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=PMC12365549&blobtype=pdf | # Correction: Pre-activation of toll-like receptor 2 enhances CD8+ T-cell responses and accelerates hepatitis B virus clearance in the mouse models
Yong Lin, Xuan Huang, Jun Wu, Jia Liu, Mingfa Chen, Zhiyong Ma, Ejuan Zhang, Yan Liu, Shunmei Huang, Qian Li, Xiaoyong Zhang, Jinlin Hou, Dongliang Yang, Mengji Lu, Yang Xu, Wu Liu, Ma Zhang, E Liu, Huang Li, Luwen Zhang, Ma Lin, Yang Zhang
## Abstract
In the published article, there was an error in Figure 2 as published. Panel D of Figure 2 appears in the published article by mistake. These errors have been corrected, and the updated panel D has been incorporated into the figure. The corrected Figure 2 and its caption, appear below.The authors apologize for this error and state that this does not change the scientific conclusions of the article in any way. The original article has been updated.
## FIGURE 2 (Continued)
Early application of TLR2 ligand P3C with pAAV-HBV1.2 by HI inhibits HBV replication without promoting HBV-specific immune response in the mouse model for persistent HBV replication. C57BL/6 mice received hydrodynamic injection (HI) with plasmid pAAV-HBV1.2. The mice were treated three times with 50 mg of P3C or PBS administered by subcutaneous (SC) injection at day 0, 7, and 14 (therefore designated as group D0). (A) Serological markers of HBV infection HBsAg, HBeAg, and HBV DNA were assayed at the indicated time points by ECLIA (Roche). The cut-off value of the HBsAg and HBeAg assays was set at cut-off index (COI) of 1.0. The cut-off value of the HBV DNA real-time PCR was 4.0 × 10 4 copies/ ml. (B) Positivity for HBsAg or HBeAg was defined as ≥1. (C) HBV DNA levels in the liver were measured by quantitative real-time PCR. (D) Liver tissue sections were stained with anti-HBc antibodies (magnification, ×200). The number of HBcAg positive hepatocytes was counted. (E) The serum levels of anti-HBs and anti-HBc antibodies were detected at the indicated time points by ECLIA. The cut-off value of anti-HBs antibody assay was 10 IU/L. The cut-off value of anti-HBc antibody assay was 1.0 COI (<1.0 COI indicates a positive reaction). (F, G) Lymphocytes were isolated from the mouse liver at day 10, 21, and 77 after HI. (F) The specific CD8 + T cells against HBcAg Cor 93-100 epitope were detected by Cor 93-100 peptide-loaded dimer staining. (G) The functionality of HBV-specific CD8 + T cells was determined by intracellular cytokine staining after ex vivo stimulation with peptide Cor 93-100 for 5 h. (H) Liver tissues were collected from the mouse liver at day 4, 10, 21, and 77 after HI. The mRNA expression levels of cytokines in the liver were determined by real-time RT-PCR. Beta-actin was used as an internal reference. Eight mice were analysed per group, and the experiments were repeated at least once. Data were analysed using an unpaired Student's t test. Statistically significant differences between the groups are indicated as *P < 0.05 and **P < 0.01. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12456139&blobtype=pdf | # Endogenous human herpesviruses 6A/B
Louis Flamand, Jesse Arbuckle, Pascale Bonnafous, Vincent Descamps, Joshua Hill, Ruth Jarrett, Keith Jerome, Benedikt Kaufer, Rie Koide, Anthony Komaroff, Peter Medveczky, Hiroki Miura, Yasuko Mori, Nicholas Parrish, Philip Pellett, Michael Wood, Tetsushi Yoshikawa, Danielle Zerr
## Abstract
Human herpesviruses 6A and 6B (HHV-6A/B) can integrate into the germline, resulting in inherited viral DNA-now proposed to be called "endogenous HHV-6A/B (eHHV-6A/B). " Present in 0.2-3% of humans, this integrated DNA is passed to offspring and may reactivate, posing health risks such as angina or lupus. To reduce confusion caused by varied terminology, researchers advocate using "eHHV-6A/B" for inherited forms and reserving "chromosomally integrated" for somatic integrations only.
M ost humans are infected with human herpesviruses 6B and/or 6A (HHV-6B/A) and are at risk of reactivation of these viruses, especially when immunocompromised (1)(2)(3). On numerous occasions in human (and possibly hominin [4]) history, HHV-6A/B integrated their genomes into the host germline. Therefore, the virally derived DNA has been passed on to descendants in a Mendelian manner and is present in the chromo somes of every diploid cell.
The viral genomic DNA integrates into the telomeric region of chromosomes and is present in the same chromosomal location in members of the same family (5)(6)(7)(8)(9). This inherited HHV-6A/B DNA is somewhat analogous to the endogenous retroviral DNA that constitutes about 8% of the human genome.
Chromosomally integrated HHV-6A/B can reactivate, express viral genes, and even produce infectious progeny, as isolation of infectious virus identical in sequence to the endogenous HHV-6 has been documented (10)(11)(12)(13). Viral DNA integration represents one way to achieve latency: the conventional means of achieving latency in herpesviruses, episomal viral DNA, has not yet been demonstrated in HHV-6A/B.
Depending on ancestry and geographic region, the prevalence of humans carrying endogenous HHV-6A/B varies between 0.2 and 3% (14)(15)(16)(17)(18). The related health conse quences of carrying inherited HHV-6A/B, and the implications for medical practice, are the subject of active investigation (19). Preliminary evidence suggests that the condition may increase the risk for angina pectoris (17,20), preeclampsia (21), and systemic lupus erythematosus (18).
Scientists have used several names for this condition, including "chromosomally integrated HHV-6 (or ciHHV-6), " "inherited chromosomally integrated HHV-6 (or iciHHV-6), " or, more recently, "endogenous HHV-6 (eHHV-6)" (4,18). This has created confusion in the literature.
To avoid confusion, the undersigned authors who study HHV-6A/B urge that all investigators adopt the term "eHHV-6A/B" when referring to inherited HHV-6A/B DNA. For the next few years, after using eHHV-6A/B for the first time in a manuscript, authors should add a parenthetical "(previously called ciHHV-6 or iciHHV-6), " for clarification. As virologists and many health care providers and biomedical scientists are familiar with the term "endogenous retroviral DNA, " the use of the term eHHV-6A/B is simpler, shorter, and also clearer.
Similarly, the term "ciHHV-6" should be used only to refer to somatic chromosomal integrations.
A draft of this letter was posted on the HHV-6 Foundation website, inviting scientists to join in signing this letter (22).
The undersigned commit ourselves to using this terminology exclusively, henceforth.
## References
1. Zerr, Gooley, Yeung et al. (2001) "Human herpesvirus 6 reactivation and encephalitis in allogeneic bone marrow transplant recipients" *Clin Infect Dis*
2. Hill, Koo, Suarez et al. (2012) "Cord-blood hematopoietic stem cell transplant confers an increased risk for human herpesvirus-6-associated acute limbic encephalitis: a cohort analysis" *Biol Blood Marrow Transplant*
3. Ogata, Satou, Inoue et al. (2013) "Foscarnet against human herpesvirus (HHV)-6 reactivation after allo-SCT: breakthrough HHV-6 encephalitis following antiviral prophylaxis" *Bone Marrow Transplant*
4. Aswad, Aimola, Wight et al. (2021) "Evolutionary history of endogenous human herpesvirus 6 reflects human migration out of Africa" *Mol Biol Evol*
5. (2025) *Commentary Journal of Virology*
6. Daibata, Taguchi, Nemoto et al. (1999) "Inheritance of chromosomally integrated human herpesvirus 6 DNA" *Blood*
7. Tanaka-Taya, Sashihara, Kurahashi et al. (2004) "Human herpesvirus 6 (HHV-6) is transmitted from parent to child in an integrated form and characterization of cases with chromosomally integrated HHV-6 DNA" *J Med Virol*
8. Arbuckle, Medveczky, Luka et al. (2010) "The latent human herpesvirus-6A genome specifically integrates in telomeres of human chromosomes in vivo and in vitro" *Proc Natl Acad Sci*
9. Arbuckle, Pantry, Medveczky et al. (2013) "Mapping the telomere integrated genome of human herpesvirus 6A and 6B" *Virology (Auckl)*
10. Wight, Aimola, Aswad et al. (2020) "Unbiased optical mapping of telomere-integrated endogenous human herpesvirus 6" *Proc Natl Acad Sci*
11. Endo, Watanabe, Ohye et al. (2014) "Molecular and virological evidence of viral activation from chromosomally integrated human herpesvirus 6A in a patient with X-linked severe combined immunodeficiency" *Clin Infect Dis*
12. Peddu, Dubuc, Gravel et al. (2019) "Inherited chromosomally integrated human herpesvirus 6 demonstrates tissuespecific RNA expression in vivo that correlates with an increased antibody immune response" *J Virol*
13. Jvi
14. Wood, Veal, Neumann et al. (2021) "Variation in human herpesvirus 6B telomeric integration, excision, and transmission between tissues and individuals" *Elife*
15. Hannolainen, Pyöriä, Pratas et al. (2025) "Reactivation of a transplant recipient's inherited human herpesvirus 6 and implica tions to the graft" *J Infect Dis*
16. Leong, Tuke, Tedder et al. (2007) "The prevalence of chromosomally integrated human herpesvirus 6 genomes in the blood of UK blood donors" *J Med Virol*
17. Potenza, Barozzi, Masetti et al. (2009) "Prevalence of human herpesvirus-6 chromosomal integration (CIHHV-6) in Italian solid organ and allogeneic stem cell transplant patients" *Am J Transplant*
18. Hubacek, Muzikova, Hrdlickova et al. (2009) "Prevalence of HHV-6 integrated chromoso mally among children treated for acute lymphoblastic or myeloid leukemia in the Czech Republic" *J Med Virol*
19. Gravel, Dubuc, Morissette et al. (2015) "Inherited chromosomally integrated human herpesvirus 6 as a predisposing risk factor for the development of angina pectoris" *Proc Natl Acad Sci*
20. Sasa, Kojima, Koide et al. (2025) "Blood DNA virome associates with autoimmune diseases and COVID-19" *Nat Genet*
21. Pellett, Ablashi, Ambros et al. (2012) "Chromosomally integrated human herpesvirus 6: questions and answers" *Rev Med Virol*
22. Wood, Bell, Young et al. (2025) "Inherited chromosomally integrated human herpesvirus 6: regional variation in prevalence, association with angina, and identification of ancestral viral lineages in two large UK studies" *J Virol*
23. Gaccioli, Lager, De Goffau et al. (2020) "Fetal inheritance of chromosomally integrated human herpesvirus 6 predisposes the mother to pre-eclampsia" *Nat Microbiol*
24. Hhv-6 (2025) "Leading HHV-6 investigators encourage using new name for iciHHV-6" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12831225&blobtype=pdf | # Rhinovirus Infects B and CD4 T Lymphocytes in Hypertrophic Tonsils in Children
Ronaldo Martins, Flavia De Paula, | Talita, B Mitchell, Miria Criado, Ricardo Cardoso, Bruna Jesus, Italo Castro, | Murilo, Henrique Cassiano, Daniela Méria, Ramos Rodrigues, Noilson Oliveira, Lucas Carenzi, | Fabiana, C Valera, | Tamashiro, Wilma Anselmo-Lima, Eurico Arruda
## Abstract
Prolonged detection of rhinovirus (RV) in secretions after a typical cold and asymptomatic shedding are frequently reported. Although RV has been detected in human hypertrophic tonsils, its replicative status and host cell range remain unclear. In this study, we analyzed RV replication, infected cell types, and recovery of infectious virus in adenoids, palatine tonsils, and respiratory secretions from 293 children with tonsillar hypertrophy undergoing tonsillectomy. Samples were screened by real-time RT-PCR, and RV-positive samples were analyzed using immunohistochemistry (IHC), chromogenic in situ hybridization (CISH), flow cytometry, and RV isolation in cell culture. RV genotypes from species A, B, and C were identified in adenotonsillar samples. RV antigenome and structural proteins were detected in tonsillar epithelial surfaces, parenchyma, and in CD4 + T and B lymphocytes. Infectious RV was recovered from adenoids and respiratory secretions. In vitro infection of tonsillar mononuclear cells with RV-16 and RV-1A resulted in viral progeny production and secretion of distinct cytokine profiles. These findings demonstrate that RV infects tonsillar T and B lymphocytes, suggesting that tonsils can serve as sites of prolonged infection and sources of RV shedding. RV infection of immune cells may have potential impact on the local immune microenvironment.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.
## 1 | Introduction
Rhinovirus (RV) is the most frequent cause of common cold, a mild acute respiratory infection (ARI) often complicated by asthma exacerbation, acute otitis media, or sinusitis [1,2]. We have previously reported the detection of RV RNA by real-time polymerase chain reaction (RT-PCR) in 25% of tissues removed from adenoids, 12% from palatine tonsils, and 35% in nasopharyngeal secretions (NPS) from children with chronic tonsillar hypertrophy undergoing adenotonsillectomy, in the absence of ARI symptoms [3]. Similar findings were later reported by other groups [4,5]. However, these earlier studies focused predominantly on viral RNA detection, and did not establish whether the virus was actively replicating, which cell types were infected, or whether infectious virus could be recovered from these tissues. Moreover, the immunological implications of RV infection in tonsillar lymphocytesparticularly CD4⁺ T and B cells-remain largely unexplored.
In this study, we aimed to fill these knowledge gaps by analyzing not only the presence of RV genomes but also the replicative status of the virus, the phenotypes of infected cells, and the recovery of infectious RV from tonsillar tissues and respiratory secretions of a large cohort of children (n = 293) undergoing tonsillectomy. Our findings demonstrate active RV replication in tonsillar tissues, including in B and CD4⁺ T lymphocytes, and provide compelling evidence that both adenoids and palatine tonsils can act as reservoirs for infectious RV. In several patients, the same infectious RV genotypes were simultaneously identified in adenoids, palatine tonsils, and nasopharyngeal secretions, supporting the role of tonsils as potential viral reservoirs. Furthermore, RV infection of tonsillar lymphocytes resulted in the production of infectious progeny and triggered distinct pro-inflammatory cytokine responses, including elevated levels of IL-17, TNF-α, and IFN-γ, suggesting that RV may modulate the local immune microenvironment within lymphoid tissues. Collectively, the findings suggest that tonsils could be reservoirs of RV. This represents a novel insight into the biology of RV, with potential implications for understanding prolonged shedding, virus persistence, and local immune modulation.
## 2 | Methods
## 2.1 | Study Design and Sample Processing
This prospective cross-sectional study enrolled 293 children (54.6% boys) 3 to 13 years of age (median 6 years; mean 5 years) who underwent adeno-tonsillectomy due to tonsillar hypertrophy or recurrent tonsillitis, at the Division of Otorhinolaryngology, Clinical Hospital of the University of São Paulo School of Medicine, in Ribeirão Preto, Brazil. Exclusion criteria were the presence of ARI symptoms or antibiotic treatment within 1 month before surgery. Informed consents were obtained from parents/guardians in compliance with the Human Research Ethics Committee of the University of Sao Paulo Clinical Hospital, Ribeirão Preto (number 10466/2008). This was an exploratory study aimed at characterizing rhinovirus infection in tonsillar tissues. Mucosal swabs were collected from the regions of adenoids and palatine tonsils under direct view after anesthesia, and fragments of tonsillar tissues were obtained during surgery. Specimens were transported on ice in less than 2 h to the virology laboratory, where they were processed and tested by real-time polymerase chain reaction (RT-PCR) for RV. Briefly, fragments of tissues of approximately 1 g including epithelial surface and parenchyma were preserved in RNAlater (Invitrogen, Carlsbad, CA, USA) for nucleic acid extraction. From the same tissues, a second similar-size fragment was placed in freezing medium for later virus isolation, another was dissociated by enzymatic digestion to purify tonsillar mononuclear cells (TMNCs), and yet another was fixed and paraffin-embedded for histology.
TMNCs were prepared by tissue digestion with dispase (0.6 U/ mL) and collagenase-I (100 U/mL) (both from Gibco, Grand Island, NY, USA). The fragments destined to histology were fixed for 4 h in Carnoy's fixative, composed of 60% ethanol, 30% acetic acid, and 10% chloroform (Merck, Darmstadt, Germany), then dehydrated in ethanol series (JT Baker, Mexico) for 1 h each (70%, 80%, 90%, 95%, and three changes of 100%), followed by 1:1 xylene:ethanol for 20 min, and three 30-min changes of pure xylene (Synth, Diadema, SP, Brazil). The tissues were then paraffin-embedded by dipping in two changes of paraffin (Merck) for 2 h each, and the blocks were used to prepare 3 µm sections on positively charged slides (Fisher, Pittsburgh, USA). Nasopharyngeal swabs (NPS) were eluted in PBS and the eluate was split for extraction by Trizol and for freezing at -80°C until further testing.
A schematic flowchart summarizing the sample collection and processing workflow is presented in Supporting Information Figure 1.
## 2.2 | Detection of RV Genomes and Genotyping of RV Species
The detection of RV and EV genomes was performed using TaqMan real-time RT-PCR, with total nucleic acids extracted via Trizol® from 250 µL of homogenized tissue samples or swab eluates, in accordance with protocols previously published by our group [3,6]. Specific primers and probes for RV and EV, as well as for the housekeeping genes β-actin and RNase P, are listed in Table 1. Reverse transcription was done on 1 µg of extracted RNA using random hexamers and Multiscribe Reverse Transcriptase (Applied Biosystems, Foster City, CA, USA). All real-time PCR assays were done on a Step-One Plus real-time PCR thermocycler (Applied Biosystems, Foster City, CA, USA) in a final volume of 15 µL using 3 µL of cDNA, 10 µM forward and reverse primers, 5 µM probe, 0.15 µL of Rox and 7.5 µL of TaqMan master mix (Sigma-Aldrich, St. Louis, MO, EUA). The cycling was 95°C for 3 min, followed by 45 cycles of 95°C for 10 s, and 60°C for 30 s, with final soaking at 10°C. Realtime PCR for enterovirus (EV) was done in a final volume of 10 µL, with 3 µL of cDNA, 10 µM forward and reverse primers, 5 µM probe, 5 µL of TaqMan master mix (Applied Biosystems, Foster City, CA, USA), with cycling at 95°C for 10 min, followed by 45 cycles of 95°C for 15 s, and 60°C for 1 min. RV genotyping was done in the RV-positive samples by conventional RT-PCR using primers targeting conserved sequences in the 5′UTR of the RV genome, followed by sequencing of the PCR products [1]. Briefly, RV amplicons to be sequenced were generated by two sequential amplification steps, following a previously published protocol and a mix of three forward primers (B1-3) plus one reverse primer R2 (Table 1). Reactions were done at the final concentration of 25 μM, with the enzyme Platinum PCR supermix HF (Invitrogen), on 100 ng of cDNA, according to the following parameters: touch-down with two cycles of incubation at 94°C for 2 min and 20 s, plus 68°C for 70 s. The cycling was repeated eight times at sequential annealing temperatures of 66°C, 64°C, 62°C, 60°C, 58°C, 56°C, 54°C, and 52°C for two cycles each, and a final 10-min extension at 68°C. After the first round of PCR, a second round was performed using 2.5 µL of the product of the first round, with the same primers, using the following parameters: 94°C for 2 min, 94°C for 20 s, 52°C for 30 s, and 68°C for 40 s, with 27 repetitions and a final extension at 68°C for 3 min. The PCR products were treated with ExoSAP-IT® (Affymetrix, Santa Clara, CA, EUA) with incubation at 37°C for 25 min, and then at 90°C for 15 min. The sequencing reaction was carried out with 3 μL of the PCR product using 2 μL of BigDye Terminator v3.1 Cycle Senquencing RR-100 (Applied Biosystems, Foster City, CA, EUA), 10 pmol of the reverse primer R2 and 3 μL of buffer, at the final volume of 20 μL, in an ABI Hitachi 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA, EUA). All sequences were analyzed by DNASTAR Lasergene Genomics software and the FASTA sequences were used in a BLAST search using the cut-off parameters of identity higher than 95% and e-value higher 10⁻⁵.
## 2.3 | Serial Immunohistochemistry (SIMPLE)
To enhance the detection of multiple rhinovirus genotypes in naturally infected samples, a 1:1 antibody blend (by volume and final concentration) of anti-VP2 (mabR16-7, QED Bioscience, San Diego, CA, USA) and anti-VP1 (mab8430, Millipore, Temecula, CA, USA) was used at a final dilution of 1:1000 in PBS/BSA containing 0.1% Triton X-100 (Sigma, St. Louis, MO, USA). This antibody combination was chosen based on the manufacturer's specifications and our own validation experiments. Specifically, mabR16-7 was confirmed to selectively detect RV-16 and RV-1A, whereas mab8430, originally developed against enterovirus VP1, displayed cross-reactivity with stocks of RV-14, RV-39, RV-16, and RV-1A maintained in our laboratory. The rationale for using this blend was to increase the detection across rhinovirus genotypes frequently observed in clinical specimens, acknowledging the potential for cross-reactivity with other enteroviruses. To minimize this, antibody blends used for IHC were applied only to samples that tested positive for RV by real-time PCR, but not for enteroviruses. Tissues positive for both RV and enteroviruses by PCR were excluded from IHC analysis. Tissue sections were then incubated with horse anti-mouse biotinylated antibody (BA-2000, Vectastain ABC Kit, Vector Laboratories, Burlingame, CA). Signal amplification was attained by incubation with polymer conjugated with streptavidin-peroxidase (s2438, Sigma-Aldrich), and color development was done with AEC peroxidase system (SK-4800, Vector Laboratories, Burlingame, CA), resulting in redpurple staining in positive cells, followed by tissue counterstaining with Harris hematoxylin (Vector) [7]. To determine the types of RV-infected lymphomononuclear cells, sequential immunoperoxidase labeling and erasing (SIMPLE) [8] was done with the above-mentioned MAbs for virus structural protein, followed by staining with rabbit polyclonal antibodies for CD3, CD4 (AB 133616, dilution 1:100, Abcam), CD8 (AB 4055, dilution 1:100, Abcam), CD20 (AB 27093, dilution 1:100, Abcam), CD11c (AB-52632, dilution 1:100, Abcam). Goat anti-rabbit biotinylated antibody (ab64256, Abcam) was used as a secondary antibody for the immune phenotype staining, and color development was generated with the Vector AEC peroxidase system (SK-4200, Vector Laboratories, Burlingame, CA). After counterstaining with Harris hematoxylin, slides were mounted with coverslips with an aqueous mounting medium. After whole-slide highresolution scanning, coverslips were removed in distilled water and slides were dehydrated through an ethanol gradient to 95% ethanol. Slides were then incubated in ethanol series until the complete erasing of the AEC color. Following rehydration, previous antibodies were eluted by incubation of the tissue sections in 0.15 mM KMnO₄/0.01 M H₂SO₄ solution for 2 min, immediately followed by a distilled water wash. The same tissue sections were then stained for another antigen, beginning at the blocking step. RV-infected and uninfected HeLa cells were included as positive and negative controls in all tested batches.
## 2.4 | Chromogenic In Situ Hybridization
Chromogenic in situ hybridization (CISH) was adapted from published procedures [9]. Using a locked nucleic acid (LNA) probe consisting of a sense oligonucleotide (Dig/ GCACTTCTGTTTCCCC/Dig) targeting the RV negative-strand replicative intermediate anti-genome RNA at a sequence within the 5′UTR. The tissues were deparaffinized in three serial xylene baths (Synth) for 5 min each, and then rehydrated in decreasing concentrations of ethanol (100%, 96%, and 70%) (JTBaker), in two changes of each concentration, the first for 1 min, and the second for 5 min. The sections were then washed for 5 min in PBS and incubated with 10 µg/mL proteinase K
## 2.5 | Flow Cytometry
The types of TMNCs were determined by flow cytometry using the antibodies CD3-PE-CF594 (562406), CD4-PerCP-Cy5.5 (560650), and CD20-PE-Cy7 (560735), all acquired from BD Horizon (Franklin Lakes, NJ, USA). Internal flow cytometry controls were done by incubating RV-infected cells with the fluorescence-conjugated secondary antibody in the absence of primary antibody. Cell suspensions were fixed with 4% paraformaldehyde for 20 min, followed by incubation with permeabilization buffer (0.1% saponin in PBS) for 5 min, and then with blocking solution (5% BSA in PBS) for 30 min. For the staining of viral structural protein, cells were incubated with a blend of anti-VP2 (mabR16-7) and anti-VP1 (mab8430) diluted 1:300, for 1 h at RT, followed by three washes in PBS, and then incubated with FITC-labeled secondary goat anti-mouse IgG Apl24F antibody (Millipore, Temecula, CA, USA) for 30 min at RT. Flow cytometry was performed on a Becton Dickinson 6 FACS-Canto machine, and data analyzes were done with FlowJo version 9.4.3, after appropriate gating on CD3 + /CD4 + and CD20 + populations. As previously stated, a 1:1 blend of anti-VP2 (mabR16-7) and anti-VP1 (mab8430) antibodies was used to enhance detection sensitivity across a broader range of RV genotypes. Flow cytometry data were acquired from three independent experiments, with a minimum of 20,000 gated events per sample to ensure statistical robustness.
## 2.6 | Rhinovirus Isolation
Isolation of RV from tissue fragments and NPS was attempted in HeLa-I and WI-38 cells, based on published procedures [10]. Briefly, tissue homogenized on TissueLyser LT (Qiagen) was snap-frozen in liquid nitrogen and then thawed, clarified by centrifugation (10 min at 1000g at 4°C), and filtered through 0. Slides were defrosted and incubated with a permeabilizing/ blocking solution (PBS with 0.01% Triton, 1% BSA Fraction V (Sigma-Aldrich), and 5% goat serum for 5 min at RT. Slides were washed in PBS, incubated with anti-VP2/anti-VP1 diluted 1:300 in PBS with 1%BSA at RT for 30 min, washed three times with PBS and incubated with FITC-labeled goat anti-mouse IgG antibody Apl24F (Millipore, Temecula, CA, USA) diluted 1:100 in PBS, for 30 min at room temperature protected from light, and then washed three times with PBS. Slides were incubated for 5 min at RT with DAPI (1 µg/ml) for nuclei staining, and then mounted in mounting media (Dako) to be examined in a fluorescence microscope. Cross-reactivity of anti-VP1 EV blend with RV was confirmed using major and minor types RV-16 and RV-1A (Supporting Information Figure 2). Samples that did not cause CPE in cultures were subject to 3 additional blind passages in fresh monolayers before being discarded as negative.
## 2.7 | Infection of TMNCs With RV In Vitro
TMNCs prepared from tonsils negative for RV and EV by RT-PCR were obtained by enzymatic dissociation of tonsillar tissue fragments with dispase-collagenase and Ficoll-Paque Plus® (Merck Millipore, EUA) density-gradient centrifugation. Briefly, RV-and EV-negative TMNCs were inoculated in suspension (MOI = 1) by incubation for 2 h at 33°C with RV-16, RV-1A, respectively major and minor RV genotypes, or mockinoculated with extracts of uninfected HeLa cells. The MOI of 1 was chosen to maximize the exposure of the heterogeneous TMNC population, and enable the detection of viral replication and cytokine response within a short period of time post infection. After inoculation, TMNCs were washed in the Hanks' balanced salt solution (HBSS) buffer, resuspended in RPMI, and cultured for 24 h at 33°C in 5% CO 2 . Virus titers were determined in cells and supernatants by TCID 50 in HeLa cells at 2, 4, 6, 8, 12, 16, and 24 h post-infection. The cytokines IL-2, IL-4, IL-6, IL-10, TNF, IFNγ, and IL-17A were quantified in the supernatants from RV-infected TMNCs using a human cytometric bead array (CBA) Th1/Th2/Th17 Cytokine Kit (BD Biosciences) following the manufacturer's instructions, using BD Accuri C6 Plus Personal Flow Cytometer.
## 3 | Results
## 3.1 | The Detection and Typing of RV in Adenoids and Palatine Tonsils
Different RV genotypes were detected by RT-PCR in at least one sampling site from 137 of 293 (46.7%) patients. Overall, RV was detected more frequently in adenoids (97 of 293; 33%) than palatine tonsils (74 of 293; 25.2%), but the difference was not significant (p-value 0.9820). RV was detected in 27% (78 out of 293) samples of NPS (Supporting Information Figure 3A). RV was detected simultaneously in the three sample sites in 24 of the 137 RV-positive patients (18%) (Supporting Information Figure 3B). For the purpose of this study of rhinoviruses, all samples that were also positive for EVs by RT-PCR were excluded from further analyzes: 58 of 74 RV-positive (78,4%) palatine tonsils, 64 of 97 (66%) RV-positive adenoids, and 9 of 78 (11.5%) RV-positive NPS. After applying this exclusion criterion, a total of 55 samples remained RV-positive and EV-negative and were included in the downstream analyzes. Sequencing of the 5′ UTR RT-PCR products enabled RV genotyping in 46 samples (Supporting Information Table 1). The analysis of RV genotypes revealed RV of species A in 21/46 (42%), species B in 18/46 (36%), and species C in 11/46 (22%) samples (Supporting Information Table 2). Among the RV genotypes identified, no significant difference was observed between the detection frequencies of RV-A, RV-B, and RV-C (p-value 0.157), likely reflecting the limited number of genotyped samples. Several genotypes of RV species A and B were detected simultaneously in all three sampling sites, and in 11 patients the same RV genotype was detected simultaneously from both tonsil types and corresponding NPS (Supporting Information Table 3). The 5′UTR sequencing strategy used does not allow for accurate genotyping of RV species C, which is a limitation of the study.
## 3.2 | In Situ RV Detection by IHC and CISH
Tonsillar tissues that were positive for RV and negative for EV by RT-PCR were investigated by IHC for RV capsid proteins VP1 or VP2, and by CISH with a probe specific for the RV antigenome (Figure 1). Of 16 palatine tonsils and 33 adenoids tested, respectively 9 (56%) and 20 (60%) were positive by IHC for RV antigen. In adenoids, IHC signal was abundant in the pseudostratified ciliated epithelium (Figure 1A), and also detected, yet less abundantly, in lymphomononuclear cells within lymphoid follicles and in extra-follicular regions (Figure 1B). In palatine tonsils, RV antigens were detected predominantly in the cytoplasm of cells in lymphoid compartments, both follicular and extra-follicular, but also in patches of crypt reticulated epithelium and surface squamous epithelium.
To confirm the presence of RV replication in tonsils, CISH was done with a probe specific for the anti-genome on RV-positive/ EV-negative palatine tonsils and adenoids. Tissues from 13 patients were available for testing with the probe for the RV antigenome: palatine tonsil and adenoid from 3, only palatine tonsil from 4, and only adenoid from 6 patients. All tissues from the 13 RV-positive/EV-negative patients were positive by CISH, with signal on the surface epithelia and in lymphoid compartments, both in and out of lymphoid follicles (Figure 1C). RVinfected and uninfected HeLa cells were included as positive and negative controls in all tested batches, and control tonsillar tissues (n = 6) that tested negative for RV and EV by RT-PCR were also negative by IHC and CISH (Supporting Information Figure 4).
## 3.3 | RV Isolation
To assess the presence of infectious RV in tonsillar tissues and secretions, available samples of tissue macerates and NPS from 17 RV-positive/EV-negative patients were inoculated in cell cultures, and RV was recovered from 10 of them (58.8%). The same RV genotypes, RV-4 and RV-14, were isolated simultaneously from NPS and adenoid tissues in two patients, while in the remaining eight patients RV was isolated exclusively from adenoid tissues (Table 2).
## 3.4 | Cell Types Infected by RV and Tonsillar Cytokine Profile
The RV structural protein VP2 was detected in epithelial cells, as evidenced by the serial staining for epithelial cell surface antigen EPCAM (Figure 2A-C). In the parenchyma, sequential IHC showed that follicular VP-2 positive cells were also positive for CD20 (Figure 2D-F), confirming that RV infects B lymphocytes in human tonsils. Moreover, VP2-positive cells (Figure 2I) were also positive for CD3 (Figure 2G) and CD4 (Figure 2H), thus confirming that RV infects CD4 + T lymphocytes (Figura 2J) in human tonsils in vivo. Of note, RV VP2 was not detected in CD8⁺ or CD11c⁺ cells in our experimental setting (Supporting Information Figure 5), suggesting that RV infection may be specific to certain lympho-hematopoietic cell types in this kind of tissue, and in our experimental conditions. To further assess celltype susceptibility, we performed in vitro infections of TMNCs dissociated from tissues with RV-16 and RV-1A, followed by RIFI staining. Again, no RV antigen was detected in CD8⁺ or CD11c⁺ cells, in further support of the findings obtained in naturally infected tissues (Supporting Information Figure 6). Further studies will be needed to verify whether this restriction is generalizable to other immunological settings. Flow cytometry experiments were conducted to determine the relative proportions of different types of RV-infected cells in the tissues and revealed that approximately 25.57% (median: 23.0%) of CD4 + T lymphocytes and approximately 39.78% (median: 41.55%) of B lymphocytes were positive for RV VP2 (Figure 2J,L), further supporting the data obtained in situ.
To obtain further evidence of susceptibility and permissiveness of tonsillar cells to RV, primary cultures of TMNCs dissociated from tonsils negative by real-time PCR for both RV and EV were infected in vitro with RV-16 or RV-1A (MOI = 1) and analyzed 24 h post-infection by TCID 50 assay. There was substantial increase in RV-16 progeny production in primary TMNCs (Figure 3A,B), while titers of RV-1A progressively decreased by approximately 100-fold in the first 24 h (Figure 3C,D). Interestingly, both RV genotypes significantly impacted the production of cytokines by TMNCs, albeit in distinct ways. Both RV-16 and RV-1A induced significant increase in production of IL-17 and TNF-α in TMNCs at 24 h post-infection (hpi) compared to mock-inoculated cultures. However, RV-16 induced a significant secretion of IFN-γ in addition to IL-17 and TNF-α (Figure 3E), whereas RV-1A significantly induced the secretion of IL-6 (Figure 3F).
## 4 | Discussion
In the present study, RV genome was detected by RT-PCR in nearly one-third of adenoids and about one-quarter of palatine tonsils and NPS from children undergoing tonsillectomy for the treatment of tonsillar hypertrophy or recurrent tonsillitis, in the absence of ARI symptoms for at least 1 month before surgery. The replication of RV in respiratory epithelial cells has been known since our early studies by in situ hybridization of human tissues infected ex vivo and in vivo [11,12]. Those studies included infection of adenoid explants ex vivo [11] and nasal epithelium from experimentally infected volunteers [12].
This study expands knowledge of RV-susceptible host cells to include lymphoid cells in both follicular and extrafollicular compartments of naturally infected tonsils, alongside epithelial cells. Unlike a previous finnish study detecting picornaviruses [14], this study confirmed RV-specific localization by excluding enteroviruspositive tonsils via RT-PCR. In addition, the present study showed that RV infects B and CD4⁺ T lymphocytes in tonsils. Remarkably, RV was detected in nearly 40% of the CD20⁺ B lymphocytes from naturally infected tonsils, which may be due to the sheer predominance of B lymphocytes among tonsillar lymphoid cells [15]. In situ hybridization for RV antigenomes strongly supports the presence of RV active replication, not only in epithelial cells, but also in tonsillar lymphomononuclear cells. Further investigation will be needed to uncover the subsets of B and CD4⁺ T lymphocytes that are infected by RV, as well as their stages of maturation.
It is also noteworthy that in 18% of the RV-positive patients, the virus was detected simultaneously in both tonsil tissues and nasopharyngeal swabs, and sometimes the same genotype of RV was detected in the three sample types, suggesting that tonsils may be sources of shedding of RV into nasopharyngeal secretions. This is also supported by the isolation of RV, including cases with the same RV genotype from tonsillar tissues and nasopharyngeal secretions, confirming that there was production of infectious RV progeny in the tissues. Notably, the infiltration of tonsillar reticulated epithelium by lymphocytes, which is crucial for the initiation of immune responses [16], may conceivably create opportunities also for the passage of RV from lymphoid to polarized epithelial cells, in which the virus may replicate efficiently and produce progeny that appear in secretions. This finding underscores the importance of considering multiple sampling methods to comprehensively assess the prolonged detection of RV and its genotype diversity.
This study did not include follow-up investigations to assess secondary transmission. Nonetheless, the detection of RV in tonsillar tissues from asymptomatic children raises the possibility that such individuals could serve as sources of silent viral shedding. While the influence of weather and climate on RV epidemiology remains uncertain, the present findings suggest that further studies are needed on the roles of asymptomatic carriers/shedders in the onset of outbreaks in situations that entail exposure of susceptible individuals, such as the beginning of school term in the fall in temperate regions [17,18]. It should be kept in mind that the present cross-sectional study design makes it impossible to demonstrate virus persistence, a practical limitation that stems from the impossibility of sampling patients tissues over time.
Genome sequencing revealed the presence of all three rhinovirus species-RV-A, RV-B, and RV-C-in lymphoid tissue. Notably, since CDHR3-the receptor for RV-C-is poorly expressed in immune cells, the detection of RV-C RNA may reflect nonspecific uptake, infection of rare susceptible cells, or alternative entry mechanisms. In vitro infection of TMNCs from RV-negative tonsils showed differential permissiveness, with progeny production by RV-16 but not by RV-1A. Accordingly, the cytokine profiles induced by in TMNCs were also different between RV-16 and RV-1A, what may be correlated with the difference in viral progeny production, but could also result from dissimilar cell activations by the binding of the viruses to different receptors [18,19]. Previous studies have shown that high IFN-α and low RORC2 expressions were associated with RV-C, while RV-B was associated with low Tbet and high IFN-γ expressions [4,13]. Also, altered expression of inflammatory cytokines by macrophage exposure to different RVs indicated that RV-16 induced higher expression of CCL20, CCL2, CXCL10, and IL-10 by macrophages when compared to RV-1A [18]. It is important to note that the in vitro experiments were done only with RV-16 and RV-1A, cautioning against generalization to all rhinoviruses.
Previous in vitro studies done with peripheral blood mononuclear cells provided glimpses at how immune cells respond to RV, with intense proliferation of B cells [20], and activation of CD4 and CD8 + T lymphocytes, even in the absence of antigen-presenting cells [21]. Thus, it is conceivable that RV infection of different types of immune cells may activate them in tonsils. Moreover, RV infection in CD4 + T and B lymphocytes in tonsils may interfere with the mounting of local immune responses to different antigens and allergens. In that regard, it is remarkable that adenoid hypertrophy has been associated with an unfavorable course of childhood asthma, and that adenotonsillectomy has been associated with significant reductions in asthma attacks in children [22].
Similar to other persistent viral infections, the prolonged asymptomatic carriage of RV in tonsils could also entail its reactivation in immunosuppressed patients, which befits results from previous studies that showed persistent shedding of RV of the same genotype in children and immunocompromised adults over periods as long as 100 days [23][24][25][26]. The study's cross-sectional design, with single-time tissue sampling, limited confirmation of persistent RV infection. Long-term persistence would require repeated sampling from the same subjects, which is not feasible.
Based on current evidence, the use of RV detection in secretions by RT-PCR as confirmation of etiology of current respiratory symptoms should be regarded with caution. Early studies based only on virus isolation in cell culture reported infectious RV shedding from asymptomatic subjects ranging from 0.7% to 1.6% in adults and 4.7% in children [27,28]. Considering only results of RV detection by RT-PCR in secretions, which is the most readily available routine sample in clinical settings, the rate of positivity in children with tonsillar hypertrophy was close to 12%, which is similar to the RV detection rates in asymptomatic children reported in previous studies in the same city [29]. It is reasonable to consider that RV may stay in tonsillar tissues for prolonged times after an acute infection, which is in agreement with previous data on RV RNA detection by RT-PCR [30,31]. Yet, RT-PCR detection does not necessarily indicate virus replication at the time of sampling.
The clinical relevance of human rhinovirus (RV) in respiratory diseases extends well beyond the common cold [32]. RV infection has been linked to exacerbations and increased morbidity in chronic respiratory diseases such as COPD and asthma, with certain genotypes-particularly RV-A and RV-C-associated with more severe outcomes [33]. Co-infections, including with SARS-CoV-2, may further worsen asthma [34], highlighting speciesspecific differences in RV pathogenesis [34,35]. These findings align with our study, which not only identified all three RV species (A, B, and C) in tonsillar tissues from children without acute symptoms, but also demonstrated active infection and replication in CD4⁺ T and B lymphocytes, accompanied by the release of proinflammatory cytokines such as IL-17, TNF-α, and IFN-γ.
Although the cross-sectional design of our study does not permit to establish viral persistence, the presence of RV in immune cells from asymptomatic individuals suggests that tonsillar lymphoid tissues may transiently harbor the virus, what may contribute to the dynamics of RV infection and transmission. In summary, tonsillar lymphoid tissues may be sites of asymptomatic RV replication, and it is conceivable that the consequent immune activation might interfere in recurrent airway inflammation associated with chronic upper airway diseases and allergy. and the surface markers CD8 and CD14. Supporting Figure 6: Ex vivo infection of tonsillar immune cells with rhinovirus (RV) and analysis of infected cell subsets by immunofluorescence. Supporting Table 1. Supporting Table 2. Supporting Table 3.
## References
1. Lee, Lemanske, Evans (2012) "Human Rhinovirus Species and Season of Infection Determine Illness Severity" *American Journal of Respiratory and Critical Care Medicine*
2. Gern (2010) "The ABCs of Rhinoviruses, Wheezing, and Asthma" *Journal of Virology*
3. Proenca-Modena, Pereira Valera, Jacob (2012) "High Rates of Detection of Respiratory Viruses in Tonsillar Tissues From Children With Chronic Adenotonsillar Disease" *PLoS One*
4. Jartti, Palomares, Waris (2014) "Distinct Regulation of Tonsillar Immune Response in Virus Infection" *Allergy*
5. Faden, Callanan, Pizzuto (2016) "The Ubiquity of Asymptomatic Respiratory Viral Infections in the Tonsils and Adenoids of Children and Their Impact on Airway Obstruction" *International Journal of Pediatric Otorhinolaryngology*
6. Martins, Rocha, Prates (2017) "Respiratory DNA Viruses are Undetectable in Nasopharyngeal Secretions From Adenotonsillectomized Children" *PLoS One*
7. Castro, Jorge, Ferreri (2020) "Silent Infection of B and CD8+ T Lymphocytes by Influenza A Virus in Children With Tonsillar Hypertrophy" *Journal of Virology*
8. Glass, Papin, Mandell (2009) "SIMPLE: A Sequential Immunoperoxidase Labeling and Erasing Method" *Journal of Histochemistry and Cytochemistry*
9. Jørgensen, Baker, Møller et al. (2010) "Robust One-Day in Situ Hybridization Protocol for Detection of Micrornas in Paraffin Samples Using LNA Probes" *Methods*
10. Arruda, Crump, Rollins et al. (1996) "Comparative Susceptibilities of Human Embryonic Fibroblasts and HeLa Cells for Isolation of Human Rhinoviruses" *Journal of Clinical Microbiology*
11. De Arruda, Mifflin, Gwaltney et al. (1991) "Localization of Rhinovirus Replication In Vitro With In Situ Hybridization" *Journal of Medical Virology*
12. Arruda, Boyle, Winther (1995) "Localization of Rhinovirus Replication in the Respiratory Tract by In Situ Hybridization" *Journal of Infectious Diseases*
13. Mikola, Palomares, Turunen (2019) "Rhinovirus Species and Tonsillar Immune Responses" *Clinical and Translational Allergy*
14. Suvilehto, Roivainen, Seppänen (2006) "Rhinovirus/Enterovirus RNA in Tonsillar Tissue of Children With Tonsillar Disease" *Journal of Clinical Virology*
15. Nave, Gebert, Pabst (2001) "Morphology and Immunology of the Human Palatine Tonsil" *Anatomy and Embryology*
16. Van Kempen, Rijkers, Van Cauwenberge (2000) "The Immune Response in Adenoids and Tonsils" *International Archives of Allergy and Immunology*
17. Gwaltney (1985) "The Jeremiah Metzger Lecture: Climatology and the Common Cold" *Transactions of the American Clinical and Climatological Association*
18. Gwaltney (1983) "Rhinovirus Colds: Epidemiology, Clinical Characteristics and Transmission," supplement" *European Journal of Respiratory Diseases Supplement*
19. Schuler, Schreiber, Li (2014) "Major and Minor Group Rhinoviruses Elicit Differential Signaling and Cytokine Responses as a Function of Receptor-Mediated Signal Transduction" *PLoS One*
20. Aab, Wirz, Van De Veen (2017) "Human Rhinoviruses Enter and Induce Proliferation of B Lymphocytes" *Allergy*
21. Ilarraza, Wu, Skappak et al. (2013) "Rhinovirus has the Unique Ability to Directly Activate Human T Cells In Vitro" *Journal of Allergy and Clinical Immunology*
22. Bhattacharjee, Choi, Gozal et al. (2014) "Association of Adenotonsillectomy With Asthma Outcomes in Children: A Longitudinal Database Analysis" *PLoS Medicine*
23. Kaiser, Aubert, Pache (2006) "Chronic Rhinoviral Infection in Lung Transplant Recipients" *American Journal of Respiratory and Critical Care Medicine*
24. Ambrosioni, Bridevaux, Aubert et al. (2015) "Role of Rhinovirus Load in the Upper Respiratory Tract and Severity of Symptoms in Lung Transplant Recipients" *Journal of Clinical Virology*
25. Piralla, Zecca, Comoli et al. (2015) "Persistent Rhinovirus Infection in Pediatric Hematopoietic Stem Cell Transplant Recipients With Impaired Cellular Immunity" *Journal of Clinical Virology*
26. Engelmann, Dewilde, Lazrek (2017) "In Vivo Persistence of Human Rhinoviruses in Immunosuppressed Patients" *PLoS One*
27. Gwaltney, Hendley, Simon et al. (1966) "Rhinovirus Infections in an Industrial Population: The Occurrence of Illness" *New England Journal of Medicine*
28. Hendley, Gwaltney, Jordan (1968) "Rhinovirus Infections in an Industrial Population. IV. Infections Within Families of Employees During Two Fall Peaks of Respiratory Illness" *American Journal of Epidemiology*
29. Camara, Silva, Ferriani (2004) "Risk Factors for Wheezing in a Subtropical Environment" *Journal of Allergy and Clinical Immunology*
30. Jartti, Lehtinen, Vuorinen et al. (2004) "Persistence of Rhinovirus and Enterovirus RNA After Acute Respiratory Illness in Children" *Journal of Medical Virology*
31. Kling, Donninger, Williams (2005) "Persistence of Rhinovirus RNA After Asthma Exacerbation in Children" *Clinical and Experimental Allergy*
32. Alsayed, Abed, Al Shawabkeh et al. (2024) "Human Rhinovirus: Molecular and Clinical Overview" *Pharm Pract (Granada)*
33. Alsayed, Abed, Khader et al. (2024) "The Role of Human Rhinovirus in Copd Exacerbations in Abu Dhabi: Molecular Epidemiology and Clinical Significance" *Libyan Journal of Medicine*
34. Al-Dulaimi, Alsayed, Maqbali et al. (2022) "Investigating the Human Rhinovirus Co-Infection in Patients With Asthma Exacerbations and COVID-19" *Pharmacy Practice*
35. Alsayed, Abed, Abu-Samak et al. (2023) "Etiologies of Acute Bronchiolitis in Children at Risk for Asthma, With Emphasis on the Human Rhinovirus Genotyping Protocol" *Journal of Clinical Medicine* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772236&blobtype=pdf | # Serological evaluation of coronavirus IgA and IgG antibodies in a repeated cross-sectional cohort of unvaccinated and vaccinated pregnant individuals over three months following SARS-CoV-2 infection
Guadalein Tanunliong, Ana Citlali Márquez, Hind Sbihi, Tamara Pidduck, Yin Chang, Fang Li, Danielle Luk, Lucia Forward, Elisabeth Mcclymont, Chelsea Elwood, Mel Krajden, Agatha Jassem, Deborah Money, Inna Sekirov
## Abstract
Antibody surveillance provided valuable public health insights during the Coronavirus Disease 2019 (COVID-19) pandemic. Severe Acute Respiratory Syn drome Coronavirus 2 (SARS-CoV-2) infection, vaccination, and cross-reactive responses from endemic coronaviruses (human coronaviruses [HCoVs]) can influence SARS-CoV-2 antibody responses, impacting reliability and interpretation of serological findings. Here, we investigated population-level SARS-CoV-2 and HCoV IgA and IgG responses following SARS-CoV-2 infection among unvaccinated and vaccinated pregnant individuals over 3 months. Using residual sera from routine antenatal screening of pregnant individuals in British Columbia between March 2020 to May 2022, we designed a retrospective repeated cross-sectional cohort of infected individuals either unvaccinated (UV + COV, N = 171) or two-to three-dose vaccinated (V + COV, N = 137). A total of 30 pre-pandemic sera served as negative controls. Sera were collected within 3 months of respiratory PCR-positivity in half-month intervals and tested for IgA and IgG against Spike (S) of alpha-HCoV (HCoV-229E, HCoV-NL63) and beta-HCoV (HCoV-HKU1, HCoV-OC43) and S, receptor binding domain, and nucleocapsid (N) of SARS-CoV-2 using a multiplex immunoassay. Following SARS-CoV-2 infection, V + COV had lower anti-N IgG levels (P = 0.004) and seropositivity rates than UV + COV. One month post-infection, V + COV (38%) had lower anti-N IgA seropositivity than UV + COV (73%). Both groups had significantly higher anti-S IgA and IgG levels against beta-HCoV vs controls, with signals correlating positively with SARS-CoV-2 anti-S levels for each isotype. These results suggest that neither IgA nor IgG can reliably identify recent infections in vaccinated populations, emphasizing the importance of considering complex interplay of antibody responses when interpreting serological data and recognizing the potential and limitations of serological testing for diagnostics and surveillance.
IMPORTANCEThis study provides key insights into how Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) vaccination and infection shape the magnitude and longevity of IgA and IgG antibody responses following infection in pregnant individuals. It also highlights the interplay of serological responses to related viru ses, such as the human coronaviruses. By leveraging population-level antenatal sera, our findings highlight important considerations for the design and interpretation of future seroprevalence studies, antibody-based surveillance, and diagnostic strategies. As SARS-CoV-2 transitions into endemic circulation, understanding the complexity of SARS-CoV-2 antibody responses provides additional insights into the strengths and limitations of serological data interpretation in a real-world setting.
KEYWORDS vaccines, COVID-19, endemic coronavirus, pregnancy, serology, SARS-CoV-2, antibodies S urveillance of antibody responses (serosurveillance) generated from Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) vaccination and infection provided valuable public health information during the Coronavirus Disease 2019 (COVID- 19) pandemic (1). However, antibody responses across individuals exhibit significant heterogeneity due to varying infection, vaccination, and exposure histories, impacting the nature and breadth of serological findings. Understanding the antibody responses to SARS-CoV-2 and factors that influence these responses is crucial for accurate serological interpretations, which can help guide public health and vaccine policies (2).
Most SARS-CoV-2 serosurveillance studies rely on the detection of circulating immunoglobulin G (IgG) antibodies against the immunodominant regions of SARS-CoV-2, including the Spike (S) protein, the receptor binding domain (RBD) within the S region, and the nucleocapsid (N) protein, for estimates of infection and vaccination in the population (3,4). Vaccinations in Canada have been S-based; thus, vaccinated individuals are expected to elicit antibody responses against S and RBD, permitting the detection of anti-N IgG as a marker for infection-induced antibody responses (3, 4). However, while SARS-CoV-2 infection following S-based vaccination typically results in higher antibody concentrations against the S protein compared to unvaccinated individuals, the opposite effect was observed for anti-N antibodies, with lower anti-N IgG seroconversion rates observed following SARS-CoV-2 infection among previously vaccinated individuals compared to those unvaccinated (5,6). This prompts questions about the possible underestimation of infected individuals in vaccinated populations in serosurveillance studies (7), which must be further investigated, given that seroprevalence estimates obtained through these studies are used to guide vaccine booster policies.
The SARS-CoV-2 antibody response can also be influenced by cross-reactive antibodies to the highly genetically related endemic human coronaviruses (HCoVs) (8), including alpha-HCoV (HCoV-229E and HCoV-NL63) and beta-HCoV (HCoV-HKU1 and HCoV-OC43), the latter of which exhibits higher sequence similarities with SARS-CoV-2 (50%-51% sequence identity) (9). These four seasonal HCoVs have long been recognized to cause a substantial proportion of common colds (10,11), with nearly everyone expected to develop and periodically boost immune memory to HCoVs through repeated re-exposures throughout their lifetime (11,12). Cross-reactivity and back boosting of HCoV IgG antibodies have been previously demonstrated follow ing SARS-CoV-2 infection, though their roles in either protection or exacerbation of infection remain inconsistent (13)(14)(15)(16)(17)(18). Additionally, their potential influence in SARS-CoV-2 serology, whether serving as a supplemental serological marker or potentially hindering accurate detection of seroconversion due to the masking of a true SARS-CoV-2 response, needs to be further studied.
SARS-CoV-2's transition into endemicity in a highly immunized population has overall brought to light several shortcomings of IgG as a serological marker of infection, including their inability to identify recent infections due to IgG's persistence in circula tion, potential underestimations of infection due to blunted anti-N responses following vaccinations, and the potential influence of cross-reactive HCoV responses. As such, there has been growing interest in expanding beyond IgG and using IgA as a potential marker to identify recent SARS-CoV-2 infection (19)(20)(21). While IgA is more commonly known in its dimeric and polymeric forms as a crucial player of mucosal defense, with mucosal IgA often associated with protection from viral replication and transmission at the local site of infection (22), systemic IgA can also exist in monomeric form in serum with a generally shorter half-life than IgG and has been variably associated with both disease severity and protection (23,24). Both SARS-CoV-2 vaccination and infection generate IgA responses against S, RBD, and N (25,26); however, the serum IgA dynamics over time following infection of previously vaccinated individuals is still unclear, and the cross-reactivity with HCoV at the IgA level, as well as serum IgA's potential utility as a marker of recent infection, remains poorly understood.
Herein, we report on investigations of population-level IgA antibody dynamics to SARS-CoV-2 and HCoV, with respect to IgG, in a repeated cross-sectional cohort of vaccinated and unvaccinated pregnant individuals up to three months following SARS-CoV-2 infection. This work leveraged residual sera collected in British Columbia (BC), Canada, as part of routine antenatal communicable disease screening. Greater than 95% of all pregnant individuals in BC undergo antenatal screening during the first trimester, capturing diverse demographic and risk groups (27). This provides a relatively generalizable model to study SARS-CoV-2 antibody dynamics at a population level, given that trends of antibody waning remain consistent across populations. We hypothe sized that IgA and IgG antibody responses to SARS-CoV-2 and HCoV vary according to the timing of serum collection following PCR-confirmed SARS-CoV-2 infection and the individual's SARS-CoV-2 vaccination status pre-infection. As SARS-CoV-2 transitions into endemic circulation, our findings provide additional insights into the utility and limitations of serological testing for diagnostics and public health surveillance, which could help guide public health and vaccine policies.
## MATERIALS AND METHODS
## Study population
We used residual serum specimens collected during first trimester antenatal testing of pregnant individuals in BC, Canada, to design and implement a retrospective repea ted cross-sectional study to investigate SARS-CoV-2 antibody responses to infection, vaccination, and previous HCoV antibodies. A schematic of our retrospective selection strategy is presented in Fig. S1. Sera were selected using linkages with BC's integrated COVID-19 PCR testing data and COVID-19 vaccine registry to determine SARS-CoV-2 infection and vaccination histories and included individuals residing in all BC health regions, with the two most densely populated health regions notably over-represented in the selected cohort (Table 1). This study was approved by the Clinical Research Ethics Board of the University of British Columbia with a waiver of informed consent (H21-03566).
A total of 338 serum specimens from 338 participants were included. Of these, 30/338 were collected prior to the index case in BC (28 January 2020) and served as pre-pan demic negative controls. A total of 308/338 sera were from individuals with a history of PCR-confirmed SARS-CoV-2 infection collected between March 2020 and May 2022. Of note, the provincial PCR testing strategy prior to December 2021 was to offer broad and accessible PCR testing to all BC residents, with a special emphasis on testing pregnant individuals. From December 2021 onward, PCR testing was only indicated for those requiring hospitalization or hospitalized, whereas rapid antigen test kits were largely distributed for community-based self-testing.
Among the 308 infected individuals, 44% (137/308) of the SARS-CoV-2 infected individuals were vaccinated prior to infection (V + COV group) with either two (2V + COV group, N = 75) or three (3V + COV group, N = 62) vaccine doses. The remain ing 56% (171/308) of SARS-CoV-2 infected individuals were unimmunized, having no documented vaccine against SARS-CoV-2 on file prior to infection (UV + COV group). All vaccinations were Health Canada-approved at the time of administration (Vaxzevria, COVISHIELD, Spikevax, or COMIRNATY) and only expected to elicit an antibody response against the S and RBD proteins. All 308 sera were further grouped according to the timeframe between their SARS-CoV-2-positive PCR result and serum collection date at approximately half-month intervals, beginning at 0.5 months and ending at 3 months (Table S1). Individuals who had a documented positive SARS-CoV-2 PCR test prior to their vaccinations were excluded from the V + COV group of the study. See Table 1 for a detailed breakdown of demographic characteristics of all included participants.
## Detection of IgA and IgG antibody against ancestral SARS-CoV-2 and HCoV proteins using a multiplex electrochemiluminescent immunoassay
All residual sera (N = 338) were tested for antibodies against S, N, and S1 RBD of ancestral SARS-CoV-2, and the S proteins of four HCoVs (Alpha-HCoVs: HCoV-229E and HCoV-NL63, and Beta-HCoVs: HCoV-HKU1 and HCoV-OC43) using the quantitative multiplex V-PLEX Coronavirus Panel 2 Assay from Meso Scale Diagnostics (MSD, Rockville, USA), which was previously validated in-house for both IgG and IgA (Fig. S2). Assays were performed according to the manufacturer's protocol as described previously (28,29), briefly outlined below.
Ten-spot assay plates from MSD with antigens against S, RBD, and N regions of SARS-CoV-2, and S proteins of HCoV-229E, HCoV-NL63, HCoV-HKU1, and HCoV-OC43 a Health authority refers to the specific regional organizations within the province responsible for managing and delivering public health services. UV + COV = unvaccinated prior to infection with SARS-CoV-2. V + COV = vaccinated with two or three doses prior to infection with SARS-CoV-2. b P-value from Wilcoxon rank sum test. c P-value from Chi-square test. d NA, not applicable.
printed on each spot were used for antibody testing. Assay plates were first blocked with MSD Blocker A for 30 minutes and then washed. Reference standard, assay controls, and sera (diluted 1:5,000 in MSD Diluent 100) were added and incubated for 2 hours before washing. Plates were then incubated with either the SULFO-TAG anti-human IgG or IgA detection antibody for 1 hour. Lastly, MSD Gold Read Buffer B was added to the plate following a final wash, and signals were immediately measured on the MSD QuickPlex SQ120. All incubation steps were carried out shaking at 700 rpm at room temperature. All wash steps were performed three times with MSD wash buffer prior to the addition of subsequent reagents outlined above. Raw signals generated by the MSD QuickPlex SQ120 were processed using the MSD Discovery Workbench software (Version 4.0) and then imported into R (Version 3.6.2) and RStudio (Version 1.2.5033) to interpret signal cutoff values. IgG positivity threshold cutoffs for SARS-CoV-2 were provided by the manufacturer as follows: anti-SARS-CoV-2 S values above 1,960 AU/mL, anti-SARS-CoV-2 N values above 5,000 AU/mL, and anti-SARS-CoV-2 S1 RBD values above 538 AU/mL, and have been previously validated in our laboratory to demonstrate excellent sensitivity and specificity (28). Threshold cutoffs for IgA positivity for SARS-CoV-2 were previously developed and validated in-house using 30 pre-pandemic individuals and 51 acute COVID-19 patients and were as follows: anti-SARS-CoV-2 S values above 1,245 AU/mL, anti-SARS-CoV-2 N values above 1,250 AU/mL, and anti-SARS-CoV-2 S1 RBD values above 530 AU/mL (Fig. S2).
## Statistical analysis
All statistical analysis and data visualizations were conducted on R (Version 3.6.2) and RStudio (Version 1.2.5033). Processed data obtained from the MSD immunoassay was visualized using the ggplot2 (Version 3.3.3) and ggpubr (Version 0.6.0) packages on RStudio. Reactivity cutoffs for IgA positivity were determined by taking the value two standard deviations above the geometric mean antibody level of the pre-pandemic negative individuals for each of S, RBD, and N. Sensitivity, specificity, positive predictive values, and negative predictive values were calculated using the cutpointR (Version 1.1.1) and epiR (Version 2.0.41) packages.
Kruskal-Wallis, Dunn's post hoc multiple comparisons, chi-squared, Wilcoxon ranksum, and Spearman's correlation calculations were conducted using ggpubr, gtsummary (Version 1.5.1), and stats (Version 3.6.2) packages. Antibody trends were fitted using locally weighted regression (loess) between points on ggpubr. All antibody levels and reactivity cutoff values were log10 transformed for analysis.
## RESULTS
## Study population
Sera from 308 pregnant individuals were collected between March 2020 and May 2022. All 308 individuals had documented PCR-confirmed SARS-CoV-2 infections, classified as vaccinated (V + COV, N = 137) or unvaccinated (UV + COV, N = 171) for SARS-CoV-2 prior to infection, according to the provincial COVID-19 vaccine registry. Individuals were further grouped based on the timing of their serum collection relative to their PCR-con firmed SARS-CoV-2 infection date, in approximately half-month intervals. The median (interquartile range [IQR]) age between both groups does not differ significantly, at 33 years (30,35) for the UV + COV group and 32 years (28,36) for the V + COV group. The sampling windows were from March 2020 to November 2021 for UV + COV, and August 2021 to April 2022 for V + COV, with median (IQR) days from infection to serum collection at 58 days (32,77) for UV + COV and 43 days (28, 64) for V + COV. Given that only PCR-confirmed cases were included, not all infections in the province may have been captured in this data set (Dataset S1). Furthermore, serum from 30 additional individuals prior to the index case in BC was included and served as pre-pandemic negative controls.
## Population IgG responses against S, RBD, and N of SARS-CoV-2 over 3 months following SARS-CoV-2 infection in vaccinated and unvaccinated pregnant individuals
Following infection, V + COV individuals demonstrated significantly higher anti-S (P < 0.001) and anti-RBD IgG (P < 0.001) but lower anti-N IgG (P < 0.004), compared to UV + COV individuals (Fig. 1A through C). Anti-S and anti-RBD IgG remained consistently higher among V + COV across 3 months (Fig. S3), with stable IgG trends irrespective of two-or three-dose vaccinations (Fig. 1D andE; Fig. S4). IgG seropositivity rates remained high in V + COV serum over 3 months post-infection, ranging from 97% to 100% for both S and RBD (Table 2). Interestingly, among UV + COV individuals, IgG seropositivity rate was at 69%-100% for S and 77%-100% for RBD. Notably, the lower seropositivity rates at 69% for S and 77% for RBD among the UV + COV group were both observed during the 0.5-month time point. IgG seropositivity rates increased to 92%-100% for S and 96%-100% for RBD after 1 month post-infection (Table 2), consistent with the evolving seroconversion rates in the population following infection.
In contrast to anti-S and anti-RBD, we observed significantly lower anti-N IgG levels among V + COV compared to UV + COV individuals (P = 0.004) (Fig. 1C). When compared at each time point, V + COV individuals consistently exhibited lower median anti-N levels than UV + COV individuals, though these trends were not statistically significant (Fig. S3). Median antibody levels for anti-N remained stable over 3 months (Fig. 1F) for both UV + COV and V + COV, irrespective of two-or three-dose vaccinations (Fig. S4). Anti-N IgG seropositivity rates were lower at 60%-85% for V + COV, compared to 77%-91% for UV + COV (Table 2).
## Population IgA responses against S, RBD, and N of SARS-CoV-2 over 3 months following SARS-CoV-2 infection in vaccinated and unvaccinated pregnant individuals
We observed significantly higher anti-S (P < 0.001) and anti-RBD (P < 0.001) IgA (Fig. 2A andB), but not anti-N IgA (Fig. 2C), following SARS-CoV-2 infection among V + COV compared to UV + COV individuals (Fig. S5). Anti-S and anti-RBD IgA trends appeared stable over 3 months for the V + COV group (Fig. 2D andE), irrespective of two-or three-dose vaccinations (Fig. S4), but declined over time in the UV + COV group, as demonstrated by the loess curves (Fig. 2D andE). The V + COV group exhibited higher IgA seropositivity rates at 95%-100% for both S and RBD for each timing group, in contrast to the lower UV + COV IgA seropositivity rates at 44%-85% for S and 75%-100% for RBD (Table 3). Notably, anti-S IgA seropositivity rates for the UV + COV group for the 0.5 and 1-month post-infection groups were both at 85%, with decreases in the seropositivity rates seen as the timeframe increases, reaching 44% by 3 months post-infection (Table 3).
V + COV individuals exhibited lower anti-N IgA concentrations relative to UV + COV individuals, but this difference was not statistically significant (P = 0.848) (Fig. 2F; Fig. S5). Anti-N IgA seropositivity rates declined over 3 months for both UV + COV and V + COV (Table 3), though this decline occurred earlier for the V + COV group (73% at 0.5 months to below 38% at 1 month onward) compared to the UV + COV group (81% at 0.5 months, to 73% at 1 month, to below 37% at 1 month onward) (Table 3).
## Cross-reactive HCoV IgA and IgG responses following SARS-CoV-2 infection in vaccinated and unvaccinated pregnant individuals
When compared to pre-pandemic controls, both UV + COV and V + COV individuals exhibited significantly higher IgA levels against beta-HCoVs HCoV-HKU1 (P < 0.001) and HCoV-OC43 (P < 0.001), but not alpha-HCoVs (Fig. 3A through D). Similarly, for IgG, compared to pre-pandemic controls, anti-OC43 IgG levels were higher among UV + COV (P < 0.001) and V + COV (P = 0.008), and anti-HKU1 IgG levels were higher among the UV + COV group (P = 0.031), while no significant differences were observed for alpha-HCoV IgG (Fig. 3E through H). Notably, anti-NL63 IgA, but not IgG, was lower following SARS-CoV-2 infection for both UV + COV (P < 0.001) and V + COV (P < 0.001) relative to levels observed in sera collected pre-pandemic (Fig. 3B andF).
Correlation analysis demonstrates that compared to infected individuals, prepandemic controls exhibited stronger positive correlations between SARS-CoV-2 S IgA and beta-HCoV S IgA levels (HCoV-HKU1: Rho = 0.685, P < 0.001; HCoV-OC43: Rho = 0.525, P = 0.003), but not for alpha-HCoV (Fig. 4A through D), suggesting pre-existing cross-reactivity at the IgA level for beta-HCoV (Fig. 4C andD). Infected individuals exhibited mildly significantly positive correlations between SARS-CoV-2 S and beta-HCoV a UV + COV = unvaccinated prior to infection with SARS-CoV-2. V + COV = vaccinated with two or three doses prior to infection with SARS-CoV-2.
S levels for both IgA and IgG, but not alpha-HCoV (Fig. 4E through H). In contrast to IgA, pre-pandemic SARS-CoV-2 S IgG and beta-HCoV S IgG were not significantly correlated (Fig. 4G andH). Notably, SARS-CoV-2 S IgA was significantly positively correlated with anti-NL63 IgA (R = 0.258, P < 0.001) for the UV + COV group (Fig. 4B).
## DISCUSSION
The COVID-19 pandemic highlighted the value of serological testing for public health surveillance and guiding vaccine policy, underlining the importance of understanding antibody responses in SARS-CoV-2 infection. Here, we utilized residual sera collected from routine antenatal screening in BC to investigate the SARS-CoV-2 and HCoV IgA and a UV + COV = unvaccinated prior to infection with SARS-CoV-2. V + COV = vaccinated with two or three doses prior to infection with SARS-CoV-2.
IgG antibody responses in vaccinated and unvaccinated pregnant individuals over 3 months following SARS-CoV-2 infection. We observed that following infection, vaccinated individuals had higher and more stable levels of anti-S and anti-RBD IgG but significantly lower levels of anti-N IgG, compared to those unvaccinated. The lower levels of anti-N IgG and lower rates of anti-N seropositivity among vaccinated individuals were consistent with other studies showing anti-N blunting and lower seroconversion rates following SARS-CoV-2 infection of previously vaccinated individuals (5)(6)(7)30). When compared by timing groups, while vaccinated individuals consistently demonstrated lower anti-N IgG trends compared to those unvaccinated, this difference did not reach statistical significance, which could be attributed to the relatively small sample size. Nonetheless, this consistent trend of blunted anti-N IgG has been attributed to a rapid and stronger recall of anamnestic anti-S responses following infection of vaccinated individuals, leading to limited viral entry, limited viral replication, faster viral clearance, and an overall lower induction of anti-N antibodies (7,30). As such, anti-N IgG responses may be a less reliable serological indicator of past infection in vaccinated populations, and potentially in previously infected populations. Given that many serological assay validations were designed and evaluated in unvaccinated populations prior to the deployment of vaccines, it is imperative that assay performance be assessed in a real-world setting, particularly in vaccinated populations, to prevent underestimation of infected individuals in seropreva lence studies. With the rapid emergence of variants that continue to change the SARS-CoV-2 immune landscape, infections and vaccinations may induce differentially boosted binding antibodies and neutralization responses across variants (31,32). This highlights the importance of ongoing investigations of both the SARS-CoV-2 serological landscape and the performance of serological assays to ensure accurate serological interpretations for diagnostics and surveillance. Furthermore, it may also be necessary to adjust assay cutoffs for serological positivity given vaccination status and/or undertake repetitive sampling to infer seroprevalence more reliably over shorter time frames, rather than consider that seroprevalence estimates hold over long periods of time.
IgG has been commonly used as a serological marker to identify past infection and/or vaccination due to their systemic persistence, which can be attributed to both the ongoing systemic IgG production by memory B cells and IgG's own recycling mechanism (33). Specifically, IgG's interactions with the neonatal fragmental crystallizable receptor (FcRn) promote its protection from intracellular lysosomal degradation (33), recycling IgG back into circulation, and supporting its serological use for identification of past infection and/or vaccination. In contrast, IgA's shorter half-life can be attributed in part to its lack of protection from degradation due to a lack of this FcRn region (34,35), suggesting that high levels of circulating IgA may potentially be correlated with recent infections and/or vaccination. In our study, we found that unvaccinated individuals demonstrated primary IgA responses to S, RBD, and N that declined below seropositivity cutoffs within 3 months post-infection, suggesting that higher population IgA levels may indeed be better correlated with recent infection among unvaccinated populations. However, given that vaccinated individuals exhibited stable levels of anti-S and anti-RBD IgA over 3 months following infection, IgA's potential use in identifying recent infections may not be applicable in highly vaccinated settings. While IgA's short-lived nature had consistently gathered interest for its potential utility in identifying recently infected individuals, given the large heterogeneity in population seropositivity rates and the stable IgA recall responses observed among vaccinated individuals, individual IgA antibody levels are not likely to be a reliable marker of recent infections, especially within a recently vaccinated population. SARS-CoV-2's transition into endemicity prompts the need for an improved under standing of its serological response dynamics and its interactions with the immune response to other HCoVs to better understand non-specific and cross-reactive responses across coronaviruses. We observed a weakly positive correlation between beta-HCoV anti-S IgG and SARS-CoV-2 anti-S IgG following infection, irrespective of prior vacci nation status. This correlation was absent in pre-pandemic individuals, suggesting either antibody cross-reactivity or a boosting of specific beta-HCoV IgG antibodies upon SARS-CoV-2 infection. Our findings were consistent with other studies that have described an induction of beta-HCoV memory B cells upon SARS-CoV-2 infection (17,36,37), specifically against the conserved S2 region within the S protein (8,38), suggesting induction of beta-HCoV memory responses beyond antibody cross-reactivity in circulation. Furthermore, several studies have evaluated the association between cross-reactive beta-HCoV responses with COVID-19 severity (8,15,39), though their role in protection from severe COVID-19 remains uncertain. Similar to IgG, we observed that infected individuals had higher beta-HCoV IgA compared to pre-pandemic individuals, and these beta-HCoV IgA were positively correlated with SARS-CoV-2 anti-S IgA, though weaker than the correlations observed in pre-pandemic control individuals. This could indicate higher baseline cross-reactivity for beta-HCoV IgA, relative to IgG, and the decrease in IgA cross-reactivity following infection could be due to a relatively higher proportion of SARS-CoV-2-specific S1 (rather than the conserved S2) IgA responses post-SARS-CoV-2 infection. Of note, we also observed lower levels of the alpha-HCoV anti-NL63 IgA, but not IgG, among SARS-CoV-2-infected individuals when compared to pre-pandemic sera. While unexpected, the decrease in IgA, but not IgG, may be due to lower overall circulation of respiratory viruses in BC when COVID-19 pandemic measures had been in place, and the decline in IgA, but not IgG, may reflect the more rapid waning kinetics of IgA relative to IgG (40). To date, we are one of the few studies to demonstrate increased HCoV IgA responses following acute infection in this specific population (41,42). Future studies should consider if this reflects cross-reactive induction of IgA by conserved epitopes or boosting of HCoV-specific memory responses and whether this confers cross-protective effects against infection and disease. While we observed an increase in beta-HCoV IgA and IgG following SARS-CoV-2 infection, similar to findings in early pandemic studies, we are unable to discern whether this is due to cross-reactivity or specific boosting of beta-HCoV responses. Several studies have suggested an association of elevated beta-HCoV responses with either disease severity or protection from infections and severe disease. While Guo et al. ( 43) observed an association with disease severity where higher HCoV-OC43 S IgG levels were associated with patients with more severe disease, Ortega et al. ( 44) demonstrated an association with protection, where higher trends of anti-HCoV N IgA and IgG were observed among asymptomatic compared to symptomatic participants following SARS-CoV-2 infection. In contrast, Anderson et al. (45) demonstrated that while beta-HCoV antibodies were boosted following infection with SARS-CoV-2, these were not associated with disease severity or protection. The heterogeneity in these findings could be partially attributed to the variability in methodologies, such as testing of antibodies against different HCoV targets, the use of various assays of distinct sensitivities and specificities (46), and other variations in population characteristics and exposure histories. Further studies are needed to investigate these cross-reactive HCoV responses and their influences on the SARS-CoV-2 antibody responses and COVID-19 outcomes.
Our study has several limitations, including the small sample size and the lack of paired longitudinal samples from the same individual over time, hindering antibody level comparisons at an individual level. Nonetheless, our use of repeated cross-sectional sampling produces estimates of aggregate population-level changes over time by use of a relatively unbiased representative sample at every time point, improving its external validity. Additionally, we acknowledge that there was a substantial underestimation of infectious case-based surveillance reporting described during the Omicron period in 2022 (4,27), as the broadly accessible PCR testing strategy in BC had changed from December 2021 onward to being indicated only for those hospitalized; thus, by using PCR as a criterion for infection, we likely missed asymptomatic or mildly symptomatic infections and individuals who chose not to seek testing during this time. Nevertheless, our study did not intend to estimate BC's overall infection prevalence and vaccine coverage; but rather, we aimed to design a repeated cross-sectional study to specifically compare the changes in population-level antibody responses to SARS-CoV-2 infection over time among previously vaccinated vs unvaccinated individuals.
We also acknowledge that another limitation of our study is the wide sampling timeframe, with the V + COV group mostly sampled at a later date than UV + COV. As our assay was designed for ancestral SARS-CoV-2, antibody responses to later variants may exhibit reduced binding affinity and therefore be underestimated; however, there remains substantial cross-reactivity in antibody responses across conserved epitopes, including N; thus, this is unlikely to have a substantial impact on our conclusions. Furthermore, despite the timeframe to serum collection being statistically significantly different between the UV + COV and V + COV groups at a median (IQR) of 58 days (32, 77) and 43 days (28,64), respectively, the large overlap in spread between both groups suggests that this statistical difference is unlikely to translate into biological differences. Specifically, antibody levels have been shown to peak and plateau within the sampling timeframe assessed in this study; thus, this difference is unlikely to bias our findings. In addition, differences in timing between previous vaccination prior to subsequent infections could contribute to some variability in measured antibody levels among vaccinated individuals, reflecting real-world population heterogeneity.
Finally, our study uses a pregnant population, which includes individuals of child bearing age. Clinical and demographic characteristics such as smoking status, disease severity, underlying comorbidities, and co-medications could not be collected for this cohort within the parameters of the study design and therefore could not be accounted for in the analysis. Nonetheless, while our findings are not directly generalizable to other age groups, such as children, older adults, males, and more vulnerable populations, antibody waning occurs in all these populations, so the trends are relatively generaliza ble as well. Additionally, in BC, prenatal screening has an exceptional uptake of >95% and represents individuals from a wide range of demographic characteristics, geographic jurisdictions, and socioeconomic and cultural backgrounds (27). It is also a cohort of generally healthy individuals, in contrast to many other types of residual sera collected clinically that might be biased toward individuals with underlying health conditions. Thus, our cohort presents a unique and less biased sample set relative to hospitalized patients for seroprevalence studies utilizing residual clinical specimens (27).
A major study strength lies in the use of a highly sensitive assay for simultaneous detection of SARS-CoV-2 and HCoV antibodies (28), minimizing biases that might arise due to variable specimen handling during the testing process. Another strength is the use of the provincial public health laboratory SARS-CoV-2 PCR testing database and the provincial COVID-19 vaccine registry to classify individuals according to SARS-CoV-2 PCR testing results and vaccination status, minimizing misclassification biases that could arise from self-reported results.
Overall, we found differences in IgA and IgG antibody levels against S, RBD, and N following SARS-CoV-2 infection, which was associated with both vaccination history and the timing of serum collection relative to infection. We also observed higher beta-HCoV HKU1 and OC43 IgA and IgG levels following confirmed SARS-CoV-2 infection relative to pre-pandemic levels. Our findings highlight the importance of considering these factors and the continuous re-evaluations of the coronavirus serological landscape to enable more reliable serological interpretations and guide clinical, public health, and vaccine policies.
## References
1. O'brien, Asamoah-Boaheng, Grunau et al. (2024) "Canada's approach to SARS-CoV-2 sero-surveillance: lessons learned for routine surveillance and future pandemics" *Can J Public Health*
2. Haselbeck, Im, Prifti et al. (2022) "Serology as a tool to assess infectious disease landscapes and guide public health policy" *Pathogens*
3. Van Den Hoogen, Smits, Van Hagen et al. "Rots NY, van der Klis FRM, van Binnendijk RS, den Hartog G. 2022. Seropositivity to Nucleoprotein to detect mild and asymptomatic SARS-CoV-2 infections: a complementary tool to detect breakthrough infections after COVID-19 vaccination?" *Vaccine (Auckl)*
4. Skowronski, Kaweski, Irvine et al. (2022) "Serial cross-sectional estimation of vaccine-and infection-induced SARS-CoV-2 seroprevalence in British Columbia" *Canada*
5. Allen, Brady, Martin et al. (2021) "Serological markers of SARS-CoV-2 infection; anti-nucleocapsid antibody positivity may not be the ideal marker of natural infection in vaccinated individuals" *J Infect*
6. Follmann, Janes, Buhule et al. (2022) "Antinucleocapsid antibodies after SARS-CoV-2 infection in the blinded phase of the randomized, placebo-controlled mRNA-1273 COVID-19 vaccine efficacy clinical trial" *Ann Intern Med*
7. Kim, Cha, Kwon et al. (2023) "The difference in anti-nucleocapsid protein antibody responses between vaccinated and unvaccinated individuals after asymptomatic, mild, or moderate COVID-19 infection" *Korean J healthc assoc Infect Control Prev*
8. Aydillo, Rombauts, Stadlbauer et al. (2021) "Immunological imprinting of the antibody response in COVID-19 patients" *Nat Commun*
9. Cicaloni, Costanti, Pasqui et al. (2022) "A bioinformatics approach to investigate structural and non-structural proteins in human coronaviruses" *Front Genet*
10. Fung, Liu (2021) "Similarities and dissimilarities of COVID-19 and other coronavirus diseases" *Annu Rev Microbiol*
11. Huang, Garcia-Carreras, Hitchings et al. (2020) "A systematic review of antibody mediated immunity to coronaviruses: kinetics, correlates of protection, and association with severity" *Nat Commun*
12. Tanunliong, Liu, Kaweski et al. (2022) "Age-associated seroprevalence of coronavirus antibodies: populationbased serosurveys in 2013 and 2020" *Canada. Front Immunol*
13. Tso, Lidenge, Peña et al. (2021) "High prevalence of preexisting serological cross-reactivity against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in sub-Saharan Africa" *Int J Infect Dis*
14. Shrwani, Sharma, Krishnan et al. (2021) "Detection of serum cross-reactive antibodies and memory response to SARS-CoV-2 in prepandemic and Post-COVID-19 convalescent samples" *J Infect Dis*
15. Sagar, Reifler, Rossi et al. (2021) "Recent endemic coronavirus infection is associated with less-severe COVID-19" *J Clin Invest*
16. Ratcliffe, Tiley, Longet et al. (2023) "Serum HCoV-spike specific antibodies do not protect against subsequent SARS-CoV-2 infection in children and adolescents"
17. Aguilar-Bretones, Westerhuis, Raadsen et al. (2021) "Seasonal coronavi rus-specific B cells with limited SARS-CoV-2 cross-reactivity dominate the IgG response in severe COVID-19" *J Clin Invest*
18. Röltgen, Nielsen, Silva et al. (2022) "Immune imprinting, breadth of variant recognition, and germinal center response in human SARS-CoV-2 infection and vaccination" *Cell*
19. Esmat, Jamil, Kheder et al. (2024) "Immunoglobulin A response to SARS-CoV-2 infection and immunity"
20. Rangel-Ramírez, Macías-Piña, Servin-Garrido et al. (2022) "A systematic review and meta-analysis of the IgA seroprevalence in COVID-19 patients: Is there a role for IgA in COVID-19 diagnosis or severity?" *Microbiol Res*
21. Guo, Ren, Yang et al. (2020) "Profiling early humoral response to diagnose novel coronavirus disease (COVID-19)" *Clin Infect Dis*
22. Marking, Bladh, Aguilera et al. (2025) "Impact of systemic SARS-CoV-2 vaccination on mucosal IgA responses to subsequent breakthrough infection" *EBioMedicine*
23. Zervou, Louie, Stachel et al. (2021) "SARS-CoV-2 antibodies: IgA correlates with severity of disease in early COVID-19 infection" *J Med Virol*
24. Hennings, Thörn, Albinsson et al. (2022) "The presence of serum anti-SARS-CoV-2 IgA appears to protect primary health care workers from COVID-19" *Eur J Immunol*
25. Wisnewski, Luna, Redlich (2021) "Human IgG and IgA responses to COVID-19 mRNA vaccines" *PLoS One*
26. Zurac, Nichita, Mateescu et al. (2021) "COVID-19 vaccination and IgG and IgA antibody dynamics in healthcare workers" *Mol Med Rep*
27. Mcclymont, Elwood, Sekirov et al. (2021) "Population SARS-CoV-2 seroprevalence using antenatal serum samples in British Columbia" *Canada. J Obstet Gynaecol Can*
28. Li, Liu, Gibbs et al. (2022) "A novel multiplex electrochemiluminescent immunoassay for detection and quantification of anti-SARS-CoV-2 IgG and anti-seasonal endemic human coronavirus IgG" *J Clin Virol*
29. Sekirov, Barakauskas, Simons et al. (2021) "SARS-CoV-2 serology: validation of high-throughput chemiluminescent immunoassay (CLIA) platforms and a field study in British Columbia" *J Clin Virol*
30. Dhakal, Yu, Yin et al. (2023) "Reconsideration of Antinucleocapsid IgG antibody as a marker of SARS-CoV-2 infection postvaccination for mild COVID-19 patients" *Open Forum Infect Dis*
31. Jiang, Song, Shi et al. (2024) "Variant-specific antibody response following repeated SARS-CoV-2 vaccination and infection" *Cell Rep*
32. Woelfel, Dütschler, Junker et al. "STAR SIGN Study Investigators. 2025. XBB.1.5-adapted COVID-19 mRNA vaccines but not infections with previous omicron variants boost neutralisation against the SARS-CoV-2 JN.1 variant in patients with inflammatory bowel disease" *Aliment Pharmacol Ther*
33. Pyzik, Rath, Lencer et al. (2015) "FcRn: the architect behind the immune and nonimmune functions of IgG and albumin" *J Immunol*
34. De Sousa-Pereira, Woof, Van Tetering et al. (2019) "Fc engineering strategies to advance IgA antibodies as therapeutic agents" *Antibodies (Basel)*
35. Nguyen-Contant, Embong, Kanagaiah et al. (2020) "S protein-reactive IgG and memory B cell production after human SARS-CoV-2 infection includes broad reactivity to the S2 subunit" *mBio*
36. Song, He, Callaghan et al. (2021) "Cross-reactive serum and memory B-cell responses to spike protein in SARS-CoV-2 and endemic coronavirus infection" *Nat Commun*
37. Hederman, Natarajan, Wiener et al. (2022) "SARS-CoV-2 mRNA vaccination elicits broad and potent Fc effector functions to VOCs in vulnerable popula tions"
38. Fillmore, Szalat, La et al. (2022) "Recent common human coronavirus infection protects against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection: a veterans affairs cohort study" *Proc Natl Acad Sci*
39. Canada, Ca (2022) "Respiratory Virus Detections/Isolations in Canada 2021 -2022"
40. Escalera, Rojo-Fernandez, Rombauts et al. (2024) "SARS-CoV-2 infection induces robust mucosal antibody responses in the upper respiratory tract"
41. Wang, Young, Li et al. (2022) "Broad cross-reactive IgA and IgG against human coronaviruses in milk induced by COVID-19 vaccination and infection" *Vaccines (Basel)*
42. Guo, Wang, Kang et al. (2021) "Cross-reactive antibody against human coronavirus OC43 spike protein correlates with disease severity in COVID-19 patients: a retrospective study" *Emerging Microbes & Infections*
43. Ortega, Ribes, Vidal et al. (2021) "Seven-month kinetics of SARS-CoV-2 antibodies and role of pre-existing antibodies to human coronaviruses" *Nat Commun*
44. Anderson, Goodwin, Verma et al. (2021) "Seasonal human coronavirus antibodies are boosted upon SARS-CoV-2 infection but not associated with protection" *Cell*
45. Becker, Strengert, Junker et al. (2021) "Exploring beyond clinical routine SARS-CoV-2 serology using MultiCoV-Ab to evaluate endemic coronavirus cross-reactivity" *Nat Commun* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724184&blobtype=pdf | # Evidence for G-quadruplex-mediated transactivation by the immediate-early 2 protein of human cytomegalovirus
Shuang Gong, Daegyu Park, Woo-Chang Chung, Jin-Hyun Ahn
## Abstract
G-quadruplex (G4) formation in gene promoters regulates promoter activity either positively or negatively. Although G4s have been shown to enhance promoter activation by serving as structural elements that recruit transcription factors, similar information on viral promoters is limited. We recently found that the human cytomegalo virus (HCMV) 86 kDa immediate-early 2 (IE2) protein, which acts as a viral transactivator, binds to a parallel G4 structure in the viral lytic replication origin, oriLyt, playing a role in initiating viral DNA replication. Here, we show that IE2 can transactivate a viral promoter by targeting G4. In HCMV-infected cells, the transcript for the UL146-UL132 locus, known to be upregulated by IE2, was significantly suppressed by treatment with N-methyl mesoporphyrin IX (NMM), a parallel G4-binding ligand. A putative G4-forming sequence, also called G4 motif, was identified in the UL146 promoter. Circular dichroism, fluorescence turn-on, and gel electrophoresis analyses confirmed the stable formation of a parallel G4 from this motif. In G4 pull-down assays, IE2 bound to G4, and NMM inhibited the G4 binding of IE2. In reporter assays, the G4 motif was necessary for activity of the UL146 promoter, and NMM treatment suppressed IE2-mediated transactivation of the UL146 promoter. UL146 transcription was significantly reduced in a recombinant virus with G4-disrupting mutations in the UL146 promoter, especially at low multiplicity of infection, along with decreased G4 formation and IE2 binding. Our findings demonstrate that G4 can act as a positive structural element in viral promoters, uncovering a novel mechanism of IE2-mediated transactivation. IMPORTANCE G4-quadruplex (G4) is a noncanonical nucleic acid secondary structure that forms in single-stranded DNA or RNA. G4 structures are involved in various cellular processes, including transcription, translation, and DNA replication. It has been shown that G4 formation in gene promoters can inhibit promoter activation. Recently, a positive role of G4 in promoters has also been discovered, but evidence in viral promoters remains limited. In this study, we demonstrate that a G4 structure forms in the human cytomegalovirus UL146 promoter and acts as a binding site for the viral transactivator IE2, which plays a key role in viral gene expression. Using recombinant viruses, we confirm that IE2 binding to G4 is necessary for efficient UL146 transcription during virus infection. This study provides evidence that G4 structures in promoters can positively regulate viral gene expression and uncovers a novel mechanism by which IE2 activates gene expression.
or multiple strands and adopt various topologies, including parallel, antiparallel, and hybrid conformations, depending on the sequence composition and environmental conditions (3). Since their initial discovery in telomeric regions, DNA or RNA G4s have been identified across genomes, as well as in coding and noncoding RNAs, playing regulatory roles in various cellular processes, including transcription, translation, DNA replication, and genome stability (3,4).
Human cytomegalovirus (HCMV), a member of the betaherpesvirinae subfamily, is a common human pathogen with global seroprevalence estimates ranging from 60% to greater than 90%. HCMV establishes lifelong latency in hematopoietic progenitor and myeloid lineage cells. The reactivation of HCMV can lead to serious illness in immuno compromised individuals, including transplant recipients and newborns with congenital infection. The HCMV genome is approximately 235 kb in length and contains more than 200 open reading frames. Viral genes are expressed in a temporally regulated cascade of immediate-early (IE), E, and late gene expression that governs viral replication and host interaction (5).
Genome-wide studies have shown that HCMV contains many putative G4-forming sequences (PQSs), also known as G4 motifs, with a high frequency in repeated regions, gene promoters, and the lytic replication origin (6)(7)(8). G4s in promoters are known to suppress transcription by inhibiting the recruitment of transcription factors (9). We previously demonstrated that several G4 motifs are present in the HCMV promoter regions and can form stable G4 structures. Treatment with the G4 ligand N-methyl mesoporphyrin IX (NMM) suppressed several G4-containing promoters, indicating that G4 ligands suppress G4-containing viral promoters in a promoter context-dependent manner. By analyzing a G4 motif-deleted recombinant virus, we demonstrated that G4 formation in UL35 inhibits promoter activity and that the G4 ligand can suppress the promoter through G4 stabilization (7). Similarly, G4 formation in the HCMV miR-US33 promoter has been shown to suppress promoter activity (10).
Recently, it has also been shown that G4 formation can positively regulate promoter activity. G4s in the human genome are common binding hubs for various transcription factors and promote increased transcription (11). G4 formation in the cMyc oncogene promoter increases gene transcription (12). In herpes simplex virus type 1, the viral IE transcription factor ICP4 specifically binds to parallel-stranded G4 structures loca ted in the IE gene promoters, enhancing their activation. Disruption of G4 formation impairs ICP4 recruitment and diminishes gene expression, supporting a role in activation (13). The promoters of viral Bcl-2 homologs, KS-Bcl-2 of Kaposi's sarcoma-associated herpesvirus and BHRF1 of Epstein-Barr virus, contain G4 motifs. Treatment with the G4 ligand pyridostatin enhances the activity of KS-Bcl-2 and BHRF1 promoters, indicating a positive role for G4 formation in promoter activation (14). Recent integrative analyses of the human genome have demonstrated that promoter-associated G4s are subject to selective constraint and often correlate with high-level gene expression, supporting their stimulatory role in gene expression (15).
Although G4s have been linked to transcriptional regulation in the HCMV genome, their ability to positively regulate HCMV promoters remains underexplored. We recently demonstrated that the 86 kDa IE2 protein, a key regulator of HCMV gene expression and DNA replication (16,17), directly binds to a parallel G4 in essential region I (ER-I) within the viral lytic origin of replication (oriLyt), facilitating DNA replication (8). During HCMV infection, IE2 has been shown to transactivate several viral and cellular genes by interacting with general or specific transcription factors in host cells (18)(19)(20). Recent studies demonstrated that IE2 drives host RNA polymerase II (Pol II)-mediated transcrip tion initiation in a subset of viral late infection promoters, either dependent or independ ent of viral late-acting transcription factors (LTFs) (21,22). IE2 has also been shown to activate or repress transcriptional initiation and modulate Pol II elongation by directly binding to viral DNA (16).
The finding that IE2 interacts with a G4 led us to explore G4 targeting as a possible mechanism of IE2-mediated transactivation. In this study, we show that the UL146 promoter, significantly activated by IE2 during the late phase of infection independent of viral LTFs, is effectively suppressed by a parallel G4-binding ligand. We show that a parallel G4 is formed in the UL146 promoter and is directly targeted by IE2 and that this interaction is required for efficient activation of UL146 transcription during HCMV infection. This highlights a positive role for G4s in viral promoters and uncovers a novel G4-dependent mechanism of IE2-mediated transactivation.
## RESULTS
## Identification of the UL146-UL132 locus as an HCMV gene suppressed by NMM
We previously demonstrated that the addition of a G4 ligand, NMM, to HCMV (Toledo strain)-infected cells effectively suppressed the transcription of several viral genes and lowered the production of progeny virions (7). NMM is an asymmetric anionic porphyrin G4 ligand, which binds to G4 through π-π stacking and is highly specific to parallel G4 over antiparallel G4 and duplex DNA (23,24) (Fig. 1A). To further investigate the effect of prolonged NMM treatment on the accumulation of viral transcripts, we infected human fibroblast (HF) cells with HCMV (Toledo strain) at a multiplicity of infection (MOI) of 5 and analyzed viral transcript accumulation at 96 h after infection using RNA-seq analysis. Although the RNA-seq data were obtained from a single experimental set, among the viral genes downregulated, UL146, UL132, UL147, UL148, and UL147A, which are expressed as a single transcript (25), were most effectively suppressed by NMM (Fig. 1B andC; File S1). Thus, NMM appeared to inhibit the transcription at the UL146-UL132 locus by affecting promoter activation. When we determined the effect of NMM on the viral mRNA levels in HCMV-infected cells by quantitative real-time PCR (qRT-PCR), NMM suppressed the accumulation of UL146 transcripts in a dose-dependent manner, while it did not affect the accumulation of UL112 transcripts, as seen in RNA-seq data (Fig. 1D). This indicates that NMM effectively suppresses transcription of the UL146-UL132 locus during virus infection.
## G4 formation in the HCMV UL146 promoter
The UL146 promoter, which drives transcription of the UL146-UL132 locus during earlylate kinetics, lacks a typical TATA box (25) or a TATT box required for transactivation by viral LTFs (21,22). We identified a G-rich sequence from -61 to -19 bp upstream of the UL146 transcription start site (Fig. 2A). This G4 motif in the UL146 promoter is highly conserved in different HCMV strains (Fig. 2B). Several noncanonical PQSs, designated PQS1 to PQS4, were identified (Fig. 2C). To investigate G4 formation from these sequen ces, we synthesized the corresponding oligodeoxynucleotides (ODNs) and their mutant versions, where four guanines were replaced with adenines to disrupt G4-forming potentials (Fig. 2C). These were then used for circular dichroism (CD) analysis.
When the wild-type and mutant ODNs were incubated to form G4 structures and subjected to CD analysis, all the PQSs (PQS1 to PQS4) showed a positive peak at 265 nm and a negative peak at 240 nm, indicating the formation of a parallel G4. However, their mutants showed alteration in positive peak positions, suggesting reduced G4 formation. As controls, the wild-type and mutant cMyc G4 motifs exhibited similar patterns (Fig. 3A). NMM fluorescence intensity increases significantly upon binding to G4 relative to ssDNA (26). Consistent with the CD results, fluorescence turn-on assays with NMM showed that intact PQS1-PQS4, but not their mutants, displayed a typical G4-bound NMM fluores cence pattern under KCl buffer conditions, which stabilizes G4 structures, but not under LiCl conditions, where G4s are not formed. As a negative control, poly-A sequence failed to display the G4-bound NMM fluorescence pattern under KCl buffer conditions. Similar patterns were observed with the wild-type and mutant cMyc G4 motifs (Fig. 3B).
G4s can be formed intramolecularly in single-stranded DNA or intermolecularly by multiple strands. In native polyacrylamide gel electrophoresis (PAGE) analysis with G4preformed wild-type and mutant ODNs, PQS1, PQS2, and PQS3 produced slowly migrating ladders, whereas their mutants yielded a band of unstructured single-stranded DNA. We reason that the weak signals for the wild-type sequences are due to the effective formation of intermolecular G4s. This indicates that PQS1, PQS2, and PQS3 favor the formation of intermolecular G4s in vitro. PQS4 formed both slowly migrating ladders and a faster migrating band, suggestive of intramolecular G4, in comparison to its mutant (Fig. 4A). The formation of intramolecular G4s from PQS2, PQS3, and PQS4 became apparent when G4s were formed under molecular crowding conditions with 40% (wt/vol) polyethylene glycol (PEG) 200 and analyzed by native PAGE, although the bands were shifted, probably due to PEG (Fig. 4B). As controls, the wild-type cMyc G4 motif also showed a faster migration pattern than its mutant version, cMyc-G4m (Fig. 4C). Collectively, the in vitro analyses results demonstrate that a parallel intramolecular G4 is formed in the UL146 promoter.
## NMM inhibits the binding of IE2 to UL146 G4
Since a G4 is formed in the UL146 promoter, which is thought to be effectively sup pressed by NMM, and IE2 has parallel G4-binding activity (8), we examined whether IE2 binds to the UL146 promoter G4. Although the full-length IE2 was difficult to purify in bacterial cells due to low solubility, we could purify the N-terminal 86 amino acid-deleted IE2 form as a soluble protein. In G4 pull-down assays, IE2 (87-579) could bind to all G4s formed by PQS1 to PQS4 ODNs with a binding strength comparable to or greater than that formed by the oriG4-1 sequence in HCMV oriLyt ER-I. However, it did not interact with mutant ODNs, indicating that IE2 directly binds to the UL146 promoter G4 (Fig. 5A).
Given that NMM suppressed UL146 transcription, we tested whether NMM can interfere with IE2 binding to the UL146 promoter G4. In G4 pull-down assays, NMM inhibited IE2 binding to the G4s formed by PQS1, PQS2 (though not statistically significant), and PQS4, while it did not affect IE2 binding to the G4 formed by PQS3 (Fig. 5B). PQS4m showed a significant reduction in IE2 binding by NMM treatment. In addition, PQS4, longer than other PQSs, formed an intramolecular G4 effectively. Therefore, we used PQS4 for further analysis of IE2 binding. Unlike NMM, which binds to a parallel G4 through a G-quartet (26), the G4 groove binder Peimine (dihydroisoimperia line) (27) did not inhibit the binding of IE2 to the G4 (Fig. 5C andD). This indicates that IE2 binding to UL146 G4 may involve G-quartet binding. Collectively, these results demonstrate that the ability of IE2 to bind UL146 G4 can be inhibited by NMM.
## IE2 transactivates the UL146 promoter through G4
We next examined whether IE2 transactivates the UL146 promoter by targeting G4. To achieve this, we created reporter plasmids containing the luciferase reporter gene driven by the intact UL146 promoter (UL146p) or the G4-defective mutant promoter (UL146(G4m)p), in which the PQS4 sequence was replaced with the PQS4m sequence (Fig. 6A). HF cells were transfected with these reporter plasmids via electroporation and then infected with HCMV, with or without NMM treatment, followed by luciferase assays. The UL146(G4m) promoter had a 12-fold reduction in activity compared to the wild-type promoter in virus-infected cells, indicating that the G4 motif is essential for efficient UL146 promoter activation (Fig. 6B). We also found that NMM significantly suppressed the UL146 promoter activity by 27-fold, while it reduced the activity of the UL146(G4m) promoter by 10-fold (Fig. 6B). These findings suggest that NMM primarily inhibits the UL146 promoter through G4, although it can also suppress promoter activity in a G4independent manner, likely by downregulation of other viral genes necessary for UL146 promoter activation.
To assess the effect of NMM on IE2-mediated UL146 promoter activation, we also conducted co-transfection reporter assays in U373-MG cells, which are semi-permissive to HCMV infection and exhibit high transfection efficiency. We observed that IE2 alone could transactivate the UL146 promoter but not the UL146(G4m) promoter and that NMM effectively suppressed the UL146 promoter activated by IE2 (Fig. 6C). These findings demonstrate that IE2 transactivates the UL146 promoter through G4, and NMM interferes with IE2-mediated UL146 transactivation.
## Evaluation of G4-dependent IE2 transactivation of the UL146 promoter using recombinant viruses
Activation of the UL146 promoter by IE2 through G4 was further examined by producing a recombinant HCMV (UL146(G4m)) with G4-disrupting mutations (PQS4m sequence) in the UL146 promoter, along with its revertant virus. The HCMV (Toledo strain) bacmids containing PQS4m and its revertant were produced by bacmid mutagenesis using the counter-selection marker rpsL-neo (Fig. 7A). The recombinant viruses were grown in HF cells that received the bacmid DNA via electroporation. UL146 encodes a viral chemo kine, vCXCL1, which is not essential for viral growth in cell culture (28,29). Consistently, the UL146(G4m) virus exhibited a growth curve similar to those of the wild-type and revertant viruses at a MOI of 2 or 0.2 (Fig. 7B). However, the UL146 transcript levels measured by real-time quantitative PCR (RT-qPCR) were about sixfold lower in UL146(G4m) virus infection at 5 or 7 days post-infection than in wild-type and revertant virus infections at an MOI of 2. At an MOI of 0.2, this reduction was more pronounced to approximately 3,000-fold at 12 days (Fig. 7C). Similarly, the UL132 mRNA levels were about 2.5-fold and 124-fold lower at MOIs of 2 (7 days) and 0.2 (12 days), respectively, during UL146(G4m) virus infection (Fig. 7D). These results indicate that G4 formation in the UL146 promoter is necessary for efficient transcription of the UL146-UL132 locus during virus infection, especially at low MOI.
To confirm G4 formation in the UL146 promoter and IE2 binding to G4 during virus infection, HF cells were infected with recombinant viruses, and chromatin immunoprecipitation (ChIP) assays were conducted using antibodies for G4 and IE2.
Since the IE2specific antibody we used was ineffective for ChIP assays, we used an antibody that detects the common N-terminal region of IE1 and IE2. We observed that G4 formation in the UL146 promoter region and the IE2 protein level associated with the promoter were significantly lower in the UL146(G4m) virus infection compared to wildtype and revertant virus infections (Fig. 7D). These findings with recombinant viruses demonstrate that IE2 transactivates the UL146 promoter through G4 during virus infection.
## DISCUSSION
In this study, we demonstrate that HCMV IE2 transactivates the UL146 promoter by targeting a G4 structure located immediately upstream of the transcription start site. The G4-forming sequence identified in the UL146 promoter was not predicted in our previous genome-wide search for G4 motifs, in which we used a G (3-6) N (1-7) G (3-6) N (1- 7) G (3-6) N (1-7) G (3-6) schema (7). However, the findings that UL146 transcription depends on IE2 (21,22); that IE2 binds to a parallel G4; and that NMM, a parallel G4-binding ligand, effectively suppresses UL146 transcription (this study), led us to investigate potential G4 formation within the UL146 promoter. We found a noncanonical G4 motif in the UL146 promoter and confirmed the formation of a parallel G4 structure by analyzing the corresponding ODN in CD analysis, G4-ligand fluorescence turn-on assay, and native Full-Length Text PAGE in vitro. Additionally, G4 formation in the viral genome during infection was validated by G4-ChIP assays using a G4specific antibody in recombinant virus-infected cells. It remains unclear whether the G4 formed in the UL146 promoter has a structure with two G-quartets or a bulged structure with three G-quartets. However, our native PAGE analysis showed that the G4 motif can adopt an intramolecular G4 structure, which may be important for transcriptional control in the viral genome during infection. We demonstrated that IE2 binds directly to the UL146 promoter G4 in vitro. The strength of IE2 binding to this G4 was comparable to that of the oriLyt ER-I G4. Consistent with its inhibitory effect on IE2-mediated UL146 transcription during virus infection, NMM inhibited IE2 binding to the UL146 promoter G4. Interestingly, NMM inhibited IE2 binding to the G4 formed by PQS1, 2, and 4 but not that by PQS3, although IE2 bound to the G4s formed by all PQS ODNs. This suggests that the G4 structure formed by PQS3 may differ from those formed by the other ODNs, and the G4-binding modes of IE2 are also different. We previously demonstrated that, when IE2 binds to a parallel G4 structure in oriLyt ER-I, G-quartet binding is more important than groove binding (8). IE2 seems to recognize UL146 G4 in a similar way, because Peimine, which binds to G4 grooves, did not inhibit its binding to UL146 G4. TMPyP4 (tetra(N-methyl-4-pyridyl)porphyrin) is a potent porphyrin G4 binder with a 4 + charge and a planar core that enables interaction with G4 through electrostatic interactions (30), although it shows low selectivity for G4 versus double-stranded DNA (dsDNA) (31). Indeed, TMPyP4 was more effective at inhibiting the binding of IE2 to oriLyt ER-I G4 than NMM (8). When we tested the effect of TMPyP4 on IE2 binding to UL146 G4 and IE2-mediated transactivation of the UL146 promoter, TMPyP4 more effectively inhibited the UL146 G4 binding by IE2, but its suppression of IE2-mediated UL146 promoter activation was similar to that of NMM (S. Gong and J.-H Ahn, unpublished data). We deduce that, although NMM has relatively low affinity for G4 because of its negative charge and nonplanarity, it functions as a good G4 binder in cells, likely due to its efficient translocation. Our findings that IE2 transactivates the UL146 promoter by targeting a promoter G4 structure support the view of G4 as a positive structural element that regulates promoter activity by acting as a binding site for transcription factors (11,15). We previously showed that treatment with G4 ligands suppresses G4-containing HCMV promoters in a context-dependent manner (7). This suppressive effect can be explained in two ways. G4 can act as a negative element for promoter activation by inhibiting the RNA Pol II complex, and G4 stabilization by ligands further suppresses promoter activation. Conversely, G4 can be a positive element that recruits transcription factors. However, G4 ligands may interfere with the G4 binding of transcription factors. The results of the current study provide evidence that G4 can act as a positive element in promoter activation in the HCMV genome. Using genome-wide functional studies of a specific G4 structure with recombinant viruses, we demonstrated G4 formation in the UL146 promoter and its binding to IE2. We also demonstrated the positive role of G4 binding by IE2 in UL146 transcription during virus infection. Interestingly, we found that the mutant virus containing the UL146 promoter lacking G4-forming potential exhibited a more severe reduction of UL146 transcription at low MOI than at high MOI. This finding indicates that UL146 promoter activation mainly depends on G4 targeting by IE2 at low MOI but may also involve G4-independent mechanisms at high MOI. IE2 binds to a 14 bp cis-repression sequence (crs), a CG dinucleotideflanked AT-rich sequence, within the major IE promoter, repressing its transcription through negative feedback regulation (32)(33)(34)(35)(36)(37). The binding of IE2 to the crs-like site appears to be involved in the activation of some viral promoters by IE2 (16,(38)(39)(40). We found one crs-like site and several putative binding sites of alkaline phosphatase (AP)-1, AP-2, and CREB, which are known to interact with IE2, in the UL146 promoter region (Fig. 2A). IE2 could not activate the G4 motif-disrupted UL146 promoter in our reporter assays with co-transfected cells. However, it cannot be ruled out that the presence of these sites contributes to the significant activation of the G4-less UL146 promoter during high MOI infection. IE2 has been shown to bind the crs as a dimer or oligomer in vitro (41,42). Whether IE2 binds to G4 as a monomer, dimer, or oligomer remains to be addressed. It would be intriguing to assess whether different states of IE2 multimerization influence its binding to different nucleic acid structures in the viral genome.
IE2 has also been shown to bind directly to several viral early promoters and cooperate with cellular transcriptional factors in activating transcription (39,40,43,44). A recent study on genome-wide IE2 occupancy has demonstrated that IE2 binds to duplex DNA near core promoter regions to activate many early-late and late promoters (16). Our study for the first time provides evidence that IE2 is also able to transactivate a viral early-late promoter that lacks an upstream TATA or TATT sequence by directly binding to a G4 structure formed in the promoter. This work underscores the different modes of IE2 association with the viral genome and the importance of G4 structures in HCMV gene regulation. Given that G4-binding ligands can effectively inhibit the binding of IE2 to G4, our study also highlights these compounds as promising antivirals capable of disrupting the activation of select IE2-dependent G4-containing viral and cellular promoters.
## MATERIALS AND METHODS
## Cell culture, virus, and chemicals
Primary HF cells and human glioblastoma U373-MG cells (ATCC) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin. The cells were maintained at 37°C in a humidified incubator with 5% CO₂. The recombinant HCMV (Toledo strain) was grown in HF cells that received bacmid DNA as previously described (45). NMM was purchased from Santa Cruz Biotechnology. TMPyP4 was purchased from GLPbio. Peimine was obtained from Cayman Chemical.
## Plasmids, transfection, and electroporation
pMP18 expressing IE2 was previously described (46). To produce the UL146 pro moter-driven luciferase reporter plasmid (UL146p-Luc), the UL146 promoter region (-927 to +108) was PCRamplified as an NheI/NcoI fragment and cloned into pGL3basic (Addgene). UL146(G4m)p-Luc containing the PQS4m sequence was produced by site-directed mutagenesis. U373-MG cells were transfected using polyethylenimine (Sigma-Aldrich). HF cells were transiently transfected via electroporation at 1,700 V for 20 ms using a Microporator MP-100 (Digital Bio) according to the manufacturer's instructions.
## Infectious center assay
Virus titers were assessed using infectious center assays. HF cells (1 × 10⁵ cells) were plated in 24-well plates and incubated for 24 h before infection. Serial dilutions (10-fold) of viral stocks (200 µL per well) were added and adsorbed to cells for 1 h. The inoculum was then removed, and 1 mL of fresh complete medium was added, followed by an additional 24-h incubation. Cells were fixed with 500 µL methanol at -20°C for 10 min, washed three times with cold phosphate buffered saline (PBS), and incubated with 200 µL of anti-IE1 rabbit polyclonal antibody in PBS at 37°C for 1 h. Subsequently, the cells were treated with a phosphatase-labeled anti-rabbit IgG antibody under the same conditions. For visualization, 200 µL of AP buffer (100 mM Tris-HCl, 100 mM NaCl, and 5 mM MgCl₂) was mixed with BCIP/NBT substrate (1:1 ratio, Millipore) and applied to the cells. IE1-positive cells were counted in five fields per well using a light microscope at 200 × magnification.
## RNA-seq analysis
Total RNA was extracted using TRIzol Reagent (Thermo Fisher Scientific) and treated with DNase. For total RNA-seq, ribosomal RNA was depleted (Ribo-Zero rRNA removal), and the remaining RNA was randomly fragmented and reverse-transcribed into cDNA, followed by ligation with Illumina adapters, PCR amplification, and size selection to obtain ~200-400 bp inserts. Libraries were prepared with the Illumina TruSeq Stranded Total RNA LT Sample Prep Kit (Illumina) and sequenced by Macrogen on an Illumina platform (paired-end, 101 bp reads).
## CD spectroscopy
ODNs were dissolved at a concentration of 15 µM in buffer containing 10 mM Tris-HCl (pH 7.5) and either 100 mM KCl or 100 mM LiCl. The samples were heated to 95°C for 5 min to denature, cooled, and annealed overnight at room temperature. CD spectra were acquired at 25°C using a Jasco J-810 spectropolarimeter equipped with a Peltier temperature controller. Measurements were performed from 220 to 320 nm, averaging three accumulations, with a response time of 1 s, a scanning speed of 100 nm/min, and a data pitch of 1 nm. The wild-type and mutant cMyc-G4 motifs were used as controls: 5′-TGAGGGTGGGTAGGGTGGGTAA-3′ (22-mer, for cMyc-G4) and 5′-TGAGAGTGAGTAGAGT GAGTAA-3′ (22-mer for cMyc-G4m).
## NMM fluorescence turn-on assay
ODNs were prepared at a final concentration of 2 µM in G4 folding buffer containing 10 mM Tris-HCl (pH 7.5) and 100 mM KCl. Samples were heated at 95°C for 5 min to denature secondary structures and then slowly cooled to room temperature overnight.
NMM was then added to each sample at a final concentration of 4 µM, followed by incubation at room temperature in the dark for 10 min to allow binding. Fluorescence emission spectra were recorded using a BioTek Synergy Neo (Agilent) spectrofluorometer with excitation at 394 nm and emission collected from 550 to 750 nm. The slit widths for both excitation and emission were set to 2 nm.
## Native PAGE
ODNs (5 µM) were dissolved in 10 mM Tris-HCl buffer (pH 7.5, room temperature), denatured by heating at 95°C for 5 min, and then cooled overnight at room temperature. The samples were analyzed via native 12% PAGE with or without 40% (wt/vol) PEG 200.
The gel was stained with SYBR Gold and scanned using a GelDoc XR+ System (Bio-Rad).
## G4 pull-down assay
Biotin-tagged ODNs (11.3 µM) preincubated to form structures were incubated with streptavidin agarose beads (30 µg, Sigma S1638) in buffer (10 mM Tris, pH 7.5, 1 mM EDTA, 1 M NaCl, and 0.003% NP40) at room temperature for 30 min with rotation. After centrifugation, the immobilized ODNs were blocked with bovine serum albumin-con taining buffer to minimize nonspecific interactions. ODNs were then incubated at 4°C for 1 h with bacterially purified IE2 proteins (2 µg) in binding buffer. After three washes, the ODN/protein complexes were resuspended in SDS-PAGE sample buffer, incubated at 37°C for 15 min, and then boiled at 97°C for 7 min, followed by SDS-PAGE and immunoblot analysis.
## Real-time qPCR
Total RNA was extracted from the cells using TRIzol Reagent (Thermo Fisher Scientific) and MaxTract High Density (Qiagen). A QuantiTect Reverse Transcription kit (Qiagen) was used to generate cDNA. RT-qPCR was conducted using Power SYBR Green PCR Master Mix and a QuantStudio Real-Time PCR System. The primers used to amplify the UL146 mRNA were 5′-ATAAGCGGGAGATGTGGATG-3′ (forward) and 5′-ACATTAAATATTATCCTCTA ACACCTA-3′ (reverse). The primers used for the GAPDH mRNA were 5′-AAATCCCATCACCA TCTTCCA-3′ (forward) and 5′-AGGGGCCATCCACAGTCTTCT-3′ (reverse).
## Luciferase reporter assay
Cells were collected and lysed through three freeze-thaw cycles in 100 µL of 0.25 M Tris-HCl (pH 7.9) with 1 mM dithiothreitol. After centrifugation to clarify the extracts, 30 µL of cell extract was incubated with 350 µL of reaction buffer A (25 mM glycylglycine [pH 7.8], 15 mM ATP, and 4 mM EGTA). The mixture was then combined with 100 µL of 0.25 mM luciferin (Sigma-Aldrich) in reaction buffer A. Luciferase activity was measured using a TD-20/20 luminometer (Turner Designs) over a 10-s assay, recorded in relative light units.
## HCMV bacmid mutagenesis
UL146 mutant and revertant Toledo-bacmids were generated using the Red/ET recombination system with a BAC Modification Kit (Gene Bridges). Briefly, an rpsL-neo selection cassette flanked by 100-nucleotide homology arms corresponding to the regions upstream and downstream of the UL146 target site was amplified by PCR.
The resulting PCR products were purified and introduced into Escherichia coli DH10B cells harboring the wild-type HCMV (Toledo)-bacmid by electroporation (Gene Pulser II, Bio-Rad). Transformants containing the cassette were selected on Luria-Bertani (LB) agar plates supplemented with kanamycin (50 µg/mL). To generate the desired UL146 mutations, two complementary single-stranded ODNs encoding the mutant sequence were annealed to form dsDNA fragments, which were then introduced by homologous recombination to replace the rpsL-neo cassette. Recombinant colonies were selected on LB agar plates containing streptomycin (100 µg/mL), and correct integration was confirmed by PCR amplification and direct sequencing of the UL146 region. To generate the revertant bacmid, the rpsL-neo cassette was re-inserted at the UL146 locus of the mutant bacmid, followed by replacement with annealed ODNs containing the wild-type UL146 sequence using the same recombination strategy described above. All modified bacmids were verified by sequencing before use.
## ChIP assay
The ChIP Assay Kit (Millipore, #17-295) was used according to the manufacturer's instructions. HF cells (2 × 10⁶ cells) infected with HCMV were harvested at 4 days post-infection. Cells were cross-linked with 1% formaldehyde for 10 min at room temperature and then lysed using the buffer provided in the kit. Chromatin was sheared by sonication to an average size of 200-600 bp. Immunoprecipitations were carried out using G4specific BG4 antibody, anti-IE1/IE2 antibody, or control IgG. After immunopreci pitation and washing, cross-links were reversed, and DNA was purified. The immunopre cipitated DNA was analyzed by qPCR using primer sets amplifying the G4-containing UL146 promoter region. The primers for UL146 were 5′-ATCGATTTTGAAACCTAATTGA-3′ (forward) and 5′-TACCAGTAATTCGTAATATC-3′ (reverse). PCR amplification was performed under the following conditions: 95°C for 5 min, followed by 50 cycles of 95°C for 30 s and 55°C for 30 s.
## Statistical analysis
Samples were compared using Student's t-test, and significance was defined as P < 0.05 (*P < 0.05, **P < 0.01, or ***P < 0.001).
## References
1. Burge, Parkinson, Hazel et al. (2006) "Quadruplex DNA: sequence, topology and structure" *Nucleic Acids Res*
2. Fay, Lyons, Ivanov (2017) "RNA G-quadruplexes in biology: principles and molecular mechanisms" *J Mol Biol*
3. Varshney, Spiegel, Zyner et al. (2020) "The regulation and functions of DNA and RNA G-quadruplexes" *Nat Rev Mol Cell Biol*
4. Rhodes, Lipps (2015) "G-quadruplexes and their regulatory roles in biology" *Nucleic Acids Res*
5. Mocarski, Shenk, Griffiths et al. (2013) "Cytomegaloviruses"
6. Biswas, Kandpal, Jauhari et al. (2016) "Genome-wide analysis of G-quadruplexes in herpesvirus genomes" *BMC Genomics*
7. Ravichandran, Kim, Bansal et al. (2018) "Genome-wide analysis of regulatory G-quadruplexes affecting gene expression in human cytomegalovirus" *PLoS Pathog*
8. Park, Chung, Gong et al. (2024) "G-quadruplex as an essential structural element in cytomega lovirus replication origin" *Nat Commun*
9. Li, Wang, Yin et al. (2021) "Ligand-induced native G-quadruplex stabiliza tion impairs transcription initiation" *Genome Res*
10. Kumar, Choudhary, Patra et al. (2020) "Analysis of G-quadruplexes upstream of herpesvirus miRNAs: evidence of G-quadruplex mediated regulation of KSHV miR-K12-1-9,11 cluster and HCMV miR-US33" *BMC Mol Cell Biol*
11. Spiegel, Cuesta, Adhikari et al. (2021) "G-quadruplexes are transcription factor binding hubs in human chromatin" *Genome Biol*
12. Esain-Garcia, Kirchner, Melidis et al. (2024) "Gquadruplex DNA structure is a positive regulator of MYC transcription" *Proc Natl Acad Sci*
13. Frasson, Soldà, Nadai et al. (2021) "Parallel Gquadruplexes recruit the HSV-1 transcription factor ICP4 to promote viral transcription in herpes virus-infected human cells" *Commun Biol*
14. Kumar, Ramamurthy, Choudhary et al. (2022) "Contrasting roles for G-quadruplexes in regulating human Bcl-2 and virus homologues KSHV KS-Bcl-2 and EBV BHRF1" *Sci Rep*
15. Li, Su, Wang et al. (2023) "Integrative genomic analyses of promoter G-quadruplexes reveal their selective constraint and association with gene activation" *Commun Biol*
16. Ball, Li, Parida et al. (2022) "Human cytomegalovirus IE2 both activates and represses initiation and modulates elongation in a context-dependent manner" *mBio*
17. Pari (2008) "Nuts and bolts of human cytomegalovirus lytic DNA replication" *Curr Top Microbiol Immunol*
18. Kim, Kim, Huh et al. (2010) "Role of noncovalent SUMO binding by the human cytomegalovirus IE2 transactivator in lytic growth" *J Virol*
19. Lukac, Harel, Tanese et al. (1997) "TAF-like functions of human cytomegalovirus immediate-early proteins" *J Virol*
20. Stinski, Petrik (2008) "Functional roles of the human cytomegalovi rus essential IE86 protein" *Curr Top Microbiol Immunol*
21. Li, Hu, Collins et al. (2021) "Cytomegalovirus late transcription factor target sequence diversity orchestrates viral early to late transcription" *PLoS Pathog*
22. Li, Ball, Collins et al. (2020) "Human cytomegalovirus IE2 drives transcription initiation from a select subset of late infection viral promoters by host RNA polymerase II" *PLoS Pathog*
23. Umar, Chan, Kwok (2019) "G-quadruplex-based fluorescent turn-on ligands and aptamers: from development to applications" *Molecules*
24. Nicoludis, Miller, Jeffrey et al. (2012) "Optimized end-stacking provides specificity of Nmethyl mesoporphyrin IX for human telomeric G-quadruplex DNA" *J Am Chem Soc*
25. He, Ma, Qi et al. (2012) "Characterization of human cytomegalovirus UL146 transcripts" *Virus Res*
26. Yett, Lin, Beseiso et al. (2019) "N-methyl mesopor phyrin IX as a highly selective light-up probe for G-quadruplex DNA" *J Porphyr Phthalocyanines*
27. Li, Xiang, Li et al. (2009) "Stabilizing parallel G-quadruplex DNA by a new class of ligands: two non-planar alkaloids through interaction in lateral grooves" *Biochimie*
28. Penfold, Dairaghi, Duke et al. (1999) "Cytomegalovirus encodes a potent alpha chemo kine" *Proc Natl Acad Sci*
29. Berg, Wedemeyer, Melynis et al. (2022) "The non-ELR CXC chemokine encoded by human cytomegalovirus UL146 genotype 5 contains a C-terminal β-hairpin and induces neutrophil migration as a selective CXCR2 agonist" *PLoS Pathog*
30. (2025) *Full-Length Text Journal of Virology*
31. Nicoludis, Barrett, Mergny et al. (2012) "Interaction of human telomeric DNA with N-methyl mesoporphyrin IX" *Nucleic Acids Res*
32. Ren, Chaires (1999) "Sequence and structural selectivity of nucleic acid binding ligands" *Biochemistry*
33. Macias, Stinski (1993) "An in vitro system for human cytomegalovi rus immediate early 2 protein (IE2)-mediated site-dependent repression of transcription and direct binding of IE2 to the major immediate early promoter" *Proc Natl Acad Sci*
34. Chiou, Zong, Waheed et al. (1993) "Identification and mapping of dimerization and DNA-binding domains in the C terminus of the IE2 regulatory protein of human cytomegalovirus" *J Virol*
35. Cherrington, Khoury, Mocarski (1991) "Human cytomegalovirus ie2 negatively regulates alpha gene expression via a short target sequence near the transcription start site" *J Virol*
36. Lang, Stamminger (1994) "Minor groove contacts are essential for an interaction of the human cytomegalovirus IE2 protein with its DNA target" *Nucleic Acids Res*
37. Pizzorno, Hayward (1990) "The IE2 gene products of human cytomegalovirus specifically down-regulate expression from the major immediate-early promoter through a target sequence located near the cap site" *J Virol*
38. Liu, Hermiston, Stinski (1991) "A cis-acting element in the major immediate-early (IE) promoter of human cytomegalovirus is required for negative regulation by IE2" *J Virol*
39. Rodems, Spector (1998) "Extracellular signal-regulated kinase activity is sustained early during human cytomegalovirus infection" *J Virol*
40. Arlt, Lang, Gebert et al. (1994) "Identification of binding sites for the 86-kilodalton IE2 protein of human cytomegalovirus within an IE2-responsive viral early promoter" *J Virol*
41. Scully, Sommer, Schwartz et al. (1995) "The human cytomegalovirus IE2 86-kilodalton protein interacts with an early gene promoter via sitespecific DNA binding and protein-protein associations" *J Virol*
42. Teng, Bolovan-Fritts, Dar et al. (2012) "An endogenous accelerator for viral gene expression confers a fitness advantage" *Cell*
43. Waheed, Chiou, Ahn et al. (1998) "Binding of the human cytomegalovirus 80-kDa immediate-early protein (IE2) to minor groove A/T-rich sequences bounded by CG dinucleotides is regulated by protein oligomerization and phosphorylation" *Virology (Auckl)*
44. Schwartz, Sommer, Scully et al. (1994) "Sitespecific binding of the human cytomegalovirus IE2 86-kilodalton protein to an early gene promoter" *J Virol*
45. Rodems, Clark, Spector (1998) "Separate DNA elements containing ATF/CREB and IE86 binding sites differentially regulate the human cytomegalovirus UL112-113 promoter at early and late times in the infection" *J Virol*
46. Park, Kim, Shamim et al. (2022) "Analysis of novel drug-resistant human cytomegalovirus DNA polymerase mutations reveals the role of a DNA-binding loop in phosphonoformic acid resistance" *Front Microbiol*
47. Pizzorno, Hare, Sha et al. (1988) "Transactivation and autoregulation of gene expression by the immediateearly region 2 gene products of human cytomegalovirus" *J Virol*
48. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12646007&blobtype=pdf | # Recent advances in lateral flow devices and point-of-care diagnostics for highly pathogenic avian influenza A viruses
Wei Lin, Irwin Quintela, Shyam Sablani, Chih-Sheng Lin, Vivian Wu
## Abstract
There has been a resurgence and ongoing outbreak of avian influenza since early 2024. Avian influenza, caused by influenza A viruses, poses significant threats to both avian populations and public health due to its zoonotic potential. Highly patho genic avian influenza (HPAI) virus, such as the H5N1 and H7 subtypes, has a high mortality rate. Traditional detection methods, i.e., virus isolation and reverse transcrip tion quantitative PCR (qRT-PCR), are reliable for diagnosis but time-consuming and labor-intensive. Rapid and accurate detection of avian influenza A viruses is crucial to prevent widespread outbreaks and minimize economic losses. Field-ready and pointof-care (POC) diagnostics, such as lateral flow assays (LFAs) offer a rapid, early, and large-scale approach for detecting avian influenza A virus infections. Early detection in HPAI management is a key factor in improving treatment effectiveness and reduc ing the negative impact on animal health. This minireview introduces the principles and techniques of current field-ready and POC diagnostics, emphasizing LFAs for HPAI detection. It also outlines and compares their development, applications, and availa bility in the market. Notably, advanced techniques such as LFAs-integrated clustered regularly interspaced short palindromic repeats (CRISPR) have expanded HPAI diagnostic capabilities and have also been reviewed. CRISPR-based LFAs use guide RNA to detect viral sequences, activating Cas enzymes that generate a visible signal on test strips, enabling a rapid and sensitive detection method. To our knowledge, this is the first comprehensive review summarizing LFA-based HPAI diagnostics in the context of the 2024 resurgence, offering timely insight into their potential roles in outbreak prepared ness and response.
KEYWORDS HPAI, avian influenza, H5N1, lateral flow assay, CRISPRA vian influenza, commonly known as bird flu, is an infectious viral disease affecting bird species worldwide (1, 2). Caused by influenza A viruses, avian influenza poses significant threats to both avian populations and public health due to its potential for zoonotic transmission (3). Particularly, highly pathogenic avian influenza (HPAI), subtypes like H5N1 and H7 viruses, exhibit a high mortality rate (4). The rapid and accurate detection of avian influenza viruses (AIVs) is crucial for implementing control measures, preventing widespread outbreaks, and minimizing economic losses in the poultry industry (5,6).H5N1 was first detected in geese in Guangdong, China, 1996 (7). The first human H5N1 cases were described in Hong Kong in 1997 (8). In recent years, the global landscape of avian influenza has been marked by a series of notable outbreaks (Table 1). In May 2021, the HPAI H5N1 virus was detected in wild foxes at a rehabilitation center in the Netherlands during an outbreak of HPAI in wild birds (9). From late 2021 to 2022, the predominant HPAI H5 virus causing poultry outbreaks worldwide was the wild bird-adap ted HPAI H5N1 virus, according to the World Organization for Animal Health (10). In 2022, HPAI H5N1 re-emerged in commercial poultry, with a U.S. outbreak in turkeys
and two asymptomatic human cases reported in Spain among poultry workers exposed to infected flocks, marking the first U.S. detection since 2020 and raising concern over zoonotic transmission (10,11).
Between 2023 and early 2025, H5N1 continued to expand its host range and geographic impact. In February 2023, Cambodia reported two human infections, including the death of a child, linked to clade 2.3.2.1c (12). By December, H5N1 was identified in a polar bear in the Arctic (10). In March 2024, the U.S. detected H5N1 in dairy cows, followed by human infections in dairy and poultry workers (13). From September to December 2024, North America reported 45 human cases of clade 2.3.4.4b (14). As of February 2025, over 34 countries have reported outbreaks, disrupting poultry industries, while Brazil's exports surged amid global shortages (15).
Early detection is essential for effective disease management, enabling timely control measures and reducing the impact on animal health. Recent events highlight the evolving nature of highly pathogenic avian influenza viruses (HPAIVs) and reinforce the urgent need for robust surveillance systems and reliable diagnostic tools. Traditional diagnostic methods for HPAIVs, such as virus isolation and reverse transcription-PCR (RT-PCR), are highly sensitive and specific (16,17). However, they require specialized laboratory facilities and trained personnel, which are time-consuming and may delay outbreak response efforts (18). In contrast, lateral flow assays (LFAs) offer a rapid, user-friendly, and cost-effective alternative for the on-site detection of HPAIVs (19,20). LFAs are designed to detect specific antigens or antibodies in samples, providing results within minutes without requiring complex equipment (21).
The application of LFAs in avian influenza detection has gained momentum, especially in resource-limited settings and during field investigations. Their porta bility and ease of use make them invaluable for immediate decision-making and • Mammal-to-mammal transmission potential is increasing implementation of control measures. Moreover, advancements in LFA technology have led to improved sensitivity and specificity, enhancing its reliability as a diagnostic tool. This review provided a comprehensive overview of the application of LFAs in avian influenza detection, focusing on HPAI strains. In light of the resurgence and ongoing outbreaks since early 2024, impacting poultry, dairy cattle, and even leading to zoonotic infections, there is an urgent need for rapid, accessible diagnostics that complement traditional laboratory-based methods such as virus isolation and reverse transcription quantitative PCR (qRT-PCR). LFAs offer practical advantages for early detection, largescale screening, and on-site decision-making during outbreaks. To our knowledge, this is the first comprehensive review of LFA-based diagnostics for HPAI conducted in the context of the 2024 outbreak. As HPAI continues to threaten both animal and human health, integrating rapid diagnostic tools like LFAs into surveillance programs will be pivotal in mitigating the impact of future outbreaks.
## ZOONOTIC HIGHLY PATHOGENIC AVIAN INFLUENZA A VIRUSES AND THE TROPISM
Influenza A viruses belong to the Orthomyxoviridae family and possess an eight-seg mented, negative-sense RNA genome (22). The virus comprises eight gene segments encoding essential proteins, including hemagglutinin (HA) and neuraminidase (NA), which determine viral subtype and host specificity (23) (Fig. 1). The segmented genome allows reassortment, facilitating the emergence of novel strains with zoonotic and pandemic potential (24).
Influenza A viruses are divided into subtypes based on two surface proteins of the virus -HA and NA. There are 18 known HA subtypes and 11 known NA subtypes (25,26). Various subtypes of AIVs have been associated with zoonotic transmission, causing sporadic human infections and posing pandemic threats (27). These viruses primarily circulate among birds but have demonstrated the ability to infect humans, sometimes leading to severe disease outbreaks (28). Continuous surveillance and research are essential to understand their evolution, transmission, and potential impact on public health (Fig. 2).
## H5N1 viruses
HPAI H5N1 has caused severe outbreaks in poultry and sporadic human cases since its emergence in 1997 in Hong Kong (29). This virus has demonstrated a high fatality rate in humans, often exceeding 50%, making it a significant concern for public health (30). Transmission occurs primarily through direct contact with infected birds or contamina ted environments, with no sustained human-to-human transmission reported to date (31). However, the virus continues to evolve through genetic reassortment and mutation, raising concerns over its pandemic potential (32).
H5N1 has spread across multiple continents, affecting poultry farms, wild birds, and, occasionally, mammals (33). Several clades and subclades of H5N1 have been identified, each with varying levels of virulence and transmissibility (34,35). H5N1 primarily infects the respiratory and gastrointestinal tracts of birds, causing severe systemic infections (30). In humans, H5N1 preferentially binds to α2,3-linked sialic acid receptors found in the lower respiratory tract, leading to severe pneumonia, acute respiratory distress syndrome (ARDS), and multiorgan failure (36). The virus has been isolated from the brain, intestines, and other organs, indicating neurotropism and systemic dissemination (37). Its ability to infect mammals, including humans, raises concerns about zoonotic spillover and pandemic potential (33,38) (Table 2). Efforts to control H5N1 include poultry vaccination programs, culling infected flocks, and strict biosecurity measures. Moreover, antiviral treatments such as oseltamivir have been used in infected individuals, though resistance remains a concern. Given its ongoing evolution, continuous surveillance and research are critical to mitigating the risks associated with H5N1.
## H7N9 viruses
H7N9 is an avian influenza virus that emerged in humans in China in 2013, causing severe respiratory illness with a high mortality rate (49). H7N9 was classified as a low pathogenic avian influenza virus (LPAIV) in birds, meaning infected poultry may show little to no symptoms (50). However, in March 2017, the United States Department of Agriculture (USDA) confirmed the presence of HPAI H7N9 in a broiler chicken breeder flock in the USA (51). Besides, when transmitted to humans, the virus can cause severe pneumonia, ARDS, and multi-organ failure (49). The case fatality rate of H7N9 has varied across outbreaks, with some waves of infection exhibiting mortality rates above 30% (52).
H7N9 predominantly infects the respiratory and gastrointestinal tracts of avian species (43,44). In humans, H7N9 exhibits a preference for α2,3-linked sialic acid receptors, primarily found in the lower respiratory tract, contributing to severe pneumo nia and ARDS (45,46). The virus has shown evidence of limited systemic spread, with detection in blood and other organs. Its ability to bind both avian and human receptors suggests a risk for adaptation and potential human-to-human transmission (53, 54) (Table 2).
Genetic studies have revealed that H7N9 possesses mutations that enhance its binding affinity to human respiratory receptors, increasing concerns about its pandemic potential (55). Vaccination efforts have been implemented in poultry to reduce the virus's circulation, and antiviral treatments, including NA inhibitors, remain the primary therapeutic options for infected individuals (56). Continuous monitoring and research are essential to prevent further zoonotic spillovers and potential adaptation for humanto-human transmission.
## Other subtypes
H1N1 is also known for zoonotic and reassortment potential, infects humans and animals (42). The 2009 pandemic highlighted its adaptability, stressing the need for vaccina tion, biosecurity, and ongoing surveillance to prevent outbreaks (57). H1N1 primarily targets the respiratory epithelium, binding to both α2,3-and α2,6-linked sialic acid receptors, facilitating interspecies transmission (58). In swine, H1N1 replicates efficiently in the intestines, enabling fecal-oral transmission (41). In humans, it infects the upper respiratory tract, leading to mild to severe respiratory illness (42).
H9N2 is a widespread LPAIV that circulates among poultry worldwide (59). In poultry, H9N2 primarily replicates in the respiratory and gastrointestinal tracts, causing mild to moderate disease (47). Human infections are typically mild, with virus replication observed in the upper respiratory tract due to its affinity for both α2,3-and α2,6-linked sialic acid receptors (48). Although it primarily affects birds, sporadic human infections of H9N2 have been reported, particularly in individuals with occupational exposure to infected poultry (60). H9N2 infections in humans typically result in mild respiratory symptoms, but the significance lies in the role as a genetic donor to other zoonotic influenza viruses (61). H9N2 viruses have been implicated in the evolution of highly pathogenic strains, such as H5N1 and H7N9, through genetic reassortment (62,63). The virus possesses internal gene segments that have contributed to the emergence of novel influenza strains with increased human infectivity (64).
While H5, H7, and H9 avian influenza A viruses are most commonly associated with human infections (65), other subtypes pose zoonotic risks. Historical examples include the 1957 H2N2 Asian flu, the 1968 H3N2 Hong Kong flu, and the 1977 Russian flu (66). More recently, an avian-origin H4N6 virus was detected in US pigs in 2015 (67), and H5N6 infections have been reported in children (68). The ability of these viruses to reassort highlights the need for ongoing surveillance. Furthermore, some H10N4 and H10N8 viruses have demonstrated the capacity to bind to human-like receptors (69). H11N9 viruses can be transmitted directly to hunters from ducks (70). These avian influenza subtypes primarily circulate in birds, but human infections occur through exposure to infected birds or contaminated environments (71). Most AVIs exhibit a strong affinity for α2,3-linked receptors, limiting human-to-human transmission but posing a zoonotic threat (58,72). The diverse host tropism of these subtypes underscores the ongoing risk of interspecies transmission and potential pandemics.
Therefore, global surveillance, robust biosecurity measures, and effective vaccination strategies are crucial for minimizing zoonotic risks. A deeper understanding of the genetic and epidemiological characteristics of these viruses is essential for predicting and preparing for future outbreaks.
## FIELD-READY AND POINT-OF-CARE (POC) TESTING FOR HPAI
The progression of avian influenza from low pathogenic to highly pathogenic strains in poultry was first recorded in domestic geese in 1996 (73,74), and the emergence of HPAI viruses of the H5N1 subtype in various animal species presents a potential pandemic risk (75). While qRT-PCR coupled with spin column RNA extraction remains the gold standard for HPAI virus surveillance (76), other influenza virus detection and diagnostic testing methods such as (1) rapid tests for antigen, (2) rapid molecular assays for viral RNA, (3) immunofluorescence assays, (4) viral culture, and (5) serological tests can also be implemented. As the demand for on-site assays grows, point-of-care (POC) testing has become a viable alternative to traditional methods, mainly due to its fast results, ease of use, affordability, and minimal infrastructure requirements. One of the main types of POC testing devices is paper-based technology. Paper-based devices are more suitable for simpler, often single-step reactions and are less expensive, which makes them widely used in areas with limited resources. Among the different paper-based POC optionsdipsticks, LFA, and microfluidic devices, LFA stands out due to its ease of production, simple operation, robustness, and user-friendliness. Lateral flow assay utilizes capillary action coupled with binding or capture elements to detect the target analyte(s) in a strip format. Due to its demonstrated advantages, LFA has garnered significant interest from researchers, inventors, and healthcare professionals, expanding its applications across various POC fields.
Recent research underscores the potential of the clustered regularly interspaced short palindromic repeats (CRISPR) system, the CRISPR sequences and associated nucleases system, as an advanced platform for POC testing (6). These systems have been applied to detect a wide range of targets, including viruses, bacteria, parasites, cancer muta tions, genotypes, and small molecules, positioning CRISPR technology at the forefront of next-generation diagnostics (77,78). In particular, immune-based methods and CRISPR-Cas13a-based detection technologies have garnered significant attention for HPAI (79). CRISPR-based approaches have been successfully integrated into LFA systems at the laboratory scale, demonstrating strong potential for early-stage detection of HPAI during outbreak scenarios.
## Antibody-based LFAs for HPAI
Immuno-based serological tests, such as enzyme-linked immunosorbent assay (ELISA) can detect elicited antibodies, which are a direct result of the host's immune response to infection. Immunoglobulin G (IgG) ELISAs have been used to measure the recognition and binding of antibodies to a wide range of seasonal and avian HA proteins (80). However, these tests may not be reliable, especially during the early stages of infection, due to their detection limit. On the other hand, antigen detections of HA and NA proteins of HPAI viruses via ELISA can be utilized at the initial phase of an outbreak but require specialized equipment, reagents, and technical expertise (81,82).
Rapid and precise diagnosis is one of the key approaches to control HPAI viruses, but the sensitivity of HPAI diagnostic tools has progressively reduced over time due to broad antigenic variations during the evolution of HPAI viruses. To overcome this, Nguyen, Nakaishi (83) designed a rapid LFA detection kit by combining two anti-H5 HA monoclo nal antibodies (mAbs), (1) A64/from Linjudge Flu A/H5 and (2) the novel mAb A32/2, which was generated from clade 2.3.4.4 H5 HPAI. This new approach has improved the sensitivity and specificity of the original Linjudge Flu A/H5, as demonstrated by its successful detection of antigens from swabs and tissue homogenates of naturally and experimentally infected birds with H5N6 HPAIVs from the genetic clade 2.3.4.4 (10 2.2 -10 3.4 TCID 50 /test). No cross-reactivity was observed when a panel of 18 IAV reference strains (H1-H16 subtypes) and two strains of influenza B viruses were tested. Recently, Mata Calidonio, Maddox (75) designed a low-cost paper-based immunoassay that can detect H5N1 HA protein from a more comprehensive set of HPAI-related sample matrices such as sera from humans, sheep, horses, poultry, dairy products (eggs and milk), and wild birds (oral, cloacal, and fecal samples). This study also included a strain belonging to HPAI H5 clade 2.3.4, which still possesses similar antigenic properties to the descendant clade 2.3.4.4 and accounts for most cases of the current outbreak. Functionalized gold nanospheres with α-HA IgG Abs established a direct colorimetric response with a limit of detection (LOD) of 0.16 nM and 1.72 nM in human serum and whole milk, respectively. However, the immunoassay did not perform relatively well when a highly viscous matrix (heavy cream) was tested (Table 3).
The viral internal nucleoprotein (NP) of HPAI has also become a target for diagnos tic development since it is highly conserved among AVIs with less susceptibility to mutations as compared to HA and NA proteins (84,88,89). A porous silica nanoparticlebased chemiluminescent LFA system with signal-amplifiable capability was designed by Lee,Kim (84) to detect NP of HPAI (H5N9). This LFA system was able to detect 20-to 100-fold lower AIV levels from chickens' cloacal and oropharyngeal swab samples than a commercial rapid kit/method with 10 4 (50% egg infective dose; EID 50 )/mL LOD. Its signal-amplifiable sensing probes rely on the optimum pore sizes of silica nanoparticles, such as sizable cavities, where larger biomolecules, such as antibodies, were conjugated onto the outer wall, increasing the chance of binding with HPAI's NPs. However, since NP is concentrated inside the envelope of HPAI viruses, an additional lysis step is needed to facilitate its release (Table 3).
There are commercially available immune-based LFA systems specifically designed for qualitatively detecting H5N1 antigens, such as Avian Influenza Virus H5N1 Antigen Lateral Flow Assay Kit (Elabscience, Houston, TX, USA) for tracheal or cloacal poultry secretions. Another gold immunochromatographic assay, Avian Influenza H5N1 Virus Rapid Test Kit (Abbexa, Cambridge, UK), can also qualitatively detect H5N1 antigens from the stool and saliva of chickens. Cows that are infected by H5N1 often display clinical symptoms such as respiratory distress, decreased appetite, altered stool consistency, reduced and anomalous milk production (90). Milk samples can be tested for H5N1 antigen using the AIV-H5N 1 Antigen Rapid Test Kit (Reagen, San Diego, CA, USA) in conjunction with auxiliary diagnosis of infection. Furthermore, the USDA approved one commercial product FluDETECT Avian (Zoetis, Parsippany, NJ, USA), for chicken and turkey. These conventional commercial immune-based LFA systems generate line signals when the target antigens are more abundant than the LOD. However, non-specific binding can be caused by sample matrix components such as proteins and salts that may adhere to the nitrocellulose membrane or Abs. Additionally, fecal and tissue lysates may clog the membranes and create uneven flow, leading to artifact formation (Table 4).
## CRISPR-based LFA for avian influenza
The CRISPR system is a powerful gene-editing technology derived from a bacterial immune defense mechanism that uses RNA-guided enzymes, Cas proteins, to precisely recognize and cut specific DNA or RNA sequences (91). When integrated into LFAs, CRISPR-based platforms use guide RNA to identify viral or bacterial sequences, activat ing Cas enzymes that produce a detectable signal on the test strip (92). This combi nation allows rapid, sensitive, and portable detection of pathogens at the point of care, enhancing traditional LFA capabilities with molecular-level precision. Detecting DNA and RNA from various environmental sources (e.g., water, feces, etc.) is a promis ing approach to monitor potential wildlife pathogens with limited disturbance to the organisms or their habitats. Minimally invasive methods like swabbing or brushing animals and noninvasive methods such as environmental DNA sampling show great potential for surveying various species, including hidden invasive and at-risk species (93,94). CRISPR-based diagnostics are powerful tools that can successfully detect unique DNA and RNA sequences from associated target pathogens; however, their applications in wildlife disease management have been sluggish (94). LFA has been investigated as a compatible platform for integrating CRISPR-Cas13a technology into field-ready/POC diagnostics and transforming its results into visible and easy-to-interpret signals.
Li et al. ( 79) designed a recombinase-aided amplification (RAA) coupled with CRISPR-Cas13a and LFA system for detecting H5 avian influenza virus from clinical samples. The methods achieved a LOD of 0.1 copy/μL, which was in line with qRT-PCR. No cross-reactivity with other AVIs (H3, H7, H9, and H10), was observed. Similarly, Yang, Yang (85) developed two rapid detection methods for avian influenza virus based on CRISPR-Cas13a. These methods detect avian influenza virus through the M gene and identify H5, H7, and H9 subtypes via the HA gene. The first uses reverse transcription (RT)-RAA with Cas13a for amplification and qRT-PCR, achieving a LOD of 1 copy/μL. The second technology combines reverse transcription recombinase-aided amplification (RT-RAA) with Cas13a and a LFA system, achieving a LOD of 10 copies/μL. Both methods show high sensitivity and specificity, potentially surpassing qRT-PCR for clinical use. A very similar approach was implemented by Zhang, Wang (86), which tested additional non-targets such as infectious bronchitis virus, infectious laryngotracheitis virus, and Newcastle disease virus, wherein no cross-reactivity was reported. Moreover, clinical samples (e.g., throat swabs) from individual chickens presenting mild or severe respiratory symptoms associated with avian influenza infection were also tested for avian influenza virus H5, H7, and H9 using RT-RAA-LF dipstick. They showed a LOD of 10 1 copies/µL -consistent with the results of real-time fluorescence qRT-PCR (87), but this approach was employed, excluding the CRISPR-Cas13a method (Table 3). For the centrifuge-free H5Nx avian influenza detection method, Song et al. ( 76) combined magnetic bead-based RNP purification using anti-NP antibodies with a PAM-independent CRISPR-Cas12a system, enabled by DMSO-mediated amplicon denaturation. This eliminates spin columns and specific PAM sequences, increasing flexibility. The assay detected H5Nx with high specificity, achieving sensitivities of 10 1 EID 50 (fluorescence) and 10 2 EID 50 (lateral flow), and clinical sample sensitivity of 80% and specificity of 100%. This approach offers a promising on-site diagnostic tool for rapidly mutating RNA viruses (Table 3).
Pre-amplification method (e.g., RAA) for nucleic acids increases the overall sensitivity of CRISPR-Cas even in the presence of PCR interferents. By combining CRISPR's ability to precisely target specific nucleic acids with the user-friendliness and speed of LFA, the resulting platform offers accurate, fast, cost-effective, and portable testing for HPAI and a wide range of disease-causing viral agents. Their combination excels in speed, sensitivity, and accessibility and is highly adaptable for both field-ready and point-of-care diagnostics.
While CRISPR-based LFAs for HPAI are not yet commercially available on the market, active research investigations are currently exploring this technology, and it may only be a matter of time before these highly sensitive and rapid tests are deployed for detecting HPAI.
## CHALLENGES AND LIMITATIONS OF LFA-BASED TECHNOLOGY FOR AVIAN INFLUENZA DETECTION
LFA-based technology has been widely adopted for virus detection due to its rapid results, ease of use, and cost-effectiveness (19). However, several challenges and limitations hinder its broader application in disease surveillance and outbreak control. A primary concern with conventional antibody-based LFAs is their suboptimal sensitiv ity and specificity, especially when detecting low-abundance analytes or distinguish ing closely related viral targets (95). These limitations are particularly problematic in early-stage infections, where low viral loads can result in false negatives, and cross-reac tivity with other influenza subtypes may lead to false positives (96,97). Furthermore, environmental factors such as temperature and humidity may impact test performance, reducing reproducibility in field conditions (98).
Another challenge is the quantitative limitation of traditional LFAs. Most LFA-based tests provide qualitative or semi-quantitative results, making it difficult to assess viral load, which is crucial for determining infection severity and guiding disease manage ment strategies (99). Additionally, current LFA technologies often struggle to differentiate between highly pathogenic and low pathogenic avian influenza strains, which is vital for disease control measures (100). Determination between highly pathogenic and low pathogenic avian influenza A virus still relies on nucleotide sequencing or mass spectrometry (101,102). To address these challenges, recent advances in POCT have introduced promising innovations (103). CRISPR-based LFAs, in particular, offer improved sensitivity, specif icity, and adaptability, helping to overcome the shortcomings of traditional assays (104). Future POCT developments may incorporate nanomaterials, smartphone-based analysis, and CRISPR-Cas technology to enhance sensitivity, specificity, and quantita tive capabilities. Integration with internet of Things and artificial intelligence-driven diagnostics can also improve real-time data analysis and epidemiological tracking (105)(106)(107).
While LFA-based technology remains a valuable tool for avian influenza detection, addressing its current limitations through advanced POCT innovations will enhance its effectiveness in disease surveillance and outbreak management. Future research should focus on improving detection accuracy, quantification capabilities, and real-time connectivity to strengthen global avian influenza control efforts.
In the regulatory context, point-of-care pathogen diagnostics used in veterinary settings are classified as veterinary biologics and fall under the jurisdiction of the USDA's Center for Veterinary Biologics (108). These products must meet specific sensitivity and specificity criteria to be approved for distribution. However, reference laboratory assays, including those for biomarkers, hormones, complete blood counts, and clinical chemistry panels, are currently not regulated by the USDA (109). Furthermore, there appears to be no standardized regulatory framework for animal disease detection tools, highlighting the need for clearer guidelines as diagnostic technologies evolve.
## CONCLUSION
LFAs have emerged as valuable tools for the rapid detection of avian influenza, providing a cost-effective and user-friendly alternative to conventional diagnostic methods. However, their broader utility in disease surveillance is currently limited by sensitiv ity, quantification, and strain differentiation challenges. Emerging innovations, such as CRISPR-based detection systems, smartphone-integrated platforms, and AI-enabled data analysis, hold promise for overcoming these limitations. Ongoing research and technological advancements continue to enhance the performance of LFAs, improving their reliability for field applications. Future efforts should focus on increasing sensitiv ity, expanding multiplexing capabilities, and integrating LFAs with digital platforms for real-time data sharing and surveillance.
## References
1. Cdc (2024)
2. Usda (2025) "Avian Influenza"
3. Abdelwhab, Mettenleiter (2023) "Zoonotic animal influenza virus and potential mixing vessel hosts" *Viruses*
4. Mj, Hurtado, Shepherd et al. (2016) "Pathogenicity and transmission of H5 and H7 highly pathogenic avian influenza viruses in mallards" *J Virol*
5. Duan, Li, Ren et al. (2023) "An overview of avian influenza surveillance strategies and modes" *Sci One Health*
6. Zhou, Wang, Ma et al. (2024) "On-site and visual detection of the H5 subtype avian influenza virus based on RT-RPA and CRISPR/Cas12a" *Viruses*
7. Wan (2012) "Lessons from emergence of A/goose/Guangdong/1996like H5N1 highly pathogenic avian influenza viruses and recent influenza surveillance efforts in southern China" *Zoonoses Public Health*
8. Castillo, Fasce, Parra et al. (2023) "The first case of human infection with H5N1 avian Influenza A virus in Chile" *J Travel Med*
9. Rijks, Hesselink, Lollinga et al. (2021) "Highly Pathogenic Avian Influenza A(H5N1) Virus in Wild Red Foxes, the Netherlands"
10. Cdc (2020) "Highlights in the History of Avian Influenza"
11. Usda (2025) "Confirmations of Highly Pathogenic Avian Influenza in Commercial and Backyard Flocks"
12. Cidrap (2023) "Cambodia Reports 2 H5N1 Avian Flu Cases, 1 Fatal"
13. Who (2024) "Avian Influenza A(H5N1) -United States of America"
14. Ecdc (2024) "Avian Influenza Overview"
15. Reuters (2025) "Global Avian Influenza Outbreaks Give Brazil Chicken Exports an Edge"
16. Mirzaei, Shoushtari, Nouri (2020) "Development and evaluation of real-time reverse transcription polymerase chain reaction test for quantitative and qualitative recognition of H5 subtype of avian influenza viruses" *Arch Razi Inst*
17. Fu, Wang, Ma et al. (2023) "Advances in detection techniques for the H5N1 avian influenza virus" *IJMS*
18. Caffry, Selby, Barr et al. (2023) "The QuantuMDx Q-POC SARS-CoV-2 RT-PCR assay for rapid detection of COVID-19 at point-of-care: preliminary evaluation of a novel technology" *Sci Rep*
19. Ince, Sezgintürk (2022) "Lateral flow assays for viruses diagnosis: Up-to-date technology and future prospects" *Trends Analyt Chem*
20. Wang, Kim, Kang et al. (2024) "Development and evaluation of two rapid lateral flow assays for on-site detection of African swine fever virus" *Front Microbiol*
21. Koczula, Gallotta (2016) "Lateral flow assays" *Essays Biochem*
22. Bouvier, Palese (2008) "The biology of influenza viruses" *Vaccine (Auckl)*
23. Kosik, Yewdell (2019) "Influenza hemagglutinin and neuraminidase: Yin-Yang proteins coevolving to thwart immunity" *Viruses*
24. Kessler, Harder, Schwemmle et al. (2021) "Influenza A viruses and zoonotic events-are we creating our own reservoirs?" *Viruses*
25. Wang, Song, Li et al. (2019) "Different subtypes of influenza viruses target different human proteins and pathways leading to different pathogenic phenotypes" *Biomed Res Int*
26. Wu, Wilson (2020) "Influenza hemagglutinin structures and antibody recognition" *Cold Spring Harb Perspect Med*
27. Health (2025) "Preparedness, prevention and control related to zoonotic avian influenza" *EFSA J*
28. Abubakar (2023) "Avian influenza virus tropism in humans" *Viruses*
29. Garg, Reinhart, Couture et al. (2025) "Highly pathogenic avian influenza A(H5N1) virus infections in humans" *N Engl J Med*
30. Cdc (2024) "Highly Pathogenic Avian Influenza A(H5N1) Virus: Interim Recommendations for Prevention, Monitoring, and Public Health Investigations"
31. Cdc (2025) "H5 Bird Flu Response: Focus Areas for Ongoing Public Health Risk Assessment"
32. Chakraborty, Bhattacharya (2024) "Evolution and mutational landscape of highly pathogenic avian influenza strain A(H5N1) in the Minireview Journal of Virology November"
33. "the USA and global landscape"
34. Peacock, Moncla, Dudas et al. (2025) "The global H5N1 influenza panzootic in mammals" *Nature*
35. Hermann, Krammer (2002) "Clade 2.3.4.4b H5N1 neuraminidase has a long stalk, which is in contrast to most highly pathogenic H5N1 viruses circulating between" *mBio*
36. Neumann (2015) "H5N1 influenza virulence, pathogenicity and transmissibility: what do we know" *Future Virol*
37. Nelli, Harm, Siepker et al. (2024) "Sialic acid receptor specificity in mammary gland of dairy cattle infected with highly pathogenic avian influenza A(H5N1) virus" *Emerg Infect Dis*
38. Murawski, Fabrizio, Ossiboff et al. (2024) "Highly pathogenic avian influenza A(H5N1) virus in a common bottlenose dolphin (Tursiops truncatus) in Florida" *Commun Biol*
39. Bordes, Vreman, Heutink et al. (2023) "Highly pathogenic avian influenza H5N1 virus infections in wild red foxes (Vulpes vulpes) show neurotropism and adaptive virus mutations" *Microbiol Spectr*
40. Peiris, Jong, Guan (2007) "Avian influenza virus (H5N1): a threat to human health" *Clin Microbiol Rev*
41. Carrasco, Gröne, Van Den Brand et al. (2024) "The mammary glands of cows abundantly display receptors for circulating avian H5 viruses" *J Virol*
42. Koçer, Obenauer, Zaraket et al. (2013) "Fecal influenza in mammals: selection of novel variants" *J Virol*
43. Jilani (2025) "H1N1 Influenza"
44. Jones, Sonnberg, Webby et al. (2015) "Influenza A(H7N9) virus transmission between finches and poultry"
45. Wiwanitkit (2013) "influenza: the emerging infectious disease" *N Am J Med Sci*
46. Li, Weng, Lan et al. (2018) "Comparison of patients with avian influenza A (H7N9) and influenza A (H1N1) complicated by acute respiratory distress syndrome" *Medicine (Baltimore)*
47. Ramos, Krammer, Hai et al. (2013) "H7N9 influenza viruses interact preferentially with α2,3-linked sialic acids and bind weakly to α2,6linked sialic acids" *J Gen Virol*
48. Oladokun, Alizadeh, Mallick et al. (2025) "Influenza a virus subtype H9N2 infection induces respiratory microbiota dysbiosis in chickens via type-I interferonmediated mechanisms" *FEMS Microbes*
49. Pusch, Suarez (2018) "The multifaceted zoonotic risk of H9N2 avian influenza" *Vet Sci*
50. Li, Zhou, Zhou et al. (2014) "Epidemiology of human infections with avian influenza A(H7N9) virus in China" *N Engl J Med*
51. Ecdc (2023) "Factsheet on A(H7N9"
52. Bertran, Lee, Criado et al. (2018) "Pathobiology of Tennessee 2017 H7N9 low and high pathogenicity avian influenza viruses in commercial broiler breeders and specific pathogen free layer chickens" *Vet Res*
53. Tanner, Toth, Gundlapalli (2015) "The pandemic potential of avian influenza A(H7N9) virus: a review"
54. Bisset, Hoyne (2020) "Evolution and adaptation of the avian H7N9 virus into the human host"
55. Xu, Peng, Zhang et al. (2019) "Avian-to-human receptorbinding adaptation of avian H7N9 influenza virus hemagglutinin" *Cell Rep*
56. Huang, Wang (2020) "The effects of genetic variation on H7N9 avian influenza virus pathogenicity" *Viruses*
57. Sahly, Anderson, Jackson et al. (2025) "Antineuraminidase and anti-hemagglutinin stalk responses to different influenza a" *Vaccine (Auckl)*
58. Sullivan, Jacobson, Dowdle et al. (2010) "2009 H1N1 influenza" *Mayo Clin Proc*
59. Liu, Huang, Smits et al. (2022) "Human-type sialic acid receptors contribute to avian influenza A virus binding and entry by hetero-multivalent interactions" *Nat Commun*
60. Peacock, James, Sealy et al. (2019) "A global perspective on H9N2 avian influenza virus"
61. Cdc (2024) "Vietnam Reports First Human Infection with Avian Influenza H9N2 Virus"
62. Carnaccini, Perez (2020) "H9 influenza viruses: an emerging challenge" *Cold Spring Harb Perspect Med*
63. Arai, Ibrahim, Elgendy et al. (2019) "Genetic compatibility of reassortants between avian H5N1 and H9N2 influenza viruses with higher pathogenicity in mammals" *J Virol*
64. Pu, Yin, Liu et al. (2021) "Reassortment with dominant chicken H9N2 influenza virus contributed to the fifth H7N9 virus human epidemic"
65. James, Bhat, Walsh et al. (2022) "The origin of internal genes contributes to the replication and transmission fitness of H7N9 avian influenza virus" *J Virol*
66. Cdc (2024) "Avian Influenza Type A"
67. Gagnon, Acosta, Hallman et al. (2018) "Pandemic paradox: early life H2N2 pandemic influenza infection enhanced susceptibility to death during the 2009 H1N1 pandemic" *mBio*
68. Abente, Gauger, Walia et al. (2017) "Detection and characterization of an H4N6 avianlineage influenza A virus in pigs in the Midwestern United States" *Virology (Auckl)*
69. Yu, Hou, Feng et al. (2021) "Emergence of a young case infected with avian influenza A (H5N6) in Anhui Province, East China during the COVID-19 pandemic" *J Med Virol*
70. Wang, Wang, Zhang et al. (2022) "Emergence, evolution, and biological characteristics of H10N4 and H10N8 avian influenza viruses in migratory wild birds detected in Eastern China in 2020"
71. Ge, Yao, Wang et al. (2019) "Detection of reassortant avian influenza A (H11N9) virus in wild birds in China"
72. Tripathi, Sendor, Sapra (2025) "Avian Influenza"
73. Kristensen, Larsen, Trebbien et al. (2024) "The avian influenza A virus receptor SA-α2,3-Gal is expressed in the porcine nasal mucosa sustaining the pig as a mixing vessel for new influenza viruses" *Virus Res*
74. Xiao, Oong, Chen et al. (2025) "Reverse genetics-derived cattle H5N1 virus from Clade 2.3.4.4b shows enhanced systemic infectivity and pathogenicity than an older Clade 1 H5N1 virus in BALB/c mice" *Emerging Microbes & Infections*
75. Xu, Cox, Guo (1999) "Genetic characterization of the pathogenic influenza A/Goose/Guangdong/1/96 (H5N1) virus: similarity of its hemagglutinin gene to those of H5N1 viruses from the 1997 outbreaks in Hong Kong" *Virology (Auckl)*
76. Mata Calidonio, Maddox, Patel et al. (2025) "Development of an immunoassay for highly pathogenic avian influenza (H5N1) across diverse sample matrices" *ACS Nanosci Au*
77. Song, Son, Kim et al. (2025) "Rapid and specific on-site H5Nx avian influenza diagnosis via RPA and PAM-independent CRISPR-Cas12a assay combined with anti-NP antibody-based viral RNA purification" *Front Vet Sci*
78. Kaminski, Abudayyeh, Gootenberg et al. (2021) "CRISPR-based diagnostics" *Nat Biomed Eng*
79. Mustafa, Makhawi (2021) "SHERLOCK and DETECTR: CRISPR-cas systems as potential rapid diagnostic tools for emerging infectious diseases" *J Clin Microbiol*
80. Li, Shang, Luo et al. (2023) "Rapid detection of H5 subtype avian influenza virus using CRISPR Cas13a based-lateral flow dipstick" *Front Microbiol*
81. Klim, William, Mellors et al. (2024) "Serological analysis in humans in Malaysian Borneo suggests prior exposure to H5 avian influenza near migratory shorebird habitats" *Nat Commun*
82. Mainenti, Siepker, Magstadt et al. (2025) "Distribution of lesions and detection of influenza A(H5N1) virus, clade 2.3.4.4b, in ante-and postmortem samples from naturally infected domestic cats on U.S. dairy farms" *J Vet Diagn Invest*
83. Xiao, Yang, Liu et al. (2021) "Development of an antigen-ELISA and a colloidal gold-based immunochromato graphic strip based on monoclonal antibodies for detection of avian influenza A(H5) viruses" *J Vet Diagn Invest*
84. (1177)
85. Nguyen, Nakaishi, Motojima et al. (2017) "Rapid and broad detection of H5 hemagglutinin by an immunochromatographic kit using novel monoclonal antibody against highly pathogenic avian influenza virus belonging to the genetic clade 2" *PLoS One*
86. Lee, Kim, Lee et al. (2020) "A selfcalibrating electrochemical aptasensing platform: Correcting external interference errors for the reliable and stable detection of avian influenza viruses" *Biosens Bioelectron*
87. Yang, Yang, Zhang et al. (2025) "Rapid detection of pan-avian influenza virus and H5, H7, H9 subtypes of avian influenza virus using CRISPR/Cas13a and lateral flow assay" *Poult Sci*
88. Zongshu, Wang, Chen et al. (2024) "Based on CRISPR-Cas13a system, to establish a rapid visual detection method for avian influenza viruses" *Front Vet Sci*
89. Zhang (2024) "Detection method for reverse transcription recombinase-aided amplification of avian influenza virus subtypes H5, H7, and H9" *BMC Vet Res*
90. Greco, Ravenswater, Ruiz-Raya et al. (2025) "Asymptomatic infection and antibody prevalence to co-occurring avian influenza viruses vary substantially between sympatric seabird species following H5N1 outbreaks" *Sci Rep*
91. Myers, Conlon, Gallagher et al. (2025) "Analysis of polyclonal and monoclonal antibody to the influenza virus nucleoprotein in different oligomeric states" *Virus Res*
92. Caserta, Frye, Butt et al. (2024) "Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle" *Nature*
93. Saeed, Ayub, Durrani et al. (2025) "CRISPR Cas systems: From bacterial defense mechanisms to revolutionary tools reshaping genetic research and translation therapeutics" *The Microbe*
94. Zhou, Li, Olajide et al. (2024) "CRISPR/Cas-based nucleic acid detection strategies: Trends and challenges"
95. Duarte, Simões, Costa (2023) "Current status and topical issues on the use of eDNA-based targeted detection of rare animal species" *Sci Total Environ*
96. Pérez, Vazquez-Meves, Hunter (2024) "Early Detection of Wildlife Disease Pathogens Using CRISPR-Cas System Methods" *CRISPR J*
97. Abu, Bakhori, Shueb (2023) "Lateral flow assay for hepatitis B detection: a review of current and new assays" *Micromachines (Basel)*
98. Soomro, Shah, Saboor et al. (2021) "Sensitivity and specificity of hepatitis B virus screening via rapid immunoassay chromatographic test" *Cureus*
99. Sadeghi, Shirazi, Dehbashi et al. (2024) "Development of a rapid LFA test based on direct RT-LAMP for diagnosis of SARS-CoV-2" *Pract Lab Med*
100. Carrell, Kava, Nguyen et al. (2018) "Beyond the lateral flow assay: A review of paper-based microfluidics" *Microelectron Eng*
101. Li, Chen, Wang (2019) "Lateral flow assay ruler for quantitative and rapid point-of-care testing" *Analyst*
102. Payungporn, Chutinimitkul, Chaisingh et al. (2006) "Discrimination between highly pathogenic and low pathogenic H5 avian influenza A viruses" *Emerg Infect Dis*
103. Filaire, Sécula, Bessière et al. (2024) "High and low pathogenicity avian influenza virus discrimina tion and prediction based on volatile organic compounds signature by SIFT-MS: a proof-of-concept study" *Sci Rep*
104. Spronken, Funk, Gultyaev et al. (2024) "Nucleotide sequence as key determinant driving Minireview Journal of Virology November"
105. "A virus hemagglutinin cleavage sites"
106. Vealan, Joseph, Alimat et al. (2023) "Lateral flow assay: a promising rapid point-of-care testing tool for infections and non-communicable diseases" *Asian Biomed (Res Rev News)*
107. Sun, Zhong, Nie et al. (2025) "State-of-the-art in food safety: the potential of CRISPR/Cas-based lateral flow assay technology in the field of food analysis" *Food Safety and Health*
108. Hassan, Mohamed, Hassan et al. (2025) "Recent developments and future directions in point-of-care next-generation CRISPR-based rapid diagnosis" *Clin Exp Med*
109. Moulahoum, Ghorbanizamani, Beduk et al. (2023) "Emerging trends in nanomaterial design for the development of point-of-care platforms and practical applications" *J Pharm Biomed Anal*
110. Nguyen (2022) "An internet of things-based point-of-care device for direct reverse-transcription-loop mediated isothermal amplification to identify SARS-CoV-2" *Biosens Bioelectron*
111. Dzanis, Greenwald (2025) "Veterinary products"
112. (2025) "#:~:text=Veterinar y%20diagnostics%20that%20are%20point" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645976&blobtype=pdf | # ARGLU1 is a negative regulator of the adenoviral replicative cycle
Amit Koul, Lauren Fulham, Nikolas Akkerman, Drayson Graves, Esha Kaul, Khadija Khadija, Peter Pelka
## Abstract
We have previously reported that human adenovirus E1A interacts with ARGLU1, a small disordered cellular protein. The consequences of this interaction for virus replication were unclear. E1A is the first protein produced during adenovirus infection. Via protein-protein interactions, E1A modifies the cellular environment to create optimal conditions for viral replication. To better understand the molecular mechanisms driving viral infection, we further investigated the functional consequences of the interaction between ARGLU1 and E1A. ARGLU1 interacts with E1A directly as determined by a GST-pulldown assay with recombinant proteins. Importantly, ARGLU1 was found to act as a transcriptional repressor when it was localized to viral promoters. Repression was driven by enhanced promoter-proximal RNA polymerase II pausing. Significantly, ARGLU1-induced repression of viral promoters is likely an unintended consequence of ARGLU1 binding to E1A, as a mutant of E1A (dl1102), unable to bind to ARGLU1, did not show reduced viral gene expression. Furthermore, ARGLU1 was found to colocalize with E1A in infected cell nuclei. Finally, the binding of E1A to ARGLU1 appears to reduce DNA damage repair in bleomycin-treated and adenovirus-infected cells, suggesting that E1A binding to ARGLU1 is important for DNA damage response inhibition. Overall, this study underscores the complexity of virus-host interactions and reveals a novel role for ARGLU1 during adenoviral infection.
IMPORTANCEThis study uncovers a novel antiviral role for ARGLU1, known for its involvement in splicing, transcriptional regulation, and DNA damage repair. Our study demonstrates that ARGLU1 functions as a viral restriction factor by reducing virus growth through the inhibition of viral gene expression and enhanced DNA damage response, leading to reduced viral growth. These findings not only highlight an important role for ARGLU1 in host antiviral mechanisms but also emphasize the utility of adenovirus as a tool to uncover critical cellular pathways. KEYWORDS ARGLU1, human adenovirus, E1A, transcription, DNA damage response A rginine and glutamate-rich protein 1 (ARGLU1) is a small intrinsically disordered protein (IDP) involved in regulating transcription, RNA processing, and DNA damage repair (1-3). Despite its disordered nature, ARGLU1 is evolutionarily conserved and features two distinctly charged regions-a positively charged arginine-rich N-termi nus and a negatively charged C-terminus. These regions enable interactions with the spliceosome and nuclear receptors, respectively (2, 4). Initially, ARGLU1 was identified as a co-regulator of estrogen receptor-mediated transcription through its interaction with mediator subunit 1 (MED1), promoting breast cancer cell growth (1). Subsequent research revealed its role as a modulator of glucocorticoid signaling in the central nervous system, where its deletion significantly alters gene expression and alterna tive splicing linked to neurogenesis (2). Moreover, loss of ARGLU1 function has been associated with global splicing defects and neuronal deficiencies (5). Recent findings
have extended the role of ARGLU1 in viral gene regulation and DNA damage repair (3). Knockdown of ARGLU1 enhances viral gene expression and growth, whereas its presence supports the DNA damage repair response (3). Mechanistically, ARGLU1 impedes transcription elongation by promoting promoter-proximal pausing of RNA polymerase II (RPII). This occurs through its binding to Jumonji-domain-containing protein 6 (JMJD6), which displaces Bromodomain-containing protein 4 (BRD4), a key partner required for transcriptional elongation (3). By stabilizing RPII pausing, ARGLU1 enhances resistance to genotoxic drugs and facilitates DNA repair processes (3). RNA polymerase II pausing is a regulatory step in gene regulation and productive RNA synthesis (6). DNA lesions in the template strand cause stalling of RNA polymerase II, leading to genome-wide transcription arrest until the damage is mended via transcrip tion-coupled repair mechanisms (7). Together, these findings suggest that ARGLU1 plays a diverse role in multiple cellular processes, a common feature of IDPs in cells (8).
Human adenovirus (HAdV) is a small, nonenveloped DNA virus with a double-stran ded linear genome that primarily infects terminally differentiated epithelial cells (9). As the first viral protein expressed during infection, the immediate early gene product Early protein 1A (E1A) plays a central role in activating the viral transcriptional program and reprogramming the host cell environment to facilitate viral replication (10,11). E1A achieves this by interacting with a wide array of cellular proteins, including cell cycle regulators, transcription factors, chromatin remodelers, and transcriptional coregulators, thereby disrupting normal cellular homeostasis and driving viral gene expression (10,11). In addition to its pivotal role in viral replication, E1A has been effectively employed as a molecular probe to uncover key regulators of cellular pathways, many of which are implicated in cancer biology, for example, FUBP1, Nek9, DREF, Ku70, and RuvBL1 (12)(13)(14)(15)(16). Notably, E1A was recently found to bind ARGLU1 in HAdV-B7-infected cells through the N-terminus of E1A, with unknown consequences for the adenoviral replicative cycle (3). Nevertheless, these findings add to the growing understanding of how E1A manipulates host cell machinery and highlight the broader utility of HAdV in dissecting complex cellular regulatory networks (9,17,18).
Adenoviruses have evolved multiple strategies to subvert the host DNA damage repair (DDR) response, a key genome integrity-maintaining mechanism that is deleteri ous to the replication of many viruses (19). The presence of viral genomes will trigger DDR in the infected cells; to overcome this, HAdVs have adapted to evade or inhibit these responses to ensure efficient replication. Interference with the activities of the Mre11-Rad50-Nbs1 (MRN) complex-a central sensor consisting of three subunits, Mre11, Rad50, and Nbs-1-is crucial for efficient viral replication (20)(21)(22). Additionally, adenoviral protein E1B-55K, together with E4orf6, promotes the degradation of DDR components, including Tip60 (23), an ATM activator, thereby suppressing DDR signaling. Structurally, adenoviral genomes and replicating viral genomes resemble DNA doublestranded breaks (DSBs) due to their linear nature, but the presence of a covalently attached terminal protein (TP) at the 5' ends protects them from recognition by DNA break sensors (19). Similarly, replication intermediates containing extensive single-stran ded DNA could potentially trigger ATR signaling; however, they are shielded by the adenoviral DNA-binding protein (DBP) (19). Moreover, wild-type HAdV genomes do not undergo concatemerization (a process commonly associated with unprotected DNA ends), suggesting that adenoviral proteins actively suppress the DDR machinery (19,24). Despite these suppression strategies, the DDR can still be inadvertently activated. For example, the covalent linkage of TP to the viral DNA may be recognized as a DNAprotein crosslink (DPC), triggering DPC repair pathways (21), exposing the adenoviral genome and activation of DDR. This can lead to TP removal and consequently impair viral genome integrity and replication efficiency.
In the present study, we investigated the functional significance of the interaction between ARGLU1 and E1A. We demonstrate that E1A and ARGLU1 interact directly via a GST-pulldown assay with bacterially expressed proteins. Overexpression of ARGLU1 resulted in a significant reduction of viral gene expression in cells infected with the phenotypically wild-type HAdV5 dl309, but not in cells infected with the HAdV5 mutant dl1102, which lacks the ability to bind ARGLU1. Consistently, ARGLU1 overexpression also led to significant changes in adenoviral protein expression. Although viral genome copy numbers were not significantly different between dl309 and dl1102-infected HT1080 (HT) and HT1080 cells overexpressing ARGLU1 (HT-A) cells, ARGLU1 overexpression substantially impaired viral growth. Additionally, chromatin immunoprecipitation (ChIP) further demonstrated that ARGLU1 is recruited to adenoviral promoters in dl309-infec ted HT-A cells. Importantly, comet assays revealed reduced DNA damage repair in dl309-infected cells, suggesting that disruption of DNA damage repair-related ARGLU1 activities is important for viral replication.
## MATERIALS AND METHODS
## Antibodies
Mouse monoclonal anti-E1A M2, M37, M58, and M73 antibodies were previously described (25) and were grown in-house and used as the hybridoma supernatant. For immunoprecipitations (IPs), 25 µL was used, and for western blot assays, a dilution of 1:400 was used. 12CA5 anti-hemagglutinin (anti-HA) mouse monoclonal antibody was grown in-house and used as hybridoma supernatant; 25 µL of hybridoma supernatant was used in chromatin immunoprecipitation (ChIP) experiments. Mouse monoclonal anti-72k DNA-binding protein (DBP) antibody was previously described (26) and was used at a dilution of 1:400 for western blotting. Viral structural and late proteins were detected with anti-adenovirus type five antibody from Abcam (cat. # ab6982) at a dilution of 1:5,000. Rat monoclonal anti-hemagglutinin (anti-HA) clone 3F10 was obtained from MilliporeSigma (cat. #11867423001). Anti-RNA polymerase II (CTD) was obtained from Abcam (cat. # ab252854). Antibody for human ARGLU1 was purchased from Invitrogen (cat. # PA5-66041); additionally, antibody for human ARGLU1 was also generated by Pacific Immunology using the following peptide sequence: Cys-KEEQ KIILGKGKSRPKLSFSLKTQD. The antibody was affinity-purified using peptide columns before use. Antibodies for ATM (cat. # 2873), phospho-serine 1981-ATM (cat. # 13050), ATR (cat. # 13934), phospho-threonine 1989-ATR (cat. # 30632), Chk1 (cat. # 2360), phospho-serine 345-Chk1 (cat. # 2348), p53 (cat. # 9282), and p21 (cat. # 2947) were purchased from Cell Signaling Technologies. Anti-53BP1 antibody (cat. # ab175933) was purchased from Abcam. Secondary antibodies were purchased from Jackson ImmunoRe search.
## Cell and virus culture
HT1080 cells (ATCC# CCL-121) were cultured in Dulbecco's modified Eagle's medium (MilliporeSigma) supplemented with 5% fetal bovine serum, streptomycin, and penicillin (Corning). Unless stated otherwise, all viral infections were performed at a multiplicity of infection (MOI) of 10 in serum-free medium for 1 h, after which fresh complete medium was added without removing the infection medium. HT1080 cells overexpress ing ARGLU1 (HT-A) were generated as previously described (3). HAdV5 dl309 was previously described (27), as was dl1102 mutant (28).
## Protein purification and GST pulldown assay
Human ARGLU1 was expressed as a GST-tagged fusion protein (GST-ARGLU1) in E. coli BL21 (DE3)-CodonPlus RIPL (Agilent Technologies). Cells were lysed on an EmulsiFlex C3 (Avestin) in Buffer-A (20 mM HEPES (pH 7.4), 300 mM NaCl, 0.1 mM EDTA, 10% (vol/vol) glycerol, 2 mM DTT, and 1% Triton X-100). Crude lysates were clarified by centrifugation at 35,000 × g for 30 min, at 4°C. Clarified lysates were affinity-purified using glutathione Sepharose (Thermo Fisher Scientific), followed by elution with 40 mM reduced gluta thione. The eluted protein was dialyzed in Buffer-B (20 mM HEPES (pH 7.4), 100 mM NaCl, 0.1 mM EDTA, 5% (vol/vol) glycerol, 2 mM DTT) overnight at 4°C. Protein was concentrated to 1 mg/mL, concentration was determined via Bradford assay, and protein was stored at 4°C. HAdV5 E1A289R was expressed as a 6xHis-tagged fusion protein (6xHis-E1A289R) in BL21 (DE3)-CodonPlus RIPL (Agilent Technologies) and purified with an Ni-NTA affinity column as described previously (26,29). Recombinant 6xHis-E1A289R was concentrated to 1 mg/mL, concentration was determined via Bradford assay, and protein was stored at -20°C. GST pulldowns were performed as described previously (29,30).
## Western blot
Protein samples were boiled in sample buffer containing 100 mM DTT at 95°C for 5 min. Samples were resolved on a Bis-Tris Bolt Plus 4%-12% protein gel (Invitrogen) using either MOPS or MES buffer (Invitrogen), depending on the size of the target protein. Following electrophoresis, proteins were transferred to a PVDF membrane using the eBlot L1 blot transfer apparatus (Genscript) using the default protocol. Membranes were then blocked for 1 h in 5% skim milk powder dissolved in TBST. Primary antibodies were applied in 1% BSA or 5% skim milk powder in TBST, shaking, overnight at 4°C. Horse radish peroxidase (HRP)-conjugated secondary antibodies (Jackson ImmunoResearch) were applied in blocking buffer at a 1:100,000 dilution. Protein bands were visualized using the Azure C600 digital imager (Azure Biosystems) with Luminata Forte ECL reagent (MilliporeSigma).
For the phosphoprotein blot, the lysis buffer used was the NP-40 lysis buffer supplemented with 10 mM sodium phosphate, 10 mM sodium orthovanadate, and 10 mM beta-glycerophosphate; protein samples were resolved on a Bis-Tris Bolt Plus 4%-12% protein gel (Invitrogen) using MOPS (Invitrogen); and a western blot was performed as above.
## Chromatin immunoprecipitation
The formaldehyde cross-linking and chromatin immunoprecipitation (ChIP) were carried out as described previously (3,31). Cells were infected with HAdV5 strain dl309 or dl1102 at an MOI of 50 and harvested 24 h after infection for ChIP analysis. Mock-infected cells were harvested at the same time. For immunoprecipitation of E1A, a combination of monoclonal M73 and M58 antibodies was used. Rabbit anti-rat antibody was used as a negative IgG control (MilliporeSigma). qPCR reactions were carried out using Powerup SYBR Green Master Mix (Thermo Fisher Scientific) on 2.6% of total eluted ChIP DNA as template, by BioRad CFX96 Real Time PCR instrument (Bio-Rad) according to manufac turer's instructions. The annealing temperature used was 69°C, and 40 cycles were run. Primers for viral and cellular promoters are provided in Table 1.
## Immunofluorescence
HT or HT-A cells were plated at low density (∼40,000 cells per chamber) on Nunc Lab Tek-II chamber slides (Thermo Fisher Scientific) and subsequently infected as described above. Twenty-four hours after infection, the cells were fixed in 4% formaldehyde, permeabilized using 0.1% Triton X-100, and blocked in blocking buffer (1% normal goat serum, 1% BSA, 0.2% Tween-20 in PBS), and stained with specific primary antibodies. M73 was used neat (hybridoma supernatant), E2 DBP antibody was used at a 1:50 dilution (hybridoma supernatant), and ARGLU1 antibody was used at a dilution of 1:100. Alexa Fluor 488 and 594-conjugated secondary antibodies (Jackson ImmunoResearch) were applied at a 1:600 dilution. Following three 10 min washes with PBS-T, slides were mounted using Prolong Gold with DAPI (4' ,6-diamidino-2-phenylindole) (Invitrogen) and imaged using Zeiss LSM700 confocal laser scanning microscope. Images were analyzed using the Zeiss ZEN software package version 8.
## Real-time gene expression analysis
HT and HT-A cells were cultured in a 6-well plate (~1 million cells per well) and infected with dl309 and dl1102 at an MOI of 10. Total RNA was extracted at the indicated time points using TRIzol Reagent (MilliporeSigma), following the manufacturer's instructions. To eliminate contaminating DNA, the total extracted RNA was treated with the TURBO DNA-free kit (Invitrogen) according to the manufacturer's protocol. RNA concentration was measured using A260, and its quality was assessed based on the A260/A280 ratio. For cDNA synthesis, 1 µg of total RNA was reverse-transcribed using the SuperScript VILO reverse transcriptase master mix (Invitrogen) with random hexanucleotide primers, following the manufacturer's protocol. Next, 2% of the total cDNA was then used for real-time expression analysis on a BioRad CFX96 real-time thermocycler (Bio-Rad) using BioRad Sso Advanced Universal SYBR Green Supermix in a 10 µL reaction volume, with the standard cycling program recommended for this reagent. Data acquisition was performed using BioRad CFX Manager software (version 3.1). Each sample was analyzed with at least three biological and two technical replicates, unless stated otherwise in the figure legend. Analysis of expression data was carried out using the percentage of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) method, with raw Cq values for GAPDH serving as an internal quality control, as its expression remained stable across experimental conditions. The specificity of PCR amplicons was confirmed via melt curve analysis. To verify the absence of contaminating DNA, RNA samples were also subjected to qPCR analysis without reverse transcription. Post-acquisition data analysis was conducted using Excel for Office 365 for Windows.
## Virus growth assay
HT and HT-A cells were infected with HAdV5 dl309 at an MOI of 10. The virus was allowed to adsorb for 1 h at 37°C under 5% CO₂. Following adsorption, the infection media were topped off with fresh complete media. The cells were then incubated at 37°C under 5% CO₂ for the remainder of the experiment. Each time point was done in three independent biological replicates. Viral titers were measured at 48 and 72 h after infection by plaque assays on 293 cells; titers for each biological replicate were determined in triplicate, and the average of these technical replicates was used in the final calculation. Ratios of virus growth in HT-A versus HT cells were calculated by dividing the viral titer obtained in HT-A cells with each virus at a given time point by the viral titer for the same virus and time point obtained in HT cells. Ratio values of less than 1 represent reduced growth in HT-A cells compared with HT, values greater than 1 represent enhanced growth in HT-A cells versus HT, and values equal to or near 1 indicate similar growth in HT-A cells compared with HT.
## Comet assay
HT and HT-A cells (~100,000 cells per well) were infected with HAdV5 dl309 or dl1102 at an MOI of 500 for 24 h. Following infection, the cells were treated with 5 µM bleomycin (Selleck Chemical) for 15 min, then allowed to recover for 0-30 min. After the recovery period, DNA damage assessment was performed using the Comet Assay Kit (Abcam, cat. ab238544) according to the manufacturer's protocol. Comet tails were resolved through alkaline electrophoresis, stained, and imaged using the ImageXpress Micro 4 Imager (Molecular Devices) with MetaXpress software (version 6.7.0.211). Image analysis and quantification of comet parameters were conducted using ImageJ software with the OpenComet plug-in (32) or TriTek CometScore 2.0.
## Statistical analysis
## PCR primers
## RESULTS
## ARGLU1 directly interacts with E1A and inhibits virus growth
Mass spectrometry analysis of protein complexes associated with HAdV-B7 E1A has previously identified ARGLU1 as a binding partner that interacts with the N-terminus of E1A (3). To determine whether this was a direct interaction, we performed a GST-pull down assay with bacterially expressed 6xHis-E1A289R and GST-ARGLU1 (Fig. 1). Although 6xHis-E1A289R was efficiently pulled down by GST-ARGLU1, GST alone did not pull down significant amounts of 6xHis-E1A289R (Fig. 1). These results demonstrate that E1A directly interacts with ARGLU1.
To explore the role of ARGLU1 during adenovirus infection, we examined its impact on viral replication using phenotypically wild-type HAdV5 dl309 in HT and HT-A cells and similarly a deletion mutant, dl1102, that expresses E1A deficient for ARGLU1 binding (3). HT-A cells displayed significantly reduced dl309 growth, with approximately 8-fold reduction in virus titers at 48 and 72 h after infection compared with HT cells (Fig. 2). The reduction in viral titers observed in HT-A cells suggests that ARGLU1 may act as a viral restriction factor, impeding adenoviral growth or assembly, without impacting viral genome replication (data not shown). Conversely, dl1102 was minimally, and not significantly, affected in growth by ARGLU1 overexpression at 48 and 72 h after infection (Fig. 2). Finally, we determined whether infection affects endogenous and overexpressed ARGLU1 protein levels (Fig. 2); no significant difference was observed in ARGLU1 levels early in infection, with a minimal reduction late in infection. Collectively, these findings indicate that ARGLU1 directly interacts with E1A and plays a negative role in HAdV5 replication.
## ARGLU1 affects viral gene and protein expression
The reduced viral growth observed in HT-A cells (Fig. 2) and enhanced growth when ARGLU1 is knocked down (3) suggest that ARGLU1 may either directly interfere with viral replication or contribute to host innate antiviral response. We therefore wanted to investigate whether overexpression of ARGL1 affects viral gene expression, as our earlier studies with ARGLU1 knockdown showed enhanced viral gene expression with reduced ARGLU1 levels (3). To elucidate this, HT and HT-A cells were infected with HAdV5 dl309 or dl1102, and total RNA was extracted, and expression of viral genes was analyzed at 16, 24, 48, and 72 h after infection. Most viral genes are negatively affected by ARGLU1 overexpression in dl309-infected cells (Fig. 3). In line with our observation of reduced viral titers in HT-A, compared with HT cells, we observed a significant overall reduction in viral gene expression in HT-A cells infected with dl309 (Fig. 3). At 16 h, DBP mRNA in HT-A showed the largest reduction in gene expression, followed by E1A and E1B compared with HT cells (Fig. 3A). Expression of viral genes, E1A, E1B, E3A, E4orf3, fiber, and hexon, was also significantly reduced at 24 h in HT-A compared with HT cells (Fig. 3A). Additionally, significant reduction in gene expression was observed at 48 h for E1A, E1B, and E3A (Fig. 3). However, mRNA levels of E4orf3, fiber, DBP, and hexon showed a modest reduction at 48 h. Furthermore, at 72 h, with the exception of mRNAs for E4orf3 and fiber, all viral genes showed significant reduction in gene expression in HT-A in comparison to HT cells. Our observation that ARGLU1 overexpression in HT-A cells led to decreased levels of E3A and E4orf3 mRNAs, compared with HT cells, suggests that ARGLU1 may downregulate some elements of viral gene expression. The observed results are unlikely to be attributed to variability in the normalization control (GAPDH), as the raw quantification cycle (Cq) values remained consistent across samples. Importantly, HAdV5 mutant dl1102 that deletes E1A residues 26-35 (28,33) required for ARGLU1 binding by E1A (3) was assayed for viral gene expression in HT and HT-A cells. With this mutant, we did not observe any significant reduction in viral gene expression in HT-A compared with HT cells (Fig. 3).
To determine whether altered levels of mRNA translate to protein expression, we performed western blotting for E1A, DBP, and the late proteins in HT and HT-A cells at 16, 24, 48, and 72 h after infection with either dl309 or dl1102 (Fig. 3). We observed a reduction in levels of E1A proteins in HT-A compared with HT cells with both viruses. A major reduction in E1A protein expression was observed very early in the infection at 16 h, whereas E1A showed a modest reduction at 48 h after infection, consistent with the mRNA levels. The observed decrease in E1A levels at 16 h after infection for HT-A cells correlates with decreased mRNA expression, suggesting that the decline may be transcriptionally driven. Levels of DBP were broadly similar between HT and HT-A cells with minimal differences throughout infection, except at 24 h in dl309-infected cells, where slightly higher levels were observed in HT-A cells compared with HT, whereas the opposite was observed in dl1102-infected cells. We have also observed a marked decrease in protein levels for hexon and penton at 48 and 72 h after infection (Fig. 3). The reduced protein levels for hexon consistently correlated with mRNA levels at both 48 and 72 h time points. One major difference observed between dl309 and dl1102 was that overall protein levels were considerably higher in the latter, consistent with higher gene expression. Protein levels of E1A and DBP sustained higher expression late in infection, whereas dl309-infected samples showed greater reduction late in infection. Together, these results demonstrate that ARGLU1 is a negative regulator of viral gene and protein expression during infection.
## ARGLU1 sub-cellular distribution is altered during infection
Since ARGLU1 influences viral replication and gene expression, we examined its subcellular localization in infected HT and HT-A cells. In uninfected cells, ARGLU1 showed predominantly a diffuse nuclear localization with some areas of concentration or speckling (Fig. 4), which is consistent with our previous observations (3). In infected cells, two distinct ARGLU1 phenotypes (phenotype-A & phenotype-B) were observed when cells were stained for ARGLU1 and E1A or DBP, the latter to label viral replication compartments (Fig. 4). ARGLU1 was found either as a diffuse nuclear protein with no specific foci, similar to what was observed in uninfected cells (Fig. 4) or it was re-localized to distinct spots within the nucleus, some of which co-localized with viral replication centers (Fig. 4). The re-localization of ARGLU1 within the infected cells was not depend ent on the ability of ARGLU1 to bind to E1A as the mutant dl1102 showed similar ARGLU1 phenotype to dl309, suggesting that it is infection or virus replication that is causing alteration in sub-nuclear ARGLU1 distribution rather than E1A binding per se. Overall, ARGLU1 phenotypes in infected and uninfected HT cells (Fig. 4) were similar to those observed in the ARGLU1-overexpressing HT-A cells. Variability in E1A distribution was also observed, but this is consistent with our previous observations (14,30) and not a direct consequence of ARGLU1 overexpression. Collectively, these results demonstrate that during viral infection, the sub-nuclear distribution of ARGLU1 is altered.
## ARGLU1 is recruited to viral promoters during HAdV infection
The observed changes in viral gene expression upon ARGLU1 overexpression in HT-A cells suggested that ARGLU1 may play a role in transcriptional regulation. This is consistent with previous findings showing that ARGLU1 knockdown enhances viral gene expression (3), whereas our data suggested that overexpression of ARGLU1 significantly reduced viral gene expression (Fig. 3). Based on these findings, we decided to investigate if the association of ARGLU1 with E1A may enable it to associate with viral promoter regions. To determine whether ARGLU1 associates with viral promoters, and what effect this has on the presence of RPII proximal to these promoters, we performed ChIP assay on dl309and dl1102-infected HT-A cells (Fig. 5). ARGLU1 was found to occupy the E2, E3, and E4 early gene promoters (Fig. 5) in dl309-infected HT-A cells, whereas the levels of ARGLU1 associated with these promoters were significantly reduced in dl1102-infec ted HT-A cells (Fig. 5). Interestingly, the levels of RPII at viral promoters were inversely correlated with viral gene expression, with higher levels observed in dl309-infected cells compared with dl1102-infected cells (Fig. 5). Additionally, E1A was consistently present at viral promoters with no significant difference observed between dl309 and dl1102 viruses.
## E1A binding to ARGLU1 inhibits DNA damage repair
We have previously demonstrated that ARGLU1 overexpression enhances DDR and increases cancer cell resistance to genotoxic drugs (3). One possibility for why E1A targets ARGLU1 is to interfere with DNA damage repair-related activities of ARGLU1, since DDR could have negative consequences for viral replication (19). We therefore investigated whether the binding of ARGLU1 by E1A during HAdV5 infection affects DDR.
To explore this, bleomycin was used to induce DSBs in HT or HT-A cells thatb infected for 24 h with dl309 or dl1102, as we have done in the past (3). Comparison of DNA damage, performed by the comet assay as we did previously (3), shortly after removal of bleomycin showed that in dl309-infected cells, there were significantly higher levels of DNA damage compared with both mock and dl1102-infected cells (Fig. 6). The reduced repair in dl309-infected cells, but not dl1102-infected cells, suggests that the interaction of ARGLU1 with E1A potentially impairs the ability of ARGLU1 to contribute to DNA damage repair. DNA damage response is a complex signal transduction network that senses DNA damage via different members of the phosphatidylinositol 3-kinase-like protein kinase (PIKK) family, which includes ATM and ATR that become activated by autophosphory lation depending on the type of DNA damage (19,31). Given our previous findings, we wanted to investigate whether ARGLU1 sequestering by E1A impairs DNA damage response by affecting activation, and therefore, phosphorylation, of ATM, ATR, and downstream Chk1, as well as downstream effectors such as p21 and 53BP1 (Fig. 7A). To explore this, we infected HT and HT-A cells with dl309 or dl1102 and assessed both total and phosphorylated protein levels of DNA damage response markers 24 h after infection. Levels of ATM and ATR were unchanged regardless of infection. However, levels of ATM were consistently higher in HT-A cells versus HT. Importantly, phosphorylation of ATM and ATR was enhanced in dl309-infected cells but largely unaffected by dl1102-infected cells compared with mock. The downstream target Chk1 was phosphorylated the most in dl309-infected cells, whereas Chk1 was moderately phosphorylated in dl1102-infected cells compared with mock. We also investigated other stress and DNA damage markers, specifically p21 and 53BP1. There was an overall higher level of the cell cycle inhibitor p21 in HT-A cells compared with HT. Levels of the DNA damage response factor 53BP1 were also higher in HT-A cells compared with HT cells, where the protein was undetecta ble. Infection had no significant impact on the overall levels of these proteins. Finally, we investigated transcription of p53 target genes in HT and HT-A cells, as the elevated levels of p21 observed in HT-A cells suggested that p53-regulated genes may be activated (Fig. 7). Indeed, levels of p53-regulated gene transcripts for TP53, MDM2, CDKN1A, and PIG3 were significantly upregulated in HT-A cells versus HT cells. This suggests that HT-A cells are more primed for stress response compared with the parental HT cells. Overall, these results suggest that targeting of ARGLU1 by E1A reduces the DDR response in infected cells downstream of ATM/ATR activation and suggest that downstream stress effectors are upregulated in HT-A cells versus HT.
## DISCUSSION
In the present study, we have identified ARGLU1 as a direct interacting partner of HAdV5 E1A (Fig. 1). Binding of E1A to ARGLU1 recruits ARGLU1 to viral promoters, reducing viral gene expression likely via enhanced promoter-proximal pausing of RNA polymerase II, with subsequent reduction of viral protein levels. Consequently, we show that overex pression of ARGLU1 resulted in a significant reduction in virus growth without affecting viral genome replication (Fig. 2). Furthermore, we demonstrate that during infection, ARGLU1 sub-nuclear distribution is altered into distinct phenotypes not observed in uninfected cells (Fig. 4). Finally, our results suggest that targeting of ARGLU1 by E1A is important for inhibition of the DNA damage response pathway. Enhanced DNA damage repair may be responsible for reduced virus growth in ARGLU1-overexpressing HT-A cells.
Our previous studies showed that ARGLU1 promotes enhanced DNA damage repair through an increase in promoter proximal pausing of RPII after DNA damage has been induced (3). Enhanced DNA damage repair is undesirable for most viruses, as this can lead to viral genome replication defects and induce cell cycle arrest or apoptosis, leading to abortive infection (19). Therefore, viruses, including HAdV, aim to block this response during infection. Indeed, infection with HAdV dl309 led to reduced DNA damage repair following bleomycin treatment, but this was not the case with ARGLU1-binding-deficient E1A-expressing mutant dl1102 (Fig. 6). This suggests that E1A binding to ARGLU1 inhibits its activities that promote enhanced DNA damage repair. Curiously, binding of ARGLU1 by E1A leads to recruitment of ARGLU1 to viral promoters, reducing viral gene expression (Fig. 5), likely an undesirable outcome overall. However, this may be a consequence that ensures that DNA damage response is suppressed, and viral gene expression may not be the limiting factor in virus replication regardless. Importantly, recruitment of ARGLU1 to viral promoters correlates with enhanced presence of RPII on these promoters but overall reduced gene expression. This is not unexpected and is consistent with our earlier observations of ARGLU1 enhancing promoter proximal RPII pausing (3). RPII would not be expected to linger on viral promoters when it quickly transitions to elongating phase (7), but if it is paused there, it would be expected to have higher occupancy as we observed by ChIP (Fig. 5). We also aimed to determine at which point in the DNA damage response cascade ARGLU1 may participate, beyond promoting RPII pausing, by investigating activation of the DNA damage sensor kinases ATM and ATR and their downstream target Chk1. Although we observed no significant difference in the activation of ATM, ATR, and Chk1 upon infection, where infection with dl309 and dl1102 induced activating phosphorylation of these kinases, we did observe an unexpectedly higher level of ATM in HT-A cells compared with the parental cell line HT (Fig. 7A). This could increase the sensitivity of these cells to DNA damage and promote more efficient repair, but it does not provide a clear answer as to how ARGLU1 binding to E1A blocks DNA damage response beyond its impact on RPII pausing on viral promoters. It is possible that downstream signals may be affected, and this merits further investigation. Indeed, stress response effector molecules, such as p21 and 53BP1, were expressed at higher levels in HT-A cells compared with HT cells, suggesting that these cells are primed and ready to rapidly respond to stress or DNA damage.
Our previous characterization of the mutant dl1102 demonstrated that it is capable of replication to levels similar to those of wild-type-like dl309 (30). Interestingly, in that study, we observed that dl1102 expresses most of its viral genes to levels that are higher than dl309, which we have also observed here. At the time, it was unclear why that was, but our current study provides a plausible explanation. The loss of binding to ARGLU1 would result in less RPII pausing at the viral promoters, enhancing expression, as we observed here and in the past (3,30). One possibility is that recruitment of ARGLU1 is not merely a serendipitous incident but is a consequence of natural selection that drives E1A to finely tune the viral transcriptional program. It is reasonable to assume that overactivation of certain genes could affect the host cell in undesirable ways and overall inhibit viral replication. By moderating transcription through ARGLU1 recruitment, the virus may optimize its gene expression for maximal viral replication in a living host. Curiously, dl1102 did not exhibit a growth deficiency or inability to drive S-phase in arrested IMR-90 cells (30). It is possible that higher viral gene expression may overcome any inability to block DNA damage response due to a deficiency in E1A binding to ARGLU1 in this mutant. Indeed, we have observed longer and more sustained expression of E1A in dl1102-infected cells (Fig. 3). Another possibility is that some degree of DNA damage response activation may be beneficial to the virus, and combined with other mechanisms, this may overcome any deleterious effects associated with loss of ARGLU1 binding to E1A. Indeed, activation of the FANC pathway has previously been shown to be beneficial to the viral replicative cycle (34). Clearly, the interaction of HAdV with the DNA damage response pathways is complex, and a deeper investigation will resolve these intriguing questions. Curiously, late protein levels were negatively impacted by ARGLU1 overexpression regardless of the virus infecting the cell (Fig. 3). Although somewhat unexpected, this demonstrates that late protein products are likely in vast excess and do not significantly impact viable viral yields. This may be a consequence of ARGLU1 impacting late transcript splicing, as its role in splicing regulation has been well established (2), and we have previously shown HAdV-dependence on splicing factor availability for specific splice variant production (35). This may lead to aberrant splicing that is not detectable by qPCR but leads to transcripts that are not properly translated into proteins.
In conclusion, the present study provides new insights into how HAdV re-tasks ARGLU1 to advance the viral replicative program. Our study identified ARGLU1 as a direct binding partner of HAdV5 E1A that is re-localized in the nucleus during viral infection, regardless of E1A binding, while providing further mechanistic insights into how ARGLU1 enhances cellular stress response. Moreover, our data further support the notion that we observed previously that ARGLU1 is a multifunctional regulator of viral transcription and cellular DNA damage response, with important implications for our understanding of virus-host interactions and suggests that ARGLU1 may be a valuable host restriction factor for antiviral therapy.
## References
1. Zhang, Jiang, Xu et al. (2011) "Arginine and glutamate-rich 1 (ARGLU1) interacts with mediator subunit 1 (MED1) and is required for estrogen receptor-mediated gene transcription and breast cancer cell growth" *J Biol Chem*
2. Magomedova, Tiefenbach, Zilberman et al. (2019) "ARGLU1 is a transcriptional coactivator and splicing regulator important for stress hormone signaling and development" *Nucleic Acids Res*
3. Bachus, Akkerman, Fulham et al. (2024) "ARGLU1 enhances promoter-proximal pausing of RNA polymerase II and stimulates DNA damage repair" *Nucleic Acids Res*
4. Bessonov, Anokhina, Will et al. (2008) "Isolation of an active step I spliceosome and composition of its RNP core" *Nature*
5. Yao, Huang, Liu et al. (2023) "Deletion of ARGLU1 causes global defects in alternative splicing in vivo and mouse cortical malformations primarily via apoptosis" *Cell Death Dis*
6. Core, Adelman (2019) "Promoter-proximal pausing of RNA polymerase II: a nexus of gene regulation" *Genes Dev*
7. Van Den Heuvel, Van Der Weegen, Boer et al. (2021) "Transcription-coupled DNA repair: from mechanism to human disorder" *Trends Cell Biol*
8. Wright, Dyson (2015) "Intrinsically disordered proteins in cellular signalling and regulation" *Nat Rev Mol Cell Biol*
9. Berk (2013) *Adenoviridae*
10. Pelka, Ablack, Fonseca et al. (2008) "Intrinsic structural disorder in adenovirus E1A: a viral molecular hub linking multiple diverse processes" *J Virol*
11. Frisch, Mymryk (2002) "Adenovirus-5 E1A: paradox and paradigm" *Nat Rev Mol Cell Biol*
12. Frost, Mendez, Soriano et al. (2018) "Adenovirus 5 E1A-mediated suppression of p53 via FUBP1" *J Virol*
13. Jung, Radko, Pelka (2016) "The dual nature of Nek9 in adenovirus replication" *J Virol*
14. Frost, Olanubi, Cheng et al. (2017) "The interaction of adenovirus E1A with the mammalian protein Ku70/XRCC6" *Virology (Auckl)*
15. Olanubi, Frost, Radko et al. (2017) "Suppression of type I interferon signaling by E1A via RuvBL1/Pontin" *J Virol*
16. Graves, Akkerman, Fulham et al. (2024) "Molecular insights into type I interferon suppression and enhanced pathogenicity by species B human adenoviruses B7 and B14" *mBio*
17. Crisostomo, Soriano, Mendez et al. (2019) "Temporal dynamics of adenovirus 5 gene expression in normal human cells" *PLoS One*
18. Macneil, Dodge, Evans et al. (2023) "Adenoviruses in medicine: innocuous pathogen, predator, or partner" *Trends Mol Med*
19. Kleinberger (2020) "En guard! The interactions between adenoviruses and the DNA damage response" *Viruses*
20. Paull, Gellert (2000) "A mechanistic basis for Mre11-directed DNA joining at microhomologies" *Proc Natl Acad Sci*
21. Kobayashi, Antoccia, Tauchi et al. (2004) "NBS1 and its functional role in the DNA damage response" *DNA Repair (Amst)*
22. Lamarche, Orazio, Weitzman (2010) "The MRN complex in double-strand break repair and telomere maintenance" *FEBS Lett*
23. Gupta, Jha, Engel et al. (2013) "Tip60 degradation by adenovirus relieves transcriptional repression of viral transcriptional activator EIA" *Oncogene*
24. Karen, Hoey, Young et al. (2009) "Temporal regulation of the Mre11-Rad50-Nbs1 complex during adenovirus infection" *J Virol*
25. Harlow, Franza, Schley (1985) "Monoclonal antibodies specific for adenovirus early region 1A proteins: extensive heterogeneity in early region 1A products" *J Virol*
26. Radko, Koleva, James et al. (2014) "Adenovirus E1A targets the DREF nuclear factor to regulate virus gene expression, DNA replication, and growth" *J Virol*
27. Jones, Shenk (1979) "An adenovirus type 5 early gene function regulates expression of other early viral genes" *Proc Natl Acad Sci*
28. Jelsma, Howe, Evelegh et al. (1988) "Use of deletion and point mutants Full-Length Text Journal of Virology November"
29. "spanning the coding region of the adenovirus 5 E1A gene to define a domain that is essential for transcriptional activation" *Virology (Auckl)*
30. Pelka, Ablack, Torchia et al. (2009) "Transcriptional control by adenovirus E1A conserved region 3 via p300/ CBP" *Nucleic Acids Res*
31. Costa, Akkerman, Graves et al. (2020) "Characterization of adenovirus 5 E1A exon 1 deletion mutants in the viral replicative cycle" *Viruses*
32. Turnell, Grand (2012) "DNA viruses and the cellular DNA-damage response" *J Gen Virol*
33. Gyori, Venkatachalam, Thiagarajan et al. (2014) "OpenComet: an automated tool for comet assay image analysis" *Redox Biol*
34. Egan, Jelsma, Howe et al. (1988) "Mapping of cellular protein-binding sites on the products of early-region 1A of human adenovirus type 5" *Mol Cell Biol*
35. Cherubini, Naim, Caruso et al. (2011) "The FANC pathway is activated by adenovirus infection and promotes viral replication-dependent recombination" *Nucleic Acids Res*
36. Graves, Akkerman, Bachus et al. (2021) "Differential splicing of human adenovirus 5 E1A RNA expressed in cis versus in trans" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12780214&blobtype=pdf | # npj | vaccines Article
Dan Zhou, Yong Wang, Yanfeng Yao, Wenhua Kuang, Rao Cheng, Gan Zhang, Hang Liu, Xin Li, Sandra Chiu, Zengqin Deng, Haiyan Zhao
## Abstract
Nipah virus (NiV) and Hendra virus (HeV), two highly pathogenic Henipaviruses (HNVs), pose a significant public health threat. The attachment glycoprotein (G) plays a crucial role in viral attachment and entry, making it an attractive target for vaccine and therapeutic antibody development. However, the antigenic landscape and neutralization sensitivity of the diverse HNV G proteins remain poorly defined. Here, we systematically characterize 27 monoclonal antibodies (mAbs) elicited by NiV G head (G H ) nanoparticle-immunized mice. Among these, 25 mAbs exhibit neutralizing activity against two major NiV strains, NiV-Malaysia and NiV-Bangladesh, with five mAbs also cross-inhibiting HeV infection. Notably, mAbs from two distinct groups conferred complete protection to hamsters against lethal NiV-Malaysia challenge. Structural analysis of NiV G H in complex with representative Fabs reveals four non-overlapping epitopes, including two novel antigenic sites and one public protective epitope shared across species. MAbs targeting the novel sites bind to the top or side faces of G protein's β-propeller and inhibit viral infection by blocking either receptor engagement or membrane fusion. MAbs recognizing the public epitope block the receptor binding directly. Our study provides a comprehensive antigenic map of the NiV G H and offers new insights and opportunities for antibodybased therapies and rational vaccine development.
Nipah virus (NiV) and Hendra virus (HeV) are two highly pathogenic zoonotic henipaviruses (HNVs) known to cause severe encephalitis and respiratory diseases in humans with a high case fatality rate [1][2][3] . Since their first identification in the mid to late 1990s, recurrent NiV outbreaks have been reported in Bangladesh and India 4,5 . Although HeV outbreaks have been sporadic in Australia, HeV spillover from bats to horses has been frequently reported, resulting in several human exposures 6,7 . HNVs can be transmitted to humans through contact with infected animals (such as bats, pigs, and horses) or humans, or consumption of contaminated food, underscoring the potential global health risk posed by HNVs 8 . Although mortality rates of 50-100% from HNV infection have been documented, there are still no clinically approved vaccines or specific therapeutics available for humans against HNVs 2,9 . Thus, there is an urgent need to develop prevention and therapeutic strategies against HNV infections.
HNVs are negative-sense, single-stranded paramyxoviruses that possess two viral membrane proteins: the attachment glycoprotein (G) and the fusion protein (F) 10 . The G protein exists as a tetramer on the viral surface and is a class II membrane protein, which can be structurally divided into four regions: the N-terminal cytoplasmic tail, the transmembrane domain, the stalk, and the C-terminal head domain (G H ) 11 . The C-terminal head domains of NiV and HeV are responsible for viral attachment to target cells by interacting with the receptors ephrin-B2 or ephrin-B3 [12][13][14] . The F protein is a trimeric class I fusion protein, composed of three domains: the C-terminal cytoplasmic tail, the transmembrane domain, and the N-terminal ectodomain, which can be further divided into a head domain (DI, DII, and DIII) and a C-terminal stalk [15][16][17] . It is believed that G and F proteins are associated with each other; however, the organization of these two proteins on the whole virion remains unclear. Previous studies suggest that receptor engagement with the G protein likely promotes a conformational transition in the G protein, subsequently triggering F proteinmediated viral fusion with cell membranes to initiate infection [18][19][20][21] .
Since the G and F proteins play critical roles in viral entry into host cells, they serve as the primary antigens for humoral immune response and vaccine development. Both G-and F-based vaccines have been investigated against HNV infections 7,9,[22][23][24] , showing robust immune responses and protective activity in animal models challenged with HNVs 7,9 . In terms of monoclonal antibody (mAb) development, both human and mouse anti-F mAbs have been characterized in vitro or in vivo, with neutralizing antibodies targeting diverse epitopes across all three domains of the F head protein 16,[25][26][27][28][29] . While several mAbs targeting the G protein have demonstrated significant in vitro neutralizing effects and in vivo protection against NiV infection [30][31][32] , most anti-G mAbs inhibit HNV infections by targeting the receptor ephrin-B2 binding site. Therefore, our understanding of the epitope landscape on the HNV G glycoproteins targeted by antibodies is incomplete, especially for those with different inhibitory mechanisms.
We recently designed a ferritin-based NiV G head nanoparticle (NiV G H -ferritin) and generated a panel of mAbs from mice immunized with NiV G H -ferritin nanoparticle 24 . Here, we further investigate the epitopes and neutralizing mechanisms of these newly identified mAbs to characterize the antigenicity of the NiV G head protein. We found that most G-reactive mAbs exhibit potent neutralizing activity against two authentic NiV strains, representing the two major NiV lineages: NiV Malaysia (NiV M ) and NiV Bangladesh (NiV B ). Five mAbs from two distinct competition groups demonstrate cross-neutralizing activity against another highly pathogenic HNV, HeV. Mechanistically, the anti-NiV G mAbs inhibit viral infection through either direct or indirect receptor blockade or by fusion inhibition. Structural analysis of NiV G H bound to representative Fabs from four groups reveals four non-overlapping epitopes within the G H protein, including two previously unreported sites. Overall, our study expands the vulnerable epitope library of G protein and provides important insights for the development of NiV antibody therapeutics and rational vaccine design.
## Results
Most anti-NiV G mAbs are potent neutralizers to NiV We previously developed 27 NiV G-reactive mAbs from NiV M G H nanoparticle-immunized mice, of which 25 showed potent neutralizing activity against vesicular stomatitis virus (VSV)-based NiV Malaysia pseudovirus (VSV-NiV M ) 24 . Although these 27 mAbs were classified into four competition groups (groups 1-4), their binding properties and neutralization sensitivity against distinct henipaviruses (HNVs) remain unclear. To further characterize the mAbs, we first assessed their cross-reactivity against G head proteins from multiple henipaviruses using enzyme-linked immunosorbent assay (ELISA). Of the 27 mAbs, 26 displayed strong binding signals to the recombinant G head proteins (residues 176-602) from two representative NiV strains: NiV M and NiV B , at concentrations of 0.5 μg/mL or 1 μg/mL (Supplementary Fig. 1A,B). We also assessed the binding capacity of the mAbs to the ectodomain of the NiV M G protein (residues 96-602, G ecto ) and found that all the mAbs exhibited similar binding capacity to G ecto as compared to the NiV M G H protein (Supplementary Fig. 1C). Additionally, six mAbs were able to recognize G H from the heterologous HeV (Supplementary Fig. 1D). None of the mAbs crossreacted with G H from two other related henipaviruses, Langya virus (LayV) or Mòjiāng virus (MojV) (Supplementary Fig. 1,E,F).
Next, we evaluated the neutralizing activity of the mAbs against NiV B and HeV pseudoviruses, as well as authentic viruses (Fig. 1; Table 1). Of the 27 mAbs, 25 effectively inhibited infection of the NiV B pseudovirus, with half-maximal inhibitory concentration (IC 50 ) values of less than 0.3 μg/mL. Most mAbs (23 out of 27) were particularly potent, with IC 50 values ≤ 0.015 μg/mL (Fig. 1A). For HeV pseudovirus neutralization, five mAs exhibited varying levels of neutralizing activity, with IC 50 values ranging from 0.001 to 9.86 μg/mL (Fig. 1B,C; Table 1). Consistent with the pseudovirus neutralization results, 25 of the 27 mAbs potently inhibited authentic NiV M and NiV B , and five also neutralized authentic HeV (Fig. 1D-F). Although IC 50 values against authentic viruses were markedly higher than those against pseudoviruses, this difference likely reflects the distinct sensitivities of the two assay systems. Additionally, three previously reported mAbs (HENV-26, HENV-32, and nAH1.3) 11,30 , which target distinct epitopes on G H , were included in both binding and neutralization assays. We previously demonstrated that HENV-26 belongs to group 3, nAH1.3 to group 4, and none of our mAbs competed with HENV-32 for G H binding 24 . Compared to the three control mAbs, 15 of our mAbs exhibited lower IC 50 values against authentic NiV M , and 9 mAbs showed comparable or lower IC 50 values against authentic NiV B (Fig. 1; Table 1). These findings suggest that potently neutralizing mAbs are consistently elicited in mice immunized with NiV M G H nanoparticle.
Neutralizing mAbs in groups 2, 3, and 4 block receptor binding via diverse mechanisms Viral infection can be inhibited by mAbs at multiple stages, but it is particularly attractive and crucial for mAbs to interrupt viral infection at an early stage, before the virus begins to replicate. These include inhibiting receptor binding and viral-cell membrane fusion. To determine whether the mAbs block the host receptor from binding to the NiV attachment G protein, we performed a competition assay using Bio-Layer Interferometry (BLI) with recombinant NiV G H protein. The BLI analysis showed that the binding of group 3 mAbs to G H protein completely abolished the interaction between G H and the ephrin-B2 receptor (Fig. 2A). By contrast, ephrin-B2 was still able to efficiently bind to G H in the presence of group 1, 2, or 4 mAbs. These results suggest that the group 3 mAbs inhibit virus infection through direct receptor binding inhibition, whereas the mAbs in the other three groups seem to neutralize NiV through distinct mechanisms.
Notably, our most potent neutralizing mAbs, including all five HeV cross-reactive mAbs, were derived from groups 3 and 4, which consist of 7 and 17 mAbs, respectively (Fig. 1; Table 1). By contrast, group 2 contains only one mAb (S2B10), which did not block ephrin-B2 binding to recombinant G H and exhibited only moderate inhibitory activity against NiV M and NiV B . Group 1 mAbs demonstrated moderate to weak neutralization and similarly did not block ephrin-B2 binding to recombinant G H , as assessed by competition BLI.
We further assessed receptor binding inhibition by flow cytometry using full-length NiV G and NiV G/F-transfected cells to mimic the organization of NiV surface glycoproteins. Live NiV G or NiV G/Fexpressing CHO cells were first incubated with individual mAbs and then stained with a fluorescence-labeled secondary antibody. Flow cytometric analysis showed that all 27 mAbs recognized G proteins displayed on the cell surface to varying degrees (Supplementary Fig. 2). We then analyzed the binding of ephrin-B2 to NiV G or NiV G/Fexpressing CHO cells in the presence or absence of the mAbs (Fig. 2B,C). Consistent with the BLI competition assay, pre-incubation of the cells with potent neutralizing mAbs in group 3 eliminated ephrin-B2 receptor binding, except for one modestly neutralizing mAb, LN3E5 (IC 50 = 14.8 μg/mL against authentic NiV M ). Unexpectedly, all group 4 mAbs (16 mAbs) decreased ephrin-B2 binding to cells by approximately 30-50%, and the group 2 mAb S2B10 also showed a slight reduction in ephrin-B2 binding. However, no receptor binding blockade was observed for group 1 and group 5 mAbs. Together with the BLI competition results, we conclude that although neutralizing mAbs from groups 2 and 4 cannot directly block the binding of ephrin-B2 to recombinant G H protein, their interactions with full-length G on the cell surface likely cause steric interference with ephrin-B2 binding.
## Neutralizing mAbs in group and 5 inhibit viral fusion
To investigate whether the mAbs could inhibit viral fusion, we conducted a NiV glycoprotein-mediated cell-cell fusion assay using a dual split protein (DSP) system by monitoring renilla luciferase (Luc) activity (Fig. 3A-E) and green fluorescent protein (GFP) signal (Fig. 3F,G). While mAbs from group 1 did not inhibit receptor binding (Fig. 2), the neutralizing mAbs from this group (S1E2 and LN1D1) effectively reduced syncytium formation and cell-cell fusion with lowest 50% effective concentration (EC 50 ) value of 287 ng/mL (Fig. 3A). Despite group 2 mAb (S2B10) likely had steric interference with receptor binding to the G protein similarly as mAbs from group 4, S2B10 did not inhibit NiV glycoprotein-mediated cell-cell fusion (Fig. 3B). By contrast, mAbs from group 4 exhibited potent fusion blockade activity similar to S1E2 from group 1 (Fig. 3A,D). Although group 3 mAbs strongly inhibited receptor binding to the G protein, they only demonstrated weak fusion inhibition activity (~30-50%
decreases fusion) at high concentrations (1-10 μg/mL) (Fig. 3C). HENV-32 (group 5) also inhibited fusion potently (Fig. 3E), but did not show any blockage activity for ephrin-B2 binding (Fig. 2). Consistent with the Luc activity, 1 and 10 μg/mL of mAbs exhibited similar fusion inhibition profiles as measured by GFP intensity (Fig. 3F,G).
## Potently neutralizing mAbs display in vivo protection against lethal NiV infection
We next assessed the protective activity of representative mAbs from each group against lethal NiV M infection in vivo, both as pre-exposure prophylaxis and post-exposure treatment (Fig. 4). The most potent neutralizing mAb against authentic NiV M was selected from each group: S1E2 (Group 1), S2B10 (Group 2), LN1F9 (Group 3), and LN1D11 (Group 4). Although LN1D11 and LN3E2 from Group 4 displayed comparable neutralization potency against authentic NiV M , LN1D11 was chosen due to its stronger activity against NiV B and higher production yield. For the pre-exposure prophylaxis, six-weekold Syrian golden hamsters were administered 30 mg/kg of antibody of either the tested mAb or an isotype control mAb S2A5 (an SFTSV mAb 33 ) intraperitoneally (i.p.) 24 h prior to NiV infection 33 . The survival rate and body weight of the hamsters were subsequently monitored. All animals that received LN1F9 (group 3 mAb) or LN1D11 (group 4 mAb) survived without observed weight loss over the 14-day observation period. By contrast, all hamsters in the isotype control-treated group succumbed to the infection within 6 days post-infection (dpi). The administration of mAb S1E2 (group 1 mAb) or S2B10 (group 2 mAb) conferred partial protection against NiV challenge, as evidenced by the extended survival time of the hamsters and the fact that 1 out of 5 mAb-administered hamsters remained alive at the end of the study (Fig. 4B,C). For post-exposure therapy, six-week-old hamsters were challenged with NiV, followed by the administration of the tested mAbs at 24 and 72 h post-infection (a total of 30 mg/kg) via the intraperitoneal route (i.p.). Treatment with the non-binding isotype control mAb failed to protect the hamsters, with all animals succumbing to infection by 6-8 dpi. We observed 83%, 67%, and 33% survival rates in hamsters treated with LN1D11 (group 4 mAb), S1E2 (group 1 mAb), and S2B10 (group 2 mAb), respectively. However, only the group 3 mAb, LN1F9, exhibited 100% protection when administered as post-exposure therapy (Fig. 4D). For the surviving hamsters, steady weight gain was observed across all mAb treatment groups throughout the course of the experiment (Fig. 4E). Time (s)
$$B C A S1E2 LN1D1 LN3G9 S2B10 HENV-26 LN4H8 LN1F9 LN2B7 LN3A12 LN3B1 LN3C3 LN3E5 nAH1.3 LN1A12 LN1B1 LN1D11 LN3A8 LN3B9 LN3D3 LN3D9 LN3D10 LN3E2 LN3F5 LN3F7 LN4A8 LN4C11 LN4E10 LN4F1LN4G9$$
$$HENV-32 G H ephrin-B2 S1E2 LN1D1 LN3G9 S2B10 HENV-26 LN4H8 LN1F9 LN2B7 LN3A12 LN3B1 LN3C3 LN3E5 nAH1.3 LN1A12 LN1B1 LN1D11 LN3A8 LN3B9 LN3D3 LN3D9 LN3D10 LN3E2 LN3F5 LN3F7 LN4A8 LN4C11 LN4E10 LN4F1LN4G9$$
## Anti-NiV G mAbs recognize four distinct antigenic sites on G H
To gain insights into G-specific mAbs neutralization, we selected a representative mAb from each group (S1E2 from group 1, S2B10 from group 2, LN1F9 from group 3, and LN3D3 from group 4) and determined the complex structures of the G head protein (G H ) bound to the antigen-binding fragments (Fabs) of these mAbs using cryo-electron microscopy (cryo-EM) or X-ray crystallography. Since the inclusion of LN1D11 in the complex resulted in particle aggregation after cryo-EM sample freezing, we proceeded with LN3D3, a sister clone with identical heavy and light chain germline genes and the same CDR lengths. Initially, we aimed to determine the structure of the G H protein complexed with all four Fabs simultaneously, as mAb competition assays suggest that these Fabs could bind concurrently. However, obtaining a high-resolution quinary complex structure proved challenging due to the preferred orientation of the sample. Consequently, we determined the cryo-EM structure of G H bound to three Fabs (S1E2, S2B10, and LN3D3) at a resolution of 3.01 Å (Supplementary Fig. 3 and Supplementary Table 1), and the crystal structure of G H complexed with the LN1F9 Fab at 3.0 Å resolution (Supplementary Table 2). The quaternary complex structure reveals that the three Fabs could bind to G H simultaneously without steric clash (Fig. 5A), consistent with the competition biolayer interferometry results 24 . N-linked oligosaccharides were observed on the G H protein in the cryo-EM map, but none are located within the G-Fab interfaces. The surface areas on G H buried by S1E2, S2B10, and LN3D3 are 764.3 Å 2 , 1007.7 Å 2 , and 900.9 Å 2 , respectively (Fig. 5C).
Specifically, S1E2 interacts with the side of the G protein's β-propeller (Fig. 5D,H), engaging 18 residues distributed across four discontinuous segments: S311-L314, 11 residues on β3S1a strand and connecting loop of β2S4a-β3S1a (L337-R344 and D346-V348), E374, and two residues in the β3S3-β3S4 loop (N423 and K425). To corroborate the structural findings, we generated site-directed alanine substitutions of selected residues in NiV G and assessed the binding ability of S1E2 to full-length G-expressing cells by flow cytometry. S1E2 showed markedly decreased binding when alanine mutation was introduced at epitope residue K342 (Supplementary Fig. 4A). Analysis of S1E2 contact residues on G protein reveals that the interaction is predominantly mediated by heavy chain, which contributes ~94% contacts (17 out of 18 residues), including four residues (S311, L337-S339) contacted by both the heavy and light chains (Supplementary Tables 3,4). S2B10 binds to the top of the G protein β-propeller by engaging 24 residues across four discontinuous elements: the N-terminal region (Q212-C216), 10 residues in the β1S2-β1S3 loop and the N-terminal of the β1S3 strand (I237-C240 and R242-Q247), the β1S4-β2S1 loop (T272-N277), and the β6S2-β6S3 loop (D585, N586 and I588) (Fig. 5E,H). Consistent with the structural analysis, S2B10 exhibited reduced binding phenotypes ( ~65%) to variant V244A (Supplementary Fig. 4B). Both the heavy and light chains contribute almost equally to the interactions, with 10 residues engaged by the heavy chain, 11 residues by the light chain, and 3 residues by both heavy and light chains (Supplementary Tables 3,4).
LN3D3 contacts the bottom loops of the G protein, including the N-terminal linker (L184-I188 and L190-Q191) as well as the C-terminal residue Q600, loop of β5S1-β5S2 (D515, I517-N518, and I520), loop of β5S3-β5S4 (K541-N543), and loop of β6S1-β6S2 (N570) (Fig. 5G,H). Consistently, a single alanine substitution of N186 in G protein diminished >80% interaction of LN3D3 with G-expressing cells (Supplementary Fig. 4D). All six complementarity-determining regions (CDRs) from the heavy and light chains of LN3D3 are involved in G H interaction, with the heavy chain contributing 60% of the buried surface at the interface of the G H -LN3D3 complex (Supplementary Tables 3,4). Sequence comparison with LN1D11 shows that all hydrogen-bonding residues in the CDRs identified in the LN3D3-G H complex are fully conserved in LN1D11, with only two residues in CDR-L3 differing and contributing to van der Waals contacts. These minor variations may fine-tune binding affinity or complex stability, likely accounting for the observed differences in neutralization potency.
The crystal structure of G H -LN1F9 demonstrates that LN1F9 binds to the top loops of the G protein (Fig. 5B,F). Docking of the G H -LN1F9 complex onto the quaternary cryo-EM structure of G H -S1E2-S2B10-LN3D3 indicates that the four mAbs target non-overlapping epitopes on G H , allowing all 4 mAbs to bind G H simultaneously (Fig. 5C). The epitope recognized by LN1F9 includes residues Y389, K391, E393, L397, and P403-N404, which are located within the β3S2-β3S3 loop, as well as residues S491-Q492 and E501-E505 in the β4S4-β5S1 loop (Fig. 5F,H). Consistently, mutating E501 and E505 to alanine in G protein resulted in partial loss-of-binding phenotype, and LN1F9 further completely lost binding to a G protein variant with four amino acid substitutions: E391A/E393A/E501A/E505A (Supplementary Fig. 4C). Although LN1F9 interacts a total of 13 residues of G protein, all contacts are engaged by the heavy chain, with only one single residue contacted by both light and heavy chains. The surface area on G H buried by LN1F9 is 627.3 Ų, with 90% of this area covered by the heavy chain (Supplementary Tables 3,4). Group 1 and 2 mAbs target two novel epitopes on NiV G H, To understand the antigenic features of G protein, we compared the epitopes recognized by our mAbs from four groups with the receptor ephrin-B2 and structurally available antibodies (Fig. 6). Alignment of the structures of ephrin-B2-bound and mAb-bound NiV/HeV G with our G H -LN1F9 and G H -S1E2-S2B10-LN3D3 complexes revealed that G H -specific mAbs target at least six distinct antigenic regions on the NiV/HeV G protein. Notably, two of these epitopes, targeted by S1E2 (group 1) and S2B10 (group 2), represent novel sites that have not been previously described (Fig. 6). The footprints of group 2 mAb (S2B10) are near the receptor ephrin-B2 binding site and also reside on the top of the G propeller (Fig. 6A,B). Detailed analysis the binding footprints shows that several S2B10 contacts on G H are also recognized by the ephrin-B2 and some group 3 mAbs (Fig. 6G). In comparison to S2B10, the ephrin-B2 and group 3 mAbs bind to G H from different orientations, and S2B10 binding to G protein results in varying degrees of steric clash with ephrin-B2 and group 3 mAbs (Supplementary Fig. 5). Group 1 (S1E2), group 5 (HENV-32 and n425), and group 6 (hAH1.3) target distinct regions on the side of the G propeller 30,34,35 . Group 4 mAbs (LN3D3 and nAH1.3) target the bottom side of the G propeller, opposite to the ephrin-B2 binding site (Fig. 6C-G) 11 . Interestingly, some mAbs from groups 3, 4, and 5 displayed cross-reactivity with the G proteins of both NiV and HeV, with potent cross-neutralization observed. In contrast, the epitopes recognized by mAbs from groups 1, 2, and 6 appeared to be specific to either NiV or HeV 35 .
Group 3 mAbs share epitopes with human and macaca mAbs Group 3 mAbs target the top side of the G propeller, overlapping greatly with the footprint of receptor ephrin-B2, and inhibit HNV infection through mimicking receptor binding (Fig. 6). This group includes one macaca-derived mAb (1E5) and three human mAbs (HENV-26, m102.3, and 41-6) 30,31,36,37 . The buried surface areas of these epitopes on the G head are 1,163.4 Š2 for HENV-26, 969.7 Š2 for 41-6, 865.9 Ų for m102.3, and 1,746.0 Š2 for 1E5 (Supplementary Fig. 5G). Similar to LN1F9, group 3 mAbs from both humans and macaques primarily engage the G H region through their heavy chains, with 70%, 95%, 95%, and 76% of the contacting residues contributed by the heavy chains of HENV-26, 41-6, m102.3, and 1E5, respectively (Supplementary Table 5).
Structural analysis revealed that both the human and macaca mAbs utilize their relatively long CDR-H3 loops (18-23 amino acids) to insert into the central cavity of the G propeller, closely mimicking the binding mode of ephrin-B2. By contrast, our mouse mAb LN1F9, with a shorter CDR-H3 loop (13 amino acids), only marginally protrudes into the G propeller's cavity (Supplementary Fig. 5). Notably, the amide nitrogen of residue G506 forms hydrogen bonds with CDR-H3 residue L106 in HENV-26, and with P107 in m102.3 and 41-6. The macaque mAb 1E5 also contributes three van der Waals contacts with G506, whereas this interaction is absent in LN1F9. In addition, we observed that residue W504 within the cavity is consistently engaged by all group 3 mAbs through extensive van der Waals contacts or hydrogen bonding, paralleling its role in receptor interaction (Fig. 6G and Supplementary Table 5).
Although all group 3 mAbs recognize overlapping epitopes and directly block ephrin-B2 binding, each mAb also has unique interactions on the G H . Structural comparison further revealed two common loops ( 488 PGQSQ 492 and 501 EICWEG 506 ) for group 3 mAbs recognition of the ephrin-B2 binding site, suggesting that these two regions may represent a public epitope across species. These two loops form extensive hydrogen interactions with ephrin-B2 (5 hydrogen bonds) and group 3 mAbs (3-7 hydrogen bonds) 30,31,37 , except for mAb 41-6 (one hydrogen bond) 36 . Importantly, like other group 3 mAbs, LN1F9 demonstrated protective efficacy in NiV-challenged animal models and showed cross-neutralization against both HeV and NiV, highlighting the conservation and protective nature of this epitope.
The receptor-binding site recognized by group 3 mAbs is an immunodominant antigenic site in mice and hamster Syrian golden hamster is a well-studied animal model for HNVs infection and vaccine evaluation prior to the pre-clinal trial of anti-HNVs vaccines. We then assessed the antibody composition in sera from mice and hamsters immunized twice with NiV sG H and NiV G H -ferritin by serum competition ELISA assay. 8 mAbs from five NiV G-bound groups (group 1: S1E2 and LN1D1; group 2: S2B10; group 3: LN3B1 and LN3A12; group 4: LN3E2 and LN3D3; group 5: HENV-32) were selected, and we monitored the binding signal of antisera from immunized animals to the NiV G head protein in the presence or absence of indicated mAbs (Fig. 7).
Serum competition analysis showed that NiV sG H and NiV G H -ferritin could induce polyclonal antibodies of group 1-4, but not group 5 antibodies in both 1 μg/dose or 10 μg/dose immunized mice (Fig. 7A,B). When preincubated G H with mAbs first, the sera of NiV sG H immunized mice lost ~15%, ~60%, ~40% and ~25% binding to G H in the presence of mAbs from groups 1-4 (Fig. 7A). For NiV G H -ferritin immunized mouse serum response, group 3 mAbs demonstrated the strongest binding inhibitory activity (Fig. 7B). In hamster, NiV sG H -elicted sera recognize all five antigenic sites (Fig. 7C), whereas NiV G H -ferritin-elicited sera primarily target antigenic sites within groups 1-3 (Fig. 7D). This suggests that the antigenicity of the G head domain differs when presented as a nanoparticle compared to its soluble form. Additionally, group 1 and 3 mAbs showed dominant blockade activity against sera from NiV G H -ferritin immunized hamster. Taken together, these results demonstrate that group 3 mAbs are readily elicited in NiV G-immunized hamsters and mice.
$$G H LN1F9 (group 3) V H V L N C N518 N543 N186 N187 P185 I517 F R338 E341 K342 L314 S311 I340 C N C N S213 R242 N586 S239 P274 S245 Q247 V H S1E2 (group 1) G H S2B10 (group 2) V L V H V L V H V L LN3D3 (group$$
## Discussion
Although several studies have generated mAbs targeting the G proteins of HNVs, these mAbs were derived from HeV-G vaccinated human 30 , naïve human or human single-domain antibody libraries 34,36 , or G-immunized macaque 31 . There remains a need to systematically map the epitope landscape and neutralization sensitivity of the HNV G proteins. In this study, we characterize 27 mouse mAbs elicited by NiV G H -ferritin both structurally and functionally, revealing two novel antigenic sites on G H and one protective, shared epitope across multiple species. Our newly isolated mAbs, along with previously reported mAbs, recognize at least five distinct epitopes on NiV G H and inhibit NiV infection by either blocking receptor recognition through epitope competition or steric hindrance, or by inhibiting fusion. This work provides antigenic knowledge of NiV G H and uncovers the relationships between epitopes, functions, and the mechanisms of action of anti-NiV G mAbs.
Most mAbs in our panel exhibited potent neutralizing activity against both pseudotyped and authentic NiV strains (NiV M and NiV B ), and five mAbs also cross-neutralized pseudotyped and authentic HeV in vitro. Additionally, representative neutralizing mAbs from 4 groups demonstrated partial to complete protection against lethal NiV M challenge in vivo, with groups 3 (LN1F9) and 4 (LN1D11) mAbs showing 100% protection as prophylaxis. Notably, all five HeV cross-reactive mAbs belong to groups 3 and 4, though no cross-reactivity was observed against two other potential human henipaviruses, Langya virus and Mòjiāng virus, likely due to their relatively low sequence identity ( < 30%). The cross-neutralizing activity of LN1F9 and three clonally related group 3 mAbs (LN2B7, LN3C3, LN4H8) likely reflects the high conservation of their epitope, with only two amino acid differences at the G contact sites between NiV and HeV (P403S and I502V). These residues contribute five (P403S) and eight (I502V) van der Waals contacts, respectively (Fig. 6G and Supplementary Tables 3,4). By contrast, the cross-neutralizing mechanism of the group 4 mAb LN4A8 remains unresolved, as its genetic origin differs from structurally characterized group 4 mAbs LN3D3 and nAH1.3. While further studies are needed, we speculate that the cross-protection against HeV by mAbs from these two groups is desirable.
As their names suggest, the attachment glycoprotein (G) primarily mediates viral attachment to host cells, while the fusion glycoprotein (F) initiates viral fusion with the host cell membrane during NiV infection. To evaluate receptor-binding inhibition, we used an "inverted" assay format in which G or G/F is displayed on cells and the receptor is supplied in soluble form. Although assays measuring virion binding to receptor-expressing cells would more directly capture antibody-mediated inhibition of attachment, such assays are not feasible due to biosafety restrictions. Given the unresolved native architecture of the HNV G-F complex, this strategy retains key aspects of glycoprotein organization and has been successfully applied in prior studies 30,38 . Nevertheless, the exact stoichiometry and interplay of G and F on the cell surface remain undefined, complicating precise quantification of receptor and antibody binding in this system.
Although our mAbs bind to four distinct epitopes on the NiV G protein with varying neutralizing potency, potent inhibitory mAbs from different groups can block either viral fusion or receptor recognition. A previously reported group 5 mAb, HENV-32, provided post-exposure protection against NiV challenge in ferrets, though the mechanistic basis of its protection remains unclear 30 . We here demonstrate that HENV-32 has minimal effect on recombinant ephrin-B2 binding to NiV G, whether in its soluble form or displayed on the cell surface, mimicking the glycoprotein configuration on the virion surface. Similar to group 1 neutralizing mAbs, HENV-32 efficiently reduced NiV glycoprotein-mediated cell-cell fusion without affecting receptor recognition.
Structural analysis revealed that S1E2 (group 1 mAb) and HENV-32 (group 5 mAb) bind to distinct regions on the side of the G propeller, with HENV-32 likely causing a steric clash with the stalk or G head domain of adjacent G proteins. By contrast, S1E2's epitope on the G tetramer is fully exposed, and no steric clash occurs when it binds to any G subunit. Despite targeting different epitopes on the G protein, both mAbs share a common fusion inhibition mechanism, likely by preventing the conformational transition of G required to trigger fusion or by impacting the interaction between G and F. Similarly, many mAbs recognize distinct antigenic sites on viral structural proteins, yet exhibit similar attachment or fusion inhibition phenotypes, as seen in different viruses 33,[39][40][41] .
Group 5 mAbs have been reported in HeV G-vaccinated humans or naïve human single-domain library 30,34 ; however, we did not observe the presence of group 5 mAbs in the sera of NiV sG H -or G H nanoparticleimmunized mice. Furthermore, none of the 27 mAbs showed competition with the group 5 mAb HENV-32 for NiV G H binding, as detected by BLI, consistent with structural comparison data. Notably, 17 out of the 27 mouse mAbs belong to group 4, with only one mouse-derived mAb nAH1.3 from hybridomas previously reported falling into this group 11,42 . While a larger panel of human and mouse mAbs is needed for further investigation, these findings suggest that mice do not fully recapitulate the human antibody response to HNV G proteins, likely due to differences in B cell repertoires between the two species.
The group 3 mAb LN1F9 provided both pre-and post-exposure protection against lethal NiV M infection in hamsters. Similar group 3 mAbs have been isolated from HeV G-vaccinated humans 30 and NiV/HeV G-immunized macaques 31 , indicating that the epitope targeted by the receptor ephrin-B2 and group 3 mAbs represents an immunodominant 'public epitope' recognized across species. The group 4 mAb LN1D11 conferred complete protection when administered prophylactically and 83% protection when used therapeutically. Our structural and mechanistic studies reveal that mAbs in groups 3 and 4 bound non-overlapping epitopes on G H and inhibited NiV infection through distinct strategies. Notably, reported mAbs from these two groups also exhibited crossreactive and inhibitory activity against three pathogenic HNVs: NiV M , NiV B , and HeV. Because the G H residues recognized by protective human mAbs significantly overlap with the LN1F9 epitope, these sites may be prone to escape mutants under human antibody selective pressure. In contrast, the epitope targeted by group 4 mAbs appears to be mousespecific and may experience relatively weaker selection pressure compared to the more dominant human epitopes. Therefore, a combination of mAbs from these two groups would be an effective countermeasure against HNV infections in the future.
In summary, our study reveals the antibody components and mechanisms of the humoral immune response at both monoclonal and polyclonal levels induced by NiV G H immunogen, further demonstrating that NiV G H -ferritin can serve as a promising vaccine candidate against NiV infection. Our characterization of neutralizing mAbs, which exhibit distinct neutralization profiles against different HNVs, target diverse epitopes on G H , inhibit NiV infection at multiple stages, and demonstrate therapeutic 32). The binding of the sera in the presence of an isotype control mAb T3D9 (anti-SFTSV mAb, unpublished) was used as a control, and the OD450 (absorbance at 450 nm) value was defined as 100%. Each graph represents the mean and SD from one experiment and is representive of two independent experiments performed in duplicate.
potential, provides valuable insights for the rational design of vaccines against highly lethal HNVs.
## Methods
## Ethics statement
Hamster experiments were approved by the Wuhan Institute of Virology, Chinese Academy of Sciences (approval number: WIVA21202301). Authentic virus infections were performed in the animal biosafety level 4 (ABSL-4) facility at the National Biosafety Laboratory (Wuhan), Chinese Academy of Sciences.
## Viruses and cells
The NiV M , NiV B , and HeV strains were obtained from the National Virus Resource Center at the Wuhan Institute of Virology for authentic virus neutralization assays and challenge studies. All authentic viruses were propagated in Vero E6 cells. Vero E6 and 293T cells were cultured at 37 °C with 5% CO 2 in DMEM (Monad Biotech) containing 8% or 10% FBS (ExCell Bio). CHO-K1 cells were cultured in DMEM/F12 (1:1) medium (Biosharp, Cat#: BL305A) with 10% FBS. Expi293 cells were cultured at 37 °C with 8% CO 2 in SMM 293-TII Expression Medium (Sino Biological, Cat#: M293TII) or 293F Hi-exp Medium (OPM Biosciences, Cat#: AC601501).
## Plasmids
DNA segments encoding the full-length NiV M G (NCBI accession number: NP_112027), NiV M F (NP_112026), NiV B G (AAY43916), NiV B F (AAY43915), HeV G (NP_047112), HeV F (NP_047111), MojV G (YP_ 009094095), LayV G (UUV47206) and human ephrin-B2 (residues 27-167; NP_004084) were codon-optimized for expression and synthesized by Tsingke Biotechnology Co. The coding regions for NiV M G head domain (residues 176-602), NiV M G ectodomain (residues 96-602), NiV B G head domain (residues 176-602), HeV G head domain (residues 176-605), LayV G head domain (residues 176-625), and MojV G head domain (residues 176-625) were individually cloned into a mammalian expression vector with a signal peptide, an N-terminal 6×His-tag, and an N-terminal HRV 3 C protease cleavage site. The coding regions for human ephrin-B2 (residues 27-167) were individually cloned into a mammalian expression vector with a signal peptide and a C-terminal 6×His-tag. Coding regions for the fulllength G and F of NiV M , NiV B , and HeV were cloned into pCAGGS vector, respectively, for VSV-based pseudovirus generation. To increase the pseudovirus titer, S207L and G252D point mutations were introduced on the NiV B F gene, and a truncated variant was introduced on the HeV F gene, leaving 5 residues in the cytoplasmic tail 43 . The variable regions of heavy (VH) and light chains (VL) were cloned into AbVec2.0-IGHG1 (Addgene) and AbVec1.1-IgKC (Addgene) or AbVec2.1-IGLC2 (Addgene) expressing vectors, respectively, for mAb generation. To generate Fab, the VH segments were also cloned into a modified AbVec2.0-IGHG1 expressing vector with a C-terminal 6×His-tag after the CH1 constant region.
## Protein expression and purification
Proteins were generated in Expi293 cells by transient transfection using CarpTrans (OPM Biosciences, Cat#: AC501302). Briefly, 200 μg expression plasmids of human ephrin-B2, HNVs G, mAbs, or Fabs were mixed with 800 μL CarpTrans following the manufacturer's protocol and added into 200 mL Expi293 cells. For Fabs and mAbs generation, the paired heavy-and light-chain plasmids were co-transfected into Expi293 cells at a molar ratio of 1:1 (for Fab) or 1:1.2 (for mAb). Five or six days after transfection, the cell supernatants containing proteins or Fabs were harvested and filtered through 0.45 μm filters. The Fabs were purified using Ni-Charged Resin (GenScript, Cat#: L00666) and size exclusion chromatography (Superose 6 Increase 10/300 GL column or Superdex 200 Increase 10/300 GL column). MAbs were purified using rProtein A Beads (Smart Lifesciences, Cat#: SA015100).
To generate the NiV M G head protein for crystallization, 5 μM kifunensine (Toronto Research Chemicals, Cat#: K450000) was added to the culture medium before transfection. The purified protein was treated with Endo HF (New England Biolabs, Cat#: P0703L) and HRV 3 C protease to produce a homogeneously deglycosylated NiV M G head domain without the His-tag.
## Enzyme-linked immunosorbent assay (ELISA)
For binding activities of NiV G mAbs to different HNVs G proteins, 96-well ELISA plates (Corning, Cat#: 9018) were coated with 3 μg/mL of the purified HNV G head protein in coating buffer (0.1 M carbonate, pH 9.6) at 4 °C overnight. The plates were then washed with PBS-T (PBS with 0.05% Tween 20) and blocked with blocking buffer (PBS-T containing 1% BSA) at 37 °C for 2 h. After blocking, the buffer was replaced with fresh blocking buffer containing NiV G mAbs (0.5 μg/mL for NiV M G head domain and NiV M G ectodomain, and 1 μg/mL for other HNVs G head domain) and incubated at 37 °C for 2 h. Following four washes, the plates were incubated with HRPconjugated goat anti-human IgG (ABclonal, Cat#: AS002) at 37 °C for 1 h. After four additional washes, the plates were developed with onecomponent TMB chromogen solution (NCM Biotech, Cat#: M30500) at 37 °C for 10-30 min, and the reaction was stopped with 1 M HCl. The absorbance at 450 nm was measured using a Varioskan LUX (Thermo Scientific).
For the proportion of different group antibodies in sera from hamsters and mice, 3 μg/mL NiV M G head domain was coated on the ELISA plates. After blocking, 10 μg/mL representative G-specific mAbs from groups 1-5 were used to occupy the corresponding epitope. Wells with 10 μg/mL T3D9 (an anti-SFTSV mAb, unpublished) served as controls, with their absorbance at 450 nm defined as 100%. After 90 min incubation at 37 °C, 50 μL sera from NiV sG H or NiV G H -ferritin vaccinated mice or hamsters, at appropriate dilution, were added into the ELISA plates and incubated at 37°C for 30 min. HRP-conjugated goat anti-hamster or anti-mouse IgG (Thermo Scientific, Cat#: PA128823; Cat#: 31430) were used for hamster or mouse sera detection, respectively.
## Biolayer interferometry assay (BLI)
The competition biolayer interferometry (BLI) assay was conducted on an Octet Red 96 device (Pall ForteBio) to evaluate whether NiV G mAbs can block the binding of human ephrin-B2 to purified NiV G H protein. Briefly, 10 μg/mL of NiV G mAbs were first loaded onto ProA Biosensors (Sartorius, Cat#: 18-5010). After a 10-s wash with running buffer, the biosensor tips were dipped into wells containing 500 nM NiV M G H protein for 120 s. All proteins were diluted in running buffer (10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 0.05% Tween-20, and 1% BSA, pH 7.4). Subsequently, the tips were then immersed in a buffer containing 100 μg/mL of human ephrin-B2 for 120 s to monitor the binding signal of ephrin-B2 to the mAb-captured G H . Tips without mAbs loading were run in parallel to define the background signal. Data were analyzed using Octet data analysis software (version 12.2.0.20).
## Pseudovirus packaging and neutralization assays
The VSV-based HNV (NiV M , NiV B , and HeV) pseudotyped viruses were generated following previously published protocols 24,44 . Briefly, pCAGGS plasmids encoding the full-length HNV G and F genes were co-transfected into 293 T cells using Gene Twin (Biomed, Cat#: TG101). After transfection, the cells were cultured for 16-24 h and then infected with VSVΔG-eGFP for 4-6 h. Following infection, the cells were washed with PBS and incubated with 1 μg/mL of anti-VSV G mAb I1 45 diluted in DMEM with 4% FBS. 24-30 h post-infection, the pseudovirus supernatants were harvested and aliquoted before storage at -80 °C.
For the pseudovirus neutralization assay, Vero E6 cells were seeded in the 96-well plates at a density of 1.5 × 10 5 cells/mL the day before the experiments. Serially diluted NiV G mAbs or Fabs were mixed with pseudovirus and incubated at 37 °C for 1 h. MAb or Fab-pseudovirus mixtures were then added to the Vero E6 cells. The following day, cells were fixed with 4% paraformaldehyde (PFA), and the green fluorescent dots were counted using a CTL-S6 Universal M2. IC₅₀ (half-maximal inhibitory concentration) was calculated using GraphPad Prism (v.8.0) with a nonlinear regression model.
## Flow cytometric assay
A flow cytometric assay was used to identify critical residues on NiV G for mAb binding. Expi293 cells were transiently transfected with wild-type NiV M G (positive control), G mutant plasmids, or pCAGGS vector (negative control). At 48 h post-transfection, the cells were fixed with 4% PFA for 10 min, and the cells were then washed and incubated with 10 μg/mL of the indicated NiV G mAbs at 4 °C for 30 min. Following washing, the cells were stained with Alexa Fluor 488 anti-human IgG antibody (Thermo Scientific, Cat#: A-11013) at 4 °C for 30 min. The cells were then washed again, resuspended, and subjected to analysis using a CytoFLEX S (Beckman). The binding capacity of the indicated NiV G mAbs to G point mutations was calculated relative to wildtype G and normalized by the G protein expression levels. Cells stained with 10 μg/mL of NiV G mAb-mix (mix of representative mAbs from 5 groups: S1E2, S2B10, LN1F9, LN3D3, and HENV-32, 2 μg/mL of each antibody) were used to calculate relative expression levels of G point mutations compared to the wildtype G.
For the competitive flow cytometric assay, the protocol was adapted from previous studies with modifications 30,38 . CHO-K1 cells first were transfected with plasmids encoding full-length NiV M G or co-transfected with plasmids encoding both full-length NiV M G and F. At 24 h posttransfection, the cells were dissociated from the plates using 5 mM EDTA. After washing, the cells were incubated with 10 μg/mL of the indicated NiV G mAbs, isotype control mAb T3D9 (an anti-SFTSV mAb, unpublished), or FACS buffer without mAb (negative control) for 30 min. Subsequently, 50 μg/mL of biotinylated human ephrin-B2 were added to the cells and incubated for 30 min, followed by washing and incubation with Streptavidin-APC (BD Biosciences, Cat#: 554067) for 30 min. The cells were then washed again, resuspended, and analyzed by flow cytometry (Cyto-FLEX S, Beckman). The binding of ephrin-B2 to the G-or G/F-expressing cells in the presence of mAbs was compared with negative control cells incubated with FACS buffer.
## Authentic virus neutralization assay
The NiV G mAbs were prepared in a three-fold serial dilution in DMEM containing 2% FBS and incubated with either 100 TCID 50 NiV M , 100 TCID 50 NiV B , or 100 TCID 50 HeV for 1 h at 37 °C. The virus-mAb mixtures were then added to Vero E6 cells in 96-well plates and incubated for 1 h at 37 °C. After incubation, the cells were washed and cultured in DMEM supplemented with 2% FBS until the cytopathic effect was observed approximately five days post-infection. Each mAb dilution was set up in four replicates. IC 50 values were calculated using IBM SPSS Statistics 27.
## Crystallization and structure determination of NiV G-LN1F9 Fab complex
The purified NiV M G head domain was mixed with LN1F9 Fab in a molar ratio of 1:1.2, and the complex was further purified to homogeneity by size exclusion chromatography using a HiLoad 16/600 Superdex 200 pg column (Cytiva). Crystallization of the NiV G-LN1F9 complex was performed by sitting drop vapor diffusion at 16 °C. Typically, 25 or 35 mg/mL protein was mixed with the precipitant/reservoir solution at a 1:1 volume ratio in a 0.6-μl drop. Crystals appeared in the precipitant/reservoir solution of 0.3 M ammonium formate, 0.1 M HEPES pH 7.0, and 20% (vol/vol) Sokalan CP 5 within 1 week. Crystals were stepwise transferred to a cryostabilizer solution (precipitant solution supplemented with 35% [vol/vol] glycerol) and then flash-cooled in liquid nitrogen before data collection.
The X-ray diffraction data were collected at the BL10U2 beamline of Shanghai Synchrotron Radiation Facility (SSRF) with a wavelength of 0.9792 Å and a temperature of 100 K. A total of 360 degrees of data were collected in 0.5°oscillation steps. The diffraction data were automatically processed by the pipeline Xia2 46 at the beamline and scaled with Aimless 47 in the CCP4 suite 48 . Phasing was obtained by molecular replacement using PHASER 49 with the crystal structure of NiV G H (PDB ID: 7TXZ) and the AlphaFold2-predicted Fab model as search models 50 .
## Cryo-EM sample preparation
NiV M G head domain was incubated with a molar excess of S1E2, S2B10, and LN3D3 Fabs in a buffer containing 20 mM Tris-HCl pH8.0 and 150 mM NaCl on ice for 1 h. The complex was further purified on a Superdex 200 Increase 10/300 GL column (Cytiva) before cryo-EM grid preparation. Cryo-EM grids were prepared on a Thermo Scientific Vitrobot Mark IV at 4 °C and 100% humidity. A total of 3.5 μL purified complex was applied to a freshly glow-discharged Cu 200 mesh R1.2/1.3 holey carbon grid (Quantifoil). After incubation for 20 s, the grids were blotted for 2 s at 100% humidity and 4 °C, and plunge-frozen in liquid ethane.
## Cryo-EM data collection and image processing
All data were collected using the CRYO ARM 300 electron microscope (JEOL, Japan) equipped with a K3 direct electron detector (Gatan, USA). Cryo-EM movies were recorded automatically using Serial-EM software in a super-resolution mode with a pixel size of 0.475 Å/pixel at a calibrated magnification of ×50,000 over a defocus range of -0.5 to -2.5 μm. Data were collected at a frame rate of 40 frames per second with a total electron dose of 40 e/Å 2 .
Recorded movies were input into cryoSPARC for patch motion correction and CTF estimation 51 . 4784 micrographs were selected for further data processing. Particles were picked using the Topaz picker. 2,248,175 particles were extracted for 2D classification using a particle box size of 300 pixels. After two rounds of 2D classification, 1,207,684 particles were selected for two rounds of heterogeneous refinement. One class (145,635 particles) from the second round of heterogeneous refinement with good features was selected for nonuniform refinement (NU-refinement), yielding an overall resolution of 3.01 Å map.
## Model building
For the cryo-EM structure, the NiV M G head domain model and AlphaFold2-predicted Fab models of variable regions were docked into the cryo-EM map using Chimera. For the crystal structure, initial models were fitted into the density in Coot. Iterative model building and refinement were performed in Coot 52 and PHENIX 53 . The data collection and refinement statistics for the final models are listed in Supplementary Tables 1,2.
## Fusion inhibition assay
A dual-functional split-reporter system, which includes RL-DSP1-7 and RL-DSP8-11 expression vectors, was used for the fusion inhibition assay as described previously 33,44 . Briefly, 293 T cells were seeded into 6-well plates one day before transfection. Effector cells were co-transfected with the fulllength NiV M G, F, and RL-DSP1-7 expression vectors, while the target cells were transfected with the RL-DSP8-11 expression vector. 6 h post-transfection, the effector and target cells were trypsinized and mixed into 96-well plates. NiV G mAbs/Fabs diluted in DMEM containing 10% FBS were added to the cells. Cells treated with DMEM containing 10% FBS without mAbs/Fabs were used as controls. For luciferase activity detection, after approximately 14-16 h of incubation, the cell culture medium was discarded and replaced with fresh DMEM containing 10% FBS and 20 μM EnduRen live-cell substrate (Promega, Cat#: E6482). The cells were incubated for at least 2 h, and live-cell luciferase activity induced by glycoprotein-mediated cell-cell fusion was detected using a Varioskan LUX (Thermo Scientific). For GFP detection, the cells were fixed by 4% PFA after 24-36 h of incubation, and the nuclei were stained with Hoechst 33342 (Thermo Scientific, R37165). Images of the same position in different experimental wells were captured using CTL-S6 Universal M2.
## Animal experiments-Syrian hamsters
Six-week-old female Syrian hamsters from Beijing Vital River Laboratories were randomly divided into five groups (n = 6). For prophylactic treatment, hamsters were administered the indicated mAbs at 30 mg/kg via the intraperitoneal (i.p.) route, followed by a challenge with 1000 LD 50 of NiV Malaysia strains via i.p. injection 24 h later. For therapeutic treatment, hamsters were challenged with 1000 LD 50 NiV M via i.p. injection. Then, hamsters were injected with tested NiV G mAb or isotype control mAb S2A5 (an anti-SFTSV mAb 33 ) twice at a dosage of 15 mg/kg through the i.p. route, once on day 1 and again on day 3 post-challenge. Hamsters were monitored daily post-challenge for survival over three weeks and for weight changes over two weeks. All animals were anesthetized with 5% isoflurane prior to mAb or virus administration. At the end of the study, animals were euthanized by cervical dislocation under isoflurane anesthesia.
## References
1. Faus-Cotino, Reina, Pueyo (2024) "Nipah Virus: A multidimensional update" *Viruses*
2. Gómez Román (2022) "Medical countermeasures against henipaviruses: a review and public health perspective" *Lancet Infect. Dis*
3. Eaton, Broder, Middleton et al. (2006) "Hendra and Nipah viruses: different and dangerous" *Nat. Rev. Microbiol*
4. Sharma, Kaushik, Kumar et al. (2019) "Emerging trends of Nipah virus: a review" *Rev. Med Virol*
5. Conroy (2023) "Nipah virus outbreak: what scientists know so far" *Nature*
6. Amaya, Broder (2020) "Vaccines to Emerging Viruses: Nipah and Hendra" *Annu Rev. Virol*
7. Spengler, Lo, Welch et al. (2024) "Henipaviruses: epidemiology, ecology, disease, and the development of vaccines and therapeutics" *Clin. Microbiol. Rev. e*
8. Nikolay (2019) "Transmission of Nipah Virus-14 Years of Investigations in Bangladesh" *N. Engl. J. Med*
9. Mishra, Prajapat, Nayak (2024) "Advancements in Nipah virus treatment: analysis of current progress in vaccines, antivirals, and therapeutics" *Immunology*
10. Aguilar, Iorio (2012) "Henipavirus membrane fusion and viral entry" *Curr. Top. Microbiol Immunol*
11. Wang (2022) "Architecture and antigenicity of the Nipah virus attachment glycoprotein" *Science*
12. Bowden (2008) "Structural basis of Nipah and Hendra virus attachment to their cell-surface receptor ephrin-B2" *Nat. Struct. Mol. Biol*
13. Xu (2008) "Host cell recognition by the henipaviruses: crystal structures of the Nipah G attachment glycoprotein and its complex with ephrin-B3" *Proc. Natl Acad. Sci. USA*
14. Negrete (2005) "EphrinB2 is the entry receptor for Nipah virus, an emergent deadly paramyxovirus" *Nature*
15. Xu (2015) "Crystal structure of the pre-fusion Nipah virus fusion glycoprotein reveals a novel hexamer-of-trimers assembly" *PLoS Pathog*
16. Dang (2019) "An antibody against the F glycoprotein inhibits Nipah and Hendra virus infections" *Nat. Struct. Mol. Biol*
17. Wong, Paterson, Lamb et al. (2016) "Structure and stabilization of the Hendra virus F glycoprotein in its prefusion form" *Proc. Natl Acad. Sci. USA*
18. Lee, Ataman (2011) "Modes of paramyxovirus fusion: a Henipavirus perspective" *Trends Microbiol*
19. Liu (2013) "Unraveling a three-step spatiotemporal mechanism of triggering of receptor-induced Nipah virus fusion and cell entry" *PLoS Pathog*
20. Liu (2015) "Nipah virus attachment glycoprotein stalk C-terminal region links receptor binding to fusion triggering" *J. Virol*
21. Wong (2017) "Monomeric ephrinB2 binding induces allosteric changes in Nipah virus G that precede its full activation" *Nat. Commun*
22. Lu (2023) "Both chimpanzee adenovirus-vectored and DNA vaccines induced long-term immunity against Nipah virus infection" *NPJ Vaccines*
23. Lu (2023) "Vaccines based on the fusion protein consensus sequence protect Syrian hamsters from Nipah virus infection" *JCI Insight*
24. Zhou (2024) "An attachment glycoprotein nanoparticle elicits broadly neutralizing antibodies and protects against lethal Nipah virus infection" *NPJ Vaccines*
25. Byrne (2023) "Structural basis for antibody recognition of vulnerable epitopes on Nipah virus F protein" *Nat. Commun*
26. Dang (2021) "Broadly neutralizing antibody cocktails targeting Nipah virus and Hendra virus fusion glycoproteins" *Nat. Struct. Mol. Biol*
27. Avanzato (2019) "A structural basis for antibody-mediated neutralization of Nipah virus reveals a site of vulnerability at the fusion glycoprotein apex"
28. Avanzato (2024) "A monoclonal antibody targeting the Nipah virus fusion glycoprotein apex imparts protection from disease" *J. Virol*
29. Mire (2020) "A cross-reactive humanized monoclonal antibody targeting fusion glycoprotein function protects ferrets against lethal Nipah Virus and Hendra virus infection" *J. Infect. Dis*
30. Dong (2020) "Potent henipavirus neutralization by antibodies recognizing diverse sites on Hendra and Nipah virus receptor binding protein" *Cell*
31. Fan (2024) "A potent Henipavirus cross-neutralizing antibody reveals a dynamic fusion-triggering pattern of the G-tetramer" *Nat. Commun*
32. Geisbert (2014) "Therapeutic treatment of Nipah virus infection in nonhuman primates with a neutralizing human monoclonal antibody" *Sci. Transl. Med*
33. Ren (2024) "A broadly protective antibody targeting glycoprotein Gn inhibits severe fever with thrombocytopenia syndrome virus infection" *Nat. Commun*
34. Wang (2024) "Fully human single-domain antibody targeting a highly conserved cryptic epitope on the Nipah virus G protein" *Nat. Commun*
35. Wang (2022) "Potent monoclonal antibody-mediated neutralization of a divergent Hendra virus variant" *Proc. Natl Acad. Sci. USA*
36. Chen (2024) "Potent human neutralizing antibodies against Nipah virus derived from two ancestral antibody heavy chains" *Nat. Commun*
37. Xu (2013) "Crystal structure of the Hendra virus attachment G glycoprotein bound to a potent cross-reactive neutralizing human monoclonal antibody" *PLoS Pathog*
38. Doyle (2021) "Cooperativity mediated by rationally selected combinations of human monoclonal antibodies targeting the henipavirus receptor binding protein" *Cell Rep*
39. Kim (2019) "Protective antibodies against Eastern equine encephalitis virus bind to epitopes in domains A and B of the E2 glycoprotein" *Nat. Microbiol*
40. Chen (2023) "Broadly neutralizing antibodies to SARS-CoV-2 and other human coronaviruses" *Nat. Rev. Immunol*
41. Rey, Stiasny, Vaney et al. (2018) "The bright and the dark side of human antibody responses to flaviviruses: lessons for vaccine design" *EMBO Rep*
42. Borisevich (2016) "Escape From Monoclonal Antibody Neutralization Affects Henipavirus Fitness In Vitro and In Vivo" *J. Infect. Dis*
43. Aguilar (2007) "Polybasic KKR motif in the cytoplasmic tail of Nipah virus fusion protein modulates membrane fusion by inside-out signaling" *J. Virol*
44. Xiong (2022) "Close relatives of MERS-CoV in bats use ACE2 as their functional receptors" *Nature*
45. Lefrancios, Lyles (1982) "The interactionof antiody with the major surface glycoprotein of vesicular stomatitis virus. I. Analysis of neutralizing epitopes with monoclonal antibodies" *Virology*
46. Gildea (2022) "xia2.multiplex: a multi-crystal data-analysis pipeline" *Acta Crystallogr. D: Struct. Biol*
47. Evans, Murshudov (2013) "How good are my data and what is the resolution?" *Acta Crystallogr. D: Biol. Crystallogr*
48. Winn (2011) "Overview of the CCP4 suite and current developments" *Acta Crystallogr. D: Biol. Crystallogr*
49. Mccoy (2007) "Phaser crystallographic software" *J. Appl. Crystallogr*
50. Jumper (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
51. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat. Methods*
52. Emsley, Cowtan (2004) "Coot: model-building tools for molecular graphics" *Acta Crystallogr. D. Biol. Crystallogr*
53. Adams (2010) "PHENIX: a comprehensive Python-based system for macromolecular structure solution" *Acta Crystallogr. D: Biol. Crystallogr* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548452&blobtype=pdf | # Sensing of Ebinur Lake virus by distinct pattern recognition receptors dictates cell-type specific innate immunity and pathogenesis
Jia-Peng Zou, Su-Yun Wang, Han Xia, Zhi-Sheng Xu, Wei-Wei Luo, Yan-Yi Wang
## Abstract
Ebinur Lake virus (EBIV) is a recently identified orthobunyavirus with broad host range and zoonotic potential, posing a public health risk. However, the mechanisms underlying EBIV pathogenesis and host innate immune responses remain unclear. Here, we investigated the pattern recognition receptors (PRRs) responsible for sensing EBIV infection and subsequent pathogenesis. EBIV infects diverse cell types and exhibits broad tissue tropism in vivo. In vitro, RIG-I was essential for type I interferon (IFN-I) and inflammatory responses in HEK293 and A549 cells. In contrast, both RIG-I and MDA5 contributed to IFN-I induction in Huh-7 and HCT116 cells, correlating with the specific accumulation of viral dsRNA intermediates in these cell types. Both RIG-I and MDA5 preferentially recognize RNA derived from the viral S segment; however, they have different abilities in sensing incoming viral genomic RNA bearing a 5'-phosphate motif and the replication intermediates. In vivo, RIG-I deficiency severely impairs host defense, while MDA5 deficiency has a more restricted effect in the spleen and liver. In addition to RIG-I and MDA5, TLR7, which is predominantly expressed in dendritic cells, also plays a crucial role for host defense by mediating systemic inflammatory cytokine production without significantly impacting IFN-I response. Our findings suggest that multiple innate sensing receptors, including RIG-I, MDA5, and TLR7, are differentially involved in host defense against EBIV by mediating IFN-I and inflammatory responses, respectively, in a cell-specific manner. IMPORTANCE This study elucidates the complex mechanisms by which host RIG-I, MDA5, and TLR7 sense the emerging EBIV and trigger cell-specific immune responses. These findings not only clarify crucial aspects of EBIV-host interactions, particularly the differential sensing of viral RNA by distinct PRRs, but also underscore how this differential sensing dictates cell-specific innate immune activation (IFN-I vs. inflammatory responses) and viral pathogenesis, providing critical insights for understanding and combating EBIV and related emerging bunyaviruses. KEYWORDS pattern recognition receptors, pathogen-associated molecular patterns, innate immunity, host-pathogen interactions, orthobunyavirus T he emergence of pathogenic viruses poses a critical challenge to public health (1).Ebinur Lake virus (EBIV) is a recently identified member of the Bunyavirales order (2), a group of viruses known for their segmented RNA genomes and their potential to cause severe diseases in humans and animals (3). EBIV was first isolated from Culex mosquitoes in the Ebinur Lake region of Northwest China (2, 4). Recent studies have demonstrated that EBIV can efficiently replicate and cause cytopathic effects in a wide variety of cells derived from mosquitoes, rodents, monkeys, and humans (4, 5). Moreover, EBIV infection at low dose (one plaque-forming unit [PFU]) causes necrosis and hemorrhage of multiple
organs, including the liver, lung, intestine, and kidney, and death in wild-type BALB/c mice (4,5). Serological surveys among local residents have shown positive antibody response against EBIV, indicating its potential risk for public health (4). Despite these findings, the mechanisms of viral pathogenesis and host innate immune response to EBIV infection remain poorly understood.
Viral infection is sensed by pattern recognition receptors (PRRs) of the host innate immune system, which recognize the conserved viral structural components called pathogen-associated molecular patterns (PAMPs). Sensing of PAMPs by PRRs triggers innate immune signaling pathways that lead to induction of downstream antiviral genes, including type I interferons (IFN-I) and proinflammatory cytokine genes (6). Among the known PRRs, several are specialized in recognizing viral RNA, including members of the Toll-like receptor (TLR), RIG-I-like receptor (RLR) and NOD-like receptor (NLR) families, and double-stranded RNA activated protein kinase R (PKR) (6). TLRs are receptors located on cell membranes or within endosomal compartments, among which TLR3 recognizes dsRNA, a replication intermediate for many RNA viruses (7). Endosomal TLR7 and TLR8 detect single-stranded RNA (ssRNA) and are involved in recognizing viral RNA derived from various viruses, such as influenza virus and SARS-CoV-2 (7). Recent studies show that some NLRs, such as NLRP6 and NLRP9b, in complex with specific DEAH-box RNA helicases, can sense viral dsRNA and activate type I IFN-and/or inflammasome-dependent antiviral responses (8,9). In addition, PKR also recognizes dsRNA, leading to the phosphorylation and inactivation of eIF2α to inhibit protein synthesis in infected cells (10).
RLRs, including RIG-I (retinoic acid-inducible gene I) and MDA5 (melanoma differentiation-associated protein 5), are sensors that detect viral RNA in the cytosol, where many RNA viruses replicate (11). These receptors share a homologous structure, including a central helicase domain and a carboxy-terminal domain (CTD) that is responsible for RNA recognition. RIG-I and MDA5 additionally have two N-terminal caspase activation and recruitment domains (CARDs) that are essential for signal transduction (12,13). Although structurally homologous, RIG-I and MDA5 discriminate among different ligands to trigger innate immune response. RIG-I recognizes short dsRNA and ssRNA with a 5′-triphosphate end, which is common in viral, but not, host RNAs (14,15). In the case of DNA virus infection, poly(dA:dT) fragments of viral genomic DNA can be transcri bed by RNA polymerase III, and the resulting RNA is detected by RIG-I to trigger innate immune response (16,17). MDA5 preferentially detects long irregular dsRNA that presumably exists in a complex higher-order configuration (18,19). Accumulating evidence demonstrates that RIG-I and MDA5 can act distinctly or cooperatively to sense infection of certain viruses. Negative-strand (ns) RNA viruses of the Paramyxoviridae, Orthomyxoviridae, Rhabdoviridae, Bunyaviridae, and Filoviridae families are mainly sensed by RIG-I, whereas positive-strand RNA viruses of the Picornaviridae family are sensed by MDA5 (18,20,21). The sensing of Reoviridae and Flaviviridae family members is mediated by both RIG-I and MDA5, highlighting the collaboration of the two receptors in antiviral immunity (21,22). Upon binding to viral RNA, RIG-I or MDA5 undergoes conformational change, leading to its recruitment to the CARD domain-containing adaptor VISA (also called MAVS, IPS-1, and Cardif ), located either at the mitochondria or the peroxisomes (23)(24)(25)(26)(27). This triggers signaling that results in translocation of the transcription factors IRF3 and NF-κB into the nucleus, initiating transcriptional induction of downstream antiviral effectors including type I IFNs and proinflammatory cytokines. Secreted type I IFNs bind to their receptors, activating the JAK/STAT signaling pathways to induce the expression of numerous IFN-stimulated genes (ISGs) (28). The ISGs encode various proteins that inhibit viral replication, induce cell death, boost antigen presentation, and modulate the immune response (28).
In this study, we demonstrate that EBIV infects a broad range of human and murine cell types. We show that RIG-I, MDA5, and TLR7 are differentially involved in sensing of EBIV in different cell types, which are determined by exposure to differential PAMPs during EBIV infection, including the viral genomic RNA containing 5' triphosphate or bisphosphate motif and dsRNA produced during replication. Experiments with gene knockout mice reveal that RIG-I, MDA5, and TLR7 differentially contribute to host innate immune and inflammatory responses in responding to EBIV. Our findings elucidate the precise mechanisms of innate immune sensing and signaling of EBIV, which is critical to understand EBIV-host interactions.
## RESULTS
## Tropism of EBIV in cells and tissues
To investigate the tissue tropism of EBIV in vivo, we systematically collected various tissues (kidney, spleen, intestine, brain, lung, liver, and lymph nodes) from C57BL/6 mice infected with EBIV and assessed viral presence using quantitative PCR (qPCR) for viral RNA and immunohistochemistry (IHC) with an anti-NP antibody. The results indicated that EBIV infected and replicated in a broad range of tissues in vivo (Fig. 1A andB). To complement tissue-level analysis, we examined EBIV infectivity in primary murine cells. The results indicated that mouse lung fibroblasts (MLFs), mouse embry onic fibroblasts (MEFs), and mouse bone marrow-derived dendritic cells (BMDCs) were susceptible to EBIV infection, while mouse bone marrow-derived macrophages (BMDMs) were largely non-permissive (Fig. S1A through C). Next, we evaluated the infectivity of EBIV in human cells. We first evaluated the infection and replication of GFP-tagged EBIV in various human cell lines by monitoring GFP expression levels and the percent age of GFP-positive cells. Our results indicated that seven human cell lines, including brain glioblastoma U-87 MG, liver carcinoma Huh-7, embryonic kidney HEK293, cervical epithelial cancer HeLa, adrenal gland and cortex epithelial SW-13, colon cancer HCT116, and lung epithelial A549 cells, were susceptible to EBIV infection to varied degrees, whereas minimal infection was observed in human monocytic THP-1 and U937 cells (Fig. 1C andD). Moreover, cytopathic effects were observed in the susceptible cells 48-60 hours post-infection with EBIV (Fig. S1D). Productive replication in these cell lines was confirmed by measuring progeny virus titers (Fig. 1E) and detecting viral S, M, and L segment RNAs and N protein expression (Fig. 1F). We also further tested the infection tropism of EBIV in primary human cells. We found that human umbilical vein endothelial cells (HUVECs) were permissive to EBIV infection, whereas human foreskin fibroblasts (HFFs) and peripheral blood mononuclear cells (PBMCs) were largely non-permissive under our experimental conditions (Fig. 1E; Fig. S1D). Together, these results suggest that EBIV exhibits a broad tropism, infecting multiple tissues in vivo as well as diverse human and murine cell lines and specific primary cell types, including endothelial cells and dendritic cells.
## EBIV triggers innate immune response in infected cells
Sensing of viral infection by PRRs leads to induction of numerous antiviral genes and innate immune response. To determine whether EBIV infection effectively activates cellular innate immune response, we examined the activation of PRR-triggered signaling pathways in HEK293 cells, which are highly susceptible to EBIV infection. We found that phosphorylation of IKKα/β, p65, TBK1, and IRF3, which are hallmarks of activation of innate immune signaling, was markedly induced following EBIV infection in HEK293 cells (Fig. 2A). Notably, phosphorylation of IKKα/β and p65, which are essential for activation of the NF-κB signaling pathway, was initiated at 24 hours after EBIV infection, whereas phosphorylation of TBK1 and IRF3, which drive the expression of ISRE-containing genes, was only detectable at 48 hours after EBIV infection (Fig. 2A). Consistently, EBIV infection induced the transcription of downstream IFNB1, CXCL10, and TNF genes in HEK293 and Huh-7 cells at 24-48 hours after infection (Fig. 2B). To investigate the contribution of viral replication to this response, we treated EBIV with UV light or heat prior to infection. UV treatment, which primarily damages viral nucleic acids, significantly reduced (by ~60%) the virus's ability to induce IFNB1 transcription, while heat treatment, which denatures proteins and inhibits entry/replication, completely abolished induction (Fig. 2C). To confirm this, we examined the effects of viral RNA polymerase (RdRp) inhibitor Favipiravir (T705) on EBIV-induced innate immune signaling. The results indicated that pretreat ment of cells with Favipiravir inhibited EBIV replication in a dose-dependent manner and correspondingly suppressed EBIV-induced IFNB1 mRNA expression (Fig. 2D). These results suggest that viral replication is required for efficiently initiating innate immune response. We next assessed whether the incoming viral genome itself possesses immunostimulatory activity. Indeed, transfection of purified viral genomic RNA (vRNA) into cells induced IFNB1 transcription in a dose-dependent manner compared to control RNA (Fig. 2E). Collectively, these results demonstrate that EBIV infection activates canonical innate immune signaling pathways, and while incoming vRNA can trigger a baseline response, active viral replication is necessary for the full induction of antiviral genes.
## RLRs are essential for antiviral innate immune response to EBIV in vivo
We next investigated the in vivo roles of the RLR sensors RIG-I and MDA5, and their essential adaptor VISA, in host defense against EBIV. We infected 8-week-old WT, RIG-I-deficient (Rig-I -/-), MDA5-deficient (Mda5 -/-), and VISA-deficient (Visa -/-) mice with a lethal dose of EBIV intraperitoneally (i.p.) and monitored their survival. Compared to wild-type mice, Rig-I -/-mice were highly susceptible to EBIV, showing markedly accelerated mortality compared to WT mice infected with EBIV. Mda5 -/-mice displayed increased susceptibility compared to WT mice but were less susceptible compared to Rig-I -/-mice. In these experiments, Visa -/-mice exhibited the highest susceptibility and lethality to EBIV infection among all the examined groups (Fig. 3A). Consistently, knockout of RIG-I significantly reduced EBIV-induced secretion of serum IFN-β, IP-10, IL-6, and TNF-α compared to wild-type mice (Fig. 3B). In contrast, knockout of MDA5 moderately reduced EBIV-induced secretion of IFN-β, IP-10, and IL-6, but did not have marked effects on serum TNF-αlevel (Fig. 3B). In these experiments, knockout of VISA reduced EBIV-induced secretion of serum IFN-β, IP-10, IL-6, and TNF-α to lower levels than knockout of either RIG-I or MDA5 (Fig. 3B). Consistent with impaired antiviral responses, viral replication and production, assessed by S segment RNA levels and measuring progeny virus titers, were significantly elevated in multiple tissues (including spleen, liver, kidney, and lung) of Rig-I -/-and Visa -/-mice compared to WT controls (Fig. 3C andD). A significant increase in viral loads in Mda5 -/-mice compared to wild-type mice was only detected in the spleen and liver, but not in other examined tissues (Fig. 3C andD). Correspondingly, histopathological analysis revealed more severe inflammatory infiltrates and tissue damages in the spleen, liver, kidney, and lung of Visa -/-mice compared to WT mice, correlating with the higher viral burdens in these tissues (Fig. 3E). RIG-I deficiency resulted in significantly increased pathology scores specifically in the spleen and lung, but not in the liver, kidney, or brain compared to WT mice (Fig. 3E). Notably, MDA5 deficiency led to significantly elevated pathology scores only in the spleen (Fig. 3E). The observation that VISA deficiency exacerbates pathology in the liver and kidney, while deficiency of either RIG-I or MDA5 does not, suggests potential complementary or redundant roles for RIG-I and MDA5 in limiting immunopathology specifically within these organs. Collectively, these results demonstrate that VISA-dependent signaling is critical for controlling EBIV replication and limiting immunopathology in vivo, with RIG-I playing a major role, while MDA5 appears more restricted in certain tissues like the spleen and liver.
## TLR7 is important for inflammatory response induced by EBIV in DCs
The above in vivo experiments indicated that deficiency of the RLR-VISA axis did not completely abolish EBIV-induced production of innate immune and inflammatory cytokines (Fig. 3B). This led us to hypothesize that additional innate immune sensors are involved in host defense to EBIV. Recent studies suggest that certain RNA viruses induce mitochondrial stress, leading to release of mtDNA into the cytosol to trigger innate immune and inflammatory response via the cGAS-MITA/STING axis (29)(30)(31). However, infecting cGas -/-and Mita -/-mice with lethal doses of EBIV revealed comparable lethality to WT mice (Fig. 4A), suggesting that the cGAS-MITA axis is dispensable for host defense against EBIV. We next considered endosomal TLRs known to recognize viral RNA, namely TLR3, TLR7, and TLR8 (7). Since these TLRs signal through the adaptor proteins MyD88 and/or TRIF (7), we examined the roles of these adaptors in host defense against EBIV in vivo. MyD88-deficient mice exhibited significantly increased susceptibility to EBIV compared to WT mice, whereas TRIF-deficient mice showed comparable lethality (Fig. 4A). Given that TLR7/8 signals through MyD88, while TLR3 signals through TRIF (7), we therefore hypothesized that TLR7 and/or TLR8, but not TLR3, were involved in antiviral response in EBIV-infected mice. Subsequent experiments showed that TLR7-deficient mice were significantly more susceptible to EBIV infection, while TLR8-deficient mice exhibited survival similar to WT mice (Fig. 4A). These results suggest that TLR7-MyD88 signaling plays a role in host defense against EBIV in vivo. Consistent with increased susceptibility, Tlr7 -/-mice displayed significantly higher viral burden compared to WT mice at 48 hours post-infection, specifically in the spleen, lung, and liver, as measured by both S segment RNA levels (Fig. 4B) and infectious virus titers (TCID50) (Fig. 4C). In contrast, viral loads and titers in the kidney, brain, and intestine were not significantly different between Tlr7 -/-and WT mice (Fig. 4B andC). Correspondingly, histopathological analysis revealed significantly higher pathology scores in the spleen and lung of Tlr7 -/-mice compared to WT controls, while scores in the liver, kidney, and brain were comparable between the genotypes (Fig. 4D). These findings suggest that TLR7 plays a tissue-specific role in controlling EBIV replication and limiting associated pathology.
Since TLR7 is predominantly expressed in dendritic cells (DCs) (7) and EBIV efficiently infects BMDCs (Fig. S1C), we investigated TLR7-MyD88 signaling in EBIV-infected BMDCs. qPCR revealed that EBIV-induced transcription of Ifnb1 and Cxcl10 was impaired in BMDCs lacking RIG-I or VISA, but not MDA5 (Fig. 4E). Conversely, RIG-I or VISA deletion had minimal impact on the induction of inflammatory genes Tnf and Il6 in BMDCs (Fig. 4E). In contrast, knockout of TLR7 or MyD88 did not affect expression of Ifnb1 and Cxcl10 but significantly inhibited expression of Tnf and Il6 induced by EBIV (Fig. 4E). This pattern was mirrored in systemic cytokine production in mice. Serum levels of the inflammatory cytokines TNF-α and IL-6 were markedly reduced in Tlr7 -/-or Myd88 -/-mice compared to WT mice post-infection (Fig. 4F). Importantly, knockout of TLR7 or MyD88 did not significantly affect EBIV-induced serum levels of IFN-β or IP-10 (Fig. 4F). Together, these findings suggest that the TLR7-MyD88 axis plays a crucial role specifically in the inflammatory cytokine response to EBIV, thereby complementing the primary antiviral IFN/ISG response driven by the RLR-VISA pathways.
## RIG-I is essential for innate immune response to EBIV in HEK293 and A549 cells
Having established the critical roles of RLR-VISA and TLR7-MyD88 signaling in vivo, we next investigated the specific contributions of the RLR sensors RIG-I and MDA5 in mediating the innate immune response within different human cell lines in vitro. We used CRISPR-Cas9 to knock out RIG-I, MDA5, or VISA in HEK293 cells and assessed their roles in innate immune response to EBIV. RT-qPCR results showed that knockout of RIG-I or VISA reduced the induction of IFNB1, CXCL10, and TNF genes induced by EBIV in HEK293 cells, whereas knockout of MDA5 had no marked effects (Fig. 5A). Additionally, EBIV-induced phosphorylation of TBK1, IRF3, and p65 was eliminated in RIG-I-and VISAdeficient HEK293 cells, whereas knockout of MDA5 had no marked effects (Fig. 5B). These results suggest that the RIG-I-VISA axis, rather than the MDA5-VISA axis, is required for mounting an effective innate immune response to EBIV in HEK293 cells. The essential role of RIG-I-VISA in driving the downstream response was further confirmed by rescue experiments, where re-expression of RIG-I in RIG-I -/-HEK293 cells or VISA in VISA -/- HEK293 cells restored IFN-β induction upon EBIV infection (Fig. S2A). Consistent with these findings, infection and replication of EBIV were enhanced in RIG-I-and VISAdeficient HEK293 cells, as evidenced by the elevated mRNA level of S segment and increased titers of progeny viruses in the culture supernatants (Fig. 5C andD). These findings suggest that RIG-I-VISA-mediated innate immune responses restrict EBIV replication in HEK293 cells. Similarly, knockout of RIG-I, but not MDA5, impaired the transcription of IFNB1, CXCL10, and TNF genes induced by EBIV in A549 cells (Fig. S2B). Furthermore, replication and production of EBIV were increased in RIG-I-but not MDA5deficient A549 cells (Fig. S2C andD). These results demonstrate that RIG-I, signaling via VISA, is the essential cytoplasmic sensor responsible for triggering the type I IFN response to EBIV infection in human kidney and lung epithelial cell lines.
## Both RIG-I and MDA5 are involved in innate immune response to EBIV in Huh-7 and HCT116 cells
In contrast to our findings in HEK293 and A549 cells, results from Huh-7 cells revealed a different pattern of innate immune recognition of EBIV. We found that the knockout of both RIG-I and MDA5 only partially impaired the transcription of IFNB1, CXCL10, and TNF genes, as well as the phosphorylation of TBK1, IRF3, and p65 induced by EBIV in Huh-7 cells (Fig. 5E andF). However, expression of downstream genes and phosphorylation of these proteins induced by EBIV were completely blocked in VISA-deficient cells (Fig. 5E andF). Rescue experiments validated these findings, showing that re-expression of either RIG-I, MDA5, or VISA in their respective knockout cells restored IFN-β induction (Fig. S2E). In contrast, replication and production of EBIV were increased in both RIG-Iand MDA5-deficient Huh-7 cells, as indicated by elevated mRNA levels of S segment and higher titers of progeny viruses in the culture supernatants (Fig. 5G andH). Knockout of VISA resulted in an even greater increase in replication and production of EBIV compared to the knockout of RIG-I or MDA5 (Fig. 5G andH). Similar results were obtained in HCT116 cells. Knockout of RIG-I or MDA5 partially inhibited EBIV-induced transcription of IFNB1, CXCL10, and TNF genes, which was completely abolished in VISA-deficient HCT116 cells (Fig. S2F). Enhanced EBIV production was observed in these cells, where viral replication was higher in VISA-deficient cells than in RIG-I-and MDA5-deficient cells (Fig. S2G andH). These results suggest that RIG-I and MDA5 are functionally redundant upstream of VISA in the innate immune response to EBIV in human liver and colon carcinoma cell lines.
## Specificity of recognition of EBIV RNA species by RIG-I and MDA5
vRNA or its replication intermediates are well-characterized PAMP that triggers RLR signaling (32). We aimed to determine the specific EBIV RNA species recognized by RIG-I and MDA5. Similar to other bunyaviruses, EBIV possesses a negative-sense RNA genome consisting of three segments: a large (L) segment encoding RNA-dependent RNA polymerase (RdRP), a medium (M) segment encoding glycoproteins, and a small (S) segment encoding the nucleocapsid protein and non-structural proteins (4). We performed photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipi tation (PAR-CLIP) using antibodies against RIG-I or MDA5 in EBIV-infected Huh-7 and HEK293 cells, followed by PCR and RT-qPCR analysis (Fig. 6A). In HEK293 cells, where RIG-I is the primary PRR for EBIV, RIG-I bound specifically to the S segment, but not to the M or L segments (Fig. S3A). Further experiments indicated that RIG-I predominantly bound to the 121-790 region of the S segment in HEK293 cells (Fig. 6B; Fig. S3C andS4A). MDA5 did not bind significantly to any segment in HEK293 cells (Fig. 6B; Fig. S3A andS4A), RT-qPCR analysis of the indicated genes in BMDCs from WT, Rig-i -/-, Mda5 -/-, Visa -/-, Tlr7 -/-, and Myd88 -/-mice after EBIV infection (MOI = 1) for the indicated times.
(F) ELISA analysis of the indicated cytokines in sera of WT, Tlr7 -/-, and Myd88 -/-mice infected for 48 hours by i. which was not affected by type I IFN pretreatment (Fig. S4C). In Huh-7 cells, where both RIG-I and MDA5 contribute to sensing of EBIV, both receptors bound to the S segment but not M or L segments (Fig. S3B), mapping predominantly to the nt 121-790 region (Fig. 6C; Fig. S3D andS4B). Furthermore, RNA immunoprecipitated by RIG-I from HEK293 cells, or by either RIG-I or MDA5 from Huh-7 cells, induced transcription of IFNB1 and CXCL10 genes upon transfection into cells (Fig. 6D). These results indicate that RLRs preferentially recognize RNA derived from the EBIV S segment, thereby initiating antiviral innate immune responses.
