell nuclei (25). None of these shared core genes have been experimentally studied in nudiviruses, whereas all have been characterized in model baculoviruses (26). Genome sequencing of V. canescens followed by a chromosome-level assembly of the genome identifies 78 nudivirus genes (8,27). Most reside in six clusters on three chromosomes, while 13 are dispersed. Sixteen of the 21 core genes shared with baculoviruses are present, including the following: helicase, required for replication of baculovirus genomes (28); lef-4, lef-5, lef-8, lef-9, and p47, which encode subunits of a DNA-dependent RNA polymerase that transcribes other viral genes (26,29); and several pif genes that encode envelope proteins (30). Two shared core genes that are known to encode capsid proteins in baculoviruses and nudiviruses (vp39 and 38K) (5,26) are pseudogenized in V. canescens, which likely explains why VcVLPs lack a capsid (31). Reciprocally, RNAi assays indicate the nudivirus-derived RNA polymerase subunits in V. canescens retain conserved functions by producing a holoenzyme that transcribes other DEV genes (32).
Most of the other nudivirus genes in the V. canescens genome share no sequence similarity with other genes and are thus named by the OrNV homolog they are most closely related to (8,27). Some of these genes are alphanudivirus-specific, others are in some but not all sequenced nudiviruses, while five are core genes present in all sequenced nudiviruses but not baculoviruses (8,27). Three of these genes are sin gle copy (OrNVorf18-like, OrNVorf61-like, and OrNVorf76-like), while OrNVorf41-like has expanded into a six-member gene family and OrNVorf47-like has expanded into a three-member family (27). The aforementioned proteomic analysis of VcVLPs detected proteins corresponding to OrNVorf61-like and some of the OrNVorf47-like and OrNVorf41like family members, but did not detect products from OrNVorf18-like or OrNVorf76-like (8). Here, we conducted studies to further characterize these nudivirus core genes. We report that OrNVorf18-like, two OrNVorf41-like family members, and OrNVorf61-like are required for normal VcVLP formation, while the OrNVorf47-like family is essential for VcVLP attachment to eggs. We also show that each of these genes is required for protecting V. canescens eggs from encapsulation.
## RESULTS
## Each of the OrNV-like nudivirus core genes encodes full-length proteins
Nudivirus, baculovirus, and BV genes with replication functions are transcribed in temporally ordered cascades (4,15,26,33,34). Early genes expressed at the beginning of a replication cycle include the RNA polymerase subunits, while many late genes are transcribed by the viral RNA polymerase and encode virion components. Transcriptome data similarly classify the RNA polymerase subunits and select other V. canescens DEV genes as early because they are upregulated in the ovaries of younger pupae, while the five genes we focused this study on, plus several others, are classified as late because they are upregulated in mid-stage pupae (8,32). We began this study by asking if any of the nudivirus core gene products we prioritized for investigation had transmembrane domains, as might be expected if they encode VLP structural proteins. A single hydro phobic domain was present in the N-terminus of OrNVorf47-like-3 and OrNVorf76-like, while internal hydrophobic domains were present in OrNVorf41-like-1-6 and OrNVorf61like (Fig. S1). In contrast, no hydrophobic domains were present in OrNVorf18-like, OrNVorf47-like-1, or OrNVorf47-like-2. None of these domains were predicted to be signal peptides by SignalP 6.0, while all were predicted to form alpha transmembrane domains by DeepTMHMM-1.0. Alignments indicated each of these genes was full length and retained conserved residues when compared to corresponding homologs in OrNV and other nudiviruses (Fig. S2 to S6). The short length of OrNVorf41-like family mem bers in V. canescens made phylogenetic resolution of relationships difficult (Fig. S2). In contrast, results well-supported clustering of the three OrNVorf47-like family members, which suggested this family arose by duplication (Fig. S3). Each of the genes in V. canescens also clustered most closely to homologs in alphanudiviruses, which supported earlier conclusions that VcVLPs evolved from a virus in this genus (8).
## VcVLP morphogenesis proceeds through three phases in females injected with ds-egfp
Previous TEM data describe VcVLPs in calyx cells from older pupae and newly emerged adults (8,22,23,32). Calyx cell nuclei contain structures originally named dense bodies, VLP envelopes, and mature VLPs that are distinguished by electron-dense material that is packaged into envelopes (22,23). Mature VLPs exit calyx cell nuclei, migrate through the cytoplasm, and bud through the plasma membrane into the calyx lumen (8,22,23). Rearing conditions for V. canescens in our laboratory result in an 8 day pupal stage (P1-P8) followed by emergence of day 1 adult females (A1). In preparation for RNAi knockdown experiments, we designed a double-stranded (ds) RNA to egfp as a negative control treatment. We first used ds-egfp by injecting it into day 1 V. canescens pupae (P1) to assess whether it had any effects on the duration of the pupal stage and to character ize the chronology of VcVLP morphogenesis, which had not previously been assessed. No differences were observed in the duration of the pupal stage, which remained 8 days in duration. Dissection of ovaries from pupae at different times post-treatment followed by TEM indicated no VLP components were present in calyx cell nuclei from P1-P4 pupae (Fig. 1A). We referred to this period as Phase 1. Phase 2 was distinguished by the appearance of dense bodies, a few VLP envelopes, and a few mature VLPs in calyx cell nuclei. These events occurred in P6 pupae (Fig. 1B). Phase 3 was distinguished by the presence of dense bodies plus large numbers of VLP envelopes and mature VLPs in the nuclei of most calyx cells in P8 pupae and A1 adults (Fig. 1C andD; Fig. S7A). VLP envelopes in the process of accumulating electron-dense material along with empty envelopes and mature VLPs were often observed near dense bodies during Phase 3 (Fig. S7B). However, large numbers of empty envelopes and mature VLPs were also present elsewhere in the nucleus (Fig. S7C). Mature VLPs exited nuclei by budding through the nuclear membrane, which results in acquisition of a second membrane (Fig. S7D). This second membrane is then lost as VLPs migrate through the cytoplasm to microvilli where they egress into the calyx lumen (Fig. S7E). We thus concluded injection of ds-egfp does not alter the duration of the pupal stage or the morphology of dense bodies, VLP envelopes, and mature VLPs when compared to previously published descriptions of calyx cells from untreated V. canescens females.
## RNAi knockdown of OrNVorf61-like causes alterations in VLP morphogenesis, but knockdown of OrNVorf76-like does not
Detection of OrNVorf61-like, OrNVorf47-like-3, and two OrNVorf41-like family members (-1 and -2) in VcVLPs by proteomic analysis (8), together with most of these pro teins also containing transmembrane domains, suggested each could be a structural component. OrNVorf76-like and the other four OrNVorf41-like family members were not detected in VcVLPs (8), but the presence of transmembrane domains in each suggested they could also have structural roles that are required for normal VLP formation. We tested this by designing dsRNAs and focusing first on the two single-copy genes (OrNVorf61-like and OrNVorf76-like). Each was injected into P1 females followed by collecting the paired ovaries from A1 adults. One ovary per individual was used in qRT-PCR assays, which showed that each dsRNA significantly reduced transcript abundance of its target gene (80%-95%) when compared to control P1 females that were injected with ds-egfp (Fig. 2A andB). We then used TEM to examine calyx cell nuclei and the calyx lumen. Knockdown of OrNVorf61-like resulted in calyx cell nuclei that contained VLP envelopes like calyx cells from control females treated with ds-egfp (Fig. 2C). However, mature VLPs in calyx cell nuclei were larger, less spherical in shape, and less abundant than in control females, while numerous vesicles that sometimes enveloped mature VLPs were also present (Fig. 2C). In contrast, no vesicles in calyx cell nuclei were observed to traverse the nuclear membrane into the cytoplasm. Examination of the calyx cell lumen showed that some enlarged, mature VLPs were present, but the abundance was consistently lower than the number of mature VLPs that were present in the calyx lumen of control females (Fig. 2D). Numerous vesicles and other cellular debris were also present. However, none of these vesicles surrounded mature VLPs. Unlike the preceding outcome, knockdown of OrNVorf76-like resulted in no visible alterations in calyx cells or the calyx lumen when compared to control females injected with ds-egfp (Fig. 2E andF). Like control females, the high density of VLPs in the calyx lumen after OrNVorf76-like knockdown females was also associated with little or no visible cellular debris (Fig. 2F).
## RNAi knockdown of OrNVorf41-like-1 and -2 also alters VLP morphogenesis
We next designed a cocktail of dsRNAs that were co-injected into P1 pupae to knock down all OrNVorf41-like family members. We assessed the efficacy of this cocktail by measuring the transcript abundance of one family member (OrNVorf41-like-1), which was reduced more than 85% in A1 females when compared to the negative control (Fig. 3A). Examination of calyx cell nuclei from A1 females showed that some empty VLP envelopes and mature VLPs with a normal morphology were present (Fig. 3B). However, many of the VLP envelopes in calyx cell nuclei were also larger and more electron-dense than the envelopes in control females (Fig. 3B). The calyx lumen from females treated with the ds-OrNVorf41-like cocktail contained few mature VLPs, while an abundance of vesicles and other cellular debris was visible (Fig. 3C). We next injected dsRNAs specific for each OrNVorf41-like family member into P1 pupae, followed by assessment of A1 adults. Knockdown of OrNVorf41-like-1 or -2 resulted in the same alterations observed in females treated with ds-OrNVorf41-like-cocktail, with the exception that a few more mature VLPs were consistently observed in the calyx lumen versus females injected with the ds-OrNVorf41-like-cocktail (Fig. 3D through I). In contrast, knockdown of OrNVorf41-like-3, -4, -5, or -6 resulted in no visible alterations in VLP assembly in calyx cell nuclei or the morphology of mature VLPs in nuclei or the calyx lumen (Fig. S8).
The dsRNAs designed to knock down OrNVorf41-like-1 or -2 were made to regions of these paralogs that share no homology. However, because phenotypes after treatment were very similar, we checked for off-target effects by comparing transcript abundances in A1 females treated with ds-OrNVorf41-like-1 or -2 to females treated with ds-egfp.
Results indicated no off-target effects occurred since ds-OrNVorf41-like-1 significantly reduced the transcript abundance of OrNVorf41-like-1, but not OrNVorf41-like-2, while ds-OrNVorf41-like-2 significantly reduced OrNVorf41-like-2, but not OrNVorf41-like-1 (Fig. S9). We thus concluded OrNVorf41-like-1 and -2 are both required for normal VLP formation, whereas OrNVorf41-like-3 to -6 are not.
## RNAi knockdown of OrNVorf47-like family members has no effect on VLP formation, whereas knockdown of OrNVorf18-like inhibits VLP formation
The last genes we examined were the three-member OrNVorf47-like family, where one paralog has a transmembrane domain (OrNVorf47-like-3) but two do not (OrNVorf47like-1 and -2), and OrNVorf18-like, which also lacks a transmembrane domain. As with the OrNVorf41-like family, we first injected a ds-OrNVorf47-like cocktail into P1 pupae followed by assessment of knockdown of one family member (OrNVorf41-like-1) in the ovaries of A1 adults. Transcript abundance of this family member was reduced more than 85% when compared to females treated with ds-egfp, but TEM data showed no alterations in VLP formation in calyx cell nuclei or the accumulation of mature VLPs in the calyx lumen (Fig. 4A through C). In contrast, knockdown of ds-OrNVorf18-like resulted in most calyx cell nuclei containing no VLP envelopes or mature VLPs, while a single dense body that was much larger than the dense bodies in calyx cell nuclei from untreated females or females treated with ds-egfp was present (Fig. 4E). In turn, the calyx lumen contained an abundance of vesicles and other debris but almost no VLPs (Fig. 4F). Examination of calyx cells showed extensive budding from the plasma membrane, which formed vesicles of varying size that were either empty or contained cytoplasm (Fig. S10).
The vesicles and other cellular debris in the calyx lumen were also similar to the debris
## Knockdown of OrNVorf47-like gene family and several other nudivirus core genes disables VLP attachment to V. canescens eggs
Follicle cells surrounding oocytes secrete a chorion upon reaching the proximal ovarioles, followed by deposition of a material that forms projections on the surface of the chorion when viewed by TEM (22,35). These projections were also previously noted to extend from the posterior end of eggs to form a fan-like structure when viewed by light microscopy (22). Follicle cells then degenerate, which is followed by eggs passing through the calyx lumen and storage in the lateral oviducts (22,24). TEM detected mature VLPs in proximity to eggs in the calyx lumen and lateral oviducts but could not determine if VLPs bind to the surface of eggs (22). To address this question, we first used low-magnification light microscopy to image the reproductive tract from A1 females injected with ds-egfp. Ovaries contained an abundance of eggs in different stages of development, with mature eggs individually passing through the calyx before storage in the lateral oviducts (Fig. 5A). When viewed by light microscopy, mature VLPs appear as blue-colored "calyx fluid, " which was observed to exit the calyx lumen and surround stored eggs in the lateral oviducts (Fig. 5A). The fan-shaped structure extending from the posterior of eggs was also visible by light microscopy, whereas no other material on the surface of eggs could be seen (Fig. 5B). In contrast, scanning electron microscopy (SEM) clearly showed that material forming 3-5-µm-long filaments covered the surface of eggs, while the fan-like structure visible by light microscopy was formed by the same material forming 20-40-µm-long filaments (Fig. 5C through E). The distinct differences in the length led us to name the former egg filaments and the latter posterior filaments. SEM also clearly showed that large numbers of VLPs were bound to the distal tips of egg filaments and along the length of posterior filaments (Fig. 5D andE).
We next examined eggs from females injected at P1 with the ds-OrNVorf47-like-cock tail, which caused no alterations in the formation or accumulation of mature VLPs in the calyx lumen. Stored eggs from the lateral oviducts looked normal when examined by light microscopy, with the exception that no fan-shaped structure was visible (Fig. 5F). SEM showed that morphologically normal egg and posterior filaments were present, but almost no VLPs were attached (Fig. 5F through I). This finding strongly suggested an essential role for OrNVorf47-like gene products in binding of VLPs to filaments, while also suggesting bound VLPs are required to see the fan-like structure that is visible by phase-contrast microscopy. Alignments identified no definitive options for designing dsRNAs that did not result in 20 bp domains that overlapped with domains present in other family members. We thus expected each dsRNA we designed would cross-silence other family members, which occurred when we conducted an off-target assay (Fig. S11). Our results, therefore, identify the OrNVorf47-like family as essential for VLP binding to egg and posterior filaments but do not distinguish whether all or only certain paralogs are required.
We also assessed whether bound VLPs are required to see the posterior fan-like structure on eggs when viewed by light microscopy by using females injected at P1 with ds-OrNVorf61-like, ds-OrNVorf41-like-cocktail, or ds-OrNVorf18-like. Since the knockdown of each of these genes caused defects that reduced or eliminated VLPs from the calyx lumen, we expected each to also result in few or no VLPs bound to filaments on eggs. This was fully supported by light microscopy and SEM analysis of eggs (Fig. S12A through F). In contrast, eggs from females treated with ds-OrfNVorf76-like, which caused no alterations in VLP morphogenesis or accumulation in the calyx lumen, were indistin guishable from control eggs (Fig. S12G andH). Treating females with ds-RNAs specific for each OrNVorf41-like family member followed by inspection of eggs from A1 adults similarly showed that knockdown of OrNVorf41-like-1 or -2 resulted in loss of visibility of the fan-shaped structure, whereas knockdown of OrNVorf41-like-3, -4, -5, or -6 did not (Fig. S13A through F). This outcome was consistent with OrNVorf41-like-1 or -2 also being the only family members that adversely affected VLP formation and accumulation in the calyx lumen. Thus, the OrNVorf47-like family plus all of the other nudivirus core genes that adversely affected VLP formation resulted in few or no VLPs binding to eggs. We further concluded posterior filaments were only visible by phase-contrast microscopy if large numbers of VLPs were attached.
## Host hemocytes encapsulate most eggs lacking bound VLPs
A key host defense response against wasps is encapsulation: a process whereby immune cells (hemocytes) surround and kill eggs or larvae (36,37). Previous experiments showed that E. kuehniella hemocytes do not encapsulate eggs that have passed through the calyx lumen and are stored in the lateral oviducts but do encapsulate eggs collected from ovarioles before passage through the calyx lumen (23,24,38). This outcome formed the basis for concluding VLPs protect eggs from encapsulation. In contrast, results from knocking down the OrNVorf47-like family indicate VLPs cannot bind to eggs but otherwise show no alterations in morphology or abundance. We thus asked if eggs are only protected from encapsulation if VLPs are bound to filaments by injecting females with ds-egfp or ds-OrNVorf47-like cocktail as P1 pupae followed by allowing A1 adults to parasitize fifth-instar E. kuehniella larvae. Dissection of hosts 48 h later showed that no hosts parasitized by control females contained encapsulated eggs, whereas 70% of hosts parasitized by treatment females contained partially or fully encapsulated eggs (Fig. 6A). Phase-contrast microscopy showed that no hemocytes were present on eggs laid by control females while posterior filaments were also still visible, which strongly suggested VLPs remained present (Fig. 6B). In contrast, eggs laid by treatment wasps, which lack visible posterior filaments, showed a progression of encapsulation states at 48 h. Some eggs lacked any bound hemocytes, while others showed hemocytes preferentially bound to the posterior end of the egg or that extended from the posterior to the anterior end (Fig. 6C andD). Some eggs were also fully surrounded by hemo cytes that formed a complete capsule (Fig. 6E). Given these results, we also compared encapsulation of eggs laid by females treated with ds-OrNVorf61-like, ds-OrNVorf41-likecocktail, or ds-OrNVorf18-like, which each disabled VLP production in calyx cells and binding to eggs. Again, almost no eggs laid by control females treated with ds-egfp were encapsulated, while most eggs laid by treatment females were encapsulated (Fig. 6F through H). In contrast, females treated with ds-OrNVorf76-like, which did not affect VLP production or binding to filaments, laid eggs like control females that were almost never encapsulated (Fig. 6I).
## DISCUSSION
The alphanudivirus-derived DEV in V. canescens produces VcVLPs, which associate with eggs that are resistant to encapsulation in permissive hosts like E. kuehniella (8,(22)(23)(24)38). Our goal in conducting this study was to identify genes unique to nudiviruses that are required for VLP assembly, egg association, and/or protection from host hemocytes. Nudivirus genes not shared with baculoviruses lack any sequence homology to suggest what their functions might be. We thus focused this study on the five genes in the V. canescens genome that all nudiviruses encode but are not core genes that are shared with baculoviruses. Our premise for selecting these targets was that each is likely essential to the VcVLP ancestor and has been maintained in V. canescens because they remain essential for VLP production and/or function. Proteomic analysis of nudivirus virions is currently restricted to one species (Tipula oleracea nudivirus [ToNV]), where homologs of OrNVorf41, OrNVorf47, OrNVorf61, and OrNVorf76 are detected (39). This finding suggests each could be a required structural protein. Detection of OrNVorf61-like, OrNVorf41-like-1 and -2, and OrNVorf47-like-1 and -3 in VcVLPs (8) along with the presence of transmembrane domains in most of these proteins suggests each could also retain structural functions as VLP envelope components. In contrast, diversification of OrNVorf41-like and OrNVorf47-like into multimember families in V. canescens could suggest some paralogs retain ancestral functions while others have been repurposed or degraded. Detection of an OrNVorf76 homolog in ToNV virions but not VcVLPs could also reflect structural differences between each, while the absence of a transmembrane domain in OrNVorf18-like along with not detecting this protein in VcVLPs or ToNV virions at minimum suggests it is not a structural factor.
We characterized the timing of VcVLP formation in females treated with ds-egfp because prior studies primarily focused on morphogenetic events in adult females. Our analysis identified an early stage (Phase 1) during the pupal stage when no VLP components are present in calyx cells, a mid-stage (Phase 2) when VLP components first appear, and a late stage (Phase 3) when large numbers of mature VLPs are present. The timeline for these phases is consistent with a previous transcriptome analysis showing that the RNA polymerase subunits and select other DEV genes are expressed in early-stage pupae while all other DEV genes are upregulated in mid-and late-stage pupae, including several encoding products detected in VLPs (32). Our results indicate that structures originally named dense bodies (23) first appear in calyx cell nuclei during Phase 2 when VLP envelopes also first appear. The morphology of dense bodies somewhat resembles electron-dense virogenic stroma that forms in baculovirus and nudivirus-infected cell nuclei and is known to contain components required for virion assembly and DNA replication (40)(41)(42). The close association of dense bodies with VLPs in different stages of maturation during Phase 3 suggests this structure contains factors that are packaged into VLP envelopes: a conclusion supported by immunogold labeling studies that detected some VLP components in dense bodies (8). In contrast, the first VLP envelopes that form during Phase 2 and most envelopes present in Phase 3 are not near dense bodies, which suggests dense bodies are not the source of the nudivirus gene products that are envelope components.
We organized our RNAi assays to first assess whether each gene we targeted for study is required to produce VcVLPs with a normal morphology. Our TEM results show that knockdown of OrNVorf61-like, two OrNVorf41-like family members, and OrNVorf18-like results in different defects, while knockdown of OrNVorf76-like and the OrNVorf47-like family does not. The larger but fewer VLPs that form in OrNVorf61-like knockdown wasps are consistently associated with formation of vesicles in the nuclei of calyx cells that are never observed in control females. The presence of mature VLPs in these vesicles shares similarities with vesicles that form during morphogenesis of Oryctes rhinoceros nudivirus (OrNV) virions (41). However, the vesicles that surround OrNV virions mediate nucleocytoplasmic transport and egress, whereas the vesicles that form after OrNVorf61like knockdown were never observed to exit calyx cell nuclei.
Our assays using a ds-OrNVorf41-like cocktail suggest one or more members of this gene family cause defects in VLP envelope size and thickness in calyx cell nuclei plus an increase in the abundance of empty envelopes but a decrease in the abundance of mature VLPs in the calyx lumen. Treatment with ds-OrNVorf41-like-1 or -2 alone generates similar defects, which indicates a requirement for both to produce VLPs with a normal morphology rather than full functional redundancy and raises the possibil ity that these homologous proteins cooperate to properly function. Consistent with causing envelope defects, OrNVorf41-like genes in nudiviruses and V. canescens share some sequence features with the 11k gene family present in several large DNA viruses including baculoviruses where family members named Ac145 and Ac150 localize to occlusion-derived virus envelopes (5,26,43). The significance of two or more 11k-like family members in viral genomes or DEVs is unclear, with information from baculoviruses indicating that functional redundancy or compensation of individual gene products can differ depending upon the identity of the infected host (44). Our evidence that OrNVorf41-like-3, -4, -5, or -6 are similarly transcribed as late genes in V. canescens calyx cells like OrNVorf41-like-1 and -2 (27,32) but cause no defects when individually knocked down suggests repurposing for still unknown functions or neutral evolution. Knockdown of OrNVorf18-like resulted in the strongest defects in VLP morphogenesis with no formation of envelopes or mature VLPs. These alterations strongly suggest an essential role for the protein encoded by OrNVorf18-like in envelope formation, although no detection of this gene product in VLPs (8) also suggests it is not a structural protein. We further speculate dense body enlargement associated with OrNVorf18-like knockdown reflects additional accumulation of electron-dense products that are normally packaged into envelopes to produce mature VLPs. As earlier noted, knockdown of OrNVorf18-like was also associated with an abundance of cellular debris in the lumen that our results indicate is due in part to budding of calyx cell plasma membranes. Cellular debris from dead follicle cells was earlier noted to accumulate in the calyx lumen of V. canescens (22,23). Cellular debris was also readily apparent in the calyx lumen of females after knockdown of OrNVorf61-like and OrNVorf41-like-1 and -2 where VLP formation was greatly reduced but not fully eliminated. In contrast, no accumulation of cellular debris was observed when we knocked down OrNVorf76-like or OrNVorf41-like-3, -4, -5, and -6, which caused no visible defects in VLP formation or morphology. These observations collectively suggest blebbing of calyx cell plasma membranes in A1 adults is related to reduced or absent VLP formation, but our results do not identify why this occurs.
We followed our analysis of VLP assembly defects by using SEM to compare the surface of eggs from control and knockdown females for the above treatments. Results advance early TEM observations by showing that VLPs from control females directly bind to the distal tips of egg filaments and along the entire length of posterior filaments. The most significant finding is the discovery that the OrNVorf47-like family is not required for normal VLP formation, but fully disables binding of mature VLPs to eggs, which also results in most eggs being encapsulated in E. kuehniella. These outcomes are interesting because they indicate the OrNVorf47-like family encodes products that VLPs require for binding to egg filaments, while also providing experimental evidence that inhibition of VLP binding disables protection from the host immune system given unbound VLPs in these females are injected into hosts. However, it is also possible the OrNVorf47-like family has functions in addition to binding to filaments that prevents host hemocytes from encapsulating eggs. Histochemical assays indicate glucosaminoglycans (GAGs) are present on the filaments coating V. canescens eggs (22), while the literature at large indicates a wide variety of proteins bind GAGs (45). Further studies will be needed to determine if OrNVorf47-like family members encode GAG-binding proteins and the mechanism(s) underlying why VLP binding to filaments prevents host hemocytes from forming capsules. Shared homology prevented us from being able to knock down individual family members. Thus, future studies will also be needed to determine if all or only certain members of the OrNVorf47-like family are required for binding to eggs and immune protection. That knockdown of OrNVorf61-like, OrNV41orf-like-1 and -2, and OrNVorf18-like also results in an absence of VLPs on eggs and encapsulation, which is consistent with defects in VLP formation that reduce or prevent VLPs from accumulating in the calyx lumen. In turn, encapsulation of most eggs laid by females where these genes were knocked down is consistent with VLPs being required for protection of eggs from host hemocytes.
In conclusion, reported results provide new insights into the functional roles of four nudivirus core genes in VcVLP formation and function. Our results also likely provide insights into the roles of these genes in nudiviruses, where we hypothesize OrNVorf18, OrNVorf41, and OrNVorf61 are required for envelope formation and OrNVorf47 may be involved in binding to cells during infection. In contrast, our results do not provide any insights into the function of OrNVorf76-like, where knockdown causes no defects in VLP formation, binding to eggs, or protection against encapsulation. Lastly, we note that only some of the nudivirus core genes studied here have been retained by BVs and other DEVs of nudivirus origin that have been identified in parasitoids (5). Which of these genes have been retained versus lost in these other DEV lineages may reflect morphological and functional differences in the virions or VLPs each produces.
## MATERIALS AND METHODS
## Insects
Two permissive hosts for V. canescens, E. kuehniella and Plodia interpunctella were reared in the laboratory (27). Both species were maintained at 26°C, 40%-62% humidity, and a 12 h light: 12 h dark photoperiod. Larvae were fed a diet consisting of a 2:2:1 ratio (by weight) of chick starter mash, cornmeal, and glycerol. Adults were transferred to jars with screen lids. After mating, females laid eggs in the jars, which were transferred to containers with rearing diet where larvae developed and pupated. P. interpunctella was primarily used for maintenance of the V. canescens culture, while E. kuehniella larvae were used in encapsulation assays because prior studies used this species (22)(23)(24). Parasitized larvae for rearing were allowed to pupate. The pupae were then sifted, and the wasps eclosed in a square mesh cage where they were fed honey and water in 0.5% agar.
## Sequence analyses
Homologs of each gene encoded by nudiviruses were identified using Blast-P. Sig nal peptide and transmembrane domains were identified using SignalP 6.0 and DeepTMHMM-1.0 (https://services.healthtech.dtu.dk/services/DeepTMHMM-1.0/) with default settings (46,47). Amino acid sequences were aligned with MAFFT using the l-INS-i model (48). Poorly aligned positions were excluded by trimAI V1.3 (strict setting) (49). Substitution models for each gene were determined with ModelTest-NG (50). Maximum likelihood trees were constructed with RAxML-HPC2 via the CIPRES Science Gateway portal (51,52).
## RNAi assays and quantification of target gene transcript abundances
RNAi assays were performed as previously detailed (14,15). Briefly, forward and reverse primers with added T7 promoter adapters were designed (Table S1) to generate dsRNAs using cDNA prepared from A1 adult wasp ovaries with Superscript III (Invitrogen) as the template and the MegaScript RNAi Kit (Ambion). Resulting 300-400 bp dsRNA products or ds-egfp (control) were individually injected into the abdomen of day 1 V. canescens pupae (P1) at a volume of 0.5-1 µL with dsRNA concentration adjusted to 400-500 ng per µL. For dsRNA cocktails targeting multiple genes from a gene family, dsRNAs were mixed together in equal concentrations and injected as above. For the first knockdown experiment, expression of all gene family members was measured with qPCR (Fig. S14). For all subsequent experiments using dsRNA cocktails, only a single targeted gene was used to verify the knockdown efficacy. After emerging as adults (A1), the paired ovaries in treated females were explanted in phosphate-buffered saline (PBS) by dissection. One ovary was removed for total RNA isolation using the QuickRNA Mini-prep kit (Zymo) to assess knockdown of the targeted gene. Across all experiments performed in this study, the amount of total RNA isolated from each ovary was not significantly different between control and knockdown ovaries, indicating consistent RNA isolation efficiency (Fig. S15). Total RNA was synthesized into cDNA, which was used as the template in reverse transcriptase (RT) quantitative (q) PCR assays that used primers corresponding to 70-160 bp regions of each targeted gene (Table S1). Within an experiment, an internally consistent amount of total RNA was used for each cDNA reaction. Each PCR product was cloned into pSC-A-amp/kan that was Sanger-sequenced to confirm the identity and used to generate an absolute standard curve by serially diluting the plasmid (10² to 10 7 copies) using the Rotor-Gene SYBR Green PCR kit and specific qPCR primers (Table S1). Copy number of each transcript from treatment samples was determined by fitting the RT-qPCR data to the standard curve. For each gene, at least two females were examined (i.e., two biological replicates) while each RT-qPCR assay was run in quadruplicate (four technical replicates). For statistical analysis, the copy number of each gene was compared between females that were injected with ds-RNA to the target gene versus females that were injected with egfp (control) by first log-transforming the data to normalize and then analyzing using a two-tailed unpaired t-test. Statistical analyses were performed using R v4.4.2 with a P-value ≤ 0.05 considered significant.
## Transmission electron microscopy (TEM)
Two types of samples were processed for TEM: (i) ovaries from pupae (P4, P6, and P8) and newly emerged adult females (A1), and (ii) ovaries from A1 wasps injected with dsRNA as P1 pupae. All ovaries used for TEM were fixed in 3% glutaraldehyde in 100 mM Sorenson's Phosphate Buffer (pH 7.4) at 4°C for 1 week. They were then rinsed 3 times for 15 min with 100 mM Sorenson's Phosphate Buffer containing 300 mM sucrose. Samples were post-fixed in 1% osmium tetroxide in 100 mM Sorenson's Phosphate Buffer for 1 h. Samples were then washed three times for 10 min in DI water followed by dehydration using graded ethanol solutions in DI water (30%, 50%, 75%, two 95%, and two 100% ethanol) each for 15 minutes. The samples were further dehydrated with two 10 minute 99.5% acetone washes and two 15 minute propylene oxide washes. Epoxy (a mixture of Epon 812: Araldite: DDSA: DMP-30 mixed by weight at 1:1:2.044:0.0932) was gradually infiltrated into the sample using the following mixes of polypropylene : epoxy for 2 hours each: 2:1, 1:1, 1:2, and finally two 2 hour washes of epoxy. Samples were embedded in epoxy and dried overnight in an oven at 65°C. Ovaries were thin-sectioned using an ultramicrotome, mounted on copper grids, and post-stained with uranyl acetate and lead citrate. Samples were then examined using a JEM-1011 Transmission Electron Micro scope, equipped with an XR80M Wide-Angle Multi-Discipline Mid-Mount CCD Camera from AMT (Advanced Microscopy Techniques).
## Scanning electron microscopy (SEM)
V. canescens eggs were dissected from the lateral oviducts of A1 adult females in PBS that were either untreated or injected with different dsRNAs as P1 pupae. Eggs were fixed in 3% glutaraldehyde in 100 mM Sorenson's Phosphate Buffer (pH 7.4) for 2 days at 4°C followed by rinsing two times for 10 min in 100 mM Sorenson's Phosphate Buffer containing 300 mM sucrose. Eggs were then placed on coverslips precoated with poly-L-lysine (Electron Microscopy Sciences) overnight in 100 mM Sorenson's Phosphate Buffer containing 300 mM sucrose. Samples were post-fixed in 1% osmium tetroxide in 100 mM Sorenson's Phosphate Buffer for 1 h, washed three times for 10 min in DI water followed by dehydration using graded ethanol solutions in DI water (30%, 50%, 75%, two 95%, and two 100% ethanol). Samples were then washed twice for 15 min in 100% ethanol and hexamethyldisilazane (HMDS) in ratios of 2:1, 1:1, and 1:2, followed by two 15 min washes in 100% HMDS and air-drying overnight. Eggs were coated with a mixture of gold and palladium to 30 nm thickness using a Leica EM ACE600 sputter coater. Samples were then examined using a Thermo Fisher Scientific Teneo Field Emission-Scanning Electron Microscope.
## Light microscopy
Eggs collected from the lateral oviducts as described for scanning electron microscopy and ovaries were also examined unfixed by placement in PBS on slides using a Keyence VHX-E100 digital microscope with phase-contrast optics. Debris was removed from the background of the ovary image using Photoshop.
## Encapsulation assays
Females injected with dsRNAs as P1 pupae were maintained until adult emergence. Three to four females for each ds-RNA that knocked a given nudivirus gene were placed individually in a small container to which E. kuehniella fifth instars were added sequen tially. Three to four females treated with ds-egfp served as a negative control. Larvae observed to be parasitized in each container were transferred to another container with medium and held under conditions used for rearing for 48 h. Each larva was then dissected in PBS to isolate the V. canescens egg that was present. Each wasp egg was then examined by light microscopy, as described above, and scored as unencapsulated if no hemocytes were bound to its surface or encapsulated if bound hemocytes were present over all or a portion of the egg. The percentage of eggs that were encapsulated for each knockdown treatment was then compared to the percentage of eggs that were encapsulated from females that were treated with ds-egfp by a Fisher's exact test using JMP 16.0.
## References
1. Forbes, Bagley, Beer et al. (2018) "Quantifying the unquantifiable: why Hymenoptera, not Coleoptera, is the most speciose animal order" *BMC Ecol*
2. Hambäck, Janz, Braga (2024) "Parasitoid speciation and diversification" *Curr Opin Insect Sci*
3. Burke, Sharanowski (2024) "Parasitoid wasps" *Curr Biol*
4. Petersen, Bézier, Drezen et al. (2022) "The naked truth: an updated review on nudiviruses and their relationship to bracoviruses and baculoviruses" *J Invertebr Pathol*
5. Drezen, Bézier, Burke et al. (2022) "Bracoviruses, ichnoviruses, and virus-like particles from parasitoid wasps retain many features of their virus ancestors" *Curr Opin Insect Sci*
6. Bézier, Annaheim, Herbinière et al. (2009) "Polydnaviruses of braconid wasps derive from an ancestral nudivirus"
7. Burke, Simmonds, Sharanowski et al. (2018) "Rapid viral symbiogenesis via changes in parasitoid wasp genome architecture" *Mol Biol Evol*
8. Pichon, Bézier, Urbach et al. (2015) "Recurrent DNA virus domestication leading to different parasite virulence strategies"
9. Volkoff, Jouan, Urbach et al. (2010) "Analysis of virion structural components reveals vestiges of the ancestral ichnovirus genome" *PLoS Pathog*
10. Murphy, Banks, Whitfield et al. (2008) "Phylogeny of the parasitic microgastroid subfamilies (Hymenoptera: Braconidae) based on sequence data from seven genes" *Mol Phylogenet Evol*
11. Gundersen-Rindal, Dupuy, Huguet et al. (2013) "Parasitoid polydnaviruses: evolution, pathology and applications" *Biocontrol Sci Technol*
12. Strand, Burke (2014) "Polydnaviruses: nature's genetic engineers" *Annu Rev Virol*
13. Arvin, Lorenzi, Burke et al. (2021) "MdBVe46 is an envelope protein that is required for virion formation by Microplitis demolitor bracovirus" *J Gen Virol*
14. Burke, Thomas, Eum et al. (2013) "Mutualistic polydnavi ruses share essential replication gene functions with pathogenic ancestors" *PLoS Pathog*
15. Lorenzi, Arvin, Burke et al. (2023) "Functional characteriza tion of Microplitis demolitor bracovirus genes that encode nucleocapsid components" *J Virol*
16. Beck, Inman, Strand (2007) "Microplitis demolitor bracovirus genome segments vary in abundance and are individually packaged in virions" *Virology (Auckl)*
17. Bitra, Zhang, Strand (2011) "Transcriptomic profiling of Microplitis demolitor bracovirus reveals host, tissue and stage-specific patterns of activity" *J Gen Virol*
18. Poelman, Cusumano (2022) "Impact of parasitoid-associated polydnaviruses on plant-mediated herbivore interactions" *Curr Opin Insect Sci*
19. Strand (2010) "Polydnaviruses"
20. Wu, Wu, Wang et al. (2025) "A symbiotic gene stimulates aggressive behavior favoring the survival of parasitized caterpillars" *Proc Natl Acad Sci*
21. Hellers, Beck, Theopold et al. (1996) "Multiple alleles encoding a virus-like particle protein in the ichneumonid endoparasitoid Venturia canescens" *Insect Mol Biol*
22. Rotheram (1973) "The surface of the egg of a parasitic insect. I. The surface of the egg and first-instar larva of Nemeritis" *Proc R Soc Lond B*
23. Rotheram (1973) "The surface of the egg of a parasitic insect. II. The ultrastructure of the particulate coat on the egg of Nemeritis" *Proc R Soc Lond B*
24. Salt (1965) "Experimental studies in insect parasitism XIII. The haemocytic reaction of a caterpillar to eggs of its habitual parasite" *Proc R Soc Lond B*
25. Harrison, Herniou, Bézier et al. (2020) "ICTV virus taxonomy profile: Nudiviridae" *J Gen Virol*
26. Rohrmann (2019) "Baculovirus molecular biology"
27. Mao, Simmonds, Stouthamer et al. (2023) "A chromosome scale assembly of the parasitoid wasp Venturia canescens provides insight into the process of virus domestication"
28. Kool, Voeten, Goldbach et al. (1994) "Functional mapping of regions of the Autographa californica nuclear polyhedrosis viral genome required for DNA replication" *Virology (Auckl)*
29. Garavaglia, Miele, Iserte et al. (2012) "The ac53, ac78, ac101, and ac103 genes are newly discovered core genes in the family Baculoviridae" *J Virol*
30. Boogaard, Van Oers, Van Lent (2018) "An advanced view on baculovirus per os infectivity factors" *Insects*
31. Leobold, Bézier, Pichon et al. (2018) "The domestication of a large DNA virus by the wasp Venturia canescens involves targeted genome reduction through pseudogeniza tion" *Genome Biol Evol*
32. De Araujo, Leobold, Bézier et al. (2022) "Conserved viral transcription plays a key role in virus-like particle production of the parasitoid wasp Venturia canescens" *J Virol*
33. Burke, Strand (2012) "Polydnaviruses of parasitic wasps: domestication of viruses to act as gene delivery vectors" *Insects*
34. Gauthier, Boulain, Van Vugt et al. (2021) "Chromosomal scale assembly of parasitic wasp genome reveals symbiotic virus colonization" *Commun Biol*
35. Reineke, Asgari, Schmidt (2006) "Evolutionary origin of Venturia canescens virus-like particles" *Arch Insect Biochem Phys*
36. Strand (2008) "The insect cellular immune response" *Insect Sci*
37. Strand, Pech (1995) "Immunological basis for compatibility in parasitoid-host relationships" *Annu Rev Entomol*
38. Salt (1973) "Experimental studies in insect parasitism. XVI. The mechanism of the resistance of Nemeritis to defence reactions" *Proc R Soc Lond B*
39. Bézier, Harichaux, Musset et al. (2017) "Qualitative proteomic analysis of Tipula oleracea nudivirus occlusion bodies" *J Gen Virol*
40. Nagamine, Kawasaki, Matsumoto (2006) "Induction of a subnuclear structure by the simultaneous expression of baculovirus proteins, IE1, LEF3, and P143 in the presence of hr" *Virology (Auckl)*
41. (2025) *Full-Length Text Journal of Virology*
42. Velamoor, Mitchell, Humbel et al. (2020) "Visualizing nudivirus assembly and egress" *mBio*
43. Williams, Faulkner (1997) "Cytological changes and viral morpho genesis during baculovirus infection"
44. Beperet, Simón, Williams et al. (2015) "The "11K" gene family members sf68, sf95 and sf138 modulate transmissibil ity and insecticidal properties of Spodoptera frugiperda multiple nucleopolyhedrovirus" *J Invertebr Pathol*
46. Lapointe, Popham, Straschil et al. (2004) "Characterization of two Autographa californica nucleopolyhe drovirus proteins, Ac145 and Ac150, which affect oral infectivity in a host-dependent manner" *J Virol*
47. Xu, Shi, Qin et al. (2022) "A mutated glycosaminogly can-binding domain functions as a novel probe to selectively target heparin-like epitopes on tumor cells" *J Biol Chem*
48. Hallgren, Tsirigos, Pedersen et al. (2022) "DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks" *bioRxiv*
49. Nielsen, Engelbrecht, Brunak et al. (1997) "Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites" *Protein Eng*
50. Katoh, Misawa, Kuma et al. (2002) "MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform" *Nucleic Acids Res*
51. Capella-Gutiérrez, Silla-Martínez, Gabaldón (2009) "trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses" *Bioinformatics*
52. Darriba, Posada, Kozlov et al. (2020) "ModelTest-NG: a new and scalable tool for the selection of DNA and protein evolutionary models" *Mol Biol Evol*
54. Stamatakis (2014) "RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies" *Bioinformatics*
55. Miller, Pfeiffer, Schwartz (2010) "USA Full-Length Text Journal of Virology December" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12691679&blobtype=pdf | # Cepharanthine inhibits enterovirus entry by endolysosomal deacidification and exhibits protective activity in vivo
Ta-Chou Weng, Bang-Yan Hsu, Szu-Hao Kung
## Abstract
Enteroviruses, which belong to the Picornaviridae family, are linked to a range of clinical illnesses that vary from mild to severe, including life-threatening diseases. Among these, enterovirus 71 (EV71) infections in infants and young children can lead to serious neurological conditions, posing a significant public health risk due to the absence of approved treatments. In this study, we assessed the anti-EV activities of four bisbenzylisoquinoline alkaloids (BBAs): tetrandrine (TET), cepharanthine (CEP), fangchinoline (FAN), and berbamine (BER), as well as their mechanisms of action. In all cases, we observed a dose-dependent decrease in EV71 protein levels and viral titers. TET and CEP exhibited lower 50% inhibitory concentrations and higher selectivity indexes among the tested BBAs. Therefore, we prioritized TET and CEP for mechanistic investiga tion and in vivo evaluation. Mechanistic studies revealed that TET and CEP inhibited EV71 infection primarily at the entry stage, without impacting viral binding, internalization, or post-entry processes. Further studies demonstrated that TET and CEP disrupted viral trafficking along the endolysosomal pathway. Both compounds were found to effectively neutralize low pH levels in endolysosomes, which corresponded to the reduced antiviral effects caused by the acidic replenishment of the medium. The antiviral effects of TET and CEP were observed against various serotypes of EV. Remarkably, administering CEP at a dose of 10 mg/kg provided complete protection to mice infected with EV71 from lethal challenges, significantly reducing viral titers, viral RNA levels, and pathologi cal scores. Collectively, these findings highlight CEP as a promising candidate for the treatment of EV infections. KEYWORDS enterovirus, antiviral drug, bisbenzylisoquinoline alkaloids, cepharanthine, viral entry, lysosome E nteroviruses are part of the Picornaviridae family and are characterized as non- enveloped, positive-sense RNA viruses. The Enterovirus (EV) genus includes several significant human pathogens, such as polioviruses, coxsackieviruses types A and B, echoviruses, numbered enteroviruses, and rhinoviruses. While most EV infections result in mild or asymptomatic illnesses, they can also lead to serious and potentially lifethreatening conditions, including myocarditis, pancreatitis, meningitis, encephalitis, and acute paralysis (1, 2). Among these viruses, EV71 (or EV-A71) is a significant cause of hand, foot, and mouth disease (HFMD), particularly affecting young children and infants. A notable proportion of those infected with EV71 may experience severe neurological complications, and in some cases, this can lead to death, especially in the Asia-Pacific region (3). Currently, vaccines are available for poliovirus and EV71 infections (4, 5); however, developing vaccines for all enteroviruses is challenging due to the wide variety of serotypes. Additionally, there are no approved antiviral treatments for EV infections at this time, highlighting the urgent need for broad-acting antiviral medications to address the diverse and potentially severe effects of EVs.
The life cycle of all EVs begins with their specific binding to one or more cell surface receptors, initiating the process of receptor-mediated endocytosis (6)(7)(8). After a virus attaches to its receptor, changes in pH within the endosome facilitate virus trafficking, resulting in the release of the viral genome into the cytoplasm (6,9,10). Upon enter ing the cytosol, the positive-stranded viral genome acts as an mRNA and is translated into a polyprotein. This translation utilizes an internal ribosome entry site (IRES) within the 5′ untranslated region of the viral genome, allowing for efficient, cap-independ ent translation. The resulting polyprotein is then effectively cleaved by virus-encoded proteases, specifically 2A and 3C, generating vital structural proteins known as VP1-4, along with essential non-structural proteins, including proteases and polymerases. The replication of the viral genome is propelled by the viral RNA-dependent RNA polymerase (3D). The newly formed positive-sense progeny viral RNAs are meticulously packaged into capsids to create new virions, which are subsequently released from host cells through either lytic or non-lytic mechanisms (11,12).
Bisbenzylisoquinoline alkaloids (BBAs) are a class of natural products derived from plants of Berberidaceae, Monimiaceae, and Ranunculaceae families. These compounds are formed by linking two benzyl isoquinoline units through one, two, or three ether linkages.
BBAs are known for their diverse biological activities, which include anti-inflammatory effects (13), antioxidant properties (14), and anti-tumor responses (15), as well as anti-infective actions against bacteria (16), parasites (17), and viruses. Research has shown that BBAs, including tetrandrine (TET), cepharanthine (CEP), fangchinoline (FAN), and berbamine (BER), specifically target different stages of enveloped viral infections. For instance, TET disrupts the transport of viruses like Ebola, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the African swine fever virus (ASFV) from early endosomes to late endosomes (18)(19)(20). Similarly, CEP inhibits the entry phase of hantavirus and SARS-CoV-2 (21)(22)(23) , while also affecting the post-entry phases of herpes simplex virus 1 (HSV-1) and hepatitis B virus (HBV) (24,25). Additionally, CEP can simultaneously block both entry and post-entry replication of the HIV through distinct mechanisms (26,27). On the other hand, FAN inhibits the replication of the porcine epidemic diarrhea virus by obstructing autophagic flux (28). Meanwhile, BER has been shown to prevent Japanese encephalitis virus (JEV) and SARS-CoV-2 infections by interfering with the endolysosomal trafficking of viral receptors, such as the low-density lipoprotein receptor and angiotensin-converting enzyme 2, respectively (29).
The antiviral activities of BBAs are primarily effective against enveloped viruses; nevertheless, a recent report indicated that FAN inhibits several serotypes of EV in cell cultures (30). In this study, we aimed to evaluate the antiviral potency of TET, CEP, FAN, and BER, as well as to explore the underlying mechanisms of action. Our findings revealed that CEP exhibited the most potent antiviral activity among the tested BBAs, demonstrated by the lowest 50% inhibition concentrations (IC 50 ) value and the highest selectivity index (SI). We focused on CEP and showed that it likely inhibits EV replica tion by neutralizing the low-pH endolysosomal pathway and interfering with the virus endolysosomal trafficking, resulting in blocking the viral entry. Notably, CEP displayed broad-spectrum anti-EV effects. Moreover, we demonstrated that CEP administration significantly reduced the lethal effects of EV71 infection and lowered the viral levels in a mouse model, suggesting its potential development as a novel antiviral treatment for EV infections.
## MATERIALS AND METHODS
## Cells and viruses
HeLa cells (ATCC, CCL-2) and Rhabdomyosarcoma (RD) (ATCC, CCL-13) cells were cultured at 37°C in minimum essential medium (MEM) (Gibco-BRL, Inc.) supplemented with 10% fetal bovine serum (31). EV stocks used in this research included EV71 (BrCr strain), coxsackievirus A16 (CVA16), CVB1, CVB3, echovirus serotype 9 (Echo9), Echo30, and EV68 (Fermon strain), as documented (32). A mouse-adapted EV71 MP4 strain used for the mouse model was kindly provided by Dr. Jen-Ren Wang from National Cheng Kung University, Taiwan.
## Chemicals
Tetrandrine (518-34-3), cepharanthine (481-49-2), fangchinoline (436-77-1), berbamine (6078-17-7), and emetine (7083-71-8) were purchased from Cayman Chemical. Chlor oquine (C6628-25G) was purchased from Sigma. All compounds were dissolved in dimethyl sulfoxide (DMSO), and the final DMSO concentration in the culture medium did not exceed 0.05%, a concentration tolerated by all tested cell lines. All drug-free controls contained 0.05% DMSO.
## Virus infection and titration
Virus infections were conducted in MEM supplemented with 2% FBS at 37°C unless otherwise indicated. The supernatant containing extracellular viruses was collected from the EV71-infected cell cultures and centrifugated at 5,700 × g for 5 min to remove cell debris. Cell-associated viruses were prepared from cell lysates collected after freeze-andthaw cycles and centrifugation at 15,300 × g for 10 min. The total virus is the combined mixture of supernatant and cell-associated virus prepared from above. Infectious viral titer was measured using the 50% tissue culture infectious dose (TCID 50 ) method on RD cells by the method of Reed-Muench (33).
## Western blot
A western blot analysis was conducted as previously described (31). The primary antibodies (Ab) were a Mouse EV71 monoclonal VP1 Ab (1:3,000, GTX633390, GeneTex) and a rabbit anti-α tubulin primary Ab (1:10,000, GTX112141, GeneTex), the latter serving as the loading control. Secondary Abs included an HRP-conjugated goat anti-mouse polyclonal Ab (1:1,000, sc-2030, Santa Cruz) and a goat anti-rabbit HRP-conjugated secondary Ab (1:3,000, sc-2004, Santa Cruz). Proteins were detected using an enhanced chemiluminescence western blot kit (GTX14698, GeneTex). The band intensities were quantified using ImageJ software.
## Immunofluorescence assay
HeLa cells were seeded in 24-well plates and pre-treated with various concentration of test compounds for 1 h. After the culture media was removed, the cells were subsequently infected with EV71 at a multiplicity of infection (MOI) of 0.1 for 1 h. Following adsorption, the inoculum was discarded, and the cells were washed twice with phosphate-buffered saline (PBS). Medium containing the same concentration of the compound was then added. At 12 h post-infection (p.i.), cells were fixed with 4% paraformaldehyde and permeabilized using 0.2% Triton X-100. Mouse anti-EV71 monoclonal Ab (1:2,000, MAB979, Millipore) and FITC-conjugated goat anti-mouse Ab (1:200, 115-095-062, Jackson ImmunoResearch) were used as a primary Ab and secondary Ab, respectively. The nuclei were stained with DAPI-Aqueous (ab104139, Abcam). The cells were viewed under a fluorescent microscope (Leica DM6000B) equipped with both fluorescein isothiocyanate (FITC) and UV filters. EV antigen-posi tive cells and 4′,6-diamidino-2-phenylindole (DAPI) positive cells from each field were counted and analyzed using the associated MetaMorph software.
## Cell viability assay
Compound concentrations at 1, 3, 10, 30, and 100 µM were tested. Cell viability was assessed after 18 h of treatment using the CellTiter 96 AQueous Cell Proliferation Assay (Promega), as described (32). The 50% cytotoxic concentration (CC 50 ) was determined using GraphPad Prism 9 (GraphPad Software).
## RNA extraction and reverse transcriptase quantitative PCR
RNA preparation and RT-qPCR followed the protocol previously detailed (31). Total cellular and viral RNA was extracted using a TRIzol reagent (ThermoFisher). The reverse transcription (RT) reaction and real-time PCR were performed using an AMV Reverse Transcriptase XL (Takara) and the FastStart Universal SYBR Green Master kit (Roche Applied Science), respectively, according to the manufacturer's instructions. PCR primer pairs for the VP1 region of the EV71 (BrCr strain) genome and human β-actin were reported.
## Binding and internalization assay
HeLa cells were pre-treated with 5-and -10 µM of TET and CEP at 37°C for 2 h. After this incubation, the cells were placed at 4°C and incubated for 10 min in 1 mL of binding buffer composed of PBS, 1% bovine serum albumin (BSA), and 0.1% sodium azide. Following the binding buffer incubation, the cells were infected with EV71 at an MOI of 50 and kept at 4°C for 1 h to facilitate viral adsorption. Unbound virus was then washed off using PBS at 4°C, and the cells were treated with 0.25% trypsin (Gibco) to detach any virus bound to the cell surface. For the internalization assay, the unbound virus was removed after the binding period and replaced with pre-warmed medium containing the respective concentrations of the compounds. The cells were then incubated at 37°C for 1 h. To remove any surface-bound viruses, the cells were trypsinized at 37°C for 3 min. Finally, viral RNA was quantified using RT-qPCR to assess the viral level.
## EV71 replicon assay
The pSVA-EV71-GFP plasmid (6), a sub-genomic replicon of EV71 that encodes green fluorescent protein (GFP) reporter (a kind gift from Dr. Lih-Hwa Hwang, National Yang-Ming University), was linearized using the Not I restriction enzyme and subse quently utilized for in vitro transcription with the RiboMAX Large Scale RNA Production System-T7 (Promega, P1320). Following transcription, the RNA was precipitated by LiCl (Cat. #7447-41-8, Ambion AM9480) and washed with ice-cold 70% ethanol. HeLa cells were transfected with the pSVA-EV71-GFP RNA using Lipofectamine 2000 (ThermoFisher) according to the manufacturer's instructions. After 30 min, the transfection reagent was removed and replaced with a medium containing the TET or CEP at 5 and 10 µM. The cells were incubated for 6 h, after which they were fixed with 4% paraformalde hyde and permeabilized with 0.2% Triton X-100. Nuclei were stained using an aqueous DAPI-containing mounting medium. The cells were then observed under a fluorescent microscope (Leica DM6000B) equipped with both FITC and UV filters. GFP-positive cells and DAPI-stained nuclei were counted and analyzed using MetaMorph software.
## EV71 movement analysis
HeLa cells were seeded in 24-well plates and incubated with EV71 at an MOI of 300 at 4°C for 1 h. After incubation, the cells were washed three times with PBS and then treated with either compound-containing or compound-free media for intervals of 15, 30, 45, 60, and 90 min. Following this, the cells were fixed with 4% paraformaldehyde and permeabilized using 0.2% Triton X-100. To detect the viral antigen, we used rabbit anti-EV71 VP2 polyclonal Ab (1:2,000, GTX132340, GeneTex) as the primary Ab and Goat Anti-Rabbit IgG H&L (Alexa Fluor 488) (1:500, ab150077, Abcam) as the secondary Ab. Additionally, cells were co-stained with either EEA1 Ab Alexa Fluor 546 (G-4, 1:500, sc-137130, Santa Cruz) or LAMP1 Ab Alexa Fluor 546 (H4A3, 1:500, sc-20011, Santa Cruz). The cells were also stained with DAPI to indicate the location of the nucleus. The stained samples were then examined using a confocal laser scanning microscope (Carl Zeiss LSM 880 META). Green fluorescence represented the viral antigen, red fluorescence corresponded to the early endosome antigen (EEA1) or lysosomal-associated membrane protein 1 (LAMP1), and blue fluorescence marked the nucleus. The percentage of colocalization was calculated using ZEN 2 software.
## Acidic organelle staining
Cells were cultured in 24-well plates and incubated with TET and CEP at the indicated concentrations at 37°C for 2 h. Lysotracker DND-99 (ThermoFisher, L7528) was used to stain acidic organelles within the cells. After removing the compound-containing medium, the cells were replaced with fresh medium containing the same compound and 500 nM Lysotracker DND-99, followed by an additional incubation at 37°C for 1 h. The cells were then washed three times with PBS, fixed with 4% paraformaldehyde, and permeabilized with 0.2% Triton X-100. Nuclei were stained using an aqueous DAPI-con taining mounting medium. Fluorescence images were acquired on a Leica DM6000B microscope equipped with a red filter set (excitation 545 ± 30 nm; emission 610 ± 75 nm) specific for LysoTracker DND-99. Fluorescence intensity was quantified using ImageJ software.
## Low-pH exposure assay
We followed a protocol reported in a previous study with some modifications (31). HeLa cells were pre-treated for 1 h at 37°C with 10 µM TET or 8 µM CEP for 1 h at 37°C. After pretreatment, the cells were infected with EV71 at an MOI of 0.5 in the presence of 10 µM of either TET or CEP at 4°C. The inoculum was then removed, and unbound virus was washed away with ice-cold PBS. The cells were incubated for 1 h at 37°C in the presence of the respective compound. Afterward, the supernatant was discarded, and media with different pH levels (7.4, 6.5, 5.5, and 5.0), along with the corresponding compound, were added for 10 min. Finally, the media were removed, and the cells were incubated in a medium (pH 7.4) containing the respective compound at 37°C for 6 h.
## Mouse infection and sample analysis
The mouse study followed a protocol with some modifications (34). Seven-day-old ICR mice were infected with the MP4 virus via intraperitoneal injection at a dose of 1 × 10 7 TCID 50 .
TET was administered at a dose of 10 mg/kg, while CEP was given at either 5 mg/kg or 10 mg/kg, with all treatments dissolved in DMSO. DMSO alone was also used as a control. The treatment was administered intraperitoneally 12 h prior to infection and continued every 12 h for four additional doses following the viral inoculation. The mice were monitored for body weight, survival rate, and clinical symptoms over a period of 7 days. Clinical severity was scored as follows: 0-healthy; 1-ruffled fur and hunchback appearance; 2-wasting; 3-limb weakness; 4-limb paralysis; and 5-moribund or death. Five mice in each group were euthanized at 7 days p.i. Their brain, spinal cord, and hind-limb muscle tissues were collected for viral titration and assessment of viral RNA levels using the TCID 50 assay and RT-qPCR, respectively. Additionally, hematoxylin-andeosin (H&E) staining was performed to evaluate histopathology. The lesion areas were quantified using ImageJ software, calculating the percentage of brain vacuole regions or myositis/myonecrosis regions in relation to the total tissue area within each field of view, as referenced in 35, 36.
## RESULTS
## BBAs effectively inhibit EV71 infections with CEP showing superior potency
The chemical structures of BBAs, including TET, CEP, FAN, and BER, consist of two benzyl isoquinoline units linked together by oxygen bridges (Fig. 1A). We evaluated the anti-EV71 potency of these BBAs using Western blot analysis at specified concentrations in HeLa cells. In all cases, we observed a dose-dependent inhibition of viral protein (Fig. 1B). Moreover, we measured viral titers and protein levels in response to BBA treatments using the TCID 50 assay (Fig. 1C) and an IFA (Fig. 1D; Fig. S1), respectively, and the IC 50 values were determined using both methods (Table 1). Additionally, a cell viability assay was performed to determine the CC 50 for each compound, along Cell lysates were prepared and subjected to Western blot analysis using an anti-EV71 VP1 Ab and an anti-tubulin Ab as an internal control. (C) HeLa cells were pretreated with TET, CEP, FAN, or BER at the indicated doses for 1 h, followed by infection with EV71 at an MOI of 0.5 for 8 h. Cell lysates and supernatants were collected, and total viral titers were determined by the TCID₅₀ assay. Data represent the mean ± standard deviation (SD) of three independent experiments performed in duplicate (n = 2). Statistical analysis was performed using one-way ANOVA; ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. (D) HeLa cells were pretreated for 1 h at the indicated concentrations of the test compounds and then infected with EV71 stocks at an MOI of 0.1, with test compounds maintained throughout the infection. At 12 p.i., the cells were analyzed using an IFA. For each condition, the percentage of infection was calculated as the ratio of the number of infected cells stained for viral VP1 to the number of cells stained with DAPI. Drug cytotoxicity was determined by treating HeLa cells with increasing concentrations of the indicated drugs (1, 3, 10, 30, and 100 µM) for 18 h. Cell viability was measured using a cell proliferation assay and expressed as the percentage of drug-free cells. IC 50 and CC 50 values were calculated using GraphPad Prism9 software. The solid circle and empty circle represented the inhibition rate (%) and cell viability (%), respectively. Data represented the means of triplicated experiments and the standard error of the mean (SEM). Drug-free wells contained 0.05% DMSO.
with calculating the SI, defined as SI = CC 50 /IC 50 . The results showed that the antiviral potencies of the compounds were ranked as follows: CEP > TET > FAN > BER, regardless of the methods employed. Notably, CEP exhibited low sub-micromolar IC 50 values, with SI values reaching as high as 287 (by TCID 50 assay) or 163 (by IFA).
## Time-of-addition analysis of TET and CEP
We next focused on the two most effective drugs: TET and CEP. To better understand how these compounds affect the viral replication cycle, we conducted a time-of-addition assay (Fig. 2A). TET and CEP were administered to HeLa cells at concentrations of 5 or 10 µM during different phases of the infection process: (a) 1 h before and during the 1 h virus adsorption step, with treatment maintained throughout the subsequent 8 h infection period; (b) 1 h before and during the 1 h virus adsorption step, followed by an additional 1 h of treatment before removal (entry step); or (c) 2 h after the completion of adsorption step, continuing treatment for the remaining 6 h of infection (post-entry). For each condition, we analyzed viral protein levels using Western blotting. Our results demonstrated that treatment with TET (Fig. 2B) or CEP (Fig. 2C) primarily reduced viral protein levels during the entry phase, with minimal effects observed during the post-entry phase.
## TET and CEP do not impede the attachment, internalization, or post-entry processes of EV71
To determine the specific steps of EV71 viral entry that are affected by TET and CEP, we investigated their impact on virus binding and internalization. For the virus binding assay, we incubated the cells with the virus at 4°C (to prevent virus internalization) for 1 h, either with or without the test compounds (Fig. 3A). Following this, we conducted the virus internalization assay (Fig. 3B) by increasing the temperature to 37°C for 1 h, allowing the internalization of the bound virus particles. After this step, we performed trypsinization to remove any uninternalized surface virus particles. We measured the viral RNA levels for each condition using RT-qPCR. Our results showed no detecta ble difference between the compound-treated group and the control group without compounds, indicating that neither TET nor CEP affected virus binding or internalization (Fig. 3A andB).
Next, a sub-genomic replicon of EV71 that encodes GFP reporter (pSVA-EV71-GFP) was used to evaluate the antiviral effects of TET and CEP in HeLa cells. This replicon system allows us to bypass receptor-mediated entry by delivering viral RNA directly into the cytoplasm through transfection (37). At 6 h post-transfection, we observed strong GFP expression in control cells that received no compounds. In contrast, emetine (EMT), a known inhibitor of EV IRES-mediated translation (38), significantly reduced the fluorescence signal. However, TET and CEP, at all tested concentrations, did not impact GFP expression (Fig. 3C andD). This suggests that these compounds do not interfere with viral translation or genome replication. Overall, our results indicate that TET and CEP do not disrupt the stages of viral attachment, internalization, or post-entry events of EV71.
## TET and CEP block the trafficking of EV71 in the endolysosomal pathway
Given that viral binding and internalization are independent of the antiviral effects of TET and CEP on viral entry, we next examined viral trafficking, a specific sub-step of viral entry. HeLa cells were incubated with EV71 at 4°C for 1 h, followed by treatment with either TET or CEP at concentrations of 2, 5, or 10 µM at 37°C for 90 min. The viral signal increased with higher compound concentrations, as shown by imaging analyses (Fig. 4A) and their quantification (Fig. 4B) in the IFA. Additionally, changes in viral signals were observed at various time points within a 120 min timeframe following treatment with 10 µM of TET or CEP. In the control group, viral signals peaked at 30 min, then gradually declined until reaching low levels at 60 min, where they remained thereafter. In contrast, the viral signal in the group treated with either TET or CEP increased over time (Fig. 4C; Fig. S2), suggesting that the virus may be sequestered within the endocytic organelle.
We then conducted an analysis of the motility of EV71 within the endolysosome during the viral entry process upon the treatment with CEP or left untreated (DMSO control). Using confocal microscopy, we examined the localization of viral protein in EMT was used as a control for post-entry targeting. (D) For each condition, the percentage of GFP (+) cells was calculated as the ratio of the number of GFP (+) cells to those stained with DAPI, compared to untreated controls (DMSO). Quantitative results are presented as mean ± SD (n = 4). Unpaired one-way ANOVA analysis was conducted. ****P < 0.0001; ns indicates non-significant. relation to two key proteins: EEA1, which is associated with early endosomes, and LAMP1, which is found in lysosomes. In the control group, we observed that the viral protein primarily colocalized with EEA1 at the 15 min mark, reaching nearly 15%, before gradually declining. In contrast, the CEP-treated group displayed significantly lower colocalization of the viral protein with EEA1 at the same time point, peaking at only about 8% at 45 min, and then decreasing over time. This observation suggests that CEP delays the trafficking of EV71 particles within the endosomes (Fig. 4D; Fig. S3). Regarding LAMP1, the viral protein colocalized with it from 30 to 45 min, reaching a peak of nearly 45% at 45 min before dropping to about 20% at later time points in the control group. In contrast, the CEP-treated group maintained a consistently high level of viral protein-LAMP1 colocalization, remaining around 40% from 30 to 90 min. This indicates that while many viral particles in the control group exited the lysosome by 60 min, those in the CEP-treated group remained mostly trapped inside the lysosomes (Fig. 4E; Fig. S3). Overall, these results imply that BBAs, specifically TET and CEP, may inhibit the movement of EV71 particles within the endolysosomal system.
## BBAs prevent the entry of EV71 by increasing lysosomal pH
In the context of EV infections, the acidic pH of lysosomes plays crucial roles in the virus trafficking and the structural changes of virions that lead to virus uncoating (11). Since BBAs, particularly TET, CEP, and FAN, have been shown to disrupt lysosomal acidification in certain cell types (39-41), we examined whether treatment with TET or CEP caused elevation of pH in acidic cellular compartments in HeLa cells. These compartments were labeled with the LysoTracker dye, which emits red fluorescence only in acidic environments, such as endolysosomes. Chloroquine (CQ), a well-known inhibitor of endosomal acidification, was used as a control. The results indicated that, compared to the untreated control, significant decreases in fluorescence intensity were observed following treatment with TET or CEP at a concentration of 10 µM, similar to the results seen with CQ (Fig. 5A andB). Notably, treatment with CEP at 1 µM resulted in significant fluorescence reduction, while treatment with TET at the same concentration showed no detectable fluorescence reduction, which aligns with their respective antiviral activities. As TET and CEP elevated the pH levels in acidic intracellular compartments, we hypothesized that the antiviral effects of both compounds could stem from their ability to block endosomal acidification. To test this, HeLa cells were exposed to low-pH media after virus binding. The cells were pretreated with either compound before being inoculated with EV71 stock while kept at 4°C. After 1 h p.i., the inoculum was removed, and any unbound viruses were washed away. We then allowed the viruses to enter the cells by incubating them with either compound at 37°C for 1 h. Following that, the cells were incubated for 10 min in compound-containing media of pH 7.4, 6.5, 5.5, or 5.0, followed by a 6 h recovery in pH 7.4 medium that also contained the compound. We next prepared the cell lysates for Western blot analysis to assess the levels of viral protein at 6 h p.i. (Fig. 5D andE). Our findings revealed that viral replication, indicated by the levels of viral protein, could be significantly enhanced by low-pH shocks (pH 5.5 and 5.0) in a pH-dependent manner. In stark contrast, exposure to media at pH 6.5 did not lead to a significant recovery of viral replication, supporting the idea that early endosomes (with a pH of 6.0-6.5) play a minor role in viral uncoating. This result is consistent with our earlier observation that the BBA treatment caused the alkalization of the cellular acidic compartment (Fig. 5A andB). Taken together, these results suggest that TET and CEP likely target a low pH-dependent step in the viral entry process.
## Antiviral activity of TET and CEP against other EV serotypes
We investigated the antiviral effects of TET and CEP against a range of EV serotypes, including CVA16, CVB1, CVB3, Echo9, Echo30, and EV68, in addition to EV71. Total viral titers were determined using cell lysates and culture supernatants collected at 16 h p.i. Both TET and CEP demonstrated antiviral activity in HeLa and RD cells, though with varying degrees of effectiveness (Fig. 6; Table 2).
## CEP prevents MP4 infection in mice
To evaluate the in vivo antiviral efficacy of TET and CEP against EV71 infection, we utilized a 7-day-old ICR mouse model infected with a mouse-adapted strain MP4 of EV71 (Fig. 7A). Mice were given a single intraperitoneal dose of either TET, CEP, or vehicle control 12 h before viral inoculation. This was followed by additional doses of the same compound or a vehicle control at 12, 24, 48, and 72 h p.i. In comparison to the vehicle control group, which had a survival rate of 47% (7 out of 15 mice), the group treated with CEP at a dose of 10 mg/kg achieved an impressive 100% survival rate (8 out of 8 mice), significantly improving the survival rate (Fig. 7B). In contrast, treatment with TET at 10 mg/kg resulted in a survival rate of only 62.5% (5 out of 8 mice), while treatment with CEP at 5 mg/kg led to a survival rate of 75% (6 out of 8 mice); these results did not significantly enhance the survival rate (Fig. 7B). Additionally, there were no significant differences in body weight between the treated and control groups (Fig. 7C). Limb paralysis was observed in the control group (8 mice), as well as in the groups treated with CEP at 5 mg/kg (2 mice) and TET (3 mice), 5 days after infection. In contrast, the group treated with CEP at 10 mg/kg did not show any signs of paralysis and had significantly lower clinical scores (Fig. 7D). Viral titers and RNA levels were measured in the brain, spinal cord, and limb muscles using the TCID 50 assay and RT-qPCR, respectively. The results showed that treatment with CEP at a dose of 10 mg/kg significantly reduced both viral titers and RNA levels compared to those of the control group, while the other treatment groups did not show this effect (Fig. 7E andF). We also performed H&E staining and analysis of brain and limb muscle tissues of the mice (Fig. 7G). In comparison to the control group devoid of compound treatment, the brain tissues of the mice treated with 10 mg/kg of CEP exhibited much milder signs of encephalitis, including reduced neutrophilic infiltration and vacuolation. Additionally, the limb muscle tissue from this treated group showed a significantly lower number of pathological features. Specifically, there was less necrosis of muscle fibers, fewer infiltrating inflammatory cells, and a reduction in fibrotic scars that typically replace damaged muscle fibers. In contrast, the tissues from the virus-infected mice that were treated with either 10 mg/kg of TET or 5 mg/kg of CEP demonstrated much less pronounced reversal of the pathological effects.
In line with the findings, ImageJ quantification revealed a dose-dependent decrease in the percentage of lesion area in the hindlimb muscle tissues of mice treated with CEP at doses of 5 mg/kg and 10 mg/kg, compared to the virus-infected control group that received no treatment. Notably, a reduction in brain lesions was observed only in the mice treated with CEP at the higher dose of 10 mg/kg. In contrast, mice treated with TET at a dose of 10 mg/kg did not show a significant reduction in lesions in either the brain or hindlimb muscle (Fig. 7H). In conclusion, treatment with CEP at 10 mg/kg effectively inhibits EV71 replication and its associated pathogenesis in this murine model.
## DISCUSSION
Enteroviruses of various serotypes continue to pose significant global health threats, especially among immunocompromised adults and pediatric populations. Among these, EV71 is particularly known for causing severe neurological complications in infants and young children across the Asia-Pacific region. There are currently no approved antiviral drugs available to treat EV infections. Additionally, there is an urgent need to develop broad-spectrum anti-EV drugs, as no single EV serotype is uniquely associated with specific clinical manifestations. In this study, we explored the antiviral potential of BBAs against EVs. All tested BBAs effectively inhibited EV71 infection in a dose-dependent manner. Notably, TET and CEP showed lower IC 50 values and higher SI values (Table 1). As a result, we have focused on TET and CEP to investigate their mechanism of action.
Our findings demonstrated that TET and CEP primarily exerted their antiviral effects at the entry stage of the viral lifecycle (Fig. 4A), while excluding virus binding, internaliza tion, and post-entry processes (Fig. 3). Specifically, both compounds immobilized viral particles during endolysosomal transport, thereby hindering effective virus uncoating (Fig. 4). This effect is at least partly attributed to the compounds' ability to neutralize the low pH levels in endolysosomal vesicles (Fig. 5A), as acidic replenishment of the medium restored the viral infection in cells pretreated with TET or CEP (Fig. 5B). Together, these data indicate a mechanism by which BBAs neutralize the pH levels of the endolysosomal pathway and disrupt its function for transporting viruses, which is crucial for the entry of EVs. The IC 50 values obtained from the antiviral treatment over an 8 h period (Fig. 1)
of drug administration and low-pH medium exposure: HeLa cells were pretreated with 10 µM TET (D) and 8 µM CEP (E) for 1 h. EV71 was used to inoculate the cells at an MOI of 0.5 in media containing the corresponding compounds at 4°C for 1 h. The cells were then washed and replaced with media containing the compounds, followed by incubation at 37°C for 1 h. Subsequently, cells were replaced with media at different pH values (7.4, 6.5, 5.5, and 5.0) containing the drugs at 37°C for 10 min, after which they were again replaced with compound-containing media (pH 7.4) for 6 h. Cell lysates were prepared for Western blot analysis to probe for viral VP1 and tubulin. The percentages shown below each lane represent the intensity of VP1 relative to that of tubulin, compared to DMSO-treated controls.
are significantly lower than those measured after a 16 h treatment (Fig. 6). To further investigate this issue, we conducted an additional experiment in which the compound was replenished at 8 h p.i. to maintain a total infection period of 16 h. However, this replenishment did not restore the antiviral effect (Fig. S5; Table S2), indicating that the time-dependent reduction in antiviral activity is not related to the stability of the compound. This decrease in antiviral efficacy over time has been documented, particularly with agents targeting the early stages of replication (42,43). Once again, the potential for compound instability has been ruled out (42), and the underlying mechanisms behind this phenomenon remain to be explored. CEP is known to demonstrate multiple antiviral mechanisms against various enveloped viruses. It has been shown to suppress SARS-CoV-2, dengue virus, and ASFV in the early stages of infection. In contrast, CEP interferes with the virus, HSV-1, HBV, and ASFV during the mid-stage of infection. For SARS-CoV-2, CEP blocks the binding of the viral spike protein to the angiotensin-converting enzyme 2 (ACE2) receptor (21) and prevents Ca 2+ -mediated membrane fusion (22), thereby hindering viral entry. In the case of ASFV, CEP inhibits viral internalization by impairing the functions of late endosomes and lysosomes (20). Additionally, CEP is likely to inhibit dengue virus replication by interfering with the viral internalization process although the precise mechanisms remain largely unknown. Moreover, CEP impairs mid-stage viral replication through its interactions with host proteins or pathways essential for viral reproduction. In HSV-1 infections, CEP downregulates the PI3K/Akt and p38 MAPK signaling pathways, causing infected cells to halt in the G2/M phase and undergo apoptosis (24). For HBV, CEP disrupts the function of the host heat shock cognate 70 protein, which is critical for viral replication (25). While studies on HSV-1 and HBV have shown post-entry inhibition by CEP, they did not conduct time-of-addition assays, leaving open the possibility that CEP may also affect the entry stage of these infections. Similarly, CEP impairs the Hsp90-Cdc37 complex and inhibits AKT signaling, resulting in the blocking of ASFV replication during the mid-stage (44). Collectively, CEP demonstrates a range of inhibitory effects against various enveloped viruses during both the entry and post-entry stages of infection. However, most of the reports referenced are primarily based on cell culture studies, except for the ones involving SARS-CoV-2, where in vivo infections and treatments were conducted to observe the effects in mice (45). Further investigation is required to validate the in vivo antiviral effects of CEP across different virus infection models.
Among the BBAs, it was reported that FAN has inhibitory effects against EVs of various serotypes by targeting the early stages of infection (30). This is the only report indicating that a BBA possesses antiviral activity against non-enveloped viruses, specifically EVs. The IC 50 values of FAN against different EVs align with our findings. When selecting FAN-resistant EV71, mutations were observed in structural proteins, with the E145G mutant in VP1 displaying particularly strong resistance. The variation of E145G can bind the virion to the heparan sulfate (HS) attachment receptor, which enhances the virus's infectivity (46,47). Moreover, our structure-based docking study shows that FAN binds effectively to VP1 E145, but it cannot bind to VP1 E145G. On the other hand, CEP does not bind to either VP1 E145 or VP1 E145G (Fig. S6). This indicates that the VP1 protein could be a potential target for FAN, distinguishing it from our findings that CEP action is independent of viral binding (Fig. 3A).
BBAs exhibit structural and physicochemical properties characteristic of lysomotropic cationic amphiphilic drugs (CADs) as shown in Fig. 1A (39,48,49). CADs consist of a hydrophobic aromatic ring along with hydrophilic segments that contain an ionizable amine functional group. Due to their amphiphilic nature, CADs can easily permeate cellular membranes, undergo protonation, and become trapped within acidic intracellu lar compartments, such as late endosomes and lysosomes. This accumulation can cause alkalization (50) and impair the endolysosomal pathway (51), which is essential for EV entry. Two quantitative measurements of the physicochemical properties of CADs are the calculated logP (ClogP) value, which indicates the compounds' lipophilicity, and the basic pKa, the acid dissociation constant for the conjugate acid of the weak base. These measurements influence both the extent of lysosomal trapping and the kinetics of passive permeation. Research has indicated that compounds with a ClogP greater than 3 and a basic pKa greater than 7.4 are often classified as CADs (51). However, some studies suggest that less stringent criteria-a ClogP greater than 3 and a basic pKa greater than 6-can also be applicable. Both TET, which has a ClogP of 3.23 and a basic pKa of 7.70, and CEP, with a ClogP of 6.29 and a basic pKa of 7.61, meet these criteria (Table S1) (52). This may explain a mechanism involved in the pH elevation of the endolysosomal system by TET or CEP.
Two-pore channels (TPCs) and transient receptor potential mucolipin (TRPML) are important ion channels located within the endolysosomal system. They primarily regulate calcium flux, which, in turn, influences endolysosomal pH levels and vesicular dynamics. TPCs and TRPML are essential for the entry and intracellular trafficking of various pathogens including viruses (53,54). Previous reports have identified TET as a TPC inhibitor that can hinder the entry of the Ebola virus and coronaviruses (18,55,56). Molecular docking analyses suggest that CEP may also bind TPC2, thereby impairing its functions (57), although this has not yet been empirically validated. On the other hand, BER is known to block the infection of SARS-CoV-2 and flaviviruses by disrupting the endolysosomal trafficking of viral receptors mediated by TPRMLs (57). However, BER exhibited relatively modest antiviral effects in the present study (Fig. 1B through D). We proposed that the primary inhibitory mechanism against EV71 might also involve TPCs though it's less likely to involve TRPMLs. Further mechanistic investigations are needed to determine if CEP and TET influence the acidity-dependent endocytosis of EV entry via TPCs.
We demonstrated that the prophylactic administration of CEP in a mouse model provided protection against a lethal EV71 challenge by decreasing mortality, viral load, and tissue damage (Fig. 7). To the best of our knowledge, this is the first study to identify a BBA as an effective inhibitor of EV71 in vivo. CEP is approved in Japan for of 1 × 10 7 TCID 50 per mouse. At 12, 24, 48, and 72 h p.i., the mice were administered the corresponding compound and monitored for 7 days. The survival rate (B), weight (C), and clinical scores (D) are presented with mean values and the SEM. Survival rates were assessed using the log-rank test, with significance indicated as *P < 0.05. Following this period, the mice were euthanized, and tissues were collected for H&E staining and viral titer analysis. At 7 days p.i., 5 mice in each group were euthanized and their brain, spinal cord, and hind-limb muscle tissues were harvested. Viral titers from harvested tissues were determined by TCID₅₀ assay (E) and RT-qPCR (F). A histopathological examination of the harvested tissues was performed (G). In the infected group devoid of compound treatment, brain areas with vacuoles, typically filled with monocyte/macrophage infiltration, are indicated by the dark arrows. In the limb muscle of the infected control group without the compounds, the dark arrows highlight regions of myositis and myonecrosis, while the white arrow points to cells that did not undergo complete myositis and myonecrosis. Virus-infected groups administered with TET (10 mg/kg) or CEP (5 or 10 mg/kg) are indicated, along with an age-matched group treated with DMSO that is mock-infected. (H) The quantification of lesions in the brains and muscles was performed using ImageJ software on the groups indicated. The results are displayed as the percentage of area measured from a minimum of six locations in each of five individual mice. (E, F, H) Statistical comparisons were made using a one-tailed t test with ****P < 0.0001, ***P < 0.001, *P < 0.05. the treatment of radioactive leukopenia, alopecia, venomous snake bites, and chronic exudative otitis media (58)(59)(60). Although the exact mechanisms of action of CEP are not fully clarified, repurposing an approved drug significantly accelerates the process of advancing a candidate drug into clinical practice. This is due to the fact that the safety and pharmacokinetic profiles of these approved drugs have already undergone extensive evaluation, paving the way for a more efficient transition. Studies have shown that intravenous administration of 100 mg of CEP in humans can achieve a maximum concentration (C max ) of drug in serum of 1,464 ± 364 ng/mL, which is approximately 2.41 µM (61). In mice, an intraperitoneal injection at a dose of 21 mg/kg results in a C max of 874 ng/mL (approximately 1.44 µM) (62). The C max values of CEP in humans and mice are higher than the in vitro IC 50 values observed in the 8 h antiviral studies (Fig. 1; Table 1). However, these C max values are similar to or lower than those recorded in the antiviral studies conducted over a 16 h infection period (Fig. 6; Table 2). Therefore, successful clinical translation will likely depend on dosing strategies that maintain effective drug levels over time. A notable challenge with the oral delivery of CEP is its poor solubility, which significantly hinders its effectiveness in vivo (63). To address this issue, further research is essential to improve its bioavailability, particularly through oral administra tion.
In sum, our study demonstrates that BBAs, particularly CEP, displayed broad-spectrum activity against multiple EV serotypes (Fig. 6), likely by disrupting lysosomal acidification and interfering with virus trafficking, resulting in effective inhibition of viral entry. The pH-dependent endolysosomal pathway is a common mechanism shared by many different virus groups. Therefore, further investigation into CEP's antiviral spectrum via this pathway is warranted. Additionally, as a host-directed antiviral agent, CEP presents a greater challenge for the development of drug resistance. These findings broaden the scope of BBA-based antiviral strategies and strongly support the need for further optimization and clinical investigation of CEP and similar compounds. In a landscape where therapeutic options are limited, utilizing the lysosomotropic and host-targeting properties of BBAs could provide an effective method for combating EV infections.
$$Fig 5 (Continued)$$
## References
1. Ooi, Wong, Lewthwaite et al. (2010) "Clinical features, diagnosis, and management of enterovirus 71" *Lancet Neurol*
2. Solomon, Lewthwaite, Perera et al. (2010) "Virology, epidemiology, pathogenesis, and control of enterovirus 71" *Lancet Infect Dis*
3. Liu, Wang, Huang et al. (2016) "Disease burden of enterovirus infection in Taiwan: implications for vaccination policy" *Vaccine (Auckl)*
4. Zhu, Li, Li et al. (2023) "Current status of hand-foot-and-mouth disease" *J Biomed Sci*
5. Fan, Liao, Jiang et al. (2020) "Study of integrated protective immunity induced in rhesus macaques by the intradermal administration of a bivalent EV71-CA16 inactivated vaccine" *Vaccine (Auckl)*
6. Yamayoshi, Iizuka, Yamashita et al. (2012) "Human SCARB2-dependent infection by coxsackievirus A7, A14, and A16 and enterovirus 71" *J Virol*
7. Lin, Lin, Tsou et al. (2012) "Human SCARB2-mediated entry and endocytosis of EV71" *PLoS One*
8. Kobayashi, Koike (2020) "Cellular receptors for enterovirus A71" *J Biomed Sci*
9. Galitska, Jassey, Wagner et al. (2023) "Enterovirus D68 capsid formation and stability requires acidic compartments"
10. Liu, Sheng, Van Vliet et al. (2018) "Molecular basis for the acid-initiated uncoating of human enterovirus D68" *Proc Natl Acad Sci*
11. Baggen, Thibaut, Strating et al. (2018) "The life cycle of non-polio enteroviruses and how to target it" *Nat Rev Microbiol*
12. Van Der Linden, Wolthers, Van Kuppeveld (2015) "Replication and inhibitors of enteroviruses and parechoviruses" *Viruses*
13. Paudel, Karki, Kim (2016) "Cepharanthine inhibits in vitro VSMC proliferation and migration and vascular inflammatory responses mediated by RAW264.7" *Toxicol In Vitro*
14. Gülçin, Elias, Gepdiremen et al. (2010) "Antioxidant activity of bisbenzylisoquinoline alkaloids from Stephania rotunda: cepharanthine and fangchinoline" *J Enzyme Inhib Med Chem*
15. Harada, Harada, Ueyama (2012) "The enhancement of tumor radioresponse by combined treatment with cepharanthine is accompa nied by the inhibition of DNA damage repair and the induction of apoptosis in oral squamous cell carcinoma" *Int J Oncol*
16. Samita, Ochieng, Owuor et al. (2017) "Isolation of a new β-carboline alkaloid from aerial parts of Triclisia sacleuxii and its antibacterial and cytotoxicity effects" *Nat Prod Res*
17. Desgrouas, Chapus, Desplans et al. (2014) "In vitro antiplasmodial activity of cepharanthine" *Malar J*
18. Sakurai, Kolokoltsov, Chen et al. (2015) "Ebola virus. Twopore channels control Ebola virus host cell entry and are drug targets for disease treatment" *Science*
19. Liu, Wang, Wang et al. (2020) "Antiviral effects and tissue exposure of tetrandrine against SARS-CoV-2 infection and COVID-19" *MedComm*
20. Zhu, Chen, Gao et al. (2024) "Bis-benzylisoquinoline alkaloids inhibit African swine fever virus internalization and replication by impairing late endosomal/lysosomal function" *J Virol*
21. Ohashi, Watashi, Saso et al. (2021) "Potential anti-COVID-19 agents, cepharanthine and nelfinavir, and their usage for combination treatment"
22. He, Huang, Wang et al. (2021) "Identification of bis-benzylisoquinoline alkaloids as SARS-CoV-2 entry inhibitors from a library of natural products" *Signal Transduct Target Ther*
23. Wen, Zhang, Liu et al. (2022) "Screening and identification of HTNV pv entry inhibitors with high-throughput pseudovirus-based chemiluminescence" *Virol Sin*
24. Liu, Chen, Liu et al. (2021) "Cepharanthine suppresses herpes simplex virus type 1 replication through the downregulation of the PI3K/Akt and p38 MAPK signaling pathways" *Front Microbiol*
25. Zhou, Wang, Zhang et al. (2012) "In vitro activity of cepharanthine hydrochloride against clinical wild-type and lamivudine-resistant hepatitis B virus isolates" *Eur J Pharmacol*
26. Okamoto, Ono, Baba (1998) "Potent inhibition of HIV type 1 replication by an antiinflammatory alkaloid, cepharanthine, in chronically infected monocytic cells" *AIDS Res Hum Retroviruses*
27. Matsuda, Hattori, Komizu et al. (2014) "Cepharanthine inhibited HIV-1 cell-cell transmission and cell-free infection via modification of cell membrane fluidity" *Bioorg Med Chem Lett*
28. Zhang, Shen, Wang et al. (2023) "Fangchinoline inhibits the PEDV replication in intestinal epithelial cells via autophagic flux suppression" *Front Microbiol*
29. Huang, Li, Ye et al. (2021) "Berbamine inhibits Japanese encephalitis virus (JEV) infection by compromising TPRMLs-mediated endolysosomal trafficking of low-density lipoprotein receptor (LDLR)" *Emerg Microbes Infect*
30. (2025) *Full-Length Text Antimicrobial Agents and Chemotherapy*
31. Zhang, Li, Li et al. (2024) "Identification of fangchinoline as a broadspectrum enterovirus inhibitor through reporter virus based highcontent screening" *Virol Sin*
32. Tseng, Hsu, Ling et al. (2022) "Antidepres sant sertraline is a broad-spectrum inhibitor of enteroviruses targeting viral entry through neutralization of endolysosomal acidification" *Viruses*
33. Hou, Lu, Wu et al. (2016) "Idarubicin is a broadspectrum enterovirus replication inhibitor that selectively targets the virus internal ribosomal entry site" *J Gen Virol*
34. Reed, Muench (1938) "A simple method of estimating fifty per cent endpoints12" *Am J Epidemiol*
35. Wang, Chou, Lei et al. (2004) "A mouse-adapted enterovirus 71 strain causes neurological disease in mice after oral infection" *J Virol*
36. Himmler, Mladinich, Conde et al. (2025) "Passage-attenuated Powassan virus LI9P protects mice from lethal LI9 challenge and links envelope residue D308 to neurovirulence" *mBio*
37. Lu, Lin, Kuo et al. (2021) "Muscle tissue damage and recovery after EV71 infection correspond to dynamic macrophage phenotypes" *Front Immunol*
38. Gigante, Li, Junghänel et al. (2019) "Non-viral transfection vectors: are hybrid materials the way forward?" *Medchemcomm*
39. Tang, Li, Du et al. (2020) "Emetine protects mice from enterovirus infection by inhibiting viral translation" *Antiviral Res*
40. Lyu, Yang, Head et al. (2017) "Pharmacological blockade of cholesterol trafficking by cepharanthine in endothelial cells suppresses angiogenesis and tumor growth" *Cancer Lett*
41. He, Liu, Jiang et al. (2025) "Tetrandrine augments melanoma cell immunogenicity via dual inhibition of autophagic flux and proteasomal activity enhancing MHC-I presentation" *Acta Pharmacol Sin*
42. Ren, Song, Xian et al. (2023) "Identification of Fangchinoline as a novel autophagy inhibitor with an adjuvant of chemotherapy against lung cancer" *Toxicol Appl Pharmacol*
43. Wolkerstorfer, Kurz, Bachhofner et al. (2009) "Glycyrrhizin inhibits influenza A virus uptake into the cell" *Antiviral Res*
44. Zhang, Zhou, Li et al. (2021) "The antiviral effect of novel steroidal derivatives on flaviviruses" *Front Microbiol*
45. Su, Su, Luo et al. (2024) "Cepharanthine inhibits African swine fever virus replication by suppressing AKT-associated pathways through disrupting Hsp90-Cdc37 complex" *Int J Biol Macromol*
46. Zhang, Huang, Ren et al. (2022) "Comparison of viral RNA-host protein interactomes across pathogenic RNA viruses informs rapid antiviral drug discovery for SARS-CoV-2" *Cell Res*
47. Nishimura, Lee, Hafenstein et al. (2013) "Enterovirus 71 binding to PSGL-1 on leukocytes: VP1-145 acts as a molecular switch to control receptor interaction" *PLoS Pathog*
48. Kobayashi, Sudaka, Takashino et al. (2018) "Amino acid variation at VP1-145 of enterovirus 71 determines attachment receptor usage and neurovirulence in human scavenger receptor B2 transgenic mice" *J Virol*
49. Shiraishi, Akiyama, Nakagawa et al. (1987) "Effect of bisbenzylisoquinoline (biscoclaurine) alkaloids on multidrug resistance in KB human cancer cells" *Cancer Res*
50. Nagatsuka, Nakazawa (1982) "Effects of membrane-stabilizing agents, cholesterol and cepharanthin, on radiation-induced lipid peroxidation and permeability in liposomes" *Biochimica Biophysica Acta*
51. Salata, Calistri, Parolin et al. (2017) "Antiviral activity of cationic amphiphilic drugs" *Expert Rev Anti Infect Ther*
52. Muehlbacher, Tripal, Roas et al. (2012) "Identification of drugs inducing phospholipidosis by novel in vitro data" *ChemMedChem*
53. Lu, Nixon (2016) "Autophagy Enhancers, are we there Yet?"
54. Marchant, Patel (2015) "Two-pore channels at the intersection of endolysosomal membrane traffic" *Biochem Soc Trans*
55. Ruas, Rietdorf, Arredouani et al. (2010) "Purified TPC isoforms form NAADP receptors with distinct roles for Ca 2+ signaling and endolysosomal trafficking" *Curr Biol*
56. Gunaratne, Yang, Li et al. (2018) "NAADPdependent Ca 2+ signaling regulates Middle East respiratory syndromecoronavirus pseudovirus translocation through the endolysosomal system" *Cell Calcium*
57. Gunaratne, Brailoiue, He et al. (2021) "Essential requirement for JPT2 in NAADPevoked Ca 2+ signaling" *Sci Signal*
58. Hijikata, Shionyu-Mitsuyama, Nakae et al. (2022) "Evaluating cepharan thine analogues as natural drugs against SARS-CoV-2" *FEBS Open Bio*
59. Kumazawa, Kaneko, Inagaki et al. (1990) "Accelerated recovery from γ-irradiation-induced leukopenia in mice by the biscoclaurine alkaloid, cepharanthin -Comparison with recombi nant human granulocyte colony-stimulating factor" *Int J Immunophar macol*
60. Bailly (2019) "Cepharanthine: an update of its mode of action, pharmacological properties and medical applications" *Phytomedicine*
61. Rogosnitzky, Okediji, Koman (2020) "Cepharanthine: a review of the antiviral potential of a Japanese-approved alopecia drug in COVID-19" *Pharmacol Rep*
62. Yasuda, Moro, Akasu et al. (1989) "Pharmacokinetic disposition of cepharanthin following single and multiple intravenous doses in healthy subjects" *Rinsho yakuri/Jpn J Clin Pharmacol Ther*
63. Desgrouas, Desbordes, Dormoi et al. (2014) "Quantitative analysis of cepharanthine in plasma based on semiauto matic microextraction by packed sorbent combined with liquid chromatography" *J Anal Methods Chem*
64. Liang, Li, Du et al. (2022) "Pharmacological effects and clinical prospects of cepharanthine" *Molecules* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12567652&blobtype=pdf | # Comparative Transcriptomics Analyses Identify DDX43 as a Cellular Regulator Involved in Suppressing HSV-2 Replication
Ranqing Cheng, Yuncheng Li, Yuhao Chen, Mudan Zhang, Qinxue Hu, Yalan Liu
## Abstract
HSV-2 is the main pathogen causing genital herpes, and its infection increases the infection and transmission of HIV-1. Currently, there are no vaccines to prevent HSV-2 infection or treatment that can fully cure it. Mining key host factors that regulate HSV-2 replication and elucidating their specific regulatory mechanisms are crucial for understanding virus-host interactions and discovering new antiviral targets. In the current study, we identified DDX43 as a cellular factor involved in the suppression of HSV-2 replication through comparative transcriptomic analyses of HSV-2-infected epithelial cells, followed by experimental validation. Comprehensive transcriptomic profiling revealed distinct host cellular gene expression patterns in HeLa and ARPE-19 cell lines post HSV-2 infection. Subsequent orthogonal partial least-squares discriminant analysis (OPLS-DA) pinpointed DDX43 as one of the principal mediators distinguishing the host response between HSV-2-infected HeLa and ARPE-19 cells. Furthermore, overexpression of DDX43 inhibited HSV-2 replication, whereas knockdown of endogenous DDX43 enhanced HSV-2 replication. Additional experiments revealed that human DDX43 inhibits HSV-2 replication in an interferon-independent manner. This study demonstrates that DDX43 serves as a host regulator against HSV-2 infection, underscoring the power of comparative transcriptomics in identifying novel host proteins that modulate viral replications.
## 1. Introduction
Herpes simplex virus type 2 (HSV-2) is the main pathogen causing genital herpes, primarily infecting the genital epithelium, and can be transmitted to the peripheral nervous system, where it establishes life-long latent infection [1]. Infection with HSV-2 can increase the risk of bacterial vaginosis and cervical cancer. Additionally, patients infected with HSV-2 are at a higher risk of transmitting HIV-1 [2]. Currently there are no vaccines to prevent HSV-2 infection, nor are there drugs that can completely cure it. Therefore, a thorough exploration of key host factors that modulate HSV-2 replication and mechanisms underlying their regulation is crucial for understanding the virus-host interaction network and identifying new antiviral targets.
To date, several host restriction factors that inhibit herpesvirus replication have been identified, with a particular focus on HSV-1. For instance, transmembrane protein with epidermal growth factor (EGF)-like and two follistatin-like domains 1 (TMEFF1) has been characterized as a neuron-specific restriction factor that impairs HSV-1 entry into neurons [3,4]. Peptidylarginine deaminase 3 (PAD3) exhibits a marked inhibitory activity against HSV-1 [5]. Human myxovirus resistance protein B (MxB) restricts replication of herpesviruses by inhibiting the delivery of incoming viral DNA into the nucleus [6,7]. Nevertheless, current understanding of host factors that restrict HSV-2 replication remains limited.
A variety of methodologies have been developed to identify host cell factors that suppress viral replication. With the rapid advancement of high-throughput sequencing technology and bioinformatics, omics approaches have become crucial tools for elucidating virus-host interactions. The transcriptome serves as an information-rich intermediate layer, offering critical insights into disease pathogenesis [8]. Transcriptomics allows for the systematic acquisition of both quantitative and qualitative data on gene transcript abundance and sequence variations across defined spatiotemporal contexts. It also facilitates dynamic tracking gene expression reprogramming in host cells following viral infection, providing molecular evidence to elucidate viral replication mechanisms, host immune response, and more [9][10][11]. Comparative analysis of transcriptome enables the identification of host genes and molecular pathways that play functional roles in viral infection process. In this context, global gene expression changes in cells in response to HSV-2 infection may reflect early and mechanistically significant cellular events.
HSV-2 exhibits broad tropism, predominantly infecting epithelial cells of the skin and mucosa, with a particular preference for genital mucosal cells. In vitro studies have demonstrated that multiple established cell lines are permissive to HSV-2 infection. However, the replication efficiency of HSV-2 varies significantly across different cell lines [12], and the mechanisms underlying these differences has yet to be elucidated. In this study, we found that viral yields and cytopathic effect (CPE) differ significantly between two HSV-2 permissive cell lines, HeLa and ARPE-19. Transcriptome sequencing was utilized to systematically compare and analyze the cellular transcriptomic profiles of these two HSV-2-susceptible cell types across multiple time points post infection, with the aim of identifying potential host cellular factors governing HSV-2 infection. By comparing the transcriptomic profiles across different biological states, such as HSV-2-infected HeLa cells vs. HSV-2-infected ARPE-19 cells, or various times post infection, we identified differentially expressed genes (DEGs) that exhibit both statistical significance and high specificity. As the result, the host protein DDX43 was identified and further validated through experimental assays to explore its anti-viral role in HSV-2 replication.
## 2. Materials and Methods
## 2.1. Cell Lines, Viruses and Antibodies
Human embryonic kidney 293T (HEK 293T) cells, human cervical epithelial cell line HeLa and African green monkey kidney cell line Vero were maintained in Dulbecco's modified Eagle's medium (DMEM, Huiying Biotech, Thermo Fisher Scientific, Shanghai, China) supplemented with 10% fetal bovine serum (FBS), 100 Units/mL penicillin and 100 Units/mL streptomycin at 37 • C in 5% CO 2 . Adult Retinal Pigment Epithelial cell line-19 (ARPE-19) was maintained in a 1:1 mixture of Dulbecco's modified Eagle's medium and F12 medium (Gibco) supplemented with 10% fetal bovine serum (FBS), 100 Units/mL penicillin and 100 Units/mL streptomycin at 37 • C in 5% CO 2 .
HSV-2 (strain G) was obtained from LGC standards. HSV-2-GFP carrying the complete genome of HSV-2 and green fluorescent protein (GFP) was kindly provided by Dr. Yasushi Kawaguchi, University of Tokyo, Japan. All viruses were propagated in African green monkey kidney cells (Vero). Virus titters were determined by a plaque-forming assay on Vero cells as previously described [12].
Antibody (Ab) against DDX43 was purchased from Proteintech (17591-1-AP and 68454-1-Ig, Wuhan, China). Ab against Flag tag was purchased from Sigma-Aldrich (F1804, Saint Louis, MO, USA). Ab against HSV-2 gD was purchased from Santa Cruz Biotechnology (sc69802, Sata Cruz, CA, USA). Ab against HSV2 gB was purchased from Santa Cruz Biotechnology (sc56987, Sata Cruz, CA, USA). Ab against HSV-2 major capsid protein ICP5 was purchased from Abcam (Ab6508, Cambridge, MA, USA). Abs against β-actin was purchased from Proteintech (66009-1-Ig, Wuhan, China).
## 2.2. RNA Extraction, Transcriptome Sequencing and Quantitative Real-Time PCR (qPCR)
HeLa and ARPE-19 cells were washed with PBS, followed by the addition of HSV-2 (MOI = 1) for an incubation at 37 • C for 1 h. After the removal of viruses, cells were washed three times with PBS and maintained in fresh medium supplemented with 2% FBS. The cells were collected at different time points post infection. Total RNAs were isolated from the cells using the TRIzol ® Reagent (15596026CN, Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocol, then subjected to transcriptome sequencing and qPCR. The transcriptome library construction and RNA sequencing (NovaSeq6000, Illumina, San Diego, CA, USA) were performed by Wuhan Zhiyuan Biotechnology Co., Ltd. (Wuhan, China). The sequencing depth for RNA-Seq is >30X reads/sample, yielding a total data output > 60M reads. cDNA was then synthesized using HiScript II Q RT SuperMix (Vazyme, R223-01, Nanjing, China). The newly synthesized cDNAs were used as templates for the amplification of a highly specific nucleotide region of target genes in qPCR assay. For the detection of HSV-2 genomic DNA, viral DNA was extracted from either HSV-2-infected or mock-infected HeLa and ARPE-19 cells using the QIAamp DNA Blood Mini Kit (51104, QIAGEN, Hilden, Germany). The extracted DNAs were served as templates for the amplification of highly specific nucleotide regions within the ICP0 or gB gene, with GAPDH employed as an internal control. Relative qPCR was performed on a CFX Connect Real-Time PCR System (Bio-Rad, Hercules, CA, USA) using ChamQ SYBR qPCR Master Mix (High ROX Premixed) (Vazyme, Q341-02, Nanjing, China). The final reaction conditions were as follows: 95 • C for 1 min, followed by 40 cycles of 95 • C for 15 s, 60 • C for 15 s, and 72 • C for 45 s. The difference in gene expression was calculated on the basis of 2 -∆∆CT values. The primers used in this study were shown in Appendix A (Table A1).
## 2.3. Transcriptome Data Processing
The human reference genome and annotation files were downloaded from website (http://ftp.ensembl.org/pub/current_gtf/homo_sapiens/Homo_sapiens.GRCh38.10 4.gtf.gz (accessed on 16 August 2023)). HISAT2 (v2.0.5) software was used to map the RNA-seq reads to the reference genome. After obtaining gene read counts using the Ge-nomicFeatures (v1.30.3) and GenomicAlignments (v1.14.2) packages, differential expression analysis between the 2/4/6/8/16 h time points and the 0 h reference group was performed using the DESeq2 (v1.18.1) package. Differential gene expression analysis was conducted using the DESeq2 package with the following criteria: |log2 fold change (FC)| > 0.25 and adjusted p-value < 0.05. All identified differentially expressed genes were subsequently integrated for downstream analytical processing.
## 2.4. Bioinformatic Analysis
The following software versions were employed for transcriptomic data analysis: R (version 4.2.2) and Python (version 3.9). Differentially expressed features were identified from each omic dataset as referred in database searching and data management, as well as RNA sequencing and data processing. Principal component analysis (PCA) was performed with all identified features to explore the largest sources of variation within each omics dataset. Orthogonal Partial Least-Squares Discriminant Analysis (OPLA-DA) was performed using the MetaboAnalyst platform. Time-series data were analyzed using fuzzy c-means clustering with the Mfuzz package (v2.60.0) (parameters: c = 9, m = 1.25). Data visualization was conducted using ggplot2 and Cytoscape (v3.9.1).
## 2.5. Plasmid Construction and Transfection
Human DDX43 gene and Flag-tagged DDX43 were cloned into pcDNA3.1 (+), respectively. The Flag-tagged plasmid expressing IRF-3/5D, RIG-I, MAVS, TBK-1, or cGAS/STING, the internal control plasmid phRL-TK and the reporter plasmids p125-Luc were described in our previous study [13]. All the plasmids were verified by DNA sequencing analysis (Sunny Biotechnology, Shanghai, China). Transfection of plasmids into cells was carried out using ExFect Transfection Reagent (Vazyme, T101-01, Nanjing, China) according to the manufacture's protocol.
## 2.6. Viral Plaque Assay
HeLa and ARPE-19 cells were infected with HSV-2-GFP (MOI = 0.0001) for 1 h. After the removal of viruses, cells were washed three times with PBS and maintained in fresh medium supplemented with 2% FBS. At 48 hpi, cells were washed twice with PBS, followed by fixation with 4% (w/v) cold paraformaldehyde for 15 min at room temperature. Nuclei were dyed by DAPI (AR1177, Boster, Wuhan, China). Viral plaques were detected using confocal microscopy (STELLARIS 8 WLL, Leica, Wetzlar, Germany) with a 10× objective.
## 2.7. Western Blotting
The collected cells were lysed in the lysis buffer (1 M Tris-HCl, 150 mM NaCl, 200 mM EDTA.2Na, 1% Triton X-100) supplemented with protease inhibitor cocktail (Roche, 11697498001, Basel, Switzerland). The proteins in supernatants were separated by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto 0.45 µm polyvinylidene difluoride membranes (Millipore, Boston, MA, USA). Nonspecific binding was blocked using 5% non-fat milk in TBST for 1 h at room temperature. The membranes were incubated with appropriate antibodies overnight at 4 • C and then washed three times with TBST, followed by incubation for 1 h with HRP conjugated Goat anti-Rabbit IgG (H + L) (Proteintech, SA00001-2, Wuhan, China) or Goat anti-Mouse IgG (H + L) (Proteintech, SA00001-1, Wuhan, China). After three washes with TBST, the bands were visualized by exposure to ChemiScope System (Clinx, Shanghai, China) following the addition of chemiluminescent substrate (Super ECL Plus, S6009M, US EVERBRIGHT, Shanghai, China; Westernbright ECL, K-12045-D50, Advansta, San Jose, CA, USA). The grayscale values of the Wb bands were analyzed using ImageJ (Version 1.33 h). The relative protein expression levels of DDX43, gB, gD, or ICP5 were quantified through the following procedure: First, the gray value of each target protein band (DDX43, gB, gD, or ICP5) was normalized to the corresponding actin band gray value to obtain a lane-specific ratio. Subsequently, the control sample ratio was designated as 1, and the ratios of all other experimental samples were calibrated by dividing them by the control ratio to yield the final normalized values. The triplicate measurements were then averaged, with the mean value presented beneath the corresponding protein band.
## 2.8. RNA Interference
Three specific shRNAs targeted DDX43 and negative control shRNA (Table A2) were designed and constructed into the pLKO.1 vector according to the manufacturer's in-struction. These plasmids and helper plasmids were cotransfected into 293T cells (0.8 µg pMD2.G, 6.67 µg pSPAX2 and 7.5 µg pLKO.1 per dish) for 48 h to produce lentiviruses. At 2 dpi, the culture medium containing the viruses was harvested, collected and filtered by 0.45 µm membrane, followed by ultracentrifugation at 24,000 rpm for 2 h at 4 • C. The purified lentiviruses were aliquoted and stored at -80 • C. For shRNA knockdown of DDX43, lentiviruses and puromycin (5 µg/mL) were added to HeLa cells pre-seeded in culture dishes for an additional 7 days. Cells were harvested and then subjected to WB assay to measure the expression level of DDX43.
## 2.9. Statistical Analysis
Normality of quantitative data was first analyzed using Graphpad Prism 10.4.1 (Graph-Pad Software, Inc.). Comparisons among groups in experiments data were performed by a two-way ANOVA followed by Bonferroni's multiple comparisons test. ns, not significant; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. All analyses were based on at least three biological replicates. All experiments were repeated at least three times. Data are presented as mean values ± SD of three independent experiments.
## 3. Results
## 3.1. Comprehensive Transcriptomic Analysis of Cellular Gene Expression Profiles of HeLa and ARPE-19 Cells in Response to HSV-2 Infection
HeLa and ARPE-19 are commonly used model cell lines in HSV-2 study. Although both cell lines are susceptible to HSV-2, there were significant differences in the yields of progeny viruses and cytopathic effect (CPE) following infection (Figure 1). The yield of progeny viruses in HeLa cells infected with HSV-2 at the same multiplicity of infection (MOI) was significantly lower than that in ARPE-19 cells ((5.5 ± 0.16) × 10 5 vs. (7.5 ± 0.25) × 10 6 PFU/mL) (Figure 1A). Additionally, the size of plaques and syncytia in HeLa cells was significantly smaller than that in ARPE-19 cells (0.02 ± 0.002 Vs. 2.78 ± 0.39 mm 2 ), while the number of syncytia in HeLa cells was markedly reduced (24 ± 1.25 Vs. 35 ± 2.87 per well) (Figure 1B). To elucidate the mechanisms underlying the differential replication of HSV-2 in various host cell types, we first analyzed the host cellular gene expression profiles in response to HSV-2 infection. HeLa and ARPE-19 cells were collected for transcriptome sequencing at different time points (0, 2, 4, 6, 8, 16 h) post infection, with the 0 h time point serving as the control group. After assessing the quality of raw sequencing data and removing low-quality reads, a thorough analysis of the transcriptome sequencing data was performed. The Venn diagram representing the transcriptomes of HeLa and ARPE-19 cells identified a total of 24,135 transcripts. Among these, 28 transcripts in HeLa cells were consistently present across the samples from five distinct time points: 2, 4, 6, 8, and 16 h. As for ARPE-19 cells, only 3 transcripts were consistently detected across samples from five distinct time points (Figure 2A). Principle component analysis (PCA) revealed that the control group (0 h) was clearly separated from the other groups (2, 4, 6, 8 and 16 h) (Figure 2B), suggesting significant transcriptomic expression disparities among the samples at various time points post infection, with strong reproducibility. The clustering heatmap illustrated the genes that were upregulated or downregulated in the transcriptome samples at different time points post-infection (Figure 2C). These differentially expressed genes are likely to serve as potential regulators influencing HSV-2 replication.
## 3.2. Comparative Transcriptomic Profiling Identifies DDX43 as One of the Principal Mediators Distinguishing HSV-2-Infected HeLa and ARPE-19 Cell Lines
An inter-group analysis was subsequently carried out on the transcriptomic sequencing data obtained from ARPE-19 and HeLa cells infected with HSV-2. The PCA scatter plot shows significant differences in the transcriptional expression levels between ARPE-19 and HeLa cells following HSV-2 infection (Figure 3A). Subsequently, orthogonal partial least-squares discriminant analysis (OPLS-DA) was performed on the transcriptome data of the two cell lines using MetaboAnalyst. Based on the OPLS-DA model, the genes with the highest variable importance in the projection (VIP) scores, which contributed to the differences between the two groups, were predicted (Figure 3B,C). As shown in Figure 3B, R2 is the correlation coefficient after PLS-DA analysis, indicating the fitting of the model, that is, the extent to which the established model (component) can represent the real data (generally, when R2 is between 0.7 and 0.8, it indicates that the model has good explanatory power). Q2 represents the predictive performance of the PLS-DA model. In general, a Q2 value greater than 0.5 indicates good predictive ability, and the values of R2 and Q2 should be relatively close. As the #2 component achieved the highest score, it was selected for prediction. Using the #2 component, the top 25 genes with the highest VIP scores were identified, as these genes significantly contribute to the divergence between the two comparative groups (Figure 3C). The higher the VIP value, the more influential the gene is in distinguishing between different groups. Among the top 25 differentially expressed genes, sortilin-related VPS10 domain containing receptor 3 (SORCS3) exhibited the highest score, followed by DEAD-box helicase 43 (DDX43). We focused on genes annotated with antiviral-related functions, such as interferon-stimulated genes (ISGs), and found that neither of these two genes have been previously linked to viral infection. Several members of the DDX family have been demonstrated to play pivotal roles in antiviral defense [14][15][16]. Notably, DDX43 was shown to modulate the interferon (IFN) signaling pathway in Oreochromis niloticus [17], suggesting that it may influence HSV-2 infection. was carried out on the transcriptomic sequencing data obtained from HSV-2-infected ARPE-19 and HeLa cells, with each group consisting of three biological replicates. (A) The PCA scatter plot revealed significant differences in the transcriptional expression levels between ARPE-19 and HeLa cells following HSV-2 infection. 1 is ARPE-19, and 2 is HeLa. (B) OPLS-DA was performed on the transcriptome data of the two cell lines using MetaboAnalyst. R2 is the correlation coefficient after PLS-DA analysis, indicating the fitting of the model. Q2 represents the predictive performance of the PLS-DA model. Generally, the values of R2 and Q2 should be relatively close. Among all components, Component #2 shows the closest agreement between R2 and Q2 (indicated by the red inverted triangle). (C) The genes with the highest VIP scores, which contributed to the divergence in viral replication levels between the two comparative groups, were identified by using #2 component in (B).
## 3.3. Verification of the Dynamic Expression of DDX43 in Response to HSV-2 Infection
DDX43 (also named HAGE), belonging to the DEAD-box helicase subfamily, is a potential prognostic marker in patients with breast cancer [18,19]. It is a dual RNA-DNA helicase, and its KH domain is required for its full unwinding activity [20]. DDX43 regulates RAS protein expression and AKT activation [21], prevents the expression of PML in ABCB5+ malignant melanoma-initiating cells [22], and plays a critical role in the male sex differentiation of channel catfish (Ictalurus punctatus) [23]. In mammals, DDX43 plays an important role in the malignant proliferation and immune response of tumor cells [24]. Currently, there is no direct evidence that viral infection affects the expression of DDX43. To verify the accuracy of the aforementioned results from transcriptomics analysis, the dynamics of DDX43 expression in response to HSV-2 infection was assessed. HeLa and APRE-19 cells were infected with or without HSV-2, and collected at different time points post infection (0, 2, 4, 6, 8 and 16 hpi), respectively. The mRNA levels of DDX43 were examined using quantitative real-time PCR (qPCR). As shown in Figure 4A,B, the mRNA level of DDX43 exhibited a progressive increase in HeLa cells, while it gradually declined in ARPE-19 cells during the course of infection, consistent with the initial transcriptome sequencing results (Figure 4C,D). The expression dynamics of genes that are rapidly upregulated in the early stages of infection and remain continuously elevated are more likely to represent key host regulators. The protein level of DDX43 was also examined at different time points following HSV-2 infection, showing that HSV-2 infection resulted in an increase in DDX43 at the protein level in HeLa cells (Figure 4E). The observed concurrent regulation of both mRNA and protein levels of DDX43 during HSV-2 infection suggests that these transcriptional and translational alterations in DDX43 expression are likely mechanistically involved in the pathophysiological processes of HSV-2 infection.
## 3.4. Overexpression of DDX43 Inhibits HSV-2 Replication
We next sought to determine whether the continuously increasing expression of DDX43 in HSV-2-infected HeLa cells contributes to the lower viral yield. To investigate whether DDX43 functions as a cellular regulator influencing HSV-2 proliferation, HeLa cells were transfected with an empty vector or a plasmid encoding DDX43, followed by HSV-2 infection. At 18 hpi, the cells were collected for qPCR, western blotting, and plaque assays, respectively. As shown in Figure 5, overexpression of DDX43 reduced the viral yields ((6.4 ± 0.29) × 10 5 vs. (2.0 ± 0.05) × 10 5 PFU/mL) (Figure 5A), significantly decreased the relative abundance of HSV-2 genome (as indicated by the gene copies of immediately early protein ICP0 and envelope glycoprotein B (gB)) relative to internal reference protein GAPDH (Figure 5B), and downregulated the protein expression levels of major capsid protein ICP5, gB and envelope glycoprotein D (gD) (Figure 5C). To further substantiate the occurrence of DDX43 overexpression, a tagged DDX43 construct was employed to distinguish it from endogenous levels, and the overexpression experiments were repeated. As shown in Figure 5D-F, the results obtained with Flag-DDX43 overexpression were consistent with the previous findings. To explore the correlation between DDX43 expression and HSV-2 proliferation, HeLa cells were transfected with varying concentrations of empty vector and a plasmid encoding Flag-DDX43, followed by HSV-2 infection. As shown in Figure 5G, overexpression of Flag-DDX43 significantly reduced the yield of HSV-2 ((6.7 ± 0.65) × 10 5 vs. (2.0 ± 0.16) × 10 5 vs. (1.8 ± 0.08) × 10 5 vs. (1.4 ± 0.12) × 10 5 PFU/mL). Moreover, the extent of viral reduction is positively correlated with the transfection dosage of Flag-DDX43. The antiviral activity of DDX43 was further confirmed in ARPE-19 cells. Briefly, ARPE-19 cells were transfected with either an empty vector or a Flag-DDX43-expressing plasmid, followed by HSV-2 infection. At 18 hpi, samples were collected for western blotting and plaque assays, respectively. The results are consistent with those on HeLa cells. Overexpression of Flag-DDX43 reduced the viral yields ((4.7 ± 0.12) × 10 6 vs. (1.4 ± 0.12) × 10 6 PFU/mL) (Figure 5H), decreased the relative abundance of HSV-2 genome (Figure 5I) and downregulated the protein expression levels of ICP5, gB and gD (Figure 5J). These data suggest that DDX43 is a cellular regulator that suppresses HSV-2 replication. The HSV-2 genome was detected by qPCR using ICP0 and gB primers (E). The protein expression levels of DDX43, HSV-2 gD, gB, and ICP5 were measured by Western blot (F). (G) HeLa cells were transfected with varying concentrations of plasmid encoding Flag-DDX43, followed by infection with HSV-2. At 18 hpi, the viral yields were detected by plaque assay and the protein expression level of DDX43 was measured by Western blot. (H-J) ARPE-19 cells transfected with pcDNA3.1 (+) empty vector or plasmid expressing Flag-DDX43. At 24 h post transfection, cells were infected with HSV-2 (MOI = 3) for 18 h, and the cell and culture medium were collected. The viral yields were measured by plaque assay (H). The HSV-2 genome was detected by qPCR using ICP0 and gB primers (I). The protein expression levels of DDX43, HSV-2 gD, gB, and ICP5 were measured by Western blot (J). For graphs, data shown are mean ± SD of three independent experiments with each condition performed in triplicate (* p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001). For images, one representative experiment out of three is shown.