We next determined the specificity of RIG-I and MDA5 in sensing EBIV genome RNA (vRNA) and replication intermediate RNA (riRNA). We extracted RNA from EBIV virions in culture medium (vRNA), or from infected cells, which mostly contain viral replication intermediates (riRNA). Transfecting these RNAs into HEK293 or Huh-7 WT or knockout cells revealed distinct patterns (Fig. 7A). Both vRNA and riRNA derived from EBIV-infected HEK293 cells strongly induced IFNB1 transcription upon transfection, an effect depend ent on RIG-I and VISA, but not on MDA5 (Fig. 7B). Similarly, vRNA derived from infected Huh-7 cells activated responses solely via RIG-I and VISA (Fig. 7C). However, riRNA derived from infected Huh-7 cells induced IFNB1 transcription, which was partially impaired by deficiency of either RIG-I or MDA5, and completely abolished by VISA deficiency in both recipient cell types (Fig. 7C). These results suggest that incoming EBIV vRNA primarily activates RIG-I, while replication in Huh-7 cells generates distinct RNA species capable of activating both RIG-I and MDA5.
It has been reported that the recognition of viral RNA by RIG-I requires a 5' triphos phate or bisphosphate motif (15,33,34). Removing all phosphate groups using calf intestinal phosphatase (CIP) abolished transcription of the IFNB1 gene induced by vRNA of EBIV or Sendai virus (SeV) (Fig. 7D), an ssRNA virus that requires 5' triphosphate to activate RIG-I signaling (20) in HEK293 cells. Furthermore, treatment with 5'-polyphos phatase, an enzyme that sequentially removes the γ-and β-phosphates but does not affect the α-phosphate group, dramatically inhibited transcription of the IFNB1 gene by EBIV vRNA (Fig. 7D). These results suggest that the 5' triphosphate or bisphosphate motif present in EBIV vRNA is required for RIG-I activation.
Previously, it has been shown that MDA5 preferentially senses long irregular dsRNA that presumably exists in a complex higher-order configuration (18,19). Since Huh-7derived riRNA activated MDA5 while HEK293-derived riRNA did not, we investigated whether these cells have differential formation of viral dsRNA intermediates following EBIV infection. Immunofluorescence analysis using a dsRNA-specific antibody revealed detectable dsRNA accumulation in EBIV-infected Huh-7 and HCT116 cells, increasing over time (24-48 hours), but dsRNA was barely detected in infected HEK293 or A549 cells at any time point (Fig. 7E; Fig. S5A). As positive controls, encephalomyocarditis virus (EMCV), a small non-enveloped single-stranded RNA virus that generates dsRNA during infection (20,35), induced strong dsRNA staining in all cell types (Fig. 7E; Fig. S5A). Furthermore, dsRNA immunoprecipitation from infected Huh-7 cells followed by RT-qPCR confirmed the enrichment of viral S segment RNA in the dsRNA fraction (Fig. S5B). In contrast, no significant enrichment of viral S RNA was observed in infected HEK293 cells (Fig. S5B). Importantly, the host mitochondrial RNA MT-ND1 was not enriched by dsRNA-specific antibody in either cell line under infected conditions (Fig. S5B). These results indicate that the dsRNA accumulating in Huh-7 cells during EBIV infection is predominantly of viral origin rather than abundant host RNAs. Consistently, treatment of riRNA from EBIV-infected Huh-7 cells with dsRNA-specific RNase III inhibited its ability to induce IFNB1 transcription (Fig. 7F). In contrast, treatment of riRNA from EBIV-infected HEK293 cells with RNase III had no marked effects on transcription of the IFNB1 gene (Fig. 7F). In these experiments, treatment of riRNA from both EMCV-infected Huh-7 and HEK293 cells with RNase III abolished its ability to induce transcription of the IFNB1 gene (Fig. 7F). Collectively, these results suggest that EBIV infection leads to the accumulation of immunostimulatory viral dsRNA intermediates specifically in Huh-7 and HCT116 cells, which are subsequently sensed by MDA5.
## DISCUSSION
EBIV is a recently identified virus within the Bunyavirales order (2), a family of viruses known for their segmented RNA genomes and their potential to cause severe diseases in humans and animals. Recent studies have demonstrated that EBIV can successfully infect several cell lines from different species, including mosquitoes, rodents, monkeys, and humans (4,5). Serological surveys among local populations have indicated positive antibody responses against EBIV, highlighting a potential public health risk (4). However, the cellular and tissue tropisms of EBIV and the mechanisms by which it triggers innate immune responses remain poorly understood. Here, our findings reveal that EBIV infects and replicates in various human cell lines, including U-87 MG, Huh-7, HEK293, HeLa, SW-13, HCT116, HUVEC, and A549. Conversely, EBIV exhibits minimal infection in human monocytes, such as THP-1, U937, and PBMC, as well as murine macrophage BMDMs, but efficient infection in BMDCs. Furthermore, intraperitoneal injection of EBIV results in infection in several murine organs, including the kidney, spleen, intestine, brain, lung, liver, and lymph nodes. These results suggest that EBIV can infect a wide range of non-immune cell types in humans and mice but selectively infects certain immune cells, indicating the presence of specific receptors or entry pathways in a broad range of cell types.
Our findings demonstrate that multiple PRR signals are involved in antiviral innate immune responses to EBIV infection in vitro and in vivo. Our in vitro studies using knockout human cell lines delineated cell type-specific RLR sensing. In HEK293 and A549 cells, RIG-I signaling via VISA was essential for inducing type I IFN and inflammatory responses upon EBIV infection, while MDA5 appeared dispensable. This dominant role of RIG-I aligns with its established function in sensing negative-sense RNA viruses (36). In contrast, in Huh-7 and HCT116 cells, both RIG-I and MDA5 contributed to the induction of type I IFN and inflammatory responses. Deficiency of either sensor resulted in a partial reduction of the response, whereas VISA knockout completely abolished it, suggesting functional contribution and potential cooperation or redundancy between RIG-I and MDA5 upstream of VISA in these specific cell types. Our in vivo studies using knockout mice highlighted the physiological relevance and complexity of PRR engagement in controlling EBIV pathogenesis. Consistent with VISA being the central RLR adaptor, VISA deficiency resulted in the most severe phenotype, characterized by rapid mortality, broadly increased viral loads across multiple organs (spleen, liver, kidney, lung, and brain), widespread tissue pathology, and severely impaired systemic cytokine responses. RIG-I deficiency also led to high susceptibility, significantly increased viral loads in most organs tested, severe pathology (especially in the spleen and lung), and broadly impaired systemic cytokine production, confirming its major protective role in vivo. MDA5 deficiency conferred increased susceptibility (though less severe than RIG-I or VISA knockout), higher viral loads/titers primarily restricted to the spleen and liver, and significantly reduced serum IFN-β, IP-10, and IL-6, indicating a more restricted, organ-specific role for MDA5 in vivo. Notably, the observation that VISA deficiency caused severe pathology in the liver and kidney, while individual RIG-I or MDA5 deficiency did not significantly increase pathology scores in these specific organs, supports the possibility of complementary or redundant functions between RIG-I and MDA5 in limiting immunopathology in vivo, particularly within the liver and kidney. Alternatively, this finding suggests that other immune receptors may be involved in sensing EBIV. Beyond the RLR pathway, our in vivo data revealed a critical, non-redundant role for TLR7 in host defense. Tlr7 -/-mice exhibited significantly increased susceptibility to lethal EBIV infection, associated with increased viral loads specifically in the spleen, lung, and liver and exacerbated pathology scores in the spleen and lung. These data suggest TLR7 contributes to viral control in specific tissues. Mechanistically, this pathway was distinct from RLRs; TLR7/MyD88 deficiency did not impact systemic IFN-β or IP-10 levels or their gene induction in BMDCs. Instead, TLR7 and MyD88 were essential for producing key inflammatory cytokines TNF-α and IL-6, both systemically and in BMDCs. This clearly distinguishes the role of TLR7-MyD88, likely operating primarily in DCs to drive inflammation, from the RLR-VISA pathway mediating the main antiviral IFN/ISG response.
The molecular basis for this differential RLR engagement appears linked to the specific viral PAMPs generated or recognized in different cellular contexts. Our PAR-CLIP experiments indicate that RLRs preferentially bind RNA derived from the EBIV S segment. Further investigation revealed that incoming vRNA primarily activates RIG-I, likely via its 5' triphosphate or bisphosphate motif, as demonstrated by transfection and phospha tase treatment experiments. However, riRNA derived from infected cells, particularly from Huh-7 cells, could activate both RIG-I and MDA5. This dual activation by Huh-7 riRNA correlated with the detectable accumulation of dsRNA intermediates in Huh-7 and HCT116 cells, but not in HEK293 or A549 cells, during EBIV infection. This accumu lating dsRNA was confirmed to be predominantly of viral origin (S segment) rather than host mitochondrial RNA and was functionally relevant, as the immunostimulatory activity of Huh-7 riRNA was sensitive to RNase III digestion. These findings suggest that certain cellular environments, like those in Huh-7 and HCT116 cells, permit the formation or accumulation of sufficient viral dsRNA intermediates during EBIV replica tion to engage MDA5 signaling, complementing RIG-I sensing. Despite limitations in precisely separating vRNA from riRNA using cellular fractionation, the data collectively suggest that incoming vRNA is sensed by RIG-I and replication intermediates (including dsRNA in specific cells) are sensed by RIG-I and/or MDA5. The underlying reasons for cell type-specific dsRNA accumulation likely involve differences in intrinsic host factors regulating dsRNA stability or processing, such as RNA-binding proteins, helicases, or cellular ribonucleases, representing an important area for future investigation. Alterna tively, the differential RLR engagement could be determined by the stoichiometry of viral PAMPs relative to endogenous host RNA-binding proteins, some of which may be IFN-inducible. In this model, MDA5 activation might only occur when dsRNA levels surpass the binding capacity of these competing cellular factors. Thus, cell type-specific differences in the expression of these competing proteins could be a key determinant of MDA5 engagement.
This study has certain limitations. While providing valuable mechanistic insights, immortalized cell lines do not fully recapitulate the complexity of primary cells or tissue environments. Our method for distinguishing vRNA and riRNA relied on cellular fractionation and has inherent limitations. Future studies employing more sophisticated techniques to precisely identify the structure and origin of RLR ligands, and to pinpoint the specific host factors responsible for differential dsRNA accumulation, will be essential to fully elucidate the mechanisms of EBIV sensing. Many bunyaviruses suppress host innate immune responses by their NSs proteins (37)(38)(39)(40), and functionally characterizing the EBIV NSs protein will further deepen our understanding.
In conclusion, our studies have identified the cell tropisms of EBIV, a newly discovered orthobunyavirus. Our findings have elucidated the immune recognition mechanisms of EBIV by different PRRs in multiple cell types in vitro and in mice, which are determined by production of distinct PAMPs and intrinsic cellular factors during EBIV infection. Our study lays the foundation for understanding the mechanisms of EBIV-host interactions and provides a prototypic model on how RNA viruses are differentially sensed by distinct PRRs for complicated innate immune and inflammatory responses in vivo.
## MATERIALS AND METHODS
## Mice
Rig-I flox/flox (S-CKO-07307) mice were purchased from Cyagen. Mda5 -/-(T013946), Tlr7 -/- (T006737), and Tlr8 -/-(T054108) mice were purchased from GemPharmatech. cGas -/-, Mita -/-, and Visa -/-mice were kindly provided by Prof. Hong-Bing Shu (Wuhan University) as previously described (30,41). Myd88 -/-mice were kindly provided by Dr. Chunsheng Dong (Soochow University). Primer information for identification of different gene knockout mice is presented in the Table S1. All mice were housed in groups of five mice per cage under a 12-hour light/dark cycle in a temperature-controlled specific pathogen-free (SPF) room (23-25°C, relative humidity of 40%-70%) with free access to water and food. At the experimental endpoint, animals were euthanized by cervical dislocation after isoflurane anesthesia. Viral infection experiments were performed in the ABSL-2 facility at the Wuhan Institute of Virology.
## Cells and viruses
Huh-7, HUVEC, and A549 cell lines were obtained from the National Virus Resource Center (NVRC, Wuhan, China). U-87 MG, HEK293, SW-13, HeLa, HCT116, THP-1, and U937 cell lines were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). Huh-7, HUVEC, A549, U-87 MG, HEK293, SW-13, HeLa, and HCT116 were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco). THP-1 and U937 cells were cultured in RPMI 1640 medium (Hyclone). Bone marrow cells were isolated from the tibiae and femora. BMDCs were generated from mouse bone marrow cells induced by GM-CSF in vitro. In all cases, the medium was supplemented with 10% (vol/vol) fetal bovine serum (FBS) and 1% (vol/vol) penicillin-streptomycin (Gibco). The primary MEF and MLF cells were isolated and cultured as previously described (30,42). Primary human PBMCs were isolated from whole blood of healthy donors with SepMate-15 (STEMCELL Technologies), according to the manufacturer's instructions as previously described (43). Cells were maintained at 37°C in a humidified incubator containing 5% CO 2 . All cell lines were tested and found to be free of mycoplasma contamination using the MycoBlue Mycoplasma Detector Kit (#D101, Vazyme).
EBIV (Cu-XJ20 isolate) and rEBIV/eGFP/S were provided by Dr. Han Xia (Wuhan Institute of Virology). EMCV and SeV have been previously described (37).
## Reagents and antibodies
Lipofectamine 2000 (Thermo Scientific, 11668), polybrene transfection reagent (Specialty Media, TR-1003-G), protein G sepharose (Cytiva, 17061805), 4-thiouridine (MCE, HY-W011793), favipiravir (TargetMol, T6833), RNase I (Thermo Scientific, EN0601), RNase inhibitor (Invitrogen, AM2694), protein K (Thermo Scientific, 25530-049), RNA 5' polyphosphatase (Biosearch, RP8092H), Quick CIP (NEB, M0525S), and RNaseIII (Invi trogen, AM2290) were purchased from the indicated companies. Information on the commercially available antibodies used in this study is provided in the Table S2.
## CRISPR/Cas9 knockout
Gene editing was performed using the CRISPR/Cas9 system. Briefly, double-stranded oligonucleotides corresponding to the target sequences were cloned into the lenti-CRISPR-V2 vector, which was co-transfected with the packaging plasmids psPAX2 and pMD2.G into HEK293 cells. Two days after transfection, lentiviruses were harvested and used to infect target cells in the presence of polybrene (8 µg/mL). The infected cells were selected with puromycin for at least seven days. The sequences of gRNAs are provided in Table S3. Mutation and deficiency of the target gene were confirmed by Sanger sequencing and immunoblots, respectively.
## RT-qPCR
Total RNA from the cells was isolated using RNAiso Plus (#9109, TaKaRa), and reverse transcription of 1 µg of RNA was conducted using a cDNA synthesis kit (#R222, Vazyme) according to the manufacturer's instructions. For the extraction of viral RNA in cell culture supernatant, TaKaRa MiniBEST Viral RNA/DNA Extraction Kit Ver.5.0 was used (#9766, TaKaRa). RT-qPCR was performed as previously described. The threshold cycle (Ct) for the indicated genes was normalized to that of the housekeeping gene GAPDH and is presented as the relative mRNA level. Gene-specific primers used in this study are listed in Table S4.
## Immunoblots
Cells were lysed in lysis buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40) supplemented with a complete protease inhibitor mixture (Targetmol) and incubated on ice for 15 minutes. Insoluble materials were removed by centrifugation. The lysates were fractionated by SDS-PAGE and transferred to a nitrocellulose filter membrane (Millipore).
## Analysis of dsRNA by confocal microscopy
HEK293 and Huh-7 cells were infected with EBIV and EMCV for the indicated time points. After infection, the cells were fixed and stained as previously described (44). The cells were incubated with the anti-dsRNA antibody (1:200) (Nordic Mubio, 10010200) for 1 hour. After washing, Alexa Fluor 555-conjugated secondary antibody was applied at a 1:2000 dilution for 1 hour. The nuclei were stained with DAPI for 2 minutes. Images were acquired using MSHOT microscope with a 40 × lens objective.
## Flow cytometry analysis for EBIV-GFP infected cells
The cells were infected with EBIV-GFP for the indicated times, harvested, and washed with PBS. The cells were then counted (10,000 cells per sample) and analyzed using flow cytometry. Data were analyzed and visualized using FlowJo software.
## TCID50 assay
Cells were plated in 96-well plates overnight before being subjected to TCID50 assays. Briefly, EBIV stock was serially diluted from 1:10 to 1:10 6 with DMEM. Cells were incubated with 100 mL of each diluted virus stock for 1 hour. The virus was then removed, and 150 µL of DMEM containing 2% FBS was added to cells to maintain cell growth for 5 days. Cytopathic effects were observed, and viral titers were calculated using the Reed-Muench method.
## PAR-CLIP
Cells were infected with EBIV (MOI = 10) in the presence of 4SU (100 µM). The cells were washed with PBS and exposed to 0.15 J/cm 2 365 nm UV light to crosslink the labeled RNA to RNA-binding proteins. Cells were harvested and lysed in Nonidet P-40 lysis buffer (50 mM HEPES, 150 mM KCl, 1 mM NaF, 10 mM ZnCl 2 , 0.5% NP-40, 0.5 mM DTT, protease inhibitor, pH 7.5) for 10 minutes on ice. The lysate was cleared by centrifugation, and endogenous proteins were immunoprecipitated for 4 hours with the respective antibodies (1 µg/ mL). The immunoprecipitates were further digested with 100 U ml -1 RNase I. Beads were washed five times with high-salt wash buffer (50 mM HEPES, 500 mM KCl, 0.05% NP-40, 0.5 mM DTT, protease inhibitor, pH 7.5) and incubated with proteinase K (Thermo Scientific) for 30 minutes at 55°C. The RNA was isolated by phenol/chloro form/isoamyl alcohol extraction and subjected to further RT-qPCR analysis.
## Viral infection in mice
For measurement of the survival rate, age-and sex-matched mice were infected i.p. with EBIV (1 × 10 3 PFU per mouse) and monitored daily for up to 10 days. For in vivo viral infection, mice were i.p. infected with EBIV (1 × 10 3 PFU per mouse). Lungs and livers from the control or virus-infected mice were harvested 48 hours post-infection for histological analysis. The tissue mRNA levels of the indicated genes were determined by RT-qPCR assays. Mouse serum was collected 48 hours post-infection to measure cytokine production by ELISA.
## Histological and immunohistochemical analysis
Tissue samples were fixed in 4% paraformaldehyde, dehydrated in graded ethanol, cleared in xylene, and embedded in paraffin. Serial sections (2 µm) were subjected to Hematoxylin and Eosin (H&E) staining following standard procedures. Histopathological parameters, including tissue damage/necrosis, inflammatory cell infiltration, vacuoliza tion, and hemorrhage, were assessed and semi-quantitatively scored on a scale of 0 to 3, with the following criteria: 0 denotes no abnormality; 1 denotes mild change (necrosis <10% of the tissue area, inflammatory cells < 10% of the field of view, vacuolization <25% of the tissue area, or petechial hemorrhage); 2 denotes moderate change (necrosis or infiltration 10-30%, vacuolization 25-50%, or patchy hemorrhage); and 3 denotes severe change (necrosis or infiltration > 30%, vacuolization > 50%, or massive hemorrhage).
For immunohistochemistry (IHC), sections were repaired antigenically by high pressure treatment in repair solution (pH 9.0 EDTA). Sections were then incubated with a primary mouse anti-EBIV NP antibody (1:2,000) overnight at 4°C, followed by incubation with a goat anti-mouse IgG-HRP secondary antibody (Abcam, ab205719, 1:2000) at 37°C for 45 minutes. Immunoreactivity was visualized using DAB chromogenic reagent (DAB-4033, diluted 1:20), and slides were counterstained with hematoxylin. Digital images of stained sections were captured using a NanoZoomer S360 scanner.
## Enzyme-linked immunosorbent assay (ELISA)
Concentrations of cytokines in mouse serum were measured using mouse IFN-β (Biolegend, 439407), IP-10 (4A Biotech, CME0016), TNF-α (Biolegend, 430904), and IL-6 (Biolegend, 431304) ELISA Kit according to the manufacturer's protocol.
## Statistics and reproducibility
For in vivo mouse experiments, data represent biological replicates (n ≥ 4), where n denotes samples from distinct individual mice, and statistical analyses were conducted on these biological replicates. For in vitro cell culture experiments, data shown are from a representative experiment performed with multiple independent cell samples (n ≥ 3 wells or dishes prepared and treated independently in parallel), which served as replicates for statistical analysis within that specific experiment. Differences between experimental and control groups were determined by unpaired two-tailed Student's t-test (when two groups of data were compared) or one-way ANOVA analysis (when more than two groups of data were compared). Statistical analysis of survival curves was performed using the log-rank test (Mantel-Cox). The key findings were confirmed through independent biological repetitions (typically 2-3 times) with consistent results. Statistically analyzed data are expressed as mean ± standard deviation (S.D.). A P value < 0.05 was considered statistically significant. Statistical analysis was performed using Prism software (GraphPad version 9.5.1).
## References
1. Knight, Glover, Mcquaid et al. (2021) "Antimicrobial resistance and COVID-19: intersections and implications" *Elife*
2. Liu, Zhang, Yang et al. (2014) "Genome sequence of Abbey Lake virus, a novel orthobunyavirus isolated from China" *Genome Announc*
3. Horne, Vanlandingham (2014) "Bunyavirus-vector interactions" *Viruses*
4. Xia, Liu, Zhao et al. (2019) "Characterization of Ebinur Lake Virus and its human seroprevalence at the China-Kazakhstan Border" *Front Microbiol*
5. Zhao, Luo, Huang et al. (2020) "Pathogenesis and immune response of Ebinur Lake Virus: a newly identified orthobunyavi rus that exhibited strong virulence in mice" *Front Microbiol*
6. Ma, Jiang, Zhou (2024) "DAMPs and DAMP-sensing receptors in inflammation and diseases" *Immunity*
7. Kawai, Ikegawa, Ori et al. (2024) "Decoding Toll-like receptors: recent insights and perspectives in innate immunity" *Immunity*
8. Wang, Zhu, Yang et al. (2015) "Nlrp6 regulates intestinal antiviral innate immunity" *Science*
9. Zhu, Ding, Wang et al. (2017) "Nlrp9b inflammasome restricts rotavirus infection in intestinal epithelial cells" *Nature*
10. García, Gil, Ventoso et al. (2006) "Impact of protein kinase PKR in cell biology: from antiviral to antiproliferative action" *Microbiol Mol Biol Rev*
11. Rehwinkel, Gack (2020) "RIG-I-like receptors: their regulation and roles in RNA sensing" *Nat Rev Immunol*
12. Rothenfusser, Goutagny, Diperna et al. (2005) "The RNA helicase Lgp2 inhibits TLR-independent sensing of viral replication by retinoic acid-inducible gene-I" *J Immunol*
13. Venkataraman, Valdes, Elsby et al. (2007) "Loss of DExD/H box RNA helicase LGP2 manifests disparate antiviral responses" *J Immunol*
14. Schlee, Roth, Hornung et al. (2009) "Recognition of 5' triphosphate by RIG-I helicase requires short blunt double-stranded RNA as contained in panhandle of negative-strand virus" *Immunity*
15. Hornung, Ellegast, Kim et al. (2006) "5'-Triphosphate RNA is the ligand for RIG-I" *Science*
16. Ablasser, Bauernfeind, Hartmann et al. (2009) "RIG-I-dependent sensing of poly(dA:dT) through the induction of an RNA polymerase III-transcribed RNA intermediate" *Nat Immunol*
17. Chiu, Macmillan, Chen (2009) "RNA polymerase III detects cytosolic DNA and induces type I interferons through the RIG-I pathway" *Cell*
18. Kato, Takeuchi, Mikamo-Satoh et al. (2008) "Length-dependent recognition of double-stranded ribonucleic acids by retinoic acidinducible gene-I and melanoma differentiation-associated gene 5" *J Exp Med*
19. Pichlmair, Schulz, Tan et al. (2009) "Activation of MDA5 requires higher-order RNA structures generated during virus infection" *J Virol*
20. Hiroki, Takeuchi, Sato et al. (2006) "Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses" *Nature*
21. Loo, Fornek, Crochet et al. (2008) "Distinct RIG-I and MDA5 signaling by RNA viruses in innate immunity" *J Virol*
22. Errett, Suthar, Mcmillan et al. (2013) "The essential, nonredundant roles of RIG-I and MDA5 in detecting and controlling West Nile virus infection" *J Virol*
23. Dixit, Boulant, Zhang et al. (2010) "Peroxisomes are signaling platforms for antiviral innate immunity" *Cell*
24. Seth, Sun, Ea et al. (2005) "Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-κB and IRF3" *Cell*
25. Xu, Wang, Han et al. (2005) "VISA is an adapter protein required for virus-triggered IFN-β signaling" *Mol Cell*
26. Kawai, Takahashi, Sato et al. (2005) "IPS-1, an adaptor triggering RIG-I-and Mda5-mediated type I interferon induction" *Nat Immunol*
27. Meylan, Curran, Hofmann et al. (2005) "Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus" *Nature*
28. De Veer, Holko, Frevel et al. (2001) "Functional classification of interferon-stimulated genes identified using microarrays" *J Leukoc Biol*
29. Zhang, Zhong (2022) "Regulation and function of the cGAS-MITA/ STING axis in health and disease" *Cell Insight*
30. (1016)
31. (2025) *Full-Length Text Journal of Virology*
32. Tong, Zou, Wang et al. (2024) "Activation of the cGAS-STING-IRF3 axis by type I and II interferons contributes to host defense" *Adv Sci (Weinh)*
33. Hu, Shu (2023) "Mitochondrial DNA-triggered innate immune response: mechanisms and diseases" *Cell Mol Immunol*
34. Ablasser, Hur (2020) "Regulation of cGAS-and RLR-mediated immunity to nucleic acids" *Nat Immunol*
35. Pichlmair, Schulz, Tan et al. (2006) "RIG-I-mediated antiviral responses to single-stranded RNA bearing 5'-phosphates" *Science*
36. Goubau, Schlee, Deddouche et al. (2014) "Antiviral immunity via RIG-I-mediated recognition of RNA bearing 5′diphosphates" *Nature*
37. Weber, Wagner, Rasmussen et al. (2006) "Double-stranded RNA is produced by positive-strand RNA viruses and DNA viruses but not in detectable amounts by negative-strand RNA viruses" *J Virol*
38. Baum, García-Sastre (2011) "Differential recognition of viral RNA by RIG-I" *Virulence*
39. Feng, Deng, Hu et al. (2019) "Heartland virus antagoni zes type I and III interferon antiviral signaling by inhibiting phosphoryla tion and nuclear translocation of STAT2 and STAT1" *J Biol Chem*
40. Li, Duan, Yu et al. (2024) "Bunyavirus SFTSV NSs utilizes autophagy to escape the antiviral innate immune response" *Autophagy*
41. Moriyama, Igarashi, Koshiba et al. (2018) "Two conserved amino acids within the NSs of severe fever with thrombocytopenia syndrome phlebovirus are essential for antiinterferon activity" *J Virol*
42. Ning, Feng, Min et al. (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*
43. Luo, Tong, Cao et al. (2022) "Transcription-independent regulation of STING activation and innate immune responses by IRF8 in monocytes" *Nat Commun*
44. Luo, Li, Li et al. (2016) "iRhom2 is essential for innate immunity to DNA viruses by mediating trafficking and stability of the adaptor STING" *Nat Immunol*
45. Xu, Du, Wang et al. (2024) "LDLR is an entry receptor for Crimean-Congo hemorrhagic fever virus" *Cell Res*
46. Zheng, Fu, Wang et al. (2022) "Herpes simplex virus protein UL56 inhibits cGAS-Mediated DNA sensing to evade antiviral immunity" *Cell Insight* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772286&blobtype=pdf | # HERV6196 as an enhancer with oncogenic potential in rectal cancer
Yi-Xiu Gan, Xin Jiang, Zhi-Yu Wang, Yi-Lin Yu, Ling-Dong Shao, Jian-Min Wang, Jun-Xin Wu
## Abstract
Colorectal cancer (CRC) is among the most prevalent malignancies. However, the regulatory networks involved in tumor occurrence and development are still poorly understood. Human endogenous retroviruses (HERVs), a class of transposa ble elements, have been implicated in the development and progression of various human cancers. This study presents the first comprehensive locus-specific profiling of the expression of HERV gene transcripts in rectal cancer, revealing significantly dysregulated HERVs. Analysis of data from three Gene Expression Omnibus data sets revealed 25 upregulated HERVs and 7 downregulated HERVs. Dysregulation of HERV6196, a type of HERVH, was validated through reverse transcription quantitative PCR and droplet digital PCR in cells and tissues. Additionally, HERV6196 promoted the proliferation, inhibited the apoptosis, enhanced the colony formation ability, and enhanced the migration capability of rectal cancer cells. Moreover, HERV6196 functioned as an enhancer, promoting the expression of neighboring genes and the development of CRC. In summary, the present results revealed that HERV6196 is involved in the pathogenesis of rectal cancer, indicating the potential contribution of dysregulated HERVs to the development and progression of CRC through gene expression modulation.
IMPORTANCEThe role of human endogenous retroviruses (HERVs) in colorectal cancer (CRC) remains insufficiently understood. The present study revealed aberrant expres sion of HERV gene transcripts in cancerous tissues compared with non-cancerous tissues. HERV6196 contributes to CRC progression by regulating the expression of neighboring genes. These findings suggest that HERVs may serve as enhancers and regulate oncogenic gene expression, providing new insights for rewiring transcriptional regulatory networks in CRC pathogenesis.
KEYWORDS colorectal cancer, endogenous retrovirus, dysregulated expressionC olorectal cancer (CRC) is a leading malignant neoplasm worldwide. According to the 2020 data of the World Health Organization, CRC is the third most common cancer in terms of incidence and the second most common in terms of mortality (1). The majority of CRC cases arise from the gradual progression of precancerous lesions (2). Early diagnosis and treatment of CRC are best achieved through the use of tumor markers, imaging techniques, and genetic testing, which have the potential to reduce disease morbidity and mortality (3-5). However, current tests cannot rapidly and accurately detect CRC (6, 7). Thus, there is an urgent need to identify novel markers that can predict the occurrence of CRC.Human endogenous retroviruses (HERVs) originated 30 million years ago when exogenous retroviruses infected human ancestors. Through continuous evolution, HERVs have integrated into the human genome and become a part of it. HERVs make up approximately 8%-9% of the human genome (8-10). A typical HERV structure consists of LTR-gag-pol-env-LTR, where the long terminal repeats (LTRs) serve as highly active
regulatory elements for transcription (11). HERVs are extensively distributed through out the human body. Through genome-wide association studies, researchers have systematically identified 13,889 HERVs expressed in various normal tissues, revealing specific expression patterns associated with body part, sex, race, and age (12). Further more, studies have demonstrated that HERVs are reactivated during aging, leading to cellular aging and inflammation through the activation of the cGAS/STING innate immune pathway (13). In cancer research, HERVs may function as potential enhanc ers of neighboring oncogene expression, thereby contributing to the development of leukemia (14). In CRC, the activation of the LTR10 enhancer of the HERV family has been shown to contribute to transcriptional dysregulation in response to oncogenic signaling (15). However, the carcinogenic characteristics of individual HERV loci have not been comprehensively studied in CRC. The current knowledge on the expression of HERV gene transcripts and its correlation with rectal cancer progression is limited. Investigating HERV transcriptional patterns in rectal cancer could reveal novel virus-related cancer biomarkers and guide the development of more effective treatment strategies.
The repetitive nature of HERVs presents challenges in accurately measuring their expression across the genome via sequencing, hindering comprehensive research and progress (16). ERVmap is a pipeline designed for the quantification of proviral endogenous retroviruses in RNA-seq data, which encompasses all known proviral HERV sequences and employs algorithms with stringent filtering criteria to increase the reliability of HERV site identification. The application of ERVmap across a range of diseases and experimental conditions holds promise for identifying novel disease markers (17).
In this study, we employed ERVmap to analyze raw RNA-seq data from rectal cancer patients extracted from the NCBI Sequence Read Archive (SRA) database, which revealed abnormal expression of HERV gene transcripts in rectal cancer tissues compared with adjacent tissues, and these findings were validated through reverse transcription quantitative PCR (RT-qPCR) and droplet digital PCR (ddPCR). In addition, the present results indicated that HERV6196 promotes the proliferation and migration of rectal cancer cells while inhibiting their apoptosis. Finally, the present study suggested that HERV6196 contributes to the initiation and progression of rectal cancer by influencing neighboring oncogenes.
## RESULTS
## Differential expression of HERVs in rectal cancer tissues and adjacent non-cancerous tissues
In this study, we obtained raw RNA sequencing data from three Gene Expression Omnibus data sets, namely, GSE50760, GSE104836, and GSE142279. We subsequently quantitatively analyzed the RNA-seq data from rectal cancer and adjacent normal tissues via the ERVmap tool, referencing a gene locus-specific expression profile derived from a database of 3,220 HERVs. We analyzed the differentially expressed HERVs via the DESeq2 package (Table S1). In the GSE50760 data set, 107 HERVs exhibited differential expression, with 79 upregulated HERVs and 28 downregulated HERVs (Fig. 1A). Similarly, the GSE104836 data set revealed 86 upregulated HERVs and 47 downregulated HERVs. In the GSE142279 data set, 239 HERVs were differentially expressed, with 123 upregulated HERVs and 116 downregulated HERVs. To further investigate HERVs related to rectal cancer, we conducted an analysis of the intersection of the GSE50760, GSE104836, and GSE142279 data sets, which revealed 25 significantly upregulated HERVs (HEV6196, HERV4098, HERV6114, HERV1066, HERV4849, HERV1412, HERV557, HERV915, HERV3452, HERV1074, HERV526, HERV5305, HERV2916, HERV4270, ERVH13q33.3, HERV1475, HERV2003, HERV3730, HERV858, HERV3732, HERV4627, HERV4825, HERV5290, HERV2415, and HERV4719) (Fig. 1B) and seven significantly downregulated HERVs (HERV4152, HERV6007, HERV559, HERV1520, HERV2674, HERV6119, and HERV4457; Fig. 1C). Heatmaps were generated to visualize the differential expression of HERVs (Fig. 1D).
## Verification of the differential expression of HERVs in rectal cell lines and tissues
To further validate the differential expression of HERVs at the cellular level, we ran domly selected HERVs with a large |log2FoldChange| for RT-qPCR verification. HERV6196 expression was greater in three distinct rectal cancer cell lines (HT29, HCT116, and SW480) than in HCoEpiC (Fig. 2A). In contrast, compared with HCoEpiC, HERV6119 was expressed at lower levels in three distinct rectal cell lines (Fig. 2B). These findings were consistent with the results of the bioinformatics analysis conducted via ERVmap. Additionally, we collected 18 cancerous and adjacent tissues from patients with CRC and conducted ddPCR experiments. The results expression of HERV6196 was greater in rectal cancer tissues than in adjacent non-cancerous tissues (P = 0.0019; Fig. 2C).
## Differentially expressed genes associated with HERV6196 are significantly enriched in multiple pathways related to CRC
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the differentially expressed genes that were correlated with HERV6196. GO pathway enrichment analysis revealed that HERV6196 was associated primarily with the modulation of chemical synaptic transmission, the regulation of membrane potential, cytokine activity, and extracellular matrix structural constituents (Fig. 2D). Moreover, KEGG analysis indicated that HERV6196 was enriched primarily in the cAMP signaling pathway, the cell cycle, the IL-17 signaling pathway, and the extracellular matrix (ECM)-receptor interaction (Fig. 2E). These pathway enrichment results suggested that HERV6196 may be closely associated with the development of cancer.
## Silencing HERV6196 inhibits rectal cancer cell proliferation but promotes rectal cancer cell apoptosis
Cell functional assays were conducted to further investigate the role of HERV6196 in the occurrence and progression of rectal cancer. To evaluate the effect of HERV6196 on cell proliferation, we employed short hairpin RNA (shRNA) to inhibit HERV6196 expression (Fig. 3A). In the Cell Counting Kit-8 (CCK-8) assay, cell viability was assessed at 24 h, 48 h, 72 h, 96 h, 120 h, and 144 h. Compared with the control group, the transfected group had significantly lower proliferation at 72 h, 96 h, 120 h, and 144 h (Fig. 3B). Additionally, cell cycle and apoptosis analyses were conducted. Flow cytometry was used to assess the impact of HERV6196 knockdown on the cell cycle within 24 h, which revealed that HERV6196 knockdown significantly altered HCT116 cell cycle progression. Compared with that of sh-NC cells, the proportion of sh-HERV6196 cells in the G1 phase was increased (sh-HERV6196 vs sh-NC: 60.0% vs. 38.0%), whereas the proportion of sh-HERV6196 cells in the S phase was decreased (sh-HERV6196 vs sh-NC: 16.4% vs 31.6%). For the G2/M phase comparison, the proportion of sh-HERV6196 cells was decreased compared with sh-NC cells (sh-HERV6196 vs sh-NC: 17.0% vs 24.7%, respec tively), and the proportion of sh-HERV6196 cells in the G2 + S phase was decreased compared with that of sh-NC cells (sh-HERV6196 vs sh-NC: 33.4% vs 56.3%, respec tively). Compared with the control group, the knockdown group exhibited reduced S phase progression and diminished DNA synthesis (Fig. 3C). Flow cytometry was also used to detect apoptosis 48 h after transfection. In Fig. 3D, quadrant Q1 represented necrotic cells, quadrant Q2 represented late apoptotic cells and dead cells, quadrant Q3 represented early apoptotic cells, and quadrant Q4 represented viable cells. Compared with the control group, the knockdown group presented significantly higher apoptosis rates (Q2 + Q3; sh-HERV6196 vs sh-NC: 27.75% vs 14.08%; (Fig. 3D). The proportion of early and late apoptotic sh-HERV6196 cells was increased compared with sh-NC cells (P = 0.0039). These results indicated that HERV6196 promoted rectal cancer cell proliferation but inhibited rectal cancer cell apoptosis.
## Silencing HERV6196 inhibits the colony formation and migration of rectal cancer cells
A colony formation assay was used to compare the colony formation efficiency of the transfected and control groups. Compared with the control group, the sh-HERV6196transfected group presented significantly lower colony formation ability (Fig. 4A). In the control group, cell migration occurred at a faster rate, and scratch healing was more pronounced. Conversely, the knockdown group exhibited considerably slower cell migration and a delayed scratch healing process. These findings suggested that the knockdown of HERV6196 inhibits the migration ability of HCT116 cells (P < 0.05; Fig. 4B). Transwell invasion assays were used to assess the invasive migration ability of HCT116 cells. Compared with the control cells, the HERV6196 knockdown cells exhibited decreased invasion ability (Fig. 4B). These results indicated that HERV6196 promotes the proliferation, migration, and invasion of rectal cancer cells.
## HERV6196 may act as a potential enhancer to influence the occurrence of cancer by affecting neighboring genes
Analysis of the GeneHancer database revealed that HERV6196 has two enhancer fragments, namely, GH01J221965 and GH01J221969. CRC chromatin immunoprecipita tion followed by sequencing (ChIP-seq) and assay for transposase-accessible chromatin using sequencing (ATAC-seq) data were downloaded and visualized using the Integrative Genomics Viewer (IGV). Compared with adjacent normal tissues, the ChIP-seq results revealed that CRC tissues presented a greater signal intensity for HERV6196. In addition, ATAC-seq also revealed a high signal strength of HERV6196 in CRC tissues. Thus, these findings suggested that HERV6196 may function as a potential enhancer (Fig. 5A). The ChIP-seq results for the adjacent non-cancerous tissues are displayed in Fig. S1.
In order to verify the enhancer function of HERV6196, a dual luciferase reporter assay was performed. The dual-luciferase reporter assay demonstrated that the relative luciferase activity of the pGL4-HERV6196 group, which carries the HERV6196 sequence, was significantly higher than that of the null-loaded pGL4-control group (P < 0.01; Fig. 5B). This result confirms the enhancer activity of the HERV6196 sequence and suggests its potential involvement in rectal cancer progression through the regulation of neighboring genes.
To further investigate the mechanism by which HERV6196 promotes rectal cancer occurrence, we analyzed the neighboring genes of HERV6196. Correlation analysis revealed that HERV6196 was positively correlated with various neighboring genes, including ADAMTS, BLACT1, DTL, KIF14, LEMD1, NEK2, LINC02474, and KIF26B (Fig. S2). We subsequently confirmed the expression levels of neighboring genes following HERV6196 knockout in HCT116 cells. The knockdown sequence of HERV6196 (sh-HERV6196) was identified and validated via RT-qPCR. RT-qPCR analysis revealed a significant decrease in the expression of neighboring genes in the transfected group compared with the control group (Fig. 5C). Furthermore, western blot analysis revealed that the expression of neighboring genes was reduced following HERV6196 knockdown (Fig. 6A through E). These results suggested that HERV may function as an enhancer, promoting the development of rectal cancer by influencing neighboring oncogenes.
## DISCUSSION
In the past, HERVs were regarded as "junk DNA" owing to limitations in sequencing technology and bioinformatics software (18). However, several studies have demon strated that HERVs are involved in gallbladder cancer, chronic lymphocytic leukemia, hepatocellular carcinoma, esophageal squamous cell carcinoma, and glioblastoma, highlighting significant differences and imbalances associated with cancer (19)(20)(21)(22)(23). With the advancement of multiomics big data, the functional roles of HERVs in vari ous diseases, particularly cancer, have garnered increasing attention as a new area of research. To date, no comprehensive study has explored the locus-specific characteristics and mechanisms of HERVs in CRC. In the present study, we conducted a comprehensive analysis of HERVs in rectal cancer via ERVmap, a locus-specific identification pipeline that includes 3,220 HERVs, which revealed significant dysregulation of HERVs in rectal cancer, with 25 upregulated HERVs and 7 downregulated HERVs. Moreover, these 32 differential HERVs were identified across all three data sets, confirming the reliability of the differential analysis results. We further validated the results via RT-qPCR and confirmed their accuracy. Additionally, we collected clinical samples for precise quantification via ddPCR. Notably, HERV6196 was highly expressed in cancerous tissues but had low expression levels in adjacent non-cancerous tissues, a result that is consistent with previous studies. According to the ERVmap database, HERV6196 belongs to the HERVH type. Alves et al. reported that the HERVH gene on the X chromosome is selectively transcribed in 60% of colon cancer patients, with metastatic colon cancers presenting a greater percentage (24). Wentzensen et al. reported a greater proportion of metastatic colon cancers, and they observed the expression of HERV-H RNA sequences in various gastrointestinal tract tumors (25). These findings indicate that aberrant expression of HERVH elements is associated with cancer, a phenomenon that is particularly pro nounced in cases of CRC. Thus, HERV6196 is a potential novel clinical diagnostic bio marker.
Pathway enrichment analysis of HERV6196-related differentially expressed genes revealed significant enrichment in CRC-related pathways. For example, cytokine activity and extracellular matrix structural constituent pathways affect the occurrence of CRC by regulating the tumor immune microenvironment (26,27). The cAMP/PKA signal ing pathway regulates mitochondrial autophagy, thereby promoting the onset and progression of CRC (28). In the IL-17 signaling pathway, increased expression of interleukin-17A is associated with poor prognosis of CRC patients, whereas blocking IL-17A inhibits CRC progression in preclinical cancer models (29). In addition, the ECM-receptor interaction promotes the metastatic potential of CRC (30). These results suggest a potential role for HERV in CRC. The present results demonstrated the role of HERV6196 in promoting rectal cancer cell proliferation and migration while reducing rectal cancer cell apoptosis. Thus, HERV6196 may serve as a potential biomarker and novel therapeutic target for CRC. We also explored the mechanisms through which HERV6196 enhances the prolifera tion of rectal cancer. Research has demonstrated that HERVs have oncogenic effects through various mechanisms, including direct involvement in the maintenance of the tumor phenotype, inactivation of oncogenes, activation of oncogenes, mediation of cell fusion, and activation of tumor signaling pathways (31)(32)(33)(34). HERVs have the capacity to function as promoters or enhancers, thereby triggering the activation of host genes and promoting the expression of cancer-associated genes (11). Jönsson et al. demonstrated that the LTR of HERVs acts as an alternative promoter to drive the expression of host genes (35). The activated LTR of HERV functions as an alternative promoter for neighbor ing genes, exerting transcriptional regulation on DNA methylation processes. Similarly, the inactivation of HERVs has been shown to directly alter gene expression in AML cell lines, suggesting that HERVs, acting as enhancers, are utilized by cancer cells to drive tumor heterogeneity and evolution (14). However, the role of HERVs in rectal cancer remains underreported in the scientific literature. In the present study, we conducted ChIP-seq and ATAC-seq analyses on HERV6196, and we utilized the GeneHancer (36) database, a comprehensive and integrated enhancer database. The present findings indicated that two segments in GeneHancer overlapped with HERV6196, suggesting that HERV6196 has potential enhancer functionality. Moreover, ChIP-seq and ATAC-seq analyses revealed significantly increased HERV6196 signals, further confirming the enhancer function of HERV6196.
Further exploration revealed that HERV6196 was positively correlated with several neighboring genes, such as NEK2, LINC02474, LEMD1, and ADAMTS4. Suzuki et al. (37) demonstrated that NEK2 regulates cell division and mitosis through centrosome division. The combination of NEK2 small interfering RNA and cisplatin has been shown to inhibit the growth of rectal cancer cells. Furthermore, elevated NEK2 has been demonstrated to play a pivotal role in tumorigenesis and tumor progression by regulating chromosomal instability and aneuploidy, signaling pathways, mRNA selective splicing, p53, ciliolysis, and tumor immune escape (38). The LINC02474 oncogene prevents apoptosis and promotes metastasis in CRC by inhibiting GZMB expression, a process that is associated with the poor prognosis of rectal cancer patients (39)(40)(41)(42). Moreover, LEMD1 promotes the migration of CRC cells through the RhoA/ROCK signaling pathway (43). In addition, high KIF26B expression is significantly associated with shorter survival of CRC patients, and the cancer-promoting roles of KIF14 (44) in rectal cancer have been confirmed. In addition to its role in the growth of early-stage lung cancer, ADAMTS4 promotes tumor progression in hepatocellular carcinoma, lung cancer, and CRC (45). Furthermore, we confirmed that HERV6196 acts as an enhancer, promoting the expression of oncogenes and contributing to the development of rectal cancer. These findings offer new insights into the mechanisms underlying the development of CRC.
In summary, the present study demonstrated that HERV6196 is significantly upregulated in various rectal cancer cell lines. Functional analyses further demonstrated that the knockdown of HERV6196 expression significantly inhibits tumor proliferation, invasion, and migration. These findings underscore the association between HERV6196 expression and the progression of rectal cancer, suggesting that HERV6196 may serve as a novel biomarker for this disease. Additionally, HERV6196 may function as an enhancer, playing a role in the regulatory mechanisms of CRC. We will further explore the enhancer mechanisms in future studies using methods such as reporter assays and chromosome conformation capture. Future research is needed to explore the upstream mechanisms influencing HERV occurrence. Additionally, HERV6196 should be validated as a novel diagnostic biomarker for CRC using a larger clinical sample size.
## Conclusions
HERVs are implicated in the development of various cancers. However, comprehensive single-site HERV studies in CRC are lacking. In the present study, we validated the differential expression of HERVs in rectal cancer via ERVmap, RT-qPCR, and ddPCR. Subsequent functional analysis revealed that the knockdown of HERV6196 expression significantly inhibits tumor proliferation, invasion, and migration. These studies suggest that HERV6196 could serve as a novel biomarker for rectal cancer. Mechanistically, HERV may function as an enhancer that promotes neighboring genes, such as NEK2, LINC02474, LEMD1, and ADAMTS4. In conclusion, the present research highlights the potentially crucial role of human endogenous retroviruses in the biology of rectal cancer.
## MATERIALS AND METHODS
## Data collection
The raw RNA sequencing data for rectal cancer patients were extracted from the NCBI SRA database (https://www.ncbi.nlm.nih.gov/sra). The following data sets were utilized to identify differentially expressed HERVs: GSE50760, which comprises 18 cancer tissues and 18 adjacent normal tissues; GSE104836, which includes 10 cancer tissues and 10 adjacent normal tissues; and GSE142279, which contains 20 cancer tissues and 20 adjacent normal tissues.
## Sample collection
Plasma and tissues were obtained from patients with rectal cancer at Fujian Cancer Hospital between January 2021 and December 2024. All samples were obtained prior to treatment, and a total of 18 cancerous and adjacent normal tissues were collected for ddPCR. Tissue gDNA was extracted from formalin-fixed paraffin-embedded (FFPE) or fresh tumor samples via a QIAamp FFPE tissue kit (Qiagen, Hilden, Germany).
## Calculation of HERV gene transcript expression using ERVmap
The expression of HERV gene transcripts was analyzed via ERVmap software. Specifically, the RNA sequencing reads were first mapped to the human genome (hg38) via the Burrows-Wheeler Aligner. The aligned reads were subsequently filtered according to the stringent criteria of ERVmap via a script designed specifically for ERVmap, and the reads were subsequently mapped to ERV loci. Finally, the expression of HERV gene transcripts was normalized according to the ERVmap protocol.
## Differential expression of HERV gene transcripts
Differential expression analysis of HERVs in rectal cancer tissues vs adjacent normal tissues was conducted via the DESeq2 R package. The data sets analyzed included the GSE50760, GSE104836, and GSE142279 data sets. The screening criteria were set at |log2FoldChange| > 1 and padj < 0.05. To visualize expression patterns, TBtools software was utilized (https://github.com/CJ-Chen/TBtools) to generate a heatmap of HERV gene transcript expression levels.
## Correlation analysis of HERVs with neighboring genes
To investigate the relationship between HERVs and neighboring genes, the expression of genes and HERVs was initially computed via ERVmap software. Finally, Spearman correlation analysis was performed to assess the relationship between HERVs and adjacent genes, applying thresholds of a correlation coefficient |R| > 0.3 and P < 0.05.
## GO and KEGG pathway enrichment analyses
The clusterProfiler package in R was utilized to conduct GO and KEGG enrichment analyses. Additionally, the results were visualized via histograms.
## Cell lines and culture
Human rectal adenocarcinoma cell lines (SW480, HCT116, HT29, LOVO, and SW620) and normal human colorectal epithelial cells (HCoEpiC) were purchased from Wuhan Pricella Biotechnology Co., Ltd., China. All the cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS; Meisen Chinese Tissue Culture Collections) at 37°C in a 5% CO 2 atmosphere.
## RNA extraction and reverse transcription
RNA extraction was performed via the TRIzol Up Plus RNA Kit (Beyotime) according to the manufacturer's instructions. The purity and concentration of the extracted RNA were determined via UV spectrophotometry, with an OD260/OD280 ratio (R value) within the range of 1.8-2.2, indicating high purity. The RNA was subsequently reverse transcribed to cDNA via the HiScript II Q RT SuperMix for qPCR (+gDNA wiper) kit (Novozymes).
## RT-qPCR and primer design
RT-qPCR was conducted using the Hieff qPCR SYBR Green Master Mix Kit (Yeasen) and an ABI 7500 Fast Real-Time PCR system (Applied Biosystems). The cycling conditions were as follows: initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 35 s. The relative expression of RNA was calculated via the 2 -ΔΔCt method. Each experiment was independently repeated three times. The cDNA products were stored at -80°C. Primers for HERVs and genes were designed via the NCBI Primer-BLAST tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Table S2 lists the primers used in this study.
## Droplet digital PCR
DdPCR was performed via the Probe One-Step RT-ddPCR Advanced Kit (Bio-Rad, Hercules, CA, USA). The reaction mixture had a final volume of 22 µL, with 5.5 µL of super mixture, 2.2 µL of 20 U/µL reverse transcriptase, 1.1 µL of 15 mM dithiothreitol, 900 nM Oropouche virus (OROV) primer, 250 nM probe, 2 µL of 20× RPP30 analysis, and 7 µL of RNA template. A QX200 droplet generator (Bio-Rad, Hercules, CA, USA) was used to convert 20 µL of each reaction mixture into droplets. The droplet aliquots were transferred to a 96-well plate, sealed, and processed in a C1000 touch thermal cycler (Bio-Rad). The following cycling protocol was used: holding at 25°C for 3 min; reverse transcription at 50°C for 60 min; enzyme activation at 95°C for 10 min; 50 cycles of denaturation at 95°C for 30 s and annealing/extension at 60°C for 60 s; enzyme inactivation at 98°C for 10 min; and holding at 4°C for 30 min. The amplified samples were then transferred and read in the FAM (OROV) and HEX (RPP30) channels via a QX200 reader (Bio-Rad, Hercules, CA, USA). The data were analyzed via QXManager 1.2 standard edition software (Bio-Rad, Hercules, CA, USA) and expressed as copy number/µL (cp/µL) in the ddPCR.
## CCK-8 assay
The cells were seeded into 96-well plates and incubated for 24 h, 48 h, 72 h, 96 h, 120 h, or 144 h. After treatment, 10 µL of CCK-8 reagent (Beyotime Biotechnology) was added to each well and incubated at 37°C for 1-4 h. The absorbance was measured at 450 nm via a microplate reader to assess cell viability.
## Colony formation assay
The cells were plated in six-well plates at a low density (500 cells/well) and cultured for 10-14 days. Colonies were fixed with 4% paraformaldehyde, stained with crystal violet, and counted to evaluate their proliferative capacity.
## Invasion assay
Transwell chambers coated with Matrigel were used for the invasion assay. The cells suspended in serum-free medium were added to the upper chamber, and medium containing 10% FBS was added to the lower chamber. After incubation for 48 h, the invading cells were fixed, stained, and counted under a microscope.
## Scratch assay
A monolayer of cells was scratched with a pipette tip to create a wound. The cells were cultured in medium with reduced serum, and images were captured at 0 h and 48 h. The migration distance was measured to assess cell motility.
## Apoptosis assay
The cells were stained with Annexin V-FITC and propidium iodide (Cell Cycle and Apoptosis Detection Kit, Beyotime) according to the manufacturer's protocol. Flow cytometry was used to quantify apoptotic cells by measuring fluorescence signals.
## Identification of enhancers
ChIP-seq data for three CRC and adjacent tissue pairs were downloaded from the GSE166254 database. We downloaded the hg19 normalized bw files from GSE166254 and used CrossMap to convert the hg19 normalized bw files to hg38 normalized bw files, which were then exported to IGV for display. Additionally, ATAC-seq data for three CRC cases were obtained from The Cancer Genome Atlas (TCGA). Both the ChIP-seq and ATAC-seq data were visualized via the IGV. Potential enhancer elements were also downloaded from the GeneHancer database.
## Western blot analysis
Protein lysates were extracted from cells via radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with protease and phosphatase inhibitors. Protein concen trations were quantified via the bicinchoninic acid assay. Equal amounts of protein (20-40 µg) were resolved on SDS-PAGE gels and transferred onto polyvinylidene difluoride membranes. The membranes were blocked with 5% nonfat milk or BSA in TBST buffer and incubated overnight at 4°C with primary antibodies. After wash ing, the membranes were incubated with horseradish peroxidase-conjugated secon dary antibodies, and signals were detected via enhanced chemiluminescence. Band intensities were quantified by ImageJ software for further analysis.
## Dual-luciferase reporter assay
The detection of HERV6196 enhancer activity was accomplished through the utilization of a dual luciferase reporter gene assay system. Initially, the HERV6196 sequence was engineered into the pGL4 luciferase reporter vector, thereby constructing the recombi nant reporter plasmid pGL4-HERV6196. Subsequently, the aforementioned vectors and the empty vectors were transferred into 24-well plates of 293T cells using Lipofectamine 3000 (Thermo Fisher; four wells for each vector). Four replicate wells were utilized for each vector. Concurrently, the pRL-TK Luciferase Vector was utilized as an inter nal reference and co-transfected with the aforementioned luciferase reporter vectors. Following a 24 h transfection period, the activities of Firefly Luciferase and Renilla Luciferase were measured sequentially in accordance with the Promega instructions. This experiment was repeated on three occasions, and the ratio of Firefly Luciferase activity to Renilla Luciferase activity was employed for normalization.
## Statistical analysis
Statistical analysis was conducted via GraphPad Prism (version 8.0.2) and R (version 4.2.1). The data were analyzed via Student's t-test for comparisons between two groups. Each experiment was independently repeated three times. The Wilcoxon signed-rank test was employed to compare differential expression between cancerous and normal samples. Correlation analysis was performed via Spearman's rank correlation coefficient, and a P-value less than 0.05 was considered statistically significant.
## References
1. Sung, Ferlay, Siegel et al. (2021) "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries" *CA Cancer J Clin*
2. Nguyen, Goel, Chung (2020) "Pathways of colorectal carcinogen esis" *Gastroenterology*
3. Siegel, Wagle, Cercek et al. (2023) "Colorectal cancer statistics, 2023" *CA Cancer J Clin*
4. Lin, Perdue, Henrikson et al. (2021) "Screening for colorectal cancer: updated evidence report and systematic review for the US preventive services task force" *JAMA*
5. (2026) *Research Article Microbiology Spectrum*
6. Kastrinos, Samadder, Burt (2020) "Use of family history and genetic testing to determine risk of colorectal cancer" *Gastroenterology*
7. Fernandes, Gollub, Brown (2022) "The importance of MRI for rectal cancer evaluation" *Surg Oncol*
8. Mo, Dai, Wang et al. (2023) "Early detection and prognosis prediction for colorectal cancer by circulating tumour DNA methylation haplotypes: a multicentre cohort study" *EClinicalMedicine*
9. Kyriakou, Magiorkinis (2023) "Interplay between endogenous and exogenous human retroviruses" *Trends Microbiol*
10. Alcazer, Bonaventura, Depil (2020) "Human endogenous retrovi ruses (HERVs): shaping the innate immune response in cancers" *Cancers (Basel)*
11. Grandi, Tramontano (2018) "Human endogenous retroviruses are ancient acquired elements still shaping innate immune responses" *Front Immunol*
12. Jakobsson, Vincendeau (2022) "SnapShot: human endogenous retroviruses" *Cell*
13. She, Du, Xu et al. (2022) "The landscape of hervRNAs transcribed from human endogenous retroviruses across human body sites" *Genome Biol*
14. Liu, Liu, Wu et al. (2023) "Resurrection of endogenous retroviruses during aging reinforces senescence" *Cell*
15. Deniz, Ahmed, Todd et al. (2020) "Endogenous retroviruses are a source of enhancers with oncogenic potential in acute myeloid leukaemia" *Nat Commun*
16. Ivancevic, Simpson, Joyner et al. (1218) "Endogenous retroviruses mediate transcrip tional rewiring in response to oncogenic signaling in colorectal cancer" *Sci Adv*
17. Treangen, Salzberg (2011) "Repetitive DNA and next-generation sequencing: computational challenges and solutions" *Nat Rev Genet*
18. Tokuyama, Kong, Song et al. (2018) "ERVmap analysis reveals genome-wide transcription of human endogenous retroviruses" *Proc Natl Acad Sci*
19. Li, Qian, Yan et al. (2024) "HervD Atlas: a curated knowledgebase of associations between human endogenous retroviruses and diseases" *Nucleic Acids Res*
20. Wang, Ren, Yu et al. (2022) "Single-cell RNA sequencing highlights the functional role of human endogenous retroviruses in gallbladder cancer" *EBioMedicine*
21. Ferlita, Nigita, Tsyba et al. (2023) "Expression signature of human endogenous retroviruses in chronic lymphocytic leukemia" *Proc Natl Acad Sci*
22. Chang, Hsu, Chung et al. (2023) "Comprehensive analysis and drug modulation of human endogenous retrovirus in hepatocellular carcinomas" *Cancers (Basel)*
23. Shi, Lu, Li et al. (2025) "The transcrip tional landscape and clinico-biological characterization of human endogenous retroviruses in esophageal squamous cell carcinoma" *Int J Cancer*
24. Li, He, Xue et al. (2024) "Impact of endogenous viral elements on glioma clinical phenotypes by inducing OCT4 in the host" *Front Cell Infect Microbiol*
25. Alves, Lévy, Stevenson et al. (2008) "Identification of tumor-associated antigens by large-scale analysis of genes expressed in human colorectal cancer" *Cancer Immun*
26. Wentzensen, Coy, Knaebel et al. (2007) "Expression of an endogenous retroviral sequence from the HERV-H group in gastrointestinal cancers" *Int J Cancer*
27. Sun, Liu, Wu et al. (2024) "Galectin-7 induction by EHMT2 inhibition enhances immunity in microsatellite stability colorectal cancer" *Gastroenterology*
28. Huang, Ou, Chen et al. (2021) "Tumor cell-derived SPON2 promotes M2-polarized tumor-associated macrophage infiltration and cancer progression by activating PYK2 in CRC" *J Exp Clin Cancer Res*
29. Tang, Peng, Ji et al. (2023) "GPR176 promotes cancer progression by interacting with G protein GNAS to restrain cell mitophagy in colorectal cancer" *Adv Sci (Weinh)*
30. Dmitrieva-Posocco, Dzutsev, Posocco et al. (2019) "Cell-type-specific responses to interleukin-1 control microbial invasion and tumor-elicited inflammation in colorectal cancer" *Immunity*
31. Machackova, Vychytilova-Faltejskova, Souckova et al. (2020) "MiR-215-5p reduces liver metastasis in an experimental model of colorectal cancer through regulation of ECM-receptor interactions and focal adhesion" *Cancers (Basel)*
32. 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*
33. Vergara Bermejo, Ragonnaud, Daradoumis et al. (2020) "Cancer associated endogenous retroviruses: ideal immune targets for adenovirus-based immunotherapy" *Int J Mol Sci*
34. Grandi, Tramontano (2018) "HERV envelope proteins: physiological role and pathogenic potential in cancer and autoimmunity" *Front Microbiol*
35. Kassiotis, Stoye (2017) "Making a virtue of necessity: the pleiotropic role of human endogenous retroviruses in cancer" *Philos Trans R Soc Lond B Biol Sci*
36. Jönsson, Brattås, Gustafsson et al. (2019) "Activation of neuronal genes via LINE-1 elements upon global DNA demethylation in human neural progenitors" *Nat Commun*
37. Fishilevich, Nudel, Rappaport et al. (2017) "Gene Hancer: genome-wide integration of enhancers and target genes in GeneCards"
38. Suzuki, Kokuryo, Senga et al. (2010) "Novel combination treatment for colorectal cancer using Nek2 siRNA and cisplatin" *Cancer Sci*
39. Xia, Zhao, Edmondson et al. (2025) "Role of NEK2 in tumorigenesis and tumor progression" *Trends Mol Med*
40. Kap, Seibold, Scherer et al. (2016) "SNPs in transporter and metabolizing genes as predictive markers for oxaliplatin treatment in colorectal cancer patients" *Int J Cancer*
42. Duan, Xia, Li et al. (2022) "Identification of autophagyrelated LncRNA to predict the prognosis of colorectal cancer" *Front Genet*
43. Wang, Zhou, Xu et al. (2018) "A 15-lncRNA signature predicts survival and functions as a ceRNA in patients with colorectal cancer" *Cancer Manag Res*
44. Du, Gao, Zhao et al. (2021) "Long non-coding RNA LINC02474 affects metastasis and apoptosis of colorectal cancer by inhibiting the expression of GZMB" *Front Oncol*
45. Zhang, Xu, Jiang et al. (2022) "A three-genes signature predicting colorectal cancer relapse reveals LEMD1 promoting CRC cells migration by RhoA/ROCK1 signaling pathway" *Front Oncol*
46. Neska-Długosz, Buchholz, Durślewicz et al. (2021) "Prognostic impact and functional annotations of KIF11 and KIF14 expression in patients with colorectal cancer" *Int J Mol Sci*
47. Asano, Edamatsu, Hatipoglu et al. (2018) "Host-produced ADAMTS4 inhibits earlystage tumor growth" *Acta Med Okayama* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12495535&blobtype=pdf | # Novel Mechanism of the L-Type Voltage-Gated Channels/Calpains Axis in Influenza A Virus-Induced Autophagosome Activity
Hannah Murphy, Hinh Ly
Seasonal influenza viruses can cause up to 1 billion human infections and can result in 290 000-650 000 deaths annually worldwide [1]. The global influenza burden varies widely due to a complex interplay of factors, such as viral strain characteristics (i.e., transmissibility, severity, and antigenic drift), seasonal patterns (i.e., temperature, humidity, and human behavior), vaccine effectiveness, and immunization coverage [2,3]. While vaccines are the most effective means for preventing seasonal influenza A virus (IAV) infections in healthy adults, they provide suboptimal protection for high-risk groups and can be ineffective when antigenic predictions in seasonal vaccine formulation happen to be mismatched with the circulating or emerging virus strains [4]. In addition to vaccines, antivirals are widely used to control IAV infections, with the main antiviral classes targeting the viral neuraminidase and cap-dependent endonuclease, whereas the M2 inhibitors not being recommended due to the rapid emergent of viral resistance [5,6]. The circulating IAV strains continuously develop resistance to all available forms of antivirals that highlight the need for new approaches against IAV infection.
The recently published article by Tian et al., entitled "Influenza A Virus Induces Autophagosome by Inhibiting LTCC/Calpain 2/LC3A Signaling to Promote Viral Replication" in the Journal of Medical Virology [7], investigated a novel mechanism of the L-type voltage-gated channels (LTCC)/calpains axis in IAV-induced autophagosome activity. Using the PR8 (H1N1) IAV isolate, the authors demonstrated, through a series of well-thought-out experiments, that IAV infection reduces LTCC-mediated Ca 2+ influx; in human lung adenocarcinoma A549 cells, Cav1.3 is the predominant LTCC isoform, and its knockdown (KD) phenocopies LTCC blockade, consistent with suppression of Cav1.3 activity during IAV infection.
Briefly summarized, the authors demonstrated that under normal conditions (Figure 1, left), Cav1.3 induces Ca 2+ influx, activating calpain-2, which cleaves LC3A and maintains normal levels of autophagosomes. However, in PR8 (H1N1) IAVinfected cells (right), Cav1.3 is suppressed, which decreases the Ca 2+ influx, leading to calpain-2 being inactivated. LC3A is therefore being left uncleaved, and autophagosome accumulation occurs concurrently with IAV blocking the fusion of autophagosomes and lysosomes into autolysosomes. The authors showed that the accumulation of autophagosomes could promote IAV viral replication in human lung adenocarcinoma A549 cells. Under physiological conditions, LC3A is primed at the C-terminus by ATG4 proteases for subsequent lipidation [6]. This study [7] extended those findings by showing that calpain-2 can cleave LC3A (aa112-118) to suppress autophagosome formation. The authors also showed that PR8 (H1N1) IAV infection of A549 cells suppresses the Cav1.3-calpain-2 axis, which results in LC3A-dependent autophagosome accumulation without lysosomal fusion, thus promoting viral replication (Figure 1, right).
Experimentally, the authors used PR8 (H1N1) IAV-infected cells and pharmacological agonists and antagonists to show that cellular autophagy is inhibited by IAV at a late step in the autophagic pathway, that is, the fusion of autophagosomes and lysosomes. By using live-cell Ca 2+ imaging throughout the virus infection cycle, they showed a persistent level of decline in cytoplasmic Ca 2+ levels, suggesting an upstream ion-channel regulation. When pharmacologic activation of LTCCs was performed, Ca 2+ levels were partially restored, reducing autophagosome formation and therefore hindering PR8 (H1N1) IAV replication. On the contrary, LTCC inhibition yielded the opposite effects, indicating that LTCCs sit upstream of the IAVinduced block of cellular autophagy. Knockdown (KD) of Cav1.3, an LTCC isoform, resulted in increased autophagosome formation and enhanced viral replication, which led the authors to investigate the role of calpains, which are downstream targets of LTCC and are known to be important in autophagy. KD of calpain-2, but not calpain-1, significantly upregulated the expressions of the cellular LC3-II and the viral M1 gene, which is an abundantly expressed IAV protein and can be used as a marker for viral replication, thereby leading the authors to conclude that calpain-2 is mediating the effect of LTCC/Cav1.3 on autophagosome formation (i.e., by cleaving LC3), as well as on PR8 (H1N1) IAV replication. Lastly, the authors used a combination of in-silico site prediction with tag-orientation assays and a cleavage-resistant mutant to localize LC3A cleavage to the C-terminus (aa 112-118), which suppresses autophagosome formation. Traditionally, autophagy has been viewed as an antiviral defense pathway (e.g., via antigen presentation, xenophagy) [8]; however, many viruses can manipulate the host autophagy machinery to enhance their own replication [9,10]. IAV is a clear example as it can drive autophagosome accumulation and blocks autolysosome formation, to benefit viral replication [11]. However, pharmacologic activation of LTCC via the BAY K8644 compound restores Ca 2+ signaling, attenuates autophagosome build-up, and suppresses IAV replication in vitro. As such, this study [7] highlights a potential practical pharmacological intervention to functionally separate the autophagy mechanism as "friend" versus "foe." Beyond understanding the underlying mechanism, the translational capabilities are numerous not only for IAV but also for many other human respiratory viruses that mirror IAV's ability to exploit the increased autophagosome accumulation level and to stall autophagosome-lysosome fusion. Some of these examples include SARS-CoV-2 (e.g., ORF3a blocking autophagosome and/or amphisome fusion with lysosomes) [12], human parainfluenza virus type 3 (HPIV3) (e.g., HPIV3 phosphoprotein prevents host SNARE proteins from mediating autophagosome-lysosome fusion) [13], and respiratory syncytial virus (RSV) (e.g., RSV inhibits autophagosome-lysosome fusion but IL-22 restores cellular autophagy) [14].
While the authors of the current study have described the involvement of the LTCC-calpain-LC3A axis in IAV infection [7], several mechanistic details remain uncharacterized, specifically how IAV selectively downregulates Cav1.3. Downregulation of Cav1.3 by IAV may occur at various levels, including transcriptional/translational control, posttranslational removal, or modulation of the L-type voltage-gated channels. Experiments to carefully discriminate between the possible mechanisms and levels of Cav1.3 downregulation will help determine whether virusinduced LTCC agonism is a result of a trafficking deficit or a gating block. While the authors have provided compelling evidence via Western blotting analysis to support the cleavage of LC3A by calpain-2, adding biochemical analysis, for example, cleavage mapping by mass spectrometry, would clarify the mechanism further.
Additionally, expanding the in vivo mouse lung transmission electron microscopy data to include some biological readouts, such as viral titrations, animal survival curves, and histopathological analysis, would further strengthen the conclusions. It is important to also note that this study utilized a laboratory strain of IAV (PR8 H1N1) and at a relatively high multiplicity of infection (MOI = 2), which can limit its generalizability. Future confirmation studies in primary human airway cell cultures and using currently circulating IAV isolates at lower (multi-cycle) MOIs could help address physiologically relevant concerns. Additional studies of the specifics of which LC3 paralog is most important in airway epithelium and whether there are any redundant mechanisms present in this LTCC-calpain-LC3A axis need to be done. Finally, while the M2 protein of IAV has been shown to block autophagosome-lysosome fusion [15] and that it contains an LC3-interacting region that sequesters LC3 to viruscontrolled membranes [16], the specific viral protein(s) responsible for the downregulation of Cav1.3 remains to be identified.
Overall, the current study [7] provides some compelling evidence for a model where PR8 (H1N1) IAV can manipulate the LTCC/calpain-2/LC3A pathway to enhance its own replication. Future studies are needed to characterize the level (i.e., gene regulation, trafficking, or gating) at which Cav1.3 is being downregulated and to pinpoint the specific viral protein(s) responsible, which will help clarify the mechanism and increase the translatability factor of the study.
## References
1. (2025) *Influenza (Seasonal)*
2. Cdc (2024) *About Estimated Flu Burden. Flu Burd*
3. Bekkat-Berkani, Romano-Mazzotti (2018) "Understanding the Unique Characteristics of Seasonal Influenza Illness to Improve Vaccine Uptake in the US" *Vaccine*
4. Houser, Subbarao (2015) "Influenza Vaccines: Challenges and Solutions" *Cell Host & Microbe*
5. Cdc (2025) "Influenza Antiviral Medications: Summary for Clinicians. Influenza Flu"
6. Jones, Yen, Adams et al. (2023) "Influenza Antivirals and Their Role in Pandemic Preparedness" *Antiviral Research*
7. Tian, Liu, Zheng (2025) "Influenza A Virus Induces Autophagosome by Inhibiting LTCC/Calpain 2/LC3A Signaling to Promote Viral Replication" *Journal of Medical Virology*
8. Shoji-Kawata, Levine (2009) "Autophagy, Antiviral Immunity, and Viral Countermeasures" *Biochimica et Biophysica Acta (BBA) -Molecular Cell Research*
9. Chen, Tu, Ding et al. (2023) "The Role of Autophagy in Viral Infections" *Journal of Biomedical Science*
10. Choi, Bowman, Jung (2018) "Autophagy During Viral Infection -A Double-Edged Sword" *Nature Reviews Microbiology*
11. Zhou, Zhang, Dong et al. (2022) "The Battle for Autophagy Between Host and Influenza A Virus" *Virulence*
12. Miao, Zhao, Li (2021) "ORF3a of the COVID-19 Virus SARS-CoV-2 Blocks HOPS Complex-Mediated Assembly of the SNARE Complex Required for Autolysosome Formation" *Developmental Cell*
13. Ding, Zhang, Yang (2014) "Phosphoprotein of Human Parainfluenza Virus Type 3 Blocks Autophagosome-Lysosome Fusion to Increase Virus Production" *Cell Host & Microbe*
14. Das, St, Croix et al. (2020) "Interleukin-22 Inhibits Respiratory Syncytial Virus Production by Blocking Virus-Mediated Subversion of Cellular Autophagy"
15. Gannagé, Dormann, Albrecht (2009) "Matrix Protein 2 of Influenza A Virus Blocks Autophagosome Fusion With Lysosomes" *Cell Host & Microbe*
16. Beale, Wise, Stuart et al. (2014) "A LC3-Interacting Motif in the Influenza A Virus M2 Protein Is Required to Subvert Autophagy and Maintain Virion Stability" *Cell Host & Microbe* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772335&blobtype=pdf | # Erratum for Nachman et al., "Increased fungal burden in the gastrointestinal tract of brain-dead organ donors"
Erika Nachman, Colleen Ardis, A Kyle, B Ardis, Javier Nieto, Madeline Bresson, Clare Robertson, Maggie Seale, Nora Villafuerte, Zhe Lyu, Eva Preisner, Heather Danhof, Sara Rienzi, Yolanda Becker, Robert Britton |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817915&blobtype=pdf | # Short-chain fatty acids butyrate and acetate limit Zika virus replication and associated ocular manifestations via the Gprotein coupled receptor 43/FFAR2
Nikhil Deshmukh, Prince Kumar, Lal Kumar, Vaishnavi Balendiran, Pawan Singh
## Abstract
Short-chain fatty acids (SCFAs) are gut microbial metabolites produced by gut microbiota from dietary fiber. SCFAs have shown both pro-and anti-viral roles among different viruses and are known to regulate immune functions during infections. However, their role against the Zika virus (ZIKV), in general, and ocular infection, in particular, has never been investigated. In the present study, we aimed to examine the role of three SCFA derivatives: phenylbutyrate (PBA), sodium butyrate (NaB), and sodium acetate (NaAC), on ZIKV replication and associated ocular complications using primary human trabecular meshwork cells (HTMCs) and an IFNAR1-deficient mouse model of ocular infection. Our findings reveal that PBA and NaAc treatment dramatically suppressed the ZIKV replication in HTMCs. NaB showed a slightly less effect than PBA and NaAc. PBA and NaAc treatment significantly attenuated the ZIKV-induced inflammatory cytokine, interferons, and interferon-stimulated genes response via antagonizing the RIG-I/NFκB/MAPKs/STAT1-3 signaling pathways. We discovered that ZIKV induces the expression of free fatty acid receptor 2 (FFAR2)/GPR43 in HTMCs, which is further potentiated by PBA/NaAc. Pharmacological inhibition of FFAR2 abrogated the protec tive abilities of PBA/NaAc and significantly increased viral replication. Blocking FFAR2 receptors promoted ZIKV-induced cell death, which was suppressed by PBA and NaAc. Butyrate and acetate also inhibited ZIKV binding and cellular entry and inactivated the virus before internalization. PBA and NaAc treatment in mice attenuated the ZIKV-induced ocular manifestations (intraocular pressure, RPE/retinal atrophy, and TM/ anterior segment inflammation), which was abrogated by FFAR2 inhibition by 4-CMTB, a selective pharmacological inhibitor of FFAR2. Collectively, our findings indicate that SCFA treatment is an effective approach to limit ZIKV replication and associated ocular damage and may be worth exploring as a means to treat or prevent ZIKV-induced ocular complications/glaucoma in humans.IMPORTANCE ZIKV is known to cause severe ocular manifestations in in-utero exposed infants; however, the molecular mechanisms of ZIKV-induced ocular complications remain unknown. SCFAs have demonstrated both pro-and anti-viral roles against different viruses; however, their role against ZIKV is unknown. We showed that SCFAs butyrate and acetate suppress ZIKV transmission and associated ocular complications. The anti-ZIKV activity of these SFACs is mediated via FFAR2, and pharmacological inhibition of FFAR2 promotes ZIKV-induced inflammatory and cell death responses, as well as ocular malformations.
## RESULTS
## SCFAs phenylbutyrate and sodium acetate restrict ZIKV replication in HTMC
SCFAs butyrate and acetate have demonstrated differential antiviral activities against different viruses. However, their role in flaviviruses, such as ZIKV infection, is entirely unknown. Here, we investigated the role of three SCFAs: PBA, NaB, and NaAc in ZIKV replication using primary human trabecular meshwork cells (HTMCs). Previously, we demonstrated that ZIKV has high tropism toward the anterior segment of the eye and can permissively infect HTMCs (11). To investigate the role of these SCFAs, we first pretreated HTMC with PBA, NaB, and NaAc prior to ZIKV infection. We observed that PBA and NaAc significantly inhibited the ZIKV replication as measured by ZIKV-envelope (E) antigen (4G2) immunofluorescence staining (Fig. 1A). To further confirm the inhibition of ZIKV replication by these SCFAs, we performed immunoblotting for ZIKV nonstructural protein NS3. Our data show that PBA and NaAc significantly inhibited ZIKV replication, as indicated by the suppression of NS3 protein in treated cells compared to untreated cells (Fig. 1B andC). NaB also demonstrated ZIKV inhibitory properties although to a slightly lesser extent than PBA and NaAc (Fig. 1A through C). To further examine the effect of butyrate and acetate on the number of replicating virions, we performed a plaque assay. Our results revealed significant inhibition of ZIKV plaques by PBA, NaB, and NaAc compared to untreated ZIKV infected cells, confirming their anti-ZIKV role (Fig. 1D). Together, our findings confirm that SCFAs phenylbutyrate and sodium acetate can inhibit ZIKV replication and transmission.
## SCFAs butyrate and acetate diminish ZIKV-induced pro-inflammatory cytokines and IFNs/ISGs response
SCFAs have been demonstrated to modulate inflammatory and antiviral responses although different studies have shown contentious results. However, how SCFAs modulate inflammatory and antiviral responses in ocular milieu upon ZIKV infection remains unknown. Previously, we demonstrated that ZIKV infection in HTMC induces a dysregulated cytokine and IFN response (11). Thus, to gain molecular insight into the effect of butyrate and acetate on the inflammatory cytokine and IFNs/ISGs response to ZIKV infection, we challenged HTMCs with and without PBA/NaB/NaAc and measured the host innate inflammatory and antiviral response 48 h post-infection via qPCR. Our results show that NaAc, PBA, and NaB significantly reduced the mRNA expression levels for classical pattern recognition receptors (PRRs) (e.g., RIG-I, TLR-3, and MDA5), inflammatory cytokines/chemokines (e.g., IL-6, IL-1β, and CCL-4), IFNs (e.g., IFN-α2, IFN-β1, and IFN-γ), and ISGs (ISG-15, OAS2, and MX1) as compared to ZIKV-infected cells (Fig. 2A).
To further investigate if the observed reduction in cytokines and anti-viral response is due to decreased viral replication, we tested the expression of these mediators in a nonvirally stimulated condition. We challenged HTMCs with Poly I:C in the presence and absence of PBA/NaAc and measured the mRNA expression of these PRRs, cytokines, IFNs, and ISGs via qPCR. Similar to ZIKV, PAB and NaAc significantly suppressed the Poly I:C induced PRRs (RIG-I, TLR-3, and MDA5), inflammatory cytokines/chemokines (IL-6, IL-1β, CCL-4), IFNs (IFN-α2, IFN-β1, and IFN-γ), and ISGs (ISG-15, OAS2, and MX1) response (Fig. 2B), indicating the decreased level of innate response is not only due to decreased viral replication.
NFκB, MAPK, and STATs signaling regulate the expression of viral-mediated cytokines, IFNs, and ISGs responses. Thus, to identify the mechanism by which NaAc, PBA, and NaB block the expression of these PRRs, cytokines, IFNs, and ISGs, we measured the activation of NFκB, MAPK (ERK1/2 and p38), STAT1, STAT2, and STAT3 signaling pathways in the presence and absence of these SCFAs following ZIKV infection. The western blot analysis of ZIKV-infected HTMC demonstrated increased phosphorylation of NFκB, ERK1/2, p38, STAT1, STAT2, and STAT3 (Fig. 3). In contrast, NaAc, NaB, and PBA treatment significantly inhibited the activation of all of these signaling pathways (Fig. 3). We next assessed whether butyrate and acetate regulated the expression of antiviral/ISGs mediators and found that PBA, NaB, and NaAc treatment suppressed the expression of critical antiviral/ ISGs mediators, such as RIG-I, IRF3, and IFIT2 (Fig. 3). Together, our data indicate that SCFAs butyrate and acetate inhibit the expression of pro-inflammatory cytokines, interferons, and ISGs by inhibiting the NFκB, MAPK, and STAT1/2/3 signaling pathways.
## FFAR2/GPCR43 inhibition enhances the HTMC susceptibility toward ZIKV
SCFAs, acetate and butyrate, are known ligands for FFAR2/GPCR43. FFAR2 is expressed by the intestinal epithelium, brain, and various subsets of immune cells such as T cells, monocytes, macrophages, neutrophils, and dendritic cells (31,36). However, their expression by the trabecular meshwork (TM) is unknown. Activation of FFAR2 by SCFAs has been shown to diminish susceptibility toward various microbial pathogens, including bacteria and viruses (37, 38), but their role in flaviviral infections such as ZIKV has never been demonstrated. Therefore, here we aimed to investigate the role of FFAR2 in ZIKV infectivity. To assess the role of FFAR2 in HTMCs, we pretreated the cells with PBA and NaAc prior to ZIKV infection and measured the expression of FFAR2 via qPCR, western blotting, and immunofluorescence staining. Our results from all three assays demonstra ted that ZIKV significantly induces the expression of FFAR2 at transcript as well as protein levels in HTMCs, which is potentiated by its ligands PBA and NaAc (Fig. 4A through C). To further test the role of FFAR2 in ZIKV infectivity, we blocked the FFAR2 receptor using a selective pharmacological inhibitor of FFAR2, 2-(4-chlorophenyl)-3-methyl-N-(thiazol-2yl)butanamide (4-CMTB) (31,39). Our data show that the inhibition of FFAR2 either by 4-CMTB alone or in the presence of PBA/NaAc substantially increased the ZIKV replication in HTMCs, as revealed by increased ZIKV-4G2 immunofluorescence staining (Fig. 4D). Immunoblotting for ZIKV-NS3 protein further confirmed increased viral replication with FFAR2 inhibition (Fig. 4B). The inhibition of the FFAR2 receptor by 4-CMTB was confirmed by qPCR, western blotting, and immunostaining (Fig. 4A through C). Together, our results indicate that PBA and NaAc mediate their antiviral activity via FFAR2 signaling, and inhibition of FFAR2 dramatically enhances the HTMCs susceptibility towards ZIKV infection.
## SCFA treatment suppresses, while FFAR2 inhibition promotes, ZIKV-induced cell death
Previously, we demonstrated that ZIKV induces TM cell death in vitro as well as in mouse eyes (11). Here, we sought to test if SCFAs PBA, NaB, and NaAc have any effect on TM cell death and if FFAR2 inhibition modulates ZIKV-induced cell death. To test this, we infected HTMCs with ZIKV in the presence and absence of PBA and NaAc with and without FFAR2 inhibition and performed a TUNEL assay to assess cell death. Our results show that PBA and NaAc treatment significantly protected the HTMCs from ZIKV-induced cell death, as evidenced by decreased TUNEL-positive cells compared to ZIKV-infected/untreated cells (Fig. 5A andB). However, the FFAR2 blocking via 4-CMTB abolished the cell protective ability of PBA and NaAc and significantly enhanced the ZIKV-induced cell death (Fig. 5A andB). Our findings indicate that PBA and NaAc protect HTMCs from ZIKV-induced cell death, and the cell protective ability of these SCFAs is mediated via FFAR2.
## SCFAs inhibit ZIKV infection by direct viral inactivation and impairment of cellular binding and entry
SCFAs have shown differential antiviral activities against different viruses. Some studies have shown that SCFAs supports viral replication by suppressing the interferon response, while some have suggested suppressed viral replication despite the reduction of inflammatory and antiviral response (30,40). We were puzzled by our observations that butyrate and acetate inhibit ZIKV replication in HTMCs as well as suppress the inflammatory/IFNs/ISGs response. This indicates an alternative antiviral mechanism of SCFAs independent of IFNs/ISGs signaling. Thus, we aimed to determine the mode of action of these SCFAs on ZIKV. SCFAs have been shown to influence viral entry, replication, and reactivation (31,41,42). Therefore, we decided to test whether SCFAs, butyrate and acetate, alter ZIKV binding and entry to HTMCs. To test this, we performed viral attachment and entry assays by measuring the viral RNA copy numbers following incubation with SCFAs as described in the methodology section. Our results from the viral attachment assay revealed a significant decrease in viral copy numbers with SCFAs treatment (Fig. 6A). Similarly, the entry assay also demonstrated a substantial decline in viral copy numbers (Fig. 6B), confirming the ability of butyrate and acetate to modulate viral internalization. To assess the direct virucidal effect of SCFAs, ZIKV was incubated with PBA, NaB, or NaAc in a cell-free medium for 2 and 4 h. Following incubation, the drug-viral mixtures were serially diluted (2-5 log), and the viral infectivity was quantified by plaque assay. Our data revealed that all three SCFAs significantly reduced the number of plaques (> 2 log 10 ) compared to untreated control, indicating viral inactivation (Fig. 6C).
To further confirm the effect of SCFAs on viral internalization, we tested the anti-ZIKV activity of PBA and NaAc in pre-infection (before viral internalization) and 6 and 12 h post-infection (after viral internalization) treatment settings. We observed that PBA and NaAc suppressed the viral replication in pretreated groups; however, ZIKV positivity was considerably higher in post-treatment settings in comparison to the pretreated groups (Fig. 6D). Though, both pre-and post-treatment with butyrate and acetate suppressed the viral replication up to some extent compared to ZIKV-infected and untreated cells (Fig. 6D). Collectively, these results indicate that SCFAs can manipulate early viral life cycle by modulating viral attachment and entry to the HTMCs.
## SCFA treatment attenuates, while FFAR2 inhibition exacerbates, ZIKVinduced ocular manifestations in mice
Previously, we showed that ZIKV induces anterior segment (AS) inflammation leading to TM damage and elevated intraocular pressure (11). We also demonstrated that ZIKV migrates from the anterior chamber to the back of the eye and causes chorioretinal atrophy, retinal and optic nerve damage (10,11). After observing the anti-ZIKV effect in vitro in HTMCs, we reasoned to test their therapeutic potential against ZIKV-induced ocular complications in mice. To test this, we pretreated (1 day before) mice with PBA and NaAc via i.p. administration, followed by ZIKV infection and subsequent treatment for 4 consecutive days. For FFAR2 inhibition groups, we pretreated mice with 4-CMTB via i.p. injections 1 day before PBA/NaAc pre-treatment and once again on the 3 rd day after the first administration for consistent FFAR2 blocking. To test the baseline effect of FFAR2 inhibition on ZIKV-induced ocular pathology, we treated another group of animals with 4-CMTB alone without any SCFAs treatment. The timeline for the 4-CMTB/ SCFAs treatment and ZIKV infection is shown in Fig. 7A. As anticipated, ZIKV infection significantly elevated the IOP (Fig. 7B), caused severe RPE/chorioretinal atrophy (Fig. 7C), and RPE/outer retinal layer disruption (Fig. 7D) in comparison to uninfected controls. In contrast, PBA and NaAc treatment significantly reduced the IOP (Fig. 7B) and attenuated the ZIKV-induced RPE/retinal damage as revealed by fundus (Fig. 7C) and optical coherence tomography (OCT) (Fig. 7D) imaging. Remarkably, pharmacological inhibition of FFAR2 by 4-CMTB antagonizes the protective ability of PBA/NaAc and aggravates ZIKV-induced RPE/chorioretinal atrophy, outer retinal layer disruption, and IOP elevation (Fig. 7B through D). The fundus and OCT exam on the 4-CMTB alone-treated group showed dramatically enhanced ZIKV-induced pathology compared to the rest of the treatment groups (Fig. 7C andD).
To further test the role of PBA and NaAc and their receptor FFAR2 on ZIKV-induced AS inflammation, we measured the expression of various inflammatory mediators in the AS tissue via qPCR with and without FFAR2 inhibition. Our result reveals that butyrate and acetate treatment significantly diminished the mRNA expression of ZIKV-induced PRRs (RIG-I, TLR3), inflammatory cytokines/chemokines (IL-6, IL-1β, CCL-4), IFNs (IFN-α2, IFN-β1), and ISGs (ISG-15, OAS2, MX1) in mouse AS tissue (Fig. 8A). Blocking the FFAR2 receptor by 4-CMTB reversed the anti-inflammatory effect of PBA and NaAc and significantly enhanced the expression of these inflammatory mediators. Since we observed an antagonizing effect on inflammatory and ISGs pathways by PBA and NaAc in vitro, we tested the role of these SCFAs on these pathways in our in vivo model. Similar to HTMCs, PBA/NaAc inhibited the NFκB, MAPKs (p38, ERK1/2), and STATs (STAT1, STAT3) and ISGs (RIG-I, IRF7) pathways activation in mouse AS/TM tissue, while FFAR2 inhibition reversed these inhibitory properties (Fig. 8B andC). Together, these findings indicate that SCFA suppresses ZIKV-induced AS inflammation and ocular manifestations, and the in vivo protective effect of acetate and butyrate is also mediated via FFAR2.
## DISCUSSION
Recent advancements in the microbiome-immunity axis have uncovered the involve ment of gut microbiomes and their metabolites in multiple diseases. A recent study demonstrated alterations in gut microbiota in immunocompetent mice (43) and enhanced peripheral and nervous system inflammation due to decreased SCFA levels in macaques (44) during ZIKV infection. SCFAs, acetate, butyrate, and propionate have demonstrated promising therapeutic potential against multiple microbial pathogens, including bacteria and viruses. Although due to both pro-and anti-viral properties, SCFAs have shown very complex and contrary roles with different viruses (17). Emerging literature suggests an interplay between gut microbiota and ocular health (45,46). However, the role of SCFAs against ZIKV ocular infection remains unknown. Here, we showed a previously unreported role of butyrate and acetate and their receptor FFAR2 in ZIKV ocular pathogenesis. In this study, we discovered that SCFAs, PBA, NaB, and NaAc restrict ZIKV transmission and modulate viral attachment and internalization to the cells. We further uncovered that PBA and NaAc mediate their anti-ZIKV activity via FFAR2, and pharmacological inhibition of FFAR2 antagonizes the protective abilities of PBA and NaAc. SCFA treatment attenuates ZIKV-induced pro-inflammatory response, cell death, and ocular complications.
ZIKV infection is linked to multiple congenital malformations, including microcephaly and ocular manifestations. In utero exposure to ZIKV caused severe ocular abnormalities, including chorioretinal atrophy, RPE mottling, optic nerve damage, and congenital glaucoma (10,(14)(15)(16). Trabecular meshwork (TM) regulates the intraocular pressure (IOP), and pathologic stress may cause TM damage, resulting in increased glaucoma phenotype (11,47). Recently, we discovered that ZIKV has high tropism toward the anterior segment of the eye and can cause TM damage, resulting in increased IOP and retinal ganglion cell (RGC) loss (11). In this study, we aimed to test whether SCFAs can protect TM from ZIKV infection. We found that SCFA derivatives PBA, NaB, and NaAc dramatically reduced ZIKV replication and transmission. We further discovered that butyrate and acetate inhibit viral binding and cellular entry to the TM cells. The SCFAs also protected TM from ZIKV-induced cell death. Our study corroborated with previous findings where SCFA treatment has been shown to suppress replication of SARS-CoV-2 (22), HBV (35), HSV-1 (32-34), and porcine epidemic diarrhea virus (PEDV) (48), and exacerbate disease severity. Similarly, butyrate has been shown to effectively reduce rotavirus-induced cell death and provide protection against intestinal epithelial barrier damage (42,49). In contrast to these findings, a few other studies have shown that butyrate could promote the replication of EBV (50), IAV, HIV-1, hMPV, VSV (30), and TGEV (51), whereas it has no effect on SeV replication (30). These findings suggest that, given the diverse role of SCFAs among different viruses, it is imperative to interpret their pro-or anti-viral activity within specific relevant physiological contexts.
Upon viral infection, host cells induce a pro-inflammatory and IFN response, which plays a crucial role in curbing infection. However, uncontrolled immune activation upon infection may lead to persistent inflammation, resulting in tissue damage, morbidity, and mortality. The recurrent inflammation of ocular tissue, an immune-privileged organ, is detrimental and results in vision loss. ZIKV has been shown to induce inflammatory cytokines, IFNs, and ISGs in different models (52). We recently observed a dysregulated immune response via ZIKV in TM, resulting in trabeculitis and TM damage (11). SCFAs are well-characterized for their anti-inflammatory activities via HDAC inhibition (17,51). In addition, PBA is also a well-known endoplasmic reticulum stress inhibitor and has been shown to modulate CREB3 pathways in HSV-induced ocular infection, suppress viral replication, and associated inflammation (32). Recent studies have demonstrated that SCFAs can suppress inflammatory cytokines and virus-induced IFNs/ISGs response to confer protection from tissue damage (17,33,34). In line with these studies, we observed that PBA, NAB, and NaAc can suppress ZIKV-induced inflammatory (IL-6, IL-1β, CCL-4), (48). Similarly, acetate has shown enhanced IFN-β production against RSV (38) and NLRP3-IFN-I-mediated antiviral response against IAV (56). However, our finding corroborated with other studies showing the suppression of JAK/STAT pathways in LPS-stimulated corneal fibroblasts (57) and RIG-I signaling in TGEV and SARS-CoV-2 infection models (51,58). These differential effects could be potentially due to either the stressor (IFN-β vs PEDV vs ZIKV infection) or the cell models (transformed colon/lung cell line vs porcine epithelial cells vs human primary TM) used in these studies, vs our study. As discussed by Yin et al., the differential effect of SCFAs could also be attributed to the distinct SCFAs treatment methods (51). Indeed, we observed alteration of viral binding and cellular entry by SCFAs and, therefore, differential effects in viral dissemination in pre-vs post-treatment. Similarly, one other study has demonstra ted the differential impact of acetate treatment in mice with simultaneous, before, or after IAV infection (56). Collectively, these findings imply the pleiotropic role of SCFAs on viral infection and immune function and underline careful consideration when applying the findings to human patients with different viral infections.
Canonically, SCFAs mediate their activity by activating the GPCRs, also known as the free fatty acid receptors (FFAR). FFAR2 is expressed by multiple cells, including immune cells, skeletal muscles, smooth muscles, and neurons, and, therefore, assists in maintaining homeostasis by SCFAs in the colon, kidney, joints, lungs, and brain. In the eye, FFAR2 is expressed by corneal epithelium and endothelium, iris, ciliary body, inner nuclear, outer nuclear, and ganglion cell layers (18). FFAR2 plays a critical role in regulating gut homeostasis, energy utilization, immune cell functions, and suppression of inflammatory response. It has also been shown to play a protective role in non-infec tious ocular diseases such as uveitis (18). However, their role in ZIKV ocular pathogenesis is entirely unknown. Here, we observed for the first time that TM cells express the FFAR2 receptor and ZIKV modulates its expression. Our study revealed that butyrate and acetate potentiated the FFAR2 expression on TM, and pharmacological inhibition of FFAR2 promoted ZIKV replication. Furthermore, FFAR2 inhibition enhanced the ZIKVinduced cell death, which was suppressed by PBA and NaAc. Our findings corroborated with previous studies where FFAR2 knockdown has shown enhanced susceptibility to C. rodentium (59), Klebsiella pneumoniae lung infection (60), and TLR-induced keratitis (18). Activation of FFAR2 in pulmonary epithelial cells has shown reduced virus-induced cytotoxicity (38). Similarly, targeted activation of FFAR2 has shown diminished suscepti bility toward S. aureus, RSV, and IAV and Streptococcus pneumoniae superinfection (28,37,38,61). In summary, our work reveals an antiviral role of SCFAs, butyrate, and acetate, restricting ZIKV transmission and associated ocular manifestations via FFAR2 signaling (Fig. 9). The inhibition of FFAR2 reverts the protective effect of butyrate and acetate and promotes ZIKV-induced cell death. Mechanistically, butyrate and acetate not only directly inactivate ZIKV in a cell-free environment but also inhibit viral binding and cellular entry into the HTMCs. Collectively, our findings, along with those of others, indicate that SCFAs have multiprong effects; they can modulate the host responses as well as viral infection by interfering with early stages of the viral lifecycle, including viral binding and entry to the cells. Our study could aid the future design of therapeutic interventions to treat or prevent ZIKV infection and associated ocular complications in humans. However, given the diverse effects of SCFAs on different viruses, their safety and efficacy must be ensured prior to therapeutic use in human subjects with various viral infections. Although we demonstrated the role of FFAR2 in ZIKV transmission using a potent and selective inhibitor of FFAR2, 4-CMTB, future studies using FFAR2 knockout are essential to decipher the role of FFAR2 in ZIKV pathogenesis. Similarly, dietary manipulations or appropriate supplements to augment SCFA production could be explored as potential antiviral strategies against ocular viral infection.
## MATERIALS AND METHODS
## Antibodies and reagents
Antibodies used in this study were purchased from the following sources: 4G2 (GeneTex, #GTX57154), NS3 (GeneTex, #GTX133309), β-actin (Millipore Sigma, #A2228), pNFκB (#3033), NFκB (#6956) pERK1/2 (#4370), ERK (#4695), pP38 (#4511), P38 (#8690), pSTAT1 (#9167), STAT1 (#14994), pSTAT2 (#88410), STAT2 (#72604), pSTAT3 (#9145), STAT3 (#9139), RIG-I (#3743), pIRF3 (#79945), IRF3 (#4302), and IFIT2 (#92633) antibodies were purchased from Cell Signaling Technology (Danvers, MA). SCFAs NaAc (Thermofisher, #A13184-30), 4-PBA (#11323), NaB (#13121), and FFAR2 receptor antagonist 4-CMTB (#29680) were purchased from Cayman Chemicals (Ann Arbor, MI). Poly I:C (#tlrl-pic) was purchased from Invivogen (San Diego, CA).
## Cell culture
Primary human trabecular meshwork cells (HTMCs) (11) were cultured in Dulbecco's minimal essential medium (DMEM) low glucose GlutaMAX medium, supplemented with 10% fetal bovine serum (FBS), 1× penicillin-streptomycin solution in a CO 2 (5%) incubator at 37°C. Vero E6 cells (ATCC CRL-1586) were cultured in DMEM low glucose GlutaMAX medium containing 1× penicillin-streptomycin supplemented with 10% FBS. For virus propagation, Aedes albopictus, C6/36 cells (ATCC CRL-1660) were cultured in Eagle's Minimum Essential Medium (EMEM) containing 1× penicillin-streptomycin supplemen ted with 10% FBS in a CO 2 incubator at 30°C.
## Zika virus strain and infection procedure
The ZIKV strain PRVABC59 (NR-50240) was obtained from BEI Resources, National Institute of Allergy and Infectious Diseases (NIAID), and propagated in Aedes albopictus, C6/36 cells. The viral titers were determined using the plaque assay. Small aliquots were prepared and stored in a -80°C freezer for infection studies.
For in vitro experiments, HTMCs were challenged with ZIKV at a multiplicity of infection (MOI) of 1. For pharmacological inhibition/activation studies, cells were pretreated with respective drugs/antagonists for 1 h, followed by virus adsorption in serum-free media for 1 h, until indicated otherwise. After adsorption, cells were replen ished with media containing drugs and incubated until the desired endpoint. For Poly I:C challenge studies, HTMCs were treated with Poly I:C (100 ng/mL) for 48 h.
## Mouse infection and SCFA treatment
For SCFA treatment, mice were pretreated with PBA/NaAc (100 mg/kg) via i.p injections 1 day prior to infection and 3 consecutive days post-ZIKV infection. In the FFAR2 inhibition groups, mice were pretreated with 4-CMTB (10 mg/kg) via i.p. injection 1 day prior to PBA/NaAc treatment and another dose on day 3 (39) since the first administration as per the timeline shown in Fig. 7A. For the ZIKV infection, anesthetized animals were inoculated with 1 × 10 4 PFU of ZIKV via intracameral injections as described previously (11). Seven days post-infection, IOP was recorded using a Tonolab tonometer, and fundus imaging was performed using Micron IV (Phoenix-Micron Inc., Bend, OR). Animals were euthanized, and anterior segment tissue was harvested for inflammatory cytokines/path ways assessment via qPCR and immunoblotting.
## Viral attachment, entry, and inactivation assay
The viral attachment and entry assay was performed as described previously (62). Briefly, HTMCs were pre-incubated with PBA (3 mM), or NaB (3 mM), or NaAc (200 µM) at 4°C (for attachment assay) or 37°C (for entry assay) for 1 h, followed by ZIKV infection (MOI:1). After viral adsorption for 2 h, the cells were washed three times with fresh DMEM media, and total RNA was isolated. The viral RNA copy numbers were determined from whole-cell RNA using a TaqMan probe against the ZIKV envelope (E) gene via qPCR.
For the direct viral inactivation assay, ZIKV (1 × 10 6 PFU/mL) was incubated with PBA (3 mM), or NaB (3 mM), or NaAc (200 µM) in a cell-free medium (serum-free DMEM) at 37°C for 2 or 4 h. ZIKV incubated without any drug was used as an untreated control. After incubation, 100 µL of each viral-drug mixture was serially diluted (10 -2 to 10 -5 fold), and viral infectivity was determined by plaque assay.
## Plaque assay
ZIKV plaque assay was performed using a protocol we described recently (2). Briefly, a confluent monolayer of Vero cells was infected with serial dilutions of ZIKV stock culture. One hour following viral adsorption, the cell monolayer was overlayed with the first overlay media containing a 1:1 mixture of 2× EMEM, 4% FBS, 2× P/S, 20 mM MgCl 2 , and 1.6% Noble Agar. The following day, a second overlay media containing DMEM, 1 mg/mL BSA, 40 mM MgCl 2 , 0.2% glucose, 2 mM sodium pyruvate, 4 mM L-glutamine, 4 mM oxaloacetic acid, 1× P/S, and 0.1% sodium bicarbonate was added. The plates were incubated at 37°C for 5 days in a CO 2 incubator. Following incubation, cells were fixed with 10% Tricarboxylic Acid (TCA) for 20 min, and the agar overlay was removed gently without disturbing the cell monolayer. Viral plaques were stained with 0.2% crystal violet for 20 min, followed by a wash with MilliQ water. The plaques were counted, and titers were estimated as log 10 PFU/mL.
## Immunofluorescence assay
For IFA, cells were seeded in a Nunc four-well chamber slide (Fisher Scientific, Rochester, NY) and infected with ZIKV at an MOI of 1 at ~70%-80% confluency. Mock-treated/unin fected cells were used as controls. At desired endpoints, cells were fixed using 4% paraformaldehyde for 10 min at room temperature (RT). After three washes with 1× PBS, cells were blocked and permeabilized using 1% (wt/vol) BSA with 0.4% (vol/vol) Triton X-100 made in PBS (blocking buffer) for 1 h at RT in a humidified chamber. Cells were then incubated with the primary mouse/rabbit antibodies in the blocking buffer (1:100 dilution) overnight at 4°C in a humidified chamber. Subsequently, after three washes with 1× PBS, cells were incubated with anti-mouse/rabbit Alexa Fluor 488/594-conjuga ted secondary antibodies (1:200 dilutions) for 1 h at RT. Finally, cells were washed four times with 1× PBS and mounted in Vectashield anti-fade mounting medium containing DAPI (Vector Laboratories, Burlingame, CA). The slides were visualized and imaged using a Keyence (Keyence, Itasca, IL) fluorescence microscope.
## TUNEL assay
The cell death was estimated using the TUNEL assay (63). HTMCs were grown in Nunc four-well chamber slides (Fisher Scientific, Rochester, NY) and infected with ZIKV in the presence or absence of SCFAs or 4-CMTB. TUNEL assay was performed using ApopTag Fluorescein In Situ Apoptosis Detection Kit (#S7110) per the manufacturer's instructions (Millipore Sigma, Billerica, MA). Following TUNEL staining, cells were immunostained with ZIKV-4G2 antibodies using the IFA protocol described above. The cells were visualized and imaged using a Keyence microscope (Keyence, Itasca, IL).
## Immunoblotting
Immunoblotting was performed using a method we described previously (63). Briefly, the treated and untreated cells were washed using ice-cold PBS and lysed using RIPA lysis buffer containing Halt protease and phosphatase inhibitor cocktail (Thermo Scientific, Rockford, IL). The total protein concentration was estimated using BCA protein esti mation kit per the manufacturer's instructions (Thermo Scientific, Rockford, IL). Forty micrograms of total protein were resolved on SDS-PAGE gels and transferred onto nitrocellulose or PVDF membranes. Following transfer, the membrane was blocked using a blocking buffer containing 5% non-fat skim milk in 1× TBST. Subsequently, blots were incubated with anti-rabbit/mouse primary antibodies (1:1,000 dilutions) diluted in 3% BSA prepared in 1× TBST, overnight at 4°C with gentle agitation. After incuba tion with the primary antibodies, the membranes were washed three times using 1× TBST, followed by incubation with anti-mouse or anti-rabbit HRP-conjugated secondary antibodies (1:2,000 dilutions) at RT for 1 h. Following three washes with 1× TBST, the blots were developed using Supersignal West Femto chemiluminescent substrate and imaged using iBright FL1500 imager (Thermo Fisher Scientific, Rockford, IL).
## RNA isolation and real-time qPCR
Following treatment, cells were collected in TRIzol reagent (#15596018), and the RNA was isolated per the manufacturer's instructions (Thermo Scientific, Rockford, IL). RNA (1 µg) was reverse transcribed to cDNA using a Maxima first-strand cDNA synthesis kit, per the manufacturer's instructions (Thermo Scientific, Rockford, IL). The cDNA was amplified using mouse or human gene-specific PCR primers in a 96-well plate using QuantStudio 3 Real-Time PCR system (ThermoFisher Scientific, Rockford, IL). The relative mRNA expressions of the gene of interest were normalized with respect to the house keeping 18sRNA gene. The data were analyzed using the 2 -ΔΔC T method and represented as relative fold change.
## Statistical analysis
The statistical analysis was performed using GraphPad Prism 10 V10.1.2 (GraphPad Software, La Jolla, CA). The one-or two-way ANOVA was used to compare the statisti cal differences between the experimental groups. A P value of <0.05 was considered statistically significant. The data were expressed as mean ± SD from three biological replicates unless indicated otherwise.
## References
1. "background, MMRRC Strain # 032045-JAX) were originally purchased from Jackson Laboratories and bred in-house in a germ-free University of Missouri (MU) Office of Animal Resources (OAR) facility. Mice aged 6-10 weeks (male and female) were used in this study. Animals were housed in a restricted-access AAALAC-accredited animal facility. All animals were maintained in a 12:12 h light/dark cycle"
2. (2026) "SCFAs butyrate and acetate exhibit anti-ZIKV effects via FFAR2 signaling. Butyrate and acetate suppress ZIKV replication and transmission and reduce associated ocular manifestations. The anti-ZIKV effects of these SCFAs are mediated through FFAR2 signaling, as pharmacological inhibition of FFAR2 enhances ZIKV replication, promotes cell death, and exacerbates ZIKV-induced ocular damage. The schematic diagram was created using BioRender software (biorender.com)" *Full-Length Text Journal of Virology*
3. Camargos, Foureaux, Medeiros et al. (2019) "In-depth characterization of congenital Zika syndrome in immunocompetent mice: antibody-dependent enhancement and an antiviral peptide therapy" *EBioMedicine*
4. Singh, Ahmad, Aruri et al. (2024) "Novel quinoline substituted autophagy inhibitors attenuate Zika virus replication in ocular cells" *Virus Res*
5. Hasan, Saeed, Panigrahi et al. (2019) "Zika virus: a global public health menace: a comprehensive update" *J Int Soc Prev Community Dent*
6. Landry, Raman, Kennedy et al. (2017) "Zika Virus (ZIKV), global public health, disability, and rehabilitation: connecting the dots" *Phys Ther*
7. Sneha, Wright, Iii et al. (2024) "Targeting ABCG1 and SREBP-2 mediated cholesterol homeostasis ameliorates Zika virus-induced ocular pathology"
8. Villarroel, Hamel, Gumpangseth et al. (2024) "Global seroprevalence of Zika virus in asymptomatic individuals: a systematic review" *PLoS Negl Trop Dis*
9. Chakhtoura, Hazra, Spong (2018) "Zika virus: a public health perspective" *Curr Opin Obstet Gynecol*
10. Marbán-Castro, Goncé, Fumadó et al. (2021) "Zika virus infection in pregnant women and their children: a review" *Eur J Obstet Gynecol Reprod Biol*
12. Grant, Flechelles, Elenga et al. (2022) "Consequences of in utero Zika virus exposure and adverse pregnancy and early childhood outcomes: a prospective cohort study" *Viruses*
13. Singh, Guest, Kanwar et al. (2017) "Zika virus infects cells lining the blood-retinal barrier Full-Length Text Journal of Virology"
14. "and causes chorioretinal atrophy in mouse eyes" *JCI Insight*
15. Singh, Kasetti, Zode et al. (2019) "Zika virus infects trabecular meshwork and causes trabeculitis and glaucomatous pathology in mouse eyes. mSphere 4:mSphere"
16. Singh, Farr, Kumar (2019) "Ocular manifestations of emerging flaviviruses and the blood-retinal barrier" *Viruses*
17. Ahmad, Deshmukh, Webel et al. (2023) "Viral infections and pathogenesis of glaucoma: a comprehensive review" *Clin Microbiol Rev*
18. De, Freitas, De Oliveira Dias et al. (2016) "Ocular findings in infants with microcephaly associated with presumed Zika virus congenital Infection in Salvador, Brazil" *JAMA Ophthalmol*
19. Yepez, Murati, Pettito et al. (2017) "Ophthalmic manifestations of congenital Zika syndrome in Colombia and Venezuela" *JAMA Ophthalmol*
20. De, Freitas, Ko et al. (2017) "Glaucoma and congenital Zika syndrome" *Ophthal mology*
21. Tsui, Wu, Zhang et al. (2025) "Short-chain fatty acids in viral infection: the underlying mechanisms, opportunities, and challenges" *Trends Microbiol*
22. Wu, Chen, Grau et al. (2024) "Short chain fatty acids inhibit corneal inflammatory responses to TLR ligands via the ocular G-protein coupled receptor 43" *Ocul Surf*
23. Alvarez-Curto, Milligan (2016) "Metabolism meets immunity: the role of free fatty acid receptors in the immune system" *Biochem Pharmacol*
24. Yao, Cai, Fei et al. (2022) "The role of short-chain fatty acids in immunity, inflammation and metabolism" *Crit Rev Food Sci Nutr*
25. Bartoszek, Moo, Binienda et al. (2020) "Free fatty acid receptors as new potential therapeutic target in inflammatory bowel diseases" *Pharmacol Res*
26. Brown, Sanidad, Lucotti et al. (2022) "Gut microbiota-derived metabolites confer protection against SARS-CoV-2 infection" *Gut Microbes*
27. Silva, Bernardi, Frozza (2020) "The role of short-chain fatty acids from gut microbiota in gut-brain communication" *Front Endocrinol (Lausanne)*
28. Lan, Tang, Lu et al. (2024) "The role of short-chain fatty acids in central nervous system diseases: a bibliometric and visualized analysis with future directions"
29. Huang, Wang, Miao et al. (2024) "Short-chain fatty acids: Important components of the gut-brain axis against AD" *Biomed Pharmacother*
30. Chen, Wu, Wang et al. (2021) "Short chain fatty acids inhibit endotoxin-induced uveitis and inflammatory responses of retinal astrocytes" *Exp Eye Res*
31. Singh, Singh, Kumar (2023) "Butyrate ameliorates intraocular bacterial infection by promoting autophagy and attenuating the inflammatory response" *Infect Immun*
32. Schlatterer, Peschel, Kretschmer (2021) "Short-chain fatty acid and FFAR2 activation -a new option for treating infections?" *Front Cell Infect Microbiol*
33. Alwin, Karst (2021) "The influence of microbiota-derived metabo lites on viral infections" *Curr Opin Virol*
34. Chemudupati, Kenney, Smith et al. (2020) "Butyrate reprograms expression of specific interferon-stimulated genes" *J Virol*
35. Wang, Jiang, Wang et al. (2020) "The G protein-coupled receptor FFAR2 promotes internalization during influenza A virus entry" *J Virol*
36. Yadavalli, Suryawanshi, Koganti et al. (2020) "Standalone or combinatorial phenylbutyrate therapy shows excellent antiviral activity and mimics CREB3 silencing" *Sci Adv*
37. Sumbria, Berber, Rouse (2021) "Supplementing the diet with sodium propionate suppresses the severity of viral immuno-inflammatory lesions" *J Virol*
38. Koganti, Yadavalli, Sutar et al. (2019) "Butyrate inhibits HBV replication and HBV-induced hepatoma cell proliferation via modulating SIRT-1/Ac-p53 regulatory axis" *Mol Carcinog*
39. Ruhnke, Beyer, Kaemmerer et al. (2024) "Expression of free fatty acid receptor 2 in normal and neoplastic tissues" *Exp Mol Pathol*
40. Sencio, Barthelemy, Tavares et al. (2020) "Gut dysbiosis during influenza contributes to pulmonary pneumococcal superinfection through altered short-chain fatty acid production" *Cell Rep*
41. Krist, Fachi, De Paula et al. (2019) "Microbiotaderived acetate protects against respiratory syncytial virus infection through a GPR43-type 1 interferon response" *Nat Commun*
42. Binienda, Owczarek, Sałaga et al. (2024) "Synthetic free fatty acid receptor (FFAR) 2 agonist 4-CMTB and FFAR4 agonist GSK13764 inhibit colon cancer cell growth and migration and regulate FFARs expression in in vitro and in vivo models of colorectal cancer" *Pharmacol Rep*
43. Krist, Stein, Franceschina et al. (2022) "Short-chain fatty acid acetate triggers antiviral response mediated by RIG-I in cells from infants with respiratory syncytial virus bronchiolitis" *EBioMedicine*
44. Li, Richards, Handberg et al. (2021) "Butyrate regulates COVID-19-relevant genes in gut epithelial organoids from normotensive rats" *Hypertension*
45. Dong, Wang, Zhu et al. (2023) "Sodium butyrate protects against rotavirus-induced intestinal epithelial barrier damage by activating AMPK-Nrf2 signaling pathway in IPEC-J2 cells" *Int J Biol Macromol*
46. Corrêa, De Oliveira Santos, Braz-De-Melo et al. (2021) "Gut microbiota modulation induced by Zika virus infection in immunocompetent mice" *Sci Rep*
47. Miller, Manuzak, Gustin et al. (2023) "Elevated peripheral and nervous system inflammation is associated with decreased short-chain fatty acid Full-Length Text Journal of Virology"
48. "Zika-virus infected macaques"
50. Ciurariu, Tirziu, Varga et al. (2025) "Short-chain fatty acids and the gut-retina connection: a systematic review" *Int J Mol Sci*
51. Kammoun, Rekik, Dlensi et al. "Bouayed Abdelmoula N. 2024. The gut-eye axis: the retinal/ocular degenerative diseases and the emergent therapeutic strategies" *Front Cell Neurosci*
52. Wu, Shui, Liu et al. (2022) "Trabecular meshwork mitochondrial function and oxidative stress: clues to racial disparities of glaucoma" *Ophthalmol Sci*
53. He, Fan, Shen et al. (2023) "Butyrate limits the replication of porcine epidemic diarrhea virus in intestine epithelial cells by enhancing GPR43-mediated IFN-III production" *Front Microbiol*
54. Zhao, Hu, Jiang et al. (2021) "Protective effects of sodium butyrate on rotavirus inducing endoplasmic reticulum stressmediated apoptosis via PERK-eIF2α signaling pathway in IPEC-J2 cells" *J Anim Sci Biotechnol*
55. Saemundsen, Kallin, Klein (1980) "Effect of n-butyrate on cellular and viral DNA synthesis in cells latently infected with Epstein-barr virus" *Virology (Auckl)*
56. Yin, Liu, Yao et al. (2024) "Gut microbiota-derived butyrate promotes coronavirus TGEV infection through impairing RIG-I-triggered local type I interferon responses via class I HDAC inhibition" *J Virol*
57. Lee, Komarasamy, Adnan et al. (2021) "Hide and seek: the interplay between Zika virus and the host immune response" *Front Immunol*
58. Labib, Chigbu (2022) "Pathogenesis and manifestations of Zika virus-associated ocular diseases" *TropicalMed*
59. Plociennikowska, Frankish, Moraes et al. (2021) "TLR3 activation by Zika virus stimulates inflammatory cytokine production which dampens the antiviral response induced by RIG-I-like receptors" *J Virol*
60. Majumdar, Venkatesh, Swarup et al. (2024) "Short-chain fatty acids abrogate Japanese encephalitis virus-induced inflammation in microglial cells via miR-200a-3p/ZBTB20/IKβα axis" *mBio*
61. Niu, Cui, Yang et al. (2023) "Microbiota-derived acetate enhances host antiviral response via NLRP3" *Nat Commun*
62. Wang, Shen, Han et al. (2024) "Sodium butyrate alleviates lipopolysaccharide-induced inflammation through JAK/STAT signalling in primary human corneal fibroblasts" *Eur J Pharmacol*
63. Pascoal, Rodrigues, Genaro et al. (2021) "Microbiota-derived short-chain fatty acids do not interfere with SARS-CoV-2 infection of human colonic samples" *Gut Microbes*
64. Yang, Xiao, Huang et al. (2019) "Microbiota metabolite short-chain fatty acids facilitate mucosal adjuvant activity of cholera toxin through GPR43" *J Immunol*
65. Galvão, Tavares, Corrêa et al. (2018) "The metabolic sensor GPR43 receptor plays a role in the control of Klebsiella pneumoniae infection in the lung" *Front Immunol*
66. Schlatterer, Beck, Schoppmeier et al. (2021) "Acetate sensing by GPR43 alarms neutrophils and protects from severe sepsis" *Commun Biol*
67. Singh, Singh, Farr et al. (2019) "Interferon-stimulated gene 15 (ISG15) restricts Zika virus replication in primary human corneal epithelial cells" *Ocul Surf*
68. Monu, Ahmad, Olson et al. (2024) "SARS-CoV-2 infects cells lining the blood-retinal barrier and induces a hyperinflammatory immune response in the retina via systemic exposure" *PLoS Pathog* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12505990&blobtype=pdf | # Humoral immunity to current variants of SARS-CoV-2 in exposed adults, September 2023 to September 2024
Lara Jeworowski, Barbara Mühlemann, Felix Walper, Marie Schmidt, Jenny Jansen, Andi Krumbholz, Terry Jones, Victor Corman, Christian Drosten
## Abstract
SARS-CoV-2 evolves variants that evade population immunity. Monitor ing group-level immunity is critical for assessing population susceptibility to newly circulating variants and updateability of imprinted immunity after exposure. We established a closely monitored cohort of 58 fully vaccinated adults in Berlin, Germany. Of these, 49 had at least one previous Omicron infection. In September 2023 and again in September 2024, we analyzed neutralizing antibody responses using full-virus plaque reduction neutralization tests against seven SARS-CoV-2 variants: B.1, BA.2, BA.5, EG.5.1, JN.1, KP.3.1.1, and XEC. Vaccination and exposure histories were traced using medical records, RT-PCR testing of any episode of respiratory tract infection, and serological testing for subclinical infections. Infecting variants were determined by sequencing or from unequivocal variant circulation at the time of positive testing. Titers from Septem ber 2023 included responses to both then-current and future variants. Over the study period, 13 subjects received monovalent XBB.1.5 vaccine. Thirty-four had one, and five more than one SARS-CoV-2 infection. None of the subjects was exposed to the most recent variant, XEC. Neutralization titers against all tested variants increased over time. Highest fold increases were seen against KP.3.1.1 and XEC. Reactivity profiles differed by exposure histories reflecting the most recent variant contact. Exposure to new variants leads to relative updates in population-level neutralizing antibody activity. Despite these updates, absolute group-level neutralization activity was low in September 2024 due to low titer levels against currently circulating variants KP.3.1.1 and XEC. Ongoing monitor ing is needed to assess the need for further vaccine updates. IMPORTANCE As new Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) variants continue to emerge, understanding how population immunity evolves is essential to guide vaccine updates and public health strategies. Our study follows a group of fully vaccinated adults for 1 year (September 2023 to September 2024) to track how infection and vaccination affect the ability to neutralize new viral variants. Despite the continued emergence of immune escape variants, the results show that infection with recent variants helps to "update" immunity at the group level, even against newer variants such as KP.3.1.1 and XEC, although titers to new variants were low, confirming the existence of immune imprinting. These findings suggest that exposure to new variants adapts the immune system over time. This provides valuable insight into how populations build resilience against SARS-CoV-2 and whether updated vaccines are needed.
were first reported in multiple countries (https://www.who.int/docs/default-source/ coronaviruse/21112023_ba.2.86_ire.pdf, https://www.who.int/docs/default-source/coron aviruse/18122023_jn.1_ire_clean.pdf?sfvrsn=6103754a_3). Compared to their ancestral variant (BA.2), variants BA.2.86 and JN.1 had 32 and 33 amino acid substitutions in the spike protein, respectively, and JN.1 rapidly became the globally dominating variant in January 2024 (https://gisaid.org/hcov-19-variants-dashboard/). In early 2024, three JN.1 descendant variants JN.1, KP.2, KP.3, and its descendant KP.3.1.1, arose, with KP.3.1.1 dominating circulation worldwide by September 2024 (https://cov-spectrum.org/ explore/World/AllSamples/from%3D2024-06-01%26to%3D2025-04-08/variants?nextclad ePangoLineage=KP.3.1.1*&). All have three to four amino acid substitutions in the spike protein compared to JN.1 and were classified as variants under monitoring by the WHO (https://www.who.int/activities/tracking-SARS-CoV-2-variants/). In June 2024, the recombinant XEC variant arose and co-circulated during the second half of 2024 at high levels with KP.3.1.1 (https://www.who.int/docs/default-source/coronaviruse/ 09122024_xec_ire.pdf). XEC is a recombinant of KS.1.1 and KP.3.3 and differs from KP.3.1.1 at three positions in the spike protein (1).
The ongoing evolution of immune escape variants has been countered by recurring vaccine updates. Since April 2024, the WHO has recommended the use of JN.1 lineage vaccines, including monovalent JN.1, KP.2, and most recently LP.8.1 vaccines (https://www.who.int/news/item/26-04-2024-statement-on-theantigen-composition-of-covid-19-vaccines, https://www.who.int/news/item/15-05-2025-statement-on-the-antigen-composition-of-covid-19-vaccines). It is currently unclear whether the future trajectory of population immunity will require further vaccine updates. Whereas mounting population-level cellular immunity may confer additional protection, it remains a subject of debate whether and to what extent re-exposure to current variants confers an update toward group-level immunity that can overcome the imprinting on B-cell immunity by past variants. One major challenge in answering this question is the scarcity of relevant SARS-CoV-2 data following the discontinuation of testing and tracing programs.
To provide insight into the development of group-level immunity upon re-exposure with current variants, we assembled a well-characterized longitudinal cohort of adults living in Berlin (2). We followed the development of group-level neutralizing immunity from 2023 to 2024 with steady monitoring of respiratory tract infection (RTI) episodes and virus neutralization testing against recently circulating variants. While we observe clear neutralization escape by recent variants, there is also a clear indication of an updating of group-level immunity due to natural exposure to recent variants.
## RESULTS
## Cohort
To assess changes in neutralization activity against emerging SARS-CoV-2 Omicron variants in a cohort with known exposure histories, we collected sera from 58 indi viduals in September 2023 and re-sampled the same individuals 1 year later. The individuals represented a predominantly young and healthy population resident in Berlin and surrounding areas (median age 35 years, age range 22-60 years, 57% female) (Table 1). By September 2023, all individuals had received at least three vaccinations against the SARS-CoV-2 wild-type variant using mRNA (Comirnaty, Pfizer, New York, the United States/BioNTech, Mainz, Germany or Spikevax, Moderna Bio tech, Madrid, Spain) or vector-based (Vaxzevria, AstraZeneca, Cambridge, the United Kingdom) vaccines and 49 had at least one infection identified by PCR, rapid anti gen tests, and/or detection of anti-SARS-CoV-2 nucleocapsid antibodies. All reported infections had taken place during the time of circulation of the Omicron variants. Between September 2023 and September 2024, 13 individuals received a monovalent booster vaccination against XBB.1.5 (Comirnaty, Pfizer, New York, the United States/ BioNTech, Mainz, Germany) and 34 had at least one and 5 at least two SARS-CoV-2 infections, as confirmed by PCR or rapid antigen test or identified by increase of titers against its nucleocapsid protein. When sera were taken in September 2023, the EG.5.1 variant had dominated circulation during the preceding months in Germany. Between our two sampling time points, BA.2.86 and JN.1 dominated circulation in Germany until April, when circulation of JN.1 descendants, mainly KP.3 and KP.3.1.1, increased rapidly (https://public.data.rki.de/t/public/views/IGS_Dashboard/Dashboard Variants?%3Aembed=y&%3AisGuestRedirectFromVizportal=y). XEC was first detected at the end of June (https://public.data.rki.de/t/public/views/IGS_Dashboard/DashboardVar iants?%3Aembed=y&%3AisGuestRedirectFromVizportal=y). We assume no XEC contact occurred in our cohort, as infections were either sequenced or infections took place in July or earlier when XEC was barely detected.
## Serum neutralization in September 2024 compared to September 2023
Neutralization titers of all sera were measured by PRNTs (Fig. 1A through C). In 2023 sera, titers were highest against the wild-type B.1 followed by BA.2 and BA.5 variants. Titers were substantially lower against concurrent EG.5.1 and JN.1 and future KP.3.1.1 and XEC. Of 58 individuals, 19 and 14 had detectable titers against KP.3.1.1 and XEC, respectively, even though these variants had not circulated in Germany when the sera were taken. In 2024, titer magnitude against B.1 was similar, while titers against other variants were generally higher than in 2023, with a broadly similar reactivity profile. An increased number of individuals had detectable titers against KP.3.1.1 and XEC (39 and 38, respectively). The largest fold increase in titers between 2023 and 2024 was observed for XEC, KP.3.1.1, and EG.5.1.
## Changes of serum neutralization depending on exposure
Of the 58 individuals included in this study, 41 were exposed to SARS-CoV-2 antigens at least once between September 2023 and September 2024, through vaccination (n = 2), infection (n = 28), or both (n = 11). In the remaining 17 individuals, no exposure was reported or detected by an increase of anti-SARS-CoV-2 nucleocapsid antibody titers (Table 1). When analyzing the individuals based on exposure during the study period, we find similar or slightly decreased titers in September 2024 compared to September 2023 for the individuals without an exposure (Fig. 1D andE), in line with waning antibodies over time. We find increased titers for the individuals that were exposed to SARS-CoV-2, with titers increasing the most for KP.3.1.1 and XEC, and only a limited increase in titers for B.1 (Fig. 1E andF).
## Impact of the variant exposure on serum neutralization
To determine the impact of the variant an individual was exposed to, we split our 2024 cohort according to the most recent variant each person had encountered (Fig. 1G through L). Fifteen individuals were exposed to the XBB/EG.5.1 group of antigenically related viruses (referred to as EG.5.1 contact group), 11 to BA.2.86 and JN.1 sublineages (JN.1 contact group), and 9 to KP.3 and its descendants (KP.3 contact group) (Table 1). None of the individuals had contact with XEC. We found that where the most recent antigen contact was with the EG.5.1 variant, titers against EG.5.1, JN.1, KP.3.1.1, and XEC were boosted similarly, with lesser titer increase against BA.5, BA.2, and B.1 (Fig. 1J). The individuals with JN.1 as the most recent contact showed similar fold change against the EG.5.1, JN.1, KP.3.1.1, and XEC variants, and no evidence for an increase in titers for B.1, BA.2, and BA.5 (Fig. 1K). In contrast, in the KP.3 contact group, the highest fold increase was for the homologous KP.3.1.1 variant (Fig. 1L). Reactivity against BA.2, BA.5, JN.1, and XEC also increased, while reactivity against B.1 did not. Titers and fold increase in the KP.3 contact group was highest, consistent with these individuals having been exposed most recently.
## DISCUSSION
Monitoring group-level neutralizing antibody activity against SARS-CoV-2 is important for assessing population susceptibility and the need for vaccine updates. Previous vaccine or virus exposure may change reactivity to certain substitutions and hence affect individual susceptibility to recent virus variants (3). The extent to which immu nological imprinting hampers the ability of neutralizing antibodies to adapt to recent variants is currently under debate. For instance, a recent study found XBB.1.5 monova lent vaccine elicited predominantly cross-reactive antibodies on the basis of previous B.1-type vaccination, as opposed to XBB.1.5-specific antibodies (4). However, others found that a subgroup of subjects developed XBB.1.5-specific neutralizing antibodies in a similar setting (5).
Many studies stratify cohorts according to vaccine exposure, often discriminating between groups with or without booster vaccinations subsequent to first course immunization, and often recording concomitant natural exposures without reference to variant. We focused on a cohort with age-and time-typical vaccination history that is precisely stratified for natural infection with a relevant panel of most recent viruses. Focusing on the most recent evolution of immunity during September 2023 to September 2024, we found increased group-level neutralization against all variants and an overall highest fold increase against the most recent variants KP.3.1.1 and XEC. Since none of the individuals were exposed to the then upcoming variant XEC, and only nine were exposed to concurrent KP.3.1.1, this indicates the induction of cross-neutraliz ing antibodies to descendant variants by exposure to immediate precursors of these variants. This is in accordance with recent findings showing that existing B memory cells elicited by older vaccines can adjust their specificity toward newly evolving variants, as shown by the example of Omicron BA.5 exposure that improved group-level immunity toward subsequent HK.3 and JN.1 strains (6). Another study found that natural exposure with recent Omicron strains, particularly JN.1, in multiply-vaccinated subjects induced a broadened response to all variants up to JN.1 (subsequent variants were not assessed) (7). Despite the clear existence of immunological imprinting in these studies and ours, the collective results suggest an immunity-updating effect conferred by recent natural exposure to circulating viruses, affecting these viruses and their descendants.
Generally low titers and high fold increase against KP.3.1.1 and XEC variants indicates an antigenic difference of these two variants compared to previously circulating variants, though we find no evidence of additional escape of XEC compared to KP.3.1.1. This is consistent with other studies assessing neutralizing antibodies in sera from XBB.1.5 and JN.1 exposed individuals using full virus (8) or pseudovirus neutralization assays (1,(9)(10)(11). In contrast, others have shown a slightly stronger immune evasion of XEC compared to KP.3.1.1 in individuals with JN.1 exposure (12,13). In sera from KP.3.1.1 and KP.3.3 breakthrough infections, two studies describe a slightly enhanced immune escape of XEC compared to KP.3.1.1 (1,12). In our KP.3 exposure sub-group, we also see slightly reduced titers against XEC compared to KP.3.1.1, highlighting the impact of the infecting variant on the reactivity profile of sera. When analyzing serum reactivity according to the variant of exposure, we found that sera from individuals exposed to JN.1 had on average lower titer increase, in particular against B.1, BA.2, and BA.5 compared to individuals exposed to EG.5.1 and KP.3.1.1. Broader reactivity after EG.5.1 compared to JN.1 infection is also supported by hamster sera (14). However, higher reactivity may also reflect more recent antigen contacts, as possibly indicated by the relatively higher reactivity after KP.3.1.1 infection.
Our study has several limitations. The overall age composition might not quantita tively reflect older populations with underlying disease and a lesser degree of immu nological adaptability to novel variants. Also, while neutralizing antibody activity is considered a valid surrogate of variant-specific immunity, cellular immune functions are not affected by viral antigenic drift to the same degree and their contribution to group-level immunity is not considered here. While evaluation of neutralization at specific timepoints provides valuable insights into population immunity, it is less informative regarding the impact of the exposure to specific variants, as effects might be masked by waning of antibodies over time. The small cohort size and the considerable number of individuals with multiple variant contact did not allow us to evaluate single contact categories. In addition, the occurrence of unrecognized infections detected by increases of anti-SARS-CoV-2 nucleocapsid antibody levels and the lack of sequencing data for some of the reported infections restricted the exposure classification and might have led to misclassification. Despite the limitations, our results provide a comprehensive overview of the evolution of neutralization titers in a well-characterized adult cohort in Berlin, Germany.
## MATERIALS AND METHODS
## Subjects and samples
Serum samples were collected from the same 58 subjects in September/October 2023 (09/08-10/13) and August/September 2024 (08/29-09/26). SARS-CoV-2 vaccina tion histories and anamnestic histories of any SARS-CoV-2 infection in the preceding year were recorded upon both visits. Each subject additionally submitted self-sampled respiratory swab specimens for broad-range (RT-)PCR testing during every RTI episode in the observation period (September 2023 to September 2024). To identify asympto matic or subclinical virus infections, antibody levels against the SARS-CoV-2 nucleocapsid protein were monitored every 3 months in every subject during the observation period using the Elecsys anti-SARS-CoV-2 test (Roche, Mannheim, Germany) (Table S1).
The variant an individual was infected with was assigned based on the genome sequence recovered from nasopharyngeal swabs in voluntary testing, if available, or according to predominance of the respective variant at the time of infection in cases of subclinical exposure. In times of codominance of multiple variants, no variant assign ment was made.
Exposures (vaccinations and infections) were categorized according to antigenic similarity as seen in the clustering of variants in SARS-CoV-2 antigenic maps (15,16): the EG.5.1 exposure group includes contact with XBB and EG.5.1 sublineages, the JN.1 exposure group includes BA.2.86 and JN.1 sublineage contact, and the KP.3 exposure group includes KP.3 descendants like KP.3.1.1 but not XEC. Individuals were classified according to their most recent antigen contact. Hence, some individuals of the JN.1 and KP.3 contact groups had an additional recorded exposure to earlier variants (EG.5.1 or JN.1, respectively).
Prior to testing, serum samples were heat-inactivated at 56°C for 30 min.
## Cell culture and viruses
Vero E6 cells expressing the transmembrane serine protease TMPRSS2 (Vero E6/ TMPRSS2, NIBSC 100978) were grown in Dulbecco's modified Eagle's medium (DMEM, Gibco, Darmstadt, Germany) supplemented with 10% fetal calf serum (Gibco) and 1.0 mg/mL geneticin (G418, Gibco). The cells were incubated at 37°C in a 5% CO 2 atmosphere. For the experimental trials, the cells were seeded without G418. Cells tested negative for Simian virus 5 and mycoplasma contaminations. Omicron KP.
## Plaque reduction neutralization tests
The neutralizing activity of each serum against the different virus strains was determined by plaque reduction neutralization test (PRNT), with small alterations compared to reference 17. In short, Vero E6/TMPRSS2 cells (1.6 × 10 5 cells/well) were seeded in 24-well plates and incubated for ~24 h. Human sera were serially diluted in OptiPro medium (Gibco) and mixed with medium containing 100 plaque-forming units of the respective virus, incubated at 37°C for 1 h, and then added to the Vero E6/TMPRSS2 cells in duplicate. After a further hour at 37°C, supernatants were discarded, and the cells washed once with phoshpate-buffered saline (PBS) and supplemented with 1.2% Avicel solution in DMEM. After 2 (B.1) or 3 days (all other viruses) at 37°C, the supernatants were removed, and the plates were fixed using a 6% formaldehyde/PBS solution and stained with crystal violet. Plaques were counted for each well or up to two dilutions without plaque reduction, and additional wells were treated as being equal to the seeding dose. If endpoint titers were not reached, titrations using additional serum dilutions were performed and included in the analysis.
## Analysis of neutralization titers
Titers were determined as the dilution where 50% of plaques were neutralized using the "neutcurve" package (version 2.1.0, https://jbloomlab.github.io/neutcurve/) in Python (version 3.12.4, https://www.python.org/), constraining the lower end of the neutraliza tion curve at zero. For the maximal number of plaques for one set of titrations, the average of the number of plaques in the virus control (incubated without serum) or the average of all dilutions with plaque counts above the virus control were used if multiple dilutions had higher plaque counts than the virus control. Geometric mean titers (GMTs) and fold changes were estimated using the gmt and log2diff functions, respectively, in the "titertools" package (version 0.0.0.9003, https://github.com/shwilks/ titertools) in R (version 4.4.1, https://www.r-project.org/), as described in reference 3. We used the following parameters: ci_method="HDI" (highest density interval), ci_level = 0.95, dilution_stepsize = 0, the prior for the mean was a normal distribution with mean of 0 and standard deviation of 100, the prior for the standard deviation was an inverse gamma distribution with a shape parameter of 2 and a scale parameter of 0.75.
## References
1. Kaku, Okumura, Kawakubo et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium, Ito J, Sato K. 2024. Virological characteristics of the SARS-CoV-2 XEC variant" *Lancet Infect Dis*
2. Jeworowski, Mühlemann, Walper et al. (2023) "Humoral immune escape by current SARS-CoV-2 variants BA.2.86 and JN" *Euro Surveill*
3. Wilks, Mühlemann, Shen et al. (2023) "Mapping SARS-CoV-2 antigenic relationships and serological responses" *Science*
4. Carreño, Lerman, Singh et al. "PVI study group. 2025. XBB.1.5 monovalent vaccine induces lasting cross-reactive responses to SARS-CoV-2 variants such as HV.1 and JN.1, as well as SARS-CoV-1, but elicits limited XBB.1.5 specific antibodies"
5. Pušnik, Monzon-Posadas, Osypchuk et al. (2024) "Effect of XBB.1.5-adapted booster vaccination on the imprinting of SARS-CoV-2 immunity" *NPJ Vaccines*
6. Kotaki, Moriyama, Oishi et al. (2024) "Repeated Omicron exposures redirect SARS-CoV-2-specific memory B cell evolution toward the latest variants" *Sci Transl Med*
7. Suntronwong, Kanokudom, Duangchinda et al. (2025) "Neutralization of omicron subvariants and antigenic cartography following multiple COVID 19 vaccinations and repeated omicron non JN.1 or JN.1 infections" *Sci Rep*
8. Fossum, Vikse, Robertson et al. (2025) "Low levels of neutralizing antibodies against SARS-CoV-2 KP.3.1.1 and XEC in serum from seniors in May 2024" *Influenza Resp Viruses*
9. Wang, Guo, Mellis et al. (2025) "Antibody evasiveness of SARS-CoV-2 subvariants KP.3.1.1 and XEC" *Cell Rep*
10. Uriu, Kaku, Uwamino et al. (2025) "Antiviral humoral immunity induced by JN.1 monovalent mRNA vaccines against SARS-CoV-2 omicron subvariants including JN.1, KP.3.1.1, and XEC" *Lancet Infect Dis*
11. Li, Faraone, Hsu et al. (2025) "Role of glycosylation mutations at the Nterminal domain of SARS-CoV-2 XEC variant in immune evasion, cell-cell fusion, and spike stability" *J Virol*
12. Arora, Happle, Kempf et al. (2024) "Impact of JN.1 booster vaccination on neutralisation of SARS-CoV-2 variants KP.3.1.1 and XEC" *Lancet Infect Dis*
13. Liu, Yu, Jian et al. (2025) "Enhanced immune evasion of SARS-CoV-2 variants KP.3.1.1 and XEC through N-terminal domain mutations" *Lancet Infect Dis*
14. Wang, Bhushan, Paz et al. (2024) "Antigenic cartography using hamster sera identifies SARS-CoV-2 JN.1 evasion seen in human XBB.1.5 booster sera" *bioRxiv*
15. Rössler, Netzl, Lasrado et al. (2025) "Nonhuman primate antigenic cartography of SARS-CoV-2" *Cell Rep*
16. Jian, Wang, Yisimayi et al. (2025) "Evolving antibody response to SARS-CoV-2 antigenic shift from XBB to JN.1" *Nature*
17. Wölfel, Corman, Guggemos et al. (2020) "Virological assessment of hospitalized patients with COVID-2019" *Nature* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548071&blobtype=pdf | # The effect of preexisting antibodies from Tdap vaccination during pregnancy on infant antibody responses to the pertussis-containing vaccines
Chacriya Sereeyothin, Stephen Kerr, Yong Poovorawan, Kirsten Maertens, Nasamon Wanlapakorn
## Abstract
Tetanus-diphtheria-acellular pertussis (Tdap) vaccination during pregnancy protects newborns from pertussis in the early months of life. Previous research indicated that Tdap vaccination during pregnancy may blunt Bordetella pertussis (B. pertussis)-specific antibody responses in infants following acellular (aP) and whole cell pertussis (wP) vaccination. However, the effect of preexisting antibodies on infants' responses to diphtheria toxoid (DT), tetanus toxoid (TT) and B. pertussis antigens is less wellunderstood. This study aims to quantify the effect of preexisting anti-DT, anti-TT, and B. pertussis-specific antibody levels from Tdap vaccination during pregnancy on infants' post-primary and post-booster responses to aP-and wP-containing vaccines. This retrospective analysis utilized data collected from a randomized controlled trial (NCT02408926) between 2015 and 2018. Pregnant women received Tdap vaccination between 27 and 36 weeks gestation. Their term infants were randomized to receive either a pentavalent DTwP-HB-Hib (wP) or a hexavalent DTaP-HB-Hib-IPV (aP) vaccine. Preexisting immunity was defined as the levels of anti-DT, anti-TT, anti-pertussis toxin (PT), anti-filamentous hemagglutinin (FHA), and anti-pertactin (PRN) IgG at month 2 (prevaccination). Blood samples were collected at birth, month 2, month 7 (following primary series vaccination at 2, 4, 6 months), month 18 (pre-booster), and month 19 (post booster). A total of 132 aP-vaccinated and 123 wP-vaccinated children completed this study. High levels of pre-vaccination antibody levels correlated with lower geometric mean ratios (GMRs) at post-primary and post-booster following wP-and aP-containing vaccination. This effect was observed consistently across all vaccine antigens following primary and booster doses. The clinical significance of this observation requires further investigation.
## Introduction
Pertussis is a respiratory bacterial infection that can cause severe disease in infants who have yet to develop their own immune defenses. 1 Maternal immunization with pertussis-containing vaccines during pregnancy provides passive immunity to infants too young to be vaccinated and has proven to be more effective than the cocooning strategy, in which mothers and close contacts receive pertussis vaccination prior to delivery or during the immediate postpartum period. Maternal immunization has already demonstrated substantial benefits in reducing mortality and preventing infectious diseases in infants, not only for pertussis but also for tetanus, influenza, coronavirus disease 2019 (COVID- 19), and most recently, respiratory syncytial virus (RSV). 2,3 One key consideration for vaccination during pregnancy is that maternally derived, transplacentally transferred antibodies may inhibit the infant's antibody responses to their own vaccinations, a phenomenon called the immunological "blunting effect." 2,4 Previous studies have shown that maternal antibodies can interfere with tetanus toxoid (TT) immunization in infancy, although the magnitude of this effect is not fully characterized. 5,6 Nevertheless, following the full primary series of tetanus-containing vaccines, infants generally achieve comparable immune responses, regardless of baseline antibody levels.
Evidence regarding blunting in pertussis vaccination is more heterogenous. A study by Englund et al. showed a strong and significant negative association between pre-existing anti-pertussis toxin (PT) antibody levels and post-immunization antibody levels in infants receiving the diphtheria-tetanus-whole cell pertussis (DTwP) vaccine, whereas this effect was not observed among diphtheria-tetanus-acellular pertussis (DTaP) recipients. 6 The results varied for other vaccine antigens. In DTaP recipients, a negative effect was found for anti-filamentous hemagglutinin (FHA), anti-pertactin (PRN), and anti-diphtheria toxoid (DT) (p < .001), but not for anti-TT. In DTwP recipients, anti-FHA and anti-DT showed significant negative correlations between pre-and post-immunization antibody levels. 6 A recent study on DTaPvaccinated infants reported that higher levels of preexisting maternal antibodies, including anti-PT, were associated with lower antibody responses to DTaP vaccination compared to infants with lower preexisting immunity, except for tetanus. 7 Thus, there is mixed characterization of the blunting effect in the acellularcontaining and whole-cell pertussis containing vaccines, and this topic warrants further exploration.
While aP vaccines are predominantly used in high-income countries (HICs), wP vaccination remain the predominant vaccine in lower-middle-income countries (LMICs). 4 Notably, the number of LMICs that recommend Tdap vaccination during pregnancy and subsequently vaccinate infants with DTwP vaccines is increasing. 8 Therefore, a better understanding of how maternal Tdap vaccination influences infant immune responses to wP vaccination is critical to inform immunization policies globally.
This study aimed to quantify the effect of vaccine-induced preexisting anti-DT, anti-TT, and B. pertussisspecific antibody levels from maternal Tdap vaccination during pregnancy on infants' post-primary and post-booster responses following immunization with either aP-or wP-containing vaccines.
## Material and methods
We retrospectively analyzed data collected between 2015 and 2018 from a randomized control trial (RCT) (ClinicalTrials.gov NCT02408926), conducted at King Chulalongkorn Memorial Hospital in Bangkok, Thailand. In this trial, a cohort of 370 healthy pregnant women consented to Tdap vaccination (Boostrix®) between gestational age 27-36 weeks. Parents then gave informed consent to enroll their infants into the study. After delivery, full-term and late-preterm infants with a birth weight >2,500 g (n = 315) were randomized to receive either an aP-containing (DTaP-HB-Hib-IPV; Infanrix hexa®) or a wP-containing (DTwP-HB-Hib; Quinvaxem®) vaccine. The study was not blinded because infants in the wP group received an oral polio vaccine (OPV), while those in the aP group received inactivated polio vaccine (IPV) as part of the hexavalent vaccine. Details of inclusion and exclusion criteria have previously been described. [9][10][11] The primary objective of the original RCT study was to compare antibody levels of infants after priming and first booster with either wP-or aP-containing vaccines (i.e., end-point comparison between aP and wP groups). The sample size for the original RCT study was calculated from a significance level of 0.05 and power of 0.90 under an assumption that the geometric mean concentration of anti-PT IgG would be 20% less in the wP than the aP group. 9 A population of 130 infants in both arms was sufficient. By 19 months (one-month post-booster), serology data was collected from 132 aP and 123 wP-vaccinated infants (n = 255). Baseline characteristics including maternal age, infant gestational age, birthweight, and sex revealed no significant differences between the groups as previously described. 9 Infants in both the wP and aP vaccine groups received their primary vaccination series at 2, 4, and 6 months, and a booster dose at 18 months of age. Serology samples were collected at birth (cord data), 2 months (prior to first vaccination), 7 months (1 month post-primary series), 18 months (pre-booster), and 19 months (1 month post-booster). Antibodies against PT, FHA, PRN, DT, and TT were measured by enzyme-linked immunosorbent assay (EUROIMMUN, Lübeck, Germany), as previously described. 9,12 The study was approved by the Institutional Review Board at Chulalongkorn University (IRB no. 604/57, 2 April 2015) and the Ethical Committee of the University of Antwerp (IRB no. 14/49/511, 23 March 2015).
The objective of the present retrospective analysis was to quantify the effect of preexisting anti-DT, anti-TT, anti-PT, anti-FHA and anti-PRN IgG levels on infants' post-primary and post-booster responses (i.e., within-group comparisons relative to baseline at 2 months of age). Preexisting immunity was defined as the antibody concentration at 2 months of age, prior to administration of the first primary vaccination dose. Within each study group (aP and wP), infants were categorized into low and high baseline IgG groups for each antibody, based on cutoffs reported by Knuutila et al. (with the exception of anti-FHA, where adjustments were made due to differences in baseline IgG levels). 7 Because this analysis makes use of data generated in a previously published RCT, no additional sample size calculation was performed.
Comparisons of post-vaccination antibody levels between low and high baseline IgG groups were conducted using log-transformed IgG values and one-way ANOVA with Bonferroni's correction for multiple comparisons. In addition, infants were stratified into five quintiles based on their preexisting antibody levels, using cutoffs adapted from a previously published article 7 to ensure a roughly equal distribution of participants across categories. Geometric mean ratios (GMRs) at each time point were calculated using generalized estimating equations (GEEs), stratified by baseline immunity level. GMRs were expressed relative to the geometric mean antibody concentration at 2 months of age as the reference. IgG concentrations were log transformed and used as outcome variables, and model parameter estimates were back transformed to derive GMRs with 95% confidence interval (CI). Statistical significance was defined as P < .05. The analysis was performed using StataNow version 19.0 (StataCorp LLC, Texas, USA).
## Results
## Participant characteristics
A total of 370 pregnant women, recruited between April 2015 and September 2016, received Tdap vaccination during pregnancy (GA 27-36 weeks). Of these, 315 healthy late preterm and term infants (gestational age 36 weeks and above) met the eligibility criteria, and their parents consented to their continued participation in the study after birth. Infants were randomized to receive either Infanrix hexa (aP group; n = 156 term and 2 late preterm) or Quinvaxem (wP group; n = 155 term and 2 late preterm). Baseline characteristics revealed no significant differences between the groups, as previously described. 9 A total of 132 aP-vaccinated children and 123 wP-vaccinated children completed the study at 19 months of age. The trials flow diagram and the drop-out rate were previously published. 9
## Effect of pre-existing antibody levels on immune response to DTaP and DTwP-containing vaccine
Infants were classified into two groups (high vs. low baseline IgG) based on their IgG levels at 2 months of age (indicating preexisting antibody levels). This classification enabled evaluation of interference from preexisting antibodies on infant antibody responses following post-primary series vaccination (at 7 months of age) and post-booster vaccination (at 19 months of age). Cutoff values for high and low baselines varied for each antibody and were determined according to Knuutila et al., except for anti-FHA as previously described (anti-PT = 10 IU/mL, anti-FHA = 80 IU/mL, anti-PRN = 20 IU/mL, anti-DT = 0.1 IU/mL, anti-TT = 1.0 IU/mL; Knuutila's cutoff for anti-FHA was 10 IU/mL). 7 In the DTwP-vaccinated infant group, there was a trend of lower geometric mean concentrations (GMCs) of anti-PT, anti-DT, and anti-TT IgG at month 7 among the high baseline IgG group; however, this did not reach statistical significance (Table 1). By month 19, these differences disappeared, as both the low and high baseline groups achieved comparable GMCs for all antigens. Notably, the anti-FHA IgG GMCs achieved at month 7 were significantly higher among the infants with high baseline titers (P = .002).
In the DTaP-vaccinated group, there were significant differences in the GMCs of achieved anti-DT and anti-PRN IgG at month 7. Infants with low baseline titers had higher GMCs of anti-DT and anti-PRN IgG (P < .001 for anti-DT IgG and P = .002 for anti-PRN IgG) (Table 2). By month 19, these disparities disappeared, with both low and high baseline groups achieving comparable GMCs for all antigens.
To further explore the relationship, infants were stratified into five groups for each antigen based on the preexisting antibody levels at month 2 as shown in Figure 1. Each antibody was binned differently as previously described. 7 The number of infant samples for each quintile is described in the Supplementary Table . Overall, higher pre-vaccination antibody titers correlated with lower GMRs at both post-primary and Table 1. Geometric mean concentrations of the five antibodies in infants who received whole-cell pertussis vaccination.
Vaccines were administered at 2, 4, 6, and 18 months (after serum collection). The time point to classify infants as having high vs. low baseline IgG was based on their IgG levels at 2 months of age prior to the first pertussis-containing vaccines, which indicated preexisting antibody titers. Statistical analyses were conducted on the log-transformed values of IgG using one-way ANOVA with Bonferroni's correction for multiple comparisons. Given that the number of comparisons is 10, a p-value of less than .005 is considered statistically significant. post-booster timepoints following DTwP-and DTaP-containing vaccination. This effect was observed consistently across all vaccine antigens following primary and booster doses, and the effect was graded, such that infants who received more antibodies from their immunized mothers showed a lower fold change in GMRs following the primary and booster immunization.
In DTwP-vaccinated infants (Figures 1 and2), the effect of preexisting antibodies was most pronounced for anti-PT and anti-DT IgG. For anti-PT IgG at month 7, the lowest preexisting antibody group, with titers between 0 and 4.99 IU/mL, showed a 24.2-fold change (95% CI: 14.0-41.8), while the highest preexisting antibody group, with titers over 50 IU/mL, showed a drop below baseline levels, with only a 0.2-fold change (95% CI: 0.1-0.3) at month 7. At month 19, the lowest preexisting antibody group showed a 58.2-fold change (95% CI: 33.8-100.5) from baseline and the highest antibody group showed only a 1.5-fold change (95% CI: 0.9-2.5). Each preexisting antibody titer quintile followed in a linear pattern, with the second highest preexisting antibody group showing second highest fold changes, and so on. Conversely, the smallest differences in GMRs between each preexisting antibody quintile were observed for anti-TT IgG, where the range in GMR for the highest preexisting antibody quintile was a 0.4-fold change (95% CI: 0.3-0.5) post-primary and a 2.7-fold change (95% CI: 2.1-3.4) post-booster, and the lowest preexisting antibody quintile was a 3.7-fold change (95% CI: 2.6-5.2) post-primary and a 13.8-fold change (95% CI: 9.7-19.5) post-booster.
Trends observed in the DTaP-vaccinated group were similar to those described for the DTwP-vaccinated group (Figures 1 and3): higher preexisting antibody levels correlated with lower responses to post-primary and post-booster vaccinations across all antibodies tested. The effect was most pronounced for anti-PT and anti-DT IgG. For anti-PT IgG at month 7, the lowest preexisting antibody group, with titers between 0 and 4.99 IU/mL, showed a 34.0-fold change (95% CI: 21.7-53.1), while the highest preexisting antibody group, with titers over 50 IU/mL, showed a drop below baseline levels, with only a 0.5-fold change (95% CI: 0.3-0.9). Similar trends were observed at 19 months post-booster, with the lowest preexisting antibody group showing a 50.7-fold change (95% CI: 32.1-80.3) from baseline and the highest antibody group showing only a 0.9-fold change (95% CI: 0.5-1.7). Anti-DT GMRs ranged from 0.6 (95% CI: 0.3-1.0) to 40.1 (95% CI: 26.1-61.6) at month 7 and from 2.0 (95% CI: 1.2-3.3) to 96.6 (95% CI: 63.6-146.8) at month 19.
The smallest GMR differences between quintiles were seen in anti-TT IgG. Anti-TT IgG GMRs ranged from 0.5 (95% CI: 0.4-0.6) to 3.1 (95% CI: 2.1-4.4) at month 7 between the lowest and highest quintiles, and from 2.2 (95% CI: 1.7-2.9) to 15.0 (95% CI: 10.6-21.3) at month 19.
Across both vaccine groups, GMRs generally increased after the 19-month booster compared with the primary vaccination at month 7 for all quintiles, demonstrating effective boosting (Figures 2 and3). This was observed for all antigens, with the most consistent response seen for anti-TT IgG across all quintiles. However, in the wP group, booster vaccination did not significantly increase anti-FHA and anti-PRN IgG above baseline levels for infants in the three highest preexisting immunity level quintiles.
## Discussion
Understanding the factors that influence infant immune responses to vaccines is crucial for designing infant immunization schedules. One such factor is the presence of preexisting antibodies prior to vaccination. 13 In this study, we assessed the effect of maternally derived antibodies on infant responses to DT, PT and B. pertussis antigens in a randomized controlled trial where pregnant women received Tdap vaccination during pregnancy, and their infants were randomized to receive either DTaP-or DTwP-containing vaccines as their primary and first booster vaccinations. The previously published data from this cohort revealed that infants born to Tdap-vaccinated mothers had significantly higher geometric mean concentrations (GMCs) of all B. pertussis-specific antibodies in the aP group compared to the wP group after the primary immunization (P < .001). 9 At post-booster, anti-PT IgG levels were similar between groups, but the aP group maintained significantly higher titers of anti-FHA (P < .001) and anti-PRN (P < .001). 9 Inspired by the approach of a recent report by Knuutila et al., we reanalyzed this previously published data to gain a more nuanced understanding of the nature of the blunting effect. 7 In the current analysis, we found that preexisting antibodies at 2 months of age modulated infant antibody responses to DT, TT, and B. pertussis antigens after primary and booster vaccinations. In DTwP recipients, higher baseline antibody levels were associated with a non-significant trend toward lower ). Levels of preexisting anti-DT, anti-TT, anti-PT, and anti-PRN IgG were stratified into roughly proportional quintiles. For each quintile, the geometric mean ratios (GMRs) of antibodies tested at month 7 (one month after primary series vaccination) and at month 19 (postbooster) were compared to baseline levels (month 2). PT = pertussis toxin, FHA = filamentous hemagglutinin, PRN = pertactin, DT = diphtheria toxoid TT = tetanus toxoid.
antibody levels for anti-PT, anti-DT, and anti-TT. In contrast, in DTaP recipients, infants with low baseline titers achieved higher levels anti-DT and anti-PRN after the primary series, although this difference disappeared by month 19. When further categorizing preexisting antibodies into five bins and examining the geometric mean ratios, we observed that higher pre-vaccination antibody titers correlated with lower GMRs following both DTwP and DTaP vaccinations.
Our results align with earlier reports showing maternal antibody interference. Englund et al. demonstrated that preexisting anti-PT negatively affected DTwP responses, while in DTaP recipients, blunting was seen for anti-FHA, anti-PRN, and anti-DT, but not for anti-TT. More recently, a large meta-analysis involving over 7,360 infants who received DTaP or DTwP vaccines found that naturally acquired maternal antibodies interfered with infant responses to priming doses for 20 of 21 antigens. For B. pertussis-related antibodies, a two-fold increase in maternal antibodies was associated with an 11% reduction in post-vaccination antibody levels for anti-PT and anti-FHA, and a 22% reduction for anti-PRN. 14 For tetanus and diphtheria, reductions were 13% and 24%, respectively. 14 Maternal antibodies continued to influence responses to booster doses of acellular pertussis, inactivated polio, and diphtheria vaccines at 12-24 months of age. 14 Another meta-analysis examining the effect of preexisting antibodies in a cohort of Tdap-vaccinated pregnant women whose children received an aP-containing vaccine found that two-fold higher IgG levels of certain vaccine antigens prior to primary immunization were associated with 8-17% lower post-primary immunization levels. The most significant effect was observed for anti-DT IgG (17%), while the lowest effect was noted for anti-PT IgG (8%). 15 Although each study employed different analytical methods and involved immunized mothers and mothers with naturally acquired immunity, the same trend was observed: maternally derived antibodies to DT, TT, and B. pertussis interfered with both primary and booster infant immune responses.
Knuutila et al. further showed that high baseline titers in aP-vaccinated infants (Infanrix hexa administered in two doses at 3 and 5 months) born to mothers who either received or did not receive Tdap vaccination during pregnancy, were associated with reduced post-primary antibody responses, particularly for anti-DT, anti-PT, anti-FHA, and anti-PRN, but not anti-TT. 7 Similarly, in our infant cohort, which received a 3-dose series of aP-containing vaccines, significant differences were found between the low and high baseline titer groups only for anti-DT and anti-PRN IgG, with this effect disappearing after the booster. It is reassuring that anti-PT IgG levels were not affected in our infant cohort, as anti-PT is considered critical for protection against pertussis. Several studies have demonstrated that low levels of anti-PT correlate with susceptibility to pertussis, 16,17 and PT-only vaccines have proven efficacious in clinical studies and in baboon models of maternal vaccination during pregnancy. [18][19][20] When comparing our results to those of Knuutila et al., it is important to note that all participating mothers in our analysis received the Tdap booster, while in their study, one arm received the Tdap booster, and the other did not. Consequently, the range of preexisting antibodies within each quintile is considerably higher in our study. For instance, the highest baseline level quintile for anti-FHA IgG reported by Knuutila et al. is >50 IU/mL, whereas in our study, the second lowest baseline quintile is already at 30-79.99 IU/mL. Notably, trends indicating lower antibody responses in individuals with high preexisting immunity are similar in both studies.
While there was a negative effect of preexisting antibodies on the post-immunization levels for anti-DT IgG and, to a lesser extent, anti-TT IgG, over 99% of infants achieved seroprotective concentrations for both diphtheria and tetanus 1 month post booster. This is reassuring as it indicates that even with high levels of maternally derived antibodies, infants can mount adequate immune responses following the booster. 12 The emergence of new generation pertussis vaccines, specifically genetically detoxified acellular pertussis vaccine (aP gen ) and combined tetanus, reduced-dose diphtheria, and genetically detoxified pertussis vaccine (Tdap gen ), licensed in many countries for pregnant women, adolescents, and adults, including Thailand, adds further complexity. These vaccines are more immunogenic and induce higher antibody levels in mothers and neonates than the traditional chemically inactivated Tdap vaccine (Tdap chem ). When administered to pregnant women, Tdap gen vaccines induce high B. pertussis-specific antibodies that are passed to their infants, as evidenced by elevated levels of B. pertussis-specific antibodies in the cord blood. 21,22 Early data from a Thai cohort of wP-primed, Tdap gen -vaccinated pregnant women whose infants received a wP-containing vaccine suggests no major differences in infant antibody responses between infants born to mothers receiving Tdap gen or Tdap chem , though anti-PT IgG GMC at month 7 were the lowest at 9.8 IU/mL among infants born to mothers vaccinated with the highest dose of Tdap gen (5 µg PT gen and 5 µg TdaP5 gen ). 23 Whether preexisting antibody interference occurs in these contexts warrants further investigation.
The strength of this study is the longitudinal data collection in aP and wP-vaccinated infants covering their primary series and booster vaccinations. Limitations include reliance on quantitative antibody data without assessing qualitative and functional properties (e.g. antibody subclass distribution, avidity, glycosylation patterns, neutralizing activity, and the capacity to activate innate immune responses were not assessed). A randomized, controlled, double-blind, phase 4 trial of Tdap-IPV immunization during pregnancy demonstrated that while Tdap-IPV was associated with relative blunting of the immune response to the DTwP primary vaccination series, B. pertussis-specific antibody quality and memory B-cell responses were nevertheless preserved, highlighting the importance of these qualitative analyses. 24 Additionally, the plausible underlying mechanisms of the interference effect, such as epitope masking, neutralization, accelerated clearance of vaccine antigens, or inhibitory signaling via immune complexes, 2,4 were beyond the scope of this analysis. Further studies using systems serology approaches are needed to address these gaps in understanding the potential impact of maternal immunization on both the quality and the mechanisms of infant immune responses. Small subgroup sizes in quintile analyses also limited statistical power. Larger studies, including controls born to non-recently Tdap vaccinated mothers, are needed to compare naturally acquired versus vaccine-induced maternal antibodies. Lastly, although none of the children in this study developed pertussis, the cohort size was too limited to draw definitive conclusions about the clinical significance of antibody interference. Assessing the clinical significance of blunting in infants vaccinated with wP vaccines in Thailand remains challenging. This is partly because Tdap vaccination in pregnancy is not included in the Thailand Expanded Program on Immunization (EPI), resulting in very low maternal vaccine coverage. Additionally, limited surveillance programs and inadequate diagnostic resources for pertussis cases further hinder accurate assessment. These challenges underscore the need for a strengthened surveillance system to monitor the epidemiological burden of pertussis in Thailand.
In conclusion, the present study showed that high maternally derived antibody levels from Tdap vaccination during pregnancy are associated with reduced infant immune responses to both aP-and wPcontaining vaccines, particularly for DT, PRN and FHA. However, for pertussis, where no established immune correlates of protection exist, current data suggest that lower antibody levels in infants born to mothers vaccinated against pertussis during pregnancy may not be clinically significant, as no increased susceptibility to pertussis was detected in children born to Tdap-immunized mothers. 25,26 Nevertheless, long-term monitoring of vaccine-induced antibody quantity and quality, along with epidemiological studies and surveillance data, is important to determine if adjustments to immunization programs are warranted. These findings reinforce the overall benefit of maternal Tdap immunization, while highlighting the importance of continued monitoring of antibody quality, durability, and clinical outcomes. Our results provide valuable evidence for shaping maternal and infant immunization strategies, particularly in countries implementing or considering antenatal pertussis immunization.
## References
1. (2015) "Pertussis vaccines: WHO position paper" *Vaccine*
2. Quincer, Cranmer, Kamidani (2024) "Prenatal maternal immunization for infant protection: a review of the vaccines recommended, infant immunity and future research directions" *Pathogens*
3. Ke, Jones, Roper et al. (2023) "Use of the Pfizer respiratory syncytial virus vaccine during pregnancy for the prevention of respiratory syncytial virus-associated lower respiratory tract disease in infants: recommendations of the Advisory Committee on Immunization Practices -United States" *MMWR Morb Mortal Wkly Rep*
4. Saso, Kampmann (2025) "What is the impact of pertussis immunization in pregnancy on the quantity, quality and longevity of infant vaccine responses?: A review of the current evidence" *Pediatr Infect Dis J*
5. Sarvas, Kurikka, Seppälä et al. (1992) "Maternal antibodies partly inhibit an active antibody response to routine tetanus toxoid immunization in infants" *J Infect Dis*
6. Englund, Reed, Decker et al. (1995) "The effect of maternal antibody on the serologic response and the incidence of adverse reactions after primary immunization with acellular and whole-cell pertussis vaccines combined with diphtheria and tetanus toxoids" *Pediatrics*
7. Knuutila, Barkoff, Ivaska et al. (2023) "Effect of immunization during pregnancy and pre-existing immunity on diphtheria-tetanus-acellular pertussis vaccine responses in infants" *Emerg Microbes Infect*
8. Abu-Raya, Edwards (2020) "Interference with pertussis vaccination in infants after maternal pertussis vaccination" *Pediatrics*
9. Wanlapakorn, Maertens, Vongpunsawad et al. (2020) "Quantity and quality of antibodies after acellular versus whole-cell pertussis vaccines in infants born to mothers who received tetanus, diphtheria, and acellular pertussis vaccine during pregnancy: a randomized trial" *Clin Infect Dis*
10. Wanlapakorn, Maertens, Chaithongwongwatthana et al. (2018) "Assessing the reactogenicity of Tdap vaccine administered during pregnancy and antibodies to Bordetella pertussis antigens in maternal and cord sera of Thai women" *Vaccine*
11. Wanlapakorn, Sarawanangkoor, Srimuan et al. (2024) "Antibody persistence to diphtheria toxoid, tetanus toxoid, Bordetella pertussis antigens, and Haemophilus influenzae type b following primary and first booster with pentavalent versus hexavalent vaccines" *Hum Vaccin Immunother*
12. Wanlapakorn, Maertens, Thongmee et al. (2020) "Levels of antibodies specific to diphtheria toxoid, tetanus toxoid, and Haemophilus influenzae type b in healthy children born to Tdap-vaccinated mothers" *Vaccine*
13. Zimmermann, Curtis (2019) "Factors that influence the immune response to vaccination" *Clin Microbiol Rev*
14. Voysey, Kelly, Fanshawe et al. (2017) "The influence of maternally derived antibody and infant age at vaccination on infant vaccine responses: an individual participant meta-analysis" *JAMA Pediatr*
15. Abu-Raya, Maertens, Munoz et al. (2021) "Factors affecting antibody responses to immunizations in infants born to women immunized against pertussis in pregnancy and unimmunized women: individual-participant data meta-analysis" *Vaccine*
16. Taranger, Trollfors, Lagergård et al. (2000) "Correlation between pertussis toxin IgG antibodies in postvaccination sera and subsequent protection against pertussis" *J Infect Dis*
17. Storsaeter, Hallander, Gustafsson (1998) "Levels of anti-pertussis antibodies related to protection after household exposure to bordetella pertussis" *Vaccine*
18. Trollfors, Taranger, Lagergård et al. (1995) "A placebo-controlled trial of a pertussis-toxoid vaccine" *N Engl J Med*
19. Kapil, Papin, Wolf et al. (2018) "Maternal vaccination with a monocomponent pertussis toxoid vaccine is sufficient to protect infants in a baboon model of whooping cough" *J Infect Dis*
20. Gregg, Merkel (2019) "Pertussis toxin: a key component in pertussis vaccines?" *Toxins (Basel)*
21. Chaithongwongwatthana, Wijagkanalan, Wanlapakorn et al. (2024) "Transplacental transfer of maternal antibodies following immunization with recombinant pertussis vaccines during pregnancy: real-world evidence" *Int J Infect Dis*
22. Chokephaibulkit, Puthanakit, Chaithongwongwatthana et al. "Effective and safe transfer of maternal" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793436&blobtype=pdf | # P-2172. Real-World Use of Letermovir for Prophylaxis of Cytomegalovirus in High-Risk Kidney Transplant Recipients
Athena Matsikas, Srijana Jonchhe, Sapna Mehta, ; Fainareti Zervou
Background. Letermovir (LTM) is FDA approved for cytomegalovirus (CMV) prophylaxis (PPX) in high risk (D+/R-) kidney transplant recipients (KTR). Compared to valganciclovir (VGCV), LTM causes less myelosuppression and does not require renal dose adjustment. A dose of LTM costs over $300, creating operational challenges in LTM access. This study evaluates the efficacy of LTM in preventing CMV infection in D+/R-KTR, and highlights barriers to use in clinical practice. Methods. Single-center retrospective chart review of all CMV D+/R-KTR transplanted between January 1, 2024 and October 31, 2024. The primary outcome was efficacy of LTM in preventing CMV viremia. Secondary outcomes include incidence of medication access barriers including prior authorization (PA) and financial assistance (FA), and description of LTM course. Results. Fifty CMV D+/R-kidney transplants were performed in the study interval. Four patients were excluded (Figure 1), leaving 46 patients in the analysis. Most KTR (89.1%) received antithymocyte globulin induction, and de-novo belatacept based immunosuppression was common (41.3%). LTM PPX was effective, with no breakthrough viremia in 45 KTR (97.8%). One KTR (2.2%) transitioned to CMV treatment for a viral load (VL) of 223 IU/mL in the setting of bacterial infection. Half the cohort (50%) required PA for LTM and 12 KTR (26.1%) needed FA due to high copay. Due to delay in accessing LTM beyond post-operative day 4, 19 KTR (41.3%) received valganciclovir (VGCV) while waiting for PA or FA. Six patients (13%) never started LTM. Details of access barriers are listed in table 2. At time of analysis, 33 KTR (71.7%) had stopped LTM. Premature discontinuation of LTM was uncommon, occurring in 4 KTR (12.1%), details of LTM courses are listed in table 3.
Conclusion. LTM is an effective agent for CMV PPX in D+/R-KTR. However, LTM access remains challenging, with high PA and FA needs. These barriers required VGCV while access was pending. Broader data supporting the use of LTM for CMV PPX may help improve access and reduce operational barriers in the future.
Disclosures. All Authors: No reported disclosures Prior to specimen collection, participants were asked about ARI symptoms (cough, fever, congestion, runny nose, shortness of breath, sore throat, and wheezing) in the past 7 days. Specimens from participants with no ARI symptoms (i.e., asymptomatic) were included in the analysis. Differences between positive (detecting ≥1 virus) and negative specimens were assessed. We used multilevel logistic models to compare odds of viral detection adjusting for school-level and season. S1320 • OFID 2026:13 (Suppl 1) • Poster Abstracts |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793576&blobtype=pdf | # Abstract citation ID: ofaf695.2355 P-2192. The Burden of Respiratory Syncytial Virus Among Young Children in the United States is Not Well-Documented Across Settings: A Systematic Literature Review
Mina Suh, Naimisha Movva, Regulatory Affairs, Ruvim Izikson, William Via, Susan Pastula, Marina Amaral De Avila Machado, Thomas Shin, Christopher Rizzo
Methods. This SLR followed PRISMA guidelines and was pre-registered on PROSPERO (#CRD42024599190). Literature published from 2009-2024 were evaluated for outcomes including RSV and LRTI rates in outpatient, urgent care, or emergency department (ED); RSV and LRTI hospitalization rates; and RSV laboratory testing practice and patterns.
Results. This review identified 2085 records; based on the eligibility criteria, 101 studies were included (Figure 1). Of the 101 studies, 34 were prospective cohort, 62 were retrospective cohort, and the remaining 5 were of other designs (1 trial, 1 casecontrol, 3 cross-sectional surveys). 35 studies provided national-level data; 64 were on various states; 2 were of unknown geographical location. All rate numbers were heterogeneous. 8 studies reported rates in the outpatient or ED settings. In urgent care, no data were available. RSV/LRTI outpatient rates ranged 1.5 to 277.8 per 1000. In the ED, the rates ranged 10 to 84.6 per 1000. RSV hospitalization rates were reported in 26 studies, and the rates were highly variable. RSV laboratory testing patterns were reported in 7 studies with only 1 study providing outpatient data. Though limited, underestimation of RSV is indicated in the outpatient compared to the inpatient setting (testing rates: 69-77% vs. 70-100%, respectively).
Conclusion. This systematic review underscores the significant impact of RSV in US children 8 months through < 5 years of age in all healthcare settings. No data are available for urgent care, and data from outpatient and ED settings remain limited while hospital data are variable. Inconsistent testing and reporting practices may be contributing factors. Given the variable disease burden estimates, additional studies are essential to assess healthcare utilization and impacts in this population. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793492&blobtype=pdf | # Prognostic Model for Yellow Fever-Related Acute Liver Failure
Carolina Vieira, ; Bráulio, R Couto, Wanessa Clemente
Background. A 2017-2018 Yellow Fever (YF) outbreak in southeastern Brazil saw several cases develop acute liver failure (ALF), emphasizing the importance of prognostic tools. Current scoring systems guide treatment and liver transplant decisions, but their use in YF-related ALF is uncertain. This study presents the Y.E.L.L.O.W. Score (Yellow Fever End-stage Liver and Organ Worsening Score) for estimating death probability in YF patients with ALF.
Table 1 Logistic regression models for the Y.E.L.L.O.W. Score to predict death due to yellow fever. Logistic regression models for the Y.E.L.L.O.W. Score to predict death due to yellow fever. Results. Of 229 patients, 197 (86%) were male; median age was 47 (range: 18-82, SD: 13.4). Median time to hospitalization from symptom onset was 4 days (range: 0-40). Bleeding occurred in 41 patients (17.9%), and sepsis in 6 (2.6%). Transfusions were given to 64 (27.9%), and 26 (11.4%) needed dialysis. Multivariate analysis included: age, encephalopathy grade, hemoglobin, platelets, leukocytes, neutrophils, INR (International Normalized Ratio), AST (aspartate transaminase), direct bilirubin, creatinine, and lactate. Logistic regression identified 6 risk factors and 1 protective factor for YF death. To maintain consistency across time points (24, 36, and 48 hours post-admission), 7 variables were kept in models, even without significance (Table 1). Mortality probability is estimated from logistic regression models (QR code in Fig. 1). Fig. 2 shows spreadsheet examples for Y.E.L.L.O.W. Score mortality probability. All 3 models highly predict YF death (Fig. 3). |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792845&blobtype=pdf | # P-1798. Identifying Early Influenza Virus Infections in a Human Infection Model
Alana Brown, Jack Anderson, Nicholas O'grady, Clare Camilleri, Sepideh Naderi, Rachel Myers, ; Wilbur, Woodhouse Iii, Julie Steinbrink, Thomas Burke, Christopher Woods, Micah Mcclain
Background. Early, accurate detection of respiratory viral infections is critical for clinical and public health interventions. We previously developed a whole blood mRNA detection assay (HR-B/V) measuring host gene expression biomarkers in circulating leukocytes that accurately discriminates between viral and bacterial acute respiratory infections. Herein, we evaluated performance of these immune biomarkers for detection of viral infection at early timepoints following influenza exposure using controlled human infection models (CHIMs).
Methods. Serial samples from 30 symptomatic, infected subjects across five influenza (H1N1 and H3N2) CHIM studies were analyzed using the Franklin Integrated Sample Prep (ISP), a sample-to-answer real-time (RT) PCR platform with ∼1 hour run-time. Blood samples (from pre-inoculation through resolution of disease) were loaded into ISP cartridges for RNA extraction and multiplex RT-PCR. Cycle threshold values were normalized, and gene expression levels were compared to clinical and virologic outcomes.
Results. Timing of symptom onset after inoculation was variable, but occurred approximately 50 hours after viral exposure, with peak symptoms occurring on average at ∼90 hours. Expression of genes depicting canonical antiviral immune responses was seen as early as 21 hours after inoculation. Most subjects (29/30) exhibited upregulation of ≥ 1 antiviral response gene at or before 36 hours post inoculation, and 73% (22/30) had ≥ 1 antiviral response genes exhibit increased expression prior to symptom onset. A viral model built from these genes correctly diagnosed infection in 29/30 subjects at ≥ 1 acute timepoint.
Conclusion. A host response-based mRNA assay targeting genes involved in canonical antiviral innate immune responses and measured on a rapid, field-deployable platform, shows promise for detection of early, even pre-symptomatic influenza infection.
Disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792677&blobtype=pdf | # other known risk factors, much of the risk remains unexplained. Latent cytomegalovirus (CMV) has been hypothesized to affect lung function through effects on natural killer cells, systemic inflammation, and direct effects on lung tissue. Prior analyses have been performed in populations with limited racial/ethnic, biologic sex, and/or geographic representation. The National Health and Nutritional Examination Survey (NHANES III) was a population-based sample from the United States. 1 /forced vital capacity (FVC) < 0.7]; 2) FEV 1 /FVC (continuous), and FEV 1 (continuous). Binary outcomes were analyzed using logistic regression and continuous outcomes were analyzed using linear regression. All analyses used robust variance estimation for confidence intervals and P-values and were adjusted for important covariates listed in the tables below
Mario Torres, Joanna Nelson, Koray Demir, Alex Zimmet, Aruna Subramanian, Thomas Dieringer
Background. Cytomegalovirus (CMV) is a significant complication in lung transplant recipients (LTRs). While valganciclovir is commonly used for prophylaxis, its use is often limited by toxicity. Letermovir (LET) has shown promise in renal and hematopoietic stem cell transplant recipients, but data in LTRs are limited. Here, we report on the safety and effectiveness of letermovir for CMV prophylaxis in lung transplant recipients.
## Table 1 Demographics
## Table 2
Results. A total of 43 lung transplant recipients who received letermovir prophylaxis were included. The median age was 56 years, and 62.8% were male. CMV serostatus was D+/R-in 39.5%, D+/R+ in 44.2%, D-/R+ in 9.3%, and D-/R-in 7.0% (Table 1). LET was started in 83.3% (N=35) due to valganciclovir-associated leukopenia. The standard dose of letermovir was 480 mg.
CMV reactivation was defined as CMV DNA detected above institutional lab threshold ( >135 copies/mL) and occurred in 2.3% (N=1), attributed to a dose reduction from 480 mg to 240 mg daily, due to drug interaction with cyclosporine.
LET was started at a median of 317 days post-transplant (IQR 64-495), with median duration of prophylaxis being 378 days (IQR 100-526) (Figure 1). LET was stopped in 14% (N=6), primarily due to cost or insurance issues. No discontinuations were due to adverse effects.
Mortality while on LET occurred in 20.9% (N=9); however, none of the deaths were attributed to CMV infection. (Table 2) Conclusion. Letermovir prophylaxis was associated with a low incidence of clinically significant CMV reactivation or infection in LTRs, along with a favorable safety profile. These findings support its potential as a viable alternative to traditional CMV prophylaxis in this population. However, further studies are warranted to confirm its long-term effectiveness and broader applicability.
Disclosures. Aruna Subramanian, MD, Gilead: Grant/Research Support| Moderna: Grant/Research Support |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792338&blobtype=pdf | # P-1805. Novel Cytomegalovirus Mutations in Immunocompromised Patients: Genotypic Insights and Clinical Correlation
Abdulrahman Alsweed, Madain Alsanea, Fatimah Alhamlan, Ahmed Alqahtani, Reem Almaghrabi
Background. Cytomegalovirus (CMV) is a substantial cause of morbidity in immunocompromised populations, particularly among solid organ and bone marrow transplant recipients. The emergence of novel mutations, particularly within conserved genomic regions, presents challenges for interpreting genotypic resistance and guiding therapy. The aim is to support the correlation of clinical variability with known genotypic resistance, enhance the understanding of viral behavior among different host factors that play the utmost role in managing CMV infection, and investigate the potential impact of the novel mutations.
Poster Abstracts • OFID 2026:13 (Suppl 1) • S1113 Novel CMV mutations found in UL54, UL97, and UL56 aligned along conserved regions The figure illustrates conserved regions among UL54, UL97, and UL56 genes. The novel mutations marked in red align along conserved regions.
## Cases summary with novel CMV mutations
The table shows patients with novel CMV mutations, their background, clinical antiviral course, and outcomes. *Refractory CMV infection: Increase by > 1 log10 CMV DNA levels in blood or plasma after at least 2 weeks of appropriately dosed anti-CMV medication. Abbreviations: GCV: Ganciclovir, VALG: Valganciclovir, FOS: Foscarnet, R: recipient of transplanted organ/bone marrow, D: donor. Reported mutations are natural polymorphisms except for marked mutations; bold indicates a known drug resistance mutation; red indicates a novel mutation. ^IC50 for GCV was reported in the literature as 3.1 (intermediate level)
Methods. Fifty-one CMV-positive clinical specimens from immunocompromised patients were subjected to UL97, UL54, and UL56 sequencing. Variants were aligned to reference genomes (Merlin strain) and assessed via BLAST for novelty and conservation. Clinical outcomes, antiviral regimens, and virologic responses were reviewed retrospectively.
Results. Thirteen patients exhibited previously unreported mutations in UL54, UL97, and UL56. Although four cases met the criteria for refractory DNAemia, some have responded favorably to first-line antivirals. Among the mutations analyzed, G579C in UL97 and A835T in UL54 were identified in conserved regions essential for substrate binding and polymerase activity, raising concern for functional relevance. One case harbored UL54 P522S, a known mutation associated with intermediate ganciclovir resistance. Two patients with profound immunosuppression and refractory DNAemia died, highlighting the critical influence of host factors on clinical outcomes.
Conclusion. CMV drug resistance mutations (DRMs) must be analyzed cautiously, as host response can be the main determining factor for DNAemia clearance. Early and specific DRM reporting is crucial, especially in immunocompromised hosts; hence, genotyping is the best modality. However, we suggest interpreting these findings according to the clinical response and reported known recombinant phenotypic testing (EC50/IC50). The additional benefit of classification for DRMs to low, intermediate, or high can aid physicians in deciding about medication switch versus dose adjustment of ongoing antiviral therapy, especially in the initial response to treatment.
Disclosures. All Authors: No reported disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12700697&blobtype=pdf | Vincent Soler
## Abstract
Could a macular telangiectasia type 2-like phenotype characterize retinal JC virus infection?We read with great interest the recent correspondence article by Hupin et al.[1] describing a case with ocular involvement of JC virus (JCV) infection. We very recently reported a case of JCV-related retinopathy for an individual with common variable immunodeficiency [2]. Clinically, our case presented with rapid progression of unilateral retinopathy involving all retinal layers [2]. JCV was detected exclusively in ocular samplesspecifically in the aqueous humor, vitreous, and retinaby PCR and positive immunodetection of the SV40 antibody (which cross-reacts with JCPyV). Positivity was predominantly within the retinal ganglion cell layer, but was also detected in some nuclei of cells within the retinal inner nuclear layer. All other biological samples, including blood, urine, and cerebrospinal fluid, tested negative for the virus.
Regarding the diagnosis proposed in Hupin et al. [1], we agree that some multimodal imaging features are consistent with macular telangiectasia type 2 (MacTel 2). However, contrary to the authors' suggestion, we believe that the angiographic finding of macular fluorescein leakage is not incompatible with MacTel 2. Indeed, macular fluorescein leakage has been described as a relatively common feature of MacTel 2 [3]. Concerning macular optical coherence tomography (OCT) findings, Hupin et al. [1] report asymptomatic outer retinal layer abnormalities 11 months after infection that do not evoke a MacTel 2 phenotype [3]. Briefly, features included unilateral widening and blurring of the interdigitation and ellipsoid zones and bilateral hyperreflectivity within the outer nuclear layer. Two-and-a-half years later, macular thickening was observed; again, not typical of MacTel 2 [4]. Subsequently, the authors describe bilateral retinal thinning with the presence of low reflective spaces. Interestingly and similarly to our reported case [2], these OCT abnormalities involved all retinal layers for the case reported by Hupin et al. [1]. On the other hand, the imaging features differed: a MacTel 2-like phenotype for Hupin et al. [1], versus diffuse retinal atrophy for our case [2]. A retinal biopsy was absent from Hupin et al. [1] and would have been helpful for confirming this rare manifestation of JCV infection in the retina. This means for now we can only speculate that ocular involvement of JCV could affect the retina and could manifest via different clinical presentations.
From a pathophysiological standpoint, MacTel 2 is thought to arise from early M€ uller cell dysfunction [5]. M€ uller cells reside in the retinal inner nuclear layer [6]. In our report [2], as mentioned above, we demonstrate JCV expression within this same retinal layer without actually identifying the specific immunostained cells. We therefore support the hypothesis put forward by Hupin et al. [1] suggesting a potential pathogenic role of JCV in the retina. Indeed, while it is possible that MacTel 2 developed coincidentally alongside ocular JCV infection in the case reported by Hupin et al. [1], it is also plausible that JCV infection actually have contributed to the development of the MacTel 2-like phenotype. This MacTel 2-like phenotype could thus specifically represent a distinct clinical entity directly attributable to JCV infection.
Overall, these two reports combined [1,2] reinforce the concept that the eye may serve as both a potential viral reservoir for JCV and a target organ, and that ocular JCV invasion may affect the retina differently. Indeed, could these two distinct clinical presentations result from infection with different JCV variants? Future reports on ocular involvement of JCV, particularly including longterm multimodal imaging findings, will be essential for deciphering the different retinal features that develop, as well as when they develop and following infection with which specific JCV variant. In the meantime, in clinical practice we aim to investigate JCV infection systematically when confronted with atypical forms of progressive retinopathy.
## References
1. Hupin, Hoogewoud, Sallo et al. (2025) "Persistence of intraocular JC-virus associated with a MacTel phenotype" *AIDS*
2. Varenne, Vergnon, Lhomme et al. (2025) "JC virus-related retinopathy" *JAMA Ophthalmol*
3. Chew, Peto, Clemons et al. (2023) "Macular telangiectasia type 2: a classification system using multimodal imaging MacTel Project Report Number 10" *Ophthalmol Sci*
4. Kim, Chung, Oh et al. (2020) "Optical coherence tomographic features of macular telangiectasia type 2: Korean Macular Telangiectasia Type 2 Study" *Sci Rep*
5. Powner, Gillies, Zhu et al. (2013) "Loss of M€ uller's cells and photoreceptors in macular telangiectasia type 2" *Ophthalmology*
6. Arrigo, Cremona, Aragona et al. (2025) "M€ uller cells trophism and pathology as the next therapeutic targets for retinal diseases" *Prog Retin Eye Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12395798&blobtype=pdf | # E D I TO R I A L E X P R E S S I O N O F CO N C E R N
22:296 https://doi.org/10.1186/s12985-025-02894-5
Editorial Expression of Concern: Virol J 2, 63 (2005). h t t p s : / / d o i . o r g / 1 0 . 1 1 8 6 / 1 7 4 3 -4 2 2 X -2 -6 3.
The Editors-in-Chief are issuing an editorial expression of concern to alert readers to issues raised after the publication of this article. Specifically, there appear to be repeated elements within the bands and background of blots presented in Fig. 2. The Publisher was not able to contact the authors for an explanation. Readers are urged to take caution when interpreting the content and conclusions of this article.
The authors have not responded to correspondence from the publisher regarding this retraction. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12729333&blobtype=pdf | # Multipurpose Passive Surveillance of Bat-Borne Viruses in Hungary: Lyssaviruses and Filoviruses in Focus
Anna Szabó, Zsófia Lanszki, Gábor Kemenesi, Alexandra Nándori, Péter Malik, Krisztián Bányai, Henrik Károlyi, Ágnes Nagy, Endre Sós, Pavle Banović, Tamás Görföl
## Abstract
This study describes the establishment of a new passive surveillance system to monitor diseases in bats in Hungary, and reports the first scientific results obtained using this approach. We retrospectively screened bat specimens collected over six years to assess the presence of viruses with zoonotic potential in Europe. European bat lyssavirus type 1 was detected in three serotine bats, while all samples tested negative for Lloviu virus, the only filovirus known to circulate enzootically in Europe. These findings provide new insights into ongoing viral surveillance efforts and highlight the importance of systematic retrospective screening in European bat populations.
## 1. Introduction
Global anthropogenic changes resulting in biodiversity loss and ecosystem degradation contribute to the emergence and spread of diseases [1]. One Health is an integrated approach that seeks to balance human, animal and ecosystem health by combining traditional disease-based surveillance with monitoring the factors driving disease emergence, thereby enhancing the prevention and mitigation of spill-over events [2]. It is known that preventing these spill-over events is much more cost-effective than relying on response activities [3].
Bats are mammals with unique adaptations and attributes that play critical roles in ecosystems, making them a key focus of One Health research. These animals possess a highly specialized immune system that enables them to harbor diverse pathogens, including viruses, while typically remaining asymptomatic [4,5]. Several viral groups contain large numbers of bat-borne viruses, like the Paramyxoviridae and Coronaviridae families, while other groups exhibit much lower genetic diversity. A good example of the latter is the Filoviridae family, that contains few but very important members, like Ebola-and Marburgviruses. The above mentioned viruses caused many deadly human outbreaks worldwide, and this has led to a growing interest in the role of bats as natural reservoirs for zoonotic diseases, making them a vital link in understanding the intricate connections between animal, human, and environmental health within the One Health framework [6][7][8].
Among the most well-known viruses are lyssaviruses, a diverse group of neurotropic viruses that can cause rabies, an invariably fatal acute encephalitis in mammals. These negative single-stranded, non-segmented RNA viruses are members of the Rhabdoviridae family in the order Mononegavirales [9]. There are 18 described species in the genus Lyssavirus, according to the International Committee on Taxonomy of Viruses (ICTV). Lyssaviruses can be divided into three phylogroups. Phylogroup I contains Lyssavirus australis (ABLV), L. aravan (ARAV), L. bokeloh (BBLV), L. duvenhage (DUVV), L. hamburg (EBLV-1), L. helsinki (EBLV-2), L. gannoruwa (GBLV), L. irkut (IRKV), L. kotalahti (KBLV), L. khujand (KHUV), L. formosa (TWBLV) [10], and L. rabies (RABV) viruses, while L. lagos (LBV), L. mokola (MOKV), and L. shimoni (SHIBV) are members of phylogroup II [11]. Based on phylogenetic distance, the most genetically distinct lyssaviruses are L. ikoma (IKOV), L. lleida (LLEBV), and L. caucasicus (WCBV), forming the putative phylogroup III. Lyssaviruses can enter the body through multiple routes, with the most common being through a bite from an infected animal [12][13][14].
Once the clinical signs are present, there is an almost 100% fatality rate. Therefore, pre-and post-exposure vaccines and immune globulin treatments are the only options for disease prevention. Rabies vaccines have been shown to provide protection against lyssaviruses within Phylogroup I, and some vaccines may offer partial cross-protection against viruses belonging to Phylogroups II and III [15,16]. In most European countries, it is advised that most individuals who are professionally or recreationally exposed to bats undergo pre-exposure immunization against rabies [17].
Since EBLV was first isolated in Germany in 1954 [18] and Yugoslavia in 1955 [19], more than a thousand rabid bat cases have been reported, especially in the western part of Europe (e.g., [20][21][22]). EBLV-1 has been linked to Cnephaeus serotinus and C. isabellinus, EBLV-2 to Myotis daubentonii and M. dasycneme, while LLEBV and WCBV to Miniopterus schreibersii [23]. Most lyssavirus-infected bats have been discovered in Northern and Central Europe, specifically in the UK, Germany, Finland, and Denmark [20,[24][25][26]. This is probably due to the more extensive surveillance in these areas rather than significant differences in epidemiology [27], and thus, differences in sampling intensity should be considered when interpreting current data.
Lloviu virus (LLOV), the only known European member of the Filoviridae family, was first identified during large bat die-off events on the Iberian Peninsula in the early 2000s [28]. The virus re-emerged in Hungary in the 2010s and has since been detected in several European countries, including Italy [29][30][31]. Its host range, geographic distribution, ecological and epidemiological characteristics, and zoonotic potential remain unknown, making further research on this Ebola-related virus critically important.
Combined serological and virus detection methods have revealed that at least three lyssaviruses are in circulation in bat populations in Hungary; however, with molecular methods, only EBLV-1 has been detected from a single species, C. serotinus [32]. A wide-range serological study of Miniopterus schreibersii revealed that bats from Hungary were infected with WCBV and LLEBV [22]. There are 29 bat species in Hungary, and several of them are hosts for diverse viruses, e.g., filoviruses, adenoviruses, picornaviruses, caliciviruses, astroviruses, herpesviruses, and coronaviruses as revealed with targeted molecular screenings [32][33][34][35].
In the context of wildlife, passive surveillance involves virological testing of animals that have died as a result of injury or debilitation [36]. This method is of paramount importance, as some of the animals found may have been infected with viruses that are responsible for the disease signs. Clinically asymptomatic rabies infections have also been observed in bats, highlighting the importance of thorough surveillance [37,38]. Passive surveillance systems for bat viruses in Europe focus on detecting diseases, such as European bat lyssaviruses in naturally deceased or ill bats, crucial for understanding zoonotic risks. Countries like Germany and France have well-established passive surveillance programs, monitoring rabies-related viruses in bats since the late 20th century. For instance, Germany's surveillance has detected both EBLV-1 and EBLV-2, contributing significantly to phylogenetic studies of these pathogens [39]. Researchers in France noted an unusual bat mortality event and several EBLV cases in the frame of their passive surveillance system [40]. The UK employs a system that has tested over several thousand bats for EBLV since the 1980s, demonstrating the importance of ongoing surveillance to track viral distribution [41] and resulting in the identification of the first case of EBLV-1 in the country [26]. Importantly, findings emphasize the interconnectedness of wildlife health and human risk, supporting One Health principles. By identifying patterns of viral circulation, these systems enable early warnings for public health interventions, as recently observed in the Netherlands [42].
In Hungary, rabies-suspected bats were subjected to investigation at the National Reference Laboratory for Rabies of the National Food Chain Safety Office, Budapest. This system provided a safe protocol to prevent human cases but did not allow a widerange passive surveillance of bats and to search for other pathogens. Here we present our passive surveillance system of Hungarian bats and the first results regarding viral disease monitoring.
## 2. Materials and Methods
## 2.1. Samples
We examined bat carcasses from Hungary, Central Europe that had died due to various factors such as disease, human activities, and other threats. The specimens primarily came from the rescue centre of the Budapest Zoo and Botanical Garden, other rescue centres, national parks, or veterinary clinics, collected by various experts, bat researchers and conservationists between 2018 and 2024. Altogether 15 species comprising 208 individuals were included in the study, serotine bat (Cnephaeus serotinus) (n = 26), Savi's pipistrelle (Hypsugo savii) (n = 12), Daubenton's bat (Myotis daubentonii) (n = 4), greater mouse-eared bat (M. myotis) (n = 4), whiskered bat (M. mystacinus) (n = 1), Natterer's bat (M. nattereri) (n = 1), lesser noctule (Nyctalus leisleri) (n = 1), common noctule (N. noctula) (n = 43), Kuhl's pipistrelle (Pipistrellus kuhlii) (n = 94), common pipistrelle (P. pipistrellus) (n = 3), Nathusius' pipistrelle (P. nathusii) (n = 3), grey long-eared bat (Plecotus austriacus) (n = 7), brown long-eared bat (P. auritus) (n = 2), lesser horseshoe bat (Rhinolophus hipposideros) (n = 1), parti-coloured bat (Vespertilio murinus) (n = 6) (Table 1). Until further processing, the bat carcasses were stored at -80 • C. As all bat species in Hungary are protected, the relevant authorities issued research and sample collection permits to the Hungarian Natural History Museum (PE-KTF/736-6/2017, PE-KTFO/329-16/2019, PE-KTFO/1568-18/2020, PE-KTFO/1403-3/2022). Bat species names were used according to the latest taxonomic reference and serotinus was placed in Cnephaeus instead of its former genus, Eptesicus [43].
$$Year/Habitat U U U/R U/R R R R U (R) U R (U) R (U) R U U U 2018 - - - - - - 1 - 1 - - - - - - 2 2019 - - - - 1 - - 1 1 - - - - - - 3 2020 1/1 3 1 3 - - - 3 2 - - - - - - 13 2021 2 - - - - 1 - 21 12 - 1 - 2 - 4 43 2022 6 5 1 - - - - 7 59 1 - - 3 1 1 84 2023 7/1 - - 1 - - - 6 12 2 1 1 1 - - 31 2024 1/1 1 1 - - - - - - - - - - - - 3 2018-2024 (no specific date) 9 3 1 - - - -5 7 - 1 1 1 - 1 29$$
## 2.2. Bat Necropsy
To ensure maximum safety, the necropsy of the bats was carried out under BSL-2+ conditions. Accordingly, the personal protective equipment (PPE) included double gloves (with the inner layer sealed), laboratory scrubs and a laboratory coat worn on top.
The dissected organs were as follows: brain, lungs, heart, liver, spleen, kidneys, rectum, testicles and muscle. Dissections were performed in a consistent order, beginning with the brain and subsequently proceeding to the other organs, with sterile scissors and tweezers used throughout. To further reduce the likelihood of cross-contamination, instruments were exchanged after the brain dissection, before proceeding with the remaining organs. Each sample was divided and placed into two 1.5 mL Eppendorf tubes without any preservatives or snap freezing and stored directly at -80 • C until further processing. For the testicle samples, one testis was frozen and stored natively at -80 • C, while the other testis was fixed in 6% formaldehyde for subsequent research purposes. External and internal characteristics (e.g., sex, age, condition of organs), abnormalities visible on the animals (e.g., injuries, weak condition) were recorded before and during the necropsy.
## 2.3. Nucleic Acid Extraction and PCR Reactions
Phosphate-buffered saline (PBS, 300 µL) was added to approximately half of the brain and lung samples and were homogenized for three minutes using a TissueLyser LT (Qiagen, Hilden, Germany). The total RNA was extracted using Monarch Total RNA Miniprep Kit (New England Biolabs, Ipswich, MA, USA) and finally eluted in 40 µL of nuclease-free water.
Following nucleic acid extraction, the brain samples were screened with a universal RT-nPCR targeting a conserved region of the N (nucleoprotein) gene for the representatives of the genus Lyssavirus. First round of the PCR was used with the Luna Universal One-Step RT-qPCR Kit (New England Biolabs, Ipswich, MA, USA) with the reaction setup as follows: at 55 • C for 10 min, and 95 • C for 1 min, followed by 30 cycles of 95 • C for 10s, 53 • C for 30s, 60 • C for 1 min, finally 60 • C for 6 min, and 4 • C ∞. The following primer sets were used: GRAB1F and GRAB1R [44]. The second round was used with GoTaq G2 Flexi DNA Polymerase Kit (Promega, Madison, WI, USA) with endpoint PCR. The reaction setup for the second round: 94 • C for 2 min, followed by 30 cycles of 94 • C for 1 min, 53 • C for 1 min, and 72 • C for 1 min, finally 72 • C for 5 min, and 4 • C ∞. The following primers were used: GRAB2F and GRAB2R [44]. The PCR products were analysed by standard gel electrophoresis using a 1.5% agarose gel. In case of lyssavirus positivity, nucleic acid purification and PCR reaction were conducted on all the necropsied organs (heart, lungs, liver, kidneys, spleen, testicle), using the same method.
The lung samples were screened with a Lloviu virus-specific real-time RT-PCR system. The PCR was used with the qRT-PCR Brilliant III Probe Master Mix (Agilent Technologies, Santa Clara, CA, USA) with the reaction setup as follows: at 50 • C for 10 min, and 95 • C for 3 min, followed by 50 cycles of 95 • C for 5 s, 60 • C for 30 s [30].
## 2.4. Viral Enrichment and RNA Extraction
A viral enrichment process was used to increase the proportion of viral reads. We followed a modified version of the method published by Conceição-Neto et al. [45]. The PCR-positive homogenates were centrifuged at 16,000× g for 10 min. From the sample, 160 µL supernatant was filtered through a 0.45 µm Ultrafree-CL Centrifugal Filter (Merck Millipore, Burlington, MA, USA). 150 µL of the sample was treated for 2 h at 37 • C with a cocktail of 1 µL micrococcal nuclease (New England Biolabs, Ipswich, MA, USA), 2 µL of benzonase (Merck Millipore, Burlington, MA, USA), 4.5 µL Turbo DNase and 15.5 µL Turbo DNase Buffer (Invitrogen, Waltham, MA, USA), and extracted with the Direct-Zol RNA MiniPrep Kit (Zymo Research, Irvine, CA, USA).
## 2.5. Illumina Sequencing
RNA library was generated using the NEBNext Ultra II Directional RNA Library Prep for Illumina (New England Biolabs, Ipswich, MA, USA). Briefly, 10 ng of total RNA was used as input for the fragmentation step, and the cDNA generation was performed using random primers. Thereafter, the cDNA was end-prepped and adapter-ligated, then the library was amplified according to the manufacturer's instructions. The quality of the libraries was checked on an Agilent 4200 TapeStation System using D1000 Screen Tape (Agilent Technologies, Santa Clara, CA, USA), and the quantity was measured on Qubit 3.0 (Thermo Fisher Scientific, Waltham, MA, USA). Illumina sequencing was performed on a NovaSeq 6000 instrument (Illumina, San Diego, CA, USA) with 2 × 151 run configuration.
## 2.6. Phylogenetic Analysis
The Illumina sequencing reads were aligned to the closest complete rabies virus genomes in Geneious Prime v2024.0.4. After manual corrections, the complete genomes were obtained and deposited in the NCBI GenBank under PV454706-PV454708 accession numbers. They were aligned with other complete Lyssavirus genomes downloaded from the GenBank using MAFFT v7.505 software. IQ-TREE v2.3.6 [46] was used to find the best fitting substitution model (GTR + F + G4 and JC in case of EBLV-1 and Lyssavirus trees, respectively) and generate a Maximum Likelihood tree with 1000 bootstrap support. The tree was visualized using iTol v7.0 online [47].
## 3. Results
## 3.1. A Passive Surveillance System to Detect Bat-Borne Viruses
To date, no passive surveillance has been conducted in Hungary for the virological examination of deceased bats. Therefore, the newly established passive surveillance system in Hungary operates through two main laboratories: the Department of Virology at the Directorate of the Veterinary Diagnostic Laboratory of the National Food Chain Safety Office (=National Reference Laboratory for Rabies, located in Budapest, the capital) and the National Laboratory of Virology at the University of Pécs (situated in Pécs, southern Hungary) (Figure 1). Animal rescue centres, which receive a variety of animals including bats, are primary contributors of specimens. Another significant source of bat submissions come from bat researchers, who are often the first professionals to interact with these animals. In collaboration with bat rehabilitation centres, we organized workshops, prepared documentation, and provided training to educate researchers about the potential health hazards associated with bats. Veterinarians and national park staff, including rangers, were also integrated into the surveillance network due to their frequent encounters with injured or deceased bats.
It is essential to collect as much information as possible about the bats and the circumstances of their discovery. This information serves two critical purposes: it allows for timely contact with relevant contact persons in case a bat tests positive for rabies, it aids decision about where to send the bats for investigation, and it can also support further research efforts.
If a bat is suspected of being infected with the rabies virus, it is sent to the National Reference Laboratory for Rabies. Investigations at the National Reference Laboratory for Rabies are conducted promptly, particularly in cases involving potential human exposure. If rabies infection is not suspected, the bat is sent to the National Laboratory of Virology at the University of Pécs. Here, additional viral groups are investigated besides lyssaviruses.
## 3.2. Lyssavirus Survey
During passive surveillance of lyssaviruses in bats collected and tested in Hungary between 2018 and 2024, three out of 208 brain samples were positive for lyssavirus RNA. All positive samples were detected in serotine bats. We also tested the lungs, hearts, livers, spleens, and kidneys of two positive bats, as well as a testicle from a male bat, for the presence of lyssavirus. All the tested organs of the female bat were positive, except the spleen tissue. In the case of the male bat, all the tested organs, including the testicle, were positive for lyssavirus.
The (near) full-length genomic sequence of all three viral RNA positive samples could be determined from the Illumina sequencing outputs. Remapping of reads to the newly generated genomes resulted in mean coverages 23452.2X (1,650,328 reads), 76.1X (5810 reads), and 91.4X (7296 reads) in case of PC454706, PV454707, and PV454708, respectively. The finalized consensus genomes were as follows: PV454706, 11966 nt; PV454707, 11966 nt; PV454708, 11931 nt. The genomic structure of the three Hungarian sequences was similar to those of other EBLV-1a strains (N, 1356 nt, 452 aa; P, 897 nt, 299 aa; M, 609 nt, 203 aa; G, 1575 nt, 525 aa; L, 6384 nt, 2128 aa). The PV454708 sequence was incomplete, with a 23 nt gap in the N gene. The order and length of genes were conserved and virtually identical to the reference EBLV-1 strain, RV9, an German isolate from 1968 (GenBank #, NC_009527). No other viruses were found in the samples of these three bats.
The phylogenetic analysis of the complete genomes confirms the results obtained from the partial sequence fragments, indicating that the three viruses belong to the EBLV-1a phylogenetic cluster (Figure 2). The genomes showed the highest similarity to EBLV-1a viruses previously isolated from serotine bats in Hungary in 2011 and 2015 [32] and clustered with sequences of EBLV-1a retrieved from C. serotinus from Slovakia and Poland between 1990 and 2014. Two strains (GT2064 from 2020 and GT3847 from 2023) clustered with other Hungarian strains isolated in 2011 and 2015, while the third strain (GT4136 from 2024) clustered with EBLB-1a strains detected in 1990 and 2014 in Poland. The genome wide sequence homology values between the two sister branches were ~98.5%, or so. This pattern of tree topology suggests that EBLV-1a strains identified thus far in Hungary may have originated from diverse geographic location through distinct flyways of migrating bats.
We also created a phylogenetic tree that provides a clearer overview of the diversity across all lyssavirus species, highlighting the evolutionary relationship that define each major lineage (Figure 3). Our three sequences consistently clustered with EBLV-1, forming a well-supported subgroup that reflects their close genetic affinity to this lineage. Notably, their position also placed them near several geographically distinct viruses; DUVV, TWBLV and IRKV, suggesting shared ancestral origins while maintaining characteristic separation.
Although we did not receive exact collection localities for all bats, the habitat preference of all species is well documented, and most bat species from this study are linked to urban habitats. Thus, over 90% of those are representatives of partly (Nyctalus noctula, Myotis myotis, Rhinolophus hipposideros, Myotis daubentonii, Pipistrellus nathusii, Pipistrellus pipistrellus) or exclusively building-dweller bat species (Cnephaeus serotinus, Pipistrellus kuhlii, Hypsugo savii, Vespertilio murinus, Plecotus austriacus) where human-bat encounters are more probable.
## 3.3. Surveillance Data on EBLV-1a Positive Bats
On 4 June 2020, near Sukoró, Hungary, a female bat was pulled underwater by a frog. Serendipitously, pedestrians witnessed the incident and managed to rescue the bat. The animal was subsequently transferred to the Animal Rescue Centre of the Budapest Zoo and Botanical Garden for veterinary care. No clinical signs of rabies were observed for several months. The bat exhibited normal behaviour during autumn and winter. Hibernation was conducted under controlled conditions, with monthly monitoring and feeding. On 1 March 2021, its weight was recorded at 26.2 g, yet by 3 April 2021, it had significantly declined to 16 g. Despite continued veterinary care, the bat died on 12 April 2021, over ten months after its initial admission to the rehabilitation centre. The second case originated from 29 September 2023, in Budapest. The male serotine bat was discovered trapped inside a police station. The animal was promptly taken to a veterinary clinic for assessment and care. However, despite medical attention, the bat was found deceased the following day. The third serotine was found on 2 July 2024, in Budapest. The bat was weak and there was a haemorrhage on the right forearm. It was fed for four days, but on 6 July, it refused both food and water. It was euthanized on 8 July. No human contacts were reported, however, post-exposure rabies prophylaxis was recommended in the latter two cases, as the bats died shortly after being brought them into expert care.
## 3.4. Lloviu Virus Survey
In addition, for the lyssavirus analyses, lung tissue samples from all bats were included in the surveillance and were tested for the presence of Lloviu virus. All samples were negative, indicating no evidence of Lloviu virus in the examined bat species during the study. These findings are consistent with current knowledge that Lloviu virus has so far only been detected in Schreibers's bats (Miniopterus schreibersii) and has not been reported from other bat species.
## 4. Discussion
Lyssaviruses have caused rabies in humans for centuries, leading researchers into creating effective strategies to prevent infection. Passive surveillance has a major importance in studying viruses circulating in the bat fauna, and as most bats found are from urban areas, this especially applies for the study of bats close to humans. This method has been used successfully by several countries in Europe since the 1980s [20,41,48,49]. However, these surveillance programs often concentrate only on lyssaviruses and other viral groups are studied occasionally. Although no other viruses were detected in this study, we conducted a proof-of-concept investigation to identify potential new hosts of Lloviu virus. Lloviu virus has been associated with mortality events in Schreiber's bats, although not yet conclusively [28]. We believe that reporting negative findings is also important, as it contributes to a more complete understanding of viral ecology. Thus, the assumption that it may occur in animals collected during passive surveillance and provide useful samples for understanding the host spectrum outside of Schreiber's bats is a novel approach to Lloviu virus research.
Our main goal was to develop a new surveillance system based on the well-performing Hungarian rabies surveillance program, to be able to respond rapidly to rabies cases as well as to be able to study the bat viruses circulating in Hungary more deeply. Based on the One Health concept, which acknowledges that human health is strongly linked to the health of animals, plants and all its environment, the integrated work of different sectors and regions should rely on detection and response infrastructures [50]. During our research, we were able to work together in a collaborative way with many bat experts, rescue centres, national park directorates and veterinary clinics. This acquaintance and cooperation between our research facilities and these partners gives the system an unparalleled opportunity to effectively study the bat-borne viruses-including lyssaviruses and Lloviu virus-circulating in Hungarian bats. The passive surveillance system that we aimed to establish in Hungary is particularly valuable and important because it fills a critical gap that has not yet been addressed, providing a solution to a long-standing deficiency.
Based on our experience, recording detailed information about the circumstances in which a bat is found can be crucial, as it may indicate the presence of pathogens. This should also include information on the contact person, to be able to reach them in case of positivity. In our study, we were able to include 15 of the 29 bat species found in Hungary. Previous studies have shown that successful passive surveillance is dependent on recovering bats from a geographical range that reflects their natural distribution [9].
Raising awareness of effective rabies control measures, both among public health professionals and the public, to reduce the risk of human infection is a very important task of this system. Misconception or misinformation are a significant issue in the realm of the public's understanding of how bats pose a risk to humans, underscoring the necessity of enhanced risk communication strategies and avoiding negative framing of bats [51], while remaining vigilant in cases of bite exposure [42,52]. We also wanted to make suggestions to bat conservationists, chiropterologists, and handlers on the honest and accurate risks of zoonotic virus spillover from bats and how to prevent such cases. Cautious handling is one of the most crucial parts of dealing with these animals, where personal protective equipment, such as gloves, masks, and in special cases hazard suits can prevent unfortunate accidents. Furthermore, individuals at rescue centres and bat researchers should be vaccinated against rabies and this is also recommended for any other person who is working with bats regularly (e.g., veterinarians, biologists, and other professionals), and we additionally recommend periodic antibody titre testing to verify adequate post-vaccination immunity.
Although several viral groups are investigated within our passive surveillance system, our case study focused on the results of our lyssavirus and Lloviu virus survey. The lyssavirus findings are consistent with previous studies, as the serotine bat is the main host for EBLV-1 [20]. The virus is still present and circulating amongst these animals in Hungary, therefore it is crucial to continue the monitoring to gain a better understanding of the virus's prevalence. In the frame of our study, we have found three serotine bats that were infected with EBLV-1. The complete viral genomes were retrieved from the brains of these animals, all belonging to the EBLV-1a strain. This is in line with the other two Hungarian strains from rabid bats found earlier [32]. The new viruses were found to be closely related to those previously detected in Hungary, Poland, and Slovakia. These closely related lineages were also derived from Cnephaeus serotinus bats, comprising Polish isolates collected in 1990 and 2014, as well as a Slovak strain recovered in 2001. Long-term surveillance efforts targeting lyssaviruses in European bat populations have been ongoing since 1977, consistently revealing that Cnephaeus serotinus is the primary reservoir species. Throughout these surveillance programs, most detected infections across Europe have been attributed to European bat lyssavirus type 1 (EBLV-1), underscoring its widespread circulation and long-standing presence within serotine bat populations [24,37,[53][54][55]. Examination of the other organs of two positive bats revealed discrepancy in lyssavirus presence between the tested organs (spleen was positive in one animal and negative in the other). This may be the result of different tissue tropism within the host or other factors [56].
Our findings align with recent European surveillance data on lyssaviruses, highlighting their sporadic but widespread circulation across the region. In contrast, only limited surveillance data are currently available for filoviruses in European bats [30,57] underscoring the need for further surveillance. Tracking and monitoring human-bat interactions are essential components of the One Health approach, helping to predict and prevent potential outbreaks or pandemics particularly in Europe, where these interactions are frequent [52,55,58,59]. To achieve this goal, we aimed to establish a passive surveillance system for bat-borne viruses in Hungary, and to study the bat lyssavirus situation in the country with the help of the public and through the collaboration of healthcare and research institutions. Moreover, considering the results of our study, we hope this paper will serve as a recommendation for both the public and professionals working with bats (e.g., veterinarians, conservation biologists), emphasizing that caution is one of the most important aspects of handling these animals. A limitation of the study is that not all bat species present in Hungary could be included in the sampling, and the number of collected samples for each species fluctuated between years, which may influence the representativeness and comparability of the results.
## 5. Conclusions
Collectively, this study represents the implementation of a One Health approach surveillance system in a Central European country. The future objective is to expand the research on a broader scale, investigate the presence of additional viruses, or investigate emerging bat-related infectious diseases within the framework of the established passive surveillance system.
## References
1. Pfenning-Butterworth, Buckley, Drake et al. (2024) "Interconnecting global threats: Climate change, biodiversity loss, and infectious diseases" *Lancet Planet Health*
2. *OHHLEP*
3. Hayman, Adisasmito, Almuhairi et al. (2023) "Developing One Health surveillance systems" *One Health*
4. *OHHLEP*
5. Markotter, Mettenleiter, Adisasmito et al. (2023) "Prevention of zoonotic spillover: From relying on response to reducing the risk at source" *PLoS Pathog*
6. Irving, Ahn, Goh et al. (2021) "Lessons from the host defences of bats, a unique viral reservoir" *Nature*
7. Seal, Dharmarajan, Khan (2021) "Evolution of pathogen tolerance and emerging infections: A missing experimental paradigm"
8. Latinne, Nga, Long et al. (2023) "One Health Surveillance Highlights Circulation of Viruses with Zoonotic Potential in Bats, Pigs, and Humans in Viet Nam" *Viruses*
9. Mubareka, Amuasi, Banerjee et al. (2023) "Strengthening a One Health approach to emerging zoonoses"
10. Gonzalez, Hurtado-Monzón, O'krafka et al. "Studying bats using a One Health lens: Bridging the gap between bat virology and disease ecology" *J*
11. Harris, Brookes, Jones et al. (2006) "European bat lyssaviruses: Distribution, prevalence and implications for conservation" *Biol. Conserv*
12. Hu, Hsu, Lee et al. (2018) "Lyssavirus in Japanese Pipistrelle"
13. Badrane, Bahloul, Perrin et al. (2001) "Evidence of two Lyssavirus phylogroups with distinct pathogenicity and immunogenicity" *J. Virol*
14. Bell, Moore (1971) "Susceptibility of carnivora to rabies virus administered orally" *Am. J. Epidemiol*
15. Fisher, Streicker, Schnell (2018) "The spread and evolution of rabies virus: Conquering new frontiers" *Nat. Rev. Microbiol*
16. Ramsden, Johnston (1975) "Studies on the oral infectivity of rabies virus in carnivora" *J. Wildl. Dis*
17. Ben Hamed, Myers, Chandwani et al. (1107) "Toward the Development of a Pan-Lyssavirus Vaccine" *Viruses*
18. Fisher, Lowe, Smith et al. "Lyssavirus Vaccine with a Chimeric Glycoprotein Protects across Phylogroups"
19. Mohr, Tollwut
20. (1957) *Med. Klin*
21. Jelesic, Nikolic (1956) "Isolation of rabies virus from insectivorous bats in Yugoslavia" *Bull. World Health Organ*
22. Schatz, Fooks, Mcelhinney et al. (2013) "Bat rabies surveillance in Europe" *Zoonoses Public Health*
23. Freuling, Beer, Conraths et al. (2011) *Novel lyssavirus in Natterer's bat, Germany. Emerg. Infect. Dis*
24. Leopardi, Dacheux, Serra-Cobo et al. (2025) "European distribution and intramuscular pathogenicity of divergent lyssaviruses West Caucasian bat virus and Lleida bat lyssavirus"
25. Dundarova, Ivanova-Aleksandrova, Bednarikova et al. "Phylogeographic Aspects of Bat Lyssaviruses in Europe: A Review"
26. Klein, Calvelage, Schlottau et al. (1538) "Retrospective Enhanced Bat Lyssavirus Surveillance in Germany between 2018-2020" *Viruses*
27. Kohl, Nitsche, Kurth "Update on Potentially Zoonotic Viruses of European Bats. Vaccines 2021"
28. Golding, Wu, Wilkie et al. (2024) "Investigating the emergence of a zoonotic virus: Phylogenetic analysis of European bat lyssavirus 1 in the UK. Virus Evol"
29. Van Der Poel, Lina, Kramps (2006) "Public health awareness of emerging zoonotic viruses of bats: A European perspective. Vector Borne Zoonotic Dis"
30. Negredo, Palacios, Vázquez-Morón et al. (2011) "Discovery of an Ebolavirus-like Filovirus in Europe" *PLoS Pathog*
31. Kemenesi, Kurucz, Dallos et al. (2016) "Re-emergence of Lloviu virus in Miniopterus schreibersii bats" *Emerg. Microbes Infect*
32. Tóth, Hume, Suder et al. "Isolation and genome characterization of Lloviu virus from Italian Schreibers's bats"
33. Goletic, Goletic, Omeragic et al. (2023) "Metagenomic sequencing of Lloviu virus from dead Schreiber's bats in Bosnia and Herzegovina. Microorganisms"
34. Forró, Marton, Fehér et al. (2021) "Phylogeny of Hungarian EBLV-1 strains using whole-genome sequence data" *Transbound. Emerg. Dis*
35. Vidovszky, Boldogh (2011) "Detection of adenoviruses in the Northern Hungarian bat fauna" *Magy. Állatorvosok Lapja*
36. Kemenesi, Dallos, Görföl et al. (2014) "Molecular survey of RNA viruses in Hungarian bats: Discovering novel astroviruses, coronaviruses, and caliciviruses"
37. Kemenesi, Zhang, Marton et al. (2015) "Genetic characterization of a novel picornavirus detected in Miniopterus schreibersii bats" *J. Gen. Virol*
38. Aguilar-Vargas, Solorzano-Scott, Baldi et al. "Passive epidemiological surveillance in wildlife in Costa Rica identifies pathogens of zoonotic and conservation importance" *PLoS ONE*
39. Fooks, Brookes, Johnson et al. (2003) "European bat lyssaviruses: An emerging zoonosis" *Epidemiol. Infect*
40. Speare, Luly, Reimers et al. (2013) "Antibodies to Australian bat lyssavirus in an asymptomatic bat carer" *Intern. Med. J*
41. Schatz, Freuling, Auer et al. (2014) "Enhanced passive bat rabies surveillance in indigenous bat species from Germany-A retrospective study" *PLoS Negl. Trop. Dis*
42. Picard-Meyer, Servat, Wasniewski et al. (2017) "Bat rabies surveillance in France: First report of unusual mortality among serotine bats" *BMC Vet. Res*
43. Harris, Brookes, Jones et al. (2006) "Passive surveillance (1987 to 2004) of United Kingdom bats for European bat lyssaviruses" *Vet. Rec*
44. Eblé, Dekker, Van Den End et al. (2024) "A case report of a cat infected with European bat lyssavirus type 1, the Netherlands" *Euro Surveill*
45. Simmons, Cirranello (2025) "Bat Species of the World: A Taxonomic and Geographic Database, version 1.9"
46. Vázquez-Morón, Avellón, Echevarría (2006) "RT-PCR for detection of all seven genotypes of Lyssavirus genus" *J. Virol. Methods*
47. Conceição-Neto, Zeller, Lefrère et al. (2015) "Modular approach to customise sample preparation procedures for viral metagenomics: A reproducible protocol for virome analysis" *Sci. Rep*
48. Minh, Schmidt, Chernomor et al. (2020) "IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era" *Mol. Biol. Evol*
49. Letunic, Bork (2024) "Interactive Tree of Life (iTOL) v6: Recent updates to the phylogenetic tree display and annotation tool" *Nucleic Acids Res*
50. Wise, Marston, Banyard et al. (2005) "Passive surveillance of United Kingdom bats for lyssaviruses" *Epidemiol. Infect*
51. Leopardi, Priori, Zecchin et al. (2018) "Active and passive surveillance for bat lyssaviruses in Italy revealed serological evidence for their circulation in three bat species" *Epidemiol. Infect*
52. Hill, Stentiford, Walker et al. (2024) "Realising a global One Health disease surveillance approach: Insights from wastewater and beyond" *Nat. Commun*
53. Macfarlane, Rocha (2020) "Guidelines for communicating about bats to prevent persecution in the time of COVID-19" *Biol. Conserv*
54. Van Gucht, Verlinde, Colyn et al. (2013) "Favourable outcome in a patient bitten by a rabid bat infected with the European bat lyssavirus-1" *Acta Clin. Belg*
55. Smreczak, Orłowska, Marzec et al. (2018) "Bokeloh bat lyssavirus isolation in a Natterer's bat, Poland" *Zoonoses Public Health*
56. Schatz, Ohlendorf, Busse et al. (2014) "Twenty years of active bat rabies surveillance in Germany: A detailed analysis and future perspectives" *Epidemiol. Infect*
57. Regnault, Evrard, Plu et al. (2022) "First Case of Lethal Encephalitis in Western Europe Due to European Bat Lyssavirus Type 1" *Clin. Infect. Dis*
58. Schatz, Teifke, Mettenleiter et al. (2014) "Lyssavirus distribution in naturally infected bats from Germany" *Vet. Microbiol*
59. Kemenesi, Tóth, Mayora-Neto et al. (1706) "Isolation of infectious Lloviu virus from Schreiber's bats in Hungary" *Nat. Commun*
60. Parize, Travecedo Robledo, Cervantes-Gonzalez et al. (2020) "Circumstances of Human-Bat interactions and risk of lyssavirus transmission in metropolitan France" *Zoonoses Public Health*
61. Vodopija, Lojkić, Hamidović et al. (2024) "Bat Bites and Rabies PEP in the Croatian Reference Centre for Rabies 1995-2020" *Viruses*
62. "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=PMC12793531&blobtype=pdf | # P-2200. A Retrospective Chart Review of Respiratory Syncytial Virus-related Hospitalizations in Adults
Stephen Baker, Abdullah Salama, Luke Shea, Ross Bernstein, Katherine Getman, Quinlan Wu, Shane Sacco, Jessica Abrantes-Figueiredo, Eun Lee, Kevin Dieckhaus
Hartford Hospital, Hartford, Connecticut; 9 UConn Health, Southington, Connecticut Session: 238. Virology Wednesday, October 22, 2025: 12:15 PM Background. Respiratory syncytial virus (RSV) causes annual outbreaks of respiratory illness with peak incidence occurring from late fall to early spring. While RSV is a leading cause of pediatric hospitalizations, it also poses a significant burden on older adults.
## Table 1. Cohort Characteristics
Table 2. Multivariable-adjusted risk of supplemental oxygen, ICU needed, and stay length Methods. We reviewed electronic medical records for hospitalized adults with laboratory-confirmed RSV between July 1, 2020, and June 30, 2024. Demographics, risk factors and outcomes were recorded. We created logistic and linear regression models to obtain multivariable-adjusted risks of supplemental oxygen needs, ICU level care, and increased length of stay (LOS). |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792641&blobtype=pdf | # P-1814. Comparative Analysis of Receptor Binding Affinity and Potential Zoonotic Spillover of H5Nx Clade 2.3.4.4b Avian Influenza Virus from Bangladesh
Subyeta Binte Sarwar, Ayman Bin, Abdul Mannan, Akash Saha, Md Hasibul Hassan, Mohammad Hossain, Sukanta Chowdhury, Mohammed Rahman
## Abstract
2 -transformed fold changes in antibody levels for each virus: (A) Human metapneumovirus (HMPV), (B) Influenza A, (C) Influenza B, and (D) Respiratory Syncytial Virus (RSV). Individual points represent participant-level values. A log 2 fold change of 0 indicates no change in antibody level; values below 0 indicate waning, while values above 0 indicate an increase. Comparisons between quartile 1 and quartile 4 were assessed using the Wilcoxon rank-sum test. All comparisons had a p<0.0001. 2 -transformed fold change in gMean EBS between timepoints was compared for each RV between Q1 and Q4. Q1 fold changes were significantly greater than Q4 for all RVs, indicating greater waning (Fig. 2A-D, P< 0.0001).
Background. Highly pathogenic avian influenza H5Nx viruses of clade 2.3.4.4b have caused extensive outbreaks in wild birds and poultry worldwide, with sporadic human cases reported globally since 2022. Viruses of this clade possess the ability to breach the avian-mammalian species barrier. To date, no human cases have been reported in Bangladesh. Therefore, we investigated the mutational profile, receptor-binding affinity, and stability of circulating 2.3.4.4b avian strains in Bangladesh and compared them to 2.3.4.4b human strains isolated globally.
Poster Abstracts • OFID 2026:13 (Suppl 1) • S1119 Methods. We retrieved complete sequences of clade 2.3.4.4b H5Nx avian strains (n=50, from Bangladesh) and human strains (n=89, worldwide) from GISAID and screened for mutational markers in the hemagglutinin (HA) segments. To assess the impact of these mutations, we conducted molecular docking of HA with avian and human receptor analogs-3′-sialylacetyllactosamine (3′-SLN) and 6′-sialylacetyllactosamine (6′-SLN)-using the GlideXP module in Maestro. We further analyzed dynamic stability through molecular dynamics simulations for up to 100 ns using the Desmond module in the Schrödinger suite.
Results. All 2.3.4.4b viruses from Bangladesh had the polybasic cleavage site REKRRKRGLF. Despite the presence of several markers associated with increased binding to 6′-SLN in all strains, human-derived 2.3.4.4b strains exhibited additional adaptive mutations, including 104G (33%), 120M (40%), 131Q (39%), 172A (98%), 211I (37%), 226A (40%), 336N (29%), and 526V (39%). Docking analysis revealed that the mean binding affinities of avian 2.3.4.4b strains were -7.87 (95% CI: -7.64 to -8.09) for 3′-SLN and -7.71 (95% CI: -7.40 to -8.02) for 6′-SLN, whereas human 2.3.4.4b viruses showed stronger binding affinities of -8.2 (95% CI: -8.0 to -8.5) for 3′-SLN and -9.5 (95% CI: -8.8 to -10.3) for 6′-SLN.
Conclusion. Our findings suggest that while Bangladeshi avian 2.3.4.4b viruses share important molecular markers with human strains and display moderate binding affinity toward human-like receptors, additional mutations-particularly 131Q and 172A found in human isolates-may enhance binding to human-type receptors, potentially facilitating zoonotic transmission and serious infections in humans.
Disclosures. All Authors: No reported disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792117&blobtype=pdf | # P-1602. SARS-CoV-2 Reinfection and Antibody Kinetics Among Healthcare Workers and Community Members in Ethiopian: A Two-Year Longitudinal Study
Esayas Kebede Gudina, Eyob Girma, Rebecca Kisch, Kira Elsbernd, Solomon Ali, Wondimagegn Adissu, Christof Geldmacher, Céline Pellaton, Andreas Wieser
Background. SARS-CoV-2 remains a significant public health issue in Ethiopia, where data on reinfection rates, immune responses, and transmission patterns remain limited. This study aimed to analyse antibody responses, reinfection rates and their determinants, and evaluate the seasonality of SARS-CoV-2 infections in Ethiopia. Methods. A longitudinal study was conducted from October 2022 to December 2024 involving 500 healthcare workers (HCWs) and 500 community members. Data collection included baseline questionnaires, quarterly SARS-CoV-2 anti-nucleoside and anti-spike antibody monitoring, annual Interferon-Gamma Release Assays, and neutralization assays against various variants. Reinfection was defined serologically. Nasopharyngeal swab of those with symptoms was tested for respiratory pathogens using Multiplex Real-Time PCR.
Results. Only 49.1% of the participants have ever received COVID-19 vaccine. At baseline, 97.8% of the participants had evidence of prior SARS-CoV-2 infection. Reinfection was frequent, with peaks observed during months of December to February each year. Overall, 91.3% of the participants were reinfected at least once during the follow-up. HCWs exhibited lower odds of reinfection compared to community members (OR=0.73; 95% CI: 0.585-0.907; p=0.005). Vaccination lowered reinfection risk by ∼30%, with 2-dose (OR=0.7; 95% CI: 0.537-0.900; p=0.006) and 3-dose (OR=0.68; 95% CI: 0.478-0.971; p=0.034) recipients having reduced odds compared to unvaccinated individuals. Neutralization assays indicated that antibody responses varied by SARS-CoV-2 variant, vaccine type, number of doses, and history of reinfection. Individuals experiencing reinfection showed higher subsequent neutralization activity, particularly against newer variants. Neutralizing activity generally increased over the 2-year study period, likely due to repeated infections.
Conclusion. This study reveals high SARS-CoV-2 reinfection rates with some seasonal variability. Vaccination provides protection against reinfection. The findings indicate the need for enhanced surveillance, continued vaccination, particularly for high-risk groups, and robust pandemic preparedness strategies tailored to the Ethiopian context.
Disclosures. All Authors: No reported disclosures
Pfizer, Collegeville, PAfoot_1 Pfizer Inc, Collegeville, Pennsylvania 3 Pfizer Inc., Collegeville, Pennsylvania
$$1 1 1 1 1 2 2 3 1$$ |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793678&blobtype=pdf | # P-2193. Epidemiology and healthcare resource utilization associated with Respiratory Syncytial Virus (RSV), Human Metapneumovirus (hMPV) and Parainfluenza Virus (PIV) infection in Central New York, USA
Stephen Thomas, Dongliang Wang, Joy Higuchi, Kelley Mooney, Rachael Cavelli, Jacqueline Malay, Michelle Klick, Samantha Gallup, Danning Huang, Oliver Martyn
Background. Respiratory viruses such as human metapneumovirus (hMPV), respiratory syncytial virus (RSV), and parainfluenza virus (PIV) are significant pathogens associated with respiratory infections, particularly in older adults and those with medical co-morbidities. These infections can lead to severe outcomes, including respiratory failure, exacerbation of chronic diseases, and increased mortality. Despite their impact, comprehensive data on the clinical features and healthcare utilization associated with these viruses in older adults is limited owing to a lack of widespread testing and diagnosis, especially so for hMPV and PIV.
Table 1. The total number of outpatient, ER and hospital were calculated by counting out-pt_dx_visit_not_applic==0, er_not_applic==0, and inpt_no_applic==0, respectively. Totals include coinfections, the values under each column represent mono-infections.
Methods. This retrospective chart review analyzes the clinical features and healthcare utilization of adults aged 50 years and older diagnosed with hMPV, RSV, and PIV 1-4 in Central New York State between 2021 and 2024. SUNY Upstate Medical University uses the Biofire multiplex platform, capable of detecting over 15 respiratory pathogens, on all respiratory samples. The study includes patients with positive test results for hMPV, RSV, or PIV 1-4 obtained in outpatient, emergency room and hospital settings. Key data points collected include positivity rates, demographics, comorbidities, healthcare encounters, and clinical outcomes. Statistical analyses are performed using R.
Results. Positivity by care setting and pathogen over the study period are presented in table 1. Analysis of clinical characteristics, healthcare utilization and outcomes are ongoing.
Conclusion. Preliminary results suggest similar rates of RSV, hMPV and PIV1-4 over the study period, although hMPV was diagnosed less frequently in the hospital setting. Analyses are ongoing and full results are expected by Q3 2025. Enhanced understanding of these infections can inform future public health strategies and vaccine development.
Disclosures. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12791608&blobtype=pdf | # P-1812. Understanding the role of the non-coding control region in JC polyomavirus cell tropism and central nervous system entry
Elizabeth Wagstaff, ; Sabrain Tan
Background. JC Polyomavirus (JCPyV) infects 40-90% of the adult population in a benign kidney infection. However, in a subset of immunocompromised individuals it causes a rapid and often fatal infection known as Progressive Multifocal Leukoencephalopathy (PML). Currently there are no treatments for PML and those that survive the initial infection are left with severe lifelong complications. Much of the process the virus must undergo to travel to the brain and establish infection is currently unknown. The non-coding control region (NCCR) of the viral genome is thought to have an important role in establishing infection in the central nervous system (CNS) as it contains many transcription factor binding sites. In neurotropic strains of the virus there are many insertions, duplications, and deletions in this region that may contribute to cell tropism and increased replication in the CNS. However, two published cases have reported JCPyV in the CNS without NCCR rearrangement suggesting the role of the region may be more complex than previously hypothesized. Studies in this area have been limited previously due to a lack of realistic models for infection and there is confusion over which cells in the CNS are being infected and what changes in the virus permit infection. Methods. We have developed several novel primary cell infection models, including a brain organoid model, to test infection by wildtype virus with and without NCCR rearrangements. Additionally, we are using advanced molecular techniques including single cell RNA sequencing to identify the interactions between JCPyV and host cells.
## JC virus infection of primary astrocytes
Results. Primary oligodendrocytes and tubule epithelial cells can be infected by wildtype archetype and neurotropic JCPyV. Rearranged JCPyV infects both cell types at a significantly higher level than archetype. Virus with non-rearranged NCCR is capable of infecting oligodendrocytes, but is less efficient than rearranged virus. Brain organoids appear to be infectable with JCPyV neurotropic virus (MAD-1).
Conclusion. The knowledge gained from these studies will aid in understanding how JCPyV changes from a benign kidney disease to a deadly neuro pathogen and potentially lead to innovative therapeutics and preventatives for patients.
Disclosures. All Authors: No reported disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12798066&blobtype=pdf | # Isolation of a Gordonia rubripertincta bacteriophage, BluerMoon, from topsoil in Lubbock, Texas
Laurissa Miller, Natalie Block, Whitney Dickens, Carson Bellew, Christian Deluna, Francesca Makilan, Malli Bhakta, Ashleigh Crawford, Trinity Criner, Chase Drucker, Aqsa Fayyaz, Jasmine Goh, Caitlyn Guetersloh, Claire Jansen, Dana Pham, Andrea Resendez, Austen Rowell, Fahareen Mosharraf, Allie Smith, Lisa Bono
## Abstract
We isolated, annotated, and analyzed the genome of an environmental bacteriophage. BluerMoon was discovered from topsoil in Lubbock, Texas, as part of the Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Sciences (SEA-PHAGES) program. BluerMoon is a member of the DJ cluster according to the Actinobacteriophage database, PhagesDB (https://phagesdb.org).
, HHPRED v3.0 (16) (using the PDB_mmCIF70), and Pfam-v36.0 (17) were utilized to predict gene function. All software applications mentioned were used with default configurations (Table 1).
BluerMoon was isolated from an enrichment culture on G. rubripertincta. BluerMoon is part of the DJ cluster within the Actinobacteriophage database, PhagesDB (https:// phagesdb.org), and is lytic. DJ cluster phage has an average GC content of 51.5%, an average genome size of 60,476 bp, and infects Gordonia spp. We manually classified BluerMoon as having a 3' sticky overhang genome termini as previously described (18). The lack of tRNAs found within BluerMoon was determined using tRNAscan-SE v2.0 and ARAGORN v1.2.38, consistent with all phages in the DJ cluster (19,20).
## References
1. Ventura, Canchaya, Tauch et al. (2007) "Genomics of actinobacteria: tracing the evolutionary history of an ancient phylum" *Microbiol Mol Biol Rev*
2. Frantsuzova, Bogun, Solomentsev et al. (2023) "Whole genome analysis and assessment of the metabolic potential of Gordonia rubripertincta strain 112, a degrader of aromatic and aliphatic compounds" *Biology (Basel)*
3. Pope, Mavrich, Garlena et al. (2017) "Science Education Alliance-Phage Hunters Advancing Genomics and Evolution ary Science (SEA-PHAGES)" *mBio*
4. Poxleitner, Pope, Jacobs-Sera et al. (2018) "Phage discovery guide"
5. Coclet, Camargo, Roux (2024) "MVP: a modular viromics pipeline to identify, filter, cluster, annotate, and bin viruses from metagenomes"
6. Chen, Zhou, Chen et al. (2018) "Fastp: an ultra-fast all-in-one FASTQ preprocessor" *Bioinformatics*
7. Bankevich, Nurk, Antipov et al. (2012) "SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing" *J Comput Biol*
8. Quinlan, Hall (2010) "BEDTools: a flexible suite of utilities for comparing genomic features" *Bioinformatics*
9. Li, Handsaker, Wysoker et al. "Genome project data processing subgroup" *Bioinformatics*
10. Pope, Jacobs-Sera, Clokie et al. (2018) "Annotation of bacteriophage genome sequences using DNA master: an overview"
11. (2016) "PECAAN: phage evidence collection and annotation network user guide"
12. Salisbury, Tsourkas (2019) "A method for improving the accuracy and efficiency of bacteriophage genome annotation" *Int J Mol Sci*
13. Mount (2007) "Using the basic local alignment search tool (BLAST)"
14. Cresawn, Bogel, Day et al. (2011) "Phamerator: a bioinformatic tool for comparative bacteriophage genomics" *BMC Bioinformatics*
15. Yang, Derbyshire, Yamashita et al. (2020) "NCBI's conserved domain database and tools for protein domain analysis" *CP in Bioinformatics*
16. Söding, Biegert, Lupas (2005) "The HHpred interactive server for protein homology detection and structure prediction" *Nucleic Acids Res*
17. Mistry, Chuguransky, Williams et al. (2021) "Pfam: the protein families database in 2021" *Nucleic Acids Res*
18. Russell (2018) "Sequencing, assembling, and finishing complete bacteriophage genomes" *Methods Mol Biol Clifton NJ*
19. Chan, Lin, Mak et al. (2021) "tRNAscan-SE 2.0: improved detection and functional classification of transfer RNA genes" *Nucleic Acids Res*
20. Cheng, Li, Zhang et al. (2024) "Genomic diversity of phages infecting the globally widespread genus Sulfurimonas" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817928&blobtype=pdf | # Demonstration of in vivo efficacy, cryo-EM-epitope identification, and breadth of two anti-alphavirus bispecific single domain antibodies
Christina Gardner, Sergei Pletnev, Jinny Liu, George Anderson, Lisa Shriver-Lake, Tatsiana Bylund, Courtney Green, Tyler Stephens, Matthew Sutton, Yaroslav Tsybovsky, Mario Roederer, Peter Kwong, Tongqing Zhou, Ellen Goldman, Crystal Burke
## Abstract
Venezuelan equine encephalitis virus (VEEV) is an arbovirus that causes a disease in which 4%-14% of individuals can develop neurological symptoms. Prior to 1970, VEEV was developed as a biological threat agent due to its stability and high morbidity when administered by aerosol. Currently, no FDA-licensed vaccines nor therapeutics for VEEV exist. Single-domain antibodies (sdAbs) may provide a therapeutic option due to their small size and ability to bind recessed epitopes not recognized by conventional antibodies. This study identified two bivalent sdAbs that were able to protect mice from a lethal challenge against both epizootic and enzootic subtypes of VEEV. Cryo-EM structures of sdAb-VEEV complexes revealed the sdAbs that comprised the bivalent sdAbs to recognize a mixture of conserved and non-conserved regions of the VEEV envelope proteins. While all three of the cryo-EM-characterized epitopes were unique in terms of their recognized VEEV residues, two sdAbs, V2B3 and V2C3, overlapped sterically, explaining why only their combinations with the non-sterically overlapping sdAb V3A8f, which composed the bivalent sdAbs described here, were so particularly effective. Binding and neutralization studies found that the bivalent sdAbs have the potential to be broad-spectrum anti-alphavirus therapeutics as they cross-neu tralize multiple alphaviruses. IMPORTANCE Alphaviruses are no longer geographically constrained to one region of the world but are expanding to be of global concern. In many regions of the world, multiple alphaviruses co-circulate; therefore, having a therapeutic that is pan-alphavirus is important. A cocktail of multiple pan-alphavirus binding/neutralizing antibodies (Abs) may provide optimal coverage against alphaviruses while decreasing the prevalence of viral escape mutants, which could cause the therapeutic to no longer be efficacious. Structures of these Abs, defining their recognition, could assist in identifying optimal combinations. A bivalent pan-alphavirus single-domain antibody could be used in a cocktail with already identified alphavirus IgG antibodies. KEYWORDS alphavirus, Venezuelan equine encephalitis virus, single-domain antibody S ingle-domain antibodies (sdAbs, also called nanobodies or VHHs) are the recombi nantly expressed variable domain from camelid heavy chain-only antibodies and provide rugged recognition elements with excellent affinities that can be engineered toward specific applications (1). SdAb properties that are advantageous for therapeutics include good tissue penetration in vivo, low immunogenicity, recognition of cryptic epitopes, and ability to tune serum half-life through genetic fusions or PEGylation (2, 3). Additionally, sdAbs have a proven safety profile with the first sdAb-based product,
caplacizumab, receiving FDA approval in 2019 (2). Recently, sdAbs have become widely recognized for their potential use as antiviral therapeutics (3,4), with sdAb-based constructs reported for a range of viruses, including alphaviruses (5,6), influenza (7)(8)(9), coronaviruses (10)(11)(12), and filoviruses (13). Constructs in which several sdAbs are linked to form multivalent reagents often provide greater viral neutralization than the un-linked sdAbs (6,7,9).
Venezuelan equine encephalitis virus (VEEV) is a mosquito-borne virus that causes periodic epizootic and epidemic outbreaks in equines and humans and has been widely studied because of its potential use as a biothreat agent (14). Although not usually fatal in humans (<1%), VEEV has a high morbidity rate due to the subclinical to clinical ratio being 1:1, can cause neurological disease in 4%-14% of cases, and is fatal in 19%-83% of equine cases (15). All subtypes of VEEV can cause disease in humans. The VEEV IAB and IC subtypes are epizootic strains capable of causing epidemics, while VEEV ID and IE subtypes are enzootic strains mainly transmitted between the mosquito vector and rodent reservoir host. Currently, there are no FDA-approved vaccines or therapeutics for VEEV, and the only treatment after VEEV infection is supportive care. Monoclonal antibodies (mAbs) can be effective in preventing VEEV infection in mice when given therapeutically (16,17). Additionally, a mAb has been shown to prevent non-human primates from severe disease even when administered 48 h post-exposure (18).
We previously described bivalent sdAb constructs that are able to bind and neu tralize VEEV in vitro (6). Assayed through plaque reduction and neutralization testing (PRNT), the best constructs reduced 50% of plaques at around 1 ng/mL or lower for both TC-83, the BSL-2 vaccine strain, and the wild-type Trinidad donkey (TrD) strain of the VEEV IAB subtype. Here, we genetically coupled the two-lead bivalent sdAb constructs to an albumin-binding domain and assessed their serum half-life, ability to protect mice from a subcutaneous (SC) and aerosol (AE) VEEV challenge, and ability to protect against multiple subtypes of VEEV. Further, we solved cryo-EM structures to delineate epitope-binding sites, assessed the diversity of the recognized epitopes, and evaluated breadths of binding and neutralization against multiple alphaviruses. Both constructs demonstrated efficacy against multiple subtypes of VEEV and different routes of exposure. Additionally, the sdAbs could bind and/or neutralize multiple different alphaviruses.
## RESULTS
## Single-domain antibody constructs
Previously, bivalent sdAb constructs capable of neutralizing VEEV as determined by PRNT were identified (6). Here, the two most potent neutralizing constructs, CA (composed of sdAb V2C3 linked to sdAb V3A8f ) and BA (composed of sdAb V2B3 linked to sdAb V3A8f ) were engineered and expressed as genetic fusions with an albumin-binding domain to increase serum half-life (19). The constructs were expressed as fusions with the Alb1 anti-albumin sdAb (20) or the albumin-binding domain (abd) from streptococcal protein G (21) in order to ensure production of a construct that retains potency while integrating a half-life extension property. Both fusion constructs retained the ability to neutralize VEEV as determined by PRNT (Table S1), however, higher yield protein expression was achieved for the bivalent sdAb constructs linked with the abd; therefore, the fusions with the abd were used in follow-on studies to determine the therapeutic potential of the bivalent sdAb constructs (Fig. S1).
because (ii) it enabled imaging of the same mouse over time, providing a more accurate estimation of the PK in individual animals. Prior to administering the VivoTag 680 XL-tagged sdAb, mice were imaged to account for any background autofluorescence. After background image collection, four mice received 200 µg of tagged CA-abd through intraperitoneal injection and then were imaged at pre-determined times (0.5, 2, 4, 8, 24, and 29 h) to quantitate the fluorescent signal (Fig. 1A through C). As expected, VivoTag 680 XL-tagged CA-abd was observed disseminated throughout the entire mouse (Fig. 1A). Using the Living Image Software, the fluorescent signal in particular regions of interest (ROIs) was measured. Whole body and head only ROIs (Fig. 1B andC, respec tively) were graphed to determine the longevity of CA-abd. These data suggest the half-life of the CA-abd antibody is greater than 29 h.
A potential pitfall with determining serum half-life of the sdAb-abd constructs through in vivo imaging is that fluorescently tagging the protein may disrupt the ability of the abd to bind albumin, thereby decreasing the half-life of the sdAb construct. The chemistry used to tag the proteins preferentially targets lysine residues, and the abd contains five lysines. Additionally, a traditional PK study examining the levels of both CA-abd and BA-abd in the serum and the brain at time points up to 24 h post-injec tion was performed. SdAb constructs were quantified by PRNT showing that the sdAb measured in the serum was functional and able to neutralize the VEEV TrD virus. The serum PK results suggest that both CA-abd and BA-abd possess serum half-lives over 24 h (Fig. 1D), confirming our IVIS PK results. Interestingly, neutralizing antibodies could be detected in brain tissue from animals receiving the CA-abd antibody but not the BA-abd, demonstrating that at least one of the antibodies could reach the central nervous system (Fig. 1E).
## Efficacy against multiple VEEV subtypes
In the VEEV antigenic complex, there are 14 antigenic lineages grouped into six subtypes, with only two lineages that are epizootic strains (IAB and IC), while the rest are enzootic strains (ID-F, II-VI) (22). In just the E1 glycoprotein, there can be up to 29% nucleotide divergence between the six different subtypes of VEEV (23). The FDA guidance for product development under the Animal Rule states that the challenge agent used in efficacy studies should reflect the etiologic agent responsible for the human disease or condition. The VEEV IAB and IC subtypes are epizootic strains of VEEV that have either historically or currently, respectively, caused epidemics. To evaluate the ability of the sdAbs to protect against a wide range of VEEV subtypes, a series of studies were conducted in the BALB/c mouse model of VEEV. Challenges were conducted using a subcutaneous (SC) footpad inoculation to mimic natural infection through mosquito bite or an aerosol exposure (AE) to model a nefarious use of VEEV as a threat agent. The first study evaluated the ability of the sdAbs to protect against a SC inoculation with the VEEV INH-9813 strain, a IC subtype isolated from a human infection in 1995 (24). Mice were inoculated with 1,000 PFU VEEV INH-9813 in the rear footpad, and therapeutic sdAb was administered intraperitoneally 1 h post-challenge. Control mice received either an irrelevant antibody (negative control) or the anti-VEEV IgG 1A3B-7 (positive control), which has previously shown efficacy in the mouse model against multiple strains of VEEV (25) and in the NHP model against VEEV TrD (18). Treatment with either CA-abd or BA-abd protected 90%-100% of the mice not only from lethality (P < 0.0001; Fig. 2A) but also from weight loss (P ≤ 0.002; Fig. 2B) and development of clinical signs of disease (P ≤ 0.005; Fig. 2C).
Since all subtypes of VEEV are capable of causing disease in humans (22), study 2 aimed to determine the breadth of CA-and BA-abd to protect against enzootic subtypes of VEEV. The VEEV ID subtype encompasses enzootic strains of VEEV that are mainly transmitted between the mosquito vector and the rodent reservoir host (26). Here, mice were SC inoculated with 1,000 PFU VEEV ZPC738 in the rear footpad, and again, thera peutic sdAb was administered intraperitoneally 1 h post-challenge. Similar to the IC subtype study, treatment with CA-or BA-abd was able to protect 100% of the mice not only from lethality (P < 0.0001; Fig. 2D) but also from weight loss (P ≤ 0.00008; Fig. 2E), development of clinical signs of disease (P ≤ 0.0001; Fig. 2F), and serum viremia (P ≤ 0.008; Fig. S2A). These data suggest that the two bivalent sdAbs have pan-anti-VEEV potential.
Previously, VEEV was developed as a threat agent due to the high rate of morbidity in humans with a subclinical to clinical symptom ratio of 1:1 (14,27,28). Past efforts to develop VEEV as an aerosol disseminated threat agent have resulted in the need to test any VEEV-specific medical countermeasures for the ability to protect against this challenge route. For this reason, studies 3 and 4 evaluated if CA-or BA-abd could protect mice against aerosol exposures to epizootic strains of VEEV (IAB and IC subtype, respec tively). First, BALB/c mice were aerosol exposed to a target dose of 1,000 PFU VEEV TrD, an IAB strain initially isolated in 1943 (29) with a history of causing laboratory exposures (30,31). Both CA-and BA-abd were able to protect 80% of mice from lethality (P ≤ 0.001; Fig. 2G), weight loss (P ≤ 0.01; Fig. 2H), development of clinical signs of disease (P ≤ 0.01; Fig. 2I), and serum viremia (P ≤ 0.02; Fig. S2B). Demonstrating protection against the IAB subtype, study 4 evaluated the sdAbs against the currently circulating epizootic subtype (IC) through an aerosol exposure with a target dose of 1,000 PFU VEEV INH-9813. Contrary to earlier experiments, the sdAbs were less effective at protecting mice against the VEEV IC aerosol exposure (Fig. 2J through L). Although there was no significant difference in the mean time-to-death between the negative control group (7.5 days) and the CA-abd-treated group (7.2 days), there was a significant reduction in the percent mortality (P = 0.03; Fig. 2J). Both the CA-abd-and BA-abd treated groups lost significantly less weight compared with the negative control group (P ≤ 0.004; Fig. 2K) and a significant delay in the development of clinical signs of disease compared with the negative control group (P ≤ 0.0006; Fig. 2L). Surprisingly, the 1A3B-7-positive control, which consistently provides 90-100% protection against aerosol exposure with VEEV TrD (Fig. 2G through I and (16,25)) also failed to provide complete protection from the VEEV IC aerosol challenge (Fig. 2J). One explanation of the reduced protection of 1A3B-7-positive control against a VEEV IC aerosol is that 1A3B-7 has a 16-fold higher PRNT80 against VEEV IAB (0.049 µg/mL) than VEEV IC (0.781 µg/mL). Taken together, the data suggest that CAand BA-abd bivalent sdAbs provide significant protection against multiple lineages and routes of exposures of VEEV.
## Cryo-EM structures and epitope identification
Since both bivalent sdAbs provided protection from VEEV in vivo, structural insight of the binding epitopes of the three individual sdAb components of CA-and BA-abd against VEEV was sought to determine if each bound to unique or overlapping epitopes. Cryo-grids for the complex of sdAb V2B3, from the BA-abd, with VEEV virus-like particle (VLP) were prepared, and single particle cryogenic-electron microscopy (cryo-EM) data were collected on a Titan Krios. From a total of 41,401 particles, ab initio reconstruction for the VEEV VLP (strain TC83) utilizing icosahedral symmetry was performed. To visualize the sdAb V2B3, C1 symmetry expansion was performed with mask generation around four VLP spikes surrounding the icosahedral threefold axis, and signal subtraction and local refinement were performed within the mask to obtain a 4.6 Å reconstruction of the V2B3-VLP spike complex (Fig. 3A; Fig. S3; Table S2). V2B3 bound to the interface between three molecules: E1 and E2 of one trimeric spike [recognized by complementar ity determining region 1 (CDR1), CDR2, and CDR3], as well as E1 from an adjacent trimeric spike (recognized by the sdAb N terminus as well as by CDR3) (Fig. 3B). This recognition appeared to lock trimers of the VLP together, suggesting a potential mechanism of blocking VLP disassembly as a means to inhibit virus entry.
For the structure of V2C3, from the CA-abd, with VEEV VLP, a similar approach was used, making cryo-grids for the complex of V2C3 sdAb and VEEV VLP and collecting single particle cryo-EM data. From a total of 31,101 particles, ab initio reconstruction for the VEEV VLP utilizing icosahedral symmetry was performed, and after masking, signal subtraction, and local refinement within the mask, a 6.0 Å reconstruction of the V2C3-VLP spike complex was determined (Fig. 3C; Fig. S4; Table S3). V2C3 bound to a site that overlapped the fusion loop, at the interface of E1 (recognized by CDR1-3) and E2 (recognized by CDR2) (Fig. 3D). This fusion peptide recognition likely disrupts the fusion of VLP and cellular membranes, thereby providing a potential mechanism for virus entry inhibition by this sdAb.
For the structure of V3A8f, present in both BA-and CA-abd, with VEEV VLP, cryo-grids for the complex of V3A8f and VEEV VLP were made, and single particle cryo-EM data were collected. From a total of 66,222 particles, ab initio reconstruction for the VEEV VLP utilizing icosahedral symmetry was performed, and after masking, signal subtraction and local refinement within the mask, a 4.3 Å reconstruction of the V3A8f-VLP spike complex was obtained (Fig. 3E; Fig. S24; Table S4). V3A8f also bound to a site that overlapped the fusion loop, at the interface of E1 (recognized by CDRs 2 and 3) and E2 (recognized by CDRs 2 and 3), likely inducing inhibition of the fusion mechanism, as a potential mecha nism for inhibiting virus entry (Fig. 3F). Despite both V2C3 and V3A8f sdAbs recognizing sites that overlapped the fusion loop, the two nanobodies used different interactive surfaces and different recognition chemistries.
Having determined the structures of V2B3, V2C3, and V3A8f sdAbs in complex with VEEV VLP, we could now compare their epitopes to the location of the binding site for the LDLRAD3 receptor as well as to other VEEV-neutralizing antibodies whose structures have been determined in complex with VEEV VLP. V2C3 epitope partially overlapped the LDLRAD3 binding site, while both V2B3 and V3A8f epitopes were proximal but did not overlap (Fig. S3; Table S2). In addition, we compared the sdAb epitopes with those of SKV09, SKV16, SKT05, and SKT20, whose structures were previously defined (32); we observed overlap between the epitopes of SKT20 Ab and V2B3, V2C3, and V3A8f sdAbs (Fig. S4; Table S2). The epitope of SKT20 Ab also overlapped with the LDLRAD3-binding site (Table S2).
Having determined the structures of V2B3, V2C3, and V3A8f sdAbs in complex with VEEV VLP, the structures were used to understand constraints to their linkage as bispecifics. Each of the separate structures of sdAbs was combined with the VEEV VLP to create a composite model (Fig. 4A), in which sdAbs V2B3 and V2C3 occupied overlapping positions. Calculation of overlap indicated 34% for V2B3 and V2C3, whereas V3A8f did not overlap either V2B3 or V2C3 (Fig. 4B). As there was no overlap with V3A8f, models of VEEV with this sdAb were examined, and either V2B3 (Fig. 4C) or V2C3 (Fig. 4D), specifically focusing on sdAbs that were within 60 Å of the N terminus of V3A8f, as these could then be covalently attached by a 20-residue linker. In both cases, only a single pair of sdAbs could be linked. For V2B3 linked to V3A8f, the calculated termini distance was 38 Å (Fig. 4E). For V2C3 linked to V3A8f, the calculated termini distance was 51 Å (Fig. 4F), although the physical distance was much closer, with side chains from framework regions 1 and 3 of V2C3 in close contact with side chains from the CDR2 region of V3A8f (Fig. 4G). Thus, structural modeling indicated two bispecific sdAbs were possible: V2B3-V3A8f, which comprised the BA-abd bivalent bispecific, and V2C3-V3A8f, which comprised the CA-abd bivalent.
## Breadth of binding and neutralization
With knowledge of the sdAbs VEEV VLP-binding epitopes, next the E1-E2 sequences of multiple alphaviruses were aligned to determine if the sdAbs bound conserved epitopes present in other alphaviruses (Fig. 5). For V3A8f, the epitope on E1 was highly conserved, but the region on E2 was not so conserved. Interactions with E2 involved both main chain and side chain interaction, suggesting the V3A8f would bind less well to divergent alphaviruses, with changes in E2. For V2B3, the recognized epitope was less conserved on E1 and quite variable on E2. For V2C3, the recognized portion of the epitope that overlaps with the fusion loop on E1 was conserved, while the recognized surface of domain II portion on E1 was less conserved. Based on the alignment, the CA-abd binds to more conserved epitopes than the BA-abd bivalent sdAb. Previously, we demonstrated that these sdAbs did not neutralize the other encephalitic alphaviruses western or eastern equine encephalitis virus (WEEV or EEEV, respectively) but did observe a reduction in plaque size against EEEV (6). Therefore, the ability of CA-abd and BA-abd to bind WEEV, EEEV, and the arthritogenic alphavirus chikungunya virus (CHIKV) VLP was evaluated since antibody binding with poor neutralization can still provide in vivo efficacy (18,(32)(33)(34)(35). In line with previously published data (Table S1) (6), both antibodies were able to bind the VEEV IAB VLP (Fig. 6A). Not unexpectedly, neither sdAb bound to the WEEV VLP (Fig. 6A), which agrees with our previous observation when assessing neutralization against WEEV (6). Interestingly,the CA-abd had strong binding to EEEV VLP and CHIKV VLP (Fig. 6A), while BA-abd did not bind to EEEV VLP and bound weakly to both the CHIKV VLP and CHIKV virus compared with stronger binding by CA-abd (Fig. 6A).
To determine the potential neutralizing breadth of the two bispecific sdAbs, the ability of the two sdAbs to neutralize multiple different alphaviruses, both encephalitic and arthritogenic, was evaluated. As expected, both sdAbs were potent neutralizers of VEEV IAB and VEEV IC strains (Fig. 6B and7); however, the potency of neutralization was reduced for VEEV IC in comparison to VEEV IAB (≥10-fold difference) (Fig. 6B and7). The reduced potency of the sdAbs against VEEV IC likely explains the reduced efficacy of these sdAbs against a VEEV IC aerosol challenge (Fig. 2J). While both bivalent sdAbs can neutralize multiple alphaviruses, CA-abd has the greatest breadth (10 of 12 viruses evaluated) and more potent neutralizing capabilities (Fig. 6 and7) in comparison to BAabd, which only neutralizes six of the viruses evaluated. Overall, these data suggest that the bispecific CA-abd antibody has the potential to be a pan-alphavirus therapeutic.
## DISCUSSION
SdAbs offer several advantages over conventional IgG including a reduced molecular mass allowing for better tissue penetration (36), increased thermostability (37), lower immunogenicity due to lack of Fc binding domain (2) and scalability for manufacturing due to flexibility in expression systems and purification methods (37). One disadvantage is a reduced half-life, a result of the smaller size, but this can be overcome with genetic fusions or PEGylation (3). By fusing the albumin binding domain (abd) to the bivalent sdAb constructs, we demonstrated we were able to maintain neutralization levels comparable to the bivalent constructs without the domain (Table S1) and have a tissue and serum half-life of ≥29 h (Fig. 1).
With the limited number of cases of VEEV per year, licensure of a therapeutic will likely occur under guidance of the FDA Animal Rule. Current guidance suggests challenge agents used in pivotal efficacy studies should replicate the etiological agent that causes disease in humans with no animal passaging and minimal cell culture passaging in the history of stock production (24). Frequently, vaccine and therapeutic efficacy studies utilize VEEV TC-83 or TrD strains as the challenge agent. While these strains are histori cally important, they are of the IAB subtype. There are two subtypes of VEEV that have been responsible for epidemic outbreaks: IAB and IC. Viral isolates of the IAB subtype have not caused outbreaks since the 1970s, instead strains from the IC subtype have been responsible for recent epidemic outbreaks (38). For that reason, we utilized VEEV INH-9813, a human isolate of the IC subtype with minimal cell culture passaging and no animal passaging (39), for our first efficacy study (Fig. 2A through C). A single dose of the bispecific sdAbs administered 1 h post-challenge was enough to protect the mice from a lethal SC VEEV IC challenge (Fig. 2A), providing protection from morbidity as well.
Considering the significant sequence diversity across VEEV subtypes, it was important that the bispecific sdAbs were evaluated for efficacy against other lineages of VEEV, especially since all lineages are capable of causing disease in humans (22). Weaver et al. have hypothesized that mutations in the major antigenic sites of the E2 glycoprotein of enzootic VEEV strains result in emergence of epizootic strains responsible for some of the major human outbreaks (40,41). Like the SC VEEV IC challenge, the two bispecific sdAbs were able to protect against both morbidity and mortality after a SC VEEV ID challenge (Fig. 2D through F). Since VEEV is on the CDC Select Agent list due to the potential of its use for nefarious purposes, we tested the ability of bispecific sdAbs to protect against an AE challenge against the two epizootic VEEV lineages as well. Both sdAbs protected 80% of the mice from a VEEV IAB AE challenge (Fig. 2G); however, less efficacy was observed against the VEEV IC AE challenge (Fig. 2J). Despite reduced ability to protect against lethality after a VEEV IC AE challenge, the sdAbs were able to significantly delay weight loss and onset of clinical signs of disease (Fig. 2K andL). In vitro neutralization assays found the CA-abd and BA-abd sdAbs neutralized VEEV IC less effectively than VEEV IAB (Fig. 6B). This result is somewhat expected as these sdAbs were originally identified based on binding and neutralization against VEEV TC-83, a VEEV IAB strain. Indeed, most VEEV-specific mAbs have been identified by panning against a VEEV IAB strain or were generated against a VEEV IAB antigen (32,(42)(43)(44)(45); therefore, the gold standard for testing in vivo efficacy of VEEV-specific mAbs is to challenge with VEEV IAB (16,25,42,(46)(47)(48).
For example, 1A3B-7, the mAb used as a positive control in these studies with demon strated efficacy in NHP studies (18), provided complete protection against AE to VEEV IAB only provided partial protection against VEEV IC. Few mAbs have actually been evaluated against other lineages of VEEV by SC (25,42) or AE challenge (17,45). Nonetheless, CAand BA-abd are potent neutralizers of VEEV IC. Future studies should test if efficacy could be increased with the addition of a second administration of sdAb ~3-4 days postchallenge. Our results highlight the need to expand testing and evaluation of new anti-VEEV Ab candidates against multiple VEEV subtypes by AE to identify the most efficacious protective anti-VEEV Abs.
The sdAbs have the potential for being a pan-alphavirus antibody. The V2C3 component of the CA antibody is in the same sequence family as the CHIKV CC3 sdAb that has been shown to neutralize CHIKV, RRV, and MAYV in vitro (5,6). We expand on these results to determine how cross-protective the VEEV sdAbs are against not only other strains of VEEV but other alphaviruses such as CHIKV (Fig. 6). The CA-abd sdAb demonstrated the widest breadth and better neutralization against multiple alphaviruses compared with the BA-abd sdAb (Fig. 6). Based on sequence alignment, the difference in breadth of neutralization among VEEV strains by the sdAbs was due to recognizing epitopes that were conserved among all VEEV strains of interest, except for the more divergent VEEV IIIA strain, which had sequence differences in epitope residues for all three sdAbs (Fig. 5); with only the V3A8f epitope showing sequence variation with an E to K change in domain B of E2 for VEEV strain IC.
Additionally, although the different lineages of CHIKV are more closely related to each other than the other arthritogenic alphaviruses (Fig. 6C), there is sequence variation and structural variation across the CHIKV lineages at the binding sites for both sdAbs (Fig. 5), which could explain why the two sdAbs have different neutralization profiles against the CHIKV Indian Ocean lineage compared with the other two lineages (Fig. 6B). In follow-up studies, we would like to evaluate the CA-abd bispecific sdAbs ability to protect in vivo against other alphaviruses to demonstrate that these bispecifics could be utilized as a pan-alphavirus therapeutic. Although the ability to bind/neutralize multiple alphaviruses could decrease the likelihood of viral escape mutants, a cocktail of multiple pan-alphavirus binding/neutralizing Abs (32,33,49) may provide optimal coverage against alphaviruses while decreasing the prevalence of viral escape mutants, which could cause the therapeutic to no longer be efficacious.
## MATERIALS AND METHODS
## SdAb constructs and protein purification
The sdAb constructs utilized in this work are derivatives of constructs that have been previously published. The CA-abd and BA-abd are derivatives of the VEEV binding, bivalent sdAbs V2C3-V3A8f and V2B3-V3A8f, respectively appended with an albumin binding domain (abd) (6,21). The protein sequence of the constructs is provided in Fig. S1.
Expression plasmids were transformed into ClearColi BL21(DE3), from Lucigen (Middleton, WI, USA), for protein production. Overnight cultures were started by inoculating either a freshly transformed colony, or 10 µL of a frozen cell stock, into 50 mL of LB Miller containing 100 µg/mL ampicillin and grown at 37°C overnight. The next day, the overnight culture was poured into 450 mL LB Miller (100 µg/mL ampicillin) in a 2-L baffled flask. Cultures were grown between 6 and 8 h at 30°C before the temperature was lowered to 25°C and the cultures induced by the addition of 0.5 mM IPTG, and grown overnight at 25°C.
After growth and induction, the solution was centrifuged to pellet the cells and the supernatant discarded. Pelleted cells from each 500 mL shake flask culture were resuspended in 14 mL of Tris-Sucrose buffer (100 mM Tris, 0.75M sucrose pH 7.5) by gently compressing them with a spatula. One milliliter of lysozyme (1 mg/mL) made up in Tris-Sucrose was added to the homogenized cells. Next, 28 mL of 1 mM EDTA was added drop-wise to the solution while the centrifuge tubes held in crushed ice were shaking on a rotating platform. After the addition of the EDTA, 0.25 mL of 5% deoxycholate was added, and the cells were gently swirled for another half hour. Finally, 1 mL of 0.5M MgCl 2 was added to bind the free EDTA, and the mix was incubated for another 15 min prior to pelleting the spheroplasts. The supernatant was poured into a 50 mL conical tube containing 5 mL of 10× IMAC buffer (0.2 M Na 2 HPO 4 , 4 M NaCl, 0.2 M imidazole, pH 7.5) and 0.5 mL of Ni Sepharose (GE Healthcare) that had been washed in 1× IMAC. The mixture was tumbled at least 1 h at 4°C on a tube rotisserie. The resin was washed twice with 25 mL 1× IMAC buffer, and then the resin was incubated with 5 mL of PBS containing 0.2% Triton (TX)-114 for 1 h. Between 0.5 and 1 mL of the nickel resin was loaded onto a column and washed with PBS 0.2% TX-114 (10 mL) followed by two washes, 5 mL each, of 1× IMAC buffer to remove detergent. SdAb constructs were eluted from the nickel resin using 2 mL of elute buffer (1× IMAC plus 250 mM Imidazole) and collected into a 2 mL tube. The eluted material was applied straight onto a MEP HyperCel resin (Sartorius; 2 mL) column that had been washed with 1 mL of NaOH followed by endotoxin-free water and then endotoxin-free PBS. Once the sample was loaded, the MEP HyperCel column was washed with 30 mL of endotoxin-free PBS. The sdAb was eluted with 50 mM Na-Citrate pH 3.5, and sdAb containing fractions determined via monitoring OD280. The sample was neutralized using 10% vol of 1.0 M Tris-HCl pH 7.5. The neutralized elute from the MEP HyperCel was loaded onto a column containing 1 mL of fresh nickel resin that had been washed with endotoxin-free PBS. The sdAb was eluted slowly using elute buffer. For the final steps of purification, the FPLC system was equilibrated with endotoxin-free PBS using a SEC 650 column that has been kept endotoxin-free.
Purified sdAb was frozen for later use or filtered through a Mustang E filter (Pall), to remove residual lipids/endotoxin prior to freezing. The yield of the sdAb was deter mined by UV spectroscopy using a Nanodrop (Thermo). Extinction coefficients can be calculated by an on-line tool such as ExPASy (50) or approximated to be the same as a traditional IgG (1.4).
When purifying the CA-abd and BA-abd constructs, we utilized pyrogen-free pipette tips and pyrogen-free tubes in an effort to minimize contamination of the protein preparations with endotoxins. These constructs were purified on a SEC 650 10 × 300 column that was kept endotoxin-free. Before purification, the column and loop were cleaned by loading 1 mL of 1M NaOH into the loop and then manually injecting it onto the column and running 30 mL of endotoxin-free PBS through the column until the NaOH eluted. The loop was extensively rinsed with endotoxin-free water prior to loading the sample. Protein was collected into pyrogen-free microfuge tubes.
## Measuring endotoxin levels
The Charles River Endosafe-PTS instrument was used to measure endotoxin levels in sdAb preparations in accordance with manufacturer instructions.
## PK studies
Mice were administered 200 µg of either CA-abd or BA-abd antibody intraperitoneally. At 0.5, 2, 4, 8, and 24 h post-injection, sera and brains were collected (n = 4/time point) for a plaque reduction neutralization assay (PRNT). Sera and brains from two additional mice for each antibody evaluated were used as a negative control for the assays. Sera were diluted initially 1:10 in minimum essential medium (MEM) with 2% heat-inactivated (HI)-FBS, 1% HEPES, and 2% Pen/Strep, then serial 1:2 dilutions were made. VEEV-TrD stocks were diluted to a concentration of 2.0 × 1,000 PFU/mL and added 1:1 to the serially diluted samples (resulting in a 1:20 initial dilution) or control wells containing media alone for the virus only control. The entire mouse brain was homogenized and diluted 1:1 with virus. Samples were incubated overnight at 4°C. Six-well plates of VERO76 cells were infected with 0.1 mL of each serial dilution per well in duplicate, and plates were incubated at 37°C for 1 h. After 1-h incubation, cells were overlaid with 0.6% agarose in Basal Medium Eagle (BME) with 10% HI-FBS and 2% Pen/Strep, and incubated for ~24 h at 37°C, 5% CO 2 . A second overlay containing 0.6% agarose in BME with 10% HI-FBS, 2% Pen/Strep, and 4% of total volume neutral red vital stain was added to the wells and further incubated for 18-28 h to visualize plaques. Plaques were counted following incubation with stain overlay. The virus-only control was counted, and the endpoint titer was determined to be the highest dilution with ≥80% reduction (PRNT80) or ≥50% reduction (PRNT50) in the number of plaques observed relative to virus-only control wells. Neutralization titer was reported as the reciprocal of dilution for the serum. The limit of detection for serum was 1:20. VEEV ATCC hyperimmune mouse ascites fluid was utilized as a positive control. Normal human serum was used as a negative control.
## In vivo imaging system (IVIS)
Two milligrams of the CA-abd was labeled with VivoTag 680 XL (PerkinElmer) following the manufacturer's protocol with a few adjustments. Briefly, we used a dye to protein ratio of ~3:1 for the labeling; 220 µg dye was brought up in DMSO and added to the CA-abd. After incubating for 30 min at room temperature, we added 20% vol of 0.1 M borate pH 9 and incubated for an additional 30 min. The reaction was stopped with 10 µl of 1 M ethanolamine. To separate the dye labeled CA-abd from free dye, the sample was added to a Zeba spin column (Pierce), which had been washed twice with endotoxin-free water by gravity, then three times with endotoxin-free PBS by centrifugation. The final dye to protein ratio was determined to be ~2 from the UV/visible spectra of the labeled material and calculated following the manufacturer's protocol. Labeled CA-abd was quick frozen on dry ice and stored at -80°C until needed. PRNT values for the labeled CA-abd were confirmed to be similar to that of the unlabeled antibody before IVIS imaging on the IVIS Spectrum CT (PerkinElmer). Mice were administered 200 µg of the labeled CA-abd intraperitoneally (IP). At 0.5, 2, 4, 8, 24, and 29 h post-administration, the mice were anesthetized with isoflurane and imaged. Living Image Software (PerkinElmer) was used to analyze the images. Region of interest (ROI) of the same size was used to calculate the total radiant efficiency ([P/s]/[µW/cm²]) for both the whole body and head only.
## Efficacy studies
Mice were anesthetized with isoflurane and then subcutaneously (SC) infected with 1,000 PFU of either VEEV INH-9813 or VEEV ZPC738 in the rear footpad. For aerosol exposure to either VEEV TrD or VEEV INH-9813, mice were exposed to a target dose of 1,000 PFU using a whole-body exposure system. The aerosol challenge was generated using a Collison Nebulizer to produce a highly respirable aerosol (flow rate 7.5 ± 0.1 L/ min). The system generates a target aerosol of 1 to 3 µm mass median aerodynamic diameter. The negative control (anti-EBOV IgG [VEEV IC challenges] or E2C2-abd [VEEV ID and VEEV IAB challenges]), the positive control (anti-VEEV IgG 1A3B-7), and sdAbs (CA-abd and BA-abd) mice received a single 200 µg administration IP 1 h post-infection.
Mice were weighed daily and scored for clinical signs of disease at least once daily until reaching a clinical score of ≥3 and then were monitored twice daily. Clinical signs of disease included ruffled fur, hunched posture, lethargy, and neurological signs such as hindlimb paralysis. When animals reached a clinical score of 5, they were humanely euthanized. Survival curve statistics were determined by using the Log-rank (Mantel-Cox) test (GraphPad). A one-sided χ analysis was used to determine differences in overall percentage of survival in each group (GraphPad). Multiple Mann-Whitney tests using Holm-Šídák method and two-way ANOVA analysis were used to determine statistical differences in weight (GraphPad). Multiple Mann-Whitney tests using Holm-Šídák method were used to determine statistical differences in clinical score (GraphPad).
## Viruses
VEEV INH9813 (IC strain) stock was passaged three times on VERO cells. VEEV Trinidad donkey (IAB strain) stock was received from DynPort Vaccine Company (DVC) and prepared by Commonwealth Biotechnologies Inc. VEEV ZPC738 (ID strain) stock was passaged once on BHK cells. VEEV TC83 (vaccine strain) was obtained from BEI (NR-63) and passaged one additional time. The following reagent was obtained through BEI Resources, NIAID, NIH and passaged 1×: VEEV TC83 (IAB vaccine strain; NR-63), UNAV Mac150 (NR-49912), MAYV Guyane (NR-49911), RRV Raratonga (NR-51647), CHIKV LR 2006-OPY1 (Indian Ocean lineage; NR-49741), CHKV PM 2951 (West African lineage; NR-49905), and ONNV UgmP30 (NR-51661). These viruses were obtained from WRCEVA NIAID and passaged 1×: SINV Eg339 (TVP21615) and SFV (TVP20172). VEEV Mucambo (IIIA) was passaged 4× on BHK cells. CHIKV 15561 (Asian lineage) has an unknown passage history.
## Serum viremia plaque assay
Sera were diluted in MEM with 2% HI-FBS, 1% HEPES, and 2% Pen/Strep, and serial (1:10) dilutions were made. Six-well plates of VERO76 cells were infected with 0.2 mL of each serial dilution per well in duplicate wells, and plates were incubated at 37°C for 1 h. After 1-h incubation, cells were overlaid with 0.6% agarose in BME with 10% HI-FBS, and 2% Pen/Strep, and incubated for ~ 24 h at 37°C, 5% CO 2 . A second overlay containing 0.6% agarose in BME with 10% HI-FBS, 2% Pen/Strep, and 4% of total volume neutral red vital stain was added to the wells, and plates were further incubated for 18-28 h for visualization of plaques. Plaques were counted following incubation with stain overlay. Virus stock was utilized as a positive control, and media were used only as a negative control. The limit of detection of the assay is 2.5E + 01 PFU/mL.
## Cryo-EM sample preparation, data collection, and processing
VEEV-V2B3, VEEV-V2C3, and VEEV-V3A8f complexes were prepared by mixing VEEV VLP (TC83 strain) with the sdAb at a 1:2 molar ratio to a final total protein concentration of 0.4 mg/mL. Quantifoil R2/2 mesh 200, 2 nm carbon gold grids were glow-discharged in PELCO easiGlow glow discharge unit (Ted Pella, Inc.) set to air pressure of 0.40 bar, current of 10 mA, and duration of 10 s.
Due to a low concentration of the VLP complex stock solution, before freezing, the grid was saturated by three consecutive applications of 2.7 µL of the sample, followed by edge blotting. The sample was then vitrified in liquid ethane using a FEI Vitrobot Mark IV set to 4°C chamber temperature, 95% chamber humidity, blot time of 1.5 or 2 s, and a blot force of -5. Cryo-EM data were collected on a Titan Krios G1 transmission electron microscope (FEI Company, Inc.) operating at 300 keV, equipped with an Apollo direct electron detection device (Direct Electron, Inc.). Exposures were taken in a movie mode, at a pixel size of 1.11 Å/pix, with a total dose of 40 e -/Å 2 fractionated over 40 raw frames, defocus values set to cycle between -0.75 and -2 mm with SerialEM (51).
All data sets were processed with the cryoSPARC 4.4 software package (52). Movies were aligned and dose-weighted with patch motion correction, and the micrograph contrast transfer function parameters were gaged with patch CTF estimation. Curate exposures were used to filter out bad micrographs. Particles were picked with the blob picker, cleaned with inspect particle picks, extracted from micrographs, subjected to 2D classification, and the best classes were selected. Ab initio reconstruction, heterogeneous, homogeneous, and non-uniform 3D refinements were run with the "I" symmetry imposed. To improve the resolution of the final map, particles were symmetry expanded, and local refinement was performed using a mask covering the region of four spikes clustered around the icosahedral threefold axis and electron density corresponding to the nanobodies. DeepEMhancer (53) was used for map post-processing.
## Model building, refinement, and structural analysis
Homology models of the V2B3, V2C3, and V3A8f sdAbs were gener ated with the Alphafold2 algorithm (54) incorporated into ColabFold (55) server (https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/beta/ AlphaFold2_advanced.ipynb). To obtain initial atomic models of the complexes, previously deposited VEEV structure (PDB: 7FFE) (56) and in silico generated models of the sdAbs were docked into corresponding cryo-EM maps with UCSF Chimera (57). The structures were refined by rounds of real-space refinement in Phenix (58) alternating with model building in Coot (59). Structure validation was performed with Molprobity (60, 61) built-in into the Phenix suite and the wwPDB validation service (https://validatercsb-2.wwpdb.org/). The analysis of VLP-sdAbs interfaces was done with PISA (62) incorporated into the PDBePISA service (https://www.ebi.ac.uk/pdbe/pisa/pistart.html). The assessment of cryo-EM reconstruction quality and model refinement statistics is summarized in Fig. S3 to S5, and Tables S2 to S4.
Local superpositions of the structures were done in Coot. Complementarity-determin ing regions of the sdAbs were mapped with the abYsis key annotation service (http:// www.abysis.org/abysis/index.html). Figures were generated with PyMOL (Schrodinger; http://www.pymol.org) and UCSF Chimera. Structural variation was calculated in Pymol after aligning the structures, sequence variation was calculated as normalized Shannon's entropy among the sequences and done with in-house script.
Binding VLP 96-well ELISA plates (Nunc) were coated with 2 µg/mL of virus-like particles (VLPs; WEEV CBA87 strain, EEEV VLP PE6 strain, VEEV IAB TC83 strain, CHIKV VLP 37997 strain) in PBS, pH 7.4, incubated at 4°C overnight, and blocked with PBS containing 5% skim milk (Difco) and 2% BSA (Fisher Scientific) (blocking buffer) at room temperature for 1 h. Each mAb was serially diluted with 5% skim milk and 2% BSA in PBS with 0.05% Tween-20 (PBST) dilution buffer and was added to the plate and incubated at room temperature for 1 h. After washing, horseradish peroxidase (HRP)-conjugated VHH domain-specific goat anti-alpaca IgG (Jackson ImmunoResearch Laboratories; cat#128-035-232) was added and incubated at room temperature for 1 h. Tetramethylbenzidine (TMB, SeraCare) HRP substrate was added to each well, and the reaction was stopped after 10 min by adding 1 M H 2 SO 4 . The absorbance was measured at 450 nm.
## Virus
96-well ELISA plates (Thermo; Immulon 2HB) were coated with 3 µg/mL of sucrose purified CHIKV strain PM2951 in PBS, incubated at 4°C overnight. Next, coated plates were fixed with 10% formalin for 1 h. After the fixative was removed, plates were washed 3× with PBS + 0.02% Tween-20 (PBST), and blocked with Neptune buffer + 3% normal goat serum at ambient temperature for 5 h. After blocking, plates were washed 3× with PBST. Samples were diluted twofold in blocking buffer starting at 50 µg/mL. Plates were incubated overnight at 4°C. Following incubation, plates were washed 3× with PBST, a secondary anti-alpaca horseradish peroxidase-conjugated antibody (Jackson ImmunoRe search Laboratories; cat#128-035-232) diluted in blocking buffer was added, and plates were incubated for 1 h at ambient temperature. Next, plates were washed 3× with PBST, TMB substrate was added, and plates were incubated for ~5 min at ambient temperature. Finally, the reaction was stopped with Stop Solution, and absorbance was read using a Spectramax M5 instrument set at 450 nm.
## Plaque reduction neutralization assay (PRNT)
## All PRNTs except ONNV
CA-abd and BA-abd were diluted at an initial concentration of 100 µg/mL in MEM with 2% HI-FBS, 1% HEPES, and 2% Pen/Strep, and serial 1:2 dilutions were made. Virus stocks were to a concentration of 2,000 PFU/mL and added 1:1 to the serially diluted samples (resulting in a 50 µg/mL initial dilution) or control wells containing media alone for the virus only control. Samples were incubated overnight at 4°C. Six-well plates of Vero 76 cells were infected with 0.1 mL of each serial dilution per well in duplicate, and plates were incubated at 37°C for 1 h. After 1 h incubation, cells were overlaid with 0.6% agarose in BME with 10% HI-FBS and 2% Pen/Strep and incubated for ~24 h at 37°C, 5% CO 2 . A second overlay containing 0.6% agarose in BME with 10% HI-FBS, 2% Pen/Strep, and 4% of total volume neutral red vital stain was added to the wells and further incubated for 18-28 h to visualize plaques. Plaques were counted following incubation with stain overlay. The virus-only control was counted, and the endpoint titer was determined to be the highest dilution with ≥80% reduction (PRNT80) or ≥50% reduction (PRNT50) in the number of plaques observed relative to virus-only control wells. Neutralization titer was reported as the reciprocal of dilution for the serum. The limit of detection for serum was 50 µg/mL. Virus-specific ATCC hyperimmune mouse ascites fluid was utilized as a positive control, except for UNAV, as an UNAV-specific hyperimmune mouse ascites fluid was not commercially available. For UNAV, a CHIKVspecific ATCC hyperimmune mouse ascites fluid was utilized. ACVE sdAb was used as a negative control.
## ONNV PRNT
CA-abd and BA-abd were diluted at an initial concentration of 100 µg/mL in MEM with 2% HI-FBS, 2% HEPES, 2% L-glutamine/GlutaMAX, and 2% Pen/Strep and then serially diluted (1:2). Virus stocks were diluted to a concentration of 2.0 × 10 3 PFU/mL and added 1:1 to the serially diluted samples (resulting in a 50 µg/mL initial dilution) or control wells containing media alone for the virus only control. Samples were incubated overnight at 4°C. Six-well plates of Vero CCL-81 cells were infected with 0.1 mL of each serial dilution per well in duplicate, and plates were incubated at 37°C for 1 h. After 1 h incubation, cells were overlaid with 0.6% agarose in BME with 20% HI-FBS, 2% non-essential amino acids, 2% L-glutamine/GlutaMAX, and 2% Pen/Strep, and incubated for ~48 h at 37°C, 5% CO 2 . A second overlay containing 0.6% agarose in BME with 10% HI-FBS, 2% non-essential amino acids, 2% L-glutamine/GlutaMAX, 2% Pen/Strep, and 4% of total volume neutral red vital stain was added to wells, and the plates were further incubated for 18-28 h for visualization of plaques. Plaques were counted following incubation with stain overlay. The virus-only control was counted, and the endpoint titer of test sera was determined to be the highest dilution with ≥80% reduction (PRNT80) or ≥50% reduction (PRNT50) in the number of plaques observed relative to virus-only control wells. Neutralization titer was reported as the reciprocal of dilution for the serum. The limit of detection for serum was 50 µg/mL. A CHIKV-specific ATCC hyperimmune mouse ascites fluid was utilized for a positive control as an ONNV-specific hyperimmune mouse ascites fluid was not commercially available. ACVE sdAb was used as a negative control.