## 3.5. Knockdown of Endogenous DDX43 Enhances HSV-2 Replication
To further validate the function of DDX43 in HSV-2 infection, retroviral vectors expressing DDX43 shRNA or control shRNA were used. Western blot showed that, at 48 h post transfection, both #1 and #2 DDX43 shRNAs effectively downregulated the expression of DDX43, with #1 DDX43 shRNA demonstrating superior knockdown efficiency compared to #2 shRNA (Figure 6A). Subsequently, #1 DDX43 shRNA was used in the subsequent experiments. HeLa cells transduced with DDX43 shRNA (designated as HeLa-DDX43-KD) or control shRNA (designated as HeLa-Ctrl) were infected with HSV-2. As shown in Figure 6, knockdown of endogenous DDX43 enhanced viral production (Figure 6B) and increased the relative abundance of HSV-2 genome (as indicated by the gene copies of ICP0 and gB) (Figure 6C). Furthermore, HeLa cells transfected with DDX43 shRNA exhibited markedly higher protein expression levels of HSV-2 gD, gB, and ICP5 compared to cells treated with control shRNA, or wild-type HeLa cells (Figure 6D). The anti-HSV-2 effect of DDX43 was further confirmed by reintroducing DDX43-expressing plasmids into HeLa-DDX43-KD. After replenishing DDX43 in the DDX43-knockdown cells, the inhibition of viral replication was restored. The viral yields (Figure 6E), the relative abundance of HSV-2 genome (as indicated by the gene copies of ICP0) (Figure 6F), and the protein expression levels of HSV-2 gD (Figure 6G) in the HeLa-DDX43-KD with DDX43 replenishment were comparable to those in HeLa-Ctrl, but remained lower than those in HeLa-DDX43-KD. These findings further confirmed that DDX43 serves as an intrinsic cellular regulator that suppresses HSV-2 replication. E-G) DDX43 expressing plasmid was transfected into HeLa-DDX43-KD (designated as HeLa-DDX43-KD + DDX43). At 24 h post transfection, HSV-2 was used to infect HeLa-Ctrl, HeLa-DDX43-KD and HeLa-DDX43-KD + DDX43 (MOI = 3). At 18 hpi, the viral yields were measured by plaque assay (E), the HSV-2 genomes were detected by qPCR using ICP0 primers (F), and the protein expression levels of HSV-2 gD and DDX43 were measured by Western Blot (G). For graphs, data shown are mean ± SD of three independent experiments with each condition performed in triplicate (* p < 0. 05; ** p < 0.01; *** p < 0.001; **** p < 0.0001). For images, one representative experiment out of three is shown.
## 3.6. Human DDX43 Inhibits Viral Replication in an Interferon-Independent Manner
Although Oreochromis niloticus DDX43 (OnDDX43) has been demonstrated to activate the IFN-β signaling pathway, the specific biological function of human DDX43 remains unclear due to the relatively low amino acid sequence identity (49%) between the Oreochromis niloticus and Homo sapiens orthologs [17]. It remains uncertain whether human DDX43 can similarly regulate the expression of IFN, thereby modulating HSV-2 replication. To investigate the molecular mechanisms by which DDX43 suppresses HSV-2 replication, we firstly conducted the temporal analysis of transcriptome of HeLa cells following HSV-2 infection at 0, 2, 4, 6, 8 and 16 hpi. Using fuzzy c-means clustering, 214 pathways were identified and clustered into 8 discrete expression clusters to illustrate the relative expression changes in genes in HeLa cells infected with HSV-2. As shown in Figure 7A, the top 3 pathways enriched in each cluster in the DAVID database were listed next to the clusters. The key pathways highlighted in red are considered to be continuously upregulated during HSV-2 infection, while those highlighted in blue are considered to be continuously downregulated. Considering the antiviral function of DDX43, the three continuously upregulated pathways were selected as the targets for further investigation. It is noteworthy that all three continuously upregulated pathways are closely associated with the IFN-related signaling pathway. Following the transfection of HEK 293T and Vero cells with either pcDNA3.1 (+) or a plasmid expressing DDX43, along with IFN-β-Luc promoter reporter plasmid p125-Luc and internal control plasmid phRL-TK, the dual luciferase report (DLR) assay revealed that overexpression of DDX43 promoted the activation of the IFN-β promoter in 293T cells (Figure 7B). However, due to a spontaneous genetic deletion, the Vero cell line is deficient in the production of IFN-α and IFN-β, and thus overexpression of DDX43 did not enhance the activation of the IFN-β promoter in these cells (Figure 7C).
In viral infections, the IRF-3 mediated signaling pathway is widely recognized for its pivotal role in inducing the expression of type I interferon [25,26]. To identify the potential mechanism by which DDX43 activates IFN-β promoter, the plasmid expressing RIG-I, MAVS, TBK-1, IRF-3/5D or cGAS/STING, which are inducers of IFN-β in the IRF-3 mediated signaling pathway, was transfected into HEK 293T cells together with p125-Luc, phRLTK, plasmid expressing DDX43, or empty vector. As shown in Figure 7D-H, overexpression of RIG-I (Figure 7D), MAVS (Figure 7E), TBK-1 (Figure 7F), IRF-3/5D (Figure 7G) or cGAS/STING (Figure 7H) directly induced the activation of IFN-β promoter. In the presence of DDX43, the activation of IFN promoter by the above components was enhanced, showing that DDX43 promotes RIG-I, MAVS, TBK1, IRF3/5D or cGAS/STING induced IFN-β promoter activation. We further investigated the effect of DDX43 on IFNβ induction. However, DDX43 did not enhance the production of IFN-β at the mRNA level (Figure 7I). Taken together, our data collectively demonstrate that human DDX43 suppresses viral replication through an interferon-independent mechanism and is likely to engage in synergistic interactions with components of the IRF-3-mediated signaling pathway via an undefined mechanism.
## 4. Discussion
Following viral infection, host cells initiate specific defensive responses, often leading to dysregulated gene expressions that characterize viral pathogenesis. Currently, few studies have investigated the differences in HSV-2 replication cross distinct cell lines [12], and the underlying mechanisms responsible for the substantial variability in replication efficiency among these cell types remain poorly understood. In this study, we observed that the yield of progeny viruses in HeLa cells infected with HSV-2 at the same multiplicity of infection (MOI) was significantly lower than that in ARPE-19 cells. Additionally, the size of plaques and syncytia in HeLa cells was notably smaller, and the number of syncytia was significantly reduced compared to ARPE-19 cells. This phenomenon is likely due to differences in the intrinsic regulatory mechanisms within distinct host cells. RNA sequencing was employed on HeLa and ARPE-19 cells at various time points post HSV-2 infection, followed by transcriptomics analysis, to investigate the dynamic changes in cellular components of HeLa and ARPE-19 cells during HSV-2 infection. In the (O)PLS-DA analysis, 25 host genes with the most significant differences in expression were identified. Among these, DDX43 and SORCS3 exhibited the greatest changes in expression, and, to date, no studies have linked them to HSV-1/2 infection. A previous study demonstrated that DDX43 recruits TRIF or IPS-1 as an adaptor to activate the IFN-β pathway in Nile tilapia (Oreochromis niloticus) [14]. In this study, we focus on investigating the antiviral function of DDX43 to validate the reliability of the omics strategy and data. It is worthy noting that both SORCS3 and DDX43 are of interest to us, with SORCS3 being investigated in a separate ongoing project.
The DExD/H-box (DDX) helicase family shares a conserved catalytic core, with members exhibiting complex roles in viral infections. While some DDXs have been shown to possess antiviral activities, others have been implicated in facilitating viral infection. Notably, several DDXs exhibit dual functions, both inhibiting and promoting viral replication. For instance, DDX3 has been reported to facilitate viral replication across a wide range of pathogens [27][28][29][30][31][32][33]. Additionally, DDX3 has been shown to exhibit antiviral roles by stimulating interferon production [34] and inhibiting HIV-1 replication [14]. DDX5 has been found to promote viral replication [35][36][37][38][39] and also been shown to inhibit the replication of two DNA viruses, HBV [40] and myxoma virus (MYXV) [41]. Similarly, DDX56 acts as a positive regulator for the replication of several viruses [42][43][44], but serves as a negative regulator for the replication of PRV [15] and Chikungunya virus [16]. Together, these findings suggest that specific members of the DDX family play distinct functional roles during infections caused by different viruses. As a member of the DDX family, the role of DDX43 in viral infections has not been reported. We speculated that the divergent alterations in DDX43 expression levels observed in HSV-2-infected HeLa and ARPE-19 cells may contribute significantly to the difference in progeny virus yield between these two cell lines. DDX43 expression was reduced in HSV-2-infected ARPE-19 cells, which appeared to facilitate the production of HSV-2 progeny viruses. In contrast, the increased expression of DDX43 in HSV-2-infected HeLa cells seemed to inhibit HSV-2 progeny virus production. Further wet experiments confirmed this hypothesis. While overexpression of DDX43 in HeLa cells led to a decrease in the yield of HSV-2 progeny viruses, knockdown of endogenous DDX43 in HeLa cells resulted in an increase in progeny virus yield. These results demonstrated that DDX43 functions as a host regulator to inhibit HSV-2 replication. Our investigation, however, has been limited to exploring the role of DDX43 during HSV-2 infection. Given that various members of the DDX family (including, but not limited to, DDX3, DDX5, and DDX56) exhibit distinct regulatory effects on different viral infections, further studies are warranted to determine whether DDX43 exerts a proviral or antiviral effect on other viruses.
The identification of DDX43 as a host regulator that inhibits HSV-2 replication enhances our understanding of the complex host-pathogen interactions underlying viral pathogenesis, while also offering a promising novel therapeutic target for antiviral intervention. Nevertheless, this study has certain limitations. For instance, in investigating virus-host interactions, single-omics analyses can provide molecular insights at a specific level but often fail to fully capture the complexity of the dynamic regulatory networks between viruses and hosts. Further research should integrate molecular data across various levels, including genomics, transcriptomics, proteomics, and metabolomics to enable a systematic dissection of the spatiotemporal dynamics of virus-host interactions. Such an integrative approach would provide a powerful framework for elucidating viral pathogenesis, characterizing host immune responses, and identifying potential therapeutic targets. Nevertheless, the mechanism by which DDX43 suppresses HSV-2 remains to be further investigated. Given that HSV-2 replication is regulated at multiple stages, including viral attachment, entry, immediate-early gene expression, early gene expression, DNA replication, late gene expression, virion assembly, and egress. Elucidating the underlying mechanism could provide valuable insights for identifying potential therapeutic targets. Based on the observed reductions in gene copy numbers of the immediate-early protein ICP0 and the late protein gB, along with the downregulation of protein expression levels for the major capsid protein ICP5, gB, and gD, it is highly likely that DDX43 exerts its repressive effect at the immediate-early gene expression stage. Further investigations will be conducted to elucidate the underlying molecular mechanisms.
The innate immune system serves as the host's primary defense against pathogens, operating through rapid detection of pathogen-associated molecular patterns and subsequent induction of effector molecules such as IFNs. IFNs establish a broad-spectrum antiviral state by activating downstream signaling cascades while coordinating adaptive immune responses. Although Oreochromis niloticus DDX43 has been shown to activate the IFN-β signaling pathway, our results suggest that human DDX43 may have a distinct functional role on IFN production, as evidenced by the relatively low amino acid sequence identity (49%) between the O. niloticus and H. sapiens orthologs. In our study, DDX43 was found to potentiate the activation of the IFN-β promoter induced by RIG-I, MAVS, TBK1, IRF3/5D, or the cGAS/STING pathway, yet it did not directly elevate the mRNA expression levels of IFN-β (Figure 7). Our findings suggested that human DDX43 suppresses viral replication through an interferon-independent mechanism and is likely to engage in synergistic interactions with components of the IRF-3-mediated signaling pathway, thereby enhancing IFN-β promoter activity via an as-yet undefined mechanism. However, given that multiple pathways influence IFN production, it cannot be ruled out that human DDX43 may activate IFN-β expression through other mechanisms. Beyond the scope of the present study, additional research will investigate other specific mechanisms through which DDX43 inhibits HSV-2 replication.
## DDX43
## References
1. Brown (2000) "HSV-2 specific serology should be offered routinely to antenatal patients" *Rev. Med. Virol*
2. Silhol, Coupland, Baggaley et al. (2021) "What Is the Burden of Heterosexually Acquired HIV Due to HSV-2? Global and Regional Model-Based Estimates of the Proportion and Number of HIV Infections Attributable to HSV-2 Infection" *J. Acquir. Immune Defic. Syndr*
3. Chan, Liu, Bastard et al. "Human TMEFF1 is a restriction factor for herpes simplex virus in the brain" *Nature*
4. Dai, Idorn, Serrero et al. "TMEFF1 is a neuron-specific restriction factor for herpes simplex virus" *Nature*
5. Pasquero, Gugliesi, Biolatti et al. (2023) "Citrullination profile analysis reveals peptidylarginine deaminase 3 as an HSV-1 target to dampen the activity of candidate antiviral restriction factors" *PLoS Pathog*
6. Crameri, Bauer, Caduff et al. (1980) "MxB is an interferon-induced restriction factor of human herpesviruses" *Nat. Commun*
7. Schilling, Bulli, Weigang et al. (2018) "Human MxB Protein Is a Pan-herpesvirus Restriction Factor" *J. Virol*
8. De Jong, Bosco (2021) "Unlocking immune-mediated disease mechanisms with transcriptomics" *Biochem. Soc. Trans*
9. Troha, Im, Revah et al. (2018) "Comparative transcriptomics reveals CrebA as a novel regulator of infection tolerance in D. melanogaster" *PLoS Pathog*
10. Arnold, Guito, Altamura et al. (2018) "Transcriptomics Reveal Antiviral Gene Induction in the Egyptian Rousette Bat Is Antagonized In Vitro by Marburg Virus Infection" *Viruses*
11. Sorce, Nuvolone, Russo et al. (2020) "Genome-wide transcriptomics identifies an early preclinical signature of prion infection" *PLoS Pathog*
12. Liu, Guan, Li et al. (2018) "HSV-2 glycoprotein J promotes viral protein expression and virus spread" *Virology*
13. Zhang, Liu, Wang et al. (2015) "HSV-2 immediate-early protein US1 inhibits IFN-beta production by suppressing association of IRF-3 with IFN-beta promoter" *J. Immunol*
14. Gringhuis, Hertoghs, Kaptein et al. (2016) "HIV-1 blocks the signaling adaptor MAVS to evade antiviral host defense after sensing of abortive HIV-1 RNA by the host helicase DDX3" *Nat. Immunol*
15. Xie, Li, Yang et al. "DDX56 inhibits PRV replication through regulation of IFN-β signaling pathway by targeting cGAS"
16. Taschuk, Tapescu, Moy et al. "DDX56 Binds to Chikungunya Virus RNA To Control Infection"
17. Zhou, Gao, Lu et al. (2022) "DDX43 recruits TRIF or IPS-1 as an adaptor and activates the IFN-β pathway in Nile tilapia (Oreochromis niloticus)" *Mol. Immunol*
18. Abdel-Fatah, Mcardle, Johnson et al. (2014) "HAGE (DDX43) is a biomarker for poor prognosis and a predictor of chemotherapy response in breast cancer" *Br. J. Cancer*
19. Amer, Khairat, Hammad et al. "DDX43 mRNA expression and protein levels in relation to clinicopathological profile of breast cancer" *PLoS ONE*
20. Talwar, Vidhyasagar, Qing et al. (2017) "The DEAD-box protein DDX43 (HAGE) is a dual RNA-DNA helicase and has a K-homology domain required for full nucleic acid unwinding activity" *J. Biol. Chem*
21. Ambrosini, Khanin, Carvajal et al. (2014) "Overexpression of DDX43 mediates MEK inhibitor resistance through RAS Upregulation in uveal melanoma cells" *Mol. Cancer Ther*
22. Mathieu, Miles, Ahmad et al. (1061) "The helicase HAGE prevents interferon-alpha-induced PML expression in ABCB5+ malignant melanoma-initiating cells by promoting the expression of SOCS1" *Cell Death Dis*
23. Tian, Tan, Bao et al. (2017) "DExD/H-box RNA helicase genes are differentially expressed between males and females during the critical period of male sex differentiation in channel catfish" *Comp. Biochem. Physiol. Part D Genom. Proteom*
24. Linley, Mathieu, Miles et al. (2012) "The helicase HAGE expressed by malignant melanoma-initiating cells is required for tumor cell proliferation in vivo" *J. Biol. Chem*
25. Hiscott (2007) "Triggering the innate antiviral response through IRF-3 activation" *J. Biol. Chem*
26. Au, Moore, Lowther et al. (1995) "Identification of a member of the interferon regulatory factor family that binds to the interferon-stimulated response element and activates expression of interferon-induced genes"
27. Khadivjam, Bonneil, Thibault et al. (2023) "RNA helicase DDX3X modulates herpes simplex virus 1 nuclear egress" *Commun. Biol*
28. Wang, Kim, Ryu (2009) "DDX3 DEAD-Box RNA helicase inhibits hepatitis B virus reverse transcription by incorporation into nucleocapsids" *J. Virol*
29. Angus, Dalrymple, Boulant et al. (2010) "Requirement of cellular DDX3 for hepatitis C virus replication is unrelated to its interaction with the viral core protein" *J. Gen. Virol*
30. Li, Ge, Li et al. (2014) "Cellular DDX3 regulates Japanese encephalitis virus replication by interacting with viral un-translated regions" *Virology*
31. Yedavalli, Neuveut, Chi et al. (2004) "Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function" *Cell*
32. Thulasi Raman, Liu, Pyo et al. (2016) "DDX3 Interacts with Influenza A Virus NS1 and NP Proteins and Exerts Antiviral Function through Regulation of Stress Granule Formation" *J. Virol*
33. Loureiro, Zorzetto-Fernandes, Radoshitzky et al. (2018) "DDX3 suppresses type I interferons and favors viral replication during Arenavirus infection" *PLoS Pathog*
34. Schroder, Baran, Bowie (2008) "Viral targeting of DEAD box protein 3 reveals its role in TBK1/IKKepsilon-mediated IRF activation" *EMBO J*
35. Chen, Chen, Poon et al. (2009) "Interaction between SARS-CoV helicase and a multifunctional cellular protein (Ddx5) revealed by yeast and mammalian cell two-hybrid systems" *Arch. Virol*
36. Zhao, Ge, Wang et al. (2015) "The DEAD-box RNA helicase 5 positively regulates the replication of porcine reproductive and respiratory syndrome virus by interacting with viral Nsp9 in vitro" *Virus Res*
37. Zhou, Luo, Mills et al. (2013) "DDX5 facilitates HIV-1 replication as a cellular co-factor of Rev" *PLoS ONE*
38. Goh, Tan, Lim et al. (2004) "Cellular RNA helicase p68 relocalization and interaction with the hepatitis C virus (HCV) NS5B protein and the potential role of p68 in HCV RNA replication" *J. Virol*
39. Li, Ge, Li et al. (2013) "The DEAD-box RNA helicase DDX5 acts as a positive regulator of Japanese encephalitis virus replication by binding to viral 3 ′ UTR" *Antivir. Res*
40. Li, Rahman, Bi et al. "RNA Helicase DDX5 in Association With IFI16 and the Polycomb Repressive Complex 2 Silences Transcription of the Hepatitis B Virus by Interferon" *J. Med*
41. Rahman, Bagdassarian, Ali et al. (2017) "Identification of host DEAD-box RNA helicases that regulate cellular tropism of oncolytic Myxoma virus in human cancer cells" *Sci. Rep*
42. Pirincal, Turan (2021) "Human DDX56 protein interacts with influenza A virus NS1 protein and stimulates the virus replication" *Genet. Mol. Biol*
43. Xu, Anderson, Hobman (2011) "The capsid-binding nucleolar helicase DDX56 is important for infectivity of West Nile virus" *J. Virol*
44. Xu, Xie, Zhang et al. (2022) "DDX56 antagonizes IFN-beta production to enhance EMCV replication by inhibiting IRF3 nuclear translocation" *Vet. Microbiol*
45. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12542282&blobtype=pdf | # Impact of Asaia bacteria on Leishmania major development in sand flies: implications for vector control strategies
Marketa Stejskalova, Magdalena Jancarova, Katerina Pruzinova, Kristina Capova, Ilaria Varotto-Boccazzi, Sara Epis, Petr Volf
## Abstract
Background Asaia spp., bacteria originally isolated from tropical plants, have also been identified in various insect species, including blood-feeding ones. Their ability to colonize different host tissues and transmit vertically between generations makes these bacteria good candidates for paratransgenesis. However, most existing data derived from studies on mosquitoes and other important vectors, such as phlebotomine sand flies (Diptera: Psychodidae), remain understudied. In this study, we investigated the ability of wild-type Asaia siamensis, Asaia krungthepensis, and a genetically modified strain of Asaia expressing the Wolbachia surface protein (Asaia WSP ) to colonize Phlebotomus duboscqi. In addition, we studied their vertical transmission and their interactions with Leishmania major during superinfection.Methods Phlebotomus duboscqi females were provided with Asaia via a sugar meal. Bacterial presence and vertical transmission were assessed using both cultivation and polymerase chain reaction (PCR). In superinfection experiments, females were first offered sugar containing Asaia, followed by a blood meal infected with Le. major. The outcomes of superinfection were assessed by cultivation, PCR, and microscopically. Statistical analyses were performed using Fisher's exact or Chi-squared tests.
ResultsAll tested Asaia strains colonized the gut of Ph. duboscqi. Both A. siamensis and A. krungthepensis were vertically transmitted to the progeny via egg smearing. These bacteria did not affect the infection rate and intensity of Le. major infection on days 2 and 5 post blood meal (pbm). However, by day 8 pbm, both species significantly reduced Le. major infection intensity. Moreover, A. krungthepensis significantly increased the proportion of metacyclic forms. Interestingly Asaia WSP did not have a significant effect on Le. major development in Ph. duboscqi.
ConclusionsWe demonstrated for the first time that A. siamensis and A. krungthepensis can infect Ph. duboscqi and be vertically transmitted to the next generation via egg smearing. These bacteria affect the late phase of Le. major infection, which could have important epidemiological consequences.
pathogenic, when the microbes reduce their host fitness (i.e., antagonistic interactions); neutral, with no effect on the host fitness (i.e., commensal interactions); or beneficial, when the microbes enhance host fitness (i.e., mutualistic interactions) [4].
Acetic acid bacteria of the genus Asaia, originally isolated from tropical plants [5][6][7], are also able to colonize sugar-feeding insects from phylogenetically distant taxonomic groups [8][9][10]. Among blood sucking arthropods, Asaia sp. were detected in different mosquito and sand fly species [8,[11][12][13][14][15][16], reduviids [17], Culicoides spp. midges, and the tick Haemaphysalis longicornis [18]. The potential infectivity and risk of Asaia spp. to humans has also been raised as a concern, given that several reports have documented human infections, likely in the context of opportunistic infections, as Asaia does not circulate or establish stable transmission within the human population [19][20][21][22][23][24][25]. The most detailed insights into Asaia come from studies conducted on mosquitoes, where Asaia has been identified as a highly promising candidate for the paratransgenic control of vector-borne diseases [11,[26][27][28]. These bacteria colonize multiple mosquito tissues, including the gut, crop, salivary glands, and reproductive organs of both male and female mosquitoes [11,12,29,30]. Asaia is transmitted between mosquitoes of the genus Anopheles during copulation, vertically from mother to offspring through egg-smearing, and horizontally during cofeeding [11,12,29]. Nevertheless, relatively little is known about the physiological role of this bacterium in mosquitoes. In addition to being a component of the microbiome, Asaia influences Anopheles mosquito larval development and adult longevity [31][32][33], causes transcriptomic changes [32,33], and induces the activation of mosquito immunity [34]. However, the underlying mechanisms of these effects remain unknown.
Sand flies (Diptera: Psychodidae) are small nocturnal insects; both sexes feed on plant sap, nectar, and honeydew, while females also take blood meals on various vertebrate hosts. These insects have significant veterinary and medical importance as vectors of various pathogens infecting humans or domestic and wild animals. The most important sand-fly-borne pathogens are Leishmania spp., the causative agents of leishmaniasis (reviewed in [35]). There are three main clinical forms of the disease: (i) visceral, which is almost always fatal without treatment; (ii) cutaneous, which causes skin ulcers; and (iii) mucocutaneous, which leads to partial/total destruction of the nose, mouth, and throat mucous membranes. Every year, there are about 700,000 to a million new cases [36]. Although Asaia sp. have been repeatedly found in sand flies [13,37], their effect on sand flies or transmitted pathogens is unknown.
As far as we know, this is the first study to examine in such detail the tripartite interactions between sand flies, Asaia sp., and Leishmania in vivo. Firstly, we infected females of Phlebotomus duboscqi, a vector of cutaneous leishmaniasis, with Asaia krungthepensis and Asaia siamensis, monitoring their survival and transmission to next generations of sand flies. Subsequently, we assessed the effect of these bacteria on the development of Leishmania major. In addition to using wild-type Asaia strains, we also investigated the effect of Asaia spp. expressing the Wolbachia surface protein (WSP) [28,38].
## Methods
## Sand flies
The laboratory colony of Phlebotomus duboscqi (originally from Senegal) was maintained at the Laboratory of Vector Biology at Charles University in Prague under standard conditions [39]. Briefly, sand flies were kept at a temperature of 26 °C, a 14/10 light/dark photoperiod, and had access to a 50% sucrose solution ad libitum.
## Leishmania
Leishmania major LV561 (MHOM/IL/67/LRC-L137 Jericho II) was cultured at 23 °C in M199 medium (Sigma-Aldrich) supplemented with 10% fetal bovine serum (Gibco), 1% BME vitamins (Sigma-Aldrich), 2% filtered human urine, and amikacin (250 μg/mL).
## Bacteria
Four Asaia strains were used. Wild strains of Asaia siamensis (7132 T ) and Asaia krungthepensis (7333 T ) [6,40] were provided by the Czech Collection of Microorganisms (Brno, Czech Republic). The genetically modified Asaia bacteria were kindly provided by Prof. Sara Epis from the University of Milan. The gene encoding the Wolbachia surface protein (WSP) was inserted into plasmid pHM4 and introduced into Asaia SF2.1, a wild-type strain derived from mosquitoes, resulting in the recombinant strain Asaia WSP . The control strain, Asaia pHM4, carried the empty plasmid [28]. Asaia spp. were cultivated on GLY medium (25 g/L glycerol [Lach-Ner s.r.o.], 10 g/L yeast extract, pH 5; Sigma-Aldrich). Agarose (20 g/L) was added when a solid medium was required to maintain colonies and detect bacteria during experiments. Kanamycin (100 µg/mL) was added to the medium to grow genetically modified bacteria carrying the plasmid.
## Bacteria preparation for sugar meal infection
Bacterial strains, regardless of species, were inoculated from agar plates into liquid GLY medium and incubated at 30 °C with shaking at 200 rpm until reaching an optical density corresponding to an infectious dose in the range of 1.8-4 × 10 8 CFU/mL. Afterwards, the cultures were centrifuged twice at 6500 rpm for 5 min and washed with physiological saline solution to stop further bacterial growth. Separately, a 20% solution of cane sugar and distilled water was boiled in a microwave oven three times. The blue food coloring (AROMA a.s.) was added to the cooled solution, which was filtered using a sterile membrane filter (0.22 µM, Millipore Millex). The resulting bacterial pellets were mixed with a 20% sucrose solution, rather than the 50% sucrose used for colony maintenance, to minimize osmotic pressure. The mixture of pellet and sugar was stained with a food dye, which allowed visual confirmation of successful feeding.
## Experimental infection of sand flies with Asaia spp.
In each experiment, 120-150 Ph. duboscqi females (1-3 days old) were first starved for 24 h before the experiment and then offered a mixture of sugar solution with Asaia bacteria for another 24 h. Unfed females were removed from the cage, and infected females were maintained in standard conditions with access to a 50% sucrose solution (noninfected). On the second day post sugar meal (psm), 15-30 females per group were dissected to confirm the ability of Asaia spp. to survive in the digestive tract of Ph. duboscqi (Fig. 1A). Females were anesthetized on ice, and their legs were removed. To minimize external contamination, the surface of each sand fly was washed with distilled water containing a small amount of mild detergent, rinsed twice in distilled water, immersed in 70% ethanol for 15 s, and finally washed again with physiological saline. Then, females were dissected in a drop of phosphate-buffered saline (PBS) under a binocular microscope using a clamp and dissecting tools; the head was removed, and the gut (midgut plus hindgut with Malpighian tubules) was pulled out from the abdomen. Each gut was rinsed in a fresh drop of physiological saline, transferred individually into a microcentrifuge tube containing 100 µL of sterile physiological saline, and then manually homogenized with a pestle. From each homogenate sample, 10 µL was inoculated onto agar, and the remaining volume was used for PCR testing to verify both the presence and viability of the bacteria. A sample was considered positive if Asaia DNA was detected by PCR and bacterial growth was observed on solid medium (Fig. 1A).
On day 6 psm, Ph. duboscqi females were fed for 60-120 min with heat-inactivated sheep blood (Lab-MediaServis) through a chicken skin membrane on a glass feeder as described by Volf and Volfova [39] (day 0 post blood meal, i.e., day 0 pbm). On days 2, 5, and 8 pbm, 10-30 females from each group were dissected and processed as described above. The experiments were repeated three times for wild-type strains and twice for genetically modified bacteria.
## Molecular detection of Asaia spp.
DNA was isolated from homogenized samples using the High Pure PCR Template Preparation Kit (Roche) according to the manufacturer's protocol. PCR reactions were performed in a total volume of 20 μL, containing 10 μL Emerald Amp GT PCR Master Mix Green (TaKaRa BIO INC.), 1 μL forward primer ASAFOR (5′-GCG CGT AGG CGG TTT ACA C-3′; Sigma-Aldrich), 1 μL reverse primer ASAREV (5′-AGC GTC AGT AAT GAG CCA GGTT-3′; Sigma-Aldrich), [11], and 2 μL template DNA and nuclease-free water to make up the final volume. The amplification protocol consisted of an initial denaturation at 95 °C for 2 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 60 s, and extension at 72 °C for 60 s, with a final extension step at 72 °C for 4 min. The resulting PCR product was analyzed by agarose gel electrophoresis. PCR products were loaded onto a 1% agarose gel prepared from agarose powder (BioReagent, Sigma-Aldrich), 50× TAE electrophoresis buffer (Thermo Fisher Scientific), and SYBR Safe DNA Gel Stain (Thermo Fisher Scientific), which allows DNA visualization. For each sample, 10 μL of PCR product was loaded into the wells of the gel. The first and last wells were loaded with 10 µL of GeneRuler 100 bp DNA Ladder (Thermo Fisher Scientific) as a size standard. Electrophoresis was performed at 90-120 V for 20-55 min, depending on the size of the gel. The gel was visualized and photographed under blue light using a VILBER imaging system and then analyzed.
## Asaia spp. detection by cultivation
In total, 10 µL of homogenate were inoculated onto an agar plate with GLY medium and spread with a sterile glass stick over the entire surface to obtain separate colonies. The plates were then incubated in a thermostat for 48 h at 30 °C.
## Transmission experiments of Asaia spp. in Ph. duboscqi
To investigate the possibility of transovarial transmission of Asaia spp. in Ph. duboscqi, three experimental groups of 150 females were established. The females were fed on sugar with (1) A. krungthepensis, (2) A. siamensis, or (3) without bacteria (control group). On day 6 psm the sand flies were fed on the blood (0 pbm) (Fig. 1B). On day 6 pbm, individual females were placed into glass vials containing filter paper moistened with distilled water, and the inlet was covered with monofilament. Vials with females were placed in a plastic box, which was also lined with moist filter paper, and maintained at 26 °C and 70% relative humidity. Viability of A. siamensis and A. krungthepensis in females was controlled as described above. As soon as the female died, its carcass and laid eggs were separately sampled, homogenized, and tested by PCR, and inoculated on agar plates. Again, positive detection by PCR must be accompanied by positive cultivation. The experiment was repeated twice.
## Vertical transmission of Asaia spp.
After laying eggs, dead females were removed, and the eggs were divided into four groups: (1) nonhomogenized, unwashed clutches; (2) homogenized, washed clutches;
(3) homogenized, unwashed clutches; and (4) nonhomogenized, washed clutches. The washing solution was prepared using a modified protocol by Poinar and Thomas [41]. Briefly, eggs were rinsed in distilled water, immersed in 500 µL of 70% ethanol for 5 min, and then in a 10% solution of sodium hypochlorite for the same period. After each chemical treatment, the eggs were rinsed in distilled water for 5 min to remove residual disinfectants. In addition, groups 2 and 3 were homogenized for 5 min using a homogenizer with an iron bead to assess whether Asaia bacteria were localized inside the eggs or merely adhered to their surface. The presence of Asaia in the homogenate was again evaluated by both PCR and agar plates. The experiments were repeated twice.
## Superinfection of Ph. duboscqi with Asaia spp. and Leishmania major
Groups of 120-150 Ph. duboscqi females (1-3 days old) were first infected with Asaia spp. using an infection dose of 1.8-4 × 10 8 CFU/mL via sugar feeding. On day 6 psm/0 pbm, Ph. duboscqi females were fed for 60-120 min with heat-inactivated sheep blood (LabMediaServis), seeded with 1 × 10 6 cells/mL promastigotes of Le. major, through a chicken skin membrane on a glass feeder. Non-bloodfed females were removed, and blood-fed ones were maintained at standard conditions. Between 15 and 30 Ph. duboscqi females were dissected on days 2, 5, and 8 post blood meal (pbm). Each female was washed as described above, and then the gut was dissected into a drop of PBS under a binocular microscope. The guts were examined microscopically for the presence, intensity, and localization of Leishmania infection. The intensity of infection was graded as (1) no infection, (2) weak (1-100 parasites/gut), ( 3) moderate (100-1000 parasites/gut), and (4) heavy (> 1000 parasites/gut) [42]. Three different localizations were distinguished, Leishmania either stayed in the abdominal midgut (AMG) or migrated anteriorly and also colonized the thoracic midgut (TMG) and stomodeal valve (SV). Subsequently, the dissected guts were rinsed from the microscope slide and coverslip using a pipette with 100 μL physiological saline solution, collected into a microtube, and homogenized. The homogenate was used for cultivation on agar plates and for PCR analysis to detect and confirm the presence of viable Asaia spp.
In addition, to evaluate Leishmania morphological forms present in the midgut, smears were made from Le. major positive guts using a coverslip. After drying, the sample was fixed with methanol and stained with Giemsa (Sigma-Aldrich).
Promastigotes of Le. major were classified as metacyclic forms when flagellum length was ≥ 2 times body length, leptomonad forms when flagellum length was < 2 times body length and body length was < 14 μm, and elongated nectomonads when flagellum length was < 2 times body length and body length was ≥ 14 μm according to [43]. The experiments were repeated three times for wild-type strains and twice for genetically modified bacteria (Fig. 1C).
## Statistical evaluation
The data were tested in StudioR (http:// cran.r-proje ct. org) [44]; owing to the type of data, Fisher's exact or Chi-squared tests were used to assess the statistical significance.
## Results
## Colonization of Ph. duboscqi with Asaia spp.
All four Asaia strains tested (A. siamensis, A. krungthepensis, Asaia WSP , and Asaia pHM4) successfully colonized the gut of Ph. duboscqi following both the sugar meal (psm) and the subsequent blood meal (pbm) throughout the experimental period (Fig. 2). The lowest colonization was observed for A. siamensis, with infection rates of 66.7%, 56.7%, 62.5%, and 40% on day 2 psm and days 2, 5, and 8 pbm, respectively. Asaia krungthepensis showed a high initial infection rate on day 2 psm (93.3%), which dropped to 60% on day 2 pbm but increased again to nearly 80% on days 5 and 8 pbm, following blood defecation. A similar trend was observed for Asaia pHM4, though with slightly lower infection rates ranging from 71% to 85%. In contrast, Asaia WSP maintained relatively stable and high infection rates, varying between 77% and 90% across all tested time points (Fig. 2).
## Vertical transmission of wild-type Asaia spp.
Out of 128 females infected with A. siamensis, 78 were positive and laid eggs, yielding 28 positive clutches. The highest positivity (80%) was detected in egg clutches laid on day 9 pbm (Fisher's exact test, P = 0.001). Out of 128 females infected with A. krungthepensis, 69 were positive and laid eggs, and 41 of the clutches were infected. The positivity of egg clutches (Table 1) did not differ significantly across various days (Fisher's exact test, P = 0.824) but was significantly higher compared with A. siamensis (χ 2 = 7.2168, df = 1, P = 0.007).
In both Asaia species the vertical transmission was through egg-smearing. Regardless of homogenization, all test groups without egg washing were PCR positive, while washed eggs were all negative.
## Leishmania major infection in Ph. duboscqi infected by wild-type Asaia spp.
On day 2 pbm, i.e., before defecation, Le. major promastigotes were present in the endoperitrophic space, and the percentage of infected females ranged between 70% and 80% across all three experimental groups (Fig. 3A). Both the infection rates and parasite loads (i.e., infection intensities) did not significantly differ between females in the control group (without Asaia) and those colonized by A. siamensis (χ 2 = 0.426, df = 1, P = 0.514; Fisher's exact test, P = 0.729), nor between the control and the group colonized by A. krungthepensis (χ 2 = 0.809, df = 1, P = 0.369). Similarly, on day 5 pbm (i.e., after defecation), no significant differences were observed among the groups in either the percentage of Leishmania-infected females or parasite loads (A. siamensis: χ 2 = 0, df = 1, P = 1; A. krungthepensis: χ 2 = 0.026, df = 1, P = 0.873; Fig. 3A) or intensity of infection (Fig. 3A) (A. siamensis: Fisher's exact test, P = 0.297; A. krungthepensis: Fisher's exact test, P = 0.318). In late-stage infections (day 8 pbm), colonization by A. siamensis significantly reduced the intensity of Le. major infection (Fisher's exact test, P = 0.002), but the percentage of infected females did not differ significantly (χ 2 = 0.256, df = 1, P = 0.613). A similar effect was observed with A. krungthepensis, which also significantly reduced the incidence of heavy infections (Fisher's exact test, P = 0.038), with no difference in percentage of infected females (χ 2 = 0.014, df = 1, P = 0.905; Fig. 3A).
Regarding infection localization (Fig. 3B), neither Asaia strain had a significant effect on any of the days tested. For A. siamensis, no differences were found on day 5 pbm (Fisher's exact test, P = 0.538) or day 8 pbm (Fisher's exact test, P = 0.410). Similarly, for A. krungthepensis, no significant effect was observed on day 5 pbm (Fisher's exact test, P = 0.705) or day 8 pbm (Fisher's exact test, P = 0.839; Fig. 3B).
As reported in Fig. 4A, three morphological forms of Leishmania (long nectomonads, leptomonads, and metacyclics) were identified on Giemsa-stained smears from Ph. duboscqi midguts colonized by Asaia. On day 5 pbm, colonization with A. siamensis and A. krungthepensis resulted in a significantly increased proportion of leptomonads at the expense of long nectomonads in both groups (Fisher's exact test, P = 0.000). The proportion of metacyclic promastigotes was significantly higher in the group infected with A. krungthepensis (9%) compared with the Asaia-negative control group (3%). On day 8 pbm, females colonized with A. krungthepensis maintained the elevated proportion of leptomonads (Fisher's exact test, P = 0.000), whereas no significant differences in the leptomonad-to-nectomonad ratio were observed between the A. siamensis group (Fisher's exact test, P = 0.277) and the Asaia-negative control group (Fig. 4A). The proportion of metacyclic forms remained significantly higher in the A. krungthepensis group 17% compared with the control 4% (Fisher's exact test, P = 0.011).
## Leishmania major infection in Ph. duboscqi infected by genetically modified Asaia
On day 2 pbm, the Leishmania infection rates in sand fly groups superinfected with Asaia WSP and Asaia pHM4 were 84% and 77%, respectively comparable to the infection rate observed in the control group without Asaia at 87% (Fisher's exact test, P = 0.834; Fig. 5A). On days 5 and 8 pbm, Leishmania infection rates in females superinfected with Asaia WSP and Asaia pHM4 fluctuated between 57 and 90%; however, no significant differences were detected compared to the Asaia-negative control (5 pbm: Fisher's exact test, P = 0.1657; 8 pbm: P = 0.5731; Fig. 5A).
Asaia WSP did not significantly affect the intensity of Le. major infection in any of the tested time points when compared either with the control group without bacteria (Fisher's exact test, day 2 pbm: P = 0.131, day 5 pbm: P = 0.796, and day 8 pbm: P = 0.599) or with the group infected with Asaia pHM4 (Fisher's exact test, day 2 pbm: P = 0.904, day 5 pbm: P = 0.078, and day 8 pbm: P = 0.565).
Similarly, there were no significant differences in Le. major localization among Ph. duboscqi sand flies infected with Asaia WSP and Asaia pHM4 (Fisher's exact test, day 5 pbm: P = 0.222 and day 8 pbm: P = 0.280), or in the control group without Asaia (Fisher's exact test, day 5 pbm: P = 0.828 and day 8 pbm: P = 0.354; Fig. 5B). On day 5 pbm, sand flies infected with Asaia WSP showed a higher proportion of leptomonad and fewer nectomonad forms compared with those infected with Asaia pHM4 (Fisher's exact test, P = 0.005; Fig. 4A), in contrast to the comparison between Asaia WSP and the group without Asaia (Fisher's exact test, P = 0.409). On day 8 pbm, no significant differences were observed between the Asaia WSP group and the control group without Asaia (Fisher's exact test, P = 0.912), nor between Asaia WSP and Asaia pHM4 (Fisher's exact test, P = 0.780; Fig. 4B).
The proportions of metacyclic forms of Le. major in Asaia WSP -infected sand flies were similar to those in the Asaia-negative control group on day 5 (Fisher's exact test, P = 0.721) and day 8 pbm (Fisher's exact test, P = 0.446). Likewise, no significant differences were observed in the proportion of metacyclic promastigotes between Asaia WSP and Asaia pHM4 groups on day 5 (Fisher's exact test, P = 0.154) or day 8 pbm (Fisher's exact test, P = 0.458).
## Discussion
In our experiments, all four Asaia species and strains tested were able to colonize the midgut of Ph. duboscqi and persisted throughout the observation period, after infection, and following blood feeding and subsequent defecations. However, their infection and survival dynamics differed over time.
The wild-type species, A. siamensis and A. krungthepensis showed similar colonization patterns, a decrease in infection rate following female blood feeding, a slight increase after defecation, and a subsequent decrease. A decline in microbial richness (i.e., the number of distinct operational taxonomic units) after blood feeding in sand flies was described by Kelly et al. [45], nevertheless, they showed that the richness was restored after defecation to a level seen with the sucrose-fed controls. At first glance, this seems inconsistent with our results, as the infection rate of Asaia continued to decline after defecation. However, while Kelly et al. [45] focused on overall species richness, our data specifically concern Asaia. There is a limited information on the behavior of Asaia in blood-feeding insects. Egyirifa and Akorli [46] described a decrease of A. siamensis in blood-fed Anopheles gambiae, and similar trends were observed in females of Aedes aegypti, Aedes albopictus, Mansonia humeralis, and Asaia sp. [47,48].
Blood feeding alters the gut environment, inducing a shift in the microbiome abundance. Bacteria that thrive within the blood meal possess large genetic redox capacity to cope with oxidative and nitrosative stress associated with blood meal digestion [49]. It seems that wild-type species of Asaia are not among these species and may prefer non-blood-fed guts or other tissues. In our study, both A. siamensis and A. krungthepensis also infected the sand fly reproductive system, facilitating vertical transmission via contamination of the egg surface. Then the bacteria are ingested by the hatching larvae, a mechanism previously suggested for Asaia in mosquitoes [12] and leafhoppers [8].
In sand flies, similar vertical transmission was proven for Psychodiella gregarines [50]; however, to the best of our knowledge, this is the first study to rule out Asaia penetration through the egg membrane and confirm transmission via egg surface contamination. It has been shown that native Asaia strains can activate mosquito immunity, potentially affecting pathogen development [34,51]. Asaia introduction triggered mosquito immune responses that reduced Plasmodium berghei development in Anopheles stephensi [34], moreover, precolonization with Asaia sp. significantly reduced Leishmania mexicana populations in Lutzomyia longipalpis [16]. In our study, we tested the effect of A. siamensis and A. krungthepensis on the development of Le. major, the causative agent of zoonotic cutaneous leishmaniasis, in Ph. duboscqi. Both bacterial species significantly reduced parasite load during the late phase of infection. Notably, A. krungthepensis also altered the composition of the Le. major population, particularly increasing the proportion of metacyclic promastigotes. Leishmania development in sand flies proceeds through a series of promastigote morphotypes, among which, metacyclic promastigotes and possibly haptomonads are the main forms transmitted to mammalian hosts during blood feeding [52]. For transmission to mammals, both infection rate and infection intensity are important factors to consider. The presence of Asaia seems to modify this development, when both species decrease the intensity of infection in the late phase; nevertheless, A. krungthepensis stimulates metacyclogenesis.
Whether this effect is due to immune changes, production of anti-leishmanial molecules, microbiome alteration, or a combination of these factors remains to be determined and warrants further research. Nevertheless, the effect of Asaia spp. is less pronounced than that of the Delftia tsuruhatensis TC1 strain, which strongly inhibited Le. major development in Ph. duboscqi; importantly, Leishmania-infected sand flies fed with D. tsuruhatensis were significantly less able to transmit Le. major parasites and cause disease in mice [53].
Regarding the engineered strains, Asaia pHM4 and Asaia WSP , we showed that they retain the ability to colonize sand flies. Interestingly, especially in the case of Asaia WSP , abundance increased after the blood meal, contrasting with wild strains. Epis and collaborators [28] observed a similar pattern in Aedes aegypti mosquitoes. Despite Asaia WSP -induced activation of the host immune response in mosquitoes inhibiting Dirofilaria immitis in Ae. aegypti [28], in the present study, we found no impact on Le. major infection parameters in Ph. duboscqi or Asaia pHM4, and neither did Asaia WSP affect the infection and development of Le. major in Ph. duboscqi.
Wolbachia has been repeatedly shown to activate the immune system in mosquitoes and negatively affect pathogen transmission [54][55][56][57]. However, its effect on sand flies remains unclear. Wolbachia have been repeatedly detected in sand flies [37,[58][59][60], but data on its impact on vector competence are still lacking.
Nevertheless, Rosário and collaborators [61] detected Wolbachia and Leishmania infantum coinfections in 37% of Nyssomyia whitmani collected in the field, suggesting that co-existence in the same vector is possible. Similarly, an in vitro study showed that Wolbachia presence in sand fly cells did not significantly impact Le. infantum infection [62]. However, it should be noted that in our experimental model using Asaia WSP , the protein came from the filarial nematode Dirofilaria immitis [28], rather than an insect-associated strain. This exogenous WSP was intentionally used to induce a stronger immune response, but it may not accurately reflect the effects of naturally occurring Wolbachia infections in sand flies.
## Conclusions
We provided the first evidence that wild-type Asaia bacteria (A. siamensis and A. krungthepensis) can infect Ph. duboscqi and can be vertically transmitted via egg smearing. These bacteria reduce the late-phase infection of Le. major, which may have important epidemiological consequences and can be used for future vector control strategies. Unexpectedly, the Asaia strain expressing Wolbachia surface protein has no measurable effect on Le. major infection in this sand fly species.
## References
1. Shi, Yu, Cheng (2023) "Impact of the microbiome on mosquito-borne diseases" *Protein Cell*
2. Motta, Moran (2024) "The honeybee microbiota and its impact on health and disease" *Nat Rev Microbiol*
3. Rajendran, Vinayagam, Sekar et al. (2024) "Symbiotic bacteria: Wolbachia, midgut microbiota in mosquitoes and their importance for vector prevention strategies" *Microb Ecol*
4. Lewis, Lizé (2015) "Insect behaviour and the microbiome" *Curr Opin Insect Sci*
5. Yamada, Katsura, Kawasaki et al. (2000) "Asaia bogorensis gen nov., sp. nov, an unusual acetic acid bacterium in the alpha-Proteobacteria" *Int J Syst Evol Microbiol*
6. Katsura, Kawasaki, Potacharoen et al. (2001) "Asaia siamensis sp. nov., an acetic acid bacterium in the alpha-proteobacteria" *Int J Syst Evol Microbiol*
7. Bassene, Niang, Fenollar et al. (2020) "Role of plants in the transmission of Asaia sp., which potentially inhibit the Plasmodium sporogenic cycle in Anopheles mosquitoes" *Sci Rep*
8. Crotti, Damiani, Pajoro et al. (2009) "Asaia, a versatile acetic acid bacterial symbiont, capable of cross-colonizing insects of phylogenetically distant genera and orders" *Environ Microbiol*
9. Li, Hua, Ali et al. (2019) "Characterization of a bacterial symbiont Asaia sp. in the White-Backed planthopper, Sogatella furcifera, and its effects on host fitness" *Front Microbiol*
10. Woruba, Morrow, Reynolds et al. (2019) "Diet and irradiation effects on the bacterial community composition and structure in the gut of domesticated teneral and mature Queensland fruit fly, Bactrocera tryoni (Diptera: Tephritidae)" *BMC Microbiol*
11. Favia, Ricci, Damiani et al. (2007) "Bacteria of the genus Asaia stably associate with Anopheles stephensi, an Asian malarial mosquito vector" *Proc Natl Acad Sci U S A*
12. Damiani, Ricci, Crotti et al. (2010) "Mosquitobacteria symbiosis: the case of Anopheles gambiae and Asaia" *Microb Ecol*
13. Akhoundi, Bakhtiari, Guillard et al. (2012) "Diversity of the bacterial and fungal microflora from the midgut and cuticle of Phlebotomine sand flies collected in north-western Iran" *PLoS ONE*
14. Epis, Gaibani, Ulissi et al. (2012) "Do mosquitoassociated bacteria of the genus Asaia circulate in humans?" *Eur J Clin Microbiol Infect Dis*
15. De Freece, Damiani, Valzano et al. (2014) "Detection and isolation of the α-proteobacterium Asaia in Culex mosquitoes" *Med Vet Entomol*
16. Santanna, Diaz-Albiter, Aguiar-Martins et al. (2014) "Colonisation resistance in the sand fly gut Leishmania protects Lutzomyia longipalpis from bacterial infection" *Parasit Vectors*
17. Tarabai, Floriano, Zima et al. (2023) "Microbiomes of blood-feeding triatomines in the context of their predatory relatives and the environment" *Microbiol Spectr*
18. Xiao, Yao, Guan et al. (2024) "Viromes of Haemaphysalis longicornis reveal different viral abundance and diversity in free and engorged ticks" *Virol Sin*
19. Snyder, Ruhe, Kobrin et al. (2004) "Asaia bogorensis peritonitis identified by 16S ribosomal RNA sequence analysis in a patient receiving peritoneal dialysis" *Am J Kidney Dis*
20. Tuuminen, Heinäsmäki, Kerttula (2006) "First report of bacteremia by Asaia bogorensis, in a patient with a history of intravenous-drug abuse" *J Clin Microbiol*
21. Abdel-Haq, Savaşan, Davis et al. (2009) "Asaia lannaensis bloodstream infection in a child with cancer and bone marrow transplantation" *J Med Microbiol*
22. Alauzet, Teyssier, Jumas-Bilak et al. (2010) "Gluconobacter as well as Asaia species, newly emerging opportunistic human pathogens among acetic acid bacteria" *J Clin Microbiol*
23. Juretschko, Beavers-May, Stovall (2010) "Nosocomial infection with Asaia lannensis in two paediatric patients with idiopathic dilated cardiomyopathy" *J Med Microbiol*
24. Carretto, Visiello, Bardaro et al. (2016) "Asaia lannensis bacteremia in a 'needle freak' patient" *Future Microbiol*
25. Mohammad, Hashim (2024) "A rare case of persistent bacteremia caused by Asaia spp. in an infant" *Cureus*
26. Shane, Grogan, Cwalina et al. (2018) "Blood meal-induced inhibition of vector-borne disease by transgenic microbiota" *Nat Commun*
27. Asgari, Ilbeigikhamsehnejad, Rismani et al. (2020) "Molecular characterization of RNase III protein of Asaia sp. for developing a robust RNAi-based paratransgensis tool to affect the sexual life-cycle of Plasmodium or Anopheles fitness" *Parasit Vectors*
28. Epis, Varotto-Boccazzi, Crotti et al. (2020) "Chimeric symbionts expressing a Wolbachia protein stimulate mosquito immunity and inhibit filarial parasite development" *Commun Biol*
29. Damiani, Ricci, Crotti et al. (2008) "Paternal transmission of symbiotic bacteria in malaria vectors" *Curr Biol*
30. Villegas, Radl, Dimopoulos et al. (2023) "Bacterial communities of Aedes aegypti mosquitoes differ between crop and midgut tissues" *PLoS Negl Trop Dis*
31. Chouaia, Rossi, Epis et al. (2012) "Delayed larval development in Anopheles mosquitoes deprived of Asaia bacterial symbionts" *BMC Microbiol*
32. Mitraka, Stathopoulos, Siden-Kiamos et al. (2013) "Asaia accelerates larval development of Anopheles gambiae" *Pathog Glob Health*
33. Mancini, Damiani, Short et al. (2020) "Inhibition of Asaia in adult mosquitoes causes male-specific mortality and diverse transcriptome changes" *Pathogens*
34. Cappelli, Damiani, Mancini et al. (2019) "Asaia activates immune genes in mosquito eliciting an anti-Plasmodium response: implications in malaria control" *Front Genet*
35. Maroli, Feliciangeli, Bichaud et al. (2013) "Phlebotomine sandflies and the spreading of leishmaniases and other diseases of public health concern" *Med Vet Entomol*
36. Karimian, Koosha, Choubdar et al. (2022) "Comparative analysis of the gut microbiota of sand fly vectors of zoonotic visceral leishmaniasis (ZVL) in Iran; host-environment interplay shapes diversity" *PLoS Negl Trop Dis*
37. Varotto-Boccazzi, Epis, Arnoldi et al. (2020) "Boosting immunity to treat parasitic infections: Asaia bacteria expressing a protein from Wolbachia determine M1 macrophage activation and killing of Leishmania protozoans" *Pharmacol Res*
38. Volf, Volfova (2011) "Establishment and maintenance of sand fly colonies" *J Vector Ecol*
39. Yukphan, Potacharoen, Tanasupawat et al. (2004) "Asaia krungthepensis sp. nov., an acetic acid bacterium in the α-Proteobacteria" *Int J Syst Evol Microbiol*
40. Poinar, Thomas (1984) "Laboratory guide to insect pathogens and parasites"
41. Myskova, Votypka, Volf (2008) "Leishmania in sand flies: comparison of quantitative polymerase chain reaction with other techniques to determine the intensity of infection" *J Med Entomol*
42. Sádlová, Price, Smith et al. (2010) "The stage-regulated HASPB and SHERP proteins are essential for differentiation of the protozoan parasite Leishmania major in its sand fly vector, Phlebotomus papatasi" *Cell Microbiol*
43. Core (2018) "R: A language and environment for statistical computing R Foundation for Statistical Computing"
44. Kelly, Bahr, Serafim et al. (2017) "The gut microbiome of the vector Lutzomyia longipalpis is essential for survival of Leishmania infantum" *MBio*
45. Egyirifa, Akorli (2024) "Two promising candidates for paratransgenesis, Elizabethkingia and Asaia, increase in both sexes of Anopheles gambiae mosquitoes after feeding" *Malar J*
46. Díaz, Camargo, Avila (2021) "Characterization of the reproductive tract bacterial microbiota of virgin, mated, and blood-fed Aedes aegypti and Aedes albopictus females" *Parasit Vectors*
47. Da Fonseca Meireles, Mdeo, Montenegro et al. (2025) "Do the microbiota of larval breeding site and the blood meal influence the composition and diversity of bacterial communities in the midgut of Mansonia humeralis (Diptera: Culicidae) from the western Amazon?" *Braz J Microbiol*
48. Wang, Iii, Kukutla et al. (2011) "Dynamic gut microbiome across life history of the malaria mosquito Anopheles gambiae in Kenya" *PLoS ONE*
49. Lantova, Volf (2012) "The development of Psychodiella sergenti (Apicomplexa: Eugregarinorida) in Phlebotomus sergenti (Diptera: Psychodidae)" *Parasitol*
50. Capone, Ricci, Damiani et al. (2013) "Interactions between Asaia, Plasmodium and Anopheles: new insights into mosquito symbiosis and implications in malaria symbiotic control" *Parasit Vectors*
51. Catta-Preta, Ghosh, Sacks et al. (2024) "Single-cell atlas of Leishmania development in sandflies reveals the heterogeneity of transmitted parasites and their role in infection" *Proc Natl Acad Sci*
52. Cecilio, Rogerio, Serafim et al. (2025) "Leishmania sand fly-transmission is disrupted by Delftia tsuruhatensis TC1 bacteria" *Nat Commun*
53. Moreira, Iturbe-Ormaetxe, Jeffery et al. (2009) "A Wolbachia symbiont in Aedes aegypti limits infection with dengue, Chikungunya, and Plasmodium" *Cell*
54. Walker, Johnson, Moreira et al. (2011) "The wMel Wolbachia strain blocks dengue and invades caged Aedes aegypti populations" *Nature*
55. Aliota, Walker, Yepes et al. (2016) "The wMel strain of Wolbachia reduces transmission of Chikungunya virus in Aedes aegypti" *PLoS Negl Trop Dis*
56. Vandana, Dong, Sheth et al. (2024) "Wolbachia infection-responsive immune genes suppress Plasmodium falciparum infection in Anopheles stephensi" *PLoS Pathog*
57. Bordbar, Soleimani, Fardid et al. (2014) "Three strains of Wolbachia pipientis and high rates of infection in Iranian sandfly species" *Bull Entomol Res*
58. Lozano-Sardaneta, Marina, Torres-Monzón et al. (2023) "Molecular detection of Wolbachia and Bartonella as part of the microbiome of phlebotomine sand flies from Chiapas" *Parasitol Res*
59. Torres-Llamas, Díaz-Sáez, Morales-Yuste et al. (2025) "Assessing Wolbachia circulation in wild populations of phlebotomine sand flies from Spain and Morocco: implications for control of leishmaniasis" *Parasit Vectors*
60. Aado, Dias-Lima, Lambert et al. (2022) "Identification and molecular characterization of Wolbachia strains and natural infection for Leishmania sp. in neotropical Phlebotominae (Diptera: Psychodidae) species, leishmaniasis vectors" *Acta Trop*
61. Da Silva Gonçalves, Iturbe-Ormaetxe, Martins-Da-Silva et al. (2019) "Wolbachia introduction into Lutzomyia longipalpis (Diptera: Psychodidae) cell lines and its effects on immunerelated gene expression and interaction with Leishmania infantum" *Parasit Vectors* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12305883&blobtype=pdf | # A bivalent Mpox nanoparticle vaccine induces robust immune response and provides long-lasting protection against vaccinia virus challenge
Yanhai Wang, Hao Sun, Huixu Zhou, Wenxue Yang, Sha Li, Mengchan Hao, Haiyan Chang, Yuan Zhang, Yun Wang, Jianjun Chen
## Abstract
The 2022 and 2024 monkeypox (mpox) outbreak highlighted the urgent need for effective, durable, and safe vaccines. In addition to the traditional smallpox vaccines that could bring cross-protection against mpox, mRNA and protein-subunit mpox vaccines were extensively studied after the outbreak of mpox. In this study, we engineered monkeypox virus (MPXV) nanoparticle vaccines by conjugating the M1R and A35R, two well-characterized protective antigens to the mi3 nanoparticle using the SpyTag-SpyCatcher system, generating mi3-M1R and mi3-A35R constructs. An equimolar mixture of mi3-M1R and mi3-A35R formed a bivalent MPXV vaccine candidate, termed mi3-AM. When administered intraperitoneally with the Mn adjuvant, the mi3-AM vaccine induced robust humoral and antigen-specific cellular immune responses. Notably, the mi3-AM vaccine provided long-lasting protection against a lethal challenge with vaccinia virus Western Reserve strain (VACV-WR) in mice. With ongoing mpox outbreaks and the limitations of current vaccines, our candidate represents a promising, deployable solution with potential to bridge existing gaps in global orthopoxvirus prevention.
## Introduction
The mpox outbreak in 2022 and remerge in 2024 raised significant global concern. This zoonotic disease, caused by the MPXV, typically manifests in humans with symptoms including fever, skin rash, and lymphadenopathy [1,2], probably lead to severe clinical symptoms and even death. Since the initial identification of human cases in the 1970s, MPXV infections have predominantly occurred in Central and West African countries [3,4]. The virus was first appeared outside Africa in 2003, with sporadic cases subsequently reported in various locations beyond Africa [3]. However, since May 2022, MPXV infections have surged outside Africa, prompting the World Health Organization (WHO) to declare mpox a Public Health Emergency of International Concern (PHEIC) [5]. By April 30, 2025, over 142,233 cases had been reported across 133 countries, resulting in more than 328 deaths globally [6].
MPXV is an enveloped double-stranded DNA virus classified within the Orthopoxvirus genus of the Poxviridae family, which also includes the variola virus (VARV), vaccinia virus (VACV), rabbitpox virus (RPXV), and cowpox virus (CPXV) [7]. Most genomes of orthopoxviruses are highly conserved, and immunity or infection with one virus often provides cross-protection against other members in the same genus [8][9][10][11][12]. Although no specific antiviral treatments or vaccines currently available for MPXV, studies have demonstrated that the smallpox vaccine effectively prevents infections from related orthopoxviruses, including MPXV [9,11], VACV [10], and RPXV [12]. As a result, two live-attenuated VACVbased smallpox vaccines -ACAM2000 and JYNNEOS -have been approved by the U.S. Food and Drug Administration (FDA) for MPXV prevention during the ongoing outbreak [13,14]. However, challenges such as suboptimal neutralizing antibody levels and significant adverse effects [15][16][17][18][19] have limited the global effectiveness of these vaccines. Therefore, there is an urgent need to develop an effective, safe, and accessible MPXV-specific vaccine to address these limitations.
Like other orthopoxviruses, MPXV exists in two distinct infectious forms: mature virions (MV) and enveloped virions (EV). Both forms are capable of causing disease [4], but they exhibit distinct surface antigens. For instance, MV-specific antigens include M1R, H3L, E8L, and A29L, whereas EV-specific antigens include B6R and A35R. These surface proteins play key roles in viral infection and immune response induction. M1R, B6R, and A35R have been identified as critical for eliciting neutralizing antibodies [20][21][22], while H3L and E8L are involved in cellular immune responses [23] and virion attachment [23,24], respectively. After the 2022 mpox outbreak, vaccines against MPXV were rapidly developed, primarily including mRNA vaccines and protein subunit vaccines. Given the diverse functions of MPXV antigens, current MPXV vaccine platforms universally employ a multi-antigen co-immunization strategy combining at least one MV-specific and one EVspecific antigen [20,21,[25][26][27]. While the multi-antigen co-immunization strategy offers theoretical advantages, it presents significant challenges in practice. First, this approach lacks precise control over the utilization efficiency of individual antigens, often resulting in imbalanced and unsustainable immune responses across different immunogens. Moreover, from a manufacturing perspective, multi-antigen formulations substantially increase production complexity and cost, creating significant barriers to industrial-scale vaccine production. These economic and technical constraints may ultimately limit the translational potential of such vaccine candidates. Selection of the minimal protective antigen combination can enhance antigen utilization efficiency, improve vaccine stability, and simplify manufacturing processes for industrial-scale production. Through comprehensive screening and immunological evaluation, we have identified an optimized minimal antigen combination consisting of MV-specific antigen M1R and EV-specific antigen A35R. MV-specific antigen M1R and EV-specific antigen A35R have been shown to elicit potent neutralizing antibodies and provide protection against MPXV infection in animal models [25,27].
Self-assembled nanoparticles are ideal vaccine carriers due to their polymer structure, which is particularly well-suited for presenting multiple antigens while enhancing antigen stability and delivery efficiency. Therefore, utilizing them as carriers for MPXV antigens is a promising strategy. Specifically, we employed the mi3 nanoparticle, derived from 2-keto-3-deoxyphosphogluconate (KDPG) aldolase, which can selfassemble into a dodecameric cage scaffold comprising 60 subunits [28,29]. This structure offers an ideal platform for vaccine antigens presentation. Conjugation of antigens to nanoparticle can be accomplished through the SpyTag-SpyCatcher protein covalent bonding strategy [28,30]. This strategy can achieve rapid expression and purification of vaccine antigens and nanoparticles, significantly shortening the time required for development and manufacturing of a new vaccine [31].
In this study, we developed a bivalent mpox nanoparticle vaccine, designated mi3-AM. Specifically, MV-specific antigen M1R and EV-specific antigen A35R were covalently conjugated to the surface of mi3 nanoparticles using the SpyTag-SpyCatcher strategy, yielding monovalent nanoparticle vaccines mi3-M1R and mi3-A35R, respectively. These two nanoparticle formulations were then combined at an equimolar ratio to generate the bivalent vaccine mi3-AM. For comparative evaluation, a control bivalent subunit vaccine (AM) was prepared by mixing M1R and A35R. Immunization studies in BALB/c mice demonstrated that the mi3-AM nanoparticle vaccine elicited significantly enhanced immune responses compared to the monomeric AM vaccine. Both vaccines provided complete protection against a lethal challenge with VACV-WR. Notably, the mi3-AM nanoparticle vaccine also exhibited superior durability of protection compared to the monomeric AM vaccine. Overall, our results highlight the potential of the mi3-AM nanoparticle vaccine as a promising candidate for controlling the global spread of MPXV.
## Materials and methods
## Cells and viruses
Vero-E6 cells were cultured in Dulbecco's Modified Eagle medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco) and penicillin (100 U/ml)-streptomycin (100 mg/ml) (P/S, Thermo Fisher) at 37°C in a humidified 5% CO 2 atmosphere.
The VACV-WR and MPXV were propagated in Vero-E6 cells using DMEM supplemented with 2% FBS and 1% P/S at 37°C in a humidified 5% CO 2 atmosphere.
## Gene synthesis and plasmid construction
The amino acid sequences of MPXV (A35R, M1R) were fused with SpyTag (AHIVMVDAYKPTK) and a hexahistidine tag at the C-terminus. The Kozak consensus sequence (GCCACC) and a signal peptide were incorporated at the N-terminus of the MPXV sequences. All sequences were codon-optimized for human cells and cloned into the eukaryotic expression vector pCAGGS between the EcoRI and XhoI sites.
The amino acid sequences of mi3, fused with Spy-Catcher at the N-terminus and a hexahistidine tag at the C-terminus (GenBank: MH425515), were codonoptimized for E. coli and cloned into the prokaryotic expression vector pET28a + between the NotI and XhoI sites.
All genes were synthesized by Sangon Biotech (Shanghai).
## Protein expression and purification
The A35R-SpyTag and M1R-SpyTag proteins were produced in HEK293F cells. Briefly, the HEK293F cells were cultured in Union 293 medium at 37 °C, 80-90% humidity, and 5% CO 2 , with rotation at 120 rpm for expansion. When the cells reached a density of 1.0 × 10 6 cells/ml, they were transiently transfected with 1 mg of expression plasmid per litre using polyetherimide (PEI). Five days later, the supernatants were collected and centrifuged at 8000 g and 4°C for 1 h to remove cellular debris. The supernatant was then filtered through a 0.22 μm membrane. The cleared supernatants were passed through Ni-NTA agarose to enrich His-tagged target proteins, followed by elution with an imidazole-containing HEPES buffer (50 mM HEPES, pH 8.0, 300 mM imidazole, and 300 mM NaCl). To obtain purer proteins, size-exclusion chromatography (SEC) was performed using Superose 6 Increase 10/300 GL gel filtration column (GE Healthcare) with a buffer comprising 50 mM HEPES, pH 8.0, and 300 mM NaCl. Fractions of the target peak were pooled, concentrated using a centrifugal cartridge (10 kDa cutoff, Millipore), and stored at 4°C for future use.
The SpyCatcher-mi3 nanoparticle proteins were expressed in E. coli. Briefly, pET28a + expression plasmids encoding SpyCatcher-mi3 were transformed into BL21 (DE3) competent cell (Vazyme). After overnight incubation at 37°C on LB agar plates, a single clone was amplified in LB containing kanamycin at 37°C with shaking. Bacteria solutions were supplemented with 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) to induce protein expression when the OD600 value reached 0.6-0.8, followed by induction at 20°C for 20 h with shaking at 150 rpm. Subsequently, bacterial precipitates were collected by centrifugation and lysed by sonication using a lysis buffer consisting of 50 mM HEPES, pH 8.0, 300 mM NaCl, 30 mM imidazole, 50 μg/mL deoxyribonuclease (DNase), 0.75% 3-((3-cholamidopropyl) dimethylammonio)-1-propanesulfonate (CHAPS) and 1 mM phenyl methane sulphonyl fluoride (PMSF). After lysis, the cell debris was removed by centrifugation at 8,000 g for 1 h, and the His-tagged target proteins in the supernatants were purified using Ni-NTA agarose and SEC. Fractions of the target peak were pooled, concentrated using a centrifugal cartridge (100 kDa cutoff, Millipore), and stored at 4 °C for future use.
## Preparation of nanoparticle vaccines
A covalent peptide bond was formed between SpyTag and SpyCatcher. To display the MPXV proteins on the surface of the nanoparticle (SpyCatcher-mi3), the Spy-Catcher-mi3 nanoparticle proteins were incubated with a 1:4 molar excess of the MPXV proteins (A35R-SpyTag, M1R-SpyTag) overnight at 4°C. After incubation, these nanoparticle vaccines (mi3-A35R, mi3-M1R) were purified by SEC using phosphate-buffered saline (PBS) as the buffer, and excess MPXV proteins were removed. An equimolar mixture of mi3-A35R and mi3-M1R was prepared to form the bivalence nanoparticle vaccine (mi3-AM).
## Dynamic light scattering (DLS)
DLS was performed using a Zetasizer Nano ZS ZEN3600 (Malvern Instruments Ltd.) to characterize the diameters of the nanoparticles. The purified proteins were centrifuged at 16,250 g for 10 min at 4°C, and 100 μL of the supernatant was loaded into a quartz cuvette to measure the nanoparticle diameters. Each sample was analyzed in triplicate.
## Transmission electron microscope (TEM)
10 μL of purified nanoparticles were adsorbed onto freshly glow-discharged carbon-coated grids for 1 min. Excess liquid was removed by blotting with filter paper. The grids were stained with 2.0% phosphotungstic acid (PTA, pH 7.0) for 5 min and then dried at room temperature. The grids were imaged using a Talos L120C TEM operating at 120 kV.
## Immunization of mice
Female BALB/c mice aged 6-8 weeks were randomly assigned to three groups for the animal experiments. Each group received two intraperitoneal injections, with a three-week interval. The immunization dose comprised 10 μg of the bivalent nanoparticle vaccine mi3-AM (mi3-A35R 4.73 μg, mi3-M1R 5.27 μg), and the bivalent MPXV protein vaccine AM (A35R 1.54 μg, M1R 2.08 μg) served as controls. The mi3 nanoparticles without attached MPXV proteins served as a negative control. Mn adjuvant was added to each vaccine group at a final concentration of 1 mg/mL prior to immunization. Blood samples were collected on 14, 35, 95, and 155 days after immunization and incubated at a temperature of 37°C for 60 min to allow coagulation and centrifuged at 5000 g and 4°C for 10 min. The serum was carefully collected and stored at -20°C until further use.
## Enzyme-linked immunosorbent assay (ELISA)
IgG antibody titres against MPXV-specific antigens A35R and M1R were determined using ELISA. Briefly, 96-well ELISA plates were coated with 1 μg/ mL of A35R or M1R proteins in 0.05 M carbonatebicarbonate buffer (pH 9.6) and incubated overnight at 4°C. The following day, the plates were washed with PBST and blocked with ELISA blocking buffer for 1 h at 37°C. Mouse serum samples were serially diluted in a 2-fold gradient with blocking buffer, starting at 1:100, and added to each well. After incubation for 1 h at 37°C, the plates were washed, and 100 µL of a 1:10,000 dilution of HRP-conjugated goat anti-mouse IgG antibody (Proteintech) was added, followed by incubation for 1 h at 37°C. Next, 100 µL of TMB substrate solution (Solarbio) was added and incubated for 20 min. The reaction was quenched with 100 µL of stop solution (Solarbio). Absorbance at 450 nm was recorded using a microplate reader (PerkinElmer, USA), and the endpoint titres were defined as the highest reciprocal dilution of serum where the absorbance was greater than 2-fold the absorbance of the negative control.
## Plaque reduction neutralization test (PRNT)
The plaque reduction neutralization test was performed to assess the neutralizing antibody titres in serum samples. Briefly, serum samples were heat-inactivated at 56 °C for 30 min. For MPXV PRNT, 100 μL of 4fold serum dilution series (starting from 1:40) was mixed with an equal volume of approximately 100 PFUs of MPXV. For VACV-WR PRNT, 100 μL of 2-fold serum dilution series (starting from 1:20) was mixed with an equal volume of approximately 100 PFUs of VACV-WR. After incubated for 1 h at 37°C, the virus-serum mixtures were transferred to 24-well plates containing Vero-E6 cells and further incubated for 1 h at 37°C with 5% CO 2 . After incubation, the virus-serum mixtures were removed, and semisolid 0.9% methylcellulose in DMEM containing 2% FBS and penicillin (100 U/mL)-streptomycin (100 mg/mL) were added to the plates. The plates were then incubated for 4 days at 37°C in a humidified 5% CO₂ atmosphere. After 4 days, the plaques were fixed with 8% formaldehyde for 1 h and visualized by crystal violet staining. The number of plaques was counted, and the 50% plaque reduction neutralization antibody titre (PRNT50) was calculated using GraphPad Prism 9.0 software.
## Flow cytometry analyses
Intracellular cytokine staining was performed by flow cytometry assay to measure MPXV antigen-specific T cells in immunized mice. Briefly, approximately 2,000,000 mouse splenocytes were added to the 24wells plates and then stimulated with 10 μg/mL purified A35R or M1R proteins for 16 h at 37°C in 5% CO 2 . After incubation, Brefeldin A (protein transport inhibitor, Abs810012) was added to each well, and the cells were incubated for an additional 8 h at 37°C in 5% CO₂. The cells were then harvested, and 0.5 μL of Fixable Viability Stain 510 (BD Biosciences, USA) was added, followed by incubation for 20 min at 4°C in the dark. After incubation, the cells were washed twice with PBS, and stained with fluorescently conjugated antibodies to CD3 (FITC, BD Biosciences, USA), CD4 (BV421, BD Biosciences, USA), and CD8 (PerCP-CY5.5, BD Biosciences, USA) for 30 min at 4°C in the dark. The cells were then washed twice with PBS, fixed, and permeabilized using the Cytofix/Cytoperm kit (BD Biosciences, USA). Afterward, the cells were stained with fluorescently conjugated antibodies to interferon-γ (IFN-γ) (PE-CY7, BD Biosciences, USA), interleukin-4 (IL-4) (PE, BD Biosciences, USA), and tumour necrosis factor alpha (TNF-α) (APC, BD Biosciences, USA) for 30 min at 4°C in the dark. Following a final wash, the cells were analyzed using a BD FACSCanto flow cytometer and the data were processed with FlowJo software.
## Virus challenge
At 42 or 160 days after initial immunization, mice were intranasally administered 10 LD 50 of the VACV-WR in 20 μL. After infection, weight changes and survival rates were monitored over a 14-day period. At 5 days post-infection, three mice from each group were randomly euthanized. Lung tissues were collected, weighed, and either preserved in 1 mL of DMEM for viral titration or placed in tissue fixation fluid for histological analysis. Lung tissue samples in DMEM were homogenized using a tissue homogenizer, followed by three freeze-thaw cycles to release viral particles. Viral titres in the lung were determined by plaque assay, and viral loads were quantified by quantitative polymerase chain reaction (qPCR). The primers used for qPCR were as follows: Forward, TCCACAACAGACGAGACTCC; Reverse, GGTTGTACTACCGCCTACAGT. Lung tissues fixed in tissue fixation fluid were embedded in paraffin, sectioned, and stained with haematoxylin and eosin (H&E) to assess pathological changes.
Mice were euthanized if they experienced a body weight loss exceeding 25% or at the conclusion of the experiment.
## Statistical analysis
Statistical analyses were conducted using GraphPad Prism 9.0 (GraphPad Software). For comparisons involving more than two groups in the in vitro assays, one-way analysis of variance (ANOVA) was applied. Data are presented as the mean ± standard error of the mean (SEM). Differences were considered statistically significant when the p-value was less than 0.05.
## Results
## Design and production of MPXV nanoparticle vaccines
To develop MPXV nanoparticle vaccines, we first expressed and purified the relevant proteins. The mi3 coding sequence, featuring an N-terminal Spy-Catcher fusion and a C-terminal hexahistidine tag, was cloned into the prokaryotic expression vector pET28a + and subsequently transformed into Escherichia coli to produce SpyCatcher-mi3 protein (Figure 1(a)). The MPXV antigens, M1R and A35R, were both fused with a signal peptide at the N-terminus and a SpyTag followed by a hexahistidine tag at the C-terminus. These constructs were cloned into the eukaryotic expression vector pCAGGS to express A35R-SPT and M1R-SPT proteins in HEK293F cells respectively (Figure 1(a)). Subsequently, the Spy-Catcher-mi3 nanoparticle protein was incubated with a 1:4 molar excess of A35R-SPT or M1R-SPT protein overnight at 4°C to form a covalent peptide bond, thereby enabling the display of the MPXV antigens on the nanoparticle surface (Figure 1(b)).
## Characterization of MPXV nanoparticle vaccines
The mi3-A35R and mi3-M1R nanoparticles were purified by SEC. As shown in Figure 1c, the retention volume peaks of both mi3-A35R and mi3-M1R shifted forward compared to mi3, indicating an increase in molecular weight due to antigen conjugation. Additionally, the hydrodynamic diameters of mi3, mi3-A35R, and mi3-M1R were measured using DLS. It's showed in Figure 1d shows that the hydrodynamic diameters of both mi3-A35R and mi3-M1R were larger than that of mi3, demonstrating antigen conjugation. Structural analysis by TEM revealed that all three nanoparticles -mi3, mi3-A35R, and mi3-M1R -formed uniform, stable, and dispersed particles (Figure 1(e)). Notably, in contrast to mi3, proteins attachment was clearly visible on the surfaces of mi3-A35R and mi3-M1R nanoparticles.
To assess the thermal stability of A35R, M1R, mi3, mi3-A35R, and mi3-M1R proteins, we performed a comprehensive stability analysis. The protein samples were subjected to various storage conditions, including 37°C, 25°C, 4°C, and -80°C, for a duration of one week. The A35R monomer protein exhibits inherent instability and is prone to degradation, even at low temperatures, as evidenced by electrophoretic band tailing (Figure S1). However, upon conjugation with mi3 nanoparticles, the stability of A35R is markedly improved, and the band tailing phenomenon is eliminated (Figure S1). Furthermore, M1R, mi3, mi3-A35R, and mi3-M1R demonstrated remarkable structural integrity and stability under all tested storage conditions, with no observable degradation detected at any temperature (Figure S1).
These results demonstrate the successful development of stable mi3-A35R and mi3-M1R nanoparticle vaccines against MPXV.
## MPXV nanoparticle vaccine induces a robust humoral immune response in mice
An equimolar mixture of mi3-A35R and mi3-M1R was prepared to form a bivalence nanoparticle vaccine, termed mi3-AM, while an equimolar mixture of A35R and M1R was used as a monomer vaccine control, named AM (Figure 2(a)).
To assess the immunogenicity and efficacy of the MPXV nanoparticle vaccine, female BALB/c mice (6-8 weeks old) were immunized twice with mi3-AM, AM, or mi3, with 1 mg/mL Mn adjuvant [32], at three-week intervals (Figure 2(b)). Sera were collected at 35 days post-immunization to assess the humoral immune responses. ELISA analysis demonstrated that mi3-AM and AM induced high antibody titres (IgG, IgG1, and IgG2a) against MPXV A35R, and mi3-AM induced significantly higher levels of MPXV A35Rspecific IgG2a compared to AM (Figure 2(c)). Additionally, the titres of MPXV A35R-specific IgG and IgG1 induced by mi3-AM were marginally higher than those induced by AM, although the difference was not statistically significant (Figure 2(c)). Regarding MPXV M1R-specific binding antibody, both mi3-AM and AM induced significantly strong antibody responses, and the antibody responses of IgG1 and IgG2a induced by mi3-AM were significantly stronger than those in the AM group (Figure 2(d)). Furthermore, the titres of MPXV M1R-specific IgG were similar between mi3-AM and AM (Figure 2(d)). The IgG1/ IgG2a ratio, which reflects the type of immune response, exceeded than 1 for both A35R and M1R, indicating a Th2-favored response [33] (Figure 2(c,d)).
Additionally, neutralization capacity of day 35 sera was evaluated through PRNT. High neutralizing antibody titres against both VACV-WR and MPXV were observed in the sera of both AM-and mi3-AM-immunized mice, and the titres of neutralizing antibody in the sera of mi3-AM-immunized mice were higher than that in AM group, but no statistical difference was shown (Figure 2(e,f)). No neutralizing activity against the VACV-WR or MPXV was detected in the mi3-immunized group.
These results demonstrate that both mi3-AM and AM vaccines induce strong humoral immune responses and exhibit high immunogenicity and efficacy in mice.
## MPXV nanoparticle vaccine elicits potential cellular immune responses in mice
In addition to humoral immunity, cellular immune responses are crucial for effective protection against viral infections. To evaluate cellular immunity, we performed intracellular cytokine staining (ICS) followed by flow cytometry to assess cytokine production by splenocytes collected 42 days post-immunization. We focused on IFN-γ, IL-4, and TNF-α production in response to MPXV antigens M1R and A35R.
For flow cytometry analysis, cell populations were gated based on the strategy outlined in Figure S2. Representative scatter plots depicting intracellular cytokine expression in antigen-specific CD8 + and CD4+ T cells induced by the MPXV vaccines are shown in Figure 3a-d. The total cellular responses to A35R and M1R stimulation in CD8 + and CD4+ T cells are summarized in Figure 3e andf, respectively. These results clearly demonstrate that the MPXV vaccines induced strong antigen-specific CD8 + and CD4+ T cells responses, characterized by a Th1-biased immune profile with elevated levels of cytokines such as IFN-γ and TNF-α in the spleen. Notably, TNF-α levels in A35R-stimulated CD8+ T cells and M1R-stimulated CD4+ T cells from the spleens of mi3-AM-immunized mice were significantly higher than those from AMimmunized mice. Additionally, no significant difference was observed in the induction of antigen-specific CD8 + and CD4+ T cells producing the type 2 (Th2) cytokine IL-4 among mice immunized with mi3-AM, AM, or mi3.
Collectively, these results demonstrate that the MPXV vaccines elicit robust cellular immune responses in mice.
## MPXV nanoparticle vaccine protects mice from lethal VACV-WR challenge
To assess the protective efficacy of the MPXV nanoparticle vaccine, BALB/c mice immunized with mi3-AM, AM, or mi3 were intranasally challenged with a 10 LD 50 dose of VACV-WR on 42 days post-immunization (Figure 2(b)). The changes of body weight and survival of mice were monitored for 14 days post-infection.
During the first two days post-infection, mice in the mi3-AM and AM groups experienced slight weight loss, but subsequently regained weight and stabilized. In contrast, mi3-immunized mice continued to lose weight until reaching the euthanasia threshold (25% weight loss from baseline) (Figure 4(a)). While no adverse symptoms were observed in mi3-AM or AM-immunized mice, mi3immunized mice exhibited symptoms such as hair loss, shivering, loss of appetite, and all died within 7 days post-infection (Figure 4(b)).
At 5 days post-infection, three mice from each group were euthanized to assess viral replication in lung tissues. The viral titres were determined by plaque assay and the viral loads were measured by qPCR. No infectious virus particles were detected in the lung tissue samples of mi3-AM-or AM-immunized mice, whereas viral titres in the lungs of mi3immunized mice reached 10 8 PFU/g (Figure 4(c)). Consistent with these results, the viral DNA was detected in the lung tissues of mi3-immunized mice but was absent in mi3-AM-or AM-immunized mice (Figure 4(d)).
To further evaluate the protective effect of the MPXV vaccines, lung tissues were subjected to haematoxylin and eosin (H&E) staining to assess pathological changes induced by VACV-WR infection. Histopathological analysis of lungs indicated that VACV-WR infection induced severe lung lesions in mi3-immunized mice, which were characterized by thickened alveolar septa and significant inflammatory cells infiltration. In contrast, lung tissues of mi3-AM-immunized mice and AM-immunized mice showed no pathological changes (Figure 4(e)).
These results indicate that both mi3-AM and AM vaccines provide robust protection against lethal VACV-WR infection, effectively preventing weight loss, viral replication, and pathological lung damage.
## MPXV nanoparticle vaccine induces durable antibody responses in mice
To evaluate the durability of antibody responses and protective efficacy of vaccines, female BALB/c mice aged 6-8 weeks were immunized with mi3-AM, AM or mi3 on day 0 and 21. Sera were collected on day 14, 35, 95, and 155. At day 160, mice were intranasally challenged with a 10 LD 50 dose of VACV-WR (Figure 5(a)).
The sustained antibody responses against MPXV antigens (A35R, M1R) were assessed by ELISA (Figure 5(b,c)). At day 14, mi3-AM-immunized mice exhibited high titres of MPXV A35R-specific IgG (∼1:56,320) and MPXV M1R-specific IgG (∼1:2,621,440), whereas AM-immunized mice showed significantly lower titres (∼1:5,200 for A35R, ∼1:48,640 for M1R). After boosting, both vaccines induced a rapid increase in antibody levels. By day 95, antibody titres of A35R-and M1R-specific IgG in mi3-AM-immunized mice had decreased by only 1.5 times compared to day 35, whereas AMimmunized mice showed a 5.8-fold reduction for A35R-specific IgG and a 2-fold reduction for M1R- specific IgG. At day 155, mi3-AM immunization still resulted in higher antibody titres than AM immunization, indicating more robust and durable antibody responses.
The durability of neutralizing antibody responses against VACV-WR and MPXV were assessed by PRNT assay (Figure 5(d,e)). At day 14, only mi3-AM-immunized mice displayed weak neutralizing activity against both VACV-WR and MPXV. After boosting, neutralizing antibody levels were significantly enhanced in both vaccine groups at day 35. However, by day 155, neutralizing antibody titres had decreased in both groups, with a more significant reduction observed in AM-immunized mice.
These results demonstrate that the mi3-AM nanoparticle vaccine induces durable antibody responses in mice.
## MPXV nanoparticle vaccine provides long-Lasting protection in mice
To determine the durability of protective abilities, mice were challenged intranasally with a 10 LD 50 dose of VACV-WR on 160 days post-immunization, and weight changes were recorded for 14 days. As is shown in Figure 6a, the mi3-immunized mice showed a significant weight loss of more than 25% within 9 days after infection, the AM-immunized mice experienced obvious weight loss as much as 14.83% within 7 days after infection, while mi3-AM-immunized mice only lost 7.41% of body weight within 3 days and showed no signs of severe infection. Notably, one of the AM-immunized mice died at day 9, while the mi3-AM immunization completely protected the mice from VACV-WR challenge (Figure 6(b)).
Lung tissues were collected at 5 days post-infection to assess the viral titres and the viral loads. The infectious virus particles and viral DNA were detected in the lung tissues of mi3-and AM-immunized mice, but were absent in mi3-AM-immunized mice (Figure 6(c,d)). H&E staining revealed severe lung lesions in mi3-immunized mice, characterized by thickened alveolar septa and inflammatory cell infiltration. AM-immunized mice exhibited milder lung damage, while mi3-AM-immunized mice exhibited no pathological changes (Figure 6(e)).
These results demonstrate that the mi3-AM nanoparticle vaccine provides long-term protection against MPXV in mice.
## Discussion
In this study, we developed a bivalent mpox nanoparticle vaccine, mi3-AM, by conjugating M1R and A35R to the mi3 nanoparticle using the SpyTag-SpyCatcher system. After intraperitoneal immunization of BALB/c mice with a Mn adjuvant, the mi3-AM nanoparticle vaccine effectively induced potent humoral and cellular immune responses. Furthermore, mi3-AM provided robust, long-lasting protection against lethal challenge with the VACV-WR strain, demonstrating the effectiveness of the nanoparticle vaccine. These findings position mi3-AM as a promising next-generation vaccine candidate that could be rapidly deployed to combat emerging mpox outbreaks, with potential for adaptation against other orthopoxviruses.
Since the global eradication of smallpox in 1980 [34], the live-attenuated VACV-based smallpox vaccine has been regarded as one of the most successful vaccines in medical history. The genomes of all orthopoxviruses exhibit high homology, with MPXV and VACV sharing 96.3% sequence similarity [35,36], suggesting that VACV-based vaccines have potential cross-protection effects against MPXV. However, the two FDA-approved live-attenuated VACV-based vaccines, ACAM2000 and JYNNEOS, which are authorized for emergency use [37], have shown limited cross-protection against MPXV [15] and may cause serious side effects [19]. While live-attenuated vaccines induce immunity against the entire pathogen, mRNA and protein-subunit vaccines focus on key components, offering safer alternatives. The rapid development of MPXV mRNA vaccines following the 2022 mpox outbreak has demonstrated their ability to induce immune responses and provide protection in animal models, including mice and macaques [20,25,26,[38][39][40]. However, epidemiological investigations have revealed that during the 2022 mpox outbreak, over 90% of reported mpox cases occurred among gay, bisexual, and other men who have sex with men, with more than 35% of these cases also co-infected with HIV [41][42][43][44]. This population often has underlying immune deficiencies, which may limit the ability of mRNA vaccines to induce robust immune responses. Furthermore, mRNA vaccines face additional challenges, including relatively short-lived immunity, the need for multiple doses to sustain protection, and high dependency on cold-chain storage and transport [45,46], which remain significant limitations of the technology.
In contrast, protein-subunit vaccines, which express key pathogen antigens in a controlled system, are known for their high safety, stability, and efficacy. The immunogenicity of protein subunit vaccines can be effectively enhanced by structure-based design strategies, such as antigen polymerization, or adjuvant stimulation [47]. For instance, the ZF2001 vaccine against SARS-CoV-2, based on the spike protein receptor binding domain (RBD), demonstrated excellent safety and efficacy in clinical trials [48]. Moreover, a structure-guided, multi-antigen fusion strategy has been employed to develop MPXV subunit vaccines, such as the DAM vaccine, which protected mice from lethal VACV challenges [27]. These findings highlight the feasibility of developing MPXV vaccines based on the MV M1R and EV A35R antigens.
Self-assembled nanoparticles, such as the mi3 nanoparticle platform employed in this study, represent ideal vaccine carriers. By promoting the multimerization of viral antigens, nanoparticles enhance the immunogenicity, stability, and overall efficacy of vaccines, as evidenced by their successful application in COVID-19 [49], influenza [50], and HIV [51] nanoparticle vaccines. In this study, the mi3 nanoparticle platform was used to present MPXV antigens M1R and A35R, creating a bivalent nanoparticle vaccine (mi3-AM). After two doses of intraperitoneal immunization in mice with the Mn adjuvant, mi3-AM elicited significantly stronger antibody and cellular immune responses and provided complete protection against a lethal challenge with VACV-WR compared to mi3 immunization. Interestingly, even without the mi3 nanoparticle, AM alone also induced strong antibody and cellular immune responses and provided complete protection against the VACV-WR challenge. This could be attributed to the enhanced immune response facilitated by the Mn adjuvant [52]. Notably, after a single dose of intraperitoneal immunization in mice with the Mn adjuvant, mi3-AM elicited a stronger antibody response than AM and mi3, highlighting the advantages of nanoparticle-based vaccines and suggesting that protection against VACV-WR infection in mice may be achieved with just one dose of mi3-AM. Furthermore, mi3 nanoparticles significantly enhanced the durability and stability of antibody responses, maintaining high titres over time. On day 160 post-immunization, mi3-AM provided complete protection against VACV-WR challenge, whereas the antibody titres elicited by AM decreased significantly and no longer conferred complete protection. These results demonstrate that mi3-AM is a promising candidate for the prevention and control of global MPXV outbreaks.
A limitation of this work is the absence of MPXV challenge data in murine models. Animal models are indispensable tools for investigating the aetiology, immunological responses, and pathological mechanisms of infectious diseases, as well as for evaluating the efficacy of vaccines, therapeutic antibodies, and pharmacological interventions. Although mouse models are widely regarded as the most accessible and cost-effective experimental system, their utility in studying MPXV infection is limited due to the manifestation of only mild, self-limiting symptoms in infected mice [53,54], which may not adequately recapitulate the pathogenesis observed in humans. In contrast, non-human primates, particularly macaques, are considered the most physiologically relevant animal models for studying human orthopoxvirus infections [55], as they accurately recapitulate key clinical manifestations of both smallpox and mpox, including the characteristic disseminated cutaneous lesions and systemic disease progression observed in human patients. While we have demonstrated that the mi3-AM vaccine provides complete protection against VACV-WR in mice, further studies are needed to evaluate its cross-clade protection in non-human primates.
Overall, we have developed a bivalent nanoparticle vaccine against MPXV that induces strong immune responses in mice and provides durable protection against VACV-WR challenge. The mi3 nanoparticle platform enables the rapid manufacturing and large-scale production of next-generation vaccines. Given the ongoing MPXV outbreaks, this vaccine represents a promising candidate for global MPXV prevention, with the potential to significantly impact public health.
## References
1. Zhu, Ji, Shi (2022) "Unusual global outbreak of monkeypox: what should we do" *Front Med*
2. (2022) "Monkeypox virus outbreak: can evolution guide us to new treatments or vaccines?" *EBioMedicine*
3. Papukashvili, Rcheulishvili, Liu (2022) "Strategy of developing nucleic acid-based universal monkeypox vaccine candidates" *Front Immunol*
4. Lum, Torres-Ruesta, Tay (2022) "Monkeypox: disease epidemiology, host immunity and clinical interventions" *Nat Rev Immunol*
5. Liu, Virology (2014) *Clin Infect Dis*
6. Petersen, Harms, Reynolds (2015) "Use of vaccinia virus smallpox vaccine in laboratory and health care personnel at risk for occupational exposure to Orthopoxviruses"
7. Hooper, Thompson, Wilhelmsen (2004) "Smallpox DNA vaccine protects nonhuman primates against lethal monkeypox" *J Virol*
8. Wyatt, Earl, Eller (2004) "Highly attenuated smallpox vaccine protects mice with and without immune deficiencies against pathogenic vaccinia virus challenge" *Proc Natl Acad Sci U S A*
9. Hatch, Graham, Bewley (2013) "Assessment of the protective effect of Imvamune and Acam2000 vaccines against aerosolized monkeypox virus in Cynomolgus macaques" *J Virol*
10. Garza, Hatkin, Livingston (2009) "Evaluation of the efficacy of modified vaccinia Ankara (MVA)/ IMVAMUNE against aerosolized rabbitpox virus in a rabbit model" *Vaccine*
11. Rizk, Lippi, Henry (2022) "Prevention and treatment of monkeypox" *Drugs*
12. Zaeck, Lamers, Verstrepen (2023) "Low levels of monkeypox virus-neutralizing antibodies after MVA-BN vaccination in healthy individuals" *Nat Med*
13. Montalti, Valerio, Angelini (2023) "Safety of monkeypox vaccine using active surveillance, twocenter observational study in Italy" *Vaccines (Basel)*
14. Rao, Petersen, Whitehill (2022) "Use of JYNNEOS (smallpox and monkeypox vaccine, live, nonreplicating) for preexposure vaccination of persons at risk for occupational exposure to orthopoxviruses. Recommendations of the Advisory Committee on Immunization Practices -United States" *MMWR Morb Mortal Wkly Rep*
15. Thornhill, Barkati, Walmsley (2022) "Monkeypox virus infection in humans across 16 countries" *N Engl J Med*
16. Zhang, Wang, Zhu (2023) "Rational development of multicomponent mRNA vaccine candidates against mpox" *Emerging Microbes Infect*
17. Zeng, Li, Jiang (2023) "Mpox multi-antigen mRNA vaccine candidates by a simplified manufacturing strategy afford efficient protection against lethal orthopoxvirus challenge" *Emerging Microbes Infect*
18. Franceschi, Parker, Jacca (2015) "BoHV-4-based vector single heterologous antigen delivery protects STAT1(-/-) mice from monkeypoxvirus lethal challenge" *PLoS Negl Trop Dis*
19. Wang, Yang, Zhou (2023) "Immunogenic proteins and potential delivery platforms for mpox virus vaccine development: a rapid review" *Int J Biol Macromol*
20. Alkhalil, Strand, Mucker (2009) "Inhibition of monkeypox virus replication by RNA interference" *Virol J*
21. Zuiani, Dulberger, Silva (2024) "A multivalent mRNA monkeypox virus vaccine (BNT166) protects mice and macaques from orthopoxvirus disease" *Cell*
22. Hou, Zhang, Liu (2023) "mRNA vaccines encoding fusion proteins of monkeypox virus antigens protect mice from vaccinia virus challenge" *Nat Commun*
23. Wang, Yin, Zheng (2024) "Rational design of a 'two-in-one' immunogen DAM drives potent immune response against mpox virus" *Nat Immunol*
24. Bruun, Andersson, Draper (2018) "Engineering a rugged nanoscaffold to enhance plugand-display vaccination" *ACS Nano*
25. Hsia, Bale, Gonen (2016) "Design of a hyperstable 60-subunit protein icosahedron" *Nature*
26. Zakeri, Fierer, Celik (2012) "Peptide tag forming a rapid covalent bond to a protein, through engineering a bacterial adhesin" *Proc Natl Acad Sci USA*
27. Anuar, Banerjee, Keeble (2019) "Spy&Go purification of SpyTag-proteins using pseudo-SpyCatcher to access an oligomerization toolbox" *Nat Commun*
28. Zhang, Wang, Guan (2021) "Manganese salts function as potent adjuvants" *Cell Mol Immunol*
29. Mills, Kincaid, Alt (2000) "M-1/M-2 macrophages and the Th1/Th2 paradigm" *J Immunol*
30. Gubser, Hué, Kellam (2004) "Poxvirus genomes: a phylogenetic analysis" *J Gen Virol*
31. Shchelkunova, Shchelkunov (2022) "Smallpox, monkeypox and other human orthopoxvirus infections" *Viruses*
32. Sang, Zhang, Liu (2023) "Monkeypox virus quadrivalent mRNA vaccine induces immune response and protects against vaccinia virus"
33. Xia, He, Zhan (2023) "Mpox virus mRNAlipid nanoparticle vaccine candidates evoke antibody responses and drive protection against the vaccinia virus challenge in mice" *Antiviral Res*
34. Tang, Liu, Lu (2023) "Recombinant proteins A29L, M1R, A35R, and B6R vaccination protects mice from mpox virus challenge" *Front Immunol*
35. Tarín-Vicente, Alemany, Agud-Dios (2022) "Clinical presentation and virological assessment of confirmed human monkeypox virus cases in Spain: a prospective observational cohort study" *Lancet*
36. Philpott, Hughes, Alroy (2022) "Epidemiologic and clinical characteristics of monkeypox cases -United States" *MMWR Morb Mortal Wkly Rep*
37. Curran, Eberly, Russell (2022) "HIV and sexually transmitted infections among persons with monkeypox -eight U.S. jurisdictions. MMWR Morb Mortal Wkly Rep"
38. Liu, Rakhmanina, Yang (2024) "Mpox (Monkeypox) virus and its co-infection with HIV, sexually transmitted infections, or bacterial superinfections: double Whammy or a new prime culprit?" *Viruses*
39. Barbier, Jiang, Zhang (2022) "The clinical progress of mRNA vaccines and immunotherapies" *Nat Biotechnol*
40. Pardi, Hogan, Porter (2018) "mRNA vaccinesa new era in vaccinology" *Nat Rev Drug Discov*
41. Wang, Shang, Jiang (2020) "Subunit vaccines against emerging pathogenic human coronaviruses" *Front Microbiol*
42. Dai, Gao, Tao (2022) "Efficacy and safety of the RBD-dimer -based Covid-19 vaccine ZF2001 in adults" *N Engl J Med*
43. Kang, Sun, Zhuang (2021) "Rapid development of SARS-CoV-2 spike protein receptor-binding domain self-assembled nanoparticle vaccine candidates" *ACS Nano*
44. Kanekiyo, Wei, Yassine (2013) "Self-assembling influenza nanoparticle vaccines elicit broadly neutralizing H1N1 antibodies" *Nature*
45. Cohen, Rosa, Fulp (2023) "A first-in-human germline-targeting HIV nanoparticle vaccine induced broad and publicly targeted helper T cell responses" *Sci Transl Med*
46. Zhang, Wang, Guan (2021) "Manganese salts function as potent adjuvants" *Cell Mol Immunol*
47. Osorio, Iams, Meteyer (2009) "Comparison of monkeypox viruses pathogenesis in mice by in vivo imaging" *PLoS One*
48. Hutson, Abel, Carroll (2010) "Comparison of West African and Congo Basin Monkeypox viruses in BALB/c and C57BL/6 mice" *PLoS One*
49. Cann, Jahrling, Hensley (2013) "Comparative pathology of smallpox and monkeypox in man and macaques" *J Comp Pathol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12757852&blobtype=pdf | # Monkeypox Virus: Epidemiology, Virology, Diagnosis, Vaccine, and Therapeutics
Yunzheng Yan, Yaqin Sun, Guangyan Sun, Cheng Niu, Xinyuan Zhao, Ming Zhao, Tongyao Liu, Suyue Zhang, Hui Zhai, Ankang Liu, Shouzhi Yu, Shuyuan Pan, Wu Zhong, Yuntao Zhang, Song Li
## Abstract
Since 2022, mpox epidemics have been sustaining and escalating over the world, posing a significant public health challenge. While significant progress has been made in diagnostic methodologies, prophylactic vaccines, and therapeutic interventions to mitigate monkeypox virus (MPXV) infection, scientific understanding of MPXV and related orthopoxviruses continues to evolve progressively. In order to keep pace with recent advancements, herein we review progress in mpox research from five key perspectives. This article first summarizes the latest epidemiological profiles, incorporating different viral lineages globally and in China, while highlighting their evolutionary history and distinct clinical characteristics. The virological profiles of MPXV shed light on its complete infectious lifecycle and the formation of distinct virus particle types. Clinically approved classical detection methods and emerging novel testing techniques are provided, establishing a framework for early diagnosis of mpox patients. The efficacy and safety of both licensed vaccines and those under development are analyzed to underscore their value in preventing mpox infection. Additionally, progress in approved and newly identified potential therapeutic agents is summarized and discussed, aiming to provide insights for further drug development and clinical treatment strategies.
## 1 Introduction
Monkeypox virus (MPXV) is a double-stranded DNA virus belonging to the genus Orthopoxvirus, the family Poxviridae [1].
The disease caused by MPXV, namely mpox, is a zoonotic viral disease with smallpox-like clinical manifestations [2]. MPXV can infect humans via various routes, such as the skin, respiratory tract, and mucous membranes under conditions of close human-to-human contact or environmental exposure; the specific transmission route significantly affects the duration of the incubation period [2]. In addition, virus strain, expo-sure route, infection dose, and the host's immune status all have significant impacts on the clinical presentations. Before skin lesions appear, the majority of patients exhibit prodromal symptoms such as fever, lymphadenopathy, myalgias, malaise, and headache; subsequently, the classic development process of the skin lesions, from the stages of macules, papules, vesicles, and pustules to scabs, could resolve within 2-4 weeks [3][4][5][6].
It is noteworthy that immunocompromised populations (HIVinfected individuals, children, etc.) account for a high proportion among mpox patients, increasing susceptibility to severe disease and the risk of death [7][8][9][10].
Since the 1970s, mpox has emerged or reemerged in Central and West Africa, posing a significant threat to public health and safety [11]. Unprecedentedly, a global mpox outbreak began in May 2022, and the World Health Organization (WHO) declared it a "Public Health Emergency of International Concern (PHEIC)" in July 2022 and August 2024, respectively [12,13]. On August 12, 2022, the WHO classified MPXV into two major clades: Clade I (formerly Central African clade) and Clade II (formerly West African clade) [14]. Clade II was further subdivided into subclades IIa and IIb [15]. In April 2024, Clade I was further subdivided into Clade Ia and Clade Ib following the detection of genetically distinct variants [16]. Notably, these two global outbreaks were driven by distinct MPXV strains, with significant differences in demographic distribution, disease severity, and clinical course [17]. For instance, the Clade IIb-driven epidemic predominantly affected male populations, particularly men who have sex with men (MSM). Clinical manifestations included genital/perianal lesions, fever, and lymphadenopathy, with the most frequent number of skin lesions ranging from 2 to 10 [18,19]. In contrast, Clade Ib infections are associated with more severe phenotypes: the median maximum number of skin lesions in Clade Ib patients reaches 60 [20], reflecting a more acute and aggressive clinical course.
To provide a comprehensive understanding of MPXV and mpox, this review first updates on recent advancements in epidemiological profiles from the perspective of different strains, both globally and in China. We also outline the virological characteristics of MPXV, including its morphology, genome, and life cycle. Additionally, we systematically summarize and analyze progress in mpox diagnosis, vaccination, and therapeutic strategies, along with perspectives on future research directions. Together, these efforts aim to inform the prevention and control of mpox by offering valuable insights.
## 2 Epidemiology
MPXV was first identified in cynomolgus monkeys in 1958 [21].
In 1970, scientists isolated MPXV from a smallpox-like patient in the Democratic Republic of the Congo (DRC; formerly Zaire, 1971Zaire, -1997)), marking the first recognized human mpox case in medical history [22]. Since then, mpox has remained endemic in Africa, and after 2003, it has gradually spread globally [21], including to China. Between January 1, 2022 and September 30, 2025, 140 countries reported 168,974 laboratory-confirmed mpox cases and 441 deaths to the WHO, concurrently with the evolution of distinct viral lineages [23]. Notably, dominant circulating strains have varied temporally and geographically, differing in susceptible populations and mortality rates. This section details the global and Chinese epidemiological profiles of these MPXV clades.
## 2.1 Global Epidemiology of Mpox
Following its initial detection in Zaire, mpox remained endemic in Central and West Africa for decades. Between 1970 and 1997, over 700 confirmed mpox cases were reported in Africa, primarily in Zaire, with case fatality rates (CFRs) as high as 9.8%-17% [24][25][26]. Phylogenetic analyses later attributed these outbreaks with high mortality to MPXV Clade I; prior to 2022, the overall CFR of Clade I was approximately 10.6% [27]. Historically, Clade I infection presented the following clinical features: prodromal fever followed by a characteristic monomorphic, centrifugal rash often initiating on the face [28,29].
Notably, in 2003, the first mpox case outside Africa was reported in the United States, which was linked to imported animals from Ghana, marking the onset of global spread of mpox [30,31]. An analysis of 35 cases from this outbreak found that 85% of them presented with prodromal fever or fever with lesions; more than 77% had respiratory symptoms; 69% exhibited lymphadenopathy; and lesions were distributed on the head, trunk, arms, and legs [17]. This outbreak was confirmed to be caused by MPXV Clade IIa [17,32]. To date, Clade IIa remains primarily associated with sporadic zoonotic transmission via human-animal contact, with an overall CFR of approximately 1% as of 2023 [33].
Clade IIb first emerged in 2017-2018 in Nigeria, causing an outbreak with 122 confirmed and suspected cases [34]. Subsequent phylogenetic analyses further defined the clade associated with this outbreak as Lineage A [35,36]. Patients presented with cutaneous rashes on the face, legs, trunk, and genitals, accompanied by fever and lymphadenopathy [34]. Subsequently, Lineage B.1 of Clade IIb triggered a global outbreak in 2022 [36]. By July 2022, over 16,000 cases had been reported to the WHO from 75 countries/regions, prompting the first declaration of mpox as a PHEIC [37]. Characterized by exclusive human-to-human transmission, this outbreak predominantly affected MSM, including gay and bisexual populations [38]. The mean incubation period ranged from 2 to 21 days [38], with primary symptoms such as cutaneous rashes, fever, headache, and lymphadenopathy. Symptoms typically resolved within 2-4 weeks, but persisted longer in immunocompromised individuals [16]. Unlike prior outbreaks, this Clade IIb manifestation was relatively mild, with rashes mainly on the genital region [39,40]. Although reported cases have since plateaued, Clade IIb transmission persists globally, with an overall CFR of approximately 0.1% as of 2023 [40]. A recent study indicated that countries with robust public health systems reported fewer mpox cases and deaths [41].
In 2024, a pandemic driven by MPXV Clade I broke out in the DRC and subsequently spread to neighboring countries as well as non-African regions (e.g., Sweden, the United Kingdom, Thailand) [42]. This resurgence prompted the WHO to redeclare mpox a PHEIC in August 2024 [13]. Distinct from Clade IIb, Clade Ib outbreaks involve a significant proportion of women and children [13]. Clinically, a study of 226 suspected Clade Ib cases in South Kivu Province, DRC, revealed universal cutaneous rashes, with 59% developing fever, 40% exhibiting lymphadenopathy, and nonspecific symptoms (e.g., headache, myalgia) in patients [43]. Prospective studies also confirm that Clade Ib infection is associated with generalized rashes, fever, lymphadenopathy, and a notable risk of fetal loss during pregnancy [17,[43][44][45]. From January 2024 to October 24, 2025, 41 countries reported Clade Ib cases to the WHO, mostly concentrated in Africa [23].
In July 2025, the WHO released its latest epidemiological update clarifying the global clade landscape: Clade Ia remains endemic in the DRC with persistently high mortality, particularly among children (2%-3% CFR); Clade Ib circulates predominantly in Africa, with travel-associated cases in other regions and an approximate 0.5% CFR; Clade IIa is rarely reported; and Clade IIb maintains stable transmission in non-African regions (500-1,000 monthly cases; ∼0.5% CFR), but recently triggered an outbreak in Sierra Leone (African region) [46]. Notably, actual cases may be underestimated due to potential underreporting caused by diagnostic and surveillance incapacities. The global distribution of mpox clades is illustrated in Figure 1.
## 2.2 Epidemiology of Mpox in China
Mainland China reported its first imported mpox case in September 2022 [47]. The first indigenous case emerged in June 2023, followed by a surge peaking in August 2023; all cases during this surge were attributed to Clade IIb [48]. An analysis of epidemiological characteristics (June-December 2023) identified 1,702 male cases (99.42%) and 10 female cases (0.58%), with a median age of 31 years (range: 15-71 years). Approximately 90% of cases reported same-sex male sexual contact within 21 days prior to symptom onset, consistent with global transmission patterns. Among 1,666 tested individuals, 42.56% (n = 709) were HIV-positive [48]. A multicenter observational study (n = 286, September 2022-October 2024) further characterized the affected population, identifying MSM aged 27-37 years in coastal and economically developed regions as the primary group. This study also demonstrated that HIV coinfection was associated with severe disease, including elevated AST/CRP levels and reduced CD4 + T cell and NK cell counts [49].
In January 2025, the China Center for Disease Control and Prevention (CDC) reported a Clade Ib outbreak traced to a returnee from the DRC, resulting in four secondary cases through close contact [50]. One secondary case was a female presenting with limb pustules and trunk/buttock lesions, without fever or genital rash [51]. This clinical presentation differed from typical Clade IIb cases, which commonly feature fever, lymphadenopathy, and genital involvement [52,53]. As of October 24, 2025, China had reported 29 Clade Ib cases to the WHO [23]. Additionally, China reported travel-associated Clade Ia cases to the WHO in 2025 [23].
## 3 Virology
MPXV exhibits close relationships with other members of the Orthopoxvirus genus, such as vaccinia virus (VACV) and variola virus (VARV), sharing conserved lifecycle features and similar morphological characteristics. An in-depth understanding of MPXV virology is crucial for elucidating its replication strategies, immune evasion mechanisms, as well as informing the development of effective vaccines and antiviral agents. This section systematically reviews the fundamental virological
## FIGURE 2
The viral structure and life cycle of MPXV replication in host cells. The mpox virion consists of four major components: the core, lateral bodies, outer membrane, and outer lipoprotein envelope, with the core surrounded by a palisade layer and the outer membrane decorated with surface tubules. Its replication cycle proceeds through sequential stages: attachment and entry, uncoating, DNA transcription and replication, viral protein synthesis, virion assembly, and release. Intracellular mature virions (IMV) are either released directly via cell lysis or acquire additional envelopes from the Golgi apparatus to form intracellular enveloped virions (IEV). IEV are secreted either as cell-associated enveloped virions (CEV), which remain attached to the cell surface, or as extracellular enveloped virions (EEV). characteristics of MPXV, including virion morphology, genome organization, and lifecycle (Figure 2).
## 3.1 Morphology
VARV shows a brick-shaped architecture of roughly 250-350 × 200 nm as determined by diagnostic electron microscopy [54]. Furthermore, recent advances using cryo-electron tomography have demonstrated that the majority of mature MPXV particles exhibit an oblate ellipsoidal shape with dimensions of approximately 313 × 267 × 236 nm, in contrast to the more rectangular morphology of VACV particles (around 347 × 260 × 240 nm), with statistically significant differences in both the long and intermediate axes [55]. Structurally, these virions consist of four major components: core, lateral bodies, outer membrane, and outer lipoprotein envelope [56,57].
The core, which appears as a biconcave structure under electron microscopy, contains the linear DNA genome, core fibrils, essential viral enzymes, and associated transcription factors [58][59][60]. Surrounding the core, there is a dense, rod-like structural layer known as the palisade layer [61], composed of A10 trimers with dimensions of ∼8 × 13 nm (width × height) [55]. MPXV possesses two lateral bodies, located symmetrically on either side of the core [60]. Following fusion between the viral envelope and host cell membrane, lateral bodies detach from the viral core to deliver effector proteins into the host cytosol, while the released core functions as an early transcription factory [55].
The outer membrane comprises a 50-55 nm lipoprotein bilayer (cholesterol and phospholipid) and its surface displays randomly distributed tubules (average 7 nm in width and 100 nm in length), forming a characteristic textured morphology [57,62]. The cholesterol and phospholipid content of the bilayer, as well as the existence of surface tubules, are important to maintain the integrity of the outer membrane structure [63]. The virion, composed of a core, lateral bodies, and an outer membrane, is infectious; however, under specific infection conditions such as those involving extracellular enveloped virions (EEV), the particle acquires an additional, structurally distinct lipoprotein envelope [57]. In 1969, Boulter first elucidated the significance of the poxvirus envelope, demonstrating that its absence abrogates the protective efficacy of inactivated poxvirus vaccines upon challenge [64]. Subsequent studies confirmed that envelope-associated viral antigens would elicit immune responses across systems, conferring effective protection against poxvirus infection [65][66][67].
## 3.2 Genome
MPXV has a linear genome with a length of ∼197 kbp [68,69]. In comparison, VACV and VARV possess genomes of 165-199 kbp [70] and ∼186 kbp [71,72], respectively. Despite these differences in size, all three genomes contain two functional regions, specifically including a highly conserved central coding region and variable regions on both sides [69,73]. In addition, both termini of the genome are characterized by inverted terminal repeats, which include the hairpin loop, NR1 and NR2, short tandem repeats, as well as several open reading frames (ORFs) [69,74]. The above terminal genes play a vital role in regulating host immune responses to achieve immune evasion [75]. The ORFs in the central conserved regions share at least 90% sequence identity with those of other orthopoxviruses, which primarily encode viral proteins involved in the process from viral replication to particle release [69].
## 3.3 Life Cycle
Among orthopoxviruses, VARV's sole natural reservoir is humans; however, MPXV and VACV possess broad natural host ranges, with MPXV mainly infecting primates and rodents [76] and VACV lacking a clearly defined reservoir but possibly involving rodents and marsupials [77,78]. The MPXV genome encodes approximately 190-200 viral proteins, including structural and nonstructural proteins, which aid in viral replication and the entire lifecycle [69,79,80]. There are two kinds of infectious viral particles of poxviruses: EEV and intracellular mature virions (IMV) [81]. Distinct from the IMV's singlemembrane structure, EEV has a double-membrane structure, and its surface possesses different proteins and membrane components. As a result, these two types of viral particles rely on different modes to achieve viral attachment, hemifusion, and core entry [82]. Additionally, due to the absence of the lipid membrane layer, IMV exhibits more stable properties, which facilitates transmission between different individuals, whereas EEV is considered to mainly play a role in intercellular transmission [83]. Although MPXV is a DNA virus, its replication process is completed in host cells' cytoplasm rather than in the nucleus.
The initial step of MPXV infection is attachment, primarily mediated by interactions between viral surface proteins and host cell glycosaminoglycans (GAGs) [76,84]. Although specific cellular receptors for poxviruses remain incompletely characterized, studies on the closely related VACV have revealed multiple viral proteins involved in viral attachment [85]. In MPXV, proteins E8, A29, A28, and H3-homologous to VACV D8, A27, A26, and H3, respectively-play essential roles in mediating attachment [86][87][88][89][90]. Following attachment, the virus proceeds to the membrane fusion stage, a process mediated by the highly conserved Entry Fusion Complex (EFC) in poxviruses, which comprises proteins ranging from 4 to 43 kDa [82,91,92]. The EFC contains at least 11 components, including A16, A21, A28, F9, G3, G9, H2, J5, L1, L5, and O3 of VACV [85,93]. These proteins do not function independently, and currently three bimolecular interactions have been identified: A28-H2, A16-G9, and G3-L5 [85,94,95]. For example, the A16-G9 subcomplex can interact with either the viral A56-K2 complex or the A26 protein on the surface of mature virions to inhibit fusion [96]. Upon membrane fusion, the viral core is released into the cytoplasm, where EFC-associated proteins mediate its uncoating to initiate viral biosynthesis [85,97]. The exposed viral core is transported to the periphery of the cell nucleus through microtubule structures [76,98].
Subsequently, the viral genome, together with enzymes as well as regulatory factors, is released from the core of virions, which enables the synthesis of early mRNAs within minutes postinfection [99]. The replication region called the factory can be detected within 2 h postinfection. Multiple viral proteins are involved in poxvirus DNA replication, such as polymerase, helicaseprimase, uracil DNA glycosylase, protein kinase, and so on [100]. Furthermore, the transcription system of poxviruses is highly complicated, and this process is regulated in a cascade mechanism involving transcription initiation at early, intermediate, and late stages of infection [101]. In the stages of virus assembly and release, newly synthesized viral components assemble into mature virus particles called IMV [61]. While some IMVs accumulate intracellularly, leading to cell lysis, others traverse the trans-Golgi network (TGN) or nuclear membranes to form intracellular enveloped viruses (IEV) [102][103][104]. IEVs transported to the cell periphery fuse with the plasma membrane via exocytosis, generating cell-associated enveloped viruses (CEV) [105].
CEV particles either remain surface-bound or are released as EEV through actin-and microtubule-mediated transport [76,106,107].
## 4 Diagnosis of Mpox
MPXV infection is clinically indistinguishable from other eruptive diseases like varicella [108,109]. Additionally, coinfection with other pathogens, for example, syphilis or herpes simplex virus (HSV), has been documented in 12.1% of hospitalized cases [110]. Consequently, early diagnosis of mpox is both necessary and challenging. Current diagnostic methods include nucleic acid detection, immunological assays, and other approaches. Herein, we summarize the characteristics of MPXV diagnostic methods from three aspects (sensitivity/specificity, cost/requirement, and application scenarios) (Table 1).
## 4.1 Nucleic Acid Detection
RT-qPCR stands as the gold standard for MPXV nucleic acid detection [126]. To date, multiple RT-qPCR-based MPXV detection kits have been approved for clinical use globally. Among these kits, the F3L gene is the most commonly targeted region [127]. Beyond target selection, commercial kits share consistent characteristics, as exemplified by four kits approved by China's National Medical Products Administration (NMPA) as of May 2025: a sensitivity of 200 genome copies/mL, an amplification time of 30-40 min, and integration of a dUTP-UDG enzyme system to prevent carryover contamination from PCR amplicons [128][129][130][131]. These kits prioritize skin lesion specimens (rash swabs or exudates), aligning with the WHO guidelines [126,132]. The WHO additionally recommends oropharyngeal or anal swabs in the absence of skin lesions [126,132]. Furthermore, due to the relatively short duration of viremia compared to symptom persistence, the WHO advises against prioritizing blood samples for PCR [132]. Despite its high sensitivity, RT-qPCR's technical complexity confines it primarily to centralized laboratories [112].
In addition to RT-qPCR, other nucleic acid amplification-based technologies are increasingly being explored for MPXV detection. Among these, the most extensively studied include digital PCR and isothermal amplification technologies-such as LAMP and RPA. Digital PCR assays exhibit higher sensitivity than RT-qPCR, with a limit of detection (LOD) of six copies per reaction and a limit of quantitation (LOQ) of 38 copies/reaction [113]; however, this technology limits clinical translation due to its lengthy protocol [108]. Both LAMP and RPA enable nucleic
## Complex operation and long time-consuming process
Assessment of MPXV's spatial and temporal distribution [125] acid amplification under isothermal conditions, eliminating the need for thermal cycling equipment. LAMP, in particular, holds significant promise for POC applications due to its high sensitivity and operability under isothermal conditions [108,112]. For instance, an A27L/F3L-targeted LAMP assay demonstrates an LOQ of 20 copies/reaction with no cross-reactivity with 21 other pathogens, including related orthopoxviruses [117]. RPA offers shorter turnaround times (results within 10 min) than LAMP, but exhibits nonspecific amplification, particularly in complex clinical matrices [111].
Furthermore, several emerging technologies that do not rely entirely on nucleic acid amplification principles are being developed for the diagnosis of MPXV, including WGS and CRISPR-Cas systems [108]. WGS enables definitive MPXV identification, distinguishing it from other orthopoxviruses [119], and facilitates evolutionary tracking; however, its high computational cost restricts use to research settings [133]. CRISPR-Cas technologies identify MPXV DNA and leverage the endonuclease activity of Cas enzymes for viral DNA elimination [120]. Two notable applications, SHERLOCK (Cas13a-based) and DETECTR (Cas12a-based), are recommended for diagnostic use [134,135].
Despite their rapid detection capability, CRISPR-Cas technologies exhibit relatively lower sensitivity compared to PCR-based methods due to their non-amplification-based principle.
Combinatorial diagnostic approaches, which integrate multiple technologies, have been developed to enhance diagnostic performance. A recent study combined Multiple Enzyme Isothermal Rapid Amplification (MIRA) with CRISPR-Cas12b, yielding a single-tube assay with an LOD of four copies/reaction, 98.5% sensitivity, and 97.0% specificity, with no cross-reactivity to enterovirus A71 or HSV [136]. Other similar strategies merging MIRA-Cas13a [137] or LAMP-Cas12b [138] highlight the potential of next-generation diagnostic platforms for POC applications. Additionally, workstation-integrated diagnostic platforms have been deployed for MPXV diagnosis, exemplified by GeneXpert and Dragonfly systems. GeneXpert, a portable rapid diagnostic device, integrates orthopoxvirus-and MPXV-specific PCR assays with a 90-min turnaround time, demonstrating 98.8% sensitivity and 100% specificity [139]. The Dragonfly system combines power-free nucleic acid extraction with LAMP, achieving an LOD of 100 genome copies/mL, with sensitivities of 96.1% for orthopoxviruses and 94.1% for MPXV, alongside 100% specificity [140].
## 4.2 Immunological Assay
In addition to nucleic acid testing, immunological assays play a critical role in mpox diagnostics. Early studies utilized ELISA, immunohistochemistry, and immunofluorescence assays for MPXV confirmation [120]. However, ELISA-based detection of postinfection IgG/IgM is hampered by cross-reactivity with antibodies induced by prior smallpox/vaccinia vaccination [124], limiting its POC utility [108,123,141].
Inspired by COVID-19 antigen rapid diagnostic tests (Ag-RDTs), researchers have attempted to develop mpox-specific antigen or antibody assays. However, cross-reactivity with other orthopoxviruses poses a significant challenge [112]. Consequently, the WHO currently does not endorse antigen or antibody detection for primary diagnosis. A recent evaluation of three Clade IIb Ag-RDTs showed an LOD of 1.0 × 10 4 PFU/mL and 100% specificity but critically low sensitivity (0.00%-15.79% for skin samples, 0.00% for respiratory samples), thereby strongly discouraging their use for diagnosis or screening [142].
Despite these limitations, technologies related to Ag-RDTs have been advancing continuously. Specifically, assays targeting the A29L antigen have demonstrated promising sensitivity and specificity [143]. For instance, a lateral flow immunoassay (LFIA) based on two hybridoma-derived anti-A29L monoclonal antibodies (mAbs) differentiated MPXV from VACV, with an LOD of 7.5-15 ng/mL [143]. Another LFIA achieved an LOD of 50 pg/mL with no cross-reactivity to orthopoxviruses or respiratory pathogens like SARS-CoV-2 or influenza A/B [144]. Additionally, a recent study reported that the pixel-diverse interferometric reflectance imaging sensor enabled quantification of MPXV infection via targeting the A29 locus, with an LOD of 200 PFU/mL. This platform discriminated between HSV and CPXV, offering POC potential [116]. Antigen detection kits targeting other loci, including the A5L locus, are also under development [145].
## 4.3 Other Detection Methods
Other approaches for MPXV detection include virus isolation, TEM for direct visualization of poxvirus particles, and WBS [108,141,146,147]. Virus isolation is a critical method for confirming viral infections; however, MPXV isolation and cultivation require BSL-3 containment and rely on cell lines (e.g., Vero, Vero-E6, BSC-1, HeLa) or chick embryos, restricting this method to reference laboratories [133,148]. TEM enables direct visualization of characteristic brick-shaped poxvirus particles but is limited by low throughput and specialized equipment requirements [133].
WBS for viral shedding has emerged as an indicator of community exposure. MPXV DNA has been detected in wastewater (e.g., the Netherlands, Italy), supporting its potential for population-level monitoring [125]. However, the correlation between wastewater viral loads and clinical outbreak dynamics remains incompletely characterized, necessitating further longitudinal studies.
## 5 Mpox Vaccines
Vaccines stand as the most effective strategy for preventing infectious diseases. Currently, four vaccines have been approved for mpox and smallpox prevention: ACAM2000, MVA-BN, LC16m8, and OrthopoxVac (Table 2). According to the WHO classification, ACAM2000 is categorized as a second-generation vaccine, whereas MVA-BN and LC16m8 are classified as third-generation vaccines [149]. OrthopoxVac, developed through genetic engineering with the deletion of virulence-associated genes, is designated as a fourth-generation vaccine [32]. The approval of these vaccines underscores the availability of active intervention strategies for mpox prevention and control.
## 5.1 MVA-BN
MVA-BN, a third-generation live-attenuated non-replicating vaccine developed by Bavarian Nordic A/S, is administered as a additional route alongside the previously approved subcutaneous administration [152] two-dose regimen with a 4-week interval. It is indicated for use in individuals aged 18 years and older who are at high risk of smallpox or mpox infection (Table 3). Compared with traditional replicating vaccines, MVA-BN exhibits significantly improved safety profiles, enabling its use in populations with compromised immune systems. In 2024, the WHO added MVA-BN to its prequalification list as the first vaccine against mpox and designated it as a priority vaccine for mpox outbreak response [126,156].
The MVA-BN strain originates from the Modified Vaccinia Ankara (MVA) virus, a highly attenuated VACV strain derived from the Ankara vaccinia virus (CVA) through over 500 serial passages in chicken embryo fibroblasts (CEF) cells [157][158][159]. A key characteristic of MVA is its host restriction phenotype-an inability to infect and replicate in mammalian cells [160]. In 1998, Bavarian Nordic further passaged the MVA strain and renamed it MVA-BN [161][162][163][164].
The effectiveness of the MVA-BN vaccine in the nonclinical phase is primarily supported by findings from MPXV challenge studies in nonhuman primates (NHPs). As detailed in FDAreleased documents related to the JYNNEOS application, a total of 11 preclinical studies were conducted in cynomolgus macaques using three primary MPXV challenge models (intravenous [IV], intratracheal, and aerosol) [152]. These studies demonstrated that MVA-BN elicits the production of MPXV-specific binding antibodies and neutralizing antibodies in macaques, thereby affording protection against severe mpox symptoms and mortality following MPXV challenge.
To date, over 20 clinical trials have been conducted to evaluate the efficacy and safety of MVA-BN. A Phase III clinical trial (NCT01913353) directly compared MVA-BN with ACAM2000 to assess the safety and efficacy of MVA-BN in vaccinia-naive healthy individuals. Peak plaque reduction neutralization test (PRNT) results demonstrated that two doses of MVA-BN administered 28 days apart are non-inferior to ACAM2000 in preventing smallpox in individuals with no prior smallpox vaccination history. Additionally, the incidence of adverse events (AEs), particularly Grade 3 or higher, in the MVA-BN group during the two-dose vaccination course was lower than that in the ACAM2000-only group [152,165]. In addition, given the risk of severe complications from replicating smallpox vaccines in immunocompromised individuals, Bavarian Nordic conducted Phase II trials in atopic dermatitis (AD) patients and in HIVinfected individuals, respectively. Data from these studies supported the use of MVA-BN in immunocompromised individuals [166,167].
Following a comprehensive evaluation of the efficacy and safety of MVA-BN, the US FDA approved the MVA-BN vaccine for use in individuals aged 18 years and older at high risk of smallpox or mpox infection [152]. In 2022, the FDA granted an EUA for MVA-BN in individuals under the age of 18, and also approved intradermal administration (one-fifth of the subcutaneous dose) for adults over the age of 18 [152]. Currently, a subsequent Phase II trial (NCT05740982) in adolescents aged 12-17 years is ongoing, with interim analysis showing non-inferior antibody titers compared to adults aged 18-50 and no serious vaccine-related adverse events [168].
In real-world settings, Mason et al. analyzed data from 16 MVA-BN vaccine studies conducted from January 2022 to February 2024, reporting that for pre-exposure prophylaxis, the estimated effectiveness of a single dose ranged from 35% to 86%, and two doses from 66% to 90%; for post-exposure prophylaxis, the estimated effectiveness of one dose was 78% and 89% derived from two separate studies, respectively [169]. Similarly, Pischel et al. evaluated the effectiveness of third-generation vaccines in studies spanning 1970-2023, with findings showing that the estimated effectiveness of one dose of MVA-BN was 76% (95% CI: 64%-88%), and that of two doses was 82% (95% CI: 72%-92%) [170,171].
MVA-BN is now included in the US Routine Adult Immunization Schedule for any person at risk for mpox infection [172]. The most common adverse reactions following subcutaneous administration of MVA-BN include solicited injection site reactions such as pain, redness, swelling, induration, and itching. For intradermal administration, common adverse reactions include injection site erythema, induration, pruritus, pain, fatigue, headache, myalgia, nausea, axillary pain, anorexia, arthralgia, chills, and axillary swelling. Importantly, adverse reactions observed in HIV-infected adults and those with AD are comparable to those in healthy adults [152].
## 5.2 ACAM2000
ACAM2000 is a second-generation live attenuated replicating vaccine derived from the conventional calf lymph-derived smallpox vaccine Dryvax (Wyeth Laboratories, Inc.) [173]. Dryvax played a pivotal role in smallpox eradication across the Western Hemisphere and Africa but was limited by manufacturing processes that failed to meet modern regulatory standards, leading to substantial impurities and AEs [174][175][176][177]. Concerns over Dryvax's safety, depleting stockpiles, and the imperative to address potential smallpox bioterrorism threats prompted Acambis, Inc. to collaborate with the US CDC in July 1999 to develop the cell culture-based ACAM2000 [178,179].
The master seed virus for ACAM2000 was generated by passaging material from three distinct Dryvax batches in MRC-5 cells, followed by plaque purification, assessments of skin and neurotoxicity in animal models, and evaluation of attenuated phenotypes [178,179]. Scaled-up production was conducted in Vero cells, with subsequent concentration, purification, and lyophilization to form the vaccine bulk [179].
Initially developed for smallpox prevention, ACAM2000 underwent pre-licensure safety and efficacy evaluations via head-tohead comparisons with Dryvax. Nonclinical efficacy assessments focused on immunogenicity profiling in monkey and mouse models, and protective efficacy against lethal vaccinia virus (WR strain) challenge in mice. Results showed that ACAM2000 induced comparable immunogenicity, neutralizing antibody levels, and T-cell responses, as well as equivalent protective efficacy to Dryvax [178]. Safety evaluations encompassing rabbit skin toxicity (assessed by erythema and lesion diameters) and neurotoxicity in mice and monkeys (assessed by survival times) demonstrated that ACAM2000 exhibited lower toxicity than Dryvax [178,180,181].
During clinical development, Phase I-III trials enrolled vaccinianaive and previously vaccinated individuals, most of whom were in parallel-group, double-blind studies using Dryvax as the active control. Efficacy evaluations demonstrated that ACAM2000 induced robust cutaneous reaction rates and seroconversion rates, with non-inferiority to Dryvax [40]. However, it elicited a slightly lower Day-30 neutralizing antibody geometric mean titer (GMT) compared to Dryvax. Both GMT and seroconver-sion rates decreased with vaccine dilution-a phenomenon not observed with Dryvax [182]. In safety assessments, ACAM2000 recipients exhibited significantly less erythema at the vaccination site than Dryvax recipients [183]. Notably, cardiac AEs (myo-/pericarditis)-previously associated with NYCBH strain-based vaccines-were observed in both groups. The incidence rate was 5.73 per 1,000 vaccinations for ACAM2000, numerically lower than that in Dryvax recipients (10.38 per 1,000) but not statistically significant [175,179,183].
Following a comprehensive benefit-risk assessment of these studies, the FDA approved ACAM2000 in August 2007 for active immunization against smallpox in high-risk individuals aged ≥ 18 years [184]. The recommended regimen is a single dose, administered via percutaneous inoculation with 15 punctures using a sterile bifurcated needle-a technique specific to orthopoxvirus vaccination [153]. However, due to adverse reactions observed in clinical trials, a black box warning covering conditions such as myo-/pericarditis, severe vaccinial skin reactions, accidental eye infection, and risks during pregnancy was included in the labeling of ACAM2000 [153]. As a live replicating vaccine, ACAM2000 is also contraindicated in immunocompromised individuals.
In response to the global mpox outbreak, in August 2024, the FDA expanded the authorized use of ACAM2000 to include mpox prevention under an Expanded Access Investigational New Drug (EA-IND) designation [185]. This expansion was based on its established human safety data from prior use and findings from well-controlled preclinical studies demonstrating protective efficacy against mpox [185]. Concurrently, ACAM2000 has been deployed in efforts to respond to mpox outbreaks across Africa. However, the WHO recommends ACAM2000 only for specific individuals following individual risk assessments, primarily when alternative vaccines are unavailable [126].
## 5.3 LC16m8
LC16m8 (also known as LC16 or LC16 KMB) is a third-generation vaccine initially approved by Japan's Pharmaceuticals and Medical Devices Agency (PMDA) in 1975 for the indication of "prevention of smallpox" without age restrictions. In 2022, the PMDA extended its indication to include "prevention of mpox" [186]. The WHO has granted the emergency use listing (EUL) to the LC16m8 mpox vaccine, making it the second mpox vaccine endorsed by the WHO following the PHEIC declaration [187].
Derived from a natural variant of the Lister (Elstree) strain, LC16m8 exhibits high genetic concordance with the parental strain in the core genomic region [188]. Complete genome sequencing revealed that LC16m8 harbors a mutation in the immunogenic membrane protein B5R gene, which results in the lack of a full-length B5R protein. Relative to its parental strain, LC16m8 exhibits favorable biological properties characterized by reduced virulence and enhanced safety [189].
To evaluate the efficacy and immunogenicity of LC16m8, a series of studies was conducted, including comparative assessments with Dryvax or the Lister strain. In the MPXV-infected Macaca fascicularis model, LC16m8 demonstrated protective efficacy comparable to that of the Lister strain [190]. Similarly, in the BALB/c mouse model, LC16m8 and Dryvax exhibited equivalent immune responses [191]. Moreover, Iizuka et al. reported that a single dose of LC16m8 confers protection against MPXV for at least 1 year in M. fascicularis [192]. Following the global mpox outbreak, Kobiyama et al. conducted extensive immunological evaluations in mice (BALB/c, C57BL/6J, and CAST/EiJ) and NHPs, demonstrating that LC16m8 elicits robust immune responses across various preclinical models and further providing preclinical and early clinical insights into the efficacy and safety of LC16m8 against mpox [193].
In the clinical stage, a Phase I/II trial comparing the safety and immunogenicity of LC16m8 with Dryvax confirmed that the two vaccines have comparable safety profiles, with both inducing similar local and systemic reactions and no serious AEs or cardiac toxicity. Immunologically, both groups developed robust neutralizing antibodies against CPXV, MPXV, and VARV (titers > 1:40) with 180-day immune memory; LC16m8 achieved 100% vaccination success rate versus 85% for Dryvax [194]. In addition, LC16m8 exhibits favorable safety and immunogenicity in healthy adults as well as well-controlled HIV-positive populations, and even shows a degree of protective efficacy in MPXV-HIV coinfection patients. In healthy adults, LC16m8 induced seroconversion rates of 72%, 70%, and 88% against the MPXV ZR599 strain, MPXV Liberia strain, and LC16m8 strain by Day 28 (declining to 30%, 30%, and 76% by Day 168), with a 94% 14-day vaccination success rate. While 98% of participants reported mild-to-moderate adverse reactions, no serious events or mpox cases occurred, confirming favorable immunogenicity and safety [195]. An open-label randomized trial validated LC16m8's safety and immunogenicity in well-controlled HIV-positive populations, supporting its use in high-risk groups despite unconfirmed efficacy in high-incidence settings [196]. In post-exposure prophylaxis research, a small study of six close contacts of mpox cases receiving LC16m8 showed no mpox development within 21 days, mild adverse reactions, and uneventful recovery of two appropriately treated HIV-infected individuals within 28 days [197,198].
Collectively, the aforementioned data demonstrate that LC16m8 exhibits favorable immunogenicity, protective efficacy, and an acceptable safety profile, with potential utility in post-exposure prophylaxis. The WHO expert panel recommends its use in children and individuals at high risk of exposure, though it is contraindicated in pregnant women and immunocompromised individuals [199]. Ongoing clinical trials are further evaluating the mpox-preventive efficacy of LC16m8 and conducting extended safety assessments [200,201].
## 5.4 OrthopoxVac
OrthopoxVac, a fourth-generation smallpox vaccine developed by the State Research Center of Virology and Biotechnology (SRC VB) VECTOR in Russia, was officially licensed in the Russian Federation in November 2022 for immunization against VARV, MPXV, and other orthopoxvirus infections [155]. The approval of this vaccine provides a new option for addressing the global mpox epidemic.
OrthopoxVac is a live-attenuated vaccine based on the VACΔ6 strain, generated via targeted deletion of six key genes (A56R, B8R, J2R, C3L, N1L, and A35R) from the parental LIVP vaccinia virus. Maksyutov et al. employed PCR and sequencing analysis to show that after 15 passages in 4,647 cells, all viral DNA sequences remained unchanged, confirming the genomic stability of the VACΔ6 strain [202]. Notably, compared to the parental LIVP strain, the VACΔ6 strain exhibits significantly reduced reactogenicity and neurovirulence while exhibiting enhanced immunogenicity [203].
The licensing of OrthopoxVac is supported by a comprehensive portfolio of nonclinical and clinical studies demonstrating its efficacy in preventing smallpox and other orthopoxvirus infections [204]. In clinical practice, the vaccine is administered as a single 0.2 mL intradermal injection into the lateral third of the deltoid muscle. The vaccination response typically manifests as a distinct local skin reaction with progression from erythema and swelling to vesiculation, reaching maximal size between Days 8 and 10. Subsequently, the reaction resolves with scab formation, and upon scab detachment (typically by the third week), no residual scarring is observed [155].
OrthopoxVac is contraindicated in immunocompromised individuals and those receiving immunosuppressive therapy. The most frequent adverse reaction (incidence > 10%) is pain at the injection site. Less common adverse effects, observed in fewer than 1% of patients, include localized lymphadenitis, headache, malaise, fatigue, and elevated body temperature exceeding 37 • C [155].
## 5.5 Other Potential MPXV Vaccines
Although approved mpox vaccines have demonstrated effectiveness, persistent concerns remain regarding the safety profile of live-attenuated mpox vaccines, highlighting a critical opportunity for the development of novel mpox vaccine candidates.
According to the WHO Mpox Vaccine Tracker, over 40 mpox vaccine candidates are currently in development, encompassing diverse technological platforms such as mRNA, subunit, and DNA vaccines. The majority of these candidates remain in the preclinical phase. Selected mpox vaccine candidates currently in clinical evaluation are summarized in Table 2.
BNT166 and mRNA-1769 are mpox vaccines developed using mRNA technology and are currently in the Phase I/II clinical stage. BNT166 is a multivalent mRNA orthopoxvirus vaccine encoding MPXV antigens A35, B6, M1, and H3. In preclinical studies, it elicited robust antibody and T-cell responses in murine and NHP models, and conferred protection against MPXV in virus challenge experiments [205]. mRNA-1769 encodes MPXV surface proteins A35, B6, M1, and A29, and has been demonstrated to provide protection against MPXV challenge in NHP models. Notably, compared to MVA, mRNA-1769 resulted in fewer lesions and reduced viral replication, while inducing enhanced neutralizing and functional antibodies [206].
The MVA strain monkeypox live attenuated vaccine's IND application was approved by the NMPA in September 2024, making it the first mpox vaccine authorized for clinical trials in China.
This vaccine is currently in Phase I clinical trials (CTR20250074).
The replication-defective monkeypox vaccine was developed by deleting 26 gene fragments from the smallpox vaccine Tian Tan strain (VTT), rendering the vaccine unable to replicate in human cells and thereby enhancing its safety profiles [207]. Preclinical data demonstrated that this replication-defective vaccine can successfully induce cross-neutralizing activity against MPXV in mouse and rhesus monkey models [208]. JT118, a recombinant protein vaccine composed of the MPXV antigen A35 and M1, has also been approved by the NMPA for its IND application [209].
## 6 Treatments for MPXV Infection
Mpox, distinct from smallpox, is a self-limiting disease, and the majority of cases are relatively mild; therefore, supportive treatments, including symptom management and rash care, are often adopted in clinical management for mpox patients [210,211]. Notably, several specific antiviral agents initially developed for smallpox have been evaluated for their efficacy against MPXV. Herein, we systematically summarize the development history, efficacy, and safety profile of several key drugs, as well as the clinical trials conducted on mpox of these drugs (Table 4), along with the research progress of candidate compounds in the preclinical development stage.
## 6.1 Tecovirimat
Tecovirimat is a tetracyclic acylhydrazide compound exhibiting broad-spectrum antiviral activity against VARV, MPXV, and CPXV [221]. This molecule was screened from 356,240 compounds through in vitro anti-VACV activity assays via highthroughput drug screening in 2002 [222]. Further studies showed that tecovirimat induces the viral VP37 protein homodimerization and blocks the interaction of VP37 with cellular Rab9 and TIP47 proteins, preventing the formation of egress-competent forms (i.e., IEV) of orthopoxviruses (Figure 3) [213][214][215].
TPOXX, the trade name of tecovirimat, has been developed and marketed by SIGA Technologies, Inc. in collaboration with the US Biomedical Advanced Research and Development Authority (BARDA), and is available in two dosage forms: oral capsules and an IV formulation [223]. Oral tecovirimat capsules are the first FDA-approved specific therapeutic agents for VARV infections.
As VARV was eradicated worldwide in 1980, clinical trials for tecovirimat capsules were primarily conducted in accordance with the FDA Animal Rule Guideline [215,224].
Specifically, two key research components were conducted in the Phase II clinical trials to evaluate the clinical efficacy of tecovirimat capsules [225]. On the one hand, the fully effective protective dose of tecovirimat was investigated using two animal infection models: MPXV-infected NHPs (NHP/MPXV model) and rabbitpox virus (RPXV)-infected rabbits (Rabbit/RPXV model).
With survival rate as the primary efficacy endpoint, the fully effective doses derived from two key pharmacodynamic (PD) studies (Nos. AP-09-026G and SR14-008F) were 10 mg/kg in the NHP/MPXV model and 40 mg/kg in the Rabbit/RPXV model, respectively [225,226]. Meanwhile, the pharmacokinetic (PK) characteristics, various treatment initiation times, and the min- imum treatment duration of tecovirimat against virus infection using these animal models were evaluated and collected [226]. On the other hand, human dosing regimens that covered the exposure levels of a fully effective protective dose in animals were investigated in healthy volunteers [225,227]. Based on data from these two components, the effective human dosage of tecovirimat was further estimated by establishing NHP and human population pharmacokinetic (Pop PK) models [225,227].
Subsequently, a Phase III clinical study (No. SIGA-246-008) was conducted, in which researchers evaluated a dosing regimen of 600 mg twice daily for 14 consecutive days to validate tecovirimat's efficacy [228]. Comparison results showed that human exposure levels of tecovirimat were several times higher than those corresponding to the fully effective dose exposure levels in the NHP/MPXV model and the Rabbit/RPXV model [229]. Based on these PK results and safety data, the dosing regimen for tecovirimat capsules was defined as 600 mg twice daily for 14 days [212], and tecovirimat capsules were ultimately approved by the FDA for smallpox disease in 2018 [230].
Tecovirimat injection was developed based on the oral formulation and approved through the 505(b) regulatory pathway. SIGA Technologies, Inc. primarily compared the bioequivalence of the IV formulation and oral capsule formulation via clinical trial SIGA-246-IV-202 and a Pop PK model [231,232]. Overall, the IV dosing regimen of 200 mg, single infusion for 6 h, twice daily for 14 consecutive days [233], was demonstrated to be bioequivalent to TPOXX capsules in humans, supporting its market approval. It is now listed by the FDA and indicated for VARV infection (Table 5).
Recently, TPOXX has been approved for mpox indication in the EU, UK, and Japan based on efficacy data from animal models,
## TABLE 4
Main anti-MPXV drugs.
## Basic information Drug name
## Structure
## Mechanism of action
## Dosage and administration
## References
## TABLE 5
Approval information of tecovirimat.
## Dosage form Country Proprietary name Approval date Indication References
Capsule US TPOXX Jul 2018 Smallpox disease in adults and pediatric patients weighing at least 13 kg.
[230]
Capsule CA TPOXX Nov 2021 Smallpox disease in adults and pediatric patients weighing at least 13 kg.
[234]
Capsule EU Tecovirimat SIGA Jan 2022 Smallpox, mpox, cowpox, and complications due to the replication of VACV following vaccination against smallpox in adults and pediatric patients weighing at least 13 kg.
[235]
Capsule the United Kingdom (UK) Tecovirimat SIGA Jun 2022 Smallpox, mpox, cowpox, and complications due to the replication of VACV following vaccination against smallpox in adults and pediatric patients weighing at least 13 kg.
[236]
Capsule Japan TEPOXX Dec 2024 Smallpox, mpox, cowpox, and complications due to the replication of VACV following vaccination against smallpox in adults and pediatric patients weighing at least 13 kg.
[237]
Injection US TPOXX May 2022 Smallpox disease in adults and pediatric patients weighing at least 3 kg. [233] while in the US, clinical use of tecovirimat has been authorized under the EA-IND protocol. Several studies have reported that most patients with MPXV infection experienced symptom improvement during tecovirimat treatment and achieved clinical recovery after a 14-day treatment course [238][239][240][241][242][243][244][245][246]. However, it should be noted that most of the above studies lacked a control group, and thus, the definitive efficacy of tecovirimat in human mpox remains unconfirmed. The PALM007 trial was a randomized double-blind placebo-controlled study conducted in the DRC to evaluate the efficacy of tecovirimat against MPXV infection, and the results showed that the drug did not shorten lesion duration in patients infected with Clade I MPXV compared with the placebo group [220]. Similarly, interim analysis of another trial named STOMP revealed that tecovirimat did not reduce lesion duration in Clade II infections [219]. Nevertheless, secondary analyses from these trials indicated potential clinical benefits in mitigating severe disease progression and in patients administered during early infection stages [219,247]. These collective findings underline the need for further investigation into tecovirimat's therapeutic potential.
Robust safety and favorable tolerability of TPOXX have been proven. The SIGA-246-008 trial revealed that tecovirimat capsules were generally well-tolerated, with the most common treatmentemergent AEs being mild in severity, including headache (12%), nausea (5%), abdominal pain (2%), and vomiting (2%) [225]. A retrospective analysis of 196 treated patients revealed similar patterns, with the most frequent adverse events being headache (6.1%), nausea (5.1%), and abdominal pain (4.1%) [248]. The PALM007 trial found comparable safety profiles between the treatment and control groups, with AEs reported in 72.9% of tecovirimat recipients versus 70.5% of placebo recipients [220]. The results from Phase I trials of TPOXX injection reported only adverse effects such as infusion-site pain (associated with prolonged administration) and headache [231,232].
## 6.2 Cidofovir and Brincidofovir
Cidofovir (CDV; trade name: Vistide), a cytosine nucleotide analog developed by Avet Pharmaceuticals Inc., was first approved by the FDA in 1996 for the treatment of cytomegalovirus (CMV) retinitis in HIV-infected patients [249]. Beyond its original indication, CDV exhibits broad-spectrum antiviral activity against a range of DNA viruses. Recent clinical studies and case reports have demonstrated its efficacy in treating infections caused by adenovirus (AdV) in transplant recipients, HSV infections, and varicella-zoster virus (VZV) infections [250][251][252]. The antiviral mechanism of CDV involves intracellular phosphorylation to its active metabolite, CDV diphosphate (CDV-PP), which is a competitive inhibitor of viral DNA polymerase, thereby preventing viral DNA synthesis and inhibiting viral replication [253].
CDV has emerged as a promising therapeutic candidate against MPXV due to its broad-spectrum activity against DNA viruses, particularly against orthopoxviruses, both in vitro and in vivo [254][255][256][257][258]. CDV is available in two dosage forms: IV and topical. Systemic IV administration is a primary route for mpox patients and has shown potential therapeutic benefits. For example, Mondi et al. reported that four severe mpox patients in Italy treated with IV CDV exhibited rapid symptom improvement, achieving clinical recovery within 4-18 days [246]. In another study, an MPXV-HIV coinfection patient showed accelerated lesion resolution and full recovery by Day 21 under treatment with CDV injection [259]. Clinical potential of topical CDV (1% cream formulated with Beeler base as an excipient) has been demonstrated in treating mpox skin lesions. A prospective study involving 24 mpox patients showed that topical CDV reduced the median lesion healing time to 12 versus 18 days in the supportive treatment group, and significantly decreased the positive rate of skin lesions (10% positive rate in the topical CDV group versus 62.5% positive rate in the supportive treatment group) [217]. It should be noted that the adverse effects and limited clinical data restrict the clinical application of CDV. Specifically, CDV exhibited dose-dependent nephrotoxicity, which leads to proteinuria, glucosuria, elevated serum creatinine, and so forth [249,260]. Consequently, renal function must be closely monitored during treatment. In addition, CDV has also been associated with other possible adverse reactions, including nausea, vomiting, neutropenia, elevated alanine aminotransferase levels, and localized burning sensation (with topical administration) [246,261,262].
To overcome the shortcomings of CDV, a long-chain lipidconjugated prodrug of CDV, namely Brincidofovir (trade name: TEMBEXA; CMX001, BCV), was developed by Chimerix, Inc. in collaboration with BARDA. This structural modification reduces nephrotoxicity and markedly enhances oral bioavailability compared with CDV [263].
There are two dosage forms, including tablets and oral suspension, which were approved for the indication of smallpox by the FDA in 2021. Similar to tecovirimat, BCV was also developed under the FDA Animal Rules, and its efficacy was established using two animal models: the Rabbit/RPXV model and ectromelia virus (ECTV)-infected mice (Mice/ECTV) model. With survival rate as the primary efficacy endpoint, the fully effective doses from two PD studies (Nos. CMX001-VIR-106 and CMX001-VIR-044) were 20/5/5 mg/kg (administered every 48 h for three doses) in the Rabbit/RPXV model and 10/5/5 mg/kg (administered every 48 h for three doses) in the Mice/ECTV model, respectively. Under these dosing regimens, lower mortality was observed in the treatment group compared with the placebo group [264].
According to the PK/PD data from animal studies, as well as PK parameters from clinical studies of healthy and nonorthopoxviral-infected subjects following oral administration of BCV tablets or suspension, Pop PK modeling was used to determine a final weight-based dosing regimen. For general adult patients (≥ 48 kg), the recommended dose for both oral suspension and tablet formulations is 200 mg once weekly for 2 weeks (administered on Days 1 and 8) [264]. During the 2022 global mpox outbreak, BCV was included in the WHO list of potential therapeutic agents. Similarly, in the US, BCV became accessible for mpox treatment via an Investigational New Drug (e-IND) application [210]. Recently, a Phase III, randomized, doubleblind, placebo-controlled study, namely "MOSA," is underway in Africa to assess the safety and therapeutic efficacy of BCV in mpox patients [265].
Regarding safety, BCV reduces the risk of nephrotoxicity compared to CDV, but may induce abnormal liver function, potentially affecting treatment tolerance in some patients [39,263]. Thus, liver function should be regularly assessed during BCV treatment to prevent irreversible damage [2]. Other adverse reac-tions are generally mild to moderate. In Phase II and III clinical trials of BCV, common adverse reactions included diarrhea (8%), nausea (5%), vomiting (4%), and abdominal pain (3%) [264]. Additionally, the drug exhibits embryotoxicity and teratogenicity in animal studies; thus, it should be used cautiously in pregnant women and neonates [264].
## 6.3
Other Candidate Drugs
## 6.3.1 Trifluridine
Trifluridine (TFT) is a thymidine nucleoside analog that exerts antiviral activity by inhibiting viral DNA synthesis. TFT exhibits potent inhibitory efficacy against MPXV infection in both cutaneous and ocular cell models, with additional activity against tecovirimat-resistant virus strains [266]. It is reported that MPXV infection may give rise to various ophthalmic complications such as conjunctivitis and keratitis [266,267]. Studies have shown that three of five patients who received tecovirimat-TFT combination therapy showed improvement in their ophthalmic symptoms [268,269]. Due to its poor corneal penetration and extremely low systemic absorption, systemic adverse reactions to TFT are rare, and common local adverse reactions include corneal inflammation, eyelid edema, and burning sensation [270].
## 6.3.2 Tecovirimat Analog
Given the low solubility of tecovirimat, which may impact its pharmacological properties, researchers have focused on improving oral bioavailability through structural modifications.
NIOCH-14, designated by the Novosibirsk Institute of Organic Chemistry, is a prodrug of tecovirimat that can be metabolized in vivo to exert anti-orthopoxvirus activity. In the mouse model, NIOCH-14 demonstrated 1.9-fold higher absolute oral bioavailability than tecovirimat (22.8% vs. 12.1% at 50 mg/kg). Furthermore, NIOCH-14 exhibited an excellent safety profile. Oral administration at 5 g/kg in mice caused no mortality or signs of toxicity, and both single-dose and repeated-dose studies in mice and rats showed no treatment-related adverse effects on hematological or histopathological parameters [271]. NIOCH-14 is currently in the Phase I clinical trial stage (NCT05976100 ASC10, an oral prodrug of molnupiravir developed by Ascletis Pharma, Inc., is converted to ASC10-A-TP in vivo, which binds to RNA-dependent RNA polymerase and induces mutations in viral RNA, thereby inhibiting viral replication. The Phase I trial in healthy Chinese participants revealed that ASC10 was safe and well-tolerated, exhibiting dose-proportional exposure and minimal food effects [276]. A Phase Ib clinical trial is currently underway in the US for the treatment of mpox patients [277].
## 6.3.4 Vaccinia Immunoglobulin Intravenous
Vaccinia immunoglobulin intravenous (VIGIV) is a polyclonal antibody preparation derived from the plasma of smallpox vaccine recipients, containing high-titer antibodies against VACV [278]. It exerts its effect by neutralizing viral particles and inhibiting viremia [262]. In 2005, VIGIV was approved for the treatment of complications related to smallpox vaccination [279,280]. During the 2022 global mpox outbreak, based on the antigenic cross-reactivity within the orthopoxvirus genus, the US CDC implemented an extended access protocol allowing the use of VIGIV for the treatment of other orthopoxvirus infections, including MPXV [281,282]. Accordingly, the US CDC recommended VIGIV for mpox patients with immune deficiencies, such as HIV-infected individuals with CD4 + T-cell counts below 350. However, further research is needed to verify its efficacy in MPXV infection [260].
## 6.3.5 Other Reagents
Several compounds have also been reported to exhibit potential efficacy against MPXV infection. For instance,
## 7 Conclusion and Prospects
Since its isolation nearly 70 years ago, MPXV has diverged into distinct lineages with marked differences in virulence and transmission. Clade I (Ia/Ib) exhibits higher pathogenicity than Clade II (IIa/IIb), driven by genetic variations modulating host-virus interactions [33]. While Clade I-associated mortality (historically ∼10% in Africa) has declined due to viral virulence attenuation and improving clinical management, its persistent circulation (e.g., Clade Ia in the DRC) and Clade Ib's retained PHEIC status underscore ongoing global risks, amplified by cross-border spread via travel. Though Clade IIb currently has low global prevalence, it previously caused a global outbreak and shows recent resurgence trends in selected regions (e.g., Sierra Leone). Notably, it disproportionately affects MSM, with mortality risks significantly elevated among those with HIV coinfection. In Uganda, 55% of MPXV-related fatalities occurred in HIV-positive individuals [46]. These divergent epidemiological patterns underscore the urgent need to advance understanding of pathogenesis, diagnostics, vaccines, and therapeutics.
To address these priorities, a systematic grasp of MPXV's fundamental virological characteristics is critical. Here, we review key features of virion morphology, genome organization, and lifecycle; however, significant gaps in virological understanding remain. For instance, the majority of studies on the poxvirus lifecycle are based on VACV, raising questions about whether MPXV employs unique viral mechanisms yet to be identified. Future research should prioritize functional conservation assessment between MPXV-VACV homologous proteins through expanded homology-based screening, coupled with systematic dissection of MPXV-specific host interaction mechanisms. This will support the development of targeted antivirals and vaccines via novel strategies.
Beyond virological characterization, effective outbreak control hinges on robust diagnostic capacity. While the WHO has reported declining mpox cases in African regions, it emphasizes that actual cases may be undercounted due to gaps in diagnosis and surveillance [46]. Pragmatically, scarcity of diagnostic reagents and lack of POC rapid testing technologies hinder timely mpox detection, exacerbating transmission. Although RT-qPCR remains the gold standard, its reliance on specialized infrastructure limits its utility in resource-poor settings. Emerging isothermal amplification methods (e.g., LAMP) offer promising decentralized alternatives, with high sensitivity, operational simplicity, and minimal equipment requirements, supported by combinatorial diagnostic approaches. Accelerated clinical validation across diverse epidemiological contexts and equitable access initiatives are now essential to translate these technological advances from bench to field in high-need regions.
Mpox vaccines serve as crucial public health tools for curbing the spread of MPXV, with their significance becoming particularly prominent amid the global spread of the epidemic in recent years. Currently, four vaccines have been approved for mpox prevention: ACAM2000 and MVA-BN from the US, LC16m8 from Japan, and OrthopoxVac from Russia. ACAM2000, a second-generation replicating smallpox vaccine, exhibits efficacy comparable to MVA-BN but carries a higher risk of adverse reactions, including myo-/pericarditis and severe vaccinial skin reaction [153,165]. Consequently, it is only recommended when no alternative vaccines are available. In contrast, both MVA-BN and LC16m8 are prioritized by the WHO [126]. MVA-BN-a non-replicating vaccine-stands out for its superior safety profile, making it suitable for immunocompromised individuals, and LC16m8 is approved for children over 1 year old. While these four vaccines mark a transition to proactive intervention, technical limitations and global health equity challenges persist. The landscape of mpox vaccine development is evolving toward next-generation technologies and rational design strategies to address the drawbacks of live-attenuated vaccines. For example, China's NMPA granted clinical approval to a replication-deficient mpox vaccine derived from the VTT, which may provide a safe alternative for high-risk populations in the future [207]. In addition, two mpox vaccines (mRNA-1769 and BNT166) based on the increasingly promising mRNA technology are currently in Phase I/II clinical trials [205,206].
Antiviral treatments play a vital role in the control of mpox and smallpox, particularly with drugs such as tecovirimat, CDV, and brincidofovir. Although tecovirimat is approved for mpox in the EU, UK, and Japan, its therapeutic efficacy still needs further research and evaluation. The PALM007 study, the first randomized controlled trial, was conducted to evaluate tecovirimat's safety and efficacy against MPXV infection [220]. Results showed no statistical significance for the primary endpoint: lesion resolution in Clade I mpox when comparing tecovirimat to placebo. Similarly, interim analysis of the STOMP trial (NCT05534984) revealed that oral tecovirimat did not reduce the time to lesion resolution compared to placebo in adults with mild-to-moderate Clade II mpox [219]. It is worth noting that these studies may have used less stringent eligibility criteria due to humanitarian considerations, so as to ensure that the maximum number of mpox patients receive treatment. For example, lesion counts of mpox patients in the PALM007 trial ranged from 1 to 10,264, potentially influencing the efficacy of tecovirimat due to variable disease severity. Nevertheless, the PALM007 trial demonstrated a strong safety profile for tecovirimat among all patients, and subgroup analysis suggested that early treatment or use in severe cases might offer clinical benefits [220]. These findings align with other reports. For instance, Aldred et al. observed that HIVinfected patients treated with tecovirimat within 7 days had lower disease progression rates compared with those treated later or not at all [245]. Similarly, Karmarkar et al. found that in patients with severe disease, administering tecovirimat within 5 days of symptom onset resulted in faster symptomatic improvement compared with no tecovirimat treatment (-5.5 days; p = 0.04) [286]. Thus, additional clinical trials under refined research conditions are needed to further clarify the efficacy of tecovirimat against MPXV in humans, considering patient characteristics, disease course, health status, efficacy endpoint, and impacts of mpox genetic variants. Furthermore, combination therapy may improve MPXV treatment outcomes. For instance, in severe mpox patients with advanced HIV (CD4 < 200 cells/µL), early antiretroviral therapy, along with multipronged mpox treatment (e.g., tecovirimat combined with cidofovir or brincidofovir), may enhance efficacy, despite the risk of hepatotoxicity and nephrotoxicity, necessitating monitoring [287,288].
To sum up, we systematically review recent advancements of MPXV research from five different perspectives, hoping that researchers gain insights from reviewing this article and develop more effective diagnostic methods, vaccines, as well as treatments to control and combat mpox.
## References
1. Ictv (2024) "Virus Taxonomy"
2. Rizk, Lippi, Henry et al. (2022) "Prevention and Treatment of Monkeypox" *Drugs*
3. Patel, Bilinska, Tam (2022) "Clinical Features and Novel Presentations of Human Monkeypox in a Central London Centre During the 2022 Outbreak: Descriptive Case Series" *BMJ*
4. Jezek, Szczeniowski, Paluku et al. (1987) "Human Monkeypox: Clinical Features of 282 Patients" *Journal of Infectious Diseases*
5. Titanji, Hazra, Zucker (2024) "Mpox Clinical Presentation, Diagnostic Approaches, and Treatment Strategies: A Review" *JAMA*
6. Tarín-Vicente, Alemany, Agud-Dios (2022) "Clinical Presentation and Virological Assessment of Confirmed Human Monkeypox Virus Cases in Spain: A Prospective Observational Cohort Study" *Lancet*
7. Ahmed, Mohamed, Dabou (2023) "Monkeypox (Mpox) in Immunosuppressed Patients" *F1000Res*
8. Islam, Mumin, Haque (2023) "Monkeypox Virus (MPXV): A Brief Account of Global Spread, Epidemiology, Virology, Clinical Features, Pathogenesis, and Therapeutic Interventions"
9. Islam (2022) "Old Enemy With a New Face: Re-Emerging Monkeypox Disease-An Update"
10. Islam (2022) "An Estimation of Five-Decade Long Monkeypox Case Fatality Rate: Systematic Review and Meta-Analysis"
11. Al-Musa, Chou, Labere (2022) "The Resurgence of a Neglected Orthopoxvirus: Immunologic and Clinical Aspects of Monkeypox Virus Infections Over the Past Six Decades" *Clinical Immunology*
12. Dewitt, Polk, Williamson (2022) "Global Monkeypox Case Hospitalisation Rates: A Rapid Systematic Review and Meta-analysis"
13. Who (2024) "Mpox Outbreak"
14. Who (2022) "Monkeypox: Experts Give Virus Variants New Names"
15. Happi, Adetifa, Mbala (2022) "Urgent Need for a Non-Discriminatory and Non-Stigmatizing Nomenclature for Monkeypox Virus" *PLoS Biology*
16. Soleimani, Motamed (2024) "An Overview of Monkey Pox Disease as a Current Important Disease in the World" *Archives of Razi Institute*
17. Callaby, Belfield, Otter (2025) "Mpox: Current Knowledge and Understanding-A Scoping Review" *FEMS Microbiology Review*
18. Angelo, Smith, Camprubí-Ferrer (2023) "Epidemiological and Clinical Characteristics of Patients With Monkeypox in the GeoSentinel Network: A Cross-Sectional Study" *Lancet Infectious Diseases*
19. Suvvari, Ghosh, Lopinti et al. (2023) "Hematological Manifestations of Monkeypox Virus (MPOX) and Impact of Human MPOX Disease on Blood Donation-What We Need to Know?" *New Microbes and New Infections*
20. Brosius, Vakaniaki, Mukari (2025) "Epidemiological and Clinical Features of Mpox During the Clade Ib Outbreak in South Kivu, Democratic Republic of the Congo: A Prospective Cohort Study" *Lancet*
21. Hajjo, Abusara, Sabbah et al. (2025) "Advancing the Understanding and Management of Mpox: Insights Into Epidemiology, Disease Pathways, Prevention, and Therapeutic Strategies" *BMC Infectious Diseases*
22. Ladnyj, Ziegler, Kima (1972) "A Human Infection Caused by Monkeypox Virus in Basankusu Territory, Democratic Republic of the Congo" *Bulletin of the World Health Organization*
23. Who (2025) "Global Mpox Trends"
24. Breman, Kalisa, Steniowski et al. (1980) "Human Monkeypox, 1970-79" *Bulletin of the World Health Organization*
25. Hutin, Williams, Malfait (2001) "Outbreak of Human Monkeypox, Democratic Republic of Congo, 1996 to 1997" *Emerging Infectious Diseases*
26. Cdc (1996) "Human Monkeypox-Kasai Oriental, Democratic Republic of Congo"
27. Bunge, Hoet, Chen (2022) "The Changing Epidemiology of Human Monkeypox-A Potential Threat? A Systematic Review" *PLOS Neglected Tropical Diseases*
28. Whitehouse, Bonwitt, Hughes (2021) "Clinical and Epidemiological Findings From Enhanced Monkeypox Surveillance in Tshuapa Province, Democratic Republic of the Congo During 2011-2015" *Journal of Infectious Diseases*
29. Ogoina, Damon, Nakoune (2023) "Clinical Review of Human Mpox" *Clinical Microbiology and Infection*
30. (2003) "Multistate Outbreak of Monkeypox-Illinois" *MMWR Morbidity and Mortality Weekly Report*
31. Reed, Melski, Graham (2004) "The Detection of Monkeypox in Humans in the Western Hemisphere" *New England Journal of Medicine*
32. Shchelkunova, Shchelkunov (2022) "Smallpox, Monkeypox and Other Human Orthopoxvirus Infections" *Viruses*
33. Andrei, Snoeck (2023) "Differences in Pathogenicity Among the Mpox Virus Clades: Impact on Drug Discovery and Vaccine Development"
34. Yinka-Ogunleye, Aruna, Dalhat (2019) "Outbreak of Human Monkeypox in Nigeria in 2017-18: A Clinical and Epidemiological Report"
35. Ndodo, Ashcroft, Lewandowski (2023) "Distinct Monkeypox Virus Lineages Co-Circulating in Humans Before 2022" *Nature Medicine*
36. Gigante, Korber, Seabolt (2022) "Multiple Lineages of Monkeypox Virus Detected in the United States, 2021-2022" *Science*
37. Who (2022) "WHO Director-General's Statement at the Press Conference Following IHR Emergency Committee Regarding the Multi-Country Outbreak of Monkeypox-23"
38. Who (2024) "Surveillance, Case Investigation and Contact Tracing for Mpox (Monkeypox)"
39. Adler, Gould, Hine (2022) "Clinical Features and Management of Human Monkeypox: A Retrospective Observational Study in the UK" *Lancet Infectious Diseases*
40. Saalbach (2024) "Treatment and Vaccination for Smallpox and Monkeypox" *Advances in Experimental Medicine and Biology*
41. Hasan, Mohanasundaram, Bhattacharya et al. (2023) "Exploring the Relationship Between the Global Health Security Index and Monkeypox: An Analysis of Preparedness and Response Capacities" *Global Health*
42. Who (2024) "Mpox-African Region"
43. Katoto, Muttamba, Bahizire (2024) "Shifting Transmission Patterns of Human Mpox in South Kivu, DR Congo" *Lancet Infectious Diseases*
44. Nizigiyimana, Ndikumwenayo, Houben (2024) "Epidemiological Analysis of Confirmed Mpox Cases" *Euro Surveillance*
45. Mcquiston, Luce, Kazadi (2024) "U.S. Preparedness and Response to Increasing Clade I Mpox Cases in the Democratic Republic of the Congo-United States, 2024" *Mmwr Morbidity and Mortality Weekly Report*
46. Who (2005) "Emergency Committee Regarding the Upsurge of Mpox 2024"
47. Zhao, Wang, Zhao (2022) "The First Imported Case of Monkeypox in the Mainland of China-Chongqing Municipality" *China CDC Weekly*
48. Ren, Li, Bai (2023) "The Epidemiological Characteristics of Mpox Cases-China" *China CDC Weekly*
49. Hu, Zheng, Ruan (2025) "Clinical, Epidemiological, Virological Characteristics and Outcomes of 286 Patients Infected with Monkeypox Virus in China" *Allergy*
50. Song, Yan, Xu (2025) "Complete Genome Sequence Analysis of the First Imported Mpox Virus Clade Ib Variant in China" *Pathogens*
51. Pan, Ge, Zhu (2025) "The First Reported Case of Monkeypox Virus Clade Ib Infection in China-Clinical Presentation and Epidemiologic Implications" *International Journal of Infectious Diseases*
52. Philpott, Hughes, Alroy (2022) "Epidemiologic and Clinical Characteristics of Monkeypox Cases-United States" *Mmwr Morbidity and Mortality Weekly Report*
53. Girometti, Byrne, Bracchi (2022) "Demographic and Clinical Characteristics of Confirmed Human Monkeypox Virus Cases in Individuals Attending a Sexual Health Centre in London, UK: An Observational Analysis" *Lancet Infectious Diseases*
54. Gelderblom, Madeley (2018) "Rapid Viral Diagnosis of Orthopoxviruses by Electron Microscopy: Optional or a Must?" *Viruses*
55. Hong, Huang, Zhang (2024) "Molecular Architecture of Monkeypox Mature Virus" *Cell Discovery*
56. Huang, Mu, Wang (2022) "Monkeypox: Epidemiology, Pathogenesis, Treatment and Prevention" *Signal Transduction and Targeted Therapy*
57. Buller, Palumbo (1991) "Poxvirus Pathogenesis" *Microbiological Reviews*
58. Lu, Wu, Jiang et al. (2022) "The Current Emergence of Monkeypox: The Recurrence of Another Smallpox?" *Biosafety and Health*
59. Odom, Hendrickson, Lefkowitz (2009) "Poxvirus Protein Evolution: Family Wide Assessment of Possible Horizontal Gene Transfer Events" *Virus Research*
60. Alakunle, Moens, Nchinda et al. (2020) "Monkeypox Virus in Nigeria: Infection Biology, Epidemiology, and Evolution"
61. Kumar, Guruparan, Karuppanan et al. (2024) "Comprehensive Insights Into Monkeypox (Mpox): Recent Advances in Epidemiology, Diagnostic Approaches and Therapeutic Strategies" *Pathogens*
62. Stern, Dales (1974) "Biogenesis of Vaccinia: Concerning the Origin of the Envelope Phospholipids" *Virology*
63. Blough, Tiffany (1973) "Lipids in Viruses" *Advances in Lipid Research*
64. Boulter (1969) "Protection Against Poxviruses" *Proceedings of the Royal Society of Medicine*
65. Payne (1980) "Significance of Extracellular Enveloped Virus in the In Vitro and In Vivo Dissemination of Vaccinia" *Journal of General Virology*
66. Payne, Kristensson (1985) "Extracellular Release of Enveloped Vaccinia Virus From Mouse Nasal Epithelial Cells In Vivo" *Journal of General Virology*
67. Thornton (1980) "The Antigenicity and Immunogenicity of the Intracellular and Extracellular Forms of Vaccinia Virus. I: The Production of High-Titre Vaccinia Extracellular Virus and Its Antigenicity After Inactivation" *British Journal of Experimental Pathology*
68. Lum, Torres-Ruesta, Tay (2022) "Monkeypox: Disease Epidemiology, Host Immunity and Clinical Interventions" *Nature Reviews Immunology*
69. Shchelkunov, Totmenin, Safronov (2002) "Analysis of the Monkeypox Virus Genome" *Virology*
70. Shen-Gunther, Cai, Wang (2025) "Genomic and Antigenic Differences Between Monkeypox Virus and Vaccinia Vaccines: Insights and Implications for Vaccinology" *International Journal of Molecular Sciences*
71. Massung, Liu, Qi (1994) "Analysis of the Complete Genome of Smallpox Variola Major Virus Strain Bangladesh-1975" *Virology*
72. Shchelkunov, Totmenin, Loparev (2000) "Alastrim Smallpox Variola Minor Virus Genome DNA Sequences" *Virology*
73. Alakunle, Kolawole, Diaz-Cánova (2024) "A Comprehensive Review of Monkeypox Virus and Mpox Characteristics" *Frontiers in Cellular and Infection Microbiology*
74. Lim, Roberts, Moso (2023) "Mpox Diagnostics: Review of Current and Emerging Technologies" *Journal of Medical Virology*
75. Esteban, Hutchinson (2011) "Genes in the Terminal Regions of Orthopoxvirus Genomes Experience Adaptive Molecular Evolution" *BMC Genomics*
76. Lu, Xing, Wang (2023) "Mpox (Formerly Monkeypox): Pathogenesis, Prevention, and Treatment" *Signal Transduction and Targeted Therapy*
77. Assarsson, Greenbaum, Sundström (2008) "Kinetic Analysis of a Complete Poxvirus Transcriptome Reveals an Immediate-Early Class of Genes" *PNAS*
78. Miranda, Borges, Campos (2017) "Serologic and Molecular Evidence of Vaccinia Virus Circulation Among Small Mammals From Different Biomes, Brazil" *Emerging Infectious Diseases*
79. Manes, Estep, Mottaz (2008) "Comparative Proteomics of human Monkeypox and Vaccinia Intracellular Mature and Extracellular Enveloped Virions" *Journal of Proteome Research*
80. Mitjà, Ogoina, Titanji (2023) "Monkeypox"
81. Locker, Kuehn, Schleich (2000) "Entry of the Two Infectious Forms of vaccinia Virus at the Plasma Membane Is Signaling-Dependent for the IMV but Not the EEV" *Molecular Biology of the Cell*
82. Moss (2016) "Membrane Fusion During Poxvirus Entry" *Seminars in Cell & Developmental Biology*
83. Vanderplasschen, Hollinshead, Smith (1998) "Intracellular and Extracellular Vaccinia Virions Enter Cells by Different Mechanisms" *Journal of General Virology*
84. Zhang, Ma, Yuan et al. (2025) "Recent Advances and Perspectives in Therapeutics for Mpox" *Bioorganic & Medicinal Chemistry Letters*
85. Senkevich, Ojeda, Townsley et al. (2005) "Poxvirus Multiprotein Entry-Fusion Complex" *PNAS*
86. Das, Bhattarai, Karn et al. (2024) "Computational Investigations of Potential Inhibitors of Monkeypox Virus Envelope Protein E8 Through Molecular Docking and Molecular Dynamics Simulations" *Scientific Reports*
87. Shi, He, Song (2022) "Kinetic and Structural Aspects of Glycosaminoglycan-Monkeypox Virus Protein A29 Interactions Using Surface Plasmon Resonance" *Molecules*
88. Hsiao, Chung, Chang (1999) "Vaccinia Virus Envelope D8L Protein Binds to Cell Surface Chondroitin Sulfate and Mediates the Adsorption of Intracellular Mature Virions to Cells" *Journal of Virology*
89. Hughes, Goldstein, Pohl (2014) "A Highly Specific Monoclonal Antibody Against Monkeypox Virus Detects the Heparin Binding Domain of A27" *Virology*
90. Zheng, Duan, Huang (2025) "Discovery of a Heparan Sulfate Binding Domain in Monkeypox Virus H3 as an Anti-Poxviral Drug Target Combining AI and MD Simulations"
91. Moss (2012) "Poxvirus Cell Entry: How Many Proteins Does It Take?" *Viruses*
92. Gong, Wang, Chuai et al. (2022) "Monkeypox Virus: A Re-Emergent Threat to Humans" *Virologica Sinica*
93. Schin, Diesterbeck, Moss (2021) "Insights Into the Organization of the Poxvirus Multicomponent Entry-Fusion Complex From Proximity Analyses in Living Infected Cells" *Journal of Virology*
94. Wagenaar, Ojeda, Moss (2008) "Vaccinia Virus A56/K2 Fusion Regulatory Protein Interacts With the A16 and G9 Subunits of the Entry Fusion Complex" *Journal of Virology*
95. Wolfe, Moss (2011) "Interaction Between the G3 and L5 Proteins of the Vaccinia Virus Entry-Fusion Complex" *Virology*
96. Chang, Shih, Tang et al. (2012) "Vaccinia Mature Virus Fusion Regulator A26 Protein Binds to A16 and G9 Proteins of the Viral Entry Fusion Complex and Dissociates From Mature Virions at Low pH" *Journal of Virology*
97. Brown, Senkevich, Moss (2006) "Vaccinia Virus F9 Virion Membrane Protein Is Required for Entry but Not Virus Assembly, in Contrast to the Related L1 Protein" *Journal of Virology*
98. Mallardo, Schleich, Locker (2001) "Microtubule-Dependent Organization of Vaccinia Virus Core-Derived Early mRNAs Into Distinct Cytoplasmic Structures" *Molecular Biology of the Cell*
99. Yang, Bruno, Martens et al. (2010) "Simultaneous High-Resolution Analysis of vaccinia Virus and Host Cell Transcriptomes by Deep RNA Sequencing" *PNAS*
100. Moss (2013) "Poxvirus DNA Replication" *Cold Spring Harbor Perspectives in Biology*
101. Yang, Reynolds, Martens et al. (2011) "Expression Profiling of the Intermediate and Late Stages of Poxvirus Replication" *Journal of Virology*
102. Sivan, Weisberg, Americo et al. (2016) "Retrograde Transport From Early Endosomes to the Trans-Golgi Network Enables Membrane Wrapping and Egress of Vaccinia Virus Virions" *Journal of Virology*
103. Alzhanova, Hruby (2006) "A Trans-Golgi Network Resident Protein, Golgin-97, Accumulates in Viral Factories and Incorporates Into Virions During Poxvirus Infection" *Journal of Virology*
104. Sodeik, Doms, Ericsson (1993) "Assembly of Vaccinia Virus: Role of the Intermediate Compartment Between the Endoplasmic Reticulum and the Golgi Stacks" *Journal of Cell Biology*
105. Blasco, Moss (1992) "Role of Cell-Associated Enveloped Vaccinia Virus in Cell-to-Cell Spread" *Journal of Virology*
106. Smith, Law (2004) "The Exit of Vaccinia Virus From Infected Cells" *Virus Research*
107. Arakawa, Cordeiro, Schleich et al. (2007) "The Release of Vaccinia Virus From Infected Cells Requires RhoA-mDia Modulation of Cortical Actin" *Cell Host & Microbe*
108. Madihi, Benani (2025) "A Comprehensive Review of Current Diagnostic Techniques for Monkeypox Virus Detection" *Biologicals*
109. Kinganda-Lusamaki, Baketana, Ndomba-Mukanya (2023) "Use of Mpox Multiplex Serology in the Identification of Cases and Outbreak Investigations in the Democratic Republic of the Congo (DRC)" *Pathogens*
110. Jiang, Xu, Guan (2025) "A Review of Mpox: Biological Characteristics, Epidemiology, Clinical Features, Diagnosis, Treatment, and Prevention Strategies" *Exploration*
111. Zhou, Chen (2023) "Mpox: A Review of Laboratory Detection Techniques" *Archives of Virology*
112. Yigci, Ergönül, Tasoglu (2025) "Mpox Diagnosis at POC" *Trends in Biotechnology*
113. Yang, Guo, Li (2023) "Development of Highly Accurate Digital PCR Method and Reference Material for Monkeypox Virus Detection" *Analytical and Bioanalytical Chemistry*
114. Specchiarello, Carletti, Matusali (2023) "Development and Validation of a Nanoplate-Based Digital PCR Assay for Absolute MPXV Quantification" *Journal of Virological Methods*
115. Huang, Xiao, Jia (2023) "Loop-Mediated Isothermal Amplification Combined With Lateral Flow Biosensor for Rapid and Sensitive Detection of Monkeypox Virus" *Frontiers in Public Health*
116. Aslan, Seymour, Brickner (2025) "A Label-free Optical Biosensor-Based Point-of-Care Test for the Rapid Detection of Monkeypox Virus" *Biosensors & Bioelectronics*
117. Feng, Xue, Cui (2022) "Development of a Loop-Mediated Isothermal Amplification Method for Rapid and Visual Detection of Monkeypox Virus" *Microbiology Spectrum*
118. Davi, Kissenkötter, Faye (2019) "Recombinase Polymerase Amplification Assay for Rapid Detection of Monkeypox Virus" *Diagnostic Microbiology and Infectious Disease*
119. Farlow, Ichou, Huggins et al. (2010) "Comparative Whole Genome Sequence Analysis of Wild-Type and Cidofovir-resistant Monkeypoxvirus" *Virology Journal*
120. Silva, Kohl, Pena et al. (2023) "Clinical and Laboratory Diagnosis of Monkeypox (Mpox): Current Status and Future Directions"
121. Sui, Xu, Liu et al. (2022) "CRISPR-Cas12a-Based Detection of Monkeypox Virus" *Journal of Infection*
122. Zhao, Hu, Fan (2023) "Rapid and Sensitive One-Tube Detection of Mpox Virus Using RPA-Coupled CRISPR-Cas12 Assay" *Cell Reports Methods*
123. Grossegesse, Stern, Hofmann (2023) "Serological Methods for the Detection of Antibodies Against Monkeypox Virus Applicable for Laboratories With Different Biosafety Levels" *Journal of Medical Virology*
124. Mccollum, Damon (2014) "Human Monkeypox" *Clinical Infectious Diseases*
125. Tiwari, Adhikari, Kaya (2023) "Monkeypox Outbreak: Wastewater and Environmental Surveillance Perspective" *Science of the Total Environment*
126. Who, Mpox (2024)
127. Find (2025) "FIND Maintains a Searchable Directory of Diagnostic Tests for Various Diseases Under the DxConnect Umbrella"
128. Daan (2024) "The Mpox Reagent (RT-qPCR)"
129. Nmpa (2025) "The Mpox Reagent (RT-qPCR)"
130. Biogerm (2024) "The Mpox Reagent (RT-qPCR)"
131. Sansure (2024) "The Mpox Reagent (RT-qPCR)"
132. Cabanillas, Murdaca, Guemari (2024) "Monkeypox 2024 Outbreak: Fifty Essential Questions and Answers" *Allergy*
133. Branda, Romano, Ciccozzi (2024) "Mpox: An Overview of Pathogenesis, Diagnosis, and Public Health Implications" *Journal of Clinical Medicine*
134. Gootenberg, Abudayyeh, Kellner et al. (2018) "Multiplexed and Portable Nucleic Acid Detection Platform With Cas13, Cas12a, and Csm6" *Science*
135. Broughton, Deng, Yu (2020) "CRISPR-Cas12-Based Detection of SARS-CoV-2" *Nature Biotechnology*
136. Hu, Wei, Guo (2025) "A Single-Tube Two-Step MIRA-CRISPR/Cas12b Assay for the Rapid Detection of Mpox Virus" *Viruses*
137. Zhang, Yu, Yin (2025) "An Ultrasensitive and Specific CRISPR-Cas13a-Mediated Point-of-Care Assay for Monkeypox Detection and PCR-Based Clade Detection" *Infectious Diseases of Poverty*
138. Guo, Shan, Hu (2025) "Rapid Visual Detection of Monkeypox Virus by One-Step LAMP-CRISPR/Cas12b Assay" *Virology Journal*
139. Li, Wilkins, Mccollum (2017) "Evaluation of the GeneXpert for Human Monkeypox Diagnosis" *American Journal of Tropical Medicine and Hygiene*
140. Cavuto, Malpartida-Cardenas, Pennisi (2024) "Portable Molecular Diagnostic Platform for Rapid Point-of-Care Detection of Mpox and Other Diseases" *Nature Communications*
141. Ishara-Nshombo, Somasundaran, Romero-Ramirez (2025) "Diagnostic Accuracy of 3 Mpox Lateral Flow Assays for Antigen Detection, Democratic Republic of the Congo and United Kingdom" *Emerging Infectious Diseases*
142. Liang, Chao, Chao (2024) "Monkeypox Virus A29L Protein as the Target for Specific Diagnosis and Serological Analysis" *Applied Microbiology and Biotechnology*
143. Ye, Lei, Xu et al. (2023) "Gold-Based Paper for Antigen Detection of Monkeypox Virus" *Analyst*
144. Kurosawa, Ozawa, Ozawa (2025) "Development and Clinical Evaluation of a MPXV Antigen-Detecting Rapid Diagnostic Test" *Journal of Virological Methods*
145. Mazzotta, Nozza, Lanini (2024) "Clinical and Laboratory Predictors of Mpox Severity and Duration: An Italian Multicentre Cohort Study (Mpox-Icona)"
146. Islam, Kumar, Sharma et al. (2024) "Wastewater-Based Surveillance of Mpox (Monkeypox): An Early Surveillance Tool for Detecting Hotspots" *Current Pollution Reports*
147. Gul, Liu, Yuan (2022) "Current and Perspective Sensing Methods for Monkeypox Virus" *Bioengineering*
148. Who (2016) "Smallpox Vaccines"
149. Ema (2022) "Summary of Product (IMVANEX) Characteristics"
150. Imvamune "Product Monograph Including Patient Medication Information (IMVAMUNE, 2025)"
151. Fda (2025) "JYNNEOS"
152. Fda (2024) "ACAM2000"
153. Pmda (2022) "LC16m8"
154. (2023) "Approves of Mpox Vaccine (OrthopoxVac)"
155. Who (2024) "WHO Prequalzifies the First Vaccine Against Mpox"
156. Volz, Sutter (2017) "Modified Vaccinia Virus Ankara: History, Value in Basic Research, and Current Perspectives for Vaccine Development" *Advances in Virus Research*
157. Herrlich, Mayr (1954) "Comparative Experimental Works on Cow Pox Virus Vaccines" *Archiv Fur Hygiene Und Bakteriologie*
158. Stickl, Hochstein-Mintzel (1971) "Intracutaneous Smallpox Vaccination With a Weak Pathogenic Vaccinia Virus" *Munchener Medizinische Wochenschrift*
159. Reina, Iglesias (2023) "Vaccines Against Monkeypox" *Medicina Clínica*
160. Volkmann, Williamson, Weidenthaler (2021) "The Brighton Collaboration Standardized Template for Collection of Key Information for Risk/Benefit Assessment of a Modified Vaccinia Ankara (MVA) Vaccine Platform" *Vaccine*
161. Kennedy, Greenberg (2009) "IMVAMUNE: Modified Vaccinia Ankara Strain as an Attenuated Smallpox Vaccine" *Expert Review of Vaccines*
162. Suter, Meisinger-Henschel, Tzatzaris (2009) "Modified Vaccinia Ankara Strains With Identical Coding Sequences Actually Represent Complex Mixtures of Viruses That Determine the Biological Properties of each Strain" *Vaccine*
163. Stittelaar, Kuiken, De Swart (2001) "Safety of Modified Vaccinia Virus Ankara (MVA) in Immune-Suppressed Macaques" *Vaccine*
164. Pittman, Hahn, Lee (2019) "Phase 3 Efficacy Trial of Modified Vaccinia Ankara as a Vaccine Against Smallpox" *New England Journal of Medicine*
165. Greenberg, Hurley, Dinh (2015) "A Multicenter, Open-Label, Controlled Phase II Study to Evaluate Safety and Immunogenicity of MVA Smallpox Vaccine (IMVAMUNE) in 18-40 Year Old Subjects With Diagnosed Atopic Dermatitis" *PLoS ONE*
166. Overton, Stapleton, Frank (2015) "Safety and Immunogenicity of Modified Vaccinia Ankara-Bavarian Nordic Smallpox Vaccine in Vaccinia-Naive and Experienced Human Immunodeficiency Virus-Infected Individuals: An Open-Label, Controlled Clinical Phase II Trial" *Open Forum Infectious Diseases*
167. Healy, Creech, Frey et al. "579. Safety and Immunogenicity of Mpox Vaccination in Adolescents" *Open Forum Infectious Diseases*
168. Mason, Betancur, Riera-Montes et al. (2024) "MVA-BN Vaccine Effectiveness: A Systematic Review of Real-World Evidence in Outbreak Settings" *Vaccine*
169. Pischel, Martini, Yu (2024) "Vaccine Effectiveness of 3rd Generation Mpox Vaccines Against mpox and Disease Severity: A Systematic Review and Meta-Analysis" *Vaccine*
170. Grabenstein, Hacker (2024) "Vaccines Against Mpox: MVA-BN and LC16m8" *Expert Review of Vaccines*
171. Cdc (2025) "Adult Immunization Schedule by Age"
172. Osborne, Silva, Frace (2007) "Genomic Differences of Vaccinia Virus Clones From Dryvax Smallpox Vaccine: The Dryvax-Like ACAM2000 and the Mouse Neurovirulent Clone-3"
173. Leduc, Becher (1999) "Current Status of Smallpox Vaccine" *Emerging Infectious Diseases*
174. Arness, Eckart, Love (2004) "Myopericarditis Following Smallpox Vaccination" *American Journal of Epidemiology*
175. Lane, Millar (1971) "Risks of Smallpox Vaccination Complications in the United States" *American Journal of Epidemiology*
176. Reina, Iglesias (2023) "Vaccines Against Monkeypox" *Medicina Clinica*
177. Monath, Caldwell, Mundt (2004) "ACAM2000 Clonal Vero Cell Culture Vaccinia Virus (New York City Board of Health Strain)-A Second-Generation Smallpox Vaccine for Biological Defense" *International Journal of Infectious Diseases*
178. Greenberg, Kennedy (2008) "ACAM2000: A Newly Licensed Cell Culture-Based Live Vaccinia Smallpox Vaccine" *Expert Opinion on Investigational Drugs*
179. Marriott, Parkinson, Morefield et al. (2008) "Clonal Vaccinia Virus Grown in Cell Culture Fully Protects Monkeys From Lethal Monkeypox Challenge" *Vaccine*
180. Fda (2000) "ACAM2000"
181. Artenstein, Johnson, Marbury (2005) "A Novel, Cell Culture-Derived Smallpox Vaccine in Vaccinia-Naive Adults" *Vaccine*
182. Handley, Buller, Frey et al. (2009) "The New ACAM2000 Vaccine and Other Therapies to Control Orthopoxvirus Outbreaks and Bioterror Attacks" *Expert Review of Vaccines*
183. Petersen, Harms, Reynolds et al. (2016) "Use of Vaccinia Virus Smallpox Vaccine in Laboratory and Health Care Personnel at Risk for Occupational Exposure to Orthopoxviruses-Recommendations of the Advisory Committee on Immunization Practices (ACIP), 2015" *Mmwr Morbidity and Mortality Weekly Report*
184. Fda (2024) "Approval Letter of ACAM2000"
185. Who (2025) "LC16m8 (Live-Attenuated Freeze-Dried Vaccinia) Smallpox and Mpox Vaccine"
186. Who (2024) "Recommendation For An Emergency Use Listing of "Freeze-Dried Smallpox Vaccine Prepared In Cell Culture Lc16" 'Kmb' (Lc16m8 Vaccine) Submitted By Km Biologics, Japan"
187. Sugimoto, Yasuda, Miki (1985) "Gene Structures of Low-Neurovirulent Vaccinia Virus LC16m0, LC16m8, and Their Lister Original (LO) Strains" *Microbiology and Immunology*
188. Morikawa, Sakiyama, Hasegawa (2005) "An Attenuated LC16m8 Smallpox Vaccine: Analysis of Full-Genome Sequence and Induction of Immune Protection" *Journal of Virology*
189. Saijo, Ami, Suzaki (2006) "LC16m8, a Highly Attenuated Vaccinia Virus Vaccine Lacking Expression of the Membrane Protein B5R, Protects Monkeys From Monkeypox" *Journal of Virology*
190. Meseda, Mayer, Kumar (2009) "Comparative Evaluation of the Immune Responses and Protection Engendered by LC16m8 and Dryvax Smallpox Vaccines in a Mouse Model" *Clinical and Vaccine Immunology*
191. Iizuka, Ami, Suzaki (2017) "A Single Vaccination of Nonhuman Primates With Highly Attenuated Smallpox Vaccine, LC16m8, Provides Long-term Protection Against Monkeypox" *Japanese Journal of Infectious Diseases*
192. Kobiyama, Utsumi, Kaku (2025) "Immunological Analysis of LC16m8 Vaccine: Preclinical and Early Clinical Insights Into mpox"
193. Kennedy, Gurwith, Dekker (2011) "Safety and Immunogenicity of LC16m8, an Attenuated Smallpox Vaccine in Vaccinia-Naive Adults" *Journal of Infectious Diseases*
194. Morino, Mine, Tomita (2024) "Mpox Neutralizing Antibody Response to LC16m8 Vaccine in Healthy Adults"
195. Okumura, Morino, Nomoto (2025) "LC16m8 for Pre-Exposure Prophylaxis Against Mpox in a High-Risk Population: An Open-Label Randomized Trial" *Clinical Infectious Diseases*
196. Yano, Terada-Hirashima, Uemura (2023) "Efficacy and Safety of the Smallpox Vaccine for Postexposure Prophylaxis in Monkeypox: Protocol for an Open-Labeled, Single-Armed Study" *JMIR Research Protocols*
197. Tomita, Terada-Hirashima, Uemura (2023) "An Open-Label, Non-Randomized Study Investigating the Safety and Efficacy of Smallpox Vaccine, LC16, as Post-Exposure Prophylaxis for mpox" *Human Vaccines & Immunotherapeutics*
198. Who (2024) "WHO Adds LC16m8 Mpox Vaccine to Emergency Use Listing"
199. (2025) "Efficacy/Effectiveness, Safety, and Immunogenicity of LC16m8 Mpox Vaccine in Colombia (MPOX-COL)" *Clinicaltrials*
200. Tomotsugu, Amaya-Mahecha, Hamana (2025) "Key Success Factors in Clinical Trial Operation of the Smallpox Vaccine LC16m8 Against Mpox in Colombia" *Global Health & Medicine*
201. Maksyutov, Yakubitskiy, Kolosova (2022) "Genome Stability of the Vaccine Strain VACΔ6" *Vavilovskii Zhurnal Genetiki i Selektsii*
202. Maksyutov, Yakubitskyi, Kolosova et al. (2017) "Comparing New-Generation Candidate Vaccines Against Human Orthopoxvirus Infections" *Acta Naturae*
203. (2023) "Russia Plans to Start Production of Smallpox Vaccine"
204. Zuiani, Dulberger, De Silva (2024) "A Multivalent mRNA Monkeypox Virus Vaccine (BNT166) Protects Mice and Macaques From Orthopoxvirus Disease" *Cell*
205. Mucker, Freyn, Bixler (2024) "Comparison of Protection Against Mpox Following mRNA or Modified Vaccinia Ankara Vaccination in Nonhuman Primates" *Cell*
206. Zhai, Han, Wang et al. (2025) "Advancements in Mpox Vaccine Development: A Comprehensive Review of Global Progress and Recent Data" *Biomedical and Environmental Sciences*
207. Chu, Huang, Li (2023) "Non-Replicating Vaccinia Virus NTV as an Effective Next-Generation Smallpox and Monkeypox Vaccine: Evidence From Mouse and Rhesus Monkey Models" *Emerging Microbes & Infections*
208. (2024) "The Clinical Trial Application for JT118 Injection Was Accepted" *Webull*
209. Cdc (2025) "Clinical Treatment of Mpox"
210. Hallo-Carrasco, Hunt, Prusinski (2023) "Pain Associated with Monkeypox Virus: A Rapid Review" *Cureus*
211. (2018) "Labeling of TPOXX"
212. Delaurentis, Kiser, Zucker (2022) "New Perspectives on Antimicrobial Agents: Tecovirimat for Treatment of Human Monkeypox Virus" *Antimicrobial Agents and Chemotherapy*
213. Vernuccio, Martínez León, Poojari (2025) "Structural Insights Into Tecovirimat Antiviral Activity and Poxvirus Resistance" *Nature microbiology*
214. Grosenbach, Jordan, Hruby (2011) "Development of the Small-Molecule Antiviral ST-246 as a Smallpox Therapeutic" *Future Virology*
215. (2000) "Labeling of Vistide"
216. Sobral-Costas, Escudero-Tornero, Servera-Negre (2023) "Human Monkeypox Outbreak: Epidemiological Data and Therapeutic Potential of Topical Cidofovir in a Prospective Cohort Study" *Journal of the American Academy of Dermatology*
217. (2021) "Labeling of TEMBEXA"
218. Siga (2024) "Interim Results from STOMP Study of SIGA's Tecovirimat in Treatment of Mpox Announced"
219. Ali, Alonga, Biampata (2025) "Tecovirimat for Clade I MPXV Infection in the Democratic Republic of Congo" *New England Journal of Medicine*
220. Russo, Grosenbach, Chinsangaram (2021) "An Overview of Tecovirimat for Smallpox Treatment and Expanded Anti-Orthopoxvirus Applications" *Expert Review of Anti-Infective Therapy*
221. Yang, Pevear, Davies (2005) "An Orally Bioavailable Antipoxvirus Compound (ST-246) Inhibits Extracellular Virus Formation and Protects Mice From Lethal Orthopoxvirus Challenge" *Journal of Virology*
222. Almehmadi, Allahyani, Alsaiari (2022) "A Glance at the Development and Patent Literature of Tecovirimat: The First-in-Class Therapy for Emerging Monkeypox Outbreak" *Viruses*
223. (2015) "Product Development Under the Animal Rule Guidance for Industry" *FDA*
224. Fda (2018) "Clinical Reviews"
225. Fda (2018) "Non-Clinical Reviews"
226. Amantana, Chen, Tyavanagimatt (2013) "Pharmacokinetics and Interspecies Allometric Scaling of ST-246, an Oral Antiviral Therapeutic for Treatment of Orthopoxvirus Infection" *PLoS ONE*
227. Laudisoit, Tepage, Colebunders (2018) "Oral Tecovirimat for the Treatment of Smallpox" *New England Journal of Medicine*
228. Merchlinsky, Albright, Olson (2019) "The Development and Approval of Tecoviromat (TPOXX()), the First Antiviral Against Smallpox" *Antiviral Research*
229. Fda (2018) "Approval Letter of TPOXX Capsule"
230. Fda (2022) "Combined Cross-Discipline Team Leader, Clinical, Clinical Pharmacology, and Division Director Review"
231. Russo, Grosenbach, Honeychurch et al. (2023) "Overview of the Regulatory Approval of tecovirimat Intravenous Formulation for Treatment of smallpox: Potential Impact on Smallpox Outbreak Response Capabilities, and Future tecovirimat Development Potential" *Expert Review of Anti-Infective Therapy*
232. Fda (2022) "Label of TPOXX Injection"
233. Tpd (2021) "TPOXX, Product Monograph of TPOXX Capsule"
234. Ema (2022) "Tecovirimat SIGA: EPAR-Medicine Overview"
235. Mhra (2022) "Summary of Product Characteristics"
236. Pmda (2024) "TEPOXX"
237. Yu, Elmor, Muhammad et al. (2024) "Tecovirimat Use Under Expanded Access to Treat Mpox in the United States, 2022-2023"
238. Raccagni, Leoni, Ciccullo (2023) "Rapid Improvement of Severe Mpox Lesions With Oral Tecovirimat" *Journal of Medical Virology*
239. O'laughlin, Tobolowsky, Elmor (2022) "Clinical Use of Tecovirimat (Tpoxx) for Treatment of Monkeypox under an Investigational New Drug Protocol-United States" *Mmwr Morbidity and Mortality Weekly Report*
240. Tempestilli, Mondi, D'avolio (2024) "Pharmacokinetics of Tecovirimat in Subjects With Mpox" *International Journal of Antimicrobial Agents*
241. Vo, Zomorodi, Silvera (2023) "Clinical Characteristics and Outcomes of Patients With Mpox Who Received Tecovirimat in a New York City Health System" *Open Forum Infectious Diseases*
242. Wu, Osborn, Bertram (2023) "Tecovirimat Use in Ambulatory and Hospitalized Patients with Monkeypox Virus Infection" *Sexually Transmitted Diseases*
243. Demir, Desjardins, Fortin (2023) "Treatment of Severe Human Mpox Virus Infection With Tecovirimat: A Case Series" *Canada Communicable Disease Report*
244. Aldred, Lyles, Scott (2024) "Early Tecovirimat Treatment for Mpox Disease Among People with HIV" *JAMA Internal Medicine*
245. Mondi, Gagliardini, Mazzotta (2023) "Clinical Experience With Use of Oral Tecovirimat or Intravenous Cidofovir for the Treatment of Monkeypox in an Italian Reference Hospital" *Journal of Infection*
246. Siga Technologies (2024) "Topline Results From PALM 007 Study of SIGA's Tecovirimat in Treatment of Mpox"
247. Mclean, Stoeckle, Huang (2023) "Tecovirimat Treatment of People with HIV During the 2022 Mpox Outbreak: A Retrospective Cohort Study" *Annals of Internal Medicine*
248. Lalezari, Stagg, Kuppermann "Intravenous Cidofovir for Peripheral Cytomegalovirus Retinitis in Patients With AIDS"
249. Randomized (1997) "Controlled Trial" *Annals of Internal Medicine*
250. Narsana, Ha, Ho (2025) "Treating Adenovirus Infection in Transplant Populations: Therapeutic Options beyond Cidofovir?" *Viruses*
251. Sallée, Boutolleau (2024) "Management of Refractory/Resistant Herpes Simplex Virus Infections in Haematopoietic Stem Cell Transplantation Recipients: A Literature Review" *Reviews in Medical Virology*
252. Piret, Boivin (2021) "Antiviral Drugs against Herpesviruses" *Advances in Experimental Medicine and Biology*
253. De Clercq, Sakuma, Baba (1987) "Antiviral Activity of Phosphonylmethoxyalkyl Derivatives of Purine and Pyrimidines" *Antiviral Research*
254. Baker, Bray, Huggins (2003) "Potential Antiviral Therapeutics for Smallpox, Monkeypox and Other Orthopoxvirus Infections" *Antiviral Research*
255. Clercq (2003) "Clinical Potential of the Acyclic Nucleoside Phosphonates Cidofovir, Adefovir, and Tenofovir in Treatment of DNA Virus and Retrovirus Infections" *Clinical Microbiology Reviews*
256. Smee (2008) "Progress in the Discovery of Compounds Inhibiting Orthopoxviruses in Animal Models" *Antiviral Chemistry & Chemotherapy*
257. Andrei, Snoeck (2010) "Cidofovir Activity Against Poxvirus Infections"
258. Prévost, Sloan, Deschambault (2024) "Treatment Efficacy of Cidofovir and Brincidofovir Against Clade II Monkeypox Virus Isolates" *Antiviral Research*
259. Fabrizio, Bruno, Cristiano et al. (2023) "Cidofovir for Treating Complicated Monkeypox in a Man With Acquired Immune Deficiency Syndrome" *Infection*
260. Rao, Schrodt, Minhaj (2023) "Interim Clinical Treatment Considerations for Severe Manifestations of Mpox-United States" *Mmwr Morbidity and Mortality Weekly Report*
261. Gupta, Talukder, Rosen et al. (2023) "Differential Diagnosis, Prevention, and Treatment of mpox (Monkeypox): A Review for Dermatologists" *American Journal of Clinical Dermatology*
262. Hershan (2025) "Virology, Epidemiology, Transmissions, Diagnostic Tests, Prophylaxis and Treatments of human Mpox: Saudi Arabia Perspective" *Frontiers in Cellular and Infection Microbiology*
263. Ciesla, Trahan, Wan (2003) "Esterification of cidofovir With Alkoxyalkanols Increases Oral Bioavailability and Diminishes Drug Accumulation in Kidney" *Antiviral Research*
264. Fda (2025) "Drug Approval Package: TEMBEXA"
265. Cdc (2025) "Enrollment Starts in Africa CDC-LED Mpox Therapeutic Study (MOSA)"
266. Cinatl, Bechtel, Reus (2024) "Trifluridine for Treatment of Mpox Infection in Drug Combinations in Ophthalmic Cell Models" *Journal of Medical Virology*
267. Abdelaal, Reda, Hassan (2023) "Monkeypox-Associated Manifestations and Complications Involving the Eye: A Systematic Review and Meta-Analysis of Previous and Current Outbreaks" *Asia-Pacific journal of ophthalmology (Philadelphia)*
268. Cash-Goldwasser, Labuda, Mccormick (2022) *Mmwr Morbidity and Mortality Weekly Report*
269. Perzia, Theotoka, Li (2023) "Treatment of Ocular-involving Monkeypox Virus With Topical Trifluridine and Oral Tecovirimat in the 2022 Monkeypox Virus Outbreak" *American Journal of Ophthalmology Case Reports*
270. Shamim, Satapathy, Padhi (2023) "Pharmacological Treatment and Vaccines in monkeypox Virus: A Narrative Review and Bibliometric Analysis" *Frontiers in pharmacology*
271. Shishkina, Mazurkov, Bormotov (2022) "Safety and Pharmacokinetics of the Substance of the Anti-Smallpox Drug NIOCH-14 After Oral Administration to Laboratory Animals"
272. Shiryaev, Skomorohov, Leonova (2021) "Adamantane Derivatives as Potential Inhibitors of p37 Major Envelope Protein and Poxvirus Reproduction. Design, Synthesis and Antiviral Activity" *European Journal of Medicinal Chemistry*
273. Quenelle, Collins, Herrod "Effect of Oral Treatment With Hexadecyloxypropyl"
274. Adenine (2007) "HPMPA] or Octadecyloxyethyl-(S)-HPMPA on Cowpox or Vaccinia Virus Infections in Mice" *Antimicrobial Agents and Chemotherapy*
275. Zhang, Wan, Guo (2025) "Novel Derivatives of Brincidofovir and (S)-9-(3-Hydroxy-2-Phosphonylmethoxypropyl)Adenine Inhibit Orthopoxviruses and Human Adenoviruses More Potently Than Brincidofovir" *Signal Transduction and Targeted Therapy*
276. Liu, Zhao, Zhai (2024) "Safety, Tolerability and Pharmacokinetics of ASC10, a Novel Oral Double Prodrug of a Broad-Spectrum Antiviral Agent, β-d-N4-Hydroxycytidine: Results From a Randomized, Double-Blind, Placebo-Controlled Phase 1 Study in Chinese Healthy Subjects" *Expert Opinion on Investigational Drugs*
277. Ascletis (2025) "Ascletis Announces IND Approval of Viral Polymerase Inhibitor ASC10 for Monkeypox Indication by U.S. FDA"
278. Martínez-Fernández, Fernández-Quezada, Casillas-Muñoz (2023) "Human Monkeypox: A Comprehensive Overview of Epidemiology, Pathogenesis, Diagnosis, Treatment, and Prevention Strategies" *Pathogens*
279. Cidrap (2005) "FDA Approves VIG for Smallpox Shot Complications"
280. Nih (2025) "CNJ-016"
281. Piparva, Fichadiya, Joshi et al. (2024) "Monkeypox: From Emerging Trends to Therapeutic Concerns" *Cureus*
282. Cdc (2025) "Informed Consent/Parental Permission Form for Vaccinia Immune Globulin Intravenous Treatment Under An Expanded Access Investigational New Drug (IND) Program"
283. Gao, Xie, Zhang (2025) "Substrate Recognition and Cleavage Mechanism of the Monkeypox Virus Core Protease" *Nature*
284. Zgarbová, Otava, Silhan et al. (2023) "Inhibitors of Mpox VP39 2'-O Methyltransferase Efficiently Inhibit the Monkeypox Virus" *Antiviral Research*
285. Chiem, Nogales, Lorenzo (2023) "Identification of In Vitro Inhibitors of Monkeypox Replication" *Microbiology Spectrum*
286. Karmarkar, Golden, Kerani (2022) "Association of Tecovirimat Therapy with Mpox Symptom Improvement: A Cross-Sectional Study-King County"
287. Garcia, Foote, Mcpherson (2024) "Severe Mpox Among People with Advanced Human Immunodeficiency Virus Receiving Prolonged Tecovirimat in New York City" *Open Forum Infectious Diseases*
288. Karan, Shah, Garrigues (2024) "Surveillance of Complicated Mpox Cases Unresponsive to Oral Tecovirimat in Los Angeles County, 2022" *Journal of Infectious Diseases* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12188950&blobtype=pdf | # Human Herpesvirus 8 in Solid Organ Transplant Donors and Recipients: Need for Screening? A Dutch Seroprevalence Pilot Study
Geesje Roo-, Xuewei Zhou, Coretta Van Leer-Buter, Marjolein Knoester, Geesje Roo-Brand
## Abstract
Human Herpesvirus 8 (HHV8) also called Kaposi's Sarcoma-associated herpesvirus (KSHV) is endemic worldwide. It is particularly pathogenic in immunocompromised patients. HHV8 related disease is a rare but dreaded complication in Solid Organ Transplant (SOT) recipients. Particularly donor-derived HHV8 infections can be life threatening and have a high mortality rate. Early recognition is crucial and therefore screening of the highest risk SOT patients could lead to better outcomes. We conducted a pilot study on the seroprevalence of HHV8 in our donor and recipient populations, to calculate the risk for HHV-8 related infections and to evaluate our current screening strategy. A total of 144 donors and 145 recipients were tested with the Indirect fluorescence assay for HHV8 Lytic IgG antibody. Seroprevalence in donors was 2.8% and 10.3% in recipients. This gives a 2.5% chance of donor positive/recipient negative (D+R-) combination, of whom a minority will develop symptomatic disease. Based on this data screening does not seem to be warranted in our center. Further multicenter studies are necessary to gain better insight in risk for symptomatic HHV8 infection in our SOT patients.
## 1 | Introduction
Human Herpesvirus 8 (HHV8), also known as Kaposi's sarcoma-associated herpesvirus, is the causative agent of Kaposi sarcoma (KS). It is globally distributed but pathogenic in a small proportion of individuals, particularly in immunocompromised. KS occurs 200 times more frequently in solid organ transplant (SOT) recipients than in the general population [1]. HHV8 infection in SOT is also linked to other diseases like Multicentric Castleman Disease, Primary Effusion Lymphoma, and KSHV Inflammatory Cytokine Syndrome [2]. HHV8 is mainly transmitted horizontally via saliva; other routes include vertically, sexually, through blood and transplant transmission [3].
In SOT recipients, infections can arise from reactivation or donor transmission, with donor-derived infections often being severe and rapidly progressing [4]. KS typically presents with skin abnormalities, but disseminated infections can present heterogeneously, sometimes lacking skin involvement, complicating diagnosis [2]. Serostatus information of the donor and recipient promotes early recognition of infection. Global HHV8 seroprevalence varies significantly, with the highest rates in Africa (up to 80%) and much lower rates in the USA and Northern Europe (below 10%) [3]. Rates also differ based on risk groups. In the Netherlands, a seroprevalence of 39%-65% among men who have sex with men and 5% among repeat male blood donors has been reported [5]. Screening for HHV8 in donors is not common in the Netherlands, but changing donor and recipient demographics necessitate reevaluation of screening strategies. This pilot study aimed to determine the current HHV8 seroprevalence among SOT donors and recipients at the University Medical Center Groningen to estimate the risk of donor positive/recipient negative (D+/R-) mismatches and to reconsider our current screening strategy.
## 2 | Methods
A retrospective pilot study was conducted using stored serum samples from SOT donors and recipients at the UMCG. The study included transplant procedures between April 2022 and September 2023. HHV8 serology testing was performed on pretransplant samples of 144 donors and 145 recipients using the Indirect Fluorescence Assay (IFA) for Human Herpesvirus 8 Lytic IgG antibody (Biocell) [6]. More than 60% of the samples were from the day before or on the day of transplantation, 20% within 2 months prior to transplantation and 20% had a longer interval. Samples were considered positive if deemed so by at least two independent lab technicians. Since donor sera are stored anonymously, donors and recipients could not be matched.
## 3 | Results
Table 1 shows the baseline characteristics of the recipients only, since donors are anonymous. Of the recipients 66% were male, the mean age was 57 years, and the distribution of transplant types reflected the proportions of SOT in our center, mostly kidney, followed by liver and lung.
Among the recipients, 10.3% (15/145) tested positive for HHV8 Lytic IgG antibody. When weak positives were included, seropositivity was 17.9%. No significant differences were found in age or gender distribution among the different IFA result categories. Among the donors, 2.8% (4/144) tested positive for HHV8 Lytic IgG antibody. Including weak positives, 10.4% tested positive. For details of test results, see Table 2. With a seroprevalence of 2.8% in donors and 10.3% in recipients, the calculated risk of a D+/R-mismatch is 2.5%.
## 4 | Discussion
The study found a higher-than-expected seroprevalence of 10.3% among SOT recipients, compared to 2.8% among donors. This discrepancy aligns with other studies showing higher seroprevalence in recipients than donors [6][7][8][9][10][11][12]. This may be attributed to differences in sociodemographic factors such as older age and increased risk of exposure to HHV8 among recipients. Additionally, the physical trauma associated with the transplant process may trigger reactivation of KSHV/HHV8, leading to elevated antibody levels and resulting in a positive classification of the recipient's sample.
The observed seroprevalence indicates an estimated risk of a 2.5% chance of a donor-positive/recipient-negative (D+R-) combination, with and even lower risk of symptomatic disease. Literature reports symptomatic infection rates from 0% in D+R + combinations [7,12] to up to 38% in D+R-combinations [6][7][8][9][10][11][12]. A recent large study by Mularoni et al. [12] found HHV8 transmission in 45% of D+R-cases, with a 26.5% rate of related disease and 8.2% mortality.
Given these data, is our policy of not screening for HHV8 still justifiable? Current guidelines do not support universal HHV8 screening due to lack of evidence. Screening options include serology before transplantation and posttransplant HHV8 DNA monitoring. An international survey [4] showed varying screening practices, with higher screening habits in highprevalence areas.
The study's limitations include its pilot nature, small sample size, and lack of donor sociodemographic data. Given our results and fewer than one clinical HHV8 infection per year out of a total of 235 SOTs in 2022 in our center, screening does not seem warranted. However, due to globalization, seroprevalence is constantly subject to changes and we need to monitor this from time to time. Further multicenter research is recommended to better assess HHV8-related disease risk and screening benefits.
## References
1. Grulich, Vajdic (2015) "The Epidemiology of Cancers in Human Immunodeficiency Virus Infection and After Organ Transplantation" *Seminars in Oncology*
2. Kates, Mcdade, Tinney et al. (2024) "HHV-8-Associated Diseases in Transplantation: A Case Report and Narrative Review Focused on Diagnosis and Prevention" *Transplant Infectious Disease*
3. Minhas, Wood (2014) "Epidemiology and Transmission of Kaposi's Sarcoma-Associated Herpesvirus" *Viruses*
4. Mularoni, Mikulska, Giannella (2021) "International Survey of Human Herpes Virus 8 Screening and Management in Solid Organ Transplantation" *Transplant Infectious Disease*
5. Van Bilsen, Zaaijer, Matser (2019) "Infection Pressure in Men Who Have Sex With Men and Their Suitability to Donate Blood" *Clinical Infectious Diseases*
6. Chiereghin, Barozzi, Petrisli (2017) "Multicenter Prospective Study for Laboratory Diagnosis of HHV8 Infection in Solid Organ Donors and Transplant Recipients and Evaluation of the Clinical Impact After Transplantation" *Transplantation*
7. Francès, Marcelin, Legendre (2009) "The Impact of Preexisting or Acquired Kaposi Sarcoma Herpesvirus Infection in Kidney Transplant Recipients on Morbidity and Survival" *American Journal of Transplantation*
8. Pietrosi, Vizzini, Pipitone (2011) "Primary and Reactivated HHV8 Infection and Disease After Liver Transplantation: A Prospective Study" *American Journal of Transplantation*
9. Riva, Barozzi, Quadrelli (2013) "Human Herpesvirus 8 (HHV8) Infection and Related Diseases in Italian Transplant Cohorts" *American Journal of Transplantation*
10. Lebbe, Porcher, Marcelin (2013) "Human Herpesvirus 8 (HHV8) Transmission and Related Morbidity in Organ Recipients" *American Journal of Transplantation*
11. Dollard, Annambhotla, Wong (2021) "Donor-Derived Human Herpesvirus 8 and Development of Kaposi Sarcoma Among 6 Recipients of Organs From Donors With High-Risk Sexual and Substance Use Behavior" *American Journal of Transplantation*
12. Mularoni, Cona, Bulati (2024) "Serologic Screening and Molecular Surveillance of Kaposi Sarcoma Herpesvirus/Human Herpesvirus-8 Infections for Early Recognition and Effective Treatment of Kaposi Sarcoma Herpesvirus-Associated Inflammatory Cytokine Syndrome in Solid Organ Transplant Recipients" *American Journal of Transplantation* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12263044&blobtype=pdf | # A bibliometric analysis of Mpox research based on Web of Science platform
Qunjiao Yan, Lei Wang
## Abstract
Mpox is still a public health emergency of international concern. A bibliometric and knowledge mapping analysis were performed to systematically examine the Mpox research landscape. Mpox-related literature was retrieved from the core collection of the Web of Science database. This study conducts a statistical analysis of related publications, examining the distribution patterns across publication years, category, journals, institutions, and authors. And CiteSpace, VOSviewer, and data-information-knowledge-wisdom were used to extract information about countries/regions, institutions, authors, category, and keywords to identify and analyze the research hotspots and trends in this field. A total of 3401 Mpox-related articles were retrieved, and since 2022, the volume of literature has increased sharply. The United States was the main publishing country, with the highest number of publications and citations. Infectious diseases and virology were the main research disciplines. Two major cooperation clusters had emerged, centered on China and the U.S., with the U.S.-led research cluster showing a multi-theme parallel advancement in thematic research. The United States Centers for Disease Control and Prevention was the most prolific institution, and its affiliated researcher Damon IK is the most productive author. Viruses-Basel has the highest number of publications. The epidemiology and public health regulation, genetic evolution, and transmission mechanism, pathogenesis and host immunity, drug research, and vaccine effectiveness were the main research topics. The study of Mpox transmission mechanism combined with ecology and artificial intelligence-based diagnostic methods are emerging research directions. Bibliometric analysis of Mpox studies enables researchers to efficiently identify research landscapes and emerging trends, providing critical references for countries to effectively allocate research priorities and identify collaborative partners, offering valuable insights for formulating epidemic containment strategies.
## 1. Introduction
Mpox virus is a zoonotic double-stranded DNA virus that belongs to the genus Orthopoxvirus, demonstrating both primary transmission between animal and human populations, as well as secondary human-to-human transmission. [1,2] As early as 1953, Reagan et al [3] focused on the electron microscopic study of chickenpox virus in monkey serum, and Mpox was officially named in 1958 when the virus was first identified in infected macaques in Copenhagen, Denmark. Mpox virus, initially named monkeypox, showed a broad propensity to infect mammals. [4] In 1970, the first case of monkeypox infection in human was diagnosed in the Democratic Republic of Congo. The geographical location of the outbreak indicates that Mpox originated from western and central African countries. However, with increasing international trade and business travel, confirmed cases have been reported in several countries since 2016. Mpox outbreaks have increased in scope and frequency since 2022. [5] The World Health Organization (WHO) declared the Mpox outbreak a global public health emergency on July 23, 2022. [6] In the same year, "monkeypox" was renamed "Mpox," the Congo Basin clade was renamed Clade I, the West African clade was renamed Clade II, and the 2 subclades of Clade II were named ІІa and ІІb. [7] In August 14, 2024, WHO director general declared Mpox outbreak a public health emergency of international concern again. Mpox is gaining heightened attention in both social and public health spheres.
Since Mpox was reported, many related literatures have been published. Mohapatra RK et al emphasized the transmission dynamics, zoonosis potential, complication and mitigation strategies for Mpox infection, and they concluded with a recommendation for a "One Health" approach to enhance the management, control, and prevention of the disease. [8] Zeeshan HF et al [9] provided an essential insight into the research response to scientific trends of Mpox based Scopus database. A limited number of studies have employed bibliometric methods to analyze Mpox research up to 2022. [10,11] However, there is currently a paucity of research focusing up-to-date on Mpox.
Yan and Wang • Medicine (2025) 104: 28 Medicine
Bibliometric analysis is a research methodology that systematically collects quantifiable, reproducible, and objective data to investigate scientific activities and developmental trends within specific research domains. This statistical technique has gained significant recognition due to its distinctive advantages and broad applicability across multiple disciplines, where it has been extensively employed to identify diverse research patterns and scientific regularities. [12][13][14] In the current study, we aimed to use bibliometric analysis to outline the historical progress, current research status, and future development trends in Mpox field and analyzed the category, journals, countries, institutions, authors, and years of monkeypox related articles. Thus, in this study, we reviewed the literature on Mpox-related based on the Web of Science (WoS). So, what about the research situation of Mpox? To identify the most relevant topics and trends in Mpox research, we addressed the following research questions: What is the distribution of Mpox studies by year of publication? What's the disciplinary distribution within the Mpox research field and the change pattern? Which are the most productive authors, institutions, and countries? What is the pattern of international collaboration and national research preferences in the Mpox research field? What are the characteristics and developmental trends in Mpox research? We hope this study can provide valuable reference information for scholars aiming to conduct further Mpox related research and as a source for those seeking information about Mpox as well as add a new reference for future human Mpox prevention.
## 2. Material and methods
## 2.1. Data sources
Data for this study were retrieved from the core collection of WoS database, using a subject search with the following search formula: ((((((TS = (Mpox)) OR TS = (MPXV)) OR TS = (Monkeypox)) OR TS = (Monkeypoxvirus*)) OR TS=("Monkeypox virus*")) OR TS=("Monkey pox virus*")) OR TS=("MPX virus*"). The type of documentation is limited to article, letter, review, and meeting abstract, and proceeding paper, a total of 3401 papers were retrieved on March 22, 2024.
## 2.2. Methods
VOSviewer, [15] CiteSpace, [16] data-information-knowledgewisdom [17] were used for data cleaning, data analysis, and visualization of the retrieved documents. Analysis of generated graphs in this study was combined with manual literature reading.
The data used in this study were from the WoS and did not involve patients. Therefore, permission was not required from the ethics committee.
## 3. Results
## 3.1. Analysis of the annual number of publications on Mpox
The annual number of publications on Mpox is shown in Figure 1. From 1953 to 2000, the number of publications on this topic was low, with an average annual number of 3.5.
As observed in Figure 1, the highest number of publications (10) before 2000 was reached in 1972. After 2000, the number of publications on Mpox has increased, especially after the first cases were reported outside Africa by the Centers for Disease Control and Prevention (CDC) in 2003. [18] In 2017, a Mpox outbreak occurred in Nigeria, and on November 3, 2017, the WHO held an informal consultation on Mpox in Geneva, which focused on assessing the situation of the Mpox epidemic and state of relevant knowledge and identifying gaps in the fight against Mpox. [19] This meeting reflected the growing international concern about Mpox. As we can see from Figure 1, the number of published papers was maintained at a high level around 2017. In 2022, the world experienced a major Mpox outbreak. Following a confirmed case of Mpox with a history of travel to Nigeria reported by the UK Health Security Agency on May 6, 2022, many countries reported confirmed cases, including the first imported cases reported in mainland China on September 16, 2022. According to the WHO, most cases were from America and Africa. [20] The 2022 outbreak led to a rapid increase in papers related to Mpox, with over 700 documents published in 2022. Mpox continued to gain traction, with more than twice as many articles published in 2023 as in 2022. Overall, the number of Mpox-related publications began to increase after 2000. Since the global Mpox outbreak in 2022, there has been a surge in research papers on the topic.
## 3.2. Analysis of the disciplinary distribution of Mpox
Subject analysis results from the WoS database (Table 1) showed that infectious diseases and virology were the main research disciplines, accounting for 20.80% and 15.34% of all Mpox disciplines, respectively. Other areas that contributed more than 10% of publications included immunology (13.62%), public environmental occupational health (12.99%), microbiology (11.72%). Among Mpox disciplines published from 2008 to 2021, virology was the top-ranked discipline, and in 2022, the infectious disease was the top-ranked discipline, with microbiology, immunology, internal general medicine, and public environmental occupational health, ranking ahead of virology (Appendix Fig. S1, Supplemental Digital Content, https://links. lww.com/MD/P409). This shift in discipline ranking can be attributed mainly to the Mpox outbreak in 2022; therefore, the research tilted from basic to applied studies.
## 3.3. Country analysis of the documents on Mpox
3.3.1. Analysis of national cooperation networks. VoSviewer was used to draw the national cooperation map of Mpox research papers. As shown in Figure 2, the size of the nodes represents the publication frequency, and the thickness of the connecting lines represents the intensity of cooperation. The United States had the highest number of publications, and 2 main country clusters emerged, 1 centered on China and the other on the United States. The China-centered country cluster comprised of India, Saudi Arabia, Pakistan, Egypt, Peru, Bangladesh, Lebanon, United Arab Emirates, Malaysia, and Colombia, while the United States-centered country cluster included the United Kingdom, Germany, Italy, Spain, Canada, France, Australia, Brazil, Nigeria, and Democratic Republic of Congo. As observed in Figure 2, the U.S.-centered cluster demonstrates stronger research capabilities. Some collaboration also exists between the 2 country clusters indicating there are channels for sharing information, knowledge, resources, and technology between countries, which are conducive to the prevention and detection of and response toward Mpox disease worldwide.
## 3.3.2. Topic analysis of national concerns.
Regarding terminologies, epidemiology, emerging infectious diseases, outbreak, orthopoxvirus, smallpox, smallpox vaccine, infection, transmission, and zoonosis were the common topics of concern in most countries. At the country level, the United States focused on various topics, as shown in Figure 3, with particular attention to Mpox, Mpox-related orthopoxvirus (e.g., smallpox and vaccinia virus), and drugs used to treat Mpox (e.g., ST-246 and Cidofovir). In Russia and Germany, the main topic of interest was Mpox-related orthopoxvirus. The United States, Germany, France, the United Kingdom, and Canada were concerned about several topics, whereas most of the other countries had research gaps in topics such as models, bioterrorism, Mpox-related HIV, efficiency, antiviral, antibodies, and drugs for Mpox, indicating that there is still much room for expansion in research in those countries. As shown in Figure 3, the U.S., Germany, France, the United Kingdom, and Canada demonstrated relatively diversified research focuses, while other countries exhibited research gaps in certain fields. This suggests these nations could strengthen collaborations with countries possessing expertise in relevant areas to expand their research directions. As for Institutes cooperation in China, it can be divided into 3 cluster (Fig. 4). Tsinghua University and Shandong University mostly cooperate with provincial centers for disease control and prevention, such as Chengdu center disease control and prevention, Anhui center disease control and prevention, Guangdong center disease control and prevention, Beijing center disease control and prevention. The cluster represented by the Chinese Academy of Sciences and Peking University is dominated by universities, such as Central South University, Capital Medical University, Fudan University, Zhejiang University, and Sun Yat Sen University.
## 3.4. Institutional analysis of the documents on Mpox
The third cluster is centered on Shanghai Jiao Tong University and there are more international universities in this cluster, such as Lebanese American University, King Saud University. As shown in Figure 4, in China, in addition to interuniversity collaborations, academic institutions tend to cooperate with provincial CDCs. This pattern may be attributed to CDCs' access to more comprehensive Mpox case data, while domestic universities also maintain research partnerships with international academic institutions.
## 3.5. Journal analysis of the documents on Mpox
The analysis of journals published on Mpox provides an understanding of the main Mpox publications groups and the preferences of different journals, which can help authors who are new to the field review corresponding papers and make submission decisions. The top 5 journals in terms of the number of published papers and citation frequency are shown in Table 3 with Viruses-Basel having the highest number of publications (53) and Journal of Virology having 2277 citations, indicating that the quality of its articles is widely recognized.
## 3.6. Authorship analysis of the documents on Mpox
Authors are the main participants in research work. In this study, we identified high-yield and high-producing teams and tried to understand the author clusters of specific topics as well as the cooperative relationship between these authors, which is conducive to further exploring the advantageous research directions of different research teams.
## 3.6.1. Analysis of major authors.
## 3.7. Analysis of research directions of documents on Mpox
## 3.7.1. Keywords analysis of the document on Mpox.
The analysis of disciplinary structure and knowledge foundations, as visualized in Figure 6, was conducted through keyword clustering analysis. This process identified 19 distinct thematic clusters that, upon systematic literature review, were consolidated into 5 primary research domains (Table 5).
## 3.7.2. Co-citation analysis of the documents on
Mpox. Further, we used direct citation relationships to develop a citation network, focusing on highly cited and high betweenness-centrality papers in each category and manually interpreted the research content as well as the core ideas.
## 3.7.2.1. Epidemiology and public health regulation. An infectious disease caused by the Mpox virus has been an
epidemic in more than a dozen African countries. The Mpox outbreak has received widespread attention since early May 2022, as Mpox viral infections have been reported in many nonendemic countries. In recent Mpox epidemics, the main affected groups are men who have sex with men (MSM). [21] According to relevant national reports, the proportion of patients with HIV among known cases of Mpox infection in the MSM population is more than 20%. [22,23]
## 3.7.2.2. Genetic evolution and mechanisms of transmission.
Mpox viruses are divided into 2 clades based on geographic factors and genetic and phenotypic differences. They are genetically conserved, with approximately 95% homology between the 2 clades. In general, Clade II has lower attenuation and higher transmission capacity than that observed in Clade I. [24] A related study analyzed the genome of Mpox virus associated with the 2022 outbreak and found that the majority of the mutations in the genome were point mutations, that is, single nucleotide substitutions. [25] This genomic change may be related to the adaptive advantage of orthopoxvirus evolution. The 2 elements of disease emergence are the introduction of pathogens into the population and the disease transmission and maintenance in the population. Disease transmission is mainly influenced by evolutionary and ecological factors. Evolutionary factors mainly refer to changes in the virulence of the pathogen during the process of growth adaption in humans and the subsequent spread from person to person. Ecological factors refer to human behavior, changes in host density, etc. Among them, ecology plays an important role in the emergence of the disease, and ecological changes may lead to an increased possibility of Mpox infections in humans. [26] 3.7.2.3. Pathogenesis and host immunity. Studying the host species, geographical distribution of hosts, and infection route of Mpox virus can lay the foundation for further understanding of the pathogenic mechanism of Mpox virus. [27] The natural host of Mpox has not been identified; however, rodents are the most likely hosts.
In the course of human infection, Mpox typically has an incubation period of 6 to 13 days or 5 to 21 days. The infection is usually characterized by fever, swollen lymph nodes, back pain, muscle pain, and weakness. The rash usually begins within 1 to 3 days after a fever and tends to be concentrated on the face, extremities, palm, and soles of the feet, and the oral mucous membranes, genitalia, conjunctiva, and corneas may be affected. [28] Symptoms typically last 2 to 4 weeks, and the duration of illness is related to the degree of viral exposure, health status of patients, and nature of complications. In recent years, the mortality rate of Mpox has ranged from 3% to 6% and is higher in young children. [29] 3.7.2.4. Drugs against Mpox research and efficacy of related vaccines. Drugs currently available for Mpox treatment include Tecovirimat, Brincidofovir, and Cidofovir, and approved vaccines include JYNNEOS and ACAM2000. [30] Cidofovir has selective activity against broad-spectrum DNA viruses, such as herpes simplex virus and vaccinia virus. [31] Further development of Cidofovir has led to the development of Brincidofovir (also known as CMX001), which is more potent but cytotoxic and has been approved by the United States Food and Drug Administration for the treatment of smallpox. [32] However, its efficacy in the treatment of Mpox has not been validated, but in efficacy evaluations in animal models, Brincidofovir showed a protective effect. [33] In vitro studies have shown that Tecovirimat (also known as ST-246) can be used as an antiviral therapeutic compound as it blocks the production and release of the virus. [34] Based on data from animal and human studies, the European Medicines Agency approved Tecovirimat in 2022, an antiviral drug for smallpox developed by SIGA for the treatment of Mpox. [35] 3.7.2.5. Research on early diagnosis and its new method. Early diagnosis of Mpox is very important as it is a transmitted disease. Currently, Mpox virus can be identified by measuring antibodies and antigens, assessing DNA, or characterization. Polymerase chain reaction (PCR) is the preferred laboratory test due to its accuracy and sensitivity. GeneXpert is a system that can combine sample preparation with PCR amplification and detection, which can minimize contamination and shorten detection time. [36] In some locations where PCR is not available, nucleic acid amplification test, a viral diagnostic test, can be used for Mpox detection and has shown high sensitivity and specificity. [36] It is worth mentioning that with the development of artificial intelligence, this test can also play a supporting role in disease diagnosis. For example, Mpox infection image datasets are used for identification and classification. [37,38] 3.7.2.6. The biological warfare potential of Mpox. Since the eradication of smallpox in the 1970s, Mpox, which has been the most prominent orthopoxvirus responsible for human epidemics, has been considered a potential biological weapon. [39] Studies have shown that naturally occurring Mpox may not pose a serious bioterrorism threat due to its low fatality rate and limited ability to spread. [40] However, the virus can be genetically manipulated to gain greater virulence and transmission capacity, so it is necessary to be wary of related manipulations based on genetic engineering. [39] 3.7. [41,42] Overall, each stage has a more dominant research content. Consistent with the disciplinary distribution of the publications, the research tilted from basic research to application-oriented research.
## 4. Discussion
In this study, 3401 papers retrieved from the WoS database were visually analyzed by VOSviewer, CiteSpace, and data-informationknowledge-wisdom bibliometric software. Regarding the total number of publications and citation frequency, the United States ranked first, with the CDC being the key institution for research on Mpox. Regarding authorship, most of the authors with high publications and high citations were from the CDC. Regarding country cooperation, the United States is at the center of cooperation. In terms of institute collaboration, the United States has more diverse partnerships such as national-level institutions and universities and companies, while Chinese institutions mostly cooperate with nation-level institutes in other countries rather than with companies and universities.
In terms of research topics, epidemiology and public health regulation, genetic evolution and mechanisms of transmission, and pathogenesis and host immunity are relatively the traditional research direction. Studies on the transmission mechanism combined with ecology and artificial intelligence-based methods for the diagnosis of Mpox are emerging research directions.
## 5. Conclusion
The current global landscape of Mpox prevention and control remains highly challenging, posing significant threats to international health security. Mpox is a mandatory reportable disease in the Democratic Republic of Congo; however, in most affected countries, especially in parts of the countries with active Mpox outbreaks, health services are under-resourced. Remote rural areas lack health care isolation areas and personal protective equipment, and many areas are active sites of conflict and civil unrest; therefore, Mpox is not systematically included in their integrated disease detection and response systems. The absence of a good disease surveillance system may hamper early detection and response, which poses significant challenges to the management of Mpox outbreaks worldwide.
Recent years have witnessed an increasingly pressing trajectory of epidemic development coupled with persistent surveillance dilemmas. In recent years, the geographic scope and frequency of Mpox outbreaks have increased, posing new challenges to public health regulation in various countries. Many factors, such as seasonal nomadism, refugee movement, and cross-border economic exchange, affect Mpox surveillance. In some severely affected countries, the lack of standardized case definitions and inadequate training in health care has prevented them from conducting systematic Mpox surveillance and collecting as well as reporting relevant data. Therefore, it is important to provide the necessary support for prevention, case detection, and laboratory development in countries where Mpox is endemic and regulatory systems are weak. There is a need to adhere to the "One Health" mentality, mechanisms for cross-border communication, and information sharing.
Regarding key priorities for containment and research, more attention should be paid to the MSM population, and the dissemination of basic knowledge of Mpox disease among the MSM population should be strengthened. On the other hand, the variation of the Mpox genome and assumption that the variation is an evolution should be monitored closely in scientific research. This will enable researchers to focus on what caused this evolution. In terms of transmission, relevant studies should emphasize ecological factors in the virus transmission model. In most studies, researchers tend to communicate that stakeholders should not be worried about Mpox being used as a biological weapon, but it is necessary to be alert to the threat posed by its modification caused by biotechnology.
From the perspective of technological empowerment, as an emerging and rapidly developing field, artificial intelligence has made significant progress in the detection, screening, diagnosis, and classification of Mpox, as well as the characterization of Mpox viral genomes and assessment of Mpox drug utility. Using artificial intelligence to identify disease clusters, monitor cases, and predict outbreaks is a new trend in disease research. New technologies and methods in Mpox research should be strengthened, with emphasis on disease surveillance, regional approaches to enhance disease prevention and control and mapping of ecological risk niches.
This study employs bibliometrics to conduct a comprehensive analysis of Mpox-related research literature. However, several limitations should be acknowledged regarding the research scope. Although the WoS database represents scientific achievements in global research, our analysis was exclusively confined to WoS-indexed publications. Notably, our study emitted other major databases like Scopus and native-language databases from non-English speaking countries. Future studies should expand the literature scope to enable more comprehensive bibliometric analyses in the Mpox field. The main obstacles that may impede progress include interpretation and analysis of keyword clustering as well as evolutionary trend analyses.