## References
1. Goldman "Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Resources, Supervision | Crystal W. Burke, Conceptualization, Formal analysis"
2. Muyldermans (2021) "A guide to: generation and design of nanobodies" *FEBS J*
3. Jovčevska, Muyldermans (2020) "The therapeutic potential of nanobodies" *BioDrugs*
4. Moliner-Morro, Mcinerney, Hanke (2022) "Nanobodies in the limelight: multifunctional tools in the fight against viruses" *J Gen Virol*
5. Wu, Jiang, Ying (2017) "Single-domain antibodies as therapeutics against human viral diseases" *Front Immunol*
6. Liu, Webb, Zabetakis et al. (2021) "Stabilization of a broadly neutralizing anti-chikungunya virus single domain antibody" *Front Med*
7. Liu, Zabetakis, Gardner et al. (2022) "Bivalent single domain antibody constructs for effective neutralization of Venezuelan equine encephalitis" *Sci Rep*
8. Hultberg, Temperton, Rosseels et al. (2011) "Llama-derived single domain antibodies to build multivalent, superpotent and broadened neutralizing anti-viral molecules" *PLoS One*
9. Gaiotto, Ramage, Ball et al. (2021) "Nanobodies mapped to cross-reactive and divergent epitopes on A(H7N9) influenza hemagglutinin using yeast display"
10. Laursen, Friesen, Zhu et al. (2018) "Universal protection against influenza infection by a multidomain antibody to influenza hemagglutinin" *Science*
11. Esparza, Martin, Anderson et al. (2020) "High affinity nanobodies block SARS-CoV-2 spike receptor binding domain interaction with human angiotensin converting enzyme" *Sci Rep*
12. Hanke, Perez, Sheward et al. (2020) "An alpaca nanobody neutralizes SARS-CoV-2 by blocking receptor interaction" *Nat Commun*
13. Zhao, He, Sun et al. (2018) "A novel nanobody targeting middle east respiratory syndrome coronavirus (MERS-CoV) receptor-binding domain has potent cross-neutralizing activity and protective efficacy against MERS-CoV" *J Virol*
14. Darling, Sherwood, Hayhurst (2017) "Intracellular crosslinking of filoviral nucleoproteins with xintrabodies restricts viral packaging" *Front Immunol*
15. Steele, Twenhafel (2010) "REVIEW PAPER: pathology of animal models of alphavirus encephalitis" *Vet Pathol*
16. Weaver, Ferro, Barrera et al. (2004) "Venezuelan equine encephalitis" *Annu Rev Entomol*
17. Goodchild, Brien, Steven et al. (2011) "A humanised murine monoclonal antibody with broad serogroup specificity protects mice from challenge with Venezuelan equine encephalitis virus" *Antiviral Res*
18. O'brien, Goodchild, Phillpotts et al. (2012) "A humanised murine monoclonal antibody protects mice from Venezuelan equine encephalitis virus, everglades virus and mucambo virus when administered up to 48 h after airborne challenge" *Virology (Auckl)*
19. Burke, Froude, Rossi et al. (2026) *Full-Length Text Journal of Virology*
20. Glass (2019) "Therapeutic monoclonal antibody treatment protects nonhuman primates from severe Venezuelan equine encephalitis virus disease after aerosol exposure" *PLoS Pathog*
21. Hoefman, Ottevaere, Baumeister et al. (2015) "Preclinical intravenous serum pharmacokinetics of albumin binding and non-half-life extended nanobodies®" *Antibodies (Basel)*
22. Roovers, Vosjan, Laeremans et al. (2011) "A biparatopic anti-EGFR nanobody efficiently inhibits solid tumour growth" *Int J Cancer*
23. Stork, Müller, Kontermann (2007) "A novel tri-functional antibody fusion protein with improved pharmacokinetic properties generated by fusing a bispecific single-chain diabody with an albumin-binding domain from streptococcal protein G" *Protein Eng Des Sel*
24. Aguilar, Estrada-Franco, Navarro-Lopez et al. (2011) "Endemic Venezuelan equine encephalitis in the Americas: hidden under the dengue umbrella" *Future Virol*
25. Powers, Brault, Shirako et al. (2001) "Evolutionary relationships and systematics of the alphaviruses" *J Virol*
26. Rusnak, Glass, Weaver et al. (2019) "Approach to strain selection and the propagation of viral stocks for Venezuelan equine encephalitis virus vaccine efficacy testing under the animal rule" *Viruses*
27. Phillpotts (2006) "Venezuelan equine encephalitis virus complexspecific monoclonal antibody provides broad protection, in murine models, against airborne challenge with viruses from serogroups I, II and III" *Virus Res*
28. Weaver (2005) "Host range, amplification and arboviral disease emergence" *Arch Virol Suppl*
29. Rusnak, Dupuy, Niemuth et al. (2018) "Comparison of aerosol-and percutaneous-acquired venezuelan equine encephalitis in humans and nonhuman primates for suitability in predicting clinical efficacy under the animal rule" *Comp Med*
30. Bronze, Huycke, Machado et al. (2002) "Viral agents as biological weapons and agents of bioterrorism" *Am J Med Sci*
31. Randall, Mills (1944) "Fatal encephalitis in man due to the venezue lan virus of equine encephalomyelitis in trinidad" *Science*
32. Meechan, Potts (2020) "Biosafety in microbiological and biomedical laboratories"
33. Dembek (2016) "USAMRIID's medical management of biological casualties handbook"
34. Sutton, Pletnev, Callahan et al. (2023) "Vaccine elicitation and structural basis for antibody protection against alphaviruses" *Cell*
35. Kim, Kafai, Winkler et al. (2021) "Pan-protective antialphavirus human antibodies target a conserved E1 protein epitope" *Cell*
36. Williamson, Gilliland, Yadav et al. (2020) "Human antibodies protect against aerosolized eastern equine encephalitis virus infection" *Cell*
37. Burke, Froude, Miethe et al. (2018) "Human-like neutralizing antibodies protect mice from aerosol exposure with western equine encephalitis virus" *Viruses*
38. Vugmeyster, Entrican, Joyce et al. (2012) "Pharmacokinetic, biodistribution, and biophysical profiles of TNF nanobodies conjugated to linear or branched poly(ethylene glycol)" *Bioconjug Chem*
39. Jin, Odongo, Radwanska et al. (2023) "NANOBODIES: a review of diagnostic and therapeutic applications" *Int J Mol Sci*
40. Guzmán-Terán, Rangel, Rodriguez-Morales et al. (2020) "Venezuelan equine encephalitis virus: the problem is not over for tropical America" *Ann Clin Microbiol Antimicrob*
41. Yu, Wiley, Kugelman et al. (2015) "Complete coding sequences of eastern equine encephalitis virus and venezuelan equine encephalitis virus strains isolated from human cases" *Genome Announc*
42. Weaver, Anishchenko, Bowen et al. (2004) "Genetic determinants of Venezuelan equine encephalitis emergence" *Arch Virol*
43. Greene, Paessler, Austgen et al. (2005) "Envelope glycoprotein mutations mediate equine amplification and virulence of epizootic venezuelan equine encephalitis virus" *J Virol*
44. Kafai, Williamson, Binshtein et al. (2022) "Neutralizing antibodies protect mice against Venezuelan equine encephalitis virus aerosol challenge" *J Exp Med*
45. O'brien, Underwood-Fowler, Goodchild et al. (2009) "Development of a novel monoclonal antibody with reactivity to a wide range of Venezuelan equine encephalitis virus strains" *Virol J*
46. Roehrig, Mathews (1985) "The neutralization site on the E2 glycoprotein of Venezuelan equine encephalomyelitis (TC-83) virus is composed of multiple conformationally stable epitopes" *Virology (Auckl)*
47. Rülker, Voß, Thullier et al. (2012) "Isolation and characterisation of a human-like antibody fragment (scFv) that inactivates VEEV in vitro and in vivo" *PLoS One*
48. Gardner, Sun, Luke et al. (2017) "Antibody preparations from human transchromosomic cows exhibit prophylactic and therapeutic efficacy against Venezuelan equine encephalitis virus" *J Virol*
49. Hunt, Bowen, Frederickson et al. (2011) "Treatment of mice with human monoclonal antibody 24 h after lethal aerosol challenge with virulent Venezuelan equine encephalitis virus prevents disease but not infection" *Virology (Auckl)*
50. Phillpotts, Jones, Howard (2002) "Monoclonal antibody protects mice against infection and disease when given either before or up to 24 h after airborne challenge with virulent Venezuelan equine encephalitis virus" *Vaccine (Auckl)*
51. Williamson, Reeder, Bailey et al. (2021) "Therapeutic alphavirus crossreactive E1 human antibodies inhibit viral egress" *Cell*
52. Gasteiger, Gattiker, Hoogland et al. (2003) "ExPASy: the proteomics server for in-depth protein knowledge and analysis" *Nucleic Acids Res*
53. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
54. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
55. (2026) *Full-Length Text Journal of Virology*
56. Sanchez-Garcia, Gomez-Blanco, Cuervo et al. (2021) "DeepEMhancer: a deep learning solution for cryo-EM volume post-processing" *Commun Biol*
57. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
58. Mirdita, Schütze, Moriwaki et al. (2022) "ColabFold: making protein folding accessible to all" *Nat Methods*
59. Ma, Huang, Ma et al. (2021) "Structure of Venezuelan equine encephalitis virus with its receptor LDLRAD3" *Nature*
60. Pettersen, Goddard, Huang et al. (2004) "UCSF chimera--a visualization system for exploratory research and analysis" *J Comput Chem*
61. Liebschner, Afonine, Baker et al. (2019) "Coot: model-building tools for molecular graphics" *Acta Crystallogr D Biol Crystallogr*
62. Emsley, Cowtan (2004) "Coot: model-building tools for molecular graphics" *Acta Crystallogr D Biol Crystallogr*
63. Davis, Murray, Richardson et al. (2004) "MOLPROBITY: structure validation and all-atom contact analysis for nucleic acids and their complexes" *Nucleic Acids Res*
64. Chen, Arendall, Iii et al. (2010) "MolProbity: all-atom structure validation for macromolecular crystallography" *Acta Crystallogr D Biol Crystallogr*
65. Krissinel, Henrick (2007) "Inference of macromolecular assemblies from crystalline state" *J Mol Biol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12802214&blobtype=pdf | # Dynamics of B-cell response in MERS-CoV patients and survivors with hybrid immunity
Hebah Al-Khatib, Fatma Ali, Hadeel Zedan, Maria Smatti, Peter Coyle, Sara Taleb, Ali Hssain, Asmaa Al-Thani, Hadi Yassine
## Abstract
Middle East respiratory syndrome coronavirus (MERS-CoV) causes a highly lethal respiratory infection for which no vaccines or antiviral therapeutics are cur rently available. Understanding the immune response is critical for designing effective therapeutics. Here, we comprehensively characterized the dynamics of B-cell responses in severely infected MERS-CoV patients and survivors with SARS-CoV-2 exposure history. Infected patients developed robust neutralizing antibody responses within 1 month of illness, with moderate-to-high cross-neutralization activity against SARS-CoV-2. The enhanced neutralization activity coincided with an increased abundance of specific mutated, class-switched IgG clones. Notably, one such clone was detected at moderate prevalence in both patients, and its expansion was accompanied by high neutraliza tion activity against both viruses. Conversely, MERS-CoV survivors demonstrated higher neutralization activity against MERS-CoV after vaccination, despite minimal changes in antibody titers and limited alterations in B-cell repertoire properties. This suggests that the enhanced neutralization activity may be mediated by the reactivation and expansion of cross-reactive memory B cells targeting conserved epitopes, originally generated in response to the virus that triggered the primary immune response. These findings provide valuable insights into the B-cell repertoire landscape during natural MERS-CoV infection and highlight the potential for identifying broadly neutralizing antibodies in individuals with hybrid immunity.
IMPORTANCEThis study examines the immune responses of MERS-CoV patients and survivors who have had confirmed exposure to SARS-CoV-2. It offers a unique oppor tunity to characterize cross-reactive B-cell responses in individuals possessing hybrid immunity to both pathogenic coronaviruses. To our knowledge, no previous studies have examined longitudinal changes in the B-cell repertoire in MERS-CoV patients or survivors before and after SARS-CoV-2 vaccination. Our findings reveal enhanced neutralization activity against both MERS-CoV and SARS-CoV-2 following infection or vaccination, which appears to be associated with distinct patterns of B-cell repertoire dynamics. Notably, the data strongly suggest the presence of potent cross-neutralizing antibody responses, particularly in MERS-CoV patients, driven by dominant B-cell clones. These results underscore the potential for identifying broadly neutralizing antibodies in individuals with hybrid immunity.
## MATERIALS AND METHODS
## Study groups
Blood samples were collected from two MERS-CoV-infected patients (group I [GI]) and four MERS-CoV survivors (group II [GII]). Serial blood samples were collected from group I patients during a month of infection, while blood samples from group II individuals were collected before and after receiving the second dose of SARS-CoV-2 mRNA vaccine (Fig. 1A; Table S1). A detailed demographic and clinical description of the two MERS-CoV cases is published on the WHO website (19). Briefly, both cases were reported in Qatar in 2022. Both cases had frequent close contact with dromedary camels and consumption of camel's milk. The first patient (M1, 50-year-old male) presented to the emergency department with cough, fever, and shortness of breath. The second patient (M2, 85-year-old male) was admitted to the emergency department with symptoms similar to those presented in the first case. On 22 March, the patient was transferred to the ICU, where he stayed for 24 days before passing away on 14 April. MERS-CoV infection was confirmed in both patients by RT-PCR targeting upE and orf1a genes. Six blood samples were collected from the first patient and two from the second patient.
## Processing of blood samples
Whole blood samples were collected into EDTA tubes. Sera were separated, aliquoted, and stored at -80°C. PBMCs were isolated using the Ficoll-Paque Plus reagent (Cytiva, USA) as per the manufacturer's protocol. In brief, a whole blood sample was diluted and layered on top of Ficoll-Paque reagent. This was followed by a centrifugation step at 400 × g for 30-40 minutes at room temperature. The buffy coat layer was aspirated and rinsed twice with 1× phosphate-buffered saline (PBS). Cells were resuspended in complete RPMI-1640 medium, counted, and subjected to a final round of centrifugation. Cell pellets were then resuspended in fetal bovine serum containing 10% DMSO and stored in the liquid nitrogen (5 × 10 6 cells per vial).
## Serological analysis
## ELISA
Titers of anti-MERS-CoV and anti-SARS-CoV-2 antibodies were measured by enzymelinked immunosorbent assay (ELISA) using full-length spike, S1-RBD, and S2 proteins (SinoBiological, USA). Proteins were coated onto ELISA plates (Thermo Fisher Scientific, USA) for 24 hours. Plates were washed and incubated with blocking buffer containing 1% bovine serum albumin for 1 hour at 37°C. Plates were washed with PBS containing 0.1% Tween-20. Serially diluted samples were then added to the coated wells and incubated for 1 hour at 37°C. Horseradish peroxidase-labeled polyclonal rabbit antibody targeting human IgG, IgM, and IgA (Abcam, USA) was added to each well and incubated for an additional 1 hour. KPL SureBlue TMB Microwell Peroxidase Substrate was then added and incubated for 5-10 minutes (Sera Care, USA). The plates were read using the BioTek Cytation 5 Microplate Reader (Agilent, USA). Antibody titers were calculated using the endpoint analysis, and cut-off values were determined to be three times the average of the blank reading. All experiments were conducted in triplicate.
## Pseudovirus neutralization assay
The neutralization activity was measured using MERS-CoV and SARS-CoV-2 (Wuhan-hu-1) pseudoviruses (PVs) generated by VSV pseudotyping systems as previously described (20,21). Cells and plasmids were kindly provided by Viral Pathogenesis Laboratory, Vaccine Research Center, National Institutes of Health. In brief, Huh7.5 and HEK293T-ACE2 cells-for MERS-CoV and SARS-CoV-2, respectively-were seeded in a 96-well white/black isoplate (PerkinElmer, USA). Heat-inactivated serum samples were diluted (twofold dilution) and incubated with the pseudovirus for 2 hours at 37°C, then added to the cells. Twenty-four hours later, cells were lysed, and 50 µL of Luciferase substrate was added to each well to measure PV entry (Bio-Glo Luciferase Assay System; Promega, USA). Luciferase activity was measured using a luminescence plate reader (Tecan Infinite 200 PRO), and the percentage of inhibition was calculated for each sample.
## B-cell repertoire analysis
## Sequencing of BCR
Total RNA was extracted from 5 × 10 6 cells using the RNeasy Mini kit (Qiagen, USA) according to the manufacturer's instructions. The quality of extracted RNA was evaluated using Agilent Bioanalyzer 2100. Amplification and sequencing of the heavy chains of B-cell receptors were performed using the iRepertoire iR-Complete Dual Index Primer Kit (iRepertoire, USA), as described in the manufacturer's protocol (Fig. 1B). In brief, reverse transcription was conducted using 400 ng of RNA, followed by dual-index PCR for sample barcoding. PCR products were purified, quality-checked with the High-Sensitivity DNA Kit (Agilent), and pooled in equimolar amounts (150 ng per sample). Libraries were quantified (JetSeq Lo-ROX kit), spiked with 10%-15% PhiX, and sequenced as paired-end reads (250 × 250 bp) using the MiSeq kit (v.2). A negative control was included in the run to ensure a contamination-free library preparation process. All samples were sequenced in duplicates. Unfortunately, the number of sequences obtained from time point 4 in M1 was not enough to conduct the BCR analysis.
## Processing of B-cell receptor sequences
Raw sequencing reads were processed using the iRweb analysis pipeline. Pre-processing sequencing reads included trimming low-quality ends and stitching the paired-end reads. Stitched reads, with an average overlapping region of 120 bp, that are not 100% identical within the overlap region were discarded. Reads were then collapsed based on nucleotide sequence identity. The stitched reads (~500 bp) were then submitted to IMGT/HighV-QUEST for V(D)J germline assignment, immunoglobulin gene use, and sequence annotation. Output results were filtered to remove non-productive reads using the Change-O toolkit (v.1.3.0) (22). Novel V genotypes were called using the TIgGER R package and used to correct V allele calls generated by the IMGT database. Samples collected at different time points from the same patient were clustered together to allow easy comparison of clusters over time. Clonotyping was performed using the Change-O toolkit by applying the cut-off value determined in the Shazam R package (v.1.1.2). Shazam R package was used to estimate the optimal distance threshold that separates clonally related from unrelated sequences using single nucleotide Hamming distance model (22). Clonotype size was inferred from the number of sequences in each unique clonotype. Clonotypes containing only one read were considered to result from sequencing bias and were removed before the subsequent analysis (except for diversity analysis).
## Clonal diversity
B-cell repertoire diversity was estimated using VDJtools (23). Diversity was estimated using the normalized data sets by resampling to the size of the smallest data set. The diversity of normalized data sets was presented as Shannon-Wiener index (diversity) and inverse Simpson index (dominance) to measure the number of unique clones and their abundance in each sample. Clonal diversity was also estimated using the D50 index as a quantitative measure of clonal expansion.
## Somatic hypermutations and clone convergence
Somatic hypermutation (SHM) frequency-including replacement and silent muta tions-per unique VDJ region per time point and isotype were calculated using the CDR region for each sample using the observedMutation function within the SHa zaM package (22). Convergent clones were identified based on shared V and J gene segments, identical CDR3 length, and a minimum of 85% amino acid sequence identity within the CDR3 region.
## Statistical analysis
Correlation analyses were conducted and visualized using the Spearman correlation test in Prism 10. Statistical significance was assessed using non-parametric tests, including the Wilcoxon test, Kruskal-Wallis test, or Mann-Whitney test, as appropriate.
## RESULTS
## Virus-specific antibody titers correlate with neutralization activity
Analysis of MERS-CoV antibody titers in both groups was quantified using the full-length S, RBD, and S2 subunit. In GI patients, anti-MERS-CoV antibodies targeting all three protein domains were detected at all time points. The first case (M1) exhibited peak antibody titers against all proteins 20 days post-symptom onset, which remained elevated for the subsequent 10 days. Similarly, the second case (M2) maintained high and stable antibody titers 1 month after symptom onset. Notably, the anti-S IgG response in M1 was predominantly driven by anti-S2 antibodies, while M2 showed a combined anti-RBD and anti-S2 response. In contrast, GII participants showed modest changes in anti-MERS-CoV antibody titers following SARS-CoV-2 vaccination, regardless of antibody isotype (Fig. 2A).
Given the reported cross-reactivity among betacoronaviruses, we also assessed anti-SARS-CoV-2 antibody levels in both groups (Fig. S1A). GI patients exhibited detectable anti-SARS-CoV-2 spike antibodies at all time points, albeit at lower levels compared to MERS-CoV antibodies. Anti-RBD antibody levels remained stable over time, irrespective of isotype. Interestingly, anti-S2 IgG antibodies showed a gradual increase, peaking around 20 days post-infection, mirroring the pattern observed for anti-MERS-CoV IgG antibodies. The anti-SARS-CoV-2 antibody response was primarily driven by anti-S2 antibodies, with a comparatively weak anti-RBD response observed in both cases. Correlation analysis of MERS-CoV and SARS-CoV-2 antibody titers in GI patients revealed a positive correlation for both IgM (rs = 0.85, P = 0.023) and IgA (rs = 0.86, P = 0.023) antibodies targeting the S2 regions of both viruses (Fig. 2B).
As anticipated, GII participants demonstrated increased anti-SARS-CoV-2 antibody titers following vaccination. This response was primarily mediated by IgG and, to a lesser extent, IgA antibodies, predominantly targeting the S2 protein (Fig. S1A). However, the elevated anti-S2 antibody titers in the GII group did not show a significant correlation with anti-S2 titers against MERS-CoV (Fig. 2B). The enhanced antibody response against the SARS-CoV-2 S2 protein can be attributed to the amino acid sequence homology in the S2 subunit between MERS-CoV and SARS-CoV-2.
We further assessed the antibody function using the pseudovirus neutralization assay. Particularly, we explored the neutralization capacity of antibodies against MERS-CoV and SARS-CoV-2 pseudoviruses (Fig. 2A). In GI patients, the neutralization activity against MERS-CoV pseudovirus mirrored the antibody titer patterns observed in ELISA. The neutralization activity increased gradually and reached its highest level after 20 days of symptom onset in M1 (87%). The neutralization activity decreased slightly at later time points but remained relatively high (more than 75%). Similarly, M2 demonstrated a remarkably high MERS-CoV neutralization activity after a month of symptom onset. Group I patients have also shown a moderate-to-high neutralization activity against SARS-CoV-2 pseudovirus at all time points. In M1, SARS-CoV-2 neutralization activity was comparable to MERS-CoV in the first two time points and stabilized at 65% in later samples. M2 exhibited a dramatic increase in SARS-CoV-2 neutralization activity from 51% at the first time point to 81% at the second time point. In GII participants, neutraliza tion activity increased two-to fivefold against both viruses following SARS-CoV-2 vaccination. This increase was comparable for both pseudoviruses, with MERS-CoV neutralization reaching 61% and 48% in M5 and M6 samples, respectively (Fig. 2A).
To evaluate the relative quality of antibodies induced by MERS-CoV infection and SARS-CoV-2 vaccination, we analyzed the relationship between antibody titers and neutralization activity. Significant positive correlations were observed between MERS-CoV IgG titers and MERS-CoV neutralization activity in GI patients (full S: r s = 0.92, P = 0.006; S2: r s = 0.93, P = 0.007) (Fig. 2C), and between SARS-CoV-2 IgG titers and SARS-CoV-2 neutralization levels in GII participants (full S: r s = 0.88, P = 0.007; S2: r s = 0.81, P = 0.021) (Fig. S1B). This pattern suggests that antibody titers may serve as indicators of neutralizing capacity, but the relationship appears to be specific to the primary antigenic stimulus.
Finally, we performed a correlation analysis between isotype usage derived from repertoire analysis and serum immunoglobulin titers obtained from ELISA. Overall, no significant correlations were seen between isotype usage and antibody titers in both groups. A weak positive correlation was seen between proportions of IgM and IgA B cells and corresponding anti-S and anti-S2 antibody titers in GII participants. The lack of correlation between BCR repertoire and serum titers, particularly in infected patients (GI), is likely to be due to differences in cellular and immunoglobulin kinetics. Following infection, IgG takes time to build up in serum, lagging behind the cellular response. Together, these findings indicate that the correlation between serum Ig and BCR repertoire may not occur, particularly in the setting of acute disease (Fig. S2).
## Reduced clonal diversity in infected patients compared to vaccinated survivors
Assessment of the clonal diversity was performed after subsampling to correct for differences in sequencing depth (Fig. 3). Diversity was assessed using Shannon index (richness and evenness), Simpson's index (clonal dominance), and D50 index (domi nant clone contribution). Overall, vaccinated individuals (GII) exhibited higher B-cell diversity compared to infected patients (GI) regardless of sample collection time (Fig. 3A). Simpson's index was then used to assess clonal expansion during infection (GI) or following vaccination (GII) due to its high sensitivity to clonal abundance. Analysis of B-cell diversity in GI patients revealed a relatively low B-cell diversity in M1 during the first 20 days of infection, which was accompanied by increased antibody titers and neutralization activity. Similar to M1, M2 demonstrated high diversity after a month of symptom onset (Fig. 3B). Vaccinated participants (GII) also exhibited a modest decrease in B-cell diversity following vaccination (Fig. 3B). This reduction was accompanied by enhanced neutralization activity against both MERS-CoV and SARS-CoV-2.
## Dynamic B-cell isotype switching during MERS-CoV infection
To investigate the kinetics of B-cell responses during MERS-CoV infection and following SARS-CoV-2 vaccination, we examined the proportions of B-cell isotypes at various time points (Fig. 4). Analysis of B-cell isotypes proportions was performed by counting each unique CDR3 sequence once to avoid bias from clonal expansion. Overall, no significant differences were observed between groups, nor were there notable changes in GII following vaccination (Fig. 4A). The longitudinal analysis in GI revealed dynamic shifts, particularly in IgG and IgM proportions (Fig. 4B). The proportions of these two isotypes showed an inverse trend over time, consistent with typical infection-associated class switching, where IgM levels decrease as IgG levels increase. The rise in IgG clones (3-fold increase at TP3 in M1 and 30-fold increase at TP2 in M2) coincided with increased antibody titers and enhanced neutralization against both viruses. IgA clones were detected at early time points but declined thereafter and were undetectable after 1 month in both patients. In contrast, GII showed less pronounced changes. As expected, IgG clone proportions increased post-vaccination (1.9-fold increase), accompanied by a modest rise in IgA and a decline in IgM. Proportions of isotype-switching clones increased over time in GI. In M1, switching to IgA dominated at TP2 (68%), while IgG dominated at TP3 (68%). In M2, class-switched clones increased to 82% after a month primarily to IgG. In GII, vaccination led to a predominance of IgG-switched clones (Fig. 4C).
## Distinct V gene usage and clonal expansion patterns in GI and GII
To identify potentially enriched V genes, the relative frequencies (number of clones) and clonal abundance (clone size) were analyzed. Clones were derived from 38 unique V genes in GI patients and 55 in GII participants. The most frequently utilized genes in both groups were IGHV3-30, IGHV4-59, IGHV1-18, and IGHV3-23 (Fig. 4D).
Longitudinal analysis of GI patients revealed dynamic changes in clonal abundance during infection. Both patients demonstrated expansion of IGHV3-43 and IGHV4-34 clones at time points corresponding to peak neutralization activity against both viruses. In M1, increased neutralization also coincided with increased abundance of IGHV1-69, IGHV3-7, IGHV3-15, IGHV3-30, IGHV3-53, IGHV4-4, and IGHV4-61 clones. In M2, the expansion of IGHV4-59 clones was particularly associated with stronger SARS-CoV-2 neutralization. In contrast, no major changes in V gene usage were observed in GII following vaccination. Moreover, longitudinal analysis of GII participants did not reveal uniform changes in clonal abundance following vaccination. Instead, neutralization activity in GII participants was linked to the expansion of distinct B-cell clones, including IGHV1-46 in M4 and M6, IGHV4-34 in M5, and IGHV3-43/IGHV3-7 in M6 (Fig. 4E).
In both groups, samples with higher neutralization activity (>30%) had significantly increased clonal abundance of IGHV3-30-3 (P = 0.04 for both viruses), IGHV3-53 (P = 0.03 for SARS-CoV-2 and P = 0.032 for MERS-CoV), and IGHV3-43 (P = 0.04 for SARS-CoV-2 and P = 0.01 for MERS-CoV) clones (Fig. 4F).
## Low levels of somatic hypermutations in most expanded clones
To assess B-cell maturation, the frequency of replacement SHM in the CDR regions was analyzed (Fig. 5). Overall, there were no significant differences in SHM frequencies between GI (mean = 0.081, SD = 0.0503) and GII (mean = 0.079, SD = 0.024). Moreover, no substantial changes in SHM frequency were observed in GII participants following vaccination. In contrast, longitudinal analysis of GI patients revealed dynamic changes in SHM frequency over time. In M1, the highest SHM frequency was seen after 15 days of symptoms (mean = 10.8%, SD = 0.06). The reduced SHM frequency at later time points suggests active selection within germinal centers. In M2, a 10-fold increase in SHM frequency was seen after a month of symptom onset. Interestingly, the rise in SHM frequency in GI coincided with enhanced neutralization activity against MERS-CoV in M1 and SARS-CoV-2 in M2 (Fig. 5A).
Analysis of SHM at the isotype level revealed dynamic changes in SHM frequency in IgG and IgM clones, while IgA clones maintained relatively stable levels. Among all isotypes, IgG clones exhibited the highest SHM frequencies in both groups (mean = 11%, SD = 5.6%) (Fig. 5B). In GI, the increase in IgG SHM observed after 15 and 26 days of infection in M1 and M2 (P < 0.001), respectively, compared to previous time points, was followed by a significant increase in neutralization activity. Similarly, the increased SHM in IgG clones after vaccination (P < 0.001) in M3, M5, and M6 participants coincided with stronger neutralization activity. Analyzing SHM in IgG clones utilizing different V genes revealed weak correla tions between MERS-CoV neutralization and SHM frequency in most abundant clones, including IGHV3-23, IGHV1-18, IGHV3-30, and IGHV3-21. In contrast, less abundant clones, IGHV3-43, IGHV3-53, IGHV4-4, and IGHV4-34, exhibited increased mutation frequencies at time points corresponding to enhanced neutralization against both viruses in both groups (Fig. 5C). The increased SHM in IGHV3-43 clones, particularly, coincided with higher neutralization activity against both viruses in both groups. Physicochemical analysis of IGHV3-43 clones showed changes indicative of improved binding affinity, including reduced CDR3 length and increased polarity and aliphatic index at time points corresponding to enhanced neutralization (Fig. 5D). In contrast, the increased SHM frequency in several IgG clones coincided with higher SARS-CoV-2 neutralization in GI (Fig. S3A). In M2, the increased SHM frequency in IGHV3-43 and IGHV4-59 IgG clones (10-and 25-fold increases relative to TP1, respectively) coincided with the significant rise in SARS-CoV-2 neutralization. These changes were accompanied by a marked decrease in CDR3 length and increased polarity, suggesting that SHM contributed to increased binding affinity and antigen specificity (Fig. 5D; Fig. S3B).
## Clone convergence between MERS-CoV patients is present but rare
Convergent clonotypes are important for unraveling viral epitopes that commonly induce antibody responses in multiple individuals. Here, we evaluated the presence and abundance of convergent clones among GI patients, GII individuals, and between the two groups (Fig. 6). Overall, GI shared 8% (M1) and 12% (M2) of their sequences with other individuals. The proportion of shared sequences in GII ranged from 3% in M3 and M5 to 8% in M6 (Fig. 6A). GI patients shared 4% of their sequences, while GII participants shared less than 1% of their sequences. M2 particularly shared 11% of sequences with GII participants (Fig. 6B). Pairwise analysis indicated the highest overlap between M1 and M2 (4%), M2 and M4 (5%), and M4 and M6 (3.8%). Comparison with the CoV-AbDab database (n = 12,536 sequences) showed minimal convergence, with shared clones making up only 0.4%-2.7% (mean = 1.2%, SD = 0.01) of total repertoires (Fig. 6C).
Five convergent clones were identified in GI, comprising 6,651 sequences. They belonged to IGHV3-43, IGHV1-69, IGHV3-23, and IGHV1-2. Notably, one of the shared IGHV3-43 clones appeared at multiple time points in both patients (abundance = 7.5% in M1 and 10.7% in M2), suggesting a potential role in the immune response. In GII, 34 convergent low-abundant clones were identified before vaccination, shared only between two individuals each. The majority (22/28) were non-class-switched and disappeared post-vaccination, indicating distinct, individual-specific responses to vaccination. Only eight convergent clones were found between groups, including two low-abundant IgG class-switched clones (Fig. 6D).
## DISCUSSION
MERS-CoV circulates in camels and causes severe sporadic outbreaks in humans, raising concerns about our preparedness for future outbreaks. Despite the extensive research, no vaccines or antiviral therapeutics are currently available. Besides, little is known about the dynamics of B-cell response during the acute phase of MERS-CoV infection. In this study, we characterized the B-cell response in two severe MERS-CoV cases with a history of SARS-CoV-2 exposure, as well as in MERS-CoV survivors both before and after SARS-CoV-2 vaccination.
The development of the immune response following MERS-CoV infection has been described in both acute and convalescent patients (24). In most cases, antibodies develop approximately 21 days after illness onset and can persist for 3-6 years after recovery (7,8,(25)(26)(27). In our study, positive seroconversion was observed in both patients; however, the exact timing of seroconversion could not be determined due to the lack of samples from earlier time points. In M1 (a 50-year-old male), anti-MERS-CoV antibodies were detected on day 12, peaked at day 20, and remained elevated for the following 14 days. These high antibody titers were associated with a significant reduction in viral RNA levels in respiratory samples collected at later time points. In contrast, M2 (an 85-year-old male) exhibited high antibody levels on days 27 and 31 of illness; however, this was accompanied by persistent viral RNA shedding, a manifestation frequently observed in older MERS-CoV patients with underlying comorbidities (24,28).
Cross-reactivity and cross-neutralization between MERS-CoV and SARS-CoV-2 are critical for understanding the role of MERS-CoV immunity in individuals with hybrid immunity (29). Cross-reactivity was clearly demonstrated in ELISA results, which showed that antibodies induced by MERS-CoV infection primarily targeted the S2 region of the SARS-CoV-2 spike protein, with low or undetectable reactivity against the RBD region. This was also evident in vaccinated individuals, as antibody titers against the full-length S and S2 proteins of both viruses showed positive correlations, further supporting enhanced cross-reactivity post-vaccination. These findings are consistent with studies reporting increased titers of cross-reactive antibodies following SARS-CoV-2 vaccination or infection in MERS-CoV survivors (17,18,30,31).
Cross-neutralization activity was observed in both groups. High MERS-CoV neutraliza tion levels in GI were accompanied by lower viral RNA levels. Both patients exhibited moderate-to-high neutralization activity (>50%) against SARS-CoV-2 despite relatively lower anti-SARS-CoV-2 antibodies, suggesting that this neutralization may have been mediated by cross-reactive MERS-CoV antibodies. Similar findings were reported in MERS-CoV survivors who experienced a boost in MERS-CoV neutralization following SARS-CoV-2 vaccination or infection without substantial changes in antibody titers (17,18,32). Similarly, vaccinated survivors exhibited a two-to fivefold increased MERS-CoV neutralization, despite the minimal increase in MERS-CoV antibody titers, suggesting the activation of vaccine-induced cross-reactive B cells. Collectively, these findings under score the potential for identifying broadly neutralizing antibodies in patients with hybrid immunity.
Changes in serum antibody kinetics were associated with dynamic shifts in B-cell repertoire in GI. A dramatic increase in IgG and IgA class-switched clones was seen in M1, a profile commonly seen during infections (33,34). This was accompanied by higher SHM frequency, clonal selection, and enhanced neutralization. The latter stage of infection was characterized by a decline in SHM frequency, stable neutralization levels, and virus clearance. This pattern reflects a transition from active germinal center reactions, which drive SHM and affinity maturation, to long-lived plasma cell production, a phenomenon also reported in other viral infections (35). Unfortunately, similar analyses could not be performed for M2 due to the lack of samples from earlier time points. However, samples collected 27 days post-infection demonstrated clonal expansion of mutated IgG clones accompanied by low diversity and high neutralization. This observation aligns with studies on severe COVID-19 cases, where extrafollicular responses dominate, leading to reduced repertoire diversity and expanded clones with high neutralization potential (36,37). GII, on the other hand, exhibited minimal alterations in B-cell repertoire following vaccination. They showed a slight decrease in diversity, an increase in IgG class-switched clones, and stable SHM frequencies. This suggests that vaccination mobilized antigenexperienced memory B cells without significantly altering overall repertoire properties, a pattern reported in studies on SARS-CoV-2 mRNA vaccines (38).
Analysis of IGHV genes revealed preferential usage of IGHV3-30, IGHV3-23, IGHV1-18, IGHV1-69, and IGHV4-59 genes in both groups, consistent with patterns observed in other viral infections (39)(40)(41). Group-specific analyses revealed distinct patterns of V gene usage. GI exhibited increased expansion of B-cell clones utilizing IGHV3-43 and IGHV4-34 genes that coincided with increased SHM frequency and enhanced neutralization. Both germlines were previously identified in MERS-CoV and SARS-CoV-2neutralizing antibodies, highlighting their role in targeting conserved viral epitopes (39,42). IGHV3-43 was also identified in pan-SARS-CoV-2-neutralizing antibodies (39,42). In M2, the expansion of mutated IgG clones utilizing IGHV4-59 coincided with a marked increase in SARS-CoV-2 neutralization. IGHV4-59 clones were commonly reported following SARS-CoV-2 infection but not in MERS-CoV patients. Studies suggest that IGHV4-59 antibodies target conserved regions within the S2 subunit of SARS-CoV-2 spike protein, enabling cross-reactivity with other betacoronaviruses (43,44).
In GII, distinct patterns emerged across individuals, with higher neutralization activity linked to the expansion of mutated IGHV4-34 clones in M5 and IGHV3-43 clones in M3 and M6. Additionally, the expansion of IGHV1-46, IGHV3-7, IGHV4-4, and IGHV3-53 clones was observed post-vaccination in some individuals. Of these, IGHV1-46 clones were identified following MERS-CoV and SARS-CoV-2 infections and in broadly neutral izing antibodies targeting the conserved S2 stem helix region of all three pathogenic coronaviruses (45,46). These findings suggest that neutralization activity arises from a synergistic effort of multiple clones, with responses varying across individuals.
Mutation analysis revealed lower SHM frequencies in GI clones compared to GII, a pattern frequently seen in neutralizing antibodies during the acute viral infections (11,47). Early immune responses prioritize rapid antibody production through extrafollicular B-cell activation rather than extensive affinity maturation in germinal centers (38). GII exhibited higher SHM frequencies, with no significant changes following vaccination. The higher baseline SHM levels in GII likely reflect the activation of memory B cells rather than de novo somatic hypermutation. Studies in vaccinated individuals reported robust B-cell responses and higher neutralization activity but no significant alteration in SHM (38,48). These results highlight how infection and vaccination shape the B-cell repertoire. While infection induces a broader repertoire with lower SHM, vaccination primarily reactivates high-affinity memory B cells.
## Conclusion
In this study, we characterized B-cell kinetics in severe MERS-CoV-infected patients and survivors. Both groups showed enhanced neutralization activity against MERS-CoV and SARS-CoV-2. Specific IgG clones were linked to enhanced cross-neutralization. These findings demonstrate that individuals with hybrid immunity are a valuable source of potent, cross-neutralizing antibodies. However, there are several limitations in our study. First, we could not compare the immune response between the two cases at the early stage of infection, as the sample collection started after a month of symptom onset in the second case. Second, we did not test the antigenic specificity nor the neutralizing capacity for the identified clones. Instead, assumptions of their function were inferred from the association between clone expansion at a certain time point and neutralization activity at that point. To characterize the functionality of the clones, we compared our data with the CoV-AbDab database, which includes 71 MERS-CoV-specific antibody sequences. However, this approach is limited in breadth due to the limited number of MERS-CoV sequences in the database. Furthermore, the majority of SARS-CoV-2 sequences in the database are biased toward neutralizing RBD-specific antibody sequences. Therefore, future work should focus on isolating B-cell subsets that recognize the spike proteins of MERS-CoV, SARS-CoV-2, or both (cross-neutralizing) for further characterization.
## 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. Who (2023) "MERS situation update"
3. Sikkema, Farag, Islam et al. (2019) "Global status of middle east respiratory syndrome coronavirus in dromedary camels: a systematic review" *Epidemiol Infect*
4. Assiri, Mcgeer, Perl et al. (2013) "Hospital outbreak of middle east respiratory syndrome coronavirus" *N Engl J Med*
5. Senga, Arabi, Fowler (2017) "Clinical spectrum of the Middle East respiratory syndrome coronavirus (MERS-CoV)" *J Infect Public Health*
6. Midgley, Alkhamis, Abedi et al. (2019) "Middle east respiratory syndrome coronavirus infection dynamics and antibody responses among clinically diverse patients, Saudi Arabia" *Emerg Infect Dis*
7. Park, Perera, Choe et al. (2015) "Kinetics of serologic responses to mers coronavirus infection in humans, South Korea" *Emerg Infect Dis*
8. Alhabbab, Algaissi, Mahmoud et al. (2023) "Middle east respiratory syndrome coronavirus infection elicits long-lasting specific antibody, T and B cell immune responses in recovered individuals" *Clin Infect Dis*
9. Zhou, Yang, Huang et al. (2019) "Advances in MERS-CoV vaccines and therapeutics based on the receptor-binding domain" *Viruses*
10. Raj, Mou, Smits et al. (2013) "Dipeptidyl peptidase 4 is a functional receptor for the emerging human coronavirus-EMC" *Nature*
11. Niu, Zhang, Zhou et al. (2018) "Ultrapotent human neutralizing antibody repertoires against middle east respiratory syndrome coronavirus from a recovered patient" *J Infect Dis*
12. Choi, Woo, Lee et al. (2020) "Characterization of a human monoclonal antibody generated from a B-cell specific for a prefusion-stabilized spike protein of middle east respiratory syndrome coronavirus" *PLoS One*
13. Corti, Zhao, Pedotti et al. (2015) "Prophylactic and postexposure efficacy of a potent human monoclonal antibody against MERS coronavirus" *Proc Natl Acad Sci*
14. Li, Wan, Liu et al. (2015) "A humanized neutralizing antibody against MERS-CoV targeting the receptor-binding domain of the spike protein" *Cell Res*
15. Luke, Wu, Zhao et al. (2016) "Human polyclonal immunoglobulin G from transchromosomic bovines inhibits MERS-CoV in vivo" *Sci Transl Med*
16. Zhao, He, Sun et al. (2018) "A novel nanobody targeting middle east respiratory syndrome coronavirus (MERS-CoV) receptor-binding domain has potent cross-neutralizing activity and protective efficacy against MERS-CoV" *J Virol*
17. Kim, Kim, Jeon et al. (2024) "Rise in broadly cross-reactive adaptive immunity against human β-coronaviruses in MERS-recovered patients during the COVID-19 pandemic" *Sci Adv*
18. Zedan, Smatti, Thomas et al. (2023) "Assessment of broadly reactive responses in patients with MERS-CoV infection and SARS-CoV-2 vaccination" *JAMA Netw Open*
19. Who (2022) "Disease outbreak news: middle east respiratory syndrome coronavirus -Qatar Online 2022"
20. Capcha, Lambert, Dykxhoorn et al. (2020) "Generation of SARS-CoV-2 spike pseudotyped virus for viral entry and neutralization assays: A 1-week protocol" *Front Cardiovasc Med*
21. Nie, Li, Wu et al. (2020) "Quantification of SARS-CoV-2 neutralizing antibody by a pseudotyped virus-based assay" *Nat Protoc*
22. Gupta, Vander Heiden, Uduman et al. (2015) "Change-O: a toolkit for analyzing large-scale B cell immunoglobulin repertoire sequencing data" *Bioinformatics*
23. Shugay, Bagaev, Turchaninova et al. (2015) "VDJtools: unifying post-analysis of T cell receptor repertoires" *PLoS Comput Biol*
24. Arabi, Balkhy, Hayden et al. (2017)
25. "Middle east respiratory syndrome" *N Engl J Med*
26. Payne, Iblan, Rha et al. (2016) "Persistence of antibodies against middle east respiratory syndrome coronavirus" *Emerg Infect Dis*
27. Ko, Müller, Seok et al. (2017) "Serologic responses of 42 MERS-coronavirus-infected patients according to the disease severity" *Diagn Microbiol Infect Dis*
28. Alshukairi, Zhao, Al-Mozaini et al. (2021) "Longevity of middle east respiratory syndrome coronavirus antibody responses in humans, Saudi Arabia" *Emerg Infect Dis*
29. Assiri, Al-Tawfiq, Aa et al. (2013) "Epidemiological, demographic, and clinical characteristics of 47 cases of middle east respiratory syndrome coronavirus disease from Saudi Arabia: a descriptive study" *Lancet Infect Dis*
30. Patrick, Petric, Skowronski et al. (2006) "An outbreak of human coronavirus OC43 infection and serological cross-reactivity with SARS coronavirus" *Can J Infect Dis Med Microbiol*
31. Alosaimi, Awadalla, Alturaiki et al. (2025) "A 10 year long-lived cellular and humoral MERS-CoV immunity cross-recognizing the wild-type and variants of SARS-CoV-2: a potential one-way MERS-CoV cross-protection toward a pan-coronavirus vaccine" *J Med Virol*
32. Hicks, Klumpp-Thomas, Kalish et al. (2021) "Serologic cross-reactivity of SARS-CoV-2 with endemic and seasonal betacoronaviruses" *J Clin Immunol*
33. Ju, Zhang, Ge et al. (2020) "Human neutralizing antibodies elicited by SARS-CoV-2 infection" *Nature*
34. Safra, Tamari, Polak et al. (2023) "Altered somatic hypermutation patterns in COVID-19 patients classifies disease severity" *Front Immunol*
35. Stewart, Sinclair, Ng et al. (2022) "Pandemic, Epidemic, Endemic: B Cell Repertoire Analysis Reveals Unique Anti-Viral Responses to SARS-CoV-2, Ebola and Respiratory Syncytial Virus" *Front Immunol*
36. He, Liu, Xu et al. (2022) "Comparative global B cell receptor repertoire difference induced by SARS-CoV-2 infection or vaccination via single-cell V(D)J sequencing" *Emerg Microbes Infect*
37. Hoehn, Ramanathan, Unterman et al. (2021) "Cutting edge: distinct B cell repertoires characterize patients with mild and severe COVID-19" *J Immunol*
38. Rao, Srivastava, Verma et al. (2022) "B cell receptor repertoire analysis unveils dynamic antibody response and severity markers in COVID-19 patients" *Immunology*
39. Paschold, Klee, Gottschick et al. (2022) "Rapid hypermutation B cell trajectory recruits previously primed B cells upon third saRS-Cov-2 mRNA Vaccination" *Front Immunol*
40. Andreano, Paciello, Piccini et al. (2021) "Hybrid immunity improves B cells and antibodies against SARS-CoV-2 variants" *Nature*
41. Pinto, Park, Beltramello et al. (2020) "Crossneutralization of SARS-CoV-2 by a human monoclonal SARS-CoV antibody" *Nature*
42. Chan, Hadlock, Foung et al. (2001) "V(H)1-69 gene is preferentially used by hepatitis C virus-associated B cell lymphomas and by normal B cells responding to the E2 viral antigen" *Blood*
43. Guo, Zhang, Yang et al. (2023) "Discovery and characterization of potent pan-variant SARS-CoV-2 neutralizing antibodies from individuals with Omicron breakthrough infection" *Nat Commun*
44. Zhou, Ma, Xie et al. (2021) "Diverse immunoglobulin gene usage and convergent epitope targeting in neutralizing antibody responses to SARS-CoV-2" *Cell Rep*
45. Chen, Gilchuk, Zost et al. (2021) "Convergent antibody responses to the SARS-CoV-2 spike protein in convalescent and vaccinated individuals" *Cell Rep*
46. Zhou, Song, He et al. (2022) "Broadly neutralizing anti-S2 antibodies protect against all three human betacoronaviruses that cause severe disease"
47. Pinto, Sauer, Czudnochowski et al. (2021) "Broad betacoronavi rus neutralization by a stem helix-specific human antibody" *Science*
48. Zhang, Yan, Luo et al. (2022) "Analysis of B cell receptor repertoires reveals key signatures of the systemic B cell response after SARS-CoV-2 infection" *J Virol*
49. Goel, Apostolidis, Painter et al. (2021) "Distinct antibody and memory B cell responses in SARS-CoV-2 naïve and recovered individuals following mRNA vaccination" *Sci Immunol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12671071&blobtype=pdf | # Oita virus rediscovered after 50 years: isolation of genetically conserved strains from bats in Southern Japan
Saiko Sawai, Kittiya Intaruck, Sho Sata, Mana Esaki, Kimitake Funakoshi, Shin Murakami, Keita Matsuno, Ryo Nakao, Kazunori Kimitsuki, Takaaki Yahiro, Akira Nishizono, Naganori Nao, Yasuko Orba, Ayato Takada, Masahiro Kajihara, Makoto Ozawa, Kosuke Okuya, Kosuke 24k18023, Okuya
## Abstract
Bats are natural reservoirs of numerous viruses, including members of the Rhabdoviridae family. Oita virus (OITV), classified within the genus Ledantevirus, was first isolated from the blood of a Japanese horseshoe bat (Rhinolophus cornutus) in the Oita Prefecture, Japan, in 1972. However, since then, no other OITV isolates have been reported. In this study, we isolated two OITV strains-OITV 321/2022 and OITV 326/2022 -from oral swab samples of R. cornutus collected in the Katano Cave, Kagoshima Prefecture, Japan. Genetic analysis revealed >98% nucleotide identity with the previous isolate, OITV 296/1972. Phylogenetic analysis confirmed their classification within the Ledantevirus subgroup C. We assessed viral growth kinetics using 17 cell lines from diverse animal species, including bats, rodents, humans, and arthropods. OITV 321/2022 showed efficient replication in primate-derived, bat-derived, and human-derived cells, but not in tick or mosquito cell lines. Experimental infection of BALB/c mice demonstra ted that OITV 321/2022 induced no overt clinical signs via either the intracerebral or intranasal routes. However, systemic infection and viral replication in the brain and lungs were observed by intracerebral inoculation. Intranasal inoculation with OITV 321/2022 and OITV 296/1972 resulted in transient pulmonary infection without systemic dissemination. This study provides evidence of OITV circulation in wild bat popula tions over 50 years, highlighting the need for continued surveillance of bat-associated rhabdoviruses in Japan. IMPORTANCE Bats are known reservoirs of zoonotic viruses, and their proximity to humans raises concerns regarding zoonotic risks. We report the first isolation of the Oita virus (OITV), a bat-associated ledantevirus, that has circulated for over 50 years. Unlike a previous blood-derived isolate, our isolates were obtained from oral swabs, suggesting their potential for respiratory transmission. OITV could infect a wide range of mammalian cells, including human-derived cells, and induce systemic infection in mice without clinical symptoms. These findings indicate that OITV possesses a broad host tropism and may circulate among microbats through the respiratory tract. Although the pathogenicity of the newly isolated strain appears to be attenuated compared with that of a historical brain-passaged strain, its ability to replicate in human cells underscores its potential zoonotic relevance, necessitating active surveillance and functional characteri zation of bat-derived rhabdoviruses to better assess emerging infectious disease threats. KEYWORDS rhabdovirus, bat, re-emerging diseases, ledantevirus B ats are natural reservoirs of various viruses, including coronaviruses and rhabdovi ruses (1, 2). To date, more than 200 viruses have been detected in or isolated from bats, many with the potential of zoonotic transmission (3). Their ability to fly enables long-distance dispersal of viruses, contributing to the widespread geographic
distribution of bat-associated pathogens (4). In addition, the formation of large colonies facilitates viral maintenance and intra-species transmission within bat populations (5). Bats are also capable of adapting to urban environments, thereby increasing the frequency of contact with humans and domestic animals and raising the risk of crossspecies transmission.
The Rhabdoviridae family comprises single-stranded negative-sense RNA viruses with five genes: nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and viral RNA polymerase (L) (6). This family exhibits high ecological diversity, with members infecting various hosts, including plants, mammals, birds, reptiles, insects, and fish (7)(8)(9)(10). Some genera within the family include important animal and human viruses, such as lyssaviruses (e.g., rabies virus) and vesiculoviruses (e.g., Chandipura virus). An increasing number of novel bat-associated rhabdoviruses have recently been identified in different regions worldwide, highlighting the growing importance of active surveil lance of bats as viral reservoirs (6).
The genus Ledantevirus, as listed by the International Committee on Taxonomy of Viruses (https://ictv.global/) in 2024, includes 21 distinct viral species. Phylogenetic analyses have subdivided the genus Ledantevirus into three major subgroups: A, B, and C (10). Many members of this genus have been isolated from bats and are thought to be transmitted by arthropods (10). Among the ectoparasites of bats, bat flies-obligate hematophagous arthropods-are of particular interest. Kanyawara virus and Bughen dera virus have been directly detected in bat flies collected from parasitized bats (11). Additionally, the Kolente virus was isolated from both bats and ticks in the Republic of Guinea in 1985 (12). Furthermore, some ledanteviruses are suspected to be associated with human diseases (6,(13)(14)(15). The first identified virus, Ledantec virus, was isolated in 1965 from the serum of a 10-year-old girl with fever and hepatosplenomegaly at Le Dantec University Hospital in Senegal (15). In China, antibodies specific to a novel ledantevirus, Rhinolophus rhabdovirus DPuer (DPRV), were detected in human serum samples collected from individuals living near bat sampling sites who had experienced fever in the past year (6).
The Oita virus (OITV), classified within the genus Ledantevirus, was first isolated in 1972 and designated as OITV 296/1972 (16). The virus was obtained from the blood of a small Japanese horseshoe bat (Rhinolophus cornutus) captured in Oita Prefecture, Japan. OITV 296/1972 was isolated by intracerebral inoculation into suckling mice, which subsequently developed lethal encephalitis, indicating neurotropism similar to that of rabies virus (16). However, the replication characteristics and growth kinetics of this virus have not been evaluated in cultured cells. As OITV has not been isolated for over 50 years since 1972, its ecology remains largely unknown.
In this study, we isolated OITV from oral swab samples collected from R. cornutus captured in Kagoshima Prefecture, Japan. We genetically characterized the OITV and virologically assessed its growth kinetics in a range of cell lines. In addition, we investiga ted the pathogenicity of OITV in mice.
## RESULTS
## OITV isolation from microbats in Kagoshima Prefecture
Eighty-four oral swab samples were collected from R. cornutus at Katano Cave, Kagosh ima, Japan (Fig. 1) and inoculated subsequently into Vero-RcACE2 cells. Cytopathic effects (CPE), characterized by cell detachment, were observed 4 days after blind passage in cells inoculated with two swab samples, Bat321/2022 and Bat326/2022 (Table 1; Fig. 2A). Furthermore, bullet-shaped virus particles, characteristic of rhabdoviruses, were observed in the supernatant of Bat321/2022-inoculated cells by transmission electron microscopy (Fig. 2B).
Full-genome sequencing revealed that these two virus isolates belonged to the genus Ledantevirus, family Rhabdoviridae, and were tentatively designated OITV 321/2022 and OITV 326/2022.
## Genetic analyses of OITV 321/2022 and OITV 326/2022
Genetic analyses revealed that the Kagoshima isolates, OITV 321/2022 and OITV 326/2022, shared a high nucleotide sequence identity (98.52%) with OITV 296/1972, which was isolated from R. cornutus in Oita Prefecture, Japan, in 1972 (16). Each viral gene exhibited >97.78% sequence identity between Kagoshima isolates and OITV 296/1972 (Table 2). Phylogenetic analyses further confirmed that OITV 321/2022 and OITV 326/2022 are classified into subgroup C of the genus Ledantevirus (Fig. 3) (10). Since the full-length genome sequences of OITV 321/2022 and OITV 326/2022 were identical, OITV 321/2022 was used as a representative strain for subsequent virological characterization in this study.
## Cell susceptibility to OITV infection
To evaluate the host range and cell susceptibility to OITV, we first assessed the replication capacity of OITV 321/2022 in a panel of cell lines commonly used in in vitro studies. A total of eight cell lines derived from various animal species were tested, including those from nonhuman primates (African green monkeys: Vero, Vero-RcACE2, and VeroE6/ TMPRSS2 cells), humans (A549 cells), bats (TB1 Lu cells), hamsters (baby hamster kidney [BHK] cells), dogs (Madin-Darby canine kidney [MDCK] cells), and pigs (PK-15 cells).
OITV 321/2022 replicated in Vero, Vero-RcACE2, and VeroE6/TMPRSS2 cells, with apparent cell detachment at 4-5 days post-infection (dpi). Although no CPE was observed in the remaining cell lines, real-time RT-PCR confirmed viral replication in A549, BHK, and MDCK cells. In contrast, OITV 321/2022 did not replicate in TB1 Lu and PK-15 cells, indicating variable susceptibility to OITV among cell lines derived from different animal species.
## Replication of OITV 321/2022
To investigate the growth kinetics of OITV, OITV 321/2022 was inoculated into six cell lines that supported viral replication (Fig. 4A andB). Culture supernatants were collected at multiple time points and subjected to virus titration using the tissue culture infectious dose (TCID 50 ) assay in Vero-RcACE2 cells.
In Vero and Vero-RcACE2 cells, both of which exhibited cell detachment following infection, the viral titers began increasing at 24 hours post-infection (hpi), plateauing at 96 hpi (Fig. 4A). In VeroE6/TMPRSS2 cells, the viral titers gradually increased and reached levels comparable to those observed in Vero and Vero-RcACE2 cells at 168 hpi (Fig. 4A). In contrast, A549, BHK, and MDCK cells did not show CPE but supported viral replication. In these cells, the titers increased from 6 to 12 hpi and plateaued between 48 and 72 hpi (Fig. 4B). Notably, BHK cells supported viral replication at levels similar to those in CPE-positive cell lines, despite no visible CPE.
Ledanteviruses have been detected in bats and are suspected to be associated with arthropod vectors (16). We further tested OITV replication in various cell lines derived from bats, ticks, and mosquitoes (Fig. 4C). Seven bat-derived cell lines from different species were examined for viral replication and growth kinetics. Among bat-derived cell lines, CPE was observed in ZFBK 11-97 and SuBK 12-08 cells at 4 dpi (Fig. S1). Viral titers began to increase at 12 hpi and plateaued at 48 hpi in DemKT1, ZFBK 15-137RA, SuBK 12-08, YubFKT1, and IndFSPT1 cells (Fig. 4C). In contrast, BKT1 and ZFBK 11-97 cells exhibited delayed viral growth, with titers increasing from 24 hpi. Notably, ZFBK 11-97 cells showed the highest viral titers among all bat cell lines, peaking at 48 hpi, while BKT1 cells displayed a gradual increase, reaching a plateau at 96 hpi (Fig. 4C). Overall, OITV 321/2022 replicated most efficiently in ZFBK 11-97 cells among the seven bat cell lines tested. In contrast, viral titers in arthropod-derived cell lines (ISE-6 and C6/36) remained below the detection limit throughout the observation period, suggesting that these cells do not support OITV 321/2022 replication.
## Infectivity of OITV 321/2022 in mice
To assess the pathogenicity of OITV 321/2022, BALB/c mice were inoculated with this strain, using OITV 296/1972 (16) as a positive control. Mice inoculated intracere brally with OITV 296/1972 exhibited weight loss, whereas those inoculated intranasally exhibited no clinical symptoms (Fig. 5A). In contrast, mice inoculated with OITV 321/2022 via either route showed no clinical signs or significant weight loss (Fig. 5A). Despite the absence of symptoms, serum samples collected on 14 dpi from mice inoculated with either OITV 296/1972 or OITV 321/2022 showed elevated titers of neutralizing antibody, except for one intracerebrally inoculated mouse in the OITV 296/1972 group (Fig. 5B). Furthermore, OITV-specific antibodies were detected by enzyme-linked immunosorbent assay (ELISA) in serum samples from all virus-inoculated mice (Fig. S2), indicating that OITV infection was established via both the intracerebral and intranasal routes. To identify the sites of viral replication, viral gene copy numbers were measured in 10 organs using real-time RT-PCR (Fig. 6). High viral loads (>10 10 copies/g) were detected in the brains of mice intracerebrally inoculated with either OITV 321/2022 or OITV 296/1972 on 3 dpi (Fig. 6A). In addition, viral RNA was detected in other organs, such as the lungs, trachea, liver, and spleen, on 3 dpi. Notably, a wide range of tissues, including the stomach and kidneys, tested positive for viral RNA in mice infected with OITV 321/2022. The brain viral load of OITV 321/2022 exceeded 10 8 copies/g on 6 dpi and remained above 10 4 copies/g on 14 dpi, whereas OITV 296/1972 maintained higher levels (10 8 copies/g) on 6 and 14 dpi.
In mice inoculated intranasally with either OITV 321/2022 or OITV 296/1972, viral genes were detected only in the lungs on 3 dpi at levels exceeding 10 7 copies/g (Fig. 6B). No viral genes were detected in the other organs, including the trachea. These results suggest that both OITV 321/2022 and OITV 296/1972 infected BALB/c mice via the intranasal route but remained largely localized and confined to the lungs and were cleared without systemic dissemination by 14 dpi.
## DISCUSSION
In the present study, two OITV strains were isolated from oral swab samples collected from R. cornutus in the Katano Cave, Kagoshima Prefecture, Japan (Table 1). Notably, this is the first isolation of OITV in 50 years since OITV 296/1972 was isolated from the blood of R. cornutus in the Oita Prefecture, Japan, in 1972 (16). Unlike previous blood-derived isolates, the Kagoshima isolates were recovered from oral swabs, suggesting viral shedding via this route. R. cornutus and other microbat species are widely distributed across Japan, from temperate to subarctic regions, and often form large colonies ranging from dozens to thousands of individuals (5). This broad geographical distribution suggests that OITV may circulate among these bat species throughout Japan and is not limited to the Kyushu region. Continued surveillance of OITV in bats across Japan will provide further insight into its epidemiology. Notably, none of the other viruses were detected from OITV-positive bats in this study, including bat sarbecoviruses, which have been isolated from the Katano Cave (17).
Among the viral genes, the P gene of OITV 321/2022 exhibited the lowest nucleotide identity (97.78%) with that of OITV 296/1972, whereas the other ORFs showed higher similarity (Table 2). Similar P gene variability has been reported for other rhabdoviruses, including rabies, Kanyawara, and Durham viruses (11,(18)(19)(20). Although the N gene is generally conserved among rhabdoviruses, a wide range of nucleotide identities (87.6%-100%) has been observed among N gene fragments of 60 RABV strains collected over a 2-year period in China (21). In contrast, the N genes of the Kagoshima isolates showed 99.15% nucleotide identity with OITV 296/1972, despite the 50-year gap, suggesting strong genetic conservation. It is known that viral mutation rates tend to be lower in insect cells than in mammalian cells (22), raising the possibility that OITV circulates primarily in arthropod hosts and is sporadically transmitted to microbats. OITV belongs to subgroup C of the genus Ledantevirus, members of which have been identified in both vertebrate and invertebrate hosts, including bat flies, supporting this hypothesis (Fig. 3) (10,11,23,24).
Several viruses in the genus Ledantevirus have been shown to replicate in cells derived from a broad range of animal species. For example, a previous study reported that DPRV replicates in BHK, A549, and African green monkey kidney MA104 cells (6). Kumasi rhabdovirus replicates efficiently in Vero, MA104, A549, and kidney and lung cells from Eidolon helvum (EidNi/EidLu cells) (2), whereas the Kolente virus replicates in BHK cells (12). In this study, OITV 321/2022 could replicate in cell lines derived from various animal species, including multiple bat species (Fig. 4), but not in mosquito (Aedes albopictus; C6/36 cells) and tick (Ixodes scapularis; ISE6 cells) cell lines. These results suggest that OITV may circulate in other arthropod vectors, such as bat flies. Notably, OITV 321/2022 replicated in a human-derived A549 cell line, indicating a potential for human infection. Previous serological studies have shown that a substantial propor tion of individuals have had prior exposure to ledanteviruses typically associated with mild or subclinical symptoms (2,6). Further serological investigations are warranted to determine whether OITV infects humans in Japan.
OITV 296/1972 is known to replicate in the brains of BALB/c mice following intracere bral inoculation, causing lethal encephalitis in mice up to 2 weeks of age and inducing only transient signs, such as ruffled fur, in mice older than 4 weeks (16). In this study, intracerebral inoculation of BALB/c mice with OITV 321/2022 did not result in any clinical symptoms (Fig. 5A). However, a systemic infection was found on 3 dpi (Fig. 6A). Although OITV 321/2022 replicated in the brain, its viral titers declined over time, in contrast to OITV 296/1972, maintaining high viral loads in the brain throughout the 14-day experimental period (Fig. 6A).
A previous study reported that DPRV exhibited increased morbidity and mortality following serial intracerebral passages in mice (6). Similarly, OITV 296/1972 used in our study had been serially passaged and maintained in mouse brains, which may have enhanced its neurovirulence compared with the original field isolate. In contrast, OITV 321/2022 was directly inoculated into mice after isolation from cultured cells, thereby providing an assessment of its pathogenicity in a state likely closer to that in nature. Importantly, both OITV 321/2022 and OITV 296/1972 established infections in the lungs following intranasal inoculation (Fig. 5B and6B; Fig. S2). Consistent with the virus isolation from oral swabs, these findings suggest that OITV may be transmitted among microbats via respiratory route.
In conclusion, OITV was isolated from the oral swabs of microbats in Kagoshima Prefecture, Japan, marking the first isolation of this virus since its original identification in 1972. Our findings demonstrate that OITV can infect and replicate in a broad range of animal cell lines, including those derived from humans, indicating a potential risk of cross-species transmission to domestic animals and humans. This study highlights the importance of continued surveillance and in-depth characterization of bat-associated viruses to better assess their potential for zoonotic transmission.
## MATERIALS AND METHODS
## Cells and medium
Cells derived from the kidneys of African green monkeys (Vero cells), Vero cells stably expressing angiotensin-converting enzyme 2 derived from R. cornutus (Vero-RcACE2 cells) (25), human adenocarcinoma cells derived from lung cancer A549 cells, and BHK cells were maintained in Dulbecco's modified Eagle's medium (DMEM) (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS), 10,000 U/mL penicillin, and 10 mg/mL streptomycin (PS) (Fujifilm Wako Pure Chemical Corporation) (10% FBS/DMEM). VeroE6 cells stably expressing human transmembrane serine protease 2 (VeroE6/TMPRSS2 cells) (26) were grown in 10% FBS/ DMEM supplemented with 100 µg/mL G418 (InvivoGen, San Diego, CA, USA). The bat cell line, TB1 Lu (Tadarida brasiliensis, lung; ATCC number CCL-88), was maintained in Eagle's minimum essential medium (EMEM) (Fujifilm Wako Pure Chemical Corporation) supplemented with 10% FBS and PS (10% FBS/EMEM). PK-15 cells were maintained in 10% FBS/EMEM supplemented with a non-essential amino acid solution (Nacalai Tesque, Kyoto, Japan). MDCK cells were maintained in MEM (Thermo Fisher Scientific) supple mented with 5% newborn calf serum, 0.9 mM sodium bicarbonate, MEM amino acids (Thermo Fisher Scientific), MEM vitamin solution (Fujifilm Wako Pure Chemical Corpora tion), 2 mM L-glutamine, and PS. After inoculation with swab samples or viral isolates, Vero, Vero-RcACE2, VeroE6/TMPRSS2, A549, and BHK cells were maintained in 2% FBS/ DMEM, whereas TB1 Lu and PK-15 cells were maintained in 2% FBS/EMEM. Bat-derived cell lines, including DemKT1 (Rousettus leschenaultii, kidney), BKT1 (R. ferrumequinum, kidney), YubFKT1 (Miniopterus fuliginosus, kidney), IndFSPT1 (Pteropus giganteus, spleen), ZFBK 11-97 (Epomophorus gambianus, kidney), ZFBK 15-137RA (Rousettus aegyptiacus, kidney), and SuBK 12-08 (M. schreibersii, kidney) were cultured in Roswell Park Memorial Institute 1640 medium (Sigma-Aldrich) supplemented with 10% FBS and PS. All bat cell lines were maintained at 37°C in an atmosphere with 5% CO 2 (27). The tick cell line ISE6, derived from embryonated eggs of I. scapularis, was cultured in L-15B medium containing 10% FBS and 5% tryptose phosphate broth (Sigma-Aldrich) at 32°C without CO 2 . The mosquito cell line C6/36, derived from A. albopictus, was maintained in MEM supplemented with 10% FBS, PS, and 1% MEM non-essential amino acids (Gibco) at 28°C in an atmosphere with 5% CO 2 .
## Sample collection
Oral swab samples were collected from microbats captured in Katano Cave, Kagoshima Prefecture, in 2022 (Fig. 1) as reported previously (17). Microbats were captured using nets and released immediately after oral swab collection. Species were identified based on morphology. Age class, adult or young, was determined according to the presence or absence of clear signs of sexual maturity. The swabs were suspended in laboratorymade transport medium (MEM supplemented with 0.5% bovine serum albumin, 2 mM L-glutamine, 0.3 mg/mL gentamicin, 2.5 µg/mL amphotericin B, and PS) and stored at 4°C or lower. The samples were immediately subjected to viral isolation on the day of collection.
## Virus isolation
Suspended swab samples were used for virus isolation (17,28). In brief, the supernatant was filtered through a 0.2 µm pore membrane (Sartorius, Göttingen, Germany) and inoculated into Vero-RcACE2 cells. After an hour of incubation at 37°C, the inoculum was replaced with fresh medium. The samples were subjected to blind passage thrice at 1-week intervals. Supernatant from cells showing detachment was collected and stored at -80°C. Viral titers were assessed in each cell line using a median TCID 50 assay.
## Transmission electron microscopy
Supernatant collected from cells infected with our isolate, OITV 321/2022, was filtered through a 0.2 µm pore membrane (Sartorius) and subjected to ultracentrifugation (Himac CS 120GX, Eppendorf Himac Technologies Co., Ltd., Ibaraki, Japan) at 24,000 rpm at 4°C for 2 h with a sucrose cushion (15% sucrose in PBS). The precipitate was fixed with 1% formaldehyde and stained negatively with phosphotungstic acid. The prepared samples were examined under an H7000KU electron microscope (Hitachi, Tokyo, Japan) operated at 80 kV.
## Full-genome sequencing
Nucleic acids were extracted from the precipitate obtained by ultracentrifugation using an innuPREP Virus DNA/RNA Kit (IST Innuscreen GmbH, Reinach, Switzerland). Doublestranded cDNA (dscDNA) was synthesized from the total RNA using the PrimeScript Double Strand cDNA Synthesis Kit (TaKaRa Bio, Kusatsu, Japan). Next, dscDNA was sequenced using a MinION Mk1B nanopore sequencer with a Flongle flow cell and a direct cDNA Sequencing Kit (SQK-DCS109) (Oxford Nanopore Technologies, Oxford, UK) or a reagent kit designed for sequencing (MiSeq Reagent Kit v3, Illumina, San Diego, CA, USA; 600 cycles). The obtained contigs were assembled de novo using the Geneious Prime software (Dottmatics, Boston, MA, USA). Consensus sequences were subjected to BLAST analysis (https://blast.ncbi.nlm.nih.gov).
## Phylogenetic analyses
Nucleotide sequences of Rhabdoviridae were retrieved from the NCBI nucleotide database (https://www.ncbi.nlm.nih.gov/nucleotide/). Sequences were aligned using the MUSCLE software (29). Phylogenetic analyses of the nucleotide sequences from our isolates and other Rhabdoviruses were conducted using IQ-TREE software (30) with bootstrap analyses of 1,000 replicates. The best-fit model (GTR + F + I + R7) was selected by IQ-TREE software.
## Real-time RT-PCR
Real-time RT-PCR was performed using an iTaq Universal SYBR Green One-Step Kit (Bio-Rad, Hercules, CA, USA) following the manufacturer's protocol. Primer sets were designed based on the sequence of the OITV 321/2022 L gene to generate 163 bp PCR products (OITV 321/2022 1797F: ACAATGGCGGATGACAT and OITV 321/2022 1958R: CCC AGGAAAGCTCCCAT). The PCR product of the OITV 321/2022 L gene was cloned into a pCR Blunt II-TOPO cloning vector (Thermo Fisher Scientific).
## Evaluation of growth kinetics based on the virus titer
OITV 321/2022 was inoculated into Vero, Vero-RcACE2, and VeroE6/TMPRSS2 cells at a multiplicity of infection (MOI) of 0.001, while A549, BHK, and MDCK cells were infected at an MOI of 1.0 calculated in Vero-RcACE2 cells. Bat-derived, tick-derived, and mosquitoderived cells were inoculated with OITV 321/2022 at an MOI of 1.0. After an hour of incubation at 37°C, the inoculum was replaced with fresh medium. Supernatants were serially collected up to 168 hpi. The viral titer in each supernatant was determined using TCID 50 assays in Vero-RcACE2 cells.
## Experimental infection in mice
Three-week-old female BALB/c mice were purchased from Japan SLC Inc. (Hamamatsu, Japan) and housed in animal biosafety level-2 facilities at the Joint Faculty of Veterinary Medicine, Kagoshima University, Japan. Overall, 34 mice were used in this study and inoculated intracerebrally (n = 7) or intranasally (n = 7) with OITV 321/2022, while the other group was inoculated intracerebrally (n = 7) or intranasally (n = 7) with OITV 296/1972. Negative controls were mice administered PBS intracerebrally (n = 3) or intranasally (n = 3). Mice were inoculated with OITV at 1.58 × 10 6 TCID 50 in 200 µL, determined using Vero-RcACE2 cells. Each mouse was monitored daily for clinical signs and weight loss until 14 dpi. Major organs (brain, heart, lungs, trachea, stomach, liver, spleen, kidneys, intestine, and colon) were collected from two mice from each OITV-ino culated group on 3, 6, and 14 dpi. Each collected organ was homogenized to prepare a 10% emulsion with a laboratory-made transport medium using a TissueLyser II (Qiagen, Hulsterweg, Netherlands). All organ homogenates were preserved at -80°C until nucleic acid extraction. RNA was extracted from organ homogenates using the MagMAX Viral/ Pathogen Nucleic Acid Isolation Kit with KingFisher Duo Prime (Thermo Fisher Scientific) following the manufacturer's instructions. The viral copy number was quantified for each specimen by real-time RT-PCR (see above). Whole blood was obtained from three mice from each OITV-inoculated group and negative control on 14 dpi. The whole blood samples were centrifuged at 10,000 rpm at 4°C for 5 min, and the serum was collected for subsequent microneutralization and ELISA. All procedures involving BALB/c mice were approved by the Institutional Animal Care and Use Committee of the Kagoshima University Experimental Animal Center (approval no. JFVM22049).
## Microneutralization assay
Serum samples collected from the BALB/c mice were serially diluted (10 × 2 0 to 10 × 2 12 ) and mixed with OITV 321/2022 or OITV 296/1972 at 240 TCID 50 . After 1 h of incubation, the serum-virus mixture was inoculated into Vero-RcACE2 cells in 96-well plates and cultured for 6 days. Wells showing CPE were scored as OITV-positive. The neutralizing antibody titer was defined as the highest serum dilution that completely prevented OITV infection.
## ELISA
Supernatants from Vero-RcACE2 cells infected with OITV 321/2022 and OITV 296/1972 were filtered through a 0.22 µm pore membrane (Sartorius). The viruses were concentra ted via ultracentrifugation at 24,000 rpm in a Beckman SW32 Ti swinging-bucket rotor (Beckman Coulter) for 2 h with a sucrose cushion (15% sucrose in PBS). The resulting pellet was resuspended in disruption buffer containing 0.05 M Tris-HCl (pH 7.6), 0.5% Triton X-100, and 0.6 M KCl. Disrupted viral particles were diluted 1,000-fold in PBS and used as ELISA antigens. ELISA plates (Nunc Maxisorp; Invitrogen, Carlsbad, CA, USA) were coated with prepared viral antigens and blocked with 3% skim milk (Becton Biosciences, Franklin Lakes, NJ, USA) in PBS. Serum samples were initially diluted 1:50 in PBS containing 1% skim milk and 0.05% Tween 20, and then subjected to twofold serial dilutions up to 1:25,600. Each dilution was plated in triplicate. Bound antibodies were detected using a horseradish peroxidase-conjugated goat anti-mouse IgG antibody (H + L) (Abcam, Cambridge, UK). Antibody-antigen interactions were visualized using EzELISA TMB (ATTO Corporation, Tokyo, Japan). Optical density was measured at 450 nm to quantify the antibody response.
## References
1. Cui, Li, Shi (2019) "Origin and evolution of pathogenic coronavi ruses" *Nat Rev Microbiol*
2. Binger, Annan, Drexler et al. (2015) "A novel rhabdovirus isolated from the straw-colored fruit bat Eidolon helvum, with signs of antibodies in swine and humans" *J Virol*
4. Moratelli, Calisher (2015) "Bats and zoonotic viruses: can we confidently link bats with emerging deadly viruses?" *Mem Inst Oswaldo Cruz*
5. Sasaki, Kajihara, Changula et al. (2018) "Identification of group A rotaviruses from Zambian fruit bats provides evidence for long-distance dispersal events in Africa" *Infect Genet Evol*
6. Funakoshi, Takeda (1998) "Food habits of sympatric insectivorous bats in southern Kyushu" *Japan. Mammal Study*
7. Li, Xu, Lu et al. (2021) "Isolation of a novel bat rhabdovirus with evidence of human exposure in China" *mBio*
8. Kuzmin, Novella, Dietzgen et al. (2009) "The rhabdoviruses: biodiversity, phylogenetics, and evolution"
9. Kuzmin, Hughes, Rupprecht (2006) "Phylogenetic relationships of seven previously unclassified viruses within the family Rhabdoviridae using partial nucleoprotein gene sequences" *J Gen Virol*
10. Gubala, Davis, Weir et al. (2010) "Ngaingan virus, a macropod-associated rhabdovirus, contains a second glycoprotein gene and seven novel open reading frames" *Virology (Auckl)*
11. Blasdell, Guzman, Widen et al. (2015) "Ledantevirus: a proposed new genus in the Rhabdoviridae has a strong ecological association with bats" *Am J Trop Med Hyg*
12. Bennett, Paskey, Kuhn et al. (2020) "Diversity, transmission, and Cophylogeny of ledanteviruses (Rhabdoviri dae: Ledantevirus) and nycteribiid bat flies parasitizing angolan softfurred fruit bats in Bundibugyo District" *Uganda. Microorganisms*
13. Ghedin, Rogers, Widen et al. (2013) "Kolente virus, a rhabdovirus species isolated from ticks and bats in the Republic of Guinea" *J Gen Virol*
14. Ashraf, Jerome, Bugembe et al. (2025) "Uncovering the viral aetiology of undiagnosed acute febrile illness in Uganda using metagenomic sequencing" *Nat Commun*
15. Shepherd, Ashraf, Salazar-Gonzalez et al. (2024) "Wide spread human exposure to ledanteviruses in Uganda: a population study" *PLoS Negl Trop Dis*
16. Karabatsos (1978) "Supplement to International Catalogue of Arboviruses including certain other viruses of vertebrates" *Am J Trop Med Hyg*
17. Iwasaki, Inoue, Tanaka et al. (2004) "Characterization of Oita virus 296/1972 of Rhabdoviridae isolated from a horseshoe bat bearing characteristics of both lyssavirus and vesiculovirus" *Arch Virol*
18. Okuya, Sata, Esaki et al. (2025) "Genetic diversity of sarbecoviruses isolated from microbats in Southern Japan" *Virol J*
19. Wang, Wu, Tao et al. (2013) "Genetic and evolutionary characterization of RABVs from China using the phospho protein gene" *Virol J*
20. Delmas, Holmes, Talbi et al. (2008) "Genomic diversity and evolution of the lyssaviruses" *PLoS One*
21. Allison, Palacios, Da Rosa et al. (2011) "Characterization of Durham virus, a novel rhabdovirus that encodes both a C and SH protein" *Virus Res*
22. Tao, Tang, Li et al. (2009) "Molecular epidemiology of rabies in Southern People's Republic of China" *Emerg Infect Dis*
23. Combe, Sanjuán (2014) "Variation in RNA virus mutation rates across host cells" *PLoS Pathog*
24. Kaneko, Inaba, Akashi et al. (1986) "Isolation of a new bovine ephemeral fever group virus" *Aust Vet J*
25. Goldberg, Bennett, Kityo et al. (2017) "Kanyawara virus: a novel rhabdovirus infecting newly discovered nycteribiid bat flies infesting previously unknown pteropodid bats in Uganda" *Sci Rep*
26. Murakami, Kitamura, Matsugo et al. (2022) "Isolation of bat sarbecoviruses" *Japan. Emerg Infect Dis*
27. Matsuyama, Nao, Shirato et al. (2020) "Enhanced isolation of SARS-CoV-2 by TMPRSS2-expressing cells" *Proc Natl Acad Sci*
28. Takadate, Kondoh, Igarashi et al. (2020) "Niemann-pick C1 heterogene ity of bat cells controls filovirus tropism" *Cell Rep*
29. Sata, Kojima, Esaki et al. (2024) "The first isolation and characteriza tion of bat jeilongviruses in Japan" *Transbound Emerg Dis*
30. Edgar (2004) "MUSCLE: multiple sequence alignment with high accuracy and high throughput" *Nucleic Acids Res*
31. Chernomor, Haeseler, Minh (2016) "Terrace aware data structure for phylogenomic inference from supermatrices" *Syst Biol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12555485&blobtype=pdf | # DU. Poster display: Public health epidemiology and genomics
M Cho, N Been, H Son
## Abstract
as well as exposure to discrimination and violence. Data are collected through face-to-face questionnaires between January and May 2025, via an outreach approach. We further investigate the determinants of occupational risk using multivariate regressions. Results: Preliminary results on a sub-sample of 729 individuals show that this population is overexposed to occupational hazards. As an example, 55.4% of delivery riders had a work accident, and 41.5% have symptoms of moderate to severe depression. In the meantime, they suffer from a great lack of access to social protection, especially health coverage: 30.8% lack health insurance coverage, and an overwhelming majority of 95.8% reported that they are not protected by insurance provided by the platform in the event of an on-the-job accident.Conclusions: Our findings will help construct an intervention to prevent occupational risks among gig delivery workers and moderate their financial impacts by improving access to health coverage. This contribution will also provide a better understanding of the construction of social inequalities in health in the context of worsening working conditions and the rise of the platform economy, with the emerging social category of "uberized" workers. Key messages:• Gig delivery workers in France are over-exposed to occupational hazards and have severely limited access to health insurance. • Our findings strongly indicate the necessity of implementing community-based occupational health interventions among platformbased delivery riders.
Background: Major variants of SARS-CoV-2 have significantly altered transmissibility and infection characteristics, posing substantial challenges to public health. In particular, structural changes in the receptor-binding domain (RBD) of the spike protein influence its binding affinity to the human ACE2 receptor. This study aimed to quantitatively assess differences in RBD features and ACE2-binding properties across SARS-CoV-2 variants to better understand their public health implications. Methods: Spike protein sequences from five major SARS-CoV-2 variants-Alpha, Beta, Gamma, Delta, and Omicron-were analysed. Full-length amino acid sequences and RBD regions were extracted and examined through multiple sequence alignment and conservation scoring. Key mutations were identified, and protein-protein docking was carried out to evaluate binding interactions between variant RBDs and human ACE2. Docking scores and confidence scores were compared. Additionally, molecular dynamics (MD) simulations were performed to assess the structural stability of the RBD-ACE2 complexes, with root mean square fluctuation (RMSF) used to quantify flexibility at the binding interface. Results: Omicron exhibited the highest number of amino acid substitutions within the RBD and the lowest sequence conservation. Mutations such as N501Y, L452R, and E484K were identified, potentially contributing to binding affinity and immune evasion. Docking analysis showed that Omicron had the strongest predicted binding to ACE2, followed by Alpha and Delta. MD simulations revealed that Gamma exhibited the highest structural flexibility, whereas Alpha and Omicron maintained relatively stable interactions. Conclusions: This study provides an integrated comparison of RBD structure and ACE2-binding characteristics among SARS-CoV-2 variants. The structural properties of the variants contribute to differences in transmissibility and may inform variant monitoring and public health response strategies.
## Key messages:
• Structural analysis of SARS-CoV-2 variants supports surveillance and public health planning. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12840300&blobtype=pdf | # Advances in Molecular Microbiology-From Recent Advances to the Future
Bruce Seal
## Abstract
The Molecular Microbiology section of Current Issues in Molecular Biology publishes original research and review articles on microbes, including bacteria, archaea, eukaryotic microorganisms, and viruses. Published investigations address the fundamental mechanisms of molecular microbiology, encompassing areas ranging from basic research to translational or applied research. Areas of interest include mechanisms of antimicrobial resistance, vaccine development, the evolution of host-pathogen interactions, microbial ecology, comparative microbiology of various hosts, and microbial genomics.Recent developments in molecular microbiology include the continuation of the search for alternatives to antibiotics, utilizing bacteriophage therapy [1,2]. This is especially true for the ESKAPE pathogens, including Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp. This group of bacteria comprises organisms that are highly predisposed to antibiotic resistance; they are therefore difficult to treat with traditional antibiotics to achieve effective infection control [3]. Although newly developed antibiotics, such as cefiderocol, have shown promise, resistance to this siderophore has been demonstrated among certain cohort groups [4]. Consequently, research into developing personalized therapies using bacteriophages as an alternative treatment in clinical settings could be an important aspect of future treatment approaches [5]. To achieve the goal of utilizing bacteriophages as antimicrobials, it will be necessary to continue supporting the isolation and characterization of bacteriophages from a variety of bacterial hosts prone to developing antibiotic resistance during infections [6][7][8][9].The Food and Agriculture Organization of the United Nations and the WHO (FAO/WHO) define probiotics as "live microorganisms which when administered in adequate amounts confer a health benefit on the host," and probiotic bacteria are now being developed for medical applications [10]. Traditionally, probiotic treatments have been associated with alleviating gastrointestinal disorders, such as those associated with Helicobacter pylori [11]. Recent proposals have included the use of probiotics in sanitizing healthcare environments [12] and employing detergents in conjunction with selected probiotics that can displace surrounding pathogens through competitive exclusion [13]. One interesting approach to control infections among crowded community environments is the use of probiotic-based sanitation [14]. This method is reportedly effective in reducing levels of fungal, bacterial, and viral pathogens as an alternative to traditional chemical disinfectants and, in one study, was even used to help sanitize a subway microbiome [15]. Such breakthroughs are particularly compelling; as an example, a high-risk, antibiotic-resistant strain of Pseudomonas aeruginosa was cultured from an urban water drain in a populated subway underpass, demonstrating the multitude of pathogens that can be found in various environments that negatively impact human and animal health [16].Research tools used by investigators to genetically modify microbial genes have traditionally been implemented using techniques such as transposon mutagenesis screens to
identify the mechanisms of specific genes in certain phenomena, including the thermoregulation of motility in organisms, for example, typhoidal Salmonella [17]. The development of high-throughput sequencing techniques [18] combined with CRISPR (clustered regularly interspaced short palindromic repeats)-Cas systems [19] has revolutionized researchers' ability to study microbial-host interactions [20]. One of the more intriguing developments is the technique of the newly developed MetaEdit (Metagenomic Editing) approach, which allows for targeted insertion of large DNA sequences into the genomes of bacteria within the mouse gastrointestinal tract [21]. The procedure involves the utilization of optimized CRISPR-associated transposases delivered by a conjugative vector system to genetically engineer a diverse ecology of commensal bacteria directly to single-nucleotide genomic resolution. The goal is to genetically modify individual bacteria within natural communities among gigabases of a metagenomic bacterial population. Moreover, techniques such as precision microbiome programming for therapeutic applications may eventually become a reality [22]. Certainly, the above breakthroughs represent potentially monumental advances in metagenomics analyses; for now, however, investigators will continue to utilize CRISPR systems to study individual pathogens [23].
Diagnostic development for clinically important microorganisms remains an essential aspect of microbial research for highly specific, rapid pathogen detection [24]. CRISPR systems have been adapted for microbial detection assays, which include the Combinatorial Arrayed Reactions for Multiplexed Evaluation of Nucleic acids (CARMEN), used for multiplexed pathogen detection [25]. This assay has recently been updated to detect 23 blood-borne pathogens critical to clinical diagnostics and public health surveillance [26]. Another technique, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), has been developed for the detection of microorganisms, and databases have been established to characterize pathogens [27,28]. The technique has been utilized for the rapid detection of pathogen-associated illnesses such as Staphylococcus aureus bloodstream infections [29] and fungal biotyping [30]. Point-of-care diagnostics utilizing molecular techniques are another critical aspect of infectious disease detection and monitoring [31]. Such testing is particularly important in locations where rapid, low-cost alternative diagnostics are needed as it does not require sophisticated equipment and has been recommended by the WHO for initial diagnosis. Consequently, methods such as the loop-mediated isothermal amplification (LAMP) technique offer a promising tool for the diagnosis of pulmonary tuberculosis in resource-limited regions of the world [32]. In addition to nucleic acid techniques, time-honored serological diagnosis of parasitic diseases, such as toxoplasmosis, remains vital in many parts of the world [33].
Exploration of microbiomes in numerous environments is an ever-evolving area of research. On a related note, Time magazine recognized Dr. Jeffrey I. Gordon for his contributions to the development of a therapeutic food designed to treat childhood malnutrition as one of the best inventions of 2025 (https://time.com/collections/best-inventions-2025/731 8496/mdcf-2/, accessed on 20 December 2025 and https://medicine.washu.edu/news/ time-magazine-names-therapeutic-microbiome-directed-food-a-best-invention-of-2025/, accessed on 20 December 2025. https://www-nature-com.oregonstate.idm.oclc.org/ subjects/microbiome, accessed on 20 December 2025). Severe global malnutrition affects 14 million children under 5 years of age, and food insecurity remains an ongoing challenge for which therapeutic interventions are needed, representing a significant societal burden in low-and middle-income countries [34]. Importantly, Dr. Gordon and collaborators created an intervention with a microbiota-directed complementary food that was more efficacious than a ready-to-use supplementary food. The results provide investigators and health personnel with a path that will provide a basis for further testing of the human microbiome to identify biomarkers and better define treatment responses for malnutrition [35,36]. Bac-teria represent the primary focus of the majority of microbiome studies. However, fungi also play important roles in gastrointestinal microbiomes, yet these organisms are not as widely studied in microbial ecology research. The basis of new applications may depend on expanding our knowledge of the evolutionary origins of the gut microbiome, which will also depend on deciphering the importance of the mycobiome [37].
Environmental microbiology is a far-reaching subject that includes a diverse array of research, including aquatic microbial systems, biofilms, plant-fungal relationships, how microbes play roles in restoration ecology, and how microbes impact a variety of insect and animal microbiomes. Within this field, bioremediation utilizing consortia of microbes holds promise as a tractable, commercially viable strategy to treat wastewater and contaminated marine or freshwater environments [38]. Moreover, these approaches may also contribute to the production of useful byproducts such as fertilizers [39]. To improve upon these approaches, progress in the world of bioinformatics will continue to play a crucial role in bacterial genomics and metagenomics [40]. Consequently, new userfriendly metagenomics analysis techniques, such as MetaXplore, have been developed as an interactive interface written in the R language to analyze complex microbial communities through high-throughput sequencing of marker gene amplicon data. This process includes determining alpha and beta diversity, taxonomic composition, differential abundance analysis, and identification of the core microbiome [41].
Discovery and research in virology will continue to play important roles in microbiological research. The impact of viral disease-causing agents, such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [42], mechanisms by which human immunodeficiency virus type 1 (HIV-1) impacts host cellular pathways [43], avian influenza in agricultural products [44], and the need for mpox diagnostics [45], combined with a "host" of other issues in virology, will continue to be a focus during the future [46]. One important aspect of viral pathogenesis research is related to the emergence of long COVID. The emergence of long-term illness after SARS-CoV-2 infection was first reported in 2020, following reports of patients with infection-associated chronic conditions, which included mainstream media coverage and official recognition of the condition among the human population [47]. The condition is defined as "an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems" that manifests as interstitial lung disease with cardiac complications [48]. Research into the causes of long COVID includes investigating persistent immune activation and proinflammatory responses [49] and potential B-cell dynamics among individuals with long COVID [50], amongst other mechanisms of the disease to aid mitigation of the syndrome [51].
Of particular interest in the future will be to monitor how artificial intelligence (AI) and machine learning impact microbiology and microbiome research [52]. Within the diagnostic laboratory, advancements in image analysis, also known as computer vision, help interpret digital images through comparison with annotated reference standards and curation of data for quality control analysis, two areas that could expedite and improve accuracy in clinical microbiology practices [53]. Natural Language Processing (NLP) is being used to analyze microbial interactions in gastrointestinal communities to predict issues such as irritable bowel disease (IBD) and diet-related effects by identifying distinct taxa [54]. Due to the complexities of artificial intelligence, machine learning, and microbial ecology, it is beneficial to have accessible guides that provide foundational information on these subjects [55,56]. Microorganisms are the most abundant living entities on our planet; they play major roles in organismal health, the environment, and in food systems. Consequently, research involving understanding the mechanisms of molecular microbiology will continue to be of utmost importance far into the future.
## References
1. Kaneko, Nakatsuka, Tsuneda (2025) "Phage Cocktails: State-of-the-Art Technologies and Strategies for Effective Design" *FEMS Microbiol. Rev*
2. Seal, Drider, Oakley et al. (2018) "Microbial-derived products as potential new antimicrobials" *Vet. Res*
3. Miller, Arias (2024) "ESKAPE pathogens: Antimicrobial resistance, epidemiology, clinical impact, and therapeutics" *Nat. Rev. Microbiol*
4. Bianco, Boattini, Cricca et al. (2024) "Updates on the Activity, Efficacy and Emerging Mechanisms of Resistance to Cefiderocol" *Curr. Issues Mol. Biol*
5. Marino, Stracquadanio, Cosentino et al. "Phage to ESKAPE: Personalizing Therapy for MDR Infections-A Comprehensive Clinical Review"
6. Zhu, Yang, Zhang et al. (2026) "Bacteriophage lysin P26ly combats multidrug-resistant Escherichia coli O157 and methicillin-resistant Staphylococcus aureus via cell wall degradation and membrane disruption" *Int. J. Biol. Macromol*
7. Ndiaye, Debarbieux, Sow et al. (2025) "Characterization of broad host range bacteriophages vKpIN31 and vKpIN32 against hospital-acquired Klebsiella pneumoniae in Dakar" *Senegal. Sci. Rep*
8. Echterhof, Dharmaraj, Blankenberg et al. (2025) "Wholebody distribution of three Pseudomonas phages characterized by a translational physiologically based pharmacokinetic model" *Antimicrob. Agents Chemother*
9. Wang, Zhao, Jiang et al. (2025) "Biological Characteristics and Genomic Analysis of Acinetobacter nosocomialis Lytic Phage XC1" *Curr. Issues Mol. Biol*
10. Vinderola, Druart, Gosálbez et al. (2023) "Postbiotics in the medical field under the perspective of the ISAPP definition: Scientific, regulatory, and marketing considerations"
11. Santacroce, Topi, Bottalico et al. (2024) "Current Knowledge about Gastric Microbiota with Special Emphasis on Helicobacter pylori-Related Gastric Conditions" *Curr. Issues Mol. Biol*
12. Denkel, Voss, Caselli et al. (2024) "Can probiotics trigger a paradigm shift for cleaning healthcare environments? A narrative review" *Antimicrob. Resist. Infect. Control*
13. D'accolti, Soffritti, Bini et al. (2024) "Tackling transmission of infectious diseases: A probiotic-based system as a remedy for the spread of pathogenic and resistant microbes" *Microb. Biotechnol*
14. Soffritti, D'accolti, Bini et al. (2025) "Probiotic-Based Approaches for Sustainable Control of Infectious Risk in Mass Transport: Current Data and Future Perspectives" *Microb. Biotechnol*
15. D'accolti, Soffritti, Bini et al. "Shaping the subway microbiome through probiotic-based sanitation during the COVID-19 emergency: A pre-post case-control study"
16. Libisch, Ozoaduche, Keresztény et al. (2025) "Detection of the ST111 Global High-Risk Pseudomonas aeruginosa Clone in a Subway Underpass" *Curr. Issues Mol. Biol*
17. Shem-Tov, Gal-Mor (2025) "HilE mediates motility thermoregulation in typhoidal Salmonella serovars at elevated physiological temperatures" *PLoS Pathog*
18. Akintunde, Tucker, Carabetta (2025) "The Evolution of Next-Generation Sequencing Technologies"
19. Benz, Beamud, Laurenceau et al. (2025) "CRISPR-Cas therapies targeting bacteria" *Nat. Rev. Bioeng*
20. Galindo, Lai (2025) "CRISPR-based genetic tools for the study of host-microbe interactions" *Infect. Immun*
21. Gelsinger, Ronda, Ma et al. (2025) "Metagenomic editing of commensal bacteria in vivo using CRISPR-associated transposases" *Science*
22. Whitaker, Russ, Stanley Shepherd et al. (2025) "Controlled colonization of the human gut with a genetically engineered microbial therapeutic" *Science*
23. Wang, Ding, Rong et al. (2024) "The Development of a CRISPR-FnCpf1 System for Large-Fragment Deletion and Multiplex Gene Editing in Acinetobacter baumannii" *Curr. Issues Mol. Biol*
24. Jhaveri, Weiss, Winkler et al. "A decade of clinical microbiology: Top 10 advances in 10 years: What every infection preventionist and antimicrobial steward should know" *Antimicrob. Steward. Healthc. Epidemiol*
25. Ackerman, Myhrvold, Thakku et al. (2020) "Massively multiplexed nucleic acid detection with Cas13" *Nature*
26. Kamariza, Mcmahon, Kim et al. (2025) "Multiplexed detection of febrile infections using CARMEN" *Nat. Commun*
27. Lasch, Beyer, Bosch et al. (2025) "A MALDI-ToF mass spectrometry database for identification and classification of highly pathogenic bacteria" *Sci. Data*
28. Xiong, Guan (2025) "Application of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in clinical testing and diagnosis" *Front. Cell. Infect. Microbiol*
29. Boattini, Guarrasi, Comini et al. (2025) "Diagnostic methods and protocols for rapid determination of methicillin resistance in Staphylococcus aureus bloodstream infections: A comparative analysis" *Eur. J. Clin. Microbiol. Infect. Dis*
30. Hernández-Torres, Torres-Mendoza, Navarro-Velasco et al. (2025) "Toward an Efficient Differentiation of Two Diaporthe Strains Through Mass Spectrometry for Fungal Biotyping" *Curr. Issues Mol. Biol*
31. Hassman, Rouchka, Sunino et al. "Molecular Point-of-Care Assay Development: Design and Considerations. Curr"
32. Rosas-Diaz, Palacios-Reyes, Godinez-Aguilar et al. (2025) "New Tool Against Tuberculosis: The Potential of the LAMP Lateral Flow Assay in Resource-Limited Settings"
33. Sołowi Ńska, Holec-G Ąsior (2025) "Development and Evaluation of Six Novel Recombinant GRA Proteins in Serodiagnosis of Human Toxoplasmosis" *Curr. Issues Mol. Biol*
34. (2023) "International Bank for Reconstruction and Development/The World Bank"
35. Hartman, Hibberd, Mostafa et al. (2024) "A microbiome-directed therapeutic food for children recovering from severe acute malnutrition" *Sci. Transl. Med*
36. Zhou, Hibberd, Lee et al. "Glycoside hydrolase-mediated glucomannan catabolism in Segatella copri, a target of microbiota-directed foods for malnourished children"
37. Van Syoc, Gomez, Davenport et al. (2025) "Gut fungal profiles reveal phylosymbiosis and codiversification across humans and nonhuman primates" *PLoS Biol*
38. Mishra, Tiwari, Kanchan et al. (2025) "Advances in Microbial Bioremediation for Effective Wastewater Treatment"
39. Butcher, Villette, Zumsteg et al. (2025) "Microbial bioremediation of persistent organic pollutants in plant tissues provides crop growth-promoting liquid fertilizer" *Nat. Commun*
40. Baev (2025) *Bioinformatics Research in Bacterial Genomics and Metagenomics. Curr. Issues Mol. Biol*
41. Bel Mokhtar, Asimakis, Galiatsatos et al. (2024) "Development of MetaXplore: An Interactive Tool for Targeted Metagenomic Analysis" *Curr. Issues Mol. Biol*
42. Gao, Ni, Hua et al. (2025) "Impact of SARS-CoV-2 Variant NSP6 on Pathogenicity: Genetic Analysis and Cell Biology" *Curr. Issues Mol. Biol*
43. Mouzakis, Petrakis, Tryfonopoulou et al. (2025) "Mechanisms of Immune Evasion in HIV-1: The Role of Virus-Host Protein Interactions" *Curr. Issues Mol. Biol*
44. Spackman, Jones, Mccoig et al. "Characterization of highly pathogenic avian influenza virus in retail dairy products in the US" *J*
45. Liu, Yang (2025) "An urgent need for diagnostic tools to address global mpox public health emergencies" *J. Clin. Microbiol*
46. Holmes, Krammer, Goodrum "Virology-The next fifty years" *Cell*
47. Al-Aly, Davis, Mccorkell et al. (2024) "Long COVID science, research and policy" *Nat. Med*
48. Ely, Brown, Fineberg (2024) "National Academies of Sciences, Engineering, and Medicine Committee on Examining the Working Definition for Long Covid. Long Covid Defined" *N. Engl. J. Med*
49. Aid, Boero-Teyssier, Mcmahan et al. (2025) "Long COVID involves activation of proinflammatory and immune exhaustion pathways" *Nat. Immunol*
50. Korobova, Arsentieva, Liubimova et al. (2025) *Cell Dynamics and Transitional B Cells in Long COVID. Curr. Issues Mol. Biol*
51. Peluso, Deeks "Mechanisms of long COVID and the path toward therapeutics" *Cell*
52. Wang, Wang, Liu "Artificial Intelligence for Microbiology and Microbiome Research. arXiv 2025"
53. Bard, Prinzi, Larkin et al. (2025) "Proceedings of the Clinical Microbiology Open 2024: Artificial intelligence applications in clinical microbiology" *J. Clin. Microbiol*
54. Pope, Varma, Tataru et al. (2025) "Learning a deep language model for microbiomes: The power of large-scale unlabeled microbiome data" *PLoS Comput. Biol*
55. Walsh, Stallard-Olivera, Fierer "Nine (not so simple) steps: A practical guide to using machine learning in microbial ecology"
56. Cao, De La Fuente-Nunez (2025) "Microbial Primer: Artificial intelligence for microbiologists" *Microbiology*
57. "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=PMC12671161&blobtype=pdf | # Gallic acid inhibits EMCV infection via targeting the interaction between TBK1 and IRF3 to promote IFN-β expression
Yan Zhang, Yanqiao Wen, Weijiao Xue, Tengyu Zhang, Jixia Hou, Xuewen Chen, Ruofei Feng, Chunxia Tan
## Abstract
Encephalomyocarditis virus (EMCV) is an important zoonotic pathogen with global distribution, which has caused huge serious economic losses to the development of the livestock industry. Gallic acid (GA) is an active ingredient of traditional Chinese medicine with wide pharmacological and biological activities and is also a potential antiviral drug. However, its antiviral effect and mechanism of anti-EMCV are still unclear.In the present study, we found that therapeutic administration of GA could significantly reduce the viral load and viral titer of EMCV to inhibit EMCV replication in a dosedependent manner. GA could also protect cells infected by EMCV and decrease the content of capsid protein VP1 of EMCV in HEK-293 cells. Additionally, the mechanistic investigations revealed that GA might exert antiviral effects by regulating the interac tion between TBK1 and IRF3 to promote the IFN-β expression. Meanwhile, GA could also alleviate EMCV-infected mice. These results indicate that GA may serve as a novel antiviral agent against EMCV infection. IMPORTANCE As a zoonotic pathogen, encephalomyocarditis virus (EMCV) causes myocarditis, encephalitis, neurological disease, reproductive disorders, and diabetes in pigs, which seriously endangers the development of the swine industry worldwide. However, due to the lack of effective commercial vaccines, there is an urgent need to develop safe and effective drugs against EMCV. Gallic acid (GA) has wide pharmacologi cal and biological activities. However, the antiviral effect and the mechanism of GA are currently unknown. Here, we demonstrated that GA had a significant anti-EMCV effect. Further research found that GA inhibited EMCV infection via targeting the interaction between TBK1 and IRF3 to promote the IFN-β expression. These findings indicate that GA could be an effective anti-EMCV drug.KEYWORDS gallic acid, EMCV, antiviral, IFN-β E ncephalomyocarditis virus (EMCV) is an important zoonotic virus (1), which is widely distributed and can be isolated from various mammals and patients with meningitis. Pigs, being susceptible animals to EMCV, mainly exhibit symptoms such as abortion in the late stage of pregnancy, stillbirth, mummification of piglets in the womb, increased mortality of weak and newborn piglets (2), as well as respiratory diseases in piglets (3). Therefore, EMCV can not only cause encephalitis and myocarditis but also lead to reproductive disorders in sows (4). The pathogenicity of EMCV isolates varies across different countries and regions, and even the same strain exhibits differing pathogenicity toward porcine fetuses of different gestational ages (5). In addition, in severe cases of human infection with EMCV, the clinical manifestations include neurological symptoms such as disturbance of consciousness, convulsions and coma, as well as cardiovascular symptoms such as heart failure and arrhythmia (6, 7). Therefore, EMCV not only has a huge impact on the livestock industry but also seriously endangers human health.
Thus, there is an urgent need to develop new drugs for the treatment of EMCV-related diseases.
During a viral infection, the innate immune response is crucial for restricting viral replication and initiating an immune response (8). Interferons (IFNs) can induce host cells to produce antiviral proteins, thereby inhibiting the replication and spread of viruses, which play a particularly important role in the innate immune response to viral infections and provide a powerful first line of defense against invading pathogens. According to sequence homology, IFNs are classified into three families: Type I, Type II, and Type III. Among them, the signal transduction of Type I IFN (IFN-I), also known as "viral interferon, " is the core of inducing antiviral innate immunity. IFN-I includes IFN-α, IFN-β, IFN-ε, IFN-κ, and IFN-ω. Currently, IFN-α and IFN-β have been proven to possess antiviral effects and immunomodulatory activities (9), and they can exhibit significant antiviral effects against a wide variety of viruses. The retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) can recognize viral RNA in the cytoplasm and initiate an immune response by inducing IFN-β. The binding of IFN-β to its receptor induces the activation of JAK1, JAK2, and tyrosine kinase 2, which can phosphorylate of STAT1 and STAT2 to form interferon-stimu lated gene (ISG) factor 3 complex (ISGF3). The ISGF3 enters the nucleus and binds to the IFN-stimulated response element, thereby activating the ISG, which induces the antiviral response of the host cell against viral infections (10)(11)(12)(13).
Traditional Chinese medicine has unique advantages in antiviral treatment owing to its broad spectrum of actions and targets, along with low side effects. Gallic acid (GA) (Fig. 1A), an organic acid, exists extensively in plants such as Galla Chinensis, Camellia sinensis, and Hevea brasiliensis. It has the highest contents in Galla Chinensis. Addition ally, GA exhibits a variety of biological activities (14,15). It has been reported that GA could inhibit the activation of the NLRP3 inflammasome and pyroptosis by enhanc ing the Nrf2 signaling pathway, thereby alleviating rheumatic arthritis (15). It can also reduce lipopolysaccharide-induced renal injury in rats by inhibiting cell pro-death and inflammatory responses (16) and exert anti-tumor effects through multiple biological pathways (17)(18)(19). However, currently, there are relatively few studies on the antiviral mechanism of GA. In this study, the inhibitory effect of GA on EMCV was explored, and its anti-EMCV mechanism was further investigated. The study aims to confirm the application prospect of GA against EMCV and provide a theoretical basis for the clinical use of GA in clinical practice and the research of related target drugs.
## MATERIALS AND METHODS
## Cells, viral strains, bacterial strains, and plasmids
The BHK-21 cells (hamster kidney fibroblast cells), HEK293 cells (human embryonic kidney cells), A549 (human non-small-cell lung cancer cells), and EMCV (PV21) virus were all preserved and provided by the Biomedical Research Center of Northwest Minzu University. The cells were cultured at a constant temperature in an incubator under conditions of 37°C and 5% CO₂. The culture media were Dulbecco's modified Eagle medium (DMEM) supplemented with 10% newborn bovine serum (NBS) or F12 medium supplemented with 15% NBS. During the cell resuscitation, culture, and passage, all operations must be carried out strictly in accordance with the aseptic principle. The competent cells of Escherichia coli BL21 were preserved and provided by the Biomedical Research Center of Northwest Minzu University. Empty vectors pCMV-HA and pCMV-Flag were provided by the Biomedical Research Center, Northwest Minzu University. The recombinant plasmids pCMV-HA-MDA5, pCMV-HA-TBK1, and pCMV-Flag-IRF3 were all constructed in-house.
## Experimental drugs and reagents
GA (149-91-7) was purchased from Shanghai Yuanye (Shanghai, China). NBS was purchased from Lanzhou Minhai (Lanzhou, China). DMEM and 0.25% trypsin were obtained from Lanzhou Bailing (Lanzhou, China). TIFIH1/MDA5 polyclonal antibody (21775-1-AP), MAVS polyclonal antibody (14341-1-AP), and IRF3 polyclonal antibody (11312-1-AP) were purchased from Proteintech (Wuhan, China). Anti-Flag tag rabbit polyclonal antibody (D110005), rabbit polyclonal antibody against the HA tag (D110004), HRP-conjugated goat anti-rabbit IgG (D110058), and HRP-conjugated goat anti-mouse IgG (D110087) were purchased from Sangon Biotech (Shanghai, China). TBK1 antibody (3013S) and phospho-IRF3 (Ser386) (37829 S) were bought from Cell Signaling Technol ogy. GAPDH was purchased from Beyotime Biotechnology (Shanghai, China). Lipofect amine 2000 was purchased from Invitrogen (Carlsbad, CA, USA). IRF3 siRNAs were designed and synthesized by Ribo Biotechnology (Guangzhou, China).
## Experimental methods
## Cytotoxicity experiment
BHK-21, A549, and HEK-293 cells were seeded in 96-well plates and incubated for 24 h. Three groups were established: blank group, control group, and treatment group. In the blank group, 10% NBS-DMEM (without cells) was added. In the control group, 10% NBS-DMEM (with cells) was added, and 10% NBS-DMEM (with cells) containing different concentrations of GA (5,10,20,40,80,160,320, and 640 µM) was added, with 200 µL per well. Each group had six replicate wells. After 24-h incubation, 10 µL of cell counting kit-8 (CCK-8) solution was added to each well. After further incubation at 37°C for 1.5 h, the absorbance was measured at 490 nm using a microplate reader, and the cell viability was calculated.
## Time addition assay
To assess the impact of GA on different stages of the EMCV life cycle, an addition-time assay was performed. The cells were treated with GA (20-160 µM) at different time points during the EMCV PC21 (100 TCID 50 ) infection, including pre-treatment, co-treat ment, and post-treatment. The experiment progress is shown in Fig. 2A. The cells and supernatants were collected at 24 h. Subsequently, the samples were subjected to freeze-thaw cycles for three times and centrifuged. The supernatant was carefully harvested, and the total viral genome was extracted using the Tiangen TIANamp DNA/RNA Kit. The extracted RNA was used as a template for reverse transcription into cDNA. Finally, qPCR was used to determine copies of viral genomes.
## Virus titer
Samples of GA at concentrations of 80 and 160 µM, with different treatment times (12,24, and 36 h), were prepared according to the treatment method described above. After undergoing repeated freezing and thawing cycles, the supernatant was collected. BHK-21 cells were infected with virus samples that had been serially diluted 10-fold in a 96-well plate, with eight replicates for each dilution gradient. Two hours later, the medium was replaced with a maintenance medium consisting of 3% NBS-DMEM. After 3-5 days, the number of cytopathic-positive wells was observed and counted under a microscope, and the virus titer was calculated using the Reed-Muench method.
## Cytoprotective effect
HEK-293 cells were seeded in 6-well plates. When the cells grew to more than 90%, they were infected with EMCV (100 TCID 50 ) for 2 h. Subsequently, the culture medium was replaced with 3% NBS-DMEM supplemented with GA (80/160 µM), and the cells were cultured for 24 h. The cell status was observed under a microscope and photographed for documentation.
## RT-qPCR
Total cellular RNA was extracted using the total cellular RNA extraction reagent TranzolUp (TransGen Biotech, Beijing, China). The total RNA was quantitatively reversetranscribed into cDNA. Then, a two-step RT-qPCR was carried out using a SYBR Green assay with an Applied Biosystems Master Mix kit in an ABI 7500 Real-Time PCR system. GAPDH was used as the internal reference. The relative gene expression levels were calculated using the 2 -ΔΔCt method, and the results were presented as the mean value ± standard deviation (SD). Table 1 shows the primer sequences used for RT-qPCR analysis.
## Western blot detection
The HEK-293 cells were lysed on ice for 30 min with a high-efficiency RIPA tissue/cell lysis buffer (Solarbio, Beijing, China). The samples were then subjected to separation via SDS-PAGE and subsequently transferred onto a PVDF membrane (Millipore Corp, Bedford, MA, United States). Next, the membrane was incubated in a blocking buffer (TBST supplemented with 2.5% skimmed milk powder) at room temperature for a duration of 2 h. After incubation, the blocking solution was discarded, and the corre sponding primary antibodies were individually added to the membrane, followed by an overnight incubation at 4°C. Subsequently, the corresponding secondary antibodies were incubated with the membrane at room temperature for 1 h. After discarding the secondary antibodies, the PVDF membrane was washed and imaged using ECL (Bio-Rad, CA, United States) chemiluminescence. The gray values of protein bands were analyzed using ImageJ software.
## Plasmid transfection
HEK-293 cells were inoculated into a 6-well cell culture plate. When the cells grew to more than 80%, the plasmid was transfected into the HEK-293 cells via the lipo some-mediated transfection approach, which the plasmid HA-TBK1 and Flag-IRF3 were undergone co-transfection during the co-immunoprecipitation assay, while all the others transfected single plasmid. Four to six hours after transfection, the culture medium was refreshed. Following this, the cells were incubated at 37°C for a period of 24 h. Subsequently, the medium was substituted with 10% NBS-DMEM, either supplemented or not supplemented with GA. After an additional 24-h culture period, detection was carried out using RT-qPCR and Western blot analysis.
## RNAi assay
siRNA targeting IRF3 was transfected into HEK-293 cells to verify the effect of IRF3 knockdown. Lipofectamine 2000 was used to transfect HEK-293 cells with siNC or siIRF3 for 24 h, followed by inoculation with 100 TCID 50 EMCV for 2 h. Subsequently, the culture medium was replaced with 3% NBS-DMEM supplemented with or without GA (160 µM), and the cells were cultured for another 24 h. The viral titers and the expression of IFN-β were tested using the TCID 50 assay and enzyme-linked immunosorbent assay (ELISA), respectively.
## Immunofluorescence
A549 cells cultured in a 12-well cell culture plate were infected with EMCV (100 TCID 50 ) for 2 h. Subsequently, the culture medium was replaced with 3% NBS-F12 medium with or without GA, and the cells were further cultured for 24 h. After that, the cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with a blocking buffer for 30 min. Later, the cells were incubated with the VP1 primary antibody (1:1,000) overnight at 4°C, followed by incubation with Cy3-conjugated goat anti-mouse IgG (diluted 1:500) at 4°C for 2 h and then stained with 4′,6-diamino-2-phenylindole (DAPI) for 10 min. Finally, images were visualized on a ZEISS LSM900 laser confocal microscope (Carl Zeiss, Oberkochen, Germany).
## Co-Immunoprecipitation
HEK-293 cells were lysed with lysate on ice for 30 min to extract total protein. A portion of the protein samples was directly subjected to Western blotting assays to verify the presence of the target protein. The tag antibody was incubated with magnetic beads on a shaker at 4°C for a duration ranging from 4 to 6 h. Subsequently, the remaining untreated protein samples were added, and the mixture was incubated on a shaker at 4°C overnight. The supernatant was carefully discarded, while the magnetic beads at the bottom layer were retained. The magnetic beads were then washed five times with pre-cooled PBS. The precipitates were combined with SDS buffer and boiled at 95°C for 5 min. Finally, the complexes were analyzed using Western blotting.
## Animal experiments
## Acute toxicity of GA in mice
The initial dose of GA was set at 1,800 mg/kg and then successively reduced in a 3/4 concentration gradient. A total of five concentration gradients were established. The mice were administered different concentrations of GA via gavage at a dosage of 0.2 mL/ 10 g body weight, and the mortality of the mice was observed. The death status of the mice at each concentration gradient was recorded to calculate the median 50% lethal dose (LD 50 ) of GA.
## Establishment of an EMCV-infected mouse model
The experimental design is shown in Fig. 6A. The BABL/c mice were randomly divided into the mock group, EMCV group, L-GA (40 mg/kg) group, and H-GA (80 mg/kg) group with 13 mice in each group. On the fifth day, when deaths occurred in the EMCV group, five mice were randomly sacrificed from each group for other tests such as copies of viral genomes determination, and other eight mice were used for plotting the survival curve. Except for the mock group, the mice in other groups were intramuscularly injected with 250 µL/10 g of 100 TCID 50 EMCV. Post-infection, the mice were gavaged daily with the corresponding dose of GA with a treatment volume of 0.2 mL/10 g and the corresponding volume of PBS solution in the mock and EMCV groups. The body weight was monitored, and the clinical score was recorded daily. The clinical symptom scoring criteria are as follows: 0: healthy; 1: the mouse is listless; 2: loss of appetite; 3: head edema and ruffled fur; 4: hind-limb paralysis and neurological symptoms; 5: death. The remaining mice were continuously cultured for 3 days after all the mice in the EMCV-infected group had died, and a survival curve was plotted.
## Enzyme-linked immunosorbent assay
The contents of IFN-β in cell supernatants and brain tissues of mice were detected using a human and mouse IFN-β ELISA kit (Multi Sciences, Hangzhou, Zhejiang, China) according to the manufacturer's instructions.
## Statistical analysis
Statistical analysis was performed with SPSS version 26.0 (IBM, Armonk, NY, United States). All data were presented as the mean ± standard deviation (SD) of at least three independent experiments unless otherwise noted. All the figures were plotted using GraphPad Prism software (Version 8.0, La Jolla, CA, United States). Statistical significance was determined via one-way ANOVA with Tukey's multiple comparison test or two-tailed t tests. Statistical significance was set at P < 0.05.
## RESULTS
## Cytotoxicity of GA
First, the cytotoxicity of GA in BHK-21, A549, and HEK-293 cells was investigated using the CCK-8 assay. The measurement results indicated that GA exerted cytotoxic effects at a concentration of 640 µM in A549 and HEK-293 cells, and cytotoxicity was detected at a concentration of 320 µM in BHK-21 cells (Fig. 1B through D). Consequently, all subsequent experiments were conducted within the safe concentration range (5-160 μM) of GA.
## GA significantly inhibits the proliferation of EMCV
To analyze the time of GA exerted antiviral effects, GA was applied at different time points for intervention. It was found that compared with the infected group, during the pre-treatment stage before EMCV infection, GA could effectively inhibit the copies of viral genomes of EMCV (Fig. 2B). When GA was applied simultaneously with EMCV infection (co-treatment), GA had no significant antiviral effect on EMCV (Fig. 2C). GA showed the best inhibitory effect during the post-treatment after EMCV infection, and the concentrations ranging from 20 to 160 µM of GA could effectively inhibit the replication of EMCV (Fig. 2D). Meanwhile, we further detected the copies of viral genomes of EMCV and viral titer to confirm the antiviral effect of GA at different time points during EMCV replication using RT-qPCR and TCID 50 assays. As shown in Fig. 2E and F, GA could significantly inhibit the copies of viral genomes of EMCV and decrease virus titer at 12, 24, and 36 h post-infection, respectively. And the antiviral effect was most pronounced at 24 h with a dose-effect relationship. Therefore, the subsequent experiments were conducted at 24 h. Simultaneously, the morphological changes of cells in each group were observed. The results showed that compared with the uninfected group, the infected group exhibited increased cell gaps, cell shrinkage, and patches of cell death, while GA could significantly inhibit the cytopathic effects induced by EMCV (Fig. 2G). In addition, the viral protein VP1 of EMCV is the main protein promoting its proliferation. Therefore, the expression of VP1 was detected by immunofluorescence to further verify the inhibitory effect of GA on EMCV. The results showed that GA could significantly reduce the VP1 level in A549 cells (Fig. 2H), indicating that GA could effectively inhibit the expression of VP1. The above results suggested that GA could significantly inhibit the proliferation of EMCV in vitro.
## GA exerts anti-EMCV effect by promoting the expression of IFN-β
After the GA treatment of EMCV-infected HEK-293 cells for 24 h, the expression of Bax, Bcl-2, TNF-α, IL-6, and IFN-β was detected by RT-qPCR and ELISA. Compared with the infected group, treatment with GA (80/160 µM) significantly increased the expression of IFN-β in EMCV-infected HEK-293 cells. However, it had no significant effect on the expressions of Bax, Bcl-2, TNF-α, and IL-6 (Fig. 3A through G). These results indicated that GA might inhibit EMCV replication by increasing the production of IFN-β. And this effect had nothing to do with the drug itself (Fig. 3H). Further studies demonstrated that GA could promote the expression of IFN-β at 12, 24, and 36 h post-infection (Fig. 3I). The above results indicated that GA could exert anti-EMCV effect by promoting the expression of IFN-β.
## GA induces the expression of IFN-β by regulating the RLR signaling pathway
After HEK-293 cells were infected with EMCV, the protein levels of TBK1 and p-IRF3 increased significantly. Compared with the mock group, the expression of TBK1, p-TBK1, IRF3, and p-IRF3 increased significantly following GA intervention (Fig. 4A). As shown in Fig. 4B through D, to further verify the target of GA, HA-MDA5, Flag-TBK1, and Flag-IRF3 (5D) were overexpressed in HEK-293 cells, and the effect of GA on the mRNA of IFN-β was examined. The results demonstrated that after overexpressing TBK1 and IRF3, GA could significantly elevate the mRNA level of IFN-β. The above results suggest that GA may promote the production of IFN-β by activating the RLR signaling pathway and regulating the expression of TBK1 and IRF3.
## GA induces the expression of IFN-β by promoting the interaction between TBK1 and IRF3
HEK-293 cells were transfected with Poly(I:C) to activate the RLR signaling pathway and then treated with GA. The results showed that after Poly(I:C) activated the RLR signaling pathway, the expressions of TBK1, p-TBK1, IRF3, and p-IRF3 were all significantly upregulated. Compared with the Poly(I:C) group, GA could further significantly enhance the expressions of TBK1, p-TBK1, IRF3, and p-IRF3 (Fig. 5A) and significantly increase the level of IFN-β mRNA (Fig. 5B). To further explore the mechanism of GA-induced expression of IFN-β, the effect of GA on the interaction between TBK1 and IRF3 was detected using the co-immunoprecipitation method. The results showed that GA could promote the interaction between TBK1 and IRF3 (Fig. 5C). In addition, an RNAi assay was performed in HEK-293. The results showed that the IRF3 (si003) expression was decreased significantly in the siRNA transfection group, as shown in Fig. 5D. Then we analyzed the effect of IRF3 knockdown on GA's anti-EMCV. As shown in Fig. 5E, following GA addition, the virus titers of the IRF3 knockdown HEK-293 cells were higher than those of the siNC control cells. Besides, the expression of IFN-β was decreased in the siIRF3 group compared to the control siNC group (Fig. 5F). The above results suggest that GA targeted IRF3 to regulate the interaction between TBK1 and IRF3 to induce the expression of IFN-β, thereby playing an anti-EMCV role.
## GA alleviates EMCV infection in mice
The LD 50 of GA was not detected in mice acute toxicity test. According to the refer ence, the therapeutic doses of GA were set at 40 mg/kg (L-GA) and 80 mg/kg (H-GA), respectively. During the experiment, mice infected with EMCV showed depression, huddling, unkempt and dull fur, reduced dietary wishes, and neurological symptoms such as hind limb paralysis on the fourth day. GA could ameliorate the above symptoms. Moreover, GA significantly increased the survival rate of mice (Fig. 6B) and effectively prevented weight loss in mice following EMCV infection (Fig. 6C). To further investigate the protective effects of GA on different organs of EMCV-infected mice, the copies of viral genomes in the brain tissues and heart of mice were detected. As shown in Fig. 6D andE, the results showed that GA could remarkably reduce the copies of viral genomes in the brain tissues and heart of mice. The TCID 50 results indicated that GA could significantly decrease the viral titer in the brain tissues of mice (Fig. 6F). Simultaneously, GA significantly promoted the production of IFN-β in the brain tissues of EMCV-infected mice (Fig. 6G). The above results suggested that GA exerted a significant antiviral effect in mice infected with EMCV.
## DISCUSSION
As a zoonotic virus, EMCV not only severely impacts the pig husbandry but also presents a potential threat to human health. There is an urgent necessity to develop novel drugs for its treatment. GA has widespread pharmacological effects, including antioxidant, anti-inflammatory, antibacterial, and anti-tumor properties. Studies have demonstrated that it could inhibit a variety of viruses (20)(21)(22)(23); however, its specific antiviral mechanism remains unclear. The results of this study indicate that GA has a certain inhibitory effect on EMCV. It can significantly decrease the viral titer of EMCV, effectively reduce cell rupture and intercellular spaces induced by EMCV infection, and display good cytoprotective effects. VP1, as the main structural protein of EMCV, is its major antigenic determinant domain, which is closely associated with the virus pathogenicity. It also interacts with receptors on the cell surface and can stimulate the body to generate neutralizing antibodies (2,24,25). Our results showed that GA can remarkably reduce the expression of the viral protein VP1 in cells.
In order to further explore the mechanism of GA's antiviral effect, the regulatory effect of GA on TNF-α, IL-6, Bcl2, Bax, and IFN-β was detected. During viral infection, the NF-κB-signaling pathway is activated. Meanwhile, the dimer of p65/p50 is released from the cytoplasm into the nucleus and induces the expression of cytokines, further recruiting immune cells to clear viral infections. In addition, the infected cells promoted apoptosis, thereby inhibiting the replication and spread of the virus (26,27). Bax, a pro-apoptotic protein within the Bcl2 family, promotes cell apoptosis by increasing the permeability of the mitochondrial membrane and releasing cytochrome. Bcl-2 is an anti-apoptotic protein, mainly distributed on the outer membrane of mitochondria, and it maintains cell survival by inhibiting apoptosis (28). Our results showed that GA had no significant effects on TNF-α, IL-6, Bcl2, and Bax, but it could significantly upregulate the expression of IFN-β during EMCV infection, which revealed that GA could inhibit the replication of EMCV by regulating the expression of IFN-β, rather than the NF-κB signal pathway and cell apoptosis.
IFN-β, as a type I IFN, plays a crucial role in the innate immune system of the body against viral infection. The host innate immunity recognizes pathogen-associated molecular patterns of viruses via pattern recognition receptors. The RLR family consists of RIG-I, melanoma differentiation-associated protein 5 (MDA5), and laboratory genetics and physiology protein 2 (LGP2) (1,29,30). Among them, MDA5 mainly recognizes long double-stranded RNA (dsRNA). The dsRNA of the replication intermediate of EMCV is approximately 7.8 kb, which falls into the category of long dsRNA. Furthermore, our previous studies have demonstrated that EMCV-infected cells mainly activate the innate immune antiviral response by recognizing its RNA through MDA5 (30,31). When MDA5 binds to dsRNA, it will promote its oligomerization and interact with MAVS. The activated MAVS recruits and activates TBK1/IKKε. TBK1/IKKε phosphorylates IRF3/IRF7 through its kinase activity, thereby inducing IFN-I to exert antiviral effects (6,(32)(33)(34). Therefore, drugs can effectively control viral infections by regulating the RLR signaling pathway. Existing studies have shown that baicalin could upregulate the expressions of IFN-I and IFN-III and their receptors under the stimulation of Poly (I:C) (35). Consistent with the above results, the findings of this study suggested that GA exerted anti-EMCV effect by promoting the expression of IFN-β through regulating the RLR signaling pathway. Specifically, GA targeted TBK1 to interact with IRF3, leading to a conformational change in phosphorylated IRF3, forming a dimer, translocating from the cytoplasm to the nucleus, binding to its co-activators, and then initiating the production of IFN-β. In summary, GA exerted an antiviral effect against EMCV infection by activating TBK1 and IRF3 in the RLR signaling pathway and promoting the production of IFN-β.
In this study, we further validated the protective effect of GA in the EMCV-infected mouse model. We measured the changes in copies of viral genomes in the brain tissues and hearts, the viral titer, and the level of IFN-β in the brain tissues of EMCV-infected mice. The results indicated that GA could significantly decrease the copies of viral
## Conclusion
This study demonstrates that GA can effectively inhibit the infection of EMCV both in vitro and in vivo. The specific mechanism of its anti-EMCV mainly involves promoting the interaction between TBK1 and IRF3 and subsequently upregulating the expression of IFN-β to inhibit the proliferation of EMCV. This research provides guidance for the clinical treatment of EMCV-related diseases, offers new insights into elucidating the pharmaco dynamic mechanism of GA, and provides a theoretical foundation for expanding its clinical applications and developing drugs against EMCV (Fig. 7).
## References
1. Amarante-Mendes, Adjemian, Branco et al. (2018) "Pattern recognition receptors and the host cell death molecular machinery" *Front Immunol*
2. Boege, Scraba (1989) "Mengo virus maturation is accompanied by C-terminal modification of capsid protein VP1" *Virology (Auckl)*
3. Murname, Wei (1986) "Encephalomyocarditis virus"
4. Shaotang (1994) "The latest understanding of encephalomyocarditis virus"
5. Carocci, Bakkali-Kassimi (2012) "The encephalomyocarditis virus" *Virulence*
6. Tesh (1978) "The prevalence of encephalomyocarditis virus neutraliz ing antibodies among various human populations" *Am J Trop Med Hyg*
7. Pope, Scott (1960) "A survey for antibodies to encephalomyocarditis virus in man and animals" *Aust J Exp Biol Med Sci*
8. Belsham (2009) "Divergent picornavirus IRES elements" *Virus Res*
9. Hervas-Stubbs, Perez-Gracia, Rouzaut et al. (2011) "Direct effects of type I interferons on cells of the immune system" *Clin Cancer Res*
11. Beadling, Guschin, Witthuhn et al. (1994) "Activation of JAK kinases and STAT proteins by interleukin-2 and interferon alpha, but not the T cell antigen receptor, in human T lymphocytes" *EMBO J*
12. Bluyssen, Durbin, Levy (1996) "ISGF3γ p48, a specificity switch for interferon activated transcription factors" *Cytokine Growth Factor Rev*
13. Lazear, Schoggins, Diamond (2019) "Shared and distinct functions of type I and type III interferons" *Immunity*
14. Platanias (2005) "Mechanisms of type-I-and type-II-interferonmediated signalling" *Nat Rev Immunol*
15. Pérez-Delgado, Villa, Fernández-Quiroz et al. (2024) "Clicking gallic acid into chitosan prolongs its antioxidant activity and produces intracellular Ca 2+ responses in rat brain cells" *Int J Biol Macromol*
16. Jiang, Pei, Zheng et al. (2022) "Gallic acid: a potential anti-cancer agent" *Chin J Integr Med*
17. Luan, Ou, Hu et al. (2023) "Gallic acid alleviates lipopolysaccharideinduced renal injury in rats by inhibiting cell pro-death and inflammatory response and its mechanism" *Cell Mol Biol*
18. Tuli, Mistry, Kaur et al. (2022) "Gallic acid: a dietary polyphenol that exhibits anti-neoplastic activities by modulating multiple oncogenic targets" *Anticancer Agents Med Chem*
19. Kaviani, Alvani, Shayanfar (2024) "Cocrystallization of 5-fluorouracil with gallic acid: a novel 5-fluorouracil cocrystal displaying synergistic anti-tumor activity both in oral and intraperito neal injection administration" *Eur J Pharm Biopharm*
20. Hassani, Ghanbari, Lotfi et al. (2023) "How gallic acid regulates molecular signaling: role in cancer drug resistance" *Med Oncol*
21. Zhang, Ye, Gan et al. (2024) "Gallic acid alleviates psoriasis keratinization and inflammation by regulating BRD4 expression" *Fol Biol*
22. Ye, Su, Li et al. (2023) "Enhanced in vivo antiviral activity against pseudorabies virus through transforming gallic acid into graphene quantum dots with stimulation of interferon-related immune responses" *J Mater Chem B*
23. Govea-Salas, Rivas-Estilla, Rodríguez-Herrera et al. (2016) "Gallic acid decreases hepatitis C virus expression through its antioxidant capacity" *Exp Ther Med*
24. Hadidi, Liñán-Atero, Tarahi et al. (2024) "The potential health benefits of gallic acid: therapeutic and food applications" *Antioxidants (Basel)*
25. Kobasa, Mulvey, Lee et al. (1995) "Characterization of Mengo virus neutralization epitopes. II. Infection of mice with an attenuated virus" *Virology (Auckl)*
26. Sin, Sung, Suh et al. (1997) "Protective immunity against heterologous challenge with encephalomyocarditis virus by VP1 DNA vaccination: effect of coinjection with a granulocytemacrophage colony stimulating factor gene" *Vaccine (Auckl)*
27. Blank, Kourilsky, Israël (1992) "NF-κB and related proteins: Rel/dorsal homologies meet ankyrin-like repeats" *Trends Biochem Sci*
28. Schottelius, Baldwin (1999) "A role for transcription factor NF-κB in intestinal inflammation" *Int J Colorectal Dis*
29. (1007)
30. King, Hohorst, García-Sáez (2023) "Expanding roles of BCL-2 proteins in apoptosis execution and beyond" *J Cell Sci*
31. Carty, Guy, Bowie (2021) "Detection of viral infections by innate immunity" *Biochem Pharmacol*
32. Song, Li, Lian et al. (2024) "Histone H1.2 inhibited EMCV replication through enhancing MDA5-mediated IFN-β signaling pathway" *Viruses*
33. Li, Ma, Wu et al. (2021) "HSP27 protein dampens encephalomyocarditis virus replication by stabilizing melanoma differentiation-associated gene 5" *Front Microbiol*
34. Jiang (2019) "Structural variability in the RLR-MAVS pathway and sensitive detection of viral RNAs" *Med Chem*
35. Chen, Hu, Xu et al. (2016) "MSX1 modulates RLRmediated innate antiviral signaling by facilitating assembly of TBK1associated complexes" *J Immunol*
36. Fu, Shao, Wang et al. (2023) "Bat MAVS involved in antiviral innate immunity via regulating IFN-beta production" *Dev Comp Immunol*
37. Li, Dong, Ye et al. (2023) "Baicalin promotes antiviral IFNs production and alleviates type I IFNinduced neutrophil inflammation" *J Nat Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12691581&blobtype=pdf | # Correction for Woodson et al., "Efficacy and in vitro pharmacological assessment of novel N-hydroxypyridinediones as hepatitis B virus ribonuclease H inhibitors"
Molly Woodson, Holly Walden, M Mottaleb, Maria Makri, Georgia-Myrto Prifti, Dimitrios Moianos, Vasiliki Pardali, Grigoris Zoidis, John Tavis
## Abstract
Table 3: The fourth column should read as shown in this correction. The EC 50 and CC 50 values in the original table were correct, but we inadvertently calculated the SI values incorrectly. This error did not materially affect any conclusions within the paper. We apologize for this mistake.
Address correspondence to John E. Tavis, john.tavis@health.slu.edu. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12729987&blobtype=pdf | # Evaluation of Three Recombinant Antigens for the Detection of Anti-Coxiella Antibodies in Cattle
Barbara Colitti, Consiglia Longobardi, Gabriela Flores-Ramirez, Chiara Nogarol, Ludovit Skultety, Gianmarco Ferrara
## Abstract
Background/Objectives: The detection of anti-Coxiella antibodies using serological methods is essential for identifying exposed ruminants and preventing this important zoonotic disease in livestock. In recent years, numerous attempts have been made to increase diagnostic performance as well as simplify the production of serological assays. Commercially available tests often use whole-cell antigens, which can decrease specificity and require high-level biosafety facilities for manufacturing. The aim of this work was to produce three Coxiella burnetii (C. burnetii) antigens in recombinant form and assess them for the detection of anti-Coxiella antibodies in ruminants. Methods: Three recombinant C. burnetii antigens (Com-1, MceB, AdaA) were selected among immunodominant antigens and produced in a heterologous system (Escherichia coli). Following purification, the proteins were utilized to coat ELISA plates and evaluated for seroreactivity against sera from both negative and positive cattle. Results: Com-1 demonstrated the greatest agreement with the commercial test, albeit moderate. MceB exhibited nonspecific reactivity against a large number of sera, while the AdaA showed reactivity against only a few positive sera. Conclusions: Our findings are consistent with previous research, indicating that utilizing a single antigen to identify exposed animals is unfeasible with current knowledge, most likely due to the complex immunological response following C. burnetii infection in cattle. Consequently, it is critical to continue testing and identifying immunoreactive antigens in order to further investigate them and, potentially, select the most appropriate.
## 1. Introduction
Coxiella burnetii (C. burnetii) is the etiological agent of Q fever in humans, a globally reported public health concern, and Coxiellosis in ruminants, which causes economic losses in the cattle industry [1,2]. This intracellular and Gram-negative bacterium has a wide host range, including almost all mammals, and has been identified as one of the most common abortive agents in ruminants [1,3]. Coxiellosis in ruminants is often asymptomatic, with the sole symptoms being reproductive problems such as abortion, metritis, and placental retention [4,5]. Ruminants are considered the primary source of infection for humans, as they shed Coxiella in high concentrations through vaginal fluids and abortion products [6,7]. Other species, although susceptible to infection, appear to have a minor role in transmission [8][9][10]. Humans become infected through inhalation or ingestion of contaminated materials (some professionals, such as veterinarians and farmers, are at high risk) and may develop an acute form, characterized by fever and flu-like symptoms, or a chronic form, which is more dangerous due to Coxiella colonization in various organs such as the liver, lungs, and heart [1,11,12].
Ruminants are the main target for surveillance and monitoring programs due to their crucial role in the epidemiological cycle of Coxiella [11,13,14]. Surveillance strategies involve the detection of Coxiella DNA through direct methods, such as polymerase chain reaction testing on milk or aborted materials [15,16]. Additionally, serological tests are used to detect specific antibodies to assess the level of exposure within a herd [5,17]. Direct approaches are typically more expensive and are mainly recommended in outbreak situations, as intermittent shedding of Coxiella can impair diagnostic sensitivity [18,19]. Serology, on the other hand, is more suitable for rapid, large-scale screening. Previously, the complement fixation test (CFT) was the preferred serological tool, but it has largely been replaced by the immunofluorescence assay (IFA) and the enzyme-linked immunosorbent assay (ELISA) [18,20]. IFA is regarded as the gold standard for human diagnosis. It allows differentiation between recent and past exposure by detecting antibodies against two distinct antigenic phases of Coxiella: phase II (typical of recent and acute infection) and phase I (indicative of chronic infection) [1,21]. However, since no commercial IFA is currently available for veterinary use, ELISA is the most commonly used tool in ruminants and is also recommended by the World Organization for Animal Health (WOAH) [22]. These tests are relatively rapid, accessible, and easily applied to herd assessment, but they also have several drawbacks. In fact, commercial ELISAs are mostly based on whole-cell antigens of a single isolate, whose manufacturing requires a biosafety level 3 laboratory (BSL3) [23,24]. Furthermore, diagnostic performance may vary depending on the ruminant species or matrix used and may be influenced by antigenic homology between C. burnetii and other common ruminant pathogens, as well as the diversity of genotypes circulating in a given area [20,23,25,26].
Improving detection tools is essential for effective infection control, and numerous efforts have been made in recent years to develop recombinant ELISAs to address performance limitations and reduce the need for high-level biosafety procedures. Important immunodominant C. burnetii antigens (Ybgf, SucB, Hstp, etc.) have been employed in recombinant ELISAs. Comparative studies with commercial ELISAs have shown high concordance rates, although these alternatives have not been deemed suitable to fully replace existing commercial tests [24,[27][28][29].
For these reasons, it is critical to continue identifying novel recombinant candidates, as the control of Q fever and Coxiellosis necessitates constant improvement of the assay used.
Several studies have proposed that major immunodominant proteins might be used to produce novel serological assays and vaccines, as they are often associated with a specific antigenic phase or form of infection [30,31]. These immunoreactive antigens include an outer membrane-associated protein (Com-1, CBU1910), acute disease antigen A (AdaA, CBU0952), and the mitochondrial Coxiella effector protein B (MceB, CBU0937) [32][33][34]. Com-1 is an outer membrane protein that participates in post-translational modification and protein turnover [35]. This antigen has been identified as immunodominant in multiple studies, demonstrating strong reactivity for both phases, and has already been used to detect anti-Coxiella antibodies in ruminants [36]. MceB and AdaA are two proteins that have demonstrated strong reactivity to phase II and are considered biomarkers of acute Coxiella infection [33].
The aim of this study was to produce these three antigens in a heterologous system (Escherichia coli), purify them, and employ them to coat ELISA plates in order to compare their diagnostic performance by testing positive and negative bovine sera.
## 2. Materials and Methods
## 2.1. Antigen Selection, Amplification, Ligation
A C. burnetii Nine Mile RSA 493 phase I strain was cultivated in axenic medium in a BSL3 laboratory in the department of Rickettiology, Slovak Academy Science, Bratislava (Slovakia) as described previously [37]. The bacteria were centrifuged at 15,000× g for 1 h at 4 • C, and genomic DNA was extracted using DNeasy blood and tissue kit (Qiagen, Venlo, The Netherlands) following the manufacturer's instructions. Specific primers were designed according to the gene sequences available in the NCBI database (https://www.ncbi.nlm.nih.gov/; accessed on 1 October 2023), excluding the signal peptide region (Table 1). The primers included recognition sites for the restriction enzymes BamHI and EcoRI for cloning. However, since a restriction site for EcoRI was present in the sequence encoding for Com-1, this enzyme was replaced with XhoI (Supplementary Table S1). Each amplification was performed using a commercial kit (HotStarTaq DNA Polymerase, Qiagen) in a total volume of 50 µL including 2.5 µL of each primer (10 µM), 5 µL of buffer 10×, 1.5 µL of MgCl 2 (50 mM), 1 µL of dNTPs (10 mM), 0.25 µL of Taq DNA polymerase, and RNase-free water. PCR amplification was carried out under the following conditions: an initial denaturation at 95 • C for 5 min, followed by 35 cycles of denaturation at 95 • C for 45 s, annealing at 52-60 • C for 30 s, and elongation at 72 • C for 1-2 min depending on the amplicon length. A final elongation step at 72 • C for 10 min was also performed. Amplicons were visualized on agarose gel to confirm the presence and size of the expected bands. Digestion was performed at 37 • C for 3 h using appropriate restriction enzymes (BamHI, EcoRI, XhoI; Thermo Scientific, Waltham, MA, USA), followed by purification and quantification assessed using NanoDrop 2000/2000c spectrophotometer (Thermo Fisher, Waltham, MA, USA). Ligation was carried out as described in previous works [24] using specific plasmid vectors (pSER or pGEX 6P-1) previously digested with the same restriction enzymes. The ligation product was utilized for transforming competent Escherichia coli BL21 C43 (DE3) cells grown in Luria-Bertani (LB) media (Thermo Scientific, Waltham, MA, USA). Positive colonies were cultivated in ampicillin-supplemented liquid LB and stimulated with 1 mM isopropyl D-1-thiogalactopyranoside (IPTG) during the mid-exponential phase. The insert's features and in-frame orientation were confirmed by colony PCR and by Sanger sequencing of the plasmid DNA extracted from a positive clone using the QIAprep Spin Miniprep kit (Qiagen, Venlo, The Netherlands). Abbreviations: + = positive; -= negative.
## 2.2. Expression and Purification
Bacterial cells were collected by centrifugation (6000× g 10 min at 4 • C) and processed with 100 µL of lysozyme (50 mg/mL) in 10 mL of STE buffer (NaCl at 100 mM, Tris HCl at 10 mM, and Na2EDTA at 1 mM; Sigma-Aldrich, Burlington, MA, USA). Proteins were extracted and purified differently based on their solubility (presence or absence in the supernatant) using either denaturing or native conditions coupled to column-based chromatography. Specifically, MceB and AdaA (cloned into pSER vector) were purified using 1 M urea (from pellet) and a nickel-affinity resin (HisPur Ni-NTA resin; Thermo Scientific). Com-1, on the other hand, was cloned into the pGEX-6P-1 plasmid and purified from the supernatant using glutathione affinity chromatography (Glutathione Sepharose 4B resin; Merck, Rahway, NJ, USA), which also allowed the cut of the fusion GST tag from purified proteins directly on-column using PreScission protease (Merck, Rahway, NJ, USA) [24]. Each purification was carried out in three steps, after 30 min of resin incubation with protein extract and 5 min of incubation with a specific elution buffer, and visualized in Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) stained with Coomassie brilliant blue R250 (Sigma-Aldrich, Burlington, MA, USA).
## 2.3. Recombinant ELISAs
Each recombinant antigen was used (100 ng for Com-1; 50 ng for MceB and AdaA) to coat the wells of 96-well plates (Nunc Maxisorp; Millipore Sigma, Burlington, MA, USA) overnight at 4 • C, followed by a blocking phase with 2.5% bovine casein to prevent nonspecific binding. Serum samples were diluted 1:20 for Com-1 and 1:100 for MceB/AdaA, and incubated for 1 h at room temperature. After three washes with a specific wash buffer, a monoclonal anti-bovine IgG-peroxidase was used as previously described [14]. After a final wash step, the enzymatic reaction was developed by adding 3,3 ′ ,5,5 ′ -tetramethylbenzidine (TMB) for 15 min and then stopped with 0.2 M H 2 SO 4 (Thermo Scientific, Waltham, MA, USA). Optical density was measured at 450 nm (OD450). A threshold for positivity was established as the mean OD of negative control sera plus four standard deviations (mean + 4SD). These operating conditions were determined by evaluating various antigen concentrations and dilutions of positive and negative control sera (crisscross serial dilution study with an initial panel of negative and positive serum samples) [14]. A panel of wellcharacterized sera belonging to apparently healthy ruminants and/or ruminants with a history of abortion was selected from a previous work [24] to investigate the diagnostic potential of recombinant antigens against a commercial ELISA (Q fever Ab test kit; IDEXX, Westbrook, ME, USA). The Ethics Committee of the Department of Veterinary Medicine and Animal Production (Centro Servizi Veterinari, Turin, Italy), University of Naples, Federico II, authorized the animal study protocol (PG/2022/0093419) on 20 July 2022. A total of 120 positive and 281 negative sera were selected from samples stored for other similar studies by choosing those best preserved and not thawed more than three times. The commercial ELISA was performed following the manufacturer's instructions. Briefly, serum was diluted 1:100 to each well, incubated for 1 h at 37 • C, and, after three washes, 100 µL of anti-ruminant IgG conjugate was added. After further incubation and washing, the reaction was read by adding TMB for 15 min and stopping solution.
## 2.4. Statistical Analysis
The commercial ELISA (Q fever Ab test kit; IDEXX) was considered as the reference assay in our statistical analysis (MedCalc v.18.11.3, https://www.medcalc.org/; accessed on 1 December 2023). The level of agreement among assays was determined by the kappa Cohen coefficient (0.81-1.00 = almost perfect agreement; 0.61-0.80 = substantial agreement; 0.41 0.60 = moderate agreement; 0.21-0.40 = fair agreement; 0.01-0.20 = slight agreement; 0.00 = no agreement).
## 3. Results
All three recombinant antigens were successfully expressed and purified in sufficient quantities for use in ELISAs. Protein concentrations, as determined by the Bradford assay, were 1.2 mg/mL for Com-1, 0.4 mg/mL for MceB, and 0.55 mg/mL for AdaA. Purification of the target proteins for ELISA preparation was successful, as evidenced by SDS-PAGE electrophoresis (Figure 1a-c), which revealed a certain level of purity despite the presence of some additional protein residues in the final eluates. For Com-1, which was expressed as a GST fusion protein, proteolytic cleavage was performed directly on-column, allowing efficient removal of the fusion tag (Figure 1c). Among the three candidates, Com-1 showed the most promising diagnostic performance, demonstrating clear reactivity with both positive and negative serum samples, and was selected for further evaluation (and tested with a larger number of serum samples). In contrast, the concordance for AdaA was poor (overall agreement of 0.54) as the antigen only reacted with 17/55 positive sera (Tables 1 and2). Similarly, MceB demonstrated a low concordance (0.56), largely due to a high number of false positives (37/60) (Tables 1 and2). In both cases, Cohen's kappa values ranged between 0 and 0.2, revealing poor agreement with the commercial test. Com-1 achieved an overall agreement of 0.83 with a Cohen's kappa value of 0.58 which was considered a moderate agreement. Furthermore, the Com-1-based ELISA appeared to be more specific than sensitive, as fewer false positives (28/281) than false negatives (40/120) were observed. In well 1, a band can be observed corresponding to approximately 50 kDa. The intensity of this band is reduced in well 2 due to protein adsorption to the resin. In wells 4, 5, and 6, the protein purified under denaturing conditions is observed in three different eluates (indicated in red). (c) SDS-PAGE analysis of the fraction obtained (Com-1): L = Ladder, 1 = total extract, 2 = total extract after resin adsorption, 3, 4, and 5 = purified and cleaved protein eluates. Well 1 shows overexpression of a Com1 band around 50 kDa, corresponding to the fusion protein. Protein adsorption to resin reduces the intensity of this band in wells 2. Wells 3, 4, and 5 showed pure and cleaved proteins (indicated in red). After removing GST, the protein returns to its original molecular weight of 25 kDa. This figure has been created with the support of Biorender (Biorender.com, accessed on 1 May 2025).
## 4. Discussion
Three immunodominant antigens of C. burnetii were produced in recombinant form in this work and were evaluated for their ability to detect specific antibodies in bovine sera. Among the candidates, Com-1 was the only antigen that yielded promising results, correctly identifying 80 out of 120 positive animals and 253 out of 281 negative animals. This antigen reached a moderate agreement (0.83) when it was compared to a commercial ELISA kit. For this reason, it was evaluated with a greater number of serum samples.
These results are consistent with a previous study in which Com-1 was produced in recombinant form, albeit with different production and purification methods. Reported diagnostic sensitivity and specificity ranged between 65 and 85% depending on the species, with better diagnostic performance in goats than in cattle [38]. A further study described the increase in diagnostic performance for the Com-1 antigen (around 83% sensitivity and 80% specificity) when it was produced synthetically and utilized in the latex agglutination test (LAT) [39].
Studies focusing on other antigens have found similar results, regardless of the test format or ruminant species. Although reported to be immunodominant in both humans and animals, antigens such as cell division coordinator CpoB (Ybgf) or dihydrolipoyllysineresidue succinyltransferase (SucB) have achieved high specificities (around 90%) but lower sensitivity (70-80%) in various ELISA formats [24,27,40]. The reported concordance values were 0.83 for SucB and 0.86 and 0.88 for Ybgf when used in double-antigen ELISA or indirect ELISA, respectively [24,41]. Currently, the most effective recombinant antigen described was heat shock protein B (HspB) with sensitivity around 80-90% and high specificity (97%) [28]. However, this antigen has only been tested in experimentally infected goats [28].
All these findings should be interpreted in the context of the complex immune response elicited by Coxiella infection. As in humans, ruminants typically generate antibodies to phase II antigens during acute infection and seroconvert to phase I antigens as the infection becomes chronic [42]. For this reason, commercial kits consist of antigen mixtures from both phases to maximize detection, although this feature can reduce specificity due to potential cross-reactivity with other pathogens. Conversely, single recombinant antigens, while potentially more specific, may lack sensitivity if they only represent one phase or stage of the immune response [27]. However, it is noteworthy that the performance of the commercial test utilized as a reference in this study was not without limitations. In the absence of a true gold standard or validated reference sera, the reliability of commercial tests has been questioned, in particular considering the high rates of false-positive and falsenegative results reported in several studies [43][44][45]. This is a common issue in the evaluation of recombinant antigens for the detection of anti-Coxiella antibodies in veterinary medicine. The WOAH manual recommends ELISA as the reference test for serology, despite its limitations associated with false positives caused by cross-reactions with organisms that share Coxiella's antigenic profile (such as Bartonella, Rickettsiae, etc.) [14].
Another explanation for the low concordance of recombinant antigens could be associated with the inability of Escherichia coli (used as the expression system) to perform post-translational modifications, such as glycosylation or methylation, that can be important for the correct epitope conformation and immune recognition [24].
Furthermore, the genetic diversity of C. burnetii strains and their association with different ruminant species and the enzootic areas may affect reactivity. For instance, more virulent strains possess the AdaA gene, while others lack it. Additionally, the AdaA gene has been identified predominantly in sheep, which could explain its low diagnostic performance in cattle sera [30].
While none of the tested antigens are yet suitable for immediate use in routine screening, Com-1 stands out as a promising candidate for further research and development. Further studies should evaluate this antigen using a larger number of well-characterized sera, including samples from experimentally infected/vaccinated animals or reference sera.
Com1 is an outer membrane protein that displays reactivity for both phases, although its specificity may be compromised by cross-reactions with some rickettsial species [34,46]. It is described in the literature as being more reactive in chronic forms, and to date, more than 15 articles have highlighted its immunoreactive properties, identifying it as a major immunodominant antigen in humans and a strong candidate for future applications in serology and vaccinology [33,36]. One human investigation showed Com-1 reacted with sera from individuals with chronic Q fever, agreeing with 92.4% (122/132) of negative clinical diagnoses and 72.2% (26/36) of positive clinical diagnoses [47]. However, veterinary research on Com-1 remains limited, and its full potential in animal diagnostics has yet to be fully explored. Improving ruminant screening tests is a veterinary priority for Q fever prevention. It is a significant concern that, despite advances in human serology (with IFA and chemiluminescent assays), which allows for fine-tuned diagnosis, differentiation of vaccinated from infected patients, determination of the infection stage, and assessment of treatment efficacy, veterinary serology still lacks reliable and standardized tools [22]. More attempts are clearly necessary, both to develop new assays and to enhance existing ones.
## 5. Conclusions
In this work, the potential of three recombinant antigens to detect specific anti-Coxiella antibodies in ELISA was tested using cattle serum samples. Com-1 showed the best agreement with a commercial ELISA kit and demonstrated potential as a candidate for further investigation. Although the data do not support the use of this Com-1-based ELISA as a screening test in ruminants at this time, they justify additional research into its applicability for serodiagnosis in veterinary medicine.
## References
1. Angelakis, Raoult, Fever (2010) *Vet. Microbiol*
2. Guatteo, Seegers, Taurel et al. (2011) "Prevalence of Coxiella burnetii Infection in Domestic Ruminants: A Critical Review" *Vet. Microbiol*
3. Bauer, Knittler, Andrack et al. (2023) "Interdisciplinary Studies on Coxiella burnetii: From Molecular to Cellular, to Host, to One Health Research" *Int. J. Med. Microbiol*
4. Gisbert, Garcia-Ispierto, Quintela et al. (1313) "Coxiella burnetii and Reproductive Disorders in Cattle: A Systematic Review" *Animals*
5. Ferrara, Longobardi, Pagnini et al. (2024) "Evaluation of the Phase-Specific Antibody Response in Water Buffalo (Bubalus bubalis) after Two Doses of an Inactivated Phase I Coxiella burnetii Vaccine" *Vet. Immunol. Immunopathol*
6. Zheng, Zadoks, Kim et al. (2026) "A Scoping Review of Coxiella burnetii Transmission Models in Ruminants" *Prev. Vet. Med*
7. Pexara, Solomakos, Govaris (2018) "Fever and Seroprevalence of Coxiella burnetii in Domestic Ruminants" *Vet. Ital*
8. Ferrara, Pagnini, Improda et al. (2024) "Pigs in Southern Italy Are Exposed to Three Ruminant Pathogens: An Analysis of Seroprevalence and Risk Factors Analysis Study" *BMC Vet. Res*
9. Ferrara, Flores-Ramirez, Palkovicova et al. (2024) "Serological and Molecular Survey of Q Fever in the Dog Population of the Campania Region, Southern Italy" *Acta Trop*
10. Epelboin, De Souza Ribeiro Mioni, Couesnon et al. (2023) "Coxiella burnetii Infection in Livestock, Pets, Wildlife, and Ticks in Latin America and the Caribbean: A Comprehensive Review of the Literature" *Curr. Trop. Med. Rep*
11. Ullah, Jamil, Saqib et al. (1530) *Q Fever-A Neglected Zoonosis. Microorganisms*
12. Eldin, Mélenotte, Mediannikov et al. (2017) "From Q Fever to Coxiella burnetii Infection: A Paradigm Change" *Clin. Microbiol. Rev*
13. Toledo-Perona, Contreras, Gomis et al. "Controlling Coxiella burnetii in Naturally Infected Sheep, Goats and Cows, and Public Health Implications: A Scoping Review" *Front. Vet. Sci*
14. Gisbert, Hurtado, Guatteo (1484) "Efficacy and Safety of an Inactivated Phase I Coxiella burnetii Vaccine to Control Q Fever in Ruminants: A Systematic Review" *Animals*
15. Woah (2024) "Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, 13th edition. World Organisation for Animal Health (WOAH)"
16. García-Pérez, Zendoia, Ferrer et al. (2025) "Combination of Serology and PCR Analysis of Environmental Samples to Assess Coxiella burnetii Infection Status in Small Ruminant Farms" *Appl. Environ. Microbiol*
17. Bauer, Hubbard, Johnson et al. (2016) "A Cross Sectional Study Evaluating the Prevalence of Coxiella burnetii, Potential Risk Factors for Infection, and Agreement between Diagnostic Methods in Goats in Indiana" *Prev. Vet. Med*
18. Szyma Ńska-Czerwi Ńska, Niemczuk, Jodełko (2016) "Evaluation of QPCR and Phase I and II Antibodies for Detection of Coxiella burnetii Infection in Cattle" *Res. Vet. Sci*
19. Walraph, Zoche-Golob, Weber et al. (2018) "Decline of Antibody Response in Indirect ELISA Tests during the Periparturient Period Caused Diagnostic Gaps in Coxiella burnetii and BVDV Serology in Pluriparous Cows within a Holstein Dairy Herd" *Res. Vet. Sci*
20. Horigan, Bell, Pollard et al. (2011) "Q Fever Diagnosis in Domestic Ruminants: Comparison between Complement Fixation and Commercial Enzyme-Linked Immunosorbent Assays" *J. Vet. Diagn. Investig*
21. Wood, Muleme, Tan et al. (2019) "Validation of an Indirect Immunofluorescence Assay (IFA)for the Detection of IgG Antibodies against Coxiella burnetii in Bovine Serum" *Prev. Vet. Med*
22. Sahu, Rawool, Vinod et al. (2020) "Current Approaches for the Detection of Coxiella burnetii Infection in Humans and Animals" *J. Microbiol. Methods*
23. Emery, Ostlund, Schmitt (2012) "Comparison of Q Fever Serology Methods in Cattle, Goats, and Sheep" *J. Vet. Diagn. Investig*
24. Ferrara, Colitti, Pagnini et al. (2022) "Characterization of Recombinant Ybgf Protein for the Detection of Coxiella Antibodies in Ruminants" *J. Vet. Diagn. Investig*
25. Paul, Toft, Agerholm et al. (2013) "Bayesian Estimation of Sensitivity and Specificity of Coxiella burnetii Antibody ELISA Tests in Bovine Blood and Milk" *Prev. Vet. Med*
26. Hogerwerf, Koop, Klinkenberg et al. (2014) "Test and Cull of High Risk Coxiella burnetii Infected Pregnant Dairy Goats Is Not Feasible Due to Poor Test Performance" *Vet. J*
27. Ferrara, Colitti, Flores-Ramirez et al. (2023) "Detection of Coxiella Antibodies in Ruminants Using a SucB Recombinant Antigen" *J. Vet. Diagn. Investig*
28. Fernandes, Rousset, Dufour et al. (2009) "Evaluation of the Recombinant Heat Shock Protein B (HspB) of Coxiella burnetii as a Potential Antigen for Immunodiagnostic of Q Fever in Goats" *Vet. Microbiol*
29. Miller, Kersh (2020) "Analysis of Recombinant Proteins for Q Fever Diagnostics" *Sci. Rep*
30. Sam, Chen, Plain et al. (2024) "Diphtheria Toxoid Particles as Q Fever Vaccine"
31. Bach, Fitzgerald, Williams-Macdonald et al. "Genome-Wide Epitope Mapping across Multiple Host Species Reveals Significant Diversity in Antibody Responses to Coxiella burnetii Vaccination and Infection"
32. Xiong, Wang, Wen et al. (2012) "Potential Serodiagnostic Markers for Q Fever Identified in Coxiella burnetii by Immunoproteomic and Protein Microarray Approaches" *BMC Microbiol*
33. Gerlach, Škultéty, Henning et al. (2017) "Coxiella burnetii Immunogenic Proteins as a Basis for New Q Fever Diagnostic and Vaccine Development" *Acta Virol*
34. Beare, Chen, Bouman et al. (2008) *Candidate Antigens for Q Fever Serodiagnosis Revealed by Immunoscreening of a Coxiella burnetii Protein Microarray. Clin. Vaccine Immunol*
35. Jiao, Xiong, Qi et al. (2014) "Serological Characterization of Surface-Exposed Proteins of Coxiella burnetii" *Microbiology*
36. Vigil, Ortega, Nakajima-Sasaki et al. (2010) "Genome-Wide Profiling of Humoral Immune Response to Coxiella burnetii Infection by Protein Microarray" *Proteomics*
37. Palkovicová, Flores-Ramírez, Quevedo-Diaz et al. "Innovative Antigens for More Accurate Diagnosis of Q Fever" *J. Microbiol. Methods*
38. Stellfeld, Gerlach, Richter et al. "Evaluation of the Diagnostic Potential of Recombinant Coxiella burnetii Com1 in an Elisa for the Diagnosis of q Fever in Sheep, Goats and Cattle"
39. Yadav, Malik, Dhaka et al. (2020) "Comparison of Two New In-House Latex Agglutination Tests (LATs), Based on the DnaK and Com1 Synthetic Peptides of Coxiella burnetii, with a Commercial Indirect-ELISA, for Sero-Screening of Coxiellosis in Bovines" *J. Microbiol. Methods*
40. Zhang, Kiss, Seshadri et al. (2004) "Identification and Cloning of Immunodominant Antigens of Coxiella burnetii" *Infect. Immun*
41. Ferrara, Colitti, Gabriela et al. (2024) "Efficiency of Recombinant Ybgf in a Double Antigen-ELISA for the Detection of Coxiella Antibodies in Ruminants" *Vet. Anim. Sci*
42. Bauer, Schwecht, Jahnke et al. (2023) "Humoral and Cellular Immune Responses in Sheep Following Administration of Different Doses of an Inactivated Phase I Vaccine against Coxiella burnetii" *Vaccine*
43. Lucchese, Capello, Barberio et al. (2015) "IFAT and ELISA Phase I/Phase II as Tools for the Identification of Q Fever Chronic Milk Shedders in Cattle" *Vet. Microbiol*
44. Lurier, Rousset, Gasqui et al. (2021) "Evaluation Using Latent Class Models of the Diagnostic Performances of Three ELISA Tests Commercialized for the Serological Diagnosis of Coxiella burnetii Infection in Domestic Ruminants" *Vet. Res*
45. Rivière, Rousset, Jourdain et al. (2025) "Harmonisation of the Diagnostic Performances of Serological ELISA Tests for C. Burnetii in Ruminants in the Absence of a Gold Standard: Optimal Cut-Offs and Performances Reassessment" *Prev. Vet. Med*
46. Deringer, Chen, Samuel et al. (2011) "Immunoreactive Coxiella burnetii Nine Mile Proteins Separated by 2D Electrophoresis and Identified by Tandem Mass Spectrometry" *Microbiology*
47. Vranakis, Mathioudaki, Kokkini et al. (2019) "Com1 as a Promising Protein for the Differential Diagnosis of the Two Forms of q Fever" *Pathogens*
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=PMC12724330&blobtype=pdf | # The crosstalk between ubiquitination and GlcNAcylation of CHAF1A regulates HIV-1 latency and reactivation
Tao Yang, Minghua Chen, Mo Zhou, Xiaohui Deng, Peiming Huang, Siyi Xie, Jianteng Zeng, Jingjing Luo, Yiwen Zhang, Xiancai Ma, Liqin Sun, Jiaye Liu, Hui Zhang, Linghua Li, Bingfeng Liu, Jie Qin, Ting Pan
## Abstract
The persistence of latent HIV-1 reservoirs remains a critical barrier to cure. Current "shock and kill" strategies are limited by ineffective latency-reversing agents (LRAs) and poor understanding of epigenetic regulation. Here, we identify chromatin assembly factor 1 subunit A (CHAF1A), a histone chaperone enforcing HIV-1 latency, as a therapeutic target regulated by antagonistic post-translational modifications: ubiquitina tion promotes its degradation, while O-GlcNAcylation stabilizes it. We demonstrate that trifluridine, a Food and Drug Administration-approved antiviral drug, reactivates latent HIV-1 by disrupting O-GlcNAcylation, triggering CHAF1A ubiquitination and proteaso mal degradation. Notably, CHAF1A expression increases with age in CD4 + T cells (>60 years), correlating with deeper proviral reservoirs. This age-dependent accumulation inversely associates with reduced O-GlcNAcase levels, suggesting O-GlcNAcylation-medi ated stabilization in aging. Our findings establish CHAF1A as both a therapeutic target and an age-stratifying biomarker, advancing trifluridine as a translatable LRA to enhance reservoir clearance in aging populations-a demographic increasingly impacted by HIV-1 persistence. IMPORTANCE HIV-1 latency continues to represent a significant barrier to achieving a cure, particularly in aging populations characterized by expanded viral reservoirs and compromised immune recovery-a challenge further intensified by the absence of therapies specifically designed to target age-related mechanisms. Current latencyreversing agents (LRAs) are insufficient in addressing the metabolic and epigenetic dysregulation that sustains viral persistence in older individuals. In this study, we reveal a dynamic interplay between ubiquitination and O-GlcNAcylation that regulates the stability of CHAF1A, a histone chaperone essential for maintaining HIV-1 latency. We identify trifluridine as a novel LRA capable of disrupting O-GlcNAcylation to degrade CHAF1A, thereby effectively reversing latency in primary cells. This research bridges a critical gap between fundamental virology and clinical gerontology. These findings establish a robust foundation for refining strategies aimed at HIV-1 eradication, with a focus on targeting host metabolic-epigenetic networks to address latency in under served aging populations.
will be 50 years or older, with the median age of this population expected to increase progressively over time (4). This demographic shift introduces unique challenges, as older individuals exhibit larger or deeper latent reservoirs, delayed viral rebound post-ART interruption, and diminished immune responses-factors that exacerbate viral persistence and complicate cure strategies (5,6). This demographic shift underscores the urgent need to identify age-specific biomarkers and therapeutic targets to tailor interventions for aging populations, who remain disproportionately underserved in HIV-1 cure research.
Chromatin assembly factor 1 subunit A (CHAF1A), a histone chaperone critical for heterochromatin formation, has emerged as a key suppressor of HIV-1 transcription (7)(8)(9). Our prior work revealed that the CAF-1 complex enforces viral latency through liquid-liquid phase separation (LLPS), forming nuclear condensates at the HIV-1 promoter that recruit repressive epigenetic modifiers (e.g., HDACs and H3K9me3 writers) to stabilize heterochromatin (10). Building on this, we now investigate the post-trans lational regulation of CHAF1A, focusing on the antagonistic interplay between two modifications: ubiquitination, which signals proteasomal degradation, and O-GlcNAcyla tion, a nutrient-sensitive modification that stabilizes chromatin-associated proteins (11,12). While CHAF1A maintains latency by depositing repressive histone H3 variants, its stability is dynamically regulated by competing post-translational modifications (PTMs). Notably, ubiquitination and O-GlcNAcylation are known to compete for substrate occupancy in other systems (12)(13)(14)(15), but their crosstalk in HIV-1 latency remains unexplored. We hypothesize that this PTM axis acts as a molecular switch governing CHAF1A stability, thereby linking cellular metabolism and aging to viral persistence.
The role of PTMs as biomarkers has revolutionized fields such as oncology, where phosphorylation and ubiquitination patterns guide diagnosis, prognosis, and thera peutic targeting (16)(17)(18)(19). However, their potential in HIV-1 research-particularly in addressing age-related latency dynamics-remains underexplored. In this study, we investigate how O-GlcNAcylation modulates CHAF1A stability, thereby influencing HIV-1 latency. Using primary CD4 + T cell models, we show that trifluridine (TFD), a clini cally approved antiviral (20)(21)(22)(23), disrupts O-GlcNAc modification, promoting CHAF1A degradation and latent HIV-1 reactivation. Critically, our findings bridge molecular mechanisms to clinical aging challenges: CHAF1A expression increases with age in older individuals (>60 years). Mechanistically, age-dependent declines in O-GlcNAcase (OGA) enhance CHAF1A stabilization via O-GlcNAcylation, linking cellular metabolism and aging to viral persistence. This age-metabolism-latency axis highlights CHAF1A as both a PTM-regulated therapeutic target and a biomarker for stratifying patients by reservoir dynamics, offering a precision framework to address aging-related HIV-1 persistence.
## RESULTS
## Multi-omics analysis reveals ubiquitination and glycosylation jointly participate in CAF-1-related HIV-1 latency
Building on prior work demonstrating CHAF1A's role in promoting latency via LLPS at proviral promoters (10), we performed proteomic approaches to dissect its regulatory mechanisms. Using the J-Lat 10.6 latency model, we observed that TNF-α, a canoni cal LRA, significantly reduced CHAF1A expression levels and its enrichment in J-Lat 10.6 cells, suggesting the presence of a potential regulatory mechanism during viral reactivation (Fig. 1A). To further explore this observation, we performed gel electropho resis and silver staining on immunoprecipitation samples following TNF-α activation. The results revealed that, compared with the IgG antibody control group, the CHAF1A antibody specifically enriched a number of differentially expressed proteins during the J-Lat 10.6 reactivation (Fig. 1B). Proteomic profiling analysis of endogenous CHAF1A complexes identified several key interaction partners, including its heterodimeric partner CHAF1B and O-GlcNAc transferase (OGT), with significant enrichment of pathways associated with histone binding and epigenetic regulation of gene expression (Fig. 1C). Furthermore, the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of the mass spectrometry data demonstrated significant enrichment of proteasomal-medi ated degradation and ribosomal-related processes, indicating that both protein turnover and translational regulation may play critical and complementary roles in maintaining HIV latency (Fig. 1D). Meanwhile, we conducted a systematic analysis of our previous transcriptome data of J-Lat 10.6 cells before and after TNF-α treatment. The results of gene set enrichment analysis (GSEA) indicated that latent cells were in a state of metabolic quiescence, with significant downregulation of fructose/mannose metabolism (NES = -2.11, P-adj = 0.002) and glycolysis (NES = -1.72, P-adj = 0.009) pathways, which was highly consistent with the nutrient-restricted state of quiescent cells (Fig. 1E andF). Additionally, this result was in line with the trend of CHAF1A expression level changes, suggesting that CHAF1A might play a stabilizing role in maintaining HIV latency by integrating epigenetic repression with metabolic dormancy. More importantly, integration of mass spectrom etry analysis with transcriptomic data revealed the co-enrichment of OGT and protea some-related pathways, suggesting that CHAF1A may serve as a central node at the intersection of two opposing PTM pathways: O-GlcNAc glycosylation, which stabilizes chromatin-associated proteins, and ubiquitination, which targets these proteins for degradation. Our multi-omics investigation of CHAF1A-mediated HIV-1 latency revealed critical insights into the crosstalk between PTMs that enforce HIV-1 latency, provid ing a mechanistic foundation for targeting the interplay between ubiquitination and glycosylation to disrupt viral persistence.
## OGT/OGA regulates CHAF1A stability via antagonistic ubiquitination-O-GlcNAcylation crosstalk
The interplay between ubiquitination and O-GlcNAcylation can dynamically regulate protein stability (24,25). Here, we investigated this crosstalk in the context of CHAF1A, focusing on the opposing roles of OGT and OGA. In the glycolytic pathway, ~2%-5% of glucose flux is diverted through the hexosamine biosynthesis pathway (HBP) to generate UDP-N-acetylglucosamine (UDP-GlcNAc), the obligate donor substrate for O-GlcNAcyla tion(26) (Fig. 2A). However, unlike other modifications, the addition and removal of this modification is performed with a single set of enzymes (27,28). O-GlcNAc glycosylation is catalyzed by OGT, while de-glycosylation is catalyzed by OGA (Fig. 2B). To investigate the impact of O-GlcNAcylation on CHAF1A stability, we genetically perturbed the O-GlcNAc cycling enzymes. Specifically, we used siRNA to knock down OGT and OGA in 293T cells and single-guide RNA (sgRNA) in J-Lat 10.6 cells. Our experiments revealed that OGT knockdown consistently decreased overall cellular O-GlcNAcylation and consequently reduced CHAF1A expression. Conversely, OGA knockdown led to increased O-GlcNAcyla tion and elevated CHAF1A levels (Fig. 2C; Fig. S1). These results strongly suggest that the OGT/OGA enzymatic pair dynamically regulates CHAF1A protein levels via O-GlcNAc modification.
To further validate the direct association between CHAF1A and these key enzymes, we performed co-immunoprecipitation (Co-IP) assays. Endogenous CHAF1A was found to interact directly with both OGT and OGA (Fig. 2D). Reciprocal Co-IP experiments further confirmed that endogenous OGT specifically complexed with CHAF1A (Fig. 2E). These findings firmly establish that CHAF1A forms protein complexes with both OGT and OGA, thereby positioning these enzymes as critical mediators of CHAF1A stability through O-GlcNAc modification. Conversely, pharmacological inhibition of OGA with PUGNAc enhanced overall O-GlcNAcylation and led to a noticeable increase in CHAF1A protein levels (Fig. 2G). This pharmacological evidence further corroborates that O-GlcNAcylation stabilizes CHAF1A, likely by competitively interfering with its ubiquitina tion.
Collectively, these findings demonstrate that OGT/OGA-mediated O-GlcNAcylation antagonistically regulates CHAF1A stability by counteracting ubiquitination. This antagonistic PTM crosstalk provides a direct molecular link between the cellular meta bolic state and the maintenance of HIV-1 latency.
## Targeted screening identifies TFD as a CHAF1A degrader via ubiquitinationglycosylation crosstalk
To identify compounds that specifically degrade CHAF1A, we screened the US DRUG COLLECTION library using a green fluorescent protein (GFP)-CHAF1A fusion protein as an indicator. We constructed a pcDNA3.1-CFP plasmid as an internal reference to transfection control (Fig. 3A). After three rounds of screening, TFD was identified as a candidate drug that significantly degraded CHAF1A (Fig. S2; Fig. 3B). Further verification through GFP fluorescence experiments and western blot experiments confirmed that TFD reduced both exogenous GFP-CHAF1A and endogenous CHAF1A in a dose-depend ent manner (Fig. 3C andD). Immunofluorescence further demonstrated diminished nuclear CHAF1A expression upon TFD treatment (Fig. 3E; Fig. S3), consistent with proteasomal degradation. Cytotoxicity profiling revealed that TFD maintained >90% cell viability in primary human peripheral blood mononuclear cells (PBMCs) and cell lines (HEK293T and TZM-bl) at concentrations up to 200 µM, which is sufficient for CHAF1A degradation (Fig. 3F; Fig. S4A andB). However, we observed higher sensitivity in certain lymphocyte cell lines (Jurkat and J-Lat 10.6) at the highest doses up to 10 µM (Fig. S4C andD). Notably, while TFD is a nucleoside analog that can affect both host and viral DNA synthesis, its safety and tolerability have been established in clinical settings when administered appropriately in combination therapies for cancer, such as TFD/tipiracil plus bevacizumab for refractory metastatic colorectal cancer (29) and TFD/tipiracil in advanced gastric or colorectal cancers (30,31). These clinical data indicate that, when properly dosed and formulated, TFD has the potential for systemic administration.
To investigate the underlying mechanism, we initially treated human T lymphocyte cell line Jurkat T cells with TFD. The results demonstrated that TFD significantly reduced global cellular glycosylation levels (Fig. 3G). This effect was similarly observed in human primary CD4 + T cells (Fig. S5). Further endogenous co-IP experiments using a CHAF1Aspecific antibody revealed that TFD treatment markedly increased CHAF1A ubiquitina tion, suggesting the involvement of the ubiquitin-proteasome system in its degradation (Fig. 3H). Moreover, sWAG analysis confirmed that CHAF1A glycosylation was substan tially diminished under TFD treatment conditions (Fig. 3I), indicating that TFD facilitated CHAF1A ubiquitination and subsequent degradation by disrupting O-GlcNAc-mediated stabilization.
These results position TFD as a dual-action compound that destabilizes CHAF1A by simultaneously suppressing O-GlcNAcylation and promoting ubiquitination. This unique mechanism highlights its potential as a precision LRA tailored to aging-associated HIV-1 reservoirs.
## The E3 ubiquitin ligase MIB1 mediates TFD-induced CHAF1A ubiquitination and degradation
E3 ubiquitin ligases are essential for substrate recognition and ubiquitin transfer in protein degradation (32). To define the role of E3 ligase in TFD-mediated CHAF1A ubiquitination, we initially used the UbiBrowser database (UbiBrowser [bio-it.cn]) and obtained several predicted proteins (Fig. 4A). Among the top 5 candidates, siRNA screening revealed that MIB1 (a RING-domain E3 ligase) and NEDD4 (a HECT-domain E3 ligase) partially reversed TFD-induced CHAF1A degradation (Fig. 4B) (33,34). While NEDD4 knockdown modestly attenuated CHAF1A protein levels, MIB1 silencing nearly abolished it, suggesting MIB1 plays a dominant role. CHAF1A immunoprecipitated from J-Lat 10.6 and HEK293T cells in the presence/absence of OGA inhibitor PUGNAc. The cells were treated with/without MG132 for proteasome inhibition. Cells were treated with 50 µM PUGNAC (an inhibitor of OGA) for 48 h, followed by an additional 6-h treatment with 20 µM MG132 prior to harvest. (G) HEK293T cells transfected with HA-tagged CHAF1A or GFP, and then treated with PUGNAc or not. The O-GlcNAcylated CHAF1A ubiquitination was analyzed by Co-IP with anti-HA beads followed by western blot with CTD110.6. GAPDH was shown as a loading control. To validate these candidates, we overexpressed MIB1 or NEDD4 in HEK293T cells. Both ligases promoted CHAF1A degradation in a dose-dependent manner (Fig. 4C). However, Co-IP assays demonstrated a direct interaction between CHAF1A and MIB1, but not NEDD4 (Fig. 4D), indicating that NEDD4's effect may occur indirectly through proteaso mal regulation. To definitively establish the requirement of MIB1 in TFD-induced CHAF1A degradation, we generated MIB1 knockout (sgMIB1) J-Lat 10.6 cells using CRISPR/Cas9 and subsequently assessed CHAF1A protein levels following TFD treatment. In control cells (sgRNA as control [sgNC]), TFD induced a clear dose-dependent reduction of CHAF1A. Strikingly, MIB1 knockout largely abolished this effect, with CHAF1A protein levels remaining stable even at higher TFD concentrations (Fig. 4E). These results conclusively confirm that MIB1 is essential for TFD-triggered CHAF1A degradation in J-Lat 10.6 cells.
By integrating loss-of-function, gain-of-function, and direct interaction analyses, we conclusively identify MIB1 as the E3 ubiquitin ligase responsible for TFD-driven CHAF1A ubiquitination and subsequent degradation. This discovery positions MIB1 as a critical node in the PTM crosstalk regulating HIV-1 latency and highlights its therapeutic potential for targeted reservoir clearance.
## Inhibiting OGT or overexpressing OGA reactivates latent HIV-1 by destabiliz ing CHAF1A
To determine whether O-GlcNAcylation directly modulates HIV-1 latency, we treated J-Lat 10.6 cells with OSMI-4, a selective OGT inhibitor (35). OSMI-4 significantly reduced global O-GlcNAcylation and reactivated latent HIV-1 (Fig. 5A), thereby demonstrating that suppressing glycosylation disrupts viral quiescence. Furthermore, pre-treatment with the OGA inhibitor PUGNAc markedly reduced OSMI-4-induced HIV-1 reactivation, suggesting that O-GlcNAcylation directly regulates HIV-1 latency (Fig. 5B). Cell viability assays using CellTiter-Glo confirmed that OSMI-4 did not induce significant cytotoxicity at concentrations up to 200 µM (Fig. 5C), ruling out non-specific toxic effects. Similarly, genetic deletion of OGT, the enzyme responsible for O-GlcNAc transfer, also enhanced viral reactivation (Fig. 5D), providing further genetic evidence that reduced O-GlcNAcyla tion destabilizes latency. Importantly, previous work has demonstrated that knockout of CHAF1A effectively reactivates latent HIV-1 in J-Lat 10.6 cells, highlighting the critical role of CHAF1A in maintaining latency (10).
We next validated the latency-reversing activity of TFD. TFD reactivated latent HIV-1 in J-Lat 10.6 cells in a concentration-dependent manner (Fig. 5E), paralleling its ability to degrade endogenous CHAF1A (Fig. 5F). Meanwhile, western blot analysis following OSMI treatment revealed a dose-dependent decrease in global O-GlcNAc levels (Fig. 5G), whereas TFD treatment resulted in a dose-dependent increase in global O-GlcNAc levels (Fig. 5H). This dual effect-suppressing O-GlcNAcylation while promoting CHAF1A ubiquitination-positions TFD as a mechanistically unique LRA. Given CHAF1A's initial identification in TNF-α-stimulated Jurkat T cells, we conducted TNF-α stimulation and receptor blockade experiments to validate our proposed mecha nism and link it to prior findings. In J-Lat 10.6 cells, combined TFD and TNF-α treatment synergistically enhanced HIV-1 activation. Importantly, while the TNF-α receptor blocker Atrosab significantly inhibited TNF-α-mediated activation, it had no appreciable effect on TFD-induced latency reversal (Fig. S6A andB). Furthermore, RNA-seq analysis of TFDtreated cells revealed no significant changes in the expression of key HIV latency regulators such as BRD4 and NF-κB (Fig. S6C), indicating that TFD acts independently of these established pathways. In parallel, CHAF1A rescue experiments in TZM-BL cells showed that CHAF1A overexpression completely reversed trifluorothymidine-induced viral reactivation (Fig. S6D), thereby confirming its role as a critical downstream effector. Collectively, the strong correlation among O-GlcNAc modification, CHAF1A dynamics, and HIV reactivation, together with the rescue data, supports the existence of a domi nant, specific signaling pathway distinct from non-specific toxic effects. These findings establish that targeting the O-GlcNAcylation-ubiquitination axis of CHAF1A effectively disrupts HIV-1 latency.
## TFD reactivates latent HIV-1 in primary cells, and CHAF1A expression correlates with aging
To validate TFD as a clinically relevant LRA, we tested its efficacy in primary CD4 + T cells isolated from three HIV-1-infected individuals on suppressive combination ART (cART) for >6 months (Fig. 6A). TFD significantly upregulated intracellular HIV-1 RNA, achieving reactivation levels comparable to the potent αCD3/αCD28 stimulation (Fig. 6B). This demonstrates TFD's ability to target latent reservoirs in patient-derived cells, a critical milestone for translational relevance. To further assess the potential of TFD as a LRA, we tested its ability to reactivate latent HIV-1 in ex vivo primary cells from ART-treated individuals using the TZM-bl-based qVOA assay (36,37). PBMCs and rectal cells were treated with TFD or anti-CD3/CD28 antibodies, and viral outgrowth was measured by luciferase activity (Fig. 6C). TFD significantly reactivated HIV-1 in both blood and rectal samples (Fig. 6D). This finding strongly supports the feasibility of exploring TFD as a localized therapeutic strategy for targeting tissue-resident reservoirs, leveraging its existing safety profile and approved route of administration for a related viral indication.
We further investigated the age-dependent expression of CHAF1A and its regulatory enzyme, OGA (MGEA5), using a public data set (GSE178670) from both healthy and HIVpositive individuals (6). In healthy controls, CHAF1A expression was significantly higher in older individuals (over 60 years) than in younger counterparts (P = 0.0036; Fig. 6E). Conversely, OGA levels significantly declined with age (P = 1.4 × 10⁻⁵; Fig. 6F). These changes showed a strong inverse correlation (r = -0.78, P = 6.8 × 10⁻⁶; Fig. 6G), suggest ing that age-related OGA loss drives CHAF1A accumulation. In HIV-positive individuals, OGA levels also declined sharply in older donors (P = 0.00019; Fig. 6H), and the CHAF1A-OGA correlation remained negative (r = -0.65, P = 0.0011; Fig. 6I). While the trend for CHAF1A to increase with age was not statistically significant in this cohort (P = 0.44; Fig. 6J), these findings suggest that chronic HIV infection or cART may modulate CHAF1A dynamics, but the core OGA-CHAF1A axis persists.
These results position CHAF1A as a potential age-associated latency stabilizer and underscore TFD's ability to counteract its accumulation in older individuals. Our findings highlight the need for age-stratified HIV-1 cure strategies.
## DISCUSSION
Our findings demonstrate that O-GlcNAcylation stabilizes CHAF1A by competitively blocking ubiquitination, thereby reinforcing heterochromatin at latent proviruses (Fig. 7). Conversely, disrupting O-GlcNAcylation-via TFD or enzymatic modulation of OGT/OGA -shifts the balance toward ubiquitin-dependent CHAF1A degradation, derepressing viral transcription (33). This PTM crosstalk not only elucidates a nutrient-sensitive "molecular switch" controlling latency but also positions CHAF1A as a dynamic therapeu tic node amenable to pharmacological intervention. The discovery of CHAF1A as a central regulator of HIV-1 latency, governed by the antagonistic interplay between flow cytometry for GFP expression. (E and F) J-Lat 10.6 cells were treated with increased concentration of TFD, PBS was used as a control. 24 h after treatment, the cells were harvested, the percentage of GFP-positive cells was measured by flow cytometry (E), and the cells were lysed, and western blot was performed with CHAF1A and β-actin antibodies (F). (G) OGT inhibitor reduces global O-GlcNAcylation in J-Lat 10.6 cells. J-Lat 10.6 cells were treated with OGT inhibitor at 0, 50, or 100 µM for 48 h.
Whole cell lysates were analyzed by western blot using an anti-O-GlcNAc antibody to assess global protein O-GlcNAcylation.
GAPDH was used as a loading control. (H) TFD treatment reduces global O-GlcNAcylation levels in a dose-dependent manner. J-Lat 10.6 cells were treated with increasing concentrations of TFD (0, 5, 10, and 20 µM) for 48 h. Whole cell lysates were analyzed by western blot using anti-O-GlcNAc antibody (CTD110.6) to assess global protein O-GlcNAcylation. GAPDH was used as a loading control. ns, not significant (P > 0.05); *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. ubiquitination and O-GlcNAcylation, provides a paradigm-shifting framework for understanding viral persistence. TFD, a clinically approved antiviral drug, emerges as a compelling LRA by uniquely targeting CHAF1A's PTM equilibrium. Unlike broad-spectrum histone deacetylase inhibitors (e.g., vorinostat) or non-specific PKC agonists, TFD's specificity for O-GlcNAcy lation pathways minimizes off-target toxicity-a critical limitation of current LRAs (3,38). By destabilizing CHAF1A, TFD selectively reactivates latent HIV-1 without global chromatin disruption, offering a safer profile for clinical translation (39). The identification of MIB1 as the E3 ligase mediating TFD-induced CHAF1A ubiquitination further validates its mechanism, providing a roadmap for optimizing small-molecule inhibitors targeting this axis (40,41). TFD functions as a novel LRA with dual mechanisms.
Our study also establishes TFD as a promising LRA with a clearly defined therapeutic window. Through rigorous cytotoxicity assessments using the CellTiter-Glo assay, we found that TFD can effectively reactivate latent HIV-1 at concentrations that do not cause significant toxicity in multiple cell lines and primary cells. This finding is crucial, as it addresses concerns about TFD's known systemic toxicity. Furthermore, we propose two distinct therapeutic strategies that leverage TFD's established clinical use and our new data. First, clinical literature shows that TFD is already safely used in systemic cancer treatment regimens, where appropriate dosing and scheduling successfully mitigate toxicity (29)(30)(31)42). This supports the feasibility of controlled systemic administration for HIV cure strategies. Second, given the drug's approved topical use for HSV, we investiga ted its potential for localized HIV-1 reservoir targeting. Our data from ex vivo rectal tissue explants demonstrated that TFD robustly reactivates latent virus in this relevant tissue compartment. This suggests a viable localized therapeutic strategy for tissue-resident reservoirs, which takes advantage of the drug's existing safety profile and approved route of administration for a related viral indication. Moreover, our identification of CHAF1A as an age-related biomarker addresses a pressing clinical challenge: older individuals (≥60 years) exhibit larger HIV-1 res ervoirs and delayed viral rebound post-ART, complicating cure strategies (43). The inverse correlation between CHAF1A and OGA levels in aging CD4 + T cells suggests that age-associated declines in OGA activity enhance O-GlcNAcylation-driven CHAF1A stabilization, perpetuating latency. This finding underscores the need for age-stratified approaches to reservoir clearance. For instance, TFD's efficacy may be heightened in older patients with elevated CHAF1A, while younger cohorts might benefit from combined OGT inhibition and immune checkpoint therapies. The finding of age-depend ent CHAF1A regulation may be an implication for HIV-1 precision medicine.
While our study establishes CHAF1A's PTM crosstalk as a key latency mechanism, several questions remain. First, the PTM of O-GlcNAcylation is nutrient-sensitive, which may link cellular metabolic state to HIV-1 latency and suggest metabolic interventions (e.g., modulating glucose/glutamine availability) could enhance LRA efficacy. Second, the discovery of MIB1 as a CHAF1A-specific E3 ligase opens avenues for developing proteolysis-targeting chimeras to enhance degradation efficiency. Finally, the tissuespecificity of CHAF1A regulation in sanctuary sites (e.g., central nervous system and lymphoid tissues) warrants exploration and long-term safety of TFD in HIV-1 patientsparticularly its impact on host chromatin stability-requires validation.
In conclusion, our work not only uncovers a novel mechanism of HIV latency reversal but also provides a compelling rationale for further exploring TFD as an LRA, especially for targeting tissue-based reservoirs. By unraveling the PTM crosstalk governing CHAF1A, this work redefines HIV-1 latency as a metabolically plastic state amenable to preci sion modulation. TFD's repurposing as a CHAF1A-targeted LRA, coupled with CHAF1A's potential as an age-related biomarker, bridges molecular discovery to critical clinical needs. As the global population of PLWH-1 ages, our findings advocate for tailored strategies that integrate metabolic regulation, aging biology, and targeted latency reversal-a triad essential for achieving functional cures.
## MATERIALS AND METHODS
## Study approval
The CD4 + T cells from healthy adult donors were provided by the Institutional Review Board of Guangzhou Blood Center (Guangzhou, Guangdong, China). The HIV-1-infec ted donors, who had undergone cART with undetectable HIV-1 viral loads in plasma (fewer than 50 copies/ml) for more than 6 months, were recruited for our study by the Department of Infectious Diseases in Guangzhou 8th People's Hospital, Guangzhou, China. All donors provided written informed consent.
## Cell culture
HEK293T (CVCL_0063; ATCC), TZM-bl (8129) cells obtained from NIH AIDS Reagent Program, were incubated in Dulbecco's modified Eagle medium (DMEM) (Thermo Fisher) supplemented with 1% penicillin-streptomycin (Thermo Fisher), 1% L-glutamine (Thermo Fisher), and 10% fetal bovine serum (FBS) (Thermo Fisher). J-Lat 10.6 cell line, which was created by Dr. Eric Verdin (The Buck Institute for Research on Aging, Novato, CA, USA) Laboratory, was obtained from Dr. Robert F. Siliciano (Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA) and was cultured in RPMI 1640 (Thermo Fisher) supplemented with 1% penicillin-streptomycin (Thermo Fisher), 1% L-glutamine (Thermo Fisher), and 10% FBS (Thermo Fisher). The CD4 + T cells from healthy donors and HIV-1-infected individuals were cultured in RPMI 1640 supplemented with 1% penicillin-streptomycin, 1% L-glutamine, 10% FBS, 1/1,000 recombinant human interleukin 2 (IL-2) (R&D Systems). All cells had been tested and confirmed to be mycoplasma-free by PCR assays. All cells were maintained in a clean incubator at 37°C and 5% CO 2 .
## Reagents and antibodies
DMEM, FBS, the Lipofectamine 2000 reagents, 4′,6-diamidino-2-phenylindole (DAPI), and penicillin-streptomycin were obtained from Gibco. The TFD, MG132, and chloro quine were purchased from Selleck. The PUGNAc and GlcN were purchased from Thermo Fisher. The following antibodies were used for this study: anti-CHAF1A antibody (17037-1-AP, Proteintech), anti-OGT antibodies (11576-2-AP, Proteintech), anti-O-GlcNAc antibody (CTD110.6; #12938, CST), anti-HA antibody (M180-3, MBL), anti-DDDDK antibody (PM020, MBL), anti-GAPDH antibody (10494-1-AP, Proteintech), IRDye 680RD goat anti-mouse IgG antibody (926-68070, LI-COR Biosciences), IRDye 800CW goat anti-rabbit IgG antibody (926-32211, LI-COR Biosciences), and anti-rabbit IgG H&L (Alexa Fluor 488) (ab150077, Abcam).
## High-throughput screening
HEK293T cells (1 × 10 5 ) were cultured in 200 µL of medium (DMEM with 10% FBS and 1% penicillin-streptomycin) in 96-well plates. The cells were co-transfected with 30 ng pEGFP-CHAF1A-C1 and 20 ng pcDNA3.1-CFP per well, in which fluorescence intensity of GFP was used to indicate degradation efficiency of drugs and CFP for quantification of cell viability (inversely correlated to compound toxicity). Four to six hours after transfection, the cells were treated with drugs from the "US DRUG COLLECTION" library at concentrations of 50 µM and dimethyl sulfoxide (DMSO) in the same volume as a negative control. Forty-eight hours later, the medium was thrown away, and the cells were washed in phosphate buffered saline (PBS), followed by detecting fluorescence intensity of GFP and CFP by Luminoskan Ascent Microplate Luminometers (Thermo Fisher). Then, the candidates of more than 50% degradation efficiency and less than 50% cell toxicity were calculated, analyzed, and picked up by normalizing with the negative control.
## Western blotting
HEK293T cells treated with TFD for 48 h were collected and lysed with NP40 lysis buffer (10 mM Tris-HCl buffered at pH 7.5, 150 mM NaCl, 0.5% NP-40, 1% Triton X-100, 10% glycerol, 2 mM EDTA, 1 mM NaF, 1 Mm Na3VO4) supplemented with protease inhibitor cocktail (Sigma Aldrich) on ice for 30 min. The lysis was vortexed every 10 min. Lysates were then clarified by centrifugation at 12,000 g for 5 min at 4°C, boiled at 100°C with loading buffer supplemented with DL-dithiothreitol (DTT) for 10 min and separa ted by SDS-PAGE. Proteins were transferred to nitrocellulose membranes (PALL). The membranes were then blocked with 5% non-fatty milk for 1 h at room temperature and incubated with anti-CHAF1A and anti-GAPDH antibodies overnight at 4°C. After three washes, membranes were incubated with IRDye secondary antibodies (LI-COR) for 1 h at room temperature and scanned with the Odyssey infrared imaging system (LI-COR).
## Co-IP assay
HEK293T cells were seeded on 6 cm plates and transfected with 3 µg CHAF1A-3HA and 4 µg 3Flag-ubiquitin. Transfection was performed using Lipofectamine 2000, according to the manufacturers' instructions. Cells were harvested and lysed with 500 µL NP40 lysis buffer supplemented with 1/100 protease inhibitor cocktail (Sigma Aldrich) and 2 M N-Ethylmaleimide (NEM) (Selleck) on ice for 30 min. The lysis was vortexed every 10 min. Lysates were then clarified by centrifugation at 12,000 g for 5 min at 4°C, and 80 µL of the lysates was taken as input control. The remaining lysates were rotated with anti-HA beads for 4 h to overnight at 4°C. The beads were then washed three to five times with 1,000 µL ice-cold STN IP wash buffer (10 mM Tri-HCl buffered at pH 7.5, 150 mM NaCl, 0.5% NP-40, 0.5% Triton X-100). Proteins were eluted with loading buffer supplemented with DTT at 100°C for 10 min. The CHAF1A ubiquitination was detected by western blotting using anti-HA (M180-3, MBL), anti-Flag (PM020, MBL), and anti-GAPDH (10494-1-AP, Proteintech) antibodies.
## Immunofluorescence assays
HEK293T cells were seeded on chambered coverglass (Thermo Fisher) and treated with TFD. Forty-eight hours later, cells were washed three times with PBS and fixed with 4% paraformaldehyde at room temperature for 10 min. The cells were washed three times with PBS and blocked with PBS containing 0.5% bovine serum albumin (BSA) at room temperature for half an hour. Cells then were incubated with anti-CHAF1A antibody (17037-1-AP, Proteintech) at 4°C overnight. Following washing with 0.1% Tween 20-PBS (PBS-T) three times, the cells were incubated with goat anti-rabbit IgG H&L (Alexa Fluor 488) (ab150077, Abcam) at room temperature for an hour. Cells then were washed with 0.1% PBS-T three times and incubated with DAPI at room temperature for 5 min, after which the cells were washed with 0.1% PBS-T three times. Fluorescent images were captured and analyzed with the N-SIM module of the NIS-Elements Advanced Research software (Nikon).
## Luciferase assay
TZM-bl cells treated with TFD for 48 h were collected and lysed with passive lysis buffer (Promega) for 30 min at room temperature. Lysates were then clarified, and luciferase in the cell lysates was measured with a luciferase reporter assay system (Promega).
## qPCR
The identity of whether the TFD influences CHAF1A mRNA, HEK293T Cells treated with TFD for 48 h were collected and lysed with TRIzol reagent (Thermo Fisher) and proceeded to cDNA synthesis with PrimeScript RT reagent Kit (Takara). Quantitative PCR (qPCR) was conducted using SYBR EX-Taq Premix (TaKaRa) in a CFX96 real-time PCR detection instrument (Bio-Rad). The data were analyzed by a SYBR green-based system (Bio-Rad) and normalized to GAPDH. Primer pairs were the following: GAPDH, forward, 5′ ATC CCA TCA CCA TCT TCC AGG 3′; reverse, 5′ CCT TCT CCA TGG TGG TGA AGA C 3′; CHAF1A, forward, 5′ TTA GAC CGA AAC TTG TCA ACG G 3′; reverse, 5′ GTC TGG CTG CTC ATT CGA GT 3′. The relative expression of each gene was calculated as 2 [Ct(Control-TRIM28)-Ct(Control-b-Actin)]-[Ct(Treatment-TRIM28)-Ct(Treatment-b-Actin)] . For the quantification of HIV-1 reactivation in primary CD4 + T cells of HIV-1-infected individuals, the procedure was previously described (31)(32)(33). Briefly, qPCR was performed for specific reverse-transcri bed HIV-1 cDNA with primer pairs: HIV-TotRNA Forward Primer: 5′-CTGGCTAACTAGGG AACCCACTGCT-3′ and HIV-TotRNA Reverse Primer: 5′-GCTTCAGCAAGCCGAGTCCTGCGTC -3′ (31). After quantitation, an in vitro-transcribed HIV-1 RNA was used as the external control for measuring cell-associated viral RNAs. The Ct of each group was converted to mass and further converted to copies. The final expression of intracellular HIV-1 RNA was represented as 10 3 copies of viral RNA per million CD4 + T cells.
## siRNA transfection
HEK293T cells were transfected with 100 nM of negative control, OGT-specific siRNA (siOGT-1: 5′-GGCACAAACTTCCGAGTGA -3′, siOGT-2: 5′-GCAGAAGCTTATTCGAATT -3′, siOGT-3: 5′-GCAGTTCGCTTGTATCGTA -3′) (Ribobio), or OGA-specific siRNA (siOGA-1: 5′-GCAGTTACTTGCTGATCTA-3′, siOGA-2: 5′-GCAAGAAGATTGTATTAGT-3′, siOGA-3: 5′-GG CACTTTCTGTTATCCAA-3′) (Ribobio), with Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Forty-eight hours later, the cells were collected and lysed for further western blot.
## shRNA-mediated knockdown and flow cytometry
The shRNA target sequence against OGT CDS was 5′ GCTGAGCAGTATTCCGAGAAA 3′. The shRNA targeting luciferase (shluc: 5′-ACCGCCTGAAGTCTCTGATTAA-3′) was set as the negative control (34). Target sequences were cloned into pLKO.3G-RFP which was derived from pLKO.3G, in which the GFP-tag was replaced with RFP-tag in pLKO.3G-RFP. pLKO.3G-RFP-shOGT and pLKO.3G-RFP-shLUC were produced in HEK293T cells by co-transfecting 3 mg of VSV-G (Addgene plasmid # 12259) glycoprotein-expression vector, 6 mg of lentiviral packaging construct pCMVDR8.2, which was a kind gift from Dr. Didier Trono (School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland) (35), and 6 µg shRNA expression lentiviral construct using Lipofectamine 2000 (Thermo Fisher) according to the manufacturer's instruction. The yield virus was concentrated into 1 mL RPMI 1640 by PEG 6000. J-Lat10.6 cells were spin-infected with shRNA virus. Forty-eight hours later, infected cells were harvested to identify knockdown efficiency by western blot, detecting infection efficiency by measuring the percentage of RFP-positive cells and confirming the reactivation effect by testing the GFP-positive cells using flow cytometry. The fluorescence value was analyzed by FlowJo software.
## Cell viability assay
The cell viability assay was conducted by measuring the percentage of amine-reactive fluorescent dye (BioLegend, 423113) non-permeant cells by flow cytometry, which indicated the percentage of viable cells.
## TZM-bl qVOA
Peripheral blood and rectal tissue were collected from ART-treated HIV-infected individuals. Rectal tissue was digested in DPBS (containing Ca² + /Mg² + ) supplemented with 1.0 mg/mL Collagenase IV, 1.5 mg/mL Dispase II, 0.5 mg/mL Hyaluronidase, 30 µg/ml DNase I, 1% BSA, and 1% penicillin/streptomycin at 37°C for 30-60 min. The tissue was gently pipetted to obtain a single-cell suspension, which was then filtered through a 70 µm cell strainer. PBMCs were isolated by Ficoll density gradient centrifuga tion after 1:2 dilution with PBS. All cells were washed twice with PBS and resuspended in culture medium. PBMCs or rectal cells were seeded at 1 × 10 5 to 2 × 10 5 per well in 24-well plates and treated with TFD (final concentrations 50 and 100 µM) and Dynabeads CD3/CD28 (12.5 µL per 1 million cells). Cells were cultured for 5 days in IMDM or RPMI 1640 with 10% FBS and 1% penicillin-streptomycin, with IL-2 supplementation. Controls received vehicle only. On day 5, activated cells were collected, counted, and washed twice with PBS. TZM-bl cells were seeded at 5 × 10 4 per well in 96-well white plates one day in advance. Activated T cells were added onto TZM-bl cells and co-cultured for 48-72 h. After co-culture, gently remove supernatants and wash wells twice with PBS. Add 100 µL Bio-Lite Luciferase Assay Working Solution to each well, incubate at room temperature (25°C) for 5 min in the dark, and immediately measure luminescence using a plate reader.
## sWGA pull-down assay
Cells were lysed in ice-cold lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 1 mM EDTA, and protease inhibitor cocktail) for 30 min on ice. Cell lysates were centri fuged at 12,000 × g for 15 min at 4°C to remove debris. The supernatants were collec ted, and protein concentrations were determined using the BCA assay (Thermo Fisher Scientific). For each sample, 250 µg total protein was incubated overnight at 4°C with 50 µL of pre-washed succinylated wheat germ agglutinin (sWGA) agarose beads (Vector Laboratories) to enrich O-GlcNAcylated proteins. After incubation, the beads were washed three times with lysis buffer containing 0.1% Tween-20 and twice with TBS to remove non-specifically bound proteins. Bound proteins were eluted by boiling in 60 µL of 3 × SDS sample buffer for 10 min at 95°C. The input and sWGA-enriched samples were analyzed by SDS-PAGE and immunoblotting with antibodies anti-O-GlcNAc (CTD110.6), anti-CHAF1A (17037-1-AP, Proteintech), and anti-GAPDH(10494-1-AP, Proteintech).
## Rescue experiments with CHAF1A overexpression
TZMBL cells were seeded in 24-well plates and first transfected with Tat expression plasmid to activate luciferase expression. After Tat transfection, cells were treated with DMSO or 20 µM TFD for a specified period. Subsequently, different amounts of CHAF1A plasmid (CHAF1A-1μg and CHAF1A-2μg) were transfected into designated wells. After incubation, single-luciferase reporter activity was measured using a luciferase assay kit according to the manufacturer's instructions. All experiments were performed in triplicate. Statistical differences between groups were analyzed by one-way analysis of variance (ANOVA) followed by post hoc tests.
CRISPR-Cas9-mediated sgRNA targeting OGA/sgRNA targeting OGT knockout in J-Lat10.6 cells sgRNA sequences targeting OGA (5′-CTTTGGGTCCATGCTCGTA-3′) and OGT (5′-GCACGC GTATAACACTGCA-3′) genes were designed using the respective gene sequences. For cloning into the lentiCRISPR v2 vector (Addgene #52961), BsmBI restriction site-compati ble overhangs were added as follows:
OGA forward: 5′-caccgCTTTGGGTCCATGCTCGTA-3′ OGA reverse: 5′-aaacTACGAGCATGGACCCAAAGc-3′ OGT forward: 5′-caccgGCACGCGTATAACACTGCA-3′ OGT reverse: 5′-aaacTGCAGTGTTATACGCGTGCc-3′ All oligonucleotides were synthesized by Ruibo Biotech (Guangzhou, China). The forward and reverse oligos were annealed and ligated into BsmBI-digested lentiCRISPR v2. Recombinant plasmids were verified by colony PCR and Sanger sequencing. Oligos were synthesized with BsmBI-compatible overhangs, annealed, and ligated into BsmBI-digested lentiCRISPR v2 vector. Recombinant plasmids were confirmed by colony PCR and Sanger sequencing. Lentiviral packaging was performed by co-transfect ing HEK293T cells with lentiCRISPR v2-sgRNA, VSV-G, and psPAX2, followed by viral concentration. J-Lat10.6 cells were infected with concentrated virus, and after 48 h, cells were selected with puromycin at the optimal concentration determined by a kill curve. Knockout efficiency was verified by western blot.
## Statistical analysis
The results of the experiments are presented as means ± standard errors of the means (SEM). Student's unpaired t-test was used to determine significance. P values are denoted in figures as ns: no significance, *: P < 0.05, **: P < 0.01, and ***: P < 0.001. clusterProfiler package (45). Gene Set Enrichment Analysis (GSEA) was performed based on the pre-ranked gene lists to identify significantly enriched pathways. The correlation between genes was evaluated using Pearson correlation analysis.
## References
1. Chun, Stuyver, Mizell et al. (1997) "Presence of an inducible HIV-1 latent reservoir during highly active antiretroviral therapy" *Proc Natl Acad Sci*
2. Siliciano, Kajdas, Finzi et al. (2003) "Long-term follow-up studies confirm the stability of the latent reservoir for HIV-1 in resting CD4+ T cells" *Nat Med*
3. Spivak, Planelles (2018) "Novel latency reversal agents for HIV-1 cure" *Annu Rev Med*
4. Smit, Brinkman, Geerlings et al. (2015) "Future challenges for clinical care of an ageing population infected with HIV: a modelling study" *Lancet Infect Dis*
5. Sauce, Larsen, Fastenackels et al. (2011) "HIV disease progression despite suppression of viral replication is associated with exhaustion of lymphopoiesis" *Blood*
6. Rinaldi, Pallikkuth, Armas et al. (2022) "Distinct molecular signatures of aging in healthy and HIV-infected individuals" *J Acquir Immune Defic Syndr*
7. Smith, Stillman (1989) "Purification and characterization of CAF-I, a human cell factor required for chromatin assembly during DNA replication in vitro" *Cell*
8. Bulger, Ito, Kamakaka et al. (1995) "Assembly of regularly spaced nucleosome arrays by Drosophila chromatin assembly factor 1 and a 56-kDa histone-binding protein" *Proc Natl Acad Sci*
9. Verreault, Kaufman, Kobayashi et al. (1996) "Nucleosome assembly by a complex of CAF-1 and acetylated histones H3/H4" *Cell*
10. Ma, Chen, Peng et al. (2021) "Histone chaperone CAF-1 promotes HIV-1 latency by leading the formation of phase-separated suppressive nuclear bodies" *EMBO J*
11. Li, Lu, Yuan et al. (2024) "O-GlcNAc of STING mediates antiviral innate immunity" *Cell Commun Signal*
12. Sun, Zhi, Ren et al. (2024) "Crosstalk between O-GlcNAcylation and ubiquitination: a novel strategy for overcoming cancer therapeutic resistance" *Exp Hematol Oncol*
13. Yin, Wang, Xu et al. (2025) "Asperuloside suppresses the progression of depression through O-GlcNAcylation of IκBα and regulating NFκB signaling" *J Pharmacol Sci*
14. Wang, Wang, Xiao et al. (2025) "Dynamic O-GlcNAcylation coordinates etoposide-triggered tumor cell pyroptosis by regulating p53 stability" *J Biol Chem*
15. Li, Zhang, Zhang et al. (2025) "Mechanism of O-GlcNAcylation regulating liver lipid synthesis in mice through FASN" *FASEB J Off Publ Fed Am Soc Exp Biol*
16. O'sullivan, Kubicek, Schreiber et al. (2010) "Reduced histone biosynthesis and chromatin changes arising from a damage signal at telomeres" *Nat Struct Mol Biol*
17. Xie, Xie, Liu et al. (2023) "Emerging role of protein O-GlcNAcylation in liver metabolism: implications for diabetes and NAFLD" *Int J Mol Sci*
18. He, Li, Chen et al. (2022) "O-GlcNAcylation and stablization of SIRT7 promote pancreatic cancer progression by blocking the SIRT7-REGγ interaction" *Cell Death Differ*
19. King, Tkachuk, Tseng (2024) "Aging on antiretrovirals: reviewing the need for pharmacologic data in elderly people with HIV" *AIDS*
20. Hyndiuk, Charlin, Alpren et al. (1978) "Trifluridine in resistant human herpetic keratitis" *Arch Ophthalmol*
21. Trifluridine (1980) "(viroptic) for herpetic keratitis" *Med Lett Drugs Ther*
22. Patel, Duh, Barghout et al. (2018) "Real-world treatment patterns among patients with colorectal cancer treated with trifluridine/tipiracil and regorafenib" *Clin Colorectal Cancer*
23. Feng, Yin, Zhang et al. (2024) "O-GlcNAcylation of E3 ubiquitin ligase SKP2 promotes hepatocellular carcinoma proliferation" *Oncogene*
24. Ruan, Nie, Yang (2013) "Regulation of protein degradation by O-GlcNAcylation: crosstalk with ubiquitination" *Mol Cell Proteomics*
25. Love, Ghosh, Mondoux et al. (2010) "Dynamic O-GlcNAc cycling at promoters of Caenorhabditis elegans genes regulating longevity, stress, and immunity" *Proc Natl Acad Sci*
26. Paneque, Fortus, Zheng et al. (2023) "The hexosamine biosynthesis pathway: regulation and function" *Genes (Basel)*
27. Dong, Hart (1994) "Purification and characterization of an O-GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen cytosol" *J Biol Chem*
28. Haltiwanger, Holt, Hart (1990) "Identification of a uridine diphospho-N-acetylglucosamine:peptide beta-N-acetylglucosaminyl transferase" *J Biol Chem*
29. Prager, Taieb, Fakih et al. (2023) "Trifluridine-tipiracil and bevacizumab in refractory metastatic colorectal cancer" *N Engl J Med*
30. Shitara, Doi, Dvorkin et al. (2018) "Trifluridine/tipiracil versus placebo in patients with heavily pretreated metastatic gastric cancer (TAGS): a randomised, double-blind, placebo-controlled, phase 3 trial" *Lancet Oncol*
31. Huang, Wu, Wei et al. (2025) "Trifluridine/tipiracil induces ferroptosis by Full-Length Text Journal of Virology December"
32. "targeting p53 via the p53-SLC7A11 axis in colorectal cancer 3D organoids" *Cell Death Dis*
33. Zheng, Shabek (2017) "Ubiquitin ligases: structure, function, and regulation" *Annu Rev Biochem*
34. Dho, Silva-Gagliardi, Morgese et al. (2019) "Proximity interactions of the ubiquitin ligase Mind bomb 1 reveal a role in regulation of epithelial polarity complex proteins" *Sci Rep*
35. Wang, Hu, Ye et al. (2020) "NEDD4 E3 ligase: functions and mechanism in human cancer" *Semin Cancer Biol*
36. Balsollier, Tomašič, Yasini et al. (2023) "Design of OSMI-4 analogs using scaffold hopping: investigating the importance of the uridine mimic in the binding of OGT inhibitors" *ChemMedChem*
37. Sanyal, Rangachar, Gupta (2019) "TZA, a sensitive reporter cellbased assay to accurately and rapidly quantify inducible, replicationcompetent latent HIV-1 from resting CD4+ T Cells" *Bio Protoc*
38. Luk, Yim, Zhou et al. (2025) "BRD9 functions as an HIV-1 latency regulatory factor"
39. Margolis, Garcia, Hazuda et al. (2016) "Latency reversal and viral clearance to cure HIV-1" *Science*
40. Archin, Margolis (2014) "Emerging strategies to deplete the HIV reservoir" *Curr Opin Infect Dis*
41. Metzger, Pruneda, Klevit et al. (2014) "RING-type E3 ligases: master manipulators of E2 ubiquitin-conjugating enzymes and ubiquitination" *Biochim Biophys Acta*
42. Berndsen, Wolberger (2014) "New insights into ubiquitin E3 ligase mechanism" *Nat Struct Mol Biol*
43. Ando, Matsuoka, Orito et al. (2025) "Comparison of standard and biweekly trifluridine/tipiracil plus bevacizumab regimens in advanced or recurrent colorectal cancer: a retrospective study" *Jpn J Clin Oncol*
44. Hossain, Lungu, De Schrijver et al. (2024) "Specific quantification of inducible HIV-1 reservoir by RT-LAMP"
45. *Commun Med (Lond)*
46. Love, Huber, Anders (2014) "Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2" *Genome Biol*
47. Yu, Wang, Han et al. (2012) "clusterProfiler: an R package for comparing biological themes among gene clusters" *OMICS* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12824426&blobtype=pdf | # Evaluation of the Humoral Immune Response Following Two Doses of a Coronavirus Disease 2019 Vector-Based Vaccine During the Initial Rollout in Bangladesh
Sharmin Sultana, Md Rahman, Abu Taher, Islam
## Abstract
Introduction: The rollout of coronavirus disease 2019 (COVID-19) vaccination was crucial in addressing the pandemic in Bangladesh, with the ChAdOx1 vaccine being the primary vaccine administered to most of the population. This study aims to assess the humoral immune response by measuring severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) IgG titers after two doses of the ChAdOx1 vaccine in the Bangladeshi population, irrespective of prior COVID-19 infection status, and to identify antibody titer levels that can predict future COVID-19 infection.Methods: This was a cross-sectional study conducted among 256 individuals exhibiting COVID-19-like respiratory symptoms who had completed two doses of ChAdOx1 between December 2021 and March 2022. Participants were tested for COVID-19 via nasopharyngeal swabs using real-time polymerase chain reaction (PCR), and SARS-CoV-2 IgG antibody titers (≥33.8 binding antibody unit (BAU)/mL was considered positive) were measured from blood samples with chemiluminescence immunoassay (CLIA) at the COVID-19 laboratory, Department of Virology, Bangladesh Medical University (BMU). Demographic and clinical data were collected, and results were analyzed using IBM SPSS Statistics software version 22 (IBM Corp., Armonk, NY), with a P-value of <0.05 deemed statistically significant.Results: The overall seropositivity rate for SARS-CoV-2 IgG antibodies was 86.7%, with a median antibody titer of 501.0 BAU/mL (range: 4.81-2080.0 BAU/mL). Individuals with a prior COVID-19 infection exhibited significantly higher antibody titers (P<0.01) (mean: 1262.12±864.46 BAU/mL, median: 1465 BAU/mL) compared to those without a prior infection. The receiver operating characteristics (ROC) curve analysis (area under the curve (AUC): 0.823, 95% confidence interval (CI): 0.769-0.877, P<0.001) identified a minimum antibody threshold of 359.5 BAU/mL for preliminary immunity against future COVID-19 infections (sensitivity: 81.3%, specificity: 65.9%, Youden's Index: 0.47). Despite vaccination and a history of previous COVID-19, 1.5% of the studied population presented as reinfection with SARS-CoV-2, confirmed by real-time PCR testing. Comorbidity variables, such as diabetes mellitus, asthma, chronic obstructive pulmonary disease (COPD), and hypertension, showed no statistically significant association with the antibody response.Conclusion: This study demonstrated that ChAdOx1 elicits robust antibody responses in the majority of individuals, with significantly stronger reactions observed among those with prior COVID-19 infection. Moreover, it provides effective protection against recurrent infection in this population, provided that a SARS-CoV-2 IgG titer is sustained above the minimum threshold of 359.5 BAU/mL. Nevertheless, additional research is warranted to characterize the SARS-CoV-2 IgG response across diverse populations receiving heterologous or multi-regimen vaccination strategies.
## Introduction
Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has remained a global challenge since its outbreak began in 2019. Approximately 778 million people worldwide have been infected, with seven million having died from coronavirus disease 2019 (COVID- 19). In Bangladesh, 2,052,210 confirmed cases and 29,530 deaths were reported as of September 24, 2025 [1,2]. The world responded quickly to this pandemic, and several highly effective SARS-CoV-2 vaccines were developed within months to contain this situation, as safe and effective vaccines made a significant contribution to its control [3][4][5]. In Bangladesh, the COVID-19 vaccine rollout began in January 2021. So far, AstraZeneca (ChAdOx1), Pfizer-BioNTech (BNT162b2; Comirnaty), Moderna (mRNA-1273; Spikevax), Sinopharm (BBIBP-CorV), Sinovac (CoronaVac), Janssen (Johnson & Johnson; Ad26.COV2.S), and Pfizer-PF (BNT162b2; Comirnaty) have been used to combat the pandemic. Among these, the Oxford-AstraZeneca (available as Covishield, Serum Institute of India) vaccine was used during the initial vaccination effort, targeting the most vulnerable populations, such as healthcare workers and those with comorbidities [6,7].
The immune response to any vaccine varies based on geographical location, age, and an individual's immunological status, with stronger immunity observed in those who have been previously exposed to the SARS-CoV-2 virus. A history of COVID-19 enhances antibody production through natural immunity, which is further boosted by vaccine-induced immune responses [8,9]. However, it emphasizes the need to follow up on vaccine-term immunity and vaccine effectiveness to determine the most effective immunization policies, especially in the event of frequent SARS-CoV-2 variant emergence [9].
Several research articles on determining the IgG responses from vaccines, regardless of the number of doses and formulations from our country, have been published [10][11][12]. A study from India showed 81.5% effectiveness of two doses of the Covishield vaccine against moderate-to-severe COVID-19 [13]. In our country, where Covishield remains one of the main vaccines administered, only a few studies have quantified SARS-CoV-2 IgG titers after two doses in Bangladeshi adults. There is also a knowledge gap regarding the determination of optimal SARS-CoV-2 protective antibody thresholds for reinfection. Therefore, this study aims to evaluate the SARS-CoV-2 IgG antibody response among individuals vaccinated with two doses of Covishield, with or without prior COVID-19 infection, and to determine antibody titers that predict subsequent COVID-19 infection.
## Materials And Methods
## Study population and settings
This was an exploratory cross-sectional study with inferential analysis conducted among individuals who visited the fever clinic at Bangladesh Medical University (BMU) (formerly Bangabandhu Sheikh Mujib Medical University (BSMMU)) from December 2021 to March 2022. Participants with COVID-19-like respiratory symptoms [14] were tested for COVID-19 using real-time polymerase chain reaction (PCR) at the COVID-19 laboratory in the Department of Virology, BMU. This laboratory was among the first of its kind, starting services on April 1, 2020, under pandemic conditions [15]. According to the study's goals, only individuals who had received two doses of the Covishield vaccine were enrolled through convenience sampling, provided they consented to participate when approached. At the time of specimen collection, no participants had received an additional booster shot, and the median time since the second vaccine shot was 190 days. Data collection during their visits involved a predefined written questionnaire that gathered demographic information, a history of previous COVID-19 infection (confirmed by COVID-19 real-time PCR), and the duration since their last vaccine dose. The detailed procedures for the study population are shown in Figure 1.
## Laboratory procedure
Maintaining biosafety precautions, a nasopharyngeal swab was collected for COVID-19 PCR testing, and approximately 3 mL of venous blood was drawn into ethylenediaminetetraacetic acid (EDTA)-containing vacutainers for SARS-CoV-2 IgG titer testing. The detection of COVID-19 infection was performed using the Novel Coronavirus (2019-nCoV) Nucleic Acid Diagnostic Kit (Sansure Biotech Inc., Changsha, China) on the QuantStudio™ 5 Real-Time PCR System (Thermo Fisher Scientific Inc., Waltham, MA). The SARS-CoV-2specific IgG antibody titer was measured at the Department of Virology, BMU, using the chemiluminescence immunoassay (CLIA) method with the LIAISON® SARS-CoV-2 TrimericS IgG assay kit (REF 311510, DiaSorin, Saluggia, Italy) on the LIAISON® XL analyzer (DiaSorin S.p.A, Saluggia, Italy), following the manufacturer's instructions. SARS-CoV-2 IgG titers were expressed in arbitrary units (AU/mL), with values <13 and ≥13 AU/mL interpreted as negative and positive for SARS-CoV-2 antibodies, respectively. This cutoff was chosen based on the assay manufacturer's instructions. In patients, seropositivity was defined as prior exposure to SARS-CoV-2 through natural infection and/or receipt of a COVID-19 vaccine. The test results were further converted to binding antibody units (BAU/mL) (AU/mL × 2.6), where values <33.8 BAU/mL and ≥33.8 BAU/mL were considered negative and positive for SARS-CoV-2 antibodies, respectively. The conversion of IgG titer values from AU/mL to BAU/mL followed the validation of the first WHO International Standard (IS) for anti-SARS-CoV-2 immunoglobulin binding activity (NIBSC 20-136) [16]. This operational definition was applied uniformly across all patient samples included in the study.
## Statistical analysis
Results were presented as numbers (n) and percentages (%), and values were calculated as means (standard deviation (SD)), medians, and ranges. The comparison of SARS-CoV-2 IgG titer values, based on COVID-19 infection history, was analyzed using the Mann-Whitney U test. A general linear regression model was employed, with SARS-CoV-2 IgG titer values as the dependent variable. Independent variables included gender, age category, presence of comorbidities, prior infection history, and time elapsed since the second vaccination. To determine the minimum COVID-19 antibody titer associated with protection against SARS-CoV-2 PCR positivity, a receiver operating characteristic (ROC) curve analysis was conducted. The COVID-19 antibody titer (BAU/mL) served as the quantitative variable, effectively discriminating against subsequent COVID-19 infection based on SARS-CoV-2 PCR positivity results. The area under the curve (AUC) was calculated to assess overall test performance, and the optimal threshold for protective antibody titer was identified using the highest Youden's J Index (J=Sensitivity+Specificity-1). The statistical analysis was performed using IBM SPSS Statistics software version 22 (IBM Corp., Armonk, NY), and a P-value of <0.05 was considered significant.
## Ethical considerations
This study was ethically approved by the Institutional Review Board of BMU (reference number: BSMMU/2021/12406, dated 11/29/21), and written informed consent was obtained from all participants before and during enrollment in the study.
## Results
A total of 256 individuals, aged 18-72 years (mean (SD): 33.67±10.03 years), participated in this study. Of these, 62.5% were male, and most participants (73.4%) were in the 21 The overall seropositivity rate for SARS-CoV-2 IgG antibodies was 86.7% (n=222), and the median antibody titer across the cohort was 501.0 BAU/mL (range: 4.81-2080 BAU/mL). Among participants, those with prior COVID-19 have significantly higher (P<0.01) antibody titers (mean: 1262.12±864.46 BAU/mL, median: 1465 BAU/mL) compared to those without previous infection (mean: 773.59±789.42 BAU/mL, median: 455 BAU/mL) (Figure 2a and Figure 2b). In this study, we used a generalized linear model to examine factors associated with SARS-CoV-2 IgG levels among individuals with different clinical and demographic characteristics (Table 2). All independent variables were categorical and interpreted using reference categories, and the model shows a moderate fit (R²=0.297, adjusted R²=0.204), allowing identification of which predictors meaningfully influence outcomes relative to their reference groups. The model identified two critical factors that were strongly linked to IgG levels. Participants who reported having had COVID-19 showed higher antibody levels than those without such a history (β=430.10, p=0.001). Another notable finding was that individuals who received their second vaccine dose more than 12 months earlier had significantly higher antibody levels compared to those vaccinated within 1-3 months (β=937.37, P=0.015). However, none of the comorbidity variables, such as diabetes mellitus, asthma, chronic obstructive pulmonary disease (COPD), or hypertension, significantly affected IgG responses. This study investigates the minimum SARS-CoV-2 titer values best to demarcate protective immunity against subsequent COVID-19 infection. The ROC curve analysis (AUC=0.823, 95% confidence interval: 0.769-0.877, P<0.001) revealed a minimum antibody threshold of 359.5 BAU/mL for preliminary protective immunity against future COVID-19 infections (sensitivity: 81.3%, specificity: 65.9%, Youden's Index: 0.47) (Figure 3).
## Variable
## Discussion
This study provides important insights into the humoral immune response to the Covishield vaccine within a Bangladeshi population during the early pandemic phase, one year after the country's vaccination campaign.
The study group was mostly male participants, which may reflect a real-world situation where men were more exposed or given priority for vaccination due to job roles. Most participants were between 21 and 40 years old, indicating a mostly young adult group. This is an important demographic in many vaccination efforts because of their high mobility and social interactions, which could make them key in virus transmission dynamics [17]. Most individuals who took part in this study completed their primary vaccination series within the last 3-6 months before sample collection. Recognized risk factors for severe COVID-19 were hypertension, diabetes mellitus, malignancy, immune suppression, and others [18]. Hypertension and bronchial asthma were the most common comorbidities. Notably, conditions such as chronic kidney or liver disease were not observed, possibly due to the generally good health profile of the population. We did not find any differences concerning these comorbid conditions.
In this study, we observed significant variability in SARS-CoV-2 IgG levels among the participants, indicating a diverse immune response. Most individuals were seroconverted, confirming that the vaccines effectively generate humoral immunity after COVID-19 vaccination [19]. The history of COVID-19 infection was a strong independent predictor of higher antibody levels, highlighting the impact of hybrid immunity. Interestingly, those vaccinated more than 12 months prior had significantly higher titers, possibly due to repeated exposure to the virus. A prior study from Bangladesh showed that the Covishield vaccine elicited robust antibody responses, regardless of previous SARS-CoV-2 exposure. Additionally, the COVID-19infected group had 2.2 times higher antibody levels than the non-infected group one month after the second dose of Covishield [11]. Therefore, the timing of measurement and prior infection status are crucial for understanding antibody responses and assessing protection against infection or severe disease. These findings align with another study suggesting that repeated antigen exposure through vaccination, natural infection, or both enhances immune protection. This concept, known as "hybrid immunity," is associated with stronger immune responses. The results imply that booster doses or natural infection in vaccination strategies may offer better protection than vaccination alone [20].
Different comorbid conditions (e.g., hypertension, cardiac disease, kidney disease, and diabetes mellitus) were also analyzed for their associations with COVID-19 vaccination, and no significant links were found. However, the presence of comorbidities did not seem to affect the vaccine's immunogenicity observed in this study negatively. One study indicates that participants with comorbid conditions showed antibody responses after the second dose that were statistically similar to those without comorbidities [12]. Another survey of Taiwanese people revealed that individuals with comorbidities had a weaker antibody response after receiving three doses of COVID-19 vaccination [21].
Recurrent COVID-19 infection, also known as a breakthrough infection, is a concern for vulnerable populations. However, it was less common among those who had previous infections or were vaccinated with strong immunity. Although reinfection can still happen, it is relatively uncommon. It is more often seen in individuals with lower immunity levels despite their prior infection or vaccination [22]. This study found that SARS-CoV-2 humoral immunity of 359.5 BAU/mL provides initial protection against future COVID-19 infections. Four individuals in the study who faced reinfection despite prior infection or vaccination had lower SARS-CoV-2 immunity levels. However, immune responses can differ based on a person's pre-existing immune status, the type and dose of vaccines, and reinfection with different SARS-CoV-2 variants. The levels of SARS-CoV-2 antibodies also varied, with 659 BAU/mL observed in a study involving two doses of mRNA-based vaccines [23]. Another survey of healthcare workers who received four doses of BNT162b2 showed their SARS-CoV-2 response was 2483 BAU/mL, which helped gauge the risk of infection [24]. A study from India showed that seropositivity to Covishield (96.7%) was significantly higher than that to Covaxin (77.1%) [25]. Overall, this immune response may vary depending on an individual's preexisting immunity, the type and amount of vaccine received, and reinfection with any SARS-CoV-2 variant.
This study evaluated the humoral immune response to Covishield and documented strong antibody responses in most individuals, especially those with a history of previous infection. It also identified the minimum protective COVID-19 antibody titer after a two-dose vaccination schedule. Therefore, monitoring antibody titers can help identify populations that would benefit from booster doses in low-resource settings, which is a public health priority. However, this study has several limitations. It was conducted at a single center, involving only the participants who attended, so the results from a small sample may not be generalized to the wider population. The timing of antibody testing after vaccination varied among participants, potentially influencing titer levels. Additionally, not testing for neutralizing antibodies or cellular immunity limits understanding of proper immune protection. Due to the cross-sectional design of this study, no causal inference or assessment of antibody decline over time was possible. Furthermore, prior infection status was self-reported from prior COVID-19 PCR results and may have been affected by recall bias or underreporting, especially among asymptomatic individuals. We were unable to perform SARS-CoV-2 variant analysis using high-throughput methods due to budget constraints. Despite these limitations, the study offers valuable insights into vaccine-induced serological responses and the potential to predict future infections based on immunity levels in a low-to middle-income country setting. Long-term follow-up may provide helpful information to help fill the knowledge gap and enhance understanding.
## Conclusions
This study found that vaccination triggers antibody responses in most individuals, especially those with prior infection, and that receiving two doses of the vaccine effectively predicts immunity. Notably, reinfection could be prevented in individuals with a SARS-CoV-2 IgG titer above 359 BAU/mL, although this should be interpreted with caution. Further studies involving larger, more diverse populations and different vaccine schedules could help clarify the impacts of comorbidities and the longevity of long-term antibody responses.
## References
1. (2025) "World Health Organization: WHO COVID-19 dashboard"
2. (2025) "COVID-19 Dashboard: COVID-19 dashboard for Bangladesh"
3. Klasse, Nixon, Moore (2021) "Immunogenicity of clinically relevant SARS-CoV-2 vaccines in nonhuman primates and humans" *Sci Adv*
4. Sadarangani, Marchant, Kollmann "Immunological mechanisms of vaccine-induced protection against COVID-19 in humans" *Nat Rev Immunol. 2021*
5. Krammer (2020) "SARS-CoV-2 vaccines in development" *Nature*
6. (2022) "COVID-19 vaccination in Bangladesh"
7. (2025) "Directorate General of Health Services: COVID-19 vaccination dashboard for Bangladesh"
8. Tut, Lancaster, Krutikov (2021) "Profile of humoral and cellular immune responses to single doses of BNT162b2 or ChAdOx1 nCoV-19 vaccines in residents and staff within residential care homes (VIVALDI): an observational study" *Lancet Healthy Longev*
9. Wang, Nair, Liu (2021) "Antibody resistance of SARS-CoV-2 variants B.1.351 and B.1.1.7" *Nature*
10. Sharmin, Majumder, Ahmed (2025) "Adverse effects and SARS-CoV-2 infection after COVID-19 vaccination among the vaccinated people of Bangabandhu Sheikh Mujib Medical University Hospital in Dhaka, Bangladesh: a pilot study" *Sci Rep*
11. Bhuiyan, Akhtar, Khaton (2022) "Covishield vaccine induces robust immune responses in Bangladeshi adults" *IJID Reg*
12. Hoque, Barshan, Chowdhury et al. (2021) "Antibody response to ChAdOx1-nCoV-19 vaccine among recipients in Bangladesh: a prospective observational study. Infect Drug Resist"
13. Thiruvengadam, Awasthi, Medigeshi (2022) "Effectiveness of ChAdOx1 nCoV-19 vaccine against SARS-CoV-2 infection during the delta (B.1.617.2) variant surge in India: a test-negative, case-control study and a mechanistic study of post-vaccination immune responses" *Lancet Infect Dis*
14. Islam, Ghosh, Begum et al. (2024) "The distribution of respiratory viral pathogens among the symptomatic respiratory tract infection patients from Dhaka city in the pre-COVID-19 pandemic era" *Cureus*
15. Islam, Akther, Sultana (2021) "Challenges in the establishment of a biosafety testing laboratory for COVID-19 in Bangladesh" *J Infect Dev Ctries*
16. (2020) "First WHO international standard for anti-SARS-CoV-2 immunoglobulin (human)"
17. Erfan Uddin, Sayeed, Ghosh et al. (2022) "Pharmacovigilance of COVID-19 vaccine among health care workers in Chittagong Medical College Hospital" *J Chittagong Med Coll Teach Assoc*
18. Zhang, Zhu, Jiang (2025) "In-depth analysis of the risk factors for persistent severe acute respiratory syndrome coronavirus 2 infection and construction of predictive models: an exploratory research study" *BMC Infect Dis*
19. Voulgaridi, Sarrou, Dadouli (2022) "Intensity of humoral immune responses, adverse reactions, and post-vaccination morbidity after adenovirus vector-based and mRNA anti-COVID-19 vaccines. Vaccines (Basel)"
20. Hammerman, Sergienko, Friger (2022) "Effectiveness of the BNT162b2 vaccine after recovery from Covid-19" *N Engl J Med*
21. Huang, Jang, Wu (2023) "Impact of comorbidities on the serological response to COVID-19 vaccination in a Taiwanese cohort" *Virol J*
22. Bergwerk, Gonen, Lustig (2021) "COVID-19 breakthrough infections in vaccinated health care workers" *N Engl J Med*
23. Sormani, Schiavetti, Inglese (2022) "Breakthrough SARS-CoV-2 infections after COVID-19 mRNA vaccination in MS patients on disease modifying therapies during the Delta and the Omicron waves in Italy" *EBioMedicine*
24. Shachor-Meyouhas, Dabaja-Younis, Magid (2023) "Immunogenicity and SARS-CoV-2 infection following the fourth BNT162b2 booster dose among health care workers. Vaccines (Basel)"
25. Dash, Subhadra, Turuk (2022) "Breakthrough SARS-CoV-2 infections among BBV-152 (COVAXIN®) and AZD1222 (COVISHIELD(TM) ) recipients: report from the eastern state of India" *J Med Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12542123&blobtype=pdf | # Correction: HPV infection incidence and genotype distribution among male patients visiting outpatient departments in Huizhou from 2014 to 2023
Xianjin Wu, Caiyi Wen, Xiaohan Yang
## Abstract
In this article [1], the bottom part of the Table 2, which includes data for all 21 HPV genotypes, were missing. For completeness and transparency, both incorrect and correct versions are displayed below.
## Incorrect Table 2:
In cases of multiple infections, subjects may be counted more than once. In the 5*2 group multiple comparisons, the alpha-splitting method was applied.At a significance level of ɑ=0.005,"a"indicates a significant difference compared to the period from 2014 to 2019,"b" indicates a significant difference compared to the year 2020,and columns without labeled letters show no significant differences in proportions
Correct Table 2:
## References
1. Wen, Yang, Wu (0272) "HPV infection incidence and genotype distribution among male patients visiting outpatient departments in Huizhou from 2014 to 2023" *Virol J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12770888&blobtype=pdf | # Study protocol: Effectiveness of the maternal RSVpreF vaccine by virus type
Anna Mensah, Rebecca Symes, Chengetai Mpamhanga, Nick Andrews, Lynne Ferguson, Rory Gunson, Katja Hoschler, Beatrix Kele, Wei Lim, Jamie Lopez Bernal, Ross Mcqueenie, Chris Robertson, Tiina Talts, Heather Whitaker, Kimberly Marsh, Conall Watson, Maria Zambon, Thomas Williams, Ram Chapagain
## Abstract
BackgroundRespiratory syncytial virus (RSV) is a virus with two antigenic types, A and B, that cause significant morbidity and mortality in infants globally. A recently developed maternal vaccination based on the prefusion F protein ("RSVpreF") could have a significant impact on disease burden, if introduced globally. Whether or not the effectiveness of this vaccine is affected by circulating viral genomic variability is currently unknown.
ObjectivesTo examine whether the vaccine effectiveness of maternal RSVpreF administration in preventing hospitalisation in infants is affected by RSV type or lineage.
MethodsWe will conduct whole genome sequencing of RSV positive samples from infants hospitalised with acute lower respiratory tract infection
Open Peer Review
(ALRI) in the 2024-2025 winter season, at multiple hospitals in England and Scotland, to calculate the relative vaccine effectiveness (rVE) of maternal RSVpreF vaccination by virus type (RSV-A and RSV-B). rVE will be calculated using a case control logistic regression with adjustment by infant age and admission date; sex, socioeconomic status and hospital location will be included as potential confounders if they are associated with a >3% change in rVE. We will also perform a test negative design to examine the VE for RSV-A and RSV-B separately, using RSV-negative controls from hospitals where cases were admitted. Finally, we will compare viral lineages in vaccinated versus unvaccinated infants.
## Results and conclusions
Our study will identify whether currently circulating RSV genomic variability impacts on rVE. Confirmation of the null hypothesis -that there is no impact of viral genomic variability on rVE -will provide reassurance to policymakers and public health bodies as RSVpreF is rolled out globally. Conversely, an association between RSV type or lineage and decreased vaccine effectiveness will highlight the need for the enhanced comprehensive national and global molecular surveillance of RSV.
## Plain language summary
Respiratory syncytial virus, often shortened to be called RSV, is a common virus that can cause severe infections in children. In babies, RSV is more likely to cause severe disease, and RSV disease is a common cause of hospital admission in this group. Recently, an effective vaccine against RSV, given to pregnant women to protect their babies after birth, has been introduced in some countries, including England and Scotland. There are two types of RSV, called RSV-A and RSV-B, which circulate in communities every year; whether the maternal vaccine antibodies transferred to the women's infants are equally protective against both types isn't known. We will conduct a study to determine whether the maternal vaccine is equally effective against RSV-A and RSV-B.
The World Health Organization has recommended that all babies should be protected against RSV infection, and maternal RSV vaccination is likely to be introduced in many countries across the world. If the vaccine is less effective against a particular RSV type, it will be important to monitor the circulation of that RSV type, and consider changing the vaccination so that it is more effective.
RSV infection causes a significant global burden of disease in babies, which could be reduced by maternal vaccination. Our study will find out whether the type of RSV that a baby is infected with makes it more or less likely for the maternal vaccine to work.
## Background
Respiratory syncytial virus (RSV) is an RNA virus with two antigenic types, RSV-A and RSV-B, that causes yearly seasonal epidemics in countries with temperate climates 1 . In the United Kingdom in a typical year, RSV cases in infants start to rise in October, peak in December, and then fall 2 ; RSV seasonality is similar in England 3 and Scotland 4 .
In the late summer of 2024, the United Kingdom, following the United States and Argentina 5 introduced maternal RSV vaccination, using a stabilised F (fusion) protein ("RSVpreF"; Abrysvo), into its routine immunisation schedule, with vaccination recommended as soon as possible after a gestation of 28 weeks 6 . At the start of the programme, which is year-round, the vaccine was offered to all pregnant women at a gestation of 28 weeks or more, with women eligible for vaccination until delivery 7 .
Estimates of maternal RSV vaccine effectiveness are now available from Argentina 8 and the UK 9 but whether vaccine effectiveness differs by RSV type remains unclear. The final analysis of the pivotal phase 3 MATISSE trial for RSVpreF 10 showed robust protection for RSV-B as an exploratory endpoint. However, this analysis was limited by the relatively small number of typed RSV samples for each group (25 severe RSV-A cases, 61 severe RSV-B cases) and it was not possible to assess protection against RSV-A cases owing to greater uncertainty and wide, overlapping confidence intervals.
RSV genomic variability has been shown to affect the binding of the monoclonal antibodies palivizumab 11 , suptavumab 12 and nirsevimab 13 , and mutations in the SARS-CoV-2 Spike protein (equivalent to the RSV F-protein) were also shown to affect the ability of neutralising, vaccine induced antibodies to bind to new variants of the virus 14 . Therefore, there is the theoretical possibility that either existing or newly emerging circulating RSV genomic variability, could impact on the effectiveness of a vaccine that depends on passive, antibody mediated-albeit polyclonal-immunity.
The World Health Organization has recently recommended that interventions to prevent RSV disease (infant passive immunisation or maternal vaccination) should be introduced to protect all infants globally 15 . Therefore, the number of vaccinated mothers is likely to increase markedly in the next few years. In this analysis, we will examine whether RSV type or lineage within each type is associated with reduced maternal RSVpreF vaccine effectiveness in infants.
## Study protocol
This protocol is structured in keeping with the principles of the STROBE statement.
## Study design
We will conduct a case control study, with samples positive for each RSV type acting as cases/controls respectively, to assess the relative effectiveness of maternal RSVpreF vaccination against hospitalisation for RSV-associated acute lower respiratory tract infection (ALRI) amongst infants born to vaccineeligible pregnant mothers. We will compare RSV types using a direct comparison to give a relative vaccine effectiveness (rVE). Sample size calculations were based on preliminary data from the BronchStop study 9 , looking at coverage in the group of mother/infant pairs where the mother was vaccinated >14 days ("fully vaccinated", see "Exposure" section below) prior to delivery (Table 1).
Table 2 shows the estimated sample size needed for 80% power, with 95% confidence intervals, to detect the difference in effectiveness by type (calculations performed in GLIM4 16 and Microsoft Excel) To detect a difference of 25% between VE for RSV-A and RSV-B, assuming a 2:1 ratio of RSV-A to RSV-B cases, and vaccine coverage of 34%, an estimated 425 RSV positive samples would need to be sequenced (highlighted row in Table 2), which was compatible with sampling from the participating schemes (HARISS/BronchStop/PHS surveillance, for details on these see below).
Additionally, a test-negative design looking at the VE separately for RSV-A and RSV-B cases, compared with test-negative controls, will be performed as a secondary analysis. For each RSV-A and RSV-B case an RSV-negative control infant, matched by admission date (epidemiological week) and age (in weeks), will be identified.
## Study population
Eligible infants will be born after August 12, 2024 (Scotland) or September 1, 2024 (England) (the same dates as the commencement of RSV maternal vaccination in the two countries) and admitted to hospital. The upper age limit for infants recruited will be 6 months, as per the age cut-off used in the MATISSE study 17 . Samples will be from infants who have been admitted to hospital, had a positive rRT-PCR test for RSV, and been assigned a primary or secondary diagnosis consistent with an ALRI, which encompasses the likely most common diagnosis of bronchiolitis (J21*; Table 3). To be included, the RSV test must be positive in the 14 days leading up to hospital admission, or 2 days after the day of admission. Samples will come from the BronchStop 18 or HARISS 19 (England) partnerships, or as part of routine public health surveillance in Scotland. Public Health Scotland will coordinate the sequencing of Scottish samples, and UKHSA will conduct sequencing of samples from across England. Test-negative controls, also with a diagnosis of ALRI but no RSV positive test, will be identified from the hospitals that admitted cases.
## Definition
for the 6-week run-in period is to allow 4 weeks for the implementation of catch-up maternal vaccination, and a further 2 weeks to allow the transfer of maternal immunity to the fetus 20 .
A fall in the proportion of vaccinated mothers to below 12% over the period of the study would mean that the case control study was no longer powered to examine the primary outcome; these inclusion criteria maximise the likelihood of including the infants of fully vaccinated mothers only (see "Exposure" section below).
## Cases and controls
A case will be an infant hospitalised with RSV-associated ALRI.
Cases are defined as a positive PCR test for RSV A or B, and a diagnosis consistent with an ALRI (see Table 3) on hospital admission identified from hospital records. Controls have the same definition as the case, but for the other RSV type (re. RSV A for RSV B). Additionally, a test-negative design looking at the VE for RSV-A and RSV-B cases, compared with testnegative controls, will be performed as a secondary analysis.
In this analysis RSV test-negative participants will be identified from admissions to the same hospital as RSV-positive cases, matching on admission date and age at admission.
## Exposure
The treatment exposure will be maternal RSVpreF receipt status prior to birth amongst both case and control patients. Vaccination status:
• Fully vaccinated: An infant will be considered fully vaccinated if the birth mother had received a dose of the RSV vaccine a least 14 days prior to giving birth.
• Partially vaccinated: RSV vaccine received within 14 days of giving birth; these samples will not be included in the analysis.
• Unvaccinated: No RSV vaccine received during the pregnancy
## Covariables
For the infant, data will be collected on date of birth, sex, location of hospital admission, recruitment scheme (HARISS/ BronchStop/PHS) and date of hospital admission and discharge.
For the infants' mother, linked data will be collected on maternal RSVpreF vaccination status (yes/no), date of vaccination if given, date of delivery, gestation at delivery, Index of Multiple Deprivation (IMD; England) and Scottish IMD (SIMD), age and ethnicity, and immunosuppressed status. This could be important in the ability to transfer transplacental antibodies, and these mother/infant pairs may need to be excluded from the analysis. Immunosuppressed mothers will be identified by selecting cohorts eligible for COVID-19 Spring vaccinations as defined in tables 3 and 4 of the COVID-19 chapter of the Green Book 21 .
## Sequencing
Cycle threshold (Ct) values for samples will be calculated using rRT-PCR prior to sequencing. Sequencing of RSV positive samples will be undertaken using the Talts et al. protocol 22 at UKHSA, and by either the Dong et al. 23 or Maloney et al. 24 protocols in Scotland. Sequences will be allocated to a lineage using the Goya et al. classification 25 .
## Datasets
RSV positive samples will be provided by the HARISS and BronchStop studies (England), and routine public health surveillance in Scotland. For all participants in England, the infant unique patient identifier (NHS number) will be linked to the Maternity Services Dataset (MSDS) to gather complete information on date of birth, gestational age at birth and identify the mother. For the test-negative design, infants matched by age (in weeks) and admission date (epidemiological week) with a negative RSV rRT-PCR test will be identified from the same hospitals cases were admitted to. In England, the NHS numbers of participant's mothers will be linked to retrieve their RSV vaccine status and other demographic information such as age, IMD, ethnicity, geographical location (IIS) and severely immunosuppressed status (CaaS). For Scotland, the Scottish Linked Pregnancy and Baby Dataset (SLiPBD) will be accessed, using the infant unique Community Health Index (CHI) number, to identify the same information.
## Statistical analysis
First, we will conduct a descriptive analysis reporting all covariables, exposure and demographic information from the mother by RSV-A/RSV-B status of the infant, and RSVpreF vaccination status of the mother. We will compare RSV Ct values for unvaccinated versus fully vaccinated infants, comparing the mean Ct values in each group using a two-sample Students t-test. We will also compare RSV type distributions for the unvaccinated versus fully vaccinated cohort, using a Fisher's exact test to look for evidence of over/under-representation of RSV-A and RSV-B. We will use a Wilcoxon rank sum test to look for over/under-representation of different RSV lineages samples from the fully vaccinated vs unvaccinated infants.
The maternal RSVpreF vaccine effectiveness against RSVassociated hospitalisation in infants for RSV-A vs RSV-B will be estimated using logistic regression. This will be calculated using the following equation: relative vaccine effectiveness = 100% x (1-incidence rate ratio [IRR]), where IRR denotes the incidence rate ratio for one group (RSV-A) versus the other (RSV-B).
The analysis will adjust for the potential confounders of age in months and admission date (as a spline). The following variables will be included if they lead to a change in the VE of more than 3%: preterm birth, sex, maternal ethnicity, hospital location (both individual site and country) and recruitment scheme. Protection is known to be higher for fully vaccinated infants 9 ; therefore, rVE will be calculated only including fully vaccinated and unvaccinated infants, as per the sample size calculations in Table 1.
The VE for the test-negative design will be calculated using the following equation: effectiveness = 100% x (1-adjusted odds ratio). The analysis will adjust for potential confounders, as outlined above. We will conduct a sensitivity analysis comparing the rVE/VE for protection against lower respiratory tract infection to that for any acute respiratory infection (Table 3).
## Ethical issues
The Impact Assessment (DPIA) DP24250011. In summary, for all three sources of samples (BronchStop, HARISS and PHS) informed consent for data collection and viral sample processing was waived as per the ethical and public health permissions outlined above, for the purposes of this public health analysis.
## Data input, storage and management
Consented data for the BronchStop study for samples from included English infants has been entered using the validated online data entry software REDCap. This software (REDCap) is hosted on the University Hospitals Bristol and Weston NHS Foundation Trust (UHBW) secure server, accessible on the Health and Social Care Network (HSCN) that is managed by NHS Digital. Data for the included infants for HARISS was collected at participating sites and shared as per the HARISS protocol using encrypted, password-protected line lists. Infant date of birth and hospital number from BronchStop and HARISS will be provided to UKHSA for linkage to the maternal dataset. Data on samples from infants admitted to hospital in Scotland is shared by NHS Health Boards with Public Health Scotland for linkage with the SLiPB dataset.
## Dissemination of results
Initial rVE estimates will be shared within UKHSA and Public Health Scotland. They will be uploaded to a preprint server (e.g. medRxiv or SSRN) and submitted for publication in a peer-reviewed journal. Results will also be disseminated via national reports by the public health agencies.
## Conclusion
The number of infants who benefit from maternal RSV vaccination is likely to rise soon. An understanding of whether RSV type potentially affects vaccine effectiveness in infants will be of interest to clinicians, public health officials and policy makers globally. A null result showing no evidence of RSV type on vaccine effectiveness will reassure stakeholders and act as a baseline for future studies; conversely, an association between RSV type or lineage and decreased vaccine effectiveness will highlight the need for the ongoing comprehensive national and global molecular surveillance of RSV.
## Michael G Ison
National Institute of Health, Rockville, MD, USA
This outlines a study protocol to assess the vaccine effectiveness of maternal RSV vaccination. Standard case-control methods and WGS will be utilized. The information will inform VE and may provide insight into drift if there are genetic changes determined throughout the study. Sample size is justified. Ethical issues and data management and dissemination are clearly outlined.
Is the rationale for, and objectives of, the study clearly described? Yes
## Is the study design appropriate for the research question? Yes
Are sufficient details of the methods provided to allow replication by others? Yes
## Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
## Reviewer Expertise: respiratory viral infections
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
Reviewer Report 29 December 2025
## References
1. Meissner (2016) "Viral bronchiolitis in children" *N Engl J Med*
2. Jcvi (2023) "Respiratory Syncytial Virus (RSV) immunisation programme for infants and older adults: JCVI full statement" *GOV.UK*
3. (2025) "UKHSA: National flu and COVID-19 surveillance report" *GOV.UK*
4. (2025) "Public Health Scotland: Viral respiratory diseases (including influenza and COVID-19"
5. Pecenka, Sparrow, Feikin (2024) "Respiratory Syncytial Virus vaccination and immunoprophylaxis: realising the potential for protection of young children" *Lancet*
6. (2025) "Respiratory Syncytial Virus: the green book, chapter 27a. The green book"
7. (2025) "United Kingdom Health Security Agency: RSV vaccination of pregnant women for infant protection: information for healthcare practitioners" *UKHSA Guidance*
8. Marc, Vizzotti, Fell (2025) "Real-world effectiveness of RSVpreF vaccination during pregnancy against RSV-associated lower respiratory tract disease leading to hospitalisation in infants during the 2024 RSV season in Argentina (BERNI study): a multicentre, retrospective, testnegative, case-control study" *Lancet Infect Dis*
9. Williams, Marlow, Cunningham (2025) "Bivalent prefusion F vaccination in pregnancy and Respiratory Syncytial Virus hospitalisation in infants in the UK: results of a multicentre, test-negative, case-control study" *Lancet Child Adolesc Health*
10. Simões, Pahud, Madhi (2025) "Efficacy, safety, and immunogenicity of the MATISSE (Maternal Immunization Study for Safety and Efficacy) maternal Respiratory Syncytial Virus prefusion F protein vaccine trial" *Obstet Gynecol*
12. Mas, Nair, Campbell (2018) "Antigenic and sequence variability of the human Respiratory Syncytial Virus F glycoprotein compared to related viruses in a comprehensive dataset" *Vaccine*
13. Simões, Forleo-Neto, Geba (2021) "Suptavumab for the prevention of Medically Attended Respiratory Syncytial Virus infection in preterm infants" *Clin Infect Dis*
14. Fourati, Reslan, Bourret (2025) "Genotypic and phenotypic characterisation of Respiratory Syncytial Virus after nirsevimab breakthrough infections: a large, multicentre, observational, real-world study" *Lancet Infect Dis*
15. Williams, Burgers (2021) "PubMed Abstract | Publisher Full Text | Free Full Text 15. World Health Organization: WHO position paper on immunization to protect infants against Respiratory Syncytial Virus disease" *Wkly Epidemiol Rec*
16. Francis, Green, Payne (1993) "The GLIM system: the statistical system for generalized linear interactive modelling"
17. Kampmann, Madhi, Munjal (2023) "Bivalent prefusion F Vaccine in Pregnancy to Prevent RSV Illness in Infants" *N Engl J Med*
18. Williams, Cunningham, Drysdale "Update to: study pre-protocol for 'BronchStart -The impact of the COVID-19 Pandemic on the timing, age and severity of Respiratory Syncytial Virus (RSV) emergency presentations; a multi-centre prospective observational cohort study"
19. (2024) "PubMed Abstract | Publisher Full Text | Free Full Text"
20. (2024) "UKHSA: Hospital-based Acute Respiratory Infection Sentinel Surveillance (HARISS) system"
21. Jasset, Zapana, Bahadir (2025) "Enhanced placental antibody transfer efficiency with longer interval between maternal Respiratory Syncytial Virus vaccination and birth" *Am J Obstet Gynecol*
22. Ukhsa (2025) "COVID-19: the green book"
23. Talts, Mosscrop, Williams (2024) "Robust and sensitive amplicon-based whole-genome sequencing assay of Respiratory Syncytial Virus subtype A and B. Microbiol Spectr" *PubMed Abstract | Publisher Full Text | Free Full Text*
24. Dong, Deng, Aziz (2023) "A simplified, amplicon-based method for whole genome sequencing of human Respiratory Syncytial Viruses" *J Clin Virol*
25. Maloney, Fernandes, Jasim "ARTIC RSV amplicon sequencing reveals global RSV genotype dynamics"
26. (2025) *Wellcome Open Res*
27. Goya, Ruis, Neher (2024) "Standardized phylogenetic classification of human Respiratory Syncytial Virus below the subgroup level" *Emerg Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817899&blobtype=pdf | # Bovine viral diarrhea virus suppresses type I IFN production by inducing MAVS degradation via autophagy mediated by the ROS-endoplasmic reticulum stress axis
Jing Wang, Jiangfei Zhou, Yixin Wang, Wenlu Fan, Xinyue Xia, Jiarui Chen, Haiyue Zhu, Qianyao Wang, Xiao Li, Yimei Liu, Jiayi Xiang, Han Yu, Moxuan Mao, Renjie Xu, Jiacun Liu, Shuo Jia, Yuan Li, Yigang Xu
## Abstract
Bovine viral diarrhea virus (BVDV) is a major animal pathogen with a broad host range, causing gastrointestinal, respiratory, and reproductive diseases in cattle worldwide. BVDV exists as two biotypes: cytopathic (cp) and non-cytopathic (ncp). Although both cpBVDV and ncpBVDV have developed sophisticated strategies to evade or subvert host antiviral innate immune response, the underlying mecha nisms remain incompletely understood. Autophagy, a process essential for maintain ing cellular homeostasis, plays an important role in regulating viral replication and antiviral immunity. In this study, we demonstrated that the induction of autophagy with rapamycin enhanced the production of infectious progeny for both cpBVDV and ncpBVDV, whereas pharmacological inhibition of autophagy with 3-MA reduced viral yields. We further showed that modulating autophagy significantly influenced the early stages of the viral life cycle and the production of type I IFN (IFN-I). Notably, overex pression of BECN1 suppressed the synthesis of IFN-α and IFN-β, thereby promoting the replication of both cpBVDV and ncpBVDV. Conversely, RNA interference-mediated knockdown of BECN1 potentiated the antiviral innate immune response and restric ted viral replication. Mechanistically, BECN1 was found to inhibit RIG-I-MAVS pathway activation by promoting ubiquitination and subsequent degradation of mitochondrial antiviral signaling (MAVS) protein, leading to suppression of IFN-I production. Addition ally, both cpBVDV and ncpBVDV were shown to induce autophagy via the ROS-endoplas mic reticulum stress axis. These findings deepen our understanding of how BVDV evades host immunity and may inform the development of preventive strategies against BVDV infection.IMPORTANCE Bovine viral diarrhea virus (BVDV), the causative agent of bovine viral diarrhea-mucosal disease, is a major global threat to cattle health. BVDV employs sophisticated strategies to evade host defense and facilitate its replication. Understand ing these mechanisms is crucial for developing effective vaccines and antiviral agents. Our study elucidates how cytopathic BVDV and non-cytopathic BVDV subvert the host's antiviral innate immune response by exploiting autophagy to inhibit the RIG-I-MAVS pathway. A key finding is that BECN1-mediated autophagy directly targets MAVS protein for degradation via a specific BECN1 and MAVS interaction. Furthermore, we demonstrate that BVDV activates autophagy through ROS-ER stress axis to promote its replication. These insights reveal a novel immune evasion mechanism of BVDV and highlight the therapeutic potential of autophagy inhibition in treating BVDV-related diseases.KEYWORDS bovine viral diarrhea virus (BVDV), ROS-ER stress-autophagy axis, mitochondrial antiviral signaling protein (MAVS), type I interferon
B ovine viral diarrhea-mucosal disease (BVD-MD), caused by bovine viral diarrhea virus (BVDV), is an economically significant contagious disease in cattle, characterized clinically by diarrhea, fever, mucosal erosion, persistent infection (PI), and immunosup pression. BVDV ranks among the most prevalent and destructive pathogens in cattle herds worldwide, leading to substantial economic losses in the global livestock industry (1,2). BVDV is categorized into two biotypes based on their cytopathic effects in cell culture: cytopathogenic (cp) BVDV (cpBVDV) and non-cytopathogenic (ncp) BVDV (ncpBVDV). The ncp biotype, which is the most frequently isolated from clinical cases, is primarily responsible for establishing PI. When cattle with PI are superinfected with antigenically related or homologous cpBVDV strain, they may develop fatal mucosal disease (3). A hallmark molecular distinction between the ncpBVDV and cpBVDV is the expression of the non-structural protein NS3 (also known as p80). This difference often arises during PIs, where ncpBVDV can acquire mutations in the NS2-3 genomic region, giving rise to cpBVDV variants that may trigger lethal disease. At the protein level, ncpBVDV produces primarily the uncleaved NS2-3 polyprotein (p125), whereas cpBVDV expresses both the full-length NS2-3 and significant amounts of the separate NS3 protein. In ncpBVDV, the NS2-3 region remains intact, preventing efficient cleavage. In contrast, cpBVDV frequently harbors insertions of cellular sequences within NS2-3, which create novel cleavage sites and lead to the constitutive processing of the polyprotein into discrete NS2 and NS3 proteins (4)(5)(6). Throughout prolonged virus-host co-evolution, both cpBVDV and ncpBVDV have evolved sophisticated strategies to evade or counteract the host's antiviral innate immune response (7). Nevertheless, the precise molecular mechanisms underlying this immune subversion remain largely unexplored.
Cellular autophagy is a highly conserved intracellular process that degrades and recycles cellular components, playing an essential role in maintaining cellular homeo stasis and modulating immune defenses. Autophagy is activated under various stress conditions, such as nutrient deprivation, reactive oxygen species (ROS) accumulation, endoplasmic reticulum (ER) stress, and pathogen infection, through the formation and lysosomal fusion of autophagosome (8). Accumulating evidence indicates that autophagy can eliminate invading viruses and support cell survival by facilitating nutrient recycling. However, several viruses have evolved strategies to hijack or suppress autophagic machinery to enhance viral replication and infectivity. Previous studies have demonstrated that BVDV infection exploits the autophagy pathway to promote viral propagation in host cells (7,9). It is well established that cellular stress resulting from viral replication often acts as a trigger for autophagy induction (10,11). Although BVDV infection has been reported to induce both ER stress and oxidative stress (12,13), the precise mechanisms linking BVDV-induced stress to autophagy regulation remain poorly defined.
As the first line of defense against viral infection, interferons (IFNs) play a pivotal role in host antiviral immunity. The mitochondrial antiviral signaling (MAVS) protein functions as a critical switch in the innate immune signaling cascade triggered by most RNA viruses. Upon viral RNA recognition, retinoic acid-inducible gene I (RIG-I) and melanoma differentiation-associated protein 5 (MDA5) activate MAVS, leading to the formation of functional prion-like aggregates that subsequently promote the phosphor ylation and nuclear translocation of IFN regulatory factor 3 (IRF3), thereby initiating IFN-mediated antiviral response (14). In previous work, we demonstrated that BVDV infection upregulates lncRNA-cylindromatosis (CYLD), which functions as a competing endogenous RNA (ceRNA) for miR-2383. This interaction enhances the expression of CYLD, thereby suppressing RIG-I-mediated type I IFN (IFN-α/β) production and facilitat ing BVDV replication (15). However, the molecular mechanisms by which BVDV evades host innate immunity are still not fully elucidated.
The interplay between autophagy and IFN response represents an important mechanism in virus-host interactions. Although autophagy can function as a cell-auton omous defense against viral infection, numerous viruses have evolved the ability to exploit the autophagic machinery to subvert host innate immunity. Accumulating evidence indicates that certain viruses activate autophagy to evade antiviral IFN signaling (16)(17)(18). A detailed understanding of the interaction between BVDV and host cellular pathways is therefore essential to elucidate the underlying mechanisms. In this study, we systematically investigated how autophagy induced by both biotypes of BVDV influences viral replication. We found that cpBVDV and ncpBVDV trigger autophagy via ROS-ER stress axis, which in turn suppresses RIG-I-MAVS pathway-mediated IFN-I production. This process was further associated with a functional interaction between MAVS and BECN1. These findings deepen our understanding of how BVDV evades innate immune defenses and offer valuable insights for the development of targeted antiviral strategies.
## RESULTS
## BVDV induces the formation of autophagosome and autophagic flux
The genomic and proteomic organizations of cpBVDV and ncpBVDV are schematically illustrated in Fig. 1A. Infection of BT cells with cpBVDV (multiplicity of infection [MOI] = 1, Fig. 1B) and ncpBVDV (MOI = 5, Fig. 2A) at the indicated time points was confirmed by immunofluorescence assay (IFA) and Western blot. Transmission electron microscopy (TEM) revealed the presence of autophagic vesicles in cpBVDV-and ncpBVDV-infected BT cells, exhibiting characteristic double-membrane structures measuring 500-1,000 nm in diameter (Fig. 1C). Similar structures were observed in rapamycin (Rapa)-treated positive control cells, displaying typical autophagic features with clearly discernible contents. In contrast, no significant autophagic vesicle-like structures were detected in mock-infected cells. To further validate the integrity of the autophagic process induced by cpBVDV/ ncpBVDV-encompassing autophagosome formation, lysosomal fusion, and autolyso some maturation-we monitored autophagic flux using confocal microscopy in BT cells co-infected with cpBVDV/ncpBVDV and the adenoviral reporter Ad-mCherry-GFP-LC3.
In this system, the acidic environment of autolysosomes quenched GFP fluorescence, while mRFP signal remained stable. Thus, autophagosomes appear as yellow puncta (GFP + mRFP + ), whereas autolysosomes are identified as red puncta (GFP -mRFP + ). As shown in Fig. 1D and 2D, infection with either cpBVDV or ncpBVDV induces yellow puncta formation, indicating autophagosome formation. The subsequent appearance of red-only puncta confirmed autophagosome-lysosome fusion and demonstrated that both biotypes of BVDV enhance autophagic flux. At the transcriptional level, cpBVDV infection significantly upregulated the mRNA expression of LC3B and BECN1 (Fig. 1E). Consistent with this, Western blot analysis showed increased protein levels of BECN1 and lipidated LC3B-II upon cpBVDV infection (Fig. 1F). A concurrent decrease in phosphoryla ted mammalian target of Rapa (p-mTOR) was observed at different time points postinfection (Fig. 1F). Similar trends were also seen in ncpBVDV-infected BT cells (Fig. 2B), suggesting that both biotypes induce autophagy via suppression of mTOR activation. Autophagic flux, reflecting the rate of autophagic degradation, was quantified by measuring LC3B-II accumulation in the presence of the lysosomal inhibitor chloroquine (CQ). Western blot analysis showed significant increases in LC3B-II levels in both cpBVDVand ncpBVDV-infected BT cells after treatment with CQ (Fig. 1G and2C), at working concentration of 0.075 mmol/L (Fig. 1H), indicating that BVDV infection promotes not only autophagosome formation but also their subsequent degradation.
## BVDV-induced autophagy inhibits IFN-I production
The interplay between cellular autophagy and viral infection is complex, with playing contradictory roles as either a host defense mechanism or a proviral factor. In this study, we investigated how autophagy influences BVDV infection using Rapa, an autophagy inducer by inhibiting the mTOR pathway (19), and 3-methyladenine (3-MA), an autoph agy inhibitor by inhibiting the PI3K class III protein (20). Viral replication (Fig. 3A andD) and mRNA expression of IFN-α (Fig. 3B andE) and IFN-β (Fig. 3C andF) were meas ured in BT cells at different time points after cpBVDV or ncpBVDV infection. At the concentrations used-10 µM for Rapa and 0.325 mM for 3-MA, as determined by the CCK-8 assay (Fig. 3G andH), Rapa significantly increased BECN1 mRNA expression, while 3-MA suppressed it in BT cells infected with cpBVDV or ncpBVDV (Fig. 3I). Furthermore, we observed that Rapa significantly enhanced cpBVDV and ncpBVDV replication, an opposite effect exerted by 3-MA (Fig. 3J). Growing evidence suggests that virus-induced autophagy can modulate antiviral immune responses (21)(22)(23). Here, Rapa downregula ted both mRNA (Fig. 3K andL) and protein levels (Fig. 3M) of IFN-α/β in BVDV-infected BT cells. Conversely, 3-MA enhanced IFN-α/β expression. We also pretreated BT cells with poly(I:C) and examined IFN-β production with or without BVDV infection in the presence of Rapa or 3-MA. As shown in Fig. 3N, IFN-α mRNA levels were significantly higher in the poly(I:C)+ BVDV + 3 MA group compared to poly(I:C)+BVDV group but markedly lower in the poly(I:C)+BVDV + Rapa group. These findings collectively demonstrate that autophagy benefits BVDV replication, consistent with previous report (24).
## BECN1 promotes BVDV replication by inhibiting type I IFN production
BECN1, a key regulator of autophagy, is involved in multiple stages of autophagosome formation, maturation, and transport. Previous studies have shown that ubiquitin-spe cific protease 19 suppresses IFN-I production by deubiquitinating BECN1, thereby disrupting RIG-I-MAVS interaction (25). Additionally, classical swine fever virus (CSFV)induced autophagy has been reported to inhibit IFN-I signaling via BECN1-MAVS association, facilitating viral replication. Building on these findings and our own data, we hypothesize that BECN1 may contribute to the suppression of IFN-I production during BVDV-induced autophagy. To test this hypothesis, we constructed a plasmid p3×-Flag-BECN1 (p3×-BECN1) for exogenous BECN1 expression. Successful overexpression was confirmed at both mRNA (Fig. 4A) and protein levels (Fig. 4B). We found that BECN1 overexpression significantly enhanced cpBVDV replication (Fig. 4C) and viral protein expression (Fig. 4D). We also designed small interfering RNA (siRNA) targeting BECN1 (siBECN1), among which siBECN1 #3 showed the highest knockdown efficiency (Fig. 4E). Knockdown of BECN1 in BVDV-infected BT cells (Fig. 4F) led to markedly reduced viral replication (Fig. 4G) and viral protein expression (Fig. 4H), supporting the notion that BECN1 promotes efficient viral replication. Furthermore, BECN1 overexpression significantly suppressed the production of IFN-α (Fig. 4I) and IFN-β (Fig. 4J) in BVDV-infec ted BT cells compared with infection alone (P < 0.01). Conversely, BECN1 knockdown enhanced IFN-α and IFN-β production (Fig. 4K). Taken together, these results suggest that the autophagy-related protein BECN1 is involved in the inhibition of IFN-I production during BVDV infection.
## BVDV-induced autophagy suppressing IFN-I production is related to the BECN1 and MAVS interaction
RIG-I acts as a primary sensor for RNA viruses by recognizing double-stranded RNA (dsRNA), 5′-triphosphorylated RNAs, and other pathogen-associated molecular patterns. This recognition triggers downstream signaling via MAVS, leading to IFN-I production.
To elucidate whether BVDV-induced upregulation of BECN1 influences the RIG-I-like receptor (RLR) signaling pathway, we examined the expression of key proteins in the RIG-I pathway in BT cells with either overexpression or knockdown of BECN1 after cp/ ncpBVDV infection. As shown in Fig. 5A andB, RIG-I protein levels remained unchanged (P > 0.05) in cp/ncpBVDV-infected BT cells regardless of BECN1 overexpression. In contrast, MAVS protein expression and IRF3 phosphorylation (pIRF3) were significantly reduced (P < 0.001 and P < 0.01, respectively) in BECN1-overexpressing cells infec ted with cp/ncpBVDV. At the mRNA level, BECN1 overexpression did not affect RIG-I expression (P > 0.05) but significantly downregulated IRF3, IFN-α, and IFN-β transcripts (P < 0.01) and upregulated MAVS expression (P < 0.01) (Fig. 5C). These results suggest that BECN1 promotes MAVS protein degradation without altering its transcription. Following siRNA-mediated BECN1 knockdown, we observed a substantial increase in MAVS and pIRF3 levels (P < 0.001), along with significant suppression of viral replica tion in both cpBVDV-(MOI = 1, Fig. 5D) and ncpBVDV-infected (MOI = 5, Fig. 5E) BT cells. Moreover, IFN-α and IFN-β mRNA levels were significantly elevated in BECN1knockdown BT cells after BVDV infection (Fig. 5F), indicating that BECN1 knockdown enhances RLR signaling and antiviral responses. To explore the mechanism of BVDVinduced MAVS degradation, BECN1-expressing BT cells infected with cpBVDV (MOI = 1, 24 h) were treated with DMSO, CQ (0.075 mmol/L, a lysosomal inhibitor), or MG-132 (50 µmol/L, a proteasome inhibitor, Fig. 5G) for 4 h. Western blot analysis revealed that CQ treatment significantly restored MAVS protein levels (Fig. 5H andI). Additionally, we found that MG-132 treatment also contributed to the inhibition of MAVS degradation. These findings suggest that both the autophagic-lysosomal and proteasomal pathways contribute to MAVS degradation, with the former playing a dominant role during BVDV infection. Furthermore, co-immunoprecipitation (Co-IP) assay confirmed an interaction between BECN1 and MAVS (Fig. 5J), and BECN1 was found to promote MAVS ubiquitina tion and subsequent degradation (Fig. 5K).
## BVDV-induced ER stress triggers autophagy
The ER, a key organelle in eukaryotic cells, undergoes morphological changes under ES stress conditions. In this study, TEM revealed that the ER in cpBVDV-infected BT cells frequently displayed swelling and rupture, features absent in mock-infected cells (Fig. 6A). To further investigate ER stress induced by BVDV infection, we evaluated the expression of GRP78 (glucose-regulated protein, 78 kDa), an ER-resident chaperone and central regulator of the unfolded protein response (UPR) under both physiological and stress conditions, at different time points after cpBVDV infection. GRP78 protein levels were markedly elevated in infected BT cells (Fig. 6B). Moreover, quantitative real-time PCR (qRT-PCR) analysis showed significant upregulation of GRP94 mRNA, another major ER chaperone, consistent with the trend observed for GRP78 mRNA (Fig. 6C). These results indicate that cpBVDV infection triggers ER stress.
The UPR is a conserved signaling network that aims to restore ER homeostasis upon stress and is mediated by three ER transmembrane sensors: protein kinase R-like ER kinase (PERK), activating transcription factor 6 (ATF6), and inositol-requiring enzyme 1 (IRE1) (26). Under ER stress, ATF6 translocates to the Golgi apparatus, where it is cleaved to release its N-terminus, which acts as a transcription factor in the nucleus (27,28). PERK and IRE1 are activated through oligomerization and trans-autophosphorylation (29). Phosphorylated PERK (p-PERK) phosphorylates eukaryotic translation initiation factor 2α (EIF2α), attenuating global protein synthesis and promoting expression of activating transcription factor 4 (ATF4). Activated IRE1, possessing endonuclease activity, excises a 26-nucleotide intron from X-box binding protein-1 (XBP1u) mRNA, generating the spliced form XBP1s (30,31). In cpBVDV-infected BT cells, we detected significant increases in p-PERK and phosphorylated EIF2α (p-EIF2α) levels (Fig. 6D). Concurrently, expression of ATF4 and GADD34, downstream effectors of the PERK pathway, was markedly upregulated (Fig. 6E). Pretreatment of BT cells with the ER stress inhibitor 4-phenylbutyric acid (4-PBA; 1.0 mmol/L, Fig. 6F) prior to cpBVDV infection (MOI = 1) significantly suppressed BECN1 mRNA expression compared to infection alone or infection with DMSO control (P < 0.001; Fig. 6G). Notably, 4-PBA treatment also reduced cpBVDV replication (P < 0.001; Fig. 6H) and enhanced IFN-I (IFN-α/IFN-β) production (P < 0.01; Fig. 6I). Furthermore, 4-PBA treatment strongly inhibited GRP78 expression and EIF2α phosphorylation (P < 0.0001; Fig. 6J), confirming effective suppression of cpBVDV-induced ER stress. Western bolt analysis also revealed upregulation of p-mTOR (P < 0.001), downregulation of BECN1 protein (P < 0.0001), and increased MAVS protein levels (P < 0.01) (Fig. 6J), suggesting that ER stress inhibition attenuates cpBVDV-induced autophagy, thereby preventing MAVS degradation and restraining viral replication. We also evaluated the activation status of the other two UPR branches. No significant changes were observed in IRE1 protein levels or phosphorylation, or in the ratio of spliced to unspliced Xbp1 (Xbp1s/Xbp1u), in cpBVDV-infected BT cells relative to mock-infected controls (Fig. 6K). Similarly, mRNA expression of EDEM1, an IRE1 down stream target, remained unchanged (P > 0.05; Fig. 6L). For the ATF6 pathway, no significant alterations were detected in full-length or cleaved ATF6 protein levels (Fig. 6M), or in the mRNA expression of its downstream targets calreticulin and calnexin proteins, which are downstream regulators of ATF6 (P > 0.05; Fig. 6N). Comparable results were observed in ncpBVDV-infected BT cells (Fig. 7A through C). Taken together, these data indicate that both biotypes of BVDV infection activate ER stress predominantly through the PERK signaling pathway.
## ROS-ER stress-autophagy axis benefits BVDV replication via promoting MAVS degradation
ROS function as crucial signaling molecules in regulating diverse physiological and pathophysiological processes, including oxidative stress and apoptosis. Previous studies have reported that cpBVDV significantly induces ROS production, while ncpBVDV does not (32). However, other evidence suggests that ncpBVDV infection can also lead to ROS accumulation (33). In the present study, we observed a marked upregulation in mRNA expression of oxidative stress-related genes HMOX-1, TXN, and PRDX-6 in cpBVDV-infected BT cells (Fig. 8A), indicating the induction of oxidative stress follow ing cpBVDV infection. Using the DCFH-DA probe combined with flow cytometry, we assessed ROS accumulation in cpBVDV-infected BT cells (Fig. 8B). Fluorescence micro scopy further confirmed ROS accumulation in both cpBVDV-and ncpBVDV-infected BT cells at various time points post-infection (Fig. 8C). Pretreatment with the antioxi dant butylated hydroxyanisole (BHA; 50 µM, Fig. 8D) prior to cpBVDV infection (MOI = 1) significantly suppressed BECN1 expression (P < 0.001; Fig. 8E) and reduced viral replication (P < 0.01; Fig. 8F), while enhancing IFN-I (IFN-α/IFN-β) production (P < 0.01; Fig. 8G). Moreover, treatment with the ER stress inhibitor 4-PBA substantially attenuated cpBVDV-induced ROS accumulation (Fig. 8H), suggesting that alleviating ER stress effectively induces oxidative stress. Importantly, antioxidant treatment with BHA in cpBVDV-or ncpBVDV-infected BT cells significantly decreased protein levels of GRP78 (P < 0.01; P < 0.001), p-EIF2α (P < 0.01; P < 0.001), and BECN1 (P < 0.001; P < 0.01), while increasing p-mTOR and MAVS levels (P < 0.001; P < 0.01), compared to virus-only and virus-plus-DMSO groups (Fig. 8I andJ). These results imply that the ROS-ER stressautophagy axis facilitates BVDV replication by promoting MAVS degradation, thereby suppressing RIG-I-MAVS-mediated IFN-I production and contributing to viral immune evasion.
## DISCUSSION
Autophagy is a critical biological process in viral pathogenesis, known for its com plex and sometimes paradoxical roles in regulating inflammation, innate immunity, apoptosis, and cellular homeostasis (34). Our previous work showed that the expression of BVDV infection modulates key autophagy-related proteins, notably downregulat ing mTOR while upregulating ATG5 and BECN1. Concomitantly, we observed a significant suppression of proteins within the RIG-I-MAVS signaling pathway (35). Although autophagy generally supports immune system development and facilitates antimicrobial innate and adaptive responses, many intracellular pathogens have evolved strategies to evade or exploit this process to enhance their own survival and replication (36). However, the potential link between BVDV-induced autophagy and the suppression of IFN-I production has remained unclear. In this study, we therefore investigated the mecha nisms through which autophagy induced by both cpBVDV and ncpBVDV antagonizes the host's IFN-I-mediated antiviral innate immunity.
Our findings demonstrate that infection with either cpBVDV or ncpBVDV mark edly enhances cellular autophagic activity, as evidenced by significant upregulation of the autophagy-related proteins BECN1 and LC3, which promotes autophagosome formation, consistent with our recent report (37). The complete autophagy process involves autophagosome formation, fusion with lysosomes, autolysosome generation, and subsequent degradation. In contrast, incomplete autophagy results from impaired autophagic lysosomal degradation, leading to accumulation of autophagic vesicles (38). Utilizing RFP-GFP-LC3 labeling systems, we observed that both cpBVDV and ncpBVDV infection induce complete autophagic flux in cells, consistent with earlier reports (33,(39)(40)(41). This differs from hepatitis C virus (HCV), another Flaviviridae member, which induces incomplete autophagy by blocking autophagosome-lysosome fusion, thereby prevent ing viral particle degradation and facilitating viral replication (42). In this study, we established that autophagy promotes the replication of both BVDV biotypes; however, the molecular mechanisms and biological implications of BVDV-induced complete autophagic flux require further investigation. It is well-known that the mTOR complex-1 acts as a central regulatory of autophagy (43). Multiple cellular signals and stressors converge on the mTOR pathway to initiate autophagy through suppression of mTOR activity. Our data showed that phosphorylation of mTOR is significantly inhibited in cpBVDV-and ncpBVDV-infected cells. Nevertheless, the precise mechanisms by which BVDV infection regulates mTOR phosphorylation to induce autophagy remain to be elucidated.
BVDV and other members of the Flaviviridae family are ER-tropic viruses known to disrupt ER homeostasis upon infection. Since cellular stress plays a critical role in autophagy regulation (44) and prolonged ER stress has been shown to trigger autophagy through IRE1α, PERK, and ATF6 signaling pathways (45), we investigated the involvement of ER stress in BVDV-induced autophagy. In this study, expression levels of the key ER chaperones GRP78 and GRP94 were significantly upregulated in BVDV-infected cells, suggesting that cpBVDV and ncpBVDV infection induces ER stress primarily through the PERK pathway, rather than via IRE1 or ATF6. This observation aligns with previous reports demonstrating PERK-mediated unfolded protein response activation during BVDV infection (46,47). Furthermore, treatment with 4-PBA (an ER stress inhibitor) markedly attenuated autophagy activation induced by cpBVDV/ncpBVDV infection and significantly suppressed viral replication of both biotypes, indicating that ER stressmediated autophagy facilitates BVDV propagation. Notably, 4-PBA also reduced the accumulation of ROS in infected cells. ROS serve as important signaling molecules involved in various physiological and pathophysiological processes and have been linked to oxidative stress, ER stress, and autophagy (48,49). Previous studies indicate that cpBVDV-induced ROS promotes viral replication, whereas antioxidant treatment suppresses its replication (50). Although ncpBVDV has been reported not to induce ROS production in one study (32), our data show that it does elicit ROS production, consistent with findings by Li et al. (33). Thus, to explore the role of ROS in BVDV infection-induced ER stress-mediated autophagy-dependent viral replication is of considerable interest. In this work, we observed that antioxidant BHA treatment reduced BVDV-induced ROS accumulation, which in turn significantly inhibited both cpBVDV-and ncpBVDV-induced ER stress and autophagy, thereby suppressing viral replication.
There is growing evidence that autophagy contributes to viral pathogenesis (51). Although BVDV has evolved sophisticated strategies to evade or subvert host antivi ral innate immunity, the mechanisms by which BVDV-induced autophagy suppresses IFN-mediated immune responses remain incompletely understood. BECN1, a key regulator of autophagy, is indispensable for autophagosome formation, maturation, and transport. Recent studies have shown that CSFV, which belongs to the same Pestivirus genus as BVDV, activates autophagy to suppress RIG-I-MAVS pathway-dependent type I IFN production via BECN1 and MAVS interaction (52). Similarly, early infection with dengue virus (DENV), another member of the Flaviviridae family, induces the ATG5-ATG12 conjugate, which attenuates MAVS-mediated IFN-stimulated gene expression, thereby helping the virus evade antiviral response prior to IFN signaling activation (53). In this study, we examined how modulating BECN1 expression-through overexpression and siRNA knockdown-affects the RLR-MAVS pathway and antiviral immunity. We found that BECN1 promotes proteolytic degradation of MAVS without affecting its transcrip tion and disrupted RIG-I-MAVS interaction, thereby inhibiting RIG-I-mediated antiviral signaling and facilitating cp/ncpBVDV replication. Co-IP confirmed the BECN1-MAVS interaction, consistent with prior reports (52). These results align with findings by Li et al. (33) indicating that ncpBVDV induces mitophagy to suppress MAVS-driven innate immunity. We further demonstrated that BVDV promotes MAVS degradation partially through the ubiquitin-proteasome pathway, corroborating a previous study in which BVDV upregulates DNA-damage-inducible transcript 3 to enhance MAVS ubiquitination and degradation (54). MAVS degradation is a common viral immune evasion strategy. For example, the H1N1 nucleoprotein triggers mitophagy to degrade MAVS and dampen host innate immunity (55), and the hepatitis B virus X protein promotes MAVS ubiquiti nation at Lys136 to inhibit IFN-I production (56). Viruses can also cleave MAVS directly: proteases such as enterovirus 71 2Apro (57), coxsackievirus B 3C (58), HCV NS3-4A (59), and hepatitis A virus 3ABC (60) specifically cleave MAVS to suppress downstream signaling. MAVS function depends on its mitochondrial localization (61,62). Notably, our study revealed that BECN1 enhances MAVS ubiquitination, although the precise molecular mechanisms warrant further investigation.
In conclusion, this study preliminarily elucidates the mechanism by which cpBVDV/ ncpBVDV infection-induced autophagy selectively degrades MAVS via BECN1-MAVS interaction, thereby suppressing the RIG-I-MAVS signaling pathway and impairing antiviral innate immunity. We further demonstrate that ROS-ER stress-autophagy axis promotes BVDV replication. As summarized in Fig. 9, BVDV targets BECN1-mediated autophagic degradation of MAVS to inhibit IFN-I signaling. These findings provide new insights into the immune evasion strategies shared by both biotypes of BVDV. How ever, the detailed mechanisms through which BVDV-induced autophagy enhances viral replication remain to be fully uncovered. Additionally, it should be clarified that cpBVDV AV69 and ncpBVDV BJ175170 are not a matched virus pair but rather separate isolates. Although we lack a matched non-cytopathic counterpart for cpBVDV AV69, we consider it methodologically valid to conduct key experiments in parallel with ncpBVDV BJ175170. The concordant results obtained from these experiments further bolster the reliability of our conclusions.
## MATERIALS AND METHODS
## Virus and cells
The cpBVDV strain AV69 (GenBank: KC695814.1) and ncpBVDV strain BJ175170 (GenBank: PX445908.1) used in this study, both belonging to genotype 1, were kept in our laboratory. BT cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Tianhang, China). Both cpBVDV and ncpBVDV infection in BT cells was detected via IFA. In this work, to ensure com parable viral replication dynamics between cpBVDV AV69 and ncpBVDV BJ175170 in subsequent experiments, we characterized replication kinetics of cpBVDV AV69 and ncpBVDV BJ175170 in BT cells using IFA and Western blot. When infected at a MOI of 1 for cpBVDV AV69 and MOI of 5 for ncpBVDV BJ175170, the two viruses exhibited similar replication levels in BT cells (Fig. 1B and2A). Therefore, to minimize the potential influence of disparities in viral replication on host responses in follow-up experiments, cpBVDV at MOI = 1 and ncpBVDV at MOI = 5 were used to infect in all subsequent assays. This adjustment ensured that any observed differences in host responses could be attributed to strain-specific characteristics rather than variations in viral load.
## Determination of viral replication
The replication levels of cpBVDV and ncpBVDV were evaluated using different methods. For cpBVDV AV69, viral titers were determined by the 50% tissue culture infective dose (TCID 50 ) assay as previously described (59). Briefly, confluent BT cell monolayers in 96-well plates were inoculated with serial 10-fold dilutions of the virus (eight replicates per dilution) for 2 h at 37°C; DMEM alone served as a mock control. After viral adsorption, cells were washed three times with phosphate-buffered saline (PBS) and maintained in DMEM with 2% FBS for 3-5 days. Cytopathic effect was monitored daily, and the TCID 50 was calculated using the Reed-Muench method. For ncpBVDV BJ175170, replication level was determined by absolute quantification of viral RNA copies at various time points post-infection. Briefly, total RNA was extracted from infected BT cells using TRIzol reagent at the indicated time points and reverse-transcribed into cDNA. A standard curve was generated from a serially diluted plasmid containing BVDV 5′UTR to enable absolute quantification. Quantitative PCR (qPCR) was performed using Taq Pro Universal SYBR qPCR Master Mix (Vazyme, China) on an Applied Biosystems 7500 RT-PCR system.
## Transmission electron microscopy
Confluent monolayers of BT cells (80%-90% confluency in six-well plates) were infected with cpBVDV (MOI = 1) or ncpBVDV (MOI = 5), using Rapa-treated cells as a positive control. At 48-h post-infection, the cells were washed twice with ice-cold PBS (0.01 M, pH 7.4), harvested using a cell scraper, and washed again with PBS before fixation in 2.5% glutaraldehyde at 4°C for 24 h. After additional PBS washes, the samples were post-fixed in 1% osmium tetroxide for 60 min, dehydrated through a graded ethanol series, and embedded. Ultrathin sections were stained with 2% uranyl acetate and lead citrate (each for 5 min). Cellular ultrastructure features, including autophagosome-like vesicles and ER morphology, were examined by TEM.
## Quantitative real-time PCR
Total cellular RNA was extracted from the variously treated cell samples using the FastPure Cell/Tissue Total RNA Isolation Kit V2 (Vazyme, China) in accordance with the manufacturer's protocol. First-strand cDNA was then synthesized with the HiScript III All-in-one RT SuperMix Perfect for qPCR (Vazyme, China). qPCR was performed on an Applied Biosystems 7500 RT-PCR system using Taq Pro Universal SYBR qPCR Master Mix (Vazyme, China). The relative mRNA expression levels of the target genes were normalized to β-actin as an internal control and calculated via the 2 -△△Ct method. All experiments were performed in triplicate. The qRT-PCR primer sequences are listed in Table 1.
## RNA interference
siBECN1 was designed and synthesized by GenePharma Co., Ltd. (Suzhou, China). A non-targeting siRNA was used as the negative control (siNC). BT cells were cultured to 80% confluence and transfected with siBECN1 using the jetPRIME in vitro DNA and siRNA transfection reagent (Polyplus, France). At 24-h post-transfection, the cells were infected with cpBVDV (MOI=1) or ncpBVDV (MOI=5). After transfection and infection, cell samples and supernatants were collected for further analysis. The knockdown efficiency of siBECN1 was assessed by qRT-PCR and Western blot. Cell viability was measured with Enhanced Cell Counting Kit-8 (CCK-8) (Beyotime, China) according to the manufacturer's instructions.
## Overexpression
Recombinant eukaryotic plasmids expressing the host proteins BECN1 and MAVS were constructed by cloning the respective coding sequences amplified from the bovine cell genome into the vector p3×-Flag and pCMV-Myc using homologous recombination. The resulting constructs were designated as p3×-BECN1 (expressing Flag-tagged BECN1) and pCMV-MAVS (expressing Myc-tagged MAVS). After transfection with these plasmids using jetPRIME in vitro DNA and siRNA transfection reagent (Polyplus, France), cells and corresponding supernatants were collected for subsequent analysis.
## Western blot
Cells subjected to various treatments in six-well plates were harvested and lysed using RIPA Lysis Buffer (P0013B, Beyotime, China) supplemented with a protease and phosphatase inhibitor cocktail (Bimake, USA). Total protein concentration was determined with a Bicinchoninic Acid Protein Assay Kit (#P0011, Beyotime, China). Protein samples were separated by SDS-PAGE, transferred onto polyvinylidene difluoride membranes, and blocked with 5% skimmed milk at 37°C for 2 h. The membranes were incubated overnight at 4°C with the following primary antibodies (all diluted at 1:1,000 unless specified): ATF6 rabbit antibody (#A0202, ABclonal, China), phospho-mTOR-S2448 rabbit antibody (#AP0115, ABclonal, China), MAVS rabbit antibody (#A25005, ABclonal, China), BiP/GRP78 rabbit antibody (#A0241, ABclonal, China), phospho-IRE1-S724 rabbit antibody (#AP1146, ABclonal, China), PERK rabbit antibody (#AP1146, Abclonal, China), XBP1 rabbit antibody (#A1731, ABclonal, China), IRF3 rabbit antibody (#ab68481, Abcam, UK), BECN1 rabbit antibody (#AF5128, Affinity Biosciences, USA), SQSTM1/p62 (#A11483, ABclonal, China), LC3B rabbit antibody (#AF5640, Affinity Biosciences, USA), RIG-I rabbit antibody (#3743, CST, USA), eIF2α (D7D3) rabbit antibody (#9079, CST, USA; 1:1,000),
## Co-immunoprecipitation
To further examine the interaction between the autophagy protein BECN1 and the host protein MAVS, a Co-IP assay was conducted. Briefly, BT cells were grown in six-well plates until reaching over 80% confluency and then co-transfected with plasmids expressing Flag-BECN1 and Myc-MAVS. After transfection, the cells were incubated at 37°C with 5% CO 2 for 36 h. Subsequently, the cells were harvested by scraping and lysed in IP lysis buffer supplemented with protease inhibitor (Beyotime, China). The lysates were centrifuged, and the supernatants were collected, precleared with anti-HA-agarose beads (Sigma, USA), and then incubated with the indicated antibodies. The beads were washed three times with PBS, followed by Western blot analysis.
## Enzyme-linked immunosorbent assay
To evaluate type I IFN production, supernatants from differently treated cells were collected. Levels of bovine IFN-α and IFN-β were then measured using specific enzymelinked immunosorbent assay (ELISA) kits (MM-3472501 for IFN-α, MM-3694801 for IFN-β; Meimian, China) following the manufacturer's protocols.
## Detection of ROS
To assess intracellular ROS levels, two loading methods for the DCFH-DA probe were employed. For the in situ loading, BT cells that were cultured in six-well plates (to ~70% confluence) and infected with BVDV were incubated at different time points post-infec tion with 1 mL of 10 μM DCFH-DA probe at 37°C for 30 min. After centrifugation, the cells were washed three times with serum-free DMEM and analyzed immediately by fluorescent microscope. For ectopic loading, cells were harvested at the indicated time points post-infection, washed twice with pre-cooled PBS, and then incubated with 1 mL of 10 μM DCFH-DA probe for 30 min at 37°C. Following centrifugation and washes with serum-free DMEM, the samples were subjected to flow cytometry analysis.
## Transfection with mRFP-GFP-LC3 adenovirus
Autophagic flux in cpBVDV/ncpBVDV-infected BT cells was monitored using the mRFP-GFP-LC3 adenoviral reporter system. This assay exploits the differential pH stability of GFP and mRFP: GFP fluorescence is quenched in the acidic environment of lysosomes, while mRFP remains stable. Thus, the autophagosomes (GFP + /mRFP + ) appear as yellow puncta in merged images, whereas autolysosomes (GFP -/mRFP + ) are identified as red puncta. Briefly, BT cells were seeded on glass coverslips in 12-well plates and grown to ~70% confluence. Cells were then co-infected with the Ad-mCherry-GFP-LC3B (C3011; Beyotime, China) at MOI=1 and either cpBVDV (MOI=1) or ncp BVDV (MOI = 5) for 2 h at 37°C. After infection, the medium was replaced with 500 μL of DMEM containing 2% FBS, and incubation continued for 24-48 h. LC3 puncta were finally visualized by laser confocal microscopy (FV1000, Olympus, Japan).
## IFA of viral dsRNA
We performed immunofluorescence staining to detect viral dsRNA in BT cells co-infected with the Ad-mCherry-GFP-LC3B. Cells were collected, fixed with 4% paraformaldehyde for 15 min at room temperature, and permeabilized with 0.2% Triton X-100 for 10 min. The cells were then incubated overnight at 4°C with a mouse anti-dsRNA antibody (#76651; CST, USA). Following primary antibody incubation, the cells were stained with an AF647-conjugated goat anti-mouse IgG antibody (A0473; Beyotime, China) for 30 min at 37°C. Nuclei were counterstained with DAPI for 3 min in the dark. Imaging was performed using a laser confocal microscopy (FV1000, Olympus, Japan), and the resulting images were analyzed using ImageJ software.
## Cell viability
The potential cytotoxicity of various compounds was assessed using the CCK-8 assay. BT cells were treated for 48 h with 0.5% DMSO (vehicle control), Rapa, 3-MA, MG132, CQ, BHA, 4-PBA, and jetPRIME reagent, each at their respective working concentrations. After treatment, the medium was replaced with fresh medium containing 10 μL of enhanced CCK-8 solution, and cell viability was measured according to the manufacturer's protocol. All experiments were independently repeated three times.
## Statistical analysis
Data are expressed as mean ± SD from at least three independent experiments. Statistical analyses were performed using GraphPad Prism 8.0 software. Differences between two groups were assessed by an unpaired Student's t-test, while compari sons across multiple groups were analyzed by one-way analysis of variance. Statistical significance is denoted as follows: * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001; ns, not significant.
## References
1. Hussen, Kandeel, Al-Mubarak et al. (2021) "Recent advances on the bovine viral diarrhea virus molecular pathogen esis, immune response, and vaccines development" *Front Vet Sci*
2. Chase, Thakur, Darweesh et al. (2015) "Immune response to bovine viral diarrhea virus--looking at newly defined targets" *Anim Health Res Rev*
3. Lanyon, Hill, Reichel et al. (2014) "Bovine viral diarrhoea: pathogenesis and diagnosis" *The Veterinary Journal*
4. Meyers, Thiel (1996) "Molecular characterization of pestiviruses" *Adv Virus Res*
5. Peterhans, Bachofen, Stalder et al. (2010) "Cytopathic bovine viral diarrhea viruses (BVDV): emerging pestiviruses doomed to extinction" *Vet Res*
6. Bazzucchi, Bertolotti, Giammarioli et al. (2018) "Complete genome sequences of a cytophatic/noncytophatic pair of bovine viral diarrhea virus subtype 1a viruses" *Arch Virol*
7. Pang, Long, Wei (2023) "Immune evasion strategies of bovine viral diarrhea virus" *Front Cell Infect Microbiol*
8. Yim, Mizushima (2020) "Lysosome biology in autophagy" *Cell Discov*
9. Shin, Han, Cho et al. (2023) "Non-cytopathic bovine viral diarrhoea virus 2 induces autophagy to enhance its replication" *Vet Med Sci*
10. Choi, Bowman, Jung (2018) "Autophagy during viral infection -a double-edged sword" *Nat Rev Microbiol*
11. Jiao, Ma, Lin et al. (2023) "SARS-CoV-2 nonstructural protein 6 triggers endoplasmic reticulum stress-induced autophagy to degrade STING1" *Autophagy*
12. Yamane, Zahoor, Mohamed et al. (2009) "Activation of extracellular signal-regulated kinase in MDBK cells infected with bovine viral diarrhea virus" *Arch Virol*
13. Schweizer, Peterhans (1999) "Oxidative stress in cells infected with bovine viral diarrhoea virus: a crucial step in the induction of apoptosis" *J Gen Virol*
14. Bai, Sui, Liu et al. (2022) "The protein arginine methyltransferase PRMT9 attenuates MAVS activation through arginine methylation" *Nat Commun*
15. Yao, Zhong, Wang et al. (2023) "Up-regulated lncRNA CYLD as a ceRNA of miR-2383 facilitates bovine viral diarrhea virus replication by promoting CYLD expression to counteract RIG-I-mediated type-I IFN production" *Int J Biol Macromol*
16. Zhu, Guan, Fang et al. (2023) "Rift valley fever virus nucleoprotein triggers autophagy to dampen antiviral innate immune responses"
17. Lee, Ou (2024) "HCV-induced autophagy and innate immunity" *Front Immunol*
18. Ma, Bai, Jiang et al. (2023) "TRAF6 autophagic degradation by avibirnavirus VP3 inhibits antiviral innate immunity via blocking NFKB/NF-κB activation" *Autophagy*
19. Rangaraju, Verrier, Madorsky et al. (2010) "Rapamycin activates autophagy and improves myelination in explant cultures from neuropathic mice" *J Neurosci*
20. Miller, Oleksy, Perisic et al. (2010) "Finding a fitting shoe for Cinderella: searching for an autophagy inhibitor" *Autophagy*
21. Li, Zhou, Zhao et al. (2023) "Porcine reproductive and respiratory syndrome virus degrades DDX10 via SQSTM1/p62-dependent selective autophagy to antagonize its antiviral activity" *Autophagy*
22. Tian, Wang, Zhao (2019) "Crosstalk between autophagy and type I interferon responses in innate antiviral immunity" *Viruses*
23. Li, Song, Wang et al. "Fan SQ. 2023. The regulation of cell homeostasis and antiviral innate immunity by autophagy during classical swine fever virus infection" *Emerg Microbes Infect*
24. Wang, Chen, Wang et al. (2023) "Effects of spatial expression of activating transcription factor 4 on the pathogenicity of two phenotypes of bovine viral diarrhea virus by regulating the endoplasmic reticulummediated autophagy process" *Microbiol Spectr*
25. Jin, Tian, Chen et al. (2016) "USP19 modulates autophagy and antiviral immune responses by deubiquiti nating Beclin-1" *EMBO J*
26. Wiseman, Mesgarzadeh, Hendershot (2022) "Reshaping endoplasmic reticulum quality control through the unfolded protein response" *Mol Cell*
27. Ye, Rawson, Komuro et al. (2000) "ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs" *Mol Cell*
28. Adachi, Yamamoto, Okada et al. (2008) "ATF6 is a transcription factor specializing in the regulation of quality control proteins in the endoplasmic reticulum" *Cell Struct Funct*
29. Preissler (2019) "Early events in the endoplasmic reticulum unfolded protein response" *Cold Spring Harb Perspect Biol*
30. Hollien, Lin, Li et al. (2009) "Regulated Ire1-dependent decay of messenger RNAs in mammalian cells" *J Cell Biol*
31. Yoshida, Uemura, Mori (2009) "pXBP1(U), a negative regulator of the unfolded protein response activator pXBP1(S), targets ATF6 but not ATF4 in proteasome-mediated degradation" *Cell Struct Funct*
32. Schweizer, Peterhans (2001) "Noncytopathic bovine viral diarrhea virus inhibits double-stranded RNA-induced apoptosis and interferon synthesis" *J Virol*
33. Li, Zhang, Zhao et al. (2024) "Non-cytopathic bovine viral diarrhea virus (BVDV) inhibits innate immune responses via induction of mitophagy" *Vet Res*
34. Deretic (2021) "Autophagy in inflammation, infection, and immunome tabolism" *Immunity*
35. Ma, Wang, Jiang et al. (2022) "Integrative transcriptomics and proteomics analysis provide a deep insight into bovine viral diarrhea virus-host interactions during BVDV infection" *Front Immunol*
36. Deretic, Levine (2009) "Autophagy, immunity, and microbial adaptations" *Cell Host Microbe*
37. Ma, Wang, Wang et al. (2025) "Pestivirus bovine viral diarrhea virus infection triggers lipophagy through the AMPK-PNPLA2/ATGL signaling pathway to promote viral replication"
38. Mizushima (2007) "Autophagy: process and function" *Genes Dev*
39. Fu, Shi, Ren et al. (2014) "Bovine viral diarrhea virus infection induces autophagy in MDBK cells" *J Microbiol*
40. Fu, Shi, Zhang et al. (2014) "Autophagy during early stages contributes to bovine viral diarrhea virus replication in MDBK cells" *J Basic Microbiol*
41. Rajput, Abdelsalam, Darweesh et al. (2017) "Both cytopathic and non-cytopathic bovine viral diarrhea virus (BVDV) induced autophagy at a similar rate" *Vet Immunol Immunopathol*
42. Sir, Chen, Choi et al. (2008) "Induction of incomplete autophagic response by hepatitis C virus via the unfolded protein response †" *Hepatology*
43. Wang, Zhang (2019) "Regulation of autophagy by mTOR signaling pathway" *Adv Exp Med Biol*
44. Albano, Montalbano, Gagliardo et al. (2023) "Autophagy/ mitophagy in airway diseases: impact of oxidative stress on epithelial cells" *Biomolecules*
45. Song, Tan, Miao et al. (2018) "Crosstalk of ER stress-mediated autophagy and ER-phagy: Involvement of UPR and the core autophagy machinery" *J Cell Physiol*
46. Jordan, Wang, Graczyk et al. (2002) "Replication of a cytopathic strain of bovine viral diarrhea virus activates PERK and induces endoplasmic reticulum stress-mediated apoptosis of MDBK cells" *J Virol*
47. Yuan, Song, Zhang et al. (2023) "Replication of standard bovine viral diarrhea strain Ore gonC24Va induces endoplasmic reticulum stress-mediated apoptosis of bovine trophoblast cells" *Cell Stress and Chaperones*
48. Wang, Zhou, Wang et al. (2022) "African swine fever virus K205R induces ER stress and consequently activates autophagy and the NF-κB signaling pathway" *Viruses*
49. Azad, Chen, Gibson (2009) "Regulation of autophagy by reactive oxygen species (ROS): implications for cancer progression and treatment" *Antioxid Redox Signal*
50. Chen, Liu, Bai et al. (2022) "Quercetin inhibits Hsp70 blocking of bovine viral diarrhea virus infection and replication in the early stage of virus infection" *Viruses*
51. Tesseraud, Avril, Bonnet et al. (2021) "Autophagy in farm animals: current knowledge and future challenges" *Autophagy*
52. Xie, Zhao, Song et al. (2021) "Induction of autophagy and suppression of type I IFN secretion by CSFV" *Autophagy*
53. Huang, Yue, Li et al. (2016) "Antibody-dependent enhancement of dengue virus infection inhibits RLR-mediated Type-I IFN-independent signalling through upregulation of cellular autophagy" *Sci Rep*
54. (2026) *Full-Length Text Journal of Virology*
55. Wang, Hou, Pan et al. (2021) "DDIT3 targets innate immunity via the ddit3-otud1-mavs pathway to promote bovine viral diarrhea virus replication" *J Virol*
56. Zhang, Xu, Liu et al. (2023) "The nucleoprotein of influenza A virus inhibits the innate immune response by inducing mitophagy" *Autophagy*
57. Wei, Ni, Song et al. (2010) "The hepatitis B virus X protein disrupts innate immunity by downregu lating mitochondrial antiviral signaling protein" *J Immunol*
58. Wang, Xi, Lei et al. (2013) "Enterovirus 71 protease 2Apro targets MAVS to inhibit anti-viral type I interferon responses" *PLoS Pathog*
59. Mukherjee, Morosky, Delorme-Axford et al. (2011) "The coxsackievirus B 3C protease cleaves MAVS and TRIF to attenuate host type I interferon and apoptotic signaling" *PLoS Pathog*
60. Li, Sun, Seth et al. (2005) "Hepatitis C virus protease NS3/4A cleaves mitochondrial antiviral signaling protein off the mitochondria to evade innate immunity" *Proc Natl Acad Sci*
61. Sun, Feng, Misumi et al. (2023) "Viral protease cleavage of MAVS in genetically modified mice with hepatitis A virus infection" *J Hepatol*
62. Jacobs, Coyne (2013) "Mechanisms of MAVS regulation at the mitochondrial membrane" *J Mol Biol*
63. Vazquez, Horner (2015) "MAVS coordination of antiviral innate immunity" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584621&blobtype=pdf | # Whole-genome sequence characterization of respiratory syncytial virus in the Johns Hopkins Health System during the 2024-2025 respiratory season
Ting Zhuang, Amary Fall, Julie Norton, Omar Abdullah, David Villafuerte, Andrew Pekosz, Eili Klein, Heba Mostafa
## Abstract
Respiratory syncytial virus (RSV) remains a leading cause of lower res piratory tract infections in young children, older adults, and immunocompromised individuals worldwide. In 2023, the first RSV vaccines and the widespread use of monoclonal antibodies were approved, underscoring the need for genomic surveillance to monitor their effectiveness and the emergence of antibody-resistant genotypes. This study details the whole-genome characterization of RSV strains circulating between September and December 2024. Whole RSV genome amplification was performed using overlapping amplicons, followed by Nanopore sequencing. Viral genomes were analyzed to determine subtypes, clade prevalence, and amino acid substitutions. Epidemiological and clinical data were collected to assess associations with specific viral variants. A total of 336 RSV-positive samples were collected for the study. RSV activity followed expected seasonal patterns, with a predominance of RSV-A (94.9%) and the regional evolution of clade A.D.1.6 (68.34%). Over half of the infections occurred in children aged 1-5 years (56.2%). Comorbidities, including immunosuppression, were significantly associated with severe clinical outcomes. Phylogenetic analyses revealed tight cluster ing and low intra-clade diversity. Our results highlight the ongoing genetic evolution of RSV-A following the coronavirus disease 2019 (COVID-19) pandemic. Amino acid substitutions were detected across surface and internal proteins, potentially affecting the effectiveness of vaccines, monoclonal antibodies, or antivirals. Predicted gains and losses of glycosylation sites may further influence antigenic presentation. These findings underscore the need for integrated RSV genomic and epidemiological surveillance. IMPORTANCE With ongoing antiviral drug development, recent approvals of new vaccines, and the continued use of protective antibody therapies, it is crucial to monitor the genetic evolution of RSV. Here, we present a long-amplicon-based whole-genome sequencing protocol for RSV-A and RSV-B, enabling genomic surveillance and represent ing the first report of whole-genome analysis of RSV strains circulating in areas served by the Johns Hopkins Health System during the 2024-2025 respiratory viral season. Our findings demonstrate the value of whole-genome surveillance in identifying emerging clades and molecular variations, and highlight the continued genomic evolution of RSV-A in the post-COVID-19 era.
modest efficacy and toxicity concerns (2). However, several vaccines were recently approved (GSK, AREXVY, Pfizer, Abrysvo, Moderna, and mRESVIA) and monoclonal antibodies (nirsevimab, AstraZeneca/Sanofi, Beyfortus) to provide protection for infants, the elderly, and the immunocompromised (3,4). A minimal number of mutations in the viral glycoproteins have been shown to be associated with host immune evasion (5). Understanding the genomic evolution of RSV and the changes that may reduce the effectiveness of recent preventive immunization approaches is essential.
RSV belongs to the Pneumoviridae family and carries a ~15.2 kb negative-sense RNA genome comprising 10 genes that encode for 11 proteins (6). RSV is classified into two major subgroups: RSV-A and RSV-B (7). The RSV surface attachment (G) protein mediates viral binding to host epithelial cells, while the fusion (F) protein facilitates membrane fusion and viral entry (8,9). These glycoproteins are major targets of neutralizing antibodies and exhibit substantial antigenic variability (10). Based on the genetic variability in the G gene, RSV-A and RSV-B have been divided into different genotypes (11,12). However, more recently, a phylogenetic classification based on whole-genome analysis defined 24 RSV-A lineages and 16 RSV-B lineages (13).
We previously described the genomic diversity of RSV during the 2023-2024 season using an analysis of the G and F genes (14). In the current study, we optimized a long-amplicon protocol that enables recovery of full genomes for both RSV-A and RSV-B, allowing comprehensive characterization of RSV strains circulating during the 2024-2025 season. The study aimed to assess RSV subtype prevalence and seasonal patterns, examine amino acid polymorphisms across surface and internal proteins, and evaluate clinical and demographic associations with infection severity. To our knowledge, this is the first report of comprehensive whole-genome RSV surveillance from the state of Maryland.
## MATERIALS AND METHODS
## Study specimens, nucleic acid extraction, and real-time RT-PCR
Standard-of-care RSV diagnostic tests used in the Johns Hopkins Health System (JHHS) include the Cepheid Xpert Xpress SARS-CoV-2/Flu/RSV test or the GenMark Dx ePlex RP1/RP2 respiratory panels (15,16), both of which target the nucleocapsid (N) gene. Nucleic acid was extracted from 300 µL of each specimen and eluted to 60 µL using the Chemagic Viral DNA/RNA 360 Kit (Revvity). A research-use-only real-time RT-PCR targeting the RSV matrix (M) gene was conducted using Luna Universal Probe One-Step RT-qPCR Kit (E3006) to determine the cycle threshold (Ct) values (17).
## Primer design and two-step multiplex RT-PCR
Consensus full-length RSV-A and RSV-B genomes were obtained from GenBank to guide primer design. Primers were designed for four overlapping long-amplicons targeting both RSV-A and RSV-B genomes. First-strand cDNA synthesis was performed using 2 µL of LunaScript RT SuperMix (New England Biolab) and 8 µL of extracted RNA (thermocy cling conditions: 2 min at 25°C, 10 min at 55°C, and 1 min at 95°C). Two multiplex PCR reactions (Pool 1 and Pool 2) were used to amplify the cDNA. Each 25 µL PCR reaction contained 5 µL of cDNA, 12.5 µL of Q5 Hot Start High-Fidelity 2× Master Mix, 3.5 µL of nuclease-free water, and primers at either 200 or 400 nM final concentration (achieved by adding 0.5 or 1.0 µL of each 10 µM primer, respectively). Thermocycling conditions were initial denaturation at 98°C for 30 s, 40 cycles of 98°C for 10 s, 50°C for 30 s, and 72°C for 4 min, and a final extension at 72°C for 10 min. Table S2 outlines the primer and probe sequences for real-time RT-PCR and multiplex RT-PCR.
## Genome assembly and phylogenetic tree construction
Libraries were prepared using the NEBNext ARTIC Library Prep Kit and the Oxford Nanopore Native Barcoding Kit 96 (v.14) and sequenced on an Oxford Nanopore GridION device. The resulting FASTQ files were processed using our in-house analysis pipeline (14). Closest reference genomes were identified via BLAST searches (MH447951 for RSV-A and OP975389 for RSV-B), and draft genomes were assembled using the mini_assemble module in Pomoxis. Consensus polishing was performed with Medaka, and sequencing depth was assessed using Samtools. Quality control filters were applied to retain only sequences with quality scores between 30 and 90 or higher. Low-quality or incomplete genomes were associated with higher average Ct values and were manually excluded after sequencing. RSV clade assignments were performed using Nextclade (v.3.15.3), which classifies sequences based on whole-genome phylogenetic placement. This tool assigns clade designations consistent with current genomic surveillance nomenclature, incorporating both sublineages and G-clade information. Sequences were aligned with MAFFT (v.7). Phylogenetic trees were generated using the maximum likelihood method in IQ-Tree (v.2.4.0) with 1,000 bootstrap replicates. Trees were visualized using FigTree (v. 1.4.4). Reference genomes used for the phylogenetic analysis are listed in Table S3.
## Amino acid sequence analysis and glycosylation prediction
RSV-A sequences were aligned and compared to the reference genome hRSV/A/ England/397/2017 (EPI_ISL_412866) due to its well-characterized full-length sequence representative of the globally dominant ON1 genotype (13). Prefusion (Protein Data Bank [PDB]: 5UDE) and post-fusion (PDB: 3RRR) F protein structures from the PDB were visualized with PyMOL (v.3.1). Amino acid variants were identified in the F protein at frequencies of ≥1% and in other viral proteins at frequencies of ≥10%. N-linked glycosylation sites (defined by the N-X-S/T motif, where X is not proline) were predicted with NetNGlyc 1.0 (threshold ≥0.5), and O-linked glycosylation sites (serine/threonine residues) were predicted with NetOGlyc 4.0 (18,19). All substitutions are detailed in Table S4.
## Statistical analysis
Clinical and demographic data were collected for RSV-positive patients in bulk. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using Microsoft Excel and mapped by Python to assess associations between clinical variables and outcomes (hospital admission, supplemental oxygen use, and intensive care unit [ICU]-level care), and statistical significance was defined as P < 0.05. Calculated odds ratios used for the forest plot are provided in Table S5.
## RESULTS
## RSV prevalence and cohort characteristics
Between June 2024 and January 2025, enterovirus/rhinovirus, influenza, RSV, and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exhibited the highest positivity rates in JHHS (Fig. 1A). Enterovirus/rhinovirus peaked at 21.65% in September, SARS-CoV-2 at approximately 9.53% in August, and RSV at 9.37% in November and 9.17% in December (Fig. 1B).
Between June 2024 and January 2025, 4.6% (2,096 out of 45,567) of respiratory tests performed within the JHHS were positive for RSV. We randomly sequenced 431 of these specimens, yielding 336 sequences suitable for analysis. RSV-A accounted for the majority (319 out of 336, 94.9%) of sequences (Table 1). The male-to-female ratio was 1.0:1.06, and over half of the patients (189 out of 336, 56.2%) were young children aged 1-5 years (median age: 2 years). Comorbidities were common, particularly lung disease (58 out of 336, 17.3%) and cancer (46 out of 336, 13.7%). Co-infections were identified in 6% (20 out of 336) of patients, most frequently with enterovirus/rhinovirus (13 out of 20, 65%). The majority of patients (320 out of 336, 95.2%) were seen in the emergency department; 20.2% (68 out of 336) required hospitalization; 15.8% (53 out of 336) received supplemental oxygen; and 4.5% (15 out of 336) required ICU-level care.
Associations between comorbidities and outcomes, including hospital admission, supplemental oxygen use, and ICU-level care, were evaluated (Fig. 2). Certain comorbidi ties were significantly associated with increased odds of admission, such as diabetes (OR: 10.13, 95% CI: 2.55-40.33; P = 0.001). For supplemental oxygen use, significant associa tions were observed with comorbidities such as heart failure (OR: 5.69, 95% CI: 1.38-23.53; P = 0.02). Regarding ICU-level care, only immunosuppression was significantly associated with increased odds of ICU admission (OR: 3.96, 95% CI: 1.28-12.23; P = 0.02).
## Full genome phylogenetic analysis
A total of 10 RSV-A clades were identified this season, with A.D.1.6 (218 out of 319, 68.34%) being predominant (Fig. 3A). RSV-A phylogeny showed tight clustering within this dominant clade. Other RSV-A clades also clustered together with their respective reference clade genomes from different geographical locations and years (Fig. 3A corresponding to G-clade GA2.3.5, and all RSV-B sequences belonged to the BA9 genotype, corresponding to G-clade GB5.0.5a. The JHHS sequences clustered with RSV genomes from Europe (Ukraine, Spain, Norway, Italy, and France) and Mexico within this season. Notably, individual-level travel history was not available, limiting our ability to determine whether these introductions were associated with international travel.
## G and F amino acid polymorphisms
We identified 28 amino acid substitutions in the G protein compared to the reference genome (Table 2), with the majority (16 out of 28, 57.1%) located in the second hyper variable region (HVR2). Two substitutions, I134K and D284G, were present in all sequen ces. Additionally, we observed the A57T substitution (11.04%) within the transmembrane domain, which was not detected in the previous season. The reference G protein contains three predicted N-linked glycosylation sites at positions 103, 135, and 237. In our sequences, the substitution S100N (11.1%) introduced an additional N-linked glycosylation site, while T137K (11.6%) resulted in the loss of one. O-linked glycosylation predictions in 50 randomly selected RSV-A sequences identified 50 predicted O-glycosylation sites in HVR2 (compared to 51 in the reference, including 10 in the duplicated region) (Fig. 4). The substitutions P217S (68.47%) and G296S (68.79%) added O-glycosylation sites in 16.0% (8 out of 50) and 70.0% (35 out of 50) of selected sequen ces, respectively; G296S lies within the duplication region. In contrast, the substitution S299N (69.52%), T319A/I, and T320A (91.43%) collectively led to the loss of three predicted O-glycosylation sites, including S299N in the duplication region.
In the F protein, 13 amino acid substitutions were identified compared to the reference genome. Two substitutions, K272N (4.4%) and S276N (2.2%), were located within antigenic site II (Fig. 5). The reference F protein contains five N-linked glycosyla tion sites at positions 27, 70, 116, 120, and 126. The substitution T122A (8.8%) resulted in the loss of one N-linked glycosylation site, and no substitutions introduced new Nglycosylation sites. Among the same 50 randomly selected sequences, no O-linked glycosylation sites were predicted in the reference genome, but six sites were predicted in our sequences at positions 99, 105, 118, 128, 244, and 248, each appearing at different frequencies. In total, 12.2% (39 out of 319) of sequences gained at least one O-glycosyla tion site, with the majority (87.2%, 34 out of 39) having a predicted site at position 244. In some RSV genomes, F substitutions (R553K: 6 out of 319, 1.88%; S554N: 21 out of 319, 6.58%) were seen as mixed populations and were not captured in the consensus sequence, leading to ambiguous base calls (N).
## Amino acid polymorphisms in other proteins
In the polymerase (L) protein, we identified 18 amino acid substitutions spanning all domains compared to the reference genome (Fig. 6A). Half of these (9 out of 18, 50%) clustered within the connector domain (CD) domain, primarily between amino acid positions 1,653 and 1,731, and were present at high frequencies. One-third of the substitutions (6 out of 18, 33.3%) were located in the RNA-dependent RNA polymerase (RdRp) domain. The Cap domain and CTD had two and one substitutions, respectively, while no substitutions were observed in the MT domain. Several L protein substitutions, P171L, R256K, Y598H, L1438Q, E1725G, and G1731D, were fixed in all sequences. Substitutions were also identified in the M, M2-1, M2-2, NS2, P, SH, and N proteins, whereas none were detected in NS1 compared to the reference genome (Fig. 6B). Notably, several substitutions became fixed in all our sequences, including V352A in the N protein, L55P in the P protein, S176P in M2-1, and S46N in M2-2.
## DISCUSSION
Public health measures implemented during the COVID-19 pandemic significantly reduced the prevalence and genetic diversity of respiratory viruses during the 2020-2021 seasons (21). As these measures were relaxed, RSV resurged with an atypical early fall peak globally (20,22,23), likely driven by reduced population immunity (24,25). Although global antibody recovery was observed following the resurgence (26), some studies reported no significant differences in population-level RSV immunity between the prepandemic and post-pandemic periods (27,28). During the 2024-2025 season, RSV activity increased in late 2024 and peaked in January 2025, reflecting a gradual return to prepandemic seasonal patterns (29)(30)(31). This season was marked by RSV-A predominance, in contrast to the previous season's RSV-B dominance, consistent with global subtype shifts influenced by herd immunity dynamics (32,33). However, RSV seasonality continues to vary across national and regional levels (34), and other studies in the U.S.A. observed RSV-B predominance during the same season (35).
In our cohort, children aged 1-5 years accounted for the majority of RSV cases, despite national and JHHS data indicating higher positivity rates among infants and older adults (36,37). Notably, this age group currently lacks access to prophylactic interventions and may experience reduced protection from maternally derived immunity (38). The predominant RSV-A clade, A.D.1.6, was responsible for most hospitaliza tions, although no significant associations with clinical severity were observed. While our univariate logistic regression analyses suggested potential associations between comorbidities and clinical outcomes, the limited sample size constrains the strength of these conclusions.
Our phylogenetic analysis revealed regional expansion of clade A.D.1.6, a subline age of the globally dominant ON1 genotype of RSV-A, which is characterized by a 72-nucleotide duplication in the G gene. This expansion is likely driven by immune evasion and altered transmissibility, as observed in SARS-CoV-2 evolution (39,40) from the globally circulating BA9 genotype carrying the hallmark 60-nucleotide G gene duplication. The reduced representation of RSV-B in our data set-compared to previous seasons when BA9-derived lineages were more prevalent within JHHS-may reflect a recent bottleneck event (42), although undersampling or geographical restriction cannot be ruled out. We also detected a few B.D.E.1 cases, a clade dominant in the prior season, indicating potential epidemiological relevance. This observation aligns with a recent B.D.E.1 surveillance report from Beijing (43).
We found that the set of HVR1 and HVR2 G protein substitutions prevalent in previous seasons persisted at high frequency (>88%) during the current season (14). Notably, substitutions such as P71L, H90Y, I134K, and S243I have been associated with severe acute respiratory infections in pediatric patients (44). Although limited information is available regarding the impact of substitutions in the transmembrane domain, the emergence of A57T and several other changes in this region have been shown to reduce the affinity of neutralizing antibodies (45). The observed gains and losses of glycosylation sites likely reflect viral adaptation under immune pressure (46,47). For instance, newly introduced glycosylation sites may help mask immunodominant epitopes (48), while the loss of specific glycans could enhance antigen presentation and T-cell activation (49). These ongoing changes in substitutions and glycosylation patterns in the G protein highlight the need for continued genomic surveillance.
In the F protein, we tracked amino acid substitutions occurring at a frequency of ≥1% due to their relevance for vaccines and monoclonal antibody therapies (50). All four substitutions detected during the previous season persisted into the current season (14). A new substitution, K272N, along with S276N, lies within antigenic site II, the target of Palivizumab. K272N has been shown to impair palivizumab neutralization (51), while S276N and S377N are located within immunodominant regions of the F protein (52).
We observed the loss of a conserved N-linked glycosylation site in some F sequences due to the T122A substitution, which may impact immune recognition (49). Although the reference F protein lacks known O-linked glycans, some of our genomes contain amino acid changes that are predicted to introduce potential O-glycosylation sites, which could possibly contribute to shielding of neutralizing epitopes (53). Importantly, none of the substitutions in our sequences affected the specific residues engineered in the new RSV F vaccines (S155C, S190F, V207L, and S290C), suggesting that vaccine-targeted epitopes remain conserved (54). However, a recent study reported limited evidence of antigenic drift in the F protein during the first season following vaccine introduction (55).
The high-frequency clustering of L protein substitutions in the RdRp and CD domains may impact viral replication and sensitivity to antivirals (56). These changes could potentially influence the binding of nucleoside analog inhibitors such as ribavirin (57), although structural insights into the RdRp and CD regions in the context of these nucleotide analogs remain limited (58). Notably, none of the previously reported resistance-associated substitutions in the CD domain (Y1631H/C, L1502Q, and H1632Q), which have been linked to reduced antiviral potency, were found in our sequences (59). During initial sequencing, one amplicon covering L residues 889-2,166 showed low coverage, which we corrected by adjusting primer concentrations. Finally, all substitu tions identified in the other RSV proteins were previously reported in Global Initiative on Sharing All Influenza Data and have been observed globally (44). The fixation of some of these changes in our genomes highlights the need for further functional investigation.
This study has several limitations, including the small number of RSV-B samples, which limited subtype comparisons, the lack of assessment for RSV-A/B co-infections, and potential sample collection bias due to clinical testing being performed primarily in certain groups or symptomatic patients. Additionally, all glycosylation-related findings are based on in silico predictions and were not experimentally validated. Individual-level data on prior RSV exposure or administration of antivirals and monoclonal antibodies were not available, limiting our ability to directly assess the impact of immune pressure on viral evolution. Nonetheless, our findings highlight the realignment of RSV seasonal ity in Maryland and underscore the value of whole-genome surveillance in detecting emerging clades and molecular variation.
## References
1. Li, Wang, Blau et al. (2022) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis" *Lancet*
2. Hoover, Eades, Lam (2018) "Pediatric antiviral stewardship: defining the potential role of ribavirin in respiratory syncytial virusassociated lower respiratory illness" *J Pediatr Pharmacol Ther*
3. Kelleher, Subramaniam, Drysdale (2025) "The recent landscape of RSV vaccine research" *Ther Adv Vaccines Immunother*
4. Langedijk, Vrancken, Lebbink et al. (2024) "The genomic evolutionary dynamics and global circulation patterns of respiratory syncytial virus" *Nat Commun*
5. Moore, Stokes, Hartert (2013) "The impact of viral genotype on pathogenesis and disease severity: respiratory syncytial virus and human rhinoviruses" *Curr Opin Immunol*
6. Battles, Mclellan (2019) "Respiratory syncytial virus entry and how to block it" *Nat Rev Microbiol*
7. Mufson, Orvell, Rafnar et al. (1985) "Two distinct subtypes of human respiratory syncytial virus" *J Gen Virol*
8. Mas, Nair, Campbell et al. (2018) "Antigenic and sequence variability of the human respiratory syncytial virus F glycoprotein compared to related viruses in a comprehensive dataset" *Vaccine (Auckland)*
9. Meng, Hotard, Currier et al. (2016) "Respiratory syncytial virus attachment glycoprotein contribution to infection depends on the specific fusion protein" *J Virol*
10. Anderson, Do, Van Kasteren et al. (2024) "The role of respiratory syncytial virus G protein in immune cell infection and pathogenesis" *EBioMedicine*
11. Muñoz-Escalante, Comas-García, Bernal-Silva et al. (2021) "Respiratory syncytial virus B sequence analysis reveals a novel early genotype" *Sci Rep*
12. Muñoz-Escalante, Comas-García, Bernal-Silva et al. (2019) "Respiratory syncytial virus A genotype classification based on systematic intergenotypic and intragenotypic sequence analysis" *Sci Rep*
13. Goya, Ruis, Neher et al. (2024) "Standardized phylogenetic classification of human respiratory syncytial virus below the subgroup level" *Emerg Infect Dis*
14. Yunker, Fall, Norton et al. (2024) "Genomic evolution and surveillance of respiratory syncytial virus during the 2023-2024 season" *Viruses*
15. Jarrett, Uhteg, Forman et al. (2021) "Clinical performance of the GenMark Dx ePlex respiratory pathogen panels for upper and lower respiratory tract infections" *J Clin Virol*
16. Mostafa, Carroll, Hicken et al. (2021) "Multicenter evaluation of the Cepheid Xpert Xpress SARS-CoV-2/Flu/RSV test" *J Clin Microbiol*
17. Fry, Chittaganpitch, Baggett et al. (2010) "The burden of hospitalized lower respiratory tract infection due to respiratory syncytial virus in rural Thailand" *PLoS One*
18. Gupta, Brunak (2001) "Prediction of glycosylation across the human proteome and the correlation to protein function"
19. Steentoft, Vakhrushev, Joshi et al. (2013) "Precision mapping of the human O-GalNAc glycopro teome through simplecell technology" *EMBO J*
20. Lavoie, Reicherz, Solimano et al. (2021) "Potential resurgence of respiratory syncytial virus in Canada" *CMAJ*
21. Holmdahl, Bents, Baker et al. (2024) "Differential impact of COVID-19 non-pharmaceutical interventions on the epidemiological dynamics of respiratory syncytial virus subtypes A and B" *Sci Rep*
22. Van Summeren, Meijer, Aspelund et al. (2021) "Low levels of respiratory syncytial virus activity in Europe during the 2020/21 season: what can we expect in the coming summer and autumn/winter?" *Euro Surveill*
23. Yeoh, Foley, Smith et al. (2021) "Impact of coronavirus disease 2019 public health measures on detections of influenza and respiratory syncytial virus in children during the 2020 Australian winter" *Clin Infect Dis*
24. Cohen, Ashman, Taha et al. (2021) "Pediatric infectious disease group (GPIP) position paper on the immune debt of the COVID-19 pandemic in childhood, how can we fill the immunity gap?" *Infect Dis Now*
25. Reicherz, Xu, Abu-Raya et al. (2022) "Waning immunity against respiratory syncytial virus during the coronavirus disease 2019 pandemic" *J Infect Dis*
26. Reicherz, Paramo, Bone et al. (2025) "Recovery of antibody immunity after a resurgence of respiratory syncytial virus infections" *J Infect Dis*
27. Lastrucci, Pacifici, Puglia et al. (2024) "Recent trends in hospitalizations for respiratory syncytial virus after the COVID-19 pandemic and before routine immunization: seasonality and severity updates from the 2023" *Int J Infect Dis*
28. Walker, Foster, Sevendal et al. (2022) "Clinical, genomic, and immunological characteriza tion of RSV surge" *Pediatrics*
29. Chen, Tsui, Gutierrez et al. (2024) "COVID-19 pandemic interventions reshaped the global dispersal of seasonal influenza viruses" *Science*
30. Li, Cong, Wei et al. (2024) "Characterising the changes in RSV epidemiology in Beijing, China during 2015-2023: results from a prospective, multicentre, hospital-based surveillance and serology study" *Lancet Reg Health West Pac*
31. Thindwa, Li, Cooper-Wootton et al. (2024) "Global patterns of rebound to normal RSV dynamics following COVID-19 suppression" *BMC Infect Dis*
32. Rios-Guzman, Simons, Dean et al. (2024) "Deviations in RSV epidemiological patterns and population structures in the United States following the COVID-19 pandemic" *Nat Commun*
33. Zambrana, Huang, Solis et al. (2024) "Spatial and temporal variation in respiratory syncytial virus (RSV) subtype RNA in wastewater and relation to clinical specimens"
34. Ye, Deng, Miao et al. (2025) "Understanding the local-level variations in seasonality of human respiratory syncytial virus infection: a systematic analysis" *BMC Med*
35. Laverriere, Behar, Sher-Jan et al. (2025) "Genomic epidemiology of respiratory syncytial virus in a new England hospital system, 2024. medRxiv"
36. Caini, Casalegno, Rodrigues et al. (2024) "Change in age profile of respiratory syncytial virus disease over the course of annual epidemics: a multi-national study" *J Infect*
37. Patton, Moline, Whitaker et al. (2024) "Interim evaluation of respiratory syncytial virus hospitalization rates among infants and young children after introduction of respiratory syncytial virus prevention products -United States" *MMWR Morb Mortal Wkly Rep*
38. Glenn-Otto, Dziacky (2025) "Prevention strategies for RSV infection in infants" *Pediatr Rev*
39. Meijers, Ruchnewitz, Eberhardt et al. (2023) "Population immunity predicts evolutionary trajectories of SARS-CoV-2" *Cell*
40. Wang, Didelot, Bi et al. (2022) "SARS-CoV-2 transmissibility compared between variants of concern and vaccination status" *Brief Bioinform*
41. Fall, Elawar, Hodcroft et al. (2021) "Genetic diversity and evolutionary dynamics of respiratory syncytial virus over eleven consecutive years of surveillance in Senegal" *Infect Genet Evol*
42. Agoti, Mwihuri, Sande et al. (2012) "Genetic relatedness of infecting and reinfecting respiratory syncytial virus strains identified in a birth cohort from rural Kenya" *J Infect Dis*
43. Wei, Wang, Li et al. (2024) "Novel imported clades accelerated the RSV surge in Beijing" *J Infect*
44. Jallow, Diagne, Sagne et al. (2023) "Respiratory syncytial virus in pediatric patients with severe acute respiratory infections in Senegal: findings from the 2022 sentinel surveillance season" *Sci Rep*
45. Walsh, Hall, Briselli et al. (1987) "Immunization with glycoprotein subunits of respiratory syncytial virus to protect cotton rats against viral infection" *J Infect Dis*
46. García-Beato, Martínez, Francí et al. (1996) "Host cell effect upon glycosylation and antigenicity of human respiratory syncytial virus G glycoprotein" *Virology (Auckland)*
47. Lee, Fang, Wang et al. (2008) "Genetic diversity and molecular epidemiology of circulating respiratory syncytial virus in central Taiwan" *Viruses*
48. Carnell, Billmeier, Vishwanath et al. (2023) "Glycan masking of a non-neutralising epitope enhances neutralising antibodies targeting the RBD of SARS-CoV-2 and its variants" *Front Immunol*
49. Efstathiou, Abidi, Harker et al. (2020) "Revisiting respiratory syncytial virus's interaction with host immunity, towards novel therapeutics" *Cell Mol Life Sci*
50. Fleming, Baral, Higgins et al. (2023) "Value profile for respiratory syncytial virus vaccines and monoclonal antibodies" *Vaccine (Auckland)*
51. Bates, Keefer, Slaughter et al. (2014) "Escape from neutralization by the respiratory syncytial virusspecific neutralizing monoclonal antibody palivizumab is driven by changes in on-rate of binding to the fusion protein" *Virology (Auckland)*
52. Bin, Liu, Tabor et al. (2019) "Emergence of new antigenic epitopes in the glycoproteins of human respiratory syncytial virus collected from a US surveillance study" *Sci Rep*
53. Patel, Tian, Flores et al. (2020) "Flexible RSV prefusogenic fusion glycoprotein exposes multiple neutralizing epitopes that may collectively contribute to protective immunity" *Vaccines (Basel)*
54. Mclellan, Chen, Leung et al. (2013) "Structure of RSV fusion glycoprotein trimer bound to a prefusionspecific neutralizing antibody" *Science*
55. Lauring, Edson, Surie et al. (2025) "Genomic characterization of RSV in the US by vaccination status" *JAMA*
56. Gilman, Liu, Fung et al. (2019) "Structure of the respiratory syncytial virus polymerase complex" *Cell*
57. Liuzzi, Mason, Cartier et al. (2005) "Inhibitors of respiratory syncytial virus replication target cotranscriptional mRNA guanylylation by viral RNAdependent RNA polymerase" *J Virol*
58. Yang, Jiang, Jiang et al. (2025) "Advances in development of antiviral strategies against respiratory syncytial virus" *Acta Pharm Sin B*
59. Bonneux, Shareef, Tcherniuk et al. (2024) "JNJ-7184, a respiratory syncytial virus inhibitor targeting the connector domain of the viral polymerase" *Antiviral Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817898&blobtype=pdf | # Stealth replication of SARS-CoV-2 Omicron in the nasal epithelium at physiological temperature
Bárbara Fonseca, Rémy Robinot, Vincent Michel, Akram Mendez, Samuel Lebourgeois, Chloé Chivé, Raphaël Jeger- Madiot, Roshan Vaid, Vincent Bondet, Elizabeth Maloney, Florence Guivel-Benhassine, Olivier Schwartz, Darragh Duffy, Tanmoy Mondal, Samy Gobaa, Lisa Chakrabarti
## Abstract
The COVID-19 pandemic was marked by successive waves of SARS-CoV-2 variants with distinct properties. The Omicron variant that emerged in late 2021 showed a major antigenic shift and rapidly spread worldwide. Since then, Omicronderived variants have maintained their global dominance, for reasons that remain incompletely understood. We report that the original Omicron variant BA.1 evolved several traits that converged in facilitating viral spread. First, Omicron displayed an early replicative advantage over previous variants when grown in a reconstructed human nasal epithelium model. The increase in Omicron replication was more marked at the physiologically relevant temperature of 33°C found in human nasal passages. Omicron also caused a decrease in epithelial integrity, as measured by transepithelial electrical resistance and caspase-3 activation. Furthermore, Omicron caused a more marked loss of motile cilia at 33°C than other variants, suggesting a capacity to impair mucociliary clearance. Omicron induced a broad transcriptional downregulation of ciliary genes but only a limited upregulation of host innate defense genes at 33°C. The lower production of type I and type III interferons in epithelia infected by Omicron compared to those infected by the Delta variant, at 33°C as well as 37°C, confirmed the increased capacity of Omicron to evade the innate antiviral response. Thus, Omicron combined replication speed, motile cilia impairment, and limited induction of innate antiviral responses when propagated in nasal epithelia at physiological temperature. Omicron has the capacity to propagate rapidly but stealthily in the upper respiratory tract, which likely contributed to the evolutionary success of this SARS-CoV-2 variant.
IMPORTANCEThe COVID-19 pandemic was initially characterized by a rapid succession of viral variants that emerged independently of each other, with each of these variants outcompeting the previous one. A major evolutionary shift occurred in late 2021, with the emergence of the highly divergent Omicron BA.1 variant. Since then, all the dominant SARS-CoV-2 variants have been derived from Omicron, for reasons that remain incompletely understood. Here, we compared the replication of SARS-CoV-2 variants in a human nasal epithelium model grown at 37°C and also at 33°C, a temperature that approximates that found in the nasal cavity. In this primary epithelial model, Omicron showed an early replicative advantage that was more marked at 33°C. However, Omicron triggered only a minimal antiviral interferon response at this temperature. Omicron could thus propagate rapidly while partly escaping the innate response at physiological nasal temperature, which helps account for the efficient dissemination of this variant worldwide.
T he COVID-19 pandemic was characterized by sequential evolutionary shifts, with a rapid succession of viral variants rising to global dominance (1,2). The initial SARS-CoV-2 Wuhan strain that emerged in late 2019 was replaced in the first semester of 2020 by a variant with higher infectivity, due to the presence of a D614G mutation in the S1 surface subunit of the Spike protein (3). Starting from late 2020, a series of variants termed Alpha, Beta, Gamma, Delta, and Omicron emerged successively and independently of each other, with each of these variants of concern (VOCs) outcompet ing previous variants. The Delta variant, which reached worldwide dominance in spring 2021, was associated with increased disease severity, which was in part ascribed to a higher fusogenic capacity of its Spike glycoprotein (4)(5)(6). A major evolutionary shift occurred in late 2021, with the emergence of the Omicron BA.1 variant that differed by more than 32 amino acids in the Spike glycoproteins from previous strains (7). Since then, all the dominant SARS-CoV-2 variants have been phylogenetically derived from Omicron, with adaptation to preexisting immunity becoming a major selective pressure for variant replacement (8,9).
Why Omicron-derived variants have maintained their global dominance remains incompletely understood. Epidemiological studies have documented a marked increase in viral fitness upon Omicron emergence, as measured by rates of variant spread over time (10). While escape from preexisting neutralizing antibodies is thought to be a key factor to explain the Omicron surge, there are signs that this variant is also intrinsically efficient at transmission from human to human (2). In particular, longitudinal analyses of nasopharyngeal swabs showed that detectable viral replication in the nasal mucosa occurs earlier for the Omicron than the Delta variant (11). Consistent with this observa tion, Omicron shows faster replication kinetics than previous variants in human airway explants (12) and in reconstructed epithelial cultures derived from primary human nasal cells (13)(14)(15).
While the Omicron variant spread efficiently, it proved slightly less pathogenic than the preceding Delta variant, as measured by hospitalization rates and pneumonia severity in individuals who were not previously vaccinated or infected by SARS-CoV-2 (16,17). These epidemiological observations are in agreement with findings in animal models, with a lower degree of lung dysfunction induced by Omicron in infected golden Syrian hamsters, ferrets, and nonhuman primates (14,(18)(19)(20). Of note, while most animal models show an attenuated Omicron phenotype, they often do not reproduce the rapid viral replication observed in the human nasal mucosa. It is also relevant that while Omicron replicates faster than Delta in human bronchus explants, it replicates more slowly than Delta in human lung explants (12) and alveolar organoids (21,22). This differential tropism fits with the notion of a bias toward the upper over the lower respiratory tract, possibly accounting for a lower pathogenic potential of Omicron.
Different mechanisms have been proposed to account for the increased replication capacity of Omicron in human airway epithelia, including a higher affinity of Spike for the ACE-2 receptor (23,24) or a better interaction of Spike with the negatively charged glycocalyx at the cellular surface (25), leading to an overall better attachment of viral particles to primary ciliated epithelial cells (26). Several studies have also documented that the Omicron entry pathway differs from that of previous variants, with a lower reliance on the surface serine protease TMPRSS2 to release the fusion peptide and an expanded usage of other proteases (15,27,28). In addition, an improved capacity to evade the innate antiviral response has also been reported for Omicron, which could contribute to its rapid replication kinetics (29)(30)(31). To further address these questions, we systematically compared the replication of Omicron and previous variants in a reconstructed human nasal epithelium model grown at the air-liquid interface (ALI). This model recapitulates key features of the nasal mucosa, including a pseudostratified epithelium structure, the presence of diverse epithelial cell types (ciliated, goblet, and basal cells), and the competence for mucociliary clearance function (32). In addition to its barrier function, the nasal epithelium has also an air conditioning function, warming and moistening inspired air to maintain the internal milieu of the lungs (33). This results in a temperature gradient from the nares to the nasopharynx, with 31°C-34°C typically measured at the posterior end of the nasal cavity in humans (34). Therefore, we chose to compare SARS-CoV-2 variant replication at 37°C but also at 33°C, a more physio logical temperature that approximates that found in the nasal cavity. We report that the Omicron replicative advantage is more marked at the temperature found in nasal passages as this variant manages to spread efficiently and induce motile cilia damage while triggering only a minimal innate interferon (IFN) response from primary epithelial cells at 33°C. These findings help account for the rapid and efficient dissemination of the Omicron variant worldwide.
## RESULTS
## The replicative advantage of Omicron in nasal epithelial cells is more marked at a physiologically relevant temperature
For infection experiments, we used reconstructed human nasal epithelia generated from pooled human donors (MucilAir, Epithelix) to minimize donor-dependent variability. Prior to infection, cultures maintained at the ALI were checked for motile cilia activity and mucus production, ensuring the proper differentiation of a pseudostratified epithelium. Infectious inocula with different SARS-CoV-2 variants were normalized based on viral RNA content and infections were monitored for 4 days.
In nasal epithelial cultures grown at 37°C, all the viruses tested showed an efficient viral replication, including the ancestral Wuhan strain, the early variant D614G, and the later variants Alpha, Delta, and Omicron BA.1 (Fig. 1A), as well as the Beta and Gamma variants (Fig. S1A andC). Viral replication reached a plateau at day 2 (D2), except for the Omicron BA.1 variant, which reached close to plateau values at day 1 (D1). Interestingly, the replicative advantage of Omicron BA.1 became more apparent when the epithelial cultures were grown at 33°C, a temperature more relevant to nasal physiology (Fig. 1B ; Fig. S1B andD). Comparison of viral RNA copy number values at D1 showed a trend for higher replication of BA.1 compared to the D614G reference variant at 37°C (not significant; Fig. 1C) and confirmed that BA.1 replication was significantly higher than that of D614G at 33°C (Fig. 1D, P = 0.016). The replicative advantage of Omicron was apparent at early time points, but the replication of other VOCs caught up with that of BA.1 at the later day 4 (D4) time point, possibly due to a limitation in the target cell availability. An additional infection experiment performed on reconstructed nasal epithelia obtained from a single human donor confirmed the replicative advantage of Omicron BA.1 (Fig. S2), consistent with findings obtained on samples generated from pooled donors.
Computation of the ratio of viral RNA copies at 33°C over those at 37°C showed that viral replication tended to be slower at 33°C for all the viruses tested, with ratios below 1 in most cases, except for Omicron BA.1 at D4 (Fig. 1E; Fig. 1E). In addition, Omicron BA.1 was the only variant that showed a ratio significantly higher than that of the D614G reference strain at D2 (P < 0.001) and D4 (P < 0.01), emphasizing the efficient replication of the Omicron variant at a temperature more physiologically relevant for nasal infection.
We then performed infectivity measurements using the S-fuse assay, which quantifies SARS-CoV-2 infection by the emission of a GFP signal, after fusion of target U2OS-ACE2 cells expressing a GFP split system (9,35,36). Analysis of viral supernatants collected from two independent infection experiments showed that Omicron BA.1 had an early replicative advantage over D614G and Delta when propagated at 33°C (Fig. S3). Compari son of viral infectivities for infections carried out at 37°C showed a more variable advantage of Omicron, which proved detectable in one out of two experiments. Consistent with viral RNA release data, these experiments confirmed the replicative advantage of Omicron at the physiological temperature found in primary nasal epithelial cells.
## Omicron infection perturbs the epithelial barrier at physiological nasal temperature
The trans-epithelial electrical resistance (TEER) measured between the upper and lower compartment of transwell cultures was used to evaluate the integrity of the epithelial barrier. All the variants tested induced a decrease in TEER at D4 as compared to mockinfected epithelia (Fig. 2A andB), indicating that SARS-CoV-2 perturbs epithelial barrier function. The extent of TEER decrease was comparable for all variants tested at 37°C, with a median decrease between 1.5-and 1.9-fold (Fig. 2A). In contrast, at the lower temperature of 33°C, the Omicron BA.1 variant caused a more marked TEER decrease than the reference D614G variant (median decrease of 2.5× vs 1.3×; P = 0.003; Fig. 2B). Thus, the Omicron variant proved particularly damaging for nasal epithelial integrity at the physiologically relevant temperature of 33°C.
We next asked whether epithelial damage was associated with cell death by measur ing the apical release of LDH. An initial analysis of the kinetics of LDH detection showed that release of this enzyme was minimal at D2 and became clearly detectable at D4 (Fig. S4). Further measurements performed at D4 showed that all the variants tested induced an increase in LDH release (Fig. 2C andD), confirming the cytopathic effect of (E) The ratio of the number of viral RNA copies measured at 33°C to that measured at 37°C is shown at different dpi, with means and SD reported (n = 4 to 14 independent samples per point).(C, D, E) Statistical comparisons were done between the D614G reference and all other variants or the mock condition, using the Kruskal-Wallis test with Dunn's correction; ns: not significant; *P < 0.05; **P < 0.01; ***P < 0.001. In (E), the only significant differences in ratios were seen between BA.1 and D614G at 2 dpi and 4 dpi. SARS-CoV-2 infection. The extent of LDH release did not differ significantly between the variants tested at 37°C (Fig. 2C). LDH release was overall lower at 33°C (Fig. 2D). However, at this lower temperature, the Delta and Omicron BA.1 variants showed a significantly increased cytopathic effect compared to the reference D614G variant (P = 0.04 for Delta; P = 0.0006 for BA.1). It was noteworthy that the Delta variant showed a clear cytopathic effect in the absence of a replicative advantage in epithelial cultures, a phenomenon that may be ascribed to the high fusogenicity of this variant (6,37).
To further investigate the cytopathic effect induced by the different SARS-CoV-2 variants, we evaluated by immunofluorescence the expression of the cleaved active form of caspase-3, an apoptosis effector protein. Cleaved caspase-3 was induced by infection at D4, and more so at 37°C than at 33°C (Fig. 3A). Fluorescence quantifica tion showed that Omicron BA.1 was the most efficient at inducing cleaved caspase-3 at both temperatures (P < 0.05 in both cases), while Delta showed a nonsignificant trend for increase as compared to the D614G and Wuhan strains (Fig. 3B andC). The Spike and cleaved caspase-3 signals colocalized in some but not all infected cells, pointing to caspase-3 activation in infected cells but also possibly in bystander cells. It was interesting to note that large Spike+ cells corresponding to syncytia appeared more prominent for the Delta variant (Fig. 3A, third row), consistent with the highly fusogenic nature of this variant. Taken together, these findings documented a grada tion in the caspase-3-mediated cytopathic effect, with maximal induction by Omicron BA.1, intermediate induction by Delta, and lower induction by other variants. Of note, Omicron significantly perturbed epithelial integrity and viability even at the physiologi cal temperature of 33°C.
## Omicron infection induces the loss of motile cilia at physiological nasal temperature
We next evaluated the impact of the different viral variants on the motile cilia layer that covers airway epithelia. Reconstructed nasal epithelia collected at D4 post-infection were labeled for the acetylated form of α-tubulin, a tubulin isoform that is enriched in the axoneme of motile cilia. Immunofluorescence analysis showed that the three variants tested induced a loss of motile cilia at 37°C (Fig. 4A, left), with a decrease that was significant only for Omicron BA.1 (Fig. 4B, P = 0.026). The loss of cilia proved less marked at 33°C (Fig. 4A, right), with again a significant decrease seen only for BA.1 (Fig. 4C; P = 0.002).
Examination of the infected epithelia by scanning electron microscopy (SEM) confirmed the disruption of the motile cilia layer by infection at D4, with the presence of multiple rounded cells with partial or complete deciliation, predominantly seen at 37°C but also detectable at 33°C (Fig. 5). The observation of membrane protrusions surround ing cilia suggested that engulfment may contribute to cilia loss. SEM analysis at higher magnification showed the accumulation of viral particles at the surface of partially deciliated cells (Fig. S5). Many of the rounded deciliated cells appeared pushed to the surface of the epithelial layer, consistent with the extrusion of dying epithelial cells. Rounded imprints corresponding to missing cells were prominent in BA.1-infected epithelia (Fig. 5, bottom left image). The detection of blebbing cells also pointed to the occurrence of apoptosis. Overall, infection by D614G, Delta, and BA.1 perturbed the ciliary layer at 37°C, while changes were more prominent for the BA.1 variant at 33°C.
## Omicron induces a broad transcriptional downregulation of ciliary genes but only a limited upregulation of host defense genes at physiological nasal temperature
To identify the transcriptional changes that may contribute to epithelial perturbation, we performed bulk RNA-seq analysis on epithelia at D2 post-infection. This early time point was chosen to better identify transcriptional patterns that may cause later functional and morphological changes.
An initial analysis of viral gene reads showed that, at 37°C, viral expression was clearly detectable at D2 for the D614G, Delta, and Omicron BA.1 variants (Fig. 6A,top). The pattern of viral gene expression was comparable for the three variants, with N transcripts being the most abundant, followed by those coding for ORF1a and ORF1ab. The percentage of viral reads tended to be higher for BA.1 compared to D614G and Delta (Fig. 6B,top). It was notable that the percentage of viral reads reached above 30% of total mapped reads for epithelia infected by BA.1, highlighting how SARS-CoV-2 could usurp cellular resources for its own replication. The analysis of epithelia collected after 2 days of infection at 33°C showed a contrasting situation (Fig. 6A andB, bottom), with minimal detection of viral reads for the D614G and Delta variants, while the extent of BA.1 replication was high (26% of total mapped reads) and close to that observed at 37°C. The RNA-seq analysis thus confirmed the replicative advantage of the BA.1 variant at physiological nasal temperature.
DEGs were defined as cellular genes that showed at least a twofold change in expression upon infection with an adjusted P value < 0.05. The number of DEGs depen ded both on the nature of the infecting viral variant and on temperature (Fig. 6C). At 37°C, transcriptional dysregulation was low for D614G (n = 35 DEGs), moderate for Delta (n = 227), and marked for BA.1 (n = 3582). Thus, the impact of infection on cellular transcripts was strongly variant-dependent, while viral transcripts at the same time point differed only to a moderate extent between variants. The cumulative effects of slight differences in viral replication kinetics may contribute to this variable effect on cellular transcriptomes. At 33°C, the D614G and Delta variants induced only minimal changes in cellular gene expression (n = 5 DEGs for both viruses), while BA.1 caused clear cellular gene dysregulation (n = 253 DEGs), which remained however lower than that observed at higher temperature (Fig. 6C).
Interestingly, functional enrichment analysis revealed that the biological processes dysregulated by infection at D2 depended on the viral variant considered (Fig. S6). At 37°C, the top two gene ontology (GO) terms associated with transcriptional changes were "defense response" to viruses and symbionts for D614G and Delta, while the top two GO terms were "cilium organization" and "cilium assembly" for BA.1. At the lower temperature of 33°C, the same cilium-related GO terms were the most dysregulated for BA.1, while changes in biological processes were minimal for D614G and Delta (counts ≤3). Based on these findings, we analyzed more in depth the transcriptional changes for genes involved in cilia structure and function (Fig. 7A). In a volcano plot analysis, we compared the expressions of genes specific for ciliated cells and goblet cells as a balance between these two cell types had been previously reported in airway epithelia (38). The analysis confirmed that Omicron BA.1 infection at 37°C induced a marked downregula tion in the expressions of ciliated cell genes, with a concomitant upregulation in the expressions of goblet cell genes (Fig. 7A, right plot). In contrast, Delta infection caused only a slight upregulation of goblet cell gene expression, and D614G infection did not modulate ciliated nor goblet cellspecific genes at this early D2 time point (Fig. 7A, middle and left plots). The same analysis carried out on epithelial samples infected at 33°C for 2 days confirmed the decrease in ciliated cell gene expression and the increase in goblet cell gene expression induced by Omicron BA.1, while changes were not detected in these two gene sets upon D614G or Delta infection (Fig. S7A). Thus, Omicron BA.1 induced early D2 changes in transcriptional patterns that could account for the loss of motile cilia observed morphologically at the later D4 time point. The slower replication kinetics of D614G and Delta may explain the limited extent of transcriptional changes observed for these two variants at D2.
We then performed a volcano plot analysis for ISGs, which represent the major gene set modulated early during antiviral defense (Fig. 7B). Upon infection at 37°C, the three variants did induce ISG transcripts at D2, with a more marked upregulation for the BA. and Delta variants compared to D614G (Fig. 7B). In contrast, at 33°C, the induction of ISG transcripts remained minimal for Omicron and undetectable for the Delta and D614G variants (Fig. S7B). To further explore this temperature-dependent effect, we directly compared gene expression patterns induced by Omicron infection at 33°C and 37°C (Fig. 7C andD). These comparisons were not performed for D614G and Delta, which showed too few DEGs at 33°C. Omicron BA.1 was found to more efficiently downregulate ciliated cell genes and upregulate goblet cell genes at 37°C than 33°C (Fig. 7C), consistent with the trend for a higher motile cilia loss observed at 37°C by immunofluorescence (Fig. 4). The most significantly downregulated cilia genes observed at 37°C coded for DYNC2H1, a microtubule-associated dynein; TRAF3IP1, a microtubule-interacting protein associated with TRAF3; RPGR, a guanyl nucleotide exchange factor involved in ciliation; and WDR, a component of the intraflagellar transport complex. Thus, both structural and regulatory components of motile cilia were affected by Omicron infection. Analyzing temperature effects on antiviral defense gene expression showed that Omicron BA.1 infection induced a markedly stronger upregulation of ISGs at 37°C than at 33°C (Fig. 7D). The most significantly upregulated ISG at 37°C included the chemokine CXCL10, which attracts immune cellular effectors; the IFIT2 antiviral protein, which inhibits the expression of uncapped viral RNA; the OASL RNA-binding protein, which potentiates IFN production; and RSAD2 or viperin, an enzyme that produces an antiviral ribonucleotide. Multiple mechanisms targeting RNA virus replication were thus induced by Omicron at 37°C. Viral transcripts were represented on the same volcano plot to evaluate the association between the triggering factors (viral RNA) and the epithelial antiviral response (ISG expression). Interestingly, while the difference in ISG expression was marked at 37°C vs 33°C, the difference in BA.1 viral expression was minimal. Indeed, viral transcripts were grouped close to the origin of the volcano plot, pointing to a lack of significant changes in viral expression when comparing the two temperatures (Fig. 7D, orange symbols). These findings highlight the capacity of Omicron BA.1 to replicate efficiently at physiological nasal temperature while inducing only a limited antiviral defense reaction in the epithelium.
## Limited induction of IFNs by Omicron at physiological nasal temperature
To further evaluate the epithelial antiviral response upon SARS-CoV-2 infection, we directly measured the secretion of IFNs in apical culture supernatants at D2 and D4 (Fig. 8). Using high-sensitivity immunoassays (32,39), we detected an induction of both type I (IFN-β) and type III (IFN-λ1 and IFN-λ2/3) IFNs between D2 and D4 at 37°C, with a positive trend for all the variants tested (Fig. 8A, top row). Of note, the Delta variant induced the strongest IFN induction at D4, while IFN concentrations observed for Omicron BA.1 appeared intermediate, and those observed for the other variants Alpha and D614G remained low (IFNλ1 induction between D2 and D4: P<0.05 for Delta and for BA.1. IFNλ2/3 induction between D2 and D4: P < 0.01 for Delta; P < 0.05 for BA.1). A similar phenomenon was observed in epithelial cultures infected at 33°C (Fig. 8A, bottom row), with a more efficient induction of type I and type III IFNs by Delta compared to the other variants tested, including Omicron. Indeed, Delta was the only variant to induce a detectable production of IFN-β at physiological nasal temperature.
To compare the production of IFNs at the two temperatures tested, we computed the ratio of concentrations measured at 33°C to those measured at 37°C for the D4 timepoint (Fig. 8B). The production of IFNs proved consistently weaker at 33°C, as indicated by ratios below 1 for all the IFNs tested. For type III IFNs, the ratios tended to be higher for Delta and Omicron compared to Alpha. However, taken together, all the SARS-CoV-2 variants tested triggered lower antiviral responses at physiological nasal temperature.
An intervariant comparison of apical IFN concentrations measured at D4 at 37°C (Fig. S8A) confirmed that Delta was the only variant to induce a significantly higher IFN secretion compared to the reference variant D614G, independently of the IFN tested (P < 0.01 for IFN-β, IFN-λ1, and IFN-λ2/3). Omicron induced an intermediate IFN secretion that did not differ significantly from that measured for D614G. A similar pattern was observed at 33°C (Fig. S8B). To determine whether these findings applied to other IFNs, we measured the secretion of IFN-α at D4 in a subset of experiments (Fig. S8C). We observed a trend for a higher induction of IFN-α by Delta compared to Omicron at 37°C, consistent with findings obtained for other type I and type III IFNs. In contrast, no induction of IFN-α could be detected at 33°C for all the variants tested, suggesting that IFN-α induction was highly temperature-dependent.
It was intriguing that Omicron BA.1 induced only a moderate induction of IFNs at 33°C, while this variant had an early replicative advantage at this temperature. To further explore this point, we computed the ratio of IFN secreted at D4 reported to the number of inducing viral RNA copies measured at the early D2 time point. At 37°C, the Delta variant showed significantly higher IFN to viral RNA ratios compared to the D614G reference, while Omicron BA.1 showed ratios equivalent to those of D614G (Fig. 8C). Interestingly, at 33°C, Omicron BA.1 showed lower IFN to viral RNA ratios compared to D614G (P < 0.05 for IFNλ2/3), while a trend for higher ratios was maintained for Delta (P < 0.05 for IFNλ1). This analysis confirmed that Delta was a potent inducer of type I and type III IFNs. In contrast, Omicron BA.1 was characterized by a weak induction of IFNs relative to its replication capacity, especially at physiological nasal temperature.
## DISCUSSION
This study highlights that the SARS-CoV-2 Omicron variant evolved several traits that converged in facilitating viral spread. First, Omicron displayed an early replicative advantage in primary human nasal epithelial cells, which represent the main source of transmissible virus. Second, the increase in Omicron replication was more notable at 33°C, the temperature typically found in human nasal passages, resulting in a physio logically relevant replicative advantage. Third, while Omicron did not entirely disrupt the epithelial structure when replicating at 33°C, it did cause measurable decreases in epithelial integrity and in motile cilia coverage. These changes are likely to perturb the mechanism of mucociliary clearance, as we reported previously (32), and could thus facilitate the spread of viral particles within the airways. Fourth, at this physiological temperature, Omicron remained a moderate inducer of IFN responses relative to its replicative capacity in the nasal epithelium. Considering the key role of the type III IFN response in limiting the early spread of viruses at epithelial surfaces (40,41), it is relevant that Omicron induced comparatively less IFN-λ1 and IFN-λ2/3 than the Delta variant. The combined advantages of increased viral replication, perturbation of motile cilia, and escape from the innate antiviral response in the nasal epithelium likely contributed to the worldwide spread of the original Omicron BA.1 variant and to the continued dominance of Omicron-derived variants since 2022. While escape from the adaptive immune response clearly played an important role in Omicron emergence (8,9), our findings argue for an additional role of intrinsic viral properties in the evolutionary success of the Omicron variant.
The IFN response is a key determinant of COVID severity, as shown by the association of inborn errors in the type I IFN pathway and critical COVID pneumonia (42). The importance of the IFN pathway is also supported by the detection of auto-antibodies capable of neutralizing type I IFNs in up to 20% of patients who had fatal COVID (43). A decreased IFN response could be directly observed in the nasopharyngeal swabs from patients with severe COVID (38). In the face of the major selective pressure exerted by the IFN response, SARS-CoV-2 has evolved a variety of countermeasures, with multiple viral genes involved in dampening the induction of IFNs and/or inhibiting the function of ISGs (44,45). Indeed, SARS-CoV-2 was shown to be more efficient at escaping the innate antiviral response than common cold coronaviruses, helping explain why the former is more pathogenic (46).
Of interest, successive SARS-CoV-2 variants have shown an evolution in their capacity to escape the innate antiviral response. The early variant Alpha was shown to induce lower levels of IFNs than the original SARS-CoV-2 strain, due to increased expression of viral genes involved in innate immune antagonism, such as ORF6, ORF9b, and a truncated form of the nucleocapsid (47,48). Studies of later variants have yielded variable findings, with some studies showing an increasing capacity to escape the antiviral effects of type I IFNs from Alpha to Delta to Omicron (30), others showing a better escape for Alpha and Omicron compared to Delta (31,49), and still others reporting a more efficient escape for Delta compared to Omicron BA.1 (50). One reason for these divergent findings could be that some studies were performed on transformed cell lines such as Calu-3, in which Omicron shows an attenuated phenotype, in contrast to the fast-replicating phenotype observed in primary epithelial cells. Another confound ing factor may be the procedure used to titrate viral stocks, which often relies on serial dilutions in transformed cell lines such as Vero-E6-derived cells. As Omicron shows an attenuated phenotype in such cell lines, the resulting viral stocks tend to be underes timated in terms of infectivity for primary cells. To avoid these caveats, we chose to normalize the viral stocks based on viral RNA content, which is a trigger of the IFN response (51). Further, we chose to analyze viral replication kinetics and IFN induction in a human primary nasal cell-based model, which is more likely to yield information relevant to human physiology.
Based on this strategy, we observed that the induction of type I and type III IFNs in primary nasal epithelial cells was low for Alpha, intermediate for Omicron, and significantly higher for Delta. To consider the faster replication kinetics of Omicron, and hence the more rapid accumulation of viral RNA acting as a pathogen-associated molecular pattern (PAMP), we computed the ratio of produced IFN to early viral RNA copies. This analysis pointed to a relatively low intensity of the IFN response to Alpha and Omicron, compared to a significantly higher induction of type I and III IFN for Delta. Omicron may be qualified as a "stealthy" variant as it combines a high replicative capacity with a limited triggering of the IFN response at physiological nasal temperature. This does not imply an entire lack of epithelial antiviral response as a subset of ISGs were still induced by Omicron at 33°C but points to a dampening of the antiviral response as compared to previous SARS-CoV-2 variants. A possible reason for the strong innate response to Delta may be the highly fusogenic phenotype of this variant (6,37) as fusion is per se a triggering event for the IFN response (52). Fused syncytia are thought to have short life spans and to release damage-associated molecular patterns (DAMPs) when dying. It is thus relevant that we observed the formation of larger syncytia in the Delta-infected cultures, consistent with a highly fusogenic phenotype for this variant. A limitation of our study is that the reconstructed epithelium model used is devoid of immune cells, so we cannot evaluate the contribution of resident and inflammatory leukocytes to the mucosal IFN response. It is interesting, however, that an in vivo study of the innate response in nasopharyngeal swabs obtained during acute COVID found an inverse correlation between nasal ISG expression and COVID severity in patients infected with the ancestral and Delta variants, but not with the Omicron variant (53). These in vivo findings are compatible with a stronger impact of the IFN response on Delta than on Omicron infections, consistent with our findings in primary nasal epithelial cells. Also relevant is a recent study showing that later Omicron variants, including BA.4 and BA.5, have further increased their suppression of innate immunity compared to BA.1 and BA.2, pointing to a continued evolution of SARS-CoV-2 variants toward improved innate immune evasion (29).
All the variants tested could replicate at both 33°C and 37°C, consistent with the dual tropism of SARS-CoV-2 for the upper and lower respiratory tract (54)(55)(56). This contrasts with the preferential replication of common cold coronaviruses at 33°C, in line with their restriction to the upper respiratory tract (57). At the early D2 time point, the Omicron variant had already reached its replication plateau when grown at 33°C, while the induction of ISGs proved minimal or undetectable at this temperature. Furthermore, at D2, the production of type I and type III IFNs was not induced in Omicron-infected cultures grown at 33°C. These observations fit with the long-held notion that the IFN response is inefficient and delayed at lower temperature, mostly due to limited transcription of ISGs and IFN genes (58,59). Our data quantifying IFN-β and IFN-λ suggest that this temperature effect may also impact IFN protein secretion. It was interesting that Omicron showed an early replicative advantage over other SARS-CoV-2 variants, particularly in the low-temperature condition. This observation suggests that Omicron is not only efficient at evading the IFN response, but that it also possesses other fitnessenhancing properties independently of its innate response evasion capacity.
Structural features may contribute to Omicron's fitness at lower temperature as its Spike protein was shown to have better thermostability, with a tighter packing of the three receptor-binding domains in their down state (60,61). The viral attachment step is also likely involved as Omicron was proposed to have a higher affinity for its receptor ACE-2 (23,24) and a higher capacity to bind the plasma membrane of its target cells (25,26). The SARS-CoV-2 fusion step depends on lipid membrane reorganization and is known to be temperature-dependent, with a lower efficiency at 33°C than at 37°C (62). As the Omicron Spike was proposed by some authors to switch more easily to a fusogenic conformation, it is possible that Omicron better overcomes the temperaturedependent barrier to fusion (25). On the other hand, a more rapid switch to a fusogenic conformation may lead to viral inactivation by fusion with extracellular vesicles, which can act as traps for incoming virions (63). It is thus interesting that the release of extracellular vesicles is decreased at lower temperature, which may facilitate the spread of fusion-prone variants at 33°C. The nature of the Omicron entry pathway remains to be precisely defined as several groups have reported increased Omicron entry through a cathepsin-dependent endosomal pathway (27,28), while others have highlighted an increased dependence of Omicron entry on metalloproteases (15,64), and still others report a continued dependence on the surface protease TMPRSS2 to trigger Omicron Spike fusion (65). Further studies are thus warranted to precisely characterize the post-attachment steps involved in Omicron entry in primary cells and determine whether they contribute to the replicative advantage of this variant at physiological nasal temperature.
Of note, Omicron perturbed the integrity of the motile cilia layer more efficiently than other variants. The differences were marked at lower temperature, with a significant decrease in TEER values compared to the reference D614G variant, indicative of increased epithelial permeability. In addition, Omicron induced increased cell death and cilia loss at 33°C compared to other variants. Motile cilia loss can have several consequences on viral dissemination. At the cellular level, it can facilitate the dissemination of viral particles, which are released at the plasma membrane and will not have to cross a layer of packed cilia to reach neighboring cells and/or the mucus layer. The loss of motile cilia may also facilitate the passage of actin protrusions that push recently released SARS-CoV-2 virions toward the mucus layer (26). At the epithelium level, the loss of motile cilia inhibits mucociliary clearance as viral particles become trapped in a mucus layer that is not propelled anymore by ciliary beating (32). This may locally limit the spread of viral particles to neighboring ciliated cells as viral particles become less motile (66). On the other hand, at the organ level, the loss of mucociliary clearance prevents the removal of viral particles trapped in mucus and their redirection toward the pharynx (67,68). The loss of this clearance mechanism likely facilitates the viral invasion of epithelia situated lower in the respiratory tract (69) and thus promotes viral dissemination within the infected organism. Omicron outbreaks in Asia have been associated with a high rate of bacterial and fungal superinfections, which may also reflect an impairment of mucociliary clearance (70,71). It is interesting that the Omicron variant can perturb mucociliary clearance at the temperature found in nasal passages, suggesting that this property is physiologically relevant and beneficial to the evolutionary fitness of this variant.
In conclusion, the present study highlights multiple features that likely contributed to the efficient transmission and worldwide dissemination of Omicron. This SARS-CoV-2 variant appears particularly adapted to the upper respiratory tract based on its early replicative advantage at the temperature found in nasal passages, its capacity to perturb the motile cilia layer, and its limited induction of the IFN epithelial response. Further development of 3D-tissue models that include innate and adaptive immune cells should help achieve an integrated view of the parameters that control SARS-CoV-2 variant dissemination in the upper airways.
## MATERIALS AND METHODS
## SARS-CoV-2 infection of reconstructed human nasal epithelia
Reconstructed human nasal epithelial cultures (MucilAir) differentiated in vitro for at least 4 weeks were purchased from Epithelix (Saint Julien-en-Genevois, France). To minimize individual variability, each reconstructed epithelium was derived from nasal cells obtained from a pool of 14 donors. Epithelia generated from three distinct pools of donors were used in the present study. Information on the composition of these three pools (age and sex of donors) is provided in Table S1. Cultures were maintained in ALI conditions in transwells, with 700 μL of MucilAir medium (Epithelix) in the basal compartment and kept at 37°C or 33°C under a 5% CO 2 atmosphere. For SARS-CoV-2 infection, the apical side of ALI cultures was washed 20 min at 37°C or 33°C in Mucilair medium to remove mucus. Cells were then incubated with the equivalent of 10 6 RNA viral copies/µL of viral stock for each of the SARS-CoV-2 variants tested: Wuhan, D614G, Alpha, Beta, Gamma, Delta, and Omicron BA.1. The source of the viral isolates used is reported in Table S2. The viral input was diluted in DMEM to a final volume of 100 μL and left on the apical side for 4 h at 37°C or 33°C. Control wells were mock-treated with DMEM (Gibco) for the same duration. Viral inputs were removed by washing three times with 200 μL of phosphatebuffered saline (PBS) (5 min at 37°C or 33°C) and once with 200 μL MucilAir medium (20 min at 37°C or 33°C). The basal medium was replaced every 2-3 days. Apically released viruses were collected by incubation with 200 μL MucilAir medium (20 min at 37°C or 33°C) at 1, 2, and 4 dpi. The integrity of the epithelia was monitored by measuring the TEER at the same days. To do so, after apical wash, the transwells were transferred into a new 24-well plate and DMEM was added to both the apical (200 μL) and basal (700 μL) sides. The TEER was then measured using an Evom3 ohmmeter (World Precision Instruments). At day 4, cultures were fixed in paraformaldehyde 4% for 30 min, washed twice with PBS, and stored in PBS at 4°C until immunofluorescence staining.
## Viral RNA quantification
Culture apical supernatants collected at 1, 2, and 4 dpi were inactivated for 20 min at 80°C and then subjected to RT-qPCR analysis, using a Luna Universal One-Step RT-qPCR Kit (New England Biolabs), following the manufacturer's instructions. SARS-CoV-2 RNA was quantified in a final volume of 5 μL per reaction in 384-well plates using SARS-CoV-2 Nspecific primers (Forward 5′-CGA AGG TGT GAC TTC CAT G-3′; Reverse 5′-TAA TCA GAC AAG GAA CTG ATT A-3′) on a QuantStudio 6 Flex thermocycler (Applied Biosystems). A standard curve was performed in parallel using purified SARS-CoV-2 viral RNA (EURM019, Sigma Aldrich).
## S-Fuse infectivity assay
Infectious titers were quantified by automated imaging of cell-cell fusion in a GFP-split cell system. U2OS-ACE2 GFP1-10 and U2OS-ACE2-GFP11 cells, also termed S-Fuse cells, become GFP+ when they fuse together upon productive infection by SARS-CoV-2 (36). The S-Fuse cells tested negative for mycoplasma. S-Fuse cells were mixed (at a 1:1 GFP1-10/GFP11 ratio) and plated at 8 × 10 3 cells per well in a μClear 96-well plate (Greiner Bio-One). The tested SARS-CoV-2 infection supernatants were serially diluted and added to S-Fuse cells. Post 18 h, cells were fixed with 2% paraformaldehyde (electron microscopy # 15714-S), washed in PBS, and stained with Hoechst at a 1:1,000 dilution (Invitrogen, H3570). Images were acquired automatically using an Opera Phenix high-content confocal microscope (PerkinElmer). The GFP+ area, the number of GFP+ objects, and the number of nuclei per well were quantified using the Harmony software (PerkinElmer). The S-fuse assay has been previously validated for the comparison of SARS-CoV-2 variant infectivities and antibody neutralization capacities (9,35).
## Immunofluorescence labeling
Membranes supporting the fixed MucilAir cultures were cut in four using a scalpel blade. Membrane pieces were placed in 10 μL drops of PBS onto parafilm and then permeabilized in PBS 0.5% Triton-X100 for 20 min at room temperature (RT). Samples were blocked in PBS with 0.1% Tween, 1% bovine serum albumin (BSA), 10% fetal bovine serum, and 0.3 M glycine for 30 min at RT. Samples were then incubated overnight at 4°C with the following antibodies: pan-SARS-CoV-2 anti-Spike mAb102 human monoclo nal antibody (72) at 1 μg/mL; rabbit anti-cleaved caspase 3 primary antibody (9661, Cell Signaling; 1:100 dilution); AF647-conjugated rabbit anti-alpha acetylated tubulin (ab218591, Abcam; 1:100 dilution). After washing, a secondary anti-human IgG antibody conjugated to AF555 (A-21422, Thermo Fisher Scientific) was added at 1:400 for 1 h at RT. Primary and secondary antibodies were diluted in PBS with 0.1% Tween and 1% BSA. Samples were counterstained with Hoechst and mounted in FluoromountG (Thermo Fisher Scientific) before observation with an LSM 710 confocal microscope (Zeiss) or with a BC43 spinning disk confocal microscope (Andor).
## LDH cytotoxicity assay
Diluted apical culture supernatants (1:25) were pretreated with Triton-X100 1% for 2 h at RT for viral inactivation. LDH dosage was performed using the LDH-Glo Cytotoxic ity Assay kit (Promega) following the manufacturer's instructions. Luminescence was measured using an EnSpire luminometer (PerkinElmer).
## Cytokine measurements
Apical culture supernatants were pretreated with Triton-X100 1% (final concentration) for 2 h at RT for viral inactivation. IFN-β protein was quantified using Simoa digi tal ELISA technology as previously described (32,39). Alternatively, IFN-β and IFN-α were quantified simultaneously in a Simoa multiplexed assay. Briefly, the inactivated supernatant was diluted 1:10 in Diluent B (Quanterix, Billerica, MA, USA) containing 1% Triton-X100 (final concentration). IFN-α and IFN-β concentrations were quantified with a Simoa assay developed on an HD-X instrument with Quanterix Homebrew kits. The assay configuration was a 2-step triplex associating IFN-α, IFN-β, and IFN-λ3 (data not shown), using 150 pM SBG and 50 uL RGP reagents. The IFN-α plex quantifies all IFN-α subtypes with a similar sensitivity, as previously reported (39), with results expressed in IFN-α17 equivalents. The antibodies used were cloned from APECED/APS1 patients. The 8H1 antibody clone was used as a capture antibody after coating on paramagnetic beads (0.3 mg/mL); the 12H5 clone was biotinylated (biotin/antibody ratio = 30:1) and used as the detector at a concentration of 0.3 µg/mL. The recombinant human IFN-α17 protein (#11150, PBL Assay Science, Piscataway, NJ, USA) was used as a reference to quantify concentrations. The limit of detection was between 0.59 and 1.43 fg/mL equivalent IFN-α17, including the dilution factor. The IFN-β plex assay is described more in detail in 73. Briefly, the 710906-9 IgG1-κ mouse monoclonal antibody (PBL Assay Science) was used as a capture antibody to coat paramagnetic beads (0.3 mg/mL); the 710323-9 IgG1-κ mouse monoclonal antibody (PBL Assay Science) was biotinylated (biotin/anti body ratio = 40/1) and used as the detector antibody at a concentration of 0.3 µg/mL. The recombinant IFN-β protein (#11415, PBL Assay Science) was used as reference to quantify concentrations. The limit of detection was between 0.01 and 0.03 pg/mL IFN-β, including the dilution factor.
## Quantification of IFN-λ1 and IFN-λ2/3 proteins using Legendplex technology
On the day of assaying, supernatants were centrifuged at 300 g for 5 min. The concen tration of 13 different analytes was measured using the LEGENDplex Human Anti-Virus Response Panel (13-plex) with a V-bottom plate kit (cat. No. 740390), and the manufac turer's protocol was followed (BioLegend, San Diego, California, USA). Plates were read on a BD Fortessa II 5-color flow cytometer (Franklin Lakes, New Jersey, USA), and the resulting mean fluorescence intensity data were analyzed using the BioLegend portal to predict the analyte concentration (https://legendplex.qognit.com/).
## Scanning electron microscopy
PFAfixed epithelial samples were postfixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer for 1 h at RT. Samples were then washed in 0.1 M cacodylate buf fer and several times in water and then incubated in 1% osmium tetroxide for 1 h. After dehydration by incubation in increasing concentrations of ethanol (35%, 70%, 85%, 95%, and 100%), samples were treated with hexamethyldisilazane for 10 min for chemical critical point drying. Specimens were then sputter-coated with a gold/ palladium conductive layer up to 9 Å using a gun ionic evaporator PEC 682. Images were acquired on a JEOL JSM 6700F field emission SEM operated at 7 kV.
## References
1. Yajima, Nomai, Okumura et al. "Genotype to Phenotype Japan (G2P-Japan) Consortium. 2024. Molecular and structural insights into SARS-CoV-2 evolution: from BA.2 to XBB subvariants" *mBio*
2. Carabelli, Peacock, Thorne et al.
3. Genomics, Consortium, Peacock et al. (2023) "SARS-CoV-2 variant biology: immune escape, transmission and fitness" *Nat Rev Microbiol*
4. Korber, Fischer, Gnanakaran et al. (2020) "Tracking changes in SARS-CoV-2 spike: evidence that D614G increases infectivity of the COVID-19 Virus" *Cell*
5. Saito, Irie, Suzuki et al. (2022) "Enhanced fusogenicity and pathogenicity of SARS-CoV-2 Delta P681R mutation" *Nature New Biol*
6. Twohig, Nyberg, Zaidi et al. (2022) "Hospital admission and emergency care attendance risk for SARS-CoV-2 delta (B.1.617.2) compared with alpha (B.1.1.7) variants of concern: a cohort study" *Lancet Infect Dis*
7. Mlcochova, Kemp, Dhar et al. (2021) "SARS-CoV-2 B.1.617.2 Delta variant replication and immune evasion" *Nature*
8. Viana, Moyo, Amoako et al. (2022) "Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa" *Nature*
9. Liu, Wu, Gao (2025) "A structural voyage toward the landscape of humoral and cellular immune escapes of SARS-CoV-2" *Immunol Rev*
10. Planas, Saunders, Maes et al. (2022) "Considerable escape of SARS-CoV-2 Omicron to antibody neutralization" *Nature*
11. Van Dorp, Goldberg, Ke et al. (2022) "Global estimates of the fitness advantage of SARS-CoV-2 variant Omicron" *Virus Evol*
12. Puhach, Meyer, Eckerle (2023) "SARS-CoV-2 viral load and shedding kinetics" *Nat Rev Microbiol*
13. Hui, Ho, Cheung et al. (2022) "SARS-CoV-2 Omicron variant replication in human bronchus and lung ex vivo" *Nature New Biol*
14. Tanneti, Patel, Tan et al. (2026) *Full-Length Text Journal of Virology*
15. Weiss (2024) "Comparison of SARS-CoV-2 variants of concern in primary human nasal cultures demonstrates Delta as most cytopathic and Omicron as fastest replicating" *mBio*
17. Barut, Halwe, Kelly et al. (2022) "The spike gene is a major determinant for the SARS-CoV-2 Omicron-BA.1 phenotype" *Nat Commun*
18. Shi, Li, Lai et al. (2024) "Omicron Spike confers enhanced infectivity and interferon resistance to SARS-CoV-2 in human nasal tissue" *Nat Commun*
19. Pascall, Vink, Blacow et al. (2023) "Directions of change in intrinsic case severity across successive SARS-CoV-2 variant waves have been inconsistent" *Journal of Infection*
20. Wolter, Jassat, Walaza et al. (2022) "Early assessment of the clinical severity of the SARS-CoV-2 omicron variant in South Africa: a data linkage study" *The Lancet*
21. Rosenke, Griffin, Kaiser et al. (2025) "Pathogenesis of bovine H5N1 clade 2.3.4.4b infection in macaques"
22. Armando, Beythien, Kaiser et al. (2022) "SARS-CoV-2 Omicron variant causes mild pathology in the upper and lower respiratory tract of hamsters" *Nat Commun*
23. De Melo, Perraud, Alvarez et al. (2023) "Neuroinvasion and anosmia are independ ent phenomena upon infection with SARS-CoV-2 and its variants" *Nat Commun*
24. Masui, Hashimoto, Matsumura et al. (2024) "Micro-patterned culture of iPSC-derived alveolar and airway cells distinguishes SARS-CoV-2 variants" *Stem Cell Reports*
25. Lamers, Mykytyn, Breugem et al. (2022) "SARS-CoV-2 Omicron efficiently infects human airway, but not alveolar epithelium" *bioRxiv*
26. Mccallum, Czudnochowski, Rosen et al. (2022) "Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement" *Science*
27. Hoffmann, Krüger, Schulz et al. (2022) "The Omicron variant is highly resistant against antibody-mediated neutralization: Implications for control of the COVID-19 pandemic" *Cell*
28. Parsons, Acharya (2023) "Evolution of the SARS-CoV-2 Omicron spike" *Cell Rep*
29. Wu, Lidsky, Xiao et al. (2023) "SARS-CoV-2 replication in airway epithelia requires motile cilia and microvillar reprogramming" *Cell*
30. Meng, Abdullahi, Ferreira et al. (2022) "Altered TMPRSS2 usage by SARS-CoV-2 Omicron impacts infectivity and fusogenicity" *Nature New Biol*
31. Willett, Grove, Maclean et al. (2022) "SARS-CoV-2 Omicron is an immune escape variant with an altered cell entry pathway" *Nat Microbiol*
32. Reuschl, Thorne, Whelan et al. (2024) "Evolution of enhanced innate immune suppression by SARS-CoV-2 Omicron subvariants" *Nat Microbiol*
33. Guo, Barrett, Morrison et al. (2022) "Interferon resistance of emerging SARS-CoV-2 variants" *Proc Natl Acad Sci*
34. Nchioua, Schundner, Klute et al. (2023) "Reduced replication but increased interferon resistance of SARS-CoV-2 Omicron BA.1" *Life Sci Alliance*
35. Robinot, Hubert, De Melo et al. (2021) "SARS-CoV-2 infection induces the dedifferentia tion of multiciliated cells and impairs mucociliary clearance" *Nat Commun*
36. Elad, Wolf, Keck (2008) "Air-conditioning in the human nasal cavity" *Respiratory Physiology & Neurobiology*
37. Keck, Leiacker, Riechelmann et al. (2000) "Temperature profile in the nasal cavity" *Laryngoscope*
38. Planas, Bruel, Grzelak et al. (2021) "Sensitivity of infectious SARS-CoV-2 B.1.1.7 and B.1.351 variants to neutralizing antibodies" *Nat Med*
39. Buchrieser, Dufloo, Hubert et al. (2020) "Syncytia formation by SARS-CoV-2-infected cells" *EMBO J*
40. Rajah, Hubert, Bishop et al. (2021) "SARS-CoV-2 Alpha, Beta, and Delta variants display enhanced Spike-mediated syncytia formation" *EMBO J*
41. Ziegler, Miao, Owings et al. (2021) "Impaired local intrinsic immunity to SARS-CoV-2 infection in severe COVID-19" *Cell*
42. Rodero, Decalf, Bondet et al. (2017) "Detection of interferon alpha protein reveals differential levels and cellular sources in disease" *J Exp Med*
43. Stanifer, Pervolaraki, Boulant (2019) "Differential regulation of Type I and Type III interferon signaling" *Int J Mol Sci*
44. Sposito, Broggi, Pandolfi et al. (2021) "The interferon landscape along the respiratory tract impacts the severity of COVID-19" *Cell*
45. Zhang, Bastard, Covid Human Genetic et al. (2022) "Human genetic and immunological determinants of critical COVID-19 pneumonia" *Nature*
46. Bastard, Gervais, Voyer et al. (2021) "Autoanti bodies neutralizing type I IFNs are present in ~4% of uninfected individuals over 70 years old and account for ~20% of COVID-19 deaths" *Sci Immunol*
47. (2026) *Full-Length Text Journal of Virology*
48. Pen, Rice (2024) "The antiviral state of the cell: lessons from SARS-CoV-2" *Curr Opin Immunol*
49. Caobi, Saeed (2024) "Upping the ante: enhanced expression of interferon-antagonizing ORF6 and ORF9b proteins by SARS-CoV-2 variants of concern" *Curr Opin Microbiol*
50. 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*
51. Mears, Young, Sanderson et al. (2025) "Emergence of SARS-CoV-2 subgenomic RNAs that enhance viral fitness and immune evasion" *PLOS Biol*
52. Thorne, Bouhaddou, Reuschl et al. (2022) "Evolution of enhanced innate immune evasion by SARS-CoV-2" *Nature*
53. Laine, Skön, Väisänen et al. (2022) "SARS-CoV-2 variants Alpha, Beta, Delta and Omicron show a slower host cell interferon response compared to an early pandemic variant" *Front Immunol*
54. Bouhaddou, Reuschl, Polacco et al. (2023) "SARS-CoV-2 variants evolve convergent strategies to remodel the host response" *Cell*
55. Thorne, Reuschl, Zuliani-Alvarez et al. (2021) "SARS-CoV-2 sensing by RIG-I and MDA5 links epithelial infection to macrophage inflammation" *EMBO J*
56. Decalf, Desdouits, Rodrigues et al. (2017) "Sensing of HIV-1 entry triggers a type I interferon response in human primary macrophages" *J Virol*
57. Walsh, Miao, Owings et al. "2025. Variants and vaccines impact nasal immunity over three waves of SARS-CoV-2" *Nat Immunol*
58. Laporte, Raeymaekers, Van Berwaer et al. (2021) "The SARS-CoV-2 and other human coronavirus spike proteins are finetuned towards temperature and proteases of the human airways" *PLoS Pathog*
59. Resnick, Beer, Pekosz (2023) "Early transcriptional responses of human nasal epithelial cells to infection with influenza A and SARS-CoV-2 virus differ and are influenced by physiological temperature" *Pathogens*
60. V'kovski, Gultom, Kelly et al. (2021) "Disparate temperature-dependent virus-host dynamics for SARS-CoV-2 and SARS-CoV in the human respiratory epithelium" *PLoS Biol*
61. Otter, Tan, Khosla et al. (2023) "Infection of primary nasal epithelial cells differentiates among lethal and seasonal human coronaviruses" *Proc Natl Acad Sci*
62. Lane, Dunn, Gardner et al. (2018) "The efficacy of the interferon alpha/beta response versus arboviruses is temperature dependent" *mBio*
63. Heron, Berg (1978) "The actions of interferon are potentiated at elevated temperature" *Nature*
64. Wang, Anang, Iketani et al. (2022) "Functional properties of the spike glycoprotein of the emerging SARS-CoV-2 variant B.1.1.529" *Cell Rep*
65. Gobeil, Henderson, Stalls et al. (2022) "Structural diversity of the SARS-CoV-2 Omicron spike" *Mol Cell*
66. Dufloo, Sanjuán (2024) "Temperature impacts SARS-CoV-2 spike fusogenicity and evolution" *mBio*
67. Bisht, Velthuis (2022) "Decoding the role of temperature in RNA virus infections" *mBio*
68. Chan, Huang, Hu et al. (2023) "Altered host protease determinants for SARS-CoV-2 Omicron" *Sci Adv*
69. Mykytyn, Breugem, Geurts et al. (2023) "SARS-CoV-2 Omicron entry is type II transmembrane serine protease-mediated in human airway and intestinal organoid models" *J Virol*
70. Becker, Martin-Sancho, Simons et al. (2024) "Live imaging of airway epithelium reveals that mucociliary clearance modulates SARS-CoV-2 spread" *Nat Commun*
71. Vijaykumar, Leung, Barrios et al. (2023) "COVID-19 causes ciliary dysfunction as demonstrated by human intranasal micro-optical coherence tomography imaging" *Am J Respir Cell Mol Biol*
72. Li, Vijaykumar, Phillips et al. (2023) "Mucociliary transport deficiency and disease progression in Syrian hamsters with SARS-CoV-2 infection" *JCI Insight*
73. Bustamante-Marin, Ostrowski (2017) "Cilia and mucociliary clearance" *Cold Spring Harb Perspect Biol*
74. Murakami, Nozaki, Morosawa et al. (2024) "Difference in the impact of coinfections and secondary infections on antibiotic use in patients hospitalized with COVID-19 between the Omicron-dominant period and the pre-Omicron period" *J Infect Chemother*
75. Cheng, Li, Wu et al. (2024) "Clinical and pathogen features of COVID-19associated infections during an Omicron strain outbreak in Guangzhou" *China. Microbiol Spectr*
76. Planchais, Fernández, Bruel et al. (2022) "Potent human broadly SARS-CoV-2 neutralizing IgA and IgG antibodies effective against Omicron BA.1 and BA.2" *Immunology*
77. Llibre, Bilek, Bondet et al. (2019) "Plasma type I IFN protein concentrations in human tuberculosis" *Front Cell Infect Microbiol*
78. Andrews (2010) "FastQC A quality control tool for high throughput sequence data"
79. Chen, Zhou, Chen et al. (2018) "Fastp: an ultra-fast all-in-one FASTQ preprocessor" *Bioinformatics*
80. Kim, Paggi, Park et al. (2019) "Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype" *Nat Biotechnol*
81. Tarasov, Vilella, Cuppen et al. (2015) "Sambamba: fast processing of NGS alignment formats" *Bioinformatics*
82. (2026) *Full-Length Text Journal of Virology*
83. Liao, Smyth, Shi (2014) "featureCounts: an efficient general purpose program for assigning sequence reads to genomic features" *Bioinformatics*
84. Wu, Hu, Xu et al. "A universal enrichment tool for interpreting omics data" *Innovation (Camb)*
85. Chen, Tan, Kou et al. (2013) "Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool" *BMC Bioinformatics*
86. Wickham (2016) "Ggplot2: elegant graphics for data analysis (3e)" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724362&blobtype=pdf | # Neuropilin-1 expression modulates infection susceptibility to murine cytomegalovirus at the materno-fetal interface
Luís Fonseca Brito, Eléonore Ostermann, Julian Kottlau, Silvia Tödter, Renke Brixel, Roland Schüle, Maria Solano, Wolfram Brune, Felix Stahl, Rolf Felix, Stahl
## Abstract
Human cytomegalovirus (HCMV) is the leading infectious cause of congenital disease, but the mechanisms governing vertical transmission remain poorly defined. Murine cytomegalovirus (MCMV) infection in mice is a widely used model to study pathogenesis in vivo. However, the mouse model cannot be used to study congenital infection as viral transmission from mother to fetus in utero does not occur. In this study, we investigated tissue-specific features of the murine materno-fetal barrier that may restrict viral infection. Following high-dose intravenous challenge of wild-type and severely immunocompromised pregnant mice, MCMV replicated efficiently in the maternal liver but infected only a few cells in the placenta, suggesting an intrinsic resistance. Ex vivo analyses of primary placental cells, trophoblast stem cells, and a trophoblast cell line revealed a low susceptibility to MCMV infection compared to other permissive cell types. The resistance of trophoblast cells to MCMV infection correla ted with the absence of neuropilin-1 (NRP1), a cellular receptor required for efficient infection of several cell types. Enforced expression of NRP1 in trophoblast cells increased their susceptibility to MCMV infection and replication, suggesting that the resistance of trophoblast cells to MCMV infection is caused by the lack of a critical receptor. These findings further suggest that, in addition to immune-mediated restrictions, cell-intrinsic resistance limits MCMV infection and transmission at the murine materno-fetal interface.IMPORTANCE Congenital cytomegalovirus (CMV) infection is a major cause of developmental disabilities in newborns, yet the biological factors that influence transmission from mother to fetus remain unclear. In this study, we demonstrate that trophoblast cells of the murine placenta are naturally resistant to CMV infection as they lack expression of a host protein, neuropilin-1 (NRP1), that the virus requires for entry. By introducing this protein into resistant cells, we demonstrated that susceptibility to infection can be reinstated, indicating that the absence of NRP1 plays a key protective role at the materno-fetal barrier. These results provide insight into why mice rarely transmit CMV to their offspring and how species-specific differences in placental biology shape susceptibility. Understanding these mechanisms will aid in refining animal models and may help identify new targets to prevent congenital infection in humans.
HCMV-infected pregnant women and origin of clinical symptoms are currently under investigation and likely include both host defense mechanisms and viral pathoge nicity factors.
Several animal models are applied to study cytomegalovirus (CMV)-host interaction. The primary advantage of the rhesus macaque CMV and guinea pig CMV models for studying CMV pathogenesis lies in their capacity to replicate fetal infection within their respective species (2)(3)(4). Nevertheless, the use of these models is restricted to a limited number of research centers, primarily due to ethical restraints, the complexity of animal husbandry, and the limited availability of species-specific tools. As a result, murine cytomegalovirus (MCMV) infection in mice remains the most widely used in vivo model for CMV research. However, vertical transmission in utero does not occur in the commonly used inbred mice and laboratory MCMV strains (5)(6)(7)(8)(9). Infection of fetal membranes has been observed in severe combined immunodeficient (SCID) mice (10) and the offspring of latently infected mice (11). Therefore, current models for in utero MCMV infection involve invasive procedures such as direct injection of virus into the placenta or the fetal brain (12)(13)(14). Alternatively, infection of neonatal mice with MCMV is a well-established surrogate model for congenital CMV disease (12,13,15).
The materno-fetal interface is the site defined by the implantation and invasion of the blastocyst-derived placenta into the uterine stroma, the decidua. The placenta serves not only as a conduit for nutrients and waste but also as a critical immunological barrier that protects the developing fetus from maternal pathogens (16). Although the structural and organizational features of the human and mouse placenta differ, each species has evolved to maintain an optimized immunological equilibrium essential for sustaining a healthy pregnancy (17). Thus, mechanisms of placental resistance-structural, cellular, and immunological-are likely involved in limiting vertical transmission of MCMV. A range of host molecules, including epidermal growth factor receptor, integrin αvβ3, CD90, CD147, neuropilin-2 (NRP2), and platelet-derived growth factor receptor alpha (PDGFRα), have been characterized as receptors or co-receptors mediating HCMV entry into fibroblasts, epithelial cells, and endothelial cells (18)(19)(20)(21)(22)(23). Less is known about host receptors relevant to MCMV infection. NRP1 and the major histocompatibility complex I (MHC-I) were described as crucial host factors for MCMV infection of endothelial cells and macrophages, respectively (24,25). The roles of other putative receptors, such as Integrin β1 subunit (ITGβ1) and PDGFRα, remain to be defined (26,27).
Here, we addressed tissue-specific features of the murine materno-fetal barrier that could play a role in interfering with vertical MCMV transmission. Infection events were tracked in the materno-fetal barrier lining cells, including trophoblast cells, in vivo and ex vivo. A trophoblast cell line and murine trophoblast stem cells (mTSCs) were used to identify tissue-specific host factors for MCMV replication. The results of this study suggest that a combination of both host immunity and intrinsic properties of tissue cells contributes to MCMV resistance in the materno-fetal interface.
## RESULTS
## Tissue-specific MCMV replication in placenta and maternal liver under conditions of immune suppression
Lymphocytes, such as T and natural killer (NK) cells, as well as type I interferons, are well established to interfere with MCMV infection. Rag2 -/-Il2rg -/-mice are deficient in functional B, T, and NK cells, while Ifnar1 -/-mice lack a crucial subunit of the type-I IFN receptor, resulting in defective interferon signaling. These models of immunodeficiency exhibit augmented MCMV replication in Rag2 -/-Il2rg -/- (28,29) or Ifnar1 -/-(30) mice. To address whether impaired immunity impacts viral replication in placental tissue, Rag2 -/-Il2rg -/-and Ifnar1 -/-or wild-type (WT) pregnant mice were infected intra venously (i.v.) with MCMV-3DR, here referred to as MCMV, which is a recombinant expressing mCherry and Gaussia luciferase (see Materials and Methods). The biological fitness of this virus has been demonstrated in several in vivo studies (24,28,31,32) and is used throughout the present investigation. The dams were sacrificed on 3-9 days post-infection (dpi) to prevent a reduced health status normally occurring in Rag2 -/-Il2rg -/-and Ifnar1 -/-pregnant mice upon infection with high MCMV doses. The average observation time after infection was 5.3 days (Table 1). Maternal livers and conceptuses (embryo/fetus, placenta, and associated membranes) were investigated to monitor infection events in tissue by means of mCherry expression in single cells.
This method holds the advantage of minimizing false-positive results, which are more likely to occur with nucleic acid amplification techniques or luciferase activity assays due to contamination with maternal blood. Expression of MCMV-encoded mCherry and Gaussia luciferase correlates with virus titers in vitro (31) and in vivo (28). Histology sections of maternal livers were used to establish a standardized algorithm (see Materials and Methods) to analyze the frequency of mCherry + MCMV-infected cells per DAPI + nuclei to estimate the magnitude of virus load in tissue (Fig. 1A). Both Rag2 -/-Il2rg -/- and Ifnar1 -/-mice exhibited more infected cells in the liver as compared to WT mice, confirming that reduced host immunity allowed increased virus replication in pregnant mice (Fig. 1B). In contrast, histological analysis of conceptuses revealed low numbers of MCMV-infected cells in the visceral yolk sac and placenta (Fig. 1C andD). These data suggested that, unlike the liver, the placenta exhibited biological resistance to MCMV infection.
## Placental and decidual cells exhibit intrinsic resistance to MCMV infection ex vivo
The materno-fetal interface consists of the blastocyst-derived placenta and the decidua. The placenta is composed of successive layers of trophoblast cells, followed by fetal endothelial capillaries that separate the maternal from the fetal blood circulation (33). To assess the susceptibility of materno-fetal interface-lining cells to MCMV infection, placental and decidual cells were obtained from the progeny of WT dams mated to homozygous actb-Egfp male mice. The offspring and placentas generated by this mating strategy are heterozygous for actb-Egfp, exhibit ubiquitous GFP expression, and thus allow for distinction between maternal decidual stroma and placenta cells. Cell suspen sions obtained from the decidua-lined placenta were analyzed for GFP and CD45 expression to classify cells as fetal/placental (GFP + ) or maternal (GFP -) and hematopoietic (CD45 + ) or non-hematopoietic (CD45 -) cells (Fig. 2A andB). As the placenta invades the maternal decidua at the implantation site, in these tissue preparations, cells of maternal origin outnumbered those of fetal origin, including both hematopoietic and nonhematopoietic cells (Fig. 2C). There was a distinct CD45 -placental cell population but only very few fetal hematopoietic cells (Fig. 2C). These primary cell suspensions were infected with MCMV directly after isolation and analyzed 1 day post-infection for MCMVencoded mCherry expression. A small fraction of the cultured cells was susceptible to MCMV infection (Fig. 2D andE). MCMV-infected cells almost exclusively belonged to the placental non-hematopoietic cell population, whereas very few maternal decidual cells were mCherry + (Fig. 2F andG). Hence, ex vivo decidua and placenta cells exhibited a significant resistance to MCMV as observed in vivo in WT and immunocompromised animals.
## Placental SM9-1 cells display marked resistance to MCMV compared with diverse cell lines
To assess the relative susceptibility of placental cells to MCMV infection, the SM9-1 placental trophoblast cell line (34) was comparatively analyzed alongside a panel of cell lines derived from diverse mouse tissues. Cell lines that were described to be permissive for MCMV or even used to propagate the virus to generate virus stocks, such as the 10.1 murine embryo fibroblasts (35), M2-10B4 bone marrow stromal cells (36), SVEC4-10 lymphatic endothelial cells (37), and NMuMG mammary gland epithelial cells (38), were included in the analysis. In addition, the Hepa1-6 liver cell line (39) was added as MCMV infection of the liver was in distinct contrast to placental tissue in vivo (Fig. 1). High and robust expression of MCMV-encoded Gaussia luciferase was found in cell culture supernatants of all but the SM9-1 cell lines after 1 and 2 days of MCMV infection (Fig. 3A). On average, MCMV infection of known permissive cells led to a 40-fold and a 141-fold higher luciferase activity than infection of SM9-1 cells at 24 h and 48 h post-infection, respectively. Similarly, high percentages of mCherry + cells were detected in all cell types except for SM9-1 cells at 2 dpi (Fig. 3B). Only a small fraction of SM9-1 cells expressed the MCMV-encoded fluorophore (Fig. 3C). These data indicated that SM9-1 trophoblast cells are poorly infectible and therefore relatively resistant to MCMV infection.
## Absence of NRP1 in SM9-1 trophoblast cells correlates with resistance to MCMV infection
MCMV infection is established through successful viral entry into host cells and ulti mately leads to the production of infectious progeny. To understand potential mecha nisms of resistance to MCMV, SM9-1 trophoblast cells were analyzed for surface expression of the putative MCMV entry receptors ITGβ1, PDGFRα, β2-microglobulin (β2-M) as part of the MHC-I, and NRP1. SM9-1 trophoblast cells exhibited membrane expression of ITGβ1 and PDGFRα, whereas β2-M was barely detectable, and NRP1 was not expressed (Fig. 4A). NRP1 expression could be confirmed in MCMV-susceptible cell lines 10.1, M2-10B4, SVEC4-10, NMuMG, and Hepa1-6 cells, but not in SM9-1 trophoblast cells (Fig. 4B). NRP1 is a transmembrane glycoprotein involved in the nervous and vascular system during embryogenesis (40) and was identified as a critical host factor for MCMV infection (25).
In a reanalysis of a recently reported single-nucleus RNA sequencing data set of gestational day 12.5 murine placental cells (41), Itgb1 was ubiquitously expressed in all cell types. Pdgfra and some B2m expression were found in decidual stroma cells and trophoblasts. In contrast, Nrp1 was found to be expressed only in fetal mesenchymal and endothelial cells but virtually absent in decidual stroma and placental trophoblast cells (Fig. S1). Hypothesizing that the lack of NRP1 expression in SM9-1 cells might be responsible for the resistance to MCMV infection, the cells were transfected with plasmids expressing GFP or NRP1-GFP and subsequently infected with MCMV. NRP1 protein was detected in cells transfected with NRP1-GFP but not in cells transfected with GFP or untransfected SM9-1 cells (Fig. 4C). SM9-1 cells transfected with NRP1-GFP were more susceptible to MCMV infection than cells transfected with GFP (Fig. 4D andE), leading to a substantially higher percentage of mCherry + cells at 1 dpi (Fig. 4F). These results suggested that the resistance of SM9-1 trophoblast cells to MCMV infection may be due to a lack of NRP1 expression.
## NRP1 expression overcomes the resistance of SM9-1 trophoblast cells to MCMV
As plasmid transfection allowed vigorous but only transient overexpression of NRP1 (Fig. 3), the impact of stable NRP1 expression on the infectability of SM9-1 trophoblast cells was tested. To do this, NRP1-GFP and GFP coding sequences were inserted into a lentiviral vector that expresses BFP and a puromycin resistance marker. SM9-1 cells were transduced with these lentiviral vectors and selected with puromycin. Using the NRP1-GFP-encoding construct for transduction led to a lower frequency of BFP-expressing cells than the GFP-encoding control (Fig. 5A andB). NRP1 cell surface expression was detected by flow cytometry in cells transduced with NRP1-GFP but not in cells transduced with GFP (Fig. 5C andD). SM9-1 cells transduced with NRP1-GFP were more susceptible to MCMV infection than cells transduced with GFP (Fig. 5E), leading to a 17-fold increase in mCherry + cells at 1 dpi (Fig. 5F). To test whether NRP1-expressing SM9-1 cells are permissive for MCMV replication, we used transduced and non-transduced SM9-1 cells and 10.1 fibroblasts for multistep replication kinetics. As expected, 10.1 fibroblasts were highly permissive and produced high titers of viral progeny (Fig. 5H). By contrast, SM9-1 cells transduced with GFP or non-transduced produced relatively low MCMV titers. However, NRP1-GFP-expressing SM9-1 cells produced up to 100-fold higher titers (Fig. 5H). Together, these results demonstrated that the resistance of SM9-1 trophoblast cells to MCMV infection could be overcome by NRP1 expression.
## Murine trophoblast stem cells lack NRP1 expression and are resistant to MCMV infection
mTSCs represent cells of the trophoblast lineage and retain the capacity to differentiate in vitro (33,42). In the present study, previously described mTSCs isolated from E3.5 blastocysts were used (43). For quality control, visual confirmation by microscopy was performed together with cellular marker gene expression analysis by quantitative real-time RT-PCR (Fig. S2). Next, cell surface expression of putative MCMV entry receptors in mTSCs was analyzed by flow cytometry. Similarly to the SM9-1 cell line, mTSCs expressed both ITGβ1 and PDGFRα but little β2-M and no NRP1 (Fig. 6A). A small fraction of the mTSCs was susceptible to MCMV infection (Fig. 6B andC), leading to low levels of progeny production in replication kinetics experiments (Fig. 6D).
To test whether enforced NRP1 expression can increase MCMV infection, mTSCs were transfected with plasmids expressing GFP or NRP1-GFP (Fig. 6E andF) and subsequently infected with MCMV. Flow cytometric analysis confirmed NRP1 surface expression in mTSCs transfected with NRP1-GFP but not with GFP (Fig. 6G andH). NRP1-GFP transfec tion increased MCMV susceptibility in mTSCs, leading to a significantly higher percentage of mCherry + cells at 1 dpi than in control cells (Fig. 6I andJ). These findings demonstrate that mTSCs lack NRP1 expression and are resistant to MCMV infection, paralleling the phenotype observed in SM9-1 trophoblasts and primary placental cells.
## Inducible NRP1 expression increases MCMV infection in mTSCs
NRP1 is involved in vessel sprouting, and global deletion of NRP1 leads to lethality in mice due to impaired yolk sac vascularization (44). Overexpression of this protein may thus negatively impact placenta development. To assess whether transient expression of NRP1 is sufficient to increase MCMV infection in mTSCs, a doxycycline-inducible expres sion system was used. As several attempts to transduce mTSCs with lentiviral vectors were unsuccessful in our hands, a piggyBac transposon system allowing doxycyclineinducible transgene expression was applied (45). The transgene is followed by an internal ribosome entry site (IRES) and GFP. PiggyBac transposon plasmids encoding NRP1-IRES-GFP or BFP-IRES-GFP were constructed and used for the transfection of mTSCs. After puromycin selection, doxycycline-inducible transgene expression was analyzed by flow cytometry (Fig. 7A). Doxycycline led to GFP expression in cells transfected with either construct (Fig. 7A andB), whereas NRP1 expression was observed in cells transfected with the NRP1-IRES-GFP but not the BFP-IRES-GFP construct (Fig. 7C andD). MCMV infection led to a significant increase in the percentage of mCherry + cells in doxycyclinetreated mTSCs expressing NRP1, when compared to the BFP control cells (Fig. 7E andF). These data demonstrated that transposon-mediated genetic modification of placental trophoblast cells allowed increased MCMV infection via NRP1 expression.
## DISCUSSION
Congenital CMV infection causes a significant health burden worldwide (46). Unfortu nately, the most widely used and well-established animal model for CMV disease has limited applicability for studying virus-host interactions in the fetus. Although infection of neonatal mice with MCMV models aspects of congenital CMV disease, MCMV cannot be used to study CMV transmission to the fetus in utero and congenital CMV infection. In the present study, we addressed tissue-specific features of the murine materno-fetal barrier interfering with vertical MCMV infection. We employed an intravenous high-dose MCMV infection model, which, while limiting the observation period due to high viral loads in organs such as the liver, enabled direct exposure of cell-free virus to the materno-fetal interface. In immunocompromised mice, MCMV replicated to considerably higher virus loads in the livers than in WT mice. However, even under these conditions, MCMV-infected cells were hardly detected in cells of the materno-fetal barrier. In line, ex vivo exposure of primary placental cells, SM9-1 trophoblasts, and mTSCs to MCMV revealed a cell type-specific resistance to infection. NRP1 expression reversed this phenotype and increased MCMV susceptibility. NRP1 was recently identified as a critical host factor for MCMV, and genetic deletion of NRP1 in SVEC4-10 cells abrogated MCMV infection and gene expression (25). MHC-I expression was described to be required for MCMV infection of macrophages (24), and we found β2-M, as part of the MHC-I, was very low in SM9-1 cells and mTSCs. How enforced MHC-I expression impacts infectibility in these cells remains to be tested in future studies. We found NRP1 to be expressed in MCMV-susceptible cell lines but not in resistant SM9-1 cells and mTSCs. Expression of NRP1 abrogated the resistance toward MCMV infection in these placental cells. Interestingly, transfection or transduction of SM9-1 cells or mTSCs with NRP1 led to a lower frequency of reporter fluorophore expression than the respective controls. This may result from the detrimental effects of NRP1 expression in trophoblasts and warrants further investigation. NRP1 is one of two homologous transmembrane proteins with a short cytoplasmic domain and a large extracellular region, which is divided into three domains. The extracellular domain allows binding to several ligands, including members of the semaphorin and vascular endothe lial growth factor families, and thus NRP1 can form complexes with other transmem brane receptors to act as a co-receptor in various biological processes (40). NRP1 has been reported to act as an entry receptor for human T-cell lymphotropic virus type 1 (47,48), Epstein-Barr virus (49), Kaposi's sarcoma-associated herpesvirus (50), and the severe acute respiratory syndrome coronavirus 2 (51,52). The related NRP2 has been identified as a host factor for Lujo virus infection (53). Importantly, NRP2 is also a receptor for HCMV, allowing virus entry into epithelial and endothelial cells (22). The HCMV pentame ric envelope glycoprotein complex, which is necessary for viral infection of several cell types, including epithelial cells, has been identified as the interaction partner for NRP2 (22,54), and more recently, NRP2 has been suggested to facilitate guinea pig CMV infection (55). It seems likely that NRP1 serves as a receptor for MCMV entry into specific cells, but this has not been formally demonstrated yet. Hence, NRP1 has merely been called a host factor required for infection of particular cells (25). Interestingly, the MCMVencoded chemokine 2 (MCK2), a part of the gH-gL-MCK2 glycoprotein complex that is analogous to the HCMV pentameric complex (56), was reported not to be involved in NRP1-dependent MCMV infection (24,25). Further studies will be necessary to determine how NRP1 facilitates MCMV infection, whether it acts as an entry receptor, and which viral glycoproteins bind to it. CMVs are known to cross the placenta in humans, non-human primates, and guinea pigs but not in rodents such as mice or rats (13). The microanatomy of the maternofetal interface is composed of various cell layers with maternal and fetal origin and is substantially different in the aforementioned species. Humans, non-human primates, and guinea pigs are hemomonochorial, whereas mice are hemotrichorial, meaning that one versus three trophoblast layers separate the chorion from maternal blood, respectively (57,58). There are several possibilities for MCMV vertical transmission to the fetus such as (i) infection of trophoblast cells and further infection of fetal cells via cell-to-cell spread, (ii) cell-free virus or infected maternal cells passing through leaks of the maternofetal barrier, (iii) virus penetrating the intact syncytiotrophoblast by transcytosis, and (iv) virus-infected maternal cells penetrating the barrier by transmigration. Accordingly, together with fetal vascular endothelial cells, a virus would need to infect or permeate only two cell layers in humans, non-human primates, and guinea pigs to penetrate the materno-fetal barrier, but four cell layers in mice. In addition, CMV is a slowly replicating virus, indicating that the production of virus particles for tissue penetration may require a certain amount of time to allow infection of the placenta and transmission from mother to fetus. Indeed, it has been reported that not until 10 days after intraperitoneal MCMV application, there is evidence of virus replication in placental tissue (7). In this respect, a major difference between species where vertical CMV transmission has been observed versus mice is the comparatively short gestational time of approximately 3 weeks in mice versus ~40, ~23, and ~10 weeks in humans, rhesus macaques, and guinea pigs, respectively (59,60). Thus, although not addressed in the present study, the combination of anatomical differences together with a short gestation period presumably reduces the chance of vertical MCMV transmission.
In the murine placenta, trophoblasts provide a multilayered defense against viral infection, combining structural, receptor-level, and immune mechanisms. The hemotri chorial murine placenta could reduce cell-to-cell spread in the materno-fetal interface. Resistance is further supported by restricting the expression of viral entry receptors. While human and guinea pig trophoblasts exhibit expression of NRP2 (55,61), murine trophoblasts display low levels of NRP1. The role of adaptive immune response was observed by MCMV infection of pregnant SCID mice bearing the Prkdc scid mutation, where increased infection in E18 fetal membranes was observed (10). Likewise, we found a few placental cells to be infected in Rag2 -/-Il2rg -/-mice, which lack T and NK cells (62,63). Thus, lymphocytes likely interfere with materno-fetal MCMV infection. When viruses bypass receptor-level restriction, murine trophoblasts themselves mount innate immune responses. Trophoblasts can sense and respond to their microenvironment through pattern recognition receptors, and placenta-derived type I interferon, together with their downstream interferon-stimulated genes, contributes to protecting against viral infection (64). The role of type I interferons for placental antiviral defense is supported by the observation that pregnant Ifnar1 ⁻/⁻ mice exhibited Zika virus infection in trophoblasts (65). Further studies are needed to decipher the role of adaptive and innate immune responses and what the underlying mechanisms are.
The present study suggests that the absence of NRP1 expression in murine placental trophoblast cells confers resistance against MCMV infection, reducing tissue damage in the materno-fetal barrier and potentially decreasing the risk of vertical transmission to the fetus.
## MATERIALS AND METHODS
## Animals
All mice were on a C57BL/6 background and kept in individually ventilated cages under specific pathogen-free conditions according to the recommendations of the FELASA (66). Pasteurella pneumotropica, Helicobacter species, and murine norovirus 1 were detected in sentinel animals tested in this breeding barrier. Food and water were provided ad libitum. Rag2 -/-Il2rg -/-(B6(Cg)-Rag2 tm1.1Cgn /J-JAX008449 crossed with B6.129S4-Il2rg 2tm1Wjl /J-JAX003174), Ifnar1 -/-(B6(Cg)-Ifnar1 tm1.2Ees /J-JAX028288), and actb-EGFP (C57BL/6-Tg(CAG-EGFP)131Osb/LeySopJ-JAX006567) were bred locally, C57BL/6J WT mice (JAX000664) were purchased from Charles River Laboratories (Sulzfeld).
## Cell lines
SM9-1 cells, an immortalized trophoblast cell line, were a kind gift from Joan Hunt (University of Kansas Medical Center, USA) and Margaret Petroff (Michigan State University, USA). They originated from a gestational day 9 Swiss-Webster mouse placenta, and the non-adherent trophoblastic cell outgrowths from the placental explants were transferred for 5-6 passages. These cells are immortalized but not transformed. In this study, SM9-1 cells were grown in RPMI 1640 containing antibiotics and supplemented with 10% fetal calf serum (FCS) and 0.05 mM β-mercaptoethanol at 37°C, 5% CO2 as previously described (34). All cell lines were cultured under identical conditions to allow comparability.
mTSCs were isolated and cultured as previously described (43). Briefly, the cells were kept undifferentiated in 30 vol% TS medium (RPMI 1640 containing 20% FCS, 100 IU/mL penicillin, 100 mg/mL streptomycin, 1 mM sodium pyruvate, and 0.1 mM β-mercaptoe thanol) and 70 vol % MEF-conditioned medium (TS medium harvested from mitomycin C-treated primary MEFs) supplemented with 30 ng/mL FGF4 and 1.2 μg/mL heparin. To induce differentiation, mTSC cells were cultured in TS medium only.
## Statistical analysis
We performed statistical tests as indicated and provided levels of significance directly for each analysis of interest. Data were processed using Prism software (GraphPad).
## MCMV infection
The MCMV-3DR recombinant, which was used in all experiments, was generated from the pSM3fr bacterial artificial chromosome by BAC recombineering. It encodes Gaussia luciferase, mCherry, contains a sequence within the m164 ORF encoding the SIINFEKL peptide, and contains the complete Mck2 ORF, but its m157 ORF is replaced by the sequences for the reporter proteins (31,67,68). Virus stocks were produced on 10.1 immortalized mouse embryonic fibroblasts (69), purified by centrifugation through a sucrose cushion, and titrated on M2-10B4 mouse bone marrow stromal cells (ATCC CRL-1972). To ensure consistent infectious unit concentrations in virus stocks, plaque assays were routinely performed using both the previous and newly prepared stocks in parallel. For in vivo infection experiments, animals were anesthetized (isoflurane) and received a single intravenous injection of 10 6 PFU MCMV-3DR.
## Histology and data processing
Organs were fixed in PBS-buffered 2% paraformaldehyde with 30% sucrose. Organ slices (7 µm thick) were prepared and stained with DAPI. Images were acquired with an AxioScan Slide Scanner (Carl Zeiss) using the Colibri 7 LED light source and processed as TIFF-formatted files with ZEN (Carl Zeiss) and ImageJ (NIH) software. Single-channel exports were converted into binary images, and particles were counted as nuclei (50-200 µm 2 area for DAPI + signals) or MCMV-infected cells (90-600 µm 2 mCherry + signals), applying the "classic watershed" plugin. Ratios of mCherry + to DAPI + signals were determined for each image.
## Luciferase assay
Cell culture supernatants were measured for luciferase expression by quantification of luminescence after the addition of native coelenterazine (Synchem) with a Centro XS³ LB 960 luminometer (Berthold Technologies) essentially as described previously (15).
## Primary cell isolation
Single-cell suspensions were prepared from the placenta at gestational day 12.5. Placenta cells were generated as described in references (70,71). Briefly, after macro scopic dissociation from the fetus, placenta fragments were digested in collagenase type II-S (Sigma-Aldrich) and then passed through a 70 µm cell strainer before fur ther separation via a Percoll (Sigma-Aldrich) gradient and subsequent resuspension in DMEM/F12 medium supplemented with 10% FCS, 100 IU/mL Penicillin, 100 µg/mL Streptomycin, 1 mM sodium pyruvate, and 0.05 mM β-mercaptoethanol.
## Quantitative real-time RT-PCR
Total RNA was extracted from the cells using the innuPREP RNA Mini Kit (Analytik Jena), and contaminating DNA was removed using the TURBO DNA-free Kit (Ambion). cDNA was synthesized from 1 to 5 µg of the extracted RNA by using the RevertAid H Minus Reverse Transcriptase, oligo-dT primers, and the RNase inhibitor RiboLock (Thermo Fisher Scientific). qPCR was performed on a QuantStudio 3 (ThermoFisher Scientific) and the PowerTrack SYBR Green Mastermix (Fisher Scientific). To amplify mouse transcripts, the following primers were used: Gapdh (GGAGAAACCTGCCAAGTATGATG and GACAACCT GGTCCTCAGTGTAGC), Cdx2 (AGACAAATACCGGGTGGTGTA and CCAGCTCACTTTTCCTCC TGA), Pl1 (GACTACCCTGCTTGGTCTGG and GAAAGACAACTCGGCACCTC), and Tpbpa (C AGAGAGTGGCGATGGGTTTT and GACAATGGCACAGTGGCTGTT). Transcript levels were normalized to a housekeeping gene (Gapdh).
## Immunoblot analysis
For immunoblot analysis, cells were lysed in NP-40 buffer (50 mM Tris, 150 mM NaCl, 1% Nonidet P-40, and Complete Mini protease inhibitor cocktail [Roche]), separated by SDS-PAGE, and subsequently transferred to a nitrocellulose membrane by semi-dry blotting. Target proteins were detected using monoclonal antibodies against NRP1 (D62C6, Cell Signaling Technology) or GAPDH (14C10, Cell Signaling Technology) and combined with HRP-coupled secondary antibodies (Jackson ImmunoResearch). Recombinant mouse NRP1 (R&D Systems) was included as a positive control. Images were acquired with the Fusion Capture Advance FX7 16.15 (Peqlab) camera.
## Transfection
SM9-1 trophoblast cells were transfected with expression plasmids pEGFP-C1 (Clontech) or pCMV3-NRP1-GFPSpark (SinoBiological) plasmids using Lipofectamine 2000 (Thermo Fisher Scientific). Briefly, 3 µg of plasmid was mixed with 10 µL of Lipofectamine 2000 in Opti-MEM (GIBCO) for 10 min at room temperature before being added dropwise to 3 × 10 5 SM9-1 cells seeded the day before in a six-well plate.
mTSCs were transfected with expression plasmids pEGFP-C1 and pCMV3-NRP1-GFPSpark using GenJet transfection reagent (SignaGen). Briefly, 1 µg of plasmid was mixed with 3 µL of GenJet in DMEM for 15 min at RT before being added dropwise to 1 × 10 5 mTSCs seeded the day before in a 12-well plate.
## Transduction with lentiviral expression vectors
NRP1-GFPSpark or GFPSpark coding sequences were PCR-amplified from the pCMV3-NRP1-GFPSpark plasmid using forward (TATAGAATTCATGGAGAGGGGGCTGCCG or TATAG AATTCATGGTGAGCAAGGGCGAGGAGC) and reverse (TTAAGAATTC TTACTTGTACAGCTCG TCCATGCCG) primers containing EcoRI restriction sites. The PCR product was cleaved with EcoRI and inserted into the lentiviral vector pLeGO-iB2-puro (72). This vector contains an SFFV promoter driving the expression of the insert, followed by IRES2-BFP-P2A-puromycin. Lentiviruses were generated using standard third-generation packag ing vectors in HEK-293T cells. SM9-1 cells were transduced with the lentiviral vectors expressing NRP1-GFPSpark or GFPSpark in the presence of polybrene (Sigma). Trans duced SM9-1 cells were selected with 2.5 µg/mL puromycin (Sigma), and GFP-positive cells were sorted using FACS Aria Fusion.
## PiggyBac transposon system
NRP1-and BFP-coding sequences were PCR amplified from the pCMV3-NRP1-GFPSpark plasmid or from the pLEGO-iB2 plasmid using forward (ATGGAGAGGGGGCTGCCG or A TGAGCGAGCTGATTAAGG) and reverse (TCACGCCTCTGAGTAATTACTCTG or TTACTTGTAC ATCAGGCGC) primers and inserted into the Tet-on piggyBac vector (Addgene # 97421) modified as described in reference (73). The vector contains a doxycycline-inducible TREG3 promoter driving the expression of the insert, followed by IRES-GFP, while the EF-1α promoter drives the expression of the reverse tetracycline-controlled transactiva tor and puromycin resistance marker. 2 × 10 5 mTSCs were seeded in a six-well plate the day before transfection. mTSCs were transfected with 2 µg PB-NRP1 or PB-BFP and 1 µg of pCMV-HyPBase (Sanger Plasmid Repository) encoding a hyperactive BP transposase (74) using 6 µL GenJet transfection reagent. 72 h post-transfection, cells were selected with 0.5 µg/mL of puromycin. NRP1 or BFP and GFP expression was induced by adding 2 µg/mL doxycycline (Biomol).
## Viral replication kinetics
SM9-1 or mTSC cells (3 × 10 4 ) were infected at an MOI of 1 with MCMV-3DR in triplicates. The input virus was removed 4 h later, and fresh medium was added. The supernatants were harvested at different times post-infection and titrated by plaque assay on M2-10B4 cells.
## Flow cytometry
Cells were acquired on a LSRFortessa Cell Analyzer or FACSymphony A1 Cell Analyzer (BD Biosciences). The following antibodies were used for stainings: CD45-APC (30-F11), NRP1-PE (3E12), PDGFRα-PerCP-Cy5.5 (APA5), β2-microglobulin-APC (A16041A), integrin β1-PE-Cy7 (HMβ1-1), and NRP1-APC (3E12). Data were processed with FlowJo version 10 (BD Biosciences). Cells were pre-gated in a forward to sideward scatter plot, and cell duplicates were excluded in a second forward scatter high to area plot. Dead cells were identified by the use of a cell viability marker (Zombie Violet or NIR) and excluded from analysis.
## Reanalysis of a single-nuclei RNA-sequencing data set
A single-nuclei RNA sequencing data set (NCBI GEO [GSE152248]) (41) was reanalyzed using the R package Seurat (v 4.3.0.1). The cell subsets were annotated according to the original report. For analysis, only placentas isolated at gestational day 12.5 were included.
## FUNDING
## References
1. Rawlinson, Boppana, Fowler et al. (2017) "Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy" *Lancet Infect Dis*
3. Choi, Root, Mcgregor (2016) "A novel non-replication-competent cytomegalovirus capsid mutant vaccine strategy is effective in reducing congenital infection" *J Virol*
4. Wang, Taher, Kreklywich et al. (2025) "The pentameric complex is not required for congenital CMV transmission in seronegative rhesus macaques" *Sci Transl Med*
5. Schleiss, Fernández-Alarcón, Bierle et al. (2025) "Replication-deficient whole-virus vaccines against cytomegalovirus induce protective immunity in a guinea pig congenital infection model" *J Virol*
6. Olding, Jensen, Oldstone (1975) "Pathogenesis of of cytomega lovirus infection. I. Activation of virus from bone marrow-derived Full-Length Text Journal of Virology December"
7. "lymphocytes by in vitro allogenic reaction" *J Exp Med*
8. Olding, Kingsbury, Oldstone (1976) "Pathogenesis of cytomegalovirus infection. Distribution of viral products, immune complexes and autoimmunity during latent murine infection" *J Gen Virol*
9. Johnson (1969) "Mouse cytomegalovirus: placental infection" *J Infect Dis*
10. Slavuljica, Kveštak, Huszthy et al. (2015) "Immunobiology of congenital cytomegalovirus infection of the central nervous system-the murine cytomegalovirus model" *Cell Mol Immunol*
11. Medearis (1964) "Mouse cytomegalovirus infection. 3. attempts to produce intrauterine infections" *Am J Hyg*
12. Woolf, Jaquish, Koehrn (2007) "Transplacental murine cytomegalovirus infection in the brain of SCID mice" *Virol J*
13. Chantler, Misra, Hudson (1979) "Vertical transmission of murine cytomegalovirus" *J Gen Virol*
14. Lisnić, Tomac, Cekinović et al. (2021) "Rodent models of congenital cytomegalovirus infection" *Methods Mol Biol*
15. Weisblum, Panet, Haimov-Kochman et al. (2014) "Models of vertical cytomegalovirus (CMV) transmission and pathogenesis" *Semin Immunopathol*
16. Zhou, Mei, Wang et al. (2022) "A congenital CMV infection model for follow-up studies of neurodevelopmental disorders, neuroimaging abnormalities, and treatment" *JCI Insight*
17. Stahl, Heller, Halle et al. (2013) "Nodular inflammatory foci are sites of T cell priming and control of murine cytomegalovirus infection in the neonatal lung" *PLoS Pathog*
18. Megli, Coyne (2022) "Infections at the maternal-fetal interface: an overview of pathogenesis and defence" *Nat Rev Microbiol*
19. Ander, Diamond, Coyne (2019) "Immune responses at the maternal-fetal interface" *Sci Immunol*
20. Wang, Huong, Chiu et al. (2003) "Epidermal growth factor receptor is a cellular receptor for human cytomegalovirus" *Nature*
21. Wang, Huang, Huong et al. (2005) "Integrin alphavbeta3 is a coreceptor for human cytomegalovirus" *Nat Med*
22. Li, Fischer, Cohen (2016) "Cell surface THY-1 contributes to human cytomegalovirus entry via a macropinocytosis-like process" *J Virol*
23. Vanarsdall, Pritchard, Wisner et al. (2018) "CD147 promotes entry of pentamer-expressing human cytomega lovirus into epithelial and endothelial cells" *mBio*
24. Martinez-Martin, Marcandalli, Huang et al. (2018) "An unbiased screen for human cytomegalovirus identifies neuropilin-2 as a central viral receptor" *Cell*
25. Kabanova, Marcandalli, Zhou et al. (2016) "Platelet-derived growth factor-α receptor is the cellular receptor for human cytomegalovirus gHgLgO trimer" *Nat Microbiol*
26. Bošnjak, Henze, Lueder et al. (2023) "MCK2-mediated MCMV infection of macrophages and virus dissemination to the salivary gland depends on MHC class I molecules" *Cell Rep*
27. Lane, Guo, Fisher et al. (2020) "Necroptosis-based CRISPR knockout screen reveals Neuropilin-1 as a critical host factor for early stages of murine cytomegalovirus infection" *Proc Natl Acad Sci*
28. Kawasaki, Kosugi, Sakao-Suzuki et al. (2015) "Cytomegalovirus initiates infection selectively from high-level β1 integrin-expressing cells in the brain" *Am J Pathol*
29. Sitnik, Krstanović, Gödecke et al. (2023) "Fibroblasts are a site of murine cytomegalovi rus lytic replication and Stat1-dependent latent persistence in vivo" *Nat Commun*
30. Lueder, Heller, Ritter et al. (2018) "Control of primary mouse cytomegalovirus infection in lung nodular inflammatory foci by cooperation of interferon-gamma expressing CD4 and CD8 T cells" *PLoS Pathog*
31. Stahl, Jung, Jazbutyte et al. (2018) "Laboratory diagnostics of murine blood for detection of mouse cytomegalovirus (MCMV)-induced hepatitis" *Sci Rep*
32. Tegtmeyer, Spanier, Borst et al. (2019) "STING induces early IFN-β in the liver and constrains myeloid cellmediated dissemination of murine cytomegalovirus" *Nat Commun*
33. Marquardt, Halle, Seckert et al. (2011) "Single cell detection of latent cytomegalovirus reactivation in host tissue" *J Gen Virol*
34. Jordan, Krause, Prager et al. (2011) "Virus progeny of murine cytomegalovirus bacterial artificial chromosome pSM3fr show reduced growth in salivary Glands due to a fixed mutation of MCK-2" *J Virol*
35. Hemberger, Hanna, Dean (2020) "Mechanisms of early placental development in mouse and humans" *Nat Rev Genet*
36. Bowen, Hunt (1999) "Expression of cell adhesion molecules in murine placentas and a placental cell line" *Biol Reprod*
37. Harvey, Levine (1991) "P53 alteration is a common event in the spontaneous immortalization of primary BALB/c murine embryo fibroblasts" *Genes Dev*
38. Lemoine, Humphries, Abraham et al. (1988) "Partial characterization of a novel stromal cell-derived pre-B-cell growth factor active on normal and immortalized pre-B cells" *Exp Hematol*
39. O'connell, Edidin (1990) "A mouse lymphoid endothelial cell line immortalized by simian virus 40 binds lymphocytes and retains functional characteristics of normal endothelial cells" *J Immunol*
40. Owens, Smith, Hackett (1974) "Epithelial cell cultures from normal glandular tissue of mice" *J Natl Cancer Inst*
41. Darlington (1987) "Liver cell lines" *Methods Enzymol*
42. Wild, Staton, Chapple et al. (2012) "Neuropilins: expression and roles in the epithelium" *Int J Experimental Path*
43. Marsh, Blelloch (2020) "Single nuclei RNA-seq of mouse placental labyrinth development"
44. Singh, Gerton (2021) "Protocol for mouse trophoblast stem cell isolation, differentiation, and cytokine detection" *STAR Protoc*
45. Zhu, Hölz, Metzger et al. (2014) "Lysine-specific demethylase 1 regulates differentiation onset and migration of trophoblast stem cells" *Nat Commun*
46. (2025) *Full-Length Text Journal of Virology*
47. Kawasaki, Kitsukawa, Bekku et al. (1999) "A requirement for neuropilin-1 in embryonic vessel formation" *Development*
48. Myers, Wright, Peckner et al. (2018) "Discovery of proteins associated with a predefined genomic locus via dCas9-APEX-mediated proximity labeling" *Nat Methods*
49. Griffiths (2012) "Burden of disease associated with human cytomega lovirus and prospects for elimination by universal immunisation" *Lancet Infect Dis*
50. Jin, Alkhatib, Cornetta et al. (2010) "Alternate receptor usage of neuropilin-1 and glucose transporter protein 1 by the human T cell leukemia virus type 1" *Virology (Auckl)*
51. Lambert, Bouttier, Vassy et al. (2009) "HTLV-1 uses HSPG and neuropilin-1 for entry by molecular mimicry of VEGF165" *Blood*
52. Wang, Zhang, Zhang et al. (2015) "Neuropilin 1 is an entry factor that promotes EBV infection of nasopharyngeal epithelial cells" *Nat Commun*
53. Lu, Sun, Zhang et al. (2023) "Neuropilin 1 is an entry receptor for KSHV infection of mesenchy mal stem cell through TGFBR1/2-mediated macropinocytosis" *Sci Adv*
54. Cantuti-Castelvetri, Ojha, Pedro et al. (2020) "Neuropilin-1 facilitates SARS-CoV-2 cell entry and infectivity" *Science*
55. Daly, Simonetti, Klein et al. (2020) "Neuropilin-1 is a host factor for SARS-CoV-2 infection" *Science*
56. Raaben, Jae, Herbert et al. (2017) "NRP2 and CD63 are host factors for lujo virus cell entry" *Cell Host Microbe*
57. Wrapp, Ye, Ku et al. (2022) "Structural basis for HCMV Pentamer recognition by neuropilin 2 and neutralizing antibodies" *Sci Adv*
58. Qin, Choi, El-Hamdi et al. (2025) "Selective knockout of key CMV receptors in fetal cells blocks direct and endocytic pathways of entry in the guinea pig" *bioRxiv*
59. Wagner, Brizic, Prager et al. (2013) "The viral chemokine MCK-2 of murine cytomegalovirus promotes infection as part of a gH/gL/MCK-2 complex" *PLoS Pathog*
60. Desesso, Williams, Ahuja et al. (2012) "The placenta, transfer of immunoglobulins, and safety assessment of biopharmaceuticals in pregnancy" *Crit Rev Toxicol*
61. Furukawa, Kuroda, Sugiyama (2014) "A comparison of the histological structure of the placenta in experimental animals" *J Toxicol Pathol*
62. Bonney (2013) "Demystifying animal models of adverse pregnancy outcomes: touching bench and bedside" *Am J Reprod Immunol*
63. Silk, Short, Roberts et al. (1993) "Gestation length in rhesus macaques (Macaca mulatta)" *Int J Primatol*
64. Arutyunyan, Roberts, Troulé et al. (2023) "Spatial multiomics map of trophoblast development in early pregnancy" *Nature*
65. Shinkai, Rathbun, Lam et al. (1992) "RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement" *Cell*
66. Cao, Shores, Hu-Li et al. (1995) "Defective lymphoid develop ment in mice lacking expression of the common cytokine receptor gamma chain" *Immunity*
67. Ding, Maxwell, Adzibolosu et al. (2022) "Mechanisms of immune regulation by the placenta: role of type I interferon and interferon-stimulated genes signaling during pregnancy" *Immunol Rev*
68. Miner, Cao, Govero et al. (2016) "Zika virus infection during pregnancy in mice causes placental damage and fetal demise" *Cell*
69. Convenor, Berard, Feinstein et al. (2014) "FELASA recommendations for the health monitoring of mouse, rat, hamster, guinea pig and rabbit colonies in breeding and experimental units" *Lab Anim*
70. Lemmermann, Gergely, Böhm et al. (2010) "Immune evasion proteins of murine cytomegalovi rus preferentially affect cell surface display of recently generated peptide presentation complexes" *J Virol*
72. Stahl, Keyser, Heller et al. (2015) "Mck2-dependent infection of alveolar macrophages promotes replication of MCMV in nodular inflammatory foci of the neonatal lung" *Mucosal Immunol*
73. Harvey, Levine (1991) "P53 alteration is a common event in the spontaneous immortalization of primary BALB/c murine embryo fibroblasts" *Genes & Development*
74. Pennington, Schlitt, Schulz (2012) "Isolation of primary mouse trophoblast cells and trophoblast invasion assay" *J Vis Exp*
75. Thordarson, Folger, Talamantes (1987) "Development of a placental cell culture system for studying the control of mouse placental lactogen II secretion" *Placenta*
76. Schwarze, Głów, Sonntag et al. (2021) "Optimisation of a TALE nuclease targeting the HIV co-receptor CCR5 for clinical application" *Gene Ther*
77. Weissmann, Robitaille, Günther et al. (2025) "Polycomb group proteins protect latent Kaposi sarcoma-associated herpesvirus from episome clearance and HUSH-dependent chromatin silencing" *bioRxiv*
78. Yusa, Zhou, Li et al. (2011) "A hyperactive piggyBac transposase for mammalian applications" *Proc Natl Acad Sci*
79. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724358&blobtype=pdf | # Broad-spectrum inhibition of influenza A virus replication by blocking the nuclear export of viral ribonucleoprotein complexes
Wentao Shen, Jie Xu, Zhaoshan Chen, Yanli Wei, Qian Wang, Xiangkun Wang, Xuegang Zhang, Qiyun Zhu, Shuai Xu
## Abstract
Influenza A virus (IAV) remains a major threat to global health. The emergence of IAV strains resistant to commonly used anti-influenza medicines has made the need for novel broad-spectrum antiviral agents more urgent. The IAV nucleoprotein (NP) is one of the most conserved viral proteins encoded by IAV and is essential for multiple processes in the lifecycle of IAVs. In this study, we screened a high-affinity nanobody, Nb7, which specifically recognizes NP from different subtypes of IAV and broadly inhibits the replication of various IAVs. Mechanistically, Nb7 blocked the nuclear export of viral ribonucleoprotein (vRNP) complexes and the subsequent assembly of progeny virions by specifically binding the nuclear export sequence 1 (NES1) region of the NP. The conserved Q42/E46/K48 residues within the NES1 region of NP were identified as the key sites for the binding of Nb7 and were synergistically responsible for the nuclear export of vRNP. In addition, Nb7 fused with a trans-activating transduction peptide efficiently inhibited the replication of IAV in vitro and provided full protection against lethal IAV infection in a mouse model. Taken together, our findings revealed that the NP-specific Nb7 is a promising candidate against IAV infection. This study provides a new understanding of the nuclear export of IAV-vRNP and contributes to the develop ment of broad-spectrum anti-influenza approaches.IMPORTANCE Influenza A viruses (IAVs) cause seasonal epidemics and global pan demics, resulting in hundreds of thousands of deaths annually. The identification of conserved epitopes and the development of effective prevention and treatment strategies are crucial for addressing these challenges. The nucleoprotein (NP) protein of IAV is a key component of the viral ribonucleoprotein (vRNP) complex and is highly conserved in various subtypes. Therefore, the development of vaccines or drugs that target conserved viral proteins such as the NP is promising. We screened Nb7, which specifically recognizes NP and broadly inhibits IAV replication through blocking the nuclear export of vRNP. More importantly, the trans-activating transduction peptidefused Nb7 has promising effects for the prevention and treatment of IAV infection. Our present study provides novel insight for the further development of broad-spectrum vaccines and anti-influenza drugs. KEYWORDS influenza A viruses, nanobody, nucleoprotein, nuclear export, viral replication I nfluenza A virus (IAV) remains a significant threat to global health, with approximately 1 billion seasonal influenza cases worldwide each year, resulting in 290,000-650,000 deaths (1). Due to persistent antigenic shift and drift, IAVs continually cross the species barrier to transmit and infect mammals, including humans, cows, and sheep, posing a significant threat to public health (2). Identifying new vulnerabilities in the viral life cycle
is critical for the development of effective therapeutic strategies. Traditional anti-influenza drugs and vaccines target mainly viral surface proteins. However, surface proteins tend to mutate under immune stress, which significantly reduces the effectiveness of antiviral methods. Therefore, the development of broad-spectrum antiviral agents that target more conserved proteins of IAV is important.
IAV belongs to the Orthomyxoviridae family and encodes at least 14 viral proteins, including basic polymerase 2 (PB2), basic polymerase 1 (PB1), acidic polymerase (PA), hemagglutinin (HA), nucleoprotein (NP), neuraminidase (NA), matrix (M) proteins, and nonstructural (NS) proteins (3). The viral ribonucleoprotein (vRNP) complexes formed by polymerase proteins (PB1, PB2, and PA), the NP, and the viral genome vRNA are the basic functional units of viral replication and transcription (4). In the early stage of infection, vRNP enters the nucleus to initiate replication and transcription of the viral genome. In the late stage of infection, the newly assembled vRNPs are exported from the nucleus to form progeny virions (5).
NP, the primary protein that forms the vRNP complex, guarantees the formation of the IAV genome into vRNPs with the correct conformation and structure (6). The vRNP complex needs to form an inverse parallel double helix structure through interactions between NP oligomers. Concurrently, three nuclear localization signals (NLSs), three functional nuclear export signals (NESs), and at least one nuclear accumulation signal have been identified in the NP (7)(8)(9)(10)(11). vRNPs rely on these signal sequences of the NP to mediate its transport between the cytoplasm and the nucleus (10,(12)(13)(14)(15)(16). Therefore, as one of the most conserved proteins among IAV virions, the NP plays an integral role in the life cycle of IAV, making the NP an ideal target for developing broad-spectrum methods against IAVs.
Nanobodies (Nbs) are variable domain fragments derived from camel heavy-chain antibodies and are characterized by strong physical and chemical stability, as well as low immunogenicity (17). Nbs can be expressed efficiently in a wide range of different expression systems at a low cost (18). Compared with conventional antibodies, Nbs have a longer complementary-determining region (CDR), enabling them to bind to the channel, seam, or hidden epitopes of the target proteins, thereby modulating their functions (19). Therefore, Nbs are valuable for the development of novel antiviral agents (20).
The present study aimed to screen for broad-spectrum Nbs and identify novel targets against IAV infection. We immunized alpaca and obtained Nb7, which broadly inhibited IAV. Nb7 recognizes the NP Q42/E46/K48 sites, which are located in the NES1 region, to block the nuclear export of vRNP. Notably, the Q42/E46/K48 triple mutation abolished the ability to mediate the nuclear export of vRNP. After fusion with the trans-activating transduction (TAT) peptide, Nb7 protected the mice from lethal IAV infection. Together, this study lays the groundwork for the development of novel broad-spectrum anti-influenza strategies.
## RESULTS
## Screening of NP-specific Nbs with broad-spectrum inhibitory effects on IAVs
To identify NP-specific Nbs with broad-spectrum inhibition of IAV replication, recombi nant NP protein derived from the H1N1 IAV (PR8 strain) was expressed and purified from HEK293T cells and was used as an antigen for immunizing alpacas and screening Nbs (Fig. 1A). After three rounds of screening, 48 clones were selected for indirect enzyme-linked immunosorbent assay (ELISA) to detect the reaction with the NP protein purified from E. coli cells (Fig. S1A). The phage-ELISA results indicated that Nb7, Nb11, Nb20, Nb22, Nb30, and Nb46 were highly reactive with the NP, with an absorbance value greater than 2. Amino acid sequence analysis revealed that the CDR3 region of the six Nb strains is composed of different amino acid residues (Fig. S1B). By fusion with human Fc fragments, six recombinant Nbs, that is, Nb7-Fc, Nb11-Fc, Nb20-Fc, Nb22-Fc, Nb30-Fc, and Nb46-Fc, were successfully expressed and purified from HEK293F suspension cells (Fig. S1C). Indirect ELISA suggested that all six recombinant Nbs strongly bind to NP (Fig. 1B).
Using an IAV reporter virus (PR8-Nluc), we found that transiently expressed Nb7 significantly inhibited IAV replication, whereas other Nbs did not (Fig. 1C; Fig. S1D). Moreover, transiently expressed Nb7 significantly inhibited the replication of the wild-type H1N1-PR8 virus in cells, but the negative control Nb (K7.13), which specifically recognizes the nCoV S protein, did not affect the replication of H1N1-PR8 (Fig. 1D; Fig. S1E) (21). The indirect immunofluorescence assay (IFA) indicated that Nb7 colocalized with PR8-NP in the nucleus of cells transfected with plasmids encoding NP and Nb7 genes (Fig. 1E). Using surface plasmon resonance method to detect the affinity between the NP and Nb7, the equilibrium dissociation constant (KD) between the NP and Nb7 was determined to be 11.2 nM (Fig. 1F), which indicated that Nb7 has a high affinity with NP. In addition, purified Nb7 immunoprecipitated with NP protein in transfected cells and the purified GST-NP, but K7.13 Nb did not (Fig. 1G; Fig. S1F).
We subsequently verified the effect of Nb7 on the replication of various IAVs. The replication kinetics curve revealed that transiently expressed Nb7 significantly reduced the replication of H1N1, H3N2, H6N6, and H9N2 viruses (Fig. 2A). Moreover, compared with K7.13 and the Vec control, the expression of Nb7 significantly decreased the expression levels of viral RNA and protein in H1N1, H3N2, H6N6, and H9N2 IAV subtypes (Fig. S2A andB). Moreover, transiently expressed Nb7 did not affect the replication of other viruses, such as vesicular stomatitis virus (VSV), Sendai virus (SeV), or Newcastle disease virus (NDV) (Fig. 2B). The ELISA results confirmed that purified Nb7 specifically recognized H1N1, H3N2, H6N6, and H9N2 IAVs, but not VSV, SeV, or NDV (Fig. 2C). Additionally, both Nb7 and K7.13 were mainly distributed in the cytoplasm of transfected cells. However, the distribution of Nb7 was rearranged and colocalized with NP in the nucleus following transfection with NP or infection with H1N1, H3N2, H6N6, and H9N2 viruses (Fig. 2D andE; Fig. S2C andD).
These results indicated that Nb7 specifically binds to NP with high affinity and broadly inhibits the replication of different subtypes of IAV in cells.
## Nb7 inhibits viral replication by blocking the nuclear export of vRNP
Next, we investigated the mechanism by which Nb7 inhibits IAV replication. Previous studies have shown that the NP encapsulates viral RNAs and associates with three polymerase proteins to form the vRNP complex, which is responsible for the transcription and replication of viral genomes (5,22,23). The results of the polymerase activity assay revealed that the expression of Nb7 does not affect viral polymerase activity (Fig. 3A). Furthermore, co-immunoprecipitation (co-IP) and cross-linking RNA immunoprecipita tion assays revealed that the expression of Nb7 does not inhibit the oligomerization of NP or the interaction of NP with viral RNAs (Fig. S3A andB). However, the distribution of the NP protein revealed that there are more NP signals in the nuclei and less in the cytoplasm of cells expressing Nb7 than in those expressing K7.13 or the Vec control (Fig. 3B andC). Consistently, the cell fraction isolation assay indicated that more NP proteins were detected in the nuclear fraction from Nb7-expressing cells than in that from K7.13-expressing cells (Fig. 3D).
NP is the main component protein of the vRNP complex and is responsible for mediating the nuclear import and export of vRNP (5). The distribution of NP proteins partially represents the localization of vRNPs within cells, especially during the early stages of infection. Two hours after viral infection, the vRNPs were located in the cytoplasm and did not differ in either Nb7-or K713-expressing cells, indicating that Nb7 did not affect the attachment or internalization of IAV (Fig. 3E). At 3 h after infection, most of the NP proteins were located in the nucleus, and no significant differences were detected among the Vec-, Nb7-, or K7.13-expressing cells, indicating that Nb7 did not affect the nuclear import of vRNP. At 4 and 5 h after infection, the NP proteins were mostly distributed in the cytoplasm and tended to aggregate at the cell membrane in Vec-and K7.13-expressing cells. In contrast, most NP proteins were still located in the FIG 2 The Nb7 antibody specifically recognizes IAV NP and inhibits IAV replication. (A) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec. After 24 h, the cells were infected with the PR8 (H1N1), TA05 (H3N2), HN4 (H6N6), or HN38 (H9N2) viruses (MOI = 0.1). At the indicated times post-infection, the supernatant containing the viral particles was collected for the EID 50 assay. The significant difference between Nb7 and Vec is labeled in blue; the significant difference between K7.13 and Vec is labeled in gray. (B) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec. After 24 h, the cells were infected with PR8 (H1N1), VSV, SeV, or NDV for an additional 24 h. Total RNA was then extracted for qPCR to detect the expression of viral RNA. (C) Indirect ELISA was used to detect the ability of Nb7 to recognize different viruses. The purified viral particles (2 µg/mL) of different viruses were coated on 96-well microtiter plates, and Nb7-Fc (Continued on next page) nucleus of the Nb7-expressing cells (Fig. 3E andF). Similar results were obtained from the cell fraction isolation assay, which revealed that Nb7 promoted the accumulation of NP and polymerase proteins in the nuclei of virus-infected cells (Fig. 3G). These results demonstrated that Nb7 blocks the nuclear export of the vRNP complex.
## Nb7 specifically recognizes conserved residues in the NES1 region of the NP
To further elucidate the mechanism by which Nb7 inhibits the replication of IAV, we next identified the binding sites of Nb7 on the NP protein. We constructed multiple NP truncations based on the domain organization of the NP (24). Nb7 recognizes the RNA-binding domain (aa 1-180) of the NP (Fig. 4A andB). The complex structure of PR8-NP and Nb7 was predicted via AlphaFold 3. The prediction results for the NP protein were consistent with the known structural model, indicating that the prediction of this binding interface is highly confident. The predictive model revealed that Nb7 mainly binds to a protruding area on the antigen surface through its CDRs. The key interaction interfaces include R101, L103, and S105 of CDR3, as well as R33 and S54 of CDR2. L103 forms crucial hydrogen bonds with E46 and K48 of the NP, whereas R33, S54, R101, and S105 of Nb7 form salt bridges and hydrogen bond networks with Q42, E115, and Q122 of the NP (Fig. 4C). Notably, residues Q42, E46, K48, E115, and Q122 are located within the RNA-binding domain of the NP, which is consistent with the former results shown in Fig. 4B.
To identify the key residues of NP recognized by Nb7, a series of single and multiple mutant plasmids was constructed. Using purified Nb7-Fc as the first antibody, the IFA results confirmed that the NP-Q42A/E46A/K48A mutant was not recognized by Nb7. In contrast, the single or double mutants were still recognized by Nb7 (Fig. 4D). In addition, triple-mutant NP (Q42A/E46A/K48A) accumulated predominantly in the nucleus. In contrast, the single or double mutants were distributed in both the nucleus and the cytoplasm (Fig. 4D). Moreover, the co-IP results demonstrated that Nb7 completely lost the ability to recognize NP when the Q42, E46, and K48 residues were simultaneously mutated (Fig. 4E). Considering that Q42, E46, and K48 are located in the NES1 region of the NP, we constructed NES1 and NLS1 truncations and detected that the NES1 truncation was located mainly in the nucleus and was not recognized by Nb7 (Fig. S4A andB). Furthermore, we superimposed the vRNP structure and Nb7 (25). As shown in Fig. 4F, multiple binding sites exist between Nb7 and vRNP, all of which are exposed on the outer surface of the rod-like vRNP structure, allowing various Nb7 molecules to bind to the NP within the vRNP complex. The alignment of 76,616 NP sequences of all the IAVs from GenBank revealed that the Q42, E46, and K48 residues are highly conserved in the NP among all the IAVs (Fig. 4G).
The data above demonstrated that Nb7 recognizes the Q42, E46, and K48 residues located in the conserved NES1 region of the NP and that the triple mutation (Q42A/ E46A/K48A) increases the nuclear retention of the NP.
## TAT-Nb7 enables penetration of the cell membrane and inhibits viral replication
As NP is an intracellular viral protein, we used a TAT peptide derived from human immunodeficiency virus type 1 (HIV-1) to deliver Nb7 into cultured cells and tissues through the fusion of the TAT peptide to the N-terminus of Nb7 (Fig. 5A) (26,27). SDS-PAGE analysis confirmed that TAT-Nb7 and TAT-K7.13 were successfully expressed and purified (Fig. 5B). Indirect ELISA results revealed that TAT-Nb7 maintained high (2 µg/mL) was then added. (D) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec, along with Vec or H1N1 PR8 -NP. After 18 h, the cells were fixed and stained with anti-Flag (diluted 1:300) and a commercial anti-NP mAb (diluted 1:300). (E) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec. After 24 h, the cells were infected with PR8 (H1N1) virus (MOI = 0.1) or Mock for 24 h and stained with anti-Flag (diluted 1:300) and a commercial anti-NP mAb (diluted 1:300). The data shown represent three independent experiments (n = 3); the bars represent the means ± SDs. reactivity with the NP (Fig. 5C). The results of the CCK8 assay indicated that TAT-Nb7 did not affect cell viability at concentrations less than or equal to 20 µM (Fig. 5D). As shown in Fig. 5E and F, TAT-Nb7 was efficiently taken up by cells in a dose-dependent manner, but Nb7 was not. Furthermore, TAT-Nb7 inhibited the expression of viral proteins and the replication of the PR8-GFP virus in a dose-dependent manner (Fig. 5G andH). Similarly, TAT-Nb7 significantly inhibited the replication of the wild-type PR8 virus in cells (Fig. 5I). These data suggest that the TAT peptide efficiently delivers Nb7 into cells and inhibits the replication of IAV.
## TAT-Nb7 treatment protects mice against lethal virus challenge
We further evaluated whether TAT-Nb7 exhibits protective efficacy against IAV infection in vivo. For prophylactic assessment, mice were administered either a low (10 mg/kg) or high (30 mg/kg) dose of TAT-Nb7 or IgG intranasally at 12 h prior to infection, followed by infection with 10 4 EID 50 of the PR8 virus. For therapeutic assessment, mice were administered either a low (10 mg/kg) or high (30 mg/kg) dose of TAT-Nb7 intranasally at 2 and 24 h post-infection (Fig. 6A).
For the prophylactic group of mice challenged with the PR8 virus, TAT-Nb7 provided 100% protection at a dose of 30 mg/kg and 40% protection at a dose of 10 mg/kg (Fig. 6B andC). As shown in Fig. 6D, virus infection resulted in severe damage to the lung, characterized by diffuse alveolar damage, including thickening of the alveolar septa, marked epithelial hyperplasia in the bronchi/bronchioles, and extensive immune infiltration in the alveoli, bronchi, and vessels. The overall pathology scores were significantly lower in the mice pretreated with TAT-Nb7 (Fig. 6E). Following prophylactic treatment with TAT-Nb7, the viral load in the mouse tissues decreased significantly (Fig. 6F).
Next, we evaluated the therapeutic efficacy of TAT-Nb7 against IAV. As shown in Fig. 6B andC, TAT-Nb7 at a dose of 30 mg/kg provided complete protection for mice challenged with a lethal infection. The lung lesions of the mice treated with TAT-Nb7 were significantly alleviated (Fig. 6D andE). Moreover, following treatment with TAT-Nb7 at 30 mg/kg, the viral load in tissues was lower than that in mice treated with IgG (Fig. 6F).
Overall, the data indicated that TAT-Nb7 reduced the replication and pathogenicity of IAV, displaying promising protective efficacy against IAV infection.
## DISCUSSION
In the present study, we identified and characterized the Nb7 Nb, which specifically recognizes IAV NP. Nb7 inhibits IAV replication by blocking the nuclear export of the vRNP complex. Importantly, we found that the synergistic effects of the NP Q42, E46, and K48 residues are essential for the nuclear export of vRNP. Nb7 fused to the were transfected with Nb7-Flag, K7.13-Flag, or Vec along with pCG-PR8-NP. After 24 h, the cells were fixed and stained with a commercial anti-NP mAb (diluted 1:300). (C) The subcellular distribution of the NP in (B) was calculated via ImageJ. At least 30 cells from each group were counted. (D) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec along with PR8-NP.
After 24 h, the nuclear and cytoplasmic fractions of the cells were separated for Western blotting analysis. (E) A549 cells were transfected with Nb7-Flag, K7.13-Flag, or Vec. After 24 h, the cells were infected with the PR8 (H1N1) virus at a multiplicity of infection (MOI) of 10. At 2, 3, 4, and 5 h post-infection, the cells were fixed and stained with anti-Flag (diluted 1:300) and a commercial anti-NP mAb (diluted 1:300). (F) ImageJ was used to calculate the subcellular distribution of the NP in (E). At transmembrane peptide TAT displayed promising protection against lethal IAV challenge. These findings provide novel insights for the development of antiviral approaches.
IAVs pose a persistent threat to public health and are characterized by the capacity to undergo antigenic drift and shift rapidly. IAVs constantly cross the interspecies barrier and infect various hosts, as seen in the recent H5N1 infection in cows and sheep in the United States (28). The development of more effective and broad-spectrum strategies to cope with the frequent mutations of IAVs is urgently needed. Compared with surface proteins, internal proteins of IAV are generally more conserved, facilitating the develop ment of broad-spectrum antiviral strategies (29). NP is one of the most conserved IAV proteins, plays a crucial role in the assembly of vRNPs, and facilitates nucleus-cytoplasm shuttling. In the present study, we screened an NP-specific Nb7 with high affinity and broad-spectrum inhibition of IAV replication.
The surface proteins of IAV, that is, HA, NA, and M2, have long been the primary targets for the development of antiviral strategies (30,31). However, recent investigations revealed that small molecules and other agents targeting NP, PB1, M1, PB2, and other internal viral proteins exert promising antiviral effects (32)(33)(34)(35). More recently, Liu et al. reported that naproxen, which targets NP at conserved residues, blocked the interaction between CRM1 and NP and suppressed IAV replication in vivo and in vitro (36). Ashour et al. isolated several NP-targeting Nbs that exerted anti-IAV activity by blocking the nuclear import of vRNPs and viral transcription and replication in the nucleus (37). The high-resolution crystal structure revealed that Nbs inhibited the nuclear import of vRNPs by occupying the non-conserved surface of the NPs and ingeniously mimicked the antiviral mechanism of the host Mx protein (37)(38)(39). Our data revealed that Nb7 inhibits the nuclear export of vRNP, thereby suppressing the assembly of progeny virions and the replication of IAVs. The conserved key sites targeted by Nb7 were identified as crucial for the nuclear export of vRNP. In conclusion, conserved internal viral proteins are promising targets for the development of novel antiviral strategies.
While the conservation of the NP across all IAVs has positioned it as an attractive target, therapeutic exploitation against the NP has been hindered by a limited under standing of its functional epitopes. Our work demonstrated that Nb7 potently inhibits vRNP nuclear export by binding to the NES1 domain, a mechanism distinct from that of existing NP-targeting Nbs, such as VHH1 (39). The NES1 domain of the NP has multiple potential nuclear export motifs (LxxLxL or LxxxLxL), which mediate the nuclear export of nuclear proteins (10,40). In this study, we determined that the key binding sites of Nb7 are located between the hydrophobic amino acids of the nuclear export motif (FYIQMCTELKL) within the NES1 region of the NP (Fig. 4). Superposition of NP7 on EM-based vRNP models revealed that NP7 could mask the nuclear export motifs on NP molecules (Fig. 4). However, the flexibility and heterogeneity of vRNP complexes pose significant challenges to proving this argument. High-resolution structures of the vRNP complex with interacting proteins may still be needed for confirmation. Importantly, we revealed the synergistic role of the conserved residues Q42, E46, and K48 within NES1. When Q42/E46/K48 is co-mutated, the nuclear export of NP is blocked, as indicated by the accumulation of NP in the nucleus. Notably, the Q42/E46/K48A triple mutation is lethal for IAV. These findings highlighted the synergistic role of the Q42/E46/K48 sites in the function of NP and the life cycle of IAV. with the NP dimer. The dashed green lines represent hydrogen bonds. The predicted antigen-antibody interchain-predicted alignment error in the contact region was generally lower than 5 Å, and the pLDDTs of the key interface residues were all greater than 85. (D) A549 cells were transfected with NP or the indicated NP mutants. After 18 h, the cells were fixed and stained with a commercial anti-NP mAb (diluted 1:300) and Nb7-Fc (2 µg/mL). (E) HEK293T cells were transfected with NP-Fc along with NP or the indicated NP mutants. After 24 h, the cell lysates were co-IP with a commercial anti-NP mAb. (F) Superposition of the Nb7 structure on the vRNP model (PDB ID: 2YMN) from the work of PyMOL (Nb7: blue; NP oligomer: buff). (G) NP amino acid sequences of all subtypes of IAVs. NP sequences were downloaded from GenBank via the "collapse identical sequences" option. The logo was generated via the WebLogo3 online tool The development of antibody drugs targeting intracellular proteins faces challenges in crossing cellular or capsular barriers to directly bind to target antigens. Currently, lipid nanoparticle (LNP) delivery methods represent promising alternatives for drug delivery. The use of a lung-targeting delivery system to administer mRNAs encoding broadly neutralizing antibodies enables the direct intracellular expression of antibodies against virus infection (41). A major limitation of LNP-mRNA-expressing antibodies in infected cells is the suppression of host protein synthesis, which impairs antibody expression during the late stages of infection. Here, we employed the HIV TAT peptide, derived from successes in HIV therapeutics, to deliver Nb7 into cells. Our results demonstrated that TAT-Nb7 fully protected mice against lethal IAV replication (Fig. 6). The in vivo efficacy of TAT-Nb7 suggests that cell-penetrating peptide fusions could overcome the pharmacokinetic barriers that have hindered conventional antibody therapies targeting intracellular proteins such as NP. Despite the limitations of our present study, such as the lack of definitive structural determination of the Nb7-NP complex via cryo-EM, which may provide more detailed insights into the epitope of Nb, we successfully mapped the binding area of Nb7 as a conserved NES1 domain. In the future, we will attempt to modify Nb7 to improve its affinity and inhibitory effect on IAV replication. Better delivery methods can be employed to assess the in vivo applicability of Nb. In conclusion, this study demonstrates that the NP-specific Nb7 is a promising candidate for combating IAV infection. The conserved sites of the NP NES1 region we identified may contribute to the development of a broad-spectrum vaccine and provide insights into novel antiviral approaches.
## MATERIALS AND METHODS
## Biosafety and ethical statements
The details of the facility and the biosafety and biosecurity measures used have been previously reported (42).
## Cells, viruses, and plasmids
HEK293T and MDCK cells (ATCC) were maintained in Dulbecco's Modified Eagle's Medium supplemented with 10% fetal bovine serum (FBS; Life Technologies), 100 U/mL penicillin, and 100 µg/mL streptomycin (Life Technologies). A549 cells (ATCC) were grown in Kaighn's modified Ham F-12 nutrient mixture medium supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin. HEK293F cells were maintained in Expi Expression Medium (Gibco, 12338018). All the cells were cultured and maintained at 37°C with 5% CO 2 .
The H1N1 IAV (A/Puerto Rico/8/1934, PR8) was stored in our laboratory. H3N2 IAV (A/Swine/Shandong/TA05/2021) was isolated from Shandong Province, China, in 2021. The H6N6 virus (A/duck/Hunan/4/2018, HN4) was isolated in 2018 from a duck in Henan Province, China. The H9N2 virus (A/chicken/Hunan/38/2018, HN38) was isolated in 2018 from Gansu Province, China. The NDV (MG7 strain) was generated and stored in our laboratory (43). SeV was kindly provided by Hongkui Deng (Peking University, China). As previously described, a recombinant PR8 reporter virus expressing Nanoluciferase (PR8-Nluc) was generated (44). The PR8-GFP virus construction is described in this study (45). Virus stocks were propagated in specific-pathogen-free (SPF) chicken eggs and stored at -70°C until use.
The genes encoding NP/PB1/PB2/PA were constructed via standard molecular biology techniques. The expression of the viral RNA-like firefly luciferase gene under the control of the human RNA polymerase I promoter, via the pPol I-Luc plasmid, has been previously reported (46). All plasmid DNAs were purified from bacteria via a plasmid extraction kit (Omega).
## Reagents and antibodies
The antibodies used in this study were as follows: HRP-conjugated anti-HA (12013819001), anti-Myc (11814150001), and anti-GFP (11814460001) antibodies (Roche); rabbit anti-PB2 (GTX125926), anti-PB1 (GTX125923), and anti-PA (GTX118991) polyclonal antibodies (Genetex); HRP-conjugated anti-human IgG-Fc (SSA001) and anti-NP (11675-MM03T) antibodies (Sino Biological); anti-β actin (TA-09), anti-His tag (TA-02), and HRP-conjugated goat anti-rabbit IgG (ZB-2301) (Zsbio); HRP-conjugated goat anti-mouse secondary antibodies (ab102448); and a Cy3-conjugated anti-Flag antibody (A8592, Sigma).
The following reagents were used in this study: Freund's adjuvant was purchased from Sigma (Germany); anti-Flag agarose affinity beads (A2220), protease inhibitor cocktail (4693116001), and protein A/G agarose affinity beads (P6486/E3403) (Merck); DAPI (C1002), NP-40 (ST366), 4% paraformaldehyde fix solution (P0099), RNase A (ST578), normal rabbit IgG (A7016), and cell counting kit-8 (CCK8, C0037); Tween-80 (HY-Y1891) (MedChem-Express); and the Expi-Fectamine 293 Transfection Kit (Gibco, A14525). The jetPRIME transfection reagent was obtained from Polyplus (USA). SYBR Green I Master Mix was purchased from Roche (Germany).
## Alpaca immunization and library construction
In consistent with our previous study (47), 1-year-old male alpacas were selected for immunization with recombinant NP purified from HEK293T cells. Five consecutive immunizations were performed every 2 weeks. For the first immunization, 1 mg of purified recombinant NP (1 mg/mL) was emulsified with an equal amount of Freund's complete adjuvant to immunize alpacas. In the second to fifth immunizations, an equal amount of Freund's incomplete adjuvant was mixed with purified recombinant NP. After the final immunization, sera were collected from the immunized alpacas and used to test the titer of anti-NP antibodies via indirect ELISA, with recombinant NP (2 µg/mL) purified from E. coli cells serving as the coating antigen. Then, peripheral blood lymphocytes were isolated, and total RNA was extracted to synthesize cDNA. The VHH gene fragments were subsequently cloned and inserted into pComb3XSS via nested PCR. The recombi nant phage was converted into E. coli TG1 cells. The cells were incubated overnight at 37°C on a Luria-Bertani (LB) plate containing 2% glucose and 100 µg/mL ampicillin. On day 2, the colonies were scraped off the plate and stored in LB with 20% glycerin at -80°C.
## Screening and purification of Nbs
To obtain NP-specific Nbs, three rounds of panning were performed as previously described (47). The ELISA plate was coated with recombinant NP (2 µg/mL) purified from E. coli cells and sealed with a 1% PVA solution for 2 h. The original library obtained was added to the ELISA plate and incubated at room temperature for 2 h, after which freshly prepared 0.1 M HCl was added. The mixture was gently shaken at room temperature for 5 min, and 1 M Tris-HCl (pH 8.0) was quickly added to neutralize the mixture. The mixture was added to 1 mL of NEB5αF′ bacteria at the logarithmic stage and cultured at 200 rpm at 37°C for 1 h. M13-assisted phage was added, and the mixture was cultured at 200 rpm at 37°C for 1 h. The mixture was subsequently transferred to 2 × YT/Amp-Kan medium. The first-generation (F1) anti-NP phage library was collected by repeating the above operations. An appropriate amount of eluate was added after the third round of panning. The plate was then diluted, and the mixture was cultured overnight at 37°C. Forty-eight monoclonal antibodies were selected, inoculated into 96-well plates pre-coated with recombinant NP (2 µg/mL), and the supernatants were collected and stored at 4°C. The Nbs that reacted with the NP were screened via indirect ELISA, and all the VHHs of the positive clones were sequenced to obtain the final VHH sequences for analysis. VHH sequences were inserted into the pcDNA3.1 eukaryotic expression vector containing Fc. The plasmid was transfected into HEK293F suspension cells, and the supernatant was collected 7 days later. The supernatant was then filtered through a 0.45 µm filter. Nbs were purified with a Protein A column.
## Enzyme-linked immunosorbent assay
Purified virions (2 µg/mL) or PR8-NP proteins (2 µg/mL) were first immobilized on 96-well microtiter plates. The plates were blocked with 2% BSA for 1 h at room temperature. Gradient-diluted Nbs or Nbs with indicated concentrations were then added and allowed to react at 37°C for 2 h. HRP-conjugated goat anti-human IgG-Fc (diluted 1:8,000) was subsequently added, and the mixture was incubated at 37°C for 0.5 h. Sulfuric acid was sequentially added to the plates to stop the reaction. The data were recorded at 450 nm via an M1000 (Tecan) plate reader.
## Virus infection and titration
Viral infection was performed as previously described (46). All cells were seeded at the desired density in culture plates as per the requirements for different experiments. Viruses were inoculated into cells at a specific multiplicity of infection for different experiments. One hour after inoculation, the medium was replaced with fresh OPTI-MEM containing 0.1 µg/mL of tosylsulfonyl phenylalanyl chloromethyl ketone (TPCK)-trypsin and incubated at 37°C. Virus-containing culture supernatants were collected at the indicated timepoints for titration.
Virus titers of the stocks, cell culture supernatants, and tissue suspensions were determined by end-point titration in MDCK cells or eggs. For end-point viral titration in MDCK cells, each sample was serially diluted with OPTI-MEM containing 0.1 µg/mL of TPCK-trypsin and then inoculated into MDCK cells. Two days after inoculation, supernatants from the inoculated cells were collected and tested for their ability to agglutinate chicken erythrocytes or the expression of GFP, indicating viral replication. The infectious virus titers are reported as log 10 TCID 50 /mL and were calculated from three replicates using the Reed-Muench method. For end-point viral titration in eggs, 10-fold serial dilutions of each sample were inoculated into 9-day-old SPF eggs. 60 h after inoculation, fluid from the allantoic cavity was collected, and its ability to agglutinate chicken erythrocytes was tested as an indicator of viral replication. The infectious virus titers are reported as log 10 EID 50 /mL and were calculated from three replicates via the Reed-Muench method (48).
## TAT-Nbs production and purification
The gene fragments for Nb7-TAT or K7.13-TAT were generated by Tsingke Biotechnology Co., Ltd. The pCold II plasmid (Novagen, Darmstadt, Germany), containing a 6-histidine (6 His) tag upstream of the gene insertion site, was used as the expression vector. For Nb expression, the plasmids were transformed into E. coli BL21(DE3) after induction with 0.1 mmol/L isopropyl-D-1-thiogalactopyranoside overnight at 16°C. Centrifuged cells were resuspended in lysis buffer (1 × PBS, 0.2 mM PMSF, and 1% Triton X-100) and sonicated for 15 min. After centrifugation, the supernatant was incubated with Ni-BestaRose FF to purify the TAT-Nbs protein, as per the manufacturer's instructions.
## Co-IP analysis
HEK293T cells or A549 cells were cotransfected with the indicated plasmids with or without virus infection for 24 h. The transfected cells were then harvested and lysed in NP-40 lysis buffer (20 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1% NP-40, 1 mM EDTA with protease inhibitor cocktails). For each immunoprecipitation, 1 mL of lysate or lysate mixture was incubated for 4 h at 4°C with 0.5 µg of the indicated antibody or control IgG and 30 µL of protein A/G-Sepharose (Sigma). The beads were washed three times with 1 mL of lysis buffer containing 500 mM NaCl. The precipitates were then analyzed via standard Western blotting procedures.
## Western blotting
The cells or protein samples were lysed in RIPA buffer (Beyotime, China). Proteins were separated by 10% SDS-PAGE and transferred to a nitrocellulose membrane (Bio-Rad). The membrane was blocked for 1 h in TBST containing 5% milk and subsequently incuba ted with primary antibodies for 2 h. After a 1-h incubation with an HRP-conjugated secondary antibody, the immunoreactive bands were visualized using an e-BLOT system (e-BLOT Life Science, China). The intensities of the target bands were quantified by using the Image J program (NIH, USA).
## Affinity measurements
Affinity tests were performed on the Biacore T200 instrument. Briefly, the purified NP proteins were coupled and fixed to the M5 chip and activated. Nb7 was diluted from 25 nM to 1.5625 nM via a buffer solution (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.05% Tween 20) and flowed uniformly through the chip (20 µL/min). Finally, competitive elution was performed with a pH 4 glycine solution. At the same time, the negative control was set to remove the nonspecific binding signal, and the kinetic affinity constant was obtained by applying the binding model.
## Atomic model building and refinement
The model of Nb was done using the ColabFold version of AlphaFold 3 (25). Structures were analyzed and figures were generated using Open-Source PyMOL version 2.5.0. (http://www.pymol.org). The ClusPro 2.0 (49) server was used in "antibody mode" to produce docking models between Nb7 and NP or vRNP complex. The top 30 mod els returned by the server were ranked depending on energy and cluster size. The 10 best ClusPro docking models were further analyzed based on interactions and interface properties calculated according to PDBePISA (Proteins, Interfaces, Structures, and Assemblies, https://www.ebi.ac.uk/pdbe/prot_int/pistart.html) (50). The final model subsequently underwent manual analysis and image generation in Open-Source PyMOL version 2.5.0.
## Indirect IFA
An indirect IFA was performed as previously described (39). The infected or transfected cells were fixed with 4% paraformaldehyde and then washed three times with PBS. The cells were next permeabilized with 0.5% Triton X-100 in PBS and blocked with 5% skim milk for 1 h. Then, the cells were incubated with the indicated primary and secondary antibodies and counterstained with DAPI. The stained cells were observed with a Zeiss microscope (LSM 980). At least 30 cells were randomly selected for quantification in each experiment. The fluorescence intensity was quantified, and colocalization analysis was conducted with ImageJ software.
## Cross-linking immunoprecipitation
At 24 h post-transfection, cells were cross-linked by 1% formaldehyde solution and lysed in cell lysis buffer (10 mM Tris-HCl, pH 7.4, 100 mM NaCl, 0.5% NP-40, 1 mM DTT, 200 units/mL RNaseOUT, and EDTA-free Protease Inhibitor Cocktail). The lysed cell suspension was incubated with anti-NP antibody for 4 h at 4°C. Then, the mixture was incubated with protein A/G magnetic beads and rotated for 4 h at 4°C. The beads were then washed three times with NT2 buffer (50 mM Tris-HCl pH = 7.4, 150 mM NaCl, 1 mM MgCl 2 , and 0.05% NP-40), followed by incubation with 50 µg of proteinase K (Takara Bio) at 55°C for 30 min. Input and co-immunoprecipitated RNAs were recovered by TRIzol extraction and analyzed by qPCR.
## RNA isolation and quantitative PCR
Total RNA was extracted from the cells using TRIzol. In accordance with the manufactur er's protocol, total RNA was subsequently transcribed into cDNA using M-MLV reverse transcriptase (Promega). The levels of RNA were determined via real-time RT-PCR as described previously (46). Actin was used as an invariant control. Real-time PCR was performed using an ABI 7500 detection system (Applied Biosystems, USA). The RNA level of each gene is shown as a fold change (2 -ΔΔCT ) in the graphs. The sequences of the gene-specific primers used for qPCR are provided in Table S1.
## Dual-luciferase reporter assay
For the viral minigenome assay, HEK293T cells were transfected with pCAGGS constructs expressing the viral PB2, PB1, PA, and NP proteins from the PR8 virus; the pPol I-Luc construct; and an internal control, pRL-TK (Promega), along with other plasmids. The cells were incubated at 37°C for 24 h, and the cell lysates were subsequently prepared via the Dual-Luciferase Reporter Assay System (Promega).
## Subcellular fractionation
The cells were harvested by scraping into cell lysis buffer containing 10 mM HEPES (pH 7.4), 10 mM NaCl, 1 mM KH 2 PO 4 , 5 mM NaHCO 3 , 1 mM CaCl 2 , 0.5 mM MgCl 2 , and 5 mM EDTA, along with a complete protease inhibitor cocktail. The cell lysates were allowed to swell for 5 min, followed by homogenization for 50 strokes. Then, the cells were centrifuged at 800 × g for 5 min, generating a pellet containing nuclei and debris and a supernatant consisting of the cytosol and plasma. The pellets were resuspended in 1 mL of TSE buffer containing 10 mM Tris (pH 7.5), 300 mM sucrose, 1 mM EDTA, and 0.1% NP-40, along with a complete protease inhibitor cocktail. The suspension was then pelleted, resuspended, and washed twice with PBS. The final pellets were pure nuclei. The subcellular fractions were subsequently analyzed via Western blotting.
## Mouse study
The prophylaxis and therapy experiments in the mouse model were conducted as described previously (51). For prophylactic assessment, groups of 11 mice were intranasally (i.n.) administered a low (10 mg/kg) or high (30 mg/kg) dose of Nb7 at 12 h before challenge with 10 4 EID 50 of the PR8 virus. For therapy, groups of 11 mice were intranasally (i.n.) administered low (10 mg/kg) or high (30 mg/kg) doses at 2 and 24 h after challenge. The control mice were inoculated with IgG and infected in the same way. After inoculation, five mice in each group were monitored daily for 14 days for weight loss and mortality. The other six mice per group were euthanized on days 3 and 5 post-infection, and their lung and nasal turbinate tissues were collected for viral titration or pathological assessments by Servicebio (Servicebio Technology, China).
## Statistical analysis
The data are expressed as the means ± standard deviations. Statistical significance was determined via Student's two-tailed unpaired t test or analysis of variance (ANOVA) with GraphPad Prism software (version 9.0, San Diego, CA, USA). Differences between groups were considered significant when the P-value was <0.05 (*), <0.01 (**), <0.001 (***), or <0.0001 (****). "ns" indicates no significant difference.
## FUNDING
## References
1. Jiang, Chen, Li (2023) "Advances in deciphering the interactions between viral proteins of influenza A virus and host cellular proteins" *Cell Insight*
2. Jones, Andreev, Fabrizio et al. (2025) "Baloxavir improves disease outcomes in mice after intranasal or ocular infection with influenza A virus H5N1-contaminated cow's milk" *Nat Microbiol*
3. Zeng, Xu, Wei et al. (2021) "The PB1 protein of influenza A virus inhibits the innate immune response by targeting MAVS for NBR1-mediated selective autophagic degradation" *PLoS Pathog*
4. Luo, Zhang, Liang et al. (2018) "Phospholipid scramblase 1 interacts with influenza A virus NP, impairing its nuclear import and thereby suppressing virus replication" *PLoS Pathog*
5. Eisfeld, Neumann, Kawaoka (2015) "At the centre: influenza A virus ribonucleoproteins" *Nat Rev Microbiol*
6. Moreira, Weber, Bolte et al. (2016) "A conserved influenza A virus nucleoprotein code controls specific viral genome packaging" *Nat Commun*
7. Ozawa, Fujii, Muramoto et al. (2007) "Contributions of two nuclear localization signals of influenza A virus nucleoprotein to viral replication" *J Virol*
8. Neumann, Castrucci, Kawaoka (1997) "Nuclear import and export of influenza virus nucleoprotein" *J Virol*
9. Cros, García-Sastre, Palese (2005) "An unconventional NLS is critical for the nuclear import of the influenza A virus nucleoprotein and ribonucleoprotein" *Traffic*
11. Yu, Liu, Cao et al. (2012) "Identification and characterization of three novel nuclear export signals in the influenza A virus nucleoprotein" *J Virol*
12. Yu, Liu, Cao et al. (2012) "Identification and characterization of three novel nuclear export signals in the influenza A virus nucleoprotein" *J Virol*
13. Tarendeau, Boudet, Guilligay et al. (2007) "Structure and nuclear import function of the C-terminal domain of Full-Length Text Journal of Virology December"
14. "influenza virus polymerase PB2 subunit" *Nat Struct Mol Biol*
15. Dadonaite, Gilbertson, Knight et al. (2019) "The structure of the influenza A virus genome" *Nat Microbiol*
16. Yasuda, Nakada, Kato et al. (1993) "Molecular assembly of influenza virus: association of the NS2 protein with virion matrix" *Virology (Auckl)*
17. Elton, Simpson-Holley, Archer et al. (2001) "Interaction of the influenza virus nucleoprotein with the cellular CRM1-mediated nuclear export pathway" *J Virol*
18. Noton, Medcalf, Fisher et al. (2007) "Identification of the domains of the influenza A virus M1 matrix protein required for NP binding, oligomerization and incorporation into virions" *J Gen Virol*
19. Bannas, Hambach, Koch-Nolte (2017) "Nanobodies and nanobodybased human heavy chain antibodies as antitumor therapeutics" *Front Immunol*
20. Salvador, Vilaplana, Marco (2019) "Nanobody: outstanding features for diagnostic and therapeutic applications" *Anal Bioanal Chem*
21. Parray, Shukla, Perween et al. (2021) "Inhalation monoclonal antibody therapy: a new way to treat and manage respiratory infections" *Appl Microbiol Biotechnol*
22. Huang, Ying, Wu (2022) "Single-domain antibodies as therapeutics for respiratory RNA virus infections" *Viruses*
23. Zupancic, Desai, Schardt et al. (2021) "Directed evolution of potent neutralizing nanobodies against SARS-CoV-2 using CDR-swapping mutagenesis" *Cell Chem Biol*
25. Turrell, Lyall, Tiley et al. (2013) "The role and assembly mechanism of nucleoprotein in influenza A virus ribonucleo protein complexes" *Nat Commun*
26. Fan, Walker, Carrique et al. (2019) "Structures of influenza A virus RNA polymerase offer insight into viral genome replication" *Nature*
27. Wei, Zeng, Zhang et al. (2020) "The nucleoprotein of H7N9 influenza virus positively regulates TRAF3-mediated innate signaling and attenuates viral virulence in mice" *J Virol*
28. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
29. Vivès, Brodin, Lebleu (1997) "A truncated HIV-1 Tat protein basic domain rapidly translocates through the plasma membrane and accumulates in the cell nucleus" *J Biol Chem*
30. Yang, Ma, Song et al. (2002) "HIV-1 TAT-mediated protein transduction and subcellular localization using novel expression vectors" *FEBS Lett*
31. Shi, Kong, Cui et al. "2025. H5N1 virus invades the mammary glands of dairy cattle through "mouth-to-teat" transmission" *Natl Sci Rev*
32. Bessonne, Morel, Nevers et al. (2024) "Antiviral activity of intracellular nanobodies targeting the influenza virus RNApolymerase core" *PLoS Pathog*
33. Ray, Mohamed, Maurer et al. (2024) "Eliciting a single amino acid change by vaccination generates antibody protection against group 1 and group 2 influenza A viruses" *Immunity*
34. Momont, Dang, Zatta et al. (2023) "A pan-influenza antibody inhibiting neuraminidase via receptor mimicry" *Nature*
35. Yin, Li, Zhou et al. (2020) "Protein transduction domain-mediated influenza NP subunit vaccine generates a potent immune response and protection against influenza virus in mice" *Emerg Microbes Infect*
36. Huang, Chen, Zhang et al. (2018) "Identification of a novel compound targeting the nuclear export of influenza A virus nucleoprotein" *J Cellular Molecular Medi*
37. Wang, Lin, Zhong et al. (2021) "Cellular hnRNPAB binding to viral nucleoprotein inhibits flu virus replication by blocking nuclear export of viral mRNA"
38. Duan, Chen, Zhang et al. (2024) "A nanobody inhibiting porcine reproductive and respiratory syndrome virus replication via blocking self-interaction of viral nucleocapsid protein" *J Virol*
39. Zheng, Zhang, Jiao et al. (2019) "Naproxen exhibits broad antiinfluenza virus activity in mice by impeding viral nucleoprotein nuclear export" *Cell Rep*
40. Ashour, Schmidt, Hanke et al. (2015) "Intracellular expression of camelid single-domain antibodies specific for influenza virus nucleoprotein uncovers distinct features of its nuclear localization" *J Virol*
41. Ingram, Schmidt, Ploegh (2018) "Exploiting nanobodies' singular traits" *Annu Rev Immunol*
42. Hanke, Knockenhauer, Brewer et al. (2016) "The antiviral mechanism of an influenza A virus nucleoprotein-specific single-domain antibody fragment" *mBio*
43. La Cour, Kiemer, Mølgaard et al. (2004) "Analysis and prediction of leucine-rich nuclear export signals" *Protein Eng Des Sel*
44. Tai, Yang, Liu et al. (2023) "A lung-selective delivery of mRNA encoding broadly neutraliz ing antibody against SARS-CoV-2 infection" *Nat Commun*
45. Kong, Guan, Shi et al. (2021) "A single-amino-acid mutation at position 225 in hemagglutinin attenuates H5N6 influenza virus in mice" *Emerg Microbes Infect*
46. Ji, Liu, Du et al. (2018) "A novel genotype VII Newcastle disease virus vaccine candidate generated by mutation in the L and F genes confers improved protection in chickens" *Vet Microbiol*
47. Nogales, Ávila-Pérez, Rangel-Moreno et al. (2019) "A novel fluorescent and bioluminescent bireporter influenza A virus to evaluate viral infections" *J Virol*
48. Fukuyama, Katsura, Zhao et al. (2015) "Multi-spectral fluorescent reporter influenza viruses (Color-flu) as powerful tools for in vivo studies" *Nat Commun*
49. Wang, Xu, Shen et al. (2025) "N6-methyladnosine of vRNA facilitates influenza A virus replication by promoting the interaction of vRNA with polymerase proteins" *Proc Natl Acad Sci*
50. (2025) *Full-Length Text Journal of Virology*
51. Chen, Huang, Wang et al. (2025) "Influenza A virus H7 nanobody recognizes a conserved immunodominant epitope on hemagglutinin head and confers heterosubtypic protection" *Nat Commun*
52. Matumoto (1949) "A note on some points of calculation method of LD50 by Reed and Muench" *Jpn J Exp Med*
53. Kozakov, Hall, Xia et al. (2017) "The ClusPro web server for protein-protein docking" *Nat Protoc*
54. Krissinel, Henrick (2007) "Inference of macromolecular assemblies from crystalline state" *J Mol Biol*
55. Shen, Wang, Wang et al. (2022) "Specific monoclonal antibodies targeting unique HA epitopes block H7N9 influenza A viral replication" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12749565&blobtype=pdf | # Enabling genomic surveillance from 30 years of linked English sentinel network data: The Wellcome Quinquagenarian (QQG) Biomedical Resource
Simon De Lusignan, Praveen Sebastianpillai, Omid Parvizi, Cecilia Okusi, Mark Joy, Shuma Banik, Fatima Batool, Katja Hoschler, Beatrix Kele, Angie Lackenby, Joanna Ellis, Richard Pebody, Conall Watson, Jamie Lopez Bernal, Maria Zambon, Rakesh Mishra, Steven Holland
## Abstract
BackgroundThe World Health Organisation recommends integrating viral genome sequences and sentinel surveillance data. We report progress in linking clinical, virology, and sequence data to enable genomic surveillance of influenza, respiratory syncytial virus (RSV), and severeacute-respiratory-syndrome coronavirus-2 (SARS-CoV-2).
MethodsWe linked individual-level clinical data from the Oxford-Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) sentinel network to virology results from the UK Health Security Agency (UKHSA) reference virology laboratory. We identify where publicly accessible repositories, the Global Initiative on Sharing All Influenza Data (GISAID), or others hold viral genome sequence data from test-positive cases. Our metadata also identifies test-negative controls contemporaneous to test-positive cases. We summarise the scope of data availability in the Wellcome Quinquagenarian (QQG)
## Introduction
The emergence of new viral variants can result in the further spread of disease and reduce the effectiveness of countermeasure programs, as was seen following the spread of drug-resistant influenza in 2008 1 and during the COVID-19 pandemic. The recent introduction of vaccine and monoclonal antibody therapies for respiratory syncytial virus (RSV) will also require contemporary monitoring of viral diversity 2 . To monitor such changes, the World Health Organisation's (WHO) 10-year global genomic surveillance strategy recommends public health authorities integrate genetic sequence data (GSD) into disease surveillance 3,4 .
Integrating genomic and clinical data will enhance the genomic surveillance of viruses of public health significance 5 . Timely virological surveillance can link viral gene sequence data with clinical characteristics of circulating strains, which, when further linked to vaccine exposure and disease burden data, enables estimates of vaccine effectiveness (VE) by viral variant, age, and severity of illness being reported. Genomic surveillance has already been implemented in acute and primary care settings for variants of SARS-CoV-2 with evidence of its utility during the pandemic period 1,6,7 .
Establishing the Wellcome Quinquagenarian (QQG) resource for a range of seasonal respiratory viruses will provide closer to real-time evidence of the impact of countermeasures in a systematic and consistent framework, which can be scaled as needed with the emergence of significant variants.
We aim to create a biomedical resource that captures England's systematic genomic surveillance of influenza, RSV, and (SARS-CoV-2). Sentinel clinical surveillance started in the 1966-67 season 8 . Current practice is built upon a longstanding collaboration between the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) and the UK Health Security Agency (UKHSA) and its government public health agency predecessors, with the University of Oxford a more recent addition to this partnership 9,10 . Prospective combined virological and clinical surveillance for influenza began in the winter of 1992-3 and gradually increased its sophistication 11,12 . RSV detection was added in 2001, with sequencing of this virus added retrospectively 13 . Detection of SARS-CoV-2 was added immediately in 2020 as part of an emergency response to the pandemic, with sequence analysis of positive samples added as routine from the start 14 .
There has been no systematic curation or consolidation of these data; clinical, laboratory, and genomic data currently sit in separate repositories. The recently published Sudlow review, commissioned by England's Chief Medical Officer articulates the barriers that arise from the UK's complex and inefficient systems for managing and accessing health data and the potential role of major national public bodies with responsibility for or interest in health data to improve critical national infrastructure for data usage 15 . In the 55 th (quinquagenarian) year of primary care sentinel surveillance we conceived creating a resource based on the combined clinical and virological data arising from a long-standing national community surveillance programme that could be interrogated by independent researchers 16 . The resource was named the Wellcome Quinquagenarian (QQG) Biomedical Resource. Below, we describe its history, scope, and components.
## Methods
## Components of the Wellcome Quinquagenarian (QQG) biomedical resource
The three data components combined to make up the QQG biomedical resource were: (1) clinical data from the RSC, the English national primary care sentinel network, (2) virology data from UKHSA's Respiratory Reference Virology Laboratory information system (LIMS) arising from sampling of a subset of cases for whom clinical data was recorded, and (3) virology sequence data stored in sequence data repositories, deposited by UKHSA or predecessors (Figure 1).
## Data linkage, and virology sampling
The UK has a registration-based primary care system using a unique identifier (NHS number). This enabled the RSC to link an individual's pseudonymised computerised medical record (CMR) data to other health systems and national data, making it readily usable for surveillance and research 17 . At an individual level, pseudonymised patient records were routinely linked to a geographical location, and using the Lower Super Output Area (LSOA) from the Office of National Statistics (ONS) census which is used for reporting small geographical areas of between 1,000-3,000 people 18 , hospital attendance and admission (Hospital Episode Statistics (HES) data) 19 , and death data. The NHS Personal Demographics Service (PDS) ensures that the date of death is recorded in the GP record 20 and the Office of National Statistics (ONS) provides links to death certificate data 21 .
Most pre-school and adult vaccinations were carried out in primary care, and where vaccine exposure happens elsewhere (mainly pharmacies and schools), records of these events were either directly transferred into the primary care record or the GP was notified. This transfer is routinely in place for influenza, RSV, and COVID-19 vaccines 22 . NHS number was also used to link virology reference laboratory results to the GP CMR, each sample also had a laboratory information management system (LIMS) identifier (ID). The latter facilitated clinical, laboratory, and viral genomic sequence data linkage.
From the 1992-1993 winter season onwards, a subset of RSC practices collected virology samples from cases of influenza-like illness (ILI). Up to the COVID-19 pandemic year of 2020 this was a seasonal collection during the winter months. Virology sampling generally took place between the International Organisation for Standardisation (ISO) week 40 and week 20 of the following year. Reflecting the focus on influenza, clinically defined cases of influenza-like illness (ILI) were eligible for sampling 23 . Whilst ILI was the RSC's long-term clinical indicator for sampling new episodes of illness within 7 days of illness onset, clinical sampling was extended in 2012 to include acute bronchitis in children under 5 years old, coinciding with the first pilot of live attenuated influenza vaccine (LAIV) 9 , though there had been longer-term interest in the association of acute bronchitis and winter pressures 24 , and the importance of RSV in those who present with acute bronchitis 25 . From 2020, as part of the pandemic response, sampling changed to become year-round and included any clinical presentation of an acute respiratory infection (ARI) 10 , with larger numbers of samples collected (up to 1,000 per week) and a broader panel of viruses tested for 26.
Throughout the entire period, the RSC conducted virological sampling of the nasopharynx, using two swabs (one nasal and one throat) placed into a single vial of Virus Transport Medium (VTM), with samples sent through the postal system to the UKHSA reference laboratory. Most samples were taken by healthcare professionals, although intermittent patient self-swabbing was implemented before 2020 27 , and became a permanent parallel stream from 2020 28 .
## Respiratory virus laboratory analysis
Swabs collected in Virus Transport Media (VTM) were transported to the laboratory through the post at ambient temperature, with a mean time to arrival of 2-3 days 12 . Each sample received was given a unique LIMS identifier and processed for the molecular detection of a range of viruses, with residual sample material stored at -80C.
Assays used for the detection and characterisation of influenza A and B inevitably changed over time, to take account of genetic drift in influenza and the evolution of molecular detection techniques (Figure S1). Techniques were based on the use of reverse transcription polymerase chain reactions (RT-PCR) for the detection of viral targets in different multiplex formats, updated regularly 12,29 .
The methodology for influenza genomic sequence reporting evolved from partial genome sequencing of the viral haemagglutinin (HA) gene using Sanger sequencing, then adding viral neuraminidase (NA) genes, and from 2009 onwards completing whole genome sequencing (WGS) of influenza using Illumina platforms. Molecular analysis of influenza was accompanied by phenotypic characterisation of selected virus isolates, including analysis of antiviral susceptibility to neuraminidase inhibitors based on the culture of virus isolates from residual VTM samples. This followed the recommendations of the WHO Global Influenza Surveillance and Response System (GISRS) for testing the antiviral susceptibility of influenza viruses 30 . Scanning for altered antiviral susceptibility is now conducted using single nucleotide polymorphism (SNP) screening from WGS 31 to allow the identification of common resistance markers. RSV A and B PCR detection was targeted on the highly conserved regions of the genome, with little variation, with retrospective use of samples to generate whole genome sequences 32,33 . SARS-CoV-2 detection also involved multiple target detection of conserved regions of the genome. These included the large open reading frame (ORF1ab) that encodes viral polyproteins, the E gene that encodes the envelope protein, and probes and primers to enable detection and amplification of these regions of the SARS-CoV-2 genome 33 . Results were reported with RT-PCR cycle threshold (Ct) values provided for each assay target. RT-PCR virus assays with a cycle-threshold (Ct) value of under 40 were regarded as positive. In general, samples with a positive Ct value of <30 provided good-quality WGS data 34,35 .
## Repositories holding viral sequenced data
The Global Initiative on Sharing All Influenza Data (GISAID) has been the primary location used to hold influenza and RSV sequence data. GISAID was established in 2008 as a not-for-profit organisation to make sequenced data available for scientific study. Each set of sequenced data deposited has been provided a unique and permanent identifier 36,37 . SARS-CoV-2 sequence data were deposited with the Cloud Infrastructure for Microbial Bioinformatics (CLIMB-COVID) developed by the COVID-19 Genomics UK Consortium (COG-ID) in response to the SARS-CoV-2 pandemic. The metadata captured included the date of sampling, geographical location, and sequence technology used 38,39 . Whilst this viral genome sequence repository was used, its overlap with GISAID deposition was beyond the scope of this paper; we include an inventory of RSC-derived data deposited in GISAID only. Sequence data before 2008 were stored locally within UKHSA and were excluded at this stage from our results.
## Linking process
The GISAID number was the primary key we used to link sequence data with reference virology laboratory data and clinical data for influenza and RSV. Viral genomic sequence data were deposited in GISAID (for influenza and RSV) and CLIMB-COVID (for SARS-CoV-2) by UKHSA. These data included the UKHSA LIMS number, which is a unique identifier (ID) for the virological sample. This enabled the linkage of GISAID data to reference virology laboratory data. Virological samples with the LIMS ID were also stored with the patient's pseudonymised NHS number, the unique NHS identifier used throughout the health system, which facilitated additional linkage to the primary care CMR and other health system data.
We also set up a process to enable contemporaneous test-negative controls to be identified. The latter may be needed for any test-negative design (TND) vaccine effectiveness (VE) studies being undertaken 40 . TND is commonly used to assess VE for a range of vaccines 41,42 .
## Data summary
A data summary will be placed online. The number of sequenced samples will also become part of the RSC's Annual Report. How these data might be visualised is described (Figure S4 andS5).
## Results
## Sentinel network data
The RSC has grown in terms of size, scope, integration virological testing, and data linkage. When the RSC started sentinel surveillance in 1967, general practice members collected data on paper spreadsheets which were sent to the RCGP's Birmingham Research Unit (BRU) for collation 43 . From 1994 onwards, data flows were progressively computerised. The RSC leadership moved to the University of Surrey and in 2015, a new pseudonymised flow of data commenced 44 , and subsequently to the University of Oxford with data flowing in 2021. Data were stored in the Oxford Royal College of General Practitioners Clinical Informatics Digital Hub (ORCHID) database, hosted by the Nuffield Department of Primary Care Health Sciences, University of Oxford, a trusted research environment (TRE) 10 .
Pseudonymisation used an NHS England-approved method allowing linkage to other health datasets 45 . We used a nonreversible approach, the Secure Hash Algorithm 512 (SHA 512). SHA 512 is a commonly used approach. We convert the NHS number into a fixed-size string. Each output produces a SHA-512 length of 512 bits (64 bytes). We added a salt before hashing to make data more secure. Between 1967 and 1997, aggregated clinical data were collected from paper records onto spreadsheets in individual general practices, forwarded to BRU, and stored on a Microsoft Access database (BRU-Access). In 1994, the first Computerised Medical Record (CMR) data started flowing to BRU and was stored on a Microsoft SQL database (BRU-SQL). In 2015, BRU-SQL was replaced by the University of Surrey-based SQL Real World Evidence (RWE) database. Retrospective data from 2004 were included in this new database, to coincide with the date when pay-for-performance for chronic disease management started in primary care 46 and GP CMR systems were linked to pathology laboratories. The net effect of the Quality and Outcomes Framework (QOF), was to incentivize improved data recording in primary care, leading to better data quality particularly associated with cardiovascular comorbidities. The role of electronic laboratory links in enhancing data quality also started to be recognised at this time 47 . Such links enable the seamless transfer of pathology data into primary care records, ensuring completeness and accuracy.
Together, QOF and electronic laboratory links have contributed to the high quality of data in UK general practice, making it suitable for research and quality improvement initiatives 48 . This database moved to Oxford in 2021 and was renamed ORCHID. Figure 2 illustrates how longitudinal ILI data can be combined, the re-extraction of BRU data is very similar to that previously reported 9 .
The RSC in 1977 had 39 general practices, representing a patient population of around 200,000 28 ; rising to over 100 practices covering a population of over 1 million in 2016 29 ; then growing to 1,879 practices, a population of 17 million, 31% of the English national population in 2021 49 . The RSC has recruited its practices and the subset of virology sampling practices to be nationally representative (Figure 3).
## Respiratory virology reference laboratory data
We report on the availability of influenza, RSV, and SARS-CoV-2 sequence data from virology swab samples where these viruses were detected. Before the 2009 influenza pandemic caused by H1N1, the number of positive samples for influenza and RSV combined was generally between 100 and 700 over the course of the winter seasons, the proportion of positive varying by week across the epidemic period. During the peak of the ILI consultation rate periods, normally lasting 6-8 weeks, the rate of influenza positivity increased up to 50-60% from <5% before the onset of sustained influenza circulation. In 2009 the number of positive influenza samples rose to over 1,600 and steadily increased thereafter. From 2020, testing for SARS-CoV-2 commenced with a switch to all-year-round sampling from 2021 onwards (Figure 4).
Between 2002 and 2023, influenza A viruses, either H1N1 or H3N2 were the predominant influenza viruses detected and typically co-circulated in most seasons with varying proportions. The H1N1 (H1) subtype caused a pandemic in 2009 (H1N1pdm09), replacing the previous seasonal H1N1 virus completely during 2009, with the highest proportion of H1N1(pdm09) detected in the winter periods (2010-2012) immediately following the pandemic of 2009 (Figure 4B). There has been co-circulation of both Victoria and Yamagata lineages of influenza B over these 20 years, but the latter influenza B subtype has not been detected since 2019 through RSC sampling. The detection of transmissible influenza resistance to oseltamivir in circulating seasonal H1N1 before the 2009 H1N1 pandemic was an example of the clinical utility of testing the circulating influenza A subtypes for antiviral susceptibility, providing an estimate of the proportion of circulating viruses with altered antiviral susceptibility 50,51 . A sporadic case of influenza A H1N2v zoonotic infection from swine was also detected in 2023 through the RSC virological swabbing programme in an area of England with the densest swine population. Part of the incident response to this unexpected detection event included an escalation of virological sampling of cases of ARI in the surrounding localities 52 .
Since 2003, similar rates of RSV A and B subtypes have been identified in each season, with a gradual expansion of sampling. Between 1997 when RSV testing was introduced, and 2003 RSV A predominated. Most recently, in 2023 the RSC collected 827 positive RSV samples: 53% (n=442) RSV A and 47% (n=385) RSV B. RSV containing samples from 2008 onwards have been used for WGS analysis, if technically suitable, to underpin studies of RSV viral diversity in England.
## SARS-CoV-2 testing of virological samples was included from
March 2020 onwards, as part of the pandemic response, but only samples collected from symptomatic patients were included, with ILI as the main clinical indicator at the time. A total of 5,068 positive samples were collected, 67% (n=2,406) in the first two years of the pandemic in 2020 and 2021. These samples were all submitted for WGS analysis, if technically suitable.
## GISAID-held viral genomic sequence data
Since 1992, 22,529 submitted samples have been positive for either influenza A or B, RSV or SARS-CoV-2 of which the majority 60.7% (n=13,665) were influenza, reflecting the origin and purpose of the virological testing programme, intended as a method of monitoring the circulation of influenza A and B in the community. A smaller proportion were RSV {16.8% (n=3,791)} and 22.5% (n=5,068) were SARS-CoV-2 (Table S2). Just under a third of these samples overall 29% (n=6,556) underwent whole genome sequencing to monitor viral diversity and provide information for the selection of samples for virus isolation and antigenic analysis, selected mainly on technical suitability for sequence analysis (sufficient sample, well preserved with adequate viral load).
Over 100 H1N1 WGS were obtained in 2009 as part of a scaled-up response to the 2009 pandemic at the time, this represented a major increase in viral sequencing activity, with the use of sequence data to track viral evolution during the early course of the 2009 pandemic. Sequence analysis was reduced between 2010 to 2013, following the de-escalation of the pandemic response. From 2014 onwards, there were increasing numbers of influenza whole genome sequences (WGS) generated each year, reflecting the gradual improvement in higher throughput laboratory sequencing methodologies, up to several hundred of each influenza subtype in the years before the pandemic of 2020. The number of influenza whole genome sequences generated reduced to 69 in 2020 and 6 in 2021, as a result of interrupted influenza transmission arising from pandemic lockdown measures (Table 1). Between 2009 to 2023, a total of 2,819 influenza WGS sequences were generated (Table 1). 97.1% of these can be linked to RSC clinical data.
Over half of all influenza sequences derived from RSC sampling stored on GISAID since 2009 were the influenza A (H3N2) subtype (51.4%, n=1,449), reflecting the dominance of circulation of this subtype in England over the time period (Figure 4B). The H1N1 subtype of influenza A contributed (20.7%, n=583), with influenza B (27.9%, n=787). 33% of all UKHSA influenza sequence data stored on GISAID are represented by samples collected by the RSC, comprising a geographically representative sample of viruses circulating in the community over this period of time.
RSV viral genome sequencing (WGS) was undertaken retrospectively using RSC samples archived since 2008. (Table 2), using a variety of sequencing methodologies, (described in Talts et al 2023). These were mainly RSV B (59.2%, n=741), with the remainder RSV A (41%, n=510). The number of RSV-positive samples increased in recent years, reflecting the increased sampling of younger age groups, following the introduction of the LAIV influenza vaccine in 2013/14, with the exception during the COVID-19 pandemic years. A large proportion of RSV samples with WGS deposited in GISAID (96.8%) could be linked to clinical records. 75% of all UKHSA RSV sequence data stored on GISAID are represented by data from samples collected by the RSC, and a detailed phylogenetic analysis of RSV strain diversity over this period is underway.
Only 3.1% of all UKHSA SARS-CoV-2 sequence data stored on GISAID are represented by data from samples collected by the RSC (N=2,486) (Table 3), reflecting the massive scale-up of community sampling and viral WGS sequencing in the UK over the pandemic period. The period of maximum sequencing occurred in 2021, (N=1,365) (Table 3), with gradual de-escalation since this time period. 98.9% of these sequenced samples could be linked to their clinical record providing the most complete linkage of the three viruses included in our analysis. During this period of time, the waves of different SARS-CoV-2 viral variants could be seen (data not seen).
Table 4 provides a summary overview of all UKHSA influenza, RSV and SARS-CoV-2 samples sequenced from 2008 to 2023 held in GISAID, over 97% of all RSC samples received at the UKHSA laboratory were linked to a clinical record.
## Discussion
## Principal findings
There is international acceptance of the importance of genomic surveillance 53,54 . With calls for a global network of laboratories generating sequence data 55 . We have demonstrated how clinical records, virology results, and viral genome sequences obtained from sentinel surveillance programmes can be linked up in a systematic and consistent manner retrospectively, with pseudonymised data being made available for independent analysis. Going forward we are building community-based surveillance systems which are scalable for responses needed during pandemic periods, with intrinsic sequence data linkage to clinical metadata, to provide the analytical capability to rapidly assess circulating virus diversity against the outcome of interventions.
We have reported the number of samples collected since 1993, but focussed on samples with sequenced whole viral genomes Table 4. Summary of the number of RSC sequences in GISAID with the percentage of total virology samples sequenced.
## Year
## Influenza A (HIN1) n (%)
Influenza A (H3N2) n (%)
## Influenza B n (%)
## Total
## Influenza n (%)
RSVA n (%)
## RSVB n (%)
Total RSV n (%) There have been some projects that have created complex performance federated environments for genomic surveillance 56 , however, considerable care is required to ensure that these are privacy preserving environments, as we have developed in this project, and will continue prospectively for these three respiratory viruses 57,58 . The UK has an overall initiative in ARI genomic surveillance 59 . A Scottish study using 150 influenza A (H3N2) linked clinical and genomic sequence data was able to draw epidemiological insights 60 . Little appears to have been completed for RSV yet, though work is in progress and wastewater analyses have been conducted 61 . The introduction of public health interventions such as RSV vaccine programmes emphasise the need for systematic genomic surveillance of this virus.
$$SARS-CoV-2 n (%) 2008* - - - - 0 (0) 8 (25.8) 8 (21.6) 0 (0)2009 109 (8.1) 0 (0) 0 (0) 109 (7.4) 2 (100) 21 (15.8) 23 (17) 0 (0) 2010 12 (1.5) 0 (0) 1 (0.3) 13 (1.1) 1 (50) 15 (21.7) 16 (22.5) 0 (0) 2011 1 (0.4) 2 (11.8) 0 (0) 3 (0.5) 0 (0) 7 (14) 7 (14) 0 (0$$
## Implications of the findings
The RSC, a longstanding community surveillance programme is developing the capability for sustained genomic surveillance of viruses of public health significance to build a scalable system for pandemic and interpandemic monitoring, linking clinical disease information to virological detection and assessment of viral diversity. Such a system is needed to provide analytical capability to monitor the effectiveness of the increasingly complex vaccine delivery programmes for influenza, RSV and SARS-CoV-2, across different segments of the population.
## Comparison with the literature
The widespread testing and need for rapid data access, meant large numbers of WGS were available for COVID-19. Utilisation of information was of clear clinical benefit 62,63 . The smaller number of SARS-CoV-2 WGS in the RSC are important for assessing the timeliness of detection through sentinel surveillance, compared with large scale clinical testing, providing valuable information about potential delays arising in detection of emerging viruses, and intensity of sampling needed for rapid detection of new variants. Long-term follow-up of the sequelae of SARS-CoV-2 infection, for example, people with long covid who can readily be identified from GP records 64 is also a valuable asset for assessment of clinical outcomes. Our findings around the disappearance of the influenza B Yamagata lineage has also been reported internationally 65 .
## Strengths and limitations
The strength of these resources is the strength and longevity of the over 57-year RCGP-UKHSA partnership, relatively newly reinforced by Oxford in 2018. Over this period there was an emphasis on high-quality computerised medical record (CMR) quality, with virology sampling since the 1992/1993 winter, with viral WGS sequence data deposited in GISAID since 2008. This partnership has enabled us to create this unique resource. Some of its collections, such as an uninterrupted series of RSV viruses since 2008 prior to the introduction of a vaccination programme in 2024 will provide important insights into viral evolution when uncontested by vaccination and can be used as a comparator to viral evolution under vaccine pressure from 2024 onwards.
The main limitations of our work were the scope of our clinical data, sampling and sequencing, and the lack of federation for searching and analysing these data systematically, including for the provision of permissions to use this data asset. Our clinical data were routinely collected into primary care CMR and coded into those records, and inevitably there will have been data quality issues. The criteria for sampling changed over time, samples from 2023 can be from those with any ARI, and samples were collected all year round from 2020. Potential users will need to apply separately to each organisation for permission to use its data, even though we will create a single webpage through which to do this.
## Further research
Research using these data will strengthen the case for their further development. The growing demand for enhanced analytical capabilities may drive the sequencing of additional virology samples and would increase the statistical power of analysis.
Integration with other repositories could also be explored.
How best to federate these data and permission to use them is the critical next step to promote their usability. Given the volume of sequence data, such a federated approach will need to include a high-performance computing environment. Given the policy constraints of processing health data outside NHS England-approved Secure Data Environments 66 , UKHSA's approved secure data environment, the Enterprise Data Analytics Platform (EDAP), and other options are being explored 67,68 .
## Conclusion
This paper describes the progress made to enable English primary care sentinel data and National Public Health Institute viral sequence genomic data to be available to enable independent analysis. The limitation of this work is the limited range of years of genomic surveillance data available over the 57-year history of the RSC, the lack of a single repository of these data, and the federated environment. However, it remains an achievement that of over 22,000 virology samples nearly 7,000 of these have sequenced data available for use in GISAID, with high linkage rates of sequenced genomes to RSC clinical data. We have undertaken ambitious strides towards enabling genomic surveillance.
The Access to RSC data is restricted due to ethical and data governance requirements. The dataset includes sensitive health information, and public sharing is not permitted. The University of Oxford's Research Ethics Committee, along with the RCGP and UKHSA (as joint data controllers), have approved these data governance protocols.
Researchers may apply to access the RSC dataset through a formal process, which includes:
-Submission of a study protocol for review, -Induction onto University of Oxford IT systems (following initial approvals), -Specification of data requirements and demonstration of compliance with information governance standards for any data transfers to open environments.
Research studies typically require approval from the University of Oxford, RCGP, UKHSA, and, where applicable, the Health Research Authority (HRA). The review process generally takes 21 working days, and pre-grant submission approvals can be obtained.
For details on how to apply and current timelines, please visit the website -https://www.phc.ox.ac.uk/intranet/better-workplacegroups-committees-open-meetings/primdisc-committee-1.n
## References
1. Hill, Githinji, Vogels (2023) "Toward a global virus genomic surveillance network" *Cell Host Microbe*
2. Ruckwardt (2023) "The road to approved vaccines for respiratory syncytial virus" *PubMed Abstract | Publisher Full Text | Free Full Text*
3. "Considerations for developing a national genomic surveillance strategy or action plan for pathogens with pandemic and epidemic potential"
4. Carter, Yu, Sacks (2022) "An appeal for strengthening genomic pathogen surveillance to improve pandemic preparedness and infection prevention: the German perspective" *PubMed Abstract | Publisher Full Text | Free Full Text*
5. Robishaw, Alter, Solano (2021) "Genomic surveillance to combat COVID-19: challenges and opportunities" *PubMed Abstract | Publisher Full Text | Free Full Text*
6. Foxley-Marrable, Cruz, Meredith (2023) "Combining viral genomics and clinical data to assess risk factors for severe COVID-19 (mortality, ICU admission, or intubation) amongst hospital patients in a large acute UK NHS hospital Trust" *PubMed Abstract | Publisher Full Text | Free Full Text*
7. Andrews, Stowe, Kirsebom (2022) "COVID-19 vaccine effectiveness against the Omicron (B.1.1.529) variant" *N Engl J Med*
8. Elliot, Fleming (2006) "Surveillance of influenza-like illness in England and Wales during 1966-2006" *Euro Surveill*
9. Fleming, Elliot (2008) "Lessons from 40 years' surveillance of influenza in England and Wales" *Epidemiol Infect*
10. De Lusignan, Correa, Smith (2017) "RCGP Research and Surveillance Centre: 50 years' surveillance of influenza, infections, and respiratory conditions" *Br J Gen Pract*
11. Fleming, Chakraverty, Sadler (1995) "Combined clinical and virological surveillance of influenza in winters of 1992 and 1993-4" *BMJ*
13. Ellis, Fleming, Zambon (1997) "Multiplex reverse transcription-PCR for surveillance of influenza A and B viruses in England and Wales in 1995 and 1996" *J Clin Microbiol*
15. Zambon, Stockton, Clewley (2001) "Contribution of influenza and respiratory syncytial virus to community cases of influenza-like illness: an observational study" *Lancet*
16. Bernal, Andrews, Gower (2021) "Effectiveness of COVID-19 Vaccines against the B.1.617.2 (Delta) Variant" *PubMed Abstract | Publisher Full Text | Free Full Text*
17. Sudlow (2024) "Uniting the UK's health data: a huge opportunity for society" *Zenodo*
18. Morley, Rocher (2024) "Building infrastructure is key to unifying UK health data" *BMJ*
19. De Lusignan, Liaw, Krause (2011) "Key concepts to assess the readiness of data for international research: data quality, lineage and provenance, extraction and processing errors, traceability, and curation. Contribution of the IMIA Primary Health Care Informatics Working Group" *Yearb Med Inform*
20. "NHS Data Model and Dictionary: Lower Layer Super Output Areas (LSOA)"
21. England "Hospital Episode Statistics (HES"
22. England *Personal Demographics Service. Reference Source*
23. *Office for National Statistics. (ONS): Deaths. Reference Source*
24. De Lusignan, Hobbs, Sheikh (2024) "Lessons from the English primary care sentinel network's response to the COVID-19 pandemic" *Lancet Infect Dis*
25. Pebody, Green, Andrews (2014) "Uptake and impact of a new live attenuated influenza vaccine programme in England: early results of a pilot in primary school-age children, 2013/14 influenza season" *Euro Surveill*
26. Elliot, Cross, Fleming (2008) "Acute respiratory infections and winter pressures on hospital admissions in England and Wales 1990-2005" *J Public Health (Oxf)*
27. De Lusignan, Bernal, Zambon (2020) "Emergence of a novel coronavirus (COVID-19): protocol for extending surveillance used by the Royal College of General Practitioners Research and Surveillance Centre and Public Health England" *PubMed Abstract | Publisher Full Text | Free Full Text*
28. De Lusignan, Jones, Dorward (2020) "The Oxford Royal College of General Practitioners Clinical Informatics Digital Hub: protocol to develop extended COVID-19 surveillance and trial platforms" *PubMed Abstract | Publisher Full Text | Free Full Text*
29. Elliot, Bermingham, Charlett (2015) "Self-sampling for community respiratory illness: a new tool for national virological surveillance" *Euro Surveill*
30. Gu, Watson, Agrawal (2024) "Postpandemic sentinel surveillance of respiratory diseases in the context of the World Health Organization mosaic framework: protocol for a development and evaluation study involving the English primary care network 2023-2024" *PubMed Abstract | Publisher Full Text | Free Full Text*
31. Stockton, Ellis, Saville (1998) "Multiplex PCR for typing and subtyping influenza and respiratory syncytial viruses" *J Clin Microbiol*
32. Zambon (2013) "Surveillance for antiviral resistance" *Influenza Other Respir Viruses*
34. Galiano, Agapow, Thompson (2011) "Evolutionary pathways of the pandemic influenza A (H1N1) 2009 in the UK" *PubMed Abstract | Publisher Full Text | Free Full Text*
35. Talts, Mosscrop, Williams (2024) "Robust and sensitive amplicon-based whole-genome sequencing assay of Respiratory Syncytial Virus subtype A and B. Microbiol Spectr" *PubMed Abstract | Publisher Full Text | Free Full Text*
36. Ladhani, Chow, Janarthanan (2020) "PubMed Abstract | Publisher Full Text | Free Full Text 35. World Health Organization: Operational considerations to expedite genomic sequencing component of GISRS surveillance of SARS-CoV-2" *Global Influenza Programme (GIP)*
37. Singanayagam, Patel, Charlett (2020) "Duration of infectiousness and correlation with RT-PCR Cycle threshold values in cases of COVID-19"
38. Shu, Mccauley (2017) "GISAID: Global Initiative on Sharing All Influenza Data -from vision to reality" *Euro Surveill*
39. Bogner, Capua, Lipman (2006) "A global initiative on sharing avian flu data" *Nature*
40. Nicholls, Poplawski, Bull (2021) "CLIMB-COVID: continuous integration supporting decentralised sequencing for SARS-CoV-2 genomic surveillance" *Genome Biol*
41. Dusetzina, Tyree, Meyer (2014) "Linking Data for Health Services Research: A Framework and Instructional Guide"
42. Chua, Feng, Lewnard (2020) "The use of test-negative controls to monitor vaccine effectiveness: a systematic review of methodology" *Epidemiology*
44. Pebody, Whitaker, Ellis (2020) "End of season influenza vaccine effectiveness in primary care in adults and children in the United Kingdom in 2018/19" *Vaccine*
45. Whitaker, Tsang, Byford (2022) "Pfizer-BioNTech and Oxford AstraZeneca COVID-19 vaccine effectiveness and immune response amongst individuals in clinical risk groups" *J Infect*
46. (1977) "The Birmingham Research Unit of The Royal College of General Practitioners: Influenza. The Birmingham Research Unit of the Royal College of General Practitioners" *J R Coll Gen Pract*
47. Correa, Hinton, Mcgovern (2016) "Royal College of General Practitioners Research and Surveillance Centre (RCGP RSC) sentinel network: a cohort profile" *PubMed Abstract | Publisher Full Text | Free Full Text*
48. Mahmoud, Galal (2023) "Elbaset A: A review of various secure hash algorithms family" *Int Res J Mod Eng Technol Sci*
49. De Lusignan, Gallagher, Chan (2009) "The QICKD study protocol: a cluster randomised trial to compare quality improvement interventions to lower systolic BP in Chronic Kidney Disease (CKD) in primary care" *PubMed Abstract | Publisher Full Text | Free Full Text*
50. De Lusignan, Gallagher, Chan (2009) "The QICKD study protocol: a cluster randomised trial to compare quality mprovement interventions to lower systolic BP in Chronic Kidney Disease (CKD) in primary care" *PubMed Abstract | Publisher Full Text | Free Full Text*
51. Majeed, Molokhia (2023) "The future role of the GP Quality and outcomes framework in England"
52. Leston, Elson, Watson (2022) "Representativeness, vaccination uptake, and COVID-19 clinical outcomes 2020-2021 in the UK Oxford-royal college of general practitioners research and surveillance network: cohort profile summary" *PubMed Abstract | Publisher Full Text | Free Full Text*
53. Lackenby, Hungnes, Dudman (2008) "Emergence of resistance to oseltamivir among influenza A(H1N1) viruses in Europe" *Euro Surveill*
54. Calatayud, Lackenby, Reynolds (2009) "Oseltamivir-resistant pandemic (H1N1) 2009 virus infection in England and Scotland" *Emerg Infect Dis*
56. Cogdale, Kele, Myers (2023) "A case of swine influenza A(H1N2)v in England" *Euro Surveill*
57. Inzaule, Tessema, Kebede (2021) "Genomic-informed pathogen surveillance in Africa: opportunities and challenges" *Lancet Infect Dis*
59. Getchell, Wulandari, De Alwis (2024) "Pathogen genomic surveillance status among lower resource settings in Asia" *Nat Microbiol*
60. Hill, Githinji, Vogels (2023) "Toward a global virus genomic surveillance network" *Cell Host Microbe*
61. Alvarellos, Sheppard, Knarston (2023) "Democratizing clinical-genomic data: how federated platforms can promote benefits sharing in genomics" *PubMed Abstract | Publisher Full Text | Free Full Text*
62. Thorogood, Rehm, Goodhand (2021) "International federation of genomic medicine databases using GA4GH standards" *Cell Genom*
63. Casaletto, Bernier, Mcdougall (2023) "Federated analysis for privacypreserving data sharing: a technical and legal primer" *Annu Rev Genomics Hum Genet*
64. (2023) "PubMed Abstract | Publisher Full Text | Free Full Text 60. Makoni M: Launch of genomic surveillance system for respiratory viruses" *Lancet Microbe*
65. Goldstein, Harvey, Wilkie (2018) "Integrating patient and wholegenome sequencing data to provide insights into the epidemiology of seasonal influenza A(H3N2) viruses" *Microb Genom*
66. Allen, Reyne, Allingham (2024) "Genomic analysis and surveillance of Respiratory Syncytial Virus (RSV) using Wastewater-Based Epidemiology (WBE)" *J Infect Dis*
68. Robishaw, Alter, Solano (2021) "Genomic surveillance to combat COVID-19: challenges and opportunities" *PubMed Abstract | Publisher Full Text | Free Full Text*
69. Meredith, Hamilton, Warne (2020) "Rapid implementation of SARS-CoV-2 sequencing to investigate cases of health-care associated COVID-19: a prospective genomic surveillance study" *Lancet Infect Dis*
71. Mayor, Meza-Torres, Okusi (2022) "Developing a Long COVID phenotype for postacute COVID-19 in a national primary care sentinel cohort: observational retrospective database analysis" *PubMed Abstract | Publisher Full Text | Free Full Text*
72. Paget, Caini, Riccio (2022) "Has influenza B/Yamagata become extinct and what implications might this have for quadrivalent influenza vaccines?" *Euro Surveill*
73. De Lusignan, Leston, Ikpoh (2022) "Data saves lives: bottom-up, professionally-led endorsement would increase the chance of success" *Br J Gen Pract*
75. Riley (2023) "UKHSA data strategy. UK Health Security Agency, London, UK. 11 th" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12815052&blobtype=pdf | # Effectiveness of poliovirus environmental surveillance in Ghana: an indicator-based performance evaluation across seven regions, 2018-2022
Evangeline Obodai, Nana Afia, Asante Ntim, Ewurabena Duker, Emmanuel Gberbi, Comfort Antwi, Jude Mensah, Deborah Odame, Jessica Boakye, Sharon Bimpong, Gayheart Agbotse, Nancy Odoom, Patience Adams, Nana Acquah, Angelina Dickson, Christabel Odoom, Kwame Achempem, Isaac Baffoe-Nyarko, Godfred Egbi, Dennis Laryea, Franklin Asiedu-Bekoe, John Odoom
## Abstract
Objective To evaluate the performance of Ghana's environmental surveillance (ES) system for poliovirus (PV) detection from 2018 to 2022 using standardised indicators developed by the WHO and the US Centers for Disease Control and Prevention. Design A retrospective performance evaluation using 10 key indicators benchmarked against global targets for PV surveillance.Setting Seven regions across Ghana, participating in the national ES programme implemented under the Global Polio Eradication Initiative. Surveillance coverage Wastewater sampling was conducted at designated ES sites, supported by field collection teams and laboratory personnel responsible for sample acquisition, processing and reporting of PV detection results. Outcome measures Detection rates of PV and non-polio enteroviruses (NPEVs), timeliness of sample collection and reporting, data quality and system stability. Results A total of 738 wastewater samples were collected. The system demonstrated high sensitivity, detecting circulating vaccine-derived PV type 2 in 51 (6.9%) of samples, Sabin PV types 1 and 3 in 61 (9.5%) and 114 (17.8%), respectively, and NPEVs in 491 (66.5%) of samples. Over 80% of samples met the recommended 21-day collection-to-reporting time frame. Data quality exceeded the ≥80% threshold, and workflows remained stable throughout the evaluation period. Conclusions Ghana's ES system for PV was found to be flexible, stable and effective in generating high-quality data for early detection and public health response. These findings underscore the system's critical role in supporting polio eradication efforts and highlight its potential as a model for surveillance in similar settings.
## INTRODUCTION
Poliomyelitis is a highly infectious viral disease that primarily affects children under 5 years of age. Its clinical presentation ranges from mild, non-specific symptoms to severe acute flaccid paralysis (AFP), with permanent disability occurring in approximately 1 out of every 200-1000 infections. 1 2 Among those paralysed, 5%-10% may die due to respiratory muscle involvement. 3 Transmission occurs predominantly via the faecal-oral route, particularly in underimmunised populations and communities with poor sanitation and hygiene. 4 5 Although there is no cure for polio, safe and effective vaccines have significantly reduced its global burden. In 1988, the WHO launched the Global Polio Eradication Initiative (GPEI), prioritising mass immunisation and active surveillance of AFP in children under 15 years. This global effort successfully interrupted transmission of wild poliovirus (WPV) types 2 and 3, leaving WPV type
## STRENGTHS AND LIMITATIONS OF THIS STUDY
⇒ The study applied standardised WHO/Centers for Disease Control and Prevention indicators to assess environmental surveillance system performance. ⇒ Surveillance data were collected consistently over a 5-year period across multiple sites. ⇒ The evaluation used secondary programmatic data, minimising ethical and logistical constraints. ⇒ Geographic coverage was limited to 7 of Ghana's 16 regions, potentially affecting representativeness. ⇒ Site selection was based on predefined criteria, which may have excluded high-risk or underserved areas.
Open access 1 endemic only in Pakistan and Afghanistan. However, outbreaks of circulating vaccine-derived poliovirus (cVDPV) have emerged in 33 countries, including Ghana in 2019.
To strengthen surveillance and mitigate the risk of undetected transmission, the GPEI Strategic Plan now incorporates environmental surveillance (ES) as a complementary tool to AFP surveillance. ES involves systematic collection and testing of wastewater and sewage samples to detect poliovirus circulation, particularly in areas with low population immunity and poor sanitation. This approach enables early detection of poliovirus, even before clinical cases appear, facilitating timely public health interventions and vaccination campaigns. 6 7 Evidence from Israel illustrates the sensitivity of ES: 67 WPV1-positive sewage samples were detected between February and August 2013, despite no reported AFP cases since 2004 and 34 WPV1 in Senegal since 2007. 8 9 In Ghana, ES was piloted in 2016, formally instituted in 2018, and yielded its first detection of cVDPV2 in July 2019 from the Northern Region, prior to any clinical cases. 10 This triggered active AFP case investigations and a targeted immunisation campaign to halt transmission.
Timely analysis and dissemination of ES data to stakeholders, including the Ghana Health Service and Ministry of Health, are essential for effective response. Enhanced AFP surveillance, coupled with robust environmental monitoring, now forms a cornerstone of the global polio eradication strategy. 9 However, since its full implementation in 2018, Ghana's ES system has not undergone formal evaluation to assess its performance and effectiveness.
This study evaluates the performance of Ghana's ES system for poliovirus detection from 2018 to 2022, using standardised indicators developed by WHO and the US Centers for Disease Control and Prevention (CDC). The findings aim to determine the system's utility in complementing AFP surveillance and informing future eradication efforts.
## METHODS
## Study design and setting
This retrospective study evaluated the performance of Ghana's ES system for poliovirus detection over a 5-year period, from January 2018 to December 2022. Secondary data were extracted from two primary sources: the Disease Surveillance Department of the Ghana Health Service and the WHO-accredited Regional Reference Polio Laboratory (RRPL) at the Noguchi Memorial Institute for Medical Research (NMIMR), University of Ghana.
To support site selection and contextual analysis, sewer network maps and population estimates were obtained from the Environmental Health Department. In districts lacking formal sewer infrastructure, large open drainage systems serving multiple communities were identified as alternative sampling points.
Ghana comprises 16 regions and 261 districts. For this study, 14 districts across 7 regions were selected, each hosting one ES sampling site. These included 3 closed sewage systems and 11 open drainage systems. Site selection was guided by the following criteria: High risk of poliovirus transmission, poor performance on AFP surveillance indicators, presence of sewer networks or drainage systems serving large populations, absence of industrial waste, particularly chemical contaminants.
Site selection for ES was led by a Technical Assessment Team comprising representatives from the WHO, Ghana Health Service, Environmental Health Department and personnel from the Ghana National Polio and Regional Reference Laboratory. Final site validation was supported by WHO-AFRO environmental site coordinators. The selected sampling locations spanned seven regions: Greater Accra (Agbogbloshie, University of Ghana, Nima Freetown, Shiabu), Eastern (Akosombo, Koforidua), Ashanti (Asokore Mampong, Ahinsan), Volta (Togbe Afede, Aflao), Bono (Sunyani Zongo), Bono East (Techiman) and Northern (Nyanshegu, Koblimahgu) (figure 1).
Site characterisation included documentation of surrounding infrastructure such as schools, markets, hospitals and commercial centres. Geo-coordinates of each sampling point were recorded to ensure consistency and traceability.
## ES system overview
Ghana's ES system operates in close alignment with the AFP surveillance framework. Both systems share protocols for sample transportation to the RRPL, result reporting and emergency response coordination. Information exchange is jointly managed by ES and AFP officers in collaboration with the Disease Surveillance Department of the Ghana Health Service, with technical support from WHO.
Following WHO guidelines for ES of poliovirus circulation, 11 trained personnel collect sewage samples monthly using the grab method. Each sampling team includes a field collector from the Environmental Health Department and a supervisor from the Disease Surveillance Division, who ensures adherence to biosafety protocols and use of personal protective equipment.
Sampling is conducted between 6:00 and 7:30 am. One Litre of sewage is collected per site and transported to the RRPL via reverse cold chain (2°C-8°C). Each sample is accompanied by a submission form detailing demographic data, collection date and time, and atmospheric temperature. Data are digitally captured using the Open Data Kit (ODK) and transmitted to the WHO server.
On arrival at the laboratory, samples are assessed for integrity (temperature, volume, leakage). Laboratory processing follows WHO protocols to isolate and identify polioviruses, differentiate serotypes (PV1-3), WPVs, Sabin-like strains and VDPVs using intratypic differentiation (ITD), sequence WPV and VDPV isolates by analysing the VP1 region of the viral genome. 12 Open access Laboratory findings are disseminated to the Disease Surveillance Department, regional and district health facilities, and WHO for coordinated public health action.
## Data collection
Secondary data were extracted from completed sewage sample collection forms spanning January 2018 to December 2022. Additional information was obtained through direct observation of sample collection procedures, review of surveillance reports and WHO guidelines on ES.
The evaluation followed the US CDC framework for assessing public health surveillance systems. 13 Ten key attributes were assessed:
1. Data Quality This is defined as ES samples with complete data including recorded atmospheric temperature, collection date and time, and names of both collector and supervisor.
## 2. Simplicity
Measured by the clarity of operational workflows, sample transport logistics, personnel roles, laboratory procedures and data dissemination pathways.
## 3. Flexibility
Evaluated by the system's ability to adapt to changing operational or informational needs.
## 4. Representativeness
Evaluated by geographic distribution of ES sites and types of drainage systems used.
## 5. Stability
Assessed through staffing levels, training frequency, inventory consistency, funding availability and contingency planning.
## 6. Acceptability
Measured by stakeholder participation, punctuality of sample submission and sample condition on arrival.
## 7. Sensitivity
Defined as the proportion of samples testing positive for enteroviruses (EVs) in the wastewater sample.
## 8. Poliovirus positivity rate
Calculated as the proportion of samples testing positive for polioviruses in the wastewater samples 9. Timeliness Assessed by calculating: Percentage of samples delivered to the laboratory within 3 days of collection, percentage concentrated within 2 days of receipt, percentage analysed within 21 days of collection.
## 10. Site review and stakeholder engagement
Annual site visits were conducted to assess viability, engage personnel and determine the need for site continuation or relocation. Review meetings were held with district and regional supervisors to discuss operational challenges and share best practices.
## Data analysis
Data entry was performed using Epi Info V.3.5.4, and preliminary analyses were conducted in Microsoft Excel 2003. Descriptive statistics including frequencies, percentages and proportions were generated using Microsoft Excel 2019. Figures and tables were created to support data visualisation and reporting.
## Patient and public involvement
Patients and the public were not directly involved in the design, conduct, reporting or dissemination of this ES study. However, findings were shared with relevant public health stakeholders and community representatives through national dissemination forums and policy briefs to inform risk communication and immunisation strategies.
## Open access
## RESULTS
## Site coverage and sample collection (data quality)
Data from all validated ES sites were included in the analysis. At the initial rollout of ES in 2018, eight sites were operational across three regions (table 1). The number of sites expanded to 10 in 2019 and reached 14 by 2020, reflecting increased geographic coverage. Among these, the Shiabu site served a population exceeding 200 000 due to a well-structured sewer system encompassing Dansoman and surrounding areas. Other high-density sampling locations included Zongo communities and the University of Ghana (table 1).
Throughout the evaluation period, sample collection was conducted every 4 weeks, resulting in approximately 13 samples per site annually. As the number of sites increased, the total number of samples collected rose from 31 in 2018 to 196 in 2020, and 195 in 2022 (table 2).
Two cell lines RD (Human rhabdomyosarcoma) and L20B (Mouse fibroblast cells expressing with poliovirus receptor) were used in the polio laboratory for virus isolation. During which three tubes of RD and five tubes of L20B were used. Although both cell lines support the growth of poliovirus, the RD cell line almost exclusively isolates non-polio EVs (NPEVs). All the ES samples were concentrated using the two-phase separation method.
ES indicators were evaluated using CDC and WHO benchmarks for field and laboratory performance. Most indicators met the recommended targets, with only a few falling short of the criteria.
## Simplicity
The ES system demonstrated operational simplicity from sample collection through to result reporting, as illustrated in figure 2. Throughout the evaluation period, the sample collection team adhered to biosafety protocols while collecting specimens from all 14 sites. Samples were transported via courier services, primarily Ghana Post and FedEx, to the RRPL at NMIMR for routine analysis.
Laboratory results were reported weekly to the Disease Surveillance Department of the Ghana Health Service and the WHO. During the cVDPV2 outbreaks in 2019-2020 and 2022, the laboratory promptly notified the Disease Surveillance Department of the Ministry of Health, triggering outbreak investigations.
Processes were consistently followed by all stakeholders. Sample collectors arrived at sites on schedule and were supervised by designated field supervisors. With the exception of four sites in Greater Accra, Eastern and Volta regions that missed one or two scheduled collections in 2022 (table 2), the remaining ten sites maintained full compliance with the sampling schedule.
## Flexibility
The ES system demonstrated strong adaptability to evolving demands. One key example was the laboratory's ability to maintain its performance target for processing and reporting results for at least 80% of received samples within the stipulated timeframe, even amid changes to the testing algorithm. In 2022, the turnaround time for virus isolation was reduced from 21 days to 14 days to accelerate the detection and reporting of circulating polioviruses and enable faster outbreak response (table 3). Despite this change, the 80% benchmark was consistently met.
The system also supported multipathogen surveillance without disruption. Environmental samples originally collected for poliovirus and non-EV detection were successfully repurposed for broader studies, Open access including the identification of SARS-CoV-2 in sewage between 2021 and 2022, 14 demonstrating the system's versatility and capacity for integrated surveillance.
## Representativeness
ES was implemented in seven out of Ghana's sixteen regions, with each participating region hosting at least two collection sites, except for Bono and Bono East. These two regions were formerly a single administrative unit with two sites before their separation. In regions with multiple sites, the sites were strategically located in different districts to enhance geographic coverage.
The remaining nine regions were not included in the ES system due to the absence of suitable sewer infrastructure and drainage systems, which are essential for meeting site selection criteria. Since its inception in 2018 with only three regions, the system has demonstrated the capacity to expand and remains flexible to incorporate additional regions as they become eligible.
## Stability
The ES system operated with consistent reliability and minimal disruptions throughout the evaluation period. Sample collection and laboratory processing adhered to the annual schedule and established standard operating procedures. The surveillance programme ensured the timely provision of funds and resources, including monthly allowances for field teams to support sample collection and shipment to the laboratory.
## Open access
Personnel capacity was maintained through periodic training and refresher sessions, contributing to sustained operational performance. Although the COVID-19 pandemic in 2020 and 2021 led to occasional delays in the supply of laboratory reagents and consumables, these challenges had only a minor impact and did not significantly disrupt overall system functionality.
## Data quality
The overall quality of data collected during the study period met the target performance indicators. Across the years, key metadata including date of sample collection, atmospheric temperature at the time of collection, and date of sample submission to the reference laboratory were consistently recorded on sample forms.
However, some deviations were noted. In 2021 and 2022, a subset of samples arrived at the laboratory with temperatures outside the recommended range of 2°C-8°C, indicating lapses in cold chain maintenance. Additionally, a few sites failed to collect scheduled samples in 2018 and 2019, resulting in minor gaps in data completeness during the early phase of system implementation.
## Acceptability
Throughout the study period, the ES system demonstrated strong stakeholder engagement and operational buy-in. All key actors, including sample collection teams, courier services, laboratory personnel, the Disease Surveillance Division and WHO, actively participated, contributing to the system's broad acceptability.
In 2020, Ashanti, Eastern and Volta Regions adhered fully to the scheduled 13 sample collections per site/year. Bono and Bono East Regions, which joined the system later, each collected 10 samples per site. Greater Accra and Northern Regions exceeded their scheduled collections due to the cVDPV2 outbreak response (see table 2). The surveillance teams also ensured timely and proper transport of samples, with over 80% arriving at the laboratory in good condition except in 2020 and 2022.
## Sensitivity
To assess the sensitivity of the surveillance system, annual EV detection rates were calculated using the grab sampling method. According to WHO guidelines, 15 an EV detection rate of ≥50% is expected in at least 80% of samples processed annually per demographic catchment. Figure 3 illustrates EV detection rates across the 5-year period. The highest rate was recorded in 2018 (83.9%), Open access followed by a gradual decline, reaching 53.0% in 2021. Despite this decline, the system consistently met the WHO sensitivity benchmark. Importantly, the system demonstrated its capacity to detect cVDPV2 ahead of clinical case identification. In 2019, Ghana's first-ever cVDPV2 was detected through ES, with a total of 17 isolates, 15 from Greater Accra alone. In 2020, 22 cVDPV2 isolates were reported, with Greater Accra (14), Ashanti (4) and Bono East (4) contributing. No cVDPV2 was detected in 2021; however, there was another outbreak in 2022, predominantly in the Northern Region (see figure 4). Hence, over the period, Open access of the 738 ES samples collected, 51 (6.9%) were cVDPV2, 61 (9.5%) were Sabin 1, 114 (17.8) were Sabin 3 and 491 (66.6%) were NPEVs. All the cVDPV2 might have been shared by children who have not been vaccinated against the type 2 poliovirus as a result of the switch from tOPV to bOPV in April 2016. Genomic sequencing revealed that the cVDPV2 detected from 2019 to 2020 belonged to an emergence group NIE-JIS-1, while the 2022 cVDPV2 belonged to the NIE-ZAS-1 emergence group, as shown in figure 5. All these were importations from neighbouring countries.
## Poliovirus positivity rate
The ES system successfully isolated poliovirus from sewage samples across all sampling sites during the study period.
In 2019, the system achieved its highest poliovirus positivity rate (PPR) of 30.8%, reflecting strong correlation between sample positivity and actual virus circulation. The lowest PPR rate was observed in 2021, at 5.5%.
## Timeliness
ES protocols stipulate that sewage samples should arrive at the WHO-accredited polio laboratory within 3 days of collection. Throughout the evaluation period, this target was consistently met, with all samples reaching the laboratory within the required time frame.
Additionally, WHO guidelines recommend that at least 80% of samples be concentrated within 48 hours of receipt. Performance in this area was generally strong, though some fluctuations were observed: In 2018, 100% of samples were concentrated within the stipulated time. This declined slightly to 98.7% in 2019, and further to 90.6% in 2020 and 90.8% by 2022.
A similar trend was observed in the timeliness of virus isolation, which is expected to be completed within 21 days of sample receipt. The system performed best in 2018, 2019, and 2022, with timely analysis rates exceeding expectations.
While the target of ≥80% for sample arrival within 3 days was met in all years, the laboratory consistently concentrated over 80% of samples within 48 hours except in 2021, when only 51.2% met this benchmark. Despite this dip, the overall condition of samples on arrival consistently exceeded WHO's quality threshold of 80%.
Moreover, turnaround times for virus isolation and ITD surpassed the 80% target across the study period. As shown in table 2, the annualised EV detection rate of ≥50% was achieved each year. A summary of ES performance indicators from 2018 to 2022 is in table 3, which further presents the key ES performance indicators observed during the study period, alongside their corresponding target thresholds as established by WHO.
## Site review and review meetings
The first comprehensive site review was conducted in early 2019, led by a WHO Technical Officer. During this mission, all nine existing ES sites were visited and evaluated. The Yendi site was found to be consistently dry, with wastewater flow observed only during the rainy season. Following further investigation, the review team recommended its closure due to unsuitability for year-round sampling.
As part of the same mission, two new sites were identified and commissioned in the Northern Region. Sample collectors and supervisors were selected and trained to initiate sampling activities. Notably, later that year, the first detection of circulating vaccine-derived poliovirus type 2 (cVDPV2) occurred at these newly established sites.
Subsequent to this initial review, annual site visits and review meetings have been institutionalised. These engagements serve to assess site performance, provide refresher training and facilitate knowledge exchange among field teams. During the 2021 review, 10 out of 14 sites were digitised using the Blue Line mapping tool, and sample collectors were trained to use ODK (https://docs. getodk.org/collect-using/) for electronic data capture. Sample collection equipment was distributed, and on-site training was provided where needed.
Review meetings consistently included presentations on site performance, biosafety protocols and best practices. Challenges and observations from field visits were openly discussed. Based on performance metrics, the Aflao site in the Volta Region was recommended for closure due to persistently low EV detection rates. A replacement site Open access was identified at Kpando, also in Volta Region; however, this site was subsequently closed following the 2022 review and team recommendation.
## DISCUSSION
The ES system for poliovirus in Ghana demonstrated operational simplicity, high data quality, timeliness, sensitivity, flexibility and stability, consistently meeting WHO performance targets between 2018 and 2022.
According to WHO guidelines, at least 80% of sewage samples should arrive at the polio laboratory within 3 days of collection. 11 Ghana exceeded this benchmark throughout the study period, reflecting effective coordination between field teams, regional offices and the laboratory. Although a slight decline in performance was observed in 2020 and 2022, the target was still met. This dip may be attributed to the COVID-19 pandemic, during which the Polio Oversight Board (POB) of the GPEI directed laboratories to prioritise COVID-19 response efforts. 16 These findings align with the report by Tuma. 17 Regarding turnaround time for laboratory results (≤21 days), the polio laboratory consistently exceeded Open access expectations, underscoring its efficiency in detecting and reporting potential outbreaks for timely public health action.
The system also proved highly sensitive in detecting both polioviruses and NPEVs. Notably, circulating vaccine-derived poliovirus type 2 (cVDPV2) was first detected in environmental samples in July 2019, 10 prior to any clinical case. This early detection triggered active case searches and led to the identification of Ghana's first clinical cVDPV2 case in Chereponi, North East Region. The outbreak continued to 2020 and in all, 19 and 12 cVDPV2 were detected in 2019 and 2020 AFP cases, respectively. Again, during the 2022 cVDPV2 outbreak, the virus was first detected in the environment before 3 AFP cases were confirmed. Similar high positivity rates have been reported in Afghanistan and Pakistan. 18 Within the same period of ES evaluation (2018-2022), a total of 4232 AFP cases were received in the polio laboratory and 414 polioviruses were detected. Of this, 34 (8.2%) were cVDPV2. Samples collected from the cVDPV2 contacts also detected 34 cVDPV2 19 WHO recommends that ≥50% of sewage samples yield EVs annually. 11 Ghana consistently surpassed this threshold, with annual EV isolation rates exceeding 50%. A comparable study in Iran reported an EV isolation rate of 53.49%. 20 However, sites such as Yendi, Aflao and Kpando recorded persistently low EV rates (<50%) and were closed following the review team's recommendations. The Yendi site was opened in early 2018 and was closed after 3 months due to dryness; hence, it did not contribute to this data. The Aflao site contributed from 2018 to 2021 and was replaced by the Kpando site, but it was also closed in January 2023. This pattern mirrors observations across other countries in the WHO-AFRO Polio network, where some ES sites have underperformed in EV detection. 21 Another WHO indicator assesses the proportion of months in which at least one sample is collected per site, with a target of ≥80% (ie, 10 out of 12 months). 10 Most Ghanaian ES sites met or exceeded this benchmark, with a few showing improvement over time. This frequency of sampling supports the reliability of surveillance data in reflecting poliovirus transmission dynamics.
Data quality was generally high, though inconsistencies were noted in some years. For example, blank fields in Epi Info records created uncertainty regarding whether omissions were due to non-compliance or data entry errors. These gaps may reflect lapses in surveillance officer commitment, a challenge also reported in other sub-Saharan African countries. 222 This highlights the need for enhanced training on accurate and complete data entry.
The system demonstrated flexibility by adapting environmental sampling for SARS-CoV-2 detection. Results indicated that sewage samples could serve as a noninvasive early warning tool for COVID-19 outbreaks, complementing clinical surveillance. 14 This aligns with findings from Peccia et al, 23 who reported similar utility in wastewater-based epidemiology.
Some limitations observed in the study include (a) since the study was indicator-based evaluation, it relied on selected metrics (eg, sample positivity rates, timeliness); hence, an important factor (like community compliance) was not fully captured by chosen indicators, (b) without a control or comparison group in the study design, it was challenging to attribute changes in effectiveness directly to improvements in the surveillance system, (c) surveillance sites were clustered in urban regions (eg, Accra, Kumasi) with fewer sites in rural or hard-to-reach areas, whereas some regions had no sites limiting representativeness, (d) low EV productivity resulted in closure of some sites (eg, Aflao and Kpando) and (e) reactive surveillance intensification in response to the 2019-2020 and 2022 cVDPV2 outbreaks might have skewed performance metrics.
Overall, the Ghana Polio ES system was simple and stable. Its structure from sample collection to result reporting facilitated efficient workflows. Timeliness targets were consistently exceeded, and funding was available to support sample collection, transport and staff training. Personnel were regularly trained on updated testing algorithms and supported outbreak response efforts.
The system's timeliness, flexibility and sensitivity are commendable. Its adaptability to detect other human pathogens and EVs enhances its value for broader public health surveillance. The sustained expertise and commitment of staff likely contributed to the system's strong performance.
## CONCLUSIONS
Over the past 5 years, Ghana's poliovirus ES system has largely met key performance indicators established by WHO and CDC, including simplicity, data quality, sensitivity, timeliness, stability and flexibility. These achievements reflect the system's critical role in supporting early detection and response to poliovirus transmission.
However, sustained efforts are needed to continuously improve data quality, enhance operational efficiency and ensure timely interventions. Strengthening these areas will not only support national public health decisionmaking but also contribute meaningfully to global initiatives aimed at achieving and maintaining a polio-free world.
## References
1. Bennett, Plum, Bower (2010) "Principles of neurology"
2. Stadtländer (2005) "Poliomyelitis: pathogenesis and epidemiology"
3. Honeyman, Reilly, Tebbens (2022) "Estimating the case fatality ratio of paralytic poliomyelitis" *BMC Infect Dis*
4. Marx, Glass, Sutter (2000) "Differential diagnosis of acute flaccid paralysis and its role in poliomyelitis surveillance" *Epidemiol Rev*
5. (2004) "World Health Organization. Poliomyelitis fact sheet"
6. Asghar, Diop, Weldegebriel (2014) "Environmental surveillance for polioviruses in the Global Polio Eradication Initiative" *J Infect Dis*
7. Hovi, Shulman, Van Der Avoort (2012) "Role of environmental poliovirus surveillance in global polio eradication and beyond" *Epidemiol Infect*
8. Shulman, Gavrilin, Jorba (2013) "Molecular epidemiology of silent introduction and sustained transmission of wild poliovirus type 1, Israel" *Euro Surveill*
9. Ndiaye, Mbathio, Diop (2014) "Environmental surveillance of poliovirus and non-polio enterovirus in urban sewage in Dakar, Senegal (2007-2013)" *Pan Afr Med J*
10. Odoom, Obodai, Adjei (2019) "Detection of circulating vaccinederived poliovirus type 2 in environmental samples in Ghana" *MMWR Morb Mortal Wkly Rep*
11. (2015) "Global polio eradication initiative: guidelines on environmental surveillance for detection of polioviruses"
12. (2004) "World Health Organization (WHO). Polio laboratory manual. 4th edn"
13. (2001) "Updated guidelines for evaluating public health surveillance systems: recommendations from the Guidelines Working Group" *MMWR Recomm Rep*
14. Duker, Obodai, Addo (2024) "First Molecular Detection of SARS-CoV-2 in Sewage and Wastewater in Ghana" *Biomed Res Int*
15. (2022) "World Health Organization. Field guidance for the implementation of environmental surveillance for poliovirus"
16. Burkholder (2023) "Maintaining polio surveillance during the COVID-19 pandemic: lessons from the GPEI" *Vaccine*
17. Tuma (2021) "Impact of COVID-19 on polio surveillance systems in West Africa" *Afr Health Sci*
18. Kroiss, Ahmed, Ahmed (2018) "Environmental surveillance for polioviruses in Pakistan and Afghanistan" *J Infect Dis*
19. Obodai, Boakye, Ntim (2024) "Evaluation of the Intensive Acute Flaccid Paralysis Surveillance System in Ghana: Post the Switch from tOPV to bOPV" *Trop Med Infect Dis*
20. Khodaei (2008) "Environmental surveillance of enteroviruses in Iran: a five-year study" *Iran J Virol*
21. "Environmental surveillance performance report: 2018-2022"
22. Nsubuga (2010) "Strengthening public health surveillance and response systems in Sub-Saharan Africa" *MMWR*
23. Peccia, Zulli, Brackney (2020) "Measurement of SARS-CoV-2 RNA in wastewater tracks community infection dynamics" *Nat Biotechnol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846377&blobtype=pdf | # Genetic Diversity of SARS-CoV-2 in Kazakhstan from 2020 to 2022
Altynay Gabiden, Andrey Komissarov, Aknur Mutaliyeva, Aidar Usserbayev, Kobey Karamendin, Alexander Perederiy, Artem Fadeev, Ainagul Kuatbaeva, Dariya Jussupova, Askar Abdaliyev, Manar Smagul, Yelizaveta Khan, Marat Kumar, Temirlan Sabyrzhan, Aigerim Abdimadiyeva, Aidyn Kydyrmanov
## Abstract
Coronavirus disease 2019 , caused by SARS-CoV-2, has had major social and economic consequences worldwide. Whole genome sequencing (WGS) is essential for genomic monitoring, enabling tracking of viral evolution, detection of emerging variants, and identification of introductions and transmission chains to inform timely public health responses. Here, we compile and harmonize SARS-CoV-2 genomic data generated by multiple laboratories across Kazakhstan together with publicly available sequences to provide a national overview of genomic dynamics across successive epidemic waves from 2020 to 2022. We analyzed 4462 genomes deposited in GISAID (including 340 generated in this study), of which 3299 passed Nextclade quality filters, and summarized lineage turnover across major phases (pre-VOC, Alpha, Delta, Omicron BA.1/BA.2, Omicron BA.4/BA.5, and a later recombinant-dominant period). Sequencing intensity varied markedly over time (0.60‰ of confirmed cases during Delta vs. 11.57‰ during the Omicron BA.5 wave), suggesting that lineage diversity and persistence may be underestimated. Pre-VOC circulation included ≥12 Pango lineages with evidence of multiple introductions and sustained local transmission, including a Kazakhstan-restricted B.4.1 lineage that emerged in Nur-Sultan/Astana and disappeared after April 2020. The Tengizchevroil oilfield outbreak comprised B.1.1 viruses with phylogenetic support for ≥three independent introductions. Alpha and Omicron waves were characterized by repeated introductions and heterogeneous origins, whereas Delta was dominated by AY.122 with an additional distinct AY.122 cluster; a notable BF.7 local transmission event was observed during BA.5. We also highlight locally enriched non-lineage-defining mutations. Overall, recurrent importations and variable local amplification shaped SARS-CoV-2 dynamics in Kazakhstan, while interpretation is constrained by strongly time-skewed sequencing.
## 1. Introduction
The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) was first identified in Wuhan, China, at the end of 2019 and rapidly spread, evolving into a global pandemic [1,2]. SARS-CoV-2 is an enveloped, non-segmented, positive single-stranded RNA virus with a genome size of about 30,000 bases, featuring four structural proteins: spike (S), envelope (E), membrane (M), and nucleocapsid (N) [3,4]. Overall, SARS-CoV-2 shares approximately 79.5% and 96% genomic sequence identity with previously identified SARS-CoV and bat coronavirus, SL-CoV-RaTG13, respectively [5,6].
SARS-CoV-2 exhibits a propensity for multiple recombinations and mutations in its genome, potentially leading to changes in viral protein structure, binding affinity, virus transmission, diagnostics, vaccine efficacy, and sensitivity to antiviral drugs [7][8][9]. Since the onset of the pandemic, the SARS-CoV-2 virus has evolved into new variants of interest (VOIs) or variants of concern (VOCs). During the circulation of early virus variants such as Alpha, Beta, Delta, etc., a high mortality rate was observed among elderly individuals and those with comorbidities such as hypertension, kidney disorders, cancer, diabetes, and obesity [10]. Later variants of the virus, including the Omicron group of strains, caused less severe illness compared to previous variants but led to an increase in cases among young people and children [11].
The first confirmed case of COVID-19 in Kazakhstan was reported on 13 March 2020. This individual had traveled from Germany. The initial virus strain in Kazakhstan was the same as the original strain identified in Wuhan, China [12]. From May 2020, Kazakhstan saw an increase in cases. Early studies of the virus indicated the circulation of the original SARS-CoV-2 strain, which caused significant outbreaks around the world [13].
From the beginning of 2021, more transmissible SARS-CoV-2 variants began to spread, including the Alpha variant (B.1.1.7), first identified in the UK, and the Beta variant (B.1.351), first identified in South Africa. From July 2021 onward, Kazakhstan began to register cases caused by the Delta variant (B.1.617.2), which is characterized by higher transmissibility and partial immune evasion. Delta subsequently became the dominant variant in the country, triggering a new wave of the pandemic [14,15].
The Omicron strain was officially confirmed in Kazakhstan in January 2022. It spread quickly across the country, helped by its high R0 (reproduction rate, meaning how quickly the virus spreads). This SARS-CoV-2 variant was more contagious than previous ones, but at the same time, it caused less severe disease in most people, which led to a significant number of infections but relatively fewer severe cases [16].
In February 2022, Kazakhstan experienced a peak in cases caused by the Omicron strain. Against this background, the number of patients with COVID-19 in medical institutions increased significantly, but most cases were mild or moderate. Thus, Omicron subvariants were dominant in winter-spring 2023 [17].
From March 2022, subvariants BA.1 and BA.2 began to actively circulate. They were then joined by the BA.4 and BA.5 subvariants. These subvariants had improved immune evasion properties, which contributed to the increase in cases despite high vaccination rates and previous infections.
Thus, the purpose of this study was to analyze the dynamics of the spread of SARS-CoV-2 virus variants in Kazakhstan in the period from 2020 to 2023, as well as to identify key mutations characteristic of variants circulating in Kazakhstan.
https://doi.org/10.3390/v18010138
Viruses 2026, 18, 138
## 2. Materials and Methods
## 2.1. Sample Collection
The Reference Laboratory for Viral Infection Control (Almaty, Kazakhstan), which is the National Influenza Center in Kazakhstan, received 10% of SARS-CoV-2-positive samples from 18 regional virology laboratories in the country for sequencing every month.
The samples used in this study were nasopharyngeal and oropharyngeal swabs selected according to the following epidemiological and clinical criteria: (1) samples from individuals with symptoms of acute respiratory viral infections and COVID-19; (2) contact with confirmed cases of COVID-19; (3) samples from individuals arriving from abroad; (4) samples from home foci of infection; (5) samples from individuals with moderate/severe disease and/or hospitalized in intensive care units, etc.
From 2022, monitoring of coronavirus infection was introduced into the current epidemiological surveillance system for acute respiratory viral infections and influenza. All samples collected within the framework of the surveillance were tested in parallel for influenza and SARS-CoV-2; during routine studies for SARS-CoV-2, some negative samples were tested for influenza.
## 2.2. Screening PCR
In regional laboratories, PCR with real-time detection for identification of SARS-CoV-2 virus RNA was performed using the Amplitest SARS-CoV-2 reagent kit (CRIE, Moscow, Russia), Intifika SARS-CoV-2 (Alkor Bio Company Ltd., Saint-Petersburg, Russia), BGI (BGI Genomics, Shenzhen, China), etc. The CDC's Influenza SARS-CoV-2 Multiplex Assay reagent kit was used in the reference laboratory.
## 2.3. Virus Genome Sequencing
Viral RNA extraction was performed using the commercial PureLink RNA MiniKit (Life Technologies, Carlsbad, CA, USA) and QIAamp Viral RNA Mini Kit (Qiagen GmbH, Hilden, Germany) according to the manufacturers' recommendations. Reverse transcription and SARS-CoV-2 library preparation were performed using the Ampliseq™ cDNA Synthesis and AmpliSeq Library PLUS for Illumina kits (Illumina, San Diego, CA, USA) according to the manufacturer's recommendations. Amplification products were purified using the AMPure XP reagent (Beckman Coulter, Inc., Brea, CA, USA) according to the manufacturer's recommendations. DNA fragment library quality was determined using the Agilent High Sensitivity DNA Kit on a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer's recommendations. Quantitative analysis of cDNA libraries was performed using the NEBNext Library Quant Kit for Illumina according to the manufacturer's recommendations.
For whole genome sequencing, the MiniSeq High output Reagent Cartridge was used in PE paired-end reading mode according to the manufacturer's recommendations. Whole genome sequencing of SARS-CoV-2 viruses was performed using the Illumina MiniSeq next-generation sequencer (NGS).
## 2.4. Bioinformatics and Phylogenetics
A sequencing quality check was performed using fastqc 0.12.1 with subsequent quality trimming using trimmomatic 0.39. Consensus genome assembly was performed using BWA v. 0.7.17, Samtools v. 1.19.2, and Ivar v.1.4.2, along with custom Python 3.10.4 scripts. All sequenced genomes were timely submitted to EpiCoV GISAID. In total, 4462 SARS-CoV-2 genomic sequences available in GISAID were downloaded and quality-filtered using Nextclade [18]. Qc.overallStatus equal to "good" or "mediocre" was used as a quality filtration criterion. A total of 3299 SARS-CoV-2 genomes from Kazakhstan passed quality filtration. First we stratified SARS-CoV-2 genomes into pandemic waves using collection dates and VOC designations: 122 genomes in the pre-VOC period, 167 in the Alpha period, 717 in the Delta period, 823 in the Omicron BA.1/BA.2 period, and 1193 in the Omicron BA.4/BA.5 period. Uvaia v2.0.1 software was used to search for the closest neighboring sequences available in GISAID (mmsa-2024-02-04.tar.xz global alignment was used as an input). Briefly, each set of SARS-CoV-2 genomes from corresponding waves was aligned to the Wuhan reference genome (EPI_ISL_402124) using uvaialign (e.g., for the Alpha wave the following command was used: uvaialign -r EPI_ISL_402124.fasta VOC_alpha.fasta). Then the uvaia search was performed as follows: "uvaia -r mmsa-2024-02-04.tar.xz/2024-02-04_masked.fa -trim=230 KZ_alpha/uvaia.10dc4_alphaKZ.aln.xz -o alpha_result --nthreads=120 -x". Rank 1 and rank 2 neighbors were used for further analysis. Multiple alignment was performed using MAFFT software (Version 7.526) [19]. Phylogenetic tree reconstruction was conducted in RAxML-NG ("raxml-ng --msa MSA_dataset.fasta --prefix MSA_dataset --model GTR") [20]. Treemmer tool v0.3 [21] was used to remove redundant sequences while preserving phylogenetic diversity. Final visualization of phylogenetic trees was performed using custom R scripts (v.4.4.1 ggtree package). The overall analysis workflow is presented in Supplementary Figure S1. The final list of viruses after quality filtration is provided in Supplementary Table S1.
## 2.5. Ethical Principles
In Kazakhstan, obtaining informed consent from the patient before collecting samples A total of 4462 SARS-CoV-2 genomes were obtained from GISAID, of which 340 were generated in this study. Following quality assessment with Nextclade, 3299 genomes met high-or medium-quality thresholds and were retained for downstream analyses.
## 3. Results
## Kazakhstan
Genomic sequencing was unevenly distributed over time, with scarce data early in the pandemic and a marked increase in the second and third years (Table 1). The proportion of confirmed cases sequenced ranged from 0.60‰ (per mille) during the Delta wave-when case counts were high and sequencing capacity limited-to 11.57‰ during the Omicron BA.5 wave. After the WHO ended the COVID-19 Public Health Emergency of International Concern on 5 May 2023, case registration practices changed substantially; therefore, we did not calculate sequencing proportions for the subsequent period. Two sequences from Baikonur (Kazakhstan) cluster phylogenetically with a South Korean genome (hCoV-19/South_Korea/KDCA3546/2020) that GISAID metadata annotate as an export from Kazakhstan to South Korea. The Baikonur sequences appear closely related to two genomes from the Northern Mariana Islands.
Tengizchevroil (TCO), in the Atyrau Region, became Kazakhstan's largest workplace COVID-19 hotspot in 2020. Between March and May 2020, 1306 laboratory-confirmed infections among workers were recorded [22]. In May 2020, as the Ministry of Health considered suspending operations, TCO demobilized ~20,000 of its ~30,000 staff and instituted pre-rotation quarantine with PCR testing for the remaining ~13,000 employees. Despite these measures, cumulative infections rose to 2661 by 29 July 2020 [22]. Genomic data from the outbreak indicate that all sequenced cases belonged to Pango lineage B.1.1. Phylogenetic reconstruction supports at least three independent introductions of SARS-CoV-2 into the Tengiz oilfields (Figure 4).
## 3.2. Genetic Diversity of SARS-CoV-2 in Kazakhstan in Alpha Period (February-June 2021)
Sequenced Alpha VOC viruses had wide geographical distribution (Figure 5). The first Alpha variant viruses were reported in Kazakhstan at the end of March 2021. Retrospective sequencing revealed the oldest specimen positive for Alpha to be from the beginning of February 2021 [23]. Most of the sequenced specimens were collected in March and April 2021 (Figure 6).
The B.1.1.7 lineage harbors eighteen amino acid changes compared with the Wuhan-Hu-1 reference-four of which are shared with its parental B.1.1 lineage-along with three in-frame deletions (ORF1a:del3676/3678, S:del69/70, and S:del144). In addition to the lineage-defining set, nine nonsynonymous mutations occurred in at least 10% of Kazakhstan viruses (Table 2), with several showing higher prevalence in Kazakhstan than globally. Several mutations in the Kazakhstan Alpha dataset show markedly higher frequencies than in global datasets, suggesting potential regional enrichment or distinct local transmission dynamics. The most striking enrichment is observed for ORF1b I28T (NSP12:I37T), detected in 69/167 (41.32%) genomes from Kazakhstan yet virtually absent in global Alpha datasets (0.01%). Additional recurrent changes include ORF1a F200L (NSP13:F200L) in 28/167 (16.77%) and ORF7a P84L (NS7a:P84L) in 30/167 (17.96%), both of which were rare worldwide (0.00% and 0.28%, respectively).
Several substitutions enriched in Kazakhstan reached approximately 10-11% locally: M F100I (M:F100I) and ORF1a M1312I (NSP3:M494I) (each 19/167; 11.38%), both essentially absent globally (0.00% and 0.08%), and ORF3a Y189S (NS3:Y189S), ORF3a A99S (NS3:A99S), and ORF1a V3595D (NSP6:V26D) (each 18/167; 10.78%), all rare worldwide (0.01%, 0.06%, and 0.00%, respectively). In contrast, ORF8 Y73C (NS8:Y73C) was highly frequent both in Kazakhstan (115/167; 68.86%) and globally (94.93%), representing a characteristic Alpha mutation rather than a region-specific variant. Similarly, ORF8 K68stop (NS8:K68stop) appeared at a lower frequency in Kazakhstan (23/167; 13.77%) compared with global datasets (33.23%).
The placement of Kazakhstan B.1.1.7 sequences across the phylogeny (Figure 7) indicates multiple introductions, as they do not form a monophyletic group. Notably, many appear as singletons or in small clades, suggesting that most introductions did not result in sustained community transmission The first Delta strain cases were identified in Kazakhstan in July 2021 [24], yet reliable genomic data start from August 2021. Sequenced Delta VOC viruses had wide geographical distribution (Figure 8). Most of the sequenced specimens were collected in August, September, and November 2021 (Figure 9). The dominating lineage was AY.122 with combination of NS7a_P45L and NSP2_K81N substitutions typical for the previously described Russian sublineage of AY.122 [19]. Interestingly, a monophyletic cluster of AY.122 viruses without NS7a_P45L+ NSP2_K81N combination typical for Russian AY.122 viruses grouping with a virus of Indian origin was observed (Figure 10). Due to the low sequencing level in Russia and Central Asia at that time, it is hard to locate the plausible place of emergence of these AY.122 sublineages and the directions of their importation/exportation; nevertheless, we can speculate that AY.122 in Kazakhstan was not as homogeneous as AY.122 in Russia. The sporadic detection of AY.120, AY.121, AY.123, AY.126, and AY.127 lineages was observed. The most characteristic Delta variant mutations showed near-universal prevalence in Kazakhstan sequences: Spike D950N in 606/613 (98.86%) and NSP14 A394V in 520/613 (84.83%), both of which were also highly conserved globally (90.63% and 98.68%, respectively). Additional defining Delta mutations included Spike T478K in 553/613 (90.21%), Spike L452R in 545/613 (88.91%), and Spike G142D in 566/613 (92.33%), all with similarly high global frequencies (97.11%, 96.83%, and 60.53%, respectively).
The NS7a substitutions P45L, T120I, and V82A were detected in 509/613 (83.03%), 487/613 (79.45%), and 507/613 (82.71%) of Kazakhstan sequences, with corresponding global frequencies of 62.89%, 93.43%, and 91.53% (Table 3). These represent characteristic AY.122 lineage markers rather than Kazakhstan-specific variants. In contrast, NSP3 H1841Y showed notable enrichment in Kazakhstan, occurring in 56/613 (9.14%) sequences com-pared with only 0.30% globally, suggesting potential regional selection or distinct local transmission dynamics within the Kazakhstan SARS-CoV-2 Delta variant population. The first Omicron case was reported in Kazakhstan in January 2022, followed by a high epidemic wave and a drastic decline within less than a month. Sequenced BA.1/BA.2 viruses showed a wide geographic distribution but a strongly skewed temporal distribution: more than 50% of sequenced specimens were collected in January 2022, at the start of the Omicron BA.1/BA.2 wave in Kazakhstan (Figures 11 and12). The dominant lineage was BA.1 and its sublineages. The most prevalent BA.1.1 did not form a monophyletic cluster, with many branches of the phylogenetic tree clustering with Indian SARS-CoV-2 genomes (Figure 13). It is interesting to note that inbound tourism from India to Kazakhstan constantly grew from 1603 visitors in 2021 to 10,090 visitors in 2022 [25]. International air travel was resumed in Kazakhstan in September 2021. BA.2 genomes from Kazakhstan also clustered with SARS-CoV-2 viruses collected in India and Nepal. Phylogenetic clustering with the closest global sequences suggested a large local transmission event involving the BF.7 lineage, affecting many regions of Kazakhstan, with the nearest non-Kazakhstan BF.7 genome in the tree originating from Central America (Figure 16). In contrast, BE.1 sequences from Kazakhstan clustered mainly with European genomes (notably from Slovenia). BA.5.2 strains appeared to be of heterogeneous origin, with multiple introductions forming clusters closely related to Russian, Indian, and various European sequences. BA.4 genomes from Kazakhstan clustered with European SARS-CoV-2 sequences.
## 4. Discussion
Kazakhstan experienced five COVID-19 waves from March 2020 through 2023. We analyzed 3299 quality-filtered genomes (4462 from GISAID; 340 newly generated), stratifying by epidemic phase-pre-VOC (March 2020-9 February 2021), Alpha (9 February -June 2021), Delta (July-December 2021), Omicron BA.1/BA.2 (December 2021-May 2022), and Omicron BA.4/BA.5 (June-October 2022)-providing new insights into the spread of SARS-CoV-2 in Central Asia. At the same time, the very uneven sequencing intensity across waves (from 0.60‰ of cases during Delta to 11.57‰ during the Omicron BA.5 period) means that inferences about relative lineage diversity and persistence must be interpreted with caution, as the diversity of SARS-CoV-2 viruses in Kazakhstan is likely undersampled.
The pre-VOC phase in Kazakhstan was characterized by the co-circulation of at least 12 Pango lineages across six major clades (A, B, A.2, B.1, B.1.1, and B.4.1), mirroring the broad early diversity reported previously and consistent with multiple importations from both Asian and European/American sources [25]. The explicit association of some B.1.1 sublineages with the Tengiz oilfield outbreak is epidemiologically plausible: that complex has been documented as a major national hotspot in 2020, with >1000 infections among oilfield workers in early 2020 and strong evidence for intense workplace-associated transmission [22].
The strong concentration of B.1.1.440 in Houston, Texas (home to NASA's Johnson Space Center), together with its apparent phylogenetic connection to Baikonur (a major spaceport) via the Northern Mariana Islands (which host space-launch facilities), may reflect travel associated with space-launch operations. However, this interpretation remains speculative and cannot be confirmed with the available data.
Earlier work suggested that B.4.1 may have arisen independently in Kazakhstan; the Kazakhstan-restricted monophyletic clade that disappeared after April 2020 provides strong support for this and illustrates how geographically constrained lineages can emerge and then go extinct without contributing to the later global VOC landscape. Similar short-lived local lineages have been described elsewhere and are generally interpreted as the product of founder effects and transient ecological opportunities, which are then overwhelmed by fitter variants or by changes in mobility and control measures.
During the Alpha period, B.1.1.7's widespread geographic presence but scattered phylogenetic placement-dominated by singletons and small clades rather than one or two large monophyletic clusters-indicates that Kazakhstan experienced numerous Alpha introductions, most of which failed to generate large, sustained community transmission chains. This pattern closely parallels detailed B.1.1.7 analyses from Denmark [26], Mexico [27], and wastewater-based studies in Europe, where repeated introductions were common but only a subset of lineages achieved major expansion. The strong local enrichment of ORF1b I28T (NSP12:I37T) among Kazakhstan Alpha genomes, despite its rarity globally, likely reflects a combination of founder effects and expansion of a particular B.1.1.7 sublineage rather than clear adaptive change.
The observed Delta period in Kazakhstan fits well into the global picture of rapid Delta replacement in mid-2021 but with some striking regional nuances. Delta became globally dominant by mid-2021, accounting for nearly all sequenced infections by late August, and similar timing has been reported across Europe and the Middle East [28].
The dominance of AY.122 carrying the nsp2:K81N and ORF7a:P45L substitutions closely parallels the situation in Russia, where >90% of Delta sequences shared this mutation pair and were assigned to AY.122, a combination that remained rare in most other countries. Klink et al. also noted that Kazakhstan was among the few settings outside Russia with a high frequency of this AY.122 signature, suggesting intense epidemiological connectivity across the region during the Delta wave. The broad geographic distribution of these AY.122 viruses within Kazakhstan allows the speculation of a scenario in which one or a few successful introduction(s) of the "Russian-like" AY.122 sublineage were amplified by sustained community transmission. At the same time, the identification of a monophyletic AY.122 cluster lacking the canonical nsp2:K81N + ORF7a:P45L combination and grouping phylogenetically with a virus of Indian origin indicates that AY.122 circulation in Kazakhstan was not genetically homogeneous. Rather, it likely reflects at least two epidemiologically distinct AY.122 sources-a Russian-linked K81N+P45L sublineage and a second introduction related to an unknown source. Similar coexistence of multiple AY.* sublineages arising from separate importation events has been documented in England and other well-sampled settings, where some sublineages expand locally while others remain confined to small clusters [29].
The BA.1/BA.2 Omicron wave in Kazakhstan, first detected in January 2022, was short but intense, with a rapid rise and decline in reported cases, consistent with the high intrinsic transmissibility and immune escape properties of BA.1/BA.2 observed globally. The genomic data show that BA.1 and its sublineages, particularly BA.1.1, dominated this wave and were widely distributed across the country, but sequencing was heavily concentrated in January, which likely captures the early expansion phase while under-representing subsequent transmission. The fact that Kazakhstan BA.1.1 genomes do not form a single monophyletic clade and instead intersperse with multiple Indian sequences, together with BA.2 genomes clustering with viruses from India and Nepal, supports a scenario of repeated Omicron introductions from South Asia rather than expansion of a single local founder lineage. The resumption of international air travel in late 2021 and the marked increase in inbound tourism from India provide plausible human mobility pathways for these introductions. However, uneven temporal sampling and limited sequencing depth constrain precise reconstruction of introduction routes and onward spread.
Overall, these results show that Kazakhstan's epidemic was shaped by repeated international introductions, workplace and community amplification of a subset of those importations, and the rapid turnover of locally restricted lineages by globally successful VOCs. Sustained, more uniform genomic surveillance across regions and epidemic phases would not only improve reconstruction of past transmission dynamics in Kazakhstan but also strengthen the country's ability to detect and characterize future variants of concern.
## References
1. 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*
2. Huang, Wang, Li et al. (2020) "Clinical features of patients infected with 2019 novel coronavirus in Wuhan" *Lancet*
3. Wu, Liu, Gong (2015) "A Structural Overview of RNA-Dependent RNA Polymerases from the Flaviviridae Family" *Int. J. Mol. Sci*
4. Follis, York, Nunberg (2006) "Furin cleavage of the SARS coronavirus spike glycoprotein enhances cell-cell fusion but does not affect virion entry" *Virology*
5. Boni, Lemey, Jiang et al. (2020) "Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic" *Nat. Microbiol*
6. Li, Giorgi, Marichannegowda et al. "Emergence of SARS-CoV-2 through recombination and strong purifying selection"
7. Munnink, Worp, Nieuwenhuijse et al. (2021) "The next phase of SARS-CoV-2 surveillance: Real-time molecular epidemiology" *Nat. Med*
8. Tegally, Wilkinson, Lessells et al. (2021) "Sixteen novel lineages of SARS-CoV-2 in South Africa" *Nat. Med*
9. Meng, Kemp, Papa et al. (2021) "Recurrent emergence of SARS-CoV-2 spike deletion H69/V70 and its role in the Alpha variant B.1.1.7. Cell Rep"
10. Zhou, Yu, Du et al. (2020) "Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan"
11. (2025) *Harvard Medical School Coronavirus Resource Center. Available online*
12. Battakova, Imasheva, Slazhneva et al. (2023) "Public Health Response Measures for COVID-19 in Kazakhstan" *Disaster Med. Public Health Prep*
13. Zhussupov, Saliev, Sarybayeva et al. (2021) "Fakhradiyev, I. Analysis of COVID-19 pandemics in Kazakhstan" *J. Res. Health Sci*
14. (2026) *Viruses*
15. Kairov, Amanzhanova, Karabayev et al. "A high scale SARS-CoV-2 profiling by its whole-genome sequencing using Oxford Nanopore Technology in Kazakhstan"
16. Andre, Lau, Pokharel et al. (2023) "From alpha to omicron: How different variants of concern of the SARS-Coronavirus-2 impacted the world" *Biology*
17. Cui, Shi, Yimamaidi et al. (2023) "Dynamic variations in COVID-19 with the SARS-CoV-2 Omicron variant in Kazakhstan and Pakistan" *Infect. Dis. Poverty*
18. Tukusheva (2023) "Kazakhstan Reports First Cases of Eris COVID-19 Variant. Kursiv.media"
19. Aksamentov, Roemer, Hodcroft et al. (2021) "Clade assignment, mutation calling and quality control for viral genomes" *J. Open Source Softw*
20. Katoh, Standley (2013) "MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability" *Mol. Biol. Evol*
21. Kozlov, Darriba, Flouri et al. (2019) "RAxML-NG: A fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference" *Bioinformatics*
22. Menardo, Loiseau, Brites et al. (2018) "Treemmer: A tool to reduce large phylogenetic datasets with minimal loss of diversity" *BMC Bioinform*
23. Nabirova, Taubayeva, Maratova et al. (2020) "Factors Associated with an Outbreak of COVID-19 in Oilfield Workers" *Int. J. Environ. Res. Public Health*
24. Usserbayev, Zakarya, Kutumbetov et al. "Near-complete genome sequence of a SARS-CoV-2 variant B.1.1.7 virus strain isolated in Kazakhstan" *Microbiol. Resour. Announc*
25. Usserbayev, Abduraimov, Kozhabergenov et al. "Complete Coding Sequence of a Lineage AY.122 SARS-CoV-2 Virus Strain Detected in Kazakhstan" *Microbiol. Resour*
26. Yegorov, Goremykina, Ivanova et al. "COVID-19 Genomics Research Groupon behalf of the Semey COVID-19 Epidemiology Research Group. Epidemiology, clinical characteristics, and virologic features of COVID-19 patients in Kazakhstan: A nation-wide retrospective cohort study"
27. Michaelsen, Bennedbaek, Christiansen et al. (2022) "Introduction and transmission of SARS-CoV-2 lineage B.1.1.7, Alpha variant, in Denmark" *Genome Med*
28. Zárate, Taboada, Muñoz-Medina et al. "The Alpha Variant (B.1.1.7) of SARS-CoV-2 Failed to Become Dominant in Mexico"
29. Chen, Azman, Chen et al. (2022) "Global landscape of SARS-CoV-2 genomic surveillance and data sharing" *Nat. Genet*
30. Eales, Page, Martins et al. "SARS-CoV-2 lineage dynamics in England from September to November 2021: High diversity of Delta sub-lineages and increased transmissibility of AY"
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=PMC12782236&blobtype=pdf | # Zika Virus Outbreak -Bangladesh, September-December 2024
Jannatul Ferdous, ; Md, Abdullah Nasif, Gretchen Cowman, Drph, Immamul Muntasir, Mohammad Hassan, Omar Qayum, Kyaw Thowai, Prue Prince, Ahmed Alam, ; Manjur, Hossain Khan, ; Fariha, Masfiqua Malek, Rozina Afroz, Mahbubur Rahman, Tahmina Shirin, Jim O'neill, Althea Grant-Lenzy, Michael Berkwits, Rachel Gorwitz, Stacy Simon, Morgan Thompson, Suzanne Webb, Terraye Starr, Alexander Gottardy, Maureen Leahy, Armina Velarde, Quang Doan, Phyllis King, Moua Yang, Timothy Jones, Matthew Boulton, Carolyn Brooks, Virginia Caine, Jonathan Fielding, David Fleming, William Halperin, Jewel Mullen, Jeff Niederdeppe, Patricia Quinlisk, Patrick Remington, Carlos Roig, William Schaffner, Morgan Swanson, Kiana Cohen, Leslie Hamlin, Lowery Johnson, Will Yang
## Abstract
Zika virus infection is transmitted to humans primarily through the bite of infected Aedes species mosquitoes. Although most Zika virus disease cases are mild or asymptomatic, severe neurologic complications can occur. Infection during pregnancy can result in severe congenital anomalies. In Bangladesh, Zika virus was first detected in an archived specimen from 2014; subsequently, five cases of Zika virus disease were identified in 2023. In September 2024, in response to identification of a confirmed Zika virus disease case in Bangladesh's capital, Dhaka, in a woman aged 29 years who was initially thought to have dengue, the Institute of Epidemiology, Disease Control and Research (IEDCR) launched an outbreak investigation. After IEDCR notification to hospitals, five additional Zika virus disease cases were identified in four patients evaluated at three Dhaka hospitals and in a household contact of one of these patients. Another four Zika virus disease cases were identified through Zika virus testing of patients referred to IEDCR during a concurrent chikungunya outbreak. In total, 10 confirmed cases of Zika virus disease were detected in and around Dhaka during September-December 2024 in patients with no history of international travel. None of the patients was pregnant, and all recovered without hospitalization or complications. An entomological investigation detected Zika virus RNA in Aedes species mosquitoes in Dhaka. This investigation suggests sporadic Zika virus transmission occurs in Dhaka. Integrated testing and surveillance for arboviral diseases might improve detection of Zika virus disease and support clinical management in areas where transmission of multiple arboviral diseases occurs. Prevention of these infections through vector control and use of personal protective measures should also be emphasized.