## References
1. Xiang, White (2022) "Monkeypox virus emerges from the shadow of its more infamous cousin: family biology matters" *Emerg Microbes Infect*
2. Hraib, Jouni, Albitar et al. (2022) "The outbreak of monkeypox 2022: An overview" *Ann Med Surg*
3. Reagan, Day, Moore et al. (1953) "Electron microscopic studies of the virus of varicella (Chicken pox) from monkey serum" *Tex Rep Biol Med*
4. Parker, Buller (2013) "A review of experimental and natural infections of animals with monkeypox virus between 1958 and 2012" *Future Virol*
5. Jirong, Zhijian (2022) "The progress on monkeypox" *China Modern Doctor*
6. Peng, Xie, Kuai (2023) "Structure of monkeypox virus DNA polymerase holoenzyme" *Science*
7. (2022) "WHO recommends new name for monkeypox disease [EB/OL]"
8. Mohapatra, Singh, Branda (2024) "Transmission dynamics, complications and mitigation strategies of the current mpox outbreak: a comprehensive review with bibliometric study" *Rev Med Virol*
9. Zeeshan, Rubab, Dhlakama et al. (2022) "Global research trends on monkeypox virus: a bibliometric and visualized Study" *Trop Med Infect Dis*
10. Kamal, Farahat, Awad (1964) "Global trends of monkeypox-related articles: a bibliometric analysis over the last five decades" *J Infect Public Health*
11. Lin, Li, Zhong et al. (2022) "Bibliometric analysis of human monkeypox research from 1975 to 2022 and novel prevention and control strategies"
12. Marhl, Markovic, Grubelnik et al. (2025) "The changing world dynamics of research performance" *Scientometrics*
13. Sunahara, Perc, Ribeiro (2021) "Association between productivity and journal impact across disciplines and career age" *Phys Rev Res*
14. Sunahara, Perc, Ribeiro (2023) "Universal productivity patterns in research careers" *Phys Rev Res*
15. Van Eck, Waltman (2010) "Software VOSviewer, a computer program for bibliometric mapping" *Scientometrics*
16. Chen (2006) "CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature" *J Am Soc Inf Sci Technol*
17. Drip (2023) "Data-Information-Knowledge-Wisdom, DIKW"
18. Sale, Melski, Stratman (2006) "Monkeypox: An epidemiologic and clinical comparison of African and US disease" *J Am Acad Dermatol*
19. Who (2018) "WHO informal consultation on monkeypox 2017[EB/OL]"
20. (2023) "Monkeypox situation update [EB/OL]"
21. Vivancos-Gallego, Sánchez-Conde, Rodríguez-Domínguez (2022) "Human monkeypox in people with HIV: transmission, clinical features, and outcome" *Open Forum Infect Dis*
22. Martínez, Montalbán, Bueno (2022) "Monkeypox outbreak predominantly affecting men who have sex with men" *Euro Surveill*
23. Hoffmann, Jessen, Wyen (2023) "Clinical characteristics of monkeypox virus infections among men with and without HIV: a large outbreak cohort in Germany" *HIV Med*
24. Chen, Li, Liszewski (2005) "Virulence differences between monkeypox virus isolates from West Africa and the Congo basin" *Virology*
25. Orassay, Berdigaliyev, Sadvokassova (2023) "Recent advances on human mpox" *New Microbes New Infect*
26. Antia, Regoes, Koella et al. (2003) "The role of evolution in the emergence of infectious diseases" *Nature*
27. Li, Yuan, Jiang et al. (2023) "Animal host range of mpox virus" *J Med Virol*
28. Soheili, Nasseri, Afraie (2022) "Monkeypox: virology, pathophysiology, clinical characteristics, epidemiology, vaccines, diagnosis, and treatments" *J Pharm Pharm Sci*
29. De Clercq, Jiang, Li (2023) "Therapeutic strategies for human poxvirus infections: Monkeypox (mpox), smallpox, molluscipox, and orf" *Travel Med Infect Dis*
30. De Clercq, Holý, Rosenberg et al. (1986) "A novel selective broad-spectrum anti-DNA virus agent" *Nature*
31. (2021) "FDA approves drug to treat smallpox"
32. Linjie, Wu (2022) "Research and treatment progress of monkeypox virus" *Clin Med J*
33. Smith, Olson, Karem et al. (2009) "In vitro efficacy of ST246 against smallpox and monkeypox" *Antimicrob Agents Chemother*
34. (2022) *European Medicines Agency. Tecovirimat SIGA*
35. Li, Wilkins, Mccollum (2017) "Evaluation of the GeneXpert for human monkeypox diagnosis" *Am J Trop Med Hyg*
36. Ahsan, Uddin, Farjana et al. "Image Data collection and implementation of deep learning-based model in detecting Monkeypox disease using modified VGG16"
37. Patel, Surti, Patel et al. (2023) "Artificial intelligence (AI) in Monkeypox infection prevention" *J Biomol Struct Dyn*
38. Giulio, Eckburg (2004) "Human monkeypox: an emerging zoonosis" *Lancet Infect Dis*
39. Ketai, Lim (2007) "Radiology of Biological Weapons-Old and the New?" *Semin Roentgenol*
40. Raccagni, Mileto, Galli (2023) "HIV viral load monitoring during monkeypox virus infection among PLWH" *AIDS*
41. Labate, Brucci, Ciccarese (2023) "Nasal monkeypox virus infection successfully treated with cidofovir in a patient newly diagnosed with HIV" *Int J STD AIDS* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12800105&blobtype=pdf | # Two-year persistence of MERS-CoV-specific antibody and T cell responses after MVA-MERS-S vaccination in healthy adults
Leonie Mayer, Anahita Fathi, Hanna-Marie Weichel, Matthijs Raadsen, Christine Dahlke, Anna Mykytyn, Jordi Rodon, Gesche Gerresheim, Merel Te Marvelde, Leonie Weskamm, Ilka Grewe, Claudia Schlesner, Marc Lütgehetmann, Christian Drosten, Stephan Becker, Bart Haagmans, Svenja Hardtke, Marylyn Addo
## Abstract
MVA-MERS-S, a vaccine candidate against Middle East respiratory syndrome (MERS), was recently evaluated in a randomized, placebo-controlled, doubleblind phase 1b clinical trial to assess its safety, immunogenicity, and optimal dosing in healthy adults in Hamburg and Rotterdam. A three-dose regimen was safe and elicited robust spike-specific antibody responses. We extended this trial to assess the two-year durability of MERS-CoV-specific antibody and T cell responses in 48 study participants of the Hamburg cohort. Our findings show that immune responses remain detectable for at least 24 months after the third vaccination. Antibodies persisted at levels comparable to the peak response observed after the second vaccination and were able to cross-neutralize MERS-CoV spike mutants. Although the immune correlates of protection against MERS remain unknown, the observed durability of humoral and cellular immune responses supports the potential of MVA-MERS-S as a promising MERS vaccine candidate and highlights the importance of a booster dose in sustaining long-term immunity.The Middle East respiratory syndrome coronavirus (MERS-CoV) emerged as a human pathogen in 2012 and has a reported case fatality rate of up to 36% 1,2 . MERS-CoV continues to circulate in camels and infections in humans are reported sporadically. It poses a potential pandemic threat and is considered a priority pathogen for the development of vaccines and therapeutics 3 . Currently, no licensed vaccine is available; however, three vaccine candidates based on the MERS-CoV spike protein have been shown to be safe and immunogenic in human phase 1 clinical trials: MVA-MERS-S 4-6 , ChAdOx1 MERS 7,8 , and GLS-5300 DNA MERS-CoV 9 .We recently conducted a randomized, double-blind, placebocontrolled phase 1 clinical trial of MVA-MERS-S, a MERS vaccine candidate based on the replication-deficient Modified Vaccinia virus Ankara (MVA) viral vector. The trial aimed to determine the optimal
dose, prime-boost interval, and impact of a booster dose in 139 healthy adults (NCT04119440). The primary and secondary outcomes of the trial have been published by Raadsen et al. 10 . Prime-boost vaccination with MVA-MERS-S elicited robust binding and neutralizing antibody responses, with higher titers observed when the prime-boost interval for the high dose group (10 8 plaque-forming units [PFU]) was extended from 28 to 56 days. Notably, administering a third dose six months later significantly enhanced the antibody response, leading to peak titers that were comparable across the different vaccine doses and prime-boost intervals, but overall higher than those observed after the second dose 10 .
While the immunological mechanisms mediating protection against MERS remain mostly unknown, vaccine efficacy studies of the closely related betacoronavirus SARS-CoV-2 have demonstrated that antibody responses serve as the best immunological correlate of protection across different vaccine platforms 11,12 . However, unlike liveattenuated vaccines, which can induce lifelong immunity 13 , the antibody responses elicited by most subunit, mRNA, and non-replicating viral vector vaccines licensed to date tend to wane over time 14 . Assessing the relationship between declining antibody titers and duration of protection against COVID-19 has been challenging due to limited long-term follow-up studies and the occurrence of breakthrough infections with SARS-CoV-2 variants [15][16][17][18] .
To evaluate the durability of vaccine-induced immunity against MERS-CoV and the impact of different dosing regimens, we extended the MVA-MERS-S phase 1b clinical trial to allow for a 24-month followup of 48 study participants. Here, we show that binding and neutralizing antibodies, as well as T cell responses are maintained for at least 24 months following the third MVA-MERS-S vaccination. While we previously showed that MVA-MERS-S vaccination does not elicit crossreactive binding antibodies against SARS-CoV-2 19 , we also assessed SARS-CoV-2 antibody responses and recorded SARS-CoV-2 infections and vaccinations in this study to determine whether they influence the persistence of MERS-CoV-specific antibody responses.
## Results
## Demographics and safety
We originally enrolled 139 participants in the phase 1b clinical trial between 2021 and 2022 at the study sites in Hamburg and Rotterdam. According to the trial design, participants had been randomized by vaccine dose and V1-V2 time interval to one of four treatment groups: the 28-day 10 7 PFU group, the 28-day 10 8 PFU group, the 56-day 10 7 PFU group, the 56-day 10 8 PFU group, or to the placebo group (Fig. 1).
Of the 74 participants enrolled in Hamburg, 55 re-consented to participate in the subsequent 24-month follow-up study (Fig. 2). All 55 participants received at least one injection and were included in the safety analysis. Seven participants of the treatment groups had missed one or more MVA-MERS-S vaccinations. The remaining 48 participants had received all three MVA-MERS-S vaccinations (treatment groups) or placebo doses (placebo group), as defined by the modified intentionto-treat (mITT) criteria of the study 10 , and were included in the longitudinal immunogenicity analysis. Four participants were lost to follow-up after V3M6 or V3M12.
Table 1 summarizes the demographics of the mITT immunogenicity set at the time of screening, stratified by study group. The demographics of the safety set are shown in Supplementary Table 1. During the follow-up period, nine participants reported to have received COVID-19 vaccinations, four participants received other vaccines within four weeks before a visit, and two participants received immunosuppressive therapy (Supplementary Table 2). Six moderate to severe serious adverse events were reported in 4/55 (7%) participants (Supplementary Table 3). All events were considered to be unrelated to the study vaccination. Two participants had short-term stays in countries where MERS-CoV is endemic (Supplementary Table 4).
## Persistence of the MERS-CoV-specific antibody response
To assess the long-term persistence of vaccine-induced MERS-CoVspecific antibody responses following three doses of
Lost to follow-up after M6 (n = 1)
Lost to follow-up after M6 (n = 1)
Fig. 2 | Trial profile. All study participants enrolled at the Hamburg site were eligible for participation in the extension of the phase 1b trial. Of those, n = 55 reconsented and were included in the long-term follow-up. Seven participants missed ≥ 1 MVA-MERS-S vaccinations and were excluded from the immunogenicity analysis, resulting in a modified intention-to-treat cohort of n = 48. Four participants were lost to follow-up and a total of n = 44 participants completed the 24month follow-up. V vaccination, M month, PFU plaque-forming units. V3M12, and V3M24 timepoints. Figure 3 shows the longitudinal S1 IgG response of the immunogenicity set from baseline (pre-V1) through the last timepoint of the 24-month follow-up (V3M24), stratified by study group: 28-day 10 7 PFU (a), 28-day 10 8 PFU (b), 56-day 10 7 PFU (c), and 56-day 10 8 PFU (d), placebo (e). Geometric mean titers (GMTs) of S1 IgG at the timepoints shown in Fig. 3 are summarized in Supplementary Table 5. Robust S1 IgG responses were induced after the second dose (V2M1), were further boosted by V3 and peaked at V3M1 with comparable titers across all treatment groups 10 . Titers then gradually declined but remained detectable through V3M24 (Fig. 3). In total, 60% (24/40) of vaccinated participants remained S1 IgG seropositive at V3M24, compared to 86% at V3M1 (Supplementary As shown in Fig. 4a, peak S1 IgG titers in vaccinated individuals at V3M1 decreased 2.3-fold until 6 months after vaccination (V3M6) and then waned more gradually, with a 1.4-fold decrease by month 12 (V3M12) and a further 1.3-fold decrease by month 24 (V3M24). Notably, antibody responses remained at higher titers after V3 compared to antibody levels achieved after two immunizations (V2). Compared to the S1 IgG GMT at V2M6 (10 IU/ml, 95% CI: 7.7-14.8) the S1 IgG GMTs were significantly higher at V3M6 (68 IU/ml, 95% CI: 44.2-103.9, p < 0.0001) and V3M12 (49 IU/ml, 95% CI: 31.4-75.1, p < 0.0001), as well as at V3M24 (35 IU/ml, 95% CI: 23.4-51.8, p = 0.4436) although not reaching significance (Fig. 4a). Antibody titers remained higher in vaccinated individuals compared to the placebo group at all long-term follow-up timepoints (V3M6 (p = 0.0002), V3M12 (p = 0.0076), and V3M24 (p = 0.0376) (Fig. 4b)). No statistically significant differences in titers at V3M24 were observed between the treatment groups (Fig. 4c). At V3M24, 56%, 36%, 83%, and 63% of participants in the 28-day 10 7 PFU, 56-day 10 7 PFU, 28-day 10 8 PFU, and 56-day 10 8 PFU groups, respectively, remained seropositive (Fig. 4c, Supplementary Table 6).
Participants in the placebo group remained seronegative at all followup timepoints.
We observed a significant positive correlation between peak antibody titers (V3M1) and long-term antibody titers two years following the last vaccination (V3M24; r = 0.89, p < 0.0001, r 2 = 0.82, p < 0.0001, Fig. 4d). The GMT of high responders (defined as participants with S1 IgG titers above the mean at V3M1) decreased by 5.5-fold to 76 IU/ml at V3M24, whereas the GMT of the low responders (defined as participants with S1 IgG titers below the mean at V3M1) decreased by 3.5-fold to 17 IU/ml at V3M24. The longitudinal S1 IgG titers of high and low responders are shown in Supplementary Fig. 4. A demographic comparison of high and low responders is provided in Supplementary Table 7. S1 IgG responses at V3M24 showed a strong positive correlation with MERS-CoV-neutralizing antibody responses measured by pseudovirus assay (r = 0.82, 95% CI: 0.69-0.90, p < 0.0001, Fig. 3e) and live-virus assay (r = 0.83, 95% CI: 0.71-0.91, p < 0.0001, Fig. 3f), as well as full S IgG (r = 0.92, 95% CI: 0.87-0.96, p < 0.0001, Fig. 3g).
$$Missed V2 (n = 2) Missed V2 (n = 1) Missed V3 (n = 1) Missed V3 (n = 1) Missed V2 + V3 (n = 1)$$
## Impact of SARS-CoV-2 exposure and anti-vector immunity
In the immunogenicity set, 47/48 (98%) participants were SARS-CoV-2 seropositive at baseline before MVA-MERS-S vaccination (Supplementary Fig. 5a) and 42/48 (88%) showed an increase in SARS-CoV-2 nucleocapsid antibodies indicative of a SARS-CoV-2 infection during the 24-month follow-up (Supplementary Fig. 5b). There was no significant correlation between pre-existing SARS-CoV-2 S IgG titers and MERS-CoV S1 IgG at V3M1 (r = -0.017, 95% CI: -0.31-0.28, p = 0.91, Supplementary Fig. 5c) and V3M24 (r = 0.14, 95% CI: -0.17-0.43, p = 0.37, Supplementary Fig. 5d), as well as MERS-CoV neutralizing antibody titers at V3M1 (r = 0.020, 95% CI: -0.27-0.31, p = 0.90, Supplementary Fig. 5e) and V3M24 (r = 0.11, 95% CI: -0.20-0.40, p = 0.48, Supplementary Fig. 5f).
We observed no significant correlation between pre-existing neutralizing antibody titers against the MVA vector and peak S1 IgG titers at V3M1 (r = -0.17, 95% CI: -0.44-0.13, p = 0.25, Supplementary Fig 6a) or persistent S1 IgG titers at V3M24 (r = -0.16, 95% CI: -0.45-0.15, p = 0.29 Supplementary Fig. 6b). MVA-neutralizing titers elicited by the first two vaccinations did not correlate with peak S1 IgG titers at V3M1 (r = 0.19, 95% CI: -0.11-0.46, p = 0.20, Supplementary Fig. 6c) or persistent S1 IgG titers at V3M24 (r = 0.13, 95% CI: -0.073-0.50, p = 0.38, Supplementary Fig. 6d).
## Cross-neutralization of MERS-CoV spike variants
To assess whether MVA-MERS-S elicits cross-neutralization against clinically relevant MERS-CoV variants, we generated VSVpp harboring spike mutations D510G or I529T (Fig. 5a), which emerged in the 2015 outbreak in South Korea, and measured virus neutralization in ten serum samples. As shown in Fig. 5b, GMTs against wild-type, mutant D510G and mutant I529T spike were 830 (95% CI: 298-2312), 1091 (95% CI: 355-3355), and 1074 (95% CI: 381-3028), respectively. Compared to the wild-type, there was no significant reduction in neutralization of mutant D510G (Fig. 5c) or mutant I529T (Fig. 5d), indicating that MVA-MERS-S elicits antibodies that can cross-neutralize these variants.
## Persistence of the MERS-CoV-specific T cell response
Next, we analyzed if vaccine-induced T cell responses persist until two years after vaccination by restimulating fresh whole blood samples at the V3M24 timepoint with an overlapping spike peptide pool and measuring cytokine release by ELISA (Fig. 6a). A shown in Fig. 6b, compared to the median IFN-γ response of the placebo group (6.3 pg/ ml, 1.1-10.4), the median IFN-γ response of the MVA-MERS-Svaccinated group was significantly higher (31.5 pg/ml, 95% CI: 13.5-55.2, p = 0.0106). Similarly, compared to the median IL-2 levels of the placebo group (4.3 pg/ml, 95% CI: 2.2-10.0), the median IL-2 response of the MVA-MERS-S-vaccinated group was significantly higher (35.1 pg/ml, 95% CI: 24.3-49.0, p = 0.0023, Fig. 6c). There was a significant positive correlation between the IFN-γ and IL-2 response (r = 0.87, 95% CI: 0.77-0.93, p < 0.0001, Fig. 6c).
## Discussion
The induction of durable, protective immunity remains a major challenge in vaccine development. Although three MERS vaccine candidates have undergone phase 1 clinical trials 4,[7][8][9] , the durability of immunogenicity remains unknown. In this study, we demonstrated that three-dose vaccination with the MVA-MERS-S candidate vaccine elicits robust immune responses that persist for at least two years, with 75% and 50% of participants maintaining detectable neutralizing antibodies measured by pseudovirus and live-virus assays, respectively, and 60% of participants remaining S1 IgG seropositive. Furthermore, we could show that vaccine-induced T cells secreting IFN-γ and IL-2, indicative of a Th1-biased response, were detectable for at least two years. Two other MERS vaccine candidates, ChAdOx1 MERS and GLS-5300 DNA, have demonstrated antigen-specific immunogenicity in humans, but only limited data on long-term immunity are available. Folegatti et al. reported that one year after single-dose ChadOx1 MERS vaccination, 68% of participants maintained full-spike-specific antibody titers above the assay cut-off and the T cell responses remained significantly above baseline 8 . Booster doses of ChAdOx1 MERS have not been tested in clinical trials; however, in a preclinical model, responses elicited by ChadOx1 MERS could be significantly boosted by subsequent vaccination with an MVA-based vaccine candidate 20 . Modjarrad et al. reported that one year after three-dose GLS-5300 DNA vaccination, 3% of participants had detectable live-virus neutralizing Fig. 4 | Persistence of vaccine-induced antibody responses. a S1 IgG titers of all MVA-MERS-S-vaccinated participants (n = 39) after V2 at month 6 (V2M6) compared to after V3 at months 1 (V3M1, p < 0.0001), month 6 (V3M6, p < 0.0001), month 12 (V3M12, p = 0.0031) and month 24 (V3M24, p > 0.99), showing the foldreduction of geometric mean titers after V3. b S1 IgG titers in vaccinated individuals (purple, n = 39) compared to the placebo group (grey, n = 4) at V3M6 (p = 0.0002), V3M12 (p = 0.0076), V3M24 (p = 0.038). c Comparison of V3M24 titers in the different treatment groups. Pie charts show proportions of seropositive (black) and seronegative (grey) individuals defined as having at least 4-fold higher titers at V3M24 compared to baseline before first vaccination. Data are shown as individual points and geometric mean titers with 95% confidence interval. The dotted horizontal lines indicate the lower limit of detection (4.6 IU/ml) (a-c). d Simple linear regression of S1 IgG titers in vaccinated individuals at V3M1 and V3M24 (n = 40, r 2 = 0.82, p < 0.0001). e Spearman correlation of V3M24 S1 IgG responses of all participants (n = 44) with V3M24 (e) neutralizing responses measured by pseudovirus neutralization assay (r = 0.82, p < 0.0001) and (f) live-virus neutralization assay (r = 0.83, p < 0.0001) as well as (g) full S IgG responses (r = 0.93, p < 0.0001). S1 IgG titers were compared using Friedman test (a) or Kruskal-Wallis test (b, c) and adjusted using Dunn's multiple comparison. Associations were investigated using simple linear regression (d) and Spearman's correlation (e-g). Statistical tests are two-sided. V vaccination, M month, S spike, IU international units, nFRNT 50 normalized pseudovirus 50% focus reduction neutralization assay, VNT 100 live-virus neutralization test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data are provided as a Source Data file.
antibodies, 79% of participants remained seropositive for S1 IgG, and ~60% of participants had a positive T cell response 9 . Nonetheless, a direct comparison between these studies is difficult because of the use of different vaccination schedules and immunogenicity assays.
In our MVA-MERS-S trial, we observed that antibody titers were substantially boosted by V3 to comparable levels between treatments groups, waned more slowly, and remained higher compared to antibody levels observed after V2. Our data suggest, that the V1-V2 interval and vaccine dose do not impact peak antibody titers and antibody persistence as long as V3 is administered. However, the small sample size within each study group makes it challenging to detect significant differences. We previously reported the enrichment of activated, S-specific memory B cells and persistent antibody titers in a subset of seven participants of the MVA-MERS-S phase 1a trial, who received a booster dose (V3) one year after their initial prime-boost vaccination 5 .
Recall responses of memory B cells are more efficient at producing long-lived plasma cells, which could explain the observed increased magnitude and durability of the antibody response after V3 21 . This enhanced antibody persistence following a third dose is consistent with findings from COVID-19 vaccination studies, where a third dose translated into higher long-term effectiveness against disease [22][23][24] .
Interestingly, we found that the magnitude of the peak antibody titer at V3M1 strongly correlated with long-term antibody persistence. In line with our findings, Huttner et al. demonstrated that the magnitude of the peak binding antibody response induced by rVSV-EBOV vaccination was the strongest predictor of 5-year antibody persistence 25 . These data suggest that a vaccine's ability to elicit robust peak antibody titers could be a strong indicator of its long-term humoral immunogenicity. Overall, we observed a substantial variation in peak antibody titers between study participants, indicating that regardless of vaccination regimen, some individuals responded more robustly than others. This variation was not associated with preexisting or induced humoral immunity against the MVA viral vector.
Although human data are limited, several clinical trials have, in line with our results, reported no impact of anti-vector immunity on the induction of insert-specific humoral immunity [26][27][28] . There was no difference in sex, age or BMI between high and low responders, but the variability in vaccine response could be attributable to other demographic or genetic factors 29 , which warrant further investigation. While for COVID-19, caused by a related betacoronavirus, antibody responses have been suggested as immune correlates of vaccineinduced protection in humans 11,12 , the correlates of protection against MERS remain unknown. Humoral immunity, however, is expected to play a critical role, as MERS-CoV-specific antibody responses have been shown to correlate with protection in mice 30,31 , to correlate with reduced MERS-CoV viral load in camels 32 , and to persist in MERS survivors two years post infection 33 . Importantly, we could show that sera of MVA-MERS-S-vaccinated individuals effectively cross-neutralized MERS-CoV spike variants D510G and I529T, with no reduction in neutralizing titers compared to the wild-type virus. These polymorphisms emerged and spread in humans during the 2015 Korean outbreak 34,35 . Reduced neutralization of these mutants was previously reported in some but not all tested sera of MERS patients 34,36,37 and escape from monoclonal antibodies targeting the affected epitopes was previously shown 38 . Our data suggests, that MVA-MERS-S elicits a broad neutralizing antibody response that remains effective against variants, which is encouraging for MERS vaccine development. Furthermore, we were able to show that MVA-MERS-S generates durable T cell immunity, which is inherently less prone to viral escape and particularly relevant for protection against severe disease [39][40][41] .
Since all study participants had pre-existing immunity to SARS-CoV-2, a comparative analysis between exposed and unexposed participants was not feasible. We, however, observed no correlation between SARS-CoV-2-specific antibody titers and MVA-MERS-Sspecific immunity, and previously showed that neutralizing and S1specific antibody responses were comparable with those seen in the pre-pandemic phase 1a trial 10 . Cross-reactive responses have been reported in humans for SARS-CoV-2 and seasonal coronaviruses, but not for SARS-CoV-2 and MERS-CoV, and their contribution to crossprotection remains unclear 42,43 . If B cell or T cell clones against specific conserved betacoronavirus epitopes are preferentially re-expanded upon heterologous challenge is subject to future investigations.
The value of long-term follow-up studies of vaccine trials, such as the one presented here, lies in their ability to longitudinally characterize the durability of vaccine-induced immunity. Such data might be used in the future to predict the long-term efficacy of a vaccine and inform boosting strategies 44 . We calibrated our ELISA and pseudovirus assays to the WHO international standard, facilitating the comparison of our findings with future studies. Standardized measurements of antibody responses may also be crucial for evaluating the efficacy of novel vaccine candidates via immunobridging 45 . This is especially relevant for pathogens like MERS-CoV, where efficacy trials are not feasible due to the currently restricted number of reported cases 46 . A limitation of this study is the absence of a group that did not receive a booster dose, limiting a direct comparison of antibody persistence between the second and third vaccinations to the 6-month timepoint. Additionally, a longitudinal comparison of the T cell response was not feasible as whole blood samples were not available at the early timepoints. As this phase 1 trial was conducted in a country where MERS-CoV does not circulate and a comparator group of MERS survivors was not available, future trials will have to assess whether the observed vaccine immunogenicity is protective.
As MVA-MERS-S requires at least two doses to elicit neutralizing antibodies, this vaccine might be less optimal for emergency vaccination schemes in an acute outbreak setting. However, with the strong boosting effect of a third dose, the long-lasting humoral and cellular immunity, and the excellent safety profile of the MVA platform, MVA-MERS-S could be employed to protect the elderly, healthcare workers, abattoir workers and travelers in MERS-CoV-endemic regions.
In conclusion, we demonstrated the two-year persistence of antigen-specific antibody and T cell responses after three-dose vaccination with an MVA-based vaccine candidate against MERS. These findings further support the clinical development of MVA-MERS-S 4,10 .
## Methods
## Study design
A placebo-controlled phase 1b clinical trial was conducted between 2021 and 2022 to assess the safety, immunogenicity, and optimal dosing of intramuscularly administered MVA-MERS-S in healthy adults (NCT04119440). The trial protocol was reviewed and approved by the competent authorities in Germany (Paul-Ehrlich-Institute) and the Netherlands (Central Committee on Research Involving Human Subjects) and by the ethics committees of the Hamburg medical association and Erasmus Medical Center. The study complies with all relevant ethical regulations. Written informed consent was obtained from all participants prior to inclusion in the study. Participants were enrolled at study sites in Rotterdam and Hamburg and randomized into four treatment groups, receiving three vaccinations (V1-V3) of either the low dose (10 7 PFU) or the high dose (10 8 PFU) of MVA-MERS-S. The prime-boost interval (V1-V2) was either 28 or 56 days, followed by a third vaccination (V3) at day 224. To maintain blinding, a placebo dose was administered on day 56 in the 28-day interval groups and on day 28 in the 56-day interval groups. A placebo-only group receiving four placebo doses was included for comparison. The trial was unblinded according to protocol after the last participant last visit, which refers to the one-month timepoint after third vaccination (=V3M1). Safety (primary outcome) and humoral immunogenicity (secondary outcome) data until V3M1 were recently published 10 . The trial was subsequently extended and participants of the Hamburg study site were reconsented to monitor safety and immunogenicity for two years after V3. The trial extension was reviewed and approved by the German Competent National Authority (Paul-Ehrlich-Institute) and the Ethics Committee of the Hamburg Medical Association (reference number 2020-10180-AMG-ff).
Demographic data of trial participants are reported in Table 1 and Supplementary Table 1. Biological sex was self-reported by participants. At each follow-up visit, the occurrence of targeted illness (COVID-19 and any febrile illness), new serious adverse events, targeted concomitant treatment (COVID-19, Mpox, or MVA vaccination, any other vaccination within four weeks before a visit, as well as immunosuppressive therapy defined as >14 days treatment with immune suppressants or other immune-modifying drugs), and travel to MERS-CoV-endemic regions were documented. These data were reported in electronic case report forms (eCRF, secuTrial database version 5.1.0.20). Serum samples were collected at baseline and at months 6 (M6), 12 (M12), and 24 (M24) following V3 to assess antibody responses. Lithium-heparin whole blood samples were collected at V3M24 to assess the durability of the T cell response.
## MERS-CoV-specific ELISAs
MERS-CoV S1-and full-spike-specific binding antibody responses were measured using in-house enzyme-linked immunosorbent assays (ELISAs) 10 . 96-well plates were coated with with 1 μg/ml MERS-CoV S1 protein (Bio Techne, cat. No. 10737-CV-100) or trimerized, pre-fusion stabilized full spike protein (Keith Chapell, University of Queensland, Australia) at 4 °C overnight. After blocking the plates with 5% powdered milk in Tris-buffer for one hour, serum samples (dilution 1:100) were added and incubated for one hour. Detection was done using HRP-labeled rabbit anti-human IgG (Dako, cat. No. P0214) followed by 3,3',5,5'-Tetramethylbenzidine (TMB, KPL, cat. No. 507603). The reaction was stopped after 5 min using 0.5 N sulfuric Acid (Merck, cat. No. 1•09073•1000). The optical density was measured at 450 nm using a microplate reader (Tecan Infinite F200). Results were interpolated using a 5-parameter sigmoidal curve fit according to the MERS-CoV IgG international standard, available from the National Institute for Biological Standards and Control (NIBSC, Hertfordshire, United Kingdom, cat. No. 19/178) 47 and reported in international units (IU)/ml. The positivity cut-offs for the S1 and full-spike ELISA were defined as the mean concentration plus three times the standard error of 77 healthy, pre-pandemic sera.
## MERS-CoV-specific neutralization assays
Neutralization of the wild-type MERS-CoV was assessed using a normalized pseudovirus 50% focus reduction neutralization assay (nFRNT 50 ) and a live-virus neutralization test (VNT 100 ) 10 . For the nFRNT 50 assay, serial dilutions of heat-inactivated serum samples in Opti-MEM medium (supplemented with 10% fetal bovine serum, penicillin and streptomycin (P/S) and 10 mM Y-27632 apoptosis inhibitor (MedChemExpress, cat. No. HY-10583)) were mixed with 600 focus-forming units (FFU) of MERS-CoV spike-pseudotyped vesicular stomatitis virus (VSV) with Green Fluorescence Protein (GFP) 48 and incubated for 1 h at 37 °C. The mixture was then added to a confluent monolayer of Calu-3 cells in 96-well plates. After overnight incubation, cells were fixed (4% paraformaldehyde) stained (Hoechst 33342 dye, ThermoScientific, cat. No. 62249), imaged (automated confocal microscope Opera Phenix, Perkin Elmer), and GFP-expressing cells were quantified (image analysis software Harmony version 4.9, Perkin Elmer). Serum was considered to be neutralizing at a 50% reduction of GFP-expressing cells. Neutralizing titers were calculated using 5-parameter logistic regression and results were normalized according to the MERS-CoV IgG international standard, available from the National Institute for Biological Standards and Control (NIBSC, Hertfordshire, United Kingdom, cat. No. 19/178) 47 . The lower limit of detection for the nFRNT 50 is 16 international units (IU)/ml. Samples without neutralization were set to 16 IU/ml.
For the VNT 100 assay, serial dilutions of heat-inactivated serum samples were incubated with 100 plaque-forming units (PFU) of MERS-CoV (EMC/2012 isolate) followed by addition of Huh-7 cells (JCRB0403). Neutralization was defined as the absence of cytopathic effect four days post-infection. A human monoclonal antibody (Human anti-MERS spike, m336, Detai Bio-Tech Co., Nanjing, China) was used as a neutralization control. Neutralization titers were calculated as reciprocal geometric mean titers of three replicates. The lower limit of detection for the VNT 100 is a titer of 8. A titer of 8 is considered positive and samples without neutralization were set to a titer of 4.
## Neutralization of MERS-CoV variants
Cross-neutralization of MERS-CoV variants was assessed by FRNT 50 . Vesicular stomatitis virus-based pseudotypes (VSVpp) expressing mutant MERS-CoV spike proteins harboring the D510G or the I529T polymorphism were generated by transfection of HEK-293T cells (ATCC, CRL-3216) with plasmids carrying the mutant spike sequences and inoculation with VSV*ΔG expressing GFP and firefly luciferase 49 . VSVpp were incubated with serial dilutions of study participants' sera for 1 h at 37 °C and applied onto Calu-3 cell (ATCC, HTB-55) monolayers for 18 h (1:1 mixture; 200 FFU/well). After 4% paraformaldehyde fixation and DAPI staining, infection foci were quantified using the iSpot reader (AID) and normalized to the serum-free control. All samples were tested in duplicate.
## SARS-CoV-2-specific antibody assays
SARS-CoV-2-specific antibody responses were measured using a qualitative anti-nucleocapsid IgG/M/A assay (Elecsys, Roche) with a predefined cut-off index (COI) for positivity of ≥1 and a quantitative anti-trimeric-spike IgG assay (LIAISON, DiaSorin) with a predefined cutoff for positivity of 13 arbitrary (arb.) units /ml. The assays were performed according to the manufacturers' instructions.
## MVA-specific neutralization assay
MVA-specific antibody responses were measured using a recombinant rMVA-GFP virus 50% focus reduction neutralization assay (MVA FRNT 50 ) 10 . Serial dilutions of heat-inactivated serum samples in serumfree AdDF-12 medium (supplemented with P/S and HEPES) were incubated with 1500 FFU of rMVA-GFP at 37 °C for one hour and then added to Vero cells (seeded at a density of 30.000 cells/ well the previous day) in 96-well plates. After three days, cells were fixed (4% paraformaldehyde) stained (Hoechst 33342 dye, ThermoScientific Cat. No. 62249), imaged (automated confocal microscope Opera Phenix, Perkin Elmer), and GFP-expressing cells were quantified (image analysis software Harmony version 4.9, Perkin Elmer). A 50% reduction of GFP-expressing cells was considered neutralizing and titers were estimated around the 50% reduction threshold. Non-neutralizing serum samples were given a titre of 10. Non-neutralizing samples were given a titer of 10.
## Cytokine-release T cell assay
Spike-specific T cell responses were measured using a cytokine-release assay. Lithium-heparin whole blood was stimulated with a MERS-CoV spike peptide pool consisting of 15-mers overlapping by 11 amino acids (GenBank: JX869059; JPT Peptide Technologies; 1 μg/ml) for 20-24 h at 37 °C. After stimulation, blood samples were centrifuged for 10 min at 12,000 × g and IFN-γ and IL-2 secretion was measured in duplicate in the supernatant using a microfluidic, multiplex ELISA following the manufacturer's instructions (ELLA, ProteinSimple). Data are shown as concentrations after background subtraction of an unstimulated control for each blood sample. Samples with undetectable cytokine concentrations were set to the lower limit of quantification predefined by the assay standard curve.
## Statistical analysis
Seropositivity was defined as an antibody titer at least fourfold above baseline (pre-vaccination) levels. The Friedman test was used to compare paired samples, whereas unpaired samples were compared using the Mann-Whitney U test or the Kruskal-Wallis test. Statistical tests were two-sided and a p-value of ≤0.05 was considered statistically significant. Where applicable, statistical significance was adjusted using Dunn's multiple comparisons test. Associations between data sets were assessed using Spearman correlation or simple linear regression.
## References
1. Zaki, Van Boheemen, Bestebroer et al. (2012) "Isolation of a novel coronavirus from a man with pneumonia in Saudi Arabia" *N. Engl. J. Med*
2. Memish, Perlman, Kerkhove et al. (2020) "Middle East respiratory syndrome" *Lancet*
3. Zumla, Peiris, Memish et al. (2024) "Anticipating a MERS-like coronavirus as a potential pandemic threat" *Lancet*
4. Koch (2020) "Safety and immunogenicity of a modified vaccinia virus Ankara vector vaccine candidate for Middle East respiratory syndrome: an open-label, phase 1 trial" *Lancet Infect. Dis*
5. Weskamm (2022) "Persistence of MERS-CoV-spike-specific B cells and antibodies after late third immunization with the MVA-MERS-S vaccine" *Cell Rep. Med*
6. Fathi (2022) "Increased neutralization and IgG epitope identification after MVA-MERS-S booster vaccination against Middle East respiratory syndrome" *Nat. Commun*
7. Bosaeed (2022) "Safety and immunogenicity of ChAdOx1 MERS vaccine candidate in healthy Middle Eastern adults (MERS002): an open-label, non-randomised, dose-escalation, phase 1b trial" *Lancet Microbe*
8. Folegatti (2020) "Safety and immunogenicity of a candidate Middle East respiratory syndrome coronavirus viral-vectored vaccine: a dose-escalation, open-label, non-randomised, uncontrolled, phase 1 trial" *Lancet Infect. Dis*
9. Modjarrad (2019) "Safety and immunogenicity of an anti-Middle East respiratory syndrome coronavirus DNA vaccine: a phase 1, open-label, single-arm, dose-escalation trial" *Lancet Infect. Dis*
10. Raadsen (2024) "Safety, immunogenicity, and optimal dosing of a modified vaccinia Ankara-based vaccine against MERS-CoV in healthy adults: a phase 1b, double-blind, randomised placebocontrolled clinical trial" *Lancet Infect. Dis*
11. Khoury (2021) "Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection" *Nat. Med*
12. Gilbert (2022) "Immune correlates analysis of the mRNA-1273 COVID-19 vaccine efficacy clinical trial" *Science*
13. Amanna, Carlson, Slifka (2007) "Duration of humoral immunity to common viral and vaccine antigens" *N. Engl. J. Med*
14. Bhattacharya (2022) "Instructing durable humoral immunity for COVID-19 and other vaccinable diseases" *Immunity*
15. Srivastava (2024) "SARS-CoV-2-infection-and vaccine-induced antibody responses are long lasting with an initial waning phase followed by a stabilization phase" *Immunity*
16. Bates (2022) "Antibody response and variant cross-neutralization after SARS-CoV-2 breakthrough infection" *JAMA*
17. Hertz (2023) "Correlates of protection for booster doses of the SARS-CoV-2 vaccine BNT162b2" *Nat. Commun*
18. Wu (2022) "Long-term effectiveness of COVID-19 vaccines against infections, hospitalisations, and mortality in adults: findings from a rapid living systematic evidence synthesis and meta-analysis up to December" *Lancet Respir. Med*
19. Raadsen (2023) "Brief report: monkeypox virus crossneutralizing antibodies in clinical trial subjects vaccinated with modified vaccinia virus Ankara encoding MERS-Coronavirus spike protein" *J. Infect. Dis. jiad*
20. Alharbi (2017) "ChAdOx1 and MVA based vaccine candidates against MERS-CoV elicit neutralising antibodies and cellular immune responses in mice" *Vaccine*
21. Akkaya, Kwak, Pierce (2020) "B cell memory: building two walls of protection against pathogens" *Nat. Rev. Immunol*
22. Bar-On (2021) "Protection of BNT162b2 vaccine booster against Covid-19 in Israel" *N. Engl. J. Med*
23. Sritipsukho (2023) "Real-life effectiveness of COVID-19 vaccine during the Omicron variant-dominant pandemic: how many booster doses do we need?" *Emerg. Microbes Infect*
24. Garcia-Beltran (2022) "mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant" *Cell*
25. Huttner (2023) "Antibody responses to recombinant vesicular stomatitis virus-Zaire Ebolavirus vaccination for Ebola virus disease across doses and continents: 5-year durability" *Clin. Microbiol. Infect*
26. Rosa (2017) "MVA vaccine encoding CMV antigens safely induces durable expansion of CMV-specific T cells in healthy adults" *Blood*
27. Kreijtz (2014) "Safety and immunogenicity of a modifiedvaccinia-virus-Ankara-based influenza A H5N1 vaccine: a randomised, double-blind phase 1/2a clinical trial" *Lancet Infect. Dis*
28. Barrett (2021) "Phase 1/2 trial of SARS-CoV-2 vaccine ChAdOx1 nCoV-19 with a booster dose induces multifunctional antibody responses" *Nat. Med*
29. Pedroza-Pacheco, Mcmichael (2022) "Immune signature atlas of vaccines: learning from the good responders" *Nat. Immunol*
30. Tai (2023) "MERS-CoV RBD-mRNA vaccine induces potent and broadly neutralizing antibodies with protection against MERS-CoV infection" *Virus Res*
31. Zhao (2015) "Passive immunotherapy with dromedary immune serum in an experimental animal model for Middle East respiratory syndrome coronavirus infection" *J. Virol*
32. Tolah (2016) "Cross-sectional prevalence study of MERS-CoV in local and imported dromedary camels in Saudi Arabia" *PloS One*
33. Cheon (2022) "Longevity of seropositivity and neutralizing antibodies in recovered MERS patients: a 5-year follow-up study" *Clin. Microbiol. Infect*
34. Kim (2015) "Sequential emergence and wide spread of neutralization escape Middle East respiratory syndrome coronavirus mutants, South Korea" *Emerg. Infect. Dis*
35. Kim (2015) "Sequential emergence and wide spread of neutralization escape Middle East respiratory syndrome coronavirus mutants, South Korea" *Emerg. Infect. Dis*
36. Kleine-Weber (2019) "Mutations in the spike protein of Middle East respiratory syndrome coronavirus transmitted in Korea increase resistance to antibody-mediated neutralization" *J. Virol*
37. Addetia (2024) "Mapping immunodominant sites on the MERS-CoV spike glycoprotein targeted by infection-elicited antibodies in humans" *Cell Rep*
38. Kleine-Weber (2019) "Mutations in the spike protein of Middle East respiratory syndrome coronavirus transmitted in Korea increase resistance to antibody-mediated neutralization" *J. Virol*
39. Pereira Neto (2025) "Highly conserved Betacoronavirus sequences are broadly recognized by human T cells"
40. Sette, Sidney, Crotty (2023) "T cell responses to SARS-CoV-2" *Annu. Rev. Immunol*
41. Moss (2022) "The T cell immune response against SARS-CoV-2" *Nat. Immunol*
42. Murray (2023) "The impact of pre-existing cross-reactive immunity on SARS-CoV-2 infection and vaccine responses" *Nat. Rev. Immunol*
43. Neto (2025) "Highly conserved Betacoronavirus sequences are broadly recognized by human T cells"
44. Padmanabhan, Desikan, Dixit (2022) "Modeling how antibody responses may determine the efficacy of COVID-19 vaccines" *Nat. Comput. Sci*
45. King (2024) "Realising the potential of correlates of protection for vaccine development, licensure and use: short summary" *Npj Vaccines*
46. (2025) "MERS-CoV worldwide overview, situation update"
47. Mattiuzzo, Bentley, Seo (2020) "WHO/BS/2020.2398 Establishment of 1st WHO International Standard for Anti-MERS-CoV AntibodyGiada"
48. Mykytyn (2021) "SARS-CoV-2 entry into human airway organoids is serine protease-mediated and facilitated by the multibasic cleavage site"
49. Schroeder (2021) "Functional comparison of MERS-coronavirus lineages reveals increased replicative fitness of the recombinant lineage 5" *Nat. Commun* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12518252&blobtype=pdf | # AI-based hardware and software tools in microscopy to boost research in immunology and virology
Diego Morone, Rocco D'antuono, Giovanni Porta, Jie Yang, Rocco Antuono, D Antuono
## Abstract
The integration of computational advances in microscopy has enhanced our ability to visualise immunological events at scales. However, data generated with these techniques is often complex, multi-dimensional, and multi-modal. Data science and artificial intelligence (AI) play a key role in untangling the wealth of information hidden in microscopy data by enhancing image processing, automating image analysis, and assisting in interpreting the results. With this Review, we aim to inform the reader about the advances in the fields of fluorescence and electron microscopy with a focus on their applications to immunology and virology, and the AI approaches to aid image acquisition, analysis, and data interpretation. We also outline the open-source tools for image acquisition and analysis and how these tools can be programmed for an image-informed, AI-assisted acquisition.
## 1 Introduction
Since the first observations of microorganisms using a high-quality single-lens microscope by Antonie van Leeuwenhoek in the 17th century (1), microscopy has been instrumental in understanding diseases and alterations of the immune system. In the late 19th century, Robert Koch isolated the anthrax and tuberculosis bacteria, establishing microscopy as one of the key techniques for investigating the immune system (2). Today, microscopy offers unparalleled insights into molecular mechanisms, cellular dynamics, and tissue interactions.
Before the 20th century, observations were performed using brightfield light microscopy (3). Afterwards, several innovations have led to progress in the imaging of immune cells and molecules. Among them, we mention the optimisation of fluorescent probes in conjunction with the advancement of fluorescence optical methods (4), the development of techniques for antibody isolation (5), the discovery of natural fluorescent proteins (6), and the development of genetic manipulation techniques (7). These innovations allow today the precise imaging of cells and molecules, enabling real-time tracking of dynamics, localisation, and interactions that characterise the immune system both in vitro and in vivo. On the other hand, the use of electron microscopy (EM) to visualise structures that are relevant for immunology and virology dates back to the first images of viral particles by Ernst Ruska in 1939 (8,9). Since then, improvements in both protocols and microscope techniques have led to significant increase in resolution, maximum sample size, and tagging specificity. Among these, we should cite progresses in sample preparation, such as ultramicrotomy (10) and cryogenic techniques (11,12), advancements in microscope hardware, such as the use of digital cameras and direct detectors (13) and scanning electron microscopy (10), or finally advancements in software, for microscope automation (14)(15)(16)(17), image reconstruction (18), tomographic reconstructions (19), and alignment between different modalities (20)(21)(22)(23)(24). Today, microscopy encompasses a wide range of techniques, ranging spatial scales from the molecular detail to the whole tissue and organ, and time scales from the millisecond to days (25,26). The wealth of visible and hidden information in the images can deeply enhance our understanding of immune events, if unlocked.
Artificial Intelligence (AI) technologies are pervasive in today's world and are affecting all fields of knowledge, including science (27). AI enables machines to mimic human intelligence and perform tasks that typically require human cognition, such as learning, problem solving and perception. In microscopy, AI can significantly enhance our understanding by extracting information from images, bridging the gap between scales (28), finding hidden connections within images or between images and other types of data (e.g. genomics or proteomics, 29), and guiding the acquisition in challenging experiments and across modalities (30,31).
In the first part of this Review, we will survey main microscopy and image analysis techniques, with a focus on their use in immunology and virology. The second part introduces the reader to the concepts of AI and its main applications in microscopy, in particular for immunology and virology research. We finish with a discussion on the current challenges in AI and how its integration with microscopy can drive a new generation of tools to unlock novel insights into the immune system.
2 Microscopy and image analysis for immunology and virology 2.1 Electron microscopy 2.1.1 Room temperature EM Classic, Room-Temperature Transmission Electron Microscopy (RT-TEM) relies on chemical fixation, lipid staining with contrast agents based on heavy metals, resin embedding, and cutting in ultrathin sections. This methodology, mainly established in the 1960s (32), is an excellent way to investigate subcellular morphology and ultrastructure (33). A certain degree of threedimensional information can be acquired by capturing EM images at varying sample tilt angles and performing tomographic reconstructions (electron tomography, ET in short, 34). RT-TEM has been applied to visualise organelle changes in immune cells, such as in the investigation of the role of autophagy of the endoplasmic reticulum during plasma cell differentiation (35), the disposal of damaged mitochondria in migrating neutrophils through mitocytosis (a novel mitochondrial quality control process, 36), or the mechanisms of antigen presentation by major histocompatibility complex (MHC) in antigen-presenting cells (37)(38)(39)(40). Another classical visualisation approach relies on the negative staining of small particles, such as protein aggregates, viral particles, or extracellular vesicles, and it is used in clinical diagnostics for virus identification (41).
Immuno-Electron Microscopy (IEM) highlights specific markers on the structure of interest by immunostaining with target-specific antibodies conjugated to colloidal gold particles (42). Another way to achieve specific staining is by expressing a genetically encodable tag that triggers the deposition of a contrasting agent (43,44), which was used for example for the visualisation of the role of ectocytosis in terminating TCR signalling in cytotoxic T cells (CTLs,45).
Scanning Electron Microscopy (SEM) employs a focused electron beam that scans the surface of a sample, generating an array of secondary electrons that are detected and converted into an image (46). This results in a high-resolution image of the sample's surface. To improve contrast, the sample is frequently coated with metals. SEM has been employed with immunogold staining to map the SARS-CoV-2 receptor ACE2 distribution along the motile cilia in respiratory multiciliated cells (47). Also, SEM showed how the porosity of liver sinusoids reduces antigen recognition by effector CD8+ T cells (48) or how intercellular nanotubes enable mitochondrial trafficking from bone marrow stromal cells to CD8+ T cells, to enhance their fitness and antitumor efficacy (49).
Finally, volume Electron Microscopy (vEM) is an emerging group of techniques that offers unprecedented insights into the three-dimensional organisation and dynamics of immune cells, tissues, and molecular complexes. vEM is based on TEM or SEM. TEM-based techniques, like serial section TEM (ssTEM) and serial section ET (ssET), reconstruct volumes by acquiring sequentially ultra-thin sample slices. On the other hand, SEM-based techniques, including array tomography, serial block-face SEM (SBF-SEM) and focused ion beam SEM (FIB-SEM), scan sample surfaces to produce image stacks for 3D reconstruction (26,50,51). In immunology, SBF-SEM has been applied to reconstruct T cells (52) and to elucidate how Candida albicans exploits transcellular tunnels to invade epithelial cells while evading host immunity (53). FIB-SEM helped in clarifying how G protein subunit Gb4 negatively regulates phagocytosis by controlling plasma membrane abundance in myeloid cells (54), or was employed to create a 3D reconstruction of CTLs with target cells (45). FIB-SEM was instrumental in reconstructing, with a near-isotropic resolution of 4 nm, wholecell organelle segmentations, which resulted in the "OpenOrganelle" web repository (55-57). Among others, a notable example is the reconstruction of a CTL interacting with an ovarian cancer cell (56). Moreover, vEM can be combined with advanced labelling techniques like immuno-gold (58) or fluorescent nanoparticles to visualise specific cellular structures or molecular interactions within complex biological samples (59), providing an invaluable information for understanding the spatiotemporal organisation of immune responses at the ultrastructural level.
## 2.1.2 Cryo-EM and freeze substitution
Cryo-Electron Microscopy (cryo-EM) techniques can currently achieve the sub-nanometre range (60)(61)(62). The sample (proteins, cells, or tissues) is first flash-frozen at cryogenic temperature, allowing the creation of a layer of vitreous ice, which fixes the sample while preserving its ultrastructure (11). The vitrified sample can be visualised by different techniques. Microcrystal Electron Diffraction (MicroED) provides structural information from 3D nanocrystals (63). Single Particle Analysis (SPA) reconstructs protein structures without the need for crystallization (61,62). Cryo-Electron Tomography (cryo-ET) enables the 3D reconstruction by capturing images at varying tilt angles and performing tomographic reconstructions (34) and together with subtomogram averaging can resolve macromolecules (64). Lastly, cryo-Scanning Transmission EM (cryo-STEM) provides a tomographic reconstruction of thick lamellae with quantifiable chemical characterization (65). Cryo-EM techniques have revolutionised structural immunology, enabling the visualisation, in high-resolution, of viral particles (66,67) or SARS-CoV-2 assembly and egress (68), the TCR complex assembly (69)(70)(71)(72), the structural components of antigen processing and presentation (73), the chemokine recognition and the activation of chemokine receptors CCR5, CCR6, CCR2, CCR3 (74), and helped guiding the design of nanoparticles inducing potent neutralising antibody responses (75).
Cryo-Immuno-EM of ultrathin cryo-sections prepared from chemically fixed samples (76) allows the best preservation of protein antigenicity, as it requires chemicals only for fixation (77). When imaging surface proteins, cells can also be labelled with immunogold before cryo-fixation and imaged by cryo-ET (78). This approach opens a range of applications for the study of ultrastructural localisation of surface markers, with potential relevance for the field of immunology.
Volume EM at cryogenic conditions can capture 3D morphology in cells at near native state. It can be approached with cryo-ssET (cryo-serial section Electron Tomography) or with cryo-FIB-SEM. For example, cryo-FIB-SEM showed how growth hormone remodels 3D mitochondrial structure in macrophages (79) or the 3D ultrastructure of HIV virological synapses (80).
Finally, in the case of cells and tissues, samples can be plungefrozen (81, if less than 10-15 mm in thickness) or fixed by highpressure freezing (82,83, if between 20-200µm in thickness). After flash-freezing, the sample can be imaged at cryogenic temperature or slowly brought back to room temperature in a chemical fixation buffer, using a so-called freeze-substitution protocols (84). This approach reduces artefacts that could potentially be introduced by toxic chemical agents used as fixatives. Freeze substitution techniques are also employed in light microscopy to reduce fixation artefacts when performing subcellular diffraction-limited or super-resolved imaging (85-89).
## 2.1.3 Room-temperature CLEM
Correlative Light-Electron Microscopy (CLEM) integrates the complementary approaches of light and electron microscopy on the same portion of cell or tissue to overcome the limitations of both techniques, combining the multichannel protein localisation of light microscopy with nanometre resolution of EM (Figure 1A). The sample is usually imaged separately with the two modalities and then images are aligned with respect to each other. This poses several challenges to both acquisition and image registration. Some approaches that directly combine both modalities in the same microscope are starting to appear (93). One interesting application of CLEM is the visualisation, in the study by Baldwin et al. (49), of mitochondrial transport from bone marrow stromal cells to CD8+ T cells, by fluorescently labelling mitochondria in stromal cells and then imaging CD8+ T cells with both modalities. Also, FIB-SEM has been combined with light microscopy to visualise the virological synapse and virus-containing compartments in HIV-infected T cells (94). The combination of intravital microscopy and electron microscopy merges the dynamic information of immune cells in vivo with a more comprehensive characterisation of the same in fixed tissue. This multiscale deep phenotyping approach is reviewed in (95).
## 2.1.4 Cryo-CLEM
CLEM combined with cryogenic conditions also has great potential, due to its preservation of the native state, very high resolution and capacity to retain fluorescent signals (96). For example, this approach has been applied to investigate the intracellular trafficking of Salmonella bacteria (97). In the case of genetically engineered cells (such as when expressing GFP or other fluorescent proteins), cryo-fixation can then be followed by cryosectioning with FIB (98) or cryo-ultramicrotomy (99). Cryofluorescence light microscopy (cryo-FLM) can guide the lamella milling process (100), also with super-resolution LM (89). Finally, current protocols are extending cryo-CLEM to post-milling visualisation (101), and to 3D samples, such as organoids (102), thus providing a step towards sub-nanometre visualisation in larger 3D context, with relevant applications in immunology.
## 2.1.5 Conclusions on the use of electron microscopy in immunology and virology
All in all, electron microscopy offers insights into the ultrastructure and three-dimensional organisation of viruses, immune complexes, cells, and tissues that are unattainable with light microscopy techniques (Figures 1E,1J). Coupled with immunostaining in a correlative approach, EM also informs on the localisation of selected proteins, thus giving context to the ultrastructural detail. EM poses several challenges in terms of image analysis. High-resolution cryo-EM of viral and protein structures employs state-of-the-art algorithms to reconstruct information from low-signal images. Images of thin slices and tomographic reconstructions are frequently analysed by manual segmentation due to the complexity of the ultrastructural contrast. Volume EM reconstructions pose many challenges in terms of image reconstruction, alignment, contrast, and segmentation, due to the size and the complexity of the structures visualised. All these techniques are already or might soon take advantage of the latest AI and image analysis developments (Table 1).
## 2.2 Light microscopy
## 2.2.1 Super-resolution
By selectively tagging molecules of interest, fluorescence microscopy allows the characterisation of functional and structural features of biological samples, with the intrinsic limitation of the optical resolution limit (136,137). In fluorescence microscopy, the diffraction-limited resolution is in the order of 200-220 nm, a scale compatible with most cell and tissue imaging applications. However, the scale of many biological structures, such as organelles and molecular clusters, is at least one order of magnitude smaller (tens of nm). Super-resolution microscopy of fluorescent samples bridges the scale of light and electron microscopy, preserving sample integrity and the possibility of performing functional imaging on live samples (Figures 1E,1J). Most super-resolution methods are based on the manipulation or the analysis of the on/off state of emitters (fluorophores), which are changed either spatially or temporally (138), or on the concept of light reassignment by optical rescanning (139) or pixel reassignment (140).
Localisation microscopy is based on experimental minimisation of the number of active emitters in the field of view, either by activating only a few fluorophores at the time or by limiting the population in the ground energetic state. This allows the determination of the emitter position with the highest probability (141). Localisation microscopy, including techniques such as PALM and STORM, requires the acquisition of a high number of frames to accumulate information about biological structures, and the use of blinking fluorophores to obtain a sparse presence of emitters in the field of view. In general, the definition of biological structures in the final image improves with the number of accumulated frames (in the order of thousands). However, there are methods to optimise the acquisition parameters, such as excitation power and number of frames needed, depending on the structure dimensionality (142), or to identify artefacts in the super-resolved image based on a local error map (143). In immunology, localisation microscopy provided information on SARS-Cov-2 entry in liver spheroids (144), showed how the TCR is randomly distributed on the surface of resting antigen-experienced T cells (145), and informed on the structure of cluster in the NK cell's immune synapse (146).
One of the methods to achieve super-resolution using specific illumination patterns is the STimulated Emission Depletion (STED) microscopy, in which the illumination laser, hitting the sample as a diffraction-limited laser spot, is used together with a depletion laser illumination, shaped as a toroidal pattern that switches off fluorescence, leaving a smaller emission spot and therefore increasing the resolution as a function of the power of the depletion laser (147). In immunology, STED microscopy has been used to show the role of SWAP70 in organising actin cytoskeleton during phagocytosis (148) and how TIGIT receptor can inhibit T cell activation by forming nanoclusters (149).
Other optical super-resolution methods implemented on laserscanning systems, such as Zeiss Airyscan (150) and Nikon NSPARC (151), rely on light reassignment, assuming that higher order rings of Airy pattern can be detected with arrayed detectors and light be reassigned to the centre of the pattern where single emitters should be located (140). Finally, other methods such as iSIM and SoRa, are based on optically rescanning the point spread function to reduce its size and obtain instant super-resolution imaging on camera-based systems (152).
Instead, super-resolution methods based on the temporal analysis of fluorescence intensity fluctuations do not require blinking fluorophores and can be employed on data sets acquired on conventional microscopes (e.g. wide-field, TIRF, laser scanning confocal). They are referred to as Fluorescence-Fluctuation based Super Resolution Methods (FF-SRM), and each relies on a different statistical analysis of the temporal fluorescence fluctuations (e.g. Super-Resolution Optical Fluctuation Imaging (SOFI, 153), Super-Resolution Radial Fluctuations (SRRF, 154) or Mean-Shift Super-Resolution (MSSR, 155) to overcome specific limitations in the acquisition or in the image, such as low signal-to-noise, low number of frames, capability to reconstruct hollow structures or susceptibility to the creation of image artefacts (156).
All in all, the landscape of super-resolution microscopy ranges from methods based on light reassignment, providing moderate optical resolution increase (e.g. Airyscan), to methods based on localisation microscopy and fluorescence depletion, achieving a resolution of the order of the nanometre (e.g. RESI (157), MINFLUX (158)(159)(160) and MINSTED (161)). The use of computational methods on top of super-resolved images can further enhance super-resolution even with a limited number of frames (155).
## 2.2.2 TIRF
Total Internal Reflection Fluorescence (TIRF) microscopy uses the total internal reflection of a laser beam to create a thin illumination layer. This allows the observation of fluorescent molecules close to the coverslip surface (depth of about 100-200 nm, Figures 1F,1J162,163). Such method provides high-resolution images of the basal cell layer with minimal background noise, making it ideal for studying cellular processes such as migration, adhesion, and signalling (164).
In immunology, TIRF microscopy is commonly used to study the interactions between immune cells in antigen presentation. For example, seminal studies employed this technique to investigate TCR clusterization and activation pathways following antigen recognition (165,166). More recently, TIRF has been used to highlight that clathrin is recruited in microclusters to mediate internalisation and vesicular release of a triggered T cell receptor at the immunological synapse (167). Another application of TIRF microscopy in immunology is the study of receptor clustering, such as to show the importance of CD4+ T cell's CXCR4 nanoclusters in supporting CXCL12-mediated responses (168). On macrophages, TIRF has been employed to show the accumulation of dynamin-2 at the site of phagosome closure (169).
A recent evolution of TIRF microscopy, called quantitative dynamic footprint (qDF) and based on variable-angle TIRF, was used to visualise leukocytes rolling, adhering, and spreading with nanometre-scale z-resolution (170,171). Finally, TIRF in conjunction with SIM super-resolution microscopy showed the engagement of two spatially distinct TCR microclusters with ZAP70-bound TCR and LAT-associated signalling complex (172).
## 2.2.3 Confocal
Confocal microscopy is based on the use of an optical aperture, called a pinhole, to obtain the optical sectioning of the sample and localise the fluorescent signal in 3D (Figure 1B). It was invented by Marvin Minsky (173)a computer scientist who would later play a significant role in the development of AI concepts and methodsand has been improved with the use of lasers and scanning systems (174,175). It avoids the need to physically slice thick samples by rejecting out-of-focus light proportionally to the reduction of the pinhole aperture (176).
Thanks to its versatility, confocal microscopy can find applications on a wide range of samples, from fast visualisation of live subcellular events to reconstruction of large portions of thick tissues. On the subcellular scale, confocal live-cell microscopy was used to investigate how lymphocytes, in the absence of chemotactic signalling, orient their migration against a fluid flow (177), to characterise the force dynamics in phagocytic engulfment by cytotoxic T cells (178), and to show how the Golgi complex directs the positioning of lytic granules inside NK cells to guide their cytotoxicity (179). At the cellular level, tracking of macrophages in live-cell confocal imaging helped, together with modelling, in clarifying how these cells use a collective quorum licensing to initiate inflammation (180). In fixed tissue samples (Figure 1C), confocal microscopy has been employed to visualise macrophages in meningeal compartments of the central nervous system (181), vascular endothelium of mouse lymph node (182), virion transport to lymph nodes (183) and neutrophil accumulation (91). Moreover, it helped in defining the role of scavenging chemokines in marginal B cell zone formation (184) or the contribution of innate lymphoid cells and conventional T cells on shaping gut microbiota and lipid metabolism (185), and Treg accumulation around self-activated T cells in lymph node paracortex (186). Finally, it contributed to characterising megakaryocytes in the bone marrow niche (187), periarteriolar alignment and integrin-dependent network formation of tissueresident mast cells (188), platelets around metastatic niches in lungs (189), and tissue-resident memory T cells on the ocular surface (190). In a high-throughput manner, confocal microscopy was instrumental in isolating CAR-T cell clones with a multi-killing property against patient-derived cancer cell organoids and associating this information with their transcriptomic profile (191).
When coupled with pulsed lasers, confocal microscopy can detect fluorescence lifetimes in so-called Fluorescence Lifetime (113,134), performing segmentation to study nanoparticle delivery to alveolar macrophages (135) The table summarizes, for each technique, the resolution range, the strengths and uses cases, the applicable sample types and provides examples of application of AI methods for image analysis.
Morone and D'Antuono 10.3389/fimmu.2025.1610345
Imaging (FLIM,192). The determination of the lifetimes can be done either with exponential fit of the decay histogram or with phasor analysis (193). The measured lifetimes are concentrationindependent but microenvironment-dependent. Thus, local microenvironment changes can be assessed, such as pH and ion changes, FRET events (194) or cell membrane tension (195).
Confocal microscopy can also be used to visualise structures below the diffraction limit by means of expansion microscopy, which increases the sample size (e.g. by 10-fold), and standard confocal imaging (196) 199). For example, N&B has been applied to determine GPCR oligomerisation states in live cells (200).
Recently, new technologies in confocal imaging are being developed to increase its speed and multi-view capabilities, such as techniques for fast, super-resolution, and multi-view imaging (201) or virtual scanning light-field technologies (202,203).
## 2.2.4 Confocal and multiphoton for intravital imaging
Staying true to the microscopists' motto, "Seeing is believing" (204), intravital microscopy (IVM) addresses the need to observe events in their context (205), which provides complementary information to static 3D tissue phenotyping (206). Depending on the degree of tissue transparency and the required depth of imaging, IVM can be achieved with widefield, confocal, or multiphoton microscopy. Due to the shallow imaging capabilities of widefield microscopy, most studies are conducted using confocal or multiphoton approaches.
In the case of confocal, IVM is sometimes based on the use of a faster alternative microscope called Spinning Disk (SD, Figures 1G,1J207). Here, multiple excitation points are obtained by splitting the laser beam with microlenses on a rotating disk, while corresponding pinholes on a second rotating disk perform optical sectioning. Spinning disk microscopy has been applied to visualise Kupfer cells sequestering E. coli to show how this mitigates neonatal sepsis (208), liver-specific Treg and their re-programming of liver neutrophils (209), peritoneal macrophages (210), patrolling by alveolar macrophages (211), and mechanisms of control of dendritic cells by nociceptors (212). A promising approach to confocal IVM is the recent development of confocal light-field microscopy (203,213), which achieves real-time acquisition of whole volumes (Z-stacks) with micrometre resolution.
Multiphoton microscopy (MPM) combines laser scanning with a multiphoton near-infrared excitation. It is based on the simultaneous absorption of multiple low-energy photons, resulting in the same fluorescence emission as in conventional one-photon excitation. Multiphoton excitation increases the achievable imaging depth (Figure 1D) thanks to the use of nearinfrared wavelengths that are scattered less by the sample, eliminates out-of-focus excitation and reduces phototoxicity and photobleaching (214). MPM contributed to several major discoveries in immunology (see reviews 205,215). The most used type of multiphoton excitation is by means of two photons (also called Two-Photon Microscopy, TPM). Examples of application of TPM in immunology include the visualisation of inflammatory dendritic cells (216)(217)(218) and neutrophil efferocytosis (219) in trachea after influenza infection, innate immune responses in the skin during wound repair (220,221), macrophage aggregation in a peritoneal sterile wound model (222). Moreover, it helped in investigating chemotactic neutrophil migration bias at capillary bifurcations (223), and complement activation in draining lymph nodes following dermal infection (224). In adaptive immunity, TPM contributed to the show that T cell activation occurs in three stages (225), or to elucidate T cell regulation by innate lymphoid cells in the liver (226), corneal tissue-resident T cells localising at the surface of immune privileged eye (190), mechanism of additive cytotoxicity by CTLs (227), dynamic interaction between marginal zone B cells and red blood cells (228), and B cell control of affinity by restraining somatic hypermutation through controlled cell proliferation (229). TPM also aided ex vivo imaging, such as in the visualisation of the role of ATP in limiting protective IgA against enteropathogens (92), or in the visualisation of collagen deposition and mesothelial cell activation in the intraperitoneal gut following microbial contamination (230).
Three-and four-photon excitations have been instrumental to reconstruct the entire depth of a popliteal lymph node (231) or the deep vasculature in brain tumours (232), to the quantification of calcium events in astrocytes in deep portions of tissue (233), and to the acquisition of multichannel data sets (up to 6 channels) in tumour tissues (234). The combined use of TPM and FLIM imaging allows the characterisation of pH and metabolic changes in vivo (235). Finally, to overcome the speed limitations inherent to laser scanning systems, faster implementations have been developed that use a synthetic aperture microscopy to achieve long-term imaging at high speed (236) or with a scan-less multiphoton setup for fast, deep, imaging-based neuron voltage recordings (237). On the other hand, adaptive optics methods have been employed to limit scattering in deep tissues and correct aberrations (233,238). Finally, sample drift or organ movements can pose challenges that AI could address during or after acquisition.
## 2.2.5 Multiplex imaging
The goal of understanding biological function within the complex context of tissueswhich is of particular importance in immunologyled to the development of techniques for visualising and analysing multiple targets or markers within the same sample. Standard imaging setups are usually limited to very few markers at the same time, while multiplex imaging extends the total number of markers in the order of several tens (239).
Current approaches to multiplex imaging include fluorescence imaging, imaging mass spectrometry, or sequencing techniques. In its fluorescence declination, samples are either stained with many fluorophores simultaneously or repetitively stained with fewer fluorophores in many cycles of imaging and fluorescence bleaching (240). Samples are then acquired in widefield or confocal microscopy, to achieve a cellular resolution (Figure 1J). When combining multiple fluorophores at the same time, several techniques have been developed to ensure the separation of highly overlapping emission spectra, either based on hardware, such as the employment multiple emission windows and spectral unmixing algorithms (241,242), or by calculating spill-over with single-stain samples (239). In the case of cyclic imaging, usually two or three fluorophores are used per cycle, with repetitive staining, imaging and bleaching phases, as applied to fixed tissues (243-245), cells (246) and live samples (247). Multiplex imaging techniques require a solid antibody validation, which is addressed also by community efforts (248). Cycling imaging is time-consuming, then a possible improvement is the use of fluorescent tags with DNA barcoding: the sample is stained simultaneously with antibodies tagged with orthogonal single-stranded DNA sequences and then imaged in cycles by using an eraser strand between each cycle (249). Finally, recent and promising advances in fluorescent multiplex imaging use FLIM to increase the number of detectable fluorophores or discriminate the autofluorescence contribution (250). In this regard, techniques using AI to overcome the limitations of low photon budget when performing spectral FLIM imaging are of particular interest (115,251).
Mass-spectrometry based techniques employ a raster-scanned ionising beam to analyse a small portion of the sample that is then associated with a single pixel in the resulting image reconstruction. Material collected can be endogenous, such as proteins, metabolites, lipids, or glycans (252), or exogenous, as in the case of antibody staining with tags suitable for mass spectrometry, such as peptides or rare metal elements. For example, Imaging Mass Cytometry (IMC) and Multiplexed Ion Beam Imaging (MIBI) can reach singlecell resolution, allowing highly multiplexed spatial proteomics (253,254). With lower resolution (clusters of cells), metabolite mapping has been performed with both mass spectrometry and Raman spectro-microscopy (255). Overall, the development of these techniques greatly increased the amount and quality of data extracted from samples. Furthermore, the mentioned techniques can be combined in a multi-omics approach to increase sample information (256)(257)(258). Of interest are methods that couple automated laser microdissection with shotgun lipidomics (259) or with (epi)genomics and transcriptomics, as they integrate imaging, analysis, and hardware feedback steps to extract interesting information for subsequent analysis (260).
Data obtained with multiplex imaging techniques can pose numerous problems regarding analysis due to size, complexity, and heterogeneity. For example, in the case of fluorescence imaging, removing autofluorescence from paraffin-embedded tissues or complex tumour tissues can be hard to achieve (261). Other challenges in the image analysis include segmentation of cells in the complex tissue environment (240), and spectral separation in the case of one-shot imaging with many overlapping fluorophore spectra (262). On the data interpretation side, data clustering and dimensionality reduction are needed to navigate complex multichannel data sets and integrate these data with other multi-omics approaches (263). An example is the integration of multiplex imaging and spatial transcriptomics to follow thymic evolution (264). Finally, sharing code and protocols of the analysis pipeline ensures a dissemination of techniques and best practices, fostering the improvement of data analysis pipelines (118). All these tasks can be approached with standard or AI-assisted image analysis techniques, as discussed in the second part of this Review (see Section 3).
## 2.2.6 Light-sheet microscopy
Light-Sheet Fluorescence Microscopy (LSFM, also called Selective Plane Illumination Microscopy, SPIM) achieves 3D sectioning by illuminating the sample with a thin sheet of light and collecting fluorescence emission in a plane orthogonal to the illumination (265). A volume reconstruction can be obtained by translating the beam or the sample in a single direction or by rotating the sample to perform a tomographic reconstruction. This type of illumination achieves a very fast volume reconstruction with micrometre resolution (Figures 1H,1J), high signal, low photobleaching and phototoxicity (266). In live samples, this technique has been applied to high-throughput live imaging of T cell cytotoxic function against B-cell lymphoma or the interaction of Tregs with gastric tumour spheroids (267).
A notable application of LSFM microscopy in immunology is the reconstruction of fixed, cleared organoids and tissues. Clearing removes the unwanted tissue components and improves the uniformity of tissue refractive index, thus reducing light scattering and improving image quality at high depth (268). LSFM has been applied to reconstruct many organoid types (269). Its use on cleared tissues is particularly interesting for the whole-organ characterisation of immune landscape and vascularisation of the brain (270), clinical identification of melanoma metastasis in the human lymph node (271), and whole-mouse cleared tissue imaging (272)(273)(274)(275).
Lattice light-sheet can image subcellular details with an illumination pattern that achieves diffraction-limited isotropic resolution and high acquisition speed (276). This technique was applied, for example, to the study of the interaction between tumour-associated macrophages and CD8+ T cells (277) or to characterise the effect of antigen strength on immune synapses (31). Given the lower penetration depth of visible light compared to multiphoton illumination, the application of LSFM in vivo has been limited to investigating cleared samples, as with the organoids mentioned above, or in embryo development studies (272,278). However, the development of multiphoton light-sheet systems (279), or the recent implementation of the NIR-II illumination (1000-1700 nm) to light-sheet microscopy opened a window to the feasibility of deep tissue LSFM in vivo (280). On the other hand, a variation of visible-light LSFM called Swept Confocally Aligned Planar Excitation (SCAPE) microscopy (281), was applied to the histopathological characterisation of live tissues from their autofluorescence (282). SCAPE provides information on tissue architecture with cellular resolution, with a strong potential for diagnostic applications.
Overall, LSFM is an exciting field, but numerous challenges remain to be addressed. For instance, the vast amount of generated data renders archiving, pre-processing, visualisation, and analysis considerably more complex than other microscopy techniques (283). Many platforms have been developed to tackle the analysis of these complex and big data sets (284). As outlined in a recent review by Daetwyler and Fiolka (266), we also foresee that lightsheet microscopy, with its fast-imaging capabilities, 3D reconstruction of big volumes and generation of highly informative data sets, will take a central stage in microscopy to image cell-cell interactions in complex 3D structures such as organoids and tissues. Also, the resulting data sets are already pushing the generation of novel data analysis techniques (285,286).
## 2.2.7 Conclusions on the use of light microscopy for immunology
In conclusion, light microscopy techniques can span resolutions from the nanometre to the centimetre (Figure 1I). They significantly advanced our understanding of the immune system, by enabling researchers to visualise complex subcellular structures, cellular interactions, and dynamic processes, for immune phenotyping and dynamic live analyses. Yet, biology occurs across all spatial and temporal scales, while current techniques can only see a portion of these events (25). To cover these different scales, we need both progress in imaging techniques, as well as automated analyses that can inform and guide the capture of events across scales in real-time.
## 2.3 Image analysis
Image analysis is a crucial component of microscopy research, enabling the extraction of quantitative data from complex visual data sets in an unbiased manner (287). The standard toolbox for image analysis comprises tools for image preprocessing, segmentation, tracking, and quantification. Automating this process ensures unbiased data analysis and simplifies compliance with good practices for data and image acquisition reporting (288).
Image preprocessing seeks to minimise noise, enhance contrast, correct geometric distortions, and improve resolution, thus facilitating the subsequent image analysis steps. Denoising techniques improve the signal-to-noise ratio (289): methods vary from standard Gaussian blur and median filter to more advanced techniques such as 3D block-matching (290), non-local means (291,292), and wavelet transforms (293). Improvements in contrast and resolution can be achieved with deconvolution techniques, where the information about the point spread function of the microscope is used to remove the signal contribution from out-of-focus planes and surrounding signal sources (294). Resolution and contrast improvements can also be obtained by acquiring the same image with slight changes in the illumination beam (295,296), with general algorithms considering the noise distribution (150, 151), analysing fluorescence fluctuations with algorithms such as mean shift vector analysis (MSSR, 155), or by deconvolution, like in SUPPOSe (297) or B-SIM and Sparse-SIM for SIM images (298,299). Image registration corrects image distortions and time drift: this is especially useful when reconstructing a volume in a mosaic (300)(301)(302), aligning images in case of sample drift, as needed in intravital microscopy (303)(304)(305)(306) or aligning images acquired with different modalities, such as in the case of CLEM (51). Lastly, crosstalk correction improves channel separation, eliminating unwanted spectral bleed-through between channels (4, 174), while spectral unmixing techniques use the characterisation of the emission profile to separate many fluorophores with overlapping emission spectra (241). These approaches are particularly useful in multiplex imaging (239) and when subtracting unwanted autofluorescence contributions (242). Other methods for spectral separation include phasor analysis based on fluorescence spectral data, or on fluorescence lifetime (307). Overall, effective image preprocessing greatly simplifies image segmentation, ultimately improving the extraction of quantitative data from microscopy images.
Segmentation of image data is the process of separating the pixels of the background from the pixels of interest, labelling the objects (e.g. organelles, cells, tissue areas) so that properties such as geometrical descriptors can be measured, or intensity statistics be calculated. Image segmentation is at the basis of most automated analysis workflows (308). Segmentation techniques can be broadly categorised into two main groups: region-based and edge-based methods. Region-based segmentation algorithms group together pixels with similar attributes (e.g., intensity, colour) to form homogeneous regions within an image. These algorithms often use intensity thresholding (309), clustering (310), or watershed (311) to identify meaningful segments. On the other hand, edgebased segmentation focuses on identifying boundaries between objects in an image by detecting abrupt changes in pixel attributes, such as intensity or texture. Common edge detection methods include Canny (312,313), Sobel (314), and Laplacian of Gaussian (LoG) operators (308). The choice of segmentation technique may vary depending on the specific application, where factors to consider include image complexity, object shape, size, and contrast with respect to the background. Object segmentation can then be followed by object classification according to some measurable property, such as object position or shape factors.
Object detection localises objects or regions of interest in an image. The task typically involves identifying the object to be detected (classification) and determining its position in the image (localisation). Object detection is frequently employed to recognise areas of interest, such as a compartment in a cell or tissue (315), and to identify cells in time-lapse microscopy movies for object tracking (316).
Tracking allows the study of temporal dynamics, such as cellular or subcellular movements. Tracking objects in a movie is a two-step process, where object detection and segmentation are followed by object linking between frames (317,318). Manual or semi-automated tracking software has dominated the scene in studies of immune cell mobility upon antigen presentation, in cell culture experiments, or in lymph node imaging (319). However, when temporal sampling cannot be done at high frequencies, or the linking process is ambiguous, deep learning techniques may prove helpful in enhancing the effectiveness of classical methods (316,320).
Recent advances in bioimage analysis are significantly broadening its accessibility, allowing researchers to leverage powerful techniques with reduced reliance on programming expertise and lowered computational resource demands. These developments are driven by a growing trend toward simplified user interfaces (321)(322)(323)(324)(325)(326), and standardised analysis protocols for light microscopy (327) and electron microscopy (57). However, while these tools streamline many routine analyses, complex or novel research questions often require more sophisticated approaches. This is where the role of the bioimage analyst remains crucialbridging the gap between readily available software and advanced techniques, and facilitating the development of custom solutions to address unique research challenges (328).
## 2.4 Impact of open-source software and open hardware in image acquisition
Open-Source software (OSS) has a key role in bioimage analysis because the access to source code enables any researcher to develop customised workflows, even with a limited knowledge of programming languages, thus guaranteeing more transparency and reproducibility (329).
Image quantification has been made easy in the past decades by many graphical user interfaces (GUI) suites, among which the most renowned include ImageJ or FIJI (330), CellProfiler (323), napari (331), QuPath (332). The key aspect that makes these GUIs widely adopted is the abundance of scripts and plugins, together with the possibility to access the source code and develop custom bioimage analysis solutions, whether implemented as point and click interaction or as a script.
A fundamental role of facilitator for bioimage analysis based on scripting has been covered by development environments such as RStudiofoot_2 (based on R language), JupyterLab (333, based on Python) or visual programming suite KNIME (334). Because of the richness of the Python environment, in terms of availability of packages, many bioimage analysis solutions have been developed as Jupyter notebooks, especially in the context of machine learning (ML) and deep learning (DL), where code modification might enhance the adaptation to specific image data (118,335,336). The availability of complete notebooks where all the steps of a workflow are explained with code comments facilitates the execution by the end user, learning, and reproducibility (335).
The integration between the full control of microscope motorisation (337), image acquisition, real-time (or offline) image analysis, and the possibility to drive a new image acquisition, based on the result of the analysis, constitutes the backbone of what is called feedback microscopy (338), also referred to as smart microscopy (339). The purpose of such integration is to allow the adaptive imaging of the biological sample in the spatial and temporal dimensions (340).
In Figure 2, we present a possible workflow of feedback-based microscopy to identify infected cells and acquire them at higher resolution, minimising the overall acquisition time. A multichannel fluorescence image, including a nuclear marker and an infection reporter (Figures 2A-C), is used to identify all the cells in the field of view (Figure 2D) and measure the level of an infection reporter (Figure 2E). Using a ML clustering algorithm (e.g. k-means clustering), cells are classified based on their infection state (Figure 2F). Then, cells are selected depending on their level of infection (Figure 2G), and the microscope is instructed to navigate to the cell position (Figure 2H), switch objective (Figure 2I), and acquire a higher-resolution image (Figure 2J). This type of workflow presents the advantage of scanning larger areas to increase the number of inspected cells and use higher resolution imaging only for the infected ones, which are identified with an unsupervised ML algorithm.
Examples of software tools for feedback-based microscopy include AutoscanJ, for the detection of mitotic cells or chromosomal anomalies, based on the same principle of rescanning cells of interest, previously detected with lower magnification (341) or the real-time drift correction in intravital movies (342). In the field of RT-EM, a similar approach has been developed by SerialEM software (14) and its Python interface pyEM (343). This approach was used in the contests of immunological research to show, with tomographic reconstructions, that plasma cells in patients with multiple myeloma display elongated centrioles (344). Another application has been developed for combining light microscopy and FIB-SEM (336). In single-particle cryo-EM, automatic acquisition is even more important because thousands of images are needed to perform a structural reconstruction (17,345). In cryo-ET, machine learning approaches were used to fully automate in-situ cryo-ET workflow (346).
While some scripting tools, such as ImageJ macro language, are very popular in the bioimage analysis community (347), the complexity of back-end programming languages (e.g. C or Java) to develop software plugins may hinder the quick implementation of novel ideas. To facilitate the use of feedback microscopy, projects like Pycro-Manager (348) or pymmcore 2 have been created to implement translation layers between programming languages (in this case, Python can be used to write scripts rather than Java). On the other hand, the Open-Source Hardware (OSH) movement has allowed the implementation of cheaper solutions for image acquisition compared to proprietary microscopy software. Projects like Micro-Manager (349) to control microscope hardware have revolutionised the field, decoupling the need for commercial licences to operate devices from the mere possession of the equipment. In addition to gaining control of microscopy equipment, the possibility to trigger and modulate the image acquisition with plug-in electronics, for example based on Arduino 3 or Raspberry Pi 4 development boards, has widened the possibilities to customise every microscopy platform. However, technology development requires the developers or early adopters to carry the risks of investing resources in technologies that might have limited or delayed benefits. Then, the advantage has to be identified either in the reduced cost of existing open technology or in access to bleeding-edge techniques, which might be rewarded in terms of scientific publications (350).
Finally, Computer-Aided Design (CAD) for machining or 3D printing of microscope components or auxiliary devices has improved the use of resources to run microscopy experiments.
Examples include the open optical setup of light-sheet system openSPIM (351), the possibility of fully 3D printing experimental tools or the wide database of open hardware projects developed by the imaging community (for example, by the LIBRE hub 5 ). These can include accessories such as syringe injection motors, sample supports, frames for optical filters, enabling components based on electronics (352), or even part of the microscope body, etc. (353). The adoption of OSS or OSH solutions is also strictly dependent on their discoverability, modularity, and the standardisation of the software interface (354).
The revolution of openness in scientific software and hardware is not necessarily in opposition to the business model of microscopy companies (355). In fact, at the request of bioimaging researchers, many companies offer support for the use of open software, such as OMERO (supported by Glencoefoot_7 356), or have opened part of the software by offering an API to interact with some of the GUI modules, such as Zeiss 7 with APEER platform for deep learning (357) or Abberior 8 with the possibility to reprogram the hardware configuration (358). In addition to the highly beneficial effect on the broader research community, we believe that companies can also benefit from the openness of both software and hardware. This applies whether resources for science are scarce or research is well funded, because there is always a business model that can be adapted to provide a service for less experienced users (355), and
## the wide adoption of open imaging solutions by companies enlarges potential customer markets.
The need for openness is even more pressing when AI solutions are implemented, as AI methods are inherently based on probability and as such not prone to reproducibility, and often the general audience employs such methods without a thorough understanding of their applicability and limitations. The second part of this Review aims to clarify some of the concepts and uses of AI for microscopy and immunology.
3 Artificial intelligence for microscopy with applications in immunology and virology
## 3.1 Historical introduction to AI
The concept of artificial intelligence takes root in Leibniz's characteristica universalis, a common unified language of pure thought in which every language could be translated, and in calculus ratiocinator, a machine capable of replicating that language (359). Computers were meant to take a set of rules and input data and return some output, but could they think autonomously or even generate novel ideas (360)? This question is still guiding research in the AI field. In the 1940s and 1950s, progress in computational capacity motivated people to explore applications in the domain of pattern recognition, where the human brain excelled. In a seminal paper, McCulloch and Pitts (361) proposed the model of a network that took inspiration from the structure of the brain. This network was composed of a single input and output neuron, with an activation state that would contribute to the final output of the network. Later, developments on this original idea extended network complexity to a multi-layer network (362) and developed algorithms to train networks with more than one layer (backpropagation, [363][364][365].
In today's technologies, artificial Neural Networks (NN) are built from a collection of nodes (neurons), operating a set of transformations on the input data to learn different representations of it. Nodes can be aggregated in layers and are connected by activation functions (synapses) computing a weighted sum of their input data. When the NN is trained, some connections get stronger, causing them to acquire a higher weight, while others get weaker, thus reducing their weight. So, NN training is essentially a problem of optimisation of parameters (366,367). Training occurs iteratively: at each cycle, the network's weights are optimised, and a loss functionan objective measure of training successis measured (368). This process continues until a stopping point, such as reaching a specified value of the loss function or after a certain number of iterations. A network may perform poorly because of lack of convergence to validation data (underfitting) or lack of generalizability (overfitting). The choice of loss function is an important part of model design (369), along with the definition of layers and their connection types, which are collectively referred to as network architecture. For example, annotated tumour areas in tissue slices are used as ground truth, and the NN predicts which areas could be classified as tumour in the same slice (370). The loss function estimates how precisely the network predicts tumour areas. After the training phase, the network is applied to predict labels (tumour or healthy) on new tissue slices.
## 3.2 Machine learning or deep learning?
Machine learning (ML) is a subfield of AI (Figure 3A) that enables systems to learn from data without being explicitly programmed. It focuses on building models or algorithms that can make predictions or decisions by identifying patterns in the data and using them to improve performance over time (378). There are four main types of ML approaches: supervised, unsupervised, self-supervised, and reinforcement learning. Supervised learning is trained on a previously labelled data set that represents bona fide the desired outcome, so both the input and desired output are known. This highlights the importance of preparing a training data set that is most representative of the desired outcome, a simple task in principle but one that should be performed with great care (379). By contrast, unsupervised learning finds a structure from the data itself without any prior labelling information. This is a commonly employed technique when unbiasedly clustering information (Figure 3B: middle row) and grouping differences in classes. For example, k-means clustering or Principal Component Analysis (PCA, Figure 3B: middle row) belong to this category. Self-supervised learning finds a classification by predicting or completing parts of its input, creating labels automatically from unlabelled data. For example, a self-supervised model has been developed to automatically learn semantic relationships between genomic data and improve tasks such as gene annotation or the role of polymorphisms (380). Lastly, Reinforcement Learning (RL, Figure 3B: bottom row) involves an agent interacting with an environment and learning through trial and error. It's often used in broad AI tasks, such as AI-assisted game-playing and autonomous driving systems (376). Any ML workflow comprises a set of input data, a model architecture, and one or more loss functions. ML models can range from lowcomplexity models with few layers of data transformationshallow learningto higher complexity models involving many layers and many connections between themdeep learning.
Deep learning (DL) is a subfield of machine learning (Figure 3A) where many layers of data representation are connected to create a complex model with multiple levels of abstraction. Even though many foundational concepts and algorithms have been developed in the twentieth century, practical advancements in DL are relatively recent. Three practical steps have contributed to these advancements and to a general renaissance in the field of AI (368). First, the development of small yet significant algorithms enhanced how these deep stacks of layers can be interconnected (366,367). Second, bigger storage was available to host the growing training data sets. Lastly, cheaper hardware increased computational power. In fact, DL models are based on simple additions and multiplications of big multidimensional arrays of data (called "tensors" in mathematics), and these operations can be easily parallelised. The development of powerful graphical units (GPUs), originally designed to improve the gaming experience but with the ability to be programmed for massively parallelised calculations and scientific computing, led to the implementation of GPU-based NNs (381,382). Today, along with gaming GPUs, researchers can also leverage dedicated GPUs, optimised for DL tasks, and specialised hardware such as tensor processing units (TPUs), with the potential for requiring less computational resources and becoming an integral part of all domains of science, including microscopy. However, practical implementation of these networks still requires programming skills, with Python being the primary development language; popular frameworks include Tensorflow/Keras (368) and Pytorch (383). DL algorithms are affected by hardware bottlenecks in steps that are not hardware accelerated, therefore some attempts have been made to leverage alternative electronics boards like fieldprogrammable gate array (FPGA, 384), which offer the flexibility of reconfigurable circuitry. For applications requiring low power consumption, FPGA have shown to be from 3 to 5 times more efficient than GPU per processed image, for tasks such as image compression (385). The advent of "liquid" and more efficient hardware such as FPGA will dictate the pace by which AI methods are implemented in microscopy and other fields.
To conclude, ML and DL techniques include computational systems capable of learning from data. Shallow ML systems continue to be utilised for their rapid training, minimal computational resource requirements, and low complexity, allowing for a complete understanding of how they operate. In contrast, DL systems require significantly more computing resources but have considerably higher capabilities, thus aiding all areas of microscopy, from experimental design to image acquisition, analysis, and data mining (386). These two types of systems are often used together in algorithms mixing classic programming, traditional ML, and DL according to the task (this permeability is reflected in the dashed lines of Figure 3A).
## 3.3 Image-based machine learning
Image-based ML methods can preprocess images or segment, detect, and track objects within multidimensional microscopy images. The most used approach is supervised learning, which 371) can be used for crowded nuclei segmentation, Feed-Forward Neural Networks (FFNN) that have been adopted for cell counting (372), Autoencoder networks (AE) that have been used for several analytical tasks such as denoising and spatial interpolation in spatial omics data (373). For ML, from left to right: Clustering methods such as DBSCAN for Single Molecule Localisation Microscopy, Decision Tree methods like random forest for cell classification (322), Principal Components Analysis (PCA) for dimensionality reduction in data analysis including multiple cell measurements (374). For AI, from left to right: Natural Language Processing (NLP) for data mining and code generation (375), Reinforcement Learning (RL) for autonomous improvement of AI models and hardware control (376), and Generative Adversarial Networks (GAN) used for data augmentation to improve the efficiency of image segmentation and interpolation of imaging data sets (377).
employs a manually segmented data set to train the model. Traditional, shallow ML techniques for image segmentation include Support Vector Machines (SVM) and Decision Tree classification (Figure 3B: middle row). Such ML models can yield good results across various applications and are featured in numerous open-source image analysis platforms (ilastik (322), Fiji/ImageJ with WEKA (387) and LABKIT plugins (388), QuPath (332), CellProfiler (323), MIB (389). They are userfriendly, can be used even without programming experience, and demand fewer computational resources compared to DL (388).
On the other hand, DL approaches for image processing have expanded the range of problems that can be solved (366). In the case of image-based methods, they are primarily employing a type of NN called Convolutional Neural Network (CNN, Figure 3B: top row, 359,379). CNNs use layers to extract hierarchical representations of input data, in a way similar to how we learn information on an object by viewing it from different distances or angles. These layers implement two data processing functions: the convolutional filter (hence the name) and the max pooling function. Convolutional filters act like a magnifying glass that scans an input image while applying different kernels (a small matrix) to the image at each position. These kernels help identify specific features within the image, such as edges or corners. The result of this operation is referred to as a feature map, which highlights where these features exist in the image. Max Pool operations are used after convolutional layers to reduce the spatial size of the data while retaining important information, thereby making the network more efficient for computational analysis. Here, we will survey the landscape of current applications of image-based DL methods to the microscopy modalities that we previously discussed, highlighting, where existing, their applications to immunology or virology (Table 1).
A major theme in DL applications to many microscopy techniques has been finding ways to increase the wealth of extracted information and overcoming the limitations of the specific techniques, such as increasing resolution without sacrificing acquisition speed. In cryo-ET, DL has been used to learn structural information from single-particle cryo-ET analysis (390) or to achieve isotropic resolution without the need for subtomogram averaging (391). In single-particle cryo-EM, DL aided particle model-building by creation of intermediate-resolution maps (392) or model building automation (393). In virology, DL with single-particle cryo-EM has been instrumental to the characterization of tegument architecture in human cytomegalovirus (394). In super-resolution, and specifically in localisation microscopy, DL has been used to increase the acquisition speed, by reducing the number of images needed to reconstruct the structures of interest (109), or to help in localising multiple adjacent emitters in 3D, thus improving volumetric reconstructions (110). In immunology, DL with STED has been applied to identify Zika virus reorganization of the endoplasmic reticulum (111). In SIM microscopy, DL can increase resolution and speed (395)(396)(397), thus better capturing live-cell events with lower phototoxicity. In confocal and spinning-disk microscopy, DL can increase image resolution (113,117), reduce optical aberrations (114), and improve FLIM lifetime determination with low photon budget, as in fast live-cell imaging (115). In TIRF, DL helped in improving single molecule FRET (smFRET) by analysing single molecule traces (112). In wide-field microscopy, DL was used to enhance the resolution and optical sectioning capabilities (30,123,124), yielding confocal resolution while improving speed. In intravital, microscopy, DL approaches combined with two-photon excitation and adaptive-optics aberration correction have improved subcellular resolution without sacrificing acquisition speed (133). Finally, DL has been applied to light-sheet microscopy for physicsinformed deconvolution, i.e. a combination of DL with optical information on the microscope setup (134). In light-sheet and confocal microscopy, DL also provided axial resolution enhancement, by learning from unpaired high-resolution, 2D confocal images and low-resolution 2D images from other planes (113).
Methods to improve resolution, signal or speed often apply to a specific imaging modality and do not translate well to other image modalities. Recent DL models have tried to provide a more general approach, for example when restoring fluorescence images from all imaging modalities (398), improving resolution without additional data acquisition (399), interpolating images between frames (400) or when performing object detection (401).
Major limitations of fluorescence microscopy are photobleaching, phototoxicity and limited speed when acquiring multiple channels. DL methods can overcome these limitations by providing in-silico labelling of transmitted light images (126)(127)(128)(129). For example, DL with in-silico labelling of brightfield images has been used to improve the tracking in chemotaxis experiments (131), or to predict the lineage choice of differentiating hematopoietic progenitors (130). Finally, DL has been applied to phase, label-free imaging (125), or to achieve fast, volumetric live-cell microscopy of bioluminescent probes (402). In immunology, DL with optical diffraction tomography has been instrumental for label-free tracking of immunological synapse of CAR-T cells (403).
DL methods are now widely used for preprocessing, segmentation, detection and tracking tasks (379,404). In image preprocessing, DL models are extensively used to denoise fluorescence images, as indicated by the many examples in the literature (290,398,(405)(406)(407)(408)(409)(410)(411)(412)(413)(414). In immunology, DL denoising was instrumental to improve contrast in Imaging Mass Cytometry, thus helping in characterizing the phenotype of immune populations in human bone marrow samples (415). DL helped in separating channels for filter-free imaging (125, 416,417), and to assist tracking by improving linking accuracy (316,320,418). Also, DL is aiding cell phenotyping from multiplex immunohistochemistry images, for example to characterize tumour microenvironment in lung cancer (121) or pancreatic ductal adenocarcinoma (122). Finally, DL models can assess the quality of fluorescence images and identify artefacts (419). Moreover, DL has been applied to denoise low-dose cryo-TEM images (420).
DL is employed in supervised cell segmentation, such as in the case of general models U-Net (421), StarDist (371), Cellpose (422) and Segment-Anything Model (SAM) models (423,424), which are available as plugins in many open-source and proprietary image analysis software. Specialised models tackle intracellular organelle segmentation (327,425), segmentation of extracellular vesicles in TEM (103), HIV-1 virions in TEM (104) or mitochondria in FIB-SEM (426). In immunology and virology, DL has been applied to FIB-SEM images of SARS-CoV2 patient-derived platelets to segment a-granules or mitochondria (105). Ligh-sheet microscopy of lungs together with DL-based analysis allowed the spatial profiling of nanoparticle delivery to alveolar macrophages (135). Image segmentation is a challenging part of image analysis in multiplex imaging, where cells are often densely packed. In this case, DL was employed to perform cell segmentation, thus improving single-cell feature extraction (118,119), or to perform a detection-based classification, therefore providing phenotypic analysis without segmentation (120).
In detection tasks from fluorescence images, DL helped in recognizing apoptotic cells from intravital multiphoton movies (427), phototoxicity in widefield time-lapse experiments (428) or, together with widefield high-content screening, to detect virusinfected cells and predict if they will follow a lytic or non-lytic infection (429). Furthermore, DL with confocal microscopy helped in classifying expression of TLRs from PBMCs of HIV-positive patients under ART therapy (430). In cryo-ET, CNNs are helping annotation and feature extraction for in situ identification of structures of the molecular components of interest (431), or template matching, i.e. detection of objects with an arbitrary shape, which is the most widely used approach in cryo-ET for particle picking (106). Still in cryo-ET, DL has been applied for finding macromolecules in cellular 3D tomograms (28). DL detection algorithms can also support feedback microscopy in real time (432) by automatically detecting events to guide acquisition, generate feedback, and predict cell fate.
In tracking, CNNs with intravital multiphoton microscopy have been used to accurately measure the position and shape of CD4+ T cells interacting with plasmacytoid dendritic cells in vivo, aiming to study interaction differences in lupus nephritis (132) or to link cell tracks in intravital imaging of leukocytes (433).
One of the limitations of supervised learning is the generation of accurate training data sets, which is, in most cases, a manual task that can be time-consuming and still prone to bias. A possible approach to overcoming this limitation is the use of self-supervised methods. For example, information from the OpenCell database was used to cluster proteins into organelles and individual protein complexes (434). Similarly, in another study DL was used to segment mitochondria, based on a training data set that was generated with DL (435). Such simulation-supervised approach could be, in principle, generalisable to other organelles or even to cellular segmentation in tissues. Self-supervised or weakly supervised models are also employed for cancer prognosis and diagnosis in pathology slides (436).
Overall, image-based DL methods are assisting a wide range of microscopy techniques in tasks from acquisition to image analysis and data extraction (Table 2). Many of the above examples are widely applicable to different types of samples, including those related to immunology or virology.
## 3.4 Data-based machine learning
Data-based ML can analyse, interpret, and learn from data. In this Review, we refer to data-based methods as the ones that can be applied generally to numerical or categorical data without the specific need for data to be generated with imaging techniques. For example, data-based ML is used in microscopy when clustering data extracted from images with the previously described imagebased methods (442,444), or when combining information from microscopy with text data generated with other methodologies, such as genomics or proteomics data (445). These tasks can be approached using traditional, shallow ML or DL.
Techniques using traditional ML include linear regression, logistic regression, and decision trees, such as Random Forest. For example, logistic regression was employed to predict MHC ligand, where a binding model and an antigen processing model were combined, and results were classified according to logistic regression score (446). Random forest was applied to analyse T cell-dendritic cell interaction in a lupus nephritis model (132).
Instead, unsupervised learning techniques are employed when the desired outcome is unknown or input data are not labelled. In this case, unsupervised learning can help clustering data in groups. Notable examples are the K-means, DBSCAN, and an unsupervised version of random forest algorithms (378). Clustering techniques have been used in high-throughput screenings to highlight differences between biological conditions when segmenting and measuring cells (374), or, in immunology, to cluster signatures and perform neighbourhood analysis in multiplex imaging in tissues (206,439,447) and single cells (246).
Another set of techniques, called dimensionality reduction, is used to group variables into "super variables". Techniques falling in this category are PCA, t-SNE, and UMAP (378). Dimensionality reduction has been used to classify cell cycle and disease progression after feature extraction with CNNs (448), to segment touching cells in confocal and two-photon microscopy (449), to group clonal distribution of CD4+ T cells in gut epithelium following Listeria monocytogenes infection (450), or to obtain behavioural signatures of immune cells in intravital inflammation models, guiding the discrimination between pathogenic and non-pathogenic phenotypes (440). Also, dimensionality reduction techniques are essential in multiplex imaging when grouping cell phenotypes (240,451).
DL for data-based methods can employ different types of architecture depending on the purpose. An Autoencoder Network (AE) is a type of NN that learns to encode input data into a lowerdimensional representation and then decode it back into the original form, thus learning meaningful representations from the data. This can be helpful for tasks like dimensionality reduction or feature extraction. AE (Figure 3B: top row) has been used to combine low-dimensional representations of scRNA data, generated using large language models, with actual single-cell scRNA-seq data from different species to create a supergene classification that can bridge differences between individual single-cell experiments and different species (452), or to create deep generative models for spatial-omics analysis that can take into account the spatial relationship information (373). A Feed-forward neural network (FFNN, Figure 3B: top row) is another type of NN where information flows only in one direction (forward) through layers of interconnected nodes. FFNNs are commonly applied to regression and classification tasks, as they can learn complex nonlinear relationships between inputs and outputs. For example, a feedforward network was used to classify cell tracks in 3D biomimetic gels of immune cells co-cultured with breast cancer cells in organ-on-chip (453).
Recurrent Neural Networks (RNNs) excel at processing sequential data like natural language text, where the context of previous words influences the use of subsequent ones. After dividing data into small chunkscalled tokens -, RNNs process them recursively to generate the most likely information based on the previous information. For example, this type of network could predict the cell lineage of hematopoietic cells from brightfield images by extracting time signatures of cells from image features extracted with a CNN (130).
On the other hand, transformer networks weigh the importance of input tokens to construct a connection map without requiring sequential processing (454). They are at the basis of the Natural Language Processing (NLP, Figure 3B: bottom row) chatbots used today, such as ChatGPT, and are also called Large Language Models (LLMs). These tools can serve as a valuable resource aiding researchers in designing microscopy experiments (455) or drafting algorithms to implement the ML techniques described here. LLMs can be applied to data mining in scientific data sets, such as those generated from single-cell omics (as reviewed in 456), and some attempts exist to apply it to bioimage analysis (443,457). We also foresee that these techniques will be increasingly integrated in microscopy hardware, for AI-assisted sample exploration and acquisition.
Another type of generative network is called Generative Adversarial Network (GAN, Figure 3B: bottom row). This network comprises two networks, one generating new data and the other evaluating its suitability as output. The generating network challenges the evaluating network, thus introducing an element of randomness and "creativity" in the output (458). This broad class of networks can be used to generate models of protein structures (the general AlphaFold (459) and its open version OpenFold (460), or a more specific version for proteins of the immune system (461)) or to reconstruct molecules from cryo-ET tomograms (390,462). Generative neural networks informed on the features of highly metastatic melanoma by "reverse engineering" a supervised CNN for cell classification. In this example, a CNN is initially trained on patient-derived melanoma xenografts to classify them based on their metastatic capability. Then, a generative neural network is used to create in silico cell images with exaggerated features, which are then used to analyse which features in the CNN are most prevalent (463). This approach is particularly intriguing as it uncovers new quantitative insights within the hidden features of DL, thereby providing information that could potentially lead to the generation of new scientific hypotheses.
In summary, we outlined some applications of data-based ML techniques for microscopy (Table 2). The field is vibrant and complex, and evolving at a fast pace. Shallow ML and DL can be employed to cluster information and combine microscopy data with multi-omics data (29,441,442), or to predict molecular biomarkers from pathology images (464). These technologies could aid vaccine design, as outlined in Hederman and Ackerman (465) or to improve antibody design (466). As such, a dialogue between computer scientists, microscopists and immunologists is fundamental.
## 3.5 Impact of open-source software in AI deployment and democratization
AI can hugely facilitate the analysis of large image data sets and the characterisation of rare biological events. Because OSS solutions can rely on the contribution and critical judgement of the wide imaging community, they are indispensable for the development of
## Clustering of phenotypes or behaviours of immune cells
Traditional ML for phenotyping multiplex images (206,439). DL for clustering phenotypes in high-content screening (374). DL to cluster behaviours from intravital cell dynamics (440) Integrating microscopy data with multi-omics (29,441,442) Dialoguing with imaging data and software (443) The table summarises the main approaches employing traditional ML or DL for selected tasks in image analysis and immune phenotyping.
trustable AI applications for bioimage analysis. In this regard, several software tools have been deployed during the past years, both as plugins for the major software GUI such as FIJI (StarDist (371), Cellpose (467), DeepImageJ (468)) or napari (SAM (423)), or as code notebooks (ZeroCostDL4Microscopy (335)).
In the previous sections, we outlined the main DL applications to analyse bioimage data, many of which are distributed as opensource software. Although some of these applications were not specifically developed with immunology in mind, their usage can significantly benefit immunological imaging. For instance, existing code can be adapted for specific biological questions, or their training data can be used to enhance other domain-specific DL models.
Furthermore, publicly available image databases are key in truly open-source AI models (469), as outlined by the Open-Source AI Definition 9 . In this sense, data sets specific to the immunological field (319), or for broader imaging purposes, such as the ones hosted on BioImage Archive (470), can constitute a valuable resource of image data to test AI software tools and foster immunological research. These data sets should follow the "Findable, Accessible, Interoperable, Reusable" (FAIR) principles (471).
Finally, the use of LLMs or other AI methods to aid software creation (443) will constitute an essential part of AI deployment and democratisation, as it empowers every scientist, even without programming experience, with the "wisdom of the crowd" provided by AI training data sets that can summarise a vast amount of human knowledge.
## 3.6 AI for instrument control
The role of AI for instrument control and automation can be delineated in at least two different ways: the first being the support of AI in automating the development of open-source code (443) for feedback microscopy, while the second involves utilising the AI-based methods described above to interpret the image data, thereby revealing information that can be used to redirect the acquisition process.
The use of open-source platforms Micro-Manager (349) or Pycro-Manager (348) for microscope control can be integrated with Python packages such as Scikit-Image 10 (472) for image segmentation and Scikit-Learn 11 (473) to run ML tasks and feed results back into the acquisition software. As a practical example, DL segmentation methods have been used to identify cells and set the correct acquisition parameters (474) or switch microscopy modality (31) to image immunological synapses. Furthermore, cost-effective open hardware facilitates possible integrations of the acquisition microscope with AI feedback tools, for example to control anaesthesia, temperature, and humidity in intravital imaging (similarly to approaches tested in the clinics, 475,476), or correcting the state of the optical system by acting on adaptive optics to minimise the loss of signal (238).
The implementation of AI for instrument control is automating the execution of precise and complex hardware tasks, shifting the troublesome duties from the human to the machine. In our view, rather than totally delegating the control of the experiment and the handling of expensive instruments to the automatic agents (as a kind of hardware/software AI-equipped decision maker), the implementation of AI tools should work as advanced technology to help the human researcher in steering the course of the experiment. In this regard, the integration of large language models and feedback microscopy is foreseen as the future evolution of microscopy, where the user doesn't necessarily need to be highly skilled in all the aspects of microscope acquisition, hardware control, bioimage analysis: a microscopy platform could accept human language instructions and convert these into hardware control operations to provide the requested type of data sets (477).
## 4 Discussion
In this Review, we surveyed the primary applications of light and electron microscopy in immunology and virology for preclinical research, outlined the concepts of AI, ML, and DL, and explored their current uses in microscopy, image analysis, data analysis, and feedback microscopy. Although we mentioned studies employing microscopy and AI in immunology, the intersection of these fields remains largely unexplored. For instance, many of the tools discussed are designed for specific applications, and few attempt to integrate multiple imaging modalities or tasks. Lack of generalizability, when not due to training issues, is a major challenge in the current research on DL and microscopy. On this topic, Kawaguchi et al. showed that analytical insights into building more generalizable architectures could be drawn when using specific loss functions and concluded on the importance of human reasoning on the physical properties and engineering principles of the specific problem at hand (478). We believe this capability could arise either through AI-assisted multimodal visualisation or through a combination of direct visualisation and AI computational modelling of structures or dynamic events.
DL networks concatenate many hidden layers to generate a rich output, and although these layers are just composed of numbers and simple operations, understanding how the output relates with their inner functioning is exceptionally challenging. Also, the mathematical structure of DL makes it prone to hallucinations, i.e. the generation of plausible but incorrect output (479). These two problems could be seen as epistemically relevant if we treated the output as scientific knowledge in its own right, without further experimental verification (480). Rather than seeing this as a problem, we think that integrating DL with additional local data, for example from microscopy or other techniques, could enhance our ability to interrogate data and generate additional perspectives (27,481), potentially inspiring new scientific ideas, to be later supported by rigorous verification, or highlighting the limitations of current theories (482). Finally, the use of AI in real-time contexts requires an evaluation of the possible outcomes of AI hallucinations, with respect to model reliability and consistency. In this regard, the openness of the AI-decision making process, the Frontiers in Immunology frontiersin.org integration with traditional techniques and a dialogue with the human researcher still have a prominent role. While one might be tempted to attribute some level of understanding to AI models, it is essential to recognise that these models merely process numerical representations of data generated through mathematical transformations (368). As such, their ability to think is as limited as the size of the data set and the types of possible transformations. This highlights the importance of dataset preparation to achieve reliable outcomes, as well as the need of a community effort for more curated, freely accessible, "FAIR" (471) microscopy data. The careful evaluation of possible cognitive biases during curation of training datasets is very important (483). For example, researchers can assess the imbalance between classes, i.e. the under-representation of certain types of conditions during data generation, the uniformity of data acquisition across instruments or laboratories, the quality of annotations or the methods used for data augmentation (484). Also, the application of AI might require considerations about data privacy and the openness of AI models, especially when relying on external services for the AI processing.
Steps to make AI more accessible and broaden its usage include creating efficient models that require less hardware resources and could be used on low-cost computers, or designing better user interfaces to guide the user in all the phases of AI implementation, such as with data quality assessment and data preparation, with the choice of NN architecture, with model validation and during model usage. For example, projects like DeepImageJ (468) and its model sharing platform BioImage Model Zoo 12 are going in this direction. Finally, a rigorous determination of the amount of scientific data needed to obtain reliable training is required, as well as techniques for performing automatic choice of DL architecture and training (485).
A cross-disciplinary approach that includes skills in biology, microscopy, electronics, and software programming is necessary for implementing AI-based hardware and software tools. This approach could help shaping open, local models with efficient use of hardware resources, to the benefit of real-time hardware control. To achieve this fully integrated use of AI tools, LLMs can help in data mining and code drafting tasks but are still limited in their capabilities to precisely manipulate factual knowledge, for example to provide advice on microscopy. A step towards more specific and fact-based AI tools is represented by Retrieval-Augmented Generation (RAG) models (486). Ideally a general model, for instance a chatbot such as BioImage.IO (455), could understand better the initial request for help posed by a human researcher (e.g. "find the organelle in these cells") and could drive a more specialised model, for example a CNN, to perform specific tasks (e.g. segmenting specifically mitochondria).
Looking ahead, we can imagine a future where multiple AI models, or agents, could act on specific parts of the microscopy process (487). New DL agents could integrate fact-checked advice on experiment design with code generation models, AI-based hardware control and analysis models, to assist the user across the whole research cycle. Integrating AI, imaging data science and microscopy automation will allow the automatic monitoring of the sample, the detection of anomalies, and the adaptive change of hardware to guide the acquisition of key biological events, thus enhancing our visualisation capabilities across scales and our systemic understanding of the immune system.
## References
1. Opal (2009) "A brief history of microbiology and immunology. Vaccines: A Biography"
2. Blevins, Bronze (2010) "Robert Koch and the 'Golden age' of bacteriology" *Int J Infect Dis*
3. Croft, Ed (2006) "Under the microscope: A brief history of microscopy"
4. Lichtman, Conchello (2005) "Fluorescence microscopy" *Nat Methods*
5. Leavy (2016) "The birth of monoclonal antibodies" *Nat Immunol*
6. Tsien (1998) "The green fluorescent protein" *Annu Rev Biochem*
7. Cohen (2013) "DNA cloning: A personal view after 40 years" *Proc Natl Acad Sci*
8. Goldsmith, Miller (2009) "Modern uses of electron microscopy for detection of viruses" *Clin Microbiol Rev*
9. Kausche, Pfankuch, Ruska (1939) "Die Sichtbarmachung von pflanzlichem Virus im Übermikroskop" *Naturwissenschaften*
10. Bozzola, Russell (1999) "Dubochet J. The physics of rapid cooling and its implications for cryoimmobilization of cells"
11. Dubochet, Lepault, Mcdowall (1985) "Emerging techniques: Cryoelectron microscopy of vitrified biological specimens" *Trends Biochem Sci*
12. Amunts, Brown, Bai et al. (2014) "Structure of the yeast mitochondrial large ribosomal subunit" *Science*
13. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
14. Carragher, Kisseberth, Kriegman et al. (2000) "Leginon: An Automated System for Acquisition of Images from Vitreous Ice Specimens" *J Struct Biol*
15. Cheng, Tan, Dandey et al. "Strategies for Automated CryoEM Data Collection Using Direct Detectors" *Methods Enzymology*
16. (2016)
17. Bouvette, Huang, Riccio et al. (2022) "Automated systematic evaluation of cryo-EM specimens with SmartScope. eLife"
18. Baŕcena, Koster (2009) "Electron tomography in life science" *Semin Cell Dev Biol*
19. Ercius, Alaidi, Rames et al. (2015) "Electron Tomography: A Three-Dimensional Analytic Tool for Hard and Soft Materials Research" *Advanced Materials*
20. Wolf, Mutsafi, Dadosh et al. (2017) "3D visualization of mitochondrial solid-phase calcium stores in whole cells"
21. Pfeil-Gardiner, Rosa, Riedel et al. (2024) "Elemental mapping in single-particle reconstructions by reconstructed electron energyloss analysis" *Nat Methods*
22. Bykov, Briggs, Bartenschlager (2016) "Correlative light and electron microscopy methods for the study of virus-cell interactions" *FEBS Lett*
23. Yang, Vrbovskáv, Franke et al. (2023) "Precise 3D Localization by Integrated Fluorescence Microscopy (iFLM) for Cryo-FIB-milling and In-situ Cryo-ET" *Microscopy Microanalysis*
24. Boer, Hoogenboom, Giepmans (2015) "Correlated light and electron microscopy: ultrastructure lights up! Nat Methods"
25. Balasubramanian, Hobson, Chew et al. (2023) "Imagining the future of optical microscopy: Everything, everywhere, all at once" *Commun Biol*
26. Mccafferty, Klumpe, Amaro et al. (2024) "Integrating cellular electron microscopy with multimodal data to explore biology across space and time" *Cell*
27. Nowogrodzki (2024) "ChatGPT for science: How to talk to your data" *Nature*
28. Moebel, Martinez-Sanchez, Lamm et al. (2021) "Deep learning improves macromolecule identification in 3D cellular cryo-electron tomograms" *Nat Methods*
29. Wan, Tam, Cai et al. (2023) "Integrating spatial and single-cell transcriptomics data using deep generative models with SpatialScope" *Nat Commun*
30. Wang, Rivenson, Wei et al. (2019) "Deep learning enables cross-modality super-resolution in fluorescence microscopy" *Nat Methods*
31. Shi, Tabet, Milkie et al. (2024) "Smart lattice light-sheet microscopy for imaging rare and complex cellular events" *Nat Methods*
32. Gordon (2014) "Electron microscopy: A brief history and review of current clinical application" *Histopathology: Methods Protoc*
33. Graham, Orenstein (2007) "Processing tissue and cells for transmission electron microscopy in diagnostic pathology and research" *Nat Protoc*
34. Gan, Jensen (2012) "Electron tomography of cells" *Q Rev Biophysics*
35. Pengo, Scolari, Oliva et al. (2013) "Plasma cells require autophagy for sustainable immunoglobulin production" *Nat Immunol*
36. Jiao, Jiang, Hu et al. (2021) "Mitocytosis, a migrasomemediated mitochondrial quality-control process" *Cell*
37. Roche, Furuta (2015) "The ins and outs of MHC class II-mediated antigen processing and presentation" *Nat Rev Immunol*
38. Faigle, Raposo, Tenza et al. (1998) "Deficient peptide loading and MHC class II endosomal sorting in a human genetic immunodeficiency disease: the chediak-higashi syndrome" *J Cell Biol*
39. Swanson, Bushnell, Silverstein (1987) "Tubular lysosome morphology and distribution within macrophages depend on the integrity of cytoplasmic microtubules" *Proc Natl Acad Sci*
40. Barral, Staiano, Almeida et al. (2022) "Current methods to analyze lysosome morphology, positioning, motility and function" *Traffic*
41. Laue (2025) "Diagnostic electron microscopy in human infectious diseases -Methods and applications" *J Microscopy*
42. Mayhew, Lucocq, Griffiths (2002) "Relative labelling index: A novel stereological approach to test for non-random immunogold labelling of organelles and membranes on transmission electron microscopy thin sections" *J Microscopy*
43. Martell, Deerinck, Sancak et al. (2012) "Engineered ascorbate peroxidase as a genetically encoded reporter for electron microscopy" *Nat Biotechnol*
44. Martell, Deerinck, Lam et al. (2017) "Electron microscopy using the genetically encoded APEX2 tag in cultured mammalian cells" *Nat Protoc*
45. Stinchcombe, Asano, Kaufman et al. (2023) "Ectocytosis renders T cell receptor signaling self-limiting at the immune synapse" *Science*
46. Fischer, Hansen, Nair et al. (2012) "Scanning electron microscopy" *Curr Protoc Microbiol*
47. Miller, Wang, Grillet (2024) "SUB-immunogold-SEM reveals nanoscale distribution of submembranous epitopes" *Nat Commun*
48. Guidotti, Inverso, Sironi et al. (2015) "Immunosurveillance of the liver by intravascular effector CD8+ T cells" *Cell*
49. Baldwin, Heuser-Loy, Saha et al. (2024) "Intercellular nanotube-mediated mitochondrial transfer enhances T cell metabolic fitness and antitumor efficacy"
50. Collinson, Bosch, Bullen et al. (2023) "A quiet revolution takes shape" *Nat Methods*
51. Peddie, Genoud, Kreshuk et al. (2022) "Volume electron microscopy" *Nat Rev Methods Primers*
52. Wolf, Zoppi, Vogel et al. (2020) "Dynamics in protein translation sustaining T cell preparedness" *Nat Immunol*
53. Lachat, Pascault, Thibaut et al. (2022) "Transcellular tunnels induced by the fungal pathogen Candida albicans facilitate invasion through successive epithelial cells without host damage" *Nat Commun*
54. Winer, Settle, Yakimov et al. (2021) "Plasma membrane abundance dictates phagocytic capacity and functional cross-talk in myeloid cells" *Sci Immunol*
55. Xu, Pang, Shtengel et al. (2021) "An openaccess volume electron microscopy atlas of whole cells and tissues" *Nature*
56. Müller, Schmidt, Albrecht et al. (2024) "Modular segmentation, spatial analysis and visualization of volume electron microscopy datasets" *Nat Protoc*
57. Gopal, Chiappini, Armstrong et al. (2019) "Immunogold FIB-SEM: combining volumetric ultrastructure visualization with 3D biomolecular analysis to dissect cell-environment interactions" *Advanced Materials*
58. Loginov, Fermie, Fokkema et al. (2022) "Correlative organelle microscopy: fluorescence guided volume electron microscopy of intracellular processes" *Front Cell Dev Biol*
59. Nogales, Mahamid (2024) "Bridging structural and cell biology with cryo-electron microscopy" *Nature*
60. Nakane, Kotecha, Sente et al. (2020) "Single-particle cryo-EM at atomic resolution" *Nature*
61. Yip, Fischer, Paknia et al. (2020) "Atomic-resolution protein structure determination by cryo-EM" *Nature*
62. Nannenga, Gonen, Turonǒváb et al. (2019) "Benchmarking tomographic acquisition schemes for high-resolution structural biology" *Wolf SG, Elbaum M. CryoSTEM tomography in biology*
63. Dutta, Acharya (2024) "Cryo-electron microscopy in the study of virus entry and infection" *Front Mol Biosci*
64. Engelenburg, Shtengel, Sengupta et al. (2014) "Distribution of ESCRT machinery at HIV assembly sites reveals virus scaffolding of ESCRT subunits" *Science*
65. Mendonca, Howe, Gilchrist et al. (2021) "Correlative multi-scale cryo-imaging unveils SARS-CoV-2 assembly and egress" *Nat Commun*
66. Dong, Zheng, Lin et al. (2019) "Structural basis of assembly of the human T cell receptor-CD3 complex" *Nature*
67. Susǎc, Vuong, Thomas et al. (2022) "Structure of a fully assembled tumor-specific T cell receptor ligated by pMHC" *Cell*
68. Arechaga, Swamy, Abia et al. (2010) "Structural characterization of the TCR complex by electron microscopy" *Int Immunol*
69. Xin, Huang, Chi et al. (2024) "Structures of human gd T cell receptor-CD3 complex" *Nature*
70. Pishesha, Harmand, Ploegh (2022) "A guide to antigen processing and presentation" *Nat Rev Immunol*
71. Zhang, Chen, Tan et al. (2021) "Structural basis for chemokine recognition and receptor activation of chemokine receptor CCR5" *Nat Commun*
72. Marcandalli, Fiala, Ols et al. (2019) "Induction of potent neutralizing antibody responses by a designed protein nanoparticle vaccine for respiratory syncytial virus" *Cell*
73. Tokuyasu (1986) "Application of cryoultramicrotomy to immunocytochemistry" *Bos E, Griekspoor A. Cryo-Immunogold Electron Microscopy. Curr Protoc Cell Biol*
74. Yi, Strauss, Ke et al. (2015) "Native Immunogold Labeling of Cell Surface Proteins and Viral Glycoproteins for Cryo-Electron Microscopy and Cryo-Electron Tomography Applications" *J Histochem Cytochem*
75. Palacios, Villares, Lucas et al. (2023) "Growth hormone remodels the 3D-structure of the mitochondria of inflammatory macrophages and promotes metabolic reprogramming" *Front Immunol*
76. Earl, Lifson, Subramaniam (2013) "Catching HIV 'in the act' with 3D electron microscopy" *Trends Microbiol*
77. Mahamid, Pfeffer, Schaffer et al. (2016) "Visualizing the molecular sociology at the HeLa cell nuclear periphery" *Science*
78. Studer, Humbel, Chiquet (2008) "Electron microscopy of high pressure frozen samples: bridging the gap between cellular ultrastructure and atomic resolution" *Histochem Cell Biol*
79. Thompson, Walker, Siebert et al. (2016) "An introduction to sample preparation and imaging by cryo-electron microscopy for structural biology" *Methods*
80. Biazik, Vihinen, Anwar et al. (2015) "The versatile electron microscope: An ultrastructural overview of autophagy" *Methods*
81. Laporte, Klena, Hamel et al. (2022) "Visualizing the native cellular organization by coupling cryofixation with expansion microscopy (Cryo-ExM)" *Nat Methods*
82. Mine-Hattab (2023) "When fixation creates fiction. eLife"
83. Irgen-Gioro, Yoshida, Walling et al. (2022) "Fixation can change the appearance of phase separation in living cells. eLife"
84. Phillips, Harkiolaki, Pinto et al. (2020) "CryoSIM: super-resolution 3D structured illumination cryogenic fluorescence microscopy for correlated ultrastructural imaging" *Optica*
85. Tuijtel, Koster, Jakobs et al. (2019) "Correlative cryo superresolution light and electron microscopy on mammalian cells using fluorescent proteins" *Sci Rep*
86. Antuono (2022) "Basic Digital Image Acquisition, Design, Processing, Analysis, Management, and Presentation"
87. Pilato, Palomino-Segura, Mejıás-Peŕez et al. (2014) "ATP released by intestinal bacteria limits the generation of protective IgA against enteropathogens" *NPJ Vaccines*
88. Wang, Eng, Law et al. (2017) "Visualization of HIV T cell virological synapses and virus-containing compartments by three-dimensional correlative light and electron microscopy" *J Virol*
89. Hor, Germain (2022) "Intravital and high-content multiplex imaging of the immune system" *Trends Cell Biol*
90. Turk, Baumeister (2020) "The promise and the challenges of cryo-electron tomography" *FEBS Lett*
91. Andrian, Bakkum, Van Elsland et al. (2021) "Chapter 13-Super-resolution correlative light-electron microscopy using a clickchemistry approach for studying intracellular trafficking"
92. Marko, Hsieh, Schalek et al. (2007) "Focused-ion-beam thinning of frozen-hydrated biological specimens for cryo-electron microscopy" *Nat Methods*
93. Al-Amoudi, Norlen, Dubochet (2004) "Cryo-electron microscopy of vitreous sections of native biological cells and tissues" *J Struct Biol*
94. Pierson, Yang, Wright (2024) "Recent advances in correlative cryo-light and electron microscopy" *Curr Opin Struct Biol*
95. Klein, Wachsmuth-Melm, Winter et al. (2021) "Cryocorrelative light and electron microscopy workflow for cryo-focused ion beam milled adherent cells" *Methods in Cell Biology*
96. Beer, Daviran, Roverts et al. (2019) "Precise targeting for 3D cryo-correlative light and electron microscopy volume imaging of tissues using a FinderTOP" *Commun Biol*
97. Rey, Li, Bryer et al. (2021) "Deep-learning in situ classification of HIV-1 virion morphology" *Comput Struct Biotechnol J*
98. Matharu, Nordmann, Ottman et al. (2023) "Deep learning, 3D ultrastructural analysis reveals quantitative differences in platelet and organelle packing in COVID-19/SARSCoV2 patient-derived platelets" *Platelets*
99. Martinez-Sanchez (2025) "Template matching and machine learning for cryo-electron tomography" *Curr Opin Struct Biol*
100. Seifert, Markert, Britz et al. "DeepCLEM: automated registration for correlative light and electron microscopy u"
101. Belevich, Deepmib (2021) "User-friendly and open-source software for training of deep learning network for biological image segmentation" *PloS Comput Biol*
102. Ouyang, Aristov, Lelek et al. (2018) "Deep learning massively accelerates super-resolution localization microscopy" *Nat Biotechnol*
103. Nehme, Freedman, Gordon et al. (2020) "DeepSTORM3D: Dense 3D localization microscopy and PSF design by deep learning" *Nat Methods*
104. Long, Moriarty, Cardoen et al. (2020) "Super resolution microscopy and deep learning identify Zika virus reorganization of the endoplasmic reticulum" *Sci Rep*
105. Wanninger, Asadiatouei, Bohlen et al. (2023) "Deep-LASI: deep-learning assisted, single-molecule imaging analysis of multi-color DNA origami structures" *Nat Commun*
106. Park, Na, Kim et al. (2022) "Deep learning enables reference-free isotropic super-resolution for volumetric fluorescence microscopy" *Nat Commun*
107. Guo, Wu, Hobson et al. (2025) "Deep learning-based aberration compensation improves contrast and resolution in fluorescence microscopy" *Nat Commun*
108. Heĺiot, Leray (2021) "Simple phasor-based deep neural network for fluorescence lifetime imaging microscopy" *Sci Rep*
109. Qiao, Liu, Wang (2025) "A neural network for long-term super-resolution imaging of live cells with reliable confidence quantification" *Nat Biotechnol*
110. Bai, Yu, Peng et al. (2020) "3D Imaging Restoration of Spinning-Disk Confocal Microscopy Via Deep Learning" *IEEE Photon Technol Lett*
111. Schapiro, Sokolov, Yapp et al. (2022) "MCMICRO: A scalable, modular image-processing pipeline for multiplexed tissue imaging" *Nat Methods*
112. Greenwald, Miller, Moen et al. (2022) "Wholecell segmentation of tissue images with human-level performance using large-scale data annotation and deep learning" *Nat Biotechnol*
113. Sultan, Gorris, Martynova et al.
114. Wilpe (2025) "ImmuNet: A segmentation-free machine learning pipeline for immune landscape phenotyping in tumors by multiplex imaging" *Biol Methods Protoc*
115. Zheng, Sadeé, Ozawa et al. (2025) "Single-cell multimodal analysis reveals tumor microenvironment predictive of treatment response in nonsmall cell lung cancer" *Sci Adv*
116. Fassler, Abousamra, Gupta et al. (2020) "Deep learning-based image analysis methods for brightfield-acquired multiplex immunohistochemistry images" *Diagn Pathol*
117. Wu, Rivenson, Wang et al. (2019) "Threedimensional virtual refocusing of fluorescence microscopy images using deep learning" *Nat Methods*
118. Li, Tan, Dong et al. (2021) "Deep-3D microscope: 3D volumetric microscopy of thick scattering samples using a wide-field microscope and machine learning" *Biomed Optics Express*
119. Christiansen, Yang, Ando et al. (2018) "In silico labeling: predicting fluorescent labels in unlabeled images" *Cell*
120. Ounkomol, Seshamani, Maleckar et al. (2018) "Label-free prediction of three-dimensional fluorescence images from transmitted-light microscopy" *Nat Methods*
121. Cheng, Fu, Kim et al. (2021) "Single-cell cytometry via multiplexed fluorescence prediction by label-free reflectance microscopy" *Sci Adv*
122. Li, Zhang, Qiao et al. (2021) "Unsupervised contentpreserving transformation for optical microscopy" *Light: Sci Appl*
123. Buggenthin, Buettner, Hoppe et al. (2017) "Prospective identification of hematopoietic lineage choice by deep learning" *Nat Methods*
124. Antonello, Morone, Pirani et al. (2023) "Tracking unlabeled cancer cells imaged with low resolution in wide migration chambers via U-NET class-1 probability (pseudofluorescence)" *J Biol Eng*
125. Liarski, Sibley, Van Panhuys et al. (2019) "Quantifying in situ adaptive immune cell cognate interactions in humans" *Nat Immunol*
126. Zhang, Yu, Lin et al. (2023) "Deep tissue superresolution imaging with adaptive optical two-photon multifocal structured illumination microscopy" *PhotoniX*
127. Wijesinghe, Corsetti, Chow et al. (2022) "Experimentally unsupervised deconvolution for light-sheet microscopy with propagation-invariant beams" *Light: Sci Appl*
128. Yang, Liu, Kumar et al. (2024) "LungVis 1.0: an automatic AI-powered 3D imaging ecosystem unveils spatial profiling of nanoparticle delivery and acinar migration of lung macrophages" *Nat Commun*
129. Inoués (2006) "Foundations of confocal scanned imaging in light microscopy"
130. Schermelleh, Ferrand, Huser et al. (2019) "Super-resolution microscopy demystified" *Nat Cell Biol*
131. Sahl, Hell, Jakobs (2017) "Fluorescence nanoscopy in cell biology" *Nat Rev Mol Cell Biol*
132. Roth, Sheppard, Wicker et al. (2013) "Optical photon reassignment microscopy (OPRA)"
133. Sheppard, Mehta, Heintzmann (2013) "Superresolution by image scanning microscopy using pixel reassignment" *Optics Lett*
134. Lelek, Gyparaki, Beliu et al. (2021) "Singlemolecule localization microscopy" *Nat Rev Methods Primers*
135. Fox-Roberts, Marsh, Pfisterer et al. (2017) "Local dimensionality determines imaging speed in localization microscopy" *Nat Commun*
136. Culley, Albrecht, Jacobs et al. (2018) "Quantitative mapping and minimization of super-resolution optical imaging artifacts" *Nat Methods*
137. Stebbing, Nievas, Falcone et al. (2021) "JAK inhibition reduces SARS-CoV-2 liver infectivity and modulates inflammatory responses to reduce morbidity and mortality" *Sci Adv*
138. Rossboth, Arnold, Ta et al. (2018) "TCRs are randomly distributed on the plasma membrane of resting antigen-experienced T cells" *Nat Immunol*
139. Blaha, Skaĺovát, Kalouskováb et al. (2022) "Structure of the human NK cell NKR-P1:LLT1 receptor:ligand complex reveals clustering in the immune synapse" *Nat Commun*
140. Hell (2007) "Far-field optical nanoscopy" *Science*
141. Baranov, Revelo, Dingjan et al. (2016) "SWAP70 organizes the actin cytoskeleton and is essential for phagocytosis" *Cell Rep*
142. Worboys, Vowell, Hare et al. (2015) "The Airyscan detector from ZEISS: Confocal imaging with improved signal-to-noise ratio and super-resolution" *Nat Methods*
143. Delattre (2023) "Igniting new confocal imaging potentialnikon AX R series with NSPARC" *Microscopy Today*
144. Curd, Cleasby, Makowska et al. (2015) "Construction of an instant structured illumination microscope" *Methods*
145. Dertinger, Colyer, Iyer et al. (2009) "Fast, background-free, 3D super-resolution optical fluctuation imaging (SOFI" *Proc Natl Acad Sci*
146. Gustafsson, Culley, Ashdown et al. (2016) "Fast live-cell conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations" *Nat Commun*
147. Torres-Garcıá, Pinto-Caḿara, Linares et al. (2022) "Extending resolution within a single imaging frame" *Nat Commun*
148. Alva, Brito-Alarcoń, Linares et al.
149. Caḿara (2022) "Fluorescence fluctuation-based super-resolution microscopy: Basic concepts for an easy start" *J Microscopy*
150. Reinhardt, Masullo, Baudrexel et al. (2023) "Ångström-resolution fluorescence microscopy" *Nature*
151. Sahl, Matthias, Inamdar et al. (2024) "Direct optical measurement of intramolecular distances with angstrom precision" *Science*
152. Balzarotti, Eilers, Gwosch et al. (2017) "Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes" *Science*
153. Wolff, Scheiderer, Engelhardt et al. (2023) "MINFLUX dissects the unimpeded walking of kinesin-1" *Science*
154. Scheiderer, Der Emde, Hesselink et al. (2024) "MINSTED tracking of single biomolecules" *Nat Methods*
155. Axelrod (1981) "Cell-substrate contacts illuminated by total internal reflection fluorescence" *Curr Protoc Cytometry*
156. Brameshuber, Klotzsch, Ponjavic et al. (2022) "Understanding immune signaling using advanced imaging techniques" *Biochem Soc Trans*
157. Grakoui, Bromley, Sumen et al. (1999) "The immunological synapse: A molecular machine controlling T cell activation" *Science*
158. Dustin, Baldari "The immune synapse: past, present, and future"
159. (2017) *Methods Mol Biol*
160. Kvalvaag, Valvo, Ceśpedes et al. (2025) "Clathrin mediates both internalization and vesicular release of triggered T cell receptor at the immunological synapse" *Proc Natl Acad Sci*
161. Martıńez-Muñoz, Rodrıǵuez-Frade, Barroso et al. (2018) "Separating actin-dependent chemokine receptor nanoclustering from dimerization indicates a role for clustering in CXCR4 signaling and function" *Mol Cell*
162. Mazzolini, Herit, Niedergang (2016) "Dynamin-actin cross talk contributes to phagosome formation and closure" *Traffic*
163. Sundd, Gutierrez, Koltsova et al. (2012) "Slings' enable neutrophil rolling at high shear" *Nature*
164. Sundd, Gutierrez, Pospieszalska et al. (2010) "Quantitative dynamic footprinting microscopy reveals mechanisms of neutrophil rolling" *Nat Methods*
165. Yi, Balagopalan, Nguyen et al. (2019) "TCR microclusters form spatially segregated domains and sequentially assemble in calcium-dependent kinetic steps" *Nat Commun*
166. Minsky (1988) "Memoir on inventing the confocal scanning microscope" *Scanning*
167. Jonkman, Brown, Wright et al. (2020) "Tutorial: Guidance for quantitative confocal microscopy" *Nat Protoc*
168. Elliott (2020) "Confocal microscopy: principles and modern practices" *Curr Protoc Cytometry*
169. Sheppard (2020) "Scanning optical microscopy"
170. Valignat, Nègre, Cadra et al. (2014) "Lymphocytes can self-steer passively with wind vane uropods" *Nat Commun*
171. Vorselen, Wang, De Jesus et al. (2020) "Microparticle traction force microscopy reveals subcellular force exertion patterns in immune cell-target interactions" *Nat Commun*
172. Muldoon, Chuang, Bagheri et al. (2020) "Macrophages employ quorum licensing to regulate collective activation" *Nat Commun*
173. Amann, Fell, Monaco et al. (2024) "Extrasinusoidal macrophages are a distinct subset of immunologically active dural macrophages" *Sci Immunol*
174. Brulois, Rajaraman, Szade et al. (2020) "A molecular map of murine lymph node blood vascular endothelium at single cell resolution" *Nat Commun*
175. Reynoso, Weisberg, Shannon et al. (2019) "Lymph node conduits transport virions for rapid T cell activation" *Nat Immunol*
176. Radice, Ameti, Melgrati et al. (2020) "Marginal zone formation requires ACKR3 expression on B cells" *Cell Rep*
177. Mao, Baptista, Tamoutounour et al. (2018) "Innate and adaptive lymphocytes sequentially shape the gut microbiota and lipid metabolism" *Nature*
178. Wong, Park, Gola et al. (2021) "A local regulatory T cell feedback circuit maintains immune homeostasis by pruning selfactivated T cells" *Cell*
179. Gaertner, Ishikawa-Ankerhold, Stutte et al. (2024) "Plasmacytoid dendritic cells control homeostasis of megakaryopoiesis" *Nature*
180. Kaltenbach, Martzloff, Bambach et al. (2023) "Slow integrin-dependent migration organizes networks of tissue-resident mast cells" *Nat Immunol*
181. Garcia-Leon, Liboni, Mittelheisser et al. (2022) "Corneal tissue-resident memory T cells form a unique immune compartment at the ocular surface" *Nat Commun*
182. Dekkers, Alieva, Cleven et al. (2023) "Uncovering the mode of action of engineered T cells in patient cancer organoids" *Nat Biotechnol*
183. Datta, Heaster, Sharick et al. (2020) "Fluorescence lifetime imaging microscopy: Fundamentals and advances in instrumentation, analysis, and applications"
184. Lakowicz, Ed (2006) "Principles of fluorescence spectroscopy"
185. Colom, Derivery, Soleimanpour et al. (2018) "A fluorescent membrane tension probe" *Nat Chem*
186. Gambarotto, Zwettler, Guennec et al. (2019) "Imaging cellular ultrastructures using expansion microscopy (U-ExM)" *Nat Methods*
187. Bacia, Kim, Schwille (2006) "Fluorescence cross-correlation spectroscopy in living cells" *Nat Methods*
188. Wiseman "Image correlation spectroscopy: Principles and applications"
189. (2015)
190. Digman, Dalal, Horwitz et al. (2008) "Mapping the number of molecules and brightness in the laser scanning microscope" *Biophys J*
191. Isbilir, Serfling, Möller et al. (2021) "Determination of G-protein-coupled receptor oligomerization by molecular brightness analyses in single cells" *Nat Protoc*
192. Wu, Han, Su et al. (2021) "Multiview confocal super-resolution microscopy" *Nature*
193. Lu, Liu, Luo et al. (2023) "Virtual-scanning light-field microscopy for robust snapshot high-resolution volumetric imaging" *Nat Methods*
194. Lu, Zuo, Shi et al. (2024) "Long-term intravital subcellular imaging with confocal scanning light-field microscopy" *Nat Biotechnol*
195. Bajeńoff, Germain (2007) "Seeing is believing: A focus on the contribution of microscopic imaging to our understanding of immune system function" *Eur J Immunol*
196. Germain, Robey, Cahalan (2012) "A decade of imaging cellular motility and interaction dynamics in the immune system"
197. Germain (2022) "Imaging the immune system redux" *Immunol Rev*
198. Pawley, Ed (2006) "Handbook of biological confocal microscopy"
199. David, Atif, Vargas et al. (2024) "Kupffer cell reverse migration into the liver sinusoids mitigates neonatal sepsis and meningitis" *Sci Immunol*
200. Umeshappa, Solép, Surewaard et al. (2021) "Liver-specific T regulatory type-1 cells program local neutrophils to suppress hepatic autoimmunity via CRAMP" *Cell Rep*
201. Hossain, Shim, Lee et al. (2022) "Gata6+ resident peritoneal macrophages promote the growth of liver metastasis" *Nat Commun*
202. Neupane, Willson, Chojnacki et al. (2020) "Patrolling alveolar macrophages conceal bacteria from the immune system to maintain homeostasis" *Cell*
203. Hančp, Gonzalez, Mazo et al. (2023) "Multimodal control of dendritic cell functions by nociceptors" *Science*
204. Hansen, Gong, Wachten et al. (2021) "Multifocal imaging for precise, label-free tracking of fast biological processes in 3D" *Nat Commun*
205. Zipfel, Williams, Webb (2003) "Nonlinear magic: Multiphoton microscopy in the biosciences" *Nat Biotechnol*
206. Pittet, Garris, Arlauckas et al. (2018) "Recording the wild lives of immune cells" *Sci Immunol*
207. Palomino-Segura, Virgilio, Morone et al. (2018) "Imaging cell interaction in tracheal mucosa during influenza virus infection using two-photon intravital microscopy" *J Vis Exp*
208. Palomino-Segura, Perez, Farsakoglu et al. (2019) "Protection against influenza infection requires early recognition by inflammatory dendritic cells through C-type lectin receptor SIGN-R1" *Nat Microbiol*
209. Chatziandreou, Farsakoglu, Palomino-Segura et al. (2017) "Macrophage death following influenza vaccination initiates the inflammatory response that promotes dendritic cell function in the draining lymph node" *Cell Rep*
210. Lim, Kim, Trzeciak et al. (2020) "In situ neutrophil efferocytosis shapes T cell immunity to influenza infection" *Nat Immunol*
211. Doni, Musso, Morone et al. (2015) "An acidic microenvironment sets the humoral pattern recognition molecule PTX3 in a tissue repair mode" *J Exp Med*
212. Kratofil, Shim, Shim et al. (2022) "A monocyteleptin-angiogenesis pathway critical for repair post-infection" *Nature*
213. Zindel, Peiseler, Hossain et al. (2021) "Primordial GATA6 macrophages function as extravascular platelets in sterile injury" *Science*
214. Wang, Hossain, Bogoslowski et al. (2020) "Chemotaxing neutrophils enter alternate branches at capillary bifurcations" *Nat Commun*
215. Prat-Luri, Neal, Passelli et al. (2022) "The C5a-C5aR1 complement axis is essential for neutrophil recruitment to draining lymph nodes via high endothelial venules in cutaneous leishmaniasis" *Cell Rep*
216. Mempel, Henrickson, Von Andrian (2004) "T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases" *Nature*
217. Fumagalli, Venzin, Lucia et al. (2022) "Group 1 ILCs regulate T cell-mediated liver immunopathology by controlling local IL-2 availability" *Sci Immunol*
218. Weigelin, Boer, Wagena et al. (2021) "Cytotoxic T cells are able to efficiently eliminate cancer cells by additive cytotoxicity" *Nat Commun*
219. Liu, Winer, Chou et al. (2024) "Dynamic encounters with red blood cells trigger splenic marginal zone B cell retention and function" *Nat Immunol*
220. Pae, Schwan, Ottino-Loffler et al. (2025) "Transient silencing of hypermutation preserves B cell affinity during clonal bursting" *Nature*
221. Zindel, Mittner, Bayer et al. (2021) "Intraperitoneal microbial contamination drives post-surgical peritoneal adhesions by mesothelial EGFR-signaling" *Nat Immunol*
222. Schubert, Soyka, Tamimi et al. (2024) "Deep intravital brain tumor imaging enabled by tailored three-photon microscopy and analysis" *Nat Commun*
223. Streich, Boffi, Wang et al. (2021) "Highresolution structural and functional deep brain imaging using adaptive optics threephoton microscopy" *Nat Methods*
224. Bakker, Weischer, Ortas et al. (2022) "Intravital deep-tumor single-beam 3-photon, 4-photon, and harmonic microscopy. eLife"
225. Morone, 'autilia, Schorn et al. (2020) "Evaluation of cell metabolic adaptation in wound and tumour by Fluorescence Lifetime Imaging Microscopy" *Sci Rep*
226. Zhao, Zhou, Liu et al. (2023) "Two-photon synthetic aperture microscopy for minimally invasive fast 3D imaging of native subcellular behaviors in deep tissue" *Cell*
227. Sims, Bendifallah, Grimm et al. (2024) "Scanless two-photon voltage imaging" *Nat Commun*
228. Pinkard, Baghdassarian, Mujal et al. (2021) "Learned adaptive multiphoton illumination microscopy for large-scale immune response imaging" *Nat Commun*
229. Gerner, Kastenmuller, Ifrim et al. (2012) "Histo-cytometry: A method for highly multiplex quantitative tissue imaging analysis applied to dendritic cell subset microanatomy in lymph nodes" *Immunity*
230. Semba, Ishimoto (2024) "Spatial analysis by current multiplexed imaging technologies for the molecular characterisation of cancer tissues" *Br J Cancer*
231. Zimmermann, Rietdorf, Pepperkok (2003) "Spectral imaging and its applications in live cell microscopy" *FEBS Lett*
232. Sadashivaiah, Tippani, Page et al. (2023) "SUFI: An automated approach to spectral unmixing of fluorescent multiplex images captured in mouse and post-mortem human brain tissues" *BMC Neurosci*
233. Klevanski, Herrmannsdoerfer, Sass et al. (2020) "Automated highly multiplexed super-resolution imaging of protein nanoarchitecture in cells and tissues" *Nat Commun*
234. Cole, Del Castillo, Gut et al. (2018) "Characterization of the neurogenic niche in the aging dentate gyrus using iterative immunofluorescence imaging. eLife" *Nat Commun*
235. Ko, Wilkovitsch, Oh et al. (2022) "Spatiotemporal multiplexed immunofluorescence imaging of living cells and tissues with bioorthogonal cycling of fluorescent probes" *Nat Biotechnol*
236. Quardokus, Saunders, Mcdonough et al. (2023) "Organ Mapping Antibody Panels: A community resource for standardized multiplexed tissue imaging" *Nat Methods*
237. Schueder, Unterauer, Ganji et al. (2020) "DNA-barcoded fluorescence microscopy for spatial omics" *PROTEOMICS*
238. Hwang, Raymond, Mcpartland et al. (2024) "Fluorescence Lifetime Multiplexing (FLEX) for simultaneous high dimensional spatial biology in 3D" *Commun Biol*
239. Gouzou, Taimori, Haloubi et al. (2024) "Applications of machine learning in time-domain fluorescence lifetime imaging: A review" *Methods Appl Fluorescence*
240. Alexandrov (2020) "Spatial metabolomics and imaging mass spectrometry in the age of artificial intelligence" *Annu Rev Biomed Data Sci*
241. Angelo, Bendall, Finck et al. (2014) "Multiplexed ion beam imaging of human breast tumors" *Nat Med*
242. Giesen, Hao, Schapiro et al. (2014) "Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry" *Nat Methods*
243. Du, Su, Qian et al. (2020) "Raman-guided subcellular pharmaco-metabolomics for metastatic melanoma cells" *Nat Commun*
244. Phillips, Matusiak, Gutierrez et al. (2021) "Immune cell topography predicts response to PD-1 blockade in cutaneous T cell lymphoma" *Nat Commun*
245. Roider, Baertsch, Fitzgerald et al. (2024) "Multimodal and spatially resolved profiling identifies distinct patterns of T cell infiltration in nodal B cell lymphoma entities" *Nat Cell Biol*
246. Zhang, Deng, Kukanja et al. (2023) "Spatial epigenome-transcriptome co-profiling of mammalian tissues" *Nature*
247. Varga-Zsıŕos, Migh, Marton et al. (2023) "Development of a laser microdissection-coupled quantitative shotgun lipidomic method to uncover spatial heterogeneity" *Cells*
248. Brasko, Smith, Molnar et al. (2018) "Intelligent image-based in situ single-cell isolation" *Nat Commun*
249. Parra, Uraoka, Jiang et al. (2017) "Validation of multiplex immunofluorescence panels using multispectral microscopy for immuneprofiling of formalin-fixed and paraffin-embedded human tumor tissues" *Sci Rep*
250. Seo, Sim, Kim et al. (2022) "PICASSO allows ultra-multiplexed fluorescence imaging of spatially overlapping proteins without reference spectra measurements" *Nat Rev Genet*
251. Yayon, Kedlian, Boehme et al. (2024) "A spatial human thymus cell atlas mapped to a continuous tissue axis" *Nature*
252. Stelzer, Strobl, Chang et al. (2021) "Light sheet fluorescence microscopy" *Nat Rev Methods Primers*
253. Daetwyler, Fiolka (2023) "Light-sheets and smart microscopy, an exciting future is dawning" *Commun Biol*
254. Rocha, Basto, Ijsselsteijn et al. (2021) "Immunophenotype of gastric tumors unveils a pleiotropic role of regulatory T cells in tumor development" *Cancers*
255. Richardson, Lichtman (2015) "Clarifying tissue clearing" *Cell*
256. Dekkers, Alieva, Wellens et al. (2019) "High-resolution 3D imaging of fixed and cleared organoids" *Nat Protoc*
257. Zhang, Cleveland, Krimitza et al. (2024) "Spatial analysis of tissue immunity and vascularity by light sheet fluorescence microscopy" *Nat Protoc*
258. Merz, Jansen, Ulankiewicz et al. (2021) "High-resolution three-dimensional imaging for precise staging in melanoma" *Eur J Cancer*
259. Mcdole, Guignard, Amat et al. (2018) "In toto imaging and reconstruction of post-implantation mouse development at the single-cell level" *Cell*
260. Cai, Kolabas, Pan et al. (2023) "Whole-mouse clearing and imaging at the cellular level with vDISCO" *Nat Protoc*
261. Mai, Luo, Hoeher et al. (2024) "Whole-body cellular mapping in mouse using standard IgG antibodies" *Nat Biotechnol*
262. Vladimirov, Voigt, Naert et al. (2024) "Benchtop mesoSPIM: A next-generation open-source light-sheet microscope for cleared samples" *Nat Commun*
263. Chen, Legant, Wang et al. (2014) "Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution" *Science*
264. Kersten, Hu, Combes et al. (2022) "Spatiotemporal co-dependency between macrophages and exhausted CD8+ T cells in cancer" *Cancer Cell*
265. Huisken, Swoger, Bene et al. (2004) "Optical sectioning deep inside live embryos by selective plane illumination microscopy" *Science*
266. Mahou, Vermot, Beaurepaire et al. (2014) "Multicolor two-photon lightsheet microscopy" *Nat Methods*
267. Wang, Zhong, Bruns et al. (2024) "In vivo NIR-II fluorescence imaging for biology and medicine" *Nat Photonics*
268. Bouchard, Voleti, Mendes et al. (2015) "Swept confocally-aligned planar excitation (SCAPE) microscopy for high-speed volumetric imaging of behaving organisms" *Nat Photon*
269. Patel, Liang, Casper et al. (2022) "High-speed light-sheet microscopy for the in-situ acquisition of volumetric histological images of living tissue" *Nat BioMed Eng*
270. Mimori-Kiyosue (2025) "Journey from image acquisition to biological insight: handling and analyzing large volumes of light-sheet imaging data" *Microscopy*
271. Jain, Ulman, Krumnikl et al. (2024) "Image processing and analysis of light sheet microscopy data"
272. Medeiros, Ortiz, Strnad et al. (2022) "Multiscale light-sheet organoid imaging framework" *Nat Commun*
273. Marin, Wang, Collison et al. (2024) "Navigate: An open-source platform for smart light-sheet microscopy" *Nat Methods*
274. Cimini, Bankhead, 'antuono et al. (2024) "The crucial role of bioimage analysts in scientific research and publication" *J Cell Sci*
275. Boehm, Nelson, Brown et al. (2021) "QUAREP-LiMi: A community endeavor to advance quality assessment and reproducibility in light microscopy" *Nat Methods*
276. Meiniel, Olivo-Marin, Angelini (2018) "Denoising of microscopy images: A review of the state-of-the-art, and a new sparsity-based method" *IEEE Trans Image Process*
277. Dabov, Foi, Katkovnik et al. (2006) "Image denoising with blockmatching and 3D filtering. Image Processing: Algorithms Systems Neural Networks Mach Learn"
278. Buades, Coll, Morel (2005) "A non-local algorithm for image denoising"
279. Boulanger, Kervrann, Bouthemy et al. (2010) "Patchbased nonlocal functional for denoising fluorescence microscopy image sequences" *IEEE Trans Med Imaging*
280. Luisier, Vonesch, Blu et al. (2010) "Fast interscale wavelet denoising of Poissoncorrupted images. Signal Process" *Methods*
281. Dehez, Pichém, Koninck (2013) "Resolution and contrast enhancement in laser scanning microscopy using dark beam imaging" *Optics Express*
282. You, Kuang, Rong et al. (2014) "Eliminating deformations in fluorescence emission difference microscopy" *Optics Express*
283. Toscani, Martıńez, Martıńez (2023) "Resolution, accuracy and precision in super-resolved microscopy images using SUPPOSe" *Optics Lasers Eng*
284. Saurabh, Brown, Bryan et al. (2025) "Approaching maximum resolution in structured illumination microscopy via accurate noise modeling" *NPJ Imaging*
285. Zhao, Zhao, Li et al. (2022) "Sparse deconvolution improves the resolution of live-cell super-resolution fluorescence microscopy" *Nat Biotechnol*
286. Preibisch, Saalfeld, Tomancak (2009) "Globally optimal stitching of tiled 3D microscopic image acquisitions" *Bioinformatics*
287. Chalfoun, Majurski, Blattner et al. (2017) "MIST: accurate and scalable microscopy image stitching tool with stage modeling and error minimization" *Sci Rep*
288. Muhlich, Chen, Yapp et al. (2022) "Stitching and registering highly multiplexed whole-slide images of tissues and tumors using ASHLAR" *Bioinformatics*
289. Goḿez-Conde, Caetano, Tadokoro et al. (2015) "Stabilizing 3D in vivo intravital microscopy images with an iteratively refined soft-tissue model for immunology experiments" *Comput Biol Med*
290. Goobic, Tang, Acton (2005) "Image stabilization and registration for tracking cells in the microvasculature" *IEEE Trans Biomed Eng*
291. Dunn, Lorenz, Salama et al. (2014) "IMART software for correction of motion artifacts in images collected in intravital microscopy" *IntraVital*
292. Thevenaz, Ruttimann, Unser (1998) "A pyramid approach to subpixel registration based on intensity" *IEEE Trans Image Process*
293. Malacrida (2023) "Phasor plots and the future of spectral and lifetime imaging" *Nat Methods*
294. Szeliski (2022) "Computer vision: algorithms and applications"
295. Otsu (1979) "A threshold selection method from gray-level histograms" *IEEE Trans Systems Man Cybernetics*
296. Mittal, Pandey, Saraswat et al. (2022) "A comprehensive survey of image segmentation: Clustering methods, performance parameters, and benchmark datasets" *Multimedia Tools Appl*
297. Kornilov, Safonov, Yakimchuk (2022) "A review of watershed implementations for segmentation of volumetric images" *J Imaging*
298. Rong, Li, Zhang et al. (2014) "An improved Canny edge detection algorithm"
299. Canny (1986) "A computational approach to edge detection"
300. Sobel (1990) "Machine vision for three-dimensional scenes"
301. Manzo, Mf (2015) "A review of progress in single particle tracking: From methods to biophysical insights" *Rep Prog Phys*
302. Midtvedt, Helgadottir, Argun et al. (2021) "Quantitative digital microscopy with deep learning" *Appl Phys Rev*
303. Jaqaman, Loerke, Mettlen et al. (2008) "Robust single-particle tracking in live-cell time-lapse sequences" *Nat Methods*
304. Tinevez, Perry, Schindelin et al. (2017) "TrackMate: An open and extensible platform for single-particle tracking" *Methods*
305. Pizzagalli, Farsakoglu, Palomino-Segura et al. (2018) "Leukocyte Tracking Database, a collection of immune cell tracks from intravital 2-photon microscopy videos. Sci Data"
306. Sugawara, Çevrim, Averof (2022) "Tracking cell lineages in 3D by incremental deep learning. eLife"
307. Miura, Tosi, Möhl et al. (2016) "Bioimage analysis tools"
308. Berg, Kutra, Kroeger et al. (2019) "ilastik: Interactive machine learning for (bio)image analysis" *Nat Methods*
309. Stirling, Swain-Bowden, Lucas et al. (2021) "CellProfiler 4: Improvements in speed, utility and usability" *BMC Bioinf*
310. D'antuono, Pisignano (2022) "ZELDA: A 3D image segmentation and parentchild relation plugin for microscopy image analysis in napari" *Front Comput Sci*
311. Cross, Fisher, Jepson (2024) "ModularImageAnalysis (MIA): assembly of modularised image and object analysis workflows in imageJ" *J Microscopy*
312. Gerst, Cseresnyeś, Figge (2023) "JIPipe: Visual batch processing for ImageJ" *Nat Methods*
313. Lefebvre, Sturm, Lin et al. (2025) "Automated organelle segmentation, tracking and hierarchical feature extraction in 2D/3D live-cell microscopy" *Nat Methods*
314. Martins, Cordelières, Colombelli et al. (2021) "Highlights from the 2016-2020 NEUBIAS training schools for Bioimage Analysts: A success story and key asset for analysts and life scientists. F1000Research"
315. Levet, Carpenter, Eliceiri et al. (2021) "Developing open-source software for bioimage analysis: opportunities and challenges. F1000Research"
316. Schindelin, Arganda-Carreras, Frise et al. (2012) "Fiji: an open-source platform for biological-image analysis" *Nat Methods*
317. Sofroniew, Lambert, Evans et al. (2022) "Napari: A multi-dimensional image viewer for python"
318. Bankhead, Loughrey, Fernańdez et al. (2017) "QuPath: Open source software for digital pathology image analysis" *Sci Rep*
319. Kluyver, Ragan-Kelley, Peŕez et al. (2016) "Jupyter Notebooksa publishing format for reproducible computational workflows"
320. Berthold, Cebron, Dill et al. (2009) "KNIMEthe konstanz information miner" *ACM SIGKDD Explor Newsletter*
321. Chamier, Laine, Jukkala et al. (2021) "Democratising deep learning for microscopy with ZeroCostDL4Mic" *Nat Commun*
322. Lleti, Steyer, Schieber et al. (2023) "CLEMSite, a software for automated phenotypic screens using light microscopy and FIB-SEM" *J Cell Biol*
323. Edelstein, Amodaj, Hoover et al. (2010) "Computer control of microscopes using mManager" *Curr Protoc Mol Biol*
324. Tischer, Hilsenstein, Hanson et al. (2014) "Adaptive fluorescence microscopy by online feedback image analysis" *Methods Cell Biol*
325. Scherf, Huisken (2015) "The smart and gentle microscope" *Nat Biotechnol*
326. Royer, Lemon, Chhetri et al. (2016) "Adaptive light-sheet microscopy for long-term, high-resolution imaging in living organisms" *Nat Biotechnol*
327. Tosi, Lladóa, Bardia et al. (2021) "AutoScanJ: A suite of imageJ scripts for intelligent microscopy" *Front Bioinf*
328. Vladymyrov, Abe, Moalli et al. (2016) "Real-time tissue offset correction system for intravital multiphoton microscopy" *J Immunol Methods*
329. Schorb, Haberbosch, Hagen et al. (2019) "Software tools for automated transmission electron microscopy" *Nat Methods*
330. Dittrich, Köhrer, Schorb et al. "A high-throughput electron tomography workflow reveals over-elongated centrioles in Morone and D"
331. (2022) */refractory multiple myeloma. Cell Rep Methods*
332. Bepler, Borst, Bouvette et al. (2022) "Smart parallel automated cryo-electron tomography" *J Struct Biol*
333. Schroeder, Dobson, Rueden et al. (2021) "The imageJ ecosystem: open-source software for image visualization, processing, and analysis" *Protein Sci*
334. Pinkard, Stuurman, Ivanov et al. (2021) "Pycromanager: open-source software for customized and reproducible microscope control" *Nat Methods*
335. Edelstein, Stuurman, Tsuchida et al. (2014) "Advanced methods of microscope control using mManager software" *JBM*
336. Munck, Bo, Cawthorne et al. (2024) "Innovating in a bioimaging core through instrument development" *J Microscopy*
337. Gualda, Vale, Almada et al. (2013) "OpenSpinMicroscopy: an open-source integrated microscopy platform" *Nat Methods*
338. Oellermann, Jolles, Ortiz et al. (2022) "Open hardware in science: the benefits of open electronics" *Integr Comp Biol*
339. Boulter, Colombelli, Henriques et al. (2022) "The LEGO ® brick road to open science and biotechnology" *Trends Biotechnol*
340. Bonvoisin, Molloy, Häuer et al. (2020) "Standardisation of practices in open source hardware" *J Open Hardware*
341. Wenzel (2023) "Open hardware: from DIY trend to global transformation in access to laboratory equipment" *PloS Biol*
342. Allan, Burel, Moore et al. (2012) "OMERO: flexible, model-driven data management for experimental biology" *Nat Methods*
343. Dang, Le, Irmer et al. (2021) "APEER: an interactive cloud platform for microscopists to easily deploy deep learning"
344. Pereira, Sousa, Almeida et al. (2019) "Coherent-hybrid STED: high contrast sub-diffraction imaging using a bi-vortex depletion beam" *Optics Express*
345. Skansi (2018) "Introduction to deep learning: from logical calculus to artificial intelligence"
346. Turing (1950) "Computing machinery and intelligence" *Mind*
347. Mcculloch, Pitts (1943) "A logical calculus of the ideas immanent in nervous activity" *Bull Math Biophysics*
348. Rosenblatt (1958) "The perceptron: A probabilistic model for information storage and organization in the brain" *psychol Rev*
349. Werbos (1974) "Beyond regression: new tools for prediction and analysis in the behavioral sciences"
350. Parker (1985) "Learning logic report TR-47"
351. Lecun (1985) "Une procedure d'apprentissage pour Reśeau à seuil assymetrique in Cognitiva 85: A la Frontière de l'Intelligence Artificielle" *Des Sciences de la Connaissance et des Neurosciences*
352. Lecun, Bengio, Hinton (2015) "Deep learning" *Nature*
353. Schmidhuber (2015) "Deep learning in neural networks: an overview" *Neural Networks*
354. Chollet (2018) "Deep learning with python"
355. Brooks (2024) "Inside the maths that drives AI" *Nature*
356. Esteva, Kuprel, Novoa et al. (2017) "Dermatologist-level classification of skin cancer with deep neural networks" *Nature*
357. Weigert, Schmidt, Sjöström et al. (1999) "Nuclei instance segmentation and classification in histopathology images with starDist"
359. Tian, Zhang, Lin et al. (2024) "Dependency-aware deep generative models for multitasking analysis of spatial omics data" *Nat Methods*
360. Stirling, Carpenter, Cimini (2021) "CellProfiler Analyst 3.0: Accessible data exploration and machine learning for image analysis" *Bioinformatics*
361. Simon, Swanson, Zou (2024) "Language models for biological research: A primer" *Nat Methods*
362. Sutton, Barto, Learning (2018)
363. Baniukiewicz, Lutton, Collier et al. (2019) "Generative adversarial networks for augmenting training data of microscopic cell images" *Front Comput Sci*
364. Greener, Kandathil, Moffat et al. (2022) "A guide to machine learning for biologists" *Nat Rev Mol Cell Biol*
365. Moen, Bannon, Kudo et al. (2019) "Deep learning for cellular image analysis" *Nat Methods*
366. Gündüz, Binder, To et al. (2023) "A self-supervised deep learning method for data-efficient training in genomics" *Commun Biol*
367. Ciresan, Meier, Masci et al. (2011)
368. Flexible "high performance convolutional neural networks for image classification"
369. Krizhevsky, Sutskever, Hinton (2017) "ImageNet classification with deep convolutional neural networks" *Commun ACM*
370. Paszke, Gross, Chintala et al. (2017) "Automatic differentiation in PyTorch"
371. Cittadini, Marinoni, Buttazzo (2025) "A hardware accelerator to support deep learning processor units in real-time image processing" *Engineering Applications of Artificial Intelligence*
372. Caba, Dıáz, Barba et al. (2020) "FPGA-Based On-Board Hyperspectral Imaging Compression: Benchmarking Performance and Energy Efficiency against GPU Implementations"
373. Meijering, Arganda-Carreras, Kaynig et al. (2017) "Trainable Weka Segmentation: A machine learning tool for microscopy pixel classification" *Comput Struct Biotechnol J*
374. Arzt, Deschamps, Schmied et al. (2022) "LABKIT: labeling and segmentation toolkit for big image data" *Front Comput Sci*
375. Belevich, Joensuu, Kumar et al. (2016) "Microscopy image browser: A platform for segmentation and analysis of multidimensional datasets" *PloS Biol*
376. Powell, Davis (2024) "Learning structural heterogeneity from cryo-electron sub-tomograms with tomoDRGN" *Nat Methods*
377. Liu, Zhang, Wang et al. (2022) "Isotropic reconstruction for electron tomography with deep learning" *Nat Commun*
378. He, Lin, Chen et al. (2022) "Model building of protein complexes from intermediate-resolution cryo-EM maps with deep learning-guided automatic assembly" *Nat Commun*
379. Jamali, Käll, Zhang et al. (2024) "Automated model building and protein identification in cryo-EM maps" *Nature*
380. Jih, Liu, Liu et al. (2024) "The incredible bulk: Human cytomegalovirus tegument architectures uncovered by AI-empowered cryo-EM" *Sci Adv*
381. Christensen, Ward, Lu et al. (2021) "ML-SIM: Universal reconstruction of structured illumination microscopy images using transfer learning" *Biomed Optics Express*
382. Qiao, Li, Guo et al. (2021) "Evaluation and development of deep neural networks for image super-resolution in optical microscopy" *Nat Methods*
383. Jin, Liu, Zhao et al. (1934) "Deep learning enables structured illumination microscopy with low light levels and enhanced speed" *Nat Commun*
384. Ma, Tan, He et al. (2024) "Pretraining a foundation model for generalizable fluorescence microscopy-based image restoration" *Nat Methods*
385. Qiao, Zeng, Meng et al. (2024) "Zero-shot learning enables instant denoising and super-resolution in optical fluorescence microscopy" *Nat Commun*
386. Priessner, Gaboriau, Sheridan et al. (2022) "Content-aware frame interpolation (CAFI): Deep learning-based temporal superresolution for fast bioimaging" *Nat Methods*
387. Morales-Curiel, Gonzalez, Castro-Olvera et al. (2022) "Volumetric imaging of fast cellular dynamics with deep learning enhanced bioluminescence microscopy" *Commun Biol*
388. Lee, Lee, Song et al. (2020) "Deep-learning-based threedimensional label-free tracking and analysis of immunological synapses of CAR-T cells. eLife"
389. Ravindran (2022) "Five ways deep learning has transformed image analysis" *Nature*
390. Krull, Buchholz, Jug (2019) "Noise2Void-Learning denoising from single noisy images"
391. Weigert, Schmidt, Boothe et al. (2018) "Contentaware image restoration: Pushing the limits of fluorescence microscopy" *Nat Methods*
392. Li, Su, Guo et al. (2022) "Incorporating the image formation process into deep learning improves network performance" *Nat Methods*
393. Zhang, Li, Zhang et al. (2023) "Bio-friendly long-term subcellular dynamic recording by self-supervised image enhancement microscopy" *Nat Methods*
394. Li, Hu, Chen et al. (2023) "Spatial redundancy transformer for self-supervised fluorescence image denoising" *Nat Comput Sci*
395. Rehman, Zhovmer, Sato et al. (2024) "Convolutional neural network transformer (CNNT) for fluorescence microscopy image denoising with improved generalization and fast adaptation" *Sci Rep*
396. Qiao, Li, Liu et al. (2023) "Rationalized deep learning super-resolution microscopy for sustained live imaging of rapid subcellular processes" *Nat Biotechnol*
397. Chen, Sasaki, Lai et al. (2021) "Three-dimensional residual channel attention networks denoise and sharpen fluorescence microscopy image volumes" *Nat Methods*
398. Chong, Cheng, Li et al. (2023) "Unsupervised image denoising for real-world optical and electron microscopy data" *Comput Biol Med*
399. Laine, Jacquemet, Krull (2021) "Imaging in focus: An introduction to denoising bioimages in the era of deep learning" *Int J Biochem Cell Biol*
400. Lu, Oetjen, Bender et al. (2023) "IMC-Denoise: a content aware denoising pipeline to enhance Imaging Mass Cytometry" *Nat Commun*
401. Dai, You, Wang et al. (2025) "Deep learning-enabled filter-free fluorescence microscope" *Sci Adv*
402. Liu, Zou, Sha et al. (2024) "Deep learning-enhanced snapshot hyperspectral confocal microscopy imaging system" *Optics Express*
403. Yao, Smal, Grigoriev et al. (2020) "Deep-learning method for data association in particle tracking" *Bioinformatics*
404. Andhari, Rinaldi, Nazari et al. (2024) "Quality control of immunofluorescence images using artificial intelligence" *Cell Rep Phys Sci*
405. Buchholz, Krull, Shahidi et al. (2019) "Content-aware image restoration for electron microscopy" *Methods Cell Biol*
406. Falk, Mai, Bensch et al. (2018) "U-Net: Deep learning for cell counting, detection, and morphometry" *Nat Methods*
407. Stringer, Pachitariu (2025) "Cellpose3: one-click image restoration for improved cellular segmentation" *Nat Methods*
408. Archit, Freckmann, Nair et al. (2025) "Segment anything for microscopy" *Nat Methods*
409. Kirillov, Mintun, Ravi et al. (2023) "Segment anything"
410. Morone, Marazza, Bergmann et al. (2020) "Deep learning approach for quantification of organelles and misfolded polypeptide delivery within degradative compartments" *Mol Biol Cell*
411. Suga, Nakamura, Nakanishi et al. (2023) "An interactive deep learning-based approach reveals mitochondrial cristae topologies" *PloS Biol*
412. Pulfer, Pizzagalli, Gagliardi et al. (2024) "Transformer-based spatial-temporal detection of apoptotic cell death in live-cell imaging. eLife"
413. Gómez-De-Mariscal, Rosario, Pylvänäinen et al. (2024) "Harnessing artificial intelligence to reduce phototoxicity in live imaging" *J Cell Sci*
414. Andriasyan, Yakimovich, Petkidis et al. (2021) "Microscopy deep learning predicts virus infections and reveals mechanics of lyticinfected cells. iScience"
415. Omange, Kim, Kolhatkar et al. (2025) "AI discovery of TLR agonist-driven phenotypes reveals unique features of peripheral cells from healthy donors and ART-suppressed people living with HIV" *Front Immunol*
416. Chen, Dai, Sun et al. (2017) "Convolutional neural networks for automated annotation of cellular cryo-electron tomograms" *Nat Methods*
417. Li, Li, Zhou et al. (2023) "Real-time denoising enables highsensitivity fluorescence time-lapse imaging beyond the shot-noise limit" *Nat Biotechnol*
418. Moghadam, Chen (2021) "Tracking leukocytes in intravital time lapse images using 3D cell association learning network" *Artif Intell Med*
419. Kobayashi, Cheveralls, Leonetti et al. (2021) "Self-supervised deep learning encodes high-resolution features of protein subcellular localization" *Nat Mach Intell*
420. Wang, Zhao, Marostica et al. (2024) "A pathology foundation model for cancer diagnosis and prognosis prediction" *Nature*
421. Bortolomeazzi, Montorsi, Temelkovski et al. (2022) "A SIMPLI (Single-cell identification from multiPLexed images) approach for spatially-resolved tissue phenotyping at single-cell resolution" *Nat Commun*
422. Song, Kim, Lee et al. (2024) "VONet: A deep learning network for 3D reconstruction of organoid structures with a minimal number of confocal images" *Patterns*
423. Palla, Spitzer, Klein et al. (2022) "Squidpy: A scalable framework for spatial omics analysis" *Nat Methods*
424. Crainiciuc, Palomino-Segura, Molina-Moreno et al. (2022) "Behavioural immune landscapes of inflammation" *Nature*
425. Acharya, Mukhopadhyay (2024) "A comprehensive review of machine learning techniques for multi-omics data integration: Challenges and applications in precision oncology" *Briefings Funct Genomics*
426. Ballard, Wang, Li et al. (2024) "Deep learning-based approaches for multi-omics data integration and analysis" *BioData Min*
427. Royer, Driscoll, Zaritsky (2021) "Omega-Harnessing the power of large language models for bioimage analysis" *Nat Methods*
428. Abdelaziz, Ismail, Mabrouk et al. (2024) "Multi-omics data integration and analysis pipeline for precision medicine: Systematic review" *Comput Biol Chem*
429. O'donnell, Rubinsteyn, Laserson (2020) "MHCflurry 2.0: improved pan-allele prediction of MHC class I-presented peptides by incorporating antigen processing" *Cell Syst*
430. Patrick, Canete, Iyengar et al. (2023) "Spatial analysis for highly multiplexed imaging data to identify tissue microenvironments" *Cytometry Part A*
431. Eulenberg, Köhler, Blasi et al. (2017) "Reconstructing cell cycle and disease progression using deep learning" *Nat Commun*
432. Pizzagalli, Gonzalez, Krause (2019) "A trainable clustering algorithm based on shortest paths from density peaks" *Sci Adv*
433. Bilate, London, Castro et al. (2020) "T cell receptor is required for differentiation, but not maintenance, of intestinal CD4+ Intraepithelial lymphocytes" *Immunity*
434. Lewis, Asselin-Labat, Nguyen et al. (2021) "Spatial omics and multiplexed imaging to explore cancer biology" *Nat Methods*
435. Rosen, Brbićm, Roohani et al. (2024) "Toward universal cell embeddings: Integrating single-cell RNA-seq datasets across species with SATURN" *Nat Methods*
436. Mencattini, Giuseppe, Comes et al. (2020) "Discovering the hidden messages within cell trajectories using a deep learning approach for in vitro evaluation of cancer drug treatments" *Sci Rep*
437. Vaswani, Shazeer, Parmar et al. (2017) "Attention is all you need" *Adv Neural Inf Process Syst*
438. Lei, Fuster-Barcelóc, Reder et al. (2024) "IO Chatbot: A community-driven AI assistant for integrative computational bioimaging" *Nat Methods*
439. Szałata, Hrovatin, Becker et al. (2024) "Transformers in single-cell omics: A review and new perspectives" *Nat Methods*
440. Zhang, Dai, Huang et al. (2024) "Multimodal large language models for bioimage analysis" *Nat Methods*
441. Gui, Sun, Wen et al. (2023) "A review on generative adversarial networks: algorithms, theory, and applications" *IEEE Trans Knowledge Data Eng*
442. Abramson, Adler, Dunger et al. (2023) "Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization" *Nat Methods*
443. Rangan, Feathers, Khavnekar et al. (2024) "CryoDRGN-ET: Deep reconstructing generative networks for visualizing dynamic biomolecules inside cells" *Nat Methods*
444. Zaritsky, Jamieson, Welf et al. (2021) "Interpretable deep learning uncovers cellular properties in label-free live cell images that are predictive of highly metastatic melanoma" *Cell Syst*
445. El Nahhas, Loeffler, Carrero et al. (2024) "Regression-based Deep-Learning predicts molecular biomarkers from pathology slides" *Nat Commun*
446. Hederman, Ackerman (2023) "Leveraging deep learning to improve vaccine design" *Trends Immunol*
447. Pertseva, Gao, Neumeier et al. (2021) "Applications of machine and deep learning in adaptive immunity" *Annu Rev Chem Biomolecular Eng*
448. Stringer, Wang, Michaelos et al. (2021) "Cellpose: A generalist algorithm for cellular segmentation" *Nat Methods*
449. Goḿez-De-Mariscal, Garcıá-Loṕez-De-Haro, Ouyang et al. (2021) "DeepImageJ: A user-friendly environment to run deep learning models in ImageJ" *Nat Methods*
450. Maffulli (2025) "Open source' AI isn't truly open-Here's how researchers can reclaim the term" *Nature*
451. Hartley, Kleywegt, Patwardhan et al. (2022) "The bioImage archivebuilding a home for life-sciences microscopy data" *Comput Resour Mol Biol*
452. Wilkinson, Dumontier, Aalbersberg et al. (2014) "The FAIR Guiding Principles for scientific data management and stewardship. Sci Data"
453. Pedregosa, Varoquaux, Gramfort et al. (2011) "Scikit-learn: machine learning in python" *J Mach Learn Res*
454. Wang, Wang, Zhao et al. (2024) "3D live imaging and phenotyping of CAR-T cell mediated-cytotoxicity using high-throughput Bessel oblique plane microscopy" *Nat Commun*
455. Wingert, Lee, Cannesson (2021) "Machine learning, deep learning, and closed loop devices-Anesthesia delivery" *Anesthesiology Clinics*
456. Singh, Nath "Artificial intelligence and anesthesia: A narrative review"
457. Saudi, Anaesthesia (2022)
458. Carpenter, Cimini, Eliceiri (2023) "Smart microscopes of the future" *Nat Methods*
459. Kawaguchi, Bengio, Kaelbling "Generalization in Deep Learning"
460. Grohs, Kutyniok (2022) "Mathematical Aspects of Deep Learning, 1st"
461. Banerjee, Agarwal, Singla (2024) "LLMs will always hallucinate, and we need to live with this. arXiv"
462. Duede (2023) "Deep learning opacity in scientific discovery" *Philosophy Sci*
463. Germain, Van Allen, Trynka et al. (2024) *AI and immunology. Immunity*
464. Meade (2019) "Reply to: One neuron versus deep learning in aftershock prediction" *Nature*
465. Lee, Eisenman, Khuon et al. (2024) "Believing is seeingthe deceptive influence of bias in quantitative microscopy" *J Cell Sci*
466. Durkee, Abraham, Clark et al. (2021) "Artificial intelligence and cellular segmentation in tissue microscopy images" *Am J Pathol*
467. Salehin, Islam, Saha et al. (2024) "AutoML: A systematic review on automated machine learning with neural architecture search" *J Inf Intell*
468. Lewis, Perez, Piktus et al. (2020) "Retrievalaugmented generation for knowledge-intensive NLP tasks"
469. Gottweis, Weng, Daryin et al. (2025) "Towards an AI co-scientist" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584909&blobtype=pdf | # Introduction to XIV updating course of antimicrobials and infectious diseases Introducción al XIV curso de actualización en antimicrobianos y enfermedades infecciosas
Francisco Candel, Mayra Matesanz, Francisco Javier
Last February, the 14th Course on Antimicrobials and Infectious Diseases Update took place at the Hospital Clínico San Carlos in Madrid. This scientific activity was accredited by the Community of Madrid (Continuing Education Commission for Healthcare Professions of the Community of Madrid, file Exp. 07-AFOC-07063.7/2024) and endorsed by the Spanish Society of Clinical Microbiology and Infectious Diseases (SEIMC), the Spanish Society of Chemotherapy (SEQ), and the Madrid Society of Clinical Microbiology (SMMC). In this edition, the course was held in a hybrid format-both in-person and online-achieving participation from 155 attendees and 786 continuous online connections from Spain and several European and American countries (Figure 1). The audience consisted of multidisciplinary professionals from all infection-related specialities, and the faculty provided an update on the most relevant aspects of bacteriology, mycology, and virology. This issue of the journal includes summaries of the lectures presented.
## Diagnostic update
González del Castillo at al. analyzed the usefulness of the Monocyte Distribution Width (MDW) parameter as an early biomarker of infection in critically ill patients seen in emergency departments. This index, integrated into automated blood counts, showed a diagnostic performance comparable to or better than classical biomarkers, with a sensitivity of 73% and a specificity of 72%. Its immediate availability and low cost make it a promising tool for early detection of sepsis (particularly in elderly patients) supporting its integration into clinical algorithms for suspected infection [1][2].
Cercenado and Candel reviewed the main characteristics of the new β-lactamase inhibitors (enmetazobactam, avibactam, relebactam, durlobactam, zidebactam, nacubactam, vaborbactam, taniborbactam, and xeruborbactam), their spectrum of activity, and the available therapeutic combinations. They emphasized how these molecules expand coverage against class A, C, and D enzymes and restore the usefulness of classic β-lactams against Enterobacterales, Pseudomonas aeruginosa, and Acinetobacter baumannii. However, they also warn that emerging resistance during treatment (through KPC mutations, AmpC modifications, or porin loss) requires prudent, microbiologically guided use. Overall, the new BL-BLI combinations (ceftazidime-avibactam, meropenem-vaborbactam, imipenem-relebactam, aztreonam-avibactam, sulbactam-durlobactam, or meropenem-xeruborbactam) represent a crucial advance in managing infections caused by carbapenemase-producing microorganisms, always within an active surveillance framework [3][4][5].
Cantón and Halperín analyzed the importance of early diagnosis and diagnostic stewardship programs, highlighting the need for 24/7 active laboratories and the incorporation of rapid technologies such as MALDI-TOF or molecular panels. These tools reduce turnaround time, improve therapeutic adequacy, and strengthen collaboration between clinicians and microbiologists in the context of sepsis [6][7].
Sanz, Burillo, and García-Lechuz examined the role of syndromic platforms in the diagnosis of critically ill patients. The simultaneous detection of pathogens and resistance mechanisms through multiplex PCR or microarrays enables early treatment adjustment and reduces hospital stay, consolidating their use as an essential tool in antimicrobial stewardship programs [8][9].
Maldonado-Barrueco and colleagues updated advances in the diagnosis of invasive fungal infections, emphasizing the importance of galactomannan, β-D-glucan, and molecular techniques-together with mass spectrometry to improve early detection of pathogenic fungi and reduce mortality [10][11][12].
Muñoz-Echeverría and colleagues highlighted the importance of early clinical recognition and immunological and genetic characterization for the diagnosis and management of primary immunodeficiencies in adults, diseases that are often underdiagnosed [13][14].
## Therapeutic update
Pina-Sánchez and colleagues reviewed therapeutic optimization strategies against bacteria carrying metalloenzymes, emphasizing the relevance of combinations with new β-lactamase inhibitors and the application of pharmacokinetic and pharmacodynamic principles in clinical decision-making [15][16][17].
Maseda and Suárez de la Rica focused their analysis on the epidemiological and therapeutic evolution of nosocomial peritonitis, highlighting the increase in multidrug-resistant pathogens and emerging antimicrobial options (ceftazidime-avibactam, meropenem-vaborbactam, imipenem-relebactam, or eravacycline). They addressed the usefulness of rapid molecular techniques for early pathogen identification, the limited role of echinocandins in intra-abdominal candidiasis, and the need to shorten therapy duration through strategies guided by procalcitonin and fungal biomarkers. Their analysis synthesizes a rational, up-to-date approach to complicated intra-abdominal infection in the ICU [18][19].
Soriano-Cuesta and colleagues underlined the importance of empirical treatment adjusted to risk and early de-escalation according to microbiological results. The new antibiotics active against carbapenemases and Acinetobacter spp. represent a key tool for improving clinical outcomes [20][21].
Blanes-Hernández and colleagues challenged the classical dogmas of antibiotic therapy (monotherapy versus combination, treatment duration, or actual impact on resistance), promoting a more rational use based on scientific evidence [22][23].
From the viral infection perspective, Fortún analyzed recent advances in the prophylaxis, monitoring, and treatment of cytomegalovirus, especially in the transplant setting. He highlighted the introduction of next-generation antivirals and the importance of personalized strategies according to risk and resistance [24][25]. Finally, Núñez-Orantos reviewed the latest advances in HIV infection, including long-acting therapies, simplified regimens, and combined prevention strategies (PrEP and vaccination) [26][27].
## References
1. Ciaccio, Agnello, Sasso et al. (2023) "Monocyte distribution width (MDW) as a biomarker of sepsis: an evidenced-based laboratory medicine approach" *Clin Chim Acta*
2. Encabo, Hernández-Álvarez, Oteo et al. (2023) "Monocyte distribution width (MDW) as an infection indicator in severe patients attending in the Emergency Department: a pilot study" *Rev Esp Quimioter*
3. (2016) "Assessing the health burden of infections with antibioticresistant bacteria in the EU/EEA"
4. Peñalva, Cantón, Pérez-Rodríguez et al. (2025) "Burden of bacterial antimicrobial resistance among hospitalised patients in Spain: findings from three nationwide prospective studies" *Lancet Reg Health Eur*
5. Bush (2018) "Past and present perspectives on β-lactamases" *Antimicrob Agents Chemother*
6. Kumar, Roberts, Wood et al. (2006) "Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock" *Crit Care Med*
7. Peri, Chatfield, Ling et al. (2024) "Rapid diagnostic tests and antimicrobial stewardship programs for the management of bloodstream infection: what is their relative contribution to improving clinical outcomes? a systematic review and network meta-analysis" *Clin Infect Dis*
8. Khare, Kothari, Castagnaro et al. (2020) "Active monitoring and feedback to improve blood culture fill volumes and positivity across a large integrated health system" *Clin Infect Dis*
9. Candel, Salavert, Cantón et al. (2024) "The role of rapid multiplex molecular syndromic panels in the clinical management of infections in critically ill patients: an experts-opinion document" *Crit Care*
10. Boch, Reinwald, Spiess et al. (2018) "Detection of invasive pulmonary aspergillosis in critically ill patients by combined use of conventional culture, galactomannan, 1-3-beta-D-glucan and Aspergillus-specific nested polymerase chain reaction in a prospective pilot study" *J Crit Care*
11. Leeflang, Debets-Ossenkopp, Wang et al. (2015) "Galactomannan detection for invasive aspergillosis in immunocompromised patients" *Cochrane Database Syst Rev*
13. Wu, Wang, Tan et al. (2021) "Diagnostic value of galactomannan in serum and bronchoalveolar lavage fluid for invasive pulmonary aspergillosis in non-neutropenic patients" *Diagn Microbiol Infect Dis*
14. Sullivan, Orange (2024) "Inborn errors of immunity (primary immunodeficiencies): classification. UpToDate. Last updated Apr 15"
15. Tangye, Al-Herz, Bousfiha et al. (2022) "Human inborn errors of immunity: 2022 update on classification" *J Clin Immunol*
16. Boattini, Gaibani, Comini et al. (2025) "In vitro activity and resistance mechanisms of novel antimicrobial agents against metallo-β-lactamase producers" *Eur J Clin Microbiol Infect Dis*
17. Grabein, Arhin, Daikos et al. (2024) "Navigating the current treatment landscape of metallo-β-lactamase-producing Gram-negative infections: what are the limitations?" *Infect Dis Ther*
18. Hidalgo-Tenorio, Bou, Oliver et al. (2024) "The challenge of treating infections caused by metallo-β-lactamaseproducing Gram-negative bacteria: a narrative review" *Drugs*
19. Blot, Antonelli, Arvaniti et al. (2019) "Epidemiology of intra-abdominal infection and sepsis in critically ill patients: "AbSeS", a multinational observational cohort study and ESICM Trials Group project" *Intensive Care Med*
20. Pascale, Antonelli, Deschepper et al. (2022) "Abdominal Sepsis Study (AbSeS) group and the Trials Group of the European Society of Intensive Care Medicine. Poor timing and failure of source control are risk factors for mortality in critically ill patients with secondary peritonitis" *Intensive Care Med*
21. Tamma, Heil, Justo et al. (2024) "Infectious Diseases Society of America 2024 guidance on the treatment of antimicrobial-resistant Gram-negative infections" *Clin Infect Dis*
22. Timsit, Bassetti, Cremer et al. (2019) "Rationalizing antimicrobial therapy in the ICU: a narrative review" *Intensive Care Med*
23. Lam, Bourassa-Blanchette (2023) "Ten common misconceptions about antibiotic use in the hospital" *J Hosp Med*
24. Mccreary, Johnson, Jones et al. (2023) "Antibiotic myths for the infectious diseases clinician" *Clin Infect Dis*
25. Ruiz-Arabi, Cisneros, Aguilera et al. (2024) "Management of cytomegalovirus in adult solid organ transplant patients: GESITRA-IC-SEIMC, CIBERINFEC, and SET recommendations update" *Transplant Rev*
26. Piñana, Giménez, Vázquez et al. (2024) "Update on cytomegalovirus infection management in allogeneic hematopoietic stem cell transplant recipients: a consensus document of the Spanish Group for Hematopoietic Transplantation and Cell Therapy (GETH-TC)" *Mediterr J Hematol Infect Dis*
27. Kityo, Mugerwa, Walimbwa et al. (2024) "Switch to long-acting cabotegravir and rilpivirine in virologically suppressed adults with HIV in Africa (CARES): week 48 results from a randomized, multicentre, open-label, non-inferiority trial" *Lancet Infect Dis*
28. Bekker, Das, Karim et al. (2024) "PURPOSE 1 Study Team. Twice yearly lenacapavir for HIV prevention in cisgender women" *N Engl J Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12539661&blobtype=pdf | # The Genetic Diversity and Drug Resistance Patterns of HIV-1 Pol Gene in East Africa
Aneth Nzinyangwa Kavuraya, Teddy Mselle, Fulgence Mpenda
## Abstract
Human immunodefciency virus-1 (HIV-1) is among the most genetically diverse pathogens due to expeditious molecular evolution. Te rapid change in HIV genomes intricates HIV transmission and progression and attributes HIV resistance to antiretroviral therapy (ART). In East Africa, as in other parts of the globe, HIV-1 occurs in various subtypes, circulating recombinant form (CRF) and unique recombinant forms, with subtype A1 being the most predominant. Surveillance of HIV-1 molecular diversity and drug resistance mutations (DRMs) is a linchpin for monitoring viral evolution and treatment efciency. However, consolidated reports on the same are limited, and therefore, the pursuit of meta-analysis was sought to analyze genetic diversity and drug resistance patterns of HIV-1 pol gene and their geographical distributions in four East African countries (Kenya, Uganda, Tanzania, and Ethiopia). We retrieved 7614 HIV-1 pol gene sequences, deposited between 2015 and 2025 from the Los Alamos HIV databases. Te predominant HIV-1 subtypes were A1 (40.2%), C (21.5%), and D (17.7%), with geographical variability. A notable frequency of inter-subtype recombinant was observed with recombinants A1D (9.5%) and A1C (2.94%) being prevalent. Few CRFs (> 0.1%) were identifed. DRM were present in 42.8% of the sequences, with the majority associated with NNRTIs (36.5%) and NRTIs (25.5%). Te most frequently associated mutations were K103N and M184V. Although resistance to INSTI (3.7%) remained minimal, its presence warrants continued monitoring. A signifcant association between HIV-1 subtypes and DRM prevalence was observed (χ 2 �102.43, p < 0.0001), with subtypes showing varied resistance burdens. Tese fndings underscore the variability in HIV-1 genetic diversity across studied East African countries, highlighting the need for region-specifc interventions, to optimize HIV-1 control in this region.
## 1. Introduction
Human immunodefciency virus (HIV), the causative agent of acquired immunodefciency syndrome (AIDS), is the most serious public health challenges. Taken as an example, globally, UNAIDS extrapolated that about 39.9 million people were living with HIV in 2023. Intriguing, in the same year, 1.3 million new infections and 630,000 HIV/AIDS attributed deaths were reported by Joint United Nations Programme on HIV/AIDS [1]. Although the distribution of HIV is skewed, Africa is particularly the most afected region accounting for about 65% of the global population of people living with HIV [2]. Likewise, the data from East and South Africa is alarming, in particular about 52% (20.8 million) of people living with HIV/AIDS reported in 2023 were from East and South Africa [3]. Sundry factors like high-risk cultural practices, socioeconomic challenges, and limited access to healthcare have been portrayed to account for skewed burden of HIV in Africa, but the intrinsic factors like the HIV genetic variations and the temporal efects cannot be underestimated.
Te HIV genome is approximately 9.7 kilobases in length and is composed of single-stranded positive-sense RNA [4]. Tere are two variants of HIV, which are HIV-1 and HIV-2. HIV-1 is responsible for the majority of HIV infections worldwide and is categorized into four genetic groups: M (major or main), N (non-M, non-O) [5], O (outlier) [4], and P [6]. Te groups of HIV-1 have diferent geographical distributions, but have similar clinical symptoms [7]. Group M is the major group associated with the global HIV pandemic [8]. Tis group is further divided into 10 subtypes (A-D, F-H, and J-L) and 8 sub-subtypes (A1, A2, A3, A4, A5, A6, F1, and F2) [9,10]. Of interest, the subtypes/subsubtype may recombine to exist in recombinant forms and about 118 circulating recombinant forms (CRFs) have been reported. Nevertheless, some of the recombinants have not been documented, and therefore they are regarded as unique recombinant forms (URFs). Te URFs are less prevalent and very unlikely to be detected. On the other hand, HIV-2 is antithetical to HIV-1 because HIV-2 is less common and less virulent. HIV-2 is mainly found in Western Africa with about 2 million people infected [11]. Te present information refects the remarkable genetic diversity of HIV-1, which varies widely across regions and populations.
Being among the most genetically diverse pathogens, HIV-1 undergoes constant molecular evolutions [12,13]. Tis appreciable viral diversity within HIV-1 is due to highly mutational escape and its adaptation to both immune activity and antiretroviral therapy (ART) [14]. Te extensive genetic diversity of HIV-1 is due to the following factors: (i) the high replication rate; (ii) the activity of reverse transcriptase (RT), which favors the accumulation of transcription errors that the enzyme is unable to correct since it lacks 3′ to 5′ exonuclease activity; (iii) the recombination events that may occur during virus replication; (iv) host selective immune pressure. [14][15][16]. Tis results in a dramatic change in genetic diversity and an increase in both intrahost variability (6%-19%) and interhost variability (2%-5%) [17]. As compared with other viral components, the Pol protein has higher mutation rate, and its role in viral survival has been highly studied.
Te HIV-1 pol gene encodes the three enzymes needed for viral replication: protease (PR), RT, and integrase (IN). Tese proteins have essential roles in the viral cycle and are the main targets of antiretroviral drugs (ARV) [16]. ARV drugs targeting the pol gene enzymes are divided into four groups: protease inhibitors (PIs), non-nucleoside reverse transcriptase inhibitors (NNRTI), nucleoside reverse transcriptase inhibitors (NRTI), and recently introduced integrase transfer inhibitors (INSTIs) [18]. Tere are fve PIs, ten RT inhibitors, and fve INSTIs FDA approved and recommended in the U.S. Department of Health and Human Services (HHS) HIV/AIDS medical practice guidelines [19]. Te identifcation of Pol mutations associated with ART through molecular detection and the use of online resistance algorithms has signifcantly enhanced resistance monitoring and ART selection for individuals with HIV [20][21][22].
Several studies have explored the genetic diversity and prevalence of HIV-1 subtypes across East Africa (EA). In EA, the HIV subtype A has remained persistent over long epoch, and available reports highlights that more than half of HIV infections among the EA individuals are attributed to subtype A. However, other subtypes like subtypes C and D have relatively noticeable proportion of HIV infections in EA [17,23]. In addition, few CRFs and URFs that contribute to the genetic complexity of HIV-1 in the region are documented [24,25]. HIV-1 subtypes and recombinants may be associated with various phenotypes such as drug resistance evolution, disease progression, transmission patterns, and neuropsychological outcome. [26]. In terms of drug resistance, studies on East African population have consistently shown high resistance mutations rates to NNRTIs and NRTIs and low prevalence of resistance to PIs and INSTIs [27,28] Despite substantial global research on HIV diversity and drug resistance, comprehensive information in the region is limited and conjured the present study. Tis study analyzed HIV-1 pol gene sequences from the Los Alamos HIV Database to characterize genetic diversity, subtypes, and drug resistance mutations (DRMs) across EA, with the fndings providing insights into ART, enhance drug resistance surveillance, inform vaccine design, and strengthen regional HIV prevention and control strategies.
## 2. Materials and Methods
## 2.1. Study Overview.
Tis study employed a meta-analysis approach to evaluate the subtype distribution, diversity and drug resistance of the HIV-1 pol gene across EA. Data on HIV-1 pol gene sequences were retrieved from public database. Using appertaining software, the sequences were analyzed for assessment of DRMs in the pol gene enzymes (PR, RT, and IN).
## 2.2. Acquisition of HIV-1 Pol Gene Sequences From Public
Database. In June 2025, HIV-1 pol gene sequences deposited between 2015 and 2025 from four East African countries were retrieved in FASTA format with subtyping information from the Los Alamos National Laboratory (LANL) HIV Database (https://www.hiv.lanl.gov). Duplicate accession numbers were removed, and only unique sequences were retained. Sequences with more than 3% IUPAC nucleotide ambiguity codes (e.g., M, R, Y) across the full pol gene were excluded from the dataset. Te subtype result per country on a map in Figure 1 was created by using ArcMap software.
## 2.3. Assessment of DRMs in PR, RT, and IN Regions.
Identifcation and classifcation of mutations associated with drug resistance in the PR, RT, and IN regions of the pol gene were carried out using the Stanford University HIV Drug Resistance Database (https://hivdb.stanford.edu/). Te Stanford HIV Drug Resistance Database (HIVDB) is curated public database and essential resource for public health ofcials monitoring Acquired Drug Resistance and Transmitted Drug Resistance, for scientists developing new ARV drugs, and for HIV care providers managing patients with HIV. In addition, chi-square (χ 2 ) tests were performed using Stata software version 16.0 to assess the relationship between HIV-1 subtypes and drug resistance. Statistical signifcance was determined at a 95% confdence level within a marginal error of 0.05.
## 3. Results
## 3.1. Te HIV-1 Pol Gene Sequences Analyzed From EA in the Present Study.
A total of 7724 HIV-1 pol gene sequences were initially retrieved from LANL HIV sequence database, comprising sequence collected and deposited between 2015 and 2025. Following quality fltering which involved the removal of duplicate entries and sequences with high levels of ambiguous nucleotide characters, 110 sequences were excluded, resulting in 7614 sequences for downstream analysis. Tese included 5730 PR/RT, 734 RT, and 1594 IN sequences. Te sequences were sourced from four EA countries: Kenya, Uganda, Tanzania, and Ethiopia. Te distribution of HIV-1 pol gene sequences by country is presented in Table 1.
## 3.2. Genetic Diversity of HIV-1 Pol Gene Subtypes Across East
African Populations. Subtype with the greatest presentation among HIV-1 pol gene sequences retrieved from four East African countries was subtype A1 accounting for 40.2% of all sequences. Of these, 42.7% originated from Kenya, 36.0% from Uganda, and 21.0% from Tanzania. Subtype C represented 24.2% of sequences, with majority from Ethiopia (55.02%), followed by Tanzania (35.9%). Subtype D accounted for 18.1% with the majority from Uganda (64.8%) followed by Kenya (18%) and Tanzania (17.3%), as shown in Table 2. Subtype A1 dominated in Kenya and Uganda, while subtype C was most common in Ethiopia and Tanzania. Inter-subtype recombinants were notable, with subtype A1D comprising 9.5% (724 sequences) of the dataset, followed by A1C (2.9%) and CD (1.5%). CRFs such as 10_CD (0.07%) and 02_AG (0.05%) exhibited a low prevalence across the region. Tese fndings highlight the dominance of subtypes A1, C, and D alongside a diverse array of recombinant forms in East African HIV-1 populations.
Te geographic distribution of HIV-1 variants across the four countries, based on available pol sequences from the LANL, is illustrated in Figure 1.
## 3.3. Prevalence and Profles of DRMs in HIV-1 Pol Gene
Sequences. Te overall prevalence of HIV-1 pol gene sequences with at least one mutation across four East African countries was 42.8% (3266 sequences). Te prevalence varied by country, with the highest observed in Kenya 51.2% (1045/2040), followed by Tanzania at 48.6% (880/1811), Uganda at 38.8% (1038/2673), and Ethiopia at 27.8% (303/1090). Mutations associated with NNRTI and NRTI were the most prevalent, accounting for 36.5% and 25.5% of all sequence analyzed, respectively. In contrast, sequences with PI-and INSTI-associated mutations were less commonly identifed in 6.2% and 3.7%, respectively. Notably, 280 out of the 3266 sequences (8.7%) harbored multiple mutations across more than one drug class region of the pol gene. Te frequency distribution of sequences with at least one mutation by country is illustrated in Figure 2.
Out of 7614 reanalyzed sequences, 2779 (36.5%) harbored at least one NNRTI resistance-associated mutations, with a total of 5432 individual mutations observed. Te most frequent NNRTI mutations were K103N, Y181C, and G190A. Variable levels of resistance to NNRTIs were identifed across the dataset, with high-level resistance most commonly observed against nevirapine (NVP) and efavirenz (EFV). Intermediate and low-level resistance patterns were also noted, particularly for doravirine (DOR), etravirine (ETR), and rilpivirine (RPV) (Figure 3(a)).
Among the studied sequences, 1941 (25.5%) sequences had at least one NRTI resistance-associated mutation, with a total of 5396 individual mutations observed. M184V, K56R, and D67N were among the most frequent NRTI mutations observed. A substantial proportion of sequences exhibited high-level resistance to emtricitabine (FTC), lamivudine (3TC), and didanosine (DDI), as illustrated in Figure 3(b). Other levels of resistance to the mentioned NRTI drugs were also observed among the retrieved pol gene sequence, as illustrated in Figure 3(b). PI resistance-associated mutation was found in 469 (6.2%) sequences, comprising 451 major and 391 accessory mutations, giving a total of 842 individual PIs mutations. Te most frequent PIs mutations including M46I, I54V, and L33F cofered high-level resistance primarily to fosamprenavir/ritonavir (FPV/r), nelfnavir (NFV), and indinavir (IDV/r). A smaller proportion of sequences showed intermediate-and low-level resistance to PIs drugs (Figure 3(c)). A total of 280 (3.7%) sequences were having INSTI resistance-associated mutations, with a total of 322 individual mutations. Te most common INSTI mutations were E138K and G140R, with few high-to intermediate-level resistance observed in elvitegravir (EVG) and cabotegravir (CAB), while other resistance levels were rare (Figure 3(d)).
## 3.4. DRMs Among Diferent HIV-1 Subtypes.
A total of 7614 HIV-1 sequences were analyzed, of which 3266 (42.9%) harbored DRMs. Subtype A1 was the most prevalent, comprising 40.2% (3060/7614) of all sequences and contributing the largest absolute number of resistance strains (1464/3266; 44.8%). Followed by subtypes, C (21.5%; 703/ 3266) and D (17.7%; 578/3266) were also frequent, with 703 and 578 resistant sequences, respectively, as shown in Table 3. Statistical results show that there was signifcant association between the HIV-1 genotype and drug-resistant mutations (χ 2 � 102.43; p ≤ 0.0001).
When resistance was evaluated relative to the number of sequences within each subtype, distinct diferent emerged. Subtype A1 showed a DRM prevalence of 47.8% (1464/
## 4. Discussion
Surveillance of HIV-1 molecular diversity and DRMs is essential for monitoring viral evolution and informing treatment strategies. However, consolidated reports on this aspect have been limited in EA. Tis meta-analysis sought to evaluate the genetic diversity and drug resistance patterns of HIV-1 pol gene and their geographical distributions using publicly available data from 2015 to 2025 in the region.
Tese fndings highlight variation in HIV-1 diversity and geographical distribution across East African countries. Six subtypes, more than fve inter-subtype recombinants and few CRFs were identifed. Te observed diversity is generally comparable to other reports from Africa, with the exception of West Africa where diversity of CRFs surpasses that of pure subtypes [17]. Te predominance of subtype A1 (40.2%) specifcally in Kenya and Uganda aligns with previous reports identifying A1 as the dominant HIV-1 subtype in EA, likely due to high transmissibility and historical spread [12,17,23]. Te substantial presentation of subtype C (24.3%) particularly in Ethiopia and Tanzania, and subtype D (18.1%) in Uganda was noted. Te current study confrmed the predominance of HIV-1 subtype C in Ethiopia similar to previous study suggesting that subtype C accounts for 97% of the pandemic in the country [29].
Te presence of inter-subtype recombinants, notably A1D (9.5%), A1C (2.9%), and CD (1.5%), indicates ongoing recombination as a driver of HIV-1 genetic diversity in EA and the possible introduction of other HIV-1 subtypes from Neighboring countries [30]. Despite being low, the infux of other variants, including subtype B (0.78%), G (0.37%), and CRFs such as 01_AE, 02_AG, 10_CD, and 16_A2D (< 0.1% each), has been observed. Te presence of other recombinant forms including URFs (0.78%) indicates sporadic and novel recombination events, expanding the genetic complexity of HIV-1 in the region [17]. Te occurrence of these other variants indicates the possible introduction of other HIV-1 subtypes, and therefore preventive measures should always be there to prevent the possibility of further introductions of new subtypes in the region.
HIV-1 drug resistance signifcantly challenges efective treatment, often leading to therapy failure and necessitating alternative therapeutic strategies [28,31]. In this analysis, a total of 11,992 individual DRMs were identifed in 3266 (42.8%) of the sequences. Majority of these mutations were associated with NNRTIs (36.5%) and NRTIs (25.5%), refecting the prolonged reliance and associated resistance challenges posed by these drug classes since the inception of ART, as reported both globally and in regions [16,27,28,32]. Tis highlights the signifcance of the current WHO guidelines to substitute NNRTIs in the fxed-dose combinations with dolutegravir for HIV treatment (World Health organization, 2019). All East African countries have adopted the tenofovir, lamivudine, and dolutegravir (TLD) regimes as the frst-line therapy [33,34].
High-level resistance to nevirapine (NVP) and efavirenz (EFV) were observed largely among NNRTI due to key mutations such as K103N, Y181C, and G190A [35][36][37]. M184V was the most common NRTIs-associated mutation conferring high-level resistance to FTC and 3TC [22,38,39]. Tymidine analogue mutations (TAMs) such as T215Y, D67N, M41L, K219K/Q, and K70R were also identifed among RT sequences, echoing previous reports from sub-Saharan Africa [28,34].
PI resistance mutations were less frequent (6.2%) suggesting that PIs may remain efective in this region. However, mutations such as M46I, I54V, and L33F indicate emerging resistance which can compromise the efcacy of certain PI-based regimens. INSTIs resistance was rare, with only 280 (3.7%) sequences harboring mutations, suggesting that this class remains a viable option, particularly for salvage therapy. Notably, INSTI-associated mutations such as E138K and G140R were detected. Although they generally have limited impact when presented alone, they have been associated with reduced susceptibility to Elvitegravir and Raltegravir in few cases, especially when occurring in combination with other resistance mutations particularly those at position 148 [40][41][42].
Studies reported that variation in DRM among HIV-1 patients may be infuenced by the genetic diversity of circulating subtypes and recombinants [43]. In this study, although subtype A1 contributed the highest absolute number of DRM-harboring sequences (44.8%), this pattern is largely attributable to its predominance in the dataset (40.2%). When resistance was normalized by subtype frequency, other variants revealed comparable or even higher proportions of DRMs. Subtype G (42.8%) and the diverse pool of recombinants (up to 58.3% in other recombinants, and 49.2% in CD forms) showed disproportionate resistance burdens relative to their population size. Tese fndings refne the reports that subtype A1 may be associated with DRMs in the EA region [38] by clarifying that this impact refects its epidemiological dominance. Subtypes C and D also showed substantial frequencies of DRMs, with 703 (21.5%) and 578 (17.7%) mutated sequences, respectively.
Te high resistance levels among recombinants were also notable, as recombination may accelerate the accumulation and dissemination of DRM patterns, thereby complicating both treatments monitoring and vaccine design [26,44]. Te complete resistance observed in CRF16_A2D, although based on very few sequences, highlights the need for close surveillance of rare forms that may escape detection in routine monitoring. Subtype C, the second most common subtype, showed a slightly lower proportion of DRMs (38.0%) compared with subtypes A1 (47.8%) and D (41.2%), which may refect diferences in drug exposure histories, treatment regimens, or intrinsic viral factors infuencing mutational pathways. Tese subtype-specifc patterns could contribute to diferential resistance profles, as mutations accumulate at varying rates and positions across subtypes potentially impacting drug efcacy [32]. Terefore, the observed frequency of DRMs across subtypes underscores the importance of continued molecular surveillance and resistance monitoring to inform treatment strategies in the region.
Te presented genetic diversity of HIV-1 in EA is characterized by the predominance of pure subtypes and a notable presence of recombinant forms. Te occurrence of DRMs particularly to NNRTIs and NRTIs, alongside a signifcant association between viral subtypes and resistance patterns, underscores the infuence of genetic diversity on ART outcomes. Although the public database biases such as over-representation of ART-failure cases may infate the observed resistance estimates, these fndings emphasize the need for enhanced molecular surveillance tailored to regional viral diversity. Future studies should incorporate nonpublic datasets and real-time sequencing to better capture evolving HIV-1 dynamics, informing targeted ART strategies to strengthen HIV-1 control across EA.
## 5. Conclusion
In this meta-analysis, we evaluated the genetic diversity and geographical distribution of HIV-1 pol gene in EA by analyzing publicly available sequences from 2015 to 2025. Tis was also the period where new HIV drug was introduced in the combination of frst-line ART regime (INSTI-Dolutegravir in 2019). Te continued dominance of subtype A1 and the emergence of recombinant forms refect the dynamic and evolving nature of HIV-1 in the region. Te observed prevalence of resistance mutations highlights the need for region-specifc interventions, such as routine drug resistance testing to guide ART selection and enhanced surveillance of subtype-specifc resistance patterns, to optimize HIV-1 control in EA. While resistance to newer drugs like INSTIs remains limited, its emergence signals the need for cautious use and sustained monitoring to preserve their efectiveness. However, given the limitations of public databases, particularly their incomplete representation of the broader HIV epidemic, future studies should incorporate nonpublic datasets and real-time surveillance protocol. Tis will improve our understanding of HIV-1 evolution and resistance trends and support the development of tailored interventions to control the epidemic more efectively in EA.
## 6. Limitations
A signifcant limitation of this study is its reliance on public databases, which capture only a fraction of available HIV-1 sequence data and are subjected to sampling biases. Additionally, the exclusion of nonpublic datasets, such as those from large-scale initiatives like the PANGEA program in Uganda, which have sequenced thousands of HIV-1 samples, limits the generalizability of our fndings. Globally, it is estimated that over 90% of HIV-1 sequences are not deposited in public repositories due to patient privacy concerns and ethical restrictions. Moreover, the available sequence data in public repositories is inherently biased. Many of the sequences were generated as part of research studies that often target specifc subpopulations, such as individuals experiencing ART failure or those participating in clinical trials. As a result, the dataset may be skewed toward viruses with DRMs, limiting the generalizability of the results to the broader HIV-positive population, and also the inability to retrieve detailed metadata on treatment history, which prevent diferentiation between transmitted and acquired DRMs which is an important distinction for interpreting resistance patterns.
## References
1. (2024) "Fact Sheet 2024-Latest Global and Regional HIV Statistics on the Status of the AIDS Epidemic"
2. Who (2024) "J0482-Who-Ias-Hiv-Statistics_Aw-1_Final_Ys"
3. (2024) "EASTERN AND"
4. De Leys, Vanderborght, Vanden Haesevelde (1990) "Isolation and Partial Characterization of an Unusual Human Immunodefciency Retrovirus From Two Persons of West-Central African Origin" *Journal of Virology*
5. Simon, Mauclère, Roques (1998) "Identifcation of a New Human Immunodefciency Virus Type 1 Distinct From Group M and Group O" *Nature Medicine*
6. Plantier, Leoz, Dickerson (2009) "A New Human Immunodefciency Virus Derived From Gorillas" *Nature Medicine*
7. Hemelaar (2012) "Te Origin and Diversity of the HIV-1 Pandemic" *Trends in Molecular Medicine*
8. Cho, Min, Son (2022) "Analysis of Evolutionary and Genetic Patterns in Structural Genes of Primate Lentiviruses" *Genes & Genomics*
9. Salminen, Ehrenberg, Mascola (2000) "Construction and Biological Characterization of Infectious Molecular Clones of HIV-1 Subtypes B and E (CRF01_AE) Generated by the Polymerase Chain Reaction" *Virology*
10. Yamaguchi, Vallari, Mcarthur (2020) "Brief Report: Complete Genome Sequence of CG-0018a-01 Establishes HIV-1 Subtype L" *JAIDS Journal of Acquired Immune Defciency Syndromes*
11. Olesen, Jespersen, Da Silva (2018) "HIV-2 Continues to Decrease, whereas HIV-1 Is Stabilizing in Guinea-Bissau" *AIDS*
12. Bbosa, Kaleebu, Ssemwanga (2019) "HIV Subtype Diversity Worldwide" *Current Opinion in HIV and AIDS*
13. Buonaguro, Tornesello, Buonaguro (2007) "Human Immunodefciency Virus Type 1 Subtype Distribution in the Worldwide Epidemic: Pathogenetic and Terapeutic Implications" *Journal of Virology*
14. Santoro, Perno (2013) "HIV-1 Genetic Variability and Clinical Implications" *ISRN Microbiology*
15. Mansky (1998) "Retrovirus Mutation Rates and Teir Role in Genetic Variation" *Journal of General Virology*
16. Troyano-Hernáez, Reinosa, Holguín (2022) "Genetic Diversity and Low Terapeutic Impact of Variant-specifc Markers in HIV-1 Pol Proteins" *Frontiers in Microbiology*
17. Giovanetti, Ciccozzi, Parolin et al. (2020) "Molecular Epidemiology of Hiv-1 in African Countries: A Comprehensive Overview" *Pathogens*
18. (2019) "HIV Drug Resistance Report"
19. Medicines (2025) "HIV Treatment"
20. Faraci, Park, Dubé et al. (2023) "Full-Spectrum HIV Drug Resistance Mutation Detection by High-Resolution Complete Pol Gene Sequencing" *Journal of Clinical Virology: Te Ofcial Publication of the Pan American Society for Clinical Virology*
21. Jourdain, Wagner, Ngo-Giang-Huong (2011) "Detection of HIV-1 DNA Resistance Mutations by a Sensitive Assay at Initiation of Antiretroviral Terapy Is Associated with Virologic Failure" *Clinical Infectious Diseases: An Offcial Publication of the Infectious Diseases Society of America*
22. Kattan (2012) "Bayesian Evidence Synthesis" *Encyclopedia of Medical Decision Making*
23. Hemelaar, Elangovan, Yun (2019) "Global and Regional Molecular Epidemiology of HIV-1, 1990-2015: a Systematic Review, Global Survey, and Trend Analysis"
24. Faria, Vidal, Lourenco (2019) "Distinct Rates and Patterns of Spread of the Major HIV-1 Subtypes in Central and East Africa" *PLoS Pathogens*
25. Lihana, Abimiku (2014) "Update on HIV-1 Diversity in Africa: A Decade in Review"
26. Huang, Hogan, Istrail et al. (2012) "Global Analysis of Sequence Diversity within HIV-1 Subtypes across Geographic Regions" *Future Virology*
27. Nkengasong, Adje-Toure, Weidle (2004) "HIV Antiretroviral Drug Resistance in Africa" *AIDS Reviews*
28. Ntamatungiro, Kagura, Weisser et al. (2022) "Pre-treatment HIV-1 Drug Resistance in Antiretroviral Terapy-Naive Adults in Eastern Africa: A Systematic Review and Meta-Analysis" *Journal of Antimicrobial Chemotherapy*
29. Kiros, Biset, Gebremariam et al. (2023) "Trends in HIV-1 Pretreatment Drug Resistance and HIV-1 Variant Dynamics Among Antiretroviral Terapy-Naive Ethiopians From 2003 to 2018: A Pooled Sequence Analysis" *Virology Journal*
30. Nchinda, Elangovan, Yun et al. (2023) "Global Associations of Key Populations With HIV-1 Recombinants: a Systematic Review, Global Survey, and Individual Participant Data Meta-Analysis" *Frontiers in Public Health*
31. Machila, Libonda, Habineza (2023) "Prevalence and Predictors of Virological Failure in Pediatric Patients on HAART in Sub-Saharan Africa: A Systematic Review and Meta-Analysis" *Pan African Medical Journal*
32. Niama, Vidal, Diop-Ndiaye (2017) "HIV-1 Genetic Diversity and Primary Drug Resistance Mutations before Large-Scale Access to Antiretroviral Terapy, Republic of Congo" *BMC Research Notes*
33. (2022) "Guidelines for the Use of Antiretroviral Agents in Adults and Adolescents with HIV How to Cite the Adult and Adolescent Guidelines: Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in Adults an"
34. Hudson, Mulenga, Westfall (2019) "Evolution of HIV-1 Drug Resistance after Virological Failure of First-Line Antiretroviral Terapy in Lusaka, Zambia" *Antiviral Terapy*
35. Omondi, Muigai, Ngayo et al. (2022) "Nevirapine Plasma Concentration Is Associated with Virologic Failure and the Emergence of Drug-Resistant Mutations Among HIV Patients in Kenya: A Cross Sectional Study" *Medicine (United States)*
36. Pujol-Hodge, Salazar-Gonzalez, Ssemwanga (2022) "Detection of HIV-1 Transmission Clusters from Dried Blood Spots within a Universal Test-And-Treat Trial in East Africa" *Viruses*
37. Wensing, Calvez, Ceccherini-Silberstein (2022) "2022 Update of the Drug Resistance Mutations in HIV-1" *Topics in Antiviral Medicine*
38. Ng'ong'a, Ayodo, Kawaka et al. (2022) "Antiretroviral (ARV) Drug Resistance and HIV-1 Subtypes Among Injecting Drug Users in the Coastal Region of Kenya" *Advances in Virology*
39. Nucleoside, Inhibitor, Inhibitor (2022) "Major HIV-1 Drug Resistance Mutations"
40. Bwire, Aiko, Mosha (2023) "High Viral Suppression and Detection of Dolutegravir-Resistance Associated Mutations in Treatment-Experienced Tanzanian Adults Living with HIV-1 in Dar Es Salaam" *Scientifc Reports*
41. Inzaule, Hamers, Noguera-Julian (2018) "Primary Resistance to Integrase Strand Transfer Inhibitors in Patients Infected with Diverse HIV-1 Subtypes in Subsaharan Africa" *Journal of Antimicrobial Chemotherapy*
42. Kiros, Tefera, Andualem (2023) "Low Level of HIV-1C Integrase Strand Transfer Inhibitor Resistance Mutations Among Recently Diagnosed ART-Naive Ethiopians" *Scientifc Reports*
43. Gartner, Roche, Churchill et al. (2020) "Understanding the Mechanisms Driving the Spread of Subtype C HIV-1" *EBioMedicine*
44. Peeters, Aghokeng, Delaporte (2010) "Genetic Diversity Among Human Immunodefciency Virus-1 Non-B Subtypes in Viral Load and Drug Resistance Assays" *Clinical Microbiology and Infection* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12625350&blobtype=pdf | # Molecular features of congenital cytomegalovirus infection in neonatal mouse brain at single-cell resolution
Acta Communications, Wen Zhou, Xuan Jiang, Simon Rayner, Han Cheng, Min-Hua Luo, Meng-Jie Mei, Yue-Peng Zhou, Yu-Ting Pan, Jin-Yan Sun, Wen-Bo Zeng, Tong Wu, Michael Mcvoy, William Britt, Bo Yang
## Abstract
Cytomegalovirus is the leading viral cause of congenital infection with neurological sequelae. Effective medical treatments are limited due to an inadequate understanding of the underlying pathogenesis. Here, we applied single-cell transcriptomics to analyze neonatal mouse brains with congenital cytomegalovirus infection (cCMV). We profiled cCMV in 22 cell types and identified neural progenitor cells (NPCs) and monocyte-derived macrophages (MDMs) as the most commonly infected cells. Infected NPCs exhibited dysregulated neurodevelopment-associated signaling pathways, correlating with viral transcript levels that indicate viral replication levels. Genes associated with phagocytosis and antigen presentation were downregulated exclusively in infected MDMs but remained largely unaffected in microglia and barrier-associated macrophages regardless of infection status. Analysis of intrinsic and induced interferon-stimulated gene expression revealed great heterogeneity across cell types but no direct correlation with cCMV susceptibility. Furthermore, our findings indicate that interferon type II is crucial for the control of cCMV and consequent cortical damage and calcification in the neonatal brain. This study advances our understanding of cCMV tropism and the molecular details of cCMV-induced neurodevelopmental impairment, cerebral immune response, and brain pathology.
## Introduction
Congenital cytomegalovirus infection (cCMV) is the most common congenital infection globally, affecting approximately 0.5-3% of live births [86]. Although the majority of newborns with cCMV are asymptomatic at birth, up to 18% may experience long-term neurodevelopmental sequelae, including neurodevelopmental delays, vision impairment, and most commonly, hearing loss [71]. Despite decades of research, effective preventive strategies and interventions remain limited, in part due to an incomplete understanding of the molecular mechanisms underlying fetal brain pathogenesis.
Neuropathological studies from human fetal brains have revealed that human cytomegalovirus (HCMV) preferentially infects neural progenitor cells (NPCs) [57,74]. NPCs are fundamental to neurodevelopment, as their proliferation contributes to brain volume, while their differentiation gives rise to neurons and glial cells. Due to the ethical and practical challenges associated with the use of human fetal samples, NPC in vitro models have been established to investigate the effects of HCMV infection on NPC fate [7,14,46,47]. These in vitro models have shown that HCMV actively manipulates the host cellular machinery, leading to premature and abnormal differentiation of NPCs. HCMV disrupts the Notch signaling pathway by dysregulating key components, including SOX2, Hes1, Notch1, and Jag1, and leads to loss of NPC characteristics and abnormal differentiation [27,40,42,81]. In addition, recent studies have revealed that HCMV-induced changes in connexin 43 and SOCS3 protein levels impair NPC migration [26,79].
Despite the valuable insights gained from NPC models, in vitro systems are incapable of recapitulating the complexity of in vivo infection. Hence, animal models are important for studying the pathogenesis of cCMV [53]. Due to the species-specific nature of HCMV, murine cytomegalovirus (MCMV) has been used to model HCMV infections. However, the three-layer structure of the mouse placenta prevents vertical transmission of MCMV [32]. Early studies attempted to simulate cCMV infection by injecting the virus into the placenta or fetal brain during embryonic development [76], but this approach faced challenges due to low survival rates and short lifespans, making long-term studies on neurodevelopmental damage difficult. To overcome these limitations, a mouse cCMV model was developed by intraperitoneally inoculating newborn mice with MCMV on postnatal day 0 (P0) with subsequent dissemination to the brain [33]. While this model mimics key clinical manifestations of human cCMV, the majority of brain damage in this model occurs in the cerebellum with little cortical impairment. This lack of cortical pathology is likely due to infection occurring at a later stage of brain development, which has been associated with less severe malformations of the brain following cCMV in humans [11]. We have recently established a cCMV mouse model by intracranial injection of MCMV at embryonic day 13.5 (E13.5), allowing infection at an earlier stage of brain development and more closely reflecting the timing of severe cCMV in humans [87]. This model achieves high rates of live birth with long survival periods, demonstrates cortical impairment, and recapitulates the clinical manifestations of cCMV in infants, including growth restriction, deficits in cognitive and learning-memory abilities, and hearing loss [87]. However, the detailed molecular mechanisms underlying these cCMV-induced neurodevelopmental defects remain unclear.
Single-cell technologies have emerged as powerful tools for delineating virus-host interactions and viral pathogenesis with unprecedented resolution. While pioneering studies using single-cell technologies have uncovered heterogeneity in herpesvirus infections and antiviral responses, these experiments were typically carried out in vitro with a homogeneous cell type [18,25,63,67,83]. Whether these findings accurately reflect how different cell types respond to infection of the developing brain and contribute to the pathogenesis of cCMV remain unresolved. In this study, we employed single-cell RNA sequencing (scRNA-seq) to characterize the molecular features of murine cCMV in neonatal brain at single-cell resolution. We profiled cCMV across 22 cell types and found that NPCs and monocyte-derived macrophages (MDMs) are the most susceptible cells to MCMV infection. Characterization of viral transcription in individual cells revealed distinct infection states, which were further examined for their effects on the fates of NPCs and brain macrophages. In addition, we detailed both intrinsic and induced expression of interferon-stimulated genes (ISGs) across all cell types to understand cell-specific antiviral defenses against cCMV. Our analyses revealed that the type II interferon (IFN) response, rather than the type I IFN response, plays a major role in controlling cCMV in the neonatal brain.
## Materials and methods
## Virus, mice, and congenital infection
An enhanced green fluorescent protein (eGFP) -tagged MCMV K181-eGFP was reconstituted by transfection of bacterial artificial chromosome DNA into NIH3T3 cells and harvesting as cell culture supernatants 10 days posttransfection as previously described [59]. Viral titers were determined in triplicate by plaque-forming assay using NIH3T3 cells as previously described [42]. Pathogenfree ICR mice were purchased from Beijing Vital River Laboratory Animal Technology. Eight-week-old ICR mice were mated and pregnant females were identified by the presence of vaginal plugs within 12 h for subsequent infection. Infection of fetuses and pups was performed as previously described [87]: fetuses were congenitally infected at E13.5 by intracranial inoculation of 100 pfu of K181-eGFP in utero. Neonatal pups received an intracranial inoculation of 500 pfu of K181-eGFP at P0. For the mock-infected controls, an equivalent volume of conditioned medium from uninfected NIH3T3 cultures was administered. All mice were bred and housed in an A2 animal laboratory with a 12 h light/12 h dark cycle.
## Tissue dissociation and single-cell preparation
Newborn P7 mice were anesthetized by placing them on ice and subsequently sacrificed via cardiac perfusion with cold Dulbecco's PBS. The cerebrums were carefully removed and transferred to pre-cooled HBSS (Gibco, Cat#14170-112) at 2-4 °C. Tissues were dissociated using the Neural Tissue Dissociation Kit (Miltenyi Biotec, Cat#130-092-628). After incubation at 37 °C for 15 min, enzymatic digestion was neutralized by addition of three volumes of HBSS supplemented with 1 mM sodium pyruvate (Gibco, Cat# 11360070), 0.01 M HEPES (Gibco, Cat# 15630-080), 100 U/ml penicillinstreptomycin (Gibco, Cat# 15140122), and 0.06% D-(+)glucose solution (Sigma, Cat# G8769). The dissociated cells were filtered through 70 μm strainers and cell debris was removed using the MACS Debris Removal Kit (Miltenyi Biotec, Cat# 130-109-398). The cell pellet was resuspended in red blood cell lysis solution (Miltenyi Biotec, Cat#130-094-183), incubated for 5 min on ice, then centrifuged at 300 x g for 5 min. The cell pellets were resuspended in HBSS and filtered through 40 μm strainers to ensure single-cell suspensions. Viability was assessed using trypan blue (Gibco, Cat# 15250061) and cell density was adjusted to 1000 cells/µl, either in staining buffer (BioLegend, Cat. #420201) for flow cytometry, or in PBS without calcium or magnesium for scRNA-seq.
## Flow cytometry
PBS-perfused mouse cerebra at P7 were digested with a Neural Tissue Dissociation Kit (Miltenyi Biotec, Cat#130-092-628), a single-cell suspension containing approximately 10,000 cells was stained with Fixable Viability Stain 700 (BD Biosciences, Cat# 564997) in 100 µl cell staining buffer (Absin, Cat#abs9475) for 5 min at room temperature to detect live cells. MG and MDMs were isolated by using a percoll density gradient protocol [51]. Cells collected at the 37%-70% density gradient interface were pelleted by centrifugation, then resuspended in cell staining buffer supplemented with Mouse FcR Blocking Reagent (STARTER, Cat# S0B0599) and incubated at room temperature for 15 min. Then, cells were stained with antibodies targeting mouse Ly-6G BV650 (BD Biosciences, Cat# 740554), CD45 APC-Cy7 (BD Biosciences, Cat# 557659), CD11b BV421 (BD Biosciences, Cat# 562605), and Fixable Viability Stain FVS510 (BD Biosciences, Cat# 564406) for 30 min at room temperature, followed by two washes with cell staining buffer. Finally, the stained cells were resuspended in 300 µl of cell staining buffer and applied to a BD LSRFortessa Flow Cytometer (BD Biosciences). The data were analyzed using FlowJo software (version 10.6.2).
## Single-cell sequencing
The BD Rhapsody system [69] was used to capture single-cell transcripts. Single-cell suspensions containing approximately 20,000 cells were loaded on arrays of over 200,000 microwells using a limiting dilution approach. Barcoded oligonucleotide-conjugated beads with hybridized polyadenylated RNA were retrieved and pooled for reverse transcription. Whole transcriptome libraries were prepared from the captured cells, followed by second-strand cDNA synthesis, adaptor ligation, and universal amplification through 22 cycles of PCR. Sequencing libraries were generated using random priming PCR to enrich for the 3′ ends of transcripts linked to the barcodes and unique molecular indices (UMI). Quality control was conducted with the Agilent DNA 2100 Tapestation assay, and the libraries were sequenced on a HiSeq4000 (Illumina) using 150 × 2 chemistry.
## scRNA-seq read mapping, data quality control, dimensionality reduction, clustering, and visualization
The raw sequencing data was processed with the BD WTA Rhapsody Analysis Pipeline Version 1.9.1. This analysis involved the joint alignment of scRNA-seq reads for both host and viral genomes, followed by transcript quantification, using modifications as previously described [22,72]. A joint reference genome was constructed that included the Mouse Genome assembly GRCm38 (mm10) and a modified MCMV strain K181 genome (NCBI: Taxonomy ID: 69156) [65]. Single-cell reads were aligned to this joint reference genome using STAR [17].
Downstream analyses, quality control (QC), normalization, integration, dimension reduction, clustering, and visualization of the scRNA-seq data were conducted using Seurat v5.1. Only host gene count matrices were used for QC, normalization, and integration. QC metrics for nFeature, nCount, and percentage of mitochondria genes were visualized by violin plots to filter out cells with low-quality reads. Remaining cells were merged and normalized using SCTransform (SCT) [24] followed by principal component analysis (PCA). The Harmony integration algorithm [34] was used to integrate multiple scRNA-seq datasets from different samples into a shared space for unsupervised clustering and to minimize bias from dataset-specific conditions. Based on elbow method in the "FindNeighbors" function, the first 20 principal components were used to compute the shared nearest neighbor graph. Cell clustering and sub-clustering analyses were conducted using the "FindClusters" function (Louvain algorithm), with appropriate resolution (0.14) determined by the "Clustree" function [8]. Uniform manifold approximation and projection (UMAP) function was employed with Harmony to visualize cell clusters. To identify cluster-specific marker genes, differential expression gene (DEG) analysis was performed using the "FindAllMarkers" function, comparing each cluster to all other clusters with parameters set for the MAST test.
In the re-clustering analysis for lymphocytes, NPCs, and MDMs, the RNA matrices were normalized and scaled using the "NormalizeData" and "ScaleData" function in the Seurat R package, and highly variable genes were identified for subsequent PCA using the "FindVari-ableGenes" and "RunPCA" functions in Seurat R package.
## Cell type annotation
To annotate cells, we first curated a predefined set of marker genes for cell types from PanglaoDB [20] and CellMarker [85] databases and relevant literature [45,60,84]. To increase the annotation accuracy, five automatic annotation methods (SCSA, SCINA, AUCell, scmap-cell, and scmap-cluster) were employed to generate preliminary annotation results (Supplementary data S3). Most cells received consistent annotations across the different methods, allowing for consensus-based annotation. Cells with discrepancies in their annotations were manually annotated based on their cluster-specific marker genes. Specifically, microglia (MG), MDMs, and barrier-associated macrophages (BAMs) were annotated based on established markers for brain macrophages [55]. NPCs and neuroblasts were distinguished by their expression levels of Ascl1 and Mki67 mRNAs.
## Viral transcription analysis
Viral gene count matrices for each sample were extracted from the joint-reference mapping results after normalization using the SCT algorithm. False positive mappings of reads to the MCMV genome in mock-samples occurred due to sequence similarities between the host and viral genomes and were considered mapping noise. These false mappings were removed through a two-step process: First, viral genes with expression values greater than zero were identified and these background values were extracted from all mock-samples. Second, the maximum background expression values of these genes in mock-samples were subtracted from their normalized expression in infected samples, with any resulting negative values corrected to zero.
To analyze viral and host genes at the same normalized level, we added an assay called "advSCT" to the Seurat object. This assay included both the corrected SCT-normalized viral gene count matrix and the host gene count matrix. Expression levels of each gene were represented by SCT-normalized UMI counts. The percentage of viral reads was calculated as the proportion of viral reads to total reads in each cell. A cell was classified as "infected" if its viral read percentage was greater than zero. The bimodal distribution of log10 normalized viral read percentages in infected cells was modeled using the "normalmixEM" function from the "mixtools" package in R. The cutoff for distinguishing infection levels was calculated using the formula:
, where µ1 and µ2 represent the means of the two modes (peaks) in the bimodal distribution and sd is the standard deviation. The criteria for different infection statuses were defined as follows: (1) "high viral transcript (HVT) cells" with the viral read percentage above the cutoff; (2) "low viral transcript (LVT) cells" with the percentage below the cutoff; (3) "bystander cells" from infected brains without detectable viral mRNA; and (4) "mock cells" from mockinfected brains.
The CMV temporal classes for immediate early (IE), early (E), and late (L) gene sets were derived from the temporal classification of MCMV genes established in previous literature [16,39,43,50]. The relative gene expression levels of a single viral gene or a class of viral genes was calculated as the percentage of reads corresponding to that gene or class out of the total viral reads in each cell.
$$(µ 1+3sd)+(µ 2-3sd) 2$$
## DEG, ISG, and enrichment analysis
DEGs in different infection status were identified using the "FindMarkers" function with the Wilcoxon method and default parameters in the Seurat package. The list of ISG genes was derived from previous research [81]. Intrinsic ISG levels were calculated as the percentage of ISG reads relative to the total reads in each mock cell. Induced ISG expression in cells from infected brains was determined by dividing the average expression of all ISGs in that cell by the average expression of these genes in cells from mock-infected brains.
The R package ClusterProfiler (version 4.12.6) was used for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses of the DEGs among subclusters. All gene sets were derived from the Molecular Signatures Database [41]. Enrichment Analysis (GSEA) was performed using the fgsea package (v1.30.0) in R on all genes ranked by log2-fold change (log2FC) from the DEG analyses. The parameters used were 1,000,000 permutations, a minimal set size of 15, and a maximal set size of 500. KEGG and HALLMARK gene sets were downloaded from h t t p : / / s o f t w a r e . b r o a d i n s t i t u t e . o r g / g s e a / m s i g d b /. Gene sets with false discovery rate (FDR) < 0.05 were considered significant.
## Expression scoring of gene sets
Expression scores for particular gene sets were determined by calculating the percentage of reads associated with that gene set relative to all reads in each cell, using the "PercentFeatureSet" function in the Seurat package. Gene sets related to NPC proliferation (GO: 0061351), differentiation (GO: 0048863), or development (GO: 0048864) were derived from the MGI database [4]. Type I and type II IFN-specific responses were defined based on the type-specific genes associated with IFNα (GO: 0085134), IFNβ (GO: 0035456), or IFNγ (GO: 0034341) from the same database. The list of ISGs was derived from previous research [81]. Cell cycle scores were calculated directly using the "CellCycleScoring" function in the Seurat package. The expression score of each gene across different cell types or infection statuses was determined by averaging the read counts across the respective groups.
## Quantitative reverse transcriptase PCR (RT-qPCR)
Total RNA was extracted from brain tissues using the RNAiso Plus reagent (Takara). PrimeScript RT Reagent Kit with gDNA Eraser was used for genome digestion and reverse transcription following the manufacturer's instructions (Takara). qPCR was performed in the CFX-96 Connect system (Bio-Rad) with the iQ SYBR Green Supermix kit (Bio-Rad) to quantify the expression levels of Ifna, Ifnb, Ifng, and Ie1 mRNAs relative to GAPDH mRNA using the 2 ΔΔCT method. The primers used are listed in Supplementary data S4. Samples from three independent experiments were tested in triplicate.
## Quantification of IFNs
Cerebral samples were prepared from PBS-perfused brains, minced, supplemented with protease inhibitors, and frozen as previously described [28]. Frozen cerebrum samples were thawed in 800 µl PBS and homogenized using a NOVAprep DS1000 tissue cell splitter. An equal volume of 2x RIPA cell lysis buffer containing 2x the standard dose of protease inhibitor cocktail (Roche, Cat #04693132001) was added to 200 µl of homogenate. Samples were then centrifuged at 4500 × g for 15 min at 4 °C and supernatants were collected for enzyme-linked immunosorbent assay (ELISA) to determine the concentrations of IFNγ (ABclonal, Cat #RK00019) or IFNβ (ABclonal, Cat #RK00420) in brain tissues per the manufacturer's instructions.