## Introduction
Zika virus is a single-stranded RNA virus belonging to the Flavivirideae family and was first identified in Uganda in 1947 (1). The first documented Zika virus disease outbreak occurred on the Pacific island of Yap in 2007 (2). An association between fetal microcephaly and congenital Zika virus infection was observed during a large outbreak in Brazil during 2015-2016 (3). Zika virus is primarily transmitted by the bite of infected Aedes species mosquitoes. Less common modes of transmission include intrauterine and intrapartum transmission, sexual transmission, transmission through breastfeeding, blood transfusion or laboratory transmission, and transmission through organ and tissue transplantation (4). Exposure to Zika virus in early pregnancy can result in congenital Zika syndrome, which is characterized by microcephaly, brain abnormalities, vision problems, low birth weight, and other conditions (5).
Zika virus was first detected in Bangladesh when reverse transcription-polymerase chain reaction (RT-PCR) testing of archived dengue-negative serum samples collected between 2013 and 2016 identified a single Zika-positive specimen from 2014 in a patient with no history of international travel (6). A study conducted by a nongovernmental research organization during July-December 2023 in the capital city of Dhaka identified five Zika virus disease cases among 152 febrile patients tested for dengue, Zika, and chikungunya viruses (7). No routine surveillance for Zika virus is conducted in Bangladesh, although it is a notifiable disease (8). In September 2024, the Institute of Epidemiology, Disease Control and Research (IEDCR), under the Ministry of Health and Family Welfare in Bangladesh, was notified by a private hospital in Dhaka of a laboratory-confirmed Zika virus disease case in a person with dengue-compatible symptoms and no history of international travel. This report describes the outbreak investigation that was prompted by identification of this case.
## Investigation and Results
## Identification of Index Case
On September 4, 2024, a woman in Dhaka aged 29 years who was not pregnant (patient A) and had symptoms of dengue (fever, joint pain, and rash) that began on August 31 received a positive Zika virus RT-PCR test result from a multiplex test (Genesig kit | Primer Design | United Kingdom) for dengue, chikungunya, and Zika virus while being evaluated at the outpatient department of hospital 1. The test had been requested because clinicians, after receipt of a negative rapid dengue nonstructural protein-1 (NS1) antigen diagnostic test result, suspected chikungunya. The multiplex test detected Zika virus RNA only. The hospital notified IEDCR and sent a serum sample to the IEDCR virology laboratory for confirmatory testing.* IEDCR also obtained a serum and urine sample from the patient. All samples tested positive for Zika virus RNA by RT-PCR.
## Epidemiologic Investigation and Surveillance
On September 4, 2024, IEDCR initiated an investigation to identify the source of the Zika virus infection and determine whether additional cases were occurring. Because in Bangladesh the occurrence of single laboratory-confirmed case with suspected local transmission is treated as an outbreak and prompts an outbreak investigation and response, on September 9, after confirmation of the first case, IECDR declared a Zika virus disease outbreak in Dhaka. A confirmed case was defined as detection of Zika virus RNA by RT-PCR in serum or urine in the IEDCR laboratory. This activity was reviewed by IEDCR and CDC, deemed not research, and was conducted consistent with applicable federal law and CDC policy. † After obtaining information about patient A's clinical signs and symptoms from the treating physician at hospital 1, investigators visited the woman at home and collected information including demographic data, symptom onset date, clinical manifestations, and possible sources of exposure (e.g., mosquito bite, blood transfusion, organ transplantation, sexual contact, travel history, and travel history of sexual partners). As part of the investigation, all five of patient A's asymptomatic household members (age range = 22-48 years) were interviewed about potential exposures and travel history, and all provided blood specimens. All five household contacts' blood test results were negative for Zika virus. IEDCR reported the case to the International Health Regulations national focal point in Bangladesh, informed Dhaka hospitals and the public through a media briefing, and notified the Obstetrical and Gynecological Society of Bangladesh. IEDCR issued a statement to all government hospitals in Bangladesh instructing them to refer any patient with suspected Zika virus disease, chikungunya, or dengue for testing.
After investigation of the index case, IEDCR established a suspected Zika virus disease case definition that consisted of onset of fever and maculopapular rash during the previous 2 weeks, with at least one of the following: conjunctivitis, arthralgia, or having a household member with confirmed Zika virus disease. In response to subsequent notifications of suspected Zika virus disease cases, IEDCR collected information from the reporting facilities, confirmed diagnoses by RT-PCR testing at the IEDCR virology laboratory, interviewed patients, and conducted active case finding at patients' residences, including interviewing household members and conducting laboratory testing of those with suspected Zika virus disease.
## Identification of Cases by Hospitals
On September 17, IEDCR was notified by a second hospital in Dhaka (hospital 2) of two patients with RT-PCR-confirmed Zika virus disease (patients B and C) (Table ). Both were evaluated as outpatients. Physicians had initially suspected dengue and ordered Zika virus testing after receiving negative results for dengue NS1 antigen. An investigation at patient B's home identified a household member who met the suspected case definition and was later confirmed to have Zika virus disease (patient D). On October 5 and October 20, two additional hospitals (hospitals 3 and 4) in Dhaka reported laboratory-confirmed Zika virus disease cases in outpatients (patients E and F). No additional cases were identified among these patients' household members.
## Identification of Cases Among Patients with Suspected Chikungunya Referred for Testing
During October, a concurrent chikungunya outbreak was detected in Dhaka (9). In response to IEDCR's letter requesting referral of patients with suspected Zika virus disease, chikungunya, or dengue for testing, health facilities began referring patients with presumed chikungunya as part of this outbreak for testing. IEDCR established a sample collection booth and screened patients for testing using a questionnaire that collected demographic and clinical information. During October-December 2024, a total of 394 referred patients, most of whom were suspected to have chikungunya, provided serum samples for testing by RT-PCR multiplex assay. Among these, 34 (8.6%) met the suspected Zika virus disease case definition and provided urine samples for RT-PCR testing as well. Overall, four additional Zika virus disease cases were confirmed: three had Zika virus RNA detected in urine (patients G, H, and J) and one received a positive serum test result (patient I). §
## Characteristics of Persons with Zika Virus Disease
Among the 10 patients with confirmed Zika virus disease cases identified during the outbreak, the median patient age was 37 years (range = 23-52 years); seven cases occurred in women, none of whom was pregnant (Table ). All patients were interviewed using the same questionnaire. Sexual contact, blood transfusion, and organ transplantation were ruled out as possible routes of transmission. No patient had a history of international travel within the 2 weeks preceding symptom onset, and only two cases appeared to be epidemiologically linked. All patients had relatively mild illnesses, all received supportive care, and none were hospitalized. All patients had fever (≥100.4°F [≥38.0°C)]), arthralgias, and myalgias. Nine also had a generalized rash, seven experienced headaches, and six had conjunctivitis. The median duration of illness was 7 days (range = 5-14 days). Zika virus RNA was detected in the serum of two patients, the urine of six patients, and both the serum and urine of two patients. Nine patients lived in various locations in Dhaka, and one lived in the adjacent Gazipur district (Figure). Although there were concurrent dengue and chikungunya outbreaks, no co-infections were detected in patients with confirmed Zika virus disease.
## Entomological Investigation
To ascertain whether Zika virus was present in mosquitoes in Dhaka, IEDCR conducted an entomological investigation at seven sites. Each site was within approximately 0.5 miles (1 km) of the home of a patient with confirmed Zika virus disease. ¶ Larvae were collected from one or two ponds or lakes at each site. The larvae were then combined by site, reared to adulthood in the entomology laboratory, and tested for Zika virus RNA by an RT-PCR multiplex test in the virology § Among those tested, positive results for chikungunya, dengue, and Zika virus disease were received by 138, 11, and four persons, respectively. ¶ Nine confirmed Zika virus disease cases were identified in Dhaka. Seven sites were selected, covering the nine patients' homes within a radius of approximately 0.5 miles (1 km). Two patients lived in the same household, and two patients lived within approximately 0.5 miles (1 km) of each other. No entomological investigation was carried out in Gazipur district.
## Public Health Response
The Public Health Emergency Operations Center of IEDCR coordinated the response, rapidly deploying teams to investigate each reported Zika virus disease case. IEDCR notified the media about the presence of Zika virus in Dhaka, and the information was distributed online to raise awareness among clinicians and the public. Zika virus screening was incorporated into evaluation of patients referred to IEDCR for RT-PCR testing for chikungunya. Screening for Zika virus, in addition to dengue and chikungunya, has continued during the 2025 dengue and chikungunya season (June-December), and surveillance for acute febrile illness has been initiated in six sites throughout Bangladesh where Zika has been designated a priority disease. In addition, IEDCR initiated an arboviral serosurvey in Dhaka to gain a better understanding of the presence and extent of Zika virus and other arboviruses in the city. Vector control programs for dengue, organized by city authorities are ongoing to reduce mosquito density. Although routine surveillance for Zika virus disease in obstetrics and gynecology departments of health care facilities has not been established, awareness has been raised that obstetric patients with clinical signs and symptoms compatible with Zika virus disease, dengue, or chikungunya should be referred to IEDCR for testing. Provisional data from 2025 indicate fewer than 10 confirmed Zika virus disease cases were identified in Bangladesh during June-November 2025 (IEDCR, unpublished data, November 2025).
## Discussion
This investigation found that transmission of Zika virus is occurring in Dhaka, Bangladesh, and surrounding areas. This is the first reported Zika virus disease outbreak of its size in Dhaka, and 2024 is the second consecutive year that Zika virus disease has been detected in Dhaka. Although Zika virus was first identified in an archived sample from 2014, and the first confirmed outbreak occurred in 2023, reoccurrence in 2024 highlights the potential for the virus to become endemic in Bangladesh. The widespread geographic distribution of cases in 2024 suggests established circulation of infected mosquitoes in Dhaka; this distribution is distinctly different from that in 2023, when all five patients with confirmed Zika virus disease lived within a radius of approximately 0.5 miles (1 km) (7). The number of Zika virus disease cases detected likely does not reflect the true magnitude and geographic distribution of Zika virus transmission in Bangladesh. Cases are likely underreported because of the absence of systematic surveillance, limited availability of testing, occurrence of mild or asymptomatic infections, and possible misdiagnosis of Zika virus disease cases as chikungunya or dengue. The tenth case occurred in the adjacent Gazipur district (north of Dhaka), which is not included on the map. Two patients lived in the same household.
Dengue is an important public health concern in Bangladesh, with annual seasonal outbreaks (10). The existence of a national dengue surveillance system provides an opportunity to expand surveillance to include Zika virus disease and chikungunya, thereby enabling an integrated approach to monitoring and response to these mosquito-borne infections. The potential for severe neurologic complications of Zika virus disease, along with the risks during pregnancy, including congenital Zika syndrome, underscore the critical importance of detection of Zika virus disease in Bangladesh. These findings highlight the need to reduce transmission and prevent future outbreaks by improving vector control, using personal protection, ensuring timely diagnosis, and integrating Zika virus disease surveillance into the existing dengue surveillance system.
## Implications for Public Health Practice
Raising clinician awareness of the presence of Zika virus in Bangladesh might encourage routine Zika virus testing of febrile patients with suspected dengue or chikungunya. This is particularly important for pregnant women with febrile illness or rash to ensure early detection. Efforts to increase community awareness of the importance of preventing mosquito bites should be emphasized, especially for pregnant women.
Integration of Zika virus testing and surveillance with that for other arboviral diseases in patients with compatible signs and symptoms might improve the detection of Zika virus disease and support appropriate clinical management, including antenatal monitoring of pregnant patients with Zika virus disease. This outbreak suggests that Zika virus is established in mosquitoes in Bangladesh. Establishing Zika virus surveillance, strengthening vector control measures, and educating providers and the public are important to prevent further transmission, improve case detection, and guide clinical management.
## References
1. Dick, Kitchen, Haddow (1952) "Zika virus. I. Isolations and serological specificity" *Trans R Soc Trop Med Hyg*
2. Duffy, Chen, Hancock (2009) "Zika virus outbreak on Yap Island, Federated States of Micronesia" *N Engl J Med*
3. Hennessey, Fischer, Staples (2015) "Zika virus spreads to new areasregion of the Americas" *MMWR Morb Mortal Wkly Rep*
4. Gregory, Oduyebo, Brault (2017) "Modes of transmission of Zika virus" *J Infect Dis*
5. Freitas, Souza-Santos, Carvalho (2020) "Congenital Zika syndrome: a systematic review" *PLoS One*
6. Muraduzzaman, Sultana, Shirin et al. (2017) "Introduction of Zika virus in Bangladesh: an impending public health threat" *Asian Pac J Trop Med*
7. Hasan, Hossain, Zamil (2025) "Concurrent transmission of Zika virus during the 2023 dengue outbreak in Dhaka" *PLoS Negl Trop Dis*
8. Siraj, Dewey, Hassan (2020) "The Infectious Diseases Act and resource allocation during the COVID-19 pandemic in Bangladesh" *Asian Bioeth Rev*
9. Nasif, Haider, Muntasir (2024) "The reappearance of chikungunya virus in Bangladesh" *IJID Reg*
10. Sharif, Opu, Saha (2024) "Evolving epidemiology, clinical features, and genotyping of dengue outbreaks in Bangladesh, 2000-2024: a systematic review" *Front Microbiol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12652532&blobtype=pdf | # Circulating Tumor DNA as a Biomarker for Precision Medicine in Prostate Cancer: A Systematic Review
Dario Marchetti, Hiroshi Miyamoto, El Annaz, H Abi, R Tagajdid, Nouhaila Chanhih, Abdelilah Laraqui, Salma Hassine, Ahmed Ameur, Larbi Hamedoun, Hicham El Annaz, Rachid Abi, Mohamed Tagajdid, Idriss Amine, Khalid Ennibi, Abdelaziz Benjouad, Lamiae Belayachi
## Abstract
Circulating tumor DNA (ctDNA) profiling offers non-invasive insights for personalized prostate cancer management. This systematic review provides the first comprehensive appraisal of ctDNA assay methods, genomic targets, and their clinical correlations and proposes practical recommendations to guide future standardization and validation. We searched PubMed, ScienceDirect, Scopus, and the Cochrane Library starting December 2024 following PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines. From 229 records, 44 studies (10,631 patients) met the inclusion criteria. Plasma ctDNA analyzed by NGS predominantly profiled TP53 (72.7%), AR (70.4%), BRCA1/2 (61.3%), ATM (50%), RB1 (47.7%), and PTEN (41%). ctDNA positivity and specific key alterations correlated with poorer overall and progression-free survival. BRCA1/2-mutant patients benefited from Olaparib plus Abiraterone, while persistent alterations predicted early progression. Beyond synthesizing existing evidence, we identify key gaps, such as inconsistent reporting of variant allele fractions, limited diversity in study populations, and underexplored rare alterations. We recommend unified reporting standards (e.g., variant allele frequency thresholds and panel composition) and prioritized prospective trials to validate high-impact targets. These steps will accelerate the integration of ctDNA into routine precision oncology practice worldwide.
## 1. Introduction
Precision medicine (PM) in oncology is an emerging approach that relies on distinct characteristics of a sub-population, such as genetic profile, environment, and lifestyle, to predict disease risk, diagnose disease, tailor therapies, and reduce clinical complications. This approach advances cancer care by shifting from a one-size-fits-all paradigm to molecularly guided interventions. In prostate cancer (PCa), PM tools such as OMICs have shown promise in prevention, diagnosis, prognosis, and therapies, offering increased efficacy and reduced toxicity [1].
PCa is the second most commonly diagnosed cancer among men and one of the leading causes of cancer-related mortality worldwide [2]. It is a highly heterogeneous disease, with clinical behaviors ranging from indolent to highly lethal and genetic backgrounds that influence prognosis and therapeutic decisions in advanced stages. Traditional diagnostic approaches rely on prostate-specific antigen (PSA) blood testing, digital rectal examination (DRE), and confirmation through transrectal ultrasound (TRUS)-guided biopsy. However, both PSA testing and DRE have limited specificity and sensitivity: DRE does not improve PSA-based detection [3], and PSA testing can be influenced by noncancerous conditions such as benign prostatic hyperplasia or inflammation [4], as well as by age.
Liquid biopsy has emerged as a promising alternative, providing a dynamic and noninvasive tool to characterize tumor biology and support the realization of PM [5]. It can guide the need for tissue biopsy, offer prognostic insights in advanced disease, and monitor treatment response in clinical trials. Nonetheless, its integration into clinical workflows remains limited [6].
Liquid biopsy detects circulating tumor-derived analytes such as ctDNA, cfDNA, exosomes, and proteins, reflecting tumor heterogeneity and evolution [7]. Among these, ctDNA is the most widely studied marker in clinical practice. Derived from apoptotic or necrotic tumor cells, ctDNA shows high sensitivity and specificity across multiple cancer types and stages [8]. Through genomic analysis, ctDNA enables early detection, determination of tissue of origin, prognosis, treatment monitoring, assessment of resistance, and detection of minimal residual disease [9].
In PCa, the ctDNA fraction correlates with overall survival (OS), progression-free survival (PFS), and treatment response, outperforming several established prognostic factors [10]. Detectable alterations from ctDNA, such as AR and HRR mutations, have been identified as predictive markers in castration-resistant PCa (CRPC) [11][12][13][14], while the ctDNA fraction increases with disease progression in hormone-sensitive PCa (HSPC) [15,16].
We therefore conducted a systematic review to investigate the relationship between genetic alterations detected through ctDNA sequencing and clinical outcomes in localized and metastatic PCa. Our aim is to highlight clinically meaningful alterations and assess their translational potential, thereby clarifying the role of ctDNA in precision medicine and its future integration into clinical practice.
## 2. Materials and Methods
Our review adhered to the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) guidelines (Figure 1). It was preregistered on the International Prospective Register of Systematic Reviews (CRD42025628570). Recommendations sections to provide a transparent evaluation of the potential impact on the review's findings.
We also searched for pertinent abstracts from leading multidisciplinary medical oncology associations, including the American Society of Clinical Oncology and the European Society for Medical Oncology, to uncover the latest research and newly published studies.
To address potential duplication in the data, studies with overlapping datasets were carefully evaluated and used the assistance of Covidence tools. Preference was given to the most comprehensive and recent publications, prioritizing studies based on their dataset completeness, publication date, and relevance. This approach ensured that duplicate data did not bias the findings.
To further mitigate the risk of duplication, we used The National Clinical Trial (NCT) numbers to identify publications arising from the same clinical trials. In cases where multiple publications with the same NCT number were identified, these publications were grouped, and their datasets were compared to identify overlaps. Data from the most comprehensive and recent publication were included to avoid redundancy and bias. Distinct analyses or non-overlapping data reported across publications were considered separately and appropriately documented.
Duplicate titles were first eliminated. We then conducted a title check to assess relevance. The remaining publications' abstracts were examined, and irrelevant ones were removed. For the remaining articles, full-text manuscripts and/or conference posters or presentations were reviewed. When multiple versions of the same data were available, preference was given to the most comprehensive and recent updates.
## 2.2. Inclusion and Exclusion Criteria
Studies were selected if they investigated localized or advanced PCa patients (patients), who underwent ctDNA testing (interventions) compared to those who did not undergo ctDNA testing or who did not detect any alterations in their ctDNA (comparisons) to assess the differential pathologic clinical outcomes (outcome) in case reports, clinical trials, metaanalyses, multicenter studies, observational studies, RCTs, systematic reviews, clinical studies, cohort studies, and original research (study design). Studies lacking original patient data, letters, editorial comments, replies from authors, and non-English language papers were excluded. All publications included had their references checked for relevant additional research.
## 2.3. Data Extraction
Two authors independently extracted the following data: author(s), year of publication, study design, country, sample size, inclusion/exclusion criteria, age, disease stage (localized or advanced PCa), treatment regimen, type of ctDNA extraction and detection methods, and genetic alterations identified. Data related to OS, PFS, treatment response, and any other relevant clinical outcomes (e.g., recurrence and progression) were also extracted in addition to specific ctDNA alterations identified and associated clinical outcomes. Available hazard ratios (HRs), p-values, and confidence intervals (CIs) of each clinical outcome were collected.
## 2.4. Risk-of-Bias Assessment
The risk of bias was evaluated using five validated tools, each tailored to the study design of the included papers. For RCTs, the RoB 2 tool was applied to assess bias across five domains: the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. For prognostic studies examining associations between ctDNA features and clinical outcomes-whether standalone or embedded within larger trials-the QUIPS tool was used. This framework evaluates six domains: study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding, and statistical analysis/reporting.
The NOS was used to assess the methodological quality of non-randomized cohort and observational studies, including both retrospective and prospective designs. The NOS evaluates three broad domains: selection of study groups, comparability of cohorts, and ascertainment of either the exposure or the outcome of interest. For non-randomized studies of interventions that aimed to explore causal relationships, the Risk Of Bias In Non-randomized Studies-of Interventions (ROBINS-I) tool was applied. This tool assesses the risk of bias across seven domains: confounding, selection of participants, classification of interventions, deviations from intended interventions, missing data, measurement of outcomes, and selection of the reported result. Each domain, as well as the overall study, was rated as having low, moderate, serious, or critical risk of bias.
In addition, the JBI critical appraisal checklist for analytical cross-sectional studies was used for studies with a cross-sectional design. This tool assesses methodological soundness based on inclusion criteria, measurement validity, confounding factors, and statistical analysis. Case reports, abstract-only or in vitro-only studies, and small descriptive case series were excluded from formal risk-of-bias assessment due to the lack of applicable tools and their limited generalizability.
Two authors independently performed the risk-of-bias assessments using the appropriate tool for each study type. Discrepancies were resolved through discussion and consensus.
## 3. Results and Discussion
## 3.1. Eligible Studies
A total of 229 publications were identified through database search (Figure 1). Of these, 44 (19.2%) met the inclusion criteria and were included in our systematic review.
Most studies were conducted in North America (13, 29.5%) [17][18][19][20][21][22][23][24][25][26][27][28][29] and Asia (10, 22.7%) [30][31][32][33][34][35][36][37][38][39]. Additional studies were conducted in Europe (2, 4.5%) [40,41], across multiple continents (17,38.6%) [42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58], Australia (2, 4.5%) [59,60], and South America (1, 2.3%) [61]. Overall, 17 (38.6%) studies were retrospective [18,19,[26][27][28][29][31][32][33][34][35]39,42,43,53,55,59], 7 (15.9%) were prospective [21,23,30,40,44,45,60], and 10 (22.7%) were RCTs [46][47][48][49]52,54,[56][57][58]61]. The remaining (9, 20.5%) were case reports, case series, or observational biomarker studies [20,22,23,25,37,38,50]. One study included both prospective and retrospective cohorts [51]. The longest study period was reported by Reimers et al. [19], which covered 30 years (1988-2018), while the shortest study by Du et al. [34] lasted only 8 months (October 2019-June 2020). The majority of studies focused on data collected within the last decade, with several large trials, such as those by Oya et al. [46] and Clarke et al. [47], contributing extensive patient cohorts. Study characteristics are provided in the Supplementary Material (Table S1).
## 3.2. Risk-of-Bias
Risk of bias was assessed across all eligible included studies using validated tools tailored to their respective study designs. For randomized controlled trials, the riskof-bias (RoB) 2 tool identified two studies, both with an overall risk of some concerns (Supplementary Table S2).
The QUIPS tool was applied to prognostic studies evaluating ctDNA features and clinical outcomes, with 12 out of 20 studies judged as low risk across most domains and 8 studies presenting moderate risk due to limitations in prognostic factor or outcome measurement, study participation or attrition, confounding control, or incomplete reporting (Supplementary Table S3).
Among non-randomized cohort and observational studies assessed using the NOS, two studies out of five were rated as high quality (low risk of bias), while three showed moderate quality due to weaknesses in comparability, outcome, or selection (Supplementary Table S4).
ROBINS-I was used for two non-randomized intervention studies, with one rated at moderate risk and one at serious risk of bias, largely due to confounding, missing data, and selective reporting (Supplementary Table S5). Additionally, two analytical cross-sectional studies were evaluated using the JBI tool, with one rated as serious and one showing some risk mainly due to confounding (Supplementary Table S6).
Together, these assessments ensured that the diverse study designs were evaluated using appropriate, design-specific tools. While most studies were rated as having good and moderate quality, caution is warranted when interpreting findings from non-randomized and analytical cross-sectional studies due to residual confounding, limited control over selection bias, and variable reporting standards.
## 3.3. Study and Patient Characteristics
The study and patients' characteristics are reported in the Supplementary Material (Table S1).
Metastatic CRPC (mCRPC) was analyzed in thirty-five (79.5%) studies [18][19][20][21][22]24,26,27,29,32,33,[36][37][38][40][41][42][43][45][46][47][49][50][51][52][54][55][56][57][58][59][60][61], Metastatic HSPC (mH-SPC) was analyzed in two (4.5%) studies [31,34], Aggressive-Variant Prostate Cancer (AVPC)/Small Cell Carcinoma of Prostate (SCCP) was present in two (4.5%) studies [25,39], and Non-metastatic Prostate Cancer (nmPCa) was analyzed in one (2.2%) study [35]. The remaining studies did not restrict the PCa subtype in their analysis and included all PCa subtypes (2, 4.5%) [30,33] (Figure 2). OS was the most frequent outcome (21, 47.7%) [18,19,22,23,26,29,32,36,38,[41][42][43]45,47,49,51,52,54,[58][59][60], followed by PFS (17, 38.6%) [18,19,22,23,26,32,36,38,[41][42][43]45,49,51,54,59,60], PSA Response or PSA-PFS (11, 25%) [18,19,22,23,32,36,42,43,49,51,59], rPFS (10, 22.7%) [19,22,23,36,43,45,49,51,54,59], Time To Progression (TTP) (5, 11.4%) [19,22,43,49,51], Time on Treatment (ToT) (4, 9.1%) [19,22,43,49]), and Castration Resistance-Free Survival (CRFS) (1, 2.3%) [47]. A total of 10,631 patients were analyzed across all studies. The sample sizes across studies varied widely, ranging from single-patient mCRPC case reports [20, 25,37,38,50] to large-scale clinical trials involving up to 2462 patients with advanced PCa, including metastatic and non-metastatic disease, as well as CRPC and mCSPC [53]. The median age of patients across studies ranged from 39 [25] to 94 years [18], with most studies reporting median ages between 70 and 75 years. However, 17 studies [17,21,30,31,33,43,[45][46][47][48]51,52,[54][55][56][57][58] did not provide explicit age distribution data.
PSA levels at diagnosis or follow-up varied significantly among studies. Wyatt et al. [29] reported a PSA ranging from 3.4 to 4478 ng/mL, whereas Yuan et al. [37] described a patient whose PSA decreased from 787 ng/mL at diagnosis to 8.02 ng/mL at best response, rising again to 601 ng/mL upon treatment resistance. Similarly, Fettke et al. [60] observed PSA values ranging from 0.51 to 2719 ng/mL, highlighting the heterogeneity of disease burden among study populations.
## 3.4. Methods for Specimen Detected
Sample types, detection methods, and characteristics of genomic alterations and aspect types are summarized in the Supplementary Material (Table S7).
Blood plasma ctDNA was the most primary sample type for testing to identify actionable germline and/or somatic alterations [18][19][20][21][22][23][52][53][54][55][56][57][58][59][60][61][62]. Analysis of ctDNA in both plasma and urine samples has been performed in only one study [30]. Sample type was not specified in one study [18]. Using plasma rather than serum minimizes the dilution of tumor-derived DNA with leukocyte cfDNA, thereby enhancing detection sensitivity, especially for low variant allele frequency (VAF) mutations [63][64][65][66][67].
Serum samples are more likely to miss alterations with low VAFs, with up to 44.8% of such alterations undetected compared to matched plasma, highlighting the superiority of plasma-based testing [68][69][70][71][72][73][74][75][76][77][78][79][80].
For ctDNA extraction, the majority of studies (31, 70.45%) did not explicitly mention the DNA extraction method [18][19][20]22,[24][25][26]28,30,32,35,[37][38][39][40][41][42][43][46][47][48][49][50]52,53,[56][57][58][59][60][61], while the remaining studies (13, 29.55%) reported to have commonly used either column-based or magnetic bead DNA extraction methods [21,23,27,29,31,33,34,36,44,45,51,54,55] (Figure 3). A column-based DNA extraction kit was used in 10 (22.73%) studies [21,23,29,30,34,36,40,44,51,54,59] (Figure 3). The remaining (3, 6.82%) studies used various magnetic bead extraction kits [31,33,45] (Figure 3). Column-based methods typically yield higher cfDNA recovery but capture larger DNA fragments, whereas bead-based methods favor shorter fragments and may provide lower yields [81][82][83][84][85][86][87][88][89][90][91][92]. These differences can influence assay sensitivity, particularly in settings where ctDNA concentration is low [93][94][95][96][97][98].
Nearly every study used next-generation sequencing (NGS) as the primary detection method for ctDNA [18][19][20][21][22][23][52][53][54][55][56][57][58][59][60][61][62] (Figure 4). Whereas the only study that combined digital droplet PCR (ddPCR) with NGS for targeted ctDNA analysis was performed by Conteduca et al. [41]. All studies that performed NGS used targeted sequencing (40, 91%) [18][19][20][21][22][23]59,60], except for two (4.5%) that combined targeted sequencing and lowpass whole-genome sequencing [40,51] and two (4.5%) that combined targeted sequencing and whole-exome sequencing [39,49] (Figure 4). NGS provides a tumor-agnostic approach that captures the molecular profiles of both primary and metastatic tumors. However, technical challenges remain: low ctDNA concentrations can reduce sensitivity, and accurate detection of low-frequency variants (<0.5-1% VAF) remains difficult [92,[99][100][101]. Recent advances such as molecular barcoding and in silico error suppression have improved the reliability of detecting variants with VAFs below 1% [92,101], yet robust performance across all assays is only achieved above ~0.5% VAF. This underscores the importance of using highly sensitive assays and reporting VAF thresholds consistently.
## 3.5. Frequent Genes and Their Somatic/Germline and Genomic Alteration Aspect Type
Blood plasma ctDNA was the most primary sample type for testing to identify actionable germline and/or somatic alterations. Nearly every study used NGS as the primary detection method for ctDNA. TP53 was the most frequently analyzed gene (32, 72.7%) (Table 1, Figure 5) [18,19,[21][22][23][25][26][27][28][30][31][32][33][37][38][39][40]44,45,[48][49][50][51][53][54][55][56][57][58]61], followed by AR (androgen receptor) (31, 70.4%) [18,[20][21][22][23][25][26][27][28][29][30][32][33][34][37][38][39][40]44,45,48,49,51,[53][54][55][56][58][59][60][61], BRCA1 and BRCA2 (27, 61.3%) [18,19,21,22,[26][27][28][29][31][32][33][34][35][36]38,[42][43][44][45][46][47][48][49]52,53,56,57,61], ATM (22, 50%) [18,19,21,22,[25][26][27][28][29][30][32][33][34]36,[43][44][45][46]48,49,52,53], RB1 (21, 47.7%) [20, 21,29,30,32,33,38,39,44,45,[48][49][50][51]53,55,60], and PTEN (18, 40.9%) [18,21,22,29,32,33,36,37,39,45,48,51,[53][54][55]58,60]. BRCA1, BRCA2, ATM, and CHEK2 were the most common genes (61.4%) that exhibited both somatic and germline alterations. The remaining exhibited either somatic (TP53: 72.7%, AR: 70.5%, and RB1: 47.7%) or germline (PTEN: 40.9%) mutations, with somatic being more prevalent. Somatic inactivating mutations and deletions in TP53 were observed in nearly all included publications [18][19][20][21][22][23][24][52][53][54][55][56][57][58][59][60][61][62]. When specific mutation sites are reported, they often include hotspot changes (e.g., R248Q and R273H) or large deletions spanning multiple exons, all of which disrupt the tumor-suppressor function of TP53 [18][19][20][21][22][23][24][52][53][54][55][56][57][58][59][60][61][62]. AR aberrations-spanning somatic amplifications, point mutations (e.g., L702H, T878A, V716M, W742C, and W742L), and structural rearrangements-are some of the most reported [18,[20][21][22][23][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][43][44][45][46][47][48][49][52][53][54][55][56][58][59][60][61][62]. In BRCA1/2, germline and somatic alterations were reported, including frameshift mutations, nonsense mutations, and large genomic rearrangements (LGRs) [18][19][20][21]23,24,[26][27][28][29][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][52][53][54][55][56][57][58][59][60][61][62]. Alterations in ATM encompassed both somatic and germline events, including missense mutations, splice site mutations, and LGRs [18][19][20][21]23,24,[26][27][28][29][31][32][33][34][35][36][37][38][39][40][41][43][44][45][46][47]49,50,[52][53][54][55][56][57][58][59][60][61][62]. As for RB1, alterations are mainly somatic deletions and inactivating mutations [18][19][20][21][22][23][24][52][53][54][55][56][57][58][59][60][61][62]. Somatic deletions, copy number losses, and inactivating point mutations were the commonly observed in PTEN [18][19][20][21][22][23][24][26][27][28][29][30][31][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][52][53][54][55][56][57][58][59][60][61][62]. Finally, somatic activating point mutations and amplifications were the most frequently observed in the PIK3CA gene [18][19][20][21][22][23][24][52][53][54][55][56][57][58][59][60][61][62]. Copy number variations (CNVs) (CNG/amplifications or losses/deletions) are common in genes such as TP53, AR, RB1, PTEN, and MYC [26,28,31,37,39,44,45,48,49,55,59,61]. Frameshift and nonsense mutations have been especially reported in DNA repair genes like BRCA1/2, ATM, CHEK2, PALB2, and FANCA [19,28,32,[36][37][38]49,52]. In the case of AR, multiple ligand-binding domain positions are repeatedly mutated, such as L702H, T878A, V716M, and H875Y [26,28,41,44,49,60]. Additionally, gene fusions (e.g., TMPRSS2-ERG) and various
LGRs (particularly in BRCA1/2 and ATM) are also reported [23,31,38,43,49,50]. Additional changes and mutations were also reported.
## 3.6. Association of ctDNA Detection with Outcome Survival
The associations between ctDNA-detected genomic alterations and survival outcomes (Table 2) as well as therapy response (Table 3) are presented in simplified form for the main text. A more detailed version of these associations is provided in Supplementary Table S8. A cross-study summary heatmap of gene-endpoint associations is shown in Figure 6. S8.
Impact of ctDNA state on OS: Here, 16 studies [18,21,23,27,[39][40][41][44][45][46][47][48]51,55,59,60] totaling 3193 patients reported a relationship between the ctDNA state and OS in PCa patients, including 2917 with mCRPC [18,21,23,27,40,41,[44][45][46][47]51,55,60], 250 with mHSPC [44], and 63 with AVPC [39]. Oya et al. [46] investigated the combined therapy of two molecules-Olaparib and Abiraterone-in which they screened patients' ctDNA to detect Homologous Recombination Repair (HRR) genes (BRCA1/2, ATM, CDK12, PALB2, and RAD51B/D). In this part of the PROpel trial, the combination of Olaparib plus Abiraterone demonstrated a trend toward improved OS in the intention-to-treat population of patients with mCRPC. Specifically, the median OS was 42.1 months for the Olaparib plus Abiraterone group compared to 34.7 months for the placebo plus Abiraterone group, with an HR of 0.81 (95% CI: 0.67-1.00; p = 0.0544). Notably, in the subgroup of patients with BRCA1/2 mutations, the survival benefit was more pronounced. In this group, the median OS was not reached in the Olaparib plus Abiraterone arm, whereas it was 23.0 months in the placebo plus Abiraterone arm, corresponding to an HR of 0.29 (95% CI: 0.14-0.56). Not reaching the OS median means that more than half of the patients in that group were still alive at the time of the analysis, implying a strong survival benefit from the combined treatment [46]. Agarwal et al. [48] worked on the combination of Apalutamide + ADT. Genes screened via ctDNA were AR, TP53, PTEN, RB1, PIK3CA, and HRR genes (BRCA2, ATM, etc.). The result was AR amplification at baseline, HR 1.9 or 6.7, all p < 0.05) for poor OS [48]. Clarke et al. [47] investigated the efficacy of Olaparib vs. SOC and resulted in OS (not reached) benefit in HRR+ (BRCA1/2: HR 0.29; CI: 0.14-0.56) upon screening BRCA1/2, ATM, CDK12, and CHEK2 (HRR pathway) genes in association with OS [47]. Knutson et al. [55] investigated ARPI combinations via screening AR, TP53, RB1, PTEN, MYC, and MYCN genes, resulting in worse OS for AR alterations (e.g., LBD truncations) [55]. De Bono et al. [58] evaluated 177Lu-PSMA-617 vs. ARPI and reported that AR, TP53, and PTEN alterations detected in patients' ctDNA were associated with shorter rPFS (AR: HR 1.954, 95% CI 1.333-2.865, and p < 0.001; TP53: HR 1.655, 95% CI 1.13-2.426, and p < 0.01; PTEN: HR 1.62, 95% CI 1.018-2.578, and p < 0.05) [58]. Jayaram et al. [45] explored the combined therapy of Apalutamide + ADT and reported worse OS (TP53: HR 7.13, 95% CI 2.37-21.47, and p < 0.001; RB1: HR 6.24, 95% CI 1.97-19.73, and p = 0.002; PTEN: HR 11.9, 95% CI 3.6-39.34, and p < 0.001) when persistent TP53/RB1/PTEN alterations were screened in ctDNA [45].
Impact of ctDNA state on PFS: Here, 21 studies [21,23,34-37,39-41,45-49,52,54,55,57-60] involving 3783 patients reported on the association between the ctDNA state and PFS, including 3099 with mCRPC [21,23,36,40,[45][46][47]49,52,54,55,57,58], 265 with CRPC [41], 161 with nmPC [35], 129 with mCSPC [48], and 63 with AVPC [39]. In the PROpel NCT03732820 study [47], Abiraterone combined with Olaparib significantly prolonged PFS compared with Abiraterone and placebo as first-line treatment for patients with mCRPC enrolled irrespective of HRRm status (HR 0.50, 95% CI: 0.34-0.73), with the greatest benefit in BRCA1/2-mutated subgroups (HR 0.23, 95% CI: 0.12-0.43). In the TITAN study [43], Apalutamide plus ADT improved OS, delayed castration resistance, maintained healthrelated quality of life, and improved radiographic PFS in a broad population of patients with mCSPC compared to the control. The PROpel study [56], which focused on patients with mCRPC, showed that Abiraterone combined with Olaparib significantly improved rPFS compared to Abiraterone plus placebo (HR 0.66, 95% CI: 0.54-0.81, p < 0.001). In the Alliance A031201 study [55], which analyzed ARPI combinations in mCRPC, AR alterations predicted worse rPFS, emphasizing their role in resistance. In the TITAN ctDNA study [45], the detection of ctDNA pre-treatment was significantly associated with shorter PFS (HR: 2.05, 95% CI: 1.36-3.11, and p < 0.0002). Multivariable analysis confirmed that ctDNA positivity remained an independent prognostic factor for PFS after adjusting for established clinical variables (HR 2.03, 95% CI: 1.21-3.41, and p < 0.007). The PS-MAfore NCT04689828 study reported shorter rPFS with AR (HR 1.954; 95% CI 1.333-2.865; p < 0.001), TP53 (1.655; 1.13-2.426; p < 0.01) and PTEN (1.62; 1.018-2.578; p < 0.05) alterations, linking ctDNA profiles to treatment resistance [58]. The ProBio investigation found that TP53-altered patients had inferior PFS (STR 0.76), suggesting limited efficacy of platinum therapy in genomically aggressive disease [57]. An association was observed between a positive ctDNA state and poor PFS in mCRPC followed by mCSPC and AVPC, which highly probably indicates an association between positive ctDNA state and poor PFS in patients with mCRPC [21,23,[34][35][36][37][39][40][41][45][46][47][48][49]52,54,55,[57][58][59][60].
## 3.7. Discussion
Understanding genomic landscapes of PCa is important, owing to the emergence of PM to guide treatment selection and improve survival outcomes. Integrated genomic and proteogenomic characterization of prostate tumors identifies biological insights and subtype-specific therapeutic strategies. ctDNA profiling provides non-invasive access to a malignancy's molecular landscape, overcoming the limitations of tissue biopsy [64,76]. In cases where archival tissue is not available for testing and tissue-based analysis is not clinically feasible, NGS-based HRR gene testing of ctDNA from plasma provides a minimally invasive alternative [29,[65][66][67]. There are also inherent advantages in profiling the latest available sample from a patient with advanced disease. Park et al. [68] showed that 11% of mCRPC patients harbored actionable genetic alterations by serial NGS based-ctDNA testing, with 3.5% of tests detecting a new BRCA2 genetic alteration or MSI-high [68]. ctDNA harbored some BRCA1/2 alterations not identified by tissue testing, and it was enriched in therapy resistance alterations, as well as possible clonal hematopoiesis mutations (e.g., in ATM and CHEK2) [78]. mPCa presents challenges in the collection of a tissue specimen since metastases are often confined to bone, requiring a technically difficult, invasive biopsy [71,79]. ctDNA testing holds promise for contributing to improved management of PCa from localized disease to metastatic stages, though its role is not yet fully established in routine practice. Thus, certain technical and biological challenges remain to be addressed before widespread clinical implementation in the field. False negative results can occur due to low levels of ctDNA in the patient's blood [65]. Disease burden, including a PSA > 10 ng/mL, lymph node-only disease, and normal LDH are strongly associated with detecting somatic mutations in cfDNA NGS studies [72,73]. Clonal hematopoiesis of indeterminate potential (CHIP) can also interfere with ctDNA testing, potentially leading to false positive results. In addition, accurate determination of copy number variations from ctDNA has been shown to be difficult, especially in low-tumor-fraction settings [65]. Because of these challenges, it is important to make ctDNA tests more precise. To move toward clinical integration, clear best practices, standard protocols, and quality measures for ctDNA analysis using NGS are still required. It is also important to collect ctDNA samples at the right time, especially when the cancer is clearly progressing [66,74,75].
Plasma-based ctDNA analysis has shown advantages over serum. Using plasma minimizes dilution of tumor-derived DNA compared to serum, enhancing sensitivity [76]. In serum, leukocyte cfDNA released during clotting dilutes ctDNA, reducing variant allele frequency (VAF) detection [77][78][79]. Up to 44.8% of low-VAF alterations were missed in serum compared to plasma, and concordance with somatic mutations in tissue biopsy was lower [80]. Thus, retrospective studies using serum must consider these limitations. VAFs in ctDNA are being considered one of the markers with prospective clinical utility, and the use of VAFs analysis may provide information about response to treatment and patient prognosis and help in developing optimal therapy. Recent studies demonstrate an association between high VAF level in ctDNA and shorter OS among patients with metastatic disease [81][82][83][84].
In cases where detectable alterations are found in ctDNA, the concordance level between genomic analysis of tumor tissue DNA and blood-derived ctDNA when acquired at the same time can be as high as 80-90% [85]. Shiota et al. [86] showed that the sensitivity of ctDNA testing represented by the ratio of detected gene alterations in blood among those in tissues from mPCa patients was 41.5% [86]. The sensitivities of ctDNA testing in mCSPC and mCRPC were 43.8% and 35.9%, respectively. The concordance status of alterations in the genes with a frequency of over 5% in tissue or ctDNA at pre-treatment among matched patients [86]. Among them, 33.7% alterations were concordant between tissue and ctDNA, whereas 66.3% alterations were discordant. AR alterations were detected only in ctDNA from patients with mCRPC, whereas there was no AR alteration in tissue, which was mostly obtained before hormonal therapy [87]. Simultaneous sequencing of ctDNA and tumor DNA in individuals with metastatic disease results in high levels of concordance between ctDNA and tissue profiling for driver gene analysis, suggesting the potential utility of ctDNA analysis in advanced stages of disease [87]. Wyatt et al. [29] conducted a study on the concordance of ctDNA with time-matched metastatic tissue biopsy in PCa and showed that 33% of somatic mutations were detected exclusively in liquid biopsy samples [29]. In mPC, ctDNA is a high-fidelity substitute for solid tumor tissue-derived DNA and is capable of not only recapitulating the somatic landscape of a tumor but also identifying clinically relevant driver alterations missed by a single metastatic biopsy [29,51,88]. Although archival tissue from primary tumors, including formalin-fixed, paraffin-embedded diagnostic biopsies, would be best for molecular characterization [89,90], this does not usually reflect metastatic disease that has evolved under treatment selective pressure and does not account for intra-patient heterogeneity. For example, metastatic samples are generally enriched for RB1 and TP53 mutations, while treatment-naive PCa tissues are not, but they can harbor HRR defects [91].
The choice of cfDNA extraction method is important in ctDNA studies given reported disparities in cfDNA yield and composition across commonly used kits. In our systematic review, column-based DNA extraction kits were used in 10 (22.73%) studies [21,23,29,30,34,36,40,44,51,54,59]. The remaining studies did not explicitly mention the ctDNA extraction method [18][19][20]22,[24][25][26]28,30,32,35,[37][38][39][40][41][42][43][46][47][48][49][50]52,53,[56][57][58][59][60][61], or used various magnetic bead extraction kits [31,33,45], (31, 70.45%, and 3, 6.82%, respectively). Multiple studies have evaluated the performance of the various cfDNA methods [92][93][94][95][96] [97]. Although spin column-based methods are more time consuming and more costly than magnetic bead-based approaches, they typically produce higher yields. Column-based methods showed superior cfDNA yield, albeit with a tendency to capture larger DNA fragments. Magnetic bead-based methods yielded less cfDNA but exhibited a bias toward recovering shorter DNA fragments [98]. There was a high level of consensus among studies regarding both cfDNA extraction methods, with a majority of studies using column-based methods. Several comparative studies report that bead-based kits yield higher DNA recovery rates than column-based kits, particularly at low DNA input, though absolute recovery percentages vary depending on the platform.
In regard to the sequencing method, NGS was the most utilized method accounting for the totality of studies [18][19][20][21][22][23][52][53][54][55][56][57][58][59][60][61][62]. These assays include amplicon based in 13.63% of studies [21,23,31,36,41,54] and hybridization capture-targeted NGS in 70.45% of studies [19,22,[24][25][26][27][28][30][31][32][33][34][37][38][39][42][43][44][45][46][47][49][50][51][52][53]56,59]. The remaining seven (15.9%) studies [35,40,48,57,58,60,61] did not specify the type of NGS used. Both strategies are followed by highly redundant ("deep") sequencing to allow for the relative amount of mutant and wild-type DNA molecules at each locus to be accurately counted.
NGS offers distinct advantages by allowing a tumor-agnostic approach for cfDNA analysis. This approach permits the identification of variants without the need for molecular profiling of tumor tissue, offering a systemic view of the disease by capturing the molecular profiles of both primary and metastatic tumors. However, one of the most difficult parts of accurately reporting liquid biopsy NGS results is indeed related to ctDNA samples with variants detected at low VAF. For example, Williams et al. [99] showed decreased detection of SNVs at low VAFs when using two allele-specific ddPCR and three (amplicon and hybrid capture) NGS panels [99]. As the ctDNA concentration in the blood is low, most platforms use methods to increase the signal of certain regions to increase the sensitivity of ctDNA detection at low tumoral VAFs [100]. Reliable detection of mutations below 0.5% VAF remains a key challenge for ctDNA tumor DNA sequencing approaches. Recent advancements in NGS technologies have enhanced sensitivity by implementing strategies such as molecular barcoding and in silico error suppression. These innovations allow the reliable differentiation of real mutations with VAFs < 1% from background artifacts [92]. The analytical performance evaluation of NGS-based ctDNA assays has undergone in earlier studies. Deveson et al. [101] evaluated the ctDNA detection performance of five leading ctDNA assay platforms developed based on NGS across 12 clinical and research facilities. Above 0.5% VAF, ctDNA mutations were detected with high sensitivity, precision, and reproducibility by all five assays, whereas below this limit of 0.5%, detection became unreliable, making it a key challenge for ctDNA sequencing analysis [101]. Detecting low-frequency ctDNA variants with a VAF < 1% in CRPC is important to identify early, subclonal, and clinically significant genomic alterations that may not be detectable by traditional methods, thereby improving treatment selection and prognostic risk stratification [102].
In our systematic review, TP53 was commonly detected, followed by alterations in AR, BRCA1/2, ATM, RB1, and PTEN genes. Alterations in TP53 in localized or metastatic hormone-sensitive PCa had a shorter time to CRPC, and cumulative gene hits in TP53, PTEN, and RB1 led to an incremental risk of progression with inferior OS with increasing gene hits [103]. Co-occurrence of alterations in TP53 and other important tumor suppressors, in particular RB1 or PTEN, was shown to render PCa tumor cells more aggressive or more resistant towards conventional therapies. In neuroendocrine PCa, the concomitant alteration of TP53 and RB1 is linked to aggressive clinical features [104] and responsiveness to platinum treatments [105]; however, it could be present also in CSPC, where is associated with increased risk of relapse and death [105]. In AR, W742C, W742L, L702H, and T878A were the most frequent points mutations detected. These hotspot mutations were revealed to change the binding affinity of ligands, including steroids and antiandrogens, and potentially result in altered responses to AR pathway inhibitors. W742C and W742L are associated with resistance to Bicalutamide/Enzalutamide by causing the AR antagonists to behave as agonists [106]. L702H is associated with resistance to Abiraterone/Enzalutamide and promotes the trans-activation of AR by glucocorticoids [89]. T878A is associated with resistance to Bicalutamide/Enzalutamide/Apalutamide and promiscuous activation by progesterone. The frequency of alterations in AR and HRR genes, including BRCA2 and ATM, was enriched in mCRPC compared with mCSPC in the most studies included in this review. These finding suggested that alterations in AR and HRR genes promote treatment resistance to initial ADT-based therapy in PCa [86]. Fan et al. [32] have shown that genomic alterations in AR and CDK12 genes were enriched in ctDNA with mCRPC compared with those with de novo mCSPC among Asians [32]. Previous studies showed that genomic alterations in AR, TP53, RB1, and PTEN are enriched during PCa progression [90,107,108]. Shaya et al. [18] found that mCRPC patients with more than one alteration on ctDNA analysis have inferior OS (26.1 months vs. 8.8 months, p < 0.001), suggesting that the presence and actionable alterations identified by NGS can predict OS in mCRPC [18].
Our analysis highlights the prognostic and predictive value of ctDNA in localized and metastatic PCa, particularly in the mCRPC setting. Across 16 studies evaluating OS and 21 studies examining PFS, a consistent trend emerged: ctDNA positivity or the presence of specific genomic alterations detected via ctDNA was significantly associated with worse clinical outcomes. The consistent association between positive ctDNA states and poor OS and PFS supports ctDNA as a non-invasive biomarker of disease aggressiveness. For example, alterations in TP53, RB1, and PTEN-frequently associated with poor prognosiswere strongly linked with significantly reduced survival (e.g., TP53: HR 7.13; PTEN: HR 11.9) [45], underscoring the utility of ctDNA profiling in risk stratification. Additionally, AR amplifications and truncations were associated with inferior OS and rPFS, indicating resistance to AR pathway inhibitors (ARPI) [48,55]. Other studies within our review, such as PROpel and TITAN, demonstrated that ctDNA profiling can inform therapeutic decision making. In PROpel, patients with BRCA1/2 mutations benefited significantly from Olaparib + Abiraterone combination therapy (HR for OS: 0.29) [46,47], illustrating how HRR mutations identified via ctDNA can guide the use of PARP inhibitors. Patients with BRCA1/2 mutations, who are often sensitive to therapies like PARP inhibitors, can develop acquired resistance when their cancer acquires new reversion mutations. These reversion mutations can restore a functional BRCA1/2 gene, allowing the cancer cells to repair DNA damage and continue growing, thus nullifying the effect of the combination therapy and leading to treatment resistance. This emergence of new mutations is a significant mechanism of secondary resistance in these cancers. Conversely, patients with AR, TP53, and PTEN alterations experienced limited benefits from standard therapies, suggesting a need for alternative regimens in these subgroups [55,58]. The association of AR, TP53, and PTEN alterations with shorter rPFS in several trials reflects emerging patterns of treatment resistance. These findings imply that ctDNA could serve not only as a baseline predictor but also as a real-time monitor of evolving resistance, enabling dynamic treatment adaptation. For example, the observation of persistently altered ctDNA profiles pre-and post-treatment in the TITAN study correlated with poor PFS, suggesting potential for ctDNA to inform early treatment modification [45]. The results of clinical implications and integration suggest that ctDNA-based genomic profiling has the potential to upgrade clinical management of metastatic PCa, with further evidence required for its full adoption into routine practice. Its use could (i) identify patients likely to benefit from targeted therapies (e.g., BRCA-mutated patients for PARPi), (ii) detect high-risk molecular subtypes (e.g., TP53/PTEN/RB1 alterations), and (iii) monitor treatment response and resistance emergence over time. Moreover, ctDNA offers an attractive alternative to invasive tissue biopsies, especially when tumor accessibility is limited, and may facilitate more frequent and patient-friendly molecular assessments. Importantly, this work was conducted by a Moroccan team as part of a national effort to lay the groundwork for the integration of ctDNA technologies in PCa diagnostics and treatment pathways. As a first of its kind in the region, this review aspires not only to inform clinical practice but also to ignite a larger research and policy momentum that may accelerate the adoption of PCa precision oncology in Morocco and the broader African context. In a field largely driven by data from high-income countries, this review highlights the potential for scientific leadership to emerge from North Africa, guided by both local needs and global standards.
## 3.8. Practical Recommendations
We have added a new "Practical Recommendations" subsection summarizing translational insights and clinical take-home points.
Based on the synthesis of 44 studies (10,631 patients) included in this review, we propose several practice-oriented recommendations for clinicians and researchers working with ctDNA in prostate cancer: a.
Sample type and handling: Plasma should be used rather than serum to minimize dilution by leukocyte cfDNA and to improve detection sensitivity. Pre-analytical variables (time to centrifugation, storage temperature, and number of freeze-thaw cycles) must be standardized to avoid degradation. b.
DNA extraction: Bead-based kits generally yield higher recovery rates than columnbased methods, particularly when ctDNA concentration is low. We recommend reporting recovery efficiency in each study to facilitate comparability. c.
Sequencing approaches: Next-generation sequencing (NGS) with unique molecular identifiers (UMIs) and error-suppression algorithms should be preferred for detecting low-frequency variants (<1% VAF). Targeted panels focusing on recurrent alterations (TP53, BRCA2, AR, and PTEN) are currently the most practical for clinical monitoring. d.
Interpretation of results: Variants associated with clonal hematopoiesis (e.g., ATM, CHEK2, and DNMT3A) should be carefully interpreted in parallel with matched white blood cell sequencing to reduce false positives. e.
Clinical integration: At present, ctDNA analysis is most useful to (i) identify resistance mechanisms to androgen receptor signaling inhibitors (ARSI), (ii) guide the addition of PARP inhibitors in patients with Homologous Recombination Repair (HRR) alterations, and (iii) monitor emerging mutations during treatment. f.
Reporting standards: Studies should systematically report the variant allele fraction (VAF), copy number thresholds, and assay sensitivity. Uniform reporting will accelerate meta-analyses and guideline development.
## 3.9. Limitations
This study has some limitations. Although 44 articles were included in our systematic review, a number of relevant studies had to be excluded due to insufficient data. Many lacked critical information, such as the clinical significance of specific genomic alterations, hazard ratios, and confidence intervals, which prevented us from conducting a comprehensive meta-analysis. Additionally, in several studies, ctDNA sequencing was performed on only a subset of enrolled patients, rather than the entire cohort. Some patients had only tissue biopsies without corresponding liquid biopsy data, limiting the ability to assess concordance between tissue and ctDNA findings and potentially biasing conclusions regarding their equivalence. We also came across studies that explored associations between ctDNA levels or fractions and treatment resistance or response; however, these were outside the scope of our research question. Furthermore, none of the included studies enrolled participants of African, Middle Eastern, or other under-represented ethnic backgrounds. This lack of diversity is a notable limitation, as these populations remain significantly underrepresented in prostate cancer research despite their relevance. Limited to no representation from African and Middle Eastern populations is considered a critical gap that must be addressed in future research to ensure global applicability of ctDNA-based strategies.
## 4. Conclusions and Clinical Implications
In this systematic review, we evaluated the relationship between genetic alterations detected through ctDNA sequencing and clinical outcomes in localized and metastatic prostate cancer.
Prognostic value: ctDNA positivity and alterations in TP53, AR, BRCA1/2, PTEN, and RB1 consistently correlate with poorer OS, shorter PFS, and resistance to therapy, particularly in mCRPC. Predictive value: HRR alterations (e.g., BRCA1/2) can guide treatment with PARP inhibitors, while AR pathway alterations predict resistance to AR-targeted therapies. Clinical utility: ctDNA profiling thus provides clinically meaningful information to (a) stratify prognosis, (b) guide personalized therapy selection, and (c) monitor disease progression over time. These findings directly address the main objective of our study and support the promising role of ctDNA-which requires further consolidation through standardized validation and broader clinical evidence-in advancing prostate cancer management as a non-invasive tool for precision oncology.
A significant limitation in the clinical application of ctDNA lies in the absence of unified guidelines from leading oncology societies such as National Comprehensive Cancer Network (NCCN), The American Society of Clinical Oncology (ASCO), and The European Society for Medical Oncology (ESMO). Currently, there is no standardized protocol for PCa ctDNA sample processing, sequencing methods, or result interpretation, which leads to variability across studies and challenges in cross-comparison. Furthermore, ctDNA assays exhibit technical limitations in detecting certain genomic alterations in PCa, particularly long-range reads (LRGs) and copy number variations. The LRGs often pose detection challenges due to fragmentation and low abundance in plasma, potentially resulting in missed clinically relevant targets. Addressing these gaps requires the development of harmonized workflows and improvements in assay sensitivity to fully harness the potential of ctDNA as a comprehensive biomarker in PCa.
Precision medicine is an approach that tailors treatment and prevention strategies to the individual, based on their genetic, molecular, environmental, and lifestyle characteristics, but this only works if we understand the molecular landscape across all populations. Currently, approximately 18-19% of the world's population lives in Africa, and the Middle East and North Africa (MENA) region has around 500 million people, or about 6-7% of global population. Yet in our review, no study included substantial representation from African or Middle Eastern cohorts. This is not a minor gap: genetic variation differs substantially among populations, and findings drawn mainly from European, North American, or East Asian groups may not generalize to regions with different ancestries. For example, allele frequencies, variant pathogenicity, linkage disequilibrium, and response to therapies (both in efficacy and adverse effects) can vary by population. Under--representation of African and Arab populations means that an entire ~25-30% of humanity is largely invisible to the genomic studies upon which precision oncology depends. Without more inclusive research-whole-genome sequencing, variant databases, local biobank efforts in Africa and the Middle East-we risk treating only a subset of patients well while leaving others behind. Expanding diversity is essential not just for equity, but for unlocking novel variants, improving prediction models, and increasing therapy success rates globally.
Despite promising findings, several challenges remain: heterogeneity in PCa ctDNA assay platforms and panels (e.g., 72-gene panels vs. broad 585-gene assays) limits crossstudy comparability. The clinical utility of ctDNA in PCa needs validation in prospective, randomized studies. Standardization of ctDNA thresholds and reporting is essential to integrate it into guidelines regarding PCa patient care.
## References
1. Medori, Micheletti, Gadler et al. "Omics sciences and precision medicine in prostate cancer" *Clin. Ter*
2. Bray, Laversanne, Sung et al. (2024) "Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries" *CA Cancer J. Clin*
3. Krilaviciute, Becker, Lakes et al. (2023) "Digital rectal examination is not a useful screening test for prostate cancer" *Eur. Urol. Oncol*
4. Han, Zhu, Liu et al. (2021) "Differential diagnosis of uncommon prostate diseases: Combining mpMRI and clinical information" *Insights Imaging*
5. Yang, Zhao, Dong et al. "Circulating tumor DNA and tissue complementarily detect genomic alterations in metastatic hormone-sensitive prostate cancer"
6. Ionescu, Zhang, Wang (1728) "Clinical applications of liquid biopsy in prostate cancer: From screening to predictive biomarker" *Cancers*
7. Foser, Maiese, Digumarthy et al. (2024) "Looking to the future of early detection in cancer: Liquid biopsies, imaging, and artificial intelligence" *Clin. Chem*
8. Gezer, Bronkhorst, Holdenrieder (1363) "The utility of repetitive cell-free DNA in cancer liquid biopsies" *Diagnostics*
9. Bin Riaz, Wang, Kohli (2018) "Liquid biopsy approach in the management of prostate cancer" *Transl. Res*
10. Fonseca, Maurice-Dror, Herberts et al. (1828) "Prediction of plasma ctDNA fraction and prognostic implications of liquid biopsy in advanced prostate cancer" *Nat. Commun*
11. Dong, Fan, Yang et al. (2021) "Use of circulating tumor DNA for the clinical management of metastatic castration-resistant prostate cancer: A multicenter, real-world study" *J. Natl. Compr. Cancer Netw*
12. Beltran, Romanel, Conteduca et al. (2020) "Circulating tumor DNA profile recognizes transformation to castration-resistant neuroendocrine prostate cancer" *J. Clin. Investig*
13. Romanel, Tandefelt, Conteduca et al. (2015) *Plasma AR and abiraterone-resistant prostate cancer. Sci. Transl. Med*
14. Annala, Vandekerkhove, Khalaf et al. (2018) "Circulating tumor DNA genomics correlate with resistance to abiraterone and enzalutamide in prostate cancer" *Cancer Discov*
15. Kwan, Wyatt, Chi (2022) "Towards clinical implementation of circulating tumor DNA in metastatic prostate cancer: Opportunities for integration and pitfalls to interpretation" *Front. Oncol*
16. Lopez-Valcarcel, Lopez-Campos, Zafra et al. (2024) "Liquid biopsy to personalize treatment for metastatic prostate cancer" *Am. J. Transl. Res*
17. Wyatt, Annala, Beja et al. (2016) "Genomic alterations in circulating tumor DNA (ctDNA) are associated with clinical outcomes in treatment-naive metastatic castration-resistant prostate cancer (mCRPC) patients commencing androgen receptor (AR)-targeted therapy" *Ann. Oncol*
18. Shaya, Nonato, Cabal et al. (2021) "Analysis of the prognostic significance of circulating tumor DNA in metastatic castrate-resistant prostate cancer" *Clin. Genitourin. Cancer*
19. Reimers, Yip, Zhang et al. (2020) "Clinical outcomes in cyclin-dependent kinase 12 mutant advanced prostate cancer" *Eur. Urol*
20. Hemenway, Tierno, Nejati et al. (2022) "Clinical utility of liquid biopsy to identify genomic heterogeneity and secondary cancer diagnoses: A case report" *Case Rep. Oncol*
21. Torquato, Pallavajjala, Goldstein et al. (2019) "Genetic alterations detected in cell-free DNA are associated with enzalutamide and abiraterone resistance in castration-resistant prostate cancer" *JCO Precis. Oncol*
22. Barata, Agarwal, Nussenzveig et al. (1065) "Clinical activity of pembrolizumab in metastatic prostate cancer with microsatellite instability high (MSI-H) detected by circulating tumor DNA" *J. Immunother. Cancer*
23. Dang, Chauhan, Ellis et al. (2020) "Cell-free DNA alterations in the AR enhancer and locus predict resistance to AR-directed therapy in patients with metastatic prostate cancer" *JCO Precis. Oncol*
24. Ravindranathan, Russler, Yantorni et al. (2021) "Detection of microsatellite instability via circulating tumor DNA and response to immunotherapy in metastatic castration-resistant prostate cancer: A case series" *Case Rep. Oncol*
25. Carneiro, Pamarthy, Shah et al. (2018) "Anaplastic lymphoma kinase mutation (ALK F1174C) in small cell carcinoma of the prostate and molecular response to alectinib" *Clin. Cancer Res*
26. Ledet, Lilly, Sonpavde et al. (2020) "Comprehensive analysis of AR alterations in circulating tumor DNA from patients with advanced prostate cancer" *Oncologist*
27. Dincman, Karam, Giordano et al. "Genomic amplifications identified by circulating tumor DNA analysis guide prognosis in metastatic castration-resistant prostate cancer"
28. Moses, Koksal, Ledet et al. (2020) "Evaluation of the genomic alterations in the androgen receptor gene during treatment with high-dose testosterone for metastatic castrate-resistant prostate cancer" *Oncotarget*
29. Wyatt, Annala, Aggarwal et al. (2017) "Concordance of circulating tumor DNA and matched metastatic tissue biopsy in prostate cancer" *J. Natl. Cancer Inst*
30. Chen, Jia, Chao et al. "Urine-and blood-based molecular profiling of human prostate cancer"
31. Yu, Cho, Choi et al. (2021) "Genomic mutation profiling using liquid biopsy in Korean patients with prostate cancer: Circulating tumor DNA mutation predicts the development of castration resistance" *Investig. Clin. Urol*
32. Fan, Fei, Zhu et al. (2021) "Comparative analysis of genomic alterations across castration sensitive and castration resistant prostate cancer via circulating tumor DNA sequencing" *J. Urol*
33. Bang, Won, Shin et al. (2023) "Circulating tumor DNA analysis on metastatic prostate cancer with disease progression" *Cancers*
34. Du, Fei, Wang et al. (2023) "Early serial circulating tumor DNA sequencing predicts the efficacy of chemohormonal therapy in patients with metastatic hormone-sensitive prostate cancer" *Transl. Oncol*
35. Fei, Du, Gong et al. (2023) "Early plasma circulating tumor DNA as a potential biomarker of disease recurrence in non-metastatic prostate cancer" *Cancer Res. Treat*
36. Pan, Zhao, Ni et al. (2022) "The prevalence and prognosis of next-generation therapeutic targets in metastatic castration-resistant prostate cancer" *Mol. Oncol*
37. Yuan, Liu, Yang et al. (2022) "Circulating tumor DNA genomic profiling reveals the complicated olaparib-resistance mechanism in prostate cancer salvage therapy: A case report" *World J. Clin. Cases*
38. Ma, He, Huang et al. (2018) "Response to olaparib in metastatic castration-resistant prostate cancer with germline BRCA2 mutation: A case report" *BMC Med. Genet*
39. Wang, Xu, Guo et al. (2023) "Concordance and clinical significance of genomic alterations in progressive tumor tissue and matched circulating tumor DNA in aggressive-variant prostate cancer" *Cancer Res. Commun*
40. De Laere, Oeyen, Mayrhofer et al. (2019) "TP53 outperforms other androgen receptor biomarkers to predict abiraterone or enzalutamide outcome in metastatic castration-resistant prostate cancer" *Clin. Cancer Res*
41. Conteduca, Wetterskog, Sharabiani et al. (2017) "Androgen receptor gene status in plasma DNA associates with worse outcome on enzalutamide or abiraterone for castration-resistant prostate cancer: A multi-institution correlative biomarker study" *Ann. Oncol*
42. Loehr, Hussain, Patnaik et al. (2023) "Emergence of BRCA reversion mutations in patients with metastatic castration-resistant prostate cancer after treatment with rucaparib" *Eur. Urol*
43. Chi, Barnicle, Sibilla et al. (2023) "Detection of BRCA1, BRCA2, and ATM alterations in matched tumor tissue and circulating tumor DNA in patients with prostate cancer screened in PROfound" *Clin. Cancer Res*
44. Kohli, Tan, Zheng et al. "Clinical and genomic insights into circulating tumor DNA-based alterations across the spectrum of metastatic hormone-sensitive and castrate-resistant prostate cancer"
45. Jayaram, Wingate, Wetterskog et al. (2021) "Plasma tumor gene conversions after one cycle abiraterone acetate for metastatic castration-resistant prostate cancer: A biomarker analysis of a multicenter international trial" *Ann. Oncol*
46. Oya, Armstrong, Thiery-Vuillemin et al. "Biomarker analysis and updated results from the phase III PROpel trial of abiraterone and olaparib vs abiraterone and placebo as first-line therapy for patients with metastatic castration-resistant prostate cancer" *Ann. Oncol*
47. Clarke, Armstrong, Thiery-Vuillemin et al. "Final overall survival in PROpel: Abiraterone and olaparib versus abiraterone and placebo as first-line therapy for metastatic castration-resistant prostate cancer" *J. Clin. Oncol*
48. Agarwal, Lucas, Aguilar-Bonavides et al. "Genomic aberrations associated with overall survival in metastatic castration-sensitive prostate cancer treated with apalutamide or placebo plus androgen deprivation therapy in TITAN" *J. Clin. Oncol*
49. Belic, Graf, Bauernhofer et al. (2018) "Genomic alterations in plasma DNA from patients with metastasized prostate cancer receiving abiraterone or enzalutamide" *Int. J. Cancer*
50. Sautois, Loehr, Watkins et al. (2022) "A case study of clinical response to rucaparib in a patient with metastatic castration-resistant prostate cancer and a RAD51B alteration" *Curr. Oncol*
51. Dong, Zheng, Zhang et al. "Circulating cell-free DNA-based detection of tumor suppressor gene copy number loss and its clinical implication in metastatic prostate cancer"
52. Carr, Adelman, Barnicle et al. (2021) "Homologous recombination repair gene mutation characterization by liquid biopsy: A phase II trial of olaparib and abiraterone in metastatic castrate-resistant prostate cancer" *Cancers*
53. Necchi, Cucchiara, Grivas et al. (2021) "Contrasting genomic profiles from metastatic sites, primary tumors, and liquid biopsies of advanced prostate cancer" *Cancer*
54. Goodall, Assaf, Shi et al. (2020) "Circulating tumor DNA (ctDNA) dynamics associate with treatment response and radiological progression-free survival (rPFS): Analyses from a randomized phase II trial in metastatic castration-resistant prostate cancer (mCRPC)" *J. Clin. Oncol*
55. Knutson, Luo, Kobilka et al. (2024) "AR alterations inform circulating tumor DNA detection in metastatic castration-resistant prostate cancer patients" *Nat. Commun*
56. Saad, Armstrong, Thiery-Vuillemin et al. "MP11-16 Prostate-specific antigen analyses in PROpel: Abiraterone and olaparib versus abiraterone and placebo as first-line therapy for metastatic castration-resistant prostate cancer" *J. Urol*
57. Kristiansen, Sautois, Crippa et al. "Efficacy of carboplatin in patients with metastatic castration-resistant prostate cancer: Results from the biomarker-driven, randomised, outcome-adaptive ProBio trial"
58. De Bono, Morris, Sartor et al. "Baseline ctDNA analyses and associations with outcomes in taxane-naive patients with mCRPC treated with 177 Lu-PSMA-617 versus change of ARPI in PSMAfore" *J. Clin. Oncol*
59. Lin, Mak, Yeung et al. (2021) "Overcoming enzalutamide resistance in metastatic prostate cancer by targeting sphingosine kinase"
60. Fettke, Kwan, Bukczynska et al. (2021) "Independent prognostic impact of plasma NCOA2 alterations in metastatic castration-resistant prostate cancer" *Prostate*
61. Annala, Taavitsainen, Khalaf et al. (2021) "Evolution of Castration-Resistant Prostate Cancer in ctDNA during Sequential Androgen Receptor Pathway Inhibition" *Clin. Cancer Res*
62. Sobhani, Sirico, Generali et al. (2020) "Circulating cell-free nucleic acids as prognostic and therapy predictive tools for metastatic castrate-resistant prostate cancer" *World J. Clin. Oncol*
63. Lee, Kim, Seong et al. (2020) "Plasma vs. serum in circulating tumor DNA measurement: Characterization by DNA fragment sizing and digital droplet polymerase chain reaction" *Clin. Chem. Lab. Med*
64. Chan, Yeung, Lui et al. (2005) "Effects of preanalytical factors on the molecular size of cell-free DNA in blood" *Clin. Chem*
65. Rolfo, Mack, Scagliotti et al. (2018) "Liquid biopsy for advanced non-small cell lung cancer (NSCLC): A statement paper from the IASLC" *J. Thorac. Oncol*
66. Ignatiadis, Sledge, Jeffrey (2021) "Liquid biopsy enters the clinic-implementation issues and future challenges" *Nat. Rev. Clin. Oncol*
67. Pittella-Silva, Chin, Chan et al. (2020) "Plasma or serum: Which is preferable for mutation detection in liquid biopsy?" *Clin. Chem*
68. Park, Chu, Li et al. "Repeat next-generation sequencing testing on progression in men with metastatic prostate cancer can identify new actionable alterations. JCO Precis"
69. Tukachinsky, Madison, Chung et al. (2021) "Genomic analysis of circulating tumor DNA in 3334 patients with advanced prostate cancer identifies targetable BRCA alterations and AR resistance mechanisms" *Clin. Cancer Res*
70. Wong, Mohamad, Giaze et al. (2019) "Ima-Nirwana, S. Prostate cancer and bone metastases: The underlying mechanisms" *Int. J. Mol. Sci*
71. Gandaglia, Abdollah, Schiffmann et al. (2014) "Distribution of metastatic sites in patients with prostate cancer: A population-based analysis" *Prostate*
72. Schweizer, Gulati, Bs et al. (2019) "Clinical determinants for successful circulating tumor DNA analysis in prostate cancer" *Prostate*
73. Herberts, Wyatt (2021) "Technical and biological constraints on ctDNA-based genotyping" *Trends Cancer*
74. Batool, Hsia, Beecroft et al. (2023) "Extrinsic and intrinsic preanalytical variables affecting liquid biopsy in cancer" *Cell Rep. Med*
75. Armstrong, Taylor, Haffner et al. (2024) "Germline and somatic testing for homologous repair deficiency in patients with prostate cancer (part 1 of 2)"
76. Huffman, Parikh (2021) "The Emerging Role of Circulating Tumor DNA: Will Tissue Become Obsolete?"
77. Alix-Panabières, Pantel (2021) "Liquid biopsy: From discovery to clinical application" *Cancer Discov*
78. Bayle, Peyraud, Belcaid et al. (2022) "Liquid versus tissue biopsy for detecting actionable alterations according to the ESMO Scale for Clinical Actionability of molecular Targets in patients with advanced cancer: A study from the French National Center for Precision Medicine (PRISM)" *Ann. Oncol*
79. Trujillo, Wu, Wetterskog et al. (2022) "Blood-based liquid biopsies for prostate cancer: Clinical opportunities and challenges" *Br. J. Cancer*
80. Taavitsainen, Annala, Ledet et al. (2019) "Evaluation of commercial circulating tumor DNA test in metastatic prostate cancer" *JCO Precis. Oncol*
81. Peleg, De La Fuente, Lichtman et al. (2024) "Prognostic value of variant allele frequency in circulating tumor DNA for survival outcomes in metastatic non-small cell lung cancer" *J. Thorac. Oncol*
82. Kato, Shiota, Nishimoto et al. (2025) "Prognostic significance of circulating tumor DNA alterations in advanced renal cell carcinoma from SCRUM-Japan MONSTAR-SCREEN: A nationwide genomic profiling project" *Br. J. Cancer*
83. Podany, Foffano, Gerratana et al. "Racial differences in ctDNA profiles, targeted therapy use, and outcomes in metastatic breast cancer" *JAMA Netw*
84. Lee, Kim, Hwang et al. (2024) "Longitudinal comparative analysis of circulating tumor DNA and matched tumor tissue DNA in patients with metastatic colorectal cancer receiving palliative first-line systemic anti-cancer therapy" *Cancer Res. Treat*
85. Mandel, Hoeh, Humke et al. (2024) "Feasibility of next-generation sequencing of liquid biopsy (circulating tumor DNA) samples and tumor tissue from patients with metastatic prostate cancer in a real-world clinical setting in Germany" *Eur. Urol. Focus*
86. Shiota, Matsubara, Kato et al. "Genomic profiling and clinical utility of circulating tumor DNA in metastatic prostate cancer: SCRUM-Japan MONSTAR SCREEN project"
87. Zhang, Dong, Li et al. (2020) "Consistency of genotyping data from simultaneously collected plasma circulating tumor DNA and tumor-DNA in lung cancer patients" *J. Thorac. Dis*
88. Matsubara, De Bono, Olmos et al. (2023) "Olaparib Efficacy in Patients with Metastatic Castration-resistant Prostate Cancer and BRCA1, BRCA2, or ATM Alterations Identified by Testing Circulating Tumor DNA" *Clin. Cancer Res*
89. Sumiyoshi, Chi, Wyatt (2021) "Clinical implications of genomic alterations in metastatic prostate cancer" *Prostate Cancer Prostatic Dis*
90. Lozano, Castro, Aragón et al. (2021) "Genetic aberrations in DNA repair pathways: A cornerstone of precision oncology in prostate cancer" *Br. J. Cancer*
91. Mateo, Seed, Bertan et al. (2020) "Genomics of lethal prostate cancer at diagnosis and castration resistance" *J. Clin. Investig*
92. Kloten, Rüchel, Brüchle et al. (2017) "Liquid biopsy in colon cancer: Comparison of different circulating DNA extraction systems following absolute quantification of KRAS mutations using Intplex allele-specific PCR" *Oncotarget*
93. Pérez-Barrios, Nieto-Alcolado, Torrente et al. (2016) "Comparison of methods for circulating cell-free DNA isolation using blood from cancer patients: Impact on biomarker testing" *Transl. Lung Cancer Res*
94. Sorber, Zwaenepoel, Deschoolmeester et al. (2017) "A comparison of cell-free DNA isolation kits: Isolation and quantification of cell-free DNA in plasma" *J. Mol. Diagn*
95. Diefenbach, Lee, Kefford et al. (2018) "Evaluation of commercial kits for purification of circulating free DNA" *Cancer Genet*
96. Bronkhorst, Ungerer, Holdenrieder (2020) "Comparison of methods for the isolation of cell-free DNA from cell culture supernatant" *Tumour Biol*
97. Lee, Cho, Kim et al. (2024) "Clinical practice guideline for blood-based circulating tumor DNA assays" *Ann. Lab. Med*
98. Dickinson, Sharma, Agnihotram et al. "Circulating tumor DNA and survival in metastatic breast cancer: A systematic review and meta-analysis" *JAMA Netw*
99. Williams, Forbes, Lund et al. "Validation of ctDNA quality control materials through a precompetitive collaboration of the Foundation for the National Institutes of Health. JCO Precis"
100. Panet, Papakonstantinou, Borrell et al. (2024) "Use of ctDNA in early breast cancer: Analytical validity and clinical potential" *npj Breast Cancer*
101. Deveson, Gong, Lai et al. (2021) "Evaluating the analytical validity of circulating tumor DNA sequencing assays for precision oncology" *Nat. Biotechnol*
102. Mizuno, Sumiyoshi, Okegawa et al. (2021) "Clinical impact of detecting low-frequency variants in cell-free DNA on treatment of castration-resistant prostate cancer" *Clin. Cancer Res*
103. Hamid, Gray, Shaw et al. (2019) "Compound genomic alterations of TP53, PTEN, and RB1 tumor suppressors in localized and metastatic prostate cancer" *Eur. Urol*
104. Chen, Shi, Choi et al. (2023) "Genomic alterations in neuroendocrine prostate cancer: A systematic review and meta-analysis" *BJUI Compass*
105. Aparicio, Shen, Tapia et al. (2016) "Combined tumor suppressor defects characterize clinically defined aggressive variant prostate cancers" *Clin. Cancer Res*
106. Korpal, Korn, Gao et al. (2013) "An F876L mutation in androgen receptor confers genetic and phenotypic resistance to MDV3100 (enzalutamide)" *Cancer Discov*
107. Abida, Armenia, Gopalan et al. (2017) "Prospective genomic profiling of prostate cancer across disease states reveals germline and somatic alterations that may affect clinical decision making. JCO Precis" *Oncol*
108. Armenia, Wankowicz, Liu et al. (2018) "The long tail of oncogenic drivers in prostate cancer" *Nat. Genet*
109. "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=PMC12846408&blobtype=pdf | # Genetic Diversity of the Hepatitis C Virus Among Patients with HIV in EECA Countries
Vasiliy Ekushov, Maksim Halikov, Irina Osipova, Alexei Totmenin, Ludmila Gotfrid, Vardan Arzakanyan, Siranush Martoyan, Kristine Lalayan, Tamara Hovsepyan, Lilit Petrosyan, Susan Muradyan, Hermine Hovakimyan, Aibek Bekbolotov, Elmira Narmatova, Aida Karagulova, Kunduz Momushova, Aikanysh Djusupbekova, Baarinisa Iskanova, Aida Mamirbaeva, Ulukbek Motorov, Vitalla-Victoria Minikhanova, Sergey Skudarnov, Tatyana Ostapova, Alexander Agafonov, Natalya Gashnikova
## Abstract
Against the backdrop of active efforts to combat HCV worldwide with the help of DAAs, knowledge of the genetic diversity of HCV in the general population and in groups most at risk of infection is becoming increasingly important. The aim of this study was to characterize the molecular genetic diversity of HCV among individuals with HIV in Armenia, Kyrgyzstan and the Krasnoyarsk Krai region of Russia. The study included residents of Armenia (n = 73), Kyrgyzstan (n = 180) and the Krasnoyarsk Territory (n = 141) with HIV/HCV co-infection who were under observation at AIDS centers in these countries, collected between 2021 and 2023. The Core/E1 gene fragments obtained were analyzed using the maximum likelihood method to create a phylogenetic tree. HCV subtype 3a was dominant in Armenia (56.2%) and Kyrgyzstan (51.4%). The circulation of HCV subtype 4a was detected for the first time in Armenia, while the spread of HCV genotype 2, represented by three different subtypes, was documented in Kyrgyzstan. The genetic diversity of HCV in Krasnoyarsk Krai is consistent with the findings of previous Russian studies. Phylogenetic analysis revealed the formation of HCV clusters with a high level of bootstrap support, suggesting shared transmission routes, predominantly among PWID. This suggests that there are common routes of HCV transmission between and within countries.
## 1. Introduction
The hepatitis C virus (HCV) is a leading cause of chronic liver disease worldwide. HCV infection can manifest as either an acute or chronic condition and may progress to liver cirrhosis and hepatocellular carcinoma (HCC) [1].
HCV is classified within the genus Hepacivirus of the Flaviviridae family. Its genome consists of a single-stranded, non-segmented, positive-sense RNA molecule approximately 9600 nucleotides in length, featuring a single long open reading frame that encodes a polyprotein of around 3000 amino acids. The gene order is as follows: Core/E1/E2/p7/NS2/NS3/NS4A/NS4B/NS5A/NS5B [2][3][4].
At its seventy-fifth session, the World Health Assembly approved a plan to combat viral hepatitis for the period 2022-2030. The WHO has set four targets for eliminating HCV as a public health threat by 2030, compared to baseline levels in 2015: diagnosing 90% of the population who have HCV, treating 80% of the population, reducing new HCV infections by 80%, and reducing HCV mortality by 65% [5]. However, the implementation of the WHO strategy is hindered by the extremely limited data on the prevalence, genetic diversity and level of drug resistance of HCV in EECA countries.
The prevalence of HCV infection in the WHO European Region varies significantly. In Northern, Western and Central European countries, HCV prevalence reaches 0.5%, whereas in many countries of Eastern Europe and Central Asia (EECA), it is as high as 5% [1].
Key populations play a pivotal role in the current epidemic. PWID represent the highest-risk group, exhibiting a high rate of HIV/HCV co-infection. Nearly half of HCV infections arise from unsafe injection practices among PWID [6]. Beyond parenteral transmission, substance use (not always injecting) may contribute to high-risk sexual behavior, increasing the likelihood of sexual HCV transmission. PWID are not only affected by the epidemic; they are also key indicators of active transmission. Studying this cohort offers a "sentinel lens" into the frontline of the epidemic.
Immunocompromised individuals are particularly vulnerable. Men who have sex with men (MSM) are another high-risk group, where widespread use of HIV-1 pre-exposure prophylaxis (PrEP) has been associated with rising HCV incidence. HCV prevalence among MSM ranges from 4% to 8% [7]. Notably, compared to MSM and individuals with sexually transmitted HCV, PWID are 2,5 times less likely to achieve a sustained virological response at 12 weeks (SVR 12 ) [8].
The high mobility of EECA countries' populations, as represented by labor migrants, also influences the overall nature of the epidemic and could result in the cross-border transmission of HCV. A systematic review of HCV prevalence among migrants worldwide showed that the prevalence of anti-HCV antibodies among migrants from EECA countries was around 2-3% [9].
In EECA countries, subtypes 1b and 3a are the most prevalent. The widespread circulation of subtype 3a in former Soviet Union states is linked to the expansion of injecting drug use in the 1960s and 1990s, whereas subtype 1b likely became established in the region due to historical blood transfusion practices [10,11]. Recent and comprehensive molecular epidemiology data for several EECA nations, including Armenia and Kyrgyzstan, remain scarce. This gap hinders the design of tailored prevention and treatment programs and limits understanding of cross-border transmission [10].
Treatment with direct-acting antivirals (DAAs) provides a key method of controlling HCV transmission. DAAs specifically target HCV proteins involved in viral replication, achieving SVR rates exceeding 90% [12]. Epidemic HCV subtypes served as in vitro models during DAA clinical trials, leading to the development of "pan-genotypic" DAAs effective against these subtypes [13]. However, treatment regimen selection and duration may be influenced by genotype/subtype (particularly for subtypes 1a, 1b and endemic variants), the presence of recombinant forms, cirrhosis and prior treatment failure [14,15]. Thus, HCV genotyping/subtyping is recommended before initiating DAA-based therapy [14,15]. Information on circulating HCV variants can also be used to track the cross-border transmission of the virus.
At present, a heterogeneous population of HCV circulates worldwide. According to the international nomenclature established by the ICTV, HCV is classified into 8 genotypes and 93 subtypes [16]. Additionally, there are reports of novel viruses not belonging to any previously described genotype/subtype, as well as recombinant viruses (e.g., 2k/1b) [16,17]. The distribution of genotypes/subtypes varies geographically. The most prevalent are subtypes 1a, 1b, 2a and 3a. These are termed "epidemic subtypes" and spread globally prior to the discovery of HCV in 1989 [10,13]. Other globally disseminated epidemic subtypes include 4a, 4d and 6a 13]. For many countries, data on HCV circulation is only available for the most common subtypes (1a, 1b and 3a), while the presence of other subtypes is often summarized under separate genotypes (GT2, GT4 and GT6) that are circulating in these countries. This can have a significant impact on HCV elimination strategies. HCV transmission primarily occurs through blood transfusions and/or blood products, iatrogenic procedures and injecting drug use [4,10,18]. The remaining HCV subtypes are considered "endemic," with circulation restricted to specific geographic regions.
Detailed characterization of HCV subtypes within individual countries-as well as among key populations-could inform optimized strategies for HCV prevention, diagnosis and treatment under national viral hepatitis (VH) elimination programs aligned with the WHO's goal of eliminating VH by 2030 [5].
This study aimed to characterize the molecular genetic diversity of HCV by analyzing Core/E1 gene fragments among people living with HIV in Armenia and Kyrgyzstan, as well as residents of Russia's Krasnoyarsk Krai.
## 2. Materials and Methods
## 2.1. Sample Source
The genetic diversity of HCV in the countries studied was assessed using a cohort observational study design. The patient sample was randomly selected from people living with HIV who had been diagnosed with HCV and regularly visited the AIDS Center for follow-up care and ART medication. This study aims to describe the genetic diversity of HCV circulating among individuals with HIV/HCV co-infection.
Inclusion criteria were individuals with HIV/HCV coinfection, aged 18 years or older, who are legally capable of providing informed consent to participate in the study.
Exclusion criteria were a lack of informed consent and the patient's inability to understand the meaning and content of the project and answer questions.
The consecutive sample of patients was formed when they visited the relevant health center in their country.
The study cohort for assessing HCV genetic diversity comprised individuals with HIV/HCV co-infection from Armenia (n = 73), Kyrgyzstan (n = 179) and Russia's Krasnoyarsk Krai (n = 141). AIDS Prevention and Control Centers in the respective countries and regions collected peripheral blood samples and clinical-epidemiological data from patients with HIV/HCV between 2021 and 2023. The Ethics Committee in each participating country approved the study protocols. All participants provided written informed consent, ensuring anonymized data usage in compliance with ethical guidelines. Clinical and epidemiological data included age, sex and probable HIV transmission route. Qualified specialists from the AIDS Center performed data collection after obtaining informed consent from the patient. Epidemiological data was collected through questionnaires that study participants completed when they visited a health center in their country. The question-naire data was anonymized, and each patient was assigned a unique identification number. Patients were given the option to skip any questions in the questionnaire. Information about HIV infection (route of HIV transmission) was obtained during the standardized interview with the patient during the visit to the AIDS center.
## 2.2. HCV RNA Extraction, Amplification and Sequencing
HCV RNA was extracted from plasma in two stages using RIBO-zol-C and RIBOprep kits (phenol-chloroform extraction method) (AmpliSens, InterLabService, Moscow, Russia), according to the manufacturer's instructions (volume of 100 µL of plasma was used for extraction).
The two-step nested PCR with virus-specific primers was employed to amplify the target Core/E1 region (~1000 bp).
To obtain Core/E1 gene fragments, we used the BioMaster OT-PCR-Extra kit (Biolabmix, Novosibirsk, Russia) with thermostable reverse transcriptase (RNAscribe RT (genetically modified reverse transcriptase (RT) of mouse leukemia virus (M-MuLV)), HS-Taq DNA polymerase and Pfu DNA polymerase for the first round of PCR.
For the second round of PCR, we used the BioMaster HS-Taq PCR kit (Biolabmix, Novosibirsk, Russia), which contains highly process-intensive recombinant Taq DNA polymerase.
The reaction mixture was prepared according to the instructions included in the kit. Control samples (both negative and positive) that were prepared during the RNA extraction stage were used to monitor the quality of the reaction.
We estimated the length of the PCR products and the melting temperature of the primers using Oligo v.7 software (Molecular Biology Insights, Cascade, CO, USA). The reaction was performed in a T100 ThermalCycler (Bio-Rad, Hercules, CA, USA) and amplified under specified conditions (Supplementary Tables S1-S3).
For PCR product detection, horizontal electrophoresis was performed in 1% agarose gel with ethidium bromide. The «Step 50 plus» DNA marker (Biolabmix, Novosibirsk, Russia) was used for visual assessment of the PCR product. DNA fragments were visualized using a transilluminator (VilberLourmat, Collégien, France) with ultraviolet light at a wavelength of 312 nm.
Subsequent sequencing of HCV DNA amplicons was performed using the BigDye Terminator™ v3.1 kit on an Applied Biosystems 3500 Series Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). The sequencing reaction was prepared following the current instructions.
The obtained data were analyzed using Sequencher 4.1 software (Gene Codes Corporation, Ann Arbor, MI, USA). The quality of each sequence was assessed by manually reviewing the chromatogram.
## 2.3. Genotyping
HCV RNA was extracted from plasma using phenol-chloroform extraction. A two-step nested PCR with virus-specific primers was employed to amplify the target Core/E1 region (~1000 bp). Subsequent sequencing of HCV DNA amplicons was performed using the BigDye Terminator™ v3.1 kit on an Applied Biosystems 3130xl Genetic Analyzer (Applied Biosystems, Foster City, CA, USA).
The obtained HCV sequences were aligned with reference sequences of different subtypes and recombinant forms from the International Committee on Taxonomy of Viruses (ICTV) database using MEGA11 and AliView v1.30 [16,19,20]. Multiple sequence alignment was conducted with MAFFT v7.526 (RIMD) under default parameters [21].
For identification of closely related HCV strains, nucleotide sequences were analyzed using BLAST against the GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi (accessed on 5 March 2025)). Phylogenetic analysis of the Core/E1 gene fragment (positions 368-1257 relative to H77) was performed via the maximum-likelihood method in IQ-TREE v1.6.12 [22], employing the GTR + I + G substitution model with 1000 bootstrap replicates to assess topology. The substitution model was chosen with IQ-TREE v1.6.12 (subsection: model selection). The resulting phylogenetic tree was visualized using iTOL v7 [23].
## 2.4. Data Availability
The sequences reported in this study are deposited in GenBank with the following accession numbers: PV754724 to PV754837 and OQ784306 to OQ784567.
## 3. Results
A total of 394 patients with HIV/hepatitis C virus (HCV) co-infection were enrolled in the study. The average patient age was 43 years, and 65% of patients were male. Liver cirrhosis was present in only a small percentage of the patients examined (2.0%; n = 8), which may be due to a lack of proper diagnosis. The proportion of migrant workers in the sample, especially from Kyrgyzstan, was lower than expected. This may be due to patients' reluctance to share information about their time abroad.
Subtype 3a was the predominant type in all of the countries that were studied, except for Russia (Krasnoyarsk Krai), where subtypes 3a and 1b were equally prevalent. Armenia had the highest rate of subtype 3a.
As expected, the percentage of PWID in the study sample was high (56.1%; n = 221). Despite the widespread prevalence of subtype 3a in the PWID group (this subtype has become widespread in this risk group in the countries of the former USSR), subtype 1b was significantly more prevalent among PWID than expected. In Armenia, all patients with subtype 1b were PWID. In the Krasnoyarsk region of Russia, 81.0% of patients infected HCV with subtype 1b were PWID, compared to 71.0% for subtype 3a. In Kyrgyzstan, the prevalence of both subtypes among PWID was nearly equal: 3a (70.7%) and 1b (66.0%). This is the first time that subtype 4a has been registered in Armenia. Previously, there had been no reports of this subtype circulating in the country. For the first time in the Kyrgyz Republic, GT2 was detected in circulation, represented by three subtypes: 2a, 2c and 2k.
## 3.1. Armenia
The mean age of patients was 44 years, with males predominating in the sample (90.4%) and females comprising only 9.6%. A history of injecting drug use (IDU) was reported by 52.1% of patients, all of whom were male; the specific types of drugs used were not specified. Labor migrants accounted for 20.5% of patients, with Russia (n = 12), Ukraine (n = 1) and Turkey (n = 1) being the main host countries; one labor migrant (LM) did not specify their migration destination. Individuals self-identifying as MSM comprised 4.1% of those tested, while one patient (1.4%) identified as transgender. Some 6.8% did not belong to any of the aforementioned key populations, and for 15.1% of patients, data on risk behaviors were unavailable (Table 1).
Among the study cohort, liver cirrhosis was reported in 8.2% of cases, with no cases of HCC observed. Of the patients with liver cirrhosis, three were infected with HCV subtype 3a, two with 1b and one with 1a. 1.4% (n = 1) --Phylogenetic analysis of the Core/E1 gene fragment revealed six HCV subtypes: 1a (15.1%; n = 11), 1b (20.5%; n = 15), 2a (1.4%; n = 1), 2k (5.5%; n = 4), 3a (56.2%; n = 41) and 4a (1.4%; n = 1) (Table 1).
Approximately half of all infections across subtypes occurred among PWID and their sexual partners (58.9%, n = 43). The highest proportion of PWID was observed for the epidemic subtype 3a (61.0%, n = 25). HCV subtypes 2k (n = 4), 1a (n = 11) and 1b (n = 15) were detected in both PWID and people living with HIV transmitted through sexual contact, with roughly equal distribution. Subtypes 1a and 4a were found exclusively in the PWID group.
Phylogenetic tree analysis revealed that HCV sequences from Armenia formed two distinct clusters (No. 9 and 11), suggesting a common transmission route among individuals with HIV/HCV co-infection (Figure 1).
One cluster (No. 9; bootstrap support value of 92) comprised HCV subtype 3a sequences obtained from ten patients (aged 41-61 years) at the Republican Center for Infectious Diseases of Armenia. Six patients were PWID, while two were presumably infected with HIV-1 through sexual contact (available data were insufficient to determine either the route of HCV transmission or the duration of infection), with one confirmed to have had a sexual partner who was a PWID. The cluster also contained HCV sequences from three male patients from Kyrgyzstan-two PWID and one presumably infected through sexual contact.
A second HCV cluster (No. 11; bootstrap support value of 98) contained six sequences of subtype 1a from the study cohort in Armenia. All viral variants in this group were isolated from men aged 23-48 years, containing two PWID, two MSM, one presumably infected with HIV-1 through sexual contact, and one with undetermined HIV-1 transmission route and/or risk group affiliation. The cluster also contained two HCV sequences from two male PWID (aged 34 and 39 years) from Krasnoyarsk. The only case of HCV subtype 4a infection was identified in a 57-year-old male PWID residing in the Shirak region with no history of international travel.
A 32-year-old male PWID from Yerevan was found to be infected with HCV subtype 2a. Prior to the study, the patient had been diagnosed via qPCR with HCV subtype 3a but had not received treatment, with the estimated year of infection as 2009 (based on anamnestic data).
An additional HCV sample, previously subtyped as 3a (from a 48-year-old male PWID with an estimated infection date of 2019, based on anamnestic data), was reclassified as subtype 1a in this study.
## 3.2. Kyrgyzstan
HCV genotyping was performed among two groups. The first group comprised 112 consenting patients from the Republican Clinical Hospital for Infectious Diseases (Bishkek) and the Regional AIDS Center (Osh). The mean age was 45 years, with male predominance (83%) versus females (17%). PWID accounted for 67.9% of cases (93.4% male, 6.6% female); the types of drugs used were not specified. Two tested patients (1.9%) self-identified as MSM, one woman reported having sexual partners of both sexes, and 1.8% declined to answer. One patient was a labor migrant in Russia (Table 1).
Within this cohort, only one case (0.9%) of liver cirrhosis was reported (patient with HCV subtype 3a), with no cases of HCC identified.
The second group consisted of 68 completely anonymous blood samples collected at a diagnostic laboratory. Phylogenetic analysis of HCV Core/E1 fragments from 180 sequences revealed six subtypes: 1a (2.8%; n = 5), 1b (41.1%; n = 74), 2a (2.2%; n = 4), 2c (1.7%; n = 3), 2k (1.1%; n = 2) and 3a (51.1%; n = 92) (Table 1).
The highest proportion of subtype 3a infections occurred among PWID (70.7%, n = 41), while 66.0% of patients with subtype 1b were also PWID. All patients with subtypes 1a and 2c were PWID, as was one of four patients with subtype 2a.
Phylogenetic analysis identified several statistically significant HCV sequence clusters (bootstrap support value > 90) which probably suggest common HIV/HCV transmission routes (Figure 1). Most clusters contained sequences from Kyrgyz PWID. For subtype 1b, three clusters (No. 14, 20 and 22; minimum bootstrap support 96) were identified: one predominantly comprising non-PWID patients, while the other two contained sequences from both PWID and patients not belonging to risk groups. Two clusters contained sequences from Russian (Krasnoyarsk Krai) residents.
Among three subtype 3a clusters (No. 4, 5 and 8; minimum bootstrap support 90), sequences from PWID predominated (except one cluster with predominantly missing epidemiological data). Sequences from Osh patients (mainly PWID) formed a distinct regional cluster (support 94). Only one cluster contained sequences from Russia (the Krasnoyarsk Krai) and Armenia (labor migrant with unspecified host country).
Among HCV subtype 1a sequences (No. 13), a cluster with bootstrap support value of 100 comprised four samples from Kyrgyzstan residents, with anamnestic data available only for one 40-year-old male PWID.
The single HCV subtype 2c cluster (No. 3; bootstrap support value of 100) contained viruses from three Kyrgyz patients-a 64-year-old male from the Chui region and two residents with no available anamnestic data.
Three HCV subtype 2a (No. 3) sequences from the Chui region residents-a 61-yearold male PWID, a 30-year-old woman presumably with HIV-1 transmitted through sexual contact, and one patient with missing anamnestic data-formed a cluster with a bootstrap support value of 94.
HCV subtype 2k was detected in two women with HIV women aged 40 and 45 years from Bishkek, both presumably infected through sexual contact, with one woman reporting sexual partners of both sexes.
## 3.3. Krasnoyarsk Krai
The mean age of patients included in the study was 41 years, with males predominating (68.8%) and females comprising 31.2% of samples. PWID accounted for 73.8% of cases, including 73.1% (n = 76) males and 26.9% (n = 28) females. One individual (0.7%) selfidentified as a MSM and was also a PWID. Some 24.1% (n = 34) of patients did not belong to any risk group, with presumed HIV/HCV transmission occurring through heterosexual contact with individuals with HIV/HCV (Table 1). Risk group data were unavailable for three individuals (2.1%). The cohort contained only one reported case of liver cirrhosis (patient with HCV subtype 3a), with no cases of HCC observed (though six non-HCC malignancies were recorded).
Phylogenetic analysis of Core/E1 fragments identified five HCV subtypes: 1a (8.5%; n = 12), 1b (44.7%; n = 63), 2a (2.1%; n = 3), 2k (0.7%; n = 1) and 3a (44.0%; n = 62) (Table 1).
Most infections with epidemic subtypes occurred among PWID, with particularly high proportions for subtypes 1b (81.0%, n = 51), 3a (71.0%, n = 44) and 1a (58.3%, n = 7). Subtype 2k was exclusively found in PWID, while one of three (33.3%) subtype 2a cases occurred in a PWID.
The majority of Krasnoyarsk-derived HCV clusters comprised viruses from individuals sharing common HIV transmission routes, predominantly PWID. Six subtype 1b clusters (No. 15, 16, 17, 18, 19 and 21; minimum bootstrap support value of 96) were identified, with only one cluster dominated by non-PWID sequences. Four clusters contained HCV sequences from Armenian residents, containing both labor migrants (one being a Russian citizen from Tyumen) and non-migrants. One additional cluster contained two sequences from Kyrgyzstan residents (with unavailable anamnestic data).
Three subtype 3a clusters (No. 6, 7 and 10; minimum bootstrap support value of 91) consisted primarily of PWID-derived sequences and contained no international variants (Figure 1).
A distinct subtype 1a cluster (No. 12; bootstrap support value of 100) contained nine sequences from seven males and two females (aged 37-40 years), containing five male PWID and four individuals (two males, two females) presumably infected through sexual contact with PWID. This cluster also contained one sequence from a 47-year-old Armenian male PWID. Two subtype 2a (No. 2) sequences formed a cluster, comprising one 41-year-old male PWID and one 46-year-old female infected through sexual contact. This cluster (bootstrap support value of 95) additionally contained a sequence from a 32-year-old Armenian male reportedly infected in Russia.
The sole Krasnoyarsk-derived subtype 2k sequence originated from a 46-year-old male PWID.
## 4. Discussion
Research on the genetic diversity of HCV strains circulating in Armenia has previously been limited. A 2016 study by Petruzziello et al. reported HCV genotype 3 prevalence at 37% [24], with genotypes 1 and 2 also detected. However, the study lacked fully sequenced HCV data, suggesting genotyping likely relied on qPCR. In a 2025 analysis by Mustafa et al., 18 publicly available sequences (NS3, NS5A and NS5B genes) were examined, predominantly subtype 1b (55.6%) and 3a (11.1%) [25]. This small sample size precludes robust assessment of subtype distribution across Armenia, compounded by the inclusion of sequences from Armenian migrants.
Our genotyping of 73 HCV variants from Armenia reveals striking disparities. Among patients with HIV/HCV co-infection, subtype 3a predominated (56.2%), while subtype 1b accounted for only 20.5%. Notably, ours is the first study to document subtype 4a infections in Armenia. As the patient infected with HCV subtype 4a had not traveled abroad, it is likely that he was infected domestically. This indicates that this subtype is spreading within the country.
Conventional genotyping employs qPCR targeting 5 ′ -UTR and Core regions. Discrepancies emerged between our results and prior classifications by Armenia's National Center for Infectious Diseases for two patients. One, initially genotyped as 3a, was reclassified as 2a based on our Core/E1 sequence analysis; another, also previously labeled 3a, was https://doi.org/10.3390/v18010016 identified as 1a. For the study, patients diagnosed with HCV had repeat blood samples taken. The first patient was a person who injects drugs (PWID), while the second had no known risk factors. Among PWID, reinfection with HIV or HCV-potentially altering viral dominance or generating recombinant strains-is not uncommon. However, full-genome sequencing is required to confirm recombination. Misclassification via qPCR or human error during diagnostic procedures cannot be ruled out. No other discrepancies in HCV genotyping were found. The test system correctly identified rare HCV genotypes, such as GT4 and GT2, in Armenia. However, qPCR does not involve classifying these rare genotypes into subtypes.
Two 2022 studies on individuals with HIV/HCV co-infection in Kyrgyzstan reported the following circulating HCV subtypes: 1a (2.6%), 1b (52.6%) and 3a (44.8%) [26]. Another study, analyzing publicly available HCV sequences, documented a different subtype distribution in Kyrgyzstan's HCV population: 3a (53.7%) and 1b (44.7%) [25]. Our genotyping data show minor deviations from these earlier findings. Among the studied cohort with HIV/HCV co-infection, subtype 3a similarly predominated. Notably, we report for the first time the circulation of HCV genotype 2 in Kyrgyzstan, represented by three distinct subtypes. The presence of several GT2 subtypes in Kyrgyzstan suggests that there have been several independent cases of transmission.
Our data on HCV subtypes circulating among patients with HIV/HCV co-infection in the Krasnoyarsk Krai align broadly with prior Russian studies encompassing the general population with HCV, though with slight variations in the 3a/1b ratio. The proportion of subtype 3a in our cohort was marginally higher (40.2-42.1%) than values in the literature, while subtype 1b was somewhat lower (46.9%) [25,27]. These modest discrepancies may reflect regional epidemic dynamics in Krasnoyarsk or the specific profile of patients with HIV/HCV co-infection.
Collectively, our findings highlight the greatest HCV diversity in Armenia and Kyrgyzstan, where subtypes 4a and 2c were respectively detected. Although these genetic variants were rare, their identification in people living with HIV (PLWH) with no history of travel suggests localized transmission networks for these HCV strains. Our study also revealed divergent subtype distributions: in Russia and Kyrgyzstan, subtypes 1b and 3a dominate among PLWH-with near-equal prevalence in Russia-while 3a exceeds 1b by 10% in Kyrgyzstan. In Armenia, HCV 3a predominates (56.2%), alongside significant contributions from 1b (20.5%) and 1a (15.1%).
Patients with HIV/HCV co-infection reflect an "active epidemic", in which transmission continues among IDUs (subtype 3a) and immunosuppressed patients. Earlier studies or studies of the general population reflect a "historical epidemic", in which transmission primarily occurred through blood transfusions (subtype 1b). Our research suggests that the nature of the epidemic is changing within the PWID cohort. Specifically, there has been an increase in the number of PINs infected with HCV subtype 1b across all the countries studied.
Labor migrants contribute substantially to HCV transmission dynamics, potentially acquiring infection abroad or introducing strains into host countries. Significant migratory flows connect Russia, Kyrgyzstan and Armenia, with prolonged workforce mobility between these states. Since 2022, a marked increase in Russian migrants to Kyrgyzstan and Armenia has been observed, suggesting potential epidemiological linkages in HCV spread across these regions.
Six clusters were identified, combining sequences from Russia and Armenia for subtypes 1a, 1b and 2a. Two of the six clusters (1b No. 15 and 2a No. 2) had been confirmed to include patients from Armenia who had visited Russia. Reliable conclusions cannot be drawn about the cross-border transmission of HCV variants for the remaining clusters.
However, this possibility cannot be ruled out, as patients could have been infected with an "imported" HCV variant within the country.
Similarly, five clusters were identified comprising sequences from Kyrgyzstan and the Krasnoyarsk Territory of Russia. However, the limited epidemiological data available for patients from Kyrgyzstan means we cannot confirm whether virus variants have spread from Kyrgyzstan to Russia, or whether citizens of Kyrgyzstan have become infected abroad.
The presence of a large number of independent clusters, each comprising sequences from different countries, may suggest the existence of a complex, interconnected meta-epidemic.
Since 2022, there has been a substantial increase in the number of Russians migrating to Kyrgyzstan and Armenia, suggesting potential epidemiological links in HCV transmission between these countries. Phylogenetic analysis revealed statistically significant regional clusters of HCV subtypes 1a, 1b and 3a with presumably shared HIV-1 transmission routes (though the precise HCV infection pathways remain uncertain due to a lack of seroconversion data). The overwhelming majority of individuals in these clusters were PWID. While definitive data on HCV infection timing and transmission modes are unavailable, these clusters likely represent HCV transmission through either injection drug use or sexual contact with PWID in the context of HIV-1-induced immunodeficiency.
The research has some limitations: 1. The sequential sampling design does not allow for a full characterization of the epidemic in the group with HIV/HCV infection. 2. There is a lack of information regarding the route and date of HCV infection in all countries. Many details in the survey (labor migration, route of infection) could have been concealed by patients despite the complete anonymity of the data. The disease was diagnosed primarily when patients visited AIDS centers, and, to a lesser extent, during annual medical exams. Determining the date of infection and the possible route of transmission is impossible in this regard. 3. For part of the Kyrgyzstani sample (68 patients), epidemiological data was unavailable. The samples were obtained from anonymous patients who only consented to participate in the HCV study on the conditions that they remained anonymous and were not required to answer questionnaires. 4. Analysis of the Core/E1 region does not allow conclusions to be drawn about the presence of recombinant viruses.
## 5. Conclusions
The considerable genetic diversity and heterogeneous distribution of HCV variants necessitate further investigation into how specific genotypes/subtypes influence epidemic dynamics. Emerging evidence indicates reduced effectiveness of DAA therapies against endemic subtypes, which may potentially compromise HCV elimination efforts [12]. Targeted interventions focusing on key populations-PWID, MSM and sex workers-are crucial for effective HCV control. Comprehensive understanding of genotype/subtype distribution patterns should inform the development of national HCV prevention and treatment programs.
The EECA countries are experiencing a highly dynamic and interconnected HCV epidemic, characterized by active transmission among PWID. The high genetic diversity of HCV circulating in the region means that systematic molecular monitoring is not only recommended, but also necessary for public health. When developing and implementing HCV elimination programs, the constantly changing nature of the epidemic must be taken into account. This includes factors such as an increase in the proportion of patients infected with subtype 1b and active population migration.
This study was conducted among key populations at high risk of HCV infection and reflects epidemic patterns within vulnerable groups where PWID represent the predominant transmission drivers. The genetic diversity of HCV strains circulating in the general population may exhibit distinct characteristics.
The present work was undertaken as part of regional initiatives to reduce HCV transmission among people living with HIV in Eastern Europe and Central Asia. The findings provide critical evidence for developing long-term strategies to combat HCV effectively in the most vulnerable populations across participating countries, where the burden of HCV infection remains disproportionately high.
## References
1. Advice (2016) "Systematic Review on Hepatitis B and C Prevalence in the EU/EEA"
2. Dubuisson (2007) "Hepatitis C virus proteins" *World J. Gastroenterol*
3. Schulze Zur Wiesch, Schmitz, Borowski et al. (2003) "The proteins of the Hepatitis C virus: Their features and interactions with intracellular protein phosphorylation" *Arch. Virol*
4. Choo, Kuo, Weiner et al. (1989) "Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome" *Science*
5. (2024) "Action for Access in Low-and Middle-Income Countries; World Health Organization"
6. Trickey, Fraser, Lim et al. (2019) "The contribution of injection drug use to hepatitis C virus transmission globally, regionally, and at country level: A modelling study" *Lancet Gastroenterol. Hepatol*
7. Vanhommerig, Lambers, Schinkel et al. "Risk factors for sexual transmission of hepatitis C virus among human immunodeficiency virus-infected men who have sex with men: A case-control study"
8. Phillips, Bradshaw, Packer et al. (2025) "Direct-acting antiviral treatment outcomes in people infected with endemic compared to epidemic hepatitis C virus subtypes in England" *J. Infect*
9. Greenaway, Thu Ma, Kloda et al. (2015) "The seroprevalence of hepatitis C antibodies in immigrants and refugees from intermediate and high endemic countries: A systematic review and meta-analysis" *PLoS ONE*
10. Messina, Humphreys, Flaxman et al. (2015) "Global distribution and prevalence of hepatitis C virus genotypes" *Hepatology*
11. Hatzakis (2021) "Hepatitis C: Epidemiology, Prevention and Elimination"
12. (2018) "Guidelines for the Screening Care and Treatment of Persons with Chronic Hepatitis C Infection"
13. Shah, Ahovegbe, Niebel et al. (2021) "Non-epidemic HCV genotypes in low-and middle-income countries and the risk of resistance to current direct-acting antiviral regimens" *J. Hepatol*
14. Ghany, Morgan (2020) "AASLD-IDSA. Hepatitis C guidance 2019 update: American Association for the Study of Liver Diseases-Infectious Diseases Society of America recommendations for testing, managing, and treating hepatitis C virus infection" *Hepatology*
15. Pawlotsky, Negro, Aghemo et al. (2018) "EASL recommendations on treatment of hepatitis C 2018" *J. Hepatol*
16. "International Committee on Taxonomy of Viruses: ICTV. Classification and Genotype/Subtype Assignments of Hepatitis C Virus"
17. Donald, Jens, Carla et al. (2019) "ICTV: London, UK"
18. Mbisa, Lapp, Bibby et al. (2024) "Identification of 2 novel subtypes of hepatitis C virus genotype 8 and a potential new genotype successfully treated with direct acting antivirals" *J. Infect. Dis*
19. Pybus, Cochrane, Holmes et al. (2005) "The hepatitis C virus epidemic among injecting drug users" *Infect. Genet. Evol*
20. Koichiro, Stecher, Kumar (2021) "MEGA11: Molecular evolutionary genetics analysis version 11" *Mol. Biol. Evol*
21. Larsson (2014) "AliView: A fast and lightweight alignment viewer and editor for large datasets" *Bioinformatics*
22. (2026) *Viruses*
23. Kazutaka, Rozewicki, Yamada (2019) "MAFFT online service: Multiple sequence alignment, interactive sequence choice and visualization" *Brief. Bioinform*
24. Trifinopoulos, Nguyen, Haeseler et al. (2016) "W-IQ-TREE: A fast online phylogenetic tool for maximum likelihood analysis" *Nucleic Acids Res*
25. Letunic, Bork (2024) "Interactive Tree of Life (iTOL) v6: Recent updates to the phylogenetic tree display and annotation tool" *Nucleic Acids Res*
26. Petruzziello, Marigliano, Loquercio et al. (2016) "Global epidemiology of hepatitis C virus infection: An up-date of the distribution and circulation of hepatitis C virus genotypes" *World J. Gastroenterol*
27. Mustafa, Davlidova, Abidi et al. (2025) "Prevalence of resistance-associated substitutions (RAS) in hepatitis C virus in the Former Soviet Union countries" *BMJ Open Gastroenterol*
28. Kartashov, Svirin, Bekbolotov et al. (2023) "Analysis of resistance-associated substitutions in hepatitis C virus sequences from Kyrgyzstan" *Probl. Virol*
29. Kashnikova, Bystrova, Polyanina et al. "Genetic Monitoring as a Component of Hepatitis C Surveillance" *Public Health Life Environ.-PHLE*
30. "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=PMC12665341&blobtype=pdf | # Molecular Characterization of Rift Valley Fever Virus From the 2025 Outbreak in Northern Senegal Reveals Lineage H Persistence and Key Polymerase Mutations
Moussa Moïse Diagne, | Fall, | Sall, | Sow, | Ndeye, Awa Ndiaye, | Gaye, | Mamadou, Sarr Ndao, Aboubacry Gaye, | El, Hadji Ndiaye, Mignane Ndiaye, | Seynabou, Mbaye Ba, Souna Diop, Safiétou Sankhe, | Kane, Seynabou Ndiaye, | Faye, Yoro Sall, Aliou Barry, Ibou Gueye, Marie Dior Ndione, | Diop, | Cisse, Joseph Fitchett, Ibrahima Dia, | Cheikh Loucoubar, | Ndongo, Mawlouth Diallo, | Diallo, | Ibrahima, Soce Fall, Mamadou Ndiaye, | Diawo Diallo, | Sow, | Faye, Moussa Diagne
## Abstract
Rift Valley fever virus (RVFV) is a mosquito-borne phlebovirus that causes severe febrile and hemorrhagic illness in humans. In September 2025, an outbreak in northern Senegal led to 119 confirmed infections and 15 deaths as of October 7, 2025. We performed rapid genomic sequencing to characterize the virus responsible for this epidemic. RNA from RT-qPCR-confirmed samples was sequenced using the Twist Comprehensive Viral Research Panel on an Illumina iSeq 100 platform. Consensus genomes were analyzed with and compared with all complete RVFV genomes in GenBank. Nine genomes were recovered, including five complete tripartite sequences. All clustered within lineage H, sharing > 99% nucleotide identity with Senegalese isolates from 2020 to 2022. Alongside two conservative mutations (R137K and K1111R in S and M segments, respectively), a single nonconservative D11N substitution in the L polymerase may affect replication efficiency, while Gn and Gc epitopes remained conserved. Phylogenetic analyses confirmed strong genetic continuity with earlier West African isolates, indicating local persistence rather than reintroduction. Lineage H persistence in Senegal, combined with polymerase substitutions under purifying selection, suggests subtle viral adaptation that may affect replication. Conserved glycoprotein epitopes indicate maintained vaccine relevance. Sustained genomic surveillance integrated with clinical and ecological monitoring remains essential to anticipate viral evolution and guide Rift Valley fever control.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.
## 1 | Introduction and Outbreak Context
Rift Valley fever virus (RVFV), a mosquito-borne phlebovirus of the family Phenuiviridae, was first described in Kenya in 1931 [1]. Since then, the virus has caused repeated epidemics in Africa and the Arabian Peninsula, including major outbreaks in Egypt (1977), East Africa (1997)(1998)(2006)(2007), Saudi Arabia, and Yemen (2000) [2]. In West Africa, recurrent epidemics have been reported in Mauritania, Niger, Mali, and Senegal [3,4].
Senegal has a well-documented history of RVF outbreaks. The first major outbreak occurred in 1987, at the frontier of Senegal and Mauritania, producing ~1500 human cases with 200 deaths [5]. Additional outbreaks or sporadic cases were reported in 1993-1994, 1998-1999, 2002-2003, 2012, and 2013-2014 but the impact was very minor without death [6]. In recent years, genomic studies have shown that lineage H, previously detected in Southern Africa, became established in Senegal from 2020 onward, replacing older West African lineages [7,8].
In September 2025, an outbreak of RVF was confirmed in the Saint-Louis region of northern Senegal. The first confirmed cases were identified on September 20 at the regional hospital. By September 26, 11 cases, including 4 deaths, had been confirmed. As investigations expanded, the outbreak spread to Matam and Louga regions. In parallel, veterinary surveillance reported an increase in abortions among small ruminants in the Saint-Louis area, consistent with active RVFV circulation in livestock.
The Ministry of Health classified the event as a severe epidemic and activated the Regional Epidemic Management Committee with an Incident Management System to coordinate response. By October 7, surveillance data indicated more than 600 suspected cases, of which 100 were confirmed by RT-qPCR and 19 by IgM serology, giving 119 laboratory-confirmed infections and 15 deaths overall.
Given the severe clinical presentation, the Institut Pasteur de Dakar conducted rapid genomic sequencing of patient samples to assess whether viral genetic changes could explain the outbreak's lethality.
## 2 | Materials and Methods
RNA from RT-qPCR-confirmed samples was sequenced on an Illumina iSeq 100 using the Twist Comprehensive Viral Research Panel hybrid-capture workflow, following the manufacturer's instructions as previously described [8].
For genome assembly, raw FASTQ reads were trimmed with Trimmomatic (v0.39) and L, M, and S segments aligned to references using BWA-MEM (v0.7.17). Consensus sequences were generated with iVar (v1.3.1) using a 50% allele frequency and ≥ 10× depth threshold. BAM files were annotated with LoFreq indelqual, and SNVs/indels called using LoFreq (v2.1.5). Genomes were considered complete when more than 90% of the reference segment length was recovered at a minimum mean depth of 10× across each segment.
Phylogenetic analyses were performed by combining 2025 outbreak genomes with all complete and near-complete RVFV genomes available in GenBank. Alignments were produced in MAFFT, and maximum-likelihood trees reconstructed using IQ-TREE with ModelFinder-selected substitution models and 1000 ultrafast bootstrap/SH-aLRT replicates. Trees were visualized in iTOL [9]. Mutations were examined in AliView [10], distinguishing synonymous from nonsynonymous changes via codon display and confirmed with MEGA (Nei-Gojobori method, codon-based Z-test of selection [11]).
## 3 | Results and Discussion
Nine RVFV genomes spanning the L, M, and S segments were assembled. Coverage exceeded 90% for five complete genomes (S1, S2, S3, S4, S8), while others showed partial coverage (Table 1). For analysis, only sequences with ≥ 15× mean depth were retained, yielding five complete genomes and one additional M segment (S5).
Comparative analysis showed that the 2025 outbreak genomes shared more than 99% nucleotide identity with Senegalese isolates from Fatick (2020) and Matam (2022) [7,8]. Phylogenetic reconstruction confirmed that these strains cluster within the West African branch of lineage H, indicating continued local circulation rather than a novel introduction (Figures 123, for S, M, and L segments, respectively).
Selection pressure analysis showed different evolutionary patterns across the genomes. When compared with the reference strain ZH-548 M12, the outbreak sequences showed clear signs of positive selection (Z ≈ -3.4 to -4; p < 0.01), meaning they accumulated more amino acid changes than expected. In contrast, evidence of purifying selection (Z > 0) was detected among sequences obtained during the 2025 outbreak, suggesting strong functional constraint and limited diversification within this epidemic cluster. Looking at changes over time, comparisons between 2020, 2022, and 2025 genomes suggested moderate adaptive trends, with slightly negative Z values (-1.5 to -1.8) but nonsignificant p values (0.07-0.13). Full results are provided in Tables S1-S3.
Most of the newly identified amino acid substitutions were conservative, including R137K in NSs and K1111R in NSm. Several changes in the L polymerase were shared with strains circulating in Senegal during 2020 and 2022 (Table 2). These substitutions involve residues of similar biochemical properties and are unlikely to significantly alter protein function [12].
A single nonconservative D11N mutation in the L segment, unique to the 2025 outbreak strains and located in the endonuclease domain, may subtly affect cap-snatching activity or fine-tune replication efficiency. The M120T substitution introduces a polar residue in place of a hydrophobic methionine in the N-terminal region, potentially influencing local folding or stability [13]. The NSs protein, the main interferon antagonist [14], carried only one conservative substitution, suggesting preserved function. NSm also retained conservative variation. Importantly, no amino acid changes were observed in the glycoproteins Gn and Gc at known neutralizing epitopes, suggesting that immune recognition and vaccine efficacy should remain unaffected [15].
## 4 | Discussion
Although the outbreak genomes were highly conserved, the presence of a few adaptive polymerase changes suggests ongoing fine-tuning of replication capacity under purifying selection. The persistence of lineage H in Senegal supports endemic maintenance rather than re-emergence through importation. This pattern mirrors earlier findings from Mauritania (2020) and southern Africa, where lineage H was associated with increased replication efficiency and more severe clinical outcomes [6,8].
During the 2020 Mauritania epidemic, which caused 78 confirmed human cases and 25 deaths, NSs gene sequences were identical to those of the 2020 Senegalese isolates from Fatick, yet the epidemiologic outcomes diverged sharply. While Mauritania experienced a severe epidemic with high fatality [16], Senegal reported only sporadic human cases without deaths [7]. This contrast highlights how viral genetics alone may not fully explain outbreak severity and underscores the potential influence of ecological, host-related, and behavioral factors such as patterns of animal handling, slaughtering practices, and exposure to infected livestock, on Rift Valley fever dynamics [17].
Experimental studies have also shown that lineage H replicates more efficiently in vitro than other epidemic lineages (G and C) [8], supporting its apparent fitness advantage and capacity for rapid emergence under favorable ecological conditions. The combination of intrinsic viral fitness and extrinsic environmental factors, including vector density, livestock movement, and underlying population immunity levels, may therefore determine whether lineage H produces localized transmission or large-scale epidemics. Although direct preoutbreak herd immunity data for the Saint-Louis region in 2025 are missing, serological studies from northern Senegal offer meaningful context. In particular, Durand and colleagues reported an IgG seroprevalence of ~15.3% among resident small-ruminant herds (n = 222) in the Saint-Louis/Matam area, indicating a moderate level of natural herd immunity shaped by nomadic herd movements and seasonal vector availability [18]. This incomplete immunity may reduce, but not eliminate, viral amplification potential, thereby enabling the relatively "stable" but subtly evolving lineage H viruses (characterized by mostly conservative amino-acid substitutions) to persist at low levels between outbreaks and re-emerge when ecological, behavioral, or immunological conditions align.
The amino acids changes shared among strains circulating in Senegal in 2020, 2022, and 2025, indicating ongoing but limited evolutionary change within lineage H. Although functional assays were not conducted, the identified nonconservative substitutions occur within domains essential for polymerase activity, suggesting potential impacts on replication fidelity and viral fitness. These changes may fine-tune polymerase conformation or replication efficiency while maintaining overall protein stability, as supported by the predominance of purifying selection across the genome. Importantly, no amino-acid alterations were detected within the neutralizing epitopes of the Gn and Gc glycoproteins, indicating that antigenic recognition and vaccine efficacy are likely preserved.
The specific polymerase substitution identified in the 2025 strains warrants functional characterization to evaluate its impact on replication kinetics, polymerase processivity, and host-virus interactions. Such phenotypic studies-integrating in vitro and in vivo models-will be essential to determine whether these mutations subtly enhance replication efficiency or alter host immune modulation. Linking genomic data with phenotypic and immunological profiling will provide a more complete understanding of how small molecular changes translate into variations in pathogenicity and outbreak potential.
Clinically, the elevated case-fatality rate observed during the 2025 outbreak aligns with prior lineage H epidemics, supporting a possible link between viral genetic background and disease severity [8]. This observation reinforces the need to integrate molecular findings with clinical and epidemiological data to better understand genotype-phenotype relationships in RVFV infections.
Lineage H has previously been associated with more severe outbreaks in Mauritania, South Africa, and Namibia. The 2025 epidemic in Senegal fits this pattern, with lineage H strains showing strong genetic continuity with earlier West African isolates and subtle amino-acid changes in the L polymerase that may fine-tune viral replication. Only shared with strains circulating in Senegal in 2020 and 2022 [7,8].
Together, these findings indicate that lineage H is now endemically established in West Africa, capable of causing recurrent outbreaks with varying severity. Genomic data reveal limited evolutionary change, with mostly conservative and a few nonconservative substitutions that spare the key Gn and Gc epitopes, suggesting that vaccine protection would not be compromised. Continued genomic surveillance-integrated with clinical, functional, and ecological monitoring-will be essential to detect further adaptive changes and to inform diagnostic and vaccine strategies aimed at mitigating future Rift Valley fever epidemics.
## References
1. Wright, Kortekaas, Bowden et al. (2019) "Rift Valley Fever: Biology and Epidemiology" *Journal of General Virology*
2. Tucker, Melocik, Anyamba et al. (2021) "Correction: Reanalysis of the 2000 Rift Valley Fever Outbreak in Southwestern Arabia" *PLoS One*
3. Sow, Faye, Ba (2012) "Rift Valley Fever Outbreak, Southern Mauritania" *Emerging Infectious Diseases*
4. Who, Fao, Oie (2021) "Rift Valley Fever: Epidemiology and Control"
5. Jouan, Guenno, Digoutte et al. (1988) "An RVF Epidemic in Southern Mauritania" *Annales de l'Institut Pasteur*
6. Tinto, Quellec, Cêtre-Sossah et al. (2023) "Rift Valley Fever in West Africa: A Zoonotic Disease With Multiple Socio-Economic Consequences" *One Health*
7. Bob, Barry, Diagne (2021) "Detection of Rift Valley Fever Virus Lineage H From South Africa Through the Syndromic Sentinel Surveillance Network in Senegal" *Open Forum Infectious Diseases*
8. Sene, Sagne, Bob (2024) "Re-Emergence of Rift Valley Fever Virus Lineage H in Senegal in 2022"
9. Letunic, Bork (2016) "Interactive Tree of Life (iTOL) v3: An Online Tool for the Display and Annotation of Phylogenetic and Other Trees" *Nucleic Acids Research*
10. Larsson (2014) "AliView: A Fast and Lightweight Alignment Viewer and Editor for Large Datasets" *Bioinformatics*
11. Nei, Gojobori (1986) "Simple Methods for Estimating the Numbers of Synonymous and Nonsynonymous Nucleotide Substitutions" *Molecular Biology and Evolution*
12. Borrego, Brun (2021) "A Hyper-Attenuated Variant of Rift Valley Fever Virus Generated by a Mutagenic Drug (Favipiravir) Unveils Potential Virulence Markers" *Frontiers in Microbiology*
13. Wang, Hu, Ye (2022) "Structure of Rift Valley Fever Virus RNA-Dependent RNA Polymerase" *Journal of Virology*
14. Ly, Ikegami (2016) "Rift Valley Fever Virus NSs Protein Functions and the Similarity to Other Bunyavirus NSs Proteins" *Virology Journal*
15. Fatima, Ahmad, Alamri (2022) "Discovery of Rift Valley Fever Virus Natural Pan-Inhibitors by Targeting Its Multiple Key Proteins Through Computational Approaches" *Scientific Reports*
16. Barry, Elbara, Bollahi (2020) "Lessons From a One Health Approach"
17. Barry, Metz, Krisztian (2025) "Local Drivers of Rift Valley Fever Outbreaks in Mauritania: A One Health Approach Combining Ecological, Vector, Host and Livestock Movement Data" *PLoS Neglected Tropical Diseases*
18. Durand, Lo Modou, Tran (2020) "Rift Valley Fever in Northern Senegal: A Modelling Approach to Analyse the Processes Underlying Virus Circulation Recurrence" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12493098&blobtype=pdf | # Predictive Validity of Pre-Clinical Academic Achievements in Comprehensive Basic Science Examination: A Nationwide Cohort of Iranian Medical Students
Farhang Rashidi, Reza Sattarpour, Alipasha Meysamie
## Abstract
Background: Medical education directly impacts patient care, yet the predictive validity of pre-clinical academic performance for licensure exam outcomes remains debated. This national, multi-institutional study (2019-2021) assessed the relationship between university course grades, cumulative grade point average (GPA), and Comprehensive Basic Science Examination (CBSE) scores in Iranian medical students. Methods: Course grades and GPAs of 23 medical schools were linked to CBSE outcomes of 51 medical schools across five consecutive exam periods via student national ID. Pearson's correlation, paired t-tests, ANOVA, and chi-square assessed trends. Hierarchical cluster analysis (dendrogram) examined course grade correlations. Independent CBSE total score predictors were found using multiple linear regression. Results: Of the 25,757 individual records, 9,359 (45.2% female) had complete academic and CBSE data, making them eligible for primary analyses (84.5% passed CBSE on the first attempt). The GPA was 15.11±1.74, and the CBSE score was 101.68±24.61. All course grades correlated significantly with CBSE subtests (r=0.055-0.544, P<0.001). A significant moderate association (r=0.492, P<0.001) exists between overall GPA and CBSE. Repeat examinees had considerably lower GPAs and CBSE scores (P<0.001). GPA (β=0.318), Anatomy (β=0.158), Physiology (β=0.135), Epidemiology (β=0.043), and Virology (β=0.043) were the most significant predictors in regression modeling (R²=0.426). Cluster analysis showed that academic grades in anatomy, physiology, and biochemistry were strongly correlated with CBSE subtests.
Conclusion:This study represents the first large-scale national dataset in Iran pertaining to medical education. Pre-clinical GPA and course grades exhibit overall and subject-specific, notable predictive validity for CBSE performance. To enhance medical education and licensure results, it is advisable to implement standardized, cross-institutional comparisons alongside dynamic curriculum reviews. The regression model and clustering insights provide a framework for targeted educational interventions.
## Introduction
Medical student academic performance is usually considered an indicator of future physician competence, with implications for career success and quality of patient care. Medical educators require comprehensive insight into academic performance metrics, including grade assessments, attendance patterns, and engagement in clinical experiences, to effectively identify at-risk students, optimize curricula, and foster an environment that leads to learning and success. Previous studies have explored factors affecting academic performance, such as pre-admission qualifications, learning preferences, mental health, and environmental influences. However, there is a continued necessity for large-scale, nationally representative research that encompasses diverse student populations and contexts. [1][2][3] We utilize Messick's comprehensive validity framework to evaluate the validity of academic assessments, focusing on content relevance, response processes, internal structure, relationships to other variables, and consequences. 4,5 The curriculum for pre-clinical courses in Iran includes basic science and physiopathology courses. Studies argue that pre-clinical academic performance is a necessary foundation for clinical success. Nehy et al underlined that insufficient attention to basic sciences might have a negative impact on comprehensive test scores. 2 Supporting this, Lynch found that university-level basic science courses more accurately predict performance in advanced coursework than secondary education science preparation, reinforcing the necessity of basic sciences within medical curricula. 6 However, there is a growing recognition of the need for greater integration between basic sciences and clinical learning, as well as the inclusion of interdisciplinary subjects that address the complex health needs of society. 7,8 Medical curricula of the Iranian Medical Education Department require students to take an exam called the Comprehensive Basic Science Exam (CBSE) upon completion of basic science courses. This helps to assess how well they understand the key concepts, allowing them to choose their clinical courses, given a previous study highlighting the validity of CBSE for screening students for the following academic courses. 9 Prior investigations into the predictive validity of CBSE scores in relation to course grades and GPA have yielded varied outcomes, with certain studies demonstrating robust correlations, whereas others indicate weak or inconsistent relationships. Furthermore, international evidence suggests the presence of biases and inequities in assessment systems that could influence these correlations. In contracting results, Lankarani et al found that the average scores of the comprehensive basic science exam, except for the immunology course, have consistently decreased by one to two points compared to the class scores, and even in the case of psychology, there has been a more obvious downward trend. 10 There was no correlation between average course scores and CBSE sub-scores in all these courses. Similar results were also found by Nemat Bakhsh et al in a prospective study. Evidence from medical education globally, including contexts similar to Iran, highlights that assessment systems often suffer from bias and inequity that undermine the validity of alignment between course assessments and comprehensive exams. 11 Bias in assessment can distort the predictive value of GPAs and exam scores, suggesting that apparent correlations may not fully reflect true competence or learning outcomes. Despite these insights, there is a lack of research evaluating the reliability of pre-clinical performance as a predictor of CBSE outcomes and, ultimately, clinical competence within a large, nationally representative Iranian cohort. This study investigates the correlations among basic science course scores, GPA, and CBSE results, utilizing a validity framework to assess their predictive value and educational implications.-.
## Materials and Methods
## Study Design and Setting
This comprehensive national research employed a retrospective cohort design to assess the academic performance of medical students throughout Iran, concentrating on the correlation between their university grades and scores from the Comprehensive Basic Science Examination (CBSE). The investigation took place at medical schools across the country, encompassing a wide variety of academic institutions, both public and private. The study focused on evaluating student performance over five academic cohorts spanning from 2019 to 2021, including all medical schools throughout Iran, yielding a comprehensive and representative sample of medical students with diverse educational backgrounds.
## Ethical Considerations
The study protocol was reviewed and approved by the National Agency for Strategic Research in Medical Education (NASR) Ethics Committee (approval code: IR.NASRME.REC.1402.125). Measures were implemented to ensure compliance with international publishing standards. Initially, national ID numbers were required for record linkage; however, all identifiers were permanently anonymized prior to analysis to safeguard participant privacy. Results were compiled, and personal information was excluded, focusing solely on academic characteristics. Access to secure institutional systems was restricted solely to the study team, who had password protection in place. All files remained on internal systems, and processing adhered to national data protection regulations. The ethics committee waived informed consent due to the retrospective nature of the investigation and the anonymization of the results. The inclusion of all participating https://doi.org/10.2147/AMEP.S552380
Advances in Medical Education and Practice 2025:16 university records, without preferential selection, and the identification of contextual discrepancies addressed equity concerns. The research procedures ensured the preservation of all student rights, confidentiality, and equitable representation.
## Data Collection Process
The data collection process required several coordination steps with the General Medicine Secretariat, which enabled communication with the Educational Department of the Ministry of Health and Medical Education. The dataset comprised academic performance data for students from the five most recent cohorts, specifically encompassing the 2019 (winter), 2020 (fall), 2020 (winter), 2021 (fall), and 2021 (winter) examination periods. The requested data comprised students' grades in essential medical science courses, including Physiology, Biochemistry, Microbiology, Parasitology, Entomology, and Virology, along with their cumulative GPAs. The initial objective was to gather data over a decade; however, this was restricted to five cohorts due to difficulties accessing student records and substantial curriculum reforms in the General Medicine program in the previous years. The dataset offered an overview of performance across a five-year timeframe.
After obtaining the required permission and coordination for using students' data from the NASR, all 51 medical schools were asked to provide us with the required data. However, only 23 responded to the request, submitting data across 212 distinct files in Excel or Word format. Despite initial challenges in standardizing formats, the universities provided the requested course grades and GPAs for the specified cohorts. They sent the data to the Ministry of Health and Medical Education, which shared them with us. Nonetheless, certain universities did not comply with the specified data structure, resulting in the submission of incomplete or non-standardized records. Only universities that supplied complete and standardized data were included in the analysis; institutions that did not provide accurate information or did not meet data standards were excluded. The final dataset regarding academic performance comprised 11,240 records sourced from 23 universities. Simultaneously, CBSE results were obtained from the Educational Assessment Center of the Ministry of Health, consisting of five cohorts from 51 medical schools with 23,434 individual records. These two datasets were later associated using students' national ID numbers. The ethics committee waived student consent due to using the documented data, and the anonymous nature of the analysis and presentation of the results.
## Data Processing and Standardization
The process of data cleaning and integration comprised three essential steps:
A. Cleaning University Data: Discrepancies in the submitted files, including missing or incorrect student national ID numbers, have been addressed. In instances of incomplete course grades, GPAs were recalculated utilizing the available course credit weights. B. The CBSE scores were standardized across all cohorts to address variations in difficulty levels during different examination periods. The data were integrated with university records using student national IDs, resulting in a unified dataset. C. Final Data Merging: Following the data cleaning process, 25,757 records were consolidated from the two main datasets (university grades and CBSE scores) into a unified dataset, ensuring accurate linkage of all student identifiers and performance metrics. The final dataset comprised 11,240 records of university grades and GPAs, alongside 23,434 records of CBSE results. A total of 9,359 records, representing 36.3%, contained both university grades and CBSE scores, which were incorporated into the final analysis, and the remaining were considered as missing. No significant differences were observed between the omitted individuals (due to the lack of required data to link the GPAs and CBSE scores) and the included students; thus, the data were assessed to be representative of the national scale.
## Statistical Analysis
Descriptive statistics, including mean, standard deviation, median, and interquartile range, were computed for both raw and standardized scores. It is worth mentioning that the scale of GPAs and each course's score was 10 through 20 since students could not take the CBSE in case of failing courses (which is traditionally considered as obtaining a score below 10 in any course). However, the average CBSE score falls between 70% of the top 5% of students' scores (which is the passing threshold according to the Ministry of Health's Educational Assessment Center) and 200 (which is the full mark). Moreover, due to the different difficulty scales of exams in each medical school, the course scores and GPAs were not directly comparable between different schools. Thus, CBSE scores and university grades were standardized using z-scores to ensure comparability across cohorts. Standardization involved separately standardizing CBSE scores and university grades for each cohort, employing the mean and standard deviation of the entire cohort to compute z-scores.
The final standardized scores were transformed to achieve a mean of 15 and a standard deviation of 3.5, enabling comparison across various cohorts and variables. Raw and standardized scores for CBSE and university grades were classified into three performance categories: A (≥17), B (≥14 and <17), and C (<14). This classification facilitated a more precise analysis of performance trends and group comparisons. The analysis of trends and performance over time utilized raw CBSE scores as the primary basis, given that these scores were standardized across cohorts. Various statistical tests were utilized to analyze the relationships between university grades and CBSE performance. A paired Samples t-test is employed to compare the means of university grades and CBSE scores for individual students. Using Pearson Correlation, the strength of the linear relationship between university grades and CBSE scores was evaluated. The McNemar Test was utilized to compare categorical data between paired observations. ANOVA and Independent Samples t-tests were employed to assess performance differences among various groups, such as universities or cohorts. The Chi-square Test for Trend evaluated temporal trends in academic performance among different cohorts. Regression analysis employed linear regression models to investigate the association between university grades and CBSE scores while controlling for potential confounding variables, including cohort and university type.
However, one major unpredictable confounding factor was the COVID-19 pandemic. Although major comprehensive examinations in the pandemic period were held in person, as they would have been in a normal situation, the basic science courses examinations and scorings were held online. Due to a sudden major shift toward online education and assessments during this period, the effects and influences of this shift are yet to be comprehensively studied and evaluated.
Data cleaning, standardization, and statistical analyses were conducted using SPSS version 27. The significance level was established at P<0.05 for all analyses, with results presented alongside 95% confidence intervals. Error bar plots were graphically used to represent the relationship between university grades and CBSE scores. Missing data (those with mismatching national IDs regarding GPAs and CBSE) were excluded from any analysis to maintain the robustness of the outcomes.
## Results
## Descriptive Statistics
A total of 25,757 student records were collected from 23 medical schools across the country. Among these, 9,359 records (36.3%) included both CBSE and university grade data, 14,074 records (54.6%) contained only CBSE data, and 2,324 records (9.0%) comprised solely university grades (Table 1). Out of students with both CBSE and university grade data available, 5,129 (54.8%) were male, while 4,230 (45.2%) were female. A majority of students (84.5%) attempted the CBSE once, while 10.8% did so twice, and 4.9% attempted it three or more times. In a sample of 23,433 CBSE records, the mean score was 101.68, with a standard deviation of 24.61. The subtest means varied from 0.91±0.74 in Entomology to 26.05±8.44 in Anatomy (Table 2). For university course grades, among 11,644 students with complete data, course grade means were closely grouped between 15 and 16 on a 0-20 scale, with the highest mean in Islamic Studies (18.00 ±1.75) and the lowest in Microbiology (14.91±2.56) (Table 3). The mean GPA was 15.11, with a standard deviation of 1.74. Upon stratification into grades A (≥17), B (14-17), and C (<14), Islamic Studies exhibited the highest proportion of "A" grades at 77.9%, followed by English at 51.4%. In contrast, Microbiology recorded a mere 24.6% "A" grades (Table 4). Correlation Between GPA and CBSE Paired samples t-tests showed that all courses' raw and standardized means differed significantly (P<0.001), with Pearson correlations ranging from r=0.055 (Virology) to r=0.544 (Anatomy). The overall GPA-CBSE correlation was r=0.492 (P<0.001) (Table 5). Pearson correlations among all standardized university courses and CBSE total ranged from r=0.183 (Islamic Studies vs Microbiology) to r=0.789 (GPA vs Anatomy), indicating robust interrelationships (Table 6).
## Performance by Course
Performance exhibited significant variation across subjects. Students achieving high grades in a course generally performed better on the associated CBSE subtest. CBSE Physiology scores averaged 24.47±4.80 for students receiving an "A" in university Physiology, compared to 17.95±4.84 for those with a "C" (P<0.001). In Biochemistry, the mean CBSE score was 11.08±3.99 for students receiving an "A" grade, in contrast to 7.85±2.92 for those with a "C" grade (P<0.001). Parallel trends were observed in Microbiology and other courses (all P<0.001), indicating subject-specific differences in predictive validity, where higher course grades predicted higher CBSE scores.
A multiple linear regression analysis involving 8,209 complete cases evaluated the predictive significance of university courses on the CBSE total score. A preliminary "Enter" model revealed significant positive coefficients for ten courses and a significant negative coefficient for GPA, whereas Parasitology and Entomology were non-significant. A backward-elimination model was conducted, resulting in the retention of ten predictors, which included GPA and nine courses. The model yielded R=0.653, R²=0.426, and adjusted R²=0.425 (Table 7).
## Repeat Test Patterns
Students who repeatedly attempted the CBSE generally exhibited diminished academic performance. Among two-time takers, only 1.4% achieved the top GPA category ("A"), whereas 12.8% of first-time takers reached this level. In contrast, 53.7% of students who attempted the assessment twice were categorized in the lowest GPA bracket ("C"), compared to 26.2% of those who took it once. This pattern persisted across three or more trials. Repeat examinees were overrepresented in the lowest GPA category, aligning with patterns of repeated failure.
## Time Trends
Comparative analysis of CBSE scores across exam periods from Winter 2019 to Winter 2021 revealed an increase in the proportion of students achieving grade A, rising from 4.5% in Winter 2020 to 11.9% in Winter 2021 (χ² for trend P<0.001). This trend suggests performance improvements, likely attributable to curriculum reforms (Table 8).
## Discussion
The current study is the first nationwide study evaluating the academic performance of Iranian medical students. The results revealed statistically significant correlations between academic metrics and licensing examination performance. The analysis demonstrated a moderate correlation (r=0.492, P<0.001) between cumulative GPA and total Comprehensive Basic Science Examination (CBSE) scores, with subject-specific correlations ranging from weak (Virology, r=0.055) to strong (Anatomy, r=0.544). The analysis identified GPA, Anatomy, Physiology, Epidemiology, and Virology as the strongest independent predictors of CBSE performance (R²=0.426), suggesting specific foundational sciences may particularly influence licensing examination success.
## Correlation Between Academic Performance and Licensing Examination Outcomes
Our findings align with previous studies in other countries, demonstrating that undergraduate academic metrics can predict licensure comprehensive exam outcomes. A recent study of osteopathic medical students reported that cumulative undergraduate GPA significantly predicted performance on both Level 1 and Level 2 of the COMLEX-USA exams (β coefficients comparable to ours), supporting the generalizability of GPA as a predictor across contexts. 12 Similarly, research has demonstrated that science-based test scores (comparable to our basic science course grades) exhibited the strongest correlations with medical school performance and licensing exam results. 13 However, the literature in Iranian studies is challenging. While some studies support the substantial predictive value of GPAs in comprehensive exams, [14][15][16][17] others question this notion. 11 Nematbakhsh reported a positive but weak correlation between most of the courses and CBSE scores. 11 This might be due to the localization of a single medical school and the selection bias they have had, along with the curriculum change later, which considers their data out of date. Also, a systematic review later in 2016 concluded that students' GPA is a key predictor of comprehensive exam results. 18 Interestingly, Adelmashhadsari et al 19 concluded that the students' academic performance in high school also plays a significant role in their educational progress. Unfortunately, no data regarding high school grades were available for the current study to confirm our results further. They also recommended considering high school GPA as an influential factor in selecting medical students. 19 This approach has been highly implemented in the Iranian National University Entrance Exam (Konkour) since 2020 and might be a potential approach for consideration in the Iranian National Residency Entrance Exam.
Our moderate overall GPA-CBSE correlation highlights that pre-clinical academic achievement remains a notable predictor of licensing exam success, though not exclusive. This correlation aligns with findings from a study that demonstrated earlier academic performance measures can effectively predict licensing examination scores, particularly within longitudinal curricula. 20,21 Our research further validates this predictive relationship in the context of Iranian medical education.
## Subject-Specific Predictive Patterns
Subject-specific analyses in our cohort revealed that courses with high cognitive load and integrative content, Anatomy (mean β=0.158), and Physiology (β=0.135) were among the most potent predictors of CBSE total score. This mirrors findings from Lynch et al, who demonstrated that basic science coursework more accurately forecasts advanced academic performance than pre-university science preparation. 13 Nehy et al also emphasized the critical role of basic science mastery in achieving strong comprehensive exam results, advocating for reinforced foundational teaching to improve summative assessment outcomes. 22 The clustering of Anatomy, Physiology, and Biochemistry grades with their CBSE subtests, revealed by our dendrogram analysis, further supports the curricular emphasis on integrated, clinically relevant basic sciences. This pattern suggests that certain cognitive domains share underlying structures that transfer across course performance and standardized testing. Interestingly, these findings also align with research showing that basic science mastery provides the cognitive scaffolding necessary for clinical reasoning development and examination success. 23,24 Although pre-clinical GPA and subject-specific performance demonstrate important predictive value, a significant amount of variance in CBSE outcomes remains unaccounted for. This indicates that non-cognitive factors may significantly contribute to student success. Previous studies have emphasized the impact of factors such as motivation, self-regulated learning, resilience, and test-taking strategies, alongside contextual elements like socioeconomic status and availability of academic resources. [25][26][27] The dimensions not addressed in this study likely have a significant impact on licensure examination performance and require systematic investigation. Integrating non-cognitive measures into forthcoming predictive models could enhance their explanatory capacity and offer a more comprehensive understanding of student achievement.
## Institutional Variations and Standardization Challenges
We also observed significant differences in both raw and transformed scores across medical schools, likely reflecting variability in institutional resources, faculty expertise, and student support infrastructures. Although disparate grading policies and assessment standards challenge direct inter-institutional comparisons, our standardized scoring approach (mean=15, SD=3.5) offers one model for facilitating fair comparisons. Recent research on USMLE Step 1 performance across different medical school types revealed notable variations in pass rates. In 2021, first-time takers from U.S./ Canadian MD and DO degree programs had pass rates of 96% and 94%, respectively, while non-US/Canadian schools had a pass rate of 82%. 28 Following the transition to pass/fail scoring in 2022, these rates dropped to 93%, 89%, and 74%, respectively. Similar calls for cross-institutional standardization have emerged in international settings; for instance, the Association of American Medical Colleges has advocated for standardized pre-clerkship score transformations to improve comparability across US medical schools. 29 Adopting a national framework for basic science assessment standardization in Iran may enhance equity and benchmarking.
## Implications for Educational Support
The markedly lower GPAs and CBSE scores among repeat examinees in our cohort echo patterns observed in US licensing data. Eisendrath et al reported that repeat USMLE takers displayed consistently lower pass rates and score gains, with most successful repeat attempts occurring by the fourth try. 30 While repeated exposure to exam content can confer modest score improvements, the diminished baseline performance of repeaters highlights the importance of early identification and targeted remediation for at-risk students. Our data reinforce the need for robust support systems, such as formative assessments, academic coaching, and stress management resources, to reduce the likelihood of repeat failures.
## Temporal Trends and Curriculum Reform
Temporal trends in CBSE performance showed modest improvement. "A" grade rates rose from 4.5% in Winter 2020 to 11.9% in Winter 2021 (χ² trend P<0.001), potentially reflecting curriculum reforms initiated in 2019. This aligns with global moves toward competency-based pre-clinical curricula, which integrate basic sciences with early clinical experiences and employ frequent formative assessment to support learning. 31 However, this timeframe aligns with the COVID-19 pandemic, which shifted education and assessments toward remote, online formats, potentially leading to fraud and related confounders. Although, as explained in the methods, CBSE were held in person during the pandemic era, the GPAs might have been affected explicitly.
Future longitudinal studies should assess whether these curricular innovations sustain performance gains, particularly in underperforming subtests. Research has demonstrated that learning strategy interventions, particularly those focused on concentration, can significantly impact licensing examination performance, suggesting that curriculum reforms emphasizing these strategies may yield long-term benefits. 32,33 Hierarchical Cluster Analysis and Curriculum Mapping Our hierarchical cluster analysis, revealing tight clusters between related courses and their corresponding CBSE subtests, suggests that certain domains (eg, anatomical sciences) may share underlying cognitive and pedagogical constructs. These insights can inform curriculum mapping and assessment design by identifying clusters of content that benefit from integrated teaching approaches.
This clustering pattern is supported by research on COMLEX-USA performance, which found that elective upperlevel undergraduate science courses influenced performance on licensing examinations. 34,35 The relationship between course clusters and examination subtests suggests that curricular organization that reinforces these natural knowledge structures may enhance student performance on comprehensive assessments. Additionally, research on pathology board examination performance found that higher USMLE Step 1 scores (≥90 on the 2-digit scale) perfectly predicted firstattempt success on specialty board examinations over nine years. 34,36 This perfect correlation for high-performing students further supports the notion that certain knowledge structures and test-taking abilities transfer across different assessment formats. 12,13,22,[29][30][31]37 Limitations and Implications for Future Research As mentioned earlier, the current study is the first national study of Iranian medical students' performances. However, there are some limitations, including the exclusion of roughly 60% of potential records due to incomplete or limited data, an inherent challenge in large-scale, retrospective designs, and the inability to assess other influential factors such as entrance exam rank, socioeconomic status, and non-cognitive attributes. Moreover, our reliance on national ID for data linkage while ensuring student privacy may have introduced selection bias if ID errors correlated with performance. In addition, the COVID-19 pandemic was a major, unpredictable confounding factor whose effects have not yet been comprehensively studied. Sudden shift toward online education and assessments may have caused score skewness and biased our results. Future research should incorporate prospective designs with richer covariate data and explore the impact of modern e-learning tools and competency-based assessments on predictive validity.
Implications for medical educators and policymakers include (1) reinforcing foundational basic science teaching, particularly in Anatomy and Physiology, to maximize CBSE performance; (2) implementing early identification and support programs for students at risk of repeat exam attempts; (3) adopting standardized score transformations to enable fair inter-institutional comparisons; and (4) tailoring gender-sensitive educational strategies to optimize learning and exam readiness. Our regression model and clustering insights provide actionable targets for curriculum enhancement and personalized remediation.
## Conclusion
In conclusion, this nationwide analysis highlights the moderate predictive validity of pre-clinical GPA and course grades for comprehensive basic science exam outcomes, revealing important subject-specific and demographic nuances. The results require careful interpretation due to not fully capturing all potential confounders and the unique disruptions posed by the COVID-19 pandemic. The results suggest that by integrating standardized assessments, targeted support interventions, and standard education aligned with the curriculum, medical schools can better prepare students for both academic success and clinical competence in an ever-evolving healthcare landscape. Future research should investigate these relationships in varied contexts and assess the long-term effects on clinical competence.
## References
1. *Parasitology + Entomology*
2. *Virology*
3. *Public Health and Ethics*
4. *Epidemiology*
5. *Biochemistry*
6. *Physiology*
7. *Microbiology*
8. "Abbreviations: SD, Standard Deviation; GPA, Grade Point Average"
9. (2009) *Public Health and Ethics*
10. "Abbreviation: GPA: Grade Point Average. Table 5 Paired t-Tests and Pearson Correlations Between Course Grades and CBSE Subtests Course N r p Anatomy" *Physiology*
11. *Epidemiology*
12. *Public Health and Ethics*
13. *Biochemistry*
14. *Microbiology*
15. *Mycology*
16. *Virology*
17. "Islamic Studies"
18. *Parasitology + Entomology*
19. "GPA vs CBSE 9"
20. Rueangket, Thaebanpakul, Sakboonyarat (2024) "Educational data mining: factors influencing medical student success and the exploration of visualization techniques" *Advances in Medical Education and Practice*
21. Nehy, Jaypalan, Naserpoor (2014) "Basic science educational department performance evaluation based on comprehensive basic science exams"
22. Diarsvitri, Garianto, Indrawati (2024) "Estimating students' academic success in the preclinical stage of undergraduate medical education using the admission test approach" *J Adv Med Educ Professionalism*
23. Messick (1989) "Validity"
24. Downing (2003) "Validity: on meaningful interpretation of assessment data" *Med Educ*
25. Lynch, Grant, Mcloughlin (2016) "The relevance of basic sciences in undergraduate medical education" *Irish J Med Sci*
26. Davoudi, Esmaeeli, Ahmadzadasl (2017) "Academic performance in Iranian medical students during the pre-clinical stage" *Med J Islamic Republic Iran*
27. Dabaliz, Kaadan, Dabbagh (2017) "Predictive validity of pre-admission assessments on medical student performance" *Int J Med Educ*
28. Sekandari, Akbari (2016) "Evaluation of the predictive validity of comprehensive basic science examination for the adequacy of dentistry students'" *Clin Competence*
29. Lankarani, Honarvar, Faghihi (2024) "Demanded interdisciplinary subjects for integration in medical education program from the point of view of graduated medical physicians and senior medical students: a nationwide mixed qualitative-quantitative study from Iran" *BMC Med Educ*
30. (2004) *Nemat Bakhsh M. Teacher evaluation. Iran J Med Educ*
31. Royal, Meyer, Guercio (2025) "The predictive validity of MCAT scores and undergraduate GPA for COMLEX-USA licensure exam performance of students enrolled in osteopathic medical schools" *J Osteopath Med*
32. Almarabheh, Shehata, Ismaeel (2022) "Predictive validity of admission criteria in predicting academic performance of medical students: a retrospective cohort study" *Front Med*
33. Bahramian, Negahdari, Kahnamouei (2021) "The relationship between the average score of first two years and comprehensive basic science examination score of dentistry students of Tabriz University of medical sciences" *Iran J Med Educ*
34. Khazaei, Zarin, Rezaei (2018) "Factors affecting the results of comprehensive pre-internship exam among medical students of Kermanshah University of Medical Sciences" *Korean J Med Educ*
35. Jafari, Aminizadeh, Samadpour (2015) "Evaluation of comprehensive medical basic sciences exams and their predictive factors" *Biomed Pharmacol J*
36. Esmaeili, Haghdoost (2008) "The internal consistency of medical students' scores in their physiopathology and clinical courses" *Indian J Med Sci*
37. Changiz, Asgari, Yamani (2016) "A systematic review of studies on general medicine comprehensive exams in Iran" *J Res Med Dental Sci*
38. Adelmashhadsari, Elmi, Elmi (2016) "The effect of diploma grade point average on educational evaluation status of medical students in Babol University of Medical Sciences during 2006-2011" *Res Med Educ*
39. Banal, Malijan, Solidum (2024) "The predictive value of admission qualifications on the academic performance of first-year medical students in Pamantasan ng Lungsod ng Maynila (PLM) College of Medicine" *J Educ Learn*
40. Zhong, Wang, Christensen (2021) "Early prediction of the risk of scoring lower than 500 on the COMLEX 1" *BMC Med Educ*
41. Saxena, Wright, Khalil (2024) "Gender differences in learning and study strategies impact medical students' preclinical and USMLE step 1 examination performance" *BMC Med Educ*
42. Clark (2012) "Formative Assessment: assessment Is for Self-regulated Learning" *Educ Psychol Rev*
43. Mcneil, Mashhouri, Christensen (2022) "Effects of organ system courses of the first two years of medical school on the performance of the Comprehensive Osteopathic Medical Licensing Examination of the United States (COMLEX-USA) level 2-cognitive evaluation" *Cureus*
44. Chisholm-Burns, Berg-Poppe, Spivey (2021) "Systematic review of non-cognitive factors influence on health professions students' academic performance" *Adv Health Sci Educ*
45. Privado, Pérez-Eizaguirre, Martínez-Rodríguez (2024) "Cognitive and non-cognitive factors as predictors of academic performance. Learn Individual Differences"
46. Žuljević, Buljan (2022) "Academic and non-academic predictors of academic performance in medical school: an exploratory cohort study" *BMC Med Educ*
47. Yadav, Dekhne, Harikrishnan (2023) "The pass/fail effect: a longitudinal study of United States Medical Licensing Examination (USMLE) step 1 performance over a decade" *Cureus*
48. Huang, Arianpoor, Taylor (2024) "Insights into undergraduate medical student selection tools: a systematic review and meta-analysis" *J Educ Eval Health Prof*
49. Busche, Elks, Hanson (2020) "The validity of scores from the new MCAT exam in predicting student performance: results from a multisite study" *Acad Med*
50. Cuddy, Swanson, Clauser (2007) "A Multilevel analysis of the relationships between examinee gender and United States Medical Licensing Exam (USMLE) step 2 CK content area performance" *Acad Med*
51. Wong, Ramirez, Helf (2009) "Student performance on levels 1 and 2-CE of COMLEX-USA: do elective upper-level undergraduate science courses matter?" *J Osteopathic Med*
52. West, Kurz, Smith (2014) "Are study strategies related to medical licensing exam performance?" *Int J Med Educ*
53. Picarsic, Raval, Macpherson (2011) "United States Medical Licensing Examination step 1 two-digit score: a correlation with the American Board of Pathology first-time test taker pass/fail rate at the University of Pittsburgh Medical Center" *Arch Pathol Lab Med*
54. Wong, Ramirez, Helf (2009) "Student performance on levels 1 and 2-CE of COMLEX-USA: do elective upper-level undergraduate science courses matter?" *J Am Osteopath Assoc*
55. Maholtz, Erickson, Cymet (2015) "Comprehensive osteopathic medical licensing examination-USA level 1 and level 2-cognitive evaluation preparation and outcomes" *J Am Osteopathic Assoc*
56. Rubright, Ong, Jodoin (2022) "Advances in Medical Education and Practice Publish your work in this journal Advances in Medical Education and Practice is an international, peer-reviewed, open access journal that aims to present and publish research on Medical Education covering medical, dental, nursing and allied health care professional education. The journal covers undergraduate education, postgraduate training and continuing medical education including emerging trends and innovative models linking education, research, and health care services. The manuscript management system is completely online and includes a very quick and fair peer-review system" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817916&blobtype=pdf | # A small-molecule HSP90 inhibitor, NVP-HSP990, alleviates rotavirus infection
Yi Cao, Qingmin Zhu, Xiaoping Wu, Zhunyi Xie, Chengying Yang, Yanyan Guo, Dongwei Meng, Xinyuan Zhou, Yuzhang Wu, Jintao Li, Haiyang He
## Abstract
Rotavirus (RV) infection remains a leading cause of hospitalization and mortality among infants and young children. Despite global implementation of RV vaccines, hundreds of thousands of infants and young children still succumb to this disease each year due to ineffective treatment. In this study, we demonstrated that NVP-HSP990, a novel small-molecule heat shock protein 90 (HSP90) inhibitor, inhibited RV infection with a fascinatingly higher selectivity index compared to conventional HSP90 inhibitors like geldanamycin and its derivative tanespimycin ). NVP-HSP990 effectively inhibited RV replication in vitro without blocking the initial establishment of infection. NVP-HSP990 restored host gene expression in most KEGG pathways disrupted by RV infection in Caco-2 cells, except some inflammatory pathways (such as IL-17 and TNF pathways). NVP-HSP990 significantly inhibited RV-induced activation of the MAPK pathway and prevented the disruption of tight junctions in Caco-2 cells. More importantly, NVP-HSP990 effectively suppressed RV infection in BALB/c suckling mice and significantly alleviated RV-induced diarrhea.IMPORTANCE Rotavirus (RV) infection poses a global health threat with an urgent need for targeted antiviral therapies. Here, we identified NVP-HSP990 as a next-gen eration HSP90 inhibitor with exceptional translational potential against RV infection. Compared to conventional HSP90 inhibitors, NVP-HSP990 demonstrated markedly enhanced anti-RV selectivity. NVP-HSP990 effectively reversed dysregulation of key host pathways in RV infection while selectively modulating pro-inflammatory respon ses, thereby balancing antiviral and immunopathological outcomes. NVP-HSP990 also blocked MAPK-driven tight junction disruption to preserve intestinal barrier integrity. As a result, NVP-HSP990 significantly alleviated the severity of RV-induced diarrhea. Given its excellent oral efficacy and systemic penetration previously reported, NVP-HSP990 emerges as a promising HSP90-targeted candidate capable of addressing both intestinal and possible extraintestinal RV infections, which also repositions HSP90 inhibition as a viable strategy in RV management. KEYWORDS diarrhea, MAPK signaling pathway, NVP-HSP990, rotavirus, tight junctions R otavirus (RV), a double-stranded RNA virus that belongs to the Sedoreoviridae family (1), is the leading cause of diarrhea in infants and young children (2). RV infection mainly occurs in the intestinal epithelium and causes dysfunction of enterocytes and disorder of the enteric nervous system, leading to severe dehydrating diarrhea and vomiting; moreover, beyond gastrointestinal symptoms, severe RV infection sometimes causes systemic infection through viremia, affecting critical organs such as the brain and leading to poor prognosis (3-5). Although expansion of RV vaccines into national immunization programs worldwide has led to a 59% decrease in hospitalizations and
a 36% decrease in deaths caused by RV infection (6), according to the latest global survey by WHO, RV infection still causes approximately 0.64% of hospitalizations and 208,009 deaths of children under 5 years of age in 28 low-and middle-income countries (7). Symptomatic treatments and gastrointestinal protective agents are usually the main choices for management of RV infection, while conventional antiviral drugs like nucleoside analogs and type I interferon are rarely employed due to their significant adverse side effects (8,9). This limitation sometimes results in uncontrolled RV infections and even mortality. Therefore, there is an urgent need to develop safe and effective antiviral drugs for RV infections.
Heat shock protein 90 (HSP90), a chaperone protein present in both eukaryotes and bacteria, has four isoforms: HSP90α, HSP90β, GRP94, and TRAP1. HSP90α and HSP90β are predominantly found in the cytoplasm, GRP94 is located in the endoplasmic reticulum, and TRAP1 is present in the mitochondria (10,11). HSP90 promotes viral replication by regulating cell signaling systems or by directly interacting with viral proteins such as hepatitis B virus reverse transcriptase, hepatitis C virus non-structural protein 5A, and influenza virus A RNA-dependent RNA polymerase (12)(13)(14)(15). HSP90 is involved in the entry of RV into certain tumor cell lines (16)(17)(18), and also contributes to RV replication in vitro (19,20). Therefore, it seems that HSP90 inhibitors would be promising broad-spectrum antiviral drug candidates (21); nevertheless, in contrast to their widespread application in anti-tumor therapies (22), no HSP90 inhibitors are in clinical use for antiviral therapies currently, probably due to unsatisfactory antiviral efficacy and undesirable toxicity.
NVP-HSP990 is a novel small-molecule HSP90 inhibitor fundamentally distinct from the natural HSP90 inhibitor geldanamycin (GA) and its derivative tanespimycin (17-allylamino-17-demethoxygeldanamycin, 17-AAG) in both chemical structure and mechanism (23)(24)(25). Briefly, as a fully synthetic inhibitor, NVP-HSP990 adopts a resor cinol-isoxazole scaffold, eliminating the natural-product-derived macrocyclic benzoqui none ansamycin backbone shared by GA and 17-AAG. This synthetic design avoids the redox-active benzoquinone moiety responsible for oxidative stress and hepatotoxicity in ansamycins. While GA and 17-AAG rely on a 19-membered macrocycle and benzo quinone for ATPase binding, NVP-HSP990 achieves potent HSP90 inhibition through its optimized resorcinol-based architecture. Critically, NVP-HSP990 exhibits superior oral bioavailability without quinone-driven toxicity, distinguishing it from intravenousadministered ansamycins and positioning it as a next-generation HSP90 inhibitor with improved safety and therapeutic potential (26). These advantages position NVP-HSP990 as a promising antiviral candidate even though experimental data on its effects against viral infections are currently limited.
In this study, we aimed to investigate the inhibitory effects of NVP-HSP990 on RV infection. Specifically, we evaluated its antiviral efficacy against RV in vitro and in vivo, elucidated its potential mechanisms of action, and assessed its therapeutic efficacy on RV diarrhea, thereby providing a solid foundation for the development of NVP-HSP990 as a promising therapeutic agent against RV infection.
## MATERIALS AND METHODS
## Cell culture, viral infection, and drug administration
Rhesus monkey embryo kidney cell line MA104 cells (ATCC: CRL-2378.1) were provided by Dr. Elschner (Friedrich-Loeffler-Institute). Human intestinal epithelial cell lines Caco-2 cells (ATCC: HTB-37) and HT-29 cells (ATCC: HTB-38) were from ATCC and kept in our institute. All the cells were cultured in complete DMEM (Dulbecco's modified Eagle medium [Gibco, USA] plus 10% [vol/vol] fetal bovine serum [FBS] [Gibco, USA] and 1% penicillin/streptomycin [Gibco, USA]) at 37°C with 5% CO 2 . RV Wa (G1P [8]) and SA11 (G3P [2]) strains were gifted by Professor Duan Zhaojun from the China Center for Disease Control and Prevention (Beijing, China) and propagated in MA104 cells. RV EDIM strain G16P [16] was gifted by Professor Wan Jianwei in Christophe Merieux Laboratory (Beijing) and propagated in BALB/c suckling mice. For RV infection in vitro, the viruses were diluted in DMEM, activated with 10 µg/mL trypsin (Gibco, USA) for 30 min at 37°C, and then added to target cells previously washed once with DMEM. After 1 h of incubation at 37°C, the inoculum was removed; the cells were then washed twice with DMEM (without FBS) and incubated with DMEM (without FBS) at 37°C, which was denoted as 0 h post-infection (h p.i.). HSP90 inhibitors (NVP-HSP990, GA, and 17-AAG) and Ribavirin (Selleck, USA) were dissolved in DMSO (Sigma, USA). The drugs were administered with indicated concentrations at 0 h p.i. in vitro unless otherwise stated, or orally administered in 10 µL volume using a pipette tip with the indicated quantity plus 90% (vol/vol) corn oil.
## Cytotoxicity assay
To evaluate the cytotoxicity of different HSP90 inhibitors in vitro, 10,000 MA104 cells, 30,000 Caco-2 cells, and 80,000 HT-29 cells in 100 µL complete DMEM medium were seeded in 96-well plates and cultured at 37°C and 5% CO 2 for 24 h to reach about 100% confluence, and a cell-free control was also set. After removing the initial medium, 100 µL of complete DMEM medium containing the indicated concentrations of drugs was added to each well, and the plates were incubated at 37°C and 5% CO 2 . Twenty-four hours later, the drug-containing medium was replaced with 100 µL fresh complete DMEM medium plus 10 µL 5 mM CCK-8 reagent (Beyotime, China), and OD 450 absorbance of each well was detected 3 h later using a microplate reader (Gene Company Limited, China). Relative cell viability rate = OD 450 (drug-treated cells -cell-free control) / OD 450 (0 µM drug-treated cells -cell-free control).
## In vitro assay of viral inhibitory effects of drugs
For analysis of effects of different HSP990 inhibitors on RV replication in vitro, 10,000 MA104 cells, 30,000 Caco-2 cells, and 80,000 HT-29 cells in 100 µL complete DMEM medium were seeded in 96-well plates and cultured at 37°C and 5% CO 2 for 24 h. Then the cells were infected with RV Wa or SA11 strains (MOI = 1 (PFU/cell)) for 1 h. After removing the viruses and washing cells two times with DMEM, 100 µL DMEM containing the indicated concentrations of drugs was added to each well, and the plates were incubated at 37°C and 5% CO 2 . At 24 h p.i., the infected cells and culture medium were collected, frozen/thawed twice, and subjected to centrifugation at 1,000 × g for 3 min at room temperature. Then, the supernatant was subjected to analysis of viral load by PFA. The 50% inhibitory concentration (IC 50 ) values were determined by nonlinear regression (curve fit) analysis in GraphPad Prism 9.0 (GraphPad Software). Dose-response data (drug concentrations as x-axis and normalized inhibition percentages as y-axis) were fitted to a dose-response inhibition model ([Inhibitor] vs normalized response -Variable slope). The curve was constrained to a bottom limit of 0% inhibition. The IC 50 value was automatically calculated as the x-intercept at y = 50%. Model adequacy was validated using goodness-of-fit metrics (R 2 ≥ 0.95), 95% confidence intervals, and residual distribution analysis.
## RNA-sequencing (RNA-seq) analysis
Caco-2 cells in six-well tissue culture plates (90% confluence) were infected with RV Wa or SA11 strains (MOI = 3) for 1 h. After removing the viruses and washing cells two times with DMEM, the cells were cultivated with 2 mL DMEM containing 100 nM NVP-HSP990 or an equal volume of DMSO at 37°C and 5% CO 2 . At 24 h p.i., the Caco-2 cells were harvested, and total RNA was extracted using Trizol reagent kit (Invitrogen, USA), according to the manufacturer's protocol. RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and checked using RNasefree agarose gel electrophoresis. After total RNA was extracted, eukaryotic mRNA was enriched by Oligo(dT) beads and used for library construction and next-generation sequencing.
## Immunofluorescence assay (IFA)
Caco-2 cells growing on coverslips were mock-infected with PBS or infected with RV Wa or SA11 strains (MOI = 1), then the cells were cultivated with DMEM containing 100 nM NVP-HSP990 or an equal volume of DMSO as a control for another 18 h after infection. The cells were then washed with PBS, fixed in 4% paraformaldehyde for 20 min, followed by permeabilization in 0.5% Triton X-100 for 10 min and blocking with 5% BSA (Beyotime, China) for 1 h at room temperature. The cells were then incubated with rabbit anti-ZO-1 monoclonal antibody (CST, USA) (1:100) or rabbit anti-RV VP6 polyclonal antibodies (CUSABIO, China) (1:50) for 2 h at room temperature, washed twice with PBS, and incubated with Cy3-conjugated goat anti-rabbit antibodies (Beyotime, China) (1:500) for 1 h at room temperature. In some experiments, the cells were also stained with FITC-conjugated goat anti-RV antibodies (Virostat, USA) (1:100) for 1 h at room tempera ture and followed by 5 µg/mL 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) (Sigma-Aldrich, USA) staining for 10 min. The triple-stained cells were washed twice with PBS and mounted in Prolong Gold Antifade Reagent (CST, USA). Microscopic images were acquired with an EVOS M5000 system (Thermo Fisher Scientific, USA) using the built-in Celleste software (v.6.0). Raw image files were saved in TIFF format to preserve data integrity. Minimal post-processing (image cropping and annotations) was applied through Adobe Illustrator 2019 (Adobe Systems, USA) to ensure readability. No alterations were made to raw data during this process.
## Western blot
MA104, Caco-2, and HT-29 cells were mock-infected with PBS, infected with RV Wa or SA11 strains (MOI = 3), or treated with 1 µM C16-PAF (C16) for 1 h. After removing the viruses or C16-PAF and washing cells two times with DMEM, the cells were cultivated with 2 mL DMEM containing 100 nM NVP-HSP990 (+), an equal volume of DMSO (-), or 1 µM C16-PAF at 37°C and 5% CO 2 for 20 h. Then, the cells in each well (~5 × 10 5 cells) were lysed with 100 µL RIPA lysis buffer (Beyotime, China) plus protease inhibitor cocktail (Thermo Fisher Scientific, USA). The cell lysates were collected in 1.5 mL EP tubes, sonicated five times for 15 s at 80 watts on ice, and clarified by centrifugation at 12,000 × g for 10 min at 4°C. After protein quantification with a BCA kit (Beyotime, China), the protein samples were mixed with 25 µL 5× SDS loading buffer (Beyotime, China), boiled at 100°C for 5 min, chilled on ice, vortexed for seconds, and centrifuged at 12,000 × g for 3 min at room temperature. Twenty micrograms of protein was subjected to protein electrophoresis in precast 4-20% SDS-PAGE gels (Beyotime, China), and then transferred to 0.22 µm PVDF membranes (Millipore, USA). After blocking with 5% BSA, the membranes were incubated with rabbit antibodies to MAPK components (SAPK/ JNK, phospho-SAPK/JNK, p38 MAPK, phospho-p38 MAPK, ERK1/2, phospho-ERK1/2) (CST, USA), rabbit anti-RV VP6/VP7 polyclonal antibodies (CUSABIO, China), or mouse mAb to β-actin (Servicebio, China) as an internal reference. Afterward, the membranes were incubated with horseradish peroxidase-conjugated rat anti-rabbit IgG (CST, USA) or rat anti-mouse IgG (CST, USA). Immunoreactive bands were visualized using enhanced chemiluminescence substrate BeyoECL Plus (Beyotime, China).
## RV diarrhea models in suckling mice
BALB/c mice were purchased from Gempharmatech Inc. (Suzhou, China) and housed under SPF conditions (22 ± 1°C, 12 h light/dark cycle). Suckling mice were obtained by mating male and female adult BALB/c mice (12-to 16-week-old). Only the 7-day-old BALB/c suckling mice with body weights ranging from 2.5 g to 4.5 g were selected for RV diarrhea models. For induction of RV diarrhea, 1 × 10 6 PFU of SA11 strain or 10 × DD 50 of EDIM strain in 20 µL PBS was orally inoculated using a pipette tip to 7-day-old BALB/c suckling mice regardless of their sexes, as sex has no significant influence on RV diarrhea occurrence. We used a scoring system to evaluate fecal consistency and color as described with some modification (27): no stool or brown formed stool (1 point); brown soft stool (2 points); yellow soft stool (3 points); and yellow watery stool (4 points). Mice with scores of ≥2 meant diarrhea occurrence, while mice with scores of 1 meant no diarrhea. At the end of each experiment, all infected or drug-treated suckling mice were humanely euthanized via CO 2 asphyxiation.
## In vivo RV inhibition assays
Seven-day-old BALB/c suckling mice meeting the weight criteria within each litter were randomly assigned to experimental groups. Each litter was allocated to ensure one to three suckling mice per experimental group (variations dependent on the total number of groups). Data from suckling mice assigned to the same group across multiple litters were aggregated to achieve a total sample size of n = 3-6 per experimental group. Then, the suckling mice were orally inoculated with 1 × 10 6 PFU of the SA11 strain or 10 × DD 50 of the EDIM strain, or with an equal amount of PBS as mock infection. For SA11 infection, the suckling mice were orally treated once with indicated doses of drugs or with an equal volume of DMSO as control at 2 h p.i. For EDIM infection, the suckling mice were orally treated with 0.2 mg/kg/day NVP-HSP990, 30 mg/kg/day ribavirin, or an equal amount of DMSO as control from 2 days post-infection (d p.i.) to indicated time points. Diarrhea scores and body weight were monitored from day 0 to day 7 post-infection in the SA11 infection model, and from day 0 to day 9 post-diarrhea occurrence (2-11 d p.i.) in the EDIM infection model. Blinding was applied throughout data collection and analysis.
The RV-infected suckling mice were sacrificed by CO 2 asphyxiation at the indicated times, and small intestines (including duodenum, jejunum, and ileum) or colon was homogenized in 0.3 mL DMEM, centrifuged at 12,000 × g for 5 min at 4°C, and the supernatant was collected. Virus contents in the supernatant were titrated with PFA as described above. Viral antigens in the supernatant were detected with enzyme-linked immunosorbent assay (ELISA) kits for RV (CUSABIO, China) following the manufacturer's instructions. For qPCR analysis of RV protein expression, the intestines were longitudi nally split and washed in 1 mL DMEM by shaking for 30 s. Then, the tissues were cut into 0.5 cm segments, incubated in Hank's buffer containing 2.5 mM EDTA-Na 2 and 1 mM dithiothreitol (Sangon Biotech, China) at 37°C with rotation at 200 rpm for 30 min, filtered with a 70 µm strainer. Then the epithelial cells were pelleted by centrifugation at 500 × g for 5 min, purified with 20% Percoll (Cytiva, Sweden), and lysed in Trizol for RNA extraction and further qPCR analysis.
## Histopathology and immunohistochemistry
The jejunum and ileum of suckling mice were fixed in 4% paraformaldehyde for 24 h, washed with PBS twice, and then subjected to paraffin embedding and slicing at 5 µm thickness. The sections were stained with hematoxylin and eosin (Beyotime, China) and observed with an Olympus BX51 microscope, and the data were processed by CaseViewer 2.4 software. For immunohistochemistry analysis, the fixed tissues were dehydrated in 30% sucrose, embedded in Optimal Cutting Temperature compound, frozen with liquid nitrogen, and cryosliced at 10 µm thickness. The slices were air-dried, followed by antigen retrieval using 0.01 M citrate buffer (pH 6.0) at 100°C for 30 min, permeabilization in 0.5% Triton X-100 for 10 min, and blocking with 5% BSA (Beyotime, China) for 1 h at room temperature. Then, the slices were incubated with rabbit anti-RV VP7 polyclonal antibodies (CUSABIO, China) (1:200) overnight at 4°C, washed twice with PBS, incubated with FITC-conjugated goat anti-rabbit antibodies (ebioscience, USA) (1:500) for 1 h at room temperature and followed by 5 µg/mL DAPI (Sigma-Aldrich, USA) staining for 10 min. The double-stained slices were washed twice with PBS and mounted in Prolong Gold Antifade Reagent (CST, USA). Microscopic images were acquired with an EVOS M5000 system (Thermo Fisher Scientific, USA) using the built-in Celleste software (v.6.0). All raw image files were saved in TIFF format to preserve data integrity. Mini mal post-processing (image cropping and annotations) was applied through Adobe Illustrator 2019 (Adobe Systems, USA) to ensure readability. No alterations were made to raw data during this process.
## Statistical analysis
Data were presented as means ± SEM. Statistical analysis was performed with Prism 9.0 (GraphPad). Paired Student's t-test was used for comparisons between two matched groups. One-way analysis of variance (ANOVA) was used for comparisons between multiple groups. Two-way ANOVA was used for comparisons of grouped data. P-values <0.05 were considered significant (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001); ns, not significant, P > 0.05.
## RESULTS
## Antiviral activity of NVP-HSP990 in vitro
NVP-HSP990 is much smaller than and structurally different from the traditional HSP90 inhibitors GA and 17-AAG (Fig. S1). NVP-HSP990 showed much less cytotoxicity than GA and 17-AAG at 100 µM, with a CC 50 (concentration of cytotoxicity 50%) >100 µM on all the cells; in contrast, the CC 50 values of both GA and 17-AAG were <40 µM (Fig. 1A; Fig. S2, Table S2). NVP-HSP990 inhibited potent RV inhibition with IC 50 (half-maximal inhibitory concentration) values of 0.96 ± 0.13 nM (Wa strain) and 0.84 ± 0.17 nM (SA11 strain) in MA104 cells, 4.00 ± 1.28 nM (Wa strain) and 7.47 ± 4.84 nM (SA11 strain) in Caco-2 cells, and 4.69 ± 0.90 nM (Wa strain) and 3.85 ± 1.16 nM (SA11 strain) in HT-29 cells, which were all significantly smaller than those of GA and 17-AAG (Fig. 1B; Table S2). Therefore, the selectivity index (SI) of NVP-HSP990 in RV infection was much higher than that of GA or 17-AAG in all tested cell lines (Table S2).
To further validate the antiviral efficacy of NVP-HSP990 against RV, Caco-2 cells were infected with either the RV Wa or SA11 strain at a range of MOIs from 0.001 to 10. Robust inhibition of RV replication was consistently observed across all MOIs tested at 24 h p.i. (Fig. 2A). Nevertheless, pre-treatment of Caco-2 cells with NVP-HSP990 prior to RV infection (-2 ~ -1 h p.i.) did not significantly interfere with the establishment of RV infection or subsequent viral replication (Fig. S3A through C; Fig. 2B andC). NVP-HSP990 only modestly (~1-fold) reduced the final level of RV replication when administered during the course of RV infection (-1 ~ 0 h p.i.), suggesting that it is not highly effective at blocking RV entry into host cells; on the contrary, NVP-HSP990 administration after RV infection and during the whole process of cultivation (0 ~ 20 h p.i.) resulted in a highly significant, nearly 100-fold inhibitory effect, and exhibited a synergistic effect with NVP-HSP990 administration during the course of RV infection (-1 ~ 0 h p.i.) (Fig. 2B andC). Accordingly, the antiviral efficacy of NVP-HSP990 is governed by exposure duration, with longer exposure producing progressively greater suppression of RV replication (Fig. 2D).
qPCR analysis revealed a significant reduction of RNA synthesis of RV proteins VP2, VP6, NSP4, and NSP5 in RV-infected Caco-2 cells cultured with NVP-HSP990 (Fig. 2E), indicating that NVP-HSP990 significantly inhibited the transcription of the RV genome. Accordingly, RV structural proteins VP6 was remarkably reduced in RV-infected Caco-2 cells treated with NVP-HSP990 (Fig. 2F). IFA also showed potent inhibition of RV VP6 expression in RV-infected Caco-2 cells when treated with 100 nM NVP-HSP990 (Fig. 2G).
## NVP-HSP990 alters life state of host cells
RNA-seq analysis was performed to evaluate the impact of NVP-HSP990 treatment on the host cell transcriptome during RV infection. Principal component analysis revealed that both RV infection and NVP-HSP990 treatment were key factors influencing host gene expression patterns (Fig. S4A). As anticipated, RV infection triggered substantial tran scriptional changes in host cells; however, NVP-HSP990 addition mitigated these changes (Fig. S4B). RV infection induced transcriptional modulations across various KEGG pathways; however, NVP-HSP990 effectively reversed these changes except for specific pathways like the IL-17 and TNF signaling pathways, which are associated with innate immunity or inflammation (Fig. S4C through F).
Next, we focused on the direct effects of NVP-HSP990 on host cells. We found that NVP-HSP990 induced significant transcriptional alterations in various genes, regardless of whether the host cells were mock-infected or RV-infected (Fig. 3A). These alterations were primarily associated with cell cycle, DNA replication, and various cancer-, signaling-, or metabolism-related pathways in host cells (Fig. S5A). Specifically, among these genes, 112 were upregulated and 287 were downregulated by NVP-HSP990 across all mock-, Wa-, and SA11-infected host cells (Fig. 3B; Tables S3 andS4). The upregulated genes were significantly enriched (Q value <0.05) in 6 KEGG pathways such as protein processing in the endoplasmic reticulum and antigen processing and presentation, while the downre gulated genes were significantly enriched (Q value <0.05) in 14 KEGG pathways including cell cycle, DNA replication, MAPK signaling pathway, and various cancer-related path ways (Fig. 3C andD). Notably, 75 genes were upregulated and 96 genes were downregu lated by NVP-HSP990 in both Wa-and SA11-infected host cells but not in mock-infected cells (Fig. 3B; Tables S5 andS6). These upregulated genes were nominally enriched (Q value >0.05) in metabolic processes, and the downregulated genes were significantly enriched (Q value <0.05) in nine KEGG pathways including inflammatory responses (Fig. 3E andF). These findings revealed a selective regulatory pattern, characterized by suppression of signaling and inflammation-related pathways, through which NVP-HSP990 mitigated the perturbations induced by RV in host cells.
## NVP-HSP990 modulated MAPK signaling pathway and mitigated disruption of tight junctions in RV infection
MAPK signaling pathway is mainly composed of ERK, JNK, and p38 signals and plays important roles in viral infections and host antiviral immunity (28,29). In fact, the MAPK signaling pathway is reported to be activated and critical for RV replication (20,30,31). We found that NVP-HSP990 significantly upregulated some but downregulated more MAPK-related genes (Fig. 3C andD; Fig. S5B). The upregulated MAPK-related genes included some heat shock proteins (HSPB1, HSPA1A, HSPA1B, and so on) and some cytokines/cytokine receptors (such as PGF, PDGFRA, and PDGFRB), while the downregulated genes included JUN, JUND, MYC, DUSP2, DUSP5, and so on, which are involved in signal transduction (Fig. S5B and S6, Tables S7 to S9). These results indicated that NVP-HSP990 mainly inhibited MAPK signaling pathway in RV infection. Notably, NVP-HSP990 significantly inhibited the MAPK activation (especially ERK1/2 and p38 MAPK) in Caco-2 and HT-29 but not MA104 cells (Fig. 4A; Fig. S7). These results indicated that NVP-HSP990 might specifically inhibit the MAPK activation in intestinal Full-Length Text epithelial cells, which are natural target cells of RV, thus favoring its anti-RV effect in vivo. Moreover, NVP-HSP990 inhibition of MAPK signaling pathway is not dependent on the inhibition of RV replication, because it also potently inhibited MAPK activation caused by non-infectious factors such as drug stimulation (Fig. 4B). Tight junctions are crucial for the survival and function of mature intestinal epithelial cells, and destruction of tight junctions causes intestinal inflammation and disorders (32,33). Formation of tight junctions is modulated by intracellular signaling pathways including MAPK (34,35). As NVP-HSP990 significantly inhibited the MAPK signaling pathway in intestinal cells, we wondered whether NVP-HSP990 also effectively mitigates RV-induced disruption of tight junctions. We found that NVP-HSP990 facilitated the expression of tight junction-associated proteins such as ZO-1, ZO-2, and claudin-1 (Fig. 4C). More importantly, NVP-HSP990 effectively restored structural disruption of tight junctions in RV infection (Fig. 5). Therefore, this function of NVP-HSP990 would benefit the protection of intestinal epithelium integrity.
## NVP-HSP990 alleviated RV infection in suckling mice
As NVP-HSP990 possessed potent anti-RV effects in vitro as shown above, we wondered whether it was effective for RV control in vivo. To this end, a series of doses of NVP-HSP990 were orally administered to suckling mice infected with the RV SA11 strain, and the diarrhea scores were evaluated daily. We found that the average diarrhea scores at 24 h p.i. (day 1) decreased along with the increase of NVP-HSP990 doses, and the reduction was especially remarkable at 1,000 µg/kg (Fig. 6A). NVP-HSP990 inhibited RV diarrhea occurrence with an ED 50 (50% effective dose) value of 142.3 µg/kg, and it reduced diarrhea score with an ED 50 value of 135.5 µg/kg (Fig. 6B andC). Treatment with 1 mg/kg NVP-HSP990 (once) did not hinder body growth of RV-infected suckling mice but significantly alleviated their diarrhea (Fig. 6D through F). Accordingly, infectious RV particles and viral antigens in intestinal contents, as well as transcription of RV VP6 protein in intestinal epithelial cells, were remarkably reduced in jejunum and ileum by administration of NVP-HSP990 compared to controls, though there were no significant differences in duodenum (Fig. 6G through I). We next assessed the protective effect of NVP-HSP990 against RV-induced ileum lesions. NVP-HSP990 treatment alone produced no appreciable pathological changes in the ileum of mock-infected mice. In contrast, RV infection induced pronounced foamy degeneration in the ileal epithelium. This pathological change was markedly attenuated by administration of NVP-HSP990 (Fig. 6J). These findings demonstrate that NVP-HSP990 can effectively alleviate RV infection in vivo.
Furthermore, we evaluated the therapeutic efficacy of NVP-HSP990 by administering it orally to suckling mice after diarrhea onset following infection with the mouse-derived RV strain EDIM. After 3 consecutive days of treatment, histopathological and immuno histochemical analyses revealed a marked reduction in both foamy degeneration and viral infection in the small intestinal epithelial cells of the NVP-HSP990-treated group compared to the untreated controls. However, no appreciable difference in the extent of infection or tissue damage was observed between the NVP-HSP990-and ribavirintreated groups (Fig. S8 and S9; Fig. 7A). Viral antigen (VP6 and NSP4) expression in intestinal epithelial cells was significantly reduced in the NVP-HSP990 group compared to the untreated group, and this reduction was comparable to that achieved by the conventional antiviral drug ribavirin, although no significant differences were observed in the duodenum among the groups (Fig. 7B andC). Additionally, RV viral antigens in colonic contents were significantly lower in the NVP-HSP990 group than in either the untreated or ribavirin-treated groups (Fig. 7D).
To evaluate the longer therapeutic effect of NVP-HSP990 on RV diarrhea, suckling mice were treated with the drugs continuously for 5 days after the onset of diarrhea, and their body weights and diarrhea scores were monitored during the process and for another 5 days after drug administration. The results showed that NVP-HSP990 did not adversely affect weight gain in EDIM-infected suckling mice (Fig. 7E). However, NVP-HSP990 significantly reduced diarrhea scores compared to the untreated group, and it also demonstrated a superior therapeutic effect to ribavirin (Fig. 7F andG).
## DISCUSSION
Cell stress induced by viral infection results in enhanced expression of various heat shock proteins, including HSP40, HSP70, HSP90, and so on, and viral replication is directly or indirectly dependent on one or more of these HSPs (36,37). Many viruses, including RV, require the participation of HSP90 in their replication (12-14, 19, 38, 39), so it sounds promising to develop broad-spectrum, HSP90-targeting antiviral drugs, which would be less affected by viral mutations and more stable in antiviral effect. Therefore, small-molecule inhibitors of HSP90 are currently popular in antiviral drug discovery (21,(40)(41)(42). Unfortunately, considering their unsatisfying antiviral efficacy and toxicity, there have been no HSP90 inhibitors in clinical use for viral control until now. The design of small-molecule HSP90 inhibitors is commonly based on the inhibitory mechanism of the natural HSP90 inhibitor GA or Radicicol, which targets the ATP pocket of N-terminal domain of HSP90 to block the ATPase activity. With the increasing understanding of HSP90 subtypes and the interactions between HSP90 and its client proteins, a series of novel HSP90 inhibitors have emerged, which selectively inhibit HSP90 subtypes or specifically block the interaction between HSP90 and client proteins, so as to improve their inhibitory efficacy and reduce their cytotoxicity (43,44). NVP-HSP990 is much smaller than and structurally distinct from GA and 17-AAG, and it targets the N-termi nals of both HSP90α and HSP90β with very high specificity (IC 50 : 0.6 nM and 0.8 nM, respectively). NVP-HSP990 is orally bioavailable and brain-penetrating and has been tested in the treatment of various tumors and Huntington's disease (26,(45)(46)(47). However, there have been few reports on the role of NVP-HSP990 in controlling viral infections, though a recent in silico study predicted that NVP-HSP990 might alleviate COVID-19 symptoms through anti-inflammatory effects on SARS-CoV-2-infected lung cells (48). In this study, we showed that NVP-HSP990 inhibited RV infection with high SI in vitro and effectively alleviated RV diarrhea in suckling mice. Our findings suggest that NVP-HSP990 may be a promising antiviral drug candidate for RV infection, and that targeting HSP90 should remain to be a promising strategy for antiviral drug development.
Viral infection usually triggers the modulation of host signaling pathways, resulting in transcriptional changes of various host genes to remodel host life systems for better viral replication (49)(50)(51). On the other hand, viral infection inevitably triggers the activa tion of immunity/inflammation-associated signaling pathways for host antiviral innate immunity. In this study, although RV infection triggered significant alterations in host gene transcription across various signaling pathways, NVP-HSP990 effectively mitigated these impacts on host cells. Notably, the activation of immune/inflammatory signaling pathways (such as IL-17 and TNF signaling pathways) was still significant under NVP-HSP990 addition. These results indicate that NVP-HSP990 not only potently inhibits RV replication but also preserves the host's ability to mount an innate immune response and trigger inflammation. This dual action facilitates complete viral clearance and represents a significant advantage of NVP-HSP990 as a potential antiviral candidate.
Upon NVP-HSP990 treatment, the most affected pathways in Caco-2 cells included cell cycle, DNA repair, and cancer-related pathways, which is consistent with its role as an effective anti-tumor drug. To be noted, MAPK signaling pathway was negatively regulated by NVP-HSP990 in both non-infected and RV-infected Caco-2 cells, suggesting that MAPK signaling pathway is sensitive to HSP90 inhibition by NVP-HSP990, aligning with the report that proper folding of RAF kinases of the RAS-MAPK signaling pathway relies on the interaction with HSP90 (52). RV infection induces MAPK activation, including the activation of ERK, JNK, and p38, among which ERK and p38 are critical for RV replication, while JNK is critical for host IFN-β production (30,31,53,54). In this study, we demonstrated that NVP-HSP990 potently suppressed the activation of ERK and p38 in RV-infected intestinal epithelial cells like Caco-2 and HT-29 but not in renal epithelial cells MA104, although all of them are susceptible to RV infection. Therefore, NVP-HSP990 might specifically inhibit MAPK activation in intestinal epithelial cells, which should facilitate its anti-RV efficacy in vivo. To be noted, NVP-HSP990 inhibition of the MAPK signaling pathway did not depend on RV inhibition, as it also inhibited the basal or drug-induced MAPK activation in intestinal epithelial cells.
Disruption of tight junctions leads to the necrosis or anoikis of mature intestinal epithelial cells, which is a key step in the pathogenesis of intestinal infectious diseases (55). Many studies show that activation of the p38 signaling pathway disrupts cellu lar tight junctions (35,56). In this study, we found that RV infection activated p38 signaling and disrupted cellular tight junctions in Caco-2 cells, which were consistent with previous reports (31,57). Nevertheless, NVP-HSP990 significantly inhibited p38 activation and mitigated tight junction disruption in RV-infected Caco-2 cells, endow ing NVP-HSP990 with the function of protecting intestinal epithelium in RV infection. The blood-brain barrier (BBB), primarily composed of tight junctions between brain microvascular endothelial cells, plays a crucial role in preventing pathogens from entering the brain. Pathogens or toxins can invade the brain by disrupting tight junctions of brain microvascular endothelial cells, thereby increasing BBB permeability (58). Numerous reports demonstrate the presence of viral particles and antigens in the brain during RV infection (5), suggesting a potential compromise of BBB integrity. Therefore, the protective effect of NVP-HSP990 on tight junctions could mitigate BBB disruption during RV infection, potentially lowering the risk of encephalopathy development.
Besides gastrointestinal symptoms such as diarrhea and vomiting, severe RV infections often affect important organs such as the brain, heart, kidneys, and liver; for example, RV infection usually causes epileptic seizures in children, although the mechanism is currently unclear (5). Conventional antiviral drugs such as nucleoside analogs and type I interferon are usually not included in the treatment of RV diarrhea due to adverse side effects (8,9), thus symptomatic treatment is usually the main choice, which sometimes leads to systemic RV infection and even mortality due to persistent infection. Therefore, the treatment of RV infection should not merely focus on sympto matic treatment; timely antiviral treatment shall be important to reduce complications and mortality. Here, we demonstrated that besides potently inhibiting RV infection in vitro, NVP-HSP990 also effectively suppressed intestinal RV infection and alleviated diarrhea in BALB/c suckling mice, rendering it a promising candidate for an anti-RV drug. In addition, given that NVP-HSP990 is able to penetrate BBB and very low dose (0.1 mg/ kg-0.2 mg/kg) of NVP-HSP990 shows potent antiepileptic activity (59,60), it could also represent a promising therapeutic strategy for RV encephalopathy whether mediated by direct RV neuroinvasion or potential unknown mechanisms.
Admittedly, in many regions where rotavirus remains a significant issue, clinical diagnostic tests are often unavailable, limiting the ability to confirm rotavirus as the cause of acute gastroenteritis. Moreover, rotavirus infections are typically acute and self-limiting (61), which complicates the use of antiviral treatments. However, the discovery of a potent antiviral compound like NVP-HSP990 remains highly valuable, as it opens the door to exploring more targeted clinical applications. Firstly, NVP-HSP990 could be particularly beneficial in chronic infection scenarios. For example, in immuno compromised individuals-such as those undergoing chemotherapy, organ transplant recipients, or those with primary immunodeficiencies-rotavirus infections can lead to severe, protracted enteritis that may become life-threatening (62). In these cases of chronic or persistent infection, a potent oral antiviral like NVP-HSP990 could offer a critical treatment option. Secondly, severe acute rotavirus infections can sometimes lead to multi-organ complications and be fatal (63). Timely administration of antiviral drugs may help alleviate the severity of intestinal infections and inhibit possible extraintestinal spread, thus reducing the risk of mortality and associated complications. Additionally, future research could investigate NVP-HSP990's potential as post-exposure prophylaxis, particularly for controlling outbreaks in closed environments such as pediatric wards or childcare centers.
## Conclusion
In this study, we demonstrated that the small-molecule HSP990 inhibitor NVP-HSP990 robustly blocked RV replication with low cytotoxicity in vitro, mitigated RV impacts on the host transcriptome, and repressed RV-induced MAPK activation and tight junction disruption in intestinal cells (Fig. 8), which finally contributed to effective alleviation of RV diarrhea in suckling mice. These findings, coupled with excellent oral bioavailability and brain-penetrating ability, make NVP-HSP990 a novel, promising candidate antiviral drug for alleviating RV infection.
## References
1. Matthijnssens, Attoui, Bányai et al. (2022) "ICTV virus taxonomy profile: Spinareoviridae 2022" *J Gen Virol*
2. Hasan, Nasirudeen, Ruzlan et al. (2021) "Acute infectious gastroenteritis: the causative agents, omics-based detection of antigens and novel biomarkers" *Children (Basel)*
3. Hellysaz, Hagbom (2021) "Understanding the central nervous system symptoms of rotavirus: a qualitative review" *Viruses*
4. Lorrot, Vasseur (2007) "How do the rotavirus NSP4 and bacterial enterotoxins lead differently to diarrhea?" *Virol J*
5. Dian, Sun, Zhang et al. (2021) "Rotavirus-related systemic diseases: clinical manifestation, evidence and pathogenesis" *Crit Rev Microbiol*
6. Cates, Tate, Parashar (2022) "Rotavirus vaccines: progress and new developments" *Expert Opin Biol Ther*
7. Cohen, Platts-Mills, Nakamura et al. (2022) "Aetiology and incidence of diarrhoea requiring hospitalisation in children under 5 years of age in 28 low-income and middle-income countries: findings from the global pediatric diarrhea surveillance network" *BMJ Glob Health*
8. Sun, Lin, Yang et al. (2020) "Acupoint application for rotavirus diarrhea in infants and children: a protocol for systematic review and meta analysis" *Medicine (Baltimore)*
9. Jiang, Tang, Song et al. (2023) "Advances in the develop ment of antivirals for rotavirus infection" *Front Immunol*
10. Hoter, El-Sabban, Naim (2018) "The HSP90 family: structure, regulation, function, and implications in health and disease" *Int J Mol Sci*
11. Powers, Workman (2006) "Targeting of multiple signalling pathways by heat shock protein 90 molecular chaperone inhibitors" *Endocr Relat Cancer*
12. Hu, Seeger (1996) "Hsp90 is required for the activity of a hepatitis B virus reverse transcriptase" *Proc Natl Acad Sci*
13. Okamoto, Nishimura, Ichimura et al. (2006) "Hepatitis C virus RNA replication is regulated by FKBP8 and Hsp90" *EMBO J*
14. Naito, Momose, Kawaguchi et al. (2007) "Involvement of Hsp90 in assembly and nuclear import of influenza virus RNA polymerase subunits" *J Virol*
15. Yadav, Singh, Kumar et al. (2022) "An insight to heat shock protein 90: a remedy for multiple problems" *CPD*
16. Rico, Perez, Hernandez et al. (2021) "Cell surface heat shock protein-mediated entry of tumor cell-adapted rotavirus into Full-Length Text Journal of Virology"
17. "U-937 cells" *Folia Microbiol (Praha)*
18. Pérez, Rico, Guerrero et al. (2021) "Role of heat-shock proteins in infection of human adenocarcinoma cell line MCF-7 by tumoradapted rotavirus isolates" *Colomb Med*
19. Rico, Perez, Guerrero et al. (2020) "Implication of heat shock proteins in rotavirus entry into Reh cells" *Acta Virol*
20. Dutta, Bagchi, Chatterjee et al. (2009) "The molecular chaperone heat shock protein-90 positively regulates rotavirus infectionx" *Virology (Auckl)*
21. Huang, Liao, Zhou et al. (2020) "Antiviral activities of resveratrol against rotavirus in vitro and in vivo" *Phytomedicine*
22. Wang, Wang, Li et al. (2017) "HSP90: a promising broad-spectrum antiviral drug target" *Arch Virol*
23. Xie, Zhang, Li et al. (2023) "Small-molecule dual inhibitors targeting heat shock protein 90 for cancer targeted therapy" *Bioorg Chem*
24. Gorska, Popowska, Sielicka-Dudzin et al. (2012) "Geldanamycin and its derivatives as Hsp90 inhibitors" *Front Biosci*
25. Kim, Keum, Pae (2013) "Discovery and development of heat shock protein 90 inhibitors as anticancer agents: a review of patented potent geldanamycin derivatives" *Expert Opin Ther Pat*
26. Mcbride, Levine, Xia et al. (2014) "Design, structure-activity relationship, and in vivo characterization of the development candidate NVP-HSP990" *J Med Chem*
27. Menezes, Taverna, Jensen et al. (2012) "The novel oral Hsp90 inhibitor NVP-HSP990 exhibits potent and broad-spectrum antitumor activities in vitro and in vivo" *Mol Cancer Ther*
28. Boshuizen, Reimerink, Korteland-Van Male et al. (2003) "Changes in small intestinal homeostasis, morphology, and gene expression during rotavirus infection of infant mice" *J Virol*
29. Chander, Kumar, Khandelwal et al. (2021) "Role of p38 mitogen-activated protein kinase signalling in virus replication and potential for developing broad spectrum antiviral drugs" *Rev Med Virol*
30. Macmicking (2022) "Rewiring the logic board of IFN signaling" *Sci Signal*
31. Soliman, Seo, Kim et al. (2018) "Activation of PI3K, Akt, and ERK during early rotavirus infection leads to V-ATPasedependent endosomal acidification required for uncoating" *PLoS Pathog*
32. 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*
33. Zihni, Mills, Matter et al. (2016) "Tight junctions: from simple barriers to multifunctional molecular gates" *Nat Rev Mol Cell Biol*
34. Furuse, Takai (2021) "Recent advances in understanding tight junctions" *Fac Rev*
35. Rungrasameviriya, Santilinon, Atichartsintop et al. (2024) "Tight junction and kidney stone disease" *Tissue Barriers*
36. Cong, Kong (2020) "Endothelial tight junctions and their regulatory signaling pathways in vascular homeostasis and disease" *Cell Signal*
37. Bolhassani, Agi (2019) "Heat shock proteins in infection" *Clin Chim Acta*
38. Zhang, Yu (2022) "Heat shock proteins and viral infection" *Front Immunol*
39. Evers, Chao, Zhang et al. (2012) "17-allylamino-17-(demethoxy)geldanamycin (17-AAG) is a potent and effective inhibitor of human cytomegalovirus replication in primary fibroblast cells" *Arch Virol*
40. Wang, Liu, Luo et al. (2017) "Geldanamycin reduces acute respiratory distress syndrome and promotes the survival of mice infected with the highly virulent H5N1 influenza virus" *Front Cell Infect Microbiol*
41. Koren J 3rd, Blagg (2020) "The right tool for the job: an overview of Hsp90 inhibitors" *Adv Exp Med Biol*
42. Serwetnyk, Blagg (2021) "The disruption of protein-protein interactions with co-chaperones and client substrates as a strategy towards Hsp90 inhibition" *Acta Pharm Sin B*
43. Solit, Chiosis (2008) "Development and application of Hsp90 inhibitors" *Drug Discov Today*
44. Costa, Raghavendra, Penido (2020) "Natural heat shock protein 90 inhibitors in cancer and inflammation" *Eur J Med Chem*
45. Li, Wang, You et al. (2020) "Heat shock protein 90 inhibitors: an update on achievements, challenges, and future directions" *J Med Chem*
46. Jackrel, Shorter (2011) "Shock and awe: unleashing the heat shock response to treat huntington disease" *J Clin Invest*
47. Oikonomou, Valsecchi, Quadri et al. (2023) "High-through put screening as a drug repurposing strategy for poor outcome subgroups of pediatric B-cell precursor acute lymphoblastic leukemia" *Biochem Pharmacol*
48. Spreafico, Delord, Mattos-Arruda et al. (2015) "A firstin-human phase I, dose-escalation, multicentre study of HSP990 administered orally in adult patients with advanced solid malignancies" *Br J Cancer*
49. Oh, Adnan, Cho (2021) "Drug-repurposing against COVID-19 by targeting a key signaling pathway: an in silico study" *Med Hypotheses*
50. Pleschka (2008) "RNA viruses and the mitogenic Raf/MEK/ERK signal transduction cascade" *Biol Chem*
51. Wu, Zhang, Li et al. (2021) "The regulation of integrated stress response signaling pathway on viral infection and viral antagonism" *Front Microbiol*
52. Akbay, Shmakova, Vassetzky et al. (2020) "Modulation of mTORC1 signaling pathway by HIV-1" *Cells*
53. Mitra, Ghosh, Gayen et al. (2016) "Bipartite role of heat shock protein 90 (Hsp90) keeps CRAF kinase poised for activation" *J Biol Chem*
54. Jafri, Donnelly, Mcneal et al. (2007) "MAPK signaling contributes to rotaviral-induced cholangiocyte injury and viral replication" *Surgery*
55. He, Zhou, Fu et al. (2012) "Cyclophilin A inhibits rotavirus replication by facilitating host IFN-I production" *Biochem Biophys Res Commun*
56. Paradis, Bègue, Basmaciyan et al. (2021) "Tight junctions as a key for pathogens invasion in intestinal epithelial cells" *Int J Mol Sci*
57. Kang, Zhou, Zhu et al. (2022) "Ginsenoside Rg1 mitigates porcine intestinal tight junction disruptions induced by LPS through the p38 MAPK/NLRP3 inflammasome pathway" *Toxics*
58. (2026) *Full-Length Text Journal of Virology*
59. Obert, Peiffer, Servin (2000) "Rotavirus-induced structural and functional alterations in tight junctions of polarized intestinal Caco-2 cell monolayers" *J Virol*
60. Wei, Jiang, Wang et al. (2024) "Brain endothelial GSDMD activation mediates inflammatory BBB breakdown" *Nature*
61. Sha, Chen, Deng et al. (2020) "Hsp90 inhibitor HSP990 in very low dose upregulates EAAT2 and exerts potent antiepileptic activity" *Theranostics*
62. Labbadia, Cunliffe, Weiss et al. (2011) "Altered chromatin architecture underlies progressive impairment of the heat shock response in mouse models of Huntington disease" *J Clin Invest*
63. Parashar, Hummelman, Bresee et al. (2003) "Global illness and deaths caused by rotavirus disease in children" *Emerg Infect Dis*
64. Ribeiro, Ferreira, Arrabalde et al. (2015) "Prevalence of adenovirus and rotavirus infection in immunocompro mised patients with acute gastroenteritis in Portugal" *World J Virol*
65. Tarris, Belliot, Callier et al. (2019) "Pathology of rotavirus-driven multiple organ failure in a 16-month-old boy" *Pediatr Infect Dis J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724366&blobtype=pdf | # The H4K20-mono-methyltransferase SETD8 promotes global accessibility of infecting herpes simplex virus genomes
Jesse Arbuckle, Andy Yanez, Syeda Baksh, Tovah Markowitz, Jodi Vogel, Alison Mcbride, Thomas Kristie
## Abstract
Epigenetic modulation of herpes simplex virus (HSV) immediate early (IE) genes is a critical parameter governing lytic infection, latency, and viral reactivation. Multiple factors, including epigenetic complexes associated with the cellular transcrip tional coactivator HCF-1, modulate the state of HSV-1 chromatin. Although some aspects of HSV-1 chromatin biology have been elucidated, many epigenetic factors controlling HSV-1 chromatin accessibility and transcriptional regulation remain unknown. To identify novel epigenetic regulators of the initial stage of HSV-1 infection, an epigenetic chemical probe library was screened for molecules that impacted viral IE gene expression. This screen identified several epigenetic inhibitors that modulated IE expression. Notably, UNC0379, an inhibitor of the histone H4K20 mono-methyltransferase SETD8, potently suppressed HSV-1 IE gene transcription during lytic infection. UNC0379 treatment also repressed HSV-1 reactivation in a mouse ganglia explant model, and topical applica tion suppressed primary ocular infection in vivo. Importantly, SETD8 inhibition resul ted in enhanced heterochromatin suppression and reduced viral genome accessibility, as evident by increased repressive H3K9me3 marks and decreased ATAC-seq signal across the HSV-1 genome. Reduced IE transcription was concomitant with disruption of H4K20me1 deposition and impaired recruitment of HCF-1 and RNA polymerase II (RNAPII). These findings are consistent with the established roles of SETD8 in promot ing chromatin accessibility through H4K20me1 deposition and in regulating transcrip tional elongation, suggesting that SETD8 may facilitate multiple steps in HSV-1 IE gene expression. Overall, these results identified SETD8 as an essential epigenetic modulator of HSV-1 chromatin accessibility and represent a promising novel antiviral therapeutic target.IMPORTANCE Herpes simplex virus is a highly prevalent human pathogen, with persistence of the viral genome in sensory neurons serving as a reservoir for recurrent disease and viral shedding. Epigenetic pharmaceuticals have proven to be valuable tools in elucidating the chromatin landscape of the viral genome and its impact on viral gene expression. With the use of a chemical library screen, the histone methyltransferase SETD8 was identified as a key factor required for promoting accessibility of the her pesvirus genome to transcriptional regulators. Importantly, pharmacological inhibition of SETD8 suppressed herpes simplex virus lytic infection, reduced viral reactivation in sensory neurons, and when applied topically, inhibited primary ocular infection of mice. Collectively, these findings establish SETD8 as a critical regulator of viral gene expression during lytic infection and the initiation of reactivation from latency. These results highlight SETD8 as a potential novel target for antiviral therapy.
H erpes simplex virus (HSV) has a dual replication cycle. During the primary lytic phase, infection of epithelial cells at mucosal surfaces results in a lytic gene cascade and subsequent shedding of viral progeny to the termini of sensory neurons. Seconda rily, HSV establishes a quiescent latent infection in sensory neurons, where the viral genome persists as a circular DNA episome for the life of the host. During latency, lytic gene transcription is minimal, while the viral gene product latency-associated transcript (LAT) is highly expressed (1,2). LAT is a long noncoding RNA that contributes to the establishment of latency, suppressing neuronal apoptosis, and localizing genomes to the nuclear periphery to silence latent genomes (2)(3)(4)(5)(6)(7). Latency is not static; various stress mediators can stimulate HSV to reenter the lytic cycle and shed viral particles at the initial site of infection.
Recurrent cycles of HSV latency-reactivation cause diseases of the oral and genital epithelia, as well as more severe outcomes such as herpes simplex encephalitis (HSE) and herpetic keratitis. The mortality rate of HSE in the absence of antiviral intervention is 70%, while herpetic keratitis is the primary source of viral-mediated blindness in the United States (8)(9)(10). HSV infections that occur in utero or during childbirth can result in HSE and developmental morbidities. Current antiviral therapies employ nucleoside analogs, such as acyclovir (ACV), to target the viral DNA polymerase. Antiviral treatment is administered during episodic and recurrent symptoms; however, drug resistance can arise in immunocompromised individuals (11)(12)(13)(14).
Transcriptional initiation of HSV-1 immediate early (IE) genes is a key determinant for the progression of lytic infection. Upon entry into the nucleus, the HSV-1 genome is rapidly assembled into unstable nucleosome structures by the cellular epigenetic machinery containing histone chaperones and chromatin remodelers (15)(16)(17)(18)(19)(20). Concur rently, transcription of IE genes is initiated by the viral transactivator VP16 in cooperation with an extended network of cellular transcription factors (19). The inherent dependence on IE gene transcription is demonstrated by studies analyzing IE deletion mutants (21,22). These studies showed that native functions of the lytic cascade of IE genes, Early (E) genes, DNA replication, Late (L) gene expression, and viral assembly are attenuated when IE gene expression is disrupted.
Modulation of the chromatinized HSV-1 genome and components of the cellular epigenetic machinery control the fundamental activities of the lytic, latency, and reactivation cycles. The HSV-1 genome is initially assembled into heterochromatic structures containing repressive histone signatures (H3K9me3 and H3K27me2) (23)(24)(25)(26)(27)(28). These modifications serve as dynamic binding platforms for chromatin readers to mediate sequential chromatin condensation. One key factor central to circumventing this initial repression is the cellular transcriptional coactivator Host Cell Factor-1 (HCF-1). HCF-1 serves as a scaffold that bridges multiple protein complexes to the promoters of IE genes (19,20,29). HCF-1 interacts with histone demethylases (JMJD2 and LSD1) that limit the accumulation of repressive H3K9-methylation and histone methyltransferases (SETD1A/MLL) that deposit active H3K4-methylation (24,26,(30)(31)(32). Adding further complexity, HCF-1 also recruits components of the super elongation complex (SEC), thereby facilitating transcriptional elongation of IE genes, a critical step for HSV-1 lytic infection and latency reactivation (33,34).
Chromatin accessibility and nucleosome positioning across the HSV-1 genome are key determinants of the HSV-1 replication program. It has also been established that HCF-1 is essential for IE gene expression during lytic infection as well as for in vivo reactivation from latency (35,36). Mass spectrometry and yeast two-hybrid studies have defined HCF-1-containing complexes that are essential for IE gene transcription (34,37). However, the full complement of components and mechanisms involved in the dynamic modulation of chromatin during IE gene transcriptional initiation remains largely unknown.
To identify additional epigenetic factors that modulate the initial events of IE gene transcriptions, we screened an epigenetic chemical probe library in primary human fibroblast cells infected with HSV-1 for 1.5 hours (h). This screen identified several epigenetic modulators. In particular, the chemical probe UNC0379 (38), which targets the methyltransferase SETD8 (KMT5A, SET8, and PR-Set7), potently suppressed IE gene transcription. SETD8-mediated mono-methylation of H4K20 is known to stimu late cellular gene expression by promoting chromatin accessibility and transcription elongation (Fig. 1) (39)(40)(41). Using UNC0379, we investigated the role of SETD8 in HSV-1 IE transcription and found that UNC0379 suppressed IE expression during lytic infection and blocked the initiation of latency reactivation in sensory ganglia. Suppression of IE expression was concomitant with reduced HSV-1 genome accessibility and diminished recruitment of HCF-1 and RNAPII. We propose a model in which SETD8-mediated H4K20-monomethylation is essential for promoting HSV-1 genome accessibility and limiting heterochromatin accumulation. Most significantly, in vivo topical application of UNC0379 suppressed HSV-1 yields in an ocular mouse model, identifying SETD8 as a potential novel therapeutic target.
## RESULTS
## Epigenetic inhibitor screen revealed requirement for SETD8 in HSV-1 IE gene transcription and productive infection
Chromatin character and accessibility of the HSV-1 genome influences the progression of lytic infection (15, 16, 19, 20, 23-26, 29-32, 35, 36). Similarly, modulation of viral chromatin in sensory neurons plays regulatory roles in latency and viral reactivation. Although some aspects of HSV-1 chromatin biology have been elucidated, many of the characteristics and relevant chromatin modulation machinery remain unknown. Therefore, we screened a curated epigenetic inhibitor library for impacts on HSV-1 IE gene transcription. This library, obtained from the National Center for Advancing Translational Sciences (NCATS) and supplemented with additional compounds from commercial sources, comprised 48 compounds targeting a select group of chromatin readers, chromatin remodelers, methyltransferases, and demethylases, as listed in Table S1. In this screen, HFF fibroblast cells were pretreated for 4 h with epigenetic inhibitors or vehicle control and infected with HSV-1 for 1.5 h. Inhibitors were titrated across 3 to 7 dilutions, ranging from 0.01 µM to 25 µM, as defined by the literature. HSV-1 IE gene ICP27 transcript levels were determined by RT-QPCR and normalized to cellular controls (SP1 or GAPDH).
Several epigenetic modulators of HSV-1 IE transcription were identified, including SGC-707 (targeting PRMT3), TP-064 (targeting PRMT4), TC-E5003 (targeting PRMT1), MS023 (targeting PRMT1, -3, -4, -6, and -8) and UNC0379, a chemical probe targeting the H4K20 mono-methyltransferase SETD8 (Fig. 2; Table S2) (38,(42)(43)(44)(45). UNC0379 was prioritized for further investigation due to its profound impact on ICP27 transcription, which was comparable to that of the previously established JMJD2 inhibitor ML324 and shown to suppress HSV-1 IE gene expression and latency reactivation (26). UNC0379 potently suppressed HSV-1 IE gene transcription and protein levels in a dose-dependent manner in both HFF and MRC-5 cells (Fig. 3A through C; Fig. S1 andS2). There was no apparent cytotoxicity at the evaluated concentrations or significant impact on cellular control genes (Fig. 3A through C; Fig. S3). Importantly, this suppression was not due to inhibition of viral entry as HSV-1 DNA levels in nuclear and total cell fractions were equivalent between UNC0379 and vehicle treatment (Fig. 3D). Rather than pretreating cells at 4 h prior to infection, treatment of cells at 1 hour post-infection (hpi) sufficiently reduced IE transcript levels (Fig. 3E). We also established that the suppression of HSV-1 was not MOI-dependent as increased viral input resulted in similar levels of inhibition (Fig. 3F). Additionally, treatment with NSC663284, a second compound targeting SETD8 (46), also suppressed IE gene transcription in a dose-dependent manner (Fig. 3C; Fig. S2 andS4).
To determine if SETD8 inhibition would suppress viral yields, HFF cells were pretrea ted with vehicle, SETD8 inhibitor UNC0379, or acyclovir (ACV, viral DNA polymerase inhibitor) for 12 h (Fig. 3G). Here, 8 µM UNC0379 treatment resulted in a 2.6 log reduction in viral yields as compared to vehicle and was nearly as effective as 100 µM ACV in plaque assays. Our results also indicated SETD8 inhibition reduced viral spread when cells were HSV-1-infected for 8 h to allow one replication cycle to occur before UNC0379 treatment for an additional 12 h (Fig. 3H). It was also apparent that suppression was not restricted to the alpha herpesvirus subfamily. UNC0379 treatment suppressed the transcription of UL123 (IE) and representative E and L gene expressions for the beta herpesvirus HCMV (Fig. S5) (47).
## UNC0379 suppressed HCF-1 coactivator complex recruitment to viral IE promoter/enhancer regions
The transition from heterochromatin to accessible euchromatin is essential for IE gene transcription. One particularly important component in this transition is the cellular transcriptional coactivator HCF-1 and its dynamic modulation of the chromatinized HSV-1 genome. This, in part, is mediated through HCF-1 recruitment of histone demethy lases JMJD2 and LSD1 that limit the accumulation of repressive H3K9 methylation at IE promoters (24,26,32). Given that SETD8 inhibition reduced IE gene expression and viral yields, we employed chromatin immunoprecipitation (ChIP) to assess the impact on the HSV-1 chromatin. We found that UNC0379 treatment reduced HCF-1 recruitment to the ICP0 and ICP4 IE promoter/enhancer regions (Fig. 4A). Consistent with reduced viral IE transcription and HCF-1 recruitment, there was a parallel decrease in RNAPII recruitment (Fig. 4B). Importantly, total histone H3 levels and the heterochromatin mark H3K9me3 were increased in all classes of representative viral genes (IE = ICP0, ICP4; E = UL29; L = UL44) in cells treated with UNC0379 (Fig. 4C).
It is important to emphasize that IE expression is essential for progression of the lytic cascade (2,(19)(20)(21)(22). The inhibition of SETD8 promoted heterochromatin formation across Full-Length Text all classes of representative viral genes (Fig. 4C), while RNAPII recruitment was decreased at IE promoters, as determined by ChIP-QPCR (Fig. 4B). To further characterize RNAPII distribution across the HSV-1 genome, ChIP-seq was completed on HFF cells treated with vehicle or UNC0379 and infected for 2 h. Consistent with a previous study (48), RNAPII was distributed across the viral genome with elevated peaks found at IE genes (ICP27, ICP0, ICP4, ICP22, and ICP47) with vehicle treatment (Fig. 4D). In UNC0379-treated cells, global RNAPII recruitment was reduced. Mean reads per genomic content (RPGC) clearly demonstrated a 2.7-fold decrease in RNAPII recruitment across the HSV-1 genome with UNC0379 (Fig. 4E). These results highlight a critical role for SETD8 in HSV-1 IE transcription, which in part was mediated through recruitment of HCF-1 and RNAPII as well as reduced heterochromatin (H3K9me3) accumulation.
## SETD8 promoted chromatin accessibility of the HSV-1 genome
One mechanism by which SETD8 promotes cellular gene expression is by facilitat ing chromatin accessibility through H4K20-monomethylation. Remarkably, it has been shown that H4K20me1 increases H4 tail flexibility, resulting in greater inter-nucleosomal distance and more mobile histone H4 (39). We hypothesized that SETD8 promotes HSV-1 IE transcription by increasing chromatin accessibility to the HCF-1 coactivator complex and other transcriptional regulators. To test this, we performed ATAC-seq in vehicle-or UNC0379-treated HFF cells infected with HSV-1 for 1.5 h (Fig. 5A). As anticipated, in vehicle-treated cells, the largest ATAC-seq peaks (and therefore accessible chromatin) were enriched in regions of the HSV-1 genome that overlapped with RNAPII ChIP-seq peaks (Fig. 4D), such as within the IE genes ICP0 and ICP4. Concomitant with increased H3 density and repressive H3K9me3 (Fig. 4C), UNC0379 reduced ATAC-seq reads across the HSV-1 genome (Fig. 5A). Most significantly, UNC0379 treatment decreased global chromatin accessibility across the HSV-1 genome by 2.6-fold compared to the vehicle, as measured by mean RPGC (Fig. 5B), without affecting viral genome levels. These results demonstrate that SETD8 promoted HSV-1 chromatin accessibility during lytic infection. We next explored whether reduced chromatin accessibility was initiated by a change in H4K20me1 levels on the HSV-1 genome. To this end, HFF cells were pretreated with vehicle or UNC0379 for 4 h, and chromatin was prepared at 1 hpi to capture the early stages of lytic infection. ChIP-QPCR assays revealed an UNC0379-mediated decrease in H4K20me1 levels proximal to the transcriptional start site (TSS) and regions further downstream within representative IE (ICP0 and ICP4) and E (UL29) viral genes (Fig. 6). It is important to note that decreased H4K20me1 levels were restricted to HSV-1 as no impact was observed on the cellular control gene RPL5. However, as determined through Western blot analysis, prolonged UNC0379 treatment (24 and 48 h) also reduced H4K20me1 levels in acid-extracted histones (Fig. S6), consistent with previous studies (49,50). Collectively, ChIP results suggest that SETD8 directly modulated HSV-1 chromatin accessibility by depositing the H4K20me1 mark. Notably, SETD8 inhibition reduced H4K20me1 levels on the HSV-1 genome.
## SETD8 inhibition reduced the levels of elongating viral mRNAs
To circumvent heterochromatic suppression, HCF-1 promotes multiple stages of IE gene transcription through interactions with transcription factors, epigenetic modifiers, and components of the SEC (19,20,24,26,29,30,33,34). It has been established that SEC-mediated transcriptional elongation is essential for optimal IE gene expression, and inhibition of the SEC suppresses HSV-1 lytic infection and reactivation from latency (33,34). Critically, the SEC promotes the release of RNAPII from its paused state located 30-50 bp downstream of the TSS to enable transcriptional elongation and production of full-length IE mRNAs.
In addition to modulating chromatin accessibility, SETD8 promotes transcriptional elongation of cellular genes (40,41). The H4K20me1 modification that is deposited by SETD8 recruits the MSL/MOF histone acetylase complex. The resulting histone acetyla tion increases nucleosome mobility along the DNA fiber, facilitating RNAPII elongation. Therefore, to determine if SETD8 was required for HSV-1 IE transcriptional elongation, we blocked the activity of SETD8 with UNC0379 and performed an assay that measured the quantity of "large" elongating viral mRNA (which is indicative of transcriptional elongation), compared to that of the "small" initiating mRNAs (Fig. 7A) (34). For ICP4 and ICP27, there was no impact on the quantity of initiating small RNAs with SETD8 inhibition (Fig. 7B). The positive control JQ1, a selective inhibitor of BET family of bromodomains previously shown to stimulate transcriptional elongation of HSV-1 and HIV (34,(51)(52)(53), increased the large HSV-1 IE mRNA population. In contrast, UNC0379 Full-Length Text treatment suppressed transcriptional elongation, as indicated by decreased proportions of large IE mRNA. Expressed as a ratio of the large IE mRNAs to small mRNAs, SETD8 inhibition significantly reduced the relative abundance of large IE transcripts (Fig. 7C). These results suggest SETD8 mediated both chromatin accessibility and transcription elongation of HSV-1 IE genes during lytic infection.
## The initiation of HSV-1 reactivation from latency was suppressed with SETD8 inhibition
Given the potent suppression of HSV-1 lytic infection by UNC0379 (Fig. 3; Fig. S1 andS2), we next assessed whether SETD8 inhibition would block latency reactivation. To this end, we utilized the mouse model of HSV-1 latency reactivation and infected the eyes of Balb/C mice with HSV-1. The clearance of primary infection and establishment of viral latency in sensory neurons occurs ~40 days post-infection (dpi) (54). To induce a robust reactivation, trigeminal ganglia were explanted into culture media and monitored for HSV-1 reactivation by measurement of viral IE mRNAs (Fig. 8A through C), viral yields (Fig. 8D), and the expression of the lytic viral protein UL29 (Fig. 8E andF).
Transcription of representative IE genes (ICP4 and ICP27) was reduced in ganglia explanted in the presence of UNC0379 (Fig. 8A). Previous studies showed that the BET inhibitor JQ1 and PI3K inhibitor LY294002 accelerated HSV-1 reactivation in ganglia explants (34,55,56). To further evaluate the requirement of SETD8 for the transcription of IE genes during reactivation, ganglia were explanted in the presence of either vehicle, JQ1, LY294002, or combinations of JQ1 or LY294002 with UNC0379 (Fig. 8B andC). Explanting with JQ1 or LY294002 increased IE transcript levels. Importantly, explant in the presence of SETD8 inhibitor UNC0379 suppressed IE gene transcription, even under accelerated reactivation conditions with JQ1 or LY294002. Given the reliance on IE gene expression for production of progeny virus during reactivation, we also observed a significant decrease in viral yields when determined by plaque assays (Fig. 8D). Thus, SETD8 inhibition suppressed both IE transcription and viral yields during reactivation.
To directly assess the number of neurons undergoing primary reactivation, ganglia were explanted in the presence of vehicle, ACV, UNC0379, or the combination of UNC0379 and ACV (Fig. 8E). Ganglia tissue sections were stained with an antibody that recognizes the lytic viral protein UL29 to identify those neurons undergoing reactivation (Fig. 8F). Quantitation of the number of UL29-positive [UL29(+)] single neurons, repre senting the primary reactivation event, revealed a significant decrease with UNC0379 treatment as compared to vehicle control (Fig. 8E). Under physiological (in vivo) condi tions, HSV-1 reactivation does not spread laterally through the ganglia; however, in the ganglia explant model, axotomy and explantation serve as recurrent reactivation triggers, and lateral spread is readily observed at 48 h following explantation (26, 35, 57, 58). To distinguish between effects on primary reactivation versus viral spread (clusters) in this model, ACV was included to block viral spread. Compared to ACV alone, there was a significant decrease in the number of UL29(+) single neurons in ganglia explanted in the combination of UNC0379 and ACV. Furthermore, SETD8 inhibition also suppressed reactivation spread, as measured by reduced UL29(+) clusters. Collectively, we conclude that SETD8 was required for the initiation of primary HSV-1 reactivation from latency.
## SETD8 inhibition suppressed primary HSV-1 ocular infection in vivo
The chemical probe UNC0379 revealed the importance of SETD8 in the initiation of HSV-1 reactivation from latency in the ganglia explant model and during in vitro lytic infection of fibroblast cells. To assess the effect of UNC0379 during primary in vivo infections, the eyes of Balb/C mice were pretreated topically with vehicle, 200 µM ACV, or 50 µM UNC0379 1 day prior to ocular HSV-1 infection (Fig. 9A). Eyes were continuously treated three times daily until 5 dpi. As indicated in Fig. 9B, UNC0379 treatment signifi cantly reduced the progression of ocular infection, as measured by a decrease in viral titers of ~1 log compared to the vehicle. Results demonstrated that topical application of the epigenetic inhibitor UNC0379 was as effective as ACV in reducing ocular titers, with no observable adverse effects on the eye. These findings support further investigation of SETD8 inhibition as an alternative or combinatorial antiviral therapy to HSV-1 infections.
## DISCUSSION
The lytic replication cycle, latency, and reactivation stages of HSV-1 are tightly coupled to the epigenetic machinery of the infected host cell (15, 16, 18-20, 23-28, 30-32, 34-37, 55). Importantly, the lytic replication cycle consists of a highly ordered transcriptional program that depends on the expression of IE genes. The IE proteins have functions not only in stimulating the sequential wave of E genes, genome amplification, and L genes but also in counteracting host responses to infection. These functions include altering the host splicing machinery, restricting mRNA transport, and promoting ubiquitin-medi ated degradation of immune mediators (2). Significantly, disruption of IE gene expression leads to defects in lytic replication and reactivation from latency (21-26, 31-35, 55, 58). HSV-1 genome accessibility is a key determinant for establishing the lytic replication program (15,16,18,23). This is important because, immediately after nuclear entry, nucleosomes assembled on the HSV-1 genome are marked by repressive heterochroma tin signatures (H3K9me3 and H3K27me2) (24)(25)(26)(27)(28). This initial cell-mediated repression is counteracted by the HCF-1 coactivator complex as it proceeds through interactions with multiple protein complexes. HCF-1 activity is potentiated by recruitment of transcription factors, H3K9-demethylases (LSD1 and JMJD2), and H3K4-methyltransferases (SETD1A and MLL1) to HSV-1 promoter/enhancer regions, thereby stimulating IE transcription (19,20,24,26,29,30,32,35,36). Inhibition of either LSD1 or JMJD2 results in enhanced epigenetic repression of the HSV-1 genome, blocking both lytic infection and latency reactivation (24,26,32). While elements of the HCF-1 coactivator complex have been defined, the epigenetic components and mechanisms regulating HSV-1 chromatin at IE genes are not completely understood. To identify additional epigenetic factors that regulate IE gene transcription, we screened a curated chemical library of 48 epigenetic inhibitors. We identified several epigenetic inhibitors that modulated HSV-1 IE gene transcription. Among these, the chemical probe UNC0379 targeting the H4K20 monomethyltransferase SETD8 effectively repressed IE gene transcription.
Epigenetic inhibitors have previously been utilized as chemical probes to interrogate the stages of HSV-1 lytic infection and latency reactivation (19,20,24,26,32,55,56,58). In this study, treatment of human fibroblast cells with the SETD8 inhibitor UNC0379 significantly repressed HSV-1 IE gene expression across a wide range of MOIs, reduced viral yields, and limited the spread of lytic infection. Moreover, UNC0379 treatment displayed no impact on viral entry and no indication of cytotoxicity at the evaluated concentration, thus supporting the premise that the disruption of IE gene transcrip tion was specifically targeted. We also found the initiation of HSV-1 reactivation was suppressed in latently infected mouse sensory ganglia. Most significantly, in vivo topical application of the SETD8 inhibitor suppressed HSV-1 titers in a mouse ocular model. The data presented here support the critical role of SETD8 in regulating HSV-1 IE gene expression during lytic infection and the initiation of primary reactivation from latency. One mechanism by which SETD8 promotes cellular gene expression is through H4K20-monomethylation, which leads to increased chromatin accessibility (39). The H4K20me1 modification enhances the flexibility of the H4 tail, resulting in a greater inter-nucleosomal distance and elevated histone mobility. In the context of HSV-1, we hypothesized that SETD8 facilitates the dynamics of chromatin modulation during the early events of IE gene transcription by promoting HSV-1 genome accessibility (Fig. 10). This hypothesis was supported by ChIP-QPCR assays, which showed increased H3K9me3 heterochromatin and reduced HCF-1 and RNAPII occupancy at IE genes following UNC0379 treatment. This repression was not restricted to the IE gene cluster as ChIP-seq illustrated a global decrease in RNAPII binding across the HSV-1 genome. Mechanistically, reduced H4K20me1 levels correlated with a genome-wide decrease in HSV-1 chromatin accessibility, as determined by ATAC-seq. These results suggest that decreased H4K20me1 deposition at viral IE genes leads to reduced recruitment of HCF-1 and RNAPII, likely due to diminished chromatin accessibility. While HCF-1 does not directly bind DNA, it is plausible that decreased chromatin accessibility also impairs the recruitment of transcription factors (such as VP16, OCT-1, Sp1, and GABP) that recruit HCF-1 to IE promoter regions. It is also important to note that unlike the cellular genome, the HSV-1 genome is initially assembled into unstable nucleosome structures that are highly accessible to MNase digestion or to Tn5 transposase insertion in ATAC-seq assays (15-18, 59, 60). Accordingly, we hypothesize that the high ATAC-seq coverage of shallow peaks observed in the vehicle group (Fig. 5A) reflects the overall high level of chromatin accessibility resulting from these unstable nucleosomes, particularly during the early phase of infection (1.5 hpi).
With SETD8 inhibition, the decrease in H4K20me1 levels observed by ChIP-QPCR (Fig. 6, 1 hpi) appeared to be restricted to the HSV-1 genome, with no change detected in a representative cellular control. To determine whether UNC0379 exhibits selectivity for HSV-1 chromatin over the cellular chromatin, we assessed total H4K20me1 levels in cells treated with UNC0379. Western blot analysis of acid-extracted histones revealed that prolonged UNC0379 treatment (24 and 48 h) reduced total H4K20me1 levels, while no significant change was observed at 5 h (Fig. S6). These results suggest that global changes in H4K20me1 levels on cellular chromatin require extended treatment. It is tempting to speculate that HSV-1 nucleosome assembly may involve the incorporation of histones initially free of the H4K20me1 modification, followed by the active recruit ment of SETD8 to deposit H4K20me1 on the viral genome early during infection. In contrast, cellular chromatin contains histones already marked with H4K20me1 and thus requires extended UNC0379 treatment to inhibit SETD8 activity and potentially promote turnover of H4K20me1 by the demethylase PHF8/KDM7B. Further investigation into the kinetics of H4K20me1 deposition and removal on HSV-1 and cellular chromatin remains an important area of future interest.
In addition to H4K20me1-mediated regulation of chromatin accessibility, our results suggest SETD8 controls HSV-1 replication on multiple levels, including transcriptional elongation. Under SETD8 inhibition, we observed reduced levels of large elongating IE mRNAs, suggesting a block in the release of paused RNAPII or a reduced rate of transcriptional elongation. Untangling the specific function(s) of SETD8 in transcriptional elongation of IE genes is beyond the scope of this study. However, these results are supported by evidence that H4K20me1 recruits histone acetyltransferase MSL/MOF to cellular genes, which promotes the release of paused RNAPII (40,41). In parallel, it has been clearly shown that transcription elongation of HSV-1 IE genes is essential for optimal lytic infection and reactivation (33,34).
As an additional layer of complexity, a previous study by Chen et al. found that UNC0379 treatment of THP-1 cells suppressed HSV-1 genome amplification at 10 hpi (61). Mechanistically, SETD8-mediated methylation of PCNA was proposed to be required for its stability, thereby suppressing HSV-1 DNA amplification upon UNC0379 treatment (61,62). Results from our study demonstrated a decrease in IE transcription with UNC0379 treatment as early as 1.5 hpi (Fig. 3A). Based on the kinetics of the lytic gene cascade, we anticipate that UNC0379-mediated suppression of IE genes would also lead to a decrease in HSV-1 genome amplification at later times post-infection. Furthermore, we observed that UNC0379-mediated reduction of H4K20me1 levels on the HSV-1 genome is an early event (1 hpi, Fig. 6), while Chen et al. reported no change during longer periods of infection (10 hpi) (61). Collectively, these findings suggest SETD8 targets multiple stages of the viral replication cycle (chromatin accessibility, transcriptional elongation, and genome amplification), making it a promising druggable target to suppress HSV-1 infection.
In addition to SETD8, our epigenetic inhibitor screen identified chemical probes targeting protein arginine methyltransferase 1 (PRMT1; TC-E5003) (42), PRMT3 (SGC-707) (43), PRMT4 (TP-064) (44), and Type I PRMTs (MS023; PRMT1, -3, -4, -6, and -8) (45), all of which suppressed HSV-1 IE transcription (Fig. 2; Table S2). A previous study has shown that arginine methylation of ICP27 disrupts its stability and localization, resulting in defects in HSV-1 transcription and viral titers (63). In another study, HSV-1 infection was reduced in heterozygous PRMT3 mice and after in vivo treatment with the PRMT3 inhibitor SGC-707 (64). Mechanistically, it was found that PRMT3-mediated methylation of DNA and RNA sensors cGAS, RIG-I, and MDA5 attenuates their activation during HSV-1 infection. The corollary of our epigenetic inhibitor screen highlights the need for further investigation into the roles of PRMT1, PRMT3, and PRMT4 in the HSV-1 replication cycle, and by extension, to other human herpesviruses.
In this study, we screened a curated epigenetic library containing 48 compounds to identify novel factors involved in HSV-1 IE gene transcription during lytic infection. We identified the H4K20 mono-methyltransferase SETD8 as a critical factor for promot ing viral IE gene expression during lytic infection and the initiation of viral reactiva tion from latency. Significantly, topical application of the SETD8 inhibitor UNC0379 suppressed primary ocular infection of mice in vivo. Our working model proposes that SETD8-mediated H4K20 mono-methylation increases HSV-1 chromatin accessibility, thereby facilitating recruitment of the HCF-1 transcription coactivator complex and RNAPII (Fig. 10). This study demonstrated that UNC0379 treatment increased repressive H3K9me3 heterochromatin and decreased both chromatin accessibility and H4K20me1 levels. The inhibition of SETD8 also reduced the fraction of elongating HSV-1 IE mRNAs.
Importantly, UNC0379 treatment also suppressed HCMV transcription, suggesting a common regulatory mechanism among herpesviruses. Because UNC0379 inhibited HSV-1 IE transcription across a wide range of MOIs, it is tempting to speculate that further research is warranted to evaluate UNC0379 as a potential combination therapy with ACV, particularly in cases of ACV resistance, as observed in the immunocompro mised population or during prolonged treatment for recurrent HSV ocular infections (11)(12)(13)(14)32). Taken together, these results indicate that SETD8 is an essential epigenetic modulator of HSV-1 chromatin and represents a promising novel antiviral therapeutic target.
## MATERIALS AND METHODS
## Cell cultures and viral infections
Telomerase-immortalized HFF cells (TERT-human foreskin fibroblasts), MRC-5 cells, and Vero cells were maintained according to standard protocols. Viral infections with HSV-1 (Strain 17) and human cytomegalovirus (HCMV, Towne strain) were performed by infecting cells at the indicated multiplicity of infection (MOI) in Dulbecco's modified Eagle medium (DMEM) containing 1% fetal bovine serum (FBS) and 1% penicillin/strep tomycin for 1 h at room temperature (RT). Following absorption, the inoculum was removed and then the cells were washed with PBS to remove unbound virus and then incubated for the indicated duration with DMEM containing 1% FBS at 37°C/5% CO 2 . Where specified, viral yields were determined by plaque assay on Vero cells using limiting dilution of homogenized cells or mouse ganglia or eyes.
## Epigenetic inhibitor screen
Epigenetic inhibitors were obtained from the NCATS or commercial sources (Table S1). Unless otherwise stated, HFF cells were pretreated for 4 h with a vehicle control (dimethyl sulfoxide, DMSO) or the indicated inhibitor. Treated cells were infected with HSV-1 (MOI 1) at RT for 1 h in the absence of inhibitor, followed by continued infection in the presence of inhibitor for 1.5 h at 37°C. The BET bromodomain inhibitor JQ1(+) and the JMJD2 inhibitor ML324 served as positive controls (26,34).
## Cell viability assays
HFF cells were treated with vehicle or the indicated concentrations of UNC0379, NSC663284, or saponin (cytotoxic control) for 5.5 h. Cell viability was measured using a BioTek Synergy Neo2 Hybrid Multimode Reader and the CellQuanti-MTT Kit (BioAssay Systems) according to the manufacturer's instructions.
## Animals and infections
## Viral reactivation in explants of sensory ganglia
Female 8-week-old BALB/cAnNTac (Taconic Biosciences) mice were infected ocularly with 5 × 10 6 PFU of HSV-1 (strain F). Following the establishment of latency (~40 dpi), mice were randomized, trigeminal ganglia were bisected, and paired halves were incubated in media containing control vehicle or UNC0379. Viral yields were determined at 48 h post explant by titering homogenates of ganglia on Vero cell monolayers. mRNA levels of viral and control cellular genes were determined after explant of latently infected ganglia in the presence of vehicle, UNC0379, JQ1, LY294002, or a combination of JQ1/UNC0379 or LY294002/UNC0379 for 12 h. For quantitation of neurons undergoing viral reactivation, ganglia were explanted for 48 h in the presence of vehicle, UNC0379, ACV, or UNC0379/ ACV. Ganglia were subsequently fixed in 4% paraformaldehyde for 12 h and embedded in paraffin (54). Deparaffinized sections were treated with citric acid for antigen retrieval, stained with anti-UL29 antibody, and visualized using a Leica TCS-SP8 laser scanning confocal microscope.
## Primary ocular HSV-1 infection
BALB/cAnNTac mice were pretreated ocularly by topical application (5 µL per eye in 0.1% methylcellulose) of 50 µM UNC0379, 200 µM ACV, or vehicle, followed by ocular infection with 5 × 10 5 PFU of HSV-1 (Strain F) per eye. Treatments were continued three times per day, and eyes were harvested at 5 dpi for determination of viral yields.
## RNA isolation and quantitation
For quantitation of mRNA from cultured cells, RNA was isolated using the Nucleo Spin RNA Isolation Kit (Macherey Nagel), RNA concentrations were determined with a spectrophotometer (DeNovix, DS-11FX+), and cDNAs were synthesized using qScript cDNA SuperMix (Quanta Biosciences). Specific cDNAs were quantitated in triplicate by RT-QPCR using SYBR green master mix (Roche) on a QuantStudio 3 (Applied Biosys tems; QuantStudio v1.5.3 software). Isolation of individual small (<200 nucleotides) and large (>200 nucleotides) RNA fractions was performed as previously described with minor modifications (34). Briefly, HFF cells were lysed in TriPure Isolation Reagent (Sigma Aldrich), followed by the addition of chloroform according to the manufacturer's instructions. Total RNA in the aqueous phase was recovered with ethanol added to a final concentration of 35% and applied to RNA isolation columns (Macherey Nagel). Following centrifugation, the column-bound large RNA fraction was isolated according to the manufacturer's instructions, while the final ethanol concentration of the unbound small RNA fraction was increased to 70% and applied to a second RNA isolation column. cDNAs from small RNA fractions were prepared using the qScript microRNA cDNA Synthesis kit (Quanta Biosciences), while the large RNA fraction was reverse-transcribed using qScript cDNA SuperMix (Quanta Biosciences). For quantitation of mRNAs from trigeminal ganglia, pools of five ganglia were lysed in TriPure Isolation Reagent (Sigma Aldrich) and disrupted using Lysing Matrix D (MP Biomedicals) in a FastPrep-24 bead beating grinder (6.0 m/s, 40 sec cycles, three cycles), and RNA was isolated using the NucleoSpin RNA Isolation Kit (Macherey Nagel). Trigeminal ganglia samples were normalized to mRNA levels of murine GAPDH. Primer sequences are listed in Table S3.
## Immunoblotting
HFF cells were pretreated with vehicle or 4 µM and 8 µM of either UNC0379 or NSC663284 and then infected with HSV-1 (MOI 2) for 3 h. Cells were lysed in radioim munoprecipitation assay (RIPA) buffer [50 mM Tris-HCl (pH 7.5), 250 mM NaCl, 1 mM EDTA, 1% NP40, 1% sodium deoxycholate, 0.1% SDS, complete protease inhibitors, 2 mM NaV0 4 , 1 mM NaF, and 10 mM B-glycerophosphate] and briefly agitated with a QSonica Q500 Cup Sonicator (amplitude 75%, 3 cycles of 30 s on/off, 4°C). For acid-extracted histones, cells were resuspended in Triton extraction buffer [TEB (0.5% Triton X-100, PBS, complete protease inhibitors)] and lysed on ice for 10 m. Following centrifugation (6,500×g for 10 m, 4°C) and a second wash with TEB, nuclei were resuspended in 0.2 N HCl and incubated overnight at 4°C. Acid-extracted histones were collected from the supernatant after centrifugation. Protein extracts were resolved by SDS-PAGE and analyzed by Western blot using the antibodies listed in Table S3. Blots were visualized with WesternBright Quantum (Advansta) and quantitated with a G:BOX Chemi XT4 (Syngene; GeneTools 4.3.17.0 v software).
## Impact of UNC0379 on HSV-1 entry/transport
HFF cells were pretreated with vehicle or 2 µM and 4 µM UNC0379 and then infected with HSV-1 (MOI 1) for 1.5 h. Total cellular DNA was isolated with the Quick-DNA Miniprep Plus kit (Zymo Research). For the nuclear fraction, HFF cells were resuspended in 0.4 mL buffer A (10 mM HEPES pH 7.9, 10 mM KCl, and 1.5 mM MgCl 2 ) for 15 min on ice, and then the cell membrane was lysed with 25 µL 10% NP40. After vortexing and centrifugation, DNA from nuclear cell pellets was isolated with the Quick-DNA Miniprep Plus kit (Zymo Research). Total and nuclear DNAs were quantitated in triplicate by RT-QPCR. Viral DNA levels were determined using primers to the viral UL30 gene, and samples were normalized to the levels of the cellular GAPDH gene.
## Chromatin immunoprecipitation (ChIP) and ChIP-seq assays
ChIP assays were done essentially as described (35) with the following modifications. Isolated nuclei were lysed, and samples were sonicated using a Covaris M220 Focused-Ultrasonicator (8 m, 10% duty cycle, 75 W incident power, 200 burst per cycle) to obtain DNA fragments of 200-500 base pairs (bp). Chromatin samples (350-450 µg) were precleared with Protein G Dynabeads (ThermoFisher) for 1 h at 4°C followed by incubation with 3-5 µg antibodies for 14 h at 4°C. Antibodies used are listed in Table S3. For ChIP-seq, chromatin samples were incubated with 1:1,000 dilution of anti-RNAPII-NTD (Cell Signaling 14958), and library construction was prepared by the CCR-Sequenc ing Facility, Frederick National Laboratory for Cancer Research. Libraries were sequenced on a NovaSeq X Plus 1.5B to yield 3.10-3.60 × 10 8 paired-end reads per sample.
## Assay for transposase-accessible chromatin sequencing (ATAC-seq) sample processing
HFF cells were pretreated with vehicle or 4 µM UNC0379 and infected with HSV-1 (MOI 5) for 1.5 h. Nuclear extracts, DNA transposition, and library construction were prepared by the CCR-Sequencing Facility, Frederick National Laboratory for Cancer Research according to Caravaca et al. (65). Libraries were sequenced on a NovaSeq X Plus 1.5B to yield 2.16-2.56 × 10 8 paired-end reads per sample.
## ChIP-seq and ATAC-seq analysis
Reads were processed using our chrom-seek (1.2.0) pipeline (https://github.com/ OpenOmics/chrom-seek). Reads were trimmed with Cutadapt version 4.4 (66) and then aligned to a custom reference genome composed of the human hg38 genome plus the HSV-1 reference chromosome NC_001806.2 using BWA version 0.7.17 (67). All reads aligning to the Encode hg38 blacklist regions (68) were identified and removed with Picard SamToFastq (https://broadinstitute.github.io/picard/). Reads with a mapQ score less than 6 were removed with SAMtools version 1.17 (69), and PCR duplicates were removed with Picard MarkDuplicates. Data were converted into bigwig files for viewing and normalized by reads per genomic content (RPGC) of the combined reference genome using deepTools version 3.5.1 (70). For ChIP-seq, each sample was subtractioncorrected against its input control. Coverage plots were created using karyoploteR and GenomicRanges. This work utilized the computational resources of the NIH HPC Biowulf cluster (http://hpc.nih.gov).
| Thomas M. Kristie, Conceptualization, Formal analysis, Funding acquisition, Project administration, Supervision, Visualization
## References
1. Russell, Tscharke (2016) "Lytic promoters express protein during herpes simplex virus latency" *PLoS Pathog*
2. Knipe, Heldwein, Mohr et al. (2021) "Herpes simplex viruses: mechanisms of lytic and latent infection"
3. Grams, Edwards, Bloom (2023) "A viral lncRNA tethers HSV-1 genomes at the nuclear periphery to establish viral latency" *J Virol*
4. Phelan, Barrozo, Bloom (2017) "HSV1 latent transcription and non-coding RNA: a critical retrospective" *J Neuroimmunol*
5. Perng, Jones, Ciacci-Zanella et al. (2000) "Virusinduced neuronal apoptosis blocked by the herpes simplex virus latency-associated transcript" *Science*
6. Garber, Schaffer, Knipe (1997) "A LAT-associated function reduces productive-cycle gene expression during acute infection of murine sensory neurons with herpes simplex virus type 1" *J Virol*
7. Wang, Zhou, Johnson et al. (2005) "Herpesviral latency-associated transcript gene promotes assembly of heterochromatin on viral lytic-gene promoters in latent infection" *Proc Natl Acad Sci*
8. Hjalmarsson, Blomqvist, Sköldenberg (2007) "Herpes simplex encephalitis in Sweden, 1990-2001: incidence, morbidity, and mortality" *Clin Infect Dis*
9. Kumar, Bhutta, Mendez (2024) "Herpes simplex encephalitis"
10. Arvin, Campadelli-Fiume, Mocarski et al. (2007) "Human herpesviruses: biology, therapy, and immunoprophylaxis"
11. Levin, Bacon, Leary (2004) "Resistance of herpes simplex virus infections to nucleoside analogues in HIV-infected patients" *Clin Infect Dis*
12. Piret, Boivin, Van Velzen et al. (2011) "Resistance of herpes simplex viruses to nucleoside analogues: mechanisms, prevalence, and management" *Antimicrob Agents Chemother*
13. Schalkwijk, Snoeck (2022) "Acyclovir resistance in herpes simplex viruses: prevalence and therapeutic alternatives" *Biochem Pharmacol*
14. Cabral, Cushman, Sodroski et al. (2021) "ATRX limits the accessibility of histone H3-occupied HSV genomes during lytic infection" *PLoS Pathog*
15. Hu, Depledge, Cortes et al. (2019) "Chromatin dynamics and the transcriptional competence of HSV-1 genomes during lytic infections" *PLoS Pathog*
16. Lacasse, Schang (2010) "During lytic infections, herpes simplex virus type 1 DNA is in complexes with the properties of unstable nucleo somes" *J Virol*
17. Lacasse, Schang (2012) "Herpes simplex virus 1 DNA is in unstable nucleosomes throughout the lytic infection cycle, and the instability of the nucleosomes is independent of DNA replication" *J Virol*
18. Vogel, Kristie (2013) "The dynamics of HCF-1 modulation of herpes simplex virus chromatin during initiation of infection" *Viruses*
19. Kristie (2015) "Dynamic modulation of HSV chromatin drives initiation of infection and provides targets for epigenetic therapies" *Virology (Auckl)*
20. Deluca, Mccarthy, Schaffer (1985) "Isolation and characteriza tion of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4" *J Virol*
21. Cai, Schaffer (1992) "Herpes simplex virus type 1 ICP0 regulates expression of immediate-early, early, and late genes in productively infected cells" *J Virol*
22. Arbuckle, Kristie (2014) "Epigenetic repression of herpes simplex virus infection by the nucleosome remodeler CHD3"
23. Liang, Vogel, Narayanan et al. (2009) "Inhibition of the histone demethylase LSD1 blocks alpha-herpesvirus lytic replication and reactivation from latency" *Nat Med*
24. (2025) *Full-Length Text Journal of Virology*
25. Silva, Cliffe, Chang et al. (2008) "Role for A-type lamins in herpesviral DNA targeting and heterochromatin modulation" *PLoS Pathog*
26. Liang, Vogel, Arbuckle et al. (2013) "Targeting the JMJD2 histone demethylases to epigenetically control herpesvirus infection and reactivation from latency" *Sci Transl Med*
27. Francois, Rohani, Loftus et al. (2024) "Single-genome analysis reveals a heterogeneous association of the herpes simplex virus genome with H3K27me2 and the reader PHF20L1 following infection of human fibroblasts"
28. Lee, Raja, Knipe (2016) "Herpesviral ICP0 protein promotes two waves of heterochromatin removal on an early viral promoter during lytic infection" *mBio*
29. Wysocka, Herr (2003) "The herpes simplex virus VP16-induced complex: the makings of a regulatory switch" *Trends Biochem Sci*
30. Narayanan, Ruyechan, Kristie (2007) "The coactivator host cell factor-1 mediates Set1 and MLL1 H3K4 trimethylation at herpesvirus immediate early promoters for initiation of infection" *Proc Natl Acad Sci*
31. Huang, Kent, Placek et al. (2006) "Trimethylation of histone H3 lysine 4 by Set1 in the lytic infection of human herpes simplex virus 1" *J Virol*
32. Hill, Quenelle, Cardin et al. (2014) "Inhibition of LSD1 reduces herpesvirus infection, shedding, and recurrence by promoting epigenetic suppression of viral genomes" *Sci Transl Med*
33. Alfonso-Dunn, Arbuckle, Vogel et al. (2020) "Inhibition of the super elongation complex suppresses herpes simplex virus immediate early gene expression, lytic infection, and reactivation from latency" *mBio*
34. Alfonso-Dunn, Turner, Beltran et al. (2017) "Transcriptional elongation of hsv immediate early genes by the super elongation complex drives lytic infection and reactivation from latency" *Cell Host Microbe*
35. Arbuckle, Vogel, Efstathiou et al. (2023) "Deletion of the transcriptional coactivator HCF-1 in vivo impairs the removal of repressive heterochromatin from latent hsv genomes and suppresses the initiation of viral reactivation" *mBio*
36. Narayanan, Nogueira, Ruyechan et al. (2005) "Combinato rial transcription of herpes simplex virus and varicella zoster virus immediate early genes is strictly determined by the cellular coactivator HCF-1" *J Biol Chem*
37. Vogel, Kristie (2000) "The novel coactivator C1 (HCF) coordinates multiprotein enhancer formation and mediates transcription activation by GABP" *EMBO J*
38. Ma, Yu, Li et al. (2014) "Discovery of a selective, substrate-competitive inhibitor of the lysine methyltransferase SETD8" *J Med Chem*
39. Shoaib, Chen, Shi et al. (2021) "Histone H4 lysine 20 monomethylation directly facilitates chromatin openness and promotes transcription of housekeeping genes" *Nat Commun*
40. Kapoor-Vazirani, Vertino (2014) "A dual role for the histone methyltransferase PR-SET7/SETD8 and histone H4 lysine 20 monome thylation in the local regulation of RNA polymerase II pausing" *J Biol Chem*
41. Nikolaou, Moulos, Harokopos et al. (2017) "Kmt5a controls hepatic metabolic pathways by facilitating RNA Pol II release from promoter-proximal regions" *Cell Rep*
42. Bissinger, Heinke, Spannhoff et al. (2011) "Acyl derivatives of p-aminosulfonamides and dapsone as new inhibitors of the arginine methyltransferase hPRMT1" *Bioorg Med Chem*
43. Kaniskan, Szewczyk, Yu et al. (2015) "A potent, selective and cell-active allosteric inhibitor of protein arginine methyltransferase 3 (PRMT3)" *Angew Chem Int Ed Engl*
44. Nakayama, Szewczyk, Sena et al. (2018) "TP-064, a potent and selective small molecule inhibitor of PRMT4 for multiple myeloma" *Oncotarget*
45. Eram, Shen, Szewczyk et al. (2016) "A potent, selective, and cell-active inhibitor of human type I protein arginine methyltransferases" *ACS Chem Biol*
46. Blum, Ibáñez, Rao et al. (2014) "Small-molecule inhibitors of SETD8 with cellular activity" *ACS Chem Biol*
47. Rozman, Nachshon, Samia et al. (2022) "Temporal dynamics of HCMV gene expression in lytic and latent infections" *Cell Rep*
48. Dremel, Deluca (2019) "Genome replication affects transcription factor binding mediating the cascade of herpes simplex virus transcrip tion" *Proc Natl Acad Sci*
49. Kukita, Sone, Kaneko et al. (2022) "The histone methyltransferase SETD8 regulates the expression of tumor suppressor genes via H4K20 methylation and the p53 signaling pathway in endometrial cancer cells" *Cancers (Basel)*
50. Veschi, Liu, Voss et al. (2017) "Epigenetic siRNA and chemical screens identify SETD8 inhibition as a therapeutic strategy for p53 activation in high-risk neuroblastoma" *Cancer Cell*
51. Contreras, Barboric, Lenasi et al. (2007) "HMBA releases P-TEFb from HEXIM1 and 7SK snRNA via PI3K/Akt and activates HIV transcription" *PLoS Pathog*
52. Filippakopoulos, Qi, Picaud et al. (2010) "Selective inhibition of BET bromodomains" *Nature*
53. Bartholomeeusen, Xiang, Fujinaga et al. (2012) "Bromodo main and extra-terminal (BET) bromodomain inhibition activate transcription via transient release of positive transcription elongation factor b (P-TEFb) from 7SK small nuclear ribonucleoprotein" *J Biol Chem*
54. Arbuckle, Turner, Kristie (2014) "Analysis of HSV viral reactivation in explants of sensory neurons" *CP Microbiology*
55. Cliffe, Arbuckle, Vogel et al. (2015) "Neuronal stress pathway mediating a histone methyl/phospho switch is required for herpes simplex virus reactivation" *Cell Host Microbe*
56. (2025) *Full-Length Text Journal of Virology*
57. Camarena, Kobayashi, Kim et al. (2010) "Nature and duration of growth factor signaling through receptor tyrosine kinases regulates HSV-1 latency in neurons" *Cell Host Microbe*
58. Sawtell, Thompson (2004) "Comparison of herpes simplex virus reactivation in ganglia in vivo and in explants demonstrates quantitative and qualitative differences" *J Virol*
59. Arbuckle, Gardina, Gordon et al. (2017) "Inhibitors of the histone methyltransfera ses EZH2/1 induce a potent antiviral state and suppress infection by diverse viral pathogens" *mBio*
60. Leinbach, Summers (1980) "The structure of herpes simplex virus type 1 DNA as probed by micrococcal nuclease digestion" *J Gen Virol*
61. Muggeridge, Fraser (1986) "Chromosomal organization of the herpes simplex virus genome during acute infection of the mouse central nervous system" *J Virol*
62. Chen, Yang, Tang et al. (2022) "Inhibition of histone methyltransferase SETD8 represses DNA virus replication" *Cell Insight*
63. Takawa, Cho, Hayami et al. (2012) "Histone lysine methyltransferase SETD8 promotes carcinogenesis by deregulating PCNA expression" *Cancer Res*
64. Souki, Gershon, Sandri-Goldin (2009) "Arginine methylation of the ICP27 RGG box regulates ICP27 export and is required for efficient herpes simplex virus 1 replication" *J Virol*
65. Zhu, Li, Cai et al. (2023) "Asymmet ric arginine dimethylation of cytosolic RNA and DNA sensors by PRMT3 attenuates antiviral innate immunity" *Proc Natl Acad Sci*
66. Caravaca, Mehta, Gowda et al. (2022) "ATAC sequencing protocol for cryopreserved mammalian cells" *Bio Protoc*
67. Martin (2011) "Cutadapt removes adapter sequences from highthroughput sequencing reads" *EMBnet j*
68. Li, Durbin (2009) "Fast and accurate short read alignment with Burrows-Wheeler transform" *Bioinformatics*
69. Abascal, Acosta, Addleman et al. (2020) "Expanded encyclopae dias of DNA elements in the human and mouse genomes" *Nature*
70. Danecek, Bonfield, Liddle et al. (2021) "Twelve years of SAMtools and BCFtools. Gigascience 10:giab008"
71. Ramírez, Ryan, Grüning et al. (2016) "deepTools2: a next generation web server for deep-sequencing data analysis" *Nucleic Acids Res* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817923&blobtype=pdf | # Molecular characterization of arenavirus defective viral genomes reveals sequence features associated with their formation
Thomas Hoenen, Patrick Bohn, Sebastian Herndler, Marine-Noël Klamke, Andreas Müller, Allison Groseth, Plasmidsaurus Gmbh, Seracell Pharma, Rostock, Germany
## Abstract
Defective viral genomes (DVGs) are byproducts of replication that arise during infection with diverse RNA viruses and can impact virus infection and disease outcome. To gain insight into DVG generation during arenavirus infection, we serially passaged Tacaribe virus at a high multiplicity of infection, which led to the generation of both deletion DVGs (del-DVGs) and copyback DVGs (cb-DVGs). Interestingly, specific combinations of start/stop breakpoints were highly overrepresented, resulting in certain DVGs being highly enriched within the population. Functional characterization of the most prevalent del-DVGs suggests that they are able to compete for interaction with the viral RNA synthesis machinery and that this ability is length-dependent. A closer analysis of the DVG breakpoints used to generate highly abundant DVGs revealed a role for local sequence identity in the formation of cb-DVGs, while del-DVG formation was associated with the presence of specific nucleotide triplets (i.e., TAG, AGA, and GAA). Taken together with similar findings from other virus families, this then supports the idea that DVG formation is not a random process, but rather that specific mechanisms promote their formation at certain positions. The characterization of these individual arenavirus DVG sequences, and also the identification of sequence elements associated with their production, will facilitate future work examining their impact on arenavirus biology, and also opens up the possibility of using such sequences as a part of antivi ral approaches and/or of modulating their production as a part of virus attenuation strategies.IMPORTANCE Infection with diverse RNA viruses can generate defective viral genomes (DVGs) that, while unable to support productive virus infection on their own, appear to play a crucial role in determining infection outcome. In light of this apparent biological importance, there is an urgent need to better understand the sequence characteristics of individual DVGs and the molecular mechanisms that regulate their formation to study their biological functions. We have now characterized several DVGs that are highly enriched during infection with the arenavirus Tacaribe virus. Functional analysis of a subset of these DVGs showed length-dependent competition for the viral RNA synthesis machinery, while detailed sequence analysis revealed that DVG formation involves either regions of sequence identity within the genome or the presence of specific nucleotide sequences. Understanding these mechanisms opens up the possibility to leverage DVG generation in support of antiviral and/or vaccine attenuation approaches. KEYWORDS arenavirus, Tacaribe virus, defective viral genomes, MinION, nanopore sequencing M embers of the family Arenaviridae that infect mammals (i.e., members of the genus Mammarenavirus) are maintained in persistently infected hosts, primarily rodents, from which they can be transmitted to humans, where many of them then represent important human pathogens. These include Old World arenaviruses such as lymphocytic
choriomeningitis virus (LCMV), which causes aseptic meningitis, encephalitis, and/or meningoencephalitis (1), as well as Lassa virus (LASV) and Lujo virus, both of which are causative agents of hemorrhagic fever in Africa (2,3). In addition, there are a number of arenaviruses endemic to South America (New World arenaviruses), for example, Junín virus (JUNV) and Machupo virus (MACV), that are highly pathogenic in humans and cause severe hemorrhagic fevers (4,5). However, this group also contains a number of closely related viruses, including Tacaribe virus (TCRV), that are not pathogenic in humans and are, therefore, often used to study the basic biology of these viruses under BSL2 conditions and for comparative studies to identify traits that play a role in pathogenesis.
Mammalian arenaviruses have a bi-segmented, ambisense single-stranded RNA (ssRNA) genome (6) with the S segment (ca. 3.4 kb) encoding the viral nucleoprotein (NP) and the glycoprotein precursor (GPC), while the L segment (ca. 7.1 kb) encodes the RNA-dependent RNA polymerase (L) and the matrix protein (Z) (Fig. 1A). Both genome segments contain a highly structured, noncoding intergenic region (IGR) that separates the two encoded ORFs and acts as a transcription termination signal during mRNA transcription (7). Additionally, both segments include untranslated regions (UTRs) at the 3′ and 5′ genome ends that are critically necessary for replication, as they contain the genomic and antigenomic promoters, respectively (8).
In addition to standard (full length) viral genomes, RNA viruses are capable of generating defective viral genomes (DVGs) as a by-product of their replication (reviewed in reference 9). In general, there are two broad types of DVGs that have been described: deletion DVGs (del-DVGs) and copyback DVGs (cb-DVGs). del-DVGs retain both the natural 3′ and 5′ UTRs of the parental genome but contain deletions of internal portions of the genome (Fig. 1B) that can range from a few nucleotides to almost the entire genome (reviewed in reference 10). They have been shown to occur during infection with a wide range of both positive-and negative-sense RNA viruses, including those with segmented genomes, such as influenza virus (reviewed in references 11,12). In contrast, cb-DVGs contain a portion of the genome flanked by either the 3′ or 5′ UTR, and its reverse complement (Fig. 1B). Such cb-DVGs can be generated during the replication of the genome (vRNA) or antigenome (cRNA), and correspondingly result in either 3′ UTR cb-DVGs (i.e., containing the 3′ UTR and its reverse complement, generated during replication of vRNA) or 5′ UTR cb-DVGs (i.e., comprising the 5′ UTR and its reverse complement, produced during replication of cRNA). The formation of cb-DVGs has been observed during infection with many different non-segmented negative-sense RNA viruses (i.e., members of the Paramyxoviridae, Pneumoviridae, Rhabdoviridae, and Filoviridae families) (reviewed in reference 12).
The existence of DVGs has been known for many decades, but only recently have we seen a resurgence in interest regarding their biological relevance, which has revealed a number of important roles in virus biology and disease outcome. This can occur through (i) competition with standard virus genomes for cellular and viral resources needed for virus replication, (ii) by serving as ligands for pathways associated with antiviral immun ity, and (iii) by facilitating virus persistence (reviewed in references 13,14). In particular, cb-DVGs can have a significant degree of dsRNA character and thus have the potential to be recognized by dsRNA sensors, such as those associated with interferon production (reviewed in reference 15). Consistent with the increasing recognition of their biological importance, there is also tremendous interest in identifying specific mechanism(s) that may regulate DVG formation. Intriguingly, for both influenza virus (16) and respiratory syncytial virus (RSV) (17), studies have shown that DVG formation occurs at certain genome locations where specific nucleotide sequences are present, indicating that DVG generation is not entirely random, but rather involves specific mechanisms that regulate this process to at least some extent.
For arenaviruses, subgenomic RNA production in the form of DVGs has long been known to occur both in vitro and in vivo (18)(19)(20)(21)(22), and indeed their generation and packaging into defective interfering (DI) particles is postulated to play an important role characteristics of such viral dsRNA ligands remain unclear, one potential source of these structures could be cb-DVGs. Indeed, recent work has also shown the formation of cb-DVGs for LCMV, although interestingly, they were observed to only an extremely limited extent for New World arenaviruses (24). This is surprising since dsRNA accumulation during infection with New World arenaviruses has been reported to be much more extensive than with Old World arenaviruses (32) and suggests that further work charac terizing these structures remains needed.
To gain insights into the sequence characteristics of individual DVGs being generated during arenavirus infection, as well as the mechanism(s) that underlie their generation, we repeatedly passaged the arenavirus TCRV at a high multiplicity of infection (MOI) to induce the formation and accumulation of DVGs, which were then sequenced using nanopore technology and analyzed using a set of scripts that we developed for the detection of both del-DVGs and cb-DVGs in these kinds of long single-read sequencing data sets.
## RESULTS
## Accumulation of aberrant small RNAs during serial passaging of TCRV
To encourage the formation of DVGs, we performed serial passaging of TCRV over 20 passages in three independent replicates. RT-PCR amplification of the extracted viral RNA using primers targeting the S segment 3′ and 5′ UTRs showed evidence for the accumulation of smaller viral genome products with increasing passage number, while the abundance of full-length S segment products diminished (Fig. 2A). Examination of virus titers over the course of the passaging showed an initial 2-3 log reduction in the p1 and p2 samples, after which the titer fluctuated cyclically within a 1-1.5 log range throughout the rest of the passaging experiment (Fig. 2B). Intriguingly, further passaging of the generated p20 stock for one replicate resulted in a spontaneous 3 log increase in virus titer between p22 and p23, which was associated with a reduced DVG content and increased levels of the full-length S segment RNA (Fig. 2B andC). Estimation of full-length genome content by RT-qPCR indicated that particles within the p23 stock contained substantially more intact S (i.e., 20×) and L (200× more) segments than those from the p22 stock (Fig. 2D). While DVGs can also be generated without the need for genome rearrangement through the accumulation of changes at individual genome positions that render these viral genome copies functionally defective, Sanger sequencing of both the S and L genome segments failed to reveal the accumulation of changes that were consistently present throughout the passaging process, were present as majority populations, and/or were present consistently across replicates (Fig. S1).
Given these indications that the observed truncated TCRV DVGs were exerting interfering activity during infection, we sought to more precisely define the sequences of the individual DVGs. While efforts to Sanger sequence the 1.2 kb subgenomic product were unsuccessful, the larger 2.1 kb product produced clear Sanger sequencing data that revealed that this product was an S segment-derived del-DVG lacking the majority of the GPC gene (Fig. 3). Interestingly, it was not possible to unequivocally define the precise break start/stop for this del-DVG due to identity in the sequences before (i.e., positions 1993-1995) and after the putative breakpoint (i.e., positions 3319-3321). As such, the observed AGG found at the breakpoint in the del-DVG sequence could be formed by joining several combinations of nucleotides between positions 1992-1995 and 3319-3322 (i.e., 1992::3319, 1993::3320, 1994::3321, or 1995::3322) (Fig. 3).
In addition to these two larger products, a number of smaller products of different sizes, and in varying abundance, were also observed (Fig. 2). To gain greater insight into the genetic structure of these putative DVGs, a long single-read sequencing approach was deemed necessary. The workflow for this approach is summarized in Fig. 4, with the resulting PCR products (Fig. S2) being pooled and subjected to nanopore sequencing on a MinION device. This analysis resulted in 652,000 sequencing reads, with 581,781 reads found to be specific for TCRV, based on the presence of sequences corresponding to the from an equivalent number of infectious particles from either p22 or p23 was subjected to RT using a genome-end binding primer followed by qPCR amplification using primers binding to central portions of the viral genome within the NP and L open reading frames (as a proxy for full-length S and L-segment content). The fold increase in genome content between p22 and p23 stocks was then calculated and is shown as the mean and standard deviation of three replicate experiments.
## Full-Length Text
## Breakpoint analysis: del-DVGs
Initial breakpoint analysis for DVGs sequences involved the determination of the frequency with which individual start and stop positions were used. Among del-DVGs, a clear preference could be observed for breakpoints to form in close proximity to the Amplification and sequencing: RNA was isolated from passage 20 supernatants and converted to cDNA using a universal genome end primer. This was followed by separate polymerase chain reactions using segment-specific genome end primers. The resulting products were used for library preparation using the nanopore 1D2 kit and loaded on a MinION sequencer. Read sorting: Obtained sequences were sorted using flexbar (v3.1) into S-segment (red/purple) or L-segment (blue/turquoise) sequences based on the presence of the primer sequences used for their amplification and then further subdivided based on local alignment using lastal (version 921) into those that best aligned to a (i) vRNA-sense genome, (ii) cRNA-sense genome, (iii) vRNA-cRNA concatemer, or (iv) cRNA-vRNA concatemer (schematic representations of the templates used for these local alignments are shown). Breakpoint analysis: Once sorted, reads were trimmed, and breakpoints in the DVG sequences were identified by global alignment to the appropriate reference template (i.e., vRNA, cRNA, cRNA_vRNA or cRNA_cRNA) using a high gap open penalty (i.e., 50) but a very low gap extend penalty (0.000000001) to allow for the large deletions expected in DVGs. Break start and stop points for each DVG were then extracted for further analysis. genome ends, that is, within the first/last 150 bp of the S segment (Fig. 5A; see Table S7 at https://zenodo.org/records/14900940) or the first/last 500-800 bp of the L segment (Fig. 5B; see Table S8 at https://zenodo.org/records/14900940). However, the frequency with which specific breakpoints were used within these regions was not a direct function of distance from the genome termini, but rather specific positions appeared to be preferentially used.
This was particularly evident for the S segment, where two break starts were heavily favored, that is, positions 97 and 122 (Fig. 5A; see Table S7 at https://zenodo.org/records/ 14900940), which were collectively found 424 times more often than expected for a random distribution across the genome. These break starts, in turn, paired with heavily favored break stops at positions 3314 and 3344, resulting in DVGs with lengths of 206 nt and 201 nt, respectively (Fig. 5A andC). Alone, these two del-DVGs made up 21% of all S segment del-DVGs with single break points, and when allowing for ±1 nt accuracy, this increased to 34% (see Table S9 at https://zenodo.org/records/14900940).
While there was somewhat more variability in breakpoint positions for del-DVGs derived from the L segment, also here distinct break starts were clearly preferred, in particular at positions 68, 190, and 460-463 (Fig. 5B; see Table S8 at https://zenodo.org/ records/14900940), which were found 112 times more often than expected for a random distribution across the genome. These positions also showed preferential pairing with specific break stops, that is, position 68 with 6976, position 190 with 6936, and position 460 with 6348 or position 463 with 6351 (Fig. 5B andD). Notably, however, the DVGs produced using breakpoints 460::6348 or 463::6351 correspond to identical DVG sequences (Fig. 5D), again reflecting ambiguity as to whether part of the sequence is derived from one end of the genome or the other (as we had seen with the longer 2.1 kb S segment DVG, cf. Fig. 3). As a result, our data indicate just three distinct highly preva lent L segment-derived del-DVG species, with lengths of 196, 358, and 1,216 nt, respec tively (Fig. 5D). Collectively, these three DVG species make up 8.5% of the total DVG population, which increased to 9.2% when allowing for ±1 nt variation (see Table S10 at https://zenodo.org/records/14900940).
## Breakpoint analysis: S segment-derived cb-DVGs
Detailed analysis of cb-DVGs revealed two distinct populations-those with a single break start/stop (see Table S11 at https://zenodo.org/records/14900940) and those with multiple break start/stops (see Table S12 at https://zenodo.org/records/14900940). For S segment-derived 3′ UTR cb-DVGs (i.e., starting with the NP gene), we found that the vast majority (88%) of the reads contained a single breakpoint, and that these also exhibited a strong preference for specific break positions, as well as preferential break start/stop pairings (Fig. 6A, see Table S11 at https://zenodo.org/records/14900940). Specifically, break starts at positions 33, 41, and 690 were strongly favored and preferentially paired with break stops at positions c698 (i.e., position 698 in the complementary strand), nucleotides within a short region centered around c689 (i.e., c692-686), and c41, respectively (Fig. 6B). Interestingly, a closer examination of the regions adjacent to these breakpoints revealed that nucleotides 34-41 and c698-c690 were again identical in sequences, so that these cb-DVG populations are actually equivalent in sequence (Fig. 6B), as we had also seen with the 2.1 kb del-DVG identified by Sanger sequencing (c.f. Fig. 3) and one of the highly prevalent L segment-derived del-DVGs (c.f. Fig. 5D). However, in this case, the ambiguous region was longer than in those other DVGs, having a length of 8 nt, instead of only 3 nt. While this again makes it more difficult to determine which exact positions within this region are actually being used as the break start/stops, in this case, heterogeneity was observed in the break stops surrounding position c690, which indicates that it is position 41 that is likely being preferentially used for the generation of the observed cb-DVGs (Fig. 6B). Furthermore, since the DVGs with breakpoints at 41::c690 and those with breakpoints at 690::c41 are equivalent (since both strands are present for any given DVG following PCR amplification), in the end, these data suggest the presence of only a single highly abundant cb-DVG product of the S segment (Fig. 6C).
When analyzing the much less prevalent S segment 3′ UTR cb-DVGs with multiple breakpoints (see Table S12 at https://zenodo.org/records/14900940), 85% were found to have two distinct pairs of break starts/stops (see Table S2 at https://zenodo.org/records/ 14900940). Among these reads, 68% of initial break starts also occurred in the nucleotide regions from 33 to 41, with position 41 again being heavily favored and alone accounting for 42% of initial DVG break points (Fig. 6D; see Table S13 at https://zenodo.org/records/ 14900940). However, analysis of corresponding break stop sites did not reveal a clearly preferred site, although position 239 was favored, accounting for 18% of the associated break stops. Among DVGs containing an initial break start/stop between positions 41::239, a second break start at position 251 was then highly favored, accounting for 41% of the reads, and the majority of these (76%) then showed a second break stop at position c672 (Fig. 6D). The resulting DVG species then has a composition very similar to that observed for the highly prevalent single breakpoint DVG 41::c690 (cf. Fig. 6C andE), albeit with a slightly different length and internal sequence (Fig. 6F). Analysis of S segment-derived 5′ UTR cb-DVGs (i.e., starting with the GPC gene) revealed only a small subset (10%) of reads that contained a single breakpoint (see Table S3 at https://zenodo.org/records/14900940). A detailed analysis of these reads indicated that they were highly heterogeneous, especially compared to 3′ UTR cb-DVGs for this segment (see Table S14 at https://zenodo.org/records/14900940). Reads using a break start at nucleotide 346 were the most frequently observed (5%), and always occurred together with a break stop at position c227 to produce a DVG product of 573 nt in length (Fig. 7A andB). Among DVGs containing multiple breaks (see Table S15 at https://zenodo.org/records/14900940), 51% of reads were found to contain two clear sets of break start/stops (see Table S3 at https://zenodo.org/records/14900940). Further analysis of these reads showed that two initial break points were favored, at either position 79 (14% of reads) or 109 (16% of reads) (see Table S16 at https://zenodo.org/ records/14900940). Of the reads showing an initial break at position 79, the majority (71%) showed a corresponding break stop at position 3301. Of these, the largest subset (24%) then showed a second break start at position c3301, all of which then showed a second break stop at either position c79 (83%) or c78 (17%) (Fig. 7C andD, top left). Such a structure produces a perfectly complementary dsRNA of 402 nt in length that is composed almost exclusively of the viral UTRs and complementary copies thereof. A slightly smaller proportion (16%) showed initial break start/stops at 79::3301, but a second break start at position 3397 (thereby eliminating the terminal 26 nucleotides of the 3′ UTR) before reinitiating at a second break stop at position c37 (Fig. 7C and D, bottom left). As such, while this DVG exhibits complementarity between its terminal regions, there is no complement to the internal 97 nucleotides. For DVGs starting with a breakpoint at position 109, we also saw a highly preferred corresponding break stop, in this case at position 3326 (79% of reads). However, the position of the second break start was then more variable, with abundant populations using positions 3397 (12%), c3326 (11%), and c3301 (7%), as well as several related species with a break start in the range of 3388-3390 (22%) (Fig. 7D). Of the reads with a second break starting at 3397, all had a corresponding break stop at position c37 (Fig. 7D, top center), while break starts at position c3326 paired mainly with break stops at c109 (64%) (Fig. 7D, bottom center), and a further 18% pairing with a break stop at the neighboring c108. DVGs with a second break start at position c3301 paired primarily with position c79 (71%), as we had also seen for DVGs where the first break start is at position 79 (Fig. 7D, cf. top right and top left). Finally, the more variable DVG variants with a second break start in the region 3388-3390 also showed heterogeneity in their corresponding break stop positions, although those at c45-c43 were highly preferred (81%) (Fig. 7D, bottom right). In examining the resulting products, we again saw that, as for one of the cb-DVGs starting with a breakpoint at position 79 (Fig. 7D, top left), also several of these cb-DVG species starting with breaks at position 109 have the potential to form perfect (or nearly perfect) dsRNA structures over much of their length (i.e., Fig. 7D, bottom center and top right).
## Breakpoint analysis: L segment-derived cb-DVGs
Analysis of L segment-derived 3′ UTR cb-DVGs (i.e., starting with the L gene) containing a single breakpoint, which was the case for 32% of the reads (see Table S5 at https:// zenodo.org/records/14900940), also revealed two highly favored DVG populations, that is, those with break start/stops at positions 28::c2575 or 2568::c35 (see Table S17 at https://zenodo.org/records/14900940; Fig. 8A). Again, closer examination of the regions adjacent to these breakpoints showed that, as we had seen for S segment-derived 3′ UTR cb-DVGs, positions 29-35 and c2575-c2568 are identical in sequence, so that these cb-DVGs are equivalent and represent a single population where it is challenging to clearly determine which breakpoints were actually used within this region (Fig. 8B). When Full-Length Text analyzing L segment-derived 3′ UTR cb-DVGs with multiple breakpoints (68% of the reads) (see Table S18 at https://zenodo.org/records/14900940), we found that 38% of these had two sets of break start/stops (see Table S5 at https://zenodo.org/records/ 14900940). More detailed analysis of these reads revealed that the most frequent break starts were positions 68, 460, and 461 (see Table S19 at https://zenodo.org/records/ 14900940; Fig. 8C). For DVGs starting with an initial break starting at position 68, the majority (64%) showed a break stop at position 6976, with minority populations using the nearby 6979 (14%) or 6973 (7%). Of the reads using a break stop at position 6976, 56% then had a second break start at c6982, and of those, 60% had a final break stop at position c71 (Fig. 8D, top left). In contrast, DVGs with an initial break start at 460 paired in almost all cases with 6348 (94%). The position of the second break start was variable, although it was mostly found either in the region 6373-6428 (33%) or 6574-6597 (33%).
The final break stop was similarly variable, but restricted mainly to the region from c125 to c56 (70%) (Fig. 8D, top right). Finally, DVGs with an initial break start at 461, almost exclusively paired with either 6350 (76%) or 6351 (18%). Again, the second break start positions varied, but were restricted to three specific regions, that is, 6372-6433, 6582-6605, and c6367-c6342, the first two of which are very similar to those observed for DVGs starting with break positions 460::6348, supporting the propensity of specific sites within these regions to serve as a second break start. Indeed, even for the considerably longer DVGs with a second break start at c6367-c6342, this represents a very similar genome position, but within the complementary strand. The final break stop for these DVGs was also variable, but in the majority of reads (62%) was found between c473 and c433 (Fig. 8D, center right and bottom). Other commonly used breakpoints were also observed, for instance, at positions 30, 6936, 7091, and c29/c28 (see Table S18 at https:// zenodo.org/records/14900940; Fig. 8C); however, these were present mainly in cb-DVGs with >2 breakpoints and were, therefore, not further analyzed in detail. Interestingly, the multi-breakpoint 3′ UTR cb-DVGs using break starts at 68 and 460 preferentially used the same break stops we had observed with the L-segment del-DVGs, that is, 68::6976 and 460::6348 (cf. Fig. 5D and8D). This then suggests that these DVGs might actually represent a combination of del-DVGs and cb-DVGs, although it remains unclear whether the formation of the internal deletion and the cb-DVG event would occur in parallel (i.e., during a single cycle of template replication), or whether they occur in distinct successive steps. Notably, in the case of the 3′ UTR cb-DVGs formed using a break start at position 68 (Fig. 8D, top left), and also some of those using a break start at 461 (i.e., Fig. 8D, bottom), the DVG sequences include both the 3′ and 5′ UTRs, and a complementary copy thereof. As a result, these DVGs can also form almost perfectly complementary hairpin structures containing complementary copies of both genome termini, similar to what we observed in several S segment 5′ UTR cb-DVGs (cf. Fig. 7D and8D).
When analyzing L segment-derived 5′ UTR cb-DVGs (i.e., those starting with the Z gene), we observed that reads with a single breakpoint were less frequent than what was observed for 3′ UTR cb-DVGs (i.e., starting with the L gene), and accounted for only 13% of the reads (see Table S6 at https://zenodo.org/records/14900940). Rather, this value was similar to the proportion of single breakpoint S segment-derived 5′ UTR cb-DVGs (i.e., 10%; see Table S3 at https://zenodo.org/records/14900940). Among the 5′ UTR cb-DVGs containing a single breakpoint, two break start positions were dominant, that is, 233 and 294, and these clearly paired up with c176 and c190, respectively (see Table S20 at https://zenodo.org/records/14900940; Fig. 9A andB). When analyzing cb-DVGs containing multiple breakpoints (see Table S21 at https://zenodo.org/records/14900940), which constituted 87% of the 5′ UTR cb-DVGs, 52% were found to contain two discrete pairs of break stops/starts. Also here, certain initial break start positions occurred more frequently than others, specifically 122, 127, 128, and 164 (see Table S22 at https:// zenodo.org/records/14900940; Fig. 9C andD). Interestingly, while most of these clearly had preferred corresponding break stops, that is, 122::7032/7033 (81%), 127::7036 (96%), 128::7036 (65%), the break start at 164 had no clear preference for a specific break stop Full-Length Text position. Nonetheless, there still appeared to be some restriction regarding the corre sponding break stop position, given that 88% of reads with a break start at position 164 terminated within the 155 nt region from positions 6856 to 7010 (Fig. 9D, bottom right). However, for all these products, there was variability observed in the position of the second break start and stop. Specifically, the DVGs starting with 122::7032/7033 then predominantly used a second break start in the regions 7100-c6881 (76%), with the remaining break starts scattered across positions c6727-c5889, and of those, most used a second break stop at c128 (31%), with the remainder occurring primarily in the regions c140-c120 (77%) (Fig. 9D, top left). DVGs starting with 127::7036 preferentially used a second break start in the regions 7055-c7005 (55%), and those then used a second break stop mainly either within c150-c124 (50%) or c82-c77 (25%) (Fig. 9D, top right). Similarly, DVGs starting with 128::7036 preferentially used a second break start in regions c7045-c7034 (53%), with a final break stop in the regions c139-c126 (100%), with most stops occurring either at c128 or c126 (63%) (Fig. 9D, bottom left). Finally, DVGs starting with 164::6856-7010 predominantly used second break starts in the regions c7038-c6990 and c6939-c6883 (together 73%), and 73% of those then used a final break stop in the region c181-c167 (Fig. 9D, bottom right). Despite the increased variability observed with the multi-breakpoint L segment 5′ UTR cb-DVGs, in all cases the first break stops were located prior to the 3′ UTR, so that they contain complementary copies of both UTRs, and again indicating that these DVGs are the product of a combination of deletion and copyback events, similar to what we observed for the multi-breakpoint L segment 3′ UTR cb-DVGs (starting with the L gene) and the S segment 5′ UTR cb-DVGs (starting with the GPC gene).