## Interferon blocking in neonatal mice
Newborn mice were infected intracranially at P0 with 500 pfu MCMV-K181-eGFP. Treatment with anti-mouse IFNγ (BioXCell, Cat#BE0055) or anti-IFNβ antibody (NeutraKine, Cat#69013-1-Ig) was administered via lateral ventricle injection at P1 and P3. Each injection consisted of 1 µL of the antibody at a dose of 0.2 µg, prepared in sterile PBS. Mouse IgG1 isotype (BioXCell, Cat#BE0083) was used as a negative control.
## Quantification of fluorescence intensity and calcification area
Whole brains were collected and fixed in 4% paraformaldehyde (PFA) overnight and eGFP fluorescence was imaged using a CRI Maestro imaging system. Fluorescence was quantified using the Analyze RGB function in ImageJ. Calcification in the brain at P14 was determined using HE and von Kossa staining as described previously [87]. Calcified and total tissue areas were measured using ImageJ. The percentage of calcification was calculated as the area of calcified tissue divided by the total tissue area, multiplied by 100.
## Immunofluorescence analysis
PBS-perfused brains were fixed in 4% paraformaldehyde, and coronal sections of 30 μm thickness were prepared for immunofluorescence analysis. Sections were stained with mouse monoclonal anti-Iba1 antibody (Abcam, Cat# ab178847), followed by incubation with Alexa Fluor 568 donkey anti-rabbit IgG (H + L) secondary antibody (Invitrogen, Cat# A-10042). Nuclei were counterstained using DAPI (Life Technologies).
## Statistics and correlation analysis
Pearson correlations between the percentages of viral reads and the expression levels of cellular genes or gene sets in HVT cells were computed and visualized using the ggplot package with the stat_cor function from the ggpubr R package. The default Pearson method was employed to calculate and label the correlations along with their corresponding p-values. P-value < 0.05 was considered significant. Additionally, Pearson correlations between the percentages of viral reads and intrinsic ISG levels or IE read percentages across cell types were calculated using the base cor function in R.
Data were from at least 3 independent experiments and presented as means ± SD/SEM as stated in the figure legends. GraphPad Prism (version 8.0.2) was used to perform statistical analyses, including unpaired two-tailed Student's t-test, one-way ANOVA, and the Mantel-Cox test.
## Results
## scRNA-seq uncovers cell composition changes induced by cCMV
In previous studies we injected MCMV K181-eGFP into the lateral ventricle of fetal mice at E13.5. We determined that P7 represents a critical time point in the brain's response to cCMV, characterized by peak levels of viral load, inflammatory cytokines, and leukocyte infiltration in the brain [87]. To further explore how different types of cells in neonatal brain interact with MCMV at this pivotal stage, we performed scRNA-seq analysis. We collected brain tissue samples from both mock-and MCMV-inoculated mice at P7 (Fig. 1a). The fluorescent imaging of mouse brains confirmed the presence of active MCMV infection (Fig. 1b) and flow cytometry determined that approximately 5% of brain cells were eGFP-positive (Supplementary Fig. 1a). After assessing cell viability, cDNA quality, and sequencing data quality control, we captured approximately 10,000 high-quality cells from each sample for scRNA-seq analysis (Supplementary Fig. 1b and Supplementary data S2). We employed UMAP for non-linear dimensionality reduction to visualize the data. This analysis revealed 22 distinct cell clusters through unsupervised clustering (Fig. 1c, Supplementary Fig. 1c). Due to the lack of reference cell markers for the developmental stage at P7 mouse brain, we utilized published datasets from adult mice (PanglaoDB [20] and CellMarker [85]) and relevant literature on P0 mouse brains [45,60,84] to identify specific marker genes for cell type annotation (Fig. 1d and Supplementary Fig. 1d-e). The annotated cell types were categorized into three major groups: neural cells, barrier cells, and immune cells (Fig. 1e).
Further analysis of cell composition revealed a significant increase in the proportion of immune cells, particularly MDMs, neutrophils, and lymphocytes, in the cCMV-brains compared to mock-brains. In addition, the population of brain macrophages, including MG and BAMs, also expanded in cCMV-brains. In contrast, many neural cell types, such as NPCs, oligodendrocyte precursor cells (OPCs), and oligodendrocytes (OLCs), along with barrier cell types like endothelial cells, meningeal cells, and pericytes, exhibited a reduction in composition during cCMV (Fig. 1f-g and Supplementary Fig. 1f ).
Overall, this scRNA-seq analysis indicates that cCMV induces substantial changes in the cellular composition of the developing mouse brain, characterized by immune cell infiltration and a reduction in neural and barrier cells.
## Characterization of viral transcript heterogeneity in cCMV brain cells
In vitro single-cell studies have demonstrated significant variability in viral gene expression among infected cells [18,19,25,67]. Here, we explored this heterogeneity during infection in vivo. In the cCMV-brain samples, both infected and bystander (non-infected) cells exhibited overlapping distributions among cell clusters, with no clear separation between them (Supplementary Fig. 2a). However, cells with high percentages of viral reads were concentrated within specific cell clusters (Fig. 2a). We further quantified viral transcription in the infected cells. The viral reads exhibited a bimodal distribution, in which most cells contained very low percentages of viral reads (median 0.03%) and a separate subset harbored significantly high percentages of viral reads (median 8.5%) (Fig. 2b). This pattern suggests the presence of two distinct infection states in the infected brain cells, which are designated as HVT and LVT states, respectively (Supplementary Fig. 2b).
Overall, our single-cell analysis showed that at P7, when the viral replication was at its peak, 43.8% of cells in the infected brain samples were bystander, 49.9% were in the LVT state, and only 6.3% were in the HVT state (Fig. 2c). The proportion of HVT cells is close to the eGFPpositive ratio observed in flow cytometry analysis, suggesting that only the HVT cells were supporting active viral infection. In most cell types, the HVT state was less than 5%. However, NPCs were particularly vulnerable, with up to 36.7% of NPCs being in the HVT state. MDMs and neutrophils also contained significant proportions of HVT cells, at 16.0% and 5.7%, respectively (Fig. 2d). Due to the relatively large numbers of infiltrated MDMs and activated MG, along with reduced NPCs and the scarce presence of neutrophils in the infected brain samples, MDMs accounted for the highest proportion of HVT cells (43.1%), followed by NPCs (28.7%) and MG (12.1%) (Fig. 2e). NPCs are well-established targets of cCMV in the developing brain [12]. To determine whether myeloid cells are also major targets, we performed immunohistochemistry with the Iba1 antibody on the infected P7 brains. We found that 18.5% of MCMV-infected cells were Iba1-positive (Supplementary Fig. 2c). Since both MG and MDMs express Iba1, we isolated immune cells from P7 brains of E13.5 MCMV-eGFP-infected mice using flow cytometry. Our results showed that 8.7% of MDMs (CD11b + /CD45 high ) and 1.6% of MG (CD11b + / CD45 med ) were eGFP-positive (Supplementary Fig. 2de). These findings are consistent with our scRNA-seq data, which show 16% of MDMs and 3% of MG are in HVT state.
## CMV transcription features of NPCs and MDMs in HVT state
We analyzed CMV transcription patterns in LVT and HVT cells. CMV genes were categorized into three temporal classes: IE, E, and L (Supplementary data S2) [16,39,43,50]. These classes show a stochastic distribution in the LVT state but an ordered pattern in the HVT state (Supplementary Fig. 3a), indicating that active viral replication occurred only in HVT cells. Given that NPCs and MDMs are the major HVT cells, viral transcription in these two cell types were further dissected. In the HVT state, NPCs exhibited a more heterogeneous distribution of viral transcription levels than MDMs. Two distinct NPC subpopulations with distinct viral gene expression patterns were identified and designated as NPC subgroup 1 and 2 (Fig. 3a and Supplementary Fig. 3b). NPC subgroup1 harbored significant higher percentage of viral reads than NPC subgroup 2, which had slightly lower percentage of viral reads than MDMs (Fig. 3b). Analysis of the temporal class composition revealed that both NPC subgroups maintained a similar ratio of IE genes relative to total viral transcripts, which was slightly higher than that in MDMs. In addition, NPC subgroup 1 had a much lower ratio of E genes and a significantly higher ratio of L genes compared to NPC subgroup 2 and MDMs, suggesting that NPC subgroup 1 more efficiently supported productive viral replication than the other two groups (Fig. 3b).
We further examined the relative expression of individual viral genes compared to total viral transcripts in these HVT cells. The proportions of IE genes M122/123 were similar between the two NPC subgroups and higher than in MDMs (Fig. 3c). In contrast, the other two IE genes, M126 and the newly identified M166.5 [43], were expressed at much lower levels relative to other viral genes in all three HVT cell types (Fig. 3c). L genes were more actively expressed than E genes in NPC subgroup 1, whereas NPC subgroup 2 and MDMs exhibited a similar expression pattern of E and L genes (Fig. 3d-e). We also noticed that certain viral genes exhibited a preference for expression in specific cell types. For instance, the expression of M04, M147.5 and M36 were significantly elevated in MDMs compared to NPCs (Fig. 3f ), whereas M80, M94, M98, and M99 showed greater expression in NPCs than in MDMs (Fig. 3g). These findings suggest that CMV adapts to the unique cellular environments of specific cell types and employs distinct strategies to facilitate viral replication.
## Infection status influences cCMV's impact on NPC fate
Previous in vitro studies have shown that CMV infection disrupts the normal proliferation and differentiation of NPCs [42,46]. In this study, our single-cell analyses of cCMV-brain tissue provided a unique opportunity to assess the functional impacts of cCMV on NPCs in vivo. We performed further re-clustering analysis on NPCs to generate five distinct subclusters (Fig. 4a). The distribution of NPCs across these subclusters was closely linked to their infection status. Cells from mock-infected brains were predominantly located in Clusters 0 and 3, while cells from infected brains included bystander and LVT cells in Cluster 1, HVT cells primarily in Cluster 2, and a mixture of bystander, LVT, and HVT cells in Cluster 4 (Fig. 4b and Supplementary Fig. 4a).
Analysis of gene signatures across these clusters revealed distinct functional characteristics (Supplementary Fig. 4b). Signature genes of Cluster 0 were enriched to GO terms related to neurodevelopmental processes, consistent with undisrupted developmental functions in mock-infected NPCs. In addition, a small subset of mock-infected NPCs in Cluster 3 are involved in the regulation of neurogenesis and synapse assembly. Cluster 1 displayed gene signatures of an activated antiviral defense system, which likely accounts for the non-infected and LVT states observed in these cells. Furthermore, the gene signatures and GO terms in Cluster 2 indicated the activation of protein folding and endoplasmic reticulum (ER) stress coping mechanisms, suggesting that these responses facilitate viral production in HVT cells. The GO terms of Cluster 4 are similar to those of Cluster 1 in defense against viral infection and response to type I and II IFNs, but Cluster 4 has more relevant genes involved and lower p-values, likely reflecting a stronger antiviral state.
We calculated gene expression scores for NPC proliferation, differentiation, and development based on known related genes [3] and found they are inversely correlated with viral transcription levels (Fig. 4c). This suggests that active viral replication impairs both neurodevelopmental and proliferative processes in NPCs. In line with these findings, GO analysis of down-regulated genes in the HVT state versus LVT and non-infected states revealed a significant downregulation of biological processes (BPs) related to NPC proliferation, differentiation, and neurodevelopment (Fig. 4d). We also conducted pathway analysis using the KEGG and Hallmark databases through Gene Set GSEA. Several critical pathways involved in neurodevelopment-such as the Notch, Wnt, and Hedgehog pathways-were strongly downregulated exclusively in the HVT state, but not in the non-infected or LVT states (Fig. 4e and Supplementary Fig. 4c). Additionally, the expression of the most affected genes (Notch1 and Ccnd2 in Notch, Ctnnb1 and Apc in Wnt, Ptch1 and Gli3 in Hedgehog) exhibited negative correlations with viral read levels in individual NPCs (Supplementary Fig. 4d). These findings demonstrate that cCMV directly interferes neurodevelopment programs in NPCs, particularly in the HVT state.
On the other hand, antiviral-related programs were activated in both LVT and bystander cells, likely a response to counteract viral infection (Fig. 4f ). Key upregulated ISGs include Irgm1/2, Igtp, Tgtp1, Irf1, and genes encoding guanylate-binding proteins (Gbps). In contrast, these ISGs were not dramatically induced in the HVT state, where genes related to translational machinery were upregulated to enhance viral production (Fig. 4d).
CMV infection has been shown to disrupt the cell cycle of NPCs in vitro [46]. We extended this analysis to our in vivo setting during natural infection. NPCs were categorized into G1, S, or G2/M phases based on the expression of cell cycle marker genes [36]. Bystander cells, although uninfected, showed a slight reduction in the S and G2/M phases and a prolonged G1 phase compared to mock NPCs. LVT cells exhibited a similar pattern to bystander cells. However, in the HVT state, the G2/M phase was almost absent, with a large proportion of cells arrested in G1, while the proportion in the S phase remained comparable to the mock (Fig. 4g).
## cCMV disrupts macrophage functions
The brain is a immune-privileged organ with specialized anatomical structures for various types of macrophages. MG are yolk sac-derived resident macrophages located within the brain parenchyma, whereas BAMs reside at key barrier regions such as the meninges, perivascular spaces, and choroid plexus [31]. These macrophages are crucial in safeguarding the central nervous system (CNS) against invading pathogens and injury. However, a large number of MDMs were recruited to the infection site and became primary viral targets, indicating a failure of both resident macrophage types to control cCMV.
Further cluster analysis of MDMs identified five distinct clusters (Fig. 5a-b and Supplementary Fig. 5). In mock-infected brains, macrophages were rare and exhibited gene signatures characteristic of Ly6C low /Cx3CR1 high cells (Cluster 5) and proliferation-related GO terms (Cluster 4), suggesting they represent patrolling monocytes involved in brain surveillance. In infected brains, migrating monocytes from peripheral circulation were primarily in Cluster 3, marked by high Ccr2 expression and GO terms related to monocyte migration. These monocytes differentiate into distinct macrophage subtypes in the infected brains. Cluster 1 is characterized by a gene signature that includes Mertk, Adgre1, Cd74, and H2-Aa, indicating activated functions in phagocytosis and antigen presentation. In contrast, Cluster 0 exhibits GO terms of the toll-like receptor signaling pathway and response to IFNs and high expression of Nos2, Cxcl16, Il1a, and Il1rn. These features suggest a pro-inflammatory macrophage phenotype characterized by nitric oxide synthesis and cytokine production. LVT and bystander MDMs are indistinctly distributed across these clusters, suggesting similar impacts on macrophage functions between these two infection states. HVT MDMs are concentrated in Cluster 2, which has GO terms related to autophagy and protein translation, suggesting that the virus hijacked these host pathways for virion production [73]. Consistent with the cluster analysis, a comparative examination of the transcriptomes of HVT versus bystander MDMs reveals increased translational activity and ER stress in the HVT state, while traditional macrophage functions such as antiviral response, antigen presentation, and phagocytosis were suppressed in HVT MDMs (Fig. 5c).
We further examined the effects of infection on macrophage identity and function in MDMs, MG, and BAMs. The gene signatures associated with these three macrophage types were downregulated in infected brains (Fig. 5d). In MG and BAMs, the changes were consistent across the three infection states; however, MDMs showed more pronounced downregulation in the HVT state compared with the other two states. All three macrophage types exhibited enhanced expression of genes related to phagocytosis in infected brains (Fig. 5e). MG and BAMs had higher induction of these genes than MDMs and showed similar or slightly increased gene expression in the HVT state relative to the other two states. In contrast, the induction of these phagocytic genes was suppressed in HVT MDMs. Similar patterns were also observed in the expression of genes related to antigen presentation (Fig. 5f ) and major histocompatibility complex (MHC) class I and II molecules (Fig. 5g). In addition, cytokine expression in these macrophage types was analyzed (Fig. 5h). In general, BAMs and MG had a higher cytokine expression than MDMs in the mock state. While cytokine production remained relatively stable in BAMs and MG across different infection states, it was dramatically downregulated in MDMs in the HVT state. However, Ccl5 expression was markedly elevated in MDMs compared to the other two macrophage types.
## Heterogeneity of ISG responses across cell types in cCMVbrain
Next, we delineated the antiviral characteristics of different cell types in the context of cCMV infection. The developing brain is comprised of both stem cells and differentiated cells that employ distinct mechanisms to combat viral infections [82]. Differentiated cells primarily rely on the IFN-associated response to initiate antiviral programs; in contrast, stem cells exhibit reduced sensitivity to IFN but maintain high levels of intrinsic ISGs to restrict viral infection [82]. Our in vivo data revealed that both NPCs with stem cell properties and differentiated cells, such as macrophages, are highly susceptible to cCMV. To explore the mechanisms underlying this susceptibility, we profiled ISG expression across all cell types in various infection states.
Contrary to the high intrinsic ISG expression observed in embryonic stem cells (ESCs) and neural stem cells (NSCs) in vitro [81], profiling of intrinsic ISG expression in mock-infected brain samples revealed that NPCs expressed relatively lower levels of intrinsic ISG expression compared to other cell types (Fig. 6a). Differentiation into neuroblasts, immature neurons, and OPCs had negligible impact on ISG expression; however, differentiation into OLCs, astrocytes, and Schwann cells increased ISG levels (Supplementary Fig. 6a). Overall, neural cells displayed lower intrinsic ISG levels than other cell types, with immune cells exhibiting the highest ISG levels, indicating differential intrinsic ISG expression (Fig. 6a). In line with this finding, most experimentally validated CMV restriction factors [64], except Atrx and Bclaf1, had lower expression in neural cells than in immune cells (Fig. 6b). We additionally examined several intrinsic ISGs that have been reported to be highly expressed exclusively in ESCs and NSCs [81]. However, most of these genes showed comparable expression levels between NPCs and the differentiated cells (Supplementary Fig. 6b). We further conducted correlation analysis and found no significant correlation between intrinsic ISG levels and the susceptibility of these cell types to cCMV in vivo (Fig. 6c). We then analyzed induced ISG expression across all cell types from the infected brain samples. cCMV triggered a robust induction of ISGs in all cell types, with the strongest response observed in immune cells and the weakest in neural cells (Fig. 6d). Immune cells and barrier cells induced more ISGs with greater expression changes (log2FC > 0.5 and FDR < 0.01) than neural cells (Supplementary Fig. 6c). We also observed that several ISGs were more upregulated in immune cells than in other cell types (Supplementary Fig. 6d), and three ISGs (Ifitm1, Tnfsf10, and Txnip) were specifically induced in barrier cells (Supplementary Fig. 6e). However, no ISG was identified that was specifically induced in neural cells. A core set of ISGs was significantly induced across nearly all cell types. Among them, the most highly induced ISGs were Cxcl10, Gbp2, and Gbp4 from the immunityrelated GTPase (IRG) family; ISGs associated with antigen presentation, such as B2m, Tap1, Tap2, and Psmb8, were also prominently expressed. These genes are typically activated by IFNγ. In addition, many antiviral ISGs, including Ifitm3, Ifit1/2/3, Rsad2, Bst2, and Samhd1, along with regulators of the IFN pathway such as Stat1, Stat2, Stat3, Irf1, and Irf9, were induced across various cell types, albeit to a lesser extent (Fig. 6e). ISG induction was weaker in neuroblasts and immature neurons than in NPCs, which had similar numbers of highly-induced ISGs (log2FC > 0.5, FDR < 0.01) to OPCs, OLCs, and astrocytes, but much fewer than barrier or immune cells (Supplementary Fig. 6f ).
The patterns of ISG induction varied based on the infection state. Compared to bystander cells, the ISG induction was slightly elevated in MG, MDMs, BAMs, astrocytes, OPCs, and meningeal cells in the LVT state; in contrast, ISG induction was suppressed in the HVT state, specifically in MDMs, NPCs, MG, and neutrophils (Mann-Whitney U test, p < 0.001) (Fig. 6f ). Consistent with the susceptibility of NPCs and MDMs to CMV infection, a profound inverse correlation was found between ISG induction and viral transcription levels in HVT cells (Fig. 6g). However, no significant correlation was observed in neutrophils, despite their ranking as the third cell type in terms of proportions in the HVT state (Fig. 2d).
Taken together, our data suggest that the relationship between ISG expression and viral susceptibility in vivo is more complex than previously thought. Both low intrinsic ISG level and an attenuated induced ISG response may contribute to cell vulnerability to cCMV in a celltype-dependent manner.
## IFNγ response limits cCMV in the developing brain
The ISG expression analysis showed that the most highly induced ISGs are typically activated by IFNγ in the type II IFN response. To explore the roles of type I and type II IFN responses to cCMV, we visualized cellular responses with expression scores for genes specific to either pathway, considering that many ISGs are shared between them. Both IFN responses were moderately activated in LVT and bystander NPCs (Fig. 7a). In MDMs, IFN responses were strongly activated in LVT and bystander cells but not in HVT cells (Fig. 7b), suggesting effective viral countermeasures in these cells. Notably, the type II IFN response appeared to exhibit broader and more intense activation than type I. Consistent with this observation, negligible expression of type I IFN genes (Ifna and Ifnb) was detected across all cells regardless of their infection state (Supplementary Fig. 7a-c), while type II IFN gene (Ifng) expression was exclusively detected in natural killer (NK) and T cells (Fig. 7c). These results indicate that the type I IFN response is muted during early brain development, whereas the type II IFN response is introduced when NK and T cells are recruited to the infection sites.
Consistent with the single-cell analyses, at both the transcriptional (tested by RT-qPCR) and protein levels (determined by ELISA), IFNα was barely expressed, while IFNγ was induced to a higher level than IFNβ in infected vs. mock-infected brains at P7 (Figs. 7d-e). To investigate the role of IFNγ in controlling cCMV, we inoculated 500 pfu of K181-eGFP into the lateral ventricle of newborn mice at P0. Antibodies to IFNβ, IFNγ, or an IgG isotype control were administrated intracranially at P1 and P3. Viral infection in the brains was assessed by eGFP fluorescence intensity at P7 and P14, and brain pathology was examined by HE staining with calcification evaluation at P14. Antibody to IFNβ had no effect on viral spread in the brain compared to the control, whereas brains treated with antibody to IFNγ had higher viral loads and exhibited more extensive calcification at P14 (Figs. 7f-g). In addition, IFNγ blockade exacerbated cCMV-induced ventricle enlargement, cortical thinning, cell loss, and inflammatory infiltration (Fig. 7h). Detailed analysis by von Kossa staining reveals that brain calcification areas correspond to regions of cortical cell loss and were significantly expanded by IFNγ blockade (Fig. 7i). Taken together, these findings suggest that type II, but not type I IFNs, is critical for controlling cCMV in developing mouse brains.
## Discussion
Building on our recently established mouse cCMV model, these scRNA-seq analyses provide the first characterization of cellular responses to cCMV in the neonatal brain at single-cell resolution. The analyses reveal changes in cell composition, identify CMV tropisms in the developing brain, demonstrate the functional impact of CMV on susceptible cells, uncover heterogeneous antiviral responses across multiple cell types, and reveal a crucial role for IFNγ in limiting viral spread in the neonatal brain. These findings provide new insights into the complex interactions between CMV and the developing brain.
CMV has a broad cell tropism in vivo, yet previous scRNA-seq experiments were conducted with limited cell types in an in vitro setting [13,19,25,43,67]. This work represents the first single-cell study that examines hostvirus interactions across various cell types during cCMV in the fetal and neonatal brain. Our data indicate that NPC is the most susceptible cell type that supports high levels of viral transcription. This finding is consistent with postmortem observations in human fetal brains [57,74] and reinforces our cCMV model [87] as a valid representation of cCMV in humans. In contrast, a recent alternative congenital MCMV infection model (intraperitoneally inoculating at P0) identified astrocytes as the initial targets at P7 and neurons as the major targets of CMV at the later infection stage of P17 [37]. This study relied on the GFAP-Cre mouse to label astrocytes, which is not a reliable method to distinguish astrocytes in the developing brain, because the GFAP gene is expressed not only in mature astrocytes but also in NPCs and nascent neurons [23]. In addition, immunohistochemical analyses of human fetal brains do not support neurons as a significant permissive cell type [57,74].
In line with in vitro scRNA-seq findings, cCMV in vivo also exhibits substantial heterogeneity. First, viral transcript levels exhibit a lognormal distribution, indicating a heterogeneous infection state within cell populations. Second, most cell types harbor only very low levels of viral transcripts without a typical ordered expression of CMV temporal genes. These LVT cells exhibit similar antiviral responses and cellular functions to those observed in the bystander state, suggestive of abortive infections. Third, heterogeneity may also arise from asynchronous infections, as observed in NPCs, likely reflecting different waves of infection in vivo.
IE genes play a crucial role in establishing lytic infection [13,61,67]. However, our in vivo infection data do not show a good correlation between IE gene expression and viral transcript levels across various cell types (Supplementary Fig. 6g). This discrepancy may arise from asynchronous infections in vivo in contrast to more synchronous infections in vitro. Further investigation is needed to clarify this relationship. We also observed cell type-specific expression patterns of viral genes. M04, M147.5, and M36 were expressed at higher levels in MDMs than in NPCs. M04 and M147.5 are known to interfere with antigen presentation [30,44], while M36 encodes a viral inhibitor of caspase 8-induced apoptosis to counteract host defense mechanisms [15]. Thus, high expression of these viral genes in MDMs enhances macrophage viability but diminishes their capacity to present antigens to T cells effectively. In contrast, M80, M94, and M99 expression was specifically enhanced in NPCs. M94 interacts with M99 and plays a key role in secondary envelopment of MCMV [49], while M80, homolog of HCMV UL80, is involved in viral capsid assembly [54]. The elevated expression levels of these viral genes and the upregulated host translation machinery suggest that NPCs are the primary cell type for virion production during cCMV infection.
The Notch, Wnt, and Hedgehog pathways are interconnected to regulate NPC fate decisions during neurodevelopment [27]. Our data confirm the downregulation of Notch1 and Jag1, reported in the NPC model [40], and identify new downregulated genes in the Notch pathway, including Hes5, Ccnd2, and Ncor2. As a major downstream target of Notch signaling, Hes5 preserves the undifferentiated state of NSCs and NPCs and regulates the timing of neurogenesis and gliogenesis [5]. Interfering with Hes5 expression may lead to abnormal NPC differentiation, as evidenced by decreases in the numbers of NPC-derived cells, particularly OPCs and OLCs. Cyclin D2 (Ccnd2), another Notch signaling target, facilitates G1 to S transition in NPCs [21]. The suppression of Ccnd2 is consistent with the G1 arrest in cCMV-infected NPCs. In addition, loss-of-function mutations of Ccnd2 have been linked to microcephaly [58], a common clinical manifestation in human newborns infected with HCMV. On the other side, Ncor2 (nuclear receptor corepressor 2) serves as a transcriptional corepressor that represses expression of Notch-targeted genes in the absence of Notch activation. Thus, downregulation of Ncor2 might result in aberrant Notch activation during neurodevelopment.
Wnt signaling is disrupted by HCMV in human fibroblasts through β-catenin degradation, but not by suppressing the transcription of its gene (Ctnnb1) [1]. In contrast, our data indicate significant downregulation of Ctnnb1 in NPCs. In addition, components of the destruction complex, including axin, APC, GSK-3β, and CK1, which keep β-catenin levels low in the absence of Wnt stimulation, were downregulated at the transcriptional level. We further observed decreased expression of Fzd3, which encodes the Wnt receptor Frizzled 3. These downregulations disrupt Wnt signaling and affect NPC fate decisions.
Disruption of the Hedgehog signaling pathway has been observed in the cCMV mouse model following neonatal inoculation, in which a key transcription factor, Gli1, was upregulated in the cerebellum [35]. Our embryonic cCMV infection model also showed a mild disruption of this pathway with only slight downregulation of Gli1 and Gli2. A strongly repressed gene in our data is Ptch1, which encodes a receptor for the Hedgehog ligand sonic hedgehog and keeps the pathway inactive in the absence of the ligand. While Ptch1 expression was reduced in MCMV-infected NPCs, the transcriptional repressor Gli3 was upregulated. The lack of correlation between Gli3 expression and viral loads suggests that this upregulation may represent a compensatory host response to counteract the decreased levels of Ptch1 and thus maintain pathway inactivity.
Importantly, disruption of all three pathways predominantly occurred in the HVT state, and the degree of perturbation correlates with viral load in individual cells, suggesting a direct effect of the virus rather than indirect effects of inflammatory insults [35]. In contrast, these neurodevelopmental programs in LVT and bystander NPCs were indistinguishable from those in the mockinfected NPCs, indicating relatively unaffected NPC functions and aborted infection in the LVT state.
Due to the absence of reliable markers to distinguish between resident brain macrophages and infiltrating MDMs, previous studies have struggled to effectively investigate the interactions between CMV and these distinct types of brain macrophages. BAMs have been considered the primary viral target due to their location at CNS entry sites [31], but their susceptibility to CMV infection has not yet been reported. MG, while supporting CMV replication in vitro [66], are not the primary target in cCMV human fetal brains [70,74]. In our study, we distinguished BAMs, MG, and MDMs by their specific gene signatures, which were found to be downregulated in cCMV-brains. This downregulation mirrors changes observed in brain macrophages during neuroinflammation in a mouse model of experimental autoimmune encephalomyelitis [29]. More MDMs were infected than BAMs or MG resident macrophages, and like NPCs, key macrophage functions such as phagocytosis and antigen presentation were disrupted in the HVT state, in MDMs more so than MG or BAMs. Similar functional impairment was reported during pulmonary CMV infections where the virus mainly infects resident alveolar macrophages rather than MDMs [2], suggesting a distinct tropism in different tissues but similar viral damage to core macrophage functions. The expression of cytokine genes was higher in MG and BAMs than MDMs, suggesting they are more responsible for recruiting immune cells, and potentially contributing to immunopathology in the infected brains [38]. Notably, Ccl5 expression was specifically enhanced in the HVT state. This increase is unlikely to be directly induced by the virus itself, as CMV encodes at least three viral proteins to suppress CCL5 functions [52]. Therefore, infected macrophages may sense their inability to control viral infection and secrete CCL5 to recruit additional immune cells for viral clearance.
Intrinsic ISGs have recently be demonstrated to play a crucial role in antiviral resistance of stem cells [82]. High levels of intrinsic ISGs, often measured in in vitro differentiation assays, are associated with stemness and negatively correlate with the differentiation state of cells [67,81]. In this study, we provide a comprehensive quantification of intrinsic ISG expression across various cell types in the developing brain. NPCs show lower intrinsic ISG expression than other none-neural cell types and do not necessarily have higher ISG levels compared with their differentiated offspring. While low intrinsic ISG levels in NPCs correlate with high susceptibility to CMV, the resistance observed in neuroblasts and immature neurons suggests that quantifying overall intrinsic ISG expression may not effectively predict CMV susceptibility. Another key element of intrinsic immunity is the PML nuclear body (PML-NB). PML-NB components, including PML, Daxx, Sp100, Morc3, and ATRX, are known host restriction factors targeting HCMV replication across different cell types [48,64,68,80]. However, these genes were expressed at low levels in the neonatal brain, with the exception of Atrx, which showed particularly high expression in neuroblasts and immature neurons. ATRX has been demonstrated to promote heterochromatinization of the HCMV genome, thereby restricting IE gene transcription [78]. Therefore, intrinsic high expression of Atrx in NPC-derived neuronal cells may contribute to the control of cCMV.
Our data reveal significant heterogeneity in ISG induction in response to cCMV across various cell types. Immune and barrier cells exhibited robust ISG responses, likely due to their higher intrinsic ISG levels and greater sensitivity to IFN signaling. In contrast, the magnitude of ISG induction in neural cells was markedly lower. In particular, neuroblasts and immature neurons showed the weakest ISG induction by CMV, which correlates with their lowest intrinsic ISG levels. Compared to these two cell types, NPCs generated a stronger ISG response but were more permissive to CMV infection. CMV encodes multiple viral proteins to suppress host antiviral responses [88]. Previous in vitro single-cell studies have demonstrated an inverse correlation between viral transcript levels and ISG expression [25,67]. This pattern is particularly evident in MDMs and NPCs in the HVT state. However, most other cell types do not demonstrate a clear inverse relationship in this state (Supplementary Fig. 6h). We also observed heterogeneity in individual ISG expression across diverse cell types. Though more ISGs were induced in immune and barrier cells than neural cells, a core set of ISGs with well-known antiviral functions was consistently induced across all cell types, suggesting activation of a universal antiviral program in response to cCMV. However, we did not identify any specific ISGs that could explain the high susceptibility of NPCs and MDMs to CMV infection. Our ISG expression profiling reveals the complexity and heterogeneity of ISG expression in vivo, suggesting that differential regulation of ISGs across various cellular contexts may contribute to the varied susceptibility to CMV infection.
Another important insight from this study is the critical role of IFNγ in controlling cCMV in the brain. First, induced ISGs in the infected brain are typically activated by IFNγ. Second, the expression of IFNα and IFNβ is nearly absent in all cell types, which are supported by significantly lower levels of type I IFNs compared to IFNγ in brains at P7. This attenuated type I IFN signaling in the developing brain is consistent with findings from a previous study in neonates [77]. The lesser importance of type I compared to type II IFN signaling in controlling cCMV is further supported by blocking experiments with IFN antibodies. However, a recent study reported that neutralization of IFNγ had minimal impact on MCMV load in neonatal mouse brains [38]. This discrepancy may be due to the route of administration. In our study, IFNγ antibody was administered via an intracranial route, presumably resulting in a more rapid and higher local concentration of antibody in the brain compared to the intraperitoneal route that was used in that study. In support of our findings, IFNγ knockout mice infected intraperitoneally at P0 had significantly higher brain MCMV titers and required more time to clear the virus than wild-type mice [37]. The importance of IFNγ is also supported by studies demonstrating that NK and T cells, two primary cell types responsible for its secretion, are essential for controlling CMV in newborn infection models [6,10]. Our scRNA-seq data also demonstrate a predominant expression of IFNγ in these cell types; however, low levels of IFNγ during the early stages of cCMV may allow the virus to inflict damage on the developing brain. This delayed IFNγ response is likely due to impaired maturation and function of NK cells induced by CMV [62], as well as the extended time required for recruiting effector T cells to the brain [6].
## Limitations
This study utilized intracranial inoculation to circumvent the murine vertical transmission barrier, thereby mimicking first-trimester congenital HCMV infection. However, this approach bypasses the intricate systemic and maternal-fetal immune interactions of natural transmission, potentially altering the initial viral tropism and early host immune responses. In addition, direct intracranial injection might compromise the blood-brain barrier's integrity, which could facilitate the infiltration of peripheral immune cells and lead to the overrepresentation of macrophages in our dataset. In contrast, neurons are largely absent due to their fragility and susceptibility to damage during enzymatic and mechanical dissociation for whole-cell isolation in scRNA-seq. Single-nucleus RNA sequencing, which captures RNA information from isolated nuclei, is better suited for studying neural responses to cCMV [56]. However, this limitation does not substantially alter our conclusions about CMV tropism and immune responses, as numerous studies have demonstrated that neurons in the fetal brain rarely permit productive viral replication, and that the immune response to cCMV infection in the developing brain is primarily mediated by MG, astrocytes, and infiltrating immune cells [9]. Nevertheless, we acknowledge the growing recognition of the active role of neurons in immune responses to viral infections [75]. Further investigations focusing on neuronal contributions will be invaluable fully understanding of the complex brain immune response to cCMV infection. Despite the new insights that our dataset offers into heterogeneous reactions to cCMV infection across various cell populations at the peak of viral burden and neuroinflammation in our model, a single-time-point snapshot inherently limits our ability to fully capture the dynamic progression of infection and neurodevelopmental perturbations. Future longitudinal analyses across multiple time points would provide a more comprehensive understanding of evolving host-pathogen interactions and long-term neuropathogenesis.
## Abbreviations
## References
1. Angelova, Zwezdaryk, Ferris et al. (2012) "Human cytomegalovirus infection dysregulates the canonical Wnt/betacatenin signaling pathway" *PLoS Pathog*
2. Baasch, Giansanti, Kolter et al. (2021) *Cell*
3. Baldarelli, Smith, Ringwald et al. (2024) "Mouse genome informatics: an integrated knowledgebase system for the laboratory mouse" *Genetics*
4. Baldarelli, Smith, Finger et al. "Campbell Jet al et al (2021) The mouse gene expression database (GXD): 2021 update" *Nucleic Acids Res*
5. Bansod, Kageyama, Ohtsuka (2017) "Hes5 regulates the transition timing of neurogenesis and gliogenesis in mammalian neocortical development" *Development*
6. Bantug, Cekinovic, Bradford et al. (2008)
7. "T lymphocytes control murine cytomegalovirus replication in the central nervous system of newborn animals" *J Immunol*
8. Belzile, Stark, Yeo et al. (2014) "Human cytomegalovirus infection of human embryonic stem Cell-Derived primitive neural stem cells is restricted at several steps but leads to the persistence of viral DNA" *J Virol*
9. Blondel, Guillaume, Lambiotte (2024) "Fast unfolding of communities in large networks: 15 years later" *J Stat Mech-Theory E*
10. Brizic, Hirsl, Britt et al. (2018) "Immune responses to congenital cytomegalovirus infection" *Microbes Infect*
11. Brizic, Hirsl, Sustic et al. (2019) "CD4 T cells are required for maintenance of CD8 T cells and virus control in the brain of MCMV-infected newborn mice" *Med Microbiol Immunol*
13. Chatzakis, Ville, Makrydimas et al. (2020) "Timing of primary maternal cytomegalovirus infection and rates of vertical transmission and fetal consequences" *Am J Obstet Gynecol*
14. Cheeran, Lokensgard, Schleiss (2009) "Neuropathogenesis of congenital cytomegalovirus infection: disease mechanisms and prospects for intervention" *Clin Microbiol Rev*
15. Costa, Becker, Krammer et al. (2024) "Human cytomegalovirus exploits STING signaling and counteracts IFN/ISG induction to facilitate infection of dendritic cells" *Nat Commun*
16. D' Aiuto, Maio, Heath et al. (2012) "Human induced pluripotent stem Cell-Derived models to investigate human cytomegalovirus infection in neural cells" *PLoS ONE*
17. Daley-Bauer, Roback, Crosby et al. (2017) "Mouse cytomegalovirus M36 and M45 death suppressors cooperate to prevent inflammation resulting from antiviral programmed cell death pathways" *Proc Natl Acad Sci U S A*
18. Damania, Lisnic, Cac et al. (2013) "Dual analysis of the murine cytomegalovirus and host cell transcriptomes reveal new aspects of the Virus-Host cell interface" *PLoS Pathog*
19. Dobin, Davis, Schlesinger et al. (2013) "STAR: ultrafast universal RNA-seq aligner" *Bioinformatics*
20. Drayman, Patel, Vistain et al. (2019) "HSV-1 single-cell analysis reveals the activation of anti-viral and developmental programs in distinct sub-populations"
21. Erhard, Baptista, Krammer et al. (2019) "scSLAM-seq reveals core features of transcription dynamics in single cells" *Nature*
22. Franzen, Gan, Bjorkegren (2019) "PanglaoDB: a web server for exploration of mouse and human single-cell RNA sequencing data. Database (Oxford)"
23. Glickstein, Monaghan, Koeller et al. (2009) "Cyclin D2 is critical for intermediate progenitor cell proliferation in the embryonic cortex" *J Neurosci*
24. Goodrum, Mcweeney (2018) "A Single-Cell approach to the elusive latent human cytomegalovirus transcriptome"
25. Guttenplan, Liddelow (2019) "Astrocytes and microglia: models and tools" *J Exp Med*
26. Hafemeister, Satija (2019) "Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression"
27. (1186) *Genome Biol*
28. Hein, Weissman (1038) "Functional single-cell genomics of human cytomegalovirus infection" *Nat Biotechnol*
29. Huang, Pan, Zhou et al. (2023) "Human cytomegalovirus IE1 impairs neuronal migration by downregulating connexin 43" *J Virol*
30. Huang, Zhou, Jiang et al. (2021) "Hearing loss caused by HCMV infection through regulating the Wnt and Notch signaling pathways" *Viruses-Basel*
31. Jiang, Dong, Li et al. (2018) "Proteomic analysis of Zika virus infected primary human fetal neural progenitors suggests a role for doublecortin in the pathological consequences of infection in the cortex" *Front Microbiol*
32. Jordao, Sankowski, Brendecke et al. (2019) "Single-cell profiling identifies myeloid cell subsets with distinct fates during neuroinflammation" *Science*
33. Kavanagh, Gold, Wagner et al. (2001) "The multiple immune-evasion genes of murine cytomegalovirus are not redundant: m4 and m152 inhibit antigen presentation in a complementary and cooperative fashion" *J Exp Med*
34. Kierdorf, Masuda, Jordao et al. (2019) "Macrophages at CNS interfaces: ontogeny and function in health and disease" *Nat Rev Neurosci*
36. Kirschen, Burd (2023) "Modeling of vertical transmission and pathogenesis of cytomegalovirus in pregnancy: opportunities and challenges"
37. Koontz, Bralic, Tomac et al. (2008) "Altered development of the brain after focal herpesvirus infection of the central nervous system" *J Exp Med*
38. Korsunsky, Millard, Fan et al. (2019) "Fast, sensitive and accurate integration of single-cell data with harmony" *Nat Methods*
39. Kosmac, Bantug, Pugel et al. (2013) "Glucocortiocoid treatment of MCMV infected newborn mice attenuates CNS inflammation and limits deficits in cerebellar development" *PLoS Pathog*
40. Doi
41. Kowalczyk, Tirosh, Heck et al. (2015) "Single-cell RNA-seq reveals changes in cell cycle and differentiation programs upon aging of hematopoietic stem cells"
42. *Genome Res*
43. Krstanovic, Mihalic, Rashidi et al. (0186) "Neuron-restricted cytomegalovirus latency in the central nervous system regulated by CD4 T-cells and IFN-γ"
44. Kvestak, Lisnic, Lisnic et al. "Krstanovic Fet al et al (2021) NK/ILC1 cells mediate neuroinflammation and brain pathology following congenital CMV infection" *J Exp Med*
45. Lacaze, Forster, Ross et al. (2011) "Temporal profiling of the coding and noncoding murine cytomegalovirus transcriptomes" *J Virol*
46. Li, Liu, Yang et al. (2015) "Human cytomegalovirus infection dysregulates the localization and stability of NICD1 and Jag1 in neural progenitor cells" *J Virol*
47. Liberzon, Birger, Thorvaldsdottir et al. (2015) "The molecular signatures database (MSigDB) hallmark gene set collection" *Cell Syst*
48. Liu, Yang, Huang et al. (2017) "Human cytomegalovirus IE1 downregulates Hes1 in neural progenitor cells as a potential E3 ubiquitin ligase" *PLoS Pathog*
49. Lodha, Muchsin, Jurges et al. (2023) "Decoding murine cytomegalovirus" *PLoS Pathog*
51. Loewendorf, Kruger, Borst et al. (2004) "Identification of a mouse cytomegalovirus gene selectively targeting CD86 expression on antigen-presenting cells" *J Virol*
52. Loo, Simon, Xing et al. (2019) "Single-cell transcriptomic analysis of mouse neocortical development" *Nat Commun*
53. Luo, Hannemann, Kulkarni et al. (2010) "Human cytomegalovirus infection causes premature and abnormal differentiation of human neural progenitor cells" *J Virol*
54. Luo, Schwartz, Fortunato (2008) "Neonatal neural progenitor cells and their neuronal and glial cell derivatives are fully permissive for human cytomegalovirus infection" *J Virol*
55. Ma, Yao, Wang et al. (2022) "MORC3 restricts human cytomegalovirus infection by suppressing the major immediate-early promoter activity" *J Med Virol*
56. Maninger, Bosse, Lemnitzer et al. (2011) "M94 is essential for the secondary envelopment of murine cytomegalovirus" *J Virol*
57. Marcinowski, Lidschreiber, Windhager et al. (2012) "Real-time transcriptional profiling of cellular and viral gene expression during lytic cytomegalovirus infection" *PLoS Pathog*
58. Martin, El-Behi, Fontaine et al. (2017) "Analysis of microglia and Monocyte-derived macrophages from the central nervous system by flow cytometry" *Jove-J Vis Exp*
59. Mcsharry, Avdic, Slobedman (2012) "Human cytomegalovirus encoded homologs of Cytokines, chemokines and their receptors: roles in Immunomodulation" *Viruses-Basel*
60. Moulden, Sung, Brizic et al. (2021) "Murine models of central nervous system disease following congenital human cytomegalovirus infections" *Pathogens*
61. Nguyen, Loveland, Gibson (2008) "Nuclear localization sequences in cytomegalovirus capsid assembly proteins (UL80 proteins) are required for virus production: inactivating NLS1, NLS2, or both affects replication to strikingly different extents" *J Virol*
62. Ochocka, Segit, Walentynowicz et al. (2021) "Single-cell RNA sequencing reveals functional heterogeneity of glioma-associated brain macrophages" *Nat Commun*
63. Ogrodnik, Evans, Fielder et al. (2021) "Whole-body senescent cell clearance alleviates age-related brain inflammation and cognitive impairment in mice" *Aging Cell*
64. Piccirilli, Gabrielli, Bonasoni et al. (2023) "Fetal brain damage in human fetuses with congenital cytomegalovirus infection: histological features and viral tropism" *Cell Mol Neurobiol*
65. (1007)
66. Pirozzi, Lee, Horsley et al. (2021) "Proximal variants in associated with microcephaly, short stature, and developmental delay: A case series and review of inverse brain growth phenotypes" *Am J Med Genet A*
67. Redwood, Messerle, Harvey et al. (2005) "Use of a murine cytomegalovirus k181-derived bacterial artificial chromosome as a vaccine vector for Immunocontraception" *J Virol*
68. Rosenberg, Roco, Muscat et al. (2018) "Single-cell profiling of the developing mouse brain and spinal cord with split-pool barcoding" *Science*
69. Rozman, Nachshon, Samia et al. (2022) "Temporal dynamics of HCMV gene expression in lytic and latent infections" *Cell Rep*
70. Rozmanic, Lisnic, Matesic et al. (2023) "perinatal murine cytomegalovirus infection reshapes the transcriptional profile and functionality of NK cells" *Nat Commun*
71. Rybak-Wolf, Wyler, Pentimalli et al. (1038) "Modelling viral encephalitis caused by herpes simplex virus 1 infection in cerebral organoids" *Nat Microbiol*
72. Schilling, Scherer, Stamminger (2021) "Intrinsic immune mechanisms restricting human cytomegalovirus replication" *Viruses-Basel*
73. Schoch, Ciufo, Domrachev et al. (2020) "NCBI Taxonomy: a comprehensive update on curation, resources and tools"
74. Schut, Gekker, Hu et al. (1994) "Cytomegalovirus replication in murine microglial Cell-Cultures -Suppression of permissive infection by Interferon-Gamma" *J Infect Dis*
75. Schwartz, Shnayder, Nachshon et al. (2023) "Molecular characterization of human cytomegalovirus infection with single-cell transcriptomics" *Nat Microbiol*
76. Seitz, Heusel, Stamminger et al. (2022) "Promyelocytic leukemia protein potently restricts human cytomegalovirus infection in endothelial cells" *Int J Mol Sci*
77. Shum, Walczak, Chang et al. (2019) "Quantitation of mRNA transcripts and proteins using the BD Rhapsody™ Single-Cell analysis system"
78. Slavuljica, Kveštak, Huszthy et al. (2014) "Immunobiology of congenital cytomegalovirus infection of the central nervous system-the murine cytomegalovirus model" *Cell Mol Immunol*
79. Ssentongo, Hehnly, Birungi et al. (2021) "Congenital cytomegalovirus infection burden and epidemiologic risk factors in countries with universal screening A systematic review and Meta-analysis" *Jama Netw Open*
80. Doi (2021)
81. Steuerman, Cohen, Peshes-Yaloz et al. (2018) "Dissection of Influenza Infection In Vivo by Single-Cell RNA Sequencing" *Cell Syst*
82. Taisne, Lussignol, Hernandez et al. (2019) "Human cytomegalovirus hijacks the autophagic machinery and LC3 homologs in order to optimize cytoplasmic envelopment of mature infectious particles" *Sci Rep-Uk*
84. Teissier, Fallet-Bianco, Delezoide et al. (2014) "Cytomegalovirus-induced brain malformations in fetuses" *J Neuropathol Exp Neurol*
85. Telikani, Monson, Hofer et al. (2022) "Antiviral response within different cell types of the CNS" *Front Immunol*
86. Tsutsui, Kosugi, Kawasaki et al. (1002) "Cytomegalovirus induces interferon-stimulated gene expression and is attenuated by interferon in the developing brain" *van den Pol AN*
87. Walter, Majumder, Kalejta (2024) "ATRX restricts Human Cytomegalovirus (HCMV) viral DNA replication through heterochromatinization and minimizes unpackaged viral genomes" *Plos Pathogens*
88. Wang, Wen, Zhou et al. (2019) "Human cytomegalovirus pUL97 upregulates SOCS3 expression via transcription factor RFX7 in neural progenitor cells" *PLoS Pathog*
89. Wu, Thi, Huang et al. (2018) "Intrinsic Immunity Shapes Viral Resistance of Stem Cells" *Cell*
90. Wu, Kwong, Rice (2019) "Antiviral resistance of stem cells" *Curr Opin Immunol*
91. Wyler, Franke, Menegatti et al. (2019) "Single-cell RNA-sequencing of herpes simplex virus 1-infected cells connects NRF2 activation to an antiviral program" *Nat Commun*
92. Zeisel, Munoz-Manchado, Codeluppi et al. (2015) "Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq" *Science*
93. Zhang, Lan, Xu et al. (2019) "CellMarker: a manually curated resource of cell markers in human and mouse" *Nucleic Acids Res*
94. Zhou, Li, Zhang et al. (2023) "A case of congenital human cytomegalovirus infection with placental and pulmonary calcification despite presence of intrauterine IgG" *J Med Virol*
95. Zhou, Mei, Wang et al. (2022) "A congenital CMV infection model for follow-up studies of neurodevelopmental disorders, neuroimaging abnormalities, and treatment" *Jci Insight*
96. Zhu, Chiang, Gack (2023) "Viral evasion of the interferon response at a glance" *J Cell Sci* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12802182&blobtype=pdf | # Human alphacoronavirus replication and innate immune induction in airway culture systems
Alejandra Fausto, Clayton Otter, Leonel Torres, Ebba Blomqvist, Nicole Bracci, David Renner, Hui Li, Tan, Devon Mooring, Nadine Ebert, Bettina Trüeb, Volker Thiel, Noam Cohen, James Burke, Susan Weiss
## Abstract
Compared with lethal betacoronaviruses, there is limited knowledge of how human alphacoronaviruses HCoV-NL63 (NL63) and HCoV-229E (229E) interact with host innate immune responses. We compared NL63 and 229E infections in human lung-derived cell lines, A549 ACE2 and MRC-5, and primary nasal epithelial air-liquid interface (ALI) cultures. We measured the infection rates and viral replication kinetics. Additionally, we assessed the activation of three dsRNA-induced pathways, interferon (IFN) production and signaling, oligoadenylate synthetase-ribonuclease L (OAS/RNase L), and protein kinase R (PKR), following infection with each virus. Although both 229E and NL63 replicated efficiently in nasal ALI cultures, NL63 replicated minimally in A549 ACE2 or MRC-5. In lung-derived cell lines, significant IFN mRNA induction as well as PKR activation was observed during NL63 but not during 229E infection. In contrast, in nasal ALI cultures, significant induction of both the IFN and PKR pathways was observed during 229E and NL63 infection. Notably, there was no evidence of RNase L activation during infection with either virus in cell lines or nasal ALI cultures. Infection with a recombinant 229E expressing an inactivated nsp15 endoribonuclease U (EndoU) led to increased dsRNA levels, stronger induction of all three antiviral pathways, and attenuation of replication relative to wild-type 229E. This indicates that 229E nsp15 EndoU regulates host dsRNA responses, as shown previously for porcine epidemic diarrhea virus (PEDV) and pathogenic betacoronaviruses. These findings demonstrate that NL63 and 229E differentially modulate host dsRNA-induced innate immune pathways and highlight the critical role of nsp15 EndoU in suppressing antiviral responses to facilitate efficient viral replication.IMPORTANCE Seasonal human coronaviruses (HCoVs) are the causative agents of more than 15% of common cold cases each year. However, compared with more virulent HCoVs such as SARS-CoV-2, there has been limited research on these viruses. We compared the replication of HCoV-NL63 (NL63) and HCoV-229E (229E). Additionally, we examined their interactions with interferon signaling and related innate immune pathways in lung-derived cell lines and primary nasal epithelial cultures. 229E replicates efficiently in each of these culture systems, with significant dsRNA-induced pathway induction only in nasal cells. In contrast, NL63 replicates efficiently only in nasal cell cultures but induces innate immune pathways in all three culture systems. Moreover, the conserved CoV innate immune antagonist endoribonuclease U aids in evading these responses in 229E infection. This study expands our understanding of common-cold HCoV-host interactions and provides insight into differences between seasonal and lethal HCoVs.
C oronaviruses (CoVs), a family within the Nidovirus order, are enveloped, positive- sense, single-stranded RNA viruses (1,2). To date, seven HCoVs have been identified, all of which are believed to be zoonotic, with origins in bats and mice (1,3,4). HCoVs are classified into either the betacoronavirus or alphacoronavirus genus. The betacoronavi ruses include HCoV-OC43 (OC43), HCoV-HKU1 (HKU1), severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and SARS-CoV-2. The alphacoronaviruses include HCoV-NL63 (NL63) and HCoV-229E (229E) (5). 229E and NL63 are further classified into subgenera Duvinacovirus and Setracovirus, respectively (5). Although betacoronavirus-host interactions are well-studied, we will focus on the two human alphacoronaviruses, which have been relatively understudied. 229E was first isolated in 1967 from a patient with mild upper respiratory symptoms, whereas NL63 was isolated in 2004 from a 7-month-old infant suffering from bronchio litis and conjunctivitis (6,7). NL63 and 229E, together with betacoronaviruses OC43 and HKU1, circulate globally and are estimated to cause 15%-30% of mild-to-moderate upper respiratory tract illnesses in humans (5,8). NL63 is also a leading cause of pediatric croup (laryngotracheobronchitis) (8)(9)(10)(11), and both NL63 and 229E can trigger severe lower respiratory infections in vulnerable populations, including children, the elderly, and immunocompromised patients (12,13).
All CoVs encode a single-stranded RNA genome with conserved organization. The 5′-two-thirds of the genome contains two overlapping open reading frames (ORFs), ORF1a and ORF1b, that encode 16 nonstructural proteins (nsp1-16). The remaining one-third of the genome contains ORFs encoding the viral structural proteins spike, envelope, membrane, and nucleocapsid, as well as accessory proteins that are distinct among HCoV genera and subgenera (1,2,5). CoV nonstructural proteins serve various roles in replication and transcription of CoV genomes and encode innate immune evasion functions. The 3' accessory proteins have been shown to be dispensable for viral replication in most cell lines, although some play important roles in innate immune antagonism, viral pathogenesis, and virulence in vivo as well as in cell lines with intact innate immune responses (1,2,14). NL63 and 229E (laboratory-adapted versions) encode only one and two accessory proteins, respectively (Fig. 1A). This is in contrast to the more virulent MERS-CoV, SARS-CoV, or SARS-CoV-2, which each encode at least four accessory proteins (1,(3)(4)(5)15).
CoV RNA replication takes place in replication transcription complexes (RTCs) within double-membrane vesicles (DMVs) formed from rearranged endoplasmic reticulum. As a byproduct of genome replication and subgenomic mRNA transcription, CoVs produce dsRNA, which serves as a pathogen-associated molecular pattern (PAMP) that is sensed by host cell pattern recognition receptors (PRRs), leading to activation of antiviral innate immune pathways. Thus, replication in DMVs is believed to protect the viral RNA from recognition by host sensors. Despite this shielding of CoV RNA, innate immune responses are initiated during coronavirus infection (17). Sensing of dsRNA by host melanoma differentiation-associated protein 5 (MDA5) (18) (a retinoic acid-inducible gene [RIG] Ilike receptor [RLR]) (19) leads to the expression of type I and III IFNs and the downstream induction of IFN-stimulated genes (ISGs), cytokines, and chemokines (20)(21)(22)(23). dsRNA is also recognized by protein kinase R (PKR), leading to its autophosphorylation and the subsequent phosphorylation of the eukaryotic translation initiation factor eIF2α, which shuts down protein synthesis (19,24,25). Oligoadenylate synthetases (OASs) serve as a third dsRNA sensor, producing 2'-5'-oligoadenylates, which activate the antiviral endoribonuclease RNase L, responsible for degrading both host and viral ssRNAs (26). These dsRNA-induced pathways are all independently activated; however, since PKR and OASs are themselves ISGs, both the PKR and OAS/RNase L pathways can be further induced by concurrent IFN signaling (19,24,25) (Fig. 1B).
In addition to the protection of RNA replication in DMVs, CoVs actively suppress innate immune responses by expressing both conserved nonstructural proteins (nsps) with host antagonist functions as well as genus/subgenus-specific accessory proteins (1,17). Notable among these is the conserved nsp15 endoribonuclease (EndoU), which has previously been shown to act as an innate immune antagonist during infection with all four coronavirus genera (alpha, beta, gamma, and deltacoronaviruses) (27)(28)(29)(30)(31)(32)(33)(34)(35)(36). CoV EndoU has been most thoroughly characterized in the murine coronavirus (MHV) model (30,32,37). EndoU is reported to cleave viral ssRNA, limiting the production of dsRNA byproducts and innate immune activation. Indeed, MHV recombinant viruses expressing enzymatically inactivated nsp15 produce more dsRNA, induce the IFN, PKR, and RNase L pathways more robustly, and are severely attenuated in primary macrophages as well as in vivo in the liver and spleen of mice relative to parental wild-type (WT) virus (30,32). Similarly, we have recently reported that MERS-CoV and SARS-CoV-2 mutants expressing inactive nsp15 EndoU stimulate dsRNA-induced pathway activation and are attenuated for replication relative to WT in lung-derived cell lines as well as primary nasal epithelial cells (29,38). Furthermore, recombinant 229E expressing an inactive nsp15 EndoU also exhibited a growth defect and elevated IFN-β production compared with WT 229E in human blood-derived macrophages (32), but the effect of nsp15 EndoU on 229E infection has not been assessed in respiratory epithelial cells.
Here, we compare 229E and NL63 in terms of their replication, induction of type I and type III IFN mRNA and ISGs, and activation of the PKR and OAS/RNase L path ways in human respiratory cell lines and primary nasal epithelial cell cultures grown at the air-liquid interface (ALI). These nasal ALI cultures model the initial site of viral replication in the human airway and the primary barrier to infection by respiratory viruses where innate immune responses are critical. We further characterize a 229E nsp15 mutant virus in terms of replication, dsRNA accumulation, and innate immune activation in the respiratory epithelium. These experiments contribute to our currently limited understanding of alphacoronavirus-host interactions.
## RESULTS
## NL63 and 229E differentially infect lung-derived cell lines
We first evaluated the percentage of infected cells in the lung-derived epithelial cell line A549 ACE2 (derived from a lung adenocarcinoma) (39) and the lung fibroblast cell line MRC-5 (derived from normal fetal lung tissue) during NL63 and 229E infection. A549 ACE2 cells stably express angiotensin-converting enzyme 2 (ACE2); we used this cell type because ACE2 expression is required for the replication of NL63, and thus this allowed for comparison of 229E and NL63 in the same cell line (40,41). To quantify the percentage of cells infected, cultures were infected with 229E or NL63 (multiplicity of infection [MOI] = 1 PFU/cell) at 33°C; this temperature was used for all experiments described below, as we and others have shown that common cold viruses such as NL63 and 229E replicate optimally at 33°C (nasal airway temperature) relative to 37°C (lower airway temperature) (42, 43). Cells were fixed at the indicated time points and stained with antibodies directed against 229E or NL63 nucleocapsid (N) protein for immunofluorescence (IF) or quantification via flow cytometry. Representative IF images are shown in Fig. 2A andB. In A549 ACE2 cells, both 229E and NL63 infect relatively few cells at either time point (Fig. 2A). In MRC-5 cells, NL63 infects few cells while 229E infects nearly all cells by 48 h post-infection (hpi) (Fig. 2B). To obtain a more quantitative assessment of infection, the cells were fixed and stained with anti-NL63 or anti-229E N protein APC-conjugated antibody and analyzed by flow cytometry (see Fig. S1 for gating strategy). In A549 ACE2 cells, 229E infection resulted in a slightly larger percentage of N-positive cells than NL63 at 48 hpi (6% vs 3%, respectively) (Fig. 2C). In MRC-5 cells, 229E infection led to a substantial increase in infected cells over time (~90% of cells infected by 48 hpi), whereas the percentage of infected cells following NL63 infection remained very low, consistent with IF data (Fig. 2D).
## 229E replicates more robustly than NL63 in respiratory cell culture
We next compared the kinetics of replication of 229E and NL63 in A549 ACE2 and MRC-5 cells. Cells were infected (MOI = 1 PFU/cell), and the infectious virus was collected from supernatants at 24 and 48 h post-infection (hpi) for quantification via plaque assay; peak titers were reached at 48 hpi, and a significant cytopathic effect was observed after 48 hpi. Despite the low percentage of infected cells seen for both viruses in A549 ACE2 cells, 229E replicated efficiently to peak titer of 7.5 log 10 PFU/mL at 48 hpi, whereas NL63 titers only slightly surpassed the plaque assay limit of detection (LOD) (Fig. 3A). In MRC-5 cells, 229E replicated more efficiently than NL63 (reaching titers of 7 log 10 PFU/mL at 48 hpi), but NL63 replication was more efficient compared with A549 ACE2 cells (reaching 4.5 log 10 PFU/mL by 48 hpi) (Fig. 3B).
Since 229E and NL63 use different host cell receptors for viral entry (aminopeptidase N or APN for 229E and ACE2 for NL63), we hypothesized that the observed differences in replication in respiratory cell lines may be due to endogenous receptor expression levels (or, in the case of A549 ACE2 cells, overexpressed receptor). We performed western blots on protein harvested from mock-infected A549 ACE2 and MRC-5 cells (Fig. S2). Our results indicate that receptor expression level is not the primary determinant of HCoV replication in these cell lines. APN expression was significantly higher in MRC-5 than in A549 ACE2 cells, which may explain the increased percentage of infected cells in MRC-5 (Fig. 2D) but not the high 229E titers in both cell lines. ACE2 expression is high in A549 ACE2 cells and nearly undetectable in MRC-5 cells, and thus, receptor expression levels do not correlate with NL63 replication or proportion of infected cells.
We have previously optimized a nasal epithelial culture system to model viral replication and innate immune induction at the primary barrier site to infection. Primary nasal cells are differentiated at an air-liquid interface (ALI) to recapitulate the heteroge neous cellular population and mucociliary function of the nasal airway. (42,44,45). Nasal ALI cultures were infected at an MOI of 5 PFU/cell to maximize the percentage of cells initially infected, which we have previously reported to be low (45), and apical surface liquid (ASL) was collected at 24-h intervals for quantification of shed virus titers via plaque assay. Growth curves in nasal ALI cultures were extended to 96 hpi, based on our previous observations of peak viral titers in this system (42,45). In contrast to replication data in lung-derived cell lines, both 229E and NL63 replicated efficiently in nasal ALI cultures (peak titers ~ 6 log 10 PFU/mL, although with slightly different kinetics) (Fig. 3C). Receptor expression level in nasal ALI cultures similarly does not predict replication, as ACE2 levels are lower than in A549 ACE2 (despite efficient NL63 replication). Taken together, our data highlight the limited replication of NL63 in lung-derived cell lines relative to 229E and suggest that primary epithelial cell culture systems may be required to adequately compare these viruses.
## NL63 and 229E differentially induce the interferon (IFN) and protein kinase R (PKR) pathways
To investigate the degree of IFN activation during infection with 229E or NL63, cells were infected (MOI = 5), and intracellular RNA was extracted following cell lysis at the indicated time points. RT-qPCR was used to quantify the mRNA expression of type I/III IFN genes (IFNB and IFNL1) and representative ISGs (IFIT1 and OAS2). In A549 ACE2 cells, no significant induction of IFN or ISGs was observed during 229E infection at either time point relative to mock-infected cells (Fig. 4A andD). In contrast, infection of A459 ACE2 cells with NL63 robustly induced IFNB and IFNL1 as well as IFIT1 and OAS2 by 48 hpi (Fig. 4A andD). As in A549 ACE2 , we observed more IFN and ISG mRNA induction during NL63 infection of MRC-5 cells, compared with 229E infection (Fig. 4B andE). We further examined ISG expression at the protein level via western blot with antibodies against IFIT1 and Viperin. ISG protein expression was greatest during NL63 infection of both cell types, with minimal induction during 229E infection, corroborating our results at the mRNA level (Fig. 4G andH). We performed similar experiments in nasal ALI cultures, in which both viruses replicate efficiently (42, 45). IFN and ISG mRNA expressions, as well as ISG protein expression, were induced significantly following either 229E or NL63 infection, with increased activation at the later time point (Fig. 4C, F, andI).
Next, we analyzed 229E-and NL63-infected protein lysates via western blot for activation of the PKR pathway using antibodies against phosphorylated PKR (p-PKR) and its downstream target eIF2α (p-eIF2α). In line with the IFN pathway data, we observed PKR phosphorylation in NL63-infected A549 ACE2 and MRC-5 cells, whereas only mild p-PKR above mock-infected levels was observed during 229E infection (Fig. 5A andB). Total PKR levels were also increased during NL63 infection, which was expected given that PKR is an ISG. Phosphorylation of downstream target eIF2α occurred only during NL63 infection in A549 ACE cells, whereas p-eIF2α levels were not increased above mock during NL63 infection of MRC-5 cells or during 229E infection of either cell line (Fig. 5A andB), suggesting incomplete activation of the PKR pathway or phosphorylation of eIF2α that is below the level of detection via western blot. A similar analysis of protein samples from nasal ALI cultures revealed p-PKR activation by both 229E and NL63, contrasting once more with data in cell lines. Interestingly, we observed stronger and earlier induction of p-PKR in 229E-infected nasal cells relative to NL63 (Fig. 5C). p-eIF2α was also induced in nasal ALI cultures following 229E but not NL63 infection, consistent with our previous observation that p-eIF2α is difficult to detect over background levels in nasal cultures (46). We also note that there is some cross-reactivity between antibodies against 229E and NL63 N protein, as we have observed previously. This is not surprising as the genomes of these two strains share some sequence homology (47).
We next investigated whether infection with 229E or NL63 induced the activation of the OAS/RNase L pathway, using degradation of ribosomal RNA (rRNA) as a readout for RNase L activity. Infection with Sindbis virus (SINV), an alphavirus that robustly activates RNase L, served as a positive control with robust degradation of 18S and 28S rRNA (16,48,49). rRNA remained intact in 229E-and NL63-infected A549 ACE2 or MRC-5 cells, suggesting that RNase L was not activated by either virus (Fig. S3A andB). There was also no detectable rRNA degradation observed following infection of nasal ALI cultures with either virus, although it is important to note that we have never detected RNase L activation in the nasal ALI culture system (Fig. S3C). Overall, we did not observe activation of the RNase L pathway in any cell type infected with NL63 and 229E.
## 229E nsp15 EndoU antagonizes dsRNA-induced antiviral pathways
Our data highlight relatively minimal induction of the IFN, PKR, and OAS/RNase L pathways during 229E infection, except during infection of nasal ALI cultures. Prior studies have identified the conserved CoV nsp15 endoribonuclease as a potent inhibitor of dsRNA-induced immune pathways during infection by multiple viruses, including betacoronaviruses MHV, MERS-CoV, and SARS-CoV-2, and alphacoronavirus PEDV. We sought to explore the role of nsp15 EndoU during 229E infection using a recombinant 229E expressing a catalytically inactivated nsp15 EndoU (His to Ala mutation at amino acid residue 250) (Fig. 1A). This recombinant virus was previously characterized by Kindler et al. in human blood-derived macrophages (32) but has not been characterized in airway culture systems. We compared the extent of replication, percentage of infected cells, as well as innate immune induction following infection with the isogenic recombi nant wild-type (WT) 229E (r229E) vs. r229E-NSP15 mut (32,50). These recombinant viruses contain eight amino acid substitutions when compared with the 229E virus used in the above experiments (see Materials and Methods). We chose to use MRC-5 cells and nasal ALI cultures to characterize these viruses, forgoing A549 ACE2 cells, as NL63 (which requires the ACE2 receptor) was not included in these experiments. Additionally, in contrast to our laboratory 229E strain, these recombinant 229E strains failed to replicate in A549 ACE2 .
We first quantified the percentage of infected cells following the infection of MRC-5 cells with r229E and r229E-nsp15 mut via flow cytometry and found a significant decrease in the proportion of N-positive cells during nsp15 mut relative to WT infection. This difference was largest at the later time point (33% vs 6% infected for WT vs nsp15 mut , respectively) (Fig. 6A). Given the proposed mechanism of nsp15 EndoU in limiting dsRNA accumulation during infection, we quantified dsRNA production during infection of MRC-5 cells with either virus. Representative images are shown in Fig. 6B, in which there appeared to be a larger number of smaller dsRNA puncta dispersed throughout infected cells following nsp15 mutant infection. Quantification of dsRNA levels revealed a significantly increased total area of dsRNA puncta per cell following r229E-nsp15 mut relative to r229E infection (Fig. 6C). We additionally separated individual dsRNA puncta according to area (small, medium, and large) and found an increased number of small puncta (<1.0 µm 2 ) during r229E-nsp15 mut infection (Fig. S4A), whereas more large puncta (>5.0 µm 2 ) were observed during r229E infection (Fig. S4B). This suggests that dsRNA is more dispersed throughout the cell during infection with the mutant virus.
We next evaluated the replication kinetics of these viruses in MRC-5 as well as nasal ALI cultures. Consistent with the lower percentage of infected cells, r229E-nsp15 mut is attenuated relative to r229E in MRC-5 cells by ~1 log 10 PFU/mL (Fig. 6D). This growth defect was even more pronounced in nasal ALI cultures, in which nsp15 mut was 1-3 log 10 PFU/mL attenuated (depending on time point), with mutant virus titers diminishing to nearly the limit of detection by 96 hpi (Fig. 6E). This suggests that earlier and more efficient viral clearance following r229E-nsp15 mut relative to r229E infection in nasal cultures.
We hypothesized that the increased dsRNA production during infection with the 229E nsp15 mutant would result in stronger induction of dsRNA-induced pathways, which would explain its replication defect as well as limited spread of infection (and lower percentage of infected cells). In both MRC-5 cells and nasal ALI cultures, we found that type I and III IFN mRNAs were induced more robustly during r229E-nsp15 mut relative to r229E infection (Fig. 7A andB). Representative ISG mRNA levels (IFIT1, OAS2) were also increased during infection with the nsp15 mutant in MRC-5 cells, but were induced to a similar extent in nasal ALI cultures (Fig. 7C andD). This increased IFN signature during nsp15 mut infection was also observed at the protein level in both cell types using western blots for IFIT1 and Viperin protein expression (Fig. 7E andF). When we compared these viruses in terms of induction of the PKR pathway, we observed increased PKR phosphory lation during r229E-nsp15 mut infection relative to r229E in both MRC-5 and nasal ALI cultures (Fig. 8A andB). We did not detect a significant increase in p-eIF2α signal during nsp15 mutant infection in either cell type. Corroborating our replication data, there was a clear decrease in 229E N levels during infection with r229E-nsp15 mut infection compared with r229E in nasal ALI culture, with 229E N being nearly undetectable at late times during infection with the mutant (Fig. 8B).
Finally, we evaluated rRNA degradation patterns as a readout for RNase L activation and found that r229E-nsp15 mut infection of MRC-5 cells resulted in RNase L activation at 72 hpi, whereas significant activation of this pathway was not detected during parental r229E infection (Fig. 8C). Since this was our first observation of RNase L activity during alphacoronavirus infection, we assessed the percentage of infected cells that activated RNase L by IF assays for dsRNA (infection marker) and poly[a]-binding protein-1 (PABPC1), which translocates from the cytosol to the nucleus upon activation of RNase Lmediated mRNA decay (51)(52)(53). We observed that 47% of cells infected with r229E-nsp15 mut displayed nuclear PABPC1 staining (Fig. 8D andE), whereas only 3% of cells infected with r229E displayed nuclear PABPC1. These data show that 229E nsp15 inhibits the activation of RNase L. We have previously used a fibrosarcoma cell line, HT1080, to evaluate RNase L activity, so we infected HT1080 cells with r229E and r229E-nsp15 mut and evaluated rRNA integrity. Confirming our results in MRC-5 cells, RNase L was activated following infection with the nsp15 mutant in HT1080 cells (Fig. 8F). RNase L was not activated during infection with either r229E or r229E-nsp15 mut in nasal ALI cultures, which was not surprising as we have never detected RNase L activation in these cultures (Fig. S3D). These data highlight nsp15 as a potent antagonist of dsRNA-induced immunity during 229E infection.
## DISCUSSION
Although interest in lethal coronaviruses has intensified in recent years, driven by the global impact of the SARS-CoV-2 pandemic and COVID-19 disease, the seasonal corona viruses, NL63, 229E, OC43, and HKU1, have not received extensive research attention in part due to challenges associated with traditional cell culture systems. We have recently reported methods for the propagation and quantification of OC43, 229E, and NL63 by identifying the optimal infection temperature (33°C) at which immortalized cell lines should be used to generate high-titer virus stocks and for optimal virus titration for each seasonal HCoV (43). In addition, we recently compared the cellular tropism, replication kinetics, and virus-induced cytotoxicity of lethal and seasonal HCoVs in nasal ALI cultures at 33°C (44,45). Tissue culture models for the study of pathogenic HCoVs such as MERS-CoV and SARS-CoV-2 have been extensively characterized, and these viruses' ability to evade and suppress host antiviral pathways to optimize their replication has been reported by our group and others (16,29,(54)(55)(56)(57). In contrast, there are relatively few studies that compare airway models for the characterization of NL63 and 229E and their activation or evasion of dsRNA-induced antiviral responses. To gain a more comprehensive understanding of the entire HCoV family, we characterized these human alphacoronaviruses and their interactions with dsRNA-induced immune pathways using three respiratory culture systems, the lung epithelial A549 ACE2 cell line, the lung fibroblast MRC-5 cell line, and primary nasal epithelial ALI cultures (43,44).
In comparing the replication kinetics and percentage of cells infected by each virus (Fig. 2 and3), we found that 229E robustly replicates in A549 ACE2 , MRC-5, and nasal ALI cultures, whereas NL63 only reaches high viral titers in nasal ALI cultures. This suggests that primary cell culture systems may be necessary in order to compare alphacoronavi ruses. We speculate that this may be because NL63 requires features of the in vivo airway, such as a heterogeneous cellular population and mucociliary function, for optimal viral entry and replication. Our prior experiments with NL63 have identified ciliated epithelial cells as the primary cell type infected by NL63 in nasal ALI culture (45). Additionally, we observed that 229E infected a much smaller percentage of A549 ACE2 cells relative to MRC-5 cells but achieved high viral titers in both cell types without induction of significant innate immune responses. We cannot explain the surprising observation of production of high titers of 229E despite the low percentage of infected A549 cells; we hypothesize that intracellular events apart from innate immunity may regulate the level of viral replication on a per-cell basis, allowing for high viral titers produced by a low percentage of infected cells.
It is important to note that these two viruses use different receptors for entry into host cells. NL63 utilizes angiotensin-converting enzyme-2 (ACE2), whereas 229E uses aminopeptidase N (APN) (40,(58)(59)(60). When we compared expression of these cellular receptors in each of the three airway culture systems (Fig. S2), we found that receptor expression is not the major determinant of viral replication. ACE2 expression was highest in A549 ACE2 cells, but NL63 failed to replicate efficiently in this cell line. ACE2 expression was comparatively below the limit of detection by western blot in nasal ALI cultures, in which NL63 replicated robustly. These data suggest that high levels of the cellular receptor are not necessary for productive infection, as we have demonstrated previously for SARS-CoV-2 ( 16) and MHV infections (61). The inability of NL63 to replicate efficiently in A549 ACE2 cells may be due to robust host responses (such as IFN/ISG induction) that limit viral replication and spread. Alternatively, there may be co-receptors or other cellular factors required for efficient infection in cell culture. For example, ACE2 plays a critical role in SARS-CoV-2 replication; however, ACE2 expression profiles along the airway are not always directly associated with infection patterns, and ACE2-independent alternative receptors have been reported to mediate SARS-CoV-2 entry (62). Potential alternative mechanisms for NL63 entry (which may be particularly important for tissue culture-adapted strains of NL63) have yet to be reported.
We report that NL63 significantly induced IFN signaling and PKR pathway activation in both A549 ACE2 and MRC-5 cells despite low levels of replication, whereas 229E did not appreciably induce either pathway in these cell lines (Fig. 4 and5). Conversely, in nasal ALI cultures, both 229E and NL63 induce the IFN/ISG and PKR pathway, with earlier activation during 229E infection (Fig. 4 and5). We previously observed early induction of antiviral interferon (IFN) signaling during 229E and NL63 infection of nasal ALI cultures and showed that clearance of both viruses from these cultures was IFN-mediated (46). Consistent with our findings but limited to 229E, two additional studies confirm induction of the IFN pathway following 229E infection, in bronchial epithelial ALI cultures and MRC-5 cells (63,64).
Although we observed PKR phosphorylation during NL63 infection of all three cell types and during 229E infection of nasal cell cultures, we did not consistently observe phosphorylation of its downstream mediator, eIF2a. This may be partially related to the fact that PKR is itself an ISG and is thus upregulated in the context of IFN signaling, whereas eIF2α is not induced by IFN. Additionally, PKR is not the sole kinase that phosphorylates eIF2α, but rather one of four kinases that compose the integrated stress response (ISR) (65,66). PKR-like ER kinase (PERK), general control nondepressible 2 (GCN2), and heme-regulated eIF2α kinase (HRI) can also activate eIF2α, following the accumulation of unfolded proteins, amino acid starvation, and heme deficiency, respectively (65,66). In addition, host pathways antagonizing phosphorylation of eIF2α may be activated during alphacoronavirus infection. For example, we previously reported that GADD34 activation during infection with betacoronaviruses OC43 and MERS-CoV results in dephosphorylation of eIF2α (67). Future experiments will evaluate translational shutoff (the downstream impact of PKR pathway activation), and the extent to which these additional kinases and regulators of eIF2α phosphorylation may or may not be activated during 229E and NL63 infection.
This limited activation of the IFN and PKR pathway during infection of cell lines with 229E led us to investigate innate immune antagonism during 229E infection. The conserved CoV nsp15 EndoU has been reported to serve as a potent inhibitor of dsRNA-induced pathways during infection of all four CoV genera: betacoronavirus MHV, SARS-CoV-2, MERS-CoV, alphacoronavirus PEDV infection (27)(28)(29)(30)(31)(32)(33)(34), as well as gammacoronavirus infectious bronchitis virus (IBV) (35) and porcine deltacoronavirus (PDCoV) (36). A 229E recombinant mutant expressing a defective nsp15 protein had previously been shown to be attenuated in human macrophages (32) but has not been characterized in airway culture systems. Quantification of dsRNA production during infection with the 229E nsp15 mut revealed an increase in total dsRNA production in individual infected cells relative to WT (Fig. 6B), as well as an increase in the number of small dsRNA puncta. Our findings seem to corroborate a model whereby mutation of nsp15 EndoU results in increased dispersal of dsRNA throughout the cytoplasm, as has been suggested during MHV infection (68). This increase in dsRNA resulted in increased dsRNA-induced pathway activation as well as attenuated replication during r229E-nsp15 mut infection relative to WT (Fig. 6D, E, 7, and8).
Interestingly, infection with this 229E nsp15 mutant resulted in activation of the OAS/RNase L pathway, as evidenced by total rRNA degradation in MRC-5 and HT1080 cells (Fig. 8C andF), as well as increased nuclear PABPC1 localization in MRC-5 cells (Fig. 8D andE). This is the first instance of activation of this pathway in the context of alphacoronavirus infection, as neither WT 229E nor NL63 activated OAS/RNase L in any cell culture system. Studies of betacoronaviruses have highlighted that SARS-CoV-2 induces RNase L in A549 ACE2 and Calu3 cells (16,28), whereas MERS-CoV infection only results in RNase L activation when two of its innate immune antagonists are inactivated (nsp15 EndoU and NS4b, an accessory protein with phosphodiesterase activity) (29). We have previously failed to observe RNase L activation in nasal ALI cultures during a variety of RNA virus infections or via treatment with synthetic dsRNA, poly(I:C) (28), suggesting that this cell type may be unable to activate the pathway. Indeed, we have observed previously that several primary mouse cell types are also unable to activate this pathway due to insufficient expression levels of OAS genes (the sensor responsible for OAS/RNase L activation) (69-71). We do not have an EndoU-deficient NL63 mutant available for study, but we hypothesize that infection with such a virus would result in significant attenuation of the virus, which would be most striking in nasal ALI culture, where NL63 replicates most efficiently. Studying the role of EndoU during NL63 infection would provide a better understanding of similarities and differences between human alpha-and beta-coronavirus nsp15 activity.
In addition to the conserved CoV nsp15 EndoU, accessory proteins likely contribute to the differential innate immune activation observed in response to common cold alphacoronaviruses compared with lethal betacoronaviruses. Indeed, 229E and NL63 have the smallest genome sizes of the seven HCoVs, each approximately 27.5 kb (5), and as such contain fewer accessory genes than the lethal betacoronaviruses that contain numerous accessory genes (at least four for SARS-CoV-2 and MERS-CoV) that serve as innate immune antagonists (29,54). NL63 encodes only one presumed accessory protein, encoded in ORF3. The NL63 ORF3 protein was initially thought to function similarly to SARS-CoV ORF3a, which was shown to play a role in the regulation of NFκB-dependent cytokines and modulation of S protein-mediated endocytosis (72). Müller et al. reported that the NL63 ORF3 protein colocalizes with E and M within the endoplasmic reticulum/Golgi intermediate compartment (ERGIC) and that it is N-glycosy lated at the N-terminus. Analysis of purified viral particles revealed that the ORF3 protein is incorporated into virions and, therefore, is an additional structural protein with a proposed function within the viral assembly and budding process (73), but its specific role during NL63 function has not been characterized. The prototype laboratory-adapted 229E strain has a split accessory gene, encoding the putative ORF4a and ORF4b proteins (74). ORF4a localizes to the ERGIC in infected cells and possesses ion channel activity, demonstrated in both Xenopus oocytes and yeast (75). Interestingly, an analysis of five 229E clinical isolates found that each encodes ORF4, which is a homolog of the NL63 ORF3 protein, instead of ORF4a and ORF4b proteins encoded by MERS-CoV (74), but its function has not been thoroughly characterized. Overall, there are limited data implicating NL63 and 229E accessory proteins in antagonizing innate immune responses. To understand the function of these accessory proteins during authentic infection, it will be necessary to generate and characterize recombinant viruses lacking each of these genes, as we and others have done previously with MERS-CoV and SARS-CoV-2 accessory genes (29,54).
With the potential for new pathogenic HCoVs to emerge, there is a need to fully characterize and understand all HCoVs and their host interactions to guide surveillance and aid the creation of antiviral treatments and vaccines. By characterizing human alphacoronaviruses in multiple airway culture systems, we begin to understand the virus-host interactions that may help predict pathogenesis and transmissibility. Future studies will compare alphacoronaviruses in upper (nasal) vs. lower (bronchial) airway culture systems, as well as compare recombinant alphacoronaviruses lacking their accessory genes with similar betacoronavirus mutants.
## MATERIALS AND METHODS
## Viruses
Our laboratory strain NL63 and 229E genome sequences were compared with wild-type (WT) reference sequences, NL63 (ATCC NR-470; GenBank: AY567487 or NC_005831.2) and 229E (ATCC VR-740; GenBank NC_002645), respectively. Nucleotide homologies between our laboratory strain and reference strain were 99.93% for 229E and 99.96% for NL63. The recombinant wild-type 229E (r229E) and nsp15 EndoU-deficient mutant virus (r229E-nsp15 mut ) were generated using the vaccinia virus-based reverse genetic system as previously described (32,50,76). The r229E genome is 100% identical to GenBank NC_002645 and 99.3% identical to our laboratory strain. The r229E genome, compared with our lab strain, encodes amino acid differences in ORF1a (V416A, S2359R, T25124); ORF 4 a (D94Y); E protein (T36I); M protein (L82F); and spike protein (F230C, I700L). NL63 and 229E stocks were prepared in LLC-MK2 and HUH7 cells, respectively, as previously described (43,45).
## Cell lines
Human A549 cells expressing the receptor ACE2 (A549 ACE2 ) (16,39) were cultured in RPMI 1640 (Gibco catalog no. 11875) supplemented with 10% fetal bovine serum (FBS) and 1× penicillin-streptomycin. LLC-MK2 cells were cultured in Minimum Essential Medium (MEM) α (Gibco 12571063) supplemented with 10% FBS and 1× of penicillin-streptomy cin. MRC-5 (CCL-171) (77) and HT1080 (HT1080/CCL-121) (78) cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco 11965) supplemented with 10% FBS, 1× of penicillin-streptomycin. HUH7 cells were cultured in DMEM (Gibco 11965) supplemented with 10% FBS, 1× penicillin-streptomycin, 1% 100× MEM Non-Essential Amino Acids (NEAA; Gibco 11140050), and 1% GlutaMAX Supplement (Gibco 35050079).
## Primary nasal epithelial air-liquid interface (ALI) cultures
Nasal mucosal specimens were obtained via cytologic brushing of patients' nares in the Department of Otorhinolaryngology-Head and Neck Surgery, Division of Rhinology at the University of Pennsylvania, and the Philadelphia Veteran Affairs Medical Center after obtaining informed consent. The full study protocol, including the acquisition and use of nasal specimens, was approved by the University of Pennsylvania Institutional Review Board (protocol #800614) and the Philadelphia VA Institutional Review Board (protocol #00781). Patients with a history of systemic disease or on immunosuppressive medications were excluded. ALI cultures were grown and differentiated on 0.4 µm pore transwell inserts as previously described (16,44,45). All nasal ALI cultures used in this study were differentiated after pooling nasal cells derived from four independent donors (in equal quantities) in order to limit donor-to-donor variability. ALI culture differentiation medium used for cultures was PneumaCult-ALI basal medium (Stemcell Technologies) (45).
## Viral replication kinetics and titration
For infection, immortalized cell lines were counted before seeding in 12-well plates (A549 ACE2 and HT1080 cells at 3 × 10 5 cells per well; MRC-5 cells at 5 × 10 5 cells per well). The next day, supernatant samples containing virus were diluted in serum-free RPMI (A549 ACE2 infections) or serum-free DMEM (MRC-5 infections) and added to cells for adsorption for 1 h at 33°C. After 1 h, cells were washed three times with PBS and fed with DMEM or RPMI supplemented with 2% FBS. For virus titration, 200 µL of the supernatant was collected at the times indicated and stored at -80°C for plaque assay on LLC-MK2 (NL63) or HUH7 (229E) cells as previously described (43). Nasal ALI cultures were apically infected at MOI = 5 PFU/cell with either NL63, 229E, r229E, or r229E-nsp15 mut , apical surface liquid was collected via the addition of PBS, and viral titers were quantified via standard plaque assay as previously described (43,44).
## Immunofluorescence (IF) staining for infected cells
Infections for IF staining to visualize infected cells (Fig. 2A andB) were conducted at an MOI of 1 using glass-bottom 12-well plates (Cellvis). Following infection, at indicated time points, cells were washed 3 times with 1× PBS and fixed in 4% paraformaldehyde at room temperature for 30 min. The cells were then washed 3 times with 1× PBS and permeabilized with 0.1% Triton X-100 in 1× PBS for 10 min and blocked with 2% Bovine Serum Albumin (BSA) in 1× PBS for 30 min at room temperature. Primary antibody incubation was done overnight at 4°C, followed by secondary incubation with Alexa Fluor dyes for 1 h at room temperature. See Spreadsheet S1 for the manufacturer and dilution used for each antibody. Images were acquired by immunofluorescence microscopy with a Nikon Eclipse Ti2 using a Nikon 20× Plan APO objective and Nikon DS-Qi1Mc-U3 12-bit camera. Images were processed using Fiji/ImageJ software.
## Intracellular nucleocapsid staining assay/flow cytometry
Briefly, indicated cells were cultured 24 h before infection at 33°C in T75 flasks. On the day of infection, the cells were infected with the indicated virus at an MOI of 1 or mock-infected and incubated for 1 h at 33°C. Following viral adsorption, the cells were washed, and fresh media was added. Cells were incubated until 24 or 48 hpi, at which time, the cells were harvested via trypsinization and washed twice with 1× PBS. Cells were then placed in 96-well U-bottom plates and were stained with LIVE/DEAD Fixable Aqua Dead Cell Stain at a 1:500 ratio (diluted in 1× PBS). Following initial staining, cells were fixed at 4°C for 30 min using the eBioscience fixation/permeabilization kit. After fixation and washing of cells, the cells were permeabilized using eBioscience Permeabili zation Buffer (diluted to 1X using deionized water) for 5 min in the dark. Cells are then incubated with APC-conjugated (ab201807) NL63 or 229E Nucleocapsid antibodies at a ratio of 1:1,000 in 1× permeabilization buffer at 4°C overnight in the dark. The following day, the cells were washed, pelleted, and resuspended in 200 µL of 1× PBS. All samples were analyzed on a BD LSR-II analyzer and analyzed with FlowJo X software.
## Western blotting
Cell lysates were harvested at indicated time points using RIPA buffer (50 mM Tris pH 8, 150 mM NaCl, 0.5% deoxycholate, 0.1% SDS, 1% NP40) supplemented with protease inhibitors (Roche: cOmplete mini EDTA-free protease inhibitor) and phospha tase inhibitors (Roche: PhosStop easy pack). Lysates were harvested via scraping of the well or transwell insert and incubated on ice for 20 min, centrifuged for 20 min at 15,000 RPM at 4°C, and the supernatant was mixed 3:1 with 4× Laemmli sample buffer. Samples were boiled at 95°C for 5 min, then separated via sodium dodecyl sulfate-poly acrylamide gel electrophoresis (SDS/PAGE) and transferred to polyvinylidene difluoride (PVDF) membrane. Blots were blocked in 5% BSA or milk in 1X Tris-buffered saline with 0.1% Tween 20 Detergent (TBST) and probed with antibodies as listed in Spreadsheet S1. Blots were visualized using Thermo Scientific SuperSignal West Femto Substrate. Blots were stripped using Thermo Scientific Restore Western Blot stripping buffer for 1 h at room temperature and then re-blocked and probed sequentially with antibodies (54).
## Quantitative PCR (q-PCR)
Cells were lysed at the indicated time points with buffer RLT Plus (Qiagen RNeasy Plus Kit #74106), and total RNA was extracted following the manufacturer's protocol. RNA was reverse transcribed into complementary DNA (cDNA) using the High-Capacity Reverse Transcriptase Kit (Applied Biosystems). This cDNA was amplified using specific qRT-PCR primers for each target gene, iQ SYBR Green Supermix (Bio-Rad), and the QuantStudio 3 PCR system (Thermo Fisher). See Spreadsheet S1 for primer sequences used for each target. Technical triplicates were averaged, and changes in mRNA levels were reported as fold change over mock, using the formula 2 -Δ(ΔCt) . ΔCt values were calculated using the formula ΔCt = Ct gene of interest -Ct 18S . Δ(ΔCt) was calculated by subtracting mock-infec ted ΔCt values from ΔCt values for NL63-or 229E-infected samples (54).
## rRNA degradation assay
Total RNA was harvested with buffer RLT Plus (Qiagen RNeasy Plus Kit #74106) and analyzed on an RNA chip with an Agilent Bioanalyzer using the Agilent 196 RNA 6000 Nano Kit and its prescribed protocol as we have described previously (16).
## Immunofluorescence (IF) assay for dsRNA and PABPC1 quantification
The indicated cell type was seeded onto glass coverslips (Thomas Scientific: 1203J81) and infected as described above. At the indicated time point, samples were fixed with 4% paraformaldehyde (PFA) for 30 min prior followed by permeabilization in 70% ethanol. For IFA, the samples were incubated with the indicated primary antibody diluted in 1× PBS at 4°C overnight (Spreadsheet S1). Secondary antibodies were Alexa Fluor antibodies purchased from Abcam and added at 1:1,000 in 1× PBS. Cells were washed 3 times in 1× PBS, then secondary antibodies were diluted in 1× PBS and added for 2 h at room temperature. Coverslips were then washed 3 times in 1× PBS before being mounted on coverslides (Fisher: 12-544-11) with Vectashield (Vector Laboratories: 101098-044) or ProLong Gold Antifade Mountant (Invitrogen: P36931). Images were taken on a Nikon Eclipse Ti2 with a CFI60 Plan Apochromat Lambda D 100 x Oil Immersion Objective Lens, N.A. 1.45, W.D. 0.13 mm, F.O.V. 25 mm, DIC, Spring Loaded. The filter set inclu ded: C-FL DAPI Filter Set, High-Signal-Noise, Semrock Brightline, Excitation: 356/30 nm (341-371nm). Image processing and analysis were performed using FIJI 2.16.0 (ImageJ2), and data processing was conducted in Microsoft Excel.
## smRNA-FISH for quantification of GAPDH degradation
For smFISH, the cells were seeded onto glass coverslips and infected as described for IF. After 4% PFA fixation and permeabilization of 70% EtOH, coverslips were placed in buffer A (filter-sterilized 2× SSC with 10% formamide) for 5 min. For each sample, smFISH probes (sequences provided in Spreadsheet S1) were prepared as described in (51) and added 1:100 into 50 μL of hybridization buffer: 10% dextran sulfate (Fisher Scientific Co LLC: S4030), 10% formamide (Fisher Scientific Co LLC: BP227500), and 1× nuclease-free SSC (Life Technologies Corporation: 15557044). The hybridization buffer containing the probes was added onto parafilm placed inside a Petri dish (Fisher: FB0875711A). The coverslips were then incubated with the smFISH probes overnight at 37°C. The following day, coverslips were washed, mounted on slides, and imaged.
## Statistical analysis
Plotting of data and statistical analysis were performed using GraphPad Prism software (GraphPad Software, Inc.). Statistical significance was determined by comparing mutant virus (r229E-nsp15 mut ) to WT 229E (r229E) using repeated measures two-way ANOVA for viral replication curves and qRT-PCR. Displayed significance is determined by P value, where * =P < 0.05; ** =P < 0.01; *** =P < 0.001; and **** =P < 0.0001; ns = not significant, ns is not displayed on the graph.
## References
1. Weiss, Leibowitz (2011) "Coronavirus pathogenesis" *Adv Virus Res*
2. Weiss (2020) "Forty years with coronaviruses" *J Exp Med*
3. V'kovski, Kratzel, Steiner et al. (2021) "Coronavirus biology and replication: implications for SARS-CoV-2" *Nat Rev Microbiol*
4. Perlman, Netland (2009) "Coronaviruses post-SARS: update on replication and pathogenesis" *Nat Rev Microbiol*
5. Liu, Liang, Fung (2021) "Human coronavirus-229E, -OC43, -NL63, and -HKU1 (Coronaviridae)"
6. Mcintosh, Dees, Becker et al. (1967) "Recovery in tracheal organ cultures of novel viruses from patients with respiratory disease" *Proc Natl Acad Sci*
8. Van Der Hoek, Pyrc (2006) "Human coronavirus NL63, a new respiratory virus" *FEMS Microbiol Rev*
9. Corman, Muth, Niemeyer et al. (2018) "Hosts and sources of endemic human coronaviruses" *Adv Virus Res*
10. Gorse, Connor, Hall et al. (2009) "Human coronavirus and acute respiratory illness in older adults with chronic obstructive pulmonary disease" *J Infect Dis*
11. Principi, Bosis, Esposito (2010) "Effects of coronavirus infections in children" *Emerg Infect Dis*
12. Fielding (2011) "Human coronavirus NL63: a clinically important virus?" *Future Microbiol*
13. Mayer, Nellessen, Hahn-Ast et al. (2016) "Fatal outcome of human coronavirus NL63 infection despite successful viral elimination by IFN-alpha in a patient with newly diagnosed ALL" *Eur J Haematol*
14. Nunes, Kuschner, Rabede et al. (2014) "Clinical epidemiology of bocavirus, rhinovirus, two polyomaviruses and four coronaviruses in HIV-infected and HIV-uninfected South African children" *PLoS One*
15. Liu, Fung, Chong et al. (2014) "Accessory proteins of SARS-CoV and other coronaviruses" *Antiviral Res*
16. Fung, Liu (2019) "Human coronavirus: host-pathogen interaction" *Annu Rev Microbiol*
17. Li, Renner, Comar et al. (2021) "SARS-CoV-2 induces double-stranded RNA-mediated innate immune responses in respiratory epithelial-derived cells and cardiomyocytes" *Proc Natl Acad Sci*
18. Tanneti, Stillwell, Weiss (2025) "Human coronaviruses: activation and antagonism of innate immune responses" *Microbiol Mol Biol Rev*
19. Bender, Weiss (2008) "Murine coronavirus mouse hepatitis virus is recognized by MDA5 and induces type I interferon in brain macrophages/microglia" *J Virol*
20. Hur (2019) "Double-stranded RNA sensors and modulators in innate immunity" *Annu Rev Immunol*
21. Boncristiani, Criado, Arruda (2009) "Respiratory viruses"
22. Hariri, Cohen (2016) "New insights into upper airway innate immunity" *Am J Rhinol Allergy*
23. Vareille, Kieninger, Edwards et al. (2011) "The airway epithelium: soldier in the fight against respiratory viruses" *Clin Microbiol Rev*
24. Parker, Prince (2011) "Innate immunity in the respiratory epithelium" *Am J Respir Cell Mol Biol*
25. Kikkert (2020) "Innate immune evasion by human respiratory RNA viruses" *J Innate Immun*
26. Gantier, Williams (2007) "The response of mammalian cells to double-stranded RNA" *Cytokine Growth Factor Rev*
27. Silverman (2007) "Viral encounters with 2′,5′-oligoadenylate synthetase and RNase L during the interferon antiviral response" *J Virol*
28. Ulferts, Ziebuhr (2011) "Nidovirus ribonucleases: structures and functions in viral replication" *RNA Biol*
29. Otter, Bracci, Parenti et al. (2023) "SARS-CoV-2 nsp15 endoribonuclease antagonizes dsRNAinduced antiviral signaling"
31. Comar, Otter, Pfannenstiel et al. (2022) "MERS-CoV endoribonu clease and accessory proteins jointly evade host innate immunity during infection of lung and nasal epithelial cells" *Proc Natl Acad Sci*
32. Deng, Baker (2018) "An "Old" protein with a new story: coronavirus endoribonuclease is important for evading host antiviral defenses" *Virology (Auckl)*
33. Ancar, Li, Kindler et al. (2020) "Physiologic RNA targets and refined sequence specificity of coronavirus EndoU" *RNA*
34. Kindler, Gil-Cruz, Spanier et al. (2017) "Early endonuclease-mediated evasion of RNA sensing ensures efficient coronavirus replication" *PLoS Pathog*
35. Hackbart, Deng, Baker (2020) "Coronavirus endoribonuclease targets viral polyuridine sequences to evade activating host sensors" *Proc Natl Acad Sci*
36. Deng, Van Geelen, Buckley et al. (2019) "Coronavirus endoribonuclease activity in porcine epidemic diarrhea virus suppresses type I and type III interferon responses" *J Virol*
37. Gao, Gong, Fang et al. (2021) "Inhibition of antiviral stress granule formation by coronavirus endoribonuclease nsp15 ensures efficient virus replication" *PLoS Pathog*
38. Deng, Buckley, Pillatzki et al. (2021) "Development and utilization of an infectious clone for porcine deltacoronavirus strain" *Virology (Auckl)*
39. Kindler, Thiel (2014) "To sense or not to sense viral RNA -essentials of coronavirus innate immune evasion" *Curr Opin Microbiol*
40. Otter, Bracci, Parenti et al. (2024) "SARS-CoV-2 nsp15 endoribonu clease antagonizes dsRNA-induced antiviral signaling" *Proc Natl Acad Sci*
41. Giard, Aaronson, Todaro et al. (1973) "In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors" *J Natl Cancer Inst*
42. Hofmann, Pyrc, Van Der Hoek et al. (2005) "Human coronavirus NL63 employs the severe acute respiratory syndrome coronavirus receptor for cellular entry" *Proc Natl Acad Sci*
43. Li, Sui, Huang et al. (2007) "The S proteins of human coronavirus NL63 and severe acute respiratory syndrome coronavirus bind overlapping regions of ACE2" *Virology (Auckl)*
44. Otter, Renner, Tan et al. (2024) "Interferon signaling in the nasal epithelium distinguishes among lethal and common cold coronaviruses and mediates viral clearance" *Proc Natl Acad Sci*
45. Fausto, Otter, Bracci et al. (2023) "Improved culture methods for human coronaviruses HCoV-OC43, HCoV-229E, and HCoV-NL63"
46. Otter, Tan, Weiss et al. (2023) "Infection of primary nasal epithelial cells grown at an air-liquid interface to characterize human coronavirus-host interactions" *J Vis Exp*
47. Otter, Tan, Khosla et al. (2023) "Infection of primary nasal epithelial cells differentiates among lethal and seasonal human coronaviruses" *Proc Natl Acad Sci*
48. Otter, Renner, Tan et al. (2023) "Interferon signaling in the nasal epithelium distinguishes among lethal and common cold respiratory viruses and is critical for viral clearance" *bioRxiv*
49. Pyrc, Dijkman, Deng et al. (2006) "Mosaic structure of human coronavirus NL63, one thousand years of evolution" *J Mol Biol*
50. Whelan, Parenti, Hatterschide et al. (2021) "Zika virus employs the host antiviral RNase L protein to support replication factory assembly" *Proc Natl Acad Sci*
51. Whelan, Li, Silverman et al. (2019) "Zika virus production is resistant to RNase L antiviral activity" *J Virol*
52. Eriksson, Makia, Thiel (2008) "Generation of recombinant coronaviruses using vaccinia virus as the cloning vector and stable cell lines containing coronaviral replicon RNAs" *Methods Mol Biol*
53. Cusic, Watkins, Burke (2023) "Single-cell analysis of RNase Lmediated mRNA decay" *Methods Enzymol*
54. Burke, Ripin, Ferretti et al. (2022) "RNase L activation in the cytoplasm induces aberrant processing of mRNAs in the nucleus" *PLoS Pathog*
55. Burke, Moon, Matheny et al. (2019) "RNase L reprograms translation by widespread mRNA turnover escaped by antiviral mRNAs" *Mol Cell*
56. Comar, Goldstein, Li et al. (2019) "Antagonism of dsRNA-induced innate immune pathways by NS4a and NS4b accessory proteins during MERS coronavirus infection" *mBio*
57. Thornbrough, Jha, Yount et al. (2016) "Middle East respiratory syndrome coronavirus NS4b protein inhibits host RNase L activation" *mBio*
58. Fung, Yuen, Ye et al. (2020) "A tug-of-war between severe acute respiratory syndrome coronavirus 2 and host antiviral defence: lessons from other pathogenic viruses" *Emerg Microbes Infect*
59. Sievers, Cheng, Csiba et al. (2024) "SARS-CoV-2 and innate immunity: the good, the bad, and the "goldilocks" *Cell Mol Immunol*
60. Li, Tomlinson, Wong et al. (2019) "The human coronavirus HCoV-229E S-protein structure and receptor binding"
61. Yeager, Ashmun, Williams et al. (1992) "Human aminopeptidase N is a receptor for human coronavirus 229E" *Nature*
62. Milewska, Nowak, Owczarek et al. (2018) "Entry of human coronavirus NL63 into the cell" *J Virol*
63. Bender, Phillips, Scott et al. (2010) "Murine coronavirus receptors are differentially expressed in the central nervous system and play virus strain-dependent roles in neuronal spread" *J Virol*
64. Lim, Zhang, Chang (2022) "ACE2-independent alternative receptors for SARS-CoV-2" *Viruses*
65. Duncan, Xu, Licursi et al. (2023) "Interferon regulatory factor 3 mediates effective antiviral responses to human coronavirus 229E and OC43 infection" *Front Immunol*
66. Loo, Wark, Esneau et al. (2020) "Human coronaviruses 229E and OC43 replicate and induce distinct antiviral responses in differentiated primary human bronchial epithelial cells" *Am J Physiol Lung Cell Mol Physiol*
67. Pakos-Zebrucka, Koryga, Mnich et al. (2016) "The integrated stress response" *EMBO Rep*
68. Wek (2018) "Role of eIF2α kinases in translational control and adaptation to cellular stress" *Cold Spring Harb Perspect Biol*
69. Renner, Parenti, Bracci et al. (2025) "Betacoronaviruses differentially activate the integrated stress response to optimize viral replication in lung-derived cell lines" *Viruses*
70. Deng, Hackbart, Mettelman et al. (2017) "Coronavirus nonstructural protein 15 mediates evasion of dsRNA sensors and limits apoptosis in macrophages" *Proc Natl Acad Sci*
71. Birdwell, Zalinger, Li et al. (2016) "Activation of RNase L by murine coronavirus in myeloid cells is dependent on basal Oas gene expression and independent of virus-induced interferon" *J Virol*
72. Zhao, Birdwell, Wu et al. (2013) "Cell-type-specific activation of the oligoadenylate synthetase-RNase L pathway by a murine coronavirus" *J Virol*
73. Li, Weiss (2016) "Antagonism of RNase L is required for murine coronavirus replication in Kupffer cells and liver sinusoidal endothelial cells but not in hepatocytes" *J Virol*
74. Narayanan, Huang, Makino (2008) "SARS coronavirus accessory proteins" *Virus Res*
75. Müller, Van Der Hoek, Voss et al. (2010) "Human coronavirus NL63 open reading frame 3 encodes a virion-incorporated N-glycosylated membrane protein" *Virol J*
76. Dijkman, Jebbink, Wilbrink et al. (2006) "Human coronavirus 229E encodes a single ORF4 protein between the spike and the envelope genes" *Virol J*
77. Zhang, Wang, Lv et al. (2014) "The ORF4a protein of human coronavirus 229E functions as a viroporin that regulates viral production" *Biochim Biophys Acta*
78. Thiel, Herold, Schelle et al. (2001) "Infectious RNA transcribed in vitro from a cDNA copy of the human coronavirus genome cloned in vaccinia virus" *J Gen Virol*
79. Jacobs, Jones, Baille (1970) "Characteristics of a human diploid cell designated MRC-5" *Nature*
80. Rasheed, Wa, Toth et al. (1974) "Characterization of a newly derived human sarcoma cell line (HT-1080)" *Cancer* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724244&blobtype=pdf | # B cell-intrinsic IRF-1 and conserved gammaherpesvirus protein kinase cooperate to promote murine gammaherpesvirus-driven germinal center response and splenic latent reservoir
Cade Rahlf, Christopher Jondle, Erika Johansen, Vera Tarakanova, Vera Ai165578, Tarakanova, Vera Ai182596, Erika Ai191670, Johansen
## Abstract
Gammaherpesviruses infect >90% of adults and are associated with B cell lymphomas. These viruses drive the expansion and differentiation of germinal center B cells to amplify latent viral reservoir and ensure life-long infection of memory B cells. Additionally, infected germinal center B cells seed viral lymphomagenesis. We previously demonstrated that global deficiency of Interferon Regulatory Factor-1 (IRF-1), a classically antiviral transcription factor, results in increased murine gammaherpesvirusdriven germinal center response and latent viral reservoir. In contrast, B cell-specific loss of IRF-1 expression attenuates murine gammaherpesvirus chronic infection and germinal center response. All gammaherpesviruses encode a conserved protein kinase. We previously showed that the murine gammaherpesvirus protein kinase antagonizes the antiviral activity of global IRF-1 expression to support the establishment of chronic infection. Intriguingly, in this study, we show cooperation between the gammaherpesvi rus protein kinase and B cell-intrinsic IRF-1 expression that supported optimal establish ment of splenic latent viral reservoir and murine gammaherpesvirus-driven germinal center response. Both viral protein kinase and B cell-intrinsic IRF-1 expression supported the survival and proliferation of germinal center B cells during chronic gammaherpesvi rus infection. Elevated DNA damage response or Fas/FasL expression was not observed, failing to account for elevated apoptosis of IRF-1-deficient germinal center B cells. However, B cell-intrinsic IRF-1 deficiency led to decreased MHC-II expression by germinal center B cells during chronic gammaherpesvirus infection. In summary, the results of this study illustrate the critical role of the cell type in defining the functional outcome of the host-viral interaction. IMPORTANCE Gammaherpesviruses are highly prevalent pathogens that uniquely target B cells for the establishment of life-long infection. This study demonstrates cooperation between a conserved gammaherpesvirus protein kinase and B cell-intrinsic IRF-1, a classical host antiviral factor. We show that this cooperation is critical to support proliferation and survival of germinal center B cells, a subset of B cells that is latently infected by gammaherpesviruses and is critical for both the establishment of chronic infection and viral lymphomagenesis. KEYWORDS gammaherpesvirus, germinal center B cells, chronic infection, gammaher pesvirus protein kinase, IRF-1 G ammaherpesviruses are pervasive pathogens that uniquely infect and manipulate B cell biology to establish life-long infection. Human (Epstein-Barr virus, EBV) and murine (murine gammaherpesvirus 68, MHV68) gammaherpesviruses target naïve B cells in the secondary lymphoid organs to drive differentiation of both infected and bystander B cells through the germinal center response (1-3). Proliferation of infected germinal
center B cells leads to rapid expansion of the viral latent reservoir; further differentiation into memory or plasma cells supports long-term latent infection or viral reactivation (4,5). Importantly, gammaherpesvirus-driven germinal center response underlies viral pathogenesis as many EBV-positive B cell lymphomas bear evidence of germinal center differentiation (2). Given the species specificity of gammaherpesviruses, we and others have taken advantage of the tractable MHV68 experimental system to define host and viral factors that regulate gammaherpesvirus-driven germinal center response during natural infection of an intact host (6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19).
All gammaherpesviruses encode a conserved protein kinase. While a plethora of host and viral substrates of gammaherpesvirus protein kinases have been identified in cultured cells, the function of gammaherpesvirus protein kinases in vivo during chronic infection of an intact host is less understood. We previously demonstrated that expression and enzymatic function of a conserved gammaherpesvirus protein kinase encoded by MHV68 (orf36) facilitate the MHV68-driven germinal center response during chronic infection (8,20). Correspondingly, the expression of orf36 encoded by human Kaposi's sarcoma-associated herpesvirus (KSHV) was sufficient to drive B cell differentiation and eventual lymphomagenesis in transgenic mice in the absence of infection (21), highlighting an important role of the gammaherpesvirus protein kinases in manipulation of B cell differentiation.
In contrast to that observed for MHV68 orf36, global deficiency of Interferon Regulatory Factor 1 (IRF-1), a tumor suppressor and a founding member of the IRF family, led to an exaggerated germinal center response driven by chronic infection with MHV68, but not LCMV (10). Further, IRF-1 protein expression was selectively decreased in EBV-positive but not EBV-negative B cell lymphomas arising in transplant patients (10). Given the opposite germinal center phenotypes driven by loss of MHV68 orf36 or global IRF-1 expression, the interplay between MHV68 orf36 and IRF-1 was examined using a combination of viral and host genetic approaches. Interestingly, global IRF-1 deficiency rescued the attenuated germinal center response driven by the orf36 null MHV68 mutant (N36S) and partially rescued the N36S splenic reservoir (22), indicating an antagonistic relationship between MHV68 orf36 and global IRF-1 expression.
IRF-1 is ubiquitously expressed, with the expression further potentiated by several stimuli, including IFN signaling. While IRF-1 is expressed in B cells, its role in B cell biology remains far less explored as compared to its functions in the innate immune system. Given the observed antiviral phenotype of global IRF-1 expression during chronic MHV68 infection, we generated a mouse model of B cell-intrinsic IRF-1 deficiency by combining conditional IRF-1 alleles with CD19 promoter-driven Cre recombinase knock-in allele (23). Surprisingly, and in contrast to that observed during global IRF-1 deficiency, B cell-intrinsic IRF-1 expression was proviral, as evidenced by decreased splenic latent reservoir and germinal center response driven by wild-type MHV68 infection in mice with IRF-1-deficient B cells (23).
The current study took advantage of a combination of host and viral genetics to define the mechanisms underlying the proviral role of B cell-intrinsic IRF-1 expression during chronic gammaherpesvirus infection and to determine the extent to which such mechanisms are modified by the expression of MHV68 protein kinase. Intriguingly, the proviral role of B cell-intrinsic IRF-1 expression in the establishment of MHV68 latent reservoir and MHV68-driven germinal center response was no longer evident in the absence of MHV68 orf36. A large proportion of MHV68 latent reservoir in the spleen is supported by infected germinal center B cells during the establishment of chronic infection. MHV68 orf36 and B cell-intrinsic IRF-1 expression independently promoted proliferation of germinal center B cells during chronic infection. In contrast, MHV68 orf36 and B cell-intrinsic IRF-1 played interdependent roles to attenuate the apoptosis of germinal center B cells. The observed increase in apoptosis of germinal center B cells was not accompanied by increased expression of Fas/FasL or increased activity of DNA damage response, a pro-apoptotic pathway activated in rapidly proliferating cells, such as germinal center B cells. However, B cell-intrinsic IRF-1 expression promoted germinal center B cell expression of MHC-II, a protein complex that mediates interaction between T follicular helper cells and germinal center B cells. Finally, and in contrast to that observed in the spleen, the previously reported antagonism between MHV68 orf36 and global IRF-1 expression was recapitulated in the peritoneal cells of mice with B cell-specific IRF-1 deficiency.
## RESULTS
## The interplay between conserved gammaherpesvirus protein kinase and B cell-intrinsic IRF-1 expression is modified by the anatomic site of infection
We previously demonstrated that the MHV68-encoded protein kinase orf36 antagonizes the antiviral function of global IRF-1 expression during chronic infection to promote the MHV68-driven germinal center response and the establishment of the latent reservoir (22). In contrast, we also showed that B cell-intrinsic IRF-1 expression is proviral during MHV68 infection as the establishment of chronic infection was attenuated in mice with B cell-specific IRF-1 deficiency following intranasal infection with 500 PFU of wild-type (WT) MHV68 (23). To define the relationship between MHV68 orf36 and B cell-intrinsic IRF-1 expression during chronic infection, parameters of chronic MHV68 infection were measured in Cd19 Cre/wt Irf1 fl/fl (Cre-positive) and Cd19 wt/wt Irf1 fl/fl (Cre-negative) mice 16 days after intranasal inoculation with 10,000 PFU of WT or orf36-deficient MHV68 mutant (N36S) (24). The 10,000 PFU dose was selected to enable meaningful comparisons of host parameters of infection as the N36S mutant fails to establish splenic latent reservoir at 16 days post-infection when inoculated at a lower 500 PFU dose used in our previous study (8,23).
As expected, the frequency and percent of MHV68 DNA-positive splenocytes were decreased in Cre-positive as compared to Cre-negative mice infected with WT MHV68 (Fig. 1A andB, compare groups with filled and open circles). The observed decrease (~25 fold) in the splenic latent reservoir of WT MHV68 following a 10,000 PFU inocu lum was greater as compared to the previously published study, which used a lower inoculation dose of 500 PFU (~3 fold) (23), supporting the proviral role of B cell-intrinsic IRF-1 expression. Similar to that observed previously (8), the inability to express orf36 led to a significant decrease in the frequency and percent of the N36S-infected splenocytes of the control, Cre-negative mice (Fig. 1A andB, compare groups with filled circles and squares). Surprisingly, B cell-intrinsic IRF-1 deficiency had no effect on the latent reservoir of the N36S viral mutant, despite a profound attenuation of the WT MHV68 latent reservoir observed in Cre-positive mice. Thus, the proviral role of B cell-intrinsic IRF-1 in supporting the splenic latent reservoir was no longer evident under conditions when MHV68 orf36 could not be expressed. Likewise, the proviral role of orf36 required B cell-intrinsic IRF-1 expression, suggesting cooperation between the viral and host protein in the establishment of splenic latent reservoir.
Unlike germinal center B cells that host the majority of latent MHV68 reservoir in the spleen, plasma cells primarily support MHV68 reactivation from splenocytes of immunocompetent mice (5). Consistent with a decreased latent reservoir (Fig. 1A andB), reactivation of WT MHV68 was attenuated in Cre-positive as compared to Cre-negative splenocytes (Fig. 1C). As expected (8), reactivation of the N36S mutant was attenuated as compared to WT MHV68 in control Cre-negative splenocytes (Fig. 1C). Thus, MHV68 orf36 and B cell-intrinsic IRF-1 expression were both required to support MHV68 reactivation from splenocytes.
MHV68 establishes a latent reservoir in secondary lymphoid organs, such as the spleen, and in the body cavities, such as the peritoneal cavity. MHV68 infection in the spleen is intimately tied to the differentiation of splenic B cells, which represent the B-2 lineage. B-2 B cells undergo development in the bone marrow with subsequent MHV68-driven differentiation through a T cell-dependent germinal center response, which allows for exponential increase in the splenic latent reservoir (11,25). In contrast, the majority of latent MHV68 in the peritoneal cavity is found in B-1 B cells, a distinct primordial B cell lineage in mice and humans (26). B-1 B cells develop in the embryonic yolk sac, self-renew, spontaneously produce self-and phospholipid-reactive antibodies, and primarily reside in body cavities, with limited circulation (reviewed in 27). When the MHV68 latent reservoir was examined in the peritoneal cavity, the frequency and percent of MHV68 DNA-positive peritoneal cells were decreased 2.7-fold in Cre-positive compared to Cre-negative mice infected with WT MHV68 (Fig. 1D andE, compare groups with open and filled circles). As expected, the lowest frequency of peritoneal cell infection was observed in Cre-negative mice infected with the N36S MHV68 mutant; the frequency was below the level of quantitation (Fig. 1D andE) (8,22). In contrast to the splenic latent reservoir, loss of B cell-intrinsic IRF-1 expression resulted in a partial rescue of the peritoneal latent reservoir of the N36S viral mutant (Fig. 1D andE, compare groups with open and closed squares). Thus, similar to that observed under conditions of global IRF-1 deficiency (22), B cell-intrinsic IRF-1 deficiency partially rescued the attenuated peritoneal latent reservoir of the N36S mutant.
The low frequency of ex vivo WT MHV68 reactivation from peritoneal cells trended toward further decrease in Cre-positive as compared to Cre-negative mice (Fig. 1F), consistent with the decreased peritoneal latent reservoir. Interestingly, despite partial rescue of the peritoneal latent reservoir of the N36S viral mutant in the Cre-positive mice, ex vivo reactivation of the N36S mutant remained very low to undetectable regardless of the Cre genotype (Fig. 1F). No persistent viral replication was observed in the spleen or peritoneal cells in any of the control or experimental groups (data not shown).
In summary, B cell-intrinsic IRF-1 expression supported the establishment of the splenic latent reservoir; however, this proviral mechanism was dependent on the expression of the conserved MHV68 protein kinase. Similarly, the proviral role of MHV68 orf36 during the establishment of the splenic latent reservoir was dependent on the expression of B cell-intrinsic IRF-1, highlighting a cooperative relationship in B-2 B cells. In contrast, B cell-intrinsic IRF-1 expression was partially responsible for the attenuated peritoneal latent reservoir of the N36S MHV68 mutant, suggesting an antagonistic relationship between IRF-1 and MHV68 orf36 in B-1 B cells.
## B cell-intrinsic IRF-1 expression supports the orf36-driven expansion of the germinal center response and generation of class-switched antiviral and self-reactive antibodies
Germinal center B cells host most of the MHV68 latent reservoirs at 16 days post-infec tion (25). The MHV68-driven expansion of germinal center B cells is dependent on CD4 T follicular helper cells (11,28). Having observed the requirement of MHV68 orf36 for the proviral effects of B cell-intrinsic IRF-1 in the establishment of the splenic latent reservoir, the germinal center response was examined next. Splenomegaly, as defined by the absolute number of nucleated cells per spleen, was decreased in WT MHV68-infec ted Cre-positive as compared to Cre-negative mice (Fig. 2A). In contrast, B cell-specific IRF-1 deficiency did not result in a further decrease of an already attenuated splenome galy observed in the N36S MHV68-infected mice (Fig. 2A). Consistent with decreased splenomegaly, the frequency and absolute number of germinal center B cells (Fig. 2B through D) and T follicular helper cells (Fig. 2E through G) were decreased in WT MHV68-infected Cre-positive compared to Cre-negative mice. Similarly, lack of MHV68 orf36 expression attenuated the germinal center response in control Cre-negative mice. However, B cell-intrinsic IRF-1 deficiency had no effect on the attenuated germinal center response driven by the N36S MHV68 mutant (Fig. 2B through G). Thus, like that observed for splenic latent reservoir, MHV68 orf36 and B cell-intrinsic IRF-1 expression served interdependent roles in supporting the MHV68-driven germinal center response.
In addition to physiological B cell differentiation that results in the generation of gammaherpesvirus-specific class-switched antibodies, EBV and MHV68 uniquely promote robust differentiation of B cells that produce class-switched antibodies against self-or foreign-species antigen. This nonphysiological B cell differentiation is thought to be proviral as EBV and MHV68 preferentially establish latency in B cells that do not encode a gammaherpesvirus-specific B cell receptor (29,30). Having observed decreased germinal center response, humoral parameters of MHV68-driven B cell differentiation were defined next. As previously shown, MHV68 orf36 expression did not affect the generation of class-switched anti-MHV68 antibodies (Fig. 2H andI) (8). As expected (23), B cell-specific IRF-1 deficiency resulted in decreased anti-MHV68 IgG titers following WT MHV68 infection. Interestingly, the anti-MHV68 IgG titers trended toward decreased in Cre-positive as compared to Cre-negative mice infected with the N36S mutant (Fig. 2H andI), although this difference did not reach statistical significance (P = 0.0546), suggesting that B cell-intrinsic IRF-1 expression but not MHV68 orf36 supports anti-MHV68 IgG response.
We showed that, despite having no effect on the generation of anti-MHV68 antibody, MHV68 orf36 supports differentiation of self-reactive B cells (8). The same phenotype was observed in the current study, as reflected by the decreased anti-double-stranded DNA (dsDNA) IgG titers in N36S-infected Cre-negative mice (Fig. 2J andK). Further decreases in anti-ds DNA IgG were observed in Cre-positive mice infected with either WT or N36S MHV68 (Fig. 2J andK). Thus, both MHV68 orf36 and B cell-intrinsic IRF-1 expression supported MHV68-driven differentiation of self-reactive B cells. To determine the extent to which MHV68 orf36 and B cell-intrinsic IRF-1 expression affect B cell proliferation, Ki67 expression was assessed. As expected, significantly more germinal center B cells expressed Ki67 and at a higher level, as compared to total splenic B cells (compare Fig. 3A andD, please note the difference in the X-axis scale as the same gating strategy for Ki-67 is used in both panels). Specifically, approximately 60% of germinal center B cells in the spleens of WT MHV68-infected Cre-negative mice expressed Ki67 at 16 days post-infection, consistent with the increased MHV68-driven germinal center response (Fig. 3A through C). The proportion and number of Ki67+ germinal center B cells were decreased in Cre-negative mice infected with the N36S viral mutant, indicating that MHV68 orf36 expression facilitates proliferation of germinal center B cells (Fig. 3A through C). Surprisingly, and in contrast to the well-established antiproliferative role of IRF-1 in cancer (32), B cell-specific IRF-1 deficiency resulted in decreased proportion of Ki-67+ germinal center B cells. This was observed in both WT MHV68-and N36S mutant-infected Cre-positive mice (Fig. 3A through C). Similar phenotypes were observed when the overall splenic B cell population was examined (Fig. 3D through F). Thus, MHV68 orf36 and B cell-intrinsic IRF-1 expression independ ently promoted MHV68-driven splenic B cell proliferation, including that of germinal center B cells.
## MHV68 orf36 and B cell-intrinsic IRF-1 expression independently support the MHV68-driven proliferation of splenic and germinal center B cells
## MHV68 orf36 and B cell-intrinsic IRF-1 function within the same pathway to attenuate apoptosis of germinal center B cells
In addition to robust proliferation, a significant proportion of germinal center B cells undergoes apoptosis, via a combination of intrinsic apoptotic pathways, due to genomic instability and/or the paucity of T cell-mediated survival signals, and extrinsic apoptotic stimuli, such as ligation of Fas, expressed by B cells, by FasL, expressed by T cells. To quantify germinal center B cell apoptosis, the combined enzymatic activity of caspase 3 and 7 was assessed by flow cytometry. Loss of MHV68 orf36 or B cell-intrinsic IRF-1 expression resulted in increased proportion of germinal center B cells with active caspases (Fig. 4A andB). However, the proportion of germinal center B cells with active caspases remained similarly elevated in mice infected with the N36S MHV68 mutant, regardless of the IRF-1 expression by B cells. Thus, MHV68 orf36 and B cell-intrinsic IRF-1 functioned in the same pathway(s) to promote survival of germinal center B cells during chronic infection.
We previously showed that in contrast to decreased germinal center response observed in MHV68-infected mice with B cell-intrinsic IRF-1 deficiency, germinal center B cells were not decreased during chronic LCMV infection (23). To assess whether B cell-intrinsic IRF-1 expression selectively attenuated germinal center B cell apoptosis during MHV68 infection, Cre-negative and Cre-positive mice were immunized with sheep red blood cells (SRBC) that stimulate robust germinal center responses (33). SRBC immunization induced a similar magnitude of germinal center B cell population as that observed at 16 days post-MHV68 infection (compare Fig. S1A; Fig. 2B through D), with the proportion and number of germinal center B cells similar in Cre-negative and Cre-positive SRBC-immunized mice (Fig. S1A). In contrast to that observed in WT MHV68-infected control spleens, where ~ 25% of germinal center B cells expressed active caspases 3/7 (Fig. 4), only ~7% of germinal center B cells expressed active caspases in Cre-negative SRBC-immunized animals (Fig. S1B). Importantly, caspase 3/7 activity in germinal center B cells was not increased in SRBC-immunized Cre-positive mice (Fig. S1B). Thus, B cell-intrinsic IRF-1 expression selectively attenuated caspase 3/7 activity in germinal center B cells during chronic MHV68 infection.
## Increased apoptosis of germinal center B cells in the absence of MHV68 orf36 or B cell-intrinsic IRF-1 expression is not accompanied by increased DNA damage response
Having observed increased activity of apoptotic caspases in the absence of MHV68 orf36 or B cell-intrinsic IRF-1 expression, we sought to identify the mechanism underlying increased apoptosis. Germinal center B cells are highly susceptible to both intrinsic and extrinsic apoptotic stimuli. For the former, increased proliferation is inherently associated with DNA damage, including double-stranded DNA breaks, the most catastrophic DNA lesions that activate DNA damage response. Double-stranded DNA breaks are marked by phosphorylation of serine 139 of histone variant H2AX (termed as γH2AX), with phosphorylation conferred by several cellular kinases, including ataxia-telangiectasia mutated (ATM) kinase. In addition to allowing recruitment of DNA repair machinery, signaling initiated downstream of γH2AX leads to p53-dependent cell cycle arrest and, eventually, apoptosis. Classically, IRF-1 cooperates with both ATM and p53 to support cell cycle arrest (34,35). Interestingly, MHV68 orf36 induces γH2AX directly and indirectly, the latter via activation of ATM, to support MHV68 lytic replication and establishment of chronic infection in a cell type-dependent manner (26,(36)(37)(38)(39).
To determine the extent to which DNA damage response activity in germinal center B cells is modulated during chronic MHV68 infection, we optimized γH2AX detection by flow cytometry. Optimization was performed using UV-irradiated or untreated splenocytes from naïve mice (Fig. S2). γH2AX levels in UV-irradiated B cells were used to establish the flow cytometry gating strategy to identify B cell populations with intermediate (γH2AX int) and high (γH2AX hi) γH2AX levels (Fig. S2). This gating strategy was applied to quantify γH2AX levels in germinal center B cells analyzed directly ex vivo at 16 days post-infection (Fig. 5).
Approximately 70% of all germinal center B cells in Cre-negative mice infected with WT MHV68 displayed intermediate γH2AX levels (Fig. 5B), consistent with ~60% of proliferating, Ki67 +germinal center B cells observed under the same conditions (Fig. 3B). Lack of MHV68 orf36 expression did not affect the frequency and number of germinal center B cells with intermediate γH2AX levels (Fig. 5A andB). Surprisingly, and in contrast to the classical role of IRF-1 in supporting DNA repair (40), the frequency of germinal center B cells with intermediate γH2AX levels was decreased in mice with B cell-specific IRF-1 deficiency regardless of the MHV68 genotype (Fig. 5B). Significantly fewer (~5%-10%) germinal center B cells displayed high γH2AX levels, with decreased proportion of γH2AX high germinal center B cells observed in Cre-positive mice (Fig. 5C). Finally, the median fluorescence intensity of the γH2AX signal in the overall germinal center B cell population was also reduced in mice with B cell-specific IRF-1 deficiency. Thus, increased apoptosis of germinal center B cells in the absence of MHV68 orf36 or B cell-intrinsic IRF-1 expression was not accompanied by the increase in the DNA damage response.
## MHV68 orf36 or B cell-intrinsic IRF-1 expression does not alter Fas or Fas ligand (FasL) expression on germinal center B cells during chronic infection
In the absence of increase in the DNA damage response in orf36-or IRF-1-deficient conditions, extrinsic apoptotic stimuli were examined next. FasL is expressed by most activated and differentiated CD4 T cells, including germinal center-localized CD4 T follicular helper cells, with corresponding cytotoxic activity (41). Germinal center B cells are exquisitely susceptible to Fas-mediated apoptosis; in fact, high expression of CD95 (Fas) is a defining cell surface marker for immunophenotyping of germinal center B cells (Fig. 2B). However, neither MHV68 orf36 nor B cell-intrinsic IRF-1 expression altered Fas protein levels on the cell surface of germinal center B cells at 16 days post-infection (Fig. 6A andB). When FasL expression was examined, both the proportion of T follicular helper cells with high FasL expression and the per cell FasL protein expression levels were decreased in infected Cre-positive mice, regardless of the infecting virus (Fig. 6C through E). This was surprising as only B cells are genetically targeted in this mouse model. However, the observed decreased FasL expression by T follicular helper cells in the absence of B cell-intrinsic IRF-1 expression did not explain increased apoptosis of germinal center B cells under the same conditions (Fig. 4). Thus, expression of FasL by germinal center B cells was examined next. In contrast to T cells, expression of FasL by B cells, includ ing germinal center B cells, is poorly defined, including functionally. Interestingly, FasL expression by germinal center B cells increased following MHV68 infection, albeit to a lesser extent as compared to T follicular helper cells (Fig. 6C vs. F, note the different X-axis scale and gating strategy). However, neither MHV68 orf36 nor B cell-intrinsic IRF-1 expression affected FasL cell surface protein levels or the proportion of FasL-expressing germinal center B cell population (Fig. 6F through H). Thus, neither B cell-intrinsic IRF-1 deficiency nor MHV68 orf36 expression affected Fas/FasL levels expressed by germinal center B cells.
## B cell intrinsic IRF-1 but not MHV68 orf36 supports MHCII expression on germinal center B cells during chronic MHV68 infection
Under physiological conditions, MHCII-dependent antigen presentation to CD4 T follicular helper cells is critical for the survival and further differentiation of germinal center B cells (42). In the context of MHV68 infection, both B cell-intrinsic MHCII expression and T follicular helper cells are necessary to support MHV68-driven splenic B cell differentiation (11,28,43). Because IFNγ signaling significantly increases MHC-II expression via the IRF-1/CIITA axis (44), with peak serum IFNγ levels observed at 16 days post-MHV68 infection (45), the expression of MHC-II by germinal center B cells was measured next. Both per cell protein levels of MHC-II and proportion of germinal center B cells with high MHC-II expression were decreased in mice with B cell-specific IRF-1 deficiency (Fig. 7A through C). In contrast, lack of orf36 did not affect MHC-II expression by germinal center B cells (Fig. 7A through C). The differences in MHC-II expression observed in germinal center B cells were mirrored by the differences in the serum IFNγ levels, although the IRF-1-dependent difference in the N36S-infected mice did not reach statistical significance (Fig. 7D, P = 0.2086). Thus, B cell-intrinsic IRF-1 expression supported germinal center B cell MHC-II expression during chronic MHV68 infection, independent of MHV68 protein kinase.
## DISCUSSION
Gammaherpesviruses are ubiquitous, oncogenic pathogens that, unlike other viral families, specifically target and manipulate B cells to establish lifelong infection. Gammaherpesvirus-driven germinal center-based B cell differentiation supports the establishment of life-long infection in memory B cells and seeds viral lymphomagene sis. We previously demonstrated that conserved gammaherpesvirus protein kinase and global expression of host IRF-1 exert opposite and antagonistic effects on gammaherpes virus-driven germinal center response and latent viral reservoir during chronic infection (8,10,22). Surprisingly, using a mouse model of B cell-intrinsic IRF-1 deficiency, we also showed that IRF-1 expression by B cells is proviral and supports the establishment of chronic gammaherpesvirus infection and germinal center response (23). Thus, the current study tested an intriguing hypothesis that B cell-intrinsic IRF-1 may be subverted by MHV68 using the conserved gammaherpesvirus protein kinase. Indeed, we found that the proviral functions of B cell-intrinsic IRF-1 in supporting germinal center response and MHV68 splenic latent reservoir required the expression of MHV68 orf36. While IRF-1 and orf36 expression independently supported proliferation of germinal center B cells during chronic infection, unexpectedly, IRF-1 attenuated apoptosis of germinal center B cells in an orf36-dependent manner. The cooperation between MHV68 orf36 and B cell-intrinsic IRF-1 in germinal center B cell survival was not explained by corresponding changes in the DNA damage response or Fas/FasL expression. Importantly, B cell-intrinsic IRF-1 expression, but not MHV68 orf36, supported optimal expression of MHC-II by germinal center B cells, a critical requirement to receive survival and proliferation stimuli from T follicular helper cells and promote MHV68-driven B-cell differentiation (11,42). Overall, the outcomes of the current study demonstrate the important role of the host cell type in defining functional consequences of viral-host interactions. We previously showed that neither global nor B cell-specific IRF-1 expression affected B cell differentiation during chronic infection with an unrelated virus (LCMV). Thus, the novel role(s) of IRF-1 and/or IRF-1 cooperation with MHV68-specific viral func tions/proteins may be responsible for IRF-1-dependent host phenotypes that selectively manifested during MHV68 infection. The latter is likely as we show that several viral and host IRF-1-dependent phenotypes required the expression of MHV68 orf36 (as summar ized in Table 1). Of these phenotypes, B cell-intrinsic IRF-1 expression promoted survival of germinal center B cells, but only when MHV68 orf36 expression was preserved. This was not observed in the context of SRBC immunization, where the proportion of apoptotic germinal center B cells trended toward decreased in mice with IRF-1-deficient B cells. The observed apoptosis phenotypes in immunized mice with IRF-1-deficient B cells are consistent with the classical role of IRF-1 in promoting p53-dependent apoptosis and cell cycle arrest. In contrast, the observed increase in apoptosis of IRF-1-deficient germinal center B cells during chronic MHV68 infection is opposite to the classical IRF-1 functions.
The current study probed potential IRF-1-dependent mechanisms that would support IRF-1-dependent germinal center B cell survival during MHV68 infection. Having ruled out DNA damage response and Fas/FasL system, decreased MHC-II expression by germinal center B cells was observed in infected mice with IRF-1-deficient B cells. Concurrently, there was a statistically significant or a trending decrease in serum IFNγ levels of infected Cre-positive mice (Fig. 7D). This decreased MHC-II expression offers one possible scenario by which the classical IFNγ/IRF-1/CIITA/MHC-II axis facilitates optimal interactions between germinal center B cells and T follicular helper cells, with the latter delivering survival and proliferation stimuli to B cells. Interestingly, SRBC immunization preferentially induces the type I IFN response in vivo (46), which, unlike type II IFN, plays a minimal role in stimulating MHC-II expression. Thus, it is conceivable that the lack of IRF-1-dependent germinal center phenotype following SRBC immunization (Fig. S1) is due to selective induction of type I IFN in immunized mice. Importantly, the regulation of MHC-II expression was not affected by the orf36 MHV68 genotype, failing to explain the observed dependence of IRF-1-mediated germinal center B cell survival on MHV68 orf36. Thus, future studies are needed to tease out the involved mechanisms.
IRF-1 can interact with multiple transcription factors, both ubiquitous and expressed in a cell type-dependent or inducible manner (47). Thus, IRF-1 target genes have to be
## Viral or host parameter Independent roles Interdependent roles
Splenic latent reservoir (Fig. 1) Proviral roles of MHV68 orf36 and B cell-intrinsic IRF-1 require the expression of both proteins Peritoneal latent reservoir (Fig. 1) Antagonism between IRF-1 and MHV68 orf36 MHV68-driven germinal center response (Fig. 2 defined in individual cell types and under physiologically relevant conditions. A single published study from the Lund group comprehensively defined IRF-1-dependent gene expression in transitional, follicular, and marginal zone B cell subsets of unmanipulated mice (48). Gene signature of IRF-1-deficient transitional B cells was indicative of an increased activity of NF-kB and type I IFN pathways (48). However, NF-kB is a critical host factor usurped by gammaherpesviruses, including MHV68, to establish latent infection (16,49,50). Similarly, we showed that B cell-intrinsic expression of STAT1, a transcription factor required for classical IFN antiviral responses, supports the establishment of latent MHV68 reservoir in the spleen (19). While it is possible that exaggerated NF-kB and IFN activity in IRF-1-deficient germinal center B cells is antiviral, this remains to be confirmed in the context of MHV68 infection, along with comprehensive analyses of IRF-1 target genes in germinal center B cells.
We have also, for the first time, defined the role of MHV68 orf36 in the biology of germinal center B cells during chronic infection. The current study demonstrates that MHV68 orf36 expression promotes the proliferation and survival of germinal center B cells during chronic infection, the former independent of and the latter in collaboration with IRF-1 expressed by B cells. Our observations suggest that orf36 is expressed and functions in germinal center B cells. This hypothesis is highly provocative as gamma herpesvirus protein kinases are classically defined as lytic cycle-associated proteins, in contrast to the tightly latent infection of germinal center B cells. Importantly, a recent publication from the Krug group profiled MHV68 gene expression by bulk RNA sequencing of sorted infected germinal center B cells and, indeed, detected orf36 transcript, albeit at low levels (51). In the future, it will be important to develop new tools to define the timing of MHV68 orf36 expression in specific germinal center B cell subsets in an intact animal. We previously showed that MHV68 orf36 antagonizes STAT1 antiviral function in myeloid cells to support MHV68 passage to splenic B cells (20). Thus, in a non-exclusive scenario, orf36 expression in specific infected myeloid cell types may promote the differentiation of T follicular helper cells and subsequent germinal center responses in a manner that subverts IRF-1-dependent processes.
The results of the current study underscore the complexity of the interactions between classically antiviral transcription factors and viral-encoded proteins that manipulate the physiological immune response. IRF-1 is evolutionarily conserved, representing an ancient component of the innate immune response. Likewise, gamma herpesviruses are ancient viruses, having evolved alongside the development of the modern immune response. As gammaherpesviruses, such as EBV, rely on establishing infection in host B cells, it is no surprise that these viruses are masters in manipulating and usurping the host immune response. Our study suggests that gammaherpesviruses have evolved a multifunctional conserved protein kinase to usurp the functions of the canonically antiviral transcription factor to promote the establishment of the viral latent reservoir in splenic B cells. In contrast, results of the current study confirm the antago nistic role of MHV68 orf36 and B cell-intrinsic IRF-1 in the context of peritoneal cavity infection. Thus, not only are viral-host interactions modified in a cell type-dependent manner, the B cell lineage (B-2 vs B-1) also plays a role. This is not an unprecedented finding as we showed that B cell-intrinsic STAT1 and IFNAR1 expression exert opposite effects on the establishment of MHV68 latent reservoir in splenic vs peritoneal B cells (19). IRF-1 expression is increased downstream of IFN receptors, and IRF-1 can complex with STAT1 to regulate gene expression (47). Thus, it is possible that the viral and host phenotypes driven by B cell-intrinsic IRF-1 deficiency in this study reflect the IRF-1-/ STAT1-dependent changes in B cell-intrinsic gene expression, to be defined in the future.
## MATERIALS AND METHODS
## Animal studies
All mice were housed and bred in a specific-pathogen-free facility at MCW. Cd19 cre/ wt IRF-1 loxP/loxP mice were previously described and validated (23). The presence of the conditional IRF-1 allele was detected using TGTTCTAGCAAGTTCTCAGAGG (forward) and TGGTACCCTGACTCACAACTG (reverse) primers. The presence of the CD19 Cre recombi nase allele was detected using ACGTACTGACGGTGGAGAA (forward) and CAAAAATCCCTT CCAGGGCG (reverse) primers.
## Virus infection
Virus stock titers were determined by plaque assay on NIH 3T12 cells. Infections with the N36S MHV68 mutant were controlled by the parental virus retaining a single LoxP site (referred to as wild-type in the figures and text) (24). Mice between the ages of 8 and 10 weeks were intranasally inoculated (15 µL/mouse) with 10,000 PFU of virus diluted in sterile serum-free Dulbecco's modified Eagle's medium (Corning, Tewksbury, MA) or sterile carrier (mock) under light anesthesia.
## SRBC inoculation
Mice between the ages of 8 and 10 weeks were inoculated via intraperitoneal injection with 300 uL of fresh sheep red blood cells (Colorado Serum Company, Denver, CO). Splenocytes from individual mice were analyzed at 9 days post-immunization.
## Limiting dilution assays
The frequency of MHV68 DNA+ cells was determined as previously described (52). Briefly, splenocytes or peritoneal cells were pooled from each experimental group (3-5 mice/group), and six 3-fold dilutions were made on a background of NIH 3T12 cells. Dilutions were subjected to a nested PCR (12 replicates/dilution) using primers designed against the MHV68 genome (outer forward: 5′-GAGATCTGTACTCAGGCACCT GT-3′; outer reverse: 5′-GGATTTCTTGACAGCTCCCTGT-3′; inner forward: 5′-TGTCAGCTG TTGTTGCTCCT-3′; inner reverse: 5′-CTCCGTCAGGATAACAACGTCT-3′). To determine the frequency of ex vivo MHV68 reactivation, 2-fold serial dilutions of pooled splenocytes or peritoneal cells were plated onto a monolayer of C57BL6/J mouse embryonic fibroblasts (MEFs) at 24 replicates per dilution and incubated at 37°C. To control for preformed virus, 2-fold serial dilutions of mechanically disrupted cells were plated on MEFs. Viral reactivation, as indicated by cytopathic clearing of MEFs, was assessed on day 21 of culture. The frequency of MHV68 DNA+ cells or ex vivo reactivation is determined by Poisson distribution.
## Flow cytometry
Single-cell suspensions of splenocytes were prepared in fluorescence-activated cell sorter (FACS) buffer (phosphate-buffered saline, 2% fetal bovine serum); 2 × 10 6 cells were treated with Fc block prior to extracellular staining with optimized antibody concentrations for 30 minutes on ice. For intracellular detection of γH2AX, cell perme abilization was performed using FOXP3 Fix/Perm Buffer Set (cat: 421403; BioLegend (San Diego, CA)) followed by 1 hour incubation with optimized antibody concentration at room temperature. Ki67 staining followed the BioLegend Ki-67 Flow Cytometry Staining Protocol, where cells were permeabilized by incubating at -20°C with 70% EtOH for 2 hours, followed by a 30 minute incubation with an optimized antibody concentration at room temperature. Data were acquired using Celesta flow cytometer (BD Biosciences, Franklin Lakes, NJ) and analyzed using FlowJo software (BD Biosciences, Franklin Lakes, NJ). The following list of antibodies used in this study were purchased from BioLegend (San Diego, CA): CD19-Bv421 (cat. 152415), B220-PE/Cy7 (cat. 103222), GL7-PerCP/Cy5.5 (cat. 144609), GL7-FITC (cat. 144605), CD95-PECF594 (cat. 562499), CD3-Bv421 (cat. 100531), CD4-FITC (cat. 100406), CXCR5-PECF594 (cat. 145522), PD-1-Bv605 (cat. 135220), MHCII-Bv605 (cat. 107639); Invitrogen (Carlsbad, CA): Ki67-PE (cat. 12-5698-82), FasL-APC (cat. 17-5911-82), Caspase-3/-7 Green Flow Cytometry Assay Kit (cat. C10427); or Cell Signaling Technology (Danvers, MA): yH2AX-Bv421 (cat. 9718).
## ELISA
ELISA was performed to measure serum IgM, IgG, MHV68-specific IgG, and anti-doublestranded DNA IgG, as previously described (22).
## Statistical analyses
Statistical analyses were performed using Student t-test when comparing two groups, and one-way ANOVA with Tukey's post hoc test when comparing more than two groups (Prism, GraphPad Software, Inc.).
## References
1. Flaño, Kim, Woodland et al. (2002) "γ-herpesvirus latency is preferentially maintained in splenic germinal center and memory B cells" *J Exp Med*
2. Roughan, Thorley-Lawson, Da (2009) "The intersection of Epstein-Barr virus with the germinal center" *J Virol*
3. Thorley-Lawson, Da (2001) "Epstein-Barr virus: exploiting the immune system" *Nat Rev Immunol*
4. Laichalk, Thorley-Lawson, Da (2005) "Terminal differentiation into plasma cells initiates the replicative cycle of Epstein-Barr virus in vivo" *J Virol*
5. Liang, Collins, Mendel et al. (2009) "Gamma herpesvirus-driven plasma cell differentiation regulates virus reactiva tion from latently infected B lymphocytes" *PLoS Pathog*
6. Jondle, Johnson, Aurubin et al. (2021) "Gammaherpesvirus usurps host IL-17 signaling to support the establishment of chronic infection" *mBio*
7. Lee, Cullum, Stoltz et al. (2021) "Mouse homologue of human HLA-DO does not preempt autoimmunity but controls murine gammaherpesvirus MHV68" *J Immunol*
8. Darrah, Jondle, Johnson et al. (2019) "Conserved gammaherpesvirus protein kinase selectively promotes irrelevant B cell responses" *J Virol*
9. Johnson, Lange, Jondle et al. (2019) "B cell-intrinsic SHP1 expression promotes the gammaherpesvirus-driven germinal center response and the establishment of chronic infection" *J Virol*
10. Mboko, Olteanu, Ray et al. (2015) "Tumor suppressor IRF-1 counteracts germinal center reaction driven by a cancer-associated gammaherpesvirus" *J Virol*
11. Collins, Speck (2014) "Expansion of murine gammaherpesvirus latently infected B cells requires T follicular help" *PLoS Pathog*
12. Terrell, Speck (2017) "Murine gammaherpesvirus M2 antigen modulates splenic B cell activation and terminal differentiation in vivo" *PLoS Pathog*
13. Wang, Feswick, Apostolou et al. (2022) "Gammaherpesvirus-mediated repression reveals EWSR1 to be a negative regulator of B cell responses" *Proc Natl Acad Sci*
14. Wang, Feldman, Bullard et al. (2019) "A gammaherpesvi rus microRNA targets EWSR1 (Ewing sarcoma breakpoint region 1) in vivo to promote latent infection of germinal center B cells"
15. Wang, Manzi, Feswick et al. (2023) "B cell expression of E3 ubiquitin ligase Cul4b promotes chronic gammaherpesvirus infection in vivo" *J Virol*
16. Cieniewicz, Kirillov, Daher et al. (2022) "IKKα-mediated noncanonical NF-κB signaling is required to support murine gammaherpesvirus 68 latency in vivo" *J Virol*
17. Rodrigues, Popov, Kaye et al. (2013) "Stabilization of Myc through heterotypic poly-ubiquitination by mLANA is critical for γherpesvirus lymphoproliferation" *PLoS Pathog*
18. Kim, Burkum, Cookenham et al. (2007) "Perturbation of B cell activation in SLAM-associated protein-deficient mice is associated with changes in gammaherpesvirus latency reservoirs" *J Immunol*
19. Johansen, Schmalzriedt, Avila et al. (2024) "Combination of proviral and antiviral roles of B cell-intrinsic STAT1 expression defines parameters of chronic gammaherpesvirus infection" *mBio*
20. Sylvester, Jondle, Stoltz et al. (2021) "Conserved gammaherpesvirus protein kinase counters the antiviral effects of myeloid cell-specific STAT1 expression to promote the establishment of splenic B cell latency" *J Virol*
21. Anders, Montgomery, Montgomery et al. (2018) "Human herpesvirus-encoded kinase induces B cell lymphomas in vivo" *J Clin Invest*
22. Jondle, Sylvester, Schmalzriedt et al. (2022) "The antagonism between the murine gammaherpesvirus protein kinase and global interferon regulatory factor 1 expression shapes the establishment of chronic infection" *J Virol*
24. Jondle, Johnson, Uitenbroek et al. (2020) "B cell-intrinsic expression of interferon regulatory factor 1 supports chronic murine gammaherpesvirus 68 infection" *J Virol*
25. Hwang, Kim, Flano et al. (2009) "Conserved herpesviral kinase promotes viral persistence by inhibiting the IRF-3-mediated type I interferon response" *Cell Host Microbe*
26. Collins, Speck (2012) "Tracking murine gammaherpesvirus 68 infection of germinal center B cells in vivo" *PLoS One*
27. Darrah, Kulinski, Mboko et al. (2017) "B cell-specific ATM expression promotes chronic gammaherpesvirus infection" *J Virol*
28. Rothstein, Quach (2015) "The human counterpart of mouse B-1 cells" *Ann N Y Acad Sci*
29. Collins, Speck (2015) "Interleukin 21 signaling in B cells is required for efficient establishment of murine gammaherpesvirus latency" *PLoS Pathog*
30. Decalf, Godinho-Silva, Fontinha et al. (2014) "Establishment of murine gammaherpesvirus latency in B cells is not a stochastic event" *PLoS Pathog*
31. Tracy, Kakalacheva, Lünemann et al. (2012) "Persistence of Epstein-Barr virus in selfreactive memory B cells" *J Virol*
32. Victora, Schwickert, Fooksman et al. (2010) "Germinal center dynamics revealed by multiphoton microscopy with a photoactivatable fluorescent reporter" *Cell*
33. Armstrong, Stang, Liu et al. (2012) "Interferon regulatory factor 1 (IRF-1) induces p21(WAF1/CIP1) dependent cell cycle arrest and p21(WAF1/CIP1) independent modulation of survivin in cancer cells" *Cancer Lett*
34. Ladics (2007) "Primary immune response to sheep red blood cells (SRBC) as the conventional T-cell dependent antibody response (TDAR) test" *J Immunotoxicol*
35. (2025) *Full-Length Text Journal of Virology*
36. Pamment, Ramsay, Kelleher et al. (2002) "Regulation of the IRF-1 tumour modifier during the response to genotoxic stress involves an ATM-dependent signalling pathway" *Oncogene*
37. Tanaka, Ishihara, Lamphier et al. (1996) "Cooperation of the tumour suppressors IRF-1 and p53 in response to DNA damage" *Nature*
38. Darrah, Stoltz, Ledwith et al. (2017) "ATM supports gammaherpesvirus replication by attenuating type I interferon pathway" *Virology (Auckl)*
39. Kulinski, Darrah, Broniowska et al. (2015) "ATM facilitates mouse gammaherpesvirus reactivation from myeloid cells during chronic infection" *Virology (Auckl)*
40. Tarakanova, Stanitsa, Leonardo et al. (2010) "Conserved gammaherpesvirus kinase and histone variant H2AX facilitate gammaherpesvirus latency in vivo" *Virology (Auckl)*
41. Tarakanova, Leung-Pineda, Hwang et al. (2007) "Gammaherpesvirus kinase actively initiates a DNA damage response by inducing phosphorylation of H2AX to foster viral replication" *Cell Host Microbe*
42. Prost, Bellamy, Cunningham et al. (1998) "Altered DNA repair and dysregulation of p53 in IRF-1 null hepatocytes" *FASEB J*
43. Kotov, Kotov, Goldberg et al. (2018) "Many Th cell subsets have Fas ligand-dependent cytotoxic potential" *J Immunol*
44. Barnett, Simkins, Barnett et al. (2014) "B cell antigen presentation in the initiation of follicular helper T cell and germinal center differentiation" *J Immunol*
45. Sangster, Topham, Costa et al. (2000) "Analysis of the virus-specific and nonspecific B cell response to a persistent B-lymphotropic gammaherpesvirus" *J Immunol*
46. Hobart, Ramassar, Goes et al. (1997) "IFN regulatory factor-1 plays a central role in the regulation of the expression of class I and II MHC genes in vivo" *J Immunol*
47. Lange, Jondle, Darrah et al. (2019) "LXR alpha restricts gammaherpesvirus reactivation from latently-infected peritoneal cells" *J Virol*
48. Loetsch, Warren, Laskowski et al. (2017) "Cytosolic recognition of RNA drives the immune response to heterolo gous erythrocytes" *Cell Rep*
49. Langlais, Barreiro, Gros (2016) "The macrophage IRF8/IRF1 regulome is required for protection against infections and is associated with chronic inflammation" *J Exp Med*
50. Peel, Owiredu, Rosenberg et al. (2024) "The marginal zone B cell compartment and T cell-independent antibody responses are supported by B cell intrinsic expression of IRF-1" *J Immunol*
51. Krug, Collins, Gargano et al. (2009) "NF-κB p50 plays distinct roles in the establishment and control of murine gammaherpes virus 68 latency" *J Virol*
52. Krug, Moser, Dickerson et al. (2007) "Inhibition of NF-κB activation in vivo impairs establishment of gammaherpesvirus latency" *PLoS Pathog*
53. Hogan, Owens, Reynoso et al. (2023) "B cellintrinsic STAT3-mediated support of latency and interferon suppression during murine gammaherpesvirus 68 infection revealed through an in vivo competition model"
54. Weck, Barkon, Yoo et al. (1996) "Mature B cells are required for acute splenic infection, but not for establishment of latency, by murine gammaherpesvirus 68" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12702710&blobtype=pdf | # Commentary: Constructing the optimal experimental autoimmune thyroiditis mouse model using porcine thyroglobulin
Min Li, Yingchun Zhou, Ming Cai, Rajna Minic, Suresh Kari
## Introduction
The study by Liu et al., titled "Constructing the optimal experimental autoimmune thyroiditis mouse model using porcine thyroglobulin," published in Frontiers in Immunology, provides a systematic evaluation of immunization strategies for inducing autoimmune thyroiditis (AIT) in NOD/LtJ mice (1). By comparing different antigen doses, immunization frequencies, and injection routes, the authors identify a high-dose (200 µg pTg) and high-frequency (three immunizations) protocol as optimal for modeling AIT, with tail vein injection favoring antibody production and subcutaneous injection promoting stronger histological inflammation. This work offers valuable practical guidance for researchers aiming to establish reproducible AIT models. However, several methodological and translational aspects merit further considerations.
To fully contextualize the work by Liu et al., it is important to consider the historical development of EAT models. The foundational studies, notably by Dr. Noel Rose and colleagues, successfully induced thyroiditis using heterologous thyroglobulin, a strategy that Liu et al. now seek to optimize (2). A significant subsequent advancement came from the work of Dr. Yi-chi Kong's lab, which emphasized the critical role of self-antigen (murine thyroglobulin) in conjunction with adjuvant to break tolerance, more closely mimicking the breach of self-tolerance in human disease (3)(4)(5). Furthermore, it is established that regulatory T cells (Tregs) are pivotal in maintaining tolerance, as their depletion can trigger autoimmune thyroiditis even without adjuvant. The genetic underpinnings of AIT are also well-documented; in humans, specific HLA haplotypes confer susceptibility, which is reflected in mouse models through the use of H-2K haplotype (6). Acknowledging this rich historical and mechanistic landscape allows for a more nuanced appreciation of the model presented by Liu et al. and its position within the ongoing quest to recapitulate human AIT.
## Subsections relevant for the subject
First, the use of NOD/LtJ mice as an alternative to the less accessible NOD.H-2h4 strain is pragmatically justified and enhances model accessibility. However, the genetic and immunological differences between these strains-particularly in MHC haplotype and spontaneous vs. induced disease onset-may influence translational relevance to human AIT. Future studies should include comparative transcriptomic or proteomic analyses to clarify strain-specific immune phenotypes and their alignment with human disease (7,8).
Second, the comprehensive multi-parameter assessmentincluding histopathology, serum antibodies, cytokines, and local immune cell infiltration-strengthens the model's validity. The incorporation of multiplex immunofluorescence and immunohistochemistry for Th17/Treg balance and inflammasome markers (NLRP3, Caspase-1) is particularly commendable. Nevertheless, the absence of B-cell and follicular helper T-cell (Tfh) analysis represents a significant opportunity for deeper investigation. As the reviewer rightly highlights, this is a critical aspect. Elaborating further, B cells are not only precursors to autoantibody-producing plasma cells but also function as antigenpresenting cells and regulators of T-cell responses in AIT. Similarly, Tfh cells, located in B-cell follicles, are specialized in providing help for B-cell affinity maturation and antibody class switching. Their coordinated action is pivotal for the development of tertiary lymphoid structures often observed in chronic autoimmune thyroiditis. Therefore, quantifying B-cell and Tfh infiltration and their spatial organization within the thyroid would substantially enhance our understanding of the humoral immune mechanisms at play in this model (9,10). Furthermore, enhancing the histopathological analysis would strengthen the model's characterization. The study's iconography primarily provides lowmagnification overviews of thyroid inflammation. While useful for assessing the overall inflammatory area, higher-magnification images are crucial for two key reasons: first, to better characterize the specific types of immune cells within the infiltrate, and second, to reliably distinguish genuine inflammatory foci from ectopic thymic tissue, a known histological feature in NOD mice that can be mistaken for lymphocytic infiltration. Such detailed microscopy would provide more definitive evidence of autoimmune pathogenesis and improve the accuracy of histological scoring.
Third, the study highlights the superiority of triple immunization over double immunization in inducing severe thyroiditis, which aligns with immune memory principles. We agree with the authors' own recognition that the lack of longitudinal tracking beyond the acute phase (4 weeks postimmunization) is a limitation of their study. As our commentary and the reviewers of the original article suggest, this is a highly relevant point for the field. Extending the observation period to include time-series assessments at 8-12 weeks would be invaluable to model the chronicity of human AIT, monitor potential disease progression or remission, and ultimately allow for the evaluation of lasting therapeutic interventions (11).
Fourth, while the tail vein method enhanced antibody production and NLRP3 activation, its slightly lower inflammation scores compared to subcutaneous injection suggest routedependent immune polarization. This observed route-dependent immune polarization indeed merits deeper mechanistic inquiry. To truly dissect the underlying mechanisms, future studies could employ techniques such as in vivo cell tracking of adoptively transferred antigen-pulsed dendritic cells to compare their trafficking to the spleen (systemic immunity) versus draining lymph nodes (local immunity) following IV or SC injection. Additionally, detailed immunophenotyping of the resulting immune responses in these lymphoid organs and the thyroid itself could reveal differences in T-cell polarization, germinal center formation, and the establishment of local versus systemic immune memory, providing a clearer rationale for selecting one injection route over the other based on the specific research objectives. Additionally, the use of LPS in the IV protocol may introduce systemic inflammation confounding thyroid-specific responses (12).
Fifth, the study appropriately acknowledges the sex-specific limitation of using an all-female cohort. As the reviewer notes, this is a useful starting point for further investigation. Future work should systematically compare both male and female mice to elucidate sex differences in immune response kinetics and disease severity. This approach could reveal crucial hormonal or genetic modifiers of AIT. Furthermore, integrating sex as a biological variable into more complex models, such as those combining genetic modifications with environmental factors like iodine supplementation, could powerfully recapitulate the heterogeneity seen in the human AIT patient population (13).
## Discussion
This study delivers a rigorously optimized protocol for inducing AIT in NOD/LtJ mice, balancing pathological severity with operational feasibility. The integration of immunological and histopathological endpoints provides a robust framework for model validation. However, the translational impact would be strengthened by including human thyroid tissue validations, extending observation to chronic phases, and incorporating B-cell and Tfh analyses. Furthermore, exploring combinatorial modelse.g., by integrating genetic predispositions with environmental triggers like iodine supplementation and by considering sex as a key biological variable-could better recapitulate the complex heterogeneity of human AIT. By providing a rigorously optimized protocol within the established framework of heterologous antigeninduced EAT, Liu et al. offer a valuable and accessible resource for accelerating preclinical research in autoimmune thyroiditis.
## References
1. Liu, Zhang, Meng et al. (2025) "Constructing the optimal experimental autoimmune thyroiditis mouse model using porcine thyroglobulin" *Front Immunol*
2. Wang, Jiang, Xu et al. (2023) "Selenium regulates T cell differentiation in experimental autoimmune thyroiditis in mice" *Int Immunopharmacol*
3. Elrehewy, Kong, Giraldo et al. (1981) "Syngeneic thyroglobulin is immunogenic in good responder mice" *Eur J Immunol*
4. Morris, Brown, Kong (2009) "Naturally-existing CD4(+)CD25(+)Foxp3(+) regulatory T cells are required for tolerance to experimental autoimmune thyroiditis induced by either exogenous or endogenous autoantigen" *J Autoimmun*
5. Rose (2011) "The genetics of autoimmune thyroiditis: the first decade" *J Autoimmun*
6. Weatherall, Sarvetnick, Shizuru (1992) "Genetic control of diabetes mellitus" *Diabetologia*
7. Aubin, Lombard-Vadnais, Aliesky et al. (2022) "The NOD mouse beyond autoimmune diabetes" *Front Immunol*
8. Braley-Mullen, Yu (1950) "Early requirement for B cells for development of spontaneous autoimmune thyroiditis in NOD.H-2h4 mice" *J Immunol*
9. Ippolito, Dalmazi, Pani et al. (2021) "Distinct cytokine signatures in thyroiditis induced by PD-1 or CTLA-4 blockade: insights from a new mouse model" *Thyroid: Off J Am Thyroid Assoc*
10. Qin, Zhao, Wang et al. (2020) "Roles of endogenous IL-10 and IL-10-competent and CD5+ B cells in autoimmune thyroiditis in NOD.H-2h4 mice" *Endocrinology*
11. Nagayama, Horie, Saitoh et al. (2007) "CD4+CD25+ naturally occurring regulatory T cells and not lymphopenia play a role in the pathogenesis of iodide-induced autoimmune thyroiditis in NOD-H2h4 mice" *J Autoimmun*
12. Gallay, Barras, Tobias et al. (1994) "Lipopolysaccharide (LPS)-binding protein in human serum determines the tumor necrosis factor response of monocytes to LPS" *J Infect Dis*
13. Hu, Chen, Shen et al. (2022) "Global prevalence and epidemiological trends of Hashimoto's thyroiditis in adults: A systematic review and meta-analysis" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12451637&blobtype=pdf | # Summary of taxonomy changes ratified by the International Committee on Taxonomy of Viruses (ICTV) from the Animal dsRNA and ssRNA(-) Viruses Subcommittee, 2025
Holly Hughes, Matthew Ballinger, Yiming Bao, Nicolas Bejerman, Kim Blasdell, Thomas Briese, Julia Brignone, Jean Carrera, Lander De Coninck, William Marciel De Souza, Humberto Debat, Ralf Dietzgen, Ralf Dürrwald, Mert Erdin, Anthony Fooks, Kristian Forbes, Juliana Freitas-Astúa, Jorge Garcia, Jemma Geoghegan, Rebecca Grimwood, Masayuki Horie, Timothy Hyndman, Reimar Johne, John Klena, Hideki Kondo, Eugene Koonin, Alexei Kostygov, Mart Krupovic, Jens Kuhn, Michael Letko, Jun-Min Li, Yiyun Liu, Maria Martin, Nathaniel Mull, Yael Nazar, Norbert Nowotny, Márcio Roberto, Teixeira Nunes, Arnfinn Økland, Dennis Rubbenstroth, Brandy Russell, Eric Schott, Stephanie Seifert, Carina Sen, Elizabeth Shedroff, Tarja Sironen, Teemu Smura, Camila Prestes, Dos Santos Tavares, Robert Tesh, Natasha Tilston, Noël Tordo, Nikos Vasilakis, Peter Walker, Fei Wang, Anna Whitfield, Shannon Whitmer, Yuri Wolf, Han Xia, Ye, Zhuangxin Ye, Vyacheslav Yurchenko, Mingli Zhao, Ictv Taxonomy, Summary Consortium
## Abstract
RNA viruses are ubiquitous in the environment and are important pathogens of humans, animals and plants. In 2024, the International Committee on Taxonomy of Viruses Animal dsRNA and ssRNA(-) Viruses Subcommittee submitted 18 taxonomic proposals for consideration. These proposals expanded the known virosphere by classifying 9 new genera and 88 species for newly detected virus genomes. Of note, newly established species expand the large family of Rhabdoviridae to 580 species. A new species in the family Arenaviridae includes a virus detected in Antarctic fish with a unique split nucleoprotein ORF. Additionally, four new species were established for historically isolated viruses with previously unsequenced genomes. Furthermore, three species were abolished due to incomplete genome sequence information, and one family was moved from being unassigned in the phylum Negarnaviricota into a subphylum and order. Herein, we summarize the 18 ratified taxonomic proposals and the general features of the current taxonomy, thereby supporting public and animal health responses.
2024.001M.Alpharhabdovirinae_1ng_11nspTitle: In the subfamily Alpharhabdovirinae, create nine new species in six existing genera (Alphapaprhavirus, Sigmavirus, Merhavirus, Tupavirus, Alphanemrhavirus, Alpharicinrhavirus), rename the existing genus Thriprhavirus (as Alphathriprhavirus), and create the new genus Betathriprhavirus including two new species (Mononegavirales: Rhabdoviridae)
## INTRODUCTION
RNA viruses are widely distributed and infect a broad variety of hosts. As technological advancements in high-throughput sequencing and data analysis have exponentially expanded in the twenty-first century, so has the RNA virome [1]. The International Committee on Taxonomy of Viruses (ICTV) Animal dsRNA and ssRNA(-) Viruses Subcommittee (SC) was established in 2014 to develop a taxonomy for RNA viruses detected in the kingdom Animalia. Study Groups within the Subcommittee are responsible for many viral families in the phyla Negarnaviricota (orders: Muvirales, Jingchuvirales, Mononegavirales, Goujianvirales, Elliovirales, Hareavirales and Articulavirales) and Duplornaviricota (orders: Reovirales and Ghabrivirales) and the realm Ribozyviria.
The phylum Negarnaviricota, the largest taxon falling within the SC's remit, comprises viruses that predominantly have ssRNA(-) genomes, though some have an ambisense coding arrangement. Animal viruses assigned to this phylum exhibit a diversity of genome organizations (e.g. non-segmented or containing two to eight segments, linear or circular); however, all viruses within this phylum encode homologous RNA-directed RNA polymerases (RdRP) that form a strongly supported clade in the RdRP phylogenetic tree of the kingdom Orthornavirae [2]. The subphylum Haploviricotina is distinguished by viruses that encode an RdRP with mRNA capping activity [3]. Families such as Artoviridae, Bornaviridae, Filoviridae, Lispiviridae and Rhabdoviridae are classified within this subphylum in the order Mononegavirales. Families such as Hantaviridae, Peribunyaviridae, Phasmaviridae, Arenaviridae and Leishbuviridae are classified in the subphylum Polyploviricotina, distinguished by viruses that have an RdRP with cap-snatching activity [3] and include the orders Elliovirales, Hareavirales and Articulavirales.
Tosoviridae includes a single species for a virus isolated from sea turtles [4]. This virus has a bi-segmented ssRNA(-) genome similar to certain polyploviricotines. However, phylogenetic analysis cannot definitively place this family in a subphylum, and so it is classified as an unassigned negarnaviricot.
The phylum Duplornaviricota includes viruses that possess dsRNA genomes and can be further characterized by the presence of an unusual T=1 capsid [5]. Viruses in this phylum that infect animals have genomes that are either non-segmented or have 9 to 12 segments. Viruses of the family Sedoreoviridae (order: Reovirales) have genomes of 10-12 dsRNA segments and virions that have a characteristic 'smooth' appearance [6] in contrast to viruses in the reoviral family, Spinareoviridae. This summary includes the ratified taxonomic proposals from 2024 for the Animal dsRNA and ssRNA(-) Viruses SC and is not a comprehensive summary of all taxonomy proposals for dsRNA and ssRNA(-) viruses since some are covered by the Plant Viruses SC [7] and Fungal and Protist Viruses SC [8].
A file including all the Tables of taxonomic changes below is available as a supplementary file to this article.
## MAIN TEXT CONTENTS
Create nine new species in six existing genera (Alphapaprhavirus, Sigmavirus, Merhavirus, Tupavirus, Alphanemrhavirus and Alpharicinrhavirus) for viruses recently detected in bats, shrew or various invertebrates by metagenomic sequencing. Rename the existing genus Thriprhavirus (as Alphathriprhavirus), and create a new genus Betathriprhavirus including two new species for viruses detected in thrips by metagenomic sequencing.
## Justification
The viruses cluster phylogenetically with others in the existing or proposed genera in maximum likelihood trees inferred using L protein sequences. All new species in existing genera meet established demarcation criteria. The proposed renamed and new genera for viruses detected in thrips are well-separated phylogenetically. Submitted: 09/06/24
## 2024.003M.Artoviridae_4nsp
Title: Create two new species in genus Peropuvirus and two new species in genus Hexartovirus (Mononegavirales: Artoviridae)
Authors: Økland, AL ( arnfinn. lodden. okland@ zoetis. com), Kuhn, J, Ye, G, Vasilakis, N
## Summary Taxonomic rank(s) affected Species Description of current taxonomy
The family Artoviridae currently includes two genera, Hexartovirus (two species) and Peropuvirus (seven species).
## Proposed taxonomic change(s)
Create two new species in genus Hexartovirus and two new species in genus Peropuvirus.
## Justification
The viruses proposed to be assigned to novel species have a minimum amino acid divergence of 44 % in their L proteins compared to classified family members and occupy different ecological niches. Submitted: 21/06/2024
## 2024.008M.Lispiviridae_5ngen_11nsp
Title: Create five new genera and eleven new species in the family Lispiviridae (Mononegavirales)
## 2024.009M.Mammarenavirus_1nsp
Title
## Proposed taxonomic change(s)
Establishment of one new species in genus Mammarenavirus for a virus named vello virus, identified following the sequencing of mammarenavirus-positive rodent samples collected in Argentina from 1990 to 2020. Justification Two of the L segment sequences of 13 clade C mammarenavirus genomes identified following the sequencing of mammarenavirus-positive rodent samples collected in Argentina from 1990 to 2020, meet current demarcation species criteria for the genus Mammarenavirus. We propose the two isolates described by Shedroff et al. [9] to represent a virus named 'vello virus' and to assign this virus to a new species, Mammarenavirus vello. Submitted: 24/05/2024
## References
1. Aylward, Hendrickson, Lefkowitz et al. "Funding Information This work was supported in part through Laulima Government Solutions, LLC prime contract with the U.S. National Institute of Allergy and Infectious Diseases under contract no. HHSN272201800013C. J.H.K. performed this work as an employee of Tunnell Government Services, a subcontractor of Laulima Government Solutions, LLC, under contract no. HHSN272201800013C. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Health and Human Services or of the institutions and companies affiliated with the authors. V.Y. is supported in part by the Grant Agency of the Czech Republic (GACR 24-10009S) and by the European Union Operational Program 'Just Transition"
2. Hou, He, Fang et al. (2024) "Using artificial intelligence to document the hidden RNA virosphere" *Cell*
3. Wolf, Kazlauskas, Iranzo et al. (2018) "Origins and evolution of the global RNA Virome" *mBio*
4. Kuhn, Adkins, Alioto et al. (2020) "2020 taxonomic update for phylum Negarnaviricota (Riboviria: Orthornavirae), including the large orders Bunyavirales and Mononegavirales" *Arch Virol*
5. Waltzek, Stacy, Ossiboff et al. (2022) "A novel group of negative-sense RNA viruses associated with epizootics in managed and free-ranging freshwater turtles in Florida, USA" *PLoS Pathog*
6. Mata, Luque, Gómez-Blanco et al. (2017) "Acquisition of functions on the outer capsid surface during evolution of double-stranded RNA fungal viruses" *PLoS Pathog*
7. Jaafar, Attoui, Mertens et al. (2005) "Structural organization of an encephalitic human isolate of Banna virus (genus Seadornavirus, family Reoviridae)" *J Gen Virol*
8. Rubino, Abrahamian, Aranda et al. (2025) "Summary of taxonomy changes ratified by the International Committee on Taxonomy of Viruses (ICTV) from the Plant Viruses Subcommittee" *J Gen Virology*
9. Sabanadzovic, Abergel, Ayllón et al. (2025) "Summary of taxonomy changes ratified by the International Committee on Taxonomy of Viruses (ICTV) from the Fungal and Protist Viruses Subcommittee" *J Gen Virology*
10. Shedroff, Martin, Whitmer et al. (1990) "Novel Oliveros-like Clade C mammarenaviruses from rodents in Argentina" *Viruses* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12451635&blobtype=pdf | # Virus taxonomy proposal summaries: a searchable and citable resource to disseminate virus taxonomy advances
Richard Mayne, Peter Simmonds, Donald Smith, Evelien Adriaenssens, Elliot Lefkowitz, Hanna Oksanen, Francisco Zerbini, Poliane Alfenas-Zerbini, Frank Aylward, Juliana Freitas-Astúa, R Hendrickson, Holly Hughes, Mart Krupovic, Jens Kuhn, Małgorzata Łobocka, Arcady Mushegian, Judit Penzes, Alejandro Reyes Muñoz, David Robertson, Simon Roux, Luisa Rubino, Sead Sabanadzovic, Nobuhiro Suzuki, Dann Turner, Koenraad Van Doorslaer, Arvind Varsani
## Abstract
The Microbiology Society is a membership charity and not-for-profit publisher.Your submissions to our titles support the community -ensuring that we continue to provide events, grants and professional development for microbiologists at all career stages.Find out more and submit your article at microbiologyresearch.org
## INTRODUCTION
Taxonomy is an area of systematics that provides internationally agreed-upon classification and nomenclature frameworks for animals, plants, fungi, protists, prokaryotes and viruses. A universal taxonomy is essential for scientific communication, providing reference points for studies of the evolution and ecology of organisms. Taxonomy also informs regulatory frameworks for agricultural and livestock trade, biosecurity, medicine and public health.
## TAXONOMIC CODES
In view of the importance of agreed-upon and universally used taxonomies, there has been considerable worldwide and crossdisciplinary effort to develop consistent rules for taxonomic assignments and to maintain internationally shared taxon names for classified organisms. Reflecting this need, the nomenclature of animals, plants, fungi, protists, prokaryotes (bacteria and archaea) and viruses has been coordinated and regulated by longstanding expert committees over the previous century or longer; the International Abstract Taxonomic classification of cellular organisms requires the publication of descriptions and proposed names of species and the deposition of specimens. Virus taxonomy is developed through a different system of annual submission of formal taxonomy proposals (TPs) that can be submitted by anyone but are typically prepared by a study group appointed by the International Committee on Taxonomy of Viruses (ICTV) and consisting of experts on a particular group of viruses. These are initially evaluated by an expert subcommittee and by the executive committee (EC) of the ICTV. EC-approved TPs are then submitted for evaluation and a ratification vote by the wider ICTV membership. Following ratification, the new taxonomy is annually updated in the Master Species List, associated databases and bioinformatic resources. The process is consistent, creates traceability in assignments and supports a fully evaluated, hierarchical classification and nomenclature of all taxonomic ranks from species to realms. The structure also facilitates large-scale and coordinated changes to virus taxonomy, such as the recent introduction of a binomial species nomenclature.
TPs are available on the ICTV website after ratification, but they are not indexed in bibliographic databases and are not easily cited. Authors of TPs do not receive citation credit for adopted proposals, and their voluntary contributions are largely invisible in the published literature. For greater visibility of TPs and their authors, the ICTV will commence the annual publication of summaries of all TPs from each ICTV subcommittee. These summaries will provide a searchable compendium of all annual taxonomy changes and additions as well as direct links to the Master Species List and other ICTV bioinformatic resources. Their publication will provide due credit and citations for their authors, form the basis for disseminating taxonomy decisions and promote greater visibility and accessibility to taxonomy changes for the virology community. With the exception of the ICVCN for viruses, these codes primarily regulate and recommend procedures for the formal description of species, the assignment of scientific names and the associated deposition of supporting material in international repositories. There are broad similarities among codes in the conventions used to create Latinized binomial scientific names that perpetuate a nomenclatural system first developed by Linnaeus in the eighteenth century [2]. Orthography remains largely based on mediaeval Latin grammar, and it is remarkably unchanged in its formatting and rules for word formation and declension. Codes for cellular organisms differ from each other in detail, such as numbering and naming conventions for below-species ranks, whether genus and species epithets can be identical [tautonyms; e.g. Gorilla gorilla; Savage, 1847 and indeed the subspecies G. gorilla gorilla (Western lowland gorilla) in the ICZN], the extent to which assignments at higher taxonomic ranks are supported, and in the number and type of available secondary taxon ranks. There are also differences among codes in how species may be formally described, how authority is formatted after the scientific name (for example, Sclerophrys capensis Tschudi, 1838, with variants of this format reflecting taxonomic histories, ranks and codes), how and where descriptions and nomenclature proposals are published and the requirement and nature of materials required to support a species proposal. For example, descriptions of bacterial and archaeal species are generally published in the International Journal of Systematic and Evolutionary Microbiology, which also publishes the ICNP code. There is no journal requirement for the publication of zoological and botanical species names and descriptions, the only criterion is that they are made publicly available in journal or book form.
There are many published compendia and online databases of classified species and proposed higher taxonomic ranks. For example, bacterial species are provided in an online database at https://lpsn.dsmz.de, and species of animals are listed in ZooBank (https:// zoobank.org). The ICTV similarly maintains the Master Species List (https://ictv.global/msl) and associated metadata for each classified species of virus (https://ictv.global/vmr).
Collectively, the application of these codes for cellular organisms has provided biologists with comprehensive and relatively coordinated inventories of agreed-upon species names and taxonomic frameworks that broadly fulfil the requirements of the various stakeholders in biological, evolutionary, clinical and regulatory fields. The use of scientific names to specify international trade restrictions on defined organisms in the Convention on International Trade in Endangered Species of Wild Fauna and Flora regulations (https://cites. org/eng/app/appendices.php) exemplifies how biological nomenclature provides precision and authority to regulate the international movement of vertebrate and invertebrate animals and of plants.
## ALTERNATIVE CODES
Current taxonomies of cellular organisms inherit a 300-hundred-year-long historical legacy with substantial organizational and procedural baggage arising from classification systems that predate modern scientific publishing, online databases and genome sequencing technologies. Inference of genetic relatedness is better able to reconstruct evolutionary histories independently of phenotypic properties that guided the original classification of animals, plants, bacteria and viruses and may provide a firmer basis for a robust taxonomy, particularly at higher ranks. Proposals for a change towards a purely genomics-based classification of organisms include the International Code of Phylogenetic Nomenclature (PhyloCode) developed by the International Society for Phylogenetic Nomenclature (discussed in [3]). This proposes a purely cladistic classification based on metrics of sequence similarity that might better reconstruct an organism's evolutionary history, provide a more transparent and objective classification of organisms (particularly at high taxonomic ranks) and provide the means to resolve the numerous examples of paraphyly and nomenclatural confusion (such as Escherichia coli and Shigella spp.) of currently classified species. Another taxonomic framework, SeqCode, proposes that genome sequences can be used for valid publication of names of prokaryotes, avoiding the requirement for culturability and deposition of type materials [4]. Sequence-based assignments may vastly expand the number and range of (genotypically defined) species that could be assigned in the future, particularly for what may amount to over a million species of non-cultivated bacteria and archaea that are currently excluded from prokaryotic taxonomy. Unified databases, including the Catalogue of Life (www.catalogueoflife.org) and Encyclopaedia of Life (https://eol.org), seek to catalogue the over 2 million currently classified biological species into a combined database that would break down the current organizational divisions between the zoological, botanical and microbiological taxonomy codes and databases.
## TAXONOMY OF VIRUSES
Since its inception in 1966, the ICTV has developed and maintained a classification and nomenclature framework for viruses. At the outset, the ICTV considers viruses to be equivalently classifiable as cellular life. There are equivalences in taxonomic assignments of viruses with those in other codes. The virus taxonomy code, ICVCN, does, however, have to contend with the fact that viruses originated de novo multiple times during the evolution of cellular life, and the highest taxonomic rank, realm, has been devised as best as possible to assign viruses to what are deduced to be separate origin groups [5,6]. Below this rank, viruses have been classified into seven principal ranks (kingdom, phylum, class, order, family, genus and species) using a uniform and universal orthography, including rank-specific suffixes (such as -viricetes and -virales for class and order, respectively) generally resembling those of other codes. Recently, as a result of years of work by the ICTV in consultation with the virology community, the virus code has adopted and universally applied a binomial name format for species [5,7] comprising a genus name+species epithet, although without the compulsory Latinization of terms used in other codes. The genus name bears a -virus suffix, but the species epithet is 'freeform' (although restricted to the 26 letters of the mediaeval Latin alphabet, hyphens and numbers).
Virus taxonomy has more recently embraced evolutionary systematics as the basis for classification, permitting the assignment of species and higher ranks primarily based on metrics of genetic relatedness, often independently of phenotypic characterization or descriptions [8,9]. Accordingly, species can be assigned in the absence of a specimen or isolate using coding-complete genome sequences, deposited in one of the International Nucleotide Sequence Database Collaboration databases to provide a unique exemplar equivalent to a type specimen of other taxonomic codes. Viruses known only from their genomic sequences, such as those characterized in metagenomic analyses of environmental samples, can therefore be assigned taxonomically [10]; this has paved the way for a fivefold expansion in classified virus species over the last 5 years [11].
## ICTV ORGANIZATION
The ICTV regulates both the assignment of viruses and virus-like agents at all taxonomic ranks (from species to realm) and the nomenclature of these taxa. This contrasts with other biological taxonomies where the remit of official bodies is limited to the regulation of taxonomic names, rather than having the specific focus and regulatory role in the creation and recording of scientific names for species. Moreover, with viruses, there is no equivalent of the publication-and-attribution model for describing and naming new species that is followed throughout the rest of biology. Instead, changes and additions to virus taxonomy are initiated by formal taxonomy proposals (TPs), submitted annually to the ICTV EC by ICTV-associated expert study groups or by members of the virology community. TPs are publicly posted on the ICTV website before being initially reviewed by EC members and, after necessary modification, voted on and ratified by the wider ICTV membership that includes EC members, subcommittee (SC) and study group chairs, life members and national representatives. After ratification, taxonomy changes are implemented in the Master Species List, which serves as the primary and authoritative record of virus classification. This decision-making process is an effective framework for managing biological classification. Proposals receive expert scrutiny before ratification and adoption to ensure compliance with nomenclature conventions and the use of declared and approved taxon assignment criteria, such as metrics of sequence similarity for species or other rank assignments. Problems, such as the existence of homonyms and disputes over precedence, cannot arise in virus classification. The ICTV also possesses the organizational structure to perform extensive coordinated changes to taxonomy, such as the formal adoption of a 15-rank hierarchy in 2017 [12] and the renaming of species to conform to a binomial format in 2023-2024 [5,13,14]; such changes would be difficult and slow to coordinate in other taxonomy systems.
However, the development of virus taxonomy through the submission and ratification of TPs does not create a searchable published record of the taxonomy changes in bibliographic databases. This is despite the considerable scientific effort of authors to analyse data, prepare and write the formal proposals and revise them in light of the ICTV feedback -activities often equalling that of preparing manuscripts for journal publication. Furthermore, there are often no directly citable sources for formal taxonomic changes. For example, if someone wanted to cite the origin of the current name of the species for hepatitis C virus, Hepacivirus hominis, this could only be done by reference to a TP file stored on the ICTV website (in this case, in the proposal 2022.007S.Flaviviridae_1genren_sprenamed) rather than by reference to a specific publication. Even much broader changes, such as the creation of higher taxonomic ranks or the introduction of binomial species names, lack defined publications that describe the specific changes actually made by the ICTV. With some notable exceptions, such as the annual publication of taxonomy changes to RNA viruses in the phylum Negarnaviricota [15,16], proposers and authors of TPs, who are the key constituency in the advancement of virus classification, generally receive no published acknowledgement or citations for their contributions towards what may be major changes and additions to taxonomy. The consequent 'invisibility' in the published literature may thus disincentivize many virologists from actively contributing to virus taxonomy by preparing TPs and greatly hamper individuals from finding and appropriately citing sources for taxonomy statements.
## A PUBLICATION STRATEGY FOR RATIFIED VIRUS TPS
To address these issues, The ICTV will publish taxonomy advances in the form of summaries of ratified virus TPs. These summaries are intended to be generated annually for all accepted proposals from each of the seven currently existing ICTV SCs and for general proposals immediately following the ratification vote. This will create a total of eight citable publications per year, each with a DOI, in the Journal of General Virology, a partner journal of the ICTV.
A pipeline has been developed for the automated extraction of proposal abstracts, the creation of tables itemizing the taxonomy changes and the creation of combined author and taxon keyword lists (Appendix I, available in the online Supplementary Material) that will greatly facilitate the production and ensure the accuracy of the summaries. Automation of this process is only possible because the submission of TPs to the ICTV follows a set of procedures involving strictly formatted documents from which the necessary information can be reliably extracted.
## Format of summary
A summary will contain the following elements:
• Title (Fig. 1). This follows a standard format 'Summary of taxonomy changes ratified by the International Committee on Taxonomy of Viruses (ICTV) from the {subcommittee name}, {year}' . • Author list and affiliations. This is a compendium of all authors who contributed ratified proposals to the SC, listed in alphabetical order following the SC chair as the first author. The affiliations of all authors are ordered to correspond to the author list. • Abstract. This is a general summary of the main taxonomic developments described within the summaries, written by the SC chair. • Introduction. This is a variable-length section that might include the number of taxonomy additions and changes (e.g. the number of species and higher taxa established) as well as broader information on changes of perceived significance for the field in taxon assignments or methodologies used. • Content list. This indexes the TPs in a tabulated list of the ICTV codes and titles of each ratified TP included in the document, with hyperlinks to the positions of the corresponding sections of the document. • Individual summaries (Fig. 2). Each ratified TP is listed separately and sequentially. Each of these consists of
• The unique TP code assigned by the ICTV to the proposal; • TP title;
• Proposal authors, including the corresponding author/s as indicated by an email address;
• Summary;
• Submission and revision dates;
• Tabular summary of the taxonomy additions and changes; and
• Source data, a hyperlink to the full TP documents on the ICTV website (indexed under https://ictv.global/files/proposals/ approved). • Keywords. This is an alphabetically sorted list of new or changed taxon names referred to in the summaries, including previously used terms for taxa that have been renamed or abolished. The keyword list provides indices for searches in PubMed and other bibliographic sources that enable taxa to be found and cited in general literature searching. • References to papers cited in the Introduction.
• Master table. A master table will be included as a supplementary Excel file, containing a formatted list of all tabular data for a TP summary. All data are stored in one sheet and are identical to their original representation in the TP summary, with the exception of an additional column to indicate from which TP they originate.
## Citing a taxonomy change
The summaries are designed to provide a published and citeable source for all taxonomic changes such as the assignment of new taxa or re-classifying or renaming virus taxa. We propose that citations of the summary should be supplemented with the relevant ICTV taxonomy proposal code in the format where [34] is the citation for the Taxonomy Summary, "Yimin et al." is an optional reference to the authorship of the proposal and "2024.003A" is the abbreviated ICTV code for the specific taxonomy proposal within the cited summary.
Reference to multiple proposals within a summary could be formatted as in this example:
"Two additional species of orthohantaviruses were described [34] (proposals 2024.001M and 2024.017M). "
## CONCLUSIONS
The creation of annual virus taxonomy summaries for the seven ICTV SCs and of general proposals represents an entirely new strategy for the publication of virus taxonomy changes. This approach is quite different in scope and authority from the less-regulated publication of species descriptions and nomenclature followed by other taxonomic codes. The virus taxonomy summaries will provide author attributions, searchable content and linkage to ICTV taxonomy proposals and Master Species List databases. We believe that the virus taxonomy summaries will be effective in disseminating and providing a citable source for virus taxonomy information in the future.
The software parses all of the files in a user-specified folder and generates a summary document. This is designed for use at the end of ICTV ratification rounds, with the intention of creating a publishable reference source for the year's TPs. Summaries will be created for all of the TPs within an SC each year.
The main output, hereafter 'summary documents' , are Microsoft Word files containing summarized TP data from both EMs and WMs. Accompanying tabular files are generated, containing the author list and 'master tables' , which contain all of the summarized EMs in a single sheet.
To test the summary system, TP summaries were generated for each of the six ICTV SCs with the 109 TPs from the 2025 ratification round. (There were no animal+ssRNA (S) TPs during this period.) Then, TP summaries were manually checked by the ICTV Executive Committee and development team, which iteratively refined the summary system.
## Performance
In cases for which no errors were identified in the underlying data, TP summaries were generated in ~5 s on a standard consumergrade laptop. Of the 109 individual TPs, the summary system identified 66 WM documents as having errors, of which 46 were human errors and 20 were due to incompatibilities in language pack encoding. All WM errors were manually corrected. Only one error was identified in EMs, in which a title field was not present, which was also corrected manually. In total, all six summary documents were generated, checked and disseminated within a single day.
For future work, it will be additionally possible to use the summary system to parse TPs at the point of importation into the ICTV website. The benefit would be the automatic proofing of incoming submissions.
The source code is freely available on the GPL-3.0 licence at https://github.com/Mayne941/parse_taxonomy_proposal_form
## References
1. "Department of Microbiology, University of Alabama at Birmingham, BBRB 276, 845 19th St South" *Molecular and Integrative Biosciences Research Programme, Faculty of Biological and Environmental Sciences*
2. Cassava, Fruits, Das Almas et al.
3. "CNRS UMR6047, Archaeal Virology Unit, 25 rue du Dr Roux" *Biophysics of the Polish Academy of Sciences*
4. Warsaw "USA; 15 Departamento de Ciencias Biológicas"
5. Bari (2019) "85287-4701, USA. References 1. International Code of Nomenclature of Prokaryotes" *The Biodesign Center for Fundamental and Applied Microbiomics, School of Life Sciences*
6. Cain (1993) "Linnaeus's ordines naturales" *Arch Nat Hist*
7. Queiroz, De (2006) "The PhyloCode and the distinction between taxonomy and nomenclature" *Syst Biol*
8. Hedlund, Chuvochina, Hugenholtz et al. (2022) "SeqCode: a nomenclatural code for prokaryotes described from sequence data" *Nat Microbiol*
9. Zerbini, Siddell, Mushegian et al. (2022) "Differentiating between viruses and virus species by writing their names correctly" *Arch Virol*
10. Koonin, Krupovic, Dolja (2023) "The global virome: How much diversity and how many independent origins?" *Environ Microbiol*
11. Siddell, Smith, Adriaenssens et al. (2023) "Virus taxonomy and the role of the international committee on taxonomy of viruses (ICTV)" *J Gen Virol*
12. Wolf, Silas, Wang et al. (2020) "Doubling of the known set of RNA viruses by metagenomic analysis of an aquatic virome" *Nat Microbiol*
13. Koonin, Dolja, Krupovic et al. (2020) "Global organization and proposed megataxonomy of the virus world" *Microbiol Mol Biol Rev*
14. Simmonds, Adams, Benkom et al. (2017) "Consensus statement: virus taxonomy in the age of metagenomics" *Nat Rev Microbiol*
15. Zerbini, Siddell, Lefkowitz et al. (2023) "Changes to virus taxonomy and the ICTV statutes ratified by the international committee on taxonomy of viruses" *Arch Virol*
16. Gorbalenya, Krupovic, Mushegian et al. (2020) "The new scope of virus taxonomy: partitioning the virosphere into 15 hierarchical ranks" *Nat Microbiol*
17. Postler, Clawson, Amarasinghe et al. (2017) "Possibility and challenges of conversion of current virus species names to linnaean binomials" *Syst Biol*
18. Siddell, Walker, Lefkowitz et al. (2020) "Binomial nomenclature for virus species: a consultation" *Arch Virol*
19. Kuhn, Abe, Adkins et al. (2023) "Taxonomic update of RNA-directed RNA polymerase-encoding negative-sense RNA viruses (realm Riboviria: kingdom Orthornavirae: phylum Negarnaviricota)" *J Gen Virol*
20. Kuhn, Adkins, Alkhovsky et al. (2022) "2022 taxonomic update of phylum Negarnaviricota (Riboviria: Orthornavirae), including the large orders bunyavirales and mononegavirales" *Arch Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724133&blobtype=pdf | # Characterization of HERV-K (HML-2) Rec proteins encoded in the human genome and their post-transcriptional function
Katarzyna Zurowska, Godfrey Dzhivhuho, David Grabski, David Rekosh, Marie-Louise Hammarskjold, Marie-Louise Ca206275, Hammarskjold
## Abstract
Human endogenous retrovirus K (HERV-K) proviruses of the HML-2 subgroup are the most recently integrated retroviral elements within the human genome. The HERV-K Rec protein, a functional homolog of HIV Rev, is essential for the expression of viral mRNAs with retained introns. However, their diversity and functional capacities have remained largely unexplored. We identified Rec coding sequences in the human genome and selected intact Rec proteins from 28 proviral loci for func tional characterization. Using a dual-color fluorescent reporter containing the HERV-K Rec-response element and a complementary enzyme-linked immunosorbent assay, we found that Rec proteins from nine proviral genomic loci promoted function at the post-transcriptional level. All but one of these proviruses are insertionally polymorphic in the human genome. In addition, several of the non-functional Rec proteins were trans-dominant negative. Detailed mutational analysis of the most potent inhibitory variant, encoded by the HERV-K provirus 12q14.1, displayed only two amino acid changes (N2H and E34del) relative to the prototypical functional Rec protein encoded by the reconstructed consensus HERV-K. Insertion of a glutamic acid at position 34 in this mutant fully restored the functional activity. Our results unveil an unexpected complexity in HERV-K (HML-2) post-transcriptional regulation, with a few Rec variants showing function, while others are trans-dominant negative. These findings significantly expand our understanding of the various Rec proteins that can be expressed in human cells and how they function at the post-transcriptional level.IMPORTANCE In this study, we compared Rec sequences from 58 type 2 human endogenous retrovirus K proviruses and showed that only 9 encode Rec proteins that are functional in post-transcriptional regulation of RNAs with retained introns. We also showed that several Rec proteins have trans-dominant negative activity when co-expressed with a functional Rec protein. While previous studies have demonstrated the expression of Rec mRNAs and proteins in human cells, this is the first study to define which loci have the potential to encode either functional or trans-dominant negative Rec proteins. Additionally, our reporter system will also enable future investigations to easily determine whether functional and/or trans-dominant negative Rec proteins are expressed in any cell. These findings are also important for future studies that aim to link Rec post-transcriptional function to physiological or pathological effects. KEYWORDS endogenous retroviruses, protein diversity, RNA export, HERV-K H uman endogenous retroviruses (HERVs), which constitute approximately 8% of the human genome, are a class of transposable elements that are remnants of ancient exogenous retroviruses. These viruses integrated into the primate germline between 100 and 40 million years ago (1, 2). Although HERVs have accumulated many mutations and deletions within coding sequences over time (3), many remain transcriptionally active,
contributing to cellular function either through the production of viral proteins or by serving as regulatory elements for host gene expression (3)(4)(5)(6).
Among HERVs, the HERV-K (HML-2) subgroup is the most recent and biologically active. While most HERV-K proviruses in the human genome are shared with other great apes and date back 5-30 million years, a subset is human-specific and polymorphic. Notably, the insertion of HERV-K115 has been estimated to have occurred as recently as ~350,000 years ago (5), indicating that this retroviral family remained active well into recent human evolutionary history.
HERV taxonomy is based on sequence similarity to other animal retroviruses (7)(8)(9). The HERV-K clade belongs to class II of beta retrovirus-like endogenous retroviruses, with the "K" designation referring to the lysine tRNA primer used to prime reverse transcription in these viruses. The HERV-K (HML-2) subgroup that comprises more than 100 proviruses is the most recently integrated and biologically active. In this report, HERV-K (HML-2) virus will hereafter be referred to simply as HERV-K.
Many HERV-K copies retain intact open reading frames (ORFs) with the potential to produce viral RNAs and, in many cases, viral proteins. Complete HERV-K genomes, like all intact retroviruses, encode the essential gag, pro, pol, and env genes, flanked by long terminal repeats (LTRs) (7). Additionally, like the more complex viruses, such as HTLV-1 and HIV (10), they also encode regulatory proteins (11). HERV-K proviruses are classified as type 1 or type 2. Type 2 proviruses encode the Rec protein from a doubly spliced transcript; type 1 proviruses have mutations and a 292-bp deletion in the pol-env region, which deletes the first coding exon of rec and changes the doubly spliced transcript in some of the type 1 viruses, to encode a protein named Np9 (7,(11)(12)(13)(14).
Rec interacts directly with the Rec-Response Element (RcRE), present in the 3′ end of all HERV-K mRNAs (11,15,16). Through this interaction, Rec facilitates the nucleocyto plasmic export and expression of HERV-K mRNAs that contain retained introns, using the Crm1-RanGTP pathway (11,17). Rec contains an arginine-rich nuclear location signal (NLS), as well as a nuclear export signal (NES) (17,18), and functions similarly to the HIV Rev, HTLV Rex, and mouse mammary tumor virus (MMTV) Rem proteins (19)(20)(21)(22)(23)(24). In addition to the proviruses, more than 900 HERV-K solitary LTRs are scattered throughout the human genome (5). Many of these contain RcRE sequences that can potentially interact with Rec proteins. Recent studies have identified cellular genes that contain sequences with high homology to HERV-K RcREs, suggesting that Rec may modulate cellular gene expression through interactions with these elements (5,(25)(26)(27)(28).
Some evidence suggests that HERV-K Rec may play significant roles in cancer development, potentially through interactions with cellular proteins. For example, Rec has been shown to interact with the promyelocytic leukemia zinc-finger protein (PLZF), activating c-myc proto-oncogene expression and promoting cell proliferation (13,29). Rec has also been shown to interact with the human small glutamine-rich tetratricopep tide repeat-containing protein (hSGT). This leads to increased androgen receptor activity, which may promote oncogenesis (30).
The first coding exon of Rec is the same as the first 87 amino acids of the 95 amino acid Env signal peptide (SP) (31). The second coding exon also overlaps env sequences but is translated in a different open reading frame. In the case of the closely related MMTV, it has been shown that the Env SP, which corresponds to the first coding exon of the nuclear export protein Rem, is sufficient to function in the nuclear export of viral mRNAs (32) since it contains RNA binding, as well as NLS and NES domains. The HERV-K Env SP also contains the NLS and NES sequences, which suggests that it could also be sufficient for HERV-K mRNA export. This relationship and the fact that both coding exons of Rec overlap the env coding region become particularly relevant when interpreting studies showing oncogenic properties of HERV-K Env in various cancers (33,34), as Env expression is also likely to depend on Rec expression. Even though the env mRNA is spliced, it still retains an intron, similar to Env mRNAs in other complex retroviruses, which have been shown to require a trans-acting protein to facilitate nucleo-cytoplasmic export and expression (10,(19)(20)(21)(22)(23)(24).
Despite the widespread distribution of HERV-K sequences in the human genome and the potential importance of Rec in regulating viral and cellular gene expression, an analysis of which proviral loci produce Rec proteins that can function at the post-tran scriptional level has not been performed. In this study, we performed a comprehensive functional analysis of the HERV-K Rec proteins that can be expressed from type 2 proviral loci. Our findings reveal that while some Rec variants promote the expression of RcRE-RNA with retained introns, others act as trans-dominant negative regulators, suggesting that post-transcriptional Rec regulation of HERV-K expression is complex.
## RESULTS
## Characterization and selection of HERV-K HML-2 Rec proteins for functional analysis
The identification and characterization of functional HERV-K HML-2 Rec proteins present significant challenges due to the repetitive nature and abundance of HERV sequences in the human genome. We approached this problem by first annotating key features of the reconstituted HERV-K provirus, HERV-K Con, including long terminal repeats, open reading frames, and known splice sites (Fig. 1A) (35,36). To account for both type 1 and type 2 proviruses, we generated two reference genomes: the original HERV-K Con (type 2) and a modified version with a 292-bp deletion in the pol-env region (type 1). We then aligned 91 HERV-K genomes described in Subramanian et al. (5) to these annotated references, transferring annotations to each sequence. This process enabled us to differentiate between type 2 proviruses (55 copies) containing the rec gene and type 1 proviruses with the 292-bp deletion. Table S4 shows all of the type 2 HERV-K loci analyzed in this paper, along with their chromosomal positions in the hg38 genome and their corresponding commonly used names (where available), such as HERV-K113 and HERV-K108. The loci from which Rec sequences were tested are highlighted in red.
Our overall strategy was informed by prior analyses of Rec, which cataloged several Rec open reading frames and their distribution across the HERV-K (HML-2) family (37,38). However, these studies did not include an analysis of the functional activity of the predicted Rec proteins.
To create the potential Rec proteins that could be expressed from each of the 55 loci and assayed for functional activity, we first determined if the 5′ and 3′ splice sites connecting the two rec exons were intact. If they were, we spliced the two exons together. If either splice site was not intact, we kept the first coding exon contiguous with the env sequence. These sequences are now included in FASTA format in File S1. Each sequence was then translated to determine the presence of stop codons and indels leading to frameshifts. The translated protein sequences are shown in Fig. 1B andC, and the sequences in FASTA format are listed in File S2. In performing our analysis, we found that the 3′ splice site upstream of the second coding exon of Rec in provirus 10q24.2 is mutated from AG to GG, and thus, the predicted protein sequence does not include a proper second exon. For the loci that could potentially produce functional Rec proteins, we also examined the major 5′ splice site and the 3′ splice site just upstream of the first coding exon of Rec. In all cases, these splice sites remain intact. Additionally, except for 19q11, all proviruses have mostly intact 5′ LTRs. However, expression of this provirus has been previously reported (39), suggesting that an upstream cellular promoter may be used for transcription.
This alignment revealed striking protein sequence diversity among the Rec proteins encoded by these HERV-K loci (Fig. 1B). Over half of the ORFs were truncated due to large deletions and/or premature stop codons. We identified and selected 25 proviruses that could express full-length Rec proteins, excluding 29 incomplete or heavily truncated protein sequences, as well as the 10q24.2 provirus that lacks the 3′ splice site for the second coding exon, from functional analysis. Among the selected 25 full-length proteins, we observed significant sequence diversity compared to the reference HERV-K Con Rec (Fig. 1C). However, five Rec proteins (encoded by proviruses 6q14.1, 7p22.1a, 7p22.1b, 11q22.1, and 19q11) were identical to the reference HERV-K Con Rec at the protein level, despite one to three nucleotide differences in their coding sequences (36). Thus, we initially selected 21 distinct Rec protein variants for the analysis of posttranscriptional function: 1 identical to the reference HERV-K Con Rec and 20 with amino acid variations.
## Evaluation of HERV-K HML-2 Rec protein post-transcriptional function using an RcRE-reporter assay
To evaluate the functional capacity of the selected HERV-K HML-2 Rec proteins, we developed an assay using a dual-color fluorescent reporter vector. This vector is similar to our established HIV-1 Rev-RRE dual-color reporter but contains a tandem RcRE in place of the HIV RRE (40,41). This modified reporter produces mCherry constitutively from a spliced mRNA, while GFP expression from an unspliced RNA depends on the presence of both the cis-acting RcRE and a functional trans-acting Rec protein. For a diagram of this vector, see Fig. S1. To facilitate our experiments, we generated a 293T/17-based cell line stably expressing the reporter mRNA (Fig. 2A).
To test the activity of the various Rec proteins, we cloned the ORFs for each of the 21 proteins into a murine stem cell virus-derived retroviral vector (pMSCV), positioning them upstream of an IRES-eBFP2 cassette. This bicistronic design ensures simultaneous expression of both Rec and eBFP2 from a single transcript, facilitating detection and quantification of expression levels.
If a functional Rec protein is expressed from this vector in the transduced reporter cells, its interaction with the results in the export of the unspliced mRNA and GFP expression. After flow cytometry analysis, Rec functional activity can be quantified as the ratio of GFP mean fluorescence intensity (MFI) to eBFP2 MFI. By using this ratio, the results are normalized to account for potential differences in transduction efficiency and RNA expression, since Rec and eBFP2 are translated from the same mRNA.
Transduction experiments with the vectors expressing the 21 different Rec proteins showed that, in addition to HERV-K Con Rec, encoded by the five proviruses at 6q14.1, 7p22.1a, 7p22.1b, 11q22.1, and 19q11, only two other Rec proteins (at loci 8p23.1a and 19p12b) displayed functional activity (Fig. 2B). All but one of these Rec proteins (19q11) originate from proviral loci that are polymorphic in the human genome. 19q11 lacks the Annotations (red) show coding exons 1 and 2. Red asterisks (*) indicate loci that encode functional Rec proteins. A summary of all the tested Rec proteins is included in Table S5. 5′ LTR, but viral RNA may be transcribed from an upstream promoter, since expression from this provirus has been reported in the literature (39). The remaining Rec variants showed no significant activity, as indicated by the absence of GFP expression despite eBFP2 expression (Fig. 2C). We also employed a complementary approach, using an HIV GagPol-RcRE reporter vector to further validate and quantify the activity of the functional Rec protein variants. This reporter measures p24 release as an indicator of viral protein expression mediated by a functional Rec/RcRE interaction. To perform this assay, the Gag-Pol-RcRE reporter was co-transfected into cells with increasing amounts of the plasmids expressing the Rec proteins (Fig. 2D). This confirmed the functionality of the three protein variants.
We next decided to directly investigate how functional activity was related to steady-state levels of Rec expression. Due to the lack of commercially available Rec antibodies, we generated constructs with HA tags added to the N-termini of these Rec proteins, enabling visualization of their expression by Western blot using an anti-HA antibody. As shown in Fig. 3A, all three of the tagged constructs displayed Rec activity. Western blot analysis showed some differences in protein expression levels among the HA-tagged Rec variants (Fig. 3B andC). We thus normalized the Rec activity based on the protein expression levels (Fig. 3D). These assays demonstrated that HERV-K Con displayed the highest activity, as was also demonstrated in the transduction experi ments.
## Some HERV-K HML-2 Rec proteins are trans-dominant negative
We hypothesized that some of the non-functional Rec proteins might display a transdominant negative phenotype. To test this, we selected the four non-functional Rec proteins (encoded at 3q21.2, 5p13.3, 10p14, and 12q14.1) that were the least changed (<15 aa changes) compared to HERV-K Con Rec (Fig. 4A). We cloned cDNA copies of each coding sequence, with or without N-terminal HA-tags, into expression plasmids. The HAtagged plasmids were then transfected into the RcRE-reporter cell line. Western blot analysis revealed that all four selected non-functional Rec proteins were expressed in these cells, but the Rec protein encoded by provirus 3q21.2 showed only very low levels of expression (Fig. 4B).
To analyze if these Rec proteins were trans-dominant negative, we next co-transfec ted increasing amounts (0-800 ng) of either the tagged or the non-tagged plasmids into RcRE-reporter cells, with a constant amount (100 ng) of the plasmid expressing the active non-tagged HERV-K Con Rec. An appropriate amount of empty vector was added to maintain a total of 2 µg of transfected DNA. We then measured Rec functional activity using flow cytometry.
In the experiment with both the non-tagged and HA-tagged Rec proteins, three of the four non-functional Rec proteins (5p13.3, 10p14, and 12q14.1) exhibited some degree of trans-dominant negative effect (Fig. 4C andD). However, Rec 3q21.2 showed no significant inhibitory effect. Interestingly, the 12q14.1 Rec protein that displayed a very potent trans-dominant negative effect had only two changes compared to HERV-K Con Rec: an N2H mutation and a deletion at position 34 (E34del) (Fig. 4A). Since Rec 12q14.1 demonstrated a potent trans-dominant negative effect while having only two amino acid changes compared to the HERV-K Con Rec, we sought to determine if one or both changes were responsible for the trans-dominance. We thus generated two variant mutants of this Rec protein, as shown in Fig. 5A: Variant 1, which restored a glutamic acid at position 34 (del34E), and Variant 2, which changed the histidine at position 2 to the asparagine present in HERV-K Con Rec (H2N). The variant sequences were cloned into the MSCV vector upstream of the IRES-eBFP2 region, and HA-tagged versions of the original 12q14.1 and the two variants were also generated to enable analysis of protein expression. Western blot analysis of 293T/17 reporter cells using the HA-tagged versions of these constructs showed that all three proteins were well expressed (Fig. 5B). The untagged constructs were then packaged into retroviral particles and used to transduce the RcRE-reporter cell line, in parallel with vectors that expressed the original 12q14.1 Rec and HERV-K Con Rec as a positive control. Rec activity was measured as described in Fig. 2B. This experiment revealed that Variant 1 showed functional activity similar to HERV-K Con Rec (Fig. 5C), whereas Variant 2 and the original 12q14.1 Rec were non-functional.
We also tested the activity of the HA-tagged 12q14.1 Rec variants. This was done by transfection of the HA-tagged plasmids into the RcRE-reporter cell line and measurement of GFP MFI by flow cytometry. Similar to the untagged constructs, HA-tagged Variant 1 showed Rec/RcRE functional activity, but the HA-tagged Variant 2 and the original HAtagged 12q14.1 Rec did not (Fig. 5D). These results indicate that deletion of E34 alone abolishes Rec functional activity, whereas changing the asparagine to histidine at position 2 has minimal impact on function.
We next analyzed the trans-dominant negative activities of the two variants by coexpressing increasing amounts of each variant with a constant amount of HERV-K Con Rec as described above for Fig. 3. Variant 2 (E34del) exhibited a trans-dominant negative effect nearly identical to the original Rec 12q14.1 (Fig. 5E). The HA-tagged versions gave similar results (Fig. 5F). Notably, both tagged and untagged Variant 1 showed no transdominant negative effect, and expression of this variant increased the overall activity.
However, noticeably, the increase in activity with the non-tagged Rec Variant 1 protein was considerably less than with the HA-tagged protein construct (Fig. 5E andF). This is likely due to saturation of functional activity with higher plasmid doses. Figure S2, which is the same data plotted without normalization, shows that the baseline GFP MFIs are different. The baseline value with the non-tagged Rec appears to already be at a level of saturation that is reached by the HA-tagged Rec only at higher DNA concentrations.
Based on these experiments, we conclude that the glutamic acid at position 34 (E34) is essential for Rec functional activity, and that this deletion alone causes a transdominant negative phenotype.
## Functional analysis of Rec proteins from loci identified since the Subrama nian 2011 study
Three new type 2 proviral loci have been identified since the Subramanian compilation. Two (8q24.3c and Xq21.33) have been described as insertional polymorphic (42). The third (1p36.21) is absent from hg19 but is present in the hg38 genome assembly (43).
The Rec proteins that can be expressed from these proviruses are shown in Fig. 6A, in an alignment to HERV-K Con Rec. The two polymorphic proviruses showed only a single amino acid change, whereas the third was highly mutated and C-terminally extended because of reading-frame shifts and other changes. We made retrovirus constructs for these three proteins and tested them in both transduction and transfection experiments, in comparison with HERV-K Con Rec. The results, shown in Fig. 6B andD, demonstrated that the highly mutated Rec protein (1p36.21) showed no functional activity, whereas the other two proteins functioned well, although not as well as the HERV-K Con Rec protein. Figure 6C shows eBFP2 expression levels in the transduced cells, confirming comparable transduction efficiencies. These findings further highlight that proteins with more than a few mutations relative to HERV-K Con Rec are unlikely to function at the post-transcriptional level. However, Rec encoded by new proviruses can easily be tested for function, using our reporters, to confirm or refute this.
## DISCUSSION
In this study, we show that 25 of the HERV-K type 2 proviruses in the human genome have open reading frames capable of producing full-length Rec proteins. However, only nine of these loci express proteins with post-transcriptional function. Among these, eight are insertionally polymorphic (6q14.1, 7p22.1a, 7p22.1b, 8p23.1a, 8q24.3c 11q22.1, 19p12b, and Xq21.33) in human populations. 19q11 is not polymorphic at the integra tion site, although it has been reported to be internally polymorphic (43)(44)(45)(46)(47). These functional loci all have nucleotide sequence changes in the rec gene relative to the prototypical HERV-K Con rec gene commonly used in published studies (43,(45)(46)(47). In five of these loci, the nucleotide changes are synonymous within the Rec ORF (6q14.1, 7p22.1a, 7p22.1b, 11q22.1, and 19q11), and they thus express a protein with the same amino acid sequence as HERV-K Con. For four of the loci (8p23.1a, 11q22.1, 8q24.3c, and Xq21.33), there are a few amino acid changes relative to HERV-K Con (see Fig. 1C; Fig. S3). None of the changes is in the known NLS and NES domains. The evolutionary age of the HERV-K proviruses differs significantly. The relatively younger proviral insertions, such as those at loci 6q14.1 and 19p12b, are more likely to retain functional Rec proteins due to fewer accumulated mutations (1, 2). Conversely, older proviruses are typically heavily mutated, rendering their encoded Rec proteins nonfunctional, as seen for proviruses at several well-characterized loci (3,5,12). Importantly, one older provirus, the nearly intact locus 19q11, maintains the potential to produce functional Rec and other viral proteins despite its age, indicating some exceptions to this (7,12,18).
About half of the HERV-K proviruses are type 1 and contain a deletion in Env and the overlapping Rec region. This eliminates the ability to synthesize a Rec protein, making these proviruses unable to export their unspliced genome mRNA, which requires the help of an export protein, to express Gag and GagPol proteins. Additionally, the singly spliced mRNA, which encodes an N-terminally truncated Env protein, still retains an intron, so env expression would also be expected to be impaired (10). Thus, cells (C) eBFP2 expression levels from the transductions performed in panel B. The MFI of eBFP2 expressed from each of the individual transductions was determined using flow cytometry. Data are shown as mean ± SD from three independent experiments. (D) Validation of Rec activity using a p24 release assay. 293T/17 cells were co-transfected with equal amounts of a Gag-Pol-RcRE reporter vector (1,500 ng) and increasing amounts of functional Rec vectors (50, 100, 150, and 250 ng). An empty vector was used to normalize the total DNA input to 2,000 ng. Supernatants were collected at 72 hours post-transfection, and p24 was then measured by ELISA. The HERV-K Con Rec sequence represents five identical Rec 1013 proteins (6q14.1, 7p22.1a, 7p22.1b, 11q22.1, and 19q11) (marked with an asterisk). The data shown are the mean values ± SD from three independent experiments. transcribing RNA from only type 1 viruses would be limited to expressing only the Np9 protein that is translated from a Rec-independent multiply spliced mRNA, unless the same cell expresses functional Rec from a type 2 HERV-K provirus. Similarly, type 2 proviruses lacking a functional Rec could also be complemented in this way to export all of their mRNAs with retained introns and express their viral structural proteins. This trans-complementation would be similar to our previous study, which showed that the HIV Rev protein can complement Rec function when supplied in trans, to enable nuclear export of endogenously expressed HERV-K RNAs with retained introns (28).
Several studies have implicated HERV-K in various malignancies and autoimmune disorders (13,29,30). For instance, a recent study reported robust transcription from certain HERV-K loci in lung cancer, highlighting the potential clinical significance (17,48). Several published studies have also documented HERV-K expression in specific biological contexts. However, most of these studies have focused on the expression of Env proteins (49)(50)(51). Since the mRNA for the Env proteins retains an intron and would be expected to be dependent on Rec (10), functional Rec is likely expressed in the cells as well. Expression of Rec has also been observed in germ cells (26) and certain tumors such as melanomas (52). However, none of these studies examined whether functional Rec proteins were expressed. In addition to disease-associated expression, it has been demonstrated that Rec and Np9 transcripts are present in a range of normal human tissues, indicating that transcription from HERV-K loci is not restricted to pathological contexts (38). This finding underscores the importance of considering both physiological and disease-associated expression patterns when evaluating Rec function.
As mentioned in the Introduction, a previous study implicated HERV-K Rec proteins in cancer-related processes through interactions with the PLZF, leading to c-myc upregula tion and enhanced cell proliferation (13). In addition, Rec protein interactions with the hSGT were reported to increase androgen receptor activity, potentially contributing to oncogenesis (30). It remains to be determined how many of the Rec proteins can interact with these proteins and cause these effects, since this may not require the protein to function in RNA expression. Activation of HERV-K and production of virus particles, which should include functional Rec expression, have also been implicated in melanoma progression (53). However, in one study, Rec was shown to form a regulatory feedback loop with the microphthalmia-associated transcription factor, potentially counteracting progression to a more invasive phenotype (54). This suggests that the effects of Rec in cancer may be context-dependent and not exclusively pro-oncogenic (for a review, see reference 55).
In addition to the few loci that can express a Rec protein that functions in post-tran scriptional RNA regulation, we have determined that some Rec proteins show transdominant negative activity. The most striking example is the Rec encoded by provirus 12q14.1, which differs from the consensus by only two amino acid changes (N2H and E34del), yet potently inhibits functional Rec activity. This highlights that a few amino acid changes in retroviral regulatory proteins can significantly affect function, as shown previously for HIV Rev (56). Our mutational analyses revealed that reintroducing glutamic acid at position 34 restored export function. Interestingly, the deletion is not in a region known to be important for nuclear export, import, or dimerization, and thus it inhibits Rec function through an unknown mechanism (17,57). This contrasts with the well-stud ied trans-dominant negative Rev protein (RevM10) that has mutations in the nuclear export signal (58)(59)(60).
The trans-dominant negative Rec proteins may serve as a natural "brake" in HERV-K replication and expression. This could potentially limit pathogenic HERV-K effects, and these Rec variants may thus have been positively selected during HERV-K evolution and spread. Additionally, the deletion of Rec in the type 1 HERV-K proviruses eliminates potential pathogenic effects resulting from Rec overexpression. However, since Rec can act in trans, if expressed from a type 2 virus that is active in the same cell, type 1 viruses do not need to encode Rec for replication. HERV-K proviruses have also been reported to produce viral particles capable of packaging and transmitting HERV-K-related sequences in cell culture (61). This highlights the potential for functional Rec proteins to contribute to the mobilization of endogenous retroviral elements under certain conditions.
Our discovery that active proviral copies of HERV-K can encode both active and inhibitory Rec proteins shows the complexity of Rec regulation. Levels of HERV-K expression and potential host cell effects of Rec activity will vary depending on which HERV-K loci are active, and it remains possible that Rec is an important factor in some human cancers. Further studies on the oncogenic properties of Rec and their relationship to the function of Rec as a post-transcriptional factor will be necessary to elucidate this.
## MATERIALS AND METHODS
## Identification of HERV-K HML-2 Rec ORFs
Human endogenous retrovirus K HML-2 proviral sequences (n = 91) were retrieved from the NCBI Nucleotide Database (GenBank) (see Table S4 for accession numbers) (5). Each HERV-K HML-2 genome was aligned to the annotated HERV-K Con reference genome using Geneious Prime software (version 2020.2, Biomatters Ltd., Auckland, New Zealand) to transfer annotated proviral features. Type 1 proviruses were identified based on the presence of the 292-bp deletion in the pol-env region and the GA-GT mutation in the 5′ splice site. Rec coding sequences from the identified type 2 proviruses lacking the 292-bp deletion were generated by extracting and joining the annotated Rec exons from each type 2 provirus. Sequence alignments were performed using the MUSCLE algorithm with default parameters as implemented in Geneious.
## Plasmids and vector construction
## Reporter vector system
The dual-color Rec-RcRE reporter vector was adapted from the previously described Rev-RRE HIV reporter system (40,41). The HIV RRE sequence was replaced with two tandem copies of the HERV-K RcRE (2xRcRE) while maintaining the same vector backbone and fluorescent protein reporters. The RcRE sequence was derived from a HERV-K LTR sequence previously deposited in GenBank (accession number AF179225.1) and synthesized by Integrated DNA Technologies (IDT, Coralville, IA, USA).
## Retroviral vectors
For Rec protein expression, HERV-K Rec open reading frames were commercially synthesized (Genscript, Piscataway, NJ, USA) or generated as gBlocks (Integrated DNA Technologies). They were then cloned into murine stem cell virus-based retroviral vectors containing an IRES-eBFP2 cassette (pMSCV-IRES-eBFP2). Cloning was performed using standard restriction enzyme digestion (EcoRI, XhoI; New England Biolabs, Ipswich, MA, USA) followed by T4 DNA ligase-mediated ligation (Thermo Fisher Scientific, Waltham, MA, USA) or using the NEB Builder HiFi Assembly Cloning Kit (New England BioLabs). Each plasmid was assigned a unique identifier (pHRXXXX) as detailed in Table S1. Construct integrity was confirmed by Sanger sequencing (Eton Bioscience Inc).
To generate N-terminal HA-tagged Rec expression constructs, we PCR-amplified each Rec variant using a forward primer encoding a 5′ HA epitope tag (MYPYDVPDYA) and a Kozak consensus sequence immediately upstream of the HA tag start codon. This primer also contained a 17-bp overhang complementary to the MSCV vector backbone. The reverse primer annealed near the 3′ end of the Rec open reading frame and included a complementary overhang to the MSCV vector. PCR was performed using Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific) with the following cycling conditions: initial denaturation, 94°C for 30 s; 25 cycles of 94°C for 15 s, 65°C for 30 s, and 68°C for 30 s; final extension, 68°C for 10 min. The resulting PCR products were cloned into a modified MSCV vector lacking the IRES-eBFP2 cassette using Gibson Assembly (NEB). All constructs were verified by Sanger sequencing. The primer sequences are listed in Table S2.
## Mutational analysis constructs
Rec 12q14.1 variants were generated using synthetic double-stranded DNA fragments (gBlocks; Integrated DNA Technologies) and subsequently cloned into the pMSCV vector upstream of the IRES-eBFP2 cassette via Gibson assembly following the manufacturer's protocol. Each plasmid was assigned a unique identifier (pHRXXXX) as detailed in Table S1. Construct integrity was confirmed by Sanger sequencing (Eton Bioscience Inc.).
## Cell culture and viral vector production
## Cell maintenance
293T/17 cells were maintained in Iscove's modified Dulbecco's medium (IMDM; Gibco, Thermo Fisher Scientific) supplemented with 10% bovine calf serum (BCS, Life Technolo gies) and 50 µg/mL gentamicin (Gibco). Cells were cultured at 37°C in a humidified 5% CO 2 incubator.
## Retroviral vector production
Retroviral vectors expressing Rec variants were produced in 293T/17 cells. Briefly, 8 × 10 6 cells were seeded in 15 cm plates containing 20 mL growth medium (IMDM, 10% BCS, and 50 µg/mL gentamicin). After 24 hours, cells were transfected using polyethyle neimine (PEI; Polysciences, Warrington, PA, USA, 1 mg/mL stock solution, pH 7.0) with a plasmid mixture containing 30 µg pMSCV-Rec construct, 12.84 µg MLV-Pol (pHIT) plasmid, and 5.16 µg VSV-G expression plasmid (pMD2.G, Addgene). The DNA was mixed with PEI at a ratio of 1:3 (DNA:PEI) in serum-free IMDM and incubated for 20 min at room temperature before adding to cells. Following a 6-hour incubation in serum-free IMDM without antibiotics, the transfection medium was replaced with fresh growth medium. Forty-eight hours post-transfection, the viral supernatant was collected and cleared by centrifugation (380 RCF, 4°C, 5 min) to remove cellular debris. The cleared supernatant was then aliquoted (1 mL per aliquot) and stored at -80°C until use. Lentiviral particles containing the tandem RcRE-reporter were generated similarly, except using psPAX2 (Addgene) as the packaging plasmid in a ratio of 12.84 µg psPAX2, 30 µg reporter plasmid, and 5.16 µg pMD2.G per 15 cm plate.
For viral titration, serial 10-fold dilutions were prepared in serum-and antibiotic-free culture medium. 293T/17 cells (3 × 10 4 ) were seeded in 96-well plates 24 hours before transduction. Cells were transduced with 100 µL of diluted viral stocks containing 6 µg/mL DEAE-dextran (Sigma-Aldrich) and incubated for 6 hours at 37°C and 5% CO 2 . Following medium replacement, cells were cultured for 72 hours, then harvested and resuspended in PBS containing 5% BCS. mCherry or eBFP2 expression was quantified using an Attune NxT flow cytometer with autosampler (Thermo Fisher Scientific). Viral titers were calculated as transducing units per milliliter (TU/mL) based on the percentage of fluorescent protein-positive cells in the linear range of dilutions (between 1% and 20% positive cells).
## Generation of stable reporter cell line
A stable 293T/17 reporter cell line (293T/17: pNL4-3[GFP] [HERV-K-2xRcRE] [mCherry]) was established through lentiviral transduction. Briefly, 3 × 10 6 293T/17 cells were seeded in 10 cm dishes 24 hours before transduction. Cells were transduced at an MOI of 0.1 with the lentiviral reporter construct in serum-and antibiotic-free IMDM containing 6 µg/mL DEAE-dextran. Following a 6-hour incubation at 37°C and 5% CO 2 , the virus-containing medium was replaced with standard growth medium.
Transduced cells were subjected to fluorescence-activated cell sorting using a BD Influx System (BD Biosciences, San Jose, CA, USA). Single cells expressing the highest levels of mCherry were isolated and seeded into individual wells of 96-well plates containing IMDM supplemented with 20% BCS and 50 µg/mL gentamicin. After a 2-week expansion period, clonal populations were cryopreserved in growth medium containing 10% DMSO (Sigma-Aldrich). The established clonal cell line was subsequently evaluated for consistent and robust reporter gene expression. Functional assays were performed to confirm the responsiveness of the reporter system to the presence of functional Rec proteins, indicated by GFP expression upon Rec/RcRE interaction.
## Functional analysis of Rec variants
To assess Rec/RcRE functional activity, reporter 293T/17 cells (3 × 10 4 cells/well) were seeded in 96-well plates 24 hours before transduction. Cells were transduced at an MOI of 1 with retroviral vectors expressing individual Rec variants in serum-and antibioticfree RPMI containing 6 µg/mL DEAE-dextran. Following a 6-hour incubation at 37°C and 5% CO 2 , virus-containing medium was replaced with standard growth medium. At 72 hours post-transduction, cells were harvested by trypsinization using 0.25% Trypsin-EDTA (Gibco), resuspended in PBS containing 5% BCS, and analyzed by flow cytometry. Single-color controls were included for proper compensation while measuring mCherry, GFP, and eBFP2 expression.
## Western blot analysis
To analyze protein expression levels, reporter 293T/17 cells were transfected with 2 µg of MSCV plasmids expressing either functional or non-functional Rec variants using Lipofectamine 3000 (Invitrogen, Thermo Fisher Scientific) according to the manufactur er's protocol. At 72 hours post-transfection, cells were harvested and split for parallel flow cytometry analysis and Western blot.
For Western blot samples, cells were washed three times with cold PBS and lysed in buffer containing 50 mM Tris, pH 7.4, 150 mM NaCl, 1.5% SDS, and protease inhibi tors. Lysates were passed through a 21G needle (BD), heat-treated (90°C, 10 min), and cleared by centrifugation (14,000 RCF, 10 min) using a microcentrifuge (Eppendorf, Hamburg, Germany). Samples were prepared for SDS-PAGE by combining 50 µL lysate with NuPAGE LDS Sample buffer (4×) (Invitrogen) and NuPAGE Sample Reducing Agent (10×) (Invitrogen), followed by heating at 70°C for 10 min.
Proteins were separated on Novex 4%-12% Bis-Tris gels (Invitrogen) at 120 V for 1.5 hours using an electrophoresis system and transferred to PVDF membranes at 30 V for 1 hour at 4°C. Membranes were blocked with 5% BSA (Sigma-Aldrich) in TBS (1 hour, room temperature) and incubated overnight at 4°C with primary antibodies (Mouse anti-HA Tag, Cell Signaling Technology; 1:30,000 dilution and Rabbit anti-Beta Tubulin, Cell Signaling Technology; 1:2,000 dilution) diluted in TBS-T (TBS + 0.1% Tween 20) containing 5% BSA.
Membranes were washed three times with TBS-T (10 min each) and were incubated with IRDye-conjugated secondary antibodies (donkey anti-mouse IRDye 800CW, LI-COR Biosciences; 1:30,000 dilution and goat anti-rabbit IRDye 680RD; LI-COR Biosciences, 1:5,000) for 1 hour at room temperature. After three additional washes with TBS-T, protein detection was performed using a LICOR Odyssey CLx scanner, and band intensities were quantified using Image Studio Software (version 5.2, LI-COR Biosciences). For quantification, Rec protein band intensities were normalized to β-tubulin loading controls from the same lanes. Antibodies used are listed in Table S3.
To normalize functional activity to protein expression, we divided the average mean fluorescence intensity, measured by flow cytometry, by the relative protein abundance of each Rec variant as detected by Western blotting of the HA-tagged protein. To accurately compare protein levels between samples, HA-tag signals were first normalized to β-tubulin as a loading control. The resulting normalized activity was calculated as follows:
where MFI is the average mean fluorescence intensity measured by flow cytometry, representing functional activity; HA is the average HA-tagged Rec protein level measured by Western blot; and β-tubulin is the loading control used to normalize the HA signal across samples.
$$Normalized MFI = Mean Fluorescence Intensity MFI HA Expression β-Tubulin Expression ,$$
## Flow cytometry
Cells were harvested by trypsinization and resuspended in PBS containing 5% BCS. Samples were analyzed using an Attune NxT flow cytometer with autosampler (Thermo Fisher Scientific). For each sample, a minimum of 30,000 events were collected. Singlecolor controls were used for compensation using the automated compensation feature of the FlowJo software. Data analysis was performed using FlowJo version 10.6.1 (FlowJo, LLC, BD Biosciences).
The analysis workflow included initial gating on single cells using forward and side scatter properties (FSC-A vs SSC-A) followed by doublet exclusion (FSC-A vs FSC-H). Gates for mCherry, eBFP2, and GFP-positive populations were established using untransduced 293T/17 cells as negative controls. All subsequent analyses were performed on the mCherry-positive population.
Relative Rec/RcRE functional activity was quantified as follows. For Fig. 2B and 5C, it was calculated as the ratio of GFP to eBFP2 mean fluorescence intensity. For all other figures, it was measured as GFP MFI from mCherry-positive populations.
## Trans-dominant negative Rec assays
To measure the effect of co-expressing the functional HERV-K Con with the selected non-functional Rec proteins, 2.5 × 10 5 reporter 293T/17 cells were seeded in 24-well plates and transfected with 100 ng of the pMSCV-HERV-K Con Rec, and either 0, 200, 400, or 800 ng of pMSCV-IRES-eBFP2 vector expressing each of the four test non-functional Recs. Transfections were performed using Lipofectamine 3000 Transfection Reagent (Invitrogen) according to the manufacturer's protocol. For each well, DNA was mixed with 1 µL of the P3000 and 1.5 µL of Lipofectamine 3000 in 50 µL of Opti-MEM. Variable amounts of an empty pMSCV plasmid were added to maintain a constant DNA mass of 2 µg in each transfection. Cells were incubated for 72 hours and subjected to flow cytometry to measure the expression of mCherry and GFP. The same method was used to test the effects of co-expressing HERV-K Con Rec with the Rec 12q14.1 point mutants.
## p24 assays
To assess Rec-mediated viral protein expression, 2.3 × 10 5 293T/17 cells were seeded per well in 12-well plates. Cells were transfected with a constant amount (1,500 ng) of the Gag-Pol-RcRE reporter vector and increasing amounts of Rec expression plasmids (50, 100, 150, and 250 ng). Transfections were performed using PEI as described above. At 72 hours post-transfection, culture supernatants were harvested, centrifuged at 300 × g for 5 min to remove cellular debris, and analyzed for p24 capsid protein using an in-house enzyme-linked immunosorbent assay, as previously described (62).
Briefly, 96-well plates (Nunc MaxiSorp, Thermo Fisher Scientific) were coated with 100 µL of anti-p24 monoclonal antibody (NIH AIDS Reagent Program) at a 1:10,000 dilution in PBS overnight at 37°C. Plates were washed five times with PBS and blocked with 200 µL of PBS containing 5% blocking buffer (PBS with 5% BCS) for 1 hour at 37°C. After washing, 100 µL of culture supernatants or p24 standards (NIH AIDS Reagent Program) ranging from 12.5 to 1,600 pg/mL were added to the wells and incubated for 2 hours at 37°C. Plates were washed and incubated with 100 µL of mouse anti-p24 polyclonal antibody (NIH AIDS Reagent Program) at a 1:10,000 dilution for 1 hour at 37°C. After washing, 100 µL of horseradish peroxidase-conjugated goat anti-mouse IgG (Abcam) at a 1:20,000 dilution was added and incubated for 30 min at 37°C. Plates were washed and developed with 100 µL of substrate for 30 min at room temperature in the dark. The reaction was stopped with 50 µL of 1 N H 2 SO 4 , and absorbance was measured at 450 nm using a microplate reader (BioTek Synergy HTX, BioTek Instruments). A four-parameter logistic regression standard curve was generated, and p24 concentrations in the samples were interpolated from this standard curve. Each sample was assayed in duplicate, and the average value was reported.
## Statistical analysis
Statistical analyses were performed using GraphPad Prism software (version 9.0; GraphPad Software, San Diego, CA, USA). Data are presented as mean ± standard deviation, derived from at least three independent experiments unless otherwise indicated in the figure legends. For comparisons involving more than two experimental groups, statistical significance was determined using ordinary one-way ANOVA followed by Tukey's multiple comparisons test or by non-parametric one-way ANOVA with Dunn's multiple comparisons post hoc test, depending on the outcome of data normality assessments. For comparisons between two groups, either an unpaired two-tailed Student's t-test or the non-parametric Mann-Whitney U test was used, as appropriate. Trans-dominant negative activity assays, involving multiple conditions across different expression levels, were analyzed using ordinary two-way ANOVA followed by Dunnett's multiple comparisons test against a single control (functional Rec alone), assuming pooled variance. For experiments involving multiple comparisons, P-values were adjusted using the Benjamini-Hochberg procedure to control the false discovery rate. A P-value of less than 0.05 was considered statistically significant. Significance thresholds used throughout the manuscript were defined as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. The specific statistical tests applied to individual experiments are indicated in the corresponding figure legends. Graphs were prepared using GraphPad Prism.