## Functional characterization of del-DVGs
To better understand the biological implications of DVG formation during arenavirus infection, we examined the impact of coexpression of each of the highly prevalent del-DVGs identified in our study (Fig. 5) in the context of both minigenome and transcription and replication-competent virus-like particle (trVLP) assays. To be able to sensitively detect competition between minigenomes and del-DVGs, the amounts of minigenome, NP, and L supplied in our assays were first titrated to establish the relevant ranges within which reporter activity is dependent on the amounts of each of these factors (Fig. S4). Based on the obtained curves, amounts of transfected minigenome (125 ng) were selected that supplied high levels of reporter activity, while amounts of NP (25 ng) and L (100 ng) were selected that were limiting. That this is the case was further demonstrated by including a control in which the amounts of NP and L were reduced by half, resulting in a strong impact on reporter activity (Fig. 10A). Similarly, reduced levels of viral RNA synthesis (as reflected by reporter activity) were observed in the minigenome assay following co-transfection of T7-driven expression constructs for several of the del-DVGs found to be highly enriched in our analysis. In particular, S122::3344 (201 bp), S97::3314 (203 bp), and L68::6976 (196 bp) all significantly reduced reporter activity in the minige nome system (Fig. 10A, left panel). Notably, these represent the shortest of the del-DVGs tested, with a length that is much shorter than that of the monocistronic minigenome (826 bp). In contrast, longer del-DVGs, that is, S1995::3322 (2,096 bp), L190::6936 (358 bp), and L460::6348 (1,216 bp) did not markedly impact the levels of reporter activity (Fig. 10A, left panel). For S1995::3322 and L460::6348, it was noted that they still contained intact open reading frames for NP and Z, respectively, and since both of these proteins contribute to the regulation of viral RNA synthesis, it was considered whether their expression might be impacting DVG function. However, we found that versions of these DVGs in which the start codons for NP or Z were eliminated behaved similarly to the parental constructs (Fig. 10A, right panel). Since the extremely short length of the monocistronic minigenome, which is markedly shorter than the corresponding arenavirus genome segment (3.4 kb), could affect their susceptibility to competition by del-DVGs for the viral RNA synthesis machi nery, the effect of del-DVG coexpression was also examined in trVLP assays, which utilize a longer bicistronic minigenome that is closer in size to the arenavirus S segment (i.e., 2.6 kb). Here, we saw that the inhibitory effects of S122::3344, S97::3314, and L68::6976 were further enhanced (Fig. 10B, left panel, p0). Furthermore, an inhibitory effect of L460::6348 and L190::6936 became apparent, although it remained weaker than for the shorter del-DVGs (Fig. 10B, left panel). This phenotype could also be transferred to fresh target (p1) cells upon infection with trVLPs generated by these DVG-expressing (p0) cells. Interestingly, however, this was only the case if the p1 target cells were naïve, that is, not pre-transfected with NP and L (Fig. 10B, center panel). Additional transfection of these Cells were harvested 48 h later and measured for both nLuc (viral RNA synthesis) and FF (host cell RNA synthesis) activity. (B) trVLP assay. The trVLP assay was performed essentially as described in (A) but with transfection of a bicistronic TCRV minigenome expressing GPC-T2A-nLuc and Z (250 ng). After 72 h, these transfected (p0) cells were lysed and reporter activity measured as described in (A). In addition, the supernatants of these cells were collected and used to infect fresh target (P1) cells that were either untransfected (naïve) or had been transfected with pCAGGS-TCRV NP (15 ng) and pCAGGS-TCRV L (55 ng). After a further 72 h, the p1 cells were lysed, and luciferase activity was measured for nLuc (viral RNA synthesis). The means and standard deviations of three independent experiments are shown for both (A) and (B). Statistical significance was determined using one-way ANOVA (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns = not significant). (C) Minigenome incorporation into trVLPs. RNA was isolated from supernatants of p0 cells transfected as described in (B) and including the indicated DVGs. The RNA was then reverse transcribed with a universal TCRV genome end primer and amplified by PCR using S segment-specific genome end primers (for minigenome, MG, and S segment-derived DVGs) or L segment-specific genome end primers (for L segment DVGs). The bicistronic minigenome components, which represent the minimal requirements for viral RNA synthesis, markedly increased the levels of reporter activity, but also masked the effects of the DVG coexpression (Fig. 10B, right panel).
Consistent with the ability of del-DVG coexpression to reduce reporter activity in target p1 cells, there was a clear reduction in the amount of incorporated minigenome in trVLPs produced in the presence of S122::3344 and S97::3314, and to a lesser extent also the L segment-derived DVGs L68::6976, L460::6348, and L190::6936 (Fig. 10C). No marked effects on minigenome incorporation were noted for the longest DVG S1995::3322, despite it also being robustly expressed and incorporated into the trVLPs (Fig. 10C).
## Identification of mechanisms contributing to DVG formation
The observed hyper-prevalence of a few types of DVGs with specific pairs of break start/ stops clearly suggests that a specific biological mechanism regulates their formation at these positions. We, therefore, more closely examined the sequences surrounding these sites to identify common features that might be of mechanistic significance. An analysis of the most common del-DVGs (Fig. 5) for sequence identity between break start/stop positions revealed no significant homology (Fig. S5). However, a similar analysis of the most prevalent cb-DVGs with a single set of break start/stops revealed clearly elevated levels of sequence identity surrounding the breakpoints (Fig. 11A). For the S segment 3′ UTR and 5′ UTR cb-DVGs, as well as the L-segment 3′ UTR cb-DVGs, these regions of homology were either immediately before or after the break points (11Ai through Aiii). In contrast, for the L-segment 5′ UTR cb-DVGs, the region of increased homology was more evenly centered around the break point, so that identical residues were found both before and after the break start/stop positions (Fig. 11Aiv andAv).
Given that regions of local sequence identity did not appear to be playing a role in the generation of del-DVGs (despite the fact that we also observed highly preferred sites for their formation), we further investigated a possible role of specific nucleotide triplet sequences in their formation, since this has been suggested to play a role in regulating DVG formation for other viruses. To this end, we analyzed all triplets present in a region spanning 8 nt before and after each break start/stop site for the most prevalent del-DVGs. Here we observed that three triplets, that is, TAG, AGA, and GAA, occurred in sequences flanking the break starts/stops in all six of these highly abundant del-DVGs (Fig. 11B; see Table S23 at https://zenodo.org/records/14900940). This represents a statistically improbable overrepresentation compared to both what would be expected assuming an equal distribution of triplet sequences within the genome and when taking into account the triplet frequency actually found within the TCRV genome (see Table S23 at https://zenodo.org/records/14900940). Intriguingly, these triplets also often occurred in overlapping combinations, that is, TAGA, AGAA, or TAGAA, and often directly at the break sites. In contrast, no significant enrichment of specific triplet sequences was observed for sequences flanking break start/stop sequences associated with cb-DVG generation (see Table S23 at https://zenodo.org/records/14900940).
## DISCUSSION
DVGs are defective byproducts of RNA virus replication that, unlike their parental standard viral genomes, contain modifications that make them incapable of supporting an infection on their own, and as such, they have traditionally been excluded from consideration with respect to virus biology. However, a growing body of evidence suggests that the production of DVGs can play a crucial role in determining infection outcome both by regulating virus infection, including during the establishment of persistence, as well as by helping to shape the host immune response to infection (reviewed in reference 12). In light of these contributions to virus biology, there is increasing appreciation regarding the need to better understand the generation and biological activities of such DVGs.
Currently, little is known about the structure or biological functions of DVGs during arenavirus infection, although their existence has long been suspected for LCMV, based on evidence for DI particle production, which appears to play a role in persistence both in vitro (23,33) and in vivo (22). Support for the association of DVGs with both DI formation and persistence is provided by the accumulation of diversely sized shorter genome forms during in vivo infection (18), as well as the association of persistent virus infection of cell cultures with the generation of unusual DVG forms that either lack or contain additional nucleotides in the genome termini (19). With the exception of these DVGs with modified genome termini reported for LCMV, until recently, nothing was known regarding the actual detailed sequences of the DVGs produced by arenaviruses. However, recent work in this area has shown that not only does interference occur during acute infection of A549 cells at high MOIs, but it is also associated with the formation of a wide range of DVGs (24). Most notably, this analysis led to the identification of a highly prevalent S segment DVG with a small portion of the GPC gene and part of the IGR deleted, which the authors could show is specifically enriched in high passage stocks and has potent interfering activity, which it appears to exert by reducing the expression of GPC (24). In comparison to LCMV, less is known about the biological roles of DI particles and/or DVG formation during infection with other arenaviruses; however, limited evidence also suggests that these other arenaviruses can indeed also produce DI particles, and that their formation is associated with the accumulation of non-standard viral RNA products (20,21,24,34). In particular, early work with TCRV showed that DI particles from persistently infected BHK cells lost their standard S and L segments while acquiring five to six discrete smaller RNA species, at least one of which was likely derived from the L segment based on its size (which was greater than that of the full-length S segment) (20).
To better understand the formation of DVGs during New World arenavirus infection, we employed a classical approach to enrich DVG content in viral stocks by serial passage at high MOI. As expected, this resulted in accumulation of subgenomic viral RNAs at the expense of full-length virus genomes, and this, in turn, correlated with reduced viral titers in these DVG-enriched stocks (Fig. 2). For one of the larger DVGs that was identified, with a size of approximately 2.1 kb, direct Sanger sequencing of the PCR product showed that it is missing the majority of the GPC ORF (Fig. 3). Interestingly, studies looking at persistently arenavirus-infected cell cultures have previously reported that such cells can continuously express NP (18,20,(35)(36)(37)(38)(39)(40)(41), but show reduced or no mature glycoprotein on the cell surface (20,35,36,(40)(41)(42). Such results might be explained by the presence of a DVG in these persistently infected cells, similar to the S segment-derived 2.1 kb DVG we report in this study. With an intact NP ORF (and an L segment with an intact L ORF), these DVGs would be able to continuously replicate and express NP but would not support infectious particle production, unless GPC was supplied in trans (e.g., through reinfection with a standard virus). Notably, a previous study examining the sequences of DVGs produced during LCMV and Candid#1 infection also detected products of the S segment in which large parts of either the NP or GPC gene were deleted (24), some of which then closely resemble the S segment 2.1 kb DVG we identified for TCRV (Fig. 3), supporting that such DVGs are a common feature of infection with several different arenaviruses.
As reported by others, the subgenomic RNA populations produced during arenavirus infections are often highly heterogeneous (18,21), and this is also consistent with what we observed here (Fig. 2; Fig. S2). To overcome challenges in sequencing these complex product pools, we performed nanopore-based sequencing to allow long single-read sequencing of entire DVG products following amplification by RT-PCR using primers specific for the genome ends. Based on our analysis, we could show that TCRV infec tion produces both del-DVGs and cb-DVGs from both the S and L genome segments. Interestingly, we found that the prevalence of these different DVG species differed markedly between the two segments, with the S segment producing a similar proportion of del-DVGs and cb-DVGs reads (61% vs 39%), while the L segment produced vastly more del-DVGs reads (i.e., 97% vs 3%) (Fig. S3). While this relatively small proportion of L segment cb-DVGs appears to be consistent with what was recently reported for LCMV, that study also observed only a very small proportion (<1%) of S segment cb-DVGs compared to what we observed (24). It is possible that the enhanced presence of S segment cb-DVGs in our samples represents an inherent difference between TCRV and LCMV (e.g., as New World and Old World arenaviruses, respectively) (24). However, since the same previous study also showed similarly low levels of cb-DVG formation using the New World arenaviruses Candid#1 (the JUNV vaccine strain) and Paraná virus ( 24), it appears more likely that experimental considerations are involved. One possible explanation is that the previous study examined DVG formation in samples collected after a primary infection for 48 h ( 24), while our study focused on the use of serial passaging to enhance the accumulation of DVGs. As such, it will be imperative that future studies explore both the diversity of DVG production and the temporal dynamics of their emergence during infections performed under various experimental conditions (i.e., single-cycle vs. multi-passage infection, acute vs persistent infection of cultures, infection of dead-end vs natural host cells). Analysis of the frequency with which individual breakpoints were used to generate the observed DVGs allowed us to identify hotspots for the generation of both del-DVGs and cb-DVGs. We found that these were disproportionately located close to the genome ends, and that this was the case regardless of whether the DVGs were derived from the S or L segment, or whether they were del-DVGs or cb-DVGs (Fig. 5 to 9). Use of these preferred breakpoints then correspondingly directed the generation of relatively small DVG products that for the S segment ranged in size from ca. 200 to 700 bp, while for the L segment they ranged from 200 to 2,500 bp. Interestingly, we did not observe abundant DVG populations corresponding to those recently reported for LCMV with deletions of the GPC gene end and a portion of the IGR (24). We considered that this may have been due, at least in part, to the fact that our workflow defined DVGs for further analysis as those sequences having breakpoints >100 nt (to reduce the risk that DVGs were identified as a result of minor sequencing or alignment errors); however, lowering this threshold did not appreciably alter the results (Fig. S6 through S9). Rather, for del-DVGs, the only more frequently used breakpoints we observed in the central portion of the genome corresponded to the same 2.1 kB product that we had previously characterized by Sanger sequencing. However, this product was still a clear minority population compared to del-DVGs with breakpoints in the first/last few hundred nucleotides of the genome (Fig. 5A), which resulted in DVG lengths that were mostly around 200 nt in length.
Although our passaging data already clearly suggested that DVG accumulation was associated with low virus titers (Fig. 2), the identification of highly abundant del-DVGs by our study offered the opportunity to also investigate their biological functions directly. Therefore, we analyzed the impact of del-DVG coexpression in both a minigenome system, which models viral RNA synthesis, and a trVLP assay, which allows analysis not only of viral RNA synthesis (using a longer minigenome closer in size to the naturally occurring S segment) but also packaging of viral RNAs into particles and their delivery into target cells. These data clearly indicated that del-DVGs derived from both the S and L segments can compete with genome analogs (i.e., minigenomes) for the viral replication machinery, and thereby limit the expression of the encoded protein(s) (Fig. 10A andB), and that their ability to do so is length-dependent. Furthermore, we found that this del-DVG-mediated inhibition takes place not only in the cell in which the DVGs are initially produced, but also in new target cells into which they can be introduced as a result of their efficient incorporation into viral particles, which occurs at the expense of the minigenome (Fig. 10B andC). Importantly, differences in the results obtained using naïve target cells and those pre-transfected to overexpress the components of the replication machinery (i.e., NP and L) appear to support that inhibition by these del-DVGs is occurring through competition for a limiting supply of the viral RNA machinery, since overexpressing these components largely abolishes the impact of del-DVG coexpression. As such, these data strongly support that several of the highly prevalent del-DVGs observed in our study are contributing to the negative impact of DVG accumulation on virus stock titers that we observed during virus passage.
In contrast to the approach based on nanopore sequencing used here, most studies of DVG formation to date have been based on classical short-read next-generation sequencing approaches (17,24,(43)(44)(45)(46)(47). Therefore, it is worth noting that our own data appear to correspond well to those of a recent analysis of JUNV (Candid#1) DVG formation (24), in that both reveal abundant formation of short DVGs with breakpoints near the genome termini as well as DVGs in which the majority of a single ORF within the S segment is deleted (i.e., similar to our 2.1 kb S segment DVG). This is despite the fact that DVGs in these two studies were generated using different infection approaches (i.e., single-cycle infection vs. passaging), amplified and sequenced using different methods (NGS vs RT-PCR and MinION), and analyzed using different workflows. However, while such NGS-based approaches also allow sequencing at the single-molecule level and can identify individual breakpoints, their short read lengths mean that they are limited in their ability to provide insight into more complex structural rearrangements, such as the multiple breakpoint cb-DVGs that we report here (Fig. 6 to 9). We found that many of these structures contained both the 3′ and 5′ UTR, as well as complements thereof, while lacking the majority of the sequence information between these elements. As such, they appear to have been generated by a combination of internal deletion and copyback events. Interestingly, while our data set also includes a small number of full-length genome reads (i.e., 199 S segment reads), we never observe structures containing duplications of the entire genome. This suggests that the process of internal deletion is necessary for the formation of structures with duplicated genome termini, although it remains unclear whether these deletion and copyback events are occurring within a single cycle of replication or as separate successive events. Given that these DVGs exhibit extensive potential for the formation of dsRNA structure (Fig. 7 to 9), which can in some cases include the entire DVG sequence, these may be particularly relevant as ligands for immune sensing during virus infection, and thus the ability to detect them appears to be a specific advantage of our approach. Indeed, a particularly important biological role of arenavirus DVGs with extensive dsRNA structure is suggested by the recent observation that arenaviruses possess an unusual dsRNA-specific exonuclease activity encoded by the viral NP that has been suggested to degrade viral RNAs that could otherwise serve as PAMPs for the activation of diverse RNA sensors (25)(26)(27)(28)(29)(30)(31). Supporting this, recent work has also shown that dsRNA (potentially in the form of DVGs, and especially cb-DVGs) indeed accumulates much more robustly in the cytoplasm of New World arenavirus-infected cells than for Old World arenaviruses, and that this appears to be dependent on the virus' exonuclease function (32,48). This is then potentially also consistent with the role of exacerbated pro-inflammatory responses (i.e., cytokine storm) in the severe forms of disease associated with New World arenaviruses, but not Old World arenaviruses (reviewed in reference 49). However, again here, a lack of knowledge about the characteristics of such viral dsRNA products has so far limited our ability to directly investigate their role in the virus' biology. The increasing availability of data, including our own, precisely describing the sequence characteristics of individual arenavirus DVGs now opens up the possibility to analyze their biological properties in detail, including with respect to roles as both competitors for the replication machinery, as we have already begun to do here in this study, and as immune agonists.
While for many viruses, there remains little known about what triggers the detach ment/reattachment events that drive DVG formation at specific sites, a growing body of evidence suggests that both virus and host factors can play a role, and in particular that specific sequence and/or structural elements in the viral genome can contribute to the formation of different types of DVG (reviewed in reference 13). For the IGR del-DVGs that have previously been reported for LCMV, it has been suggested that these are generated as a result of the highly structured nature of this RNA region (24). However, given that the arenavirus genome is encapsidated by NP, which should limit secondary structure formation, it appears unlikely that such a mechanism would be playing a role in other regions of the genome. Nonetheless, our data clearly show strong preferences for the use of specific breakpoints within regions outside the IGR (Fig. 5 to 8). Here, we clearly observed that the generation of cb-DVGs was associated with high levels of local sequence identity in proximity to the breakpoints (Fig. 11A). Interestingly, something similar has been observed for positive-sense RNA viruses, where partial sequence homology between break starts and stops has led to the hypothesis that for these viruses, DVG generation is driven by homologous recombination (50,51). And while such a mechanism might be considered unlikely for negative-sense RNA viruses (due to the encapsidation of their genomes within the ribonucleoprotein complex), it is important to note that arenaviruses are known to have undergone recombination in nature (i.e., to generate the Clade A/B viruses) (52), and thus a homologous recombi nation-driven DVG generation cannot be so easily excluded. In contrast, for del-DVGs, we observed that highly favored break points were associated with the presence of specific nucleotide sequences, that is, TAG, AGA, and GAA, and that these frequently also occurred in longer overlapping combinations, that is, TAGA, AGAA, or TAGAA (Fig. 11B). This then appears similar to what has been reported for RSV (17) and also influenza virus (16), where the presence of specific sequence motifs also seems to be associated with DVG formation. However, the fact that certain break start/stop combinations are also highly favored (e.g., for the S segment, the combinations 97::3314 and 122::3344 are frequently observed, whereas 97::3344 was never observed, and 122::3314 only very rarely) suggests that additional sequence and/or structural elements must play a role in the pairing of break starts and stops, although their nature remains elusive at this time. Nonetheless, taken together, this collective evidence for the existence of common sequence elements/features among genetically diverse virus families supports the hypothesis that the generation of DVGs, including by arenaviruses, is not a random process but is rather influenced by these local sequence features.
In addition to its scientific findings, the presented work establishes a new workflow for the analysis of sequencing data for the detection of del-DVGs, as well as cb-DVGs generated from either genome end, based on long single-read sequencing data, such as is generated by nanopore sequencing (Fig. 4). Indeed, to the best of our knowledge, the only tool currently available to handle long-read nanopore sequencing data is BBMAP v36; however, it does not allow for the analysis of cb-DVGs (53). Almost all other currently existing analysis tools for DVG analysis are only designed to handle short-read Illumina (or 454) next-generation sequencing data (17,(44)(45)(46)(47)54). Notably, this severely limits the ability of these approaches to detect multiple-breakpoint DVGs, which our study identified as composing a substantial proportion of cb-DVG reads from both genome segments. Furthermore, only a few of these existing DVG analysis tools are capable of detecting both del-DVGs and cb-DVGs (i.e., VODKA2 [47], DI-tector [46] and DVG-profiler [44]), and even then they contain assumptions that may bias them, for instance, in favor of the detection of only 5′ UTR copyback reads (i.e., those starting from the 3′ genome end of the cRNA, proximal to the arenavirus GPC gene) (47). While this is based on the observation that 5′ trailer (UTR) cb-DVGs appear to be more prevalent for negative-sense RNA viruses, it is not clear that this assumption is valid for ambisense viruses like the arenaviruses. In contrast, our approach based on alignment to reference genomes with both duplicated complementary 3′ UTRs or 5′ UTRs allows equally for the detection of cb-DVGs formed starting from either genome terminus. However, in this study, we initially imposed the additional restriction that only gaps between break start and stop sites of >100 bp were reported, to conservatively exclude the possibility of false positives due to minor mismatches during alignment (including due to sequence errors, which are still potentially a greater challenge for nanopore sequence data than for short-read next-generation sequencing methods). While this potentially limits our ability to detect del-DVGS with extremely short deletions like those found in the IGR of LCMV (24), repeating the analysis with thresholds of 50 or 25 nt did not result in notable changes to our data (Fig. S6 through S9). Given its ability to handle nanopore sequencing data and to detect not only del-DVGs but also both 3′ and 5′ UTR cb-DVGs, as well as its fundamentally different assumptions and biases compared to existing tools, we suggest that this workflow represents a valuable complement to the currently available resources for the study of DVGs.
Furthermore, it is worth noting that a potential caveat of current DVG sequencing approaches is their reliance on RNA to DNA conversion and DNA amplification prior to or during sequencing. In particular, deletions involving the highly structured arenavirus IGR have been reported to occur as a product of amplification by RT-PCR (55). While this potential to generate aberrant products through RT-PCR amplification itself remains a significant challenge to many studies that look at DVG formation during a single round of infection, we avoid this issue by focusing on DVGs that only accumulate over sequential rounds of passaging. Importantly, in these experiments, we do not see significant accumulation of subgenomic products during RT-PCR amplification with non-passaged virus stocks (which contain higher levels of full-length genome template) (Fig. 2), but rather these only occur as a product of passaging (and at the expense of the full-length S segment), thus strongly suggesting that the formation of these structures has a biological basis and is not an RT-PCR artifact. Furthermore, upon passaging of our p20 stock, we observed spontaneous reversion from a high DVG content/low titer (p22) to a low DVG content/high titer (p23) state (Fig. 2). Such an event is consistent with an interfering nature of these DVGs and with competition between DI and standard viruses, but inconsistent with them being RT-PCR artifacts. Nonetheless, these ampli fication steps remain a technical challenge for many of these kinds of studies, and especially those focused on studying DVG generation during a single cycle of infection. While direct RNA sequencing using nanopore technology has recently been used as a supplemental approach to avoid this issue and validate individual findings from classical next-generation sequencing-based studies of DVG formation (24), it is still currently not a viable option for de novo sequence determination, due to problematically high levels of sequencing errors that arise as a result of extensive RNA modifications (56). Nonethe less, it is to be hoped that further technical improvements may make direct nanopore sequencing of RNA a viable approach for such studies in the future. At that time, the scripts that we have developed as part of our analysis pipeline could be seamlessly adapted to such data by substituting an RNA rather than a DNA reference sequence.
Taken together, our work presents not only a novel workflow for the analysis of DVGs present in nanopore sequencing data but also provides insight regarding DVG diversity during TCRV infection by identifying the most highly enriched DVG products of both genome segments. Based on this information, we could demonstrate a lengthdependent competition for interaction with the viral RNA synthesis machinery by several short del-DVGs identified in our study. Finally, the identification of a substantial number of highly enriched del-DVG and cb-DVG species has allowed us to uncover genome sequence features that are associated with their production. As such, our work will not only enable future research investigating the biological activities of specific arenavirus DVGs, but also open up the possibility to regulate their production (i.e., by modifying the sequence elements associated with their formation) to study their biological impact and modulate infection outcome.
## MATERIALS AND METHODS
## Cells and viruses
TCRV (strain TRVL-11573) was kindly provided by Dan Kolakofsky and Dominique Garcin (University of Geneva). Vero76 cells (African green monkey kidney; Collection of Cell Lines in Veterinary Medicine [CCLV]-RIE 0228) were maintained in Dulbecco's Modified Eagle Medium (DMEM) + 10% fetal calf serum (FCS) and 100 U/mL penicillin, and 100 µg/mL streptomycin (P/S). Cells were cultured at 37°C in 5% CO 2 .
## Virus passaging
Vero76 cells were seeded for 80%-90% confluence in 12-well plates. Initial infections were performed in triplicate at MOI = 0.5 using TCRV stock virus diluted in DMEM without FCS. After 1 h at 37°C, the inoculum was exchanged for 2 mL DMEM + 2% FCS. After incubation for 7 days, 500 µL of undiluted cell culture supernatant was used to infect freshly seeded Vero76 cells using the same approach as for the initial infections. The remaining supernatant was stored at -20°C until all samples had been collected before being processed as described in the sections below.
## Plaque assay
Vero 76 cells were seeded into 12-well plates for a confluence of ~90% on the following day. They were infected for 1 h at 37°C with serial dilutions of virus stocks collected at the indicated passages. After 1 h, the inoculum was removed, and an overlay composed of a 1:1 mixture of 2× MEM + 4% FCS, and 1.4% agarose was added. Plates were fixed and stained 10 days post-infection by incubation with 10% formalin containing crystal violet.
## RT-PCR amplification for Sanger sequencing
RNA was isolated from serially passaged TCRV p1, p5, p10, p15, and p20 samples (three independent replicates each) using the QIAamp Viral RNA Mini Kit (Qiagen). cDNA was generated using SuperScript III Reverse Transcriptase (Invitrogen) with a universal L and S segment 3′ UTR primer (CGCACAGTGGATCCTAGGC). For initial analysis of DVG accumula tion, cDNA was then amplified by PCR using S segment-specific genome end primers (S segment 3′ UTR Fwd: GGCAAATTGTCTAACTCTTTCACTGAG, S segment 5′ UTR Rev: C CTAGGCATTTCTTGACCATATTTGC). The resulting PCR products were then separated by agarose gel electrophoresis and stained with ethidium bromide. Subgenomic products of >1 kb were gel extracted using the NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel) before being sent for commercial Sanger sequencing (Eurofins). The resulting data were analyzed by alignment to a TCRV S segment reference sequence (GenBank Accession number: MT081316 [57]) using Geneious Prime (Biomatters).
## RT-PCR amplification for nanopore sequencing
For nanopore sequencing of DVGs, separate PCRs using cDNA (generated as described above from p20 virus samples; three replicates) were performed for each genome segment using the iProof High-Fidelity PCR Kit (Bio-Rad) with segment-specific genome end primers (S segment primers, as above; L segment 3′ UTR Fwd: ATCCTAGGCGGCAC TTGACC, L segment 5′ UTR Rev: CCTAGGCGTTACGTGCACTC). To amplify del-DVGs, both forward and reverse primers were used for amplification of the respective segment, while for the detection of cb-DVGs (which contain only one genome end and its complement), PCR was run with either only the forward or only the reverse primer. All PCR products were purified using the NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel) prior to further analysis.
## Nanopore library preparation and sequencing
Following PCR amplification, a nanopore sequencing run (Oxford Nanopore Technolo gies) was prepared using the 1D2 kit according to the manufacturer's protocol (version LSD_9032_v11_revO_23Mar2017). Briefly, an equal mass of PCR product from each reaction was combined for a total of 1 µg, and nuclease-free water was added to a final volume of 45 µL. No barcoding was implemented in this approach, but a linker sequence, as well as an adapter (to allow binding of the DNA to the nanopores), was added to the DNA in successive reactions, with purification of the DNA using magnetic AMPure XP beads between the reactions. Finally, the prepared sample was loaded onto a MinION sequencer containing a FLO-MIN107 R9 flow cell, also as per the manufactur er's instructions. The flow cell used was confirmed to have 1,360 functional pores. For sequencing, the MinKNOW software (version 2.2.15) with integrated live base-calling was used. The sequencing run was allowed to proceed for 16 h.
## Analysis of nanopore sequencing data
Data were analyzed according to the workflow shown in Fig. 4 using the scripts detailed in the supplemental information (Supplementary Methods) running under Ubuntu 18.04.6. Following this approach, the obtained sequences were first divided into S-segment or L-segment sequences by searching for the primer sequences used in PCR amplification at positions close to the genome termini using flexbar (v3.1) (58). Flexbar parameters were optimized for this purpose using a test set of 400 sequences and gave a sensitivity of 89% and a specificity of 99% when compared manually to alignments of the reads to reference sequences for the S or L segment (GenBank accession numbers: S segment, MT081316; L segment, MT081317 [57]) using Geneious Prime (Biomatters).
Reads that mapped to a specific genome segment (i.e., S or L) were further sorted based on local alignment to four different reference sequence templates using lastal (version 921) (59) with parameters that had been optimized for nanopore data (60). These four templates corresponded to (i) the vRNA-sense genome segment, (ii) the cRNA-sense genome segment, (iii) a vRNA-cRNA concatemer, and (iv) a cRNA-vRNA concatemer. For each read, the number of local matches to each of the four templates (i.e., the vRNA, cRNA, vRNA_cRNA, or cRNA_vRNA match counts) was determined, and this information was used to deduce the DVG type. If there was only a single match, the read was discarded from the analysis, since it then corresponded to a full-length genome rather than a DVG. In all other cases, if the vRNA or cRNA match count was as high or higher than all other counts, then the read was considered a del-DVG. Impor tantly, however, while vRNA and cRNA reads were aligned and analyzed separately in the bioinformatics workflow, due to the PCR amplification step, a distinction between vRNA and cRNA input material is not possible, as either RNA species can be amplified into a double-stranded PCR product, of which one random strand was then sequenced through the nanopore. In contrast, if the vRNA_cRNA or cRNA_vRNA match count was higher than the vRNA and cRNA match count, then the read was identified as a cb-DVG. Here, the order of the local alignments can only correspond to either a vRNA_cRNA or a cRNA_vRNA reference sequence, and thus the order of these local alignments was used to determine whether reads corresponded to 3′ or 5′ UTR cb-DVGs.
## Analysis of break points
Once sorted into categories, reads and reference sequences for each local alignment were trimmed to remove barcodes and linker sequences from the reads and to ensure consistent ends. A global alignment was then performed using the appropri ate reference template (i.e., vRNA, cRNA, cRNA_vRNA, or cRNA_cRNA) and the Needle-Wunsch algorithm as implemented in needle (version: EMBOSS:6.6.0.0) (61) with a high gap open penalty (i.e., 50) but a very low gap extend penalty (i.e., 0.000000001) to allow the algorithm to tolerate the large deletions that are expected for many DVGs. The output of the Needle-Wunsch algorithm was then examined for breakpoints, with a gap of at least 100 nt being defined as a break. This conservative cutoff was chosen to avoid false-positive detection of breaks due to sequencing inaccuracies; however, subsequent analysis showed that the results remained virtually unchanged if cutoffs of 50 nt or 25 nt were used (Fig. S6 through S9). A "break start" was defined as the last nucleotide of a reference that aligns to a read before the deletion, while a "break stop" was the first nucleotide of the read that again aligns to the reference sequence after the deletion. Break start and stop points for each DVG were then extracted and further analyzed in Excel (Microsoft). Since the DVG products analyzed were generated by PCR (and thus both strands are present for any given DVG), the break positions for del-DVGs in cRNA orientation were converted into their vRNA equivalents, and the data were combined.
## Minigenome assay
Huh7 cells were seeded into 12-well plates for a confluency of ~60% on the next day. They were then transfected with a nanoluciferase (nLuc)-expressing monocistronic TCRV minigenome (125 ng [62]), pCAGGS-T7 (125 ng), pCAGGS-TCRV L (100 ng), pCAGGS-TCRV NP (25 ng), and pCAGGS-firefly-luciferase (FF, 50 ng; as a transfection control) using Transit-LT1 with a ratio of 3 µL to 1 µg of DNA. As controls, samples without pCAGGS-L (-L) and with only half the amount of transfected pCAGGS-NP and pCAGGS-L (+L 0.5×) (i.e., 50 ng pCAGGS-L and 12.5 ng pCAGGS-NP) were prepared. To assess the inhibitory activity of DVGs, 125 ng of the indicated DVG was additionally transfected. Total plasmid amounts were equalized between samples by adding empty pCAGGS. After 24 h, the medium on the transfected cells was exchanged against fresh DMEM + 10% FCS, and after a further 24 h, the cells were lysed in 1% Triton-X100, and luciferase activity was measured using Nano-Glo (for nLuc) and Beetle Juice (for FF) reagents in a Glomax Multi microplate reader. Normalized luciferase values were calculated based on the nLuc (viral RNA synthesis) and FF (host cell RNA synthesis) values.
## Transcription and replication-competent virus-like particle assay
Huh7 cells were seeded as described above for the minigenome assay. These cells (i.e., p0 cells) were transfected with a bicistronic TCRV minigenome expressing the GPC linked by a T2A sequence to nLuc (GPC-T2A-nLuc) and Z (250 ng [62]), pCAGGS-T7 (125 ng), pCAGGS-TCRV L (100 µg), pCAGGS-TCRV NP (25 ng), and pCAGGS-FF (50 ng) using Transit-LT1 at a ratio of 3 µL to 1 µg of DNA. After 24 h, the medium on the transfected cells was exchanged against fresh DMEM + 10% FCS, and after a further 48 h, supernatants were harvested, and the p0 cells were lysed in 1% Triton-X100, and luciferase activity was measured using Nano-Glo (for nLuc) and Beetle Juice (for FF) reagents in a Glomax Multi microplate reader. The supernatants were cleared of debris by centrifugation at 500 × g for 5 min and used to infect fresh Huh7 cells at a confluence of ~50% in 96-well plates (i.e., p1 cells) in triplicate. These p1 cells were either untransfected (i.e., naïve) or had been transfected the day before (i.e., pre-transfected) with pCAGGS-TCRV NP (15 ng) and pCAGGS-TCRV L (55 ng) per well using Transit-LT1. After 72 h, the p1 cells were lysed in 1% Triton-X100, and luciferase activity was measured using Nano-Glo (for nLuc) in a Glomax Multi microplate reader. Normalized luciferase values were calculated for p0 cell lysates based on the nLuc (viral RNA synthesis) and FF (host cell RNA synthesis) values, while only nLuc (viral RNA synthesis) values are shown for p1 cells.
## References
1. Barton, Hyndman (2000) "Lymphocytic choriomeningitis virus: reemerging central nervous system pathogen" *Pediatrics*
2. Ogbu, Ajuluchukwu, Uneke (2007) "Lassa fever in West African subregion: an overview" *J Vector Borne Dis*
3. Paweska, Sewlall, Ksiazek et al. (2009) "Nosocomial outbreak of novel arenavirus infection, southern Africa" *Emerg Infect Dis*
4. Charrel, De Lamballerie (2003) "Arenaviruses other than Lassa virus" *Antiviral Res*
5. Delgado, Erickson, Agudo et al. (2008) "Chapare virus, a newly discovered arenavirus isolated from a fatal hemorrhagic fever case in Bolivia" *PLoS Pathog*
6. Oldstone (2002) "Arenaviruses. I. The epidemiology molecular and cell biology of arenaviruses. Introduction, p V-XII"
7. Pinschewer, Perez, De La Torre (2005) "Dual role of the lympho cytic choriomeningitis virus intergenic region in transcription termination and virus propagation" *J Virol*
8. Perez, De La Torre (2003) "Characterization of the genomic promoter of the prototypic arenavirus lymphocytic choriomeningitis virus" *J Virol*
9. Ziegler, Botten (2020) "Defective interfering particles of negativestrand RNA viruses" *Trends Microbiol*
10. Genoyer, López (2019) "The impact of defective viruses on infection and immunity" *Annu Rev Virol*
11. Perrault (1981) "Origin and replication of defective interfering particles" *Curr Top Microbiol Immunol*
12. Vignuzzi, López (2019) "Defective viral genomes are key drivers of the virus-host interaction" *Nat Microbiol*
13. Brennan, Sun (2024) "Defective viral genomes: advances in understanding their generation, function, and impact on infection outcomes"
14. (2026) *Full-Length Text Journal of Virology*
15. Manzoni, López (2018) "Defective (interfering) viral genomes reexplored: impact on antiviral immunity and virus persistence" *Future Virol*
16. López (2014) "Defective viral genomes: critical danger signals of viral infections" *J Virol*
17. Jennings, Finch, Winter et al. (1983) "Does the higher order structure of the influenza virus ribonucleoprotein guide sequence rearrangements in influenza viral RNA?" *Cell*
18. Sun, Kim, Felt et al. (2019) "A specific sequence in the genome of respiratory syncytial virus regulates the generation of copy-back defective viral genomes" *PLoS Pathog*
19. Francis, Southern (1988) "Molecular analysis of viral RNAs in mice persistently infected with lymphocytic choriomeningitis virus" *J Virol*
20. Meyer, Southern (1997) "A novel type of defective viral genome suggests a unique strategy to establish and maintain persistent lymphocytic choriomeningitis virus infections" *J Virol*
21. Gimenez, Compans (1980) "Defective interfering Tacaribe virus and persistently infected cells" *Virology (Auckl)*
22. Johnson, Cubitt, Pfeffer et al. (2021) "Lassa virus vaccine candidate ML29 generates truncated viral RNAs which contribute to interfering activity and attenuation" *Viruses*
23. Popescu, Lehmann-Grube (1977) "Defective interfering particles in mice infected with lymphocytic choriomengingitis virus" *Virology*
24. Peralta, Lehmann-Grube (1983) "Properties of lymphocytic choriomeningitis virus interfering particles" *Arch Virol*
25. Hackbart, López (2024) "Characterization of non-standard viral genomes during arenavirus infections identifies prominent S RNA intergenic region deletions"
26. Hastie, Kimberlin, Zandonatti et al. (2011) "Structure of the Lassa virus nucleoprotein reveals a dsRNA-specific 3′ to 5′ exonuclease activity essential for immune suppression" *Proc Natl Acad Sci*
27. Qi, Lan, Wang et al. (2010) "Cap binding and immune evasion revealed by Lassa nucleopro tein structure" *Nature*
28. Jiang, Huang, Dong et al. (2013) "Structures of arenaviral nucleoproteins with triphosphate dsRNA reveal a unique mechanism of immune suppression" *J Biol Chem*
29. Zhang, Li, Liu et al. (2013) "Crystal structure of Junin virus nucleoprotein" *J Gen Virol*
30. Bohn, Waßmann, Wendt et al. (2023) "A dsRNA-binding mutant reveals only a minor role of exonuclease activity in interferon antagonism by the arenavirus nucleoprotein" *PLoS Pathog*
31. Huang, Shao, Lan et al. (2015) "In vitro and in vivo characterizations of pichinde viral nucleoprotein exoribonuclease functions" *J Virol*
32. Reynard, Russier, Fizet et al. (2014) "Exonuclease domain of the Lassa virus nucleoprotein is critical to avoid RIG-I signaling and to inhibit the innate immune response" *J Virol*
33. Mateer, Maruyama, Card et al. (2020) "Lassa virus, but not highly pathogenic new world arenaviruses, restricts immunosti mulatory double-stranded RNA accumulation during infection" *J Virol*
34. Welsh, Connell, Pfau (1972) "Properties of defective lymphocytic choriomeningitis virus" *J Gen Virol*
35. Dutko, Wright, Pfau (1976) "The RNAs of the defective interfering Pichinide virus" *J Gen Virol*
36. Lehmann-Grube, Slenczka, Tees (1969) "A persistent and inapparent infection of L cells with the virus of lymphocytic choriomeningitis" *J Gen Virol*
37. Welsh, Buchmeier (1979) "Protein analysis of defective interfering lymphocytic choriomeningitis virus and persistently infected cells" *Virology*
38. Dimock, Harnish, Sisson et al. (1982) "Synthesis of virus-specific polypeptides and genomic RNA during the replicative cycle of Pichinde virus" *J Virol*
39. Ellenberg, Edreira, Scolaro (2004) "Resistance to superinfection of Vero cells persistently infected with Junin virus" *Arch Virol*
40. Damonte, Mersich, Coto (1983) "Response of cells persistently infected with arenaviruses to superinfection with homotypic and heterotypic viruses" *Virology*
41. Ellenberg, Edreira, Lozano et al. (2002) "Synthesis and expression of viral antigens in Vero cells persistently infected with Junin virus" *Arch Virol*
42. Staneck, Trowbridge, Welsh et al. (1972) "Arenaviruses: cellular response to long-term in vitro infection with parana and lymphocytic choriomeningitis viruses" *Infect Immun*
43. Welsh, Burner, Holland et al. (1975) "A comparison of biochemical and biological properties of standard and defective lymphocytic choriomeningitis virus" *Bull World Health Organ*
44. Timm, Akpinar, Yin (2014) "Quantitative characterization of defective virus emergence by deep sequencing" *J Virol*
45. Bosma, Karagiannis, Santana-Quintero et al. (2019) "Identification and quantification of defective virus genomes in high throughput sequencing data using DVG-profiler, a novel postsequence alignment processing algorithm" *PLoS One*
46. Boussier, Munier, Achouri et al. (2020) "RNA-seq accuracy and reproducibility for the mapping and quantification of influenza defective viral genomes" *RNA*
47. Beauclair, Mura, Combredet et al. (2018) "DI-tector: defective interfering viral genomes' detector for nextgeneration sequencing data" *RNA*
48. Achouri, Felt, Hackbart et al. (2023) "VODKA2: a fast and accurate method to detect non-standard viral genomes from large RNA-seq data sets" *RNA*
49. Mateer, Paessler, Huang (2018) "Visualization of double-stranded RNA colocalizing with pattern recognition receptors in arenavirus infected cells" *Front Cell Infect Microbiol*
50. Mclay, Liang, Ly (2014) "Comparative analysis of disease pathogen esis and molecular mechanisms of New World and Old World arenavirus infections" *J Gen Virol*
51. Poirier, Mounce, Rozen-Gagnon et al. (2015) "Low-fidelity polymerases of alphaviruses recombine at higher rates to overproduce defective interfering particles" *J Virol*
52. Xiao, Liu, Yu et al. (2011) "Identification of new defective interfering RNA species associated with porcine reproductive and respiratory syndrome virus infection" *Virus Res*
53. Archer, Rico-Hesse (2002) "High genetic divergence and recombina tion in Arenaviruses from the Americas" *Virology (Auckl)*
54. Jaworski, Routh (2017) "Parallel ClickSeq and nanopore sequencing elucidates the rapid evolution of defective-interfering RNAs in Flock House virus" *PLoS Pathog*
55. (2026) *Full-Length Text Journal of Virology*
56. Routh, Johnson (2014) "Discovery of functional genomic motifs in viruses with ViReMa-a Virus Recombination Mapper-for analysis of nextgeneration sequencing data" *Nucleic Acids Res*
57. Ye, De La Torre, Martínez-Sobrido (2020) "Development of reverse genetics for the prototype new world mammarenavirus tacaribe virus" *J Virol*
58. Liu-Wei, Van Der Toorn, Bohn et al. (2024) "Sequencing accuracy and systematic errors of nanopore direct RNA sequencing" *BMC Genomics*
59. Holzerland, Leske, Fénéant et al. (2020) "Complete genome sequence of Tacaribe virus" *Arch Virol*
60. Roehr, Dieterich, Reinert (2017) "Flexbar 3.0 -SIMD and multicore parallelization" *Bioinformatics*
61. Kiełbasa, Wan, Sato et al. (2011) "Adaptive seeds tame genomic sequence comparison" *Genome Res*
62. Hoenen, Groseth, Rosenke et al. (2016) "Nanopore sequencing as a rapidly deployable ebola outbreak tool" *Emerg Infect Dis*
63. Rice, Longden, Bleasby (2000) "EMBOSS: the European molecular biology open software suite" *Trends Genet*
64. Fénéant, Leske, Günther et al. (2022) "Generation of reporterexpressing new world arenaviruses: a systematic comparison" *Viruses* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817933&blobtype=pdf | # A recurrent adaptive mutation in the transmembrane 2B protein of an insect picorna-like virus in a nonnative host
Oscar Lezcano, Lara Fuhrmann, Reinder Bos, Haitao Wang, Milou Stevens, Niko Beerenwinkel, Martijn Huynen, Ronald Van Rij
## Abstract
Virus emergence is often due to cross-species transmission and adaptation to the new host. We studied the effect of innate immune responses on shaping virus populations in native and nonnative virus-host combinations, using as a model system Drosophila melanogaster infected with either Drosophila C virus (DCV) or cricket paralysis virus (CrPV). In this host, the cGAS-like receptor 1 senses viral double-stranded RNA and produces cyclic dinucleotides (CDNs) to activate the STING protein and induce an antiviral response. Both viruses were serially passaged in three host conditions differ ing in their cGAS-STING response: wild-type (WT) flies, Sting knock-out (KO) flies, and flies with a primed immune response by CDN injection. We found no immune-related effects on virus evolution, but we uncovered the CrPV nonstructural 2B protein as a key regulator of cross-species transmission. Nucleotide diversity specifically accumulated in the 2B gene during passage of CrPV in its nonnative Drosophila host, while 2B of the flyadapted DCV displayed markedly lower and constant nucleotide diversity. In particular, the CrPV 2B D29N variant was selected in all six virus lineages evolved in WT and Sting KO flies, with an estimated selection coefficient greater than 0.2. This variant replicated faster and was more lethal than the parental virus in all three host backgrounds. 2B is a predicted transmembrane protein, which we found to be associated with cellular endomembranes and may be involved in replication organelle formation. Our findings suggest a role for the 2B protein in adaptation to a new host independent of the cGAS-STING pathway.
IMPORTANCEThe forces driving virus evolution are central to understanding cross-spe cies transmission and virus emergence. It is well established that the adaptive immune system drives virus evolution in mammals, but whether innate responses likewise drive virus evolution upon host shifts is less well understood. In this manuscript, we used Drosophila melanogaster as a model to study the evolution of a native and a nonna tive pathogen under conditions in which innate antiviral immunity is either abolished or enhanced. Using an experimental evolution approach, we find little evidence for adaptive evolution of the natural pathogen Drosophila C virus. In contrast, we observed a recurrent adaptive mutation in the viral nonstructural 2B protein in the nonnative cricket paralysis virus, independent of the antiviral cGAS/STING pathway. Our work provides insights into viral adaptation to new hosts and the characteristics of the 2B protein of dicistroviruses, a family comprising important model insect viruses.
and the immune system of the new host (1). Indeed, host shifts appear to occur most frequently among RNA viruses (2), likely due to their error-prone replica tion and adaptability. Also, within natural hosts, the immune system is an important driver of virus-host coevolution, which is especially well established for the adaptive immune response of mammals. Models suggest that immune memory selects new surface antigens unrecognized by the host immune system, forcing the virus into uncharted antigenic space (3). Other simulations proposed that the immune response accelerates viral evolution during persistent infections (4). Indeed, in hepatitis C virus and human immunodeficiency virus-1 (HIV-1) infection, significant associations were observed between the emergence of specific viral mutants and specific human leukocyte antigen (HLA) types, likely a consequence of immune-driven selection of variants with decreased antigen-HLA binding affinity (5,6). However, also in organisms without adaptive immunity, host resistance may drive the evolution of escape variants. This is exemplified by an experimental evolution study in Escherichia coli, where the bacterio phage λ repeatedly evolved the ability to use an alternative surface receptor in response to the evolution of host resistance due to reduced expression of the original LamB receptor (7).
In mammals, the main innate antiviral pathway is the interferon response (8), which can be activated by the cGAS-STING signaling pathway, an evolutionarily conserved pathway that originated in bacteria to control bacteriophage infections (9). The cGAS protein (cyclic GMP-AMP synthase) acts as a pattern recognition receptor that senses double-stranded DNA (dsDNA) in the cytosol and, in response, synthesizes cyclic di-nucleotides (CDNs), in particular 2′3′-cyclic GMP-AMP (cGAMP). CDNs are soluble molecules that act as second messengers, activating the conserved STING protein (stimulator of interferon genes). STING is an adapter protein embedded in the endoplas mic reticulum (ER). Upon activation, it translocates to the Golgi apparatus through an unknown mechanism that is essential for downstream signaling (reviewed in reference 10). This, in turn, leads to translocation of the transcription factor IRF3 (interferon regulatory factor 3) into the nucleus and the expression of type I interferons and interferon-stimulated genes.
Despite lacking the interferon response, insects have a functional cGAS-STING pathway, albeit with marked differences compared to the mammalian pathway. For instance, Drosophila melanogaster encodes two cGAS-like receptors (cGLRs) capable of producing CDNs (11,12), and different Drosophila species may even encode up to seven cGLRs (13). Of these, cGLR1 is activated by dsRNA, an intermediate in RNA virus replication and a hallmark of RNA virus infection (13). cGLR1 then produces 3′2′-cGAMP, a strong STING agonist that activates Sting-dependent signaling, leading to the translo cation of the NF-ΚB transcription factor Relish to the nucleus and the transcriptional activation of STING-regulated genes (SRGs) (14). This response has been shown to limit viral RNA accumulation and extend the survival of flies infected with several viruses, including the dicistroviruses Drosophila C virus (DCV) and cricket paralysis virus (CrPV) (15), both positive-sense RNA viruses from the genus Cripavirus (Dicistroviridae).
Here, we aimed to gain a quantitative understanding of the specific effects of host shifts and innate immune adaptation on virus evolution. We performed an experimen tal evolution study of a native and a nonnative RNA virus in three D. melanogaster lines, differing in the activity of the cGAS-STING pathway. Along with the natural fly pathogen DCV, we used the nonnative CrPV, originally isolated from Australian field crickets (Teleogryllus oceanicus and Teleogryllus commodus) (16). While CrPV has also been found in metagenomic data from Australian honey bees (17), it has to the best of our knowledge never been reported in wild-caught flies. We passaged the viruses for 10 fly generations, keeping 3 parallel lineages per virus and condition, after which the evolved lineages were characterized by RNA-sequencing. We found limited immunerelated effects on virus evolution but observed high genetic diversity in the CrPV 2B gene. In particular, the conserved aspartic acid (D) at position 29 was substituted by an asparagine (N) in all virus lineages evolved in wild-type (WT) and Sting null mutant flies. We found that the CrPV D29N variant replicated faster and was more virulent than the WT CrPV strain in all fly conditions (virulence defined here as the capacity to cause host damage [18]). Our results suggest that viruses may not readily evolve resistance to cGAS/STING immunity and highlight the importance of the nonstructural 2B protein in viral adaptation to a non-native host species.
## RESULTS
## Experimental CrPV and DCV evolution
To investigate the influence of the cGAS-STING pathway on host adaptation and virus evolution, we experimentally evolved DCV and CrPV in flies differing in their immune response: WT flies (w 1118 , hereafter WT), Sting null mutant flies generated by CRISPR/ Cas9 editing (dSTING L76GfsTer11 [15], hereafter knockout [KO]), and WT flies whose immune response was ectopically activated by injection of 3′2′-cGAMP (hereafter referred to as immune primed [IP]) (Fig. 1A). As anticipated (13)(14)(15), modulation of cGAS-STING pathway activity correlated with fly survival after DCV infection. Sting KO flies were more sensitive to infection than WT flies (P < 0.0001, log-rank test; mean survival times 6.7 days for WT and 3.7 days for KO), and IP flies were less sensitive than WT flies (P < 0.0001, log-rank test; mean survival time of 8.8 days for IP flies; Fig. 1B). The protective effect of cGAMP injection was Sting-dependent as cGAMP injection in Sting KO flies did not affect survival after DCV infection (Fig. 1B, KO-cGAMP), thereby also excluding that a possible direct interaction between cGAMP and the virus inoculum caused the protective effect of cGAMP in WT flies. Likewise, cGAMP injection extended fly lifespan after CrPV infection, although Sting inactivation did not affect survival (P < 0.001, log-rank test; mean survival times of 6.8, 7.2, and 8.7 days for WT, KO, and IP, respectively; Fig. 1C).
We allowed virus populations to evolve during 10 fly generations (passages), with three independent lineages per host condition (Fig. 1A). The DCV parental stock was a host-adapted strain generated by serial dilution of the virus evolved in WT flies for 10 passages from our previous study (19). The CrPV parental stock was generated from an infectious molecular clone (20). Virus titers were measured after each passage to standardize the inoculum for subsequent passages and to minimize population bottle necks and differences in the rate of genetic drift. Viral titers of CrPV and DCV after each passage generally showed small differences between fly conditions across most passages (CrPV: Fig. 1D; DCV: Fig. S1A), likely due to virus titers reaching a plateau at this time point (3 days post-infection). The exception was the first passage of CrPV evolution, where the viral titers in IP flies were significantly lower than in WT and KO flies (61-fold and 71-fold, respectively; P-adj <0.005, estimated marginal means for pairwise compari son; Fig. 1D). In fact, viral titers were barely high enough to inoculate the next generation with the desired 1,000 TCID 50 inoculum, and we therefore decided to increase the CrPV inoculum to 5,000 TCID 50 from passage three onwards for all lineages.
## Increasing DCV synonymous nucleotide diversity
We studied molecular evolution during the experiment by RNA-sequencing at passages 1, 2, 3, 5, and 10 and computed viral nucleotide diversity across passages. We obtained a mean read depth across the genomes of 47,501 and 29,539 for the DCV and CrPV samples, respectively, which was consistent across samples (Fig. S2). For DCV populations evolved in WT flies, we incorporated the diversity values from our previous study (19). As expected, serial dilution used to prepare the parental virus (P0) reduced nucleotide diversity, consistent with population bottlenecking (Fig. 2A). By weighing nucleotide diversity per site according to the fraction of possible synonymous or non-synonymous substitutions, we found that the serial dilution specifically reduced non-synonymous nucleotide diversity (Fig. S1C). Afterward, we observed a significant increase in nucleo tide diversity as passages progressed across all DCV populations, regardless of their host condition (Fig. 2A, P < 0.0001, linear mixed model, Table S1). Increasing nucleotide diversity was predominantly explained by the accumulation of synonymous single-nucleotide variants (SNVs) (Fig. S1B through D). Only a single SNV in the 3′ untranslated region accumulated in all lineages, but this SNV preexisted in the parental virus. However, as we observed no indications for adaptive DCV evolution during the experiment, such as the conditionspecific fixation of SNVs or the accumulation of nucleotide diversity, we focused our subsequent analyses on the evolved CrPV popula tions.
## Transient increase and subsequent collapse of CrPV nucleotide diversity
In the CrPV parental stock, we identified a total of 244 SNVs with respect to the consen sus sequence of its 9.2 kb genome, of which four occurred at frequencies above 10% (Fig. S3A). As the parental stock was produced from an infectious clone, these SNVs must have accumulated during in vitro transcription or the 48 hours of replication in Drosophila S2 cells used to prepare the stock. As observed for DCV, nucleotide diversity significantly increased for CrPV with passage number (P < 0.001, estimate = 0.05, linear mixed model, Table S2). However, there was a significant decrease in diversity between passages 5 and 10 for populations evolved in WT and Sting KO flies (P = 0.001, estimate = -0.39, linear mixed model). A significant interaction was observed between the IP background and passage 10 (P = 0.01, estimate = 0.25, linear mixed model), indicating a sustained increase in nucleotide diversity of viral populations evolved in IP flies at later passages, in contrast with the sharp decline observed in WT and KO flies (Fig. 2B; Fig. S3B). At the same time, we observed increasing numbers of SNVs from passage 1 onwards in the populations evolved in WT and IP flies (Fig. 2C) with a mean of 244 and 199 SNVs at passage 1, respectively, which increased to a mean of 291 and 286 at passage 10. In contrast, for CrPV populations evolved in Sting KO flies, the mean number of SNVs remained roughly at the same level as the parental stock. The vast majority of SNVs occurred at low frequency (median: 0.07%, Fig. 2D), whereas only between 9 and 28 SNVs occurred at a frequency above 1%, and between 1 and 5 SNVs at a frequency above 10%, across all host conditions, passages, and replicates (Fig. 2D).
## Specific accumulation of nucleotide diversity in the CrPV 2B and RdRp genes
We calculated the nucleotide diversity for each viral gene and each condition (Fig. 3A; Fig. S4) and found that the mean nucleotide diversity varied significantly between viral genes (P-adj. < 0.001, two-way mixed ANOVA), but not between host conditions (P-adj. = 0.202). Post hoc pairwise comparisons (Materials and Methods, Table S3) revealed that the 3D gene encoding the RNA-dependent polymerase (RdRp) and the 2B gene showed Full-Length Text significantly increased diversity compared to seven and ten other genes, respectively (Table S3). In addition, 2B was the only gene that showed higher diversity in the evolved populations across all three host conditions than in the parental stock (Fig. S4). Moreover, the diversity in 2B was significantly higher compared to the remaining coding region (Fig. 3B, P-adj. = 0.0004, estimated marginal means with Tukey's adjustment), indicating that the dynamics in 2B dominate the diversity in the whole genome. Conversely, in evolved DCV populations, we observed the opposite trend: the diversity in 2B was significantly lower than in the remaining coding region (Fig. 3B, P-adj. = 0.0024, estimated mar ginal means with Tukey's adjustment). Furthermore, a direct comparison of 2B diversity between DCV and CrPV populations revealed significantly lower diversity in DCV (Fig. 3B, P-adj. < 0.0001, estimated marginal means with Tukey's adjustment), suggesting the importance of the 2B gene in CrPV adaptation to the Drosophila host and stringent selection on 2B in DCV.
## Independent fixation of CrPV 2B D29N in six lineages
A complex pattern of SNV retention and emergence was observed along the CrPV genome across different lineages and passages (Fig. 3C). Many SNVs present in the parental stock were retained in most lineages (e.g., C775A, T828C, C1602A, G2760A, A3264T, C4786A), while others showed temporal variations. For instance, C8749A and Full-Length Text T8856A persisted until passage 5 but were absent in passage 10, and G4612A was retained only in populations from IP flies. We observed the de novo appearance of A3892C, a silent mutation of Arg154 in the 3C protease (AGA to CGA), in all host conditions across most passages and replicates at frequencies of 2.0%-14.9%. Another interesting de novo SNV was G1363A, resulting in an aspartic acid to asparagine substitution at position 29 of the 2B protein (D29N). This substitution emerged at passage 2 in WT flies for the first time at a frequency of 0.03% and in Sting KO flies at passage 3 at a frequency of 0.05%. In all replicate lineages in WT and Sting KO flies, this mutation was fixed at passage 10 at frequencies above 82%. In contrast, in viral populations from IP flies, G1363A was only detectable in one replicate lineage at a frequency of 0.2% at passage 3, but rather than increasing in frequency, it became extinct in the following passages. In one of the lineages from Sting KO flies (KO-a), we detected an additional synonymous mutation, G1323A, in close proximity to and at similar frequencies as the G1363A substitution. Local haplotype analysis confirmed the co-occurrence of G1363A and G1323A in a high proportion of reads that carry at least one of those mutations at passages 5 and 10 (2101/2501 and 4792/4802, respectively), suggesting that G1323A is a hitchhiking mutation.
## D29N causes a marked increase in viral virulence
The recurrent fixation of D29N in multiple independent lineages suggests that it confers an evolutionary benefit to the virus. To experimentally study the phenotype of D29N, we introduced the mutation into the infectious clone and compared the growth kinetics of WT CrPV and the D29N CrPV mutant in WT, Sting KO, and IP flies. The D29N mutation significantly increased viral RNA levels (P < 0.0001, linear mixedeffects model, Table S4), with an estimated 5-to 10-fold increase in all three host conditions compared to the WT virus (Fig. 4A). Stimulation of the cGAS-STING pathway by co-injection of 3′2′-cGAMP significantly reduced CrPV RNA levels by a magnitude of 26.7 for both variants in WT flies (P = 0.0014, linear mixedeffects model, Table S4), indicating that D29N did not affect sensitivity to cGAS/STING-mediated immunity. In line with the increased viral replication rate, survival of flies infected with the D29N mutant virus was significantly reduced compared to WT CrPV-infected flies in all host backgrounds (P < 0.0001, log-rank test; mean survival times of 8.2, 8.2, and 9.0 days for WT, KO, and IP flies infected with WT CrPV and 4.8, 5.1, and 5.7 for WT, KO, and IP infected with the CrPV D29N), although 3′2′-cGAMP co-injection still extended fly survival (Fig. 4B).
We approximated the selection coefficient of the D29N mutant relative to WT CrPV using two independent approaches: tracking changes in nucleotide frequencies during the experimental evolution study (Fig. 3C) and comparing the replication rates of WT and D29N mutant CrPV that we estimated assuming exponential growth (Fig. 4A, Materials and Methods). Both methods yielded similar results, with mean estimates of 0.287 and 0.267 for the frequency and RNA-based estimates, respectively, indicating increased fitness of the D29N mutant virus (Table S5). Overall, these data indicate that D29N causes an increase in viral replication and virulence.
## 2B is a predicted transmembrane protein that localizes to endomembranes
No experimental data exist about the function of CrPV 2B or any of its homologs, including DCV 2B that we identified using PSI-BLAST (22) or HHPred (23). Picornaviruses such as coxsackievirus and poliovirus (Enterovirus genus) also encode a protein named 2B that, as in CrPV, lies directly upstream of the 2C RNA helicase. To find evidence for homology between CrPV 2B and enterovirus 2B, we ran HHPred in the pairwise profile alignment mode, in which we compared CrPV 2B with the complete coxsackievirus A10 (CVA10) and poliovirus polyproteins. CrPV 2B specifically aligned with the poliovirus 2B protein and the CVA10 2B protein (Fig. S5A andB), albeit with high E-values (E = 0.55 and E = 0.24, respectively). The region that is most similar between the 2B proteins is a stretch of 49 amino acids that corresponds to the two predicted transmembrane (TM) helices in CrPV 2B (Fig. S5A) (24) and two experimentally determined TM helices in the coxsackievirus and poliovirus 2B proteins (25,26). The D29 residue lies upstream of these helices and appears well conserved among homologs that could be detected with BLASTp (Fig. 4C). AlphaFold structure prediction (27), whose modeling accuracy improved by the addition of lipids, as well as HHpred (Fig. 4C; Fig. S5C), suggested that all three 2B proteins are predominantly α-helical. They likely adopt a helix-turn-helix motif that traverses organelle membranes and would expose D29 toward the cytoplasm (26). Based on its tentative homology with the enterovirus 2B proteins, we hypothesized that the CrPV 2B protein localizes in the endomembrane system of the host cell (25,26). To experimentally validate this, we tagged the WT and D29N 2B proteins with eGFP at either the N or C terminus and studied their subcellular localization, using Drosophila proteins known to localize in the ER (Calnexin 99a), cis-Golgi (GM130), and lysosomes (Rab7) as markers (Fig. 5). We observed that 2B localized in the ER (red), but also showed punctate patterns corresponding to the cis-Golgi (Fig. 5A, magenta), lysosomes (Fig. 5B), and other membranous organelles, whose identity remains to be established. The D29N mutation did not seem to drastically affect the subcellular localization of the 2B protein.
Considering that human STING is embedded into the ER membrane, we analyzed whether Drosophila Sting would interact with CrPV 2B. We epitope-tagged Sting and co-expressed the fusion construct with either WT or D29N 2B-GFP fusion proteins or an eGFP control in S2 cells. The punctate pattern of 2B was highly similar to the pattern produced by the Sting protein (Fig. 6A), which was validated by colocalization analysis (median Pearson's correlation coefficient R = 0.48; Fig. 6B). Strikingly, an even stronger correlation was observed for the 2B D29N mutant (R = 0.76; Fig. 6B). The interaction between 2B and Sting was further validated by a co-immunoprecipitation assay. We found that the Sting protein was recovered in immunoprecipitations of both the WT and D29N 2B-GFP proteins, but did not coprecipitate with the eGFP control alone (Fig. 6C). Together, these data suggest that the 2B protein localizes in cellular endomembranes, where it interacts with Sting. However, given that the D29N mutation was also selected in flies lacking Sting, it is unlikely to represent an adaptation to optimize 2B interactions with Sting.
## DISCUSSION
In this study, we have explored the impact of the cGAS-STING pathway on virus evolution using D. melanogaster infected with DCV or CrPV as a model. We allowed the viruses to adapt to flies with a modulated cGAS-STING response, hypothesizing that variations in pathway activity would affect virus evolution at the genomic and phenotypic level. We found no effect of the STING pathway on DCV evolution, but instead noted a gradual increase in synonymous nucleotide diversity in all host conditions. Likewise, for CrPV, we found no evidence for adaptation that was specific to different cGAS-STING regimes. In contrast, we observed that nucleotide diversity accumulated especially in the CrPV 2B gene in all conditions, while DCV exhibited reduced diversity in 2B compared to the remaining coding regions, consistent with findings from our previous study (19). Specifically, for CrPV, we detected an SNV resulting in a D29N amino acid substitution in the 2B protein that was selected in all viral lineages from WT and Sting KO flies, while its frequency did not increase in viral populations from IP flies, despite being detected in one lineage at an early passage. CrPV mutants harboring D29N replicated faster and were more virulent than the WT virus in all host backgrounds. The appearance of parallel adaptive mutations driven by cross-species transmission has been previously observed in experimental evolution studies of DCV adapting to different Drosophila species (29) and vesicular stomatitis virus adapting to either human or dog cells (30). As Drosophila is not the natural host of CrPV, our data suggest that the 2B mutation likewise emerged to facilitate adaptation to its nonnative host.
We found that DCV nucleotide diversity continued to increase with passage number, suggesting that the virus populations had not reached equilibrium yet. Equilibrium is reached when the gain of genetic diversity is equal to the loss caused by purifying selection and random drift (31). This state seems difficult to attain in virus populations, as nucleotide diversity not only increases during previous relatively short-term evolutionary experiments (32)(33)(34) but also in long-term experiments, for example, HIV-1 experimental evolution over more than 3 years, during which mutations continue to accumulate (35). Likewise, the number of polymorphic sites also continued to accumulate in anellovirus over a period of 30 years in chronically infected patients (36). Therefore, the time scale necessary for a virus to achieve equilibrium is likely too long to be achieved in our experiment.
In 6 of the 18 virus lineages in our study, nucleotide diversity decreased from passage 5 to passage 10. A common factor of those virus populations was the fixation of the adaptive D29N mutation in the CrPV 2B gene. We hypothesize that the diversity decrease was caused by a selective sweep (37), as previously observed during the fixation of drug resistance variants in HIV-1 (38). The observation that the D29N mutant was not selected in IP flies despite its higher replication rates across all host conditions might be explained by altered evolutionary dynamics due to a reduction of the viral effective population size upon the stimulation of the cGAS-STING pathway. Under these conditions, evolution is expected to be dominated by random drift rather than deterministic selection occurring under large population sizes (39,40). While we attempted to estimate the variance effective population size from temporal changes in synonymous allele frequencies, our experiment involved selection effects and population fluctuations that violate key assumptions of standard effective population size estimation methods (41), making such estimates unreliable in our system (41)(42)(43). We did observe, nonetheless, a significantly reduced viral RNA load in IP flies, suggesting a smaller total population size, which may contribute to the observed evolutionary patterns.
It is currently unclear why the 2B mutation facilitated CrPV host adaptation. The function of the 2B protein of members of the Dicistroviridae family has not been studied. However, their corresponding relative positions in the viral genome, putative homol ogy based on sequenceprofile comparisons, and the location of (predicted) transmem brane regions suggest that CrPV 2B is homologous to the 2B protein of enteroviruses, although direct evidence is lacking. Enterovirus 2B is an ion channel-forming protein, typically 50-120 amino acids in length. It contains two hydrophobic regions that adopt a helix-turn-helix conformation, enabling insertion into cellular membranes (reviewed in reference 44). The protein also has at least one amphipathic α-helical structure that can oligomerize to form transmembrane hydrophilic pores (45,46), possibly as a tetrameric complex (47)(48)(49)(50). Enterovirus 2B has viroporin activity, which is associated with increased membrane permeabilization, disrupted calcium homeostasis, and induction of apoptosis and autophagy (51). In addition, enterovirus 2B is associated with viral replication organelles, thought to be derived from modified ER and Golgi apparatus membranes (52). We indeed observed localization of CrPV 2B in cellular endomembranes, but whether the protein is also a viroporin and how mutations in such a protein would facilitate host adaptation remains to be studied.
The 2B D29N mutation was selected in both WT and Sting KO flies, and the mutant conferred a growth advantage to CrPV also in Sting KO flies, indicating that 2B D29Nmediated host adaptation is independent of the cGAS-STING pathway. It was therefore rather unexpected to find that CrPV 2B colocalizes with Sting, and that the colocalization was even stronger for the D29N 2B mutant. While this conundrum awaits being solved, it is interesting to note that enterovirus 2B has been shown to affect STING localization in mammalian cells. Specifically, the 2B protein of rhinovirus was found to reduce Ca 2+ levels in the ER, triggering the relocation of STING to the replication organelles (53), where it acts as a proviral factor to promote viral replication (28).
Overall, our work provides new insights into viral adaptation to a new host and the characteristics of the 2B protein of dicistroviruses, a family comprising important model viruses to dissect insect virus-host interactions.
## MATERIALS AND METHODS
## Fly strains and husbandry
Flies were maintained at 25°C in standard fly food. Eggs were bleached and treated with tetracycline as described before (54), and the absence of Nora virus, DCV, Drosophila X virus, CrPV, and Wolbachia was verified as described earlier (19). Flies containing the dSTING L76GfsTer11 null allele (15) (referred to as Sting KO in this study) were a kind gift from Jean-Luc Imler (CNRS-Université de Strasbourg). w 1118 flies were used as WT, as the Sting KO flies had been isogenized in this genetic background according to the procedure described for cGLR1 and cGLR2 in reference 11. IP flies were generated by the intrathoracic injection of WT flies with 69 nL of a 1 µg/µL solution of c[A(2′,5′)pG(3′,5′)p] (3′2′-cGAMP) (BioLog). Two-to fivedayold female flies, likely mated as they were obtained from a vial containing unsorted flies, were used for the experimental evolution study, growth curves, and survival assays.
## Plasmids
The CrPV 2B coding sequence was amplified by PCR from cDNA of infected w 1118 flies and cloned into fusion vectors expressing eGFP under control of the Actin promoter using In-Fusion cloning (TAKARA). For the construction of pAct-2B-eGFP plasmid, the 2B sequence was amplified using forward primer 2B-eGFP-For (5′-GGTACCTACTAGTCCccacc ATGCGCCGAGATGAGAAAATTTCAACCTT-3′), containing the Kozak sequence and the start codon, and reverse primer 2B-eGFP-Rev (5′-GCCCTTGCTCACCATggatccacctgatccgccTTGG GTTGTTGTCTGCAAAATTTTGT-3′) encoding a linker sequence (GGSGGS). The PCR product was cloned into the pAct-eGFP-C backbone linearized with primers vectorFor (5′-AT GGTGAGCAAGGGCGAG-3′) and vectorRev (5′-GGACTAGTAGGTACCCCGATCC-3′). For the construction of the pAct-eGFP-2B plasmid, the 2B sequence was amplified using forward primer eGFP-2B-For (5′-GACGAGCTGTACAAGGGCGGATCAGGTGGATCCCGCCGAGATGAGA AAATTTCAACCTT-3′) encoding the linker peptide and reverse primer eGFP-2B-Rev (5′ -AGCTCAGGCCTTAGAttaTTGGGTTGTTGTCTGCAAAATTTTGT-3′) containing a stop codon. The PCR product was cloned into the pAct-eGFP-N backbone linearized with primers vectorFor (5′-TCTAAGGCCTGAGCTCGCT-3′) and vectorRev (5′-CTTGTACAGCTCGTCCATGC C-3′). The mutant fusion constructs pAct-2B(D29N)-eGFP and pAct-eGFP-2B(D29N) were generated by site-directed mutagenesis by amplifying either pAct-2B-eGFP or pAct-eGFP-2B with primers For (5′-GGTTTCTTTAATGATCTCAAAGGAGCAAAAGG-3′) and Rev (5′ -TAAAGAAACCTTGGGTGTAAATTCTGTTG-3′). The linearized PCR product was recombined using In-Fusion cloning (TAKARA).
The pCrPV(D29N) infectious clone was generated by amplifying the 2B(D29N) sequence from the plasmid pAct-2B(D29N)-eGFP with primers For (5′-GAGAAAATTTCAA CCTTGATTAAGAAG-3′) and Rev (5′-AATCAAGGCGGCACGATAC-3′) and the backbone from pCrPV-3 (20) with primers For (5′-CGTGCCGCCTTGATTGTAAT-3′) and Rev (5′-GTTGAAATTT TCTCATCTCGGCG-3′) using In-Fusion cloning (TAKARA).
The Sting coding sequence was amplified by PCR from cDNA of w 1118 flies and cloned into pAc5.1-V5-His-A (Invitrogen) using In-Fusion cloning (TAKARA) to produce Sting-V5. The forward primer (5′-GATCGGGGTACCTACccaccATGGCAATCGCTAGCAACGT-3 ′) contained the Kozak sequence, and the reverse primer (5′-AGGGATAGGCTTACCgga tccacctgatccgccGTTGGAAATTTCGTCAATAGTTTTGGTTTTGTTT-3′) contained a sequence encoding a linker peptide (GGSGGS).
## In vitro transcription and transfection of the CrPV infectious clone
The plasmid encoding the CrPV molecular clone (pCrPV-3 [20]) was a kind gift from Eric Jan (University of British Columbia). In vitro transcription and transfection were performed as previously described (20). Briefly, the plasmids were amplified in Stellarcompetent E. coli, purified, linearized with Ecl136II (Invitrogen), and used for in vitro transcription at 30°C for 3 h using the RiboMAX kit (Promega). The RNA was treated with DNase RG1 for 15 min at 37°C and purified using RNeasy columns (Qiagen). The viral RNA was transfected into Drosophila S2 cells (Invitrogen) in six-well plates containing 2 mL of Schneider's medium. After 48 h, the supernatant was collected, centrifuged at 300 × g for 5 min, aliquoted, titered, and stored at -80°C.
## Cells and virus stocks
The parental DCV and CrPV stocks were produced in Drosophila S2 cells (Invitrogen), which were maintained in Schneider's medium supplemented with 10% fetal calf serum (Sigma) and 50 U/mL penicillin and 50 µg/mL streptomycin (Gibco) at 27°C. DCV previously passaged in WT flies for 10 generations (19) was bottlenecked by serial dilution in a 24-well plate, selecting the highest dilution that still showed cytopathic effects to inoculate a T25 flask of confluent S2 cells containing 7 mL of medium. After 72 h, the supernatant was collected, centrifuged at 300 × g for 5 min, aliquoted, and stored at -80°C. Titers were determined by end-point dilution in 96-well plates, as described before (54), and expressed as median tissue culture infectious dose (TCID 50 ), calculated using the Reed-Muench method.
## Experimental evolution
Thirty female flies of each host condition were intrathoracically inoculated with 1,000 TCID 50 of DCV or CrPV in 69 nL of PBS, pH 7.3, using a Nanoject II microinjector (Drum mond Scientific). For IP flies, 3′2′-cGAMP was co-injected with the virus inoculum into WT flies. Three lineages per host condition were used as independent replicates. For the first passage, flies were inoculated with the respective parental stock, and from P1 onwards, all 18 lineages were kept independently (nine lineages per virus). After each passage, pools of 10 flies were snap-frozen and homogenized two times for 10 s in 220 µL of PBS, using 1 mm silica beads in a Precellys homogenizer, centrifuged at 16,000 × g for 10 min to discard fly debris, and the supernatant was transferred to a fresh tube, aliquoted, stored at -80°C, and directly titrated. The next generation of flies was then inoculated through intrathoracic inoculation. In the case of CrPV, the virus inoculum was increased to 5,000 TCID 50 from passage three onwards, as the titers after the first and second passage were barely high enough to reach the desired inoculum for the next passage in the IP flies. For RNA extraction and preparation of a next-generation sequencing (NGS) library of the parental stock, 1 mL of TRI Reagent (Sigma) was added to 100 µL of virus stock, while for the viral lineages, pools of five flies were homogenized in 1 mL of TRI Reagent (Sigma) using 1 mm silica beads in a Precellys homogenizer. RNA was isolated according to the manufacturer's instructions.
## Virus infections and survival assays
To study virus growth kinetics, 100 TCID 50 of DCV or 1,000 TCID 50 of WT or D29N CrPV were injected into 2-to 5-day-old WT, Sting KO, and IP flies. For the IP condition, 3′2′-cGAMP was co-injected with the virus into WT flies. Three pools of five flies per condition were harvested at the indicated time points and processed for RT-qPCR. To study virulence, the same dose and procedure were applied to 50 flies per condi tion. Survival was monitored daily and analyzed using the Kaplan-Meier estimator. The difference between the curves was compared using the log-rank test in GraphPad Prism 10.
## RT-qPCR
RT-qPCR was performed as previously described (19), using primers CrPVFor (5′-ACGA GGAAGCAACTCAAGG-3′) and CrPVRev (5′-GAGCCCGCTGAGATGTAAAG-3′) for CrPV RNA quantification.
## NGS library preparation
Total RNA from passages 1, 2, 3, 5, and 10 was extracted using TRi Reagent (Sigma) and reverse transcribed with Superscript IV (Thermo Fisher) and oligo(dT) primers. Viral cDNA was amplified in four overlapping amplicons of about 2.2 kb. The primers used for CrPV were as follows:
For1, 5′-CTCCCCGTGAGAAACCTTGTTT-3′; Rev1, 5′-GTGTTTGTAAGCGTCGGGTTTG-3′; For2, 5′-GATCCCGGACCGAGACATTG-3′; Rev2, 5′-TTACCGCCTGACCAACCTTG-3′; For3, 5′-ACGGATATGCTTGCCCCTTAAC-3′; Rev3, 5′-TGGGGTGAAACATAGGGAATTCTC-3′; For4, 5′-CTTCGCGCCACACTTGTTG-3′; Rev4, 5′-AAAACCTGTTAGCCCCGATG-3′. Primers to amplify DCV are provided in reference 19. Amplicons were pooled, purified with the NucleoSpin kit (MACHEREY-NAGEL), sheared in a Bioruptor Pico (Diagenode) in 1.5 mL Bioruptor tubes (Diagenode) to an average size of 200 bp following 20 cycles of 30 s of sonication and 30 s of cooling. The library was prepared using the NEB Next Ultra II DNA Library Prep kit (NEB) according to the manufacturer's instructions and multi plexed with barcodes (E7335, E7500, E7710, E7730, NEB). Library size was determined using the DNA 1000 Bioanalyzer kit (Agilent), and the concentration was measured with Qubit dsDNA High Sensitivity assay kit (Thermo Fisher). Libraries were sequenced on an Illumina NextSeq200 as paired-end 55 bp long reads and demultiplexed with bcl2fastq (RRID:SCR_015058).
## NGS data processing
Raw Illumina read data were processed using the bioinformatics pipeline V-pipe 3.0 (55), which was integrated into a custom Snakemake workflow. The CrPV and DCV parental stock data sets were aligned to their respective reference sequences (CrPV, NC_003924.1 and DCV EB, NC_001834.1), using the Burrows-Wheeler Alignment Tool BWA-MEM (56). Consensus sequences were generated for both parental stock data sets using SmallGe nomeUtilities (57). These consensus sequences were used as references for aligning raw reads from the samples at passages 1, 2, 3, 5, and 10. Sequencing errors were corrected and mutations called relative to the parental stock consensus sequence using the tool VILOCA (version 1.1.0). VILOCA achieves this by clustering sequencing reads into local haplotypes approximately as long as the read length through a finite Dirichlet process mixture model, which integrates read quality scores to reliably distinguish true variants from sequencing noise (58). Non-synonymous mutations were annotated using an adapted version of vcf_annotator (https://github.com/rpetit3/vcf-annotator). Population nucleotide diversity was calculated using SNPGenie (59), which estimates average pairwise nucleotide differences directly from the observed allele frequencies at each position after error correction. SNPGenie computes overall diversity and diversity per synonymous (piS) and non-synonymous (piN) sites, weighting each site by the fraction of possible substitutions that are synonymous or nonsynonymous. The minfreq parameter in SNPGenie was set to zero. The complete computational workflow and notebooks for generating figures are available on GitHub (https://github.com/cbg-ethz/ DCV-CrPV-cGAS-STING-pathway-data-analysis).
## Mean nucleotide diversity
A linear mixedeffects model was fitted to analyze the log 10 -transformed DCV mean nucleotide diversity (lme, nlme package in R [60]). First, a model was constructed including passage, condition, and their interaction as fixed effects, along with evolu tionary lineage replicate as a random effect to account for non-independence within replicates. To allow for heteroscedasticity between passages, a variance structure (VarIdent) was included, and since slight autocorrelation across passages within lineages was detected, an autoregressive moving average correlation structure (corARMA, q = 1) was incorporated. This model was then compared to a simpler model without the interaction term using ANOVA. Since ANOVA indicated no significant difference between the full and simplified models, the simpler model was used due to its fewer degrees of freedom. Residuals versus fitted values plots, as well as Levene's tests for homogeneity of variances across passage, genotype, and their interaction, did not indicate significant heteroscedasticity. Q-Q plots of the residuals showed approximate normality.
A similar mixedeffects model was fitted to analyze the log 10 -transformed CrPV mean nucleotide diversity. To account for the piecewise linear structure of the data (initially increasing linearly, then decreasing), passage number was included as fixed effects along with an indicator variable for passage 10. This approach allowed modeling a potential change in the effect of passage after passage 5. An interaction term between condition and this indicator variable was also included to test for conditionspecific changes after passage 10. Additionally, to account for within-lineage autocorrelation across passages, an AR(1) autocorrelation structure was incorporated in the model residuals using the corAR1 function from the nlme package. Model assumptions were confirmed using check_model from the performance package in R (61).
CrPV gene diversity was analyzed using ANOVA (R rstatix, anova_test [62]) with the dependent variable being the mean diversity over the passages for each replicate line, the between-subject factor being condition, and the within-subject factor being gene. Normality was confirmed using qq-plots, homogeneity of variances was confirmed using Levene's test (R, levene_test), and homogeneity of covariances of the between-fac tors was tested using Box's M-Test (R, box_m). As post hoc analysis, pairwise compari sons between all gene levels were performed using estimated marginal means with Bonferroni adjustment for multiple testing (emmeans package in R [63]).
## Nucleotide diversity in 2B and the remaining coding region in DCV and CrPV populations
Nucleotide diversity in the 2B gene was compared to the remaining coding region by computing the mean diversity over the passages and then using a linear mixedeffects model (lme4 package in R [64]), with the dependent variable being the log-transformed diversity for each evolutionary lineage. The fixed effects in the model were virus (DCV or CrPV), condition (included as a control variable), and genome region, while the evolutionary lineage was included as a random effect. A sensitivity analysis confirmed that removing the condition from the model did not alter the significance or effect sizes of virus or genome region, supporting its exclusion. The normality of residuals was assessed using Q-Q plots and the Shapiro-Wilk test (shapiro.test). The homogeneity of variance was evaluated using plots of residuals versus fitted values and Levene's test (car package in R). The independence of residuals was checked through plots of residuals and autocorrelation function (ACF) plots. Post hoc pairwise comparisons were performed using estimated marginal means with Tukey's adjustment for multiple comparisons between virus and genome region (emmeans package in R).
## Statistical analysis for titers and relative RNA levels
A mixed ANOVA with a within-subject factor, passage, and an independent between-sub ject factor, host condition, was used to assess variation among log-transformed viral titers between conditions and passages. Subsequently, a pairwise comparison of host conditions at each passage level was performed using estimated marginal means (R package emmeans, v1.10.7 [63]), adjusting P-values with Bonferroni correction. ANOVA assumptions were confirmed using the R package performance (61) with the function check_homogeneity (Levene's test, P > 0.05), check_sphericity (P > 0.05), and check_nor mality, where the points remained within the confidence interval, suggesting that the residuals are approximately normally distributed. Residual diagnostics, including Durbin-Watson tests (DCV: P = 0.998, CrPV: P = 0.07) and ACF plots of residuals, indicated no significant autocorrelation, supporting the use of a model without an autocorrelation structure. Nevertheless, given the temporal structure of the data, a model incorporating an AR(1) autocorrelation structure was also fitted as a sensitivity analysis. This alternative model exhibited some collinearity and some deviations from normality in residuals. However, the significance of P-values remained unchanged, suggesting that the main model inference is robust to potential autocorrelation. This analysis was performed for each virus separately.
To compare the relative RNA levels between WT and D29N CrPV, a linear mixed effects model was used on the log-transformed values (lme, nlme package in R [60]). A model was fitted that included host condition, time post-infection, and variant (D29N or WT) as additive effects. To account for measurements from the same lineage over time, a random intercept was included for time post-infection within each lineage. Additionally, to model the correlation of the measurements over time post-infection within each lineage, a firstorder autoregressive (AR(1)) correlation structure was incorporated. Model assumptions were confirmed using check_model from the performance package in R (61), revealing only mild deviations from the model assumptions. Residual autocorrela tion was assessed using the ACF, confirming that AR(1) correlation structure successfully eliminated temporal dependencies.
## Approximation of the selection coefficient of the D29N mutation
Two complementary approaches were used to estimate the fitness effect of the D29N mutation: replication rate analysis and frequency data analysis. The replication rates (r) of the WT and mutant (D29N) were computed based on the fold change (x) representing RNA levels at 96 hours post-infection (hpi) relative to 0 hpi, assuming an exponential growth model:
where t is time in hours and assuming a 10-hour generation time (65). Using the calculated replication rates, the selection coefficient (s RN A ) was determined by compar ing the growth advantage of D29N relative to P0 over one generation (66):
The selection coefficient (s freq ) was also estimated using changes in D29N frequencies over time (67):
with n being the number of total generations. For this analysis, only evolutionary lineages with multiple time points in which D29N occurred were included, as singletime-point data do not allow for the determination of frequency change over time. Positive selection coefficients indicate that the mutant grows faster than the WT, while negative coefficients indicate slower growth.
$$x = e t 10 r$$
$$S RNA = r mutant -r wildtype r wildtype$$
$$S freq = ln f mutant 1 -f mutant final -ln f mutant 1 -f mutant initial n$$
## Protein structure prediction
AlphaFold 3 (27) was used to predict the CrPV 2B structure, which was simulated with the addition of oleic acid, myristic acid, palmitic acid, and Ca 2+ as ligands and ions.
## 2B sequence logo and homology detection
The CrPV 2B protein sequence was used in a homology search using BLASTp (68) against the non-redundant databases (NRDB) with an E value cutoff of 0.05, resulting in 131 homologs that were mainly from Riboviria. A multiple alignment with all the obtained proteins was created on the NCBI site with Cobalt (69), and the first 50 amino acids of the 2B protein in the alignment were used to construct a sequence logo (70). Homology of the CrPV 2B protein with other proteins was established with PSI-BLAST ( 22) using default parameters (E < 0.005 for inclusion in the next iteration) against the NRDB database and iterating until convergence. To test the homology of CrPV 2B with poliovirus (Q1PHW2 in UniProt) and CVA10 (A0A6M2Z865 in UniProt) polyproteins, HHPred was used in the pairwise alignment mode using default parameters.
## Immunofluorescence assay
Drosophila S2 cells were seeded on 0.01% poly-D-lysine-coated coverslips in 24-well plates and incubated for 3 h at 27°C. Cells were then transfected with 0.5 µg of the indicated plasmids using 1 µL of X-tremeGENE HP (Roche), and the medium was refreshed after 24 h. For co-transfection of two plasmids, 0.25 µg of each plasmid was used instead. All following steps were performed at room temperature and in the dark when possible. At 48 h after transfection, cells were fixed in 4% paraformaldehyde for 15 min, permeabilized with 0.1% Triton-X100 in PBS for 15 min, and washed with PBS containing 0.1% Tween-20. Cells were blocked in 2% normal goat serum (NGS) for 30 min and incubated for 45 min with the primary antibody diluted in 2% normal goat serum in PBS. Calnexin 99A antibody (Developmental Studies Hybridoma Bank, Cnx99A 6-2-1; RRID:AB_2722011) was diluted 1:10, GM130 antibody (Abcam, ab30637) was diluted 1:250, Rab7 antibody (Developmental Studies Hybridoma Bank, RRID:AB_2722471) was diluted 1:10, and V5 antibody (Invitrogen, 46-0705) was diluted 1:200. After washing, cells were stained with secondary antibody goat anti-rabbit conjugated with Alexa Fluor 594 (Thermo Fisher, A11012) and goat anti-mouse with Alexa Fluor 647 (Thermo Fisher, A21235), used at a 1:500 dilution. Nuclei were stained with 1 µg/mL DAPI (Sigma) for 10 min, and coverslips were mounted on slides in ProLong Glass (Thermo Fisher). Cells were imaged on a Zeiss LSM900 Airyscan and 63× oil objective using the 405, 488, 561, and 633 excitation lasers. Images were Airyscan processed before analysis with Fiji software (71). Brightness and contrast were adjusted for visualization purposes. Raw images were used for colocalization analysis using the JACoP plug-in in Fiji (72). Images from individual cells were cropped, and the Pearson's correlation coefficient between the eGFP and V5 signal was determined by applying the Costes automatic threshold. The coefficients were compared using a one-way ANOVA with Tukey's multiple comparison test. Shapiro-Wilk normality test confirmed the normality of residuals and Levene's test confirmed equal variances.
## Immunoprecipitation
Immunoprecipitation was performed following the instructions provided with the ChromoTek GFP-Trap Magnetic Agarose Kit (Proteintech). Briefly, S2R+ cells were co-transfected with plasmids encoding eGFP, 2B-eGFP, or 2B(D29N)-eGFP in combination with a plasmid encoding Sting-V5. Cells were rinsed with PBS and lysed in ice-cold lysis buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5 mM EDTA, 0.5% NP-40, and 1 mM PMSF). The cell lysates were centrifuged at 17,000 × g for 15 min at 4°C. The supernatants were diluted with lysis buffer, equilibrated GFP-Trap beads were added, and samples were incubated with end-over-end rotation for 1 h at 4°C. The beads were then washed four times with washing buffer (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5 mM EDTA, 0.05% NP40, and 1 mM PMSF) and resuspended in Laemmli sample buffer. The samples were heated for 5 min at 95°C, resolved on 12% SDS-PAGE gels, and transferred to nitrocel lulose membranes by wet transfer. Membranes were stained with antibodies against GFP (rat, 1:1000, ChromoTek, 3h9-150), V5 tag (mouse, 1:2000, Invitrogen, 46-0705), or GAPDH (rabbit, 1:1000, Bioss, bs-8778R), followed by secondary antibodies IRDye 800CW goat anti-rat IgG (LICORBio, 926-32219), IRDye 800CW goat anti-mouse IgG (LICORBio, 926-32210), and IRDye 800CW goat anti-rabbit IgG (LICORBio, 926-32211), respectively. All secondary antibodies were used at a 1:2,000 dilution. The membranes were imaged using an Odyssey M Imaging System.
## References
1. Longdon, Brockhurst, Russell et al. (2014) "The evolution and genetics of virus host shifts" *PLoS Pathog*
2. Woolhouse, Haydon, Antia (2005) "Emerging pathogens: the epidemiology and evolution of species jumps" *Trends Ecol Evol (Amst)*
3. Marchi, Lässig, Walczak et al. (2021) "Antigenic waves of virusimmune coevolution" *Proc Natl Acad Sci*
4. Korobeinikov (2018) "Immune response and within-host viral evolution: immune response can accelerate evolution" *J Theor Biol*
5. Merani, Petrovic, James et al. (2011) "Effect of immune pressure on hepatitis C virus evolution: insights from a single-source outbreak" *Hepatology*
6. Neuner-Jehle, Zeeb, Thorball et al. (2024) "Using viral diversity to identify HIV-1 variants under HLA-dependent selection in a systematic viral genomewide screen" *PLoS Pathog*
7. Burmeister, Lenski, Meyer (2016) "Host coevolution alters the adaptive landscape of a virus" *Proc Biol Sci*
8. Seth, Sun, Chen (2006) "Antiviral innate immunity pathways" *Cell Res*
9. Cai, Imler (2021) "cGAS-STING: insight on the evolution of a primordial antiviral signaling cassette" *Fac Rev*
10. Ablasser, Chen (2019) "cGAS in action: expanding roles in immunity and inflammation" *Science*
11. Holleufer, Winther, Gad et al. (2021) "Two cGAS-like receptors induce antiviral immunity in Drosophila" *Nature*
12. Slavik, Morehouse, Ragucci et al. "2021. cGAS-like receptors sense RNA and control 3′2′-cGAMP signalling in Drosophila" *Nature*
13. Cai, Li, Slavik et al. (2023) "The virus-induced cyclic dinucleotide 2'3'c-di-GMP mediates STING-dependent antiviral immunity in Drosophila" *Immunity*
14. Goto, Okado, Martins et al. (2019) "The kinase IKKβ regulates a STING-and NF-κB-dependent antiviral response pathway in Drosophila" *Immunity*
15. Cai, Holleufer, Simonsen et al. (2020) "2'3'-cGAMP triggers a STING-and NF-κB-dependent broad antiviral response in Drosophila" *Sci Signal*
16. Reinganum, 'loughlin, Hogan (1970) "A nonoccluded virus of the field crickets Teleogryllus oceanicus and T. commodus (Orthoptera: Gryllidae)" *J Invertebr Pathol*
17. (2026) *Full-Length Text Journal of Virology*
18. Roberts, Anderson, Durr (2018) "Metagenomic analysis of Varroa-free Australian honey bees (Apis mellifera) shows a diverse Picornavirales virome" *J Gen Virol*
19. Casadevall, Pirofski (1999) "Host-pathogen interactions: redefining the basic concepts of virulence and pathogenicity" *Infect Immun*
20. Lezcano, Fuhrmann, Ramakrishnan et al. (2023) "Parallel evolution and enhanced virulence upon in vivo passage of an virus in Drosophila melanogaster" *Virus Evol*
21. Kerr, Wang, Keatings et al. (2015) "The 5' untranslated region of a novel infectious molecular clone of the dicistrovirus cricket paralysis virus modulates infection" *J Virol*
22. Valles, Chen, Firth et al. (2017) "ICTV virus taxonomy profile: dicistroviridae" *J Gen Virol*
23. Altschul, Madden, Schäffer et al. (1997) "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs" *Nucleic Acids Res*
24. Söding, Biegert, Lupas (2005) "The HHpred interactive server for protein homology detection and structure prediction" *Nucleic Acids Res*
25. Käll, Krogh, Sonnhammer (2007) "Advantages of combined transmembrane topology and signal peptide prediction-the Phobius web server" *Nucleic Acids Res*
26. De Jong, Wessels, Dijkman et al. (2003) "Determinants for membrane association and permeabilization of the coxsackievirus 2B protein and the identification of the Golgi complex as the target organelle" *J Biol Chem*
27. Martínez-Gil, Bañó-Polo, Redondo et al. (2011) "Membrane integration of poliovirus 2B viroporin" *J Virol*
28. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
29. Mcknight, Swanson, Austgen et al. (2020) "Stimulator of interferon genes (STING) is an essential proviral host factor for human rhinovirus species A and C" *Proc Natl Acad Sci*
30. Longdon, Day, Alves et al. (2018) "Host shifts result in parallel genetic changes when viruses evolve in closely related species" *PLoS Pathog*
31. Remold, Rambaut, Turner (2008) "Evolutionary genomics of host adaptation in vesicular stomatitis virus" *Mol Biol Evol*
32. Sanjuán, Domingo-Calap (2021) "Genetic diversity and evolution of viral populations"
33. Wichman, Millstein, Bull (2005) "Adaptive molecular evolution for 13,000 phage generations: a possible arms race" *Genetics*
34. Mattenberger, Vila-Nistal, Geller (2021) "Increased RNA virus population diversity improves adaptability" *Sci Rep*
35. Riemersma, Jaeger, Crooks et al. (2021) "Rapid evolution of enhanced Zika virus virulence during direct vertebrate transmission chains" *J Virol*
36. Bons, Leemann, Metzner et al. (2020) "Long-term experimen tal evolution of HIV-1 reveals effects of environment and mutational history" *PLoS Biol*
37. Kaczorowska, Timmerman, Deijs et al. (2023) "Anellovirus evolution during long-term chronic infection" *Virus Evol*
38. Wahl, Tanaka (2022) "Hazardous loss of genetic diversity through selective sweeps in asexual populations" *Am Nat*
39. Pennings, Kryazhimskiy, Wakeley (2014) "Loss and recovery of genetic diversity in adapting populations of HIV" *PLoS Genet*
40. Rouzine, Coffin (2001) "Transition between stochastic evolution and deterministic evolution in the presence of selection: general theory and application to virology" *Microbiol Mol Biol Rev*
41. Moya, Holmes, González-Candelas (2004) "The population genetics and evolutionary epidemiology of RNA viruses" *Nat Rev Microbiol*
42. Jónás, Taus, Kosiol et al. (2016) "Estimating the effective population size from temporal allele frequency changes in experimental evolution" *Genetics*
43. Nei, Kumar (2000) "Molecular evolution and Phylogenetics"
44. Nelson, Hughes (2015) "Within-host nucleotide diversity of virus populations: insights from next-generation sequencing" *Infect Genet Evol*
45. Li, Zou, Jiang et al. (2019) "Biological function and application of picornaviral 2B protein: a new target for antiviral drug development" *Viruses*
46. Nieva, Madan, Carrasco (2012) "Viroporins: structure and biological functions" *Nat Rev Microbiol*
47. Martinez-Gil, Mingarro (2015) "Viroporins, examples of the two-stage membrane protein folding model" *Viruses*
48. Cuconati, Xiang, Lahser et al. (1998) "A protein linkage map of the P2 nonstructural proteins of poliovirus" *J Virol*
49. Agirre, Barco, Carrasco et al. (2002) "Viroporin-mediated membrane permeabilization. Pore formation by nonstructural poliovirus 2B protein" *J Biol Chem*
50. De Jong, Schrama, Willems et al. (2002) "Multimerization reactions of coxsackievirus proteins 2B, 2C and 2BC: a mammalian two-hybrid analysis" *J Gen Virol*
51. Patargias, Barke, Watts et al. (2009) "Model generation of viral channel forming 2B protein bundles from polio and coxsackie viruses" *Mol Membr Biol*
52. Ao, Sun, Guo (2014) "Topology and biological function of enterovirus non-structural protein 2B as a member of the viroporin family" *Vet Res*
53. Baggen, Thibaut, Strating et al. (2018) "The life cycle of non-polio enteroviruses and how to target it" *Nat Rev Microbiol*
54. Triantafilou, Ramanjulu, Booty et al. (2022) "Human rhinovirus promotes STING trafficking to replication organelles to promote viral replication" *Nat Commun*
55. Merkling, Van Rij (2015) "Analysis of resistance and tolerance to virus infection in Drosophila" *Nat Protoc*
57. Fuhrmann, Jablonski, Topolsky et al. (2024) "V-pipe 3.0: a sustainable pipeline for within-sample viral genetic diversity estimation" *Gigascience*
58. Li, Durbin (2009) "Fast and accurate short read alignment with Burrows-Wheeler transform" *Bioinformatics*
59. Posada-Céspedes, Seifert, Topolsky et al. (2021) "V-pipe: a computational pipeline for assessing viral genetic diversity from high-throughput data" *Bioinformatics*
60. (2026) *Full-Length Text Journal of Virology*
61. Fuhrmann, Langer, Topolsky et al. (2024) "VILOCA: sequencing quality-aware viral haplotype reconstruction and mutation calling for short-read and long-read data" *NAR Genom Bioinform*
62. Nelson, Moncla, Hughes (2015) "SNPGenie: estimating evolutionary parameters to detect natural selection using pooled nextgeneration sequencing data" *Bioinformatics*
63. Kuznetsova, Brockhoff, Christensen (2015) "Package "lmertest"
64. Lüdecke, Ben-Shachar, Patil et al. "2021. performance: an R package for assessment, comparison and testing of statistical models" *J Open Source Softw*
65. Kassambara (2020) "Pipe-Friendly framework for basic statistical tests"
66. Searle, Speed, Milliken (1980) "Population marginal means in the linear model: an alternative to least squares means" *Am Stat*
67. Bates, Mächler, Bolker et al. (2015) "Fitting linear mixedeffects models using lme4" *J Stat Softw*
68. Kerr, Dalwadi, Scott et al. (2018) "Transmission of Cricket paralysis virus via exosome-like vesicles during infection of Drosophila cells" *Sci Rep*
69. Marée, Keulen, Boucher et al. (2000) "Estimating relative fitness in viral competition experiments" *J Virol*
70. Chevin (2011) "On measuring selection in experimental evolution" *Biol Lett*
71. Camacho, Coulouris, Avagyan et al. (2009) "BLAST+: architecture and applications" *BMC Bioinformatics*
72. Papadopoulos, Agarwala (2007) "COBALT: constraint-based alignment tool for multiple protein sequences" *Bioinformatics*
73. Crooks, Hon, Chandonia et al. (2004) "WebLogo: a sequence logo generator" *Genome Res*
74. Schindelin, Arganda-Carreras, Frise et al. (2012) "Fiji: an opensource platform for biological-image analysis" *Nat Methods*
75. Bolte, Cordelières (2006) "A guided tour into subcellular colocaliza tion analysis in light microscopy" *J Microsc* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12656783&blobtype=pdf | # Genetic and Statistical Study of Anelloviruses and Gyroviruses in Diarrheic Cats and Their Co-Occurrence Patterns
Turhan Turan, Hakan Işıdan, Selda Duran-Yelken, Mustafa Atasoy, Remziye Özbek, Rania El Naggar, Mohammed Rohaim
## Abstract
Members of the Anelloviridae family are increasingly being recognized for their role in veterinary and public health, with domestic cats identified as potential carriers of anelloviruses and gyroviruses. This study aimed to investigate the prevalence and genetic characteristics of these viruses in diarrheic cats from Sivas, Türkiye. A total of 91 fecal samples were analysed, initially for feline panleukopenia virus using conventional PCR, followed by screening with our Anelloviridae panel. The results revealed that 19 (20.9%) samples were positive for TTFeV1, 32 (35.2%) for CAV, 67 (73.6%) for Avian gyrovirus 2, four (4.4%) for Gyrovirus 3, and three (3.3%) for Gyrovirus 4. Statistical analyses revealed frequent co-infections among parvoviruses, anelloviruses, and gyroviruses, with a significant association between Gyrovirus chickenanemia (CAV) and Gyrovirus galga1 (AvGyV2). Notably, Gyrovirus 4 (Gyrovirus homsa3) was identified in feline stool for the first time. Phylogenetic and genomic analyses, based on partial TATA box-ORF2 sequences for anelloviruses and VP1 sequences for gyroviruses, provided further insights into viral diversity. These findings expand current knowledge of anellovirus and gyrovirus circulation in feline populations, underscoring the importance of continued surveillance for feline and public health.
## 1. Introduction
The family Anelloviridae contains negative-sense, circular, and single-stranded DNA viruses with a genome of approximately 1.6-3.9 kb in length [1]. Anelloviruses were first described in Japan in 1997 in a patient who developed hepatitis as a result of blood transfusion [2]. Since then, they have been detected in many animal species such as pigs, non-human primates, marine mammals, bats, dogs, cats, ruminants, and equids [3][4][5][6][7]. Additionally, in cervical specimens, TTVs have been discovered to co-infect with other viruses, such as human papillomavirus, indicating the possibility of interactions between several viral species [8]. The Anelloviridae family includes 30 genera with high genetic heterogenity and 156 species belonging to these genera. According to recent update on the International Committee on Taxonomy of Viruses (ICTV), feline anelloviruses have been classified under the in the genus Etatorquevirus (Etatorquevirus felid1-5) and Tettorquevirus (Tettorquevirus felid6). In addition to these two genera, four of the 10 species in Gyrovirus genus, Chicken anemia virus (CAV), also known as Gyrovirus chickenanemia; Gyrovirus galga1 (previously referred to as Avian gyrovirus 2, or AvGyV2); Gyrovirus 3 (GyV3), also called "Gyrovirus homsa1"; and Gyrovirus 6 (GyV6) have been identified in cats so far [9][10][11][12][13].
Anelloviruses are highly prevalent in human and animal populations, but their importance on health remains uncertain. Myriads of studies have been conducted on Torque teno viruses (TTVs), pinpointing their significance as novel agents that could affect the health of humans and other animals. After the first report of feline TT viruses by Okamoto et al. (2002), the CAV genome was detected in breeder and commercial chicken flocks in South Korea [5,14]. A year later, a new human virus was discovered on the human skin and named "human gyrovirus" (HGyV) due to its similarity to the chicken anemia virus [15]. In the same year, another species of gyrovirus related to CAV, known as Avian gyrovirus 2 (AvGyV2), was identified in diseased chickens in Brazil [16]. Due to the high nucleotide identity of over 93% between HGyV and AvGyV2 in the VP1-VP3 gene, they have been regarded as the same species and given the name HGyV/AGV2. The genus Gyrovirus formally includes eleven species that have been discovered in a wide range of hosts, including chickens, birds, ferrets, humans, dogs, and cats [17,18]. Chicken anemia virus is a highly pathogenic and contagious viral agent characterized by severe aplastic anaemia, subcutaneous and muscular haemorrhages, thymus atrophy, abnormal feather development, and profound immunosuppression, thereby predisposing affected young chickens to secondary infections caused by various pathogens [19]. The GyV3 was detected in both diarrheal and formed stool samples from Chilean children in the USA, while Gyrovirus 4 (GyV4), or "Gyrovirus homsa3" has been detected in human and chicken stool, also raw chicken [20,21].
While well-known viruses like protoparvovirus carnivoran1, which is an aetiological agent of feline panleukopenia (otherwise known as feline distemper), have been extensively studied due to their major impact on domestic cats, the emergence of new viruses and their potential role in gastroenteric diseases often remain unforeseen. To address this gap, we aimed to investigate the prevalence of feline torque teno viruses, specifically Torque teno felid virus 1 (TTFeV1) and 2 (TTFeV2), as well as four different Gyrovirus species, in the diarrheal fecal samples of domestic cats, providing insight into their possible contribution to feline gastrointestinal disorders.
## 2. Materials and Methods
## 2.1. Sampling
A total of 91 rectal swab samples were collected between May 2019-May 2021 from cats suffering from mild to severe diarrhea. Samples were obtained from animals which was admitted to Sivas Cumhuriyet University Faculty of Veterinary Medicine, and six other private district veterinary clinics located in the Sivas province. Of the 91 rectal swab samples, 56 were from kittens under one year old, and 35 were from adult cats aged 1 to 8 years. The collected samples were transported to the laboratory just after sampling and stored at -80 • C until being subjected to DNA isolation. Samples were diluted to 1:10 with 1 M phosphate-buffered saline solution and centrifuged at 3000× g for 5 min to remove coarse particles. After centrifugation, the supernatants were submitted to a nucleic acid extraction procedure using a GF-1 Viral Nucleic Acid Extraction Kit (Vivantis, Subang Jaya, Malaysia) according to the manufacturer's instructions. The eluted DNAs were stored at -80 • C until use.
## 2.2. Polymerase Chain Reactions (PCRs)
DNA samples were initially screened by conventional PCR targeting the VP2 gene of Protoparvovirus carnivoran1, also known as Feline panleukopenia virus (FPV) in feline species using primer set previously described by Buonavoglia et al. (2001) [22]. The same samples were further investigated to detect feline anelloviruses and gyroviruses. All primers used in this study are listed in Table 1. All Reactions were performed in 30 µL volume using a commercial PCR master mix (Thermo, Waltham, MA, USA). PCR conditions were adjusted as follows: Following pre-denaturation at 95 • C for 2 min., 45 s at 94 • C for denaturation, annealing at 48-57 • C (listed in Table 1) for 30 s and 1 min. at 72 • C for extension were set to 40 cycles. Final extension was at 72 • C for 10 min. Amplified products were run on the 1% agarose gel containing GelRed Nucleic Acid Stain (Merck, Rahway, NJ, USA) and visualised by a standard transilluminator. The samples showing the expected band size were regarded as positive. 1). PCR amplicons were purified using Wizard SV Gel and PCR CleanUp System (Promega, Madison, WI, USA) and BigDye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and ABI Prism 310 Genetic Analyzer (ABI 3100; Applied Biosystems, Foster City, CA, USA). Among the positive samples, six Torque teno felis virus 1, five CAV, five AvGyV2, four Gyv3 and three GyV4 samples were randomly selected for bidirectional dideoxy sequencing. Partial sequences of six TATA Box-ORF2 gene of TTFeV1 were compared with other available strains retrieved from GenBank. Phylogenetic analysis of partial VP1 gene sequences of gyroviruses (CAV, AvGyV2, GyV3, and GyV4) were performed and compared with other publicly available anellovirus data in the NCBI database.
The phylogenetic analysis was conducted using MEGAX (v.10.2.6) software by applying the maximum likelihood (ML) method [23]. MEGAX software was utilized to evaluate the Bayesian Information Criterion (BIC) in order to select the best-fit substitution model. The model with the lowest BIC value was chosen as it represents the best fit for the data.
Then phylogenetic trees were generated using the maximum likelihood method and the Kimura-2 parameter model (+G) with 1000 replicates [24]. The nucleotide (nt) identities were calculated using the SIAS on-line tool (http://imed.med.ucm.es/Tools/sias.html; accessed on 2 November 2024). The datasets were individually analysed for potential recombination events in both structural and non-structural genes using RDP4 v4.100 software, applying multiple algorithms (RDP, GENECONV, Chimaera, Max-Chi, BootScan, SiScan, and 3Seq) with default parameters and a p-value threshold of 0.05 [25].
## 2.4. Statistical Analyses
All statistical analyses and visualizations were performed using RStudio (version 2024.09.0) (http://www.rstudio.com/). Initially, the statistical relationship between feline panleukopenia virus (FPV) and the other viruses subjected to this study (circoviruses and anelloviruses) was examined individually using a chi-squared test, along with calculations of both phi coefficient and Cramér's V to assess the direction and strength of the associations. Then, pairwise comparisons were performed among the remaining viruses using the same methodology to identify any significant associations.
## 3. Results
## 3.1. Prevalence of Feline Panleukopenia Virus
From the 91 samples initially investigated, 37 samples were positive for feline panleukopenia virus (40.65%). The positivity rate varied significantly across age groups: only 20% (7/35) of the adult cats tested positive for FPV, while the rate peaked at 53.57% (30/56) among kittens (Figure 1a). PCR detection results for the seven viruses in adults and kittens are summarized in Table 2.
## 3.2. Prevalence of the Anelloviridae Family
PCR positivity for anelloviruses and gyroviruses was detected in 75 of 91 rectal swab samples of cats. Of the 56 kitten samples, 13 (23.21%) were positive for TTFeV1, 21 (37.50%) for CAV, 42 (75.00%) for AvGyV2, 3 (5.36%) for GyV3, and 2 (3.57%) for GyV4, whereas, of the 35 adult cat samples, 6 (17.14%) were positive for TTFeV1, 11 (31.43%) for CAV, 25 (71.43%) for AvGyV2, 1 (3.86%) for GyV3, and 1 (3.86%) for GyV4 (Figure 1a). Furthermore, a single anellovirus infection was detected in 23 cases among the 56 kittens with diarrhea. Mixed infections were observed in the remaining kittens: 15 had dual infections, eight had triple infections, and one had a quintuple infection. Furthermore, among the 35 adult cats with diarrhea, 13 had a single infection, while mixed infections were detected in the others: 14 had dual infections, and only one had a triunal infection (Figure 1b).
## 3.3. Genetic Profiling and Characteristics of Anelloviruses and Gyroviruses
After PCRs, six samples for TTFeV1, five samples for CAV, five samples for AvGyV2, four samples for GyV3, and three samples for GyV4 were successfully sequenced for molecular analysis.
## 3.3.1. Molecular Analysis of TTFeV1
A phylogenetic tree was constructed with 330 bp-partial sequences of the six TATA Box-ORF2 genes of TTFeVs in this study retrieved from GenBank. Phylogenetic analysis revealed a distinctive distribution between the TTFeV1 and TTFeV2 genera, which were supported by a high bootstrap value (100%). All Turkish strains clustered together with the strain Fc-TTV4 (NC_014072), and some other strains majorly reported from China (Figure 2). The identity percentage of TTFeV1 viruses ranged between 96.36% and 99.39% among themselves. Furthermore, five of the six TTFeV1 samples were more closely related to Chinese strains with a range of identity, while sample FeTTV28 was identical with PRA4 and chat-13 FRA (KM593801) and (EF538878) isolates previously reported from cat saliva (Figure 2).
## 3.3.2. Molecular Analysis of Chicken Anemia Virus
Pairwise sequence analysis of 351 bases revealed that the five strains were closely related, with high identity ranging from 96.30% to 100%. Notably, Fe-CAV43/TUR and Fe-CAV60/TUR were identical based on partial sequences. However, overall identity dropped to 94.02% when additional Turkish strains were included. Further analysis using the reference strain Cuxhaven-1 (NC_001427.1) identified various potential polymorphisms in the 17 deduced amino acid sequences of the Cux-1 gene and the 117 amino acid sequences of the nucleocapsid gene (ORF3). These mutations were compared with other partial sequences, including strains previously reported from Türkiye. The partial alignment results revealed point mutations at positions V75, M97, I125, K139, D144, and V157 in the capsid protein sequence (Figure 3a). Notably, these mutations were also observed in other Turkish strains. Furthermore, the C-terminal end of the Cux-1 protein was identical to the reference sequence. However, non-synonymous mutations in Fe-CAV43/TUR and Fe-CAV60/TUR resulted in a conversion from opal (TGA) to ochre (TAA), a change also observed in the isolate EB1K, which was previously identified from a broiler.
Phylogenetic analysis using 351-base nucleotide sequences classified the samples into three distinct groups (Genotypes I-III) with strong bootstrap values. Our strains primarily clustered within Group III, further branching into multiple subgroups. Fe-CAV43/TUR, Fe-CAV60/TUR, and Fe-CAV76/TUR formed a distinct branch alongside other strains previously reported from Türkiye, while Fe-CAV33/TUR and Fe-CAV46/TUR diverged from this main branch (Figure 3b). Despite being classified under the Anelloviridae genus, multiple sequence analyses of partial nucleocapsid sequences for GyV2, GyV3, and GyV4 were applied individually due to high dissimilarity within the species. The 609 bp nucleotide and 203 derived amino acid (aa) sequences of the partial VP1 gene of AvGyV2 were utilised to conduct multiple sequence alignments with reference sequence (Clone Ave3, NC_015396.1). Turkish strains had varied nucleotide identity between 92.78% and 98.36%. FeAvGyV2-31/TUR showed high identity to reference genome (99.51%), whereas FeAvGyV2-46 was identical to two Chinese strains, isolate HLJ1508 (KX708510) and strain 27-GD201810 (OQ116651), both isolated from chickens. Other two strains, FeAvGyV2-38/TUR and FeAvGyV2-49/TUR, possessed varied degree of identity to other strains dominantly isolated from China. On the other hand, FeAvGyV2-20/TUR showed slight deviation from other Turkish strains, exhibiting 98.52% identity to the isolate JP/KGSM/M0313-2Li/97, which was obtained from cryopreserved organs of a broiler in Japan in 1997. Furthermore, the deduced amino acid sequences of partial AvGyV2 VP1 sequences revealed a series of substitutions compared to the reference sequence, S154A substitution in three samples (FeAvGyV2-20, 38, 49), R209S in FeAvGyV2-49, R212K and G242R in five samples (FeAvGyV2-20,-31,-38,-46,-49), A270S in four samples (FeAvGyV2-20,-38,-46,-49), V288Q and G293Q in three samples (FeAvGyV2-38,-46,-49), and Q310E in three samples (FeAvGyV2-20,-46,-49). In addition, the V288I mutation in FeAvGyV2-20 was found to be unique.
Phylogenetic trees were built using 609 bp partial VP1 nucleotide sequences of other available strains in GenBank (n = 42). Phylogenetic analysis of AvGyV2 revealed two main clades, supported by a high bootstrap value (100%), with Turkish strains clustering in clade 1 as displayed in Figure 4 (top). Clade 1 was further divided into subclades 1a and 1b, supported by a significant bootstrap value (76%). FeAvGyV2-20/TUR was grouped into subclade 1a, while the remaining Turkish strains clustered in subclade 1b along with the reference genome.
The 642 bp nucleotide sequence and the 214 amino acid (aa) sequence derived from the partial VP1 gene of GyV3 were used to perform multiple sequence alignments with the reference sequence (strain FecGy, NC_017091). The Turkish strains exhibited a low level of nucleotide variation, ranging from 98.60% to 100%, with Fe-GyV3-29/TUR and Fe-GyV3-90/TUR being identical to each other. These two strains were closely related (99.53%) to isolate BR_DF5 (MN175607.1), identified from wild bird feces in Brazil. In contrast, Fe-GyV3-57/TUR and Fe-GyV3-75/TUR sequences showed the highest nucleotide identities (98.75%) with two GyV3 strains from China, which were isolated from domestic cat feces (isolate 17CC0704, MK089247; isolate 17CC0711, MK089248.1). The deduced amino acid sequences of Turkish strains were compared with the reference sequence (NC_017091) and all of them found to be identical. A phylogenetic tree was further constructed based on 15 available strains from GenBank, revealing two main clades as displayed in Figure 4 (middle). Clade 1 included the reference sequence and the majority of strains, including Turkish strains, with a minimum identity of 97.35%. Clade 2 comprised only two strains (GyV3/LY-GyV3-202201/CHN, OR271604 and GyV3/NC19-Gyv3-01/CHN, OR271605), sharing 92.52% to 94.24% identity with Clade 1.
The 678 bp nucleotide sequence and the 226 amino acid (aa) sequence derived from the partial VP1 gene of GyV4 were used to perform multiple sequence alignments with the reference sequence (strain GyV4/D137/CHN, NC_018401) along with six other available strains. Our three Turkish strains showed high identity, ranging from 99.41% to 99.71%, and exhibited a close percentage of identity (98.08-98.38%) with a Brazilian strain (GyV4/BRA, MT671983) isolated from chicken meat. Furthermore, the 226-deduced amino acid sequence of Turkish strains displayed various point mutations: the Y116H substitution was detected in Fe-GyV4-37/TUR and found to be unique, whereas the S153T, S175A, and L244T substitutions were detected in all three samples. Additionally, the phylogenetic tree revealed two main clades, Clade 1 and Clade 2, with the Turkish strains falling into Clade 1 as shown in Figure 4 (bottom).
## 4. Discussion
Diarrhea in kittens is frequently considered as one of the significant pre-mortem risk factors due to its high incidence, leading to malabsorption, weight loss, cachexia, and death in the early weeks of life [26]. Kitten diarrhea can be attributed to wide range of factors, including bacterial and viral diseases, such as colibacillosis [27], or parvoviruses [28], drug intoxication [29], or food sensitivity [30]. To better understand the nature of viral diarrhea in kittens and implement effective mitigation measures, it is important to address other pathogens that may serve as contributing factors. Therefore, we aimed to investigate the existence and prevalence of various anelloviruses and gyroviruses in fecal samples from domestic cats exhibiting diarrhea. Our samples were initially investigated using a universal primer set detecting feline panleukopenia virus. Our results revealed an overall FPV positivity rate of 40.65% (37/91) in the investigated samples. The prevalence was notably higher in kittens at 53.57% (30/56) but declined to 20% (7/35) in adult cats. Several endeavours have been made in various provinces of Türkiye, all of which resulted in an overall prevalence between 10% and 25%, varying based on factors such as sample size, location, methodology, cats' age, and habitat [31,32]. Similarly, many reports have been documented worldwide, including in China, with a prevalence ranging from 8.54% to 45.09% [33,34], 11.32% in India [35], 22.90% in Bangladesh [36], and 43.00% in Egypt [37]. Furthermore, young kittens are known to be more susceptible to infection than adult cats over two years old [38,39]. On the other hand, ten samples (10.99%) yielded negative results for all viruses investigated in this study, while among the 37 FPV-positive samples, six were positive only for FPV (6.59% in total) with no other viruses detected through our Anelloviridae panel. FPV may occur alone or in conjunction with other viruses that share similar pathogenic mechanisms, such as astroviruses, caliciviruses, kobuviruses, and other parvovirus species, which can either cause independent infections or contribute to concurrent infections [40][41][42]. Overall, our findings demonstrated a relatively high prevalence of FPV in the Sivas Region, particularly in kittens, which was somewhat consistent with previous reports. Nevertheless, the contribution of other major diarrheal viruses to the clinical manifestations of FPV infection remains to be elucidated.
Torque teno felis viruses, first identified in Japan by Okamoto et al. (2002) in cat serum and later detected in countries such as France and China, are a newly described virus species found at a high frequency in various animal species [5,9,12,43,44]. Zhu et al. (2011) reported that 12.5% (2/16) of serum samples tested positive for FeTTV, while Jarošová et al. (2015) later found a higher prevalence of 33.63% (37/110) in serum samples from Czechia [9,12]. A recent study conducted by Gao et al. (2023) demonstrated 36.67% (11/30) positivity rate in stool samples [43]. This study revealed an overall TTFeV-1 prevalence of 20.88% (19/91) in rectal swabs from domestic cats, with a slightly higher rate in kittens (23.21%) than in adult cats (17.14%). Phylogenetic analysis of our FeTTV sequences using the maximum likelihood method further revealed varying degrees of genetic similarity, with the sequences clustering into a large group of FeTTVs from domestic cats across China, France, the USA, and Czechia. Finally, sequencing results unveiled the genomic characteristics of six more Etatorquevirus felid1-associated strains, reflecting the genetic diversity of the virus in the region. We attempted to detect other felid torque teno virus species using predesigned primer sets, with samples previously isolated from chicken as a positive control; however, no positive results were found. These findings might serve as a foundation for future research into the rapid detection of FeTTV and the role of TTV in domestic cats.
A recent comprehensive study by Kraberger et al. (2021) pointed out the intricacy of Torque teno felid viruses' evolutionary mechanisms in the Felidae family, with reference to recombination events stemming from simultaneous infections by Anelloviridae [45]. The of multiple anelloviruses and gyroviruses have been extensively documented across various studies, including those on feline [13] and human fecal samples [46], as well as rodent spleens [47]. Furthermore, Liu et al. (2022) revealed the coexistence of a bocaparvovirus-like virus and Anelloviridae members in the lymph nodes of the Amur leopard cat (Prionailurus bengalensis euptilurus), further confirming the significant genetic similarity of the bocaparvovirus to strains previously identified in diarrheic cats [48]. In a similar vein, substantial evidence has been presented so far indicating that various viruses, including Epstein-Barr virus, papillomavirus, and hepatitis C virus, may act as facilitating agents in the replication process of TTVs, potentially contributing to the development of diseases such as multiple sclerosis and carcinoma [8,49,50]. On the other hand, TTVs have been identified as components of the microbiome in healthy pregnant women, detected incidentally in sources like raw buffalo milk, and even associated with potential benefits for human health, such as reducing the risk of schizophrenia [51][52][53]. In this study, we documented the coexistence of DNA virus species from the Parvoviridae and Anelloviridae families and also showed the co-occurrence of anellovirus and gyrovirus species in diarrheic cats. However, our further statistical analysis demonstrated no statistically significant associations were observed between TTFeV1 and the other detected viruses. Taken together, we hypothesised that TTFeVs might be a substantial component of the felid gastrointestinal flora; however, interactions with some other pathogens could also fluctuate the viral load and potentially alter the virus's role in feline diarrhea.
The first detection of the CAV-related strain (CAT-CAV, KC414026) in domestic cats was reported by Zhang and colleagues (2012), who detected recombination events within the partial C-terminus of the VP1 gene and UTR regions [10]. Then, a more recent complementary study investigated the occurrence of gyroviruses in domestic cats, identifying the presence of AvGyV2, GyV3, and GyV6 for the first time. Notably, the overall prevalence of GyVs was significantly higher in healthy cats (8.7%) than in diarrhoeic cats (3.8%) [13]. In this study, we observed an exceptionally high rate of Chicken Anemia Virus (35.16%) and AvGyV2 (73.63%) viral DNA positivity in cat stools. Furthermore, statistical analysis indicated the moderate level of association (p < 0.05) between chicken anemia virus and AvGyV2 in cat samples. A recent study by Yang et al. (2023) demonstrated the convergence of these two viruses in chickens, confirming that their co-application to SPF chickens exacerbates immunosuppression, pathogenicity, and viral replication of both viruses [54]. Taken together, we conjecture that a similar synergistic mechanism between these two gyroviruses could exist in other hosts, potentially leading to a higher prevalence of gyroviruses in domestic cats.
In this study, we successfully amplified a 351 bp nucleotide sequence, representing approximately 15% of the entire genome, which has provided valuable biological insights into the chicken anemia virus identified in the Turkish cat population. Phylogenetic analysis further demonstrated high variability among the sequences, with Fe-CAV33/TUR and Fe-CAV46/TUR clustering together in a single branch, whereas Fe-CAV43/TUR, Fe-CAV60/TUR, and Fe-CAV76/TUR formed a separate branch alongside a central Turkish cluster. A recent study by Song et al. (2024) analysed the genomic sequences of 28 CAV strains, identifying three primary groups and additional subgroups based on complete genome data [55]. Upon slight modification of this dataset and subsequent phylogenetic analysis based on partial nucleotide sequence, Groups I-III were distinctly identified; however, the bootstrap values were lower than anticipated (below 70), offering limited statistical support for the subdivision of Groups IIa-IIb and IIIa-IIIb. In conclusion, the primer set designed for this study effectively distinguishes groups; however, a longer amplicon may be required for the identification of subgroups.
VP1 gene, which encodes the only structural protein of gyroviruses, plays an integral role in receptor binding and mediating interactions between the virus and its host [56,57]. Therefore, it is plausible that mutations in the VP1 protein could influence cell tropism and contribute to the emergence of immune escape mutants. An initial study by Renshaw et al. (1996) identified various point mutations by comparing cell cultureadapted CAV (isolate ConnB) with field isolates and detected a hypervariable region (HVR) between 139 and 151 that may alter the pathogenicity and transmissibility of viral strains [58]. Later, Todd et al. (2002) identified additional point mutations associated with viral pathogenicity in vivo at positions 75, 89, 125, 141, and 144, which may be related to the virus's adaptation during interspecies transmission [59]. Although several strains have been previously reported in chickens in Türkiye [60,61], none have been identified in cats; thus, we conducted a comparative analysis of our strains alongside these reported, reference (Cuxhaven 1), and attenuated strains (clone 33, clone 34 and ConnB) based on the partial amino acid sequence of the VP1 protein. We identified several significant mutations. First, the V75I substitution was observed in Fe-CAV43/TUR, Fe-CAV60/TUR, and Fe-CAV76/TUR. Second, the I125L substitution was detected in Fe-CAV33/TUR and Fe-CAV46/TUR. Third, various alterations were found in hypervariable regions, particularly the K139Q and D144Q/E mutations. These mutations were frequently detected in other strains, including previously reported Turkish strains, and were less likely to be part of the feline host adaptation process. Therefore, we concluded that amino acid sequences clearly reflect the genetic diversity of CAV in Türkiye, highlighting the need for further genomic characterization to elucidate virus-host interactions.
AvGyV2 has been reported worldwide, and multiple studies have confirmed its presence in humans, dogs, cats, and ferrets, underscoring its high host range plasticity. [15,[62][63][64][65][66]. Our phylogenetic analysis revealed significant diversity in the VP1 gene of Turkish strains, with FeAvGyV2-20/TUR grouping into clade 1 while the remaining strains clustered in clade 2, both consisting of isolates with no apparent geographical link. Furthermore, multiple sequence analyses based on the 203 deduced amino acid sequences revealed various mutations, which were interpreted in the context of significant alterations. Yao et al. (2017) drew insights into evolutionary trends of avian gyrovirus 2 in China by comparing the structural proteins of 17 strains and annotating hypervariable regions between positions 288 and 314, as well as a replication motif between positions 325 and 329 in VP1 protein [62]. Similarly, Liu et al. (2022) investigated the dynamics of host species transmission and identified several canine-related non-synonymous mutations [64]. We observed that our strains had several mutations in hypervariable regions, based on the reference sequence, which were common in other strains and the replication motif (FAAL) was well conserved. However, we detected V288I mutation in the same HV region, which was unique for FeAvGyV2-20/TUR. Taken together, it can be inferred that random mutations may be generated in this hypervariable region, which could contribute to interspecies transmission.
We compared our partial VP1 genome of GyV3 and GyV4 with GenBank sequences and found two main clades for each species due to limited data. All Turkish GyV3 strains were into Clade 1 with reference sequences, whereas two Chinese strains were outgrouped. Furthermore, Fe-GyV3-29/TUR and Fe-GyV3-90/TUR had the highest identity with a bird originated strain (99.53%, isolate BR_DF5), whereas Fe-GyV3-57/TUR and Fe-GyV3-75/TUR were closely related to domestic cat fecal samples [13,67]. This could be attributed to different sources of infection in each individual cat. On the other hand, the deduced amino acid sequence showed that all four strains were identical to the partial VP1 protein of the reference genome, indicating that the mutations in the genome remain synonymous. The infectivity of GyV3 and its broad host-range capacity have been extensively studied chickens and mice, showing that GyV3 targets hematopoietic cells in the bone marrow, leading to severe aplastic anemia and immunosuppression, as well as hepatitis and gastroenteritis in both hosts [68,69]. Furthermore, mutations in the process of infection in mice revealed a hypervariable region in the C-terminus of VP1, which could not be detected in this study [68]. Taken together, our findings suggested that cats are naturally infected with GyV3 from multiple sources at a low prevalence, without any evidence of adaptation. Given that GyV3 has been detected in human faeces [46], its presence is particularly noteworthy due to the close interactions between humans and animals, which may facilitate viral transmission. The identification of gyrovirus DNA in fecal samples from diarrheic cats emphasizes the importance of continued monitoring to assess potential health risks.
Partial VP1 sequences were used to generate a phylogenetic tree, where Turkish GyV4 sequences grouped with the reference strain in Clade 1 and showed the closest relationship to a strain previously isolated from chicken meat in Brazil [70]. Two strains, isolated from ferrets and chickens, exhibited high identity and formed Clade 2. Then, multiple sequence alignments were conducted based on the 226 bp amino acid sequence of the VP1 gene, which is relatively closer to the N-terminal end. Our analysis revealed a unique Y21H mutation in Fe-GyV4-37/TUR, while the other two sequences exhibited common alterations. Due to limited data, we could not identify any associations between strains based on either geographical location or host species. Notably, we report the occurrence of GyV4 in cat samples for the first time.
This study has several limitations. Faecal samples were obtained from the Faculty Animal Hospital and six private veterinary clinics across different districts of Sivas, Türkiye. Most specimens originated from cats exhibiting mild to severe diarrhea; however, a standardized clinical scoring system was not applied by participating practitioners. The majority of samples were collected from rural clinics, and owners of cats that recovered typically did not return for follow-up examinations despite our recommendations. As the sampling was restricted to diarrheic animals, the prevalence of anelloviruses and gyroviruses among healthy cats could not be evaluated. Moreover, sequence and phylogenetic analyses were conducted using partial VP1 fragments, which limited the ability to perform comprehensive structural comparisons among all gyrovirus species and required separate analyses for each viral group.
## 5. Conclusions
This investigation provides new insights into the prevalence and molecular characteristics of anelloviruses and gyroviruses in diarrheic domestic cats. The detection of TTFeV1, CAV, AvGyV2, GyV3, and GyV4 broadens the recognized diversity of circular DNA viruses associated with feline disease. These findings highlight the necessity of continued molecular surveillance and broader epidemiological studies, including healthy populations, to clarify the pathogenic and ecological roles of these viruses and to support both animal and public health.
## References
1. Varsani, Opriessnig, Celer et al. (2021) "Taxonomic update for mammalian anelloviruses (family Anelloviridae)" *Arch. Virol*
2. Nishizawa, Okamoto, Konishi et al. (1997) "A novel DNA virus (TTV) associated with elevated transaminase levels in posttransfusion hepatitis of unknown etiology" *Biochem. Biophys. Res. Commun*
3. Mushahwar, Erker, Muerhoff et al. (1999) "Molecular and biophysical characterization of TT virus: Evidence for a new virus family infecting humans"
4. Okamoto, Nishizawa, Takahashi et al. (2001) "Genomic and evolutionary characterization of TT virus (TTV) in tupaias and comparison with species-specific TTVs in humans and non-human primates" *J. Gen. Virol*
5. Okamoto, Takahashi, Nishizawa et al. (2002) "Genomic characterization of TT viruses (TTVs) in pigs, cats and dogs and their relatedness with species-specific TTVs in primates and tupaias" *J. Gen. Virol*
6. Li, Giannitti, Low et al. (2015) "Exploring the virome of diseased horses" *J. Gen. Virol*
7. Polster, Lechmann, Lienhard et al. "First Report of TTSuV1 in Domestic Swiss Pigs"
8. Siahpoush, Noorbazargan, Kalantari et al. (2022) "Coinfection of torque teno virus (TTV) and human papillomavirus (HPV) in cervical samples of women living in Tehran" *Iran. Iran. J. Microbiol*
9. Zhu, Shan, Cui et al. (2011) "Molecular detection and sequence analysis of feline Torque teno virus (TTV) in China" *Virus Res*
10. Zhang, Liu, Ji et al. (2012) "Identification of a chicken anemia virus variant-related gyrovirus in stray cats in china" *BioMed Res. Int*
11. Zhang, Li, Deng et al. (2014) "Faecal virome of cats in an animal shelter" *J. Gen. Virol*
12. Jarosova, Hrazdilova, Filipejova et al. (2015) "Whole genome sequencing and phylogenetic analysis of feline anelloviruses" *Infect. Genet. Evol*
13. Niu, Yi, Dong et al. (2019) "Genomic Characterization of Diverse Gyroviruses Identified in the Feces of Domestic Cats. Sci. Rep"
14. Kim, Kwon, Bae et al. (2010) "Molecular characterization of chicken infectious anemia viruses detected from breeder and broiler chickens in South Korea" *Poult. Sci*
15. Sauvage, Cheval, Foulongne et al. (2011) "Identification of the First Human, a Virus Related to Chicken Anemia Virus" *J. Virol*
16. Rijsewijk, Santos, Teixeira et al. (2011) "Discovery of a genome of a distant relative of chicken anemia virus reveals a new member of the genus Gyrovirus" *Arch. Virol*
17. Feher, Bali, Kaszab et al. (2022) "A novel gyrovirus in a common pheasant (Phasianus colchicus) with poult enteritis and mortality syndrome" *Arch. Virol*
18. Varsani, Kraberger, Opriessnig et al. (2023) "Anelloviridae taxonomy update 2023" *Arch. Virol*
19. Karki, Bora, Manu et al. (2024) "An updated review on chicken infectious anaemia" *World Poult. Sci. J*
20. Chu, Poon, Chiu et al. (2012) "Characterization of a novel gyrovirus in human stool and chicken meat" *J. Clin. Virol*
21. Lima, Cibulski, Finkler et al. (2017) "Faecal virome of healthy chickens reveals a large diversity of the eukaryote viral community, including novel circular ssDNA viruses" *J. Gen. Virol*
22. Buonavoglia, Martella, Pratelli et al. (2001) "Evidence for evolution of canine parvovirus type 2 in Italy" *J. Gen. Virol*
23. Stecher, Tamura, Kumar (2020) "Molecular Evolutionary Genetics Analysis (MEGA) for macOS" *Mol. Biol. Evol*
24. Guindon, Lethiec, Duroux et al. (2005) "PHYML Online-A web server for fast maximum likelihood-based phylogenetic inference" *Nucleic Acids Res*
25. Martin, Murrell, Golden et al. (2015) "RDP4: Detection and analysis of recombination patterns in virus genomes" *Virus Evol*
26. Dolan, Doyle, Tran et al. (2021) "Pre-mortem risk factors for mortality in kittens less than 8 weeks old at a dedicated kitten nursery" *J. Feline Med. Surg*
27. Watson, Jacob, Flowers et al. (2017) "Association of Atypical Enteropathogenic Escherichia coli with Diarrhea and Related Mortality in Kittens" *J. Clin. Microbiol*
28. Stuetzer, Hartmann (2014) "Feline parvovirus infection and associated diseases" *Vet. J*
29. Rubini, Rubini, Bertocchi et al. (2024) "A case of severe benzalkonium chloride intoxication in a cat" *Acta Vet. Scand*
30. Guilford, Jones, Markwell et al. (2001) "Food sensitivity in cats with chronic idiopathic gastrointestinal problems" *J. Vet. Intern. Med*
31. Aydin, Timurkan (2018) "A pilot study on feline astrovirus and feline panleukopenia virus in shelter cats in Erzurum" *Turkey. Rev. Med. Vet.-Toulouse*
32. Hasircioglu, Aslim, Kale et al. "Molecular characterization of carnivore protoparvovirus strains circulating in cats in Turkey"
33. Wang, Yan, Liu et al. (2021) "Prevalence and Molecular Evolution of Parvovirus in Cats in Eastern Shandong" *Transbound. Emerg. Dis*
34. Tang, Tang, Zhu et al. (2022) "Molecular characteristics and genetic evolutionary analyses of circulating parvoviruses derived from cats in Beijing" *BMC Vet.-Res*
35. Mukhopadhyay, Nookala, Thangamani et al. (2017) "Molecular characterisation of parvoviruses from domestic cats reveals emergence of newer variants in India" *J. Feline Med. Surg*
36. Kabir, Habib, Chouhan et al. "Epidemiology and molecular characterization of Feline panleukopenia virus from suspected domestic cats in selected Bangladesh regions" *PLoS ONE*
37. Abdel-Baky, El-Khabaz, Abdelbaset et al. (2023) "Clinico-epidemiological survey of feline parvovirus circulating in three Egyptian provinces from 2020 to 2021" *Arch. Virol*
38. Rehme, Hartmann, Truyen et al. (1248) "Feline Panleukopenia Outbreaks and Risk Factors in Cats in Animal Shelters" *Viruses*
39. Truyen, Addie, Belak et al. (2009) "Feline panleukopenia. ABCD guidelines on prevention and management" *J. Feline Med. Surg*
40. Brussel, Wang, Shi et al. (2020) "Identification of novel astroviruses in the gastrointestinal tract of domestic cats" *Viruses*
41. Castro, Garcia, Fumian et al. (2015) "Detection and molecular characterization of caliciviruses (vesivirus and norovirus) in an outbreak of acute diarrhea in kittens from Brazil" *Vet. J*
42. Van Brussel, Wang, Shi et al. (2022) "The enteric virome of cats with feline panleukopenia differs in abundance and diversity from healthy cats" *Transbound. Emerg. Dis*
43. Gao, Liu, Yi et al. (2023) "Genomic Characteristics of Feline Anelloviruses Isolated from Domestic Cats in"
44. Biagini, Uch, Belhouchet et al. (2007) "Circular genomes related to anelloviruses identified in human and animal samples by using a combined rolling-circle amplification/sequence-independent single primer amplification approach" *J. Gen. Virol*
45. Kraberger, Serieys, Richet et al. (2021) "Complex evolutionary history of felid anelloviruses"
46. Phan, Li, O'ryan et al. (2012) "A third gyrovirus species in human faeces" *J. Gen. Virol*
47. Nishiyama, Dutia, Stewart et al. (2014) "Identification of novel anelloviruses with broad diversity in UK rodents" *J. Gen. Virol*
48. Liu, Sun, Tu et al. "Virome Profiling of an Amur leopard cat Reveals Multiple Anelloviruses and a Bocaparvovirus" *Vet. Sci*
49. Borkosky, Whitley, Kopp-Schneider et al. (2012) "Epstein-Barr virus stimulates torque teno virus replication: A possible relationship to multiple sclerosis" *PLoS ONE*
50. Rosa, Araujo, Savassi-Ribas et al. (2017) "Prevalence of occult hepatitis B virus infection and Torque teno virus infection and their association with hepatocellular carcinoma in chronic hepatitis C patients" *Virus Res*
51. Roperto, Paciello, Paolini et al. (2009) "Short communication: Detection of human Torque teno virus in the milk of water buffaloes (Bubalus bubalis)" *J. Dairy Sci*
52. Chan, Tam, Yeo et al. (2001) "High carriage rate of TT virus in the cervices of pregnant women" *Clin. Infect. Dis*
53. Canuti, Buka, Jazaeri Farsani et al. (2015) "Reduced maternal levels of common viruses during pregnancy predict offspring psychosis: Potential role of enhanced maternal immune activity?" *Schizophr. Res*
54. Yang, Yang, Bi et al. (2023) "The Synergy of Chicken Anemia Virus and Gyrovirus Homsa 1 in Chickens" *Viruses*
55. Song, Kim, Kwon et al. (2024) "Genetic characterization of chicken infectious anaemia viruses isolated in Korea and their pathogenicity in chicks" *Front. Cell Infect. Microbiol*
56. Noteborn, Verschueren, Koch et al. (1998) "Simultaneous expression of recombinant baculovirus-encoded chicken anaemia virus (CAV) proteins VP1 and VP2 is required for formation of the CAV-specific neutralizing epitope" *J. Gen. Virol*
57. Fang, Zhen, Su et al. (2019) "Efficacy of CpG-ODN and Freund's immune adjuvants on antibody responses induced by chicken infectious anemia virus VP1, VP2, and VP3 subunit proteins" *Poult. Sci*
58. Renshaw, Soine, Weinkle et al. (1996) "A hypervariable region in VP1 of chicken infectious anemia virus mediates rate of spread and cell tropism in tissue culture" *J. Virol*
59. Todd, Scott, Ball et al. (2002) "Molecular basis of the attenuation exhibited by molecularly cloned highly passaged chicken anemia virus isolates" *J. Virol*
60. Abayli, Tonbak, Ozbek et al. (2011) "Viral analysis of tumor-bearing chicken flocks in Turkey over the last decade" *J. Hell. Vet. Med. Soc*
61. Okay, Askar (2021) "Molecular characterization of VP2 and VP3 proteins of chicken anemia virus isolates in Turkey" *Turk. J. Vet. Anim. Sci*
62. Yao, Gao, Tuo et al. "Novel characteristics of the avian gyrovirus 2 genome" *Sci*
63. Maggi, Macera, Focosi et al. (2012) "Human gyrovirus DNA in human blood" *Italy. Emerg. Infect. Dis*
64. Liu, Lv, Li et al. (2022) "Cross-species transmission potential of chicken anemia virus and avian gyrovirus 2" *Infect. Genet. Evol*
65. Fehér, Pazár, Kovács et al. (2014) "Molecular detection and characterization of human gyroviruses identified in the ferret fecal virome" *Arch. Virol*
66. Xu, Man, Yu et al. (2024) "Molecular analysis of Gyrovirus galga1 variants identified from the sera of dogs and cats in China" *Vet. Q*
67. Duarte, Silva, Brito et al. (2019) "Faecal Virome Analysis of Wild Animals from Brazil" *Viruses*
68. Yuan, Yan, Huang et al. (2021) "Cross-species pathogenicity of gyrovirus 3 in experimentally infected chickens and mice" *Vet. Microbiol*
69. Li, Zhou, Zhao et al. (2021) "Kinetic analysis of pathogenicity and tissue tropism of gyrovirus 3 in experimentally infected chickens" *Vet. Res*
70. Cibulski, Alves De Lima, Fernandes Dos Santos et al. (2021) "A plate of viruses: Viral metagenomics of supermarket chicken, pork and beef from Brazil" *Virology*
71. "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=PMC12724275&blobtype=pdf | # TRIM29 inhibits PRRSV replication by targeting nsp11 for degradation
Wei Wen, Zhenghong Xue, Yi Lu, Yuhang Liu, Wenqiang Wang, Zhenbang Zhu, Xiangdong Li
## Abstract
Ubiquitination plays critical roles in viral infections. This study demonstrates that porcine reproductive and respiratory syndrome virus (PRRSV) endoribonuclease nsp11 undergoes K48-linked polyubiquitination specifically at the conserved catalytic residue lysine 173 (K173) during viral infection. This modification targets nsp11 for degradation via the ubiquitin-proteasome system (UPS), as evidenced by the profound stabilization of a ubiquitination-deficient K173R mutant. Remarkably, this ubiquitination mechanism targeting the endonuclease active site is evolutionarily conserved across most arteriviruses, including simian hemorrhagic fever virus and equine arteritis virus, with mouse lactate dehydrogenase-elevating virus being an exception. We further identify the host E3 ubiquitin ligase TRIM29 as a key regulator that binds PRRSV nsp11 via its coiled-coil domain and specifically promotes its K48-linked ubiquitina tion and subsequent proteasomal degradation. TRIM29-mediated degradation of nsp11 counteracts nsp11's suppression of interferon (IFN-β) and interferon-stimulated gene production. Consequently, TRIM29 significantly inhibits PRRSV replication. Collectively, these findings uncover a conserved UPS-mediated regulatory mechanism targeting a critical arteriviral endonuclease and demonstrate TRIM29 as a potent host restriction factor that antagonizes PRRSV immune evasion by degrading nsp11. IMPORTANCE This study reveals that porcine reproductive and respiratory syndrome virus (PRRSV) nsp11 undergoes K48-linked polyubiquitination at catalytic residue K173, triggering ubiquitin-proteasome system (UPS)-mediated degradation, a mechanism conserved in most arteriviruses. The host E3 ligase TRIM29 binds nsp11 via its coiledcoil domain, catalyzing this ubiquitination to degrade nsp11. This counteracts nsp11's suppression of interferon (IFN-β)/interferon-stimulated gene production and inhibits PRRSV replication. These findings identify TRIM29 as a key host restriction factor that disrupts viral immune evasion by targeting a conserved arteriviral endonuclease via the UPS.
in annual economic losses exceeding $2.5 billion worldwide (6,7). While vaccination remains the primary control strategy, the remarkable genetic plasticity of PRRSV, particularly among circulating strains in China, has rendered most commercial vaccines ineffective due to rapid viral evolution and immune escape mechanisms (8,9).
PRRSV possesses a 15.4 kb single-stranded RNA genome containing at least 10 open reading frames (ORFs) (10). The 5′-proximal ORF1a/ORF1b region spans approximately 80% of the viral genome and encodes two large replicase polyproteins (pp1a and pp1ab) that undergo extensive proteolytic processing (11). The pp1ab polyprotein gives rise to four nonstructural proteins (nsp9-12) that form the core viral replication-transcrip tion complex (12). Among these, nsp11 is the only viral factor possessing endoribonu clease activity, a function mediated by its C-terminal EndoU domain. This domain is conserved across nidoviruses and shares distant homology with the Xenopus laevis XendoU ribonuclease (13,14). Structural elucidation reveals that PRRSV nsp11 adopts an unusual asymmetric dimeric configuration, contrasting sharply with the hexameric architecture observed in coronavirus nsp15 homologs (15). The enzyme's uridylate-spe cific cleavage activity resides in subdomain A, while subdomain B maintains structural integrity and potentially mediates regulatory interactions (16). PRRSV is an immunosup pressive pathogen that weakens innate immune responses, particularly by suppressing type I interferons (IFNs) (17,18). A key factor in this immune evasion is NSP11, which possesses nidovirus-like NendoU activity critical for viral replication and host innate immune suppression (19).
Ubiquitination represents a fundamental post-translational modification that orchestrates virtually all aspects of cellular physiology through dynamic protein regulation (20). This versatile modification system achieves its remarkable functional diversity through ubiquitin, a highly conserved 76-amino acid protein containing seven lysine residues (K6, K11, K27, K29, K33, K48, and K63), which forms topologically distinct polymeric chains (21). Among these, K48-linked polyubiquitin chains serve as the canonical signal for proteasomal degradation, while K63-linked chains typically modulate protein-protein interactions and subcellular trafficking (22). The ubiquitination cascade is mediated by a hierarchical enzymatic system comprising E1 activating enzymes, E2 conjugating enzymes, and E3 ligases, with the latter conferring substrate specificity through direct recognition of target proteins (23). Notably, the ubiquitin-proteasome system (UPS) has emerged as a critical battlefield in host-pathogen interactions, with numerous studies demonstrating how viral pathogens exploit or are restricted by host ubiquitination machinery during infection (24).
Functioning as cell-intrinsic antiviral effectors, several TRIM proteins employ diverse strategies to inhibit virus replication. A common mechanism involves their E3 ligase activity, which directs viral proteins for degradation via the ubiquitin-proteasome pathway (25,26). A key example is TRIM5, which not only targets the tick-borne encephalitis virus protease for ubiquitin-dependent degradation but also restricts primate retroviruses through a proteasome-independent mechanism by binding viral capsids (27). Beyond TRIM5, other family members mediate the degradation of viral proteins from a broad spectrum of viruses, including influenza A virus and enteroviruses (28,29). TRIM29 belongs to the TRIM family of proteins that function as crucial regulators of diverse cellular processes ranging from cell cycle progression to innate immunity (30,31). Unlike most TRIM proteins that possess intrinsic E3 ubiquitin ligase activity through their RING domains, TRIM29 has evolved alternative regulatory mechanisms despite lacking this canonical domain (32). Emerging evidence positions TRIM29 as a key modulator of protein turnover, capable of directing the Lys48-linked polyubiquitina tion and subsequent proteasomal degradation of critical immune regulators, including NEMO and STING (33). This activity effectively suppresses type I interferon responses and inflammatory cytokine production, establishing TRIM29 as an important immune checkpoint molecule. However, its potential role in antiviral defense against RNA viruses, particularly in the context of arterivirus infection, remains completely unexplored.
Our investigation reveals a novel host antiviral mechanism wherein the arteriviral endoribonuclease nsp11 undergoes K48-linked polyubiquitination at a critical catalytic residue, targeting it for proteasomal degradation. Through systematic biochemical and virological analyses, we identify TRIM29 as the E3 ubiquitin ligase responsible for this post-translational modification of PRRSV nsp11. Functional studies demonstrate that TRIM29-mediated degradation of nsp11 substantially attenuates viral replication kinetics. Collectively, our work provides fundamental insights into the intricate host-virus interplay governing PRRSV pathogenesis while revealing potential therapeutic targets for intervening in arterivirus infections.
## RESULTS
## PRRSV nsp11 undergoes ubiquitination during viral infection
Ubiquitination plays a pivotal role in viral infections, with many viruses exploiting this post-translational modification to enhance replication or evade host immune responses (34,35). While ubiquitination of viral enzymes and cofactors can promote genome replication, host cells also utilize ubiquitination to degrade viral proteins or amplify innate immune defenses. During PRRSV infection, viral RNA synthesis relies on the precise assembly of replication and transcription complexes (RTCs). To determine whether PRRSV RTC components undergo ubiquitination, we co-expressed core RTC proteins with ubiquitin. Notably, we observed robust ubiquitination of nsp11 (Fig. 1A andB). Consistent with a previous study, we also found that PRRSV nsp12 underwent ubiquitination (data not shown). To validate these findings in the context of infection, we performed ubiquitination assays in PRRSV-infected MARC-145 and porcine alveolar macrophage (PAM) cells, with or without the proteasome inhibitor MG132. nsp11, migrating at high molecular weight, was covalently linked to polyubiquitin chains in PRRSV-infected cells, which was further enhanced upon MG132 treatment (Fig. 1C andD), suggesting that ubiquitination occurs during viral replication. Collectively, our results demonstrate that PRRSV nsp11 undergoes ubiquitination, highlighting its potential role in viral replication or host-pathogen interactions.
## PRRSV nsp11 is modified by K48-linked ubiquitin chain at K173
To map the ubiquitination sites on PRRSV nsp11, we enriched ubiquitinated proteins by immunoprecipitating ectopically expressed nsp11 in the presence of MG132 and subjected the samples to trypsin digestion followed by mass spectrometry (MS). This approach revealed three nsp11-derived peptides (K157, K170, and K173) carrying diglycine modifications, a hallmark of ubiquitinated lysine residues post-trypsin diges tion (Fig. 2A andB). To further verify these ubiquitination sites, we constructed different mutants of nsp11 bearing single Lys (K)-to-Arg (R) substitutions and performed immuno precipitation assays. As expected, multiple ubiquitin chains were detected in the nsp11 immunoprecipitates (Fig. 2C). Remarkably, among the three single K-to-R mutants, only the K173R mutant completely lost ubiquitination signals (Fig. 2C), demonstrating that lysine 173 is the indispensable site for nsp11 ubiquitination. The K173 site is one of the endoribonuclease active sites in PRRSV nsp11 and is essential for viral replication (16).
Polyubiquitination can be mediated through any of ubiquitin's seven lysine residues, generating structurally distinct ubiquitin chains that determine the functional outcome for modified proteins. To define the chain linkage specificity of nsp11 ubiquitination, we performed in vitro ubiquitination assays using a panel of ubiquitin mutants where all but one lysine residue were mutated (K6, K11, K27, K29, K33, K48, or K63). Notably, nsp11 ubiquitination was exclusively observed when using the K48-linked ubiquitin mutant (Fig. 3A). To assess the biological significance of nsp11 ubiquitination during PRRSV infection, we engineered a recombinant PRRSV containing the nsp11-K173R mutation via reverse genetics (Fig. 3B). While wild-type virus produced characteristic cytopathic effects (CPEs) upon serial passage, the K173R mutant virus failed to induce detectable CPEs after three passages (Fig. 3C) and no K173R mutant virus can be detected using TCID 50 assay (Fig. 3D), demonstrating that the K173R mutation is lethal for viral recovery owing to nsp11's deficient endoribonuclease activity, which is consistent with previous study (36).
## Most arterivirus nsp11 can be ubiquitinated at its endonuclease activity site
Despite low overall sequence conservation among arteriviral nsp11 proteins, the K173 residue is strictly conserved and forms an essential part of the endonuclease active site (Fig. 4A andB). Based on this structural constraint, we examined the conservation of nsp11 ubiquitination across arteriviruses and its potential specificity for the K173 residue. Intriguingly, ubiquitination was detected in all tested nsp11 homologs (Fig. 4C through F). In PRRSV-1, SHFV, and EAV, mutagenesis analyses confirmed that ubiquitination occurs at the catalytic residue within the endonuclease active site, precisely recapitulat ing the K173 modification observed in arteriviruses. Notably, LDV nsp11 constituted an exception, with ubiquitination mapped to a distinct residue (Fig. 4C through F), suggest ing potential lineage-specific divergence in regulatory mechanisms. Together, these findings uncover a remarkable evolutionary conservation of nsp11 ubiquitination among most arteriviruses, with modification predominantly targeting catalytic residues critical for endonuclease function.
## Arterivirus nsp11 undergoes ubiquitin-proteasome-mediated degradation
Given that PRRSV nsp11 is predominantly modified by K48-linked polyubiquitin chains that are typically associated with protein stability, we sought to systematically investi gate whether the turnover of this viral protein is mechanistically regulated through ubiquitin-dependent mechanisms. To definitively map the degradation pathway, we employed pharmacological dissection using chloroquine (CQ) to inhibit autophagic flux and MG132 to block proteasomal activity. As shown in Fig. 5A through D, MG132 treatment caused dramatic accumulation of all arterivirus nsp11, whereas CQ had no measurable effect on protein levels. The ubiquitination-site mutants (with the noted exception of LDV nsp11) showed minimal response to either inhibitor. These results indicated that nsp11 was degraded by the ubiquitin-proteasome system. We also employed a cycloheximide (CHX)-based half-life assay to investigate the involvement of the ubiquitin-proteasome system in the degradation of nsp11. Quantitative immuno blotting revealed striking differential stability profiles: while wild-type nsp11 exhibited progressive degradation with a half-life consistent with active proteasomal targeting, the K173R mutant demonstrated remarkable resistance to decay, maintaining near-complete stability throughout the experimental timeframe (Fig. 5E). Notably, because residue K173 is involved in both ubiquitination and the protein's endonucleolytic activity, which may confound stability assays, we designed complementary experiments using two addi tional mutants: an endonuclease-dead variant (nsp11-H129A) and a double mutant with both enzymatic inactivation and ubiquitination deficiency (nsp11-H129A-K173R). This orthogonal approach yielded results fully congruent with our initial observations: the H129A mutant displayed degradation kinetics indistinguishable from wild-type nsp11, whereas the H129A-K173R double mutant recapitulated the stabilization phenotype observed with the single K173R mutation (Fig. 5F), thereby conclusively excluding the effect of nsp11 endonuclease activity on its stability. To establish whether this regulatory paradigm extends across the arterivirus genus, we performed comparative stability analyses on nsp11 homologs from related viruses. Intriguingly, whereas all tested arterivirus nsp11 proteins were degraded in a proteasome-sensitive manner and their ubiquitination-site mutants were stabilized, LDV nsp11 proved an exception: its degrada tion pattern suggested reliance on a different ubiquitination site (Fig. 5G through I).
Collectively, these multifaceted experiments establish that arterivirus nsp11 is selectively targeted for destruction via the ubiquitin-proteasome system, while LDV appears to have evolved distinct regulatory mechanisms.
## TRIM29 induces proteasomal degradation of PPRRSV nsp11 by promoting its K48-linked ubiquitination
MS-based proteomic analysis revealed two E3 ubiquitin ligases, TRIM29 and LTN1, as potential interactors of PRRSV nsp11 (Fig. 6A). To validate these findings, co-immunopre cipitation (Co-IP) assays were performed, confirming a robust interaction between TRIM29 and nsp11 (Fig. 6A andB). Given the critical role of E3 ligases in regulating protein stability through ubiquitination, we next investigated whether TRIM29 modu lates the ubiquitination status of nsp11. Strikingly, upon TRIM29 overexpression in the presence of the proteasome inhibitor MG132, we observed a pronounced increase in nsp11 ubiquitination (Fig. 6C), suggesting that TRIM29 mediates the polyubiquitination of nsp11, thereby targeting it for proteasomal degradation. Consistent with this hypothe sis, TRIM29 overexpression led to a dose-dependent reduction in nsp11 protein levels (Fig. 6D andE), an effect that was abolished upon MG132 treatment, further supporting the notion that TRIM29 promotes nsp11 degradation via the ubiquitin-proteasome system. To illustrate the molecular determinants governing the TRIM29-nsp11 interac tion, we systematically analyzed the binding capacity of nsp11 to various TRIM29 truncation mutants. Intriguingly, both recombinant full-length TRIM29 and its truncation mutants, except for ΔCC, C, or ΔC domain of TRIM29, interacted with nsp11, implicating the coiled-coil (CC) domain of TRIM29 as the critical region mediating this association (Fig. 6F andG). Conversely, reciprocal mapping using nsp11 truncation mutants demon strated that both full-length nsp11 and its C-terminal domain were sufficient for TRIM29 binding (Fig. 6I). Furthermore, we assessed the potential interaction of TRIM29 with nsp11 proteins from other arteriviruses; however, no such interactions were detected (data not shown). Thus, even though arterivirus nsp11 is degraded by the host ubiquitinproteasome system, the mechanisms are different. Collectively, these findings establish TRIM29 as a key regulator of nsp11 stability, driving its K48-linked polyubiquitination and subsequent proteasomal degradation.
## TRIM29 inhibits PRRSV replication
Previous studies have reported that nsp11 serves as a major IFN antagonist of PRRSV (19,37). Given this finding, we sought to determine whether TRIM29 modulates nsp11mediated suppression of IFN production. HEK293T cells were transfected with an IFN promoter-driven firefly luciferase reporter plasmid, along with nsp11 and TRIM29 (where applicable), while MAVS was employed as a strong inducer of IFN production. Consistent with prior reports, nsp11 overexpression significantly suppressed IFN-β production, but this inhibitory effect was markedly counteracted by TRIM29 co-expression (Fig. 7A). To further assess the impact of TRIM29 on IFN signaling, we measured the mRNA levels of IFN-β, ISG15, and ISG56 in HEK293T cells co-transfected with MAVS and either nsp11 alone or nsp11 plus TRIM29. Notably, TRIM29 promoted nsp11 degradation, leading to higher mRNA levels of IFN-β and interferon-stimulated genes (ISGs) compared to cells expressing nsp11 alone (Fig. 7B through D). Moreover, PRRSV infection in TRIM29overexpressing cells significantly upregulated antiviral cytokines, including IFN-β, ISG15, and ISG56 (Fig. 7E). These findings suggest that TRIM29 antagonizes nsp11-mediated IFN suppression by enhancing its degradation.
To explore TRIM29's role in PRRSV replication, we examined its effects on viral protein expression and infectivity. Overexpression of TRIM29 significantly reduced viral N protein levels and virus titers (Fig. 7F andG). To confirm these observations, vector-expressing or TRIM29-expressing cells were infected with PRRSV-GFP for 36 h. Compared to control cells, which exhibited strong fluorescence, TRIM29-expressing cells displayed substantially weaker GFP signals, further supporting its antiviral role (Fig. 7H andI). Conversely, small interfering RNA (siRNA)-mediated knockdown of TRIM29 in PAM cells enhanced N protein expression and viral titers at 16 h post-infection (Fig. 7J andK), reinforcing the notion that TRIM29 restricts PRRSV replication. To determine whether (G) HEK293T cells were transfected with TRIM29 mutants and nsp11 for 30 h, and then the cells were collected for immunoprecipitation, followed by western blot analysis of both total lysates and immunoprecipitated samples. (H) Schematic representation of nsp11 truncation mutants. (I) HEK293T cells were transfected with nsp11 mutants and TRIM29 for 30 h, and then the cells were collected for immunoprecipitation, followed by western blot analysis of both total lysates and immunoprecipitated samples. PRRSV infection influences TRIM29 expression, we infected MARC-145 cells with PRRSV (MOI = 0.1) and analyzed TRIM29 levels by Western blotting at various time points. However, PRRSV infection did not significantly alter TRIM29 expression (Fig. 7L).
## DISCUSSION
Ubiquitination, a pivotal post-translational modification, regulates critical cellular processes including protein homeostasis, innate immunity, and antiviral defense (38).
To evade immune surveillance, viruses encode specialized proteins that exploit the host ubiquitination machinery. These viral effectors selectively target host antiviral defenses, such as restriction factors and key signaling molecules (e.g., MAVS and NF-κB), ultimately suppressing innate immune activation (39,40). Mechanistically, certain viral proteins achieve immune evasion by either removing ubiquitin modifications from RIG-I and STING or blocking their interactions with cognate E3 ligases, thereby impairing their activation (41)(42)(43). Conversely, the host counteracts viral infection by mobilizing E3 ubiquitin ligases to mediate proteasomal degradation of viral proteins through the UPS, effectively attenuating viral virulence (21,44,45). The present study uncovers a previously unrecognized antiviral mechanism wherein the host E3 ubiquitin ligase TRIM29 selectively targets the conserved endoribonuclease nsp11 of PRRSV for K48-linked polyubiquitination and subsequent proteasomal degradation. Our findings provide key mechanistic insights into how host cells restrict arteriviral replication by exploiting the vulnerability of a catalytically essential residue within a critical viral enzyme. This work highlights a unique interface between host ubiquitination machinery and viral enzymatic function, offering conceptual and therapeutic implications for the broader control of arteriviral infections.
A central discovery of this study is that PRRSV nsp11 is ubiquitinated at lysine 173 (K173), a residue indispensable for its endoribonuclease activity (Fig. 2C and4B). Notably, nsp11 K173 site is not only sufficient to trigger proteasomal degradation (Fig. 5A andF) but also structurally coupled to enzymatic inactivation (37,46). These findings align with emerging evidence that host cells recognize and destabilize conserved viral enzymatic motifs as a defense strategy. The selective pressure imposed by host ubiquitination at K173 may explain the high conservation of this residue across PRRSV strains. Further more, our evolutionary analyses show that this regulatory mechanism is conserved across multiple arterivirus species, including SHFV and EAV, despite sequence divergence (Fig. 4C andD). However, LDV is a notable exception, as it appears to use an alternative degradation strategy (Fig. 4E). This conserved targeting of the catalytic site suggests an evolutionary "Achilles' heel" that the host exploits to restrict a broad spectrum of arteriviruses. Mechanistically, we identify TRIM29 as the host factor responsible for directing the K48-linked polyubiquitination of PRRSV nsp11. Our data show that TRIM29 binds nsp11 via its CC domain, a region involved in substrate recognition for immune regulators like STING and NEMO. This finding underscores TRIM29's broader role in innate immune regulation. The specificity of TRIM29 for PRRSV nsp11, and not for nsp11 homologs from other arteriviruses, suggests that TRIM29-substrate recognition is likely determined by virus-specific structural or contextual motifs within the C-terminal region of nsp11.
Importantly, we show that TRIM29-mediated degradation of nsp11 reactivates the antiviral IFN pathway by relieving nsp11-mediated suppression of IFN-β and ISGs (Fig. 7A through E). Given that nsp11 has been previously established as a potent antagonist of MAVS-dependent signaling (36), its removal by TRIM29 shifts the balance in favor of host defense. Indeed, overexpression of TRIM29 leads to significant reductions in viral protein synthesis and virus titers, and TRIM29 knockdown enhances PRRSV replication (Fig. 7F through J).
From a broader perspective, our results establish a precedent for host-targeted degradation of viral enzymes via modification of their catalytic cores, which is an elegant strategy that circumvents the need for host recognition of structurally variable epitopes. The fact that TRIM29 modifies a functionally irreplaceable lysine within nsp11's active site provides a compelling example of post-translational regulation functioning as a molecular switch for viral attenuation. This discovery not only expands our under standing of host-pathogen conflict at the interface of ubiquitin biology and RNA virus replication but also suggests novel antiviral strategies. Small molecules or gene therapy approaches that mimic or enhance TRIM29 activity could represent promising therapeu tic avenues against PRRSV.
Finally, our findings raise intriguing questions about viral countermeasures. It remains to be determined whether PRRSV encodes antagonists that neutralize TRIM29 or otherwise prevent nsp11 degradation. While our data indicate that PRRSV infection does not modulate TRIM29 expression at the protein level, post-translational modifications or spatial sequestration could still underlie potential evasion strategies. Future investiga tions are warranted to explore the dynamic interplay between TRIM29, viral proteins, and other cellular quality control pathways.
In summary, this work defines a previously unappreciated host-driven restriction mechanism targeting the catalytic core of a key viral enzyme, mediated by TRIM29dependent ubiquitination. It sheds light on a conserved vulnerability in arterivirus replication machinery and opens new avenues for host-directed antiviral interventions.
## MATERIALS AND METHODS
## Cells and virus
HEK-293T cells (ATCC CRL-11268) and MARC-145 cells (African green monkey kidney cells; ATCC CRL-12231) were maintained in Dulbecco's modified Eagle's medium (HyClone, SH30022.01) supplemented with 10% fetal bovine serum (Lonsa Science, S711-001S) and 1% penicillin-streptomycin (Solarbio, P1400). PAMs and the PRRSV strain were preserved in our laboratory. The PRRSV-EGFP recombinant virus, which stably expresses enhanced green fluorescent protein (EGFP), was generously provided by Dr. Nanhua Chen (Yangzhou University).
## Plasmids, antibodies, and reagents
Plasmids encoding PRRSV nsp11 and its mutant variants were constructed and stored in our laboratory. TRIM29 was amplified from MARC-145 cell-derived cDNA using genespecific primers, and site-directed mutagenesis was performed to generate TRIM29 mutants. All constructs, including wild-type and mutant TRIM29, were cloned into the pCDNA3.1-3Flag vector (Miaolingbio, P0157) using T4 DNA ligase (Promega, M1801). Additionally, nsp11 genes from SHFV, LDV, and EAV, along with their respective mutants, were synthesized by Genewiz and inserted into the same vector backbone.
Primary antibodies, including mouse anti-HA (66006-2-Ig), rabbit anti-HA (51064-2-AP), mouse anti-Flag (66008-3-Ig), and mouse anti-α-tubulin (6031-1-Ig), were purchased from Proteintech. Secondary antibodies conjugated to Alexa Fluor 555 (goat anti-mouse, Invitrogen, A32727) and Alexa Fluor 488 (goat anti-rabbit, A32731) were purchased from Invitrogen. Rabbit anti-TRIM29 polyclonal antibody (Abclonal, A5476) was used for TRIM29 detection. The anti-PRRSV N monoclonal antibody was developed and preserved in our laboratory.
Chemical reagents included chloroquine (autophagy inhibitor; Beyotime, C1202) and MG132 (proteasome inhibitor; Selleck, S2619).
## Plasmid transfection
HEK293T and MARC-145 cells were transiently transfected with plasmid DNA using JetPRIME transfection reagent (Polyplus, PT-114-15), according to the manufacturer's instructions.
## RNA interference
siRNAs targeting specific genes were designed and synthesized by GenePharma. PAM cells were transfected with siRNAs at a final concentration of 10 nM using Lipofectamine 2000 (Invitrogen, 11668019), following the manufacturer's protocol. Cells were infected with PRRSV 36 h post-transfection.
## RT-qPCR analysis
Total RNA was extracted using TRIzol reagent (Invitrogen) and reverse-transcribed into cDNA using the HiScript III 1st Strand cDNA Synthesis Kit (Vazyme, R312). Quantitative PCR was performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711). Relative gene expression levels were calculated using the 2 -ΔΔCt method.
## Dual-luciferase reporter assay
HEK293T cells were seeded in 24-well plates 24 h prior to transfection. Cells were co-transfected with an IFN-β luciferase reporter plasmid and pTK-Renilla, along with the indicated expression plasmids. An empty vector was included as a negative control to equalize the total plasmid DNA amount. Luciferase activities were measured using a Dual-Luciferase Reporter Assay System (Beyotime, RG027) according to the manufactur er's protocol. Each experiment was performed in triplicate. Results are presented as mean ± standard deviation.
## Western blotting, Co-IP, and LC-MS/MS
Cells were lysed in lysis buffer and incubated on ice for 30 min with intermittent vortexing. Lysates were clarified by centrifugation at 12,000 rpm for 10 min. Proteins were separated by SDS-PAGE and transferred onto PVDF membranes (Roche, 46978100). Membranes were blocked in 5% skim milk and incubated with primary antibodies for 1 h at room temperature, followed by secondary antibody incubation.
For Co-IP analysis, cell lysates were incubated overnight at 4°C with specific antibod ies, then mixed with Protein A+G agarose beads (Beyotime, P2012) and further incubated at 4°C for 3 h. The resulting complexes were collected by centrifugation at 1,000 × g for 3 min, washed five times with cold lysis buffer, and analyzed by Western blotting.
For LC-MS/MS analysis, immunoprecipitated complexes were resolved via SDS-PAGE, followed by silver staining to visualize differentially expressed protein bands between experimental and control groups. Selected bands were excised and subjected to LC-MS/MS by APTBIO (Shanghai, China) to identify nsp11-interacting proteins.
## Statistical analysis
All data were analyzed using GraphPad Prism 9. Each experiment was independently repeated at least three times. Statistical comparisons between two groups were performed using a two-tailed Student's t-test. A P value <0.05 was considered statistically significant (*P < 0.05, **P < 0.01, and ***P < 0.001).
## References
1. Han, Yoo (2014) "Engineering the PRRS virus genome: updates and perspectives" *Vet Microbiol*
2. Music, Gagnon (2010) "The role of porcine reproductive and respiratory syndrome (PRRS) virus structural and non-structural proteins in virus pathogenesis" *Anim Health Res Rev*
3. Zhong, Tan, Liu (2012) "Recent progress in studies of arterivirusand coronavirus-host interactions" *Viruses*
4. Snijder, Kikkert, Fang (2013) "Arterivirus molecular biology and pathogenesis" *J Gen Virol*
5. Shaw, Huey, Mousa-Makky et al. (2024) "The neonatal Fc receptor (FcRn) is a pan-arterivirus receptor" *Nat Commun*
6. Han, Zhou, Ge et al. (2017) "Pathogenesis and control of the Chinese highly pathogenic porcine reproductive and respiratory syndrome virus" *Vet Microbiol*
7. Chai, Liu, Sun et al. (2020) "Efficacy of two porcine reproductive and respiratory syndrome (PRRS) modified-live virus (MLV) vaccines against heterologous NADC30-like PRRS virus challenge" *Vet Microbiol*
8. Bai, Wang, Xu et al. (2016) "Commercial vaccines provide limited protection to NADC30-like PRRSV infection" *Vaccine (Auckl)*
9. Chen, Zhou, Guo et al. (2021) "Efficacy of a live attenuated highly pathogenic PRRSV vaccine against a NADC30-like strain challenge: implications for ADE of PRRSV" *BMC Vet Res*
10. Gao, Guo, Yang (2004) "Genomic characterization of two Chinese isolates of porcine respiratory and reproductive syndrome virus" *Arch Virol*
11. Dokland (2010) "The structural biology of PRRSV" *Virus Res*
12. Fang, Snijder (2010) "The PRRSV replicase: exploring the multifunc tionality of an intriguing set of nonstructural proteins" *Virus Res*
13. Ivanov, Hertzig, Rozanov et al. (2004) "Major genetic marker of nidoviruses encodes a replicative endoribonuclease" *Proc Natl Acad Sci*
14. Laneve, Altieri, Fiori et al. (2003) "Purification, cloning, and characterization of XendoU, a novel endoribo nuclease involved in processing of intron-encoded small nucleolar RNAs in Xenopus laevis" *J Biol Chem*
15. Ricagno, Egloff, Ulferts et al. (2006) "Crystal structure and mechanis tic determinants of SARS coronavirus nonstructural protein 15 define an endoribonuclease family" *Proc Natl Acad Sci*
16. Shi, Li, Lei et al. (2016) "A dimerization-dependent mechanism drives the endoribonu clease function of porcine reproductive and respiratory syndrome virus nsp11" *J Virol*
17. Wang, Zhang (2014) "Antagonizing interferon-mediated immune response by porcine reproductive and respiratory syndrome virus" *Biomed Res Int*
18. Yang, Zhang (2017) "Antagonizing cytokine-mediated JAK-STAT signaling by porcine reproductive and respiratory syndrome virus" *Vet Microbiol*
19. Wang, Chen, Yu et al. (2019) "Porcine reproductive and respiratory syndrome virus nsp11 antagonizes Type I interferon signaling by targeting IRF9" *J Virol*
20. Komander (2009) "The emerging complexity of protein ubiquitination" *Biochem Soc Trans*
21. Li, Hao, Zhao et al. (2023) "The E3 ligase RNF5 restricts SARS-CoV-2 replication by targeting its envelope protein for degradation" *Sig Transduct Target Ther*
22. Castañeda, Chaturvedi, Camara et al. (2016) "Linkage-specific conformational ensembles of non-canonical polyubiquitin chains" *Phys Chem Chem Phys*
23. Buetow, Huang (2016) "Structural insights into the catalysis and regulation of E3 ubiquitin ligases" *Nat Rev Mol Cell Biol*
24. Velez-Brochero, Behera, Afreen et al. (2024) "Ubiquitination in viral entry and replication: mechanisms and implications" *Adv Virus Res*
25. Van Gent, Sparrer, Gack (2018) "TRIM proteins and their roles in antiviral host defenses" *Annu Rev Virol*
26. Hage, Rajsbaum (2019) "To TRIM or not to TRIM: the balance of hostvirus interactions mediated by the ubiquitin system" *J Gen Virol*
27. Ganser-Pornillos, Pornillos (2019) "Restriction of HIV-1 and other retroviruses by TRIM5" *Nat Rev Microbiol*
28. Fan, Mar, Sari et al. (2021) "TRIM7 inhibits enterovirus replication and promotes emergence of a viral variant with increased pathogenicity" *Cell*
29. (2025) *Full-Length Text Journal of Virology*
30. Lin, Wang, Chen et al. (2023) "TRIM21 restricts influenza A virus replication by ubiquitinationdependent degradation of M1" *PLoS Pathog*
31. Xing, Zhang, Zhang et al. (2017) "TRIM29 promotes DNA virus infections by inhibiting innate immune response" *Nat Commun*
32. Wu, Nandi, Sharma (2024) "TRIM-endous functional network of tripartite motif 29 (TRIM29) in cancer progression and beyond" *Cancer Metastasis Rev*
33. Li, Lin, Tong et al. (2018) "TRIM29 negatively controls antiviral immune response through targeting STING for degradation" *Cell Discov*
34. Xing, Weng, Yuan et al. (2016) "Identification of a role for TRIM29 in the control of innate immunity in the respiratory tract" *Nat Immunol*
35. Nisole, Stoye, Saïb (2005) "TRIM family proteins: retroviral restriction and antiviral defence" *Nat Rev Microbiol*
36. Randow, Lehner (2009) "Viral avoidance and exploitation of the ubiquitin system" *Nat Cell Biol*
37. Sun, Ke, Han et al. (2016) "Nonstructural protein 11 of porcine reproductive and respiratory syndrome virus suppresses both MAVS and RIG-I expression as one of the mechanisms to antagonize type i interferon production" *PLoS One*
38. Shi, Wang, Li et al. (2011) "Endoribonuclease activities of porcine reproductive and respiratory syndrome virus nsp11 was essential for nsp11 to inhibit IFN-β induction" *Mol Immunol*
39. Valerdi, Hage, Van Tol et al. (2021) "The role of the host ubiquitin system in promoting replication of emergent viruses" *Viruses*
40. Gu (2020) "Specificity in ubiquitination triggered by virus infection" *Int J Mol Sci*
41. Liu, Zhang, Chu et al. (2017) "The ubiquitin E3 ligase TRIM31 promotes aggregation and activation of the signaling adaptor MAVS through Lys63-linked polyubiquitination" *Nat Immunol*
42. Fan, Mao, Yu et al. (2014) "USP21 negatively regulates antiviral response by acting as a RIG-I deubiquitinase" *J Exp Med*
43. Zhou, Zheng, Liu et al. (2022) "UBR5 acts as an antiviral host factor against MERS-CoV via promoting ubiquitination and degradation of ORF4b" *J Virol*
44. Bodda, Reinert, Fruhwürth et al. (2020) "HSV1 VP1-2 deubiquitinates STING to block type I interferon expression and promote brain infection" *J Exp Med*
45. Yu, Chen, Hou et al. (2015) "VHL negatively regulates SARS coronavirus replication by modulating nsp16 ubiquitina tion and stability"
46. Wang, Dong, Luan et al. (2024) "TRIM56 restricts Coxsackievirus B infection by mediating the ubiquitination of viral RNA-dependent RNA polymerase 3D" *PLoS Pathog*
47. Dong, Zhou, Ge et al. (2018) "Porcine reproductive and respiratory syndrome virus nsp1β and nsp11 antagonize the antiviral activity of cholesterol-25-hydroxylase via lysosomal degrada tion" *Vet Microbiol*
48. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12778032&blobtype=pdf | # A comparative study on proinflammatory cytokines interleukin-17A and interleukin-17F expressions in whole blood of patients with COVID-19
Zeynab Rahni, Seyed Mohebbi, Seyed Hosseini, Shahrzad Shoraka, Kambiz Nabati, Mahsa Saeedi Niasar, Shabnam Shahrokh, Amir Sadeghi, Mohammad Zali
## Abstract
Background and Objectives: Coronavirus disease 2019 , due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has expanded rapidly to all over the world. Interleukin-17 is one of the inflammatory cytokines which is highly expressed in the blood of individuals with COVID-19. Our aim in the present survey was to assess the mRNA expression levels of cytokine IL-17A, IL-17F and TNF-α in the blood of COVID-19 patients compared with healthy control individuals. Materials and Methods: A total of 69 patients including 32 mild patients, 20 severe and 17 asymptomatic patients in comparison with 25 healthy controls were evaluated. To measure the expression profile of IL-17A and IL-17F in whole blood, quantitative PCR was used. Results: Asymptomatic, mild, and severe SARS-CoV-2 infections were found to have significantly higher levels of IL-17A and IL-17F mRNAs than the healthy group (fold change IL-17A: 3.778; p= 0.002, 4.003; p= 0.001, 2.608; p= 0.0001 respectively, fold change IL-17F: 2.967; p= 0.003, 3.819; p= 0.001, 2.617; p= 0.0012 respectively). TNF-α mRNA expression was also measured, which shows an approximately similar increase compared to IL-17 (fold change: 2.726; p= 0.002, 2.383; P= 0.001, 2.631; p= 0.001, respectively). Conclusion: SARS-CoV-2 infection severity was associated with blood levels of IL-17A and IL-17F mRNA.
## INTRODUCTION
A novel coronavirus known as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is the cause of coronavirus disease 2019 . The virus was initially identified in Wuhan, China, in December 2019 in the respiratory systems of pneumonia patients. SARS-CoV-2 is an unsegmented RNA virus with a genome size of 29.9 kb (1)(2)(3). Spike protein is a transmembrane protein found on the virus's outer surface, which helps the viral coat attach to the cell by binding to the human angiotensin-converting enzyme 2 (hACE2) (4)(5)(6). The disease advancement is more likely to affect those who have underlying issues, such as diabetes, heart disease, and high blood pressure (2). Apart from the less frequent symptoms like hemoptysis, angina, diarrhea, nausea, vomiting, and chest pain, COVID-19 patients frequently have pyrexia, coughing, exhaustion, headaches, shortness of breath, sore muscles, and sputum production (1)(2)(3).
A SARS-CoV-2 infection leads to a number of dysregulated inflammatory reactions by continuously interfering with the immune system's normal functions. Lymphopenia is the primary feature of COVID-19 in patients, particularly those with severe illness. TCD4+, TCD8+, NK, and B lymphocytes are significantly reduced in these individuals (1,7,8). Another feature of severe COVID-19 can be considered as increased cytokine production; these patients experience a significant rise in inflammatory cytokines, including IL-6, IL-1β, TNF-α, IL-10, MCP-3, IP-10, GM-CSF, IL-17, and IL-1Ra, that is suggestive of a cytokine storm (1,2,(9)(10)(11)(12).
IL-17A, IL-17B, IL-17C, IL-17D, IL-17E, and IL-17F are the cytokines that make up IL-17. The biological activities along with adaptation of IL-17F and IL-17A are among the most well-known of the cytokines (13). T helper 17 cells make up the majority of the IL-17A and IL-17F found in adaptive responses (14). Moreover, these cytokines are made by other cells, including natural killer cells, macrophages, and innate Th17 lymphocytes (15). In order to eradicate pathogens such Citrobacter rodentium, Klebsiella pneumoniae, and Staphylococcus aureus, IL-17F and IL-17A should be considered necessary (16)(17)(18). Dysregulated secretion of IL-17F and IL-17A can possibly cause inflammation, tissue damage, and autoimmune illnesses including psoriasis, inflammatory bowel disease (13) rheumatoid arthritis (RA) and multiple sclerosis (MS) (19,20). GM-CSF, TNF-α, and IL-17 cytokine levels are elevated in patients with severe COVID-19, a condition known as the Th17-type cytokine storm, which may result to organ damage. Th17 cells stimulate Th17 responses by overexpressing TNF-α. In the context of viral diseases, IL-17 can be linked to IL-6, a predictor of COVID-19 severity. Indeed, elevated amounts of IL-6 have been shown to boost Th17 cell development in mouse viral models. IL-6 and IL-17 elevated secretion may protect infected cells from apoptosis, thereby increasing viral resistance (21). Furthermore, IL-17 excess levels can induce T-cell responses and expand the concentrations of inflammatory cytokines, including IL-1β, IL-6, and TNF-α (22). These factors have motivated researchers to explore the potential of IL-17 as a remedial option in the context of COVID-19, employing cytokine inhibitors (22)(23)(24)(25).
To assess the mRNA quantities of IL-17A, IL-17F, and TNF-α in the whole blood of SARS-CoV-2 patients compared to a normal control category, we conducted this study, as cytokines, particularly IL-17, play critical roles in predicting the severity and pathogenesis of viral infections like COVID-19.
## MATERIALS AND METHODS
Study design and sample collection. 32 mild patients (14 females and 18 males) along with 20 severe patients (5 females and 15 males) were admitted to Taleghani and Imam Hossein Hospitals for this survey. Additionally, 17 asymptomatic patients (8 females and 9 males) who were clinically diagnosed and real-time PCR SARS-CoV-2 positive throat swab samples were included, in comparison to 25 healthy subjects (7 females and 18 males). From March 25 to August 25, 2020, all sample groups were collected. The Research Institute's ethics committee accepted the protocols (IR.SBMU.RIGLD.REC.1399.008, Iran). Furthermore, all participants gave their informed consent.
Extraction and cDNA preparation. To extract total RNA from whole blood samples, the Hybrid-RTM blood RNA isolation reagent (GeneAll Biotechnology, South Korea) was utilized following the manufacturer's instructions. The extraction was performed from peripheral blood based on glass fiber membrane technology (26). cDNA was synthesized using the Thermo Scientific RevertAid Strand cDNA Syn-thesis Kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) (27). IL-17A and IL-17F expression levels in the whole blood of patients and a normal control category was measured employing the SYBR Green (RealQ plus 2x Master Mix Green, Ampliqon, Denmark) technique. The reference gene was the stably expressed reference gene in whole blood, the ß2-Microglobulin. The quantitative PCR utilized pertinent forward and reverse primers (Table 1). The following steps were used to run the quantitative PCR: 15 minutes at 95°C, 40 cycles of 60°C for 60 seconds, and 15 seconds at 95°C. For calculating relative expression, the 2-ΔΔCT method was utilized.
## Measurement of anti-SARS-CoV-2 IgG and IgM by ELISA assay. Anti-SARS-CoV-2 nucleocapsid
IgM and IgG antibodies were counted in the plasma samples of mild and severe patients, using BioTek Elx800 microplate reader according to manufacturer protocol (Highland park, Winooski, Vermont, USA).
## Statistical analysis. The statistical version of the
Social Science Software Package 16 (SPSS Inc., Illinois, USA) was utilized to conduct the statistical analysis. The data was analyzed employing one-way ANOVA, Tukey's post-hoc analysis, or the Kruskal-Wallis test, followed by Dunn's post-hoc comparisons. The parametric variables were compared utilizing either the independent sample t-test or the Mann-Whitney U test. Non-parametric variables were analyzed applying the χ2. The potential use of blood IL-17A and IL-17F levels as a diagnostic marker was analyzed employing area under the curve (AUC) and receiver operating characteristic (ROC) curve analysis. The appropriate cut-off was determined by maximizing Youden's index, defined as max [sensitivity(c) + specificity(c) -1], which provides an excellent diagnostic threshold for determining whether an indi-vidual is ill or healthy (28). A logistic regression test was conducted to remove the influence of age. Charts were plotted using GraphPad Prism 8.
## RESULTS
Demographic and clinical characteristics. This study includes data from 17 asymptomatic, 32 mild, and 20 severe COVID-19 patients, as well as 25 healthy controls. Table 2 shows the demographic information for the examined subjects. Regression analysis was used to adjust for the potential confounding effect of age (Table 2). When comparing moderate and severe patients by age, there was no significant relationship between cytokine production and age (after adjustment to remove the confounding effect). Ten days after infection, samples from patients with moderate and severe infections were taken. Table 3 shows the laboratory results, whereas Table 4 describes the clinical characteristics. According to Table 3, those with severe diseases showed elevated amounts of D-dimer, CRP, and neutrophil count, as well as lower O2 saturation and lymphocyte count.
A study of the patients' medical history found that 55.8% of them had underlying medical disorders, such as chronic heart disease (19.23%), diabetes mellitus (21.15%), chronic kidney disease (9.61%0), hypertension (25%) or cancer (25%).
There was no significant relationship between the examined mRNA levels and the clinical features as well as comorbidities of moderate and severe COVID-19 individuals. The illness severity was estimated using the COVID-19 clinical therapy guidelines (29). As a result, asymptomatic patients are those who have caught SARS-CoV-2 nevertheless have not developed signs at all. Individuals with SARS-CoV-2 who need hospitalization and have unique pneumonia Patients were classified as severe based on their history, which was verified by an infectious disease physician.
## Characteristics of plasma IgM and IgG antibodies in patients. ELISA was used to detect anti-SARS-
CoV-2 nucleocapsid (N) protein antibody tests for IgG as well as IgM antibodies in moderate and severe patients' blood samples. These findings demonstrated that the positive rates for IgM and IgG were 42.30% (22 of 52) and 63.46%, respectively (33 of 52).
## Blood mRNA quantity of IL-17A, IL-17F, and TNF-α in COVID-19 patients and healthy controls.
The expression rate of IL-17A, IL-17F and TNF-α remarkably elevated in asymptomatic, mild and severe patients in comparison to the normal category (Fold change IL-17A: 3.778; p= 0.002, 4.003; p= 0.001, 2.608; p= 0.0001 respectively), (Fold change IL-17F: 2.967; p= 0.003, 3.819; p= 0.001, 2.617; p= 0.0012 respectively), (Fold change TNF-α: 2.726; p= 0.002, 2.383; p= 0.001, 2.631; p= 0.001, respectively), and mRNA levels in severe, mild and asymptomatic patients did not differ significantly (Fold change IL-17A: 0.2249; p= 0.9867, 1.171; p= 0.3845, 1.396; p= 0.1295, Fold change IL-17F: 0.8523; p= 0.5920, 0.3506; p= 0.9653, 1.203; p= 0.2487, fold change TNF-α: 0.3434; p= 0.5673, 0.0950; p= 0.8854, 0.2488; p= 0.6629, respectively), (Fig. 1A-C). The expression of IL-17F and IL-17A mRNAs is approximately equal to that of TNF-α as indicator of COVID-19 severity.
## Diagnostic value of TNF-α, IL-17F, and IL-17A
in SARS CoV-2 infection. Using the ROC curve, the diagnostic specificity and sensitivity of Interleukin 17A, 17F, and TNF-α were assessed. The AUC of IL-17A was 0.8662 (95% CI: 0.7850 to 0.9473, p-value <0.0001), with the optimal cut-off value estimated to be less than 0.0786 (Sensitivity, 76%; Specificity, 84.62%) (Fig. 2A). The AUC of IL-17F for SARS-CoV-2 diagnosis in compared with controls was 0.8681 (95% CI: 0.7881 to 0.9480, p-value <0.0001). The optimal cut-off was determined to be less than 0.0742 (Sensitivity, 72%; Specificity, 90.38%) (Fig. 2B). The optimal cut-off was established to be less than 0.1453 (Sensitivity, 92%; Specificity, 65.38%), as indicated by the AUC for TNF-α of 0.8238 (95% CI: 0.8238, P-value < 0.0001) (Fig. 2C).
## DISCUSSION
The production and secretion of numerous proinflammatory cytokines, together with T cells triggering, including CD4+ and CD8+, are all examples of competent antiviral responses in innate and adaptive immunity. T cells has been considered indispensable for controlling of propagation, the restriction of viral dissemination, the elimination of infected cells, and the reduction of inflammation (30,31). Cytokines are a class of signaling molecules which are responsible for modulating a variety of biological processes through cell surface receptors (32). They are primarily released by immune cells, such as lymphocytes, monocytes, and macrophages. Organ failure, tissue injury, and ultimately mortality may be the effect of too much inflammatory cytokines production (33). The primary functions of CD4+ T helper cells in the tissue injury are associated with autoimmune or inflammatory diseases as follows (34). Th-17 is classified as an inflammatory T-helper among the subsets of T-helper cells, as it generates IL-17. This leads to chronic inflammation in tissues and subsequent organ failure (14).
The biological association of IL-17 produced by Th17 cells and their involvement in the pathogenesis of viral infections, including HBV, HCV, HIV, and Dengue virus, has been investigated. There has been a report of an elevation in IL-17 mRNA production in PBMC during HBV infection (35). Another study reported that IL-17 mRNA levels were substantially raised in HBV patients in comparison to healthy individuals (36, 37). Th17 cells are more abundant in the peripheral blood of persistently infected HCV patients during HCV infection (38,39). Moreover, HIV infection is linked to a low percentage of Th17 cells, and IL-17 quantities are positively correlated to a Th17 cell percentage, and both IL-17 and Th17 are negatively correlated with the plasma viral load (40,41). A high serum rate of IL-23, IL-22, IL-17F and IL-17A, which are marker cytokines primarily related to Th17 cells, is produced in Dengue virus infections compared with the healthy group (42). Additionally, patients with HCV had higher levels of IL-17 than healthy controls did (38,39).
Two mechanisms generate the immune system malfunction in COVID-19 patients: the excessive pro-inflammatory cytokines production by monocytes and the aberrant generation of lymphocytes by CD4+ cells (43,44). Severe lung tissue damage has been linked to cytokine storms or cytokine release syndrome (CRS), which are marked by an over-inflammatory response and cytokines and chemokines (IL-8, IL-6, IL-17, TNF-α, IFN-γ and G-CSF) secretion into the lungs (45). It has been reported that the fundamental cause of ARDS (acute respiratory distress syndrome) can be the cytokine storm. Many coronavirus-infected people develop ARDS, which affects the liver, kidneys, and heart in addition to producing pulmonary edema (7).
Recent data suggests that the pathophysiology of COVID-19 involves host Th17 inflammatory responses. These responses include the diffusion of important cytokines like GM-CSF and IL-17 as well as other immune-boosting factors that help fight off viral infection by decreasing Treg cell counts, increasing neutrophil migration, and simultaneously triggering Th2 responses (46).
In patients with severe COVID-19, an excess of CCR4+/CCR6+ Th17 cells in the blood may have potent proinflammatory effects and encourage Th17type cytokine storm (7). It has also been shown that patients with SARS-CoV, MERS-CoV, and other beta coronavirus members have elevated Th17 responses. According to recent research MERS-CoV and SARS-CoV infections increase host's expression of IL-17 (47)(48)(49).
A biological molecule in the blood or tissues that indicates a common or uncommon operation, circumstance, or illness is referred to as a biomarker, according to the National Cancer Institute. The body's reaction to a disease treatment may be evaluated using a biomarker (50). For instance, IL-17 may be a biomarker for liver transplant recipients who have liver injury (51,52). The DNA methylation measure is generally suitable for use as a biomarker for diseases verification (53). Research has shown that suppressing the hypermethylated IL-17 promoter may stop CHB from developing and spreading. Furthermore, these results suggest that methylation of the IL-17 promoter might be a biomarker for HBV-HCC (54). Additionally, Zuñiga, Joaquín, et al. have shown the potential use of IL-17 as a biomarker for acute ZIKV infection (55).
Confirming the gene expression of T-helper 17 pathway-related elements, like IL-17F, IL-17A, and TNF-α, in normal controls and severe COVID-19 patients, moderate and asymptomatic individuals was the aim of this study. Furthermore, we evaluated TNF-α, IL-17A, and IL-17F's potential as diagnostic biomarkers for the infection. We found that the patients had considerably higher mean mRNA quanti-ties of IL-17A, IL-17F, and TNF-α than the healthy category. Furthermore, whole blood quantities of TNF-α, IL-17A, and IL-17F may be circulating biomarkers for the infection, in line with our finding. It is important to note that the assessment of IL-17F and IL-17A mRNA quantities in the infection may be clinically complicated by other respiratory diseases, including rhinovirus, adenovirus, influenza viruses, and even bacterial lung infections. Our investigation also revealed that severe patients had greater levels of disease-related variables, such CRP and D-dimer, than mild patients. These findings are supported by the findings of previous studies (56). Additionally, our severe patients had significantly greater reduced O2 saturation and cell counts (57-59), which are indicators of infection severity, than our mild patients. According to earlier research, severe patients had higher neutrophil counts than mild patients, which is consistent with our findings (60,61).
The current study showed a significant elevation in IL-17F, IL-17A, and TNF-α levels between severe patients and the healthy group, despite the fact that many studies have reported a significant difference in pro-inflammatory cytokines between asymptomatic, severe and mild patients (62,63), However, according to the limitations of the research, there was no discernible difference between individuals who were asymptomatic and those who were severe. According to earlier research, inflammatory cytokines such IL-17 is increased and influential for resistance to the infection. In line with our findings, Sadeghi et al. reported that 40 COVID-19 patients had significantly over expressed plasma quantities of Th17-related cytokines like IL-23 in addition to IL-17, than the 40 healthy controls (64). According to a different research, moderate cases of SARS-CoV-2 may have higher blood concentrations of IL-17 than the severe and control groups (65). Furthermore, in 23 COVID-19 patients, De Biasi, Sara, et al. showed substantial increases in IL-17 and TNF-α plasma levels compared with 15 healthy individuals (66). Based on a research by Valizadeh et al., mRNA levels of TNF-α were enhanced compared to healthy patients (67), additionally, increased expression of TNF-α as an indicator of COVID-19 severity, has been supported by other studies (1,68), which corroborates our data. In earlier investigations, the LDH, CRP, and D-dimer as inflammatory variables and age were identified to be connected with severity of COVID-19 (69)(70)(71)(72). Tanacan, Atakan, et al. found a positive link between inflammatory variables such as CRP and IL17 expression levels (73), however, this link was not seen in the current investigation. Furthermore, comorbidities like diabetes and hypertension are linked to the probability of severity or mortality in the infected individuals (74, 75), however cytokine expression was not connected with these parameters in our investigation. Future research should include larger sample sizes that permit stronger subgroup analyses and prospective studies that carefully evaluate and document the presence and extent of underlying disease. The study's limitations included the limited number of samples, short sampling period, specific kind of therapy, and lack of tests for cytokine levels.
## CONCLUSION
SARS-CoV-2 patients had increased quantities of IL-17A, IL-17F, and TNF-α mRNAs. The survey reveals that IL-17F, IL-17A and TNF-α have appropriate sensitivity as well as specificity for evaluating the infected patients, and might work well as a viable biomarker for SARS-CoV-2 diagnosis. More study into interleukin-17 inhibitors is also suggested as a treatment option for lowering pulmonary inflammation and lung tissue damage.
## References
1. Iran, Microbiol (2025)
2. Huang, Wang, Li et al. (2020) "Clinical features of patients infected with 2019 novel coronavirus in Wuhan" *Lancet*
3. Chen, Zhou, Dong et al. (2019) "Epidemiological and clinical characteristics of 99 cases of"
4. (2020) "na: a descriptive study" *J. MICROBIOL*
5. Xu, Wu, Jiang et al. (2020) "Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series" *BMJ*
6. Rothan, Byrareddy (2020) "The epidemiology and pathogenesis of coronavirus disease (COVID-19) outbreak" *J Autoimmun*
7. Wan, Shang, Graham et al. (2020) "Receptor recognition by the novel coronavirus from Wuhan: an analysis based on decade-long structural studies of SARS coronavirus" *J Virol*
8. Blume, Jackson, Spalluto et al. (2021) "A novel ACE2 isoform is expressed in human respiratory epithelia and is upregulated in response to interferons and RNA respiratory virus infection" *Nat Genet*
9. Xu, Shi, Wang et al. (2020) "Pathological findings of COVID-19 associated with acute respiratory distress syndrome" *Lancet Respir Med*
10. Tan, Liu, Zhou et al. (2020) "Immunopathological characteristics of coronavirus disease 2019 cases in Guangzhou" *Immunology*
11. Liu, Zhang, Huang et al. (2020) "Elevated plasma levels of selective cytokines in COVID-19 patients reflect viral load and lung injury" *Natl Sci Rev*
12. Liu, Li, Liu et al. (2020) "Longitudinal characteristics of lymphocyte responses and cytokine profiles in the peripheral blood of SARS-CoV-2 infected patients" *EBioMedicine*
13. Wang, Jiang, Chen et al. (2020) "Cytokine storm and leukocyte changes in mild versus severe SARS-CoV-2 infection: Review of 3939 COVID-19 patients in China and emerging pathogenesis and therapy concepts" *J Leukoc Biol*
14. Rotondo, Martini, Maritati et al. (2021) "SARS-CoV-2 infection: new molecular, phylogenetic, and pathogenetic insights. Efficacy of current vaccines and the potential risk of variants" *Viruses*
15. Dong (2013) "IL-17 cytokines in immunity and inflammation" *Emerg Microbes Infect*
16. Harrington, Hatton, Mangan et al. (2005) "Interleukin 17-producing CD4+ effector T cells develop via a lineage distinct from the T helper type 1 and 2 lineages" *Nat Immunol*
17. Cua, Tato (2010) "Innate IL-17-producing cells: the sentinels of the immune system" *Nat Rev Immunol*
18. Ishigame, Kakuta, Nagai et al. (2009) "Differential roles of interleukin-17A and-17F in host defense against mucoepithelial bacterial infection and allergic responses" *Immunity*
19. Cho, Pietras, Garcia et al. (2010) "IL-17 is essential for host defense against cutaneous Staphylococcus aureus infection in mice" *J Clin Invest*
20. Aujla, Chan, Zheng et al. (2008) "IL-22 mediates mucosal host defense against Gram-negative bacterial pneumonia" *Nat Med*
21. Park, Li, Yang et al. (2005) "A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17" *Nat Immunol*
22. Waite, Skokos (2012) "Th17 response and inflammatory autoimmune diseases" *Int J Inflam*
23. Hou, Jin, Kang et al. (2014) "Interleukin-6 (IL-6) and IL-17 synergistically promote viral persistence by inhibiting cellular apoptosis and cytotoxic T cell function" *J Virol*
24. Pacha, Sallman, Evans (2020) "COVID-19: a case for inhibiting IL-17?" *Nat Rev Immunol*
25. Avdeev, Trushenko, Tsareva et al. (2021) "Anti-IL-17 monoclonal antibodies in hospitalized patients with severe COVID-19: A pilot study" *Cytokine*
26. Casillo, Mansour, Raucci et al. (2020) "Could IL-17 represent a new therapeutic target for the treatment and/or management of COVID-19-related respiratory syndrome?" *Pharmacol Res*
27. Megna, Napolitano, Fabbrocini (2020) "May IL-17 have a role in COVID-19 infection?" *Med Hypotheses*
28. Ghafouri-Fard, Oskooei, Omrani et al. (2019) "Dysregulation of cytokine coding genes in peripheral blood of bipolar patients" *J Affect Disord*
29. Yang, Ouyang, Gao et al. (2020) "Enriched high-throughput reverse transcription-quantitative PCR template preparation without pre-amplification" *Mol Med Rep*
30. Yin, Tian (2014) "Joint confidence region estimation for area under ROC curve and Youden index" *Stat Med*
31. (2020) "Clinical management of COVID-19: interim guidance"
32. Li, Fan, Lai et al. (2020) "Coronavirus infections and immune responses" *J Med Virol*
33. Tufan, Güler, Matucci-Cerinic (2020) "COVID-19, immune system response, hyperinflammation and repurposing antirheumatic drugs" *Turk J Med Sci*
34. Bartee, Mcfadden (2013) "Cytokine synergy: an underappreciated contributor to innate anti-viral immunity" *Cytokine*
35. Chen, Deng, Cui et al. (2017) "Inflammatory responses and inflammation-associated diseases in organs" *Oncotarget*
36. Gaffen, Jain, Garg et al. (2014) "IL-23-IL-17 immune axis: discovery, mechanistic understanding, and clinical testing" *Nat Rev Immunol*
37. Yang, Dai, Yan et al. (2018) "Expression of interleukin-17 is associated with different immune phases in patients with chronic hepatitis B" *Eur J Inflamm*
38. Du, Zhen, Zeng et al. (2013) "Expression of interleukin-17 associated with disease progression and liver fibrosis with hepatitis B virus infection: IL-17 in HBV infection" *Diagn Pathol*
39. Wang, Chen, Xu (2011) "IL-17 expression is correlated with hepatitis B-related liver diseases and fibrosis" *Int J Mol Med*
40. Chang, Wang, Zhao et al. (2012) "Th17 cells are increased with severity of liver inflammation in patients with chronic hepatitis C" *J Gastroenterol Hepatol*
41. Fathy, Ahmed, Metwally et al. (2011) "T helper type 1/T helper type 17-related cytokines in chronic hepatitis C patients before and after interferon and ribavirin therapy" *Med Princ Pract*
42. Alvarez, Tuen, Nàdas et al. (2012) "In vitro restoration of Th17 response during HIV infection with an antiretroviral drug and Th17 differentiation cytokines" *AIDS Res Hum Retroviruses*
43. Prendergast, Prado, Kang et al. (2010) "HIV-1 infection is characterized by profound depletion of CD161+ Th17 cells and gradual decline in regulatory T cells" *AIDS*
44. Sánchez-Vargas, Hernández-Flores, Thomas-Dupont et al. (2020) "Characterization of the IL-17 and CD4+ Th17 cells in the clinical course of dengue virus infections" *Viruses*
45. Netea, Rovina, Akinosoglou et al. (2020) "Complex immune dysregulation in COVID-19 patients with severe respiratory failure" *Cell Host Microbe*
46. Mahmoudi, Rezaei, Mansouri et al. (2020) "Immunologic features in coronavirus disease 2019: functional exhaustion of T cells and cytokine storm" *J Clin Immunol*
47. Mehta, Mcauley, Brown et al. (2020) "COVID-19: consider cytokine storm syndromes and immunosuppression" *Lancet*
48. Hoe, Anderson, Nathanielsz et al. (2017) "The contrasting roles of Th17 immunity in human health and disease" *Microbiol Immunol*
49. Faure, Poissy, Goffard et al. (2014) "Distinct immune response in two MERS-CoV-infected patients: can we go from bench to bedside?" *PloS One*
50. Josset, Menachery, Gralinski et al. (2013) "Cell host response to infection with novel human coronavirus EMC predicts potential antivirals and important differences with SARS coronavirus" *mBio*
51. Mahallawi, Khabour, Zhang et al. (2018) "MERS-CoV infection in humans is associated with a pro-inflammatory Th1 and Th17 cytokine profile" *Cytokine*
52. Pierce, Mccabe, White et al. (2012) "Biomarkers: an important clinical assessment tool" *Am J Nurs*
53. Kim, Yoon, Yoo et al. (2016) "Combined detection of serum IL-10, IL-17, and CXCL10 predicts acute rejection following adult liver transplantation" *Mol Cells*
54. Niu, Liu, Yin et al. (2011) "The balance between intrahepatic IL-17+ T cells and Foxp3+ regulatory T cells plays an important role in HBV-related end-stage liver disease" *BMC Immunol*
55. Leygo, Williams, Jin et al. (2017) "DNA methylation as a noninvasive epigenetic biomarker for the detection of cancer" *Dis Markers*
56. Tian, Dai, Zhang et al. (2019) "Expression of IL-17 and its gene promoter methylation status are associated with the progression of chronic hepatitis B virus infection" *Medicine (Baltimore)*
57. Zuñiga, Choreño-Parra, Jiménez-Alvarez et al. (2020) "A unique immune signature of serum cytokine and chemokine dynamics in patients with Zika virus infection from a tropical region in Southern Mexico" *Int J Infect Dis*
58. Wang (2020) "C-reactive protein levels in the early stage of COVID-19" *Med Mal Infect*
59. Nabavi, Javidarabshahi, Allahyari et al. (2021) "Clinical features and disease severity in an Iranian population of inpatients with COVID-19" *Sci Rep*
60. Chen, Wu, Guo et al. (2019) "Clinical and immunological features of severe and moderate coronavirus disease" *J Clin Invest*
61. Wang, Deng, Gou et al. (2020) "Preliminary study to identify severe from moderate cases of COVID-19 using combined hematology parameters" *Ann Transl Med*
62. Qin, Zhou, Hu et al. (2020) "Dysregulation of immune response in patients with coronavirus 2019 (COVID-19) in Wuhan" *Clin Infect Dis*
63. Tahtasakal, Oncul, Sevgi et al. (2021) "Could we predict the prognosis of the COVID-19 disease" *J Med Virol*
64. Kwon, Kim, Kim et al. (2020) "Factors of severity in patients with COVID-19: cytokine/chemokine concentrations, viral load, and antibody responses" *Am J Trop Med Hyg*
65. Tjan, Furukawa, Nagano et al. (2019) "Early differences in cytokine production by severity of coronavirus disease" *J Infec Dis*
66. Sadeghi, Tahmasebi, Mahmood et al. (2021) "Th17 and Treg cells function in SARS-CoV2 patients compared with healthy controls" *J Cell Physiol*
67. Ghazavi, Ganji, Keshavarzian et al. (2021) "Cytokine profile and disease severity in patients with COVID-19" *Cytokine*
68. Biasi, Meschiari, Gibellini et al. (2020) "Marked T cell activation, senescence, exhaustion and skewing towards TH17 in patients with COVID-19 pneumonia" *Nat Commun*
69. Valizadeh, Abdolmohammadi-Vahid, Danshina et al. (2020) "Nano-curcumin therapy, a promising method in modulating inflammatory cytokines in COVID-19 patients" *Int Immunopharmacol*
70. Leija-Martínez, Huang, Del-Río-Navarro et al. (2020) "IL-17A and TNF-α as potential biomarkers for acute respiratory distress syndrome and mortality in patients with obesity and COVID-19" *Med Hypotheses*
71. Li, Xu, Yu et al. (2020) "Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan" *J Allergy Clin Immunol*
72. Zhang, Dong, Cao et al. (2020) "Clinical characteristics of 140 patients infected with SARS-CoV-2 in Wuhan" *Allergy*
73. Wu, Chen, Cai et al. (2019) "Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease" *JAMA Intern Med*
74. Angioni, Sánchez-Rodríguez, Munari et al. (2020) "Age-severity matched cytokine profiling reveals specific signatures in Covid-19 patients" *Cell Death Dis*
75. Tanacan, Yazihan, Erol et al. (2021) "The impact of COVID-19 infection on the cytokine profile of pregnant women: A prospective case-control study" *Cytokine*
76. De Lucena, Da, Santos et al. (2020) "Mechanism of inflammatory response in associated comorbidities in COVID-19" *Diabetes Metab Syndr*
77. Wu, Mcgoogan (2020) "Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention" *JAMA* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548413&blobtype=pdf | # MERS-CoV and SARS-CoV-2 infection in diverse human lung organoid-derived cultures
Kim Lam Chiok, Kristof Jenik, Mark Fenton, Darryl Falzarano, Neeraj Dhar, Arinjay Banerjee
## Abstract
Cell cultures are widely used to study infectious respiratory diseases and to test therapeutics; however, they do not faithfully recapitulate the architecture and complexity of the human respiratory tract. Lung organoids have emerged as an alternative model that partially overcomes this key disadvantage. Lung organoids can be cultured in various formats that offer potential for studying highly pathogenic viruses. However, the effects of these different formats on virus infection remain unexplored, leaving their relative value unclear. In this study, we generated primary lung organoids from human donor cells and used them to derive monolayers and air-liquid interface (ALI) cultures with the goal of comparing the replication kinetics of two circulating highly pathogenic coronaviruses, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and Middle East respiratory syndrome coronavirus (MERS-CoV). Infection studies revealed that organoid-derived monolayers displayed limited infection, and the innate immune response was impaired against bacterial lipopolysaccharide (LPS) but not against virus-like double-stranded RNA (dsRNA) or poly(I:C). Meanwhile, organo ids and organoid-derived ALI cultures retained viral permissivity, with ALI cultures displaying diverse antiviral immune responses against both coronaviruses. SARS-CoV-2 and MERS-CoV demonstrated differential replication kinetics in organoid and organoidderived ALI cultures. Therefore, primary organoid-derived cells in two-dimensional monolayer or three-dimensional ALI formats influence virus infection and host antiviral responses. Our study informs the selection of culture conditions for organoid-based respiratory disease research and therapeutic testing.
IMPORTANCEThe COVID-19 pandemic heralded the upsurge in human-derived lung organoid-based studies due to their cellular heterogeneity that partly emulates the cellular complexity of the respiratory tract. A major disadvantage of organoid models resides in their apical-in conformation that "hides" cells and proteins that are typically exposed to the air-liquid interface (ALI) in the airways and are targets of viruses. Here, we generated monolayers and ALI cultures to facilitate cell exposure to highly relevant pathogens and compared them to parental organoids. Organoids at the ALI captured infection and immune responses better than organoids and organoid-derived monolayer cultures. Organoids at the ALI are a viable approach to improve identification and characterization of virus infection, host responses, and therapeutic testing. KEYWORDS coronavirus, SARS-CoV-2, MERS-CoV, tissue culture models, patientderived models, respiratory disease, virus-host interactions, high containment, airway organoid, air-liquid interface cultures T raditionally, researchers have relied on two-dimensional (2D) cell culture models such as immortalized cell lines to uncover molecular mechanisms of disease, understand the host immune response against infectious viral agents, identify therapeu tic targets, and assess drug efficacy (1). Even for studying respiratory viruses, cells are
typically cultured as 2D monolayers, submerged in nutrient-rich media that supports their growth and replication. While these cell lines are important tools for studying disease mechanisms, they often fail to replicate the cellular heterogeneity, three-dimen sional architecture, and physiological responses of respiratory tissue, which is naturally exposed to environmental air in the airways. Animal models are useful to validate in vitro observations; however, these models can be ethically challenging and yield inconclusive outcomes due to anatomical and cellular differences when compared to humans. This is a major bottleneck when studying newly emerging viruses where time is critical and the absence of validated animal models impedes the evaluation of therapeutics (2). As emerging pathogens are a continuous threat to global health, human-like in vitro models are key to identify, develop, and test prophylactic and therapeutic measures.
Organoids derived from human cells have emerged as an alternate model that partly recapitulates the structural and compositional complexity of the native organ. Organoids are composed of self-renewing stem cells that differentiate into various cell types present in the tissue of origin and self-assemble into three-dimensional (3D) microtissues (3). Adult stem cells (ASC) from donor tissue are viable sources for organoid development without the lengthy differentiation process used for induced pluripotent stem cells (iPSCs) or the oncogenic mutations of cancer-derived organo ids (4). Donor-derived organoids preserve the individual-level diversity that influences immune response, susceptibility to pathogens, metabolism, and tolerance to pharma ceuticals. These advantages have poised organoids as promising approaches for use in preclinical therapeutic screening and have received authorization from the US Food and Drug Administration (FDA) for this purpose (5). Human respiratory organoids are being actively used to study respiratory pathogens like influenza A virus (IAV) (6), respiratory syncytial virus (RSV) (7), human adenovirus type 3 (HAdV-3) and type 55 (HAdV-55) (8), and the recently emerged coronaviruses, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and Middle East respiratory syndrome coronavirus (MERS-CoV) (9,10). Despite this progress, studies comparing 2D and 3D cell culture models and their impact on SARS-CoV-2 and MERS-CoV replication are limited, particularly in the case of the highly pathogenic MERS-CoV.
SARS-CoV-2 emerged in late 2019 to cause the COVID-19 (coronavirus disease 2019) pandemic. As of December 2024, the World Health Organization (WHO) has reported over 7 million deaths worldwide since the COVID-19 pandemic started, which is likely an underestimate of the true impact of this pandemic (11). COVID-19 mortality rates are estimated to be between 0.1% and 5% (12). The WHO data show that the virus continues to disseminate, propelled by the emergence of viral variants that can at least partially evade protection from existing vaccines (13). MERS-CoV emerged in 2012 and continues to cause outbreaks of severe viral pneumonia with an approximate case fatality rate of 35% (14). MERS-CoV remains a pathogen of concern and a pandemic threat. Indeed, research into SARS-CoV-2 and MERS-CoV can provide insights into coronavirus biol ogy which will inform the development of prophylactic and therapeutic interventions. Human-derived experimental models are, thus, essential to understand and mitigate risks posed by emerging coronaviruses and other respiratory pathogens.
In this study, we established human donor-derived lung organoids from which we derived traditional cell monolayers and air-liquid interface (ALI) cultures and compared infectivity using currently circulating highly pathogenic coronaviruses, SARS-CoV-2 and MERS-CoV. We used protocols, supplies, and equipment that are widely available in virology laboratories to promote the adoption of 3D systems for virological studies. We generated different 3D culture systems from the same genetic background to facilitate comparison across systems and inform the criteria for selection of models for virology research. We also aimed to expand work on MERS-CoV in 3D models due to its relevance as a virus with pandemic potential and the scarcity of studies that have used 3D cultures to study this pathogen. Lung organoids and derived monolayers and ALI cultures showed distinct differences in virus infection and transcriptional regulation of antiviral genes upon immune stimulation and virus infection. While monolayers transitioned into a virus-resistant phenotype, ALI cultures sustained viral infection and antiviral response against virus infection. Despite their common origin, differences between two-(2D) and three-dimensional (3D) cultures emphasize the need for careful selection of cell culture models for respiratory infectious disease studies and therapeutic testing.
## RESULTS
## Cell culture platforms influence organoid growth properties
We obtained lung tissue sample from a donor patient and used previously established protocols (15) to generate human lung organoids (hLO). These hLO were then cultured and maintained under three different formats-traditional two-dimensional adherent monolayers (hLOm); as three-dimensional air-liquid interface cultures (hLO ALI) or passaged as three-dimensional lung organoids (hLO) (Fig. 1A).
To characterize the distribution of different cell types in the hLOm and hLOs, we carried out immunofluorescence staining using antibodies directed to cell type-specific markers (Fig. 1B). Acetylated tubulin localized along cytoplasmic microtubules in hLOm, within the hLOs indicating an apical-in orientation, and consistent with cilia formation in the apical-out orientation in hLO ALI cultures (Fig. 1B). We identified the mucus compo nent, Muc5AC produced by Goblet cells (16), and the tight junction protein, zona occludens-1 (ZO-1) in hLO and hLO ALI cultures identified as individual cells stained with corresponding markers. Club cells (CC10) typical of bronchiolar epithelium (17) were more abundant in hLOs and hLO ALI relative to hLO monolayers. In hLO ALI, CC10 cells appeared containing secretory granules that suggested functional production and secretion of surfactant-like glycoproteins (Fig. 1B). We did not identify specific individual cells positive for Surfactant Protein C (SFTPC) in any of the cultures, suggesting limited presence of alveolar type 2 (AT-2) cells (Fig. 1B). Similarly, we observed weak staining of KRT5+ cells (basal stem cells) and the alveolar type 1 (AT-1) protein, podoplanin-1 (PDPN1). The absence of several markers in hLOm suggested that this format did not foster various cell types and structures typical of the lungs (Fig. 1B).
We also employed RT-qPCR assays to further investigate the composition of hLOs and the derived culture models (Fig. 1C). We found that markers for ciliated, upper barrier cells (FoxJ1, TUBB4B, TJP1) were expressed in 3D format hLO and hLO ALI. Transcript levels of KRT5 (marker for basal cells) were higher in hLO ALI, consistent with ALI promotion of basal stem cell growth reported previously (18). Expression of stem cell-like markers like TP63 (tumor protein 63) and NGRF (Neurogranin) mirrored each other and were mostly expressed in hLO, suggesting differentiation processes in hLO ALI that correlate with loss of stem cell like properties. Muc5AC, a marker of Goblet cells was predominant in hLO ALI, suggesting that this system can produce mucus. CC10 (or SCGB1A1) transcripts were expressed in all three systems, indicating the potential secretion of glycoproteins by club cells typical of the bronchiolar epithelium (Fig. 1C).
We determined transcript levels of AQP5 (aquaporin 5) and PDPN1 as markers for AT-1 cells. The early alveolar structural marker, PDPN1 remained low in both 3D systems. In contrast, AQP5 was predominantly expressed in hLO ALI cultures, suggesting potential metabolic activity associated with water export and hydration of the epithelium. Transcripts for the AT-2 cellular marker, SFTPC was almost undetectable in 3D models, while transcripts for SLC34A2 (encoding a phosphate transporter involved in surfactant synthesis) was predominant in hLO ALI (Fig. 1C). Therefore, our data demonstrate that cellular composition was different between hLO, hLOm, and hLO ALI cultures derived from the same parental organoids. Our results suggest that 3D formats like hLO and hLO ALI consist primarily of lung bronchial airway cells that are absent in cell culture mono layers. Culture format influences innate immune response against virus-and bacteria-like stimuli
Next, we examined the cellular response against bacteria-like and virus-like stimuli in hLOm and hLO as prototypic 2D and 3D culture systems. hLOm and hLO were transfected with multiple doses of the dsRNA analog polyinosinic-polycytidylic acid [poly(I:C)] to mimic viral infection. Poly(I:C) was delivered successfully in both hLOm and hLO as observed by live cell imaging (Fig. 2A). hLOm and hLO responded to this stimulus by upregulating the transcripts for antiviral genes like interferon-beta (IFNβ) (Fig. 2B), interferon Induced Protein with tetratricopeptide 1 (IFIT1) (Fig. 2C), 2′-5′-oligoadenylate synthetase 1 (OAS1) (Fig. 2D), and MX Dynamin Like GTPase 1 (MX1) (Fig. 2E). While most responses were comparable between hLO and hLOm, upregulation of IFNβ was nearly 100-fold higher in hLOm (Fig. 2B). We (19) and others (20) previously identified IFNβ and this set of interferon-stimulated genes (ISGs) as relevant for antiviral responses against coronaviruses. Therefore, both hLO and hLOm upregulate antiviral genes upon encountering intracellular virus-like stimuli. Lipopolysaccharide (LPS) is a major component of the outer membrane of gramnegative bacteria that induces production of tumor necrosis factor alpha (TNF-α) and interleukin 1 beta (IL-1β) to mediate inflammation and acute lung injury in cellular and animal models (21). hLOs exposed to LPS from pathogenic E. coli O111:B4 responded by upregulating the proinflammatory genes TNF-α and IL-1β (Fig. 2F). In contrast, hLOm did not upregulate these proinflammatory cytokines and remained unresponsive to LPS stimuli. These results indicate that donor-derived hLO and corresponding hLOm are immunocompetent and responsive to virus-like stimuli, but only hLOs respond to LPS. Thus, transition from 3D to 2D culture formats may influence the breadth of the innate immune response, with hLOm missing relevant pulmonary responses such as those directed against bacterial LPS.
Next, we aimed to map the distribution of coronavirus receptors in hLOm, hLO, and hLO ALI. Immunofluorescence staining revealed that distinct cells within parental hLOs expressed angiotensin converting enzyme 2 (ACE2) and TMPRSS2, the cellular receptor for SARS-CoV-2, and its entry cellular co-factor, respectively (22) (Fig. 2G). We did not detect ACE2 in monolayers, whereas TMPRSS2 appeared less intense and diffuse in cytoplasms of cells in monolayers (Fig. 2G). Individual cells in hLO ALI also expressed ACE2 and TMPRSS2, the latter of which localized to the cell membrane and cilia (Fig. 2G), similar to what has been reported previously for primary cell nasal epithelium ALI cultures (23). Despite our best efforts, we were unable to detect the DPP4 protein, the receptor for MERS-CoV (24), by immunostaining. Additional RT-qPCR assays detected transcripts for ACE2 and TMPRSS2 predominantly in hLO and hLO ALI (Fig. 2H), with higher transcript levels in hLO ALI, which is consistent with our immunostaining findings. Meanwhile, we also detected DPP4 transcripts in hLO and hLO ALI, but not in hLOm (Fig. 2H). These results suggest that in addition to differences in cellular composition and architecture, hLO cells under different culture formats may differ in their surface proteins, such as viral receptor expression profiles and, thus, susceptibility to coronavirus infec tion. KRT5 (Basal Stem Cells, white arrows), Muc5AC (mucus and mucus producing cells, Goblet Cells), CC10 (Club Cells), ZO-1 (zona occludens-1, tight junctions), PDPN1 (podoplanin-1, AT-1 cells, white arrows), SFTPC (surfactant protein C, AT-2 cells), and DAPI (nuclei, blue). Cells were imaged on a confocal microscope using a 63× objective. Scale bars correspond to 50 µm. (C) Gene expression of lung markers FoxJ1, TUBB4B, TJP1, TP63, KRT5, NGRF, Muc5AC, SCGB1A1, PDPN1, AQP5 (aquaporin 5, AT-1 cells), SFTPC and SLC34A2 (AT-2 cells) in hLO, hLOm, and hLO ALI was determined by RT-qPCR assays. Samples were assayed in technical duplicates, dCt was normalized to GAPDH, and expression was calculated by 2 -dCt . The mean of three independent samples is presented, and error bars are standard error of the mean (SEM).
## 2D and 3D culture formats influence SARS-CoV-2 infectivity
Since 2D and 3D cultures differed in cell composition, architecture, and breadth of response against pathogen-like stimuli, we next investigated whether culture format also impacts infection with respiratory viruses. We first used SARS-CoV-2 due to its recent relevance for global health. Parental hLO, hLOm, and hLO ALI cultures were infected with SARS-CoV-2 and followed for up to 7 days. Brightfield images suggested that hLO monolayers were composed of cells with epithelial-like morphology (Fig. 3A). Meanwhile, hLOs retained characteristic spherical morphology with presence of cyst-like cavities and absence of cilia on the apical aspect of the organoids (apical-in). hLO ALI displayed airway-like characteristics with beating cilia that actively moved mucus atop the culture, giving the appearance of whorls when viewed from top down (Fig. 3A). Despite the absence of apparent cytopathic effect (CPE) in all three types of culture upon SARS-CoV-2 infection (Fig. 3A), immunofluorescence staining confirmed the presence of the SARS-CoV-2 N protein in hLO and hLO ALI cultures (Fig. 3B). We evaluated tight junction status in ALI cultures by staining for the ZO-1 marker (Fig. 3B), which appeared unaffected by SARS-CoV-2 infection in our experimental conditions (Fig. 3B). We did not detect cells positive for SARS-CoV-2 N protein in hLOm (Fig. 3B). These observations were further confirmed by RT-qPCR assays which showed higher upstream of E gene (UpE) transcripts (19) in hLO and hLO ALI cultures upon infection with SARS-CoV-2 when compared to hLOm (Fig. 3C).
We detected SARS-CoV-2 transcripts in hLOm (low), hLO, and hLO ALI (UpE, Fig. 3C). However, we did not detect SARS-CoV-2 N positive cells in hLOm, whereas hLO ALI cultures infected with SARS-CoV-2 had several cells positive for N (Fig. 3B). We performed TCID 50 assays to determine whether hLOm or hLO ALI produced infectious progeny virions given that virus UpE levels were lower in these formats relative to hLO (Fig. 3D). Infected hLO ALI produced increasing amounts of infectious virus at the apical side of the ALI. In hLOm, virus was detected 2 days post-infection which then decreased over 4-and 7-days post infection (Fig. 3D). We did not detect infectious virus in ALI basolateral medium (Fig. 3D).
We then profiled the immune responses that were generated in the three culture systems upon SARS-CoV-2 infection. hLO and ALI cultures upregulated IFNβ transcripts at 7-and 4-days post-infection, respectively (Fig. 3E). SARS-CoV-2 infection led to the upregulation of several canonical antiviral ISGs in hLO ALI including IFN regulatory factor 7 (IRF7) (Fig. 3F), IFIT1 (Fig. 3G), STAT1 (Fig. 3H), MX1 (Fig. 3I), OAS1 (Fig. 3J), and Radical S-Adenosyl Methionine Domain Containing 2 (RSAD2) (Fig. 3K) at 4 days post infection. In case of hLO, SARS-CoV-2 infection led to the upregulation of IFIT1 (Fig. 3G), with the remaining ISGs failing to respond to virus infection. Analyses of key proinflammatory genes (Fig. 3L through N) demonstrated the upregulation of transcripts for Interleukin 6 (IL6) in hLO ALI at 4 dpi (Fig. 3M), whereas transcripts for IL1β were upregulated in hLO at 7 dpi (Fig. 3N). hLOm remained mostly unresponsive to virus infection and did not upregulate transcripts for our selected antiviral genes other than IRF7 at 2 dpi (Fig. 3F). Thus, hLO ALI cultures were more responsive to SARS-CoV-2 infection, relative to parental (F) hLO and hLOm were treated with either vehicle (LPS-free water) or LPS (100 ng/mL) for 6 h. RNA was collected for RT-qPCR determination of TNF-α and IL-1β transcript levels. (G) hLOm, hLO, and hLO ALI were fixed in paraformaldehyde 4%, permeabilized and stained with antibodies against ACE2 and TMPRSS2, and counterstained with DAPI (nuclei, blue). Cells were imaged using a confocal microscope at immersion (63×). Individual positive cells are presented as insets highlighted with white boxes. Scale bars correspond to 50 µm. (H) Gene expression levels of ACE2, TMPRSS2 and DPP4 in hLO, hLOm, and hLO ALI were determined by RT-qPCR assays. Samples were assayed in technical duplicates, dCt was normalized to GAPDH, and expression was calculated by 2 -dCt . The mean of three independent samples is presented. For RT-qPCR assays in B to F, samples were assayed in technical duplicates, dCt was normalized to GAPDH, and expression was calculated as fold-change relative to vehicle treated controls by 2 -ddCt method. The mean of three hLO and hLOm. These results show that despite their common origin, culture format impacts SARS-CoV-2 infection kinetics and immune response in primary donor lung organoid cells.
## 2D and 3D culture formats impact MERS-CoV infectivity
We next investigated whether culture format also impacts MERS-CoV infection kinet ics and host response. Like SARS-CoV-2, we did not observe CPE under bright field microscopy in either hLOm, hLO, or hLO ALI infected with MERS-CoV (Fig. 4A). Both hLO and hLO ALI had cells that stained positive for MERS-CoV nucleoprotein (N) (Fig. 4B). ALI cultures infected with MERS-CoV had areas of cell detachment, leaving gaps lined by infected cells and exposed cells in the basement layer that appeared as plaque-like lesions (Fig. 4B). These disrupted areas had discontinuous distribution of the ZO-1 marker, suggesting loss of tight junction integrity (Fig. 4B). RT-qPCR assays confirmed upregulation of viral gene transcripts (UpE) (25) in MERS-CoV infected hLOm, hLO, and hLO ALI cultures although hLOm and hLO ALI cultures did not show time-dependent increase of UpE (Fig. 4C).
Next, we performed TCID 50 assays to determine whether hLOm and hLO ALI cultures produced infectious progeny virions into culture media and the apical side of ALIs (Fig. 4D). Like SARS-CoV-2, hLOm produced infectious MERS-CoV at 2 dpi, which decreased over time (Fig. 4D). Infectious MERS-CoV in apical washes of hLO ALIs or in the basolateral medium remained low or undetectable (Fig. 4D). Transcripts for IFNβ (Fig. 4E) and IRF7 (Fig. 4G) were upregulated at early time points post infection in hLOm. IFIT1 transcript levels remained largely unchanged (Fig. 4F). Transcripts for antiviral genes like MX1 (Fig. 4I) and OAS1 (Fig. 4J), along with proinflammatory genes like IL1β (Fig. 4N), were upregulated following infection with MERS-CoV in hLO ALI. Transcript levels for IFIT1, STAT1, RSAD2, TNFα, and IL6 remained largely unchanged or were minimally differentially regulated upon infection with MERS-CoV (Fig. 4F, H, K, L andM). In hLOs, we noted mostly discrete downregulation of antiviral and proinflammatory gene transcripts. These results show that hLO ALI partly recapitulated host responses against MERS-CoV infection, which were not detected in hLO and hLOm cultures.
## DISCUSSION
Lung organoids are emerging tools to identify and model respiratory infectious diseases due to their tissue-like cellular heterogeneity and three-dimensional organization. Here, we generated donor-derived lung organoids and used them to make two-dimensional cell monolayers (hLOm) and three-dimensional ALI (hLO ALI) cultures. We compared infection kinetics of SARS-CoV-2 and MERS-CoV and associated host responses in these three culture formats. Relative to hLOm and parental hLO, hLO ALI were more responsive to virus infection, allowing for the identification of immune response signatures corre sponding to SARS-CoV-2 and MERS-CoV. Our comparative study shows that despite sharing a common origin and genetic background, each culture format had distinctive virus infection and host response patterns.
Although organoids do not reflect the full composition or functionality of the lungs, they can partially recapitulate key features of the respiratory tract to a better extent than traditional 2D cell lines (4). Organoid culture is achieved through growth in gel-like matrix supports to ensure survival and differentiation of stem cells into cell types typical of the tissue of origin. Growth in basement membrane matrix gel prevents exposure to air while promoting growth of organoids as multicellular aggregates typically in the apical-in orientation wherein functional cilia does not develop (26). Despite these disadvantages, organoids from PSC (10,27), iPSCs, and adult stem cells (9) have been used to generate nasal (28), tracheobronchial (27), and alveolar organoids (29) to model SARS-CoV-2 infection. Studies generating organoids from iPSCs (30), embryonic stem cells or ESCs (31), and adult lung tissue (10) have reported that the origin and culture conditions of stem cells can influence the cell population that are present in final organoids, along with differences in their immune competency (32), which, in turn, may influence which cells are infected with SARS-CoV-2. Therefore, donor-specific traits (32), origin of stem cells, and types of cells within organoids may shape the antiviral response against virus infections. The use of hESCs and hPSCs may not be feasible for every virology lab as access to these cells and reagents can be expensive, and the extensive time required to culture organoids can be limiting. Given the recent drive from the FDA (US) to de-risk preclinical testing by embracing organoid models and their inherent donor-to-donor variation (5), we used adult lung tissue as a source for stem cells. Culture of organoids from adult lung tissue follows procedures that can be easily adopted by the wider virology community and requires equipment that is already available in most virology laborato ries. In addition to SARS-CoV-2, our goal was to evaluate this platform for MERS-CoV since limited studies have used respiratory organoids for MERS-CoV research (33,34).
Access to tissues from healthy adult donors may be limiting. Biobanks cataloging and storing patient-derived cancer tissue are available through academic (e.g., Princess Margaret Living Biobank, the Hubrecht Institute) and commercial vendors (e.g., Cellesce, DefiniGEN). Biobanks for healthy tissues are not well developed and require partnership with clinicians and other researchers. While partnerships with academic hospitals is the most frequent way to secure healthy tissue for organoid-based research, regulated commercial enterprises (e.g., HUB organoids) are making headway in bringing these services to academic users.
As reported previously, we found that organoids generated from healthy donor tissue differentiated into cell types typical of bronchial airway tissue in the apical-in orientation due to the use of matrigel (Fig. 1) and were readily infectable by SARS-CoV-2 (Fig. 3) and MERS-CoV (Fig. 4). However, infecting hLOs involved the dissolution of matrigel, subsequent recovery of matrix-free organoids followed by virus infection in suspension, and organoid re-embedding in matrix, which is a complex process that does not facilitate high-throughput work. While virus infection was successful in matrigel-free whole organoids by us in this study and by others (35), methods like mechanical shearing or microinjection are also used to facilitate infection of apical-in organoids (36). Recently developed microinjection platforms can process ~90 organoids per hour (37), but require specialized equipment and lack high-throughput capabilities, which curtails microinjection use in higher containment level laboratories. Microinjection platforms can be considered for infection of organoids; however, these protocols will need to be carefully assessed by institutional biosafety committees for biosafety risks before they can be implemented. (G), STAT1 (H), MX1 (I), OAS1 (J), RSAD2 (K), TNF-α (L), IL6 (M), and IL1β (N) was evaluated by RT-qPCR. The mock infected samples for hLO ALI are the same as in Fig. 3. For RT-qPCR assays, samples were assayed in technical duplicates, dCt was normalized to GAPDH, and virus gene expression was calculated as 1/dCt. Host gene expression was calculated as fold-change relative to the mean of mock infected controls using the 2 -ddCt method. The mean of three independent samples Given current challenges, we were prompted to explore the versatility of organoids in two additional culture formats: as cell monolayers (2D) and as ALI cultures (3D). This approach examines the intra-individual differences in organoid culture under the same genetic background, facilitating comparison between organoid culture formats to inform on choice criteria for virology research. Cell line monolayers composed of homogenous cell populations expressing virus receptors are the cornerstone for studies on SARS-CoV-2 and MERS-CoV. Cell lines like Calu-3 (38,39), Huh-7 (40), MRC-5 (41), and the non-human primate cell line Vero and its derivates (42) have been widely used in coronavirus biology research. We used organoids to produce monolayers and analyzed virus infection kinetics. Unlike cell lines in which viruses cause widespread damage (cytopathic effect) due to quick and efficient replication, monolayers derived from primary organoids were infected poorly by SARS-CoV-2 (Fig. 3) and MERS-CoV (Fig. 4). Transition from organoids into monolayers led to the loss of detectable ACE2 and TMPRSS2 expression levels (Fig. 2H), resulting in a loss of viral infectivity in these cells.
hLOm were found to be immunocompetent against viral dsRNA analogs (Fig. 2A through F) but were unresponsive to bacterial LPS (Fig. 2G), indicating that monolayers do not retain the full breadth of immune response signaling. Human primary tracheo bronchial cells upregulate IL6 and IL-1β when exposed to high doses (10-100 µg/mL) of LPS in vitro (43) although high LPS dosage has also been linked to loss of viability and subsequent release of inflammatory markers (44). However, features observed in primary cell cultures cannot be extended to hLOm due to differences in culture establishment and conditions such as growth media. In our study, the addition of a single growth factor (Epidermal Growth Factor, EGF) was required to direct monolayer-like growth from dissociated organoid cells. In epithelial cells, LPS engages TLR4 at the cell surface in the canonical pathway, which results in transactivation of the EGF receptor (EGFR) with subsequent NF-κβ activation in LPS-dependent acute lung injury models (45). While additional studies are needed to clarify the mechanisms by which hLOm lose responsive ness to LPS, we surmise that it is possible that EGF occupancy or regulation of EGFR may interfere with LPS-dependent inflammation.
Our data suggest that cell monolayers derived from organoids may not be appropri ate models for mechanistic studies on pathogen-host interactions and care must be observed about the utility of these models to study antiviral response. In contrast, hLO ALI cultures could be infected with SARS-CoV-2 which led to the production of infectious progeny virions that were released at the apical side of the ALI membrane (Fig. 3). Our studies adopted a multi-pronged approach of immunofluorescence, RT-qPCR, and TCID50 assays to determine virus infection. These assays vary in their sensitivity and may present discrepancies that justify their combined use for evaluating virus infections. RT-qPCR detects low levels of transcripts, whereas immunofluorescence requires higher levels of viral proteins for visualization using microscopy. An additional hurdle arises in TCID50 assays to detect viruses in culture supernatant from matrigel-embedded organoids. Some groups have reported that viruses become entrapped in matrigel and are difficult to recover from medium (29,46), whereas other groups report no issues in virus recovery from matrigel supernatants (47). Given these conflicting reports, additional steps may be needed to recover matrigel-entrapped viruses in high-contain ment facilities, along with complementary quantification of virus infection by immuno fluorescence and RT-qPCR. We did not attempt TCID50 assays using culture supernatant from hLOs embedded in matrigel.
ALI cultures upregulated a diverse array of antiviral transcripts more frequently than parental organoids (Fig. 3E through N). Our study is consistent with previous work highlighting the improved immune response against SARS-CoV-2 in ALI cultures, including those generated from primary bronchial cells (48). The commercialization of transwell inserts and ready-made cell culture reagents have facilitated the establishment of ALI cultures from different cell sources with minimal modifications. Indeed, commer cial vendors (e.g., StemCell, Lonza, Corning) have made significant advances in pre-made media and plasticware for 3D culture, facilitating its adoption across disciplines.
Studies exploring MERS-CoV infection in organoids and ALI cultures are scarce. We investigated both culture formats and found that parental organoids (hLO) and hLO ALI were permissive to MERS-CoV infection (Fig. 4). Our findings are consistent with a recent report in human airway organoids suggesting that merbecoviruses likely infect airway organoids with low efficiency (49). Responses against MERS-CoV were centered on upregulation of antiviral transcripts for MX1 (Fig. 4I) and OAS1 (Fig. 4J) as reported in other in vitro studies (50,51). Interestingly, MX1 and OAS1 transcripts were downre gulated in parental organoids and hLOm infected with MERS-CoV. Induction of IFNβ transcripts was largely absent in hLO and hLO ALI cultures, demonstrating that 3D models can also emulate this aspect of MERS-CoV IFN antagonism (52). MERS-CoV infection induced the upregulation of transcripts for proinflammatory cytokines like IL-1β in hLO ALI cultures, but not in the other formats (Fig. 4N). Thus, in our studies, hLO ALI models better recapitulated responses that underlie hyperinflammatory syndromes in patients with severe MERS (53) and virus-mediated IFN antagonism.
MERS-CoV induces CPE or cell death in various human and non-human cell lines like Calu3, Huh7, Vero, and derivates (42,54). MERS can induce sloughing of infected cells in ALI cultures made from primary human tracheobronchial cells (55). Instead of sloughing, we identified dissolution of tight junctions at the epithelial barrier of organoid-derived ALIs where plaque-like lesions lined by infected cells had formed (Fig. 4B). Disturbances in the epithelial barrier have also been identified in human alveolar tissue infected with MERS-CoV ex vivo (56) and in postmortem histopathological exams (57). Our study suggests that organoids at the ALI can recapitulate part of the virusinduced tissue damage and immune responses better than their parental organoids. Neither SARS-CoV-2 nor MERS-CoV induced extensive CPE reported in 2D cell lines although cell lines do not necessarily reflect the diverse pathologies observed in infected individuals. Cell lines are homogeneous in their expression of virus receptors and support of virus replication, which intensifies the appearance of CPE. On the contrary, cellular and phenotypic heterogeneity in the lungs occurs at the intra-individual level, thus governing different populations of cells that are affected at any given time in an individual. In this fundamental difference, 3D models partly capture the general heterogeneity and intrinsic variability of the lungs better than traditional cell cultures.
A potential drawback of our approach is the use of tissue obtained from a single donor to generate organoids and organoid-derived cultures. Therefore, while addressing intra-donor variation across culture formats, our study does not address donor-to-donor differences that influence susceptibility and host responses against virus infections. Despite this limitation, and in the background of a shared origin and genetic makeup, 2D and 3D cell culture formats influenced pathogen kinetics and host immune responses in our studies. These differences between 2D and 3D cultures must be taken into consideration when selecting models to investigate respiratory diseases. As technology continues to advance, bioengineering approaches like microfluidic systems can help refine current respiratory models for infectious disease research and therapeutic testing. Indeed, our work informs the selection and use of human-derived models compatible with biomimetic approaches for infectious disease research and therapeutic screening. Given the limitations of organoids that typically grow in apical-in orientations, organoids at the ALI are a more reliable and reproducible model for infection and immune response studies that partially mimic the air-liquid environment of the lungs.
## MATERIALS AND METHODS
## Generation of human lung organoids
Human lung organoids (hLO) were established as described by Sachs (15) from a healthy tissue sample obtained from the upper lobe of a male patient undergoing surgery. Briefly, lung tissue was minced in Advanced DMEM/F-12 (Gibco, cat. 12634010) before single cell dissociation in ACF dissociation solution (StemCell, cat. 05426) for 1.5 h at 37°C. Undigested tissue was pelleted, and the cell suspension was strained through a 37 µm mesh reversible strainer. Red blood cells were lysed with ACK lysing buffer (Gibco, cat. A1049201), and the remaining cells were embedded in Matrigel (Corning, cat. CACB356231). Matrigel-embedded cells (P0) were dispensed as 40 µL dome/well on three wells of a pre-warmed 12-well tissue culture plate, left to solidify at room temperature for 2 min and then incubated upside down at 37°C for 10 min. Solidified domes were submerged in 400 µL of complete organoid media (15) (Table 1) supplemen ted with antibiotic-antimycotic (Gibco, cat. 15240096), with media changes every 3-4 days. hLO were passaged (1:6 ratio) every 7-10 days by dissociation at 37°C (TrypLE Express 1X) into clusters of 3-4 cells each. Cell clusters were washed in advanced DMEM, re-embedded into fresh Matrigel, and dispensed as 25 µL dome/well of a pre-warmed 24-well tissue culture plate, left to solidify as before and submerged in 250 µL of complete organoid media. To generate cell monolayers (hLOm), hLOs were dissociated into clusters of 3-4 cells (TrypLE Express), washed, and seeded in pre-warmed cell culture plates using complete hLO media supplemented with EGF (20 ng/mL). Media was changed every 3-4 days until reaching 80% confluence before passage or used in indicated experiments. Air-liquid interface cultures (hLO-ALI) were generated using hLOs dissociated as before and seeded at ~1-3 × 10 5 cells/200 µL in hLO complete media supplemented with EGF (20 ng/mL) on the top chambers of 12 mm transwell inserts (StemCell, cat. 38024). Bottom chambers were filled with 500 µL of hLO complete media supplemented with EGF, and cell-laden transwells were incubated at 37°C for 2-4 days. The top chamber was exposed to the air interface for 28 days to allow for bronchiolar airway-like differentiation, and bottom chamber media was replaced with ALI maintenance medium (Pneumacult-ALI, StemCell, cat. 05001) and changed every 2 days. The apical aspect of ALI cultures was washed in 200 µL of PBS for 10 min at 37°C at least once a week after exposure to air to remove mucus buildup.
## Virus infection
SARS-CoV-2/SB2 clinical isolate (58) and MERS-CoV isolate EMC/2012 were used in these experiments. hLOs were released from Matrigel in ice cold gentle harvesting buffer (Cultrex, cat. 3700-100-01) and infected at an MOI of 20 per hLO (~0.3 per cell) for 2 h at 37°C in Advanced DMEM/F-12. hLOs were washed twice to remove unbound virus, re-embedded in Matrigel, and cultured in hLO complete media for indicated times. Monolayers were cultured until 80% confluence, washed with PBS, and infected at MOI 0.3 for 1 h at 37°C. Cells were PBS washed and incubated for the indicated times in hLO complete media supplemented with EGF. The apical side of differentiated ALI cultures was washed with PBS pre-warmed at 37°C for 10 min to remove mucus prior to infection. ALIs were infected at MOI 0.3 for 1 h at 37°C to mimic virus entry through the air interface. Infected ALIs were washed twice and incubated in ALI maintenance medium for indicated times. After indicated times, total RNA was collected using the RNeasy Mini kit (Qiagen, cat 74104) following the manufacturer's instructions. Work with infectious virus was performed in containment level 3 (CL-3) facilities at VIDO-Intervac, University of Saskatchewan.
## Immunocompetence assays
hLOs, hLOm, and hLO-ALI were exposed to LPS 100 ng/mL (Invivogen, cat tlrl-eblps) prepared in complete media for 6 h. For transfection assays, organoids were recovered from matrigel using ice cold gentle harvesting buffer (Cultrex, cat. 3700-100-01) and transfected in suspension with poly(I:C) rhodamine (InvivoGen, cat tlrl-picr) at increasing concentrations for 6 h using lipofectamine 3000 (Invitrogen, cat. L3000015). Transfected organoids were then re-embedded in fresh matrigel and incubated for 48 h. RNA was extracted with the RNeasy Mini kit (Qiagen, cat 74104) following the manufacturer's instructions.
## RT-qPCR
cDNA synthesis from ~500 ng of RNA was generated with the iScript gDNA Clear cDNA Synthesis Kit following the manufacturer's instructions (BioRad 1725034). cDNA was diluted 1:10 in RNAse free water and used as template for qPCRs using Ssoadvanced Universal SYBR kit following the manufacturer's instructions (BioRad 1725274) using selected primers (Table 2). UpE primers for SARS-CoV-2 and MERS-CoV were utilized to detect both genomic and subgenomic RNA from actively replicating virus (19,25).
For RT-qPCR assays, samples were assayed in technical duplicates, dCt was normalized to GAPDH, and virus gene expression was calculated as 1/dCt. qPCRs were performed in a StepOne Real Time PCR System (Applied Biosciences). We normalized Ct values by GAPDH (dCt) and used 1/dCt for viral transcript levels (UpE), and 2 -ddCt method for quantitation of host gene expression relative to time-matched mocks or vehicle controls.
## Immunocytochemistry
Matrigel-free hLOs fixed in neutral buffered formalin (NBF) 10% were blocked and permeabilized for 2 h at room temperature (5% BSA, 1% Triton-X-100, 0.1% Tween-20 in PBS). Incubation with primary antibodies was performed at 1:250 dilution (Table 3) in IF buffer (0.1% BSA, 0.2% Triton-X-100 in PBS) overnight at 4°C. Samples were washed in IF buffer thrice before addition of labeled secondary antibodies (1:3,000). Nuclei were counterstained with DAPI (4 µg/mL) prepared in IF buffer, rinsed once in PBS and once in distilled water, and mounted on glass slides using ProLong Gold antifade mountant (Invitrogen, cat. P36930). Monolayers cultured in chambered slides (IBIDI 80806) were fixed with NBF 10% and processed as in hLOs. hLO ALI samples were fixed and cryoprotected in 30% sucrose for 24 h, embedded in Tissue TEK-O.C.T compound (Sakura), and frozen in cryomolds at -20°C. Transections of 4 µm were prepared by cryo-sectioning and processed for labeling as in hLOs. Samples were imaged in Leica TCS SP8.
## Virus titration (TCID 50 assays)
Apical washes and basal medium of mock and infected ALI cultures were collected for titration assays. Apical washes were obtained by adding 200 µL of PBS to the apical side of ALI cultures and incubating for 10 min at room temperature before collection. Supernatant from cell monolayers were also collected. Virus titration was performed as described previously (19). In brief, Vero76 cells were seeded at 2 × 10 4 cells per well in 96 well plates and incubated for 24 h at 37°C in complete media (DMEM supplemented with 10% FBS, 1% penicillin-streptomycin, and 1× Glutamax). Vero 76 cells were infected in triplicates with undiluted and tenfold diluted samples prepared in serum free DMEM for 1 h at 37°C. Inoculum was removed, and Vero76 cells were reconstituted with 2% FBS media and incubated at 37°C (5% CO 2 ) for up to 5 days. Cells were monitored at 3and 5-days post infection to determine CPE. Virus titers were calculated according to the
## References
1. De Souza, Bideau, Boschi et al. (2022) "Choosing a cellular model to study SARS-CoV-2" *Front Cell Infect Microbiol*
2. Rong, Liu (2023) "Development of animal models for emerging infectious diseases by breaking the barrier of species susceptibility to human pathogens" *Emerg Microbes Infect*
3. Zhao, Chen, Dowbaj et al. (2022) *Organoids. Nat Rev Methods Primers*
4. Kim, Koo, Knoblich (2020) "Human organoids: model systems for human biology and medicine" *Nat Rev Mol Cell Biol*
5. (2025) *Full-Length Text Journal of Virology*
6. Han (2023) "FDA modernization act 2.0 allows for alternatives to animal testing" *Artif Organs*
7. Hui, Ching, Chan et al. (2018) "Tropism, replication competence, and innate immune responses of influenza virus: an analysis of human airway organoids and ex-vivo bronchus cultures" *Lancet Respir Med*
8. Harford, Rezaee, Dye et al. (2022) "RSVinduced changes in a 3-dimensional organoid model of human fetal lungs" *PLoS One*
9. Zhao, Wu, Tan et al. (2023) "Generation of human embryonic stem cellderived lung organoids for modeling infection and replication differences between human adenovirus types 3 and 55 and evaluating potential antiviral drugs" *J Virol*
10. Salahudeen, Choi, Rustagi et al. (2020) "Progenitor identification and SARS-CoV-2 infection in human distal lung organoids" *Nature*
11. Lamers, Van Der Vaart, Knoops et al. (2021) "An organoidderived bronchioalveolar model for SARS-CoV-2 infection of human alveolar type II-like cells" *EMBO J*
12. (2023) "COVID-19 deaths | WHO COVID-19 dashboard"
13. (2022) "Mortality analyses"
14. Wang, Iketani, Li et al. (2023) "Alarming antibody evasion properties of rising SARS-CoV-2 BQ and XBB subvariants" *Cell*
15. Milne-Price, Miazgowicz, Munster (2014) "The emergence of the Middle East respiratory syndrome coronavirus" *Pathog Dis*
16. Sachs, Papaspyropoulos, Zomer-Van Ommen et al. (2019) "Long-term expanding human airway organoids for disease modeling" *EMBO J*
17. Kesimer (2022) "Mucins MUC5AC and MUC5B in the airways: MUCing around together" *Am J Respir Crit Care Med*
18. Blackburn, Li, Bartlett et al. (2023) "An update in club cell biology and its potential relevance to chronic obstructive pulmo nary disease" *Am J Physiol Lung Cell Mol Physiol*
19. Prescott, Pankow, De Vries et al. (2023) "A comparative study of in vitro air-liquid interface culture models of the human airway epithelium evaluating cellular heterogeneity and gene expression at single cell resolution" *Respir Res*
20. Banerjee, El-Sayes, Budylowski et al. (2021) "Experimental and natural evidence of SARS-CoV-2-infection-induced activation of type I interferon responses"
21. Blanco-Melo, Nilsson-Payant, Liu et al. (2020) "Imbalanced host response to SARS-CoV-2 drives development of COVID-19" *Cell*
22. Jeyaseelan, Chu, Young et al. (2004) "Transcriptional profiling of lipopolysaccharide-induced acute lung injury" *Infect Immun*
23. Hoffmann, Kleine-Weber, Schroeder et al. (2020) "SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor" *Cell*
24. Wu, Lidsky, Xiao et al. (2023) "SARS-CoV-2 replication in airway epithelia requires motile cilia and microvillar reprogramming" *Cell*
25. Wang, Shi, Jiang et al. (2013) "Structure of MERS-CoV spike receptor-binding domain complexed with human receptor DPP4" *Cell Res*
26. Corman, Eckerle, Bleicker et al. (2012) "Detection of a novel human coronavirus by real-time reversetranscription polymerase chain reaction" *Euro Surveill*
27. Kozlowski, Crook, Ku (2021) "Towards organoid culture without matrigel" *Commun Biol*
28. Duan, Tang, Nair et al. (2021) "An airway organoid-based screen identifies a role for the HIF1α-glycolysis axis in SARS-CoV-2 infection" *Cell Rep*
29. Li, Huang, Yu et al. (2023) "Human airway and nasal organoids reveal escalating replicative fitness of SARS-CoV-2 emerging variants" *Proc Natl Acad Sci*
30. Katsura, Sontake, Tata et al. (2020) "Human lung stem cell-based alveolospheres provide insights into SARS-CoV-2-mediated interferon responses and pneumocyte dysfunction" *Cell Stem Cell*
31. Han, Duan, Yang et al. (2021) "Identification of SARS-CoV-2 inhibitors using lung and colonic organoids" *Nature*
33. Pei, Feng, Zhang et al. (2021) "Host metabolism dysregulation and cell tropism identification in human airway and alveolar organoids upon SARS-CoV-2 infection" *Protein Cell*
34. Wu, Thi, Huang et al. (2018) "Intrinsic immunity shapes viral resistance of stem cells" *Cell*
35. Gong, Jiang, Lin et al. (2024) "Establishment of a human organoid-based evaluation system for assessing interspecies infection risk of animal-borne coronaviruses" *Emerg Microbes Infect*
36. Park, Kim, Jang et al. (2024) "Differential betacoronavirus infection dynamics in human bronchial epithelial organoids" *J Med Virol*
37. Hysenaj, Little, Kulhanek et al. (2023) "SARS-CoV-2 infection of airway organoids reveals conserved use of Tetraspanin-8 by Ancestral, Delta, and Omicron variants" *Stem Cell Rep*
39. Sharma, Thacker, Dhar et al. (2021) "Early invasion of the bladder wall by solitary bacteria protects UPEC from antibiotics and neutrophil swarms in an organoid model" *Cell Rep*
40. Williamson, Arnold, Samsa et al. (2018) "A high-throughput organoid microinjection platform to study Full-Length Text Journal of Virology October"
41. "gastrointestinal microbiota and luminal physiology" *Cell Mol Gastroen terol Hepatol*
42. Li, Ayyanathan, Dittmar et al. (2023) "SARS-CoV-2 ORF6 protein does not antagonize interferon signaling in respiratory epithelial Calu-3 cells during infection" *mBio*
43. Shapira, Monreal, Dion et al. (2022) "A TMPRSS2 inhibitor acts as a pan-SARS-CoV-2 prophylactic and therapeutic" *Nature*
44. Chen, Saccon, Appelberg et al. (2021) "Type-I interferon signatures in SARS-CoV-2 infected Huh7 cells"
45. Millet, Whittaker (2014) "Host cell entry of Middle East respiratory syndrome coronavirus after two-step, furin-mediated activation of the spike protein" *Proc Natl Acad Sci*
46. Shirato, Kawase, Matsuyama (2013) "Middle East respiratory syndrome coronavirus infection mediated by the transmembrane serine protease TMPRSS2" *J Virol*
47. Jin, Shao, Yang et al. (2023) "Airway epithelial cGAS inhibits LPS-induced acute lung injury through CREB signaling" *Cell Death Dis*
48. Nova, Skovierova, Strnadel et al. (2020) "Shortterm versus long-term culture of a549 cells for evaluating the effects of lipopolysaccharide on oxidative stress, surfactant proteins and cathelicidin LL-37" *Int J Mol Sci*
49. De, Zhou, Desantis et al. (2015) "Erlotinib protects against LPS-induced endotoxicity because TLR4 needs EGFR to signal" *Proc Natl Acad Sci*
50. Giobbe, Bonfante, Jones et al. (2021) "SARS-CoV-2 infection and replication in human gastric organoids" *Nat Commun*
51. Ekanger, Zhou, Bohan et al. (2022) "Human organotypic airway and lung organoid cells of bronchiolar and alveolar differentiation are permissive to infection by influenza and SARS-CoV-2 respiratory virus" *Front Cell Infect Microbiol*
52. Stölting, Baillon, Frise et al. (2022) "Distinct airway epithelial immune responses after infection with SARS-CoV-2 compared to H1N1" *Mucosal Immunol*
53. Chen, Yang, Si et al. (2023) "A bat MERS-like coronavirus circulates in pangolins and utilizes human DPP4 and host proteases for cell entry" *Cell*
54. Chu, Hui, Perera et al. (2018) "MERS coronaviruses from camels in Africa exhibit region-dependent genetic diversity" *Proc Natl Acad Sci*
55. Alsamman, Zayed (2020) "The transcriptomic profiling of SARS-CoV-2 compared to SARS, MERS, EBOV, and H1N1" *PLoS One*
56. Schroeder, Mache, Kleine-Weber et al. (2021) "Functional comparison of MERS-coronavirus lineages reveals increased replicative fitness of the recombinant lineage 5" *Nat Commun*
57. Kim, Choe, Park et al. (2016) "Clinical progression and cytokine profiles of middle east respiratory syndrome coronavirus infection" *J Korean Med Sci*
58. De Wilde, Vs, Oudshoorn et al. (2013) "MERS-coronavirus replication induces severe in vitro cytopathology and is strongly inhibited by cyclosporin A or interferon-α treatment" *J Gen Virol*
59. Li, Wohlford-Lenane, Bartlett et al. (2022) "Inter-individual variation in receptor expression influences MERS-CoV infection and immune responses in airway epithelia" *Front Public Health*
60. Hocke, Becher, Knepper et al. (2013) "Emerging human middle East respiratory syndrome coronavirus causes wide spread infection and alveolar damage in human lungs" *Am J Respir Crit Care Med*
61. Alsaad, Hajeer, Balwi et al. (2018) "Histopathology of Middle East respiratory syndrome coronovirus (MERS-CoV) infection -clinicopathological and ultrastruc tural study" *Histopathology*
62. Banerjee, Nasir, Budylowski et al. (2020) "Isolation, sequence, infectivity, and replication kinetics of severe acute respiratory syndrome coronavirus 2" *Emerg Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645937&blobtype=pdf | # A structural roadmap for the formation of the coronavirus nsp3/ nsp4 double membrane vesicle pore and its implications for polyprotein processing and replication/transcription
Jason Perry, Samuel Itskanov, John Bilello, Eric Lansdon
## Abstract
Coronavirus replication is understood to occur within double membrane vesicles (DMVs) that arise during viral infection. Prior work has determined that these DMVs have characteristic pores formed from the non-structural viral proteins nsp3 and nsp4, which facilitate export of newly synthesized viral RNA. Yet how the replication machinery, which is comprised of the non-structural proteins nsp7 to nsp16, is recruited to the DMV remains a mystery. Working from AlphaFold and previously determined structures, we constructed a series of models that link formation of the DMV pore to nsp5 protease processing of the polyprotein and trapping of the cleaved products within the DMV itself. We argue that the initial pore is formed from 12 subunits of nsp3 and six subunits each of the intermediate uncleaved polyproteins pp1a' (nsp4-nsp10) and pp1ab' (nsp4-nsp16). Formation of this initial structure activates the protease function of alternating nsp5 subunits within a close-packed ring, facilitating the initial trans-cleavage of the nsp4-nsp5 linkage. Maturation of the pore follows, as does formation of canonical nsp5 dimers, which can process the remainder of the polyproteins. When protease activation occurs subsequent to closure of the DMV, the cleavage products, whose stoichiometry is consistent with the previously proposed nsp15-centered hexameric replication complex, will be trapped inside.IMPORTANCE Coronaviruses, like many other positive-sense single-stranded RNA viruses, rely on intracellular membrane remodeling to create a protected environment for efficient replication to occur. The resulting double membrane vesicles (DMVs) have characteristic pores formed from the non-structural viral proteins nsp3 and nsp4, while enzymes responsible for RNA replication, including the polymerase nsp12, are contained inside. Recent structural work has elucidated the nature of the pores, but how the polymerase and other viral proteins are recruited to the DMV represents a major gap in our current knowledge. Here we present a novel, step-by-step structural model of how the pores initially form from the uncleaved polyprotein, how protease activity is initiated, and how the viral replication machinery is trapped within the DMV.
the proteins necessary for replication (2)(3)(4)(5)(6)(7)(8), thus serving the dual purposes of shielding the RNA from triggering an immune response and protecting the RNA and viral proteins from degradation. Research has focused on the viral transmembrane non-structural proteins nsp3, nsp4, and nsp6 as the principal drivers of membrane remodeling and DMV formation (9). The general picture that has emerged from this work is that nsp3 and nsp4 are primarily responsible for membrane doubling and DMV formation, while nsp6 is responsible for membrane zippering and the tethering of DMV clusters to lipid droplets and the ER (10). Initial images of severe acute respiratory syndrome (SARS) DMVs suggested a closed organelle (11); however, more detailed cryo-electron tomography (cryo-ET) images have captured multiple distinct pores embedded in the membranes of each DMV, providing a necessary mode of egress for newly synthesized RNA. Low-resolu tion cryo-ET reconstructions for both murine hepatitis virus (MHV) (12) and SARS-CoV-2 (13) clearly revealed pore structures with distinct sixfold symmetric cytosolic crowns terminating in prominent prong-like features. In the case of SARS-CoV-2, this cryo-ET work also demonstrated that DMV formation and pore formation go hand in hand, and that nsp3 and nsp4 are the minimal viral components required. A recent 4.2 Å resolution cryo-ET structure further revealed the pore in even greater detail, specifically showing the crown to be composed of 12 subunits of nsp3, anchored to a dodecameric arrangement of nsp4 spanning both inner and outer membranes (14).
Nsp3 and nsp4 can be found on the first open reading frame (ORF1) of the coro navirus genome. Encompassing over 70% of the genome, ORF1 encodes two viral polyproteins, pp1a (nsp1-nsp10) and pp1ab (nsp1-nsp16). They are distinguished by a ribosomal frameshift located between nsp10 and nsp12 (15). Both polyproteins are cleaved by viral proteases embedded in nsp3 and nsp5. Studies of MHV and SARS established that the first three non-structural proteins (nsp1-nsp3) are efficiently cleaved by the papain-like proteases of nsp3 (PLpro in SARS-CoV-2) (16,17), while the remainder of the polyprotein (nsp4-nsp10, referred to here as pp1a' , and nsp4-nsp16, referred to as pp1ab') are cleaved in a more regulated manner by the main protease (Mpro) of nsp5 (18,19). Notably, biochemical studies of cleaved nsp5 have shown the protease to be significantly more active when dimerized (20), but the factors that initiate nsp5 maturation and polyprotein processing have yet to be elucidated (21). Considering that uncleaved nsp5 is sandwiched between the membrane binding proteins nsp4 and nsp6, additional constraints on its initial activity should be expected, but these constraints have not been explored and remain poorly understood. With respect to the DMV pore, it has been clearly demonstrated that cleavage between nsp3 and nsp4 is required prior to pore formation (13,22), but no comparable study has established when in the process of pore formation cleavage between nsp4 and nsp5 occurs. Thus, whether nsp5 maturation occurs prior to pore formation or in concert with it is an open question.
Once the pp1a' and pp1ab' polyproteins are fully cleaved, the resulting individual proteins, nsp7-nsp16, constitute the viral replication machinery (23). At a minimum, RNA polymerization is carried out by a complex of (nsp12)(nsp7)(nsp8) 2 (24). Additional enzymatic domains of nsp12 (25), nsp13 (26,27), nsp14 (28,29), and nsp16 (30), aided by the cofactors nsp9 and nsp10, are responsible for other critical viral RNA replication functions, including proofreading and mRNA capping. Previously, we proposed these viral proteins assemble around the endonuclease nsp15 to form a hexameric complex of over 60 subunits (31). Our modeling suggested this complex efficiently coordinates the various functions just described. Implicitly, these proteins must be recruited to the DMV prior to its closure for replication inside the organelle to occur, although how that might be achieved is unknown, representing a major gap in our knowledge of coronavirus replication (6)(7)(8).
Here, we tackle the question of pore formation and cleavage of the polyprotein from a structural modeling approach. We first review our attempts at constructing a model of the nsp3/nsp4 pore structure, highlighting the successes and limitations of AlphaFold in such an endeavor. We ultimately propose modest changes to the modeled prongs in the published structure of Huang et al. (14). We then examine constraints on nsp5 protease activity imposed by the flanking membrane binding proteins nsp4 and nsp6 and describe our discovery of what appears to be a polyprotein precursor to pore formation which overcomes those constraints. We propose that when this pore precursor is formed from six heterodimers of pp1a' and pp1ab' , the cleaved products will be trapped within the DMV and have the proper stoichiometry to form the previously proposed hexameric replication complex. The implications of the proposed model are far-ranging, with direct relevance to membrane remodeling, regulation of nsp5 protease activity, and formation of the replication and transcription complex within the closed vesicles.
## RESULTS
## SARS-CoV-2 nsp3 monomer
At 1,945 residues, nsp3 is the largest protein produced by the SARS-CoV-2 virus. While most of the individual domains have been solved by X-ray crystallography, AlphaFold2.1 was used to predict the full-length structure (Fig. 1a). The majority of the protein, with some notable exceptions, was predicted with high confidence (Fig. S1), as measured by the predicted Local Distance Difference Test (pLDDT) (32), with most domains scoring in the 70-90 range. X-ray structures for Ubl1 (ubiquitin-like domain 1) (33), Mac1 (macro domain 1) (34), Mac3 (35), DPUP (domain preceding Ubl2 and PLpro) (36), Ubl2/PLpro (37), NAB (nucleic acid binding domain) (38), and βSM (betacoronavirus specific marker) (39) were clearly reproduced in the AlphaFold model. Underscoring the success of the method, following completion of this work, a structure (PDB 8ILC) became available for the SARS-CoV-2 C-terminal CoV-Y domain (Y2, Y3, and Y4) (40) that matched the AlphaFold prediction as well. Additional domains with no available structures for any coronavirus were predicted for TM1, 3Ecto, TM2 (a bundle more accurately described as transmembrane domain 2 or TMD2) and Y1. The pLDDT score for the region between βSM and Y1, which includes TM1-3Ecto-TMD2, was comparatively lower, at 64.5, but notably the residues immediately following βSM (1318-1364) were predicted to be α-helical, with residues 1318-1349 having amphipathic character. The residues prior to TM1 (1385-1400) were predicted to form a second amphipathic helix.
The most notable failure in the AlphaFold prediction was for the Mac2 domain, which deviated significantly from X-ray structures for the highly homologous SARS Mac2 domain (71% identity, 85% similarity) (41). This was unexpected, given the SARS structure's inclusion in the AlphaFold training data set, but it appeared to be unique to the prediction of full-length nsp3. Predictions of the isolated domain or the Mac2 and flanking domains were consistent with the SARS structures, putting its pLDDT score in line with other domains (Fig. S1). Prediction of the heterodimer of Mac2 with human Paip1 was also consistent with the existing SARS Mac2/Paip1 X-ray crystal structure (PDB 6YXJ) (42).
In general, variations in the predicted structure conformations, coupled with the pLDDT analysis, reflect significant flexibility between the domains (see Fig. S2). This is most striking for the highly variable region (HVR) between Ubl1 and Mac1 (residues 106-213), which was completely disordered in the AlphaFold prediction. In addition, the loops connecting Mac1 and Mac2 (residues 386-416) and connecting NAB and βSM (residues 1194-1242) are particularly long. Ultimately, these linkers provided a set of constraints on the maximum distance and range of motion between domains that we considered when examining the nsp3 component of the pore.
## SARS-CoV-2 nsp3 dodecamer
Early analysis of the SARS-CoV-2 and MHV pore structures determined from cryo-ET suggested that the cytosolic crown is composed of a hexamer of nsp3, with the Nterminus residing in the prongs (12,13). Due to practical limitations in the size of structures that could be predicted with AlphaFold (about 5,000 residues), it wasn't possible to generate a model for an entire nsp3 hexamer (11,670 residues) and exploration of truncated constructs had only limited success. The publication of the 4.2 Å cryo-ET structure, however, provided surprising clarity (14). The sixfold symmetric crown is instead a dodecamer of nsp3, although much of the protein structure was not resolved. Specifically, the C-terminal Y1/CoV-Y domains form the dodecameric base of the crown and the cytosolic portion of the pore. It is constructed from an inner hexame ric ring of Y1/Y2 (forming the pore), which is surrounded by another hexameric outer Y1/Y2 ring to create the base of the crown. The 12 CoV-Y Y3/Y4 domains interlace above the base, forming the bulk of the crown. The crown is embedded in the outer membrane through TMD2, interacting with nsp4. Here, we note that the clearest success among the AlphaFold predictions of multimeric nsp3 constructs was that of the inner Y1/Y2 hexameric ring, shown in Fig. S3 in comparison to the cryo-ET structure.
The N-terminal domains preceding TM1 are only partially resolved for six subunits and not resolved at all for the other six, implying a degree of disorder, which may reflect real dynamics of the pore or the lack of other key stabilizing elements, such as RNA or additional host or viral proteins. Alternatively, it may simply be an artifact of how the structure was obtained and refined (the cryo-ET structure was determined from DMVs derived from nsp3-nsp4 expression in HEK293F cells, not DMVs derived from infected cells). What is definitively resolved is that a hexameric ring of Ubl2/PLpro rests atop the CoV-Y dodecamer to build up the crown. Huang et al. ( 14) also fit each of the six prongs with the Mac2, Mac3, and DPUP domains from one subunit and the NAB domain from a neighboring subunit. However, the resolution of these prongs is considerably weaker (>9 Å), making the fit ambiguous. Several considerations led us to remodel this part of the structure. Huang et al. assumed the prong density was primarily composed of the Mac2, Mac3, and NAB domains and used AlphaFold to predict a complex of these domains, which they docked into the density. This fit put the NAB domain implausibly far from the assumed position of the βSM domain at the base of the crown. In contrast, they did not fit the Ubl1, HVR, or Mac1 domains, which Zimmermann et al. (13) previously showed was critical to the proper formation of the prongs. Considering that Mac1 corresponds to the only Mac domain in MHV nsp3, we reasoned that it was an important component of the prong density and was likely similarly positioned in both structures. Taking into account functional constraints, similarities and differences between MHV and SARS-CoV-2, and all available cryo-ET maps, including the prior maps for both MHV and SARS-CoV-2, we fit the prongs with Ubl1, Mac1, Mac2, Mac3, and DPUP, omitting only the HVR domain for lack of structural information. Furthermore, we positioned the NAB domain between PLpro domains in the rim of the crown. We also modeled the βSM domain and two amphipathic helices on the outer membrane cytosolic surface, following previously unfit density. As a proof of principle, we connected the NAB and βSM domains with the 44-residue linker. A similar model was built for the MHV nsp3 dodecamer to establish consistency across diverse coronaviruses. The final remodeled nsp3 crown is shown in Fig. 1b andc and described in greater detail in the supplemental material and Fig. S4, S5 and S6.
## SARS-CoV-2 nsp4 monomer
Our model of nsp4 (Fig. 2a) is consistent with previous AlphaFold models with four distinct domains (43). The N-terminal TM1 is a single α-helix which spans residues 1-27. This is followed by the lumenal domain, a soluble region composed of two distinct lobes connected by a 9-residue linker. The previously designated TM2/TM3/TM4 helices (44) were instead predicted by AlphaFold to be a 5-helix bundle, spanning residues 277-398. For simplicity, we'll refer to this region as TMD2, with specific helices α1-α5. Finally, the AlphaFold predicted C-terminal domain (CTD) is another soluble region that reproduces existing X-ray structures (45). The average pLDDT score for the full protein was 83.0 (Fig. S7).
In none of the predicted structures does TM1 interact with TMD2. The conformation is instead consistent with nsp4 association with the DMV double membrane, where the nsp4 N-terminus is cytosolic, TM1 sits in the outer membrane, the lumenal domain is positioned in the membrane gap, TMD2 sits in the inner membrane, and the nsp4 CTD is exposed to the DMV interior. Thus, the overall topology of the AlphaFold predicted nsp4 structure is consistent with that seen in the detailed pore structure. Details of the predicted structure, such as the lumenal domain and TMD2, were also confirmed by the cryo-ET structure, with the only notable difference being a conformational shift of helix α1 in TMD2, to be discussed below in more detail.
## SARS-CoV-2 nsp4-nsp5-nsp6-nsp7 monomer
To investigate constraints on the nsp5 protease imposed by its position within the polyprotein, we generated AlphaFold models for the uncleaved nsp4-nsp5-nsp6-nsp7 construct (Fig. 2b). The pLDDT declined slightly to an average of 77.0 (Fig. S8), but X-ray structures of nsp5 (46) and nsp7 (47) were well reproduced, and nsp6 was predicted to form an 8-helix bundle. The overall organization of the polyprotein showed some flexibility between the individual domains. However, in all predicted structures, the nsp4 TMD2 was aligned with nsp6 as if bound to a common membrane. Unlike the prediction for cleaved nsp4, the lumenal domain of nsp4 in the polyprotein adopted a conforma tion between its N-and C-lobes (linked by residues 124-132) which oriented TM1 such that it associated with the nsp6 helix bundle (Fig. S9). This would be more consistent with association with the single ER membrane rather than the DMV double membrane, making nsp5, nsp7, the nsp4 N-terminus, and the nsp4 CTD cytosolic (48). The observa tion that this conformational change is driven by the nsp4 lumenal domain is consistent with previous characterizations of this domain by Klatte et al. (49).
The nsp5 protease dimer is known to be significantly more active than the monomer (20,21), but a major conclusion from this work is that despite a relatively long flexible loop connecting nsp4 and nsp5 (residues 488-510), cis-cleavage of these proteins does not appear to be feasible. The cleavage site on the connecting loop (residues 500-501) simply cannot reach the protease active site without compromising the integrity of the protease structure (Fig. 3a). Similarly, cis-cleavage of the nsp5 and nsp6 proteins at residues 807-808 does not appear to be possible under any circumstances. Thus, in both cases, cleavage must proceed via interaction with the protease of a second polyprotein or an already cleaved nsp5 dimer.
## SARS-CoV-2 nsp4-nsp5-nsp6-nsp7 dimer
The cleaved nsp5 dimer has been exhaustively studied through X-ray crystallography for multiple coronaviruses (50)(51)(52). A key factor in the increased protease activity of the dimer as compared to the monomer is how the N-terminal amine of one monomer engages with the S1 active site pocket of the other monomer. This has been shown to be critical to stabilizing the active site in both SARS (50,53,54) and SARS-CoV-2 (52), enhancing the protease's catalytic activity. We considered if a canonical nsp5 dimer could form within the uncleaved nsp4-nsp5-nsp6-nsp7 polyprotein, and if so, would it be capable of self-cleavage. Using a truncated construct, in which the nsp4 TM1 and lumenal domains were removed, we generated an AlphaFold model of the uncleaved dimer possessing approximately C2 symmetry (Fig. 3b). In this model, nsp4 and nsp6 are aligned as if bound to a common membrane, with the canonical nsp5 dimer oriented such that its domains 1 rest against the membrane and domains 3 are most exposed to the cytosol.
The nsp4-nsp5 cleavage site P1 Q500 residue can be seen to occupy the S1 pocket of the active site. Yet the full nsp4-nsp5 cleavage peptide is not properly aligned, and critically, the P1' S501 cleavage site residue cannot be positioned in the S1' pocket due to constraints from the nsp5 dimer itself (Fig. 3c). Thus, the dimer is incapable of cleaving nsp4/nsp5. To stress this point, a 100 ns molecular dynamics (MD) simulation of the membrane-bound dimer maintained this interaction between the glutamine residue and the S1 site over the entire simulation (Fig. S10). We conclude from this analysis that not only does the uncleaved dimer not have the advantage of the cleaved nsp5 N-terminus to stabilize the S1 site and activate the protease, it has the uncleaved nsp4 C-terminal glutamine residue blocking the site from acting on any other substrate.
If additional steric and conformational constraints from the extended polyprotein or potentially an nsp3 interaction prove to be more dominant, it should be considered that this canonical dimer may not even form until some cleavage has already occurred via a different route. Underscoring this point, the nsp5 protein of MERS only weakly dimerizes, depending instead on a ligand-induced dimerization mechanism (51). Indeed, as seen in Fig. 3d, taking the same nsp4-nsp5-nsp6-nsp7 dimer and assuming at least one nsp4/ nsp5 cleavage has occurred via some yet-to-be-determined mechanism, the resulting nsp5 N-terminal amine then activates the protease of the uncleaved subunit. The nsp5- nsp6 and nsp6-nsp7 cleavage sites of the cleaved monomer can be observed to be in close proximity to the activated protease and could be potential substrates. This analysis suggests that an initial nsp4-nsp5 cleavage event is required but does not proceed through the canonical nsp5 dimer. Some other association between polyproteins must be responsible.
## SARS-CoV-2 nsp4 dodecamer
As with nsp3, our initial assumption was that the nsp4 component of the pore was hexameric. Based on our predicted structures for the nsp4 monomer and previous work suggesting nsp3 interacts with the nsp4 lumenal domain (49,55), we reasoned TM1 must sit in the outer membrane, positioning the lumenal domain in the membrane gap. This suggested the portion of the pore embedded in the inner membrane would be formed from a hexamer of the nsp4 TMD2 domains, with the nsp4 CTD exposed to the DMV interior. Thus, we worked with a truncated construct which spanned residues 256- 500 of nsp4 (TMD2 and CTD). AlphaFold indeed predicted several plausible hexameric structures with pore-like properties, but none was entirely consistent with the available cryo-ET maps. However, upon close examination of the MHV map, we identified density within the membrane gap which was similar in size to the nsp4 lumen domain, but which indicated nsp4 was in fact dodecameric, not hexameric. Unfortunately, AlphaFold predictions of dodecameric nsp4 (residues 256-500) did not produce structures with the expected C6 symmetry. The detailed pore structure from Huang et al. (14) confirmed that the nsp4 component is indeed a dodecamer, characterized by a central hexameric ring interlaced with an outer hexameric ring. While failing to produce the full dodeca meric nsp4 assembly, AlphaFold in fact accurately predicted the structure of the inner hexameric ring, as shown in Fig. S11 in comparison to the cryo-ET structure.
## SARS-CoV-2 nsp4-nsp5-nsp6 dodecamer
As the initial nsp5 cleavage events appeared to be unresolved, we considered the possibility of pore formation prior to nsp4/nsp5 cleavage. Our preliminary exploration with AlphaFold also considered hexamers of uncleaved nsp4-nsp5-nsp6 (truncated to residues 256-1,042). AlphaFold generated a particularly compelling model shown in Fig. 4a and Fig. S12, which consisted of an inner ring of nsp4 TMD2 surrounded by an outer ring of nsp6. The two transmembrane domains are linked via an inner, close-packed solvent exposed ring formed from the nsp4 CTD, which is surrounded by a larger close-packed ring formed from nsp5. Consistent with the emerging topology of the nsp3/nsp4 pore, both nsp5 and the nsp4 CTD would be exposed to the interior of the DMV in this model.
As we came to understand the stoichiometry of the nsp4 component of the pore was dodecameric instead of hexameric, we found it was straightforward to envision similar close-packed polyprotein rings of increasing size. As an example, the chikungunya virus protein nsP1 has been shown via cryo-EM to form a dodecameric ring (56). We used AlphaFold to look at a series of smaller aggregates (hexamer, octamer, decamer), and a similarly arranged ring was predicted in each case (see Fig. S13a). Unfortunately, a dodecameric arrangement of nsp4-nsp5-nsp6 or any ring larger than a hexamer was beyond the practical limits of AlphaFold. To generate such a model, we instead resorted to a simple geometric expansion of the nsp4-nsp5-nsp6 hexamer. This was executed by defining the center of mass of each subunit within the hexamer, measuring the distance between adjacent subunits, and translating and rotating duplicate sets of these subunits to form a dodecameric ring in which this distance was maintained. Structural refinement from this starting point (as detailed below) led to the fully optimized construct shown in Fig. 4b. A similar approach was applied to the AlphaFold model of hexameric chikungu nya nsP1 to yield a dodecameric model that reproduces the cryo-EM structure well (Fig. S13b).
In comparison to the earlier MHV and SARS-CoV-2 cryo-ET maps, this dodecameric arrangement appeared to be consistent with the overall diameter of the inner mem brane pore and provided the first hint as to how these pores may form. Once the detailed pore structure of Huang et al. (14) became available, the arrangement of the final fully cleaved pore was clear, and our attention turned to whether the uncleaved model was consistent with this structure.
As described above, the nsp4 dodecamer component of the pore structure can be described as an inner hexameric ring interlaced with an outer hexameric ring. Based on the locations of the C-terminal domains, a clear requirement for this structure to form is that the inner nsp4 hexamer must be cleaved from nsp5. However, it's equally clear that the outer hexameric ring can remain part of the uncleaved polyprotein and retain the integrity of the pore structure. Thus, maturation of the pore within the framework of this model would at minimum require the cleavage of alternating nsp4/nsp5 sites.
This presented an interesting topological question and an opportunity to reverse engineer the process. In the nsp4-nsp5-nsp6 hexamers, we saw a domain swapping occur in which TMD2 adopts an open conformation with the α1 helix jutting out at a significant angle (~60°) to interact with a neighboring TMD2 bundle of α2-α5. This open conformation (see Fig. S14a) is also seen in both the inner and outer hexamers of the structure solved by Huang et al., where domain swapping is clearly observed in the inner hexamer. Furthermore, examining an interlaced pair of nsp4 subunits from this structure, the open conformation has the effect of projecting one subunit inward and the other outward, causing the two to cross each other (Fig. S14b).
We recognized that there was some ambiguity as to how this may manifest itself when converting the AlphaFold predicted nsp4-nsp5-nsp6 hexamer to a dodecamer. We considered three possibilities: one in which all 12 subunits adopt the open confor mation, one in which they all adopt the closed conformation, and one in which only alternating subunits adopt the open conformation (Fig. S14c). The overall structure was unchanged except for whether the α1 helix was open or closed. From these three possibilities, it could be seen that the structure with all 12 subunits in the open conformation does not have a direct path to the interlaced final pore structure upon nsp4/nsp5 cleavage. However, in an arrangement in which only alternating subunits adopt the open conformation, upon cleavage of the nsp4-nsp5 linkages from those The dimer incorporates known interactions between the nsp proteins to the maximum extent. The nsp4 conformation is consistent with coordination to a double membrane and formation of the dodecameric ring. (d) Model of the complete DMV immature pore, prior to nsp4-nsp5 cleavage. The precursor to the pore is formed from 12 subunits of nsp3 (dark blue) and six subunits each of uncleaved pp1a' and pp1ab' . same six subunits, the structure would have a simple pathway to adopt the mature pore conformation (see the movie in the supplemental material). From this perspective, the model shown in Fig. 4b may be considered a hexameric ring of polyprotein dimers, mirroring the situation for the nsp3 dodecamer.
This model was ultimately merged with the nsp4 lumenal domains of the Huang et al. structure and our rebuilt model of the nsp3 dodecamer. The nsp4 TM1 helices, missing from the cryo-ET structure, were also added in parallel to the nsp3 TM1 helices, consistent with discernible, but previously unmodeled, density in the maps.
## SARS-CoV-2 uncleaved pp1a'/pp1ab' model
Having created a model of dodecameric nsp4-nsp6, we considered modeling the remainder of the uncleaved polyprotein. This extended model could be derived from either pp1a' (nsp4-nsp10) or pp1ab' (nsp4-nsp16) or a combination of both. We considered that certain known protein-protein interactions may be preserved in this situation, creating a degree of preorganization that would facilitate efficient formation of the replication complex upon polyprotein cleavage. From cryo-EM and X-ray structures, interactions between nsp7 and nsp8, nsp7 and nsp12, nsp8 and nsp12, nsp8 and nsp13, nsp9 and nsp12, nsp10 and nsp14, and nsp10 and nsp16 are well characterized (57-60), as is nsp15 hexamerization (61).
Working from a dimer of nsp4-nsp5-nsp6 extracted from the optimized dodecameric ring, it appeared that extending both subunits to the full pp1ab' polyprotein would be too bulky and provide only limited interactions between the subunits. In contrast, extending both to the smaller pp1a' polyprotein would be well tolerated, but offer little in additional protein-protein interactions. However, a heterodimer formed from pp1a' and pp1ab' offered the maximum opportunity to incorporate the known protein-protein interactions.
Initial exploration of polyprotein flexibility was conducted with AlphaFold on multiple constructs. The uncleaved protein dimers (e.g., nsp9-nsp10, nsp10-nsp12, etc.) revealed conformational flexibility between all pairs of proteins (Fig. S15), driven by typically unstructured linkers. This led to exploration with larger fragments, with particular attention to forming interactions that are already known. We extended the initial nsp4-nsp5-nsp6 dimer to include nsp7, consistent with a conformation seen in the AlphaFold prediction of monomeric nsp4-nsp5-nsp6-nsp7. We then found we could extend one of the subunits to include nsp8 and have that protein interact with nsp7 from the other subunit just as it does in the cryo-EM structures of the replication complex (e.g., 6XEZ) (57). Furthermore, we found we could associate nsp13 with this nsp7/nsp8 pair, again as seen in the replication complex structure (Fig. S16a). This immediately established constraints on where nsp9, nsp12, nsp14, and the second nsp8 would need to connect.
AlphaFold predictions of uncleaved nsp8-nsp9-nsp10-nsp12 led to one of the largest building blocks of the polyprotein heterodimer. The predictions showed variability in the positioning of nsp9 and nsp10 but consistency in the interaction between nsp8 and nsp12 (Fig. S16b). This mirrored the replication complex as well and put constraints on how this string of proteins from pp1ab' would be placed relative to nsp7 and nsp13. As both nsp14 and nsp16 are known to coordinate to nsp10, we further aimed to find a conformation that would allow both protein pairs to be satisfied. We proposed that one of these proteins would coordinate cis to nsp10 within the larger pp1ab' polyprotein and the other would coordinate trans to the terminal nsp10 of pp1a' . We examined both possibilities with respect to binding to the nsp8-nsp9-nsp10-nsp12 polyprotein. We concluded that only coordination of nsp16 following existing X-ray crystal structures (6W4H) (60) had the potential to satisfy the constraints of linking all the subunits. We then coordinated nsp14 and a second subunit of nsp10 to nsp12, following our previous model for the full hexameric replication complex (see Fig. S16c) (31).
These two large complexes (nsp4-nsp5-nsp6-nsp7-nsp8/nsp4-nsp5-nsp6-nsp7/ nsp13 and nsp8-nsp9-nsp10-nsp12/nsp10/nsp14/nsp16) were reasonably consistent with each other. We found they could be easily linked and optimized, connecting nsp7 to nsp8, nsp12 to nsp13, and nsp13 to nsp14. This meant only nsp9 was needed to link nsp8 and nsp10 of pp1a' and nsp15 was needed to link nsp14 and nsp16 of pp1ab' (see Fig. S16d). Protein-protein docking with Piper (62) was used to explore initial placement of the final proteins. The most promising structures were then optimized with constraints to minimize the linkage distances before finally linking the completed polyproteins and doing a refinement of the full protein heterodimer. In the case of the nsp14-nsp15 and nsp15-nsp16 linkage, significant freedom was given to nsp9-nsp10/ nsp16 to adopt a conformation that would facilitate the final linkage. The final model is shown in Fig. 4c. While the conformational space of the two polyproteins is enormous and could not be exhaustively studied, in the end, we established that multiple known protein-protein interactions could be preserved in the polyprotein heterodimer. These include pp1a'-nsp8/pp1ab'-nsp7; pp1a'-nsp8/pp1ab'-nsp13; pp1a'-nsp10/pp1ab'-nsp14; pp1ab'-nsp8/pp1ab'-nsp12; and pp1ab'-nsp10/pp1ab'-nsp16. This heterodimer was then incorporated into the dodecameric model where it went through an additional round of refinement to remove any clashes between the various components (Fig. 4d).
## SARS-CoV-2 polyprotein cleavage
The close-packed arrangement of nsp5 in the polyprotein dodecameric ring establishes a favorable environment for the protease to make the initial nsp4-nsp5 cleavage. A key observation we made was that the nsp4-nsp5 linker (residues 488-510) was long enough that it could conceivably connect nsp4 to either of two nsp5 subunits in the ring (at a distance of 35 or 52 Å). As originally constructed from the AlphaFold hexamer, the connection was to the nearer nsp5-nsp6 subunit, comparable to how the proteins are arranged in the monomeric polyprotein. However, connecting nsp4 instead to the neighboring nsp5-nsp6 subunit allowed the nsp4-nsp5 cleavage site to pass over the proximal nsp5 protease active site. A positional shift of the nsp5 proteins would still be required for cleavage to occur, and this was investigated for both subunits. We found that simply by shifting alternating nsp5 subunits while retaining the positions of everything else, we could identify a conformation that was capable of nsp4-nsp5 cleavage, consistent with available crystal structures of bound substrate (63). As depicted in Fig. 5a andb, the shifted nsp5 subunit functions as the substrate, while the nsp5 subunit that retained its original position functions as the protease. Given the constraints imposed by the full (pp1a'/pp1ab') 6 model, a repositioning of nsp5 associated with pp1ab' appears more likely than that of pp1a' suggesting the larger polyprotein is the initial substrate for cleavage.
As previously described, the protease activity of nsp5 is known to be significantly enhanced when the protein is in its canonical dimer form, in large part due to stabiliza tion of the S1 pocket by the N-terminal amine of the paired subunit (52). A striking observation from these AlphaFold-derived models was that the C-terminal domain of nsp4 forms an interaction with nsp5 which appears to mimic this aspect of nsp5 dimerization (Fig. 5c). Specifically, a conserved nsp4 basic residue (K452) forms a salt bridge with the conserved nsp5 glutamic acid (E666), replicating the interaction seen in the dimer between the same glutamic acid and the N-terminus of the paired subunit. The similar interaction seen here suggests formation of the pore prior to polyprotein cleavage both activates the nsp5 protease and positions the nsp4-nsp5 substrate to carry out the first cleavage event.
This proposed arrangement of alternating positions of nsp5 in the dodecameric ring would lead to the cleavage of half of the nsp4-nsp5 linkages. Once this occurs, the reduced constraints allow the nsp3/nsp4 pore to adopt a conformation consistent with that of the cryo-ET structure. That is, the six cleaved nsp4 proteins rotate inward to form the central hexameric pore. With this reconfiguration, the six remaining uncleaved nsp4 subunits can rotate outward to adopt the conformations seen in the cryo-ET structure. At the same time, the six partially cleaved nsp5 subunits can now form canonical nsp5 dimers with the six uncleaved nsp5 subunits, as shown in Fig. 5d ande. In this case, the protease site in the uncleaved nsp5 subunit would be considered activated by the N-terminal amine from the cleaved subunit, as previously described. From this configura tion, cleavage of additional linkages could take several pathways, with nsp5-nsp6, nsp6-nsp7, and nsp8-nsp9 linkages all within immediate range (Fig. S17).
## DISCUSSION
The events that lead to nsp5 protease cleavage of the coronavirus ORF1 polyproteins pp1a and pp1ab are at present illdefined. Furthermore, there is currently no clear hypothesis for how the resulting cleavage products, which constitute the replication machinery, are recruited to the closed DMV replication organelle. Cleavage of nsp1, nsp2, and nsp3 from the polyprotein occurs very quickly, facilitated by the nsp3 PL proteases (17). Cleavage of the remainder of the polyprotein by nsp5 appears to be a slower process, possibly requiring dimer formation or some other regulatory event to initiate (64). As part of an investigation into the structure of the SARS-CoV-2 DMV nsp3/nsp4 pore, we were struck by a particular AlphaFold-generated model of a ring formed by a multimer of uncleaved nsp4-nsp5-nsp6 polyproteins. We considered the possibility that such a structure may represent a precursor to formation of the DMV pore in an infected cell. From this starting point, we generated a series of structural models outlining a plausible path for DMV and pore formation stemming from the interaction of 12 nsp3 subunits with six pp1a' and six pp1ab' uncleaved polyproteins. We argue that it is the formation of an immature pore from these components that initiates the protease activity of nsp5 and creates a situation where the cleaved products would be positioned within the DMV prior to its closure. We made several key observations from this modeling effort. First, analysis of the nsp4-nsp5-nsp6-nsp7 monomer made clear that cis-cleavage of the nsp4-nsp5 and nsp5-nsp6 linkages is not possible, implying cleavage is facilitated by some interaction between polyproteins. Based on a great deal of biochemical and crystallographic work on cleaved constructs (18,19,21,50,52,54), it is well established that nsp5 is signifi cantly more active as a dimer. But further analysis of the nsp4-nsp5-nsp6-nsp7 dimer indicated that the constraints imposed by the flanking membrane binding proteins nsp4 and nsp6 effectively render the canonical nsp5 dimer inactive within the context of the polyprotein. Such a dimer can only be considered active following nsp4-nsp5 cleavage of at least one of the two subunits. This leads to the conclusion that some alternative interaction between the polyproteins is responsible for the initial nsp4-nsp5 cleavage event.
Second, the uncleaved polyproteins pp1a' and pp1ab' can form a favorable hetero dimer in which known protein-protein interactions are established. While pp1a' and pp1ab' homodimers could also be imagined, neither would have the advantages of the complementarity seen in the heterodimer. As driven by a conformational change in the lumenal domain of nsp4, the pp1a'/pp1ab' dimer is capable of binding to either a single membrane or a double membrane. When bound to the single membrane of the ER, the nsp4 N-terminus and CTD, nsp5, nsp7-nsp10 of pp1a' and nsp7-nsp16 of pp1ab' are cytosolic, while only the nsp4 lumen domain is in the ER interior. When bound to the double membrane of a DMV, the nsp4 N-terminus is the only element remaining exposed to the cytosol. With the nsp4 lumen domain sitting in the membrane gap, the bulk of the polyprotein dimer then resides in the DMV interior.
These heterodimers, when in the double membrane spanning conformation, are then capable of forming a dodecameric ring comprised of alternating pp1a' and ppa1b' polyproteins. This ring has a diameter comparable to that seen for the nsp4 dodeca mer in the recent nsp3/nsp4 pore structure reported by Huang et al. (14) and can associate with the nsp3 dodecamer in the same manner. Thus, we propose that this (nsp3) 12 (pp1a') 6 (pp1ab') 6 complex represents an immature form of the pore.
The polyprotein dodecameric ring leads to close-packing of nsp5 and the nsp4 CTD, providing the necessary scaffolding to facilitate trans-cleavage of the nsp4-nsp5 linkage. We identified a conformation within this framework in which the 23-residue nsp4-nsp5 linker of alternating subunits engages the proximal nsp5 protease active site. Critically, we also noticed that a conserved basic residue from the nsp4 C-terminal domain (K452) forms a salt bridge with the nsp5 glutamate residue (E666) located in the protease active site S1 pocket. This interaction mimics the situation in the nsp5 dimer, in which the same glutamate residue interacts with the N-terminal amine of the other subunit. It is thought to be critical to stabilization of the protease active site, explaining the significant enhancement of protease activity associated with dimerization (54). Thus, we conclude that the formation of this immature pore activates the nsp5 protease and sets it up for the initial nsp4-nsp5 cleavage.
Following this initial nsp4-nsp5 cleavage event, the six cleaved nsp4 proteins and the six still uncleaved nsp4 proteins are free to adopt the conformations seen in the mature pore structure. The reduced structural constraints also allow formation of canonical nsp5 dimers in which one of the two protease sites would be activated. With these nsp5 dimers having proximity to additional cleavage sites within the polyprotein, the remainder of the polyprotein can be processed, leading to the final fully cleaved (nsp3) 12 (nsp4) 12 pore.
With this model, we found that pore formation prior to nsp5 polyprotein processing is a reasonable possibility and may in fact be the regulatory step needed to ensure proper localization of the cleaved proteins that make up the replication machinery. A notable observation of the full model presented here is that the stoichiometry is consistent with our previous model for the hexameric replication complex (31). That model included 60+ subunits centered around nsp15. Specifically, the model includes six subunits of nsp7, nsp12, nsp13, nsp14, nsp15, and nsp16, and 12 subunits of nsp8 and nsp10, with up to six transiently associated nsp9 subunits. This complex is most efficiently derived from six pp1a and six pp1ab polyproteins, meaning the precursor presented here simultaneously produces one pore structure and generates one full replication complex. When viewed together, the two models imply a mechanism not only for DMV and pore formation, but for recruitment of the replicase to the DMV.
As depicted in Fig. 6, the following picture emerges from these models. Upon cleavage by the PL proteases, nsp3 then associates with the nsp4 component of either pp1a' or pp1ab' on the ER membrane (Fig. 6a). The nsp3 3Ecto and nsp4 lumenal domains are in the ER lumen, with the bulk of the proteins exposed to the cytosol. The association between nsp3 and nsp4 then induces double membrane formation, with aggregation into a tetrameric unit (nsp3) 2 (pp1a')(pp1ab') following (Fig. 6b). The exact sequence of events is unclear, but this pathway seems likely given the immedi ate proximity between nsp3 and either pp1a' or pp1ab' following PLpro cleavage and the observation that the nsp3/nsp4 interaction drives membrane rearrangement (55). Once the double membrane forms, the nsp3 3Ecto and the nsp4 lumen domains are positioned in the gap between the two membranes, and the remainder of nsp3 is on one side of the double membrane and nsp5, nsp7, and the rest of the polyprotein are on the other side. Further aggregation of these tetrameric building blocks leads to the immature pore and may facilitate the observed increased curvature of the double membrane (Fig. 6c). Formation of additional pores will ultimately yield a sealed, spherical DMV.
The picture of membrane remodeling presented here raises the question of the relative timing of nsp5 polyprotein processing and DMV formation. We see no justifica tion for a strict requirement that a DMV be fully formed for cleavage to begin. Should cleavage occur prior to DMV formation, the nsp products would be released into the cytosol. We note that in this situation, cleaved nsp6 could migrate back to the ER surface where it would be available for membrane zippering, a key step in forming the connec tors for DMV clusters (10). However, should cleavage follow closure of the DMV, the resulting products would be trapped inside (Fig. 6d). This is a scenario that would likely follow the formation of the final pore of the DMV, suggesting each DMV contains at minimum one hexameric replication complex. Interestingly, this process may ultimately become self-limiting, as free nsp5 dimers build up in the cytosol, cleavage of new polyproteins at late stages may occur without the requirement of new pore formation.
The proposed mechanism of regulated protease activation to ensure proper localiza tion of the viral proteins is similar to the well-established mechanism of HIV protease maturation (65,66). In that case, the protease resides on the gag-pol polyprotein and requires dimerization to be enzymatically active. While the cleaved protease rapidly dimerizes in a biochemical setting, dimerization and protease activation of the polypro tein only occur within the immature viral particle. This ensures that the products (which include the reverse transcriptase and the integrase) are contained within the virion and not released into the cytosol.
The implications of this model require careful consideration. Within the protected environment of the DMV, the proposed stoichiometry, pre-organization and high protein concentration (we estimate at least 1-2 µM based on a DMV diameter of 250-350 nm [11,12,67]) would be ideal for formation of a long-lived full hexameric replicase complex. As previously described, this complex should ensure that replication, proofreading, capping, and dsRNA unwinding are coordinated and efficient processes (31). Confirmation of the exact contents of DMVs has been notoriously difficult, however. Multiple studies have shown that the DMVs contain RNA, specifically identifying both dsRNA and newly synthesized RNA (3,7,8,11,67). Identification of viral protein components of the replicase within DMVs has been less conclusive, although increasing evidence supports their association with DMVs (7,8).
In contrast, the less controlled environment outside of DMVs may mean the full complex does not form or that complexes are transient. The minimally known replica tion-competent complex is formed from nsp12, nsp7, and two subunits of nsp8 (68,69). Association with nsp9 (58,70) and two subunits of nsp13 (26,57) has been documen ted as well, but characterization of other viral protein interactions has proven more elusive. Indeed, the proposed dual role of nsp15 suggests a full hexameric complex forms within the DMV but likely does not in the cytosol. Within the DMV, endonuclease activity would need to be regulated to prevent degradation of the RNA. This would be the case for the hexameric replication complex, where nsp15 is proposed to function primarily as scaffolding, with its endonuclease activity highly curtailed (31). However, in the cytosol, nsp15's endonuclease activity serves as an effective defense against the immune response by degrading cytosolic RNA (61,71), suggesting that it exists as a free hexamer and is not regulated by the larger replication complex.
In the absence of full complex formation, some functions, such as proofreading and mRNA capping, might be compromised or lost entirely. Yet plenty of evidence remains that some replication is occurring outside of the DMVs. At minimum, there is a temporal argument for this, in that replication necessarily precedes DMV formation (72). However, other lines of evidence are consistent with a two-compartment model persisting throughout infection. Notably, the groundbreaking work of Sawicki and Sawicki (73) established that coronavirus RNA replication is dependent on continuous ORF1 translation and proteolytic processing. Further work on several temperature-sen sitive mutations showed that synthesis of negative-sense RNA was halted at higher temperatures but positive-sense RNA was not (74). Donaldson et al. (75) showed that this was a result of inhibited nsp5 protease activity at the higher temperatures. More recently, Schmidt et al. (76) demonstrated that negative strand synthesis was dependent on UMPylated-nsp9 priming, facilitated by the host factor SND1. This did not appear to be the case for positive strand synthesis. With no proposed mechanism for recruitment of additional proteins to an already formed DMV, the dependence of negative-sense RNA synthesis on host factors and the continuous generation of viral replicase proteins (specifically nsp9) suggest transcription of the minus strand occurs exclusively in the cytosol. In contrast, this model would be consistent with positive strand synthesis occurring in the cytosol at early stages but predominantly occurring within the protected environment of DMVs at later stages.
In summary, in asking the question whether the nsp3/nsp4 DMV pore forms prior to nsp5 cleavage of the polyprotein, we have developed a simple framework for under standing how the replication complex is recruited to DMVs in coronavirus-infected cells. Our model directly links the dodecameric nsp3/nsp4 pore structure and the previously proposed 60+ subunit hexameric replication complex to the same polyprotein inter mediate, elucidating how the nsp5 protease is regulated to ensure proper localization of the replicase in the DMV. From a close examination of structural constraints within monomeric and multimeric forms of the polyprotein, aided in large part by AlphaFold predictions, we conclude that formation of an immature pore from 12 nsp3 subunits and six subunits each of the intermediate polyproteins pp1a' and pp1ab' initiates the protease activity of nsp5. Following cleavage of half of the nsp4-nsp5 linkages, the mature nsp3/nsp4 pore can form, and protease cleavage of the remainder of the polyproteins will release the components necessary to form the full replication complex into the DMV. While the detailed mechanism is specific to coronaviruses, we expect similar processes are taking place with other single-stranded (+) RNA viral families.
## MATERIALS AND METHODS
AlphaFold2.1 (77) (with a template cutoff date of 15 February 2021) was used to predict the structures of individual monomers of nsp3, nsp4, and uncleaved nsp4-nsp5-nsp6-nsp7 from SARS-CoV-2, as well as hexameric complexes of the C-termi nal Y1/CoV-Y domain of nsp3, hexameric complexes of the TMD2/CTD domains of nsp4 (residues 256-500), hexameric complexes of uncleaved nsp4-nsp5-nsp6 (residues 256-1042), dimeric complexes of uncleaved nsp4-nsp5-nsp6-nsp7, and various fragments and heterodimers of nsp3 and the pp1a' and pp1ab' polyproteins.
The SARS-CoV-2 sequences used are from the original Wuhan strain (NCBI acces sion number QHD43415.1 [78]). Nsp3 spans residues 819-2763 of the polyprotein, nsp4 spans 2764-3263, pp1a' (nsp4-nsp10) spans 2764-4392, and pp1ab' (nsp4-nsp16) spans 2764-7096. The MHV sequence used in the construction of the nsp3 dodecamer was from NCBI sequence AAX23975.1 (79). Additional detail on the constructs explored can be found in the supplemental material. In all cases, AMBER relaxation was turned off and structures were instead refined with software from the Schrödinger Suite, including Prime, Macromodel, and Desmond (80). In all cases, the OPLS4 force field was employed (81). A typical workflow involved construction of an initial model either from AlphaFold directly or via protein-protein docking with Piper (82) or manual manipulation of individual subunits, followed by sidechain prediction in Prime. This was followed by a stepwise minimization with Prime and/or Macromodel, in which residues at the protein-protein interface were optimized first, building up to a minimization of the full structure. In cases where large movements from the starting point occurred, a second round of optimization would be carried out starting from a structure in which the original protein subunits were re-aligned to their new positions. Stability was tested at several stages of model development with MD using Desmond, typically run for 100 ns at constant temperature and pressure (NPT) with suitable constraints. SPC water was used in these simulations, as well as a POPC membrane, if appropriate. Macromodel and Desmond were also used to predict conformations of some long loops. Construction of the pp1a'/pp1ab' dimer involved preservation of known interactions as much as possible, using AlphaFold and Piper protein-protein docking to explore possible arrangements of various subunits. Model quality was assessed with Schrödinger's Protein Reports and by MolProbity (83).
Models were guided by the cryo-ET maps deposited in the Electron Microscopy Data Bank with accession codes EMD-11514 (MHV [12]), EMD-15963 (SARS-CoV-2 [13]), and EMD-39107 and EMD-39109 (SARS-CoV-2 [14]). Ultimately, the models were merged with the structures of Huang et al. (PDB 8YAX and 8YB5) (14) and further refined using the approaches described above. While the final pore models have approximate C6 rotational symmetry, strict symmetry constraints were not imposed.
## References
1. Wolff, Bárcena (2021) "Multiscale electron microscopy for the study of viral replication organelles" *Viruses*
2. Wolff, Melia, Snijder et al. (2020) "Double-membrane vesicles as platforms for viral replication" *Trends Microbiol*
3. Snijder, Limpens, De Wilde et al. (2020) "A unifying structural and functional model of the coronavirus replication organelle: tracking down RNA synthesis" *PLoS Biol*
4. Sergio, Ricciardi, Guarino et al. (2024) "Membrane remodeling and trafficking piloted by SARS-CoV-2" *Trends Cell Biol*
5. Roingeard, Eymieux, Burlaud-Gaillard et al. (2022) "The double-membrane vesicle (DMV): a virus-induced organelle dedicated to the replication of SARS-CoV-2 and other positivesense single-stranded RNA viruses" *Cell Mol Life Sci*
6. Wang, Han, Roy et al. (2022) "Multi-color super-resolution imaging to study human coronavirus RNA during cellular infection" *Cell Rep Methods*
7. Chen, Lupan, Quek et al. (2024) "A coronaviral pore-replicase complex links RNA synthesis and export from double-membrane vesicles" *Sci Adv*
8. Andronov, Han, Zhu et al. (2024) "Nanoscale cellular organization of viral RNA and proteins in SARS-CoV-2 replication organelles" *Nat Commun*
9. Angelini, Akhlaghpour, Neuman et al. (2013) "Severe acute respiratory syndrome coronavirus nonstructural proteins 3, 4, and 6 induce double-membrane vesicles" *MBio*
10. Ricciardi, Guarino, Giaquinto et al. (2022) "The role of NSP6 in the biogenesis of the SARS-CoV-2 replication organelle" *Nature*
11. Knoops, Kikkert, Worm S Van Den et al. (2008) "SARS-coronavirus replication is supported by a reticulovesicular network of modified endoplasmic reticulum" *PLoS Biol*
12. Wolff, Limpens, Zevenhoven-Dobbe et al. (2020) "A molecular pore spans the double membrane of the coronavirus replication organelle" *Science*
13. Zimmermann, Zhao, Makroczyova et al. (2023) "SARS-CoV-2 nsp3 and nsp4 are minimal constituents of a pore spanning replication organelle" *Nat Commun*
14. Huang, Wang, Zhong et al. (2024) "Molecular architecture of coronavirus double-membrane vesicle pore complex" *Nature*
15. Sawicki, Sawicki, Siddell (2007) "A contemporary view of coronavirus transcription" *J Virol*
16. Harcourt, Jukneliene, Kanjanahaluethai et al. (2004) "Identification of severe acute respiratory syndrome coronavirus replicase products and characteriza tion of papain-like protease activity" *J Virol*
17. Graham, Denison (2006) "Replication of murine hepatitis virus is regulated by papain-like proteinase 1 processing of nonstructural proteins 1, 2, and 3" *J Virol*
18. Li, Qi, Teng et al. (2010) "Maturation mechanism of severe acute respiratory syndrome (SARS) coronavirus 3C-like proteinase" *J Biol Chem*
19. Chen, Jonas, Shen et al. (2010) "Liberation of SARS-CoV main protease from the viral polyprotein: N-terminal autocleavage does not depend on the mature dimerization mode" *Protein Cell*
20. Roe, Junod, Young et al. (2021) "Targeting novel structural and functional features of coronavirus protease nsp5 (3CLpro, Mpro) in the age of COVID-19" *J Gen Virol*
21. Noske, Nakamura, Gawriljuk et al. (2021) "A crystallographic snapshot of SARS-CoV-2 main protease maturation process" *J Mol Biol*
22. Oudshoorn, Rijs, Limpens et al. (2017) "Expression and cleavage of middle east respiratory syndrome coronavirus nsp3-4 polyprotein induce the formation of double-membrane vesicles that mimic those associated with coronaviral RNA replication" *MBio*
23. 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*
24. Hillen, Kokic, Farnung et al. (2020) "Structure of replicating SARS-CoV-2 polymerase"
25. Park, Osinski, Hernandez et al. (2025) *Full-Length Text Journal of Virology*
26. Tagliabracci (2022) "The mechanism of RNA capping by SARS-CoV-2" *Nature*
27. Chen, Wang, Malone et al. (2021) "Ensemble cryo-electron microscopy reveals conformational states of the nsp13 helicase in the SARS-CoV-2 helicase replication-transcription complex" *Biophysics*
28. Malone, Chen, Wang et al. (2021) "Structural basis for backtracking by the SARS-CoV-2 replica tion-transcription complex" *Proc Natl Acad Sci*
29. Bouvet, Imbert, Subissi et al. (2012) "RNA 3'end mismatch excision by the severe acute respiratory syndrome coronavirus nonstructural protein nsp10/nsp14 exoribonuclease complex" *Proc Natl Acad Sci*
30. Ogando, Kazzi, Zevenhoven-Dobbe et al. (2021) "Structure-function analysis of the nsp14 N7-guanine methyltransferase reveals an essential role in Betacoronavirus replication" *Proc Natl Acad Sci*
31. Decroly, Debarnot, Ferron et al. (2011) "Crystal structure and functional analysis of the SARScoronavirus RNA cap 2'-O-methyltransferase nsp10/nsp16 complex" *PLoS Pathog*
32. Perry, Appleby, Bilello et al. (2021) "An atomistic model of the coronavirus replication-transcription complex as a hexamer assembled around nsp15" *J Biol Chem*
33. Mariani, Biasini, Barbato et al. (2013) "lDDT: a local superposition-free score for comparing protein structures and models using distance difference tests" *Bioinformatics*
34. Stogios (2020) "Crystal structure of the ubiquitin-like domain 1 (Ubl1) of Nsp3 from SARS-CoV-2. Protein Data Bank"
35. Brosey, Houl, Katsonis et al. (2021) "Targeting SARS-CoV-2 Nsp3 macrodomain structure with insights from human poly(ADP-ribose) glycohydrolase (PARG) structures with inhibitors" *Prog Biophys Mol Biol*
36. Li, Zhong, Li et al. (2022) "Two binding sites of SARS-CoV-2 macrodo main 3 probed by oxaprozin and meclomen" *J Med Chem*
37. Gallo, Tsika, Fourkiotis et al. (2022) "NMR solution structure of SUD-C domain of SARS-CoV-2. Protein Data Bank"
38. Calleja, Kuchel, Lu et al. (2022) "Insights into drug repurposing, as well as specificity and compound properties of piperidine-based SARS-CoV-2 PLpro inhibitors" *Front Chem*
39. Stogios (2021) "Crystal structure of the nucleic acid binding domain (NAB) of Nsp3 from SARS-CoV-2. Protein Data Bank"
40. Stogios (2021) "Crystal structure of the Nsp3 bSM (betacoronavirus specific marker) domain from SARS-CoV-2. Protein Data Bank"
41. Li, Pustovalova, Shi et al. (2023) "Crystal structure of the CoV-Y domain of SARS-CoV-2 nonstructural protein 3" *Sci Rep*
42. Tan, Vonrhein, Smart et al. (2009) "The SARS-unique domain (SUD) of SARS coronavirus contains two macrodomains that bind Gquadruplexes" *PLoS Pathog*
43. Lei, Ma-Lauer, Han et al. (2021) "The SARS-unique domain (SUD) of SARS-CoV and SARS-CoV-2 interacts with human Paip1 to enhance viral RNA translation" *EMBO J*
44. Yan, Zheng, Zeng et al. (2022) "Structural biology of SARS-CoV-2: open the door for novel therapies" *Sig Transduct Target Ther*
45. Sparks, Lu, Denison (2007) "Genetic analysis of murine hepatitis virus nsp4 in virus replication" *J Virol*
46. Xu, Lou, Ma et al. (2009) "Crystal structure of the C-terminal cytoplasmic domain of non-structural protein 4 from mouse hepatitis virus A59" *PLoS One*
47. Jin, Du, Xu et al. (2020) "Structure of M pro from SARS-CoV-2 and discovery of its inhibitors" *Nature*
48. Zhang, Li, He et al. (2021) "Nonstructural protein 7 and 8 complexes of SARS-CoV-2" *Protein Sci*
49. Oostra, Lintelo, Deijs et al. (2007) "Localization and membrane topology of coronavirus nonstructural protein 4: involvement of the early secretory pathway in replication" *J Virol*
50. Klatte, Shields, Agoni (2022) "Modelling the transitioning of SARS-CoV-2 nsp3 and nsp4 lumenal regions towards a more stable state on complex formation" *Int J Mol Sci*
51. Yang, Yang, Ding et al. (2003) "The crystal structures of severe acute respiratory syndrome virus main protease and its complex with an inhibitor" *Proc Natl Acad Sci*
52. Tomar, Johnston, John et al. (2015) "Ligand-induced dimerization of middle east respiratory syndrome (MERS) coronavirus nsp5 Protease (3CLpro): IMPLICATIONS FOR nsp5 REGULATION AND THE DEVELOP MENT OF ANTIVIRALS" *J Biol Chem*
53. Lee, Worrall, Vuckovic et al. (2020) "Crystallographic structure of wild-type SARS-CoV-2 main protease acyl-enzyme intermediate with physiological C-terminal autoprocessing site" *Nat Commun*
54. Anand, Ziebuhr, Wadhwani et al. (2003) "Coronavirus main proteinase (3CL pro ) structure: basis for design of anti-SARS drugs" *Science*
55. Chen, Chen, Tan et al. (2005) "Severe acute respiratory syndrome coronavirus 3C-like proteinase N terminus is indispensable for proteolytic activity but not for enzyme dimerization. Biochemical and thermodynamic investigation in conjunction with molecular dynamics simulations" *J Biol Chem*
56. Hagemeijer, Monastyrska, Griffith et al. (2014) "Membrane rearrangements mediated by coronavirus nonstructural proteins 3 and 4"
57. Jones, Bragagnolo, Arranz et al. (2021) "Capping pores of alphavirus nsP1 gate membranous viral replication factories" *Nature*
58. Chen, Malone, Llewellyn et al. (2020) "Structural basis for helicase-polymerase coupling in the SARS-CoV-2 replication-transcription complex" *Cell*
59. Yan, Ge, Zheng et al. (2021) "Cryo-EM structure of an extended SARS-CoV-2 replication and transcription complex reveals an intermediate state in cap synthesis" *Cell*
60. (2025) *Full-Length Text Journal of Virology*
61. Liu, Shi, Becker et al. (2021) "Structural basis of mismatch recognition by a SARS-CoV-2 proofreading enzyme" *Science*
62. Rosas-Lemus, Minasov, Shuvalova et al. (2020) "High-resolution structures of the SARS-CoV-2 2'-O-methyltransferase reveal strategies for structure-based inhibitor design" *Sci Signal*
63. Frazier, Wilson, Krahn et al. (2022) "Flipped over U: structural basis for dsRNA cleavage by the SARS-CoV-2 endoribonuclease" *Nucleic Acids Res*
64. (2025) "Schrödinger release 2025-2: PIPER, Schrödinger, LLC"
65. Shaqra, Zvornicanin, Huang et al. (2022) "Defining the substrate envelope of SARS-CoV-2 main protease to predict and avoid drug resistance" *Nat Commun*
66. Kaptan, Girych, Enkavi et al. (2022) "Maturation of the SARS-CoV-2 virus is regulated by dimerization of its main protease" *Comput Struct Biotechnol J*
67. Tabler, Wegman, Chen et al. (2022) "The HIV-1 viral protease is activated during assembly and budding prior to particle release" *J Virol*
68. Harrison, Passos, Bruhn et al. (2022) "Cryo-EM structure of the HIV-1 Pol polyprotein provides insights into virion maturation" *Sci Adv*
69. Klein, Winter, Wachsmuth-Melm et al. (2020) "SARS-CoV-2 structure and replication characterized by in situ cryo-electron tomography" *Nat Commun*
70. Subissi, Posthuma, Collet et al. (2014) "One severe acute respiratory syndrome coronavirus protein complex integrates processive RNA polymerase and exonuclease activities" *Proc Natl Acad Sci*
71. Kirchdoerfer, Ward (2019) "Structure of the SARS-CoV nsp12 polymerase bound to nsp7 and nsp8 co-factors" *Nat Commun*
72. Slanina, Madhugiri, Bylapudi et al. (2021) "Coronavirus replicationtranscription complex: vital and selective NMPylation of a conserved site in nsp9 by the NiRAN-RdRp subunit" *Proc Natl Acad Sci*
73. Deng, Hackbart, Mettelman et al. (2017) "Coronavirus nonstructural protein 15 mediates evasion of dsRNA sensors and limits apoptosis in macrophages" *Proc Natl Acad Sci*
74. Pathak, Eliscovich, Mena et al. (2024) "Visualization of early RNA replication kinetics of SARS-CoV-2 by using single molecule RNA-FISH combined with immunofluorescence" *Viruses*
75. Sawicki, Sawicki (1986) "Coronavirus minus-strand RNA synthesis and effect of cycloheximide on coronavirus RNA synthesis" *J Virol*
76. Sawicki, Sawicki, Younker et al. (2005) "Functional and genetic analysis of coronavirus replicasetranscriptase proteins" *PLoS Pathog*
77. Donaldson, Graham, Sims et al. (2007) "Analysis of murine hepatitis virus strain a59 temperature-sensitive mutant TS-LA6 suggests that nsp10 plays a critical role in polyprotein processing" *J Virol*
78. Schmidt, Ganskih, Wei et al. (2023) "SND1 binds SARS-CoV-2 negative-sense RNA and promotes viral RNA synthesis through NSP9" *Cell*
79. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
80. Wu, Zhao, Yu et al. (2020) "A new coronavirus associated with human respiratory disease in China" *Nature*
81. Sperry, Kazi, Graham et al. (2005) "Single-amino-acid substitutions in open reading frame (ORF) 1b-nsp14 and ORF 2a proteins of the coronavirus mouse hepatitis virus are attenuating in mice" *J Virol*
82. (2023) "Schrödinger release 2023-1: Schrödinger, LLC"
83. Lu, Wu, Ghoreishi et al. (2021) "OPLS4: improving force field accuracy on challenging regimes of chemical space" *J Chem Theory Comput*
84. Kozakov, Brenke, Comeau et al. (2006) "PIPER: an FFT-based protein docking program with pairwise potentials" *Proteins*
85. Davis, Leaver-Fay, Chen et al. (2007) "MolPro bity: all-atom contacts and structure validation for proteins and nucleic acids" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12090797&blobtype=pdf | # Recognizing the Top Peer Reviewers in 2024 for the Journal of Virology
Felicia Goodrum, Matthew Aliota, David Barton, Nora Chapman, Anthony Fehr, Ronald Iorio, Florian Krammer, Richard Plemper, Bert Rima, Ashley John, Xianfang Wu, Cleveland Clinic, Shuqi Xiao, Lanzhou
## Abstract
T he Journal of Virology (JVI) remains the leading journal in virology, with 78,636 citations in 2023, reaffirming its status as the most cited journal in the Virology category, according to the latest Journal Citation Reports (Clarivate, 2024). This excellence is derived directly from JVI authors, reviewers, editors, and staff. The editors, editorial board members, and reviewers are to be commended for their hard work in ensuring a smooth review process that is constructive, rigorous, fair, and objective, as well as expedient. The median time from manuscript submission to an editor's first peer review decision letter now stands at 29 days. We thank all the JVI reviewers over the past year. We would especially like to thank the following 30 reviewers, whose service to JVI during the past year was exceptional in the total number of manuscripts reviewed, submitting reviews on time >80% of the time with an average of 11 days or less. All of these reviewers completed 10 or more reviews in 2024. A special shout-out to the reviewers who were also on the 2023 Top Reviewer list (individuals listed in bold). |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12481655&blobtype=pdf | # Correction to "Relative Efficacy, Effectiveness and Safety of Newer And/Or Enhanced Seasonal Influenza Vaccines for the Prevention of Laboratory-Confirmed Influenza in Individuals Aged 18 Years and Over: Update of a Systematic Review"
## Abstract
In the Section 3.3 (page 8 'Serious adverse events' of the MF59-adjuvanted influenza vaccine) the assignment of two cases of Guillain-Barré syndrome(GBS) to the MF59-adjuvanted group was incorrect. This should have read: 'In the RCTs [49, 74, 75], a total of 3 SAEs were identified in the MF59-adjuvanted vaccine group and 3 SAEs were found in the standard vaccine group (including 1 case of Guillain-Barré syndrome).'We apologise for this error. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12639462&blobtype=pdf | # Identification of amino acid substitutions in the hepatitis C virus core region associated with the development of hepatocellular carcinoma in treatment-naive patients in Cameroon -a 10-year retrospective cross-sectional study
Aristide Mounchili-Njifon, Abdou Modiyinji, Loique Landry, E Messanga, Moise Henri, Moumbeket Yifomnjou, Herman Mfombouot, Desmon Tsafack, Rosereine Mbouyap, Gwladys Monamele, Alain Paul, Tagnouokam-Ngoupo, Simon Lissock, Jean Paul, Assam Assam, Richard Njouom
## Abstract
In Cameroon, infection with the hepatitis C virus (HCV) is a major factor in hepatocellular carcinoma (HCC). Cirrhotic patients, even when treated with direct-acting antivirals (DAAs), may still be at risk of developing HCC. The aim of this study was to identify mutations associated with the development of HCC in treatment-naive HCVinfected patients, in order to understand the molecular mechanisms involved in this local context. From 2013 to 2023. 1065 HCV-infected Cameroonian patients provided blood samples. Plasma, isolated and stored at -80 • C, was used to amplify the Core gene using specific primers. Nucleotide sequences obtained by Sanger sequencing were analyzed with IQ-Tree for phylogenetic studies. Sequences were edited and aligned using Mega and MAFFT, and mutation searches were performed manually using AliView software. Three genotypes (1, 2, 4) were identified, with genotype 4 predominating. Several mutations known to play an oncogenic role were detected: K10R (0.66 %), R70Q (10.52 %), T72E (80.56 %), K74R (82.82 %), G77A (70.53 %) and C/L91M (0.09 %). A hundred other mutations potentially linked to response to AADs were also observed. The analysis revealed mutations significantly related to sex and year, such as (K115R, N106S, T48A) with p-values of (0.0023; 0.0006), (0.0012; 0.0004) and (0.0045; 0.0058) respectively. This study provides the first comprehensive mapping of HCV core mutations in Cameroon, identifying variants potentially linked to HCC. Although limited by the lack of clinical follow-up, it underscores the urgency of monitoring these mutations in national HCV elimination programs, in line with the WHO's goals for 2030.
## 1. Introduction
Chronic hepatitis C virus (HCV) infection is a major cause of liver fibrosis and cirrhosis, with a risk of developing hepatocellular carcinoma (HCC) (Fiehn et al., 2024). HCC is one of the most common cancers in sub-Saharan Africa and represents the second leading cause of cancer deaths (Ferlay et al., 2015). Worldwide, around 50 million people are living with HCV, with approximately 1.0 million new infections per year (WHO 2024). HCV infection is recognized as a major risk factor for HCC in Cameroon (Amougou-Atsama et al., 2020). Populations aged 65 and over, rural communities, and those in West Africa, as well as and some Central and East African countries appear to be disproportionately affected (Kassa et al., 2024).
Hepatitis C virus (HCV) is a single-stranded RNA virus with a genome of around 9600 base pairs. Taxonomically, it is subdivided into eight main genotypes (GT) and numerous subtypes (Njifon et al., 2024;Ezzat et al., 2005), whose geographical distribution varies considerably (Raimondi et al., 2009). Genotypes 1, 2, and 3 have a worldwide distribution, while the other genotypes show a more regional prevalence. A recent study in Cameroon identified three circulating genotypes (GT1, GT2 and GT4), with a predominance of GT4 (38.48 %) and GT1 (37.11 %), followed by genotype 2, with twelve different subtypes detected during the study period, with a predominance of subtypes 4f (18.95 %) and 1e (16.02 %) (Njifon et al., 2024). An Egyptian study investigating risk factors for HCC in chronically HCV-infected patients suggested that genotype 4 might expose patients to a higher risk of HCC than other HCV variants (Ezzat et al., 2005). In Cameroon, genotype 4, and in particular subtype 4f, or selected variants affecting the core HCV gene, present a higher risk of developing HCC than patients infected with genotypes 1 and 2 (Amougou- Atsama et al., 2020).
The progression of HCV infection to end-stage liver disease and HCC is extremely variable and depends on many factors, including viral ones. Worldwide, a close relationship has been highlighted between infection with several HCV genotypes and/or the presence of mutations in different HCV proteins (Raimondi et al., 2009;Silini et al., 1996). In addition, amino acid substitutions at certain positions in the core region of HCV have been reported to be important predictors of HCC development in Asian patients (Amougou-Atsama et al., 2020;El-Shamy et al., 2016).
HCV is a blood-borne virus (Farooq et al., 2024). Treatment of hepatitis C virus (HCV) with direct-acting antivirals (DAAs) results in high rates of sustained virological response (SVR), but the risk of HCC persists in people with advanced liver disease, even after SVR (Quaranta et al., 2024).
In Cameroon, genotype 4 (particularly 4f) predominates and is associated with an increased risk of HCC, in contrast to other regions like Asia where genotype 1b is predominant (Amougou-Atsama et al., 2020;Farooq et al., 2024). This genotypic divergence highlights the importance of local studies to guide screening strategies.
Few studies have been conducted worldwide, particularly in lowincome countries, using partial genomic sequences of the core region to identify possible amino acid mutations associated with the development of HCC in HCV-infected patients. In order to fill this knowledge gap, our study aimed to perform nucleotide sequencing of the HCV core region, its phylogenetic analysis, and to identify amino acid substitutions of the core protein in treatment-naive chronically HCVinfected Cameroonian patients. This knowledge is essential to reduce the incidence of terminal complications of HCV infection in this country, and to ensure effective treatment of the disease on a global scale, in order to achieve the World Health Organization (WHO)'s is goal of eliminating HCV by 2030.
## 2. Materials and methods
## 2.1. Sampling technique and HCV viral load testing
This is a cross-sectional retrospective descriptive study conducted nationally between January 2013 and October 2023 at the Pasteur Center of Cameroon (CPC). The CPC is the reference laboratory for several pathogens, including viral hepatitis in Cameroon, and plays a central role in the national HCV control program. As part of this role, patients with HCV are systematically referred to the CPC for viral load measurement and genotyping. The data presented here were collected as part of routine diagnostic activities for HCV, with no additional tests performed. Over a period of 11 years, samples were randomly and stratified by year to ensure a balanced representation of time periods and demographic characteristics. This approach aimed to minimize temporal biases, although limitations include a potential selection bias related to patients referred to the CPC, which may not perfectly reflect the general population infected with HCV in Cameroon. HCV viral loads were determined using the Abbott Real-Time HCV test and Abbott m2000 platforms (Abbott Molecular, Wiesbaden, Germany) according to the manufacturer's instructions. Briefly, the protocol calls for RNA extraction from a 0.5 ml plasma sample (from EDTA tubes) on the Abbott m2000sp, followed by amplification on the m2000rt with a detection limit set at 12 IU/mL. Sequencing, phylogenetic analysis, and mutation analysis of approximately 345 nucleotides of the core region were then carried out.
## 2.2. Amplification of the core region
Viral RNA was extracted from 140 μL of plasma using the QIAamp viral RNA mini kit (Qiagen, Courtaboeuf, France) according to the manufacturer's protocol. HCV genotyping was performed at the HCV genome level of the central region of the core gene, approximately 345 bp. This HCV fragment was amplified using semi-nested PCR amplification as described in our previous study (Silini et al., 1996).
## 2.3. Statistical analysis
Patient demographic (date of birth, sex), virological (viral load) and clinical (year of diagnosis) data were extracted from the HCV database. Statistical analyses were performed with SPSS 27.0, presenting prevalences and quantitative variables in percentages. Comparisons of categorical variables were made using Kruskal-Wallis or χ² tests, with a significance threshold set at p < 0.05. Visualization of the distribution of genotypes according to mutations was carried out with Python 3.12 via the Spyder environment (Anaconda).
## 2.4. Sequencing and phylogenetic analysis
All nested PCR products from core regions were sequenced by the Sanger method using the GenomeLab DTCS-Quick Start kit (Beckman Coulter, Paris, France) in accordance with the manufacturer's instructions. Consensus sequences were obtained after manual editing of forward and reverse sequences using CLC Main Workbench software (version 5.5.2). HCV genotype/subtype assignment was performed using the HCVnet genotyping tool (https://www.genomedetective.com/ app/typingtool/hcv/) and confirmed by BLAST. The 1065 core gene sequences were edited and aligned with MEGA v.11 (Kichatova et al., 2018). Reference sequences were downloaded from the GenBank database (NCBI). Phylogenetic reconstructions were performed in IQ-TREE v2.1 (Chouikha et al., 2021) using the maximum likelihood method, with selection of the optimal nucleotide substitution model via Model-Finder. (Hamadou et al., 2025). The robustness of the phylogenetic grouping was assessed using the SH-aLRT (1000 replicates) approximate likelihood ratio test (Sanchez et al., 2000). Bootstrap values above 70 % were reported on the consensus trees, and 1000 replicates were used to assess the robustness of the tree topology. All nucleotide sequences obtained from the HCV core gene were submitted to GenBank (access numbers are given in the data availability statement).
## 2.5. Amino acid analysis
To determine nucleotide changes, the partial sequences of the HCV core were compared with the reference genome sequence (U10190.1) available on the "Gen-Bank" website (https://www.ncbi.nlm.nih.gov/). The generated dataset was aligned using the online program MAFFT version 7 (Katoh and Standley, 2013) and then, using AliView software version 1.25 (Larsson, 2014)(manually made), the nucleotide sequences were translated into amino acid sequences. The amino acid positions of the reference sequence (U10190.1) were then compared against the partial sequences of the HCV core gene to identify the different amino acid mutations.
## 3. Results
## 3.1. Demographic characteristics and genotype distribution
A total of 1065 samples were included in this study by stratified sampling, with 1065 nucleotide sequences obtained from the sequenced core region. Of the 1065 HCV-infected patients included in our study, 647 (60.80 %) were women and 418 (39.2 %) were men, giving an M/W sex ratio of 0.64. With a mean age of 67.65 and a median age of 70, age ranged from 5 to 97 years.
Three genotypes were identified by core sequencing in 1065 blood samples from HCV-positive individuals: genotypes 1, 2, and 4. Genotype 4 was the most frequent at 41.97 % (447/1065), followed by genotype 1 at 38.50 % (410/1065) and genotype 2 at 19.53 % (208/1065).
Various HCV subtypes were also detected, namely 1c, 1e, 1 g, 1 h, 1l, 2, 4a, 4f, 4k,4l, 4p, 4r, and 4t Subtype 4f was the most predominant, with a percentage of 18.05 % (198/1065), followed by 1e with 17.84 % (164/1065). Fig. 1
## 3.2. Analysis of HCV core mutations
We then aligned the HCV amino acid sequences of 1065 successfully sequenced isolates with the reference sequence (U10190) first detected in 1994 (Bukh et al., 1994), to determine amino acid substitutions. Fig. 2 Mutations were defined as amino acid substitutions in the core of our sequences relative to the amino acids of the aligned reference sequence. In total, several amino acid mutations have been identified in this region, which are summarized in Table 1.
Many of these mutations are not yet available in the literature, so their clinical significance remains unknown. We had identified various mutations such as K10R, R70Q, T72E, K74R, G77A, C/L91M which aggravate the prognosis of the infection and increase the risk of HCC due to their oncogenic effect alongside other mutations that have already been identified in Cameroon during another study (Amougou-Atsama et al., 2020). Fig. 3
## 3.3. Factors associated with core region mutations and genotype distribution
We further analyzed the risks associated with these mutations in relation to the different parameters studied, such as gender, age, year and viral load. Our results suggest that the majority of these mutations such as: (K115R, N106S, H114R, K78Q, P75S, K74R, T72E, D68A, T48A, Q20M, G77A) were therefore significantly associated with gender, with a higher prevalence in women with p-value values respectively (0.0023; 0.0012; 0.0173; 0.0144; 0.0037; 0.0230; 0.0184; 0.0212; 0.0045; 0.0011; 0.0206). We also observed several mutations such as: (K115R, N106S, H114R, K78Q, P75S, K74R, T72E, D68A, T48A, G77A, S71P) were therefore significantly associated with year, with values respectively (0.0006; 0.0004; 0.0079; 0.0040; 0.0127; 0.0016; 0.0040; 0.0020; 0.0058; 0.0048; 0.0114), suggesting a variation in their frequency over time. Then only the H114R mutation was significantly associated with age (p = 0.0472). Finally, only the S71P mutation was significantly associated with viral load (p = 0.0401). The factors associated with mutations in the core region in relation to the various parameters are presented in Table 2.
With regard to genotype diversity, we observed that certain genotypes such as: (1l, 1e, 2nc, 4nc, 4f, 4p, 4t) are persistent over several years (2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021)(2022). Whereas mutations such as: (4a, 4l, 4k, 1c, 4r, 1 g) appear sporadically. Then mutations such as: K151R, H114R and T110N are frequent and spread over several genotypes and years. These results underline the influence of epidemiological factors (sex, time) and viral Fig. 1. Phylogenetic analysis based on partial sequences of the core gene from Cameroonian HCV isolates and reference strains. The phylogenetic tree is based on 1065 partial core gene sequences of HCV from samples collected in Cameroon between 2013 and 2023, along with reference strains. For easy identification, reference sequences are colored in red and identified by their accession number and genotype, while the Cameroonian sequences described in this study are in black and indicated in bold. The latter cluster within the corresponding genotypic clusters. Phylogenetic tree analyses were performed using the maximum likelihood method with IQ-TREE software, employing the most suitable nucleotide substitution model selected by Model Finder. The robustness of the phylogenetic topologies was assessed using a bootstrap method with 1000 replicates. By convention, only nodes with bootstrap values ≥ 70 % are annotated on the final tree. All nucleotide sequences reported in this study are available in the GenBank databases, and the sequencing data can be accessed via their accession numbers. dynamics in the distribution of mutations. Visualization confirms the persistence of certain genotypes and the dynamics of mutations over a decade, with notable annual variations. The distribution of genotypes according to mutation and year is shown in Fig. 4.
## 4. Discussion
Chronic hepatitis C is a major etiological factor in hepatic cirrhosis, hepatocellular carcinoma, and hepatic mortality. Moreover, 40-70 % of patients develop extrahepatic manifestations during the course of the infection (Nerrienet et al., 2005). At the molecular level, amino acid substitutions at key positions in the HCV capsid protein influence both therapeutic response to interferon and hepatic oncogenesis (Kichatova et al., 2018). Infection with HCV is recognized as a major risk factor for HCC in Cameroon (Amougou-Atsama et al., 2020).
In our study, HCV infection was more frequent in women (60.8 %) than in men (39.2 %), with an M/W sex ratio of 0.64. A higher frequency of HCV infection in women has been reported in several studies in Cameroon, Africa and worldwide (Njifon et al., 2024;Chouikha et al., 2021;Hamadou et al., 2025;Sanchez et al., 2000). Also, the age of patients in our study ranged from 5 to 97 years, with a mean age of 67.65 years; these results are consistent with our previous study, where the mean age was 66 years (Njifon et al., 2024), and similar to other studies carried out in Cameroon, where the mean age was between 50 and 62 years (Cantaloube et al., 2010;Pépin et al., 2010).These studies confirmed that the spread of HCV before the 1960s in certain African countries was due to a cohort effect, with prior exposure, possibly iatrogenic, hence, widespread mass treatments and/or vaccination using unsterilized equipment were a major source of HCV transmission (Pépin et al., 2010;Frank et al., 2000;Nerrienet et al., 2005;Njouom et al., 2003).
In the current study, the core gene was amplified as the region of interest because a study done in Cameroon showed a high HCV amplification rate in the core and NS5B regions with good concordance for genotyping (Tagnouokam-Ngoupo et al., 2019). This study reports the circulation of 03 genotypes thus HCV-1, HCV-2 and HCV-4 in HCV-infected Cameroonian patients. Our study, which we believe to be one of the few to have estimated the frequencies of natural mutations at the Core gene level in Cameroonian patients in order to provide optimal treatment, was carried out in 1065 DAA treatment-naive patients who were successfully amplified for Core gene fragments. The present study revealed HCV genotype 4 (HCV-4) to be predominant with a percentage of 41.97 %. Remarkably, HCV subtype 4f (HCV-4f) was dominant with a percentage of 18.59 % we observed a 21-fold higher risk of developing HCC in patients infected with HCV-4f. Our data regarding the high predominance of HCV genotype 4 are however similar to those of previously reported studies conducted on non-HCC patients in Cameroon and other studies conducted in Central African countries such as Gabon, where genotype 4e is predominant, and the Central African Republic, where genotype 4k is predominant (Amougou-Atsama et al., 2020;Ali-Mahamat and Njouom, 2015;Njouom et al., 2012;Njouom et al., 2009). The high prevalence of HCV-4f HCC cases in Cameroon suggests that HCV genotype 4 has a higher oncogenic potential than other genotypes circulating in the country. Interestingly, this result differs from that observed in HCC cases in North Africa, where HCV-1b was predominant (Brahim et al., 2013). Our results therefore show that genotype determination could be useful in predicting the outcome of infection.
In the present study, we analyzed the different amino acid sequences of the core gene of our 1065 treatment-naive HCV-infected patients to identify the different mutations in the HCV core that could be responsible for the development of HCC. A total of 278 amino acid mutations were identified in this region, with different proportions than in the first study done in Central Africa with such a high frequency. Our study revealed the presence of K10R, T72E, K74R and G77A mutations, which are known to be strongly associated with the development of HCC in HCV-infected patients (Amougou-Atsama et al., 2020;Hu et al., 2009;Jaspe et al., 2012;Seko et al., 2013). It is well known that HCV core mutations R70Q and L/C91M worsen the prognosis of infection and increase the risk of HCC due to their oncogenic effect (El-Shamy et al., 2016;Hu et al., 2009); In this study, mutations at positions 70 and 91 were significantly found. Our results are similar to a study done in Cameroon in 2020 with case controls in patients with HCC (Amougou-Atsama et al., 2020) and also similar to those observed in patients with liver disease in Botswana (Bhebhe et al., 2019). In contrast, these mutations were associated with a higher risk of HCC in patients infected with HCV genotype 1 in Japan (Akuta et al., 2011;Akuta et al., 2009).
A similar study carried out in Morocco on treatment-naive patients with HCV genotype 1b revealed 03 amino acid mutations at positions 70, 75, and 91 (Brahim et al., 2013), then three years later another study revealed some amino acid mutations such as: N161I, L36V, T49A, P71S, T75S, and T110N in Moroccan intravenous drug users among patients infected with HCV genotype 3a (Trimbitas et al., 2016) and finally another study carried out in Pakistan in 2022 in patients infected with HCV genotype 3a revealed several amino acid mutations such as: R46C, R70Q, L91C, G90E, N/S105A, P108A, N110I, S116V, G90S, A77G, and G145R (Khan et al., 2022). The significant difference in mutation frequency between previous studies and ours, which identified over a hundred mutations, can be attributed to the sample size and the number of sequenced genotypes, as we analyzed all genotypic sequences from a large sample. Furthermore, our work highlights the specificity of mutations in the African genotype 4, suggesting distinct oncogenic mechanisms related to viral genetic diversity.
The study of factors associated with mutations in the core region revealed the predominance of female sex observed for most mutations (K115R, N106S, H114R, etc.) suggests a potential influence of biological factors (hormonal, immune response) or epidemiological factors (different exposure, risk behaviors). For example, estrogens could modulate viral replication or selection pressure. The absence of linkage for T110N, R70Q, K10E/K10R could indicate that these mutations are gender-neutral. Then certain mutations (K115R, N106S, H114R, etc.) showed a significant association over the years, suggesting changes in clinical practices (antiviral treatments such as DAAs, selection pressure) and epidemiological variations (transmission of specific strains). Then, the single H114R mutation showed a significant age-related association (p = 0.0472), which may be due to an accumulation of mutations in older patients (long-term chronic infection) and an altered immune response with age, favoring certain mutations. Finally, we found a significant association with viral load and the S71P mutation (p = 0.0401), which could imply a functional impact of this mutation on viral replication and adaptation of the virus to specific cellular niches (liver, other tissues). The study of the distribution of diversity genotypes according to mutations and years revealed a persistent pattern, with some genotypes (1l, 1e, 2nc, etc.) present consistently over several years, while others appear sporadically. These results underline the influence of epidemiological factors (sex, time) and viral dynamics in the distribution of mutations. Our results are very different from those of previous studies carried out in Africa and Cameroon, which can be explained by the fact that several virological and demographic factors (sex, year, viral load and age) were not taken into account in these studies. (Amougou-Atsama et al., 2020;Brahim et al., 2013;Akuta et al., 2011).
To achieve the goal of HCV elimination by 2030, Cameroon and sub-Saharan Africa must: introduce public coverage of treatment costs, identify and rigorously monitor at-risk patients (carriers of resistance mutations such as NS5A variants, or oncogenic mutations such as Core R70Q) via molecular testing, and set up systematic monitoring of viral loads to prevent relapse and progression to HCC. This integrated approach, combining affordability, precision medicine and virological monitoring, is essential to counter the epidemic in a context of limited resources.
Despite the accumulated epidemiological evidence confirming the impact of core protein mutations on the progression of HCV infection, particularly hepatocellular carcinoma (HCC), the underlying molecular mechanisms remain uncertain for many mutations, as our study identified approximately 100 mutations. The role of amino acid substitutions at various positions within the core region is challenging to delineate, necessitating further research and longitudinal studies on mutation analysis based on circulating genotypes in Cameroon, given that some studies indicate mutations vary by genotype. Notably, our study revealed several mutations, yet the impact of these individual or combined mutations on treatment resistance or disease progression to HCC is unknown and warrants greater attention in the future. Consequently, systematic screening for mutations in the HCV core sequence in Cameroon in cases of treatment failure could partially predict progression to HCC. While this study identifies potentially oncogenic mutations, the lack of longitudinal clinical follow-up prevents establishing a direct causal link between these mutations and the development of HCC. Future studies incorporating histological and imaging data will be necessary to confirm these associations.
In conclusion, this study revealed a high frequency of amino acid mutations in the central region of HCV, making it one of the first of its kind in Central Africa, with the identification of several mutations already known to be associated with the development of HCC in Central Africa. Effective diagnostic measures to identify mutations associated with an increased risk of HCC are of paramount importance in reducing the incidence of terminal complications of HCV infection in this country, as early detection of HCC improves outcomes. The diversity of genotypes and co-occurrence of mutations underscore the genetic complexity of the virus, with potential evolutionary trends to watch out for. A better understanding of hepatocellular carcinogenesis and a better ability to identify high-risk patients would enable surveillance and prevention efforts to be focused on those most at risk of, and most likely to benefit from, HCC, helping to achieve the WHO's goal of eliminating HCV by 2030.
## Glossary
## Ethical statement
The study protocol conformed ethical guidelines of the 1975 Declaration of Helsinki, and was approved by the Regional Center's Human Health Research Ethics Committee (CRERSH/C) (number: E00571/ CERSH/ C/2023).
## Informed consent
Not applicable.
## Organ donation
Not Applicable.
## Animal treatment
Not Applicable.
## Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
## References
1. Akuta, Suzuki, Hirakawa et al. (2011) "Amino acid substitutions in hepatitis C virus core region predict hepatocarcinogenesis following eradication of HCV RNA by antiviral therapy" *J. Med. Virol. Juin*
2. Akuta, Suzuki, Hirakawa et al. (2009) "Amino acid substitutions in the hepatitis C virus core region of genotype 1b are the important predictor of severe insulin resistance in patients without cirrhosis and diabetes mellitus" *J. Med. Virol. Juin*
3. Ali-Mahamat, Njouom (2015) "High rate of infection with hepatitis C virus genotype 4 in Chad" *Central Africa. Indian J. Med. Microbiol*
4. Amougou-Atsama, Jean Adrien Atangana, Noah Noah et al. (2020) "The role of hepatitis C virus genotypes and core mutations in hepatocellular carcinoma in Cameroon" *J. Viral Hepat*
5. Bhebhe, Anderson, Souda et al. (2019) "Molecular characterization of hepatitis C virus in liver disease patients in Botswana: a retrospective cross-sectional study" *BMC. Infect. Dis*
6. Brahim, Ezzikouri, Mtairag et al. (2013) "Amino acid substitutions in the Hepatitis C virus core region of genotype 1b in Moroccan patients" *Infect. Genet. Evol*
7. Bukh, Purcell, Miller (1994) "Sequence analysis of the core gene of 14 hepatitis C virus genotypes" *Proc. Natl. Acad. Sci. U S A*
8. Cantaloube, Gallian, Bokilo et al. (2010) "Analysis of hepatitis C virus strains circulating in Republic of the Congo" *J. Med. Virol*
9. Chouikha, Khedhiri, Triki et al. (2021) "Focus on hepatitis C virus genotype distribution in Tunisia prior to elimination: a 16year retrospective study" *Arch. Virol*
10. El-Shamy, Pendleton, Eng et al. (2016) "Impact of HCV core gene quasispecies on hepatocellular carcinoma risk among HALT-C trial patients" *Sci. Rep*
11. Ezzat, Abdel-Hamid, Eissa et al. (2005) "Associations of pesticides, HCV, HBV, and hepatocellular carcinoma in Egypt" *Int. J. Hyg. Environ. Health*
12. Farooq, James, Abbott et al. (2015) "Risk factors for hepatocellular carcinoma associated with hepatitis C genotype 3 infection: a systematic review" *World J. Gastrointest. Oncol*
13. Fiehn, Beisel, Binder (2024) "Hepatitis C virus and hepatocellular carcinoma: carcinogenesis in the era of direct-acting antivirals" *Curr. Opin. Virol*
14. Frank, Mohamed, Strickland et al. (2000) "The role of parenteral antischistosomal therapy in the spread of hepatitis C virus in Egypt" *Lancet*
15. Hamadou, Njoya, Kowo et al. (2018) "Traitement de l'Hépatite C de Génotype 1 par les antiviraux d'Action Directe au Cameroun : résultats préliminaires. HEALTH SCIENCES AND DISEASE [Internet], 22 août"
16. Hu, Muroyama, Kowatari et al. (2009) "Characteristic mutations in hepatitis C virus core gene related to the occurrence of hepatocellular carcinoma" *Cancer Sci*
17. Jaspe, Sulbarán, Sulbarán et al. (2012) "Prevalence of amino acid mutations in hepatitis C virus core and NS5B regions among Venezuelan viral isolates and comparison with worldwide isolates" *Virol. J*
18. Kassa, Walker, Alamneh et al. (2024) "Prevalence, trends, and distribution of hepatitis C virus among the general population in sub-Saharan Africa: a systematic review and meta-analysis" *Liver Internat*
19. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol. Biol. Evol*
20. Khan, Nawaz, Ullah et al. (2022) "Core amino acid substitutions in HCV-3a isolates from Pakistan and opportunities for multi-epitopic vaccines" *J. Biomol. Struct. Dyn. Mai*
21. Kichatova, Kyuregyan, Soboleva et al. (2018) "Frequency of interferon-resistance conferring substitutions in amino acid positions 70 and 91 of core protein of the Russian HCV 1b isolates analyzed in the T-cell epitopic context" *J Immunol. Res*
22. Larsson (2014) "AliView: a fast and lightweight alignment viewer and editor for large datasets" *Bioinformatics*
23. Nerrienet, Pouillot, Lachenal et al. (2005) "Hepatitis C virus infection in cameroon: a cohort-effect" *J. Med. Virol*
24. Njifon, Modiyinji, Monamele et al. (2024) "A decade-long retrospective study of hepatitis C virus genetic diversity in Cameroon, 2013-2023: presence of a high proportion of unsubtypable and putative recombinant HCV strains" *Arch. Virol*
25. Njouom, Caron, Besson et al. (2012) "Phylogeography, risk factors and genetic history of hepatitis C virus in Gabon, central Africa" *PLoS. One*
26. Njouom, Frost, Deslandes et al. (2009) "Predominance of hepatitis C virus genotype 4 infection and rapid transmission between 1935 and 1965 in the Central African Republic" *J. Gen. Virol*
27. Njouom, Pasquier, Ayouba et al. (2003) "Hepatitis C virus infection among pregnant women in Yaounde, Cameroon: prevalence, viremia, and genotypes" *J. Med. Virol*
28. Pépin, Labbé, Mamadou-Yaya et al. (2010) "Iatrogenic transmission of human T cell lymphotropic virus type 1 and hepatitis C virus through parenteral treatment and chemoprophylaxis of sleeping sickness in colonial Equatorial Africa" *Clin. Infect. Dis*
29. Quaranta, Cavalletto, Russo et al. (2024) "Reduction of the risk of hepatocellular carcinoma over time using direct-acting antivirals: a propensity score analysis of a real-life cohort (PITER HCV)" *Viruses*
30. Raimondi, Bruno, Mondelli et al. (2009) "Hepatitis C virus genotype 1b as a risk factor for hepatocellular carcinoma development: a metaanalysis" *J. Hepatol*
31. Sanchez, Sjogren, Callahan et al. (2000) "Hepatitis C in Peru: risk factors for infection, potential iatrogenic transmission, and genotype distribution" *Am. J. Trop. Med. Hyg*
32. Seko, Akuta, Suzuki et al. (2013) "Amino acid substitutions in the hepatitis C Virus core region and lipid metabolism are associated with hepatocarcinogenesis in nonresponders to interferon plus ribavirin combination therapy" *Intervirology*
33. Silini, Bottelli, Asti et al. (1996) "Hepatitis C virus genotypes and risk of hepatocellular carcinoma in cirrhosis: a casecontrol study" *Gastroenterology*
34. Tagnouokam-Ngoupo, Ngoufack, Kenmoe et al. (2019) "Hepatitis C virus genotyping based on Core and NS5B regions in Cameroonian patients" *Virol. J*
35. Trimbitas, Fayssel, Serghini et al. (2016) "Molecular characterization of hepatitis C virus core region in moroccan intravenous drug users" *J. Med. Virol*
36. Who (2024) "World Health Organization. Hepatitis C, fact sheet" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817958&blobtype=pdf | # Hijacking of host Src-PI3K-Akt signaling by WSSV IE1 protein suppresses apoptotic and autophagic defenses to facilitate viral proliferation
Kaiyu Lu, Jia Zhang, Jinghua Zhu, Yongzhen Zhao, Xiuli Chen, Yueling Zhang, Defu Yao
## Abstract
The phosphoinositide 3-kinase (PI3K)-Akt pathway is a key signaling cascade regulating diverse cellular processes, including proliferation, survival, autoph agy, translation, and metabolism. White spot syndrome virus (WSSV), a major patho gen devastating global crustacean aquaculture, has been demonstrated to exploit the PI3K-Akt pathway to facilitate its proliferation. However, the precise mechanism underlying this viral modulation remained unclear. In this study, we demonstrate that WSSV infection induces activation of the PI3K-Akt pathway during the early infection stage in Penaeus vannamei. Mechanistically, we reveal that the WSSV immediate-early protein IE1 interacts with and activates host Src64B kinase via its Y 129 FTS tyrosine motif. This specific interaction promotes recruitment of the PI3K regulatory subunit alpha (PI3Kp85α), thereby triggering the downstream PI3K-Akt signaling. By activating this pathway, WSSV establishes a favorable environment for its proliferation by suppressing host apoptotic and autophagic defenses. Our findings unveil a previously unknown mechanism of WSSV immune evasion through Src-PI3K-Akt signaling hijacking and identify components of this signaling hub as potential therapeutic targets for anti-WSSV strategies.IMPORTANCE Viruses usually hijack host signaling pathways to enhance infectivity and evade immune defenses. Understanding these interactions is critical for elucidating viral pathogenesis and developing effective antiviral strategies. Here, we demonstrate that the WSSV immediate-early protein IE1 binds to and activates host Src64B kinase, which in turn recruits PI3Kp85α and activates the PI3K-Akt signaling cascade. Activation of this pathway suppresses apoptosis and autophagy, thereby facilitating viral prolifera tion. These findings advance our understanding of WSSV pathogenesis and identify the Src-PI3K-Akt signaling as a promising therapeutic target for anti-WSSV intervention.
Phosphoinositide 3-kinase (PI3K) is a heterodimeric lipid kinase comprising a regulatory subunit (p85) and a catalytic subunit (p110) (9). PI3K activation occurs when the SH2 domain of the p85 binds autophosphorylated receptor or nonreceptor tyrosine kinases at the plasma membrane. This binding activates p110, enabling phosphorylation of phosphatidylinositol-4,5-bisphosphate (PIP2) to generate phosphatidylinositol-3,4,5trisphosphate (PIP3). PIP3 acts as a second messenger, recruiting pleckstrin homology (PH) domain-containing kinases like Akt. Akt subsequently undergoes dual phosphory lation-Thr308 (mediated by PDK1) and Ser473 (mediated by mTORC2)-to achieve full activation. Once activated, Akt translocates from the plasma membrane to various intracellular compartments, where it phosphorylates downstream substrates to regulate processes such as proliferation, survival, autophagy, translation, and metabolism (10,11). Given its critical role in cellular homeostasis, the PI3K-Akt pathway is frequently hijacked by viruses to support infection. Emerging evidence highlights its multifaceted contributions to viral infections, such as facilitating viral entry, inhibiting apoptosis, enhancing viral protein synthesis, and boosting viral replication (12)(13)(14). In penaeid shrimp, studies have demonstrated that the PI3K-Akt pathway facilitates WSSV infection through several mechanisms. For instance, WSSV hijacks this pathway to induce aerobic glycolysis (the Warburg effect), thereby reprogramming host metabolism to fulfill its energy and biosynthetic demands (15). Additionally, this metabolic shift also helps maintain cellular redox balance, protecting against oxidative damage caused by reactive oxygen species during infection (16). Furthermore, WSSV leverages this pathway to induce lipid biosynthesis, which is crucial for viral morphogenesis (17). However, the exact molecular mechanism by which WSSV modulates the PI3K-Akt pathway in shrimp remains unclear.
The immediate-early (IE) genes of DNA viruses encode crucial regulatory proteins that facilitate viral infection. Among the 21 IE genes identified in WSSV (18)(19)(20), IE1, the most extensively studied, functions as a transcription factor, exhibiting transactiva tion and DNA-binding capabilities (21). IE1 also interacts with multiple host proteins, including Retinoblastoma protein (Rb) (22), signal transducer and activator of transcrip tion (STAT) (23), c-Jun N-terminal kinase (JNK) (24), β-catenin (25), prophenoloxidase (proPO) (26), and integrin-α5 (27) to modulate host signaling and enhance viral prolifer ation. Interestingly, our prior work indicated potential interactions between IE1 and components of the PI3K-Akt pathway, specifically Src64B and PI3Kp85α (26), suggesting IE1's involvement in regulating this pathway. In this study, we clarify a mechanistic cascade in which IE1 binds and activates the Src64B via its Y 129 FTS tyrosine motif, promoting PI3Kp85α recruitment and PI3K-Akt pathway activation. This virus-induced Src-PI3K-Akt signaling suppresses host apoptotic and autophagic defenses, creating a favorable environment for WSSV proliferation. Our findings reveal a novel WSSV strategy for hijacking the host Src-PI3K-Akt signaling and identify promising targets for anti-WSSV intervention.
## RESULTS
## IE1 interacts with host Src64B and PI3Kp85α
Our previous proteomics data indicated potential interactions between WSSV IE1 and Penaeus vannamei Src64B and PI3Kp85α (26). Domain analysis revealed that P. vannamei Src64B is highly conserved, featuring characteristic Src family kinase domains: an Src homology 3 (SH3) domain, an SH2 domain, and a tyrosine kinase catalytic (TyrKc) domain (Fig. S1A). In contrast, P. vannamei PI3Kp85α shows lower conservation relative to its human ortholog. Notably, it contains two additional protein kinase C conserved region 1 (C1) domains and an N-terminal sterile alpha motif (SAM) domain, which are absent in Homo sapiens PI3Kp85 (Fig. S1B). To assess physical interactions, co-immunoprecipitation (Co-IP) assays in High Five cells demonstrated that ectopically expressed IE1 specifically binds both Src64B and PI3Kp85α (Fig. 1A andB). Immunofluorescence assays showed that individually expressed IE1 localized throughout the cytoplasm and nucleus, Src64B exclusively to the plasma membrane, and PI3Kp85α predominantly to the cytoplasm (Fig. 1C). However, upon co-expression, IE1 was observed to co-localize with Src64B and PI3Kp85α at the plasma membrane (Fig. 1D). These data demonstrate functional interactions between IE1 and both Src64B and PI3Kp85α.
## IE1 activates Src64B via a tyrosine motif to recruit PI3Kp85α
Viral proteins are able to bind and activate host Src family kinases via tyrosine motifs (28,29). In this study, we identified a putative tyrosine motif at residue 129 of IE1 (Y 129 FTS), suggesting its potential to activate Src64B via binding. To test this, we co-expressed IE1 and Src64B in High Five cells and assessed Src64B kinase activity by monitoring its Full-Length Text tyrosine phosphorylation. Our results showed that overexpression of IE1 significantly increased Src64B tyrosine phosphorylation compared with the EGFP control (Fig. 2A). Furthermore, we introduced a point mutation in the Y 129 FTS motif, replacing the tyrosine residue with phenylalanine. This substitution markedly reduced both IE1-Src64B binding and Src64B tyrosine phosphorylation (Fig. 2B). These results indicate that IE1 binds and activates Src64B via its Y 129 FTS tyrosine motif.
To determine whether Src kinase activity influences the interactions between IE1 and Src64B or PI3Kp85α, we performed Co-IP experiments in the presence of the specific Src family kinase inhibitor Saracatinib. Treatment with Saracatinib significantly attenuated the interaction between IE1 and Src64B or PI3Kp85α (Fig. 2C andD). Moreover, Co-IP experiments upon co-expression of IE1, Src64B, and PI3Kp85α revealed that these proteins form a ternary complex (Fig. 2E). Notably, Src64B overexpression enhanced the interaction between IE1 and PI3Kp85α (Fig. 2E), whereas Saracatinib treatment reduced these interactions (Fig. 2F). Collectively, these data demonstrate that IE1 activates Src64B, thereby facilitating the recruitment of PI3Kp85α.
## IE1 is essential for activating Src-PI3K-Akt signaling upon WSSV infection
Previous studies have established the essential role of the PI3K-Akt pathway in promot ing WSSV infection (15)(16)(17); however, the dynamics of its activation during infection remain unclear. To address this, we quantified the mRNA expression of Src64B and PI3Kp85α, as well as monitored PIP3 generation and Akt phosphorylation (pAkt-Ser473) in hemocytes at various time points post-injection with WSSV or PBS. The results showed that a progressive increase in IE1 mRNA expression was observed following WSSV injection (Fig. 3A), indicating successful viral infection. Notably, the mRNA levels of Src64B and PI3Kp85α in hemocytes were transiently upregulated at 6 h post-infection, then declined from 12 to 24 h post-infection compared with the PBS controls (Fig. 3B andC). Moreover, in both hemocytes and gills, PIP3 generation and Akt phosphorylation, which are key markers of PI3K-Akt pathway activation (10), were significantly elevated from 6 to 24 h post-WSSV infection relative to the PBS controls (Fig. 3D through F). These findings suggest that the PI3K-Akt pathway is activated during the early stage of WSSV infection.
Given that Src and PI3Kp85ɑ are two critical components of the mammalian PI3K-Akt pathway (30,31), we investigated whether the WSSV IE1 protein regulates this pathway by interacting with P. vannamei Src64B and PI3Kp85α. We first examined the involvement of Src64B and PI3Kp85α in activating the PI3K-Akt pathway in shrimp during WSSV infection. Our results demonstrated that under WSSV infection, knockdown of Src64B or PI3Kp85α, or treatment of shrimp with the Src inhibitor Saracatinib or the PI3K inhibitor LY294002, significantly reduced PIP3 generation and Akt phosphorylation in hemocytes post-WSSV infection (Fig. S2A through H). These results underscore the evolutionary conservation of the PI3K-Akt pathway across species. Subsequently, we determined the role of IE1 in regulating this pathway. Knockdown of the IE1 gene in WSSV-infected shrimp resulted in a notable reduction in PIP3 generation and Akt phosphorylation in hemocytes and gills (Fig. 3G through J). Overall, our results indicate the essential role of IE1 in activating the Src-PI3K-Akt signaling upon WSSV infection.
## Src-PI3K-Akt signaling inhibits apoptosis and autophagy during WSSV infection
To investigate the role of the Src-PI3K-Akt signaling during infection, we performed comparative transcriptomic analyses following PI3Kp85α silencing in WSSV-infected shrimp. The results revealed significant upregulation of genes associated with proapoptotic and autophagy pathways in PI3Kp85α-silenced shrimp (Fig. S3A andB), suggesting that the PI3K pathway may act as a negative regulator of apoptosis and autophagy during infection. We then conducted RNA interference (RNAi) and overex pression experiments targeting Src64B and PI3Kp85α to validate this hypothesis. Silencing Src64B or PI3Kp85α in WSSV-infected shrimp led to a marked increase in the percentage of apoptotic cells and Caspase activation, as demonstrated by flow cytome try and Caspase 3/7 activity assays (Fig. 4A through E). Conversely, overexpression of these proteins in High Five cells reduced apoptosis (Fig. 4H through K). Autophagy levels were assessed using western blotting and Monodansylcadaverine (MDC) staining. Silencing Src64B or PI3Kp85α in WSSV-infected shrimp significantly increased the ratio of γ-aminobutyric acid receptor-associated protein II/I (GABARAP-II/I), a marker of autoph agy induction (32), and promoted autophagosome formation (Fig. 4F andG). Overex pression of these proteins, however, suppressed autophagy (Fig. 4L andM). Furthermore, inhibitor experiments showed that treating WSSV-infected shrimp with the Src inhibitor Saracatinib or the PI3K inhibitor LY294002 induced apoptotic and autophagic processes (Fig. 4N andO). These results confirm that the Src-PI3K signaling inhibits apoptosis and autophagy during WSSV infection.
To clarify whether the Src-PI3K signaling exerts its effects through its downstream effector Akt, we treated WSSV-infected shrimp with a specific Akt inhibitor MK2206. The results demonstrated that this inhibitor induced apoptosis and autophagy compared with the control group (Fig. S4A andB). Moreover, when cells overexpressing Src64B or PI3Kp85α were further treated with the Akt inhibitor, apoptosis and autophagy levels were significantly increased compared with untreated overexpressing cells (Fig. S4C through E). These results strongly indicate that the Src-PI3K-Akt signaling suppresses apoptosis and autophagy under WSSV challenge.
## IE1 inhibits apoptosis and autophagy via the Src-PI3K-Akt signaling
Building on the above findings, we investigated the role of the WSSV IE1 protein in regulating apoptosis and autophagy. Using RNAi and overexpression experiments, we evaluated the effect of IE1 on these processes. In WSSV-infected shrimp, knockdown of IE1 significantly increased the percentage of apoptotic cells and enhanced caspase 3/7 activity (Fig. 5A through D) while also elevating the GABARAP II/I ratio and promoting autophagosome formation (Fig. 5E andF). Conversely, overexpression of IE1 in High Five cells suppressed both apoptosis and autophagy (Fig. 5G through L). To assess whether the Y 129 FTS tyrosine motif is required for the suppressive function of IE1 on apoptosis and autophagy, we overexpressed the IE1-Y129F mutant in High Five cells. In contrast to the wild-type IE1, the IE1-Y129F mutant failed to suppress either apoptosis or autophagy, evidenced by the observation that apoptosis and autophagy levels in cells expressing IE1-Y129F were comparable with those in empty vector controls (Fig. 5M through O). This indicates that an intact Y129 motif is essential for IE1-mediated inhibition of apoptosis and autophagy.
To further examine whether IE1 exerts its suppressive effect through activation of the Src-PI3K-Akt signaling pathway, we treated IE1-overexpressing cells with inhibitors targeting Src, PI3K, and Akt. The results showed that apoptosis and autophagy levels were significantly higher in inhibitor-treated cells compared with untreated controls (Fig. 6A through L). Taken together, these results demonstrate that IE1 inhibits apoptosis and autophagy via the Src-PI3K-Akt signaling pathway during WSSV infection.
## IE1-induced Src-PI3K-Akt signaling facilitates WSSV infection by suppressing apoptosis and autophagy
To determine the role of the IE1-induced Src-PI3K-Akt signaling in WSSV infection, we performed individual knockdowns of IE1, Src64B, and PI3Kp85α in WSSV-infected shrimp and assessed their impact on viral infection. The results showed that silencing these genes led to a significant reduction in WSSV VP28 gene expression and viral copy numbers compared with the control groups (Fig. 7A through G), indicating the essential role of this IE1-induced signaling axis in WSSV infection. This conclusion was further supported by inhibitor studies, where treatment of WSSV-infected shrimp with inhibitors targeting Src, PI3K, or Akt also resulted in a significant decrease in WSSV copy numbers relative to untreated controls (Fig. 7H). Furthermore, to determine whether the IE1induced Src-PI3K-Akt signaling promotes WSSV infection by inhibiting apoptosis and
## DISCUSSION
Previous research has demonstrated that the PI3K-Akt pathway facilitates WSSV infection by reprogramming metabolic processes, maintaining cellular redox balance, and enhancing lipid synthesis in shrimp (15)(16)(17). However, the precise mechanism through which WSSV activates this pathway has remained elusive. Here, our study identifies the viral immediate-early protein IE1 as the key trigger. We reveal that IE1 binds to and activates the host kinase Src64B via its Y 129 FTS tyrosine motif, leading to the recruitment of PI3Kp85α to the plasma membrane and initiation of the PI3K-Akt signaling cascade. This IE1-triggered activation of the Src-PI3K-Akt axis suppresses host apoptotic and autophagic responses, thereby establishing a cellular environment conducive to viral proliferation. Critically, we provide direct functional evidence linking this molecular mechanism to phenotypic outcome: the IE1-Y129F mutant, which is impaired in Src64B binding and activation, fails to suppress apoptosis and autophagy. This establishes Y129-mediated activation of the Src-PI3K-Akt pathway as the pivotal upstream event through which WSSV neutralizes these essential host defense mechanisms. The residual PIP3 and pAkt signals observed following IE1 knockdown likely reflect incomplete gene silencing and/or the contribution of alternative WSSV-mediated PI3K-Akt path way activation mechanisms. Nonetheless, the marked reduction in pathway activation underscores the essential and direct role of IE1 in hijacking this specific signaling axis upon WSSV infection. Collectively, our findings thus offer the first mechanistic explana tion of how WSSV coopts the host PI3K-Akt pathway via IE1 to achieve immune evasion. The PI3K-Akt pathway is a core regulator of cellular homeostasis, making it a prime target for viral manipulation. Viruses such as Influenza A virus, Hepatitis C virus, and Herpes simplex virus type 1 have been shown to activate the PI3K-Akt pathway to promote their proliferation (33)(34)(35). Our data confirm that WSSV likewise activates this pathway during the early stages of infection, as evidenced by increased PIP3 generation and Akt phosphorylation during this phase. Although the outcome is conserved, the strategies employed by viruses vary. For instance, the NS1 protein of the Influenza A virus and the NS5A protein of the Hepatitis C virus directly interact with the SH3 domain of PI3Kp85 to trigger its activation (34,36). In contrast, Newcastle disease virus utilizes a more indirect mechanism, where its V protein promotes the ubiquitin-mediated degradation of PHLPP2, thereby relieving the inhibition on Akt (37). A third, recurrent tactic involves the hijacking of Src family kinases. The tegument protein VP11/12 of Herpes simplex virus type 1 and the middle T antigen (MTAg) of polyomavirus contain tyrosine motifs that mediate binding to and activation of host Src kinases. The activated Src then phosphorylates tyrosine motifs within VP11/12 and MTAg, creating docking sites for PI3Kp85 recruitment and subsequent pathway activation (28,29). Notably, we found that the WSSV IE1 protein employs a mechanism analogous to that of VP11/12 and MTAg: IE1 associates with Src64B via its Y 129 FTS tyrosine motif, triggering Src64B activation, PI3Kp85 recruitment, and PI3K-Akt signaling. Furthermore, our domain analysis revealed that P. vannamei PI3Kp85α possesses two additional C1 domains and an N-terminal SAM domain, which are absent in its human ortholog. As these domains are known to mediate protein-protein and protein-lipid interactions in other systems (38,39), we speculate that these unique structural features may facilitate WSSV-spe cific interactions, potentially enhancing the recruitment of PI3Kp85α to the plasma membrane upon IE1-Src64B complex formation. This could represent an evolutionary adaptation of WSSV to its crustacean host, optimizing viral hijacking of the Src-PI3K-Akt axis. Future structural and mutational studies focusing on these domains will be essential to elucidate their precise roles in WSSV infection.
Having established how IE1 activates the pathway, we sought to define the spe cific downstream processes it modulates to support viral proliferation. Transcriptomic profiling following PI3Kp85α knockdown in WSSV-infected shrimp revealed a significant upregulation of apoptosis-and autophagy-related genes, suggesting that IE1-driven signaling acts to suppress both processes. This finding is highly relevant, as apopto sis and autophagy are well-established host defense mechanisms that restrict viral propagation (40,41). Their inhibition is a common viral evasion strategy, particularly critical in invertebrates like shrimp, which lack a vertebrate-like adaptive immune system and rely heavily on these innate cellular defenses. Indeed, WSSV has evolved a repertoire of strategies to counteract these responses. To inhibit apoptosis, it encodes proteins such as the E3 ligase WSSV222, which mediates ubiquitination and degradation of a host tumor suppressor-like protein (42); AAP-1 (WSSV449), which binds and inhibits the effector caspase PmCasp (43), and other anti-apoptotic factors, including WSSV134, WSSV322, and ORF390, which suppress caspase-dependent apoptosis (44,45). Addition ally, WSSV employs a viral miRNA, WSSV-miR-N24, to post-transcriptionally silence host caspase 8 and repress apoptosis (46). In parallel, WSSV subverts autophagy through the tegument protein VP26, which binds SNAP29 and disrupts SNARE complex assembly, thereby blocking autophagosome-lysosome fusion and autophagic flux (47). Our present study adds a novel and overarching layer to this understanding by revealing that WSSV, through its IE1 protein, actively suppresses both apoptosis and autophagy simultane ously by hijacking the host Src-PI3K-Akt signaling pathway. This mechanism is consistent with the established role of PI3K-Akt signaling in facilitating viral replication across diverse viruses, such as Influenza A virus and Newcastle disease virus, which activate this pathway for antiapoptotic effects (37,48), and Human papillomavirus type 16 and Rotavirus, which exploit it to inhibit autophagy (49,50).
Furthermore, our discovery of IE1 as the viral trigger of the Src-PI3K-Akt axis provides a potential mechanistic explanation for the previously reported WSSV-induced metabolic reprogramming. Earlier studies have established that WSSV infection hijacks the PI3K-Akt pathway to induce aerobic glycolysis (the Warburg effect) and lipid biosynthesis, processes crucial for viral replication (15)(16)(17). Our demonstration that IE1 is the key viral factor responsible for the early activation of this pathway makes it plausible that the IE1-Src-PI3K-Akt signaling cascade we delineated serves as the upstream driver of these metabolic alterations. Although the observed increase in apoptotic and autophagic cells following Src-PI3K pathway disruption may appear numerically limited, it is biologically significant in the context of viral infection, where even a partial attenuation of prosurvival signaling can substantially impair viral yield. Moreover, this effect represents only one facet of the pathway's role. We propose that by initiating this pathway, IE1 not only directly suppresses apoptosis and autophagy but also orchestrates a broader pro-viral state, including metabolic reprogramming. This integrated model suggests that WSSV, through a single viral protein, co-opts a central host signaling hub to achieve multi-faceted immune evasion: simultaneously blocking key cellular defense pathways and reprogramming cellular metabolism to fuel its proliferation. Future research aimed at directly linking IE1 expression to specific metabolic changes will be valuable in fully elucidating this coordinated evasion strategy.
In summary, our study reveals a novel mechanism by which WSSV IE1 co-opts host Src-PI3K-Akt signaling to inhibit apoptosis and autophagy, thereby promoting viral proliferation (Fig. 8). These findings establish the Src-PI3K-Akt axis as a critical deter minant of WSSV pathogenesis and identify its components as promising therapeutic targets. Although this work defines the role of IE1 in initiating this signaling cascade, several questions remain. In particular, the structural basis for the IE1-Src64B interaction warrants further investigation. Based on the canonical activation mechanism of Src family kinases and our mutational analysis (28,35), we hypothesize that the phosphory lated Y 129 FTS motif of IE1 binds to the SH2 domain of Src64B, thereby disrupting its auto-inhibitory conformation and leading to kinase activation. Future studies employing techniques such as X-ray crystallography or cryo-EM to resolve the complex structure of IE1 bound to Src64B will be crucial to validate this hypothesis and clarify the precise molecular interfaces involved. Additionally, the potential contribution of other WSSV proteins to pathway subversion cannot be excluded, given that viruses often employ redundant mechanisms to ensure host manipulation.
## MATERIALS AND METHODS
## Experimental animals
Pacific white shrimp (P. vannamei), 5-8 g body weight, were obtained from a local shrimp farm in Shantou, Guangdong, China. Prior to experimentation, the animals were acclimated for at least 48 h in aerated, recirculating artificial seawater (salinity 5 ‰, 25°C) and fed a commercial diet once daily.
## Antibodies, inhibitors, and plasmids
The primary antibodies used in the study included mouse anti-FLAG antibody (Beyotime, China; Cat. No. AF519), mouse anti-V5 antibody (TransGen Biotech, China; Cat. No. HT401-01), rabbit anti-phosphotyrosine antibody (Blue Light, China; Cat. No. ICP9805), mouse anti-α-tubulin antibody (Sigma, USA; Cat. No. T5168), rabbit anti-p-Akt-S473 antibody (ABclonal, China; Cat. No. AP1208), rabbit anti-Akt antibody (Servicebio, China; Cat. No. GB15689-100), and rabbit anti-GABARAP antibody (Abcam, UK; Cat. No. ab109299), and mouse anti-PIP3 antibody (ThermoFisher Scientific; USA, Cat. No. A-21328). The mouse anti-VP28 antibody was provided by Prof. Shengkang Li's research group at Shantou University. For western blot analysis, secondary antibodies used were goat anti-mouse IgG-HRP (ThermoFisher Scientific, USA, Cat. No. G21040) and goat anti-rabbit IgG-HRP (ThermoFisher Scientific, USA; Cat. No. G21234). For the The V5-tagged IE1 and EGFP expression plasmids (IE1-V5 and EGFP-V5) were constructed in our previous study (26). The IE1 Y 129 FTS tyrosine motif mutant (IE1-Y129F-V5) was generated by overlap PCR and cloned into the pIZ-V5-His vector (Invitro gen, USA). For Src64B (GenBank accession No. MH397363.1) and PI3Kp85α (GenBank accession No. MH397365.1) expression plasmids, the full-length open reading frames (ORFs) were amplified and fused with a FLAG tag and then ligated into the pIZ-V5-His and pIEx-4 (Novagen, USA) vectors, respectively. All primers used for plasmid construc tion are listed in Table S1.
## Co-immunoprecipitation (co-IP) assay
High Five cells were seeded into six-well culture plates at a density of ~1 × 10⁶ cells/ well and maintained in Express Five SFM medium (ThermoFisher Scientific, USA; Cat. No. 10486025) overnight. For DNA transfection, 1 µg each of FLAG-tagged Src64B and PI3Kp85α expression plasmids (Src64B-FLAG and PI3Kp85ɑ-FLAG) were co-transfec ted with 1 µg of either IE1-V5, IE1-Y129F-V5, or EGFP-V5 (control) using FuGENE HD transfection reagent (Promega, USA; Cat. No. E2311) according to the manufacturer's protocol. At 48 h post-transfection, the cells were harvested and lysed with 200 µL of western and IP cell lysis buffer (Beyotime, China; Cat. No. P0013) for 20 min on ice, followed by centrifugation at 16,000 × g for 10 min at 4°C to obtain the supernatant. A 20 µL aliquot of the supernatant was reserved for direct western blot analysis, whereas the rest was incubated with 5 µL of anti-FLAG M2 magnetic beads (Sigma, USA; Cat. No. A2220) overnight at 4°C; the beads were then washed three times with cell lysis buffer and resuspended in 20 µL PBS plus 5 µL 5 × loading buffer before boiling at 100°C for 10 min for subsequent western blot analysis.
## Immunofluorescence assay
High Five cells seeded in six-well plates were transfected with 1 µg of plasmids encoding either IE1, Src64B, or PI3Kp85α individually, or co-transfected with IE1 plus Src64B or PI3Kp85α. After 48 h, the cells were harvested and transferred to confocal dishes for 1 h to allow adherence. Following attachment, cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.5% Triton X-100 in PBS for 20 min, and blocked with 3% BSA in PBS for 1 h at room temperature. Primary antibodies (mouse anti-V5 and rabbit anti-FLAG, 1:200 in 3% BSA) were incubated overnight at 4°C, followed by secondary antibody staining using Alexa Fluor 488-conjugated goat anti-mouse IgG and Alexa Fluor 555-conjugated donkey anti-rabbit IgG (both 1:400) for 1 h at room temperature. Nuclei were counterstained with Hoechst 33342 (Beyotime, China; Cat. No. C1022) before imaging with a Zeiss LSM confocal microscope.
## Detection of Src64B phosphorylation by western blot analysis
Due to the unavailability of specific phospho-antibodies against P. vannamei Src64B, we used a commercial phosphotyrosine antibody to detect Src64B phosphorylation in vitro. Briefly, High Five cells were co-transfected with the Src64B-FLAG construct along with either IE1-V5, IE1-Y129F-V5, or EGFP-V5 (control) for 48 h before harvesting. Following cell lysis and co-IP using anti-FLAG M2 magnetic beads as described above, the immunoprecipitated protein samples were resolved by 10% SDS-PAGE and transferred to PVDF membranes (Millipore, USA, Cat. No. IPVH00010) using a Mini Transblot wet transfer system (Bio-Rad Laboratories, USA). After blocking with 5% skimmed BSA in TBST (20 mM Tris, 150 mM NaCl, 0.1% Tween 20, pH 7.6) for 2 h at room temperature, membranes were probed with phosphotyrosine antibody overnight at 4°C. Following three TBST washes, membranes were incubated with secondary antibodies for 1 h at room temperature, washed again, and developed using ECL reagent (Millipore, USA, Cat. No. 203453), with signal detection performed on an Amersham Imager 600 (GE Healthcare).
## WSSV challenge, qPCR, western blot, and PIP3 level measurement
The WSSV stock was purified from WSSV-infected crayfish (Procambarus clarkii) and quantified by absolute qPCR as described previously (51). For the challenge experiment, shrimp were intramuscularly injected with 100 µL of WSSV virions (1 × 10 5 copies) using a sterile syringe with a 22-gauge needle, whereas control shrimp received an equal volume of sterile PBS. At 0, 6, 12, and 24 h post-infection, hemolymph was collected from each group into 600 µL of ice-cold anticoagulant solution (258 mM Sodium citrate dihydrate, 328 mM Sodium citrate, 110 mM glucose, 140 mM NaCl, pH 6.0) using a sterile syringe. The samples were immediately centrifuged at 800 × g for 10 min at 4°C to pellet the hemocytes for subsequent qPCR, western blot, and PIP3 level measurement analyses.
To determine the expression of Src64B, PI3Kp85ɑ, and IE1 following WSSV infection, the total RNA was extracted from hemocytes using an RNA rapid extraction kit (Fastagen, China; Cat. No. 220010) and reverse-transcribed into cDNA with a cDNA synthesis kit (TransGen Biotech, China; Cat. No. AT311-02). qPCR was performed using a reaction mixture containing 5 µL of 2 × RealStar Green Power Mix (GenStar, Beijing, China; Cat. No. A311), 1 µL each of forward and reverse primers, 1 µL of cDNA template, and 3 µL of nuclease-free water. Amplification was carried out on a LightCycler 480 (Roche, Switzerland) under the following conditions: 95°C for 10 min (initial denaturation), followed by 45 cycles of 95°C for 15 s and 60°C for 30 s. The relative mRNA expression levels, normalized to elongation factor 1-α (EF-1α), were calculated using the 2⁻ ΔΔCT method (all primers are listed in Table S1). For protein analysis, hemocyte lysates were subjected to western blot analysis using antibodies against Akt, including anti-phospho-Akt (Ser473) and total anti-Akt. To assess PIP3 level, hemocytes were resuspended in Insect-XPRESS medium (Lonza, Switzerland, Cat. No.12-730Q), plated on confocal dishes at 1 × 10⁵ cells/dish, and incubated for 1 h at 28°C before immunofluorescence staining with an anti-PIP3 antibody.
## Knockdown and overexpression of IE1, Src64B, and PI3Kp85α
The double-stranded RNAs (dsRNA) targeting IE1 (dsIE1), Src64B (dsSrc64B), and PI3Kp85α (dsPI3Kp85α) were synthesized in vitro using the HiScribe Quick High Yield RNA Synthesis Kit (New England Biolabs, USA; Cat. No. E2050S) according to the manufactur er's protocol, with dsEGFP serving as a negative control. For IE1 knockdown, shrimp were intramuscularly injected with 15 µg of dsIE1, followed by WSSV challenge (1 × 10⁵ copies) at 12 h post-injection. For Src64B and PI3Kp85α knockdown, shrimp received 10 µg of respective dsRNA, with WSSV inoculation (1 × 10 5 copies) performed 24 h later. Hemocytes were collected at 24 h post-infection for subsequent analyses, including Akt phosphorylation assessment, PIP3 level quantification, and evaluation of cell apoptosis, autophagy, and WSSV copy number.
Due to the lack of shrimp cell lines, overexpression studies were conducted using High Five cells. Cells were transfected with 2 µg of expression plasmids (IE1-V5, Src64B-FLAG, or PI3Kp85α-FLAG) using FuGENE HD transfection reagent according to the manufacturer's instructions. At 48 h post-transfection, the cells were harvested by centrifugation for downstream analyses of Akt phosphorylation, PIP3 level, apoptosis, and autophagy.
## Cell apoptosis assay
Apoptotic activity was assessed using a Dead Cell Apoptosis Kit with YO-PRO-1 and propidium iodide (PI) (Invitrogen, USA; Cat. No. V23201). Briefly, the cells were harvested, washed twice with cold PBS, and resuspended at a density of approximately 1 × 10⁶ cells/mL. For staining, 1 µL of YO-PRO-1 and 1 µL of PI were added per 1 mL of cell suspension, followed by incubation on ice for 20 min in the dark. Stained cells were then analyzed immediately using a flow cytometer (Accuri C6 Plus, BD Biosciences, USA). A minimum of 10,000 events was acquired per sample. Apoptotic cells (YO-PRO-1 positive, PI negative) were distinguished from live (double negative) and necrotic/late apoptotic (PI positive) cells, and the data were analyzed using FlowJo software (BD Biosciences). Meanwhile, Caspase-3/7 activity was measured using the Caspase-Glo 3/7 Assay System (Promega, USA; Cat. No. G8091). After cell counting with a hemocytometer, 1 × 10⁶ cells were aliquoted into 1.5 mL microcentrifuge tubes. An equal volume of Caspase-Glo3/7 reagent was added to each sample, followed by incubation at room temperature for 1 h in the dark with gentle shaking. Luminescence was quantified using a GloMax-Multi Detection System (Promega, USA).
## Cell autophagy assay
Autophagic activity was detected using Monodansylcadaverine (MDC) staining (Beyotime, China; Cat. No. C3018S) according to the manufacturer's protocol. In brief, cells were washed with 1 × wash buffer and incubated with 0.05 mM MDC dye-loading solution for 1 h at 37°C in the dark. After incubation, cells were washed twice with 1 × wash buffer. The mean fluorescence intensity per cell of autophagic vesicles was measured using a fluorescence microplate reader (Multiskan FC, Thermo Scientific) with excitation at 335 nm and emission at 512 nm. In parallel, autophagy induction was evaluated by western blot analysis using an anti-GABARAP antibody to monitor the conversion of GABARAP-I (cytosolic form) to GABARAP-II (lipidated form bound to autophagosomes).
## Quantification of WSSV copy number
WSSV copy numbers were quantified using absolute qPCR as described previously (52). Briefly, genomic DNA (gDNA) was extracted from shrimp hemocytes using the TIANamp Marine Animals DNA Kit (TIANGEN, Beijing, China) as per the manufacturer's instructions. The viral VP28 gene was cloned into the pMD19-T vector, and the copy number of the plasmid was determined based on its concentration and molecular weight. Standard samples were created by serially diluting the VP28 gene-containing plasmid. qPCR assays were then performed using qVP28-F/R primers (Table S1) with both standard dilutions and test gDNA samples, where Ct values from the standard samples were used to generate a standard curve, and viral copy numbers in test samples were calculated by comparing their Ct values with the standard curve.
## Inhibitor treatment assay
For overexpression studies in vitro, High Five cells were transfected with IE1-V5, Src64B-FLAG, or PI3Kp85-FLAG expression constructs, followed by treatment with pathway inhibitors (20 µM Saracatinib for Src inhibition, 20 µM LY294002 for PI3K inhibition, or 250 nM MK2206 for Akt inhibition). After 24 h of inhibitor treatment, cells were harvested by centrifugation for subsequent apoptosis and autophagy analyses. In vivo experiments, shrimp were injected with WSSV (1 × 10⁵ copies) along with the respective pathway inhibitors. Hemocytes were collected at 24 h post-infection for apoptosis, autophagy, and viral copy number quantification. In a parallel experiment, shrimp were injected with either control dsEGFP or target-specific dsRNAs (dsIE1, dsSrc64B, or dsPI3Kp85α). At 12 h for IE1 knockdown or 24 h for Src64B/PI3Kp85α knockdown, the animals were challenged with WSSV (1 × 10⁵ copies) and co-treated with either pharmacological inhibitors (20 µM Z-VAD-FMK [apoptosis inhibitor] or 2.5 mM 3-Methyladenine [autoph agy inhibitor]) or vehicle control (DMSO). Hemocytes were harvested at 24 hpi for viral copy number quantification. The cytotoxicity of all inhibitors used in both High Five cells and shrimp was evaluated using a CCK-8 kit (Beyotime, China, Cat. No. C0037) according to the manufacturer's instructions. The results confirmed that the applied concentrations did not induce significant cytotoxicity under our experimental conditions (Fig. S5)
## Statistical analysis
All data represent mean ± standard deviation (SD) from at least three independent biological replicates. Statistical comparisons were performed using two-tailed Student's t-tests. Data visualizations were conducted using GraphPad Prism 9.0 (GraphPad Software, USA).
## References
1. Dey, Dugassa, Hinzano et al. (2020) "Causative agent, diagnosis and management of white spot disease in shrimp: a review" *Reviews in Aquaculture*
2. Desrina, Prayitno, Verdegem et al. (2022) "White spot syndrome virus host range and impact on transmission" *Reviews in Aquaculture*
3. (2026) *Full-Length Text Journal of Virology*
4. Wang, Hirono, Maningas et al. (2019) "ICTV virus taxonomy profile: Nimaviridae" *J Gen Virol*
5. Yang, He, Lin et al. (2001) "Complete genome sequence of the shrimp white spot bacilliform virus" *J Virol*
6. Van Hulten, Witteveldt, Peters et al. (2001) "The white spot syndrome virus DNA genome sequence" *Virology (Auckl)*
7. Tsai, Wang, Leu et al. (2004) "Genomic and proteomic analysis of thirty-nine structural proteins of shrimp white spot syndrome virus" *J Virol*
8. Li, Weng, He (2019) "WSSV-host interaction: host response and immune evasion" *Fish & Shellfish Immunology*
9. Tran, Liang, Li et al. (2023) "Cellular responses in crustaceans under white spot syndrome virus infection" *Fish Shellfish Immunol*
10. Vanhaesebroeck, Stephens, Hawkins (2012) "PI3K signalling: the path to discovery and understanding" *Nat Rev Mol Cell Biol*
11. Han, Lin, Hu (2024) "Regulation of PI3K signaling in cancer metabolism and PI3K-targeting therapy" *Transl Breast Cancer Res*
12. Manning, Cantley (2007) "AKT/PKB signaling: navigating downstream" *Cell*
13. Dunn, Connor (2012) "HijAkt: The PI3K/Akt pathway in virus replication and pathogenesis" *Prog Mol Biol Transl Sci*
14. Diehl, Schaal (2013) "Make yourself at home: viral hijacking of the PI3K/Akt signaling pathway" *Viruses*
15. Blanco, Cameirao, López et al. (2020) "Phosphatidyli nositol-3-kinase-Akt pathway in negative-stranded RNA virus infection: a minireview" *Arch Virol*
16. Su, Huang, Chen et al. (2014) "An invertebrate Warburg effect: a shrimp virus achieves successful replication by altering the host metabolome via the PI3K-Akt-mTOR pathway" *PLoS Pathog*
17. Chen, Lee, Huang et al. (2016) "Six Hours after Infection, the metabolic changes induced by WSSV neutralize the host's oxidative stress defenses" *Sci Rep*
18. Hsieh, Chen, Li et al. (2015) "To complete its replication cycle, a shrimp virus changes the population of long chain fatty acids during infection via the PI3K-Akt-mTOR-HIF1α pathway" *Developmental & Comparative Immunology*
19. Li, Li, Ke et al. (2009) "Identification of the immediate-early genes of white spot syndrome virus" *Virology (Auckl)*
20. Lin, Huang, Xu et al. (2011) "Identification of three immediate-early genes of white spot syndrome virus" *Arch Virol*
21. Liu, Chang, Wang et al. (2005) "Microarray and RT-PCR screening for white spot syndrome virus immediate-early genes in cycloheximide-treated shrimp" *Virology (Auckl)*
22. Liu, Chang, Wang et al. (2008) "Transactiva tion, dimerization, and DNA-binding activity of white spot syndrome virus immediate-early protein IE1" *J Virol*
23. Ran, Bian, Ji et al. (2013) "White spot syndrome virus IE1 and WSV056 modulate the G 1 /S transition by binding to the host retinoblastoma protein" *J Virol*
24. Yao, Ruan, Lu et al. (2016) "Shrimp STAT was hijacked by white spot syndrome virus immediate-early protein IE1 involved in modulation of viral genes" *Fish & Shellfish Immunology*
25. Wang, Li, Weng et al. (2020) "White spot syndrome virus establishes a novel IE1/JNK/c-jun positive feedback loop to drive replication"
26. Sun, Ruan, Zhou et al. (2017) "Characterization and function of a β-catenin homolog from Litopenaeus vannamei in WSSV infection" *Developmental & Comparative Immunology*
27. Wang, Wei, Wu et al. (2022) "Proteomics analysis reveals a critical role for the WSSV immediate-early protein IE1 in modulating the host prophenoloxi dase system" *Virulence*
28. Lu, Chen, Zhang et al. (2025) "White spot syndrome virus IE1 protein blocks the integrin-FAK signaling to enhance viral infection in shrimp" *Fish & Shellfish Immunol ogy*
29. Strunk, Saffran, Wu et al. (2013) "Role of herpes simplex virus VP11/12 tyrosine-based motifs in binding and activation of the Src family kinase Lck and recruitment of p85, Grb2, and Shc" *J Virol*
30. Schaffhausen, Roberts (2009) "Lessons from polyoma middle T antigen on signaling and transformation: a DNA tumor virus contribu tion to the war on cancer" *Virology (Auckl)*
31. Luo, Zheng, Wei et al. (2021) "The fatty acid receptor CD36 promotes HCC progression through activating Src/PI3K/AKT axis-dependent aerobic glycolysis" *Cell Death Dis*
32. Liu, Guo, Jing et al. (2018) "Hepatitis B virus core protein promotes hepatocarcinogenesis by enhancing Src expression and activating the Src/PI3K/Akt pathway" *FASEB J*
33. Weidberg, Shvets, Shpilka et al. (2010) "LC3 and GATE-16/GABARAP subfamilies are both essential yet act differently in autophagosome biogenesis" *EMBO J*
34. Hale, Jackson, Chen et al. (2006) "Influenza A virus NS1 protein binds p85beta and activates phosphatidylinositol-3kinase signaling" *Proc Natl Acad Sci*
35. Street, Macdonald, Crowder et al. (2004) "The Hepatitis C virus NS5A protein activates a phosphoinositide 3-kinase-dependent survival signaling cascade" *J Biol Chem*
36. Wagner, Smiley (2011) "Herpes simplex virus requires VP11/12 to activate Src family kinase-phosphoinositide 3-kinase-Akt signaling" *J Virol*
37. Shin, Liu, Tikoo et al. (2007) "Influenza A virus NS1 protein activates the phosphatidylinositol 3-kinase (PI3K)/Akt pathway by direct interaction with the p85 subunit of PI3K" *J Gen Virol*
38. Fan, Liang, Ren et al. (2024) "Newcastle disease virus activates the PI3K/AKT signaling pathway by targeting PHLPP2 degradation to delay cell apoptosis and promote viral replication" *Vet Microbiol*
39. Schultz, Ponting, Hofmann et al. (1997) "SAM as a protein interaction domain involved in developmental regulation" *Protein Sci*
40. Ono, Fujii, Igarashi et al. (1989) "Phorbol ester binding to protein kinase C requires a cysteine-rich zincfinger-like sequence" *Proc Natl Acad Sci*
41. Orzalli, Kagan (2017) "Apoptosis and necroptosis as host defense strategies to prevent viral infection" *Trends Cell Biol*
42. (2026) *Full-Length Text Journal of Virology*
43. Choi, Bowman, Jung (2018) "Autophagy during viral infection -a double-edged sword" *Nat Rev Microbiol*
44. He, Fenner, Godwin et al. (2006) "White spot syndrome virus open reading frame 222 encodes a viral E3 ligase and mediates degradation of a host tumor suppressor via ubiquitination" *J Virol*
45. Leu, Chen, Lin et al. (2010) "Molecular mechanism of the interactions between white spot syndrome virus anti-apoptosis protein AAP-1 (WSSV449) and shrimp effector caspase" *Developmental & Comparative Immunology*
46. Lertwimol, Sangsuriya, Phiwsaiya et al. (2014) "Two new anti-apoptotic proteins of white spot syndrome virus that bind to an effector caspase (PmCasp) of the giant tiger shrimp Penaeus (Penaeus) monodon" *Fish Shellfish Immunol*
47. Wang, Hu, Yi et al. (2004) "ORF390 of white spot syndrome virus genome is identified as a novel anti-apoptosis gene" *Biochem Biophys Res Commun*
48. Huang, Cui, Zhang (2014) "Involvement of viral MicroRNA in the regulation of antiviral apoptosis in shrimp" *J Virol*
49. Liu, Jian, Jing et al. (2024) "The crustacean DNA virus tegument protein VP26 binds to SNAP29 to inhibit SNARE complex assembly and autophagic degradation" *J Virol*
50. Ehrhardt, Wolff, Pleschka et al. (2007) "Influenza A virus NS1 protein activates the PI3K/Akt pathway to mediate antiapoptotic signaling responses" *J Virol*
51. Surviladze, Sterk, Deharo et al. (2013) "Cellular entry of human papillomavirus type 16 involves activation of the phosphatidyli nositol 3-kinase/Akt/mTOR pathway and inhibition of autophagy" *J Virol*
52. Yin, Dang, Zhou et al. (2018) "PI3K-Akt-mTOR axis sustains rotavirus infection via the 4E-BP1 mediated autophagy pathway and represents an antiviral target" *Virulence*
53. Xie, Li, Xu et al. (2005) "A simple and efficient method for purification of intact white spot syndrome virus (WSSV) viral particles" *Virus Res*
54. Zhan, Aweya, Wang et al. (2018) "Litopenaeus vannamei attenuates white spot syndrome virus replication by specific antiviral peptides generated from hemocyanin" *Developmental & Comparative Immunology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724219&blobtype=pdf | # Pathogenic dengue virus TW2015 strain infection triggers anaerobic glycolysis and enhances mortality in diabetic mice
Yi-Ping Kuo, Shih-Syong Dai, En-Ju Lin, Wann-Neng Jane, Wei-Hsiang Tsai, Wan-Ting Tsai, Zaida Imana, Wan-Ju Tung, Yu-Siang Su, Chih-Feng Tien, Chia-Yi Yu, Chun-Hong Chen, Guann-Yi Yu
## Abstract
Dengue virus (DENV) strains with high pathogenicity and transmissibility pose significant public health challenges, especially in tropical and subtropical regions. Underlying conditions such as diabetes mellitus and renal diseases significantly increase the risk of severe dengue. The DENV-2 strain, responsible for a severe outbreak in Taiwan in 2015, exhibits enhanced pathogenicity and transmissibility in a mosquito-mouse transmission model. In this study, we demonstrated that pathogenic DENV infection leads to elevated lactate levels and hypoglycemia in mice, correlating with increased mortality in streptozotocin-induced diabetic models. In infected cells, pathogenic DENV induces rapid eIF2α phosphorylation, extensive ER membrane aggregation, disrupted calcium transfer to mitochondria, and mitochondrial dysfunction, which may contribute to excessive lactate production. Notably, inhibition of lactate production reduced viremia and mortality in mice. These findings highlight the role of metabolic dysregulation in DENV pathogenesis and provide insights into the mechanisms driving severe dengue, particularly in patients with underlying comorbidities. IMPORTANCE DENV is a mosquito-borne virus that can cause severe illness, particu larly in tropical and subtropical regions. In 2015, a strain of DENV-2 caused a major outbreak in Taiwan with high mortality rates. People with conditions like diabetes or kidney disease were more likely to develop severe dengue. In our study, we found that this highly pathogenic virus caused mice to have high levels of lactate and low blood sugar before death. In diabetic mice, the virus caused even higher death rates. The virus impairs cellular energy production by disrupting communication between the endoplasmic reticulum and mitochondria, potentially leading to excessive lactate accumulation. Blocking lactate production helped reduce viremia and death rate. These findings suggest that the virus's impact on metabolism may play a role in severe illness, especially for people with pre-existing health issues.
2015, resulting in 43,784 confirmed dengue cases, high mortality, and severe complica tions (3). This virus strain demonstrates high pathogenicity in Stat1 -/-mice, which are deficient in the signal transducer and activator of transcription 1 (STAT1)-a critical mediator downstream of type I and II interferon pathways-and in AGB6 mice, which lack functional type I and II interferon receptors. The TW2015 strain also demonstrates high transmissibility between mosquitoes and vertebrate hosts (4).
Epidemiological studies of the 2015 dengue outbreak in Taiwan indicate that a high proportion of fatal cases occurred in older adults with comorbid conditions, including diabetes mellitus and chronic kidney disease (3). Diabetes has been identified as a comorbidity in dengue hemorrhagic fever and dengue shock syndrome in several retrospective studies (5)(6)(7)(8)(9). Diabetes is a metabolic disorder characterized by defective glucose homeostasis. Glucose serves as a primary energy source for mammalian cells, and glycolysis is a central metabolic pathway that converts glucose into pyruvate while generating ATP and NADH. Under aerobic conditions, pyruvate typically enters mitochondria to fuel the tricarboxylic acid (TCA) cycle and oxidative phosphorylation, maximizing ATP production. However, during hypoxia, mitochondrial dysfunction, or high biosynthetic demand (as in cancer or viral infection), pyruvate is predominantly converted into lactate via lactate dehydrogenase (LDH), a process referred to as anaerobic glycolysis (10). Hyperglycemia is shown to exacerbate DENV infection by facilitating viral translation (11).
Investigating whether DENV infection exacerbates metabolic stress in diabetic hosts, potentially contributing to higher mortality, is crucial. This study demonstrates that DENV TW2015 infection leads to the accumulation of lactate, resulting in enhanced mortality in diabetic mice. The underlying mechanisms were further explored.
## RESULTS
## Pathogenic DENV infection disturbs glucose metabolism in Stat1 -/-mice
During glycolysis, glucose is metabolized into pyruvate. Under aerobic conditions, pyruvate enters to participate in the mitochondria for the citric acid cycle, while under anaerobic conditions, it is converted to lactate in the cytoplasm. The TW2015 strain has been shown to exhibit higher pathogenicity in vivo compared to other DENV-2 strains, such as 16681 and NGC (4). To investigate the impact of DENV infection on glucose metabolism, female Stat1 -/-mice were infected intradermally with DENV-2 16681 and TW2015 strains (5 × 10 4 PFU/mouse). Subsequent monitoring included body weight, clinical score, morbidity, and levels of blood glucose and lactate. Both DENV-2 strains caused significant body weight loss, with TW2015-infected mice showing greater weight loss than those infected with the 16681 strain (Fig. 1A). 16681-infected mice showed mild symptoms (score ≤2), while TW2015-infected mice exhibited more severe signs (score ≥3), indicating higher pathogenicity of the TW2015 strain (Fig. 1B). Serum virus titers were significantly higher in TW2015-infected mice than in 16681-infected mice (Fig. 1C). Blood glucose levels in DENV-2-infected mice notably decreased, particularly at 9 DPI (Fig. 1D). Concurrently, blood lactate levels were significantly elevated in TW2015infected mice at the same time point (Fig. 1E), suggesting that pathogenic DENV-2 infection may promote enhanced anaerobic glycolysis. Consistent with this, glycoly sis-related gene activation was observed in DENV-2 TW2015-infected mice (Fig. S1). Specifically, Hypoxia-inducible factor 1 α (Hif1a), Hexokinase 2 (Hk2), and Lactate Dehydro genase A (Ldha) were upregulated in the spleen at 6 DPI. In the liver, the expression of Glucose transporter 1 (Glut1), Hexokinase 1 (Hk1), and Hk2 was also significantly increased at the same point.
Given the increased risk of severe dengue and mortality associated with diabetes mellitus, the impact of DENV infection was further examined in diabetic mice. Female Stat1 -/-mice were administered a single intraperitoneal injection of streptozotocin (STZ, 200 mg/kg) to induce pancreatic islet β-cell destruction (12). Blood glucose levels in STZ-treated mice reached approximately 300 mg/dL, nearly double those of control mice (average 136 mg/dL) (Fig. S2A). Subsequently, the mice were infected with the TW2015 virus (5 × 10⁴ PFU/mouse, intradermal). Despite experiencing slightly less severe body weight loss than controls, STZ-treated mice exhibited a 100% mortality rate, in contrast to the 50% mortality rate observed in control mice (Fig. 2A andB). Serum virus titers on day 3 post-infection were reduced in STZ-treated mice (Fig. 2C). Male mice, known to be more sensitive to STZ treatment, exhibited blood glucose levels reaching up to 600 mg/dL and reduced blood lactate levels within a week (Fig. S2B). Following DENV-2 infection, STZ-treated male mice demonstrated a 100% mortality rate compared to a 25% mortality rate in control mice (Fig. 2D andE).
When the kinetics of virus titer was examined, TW2015-infected male mice exhibited significantly higher serum viral loads compared to the STZ + TW2015 group at 3 DPI (Fig. 2F). By 5 DPI, more than half of the TW2015-infected mice had undetectable viral titers, whereas the STZ + TW2015 group maintained relatively high levels (ranging from 10² to 10⁴ ffu/mL), suggesting that diabetic mice exhibit delayed viral replication kinetics. In the STZ + TW2015 group, blood glucose levels remained consistently elevated throughout the course of infection, except during the moribund state observed at 7 DPI (Fig. 2G). Additionally, lactate levels in STZ-treated mice remained low, even after DENV-2 infection (Fig. 2H), indicating impaired glycolytic flux under diabetic conditions.
To explore potential mechanisms associated with the high mortality observed in TW2015-infected diabetic mice, gene expression in mouse organs at 7 DPI was analyzed using real-time RT-qPCR. Several glycolysis-related genes were significantly upregulated in STZ-treated mice. In particular, Hk2, Pgk1 (phosphoglycerate kinase 1), and Ldha expression were elevated in the spleen, while Glut1 (glucose transporter 1), Hk2, and Ldha were upregulated in the liver. Notably, this elevated expression was not further enhanced by DENV-2 TW2015 infection in the STZ-treated mice but was elevated in control mice after TW2015 infection (Fig. 2I andJ). Impaired glucose regulation may contribute to the increased mortality observed in DENV-2-infected STZ-treated mice.
## Highly pathogenic DENV-2 virus infection increases lactate production in cells
The pathogenic DENV-2 TW2015 virus can penetrate extralymphatic organs, such as the lungs and intestines in mice (4). Lactate induction in DENV-infected Stat1 -/-mice may arise directly from infected cells or indirectly due to hypoxia associated with pulmonary dysfunction. To investigate this, the potential of the pathogenic DENV-2 virus to induce lactate production was further examined. When DENV-2 viruses were amplified in Vero cells, the culture medium of TW2015-infected cells became distinctly acidic by day 4 post-infection, as evidenced by the phenol red color change in the medium (Fig. 3A). Direct pH measurements further confirmed a significant decrease in the pH of the TW2015-infected cell culture medium on days 3 and 4 post-infection (Fig. S3). NGC-infected cells exhibited a slight pH decrease on day 4 post-infection, while 16681-infected cells showed no significant pH changes compared to mock-infected cells. All tested strains replicated efficiently in Vero cells; however, TW2015-infected cells showed relatively lower titers compared to the other two strains at 3 and 4 DPI (Fig. 3B). Since lactate, a byproduct of anaerobic glycolysis, can lower pH in culture medium, lactate levels in the medium of DENV-2-infected cells were measured on day 4. As shown in Fig. 3C, DENV infections in Vero cells led to lactate accumulation compared to mock infections, with TW2015-infected cells secreting a significantly greater amount of lactic acid into the medium. This pH reduction associated with the TW2015 virus infection was also observed in Huh7 hepatoma cells (Fig. S4). Overall, these findings suggest that pathogenic DENV-2 viruses alter glucose metabolism in infected cells.
## Pathogenic DENV-2 infection disrupts mitochondrial function
DENV infection induces mitochondrial dysfunction (13)(14)(15). To determine whether mitochondrial dysfunction contributes to lactic acid production from anaerobic glycolysis, JC-1 dye staining was used to assess mitochondrial membrane potential in DENV-2-infected cells. In healthy mitochondria, the cell-permeable JC-1 dye forms red fluorescent J-aggregates, while in dysfunctional mitochondria, it remains as green fluorescent J-monomers. Fewer than 10% of mock-infected cells showed mitochondrial dysfunction. In contrast, significant mitochondrial dysfunction was observed in DENV-2infected cells by day 3 post-infection (Fig. 3D andE). The highly pathogenic TW2015 strain induced markedly greater mitochondrial dysfunction in Vero cells compared to the 16681 and NGC strains. Given the critical role of the prM-E region in TW2015 patho genesis (4), Vero cells were infected with recombinant 16681 viruses carrying C-prM-E, prM-E, or E segments from TW2015 and were subjected to JC-1 staining. Higher levels of mitochondrial dysfunction were detected in cells infected with 16681-TW2015(E), 16681-TW2015(prM-E), and 16681-TW2015(C-prM-E) viruses compared to those infected with 16681 virus (Fig. 3F andG). Mitochondrial dysfunction induced by highly patho genic DENV-2 strains, TW2015 and 16681-TW2015(C-PrM-E), may initiate as early as day 2 post-infection (Fig. S5A). Additionally, TW2015 virus infection also induced mitochondrial dysfunction in Huh7 cells (Fig. S5B). These results suggest that lactic acid production may be driven by mitochondrial dysfunction triggered by pathogenic DENV infection.
## Pathogenic DENV-2 virus infection impairs ER-mitochondria communication
DENV replication and assembly primarily occur in the endoplasmic reticulum (ER). To explore the relationship between TW2015 virus infection, mitochondrial dysfunction, and metabolic alterations, infected Huh7 cells were stained with antibodies for the DENV E protein and organelle markers for mitochondria (AIF) and ER (PDI). As shown in Fig. 4A, E protein expression in DENV-2-infected cells predominantly co-localized with the ER marker. In TW2015-infected cells, the ER structure appeared highly condensed and aggregated to one side of the cytoplasm, a feature less pronounced in cells infected with the 16681 strain. The mitochondrial staining patterns in DENV-2-infected cells showed no significant changes in shape or length.
The distribution of the E protein in cells infected with 16681-TW2015(E), 16681-TW2015(prM-E), and 16681-TW2015(C-PrM-E) viruses was analyzed at day 2 post-infec tion (Fig. 4B; Fig. S6). The E protein distribution in 16681-TW2015(C-PrM-E)-infected cells closely resembled that of TW2015-infected cells. E protein expression levels in cells infected with 16681-TW2015(E) and 16681-TW2015(prM-E) were relatively low, requiring extended exposure times for confocal microscopy imaging in Fig. 4B. The E protein distribution patterns for these two recombinant viruses were similar to the E protein pattern of the 16681 viruses. Image quantification revealed that approximately 70% of TW2015-infected cells exhibited condensed E protein distribution, compared to less than 30% in 16681-infected cells (Fig. 4C). These results suggest that pathogenic DENV-2 infection induces an ER structural alteration where the E protein is concentrated.
Ca 2+ is primarily stored in the ER and transported to mitochondria to regulate ATP production (16). DENV infection is shown to reduce ER-mitochondria contact site and regulate respiration (17). The potential disruption of Ca² + transport between the ER and mitochondria in TW2015-infected cells was further investigated. DENV-2-infected cells were stained with Mag-Fluo-4 AM (18) and Rhod-2 AM (19) to trace the Ca 2+ levels in the ER and mitochondria, respectively. The ER Ca² + levels in TW2015-infected cells were obviously increased at 2 DPI (Fig. 4D andE). In contrast, Rhod-2 signals indicating mitochondrial Ca² + levels were significantly lower in TW2015-infected cells compared to mock or 16681-infected cells. These results indicate that TW2015 infection disrupts Ca² + transport from the ER to mitochondria. In summary, pathogenic DENV-2 infection induces ER alterations and disrupts ER-mitochondria Ca² + transport, both of which may contribute to mitochondrial dysfunction.
## Pathogenic DENV-2 infection induces ER membrane aggregation
Membrane alterations are commonly associated with virus replication and assembly (20,21). DENV infection induces ER-derived membranous structures containing nonstructural proteins for RNA replication (20,22). To investigate the structural changes induced by pathogenic DENV-2 infection, Huh7 cells infected with 16681, TW2015, and recombi nant viruses were analyzed by transmission electron microscopy at 48 h post-infection.
As shown in Fig. 5A, 16681-infected cells exhibited vacuoles compared to mock-infec ted cells. In contrast, TW2015-infected cells displayed numerous large vacuoles, with aggregated ER membranes localized at one side near the nuclei (Fig. 5A). Addition ally, mitochondria in TW2015-infected cells appeared condensed, as highlighted in the enlarged image in Fig. 5B, with most mitochondria excluded from the ER aggre gates. Large vacuoles were also observed in 16681-TW2015(C-prM-E)-infected cells, although the ER aggregates were relatively dispersed (Fig. 5A). Cells infected with 16681-TW2015(prM-E) and 16681-TW2015(E) did not exhibit pronounced ER condensa tion but still contained numerous vacuoles. Quantification of ultrastructural changes under different infection conditions was performed using scanning TEM (STEM) imaging technology by Borries (Singapore) (Fig. S7). TW2015-infected cells exhibited a higher percentage of cells with small and large vacuoles (49.4%) compared to mock-infected controls (8.2%) and 16681-infected cells (38.5%). Recombinant viruses 16681-TW2015 (C-prM-E), (prM-E), and (E) induced moderate levels of vacuole formation (35.1%-38.9%; Fig. 5C). Regarding ER aggregation, only the TW2015 strain induced prominent ER condensation (4.6%), with lower frequencies observed in 16681-TW2015 (C-prM-E) (1.7%) and (prM-E) (1.3%) (Fig. 5D). Minimal or no ER aggregation was observed in the mock, 16681, and 16681-TW2015 (E) groups. The relationship between the accumulation of autophagosome-or lysosome-like vacuoles and mitochondrial dysfunction in the context of pathogenic viral infection remains to be further investigated.
## Pathogenic DENV-2 infection stimulates eIF2α phosphorylation-associated stress response
RNA virus infection can induce ER stress and activate the unfolded protein response (UPR), leading to an autophagic response (23)(24)(25). It is plausible that ER stress contrib utes to the ER aggregation observed in pathological DENV infection. To investigate whether the ER stress response is differentially activated by pathogenic DENV, the protein expression levels of ER stress regulatory proteins (GRP78, HSP70, and ERp44), the eIF2α phosphorylation level, and the autophagy marker LC3 were evaluated in DENV-2-infected cells. In Fig. 6A, the eIF2α phosphorylation level increased in TW2015infected cells at both 24 and 48 h post-infection, earlier than in 16681-infected cells. Autophagy was activated by both TW2015 and 16681 virus infections, as indicated by the elevated LC3B-II levels at 48 h post-infection. The protein expression levels of GRP78, HSP70, and ERp44 did not show significant changes in DENV-2-infected cells. Phosphor ylation of eIF2α leads to the inhibition of protein translation as a cellular mechanism to combat stress or viral invasion. Although the infections were carried out using the same multiplicity of infection (MOI = 1), virus production in the culture supernatant was higher in TW2015-infected cells compared to those infected with the 16681 strain (Fig. 6A).
To investigate whether disturbance of eIF2α phosphorylation affects ER aggrega tion induced by TW2015 infection, salubrinal, an inhibitor of eIF2α dephosphorylation that protects cells from ER stress (26), was added during DENV-2 infection. Salubrinal treatment in Huh7 cells was well tolerated, as evidenced by the absence of significant cytotoxic effects at the concentrations used (Fig. S8A). As indicated by the condensed DENV E protein localization (Fig. 6B), ER aggregation induced by TW2015 infection significantly decreased with salubrinal treatment. Virus replication and eIF2α phosphory lation were also reduced upon salubrinal treatment (Fig. S8). The results suggest that ER aggregation may be due to continuous eIF2α phosphorylation/dephosphorylation or high levels of viral protein accumulation in the ER.
During ER stress, the unfolded protein response (UPR) is activated, leading to eIF2α phosphorylation by PKR-like ER kinase (PERK) (27). To evaluate whether the UPR serves as a critical upstream event for ER alterations induced by TW2015 virus infec tion, ER chemical chaperones tauroursodeoxycholic acid (TUDCA) and 4-phenylbutyric acid (4-PBA) (28) were used during DENV-2 infection. Treatment with these chemical chaperones did not affect the aggregation pattern of the TW2015 E protein (Fig. S9 and S10), indicating that the UPR may not be essential for ER aggregation caused by pathogenic DENV-2 infection. eIF2α phosphorylation can also be triggered by vari ous stress responses, including virus infection-associated PKR activation, to modulate translation (29). Hence, PKR activation during DENV-2 infection was assessed via immunoblotting for phosphorylated PKR (Fig. 6C). Elevated levels of PKR phosphoryla tion, correlating with eIF2α phosphorylation, were observed at early time points (24 and 30 h post-infection) in TW2015-infected cells compared to 16681-infected cells. These findings suggest that PKR activation precedes eIF2α phosphorylation during pathogenic DENV-2 infection.
## Lactate production enhances DENV-2 virus penetration in mice
With mitochondrial dysfunction, TW2015 virus-infected cells might rely on anaerobic glycolysis for energy production. Lactate dehydrogenase (LDH) converts pyruvate from glycolysis to lactate. Whether lactate accumulation affects virus replication in vitro was tested using LDH inhibitor sodium oxamate during virus infection. While sodium oxamate treatment slightly reduced cell viability, it had no significant effect on virus production in Huh7 cells (Fig. 7A andB). To assess the role of lactate secretion during pathogenic virus infection in vivo, Stat -/-mice were treated with sodium oxamate (0.75 g/kg) (30) 2 h prior to TW2015 infection and then daily post-infection. As shown in Fig. 7C, serum virus titers at 3 DPI were reduced in the sodium oxamate-treated group. Although body weight did not change significantly (Fig. 7D), sodium oxamate treatment (n = 6) reduced mortality from 57.1% in the PBS-treated group (n = 7) to 33% (Fig. 7E), suggesting that lactate overproduction may partially exacerbate DENV-induced pathology.
## DISCUSSION
The study identified that infection with highly pathogenic DENV-2 induces lactate accumulation both in vivo and in vitro and further examined the intracellular alterations driving the switch to anaerobic glycolysis. In healthy cells, the ER and mitochondria are connected through contact sites that facilitate the exchange of lipids and Ca² + , ensuring proper regulation of energy production and glucose homeostasis (31). Mitochondrial Ca² + levels regulate dehydrogenase activity in the tricarboxylic acid (TCA) cycle, directly impacting ATP production (32,33). Our findings show that pathogenic DENV-2 infection results in ER aggregation, impaired Ca² + transfer from the ER to mitochondria, and mitochondrial dysfunction. PKR activation and eIF2α phosphorylation are dominant ER stress responses induced in the infected cells. All these alterations may drive energy production to anaerobic glycolysis and lead to lactate accumulation in infected cells (Fig. 8). The accumulation of lactate from anaerobic glycolysis worsens disease severity in vivo, as inhibition of lactate production in infected mice reduces both viral titers and mortality rates. Moreover, mice under diabetic conditions exhibit heightened sensitivity to DENV-2 infection, with increased mortality, suggesting that functional glycolysis is essential for survival during infection. This study provides insights into why diabetes mellitus significantly increases the risk of mortality in severe dengue cases.
The contact sites between the ER and mitochondria, known as mitochondria-asso ciated membranes (MAMs), are essential for energy production and various cellular functions. Ca² + transport from the ER to mitochondria is facilitated by a channel complex involving inositol 1,4,5-trisphosphate receptors (IP3Rs) on the ER membrane, voltage-dependent anion channel 1 (VDAC1) on the outer mitochondrial membrane, and glucose-regulated protein 75 (GRP75) (16,34). Disruption of MAMs suppresses Ca² + transfer from the ER to mitochondria, leading to reduced ATP production. During DENV-2 infection, the protein expression levels of IP3R1, GRP75, and VDAC1 remained unchanged (data not shown). In TW2015-infected cells, the ER membrane becomes tightly aggregated, reducing its interaction with mitochondria. This may impair the formation of MAMs and limit the assembly of Ca² + transport complexes. Consistently, Freppel et al. reported that DENV infection reduces ER-mitochondria contact sites and disrupts mitochondrial respiratory metabolism at 48 and 72 h post-infection (17).
DENV infection is known to induce the formation of ER-derived convoluted membra nous structures within 24 hours post-infection. These structures, typically 2-3 µm in size, are densely packed with several nonstructural proteins and double-stranded RNAs (20,22). In contrast, ER aggregates observed in TW2015-infected cells at 48 h post-infec tion range from 7 to 12 µm in size, suggesting a more extensive ER rearrangement likely driven by infection-induced cellular stress at later stages of infection. To further elucidate the dynamics of this process, immunostaining coupled with quantitative imaging analyses will be required to characterize the kinetics of ER aggregate formation in DENV-infected cells throughout the course of infection.
Comorbidities such as cardiovascular disease, stroke, diabetes, respiratory disease, and renal disease are significant risk factors for severe dengue and increased mortality. During the 2015 dengue outbreak in Taiwan, diabetes mellitus and chronic kidney disease were notably critical underlying conditions in dengue hemorrhagic cases. The TW2015 virus-infected cells exhibited mitochondrial dysfunction, relying on anaero bic glycolysis for energy production, which may exacerbate glucose dysregulation in dengue patients with diabetes. Furthermore, excessive lactic acid accumulation might be inadequately cleared in patients with renal dysfunction. These findings provide insight into how these comorbidities can aggravate dengue infection symptoms. In our study, the STZ model was used to induce β-cell damage, thereby modeling type 1 diabetes rather than type 2 diabetes, which is a more common comorbidity in dengue patients. To address this limitation, future studies should validate the effects of pathogenic DENV infection in type 2 diabetes models, such as genetic models or high-fat diet-induced insulin resistance models.
Our findings reveal that excessive lactate accumulation contributes to the pathogen esis of virulent DENV infection. However, the mechanisms by which lactate influences DENV pathogenesis remain unclear. Beyond serving as a glycolysis byproduct, lactate plays a regulatory role in various biological processes, including antiviral responses. For example, lactate can bind to MAVS, suppressing intracellular viral RNA sensing and type I interferon production (35,36). Elevated lactate levels are also known to create an immunosuppressive environment by modulating immune cell function, as seen during tumor progression (37,38). Further investigation is needed to determine whether lactate elevation caused by pathogenic DENV infection suppresses host immunity and impairs viral clearance.
A difference in E protein size between the TW2015 and 16681 strains was observed by immunoblotting (Fig. 6), suggesting that variations in E protein amino acid compo sition or post-translational modifications-such as glycosylation-may contribute to this discrepancy. To investigate this possibility, we conducted amino acid sequence analysis, which confirmed that the major N-glycosylation sites are conserved between the TW2015 and 16681 strains. Furthermore, PNGase F digestion did not suggest any significant differences in N-glycosylation between the strains. Whether other post-trans lational modifications contribute to the differences in E protein size or influence the pathogenic properties of these strains remains to be determined and warrants further investigation.
## MATERIALS AND METHODS
## Viruses and cells
NGC and 16681 viruses were obtained from Dr. Andrew Yueh (NHRI, Taiwan). The TW2015 virus was obtained from the Centers for Disease Control. Recombinant 16681 viruses containing various TW2015 fragments (E, prM-E, or C-prM-E) were generated and described previously (4). All virus strains were amplified in Vero 76 cells and titrated by colorimetric focus-forming assay (4). Vero cells were cultured with 1× Dulbecco's modified Eagle medium (DMEM) supplemented with 2% FBS and 1% antibiotic-antimy cotic at 37°C. Huh7 cells were cultured with DMEM supplemented with 10% FBS and 1% antibiotic-antimycotic.
## Mice
Stat1
## Streptozotocin-induced diabetic mice
Diabetic mellitus in mice was induced using a single high dose of streptozotocin, following a previously described protocol (12). Briefly, female Stat1 -/-mice were fasted for 4 h prior to receiving an intraperitoneal injection of streptozotocin (200 mg/kg in sodium citrate buffer, pH 4.5). After injection, the mice were given regular food and 10% sucrose water for 12 h, followed by regular water the next day.
## Antibodies, chemicals, and reagents
The MitoProbe Jc-1 Assay Kit (Invitrogen, USA) was used to detect mitochondrial depolarization, indicated by an increase in green fluorescence (Jc-1 monomer), following the manufacturer's instructions. Mag-Fluo-4 AM (Invitrogen) and Rhod-2 AM (Invitrogen) were used to stain calcium in the ER and mitochondria, respectively. Antibodies for AIF (ab32516, Abcam, UK), PDI (#3501, Cell Signaling Technology, USA), and DENV E (YH0026, Yao-Hong Biotechnology, Taiwan) were used for immunostaining, and DAPI (H-1200, VECTASHIELD, USA) was used to label nuclear DNA. Cell images were captured using an Olympus IX73 microscope or a Leica TCS SP5II confocal microscope at the NHRI Core Instrument Center. Antibodies for p-eIF2a (CST3398, Cell Signaling Technology), LC3B (GTX127375, GeneTex, Taiwan), GRP78 (CST3177, Cell Signaling Technology), HSP70 (GTX111088, GeneTex), Erp44 (CST3798, Cell Signaling Technology), Actin (GTX109639, GeneTex), DENV E (GTX127277, GeneTex), p-PKR (sc-101783, Santa Cruz, USA), and PKR (sc-707, Santa Cruz) were used for immunoblotting. Salubrinal (S2923, Selleckchem) was used to inhibit eIF2α phosphatase. Oxamate (O2751, Sigma) was used to block lactate dehydrogenase activity.
## Transmission electron microscope (TEM)
DENV-infected cells were fixed with 2.5% glutaraldehyde for 30 min and processed at the Electron Microscope Division, Cell Biology Core Lab, Institute of Plant and Microbial Biology, Academia Sinica. Cells were examined using a Transmission Electron Microscope (FEI Tecnai G2 Spirit, 2014). For quantification, approximately 50 images per sample grid were captured using a Borries Optimus 100 Scanning TEM (STEM) at an accelerating voltage of 30 kV. Each image had a field of view of approximately 80 µm × 80 µm. Images were processed using Fiji software with histogram equalization and gamma adjustment to enhance image quality. A total of 500-600 cells were analyzed, and the percentages of cells containing small vacuoles, large vacuoles, or exhibiting ER aggregation were recorded.
## Quantitative real-time (RT)-PCR
The detailed information for RNA extraction, cDNA synthesis, and quantitative real-time PCR has been described previously (4). The primer sequences used in the study were: β-Actin(Actb): Forward gctacagcttcaccaccaca, Reverse aaggaaggctggaaaagagc; Glucose transporter protein type 1(Glut1): Forward cttgcttgtagagtgacgatc, Reverse cagtgatcc gagcactgctc; Hexokinase 2 (Hk2) : Forward attcctgcctgatactgcc, Reverse acacacacaca cacacatac; Phosphoglycerate kinase 1 (Pgk1): Forward gctgttccaagcatcaaattc, Reverse tcttttcccttcccttcttcc; Lactate dehydrogenase A (Ldha): Forward ggcactgacgcagacaag, Reverse tgatcacctcgtaggcactg; Relative mRNA expression levels were calculated using the ΔΔCq method with Actb cDNA as an internal control.
## Statistical analysis
Unless otherwise indicated, data are presented as mean ± SD. Statistical differences between treatment groups were analyzed using the Mann-Whitney U test, one-way ANOVA, two-way ANOVA, unpaired t-test, or the log-rank (Mantel-Cox) test, as appropri ate for each experimental design. *P < 0.05, **P < 0.01.
## References
1. Yang, Chang, Hsu et al. (2018) "Molecular characterization and phylogenetic analysis of dengue viruses imported into Taiwan during 2011-2016" *PLoS Negl Trop Dis*
2. Wang, Chang, Loh et al. (2016) "Consecutive large dengue outbreaks in Taiwan in 2014-2015" *Emerg Microbes Infect*
3. Chang, Huang, Lee et al. (2017) "Differences in mortality and clinical manifestations of dengue hemorrhagic fever in Taiwan in different years: a comparison for cases in 2014 and 2015 epidemics" *Am J Trop Med Hyg*
4. Lin, Chung, Dai et al. (2021) "Aggressive organ penetration and high vector transmissibility of epidemic dengue virus-2 Cosmopolitan genotype in a transmission mouse model" *PLoS Pathog*
5. Lahiri, Fisher, Tambyah (2008) "Dengue mortality: reassessing the risks in transition countries" *Trans R Soc Trop Med Hyg*
6. Lye, Lee, Sun et al. (2010) "The benign nature of acute dengue infection in hospitalized older adults in Singapore" *Int J Infect Dis*
7. Pang, Thein, Leo et al. (2014) "Early clinical and laboratory risk factors of intensive care unit requirement during 2004-2008 dengue epidemics in Singapore: a matched case-control study" *BMC Infect Dis*
8. Lee, Hwang, Chen et al. (2006) "Clinical characteristics of dengue and dengue hemorrhagic fever in a medical center of southern Taiwan during the 2002 epidemic" *J Microbiol Immunol Infect*
9. Toledo, George, Martinez et al. (2016) "Relevance of non-communicable comorbidities for the development of the severe forms of dengue: a systematic literature review" *PLoS Negl Trop Dis*
10. Vander Heiden, Cantley, Thompson (2009) "Understanding the Warburg effect: the metabolic requirements of cell proliferation" *Science*
11. Shen, Chen, Tsai et al. (2022) "Hyperglycemia exacerbates dengue virus infection by facilitating poly(A)-binding protein-mediated viral translation" *JCI Insight*
12. Furman (2015) "Streptozotocin-induced diabetic models in mice and rats" *Curr Protoc Pharmacol*
13. El-Bacha, Midlej, Da Silva et al. (2007) "Mitochondrial and bioenergetic dysfunction in human hepatic cells infected with dengue 2 virus" *Biochim Biophys Acta*
14. Hottz, Oliveira, Nunes et al. (2013) "Dengue induces platelet activation, mitochondrial dysfunction and cell death through mechanisms that involve DC-SIGN and caspases" *J Thromb Haemost*
15. Barbier, Lang, Valois et al. (2017) "Dengue virus induces mitochondrial elongation through impairment of Drp1triggered mitochondrial fission" *Virology (Auckl)*
16. Rossi, Pizzo, Filadi (2019) "Calcium, mitochondria and cell metabolism: a functional triangle in bioenergetics" *Biochim Biophys Acta Mol Cell Res*
17. Freppel, Torres, Uyar et al. (2025) "Dengue virus and Zika virus alter endoplasmic reticulum-mitochondria contact sites to regulate respiration and apoptosis"
18. (1016)
19. Rossi, Taylor (2020) "Reliable measurement of free Ca 2+ concentrations in the ER lumen using Mag-Fluo-4" *Cell Calcium*
20. Trollinger, Cascio, Lemasters (1997) "elective loading of Rhod 2 into mitochondria shows mitochondrial Ca 2+ transients during the contractile cycle in adult rabbit cardiac myocytes" *Biochem Biophys Res Commun*
21. Welsch, Miller, Romero-Brey et al. (2009) "Composition and three-dimensional architecture of the dengue virus replication and assembly sites" *Cell Host Microbe*
22. Egger, Wölk, Gosert et al. (2002) "Expression of hepatitis C virus proteins induces distinct membrane alterations including a candidate viral replication complex" *J Virol*
23. Miller, Kastner, Krijnse-Locker et al. (2007) "The non-structural protein 4A of dengue virus is an integral membrane protein inducing membrane alterations in a 2K-regulated manner" *J Biol Chem*
24. Jheng, Ho, Horng (2014) "ER stress, autophagy, and RNA viruses" *Front Microbiol*
25. Yu, Hsu, Liao et al. (2006) "Flavivirus infection activates the XBP1 pathway of the unfolded protein response to cope with endoplas mic reticulum stress" *J Virol*
26. Umareddy, Pluquet, Wang et al. (2007) "Dengue virus serotype infection specifies the activation of the unfolded protein response" *Virol J*
27. Boyce, Bryant, Jousse et al. (2005) "A selective inhibitor of eIF2α dephosphorylation protects cells from ER stress" *Science*
28. Zhang, Kaufman (2008) "From endoplasmic-reticulum stress to the inflammatory response" *Nature*
29. Upagupta, Carlisle, Dickhout (2017) "Analysis of the potency of various low molecular weight chemical chaperones to prevent protein aggregation" *Biochem Biophys Res Commun*
30. Costa-Mattioli, Walter (2020) "The integrated stress response: from mechanism to disease" *Science*
31. (2025) *Full-Length Text Journal of Virology*
32. Yang, Fan, Wang et al. (2022) "Lactate induces vascular permeability via disruption of VEcadherin in endothelial cells during sepsis" *Sci Adv*
33. Zulkipli, Dahlui, Jamil et al. (2018) "The association between obesity and dengue severity among pediatric patients: a systematic review and meta-analysis" *PLoS Negl Trop Dis*
34. Denton (2009) "Regulation of mitochondrial dehydrogenases by calcium ions" *Biochim Biophys Acta*
35. Nichols, Denton (1995) "Towards the molecular basis for the regulation of mitochondrial dehydrogenases by calcium ions" *Mol Cell Biochem*
36. Atakpa-Adaji, Ivanova (2023) "IP 3 R at ER-mitochondrial contact sites: beyond the IP 3 R-GRP75-VDAC1 Ca 2+ funnel" *Contact (Thousand Oaks)*
37. Zhang, Wang, Xu et al. (2019) "Lactate is a natural suppressor of RLR signaling by targeting MAVS" *Cell*
38. Thyrsted, Storgaard, Blay-Cadanet et al. (2021) "Influenza A induces lactate formation to inhibit type I IFN in primary human airway epithelium"
39. Chen, Huang, Gu et al. (2022) "Lactatelactylation hands between metabolic reprogramming and immunosup pression" *Int J Mol Sci*
40. Comito, Iscaro, Bacci et al. (2019) "Lactate modulates CD4 + T-cell polarization and induces an immunosuppressive environment, which sustains prostate carcinoma progression via TLR8/ miR21 axis" *Oncogene*
41. Durbin, Hackenmiller, Simon et al. (1996) "Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease" *Cell*
42. Chen, Hofman, Kung et al. (2007) "Both virus and tumor necrosis factor alpha are critical for endothelium damage in a mouse model of dengue virus-induced hemorrhage" *J Virol*
43. Orozco, Schmid, Parameswaran et al. (2012) "Characterization of a model of lethal dengue virus 2 infection in C57BL/6 mice deficient in the alpha/beta interferon receptor" *J Gen Virol*
44. Sarathy, White, Li et al. (2015) "A lethal murine infection model for dengue virus 3 in AG129 mice deficient in type I and II interferon receptors leads to systemic disease" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645916&blobtype=pdf | # The La Crosse virus Gc head domain is a major determinant of infection and pathogenesis
Ariana Dedvukaj, Nicole Rondeau, Tamara Vázquez, Alejandro Cristófalo, Molly Durawa, Matthew Lutchko, Kenneth Stapleford
## Abstract
How orthobunyaviruses establish infections and disseminate to cause disease is not well understood. In a previous study using the in vivo evolution of La Crosse virus (LACV), we discovered a cluster of mutations localizing to the LACV Gc head domain. However, we do not understand how the Gc head domain contributes to infection. Here, we generated each of the aforementioned mutations and addressed the role of the Gc head domain in viral replication and infectivity in mammalian and insect cells. We found that specific head domain residues could attenuate replication and infectivity in human neurons and reduce cell binding across different hosts, indicating an important role for the head domain during infection in vitro. Focusing on the in vitroattenuated Gc N609D variant, we infected 3-week-old wild-type (WT) C57BL/6J mice via the footpad with WT LACV or the Gc N609D variant and found that the Gc N609D virus was completely attenuated. To address whether the variant was also attenuated in a highly susceptible mouse model, we infected Ifnar1 -/-mice with WT LACV and Gc N609D. We found that virulence in mice infected with Gc N609D was delayed, with several mice surviving the infection, and while viral titers were comparable between viruses in the footpad, there was a significant reduction in virus in the brain. Together, these studies define a critical role of the Gc head domain in infectivity and pathogenesis. Studies are underway to further define how the orthobunyavirus Gc head domain contributes to infection and disease. IMPORTANCE Orthobunyaviruses are emerging arboviruses capable of severe disease and explosive outbreaks. However, our understanding of how orthobunyaviruses establish infections or cause disease is not completely understood. The orthobunyavirus Gc glycoprotein contains a variable amino-terminal head domain that forms the tip of the virion trimeric spike, yet it is unclear how the head domain contributes to infection or pathogenesis. In this study, we use LACV and a panel of Gc head domain variants to address the role of the head domain in LACV biology. We found that critical head domain regions are important for virus infectivity and pathogenesis in mice, highlighting an important role for the Gc head domain in orthobunyavirus infection and disease. KEYWORDS orthobunyavirus, La Crosse virus, head domain, virus entry T he orthobunyavirus genus (Peribunyaviridae) includes a long list of significant human pathogens (1, 2). These arthropod-borne viruses (arboviruses) are transmitted to humans primarily by mosquitoes, midges, and ticks. Recent outbreaks of Oropouche virus (OROV) in Cuba (3) and South America (4-7), along with the prevalence of La Crosse virus (LACV) in the United States, emphasize the clinical relevance of orthobunyaviruses. LACV is mainly found in the East North Central and Appalachian regions of the United States (8). As a member of the California serogroup of orthobunyaviruses, LACV is related to other neuroinvasive human viruses found worldwide, including Jamestown Canyon
virus, Inkoo virus, and Tahyna virus (9). Although the majority of cases are asympto matic and therefore underreported, LACV is the leading cause of pediatric arbo viral encephalitis in the United States. Neuroinvasive disease can result in fatality or lifelong neurological sequelae, such as recurring seizures and cognitive deficits (10)(11)(12). However, there are currently no antiviral therapies or vaccines targeting orthobunyavi ruses, highlighting the need to study orthobunyavirus biology in molecular detail.
Our lack of antiviral therapies is in part due to our incomplete understanding of the molecular mechanisms orthobunyaviruses use to establish infections. The orthobu nyavirus negative-sense RNA genome consists of S, M, and L segments. The S segment encodes the nucleoprotein and interferon antagonist non-structural protein NSs (13,14). The M segment encodes the Gn and Gc glycoproteins, along with non-structural protein NSm (1,15), and the L segment encodes the RNA-dependent RNA polymerase. The M segment proteins are important for virion assembly (15,16), entry and attachment (17)(18)(19), and cell-to-cell spread ( 20), yet we know little of how discrete domains within these proteins contribute to virus infection. Specifically, the orthobunyavirus Gc protein is a class II fusion glycoprotein similar to those of other bunyaviruses as well as alphaand flaviviruses (21)(22)(23)(24). Orthobunyavirus Gc contains a unique amino-terminal variable head domain that forms the tip of the glycoprotein spike (25,26). Previous work with Bunyamwera virus and OROV has shown that the head domain can be deleted with little impact to virus replication in vitro and that fusion and immunogenicity are enhanced, suggesting that the head domain plays an important yet unknown role in virus biology (27,28). However, even considering observations, we do not understand specifically how the Gc head domain contributes to infection or pathogenesis.
In a previous study, we hypothesized that we could use in vivo virus evolution to identify critical determinants of LACV infection as we have done previously for CHIKV (29,30) and ZIKV (31,32). We infected Aedes (Ae.) mosquitoes and suckling mice with wild-type (WT) LACV and Sanger sequenced virus populations at 7 and 3 days post-infec tion for mosquitoes and mice, respectively (17). From these studies, we identified seven mutations in the LACV Gc head domain, suggesting that the head domain may be an important determinant for LACV biology (17). Here, we generated this panel of LACV Gc head domain variants and addressed the role of these residues and the Gc head domain in vitro and in vivo. We found that specific Gc head domain residues could influence replication, infectivity, and binding in vitro. In addition, we found that the most attenuated Gc variant N609D could completely attenuate virulence of a pathogenic WT LACV strain in wild-type mice. Moreover, we found that Gc residue N609 was critical for infection and virulence in Ifnar1 -/-mice. Finally, using an evolutionary approach, we found that the LACV Gc head domain has been evolving across LACV lineages and that there are conserved elements maintained within the head domain of related orthobunyaviruses. Together, these studies provide a critical role for the orthobunyavirus Gc head domain in virus binding, infectivity, and virulence and allow us to speculate on how changes in the orthobunyavirus head domain may impact virus biology.
## RESULTS
## LACV Gc head domain variants localize to the tip of the head domain with the potential to alter folding and trimer stability
The orthobunyavirus Gc glycoprotein amino-terminal variable head domain sits atop the glycoprotein spike (Fig. 1A andB, red dashed box) (24)(25)(26). However, it is unclear how the head domain contributes to LACV infection in vitro or in vivo. In a previous study, we used in vivo evolution of LACV in Ae. mosquitoes and C57BL/6J mice to identify potential residues that may be important for virus infection (17). We identified seven variants at six head domain residues, which we hypothesized were important for LACV infection (Table 1; Fig. 1).
To understand how these residues contribute to LACV biology, we first looked at where each residue is located on the Gc head domain trimer (Fig. 1A through C) (17,26). We observed that residue N609 is positioned at the point at which all three monomers converge to form the trimer, while residues S615, W618, D619, A621, and E623 are positioned above N609 on extensions facing away from the particle (Fig. 1A through C). The location of these residues, particularly N609, allowed us to hypothesize that changes at these amino acids may impact trimer folding. To test this hypothesis in silico, we calculated the ΔΔ fold G and ΔΔ bind G for each variant to understand how individual variants may influence trimer folding and monomer binding, respectively (Table 1). We observed that the variant N609D is predicted to negatively impact trimer folding and inter-protomer binding, while other variants such as D619G and A621V are predicted to have impacts on trimer folding but not inter-protomer interactions. Finally, we addressed whether mutations at the residues in Table 1 have been observed in nature. We aligned the M segment protein regions deposited in the National Center for Biotechnology and Information (NCBI) Virus and found several of the variants we identified (W618R, D619G, and E623G) have been found in nature (Fig. 1C). Taken together, these studies highlight the potential role of the Gc head domain in LACV infection.
## LACV Gc head domain is important for virus production in human cells
Given the location of each residue and the potential influence of each variant on folding, we hypothesized that the head domain is critical for virus infection. To test this hypothesis, we generated each head domain variant in the LACV lineage I Mosquito/NC/ 1978 infectious clone M segment (33). All variants were genetically stable after passaging virus once in Vero cells to generate working stocks. To begin, we looked at plaque size of each variant on Vero cells (Fig. 2). We observed that the variants Gc N609D and W618L led to small plaques, while variants W618R, D619G, A621V, and E623A led to larger plaques compared to WT LACV (Fig. 2).
To study the role of each residue in LACV replication in vitro, we first performed multistep growth curves by infecting Vero cells with each virus at a multiplicity of infection (MOI) of 0.1 and quantifying infectious virus in the supernatant at multiple time points post infection (Fig. 3A). Vero cells were used because of their interferon signaling deficiency, allowing us to focus on the role of each Gc residue without confounding influence from the antiviral response. In Vero cells, the replication of the Gc variants separated into two groups. We found that several of the variants, including Gc N609D, S615I, and W618L, led to attenuated growth, while D619G, A621V, and E623A led to enhanced growth over WT LACV (Fig. 3A). Since several of the variants were selected for in Ae. aegypti mosquitoes, we also addressed viral growth in Aag2 Ae. aegypti cells by infecting Aag2 cells with each virus at an MOI of 0.1 and quantifying infectious virus in the supernatant by plaque assay (Fig. 3B). We observed that while several of the variants were attenuated early during infection, all of the variants caught up to WT LACV by 24 h. When we sequenced each variant at 24 h from the Aag2 cells, we found that all of the variants had reverted to the wild-type residue, suggesting that these variants are restricted early during infection, but a strong selective pressure forces them to revert to the WT residue. Together, given significant changes in Vero cells and reversion in Aag2 cells, these studies show that the Gc head domain plays an important role in replication in vitro.
We next tested whether the LACV Gc head domain residues were critical for replica tion in human neurons and myoblasts, relevant cell lines for LACV infection (34,35) (Fig. 3C andD). We infected human neuroblastoma SH-SY5Y cells and human myoblasts with WT LACV or each Gc head domain variant at an MOI of 0.1 and quantified infectious virus production at 24 h post-infection. In SH-SY5Y cells, we found that the variants Gc N609D, S615I, and W618L were attenuated in virus growth up to 100-fold compared to WT LACV, similar to the growth in Vero cells (Fig. 3A). In human myoblasts, we observed that the attenuated Gc N609D, S615I, and W618L variants in neurons and Vero cells were not as severely impacted. On the other hand, we found that the Gc variants W618R, D619G, A621V, and E623A were able to enhance infection, similar to what we saw in Vero cells. These results demonstrate that the Gc head domain can influence virus production in multiple cell types, including human cells, and that there may be cell type-specific mechanisms regulating infection.
## The LACV Gc head domain residues contribute to cell-specific infectivity and binding
Given the location of each residue on the Gc spike (Fig. 1) and our plaque size pheno types (Fig. 2), we hypothesized these residues may be important for virus infectivity. To test this hypothesis, we performed infectivity assays by incubating WT LACV and each variant with Vero cells, Aag2 cells, human neurons, or myoblasts at an MOI of 1 for 1 h to allow entry. We then added media containing 20 mM ammonium chloride (NH 4 Cl) to stop further virus spread, allowing us to address only the initial infection. In Vero cells (Fig. 4A) and Aag2 cells (Fig. 4B), we observed that many of the variants had little impact on virus infectivity. However, the LACV variant E623A led to enhanced infection in both Vero and Aag2 cells, suggesting a role in entry. On the other hand, in both human neurons (Fig. 4C) and myoblasts (Fig. 4D), we found that the Gc variant N609D and S615I had reduced virus infection, indicating that these variants may have defects in virus entry. Interestingly, the Gc variant A621V was also attenuated in the neurons (Fig. 4C) yet showed the same levels of infectious particle production in growth assays as WT LACV (Fig. 3A). These results may suggest that the Gc A621V variant could have defective cell entry, rescued by advantages in virus genome replication, assembly, or egress.
To begin to understand how the Gc head domain functions during entry, we per formed cell binding assays on Vero, Aag2, and human myoblasts (Fig. 5). We chose to omit the human neurons as they did not adhere to culture plates well enough to withstand vigorous washing. We incubated each cell line with 20 mM NH 4 Cl for 1 h to block endocytosis and then added sucrose-purified viruses (MOI of 100 based on RNA molecules) for 30 min on ice. We then washed off unbound virus and quantified the amount of bound virus S segment RNA by qPCR. We found that the Gc residues S615, W618, and E623 could alter binding in all three cell lines, while residue D619 behaved similarly to WT virus. Interestingly, residue N609 did not play a major role in binding in Vero or Aag2 cells (Fig. 5A andB) but did influence binding in myoblasts (Fig. 5C), suggesting cell type-specific roles in virus infection as we saw for infectivity. Taken together, we conclude that the LACV Gc head domain is critical for virus entry and cell binding.
## The Gc head domain variant N609D attenuates LACV virulence and replica tion in mice
The Gc head domain variant N609D displayed the strongest phenotypes in vitro with reduced replication and infectivity in neurons, the major site of infection in humans and mice. These results allowed us to hypothesize that the Gc N609D variant may significantly reduce virulence. To test this hypothesis, we infected 3-week-old wild-type C57BL/6J mice with 20,000 PFU of WT LACV or the Gc N609D via the left footpad and weighed the mice daily for 2 weeks (Fig. 6A andB). We found that while all of the mice infected with WT LACV succumbed to infection by 8 days post-infection, the mice infected with Gc N609D all survived the infection (Fig. 6A). One potential explanation for this phenotype is that the mice injected with the N609D variant did not become infected. To rule this possibility out, we addressed the presence of neutralizing antibodies in each mouse and found that the mice infected with Gc N609D generated antibodies that neutralized both WT LACV and the Gc N609D variant (Fig. 6C andD), indicating that they were infected.
Given the complete attenuation of the LACV Gc N609D variant in WT mice, we wondered whether this variant would also be attenuated in a more susceptible mouse model. We infected Ifnar1 -/-mice, which lack the type I interferon alpha receptor, with 50,000 PFU of WT LACV or the Gc N609D variant via the footpad (Fig. 7). We observed that all mice infected with WT LACV succumbed to infection at 5 days post-infection. However, we found that for mice infected with Gc N609D, survival was extended several days, with two out of seven mice surviving the infection (Fig. 7A andB). The mice surviving the infection lost weight yet recovered around day 7 and generated neutraliz ing antibodies to WT LACV and the Gc N609D variant (Fig. 7C andD).
Finally, we hypothesized that the attenuation of the Gc N609D variant in Ifnar1 -/-mice was due to reduced infection in the brain. To test this hypothesis, we infected Ifnar1 -/- mice with 20,000 PFU of WT LACV and the Gc N609D virus and measured viral titers in the footpad (site of infection) and brain at 3 days post-infection. In the footpad (Fig. 7E), both viruses replicated to similar levels, indicating that the Gc N609D variant is capable of infecting cells and replicating at the site of infection. However, when we looked at viral titers in the brain at 3 days post-infection, we found that while WT LACV replicated to high titers, many of the mice infected with the Gc N609D variant had significantly reduced viral titers (Fig. 7F) with several mice having no detectable infectious virus in the brain. These results indicate that the Gc head domain, specifically residue N609, is important for virus infection and pathogenesis.
## The Gc head domain conservation across LACV lineages and other orthobu nyaviruses
Our in vivo evolution studies identified the Gc head domain as a potential hotspot for adaptation (17). Therefore, we asked whether the LACV Gc head domain has been changing over time and between LACV lineages. We aligned the LACV Gc head domain regions (amino acids 477 to 722) of all complete LACV M segment sequences (31 in total) deposited in the NCBI Virus database (Fig. S1). For LACV lineage I, we found that the Gc head domain remained mostly constant with only two mutations present in over 60% of deposited sequences (V528I and K548R; Table 2; Fig. 8A). These two residues are found at the base of the head domain away from the mutations we had found in our in vivo study. In LACV lineage II, there were six mutations with five localized to the tip of the head domain (Table 2; Fig. 8A). Finally, in LACV lineage III, which is the most divergent lineage (36), there were 19 amino acid changes that largely localized along the side of the head domain and at the base (Fig. 8A).
Looking closer at the differing residues between lineages, many of the changes in the Gc head domain corresponded to changes in charge, leading us to hypothesize that these may alter the overall charge of the Gc head domain. To address this hypothesis, we first looked at the overall charged distribution of the LACV Gc head domain in the trimer (Fig. 8B). On one side of the Gc head domain, there is a largely negatively charged region running along the side of the head domain with a negatively charged spot on top and a positively charged region on the side (Fig. 8B). When we introduced changes for lineages I, II, and III into the LACV head domain structure, we found that these variants changed the overall charge of the Gc head domain both at the tip, the side, and the base of the head domain (Fig. 8B,boxes). These results suggest that alterations in the Gc head domain can influence charged patches on the protein surface that are critical for protein-protein interactions within the virus or host-pathogen interactions needed for infection. Finally, we asked whether the Gc head domain residues we studied in LACV were conserved across other orthobunyaviruses (Fig. 9). Looking at the head domain of the related orthobunyaviruses OROV and BUNV, we found that there are structurally similar alpha helices making up the top portion of the head domains (Fig. 9A, colored helices). Moreover, there are similar amino acid residues present in the top of the orthobunyavirus head domains, including a conserved glutamine, which we showed in LACV is critical for pathogenesis, as well as a conserved aromatic tryptophan or tyrosine. Interestingly, both LACV and OROV contain an identical glutamic acid at position 623, which we have shown is capable of enhancing LACV infection in vitro. While there are similarities between orthobunyavirus Gc head domains at the amino acid level, we found major differences in the charge distribution of the Gc head domain between viruses (Fig. 9B). These results suggest that the location of charged residues could impact critical host-pathogen interactions necessary for individual virus biology.
## DISCUSSION
Orthobunyaviruses are significant human pathogens capable of devastating outbreaks. These viruses encode a class II fusion glycoprotein Gc that is critical for virion assembly and entry. The Gc glycoprotein is functionally and structurally similar to those of alphaand flaviviruses in domain II, which includes the fusion loop. However, orthobunyaviruses are unique in that they encode a variable amino-terminal head domain that forms the tip of the virion trimeric spike. Our understanding of how the head domain functions is not well defined.
In this study, we took advantage of in vivo evolution of LACV in mice and mosquitoes, identifying several mutations clustering in the Gc head domain. Given the location of these changes and that several of them are found in nature, we hypothesized these are important for LACV and orthobunyavirus biology. We generated each variant and tested F) Three-week-old male and female Ifnar1 -/-mice were infected with 20,000 PFU of each virus or PBS via the footpad. At 3 days post-infection, mice were euthanized, and the footpad (E) and brain (F) were collected and homogenized, and the infectious virus was quantified by plaque assay. Data represent the mean and SD of two independent experiments. N = 12 mice for each virus. Mann-Whitney test. how these mutations influenced virus growth in multiple cell lines. We found that while these variants were largely genetically stable in mammalian cells, they reverted to the wild-type residue in mosquito cells, suggesting a strong selective pressure in these cells. These results are interesting as we identified several of these variants in mosquitoes. One explanation for these results could be that our previous in vivo evolution studies used the LACV lineage I strain Human/MN/1960, which differs from our infectious clone lineage I system Mosquito/NC/1978 by 29 amino acids in the M segment. Regardless, we find that specific head domain residues could influence plaque size, virus production, and infectivity in human neurons and myoblasts, indicating an important role for the head domain in virus production and spread. Interestingly, these specific phenotypes may be cell type specific through our findings of several of the variants being attenuated in neurons but not in myoblasts. Moreover, although we find that the variants W618R, D619G, A621V, and E623A can increase virus production in myoblasts, this does not seem to be due to changes in infectivity. These results may suggest that these variants are important for different steps in the viral life cycle or that these variants enhance other functions to compensate for defects. This point is particularly true when looking at cell binding. We find that many of the variants tested lead to reduced cell binding on multiple cell types, yet these results do not directly translate to similar defects in infectivity. Future studies to explore the detailed mechanisms of how the LACV Gc head domain contributes to fusion and entry will be critical to dissect this complex process. Taken together, our results show that even single amino acid changes can have major impacts on the virus life cycle, highlighting the importance of the Gc head domain in orthobunyavirus biology.
## TABLE 2 LACV Gc head domain mutations across lineages
## Lineage
In addition to roles in vitro, we found that the head domain and specifically residue N609 were important for virulence in mice. In WT mice, the Gc head domain variant N609D was completely attenuated, yet the mice started to lose weight around 7 days post-infection and retained this weight for the remainder of the experiment. These results, along with the presence of neutralizing antibodies, show that the mice are infected yet do not succumb to infection. Moreover, we find that in the highly suscepti ble type I interferon-deficient mice, the Gc N609D variant is attenuated, although mice do succumb to infection. In the two Ifnar1 -/-mice that survived the infection, both lost weight and recovered and produced neutralizing antibodies, which we hypothesize are critical for viral clearance. Looking at virus replication, we find that while the Gc N609D variant can replicate to WT levels at the site of infection, there was a reduction in viral titers in the brain of the same animals at 3 days post-infection. One explanation for these results could be that the Gc head domain is important for neuroinvasion and infection of the brain, as we do see reduced replication and infection in human neurons. Another possibility could be that the Gc head domain is important for virus dissemination from the site of infection to the brain. Similar defects in dissemination have been seen for LACV fusion loop mutants (37); however, these variants also showed replication defects in the muscle, suggesting a global defect in virus entry. Future work investigating multiple time points and organs will be important to understand how the Gc head domain influences infection in vivo. Our in vivo and in vitro results suggest that the Gc head domain is critical for cell-specific interactions and entry mechanisms that may drive virus dissemination. Together, these results show that the Gc head domain plays an essential role in LACV virulence and suggest that if the virus can be restricted to the periphery for long enough that antibodies can be produced, the host can recover. Finally, we see that the LACV Gc head domain, while variable between lineages, is largely conserved at key residues at the tip of the spike. Given their location, we hypothesize that these residues may be critical for trimer formation and stability, interactions with an unknown LACV receptor or other host factors on the cell surface to facilitate entry, and/or the structure and function of the Gn glycoprotein. In the case of the Gc N609D variant, we speculate that this variant may influence trimer formation or disassociation during entry, leading to reduced infection or infection via an alternative pathway that attenuates virus dissemination and overall pathogenesis. An additional hypothesis may be that these Gc head domain residues are important for proper spike assembly, meaning that changes in glycoprotein structure can change host-pathogen interactions needed for virus entry. Specifically, the presence of distinct positively charged patches may suggest potential interactions with negatively charged glycosaminoglycans, as is the case for other arboviruses (38)(39)(40). Changes in the charge network of the Gc head domain could explain why the different lineages of LACV differ in pathogenesis (41). One explanation may be due to the changes in the Gn-Gc spike, altering the function of the spike during entry. Previous work has shown that LACV G1 and G2 can individually influence cell-specific binding (42). Therefore, it may be that changes in Gc have downstream effects on Gn for multiple step binding. Future work will be important to investigate the role of the Gc head domain in the pathogenesis of LACV and other orthobunyaviruses. We find that the OROV Gc head domain, for example, is structurally similar to that of LACV, with similar residues maintained at the Gc tip. It will be important to interrogate how the head domain of orthobunyaviruses facilitates virus dissemination and disease to better understand how these pathogens establish infections.
## MATERIALS AND METHODS
## Cells
Vero cells (CCL-81, American Type Culture Collection [ATCC]) were grown in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% newborn calf serum (NBCS). BHK-21 BSR/T7 cells (43), a gift from Dr. Steven Whitehead at the National Institutes of Health (NIH), were grown in DMEM supplemented with 10% fetal bovine serum (FBS), 1% non-essential amino acids (NEAAs), 10 mM HEPES, and 1 mg/mL geneticin added every other passage to maintain T7 selection. Human neuroblastoma cells (SH-SY5Y and CRL-2266, ATCC) were grown in a 50:50 mix of Eagle's minimum essential medium (ATCC) and F12 medium supplemented with 10% FBS. Immortalized human myoblasts were a gift from Dr. Michael Kyba at the University of Minnesota (34). Myoblasts were grown in HAM's/F10 Nutrient Mixture supplemented with 20% FBS, 1× Glutamax, 10 ng/mL human basic fibroblast growth factor, 40 ng/mL dexamethasone, and 100 µM beta-mercaptoethanol. All mammalian cells were maintained at 37°C with 5% CO 2 . Aedes aegypti cells (Aag2), a gift from Dr. Paul Turner at Yale University, were maintained in DMEM supplemented with 10% FBS, 1% NEAA, and 10 mM HEPES at 28°C with 5% CO 2 . All cells were confirmed mycoplasma free by monthly testing.
## Viruses
The LACV infectious clone system was obtained from Dr. Whitehead (33). Gc head domain mutants were generated by site-directed mutagenesis of the M segment using the primers in Table 3. All plasmids were Sanger sequenced at Plasmidsaurus to ensure there were no second-site mutations. To generate each virus, BHK-21 BSR/T7 cells were transfected with 2 µg of each of the S, M, and L plasmids using Trans-IT LT1 transfection reagent (Mirus). Twenty-four hours post-transfection, media were replaced, and cells were incubated at 37°C for 5 days. Supernatants were collected, aliquoted, and stored at -80°C. To generate working virus stocks, virus from each transfection was amplified on a monolayer of Vero cells. Viruses were collected, centrifuged at 1,200 rpm for 5 min, aliquoted, and stored at -80°C. Viral titers were quantified by plaque assay as described below.
## RNA extractions and Sanger sequencing
RNA was extracted using Trizol (Thermo Fisher Scientific) and the Direct-Zol plus RNA extraction kit (Zymo Research) following the manufacturer's instructions. cDNA was generated using the Maxima H Minus First-Strand cDNA Synthesis Kit (Thermo Fisher Scientific) and used for Phusion PCR (Thermo Fisher Scientific) with the M segment sequencing primers in Table 3. PCR amplicons were purified with the Macherey-Nagel PCR clean-up kit and Sanger sequenced at Plasmidsaurus. PCR sequences were aligned to the infectious clone reference using SnapGene (version 8.0.3).
## Plaque assay
A total of 350,000 Vero cells were seeded in 12-well plates and incubated with 10-fold dilutions of each virus for 1 h at 37°C. Agarose (0.8%) in DMEM containing 2% NBCS and 1× antibiotic/antimycotic (Gibco) was added, and cells were incubated for 72 h. Following incubation, cells were fixed with 4% formalin for 1 h; agarose plugs were removed; and cells were stained with crystal violet. Viral titers were quantified by counting the number of plaques on the lowest countable dilution. Plaque size was quantified using Image Lab (version 6.1.0, Bio-Rad).
## LACV growth curves
Vero cells (55,000 cells/well), Aag2 cells (200,000 cells/well), myoblasts (55,000 cells/well), and SH-SY5Y cells (200,000 cells/well) were seeded in poly-L-lysine-coated 24-well plates. Cells were incubated with each virus at a MOI of 0.1 for 1 h at 37°C (mammalian cells) or 28°C (insect cells). The virus was removed; cells were washed twice with phosphatebuffered saline (PBS); and complete media were added. Supernatant was collected at the indicated time points, and infectious viral titers were quantified by plaque assay as described above.
## LACV infectivity assays and immunostaining
Myoblasts (10,000 cells/well), SH-SY5Y cells (50,000 cells/well), Vero cells (10,000 cells/ well), and Aag2 cells (150,000 cells/well) were seeded in black 96-well Costar clear bottom plates. Cells were washed once with PBS and incubated with each virus at an MOI of 1 for 1 h at 37°C. After incubation, media containing 20 mM NH 4 Cl were added to block virus spread. Cells were incubated for 24 h and fixed in 4% paraformaldehyde (PFA).
For LACV staining, cells were then washed three times with perm/wash (BD Biosciences), incubated with 0.25% Triton for 10 min, and incubated in blocking buffer (0.2% bovine serum albumin and 0.05% saponin in PBS) for 1 h at room temperature. Cells were then incubated with a 1:2,000 dilution of primary rabbit anti-LACV antibodies (a gift from Dr. Karin Peterson at the NIH) in blocking buffer for 2 h, washed extensively with perm/wash, and incubated with a 1:10,000 dilution of secondary goat antirabbit IgG Alexa488 and 4′,6-diamidino-2-phenylindole (1:1,000 dilution) for 1 h. Cells were washed three times with perm/wash, and PBS was added. The number of infected cells was quantified on the CX7 CellInsight high-content microscope.
## LACV binding assay
For LACV binding assays, clarified viral supernatants were concentrated over a 20% sucrose cushion by ultracentrifugation at 25,000 × g for 4 h. Virus pellets were resuspen ded in DMEM containing 2% NBCS. Infectious virus titers were determined by plaque assay as described above. LACV S genome segments were quantified by extracting RNA with Trizol followed by cDNA synthesis as described above. cDNA was used for RT-qPCR using Power SYBR Green (Applied Biosciences) and the primers in Table 3. An S segment DNA standard was used to generate a standard curve to quantify the number of S genome segments. Virus binding assays were performed by seeding Aag (200,000 cells/well), myoblasts (50,000 cells/well), and Vero cells (50,000 cells/well) into 24-well plates coated with poly-L-lysine. Cells were incubated for 1 h with media containing 20 mM NH 4 Cl and then placed on ice. Each virus was diluted to an MOI of 100 (based on RNA genomes) in DMEM containing 20 mM NH 4 Cl and incubated with cells for 20 min on ice. Cells were washed three times with cold PBS, and 250 µL of Trizol was added to each well. RNA was extracted, and cDNA was generated as described above. Relative RNA concentrations were calculated using Power SYBR Green and the S segment and GAPDH (mammalian cells) or actin (Aag2 cells) primers in Table 3. Relative increases in RNA binding over mock treatment controls were calculated using the 2 -ΔΔCt method against an untreated control. The amount S segment in each input RNA dilution was also quantified and used to correct for starting RNA amounts.
## Mouse infections
All animal work was completed at the NYU Grossman School of Medicine under IACUC protocol IA16-01783. For survival assays, 3-week-old wild-type C57BL/6J (Strain #000664, Jackson Laboratory) and Ifnar1 -/-mice (Strain #028288; Jackson Laboratory) were infected with 20,000 or 50,000 PFU of each virus, respectively, in the left rear footpad while under anesthesia with isoflurane. Mice were weighed daily and euthanized when the weight reached less than 20% of the starting weight or exhibited neurological symptoms. All mice were euthanized after 14 days. For viral titers, mice were infected with 20,000 PFU and euthanized at 3 days post-infection. The footpad and brain were harvested in 750 µL of plaquing media containing steel beads. The tissue was homogenized using a Bullet Blender Storm Pro (Next Advance) for 5 min (brains) or 10 min (footpad) at setting 12 and clarified by centrifugation. Viral titers were quantified by plaque assay as described above.
## LACV neutralization assays
At 14 days post-infection, mice were euthanized and serum was collected. Serum was inactivated at 56°C for 30 min, and twofold dilutions were made in DMEM containing 2% NCBS. Each dilution was mixed with 5,000 PFU of WT LACV or the Gc N609D virus and incubated at 37°C for 1 h. Following incubation, the virus mix was added to Vero cells (15,000 cells/well in a black 96-well Costar clear bottom plate) and incubated for 48 h at 37°C. Cells were then fixed with 4% PFA and stained for LACV antigen, and the number of infected cells was quantified with the CX7 CellInsight High Content microscope as described above.
## Protein structures, folding calculations, and alignments
To calculate ΔΔ fold G and ΔΔ bind G, the mutant's PDB files were generated with FoldX (44) built in YASARA software using the crystallized model of LACV Glycoprotein Gc head domain biological assembly (PDB: 6H3W). First, the WT trimer was prepared in ChimeraX by deleting water molecules and glycans (45). The impact on stability and binding of each mutation was calculated by means of ΔΔ fold G and ΔΔ bind G, employing the FoldX plugin (BuildModel and AnalyzeComplex, respectively). The positive values in ΔΔ fold G imply mutations that are detrimental for proper folding of the trimer, and positive values in ΔΔ bind G imply mutations that negatively impact inter-protomer binding. The ΔΔ bind G values estimate the difference in interaction energy between trimer chains between mutants and wild-type protein. ΔΔ bind G values were calculated according to ΔΔ bind G = ∑ mutant chains Δ bind G -∑ WT chains Δ bind G. Experiments were performed in triplicate.
PyMOL (version 3.1.1) was used for the structures of the LACV head domain trimer (PDB: 6H3W) and head domains of OROV (PDB: 6H3X) and BUNV (PDB: 6H3V). The PyMOL mutagenesis wizard was used to introduce mutations, and the APBS Electrostatics plugin was used for analyzing charge distribution at pH 7.4. SnapGene (version 8.0.3) was used to align all LACV M segment complete protein sequences found in NCBI Virus.
## Statistics
All data were analyzed using GraphPad Prism (version 10.4.2). Data are represented as the mean and SD. In vitro experiments were completed three independent times with internal duplicates where noted. In vivo experiments were completed at least two independent times with the number of animals found in the figure legends. P < 0.05 is considered statistically significant.
## References
1. Elliott (2014) "Orthobunyaviruses: recent genetic and structural insights" *Nat Rev Microbiol*
2. O'connor, Hick, Finlaison et al. (2024) "Revisiting the importance of orthobunyaviruses for animal health: a scoping review of livestock disease, diagnostic tests, and surveillance strategies for the simbu serogroup" *Viruses*
3. Benitez, Alvarez, Perez et al. (2024) *Emerg Infect Dis*
4. De Melo Iani, Pereira, De Oliveira et al. (2025) "Travel-associated international spread of Oropouche virus beyond the Amazon" *J Travel Med*
5. Daudt-Lemos, Ramos-Silva, Faustino et al. (2025) "Rising incidence and spatiotemporal dynamics of emerging and reemerging arboviruses in Brazil" *Viruses*
6. Nl (2024) "Oropouche virus: an emerging orthobunyavirus" *J Gen Virol*
7. Gaillet, Pichard, Restrepo et al. (2021) "Outbreak of Oropouche virus in French Guiana" *Emerg Infect Dis*
8. Day, Odoi, Fryxell (2023) "Geographically persistent clusters of La Crosse virus disease in the Appalachian region of the United States from 2003 to 2021" *PLoS Negl Trop Dis*
9. Evans, Winkler, Peterson (2019) "Differences in neuropathogen esis of encephalitic California serogroup viruses" *Emerg Infect Dis*
10. Edridge, Van Der Hoek (2020) "Emerging orthobunyaviruses associated with CNS disease" *PLoS Negl Trop Dis*
11. Vahey, Lindsey, Staples et al. (2021) "La Crosse virus disease in the United States, 2003-2019" *Am J Trop Med Hyg*
12. Day, Byrd, Fryxell (2023) "La Crosse virus neuroinvasive disease: the kids are not alright" *J Med Entomol*
13. Schoen, Lau, Verbruggen et al. (2020) "Elongin C contributes to RNA polymerase II degradation by the interferon antagonist NSs of La Full-Length Text Journal of Virology November"
14. Orthobunyavirus *J Virol*
15. Blakqori, Delhaye, Habjan et al. (2007) "La Crosse bunyavirus nonstructural protein NSs serves to suppress the type I interferon system of mammalian hosts" *J Virol*
16. Barker, Crump (2023) "Mechanisms of bunyavirus morphogene sis and egress" *J Gen Virol*
17. Shi, Kohl, Léonard et al. (2006) "Requirement of the N-terminal region of orthobunyavirus nonstructural protein NSm for virus assembly and morphogenesis" *J Virol*
18. Thannickal, Spector, Stapleford (2023) "The La Crosse virus class II fusion glycoprotein ij loop contributes to infectivity and replication in vitro and in vivo" *J Virol*
19. Plassmeyer, Soldan, Stachelek et al. (2007) "Mutagenesis of the La Crosse Virus glycoprotein supports a role for Gc (1066-1087) as the fusion peptide"
20. Plassmeyer, Soldan, Stachelek et al. (2005) "California serogroup Gc (G1) glycoprotein is the principal determinant of pH-dependent cell fusion and entry" *Virology (Auckl)*
21. Terhzaz, Kerrigan, Szemiel et al. (2025) "NSm is a critical determinant for bunyavirus transmission between vertebrate and mosquito hosts" *Nat Commun*
22. Guardado-Calvo, Rey (2017) "The envelope proteins of the bunyavirales" *Adv Virus Res*
23. Hellert, Aebischer, Haouz et al. (2023) "Structure, function, and evolution of the Orthobunyavirus membrane fusion glycoprotein" *Cell Rep*
24. Guardado-Calvo, Rey (2021) "The viral class II membrane fusion machinery: divergent evolution from an ancestral heterodimer" *Viruses*
25. Gao, Zhao, Li et al. (2025) "The role of Orthobunyavirus glycoprotein Gc in the viral life cycle: from viral entry to Egress" *Molecules*
26. Bowden, Bitto, Mclees et al. (2013) "Orthobunyavirus ultrastructure and the curious tripodal glycoprotein spike" *PLoS Pathog*
27. Hellert, Aebischer, Wernike et al. (2019) "Orthobunyavirus spike architecture and recognition by neutralizing antibodies" *Nat Commun*
28. Stubbs, Pontelli, Mishra et al. (2021) "Vesicular stomatitis virus chimeras expressing the Oropouche virus glycoproteins elicit protective immune responses in mice" *mBio*
29. Shi, Goli, Clark et al. (2009) "Functional analysis of the Bunyamwera orthobunyavirus Gc glycoprotein" *J Gen Virol*
30. Stapleford, Coffey, Lay et al. (2014) "Emergence and transmission of arbovirus evolutionary intermediates with epidemic potential" *Cell Host Microbe*
32. Noval, Rodriguez-Rodriguez, Rangel et al. (2019) "Evolution-driven attenuation of alphaviruses highlights key glycoprotein determinants regulating viral infectivity and dissemination" *Cell Rep*
33. Johnson, Noval, Rangel et al. (2020) "Mapping the evolutionary landscape of Zika virus infection in immunocompromised mice" *Virus Evol*
34. Rangel, Catanzaro, Thannickal et al. (2022) "Structurally conserved domains between flavivirus and alphavirus fusion glycoproteins contribute to replication and infectious-virion production" *J Virol*
35. Bennett, Gresko, Nelson et al. (2012) "A recombinant chimeric La Crosse virus expressing the surface glycopro teins of Jamestown Canyon virus is immunogenic and protective against challenge with either parental virus in mice or monkeys" *J Virol*
36. Schneider, Leung, Valenzuela-Leon et al. (2024) "Skin muscle is the initial site of viral replication for arboviral bunyavirus infection" *Nat Commun*
37. Ojha, Winkler, Leung et al. (2021) "Rottlerin inhibits La Crosse virus-induced encephalitis in mice and blocks release of replicating virus from the Golgi body in neurons" *Nat Microbiol*
38. Lambert, Fryxell, Freyman et al. (2015) "Comparative sequence analyses of La Crosse virus strain isolated from patient with fatal encephalitis" *Emerg Infect Dis*
39. Hollidge, Salzano, Ibrahim et al. (2022) "Targeted mutations in the fusion peptide region of La Crosse virus attenuate neuroinvasion and confer protection against encephalitis" *Viruses*
40. Murakami, Takenaka-Uema, Kobayashi et al. (2017) "Heparan sulfate proteoglycan is an important attachment factor for cell entry of akabane and schmallen berg viruses" *J Virol*
41. Mcallister, Liu, Silva et al. (2020) "Chikungunya virus strains from each genetic clade bind sulfated glycosaminoglycans as attachment factors" *J Virol*
42. Rangel, Mcallister, Dancel-Manning et al. (2022) "Emerging chikungunya virus variants at the E1-E1 interglycoprotein spike interface impact virus attachment and inflammation" *J Virol*
43. Wilson, López, Coutermash-Ott et al. (2021) "La Crosse virus shows strain-specific differences in pathogenesis" *Pathogens*
44. Ludwig, Israel, Christensen et al. (1991) "Role of La Crosse virus glycoproteins in attachment of virus to host cells" *Virology (Auckl)*
45. Buchholz, Finke, Conzelmann (1999) "Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter" *J Virol*
46. Schymkowitz, Borg, Stricher et al. (2005) "The FoldX web server: an online force field" *Nucleic Acids Res*
47. Meng, Goddard, Pettersen et al. (2023) "UCSF ChimeraX: tools for structure building and analysis" *Protein Sci*
48. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12797722&blobtype=pdf | # Seroprevalence of hepatitis E in general, hepatic, and pregnant populations in Nepal, Bangladesh, and Pakistan: a systematic review and meta-analysis
Birendra Gupta, Birendra Prasad Gupta, Ajay Sah, Abhay Kumar Sah, Vivek Gupta, Dinesh Ghimire, Nikita Acharya, Chandramani Wagle
## Abstract
Background Hepatitis E virus (HEV) is a major cause of viral hepatitis in low-and middle-income countries, particularly in South Asia, where poor sanitation facilitates its fecal-oral transmission. Nepal, Bangladesh, and Pakistan experience a significant HEV burden, with severe outcomes in high-risk groups like pregnant women and hepatic patients, who face elevated mortality rates. This systematic review and meta-analysis aimed to estimate HEV seroprevalence in these countries, focusing on the general population, pregnant women, and hepatic patients, to inform public health strategies.
MethodsWe searched PubMed, Embase, Cochrane Library, and ProMED-mail for studies published between 2000 and 2017, following PRISMA guidelines and a registered protocol [PROSPERO: CRD42018099558]. Studies reporting HEV seroprevalence in Nepal, Bangladesh, or Pakistan using serological or molecular methods were included. The original search period was retained as data extraction was completed in 2018, and including newer studies was not feasible. A random-effects meta-analysis was conducted using the 'metaprop' function in R, with heterogeneity assessed via I² and τ² statistics. Publication bias was evaluated using Egger's test and funnel plots.
ResultsThe meta-analysis included 64 studies, totaling 25,301 participants. Pooled HEV seroprevalence was 23.8% (95% CI: 13.7%-35.5%) in the general population (n = 9,935), 41.2% (95% CI: 27.6%-55.6%) in pregnant women (n = 4,345), and 41.8% (95% CI: 30.7%-53.4%) in hepatic patients (n = 11,021). High heterogeneity was observed (I²=98.3-99.4%), with significant country-level variation in hepatic patients (p = 0.0002). Subgroup analyses revealed country-specific variations, particularly for hepatic patients where Bangladesh and Nepal showed higher prevalence than Pakistan. Publication bias was evident in pregnant women (p = 0.0002) and hepatic patients (p = 0.0066), but not in the general population (p = 0.6235).Conclusions These findings highlight the substantial burden of HEV in South Asia, with notable differences across population groups and countries. The high seroprevalence in hepatic patients and pregnant women underscores
## Background
Hepatitis E (HE) is a leading cause of waterborne viral hepatitis worldwide, contributing significantly to the global burden of liver disease [1]. It is caused by the Hepatitis E virus (HEV), a single-stranded positive-sense RNA virus with four major genotypes that cause human disease (genotypes 1, 2, 3 and 4) [2]. Genotypes 1 and 2 primarily infect humans, while genotypes 3 and 4 can infect both humans and animals, such as pigs [3]. HE typically presents as an acute, self-limiting illness with clinical symptoms including fever, nausea, jaundice, loss of appetite, and vomiting. However, only about 30% of infected individuals exhibit these symptoms [4].
HEV poses a substantial global health challenge, particularly in low-and middle-income countries (LMICs). Globally, it causes approximately 20 million infections annually, resulting in 3.3 million symptomatic cases, 70,000 deaths and 3,000 stillbirths [5]. In Southeast Asia, viral hepatitis is a leading cause of death, with around 410,000 deaths yearly, and HEV accounts for a substantial proportion of burden [6].
The virus spreads primarily through the fecal-oral route in developing countries, often via contaminated water (genotypes 1 and 2) or undercooked pork and animal products (genotypes 3 and 4). Region with poor sanitation, such as Nepal, Bangladesh and Pakistan, experiencing recurring waterborne outbreak, especially during the monsoon season, which disproportionately affect vulnerable population [6]. Beyond its traditional transmission routes, HEV is now recognized as an emerging transfusion related infection, further complicating its control [7][8][9][10][11].
Certain groups face a higher risk of severe outcomes from HEV infection. Pregnant women, particularly in their third trimester, face mortality of 30% or higher [12]. HE during pregnancy can lead to severe complications, including fulminant hepatic failure, hepatic encephalopathy, disseminated intravascular coagulation, maternal death, fetal distress, premature delivery, or infant death [12,13]. Beyond its traditional fecal-oral transmission, Individuals with pre-existing chronic liver disease are at increased risk of acute liver failure, with mortality rates reaching 43% at 1 month and 70% at 12 months following HEV infection, while healthy individuals without liver conditions are generally at lower risk [11,14]. Similarly, immunocompromised individuals such as solid organ transplant recipients on immunosuppressive drugs are at risk to chronic HEV infections that may persist for years [15,16]. A current study shows that HIV-infected individuals are at higher risk of acquiring acute HEV infection than healthy individuals [1]. and chronic HEV has been reported in patients with compromised immunity due to HIV/AIDS or hematologic malignancies [17,18]. Additionally, HEV is also prevalent among travelers from industrialized countries visiting high-endemic areas like Nepal, Bangladesh, and Pakistan, particularly during outbreaks in the rainy season [19].
Despite the significant burden of HEV in South Asia, there is a lack of comprehensive, region-specific data on its seroprevalence, particularly in high-risk populations. Previous systematic reviews have focused on global HEV prevalence or specific populations like pregnant women in other regions, but none have synthesized seroprevalence data across diverse groups in Nepal, Bangladesh, and Pakistan [6,[19][20][21]. These countries face unique challenges, including poor water quality, inadequate sanitation infrastructure, and limited surveillance systems, which exacerbate HEV transmission and complicate public health responses [6,22]. Understanding the seroprevalence of HEV in these settings is critical for identifying high-risk groups, informing targeted interventions such as improved sanitation, vaccination, and enhanced surveillance, and reducing the associated morbidity and mortality.
This systematic review and meta-analysis aim to synthesize historical evidence (2000-2017) to estimate the seroprevalence of Hepatitis E virus (HEV) in Nepal, Bangladesh, and Pakistan among three key populations: the general population, pregnant women, and hepatic patients. By analyzing the data from studies published between 2000 and 2017, we seek to provide a comprehensive understanding of HEV exposure in these regions. Specifically, we investigated: (1) What is the pooled seroprevalence of HEV in the general population, pregnant women, and hepatic patients across these countries during the study period? (2) How does HEV seroprevalence vary geographically (subgroup analysis by country). These questions remain relevant today, as HEV continues to burden South Asia due to persistent challenges in sanitation and healthcare access. The findings will help guide interventions such as targeted vaccination programs and sanitation improvements, which are critical for mitigating the burden of HEV in South Asia in the present day.
## Methods
## Protocol and registration
The protocol for this systematic review and meta-analysis was registered with PROSPERO [CRD42018099558] in July 2018. PROSPERO is an international database of prospectively registered systematic reviews managed by the Centre for Reviews and Dissemination, University of York. The review was conducted in accordance with the registered protocol. While we could not update the registration on PROSPERO, we have refined the study by reframing the research questions and updating the search strategy to focus on targeted populations (general population, pregnant women, and hepatic patients) and specific regions (Nepal, Bangladesh, and Pakistan). The search period was retained as originally planned (January 1, 2000, to December 31, 2017) to ensure consistency with the registered protocol.
## Eligibility criteria
Studies were included if they reported HEV seroprevalence in Nepal, Bangladesh, or Pakistan, based on abstract screening. The following predefined criteria were applied (Table 1).
## Information sources and search strategy
We conducted a systematic search of the Cochrane Library, MEDLINE (via PubMed), and Embase (via Ovid) databases, as well as the ProMED-mail archive, to identify relevant studies published in English. A systematic search was conducted in accordance with our PROS-PERO-registered protocol (CRD42018099558). While the original protocol included a broader geographical scope, the search strategy was refined prior to study selection to focus specifically on Nepal, Bangladesh, and Pakistan to ensure clinical homogeneity and a focused research question. The final search strategy employed for all databases incorporated three key concepts: (1) Hepatitis E virus, (2) the target countries, and (3) seroprevalence or prevalence. In PubMed, the search string was: ("Hepatitis E" OR "acute hepatitis" OR "acute liver disease" OR "jaundice") AND ("Nepal" OR "Bangladesh" OR "Pakistan") AND ("seroprevalence" OR "prevalence" OR "seroepidemiology"). Filters were applied to limit results to studies published in English between January 1, 2000, and December 31, 2017. This strategy was adapted for the Cochrane Library and Embase. The detailed, final search strategy for each database is provided in S1 Table.
## Selection process
Eligible studies were independently screened by four researchers based on the inclusion criteria outlined in Table 1. The screening process involved an initial review of titles and abstracts to generate a list of potentially relevant studies, followed by a full-text assessment for final inclusion. Duplicate records across databases were first identified and removed using EndNote, followed by manual verification to ensure accuracy. Records retrieved from the search were screened for eligibility, with 2 articles excluded due to language restrictions and 198 fulltext articles excluded for reasons such as Not focused on HEV seroprevalence, or incomplete HEV seroprevalence data, as illustrated in the PRISMA flowchart (Fig. 1). Disagreements in the identification of relevant studies were discussed by the authors until a consensus was reached. The flow of article searches and selection, including the number of records identified, screened, and included, is presented in the PRISMA flowchart (Fig. 1). No automation tools were used in the process. Bibliographic database EndNote was used to manage references.
## Data extraction
Data were extracted independently by authors using a predefined template in Microsoft Excel.
Extracted information included article details (PMID, year of publication, first author, time of sample collection, country/region), study design (sample size, age group, target population, clinical syndrome, study endpoint), pathogen identification methods (serology and/or RT-PCR). Information on HEV antibody type (IgM, IgG, or total) was extracted as reported by each study. However, since not all studies differentiated antibody classes, seroprevalence was recorded as presented in the original article.
## Quality assessment
To ensure the reliability of study outcomes, we used the integrated quality criteria for review of multiple study designs (ICROMS) as quality assessment tool [23]. This tool allowed for a rigorous analysis of the studies, identifying limitations and considering them in the interpretation of the results. ICROMS comprises several criteria, each receiving a score based on whether the criterion is met. A score of 2 points is assigned if a criterion is fully met, 0 points if it is not met, and 1 point if it is unclear whether the criterion is met or if it is not applicable. The total score for each study is calculated by summing the points for all criteria, resulting in a global quality score.
Each study design required a minimum global quality score to be eligible for inclusion in the data analysis. Any disagreements during the quality evaluation process were discussed among the researchers until a consensus was reached.
## Effect measure
Seroprevalence was selected as the primary effect measure. For this review, seroprevalence was defined as the proportion of individuals testing positive for anti-HEV IgG and/or IgM antibodies, depending on the assay used in each study. To address the statistical challenges of synthesizing proportions (e.g., bounded scales [0, 1] and skewed variances), logit transformation was applied for meta-analysis. Following synthesis, results were back transformed to the original percentage scale for presentation, accompanied by 95% confidence intervals (CIs) to quantify the precision of individual study estimates and the pooled seroprevalence.
## Synthesis methods
A random-effects meta-analysis was conducted using the 'metaprop' function in R's meta package, employing the DerSimonian-Laird estimator to account for expected heterogeneity across diverse epidemiological settings. Heterogeneity was assessed using the I-squared (I²) statistic (to measure the percentage of variation due to between-study differences), the tau-squared (τ²) statistic (to estimate the variance of the true effect sizes), and the Cochran's Q statistic (assessing whether observed variations exceed chance levels).
Results were presented in forest plots to visually depict individual study estimates, pooled estimates, and corresponding 95% confidence intervals. Funnel plots were generated to visually explore potential publication bias, plotting study effect sizes (seroprevalence, Freeman-Tukey transformed) against their standard errors, with asymmetry indicating possible bias. Subgroup analysis by country was conducted to explore differences in seroprevalence across Nepal, Bangladesh, and Pakistan. Sensitivity analyses were planned but not conducted due to time and data constraints.
## Reporting bias assessment
Publication bias was assessed using funnel plots and Egger's regression test. Asymmetry in the funnel plot or a significant Egger's test (p < 0.05) would indicate bias toward publishing larger or statistically significant effects. However, interpretation was cautioned due to high heterogeneity, which can distort funnel plot symmetry independent of true bias.
## Certainty assessment
Given the high degree of heterogeneity observed in the meta-analysis, certainty in the pooled seroprevalence estimates is likely to be low. The wide confidence intervals indicate substantial imprecision. Although formal methods like the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) approach were not used, the high heterogeneity and imprecision suggest cautious interpretation of the estimates. Despite these limitations, the seroprevalence estimates provide valuable insights into the HEV burden in the studied populations, though future research with more consistent study designs is needed for more robust conclusions.
## Results
## Study selection
Initially, we identified a total of 1713 articles published between (2000 to 2017), of which 171 were identified from Cochrane library, 553 Medline (via PubMed), 963 Embase (via Ovid) and 26 articles were identified manually from the references of included studies. After screening for eligibility by reviewing full texts, 262 articles remained, of which 64 studies were ultimately included in the meta-analysis. The included studies comprised 16 articles on the general population, 30 on hepatic patients, and 18 on pregnant women published between 2000 and 2017 (Fig. 1).
Exclusions were based on the following criteria: review articles, book chapters, topics unrelated to HEV infection in the specified populations (e.g., animal studies), duplicates, insufficient data, and non-English languages (Table 1). The majority of included studies were hospital-based, cross-sectional designs focusing on hepatic patients, with variations in study design and setting across the dataset.
## Study characteristics
The characteristics of the included studies are summarized below, categorized by the study population. Table 2 presents the study characteristics for the general population. Table 3 for pregnant women, and Table 4 for hepatic patients. Each table includes study design, sample size, seropositive number, proportion, diagnostic methods, and other relevant data.
## Result of individual studies
## HEV in general population
A total of 16 studies reported HEV seroprevalence in the general population across Bangladesh, Nepal, and Pakistan (Table 2). Sample sizes ranged from 20 to 4,751 participants. Diagnostic methods included serology (IgG, IgM, or combined with molecular techniques), and study designs comprised cross-sectional studies and outbreak reports, with cross-sectional studies being more common. The summarized data for each study, including seroprevalence estimates and study characteristics, are presented in Table 2.
## HEV in pregnant women
Eighteen studies reported HEV seroprevalence among pregnant women (Table 3). Sample sizes varied from 5 to 2,227 participants. The majority of studies used serology, while a few incorporated combined serology and molecular methods. Study designs included cross-sectional, cohort, and randomized controlled trials. Detailed seroprevalence estimates and relevant study information are summarized in Table 3.
## HEV in hepatic patients
A total of 27 studies examined HEV seroprevalence among hepatic patients (Table 4). Sample sizes ranged from 12 to 4,751 participants. Diagnostic methods included serology (IgG, IgM, or combined with molecular techniques). Study designs included cross-sectional studies, cohort studies, and outbreak reports. The seroprevalence estimates and study characteristics for each included study are presented in Table 4.
## Characteristics and risk of bias in studies
Pooled HEV seroprevalence estimates were calculated for three population categories using a random-effects model due to significant heterogeneity. In the general population, 16 studies (N = 9935) from Bangladesh (7), For pregnant women, 18 studies (N = 4,345) were analyzed, including data from Bangladesh (4), Nepal (4), and Pakistan (10). Cross-sectional studies were predominant, with a few cohort and randomized controlled trials. Seroprevalence ranged from 3.0% to 80.4%, with one study incorporating molecular techniques. The ICROMS scores (range: 18-25, median: 22) suggest moderate study quality.
Among hepatic patients, 30 studies (N = 11,021) from Bangladesh (16), Nepal (7), and Pakistan (7) reported seroprevalence ranging from 3.7% to 97.9%. Cross-sectional designs were most common, and all studies used serology, with some incorporating molecular methods. The ICROMS scores (17)(18)(19)(20)(21)(22)(23)(24)(25) indicate moderate risk of bias, with higher-quality studies including Harun-Or-Rashid (2013) [54] and Sazzad (2017) [60]. Overall, ICROMS quality scores for included studies ranged from 18 to 25 out of a maximum of 30, indicating moderate to high methodological quality across studies.
## Result of statistical synthesis
## Seroprevalence in general population
The pooled seroprevalence of anti-HEV antibodies in general population was estimated to be 23.8% (CI: 13.7%-35.5%), indicating a high burden of HEV exposure in the region. However, there was significant heterogeneity among studies (I² = 99.4%, τ² = 0.0668), suggesting variability in prevalence across different populations and study settings. The result of the meta-analysis, including individual study estimates and the pooled effect size is presented in the forest plot (Fig. 2).
Study-specific seroprevalence estimates ranged from 1.1% (CI: 0.7%-1.7%) in Pakistan (Saeedi 2004) to 74.6% (CI: 62.5%-84.5%) in Bangladesh (Alam 2009). The prediction interval (0.0%-77.9%) suggests that future studies may report widely varying seroprevalence estimates, further supporting the presence of regional and demographic disparities in HEV exposure. Subgroup analysis by country showed differing seroprevalence estimates. In Bangladesh, the estimate was 26.72% (CI: 14.49% to 41.06%), with moderate heterogeneity (I² = 98.5%). Studies from Pakistan yielded a lower estimate of 19.44% (CI: 1.23% to 50.20%), with high heterogeneity (I² = 99.4%). The studies from Nepal showed an estimate of 23.30% (CI: 9.51% to 40.90%), with an I² of 99.2%. The test for subgroup differences between the countries was not statistically significant (p-value = 0.8809), suggesting that the variation in seroprevalence across countries may not be as large as initially expected (Table 5).
## Seroprevalence in pregnant women
The random effects model provided an overall pooled estimate of 41.23% (CI: 27.56% to 55.57%) seroprevalence. However, there was substantial heterogeneity (I² = 98.3%), indicating significant variability across the studies, which is further confirmed by the high values of tau² (0.0840) and tau (0.2898). The prediction interval ranged from 0.00% to 95.66%, suggesting wide uncertainty about the true seroprevalence (Fig. 3).
Subgroup analysis by country showed seroprevalence of 24.8% (CI: 11.6%-41.1%) in Bangladesh, 52.2% (CI: A test for subgroup differences showed no statistically significant variation between countries (Q = 3.64, p = 0.1623), suggesting that the observed differences in seroprevalence across countries may be due to chance rather than systematic differences (Table 5).
## Seroprevalence in hepatic patient
The meta-analysis of 30 studies (n = 11,021) in hepatic patients revealed a pooled HEV seroprevalence of 41.8% (CI: 30.7%-53.4%) using a random-effects model. Extreme heterogeneity was observed (I²=99.2%, Q = 3580.81, p < 0.0001), reflecting significant betweenstudy variability. The prediction interval (0%-96.3%) indicates a wide range of expected seroprevalence in future studies, highlighting regional and methodological differences (Fig. 4). Subgroup analysis reveals notable differences between countries. For studies from Bangladesh, the estimated seroprevalence was 48.07% (CI: 36.76% to 59.47%), with moderate heterogeneity (I² = 98.6%) and a statistically significant Q-test for heterogeneity (p-value < 0.001). The studies from Nepal showed a slightly lower seroprevalence estimate of 47.26% (CI: 21.85% to 73.45%), with high heterogeneity (I² = 99.1%). In contrast, studies from Pakistan reported a substantially lower estimate of 15.62% (CI: 7.18% to 26.48%), with moderate heterogeneity (I² = 94.9%). The test for subgroup differences was significant (p-value = 0.0002), indicating that the country of origin significantly influenced the seroprevalence estimates (Table 5).
## Result of publication bias
Publication bias was evaluated using Egger's test and funnel plot asymmetry analysis across all three study populations. For hepatic patients, Egger's test indicated significant funnel plot asymmetry (t = -2.94, df = 28, p = 0.0066), suggesting potential publication bias, with a bias estimate of -8.77 (SE = 2.99). Similarly, in the pregnant women subgroup, a significant small-study effect was observed (t = 4.77, df = 16, p = 0.0002), with a bias estimate of 8.04 (SE = 1.69), indicating possible bias (Fig. 5). However, for the general population, Egger's test did not detect significant asymmetry (t = 0.50, df = 14, p = 0.6235), suggesting no strong evidence of publication bias in this subgroup.
## Discussion
This systematic review and meta-analysis provide a comprehensive assessment of the seroprevalence of hepatitis E virus (HEV) in Nepal, Bangladesh, and Pakistan across three population groups: hepatic patients, general Fig. 3 Forest plot of the meta-analysis of HEV seroprevalence in the pregnant women of Nepal, Bangladesh, and Pakistan population, and pregnant women. Our findings reveal substantial HEV exposure, with pooled seroprevalence estimates of 23.8% in the general population, 41.2% in pregnant women, and 41.8% in hepatic patients. These results align with prior regional reports but highlight critical disparities in surveillance quality and epidemiological trends across South Asia [20,21].
However, our estimate for the general population (23.8%) is lower than some reports from other low-and middle-income countries (LMICs), where seroprevalence estimates often exceed 30% in certain regions, reflecting higher endemicity and past exposure [78,79]. This difference may be attributable to variations in diagnostic methods, such as serology versus combined serology and molecular testing, as well as the focus on specific countries within South Asia, which might experience lower overall exposure compared to other LMIC settings. Subgroup analysis by country revealed significant variation in hepatic patients, with seroprevalence lowest in Pakistan (15.6%) compared to Bangladesh (48.1%) and Nepal (47.3%) (p = 0.0002). In contrast, no significant country-level differences were observed in the general population (p = 0.8809) or pregnant women (p = 0.1623), suggesting that other factors, such as study design or diagnostic variability, may contribute more to the observed heterogeneity in these groups. The high seroprevalence in pregnant women (41.2%) is particularly alarming, given the elevated mortality risk associated with HEV infection during pregnancy, which can reach 20-30% in the third trimester [13]. Similarly, the 41.8% seroprevalence in hepatic patients underscores the vulnerability of this group, with prior studies reporting mortality rates of 43% at 1 month and 70% at 12 months following HEV infection in patients with preexisting liver disease [14]. The rising incidence of HEV Fig. 4 Forest plot of the meta-analysis of HEV seroprevalence in hepatic patients of Nepal, Bangladesh, and Pakistan co-infection with other hepatitis viruses (e.g., HBV, HCV) in hepatic patients, as noted in our review, further exacerbates morbidity and mortality in these populations [80].
The extreme heterogeneity across all categories (I²=98.3-99.4%) is a critical finding, likely driven by differences in study settings (e.g., hospital-based vs. community-based, outbreak vs. non-outbreak), study design (cross-sectional, cohort, or outbreak reports), population characteristics (age, urban-rural composition, highrisk groups), and diagnostic methods (e.g., IgG for past exposure in the general population and pregnant women, IgM for acute infection in hepatic patients, and variability in assay types) and regional sanitation levels. It should be noted that specific exposure or risk factor data (e.g., water quality, hygiene practices, or urban-rural differences) were not consistently reported in the included studies; therefore, the influence of these factors on heterogeneity remains speculative. The wide prediction intervals (e.g., 0.0%-96.3% for hepatic patients) highlight the uncertainty in applying these estimates to new settings, emphasizing the need for context-specific data. Publication bias was evident in the pregnant women (p = 0.0002) and hepatic patients (p = 0.0066) categories, suggesting that the pooled estimates for these groups may be overestimated due to the overrepresentation of smaller studies with higher seroprevalence. In contrast, the general population estimate showed no evidence of publication bias (p = 0.6235), enhancing its reliability. Significant publication bias detected in two subgroups suggests potential overestimation of pooled prevalence. Hence, these results should be interpreted cautiously, considering possible underreporting of studies with negative or lower prevalence findings. However, the high residual heterogeneity in Egger's regression (τ² ranging from 25.65 to 166.26) indicates that methodological variability also contributes to the observed differences, and the subjectivity of visual funnel plot inspection warrants cautious interpretation, as noted in the methods [23].
These findings have significant public health implications for South Asia, where HEV remains a major cause of acute viral hepatitis. The high seroprevalence in pregnant women and hepatic patients underscores the urgent need for targeted interventions, such as improving access to safe drinking water and sanitation infrastructure, which are critical for reducing HEV transmission [19]. The potential for HEV vaccination, which has demonstrated efficacy in endemic areas (Zhang et al., 2015), should be explored as a preventive strategy, particularly for high-risk groups like pregnant women [81]. Public health policies should also prioritize education campaigns to raise awareness of HEV transmission risks, particularly during the monsoon season when outbreaks are more frequent. For hepatic patients, early diagnosis and management of HEV infection are essential to reduce mortality, especially in Bangladesh and Nepal, where serprevalence is highest. This study's strengths include adherence to PRISMA guidelines, a comprehensive search strategy, a defined study period (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017), and robust statistical analyses. However, several limitations should be considered. The high heterogeneity across studies limits the ability to draw country-specific conclusions, and publication bias was detected in the hepatic patient and pregnant women subgroups, suggesting underrepresentation of smaller studies with negative findings. Variations in diagnostic Fig. 5 Funnel plot analysis for publication bias assessment across different study population methods (serology alone versus combined serology and molecular testing) and the predominance of hospitalbased studies may also influence prevalence estimates. Exposure-related determinants such as water quality, sanitation, hygiene practices, and urban-rural differences could not be assessed due to inconsistent reporting, though these factors likely contributed to the observed heterogeneity. Subgroup analysis by study design and sensitivity analyses were not feasible because of the imbalance in study numbers, differing designs, and diagnostic approaches. Although this review includes articles published only up to 2017, its findings remain relevant for understanding the HEV burden in South Asia.
## Conclusion
This systematic review and meta-analysis provide important insights into the seroprevalence of HEV in Nepal, Bangladesh, and Pakistan. However, given the substantial heterogeneity across studies and the evidence base limited to publications up to 2017, the findings should be interpreted with caution. The observed higher burden among hepatic patients and pregnant women underscores the continued need for strengthened surveillance, improved sanitation, and further research using standardized diagnostic approaches to better define the evolving HEV epidemiology in South Asia.
## References
1. Shrestha, Adhikari, Bhattarai et al. (2017) "Prevalence and risk of hepatitis E virus infection in the HIV population of Nepal" *Virol J*
3. Who (2025)
4. Bhattarai, Baniya, Aryal et al. (2018) "Epidemiological profile and risk factors for acquiring HBV and/or HCV in HIVinfected population groups in Nepal" *BioMed Res Int*
5. Gupta, Lama, Adhikari et al. (2016) "First report of hepatitis E virus viremia in healthy blood donors from Nepal" *Virusdisease*
6. Who (2025) "Vaccine Standardization HE"
7. Shrestha, Lama, Gupta et al. (2016) "Hepatitis E virus outbreak in postearthquake Nepal: is a vaccine really needed?" *J Viral Hepat*
8. Mushahwar (2008) "Hepatitis E virus: molecular virology, clinical features, diagnosis, transmission, epidemiology, and prevention" *J Med Virol*
9. Bajpai, Gupta (2011) "Transfusion-transmitted hepatitis E: is screening warranted?" *Indian J Med Microbiol*
10. Marano, Vaglio, Pupella et al. (2015) "Hepatitis E: an old infection with new implications" *Blood Transfus*
11. Yazbek, Kreidieh, Ramia (2016) "Hepatitis E virus in the countries of the middle East and North Africa region: an awareness of an infectious threat to blood safety" *Infection*
12. Taherkhani, Farshadpour (2016) "Epidemiology of hepatitis E virus in Iran" *World J Gastroenterol*
13. Gupta, Shrestha, Adhikari et al. (2016) "Acute hepatitis E virus infection in human immunodeficiency virus-positive men and women in Nepal: not quite a rare entity" *Hepatology*
14. Bista (2006) "Acute hepatitis E in pregnancy-study of 16 cases" *JNMA J Nepal Med Assoc*
15. Acharya, Sharma, Singh et al. (2007) "Hepatitis E virus (HEV) infection in patients with cirrhosis is associated with rapid decompensation and death" *J Hepatol*
16. Kamar, Rostaing, Izopet (2013) "Hepatitis E virus infection in immunosuppressed patients: natural history and therapy"
17. Kamar, Selves, Mansuy et al. (2008) "Hepatitis E virus and chronic hepatitis in organ-transplant recipients" *N Engl J Med*
18. Geng, Zhang, Huang et al. "Persistent hepatitis e virus genotype 4 infection in a child with acute lymphoblastic leukemia"
19. Kuniholm, Ong, Hogema et al. (2016) "Acute and chronic hepatitis E virus infection in human immunodeficiency virus-infected US women" *Hepatology*
20. Labrique, Thomas, Stoszek et al. (1999) "Hepatitis E: an emerging infectious disease" *Epidemiol Rev*
21. Labrique, Zaman, Hossain et al. (2010) "Epidemiology and risk factors of incident hepatitis e virus infections in rural Bangladesh" *Am J Epidemiol*
22. Shrestha, Lama, Karki et al. (2014) "Hepatitis e epidemic" *Emerg Infect Dis*
23. Rab, Bile, Mubarik et al. (1997) "Water-borne hepatitis E virus epidemic in Islamabad, pakistan: A common source outbreak traced to the malfunction of a modern water treatment plant" *Am J Trop Med Hyg*
24. Zingg, Castro-Sanchez, Secci et al. (2016) "Innovative tools for quality assessment: integrated quality criteria for review of multiple study designs (ICROMS)" *Public Health*
25. Alam, Azam, Mustafa et al. (2009) "Natural course of fulminant hepatic failure: the scenario in Bangladesh and the differences from the West" *Saudi J Gastroenterol*
26. Ali, Fatima, Shah et al. (2000) "Fulminant hepatic failure -a devastating problem" *J Coll Physicians Surg Pak*
27. Bryan, Tsarev, Iqbal et al. (1994) "Epidemic hepatitis E in pakistan: patterns of serologic response and evidence that antibody to hepatitis E virus protects against disease" *J Infect Dis*
28. Mahtab, Rahman, Jahan et al. (2014) "Epidemiology of hepatitis E virus in an urban community in Dhaka City" *Euroasian J Hepatogastroenterol*
30. Saeedi, Mahmood, Ziauddin et al. (2004) "Frequency and clinical course of hepatitis E in tertiary care hospitals" *J Coll Physicians Surg Pak*
31. Shrestha, Flower, Seed et al. (2016) "Hepatitis E virus seroepidemiology: a post-earthquake study among blood donors in Nepal" *BMC Infect Dis*
32. Labrique, Zaman, Hossain et al. (2009) "Population seroprevalence of hepatitis E virus antibodies in rural Bangladesh" *Am J Trop Med Hyg*
33. Pubmed
34. Sheikh, Sugitani, Kinukawa et al. (2002) "Hepatitis e virus infection in fulminant hepatitis patients and an apparently healthy population in Bangladesh" *Am J Trop Med Hyg*
35. Jafri, Yakoob, Abid et al. (2013) "Seroprevalence of hepatitis E and Helicobacter pylori in a low socioeconomic area of a metropolitan city in a developing country" *Br J Biomed Sci*
36. Izopet, Labrique, Basnyat et al. (2015) "Hepatitis E virus seroprevalence in three hyperendemic areas: Nepal, Bangladesh and southwest France" *J Clin Virol*
37. Begum, Tabassum, Islam et al. (2007) "Seroprevalence of hepatitis E virus (HEV) infection among patients attending BSMMU, Dhaka" *Bangladesh J Med Microbiol*
38. Showkath, Das, Fatema et al. (2015) "Seroprevalence of hepatitis G virus IgG antibody among blood donors, pregnant women, neonates and apparently healthy population" *Faridpur Med Coll J*
39. Karim, Billah, Al-Mahtab et al. (2012) "HEV hepatitis dominates jaundice patients in gyne wards of Bangladesh" *Hepatol Int*
40. Kmush, Labrique, Li et al. (0238) "The association of cytokines and micronutrients with hepatitis E virus infection during pregnancy and the postpartum period in rural Bangladesh" *Am J Trop Med Hyg*
41. Parveen, Begum, Akhter (2015) "Feto-maternal outcome of jaundice in pregnancy in a tertiary care hospital" *Mymensingh Med J*
42. Rana, Pradhan, Manandhar et al. (2009) "Maternal mortality over the last decade: a changing pattern of death due to alarming rise in hepatitis in the latter five-year period" *J Obstet Gynaecol Res*
43. Shrestha, Saha, Karki (2010) "Near miss maternal morbidity and maternal mortality at Kathmandu medical college teaching hospital" *Kathmandu Univ Med J (KUMJ)*
44. Shrestha, Bhandari, Sharma et al. (2009) "A study of viral hepatitis during pregnancy in Nepal medical college teaching hospital" *Nepal Med Coll J*
45. Alvi, Baig, Sabir (2015) "Etiology and epidemiology of patients presenting with liver disease during pregnancy" *Med Forum Mon*
46. Ambreen, Sheikh, Ayub et al. (2015) "Jaundice in pregnancy: a clinical study at Fatima memorial system" *J South Asian Fed Obstet Gynecol*
47. Aziz, Hamid, Iqbal et al. (1997) "Prevalence and severity of viral hepatitis in Pakistani pregnant women: a five year hospital based study" *J Pak Med Assoc*
48. Brohi, Sadaf, Perveen (2013) "Etiology, clinical features and outcome of fulminant hepatic failure in pregnancy" *J Pak Med Assoc*
49. Butt, Malik, Sharif et al. (2016) "Spectrum of liver disease during pregnancy: a cross sectional study from Pakistan" *Hepatol Int*
50. Hamid, Jafri, Khan et al. (1996) "Fulminant hepatic failure in pregnant women: acute fatty liver or acute viral hepatitis?" *J Hepatol*
51. Hossain (2012) "Hepatitis e is an important cause of jaundice during pregnancy in Pakistan" *Int J Gynecol Obstet*
52. Javed, Ullah, Hussain et al. (2017) "Hepatitis e virus seroprevalence in pregnant women in Pakistan: maternal and fetal outcomes" *East Mediterr Health J*
53. Shams, Khero, Ahmed et al. (2001) "Prevalence of hepatitis E virus (HEV) antibody in pregnant women of Karachi" *J Ayub Med Coll Abbottabad*
54. Siddiqui, Soomro, Hasnain (2012) "Severe obstetric morbidity and its outcome in patients presenting in a tertiary care hospital of Karachi" *J Pak Med Assoc*
55. Ashraf-Uz-Zaman, Begum, Asad et al. (2010) "Biochemical parameters in common viral hepatitis" *J Med*
56. Haque, Sultana, Mahtab (2016) "Sero-prevalence of acute hepatitis caused by HAV and HEV in Gazipur" *Hepatology International*
57. Harun-Or-Rashid, Akbar, Takahashi et al. (2013) "Epidemiological and molecular analyses of a non-seasonal outbreak of acute Icteric hepatitis E in Bangladesh" *J Med Virol*
58. Hoque, Islam, Al-Faruk et al. (2015) "Aetiology and outcome of sporadic acute viral hepatitis in Bangladesh" *Hepatology International*
59. Khan, Sultana, Rahman et al. (2000) "Viral hepatitis: recent experiences from serological studies in Bangladesh" *Asian Pac J Allergy Immunol*
60. Md, Moshaed Hossain, Azam et al. (0960) "Aetiology and outcomes of acute viral hepatitis in type 2 diabetic patients in Dhaka Bangladesh"
61. Paul, Nazneen, Banik et al. (2016) "Incidence of adult deaths associated with hepatitis e virus in bangladesh" *American Journal of Tropical Medicine and Hygiene*
62. (2016)
63. Rahman, Sultana, Chowdhury (2007) "Seropositivity and pattern of viral hepatitis in clinically suspected cases of hepatitis in Dhaka City" *Bangladesh Med Res Counc Bull*
64. Sazzad, Luby, Labrique et al. (2017) "Risk factors associated with blood exposure for sporadic hepatitis E in Dhaka, Bangladesh" *Am J Trop Med Hyg*
65. Sugitani, Sheikh, Suzuki et al. (2009) "Sero-epidemiology of sporadic acute hepatitis in Bangladesh: High prevalences of infection with type-B, type-E and multiple types of hepatitis virus" *Ann Trop Med Parasitol*
66. Sugitani, Tamura, Shimizu et al. (2009) "Detection of hepatitis E virus RNA and genotype in Bangladesh" *J Gastroenterol Hepatol*
67. Gurley, Halder, Streatfield et al. (2012) "Estimating the burden of maternal and neonatal deaths associated with jaundice in Bangladesh: possible role of hepatitis E infection" *Am J Public Health*
68. Haque, Banu, Chowdhury et al. (2015) "An outbreak of hepatitis e in an urban area of Bangladesh" *J Viral Hepat*
69. Rashid (2017) "Outcome of acute viral hepatitis in diabetic and non diabetic patients in Bangladesh-report from a tertiary centre" *J Clin Exp Hepatol*
70. Adhikari, Gupta, Manandhar (2015) "Co-circulation of hepatitis A, B and C with hepatitis E virus genotype 1a during major outbreak of 2014 in Nepal" *J Viral Hepat*
71. Kc (2007) "Acute liver failure caused by hepatitis E virus and Paracetamol" *JNMA J Nepal Med Assoc*
72. Shrestha, Shrestha, Tsuda et al. (2003) "Molecular investigation of hepatitis E virus infection in patients with acute hepatitis in Kathmandu, Nepal" *J Med Virol*
73. Sudhamshu (2006) "Ultrasound findings in acute viral hepatitis" *Kathmandu Univ Med J (KUMJ)*
74. Sudhamshu, Khadka, Sharma et al. (2011) "Acute pancreatitis in acute viral hepatitis" *J Nepal Med Assoc*
75. Sudhamshu, Sharma, Poudya et al. (2014) "Acute viral hepatitis in pediatric age groups" *J Nepal Med Assoc*
76. Yadav, Chaudhari, Podel et al. (2013) "Prevalence of Hepatitis E Virus (HEV) in Jaundice Patients Admitted in Different Hospital of Kathmandu Valley" *Hepatology International*
77. Ahmed, Qureshi, Arif et al. (2010) "Changing trend of viral hepatitis -a twenty one year report from Pakistan medical research Council research Centre, Jinnah postgraduate medical Centre, Karachi" *J Pak Med Assoc*
78. Bashir, Hussain, Hasnain et al. (2009) "Seroprevalence of hepatitis E virus immunoglobulin G and M antibodies in adults: a hospital-based study" *Indian J Med Microbiol*
79. Centers For Disease (2009) "Establishment of a viral hepatitis surveillance system-Pakistan"
80. Siddique, Farzand, Waheed et al. (2012) "Frequency of hepatitis E (HEV) patients in medical unit-I of a tertiary care teaching hospital Lahore" *Pak J Med Health Sci*
81. Al, Rahman, Khan et al. (2009) "HEV infection as an aetiologic factor for acute hepatitis: experience from a tertiary hospital in Bangladesh" *J Health Popul Nutr*
82. Mirzaev, Ouoba, Ko et al. (2024) "Systematic review and meta-analysis of hepatitis E seroprevalence in Southeast asia: a comprehensive assessment of epidemiological patterns" *BMC Infect Dis*
84. Qashqari (2022) "Seroprevalence of Hepatitis E Virus Infection in Middle Eastern Countries: A Systematic Review and Meta-Analysis"
85. (0905)
86. Bricks, Senise, Pott-Jr et al. (2019) "Previous hepatitis E virus infection, cirrhosis and insulin resistance in patients with chronic hepatitis C" *Braz J Infect Dis*
87. Zhang, Zhang, Huang et al. (2015) "Long-term efficacy of a hepatitis E vaccine" *N Engl J Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548456&blobtype=pdf | # Human herpesvirus 6B U65 binds to histone proteins and suppresses interferon production
Haokun Li, Hirohito Ogawa, Da Teng, Yuki Okame, Hikaru Namba, Tomoyuki Honda
## Abstract
Human herpesvirus 6B (HHV-6B), a member of the Betaherpesvirinae subfamily, is a T-lymphotropic virus that causes exanthem subitum and has been implicated in neuroinflammatory conditions such as multiple sclerosis. The tegument proteins, which are characteristic of herpesviruses, play a crucial role in the envelopment of virions and evasion of host immune defenses, such as the interferon β (IFNβ) signaling pathway. However, the precise mechanisms through which the HHV-6B tegument proteins modulate the IFNβ pathway are not yet fully understood. In this study, we identified a novel function of the HHV-6B tegument protein U65 as an inhibitor of IFNβ production. Additionally, two host histone proteins, hCG_2039566 (H2ACG) and H2AC7, were identified as positive regulators of innate immune pathways. U65 interacts with H2ACG and H2AC7, impairing their ability to promote the IFNβ pathway. Further more, we demonstrated that U65 plays critical roles during HHV-6B infection. This study highlights a critical strategy employed by HHV-6B to evade immune defenses, shedding light on its mechanisms for counteracting host responses. IMPORTANCE HHV-6B is a virus that primarily infects T cells and can cause illnesses like exanthem subitum and is linked to neurological conditions such as multiple sclerosis. Like other herpesviruses, HHV-6B likely has special proteins that help it avoid the body's immune defenses, such as the IFNβ signaling pathway, which plays a key role in fighting viral infections. However, how these viral proteins interfere with the immune response is not yet fully understood. In this study, we discovered that one of the HHV-6B proteins, U65, blocks the production of IFNβ, weakening the antiviral defenses. We also found that two human proteins, hCG_2039566 and H2AC7, promote the immune system, but U65 prevents them from doing so. This study reveals a new way that HHV-6B escapes the immune system, providing insight into how the virus efficiently establishes infections and how we might target its ability to evade immunity.
Interferons (IFNs) are antiviral cytokines crucial for the innate immune response to viral infections. In response to viral infection, cells produce and release type I IFNs that act on themselves and neighboring cells, triggering the transcription of hundreds of IFN-stimulated genes (ISGs). These gene products combat viral infections directly, e.g., by inhibiting viral replication, or indirectly by modulating subsequent immune responses (11,12). Among the various subtypes of type I IFNs, IFNβ can be produced by almost all cells in the body (12,13). During infection, dsDNA of herpesviruses is detected by pattern recognition receptors (PRRs), such as cyclic GMP-AMP synthase (cGAS). The binding of cGAS to viral dsDNA allosterically activates its catalytic activity and leads to the production of 2′3′-cyclic GMP-AMP (cGAMP), a second messenger molecule that stimulates stimulator of interferon genes (STING) (14)(15)(16)(17). The carboxyl terminus of stimulated STING recruits and activates TANK-binding kinase 1 (TBK1), which in turn phosphorylates the transcription factor interferon regulatory factor 3 (IRF3) (18,19). The phosphorylated IRF3 dimerizes and then enters the nucleus to induce expression of IFNβ. Stimulated STING also activates I kappa B kinase, which phosphorylates the I kappa B (IkB) family of inhibitors of the nuclear factor-κB (NF-κB). NF-κB has the potential to cooperate with IRF3, thereby enhancing the maximal expression of IFNβ (20,21).
The IFNβ response to herpesvirus infection is multifaceted, involving distinct temporal activation patterns at various stages of the viral life cycle, and plays impor tant roles in herpesvirus infection (22)(23)(24)(25)(26). Therefore, herpesviruses have evolved to suppress the IFNβ pathway. Consistently, several studies have reported the impact of HHV-6 on IFNβ production. HHV-6 encodes immediate-early 1 (IE1) proteins that can suppress IFNβ production by disrupting key signaling molecules such as IRF3, TBK1, and mitochondrial antiviral-signaling protein (MAVS), thereby impairing antiviral gene expression (27,28). HHV-6B IE1 further inhibits type I IFN responses by binding to signal transducer and activator of transcription 2 (STAT2), leading to nuclear accumulation of STAT2 and blockade of ISG expression (22). Additionally, HHV-6 infection alters cytokine profiles in infected monocytes and macrophages, increasing interleukin (IL)-10 while suppressing IL-12, thus shifting immune responses toward an anti-inflammatory state (29). The HHV-6B tegument protein U54 binds to the calcineurin phosphatase enzyme, disrupting the proper dephosphorylation and nuclear translocation of nuclear factor of activated T cells proteins, and thereby resulting in suboptimal IL-2 expression (30). However, the molecular mechanisms underlying the immune evasion strategies of HHV-6 remain incompletely understood.
In this study, we focused on the HHV-6B tegument protein U65, which is a homolog of HCMV UL94 (31,32). We demonstrated that U65 serves as an inhibitor of the innate immune response related to particular host histone proteins, i.e., hCG_2039566 (herein referred to as H2ACG) and H2AC7. Overexpression of U65 suppressed the induction of IFNβ triggered by a viral DNA analog. U65 interacted with H2ACG and H2AC7, suppress ing their ability to modulate the intensity of the antiviral response. Consequently, U65 impairs the H2ACG-and H2AC7-related antiviral pathways, thereby promoting HHV-6B replication. These findings enhance our understanding of the mechanisms of HHV-6B immune evasion during the infection.
## RESULTS
## HHV-6B U65 inhibits IFNβ production
HHV-6B, a dsDNA virus, has evolved multiple strategies to antagonize innate immune pathways, thereby facilitating its replication (33)(34)(35)(36). HHV-6B and HCMV belong to the same β-herpesvirus subfamily, and the HCMV tegument proteins, such as UL48, UL82/83, UL94, and UL97, are known to play roles in immune suppression (37)(38)(39)(40). Among the homologous tegument proteins, the function of U65, a homolog of HCMV UL94, in HHV-6B remains unclear. Therefore, we hypothesized that U65 might promote viral infection by overcoming the restriction by innate immune responses. To inves tigate the role of U65 in the IFNβ pathway, we first explored whether U65 could suppress the activation of the IFNβ promoter by poly(deoxyadenylic-deoxythymidylic) acid [poly(dA:dT)], a synthetic viral dsDNA analog. 293T cells were transfected with an IFNβ promoter reporter plasmid (p125-luc), along with a green fluorescent protein (GFP)-expressing plasmid or increasing amounts of a GFP-U65-expressing plasmid. At 24 h post-transfection, the cells were stimulated with poly(dA:dT). Poly(dA:dT) activated the IFNβ promoter in GFP-expressing cells, whereas the activation of the IFNβ promoter was suppressed in GFP-U65-expressing cells in a dose-dependent manner (Fig. 1A). Since p125-luc is of murine origin, we then validated these results by measuring the human IFNβ mRNA levels. The expression of the IFNβ mRNA was robustly induced by poly(dA:dT) in GFP-expressing cells, whereas the induction was suppressed in GFP-U65expressing cells in a dose-dependent manner (Fig. 1B). Dose-dependent expression of GFP-U65 was confirmed by Western blot analysis (Fig. 1C). These results suggest that HHV-6B U65 inhibits the poly(dA:dT)-triggered IFNβ production pathways.
## HHV-6B U65 does not affect the cytoplasmic IFNβ production pathway
Given that HHV-6B U65 suppresses the IFNβ production pathway (Fig. 1), we investigated its effect on the phosphorylation of signal transducer and activator of transcription 1 (STAT1), a key downstream effector of the IFNβ signaling pathway (41). Consistent with the inhibitory effect of U65 on the IFNβ production pathway, U65 expression reduced STAT1 phosphorylation (Fig. 2A andB), indicating that U65 functionally suppresses the IFNβ signaling pathway. The cytosolic DNA sensing pathways, such as the cGAS-STING pathway, play a central role in the innate immune defense against herpesvirus infections (42). Therefore, we hypothesized that U65 might suppress innate immune response by inhibiting the DNA sensing pathways. To assess the effect of U65 on the DNA sensing pathways, we examined phosphorylation of TBK1, IRF3, and IkBα, which are hallmarks of activation of the cytoplasmic process in these pathways (18). U65 expression did not inhibit the poly(dA:dT)-triggered phosphorylation of TBK1, IRF3, and IkBα (Fig. 2C andD). These results suggest that U65 does not suppress the cytoplasmic process in the DNA sensing pathway. We then examined the intracellular localization of U65 and found that U65 localized both in the cytoplasm and the nucleus of transfected 293T cells (Fig. 2E). Taken together, these results suggest that nuclear-localizing U65 likely suppresses the nuclear process of the IFNβ production pathway.
## HHV-6B U65 suppresses the H2ACG-and H2AC7-related IFNβ pathways
We next investigated the molecular mechanisms by which U65 suppresses the nuclear IFNβ pathway by identifying U65-binding proteins. We expressed GFP or GFP-U65 in 293T cells, immunoprecipitated using an anti-GFP nanobody, determined coimmuno precipitated proteins by liquid chromatography-mass spectrometry (LC-MS), and then identified proteins coprecipitated with GFP-U65 but not with GFP. Since U65 likely inhibits the nuclear IFNβ production pathway (Fig. 2), we selected 16 nuclear-localizing proteins among the identified U65-binding proteins. Notably, they included five histone variant proteins (Fig. 3A and Table 1). Histones act in the nucleus and particular histone variant proteins are reported to regulate IFNβ production (43)(44)(45)(46)(47)(48). Based on these reports, the enrichment of histone proteins in our candidates prompted us to further investigate the involvement of these histones in the IFNβ production pathway. We first examined whether identified histone proteins affect the IFNβ production pathway in our experimental setting. 293T cells were transfected with p125-luc, along with each histone protein-expressing plasmid (H2ACG, H2AC7, H2BC13, H2BC14, or ). In our experimental setting, expression of H2ACG or H2AC7 promoted the poly(dA:dT)-triggered activation of the IFNβ promoter (Fig. 3B). We further confirmed that H2ACG or H2AC7 expression promoted the IFNβ promoter activity and the IFNβ mRNA expression activated by poly(dA:dT) in a dose-dependent manner (Fig. 3C through E). The effect of H2ACG or H2AC7 expression on the IFNβ production was also confirmed at the protein level in culture media using enzyme-linked immunosorbent assay (ELISA) (Fig. 3F).
We then examined the effect of U65 on the H2ACG-or H2AC7-mediated IFNβ promoter regulation. Expression of H2ACG or H2AC7 promoted the activation of IFNβ promoter in the absence of U65 (Fig. 4A,left). On the other hand, the expression of these histone proteins did not promote the activation of the IFNβ promoter in the presence of U65 (Fig. 4A, right). Furthermore, we confirmed the interaction of H2ACG or H2AC7 with U65 using coimmunoprecipitation (Fig. 4B). To obtain physiological relevance, we also demonstrated the interaction between endogenous H2A and U65 although the antibody for H2A used in this study is not specific to H2ACG and H2AC7 (Fig. 4C). Consistent with the results described above, the codistribution of H2ACG or H2AC7 with U65 was observed in an immunofluorescence assay (IFA) (Fig. 4D). Taken together, our results indicate that the HHV-6B tegument protein U65 inhibits innate immune responses by interfering with the IFNβ production influenced by H2ACG and H2AC7.
## HHV-6B U65 contains a peptide sequence with similarity to histone proteins
To gain insights into the underlying mechanism, we compared the amino acid sequences of H2ACG and H2AC7 with those of canonical H2A and the well-known variant H2A.X (Fig. 5A). We identified that H2ACG and H2AC7 share a lysine (K) residue at position 100, which is substituted with arginine in canonical H2A and glycine in the H2A.X variant (Fig. 5A, highlighted in red). This suggests that the presence of additional lysine residues in H2ACG and H2AC7 may allow for post-transcriptional modifications (PTMs), potentially contributing to the induction of IFNβ expression (43,49,50). Consistently, the lysine residue unique to H2ACG and H2AC7 is predicted to be ubiquitinated (Fig. 5B). Next, to evaluate the possibility that U65 mimics histone proteins, we compared the amino acid sequence of U65 with canonical H2A, H2ACG, and H2AC7 (Fig. 5B). Sequence alignment revealed that U65 contains a peptide with similarity to the C-terminal histone tail domain of H2A proteins (Fig. 5B,underlined). Notably, a K residue at position 281 and a serine (S) residue at position 284 in U65 are predicted to undergo ubiquitination and phosphorylation, respectively, analogous to the ubiquitination and phosphorylation (FLAG-H2ACG or H2AC7). The cells were stimulated with poly(dA:dT) for 18 h. Total RNA was extracted for RT-qPCR analysis. (E) Dose-dependent expression of FLAG-H2ACG or H2AC7. The cell lysates were subjected to Western blot analysis with an anti-FLAG antibody. A blot with an anti-β-actin antibody was included as a loading control. (F) Effects of histone proteins on the IFNβ protein level. 293T cells were transfected with each histone-expressing plasmid (FLAG-H2ACG or H2AC7). The cells were stimulated with poly(dA:dT) for 18 h. Culture media were collected for enzyme-linked immunosorbent assay of secreted IFNβ. Values are expressed as the means + SE of at least four independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001. n.s., no significance. a Prot_score is a protein score that was calculated using Mascot software. Prot_matches are the numbers of peptides that were identified in the samples. observed in canonical H2A (51,52). These results support the hypothesis that U65 may function as a histone mimic in U65-mediated immune evasion.
## Knockdown of HHV-6B U65 restores IFNβ production and reduces viral load during HHV-6B infection
We finally examined the role of U65 in the inhibition of IFNβ production during HHV-6B infection. 293T-hCD134 cells, a cell line stably expressing an HHV-6 entry receptor, human CD134, were first transfected with a sh-U65-expressing plasmid and then infected with the HHV-6B Z29 strain at a multiplicity of infection (MOI) of 0.1 (Fig. 6A). Knockdown of U65 by sh-U65-1 or sh-U65-2 successfully decreased U65 mRNA expression (Fig. 6B, left). In this setting, IFNβ mRNA expression following HHV-6B Z29 infection was increased by U65 knockdown (Fig. 6B, right), consistent with the results described above, and the number of the infected cells (p41-positive cells) or the viral load was reduced as expected (Fig. 6C through E). Altogether, these results demonstrate that U65 mediates immune evasion during HHV-6B infection to support efficient infection.
## DISCUSSION
Herpesviruses constitute a large family of enveloped dsDNA viruses that establish a lifelong persistent infection in the host. The viral DNA in the infected cells is recognized by PRRs, which trigger IFNβ production and subsequent antiviral responses. These responses inhibit viral replication and promote the elimination of infected cells (15,17,18,53). On the other hand, herpesviruses have developed diverse strategies to counteract the innate immune responses. For example, Kaposi's sarcoma-associated herpesvirus (KSHV) tegument protein ORF52 has been shown to inhibit the enzymatic activity of cGAS, disrupting the synthesis of cGAMP (54). Similarly, the KSHV tegument protein ORF33 interacts with STING and MAVS and facilitates the recruitment of host protein phosphatase, Mg 2+ /Mn 2+ -dependent 1G to dephosphorylate phosphorylated STING and MAVS, thereby suppressing immune responses (55). The Epstein-Barr virus (EBV) tegument protein BGLF2 has been identified as a potent suppressor of the Janus kinase-signal transducer and activator of transcription pathway, through which ISGs are induced (56). Additionally, the HCMV tegument protein UL94 has been shown to interact with mediator of IRF3 activation (MITA), impairing the recruitment of TBK1 to the MITA microsome and suppressing the induction of IFNβ triggered by cytosolic dsDNA and DNA viruses (38). Based on the comparison with the homologous HCMV tegument proteins known for their inhibitory effects on the antiviral innate immune response, we selected to investigate HHV-6B U65 in this study as the effect of U65 on IFNβ production has not been reported. The HHV-6B tegument protein U65 is homologous to HCMV UL94 that acts as a core herpesvirus structural component to facilitate the secondary envelopment of virions and targets MITA to disrupt the recruitment of TBK1 to the MITA microsome, thereby evading the antiviral immune response (38,57). Expression of U65 suppressed the poly(dA:dT)-triggered IFNβ induction (Fig. 1A andB). Conversely, knockdown of U65 enhanced HHV-6B-induced IFNβ production (Fig. 6B). Additionally, we demonstrated that knockdown of U65 reduced viral replication during HHV-6B infection (Fig. 6C through E). Collectively, HHV-6B U65 plays a critical role in immune evasion during HHV-6B infection.
Histones are essential nuclear proteins that form the nucleosome, the fundamental structural unit of chromatin fibers in eukaryotes (58), and play key roles in regulating gene expression (59). The nucleosome, composed of an octamer of core histones-two each of H2A, H2B, H3, and H4-serves as the basic unit of chromatin organization (60). Structurally, the histone octamer is organized such that an H3-H4 tetramer forms the central scaffold, flanked by two H2A-H2B dimers. Each H2A-H2B dimer interacts with the H3-H4 tetramer to create a "dimer of dimers" arrangement, stabilizing the DNA-protein complex (61,62). This architecture not only provides a compact packag ing system for DNA but also serves as a dynamic platform for epigenetic regulation through histone modifications and variant incorporation (63,64). Multiple nucleosomes form higher-order chromatin fibers whose compaction states influence transcriptional activity (65). Thus, while the nucleosome represents the primary level of chromatin organization, the collective arrangement of nucleosomes and their associated proteins defines the structural and functional landscape of chromatin in the nucleus (66). Within the nucleosomes, histone variants such as H2A.X, H2A.Z, and macroH2A can replace canonical H2A, modulating chromatin structure and function (67). Recent studies have shown that nucleosomes containing the histone variant H2A.Z recruit BRD2, which facilitates transcriptional repression and suppresses IFNβ responses, potentially contributing to immune evasion (43,68). In this study, we noticed that 5 of 16 identi fied U65-binding nuclear proteins were histone variant proteins (Fig. 3A and Table 1) (60). We found that overexpression of histone proteins, H2ACG or H2AC7, promoted the production of IFNβ (Fig. 3B through F). To explore the potential mechanisms, we performed amino acid sequence alignment of H2ACG and H2AC7 with canonical H2A and the well-characterized variant H2A.X (Fig. 5A). We identified a K residue at position 100 unique to H2ACG and H2AC7. Given that lysine residues are often subjected to PTMs in histone proteins (69), we hypothesize that this lysine may undergo modifications, e.g., ubiquitination (predicted in Fig. 5B), potentially promoting IFNβ production (Fig. 7,left). Further investigation is needed to test this hypothesis.
U65 compromises the innate immune response to HHV-6B by targeting the H2ACGand H2AC7-mediated regulation of the IFNβ pathway at least in part (Fig. 4). U65 may inhibit the function of histone variants, H2ACG and H2AC7, through two distinct mechanisms (Fig. 7). First, U65 may function as a mimic of H2ACG and H2AC7, disrupting their regulation of the IFNβ pathway (Fig. 7, middle). U65 contains a peptide sequence (positions 276-284) analogous to the histone tail of these H2A variants (positions 115-123) (Fig. 5B), which may act as a histone mimic by potentially undergoing competitive ubiquitination at the K residue at position 281 and phosphorylation at the S residue at position 284 in U65, thereby suppressing ubiquitination of the K residue at position 121 and phosphorylation of the S residue at position 123 in the H2ACG and H2AC7, respec tively. In this context, in addition to the K residue at position 100, ubiquitination of the K residue at position 121 and phosphorylation of the S residue at position 123 in H2ACG and H2AC7 are likely required for their function. Alternatively, U65 may inhibit the incorporation of the IFNβ-enhancing histone variants, H2ACG and H2AC7, into the IFNβ promoter, thereby suppressing IFNβ production (Fig. 7, right). These models are plausible because recent studies have demonstrated that histone modulation by viral proteins, such as mimicking or hijacking host histone regulatory machineries, is a common strategy for efficient viral infection (70). For example, the influenza virus encodes a histone mimic that suppresses antiviral responses by disrupting host transcriptional programs (71,72). The HSV protein ICP0 facilitates histone removal from viral DNA during lytic infection (73), while it also induces chromatin de-repression of latent viral genomes through modulation of host histone marks (74). In EBV, the histone variant H2A.Z cooperates with the EBV latency maintenance protein EBNA1 to sustain a viral chromatin landscape during latency (75).
In summary, although the mechanism on how U65 compromises the H2ACG-and H2AC7-mediated regulation of IFNβ production remains unknown, we identified previously uncharacterized roles of U65 as an inhibitor of these histones in the immune evasion of HHV-6B during the infection. These findings expand our understanding of the mechanisms employed by HHV-6B to ensure persistent infection and enhance our knowledge of the virus-host interplay. A limitation of this study is that HHV-6B infection was evaluated only in 293T-hCD134 cells. Future validation using cell lines commonly employed for HHV-6B infection, such as MT-4 cells, is needed. Given the role of U65 in immune evasion and potential involvement in secondary virion envelopment, therapeu tic strategies targeting U65 represent a promising approach to controlling HHV-6B infection.
## MATERIALS AND METHODS
## Cells
293T cells (a human embryonic kidney cell line, ATCC #CRL-3216) and 293T-hCD134 cells, which are overexpressing human CD134 to facilitate HHV-6B infection (76-78), were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal bovine serum at 37°C and 5% CO 2 .
## Antibodies and compounds
Rabbit antibodies against human phosphorylated TBK1 Ser172 (clone: D52C2), phosphorylated IRF3 Ser386 (clone: E7J8G), phosphorylated IkBα Ser32 (clone: 14D4), phosphorylated STAT1 Tyr701 (clone: 9167), and histone H2A (clone: 2578) were purchased from Cell Signaling Technology (Danvers, MA, USA). Mouse antibodies against β-actin (clone: AC-15) and anti-HHV-6 p41 early antigen (clone: 9A5D12) were purchased from FUJIFILM Wako (Osaka, Japan) and Santa Cruz Biotechnology (Dallas, TX, USA), respectively. A rabbit antibody against FLAG (F7425) was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Mouse antibodies against hemagglutinin (HA) (M180-3) and FLAG (M185-3L) were purchased from MBL (Nagoya, Japan). Alexa Fluor 488-conjugated goat anti-mouse IgG (H + L) F(ab′) 2 (A-11017) and Alexa Fluor 555-con jugated goat anti-mouse IgG (H + L) F(ab′) 2 (A-21425) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (H + L) (115-035-062) was purchased from Jackson ImmunoResearch (West Grove, PA, USA). HRP-conjugated anti-rabbit IgG (H + L) (catalog no. 474-1516) was purchased from KPL (SeraCare Life Sciences, Milford, MA, USA). Anti-HA-tag mAb-Mag netic Agarose (M132-10), poly(dA:dT) (catalog no. tlrl-pic, tlrl-patn-1), and polyethyleni mine "Max" were purchased from MBL, InvivoGen (Toulouse, France), and Polysciences (Warrington, PA, USA), respectively.
## Plasmids
Human cDNAs encoding H2ACG, H2AC7, H2BC13, H2BC14, H2BU1, and the HHV-6B tegument U65 were cloned into three different vectors (pCMV-GFP, pCAGSS-HA, and pCAGSS-FLAG) based on the required tags. For a plasmid expressing shRNA against
## Luciferase reporter assays
293T cells were transfected with p125-Luc and pRL-TK (Promega, Fitchburg, WI, USA), along with the indicated expression plasmids, and incubated for 24 h at 37°C and 5% CO 2 . At 24 h post-transfection, the cells were further transfected with 0.4 µg of poly(dA:dT) using polyethylenimine "Max" (1 mg/mL) and lysed using passive lysis buffer (TOYO INK, Tokyo, Japan) at 18 h post-poly(dA:dT) stimulation. Luciferase assays were performed as previously described ( 80) with some modifications using the PicaGene Dual Sea Pansy Luminescence Kit (Wako, catalog no. 301-05584), in which luciferase activity was quantified with the LUMAT3 LB 9508 luminometer (Berthold, Bad Wildbad, Germany). Relative luciferase activity was calculated by normalizing Firefly luciferase activity to Renilla luciferase activity.
## Real-time PCR (quantitative PCR) analyses
Reverse transcription real-time quantitative PCR (RT-qPCR) was performed as previously described (81-83) with minor modifications. Total RNA from 293T cells was extracted using TRI Reagent (Molecular Research Center, Cincinnati, Ohio, USA) according to the manufacturer's protocol. mRNA was reverse transcribed into cDNA using a Verso cDNA Synthesis Kit (Thermo Fisher Scientific) with oligo-dT primers. The relative mRNA levels of human IFNβ were measured using a CFX Connect Real-Time System (Bio-Rad Labora tories, USA) with THUNDERBIRD Next SYBR (TOYOBO, Osaka, Japan) and normalized to GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression. Total DNA was extracted from infected cells using QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. Quantitative PCR (qPCR) assays for HHV-6B genomic DNA were carried out using THUNDERBIRD Next SYBR (TOYOBO) and the HHV-6B-specific primers. The gene-specific primers used in this study are as follows: human IFNβ forward, 5′-GCC GCA GTG ACC ATC TAT GA-3′; human IFNβ reverse, 5′-CTC ATG CGT TTT CCC CTG GT-3′; human GAPDH forward, 5′-AGC GAG ATC CCT CCA AAA TC-3′ (84); human GAPDH reverse, 5′-AAA TGA GCC CCA GCC TTC TC-3′ (84); HHV-6B-U65 forward, 5′-TTG CAT GCA TTG CGA GAT GG-3′; HHV-6B-U65 reverse, 5′-GCT CCG GTG TAA CAC AAT GC-3′; HHV-6B-genome forward, 5′-TTT GCA GTC ATC ACG ATC GG-3′ (85); HHV-6B genome reverse, 5′-AGA GCG ACA AAT TGG AGG TTT C-3′ (85).
## Western blot
Western blot was performed as previously described ( 86) with some modifications. Briefly, 293T cells were lysed in sodium dodecyl sulfate (SDS) sample buffer. The cell lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophore sis (SDS-PAGE) and transferred onto polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). The membranes were then blocked with 3% skim milk in phosphatebuffered saline (PBS) containing 0.05% Tween 20 for 1 h at room temperature, followed by incubation with primary antibodies for 1 h at room temperature. After washing, the membranes were incubated with secondary antibodies. The bound antibodies were visualized using Clarity Western ECL Substrate (Bio-Rad) and detected with a MultiImager II (BioTools, Gunma, Japan).
## Identification of U65-binding proteins
GFP-tagged U65-or GFP-expressing plasmids were transfected into the cells using polyethylenimine "Max". The cells were washed four times with cold PBS and subse quently lysed in lysis buffer (
## ELISA
IFNβ protein concentrations in the culture media were determined using the Authenti Kine Human IFN-beta ELISA Kit (Proteintech) according to the manufacturer's protocol. Briefly, 293T cells were transfected with the indicated plasmids. At 24 h post-transfection, the cells were further transfected with 0.4 µg of poly(dA:dT). At 18 h post-poly(dA:dT) stimulation, the culture media were collected and the IFNβ protein concentration was evaluated.
## Coimmunoprecipitation
Coimmunoprecipitation was performed as previously described ( 78) with some modifications. Briefly, 293T cells were transfected with pCAGSS-HA-U65 and pCAGSS-FLAG-H2ACG/H2AC7 or pCAGSS-HA-U65 alone. At 48 h post-transfection, the cells were washed four times with cold PBS and subsequently lysed in lysis buffer at 4°C. The supernatant was collected by centrifugation at 12,000 × g for 5 min at 4°C. The lysates were incubated with anti-HA-tag mAb-magnetic agarose on a rotator at 4°C overnight.
After washing four times with cold lysis buffer, the agarose was boiled in SDS sample buffer and subjected to SDS-PAGE. The subsequent Western blot was performed as described above.
## Sequence alignment and PTM site prediction
Amino acid sequences of human canonical H2A (UniProt ID: P04908), H2A.X (Uni Prot ID: P16104), H2ACG (UniProt ID: A0A0U1RR32), H2AC7 (UniProt ID: P20671), and U65 (UniProt ID: Q9QJ19), were retrieved from the UniProt database (https://www.uni prot.org/). Multiple sequence alignment was performed using MAFFT (version 7) with default parameters via the online server at https://mafft.cbrc.jp/alignment/server/ (89). Alignment results were manually inspected, and conserved regions were annotated. Prediction of PTM sites, including phosphorylation, ubiquitination, acetylation, and methylation, was performed using MusiDeep (SVM version) at https://www.musite.net/ (90).
## Virus infection
HHV-6B strain Z29 was propagated in cord blood mononuclear cells, and virus stocks were prepared as previously described (91). 293T-hCD134 cells were transfected with a sh-U65-expressing plasmid, along with the indicated plasmids, and incubated for 24 h. The cells were then incubated with HHV-6B Z29 (MOI = 0.1) at 37°C. After viral absorption for 1 h, the cells were washed with DMEM and then cultured for an additional 3 days. At the indicated time point, the mRNA levels of human IFNβ and viral copies in the infected cells were measured by RT-qPCR and qPCR, respectively.
## IFA
IFA was performed as previously described (86) with some modifications. Briefly, 293T-hCD134 cells (5 × 10 4 ) were plated onto chamber slides. The cells were transfected with a sh-U65-expressing plasmid using polyethylenimine "Max". At 24 h post-transfec tion, the cells were infected with HHV-6B virus (MOI = 0.1) for 1 h, washed with DMEM, then cultured in fresh medium for 3 days. The cells were then fixed in 4% paraformalde hyde for 20 min, washed with PBS, and permeabilized with 0.5% Triton X-100 for 10 min.
After washing with PBS, the cells were incubated overnight at room temperature with a mouse antibody against HHV-6 p41 early antigen. At the following day, the cells were washed with PBS and incubated with Alexa Fluor 488-conjugated goat anti-mouse IgG (H + L) F(ab′)2 at room temperature for 1 h. After washing with PBS, the cells were mounted using ProLong Diamond Antifade Mounting with 4′,6-diamidino-2-phenylin dole (Thermo Fisher Scientific). The images were captured using the BX60 microscope (Olympus Corporation, Tokyo, Japan). For the U65 localization and the codistribution of U65 with H2ACG/H2AC7 or H2A, transfected 293T cells were fixed, permeabilized, and blocked as described above. The cells were incubated with primary antibodies at 4°C overnight, followed by incubation with Alexa Fluor 488 conjugated goat anti-mouse IgG (H + L) F(ab′) 2 and Alexa Fluor 555 conjugated goat anti-mouse IgG (H + L) F(ab′) 2 for 1 h. The images were captured using the BZ-X810 microscope (Keyence, Osaka, Japan) or the LSM780 confocal laser scanning microscope (Carl Zeiss, Jena, Germany).
## Statistical analysis
Statistical significance was evaluated using two-tailed Student's t-test or analysis of variance with Tukey's post hoc test. A P value of less than 0.05 was considered significant.
## References
1. Yamanishi, Shiraki, Kondo et al. (1988) "Identification of human herpesvirus-6 as a causal agent for exanthem subitum" *The Lancet*
2. Wyatt, Balachandran, Frenkel (1990) "Variations in the replication and antigenic properties of human herpesvirus 6 strains" *J Infect Dis*
3. Ablashi, Agut, Alvarez-Lafuente et al. (2014) "Classification of HHV-6A and HHV-6B as distinct viruses" *Arch Virol*
4. Zedan, Aldewik, Elkhider et al. (2022) "Human herpes simplex virus-6 (HHV-6) detection and seroprevalence among Qatari nationals and immigrants residing in Qatar" *IJID Regions*
5. Alvarez-Lafuente, García-Montojo, Las Heras et al. (2006) "Clinical parameters and HHV-6 active replication in relapsing-remitting multiple sclerosis patients" *J Clin Virol*
6. Caselli, Zatelli, Rizzo et al. (2012) "Virologic and immunologic evidence supporting an association between HHV-6 and hashimoto's thyroiditis" *PLoS Pathog*
7. Heming, Conway, Homa (2017) "Herpesvirus Capsid Assembly and DNA Packaging"
8. Kondo, Shimada, Sashihara et al. (2002) "Identification of human herpesvirus 6 latency-associated transcripts" *J Virol*
9. Dominguez, Dambaugh, Stamey et al. (1999) "Human herpesvirus 6B genome sequence: coding content and comparison with human herpesvirus 6A" *J Virol*
10. Aktar, Arii, Tjan et al. (2021) "Human herpesvirus 6A tegument protein U14 induces NF-κB signaling by interacting with p65" *J Virol*
11. Dalskov, Gad, Hartmann (2023) "Viral recognition and the antiviral interferon response" *EMBO J*
12. Mcnab, Mayer-Barber, Sher et al. (2015) "Type I interferons in infectious disease" *Nat Rev Immunol*
13. Ji, Li, Chen et al. (2023) "The crucial regulatory role of type I interferon in inflammatory diseases" *Cell Biosci*
14. Perry, Chen, Zheng et al. (2005) "The host type I interferon response to viral and bacterial infections" *Cell Res*
15. Ablasser, Goldeck, Cavlar et al. (2013) "cGAS produces a 2′-5′-linked cyclic dinucleotide second messenger that activates STING" *Nature*
16. Diner, Burdette, Wilson et al. (2013) "The innate immune DNA sensor cGAS produces a noncanonical cyclic dinucleotide that activates human STING" *Cell Rep*
17. Sun, Wu, Du et al. (2013) "Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway" *Science*
18. Tanaka, Chen (2012) "STING specifies IRF3 phosphorylation by TBK1 in the cytosolic DNA signaling pathway" *Sci Signal*
19. Chen, Sun, Chen (2016) "Regulation and function of the cGAS-STING pathway of cytosolic DNA sensing" *Nat Immunol*
20. Ishikawa, Barber (2008) "STING is an endoplasmic reticulum adaptor that facilitates innate immune signalling" *Nature*
21. Balka, Louis, Saunders et al. (2020) "TBK1 and IKKε act redundantly to mediate STING-induced NF-κB responses in myeloid cells" *Cell Rep*
22. Jaworska, Gravel, Flamand (2010) "Divergent susceptibilities of human herpesvirus 6 variants to type I interferons" *Proc Natl Acad Sci*
23. Lubyova, Kellum, Frisancho et al. (2004) "Kaposi's sarcomaassociated herpesvirus-encoded vIRF-3 stimulates the transcriptional activity of cellular IRF-3 and IRF-7" *J Biol Chem*
24. Bodda, Reinert, Fruhwürth et al. (2020) "HSV1 VP1-2 deubiquitinates STING to block type I interferon expression and promote brain infection" *J Exp Med*
25. Vilmen, Glon, Siracusano et al. (2021) "BHRF1, a BCL2 viral homolog, disturbs mitochondrial dynamics and stimulates mitophagy to dampen type I IFN induction" *Autophagy*
26. Fruhwürth, Reinert, Öberg et al. (2023) "TREM2 is down-regulated by HSV1 in microglia and involved in antiviral defense in the brain" *Sci Adv*
27. Jaworska, Gravel, Fink et al. (2007) "Inhibition of transcription of the beta interferon gene by the human herpesvirus 6 immediate-early 1 protein" *J Virol*
28. Liu, Rao, Tian et al. (2019) "Modulation of innate immune signaling pathways by herpesviruses" *Viruses*
29. Li, Goodrich, Yang (1997) "Interferon-gamma (IFN-gamma) regulates production of IL-10 and IL-12 in human herpesvirus-6 (HHV-6)infected monocyte/macrophage lineage" *Clin Exp Immunol*
30. Iampietro, Morissette, Gravel et al. (2014) "Inhibition of interleukin-2 gene expression by human herpesvirus 6B U54 tegument protein" *J Virol*
31. Tweedy, Spyrou, Pearson et al. (2016) "Complete genome sequence of germline chromosomally integrated human herpesvirus 6A and analyses integration sites define a new human endogenous virus with potential to reactivate as an emerging infection" *Viruses*
32. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
33. Lusso (2006) "HHV-6 and the immune system: mechanisms of immunomodulation and viral escape" *J Clin Virol*
34. Bortolotti, Gentili, Caselli et al. (2020) "DNA sensors' signaling in NK cells during HHV-6A, HHV-6B and HHV-7 infection" *Front Microbiol*
35. Jiang, Tang, Guo et al. (2021) "Human herpesvirus 6B U26 inhibits the activation of the RLR/MAVS signaling pathway" *mBio*
36. Hirai, Amaliin, Huang et al. (2025) "HHV-6B ribonucleotide reductase sequesters NF-κB subunit p65 to inhibit innate immune responses"
37. Mun, Eui, Kyu et al. (2021) "Human cytomegalovirus UL48 deubiquiti nase primarily targets innermost tegument proteins pp150 and itself to regulate their stability and protects virions from inclusion of ubiquitin conjugates" *J Virol*
38. Zou, Huang, Yang et al. (2020) "Human cytomegalovirus protein UL94 targets MITA to evade the antiviral immune response" *J Virol*
39. Fu, Su, Gao et al. (2017) "Human cytomegalovirus tegument protein UL82 inhibits STING-mediated signaling to evade antiviral immunity" *Cell Host & Microbe*
40. Landolfo, Andrea, Dell'oste et al. (2016) "Intrinsic host restriction factors of human cytomegalovirus replication and mecha nisms of viral escape" *WJV*
41. Tolomeo, Cavalli, Cascio (2022) "STAT1 and its crucial role in the control of viral infections" *IJMS*
42. Deng, Xu, Li et al. (2022) "Insights on the cGAS-STING signaling pathway during herpesvirus infections" *Front Immunol*
43. Au-Yeung, Horvath (2018) "Histone H2A.Z suppression of interferon-stimulated transcription and antiviral immunity is modulated by GCN5 and BRD2"
44. Wu, Fang, Zhang et al. (2021) "Histone H2A nuclear/ cytoplasmic trafficking is essential for negative regulation of antiviral immune response and lysosomal degradation of TBK1 and IRF3" *Front Immunol*
45. Kobiyama, Takeshita, Jounai et al. (2010) "Extrachromosomal histone H2B mediates innate antiviral immune responses induced by intracellular double-stranded DNA" *J Virol*
46. Fonseca, Thillainadesan, Yousef et al. (2012) "Adenovirus evasion of interferon-mediated innate immunity by direct antagonism of a cellular histone posttranslational modification" *Cell Host & Microbe*
47. Iqbal, Ansari, Kumar et al. (2016) "Histone H2B-IFI16 recognition of nuclear herpesviral genome induces cytoplasmic interferon-β responses" *PLOS Pathog*
48. Singh, Padariya, Faktor et al. (2022) "Identification of novel interferon responsive protein partners of human leukocyte antigen A (HLA-A) using cross-linking mass spectrom etry (CLMS) approach" *Sci Rep*
49. Parekh, Maniatis (1999) "Virus infection leads to localized hyperacety lation of histones H3 and H4 at the IFN-β promoter" *Mol Cell*
50. Xue, Feng, Chen et al. (2022) "Protein Acetylation Going Viral: Implications in Antiviral Immunity and Viral Infection" *IJMS*
51. Baugé, Bazille, Girard et al. (2014) "Histone methylases as novel drug targets: developing inhibitors of EZH2" *Future Med Chem*
52. Van Attikum, Gasser (2009) "Crosstalk between histone modifications during the DNA damage response" *Trends Cell Biol*
53. Skopelja-Gardner, Elkon (2022) "Role of the cGAS-STING pathway in systemic and organ-specific diseases" *Nat Rev Nephrol*
54. Wu, Li, Shao et al. (2015) "Inhibition of cGAS DNA sensing by a herpesvirus virion protein" *Cell Host & Microbe*
55. Yu, Tian, Deng (2020) "PPM1G restricts innate immune signaling mediated by STING and MAVS and is hijacked by KSHV for immune evasion" *Sci Adv*
56. Jangra, Bharti, Lui et al. (2021) "Suppression of JAK-STAT signaling by Epstein-Barr virus tegument protein BGLF2 through recruitment of SHP1 Phosphatase and promotion of STAT2 degradation" *J Virol*
57. Wing, Lee, Huang (1996) "The human cytomegalovirus UL94 open reading frame encodes a conserved herpesvirus capsid/tegumentassociated virion protein that is expressed with true late kinetics" *J Virol*
58. Sato, Dacher, Kurumizaka (2022) "Nucleosome structures built from highly divergent histones: parasites and giant DNA viruses" *Epigenomes*
59. Benveniste, Sonntag, Sanguinetti et al. (2014) "Transcription factor binding predicts histone modifications in human cell lines" *Proc Natl Acad Sci*
60. Verdone, Agricola, Caserta et al. (2006) "Histone acetylation in gene regulation" *Briefings in Functional Genomics and Proteomics*
61. (2025) *Full-Length Text Journal of Virology*
62. Arents, Moudrianakis (1993) "Topography of the histone octamer surface: repeating structural motifs utilized in the docking of nucleoso mal DNA" *Proc Natl Acad Sci*
63. Davey, Sargent, Luger et al. (2002) "Solvent mediated interactions in the structure of the nucleosome core particle at 1.9Å resolution † †We dedicate this paper to the memory of Max Perutz who was particularly inspirational and supportive to T.J.R. in the early stages of this study" *J Mol Biol*
64. Zhou, Gaullier, Luger (2019) "Nucleosome structure and dynamics are coming of age" *Nat Struct Mol Biol*
65. Kornberg, Lorch (2020) "Primary role of the nucleosome" *Mol Cell*
66. Kalashnikova, Rogge, Hansen (2016) "Linker histone H1 and protein-protein interactions" *Biochimica et Biophysica Acta (BBA) -Gene Regulatory Mechanisms*
67. Cutter, Hayes (2015) "A brief review of nucleosome structure" *FEBS Lett*
68. Oberdoerffer, Miller (2023) "Histone H2A variants: diversifying chromatin to ensure genome integrity" *Semin Cell Dev Biol*
69. Cao, Ruan, Yang et al. (2024) "Estrogen receptor α-mediated signaling inhibits type I interferon response to promote breast carcinogenesis" *J Mol Cell Biol*
70. Bannister, Kouzarides (2011) "Regulation of chromatin by histone modifications" *Cell Res*
71. Xiao, Wang, Shan et al. (2023) "The interplay between viral molecular mimicry and host chromatin dynamics" *Nucleus*
72. Marazzi, Ho, Kim et al. (2012) "Suppression of the antiviral response by an influenza histone mimic" *Nature*
73. Tarakhovsky, Prinjha (2018) "Drawing on disorder: how viruses use histone mimicry to their advantage" *J Exp Med*
74. Cliffe, Knipe (2008) "Herpes simplex virus ICP0 promotes both histone removal and acetylation on viral DNA during lytic infection" *J Virol*
75. Coleman, Connor, Cheng et al. (2008) "Histone modifications associated with herpes simplex virus type 1 genomes during quiescence and following ICP0-mediated derepression" *J Gen Virol*
76. Josué, Chandra, Italo et al. (2025) "Histone variant H2A.Z cooperates with EBNA1 to maintain Epstein-Barr virus latent epigenome"
77. Tang, Mori (2015) "Determinants of Human CD134 Essential for Entry of Human Herpesvirus 6B" *J Virol*
78. Nagamata, Nagasaka, Kawabata et al. (2018) "Human CD134 (OX40) expressed on T cells plays a key role for human herpesvirus 6B replication after allogeneic hematopoietic stem cell transplantation" *J Clin Virol*
79. Ogawa, Fujikura, Namba et al. (2022) "Nectin-2 acts as a viral entry mediated molecule that binds to human herpesvirus 6B glycoprotein B" *Viruses*
80. Yoneyama, Suhara, Fukuhara et al. (1996) "Autocrine amplification of type I interferon gene expression mediated by interferon stimulated gene factor 3 (ISGF3)" *J Biochem*
81. Honda, Takemoto, Ueda (2019) "Identification of a retroelementcontaining human transcript induced in the nucleus by vaccination" *IJMS*
82. Nakayama, Ueno, Ueda et al. (2019) "Latent infection with Kaposi's sarcoma-associated herpesvirus enhances retrotransposition of long interspersed element-1" *Oncogene*
83. Nishikawa, Nakayama, Obika et al. (2018) "Inhibition of LINE-1 retrotransposition by capsaicin" *IJMS*
84. Hayashi, Horie, Daito et al. (2009) "Heat shock cognate protein 70 controls Borna disease virus replication via interaction with the viral non-structural protein X" *Microbes Infect*
85. Kojima, Honda, Matsumoto et al. (2014) "Heat stress is a potent stimulus for enhancing rescue efficiency of recombinant Borna disease virus" *Microbiol Immunol*
86. Wada, Kubota, Ito et al. (2007) "Simultane ous quantification of Epstein-Barr Virus, cytomegalovirus, and human herpesvirus 6 DNA in samples from transplant recipients by multiplex real-time PCR assay" *J Clin Microbiol*
87. Honda, Yamamoto, Daito et al. (2016) "Long-term expression of miRNA for RNA interference using a novel vector system based on a negative-strand RNA virus" *Sci Rep*
88. Kubala, Kovtun, Alexandrov et al. (2010) "Structural and thermodynamic analysis of the GFP:GFP-nanobody complex" *Protein Sci*
89. Katoh, Nozaki, Hartanto et al. (2015) "Architec tures of multisubunit complexes revealed by a visible immunoprecipita tion assay using fluorescent fusion proteins" *J Cell Sci*
90. Rozewicki, Li, Amada et al. (2019) "MAFFT-DASH: integrated protein sequence and structural alignment" *Nucleic Acids Res*
91. Wang, Liu, Yuchi et al. (2020) "MusiteDeep: a deep-learning based webserver for protein post-translational modification site prediction and visualization" *Nucleic Acids Res*
92. Tsukazaki, Yoshida, Namba et al. (1998) "Development of a dot blot neutralizing assay for HHV-6 and HHV-7 using specific monoclonal antibodies" *J Virol Methods*
93. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12811890&blobtype=pdf | # Resistance Analysis of Low-Level Virologic Rebound During HIV-1 Treatment With Lenacapavir and Broadly Neutralizing Antibodies Teropavimab and Zinlirvimab
Lisa Selzer, Sally Demirdjian, Brie Falkard, Jiani Li, Ross Martin, Sean Collins, Joseph Eron, Laurie Vanderveen, Christian Callebaut
## Abstract
Background. High rates of virologic suppression were observed in the Phase 1b study (NCT04811040) investigating lenacapavir and two broadly neutralizing antibodies (bNAb), teropavimab (30 mg/kg) and zinlirvimab (10 or 30 mg/kg), in virologically suppressed people with HIV-1 susceptible (IC 90 ≤ 2 μg/mL) to both (primary cohort, n = 20) or either (pilot cohort, n = 10) bNAb. We describe resistance analyses through Week (W) 26.Methods. Post-baseline resistance analyses were conducted at virologic failure, and exploratory resistance analyses performed for participants with virologic rebound. Low copy number genotyping methods for capsid and a 1 kb stretch of gp120 from rebound virus were developed, and phenotypic susceptibility assessed.Results. Virologic failure was observed in 1/30 participants. This primary cohort participant had HIV RNA 155 copies/mL at W16 and developed Q67H in capsid (lenacapavir fold-change 4.7), without resistance to bNAbs; the participant resuppressed on oral antiretrovirals. Two pilot cohort participants, experienced virologic rebound at W26 (55 and 72 copies/mL) and restarted oral antiretrovirals. In exploratory analyses, neither had emergent lenacapavir resistance or altered bNAb susceptibility.Conclusions. Lenacapavir, teropavimab, and zinlirvimab maintained a high rate of virologic suppression through W26, with rare emergent lenacapavir resistance and no bNAb resistance, supporting further Phase 2 evaluation.
Antiretroviral therapy (ART) remains a lifelong requirement for people with human immunodeficiency virus 1 (HIV-1) (PWH), for whom there are effective treatments available but no cure [1]. Adhering to daily oral ART can be challenging due to barriers including pill fatigue, forgetfulness, side effects, stigma, and mental health challenges [2,3]. Long-acting therapies are being developed to address these challenges [4]. One such long-acting agent is lenacapavir (LEN), the first-in-class HIV capsid (CA) inhibitor targeting early and late stages of the viral life cycle [5]. LEN is administered by twice-yearly subcutaneous (SC) injections, exhibits potent antiviral activity, and has demonstrated clinical efficacy combined with oral ART [5][6][7]. LEN is approved, in combination with other antiretrovirals, for heavily treatment-experienced adults with multidrug-resistant HIV and for use as pre-exposure prophylaxis [8,9].
Synchronous long-acting therapy for PWH requires partner agents with similar pharmacokinetic properties to LEN. Broadly neutralizing antibodies (bNAbs) targeting HIV envelope (ENV) neutralize a wide variety of HIV [10,11]. Two bNAbs, 3BNC117 targeting the CD4 binding site and 10-1074 targeting the V3 glycan on the HIV ENV protein gp120, demonstrated a direct, rapid antiretroviral effect in singleagent, dose-escalation studies [12][13][14]. Teropavimab (TAB; formerly 3BNC117-LS or GS-5423) and zinlirvimab (ZAB; formerly 10-1074-LS or GS-2872) have been modified to extend their half-lives, allowing for twice-yearly intravenous dosing [15]. Over 90% of subtype B HIV viruses are highly susceptible to TAB or ZAB, while >50% are highly susceptible to both (90% inhibitory concentration [IC 90 ] of ≤2 μg/mL by PhenoSense® monoclonal antibody (mAb) assay) [16].
LEN, TAB, and ZAB have been studied in a proof-of-concept Phase 1b study (NCT04811040) as the first prospective, complete, twice-yearly combination treatment for HIV. Participants were enrolled across a primary and a pilot cohort of participants with HIV highly susceptible to both or one of the bNAbs, respectively [17,18]. The interventions were well tolerated, with SC LEN-related injection site reactions being the most reported adverse events [17,18]. Virologic suppression (VS; HIV RNA <50 copies/mL) per US Food and Drug Administration Snapshot Algorithm was maintained through Week 26 in 26/30 participants; 3/30 participants had low-level virologic rebound (VR) (HIV RNA ≥50 to <1000 copies mL), and 1/30 participants had no data in the Week 26 window [17,18].
Here, we describe the Phase 1b resistance analysis through Week 26 in participants treated with a single dose of LEN, TAB, and ZAB.
## METHODS
## Study Overview
This completed, randomized Phase 1b study (NCT04811040), conducted at 13 US sites, comprised a primary cohort of participants with HIV highly susceptible to both TAB and ZAB [17], and a pilot cohort of participants with HIV highly susceptible to either TAB or ZAB (Supplementary Figure 1) [18]. The methodology and primary outcome were communicated previously [17,18].
## Participant Selection and Dosing
Eligible participants were aged 18-65 years with HIV virologically suppressed (HIV RNA <50 copies/mL) for ≥18 months, taking oral ART for ≥2 years, and with a nadir CD4 count of ≥350 cells/μL and a CD4 count of ≥500 cells/μL at study entry. Participants had proviral phenotypic susceptibility to TAB and/ or ZAB, defined as IC 90 ≤ 2 μg/mL for each antibody using the PhenoSense® mAb Assay (Monogram Biosciences, South San Francisco, CA, USA). Participants stopped oral ART and were randomly assigned (1:1) on Day 1 to receive a single dose of SC LEN 927 mg (plus LEN 600 mg oral loading on Days 1 and 2), intravenous TAB 30 mg/kg, and either intravenous ZAB 10 mg/kg or 30 mg/kg. The ZAB dose was masked for participants and investigators.
## Baseline Resistance Analyses
## HIV Proviral CA and ENV Genotyping
CA and ENV genotyping were performed at screening using the deep-sequencing HIV gag and env assays with variant detection limits of 2% and 1%, respectively (Seq-IT GmbH & Co. KG, Kaiserslautern, Germany) (Figure 1A). Briefly, DNA was extracted from purified peripheral blood mononuclear cells (PBMC), and then CA or ENV was amplified in bulk and sequenced using the Illumina MiSeq platform. Sequencing data were processed using a Gilead Sciences, Inc. developed analysis pipeline [19]. HIV-1 subtype was determined based on nucleotide identity of the consensus env gene against curated subtype reference sequences from GenBank® [20,21].
## HIV Proviral CA and ENV Phenotyping
Phenotypic HIV testing at screening was conducted on DNA extracted from PBMCs collected at screening (Figure 1A). HIV proviral CA was phenotyped with the PhenoSense® Gag-Pro assay (research use only; Monogram Biosciences) using the proviral consensus sequence of CA determined from deep sequencing. Baseline phenotypic susceptibility to TAB and ZAB was analyzed in real time by the PhenoSense® mAb DNA assay [22,23], which evaluates the neutralization susceptibility of pseudovirions generated from PBMC-derived HIV ENV DNA amplified in bulk, to TAB or ZAB. The PhenoSense® mAb assay is Clinical Laboratory Improvement Amendments validated, and appropriate for clinical decision-making.
## Post-Baseline Resistance Analyses
Resistance analyses were performed on participants in the perprotocol defined resistance analysis population, which included participants who received at least 1 dose of the complete study drug regimen and met protocol-defined criteria for virologic failure (VF; HIV RNA ≥200 copies/mL on two consecutive visits, or HIV RNA ≥200 copies/mL at study discontinuation or Week 26). Exploratory resistance analyses were performed for participants meeting criteria for virologic rebound (VR; HIV RNA ≥50 copies/mL at any post-Day 1 visit, subsequently confirmed at the following scheduled or unscheduled visit, or HIV RNA ≥50 copies/mL at study discontinuation or Week 26) but did not meet requirements for per protocol resistance analysis population (Figure 1B,C).
## HIV CA and ENV Genotyping From Plasma
Genotypic resistance testing of HIV CA and ENV was conducted at Monogram Biosciences using next-generation sequencing with reporting of variants present at >10% and >2% frequencies, respectively (GeneSeq® Gag-Pro and GenoSure® Env assays). In case of assay failure (AF), and for exploratory analyses, genotyping of gag and env was conducted for VF and VR participants using a novel genotyping assay developed at Gilead Sciences, Inc., amenable to low virus copy number samples (Figure 1C). For genotyping of HIV CA and ENV, RNA was isolated from 1.5 mL of plasma using the QIamp® Viral RNA Mini Kit (Qiagen, Hilden, Germany). Total RNA was divided into two parts and reverse transcribed with the SuperScript™ IV First-Strand Synthesis System (Invitrogen, Waltham, MA, USA) using gene-specific reverse primers for either CA or ENV (Supplementary Table 1). A partial 386 base pair (bp) stretch of the CA gene (HXB2 gag nucleotides 384-782; CA amino acids 1-128. GenBank: K03455.1) encompassing all known LEN resistance-associated residues, or the full CA gene (693 bp; HXB2 gag nucleotides 384-1193; CA amino acids 1-231. GenBank: K03455.1) were amplified via nested polymerase chain reaction using Platinum® Taq DNA High Fidelity Polymerase (Invitrogen, Waltham, MA, USA) (Supplementary Table 1). For all participants, a 994 bp stretch of ENV (HXB2 gp120 nucleotides 100-1093; ENV amino acids 34-371. GenBank: K03455.1) was amplified using the same method (Supplementary Table 1). HIV CA and ENV amplicons were deep sequenced using the Illumina platform (Novogene, Sacramento, CA, USA).
## HIV CA and ENV Phenotyping
Phenotypic resistance testing was conducted at Monogram Biosciences using the PhenoSense® Gag-Pro and PhenoSense® mAb assays to assess responses to LEN and the bNAbs, respectively. In the event of AF, and for exploratory analyses, phenotyping of CA and ENV was conducted using a novel assay developed at Gilead Sciences, Inc. (Figure 1C). To determine phenotypic susceptibility, as described above, a partial capsid gene (386 bp) or the full capsid gene (693 bp) were sequenced from rebound viruses, synthesized, and cloned into majority sequences for gag determined at baseline. In addition, a 1 kb stretch of gp120 was sequenced from rebound viruses, synthesized, and cloned into majority sequences for env determined at baseline. Synthesis for both genes was based on the majority nucleotide call derived from deep sequencing. The chimeric gag and env genes were submitted to Monogram Biosciences for phenotyping using the PhenoSense® Gag-Pro and PhenoSense® mAb assays, respectively, with each gene assessed independently.
## Genotypic Assessment of LEN and bNAb Susceptibility
For resistance analyses, CA resistance substitutions were defined according to the definitions of the International AIDS Society-USA (IAS-USA), as well as additional CA resistance substitutions reported in the literature [24][25][26]. For prediction of susceptibility to TAB and ZAB from genotypic data, baseline and post-baseline ENV sequences were analyzed for the presence of ENV amino acid signatures (Supplementary Table 2). Signatures were previously developed by combining neutralization data with virus sequence information to identify HIV ENV amino acid positions associated with increased susceptibility to TAB and ZAB [16]. This identification was performed by comparing the frequency of amino acids at each ENV residue in susceptible and non-susceptible viruses using genotypic-phenotypic datasets downloaded from the CATNAP database [27,28] and internal Gilead datasets (susceptible, IC 50 < 1 µg/mL; nonsusceptible, IC 50 ≥ 1 µg/mL) by Fisher's exact test [19].
## Clonal Phenotypic Analysis at Baseline for Participant 1 With VF
Twenty-four single proviral env genes were amplified from the baseline screening sample and cloned into individual expression vectors. TAB or ZAB were titrated against the generated clonal pseudoviruses to determine concentrations required for virus inhibition [17].
## Sensitivity to Additional bNAbs for Participant 1 With VF
Phenotypic resistance to five additional bNAbs for the baseline and post-baseline virus was assessed using the PhenoSense® The five additional bNAbs and their targets were N49P7 (CD4 binding site) [29], 1-18 (CD4 binding site) [30], PGT121 (HIV gp120 V3 base) [31], PG16 (V2/glycan apex region on the HIV ENV glycoprotein site) [32], and PGDM1400 (V2 apex region) [33].
## RESULTS
Details of the participant population have been reported previously [17,18]. Baseline phenotypic data for HIV ENV confirmed eligibility of the 30 enrolled participants, and no participant had previous LEN exposure. All participants with available data (24/30 participants) had subtype B virus, except for one participant in the pilot cohort (30 mg/kg ZAB group), with subtype AG virus. During the study, 1/30 (3.3%) participants met criteria for VF and was included in the per-protocol resistance analysis population. Resistance testing was also performed for 2/30 (6.7%) participants who did not meet the criteria for resistance testing, but who experienced VR at Week 26 with HIV RNA <200 copies/mL.
## LEN, TAB, and ZAB Resistance Analyses
Descriptions of LEN, TAB, and ZAB resistance outcomes for the three participants are summarized in 2A).
No genotypic or phenotypic resistance to TAB or ZAB was observed. Participant 1 restarted oral ART at Week 18 (rilpivirine/ emtricitabine/tenofovir alafenamide) and resuppressed by Week 20 (HIV RNA ≤50 copies/mL), with suppression maintained through subsequent study visits.
The other two participants (Participants 2 and 3) were from the pilot cohort 10 mg/kg ZAB group [18]. Participant 2 was highly susceptible to ZAB (IC 90 0.12 μg/mL) but not TAB (IC 90 5.02 μg/mL) at baseline and had VR at Week 26. Participant 3 was highly susceptible to TAB (IC 90 0.43 μg/ mL) but not ZAB (IC 90 > 50 μg/mL) at baseline and experienced VR following acute COVID-19 during Weeks 12-14. Similarly to Participant 1, planned resistance testing could not be performed for Participants 2 and 3 on samples with low copy HIV RNA of 72 and 55 copies/mL, respectively. Novel resistance testing methods detected no treatment emergent resistance to LEN after analysis of the full CA gene, and no resistance to TAB or ZAB for either participant (Figure 2B andC). Per protocol, Participant 2 resumed oral ART at Week 26 (dolutegravir/lamivudine) and was suppressed at subsequent study visits (Figure 2B). Participant 3 resumed oral ART (dolutegravir/lamivudine) at Week 26 but continued to experience low-level viremia through Week 54 (Figure 2C).
ENV genotypic data for the three participants were evaluated to predict TAB and ZAB susceptibility based on the prevalence of ENV susceptibility signatures. Supplementary Table 3 provides a summary of the prevalence of baseline and post-baseline ENV susceptibility signatures, while a detailed amino acid alignment between baseline and post-baseline ENV sequences is shown in Supplementary Figure 2 and Supplementary Table 4. Minor differences between baseline and post-baseline genotypic signatures were observed for all three participants (Supplementary Table 3). For example, for Participant 2, who was only susceptible to ZAB, residue D325 was present at baseline but not Week 26. Similarly, for Participant 3, who was only susceptible to TAB, residue Y318 was present at baseline but not Week 26. Despite these changes in the susceptibility signatures, phenotypic susceptibility to TAB and ZAB were unaffected for Participant 2 and 3.
## Participant 1: Clonal Genotypic and Phenotypic Analyses
For Participant 1, resistance analyses at the VF visit showed development of resistance to LEN (Q67H) but no change in susceptibility to either bNAb by phenotypic analysis of bulk amplified ENV. To further investigate potential pre-existing resistance to TAB and ZAB at baseline, phenotypic susceptibility to bNAbs was reassessed for clonal ENV sequences isolated from the screening visit sample. No phenotypic resistance to either bNAb was detected at baseline. Of 24 screened ENV clones, 23 had TAB IC 90 ≤ 2 μg/mL and one (clone 17) had TAB IC 90 of 3.96 μg/mL; all clones had ZAB IC 90 ≤ 2 μg/mL. (Figure 3). For TAB, the bulk screening IC 90 and clonal analysis geometric mean IC 90 were 0.22 μg/mL and 0.36 (range, 0.20-3.96) μg/mL, respectively. For ZAB, the bulk screening IC 90 and clonal analysis geometric mean IC 90 were 0.26 μg/mL and 0.13 (range, 0.06-0.29) μg/mL, respectively.
Clonal sequence analysis showed that most TAB and ZAB signature residues were present in all clones, except for A281V in A, For Participant 1, Q67H was detected as a mixed population at Week 16, with 84% H and 16% Q, and as 100% Q67H at the Week 16 retest visit (Week 16 + 15 d), based on deep sequencing data. Participants 1 and 2 resumed the same antiretroviral regimen that was in place before study entry, RPV/FTC/TAF and DTG/ 3TC, respectively. Participant 3 entered the study on ABC/DTG/3TC and resumed with DTG/3TC. Dotted line denotes the 50 copies/mL limit which defined virologic rebound in this study. Abbreviations: ABC/DTG/3TC, abacavir/dolutegravir/lamivudine; CA, capsid; COVID-19, Coronavirus-2019; DTG/3TC, dolutegravir/lamivudine; ENV, envelope; FC, fold change; IC 90 , 90% inhibitory concentration, LEN, lenacapavir; RAM, resistance associated mutation; RPV/FTC/TAF, rilpivirine/emtricitabine/ tenofovir alafenamide; TAB, teropavimab; ZAB, zinlirvimab. 5). Phylogenetic analysis included full-length and 994 bp ENV regions for clonal sequences, along with Week 16 and Week 16 retest ENV sequences from Participant 1 (Supplementary Figure 3). The Week 16 ENV sequence was identical to that of clone 18 (TAB IC 90 :0.72 μg/mL, ZAB IC 90 :0.097 μg/mL), while the Week 16 retest sequence was closely related.
## Participant 1: Susceptibility of Provirus and Rebound Virus to Additional bNAbs
To further investigate changes in virus susceptibility between baseline and the time of VF for Participant 1, provirus and rebound virus were phenotyped using five additional bNAbs targeting the CD4-binding site or the V3 or V2 Apex regions of ENV. Only minor changes in susceptibility between baseline and Week 16 were observed for bNAbs N49P7 (CD4), 1-18 (CD4), PGT121 (V3), PG16 (V2 Apex), and PGDM1400 (V2 Apex). IC 90 values for most bNAbs showed <3-fold change from baseline (Figure 4); this is consistent with the minimal changes in susceptibility observed for TAB and ZAB between baseline and Week 16.
## DISCUSSION
In this Phase 1b study of LEN, TAB, and ZAB as the first twice-yearly HIV treatment, high rates of VS were maintained through 26 weeks, including among pilot cohort participants who were highly susceptible to only one bNAb [17,18]. Emergent resistance was rare; of the three participants with VF or VR, all of whom received the lower dose of ZAB (10 mg/kg), treatment-emergent LEN resistance (Q67H) was only detected in one participant, with none showing treatment-emergent resistance to TAB or ZAB. The development of a novel low copy number genotyping assay allowed for genotypic and phenotypic resistance analyses in these three participants with low-level viremia (HIV RNA ≥50 to <1000 copies/mL), for whom attempts to sequence gag or env using planned assays resulted in AF or could not be performed due to low viral loads.
Participant 1 (primary cohort, low-dose ZAB group) developed the Q67H resistance-associated mutation (RAM) to LEN despite remaining susceptible to both bNAbs and having no pre-existing resistance mutations to LEN at baseline. Although the Q67H mutation has been observed in previous LEN treatment studies [6,[34][35][36], it is rarely seen without LEN exposure [37]. Naturally occurring LEN binding site variants in CA remain fully susceptible to LEN, indicating minimal impact of natural viral diversity on LEN efficacy [38]. LEN RAMs were identified in 19/258 PWH across three other clinical studies, including M66I, Q67H/K/N, K70H/N/R/S, N74D/ H/K, A105S/T, and T107A/C/N/S [26,34,39]. Resistance typically emerged in highly treatment-experienced participants due to functional monotherapy from insufficient active antiretrovirals (ARVs) in, or inadequate adherence to, the optimized background regimen [5,26,35].