## References
1. Bannert, Kurth (2006) "The evolutionary dynamics of human endogenous retroviral families" *Annu Rev Genomics Hum Genet*
2. Katzourakis, Rambaut, Pybus (2005) "The evolutionary dynamics of endogenous retroviruses" *Trends Microbiol*
3. Stoye (2012) "Studies of endogenous retroviruses reveal a continuing evolutionary saga" *Nat Rev Microbiol*
4. Mayer, Blomberg, Seal (2011) "A revised nomenclature for transcribed human endogenous retroviral loci" *Mob DNA*
5. Subramanian, Wildschutte, Russo et al. (2011) "Identification, characterization, and comparative genomic distribution of the HERV-K (HML-2) group of human endogenous retroviruses" *Retrovirology (Auckl)*
6. Jern, Coffin (2008) "Effects of retroviruses on host genome function" *Annu Rev Genet*
7. Bannert, Kurth (2004) "Retroelements and the human genome: new perspectives on an old relation" *Proc Natl Acad Sci*
8. Gifford, Tristem (2003) "The evolution, distribution and diversity of endogenous retroviruses" *Virus Genes*
9. Grandi, Tramontano (2018) "Human endogenous retroviruses are ancient acquired elements still shaping innate immune responses" *Front Immunol*
10. Rekosh, Hammarskjold (2018) "Intron retention in viruses and cellular genes: detention, border controls and passports" *Wiley Interdiscip Rev RNA*
11. Magin, Löwer, Löwer (1999) "cORF and RcRE, the Rev/Rex and RRE/ RxRE homologues of the human endogenous retrovirus family HTDV/ HERV-K" *J Virol*
12. Hohn, Hanke, Bannert (2013) "HERV-K(HML-2), the Best Preserved Family of HERVs: endogenization, expression, and implications in health and disease" *Front Oncol*
13. Denne, Sauter, Armbruester et al. (2007) "Physical and functional interactions of human endogenous retrovirus proteins Np9 and rec with the promyelocytic leukemia zinc finger protein" *J Virol*
14. Heyne, Kölsch, Bruand et al. (2015) "Np9, a cellular protein of retroviral ancestry restricted to human, chimpanzee and gorilla, binds and regulates ubiquitin ligase MDM2" *Cell Cycle*
15. Yang, Bogerd, Peng et al. (1999) "An ancient family of human endogenous retroviruses encodes a functional homolog of the HIV-1 Rev protein" *Proc Natl Acad Sci*
16. Magin-Lachmann, Hahn, Strobel et al. (2001) "Rec (formerly Corf ) function requires interaction with a complex, folded RNA structure within its responsive element rather than binding to a discrete specific binding site" *J Virol*
17. Magin, Hesse, Löwer et al. (2000) "Corf, the Rev/Rex Homologue of HTDV/HERV-K, encodes an arginine-rich nuclear localization signal that exerts a trans-dominant phenotype when mutated" *Virology (Auckl)*
18. Ruggieri, Maldener, Sauter et al. (2009) "Human endogenous retrovirus HERV-K(HML-2) encodes a stable signal peptide with biological properties distinct from Rec" *Retrovirology (Auckl)*
19. Malim, Hauber, Le et al. (1989) "The HIV-1 rev trans-activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA" *Nature*
20. Hadzopoulou-Cladaras, Felber, Cladaras et al. (1989) "The rev (trs/art) protein of human immunodeficiency virus type 1 affects viral mRNA and protein expression via a cisacting sequence in the env region" *J Virol*
21. Hammarskjöld, Heimer, Hammarskjöld et al. (1989) "Regulation of human immunodeficiency virus env expression by the rev gene product" *J Virol*
22. Hidaka, Inoue, Yoshida et al. (1988) "Post-transcriptional regulator (rex) of HTLV-1 initiates expression of viral structural proteins but suppresses expression of regulatory proteins" *EMBO J*
23. Seiki, Inoue, Hidaka et al. (1988) "Two cis-acting elements responsible for posttranscriptional trans-regulation of gene expression of human T-cell leukemia virus type I" *Proc Natl Acad Sci*
24. Mertz, Simper, Lozano et al. (2005) "Mouse mammary tumor virus encodes a self-regulatory RNA export protein and is a complex retrovirus" *J Virol*
25. Fuentes, Swigut, Wysocka (2018) "Systematic perturbation of retroviral LTRs reveals widespread long-range effects on human gene regulation"
26. Grow, Flynn, Chavez et al. (2015) "Intrinsic retroviral reactivation in human preimplantation embryos and pluripotent cells" *Nature*
27. Buzdin, Kovalskaya-Alexandrova, Gogvadze et al. (2006) "At least 50% of human-specific HERV-K (HML-2) long terminal repeats serve in vivo as active promoters for host nonrepetitive DNA transcription" *J Virol*
28. Gray, Jackson, Jackson et al. (2019) "HIV-1 Rev interacts with HERV-K RcREs present in the human genome and promotes export of unspliced HERV-K proviral RNA" *Retrovirology (Auckl)*
29. Galli, Sauter, Lecher et al. (2005) "Human endogenous retrovirus rec interferes with germ cell development in mice and may cause carcinoma in situ, the predecessor lesion of germ cell tumors" *Oncogene*
30. Hanke, Chudak, Kurth et al. (2013) "The Rec protein of HERV-K(HML-2) upregulates androgen receptor activity by binding to the human small glutamine-rich tetratricopeptide repeat protein (hSGT)" *Int J Cancer*
31. Lemaître, Harper, Pierron et al. (2014) "The HERV-K human endogenous retrovirus envelope protein antagonizes tetherin antiviral activity" *J Virol*
32. Byun, Halani, Mertz et al. (2010) "Retroviral Rem protein requires processing by signal peptidase and retrotranslocation for nuclear function" *Proc Natl Acad Sci*
33. Huang, Li, Wang et al. (2013) "Human endogenous retroviral K element encodes fusogenic activity in melanoma cells" *J Carcinog*
34. Zhou, Li, Wei et al. (2016) "Activation of HERV-K Env protein is essential for tumorigenesis and metastasis of breast cancer cells" *Oncotarget*
35. Löwer, Boller, Hasenmaier et al. (1993) "Identification of human endogenous retroviruses with complex mRNA expression and particle formation" *Proc Natl Acad Sci*
36. Brady, Lee, Malani et al. (2009) "Integration target site selection by a resurrected human endogenous retrovirus"
37. Mayer, Ehlhardt, Seifert et al. (2004) "Human endogenous retrovirus HERV-K(HML-2) proviruses with Rec protein coding capacity and transcriptional activity" *Virology (Auckl)*
38. Schmitt, Heyne, Roemer et al. (2015) "HERV-K(HML-2) rec and np9 transcripts not restricted to disease but present in many normal human tissues" *Mob DNA*
39. Pasternack, Doucet-O'hare, Johnson et al. (2024) "Endogenous retroviruses are dysregulated in ALS"
40. Jackson, Huang, Sharma et al. (2019) "A novel retroviral vector system to analyze expression from mRNA with retained introns using fluorescent proteins and flow cytometry" *Sci Rep*
41. Jackson, Dzhivhuho, Huang et al. (2024) "Measurement of HIV Rev-Rev response element functional activity" *Methods Mol Biol*
42. Wildschutte, Williams, Montesion et al. (2016) "Discovery of unfixed endogenous retrovirus insertions in diverse human populations" *Proc Natl Acad Sci*
43. Kyriakou, Magiorkinis (2025) "Compilation of all known HERV-K HML-2 proviral integrations" *Mob DNA*
44. Tönjes, Czauderna, Kurth (1999) "Genome-wide screening, cloning, chromosomal assignment, and expression of full-length human endogenous retrovirus type K" *J Virol*
45. Shin, Mun, Han (2023) "Human endogenous retrovirus-K (HML-2)related genetic variation: human genome diversity and disease" *Genes (Basel)*
46. Turner, Barbulescu, Su et al. (2001) "Insertional polymorphisms of full-length endogenous retroviruses in humans" *Curr Biol*
47. Kahyo, Yamada, Tao et al. (2017) "Insertionally polymorphic sites of human endogenous retrovirus-K (HML-2) with long target site duplications" *BMC Genomics*
48. Yang, Guo, Li et al. (2022) "Significant upregulation of HERV-K (HML-2) transcription levels in human lung cancer and cancer Full-Length Text Journal of Virology December"
49. *cells. Front Microbiol*
50. Zare, Mostafaei, Ahmadi et al. (2018) "Human endogenous retrovirus env genes: potential blood biomarkers in lung cancer" *Microb Pathog*
51. Lemaître, Tsang, Bireau et al. (2017) "A human endogenous retrovirus-derived gene that can contribute to oncogenesis by activating the ERK pathway and inducing migration and invasion" *PLoS Pathog*
52. Rycaj, Plummer, Li et al. (2012) "Immunotherapeutic potential of anti-human endogenous retrovirus-K envelope protein antibodies in targeting breast tumors" *J Natl Cancer Inst*
53. Büscher, Trefzer, Hofmann et al. (2005) "Expression of human endogenous retrovirus K in melanomas and melanoma cell lines" *Cancer Res*
54. Serafino, Balestrieri, Pierimarchi et al. (2009) "The activation of human endogenous retrovirus K (HERV-K) is implicated in melanoma cell malignant transformation" *Exp Cell Res*
55. Singh, Cai, Bunse et al. (2020) "Human endoge nous retrovirus K rec forms a regulatory loop with MITF that opposes the progression of melanoma to an invasive stage" *Viruses*
56. Downey, Sullivan, Wang-Johanning et al. (2015) "Human endogenous retrovirus K and cancer: innocent bystander or tumorigenic accomplice?" *Int J Cancer*
57. Dzhivhuho, Holsey, Honeycutt et al. (2022) "HIV-1 Rev-RRE functional activity in primary isolates is highly dependent on minimal context-dependent changes in" *Rev. Sci Rep*
58. Boese, Galli, Geyer et al. (2001) "The Rev/Rex homolog HERV-K cORF multimerizes via a C-terminal domain" *FEBS Lett*
59. Malim, Mccarn, Tiley et al. (1991) "Mutational definition of the human immunodeficiency virus type 1 Rev activation domain" *J Virol*
60. Legiewicz, Badorrek, Turner et al. (2008) "Resistance to RevM10 inhibition reflects a conformational switch in the HIV-1 Rev response element" *Proc Natl Acad Sci*
61. Bevec, Dobrovnik, Hauber et al. (1992) "Inhibition of human immunodeficiency virus type 1 replication in human T cells by retroviralmediated gene transfer of a dominant-negative Rev trans-activator" *Proc Natl Acad Sci*
62. Contreras-Galindo, Kaplan, Dube et al. (2015) "Human endogenous retrovirus type K (HERV-K) particles package and transmit HERV-K-related sequences" *J Virol*
63. Wehrly, Chesebro (1997) "P24 antigen capture assay for quantification of human immunodeficiency virus using readily available inexpensive reagents" *Methods*
64. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12473741&blobtype=pdf | # Evolving Dynamics of Whole-Genome Influenza A/H3N2 Viruses Isolated in Cameroon
Toutou Desmon, Tsafack, Chavely Gwladys Monamele, Henri Moïse, Moumbeket-Yifomnjou, Loique Landry, Messanga Essengue, Chanceline Bilounga, Mohamadou Ripa Njankouo, Pascal Touoyem, Ubald Tamoufe, Francioli Koro Koro, Richard Njouom
## Abstract
Background: Since 2023, Cameroon has recorded numerous cases of seasonal infuenza caused by the A/H3N2 subtype, which is the strain most commonly encountered worldwide in 2024. Methods: To describe the evolutionary dynamics of infuenza A/H3N2 viruses, whole genome sequencing was performed using the Oxford Nanopore Technologies sequencing platform and the SQK-LSK109, EXP-NBD196 reagent kit (Oxford Nanopore Technologies, catalog no. SQK-LSK109). Subsequently, mutational analysis was performed on the 8 genes of the H3N2 infuenza strains isolated between 2023 and 2024 in Cameroon by aligning our protein sequences with the reference sequences recommended by the WHO in the northern hemisphere during the 2023-2024 infuenza season using MEGA 11 software. Te trimeric and tetrameric structures of the HA, NA, and M proteins were downloaded from the protein website https://www.rcsb.org/ and imported into the PyMOL Version 2.6.1 software for visualization and annotation of the observed amino acid substitutions. Results: All Cameroonian A(H3N2) strains from 2023 to 2024 belonged to clade 3c.2a. Te mutations I208F, K156I, E66K, N112S, G69N, V239I, K292E, Q189R, G241D, A202D, T3A, S70R, N161S, N138D, N394S, and N120D were detected in most HA1 gene samples (Supporting Table S1). Among these mutations, the important A202D and N161S mutations in HA1 in 2023 and 2024 led to the virulence of the virus and consequently resulted in the rapid evolution of the A/H3N2 virus and the generation of the new clades 3C.2a1b.2a.2a.3 and 3c.2a1b.2a.2a.3a.1, respectively. Similarly, amino acid substitutions at sites I469T, I65V, and H275Y in the NA protein were observed compared to the 2024 vaccine strain A/Darwin/6/2021. We noted the presence of the H275Y substitution in 30% of Cameroonian strains associated with major resistance to neuraminidase inhibitors, particularly oseltamivir. In general, the number of amino acid mutations observed between circulating strains and the vaccine strain for the following year was higher, indicating that circulating strains would evolve away from vaccine strains for the year 2023-2024. Conclusions: Tese results highlight the evolutionary nature of the human infuenza virus.
## 1. Background
Infuenza virus infection poses a serious threat to human life and health. Each year, approximately one billion cases of seasonal infuenza are recorded, including 3-5 million severe cases, resulting in 290,000 to 650,000 respiratory-related deaths [1,2], making it one of the most signifcant public health issues worldwide. Seasonal H3N2 infuenza virus is an 8-segmented RNA virus, encompassing the genes for hemagglutinin (HA), neuraminidase (NA), matrix protein (MP), nonstructural (NS), nucleoprotein (NP), polymerase acidic (PA), polymerase basic 1 (PB1), and polymerase basic 2 (PB2). Tese genetic segments collectively encode essential proteins including HA, NA, MP 1 (M1), MP 2 (M2), nonstructural protein (NS1), nuclear export protein (NEP), NP, PA, PA-X, PB1, PB1-F2, and PB2 [3,4]. Based on the major surface antigens, there are 18 HA (H1-H18) and 11 NA (N1-N11) subtypes of the infuenza A virus [5]. Since their emergence in 1968, infuenza H3N2 viruses have been highly prevalent, with the H3 HA and N2 NA surface glycoproteins being the dominant proteins in these strains [6]. Te activities of both HA and NA are crucial for viral function, with antibodies targeting these proteins serving as the primary defense against infection [7]. HA, the main surface antigen of the H3N2 seasonal infuenza virus, comprises fve antigenic sites: Region A (amino acids [AAs] [8]. While the dynamics of infuenza epidemics are complex and not fully understood, one major driver of seasonal variation is antigenic drift [9]. For example, in Canada, antigenic drift was observed in 2008 when the A/Brisbane/10/ 2007 strain mutated into the A/Perth/16/2009 strain, which was isolated for the frst time in Australia and grouped within clade 3C.3a and was used as a vaccine strain in subsequent years [10]. In Cameroon, Monamele et al. confrmed several antigenic site mutations among H3N2 virus strains during the 2014-2016 infuenza seasons. Tis study briefy describes the frequency of infuenza cases and the evolutionary dynamics of the H3N2 seasonal infuenza virus in Cameroon over two successive years, from 2023 to 2024. In addition, it carries out comprehensive analyses of homology, evolution, and variation within the complete genomic sequences of 33 strains of the H3N2 seasonal infuenza virus over the same period. Tese results make a signifcant contribution to our understanding of the evolutionary and variation characteristics of the complete genome of the H3N2 seasonal infuenza virus, providing upto-date epidemiological data for predicting future infuenza epidemics and prevention and control strategies.
## 2. Materials and Methods
## 2.1. Sample Collection and Preparation.
In accordance with authorization no. 3971CEI-Udo/07/2023/M from the Ethical and Institutional Committee of the University of Douala, nasopharyngeal swabs were collected from patients with acute respiratory infection (ARI) at 19 infuenza sentinel surveillance sites in Cameroon between January 2023 and December 2024. ARI was defned according to the European Centre for Disease Prevention and Control (ECDC) guidelines, which include patients with infuenza-like illness (ILI) or severe acute respiratory infection (SARI) according to the WHO case defnitions. Nasopharyngeal specimens were collected using polyester swabs and stored at 4 °C in 2 mL virus transport medium and transported to the Centre Pasteur of Cameroun (CPC). Samples were processed immediately or stored at -80 °C prior to analysis.
## 2.2. Extraction of Viral Nucleic
Acids and RT-PCR. Viral RNA was extracted from nasopharyngeal swabs using the QIamp viral RNA Kit (Qiagen, Hilden, Germany) according to the manufacturer's recommendations. In summary, 140 μL of nasopharyngeal swab specimen was utilized for nucleic acid extraction. Detection and typing/subtyping of infuenza viruses were performed using the SuperScript III Platinum One-Step Quantitative RT-PCR (qRT-PCR) System (Invitrogen, USA). All samples were tested with a multiplex kit targeting infuenza A and B viruses. Positive infuenza A samples were further subtyped for A/H3N2. Amplifcation was conducted on an ABI Prism 7500 thermocycler (Applied Biosystem, Foster City, CA, USA). A 20 μL master mix was prepared, consisting of 1 µL water, 12.5 µL bufer (2X), 0.5 µL enzyme reverse transcriptase/Taq polymerase, 2 µL forward and 2 µL reverse primers (10 μM each), and 2 µL probe (2.5 µM). Five microliters of extracted RNA were added to each sample or control (negative and positive). A threshold cycle (Ct) value below 37 was considered positive. Nucleic acid testing was completed within 24 h.
## 2.3. Whole Genome Sequencing of A (H3N2).
For whole genome sequencing, viral RNA was extracted and subjected to capture and amplifcation using the ULSEN Ultrasensitive Infuenza Virus Whole Genome Capture Kit (Low Load) from Beijing We Future Technology Co., Ltd. (Catalog no. V-090417). Before the PCR amplifcation step, superscript III, which is a more thermally stable enzyme than other reverse transcriptases and can operate at higher temperatures (up to 55 °C or even 60 °C), was frst used to convert RNA into cDNA. Next, to fragment the DNA and repair the ends of these fragments, we used the Ultra II enzyme. Finally, the library was prepared by attaching adapters using Quick T4 DNA ligase. Te protocol involved incubating the sample at 42 °C for 50 min (1 cycle), followed by denaturation at 94 °C for 30 s. Tis was followed by 4 cycles of denaturation at 94 °C for 30 s, annealing at 57 °C for 30 s, and extension at 68 °C for 3 min and 30 s. Tis was followed by 10 cycles of denaturation at 94 °C for 30 s, annealing at 57 °C for 30 s, and extension at 68 °C for 3 min and 30 s. Te process was completed with a single fnal extension at 68 °C for 10 min, followed by cooling to 4 °C. Te products obtained were purifed using the AMPure XP beads nucleic acid purifcation kit (Beckman Coulter: catalog no. A63880, A63881, A63882). Nucleic acid quantifcation was performed using the Qubit 4 dsDNA HS Assay Kit fuorometer (Termo Fisher Scientifc: catalog no. Q32851-100 assays). Next, cDNA fragmentation was performed using the NEBNext Ultra II End Repair/dA-Tailing Module Kit from Oxford Nanopore Technologies (USA, catalog no. E7546). Adapter ligation was performed using the NEBNext Quick Ligation Module Kit from New England Biolabs (NEB: catalog no. E6056). Finally, whole genome sequencing was performed using the Oxford Nanopore Technologies sequencing platform and the SQK-LSK109 reagent kit, EXP-NBD196 (Oxford Nanopore Technologies, catalog no. SQK-LSK109).
2.4. Sequence Alignment and Analysis. Te sequencing data were processed and analyzed using BioEdit (Version 7.2.5). Te HA, NA, and M sequences of the northern hemisphere infuenza virus reference strains for 2023-2024 (A/Darwin/6/ 2021) used in the phylogenetic analysis were obtained from the Global Initiative on Sharing All Infuenza Data (GISAID, https://www.gisaid.org) and are listed in Supporting Table S1. All sequences were aligned using Multiple Alignment using Fast Fourier Transform (MAFFT Version 6.864) software. Te genetic analysis was based on mutations causing AA substitutions. Using MEGA Version 11 software, the sequences of the eight segments obtained were compared to the virus sequence (A/Darwin/6/2021) (H3N2) (reference strain for the 2023-2024 infuenza vaccine) as presented in Supporting Table S1. Te phenotypic properties of eacvdh identifed mutation were determined using MEGA software. Phylogenetic trees for the HA, NA and M genes were generated using the maximum likelihood (ML) method with MEGA (Version 11) and visualized with FigTree (Version 1.4.4), using the study sequences, reference sequences and representative sequences from other regions. In addition, a similarity analysis was performed between the genes and encoded proteins of the seasonal infuenza H3N2 virus in Cameroon from 2023 to 2024 and strain A/Darwin/6/2021, a vaccine strain of the seasonal infuenza H3N2 virus from the northern hemisphere. Te robustness of the tree topology was assessed using 1000 bootstrap replications, with values above 70% indicated on the branches of the tree. All sequence data analyzed in this study have been deposited in the GISAID repository, and the accession numbers are detailed in Supporting Table S1.
## 2.5. Classifcation of Subclades by Amino Acid Substitutions in HA.
All sequences were aligned to the selected reference strain. Te reference strain used in this study corresponded to the WHO-recommended vaccine strain for the Northern Hemisphere for 2023-2024, which was retrieved from the GISAID database. Te subclades of the 33 A/H3N2 strains in this study were determined by key AA substitutions in HA that defne A/H3N2 subclades, as proposed by the WHO. Key AA substitutions based on A/Darwin/6/2021, a 2024 Northern Hemisphere vaccine strain, were used to classify the HA 3C.2a group comprising 3C.2a1.2a.2a.1b, 3C.2a1b.2a.2b, 3C.2a1b.1a, and 3C.2a1b.2a.2a.3a.1. Te subclassifcation of vaccine strains for each year was performed based on HA substitutions. Genetic analysis of AA substitutions was performed using MEGA V.7.0.26 software (https://www.megasoftware.net/; accessed on July 17, 2025). In order to better understand the classifcation of clades and the grouping of viruses, a phylogenetic tree was constructed from 67 strains from other countries around the world. Tese were then compared to vaccine strains from the Northern Hemisphere belonging to known clades recommended by the WHO.
2.6. 3D Structure of HA, NA, and MP Proteins. In order to highlight the AA substitutions observed in the three genes, the codes for the HA, NA, and NA proteins, 4O5N, 1ING, and 8RNG, respectively, were uploaded to the Protein Data Bank (PDB) along with the various trimeric and tetrameric structures corresponding to these proteins. Tese three structures were then imported into the PyMOL software for visualization and annotation of the observed AA substitutions. Te transparency of the contours and the surface was adjusted to 80%.
## 3. Results
## 3.1. Description of the Socio-Demographic Characteristics of the Study Population.
Using stratifed random sampling, 33 real-time PCR-positive nasopharyngeal specimens from infuenza patients with Ct < 30 were selected and included in our study. Of these samples, 30 (90%) were from individuals attending outpatient clinics (ILI), while 3 (09%) were from hospitalized patients (SARI). Patients ranged in age from 1 to 90 years, with 19 out of 33 (57.6%) being female. Myalgias, cough, and vomiting were the major symptoms observed in 90% of the patients included in this study. Infuenza A (H3N2) viruses were identifed from both outpatients and inpatients. In contrast, the 50-64 and over-65 age groups were only slightly afected by infuenza infection during this period. Additionally, the predominance of A/H3N2 cases was greatest in the 5-14 age group (18/33), followed by the 0-1 age group (10/33). Te majority of infuenza cases were observed during the infuenza period from October to December of both study years.
## 3.2. Nucleotide Diversity of Infuenza A/H3N2 Viruses.
For the frst time ever in Cameroon, 33 complete genomes of infuenza A (H3N2) viruses have been successfully sequenced. Similarity analyses showed that the nucleotide similarity of eight gene fragments in the sequences of the 33 strains ranged from 98.1% to 100%. Compared to the A/Darwin/6/2021 strain, the nucleotide similarity of the eight gene segments ranged from 98.1% to 100%. AA similarity among the encoded proteins varied between 97.3% and 100%, as shown in Table 1. In summary, this study provides clear information on the genetic characteristics and evolutionary distances of infuenza A (H3N2) viruses, which will be used to select new vaccine strains.
## 3.3. Analysis of Genetic Variation and Evolution of H3N2
Infuenza Viruses. During our study, the genetic evolution of A/H3N2 infuenza viruses in Cameroon was analyzed using the sequences of their eight segments. For the segments (HA, NA, and M), a phylogenetic tree was constructed from the sequences of the eight segments of 33 A/H3N2 strains, including the global vaccine strain recommended by the WHO (2023-2024) and strains from other countries. Te results revealed that all sequences from these eight segments were closely related and located in the same clade, 3c.2a. It should be noted that the HA genes from these sequences all belonged to subclade 3c.2a1b.2a.2a.3a.1. Te remaining integrated strains evolved in a distinct evolutionary clade, with their HA genes belonging to subclades 3c.2a1b.2a.2a.1b, 3c.2a1b.2a.2b, and fnally to subclade 3c.2a1b.1a (Figure 1). Overall, phylogenetic analysis of the eight-segment sequences of infuenza A/H3N2 viruses in Cameroon revealed four distinct evolutionary subclades, with the HA genes of the sequences belonging either to subclade 3c.2a1b.2a.2a.1b, or subclade 3c.2a1b.2a.2b, or subclade 3c.2a1b.1a, and fnally subclade 3c.2a1b.2a.2a.3a.1.
## 3.4. Subclade Analysis by AA Substitutions and Phylogenetic
Tree Analysis of HA, NA, and MP Genes. Te 33 strains from Cameroon all had AA substitutions at sites I208F, K156I, E66K, N112S, G69N, V239I, K292E, Q189R, G241D, A202D, T3A, S70R, N161S, N138D, N394S, N120D, A16T, G78E, and T144E in HA1 compared to the 2024 vaccine strain A/Darwin/6/2021. Te results indicated that the 33 A/H3N2 viruses from Cameroon collected between 2023 and 2024 belonged to clade 3C.2a. Of the 33 strains, analysis of AA substitutions in HA showed that the majority of strains from 2023 to 2024 belonged to subclade 3c.2a1b.2a.2a.3a.1 (n � 70%). Te remaining strains belonged to subclade 3C.2a1b.2a.2a.3 (n � 30%). All of the specifc AA mutations in HA1, which defne the subclades relative to the northern hemisphere vaccine strain A/Darwin/6/2021, are summarized in Supporting Table S2. Phylogenetic analysis of HA confrmed that all Cameroon sequences grouped in clade 3c.2a1b.2a.2a.3a.1 evolved separately from the reference vaccine strain for the 2023-2024 infuenza season (Figure 1) and its subclades according to the AA defned in HA by the WHO. Furthermore, the number of AA mutations observed between circulating strains and the vaccine strain for the infuenza season was higher, which also indicates that the circulating strains were diferent from the recommended vaccine strain. In order to evaluate the genetic relationship between the strains from Cameroon and those circulating worldwide, we performed a BLASTsearch for the HA gene of 67 A/H3N2 sequences and constructed a phylogenetic tree. Te HA gene in Cameroon was closest to strains isolated in South Africa, Togo, Burkina Faso, Niger, China, Belgium, and Italy between 2023 and 2024 (Figure 1). We also noted that the A/H3N2 viruses from 2023 to 2024 in Cameroon, belonging respectively to subclades 3c.2a1b.2a.2a.3a.1 and 3C.2a1b.2a.2a.3, were closely related to subclade 3C.2a1b.2a. 2b. Similarly, AA substitutions at sites I469T, I65V, P45S, I392T, L140I, L338V, H150R, K400R, and H275Y in the NA protein were observed compared to the vaccine strain 2024 A/Darwin/6/2021. We noted the presence of the H275Y substitution in 30% of Cameroonian strains associated with major resistance to NA inhibitors, specifcally oseltamivir. Phylogenetic analysis of NA confrmed that all Cameroonian sequences grouped in clade 3c.2a1b.2a.2a.3a.1 evolved separately from the reference vaccine strain for the 2023-2024 infuenza season (Figure 2). In addition, AA substitutions at sites N85S, L59I, D24N, V27I, Y52C, F54L, L25P, S82N, E66K, and S31N in the MP protein were observed compared to the vaccine strain 2024 A/Darwin/6/ 2021. We noted the presence of the major substitution S31N in some Cameroonian strains, which is known to confer a high level of resistance to amantadine. All of the specifc AA mutations in NA are summarized in Supporting Table S3. Phylogenetic analysis of MP confrmed that all but a few of the Cameroonian sequences grouped in clade 3c.2a1b.2a.2a.3a.1 evolved separately from the reference vaccine strain (Figure 3). All of the specifc AA mutations in MP are summarized in Supporting Table S4.
## 3.5. Analysis of Comparison of AA Substitutions Between Circulating Strains and A/Darwin/6/2021 Vaccine Strain in
Eight Genes. Compared to the 2023-2024 vaccine strain A/Darwin/6/2021, the mutations I208F, K156I, E66K, N112S, G69N, V239I, K292E, Q189R, G241D, A202D, T3A, S70R, N161S, N138D, N394S, A16T, G78E, and N120D were detected in most samples of HA1 genes (Supporting Table S2). Among these mutations, the important A202D shown in blue and brown, respectively. All viruses were detected in unvaccinated outpatients and hospitalized patients. Phylogenetic trees for the MP genes were generated using the maximum likelihood (ML) method with MEGA (Version 11). Te robustness of the tree topology was assessed with 1000 bootstrap replicates, with values above 70% indicated on tree branches. and N161S mutations in HA1 in 2023 and 2024 led to virus virulence and consequently conferred rapid evolution of the A/H3N2 virus and the generation of the new clades 3C.2a1b.2a.2a.3 and 3c.2a1b.2a.2a.3a.1, respectively. Similarly, mutational analyses were extended to the seven longest viral genes (PB2, PB1, PA, NS, and NP; Supporting Tables S5-S9). In the NP gene, a total of 9 AA substitutions (E220D, L418I, L136M, I186V, A129S, N432S, S482N, K236R, and S359L) were observed. In the PA gene, a total of 18 AA substitutions (S402A, G684R, K142N, E101G, M311I, R605K, K142N, R213K, G99E, A660S, I407S, C321Y, A20T, K269R, K497R, R158K, Y277H, and L400I) were observed. Similarly, new AA substitutions not previously described have been observed in the PB1 gene (V200I, S375N, R386K, and V527I) and PB2 (R340K, D87N, V461I, N107D, T147I, M410V, and L384F). Numerous mutations have also been identifed in the NS genes, notably the substitutions E152D, N207H, I18V, N127S, N26K, A60V, R227G, I33L, I124M, A82V, V171I, L14S, and K88R. In general, the number of AA mutations observed between circulating strains and the vaccine strain for the following year was higher, indicating that circulating strains would evolve away from the vaccine strains for the year 2023-2024.
3.6. 3D Structure of HA, NA, and MP Proteins. Te threedimensional structures of the HA, NA, and MP proteins from sequences in clade 3C.2a1b.2a.2a.3a.1 were modeled. Te HA protein exhibited several mutations, including I208F, K156I, E66K, N112S, G69N, V239I, K292E, Q189R, A202D, T3A, S70R, N161S, N138D, N394S, N120D, and G241D. In the NA protein, mutations such as I469T, P45S, I65V, H275Y, I392T, L140I, L338V, H150R, and K400R were identifed. Additionally, the M protein displayed variations at sites N85S, L59I, D24N, V27I, Y52C, S31N, F54L, L25P, S82N, and E66K. Te variation sites were distinctly marked in the three-dimensional structures with diferent colors, as illustrated in Figure 4.
## 4. Discussion
Several studies have utilized whole infuenza virus genome sequences and modern software to understand the evolutionary dynamics of the infuenza A/H3N2 subtype [11][12][13]. Te replication of the RNA genome in infuenza viruses is associated with a relatively high mutation rate (2.3 × 10 -5 ), primarily because the viral RNA-dependent RNA polymerase lacks 3′-5′-exonuclease activity, leading to an absence of proofreading functions [14,15]. Tis study represents the second report of genome characterization of infuenza H3N2 in Cameroon, following an earlier study by Monamele et al. [16] that focused on partial segments of the HA, NA, and M genes from 2014 to 2016. Te WHO infuenza established a global surveillance network and recommends annual infuenza vaccine strains for the northern and southern hemispheres. Cameroon is one of the most crucial member countries for infuenza surveillance and has established a surveillance network covering all cities in the country.
In the HA genes of the isolated strains, signifcant substitutions such as E66K, I208F, N112S, N394S, G241D, G69N, K156I, V239I, K292E, Q189R, N175S, R108K, G291D, and A202D were observed. Among these mutations, the important A202D and N161S mutations in HA1 in 2023 and 2024 led to virus virulence and consequently conferred rapid evolution of the A/H3N2 virus and the generation of the new clades 3C.2a1b.2a.2a.3 and 3c.2a1b.2a.2a.3a.1. Tese fndings contrast with those observed in southern China in 2012, where the D69N, Y110H, I246V, and E296A/T substitutions involved in the rapid evolution of the A/H3N2 virus were identifed [17]. Te observed mutational diversity from year to year may be attributed to AA changes at sites associated with human leukocyte antigen (HLA), which alter the HA antiglobulin antibody recognition sites.
Similarly, the mutations involved in the virulence of Cameroonian strains circulating between 2023 and 2024 difer from those reported in 2017 by Monamele et al., who identifed the mutations N145S, Y186G, P198S, and F219S in the HA polypeptide. Tis variability highlights the signifcant antigenic diversity that exists within the binding sites of the HA gene. Tis antigenic variability observed within the HA gene can also be confrmed by the results obtained by Ramuth et al. in 2025, which showed signifcant antigenic diversity following the cocirculation of subclades 3C.2a4 and 3C.2a1 in 2017, while the predominant subclade in 2018 was subclade 3C.2a1b.1. Phylogenetic analysis of the HA gene of A(H3N2) viruses showed that the vast majority of viruses circulating in Cameroon during the 2023-2024 infuenza season belonged to clade 2 (full classifcation 3C.2a1b.2a.2) and had acquired several AA substitutions. Te diferent HA subclades were found in different regions of the world, and viruses with HA genes from several subclades cocirculated in several geographical regions in varying proportions. Among the viruses with HA genes from clade 2 that cocirculated during this period, three subclades predominated: 2a.1b (generally encoding D53G, D104G, I140K, K276R, and R299K), 2a.3a.1 (generally encoding E50K, G53N, N96S (CHO+), I140K, I192F, I223V, and N378S), and 2b (generally encoding E50K, F79V, and I140K). Among these, subclade 2a.3a.1, which includes more than 70% of Cameroonian strains (e.g., A/Massachusetts/18/2022), was predominant worldwide. Tese viruses were detected mainly in Africa, Asia, North America, and Oceania. No Cameroonian strains were grouped in subclade 2a.1b, which was detected mainly in North America and Europe, as shown in Figure 1. Clade 2b viruses circulated globally. In general, postinfection ferret antisera produced against SH 2023 vaccine viruses (A/ Darwin/6/2021 viruses propagated in cell culture and A/Darwin/9/2021 2a viruses propagated in eggs) recognized viruses expressing HA 2a genes (including subclades) well. However, some viruses expressing HA 2a.3a.1 genes, such as the majority of Cameroonian strains (70%), reacted less well with these antisera during the 2023-2024 infuenza season. Tis evolution of clade 2a.3a.1 viruses away from the A/Darwin/6/ 2021 vaccine reference virus was also observed in 2017 by Monamele et al., who showed that as the 2012 fu season approached, the HA gene sequences indicated that all Cameroonian strains had evolved away from the 3C.1-A/ Darwin/6/2021 clade. Tis observation was confrmed in 2019 in northern Cameroon by Njifon et al. [18] and also in 2022 in the city of Myanmar, Japan, by Phyu et al., who observed that the Myanmar strains difered from the Southern Hemisphere vaccine strains each year, indicating a mismatch between the vaccine strains and the circulating strains [19]. Tese results update the fndings of Monamele et al., who showed in 2017 that Cameroonian strains formed two distinct groups. Tese illustrations indicate that the molecular characterization of the infuenza virus varies from year to year, region to region, and country to country, as several authors have pointed out [20][21][22]. Furthermore, the NA and M strains of infuenza from 2023 to 2024 did not cluster with any strain but closely resembled the 2019-2020, 2021-2022, and 2022-2023 strains (3C.3a-A/ Kansas/14/2014, 3C.2a-A/Cambodia/E0826360/2020, and In 2011, a Canadian study looking at the antigenic and molecular characterization of H3N2 viruses over three seasons revealed signifcant HA mutations, as well as the nature and location of the main mutations, which played a crucial role in antigenic drift. Tis study again confrms the significant mutational variation observed in the HA binding sites of Cameroonian strains. Furthermore, the results of our study showed that 30% of Cameroonian strains, that is, 10/33, had a point mutation (cytosine to thymidine) at position 823, which causes the substitution of histidine by tyrosine at position 275 in the AA sequence of NA (H275Y). Tis substitution has been reported by Pinella et al. as conferring more than 1.5% resistance to oseltamivir in patients with ARI. We also noted the absence of the S31N substitution present in the M2 protein of many previous Cameroonian strains, which conferred resistance to amantadine. Tis resistance was confrmed in studies described by [23,24]. Our data instead confrmed the presence of the V27I mutation, which causes strong phenotypic resistance to amantadine at a level similar to that of the S31N mutation. Tis result is similar to that of Balannik et al., who reported that the S31D mutation reduced the ability of amantadine to block the M2 channel to a comparable level as the S31N mutation [25]. Monitoring amantadine resistance among A (H3N2) viruses from 1991 to 2024 revealed that the global incidence of resistance was well above 50%, while a recent study showed that the incidence of resistance had reached 96% and 72% in China and South Korea, respectively. Tis consistent increase in the frequency of resistance to amantadine in Asian countries sharing a people-fow border with Cameroon emphasizes the importance of incorporating full genome sequences alongside antigenic data to predict which infuenza strains are likely to prevail in the upcoming infuenza seasons [26]. Given that the number of antiviral drugs available to treat and prevent infuenza virus infections is very limited, it is very important to understand the mechanisms that cause resistance to INAs and to establish a program to monitor the evolution of antiviralresistant strains.
## 5. Conclusions
Te presence of numerous substitution mutations in most of the Cameroonian strains isolated in 2024, which were absent in the 2023 strains, clearly indicates that the A/H3N2 strains circulating in Cameroon are constantly evolving. Tis fnding highlights the need for systematic genomic surveillance of seasonal infuenza viruses to assess the burden of infuenza infections and to adopt efective national infuenza control and prevention strategies.
## References
1. Who "Launches New Global Infuenza Strategy"
2. Who (2023) "Grippe Saisonnière"
3. Liang (2023) "Pathogenicity and Virulence of Infuenza" *Virulence*
4. Chauhan, Gordon (2022) "An Overview of Infuenza A Virus Genes, Protein Functions, and Replication Cycle Highlighting Important Updates" *Virus Genes*
5. Hutchinson (2018) "Infuenza Virus" *Trends in Microbiology*
6. Schweiger, Zadow, Heckler (2002) "Antigenic Drift and Variability of Infuenza Viruses" *Medical Microbiology and Immunology*
7. Hirst (1942) "Te Quantitative Determination of Infuenza Virus and Antibodies by Means of Red Cell Agglutination" *Journal of Experimental Medicine*
8. Iba, Fujii, Ohshima (2014) "Conserved Neutralizing Epitope at Globular Head of Hemagglutinin in H3N2 Infuenza Viruses" *Journal of Virology*
9. Axelsen, Yaari, Grenfell et al. (2014) "Multiannual Forecasting of Seasonal Infuenza Dynamics Reveals Climatic and Evolutionary Drivers"
10. Ann, Papenburg, Bouhy et al. (2012) "Molecular and Antigenic Evolution of Human Infuenza A/H3N2 Viruses in Quebec, Canada, 2009-2011" *Journal of Clinical Virology*
11. Holmes, Ghedin, Miller (2005) "Whole-Genome Analysis of Human Infuenza A Virus Reveals Multiple Persistent Lineages and Reassortment Among Recent H3N2 Viruses" *PLoS Biology*
12. Nunes, Pechirra, Coelho et al. (2008) "Heterogeneous Selective Pressure Acting on Infuenza B Victoria-and Yamagata-Like Hemagglutinins" *Journal of Molecular Evolution*
13. Vijaykrishna, Holmes, Joseph (2015) "Te Contrasting Phylodynamics of Human Infuenza B Viruses"
14. Nobusawa, Sato (2006) "Comparison of the Mutation Rates of Human Infuenza A and B Viruses" *Journal of Virology*
15. Sanjuán, Nebot, Chirico et al. (2010) "Viral Mutation Rates" *Journal of Virology*
16. Monamele, Vernet, Njankouo (2017) "Genetic and Antigenic Characterization of Infuenza A(H3N2) in Cameroon During the 2014-2016 Infuenza Seasons" *PLoS One*
17. Zhong, Liang, Huang (2013) "Genetic Mutations in Infuenza H3N2 Viruses From a 2012 Epidemic in Southern China" *Virology Journal*
18. Njifon, Monamele, Vernet (2019) "Genetic Diversity of Infuenza A(H3N2) Viruses in Northern Cameroon During the 2014-2016 Infuenza Seasons" *Journal of Medical Virology*
19. Phyu, Saito, Kyaw (2015) "Evolutionary Dynamics of Whole-Genome Infuenza A/H3N2 Viruses Isolated in Myanmar From" *Viruses*
20. Liu, Walker (2023) "Testing for Genetic Mutation of Seasonal Infuenza Virus" *Journal of Applied Statistics*
21. Glatman-Freedman, Drori, Beni (2017) "Genetic Divergence of Infuenza A(H3N2) Amino Acid Substitutions Mark the Beginning of the 2016-2017 Winter Season in Israel" *Journal of Clinical Virology*
22. Shih, Hsiao, Ho et al. (2007) "Simultaneous Amino Acid Substitutions at Antigenic Sites Drive Infuenza A Hemagglutinin Evolution" *Proceedings of the National Academy of Sciences of the United States of America*
23. Lee (2020) "Complete Genome Sequencing of Infuenza A Viruses Using Next-Generation Sequencing" *Methods in Molecular Biology*
24. Zaraket, Kondo, Hibino (2016) "Full Genome Characterization of Human Infuenza A/H3N2 Isolates From Asian Countries Reveals a Rare Amantadine Resistance-Conferring Mutation and Novel PB1-F2 Polymorphisms" *Frontiers in Microbiology*
25. Balannik, Carnevale, Fiorin (2010) "Functional Studies and Modeling of Pore-Lining Residue Mutants of the Infuenza a Virus M2 Ion Channel" *Biochemistry*
26. Belanov, Bychkov, Benner (2015) "Genome-Wide Analysis of Evolutionary Markers of Human Infuenza A(H1N1)pdm09 and A(H3N2) Viruses May Guide Selection of Vaccine Strain Candidates" *Genome Biol Evol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC11612878&blobtype=pdf | # Rhabdomyolysis-related acute kidney injury in COVID-19: A critical concern
Safiullah Sarker
## Abstract
Rhabdomyolysis is a severe condition characterized by the breakdown of muscle tissue leading to the release of intracellular components into the bloodstream. This condition, when associated with acute kidney injury (AKI), can result in significant morbidity and mortality, particularly in the context of coronavirus disease 2019 . This editorial discusses a retrospective study on patients with COVID-19 who developed rhabdomyolysis-related AKI. The study highlights that patients with rhabdomyolysis exhibited higher inflammatory markers, such as Creactive protein, ferritin, and procalcitonin, and experienced worse clinical outcomes compared to those with other causes of AKI. The findings underscore the importance of early recognition and management of rhabdomyolysis in COVID-19 patients to improve prognosis and reduce mortality rates.
## References
1. Murt, Altiparmak (2024) "Rhabdomyolysis-related acute kidney injury in patients with COVID-19" *World J Virol*
2. Giannoglou, Chatzizisis, Misirli (2007) "The syndrome of rhabdomyolysis: Pathophysiology and diagnosis" *Eur J Intern Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12654846&blobtype=pdf | # Epidemiology and Drug Susceptibility of Nontuberculous Mycobacteria in the Province of Pavia (Northern Italy): An Overview
Mariangela Siciliano, Francesco Amisano, Jessica Bagnarino, Giulia Grassia, Patrizia Cambieri, Fausto Baldanti, Vincenzina Monzillo, Daniela Barbarini
## Abstract
Nontuberculous mycobacteria (NTM) represent a heterogeneous group of environmental opportunistic pathogens that have emerged particularly in immunocompromised individuals and patients with underlying pulmonary disorders. NTM infections primarily affect the lungs, but can also manifest as lymphadenitis, skin and soft tissue infections, and disseminated disease. This retrospective study took into consideration 425 NTM-positive samples collected between May 2011 and December 2023, analyzed by sample type, sex, and age group (0-17, 18-49, 50-65, >65 years). Antimicrobial susceptibility analysis was performed on the 223 NTM strains with greater pathogenic power and most frequently isolated, from 2016 to 2023. Pulmonary NTM disease (NTM-PD) infections were most prevalent in patients over 65 years (52.1%), while extrapulmonary NTM disease (NTM-EPD) occurred most frequently in the 0-17 age group (56.4%). Women were slightly more affected (54.4%) than men (45.6%), with the highest incidence in female individuals over 65 years old. The most frequently isolated NTM species was the Mycobacterium avium complex (MAC) (47% of isolates). Antimicrobial susceptibility testing of 223 isolates from 2016 to 2023 revealed species-specific resistance patterns, with high susceptibility to clarithromycin in MAC (94.7%) and Mycobacterium chelonae (100%), but notable resistance in Mycobacterium abscessus complex (MABC). The increasing incidence of NTM infections underscores the need for improved diagnostic techniques and targeted treatment strategies.
## 1. Introduction
Nontuberculous mycobacteria (NTM) are a group of bacteria belonging to the genus Mycobacterium, which includes species other than Mycobacterium tuberculosis complex (MTC), the causative agent of tuberculosis [1]. Currently, NTM comprise over 200 globally ubiquitous species in both natural and anthropogenic environments. NTM are generally acquired from the environment through ingestion, inhalation, and contact [2]; person-toperson transmission of NTM is considered rare [3].
NTM pulmonary disease (NTM-PD) is the most common clinical manifestation and has become a major global public health concern due to the sharp increase in incidence and prevalence worldwide. Most pulmonary infections occur in patients with predisposing factors such as chronic obstructive pulmonary disease (COPD), bronchiectasis, cystic fibrosis, or a history of TB. These conditions make individuals particularly susceptible to colonization [4]. Lung infections are primarily caused by Mycobacterium avium complex (MAC) and Mycobacterium abscessus complex (MABC), followed by Mycobacterium xenopi, Mycobacterium fortuitum, and Mycobacterium kansasii [5]. MAC includes slow-growing mycobacteria (SGM) such as Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium chimaera, the species most frequently associated with clinical disease. MABC includes rapidly growing mycobacteria (RGM), divided into three subspecies: M. abscessus subsp. abscessus, M. abscessus subsp. bolletii, and M. abscessus subsp. massiliense. NTM also cause extrapulmonary diseases (NTM-EPD) such as cervical lymphadenitis in children, skin, soft tissue and prosthesis infections [6][7][8], bone and joint complications, and disseminated disease, especially in individuals with compromised immune systems [9].
Epidemiological studies demonstrate a heterogeneous distribution of NTM-PD. Population-based data from high-income countries suggest incidence rates ranging from approximately 4 to 20 cases per 100,000 persons per year, depending on the methodology and definitions applied [10]. In the United States, a large claims-based analysis estimated a mean annual incidence of 20.1 per 100,000 population between 2010 and 2019, with a significant upward trend over time [11]. The prevalence of NTM-PD in Europe is lower than in North America, Japan, and Korea, but is still increasing [12,13]. Specifically, it ranged from 0.9 to 7.0 per 100,000 in the UK, from 1.3 to 13.6 in France, from 3.3 to 8.4 in Spain, from 3.9 to 8.2 in Germany, from 2.3 to 5.9 in the Netherlands, and from 3.8 to 10.4 in Italy [14][15][16].
NTM constitute a substantial clinical challenge across both diagnostic and therapeutic domains. This is largely attributable to the complexity of early detection, the high prevalence of antimicrobial resistance, and the adverse effects associated with prolonged multidrug regimens. Nevertheless, a considerable proportion of prevalent cases are presumed to remain underreported, owing to persistent diagnostic limitations and the absence of mandatory surveillance or notification systems. Given the limited epidemiological information currently available in Italy, this study was designed to provide an updated overview of the epidemiological distribution of NTM during 2011-2023 in the province of Pavia. Given the heterogeneous spectrum of NTM species, antimicrobial resistance testing was performed only on a limited number of isolates, specifically those most frequently recovered and regarded as clinically significant pathogens. This focused approach was adopted to optimize statistical reliability and to generate clinically meaningful resistance data.
## 2. Materials and Methods
## 2.1. Study Setting
This retrospective study was conducted at the Microbiology Laboratory of the Fondazione IRCCS Policlinico San Matteo, Pavia, Italy, and included all clinical samples that tested positive for mycobacteria between May 2011 and December 2023. Clinical specimens were obtained from patients with suspected mycobacterial infections as part of routine diagnostic procedures. A total of 425 NTM strains, comprising both SGM and RGM species, were considered. For comparison, 456 MTC strains isolated during the same period were also analyzed. For each patient, the first isolation of each year was included in the analysis. For each isolate, anonymized data were retrieved, including sample identification code, patient age, and sex. Temporal trends in isolation were assessed by calculating the annual distribution and isolation rates of individual NTM species, and these were compared with MTC isolates. For NTM, the distribution of positive samples was analyzed by specimen type, sex, and age group (0-17, 18-49, 50-65, and >65 years). Due to the large diversity of species and the limited number of isolates for less common NTM, antimicrobial susceptibility testing (AST) was performed only on the 223 most frequently isolated and clinically significant species (MAC, M. kansasii, MABC, M. chelonae). Minimum inhibitory concentrations (MICs) were determined using the broth microdilution method according to the Clinical and Laboratory Standards Institute (CLSI) guidelines [17]. For each drug and species, MIC 50 and MIC 90 values and percentages for each category (susceptibility, intermediate, and resistance) were reported.
## 2.2. Culture and Identification of NTM
Respiratory samples, including sputum, endotracheal aspirates, bronchoalveolar lavages, and lung and pleural biopsies, were examined for the diagnosis of NTM-PD. Other materials, such as urine, biopsies, cerebrospinal fluid, cavitary fluids, and pus, were processed for the diagnosis of NTM-EPD. Samples were processed following international guidelines [18]. After Kinyoun staining, decontamination using 0.25% N-acetyl-L-cysteine and 1% NaOH (NALC-NaOH) was performed, according to the MycoTB TM (Copan, Brescia, Italy) manufacturer's instructions. To improve sensitivity, samples were cultured on both solid media Löwenstein-Jensen (Termo Fisher Scientific TM , Waltham, MA, USA) and liquid media BD BACTEC TM MGIT TM 960 (Becton Dickinson, Franklin Lakes, NJ, USA). Löwenstein-Jensen cultures were incubated at 37 • C in a 5% CO 2 atmosphere for 60 days, while MGIT TM tubes were incubated in the automated BACTEC MGIT TM 960 system at 37 • C for 56 days. Positive samples were identified using the commercial PCR reverse hybridization method GenoType CM/AS from 2014 onwards, and NTM-DR was implemented from 2019 onwards (Hain Lifescience/Arnika, Nehren, Germany).
## 2.3. Drug Susceptibility Testing
Broth microdilution is the recommended method by CLSI Standards [17]. For SGM, broth microdilution assays were performed using Sensititre TM SLOMYCO2 Susceptibility Testing Plate assay (Thermo Fisher Scientific TM ) on isolates previously grown on Middlebrook 7H11 agar (Liofilchem ® , Roseto degli Abruzzi, Italy). The SLOMYCO panel included the following drugs: amikacin, clarithromycin, ciprofloxacin, doxycycline, ethambutol, ethionamide, isoniazid, linezolid, moxifloxacin, rifabutin, rifampin, streptomycin, and trimethoprim/sulfamethoxazole. For each drug, MIC values were read after 7-14 days of incubation at 35 • C. For RGM, drug susceptibility tests were conducted by using the Sensititre TM Myco RAPMYCOI AST Plate assay (Thermo Fisher Scientific TM ). The following antibiotics were tested: trimethoprim/sulfamethoxazole, ciprofloxacin, moxifloxacin, cefoxitin, cefepime, ceftriaxone, amikacin, doxycycline, tigecycline, clarithromycin, linezolid, minocycline, amoxicillin/clavulanic acid, imipenem, and tobramycin. Sensititre panel plates were incubated at 30 • C for 3-5 days. For MABC, the clarithromycin incubation period was extended to 14 days to evaluate inducible resistance to macrolides. MICs were interpreted according to the breakpoints in the CLSI document; in particular, MAC MICs were interpreted according to CLSI. "Antimycobacterial Agents and Breakpoint for Testing MAC" [17] (breakpoint criteria are available in Appendix A.1. Table A1); M. xenopi MICs were interpreted according to CLSI. "Antimycobacterial Agents and Breakpoint for Testing Slowly Growing Nontuberculous Mycobacteria Other than MAC and M. kansasii" [17] (breakpoint criteria are available in Appendix A.1. Table A2), while MABC and M. chelonae MICs were interpreted according to CLSI. "Antimycobacterial Agents and Breakpoint for Testing Rapidly Growing Mycobacteria" [17] (breakpoint criteria are available in Appendix A.1. Table A3). Notably, MACs have different amikacin breakpoints, depending on the route of administration, whether it is parenteral or inhaled.
## 2.4. Statistical Analysis
Temporal trends in the annual number of isolates were assessed using linear regression, reporting regression coefficients (β), coefficients of determination (R 2 ), and corresponding p-values. Differences in the distribution of isolates between groups (e.g., NTM vs. MTC, pulmonary vs. extrapulmonary sources) were evaluated by comparing proportions, using the chi-square test. A two-tailed p-value < 0.05 was considered statistically significant. All statistical analyses were performed using MedCalc statistical software Version 22.017.
## 3. Results
The number of MTC and NTM isolates from pulmonary and extrapulmonary samples exhibited distinct temporal patterns across the four groups (Figure 1). A total of 425 NTM isolates were identified, of which 370/425 (87.1%) were recovered from respiratory specimens, while 55/425 (12.9%) originated from non-respiratory sources. In comparison, during the same period, 456 isolates of MTC were detected, with 354/456 (77.6%) derived from respiratory samples and 102/456 (22.4%) from extrapulmonary sites. NTM-PD isolates showed the largest fluctuations, ranging from 10 to 44 cases per year. A general increasing trend was observed over the study period (linear regression: β ≈ +1.8 isolates/year, R 2 = 0.42, p < 0.05). Peaks occurred in 2013-2014, 2016-2017, and 2021-2022, whereas troughs were recorded in 2015, 2018, and 2020. MTC-PD isolates ranged from 13 to 45 per year, without a significant linear trend over time (β ≈ +0.3 isolates/year, R 2 = 0.08, p = n.s.). NTM-EPD and MTC-EPD isolates remained consistently lower, generally ≤10 cases/year. No significant temporal variation was detected in these categories (β ≈ 0, R 2 < 0.05, p = n.s.). Analysis of these data indicates a relative increase in the incidence of NTM infections compared to MTC in recent years, particularly with respect to pulmonary manifestations (p value < 0.05). The distribution across age groups differed markedly between the MTC-PD and NTM-PD groups (Figure 2). In the youngest age group (0-17 years), 6 cases were classified as MTC-PD compared with 5 in the NTM-PD group. Among adults aged 18-49 years, the majority belonged to the MTC-PD group (n = 211), whereas the NTM-PD group had only 50 cases. Conversely, in the 50-65 and >65 age groups, NTM-PD were more prevalent (50-65 years: MTC-PD n = 63, NTM-PD n = 122; >65 years: MTC-PD n = 74, NTM-PD n = 193). A chi-square test of independence revealed a significant association between age group and disease classification (χ 2 = 170.99, df = 3, p < 0.001), indicating that MTC-PD was more frequent in younger adults, while NTM-PD predominated in older age groups. The distribution of cases across age groups differed markedly between the NTM-PD and NTM-EPD groups (Table 1). In the youngest age group (0-17 years), 1.4% were classified as NTM-PD compared with 56.4% in the NTM-EPD group. Among adults aged 18-49 years, 50 (13.5%) cases belonged to NTM-PD and 8 (14.5%) to NTM-EPD. In the 50-65 and >65 age groups, NTM-PD was predominant (50-65 years: NTM-PD n = 122 (33%), NTM-EPD n = 7 (12.7%); >65 years: NTM-PD n = 193 (52.1%), NTM-EPD n = 9 (16.4%)). A chi-square test of independence revealed a significant association between age group and disease classification (χ 2 = 190.49, df = 3, p < 0.001), indicating that NTM-EPD was more frequent in the youngest age group, while NTM-PD predominated in older age groups. Specifically, NTM-PD primarily affected individuals over 65 years (52.1%), whereas NTM-EPD was more frequently observed in the 0-17 age group (56.4%).
The distribution of cases across age groups by sex is presented in Table 2. A preliminary comparison of proportions indicates that in the younger age groups (0-17 and 18-49 years), the proportions of males and females were roughly similar, whereas in the older age groups (50-65 and >65 years), females slightly outnumbered males. These patterns suggest a trend of increasing female predominance with age, although the difference appears modest in the 50-65 group and more pronounced in the >65 group. In the youngest age group (0-17 years), there were 19 males (9.8%) and 17 females (7.4%). In the 18-49 age group, 31 cases were male (16%) and 27 were female (11.7%). Among the 50-65 years group, 59 males (30.4%) and 70 females (30.3%) were recorded. In the oldest age group (>65 years), 85 males (43.8%) and 117 females (50.6%) were observed. A chi-square test of independence indicated no significant association between age group and sex distribution (χ 2 = 3.20, df = 3, p = 0.36), suggesting that the proportion of males and females did not differ significantly across age groups. Among the 425 isolates considered in the study (Figure 3), the most frequently detected species was M. avium (n = 120, 28.2%), followed by M. intracellulare (n = 67, 15.8%), while 42 (9.9%) isolates were only reported as belonging to MAC, due to the unavailability of the GenoType NTM-DR at the time of their isolation. They were followed by M. gordonae (n = 54, 12.7%) and the M. abscessus complex that comprised 36 isolates (8.5%). Other species, including M. xenopi (n = 27, 6.3%), M. fortuitum complex (n = 16, 3.8%), M. chimaera (n = 13, 3.0%), and M. chelonae (n = 11, 2.6%), were less frequently detected. Rare species included M. kansasii (n = 8, 1.9%), Mycobacterium malmoense (n = 7, 1.6%), and Mycobacterium marinum (n = 5, 1.2%). The remaining isolates were categorized as "Other" (n = 19, 4.5%). A chi-square goodness-of-fit test indicated a highly significant deviation from a uniform distribution (χ 2 = 389.84, df = 12, p < 0.001), showing that certain species, particularly M. avium, M. intracellulare, and M. gordonae, were significantly more prevalent than others.
## Antibiotic Resistance Patterns
The isolated NTM strains exhibited species-specific antibiotic resistance patterns. AST was performed on NTM strains isolated between 2016 and 2023, and here, we report only the results of 223 strains, belonging to the most frequently represented species. SGM included in the analysis were MAC and M. xenopi, while the RGM considered were MABC and M. chelonae. Following CLSI guidelines, only antibiotics with established breakpoints were considered in this study. AST was performed on 170 MAC and 19 M. xenopi isolates (Table 3). For MAC, the MIC 50 and MIC 90 values for clarithromycin were 4 µg/mL and 8 µg/mL, respectively, with 94.7% of isolates classified as susceptible, 2.9% as intermediate, and 2.4% as resistant. Intravenous amikacin showed an MIC 50 of 16 µg/mL and MIC 90 of 64 µg/mL, with 60.6% susceptibility; in contrast, liposomal amikacin susceptibility was observed in 97.6% of strains. Moxifloxacin had limited activity against MAC, with 5.8% susceptible, 72.4% intermediate, and 21.8% resistant isolates. Linezolid showed minimal activity, with only 2.9% susceptibility and 87.7% resistance. For M. xenopi, clarithromycin exhibited strong activity with an MIC 50 of 0.12 µg/mL and MIC 90 of 0.25 µg/mL; 100% of isolates were susceptible. Amikacin showed an MIC 50 of 8 µg/mL and MIC 90 of 64 µg/mL, with 84.2% susceptibility. Moxifloxacin demonstrated good activity, with 94.7% of isolates susceptible. Linezolid was highly effective (100% susceptibility). Ciprofloxacin showed moderate activity, with 63.2% susceptibility and 36.8% intermediate. Doxycycline, trimethoprim/sulfamethoxazole, and rifampicin exhibited high resistance rates (84.2%, 84.2%, and 63.2%, respectively), whereas rifabutin remained fully active (100% susceptibility). A1 for MAC and Appendix A.2. Table A2 for M. xenopi. A hyphen (-) indicates the absence of a breakpoint. Amikacin interpretation refers to the intravenous resistance breakpoint. Clarithromycin (CLA), amikacin (AMI), moxifloxacin (MXF), linezolid (LZD), ciprofloxacin (CIP), doxycycline (DOX), trimethoprim/sulfamethoxazole (SXT), rifampin (RIF), rifabutin (RFB). For each case, the analysis included only the first isolation recorded for the year. Analysis of MABC isolates, as shown in Table 4, revealed a high susceptibility rate to intravenous amikacin (92.8%), with an MIC 50 of 4 µg/mL and MIC 90 of 16 µg/mL. Line-zolid susceptibility was 67.9%, with an MIC 50 of 8 µg/mL and MIC 90 of 32 µg/mL. Clarithromycin susceptibility was 39.3%, with an MIC 50 of 8 µg/mL and MIC 90 of 16 µg/mL. Susceptibility to moxifloxacin, ciprofloxacin, and doxycycline was lower (4%). No susceptibility was observed for trimethoprim/sulfamethoxazole, cefoxitin, imipenem, and tobramycin. For M. chelonae, a broader spectrum of susceptibility was observed. All the tested strains (100%) were susceptible to clarithromycin, intravenous amikacin, and linezolid. Susceptibility to moxifloxacin was 66.7%, while 50% of strains were susceptible to ciprofloxacin and tobramycin. Susceptibility to other drugs varied, as detailed in Table 4. Comparative analysis revealed significantly higher MIC values for MABC compared with M. chelonae across several antibiotics (p < 0.05), consistent with the observed resistance patterns (p < 0.01).
## SGM (189
## 4. Discussion
Currently, in Italy, there is limited information available on the epidemiology and drug susceptibility of NTM infections. The monitoring of demographic and microbiological data related to NTM is coordinated by the Istituto Superiore di Sanità (ISS) (Italian National Institute of Health, Rome, Italy), in collaboration with a network of 42 hospital laboratories across 16 out of 20 regions, including our institution [19]. In this context, the present study delineates an extensive examination of the long-term epidemiological patterns and the antimicrobial susceptibility profiles of NTM within a specified region of Northern Italy. This analysis spans a period of 13 years and emphasizes various demographic factors, including temporal trends, sex-based differences, and age-specific variances. NTM positivity exhibited significant interannual variability, with certain periods characterized by increased detection rates and others by a decline in cases. While multiple factors may have contributed to these year-to-year fluctuations, the notable rise in both NTM-PD and NTM-EPD cases-concurrent with a decrease in MTC-PD and MTC-EPD cases during the COVID-19 pandemic-could correlate with the implementation of public health measures such as mask usage, social distancing, and school closures, which likely contributed to reduced transmission dynamics [20]. Starting in 2022, a reversal in the previously observed trend was noted, likely attributable to the progressive relaxation or removal of public health measures implemented during the COVID-19 pandemic. Regarding age-based differences, this study observed a significant disparity in positivity rates between MTC-PD and NTM-PD. MTC-PD tends to have a higher incidence and more severe clinical manifestations in children compared to adults due to several factors [21]. Tuberculosis in children is often paucibacillary, making microbiological confirmation difficult [22]. NTM-PD increases among the elderly population, particularly those aged ≥65 years. Numerous studies have documented that pulmonary NTM infections more frequently affect elderly patients [23,24]. Aging is associated with a gradual decline in both innate and adaptive immune responses [25]. Elderly individuals often present with underlying pulmonary conditions such as COPD, bronchiectasis, or previous TB-related scarring, which create a favorable environment for NTM colonization and infection. In Italy, although national surveillance is limited, several regional and multicenter studies provide valuable epidemiological insights, particularly regarding the population aged ≥ 65 years. A multicenter retrospective analysis conducted by the network IRENE (Italian Registry of Nontuberculous Mycobacteria) across 42 hospitals found that most NTM cases were observed in individuals over the age of 60, with a predominance in females (57%) and in patients with chronic pulmonary comorbidities, particularly bronchiectasis and COPD [26]. A recent study conducted by Giannoni et al. analyzed laboratory-based data from the ISS monitoring over a five-year period. The results indicated a progressive increase in NTM isolation, with age-stratified data revealing a clear overrepresentation of patients aged ≥ 60 years, suggesting increased vulnerability in older age groups [19]. Several national and multicenter studies reported the highest incidence and mortality in individuals over 65 years [27][28][29]. While in adults, pulmonary forms predominate, the highest positivity rates of NTM-EPD are most prevalent among children and adolescents. A global meta-analysis on pediatric NTM infections shows that in children (especially ages 1-5), about 71% of cases involve lymphadenitis. This indicates different infection routes-oral in children compared with inhalation in adults-and reflects the immature immune system of young children [30]. In contrast, adults and the elderly generally showed lower positivity rates. Regarding sexbased differences, a slightly higher positivity rate was observed in postmenopausal women, especially as regards NTM-PD, especially as regards non-cavitary, nodular-bronchiectatic form. After menopause, estrogen decline may impair the host's ability to control mycobacterial infections [4,31,32]. The most common NTM that causes pulmonary disease in the province of Pavia is MAC, which is consistent with its predominance in Italy and in other parts of the world [13,[33][34][35] followed by M. gordonae, M. xenopi and M. kansasii [36]. Among RGM, MABC is the most prevalent, followed by M. fortuitum and M. chelonae three species found to be the most representative in other epidemiological studies [37]. Water, soil, and dust are known environments where MAC can live. In homes, MAC is commonly found in tap water, bathrooms, potting and garden soil, and these can be sources of infection. MAC likely spreads from natural sources into households through water distribution systems. The global spread of pulmonary MAC disease might be influenced by human activities, since people can carry MAC on themselves and their belongings, contributing to its transmission through travel and trade. Although living environments are now cleaner and more comfortable than in the past-and medical advances have increased life expectancy-these changes may unintentionally support MAC survival by reducing microbial competition through disinfection [38].
Antibiotic resistance in NTM is an increasingly important public health concern. The therapeutic approach to NTM infections is based on combined antibiotic regimens, owing to the natural drug resistance of some NTM and the potential emergence of resistance during treatment [39]. Data obtained by broth microdilution assays showed that clarithromycin was the most effective drug for MAC, M. xenopi, and M. chelonae, while it appears to be less effective for MABC. One of the most significant problems in treating MABC infections is resistance to a wide range of antibiotics, including some first-line drugs such as macrolides. These drugs are commonly used to treat various NTM infections, but in the case of MABC, inducible resistance renders them largely ineffective [40]. As regards aminoglycosides, amikacin is the most effective drug against MABC, M. xenopi, and M. chelonae, while it is less effective against MAC. According to the treatment guidelines established by the American Thoracic Society and the Infectious Diseases Society of America, patients with advanced-stage or previously treated MAC pulmonary disease are recommended to receive intravenous aminoglycosides (streptomycin or amikacin). However, the administration of an intravascular aminoglycoside is restricted by the risk of ototoxicity and renal toxicity [41]. For these reasons, the FDA approved liposome amikacin for inhalation to deliver high concentrations of the drug to the lungs in patients with MAC infection who have not achieved negative sputum culture [42]. The addition of liposomal amikacin to standard regimens has been associated with decreased hospitalization rates. The observed effects in treating NTM-PD, especially those sustained by MAC, are attributed to its targeted drug delivery at high concentration and to the unique pharmacological properties [43]. However, the FDA has not yet approved liposomal amikacin for MABC disease. Nevertheless, in some expert centers, it is used as an off-label therapy aimed at improving outcomes for these patients, who have a condition that is inherently difficult to treat [44,45].
Moxifloxacin demonstrates variable efficacy against different mycobacterial species [46]. MAC isolates showed low susceptibility, with most strains being intermediate or resistant, whereas M. xenopi and M. chelonae were largely susceptible. MABC exhibited very high resistance. These data suggest moxifloxacin is potentially useful against M. xenopi and M. chelonae but not for MAC or MABC infections. These findings are corroborated by pharmacokinetic-pharmacodynamic studies, which suggest that moxifloxacin may be useful against M. xenopi and M. chelonae infections but not for MAC or MABC infections [47]. Linezolid resistance was pronounced in MAC, with high resistance rates reported in clinical isolates (up to 80-90%) [46], whereas M. xenopi and M. chelonae isolates were fully susceptible. MABC showed moderate susceptibility. Linezolid is primarily reserved for refractory or resistant NTM infections, particularly M. abscessus. Its prolonged use is restricted due to hematologic and neurologic toxicity, often emerging after 2-4 months of therapy. Therapeutic drug monitoring and close clinical follow-up are recommended to minimize adverse effects while maintaining efficacy [48]. ciprofloxacin activity was generally poor in MAC [49] and MABC [50], while M. xenopi and M. chelonae showed intermediate susceptibility [51]. For MAC and M. chelonae, ciprofloxacin is usually not recommended as a principal agent. For MABC, ciprofloxacin may show intermediate/variable activity but is rarely sufficient alone-consider it only within combination regimens guided by MIC results. Ciprofloxacin showed moderate in vitro activity against M. xenopi, with susceptibility rates higher than those of MAC but lower than for moxifloxacin. Despite some clinical use in combination regimens, current guidelines recommend moxifloxacin over ciprofloxacin due to superior potency and clinical efficacy [52]. The role of doxycycline in NTM infections is minimal. It lacks consistent in vitro activity against both SGM and RGM and is not recommended as part of empiric or guideline-based treatment regimens. Its use should be restricted to confirmed susceptible isolates, primarily of M. chelonae, and in specific, localized infections [18]. Similarly, trimethoprim/sulfamethoxazole does not appear to be a viable option for mycobacterial infections; data were limited for MAC and M. xenopi, with high resistance observed in MABC and M. chelonae. Rifampicin and rifabutin exhibit moderate activity against MAC and M. xenopi, while most RGM, including MABC and M. chelonae, are resistant. Rifabutin is generally preferred over rifampicin in MAC due to higher potency and better intracellular activity [53].
There are several limitations in this study. First, the NTM-PD definition was based only on microbiological criteria adapted from the American Thoracic Society and the Infectious Diseases Society of America. The lack of diagnostic test results, such as chest X-ray and chest computed tomography, is a fundamental limitation for pulmonary disease confirmation. Second, the inability to accurately classify MAC species before the use of GenoType NTM-DR represents another limitation of this study. Finally, the study did not include all strains for AST because species less frequently isolated over a 13-year period may not be representative for the purposes of the survey (low sample size). The standardized MIC microdilution method was introduced in our laboratory in 2016, and we therefore could not include results obtained previously using a different method. Antibiograms were not performed for all strains because some were not considered pathogenic or clinically relevant, as suggested by the guidelines [18]. Furthermore, before 2016, it was not customary to stock strains.
## 5. Conclusions
This study provides a comprehensive overview of NTM epidemiology and antimicrobial susceptibility in a Northern Italian province over a 13-year period. The data confirm a rising trend in NTM infections, and the distribution of NTM species aligns with global patterns, with MAC as the predominant cause of pulmonary disease. AST revealed speciesspecific patterns that have important clinical implications. The variability in drug susceptibility highlights the necessity of individualized treatment regimens and the potential role of emerging therapeutic strategies, such as liposomal aminoglycosides, for improving outcomes in difficult-to-treat cases. Overall, these findings emphasize the importance of accurate species identification and susceptibility-guided therapy in NTM infections. The observed epidemiological trends underscore the need for ongoing surveillance, particularly in aging populations and high-risk groups.
## References
1. Cook, Berney, Gebhard et al. (2009) *Adv. Microb. Physiol*
2. Honda, Virdi, Chan (2018) "Global Environmental Nontuberculous Mycobacteria and Their Contemporaneous Man-Made and Natural Niches"
3. Aitken, Limaye, Pottinger et al. (2012) "Respiratory Outbreak of Mycobacterium abscessus Subspecies massiliense in a Lung Transplant and Cystic Fibrosis Center" *Am. J. Respir. Crit. Care Med*
4. Cowman, Van Ingen, Griffith et al. (2019) "Non-Tuberculous Mycobacterial Pulmonary Disease" *Eur. Respir. J*
5. Prevots, Marras (2015) "Epidemiology of Human Pulmonary Infection with Nontuberculous Mycobacteria: A Review" *Clin. Chest Med*
6. Pampaloni, Tosto, Locatelli et al. "Skin and Soft Tissue Infection by Mycobacterium intracellulare in an Immunocompetent Patient"
7. Edun, Shah, Durkin et al. (2017) "Non-Tuberculous Mycobacterial Bloodstream Infections in Patients with Indwelling Vascular Catheters-The Role of Sickle Cell Anaemia" *Infect. Dis*
8. Heid-Picard, Mougari, Pouvaret et al. (2012) *Extrapulmonary Mycobacterium abscessus Infections*
9. Ricotta, Adjemian, Blakney et al. (2009) "Extrapulmonary Nontuberculous Mycobacteria Infections in Hospitalized Patients" *Emerg. Infect. Dis*
10. Grigg, Jackson, Barter et al. (2023) "Epidemiology of Pulmonary and Extrapulmonary Nontuberculous Mycobacteria Infections at 4 US Emerging Infections Program Sites: A 6-Month Pilot" *Clin. Infect. Dis*
11. Bents, Mercaldo, Powell et al. (1094) "Nontuberculous Mycobacterial Pulmonary Disease (NTM PD) Incidence Trends in the United States"
12. Sharma, Upadhyay, Epidemiology (2020) "Diagnosis & Treatment of Non-Tuberculous Mycobacterial Diseases" *Indian J. Med. Res*
13. Prevots, Marshall, Wagner et al. (2023) "Global Epidemiology of Nontuberculous Mycobacterial Pulmonary Disease" *Clin. Chest Med*
14. Hamed, Tillotson (2023) "A Narrative Review of Nontuberculous Mycobacterial Pulmonary Disease: Microbiology, Epidemiology, Diagnosis, and Management Challenges" *Expert Rev. Respir. Med*
15. Schildkraut, Gallagher, Morimoto et al. (2020) "NTM-NET] Epidemiology of Nontuberculous Mycobacterial Pulmonary Disease in Europe and Japan by Delphi Estimation" *Respir. Med*
16. Schildkraut, Zweijpfenning, Nap et al. (2021) "The Epidemiology of Nontuberculous Mycobacterial Pulmonary Disease in the Netherlands" *ERJ Open Res*
17. Woods, Brown-Elliott, Conville et al. (2018) "Susceptibility Testing of Mycobacteria, Nocardiae, and Other Aerobic Actinomycetes"
18. Daley, Iaccarino, Lange et al. (2020) "Treatment of Nontuberculous Mycobacterial Pulmonary Disease: An Official ATS/ERS/ESCMID/IDSA Clinical Practice Guideline" *Eur. Respir*
19. Giannoni, Lanni, Iacobino et al. (2023) "Epidemiology and Drug Susceptibility of Nontuberculous Mycobacteria (NTM) in Italy in 2016-2020" *Ann. Ist. Super. Sanita*
20. Doroshenko, Lee, Macdonald et al. "Decline of Influenza and Respiratory Viruses with COVID-19 Public Health Measures: Alberta" *Canada. Mayo Clin. Proc*
21. Marais, Gie, Schaaf et al. (2004) "The Natural History of Childhood Intra-Thoracic Tuberculosis: A Critical Review of Literature from the Pre-Chemotherapy Era" *Int. J. Tuberc. Lung Dis*
22. Swaminathan, Rekha (2010) "Pediatric Tuberculosis: Global Overview and Challenges" *Clin. Infect. Dis*
23. Mcshane, Glassroth (2015) "Pulmonary Disease Due to Nontuberculous Mycobacteria: Current State and New Insights" *Chest*
24. Klein, Jedlicka, Pekosz (2010) "The Xs and Y of Immune Responses to Viral Vaccines" *Lancet Infect. Dis*
25. Meyer (2001) "The Role of Immunity in Susceptibility to Respiratory Infection in the Aging Lung" *Respir. Physiol*
26. Prato, Altieri, Carlucci et al.
27. (2018) "Non-Tuberculous Mycobacterial Pulmonary Disease: An Italian National Survey" *Sarcoidosis Vasc. Diffus. Lung Dis*
28. Pedersen, Løkke, Fløe et al. "Nationwide Increasing Incidence of Nontuberculous Mycobacterial Diseases Among Adults in Denmark" *Chest*
29. Corbett, Finger, Heiß-Neumann et al. (2023) "Development of Prevalence and Incidence of Non-Tuberculous Mycobacteria in German Laboratories from 2016 to 2020" *Emerg. Microbes Infect*
30. Kim, Kim, Jung et al. (2022) "Old Age Is Associated with Worse Treatment Outcome and Frequent Adverse Drug Reaction in Mycobacterium avium Complex Pulmonary Disease" *BMC Pulm. Med*
31. Jiang, Xue, Men et al. (2024) "Nontuberculous Mycobacterial Disease in Children: A Systematic Review and Meta-Analysis"
32. Ratnatunga, Lutzky, Kupz et al. (2020) "The Rise of Non-Tuberculosis Mycobacterial Lung Disease" *Front. Immunol*
33. Dahl, Pedersen, Van Ingen et al. (2025) "Relationship between Age, Sex, Geography and Incidence of Nontuberculous Mycobacteria in Denmark from 1991 to 2022" *ERJ Open Res*
34. Wetzstein, Diricks, Anton et al. (2024) "Clinical and Genomic Features of Mycobacterium avium Complex: A Multi-National European Study" *Genome Med*
35. Andréjak, Thomsen, Johansen et al. (2010) "Nontuberculous Pulmonary Mycobacteriosis in Denmark: Incidence and Prognostic Factors" *Am. J. Respir. Crit. Care Med*
36. Dahl, Laursen, He et al. (2023) "Species Distribution among Patients with Nontuberculous Mycobacteria Pulmonary Disease in Europe" *J. Infect*
37. Fernandez-Pittol, Batista, Narváez et al. "Microbiological Profile of Slow-Growing Non-Tuberculous Mycobacteria Species Other than Mycobacterium avium Complex"
38. Hunkins, De-Moura, Eddy et al. (2023) "In Vitro Susceptibility Patterns for Rapidly Growing Nontuberculous Mycobacteria in the United States" *Diagn. Microbiol. Infect. Dis*
39. Nishiuchi, Iwamoto, Maruyama (2017) "Infection Sources of a Common Non-Tuberculous Mycobacterial Pathogen, Mycobacterium avium Complex" *Front. Med*
40. Calcagno, Coppola, Sarmati et al. "Drugs for Treating Infections Caused by Non-Tubercular Mycobacteria: A Narrative Review from the Study Group on Mycobacteria of the Italian Society of Infectious Diseases and Tropical Medicine" *Infection*
41. Jong, Wu, Chen et al.
42. Griffith, Aksamit, Brown-Elliott et al. (2007) "An Official ATS/IDSA Statement: Diagnosis, Treatment, and Prevention of Nontuberculous Mycobacterial Diseases" *Am. Respir. Crit. Care Med*
43. Meers, Neville, Malinin et al. (2008) "Biofilm Penetration, Triggered Release and In Vivo Activity of Inhaled Liposomal Amikacin in Chronic Pseudomonas Aeruginosa Lung Infections" *J. Antimicrob. Chemother*
44. Loukeri, Papathanassiou, Kavvada et al. (1620) "Amikacin Liposomal Inhalation Suspension for Non-Tuberculous Mycobacteria Lung Infection: A Greek Observational Study"
45. Swenson, Lapinel, Ali (2020) "Clinical Management of Respiratory Adverse Events Associated with Amikacin Liposome Inhalation Suspension: Results from a Patient Survey" *Open Forum Infect. Dis*
46. Zweijpfenning, Chiron, Essink et al. (2022) "Safety and Outcomes of Amikacin Liposome Inhalation Suspension for Mycobacterium abscessus Pulmonary Disease: A NTM-NET Study" *Chest*
47. Yamaba, Ito, Suzuki et al. (2019) "Moxifloxacin Resistance and Genotyping of Mycobacterium avium and Mycobacterium intracellulare Isolates in Japan" *J. Infect. Chemother*
48. Deshpande, Srivastava, Meek et al. (2010) "Moxifloxacin Pharmacokinetics/Pharmacodynamics and Optimal Dose and Susceptibility Breakpoint Identification for Treatment of Disseminated Mycobacterium avium Infection" *Antimicrob. Agents Chemother*
49. Kurz, Zha, Herman et al. (2020) "Summary for Clinicians: 2020 Clinical Practice Guideline Summary for the Treatment of Nontuberculous Mycobacterial Pulmonary Disease" *Ann. Am. Thorac. Soc*
50. González Martínez, Aguilera, Tarriño et al. (2613) "Susceptibility Patterns in Clinical Isolates of Mycobacterium avium Complex from a Hospital in Southern Spain" *Microorganisms*
51. Park, Kim, Park et al. (2008) "In Vitro Antimicrobial Susceptibility of Mycobacterium abscessus in Korea" *J. Korean Med. Sci*
52. Fung-Rong, Kwen-Tay (1998) "Topical Ciprofloxacin for Treating Nontuberculous Mycobacterial Keratitis" *Ophthalmology*
53. Marx, Fan, Morris et al. (1995) "Laboratory and Clinical Evaluation of Mycobacterium xenopi Isolates" *Diagn. Microbiol. Infect. Dis*
54. Van Ingen, Boeree, Van Soolingen et al. (2012) "Resistance Mechanisms and Drug Susceptibility Testing of Nontuberculous Mycobacteria. Drug Resist. Updates"
55. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793610&blobtype=pdf | # P-1767. A Gut Microbiome Specific Antibiotic Spectrum Index (gASI) Predicts Clostridioides difficile Infection Disease Severity
Taryn Eubank, Abdulwhab Msdi, Kevin Garey
## Abstract
Background. Antibiotics are the top modifiable risk factor for Clostridioides difficile infection (CDI) thus is a recognized antimicrobial stewardship initiative. Antibiotic spectrum index (ASI) has been developed to better quantify antibiotic impact and stewardship practices. Recently, ASI score was associated with hospitalacquired CDI. This study aims to investigate ASI score correlation with CDI disease severity and pilot a gut microbiome specific ASI (gASI).
Table 1. Score differences between ASI and gASI of top antibiotic exposures.
Methods. We performed a case-control study of adult patients hospitalized with CDI from two health systems (14 hospitals) in Houston, TX USA (2016USA ( -2024)).
Patients were selected based on CDI disease severity and matched on variables known to impact severity including immunocompromised status and age ±10 years. Additional clinical variables of interest were extracted from the electronic health record with detailed documentation of duration and type of antibiotic exposure in the previous 30 days prior to CDI diagnosis. Disease severity and CDI classification were defined according to the 2017 IDSDA/SHEA clinical guidelines. gASI development utilized previously published ASI score with additional points for anaerobic coverage to emphasize microbiota impact and antibiotic present in bile as a surrogate for microbiota exposure.
Results. A total of 100 patients (50 severe vs 50 nonsevere) with CDI were included (Female: 53%, white, non-Hispanic: 54%; Age > 65 years: 66%; hospital-acquired CDI: 50%; CDI initial episode: 89%). Higher ASI score was significantly associated with severe CDI (14.5±10.9 vs 8.3±8; p=0.002) (Figure 1). Table 1 demonstrates the score differences between ASI and gASI. gASI remained significantly associated with severe CDI (17.6±13.3 vs 10.4±10; p=0.003). Patients with severe CDI had significantly more exposure to intravenous vancomycin, cefepime, piperacillin/tazobactam, and meropenem.
Conclusion. ASI and newly developed gASI both correlate with CDI disease severity. Future work to validate and strengthen gASI score is warranted through metagenomics and/or metabolomics to capture the true impact of antibiotics on the gut microbiome.
Disclosures. Kevin W. Garey, MS;PharmD, Acurx: Grant/Research Support| Merck: Grant/Research Support|Paratek: Grant/Research Support |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12791329&blobtype=pdf | # P-283. Evaluating Missed Opportunities for HIV Diagnosis in a Statewide Health System
Timothy Adkins, Nicole Bryan, Jesse Thompson
## 1
West Virginia University, Danville, WV Session: 48. HIV: Epidemiology and Screening Monday, October 20, 2025: 12:15 PM Background. HIV is an ongoing public health burden which can be controlled with interventions such as ART. In recent years, outbreaks of HIV infection have occurred related to the opioid epidemic. Inconsistencies in screening high risk patients in rural communities at health care encounters may lead to delays in care, more opportunistic infections, and further HIV transmission. Addressing screening disparities in rural communities can improve patient outcomes and reduce future outbreaks. WVU Medicine is the largest health system in WV and includes 25 hospitals located throughout the state. This study retrospectively examines 101 recently diagnosed HIV patients who received care at WVU Medicine for patterns in missed opportunities for early HIV diagnosis. Results. 21 of the 101 patients in the study had two or more encounters with the health care system prior to diagnosis. No identified risk factor-MSM, IDU, or opportunistic infection-had a statistically significant impact on whether there were two or more healthcare contacts prior to diagnosis. The most common setting for these was the Emergency Department.
Conclusion. Missed testing opportunities in the Emergency Department are a consequence of volume and triage of priorities. No single risk factor had a statistically significant effect on the number of healthcare encounters prior to diagnosis, though due to size the statistical power of the study is small. Improving identification and screening procedures for patients at community and rural access hospitals, especially in Emergency Departments, may reduce the number of missed new HIV diagnoses.
Disclosures. All Authors: No reported disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12746540&blobtype=pdf | # Arrival of Oropouche Virus in a Nonendemic Area in Northeastern Brazil, 2024
Jean Nascimento, | Thiago, P Araújo, Mykaella Araújo, | Mateus, M Arruda, Vitória Simplicio, Emelly Calheiros, Aline Pereira E Silva, Laura Silva, Marcus Santos, | Magliones, C Lima, Hazerral Santos, | Ênio, J Bassi, Alessandra Borges, Anderson Leite, Abelardo Silva-Júnior, Thiago Araújo
## Abstract
Orthobunyavirus oropoucheense (OROV) causes Oropouche fever, which exhibits symptoms similar to those of other arboviral diseases. Although it has historically been restricted to the Amazon region, the virus has recently spread to other areas of Brazil. Alagoas state, with low socioeconomic conditions and limited health coverage, has seen an increase in febrile cases without confirmed molecular diagnoses of circulating arboviruses. By September 6, 2024, 1316 samples negative for Dengue, Zika, and Chikungunya were tested for OROV and Mayaro virus using RT-qPCR, yielding 115 (8.74%) positive results for OROV. Among these, 14 (22.22%) viral isolates were obtained in Vero cells and confirmed by RT-qPCR and immunofluorescence assay (IFA). The study generated 37 new near-complete genomic sequences corresponding to the newly characterized OROV lineage and examined selection pressures on the M gene, identifying sites under purifying selection. We identified amino acid variations in the Gc glycoprotein structure at positions 507, 552, 738, and 795, which may influence host-cell interactions. This work is the first to report OROV in Alagoas, emphasizing the need for improved monitoring and control measures to mitigate public health impacts.
| IntroductionOrthobunyavirus oropoucheense (OROV) belongs to the Orthobunyavirus genus. It was first reported in 1955 in Trinidad and Tobago and has since caused outbreaks, primarily in Peru, Panama, and Brazil [1]. In Brazil, OROV was first isolated in 1960 from sloths and Aedes serratus near the construction site of the Belém-Brasília highway [2]. The Amazon region of northern Brazil remains endemic for OROV, with 261 reported cases between 2015 and 2022 [3]. However, in 2023, the Brazilian Ministry of Health reported a significant increase in OROV cases across various regions, totaling 7497 confirmed cases by early September 2024 [4].OROV is an enveloped virus in the Peribunyaviridae family, antigenically related to the Simbu serogroup. It possesses aThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
segmented, negative-sense single-stranded RNA (ssRNA) genome consisting of three segments: small (S), medium (M), and large (L) [5,6]. Among these segments, the M segment encodes the structural glycoproteins Gn and Gc, which are recognized as significant antigenic determinants and have been implicated in mediating host-cell entry. Additionally, it encodes the non-structural protein NSm, which plays a crucial role in viral assembly and budding [7]. Successive reassortment events in South America may be related to the rise of the OROV clade (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021)(2022)(2023), which is spreading in Brazil, characterized by the M segment from the OROVbr clade (2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) and the L and S segments from the OROV Pe/CO/EC clade [8].
In segmented viruses, reassortment is crucial for generating new strains, driving viral evolution, and facilitating adaptation to new hosts. Additionally, these events enable OROV to exploit new ecological niches and adapt to novel vectors [9], facilitating its dissemination to non-endemic regions. Recent studies have shown that the BR-2015-2024 lineage exhibits increased replication in vitro and in animal models, along with partial escape from neutralizing antibodies, suggesting that reassortment events may enhance viral fitness and transmissibility [10,11]. In urban and peri-urban regions, OROV transmission is primarily associated with the biting midge Culicoides paraensis [12]. The recent outbreaks outside the Amazon may be driven by human factors such as rapid urbanization, inadequate sanitation, and stagnant water accumulation [13]. Evidence suggests that changes in land use, deforestation, and agricultural expansion increase interactions between humans and vectors, which could explain the expansion of OROV beyond its historically endemic areas [14].
OROV causes Oropouche fever, an arboviral disease characterized by high fever, headache, arthralgia, myalgia, skin rashes, malaise, nausea, and vomiting. Rare complications can include meningitis and encephalitis [1]. Recently, there has been increasing evidence of an association between congenital OROV infections and fetal loss, stillbirths, and other complications [15]. During the acute phase of the disease, viral genetic material can be detected in patient samples using molecular techniques such as quantitative reverse transcription polymerase chain reaction (RT-qPCR) or viral isolation in cell culture [1].
Recent geographic expansion changes have increased the dissemination of OROV beyond the Amazon region [16]. Alagoas, including its coastal geography and adjacent areas, is particularly vulnerable to unexpected outbreaks due to low socioeconomic indicators and limited regional health coverage, despite not being traditionally affected. Considering Brazil's recent Oropouche virus (OROV) outbreak [4] and its potential to expand to new regions, a comprehensive investigation is essential to assess its presence in additional areas. Furthermore, like other arboviruses, Oropouche fever is a neglected disease, highlighting the need for increased attention to its public health implications in Alagoas and across Brazil. Given the increasing number of febrile cases and the lack of confirmed molecular diagnoses for arboviruses, this study aimed to assess the circulation of OROV in Alagoas during 2024. This research is significant because it addresses a critical gap in the surveillance of neglected arboviruses, providing valuable insights to inform public health strategies and control measures.
## 2 | Methods
## 2.1 | Sample Selection and Molecular Diagnosis
Serum samples from reverse transcriptase quantitative polymerase chain reaction (RT-qPCR)-negative patients for Zika, Dengue, and Chikungunya were tested for Oropouche (OROV) and Mayaro (MAYV) viruses. Viral RNA was extracted using the Extracta® Kit -DNA and RNA of pathogens -MPTA MDx (Loccus) on the Extracta® 96 DNA and RNA extractor and purifier (LOCCUS), following the manufacturer's instructions. The extracted RNA was subjected to RT-qPCR.
For the molecular detection of OROV and MAYV viruses, the IBMP Mix Fit I -Mastermix OneStep kit (Instituto de Biologia Molecular do Paraná -IBMP) was used, along with previously designed oligonucleotides [17]. All runs were performed on Applied Biosystems™ 7500 real-time PCR System, with detection based on cycle threshold (Ct) values < 40.
Our study analyzed samples from patients who visited public health units in the state of Alagoas with symptoms suggestive of arbovirus infection. All suspected cases of acute arboviral infections in Alagoas are referred to the state's central laboratory, LACEN-AL (Alagoas, Brazil). All study procedures complied with the ethical standards of the Human Research Ethics Committee of the Federal University of Alagoas (protocol number 65701122.8.0000.5013).
## 2.2 | Viral Culture Cell Isolation and Indirect Immunofluorescence Assay (Ifa)
Serum samples from selected patients were included in the study, in which RT-qPCR detected OROV, and the cycle threshold was < 20. The samples were inoculated onto monolayers of VERO E6 cells (CRL-1586) and cultured in DMEM F12 medium (Gibco, USA) containing 10% heat-inactivated fetal bovine serum (FBS), 1% L-glutamine (Gluta-MAX, Thermo-Fisher Scientific), and 1% penicillin-streptomycin-amphotericin B (PSA) solution. The cells were incubated at 37°C in a CO2 incubator. These cells were inoculated with serum samples. The cell cultures were monitored daily for the development of virusinduced cytopathic effects (CPE). The aliquots were used to confirm viral isolation by RT-qPCR and IFA.
The cells from culture pellets were fixed with a 1:1 methanolacetone solution for 10 min. After washing with PBS, they were permeabilized with 0.1% Triton X-100. Samples were incubated with PBS and BSA, then treated with mouse ascitic fluid hyperimmune to anti-OROV (1:500). The slides were washed three times with PBS and then incubated with Alexa Fluorconjugated anti-mouse IgG secondary antibody (1:2000). After additional washes and staining with DAPI, the slides were mounted and examined under a fluorescence microscope.
## 2.3 | Whole Genome Sequencing and Generation of Consensus Sequences
Whole genome sequencing was performed on 37 OROVpositive samples from serum patients. Only samples with CT values of 27 or less were selected. To ensure the integrity, specificity, and reliability of the NGS results, a negative control sample (no-template control) was included in the PCR amplification and sequencing workflow.
RNA was converted to cDNA using Luna Script RT SuperMix (5x; New England Biolabs [NEB], Ipswich, MA, USA). The generated cDNA underwent multiplex PCR sequencing using Q5 High Fidelity Hot-Start DNA Polymerase (NEB) and a primer set designed for sequencing the three OROV segments, following the method described by Naveca and coworkers [8]. Amplicons were purified using AMPure XP beads (Beckman Coulter, Brea, CA, USA), and concentrations were determined using the Qubit dsDNA HS Assay Kit on a Qubit 4 Fluorometer (Thermo Fisher Scientific Corporation, Waltham, MA, USA). Library preparation was performed using the Ligation Sequencing Kit (SQK-LSK109) and Native Barcoding Expansion Kit EXP-NBD196 (Oxford Nanopore, Oxford, UK). The library was loaded onto an R10.4 flow cell and sequenced using a MinION Mk1B device. ONT MinKNOW software was used for collecting raw data. Raw files were basecalled and demultiplexed using Guppy v.6.0.1 (Oxford Nanopore Technologies). Consensus sequences were obtained through hybrid assembly using the Genome Detective online tool (https://www. genomedetective.com/). In this context, hybrid assembly refers to a computational framework that combines de novo and reference-guided assembly methods to enhance genome completeness and accuracy. This differs from the conventional hybrid sequencing and assembly approach, which typically integrates short-and long-read data from the same sample.
## 2.4 | Phylogeny and Time-Scaled Phylogenetic Tree Analysis
The S, M, and L genomic segments of OROV generated in this study were combined with corresponding segments from all published full-length OROV sequences available in GISAID up to August 2024. Sequences with identical collection locations and dates were filtered to construct a final dataset, which included our sequences and 305 representative sequences per segment. Sequence alignment was performed using MAFFT v7.526 [18] and manually curated in AliView v1.28 [19] to remove artifacts. Maximum likelihood phylogenetic trees for each segment were constructed using IQ-TREE v2.3.5 [20]. The TIM3 + F + I + R3 nucleotide substitution model, selected by ModelFinder [21], was used for the L and M segments, while the TPM3 + R2 model was applied to the S segment. Branch support was assessed using 1000 ultrafast bootstrap replicates and an SH-aLRT test (1000 replicates). Trees were visualized using the ggtree R package.
To estimate the temporal and spatial dynamics of OROV in Alagoas, a time-scaled phylogenetic tree was constructed using the Nextstrain Augur pipeline [22]. Sequences were filtered by metadata (state, municipality, and collection date), and aligned genomes from 2022 to 2025 were used to infer a maximum likelihood tree with IQ-TREE. A time-scaled phylogeny was generated with TreeTime using a strict molecular clock and marginal date inference. Geographic ancestral states were inferred with Augur traits at two resolutions. The dataset was then exported to Auspice, enabling interactive exploration of clades and dispersal routes, as well as the identification of the most recent common ancestor of the Alagoas sequences. This allowed us to estimate when the virus was introduced to the state.
Using the Recombination Detection Program (RDP5) [23], we analyzed the multiple sequence alignments of the three concatenated segments (S, M, and L). This allowed us to identify potential homologous recombination events involving template switching within the same genomic segment. Additionally, we investigated reassortment events, defined as the exchange of complete genomic segments among related viruses, separately. We performed segment-specific phylogenetic reconstructions and compared the resulting tree topologies.
## 2.5 | Selection Pressures
The investigation of selection pressures was performed in a first set of M gene sequences obtained in Brazil's Northeast region and a second set of sequences from all Brazilian regions: Midwest (n = 10), North (n = 173), Northeast (n = 111), South (n = 10), and Southeast (n = 105). The first set included 111 sequences (> 4200 nt) and the second set included 409 sequences (> 4200 nt). All sequences were obtained from human cases in 2024.
The average number of nonsynonymous substitutions (dN) and synonymous substitutions (dS) per site (dN/dS ratio) was estimated by the single likelihood ancestor counting (SLAC) [24]. SLAC estimated sites under diversifying selection (positive selection), mixed effects model of evolution (MEME) [25], and fast unconstrained Bayesian approximation (FUBAR) [26]. Sites under purifying selection (negative selection) were estimated with FUBAR and SLAC. These methods are implemented in the HYPHY platform [27] and were accessed through the Data-Monkey 2.0 web server (http://www.datamonkey.org) [28]. For the MEME and SLAC methods, the confidence level was set to a p-value of 0•05, and the FUBAR method was set to a posterior probability of 0•95.
## 2.6 | Evaluation of Glycoprotein Mutations in OROV Isolates
The cDNA sequences of M polyprotein were translated into protein sequences of the viral glycoproteins Gn (1-312 aa), Nsm (313-480 aa), and Gc (481-1420 aa) using MEGAX software version 10.0.5 [29]. The alignment was performed using the MUSCLE algorithm [30]. The resolution of OROV proteins is still unavailable, except for the Gc N-terminal region (PDB ID 6H3X), which is involved in viral fusion to the host cell. Herein, we obtained the homotrimer three-dimensional structure of the Gc protein based on the sequence OROV-AL29_2024-07-12 using Colabfold software version 1.5.2 [31] on the COSMIC2 platform (https://cosmic2.sdsc.edu:8443/gateway/login!input.action) [32]. Software basal parameters were maintained and added to the "Use templates from published PDB structures" option.
## 2.7 | Epidemiological Data and Statistical Analysis
For the descriptive analysis, information on the age, sex, and municipality of residence of patients testing positive was obtained from the Laboratory Environment Manager (GAL) database, which is dedicated to health surveillance. Age categories followed the model of the Brazilian Institute of Geography and Statistics (IBGE) (https://ibge.gov.br/).
To account for differences in population size between municipalities, incidence rates were calculated and expressed as the number of confirmed OROV cases per 100,000 inhabitants. Population estimates for 2024 were obtained from official IBGE projections and used as the denominator in the calculations.
The data were tabulated and analysed using R version 4.4.1 and RStudio 2024.04.2 + 764. The tidyverse, ggplot2, geobr, ggExtra, and cowplot packages were utilized for data manipulation, statistical analysis, and visualization.
## 3 | Results
To investigate the presence of OROV in Alagoas state, we analyzed samples that tested negative for Zika, Dengue, and Chikungunya viruses using molecular diagnostics. Among the 1,316 samples examined from April to September 2024, 115 (8.74%) tested positive for OROV (Table S1). The remaining 1,201 samples (91.26%) were negative, including all those tested for MAYV. Figure 1A illustrates the age distribution of Alagoas state's population and confirmed OROV cases. The population was predominantly younger individuals aged 44 years or less. OROV incidence analysis revealed a concentrated distribution among young adults. However, there is no statistically significant association between OROV detection and age groups (Figure 1A). Regarding the gender distribution, 61 (53.04%) of the OROV-positive patients were male, and 54 (46.96%) were female. Pearson's Chi-squared test revealed no statistically significant association between OROV positivity and patient gender (p-value > 0.05; Figure 1B, Table S1). The gender-specific findings that stand out include the identification of 37 young women between the ages of 15 and 35 who were of reproductive age, 32.17% (37/115) of whom tested positive for OROV.
The first case of OROV in Alagoas was identified in a male patient living in Japaratinga, a municipality on the northern coast of the state. Since then, the virus has been detected in a further 13 municipalities (Figure 2). After adjusting for population size, Palmeira dos Índios had the highest incidence rate, at around 126.4 cases per 100,000 people (93 cases among an estimated population of 73,596 in 2024). Tanque D'Arca followed with an incidence rate of 67.7 per 100,000 inhabitants (four cases), and Japaratinga with an incidence rate of 21.2 per 100,000 inhabitants (two cases). Other municipalities with lower incidence rates included Estrela de Alagoas (12.7 per 100,000) and Viçosa (12.3 per 100,000). Municipalities such as Atalaia, Maceió, Porto Calvo, Messias, Arapiraca, Coruripe, Igaci, Santana do Ipanema, and União dos Palmares each reported a single case, corresponding to incidence values of less than 10 per 100,000 inhabitants. These findings suggest a focal pattern of transmission, with disproportionate clustering in mid-sized municipalities.
OROV cases in Alagoas were detected in samples collected from May to August, with the first detection on May 22 and the last on August 15 (Table S2). Positive cases were reported during epidemiological weeks 21 to 33 (Figure 3), except for week 26, which had no recorded cases. Of the 115 samples positive for the viral genome by RT-qPCR, 14 (22.22%) underwent viral isolation attempts in Vero E6 cell culture. Culture supernatants were collected 3 days post-infection (dpi) due to the appearance of viral cytopathic effects (CPE) in the cell monolayers, characterized by cell detachment, cytoplasmic rounding, syncytium formation, cellular debris, and plasma membrane blebbing (Figure 4). The isolation of all samples was confirmed by RT-qPCR, with positive results for OROV, indicating isolation in the first cell culture passage (Table S3). An indirect immunofluorescence assay (IFA) was also performed on all samples subjected to isolation attempts, confirming infection of the cell monolayers.
We generated 37 new genomic sequences for each OROV segment. These sequences were obtained from cases reported during epidemiological weeks 27 to 33 (Figure 3). The average genome coverage was 89.64% for segment L, 96.20% for segment M, and 95.50% for segment S. The selected samples had an average cycle threshold (Ct) value of 17.67, with a minimum of 13.41 and a maximum of 22.57. All sequenced samples produced high-quality data (Table S4), confirming that our selection criterion (Ct ≤ 27) was appropriate for OROV genome sequencing. The sequencing data obtained from the negative control sample showed no detectable signs of contamination, thereby confirming the integrity and reliability of the experimental procedures.
Phylogenetic reconstruction of the L, M, and S segments from the newly generated genomes revealed a high degree of conservation with most Brazilian isolates, including sequences from the recent OROV outbreaks in Brazil (2023-2024, green; Figure 5). A wellsupported cluster was observed, grouping these sequences with previous Brazilian lineages (Brazil, blue; Figure 5), supporting the evolution of OROV from the western Amazon region of Brazil. The isolates from Alagoas (red) clustered closely with contemporary Brazilian strains, suggesting a recent common origin and indicating that these isolates likely resulted from the geographic spread of OROV within the country. The novel sequences generated in this study are available in the NCBI GenBank database under the following accession numbers: L- segment PV335275-PV335311, M-segment PV254768-PV254804, and S-segment PV254731-PV254767.
Discrete phylogeographic reconstruction revealed that OROV sequences from this study belonged to a Northeast Brazilian clade (Figure 6). All Alagoas genomes clustered with sequences from neighboring northeastern states, particularly Pernambuco (Figure 6a). Time-scaled phylogeny estimated the most recent common ancestor of the Alagoas cluster emerged on November 30, 2023 (90% CI: March 28, 2023-February 8, 2024), suggesting recent introduction. Spatial diffusion analysis supported a northeastern dispersal pattern, detecting two independent OROV introductions into the Northeast: one from Santa Catarina to Bahia and another from Rio de Janeiro to Pernambuco (Figure 6b). From Pernambuco, viral movement spread into Alagoas as well as to Sergipe and Paraíba (Figure 6c). These results reveal the Alagoas outbreak is part of a regional transmission network originating from neighboring northeastern states. Unfortunately, because the majority of our sequences belonged to the municipality of Palmeira dos Índios, we were unable to verify broader OROV distribution across Alagoas. This occurred because sequences from other municipalities did not yield quality data, or positive samples with Ct values greater than 27 were not selected for sequencing.
We assessed selection pressures acting on the M gene to investigate viral evolutionary dynamics. The results are shown in Table 1. In M gene sequences from Brazil's Northeastern region obtained in 2024, FUBAR and SLAC analyses indicated 60 and 38 sites under purifying selection, respectively. FUBAR supported thirty-eight sites, which SLAC supported. In sequences from all Brazilian regions, FUBAR and SLAC analysis indicated 104 and 66 sites under purifying selection, respectively. FUBAR supports all sites supported by SLAC. The mean ratio of nonsynonymous substitutions (dN) and synonymous substitutions (dS) rates (dN/dS) was used to identify signatures of negative (dN/dS < 1) or positive (dN/ dS > 1) selection. The identified distribution rates were 0.136 and 0.173 for the sequences from Brazil's Northeast and all Brazilian regions, respectively. Sequences from the Brazilian Northeast had four and eight sites under diversifying selection, as indicated by MEME and FUBAR, respectively, and SLAC identified a single site under diversifying selection in this case. Two sites, 176 (Gn glycoprotein) and 552 (Gc glycoprotein), were identified as supported by MEME and FUBAR. For sequences from all Brazilian regions, MEME, FUBAR, and SLAC analyses indicated six, eight, and five sites under diversifying selection, respectively. The three tools support four sites: 176, 552, 981, and 982.
The 37 analyzed sequences contained amino acid mutations in 10 glycoprotein regions. Gn, Nsm, and Gc had 3, 1, and 6 mutations, respectively. Based on the tridimensional Gc glycoprotein structure, mutations at amino acid positions 507, 552, 738, and 795 were in a region that may play a role in host-cell interactions (Figure 7).
## 4 | Discussion
Information on arboviruses is vital for public health monitoring. The symptoms of Oropouche fever are similar to those of Dengue and Chikungunya, which makes syndromic detection difficult [33]. In response to the increase in OROV cases across Brazil, including regions beyond the Amazon, the Coordination of Public Health Laboratories of the Ministry of Health (CGLAB/MS) implemented a strategy in 2024 for distributing a molecular diagnostic protocol using RT-qPCR to Brazil's central public health laboratories, aiming to assess the extent of viral dissemination nationally. In Alagoas specifically, all serum samples that are negative for Zika, Dengue, and Chikungunya must be tested for OROV and MAYV. This change represents a critical shift from targeted surveillance, which will likely lead to an increase in reported cases of these emerging viruses and provide a more accurate picture of their circulation beyond the Amazon region.
This study reports OROV detection in Alagoas, with 115 cases recorded. Brazil confirmed 7931 OROV cases, including two deaths, by Epidemiological Week 35 of 2024. The first confirmed case of Oropouche fever in Alagoas was reported on May 22, 2024, in a male patient from Japaratinga, a town on the state's northern coast. This area is characterized by intense population mobility driven by tourism [34] and its proximity to Pernambuco, which had already reported 92 confirmed cases by the time Alagoas's first case was detected. After the initial case, new OROV cases were reported in other municipalities, particularly in Palmeiras dos Índios. The high number of cases may be related to the traditional and historical habitation of indigenous communities, whose sociocultural dynamics favor viral circulations [35], as well as to environmental factors, since the region is a significant area of banana cultivation, which has been positively correlated with OROV cases [16]. In addition, the presence of Culicoides insignis in Alagoas, a potential vector of the virus [36], highlights the role of local ecological conditions in transmission.
Our study also found cases in young women of reproductive age (15 to 35), with 32,17% (37/115) being OROV-positive. This finding is concerning in public health, as studies have discussed the potential vertical transmission of OROV, which may significantly impact fetal development [37]. Increased surveillance and attention to pregnant women are essential. Reports of fetal deaths and vertical transmission of OROV have been confirmed in various Brazilian regions [38], although not yet in Alagoas.
Understanding arbovirus circulation and the factors predisposing to dissemination risk is essential for monitoring and mapping affected areas. Early diagnosis supports data collection, facilitating epidemiological and clinical studies. In Alagoas, OROV was first identified in May 2024, despite neighboring states, Pernambuco and Bahia, having diagnosed the virus months earlier [39]. This diagnostic delay may reflect either a late introduction of the virus in Alagoas or limited surveillance sensitivity, highlighting the need for enhanced, timely arbovirus monitoring strategies across different states. Early diagnosis also enables effective surveillance and timely healthcare, improving the prognosis and treatment of arbovirusinfected patients [40], a critical goal in Brazil. Alagoas is also known for the circulation of other arboviruses, such as Dengue, Zika, and Chikungunya, which, like OROV, are considered dengue-like viruses due to their similar symptoms: arthralgic and myalgic fever, followed by neurological and dermatological complications. This symptom overlap complicates differential diagnosis, underscoring the importance of biomolecular detection. Phylogenetic analysis of the three OROV genome segments (L, M, and S) obtained in this study, along with reference sequences, revealed a monophyletic clade comprising sequences from recent OROV outbreaks in Brazil. This finding confirms the current circulation of a viral lineage previously reported in other Brazilian regions [39]. The high sequence similarity between Alagoas strains and those from recent national outbreaks suggests limited genetic divergence, consistent with recent viral expansion. Although genetic changes are common in segmented viruses, our sequences showed no evidence of homologous recombination or reassortment. Our sequences originate from a recently circulating reassortant lineage in Brazil, which contains M segments from viruses detected in the eastern Amazon region and L and S segments from viruses found in Peru, Colombia, and Ecuador [41]. These results support OROV's geographic dispersion across Brazil, particularly in areas with no history of endemicity or infection risk.
Recent genomic analyses have demonstrated the introduction of the BR-2015-2024 lineage into Colombia, where it co-circulates with a previously established OROV lineage (OROV PE/CO/EC-2008-2021), indicating that this variant is spreading beyond Brazil throughout the Amazon Basin [42]. The simultaneous circulation of distinct lineages increases the likelihood of reassortment and may generate viruses with altered transmissibility or pathogenicity. This scenario underscores the importance of continuous genomic surveillance and cross-border monitoring to detect new introductions, characterize emerging variants, and prevent the wider dissemination of OROV across the Americas.
Residues 176 (Gn) and 552 (Gc), which showed the most homogeneous frequencies of amino acid mutations among the sequences obtained in this study [Figure 5], were also identified as under diversifying selection pressures based on both Brazilian and Northeast Brazilian sequences. Other mutations identified in this study were not under selection pressure. Additionally, under diversifying selection, residue 61 (Gn) has been identified as an emerging site of the non-synonymous substitution V61F, which may have emerged between December 2023 and April 2024 in Brazilian OROV sequences [43]. However, all 37 sequences obtained from May to August 2024 presented V61. In line with our findings, sites 66 and 86 (Gn) were also identified as undergoing diversifying and purifying selection, respectively, in another study that analyzed OROV samples from Trinidad and Tobago, Panama, Peru, and Brazil from the 1950s to the late 2000s [41].
The Gn and Gc glycoproteins play an important role in OROV architecture. Studies on crystallography and comparison of protein sequences among species have demonstrated that these proteins may form trimeric structures [44]. Thus, these glycoproteins may represent crucial targets for antigenic recognition and cell interactions, highlighting the importance of investigating variations in their amino acid residues. Our data demonstrate two and five amino acid modifications in Gn and Gc, respectively, which alter their chemical properties. In Gn, at position 40, leucine (an apolar aliphatic residue) was replaced by tryptophan. In position 119, cysteine (a polar residue) was also substituted by tryptophan, introducing aromatic properties in both cases. In Gc, modifications were observed in 507 residues from glycine to aspartic acid (negatively charged); at position 552, serine was replaced by alanine (apolar, aliphatic); at position 738, methionine was replaced by isoleucine; at position 795, threonine was replaced by alanine; and at position 1197, glutamic acid (negatively charged) was replaced by lysine (positively charged). Given this, modifications in the residues of glycoproteins may represent significant changes in the immune response and virus-cell interactions.
This study has both inherent limitations and noteworthy strengths. One of its main advantages is being the first report of Oropouche virus circulation in Alagoas, Brazil, supported by extensive molecular, virological, and genomic evidence, including the generation of 37 nearly complete genome sequences, which enhances the originality and robustness of the findings. Additionally, the epidemiological analysis provides valuable insights, particularly regarding the identification of reproductive-age women as a potentially vulnerable group, which has important public health implications. However, certain limitations must be acknowledged.
The study is primarily descriptive and lacks clinical outcome data. Moreover, the absence of serological investigations precludes the evaluation of prior or asymptomatic infections. Ultimately, despite the study highlighting the significance of enhanced surveillance, incorporating more detailed recommendations for public health officials would enhance its practical applicability.
## 5 | Conclusion
We confirmed the first case of OROV in Alagoas through molecular detection in samples that tested negative for other arboviruses. This finding corroborates recent findings from various studies, which have shown a shift in the geographic distribution of OROV cases, previously concentrated in northern Brazil, indicating an expansion of its transmission area. This highlights the need to intensify epidemiological surveillance in other regions with similar symptoms but limited testing for this arbovirus.
Herein, we recommend implementing genomic surveillance strategies nationwide to enhance monitoring and understanding of viral circulation, providing essential data for public health decision-making. This approach will enable rapid detection of potential outbreaks, improve understanding of transmission dynamics, and support the development of targeted interventions to control the virus's spread. Our findings underscore the need for coordinated actions among researchers, health authorities, and the community to address the emerging challenges posed by OROV in Brazil.
## References
1. Travassos Da Rosa, De, Souza et al. (2017) "Oropouche Virus: Clinical, Epidemiological, and Molecular Aspects of a Neglected Orthobunyavirus"
2. Pinheiro, Pinheiro, Bensabath et al. (1962) "Epidemia De Vírus Oropouche Em Belém. Revista Do" *Serviço Especial de Saúde Pública*
3. Martins-Filho, Carvalho, Santos (2015) "Spatiotemporal Epidemiology of Oropouche Fever"
4. Brasil, Da Saúde (2024) "Secretaria de Vigilância em Saúde e Ambiente. Oropouche"
5. Elliott (2014) "Orthobunyaviruses: Recent Genetic and Structural Insights" *Nature Reviews Microbiology*
6. Hughes, Adkins, Alkhovskiy (2020) "ICTV Virus Taxonomy Profile: Peribunyaviridae" *Journal of General Virology*
7. Gunter, Omoga, Bowen (2024) "A Reporter Oropouche Virus Expressing Zsgreen From the M Segment Enables Pathogenesis Studies in Mice" *Journal of Virology*
8. Naveca, Almeida, Souza (2024) "Human Outbreaks of a Novel Reassortant Oropouche Virus in the Brazilian Amazon Region" *Nature Medicine*
9. Latourrette, Garcia-Ruiz (2022) "Determinants of Virus Variation, Evolution, and Host Adaptation" *Pathogens*
10. Scroggs, Gutierrez, Reister-Hendricks et al. (2025) "Enhanced Infection and Transmission of the 2022-2024 Oropouche Virus Strain in the North American Biting Midge Culicoides Sonorensis" *Scientific Reports*
11. Merchant, De Paula, Souza et al. (2025) "Protein-Based Tools for the Detection and Characterisation of Oropouche Virus Infection" *EMBO Molecular Medicine*
12. Zhang, Liu, Wu (2024) "Oropouche Virus: A Neglected Global Arboviral Threat" *Virus Research*
13. Gómez-Marín, Usme, Parra-Henao et al. (2025) "Oropouche Virus Entomovirology and Surveillance: Where Are We and Where are We Headed?" *Acta Tropica*
14. Lorenz, Azevedo, Sallum et al. (2025) "Oropouche Fever Outbreak in Brazil: Key Factors Behind the Largest Epidemic in History" *PLoS One*
15. Martins-Filho, Carvalho, Santos (2024) "Oropouche Fever: Reports of Vertical Transmission and Deaths in Brazil" *Lancet Infectious Diseases*
16. Gräf, Delatorre, Do Nascimento et al. (2025) "Expansion of Oropouche Virus in Non-Endemic Brazilian Regions: Analysis of Genomic Characterisation and Ecological Drivers" *Lancet Infectious Diseases*
17. Naveca, Nascimento, Souza et al. (2017) "Multiplexed Reverse Transcription Real-Time Polymerase Chain Reaction for Simultaneous Detection of Mayaro, Oropouche, and Oropouche-Like Viruses" *Memórias do Instituto Oswaldo Cruz*
18. Katoh, Standley (2013) "MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability" *Molecular Biology and Evolution*
19. Larsson (2014) "AliView: A Fast and Lightweight Alignment Viewer and Editor for Large Datasets" *Bioinformatics*
20. Minh, Schmidt, Chernomor (2020) "IQ-tree 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era" *Molecular Biology and Evolution*
21. Kalyaanamoorthy, Minh, Wong et al. (2017) "Modelfinder: Fast Model Selection for Accurate Phylogenetic Estimates" *Nature Methods*
22. Hadfield, Megill, Bell (2018) "Nextstrain: Real-Time Tracking of Pathogen Evolution" *Bioinformatics*
23. Martin, Varsani, Roumagnac (2020) "RDP5: a Computer Program for Analyzing Recombination in, and Removing Signals of Recombination from, Nucleotide Sequence Datasets" *Virus Evolution*
24. Pond, Frost (2005) "Not so Different After All: A Comparison of Methods for Detecting Amino Acid Sites Under Selection" *Molecular Biology and Evolution*
25. Murrell, Wertheim, Moola et al. (2012) "Detecting Individual Sites Subject to Episodic Diversifying Selection" *PLoS Genetics*
26. Murrell, Moola, Mabona (2013) "FUBAR: A Fast, Unconstrained Bayesian Approximation for Inferring Selection" *Molecular Biology and Evolution*
27. Kosakovsky Pond, Poon, Velazquez (2019) "HyPhy 2.5 A Customizable Platform for Evolutionary Hypothesis Testing Using Phylogenies" *Molecular Biology and Evolution*
28. Weaver, Shank, Spielman et al. (2018) "Datamonkey 2.0: A Modern Web Application for Characterizing Selective and Other Evolutionary Processes" *Molecular Biology and Evolution*
29. Kumar, Stecher, Li et al. (2018) "MEGA X: Molecular Evolutionary Genetics Analysis Across Computing Platforms" *Molecular Biology and Evolution*
30. Edgar (2004) "MUSCLE: A Multiple Sequence Alignment Method With Reduced Time and Space Complexity" *BMC Bioinformatics*
31. Mirdita, Schütze, Moriwaki et al. (2022) "Colabfold: Making Protein Folding Accessible to All" *Nature Methods*
32. Cianfrocco, Wong-Barnum, Youn et al. (2017) "COSMIC2: A Science Gateway for Cryo-Electron Microscopy Structure Determination" *Practice and Experience in Advanced Research Computing*
33. Desai, Otter, Koopmans (2024) "Oropouche Virus: A Re-Emerging Arbovirus of Clinical Significance" *IJID regions*
34. Embratur "Briefing para Alagoas 2024"
35. Coimbra, Santos, Cardoso (2007) "Vigilância alimentar e nutricional para a Saúde Indígena"
36. Rios, Santarém, Ribeiro Júnior (2021) "Culicoides Insignis Lutz, 1913 (Diptera: Ceratopogonidae) Biting Midges in Northeast of Brazil" *Insects*
37. Sah, Srivastava, Mehta (2024) "Oropouche Fever Fatalities and Vertical Transmission in South America: Implications of a Potential New Mode of Transmission" *Lancet regional health*
38. Garcia Filho, Lima Neto, Maia (2024) "A Case of Vertical Transmission of Oropouche Virus in Brazil" *New England Journal of Medicine*
39. Azevedo, Silva, Silva (2024) "Genomic and Phenotypic Characterization of the Oropouche Virus Strain Implicated in the 2022-24 Large-Scale Outbreak in Brazil" *Journal of Medical Virology*
40. Files, Hansen, Herrera (2022) "Baseline Mapping of Oropouche Virology, Epidemiology, Therapeutics, and Vaccine Research and Development" *NPJ Vaccines*
41. Gutierrez, Wise, Pullan (2020) "Evolutionary Dynamics of Oropouche Virus in South America" *Journal of Virology*
42. Usuga, Limonta, Perez-Restrepo
43. (2024) "Emerging Infectious Diseases"
44. Moreira, Dutra, De Carvalho (2024) "Oropouche Virus Genomic Surveillance in Brazil" *Lancet Infectious Diseases*
45. Hellert, Aebischer, Haouz (2023) "Structure, Function, and Evolution of the Orthobunyavirus Membrane Fusion Glycoprotein" *Cell Reports* |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.