Resistance to the parent antibodies of TAB (3BNC117) and ZAB (10-1074) has been observed in the setting of monotherapy [12-14, 40, 41]. For 3BNC117, mutations in R456 and K282 were common in rebound viruses in participants receiving monotherapy [14], while resistance to 10-1074, was associated with mutations in the V3 loop binding site, ie, the 324 GDIR 327 motif and residues H330, N332, and S334 [13,40]. Bioinformatic analyses of phenotypic neutralization and genotypic ENV data have identified ENV signatures important for viral neutralization by 3BNC117 and 10-1074 [19,42]. For Participant 1, no development of phenotypic resistance or changes in these previously characterized resistance regions were detected. Clonal analysis of proviral ENV genes at baseline revealed no pre-existing geno-or phenotypic resistance to TAB or ZAB, consistent with the bulk ENV phenotyping performed at baseline. In addition, phylogenetic analysis revealed a close relationship between clone 18 (TAB and ZAB IC 90 < 2 μg/mL) and the Week 16 and Week 16 Retest ENV sequences, indicating clone 18 may have emerged during viral rebound. The emergence of resistance in Participant 1 is unlikely due to insufficient exposure to LEN, TAB, or ZAB, as the pharmacokinetic data and anti-drug antibodies were consistent with those of other study participants [17] and showed no impact on TAB or ZAB exposure or neutralization [17]. However, tissue distribution of LEN, TAB, and ZAB may have influenced viral rebound dynamics in this participant. Rebound could occur due to reactivation of latent viruses within tissue compartments, referred to as "sanctuary sites"; these sites may exhibit reduced penetration of ARVs, a highly debated topic in the field of HIV research [43,44]. Interestingly, in clinical studies of cabotegravir plus the CD4-binding bNAb VRC07-523LS for maintenance in PWH, viral rebound in the absence of VRC07-523LS resistance was observed [45], suggesting that the distribution of long-acting ARVs in tissue compartments warrants further investigation.
No LEN resistance was detected in Participants 2 and 3 from the pilot cohort low-dose ZAB group. Similarly to Participant 1, no emergent phenotypic resistance to bNAbs or changes in known resistance-associated amino acid regions were observed. The unchanged phenotypic susceptibility to TAB and ZAB for Participants 2 and 3, despite observed changes in ENV susceptibility signatures, may be due to the signatures prioritizing specificity over sensitivity, which can lead to false negatives. Since LEN, TAB, and ZAB exposures for both participants were consistent with other study participants, the detection of low-level viremia (HIV RNA copies ≥50 copies/mL) in Participants 2 and 3 may be associated with a transient increase in detectable viral RNA levels due to viral activation or clonal expansion of the latent viral reservoir, rather than onward viral replication with development of resistance to the study drugs. This is particularly plausible for the participant who contracted COVID-19, which has been associated with episodes of low-level viremia in PWH [46].
These results need to be considered in the context of several study limitations. Phenotypic analyses at the time of viral rebound focused on the predominant viral variants. However, since viral diversity is typically low during initial rebound, these predominant variants likely provide a reliable representation of the overall viral population at this stage. For Participant 1, a portion of CA amino acids 1-128 was sequenced, although this includes all known LEN-associated resistance mutations. Additionally, only part of the gp120 ENV gene was amplified and sequenced, excluding the CD4 binding site; however, all previously determined susceptibility signature regions for 3BNC117 and 10-1074 were covered by the current sequence analysis [19]. Similarly, the V2 Apex region of ENV associated with binding to bNAbs PG16 and PGDM1400, tested for Participant 1, were covered by this analysis.
Overall, this Phase 1b study showed high rates of VS through 26 weeks of treatment with the triple therapy regimen, with low-level VR in three participants, all of whom were on the lower ZAB dose. Thus, LEN, TAB, and ZAB hold promise as a complete twice-yearly combination regimen for PWH. Evaluation of the investigational regimen with the higher ZAB dose is ongoing in a Phase 2 study in participants highly susceptible to both bNAbs (NCT05729568). Ongoing and future studies will evaluate how well baseline bNAb sensitivity translates into clinical outcomes.
## References
1. Buell, Chung, Chaudhry (2016) "Lifelong antiretroviral therapy or HIV cure: the benefits for the individual patient" *AIDS Care*
2. Shubber, Mills, Nachega (2016) "Patient-reported barriers to adherence to antiretroviral therapy: a systematic review and meta-analysis" *PLoS Med*
3. Claborn, Meier, Miller (2015) "A systematic review of treatment fatigue among HIV-infected patients prescribed antiretroviral therapy" *Psychol Health Med*
4. Cooper, Rosenblatt, Gulick (2022) "Barriers to uptake of long-acting antiretroviral products for treatment and prevention of human immunodeficiency virus (HIV) in high-income countries" *Clin Infect Dis*
5. Segal-Maurer, Dejesus, Stellbrink (2022) "Capsid inhibition with lenacapavir in multidrug-resistant HIV-1 infection" *N Engl J Med*
6. Link, Rhee, Tse (2020) "Clinical targeting of HIV capsid protein with a long-acting small molecule" *Nature*
7. Gupta, Berhe, Crofoot (2023) "Lenacapavir administered every 26 weeks or daily in combination with oral daily antiretroviral therapy for initial treatment of HIV: a randomised, open-label, active-controlled, phase 2 trial" *Lancet HIV*
8. Sunlenca® (2025) "Prescribing Information"
9. Yeztugo® (2025) "(lenacapavir) tablets, for oral use. YEZTUGO® (lenacapavir) injection, for subcutaneous use"
10. Hraber, Seaman, Bailer (2014) "Prevalence of broadly neutralizing antibody responses during chronic HIV-1 infection" *AIDS*
11. Liao, Lynch, Zhou (2013) "Co-evolution of a broadly neutralizing HIV-1 antibody and founder virus" *Nature*
12. Caskey, Klein, Lorenzi (2015) "Viraemia suppressed in HIV-1-infected humans by broadly neutralizing antibody 3BNC117" *Nature*
13. Caskey, Schoofs, Gruell (2017) "Antibody 10-1074 suppresses viremia in HIV-1-infected individuals" *Nat Med*
14. Scheid, Horwitz, Bar-On (2016) "HIV-1 Antibody 3BNC117 suppresses viral rebound in humans during treatment interruption" *Nature*
15. Gautam, Nishimura, Gaughan (2018) "A single injection of crystallizable fragment domain-modified antibodies elicits durable protection from SHIV infection" *Nat Med*
16. Selzer, Vanderveen, Parvangada (2025) "Susceptibility screening of HIV-1 viruses to broadly neutralizing antibodies, teropavimab and zinlirvimab, in people with HIV-1 suppressed by antiretroviral therapy" *J Acquir Immune Defic Syndr*
17. Eron, Little, Crofoot (2024) "Safety of teropavimab and zinlirvimab with lenacapavir once every 6 months for HIV treatment: a phase 1b, randomised, proof-ofconcept study" *Lancet HIV*
18. Eron, Cook, Mehrotra (2025) "Lenacapavir plus 2 broadly neutralizing antibodies, teropavimab and zinlirvimab, for people with HIV-1 highly susceptible to either teropavimab or zinlirvimab" *J Infect Dis*
19. Moldt, Parvangada, Martin (2021) "Evaluation of broadly neutralizing antibody sensitivity by genotyping and phenotyping for qualifying participants to HIV clinical trials" *J Acquir Immune Defic Syndr*
20. Vanderveen, Selzer, Moldt (2024) "HIV-1 Envelope diversity and sensitivity to broadly neutralizing antibodies across stages of acute HIV-1 infection" *AIDS*
21. Sayers, Cavanaugh, Clark (2021) *Nucleic Acids Res*
22. Reeves, Zheng, Olefsky (2019) "Susceptibility to Anti-HIV bNAbs is Concordant in Pre-ART Plasma and On-ART PBMC Samples: ACTG NWCS413"
23. Richman, Wrin, Little (2003) "Rapid evolution of the neutralizing antibody response to HIV type 1 infection" *Proc Natl Acad Sci*
24. Margot, Pennetzdorfer, Naik (2023) "Cross-resistance to entry inhibitors and lenacapavir resistance through week 52 in study CAPELLA" *Antivir Ther*
25. Wensing, Calvez, Ceccherini-Silberstein (2025) "2025 update of the drug resistance mutations in HIV-1" *Top Antivir Med*
26. Margot, Jogiraju, Pennetzdorfer (2025) "Resistance analyses in heavily treatment-experienced people with HIV treated with the novel HIV capsid inhibitor lenacapavir after 2 years" *J Infect Dis*
27. Yoon, Macke, West (2015) "CATNAP: a tool to compile, analyze and tally neutralizing antibody panels" *Nucleic Acids Res*
28. Tolbert, Nguyen, Tehrani (2021) "Near-panneutralizing, plasma deconvoluted antibody N49P6 mimics host receptor CD4 in its quaternary interactions with the HIV-1 envelope trimer" *mBio*
29. Schommers, Gruell, Abernathy (2020) "Restriction of HIV-1 Escape by a highly broad and potent neutralizing antibody" *Cell*
30. Julien, Sok, Khayat (2013) "Broadly neutralizing antibody PGT121 allosterically modulates CD4 binding via recognition of the HIV-1 gp120 V3 base and multiple surrounding glycans" *PLoS Pathog*
31. Doores, Burton (2010) "Variable loop glycan dependency of the broad and potent HIV-1-neutralizing antibodies PG9 and PG16" *J Virol*
32. Sok, Van Gils, Pauthner (2014) "Recombinant HIV envelope trimer selects for quaternary-dependent antibodies targeting the trimer apex" *Proc Natl Acad Sci U S A*
33. Margot, Vanderveen, Naik (2022) "Phenotypic resistance to lenacapavir and monotherapy efficacy in a proof-of-concept clinical study" *J Antimicrob Chemother*
34. Margot, Naik, Vanderveen (2022) "Resistance analyses in highly treatment-experienced people with human immunodeficiency virus (HIV) treated with the novel capsid HIV inhibitor lenacapavir" *J Infect Dis*
35. Demirdjian, Naik, Margot (2025) "Phenotypic characterization of replication-impaired lenacapavir-resistant HIV clinical isolates" *J Med Virol*
36. Marcelin, Charpentier, Jary (2020) "Frequency of capsid substitutions associated with GS-6207 in vitro resistance in HIV-1 from antiretroviral-naive and -experienced patients" *J Antimicrob Chemother*
37. Hansen, Hendricks, Chang (2025) "Impact of HIV-1 capsid polymorphisms on viral infectivity and susceptibility to lenacapavir" *mBio*
38. Hagins, Berhe, Crofoot (2025) "Final efficacy and safety of twice-yearly subcutaneous lenacapavir in treatment-naïve people with HIV: randomized study" *AIDS*
39. Mendoza, Gruell, Nogueira (2018) "Combination therapy with anti-HIV-1 antibodies maintains viral suppression" *Nature*
40. Gaebler, Nogueira, Stoffel (2022) "Prolonged viral suppression with anti-HIV-1 antibody therapy" *Nature*
41. Bricault, Yusim, Seaman (2019) "HIV-1 Neutralizing antibody signatures and application to epitope-targeted vaccine design" *Cell Host Microbe*
42. Kalada, Cory (2022) "The importance of tissue sanctuaries and cellular reservoirs of HIV-1" *Curr HIV Res*
43. Jacobs, Halvas, Tosiano (2019) "Persistent HIV-1 viremia on antiretroviral therapy: measurement and mechanisms" *Front Microbiol*
44. Taiwo, Zheng, Rodriguez (2025) "Phase 2 trial of longacting cabotegravir and VRC07-523LS for viral suppression in adults with HIV-1: ACTG A5357" *Clin Infect Dis*
45. Peluso, Bakkour, Busch (2021) "A high percentage of people with human immunodeficiency virus (HIV) on antiretroviral therapy experience detectable low-level plasma HIV-1 RNA following coronavirus disease 2019 (COVID-19)" *Clin Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12685203&blobtype=pdf | # A new path to spillover: MHC-II entry of influenza A viruses
Silke Stertz, Umut Karakus
## Zoonotic spillover of influenza A viruses represents a threat to human health. Emerging research suggests that some influenza A viruses can enter host cells via MHC-II across species, potentially increasing spillover risk.
Respiratory virus infections place a significant burden on human health. Influenza A (IAV) and B viruses alone account for approximately 1 billion infections annually, leading to 3-5 million severe cases requiring hospitalization [1]. Furthermore, the threat of a respiratory virus pandemic remains ever-present. Containing outbreaks is particularly challenging due to the limited understanding of airborne transmission and the scarcity of effective countermeasures [2]. While attention had shifted to coronaviruses after the COVID-19 pandemic, focus is returning to IAV due to the expanding host range and global spread of highly pathogenic avian H5N1 strains. Particularly concerning is the recent emergence of H5N1 in dairy cattle in the USA, which has raised fears of a potential H5N1 pandemic [3].
So how can we mitigate such zoonotic threats? First, we need to identify and assess them before devising mitigation plans. This requires studying past zoonotic events to deduce the mechanisms of adaptation. Different viral features, such as viral polymerase activity or acid stability of the viral surface glycoprotein hemagglutinin, contribute to zoonotic potential. Here, we focus on receptor specificity, crucial for zoonotic spillover, as one of the key determinants [4]. Most IAVs use sialic acid, the terminal sugar on many host cell glycans, as their primary attachment receptor, mediated by viral hemagglutinin. Mammalian IAVs prefer sialic acid linked to galactose via an α2,6′-linkage, whereas avian IAVs bind sialic acid with an α2,3′-linkage. Successful adaptation from avian to mammalian hosts therefore requires a switch in sialic acid specificity.
But what if IAVs could use a different receptor? Our previous work revealed that the IAV subtypes H17N10 and H18N11, which have so far been detected exclusively in bats, use major histocompatibility complex class II (MHC-II) molecules as entry receptors instead of the conventional sialic acid receptors [5]. As a central immune regulator, MHC-II is abundantly expressed on professional antigen-presenting cells and in certain subsets of airway epithelial cells, making it a good target for respiratory viruses. Interestingly, this ability to exploit MHC-II is not limited to bat IAVs: the newly identified avian H19 subtype also lacks the ability to bind sialic acid and instead uses MHC-II [6]. Intriguingly, both avian and human strains of the IAV H2 subtype exhibit dual receptor usage; they can enter cells either via MHC-II in a sialic acidindependent manner or through the traditional sialic acid pathway [7]. Perhaps most striking is the breadth and specificity of MHC-II usage across host species. H17, H18, H19, and H2 IAVs can use MHC-II complexes found in various aquatic birds, such ducks and swans. However, only H17, H18, and H2 (but not H19) are capable of entering cells via human and swine MHC-II. The molecular determinants of this difference in breadth are not understood as the binding sites are only partially defined for a small subset of HAs and HLA-DR [7,8].
Given the established role of receptor specificity in zoonotic transmission, and the fact that avian H2 IAVs can use human MHC-II for entry, these findings raise the possibility that MHC-II tropism represents a previously unrecognized determinant of zoonotic risk. Support for this hypothesis comes from an earlier study that assessed the replicative potential of a broad panel of avian H2 viruses in mammalian systems. The authors of this study observed large differences in infectivity and replication that could not be explained by known zoonotic markers [9]. However, when we compared their data with our mapped sequence motif for MHC-II entry competence in H2 viruses, a clear positive correlation emerged [7]. Thus, the ability to use MHC-II as receptor does indeed seem to promote IAV's ability to infect and replicate in mammalian cells and should be considered for zoonotic risk assessment (Fig 1).
In light of this, it is therefore conceivable that MHC-II entry competence contributed to the zoonotic transmission of H2 IAV, which led to the 1957 H2 pandemic. But what about H1 and H3 IAVs, the two other subtypes that successfully made the jump to humans? So far, MHC-II entry competence has not been observed in the tested panel of human and avian strains of H1 and H3 IAVs, suggesting that MHC-II competence is not widespread among these subtypes. However, fewer than 10 strains of each subtype have so far been tested, so it is still possible that certain H1 or H3 strains could exhibit MHC-II entry competence. Furthermore, testing has primarily focused on human MHC-II, and it is possible that some strains can utilize MHC-II molecules from a limited subset of species, as seen with H19 strains. Even such restricted MHC-II usage could influence host tropism and facilitate adaptation in intermediate hosts. In support of this idea, a recent preprint identified MHC-II entry competence in a swine H3 strain [10]. It will therefore be crucial to test H1 and H3 viruses more comprehensively against a broad range of MHC-II complexes, particularly those strains that are closely related to pandemic viruses, to determine if MHC-II entry competence plays a more general role in zoonotic transmission. Current and future zoonotic threats, such as the widely circulating H5N1 IAVs of clade 2.3.4.4b, should also be tested for MHC-II entry competence. Given the broad host range of these viruses, it will be worth investigating whether MHC-II entry competence contributes to their success in crossing species barriers.
Beyond the prevalence and species specificity of MHC-II entry competence, questions also arise about the impact of dual receptor specificity within the host. MHC-II usage could influence IAV tropism and its interplay with the host immune system, as susceptibility to H2N2 infection is enhanced in MHC-II-expressing immune cells and airway epithelial cells [7].
Another key question that remains unanswered is whether MHC-II usage could impact the host immune response or even facilitate immune evasion, for example by modulating receptor availability or impairing antigen-presenting cell functions. Unlike conventional IAVs that rely on neuraminidase to cleave sialic acid receptors and promote viral release, an alternative receptor-destroying activity targeting MHC-II would represent a distinct adaptation of such viruses. Addressing these questions could reveal new intersections between receptor usage, immune regulation, and viral evolution, reshaping our understanding of how influenza viruses adapt and emerge across hosts.
Together, these open questions highlight that MHC-II entry competence may represent an underappreciated feature of IAV biology with implications for host range and zoonotic potential. A systematic study of MHC-II usage across subtypes, host species, and cellular contexts combined with evolutionary and immunological approaches will be needed to reveal how MHC-II entry competence contributes to cross-species transmission and to identify viral vulnerabilities that might be exploited for surveillance and intervention.
## Author contributions
Conceptualization: Silke Stertz, Umut Karakus.
Visualization: Silke Stertz, Umut Karakus.
## References
1. Who (2025)
2. Wang, Prather, Sznitman et al. (2021) "Airborne transmission of respiratory viruses" *Science*
3. Peacock, Moncla, Dudas et al. (2025) "The global H5N1 influenza panzootic in mammals" *Nature*
4. Long, Mistry, Haslam et al. (2019) "Host and viral determinants of influenza A virus species specificity" *Nat Rev Microbiol*
5. Karakus, Thamamongood, Ciminski et al. (2019) "MHC class II proteins mediate cross-species entry of bat influenza viruses" *Nature*
6. Karakus, Mena, Kottur et al. (2024) "H19 influenza A virus exhibits species-specific MHC class II receptor usage" *Cell Host Microbe*
7. Karakus, Borau, Martínez-Barragán et al. (2024) "MHC class II proteins mediate sialic acid independent entry of human and avian H2N2 influenza A viruses" *Nat Microbiol*
8. Olajide, Osman, Robert et al. (2023) "Evolutionarily conserved amino acids in MHC-II mediate bat influenza A virus entry into human cells" *PLoS Biol*
9. Jones, Baranovich, Marathe et al. (2014) "Risk assessment of H2N2 influenza viruses from the avian reservoir" *J Virol*
10. Rajao, Cardenas, Compton et al. (2025) "MHC class II is a functional receptor for H3N2 Influenza A viruses and mediates host-specificity" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12751533&blobtype=pdf | # BMC Infectious Diseases
Saeid Jamehdar, Somayeh Jalilvand, Mehrnaz Kaffashian, Amin Tadayoni, Ehsan Khadivi, Mohammad Farahmand, Zabihollah Shoja
## Abstract
Background Human papillomavirus (HPV) and Epstein-Barr virus (EBV) are two common viruses that are considered to be associated with various head and neck cancers, including laryngeal cancer. Understanding the prevalence of these viruses in patients referred for laryngoscopy can provide valuable insights into the potential role of viral infections in the progression of laryngeal diseases. This information will help elucidate the role of these viruses in upper respiratory tract pathologies and guide future diagnostic and treatment strategies. This study aims to determine the prevalence of HPV, HPV types, and EBV in biopsy samples of patients referred for laryngoscopy in the Northeast of Iran.Methods A total of 144 laryngeal biopsy samples with histopathological findings of polyps and squamous cell carcinoma were included in this study. HPV DNA identification and genotyping were carried out using the PCR with MY09/MY11 primer and nested PCR with the GP5+/6 + primer set and Real-Time PCR employing the AmpliSens® HPV HCR genotyping kit, respectively. EBV DNA detection was carried out using conventional PCR targeting the LMP1 gene. Statistical analysis was performed using the SPSS software and included Fisher's exact test.
ResultsThe overall HPV prevalence was 10.4%, with higher rates observed in laryngeal squamous cell carcinoma (LSCC) (18.0%) compared to noncancerous polyp (2.8%) samples, and a statistically significant difference was identified in this regard (P-value = 0.004). The most prevalent HPV types were HPV 33 (4.2%) and 39 (2.7%). Significant associations were observed between HPV positivity and age (≤ 50 years), smoking, and opium addiction status, supporting the role of these factors in viral prevalence and potential disease progression. Also, EBV was detected in 2.8% of the samples (5.5% of LSCC and 0% of polyp samples).
ConclusionThe prevalence of HPV infection was significantly higher in LSCC samples than in polyp samples among Iranian patients, with the main types being 33 and 39. Also, EBV was solely identified in LSCC samples.
## Introduction
Laryngeal cancer represents approximately 1-2% of all cancers globally, with elevated incidence rates noted in Central/Eastern Europe, South America, and certain regions of West Africa. Almost 184,615 new cases were recorded globally in 2020 [1]. From 2003 to 2009, the age-standardized incidence rate (ASIR) of head and neck cancers in Iran showed a noticeable increase, rising from 4.8 to between 7.4 and 8.5 cases per 100,000 individuals. During this period, laryngeal cancer was identified as the predominant subtype among both genders, with 86% of cases occurring in men [2,3]. By 2022, laryngeal cancer accounted for approximately 2% of all new cancer diagnoses in Iran, accompanied by an age-standardized mortality rate (ASMR) of 2.1 per 100,000 population, which exceeds the global average. This highlights the persistent public health challenge posed by laryngeal cancer in the region [4].
Tobacco consumption is the leading risk factor, accounting for over 70% of laryngeal squamous cell carcinoma (LSCC) cases, and possibly as high as 89% when combined with alcohol intake [5]. Viral infections have been considered a risk factor for LSCC development. The link between cancer development in the upper respiratory tract and Epstein-Barr virus (EBV) and human papillomavirus (HPV) has been well-established in human cancer studies [6,7].
HPV is a prevalent sexually transmitted infection that has been linked to the development of various cancers, including several anogenital and head and neck cancers [8]. High-risk HPV types, especially HPV16 and HPV18, are the most prevalent and account for the majority of HPV-associated cancers in the world [9]. HPV infection has been recognized as a contributor linked to head and neck squamous cell carcinomas (HNSCC). HPV-related HNSCC mainly affects the oropharynx, where the virus accounts for approximately 18-22.4% of cancers, depending on the study and region [10]. The occurrence of HPV infection in Iranian HNSCC patients ranges from 3% to 60%, influenced by the different geographical areas [11]. Studies show that high-risk HPV infection types, including 16,18,31,33,45,52, and 58 are associated with an increased risk of HNSCC. In contrast, low and intermediate-risk HPV types (like HPV-6 and HPV-11) do not significantly contribute to HNSCC [12,13].
EBV is an established oncogenic virus associated with various cancers such as nasopharyngeal carcinoma (NPC), Burkitt lymphoma, and Hodgkin lymphoma. Nevertheless, its involvement in laryngeal cancer is not as well understood in comparison to other head and neck tumors [14]. Studies have shown that EBV has been found in laryngeal carcinoma cells, suggesting that it may be a possible risk factor, even though its occurrence in laryngeal cancer is rare and difficult to detect [15]. The prevalence of EBV in laryngeal carcinoma has shown inconsistent results, with reported rates ranging from 0% to 100%. Notably, the two studies that reported a 100% prevalence utilized serological assays [16]. Additionally, variations in sample sizes, methodologies, and targets across the studies contribute to these conflicting results [17].
While HPV and EBV have been detected in a proportion of cases of laryngeal squamous cell carcinoma (LSCC), there is insufficient conclusive evidence to establish a direct causal link. Additional well-structured studies are required to elucidate the potential roles of these viruses in the etiology and advancement of LSCC [7,18,19]. Although numerous studies have been carried out in Iran regarding the prevalence of HPV in LSCC samples, the emphasis has primarily been on types 16 and 18, with other high-risk types remaining unexamined. Furthermore, the prevalence of HPV and EBV viruses in biopsy samples collected from laryngoscopy has not been investigated concurrently [20,21]. Therefore, this study examines the prevalence of EBV and HPV in biopsy samples of patients referred for laryngoscopy in the Northeast of Iran.
## Materials and methods
## Population sample
The present study is a cross-sectional study conducted in Mashhad from 2023 to 2025. To determine the sample size, the Fleiss formula was applied, which indicates the difference in the proportion of individuals with cancer and polyps in both groups. With a statistical confidence of 95% (5% error) and a test power of 80%, a minimum of 72 samples was estimated for each group. In this research, one hundred and forty-four fresh laryngeal biopsy specimens were collected from out of two hundred and fifty-two patients who had been referred to Ghaem Hospital and Kasra Clinic of Mashhad for laryngoscopy, after approval by the Ethics Committee of Tehran University of Medical Sciences (TUMS) (IR.TUMS. SPH.REC.1402.057). Inclusion criteria for patients in this study comprised a histologically confirmed diagnosis of laryngeal squamous cell carcinoma (LSCC) or noncancerous laryngeal polyps through biopsy, age 18 years or older, and no previous treatment involving radiotherapy, chemotherapy, or surgical procedures for laryngeal conditions before sample collection. Excluded from the study were patients with a history of prior malignancies or concurrent cancers apart from LSCC, those presenting with active infections or inflammatory laryngeal conditions other than polyps during biopsy, and individuals who had undergone radiotherapy, chemotherapy, or immunotherapy before sample collection. Histopathological classification was conducted before extraction, by expert pathologists according to standard morphological criteria seen in formalin-fixed paraffin-embedded (FFPE) tissue sections using routine staining and microscopy. It should be noted that because there are major differences in the etiology and neoplastic nature of papillomas versus polyps, papilloma samples were excluded from this study. All individuals in this study signed an informed consent form. Also, a questionnaire was completed for all study subjects regarding various variables, including age, addictive substance use, city of residence, and occupation.
## DNA isolation
DNA was extracted from biopsy tissues using the TRIzol reagent (Yekta Tajhiz Azma, Iran; Cat. No. YT9065) according to the manufacturer's instructions. Briefly, for every 50-100 mg of tissue, 1 ml of Trizol solution was added. Using a micropestle, the tissues were pounded and finally homogenized, followed by phase separation with chloroform to isolate DNA from interphase and organic phases. The DNA was then purified through precipitation and washing steps using ethanol and NaOH buffer prior to downstream applications.
Finally, the quality of extracted DNA was assessed by PCR for amplification of a 227 bp amplicon from the human HLA-DQ gene using HLAdQ-R/HLAdQ-F primers (Table 1).
## HPV DNA detection and typing
All specimens were analyzed via PCR with a MY09/ MY11 primer and nested PCR utilizing the GP5+/6 + primer pair to obtain a 150 bp amplicon of the L1 gene (Table 1). The PCR reactions and amplification cycles were conducted following a previously published protocol [22].
Real-Time PCR for HPV DNA genotyping was performed using the AmpliSens® HPV HCR genotyping kit (Cat. Number: R-V67-F-CE) in the Rotor-Gene system (Cat no / ID. 9001862) according to the manufacturer's instructions. This kit is designed to detect and genotype 14 high-risk HPV types, including HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68. The comprehensive coverage of clinically relevant high-risk HPV types ensures accurate genotyping relevant to oncogenic risk assessment in laryngeal squamous cell carcinoma.
Positive controls consisting of known HPV-positive DNA samples and negative controls, including reaction mixtures lacking DNA template, were included in each PCR run to monitor assay specificity and contamination.
## EBV DNA detection
All samples were tested by PCR using a LMP1-F/LMP1-R primer pair to get a 219 bp amplicon of the LMP1 gene (Table 1). The PCR reaction was done in a 50 µl reaction mixture including 10 pmol of each primer, 1.5 mM MgCl 2 , 1.5 U of Taq DNA polymerase, 50 µM of each dNTP, and 100-200 ng of DNA template. PCR amplification cycles were performed under the following conditions: 95 °C for 5 min, 40 cycles of 95 °C for 20s, 55 °C for 20s, and 72 °C for 20s, and one extension time of 72 °C for 5 min. Each PCR run included a no-template negative control and a positive control consisting of EBVcontaining DNA to ensure assay reliability and detect contamination.
## Statistical analysis
Data analysis was conducted utilizing the SPSS software. Descriptive statistics, such as means, standard deviations (SD), frequencies, and percentages, were employed to provide an overview of the demographic and clinical variables. Group comparisons between laryngeal squamous cell carcinoma (LSCC) and polyp samples for categorical variables, including virus prevalence and risk factors, were analyzed using Fisher's exact tests. Statistical significance was set at a two-tailed p-value < 0.05. Due to the limited number of EBV-positive cases (n = 4), inferential analyses of associations between EBV and clinical or demographic variables were interpreted descriptively, and p-values for these comparisons were reported cautiously, recognizing the limited statistical power.
## Results
The study included a total of 144 samples, which were detailed as follows: 72 laryngeal squamous cell carcinoma (LSCC) and 72 noncancerous laryngeal polyp samples, which were characterized as a benign proliferation of cells on the surface of the vocal cords. The average age of patients, represented as mean ± standard deviation (SD), was as follows: 47.49 ± 11.87 (noncancerous poly group) and 57.72 ± 8.17 (LSCC group).
As shown in Table 2, the prevalence of HPV and EBV was 10.4% and 2.8%, respectively. A statistically significant difference was observed between HPV infection with histopathology, age, addiction, or smoking (P < 0.05). However, no differences were found between HPV infection with gender or drinking alcohol (P > 0.05). Regarding EBV infection, no statistically significant difference was found for all analyzed variables (P > 0.05) (Table 2). The frequency of HPV was higher in LSCC samples (18.0%) compared to laryngeal polyps (2.8%), and a statistically significant difference was found in this regard (P = 0.004). The prevalence of HPV was also statistically increased in patients ≤ 50 years old (16.2%) than in those > 50 years old (3.1%) (P = 0.017). HPV was detected in 16.7% and 3% of patients with addiction and nonaddicted, respectively, and a statistically significant difference was observed for this variable (P = 0.012) (Table 2).
The most prevalent HPV types, regardless of histopathology, were HPV 33 (4.2%), 39 (2.7%), 58 (2.1%), 18 (1.4%), and 51 (1.4%). High-risk HPV types were identified in 2.8% of polyps and 15.3% of LSCC samples. Also, 2.8% of LSCC specimens were infected with low-risk HPV types (Table 3). The prevalence of different HPV types among the two investigated groups was as follows:
HPV 33/39 (1.4%) and HPV 33/45 (1.4%) in polyp group; HPV 18 (2.8%), HPV 33/51 (1.4%), HPV 39 (2.8%), HPV 33/39 (1.4%), HPV 51 (1.4%), HPV 58 (2.8%), HPV 33/58 (1.4%), and HPV 33/66 (1.4%) in LSCC group.
According to histopathological type of LSCC, the prevalence of HPV in HPV-positive samples was as follows: HPV 39 (16.7%) in poorly differentiated group; HPV 18 (5.0%), HPV 33/51 (2.5%), HPV 51 (2.5%), HPV 58 (2.5%), and HPV 33/66 (2.5%) in moderately differentiated group; HPV 39 (3.8%), HPV 33/39 (3.8%), and HPV 33/58 (3.8%) in well differentiated group. Additionally, in comparison, a greater percentage of poorly differentiated samples (33.3%) were found to be infected with HPV.
The overall prevalence of EBV was 2.8%. The EBV was identified in 5.5% of LSCC samples, but not in polyp samples. Based on the histopathology results, a greater percentage of moderately differentiated samples (7.5%) were found to be infected with EBV. However, no statistically significant differences were observed (Table 4).
## Discussion
Cancers of the head and neck represent 4% of all cancer types, with laryngeal carcinoma making up 25% to 40% of these malignancies [23]. Numerous factors, particularly the use of tobacco and the intake of alcohol, have been linked to the onset of laryngeal squamous cell carcinomas (LSCC). Additionally, it is recognized that specific viruses possess oncogenic properties, and the connection between laryngeal squamous cell carcinoma and these viruses has been a widely studied topic for many years [23][24][25]. LSCCs are aggressive tumors that have a debated relationship with HPV and EBV [6]. Although HPV is more closely associated with oropharyngeal cancer, certain studies indicate a possible involvement in LSCC [26].
In this study, the overall prevalence of HPV was 10.4%. However, it was noted that the prevalence of HPV was statistically increased in patients ≤ 50 years old (16.2%) than in those >50 years old (3.1%) (P = 0.017). Research indicates that HPV positivity is markedly elevated in individuals under 50 years, particularly within the 31-40 age range, where prevalence can approach approximately 40%, in contrast to much lower rates in older individuals. One reason for this is that younger patients are more likely to acquire HPV through sexual behaviors, including oral sex [27,28]. Thus, it is probable that if we had access to individuals with laryngeal cancer in a younger age bracket, the proportion of HPV-positive cases would have been greater. Also, most of the HPV-positive individuals were patients with addiction and/or smokers. Given that smoking diminishes the immune response and hinders HPV clearance, and that HPV positivity was more frequently observed among smokers in this study, it can be concluded that these two elements combined may elevate the risk of LSCC [29]. Additionally, opium addiction, similar to smoking, could play a role in the persistence of HPV infection through immunosuppressive effects, as indicated by findings from research on other abused substances that weaken immune responses essential for viral clearance [30]. Thus, the combination of addiction and HPV positivity can elevate the risk of LSCC [31]. Considering the prevalence of smoking and addiction among EBV-positive individuals and the role of these two factors in facilitating EBV activation, it can be said that the combination of these factors can increase the risk of LSCC [32]. As indicated in Tables 2 and86.7% of HPV positives and 100% of EBV positives were male, which can be ascribed to various factors, including increased exposure risk via sexual behavior, biological vulnerability, and characteristics of the study group [33,34].
Polyp samples can serve as normal controls in studies on laryngeal squamous cell carcinoma (LSCC) because they represent benign lesions of the laryngeal mucosa without malignant potential or epithelial dysplasia, unlike papillomas, which, although a subgroup of polyps, are epithelial neoplastic in nature [35,36]. Etiologically, laryngeal polyps are primarily associated with trauma, chronic irritation, and inflammation, while laryngeal papillomas are caused by viral infection with human papillomavirus (HPV), leading to neoplastic epithelial proliferation [37]. The prevalence of HPV in total was 10.4%. Based on histopathology specifications, HPV was identified in 2.8% and 18.0% of polyp and LSCC samples, respectively. The most prevalent HPV types were HPV 33 (4.2%) and 39 (2.7%), irrespective of histopathology. Additionally, the most common HPV types identified were HPV 33 (5.5%) and 39 (4.1%) in LSCC samples. Several studies have found that HPV16 and HPV18 are the most commonly associated high-risk types in LSCC, with HPV16 being the leading type. HPV16 prevalence is particularly high in oropharyngeal cancers and to a somewhat lesser extent in laryngeal cancers, though it remains the most frequent genotype [38,39]. As stated previously, the predominant types detected in LSCC samples in our research were 33, 39, and to a lesser degree 18, suggesting a discrepancy with other studies. Such regional variations in HPV genotype distribution are supported by emerging evidence that HPV type prevalence can vary significantly by geographic location, ethnic populations, and [38]. These discrepancies in results may be attributable to several factors, including the quality of test samples, variations in the geographical distribution of study populations, and the specificity and sensitivity of the diagnostic methods employed [40].
Considering the greater frequency of HPV in LSCC samples compared to polyp samples and research indicating a potential involvement of HPV in the development of LSCC, it can be concluded that preventing this virus may lower the risk of LSCC [41]. Given that the predominant types identified in SCC samples were types 33 and 39, it can be concluded that the Gardasil 9 vaccine is the optimal choice for the prevention of HPV-related cancers [42]. Our results indicated that a greater percentage of the poorly differentiated samples (33.3%) of LSCC were infected with HPV, which aligns perfectly with other findings in this area reported globally [43,44]. It is important to note that some subgroup analyses, particularly those involving poorly differentiated LSCC samples, are based on small numbers of cases. Consequently, these findings should be interpreted with caution as the limited sample size may reduce statistical power and increase the risk of chance associations. This result highlights the importance of further studies with larger groups centered on histopathological traits to uncover the pathogenic mechanisms and confirm these findings.
This study provides an evaluation of EBV occurrence in laryngeal specimens alongside HPV. The prevalence of EBV in total was 2.8%. EBV was detected in 5.5% of LSCC samples but not in polyp samples (P-value = 0.12). A comprehensive recent meta-analysis revealed that EBV infection is associated with an approximately threefold increased risk of developing laryngeal carcinoma (odds ratio = 2.86) [19]. However, given the lack of statistically significant difference in our study and the small number of EBV-positive cases, the etiological role of EBV in LSCC remains inconclusive and warrants cautious interpretation. Thus, while the literature suggests a potential association between EBV and LSCC, our findings do not provide sufficient evidence to support the notion that suppressing EBV would necessarily reduce LSCC risk. Further large-scale and well-powered studies are required to clarify EBV's contribution to LSCC pathogenesis and to assess the potential impact of targeting EBV in prevention or treatment strategies [45].
The most important limitations of the present study were found to be the moderately small sample size and the lack of the expression analysis of HPV and EBV transcripts or proteins.
In conclusion, the results of this study showed that HPV is more common in LSCC samples than in polyp samples, with HPV types 33 and 39 being the most frequently found in LSCC among Iranian patients. Also, EBV was only detected in LSCC samples. The important point is that both viruses may play a role in LSCC pathogenesis, but their mechanisms and clinical effects vary and are contentious, underscoring the necessity for more research, customized prevention, and monitoring approaches. Nevertheless, future studies with larger sample sizes and the expression analysis of HPV and EBV transcripts or proteins are needed.
## References
1. Sung, Ferlay, Siegel et al. (2021) "Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries" *CA Cancer J Clin*
2. Mirzaei, Hosseini, Ghoncheh et al. (2015) "Epidemiology and trend of head and neck cancers in Iran" *Glob J Health Sci*
3. Naeini, Raad, Eslami et al. (2023) "The geographical distribution of laryngeal cancer in Iran from 2004 to 2014" *J Otorhinolaryngol Facial Plast Surg*
4. Farimah, Hamideh, Fereshteh et al. (2024) "Geographic variation and trends in laryngeal cancer incidence in Iran: a comprehensive analysis" *Basic Clin Cancer Res*
5. Menach, Oburra, Patel (2012) "Cigarette smoking and alcohol ingestion as risk factors for laryngeal squamous cell carcinoma at Kenyatta National hospital, Kenya" *Clin Med Insights Ear Nose Throat*
6. De Oliveira, Bacchi, Macarenco et al. (2006) "Human papillomavirus and Epstein-Barr virus Infection, p53 Expression, and cellular proliferation in laryngeal carcinoma" *Am J Clin Pathol*
7. Schindele, Holm, Nylander et al. (2022) "Mapping human papillomavirus, Epstein-Barr virus, cytomegalovirus, adenovirus, and p16 in laryngeal cancer" *Discov Oncol*
8. Baghi, Aghbash, Rasizadeh et al. (2024) "Cancers associated with human papillomavirus: an overview of prevalence in Iran and the middle East" *Exploratory Res Hypothesis Med*
9. Brianti, Flammineis, Mercuri (2017) "Review of HPV-related diseases and cancers" *New Microbiol*
10. Roman, Aragones (2021) "Epidemiology and incidence of HPV-related cancers of the head and neck" *J Surg Oncol*
11. Karimi, Mohebbi, Mckay-Chopin et al. (2022) "Human papillomavirus and risk of head and neck squamous cell carcinoma in Iran" *Microbiol Spectr*
12. Michaud, Langevin, Eliot et al. (2014) "High-risk HPV types and head and neck cancer" *Int J Cancer*
13. Kreimer, Clifford, Boyle et al. (2005) "Human papillomavirus types in head and neck squamous cell carcinomas worldwide: A systematic review" *Cancer Epidemiol Biomarkers Prev*
14. Baumforth, Young, Flavell et al. (1999) "The Epstein-Barr virus and its association with human cancers" *Mol Pathol*
15. Samara, Athanasopoulos, Mastronikolis et al. (2024) "The role of oncogenic viruses in head and neck cancers: epidemiology, pathogenesis, and advancements in detection methods" *Microorganisms*
16. Goldenberg, Benoit, Begum et al. (2004) "Epstein-Barr virus in head and neck cancer assessed by quantitative polymerase chain reaction" *Laryngoscope*
17. De Lima, Silva, Do et al. (2021) "Epstein-Barr virus-associated carcinoma of the larynx: a systematic review with metaanalysis" *Pathogens*
18. Nouri, Larizadeh, Mollaei et al. "Assessment of Epstein-Barr virus in patients with squamous cell carcinoma of the larynx by real-time polymerase chain reaction" *J Contemp Med Sci*
19. De Lima, Silva, Do et al. (2021) "Epstein-Barr virus-associated carcinoma of the larynx: a systematic review with meta-analysis" *Pathogens*
20. Atighechi, Meybodian, Dadgarnia et al. (2016) "Investigating the prevalence of human papilloma virus in squamous cell carcinoma of the larynx and its correlation with disease prognosis" *Iran J Otorhinolaryngol*
21. Karimi, Mohebbi, Mckay-Chopin et al. (2022) "Human papillomavirus and risk of head and neck squamous cell carcinoma in Iran" *Microbiol Spectr*
22. Heydari, Oskouee, Vaezi et al. (2018) "Type-specific human papillomavirus prevalence in cervical intraepithelial neoplasia and cancer in Iran" *J Med Virol*
23. Oksüzler, Tuna, Soyaliç et al. (2009) "Investigation of the synergism between alcohol consumption and herpes simplex virus in patients with laryngeal squamous cell cancers" *Eur Arch Otorhinolaryngol*
24. Morshed (2010) "Association between human papillomavirus infection and laryngeal squamous cell carcinoma" *J Med Virol*
25. De Oliveira, Bacchi, Macarenco et al. (2006) "Human papillomavirus and Epstein-Barr virus infection, p53 expression, and cellular proliferation in laryngeal carcinoma" *Am J Clin Pathol*
26. Ndiaye, Mena, Alemany et al. (2014) "HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis" *Lancet Oncol*
27. Ghosh, Kumar, Chaudhary et al. (2023) "High-Risk human papillomavirus infection in squamous cell carcinoma of the larynx: A study from a tertiary care center in North India" *Cureus*
28. Young, Xiao, Murphy et al. (2015) "Increase in head and neck cancer in younger patients due to human papillomavirus (HPV)" *Oral Oncol*
29. Warren, Singh (2013) "Nicotine and lung cancer" *J Carcinog*
30. Minkoff, Zhong, Strickler et al. (2008) "The relationship between cocaine use and human papillomavirus infections in HIV-seropositive and HIVseronegative women" *Infect Dis Obstet Gynecol*
31. Kosciuczuk, Knapp, Lotowska-Cwiklewska (2020) "Opioid-induced immunosuppression and carcinogenesis promotion theories create the newest trend in acute and chronic pain pharmacotherapy" *Clinics*
32. Su, Siak, Leong et al. (2023) "The role of Epstein-Barr virus in nasopharyngeal carcinoma" *Front Microbiol*
33. Lewis, Markowitz, Gargano et al. (2018) "Prevalence of genital human papillomavirus among sexually experienced males and females aged 14-59 Years, united States, 2013-2014" *J Infect Dis*
34. Mazurek, Jaros, Gliwa et al. (2025) "Epstein-Barr virus (EBV) and human papilloma virus (HPV) in gastric cancers, with special reference to gastric cancer at a young Age-A pilot study in Poland" *Int J Mol Sci*
35. Gong, Shi, Zhou et al. (2013) "The composition of Microbiome in larynx and the throat biodiversity between laryngeal squamous cell carcinoma patients and control population" *PLoS ONE*
36. Guo, Zang, Fu et al. (2023) "Classification of nasal polyps and inverted papillomas using CT-based radiomics" *Insights Imaging*
37. Vasconcelos, Gomes, Araújo (2019) "Vocal fold polyps: literature review" *Int Arch Otorhinolaryngol*
38. Mohamadian Roshan, Jafarian, Ayatollahi et al. (2014) "Correlation of laryngeal squamous cell carcinoma and infections with either HHV-8 or HPV-16/18" *Pathol -Res Pract*
39. Yang, Shi, Tang et al. (2019) "Effect of HPV infection on the occurrence and development of laryngeal cancer: A review" *J Cancer*
40. Quintero, Giraldo, Uribe et al. (2013) "Human papillomavirus types in cases of squamous cell carcinoma of head and neck in Colombia" *Brazilian J Otorhinolaryngol (English Edition)*
41. Wang, Wei, Wang et al. (2020) "Role of human papillomavirus in laryngeal squamous cell carcinoma: A meta-analysis of cohort study" *Cancer Med*
42. Cheng, Wang, Du (2020) "Human papillomavirus vaccines: an updated review" *Vaccines (Basel)*
43. Laskaris, Sengas, Maragoudakis et al. (2014) "Prevalence of human papillomavirus infection in Greek patients with squamous cell carcinoma of the larynx" *Anticancer Res*
44. Brito, Cossetti, De Souza et al. (2022) "Prevalence of HPV genotypes and assessment of their clinical relevance in laryngeal squamous cell carcinoma in a Northeastern state of Brazil-a retrospective study" *PeerJ*
45. Muderris, Rota, Muderris et al. (2013) "Does Epstein-Barr virus infection have an influence on the development of laryngeal carcinoma? Detection of EBV by < span class=elsevierStyleItalic>Real-Time polymerase chain Reaction in tumour tissues of patients with laryngeal carcinoma" *Brazilian J Otorhinolaryngol (English Edition)* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12754539&blobtype=pdf | # Disclosure and peer support in prevention of mother-to-child transmission of human immunodeficiency virus: Evidence from Rwanda
Saurav Basu
## Abstract
This retrospective cohort study from Rwanda demonstrated the likelihood of maternal disclosure and peer support in preventing mother-to-child human immunodeficiency virus (HIV) transmission. High sustained maternal viral load suppression (91.0%) and exceptional infant testing uptake (100% at 6 weeks) correlated with a low 0.7% infant HIV incidence. To eliminate mother-to-child transmission of HIV, effective strategies must engage male partners in disclosure, reduce stigma, improve health literacy, and provide structural peer-support for enhancing adherence and mental health.
## INTRODUCTION
Pediatric HIV infections remain a significant global public health challenge, with over 120000 new human immunodeficiency virus (HIV) infections recorded among children under five years old in 2023 [1]. Mother-to-child transmission (MTCT), occurring during pregnancy, labor, delivery, or breastfeeding, represents the primary mode of HIV acquisition in children, with the highest risk observed in low-resource settings. It is well-established that effective interventions for the prevention of MTCT, particularly maternal antiretroviral therapy (ART), substantially reduce this risk to below 5% in breastfeeding children and less than 2% in non-breastfeeding children [2].
Achieving viral load suppression (VLS) in individuals with HIV necessitates consistent daily adherence to ART [3]. However, adverse social determinants, such as lower socioeconomic status, inadequate social support, limited education, and pervasive societal HIV-related stigma, frequently impede access to and retention in ART among mothers living with HIV [4]. Non-adherence to ART, whether due to failure of initiation, intermittent missed doses, or complete cessation of therapy, significantly elevates the risk of MTCT [2,3]. A critical gap in the research literature from developing nations is that VLS estimations are predominantly derived from cross-sectional studies, which, by their nature, capture VLS at discrete time points within the HIV care continuum.
Rwanda, a Central African nation, with a generalized HIV epidemic (adult prevalence 3%) has made major strides in reducing HIV incidence (0.08%) with improved awareness, testing, access, acceptability, and affordability of ART. These national achievements are mirrored at a local level; in the Karongi district, the HIV prevalence among pregnant women attending antenatal care services dropped from 2.7% in 2010 to just 0.3% in 2019 [5]. Furthermore, since 2015, the rate of MTCT of HIV within health facilities has reduced below 2%, primarily through expansion of ART services in existing ANC facilities [6].
## EVIDENCE OF DISCLOSURE AND PEER SUPPORT IN PREVENTION OF MOTHER-TO-CHILD TRANSMISSION OF HIV IN RWANDA
A recent 24-month retrospective cohort study from Rwanda by Bakari et al [7] highlights further progress in the nation's fight against HIV. The study found that 91% of mothers achieved sustained viral load suppression, and a perfect infant testing uptake (100% at six weeks) correlating with a low infant HIV incidence of only 0.7%. Further, the study findings indicated that mothers who were married or living with a partner, had disclosed their HIV status, or initiated ART during pregnancy were more likely to maintain VLS. Furthermore, multivariate analysis revealed a statistically significant association between assigning mothers to peer educators and more consistent infant HIV testing.
The findings from Bakari et al [7] further reveal that when women disclose their HIV status to family members, especially male partners, they experience increased familial support which in turn, is associated with improved ART initiation and adherence, as well as timely HIV testing and prophylaxis. These results reasonably align with prior research on the benefits of disclosure. A systematic review and meta-analysis (SRMA) of 21 studies in China found that people living with HIV (PLHIV) who disclosed their status were 2.59 times more likely to initiate ART than those who did not, although no significant improvement in overall adherence was observed [8]. In contrast, a similar SRMA consisting of seven studies from Ethiopia reported that adults who disclosed were 1.64 times more likely to demonstrate good ART adherence compared to non-disclosers [9]. The Bakari et al [7] study provides robust, corroborative evidence for this phenomenon from a prospective design, a higher evidentiary standard, thereby further strengthening the linkage between HIV status disclosure and improved health outcomes.
Women living with HIV frequently encounter substantial challenges related to stigma and mental health. A systematic review and meta-analysis of ten studies from sub-Saharan Africa estimated the pooled prevalence of antenatal depression among pregnant women living with HIV at 39.86% (95%CI: 34.89-44.83) [10]. The secondary benefits of disclosure include alleviation of perceived stigma and subsequent improvement in mental well-being. Women who have disclosed their HIV status to their families are also better positioned to make informed decisions regarding their reproductive health, particularly family planning. Conversely, a systematic review encompassing 26 studies from high-income countries (United States/Canada) observed that persistent stigma and concerns about disclosure negatively impacted mental health, resulting in reduced ART adherence and retention in care among women with HIV [11].
Fear of the consequences of disclosure, such as violence, abandonment, social ostracization, and lack of familial support, constitute major household-level barriers to disclosure, especially in patriarchal and male-dominant societies. Additionally, healthcare workers may lack the necessary sensitization, empathy, and training crucial for promoting disclosure by women through effectively overcoming their stigma and fears. A study from China revealed that nurses and inexperienced healthcare providers, particularly those with lower education levels, tended to also avoid contact with PLWHA [12].
Peer support is when people with similar experiences or challenges (such as having a disease in common) connect with one another to offer and receive mutual assistance, guidance, encouragement, and emotional support. Informal peer support are informal, organic processes that involve reciprocity and social benefits. Structured peer-mentors in context of PLHIV are deliberate, targeted, formal programs to achieve specific health goals including navigation of the treatment pathway from initiation to high adherence, while coping with the disease. Peer-mother interactive programs are increasingly being explored to enhance ART retention and viral suppression, although their effectiveness in directly preventing MTCT has shown mixed evidence. Peer support interventions in PLHIV as per an SRMA including 20 randomized control trials (RCTs) reported a modest but superior retention in care (RR 1.07), ART adherence (RR 1.06), and viral suppression (RR 6.24) among peer-support participants despite considerable heterogeneity in study outcomes [13].
For instance, a cluster RCT from Tanzania did not find peer-mother integration effective in reducing MTCT, despite observed improvements in ART retention, particularly in ART-naive women [14]. Similarly, another RCT conducted in Uganda (2024) demonstrated the effectiveness of peer support in improving ART retention and viral suppression but did not detect a statistically significant difference in rates of infant HIV positivity [15]. Nevertheless, peer mothers are uniquely positioned to provide emotional support to women living with HIV, aiding them in navigating pregnancy, offering practical advice on coping with ART side effects and misconceptions about disease progression. This support can alleviate stress and anxiety, which are often linked to ART non-adherence [10].
Health literacy refers to the adept accessing and application of health information, leading to informed choices, healthier behaviors, and improved health outcomes. However, among PLHIV, health literacy does not always translate into optimal and safe behaviors [16]. Peer mothers can enhance health literacy among women living with HIV by reinforcing healthcare messages and boosting the credibility and trustworthiness of health information in a relatable manner [17]. Evidence suggests that peer support can significantly improve a range of outcomes among PLHIV, with the strongest benefits observed in social and behavioral domains [18].
## CONCLUSION
Despite the significant benefits of peer support programs in improving the quality of life and health outcomes in PLHIV, especially pregnant mothers, their implementation and sustainability remain a major public health challenge. Barriers to running effective peer education programs include difficulties in recruiting, training, and compensating peer educators for their time, while ensuring optimal and uniform service quality. An additional ethical dilemma arises in maintaining the confidentiality of peer educator mothers' HIV status within the local community while simultaneously encouraging disclosure as role models. This necessitates major anti-stigma campaigns through robust community engagement and social mobilization. The evidence from the Bakari et al [7] study strongly indicates the value of vigorously advancing the development and integration of peer support programs, through capacity building, within national HIV control programs, particularly in developing countries. Future research should identify through qualitative perspectives, successful strategies for male partner engagement in the disclosure process enabling better understanding of household dynamics and ART adherence in households with PLHIV. Studies with a similar design but a longer follow-up period, extending beyond the initial six-week infant HIV testing, are needed to understand the real-world, long-term effectiveness of these interventions in preventing MTCT of HIV. Additionally, economic evaluations are necessary to determine the cost-effectiveness and feasibility of peer educator programs in resource-limited health systems, which would also better inform and complement future scaling-up efforts.
## FOOTNOTES
Author contributions: All contributions by single author.
## Conflict-of-interest statement:
The author declares no conflict of interest.
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. See: https://creativecommons.org/Licenses/by-nc/4.0/ Country of origin: India ORCID number: Saurav Basu 0000-0003-1336-8720.
## S-Editor:
Liu JH L-Editor: Filipodia P-Editor: Zhang L
## References
1. Volmink, Marais (2008) "HIV: mother-to-child transmission" *BMJ Clin Evid*
2. (2010) "Towards universal access: scaling up priority HIV/AIDS interventions in the health sector" *Geneva: World Health Organization*
3. Mcmahon, Elliott, Bertagnolio et al. (2013) "Viral suppression after 12 months of antiretroviral therapy in low-and middle-income countries: a systematic review" *Bull World Health Organ*
4. Basu, Marimuthu, Garg et al. (2024) "Anti-retroviral therapy adherence in India (2012-18): A systematic review and meta-analysis" *Indian J Sex Transm Dis AIDS*
5. Mutagoma, Theogene (2020) "Prevention of Mother-to-Child Transmission (PMTCT) of HIV in Karongi District, Rwanda: A Success Story" *Rw Public Health Bul*
6. Remera, Mugwaneza, Chammartin et al. (2021) "Towards elimination of mother-to-child transmission of HIV in Rwanda: a nested case-control study of risk factors for transmission" *BMC Pregnancy Childbirth*
7. Bakari, Sebeza, Ally et al. (2025) "Sustained maternal human immunodeficiency virus viral load suppression and cascade of human immunodeficiency virus testing among exposed infants in Rwanda" *World J Virol*
8. Fei, Zhao, Yang et al. (2025) "The association between disclosure and antiretroviral therapy among adults living with HIV in China: a systematic review and meta-analysis" *BMC Infect Dis*
9. Dessie, Wagnew, Mulugeta et al. (2019) "The effect of disclosure on adherence to antiretroviral therapy among adults living with HIV in Ethiopia: a systematic review and meta-analysis" *BMC Infect Dis*
10. Ferede, Zeleke, Assefa et al. (2025) "Depression and associated factors among human immunodeficiency viruspositive pregnant women in sub-Saharan Africa: systematic review and meta-analysis" *AJOG Glob Rep*
11. Nawfal, Gray, Sheehan et al. (2024) "A Systematic Review of the Impact of HIV-Related Stigma and Serostatus Disclosure on Retention in Care and Antiretroviral Therapy Adherence Among Women with HIV in the United States/Canada" *AIDS Patient Care STDS*
12. Lin, Li, Wan et al. (2012) "Empathy and avoidance in treating patients living with HIV/AIDS (PLWHA) among service providers in China" *AIDS Care*
13. Berg, Page, Øgård-Repål (2021) "The effectiveness of peer-support for people living with HIV: A systematic review and meta-analysis" *PLoS One*
14. Lyatuu, Naburi, Mwashemele et al. (2022) "Effect of peer-mother interactive programme on prevention of mother-to-child HIV transmission outcomes among pregnant women on anti-retroviral treatment in routine healthcare in Dar es Salaam" *PLOS Glob Public Health*
15. Amone, Gabagaya, Wavamunno et al. (2024) "Enhanced peer-group strategies to support the prevention of mother-to-child HIV transmission leads to increased retention in care in Uganda: A randomized controlled trial" *PLoS One*
16. Wawrzyniak, Ownby, Mccoy et al. (2013) "Health literacy: impact on the health of HIV-infected individuals" *Curr HIV/ AIDS Rep*
17. Sanders, Tobin, Cassells et al. (2021) "Can a brief peer-led group training intervention improve health literacy in persons living with HIV? Results from a randomized controlled trial" *Patient Educ Couns*
18. Han, Zhang, Yang et al. (2023) "The effectiveness and sustainability of peer support interventions for persons living with HIV: a realist synthesis" *BMJ Glob Health*
19. Center
20. "office@baishideng.com Help Desk" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12241321&blobtype=pdf | # Corrections & amendments Author Correction: Sarbecovirus RBD indels and specific residues dictating multi-species ACE2 adaptiveness
Jun-Yu Si, Yuan-Mei Chen, Ye-Hui Sun, Xue Gu, Mei-Ling Huang, Lu-Lu Shi, Xiao Yu, Xiao Yang, Qing Xiong, Cheng-Bao Ma, Peng Liu, Zheng-Li Shi, Huan Yan, Nature Communications |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12650009&blobtype=pdf | # Academic Editors: Shunbin Ning and
Natalia Osna, Roberto Ferrarese, Federica Novazzi, Gabriele Arcari, Angelo Genoni, Francesca Drago Ferrante, Nicola Clementi, Serena Messali, Antonino Guglielmo Pitrolo, Francesca Caccuri, Antonio Piralla, Arnaldo Caruso, Fausto Baldanti, Nicasio Mancini
## Abstract
Respiratory tract infections are a major cause of morbidity and mortality. After the SARS-CoV-2 pandemic, pathogenetic mechanisms leading to more severe outcomes were investigated, including uncontrolled viral replication in the upper airways. This was only partially investigated for other respiratory viruses. We measured mucosal expression of IFN-β1, IFN-λ1, IFN-λ2/3, IL-1β, and IL-6 in patients infected by human metapneumovirus, human rhinovirus, human respiratory syncytial virus or type A influenza virus. A total of 806 nasopharyngeal swabs were collected from patients presenting at emergency departments or hospitalized. Viral load was inferred through cycle threshold determination, whereas cytokine levels were measured through mRNA detection. Each expression pattern was correlated with age, viral load, and specific infecting virus. IFN-β1 and IFN-λ2/3 showed a negative correlation with viral load, while IFN-λ1 and IL-6 exhibited the opposite trend, suggesting increased inflammation with higher viral load. This was more evident in the ≥70-year-old group, with significantly higher IL-6 levels. Higher viral load of potentially more pathogenic viruses was associated with higher IL-6 expression. Cytokine production in the upper respiratory tract is only partially influenced by age per se, with a more relevant role played by viral load and specific infecting virus. In older patients, this response is less coordinated and prone to elicit a proinflammatory response, especially when clinically impacting viruses are involved.
## 1. Introduction
Acute respiratory tract infections (ARTIs) are a major cause of morbidity and mortality worldwide, especially in young children, older adults, and immunocompromised subjects, causing hundreds of thousands of hospitalizations and thousands of deaths annually [1]. The outcome of these infections and the resulting clinical severity are determined by numerous factors, including (i) the effectiveness of the immune response elicited in the upper respiratory tract (URT) mucosa, (ii) the avoidance of an aberrant immune response accountable for collateral damage to tissues and organs, and (iii) the ability of the respiratory viruses (RVs) to elude the immune response in the URT and to spread to the lower respiratory tract (LRT) or to other organs [2,3]. The importance of these concepts was significantly confirmed and brought back to the attention of the scientific community during the recent SARS-CoV-2 pandemic [4].
SARS-CoV-2 caused more than 777 million cases and 7 million deaths worldwide, mostly due to acute respiratory distress syndrome (ARDS) and multiple organ failure induced by a dysregulated immune response [5]. Several papers attempted to clarify the pathogenetic mechanisms leading to its most severe outcomes, suggesting the importance of prompt control of the virus in the URT to avoid the most severe complications of the infection [6]. Older age was one of the factors correlated with more severe SARS-CoV-2 infection, and higher risk was associated to changes in immune and hormone profiles, oxidative stress, mitochondrial dysfunction, altered expression of angiotensin-converting enzyme 2 (ACE2), and differential infection of specific cell types [7,8].
On the viral side, during the pandemic, SARS-CoV-2 also evolved the capacity of evading the innate immune response through several mechanisms, including antagonism of double-stranded RNA (dsRNA) sensors and interference with the mitochondrial antiviralsignaling protein (MAVS) and stimulator of interferon genes (STING) pathways [9]. In particular, the role in influencing the expression and activity of interferons (IFNs) was the subject of much attention, as IFNs are major mediators of the innate antiviral immune response. There are three major families of IFNs: IFN-I (IFN-α and -β, being the most studied), IFN-II (IFN-γ), and IFN-III (IFN-λ1-4). Type I IFNs are expressed by almost all cell types in the body and their strong antiviral effect is paralleled by a significant systemic proinflammatory activity; type II IFN is mainly produced by T lymphocytes and natural killer (NK) cells and is also characterized by a strong proinflammatory action; and type III IFNs are closely related to type I IFNs, but their activity is mostly limited to the mucosal sites with lower systemic effects. The three families of IFNs were shown to play different roles in the course of SARS-CoV-2 infection. IFN-II is significantly upregulated in patients with severe disease and is associated with an increased mortality rate, showing its role in the aberrant late proinflammatory response elicited by the RVs. On the other hand, IFN-I and IFN-III have controversial roles. The kinetics of type I and type III IFN production are important for the final effect of these mediators, with an early response being associated with milder symptoms and less severe COVID-19 [10]. On the contrary, in patients with severe COVID-19, type I and type III IFN production was delayed and paralleled by the abundant secretion of proinflammatory cytokines [11]. The consequent lack of viral control in the URT was then proposed as the main factor exacerbating the aberrant production of inflammatory cytokines (i.e., IL-1β and IL-6), leading to massive lung tissue damage and ARDS [12,13].
These points were partially investigated also for other RVs, even if in epidemiological, immunological, and clinical settings dramatically different than those observed for SARS-CoV-2 [14]. Moreover, in previous studies, the attention was mostly reserved to the systemic innate immune response, which is only partially representative of what occurs at the mucosal level [15,16].
From this perspective, we focused our attention on the local innate immune response against four different RVs with different pathogenic potential that have higher circulation in the population and can lead to emergency department (ED) visits and, in some cases, hospitalization. These include human metapneumovirus (hMPV), human rhinovirus (HRV), human respiratory syncytial virus (hRSV), and influenza type A virus (FLUA). We collected leftover material from nasopharyngeal swabs (NSs) of patients presenting to the ED with respiratory symptoms or eventually hospitalized and measured the expression levels of a selected panel of innate immune molecules, including type I IFN (IFN-β1), type III IFN (IFN-λ1 and IFN-λ2/3), and proinflammatory cytokines (IL-1β and IL-6). The cytokine expression levels were correlated with the viral load (VL) of the infecting RV and with the age of the patient, evidencing for each analyzed cytokine different age-and RV-related trends.
## 2. Materials and Methods
## 2.1. Clinical Samples for Gene Expression Analysis
NSs were collected for diagnostic purposes at three different centers in Lombardy, Northern Italy, from patients presenting at the ED or eventually hospitalized between 2022 and 2024. FLOQSwabs in UTM Universal Transport Medium were used (COPAN, Brescia; Italy). A total of 806 samples were analyzed, including 100 swabs positive for hMPV, 123 for HRV, 237 for hRSV, and 346 for FLUA. Leftover materials were used to perform the study. The characteristics of our study population are described in Supplementary Table S1.
## 2.2. Evaluation of Viral Load
VL was inferred from nasopharyngeal swabs through cycle threshold (Ct) determination with the Allplex™ Respiratory Panel (Seegene, Seoul, Republic of Korea), following the manufacturer's instructions, for samples collected at ASST dei Sette Laghi, Varese, and ASST degli Spedali Civili di Brescia, while a series of homemade real-time RT-PCR assays were used for samples collected at Fondazione IRCCS Policlinico San Matteo, Pavia. In this case, respiratory samples were tested with a panel of laboratory-developed real-time RT-PCR able to detect and quantify the following viruses: influenza virus A, respiratory syncytial virus, human metapneumovirus, and human rhinovirus. Real-time RT-PCR reactions were performed on a Rotor-Gene Q with the Quantifast ® Pathogen PCR+IC Kit (Qiagen, Heidelberg, Germany), according to the manufacturer's instructions. [17][18][19]. A representative number of samples from Fondazione IRCCS Policlinico San Matteo were also tested with the Allplex™ Respiratory Panel and no significant differences were observed in the Ct values detected.
## 2.3. RNA Extraction Protocol and Real-Time PCR for Cytokine Detection
RNA was extracted from NSs using the QIAamp DNA Blood mini kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions. Reverse transcription was performed using the SuperScript™ VILO™ cDNA Synthesis Kit (Thermo Fisher, Waltham, MA, USA), according to manufacturer's instructions. qRT-PCR analysis was then carried out with TaqMan™ Fast Advanced Master Mix for qPCR (Applied Biosystems (Waltham, MA, USA) Cat#444557) using specific Taqman™ Gene Expression Assays from Thermo Fisher. IFN-λ1 (Hs01050642_gH), IFN-λ2/3 (Hs04193047_gH), IFN-β1 (Hs01077958_s1), IL-1β (Hs01555410_m1), and IL-6 (Hs00174131_m1) expression was assessed with respect to the housekeeping gene GAPDH (Hs99999905_m1). All transcripts were tested in triplicate for each sample using the CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA).
## 2.4. Statistical Analysis
Cytokine levels were compared using the Mann-Whitney test for two-group comparisons, whereas Quade's non-parametric ANCOVA was employed for comparisons involving more than two groups to address uneven age distribution in our sample cohort. Correlations were determined using Spearman's correlation. We reported Hedges' g (Hg) and the confidence interval (CI) to measure the effect size in small subgroup correlations. In order to decrease the risk of type I error, we applied false discovery rate (FDR) correction and reported adjusted p-values (Benjamini-Hochberg adjusted p-value). Results were considered statistically significant for p-values < 0.05. Statistical analyses were performed using Prism V8.0 (GraphPad Software Inc., La Jolla, CA, USA) and IBM SPSS Statistics Version 31 (IBM, Armonk, NY, USA).
## 3. Results
## 3.1. Age, Viral Load, and Cytokine Production in URT: Cumulative Analysis
All data were initially analyzed comprehensively, without differentiating for RVs. No strong correlations were observed between the innate immune response and age, possibly suggesting that age is a factor that cannot be considered per se independently of the infecting RVs. Among IFNs, a statistically significant slight increase with age was observed for IFN-β1 (r = 0.085; p = 0.03) and IFN-λ2/3 (r = 0.094; p = 0.02), but not for IFN-λ1 (Supplementary Figure S1A-C). Among tested proinflammatory cytokine, IL-6 featured a weak positive correlation with age (r = 0.07; p = 0.06), whereas, intriguingly, IL-1β showed a slight negative correlation (r = -0.15; p = 0.02) (Supplementary Figure S1D,E).
Next, we analyzed cytokine expression levels and VL, evidencing stronger correlations than those observed with age. We observed that IFN-β1 was positively and strongly correlated with Ct values (r = 0.320; p < 0.001) (Figure 1A) and, therefore, negatively correlated with VL, possibly suggesting a role in limiting viral replication. A similar, even if weaker, trend was also observed for IFN-λ2/3 (r = 0.185; p < 0.001) (Figure 1B). On the contrary, the opposite trend was observed for IFN-λ1 (r = -0.241; p < 0.001) (Figure 1C). Proinflammatory IL-6 transcript levels were positively and coherently correlated with viral load (r= -0.173; p < 0.001) (Figure 1D), suggesting higher inflammation in the presence of less controlled viral replication, independently of the infecting RV. No significant correlation was observed for IL-1β (Figure 1E).
We also generated a correlation matrix to compare the relationship among pairs of variables in our dataset (Supplementary Table S2). A strong and significant positive correlation was observed among all investigated cytokines, evidencing their frequent common elicitation in the URT during a viral infection. However, the strongest correlation was observed between IFN-β1 and IFN-λ2/3 (r = 0.607; p < 0.0001), stressing their frequent common production in the URT and their association with lower VL. Counterintuitively, higher levels of IFN-β1 (2.9 ± 0.29 vs. -1.9 ± 0.36; p < 0.001) and IFN-λ2/3 (-3.5 ± 0.35 vs. -0.1 ± 0.31; p < 0.001) were observed in hospitalized patients compared to ED admitted patients (Supplementary Figure S2). A specific focus was reserved to pediatric patients, considering the important role played by age in the expression of the analyzed parameters. We performed the same analysis on the pediatric population as well and observed the same trends detected in the previous analysis (Supplementary Figure S4A).
To further investigate these correlations among cytokines, age, and VL and to speculate on their possible biological significance, we divided our cohort into two different age groups according to previous studies showing higher risk of severe complications of respiratory infections in patients older than 70 years [12]. Interestingly, the inverse correlation between IFN-β1 and viral load was evident only in the <70 group (r = 0.358; p < 0.001 vs. Ct), whereas it was lost in the older group (Figure 2F). A similar scenario was also observed for IFN-λ2/3 only in the younger cohort (r = 0.203; p < 0.001 vs. Ct) (Figure 2G). On the other hand, a strong direct correlation with VL in the URT was observed for proinflammatory IL-6, especially in the older cohort (r = -0.260; p = 0.006 vs. Ct), with the <70 cohort showing a similar, even if weaker, trend (r = -0.173; p < 0.001 vs. Ct) (Figure 2I). In any case, the average expression levels of IL-6 were significantly higher in the older group (p < 0.001) (Figure 2D), highlighting the presence of higher levels of inflammatory markers when the virus is less controlled in the URT, especially in older subjects. Although no statistical significance was reached in the case of IL-1β, a weak trend in the same direction was observed only in the older cohort (r = -0.170; p = 0.09 vs. Ct) (Figure 2J). Finally, in the case of IFN-λ1, higher levels were associated with higher VL both in the younger (r = -0.212; p < 0.001 vs. Ct) and, even more evidently, in the older cohort (r = -0.466; p < 0.001 vs. Ct) (Figure 2H).
## 3.2. Age, Viral Load, and Cytokine Production in Respiratory Infections: Virus-Specific Analysis
The same data were then analyzed stratifying the samples according to the infecting virus. The expression patterns observed for IFNs and proinflammatory cytokines were similar against the different viruses. However, significantly higher levels of IFN-β1 were detected in samples positive for HRV compared to samples positive for FLUA (p < 0.001), hRSV (p < 0.001) and hMPV (p < 0.001). Analogously, HRV-positive samples featured higher IFN-λ2/3 expression levels compared to FLUA-(p = 0.003) and hRSV-positive samples (p = 0.003). No statistically significant differences were observed for IFN-λ1, IL-6, and IL-1β (Figure 3). The expression levels of interferons were not significantly correlated to the viral load of HRV in the URT (Figure 4F-H), which, together with the average higher levels observed for IFN-β1 and IFN-λ2/3, could suggest the higher propensity of HRV to elicit a protective IFN response even at lower VL. Importantly, always in the case of HRV, higher VL was not paralleled by higher levels of proinflammatory cytokines in the URT (Figure 4I,J). On the contrary, in the case of potentially more pathogenic viruses, a negative correlation trend between VL (i.e., a direct correlation with Ct) and IFN-β1 levels was observed for hMPV (r = 0.200; p = 0.12 vs. Ct) (Figure 4A) and, even more significantly, for FLUA (r = 0.258; p < 0.001 vs. Ct) (Figure 4P) and hRSV (r = 0.394; p < 0.001 vs. Ct) (Figure 4K). Moreover, in the case of FLUA and hRSV, higher VL was correlated with higher levels of IFN-λ1 (r = -0.383; p < 0.001 vs. Ct and r = -0.205; p = 0.008 vs. Ct, respectively) (Figure 4R,M) and, only for FLUA, coherently with the proinflammatory potential of FLUA, with higher levels of IL6 (r= -0.138; p = 0.03 vs. Ct) (Figure 4S).
Subsequently, we generated correlation matrices to compare the relationships between pairs of variables in the population stratified by infecting virus, evidencing a strong correlation among IFNs (especially between IFN-β1 and IFNλ2/3) for the usually less pathogenic HRV compared to other viruses (i.e., FLUA) where a strong correlation between proinflammatory cytokines (IL-6 and IL-1β) was also observed (Supplementary Table S3). Interestingly, after dividing each cohort in two groups according to age, as above, different patterns of cytokine expression were observed. The negative correlation of IFN-β1 with FLUA and hRSV viral load in the URT was observed only in the <70-year-old group (r = 0.333; p < 0.001 vs. Ct and r = 0.391; p < 0.001 vs. Ct, respectively) (Figure 5K,P). A similar pattern was observed in the same age group also for IFNλ2/3 against hRSV (r = 0.189; p = 0.028 vs. Ct) but not in the older group (Figure 5L). Moreover, IFN-λ1 was positively correlated with viral load in hRSV patients aged < 70 years (r = -0.225; p < 0.001 vs. Ct) (Figure 5M) and in FLUA-infected subjects aged both < 70 and ≥70 years (r = -0.333; p < 0.001 vs. Ct and r = -0.614; p < 0.001 vs. Ct, respectively) (Figure 5R).
Analyzing the overall average cytokine expression in the two groups, we observed significantly higher levels of IL-6 elicited by FLUA in the older group compared to the younger group (p < 0.001) (Supplementary Figure S3). Furthermore, we performed the same analysis comparing cytokine expression levels between the pediatric population (<18 years) and the adult population (≥18 years), but we did not find any statistically significant difference (Supplementary Figure S4B).
## 4. Discussion
RV pathogens are the cause of hundreds of thousands of hospital admissions and fatalities annually, especially during the autumn and winter period [1]. The recent SARS-CoV-2 pandemic intensified scientific efforts to delve into the pathogenetic mechanisms of respiratory infections, investigating the role of both viral and host factors in causing the most severe clinical outcomes. Among host-related risk factors, age is certainly very important and its role was clearly demonstrated for SARS-CoV-2 and other RVs. Recent studies also gave possible mechanistic explanations for the reasons behind that, pointing out the importance of prompt containment of the infection in the URT. For example, a recent paper elegantly demonstrated that SARS-CoV-2 features an age-specific tropism for cells in the URT and that, while pediatric cells elicited a strong antiviral response that resulted in limited viral replication, adult cells had altered repair pathways and fibrosis that contributed to viral spread, shedding, and epithelial damage [20]. Another study demonstrated that the age-dependent severity of COVID-19 is due to an impaired IFN response causing a delayed, insufficient, and dysregulated innate and adaptive immune response in aged hosts, thus resulting in a more severe respiratory disease [21].
From this perspective, we aimed to analyze the expression levels of type I IFN (IFN-β1), type III IFNs (IFN-λ1 and IFN-λ2/3), and proinflammatory cytokines (IL-6 and IL-1β) in leftover material obtained from the NSs of 806 subject infected with other RVs (hMPV, HRV, hRSV, and FLUA) with symptoms requiring access to the ED or hospitalization. All of the samples were analyzed both as a single cohort and in two age-dependent groups (<70 and ≥70 years) and categorized according to the infecting virus.
The slight positive correlation of IFN-β1 and IFN-λ2/3, but not IFN-λ1, with age is coherent with what was observed by Gilbert et al. in the URT of SARS-CoV-2-infected patients, confirming in a broader viral context that IFN production is only partially influenced by age [22]. Much more important is the kinetics of mucosal IFN production, that is, the concerted sequential pattern of their expression. A possible speculative interpretation of what was observed is that the direct correlation of IFN-λ1 levels with VL and, conversely, the inverse correlation observed for IFN-β1 and IFN-λ2/3, might be related to the already described sequential IFN production in the URT. In more detail, in more severe cases, such as the ones included in our cohort of patients seeking medical support, the initial production of IFN-λ1 could be insufficient to control viral replication and could be supported by more effective, but also potentially detrimental, higher levels of IFN-β1 and IFN-λ2/3 [23]. However, this could lead to higher expression of proinflammatory cytokines, evident in potentially more pathogenic viruses such as FLUA and less in viruses such as HRV. Experiments performed using serial respiratory samples could confirm this speculation.
The higher levels of IFN-β1 and IFN-λ2/3 in hospitalized patients compared to patients only admitted to the ED could be interpreted as the need for this "second line" of innate immune defense, with all possible associated deleterious consequences. This is consistent with what was already observed regarding the potential negative effects of IFNs beyond their direct antiviral role [10,11,24].
The effect of age and VL on the expression of IFNs and proinflammatory cytokines was evident also when dividing our patient population into <70-and ≥70-year-old subgroups. As already observed also for SARS-CoV-2 [12], in our study most correlations were evident in the younger cohort, suggesting a more VL-balanced IFN and proinflammatory response in the URT of younger patients. On the contrary, in the elderly, we observed a different pattern of IFN and proinflammatory cytokine production, not balanced with VL and thus potentially resulting in a higher risk of complications already described as being mediated by inflammatory monocytes and neutrophils and by impaired resolution of inflammation by macrophages [25]. A possible general explanation of what we observed is that, in younger patients, the mucosal innate response follows a "canonical" pathway, with IFN-λ family members being predominantly expressed in the URT mucosa acting as a first line of defense to suppress the spread of RVs in the LRT and to prevent the initiation of an inflammatory process. When RVs overcome this defense line, type I IFNs come into play, leading to an inflammatory response essential for eradicating viral infections but also capable of causing tissue damage mediated by proinflammatory cytokines like IL-6 and IL-1β [2,26,27].
Our data highlight the obvious role played by the infecting RVs in this scenario. When analyzed from this perspective, HRV-infected patients exhibited significantly higher levels of IFN-β1 and IFN-λ2/3 compared to the other groups, and this could be associated with the higher capability of the innate immune system to control this RV in the URT of infected patients. As mentioned above, when compared to other viruses (i.e., FLUA), HRV showed a lower tendency to induce proinflammatory cytokines such as IL-6 [28]. In line with that, a recent in vitro study demonstrated that HRV elicits higher expression of IFN-β1, IFN-λ2/3, and IFN-λ1 compared to FLUA [29]. However, considering the clinical characteristics of our cohort, including patients seeking medical support, the high expression of more proinflammatory second-line IFNs could also give a molecular explanation of the reason why HRV is emerging as a significant viral pathogen, especially in the elderly, when the first line of defense is overcome.
The higher expression of IFN-β1 and IFN-λ2/3 was particularly evident for FLUA and hRSV hospitalized patients, confirming that their presence is associated with failure of a first line of defense, which is more common for these viruses, and that it might be associated with more severe consequences [30]. Differential behaviors of various viral types and cytokines are already reported in the literature and supported by studies indicating the presence of distinct immune evasion mechanisms [31,32].
The already described double-edge effect of different IFNs emerging from our data further confirms the need for timely administration when considering their possible therapeutic role [33].
Our study has several strengths and weaknesses. The strengths include the large sample size and inclusion of individuals across a wide age spectrum and infected by four different RVs. Moreover, we focused on an outpatient population of subjects presenting to the ED or eventually hospitalized, restricting the population to those who had evident symptoms of infection. The weaknesses of the study include the fact that we considered only statistical correlations between cytokines, without the possibility of testing our findings in an in vitro system and to determine a cause-and-effect relationship between the observed variations. Importantly, the lack of sequential samples prevented a proper understanding of the different kinetics of cytokine production.
## 5. Conclusions
ARTIs are a major cause of morbidity and mortality worldwide, causing hundreds of thousands of hospitalizations and thousands of deaths annually. In this manuscript, we analyzed the local innate immune response against four different RVs with different pathogenic potential by measuring the expression levels of type I and type III IFNs and proinflammatory cytokines in a large and wide-age-spectrum population, observing for each analyzed parameter different age-and RV-related trends. Despite the limits of our work, we think that our observations on such a large cohort of severe patients could be of help in the design of future in vitro studies using different RVs to validate cytokine kinetics in primary nasal epithelial cells from young and aged donors. Other experiments could involve the measurement of protein-level expression of key cytokines to confirm mRNA findings and establish correlations between cytokine levels and clinical severity scores for a comprehensive understanding of molecular predictors of patient outcomes. The impact of
## References
1. Chen, You, Du et al. "Global epidemiological trends in the incidence and deaths of acute respiratory infections from 1990 to 2021"
2. Klinkhammer, Schnepf, Ye et al. (2018) "IFN-λ prevents influenza virus spread from the upper airways to the lungs and limits virus transmission"
3. Troy, Bosco (2016) "Respiratory viral infections and host responses; insights from genomics" *Respir. Res*
4. Lamers, Haagmans (2022) "SARS-CoV-2 pathogenesis" *Nat. Rev. Microbiol*
5. (2025) "COVID-19 Dashboard"
6. Moradi Marjaneh, Challenger, Salas et al. (2023) "Analysis of blood and nasal epithelial transcriptomes to identify mechanisms associated with control of SARS-CoV-2 viral load in the upper respiratory tract" *J. Infect*
7. Farshbafnadi, Kamali Zonouzi, Sabahi et al. (2021) "Aging & COVID-19 susceptibility, disease severity, and clinical outcomes: The role of entangled risk factors" *Exp. Gerontol*
8. Bunyavanich, Do, Vicencio (2020) "Nasal Gene Expression of Angiotensin-Converting Enzyme 2 in Children and Adults" *JAMA*
9. Yao, Foo, Zheng et al. (1869) "Insight into the mechanisms of coronaviruses evading host innate immunity" *Biochim. Biophys. Acta Mol. Basis Dis*
10. Karki, Sharma, Tuladhar et al. (2021) "Synergism of TNF-α and IFN-γ Triggers Inflammatory Cell Death, Tissue Damage, and Mortality in SARS-CoV-2 Infection and Cytokine Shock Syndromes" *Cell*
11. Galani, Rovina, Lampropoulou et al. (2021) "Untuned antiviral immunity in COVID-19 revealed by temporal type I/III interferon patterns and flu comparison" *Nat. Immunol*
12. Sposito, Broggi, Pandolfi et al. (2021) "The interferon landscape along the respiratory tract impacts the severity of COVID-19" *Cell*
13. Blanco-Melo, Nilsson-Payant, Liu et al. (2020) "Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19" *Cell*
14. Bergeron, Hansen, Tripp (2179) "Interferons-Implications in the Immune Response to Respiratory Viruses. Microorganisms 2023"
15. Liu, Zhang, Huang et al. (2020) "Elevated plasma levels of selective cytokines in COVID-19 patients reflect viral load and lung injury" *Natl. Sci. Rev*
16. Lee, Sung, Wu et al. (2023) "Clinical characteristics and differential cytokine expression in hospitalized Taiwanese children with respiratory syncytial virus and rhinovirus bronchiolitis" *J. Microbiol. Immunol. Infect*
17. Piralla, Pariani, Rovida et al. (2011) "Segregation of virulent influenza A(H1N1) variants in the lower respiratory tract of critically ill patients during the 2010-2011 seasonal epidemic" *PLoS ONE*
18. Piralla, Mariani, Rovida et al. (2009) "Frequency of respiratory viruses among patients admitted to 26 Intensive Care Units in seven consecutive winter-spring seasons" *J. Clin. Virol*
19. Giardina, Piralla, Ferrari et al. (2022) "Molecular Epidemiology of Rhinovirus/Enterovirus and Their Role on Cause Severe and Prolonged Infection in Hospitalized Patients" *Microorganisms*
20. Chen, Kelley, Goldstein (2020) "Role of Aging and the Immune Response to Respiratory Viral Infections: Potential Implications for COVID-19" *J. Immunol*
21. Beer, Crotta, Breithaupt et al. (2022) "Impaired immune response drives age-dependent severity of COVID-19" *J. Exp. Med*
22. Gilbert, Lefeuvre, Preisser et al. "Age-Related Expression of IFN-λ1 Versus IFN-I and Beta-Defensins in the Nasopharynx of SARS-CoV-2-Infected Individuals"
23. Lazear, Schoggins, Diamond (2019) "Shared and Distinct Functions of Type I and Type III Interferons" *Immunity*
24. Domizio, Gulen, Saidoune et al. (2022) "The cGAS-STING pathway drives type I IFN immunopathology in COVID-19" *Nature*
25. Feng, Balint, Poznanski et al. (2021) "Aging and interferons: Impacts on inflammation and viral disease outcomes" *Cells*
26. Galani, Triantafyllia, Eleminiadou et al. (2017) "Interferon-λ Mediates Non-redundant Front-Line Antiviral Protection against Influenza Virus Infection without Compromising Host Fitness" *Immunity*
27. Velazquez-Salinas, Verdugo-Rodriguez, Rodriguez et al. (1057) "The role of interleukin 6 during viral infections" *Front. Microbiol*
28. Hung, Zhang, To et al. (2017) "Unexpectedly higher morbidity and mortality of hospitalized elderly patients associated with rhinovirus compared with influenza virus respiratory tract infection" *Int. J. Mol. Sci*
29. Dissanayake, Schäuble, Mirhakkak et al. "Comparative Transcriptomic Analysis of Rhinovirus and Influenza Virus Infection"
30. Major, Crotta, Llorian et al. (2020) "Type I and III interferons disrupt lung epithelial repair during recovery from viral infection" *Science*
31. Van De Sandt, Kreijtz, Rimmelzwaan (2012) "Evasion of influenza A viruses from innate and adaptive immune responses" *Viruses*
32. Ouyang, Liao, Hu et al. (2022) "Innate Immune Evasion by Human Respiratory Syncytial Virus" *Front. Microbiol*
33. Broggi, Ghosh, Sposito et al. (2020) "Type III interferons disrupt the lung epithelial barrier upon viral recognition" *Science*
34. "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=PMC12532903&blobtype=pdf | # Correction: HTLV1 infection and long term association with liver function and lipid indices; 10 years' follow-up
Majid Ghayour-Mobarhan, Farzam Kamrani, Amirhossein Esfandiari, Hojjat Ghahvechi, Samaneh Abolbashari, Zahra Mashkat, Habibollah Esmaily, Susan Darroudi
## Abstract
Following publication of the original article [1], we were notified that the third author's name was incorrectly spelled as "Hojat ghahvechi" instead of "Hojjat Ghahvechi".The original article has been corrected.
## References
1. Kamrani (1067) "-z. BMC Infectious Diseases The original article can be found" *BMC Infectious Diseases* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12585739&blobtype=pdf | # DNA binding and cleavage mechanism of DNA topoisomerase VI, an evolutionary counterpart of Spo11/Wadjet/Gabija systems
Daniel Richman, Timothy Wendorff, Fahad Rashid, Curtis Beck, Matthew Baker, Jonathan Fogg, Lynn Zechiedrich, James Berger
## Abstract
Type II topoisomerases resolve DNA supercoiling and chromosome entanglements. Type IIB topoisomerases, exemplified by Top6, are used by plants and archaea to support endoreduplication and cell proliferation, respectively; homologs of Top6 subunits initiate meiotic recombination in eukaryotes and constitute the nuclease portion of Wadjet/Gabija bacterial plasmid defense systems. To understand how such factors act upon native substrates, we determined cryoEM structures of M. mazei Top6 bound to supercoiled DNA in cleaved and uncleaved states. The structures show that Top6 deforms DNA into an ~80bp loop, explaining its preferential association with supercoiled substrates. Five holoenzyme conformations reveal new interactions and motifs that recognize bent DNA and promote cleavage, as well as an unanticipated tension sensor that couples ATPase disposition to cleavage state activation. Our observations offer unprecedented insight into interdependent structural changes that regulate DNA recognition and cleavage in type II topoisomerases and related systems. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12159595&blobtype=pdf | # A look back at the Virology Departmental Days of the Institut Pasteur (Le Touquet, May 13-15, 2024)
Rubén González, Cassandra Koh, Bérangère Virlon, Sarah Merkling, Jean-Pierre Vartanian
## Abstract
This event brought together virologists across the department for a gathering of scientific exchange and collaboration. Placing young researchers in the spotlight, the meeting featured 25 talks, 31 posters, and a keynote address. In this meeting report, we aim to introduce the department, present its current activities, and communicate its vision.
The Institut Pasteur Virology DepartmentEstablished in 1888, the Institut Pasteur is a private, non-profit foundation built on four mission pillars-research, education, public health, and innovation. These are fulfilled through the involvement of committed people, internal dedicated teams (research teams, scientific departments, technological platforms, reference centres for infectious disease monitoring, and support services and facilities), the Pasteur Network of > 30 institutes worldwide, and close cooperative ties with
department has been at the forefront of the fight against viral diseases through emergence prevention and transmission mitigation. The department, with a focus on major viral pathogens such as human immunodeficiency virus (HIV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), actively contributes to tackle research on emerging viral diseases at the Institut Pasteur. In addition, the department hosts two National Reference Centres that conduct population surveillance of viral diseases, including haemorrhagic fevers and respiratory viral infections such as influenza and COVID-19, as well as a WHO Collaborating Centre for polioviruses and non-polio enteroviruses. The Virology Department is also extensively involved in education, participating in nine infectious disease courses and seven massive open online courses (MOOCs).
The breadth of research foci within the department ( Figure 1 ) includes virus pathogenesis, dissemination and transmission, host population genetics and phylogeny, molecular and cellular mechanisms of the viral replication cycle, virus epidemiology, surveillance and control strategies, virus structural biology and evolution, and virus-host interactions.
Our department sits at the cutting edge of fundamental research, providing discoveries that can be translated into solutions to address current viral threats to global public health.
department's research is inherently interdisciplinary and continuously evolving. The following subsections outline the teams' main goals and the research presented by their members at JDV2024. Works that were published at the moment of writing this report are cited.
## Pathogenesis
Jean-Pierre Vartanian is the Principal Investigator (PI) of the Virus and Cellular Stress team. This team (i) deciphers the mitochondrial network dysfunction caused by cellular stress (viral, genotoxic, oxidative, and hypoxic) and (ii) identifies sensing molecules leading to the production of interferon, using released mitochondrial DNA as a biomarker. Théo Defresne presented his work on the impact of mitochondrial network disruption following SARS-CoV-2 Alpha and Omicron infections.
The Hepacivirus-Host Interactions team (PI: Annette Martin) links hepatitis C virus (HCV) strains and pathological disorders. Angeliki Anna Beka showed that Rattus norvegicus isolate 1 induces HCV-like pathology in rats, using this virus to successfully set up an animal model to study hepatitis-like disease.
The HIV, Inflammation, and Persistence team (PI: Michaela Müller-Trutwin) deciphers the mechanisms of protective immunity of natural killer (NK) cells to find novel immunotherapeutic strategies against viral diseases, with a primary focus on HIV. Aurelio Orta-Resendiz shed light on the role of unconventional CD8 + T cells in inflammation within individuals living with HIV-2 (Orta-Resendiz et al., 2023 ). Emma Beaumont presented evidence on the role of adaptor proteins in regulating NK cell activity during persistent SARS-CoV-2 infection.
## Molecular and cellular mechanisms of the viral cycle
The Viral Replication and Nucleic Acids team (PI: Marc Lavigne) studies how G4s (non-canonical DNA/RNA structures) regulate the replication of several DNA and RNA viruses. Marc Lavigne talked about the role of guanine quadruplexes in HIV-1 (Lista et al., 2023 ) and SARS-CoV-2 replication (Lavigne et al., 2021 ).
The Virus and Immunity team (PI: Olivier Schwartz) focuses on cellular and molecular aspects of the replication of HIV and SARS-CoV-2. William Bolland shared new insights into viral cytopathy linked to SARS-CoV-2-induced cell-to-cell fusion and the emerging role of lipid rafts in the phenotype (Bolland et al., 2025 ). In addition, Jeanne Postal lifted the curtain on the mechanisms behind the replication and entry of the seasonal human coronavirus OC43.
The Advanced Molecular Virology team (PI: Francesca Di Nunzio) aims to untangle the mechanism underlying HIV-1 membraneless organelle biogenesis and its role in viral reverse transcription, replication, and latency. Selen Ay mapped the nuclear landscape remodeling induced by HIV-1 infection and described the biogenesis of membraneless organelles as a site of nuclear reverse transcription (Scoca et al., 2022 ;Ay and Nunzio, 2023 ).
The Molecular Mechanisms of Multiplication of Pneumoviruses team (PI: Marie-Anne Rameix-Welti) seeks to illuminate the multiplication processes of the respiratory syncytial virus from RNA replication to virion release. Of note, the virus utilizes cellular microtubules to transport viral ribonucleoprotein complexes to assembly and budding sites (Cosentino et al., 2022 ).
## Transmission and One Health
The Epidemiology and Physiopathology of Oncogenic Viruses team (PI: Antoine Gessain) is dedicated to the epidemiology, physiopathology, and immunology of retroviruses, herpes viruses, and some emerging viruses. Manon Curaudeau presented her investigative trail in identifying the African mammal reservoir host for the monkeypox virus, the squirrel Funisciurus anerythrus (Curaudeau et al., 2023 ). Jeanne Pascard demonstrated the potential of mother-to-child transmission of the yellow fever virus during breastfeeding.
The Arboviruses and Insect Vectors team (PI: Anna-Bella Failloux) seeks to understand the mechanisms of arboviral emergence by developing three complementary lines of research: (i) offering scientific expertise for public health actions, (ii) deciphering the molecular mechanisms that limit/promote arbovirus infections in mosquitoes, and (iii) developing alternative strategies to complement insecticide-based control measures. Christian Mitri presented his work on fungal priming of mosquito immunity to reduce their vector competence.
The Insect-Virus Interactions team (PI: Louis Lambrechts) investigates the causes (evolutionary drivers, genetic and non-genetic factors, and mechanisms) and consequences (virus transmission dynamics and evolution) of natural variation in the ability of mosquitoes to carry arboviruses. Thomas Vial presented work on cellular and metabolic signatures of resistance to dengue virus in the mosquito Aedes aegypti (Vial et al., 2024 ). Théo Maire introduced a new assay to continuously monitor mosquito flight activity over weeks after an infectious blood-meal, which he used to test the impact of dengue virus infection on the mosquito daily flight rhythm activity.
The Viruses and RNA Interference team (PI: Carla Saleh) studies insect immunity with a focus on factors that enable insects to cope with viral infections and manipulation of insect antiviral immunity to prevent disease spread in humans. Mauro Castelló-Sanjuán presented work on how the fruit fly (an insect model system) survives and maintains persistent RNA viral infections using endogenous retrotransposons to generate viral DNA. Rubén González showed that the bacteriome composition of the fruit fly modulates host mortality upon enteric viral infection.
## Virus epidemiology and evolution
The Virus Sensing and Signaling team (PI: Nolwenn Jouvenet) has its research axis in enterovirus surveillance, epidemiology, and evolution. Felix Streicher presented how type I and type III interferon responses protect against tick-borne orthoflaviviruses through IFI6.
The Pathogen Discovery team (PI: Nolwenn Dheilly) aims to discover, characterize, and demonstrate the attributability of new or unexpected infectious agents in clinical syndromes of unknown etiology. Sarah Temmam described the circulation of sarbecoviruses in horseshoe bats in Southeast Asia, which are hosts for a high diversity of viruses, including close ancestors of SARS-CoV-2 (Temmam et al., 2023 ).
The Evolutionary Genomics of RNA Viruses team (PI: Etienne Simon-Lorière) uses viral genomics to study the evolutionary processes of RNA viruses, with a focus on respiratory viruses and arboviruses. Artem Baidaliuk applied a metatransciptomics approach to profile insect-specific viruses in mosquitoes, delving into the phylogeography of viruses and virus-host phylogenetic co-divergence.
## Surveillance and control strategies
The Viral Reservoirs and Immune Control team (PI: Asier Sáez-Cirión) aims to develop tools and strategies for HIV remission. Dr Sáez-Cirión presented how CD8 + T cell reprogramming may constitute a new therapeutic opportunity with improved persistence and durability of the response (Sáez-Cirión and Sereti, 2021 ).
## Viral structure
The Structural Biology of Infectious Diseases team (PI: Pablo Guardado-Calvo) seeks to develop novel therapies, based on a better understanding of the entry mechanisms of large DNA viruses (poxviruses and related viruses) and the mechanisms of action of antibodies that clear the infection. Dr Guardado-Calvo spoke about the use of structural virology to understand the mechanisms of antiviral drugs against poxviruses.
The Structural Virology team (PI: Félix A. Rey) focuses on enveloped viruses, including both zoonotic (flaviviruses, alphaviruses, bunyaviruses, and retroviruses) and non-zoonotic (herpesviruses) ones. They study (i) the structure of enveloped viral proteins to understand mechanisms of cell entry and how envelope glycoproteins are neutralized by antibodies and (ii) the effect of non-structural proteins on the replication of mosquito-transmitted viruses. Max Baker resolved the first structure of a human rotavirus, gaining insights into structural mechanisms underlying rotavirus infection. Ignacio Fernández presented work on the structural basis of TMPRSS2 zymogen activation and recognition by the HKU1 seasonal coronavirus (Fernández et al., 2024 ).
## Virus-host interactions
In his keynote lecture, John Gross (University of California San Francisco) presented his work on evolutionary pressures that shape virus-host interactions, focusing on the viral antagonism of innate immunity. The lecture focused on the arms race between the human APOBEC3G enzyme and HIV-1 Vif (Li et al., 2023 ).
From the RNA Biology of Influenza Viruses team (PI: Nadia Naffakh), Maud Dupont presented her investigation of the RNA-binding protein interactome with influenza A viral mRNAs. Her findings highlighted TDP-43, a protein recruited by the viral polymerase that orchestrates the assembly of viral mRNAs into infectious viral particles (Dupont et al., 2024 ). Catherine Isel described a proof-of-concept for peptidebased inhibition of protein-protein interactions as a therapeutic strategy against influenza viruses. The Interactomics, RNA and Immunity team (PI: Caroline Demeret) develops methods for detecting direct, binary interactions between pairs of proteins. Mikaël Attia constructed a 3D contactome network of SARS-CoV-2 and human proteins to better elucidate virus-host interactions.
## Translating discoveries to solutions
The department hosts two Joint Research laboratories in close collaboration with companies that bridge the gap between academic and biotechnological research. Oncovita aims to develop a therapeutic anti-cancer vaccine using technology derived from the measles Schwarz vaccine virus. From Oncovita, Aleksandr Barinov presented results on a new therapeutic vaccine with promising preclinical immuno-oncolytic activities. TheraVectys, represented by Pierre Charneau, introduced lentiviral vector technology for immuno-onco vaccine applications as well as therapeutics for emerged and emerging infectious diseases.
## Beyond science
In addition to research sessions, the JDV2024 held sessions dedicated to sustainability, scientific integrity, and funding acquisition.
## Future perspectives
The Virology Department will continue its relentless efforts to understand and combat viruses affecting human health. The department aims to perform excellent research while also strengthening interdisciplinary partnerships with the currently expanding Pasteur Network, international partners, and industry stakeholders.
We strive to extend our knowledge in fundamental virology and develop new therapies, in particular medicinal chemistry, vaccines, and diagnostic tools for future viral threats that will emerge on the planet.
We will build on the lessons learned from previous epidemics and pandemics, particularly those caused by respiratory pathogens. It is a challenge to accelerate preparedness for pandemics and emerging threats on a global scale, as effective prevention requires solid planning and coordinated action. The Virology Department of the Institut Pasteur will continue to be a key player in the fight against viral diseases through cutting-edge fundamental research developed across the different laboratories and translating groundbreaking discoveries into tangible benefits for public health and emergence preparedness.
Collective mobilization will inevitably be key to success, and the Virology Department will continue to follow the path paved by all its previous esteemed members.
## Final remarks
Following three days of intensive scientific exchange, engaging discussions, delightful meals, networking activities, and beachfront promenades, the department members have strengthened professional relationships-a crucial aspect for fostering excellent science. We eagerly look forward to our next biennial meeting.
[The JDV2024 organizing committee is grateful to the ANRS-MIE and to numerous commercial sponsors for their participation and generous funding support: Agilent, Eurofins, Eurogentec, Fisher Scientific, InvivoGen, Merck, Novogene, ThermoFisher, and VectorBuilder. The Virology Department acknowledges Christine Letellier for administrative assistance and the Institut Pasteur for its support. We acknowledge the commissioned work of Bertsy Goic ( http://www.drawinscience.fr) for the figure. We thank Louis Lambrechts and Carla Saleh for thoroughly reading and commenting on the text. We also acknowledge all the members of the department for stimulating discussions, insightful presentations, and enriching interactions.]
## References
1. Ay, Di Nunzio (2023) "HIV-induced CPSF6 condensates" *J. Mol. Biol*
2. Bolland, Marechal, Petiot (2025) "SARS-CoV-2 entry and fusion are independent of ACE2 localization to lipid rafts" *J. Virol*
3. Cosentino, Marougka, Desquenes (2022) "Respiratory syncytial virus ribonucleoproteins hijack microtubule Rab11 dependent transport for intracellular trafficking" *PLoS Pathog*
4. Curaudeau, Besombes, Nakouné (2023) "Identifying the most probable mammal reservoir hosts for monkeypox virus based on ecological niche comparisons" *Viruses*
5. Dupont, Krischuns, Giai Gianetto (2024) "The RBPome of influenza A virus NP-mRNA reveals a role for TDP-43 in viral replication" *Nucleic Acids Res*
6. Fernández, Saunders, Duquerroy (2024) "Structural basis of TMPRSS2 zymogen activation and recognition by the HKU1 seasonal coronavirus" *Cell*
7. Lavigne, Helynck, Rigolet (2021) "SARS-CoV-2 Nsp3 unique domain SUD interacts with guanine quadruplexes and G4ligands inhibit this interaction" *Nucleic Acids Res*
8. Li, Langley, Azumaya (2023) "The structural basis for HIV-1 Vif antagonism of human APOBEC3G" *Nature*
9. Lista, Jousset, Cheng (2023) "DNA topoisomerase 1 represses HIV-1 promoter activity through its interaction with a guanine quadruplex present in the LTR sequence" *Retrovirology*
10. Orta-Resendiz, Petitdemange, Schmutz (2023) "Deep phenotyping characterization of human unconventional CD8 + NKG2A/C + T cells among T and NK cells by spectral flow cytometry" *STAR Protocols*
11. Sáez-Cirión, Sereti (2021) "Immunometabolism and HIV-1 pathogenesis: food for thought" *Nat. Rev. Immunol*
12. Scoca, Morin, Collard (2022) "HIVinduced membraneless organelles orchestrate post-nuclear entry steps" *J. Mol. Cell Biol*
13. Temmam, Montagutelli, Herate (2023) "SARS-CoV-2-related bat virus behavior in human-relevant models sheds light on the origin of COVID-19" *EMBO Rep*
14. Vial, Lopez-Maestre, Couderc (2024) "Single-cell transcriptional landscapes of Aedes aegypti midgut and fat body after a bloodmeal" *bioRxiv* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12246178&blobtype=pdf | # Author Correction: An RBD-Fc mucosal vaccine provides variant-proof protection against SARS-CoV-2 in mice and hamsters Check for updates
Yanjun Zhang, Yan Wu, Meng-Qian Zhang, Haiyue Rao, Zhaoyong Zhang, Xiangyue He, Yiwen Liang, Raoqing Guo, Yaochang Yuan, Jing Sun, Helen Duyvesteyn, Elizabeth Fry, David Stuart, Jingxian Zhao, Xiaoyan Pan, Shu-Lin Liu, Jincun Zhao, Jiandong Huo |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12817941&blobtype=pdf | # VILMIR is a trans-acting long noncoding RNA that enhances the host interferon response in human epithelial cells
Kristen John, Ethan Smith, Alexandra Istishin, Nasif Mahmood, Kayleigh Diveley, Tammy Tollison, Susan Carpenter, Xinxia Peng
## Abstract
Long noncoding RNAs (lncRNAs) have been found to play significant regulatory roles within antiviral and immune responses. We previously identified the novel lncRNA virus-inducible lncRNA modulator of interferon response (VILMIR), which was found to broadly regulate the host transcriptional response to interferon-beta (IFN-β) treatment in A549 human lung epithelial cells. Here, we investigated the mechanism by which VILMIR regulates the host interferon response in trans by identi fying interacting proteins and gene regulatory networks of VILMIR. Through an RNA pull-down assay, we found that VILMIR interacted with both nuclear and cytoplasmic proteins in vitro, including the transcriptional regulators FUBP1 and PUF60 in the nucleus, as well as the antiviral proteins IFIT1 and IFIT3 and the aminoacyl-tRNA synthetases QARS1 and KARS1 in the cytoplasm. In addition, we found that the overexpression of VILMIR in A549 cells resulted in an overall enhancement of host interferon response genes and identified a core set of interferon-stimulated genes that were consistently regulated by VILMIR knockdown and overexpression. Finally, we proposed several possible mechanisms by which VILMIR may interact with the identified proteins to regulate the interferon response, such as by interacting with FUBP1 and PUF60 in the nucleus to regulate host transcription in trans or by interacting with the IFIT proteins and aminoacyl-tRNA synthetases in the cytoplasm to regulate translation. IMPORTANCE Despite thousands of long noncoding RNAs (lncRNAs) being differentially expressed after immune responses and viral infections, there is limited knowledge on their individual functions in these contexts. We previously identified a novel lncRNA, VILMIR, that was found to be an interferon-stimulated gene that regulated the host transcriptional response to interferon-beta treatment in human epithelial cells. Here, we investigated the mechanism by which VILMIR regulates the interferon response. Through in vitro studies, we identified several nuclear and cytoplasmic proteins that interact with VILMIR, including proteins involved in transcriptional and translational regulation. In addition, we demonstrated that the overexpression of VILMIR results in an enhance ment of host interferon response genes, supporting our hypothesis that VILMIR plays an activating role in the host interferon response. Finally, we propose several potential models for the mechanism of VILMIR, providing a foundation for the investigation of VILMIR as a novel therapeutic target in antiviral immunity.
nucleotides in length with low translational potential (5). The recent GENCODE V47 release estimates 35,934 lncRNA genes in the human genome (6). However, despite the large number of annotated lncRNAs, their functions are still widely unknown.
Individual lncRNAs have been identified to have significant functions within biological processes such as cell development (7), cancer (8), and inflammation (9) by regulating processes such as gene transcription and protein translation (10). There is also growing evidence that lncRNAs play important roles within antiviral and immune responses (11). For example, a recent study in 2023 identified that the overexpres sion of an lncRNA, LncRNA#61, inhibited influenza A virus (IAV) replication in human cells and even reduced viral replication in vivo after lipid nanoparticle-encapsulated delivery of LncRNA#61 in mice (12). As regulatory RNAs such as lncRNAs have low translational potential, they often function by interacting with RNA-binding proteins (RBPs) to regulate transcription (13) or act as scaffolds for protein interactions (14). While many lncRNAs have been found to act as cis regulators by regulating the expression of neighboring protein-coding genes (15), lncRNAs have also been found to function as trans regulators by regulating transcription on different chromosomes (16). Some lncRNAs can act in both cis and trans, such as lincRNA-Cox2 that was found to regulate its neighboring gene, Ptgs2, as well as a subset of immune genes in trans using mouse models (17). In addition, a single lncRNA can have multiple functions in different cellular compartments, such as PYCARD-AS1 that facilitates DNA methylation at the PYCARD promoter in the nucleus, as well as interacts with PYCARD mRNA in the cytoplasm to inhibit ribosome assembly (18). Therefore, understanding the RBP interactions of an lncRNA can help elucidate how it functions within the cell.
We previously identified a novel lncRNA named virus-inducible lncRNA modulator of interferon response (VILMIR) that was found to regulate the host transcriptional response to both interferon-beta (IFN-β) treatment and IAV infection in A549 human lung epithelial cells (19). We found that VILMIR did not regulate transcription of its neighboring protein-coding genes but rather had a broad transcriptional regulation; however, the exact mechanism of this regulation was not explored. Therefore, in this study, we aimed to identify interacting proteins and gene regulatory networks of VILMIR to better understand its molecular interactions and how it regulates the interferon response in trans. Using an RNA-pull down assay, we found that VILMIR interacts with several proteins in A549 nuclear and cytoplasmic lysates, including the transcriptional regulators FUBP1 and PUF60, as well as antiviral proteins IFIT1 and IFIT3 and aminoacyl-tRNA synthetases (ARSs) QARS1 and KARS1. In addition, we overexpressed VILMIR in A549 cells and found that the overexpression resulted in an overall enhancement of host interferon response genes, supporting our hypothesis that VILMIR plays an activating role in the host interferon response. By combining RNA-seq analyses from both VILMIR knockdown (KD) and overexpression studies, we further identified a core set of genes that are consistently regulated by VILMIR perturbation in trans. Finally, we proposed several potential models of VILMIR function, suggesting that VILMIR may function in both the nucleus and the cytoplasm to regulate host interferon responses.
## MATERIALS AND METHODS
## Cell culture
The human cancer cell line, A549 lung epithelial (CCL-185), was purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). Human embryonic kidney (HEK) epithelial 293 FT cells were ordered from Invitrogen (ThermoFisher Scientific). A549 cells were maintained in F-12K media with 10% fetal bovine serum (FBS). HEK 293 FT cells were maintained in DMEM media plus 1% GlutaMAX and 10% FBS. All cell lines were kept at 37°C in a 5% CO 2 incubator and maintained in culture, as recommended by the ATCC.
## Protein extraction
In order to collect protein lysates for subsequent RNA pull-down assay, confluent T182 flasks of A549 cells were washed with 1X Dulbecco's phosphate-buffered saline and treated with fresh A549 media containing 1 ng/mL human IFN-β recombinant protein (R&D Systems 8499IF010) for 6 hours. Nuclear and cytoplasmic protein lysates of the cells were extracted using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific) according to the manufacturer's instructions. Relative protein quantification was determined using absorbance at 280 nm by comparison with a bovine serum albumin (BSA) standard curve.
## RNA pull-down assay and mass spectrometry
Full-length VILMIR (Table S1) or its antisense sequence was cloned into the pGEM-3Z vector (Promega) downstream of the T7 promoter using EcoRI and BamHI restriction enzyme sites. The plasmids were linearized with BamHI on the 3′ end in order to facilitate in vitro transcription. Biotinylated RNA probes were in vitro-transcribed by the T7 RNA polymerase using the TranscriptAid T7 High Yield Transcription Kit (Thermo Scientific) and a 1:39 molar ratio of Biotin-16-UTP (ApexBio Technology) to standard UTP. Subsequent RNA was purified using the RNeasy Plus Mini Kit (Qiagen) with a genomic DNA eliminator column, and the size was determined with an Agilent Bioanalyzer or TapeStation 4150.
The pull-down assay was adapted for RNA from Springer Protocols (20). Briefly, 1.5 mg of prewashed Dynabeads M-280 Streptavidin (Invitrogen) was incubated with 25 µg of VILMIR sense or antisense RNA probes at room temperature with agitation for 30 min. Excess biotinylated RNA probe was removed from the bead-probe complex by several washes. Approximately 3.6 mg of nuclear or 5.3 mg of cytoplasmic protein lysate from A549 cells was incubated with 12.5 µg of antisense bead:probe complexes at 4°C, with agitation for 30 min as a preclearing step. Precleared lysates were then incubated with VILMIR sense or antisense bead:probe complexes with 50 µg tRNA competitor (Thermo Scientific) and 300 U of SUPERase•In RNase Inhibitor (Invitrogen) at 4°C with agitation for 1 h. Bead:probe:protein complexes were washed three times in washing buffer, eluted in water at 70°C for 5 min, and boiled in Laemmli buffer at 95°C for 10 min. The pull-down assay was performed in triplicate for both sense and antisense reactions.
The supernatant was run on a 6% SDS-PAGE gel for approximately 10 min, and then the gel was stained with Coomassie blue and destained to excise and prepare the samples for protein identification by mass spectrometry. Mass spectrometry was performed by the BIDMC-Harvard Mass Spectrometry Facility and Asara Laboratory using the Thermo Scientific QExactive HFx Orbitrap nano HR-LC-MS/MS following in-gel digestion of proteins with Trypsin/LysC.
To identify VILMIR sense-specific binding proteins in the nucleus or cytoplasm, we first averaged the peptide spectrum counts in the sense and antisense replicates and then calculated an approximate fold-change (FC) value by taking the difference between the sense and antisense peptide counts (sense/antisense). Proteins that were evenly distributed between the sense and antisense samples (i.e., FC of 1), enriched in the antisense samples (FC <1), or whose replicates had inconsistent counts were eliminated. VILMIR sense-specific bound proteins were identified as those proteins that had protein peptide spectrum counts in all three replicates of the sense RNA but none in the antisense replicates (unique binding), or whose peptide spectrum counts were enriched (FC >1) in the sense replicates over the antisense replicates (enriched binding). These remaining proteins were narrowed down further by removing proteins with an FC <2 in the sense replicates, as well as focusing on proteins with known associations with interferon and antiviral responses.
## Western blotting
Identified proteins from mass spectrometry were confirmed by Western blot after a pull-down assay, as described above. The starting protein lysate for VILMIR sense and antisense pull-down assay was standardized before pull-down. After pull-down, an equal volume of the eluted protein from the sense and antisense pull-down was loaded into 10% SDS-PAGE gels and transferred to polyvinylidene difluoride (PVDF) membranes (Invitrogen). Membranes were blocked for either 1 h at room temperature or 4°C overnight in 1× Tris-Buffered Saline (TBS) with 1% (wt/vol) casein and then incubated in primary antibody diluted in blocking buffer for 1 h at room temperature. The following primary antibodies were used: anti-FUBP1 1:1,000 (Proteintech, 24864-1-AP), anti-PUF60 1:1,000 (Proteintech 10810-1-AP), anti-PCNA 1:5,000 (Proteintech, 10205-2-AP), anti-IFIT1 1:500 (Cell Signaling Technology, #14769), anti-IFIT3 1:2,000 (Proteintech, 15201-1-AP), anti-GlnRS (or QARS1) 1:2,000 (Proteintech 12645-1-AP), anti-KARS 1:1,000 (Proteintech 14951-1-AP), and anti-GAPDH 1:5,000 (Proteintech, 10494-1-AP). Membranes were rinsed 2× and washed 2× for 5 min each in TBS buffer with 0.05% Tween 20 (TBS-T) and then incubated in Goat anti-Rabbit IgG (H + L) secondary antibody, HRP conjugate (Invitrogen #31460) diluted 1:5,000 in blocking buffer for 30 min at room temperature. Membranes were rinsed 3× and washed 3× for 5 min each in TBS-T buffer and then detected by chemiluminescence using either Pierce ECL Substrate (Thermo Scientific) for the FUBP1 and PCNA blots or SuperSignal West Atto Ultimate Sensitivity Substrate (Thermo Scientific) for the remaining blots, according to the detection limits of the protein. The blots were visualized using a Bio-Rad ChemiDoc MP Imaging System. Densitometry analysis was performed using ImageJ software where applicable. Briefly, the background was subtracted, and the density of each band was measured. In order to estimate a relative FC between the sense and antisense protein interactions, the sense and antisense bands were then normalized to their input band, and the average FC of three independent replicates was calculated. With FUBP1, PUF60, and QARS1 that displayed several bands, all bands were included so as to not bias the analysis. FUBP1 and PUF60 were not included in the densitometry analysis as the antisense band was too low to obtain an accurate density.
## Interferon treatment
Following the same methods as 19, A549 cells were seeded overnight between 150,000 and 175,000 cells per well in 12-well plates in 1.5 mL media. The following day, the cell monolayer was washed with 1× DBS and treated with fresh A549 media with or without human IFN-β recombinant protein at the indicated concentrations. All cells were harvested at 6 hours after treatment according to the TRIzol Reagent User Guide (Invitrogen).
## RNA isolation and quantitative PCR
Total RNA was isolated from cells following the TRIzol isolation method (Invitrogen) and quantified using NanoDrop spectrophotometry. One microgram of RNA was reversetranscribed into cDNA using the QuantiTect Reverse Transcription Kit (Qiagen) contain ing both oligo-dT and random primers. Quantitative PCR (qPCR) was performed on the cDNA using PowerUp SYBR Green Master Mix (Applied Biosystems). Relative expression of the indicated RNAs was determined using the ΔΔCt method with GAPDH as an endogenous control. Statistical analysis of significance was performed in JMP Pro 16 software (SAS Institute Inc., Cary, NC). The primer sequences used in this study are as follows: GAPDH F: GGTATCGTGGAAGGACTCATGAC; GAPDH R: ATGCCAGTGAGCTTCCCGTT CAG (21); VILMIR F: GCTCCACCCTGAAAGTC; VILMIR R: CTACACAGTGCTGAGGAAA (19).
## Plasmid construction and overexpression of VILMIR
The pSico bidirectional expression vector was a gift from Susan Carpenter and described in 22. Full-length VILMIR was cloned into the pSico vector with an EF1a promoter expressing zeocin resistance and GFP as a selection marker, using PspXI and NotI restriction enzyme sites. The sequence was confirmed by Sanger sequencing. To produce lentivirus, HEK-293FT cells were co-transfected with the pSico vector expressing VILMIR and lentiviral vectors, psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259) using the Lipofectamine 3000 Reagent (Invitrogen). In addition, an empty pSico vector was transfected as a negative control. The viral supernatant was collected 72 h post-transfec tion and filtered through a 0.22 µM syringe filter. A549 cells were transduced with the lentivirus, and stable integrants were sorted based on GFP expression at the UNC Flow Cytometry Core Facility (Chapel Hill, North Carolina) using a Becton Dickinson FACSAria II. The successful overexpression of VILMIR was confirmed by RT-qPCR.
## cDNA library construction, RNA-sequencing, and Ingenuity Pathway Analysis (IPA)
mRNA sequencing was performed in biological triplicate in A549 VILMIR-overexpressing and control cell lines treated with mock or either 1 ng/mL or 10 ng/mL human IFN-β for 6 h. Total RNA was isolated from cells following the TRIzol isolation method (Invitrogen). All samples were quantified and assayed to confirm a minimum RNA integrity number of at least 9.7 using an Agilent TapeStation 4150. Next, 500 ng of total RNA per sample underwent mRNA capture and was then fragmented at 94°C for 6 min. Sequencing libraries were prepared according to the manufacturer's protocol using 11 cycles of final amplification (KAPA mRNA HyperPrep Kit, catalog no. KK8580 and KAPA UDI Adapter Kit, catalog no. KK8727). Libraries underwent QC prior to sequencing using an Agilent TapeStation 4150. Next-generation sequencing was performed on a Complete Genomics DNBSEQ-G400C (150 bp paired end) to a targeted depth of ~20 million reads per sample. The sequencing data from VILMIR KD in 19 were also incorporated.
Complete genomics RNA-seq reads were mapped against the Hg38 using STAR version 2.7.9 a (23). Custom STAR parameters were set as follows: limitOutSAMoneRead Bytes: 1,000,000, outSAMprimaryFlag: AllBestScore, outFilterType: BySJout, alignSJover hangMin: 8, alignSJDBoverhangMin: 3, outFilterMismatchNmax: 999, alignIntronMin: 20, alignIntronMax: 1,000,000, alignMatesGapMax: 1,000,000, and outFilterMultimapNmax: 20; otherwise, default STAR parameters were used. Following read mapping, a count matrix was generated from the STAR results using R. Genes were removed from the matrix if they did not have at least 30 reads in a minimum of three samples from either the perturbation group (KD or overexpression) or the control group. Counts were normalized using the TMM normalization method via the calcNormFactors function in edgeR version 3.40.2 (24).
To conduct our differential gene expression analysis, we utilized the limma-trend approach from Limma version 3.54.2 (25). For each dose level of the IFN-β treatment, we assessed differential expression by comparing each KD or overexpression condition to its respective mock treatment. We then contrasted the differential expression results of each KD or overexpression (STAT1g1/1 ng IFN-β vs STAT1g1/Mock, VILMIRg1/1 ng IFN-β vs VILMIRg1/Mock, etc.) against that of its corresponding control (Ctrl/1 ng IFN-β vs Ctrl/Mock, Ctrl/10 ng IFN-β vs Ctrl/Mock). Genes were considered differentially expressed in a given contrast if their unadjusted P-value was less than 0.05 with no FC requirement. To identify IFN-responsive genes, we further filtered these results by requiring a log2FC greater than 1.25 following IFN-β treatment in the control cell lines for each experiment. Differential expression results were visualized using the ComplexHeatmap R package version 2.14.0 (26).
Pathway enrichment analysis was generated using QIAGEN Ingenuity Pathway Analysis (IPA) (27). A raw P-value cutoff of <0.05 was used to define genes with signif icant expression changes after VILMIR overexpression in each IFN-β treatment. Canonical pathways analysis identified the pathways from the QIAGEN IPA library of canonical pathways that were most significant to the data set. Differentially expressed genes from the data set that met the P-value cutoff of 0.05 (-log10 P-value 1.3) and were associated with a canonical pathway in the QIAGEN Knowledge Base were considered for the pathway analysis. A right-tailed Fisher's exact test was used to calculate a P-value determining the probability that the association between the genes in the data set and the canonical pathways is explained by chance alone.
## RESULTS
LncRNA VILMIR can interact with nuclear and cytoplasmic proteins from A549 epithelial cells in vitro, including FUBP1, PUF60, IFIT1, IFIT3, QARS1, and KARS1
To identify potential protein interactions of VILMIR, we performed an RNA pull-down assay. As VILMIR was localized in both nuclear and cytoplasmic compartments in A549 human lung epithelial cells (19), we investigated its potential protein interactions in both compartments. In vitro-transcribed biotinylated VILMIR as well as antisense VILMIR control RNA were incubated with nuclear or cytoplasmic lysates of A549 cells treated with IFN-β to mimic cellular interactions during a host interferon response. Interacting RBPs were then identified by mass spectrometry. The full list of identified proteins was narrowed down by prioritizing proteins that either uniquely interacted with VILMIR sense RNA compared to antisense RNA or were greater than two times enriched in the VILMIR sense RNA compared to antisense RNA according to peptide spectrum counts (see Materials and Methods). From these criteria, we identified 19 proteins in each compartment that were either unique to or enriched in the VILMIR sense RNA (Tables 1 and2). This suggests that VILMIR may have functions in both compartments.
In the nuclear A549 lysate, VILMIR sense RNA was found to interact with several proteins involved in pre-mRNA splicing and processing (CSTF1, CSTF2, CSTF3, CELF1, CPSF7, SYMPK, SF1, HNRNPM, CPSF1, CPSF2, and SCAF11), as well as transcriptional regulation (TARDBP, FUBP1, KHSRP, FUBP3, and PUF60). Using Western blot analysis, we confirmed the interaction of VILMIR sense RNA with FUBP1 and PUF60 (Fig. 1A). FUBP1, or Far Upstream Element-Binding Protein 1, acts as a transcriptional regulator and is a well-known activator of the c-Myc oncogene (28). While FUBP1 is primarily located in the nucleus, it has been found to translocate to the cytoplasm (29) and was also identified in a N/A, not applicable, as all peptide counts in the antisense were zero.
Full-Length Text Journal of Virology our cytoplasmic mass spectrometry results (Table 2). To negatively control the expression of c-Myc, FUBP1 also interacts with the FUBP-Interacting Repressor (FIR), which is an alternatively spliced variant of Poly(U)-Binding Splicing Factor 60 (PUF60), meaning that FUBP1 can be involved in both positive and negative regulation of gene expressions (28). Interestingly, in the cytoplasmic A549 lysate, VILMIR sense RNA interacted with IFIT1 and IFIT3 proteins more than the antisense RNA (Table 2). The IFIT protein family, or IFN-induced protein with tetratricopeptide repeats, includes IFN-stimulated genes (ISGs) that get induced during antiviral immune responses and consist of IFIT1, IFIT2, IFIT3, and IFIT5 in humans (31). The IFIT proteins are well-known to inhibit translation of both cellular mRNA and viral RNA by either interacting with eukaryotic initiation factor 3 (eIF3) to block translation (32) or by binding directly to the 5′ end of nonself RNAs (33,34). While IFIT2 and IFIT5 were not identified as VILMIR-interacting proteins in the mass spectrometry, the interaction of VILMIR sense RNA with IFIT1 and IFIT3 proteins was confirmed by Western blot and densitometry analysis (Fig. 1B). In addition, VILMIR sense RNA was found to interact with eight ARSs, which are enzymes responsible for pairing tRNAs with amino acids during translation, as well as ARS-interacting multifunctional protein 1 (AIMP1), which helps form the multi-tRNA synthetase complex (35) (Table 2). Two of these ARSs were confirmed by Western blot and densitometry analysis, QARS1 or glutaminyl-tRNA synthetase 1, and KARS1 or lysyl-tRNA synthetase 1 (Fig. 1C). Therefore, while several nuclear and cytoplasmic protein interactions were confirmed in vitro and in the cells, these results suggest that VILMIR could function in the nucleus by interacting All Western blots are representative of three independent replicates. An equal volume of the sense and antisense pull-down protein was loaded into the gel, and the input lane was included to estimate a relative FC between sense and antisense protein interaction. The labeled size in kilodaltons (kDa) denotes the major band for each protein, as well as a triangle for proteins that display multiple bands. FUBP1 has known shorter isoforms (30), while the additional bands for PUF60 and QARS1 are likely background bands. All bands were included so as not to bias the analysis. Proteins were detected by chemiluminescence using either Pierce ECL Substrate (Thermo Scientific) for the FUBP1 and PCNA blots (A) or SuperSignal West Atto Ultimate Sensitivity Substrate (Thermo Scientific) for the remaining blots (PUF60 in A and B-C). Densitometry analysis was not performed for panel A as the antisense band was too low to obtain an accurate density. *P < 0.05, **P < 0.01, and ***P < 0.001 (Student's t-test).
with proteins such as FUBP1 and PUF60 to regulate transcription or in the cytoplasm by interacting with IFIT1 and IFIT3 and/or QARS1 and KARS1 to regulate translation.
## VILMIR overexpression results in minimal fold change differences before interferon-β treatment in A549 epithelial cells
As our pull-down assay suggested that VILMIR could function through the transcript itself, we were interested in whether the overexpression of VILMIR in cells would cause significant changes in expression, both before and after IFN-β treatment. Therefore, we generated an A549 cell line overexpressing ectopic VILMIR and performed RNA-sequenc ing (RNA-seq) analysis of overexpressing cells treated with or without two separate concentrations of human IFN-β. Compared to the control cell line with an empty vector, we observed a 7.5-fold increase in the baseline expression of VILMIR (Fig. 2A). We first examined if VILMIR alone caused significant expression differences outside of an IFN response by comparing gene expression in the mock-treated cell lines. Using an adjusted P-value < 0.05, the only gene with a significant expression difference in the mock-treated cells after VILMIR overexpression was VILMIR itself. Therefore, the control cell line and VILMIR-overexpressing cell line appeared very similar in gene expressions before IFN-β treatment. When this analysis was expanded to use a relaxed criterion for differential expression analysis (raw P-value < 0.05 and no FC cutoff), we identified 459 genes that showed altered expression changes in the VILMIR-overexpressing cell line compared to the control cell line in the mock treatment (Fig. 2B).
Interestingly, there were several ISGs impacted by VILMIR overexpression in the mock-treated cells, including MX1, IFIT1, and IFIT2, the expressions of which were slightly downregulated, as well as IFI6, the expression of which was slightly upregulated (Fig. 2B). This may suggest that VILMIR expression regulates ISGs before IFN treatment. However, when plotting these same genes against their log2 fold change (log2FC) after IFN-β treatment in the control cell line, we observed that the log2FC differences between the mock-treated cell lines were relatively small in comparison. In fact, only 14% of the total 459 genes exhibited absolute FC greater than 1.5, and five genes exhibited an absolute FC greater than 2 (Fig. 2B). Apart from VILMIR itself, which was the highest upregulated gene as expected, the expressions of RNS7SL3, GBP1, and TUBA8 were upregulated and that of CFLAR-AS1 was downregulated with an FC greater than 2 (Fig. 2B). As VILMIR was transcribed ectopically from an overexpression vector, this suggests that VILMIR can function through its transcript, rather than just transcription at its genomic locus. In addition, these results suggest that VILMIR may have a regulatory role outside of the IFN response, particularly with transcription of RNS7SL3, GBP1, TUBA8, and CFLAR-AS1. However, as the majority of expression differences caused by VILMIR in the mock-treated cells were relatively small, we sought to determine the impact of VILMIR overexpression on host transcription in response to IFN-β treatment.
## Overexpression of VILMIR enhances the host transcriptional response to interferon-β treatment in A549 epithelial cells
Next, we determined the impact of VILMIR overexpression on the host transcriptional response to IFN-β treatment using the same RNA-seq analysis described above. When using an adjusted P-value < 0.05, the only gene that showed an altered expression change to IFN-β treatment after VILMIR overexpression was VILMIR itself. Therefore, to investigate the potentially broad regulatory roles of VILMIR, similarly as in our previous study (19), we used a relaxed criterion for differential expression analysis, i.e., raw P-value < 0.05, which allowed us to observe the overall trend of the transcriptional response after VILMIR overexpression. Using this criterion, we identified 731 genes that showed altered expression changes to IFN-β treatment after VILMIR overexpression in at least one of the two doses of IFN-β (Table S2). When analyzing the host transcriptional response to IFN-β, we observed larger FC differences after VILMIR overexpression, with 33%-35% of differentially expressed genes (DEGs) exhibiting absolute FCs greater than 1.5 in either IFN-β treatment, compared to 14% in the mock treatment, indicating that VILMIR overexpression has larger impacts during an IFN response. However, similarly to our previous study (19), the magnitude of expression changes was relatively small, with gene expression changing by an average of 1.4-fold after VILMIR overexpression compared to the control.
To identify canonical pathways enriched in the DEGs impacted by VILMIR overexpres sion, QIAGEN IPA was performed (27). There were 17 canonical pathways significantly enriched in the overexpression cell line in both IFN-β treatments with a raw enrichment P-value < 0.05 (-log10 P-value > 1.3), with the most significant of these pathways being the interferon alpha/beta signaling pathway (Fig. 3A; Table S3). In addition, IPA predicted an overall activation of this pathway, with positive z-scores of 2.121 and 1 for the 1 ng/mL and 10 ng/mL IFN-β treatments, respectively. The majority of the genes represented in this pathway, such as MX1, IFIT2, IRF7, IFIT1, RNASEL, TYK2, and OAS1, had higher FCs after VILMIR overexpression compared to the control cell line, besides IFITM1, which had a lower FC (Fig. 3B andC). However, it is possible that IFITM1 could serve as a negative regulator as it has previously been found to be a negative regulator in certain contexts, with suppression of IFITM1 inhibiting proliferation in glioma cells (36). The overall enhancement of genes within the IFN pathway after VILMIR overexpression is consistent with our previous findings that VILMIR KD suppressed ISGs (19), supporting our hypothesis that VILMIR plays an activating role in the host interferon response.
## VILMIR knockdown and overexpression consistently regulate the transcrip tion of a core set of interferon-stimulated genes in A549 epithelial cells
In order to identify a core set of genes that were differentially expressed in response to IFN-β treatment after both VILMIR overexpression and KD, we next combined the RNA-seq analysis of VILMIR overexpression with our previous analysis of VILMIR KD (19), which included two A549 VILMIR KD cell lines (VILMIRg1 and VILMIRg2) as well as a STAT1 KD cell line (STAT1g1) as a positive control for interferon response, treated with the same two IFN-β doses. To obtain a more robust set of genes that are IFN-responsive, we applied an additional filtering step of an FC greater than 1.25 after IFN-β treatment in the control cell lines for each experiment. After applying this criterion, we then examined expression changes after VILMIR KD or overexpression with a raw P-value < 0.05. This S3). Enriched pathways that met the raw enrichment P-value < 0.05 (-log10 P-value cutoff of 1.3) using a right-tailed Fisher's exact test and were associated with a canonical pathway in the QIAGEN Knowledge Base were included here (*P < 0.05 included as the reference). gave us a list of 132 IFN-responsive genes that showed altered expression changes to IFN-β treatment after either VILMIR KD or overexpression, in at least one of the two IFN-β doses (Fig. S1). Out of these genes, 96 were only differentially expressed after VILMIR KD, whereas 25 genes were only differentially expressed after VILMIR overexpression. This difference in the number of DEGs may be because the KD study contained two VILMIR KD cell lines, whereas the overexpression experiment only had a single VILMIR overexpression cell line. Another reason could be due to the difference in the method of gene perturbation as the KD targeted endogenous VILMIR, whereas the overexpression produced ectopic VILMIR transcripts lacking modifications. The remaining 11 out of 132 genes impacted by VILMIR perturbation were differentially expressed after both VILMIR KD and overexpression, which is what we chose to focus on (Fig. 4). One of these genes included VILMIR itself, which was expected.
The 10 genes that were consistently impacted by VILMIR perturbation were OAS1, IFIT1, UBE2L6, TAP2, TRIM38, APOL2, ERAP2, CASP1, IFIT2, and GBP1, which have all been associated with interferon and antiviral responses in literature (Table 3). As expected, the expressions of these genes were all upregulated after IFN-β treatment in our A549 cell (VILMIR) or control (Ctrl) A549 cell lines, all of which were treated with mock or either 1 ng/mL or 10 ng/mL human IFN-β for 6 h (n = 3). The heatmap displays 11 human genes that exhibited significant changes in their responses to IFN-β treatment after both VILMIR KD and overexpression (OE), in at least one of two doses of IFN (raw P-value < 0.05). Rows are genes, and columns are conditions and comparisons. As shown by the labels at the bottom, the log2FC after IFN-β treatment in each cell line was first calculated ("IFN response"), and then the "KD effect" or "OE effect" was calculated by comparing the "IFN response" log2FC of each KD/OE line to the "IFN response" log2FC of the control cell line. Red color indicates a positive log2FC value (i.e., upregulation) in columns above the label "IFN response, " or higher log2FC values in KD/OE cells compared to that of control cells in columns above the label "KD effect. " The blue color indicates lower log2FC values in KD/OE cells compared to that of control cells in columns above the label "KD/OE effect. " lines. However, after VILMIR KD, we observed an overall suppression of the expression of these genes, whereas after VILMIR overexpression, these same genes showed an increase in expression (Fig. 4). The opposite trend was true for UBE2L6, which showed an increase in the VILMIR KD lines and a decrease in the VILMIR overexpression line. However, one reason for this opposite trend may be that UBE2L6, a ubiquitin-conjugating enzyme, can be a negative regulator in certain contexts, such as inhibiting autophagy in cancer cells (37). Additionally, GBP1 was decreased in both VILMIR KD and overexpression after IFN-β treatment. However, as GBP1 was already upregulated in response to VILMIR overexpres sion before IFN-β treatment, it is possible that VILMIR upregulates GBP1 independently of IFN responses (Fig. 4, "Overexpression" mock column, as well as in Fig. 2B), which explains why the addition of IFN in the overexpression cell line results in a smaller FC. This is also true of VILMIR itself, which shows a negative FC difference in the IFN-treated overexpression cells. However, because VILMIR is already highly upregulated before IFN-β treatment in the overexpression cells, the addition of IFN-β does not result in a higher FC compared to the control cells. Finally, as all 10 genes consistently regulated by VILMIR are located on different chromosomes than VILMIR, these results suggest that VILMIR is a trans regulator of gene expression.
## DISCUSSION
We previously identified the human lncRNA VILMIR as a novel ISG during viral infection and found that KD of VILMIR in A549 cells resulted in a suppression of the host transcrip tional response to IFN-β treatment and IAV infection. However, the mechanism by which VILMIR regulates the host interferon response was not explored. Therefore, in this study, we aimed to identify potential protein interactions as well as gene regulatory networks of VILMIR to better understand its molecular interactions and propose models of how it may function during an interferon response.
Using an RNA pull-down assay and mass spectrometry, we found that VILMIR RNA interacted with several proteins in nuclear and cytoplasmic lysate from A549 epithelial cells treated with IFN-β. As we previously determined that VILMIR is distributed in both the nucleus and the cytoplasm (19), these new results further support that VILMIR may function in both compartments through protein interactions. Several lncRNAs have also been found to have dual functions in the nucleus and cytoplasm (18,(46)(47)(48). For example, lncRNA HOTAIR can function in the nucleus to regulate gene expression by interacting with histone methyltransferases (46), whereas in the cytoplasm, it can act as a competing endogenous RNA (ceRNA) by interacting with microRNAs (miRNAs) and regulating translation (47). Therefore, it is possible that VILMIR could function through different mechanisms in each compartment as well.
Since the KD of VILMIR in A549 cells resulted in a suppression of the host transcrip tional response to IFN-β treatment and IAV infection, we predicted that VILMIR may activate the host IFN-β response (19). Here, we analyzed the impact of VILMIR overex pression during IFN-β treatment. Using RNA-seq analysis, we identified 731 genes that showed altered expression changes to IFN-β treatment after VILMIR overexpression in at least one of the two doses of IFN-β. These DEGs were enriched for the interferon alpha/beta signaling pathway and displayed an overall enhancement of several ISGs after VILMIR overexpression, strongly supporting our hypothesis that VILMIR activates the host IFN response.
By combining our VILMIR KD and overexpression RNA-seq analysis, we obtained a list of 10 IFN-responsive genes with significant expression changes after both VILMIR knockdown and overexpression. Interestingly, seven of the ten genes have known functions related to translational regulation, post-translational regulation by ubiquitina tion, or protease activity-OAS1, IFIT1, UBE2L6, TRIM38, ERAP2, CASP1, and IFIT2. This may mean that VILMIR has a regulatory role within translational control or protein processing, which are important cellular processes during a viral infection, as the host translation is tightly regulated in order to limit viral propagation (49). These results were also interesting given our pull-down assay that confirmed the interaction of VILMIR with several proteins involved in translation regulation in vitro, such as IFIT1, IFIT3, QARS1, and KARS1. Therefore, it is possible VILMIR could be regulating ISG expression in the nucleus to modulate their protein abundances, as well as interacting with translational machinery in the cytoplasm. Future work is necessary to determine the potential impact of VILMIR on global translation, such as by proteomic analysis or ribosome profiling (50).
## Proposed models of VILMIR function
Taking these results together, we suggest several potential models for the function of VILMIR that should be further explored. First, as VILMIR was found to interact with FUBP1 and PUF60 in vitro, known transcriptional regulators in the nucleus (28), we suggest a mechanism by which VILMIR interacts with proteins such as FUBP1 and PUF60 to regulate gene transcription in trans (Fig. 5A), as we also observed that VILMIR KD and overexpres sion impacts the expression of genes through RNA-seq analysis. FUBP1 has been found to be both a positive regulator of transcription as well as a negative regulator through interacting with the repressor protein FIR or PUF60 (28) and has also been associated with virus infection (29,51). A different lncRNA, NR-109, was previously found to interact with FUBP1 by preventing ubiquitin-mediated degradation of FUBP1 and thus activating c-Myc transcription (52). Therefore, as VILMIR overexpression resulted in an activation of ISGs, VILMIR may either act as a guide to recruit FUBP1 to enhance transcription or VILMIR could act as a decoy to prevent FIR/PUF60 from negatively regulating transcription.
In the cytoplasmic lysate, we confirmed the interaction of VILMIR with IFIT1 and IFIT3 in vitro. IFIT1 and IFIT3 are known to inhibit translation of both cellular mRNA and viral RNA by either interacting with eIF3 to block translation (32) or by binding directly to the 5′ end of non-self RNAs (33,34). Therefore, we suggest a second model by which VILMIR either stabilizes the functions of IFIT1/3 to inhibit translation or interferes with their function to enhance translation (Fig. 5B). Apart from regulating translation, IFIT3 can also modulate interferon signaling by acting as a bridge between the MAVS complex and the TNFR-associated factor family member-associated NF-κB activator-binding kinase 1 (TBK1), which leads to phosphorylation of IRF3 and induction of IFN-β and ISG expres sion (53). Therefore, we suggest a third model by which VILMIR acts as a scaffold to help bridge IFIT3 to MAVS and TBK1, thus enhancing ISG expression, as we observed that VILMIR overexpression resulted in an activation of ISGs (Fig. 5C). Although the IFIT proteins often act in a complex (32), we did not identify IFIT2 or IFIT5 proteins in the mass spectrometry, so it is unknown whether this is due to the sensitivity of the assay or if VILMIR does not directly interact with these two proteins. In addition, as IFIT1 and IFIT3 were identified individually by mass spectrometry, the possibility that VILMIR interacts with these proteins in a complex needs to be further explored. While previous studies have reported lncRNAs that regulate the transcription of IFIT genes (54,55) and another study reported that a segment of the lncRNA NORAD binds to IFIT proteins (56), to our Full-Length Text knowledge, this is the first reported case of a full-length lncRNA interacting with IFIT proteins.
Finally, we also confirmed the interaction of VILMIR with two ARSs in vitro, QARS1 and KARS1. As stated above, ARSs are enzymes responsible for pairing tRNAs with amino acids during translation and also help form the MSC (35). A study in 2020 found that mascRNA, a small RNA derived from the lncRNA MALAT1, binds to QARS1 in the MSC in order to promote global protein translation by regulating QARS1 protein levels (57). Similarly, we suggest a final model by which VILMIR either stabilizes QARS1 and/or KARS1 in the MSC to promote translation or interferes with their function to negatively regulate translation (Fig. 5D). While these potential models need to be investigated further, they suggest that VILMIR may regulate host interferon responses through protein partners in both the nucleus and cytoplasm, which provides important foundational work in interrogating the specific mechanism of VILMIR.
As the pull-down assay was performed in vitro, future work is needed to confirm these protein interactions in cells, such as by an RNA immunoprecipitation (RIP) or RNA antisense purification (RAP) assay to establish their biological significance (58). Additional interacting molecules of VILMIR may also be determined by assays such as chromatin isolation by RNA purification (ChIRP) that can identify both chromatin and protein associations (58). This could also help determine if VILMIR is directly regulat ing the transcription of specific genes. In addition, immunoprecipitation assays could determine if VILMIR was binding to a unique protein or interacting with a protein complex. Finally, the biological significance of these protein interactions could be investigated by blocking or mutating the binding site on either VILMIR or the protein, such as in previous studies (14,59).
In summary, we found that VILMIR interacts with multiple proteins in nuclear and cytoplasmic lysate. We also confirmed that VILMIR plays an activating role in the host interferon response in trans through the establishment of an overexpression cell line. Finally, by compiling RNA-seq analyses, we identified a core set of genes that are consistently differentially expressed after both VILMIR KD and overexpression. We have proposed several potential models for how VILMIR may function in the host interferon response. We expect these results will serve as a guide in probing the molecular mechanisms of VILMIR in detail, providing new insights into the biological significance of VILMIR during antiviral and interferon responses.
## References
1. Catalanotto, Barbato, Cogoni et al. (2023) "The RNA-binding function of ribosomal proteins and ribosome biogenesis factors in human health and disease" *Biomedicines*
2. Schuntermann, Jaskolowski, Reynolds et al. (2024) "The central role of transfer RNAs in mistranslation" *J Biol Chem*
3. Valadkhan, Gunawardane (2013) "Role of small nuclear RNAs in eukaryotic gene expression" *Essays Biochem*
4. Huang, Du, Wen et al. (2022) "snoRNAs: functions and mechanisms in biological processes, and roles in tumor pathophysiol ogy" *Cell Death Discov*
5. Mattick, Amaral, Carninci et al. (2023) "Long non-coding RNAs: definitions, functions, challenges and recommendations" *Nat Rev Mol Cell Biol*
6. Mudge, Carbonell-Sala, Diekhans et al. (2025) "GENCODE 2025: reference gene annotation for human and mouse" *Nucleic Acids Res*
7. Flynn, Chang (2014) "Long noncoding RNAs in cell-fate program ming and reprogramming" *Cell Stem Cell*
8. Ma, Zhang, Wen (2018) "Membrane-lipid associated lncRNA: a new regulator in cancer signaling" *Cancer Lett*
9. Atianand, Hu, Satpathy et al. (2016) "A long noncoding RNA lincRNA-EPS acts as a transcriptional brake to restrain inflammation" *Cell*
10. Jarroux, Morillon, Pinskaya (2017) "History, discovery, and classification of lncRNAs" *Adv Exp Med Biol*
11. Ginn, Montagna, Wu et al. (2021) "Diverse roles of long noncoding RNAs in viral diseases" *Rev Med Virol*
12. Hu, Zhang, Zheng et al. (2023) "Long noncoding RNA #61 exerts a broad anti-influenza a virus effect by its long arm rings" *Antiviral Res*
13. (2026) *Full-Length Text Journal of Virology*
14. Carpenter, Aiello, Atianand et al. (2013) "A long noncoding RNA mediates both activation and repression of immune response genes" *Science*
15. Yoon, Abdelmohsen, Kim et al. (2013) "Scaffold function of long non-coding RNA HOTAIR in protein ubiquitination" *Nat Commun*
16. Engreitz, Haines, Perez et al. (2016) "Local regulation of gene expression by lncRNA promoters, transcription and splicing" *Nature*
17. Yan, Luo, Lu et al. (2017) "Cis-and trans-acting lncRNAs in pluripotency and reprogramming" *Curr Opin Genet Dev*
18. Elling, Robinson, Shapleigh et al. (2018) "Genetic models reveal cis and trans immune-regulatory activities for lincRNA-Cox2" *Cell Rep*
19. Miao, Wang, Zhan et al. (2019) "A long noncoding RNA distributed in both nucleus and cytoplasm operates in the PYCARD-regulated apoptosis by coordinating the epigenetic and translational regulation" *PLoS Genet*
20. John, Huntress, Smith et al. (2025) "Human long noncoding RNA VILMIR is induced by major respiratory viral infections and modulates the host interferon response" *J Virol*
21. Chaparian, Van Kessel (2021) "Promoter pull-down assay: a biochemical screen for DNA-binding proteins" *Methods Mol Biol*
22. Winterling, Koch, Koeppel et al. (2014) "Evidence for a crucial role of a host non-coding RNA in influenza A virus replication" *RNA Biol*
23. Vollmers, Covarrubias, Kuang et al. (2021) "A conserved long noncoding RNA, GAPLINC, modulates the immune response during endotoxic shock" *Proc Natl Acad Sci*
24. Dobin, Davis, Schlesinger et al. (2013) "STAR: ultrafast universal RNA-seq aligner" *Bioinformatics*
25. Robinson, Mccarthy, Smyth (2010) "edgeR: a Bioconductor package for differential expression analysis of digital gene expression data" *Bioinformatics*
26. Ritchie, Phipson, Wu et al. (2015) "Limma powers differential expression analyses for RNA-sequencing and microarray studies" *Nucleic Acids Res*
27. Gu, Eils, Schlesner (2016) "Complex heatmaps reveal patterns and correlations in multidimensional genomic data" *Bioinformatics*
28. Krämer, Green, Pollard et al. (2014) "Causal analysis approaches in ingenuity pathway analysis" *Bioinformatics*
29. Debaize, Troadec (2019) "The master regulator FUBP1: its emerging role in normal cell function and malignant development" *Cell Mol Life Sci*
30. Chien, Liao, Lin (2011) "FUSE binding protein 1 interacts with untranslated regions of Japanese encephalitis virus RNA and negatively regulates viral replication" *J Virol*
31. Jacob, Singh, Mohammad et al. (2014) "The splicing factor FUBP1 is required for the efficient splicing of oncogene MDM2 pre-mRNA" *J Biol Chem*
32. Diamond, Farzan (2013) "The broad-spectrum antiviral functions of IFIT and IFITM proteins" *Nat Rev Immunol*
33. Mears, Sweeney (2018) "Better together: the role of IFIT proteinprotein interactions in the antiviral response" *J Gen Virol*
34. Pichlmair, Lassnig, Eberle et al. (2011) "IFIT1 is an antiviral protein that recognizes 5'-triphosphate RNA" *Nat Immunol*
35. Kumar, Sweeney, Skabkin et al. (2014) "Inhibition of translation by IFIT family members is determined by their ability to interact selectively with the 5'-terminal regions of cap0-, cap1-and 5'ppp-mRNAs" *Nucleic Acids Res*
36. Rajendran, Kalita, Shukla et al. (2018) "Aminoacyl-tRNA synthetases: structure, function, and drug discovery" *Int J Biol Macromol*
37. Yu, Ng, Chow et al. (2011) "Knockdown of interferon-induced transmembrane protein 1 (IFITM1) inhibits proliferation, migration, and invasion of glioma cells" *J Neurooncol*
38. Falvey, 'donovan, El-Mashed et al. (2017) "UBE2L6/UBCH8 and ISG15 attenuate autophagy in esophageal cancer cells" *Oncotarget*
39. Harioudh, Perez, Chong et al. (2024) "Oligoadenylate synthetase 1 displays dual antiviral mechanisms in driving translational shutdown and protecting interferon production" *Immunity*
40. Zhang, Zhang (2011) "Interferon-stimulated gene 15 and the protein ISGylation system" *J Interferon Cytokine Res*
41. Mantel, Sadiq, Blander (2022) "Spotlight on TAP and its vital role in antigen presentation and cross-presentation" *Mol Immunol*
42. Xue, Zhou, Lei et al. (2012) "TRIM38 negatively regulates TLR3-mediated IFN-β signaling by targeting TRIF for degradation" *PLoS One*
43. Liao, Goh, Betts et al. (2011) "A novel anti-apoptotic role for apolipoprotein L2 in IFN-γ-induced cytotoxicity in human bronchial epithelial cells" *J Cell Physiol*
44. Saulle, Marventano, Saresella et al. (2021) "ERAPs reduce in vitro HIV infection by activating innate immune response" *J Immunol*
45. Bauer, Brighton, Mueller et al. (2012) "Influenza enhances caspase-1 in bronchial epithelial cells from asthmatic volunteers and is associated with pathogenesis" *J Allergy Clin Immunol*
46. Anonymous (2025) "The guanylate-binding protein GBP1 forms a protein coat that enwraps cytosol-invasive bacteria" *Nat Struct Mol Biol*
47. Bhan, Mandal (2015) "LncRNA HOTAIR: a master regulator of chromatin dynamics and cancer" *Biochim Biophys Acta*
48. Li, Sun, Li et al. (2023) "HuR-mediated nucleocytoplasmic translocation of HOTAIR relieves its inhibition of osteogenic differentiation and promotes bone formation" *Bone Res*
49. Katsushima, Natsume, Ohka et al. (2016) "Targeting the Notch-regulated non-coding RNA TUG1 for glioma treatment" *Nat Commun*
50. Hoang, Neault, Pelin (2021) "Emerging translation strategies during virus-host interaction" *Wiley Interdiscip Rev RNA*
51. (2026) *Full-Length Text Journal of Virology*
52. Brar, Weissman (2015) "Ribosome profiling reveals the what, when, where and how of protein synthesis" *Nat Rev Mol Cell Biol*
53. Dixit, Pandey, Liu et al. (2015) "FUSE binding protein 1 facilitates persistent hepatitis C virus replication in hepatoma cells by regulating tumor suppressor p53" *J Virol*
54. Zhang, Wei, Dai et al. (2023) "The NR_109/FUBP1/c-Myc axis regulates TAM polarization and remodels the tumor microenvironment to promote cancer develop ment" *J Immunother Cancer*
55. Liu, Wei, Shan et al. (2011) "IFN-induced TPR protein IFIT3 potentiates antiviral signaling by bridging MAVS and TBK1" *J Immunol*
56. Guo, Jiang, Wu et al. (2022) "LncRNA RP5-998N21.4 promotes immune defense through upregulation of IFIT2 and IFIT3 in schizophrenia" *Schizophrenia (Heidelb)*
57. Van Solingen, Cyr, Scacalossi et al. (2022) "Long noncoding RNA CHROMR regulates antiviral immunity in humans" *Proc Natl Acad Sci*
58. Tichon, Gil, Lubelsky et al. (2016) "A conserved abundant cytoplasmic long noncoding RNA modulates repression by Pumilio proteins in human cells" *Nat Commun*
59. Lu, Huang, Wu et al. (2020) "The tRNA-like small noncoding RNA mascRNA promotes global protein translation" *EMBO Rep*
60. Wang, Chekanova (2019) "An overview of methodologies in studying lncRNAs in the high-throughput era: when acronyms ATTACK" *Methods Mol Biol*
61. Hu, Lou, Gupta (2014) "The long non-coding RNA GAS5 cooperates with the eukaryotic translation initiation factor 4E to regulate c-Myc translation" *PLoS One*
62. (2026) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793559&blobtype=pdf | # Evolutionary Dynamics of Respiratory Syncytial Virus in Pre-pandemic, Pandemic, and Post-pandemic Periods in Houston, Texas, USA
Vasanthi Avadhanula, Daniel Agustinho, Leila Sahni, Anil Sarathu, David Henke, Harshavardhan Doddapaneni, Donna Muzny, Ginger Metcalf, Sara Javornik Cregeen, Natalie Thornburg, Heidi Moline, Ayzsa Tannis, Fritz Sedlazeck, Pedro Piedra
## 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. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793519&blobtype=pdf | Lyubov Tiegs, Madison Okuno, Jamie Forschmiedt, Ken Kunisaki, Alexa Pragman, Kyle Rudser, Chris Wendt, David Macdonald
## Abstract
Background. Chronic obstructive pulmonary disease (COPD) is the 4 th leading cause of death globally. COPD is defined by irreversible expiratory airflow obstruction (e.g., FEV1/FVC< 0.70) and severity is assessed by reduced FEV1. COPD is most often caused by inhalation of tobacco smoke, but even after controlling for smoking and
other known risk factors, much of the risk remains unexplained. Latent cytomegalovirus (CMV) has been hypothesized to affect lung function through effects on natural killer cells, systemic inflammation, and direct effects on lung tissue. Prior analyses have been performed in populations with limited racial/ethnic, biologic sex, and/or geographic representation. The National Health and Nutritional Examination Survey (NHANES III) was a population-based sample from the United States.
Table 1 Associations between CMV seropositivity and lung function. All analyses were adjusted for age, race/ethnicity, biologic sex, smoking, poverty-income ratio, education, and urban vs rural residence. FEV1, forced expiratory volume in 1-second; FVC, forced vital capacity 1: Defined as FEV1/FVC ratio < 0.7. 2: Mean differences represent differences in the outcome of interest (ratio for FEV1/FVC ratio and liters for FEV1) in participants positive for CMV compared to those negative for CMV.
Table 2 Associations between CMV seropositivity and lung function in ever smokers and never smokers. All analyses were adjusted for age, race/ethnicity, biologic sex, smoking, poverty-income ratio, education, and urban vs rural residence. All interaction p-values (not shown) were greater than 0.4. CI, confidence interval; OR, odds ratio. 1: Defined as FEV1/FVC ratio < 0.7. 2: Mean differences represent differences in the outcome of interest (ratio for FEV1/FVC ratio and liters for FEV1) in participants positive for CMV compared to those negative for CMV.
Methods. Using cross-sectional data from NHANES III, we tested associations between CMV seropositivity (ELISA optical density index values greater than 1.0) and the following outcomes: 1) airflow obstruction [FEV 1 /forced vital capacity (FVC) < 0.7]; 2) FEV 1 /FVC (continuous), and FEV 1 (continuous). Binary outcomes were analyzed using logistic regression and continuous outcomes were analyzed using linear regression. All analyses used robust variance estimation for confidence intervals and P-values and were adjusted for important covariates listed in the tables below.
Results. Among 14,279 participants with available spirometry and CMV IgG serologies, 11,000 were CMV positive and 3,279 were CMV negative. The mean age of included participants was 42.5 (SD 16.6) years, 51.2% were female, and 54.2% were ever smokers. CMV seropositivity was associated with higher odds of airflow obstruction and lower FEV 1 /FVC ratio and FEV 1 (Table 1). Results did not significantly differ in never smokers vs ever smokers (Table 2).
Conclusion. In a general population sample of the United States, CMV seropositivity was independently associated with worse lung function. If these results are confirmed in longitudinal studies, CMV vaccines and other CMV mitigation efforts could have a role in reducing the burden of COPD.
Disclosures. Ken Kunisaki, MD, MS, Nuvaira: Data and Safety Monitoring Board Member
$$1 1 1 1 1 1 1 2 2 1$$ |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12613920&blobtype=pdf | # Complete genome sequence of Bacillus cereus phage vB_BceH_LBC2, a lytic member of the genus Bequatrovirus
Eo-Jin Kim, Jeong-A Lim, John Dennehy
## Abstract
Phage vB_BceH_LBC2 is a lytic bacteriophage infecting Bacillus cereus, classified as a member of the genus Bequatrovirus (family Herelleviridae). Its genome is 157,419 base pairs, with 232 predicted genes and a GC content of 35.63 percent. No virulence, antibiotic resistance, or lysogeny-related genes were identified.
KEYWORDS bacteriophage, genome sequencing, Bacillus cereus, BequatrovirusB acillus cereus is a gram-positive, spore-forming bacterium linked to food spoilage and gastrointestinal illness (1). Its ubiquity and resistance make it a frequent target for phage-based control (2). Lytic phages offer host-specific antibacterial potential (3). Here, we report the isolation, morphology, and genome sequence of lytic phage vB_BceH_LBC2, a member of the genus Bequatrovirus.Phage vB_BceH_LBC2 was isolated from sewage samples from wastewater treatment plants in Iksan, South Korea (35.988558°N, 126.935605°E), using B cereus NCCP 14796, obtained from the National Culture Collection for Pathogens (NCCP), as the host strain. Filtered samples were mixed with 2× TSB containing 2 mM MgCl 2 and CaCl 2 , incubated with the host at 37°C for 24 h, re-filtered, and spotted onto B. cereus lawns via doublelayer agar (4). Plaques were picked and purified through three reinfection rounds. Phages were propagated in exponential-phase cultures, lysate dialyzed with SM buffer, and stored at 4°C (5).Phage morphology was confirmed by transmission electron microscopy (TEM) after negative staining with 2% (wt/vol) uranyl acetate (pH 4.0) (6). TEM showed LBC2 possesses an icosahedral capsid (length: 86.3 ± 1.2 nm; width: 90.9 ± 6.0 nm) and a long, flexible tail (205.7 ± 5.8 nm), typical of siphovirus-like morphology (Fig. 1).Genomic DNA was extracted from a high-titer phage stock using the Phage DNA Isolation Kit (Norgen Biotek, Ontario, Canada) (7). Paired-end libraries (2 × 300 bp) were prepared using the TruSeq Nano DNA Library Prep Kit (Illumina) and sequenced on the Illumina MiSeq platform by Sanigen Co., Ltd. (South Korea). Adapter trimming and quality filtering were performed using Trimmomatic v0.39 (8), and read quality was assessed with FastQC v0.11.9. A total of 852,458 reads were generated, yielding approxi mately 242 Mb of sequence data, with an average genome coverage of ~1,537× . To assess the genome termini and packaging strategy, we analyzed the raw sequencing reads using PhageTerm (9). De novo genome assembly was performed using SPAdes v3.15.2 with default parameters (10). All other software was run with default parameters unless otherwise specified. Open reading frames (ORFs) were predicted using Gene MarkS (11), and annotated using RAST (12) and BLASTp searches against the NCBI nonredundant protein database (13). In addition, tRNA and tmRNA genes were searched using ARAGORN v1.2.41 ( 14).The complete genome of phage LBC2 is 157,419 bp in length and has a GC content of 35.63%. The genome comprises 232 predicted protein-coding sequences, of which 99 were assigned putative functions. PhageTerm analysis suggested that the phage genome
is circularly permuted with multiple preferred termini, consistent with a headful packag ing mechanism. No tRNA genes or other non-coding RNA elements were identified in the genome. No virulence or lysogeny-associated genes, including integrases or repressors, were detected. Based on sequence similarity and current ICTV taxonomy (15), phage LBC2 is classified within the genus Bequatrovirus, subfamily Bastillevirinae, family Herelleviridae, and class Caudoviricetes. The closest known relative to phage LBC2 is Bacillus phage vB_BceH_LY2 (GenBank accession no. ON366411.1), with 97.68% nucleo tide identity over 82% of the genome, supporting its classification within the genus Bequatrovirus.
## References
1. Haque, Quan, Zuo et al. (2021) "Pathogenic ity of feed-borne Bacillus cereus and its implication on food safety" *Agrobiol Records*
2. Garvey (2022) "Bacteriophages and food production: biocontrol and bio-preservation options for food safety" *Antibiotics (Basel)*
3. Li, Li, Ma et al. (2024) "Isolation and characteristic of Bacillus cereus phage Z3 and its application in rice and milk" *LWT*
4. Li, Yuan, Li et al. (2020) "Isolation and characterization of Bacillus cereus phage vB_BceP-DLc1 reveals the largest member of the Φ29-like phages" *Microorganisms*
5. Peters, Harris, Davis et al. (2022) "Bacteriophage isolation, purification, and characterization techniques against ubiquitous opportunistic pathogens" *Curr Protoc*
6. Meidaninikjeh, Mohammadi, Elikaei (2024) "A simplified method of bacteriophage preparation for transmission electron microscope" *J Virol Methods*
7. Maffei, Manner, Jenal et al. (2024) "Complete genome sequence of Pseudomonas aeruginosa phage Knedl" *Microbiol Resour Announc*
8. Bolger, Lohse, Usadel (2014) "Trimmomatic: a flexible trimmer for Illumina sequence data" *Bioinformatics*
9. Garneau, Depardieu, Fortier et al. (2017) "PhageTerm: a tool for fast and accurate determination of phage termini and packaging mechanism using next-generation sequencing data" *Sci Rep*
10. Prjibelski, Antipov, Meleshko et al. (2020) "Using SPAdes de novo assembler" *CP in Bioinformatics*
11. Besemer, Lomsadze, Borodovsky (2001) "GeneMarkS: a self-training method for prediction of gene starts in microbial genomes. Implications for finding sequence motifs in regulatory regions" *Nucleic Acids Res*
12. Aziz, Bartels, Best et al. (2008) "The RAST server: rapid annotations using subsystems technology" *BMC Genomics*
13. Tatusova, Dicuccio, Badretdin et al. (2016) "NCBI prokaryotic genome annotation pipeline" *Nucleic Acids Res*
14. Laslett, Canback (2004) "ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences" *Nucleic Acids Res*
15. Zerbini, Siddell, Lefkowitz et al. (2023) "Changes to virus taxonomy and the ICTV Statutes ratified by the international committee on taxonomy of viruses" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645957&blobtype=pdf | # Vaccination protects against COVID-associated pulmonary fibrin deposition
Joanna Ireland, David Myers, Chang Huang, Cameron Allen, Gwynne Roth, Zhongcheng Zou, Ming Zhao, Motoshi Suzuki, Lisa Olano, Joshua Tan, Shreya Kanth, Julio Huapaya, Homer Twigg, Anthony Suffredini, Peter Sun
## Abstract
Understanding the protective mechanism of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines against severe COVID is important for therapeutic development to mitigate disease-associated lung pathologies. Here, we investigated the association between vaccination and the disease severity with a particular emphasis on viral-induced pulmonary fibrin formation in 43 COVID individu als. While COVID vaccination reduced the disease severity in this cohort, their plasma coagulation indices, including prothrombin time (PT), partial thromboplastin time (PTT), and D-dimer concentrations, remain unchanged between the vaccinated and non-vac cinated individuals. In contrast, vaccination lowered pulmonary inflammatory and coagulation signatures, reduced fibrinogen concentrations, and prevented prothrombin activation in bronchoalveolar lavage fluid (BALF), such that no viral-induced fibrin was observed in the vaccinated BALF. The formation of viral-induced fibrin correlated with the disease severity and was observed in non-vaccinated BALF samples containing high concentrations of fibrinogen and prothrombin, suggesting vaccination protected against the viral-induced pulmonary fibrin formation. Our finding highlights the use of pulmonary rather than plasma fibrinogen levels as a risk indicator for severe COVID disease.IMPORTANCE Understanding the protective mechanism of COVID-19 vaccines against the severity of the disease is important for therapeutic development, and thus, subject to intense investigation. Here, we studied a cohort of 43 COVID patients based on their vaccination status. We showed that (i) COVID disease severity is associated with the formation of SARS-CoV-2-induced pulmonary fibrin, (ii) vaccination protected against severe COVID disease by reducing infiltration of coagulants, preventing prothrombin activation and fibrin deposition in infected lungs, and (iii) plasma coagulation indices are not useful indicators for fibrin deposition in infected lungs. Rather, the level of pulmo nary fibrinogen provides an informative indicator for COVID-associated coagulation in lung.
inflammatory responses in severe COVID diseases. However, the clinical benefit of therapeutic use of anti-inflammatory compounds was marginal for treating severe COVID (15)(16)(17)(18). In addition, vaccination also reduced the disease severity in immunocom promised individuals (19)(20)(21)(22), suggesting the benefit of vaccination extends beyond the reduction of inflammation. The potential contributing factors to severe COVID-associated pathophysiology are likely multifaceted, involving inflammatory neutrophil and T cell infiltrations, potential formation of neutrophil extracellular traps, and the presence of other comorbidity factors, such as immunodeficiency and diabetes (23). Early autopsies showed the presence of characteristic fibrin-rich hyaline membranes associated with diffuse alveolar damage (DAD) in infected lungs as a hallmark of severe COVID-19 infections (24)(25)(26)(27). D-dimers are degradation fragments of fibrin and are used for assessing the risk of venous thromboembolism (28)(29)(30). While some studies support a correlation between higher plasma D-dimer levels and the severity of COVID-19 disease (31)(32)(33)(34), others did not (35)(36)(37)(38)(39). Furthermore, the therapeutic use of anticoagulant heparin failed to reverse the disease in critically ill patients (16).
We recently described a SARS-CoV-2 infection-induced non-classical coagulation that occurred in pulmonary tissue outside of plasma (40). The viral infection activates the TMPRSS family of serine proteases, such as ST14 and HAT, on lung epithelial cells, leading to prothrombin activation and fibrin deposition in the presence of infiltrating fibrinogen. Here, we investigated the link between COVID vaccination and the viral-induced fibrin deposition. Using bronchoalveolar lavage fluid (BALF) and plasma collected from COVID clinical studies during the early pandemic from 2020 to 2021, we investigated the impact of vaccination on the level of coagulation factors in plasma and pulmonary airways and assessed the risk of viral-induced pulmonary fibrin deposition (Fig. 1A). Our findings clarified the conflicting data regarding the use of D-dimers as the indicator for severe COVID and showed pulmonary fibrinogen concentrations correlated with viral-induced pulmonary fibrin depositions. These findings suggest a new mechanism for vaccinationassociated protection against disease severity.
## RESULTS
## Vaccination reduced COVID disease severity
Existing data support the benefit of vaccination in protection against severe COVID diseases despite the occurrence of breakthrough infections (6,7,9). To investigate the potential protective mechanisms of COVID vaccines, we investigated the risk of coagula tion in plasma and lung among a group of 43 vaccinated and non-vaccinated COVID patients during the early pandemic between 2020 and 2021. Among them, 19 individu als received either mRNA-1273 (Moderna), BNT162b2 (Pfizer), or Ad26.COV2.S (Johnson & Johnson) vaccines and are regarded as having breakthrough infections. The remainder did not receive prior vaccinations. Plasma and bronchoalveolar lavage (BAL) samples were collected to assess the risk of coagulations in blood and pulmonary space, respec tively. The samples were further grouped into acute, recovery, and convalescent categories if they were collected within the first 20 days, between 3 and 8 weeks, or after 8 weeks of COVID symptom onset (Fig. 1A). Among acute COVID patients of this study group, vaccination lessened the clinical disease severity (Fig. 1B), consistent with the protective effect of vaccination against COVID severity.
## Plasma coagulation indices did not differ with vaccination status
As coagulopathy is associated with severe COVID-19 (26,41), it is conceivable that vaccination may lower the risk of hypercoagulation associated with COVID-19. Blood coagulation occurs through extrinsic and intrinsic coagulation pathways, and its risk is measured by prothrombin time (PT) and partial thromboplastin time (PTT) (42)(43)(44). Interestingly, despite heightened inflammation in the non-vaccinated group as evi denced by the presence of higher levels of C-reactive protein (CRP) and proinflammatory cytokine (Fig. 1C; Fig. S1A), their plasma coagulation indices, PT and PTT, remained similar to those from vaccinated COVID and healthy plasma samples (Fig. 1C), suggesting inflammation did not affect their plasma coagulation indices. We further measured fibrinogen and prothrombin concentrations in vaccinated and non-vaccinated COVID plasma, as well as in healthy controls by enzyme-linked immunosorbent assay (ELISA). Individual fibrinogen varied between 2 and 8 mg/mL, and prothrombin varied between 50 and 150 µg/mL in plasma samples (Fig. 1D). Overall, the concentrations of fibrinogen, prothrombin, and D-dimer remained similar between vaccinated and non-vaccinated COVID plasma samples (Fig. 1D), consistent with the observations by others (45). Thus, vaccination did not affect coagulation in the blood circulation.
## Vaccination reduced pulmonary immune activation and coagulation signatures
Despite clear serological differences in both cellular immune responses and inflammatory cytokine activations between vaccinated and non-vaccinated populations (Fig. S1A) (11)(12)(13), plasma coagulation parameters PT and PTT did not differ between the two groups (Fig. 1C andD). As viral-induced fibrin deposition does not require the initiation of the extrinsic coagulation pathway in plasma, this prompted us to investigate if pulmonary tissue rather than plasma concentration of coagulation factors is a better indicator for COVID-associated fibrin deposition. Previous proteomic characterizations of BALF from SARS-CoV-2-infected lungs primarily focused on dysregulation of immune function and inflammatory signatures associated with COVID-19 disease severity (46)(47)(48)(49)(50). To address the effect of vaccination on fibrinogenic profiles in COVID lung fluids, we performed proteomic analyses on BAL fluids (BALF) by mass spectrometry from 6 vaccinated, 10 non-vaccinated (both acute and recovery) COVID individuals, as well as 4 convalescent COVID and two healthy individuals. In total, mass spectrometry identified approximately 1,000 common proteins from these BALF samples (Table S1). Approxi mately 200 proteins are pulmonary and plasma proteins, including surfactants and mucins, common enriched plasma proteins, coagulation factors, complement factors, serpins, and immunoglobulins (Table S1; Fig. 2A; Fig. S1B). The rest are intracellular proteins, presumably released from apoptotic cells. Among infiltrated plasma proteins, the most abundant ones are from common enriched plasma proteins, immunoglobu lins, complement factors, and serpins categories, indicating a significant presence of both adaptive and innate immune systems in lung. Differential protein abundance analysis showed an increased abundance of complement components in all COVID BALF, including vaccinated, non-vaccinated, and convalescent COVID samples compared to the healthy controls, consistent with the activation of innate immune responses to SARS-CoV-2 infections (Fig. 2D; Fig. S2A). The presence of more abundant immuno globulins in the non-vaccinated than the vaccinated lung fluids is consistent with a recent finding of non-vaccinated BALF forming clusters containing more inflammatory proteins than vaccinated BALF (Fig. 2A; Fig. S2A andB) (46). Given that immunizations in general increase antigen-specific antibody titers, we further measured the concentra tion of SARS-CoV-2 spike-specific IgG, as well as the neutralization antibody titers in non-vaccinated and vaccinated samples (Fig. 2B; Fig. S3). As expected, higher levels of SARS-CoV-2 specific IgGs are present in vaccinated than in the non-vaccinated plasma samples. To the contrary, the order of SARS-CoV-2 specific antibodies levels, as well as their neutralization antibody titers in BALF, is reversed between the vaccinated and non-vaccinated individuals (Fig. 2B). The result is counterintuitive and suggests a decreased plasma infiltration in vaccinated lungs.
## Vaccination reduced pulmonary fibrinogen, prothrombin, and D-dimer concentrations
Next, we examined the presence of coagulation factors in lung fluids. Among them, fibrinogen, prothrombin, antithrombin-III, factor XII, and plasminogen are routinely present in all BALF samples; other coagulation factors are less abundant and mostly undetectable by mass spectrometry (Table S1). Importantly, similar levels of coagulation factors are found in the vaccinated and healthy donors; both are less abundant than those in the non-vaccinated individuals (Fig. 2C). Furthermore, the abundance of common plasma proteins is also less in the vaccinated than in the non-vaccinated BALF (Fig. S2), consistent with the reduction of plasma infiltration in vaccinated COVID lungs. Further analysis of the mass spectrometry data from the longitudinal COVID-19 studies by Kanth et al. also supports the presence of lower levels of BALF coagulation factors in the vaccinated than non-vaccinated individuals. Specifically, both fibrinogen and prothrombin were significantly less abundant in the vaccinated and healthy group than in the non-vaccinated group (Fig. 3A andB). In addition, the level of fibrinogen-α, -β, and -γ in both vaccinated and non-vaccinated BALF decreased as individuals progressed from the acute to recovery phase (Fig. 3C).
To further validate the proteomic analyses on fibrinogen and prothrombin levels, we measured fibrinogen, prothrombin, and D-dimer concentration from the vaccinated, non-vaccinated COVID, as well as healthy BALF samples by ELISA (Fig. S4). Unlike their plasma concentrations, which varied less than threefold among individuals, the BALF fibrinogen and prothrombin concentrations varied up to 500-fold (Fig. 4A; Fig. S4). Overall, most fibrinogen concentrations varied from less than 50 to 500 ng/mL in the vaccinated, but to greater than 50 µg/mL in non-vaccinated COVID BALF samples. Likewise, most prothrombin concentrations ranged from 10 to 200 ng/mL in vaccinated BALF but to ~3 µg/mL in non-vaccinated samples (Fig. 4A; Fig. S4). Consistent with the mass spectrometry data, concentrations from vaccinated individuals are similar to those from healthy donors but are significantly lower than those from non-vaccinated individu als (Fig. 4A). When the vaccinated and non-vaccinated samples are further separated into the acute and recovery phases, it shows that vaccination primarily reduced the acute phase coagulation factor concentrations (Fig. 4B). Despite their wide variation in concentrations, both prothrombin and D-dimer levels are correlated with those of fibrinogen (Fig. 4C).
## Full-Length Text
## Viral-induced fibrin was absent in vaccinated BALF and correlated with COVID severity
Elevated fibrinogen and prothrombin levels can induce fibrin formation in the presence of SARS-CoV-2 infection of primary airway epithelial cells. This viral-induced fibrin depends on the activation of prothrombin by TMPRSS family proteases on infected airway epithelial cells (40). Both ELISA and proteomic data showed reduced levels of fibrinogen and prothrombin in vaccinated compared to non-vaccinated BALF, suggest ing less risk of fibrin deposition in vaccinated BALF. To address this, we evaluated the tendency of vaccinated and non-vaccinated BALF to develop fibrin clots in the presence of SARS-CoV-2 infections. Normal human bronchial/tracheal epithelial (NHBE) or small airway epithelial (HSAEC) cells were infected with SARS-CoV-2 pseudovirus at a dose of 2-5 copies viral RNA/cell in 384-well cell culture plates for 24 h before replacing the cell culture media with 1:1 mix of 20-fold concentrated BALF and clotting buffer (20 mM HEPES, 137 mM NaCl, 5 mM CaCl 2 ) for 2-4 h at 26°C to observe fibrin clotting. The samples were then fixed with 2% paraformaldehyde, and fibrin clots were imaged using a confocal microscope. Fibrin clots were observed in 3 out of 11 non-vaccinated acute BALF in the presence of the viral infections (Fig. 5; Fig. S5). In contrast, no fibrin clots were observed in any of the vaccinated BALF (Fig. 5; Fig. S5), suggesting that the concentra tions of fibrinogen and prothrombin in the vaccinated BALF were insufficient to support fibrin clot formation. These results indicate that the COVID vaccine protects against SARS-CoV-2 infection-induced pulmonary fibrin deposition. It is worth noting that the majority of non-vaccinated BALF with a lower concentration of fibrinogen did not form the viral-induced fibrin. Fibrin formations were observed only in non-vaccinated BALF containing the highest levels of coagulation factors (Fig. 4C). When the association between the presence of viral-induced fibrin and COVID clinical severity is analyzed regardless of the vaccination status, the formation of fibrin was found to correlate with COVID severity (Fig. 6A), suggesting pulmonary fibrin deposition contributes to severe COVID disease.
## Protective mechanism of vaccination against viral-induced fibrin deposition
The association between fibrin deposition and COVID severity prompted us to examine the dependence of fibrin clotting on the viral doses. We infected ACE2-293T with either titration doses of pseudotyped SARS-CoV-2 (pSARS-2) or a constant viral dose in the presence of titration amounts of neutralizing antibodies (CV664 and CV521) (51). As expected, the infection of ACE2-293T cells correlated with the viral dose and neutraliza tion antibodies CV664 and CV521 but not IgG 1 inhibited the viral infections with IC 50 of 3.4 and 4.0 nM, respectively (Fig. 6B andC; Fig. S6A through C). Likewise, the extent of fibrin clotting was also dependent on the viral dose (Fig. 6D; Fig. S7A andB). Additionally, when the supernatants from the titration dose pSARS-2 infected HSAEC cells were used to cleave a fluorogenic fibrinogen-β peptide, corresponding to the thrombin cleavage region, it resulted in a viral-dose-dependent cleavage like the peptide cleavage by thrombin (Fig. 6E andF), suggesting that the prothrombin activation depends on the viral dose. Thus, while vaccination-induced neutralization antibodies may be insufficient to prevent infections (52,53), vaccination leads to partial neutralization of SARS-CoV-2 viral titers, resulting in diminished prothrombin activation and reduced infiltration of fibrinogen compared to non-vaccinated naïve infections.
## Pulmonary fibrinogen level indicates the risk of developing viral-induced fibrin
Unlike their plasma concentrations that varied less than threefold in SARS-CoV-2 infections, pulmonary fibrinogen concentrations varied more than 500-fold from less than 20 ng/mL to greater than 10 µg/mL in non-vaccinated COVID BALF (Fig. S4). Since fibrinogen, prothrombin, and D-dimer concentrations in BALF are correlated and fibrin clotting was observed in samples with highest concentrations of all three (Fig. 4C), we then asked if pulmonary rather than plasma fibrinogen level serves a better indication for the risk of fibrin deposition in SARS-CoV-2 infected lung and if a threshold level of fibrinogen exists above which fibrin clotting becomes significant.
To address the minimum concentration of fibrinogen required to support viral-induced fibrin deposition, we assessed the extent of SARS-CoV-2-induced fibrin clot formation with titration concentrations of fibrinogen between 10 and 500 µg/mL, corresponding to approximately 0.5-25 µg/mL fibrinogen in BALF in the presence of a constant dose of a GFP-expressing omicron-spike pSARS-2 virus. Indeed, the amount of viral-induced fibrin clot formation was proportional to the fibrinogen concentration (Fig. 6G), and the viral-induced fibrin formation occurred at above ~50 µg/mL of fibrinogen (or 2-5 µg/mL of fibrinogen in BALF), below which no significant fibrin was observed (Fig. 6G; Fig. S4A andS7C). This suggests a protective mechanism of vaccination against severe COVID is through reducing pulmonary fibrinogen and prothrombin into infected lungs. Indeed, the fibrinogen concentrations in vaccinated BALF are between 50 and 5 µg/mL (Fig. 4A), generally below the threshold of fibrin clotting. Among the non-vaccinated samples, the majority of their fibrinogen concentrations are also below the fibrin clotting threshold, with only the highest ones exceeding the clotting threshold and support ing viral-induced fibrin clotting, suggesting these individuals are at risk of developing pulmonary fibrin deposition. Taken together, we suggest that pulmonary fibrinogen concentration serves as a biomarker for the risk of developing COVID-associated lung fibrin deposition.
## DISCUSSION
As COVID-19 evolves from pandemic to recurring seasonal infections, vaccines and antiviral drugs are the primary health care measures to counter SARS-CoV-2 infections. Despite the frequent occurrence of breakthrough infections, vaccines protected against severe COVID diseases. Understanding the mechanism of this protection is key to developing therapeutic treatment against severe COVID. Early publications suggested COVID vaccination reduced the overt immune responses associated with SARS-CoV-2 infections (11-13), thus lessened pulmonary immunopathology. However, vaccination also benefited the immunocompromised population (21,22). Further complicating our understanding is the lack of a reliable biomarker for severe COVID-associated hypercoa gulation.
Here, we investigated the protective mechanisms of SARS-CoV-2 vaccines against severe COVID-19 diseases. Early autopsy data showed the presence of fibrosis in diseased lungs (25)(26)(27). However, clinical use of anticoagulant, including low molecular weight heparin, failed to mitigate severe COVID-associated mortality (16). We introduced here a model for severe COVID-associated pulmonary hypercoagulation based on SARS-CoV-2 infection-induced fibrin formation and showed it correlated with disease severity (Fig. 6A) (40). Using this model, we investigated the influence of vaccination on the viral-induced fibrin deposition. We found that plasma coagulation indices, including PT, PTT, as well as plasma concentrations of fibrinogen and prothrombin remained similar between COVID and healthy, or vaccinated and non-vaccinated COVID groups. In contrast, vaccination reduced pulmonary inflammation and plasma infiltrations, resulting in lower pulmonary fibrinogen, prothrombin, and D-dimer concentrations in vaccinated than non-vaccinated COVID BALF. Importantly, vaccinations protected against the viral-induced fibrin formation, suggesting a protective mechanism of SARS-CoV-2 vaccine by reducing the risk of pulmonary fibrin deposition. The benefit of the vac cine appears twofold. First, the vaccine reduces the viral load and, consequently, less viral-induced prothrombin activation. Second, the vaccine reduces infiltration of coagulation components into infected lungs (Fig. 7).
SARS-CoV-2 infection leads to pneumonia and DAD with characteristic fibrin-rich hyaline membranes. Our finding supports that the viral-induced fibrin deposition contributed to the formation of pulmonary hyaline membranes. The formation of viral-induced fibrin deposition depends on both viral load and inflammation. Higher viral loads would lead to increased activation of TMPRSS proteases in infected lung epithelial cells, as well as increased inflammatory infiltration of fibrinogen and prothrombin into infected alveolar space, leading to the activation of prothrombin and fibrin deposition. The presence of vaccine-induced neutralization antibodies can minimize viral-induced fibrin formation through reducing both viral activation of TMPRSS proteases and inflammatory infiltration of coagulation components. As the levels of BALF fibrinogen and prothrombin correlated with viral-induced fibrin formation, we speculate that their pulmonary but not plasma concentrations serve as a better marker for clinical risk of severe COVID. We estimated the threshold fibrinogen concentration to be ~50 µg/mL for developing viral-induced pulmonary fibrin clots. Consistently, healthy donors have less than 10 µg/mL fibrinogen in their lung, well below the clotting threshold. The majority of the non-vaccinated lavages do not contain high levels of fibrinogen to result in fibrin deposition. Some non-vaccinated lavages show similar levels of fibrinogen and prothrombin as vaccinated or healthy samples, suggesting the presence of strong natural immunity is sufficient to prevent fibrin deposition and DAD. However, a small number of non-vaccinated lavages, ~10% in this small study, exhibited viral-induced fibrin formation. In summary, the investigation of pulmonary coagulation factors and their involve ment in viral-induced fibrin clot formation in vaccinated and non-vaccinated BALF revealed a vaccine-mediated protective mechanism against SARS-CoV-2 induced pulmonary fibrin deposition and suggests the use of pulmonary fibrinogen concentra tion as a biomarker for severe COVID. of Health. The details of COVID-ARC-19 patients and their responses to SARS-CoV-2 infection were described previously (46). Thirteen of the COVID BAL samples were collected through a CLIA-approved clinical BAL laboratory at Indiana University under protocol 1011003397R010. Eighteen of the COVID patients received either mRNA-1273 (Moderna), BNT162b2 (Pfizer), or Ad26.COV2.S (Johnson & Johnson) vaccinations and thus were regarded as having breakthrough infections, while 25 of the COVID patients did not receive prior vaccination before the study. Plasma samples were collected from a subgroup of the COVID-ARC-19 patients. Eighteen of the BAL samples collected within the first 20 days of COVID symptom onset are designated as acute, and 19 of the BAL samples collected between 3 and 8 weeks are designated as recovery samples, and 6 samples collected after 8 weeks are designated as convalescent samples. Most of the samples were collected during the early pandemic between 2020 and 2021, before omicron infections. BAL and plasma of healthy controls were collected from separate studies or acquired from Audubon Biosciences (New Orleans, LA). BALF was obtained by centrifugation to remove cells and other insoluble debris in BAL samples. All samples were collected with patient consent and used without individual identifications. The clinical scores of COVID severity were assigned 1 through 8 corresponding to 1-2: non-hospitalized mild symptom for no oxygen needed and home oxygen use cases; 3-5: hospitalized moderate disease for no oxygen, no oxygen but ongoing care, and oxygen by mask or nasal prongs; 6-7: hospitalized severe disease for high flow non-invasive oxygen, mechanical ventilation; 8: death after hospitalization. Blood coagulation profiles for all clinical samples, including PT, activated PTT, INR, as well as concentrations of CRP, D-dimer, and fibrinogen, were measured within ~4 weeks of symptom onset using an automated hemostasis platform (ACLTOP750 CTS, Werfen) in the Department of Laboratory Medicine at the NIH Clinical Center. Other parameters on the clinical samples, including fibrinogen, prothrombin, additional D-dimer concentrations, and SARS-CoV-2 spike-specific IgG, were measured by ELISA in-house as described below.
## MATERIALS AND METHODS
## Clinical samples
## Reagent
ACE2-expressing HEK 293T cells (referred to as ACE2-293T) were purchased from Genecopoeia, Inc., MD. Normal human bronchoalveolar epithelial (NHBE, catalog PCS-300-010) cells, human small airway epithelial cells (HSAEC, catalog PCS-301-010) and their culture media components were purchased from American Type Culture Collection (ATCC, https://www.atcc.org).
## Production of SARS-CoV-2 pseudoviruses
For the production of pSARS-2, HEK 293T cells were cultured at a density of 2.5 × 10 6 cells in 10 cm plates in Dulbecco's Modified Eagle's Medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM L-glutamine, and 1% penicillin-streptomycin in a 37°C incubator with 5% CO 2 . Upon near confluence, cells were co-transfected with a SARS-CoV-2 spike protein plasmid and an env-/nef-GFP-expressing HIV NL4-3 core plasmid, in which the viral nef gene is replaced with an EGFP coding sequence, using Lipofectamine 3000 according to the manufacturer's protocol (54,55). Plasmids encoding SARS-CoV-2 spike genes, including Wuhan (56) and omicron (B.1.1.529) BA.2 strains, were obtained from Addgene (https://www.addgene.org). Supernatants containing pseudovirus particles were harvested 48 h post-transfection and concentra ted 100-fold using the PEG Virus precipitation kit (MAK343-1KT, Sigma-Aldrich Co., MO). The titer of SARS-CoV-2 pseudovirus was estimated by reverse transcription-PCR in numbers of RNA copies/mL. In brief, RNA was extracted from 50 uL concentrated pseudovirus using the Qiagen RNeasy Mini Kit, and cDNA was generated using a C1000 Touch Thermal Cycler (BIO-RAD, CA 94547) with ABI High-Capacity cDNA Reverse Transcription Kit following the manufacturer's protocol. HIV-1 NL4-3 LTR was amplified using TaqMan HIV-1 LTR primer/probe sets (Pa03453409_s1) from ThermoFisher with 50 ng cDNA as template. Samples were run in duplicate using a QuantStudio 6 Pro Real-Time PCR System (ThermoFisher, MA 02451) together with a serial dilution of a known copy number HIV DNA as standards. The pseudovirus titers were between 10 8 and 10 9 copies of RNA/mL. The use of SARS-CoV-2 viruses was approved by the NIH inter-institute Biosafety Committee (IBC) under protocol RD-20-VI-12.
## Infection of ACE2-293T cells with SARS-CoV-2 viruses
ACE2-293T cells were grown either in 384-well or 96-well cell culture plates at appro priate densities to near confluence. For SARS-CoV-2 pseudovirus infections, cells were infected with a titration dose of pSARS-2 between 1 and 100 copies of viral RNA/cell in fresh culture media for 24-72 h, and the infections were measured by the number of GFP+ cells. For antibody neutralization of infections, near-confluent ACE2-293T cells in a 384-well plate were infected with ~10-40 copies of viral RNA/cell in their growth media in the presence of titration amounts of antibodies between 0.2 and 70 nM or IgG between 1 and 1,000 nM.
## Fibrin clotting turbidity assay
Purified fibrinogen from human plasma (Sigma-Aldrich, MO) was dissolved in 100 mM NaCl, 20 mM HEPES buffer. The solution was incubated at 37°C for 10 min, then filtered through a 0.45 um syringe filter. The solution was stored at 4°C for 30 min, then filtered again to remove aggregates. Concentration was measured using nanodrop, then the solution was aliquoted and frozen at -20°C. Clot formation was assayed using fibrinogen solution diluted to 1.5 µM concentration in clotting buffer (20 mM HEPES, 137 mM NaCl, 5 mM CaCl 2 ). Diluted fibrinogen was added to thrombin enzyme (5 U/mL, Sigma) (positive control) or infected/uninfected HSAEC cells and placed in a plate reader. The absorbance was measured at 350 nm wavelength continuously with 2 min intervals for 4-10 h with Synergy_H1 (BioTek) plate reader. Fibrin clot formation causes scattering of light that passes through the solution, which increases the turbidity. For infectioninduced fibrin clotting assays, primary NHBE or HSAEC cells were grown in 384-wells at 1,000 cells/well or in a 96-well plate at 4,000 cells/well in their culture media, consisting of airway epithelial cell basal medium supplemented with bronchial epithelial growth kit as recommended by ATCC, until 60%-80% confluence. The cells were infected with various strains (Wuhan or omicron BA.2) of pSARS-2 at ~1-20 copies of RNA per cell for 24 h before replacing growth media with either fibrinogen or concentrated BALF in clotting buffer. To use BALF samples in the fibrin clotting assay, BALF samples were first concentrated 20-fold using 3 K molecular weight cutoff Amicon Ultra-4 concentration filters (Millipore catalog UFC800396) at 4°C to approximate the content in lung epithelial lining fluid. The growth media from the infected cells were replaced with a mixture of half volume of concentrated BALF and half volume of the clotting buffer. For confocal imaging, fibrin clotting assays were monitored by absorption at 350 nm wavelength for 2-4 h and then fixed with 2% paraformaldehyde before imaging by confocal microscope.
## Enzymatic cleavage of a fluorogenic fibrinogen-β peptide
The fluorogenic fibrinogen-β peptide, FPB, was synthesized by Biomatik as dabcyl-SARGHRPLE-edans, corresponding to amino acids 42-49 of human fibrinogen-β encompassing the thrombin cleavage site. The cleavage of FPB peptide was carried out by mixing 10 µM of the peptide with 20 µL of infected or uninfected HSAEC supernatants and 5 µL of the assay buffer containing 50 mM Tris, 0.01% Tween 20 at pH 9. The enzymatic cleavage reactions were monitored using a Synergy_h1 fluorescent plate reader (BioTek) with 340 nm excitation and 490 nm emission wavelengths for 4 h at 37°C.
## Proteomics analyses of BALF by mass spectrometry
Fifteen microliter aliquots of BALF samples were mixed with 5 µL 4× LDS-sample buffer and applied onto a 4%-12% Nupage gel with MOPs running buffer. The run stopped after the samples migrated approximately ¼ distance into the gel. Each lane of the gel was sliced into smaller pieces and subjected to destaining, reducing/alkylation, and in-gel trypsin digestion. The extracted peptides were applied for liquid chromatographytandem mass spectrometry analysis using either a Thermo Orbitrap Fusion or a Thermo Orbitrap Fusion Lumos operated with an in-line Thermo nLC 1200 and an EASY-Spray ion source. Peptides were separated using a 2 cm Pepmap 100 C18 trap column and a 25 cm Easy-spray Pepmap 100 C18 analytical column. MS/MS data acquisitions were operated at a 120,000 resolution (m/z 200) with a scan range of 350-1,950 m/z and CID fragmentation. All data were processed using Proteome Discoverer v2.4 (Thermo Scientific) with a SEQUEST HT search against the Uniprot KB/Swiss-Prot Human Proteome (02/2021) and common contaminants (theGPM.org) using a 5 ppm precursor mass tolerance and a 0.5 Da fragment tolerance. Dynamic modifications included in the search were limited to oxidation [M], deamidation [NQ], and acetylation [Protein N-terminal] while carbamidomethylation [C] was the only static modification utilized. Peptides and proteins were filtered at a 1% false discovery rate using a target-decoy approach with a two-peptide per protein minimum. Relative protein abundance was estimated from an average of its top three unique peptide intensities as determined by chromatographic area-under-the-curve and normalized by total intensity of all peptides. The sum of abundances in a data set is normalized to 1,000,000. The differential abundance is calculated as a percentage of difference in abundance: by dividing the difference in abundance between a protein in one sample and the average abundance of the protein by the average abundance of the protein from all healthy samples. The list of proteins used for the differential abundance heatmap analysis includes the ones with average healthy abundance greater than 25 and all non-zero abundance in the acute COVID sample. The heatmaps display the fold change in abundance relative to the average of each protein. Figure 3 is generated from analysis of fibrinogen and prothrombin in the quantitative proteomics data collected on acute, recovery, and convalescent BALF samples using data-independent acquisition mass spectrometry (DIA-MS) by Kanth et al. (46). The sum of DIA-MS abundances from each individual sample was similarly normalized to 1,000,000.
## Measuring fibrinogen, prothrombin, D-dimer, and IgG concentrations by ELISA
ELISA assays were used to determine the levels of fibrinogen (Abcam, ab108841), total IgG (Abcam, ab195215), prothrombin (Innovative Research, IHUFIIKTT), SARS-CoV-2 Spike IgG (Invitrogen, BMS2325), and D-dimer (Abcam, ab260076) present in human BALF and serum samples. The samples were diluted with kit-specific assay diluents. BALF sample dilutions ranged from D-Dimer (1:10 and 1:100), prothrombin, SARS-CoV-2 Spike IgG, and fibrinogen (1:50 and 1:500), and total IgG (1:1,000 and 1:10,000). Serum sample dilutions were ten times more dilute for all conditions. The assays were carried out following the manufacturer's protocols.
## SARS-CoV-2 neutralization antibody titer assay
SARS-CoV-2 neutralizing antibody titers were measured by a competitive ELISA assay based on the blocking of biotinylated ACE2 binding to immobilized SARS-CoV-2 RBD by serum neutralizing antibodies. The ELISA was performed and analyzed according to the manufacturer's instructions (Thermo Fisher Scientific, catalog BMS2326). In brief, samples were run undiluted (BAL-F) or diluted 1:50 (plasma). Positive inhibition is defined as greater than or equal to 20%, while negative results are defined as less than 20%. Inhibition calculations are based on the change in absorbance relative to the negative control wells.
## Imaging of fibrin fibers by confocal microscopy
Fibrinogen was labeled with a fluorescent dye TAMRA-SE (Thermo Fisher Scientific, catalog c1171) according to the manufacturer's protocol. Trace of fluorescent TAMRAfibrinogen was added to fibrin clotting assays at 80 µg/mL concentration or mixed with unlabeled fibrinogen at a 1:6 ratio. Images were taken on a Zeiss LSM 880 confo cal microscope equipped with Plan-Apochromat 20×/0.8 M27 objective. Z-stacks were performed to image fibrin formation. After acquisition, maximum intensity projections of the z-stacks were made using Fiji.
## References
1. Baden, Sahly, Essink et al. (2021) "Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine" *N Engl J Med*
2. Hall, Foulkes, Insalata et al. (2022) "Protection against SARS-CoV-2 after COVID-19 vaccination and previous infection" *N Engl J Med*
3. Nielsen, Helms, Schelde et al. (2022) "Vaccine effectiveness against SARS-CoV-2 reinfection during periods of Alpha, Delta, or Omicron dominance: a Danish nationwide study" *PLoS Med*
4. Huapaya, Higgins, Kanth et al. (2023) "Vaccination ameliorates cellular inflammatory responses in SARS-CoV-2 break through infections" *J Infect Dis*
5. Al-Aly, Bowe, Xie (2022) "Long COVID after breakthrough SARS-CoV-2 infection" *Nat Med*
6. Lin, Gu, Xu et al. (2022) "Association of primary and booster vaccination and prior infection with SARS-CoV-2 infection and severe COVID-19 outcomes" *JAMA*
7. Feikin, Higdon, Abu-Raddad et al. (2022) "Duration of effectiveness of vaccines against SARS-CoV-2 infection and COVID-19 disease: results of a systematic review and meta-regression" *Lancet*
8. Gan, Zhang, Wu et al. (2024) "Effect of inactivated vaccine boosters against severe and critical COVID-19 during the Omicron BA.5 wave: a retrospective analysis of hospitalized patients in China" *J Med Virol*
9. Wright, Tideman, Diaz et al. (2022) "Comparative vaccine effectiveness against severe COVID-19 over time in US hospital administrative data: a case-control study" *Lancet Respir Med*
10. Haas, Angulo, Mclaughlin et al. "Alroy-Preis S. 2021. Impact and effectiveness of mRNA BNT162b2 vaccine against SARS-CoV-2 infections and COVID-19 cases, hospitalisa tions, and deaths following a nationwide vaccination campaign in Israel: an observational study using national surveillance data" *Lancet*
11. Drury, Camara, Chelysheva et al. (2024) "Multi-omics analysis reveals COVID-19 vaccine induced attenuation of inflammatory responses during breakthrough disease" *Nat Commun*
12. Zhu, Gebo, Abraham et al. (2023) "Dynamics of inflammatory responses after SARS-CoV-2 infection by vaccination status in the USA: a prospective cohort study" *Lancet Microbe*
13. Fan, Shi, Yang et al. (2022) "Clinical characteristics and immune profile alterations in vaccinated individuals with breakthrough Delta SARS-CoV-2 infections" *Nat Commun*
14. Plante, Machado, Mitchell et al. (2022) "Vaccination decreases the infectious viral load of Delta variant SARS-CoV-2 in asymptomatic patients" *Viruses*
15. Grouprc, Horbyp, Limw et al. (2021) "Dexamethasone in hospitalized patients with COVID-19" *N Engl J Med*
16. Goligher, Bradbury, Mcverry et al. (2021) "Therapeutic anticoagulation with heparin in critically ill patients with COVID-19" *N Engl J Med*
17. Lawler, Goligher, Berger et al. (2021) "Therapeutic anticoagulation with heparin in noncritically ill patients with"
18. *N Engl J Med*
19. Kalil, Patterson, Mehta et al. (2021) "Baricitinib plus remdesivir for hospitalized adults with COVID-19" *N Engl J Med*
20. Li, Moser, Aga et al. (2025) *Chew Full-Length Text Journal of Virology*
21. Kw (2023) "Immune status and SARS-CoV-2 viral dynamics" *J Infect Dis*
22. Yan, Yang, Lai (2021) "COVID-19 vaccinations: a comprehensive review of their safety and efficacy in special populations" *Vaccines (Basel)*
23. Chen, Copland, Hirst et al. (2024) "Uptake, effectiveness and safety of COVID-19 vaccines in individuals at clinical risk due to immunosuppressive drug therapy or transplantation procedures: a population-based cohort study in England" *BMC Med*
24. Sherman, Tuan, Cantos et al. (2024) "COVID-19 vaccine efficacy in participants with weakened immune systems from 4 randomized controlled trials" *Clin Infect Dis*
25. Veras, Pontelli, Silva et al. (2020) "SARS-CoV-2-triggered neutrophil extracellular traps mediate COVID-19 pathology" *J Exp Med*
26. Vasquez-Bonilla, Orozco, Argueta et al. (2019) "A review of the main histopathological findings in coronavirus disease" *Hum Pathol*
27. Carsana, Sonzogni, Nasr et al. (2020) "Pulmonary post-mortem findings in a series of COVID-19 cases from northern Italy: a two-centre descriptive study" *Lancet Infect Dis*
28. Rapkiewicz, Mai, Carsons et al. (2020) "Megakaryocytes and platelet-fibrin thrombi characterize multiorgan thrombosis at autopsy in COVID-19: a case series" *EClinicalMedi cine*
29. Konopka, Nguyen, Jentzen et al. (2020) "Diffuse alveolar damage (DAD) resulting from coronavirus disease 2019 Infection is Morphologically Indistinguishable from Other Causes of DAD" *Histopathology*
30. Wauthier, Favresse, Hardy et al. (2023) "D-dimer testing: a narrative review" *Adv Clin Chem*
31. Johnson, Schell, Rodgers (2019) "The D-dimer assay" *Am J Hematol*
32. Olson (2015) "D-dimer: an overview of hemostasis and fibrinolysis, assays, and clinical applications" *Adv Clin Chem*
33. Nemec, Ferenczy, Christie Bd 3rd et al. (2022) "Correlation of D-dimer and outcomes in COVID-19 patients" *Am Surg*
34. Poudel, Poudel, Adhikari et al. (2021) "D-dimer as a biomarker for assessment of COVID-19 prognosis: D-dimer levels on admission and its role in predicting disease outcome in hospitalized patients with COVID-19" *PLoS One*
35. Lopez-Castaneda, García-Larragoiti, Cano-Mendez et al. (2021) "Inflammatory and prothrom botic biomarkers associated with the severity of COVID-19 infection" *Clin Appl Thromb Hemost*
36. He, Yao, Chen et al. (2021) "The poor prognosis and influencing factors of high D-dimer levels for COVID-19 patients" *Sci Rep*
37. Elbadawi, Elgendy, Sahai et al. (2021) "Incidence and outcomes of thrombotic events in symptomatic patients with COVID-19" *Arterioscler Thromb Vasc Biol*
38. Auron, Porres-Aguilar, Cameron (2022) "COVID-19 and elevated Ddimer: a tale of caution" *J Gen Intern Med*
39. Porfidia, Porceddu, Talerico et al. (2021) "Second wave of the COVID-19 pandemic: D-dimer levels are not so high anymore" *J Thromb Thrombolysis*
40. Myhre, Prebensen, Jonassen et al. (2021) "SARS-CoV-2 viremia is associated with inflammatory, but not cardiovascular biomarkers, in patients hospitalized for COVID-19" *J Am Heart Assoc*
41. Chandel, Patolia, Looby et al. (2021) "Association of D-dimer and fibrinogen with hypercoagulability in COVID-19 requiring extracorporeal membrane oxygenation" *J Intensive Care Med*
42. Erickson, Huang, Allen et al. (2023) "SARS-CoV-2 infection of human lung epithelial cells induces TMPRSS-mediated acute fibrin deposition" *Nat Commun*
43. Becker (2020) "COVID-19 update: COVID-19-associated coagulopathy" *J Thromb Thrombolysis*
44. Korte, Clarke, Lefkowitz (2000) "Short activated partial thrombo plastin times are related to increased thrombin generation and an increased risk for thromboembolism" *Am J Clin Pathol*
45. Thachil, Tang, Gando et al. (2020) "ISTH interim guidance on recognition and management of coagulopathy in COVID-19" *J Thromb Haemost*
46. Winter, Greene, Beal et al. (2020) "Clotting factors: clinical biochemistry and their roles as plasma enzymes" *Adv Clin Chem*
47. Brambilla, Canzano, Valle et al. (2023) "Head-to-head comparison of four COVID-19 vaccines on platelet activation, coagulation and inflammation. The TREASURE study" *Thromb Res*
48. Kanth, Huapaya, Gairhe et al. (2024) "Longitudinal analysis of the lung proteome reveals persistent repair months after mild to moderate COVID-19" *Cell Rep Med*
49. Dentone, Vena, Loconte et al. (2021) "Bronchoalveolar lavage fluid characteristics and outcomes of invasively mechanically ventilated patients with COVID-19 pneumonia in" *BMC Infect Dis*
50. Zeng, Chen, Yan et al. (2021) "Proteomic characteristics of bronchoalveolar lavage fluid in critical COVID-19 patients" *FEBS J*
51. Saris, Reijnders, Nossent et al. "ArtDECO consortium and the Amsterdam UMC COVID study group. 2021. Distinct cellular immune profiles in the airways and blood of critically ill patients with COVID-19" *Thorax*
52. George, Amjesh, Paul et al. (2021) "Evidence of a dysregulated vitamin D endocrine system in SARS-CoV-2 infected patient's lung cells" *Sci Rep*
53. Cho, Gonzales-Wartz, Huang et al. (2021) "Bispecific antibodies targeting distinct regions of the spike protein potently neutralize SARS-CoV-2 variants of concern" *Sci Transl Med*
54. Marking, Havervall, Norin et al. (2023) "Correlates of protection and viral load trajectories in omicron breakthrough infections in triple Full-Length Text Journal of Virology November"
55. *Nat Commun*
56. Bergwerk, Gonen, Lustig et al. (2021) "COVID-19 breakthrough infections in vaccinated health care workers" *N Engl J Med*
57. Landau, Page, Littman et al. (1991) "Human immunodeficiency virus type 1 viral protein R (Vpr) arrests cells in the G2 phase of the cell cycle by inhibiting p34cdc2 activity" *J Virol*
58. Shang, Ye, Shi et al. (2020) "Structural basis of receptor recognition by SARS-CoV-2" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792788&blobtype=pdf | Jacob Kocher, Nicole Jandick, Derry Spragion, Leslee Arwood, Rachel Roberts, Max Stockslager, ; Desena, Riley Hook, ; Womble, Nate Stasko
Background. The RD-X19 is an investigational medical device that targets 420 -430 nm light to the oropharynx and surrounding tissues to treat upper respiratory viral infections. The development of the RD-X19 was translated from the biological light unit (BLU), a proprietary high-throughput in vitro test platform. Clinical trials have indicated that the RD-X19 device can reduce the time to sustained symptom resolution and the time to negative antigen tests in clinical subjects 40 years of age and over with mild COVID-19. Visible light impacts SARS-CoV-2 through modulation of host and viral factors; therefore, we expect 420 -430 nm light biomodulation to be effective against other respiratory viruses, including influenza and respiratory syncytial virus (RSV).
Methods. In this study, we utilized both the BLU and RD-X19 to test whether 420 -430 nm light a) inactivated influenza A H1N1 and H3N2, influenza B, and RSV in cell-free suspensions and b) reduced viral titers in well-differentiated models of human airway epithelia (HAE). Cell-free suspensions of each virus were illuminated with 420 -430 nm light and viral titers were determined. HAE were inoculated with virus prior to twice-daily treatment with 420 -430 nm light. Apical washes were collected and viral titers were determined.
Results. H1N1 and H3N2 and influenza type B titers in cell-free suspensions. However, 420 -430 nm light reduced RSV titers in a dose-dependent manner. In HAE, 420 -430 nm light significantly reduced influenza A H1N1 and H3N2, influenza B, and RSV titers supporting that host factors play role. The illumination of RSV with the RD-X19 device induced similar dependent inactivation of RSV as those observed with the BLU.
Conclusion. These data demonstrate the potential of the RD-X19 as a novel treatment for upper respiratory infections regardless of the infecting pathogen. Based on the comparable benchtop findings in HAE for influenza and RSV as those in the clinically-evaluated SARS-CoV-2, we anticipate that the RD-X19 treatment should achieve clinical success against influenza and RSV and warrants further clinical investigation.
Disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792316&blobtype=pdf | # P-1663. The effectiveness of forward contact tracing and backward source investigation in controlling COVID-19 transmission: a case-series study in Yamagata, Japan, between 2020-2022
Takeaki Imamura, Yoshiharu Mori, Shunji Fujii, Emiko Suzuki, Keiko Yamada, Yoshihiro Ashino, Yuichi Kato, Hidetoshi Yamashita, Junji Seto, Mayuko Saito, Taro Kamigaki, Tomoe Shimada, Motoi Suzuki, ; Katsumi Mizuta, Tadayuki Ahiko, Hitoshi Oshitani
Background. The real-world data regarding the effectiveness of forward contact tracing and backward source investigation in controlling COVID-19 transmission remains limited.
Methods. We reviewed records of forward contact tracing and backward source investigation conducted in four public health centers (PHCs) in Yamagata, Japan, regarding COVID-19 cases reported between January 2020 and February 2022. A transmission setting was defined as where an epidemiological link occurred, and onward transmission as a transmission from one transmission setting to another. Cases identified and requested to stay at home based on PHCs' contact tracing and source investigation were classified into the Intervention group, cases who took COVID-19 tests and stayed at home upon diagnosis without PHCs' intervention into the Non-intervention group, and others into the Partial-intervention group. This study was approved by Yamagata Prefectural Institute of Public Health (No. 125), Tohoku University (2023-1-252), and the National Institute of Infectious Diseases (No. 1595), Japan.
Results. Among 8,627 COVID-19 cases reported in four Yamagata PHCs, transmission settings were identified in 6,517 cases (76%). Of 6,517 cases, 3,947 cases (61%) were classified into the Intervention, 2,199 cases (34%) into the Non-intervention, and 371 cases (6%) into the Partial-intervention group. The interval between symptom onset and the stay-at-home requests was 0 days [IQR: -1 -2] among the Intervention group, which was shorter compared to that among the Non-intervention (2 days [IQR: 1-3], p< 0.001) and the Partial-intervention group (1 day [IQR: 0-2], p< 0.001). Cases belonging to the Intervention (aOR 0.62 [95%CI: 0.52-0.75], p< 0.001) and the Partial-intervention group (aOR 0.62 [95%CI: 0.44-0.88], p=0.007) were less likely to generate onward transmission compared to those in the Non-intervention group.
Conclusion. In this study, COVID-19 cases with identified transmission settings receiving intervention based on PHCs' forward contact tracing and backward source investigation were associated with earlier stay-at-home requests and less likelihood of generating onward transmission than cases without contact-tracing/ source-investigation-based intervention.
Disclosures. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12497866&blobtype=pdf | # Lactylation of the SARS-CoV-2 spike protein is required for viral infection
## Dear Editor,
The primary mechanism by which SARS-CoV-2 invades the host is the binding of the spike (S) protein to the angiotensinconverting enzyme 2 (ACE2) receptor and subsequent membrane fusion. 1 During the cell entry process, SARS-CoV-2 S protein is cleaved into the S1 (responsible for ACE2 binding) and S2 (anchoring the S protein to the membrane and mediating membrane fusion) subunits by cellular proteases. Additionally, palmitoylation of the SARS-CoV-2 S protein enhances the binding affinity of the virus for the ACE2 receptor, thereby increasing viral infectivity. 2 However, whether the SARS-CoV-2 S protein undergoes other posttranslational modifications (PTMs) that regulate S protein function remains unclear.
Protein lactylation was initially identified as occurring on histone lysine residues in 2019, 3 which has been gradually observed on nonhistone proteins as well. 4 Recent studies have reported that lactylation of host proteins during viral infection affects virus replication and reactivation. 4 Although SARS-CoV-2 infection leads to intracellular lactic acid accumulation and muscle soreness, whether SARS-CoV-2-encoded viral proteins undergo lactylation and their role and regulatory mechanisms in SARS-CoV-2 infection have not been reported.
To determine whether increased lactate levels are closely associated with SARS-CoV-2-encoded viral proteins, we examined the lactylation levels of four structural proteins and found that the S protein underwent lactylation (Fig. 1a). Furthermore, exogenous lactate (sodium lactate, NaLa) treatment promoted the S lactylation level, whereas treatment with 2-DG, a glycolysis inhibitor that reduces the intracellular lactate concentration, or oxamate, a pyruvate analog that competitively inhibits the activity of lactate dehydrogenase (LDH), reduced the S lactylation level (Fig. 1a). These findings confirm that the S protein is subject to lactylation. The specific lactylation sites K424, K776, and K1028 on the S protein were subsequently identified through mass spectrometry (MS) analysis. By immunoprecipitation (IP) experiments, we confirmed that, compared with the wild-type S protein (S-WT), the S-K424R, S-K776R, and S-K1028R mutants presented reduced lactylation levels, with a cumulative effect observed when all three sites were simultaneously mutated (Fig. 1a).
To investigate whether lactylation of the S protein affects SARS-CoV-2 infection, luciferase-expressing pseudoviruses bearing either S-WT or S mutants (S-K424R, S-K776R, S-K1028R, or S-3KR) were generated in an envelope-defective HIV-1 backbone. The entry efficiency of these pseudoviruses was examined by analyzing luciferase activity in transfected HEK293T-ACE2 cells. Compared with the S-WT pseudovirus, which was normalized by quantitative PCR, the S protein mutant pseudovirus presented a reduction in luciferase activity (Fig. 1b), indicating that the entry of SARS-CoV-2 pseudoviruses is highly dependent on the S lactylation level. We further infected HEK293T-ACE2 cells with the SARS-CoV-2 BA.5 variant to examine the effect of the S lactylation level on virus replication. NaLa treatment also promoted SARS-CoV-2 replication (Fig. 1b), whereas lactylation inhibitors (2-DG and oxamate) had the opposite effect (Fig. 1b).
To further investigate the critical role of lactylation in vivo, BalB/c mice were treated with oxamate and subsequently infected with a mouse-adapted SARS-CoV-2 strain. Oxamate treatment significantly reversed weight loss and reduced the viral load in the lungs (Fig. 1c), accompanied by substantial alleviation of pulmonary injury (Fig. 1c). These findings collectively indicate that inhibition of S protein lactylation attenuates SARS-CoV-2 infectivity.
We next investigated the mechanism by which S-lactylation affects SARS-CoV-2 infection. S-mediated membrane fusion is critical for virus entry. Therefore, we investigated whether S-lactylation affects membrane fusion efficacy. Fluorescence imaging revealed that S-WT mediated cell-cell fusion well, whereas S mutants presented a significant reduction in the fusion area. Furthermore, NaLa treatment enhanced the fusion capacity of S-WT but had little effect on the lactylation-deficient mutants (Fig. 1d).
Since the S1 subunit of the S protein is responsible for ACE2 binding, we examined the binding ability of S-lactylationdefective mutants to ACE2. Compared with the S-WT protein, three mutants presented weaker binding to ACE2, especially K424R (located in the S1 subunit) (Fig. 1d). These findings suggest that lactylation may influence S and ACE2 interactions by neutralizing key charges, altering the RBD conformation. We also examined the interaction of S mutants with the transmembrane protease serine 2 (TMPRSS2), which is responsible for S2 site cleavage to expose the fusion peptide and promote membrane fusion. The results showed that the S protein efficiently interacted with TMPRSS2, while its mutants, especially K776R and K1028R (located in the S2 subunit), had weakened binding to TMPRSS2 (Fig. 1d). Overall, S protein lactylation is crucial for ACE2 and TMPRSS2 binding.
In summary, we identified lactylation as a previously unrecognized PTM occurring on the SARS-CoV-2 S protein that enhances S protein interaction with the host receptor ACE2 or the host protease TMPRSS2, thereby promoting viral entry and infection. However, our MS data and recent studies 5 revealed that multiple PTMs occur on the S protein, including coexisting lactylation and acetylation at K424, K776, and K1028. Therefore, cross-talk between different PTMs may exist, and a potential role for acetylation cannot be completely excluded. This study highlights the important role of lactylation in SARS-CoV-2 infection and provides new insights for the development of antiviral strategies. Further research is needed to identify the specific enzymes responsible for lactylation, which may help elucidate the regulatory networks involving lactylation and other PTMs in the modulation of S protein function.
Fig. 1 Lactylation of the S protein promotes viral infection. a Lactylation occurs on the S protein of SARS-CoV-2. S/N/M/E-Flag plasmids were transfected into HEK293T cells for 48 h, after which the cells were harvested, and the lysates were coincubated with protein G-agarose and Flag antibodies to enrich the structural proteins. Protein lactylation was probed with a Pan-Kla antibody via western blot (WB). Exogenous lactate treatment increases the lactylation level of the S protein. S-Flag was transfected into HEK293T cells, and the cells were then treated with NaLa (25 mM, 50 mM) 12 h before harvest. At 48 h posttransfection, the cells were collected for IP analysis to examine the lactylation level of the S protein. 2-DG or oxamate treatment inhibits S protein lactylation. S-Flag was transfected into HEK293T cells for 36 h, after which the cells were treated with 2-DG (5 mM, 10 mM) or oxamate (10 mM, 20 mM) for another 12 h before harvest. S-WT/K424R/K776R/K1028R/3KR-Flag were transfected into HEK293T cells for 48 h, after which lactylation levels were detected. b S lactylation is essential for the infectivity of SARS-CoV-2 pseudoviruses. HEK293T-ACE2 cells were infected with S-WT or mutant pseudoviruses for 72 h. Infection efficacy was analyzed by measuring firefly luciferase activity relative to the S-WT level (set as 1) (n = 3). Statistical analysis was performed via one-way ANOVA. Exogenous NaLa treatment promoted SARS-CoV-2 infectivity, whereas 2-DG or oxamate treatment inhibits SARS-CoV-2 infectivity. HEK293T-ACE2 cells were infected with the SARS-CoV-2 BA.5 variant and then treated with NaLa (10 mM, 25 mM, or 50 mM), 2-DG (2 mM, 5 mM, or 10 mM) or oxamate (5 mM, 10 mM, or 20 mM). N protein expression levels in cell lysates and N mRNA levels in the cell supernatant were assayed 48 h later. c Representative images of H&E-stained lungs from differently treated mice. (scale bar, 100 μm). Weights of the mice monitored over the experimental duration. Viral RNA loads in mouse lungs were detected at 7 dpi by measuring the mRNA levels of M, N and E. d S-lactylation is crucial for S-mediated membrane fusion. HEK293T cells coexpressing S-WT/K424R/K776R/K1028R/3KR-Flag and GFP were cocultured with HEK293T-ACE2 cells. Cell fusion was measured by fluorescence microscopy after 24 h (scale bar, 50 μm). The fusion areas were quantitatively analyzed via ImageJ software, and statistical analysis was performed via two-way ANOVA (* for comparisons versus the control group-WT; # for comparisons versus the NaLa group-WT; and † for comparisons between selected groups). S mutants show reduced binding to ACE2 and TMPRSS2. ACE2-HA or TMPRSS2-myc were transfected into HEK293T cells along with S-WT-Flag or its mutants for 48 h, after which protein interactions were detected by co-IP. (All subfigures follow the same statistical criteria. ns, not significant, *p < 0.05 (#, †), **p < 0.01 (#, †), ***p < 0.001 (#, †), ****p < 0.0001 (#, †))
## References
1. Jackson, Farzan, Chen et al. (2022) "Mechanisms of SARS-CoV-2 entry into cells" *Nat. Rev. Mol. Cell Biol*
2. Wu (2021) "Palmitoylation of SARS-CoV-2 S protein is essential for viral infectivity" *Signal Transduct. Target Ther*
3. Zhang (2019) "Metabolic regulation of gene expression by histone lactylation" *Nature*
4. Liu (2024) "Severe fever with thrombocytopenia syndrome virus induces lactylation of m6A reader protein YTHDF1 to facilitate viral replication" *EMBO Rep*
5. Liang (2023) "SARS-CoV-2 spike protein post-translational modification landscape and its impact on protein structure and function via computational prediction" *Research* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793634&blobtype=pdf | Leka Papazisi, Holly Asbury, Jason Bose, Pushpa Gujjari, Laksmi Castro, Trudy Corriea, Loricel Champ, Brian Chase, Sung-Oui Suh, Joseph Thiriot, Kyle Young, Victoria Knight-Connoni
## Abstract
Background. Human and highly pathogenic avian influenza (HPAI) viruses pose a major public health risk due to their potential for widespread illness and economic consequences. Early detection and control of outbreaks rely on effective surveillance and diagnostic testing.
Methods. ATCC® developed a comprehensive suite of quantitative synthetic analytical reference materials (ARMs) for HPAI virus serotypes H5N1, H5N6, H7N7, H7N9, and H9N2; human influenza A virus serotypes H1N1, H3N2, and H1N1 2009 pandemic; and Influenza B virus strains. Each synthetic ARM contains the complete sequences from segments 4, 5, 6, 7, and 8, including the HA, NP, NA, M1, M2, NS1, and NEP/NS1 genes, covering 50% of the influenza genome. These segments are key diagnostic targets for molecular tests and provide sufficient genomic context for assessing assay specificity. These ARMs are manufactured using a highly reliable synthetic biology technology, verified through next-generation sequencing, and quantitated via Droplet Digital PCR (Bio-Rad Laboratories, Inc). Further, they do not contain any viable material and can be handled in BSL-1 settings. As such, they are intended to serve as safe and reliable positive controls for molecular tests for surveillance and diagnostics. Poster Abstracts • OFID 2026:13 (Suppl 1) • S1117
Results. The synthetic ARMs were experimentally evaluated using several published quantitative PCR assays, including those from the Centers for Disease Control and Prevention, the World Health Organization, the World Organization for Animal Health, and other highly cited sources. We also conducted an in silico assessment of ARM compatibility with over 250 publicly available published assays. The synthetic products displayed equal performance to genomic RNA during all experimental tests.
Conclusion. The data demonstrate that all synthetic ARMs for avian and human influenza are effectively designed and suitable for developing and validating molecular-based detection and quantification assays. Furthermore, our findings indicate that the synthetic RNA ARMs are equivalent to their corresponding genomic RNA, making them a valuable BSL-1 alternative to BSL-3-derived materials. Our results suggest that these synthetic ARMs can serve as reliable and safe controls for molecular assays used in diagnostics and surveillance.
Disclosures. All Authors: No reported disclosures 1 1 1
1 University of Texas Medical Branch, Texas City, Texas 2 = 0.198) with a 95% confidence interval of [-85.58, -2.65]. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792686&blobtype=pdf | Min Lee, David Alfego, Laura Gillim, Charles Walworth, Suzanne Dale, Kathryn Lang, Ruth Carrico, ; Colm Smart, Payman Ghasemi
## Abstract
nd 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).
Change in yearly testing volumes for SARS-CoV-2 and influenza in adult and pediatric populations Monthly vaccination to testing volumes of SARS-CoV-2 and influenza in adult and pediatric populations Methods. We retrospectively reviewed 604,977 vaccination events and 881,573 tests for SARS-CoV-2 and 4,958,425 vaccination events and 31,274 tests for influenza from 2019 to 2024 in a US population administered seasonal vaccines with respiratory diagnostic testing performed at Labcorp. Monthly vaccination volume to testing volume ratio (mVTR) was used to estimate integration of vaccinations and diagnostic testing in health service systems. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793622&blobtype=pdf | # Abstract citation ID: ofaf695.2314 P-2151. Incidence of Breakthrough HSV in Adult Allogeneic Hematopoietic Cell Transplant Recipients on Standardized Antiviral Prophylaxis
Ria Mohan, ; Johnston, Amanda Phipps, Chris Davis, Agnes Ho, Michael Boeckh, Emily Ford, Denise Mcculloch, Ted Gooley, Frank Tverdek, Steven Pergam, Molly Fischer
Background. Reactivations of herpes simplex viruses (HSV) can occur in the early post-allogeneic hematopoietic cell transplant (aHCT) period despite universal antiviral prophylaxis. Few studies have assessed HSV recurrence in the era of standardized antiviral prophylaxis, in which val/acyclovir is recommended for up to 1 year post aHCT. We evaluated the incidence and management of HSV during the first 100 days post-HCT over two decades. Results. We reviewed data from 4,358 aHCT recipients aged ≥18 years, among whom 3,749 (86%) were HSV seropositive and 30 developed HSV recurrence (cumulative incidence = 0.8%). Most (n = 22) reactivations were HSV-1 and 8 were HSV-2; 2 were unspecified. The median time from transplant to first positive test was 35 days (IQR: 20-65 days) (Figure 1). HSV was detected at multiple anatomic sites; oral recurrences were most common. In total, 14/30 (46.7%) patients developed acyclovirresistant HSV (8 virologically-confirmed, 6 clinical). Treatment duration was significantly longer for patients with resistant (median 38 days [IQR: 26-41]) compared to susceptible infections (20 days [IQR: 16-27]; p = 0.03). Few patients (n = 4) had events attributed to non-adherence/malabsorption.
Conclusion. Clinical HSV disease is rare among aHCT patients on prophylaxis in the first 100 days. Development of acyclovir resistance is uncommon but represented almost half of breakthrough cases. Our findings highlight the importance and effectiveness of universal val/acyclovir prophylaxis in the early post-transplant period.
Disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793414&blobtype=pdf | # Abstract citation ID: ofaf695.2368 P-2205. Factors Associated with Human Herpesvirus 6 DNAemia Among Allogeneic Hematopoietic Cell Transplant Recipients in the Modern Era
Yusuke Ohashi, Roy Chemaly, Fareed Khawaja, Marilyne Daher, Amy Spallone, Tali Shafat, Ella Heredia, Guy Handley
## Abstract
Background. Human herpesvirus 6 (HHV-6) DNAemia frequently occurs following allogeneic hematopoietic cell transplantation (allo-HCT). Prior studies have established risk factors for DNAemia. However, advances such as haploidentical HCT, post-transplant cyclophosphamide (PtCy) for graft-versus-host disease (GVHD) prevention, and CMV prevention may have affected the risk of DNAemia.
Table 1. Baseline characteristics of allogeneic hematopoietic cell transplant recipients. Abbreviations: ALL, acute lymphocytic leukemia; AML, acute myeloid leukemia; ATG, antithymocyte globulin; CDV, cidofovir; CLL, chronic lymphocytic leukemia; CML, chronic myeloid leukemia; CMML, chronic myelomonocytic leukemia; CMV, cytomegalovirus; DNA, deoxyribonucleic acid; EBV, Epstein-Barr virus; FOS, foscarnet; GCV, ganciclovir; GVHD, graft-versus-host disease; HCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MMUD, mismatched unrelated donor; MRD, matched related donor; MUD, matched unrelated donor; NHL, non-Hodgkin lymphoma; PCR, polymerase chain reaction, PT-Cy, post-transplant cyclophosphamide; SD, standard deviation; SLL, small lymphocytic lymphoma; VGCV, valganciclovir. Results. Antiviral therapy was initiated for 188 (35.7%) patients with DNAemia, mostly with foscarnet (174/188). ACM within 48 weeks of transplantation was higher in patients with DNAemia (39.1% vs 28.3%; p< 0.001), and in those patients with DNAemia who received therapy (50.0% vs 33.0%; p< 0.001).
Conclusion. The prevalence of HHV-6 DNAemia was high in our cohort but with low incidence of encephalitis. Although, all cause mortality was higher in patients with HHV-6 DNAemia, patients who received therapy for DNAemia had worse outcomes. While receipt of post-transplant cyclophosphamide was associated with increased DNAemia, no association was found with ATG use and primary letermovir prophylaxis.
Disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12792906&blobtype=pdf | Jun Tachikawa, Yuta Aizawa, Rie Habuka, Kotaro Tsushima, ; Nur, Irma Safitri, Tatsuki Ikuse, Masaaki Kitajima, Akihiko Saitoh
Background. Enterovirus-D68(EV-D68) causes acute asthma-like respiratory illness in children. This re-emerging virus is classified into four clades (A-D) based on the sequence of the viral protein 1 (VP1) region. Clade B strains are predominantly isolated from children, and clade D strains are more frequently detected from adults. Prediction of the next EV-D68 outbreak has become more challenging following the pandemic of the coronavirus disease 2019 . Recently, wastewater surveillance has provided community-level monitoring and has shown potential for predicting outbreaks of viral diseases, including the COVID-19.
Mean concentration of enterovirus-D68 (EV-D68) RNA in wastewater from two wastewater treatment plants and weekly number of admitted pediatric patients with wheezing episodes in Niigata City, Japan, 2024 Black line indicates the mean concentration of enterovirus-D68 RNA in wastewater from two wastewater treatment plants (normalized with pepper mild mottle virus). Grey bars indicate the weekly number of children with wheezing episodes who were admitted to 6 hospitals in Niigata City, Japan, in 2024. Nasopharyngeal swab samples were collected from the admitted patients during week 37-52 of 2024. The first detection of EV-D68 in wastewater samples and pediatric patients was in week 27 and week 37, respectively.
Methods. Influent wastewater was collected once a week from two wastewater treatment plants (WWTP) in Niigata, Japan during January-December 2024. EV-D68 RNA was detected by using the Efficient and Practical virus Identification System with Enhanced Sensitivity for Membrane (EPISENS-M) method. Concentrations of EV-D68 RNA were normalized to those of pepper mild mottle virus (PMMoV). Following the detection of EV-D68 in wastewater samples, and an increase in pediatric hospitalizations for asthma-like respiratory illness was recognized by pediatricians in Niigata City, nasopharyngeal (NP) swab samples were collected from children admitted with wheezing episodes. VP1 Sanger sequencing and phylogenetic analysis were performed on EV-D68 positive samples. The number of admitted pediatric patients with wheezing episodes was also collected from 6 hospitals in Niigata City.
Results. EV-D68 RNA was detected in wastewater from week 27 (July), peaking in week 39 (September). Pediatric wheezing admissions peaked in week 37 (September), 2024. Wastewater EV-D68 RNA concentrations correlated with pediatric admissions (r = 0.48, P < 0.001). (Figure 1) Of 31 NP swab samples collected from weeks 37-52 (September-December), 16 (51.6%) were EV-D68 positive. All EV-D68 strains detected from NP swabs were in clade B3, whereas strains detected from the wastewater included clades B3 and D1. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12737613&blobtype=pdf | # Learning from the COVID-19 Pandemic-Through Sharing and Collaboration
Julian Tang
During the COVID-19 pandemic (which some argue is still ongoing), I led three Special Issues (SIs) that were designed to cover the talking points that arose amongst many practitioners and researchers: (1) How can we maintain our usual diagnostic services in the face of an ongoing pandemic? [1] (2) Where might the next pandemic virus come from and how might we first detect it? [2] (3) How has the pandemic changed the landscape of other viruses that affect human health? [3]. The papers that were published were broadly related to the main them of each SI, though some leeway was allowed.
In the first SI, Diagnostic Virology during the COVID-19 Pandemic-Business as Usual, my intention was to look at how 'non-SARS-CoV-2' testing was being maintained despite all the pandemic-related testing that was ongoing, so none of the papers focused on SARS-CoV-2 testing directly. The papers were selected to represent a general cross-section of what diagnostic laboratories normally do in non-pandemic times, and in some cases, how they maintained such services during the pandemic. These included assay evaluations [4,5], outbreak investigations [6][7][8], routine diagnostics and assay development [9][10][11], epidemiology and surveillance testing [12][13][14]. From these studies, it is clear that researchers and their laboratories were still managing to stay active and work on non-SARS-CoV-2 projects, demonstrating that they had some capacity to spare-which bodes well for any future pandemic.
The second SI, Animal and Human Respiratory Viruses-Causes of the Next Pandemic, arose from the fervent speculation around the WHO (World Health Organization) visit in early 2021 to Wuhan, China, to look for possible origins of SARS-CoV-2 [15]. The debate is still ongoing (and may never be fully resolved) about whether the virus was of natural origins, or whether it had been produced in a laboratory experiment and released by accident [16]. This SI featured papers examining potential pandemic threats from viruses originating in pigs (swine influenza) [17] and birds (avian influenza) [18][19][20], as well as the detection and surveillance of various human viruses, including RSV (respiratory syncytial virus) [21,22], influenza [23], and coronaviruses [24][25][26]. Again, whilst SARS-CoV-2 and coronaviruses were the focus of many studies, this also demonstrates that researchers were still able study the behaviour of other respiratory viruses that were still potential pandemic candidates.
Finally, in the third SI, Impact of Pandemic Measures on the Epidemiology and Seasonality of Other Viruses, as pandemic restrictions in most countries started to ease, researchers were asking the following question: how did the pandemic and its related restrictions affect the seasonality and behaviour of other respiratory viruses? This issue was perhaps most prominently highlighted with the global outbreaks of acute hepatitis in children, many of whom had been infected with adenoviruses [27]. Although none of the contributions to this SI included adenovirus, changes in the epidemiology of seasonal influenza [28][29][30],
RSV [31][32][33], and other respiratory viruses [34][35][36] were investigated. Some studies compared surveillance data from before and after the main pandemic period; others examined the viral epidemiology from just before or since the onset of the pandemic. The various findings were likely related to the study population's prior immunity to these seasonal viruses, which determined how they responded when these viruses returned after pandemic restrictions were lifted.
I would like to sincerely thank all the authors that contributed to these SIs, especially during such a busy time, when the demands on their time were likely multiple and unrelenting.
## References
1. Tang (2025) "Diagnostic Virology during the COVID-19 Pandemic-Business as Usual"
2. Tang (2025) "Animal and Human Respiratory Viruses-Causes of the Next Pandemic"
3. Tang (2025) "Impact of Pandemic Measures on the Epidemiology and Seasonality of Other Viruses"
4. Virant, Uršič, Kogoj et al. (1530) "Evaluation of Two Broadly Used Commercial Methods for Detection of Respiratory Viruses with a Recently Added New Target for Detection of SARS-CoV-2. Viruses"
5. Herrera, Mayoral, Brites (2022) "Development and Validation of a Rapid Screening Test for HTLV-I IgG Antibodies" *Viruses*
6. Dilcher, Howard, Dalton et al. (2022) "Clinical, Laboratory, and Molecular Epidemiology of an Outbreak of Aseptic Meningitis Due to a Triple-Recombinant Echovirus" *Viruses*
7. Howard-Jones, Pham, Jeoffreys et al. (1853) "Emerging Genotype IV Japanese Encephalitis Virus Outbreak in New South Wales, Australia. Viruses"
8. Lindblad, Hänninen, Valtonen et al. (2730) "Influenza A Outbreaks in Two Professional Ice Hockey Teams during COVID-19 Epidemic" *Viruses*
9. Bird, Taylor, Cafferata et al. (2022) "Performing under Pressure: Insights into the Diagnostic Testing Burden at a UK National Health Service Clinical Virology Laboratory during the SARS-CoV-2 Pandemic" *Viruses*
10. Kitai, Sato, Shirato et al. (2022) "Variation in Thermal Stability among Respiratory Syncytial Virus Clinical Isolates under Non-Freezing Conditions" *Viruses*
11. Teo, Norhisham, Lee et al. (2208) "Towards Next-Generation Sequencing for HIV-1 Drug Resistance Testing in a Clinical Setting" *Viruses*
12. Chon, Saito, Kyaw et al. (2023) ") and B/Victoria Viruses Detected in Myanmar during the COVID-19 Pandemic in 2021" *Genome Analysis of Influenza A*
13. Foley, Sikazwe, Minney-Smith et al. (2022) "An Unusual Resurgence of Human Metapneumovirus in Western Australia Following the Reduction of Non-Pharmaceutical Interventions to Prevent SARS-CoV-2 Transmission" *Viruses*
14. Wagatsuma, Koolhof, Saito (1417) "Was the Reduction in Seasonal Influenza Transmission during 2020 Attributable to Non-Pharmaceutical Interventions to Contain Coronavirus Disease 2019 (COVID-19) in Japan? Viruses"
15. (2021) "Briefing by the International Team Studying the Origins of the COVID-19 Virus-30"
16. (2025) "Independent Assessment of the Origins of SARS-CoV"
17. Padykula, Damodaran, Young et al. "Pandemic Risk Assessment for Swine Influenza A Virus in Comparative In Vitro and In Vivo Models" *Viruses*
18. Abubakar, Amrani, Kamarulzaman et al. (2023) "Avian Influenza Virus Tropism in Humans" *Viruses*
19. Liu, Zeng, Wu et al. (2023) "Global Prevalence and Hemagglutinin Evolution of H7N9 Avian Influenza Viruses from 2013 to 2022" *Viruses*
20. Guo, Zhou, Yan et al. "Molecular Markers and Mechanisms of Influenza A Virus Cross-Species Transmission and New Host Adaptation" *Viruses*
21. Foley, Minney-Smith, Lee et al. (2023) "Respiratory Syncytial Virus Reinfections in Children in Western Australia"
22. Wagatsuma, Koolhof, Saito (1914) "Nonlinear and Multidelayed Effects of Meteorological Drivers on Human Respiratory Syncytial Virus Infection in Japan" *Viruses*
23. Chow, Tay, Chen et al. (2023) "Influenza A and B Viruses in Fine Aerosols of Exhaled Breath Samples from Patients in Tropical Singapore"
24. Del-Puerto, Rojas, Díaz Acosta et al. (1136) "The Experience of Testing for Coronavirus Disease (COVID-19) at a Single Diagnostic Center in Paraguay before the Introduction of Vaccination" *Viruses*
25. Tambe, Mathobo, Munzhedzi et al. (2023) "Prevalence and Molecular Epidemiology of Human Coronaviruses in Africa Prior to the SARS-CoV-2 Outbreak: A Systematic Review" *Viruses*
26. Zabidi, Liew, Farouk et al. (2023) "Evolution of SARS-CoV-2 Variants: Implications on Immune Escape, Vaccination, Therapeutic and Diagnostic Strategies" *Viruses*
27. (2022) "Severe Acute Hepatitis of Unknown Origin in Children-Multicountry"
28. Chon, Win, Phyu et al. (1300) "Whole-Genome Analysis of the Influenza A(H1N1)pdm09 Viruses Isolated from Influenza-like Illness Outpatients in Myanmar and Community-Acquired Oseltamivir-Resistant Strains Present from 2015 to" *Viruses*
29. Escuyer, Gowie, St George (1952) "Influenza Virus Surveillance from the 1918 Influenza Pandemic to the 2020 Coronavirus Pandemic in New York State, USA. Viruses"
30. Jugulete, Olariu, Stanescu et al. (1576) "The Clinical Effectiveness and Tolerability of Oseltamivir in Unvaccinated Pediatric Influenza Patients during Two Influenza Seasons after the COVID-19 Pandemic: The Impact of Comorbidities on Hospitalization for Influenza in Children" *Viruses*
31. Yoshioka, Phyu, Wagatsuma et al. (2023) "Molecular Epidemiology of Respiratory Syncytial Virus during 2019-2022 and Surviving Genotypes after the COVID-19 Pandemic in Japan" *Viruses*
32. Cho, Kim, Mun et al. "Impact of COVID-19 Pandemic Restrictions on Respiratory Virus Patterns: Insights from RSV Surveillance in Gwangju, South Korea" *Viruses*
33. Foley, Minney-Smith, Tjea et al. (2017) "The Changing Detection Rate of Respiratory Syncytial Virus in Adults in Western Australia between" *Viruses*
34. Davids, Johnstone, Mendes et al. (2018) "Changes in Prevalence and Seasonality of Pathogens Identified in Acute Respiratory Tract Infections in Hospitalised Individuals in Rural and Urban Settings in South Africa"
35. Di Maio, Scutari, Forqué et al. "Presence and Significance of Multiple Respiratory Viral Infections in Children Admitted to a Tertiary Pediatric Hospital in Italy" *Viruses*
36. Reddy, Simane, Mthiyane et al. (1325) "Prevalence and Seasonal Patterns of 16 Common Viral Respiratory Pathogens during the COVID-19 Pandemic in Gauteng Province, South Africa"
37. "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=PMC12691608&blobtype=pdf | # The cryptic diversity of hepadnavirus relatives
Zhen Gong, Guan-Zhu Han, Eugene Koonin
## Abstract
Enveloped hepadnaviruses, non-enveloped nackednaviruses, and the recently discovered non-enveloped proto-nackednaviruses are closely related reversetranscribing DNA viruses. Yet, the host range and diversity of proto-nackednaviruses remain unclear. Here, we report the discovery of 31 proto-nackednaviruses (PnNVs) through deep genome-scale data mining. PnNVs are predominantly identified in distinct species of rotifers, implying the probable association between PnNVs and rotifers. Phylogenetic analyses show that PnNVs form at least five paraphyletic lineages sister to nackednaviruses and hepadnaviruses, unveiling the cryptic diversity of reverse-transcrib ing DNA viruses. This study provides insights into the origin, evolution, and diversity of reverse-transcribing DNA viruses.IMPORTANCE Hepatitis B virus belongs to Hepadnaviridae and represents a serious threat to global public health. Recently, non-enveloped proto-nackednaviruses were discovered to be the closest relatives of Hepadnaviridae. However, the host range and diversity of proto-nackednaviruses remain unclear, impeding our understanding of the origin, evolution, and diversity of Blubervirales and reverse-transcribing DNA viruses in general. This work expands the hidden diversity of proto-nackednaviruses and indi cates that rotifers might be their putative hosts. This work further traces the origin of Blubervirales back to 476 million years ago. These findings deepen our understanding of the evolution of reverse-transcribing DNA viruses.
B ased on the virus taxonomy of the International Committee on Taxonomy of Viruses, reverse-transcribing DNA viruses include two viral families, namely, Hepadnaviri dae (in the order Blubervirales) and Caulimoviridae (in the order Ortervirales) (1)(2)(3). Hepadnaviruses and caulimoviruses are enveloped viruses that infect vertebrates and non-enveloped viruses that infect land plants, respectively (2)(3)(4). Hepadnaviruses and caulimoviruses are distantly related and appear to have originated independently (1).
Hepatitis B virus belongs to Hepadnaviridae and represents a serious threat to global public health. Hepadnaviruses possess small (3.0-3.3 kb), circular, partially double-stran ded DNA genomes characterized by overlapping open reading frames (ORFs) encod ing core (C), polymerase (P), surface (S), and X proteins. Their replication is initiated by the terminal protein (TP) of P and proceeded through reverse transcription of an RNA intermediate, pregenomic RNA (pgRNA), by reverse transcriptase (RT), and RNase H (RH) of P (5). Enveloped hepadnaviruses have been found in mammals, birds, reptiles, amphibians, and fish (6)(7)(8)(9)(10)(11)(12). Non-enveloped viruses closely related to hepadnaviruses have been reported recently, including nackednaviruses discovered in fish and proto-nackednaviruses identified in bat feces and permafrost (9,13). Previous studies suggested an ancient origin for Blubervirales hundreds of millions of years ago (9,13). Moreover, we discovered retroelements (known as HEART elements) closely related to Blubervirales in invertebrates, supporting the origin of hepadnaviruses from retroelements (14). Yet, the host range and diversity of proto-nackednaviruses remain mysterious, impeding our understanding of the origin, evolution, and diversity of Blubervirales and reverse-transcribing DNA viruses in general.
Here, we report the discovery of 31 proto-nackednaviruses (PnNVs) through deep mining of public genome-scale data. PnNVs are probably associated with rotifers. Phylogenetic analyses show PnNVs form several paraphyletic lineages sister to hepad naviruses and nackednaviruses, expanding the diversity of Blubervirales. The origin of Blubervirales was traced back to 476 million years ago (MYA). Our findings have implications in understanding the origin, evolution, host range, and diversity of reverse-transcribing DNA viruses.
Following our previous attempt to unveil the evolutionary history of hepadnaviruses (14), we performed deep mining of hepadnavirus-like elements in a wide range of genome-scale data, including 5,500 animal genomes, 4,912 metatranscriptomes, 12,053 global metagenomes, and 9,549 eukaryote transcriptome assemblies. We identified a total of 31 hepadnavirus-related elements in the genomes and transcriptomes of rotifers (Rotaria macrura and Rotaria tardigrada), the transcriptome of one amphipod (Gammarus fossarum), 1 freshwater metagenome, and 17 environmental metatranscriptomes (Table S1; Data Set S1). These elements can be clustered based on the nucleotide identity of 95% and the coverage of 85% into 23 virus operational taxonomic units (15). Phyloge netic analyses based on RT domains show that these hepadnavirus-like elements cluster with proto-nackednaviruses, nackednaviruses, and hepadnaviruses with strong support (ultrafast bootstrap [UFBoot] value = 98%, Bayesian posterior probability [BPP] = 1.00) (Fig. 1A; Fig. S1 to S3). These elements form five paraphyletic lineages with strong support (UFBoot value >90%, BPP >0.90), designated PnNV-1 to PnNV-5. Pliego virus and Toolik virus, the two recently discovered PnNVs (13), pertain to the PnNV-3 lineage. Moreover, PnNV lineages are sister to nackednaviruses and hepadnaviruses. Together, the discovery of PnNVs unveils the cryptic diversity of Blubervirales. These five PnNV lineages might be putatively classified as the family level.
We assembled and annotated two complete genomes of PnNVs (Data Set S2) from rotifers: one putative pregenomic RNA (corresponding to ~3,249 nt in circular genome form) in the transcriptome of R. tardigrada (R. tardigrada PnNV [RTPV]) and one PnNV (~3417 nt) in the genome of R. macrura (R. macrura PnNV [RMPV]) (16). RTPV and RMPV pertain to PnNV-5 and PnNV-2 lineages, respectively (Fig. S2 andS3). The terminal redundancy and read mapping of the head-tail junction provide strong evidence for the circular and exogenous nature of RMPV (Fig. 2A). RMPV was predicted to encode four overlapping ORFs: two major ORFs encoding P and C proteins, ORF1 without detectable similarity to known proteins, and a small ORF (smORF2) of <300 nt. The RTPV genome comprises one major ORF encoding the P protein and two small ORFs (smORF1 and smORF2), but no C protein was predicted. Sequence-based and protein structure-based analyses did not identify high-confidence homologs to PnNV smORFs and RMPV ORF1. Like nackednaviruses and hepadnaviruses, the P proteins of RMPV and RTPV contain TP, RT, and RH domains. The spacer regions between TP and RT in PnNVs are much shorter than those in hepadnaviruses but similar to those in nackednaviruses (Fig. 2A). Two direct repeats (DR1 and DR2), which are essential for hepadnavirus replication, are present in both RMPV and RTPV (5). We also predicted RNA element epsilon (ε), which is crucial for reverse transcription priming, downstream of DR1 in both RMPV (64 nt) and RTPV (55 nt) (Fig. 2A) (17). Both RMPVε and RTPVε exhibit a hepadnavi ral ε-like stem-bulge-stem-loop structure, indicating potential similarity in replication mechanisms among PnNVs, nackednaviruses, and hepadnaviruses (Fig. 2B) (17)(18)(19). For RTPV, we reconstructed a linear pgRNA sequence of 3,471 nt. This linear pgRNA contains other critical cis-regulatory elements, including TATA box and polyA tail (Fig. 2C). Notably, the P ORF of RTPV has a large extension of around 300 amino acids downstream of the RH domain, without known homologs (Fig. 2C). Production of pgRNA from the circular genome results in a short terminal redundancy of ~33 nt between TATA box and polyA tail, which contains a second copy of DR1 (DR1 * ). Furthermore, conserved motifs of C proteins were identified in PnNVs, and the predicted secondary structures also suggest conservation in α-helix 4b and α-helix 5 of PnNVs, hepadnaviruses, and nackednaviruses (Fig. 2D; Data Set S3) (20). However, no homolog to C protein was detected in RTPV, even when alternative genetic codes were used during ORF prediction. Read mapping analyses excluded the possibility of assembly mistakes (Fig. S5). Several possibilities might account for it: (i) RTPV lacks a conventional core protein; (ii) RTPV requires an undiscovered helper virus/segment; and (iii) it represents an mRNA rather than a pgRNA. Nevertheless, these results reveal the conservation and diversity in genome structures among Blubervirales.
Phylogenetic relationship and similarity in genome structures provide strong evidence that PnNVs are the closest relatives of hepadnaviruses and nackednaviruses. The discovery of complete PnNV pgRNA and genomes in different rotifer species indicates the probable association between PnNVs and rotifers, expanding the host range of Blubervirales (8)(9)(10)(11).
Although rotifers acquire a substantial amount of genetic materials through horizontal gene transfer, these foreign genes are predominantly derived from bacteria, archaea, fungi, plants, and protists (21). Additionally, high abundance and sequencing average depth of RTPV and RMPV (the number of viral reads per million total filtered reads [RPM] >1) could largely reduce the possibility that PnNVs identified in rotifers were due to contamination during the sequencing process (Table 1). Nevertheless, further studies are still needed to verify the host range of PnNVs.
To infer the evolutionary timescale of Blubervirales, we reconstructed a time-calibra ted Bayesian phylogeny based on the P proteins using the integration time of an avihepadnavirus (eAHBV-FRY) in Neoaves genomes (69 MYA) as the calibration point (9, 11) (Fig. 1B; Fig. S4). The predicted divergence time between hepadnaviruses and nackednaviruses is 380 (95% HPD [highest posterior density]: 311-449) MYA, consistent with previous estimates (9), and PnNVs emerged at 476 (95% HPD: 381-565) MYA. The timescale suggests that Blubervirales likely infected vertebrates by cross-species transmission, as the inferred divergence time between PnNVs and hepadnaviruses/nack ednaviruses (407 [95% HPD: 332-478] MYA) is much shorter than the divergence time between rotifers and vertebrates (about 686 MYA) (22) (Table S2). Previously, we proposed a stepwise model for the origin and evolution of hepadna viruses, evolving from retroelements to non-enveloped viruses to enveloped viruses (14). Building on the newly discovered diversity of proto-nackednaviruses, we pro posed a refined stepwise evolutionary trajectory (Fig. 1C): PnNVs emerged from an escaped ancient invertebrate HEART retroelement. PnNVs infected vertebrates probably by cross-species transmission, leading to the origin of nackednaviruses in fish. Then, hepadnaviruses were derived from non-enveloped hepadnaviruses by gain of envelope. The discovery of PnNVs provides further insights into the origin and evolution of hepadnaviruses, bridging the evolutionary gap between retroelements and non-envel oped viruses.
PnNVs were identified in genome-scale data using similarity search and phylogenetic analysis combined approaches. Alignments were generated by MAFFT. Phylogenetic analyses were performed using FastTree, IQTREE, and MrBayes. Genome annotation involved sequence-based and structure-based searches by CD-Search, HHPred, and AlphaFold. Viral abundance is represented by the number of viral RPM. Reads were mapped to the viral genomes using Bowtie2. Time-calibrated Bayesian phylogeny was reconstructed by BEAST. See Supplemental Material for details.
## ADDITIONAL FILES
The following material is available online. S1 andS2.
## References
1. Krupovic, Blomberg, Coffin et al. (2018) "Ortervirales: new virus order unifying five families of reverse-transcribing viruses" *J Virol*
2. Magnius, Mason, Taylor et al. (2020) "ICTV virus taxonomy profile: Hepadnaviridae" *J Gen Virol*
3. Teycheney, Geering, Dasgupta et al. "Report Consortium I. 2020. ICTV virus taxonomy profile: Caulimoviridae" *J Gen Virol*
4. Gong, Han (2018) "Euphyllophyte paleoviruses illuminate hidden diversity and macroevolutionary mode of Caulimoviridae" *J Virol*
5. Beck, Nassal (2007) "Hepatitis B virus replication" *World J Gastroen terol*
6. Dill, Camus, Leary et al. (2016) "Distinct viral lineages from fish and amphibians reveal the complex evolutionary history of hepadnaviruses" *J Virol*
7. Gilbert, Feschotte (2010) "Genomic fossils calibrate the long-term evolution of hepadnaviruses" *PLoS Biol*
8. Hahn, Iwanowicz, Cornman et al. (2015) "Characterization of a novel hepadnavirus in the white sucker (Catostomus commersonii) from the Great Lakes region of the United States" *J Virol*
9. Lauber, Seitz, Mattei et al. (2017) "Deciphering the origin and evolution of Hepatitis B viruses by means of a family of nonenveloped fish viruses" *Cell Host Microbe*
10. Schaefer (2007) "Hepatitis B virus taxonomy and hepatitis B virus genotypes" *World J Gastroenterol*
11. Suh, Brosius, Schmitz et al. (2013) "The genome of a Mesozoic paleovirus reveals the evolution of hepatitis B viruses" *Nat Commun*
12. Suh, Weber, Kehlmaier et al. (2014) "Early mesozoic coexistence of amniotes and hepadnaviridae" *PLoS Genet*
13. Buigues, Viñals, Martínez-Recio et al. (2024) "Phylogenetic evidence supporting the nonenveloped nature of hepadnavirus ancestors" *Proc Natl Acad Sci*
14. Gong, Han (2018) "Insect retroelements provide novel insights into the origin of Hepatitis B viruses" *Mol Biol Evol*
15. Roux, Adriaenssens, Dutilh et al. (2019) "Minimum information about an uncultivated virus genome (MIUViG)" *Nat Biotechnol*
16. Nowell, Almeida, Wilson et al. (2018) "Compara tive genomics of bdelloid rotifers: insights from desiccating and nondesiccating species"
17. Wang, Seeger (1993) "Novel mechanism for reverse transcription in hepatitis B viruses" *J Virol*
18. Beck, Seitz, Lauber et al. (2021) "Conservation of the HBV RNA element epsilon in nackednaviruses reveals ancient origin of proteinprimed reverse transcription" *Proc Natl Acad Sci*
19. Knaus, Nassal (1993) "The encapsidation signal on the hepatitis B virus RNA pregenome forms a stem-loop structure that is critical for its function" *Nucleic Acids Res*
20. Pfister, Rabl, Wiegand et al. (2023) "Structural conservation of HBV-like capsid proteins over hundreds of millions of years despite the shift from non-enveloped to enveloped lifestyle" *Nat Commun*
21. Eyres, Boschetti, Crisp et al. (2015) "Horizontal gene transfer in bdelloid rotifers is ancient, ongoing and more frequent in species from desiccating habitats" *BMC Biol*
22. Kumar, Suleski, Craig et al. (2022) "TimeTree 5: an expanded resource for species divergence times" *Mol Biol Evol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793691&blobtype=pdf | Jon Salmanton-Garcia, Carla Niveyro, Pedro Villalba Apestegui, Victoria Martin, Maria Tomasino, Mariana Del, Carmen Marull, Claudia Patricia, Lorena Fernandez, Fernanda Tosin, Maria Hobecker, Karen Duranona, ; Angela, Daiana Jara, Oliver Cornely, Gustavo-Adolfo Méndez
## Abstract
Background. Dengue is a single-stranded RNA virus with four serotypes, transmitted primarily through the Aedes aegypti mosquito. Haematological malignancy (HM) patients are at heightened risk for severe dengue due to their immunocompromised state, yet the infection is often underdiagnosed in this group due to overlapping symptoms with their underlying condition or treatments.Methods. This retrospective multicentre study analysed HM patients diagnosed with dengue between November 2023 and May 2024. Using data from the DANGO registry, which documented cases from seven hospitals in Argentina during the dengue epidemic 2023-2024, the study included patients aged ≥ 16 years with confirmed dengue and existing HM, collecting data on clinical presentation, laboratory findings, and outcomes. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12656787&blobtype=pdf | # Long-Term Stabilization of Dengue Virus RNA at 37 • C for 14 Months Using Silk Fibroin Films
Nuo Wang, Ping He, Bohan Xu, Hongping Wei, Junping Yu
## Abstract
Diagnosis of dengue virus infections typically relies on RT-PCR-based methods, for which reliable positive controls are essential. Viral RNA is an ideal positive control, but its inherent instability poses a major challenge. Herein, we report a simple and effective method for stabilizing dengue virus RNA by immobilizing it onto silk fibroin films (RNA-SFFs). We evaluated various substrate surfaces for RNA-SFFs preparation and found that the inner surface of sealable bags is optimal for uniform film formation and easy harvesting. Screening different silk fibroin concentrations revealed that even low concentrations (2.8%) effectively preserved RNA well and kept Ct constant for up to 16 days at 25 • C, 37 • C, and even 45 • C (extreme weather for transportations). Due to its rapid film formation and ease of peeling, 7% silk fibroin was selected. Notably, the RNA-SFFs demonstrated robust resistance to UV irradiation, with no significant Ct value changes after 4 h of exposure. Longterm stability testing at -20 • C, 25 • C, and 37 • C showed that dengue serotype 1-4 RNA-SFFs remained stable for the entire duration of the study-up to 56 weeks (approximately 14 months)-at all tested temperatures. These results demonstrate that RNA-SFFs are highly stable, portable, and practical as positive controls for dengue diagnostics, with strong potential for use in on-site and resource-limited settings.
## 1. Introduction
Dengue, an arboviral disease caused by dengue viruses and transmitted by Aedes mosquitoes, can lead to mild or severe sickness. The symptoms include fever, aches (such as eye pain, typically behind the eyes, muscle, joint, or bone pain), nausea or rashes. Approximately 5% of individuals who develop dengue will progress to severe dengue, which may result in shock, internal bleeding, and even death [1,2]. The dengue virus comprises four distinct serotypes (DENV-1 to DENV-4), contributing to its complex epidemiology. In 2024, the World Health Organization's (WHO) reported a record 14.4 million dengue cases globally (https://worldhealthorg.shinyapps.io/dengue_global/, accessed on 15 October 2025), which is more than double the 7 million cases recorded in 2023 [3]. The real situation might be more serious. One modeling estimate indicates that dengue fever affects more than 400 million people annually worldwide and causes around 22,000 deaths across over 100 countries. Given the lack of specific antiviral treatment, early diagnosis continues to play a critical role in enabling timely treatment and minimizing the risk of severe dengue [2,4].
Reverse transcription-polymerase chain reaction (RT-PCR) remains the benchmark method for the early and accurate detection of dengue virus infection [5,6]. To ensure the reliability and validity of RT-PCR, the inclusion of positive controls or reference materials in each run is essential. Genomic RNA has become a widely adopted reference standard in molecular diagnostics and assay calibration due to its full-length genomic sequence, ease of production, and suitability for scalable manufacturing [7,8]. However, RNA molecules are inherently unstable and highly susceptible to enzymatic (e.g., ribonuclease-mediated) and chemical degradation (e.g., hydrolysis, oxidation), necessitating immediate stabilization via storage at ultra-low temperatures or under desiccated conditions [7,[9][10][11]. This requirement presents significant challenges for resource-limited settings.
To circumvent these constraints, various ambient-temperature stabilization strategies have been investigated. Lyophilization (freeze-drying) [12], encapsulation within metallic capsules [10,11], or entrapment in silica-based microparticles [13] offer enhanced stability but are often prohibitively expensive, labor-intensive, or ill-suited for deployment in resource-limited or field settings.
Silk fibroin is a natural protein polymer derived from the cocoons of the domesticated silkworm Bombyx mori. In its native state, silk fibroin adopts a semicrystalline architecture comprising approximately 65% crystalline domains and 35% amorphous regions, which synergistically contribute to the mechanical resilience and structural robustness of the cocoons. Silk fibroin exhibits high solubility in water and can be processed into a regenerated material with a water-stable structure, featuring a predominant β-sheet crystalline conformation. This conformation endows the protein-based matrix with exceptional thermal stability, mechanical strength under tension and resistance to chemical degradation [14,15]. Owing to these unique physicochemical properties-particularly its biocompatibility and ability to stabilize labile biomolecules-silk fibroin has emerged as a highly promising matrix for the encapsulation and preservation of thermosensitive biological agents, including whole blood, DNA, and RNA [14,[16][17][18][19][20][21].
In this study, silk fibroin is employed to form films incorporating dengue viral RNA, serving as reference materials to support the development and evaluation of RT-PCR-based dengue virus diagnostic assays. The resulting RNA-silk fibroin films (RNA-SFFs) effectively stabilize dengue viral RNA, with no significant change in RT-qPCR Ct values observed after storage for at least 14 months at 37 • C (longer durations have not yet been evaluated). RNA-SFFs prepared from various concentrations of silk fibroin maintain RNA stability at 25 • C, 37 • C and 45 • C for 16 days. Additionally, the RNA-SFFs exhibit resistance to UV radiation, as evidenced by stable Ct values following 4 h of exposure to UV light at an intensity of over 90 µW/cm 2 . These results suggest minimal RNA degradation under thermal and environmental stress. Therefore, the developed RNA-SFFs represent a promising cold-chain-free alternative for the long-term storage and transport of reference materials, with potential utility in resource-limited and field-deployable diagnostic settings.
## 2. Materials and Methods
## 2.1. Materials
Dengue virus serotypes 1 through 4 were obtained from clinical specimens derived from infected individuals and propagated in Aedes albopictus C6/36 cells (IVCAS9.087 in National Virus Resource Center, Wuhan Institute of Virology, CAS, Wuhan, China). C6/36 cells were cultured in T75 flasks using MEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. When cells reached 70-80% confluence, they were infected with DENV-1 to DENV-4 at a multiplicity of infection (MOI) ranging from 0.01 to 0.1. The infected cultures were incubated at 37 • C in a 5% CO 2 humidified incubator until pronounced cytopathic effect (CPE) was observed (3-4 days
## Primer or Probe
Sequence (5
$$′ -3 ′ ) DENV-1-F TGTGCATTGAAGCTAAAATATCA DENV-1-R CGTCTTGTTCTTCCACCA DENV-1-P FAM-ACCACCACCGACTCAAGATGTCCAA-BHQ1 DENV-2-F CGAGAAATACGCCTTTCAATA DENV-2-R CAGCATTCCAAGTGAGAATC DENV-2-P FAM-AACCGCGTGTCAACTGTGCAAC-BHQ1 DENV-3-F CAACCAACGGAAGAAGAC DENV-3-R CGCCAACTGTGATCCAGT DENV-3-P FAM-AAACCGTCTATCAATATGCTGAAACGC-BHQ1 DENV-4-F GGTTGGTGAAGAGATTCTCA DENV-4-R GTGGGATGGAAAGGACTC DENV-4-P FAM-AGCACCATCCGTAAGGGTCCT-BHQ1$$
## 2.2. Preparation and Quantification of Silk Fibroin Solution from Silk Cocoons
Silk fibroin solution was prepared following the established protocol described by Rockwood D. et al. [16], with minor modifications. Approximately 2.5 g of Bombyx mori silk cocoons were manually cut into fragments of nail-clipping size using sterilized scissors. To facilitate degumming, 2.12 g of sodium carbonate was dissolved in 1 L of ultrapure water, which was heated to boiling to dissolve. The cocoon pieces were introduced into the boiling sodium carbonate solution and stirred vigorously at 110-200 rpm for 50 min to remove sericin. Following cooling, the degummed silk fibers were retrieved, transferred to 1 L of fresh ultrapure water, and subjected to magnetic stirring for 20 min to remove residual salts. This washing step was repeated three times with complete water replacement. The purified fibers were then dried overnight in a fume hood. The dried silk fibroin was weighed (about 1.70 g). For dissolution, 8.1 g of lithium bromide (LiBr) was dissolved in 10 mL of ultrapure water. The dried fibers were solubilized in the LiBr solution at a ratio of 1:4 (for example, 1.7 g silk fibers to 6.8 mL LiBr solution), followed by incubation at 60 • C for 4 h to ensure complete dissolution of the protein matrix. The resulting solution was dialyzed against 1 L of ultrapure water at 4 • C for 72 h, with the dialysate replaced three times daily to eliminate residual LiBr. After dialysis, the aqueous silk fibroin solution was centrifuged at 9000× g and 4 • C for 20 min to remove insoluble aggregates. The supernatant was collected and subjected to a second centrifugation under identical conditions to ensure maximal clarity. The purified silk fibroin solution was stored at 4 • C for up to one month and used in subsequent experiments as required. The concentration of the prepared silk fibroin solution was determined as follows. A clean Petri dish lid was pre-weighed (mass recorded as m 1 in grams). Exactly 500 µL of the silk fibroin solution was pipetted onto the lid and dried at 60 • C for 1 h to constant weight. The lid with dried fibroin was weighed (mass recorded as m 2 in grams, and the mass difference in mass (m 2m 1 ). The concentration of the prepared silk fibroin solution is calculated as:
The unit of the protein concentration is in % w/v. For example, if the mass difference is 0.035 g, then the concentration is 0.035 divided by 0.5, which is 7.0% w/v.
$$c silk fibroin = m 2 -m 1 0.5$$
## 2.3. Evaluation of Film Formation Efficiency of the Silk Fibroin on Various Substrate Surfaces
The prepared silk fibroin solution in Section 2.2 was quantified and adjusted to a final concentration of 7.0% w/v. To investigate the influence of substrate properties on film formation and ease of harvest, 10 µL of RNase-free water was mixed with 10 µL of the 7.0% w/v silk fibroin solution. The resulting blend was cast onto various surfaces to form circular films with a diameter of approximately 1 cm, including a Petri dish (plastic), a Petri dish (glass), aluminum foil, the inner surface of an incised new sealable plastic bag, and the interior of microcentrifuge tubes. Samples were air-dried under ambient conditions to form thin solid films. The time required for complete film formation and the ease of film detachment were assessed for each substrate.
## 2.4. RT-qPCR System and Protocol
The RT-qPCR assay was carried out in a total reaction volume of 20 µL, comprising 5 µL of template RNA, 10 µL of 2× one-step reaction mix, 1 µL of enzyme mix, 0.4 µL each of 10 µM forward and reverse primers, 0.4 µL of 10 µM fluorescent probe, and nucleasefree water to achieve the final volume. The RT-qPCR reactions were run in duplicate. Amplification and fluorescence detection were performed using the CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) or MA-6000 (Yarui Biotech, Suzhou, China) under the following thermal cycling conditions: an initial reverse transcription phase at 50 • C for 15 min, followed by initial denaturation at 95 • C for 30 s. This was succeeded by 45 consecutive cycles of amplification, each consisting of a denaturation step at 95 • C for 10 s and an annealing/extension step at 60 • C for 30 s, with fluorescence acquisition occurring during the latter phase of each cycle. The Ct thresholds for all qPCR runs were set automatically by the instrument's software (Bio-Rad CFX Manager version 3.1, Hercules, CA, USA) or MA-6000 (Yarui Biotech, Suzhou, China).
## 2.5. The Effect of Silk Fibroin Films (SFFs) on Ct Values of DENV-1
Evaluation of the effect of silk fibroin on Ct values of DENV-1 RNA through RT-qPCR: To assess potential matrix-induced effects of silk fibroin on RT-qPCR quantification, a dilution series of silk fibroin solutions was prepared from 7.0% w/v using a two-fold serial dilution method, yielding concentrations of 7.0%, 3.5%, 1.75%, 0.88%, and 0.44% w/v. Here, a wide range of the silk fibroin concentrations were used to identify the threshold at which silk fibroin might begin to interfere with RT-qPCR. For each concentration, 10 µL of DENV-1 RNA was mixed with 10 µL of the respective silk fibroin solution, resulting in a 1:1 (v/v) RNA-silk matrix. The mixtures were then directly subjected to RT-qPCR analysis without prior processing.
Assessment of DENV-1 RNA stability in the form of RNA-silk fibroin films (RNA-SFFs) prepared from silk fibroin solutions of varying concentrations: Silk fibroin solutions at concentrations of 7.0%, 5.6%, 4.2% and 2.8% w/v were prepared by serial dilution of the stock solution with nuclease-free water (Source Leaf Biotech Co., Ltd., Shanghai, China), which are different from the concentrations used above due to that very low concentrations may compromise film integrity. The mixtures of 10 µL of DENV-1 viral RNA and 10 µL of each silk fibroin formulation were spread onto the inner surfaces of sealed polyethylene bags and air-dried under ambient conditions for 30 min to 1 h to yield thin, solid composite films. The resulting RNA-SFFs were transferred into 1.5 mL microcentrifuge tubes and stored under controlled thermal conditions at 25 • C, 37 • C and 45 • C. The preparation process of dengue virus RNA-silk fibroin films (SFFs) is illustrated in Figure 1. At days 1, 3, 6, 9, 13 and 16, individual samples were retrieved, and each film was rehydrated by adding 500 µL of RNase-free water. Samples were vortexed vigorously to ensure complete dissolution and homogeneous resuspension of RNA. The recovered RNA was then immediately subjected to RT-qPCR, and the cycle threshold (Ct) values were recorded.
## 2.6. Assessment of DENV-1 RNA Stability in the Form of RNA-SFFs
To evaluate the protective effect of RNA in the form of RNA-SFFs against UV-induced degradation, samples were exposed to UV radiation at an irradiance intensity exceeding 90 µW/cm 2 for durations of 1 h, 2 h, and 4 h. The irradiance level was chosen to reflect typical conditions used for surface decontamination in biological safety cabinets and clean benches. DENV-1 RNA-SFFs were prepared following the protocol outlined in Section 2.5, and naked DENV-1 RNA in solution served as the control. Dengue virus serotypes 1-4 RNA were individually encapsulated into SFFs using the protocol described above. The prepared RNA-SFFs were stored at -20 • C, 25 • C, and 37 • C to assess long-term RNA stability under varying thermal conditions. No special precautions against light exposure were taken during storage, and temperature fluctuations were kept to a minimum to ensure consistent experimental conditions. At 1, 2, 3, 4, 6, 8, 12, 24, 48 and 56 weeks, individual films were removed from storage and dissolved in RNase-free water according to the rehydration procedure mentioned above. The dissolved RNA was then subjected to serotype-specific RT-qPCR using corresponding primer-probe sets as listed in Table 1. Ct values were recorded at each time point to monitor Dengue RNA stability.
## 2.8. Data Statistics
Data are presented as the mean ± standard deviation (SD) derived from three independent replicates. Statistical significance between experimental groups at different temperature was assessed by one-way ANOVA (and nonparametric or mixed) with Dunn's multiple comparisons test using GraphPad Prism 9.0.0 (GraphPad Software Inc., San Diego, CA, USA). A probability value of p < 0.05 was designated as statistically significant.
## 3. Results
## 3.1. Selection of Appropriate Substrate Surfaces to Form SFFs
Initially, the film formation efficiency of silk fibroin was evaluated across various substrate surfaces. The results are summarized in Table 2. To optimize the preparation process of SFFs, an ideal substrate was sought that would enable the formation of intact, easily detachable films with a short air-drying time. As shown in Table 2, the inner surface of an incised new sealable plastic bag exhibited favorable characteristics: rapid air-drying, formation of structurally uniform films, and facile delamination without fragmentation. Given these advantages, this substrate was adopted for the preparation of RNA-loaded silk fibroin films (RNA-SFFs), where the RNA is embedded within the film matrix during casting and dried into a stable, peelable sheet for subsequent rehydration and use in RT-qPCR.
## 3.2. Assessment of the Effect of Silk Fibroin at Varying Concentrations on DENV-1 Genomic RNA Detection
We first evaluated the effect of silk fibroin on Ct values of DENV-1 viral RNA by RT-qPCR. The results are shown in Figure 2. The figure displayed no significant differences between DENV-1 viral RNA alone (the control group is DENV-1 RNA with 0% silk fibroin) and samples containing silk fibroin at various concentrations of the silk fibroin, even at 7% w/v (the corresponding final concentration of silk fibroin in the RNA samples was 3.5%). This indicates that silk fibroin had no obvious effect on the Ct value of DENV-1 viral RNA.
## 3.3. Evaluation of the Stability of DENV-1 RNA in SFFs Prepared with Silk Fibroin at Varying Concentrations
DENV-1 RNA was mixed with silk fibroin solutions of four silk fibroin concentrations (7.0%, 5.6%, 4.2%, and 2.8% w/v) and dried to form RNA-silk fibroin films (RNA-SFFs) to evaluate RNA stability in the SFF format under different storage conditions at 25 • C, 37 • C and 45 • C. The results are presented in Figure 3. The data, along with statistical p values and z values, are provided in the Supplementary Materials. A p value > 0.05 indicates no significant difference, while the z value reflects the magnitude of intergroup differences. The results indicated that at all tested concentrations, RT-qPCR Ct values remained stable throughout the 16-day storage period, even under the most thermally stressful condition (45 • C), indicating that silk fibroin effectively preserves RNA well in the SFF format for at least 16 days at elevated temperatures. Notably, RNA-SFFs prepared with 7.0% w/v silk fibroin exhibited the narrowest confidence intervals and the most consistent Ct values across all time points and temperatures, as evidenced by the smallest error bars in Figure 3D (with the smallest fluctuations for all the temperatures with 7.0% w/v silk fibroin (7.0%: 29.4 ± 0.45; 5.6%: 29.7 ± 0.64; 4.2%: 30.0 ± 0.75; 2.8%: 30.2 ± 0.61). This enhanced reproducibility is likely attributable to the formation of denser, more cohesive films at the higher silk fibroin concentration, which facilitated complete and consistent delamination from the substrate surface, thereby improving sample recovery uniformity and experimental repeatability. Based on these results, a silk fibroin concentration of 7.0% w/v was selected for subsequent experiments to ensure optimal film quality and assay consistency.
## 3.4. UV Resistance of DENV-1 RNA-SFFs
DENV-1 RNA-SFFs, prepared according to the aforementioned protocol, were exposed to UV radiation at an irradiance intensity exceeding 90 µW/cm 2 , which was selected based on the typical irradiance level used for surface decontamination in biological safety cabinets and clean benches, for 1 h, 2 h and 4 h to evaluate the protective capacity of DENV-1 RNA-SFFs against UV-induced RNA degradation. In contrast to the RNA-SFFs, naked DENV-1 RNA in solution exhibited a time-dependent increase in RT-qPCR Ct values with prolonged UV exposure (Figure 4), indicating progressive RNA fragmentation and reduced amplifiability. By contrast, RNA within the silk fibroin films maintained stable Ct values across all time points, with no statistically significant differences observed after 1, 2, or 4 h of irradiation (p > 0.05). These results demonstrate that the silk fibroin films effectively shields RNA from UV-induced photodegradation. The RNA molecules are preserved well within the film, likely due to the physical barrier and radical-scavenging properties of the silk protein network. Thus, silk fibroin confers robust protection against environmental UV stress, highlighting its potential as a stabilizing biomaterial for RNA preservation under challenging conditions.
## 3.5. Long-Term Stability of Dengue Serotypes 1-4 RNA in RNA-SFFs Under Different Temperature Conditions
To evaluate the long-term stability of dengue viral RNA in RNA-SFFs, DENV-1-4 RNA-SFFs were stored under three distinct temperature conditions: -20 • C, 25 • C and 37 • C, for an extended period of 56 weeks (approximately 14 months). All data, including statistical p and z values, are available in the Supplementary Materials. Plots of Ct values vs. weeks are presented in Figure 5. The mean Ct values for the first six time points in Figure 5 are from three independent experiments, while the data for the last four time points are from two independent experiments. As shown, DENV-2 RNA (Figure 5B) exhibited a lower Ct value at the one-week time point compared to subsequent measurements. This may reflect an initial measurement variability or a transient shift in Ct values between week one and week two, after which the values stabilized and remained consistent for the remainder of the study. No significant differences were observed in Ct values between any testing time point and the reference time point (week one, or week two for DENV-2); p > 0.05, see Supplementary Materials (Data for Figure 5). Notably, all four dengue serotypes demonstrated excellent stability over the 56-week duration, with minimal fluctuation in Ct values across all storage temperatures. Importantly, statistical analysis revealed no significant differences in Ct values between samples stored at -20 • C and those maintained at 25 • C or 37 • C at any tested time points (p > 0.05). The p values are listed in the Supplementary Materials (Data for Figure 5). This indicates that RNA-SFFs effectively preserve RNA even under elevated, non-frozen conditions. Collectively, these findings demonstrate that RNA-SFFs provide robust protection for multivalent dengue viral RNA over prolonged periods, maintaining molecular stability for up to 56 weeks without the need for cold-chain storage. These results highlight the potential of SFFs as a versatile and reliable platform for ambient-temperature biostabilization of RNA-based diagnostics.
## 4. Discussion
Positive controls play a pivotal role in RT-PCR-based assays for the diagnosis of viral infections. Dengue viral genomic RNA serves as a critical positive control with advantages of containing all targets of dengue viruses to validate all RT-PCR-based diagnostic kits. However, RNA is inherently labile and highly susceptible to degradation by RNases, chemicals and environmental stressors such as heat and UV radiation. The storage and transport of RNA-based controls depend on cold-chain infrastructure, which limits the practical utility in resource-limited or field-deployable diagnostic settings.
In this study, we demonstrate that dengue RNA-SFFs can effectively stabilize dengue genomic RNA under challenging conditions. Specifically, RNA in SFFs remains stable for up to 14 months at 37 • C, up to 16 days at 45 • C, and for a minimum of 4 h under intense UV irradiation (>90 µW/cm 2 ). These results highlight the robust protective capacity of silk fibroin matrices against both thermal and photodegradation.
Notably, we found that silk fibroin, even at a final concentration of 3.5% w/v (derived from the original 7.0% w/v solution shown in Figure 2), did not significantly affect RT-qPCR Ct values. This observation contrasts with a previous report [21], in which silk concentrations exceeding 1% w/v were shown to interfere with Ct values, although this interference was mitigated by RNA purification. The discrepancy may stem from differences in experimental conditions, including variations in silk fibroin preparation methods, PCR master mix composition, or RNA type. Importantly, in our system, RNA can be directly detected from the dissolved films without prior extraction, enabling a simpler and more user-friendly workflow. Consistent with our findings, RNA in SFFs remained stable at elevated temperatures, including up to 45 • C, further supporting the potential for using RNA-SFFs-based reference materials in extreme or resource-limited environments.
To the best of our knowledge, stability for 14 months at 37 • C represents the longest reported storage duration for viral RNA reference materials under non-refrigerated conditions to date. A recent study [10] demonstrated that RNA encapsulated in dehydrated form within metallic capsules remained stable for up to 3 years at room temperature, as assessed by RT-qPCR. However, that method relies on proprietary technology developed by a commercial entity. Due to intellectual property restrictions and cost considerations, this approach is not readily adaptable for use in independently developed diagnostic assays or in-house testing systems. In contrast our method demonstrated consistent RNA protection across all four dengue virus serotypes, indicating that stability is maintained regardless of genomic sequence variation. This pan-serotypic efficacy underscores the versatility and generalizability of the silk fibroin matrix, making it particularly valuable for multivalent diagnostic kits where balanced detection sensitivity across serotypes is essential.
Several limitations of this study should be acknowledged. First, while this work provides proof-of-concept, the development of RNA-SFFs-based materials into standardized reference materials will require precise quantification of dengue viral RNA copy number in the films, followed by multi-laboratory validation to ensure reproducibility and comparability across testing sites. Second, although the materials exhibited excellent long-term stability, the upper limits of stability-particularly under higher temperatures and more extreme environmental conditions-have not yet been fully defined. Further studies using accelerated aging protocols (e.g., elevated temperatures, variable humidity) are needed to establish shelf life and define optimal storage boundaries. As well, the cost-effectiveness and scalability of large-scale production, and the effects of different batches of silk fibroin preparations, and stability tests by different persons, require further investigation. In addition, the evaluation of RNA stability across a broad dynamic range, particularly at low-copy inputs (e.g., Ct > 30) that are representative of weak-positive or near-limit-of-detection diagnostic controls should be studies further. Moreover, the molecular mechanisms underlying the exceptional thermal stability of silk fibroin-RNA films, especially at temperatures up to 45 • C, remain unclear. A deeper understanding of the interactions between silk fibroin and nucleic acids, as well as the role of water exclusion dynamics during film formation could guide the rational design of more robust biomaterials. Such insights may not only improve RNA preservation but also expand the application of silk fibroin in stabilizing other labile biologics such as vaccines and biopharmaceuticals, thereby contributing to global health.
## 5. Conclusions
Genomic RNA-based positive controls are essential for current dengue virus diagnosis but are inherently labile and require stringent cold-chain storage, limiting their use in tropical and resource-limited settings. To overcome this limitation, we developed a silk fibroin film (SFF)-based platform for the stabilization of full-length dengue viral RNA. Our results show that RNA in RNA-SFFs exhibited no significant increase in RT-qPCR Ct values across all four serotypes (DENV-1-4) after more than 14 months of storage at 37 • C. Furthermore, the RNA-SFFs exhibit robust resistance to both UV irradiation and elevated temperature (45 • C for at least 16 days). These findings establish silk fibroin as a highly effective matrix for ambient-temperature preservation of viral RNA. The RNA-SFFs platform offers a promising strategy for developing ready-to-use, cold-chain-free positive controls, thereby enhancing the reliability and accessibility of molecular diagnostics in dengue-endemic regions.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v17111452/s1, Table S1: Amplification efficiencies and limit of detection of the primer/probe sets; Data for
## References
1. Kothari, Patel, Bishoyi (2025) "Dengue: Epidemiology, diagnosis methods, treatment options, and prevention strategies" *Arch. Virol*
2. Pourzangiabadi, Najafi, Fallah et al. (2025) "Dengue virus: Etiology, epidemiology, pathobiology, and developments in diagnosis and control-A comprehensive review" *Infect. Genet. Evol. J. Mol. Epidemiol. Evol. Genet. Infect. Dis*
3. Haider, Hasan, Onyango et al. (2025) "Global dengue epidemic worsens with record 14 million cases and 9000 deaths reported in 2024" *Int. J. Infect. Dis. IJID Off. Publ. Int. Soc. Infect. Dis*
4. Bhatt, Gething, Brady et al. (2013) "The global distribution and burden of dengue" *Nature*
5. Pabbaraju, Wong, Gill et al. (2016) "Simultaneous detection of Zika, Chikungunya and Dengue viruses by a multiplex real-time RT-PCR assay" *J. Clin. Virol. Off. Publ. Pan Am. Soc. Clin. Virol*
6. Liu, Chen, Lin et al. (2025) "Performance comparison of two dengue NS1 rapid diagnostic tests against RT-PCR: Sensitivity and specificity across infections and timeframes" *J. Infect. Public Health*
7. Zheng, Xu (2025) "Pseudovirus as an Emerging Reference Material in Molecular Diagnostics: Advancement and Perspective" *Curr. Issues Mol. Biol*
8. Mattiuzzo, Bentley, Page (2019) "The Role of Reference Materials in the Research and Development of Diagnostic Tools and Treatments for Haemorrhagic Fever Viruses" *Viruses*
9. Li, Guo, Qi et al. (2025) "Nucleic acid-based reference materials: A "ruler" for precision molecular diagnostics" *TrAC Trends Anal. Chem*
10. Colotte, Luis, Coudy et al. (2025) "Room temperature storage and shipping of encapsulated synthetic RNAs as quality control materials for SARS-CoV-2 molecular diagnostic assays" *J. Virol. Methods*
11. Fabre, Colotte, Luis et al. (2014) "An efficient method for long-term room temperature storage of RNA" *Eur. J. Hum. Genet. EJHG*
12. Pisani, Le Tallec, Costanzo (2023) "Establishment of Ph. Eur. Hepatitis C Virus RNA for NAT testing BRP batch 2" *Pharmeuropa Bio Sci. Notes*
13. Puddu, Stark, Grass (2015) "Silica Microcapsules for Long-Term, Robust, and Reliable Room Temperature RNA Preservation" *Adv. Healthc. Mater*
14. Reizabal, Costa, Pérez-Alvarez et al. (2023) "Silk Fibroin as Sustainable Advanced Material: Material Properties and Characteristics, Processing, and Applications"
15. Reizabal, Costa, Saiz et al. (2021) "Processing Strategies to Obtain Highly Porous Silk Fibroin Structures with Tailored Microstructure and Molecular Characteristics and Their Applicability in Water Remediation" *J. Hazard. Mater*
16. Rockwood, Preda, Yucel et al. (2011) "Materials fabrication from Bombyx mori silk fibroin" *Nat. Protoc*
17. Li, Kluge, Guziewicz et al. (2015) "Silk-based stabilization of biomacromolecules" *J. Control. Release Off. J. Control. Release Soc*
18. Wani, Zargar, Masoodi et al. (2022) "Silk Fibroin as an Efficient Biomaterial for Drug Delivery" *Gene Therapy, and Wound Healing. Int. J. Mol. Sci*
19. Liu, Zheng, Gong et al. (2017) "DNA preservation in silk" *Biomater. Sci*
20. Nyaruaba, Hong, Li et al. (2022) "Long-Term Preservation of SARS-CoV-2 RNA in Silk for Downstream RT-PCR Tests" *Anal. Chem*
21. He, Yavuz, Kluge et al. (2018) "Stabilization of RNA Encapsulated in Silk" *ACS Biomater. Sci. Eng*
22. "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=PMC12608847&blobtype=pdf | # RSV Monitoring in Germany: A Critical Overview of Available Surveillance Systems
Ioannis Vogiatzis, Adriano Peris, Lea Bayer, Christian Brösamle, Gordon Brestrich, Bahar Najafi, Christof Von Eiff, Cornelia Hösemann, Holger Stepan, Gunther Gosch, Michael Wojcinski, Michael Abou-Dakn, Egbert Herting, Markus Rose, Martina Prelog, Rolf Kaiser
## Abstract
Respiratory syncytial virus (RSV) is a leading cause of respiratory infections in young children, elderly people, and patients with underlying diseases. Solid data on its epidemiology and burden of disease are essential for the implementation of preventive strategies. This review provides for the first time a comprehensive overview on publicly available RSV surveillance resources in Germany. Methods: Public RSV surveillance systems in Germany were identified and, where possible, exemplary data was extracted to provide an overview of the scope of available data, their strengths and limitations. Results: German RSV surveillance systems provide data on both outpatient and inpatient incidence rates, age distribution, and seasonality. Germany's public health institution, the Robert Koch Institute (RKI), documents RSV cases nationwide based on mandatory reporting. Further, sentinel surveillance by RKI captures outpatient RSV infections as well as severe hospitalized cases. Nationwide, data on inpatients is collected and reported by hospital discharge diagnostic codes. Additional surveillance systems (e.g., clinical-virology.net) provide data on RSV positivity rates stratified by age and gender. Regional surveillance efforts by ten German states provide data on the infection dynamics. Pediatric documentation of age distribution and severity of respiratory diseases via surveillance was initiated by the German Society for Pediatric Infectious Diseases. Reviewing all available sources and data underlines the high clinical burden, especially in infants and older adults during the winter season. Conclusions: Germany's RSV surveillance systems on the national and regional level support the tracking of incidence rates and seasonal patterns. Notably, pediatric data collection is more thorough, yielding a more comprehensive dataset than that available for adults. Contextualizing reported incidence rates in light of prospective or modeling studies suggests that the official documentation of RSV cases-particularly among adults-is underestimated.
## 1. Introduction
Respiratory Syncytial Virus (RSV), a member of the Pneumoviridae family, is a negative-sense, single-stranded enveloped RNA virus that regularly causes seasonal epidemics [1]. It exists in two subtypes, RSV A and RSV B, with both subtypes typically cocirculating and one subtype-mostly RSV-A-dominating each year [2,3]. RSV replicates in the epithelial cells of the upper respiratory tract and can also infect the bronchioles and/or alveoli, leading to lower respiratory tract illness (LRTI) with conditions such as bronchitis, bronchiolitis, and pneumonia. Common clinical symptoms of RSV infection (RSVI) include, but are not limited to, fever, rhinitis, cough, dyspnea, apnea, and chest tightness [1]. Additionally, neurotopic infections may occur with seizures and other symptoms of the central nervous system [4,5]. As RSV is highly contagious and is transmitted at any age through contact with saliva, contaminated surfaces, or respiratory droplets, approximately 50-70% of all children contract it within their first year of life. Nearly all children have had at least one RSV infection by the end of their second year of life [6]. Even mild RSV infections during infancy are associated with an increased risk of childhood asthma [7]. While most RSVIs result in only mild, common cold-like symptoms, a significant proportion will cause acute and severe respiratory tract illnesses, especially in premature and young infants, chronically ill individuals, and elderly subjects. The latter face a particularly high risk of experiencing cardiovascular complications, such as myocardial infarction or stroke due to activation of the thrombotic system by systemic inflammation caused by RSV [8][9][10][11]. Known risk factors for severe RSVI include underlying medical conditions such as cardiopulmonary diseases, a compromised immune response, prematurity, and very young or old age [9,[12][13][14][15]. Current treatment options for RSVI are limited to the management of symptoms. After infection, individuals do not develop long-term protective immunity, making reinfections common. RSVI thus presents a significant health risk and has a substantial financial impact on the health care system [16][17][18]. In June 2024, Germany's Standing Committee on Vaccination (STIKO) recommended RSV immunization for all infants in their first RSV season using the monoclonal antibody Nirsevimab [19]. This was followed by the STIKO recommendation of RSV immunization for all adults from 75 years of age as well as for adults aged 60-74 years with increased risk for severe RSV infections due to certain co-morbidities. The authors of STIKO's justification of its recommendation underline the underestimation of RSV in adults, emphasizing the need for a deeper understanding of the virus' epidemiology and its impact on patients' lives [20].
In Germany, several distinct nationwide or state-level public surveillance systems track RSVI dynamics. On the national level, the Robert Koch Institute (RKI) compiles data on RSV epidemiology [21][22][23]. In addition, the Federal Health Monitoring Information System (Gesundheitsberichterstattung des Bundes, GBE-Bund) compiles routine administrative data on RSV-related hospitalizations from nearly all hospitals in Germany and also provides official German Cause of Death statistics [24]. Similar data are collected in InEK data, including detailed information on hospital cases and treatments [25]. The German Society for Pediatric Infectious Diseases (DGPI) started collecting data on hospitalized RSV cases among children from children's hospitals and pediatrics departments in October 2021 [26][27][28][29]. The RespVir project, which is carried out by the Clinical Virology Network, collects and processes diagnostic data on a variety of respiratory infections from a set of laboratories in Germany, Austria, and Switzerland [30].
This review provides the first comprehensive synthesis of publicly available monitoring systems on respiratory syncytial virus (RSV) epidemiology in Germany and systematically evaluates them against a set of relevant criteria. We explore the range of existing databases and surveillance structures, examining their underlying methodologies, data collection strategies, and reporting mechanisms. Particular attention is given to the accuracy, representativeness, and timeliness of the reported data, as well as the extent to which these sources can be reliably used for different purposes, such as clinical decision-making, public health surveillance, and policy planning.
Beyond a technical assessment, we also discuss the strengths and limitations of each monitoring system, highlighting gaps in coverage, potential biases, and opportunities for improvement. In doing so, this review sheds light on how these data sources complement or overlap with one another, and how they may be integrated to form a more coherent and responsive epidemiological picture of RSV in Germany. To illustrate this, we extracted exemplary data to provide an overview of RSV seasonality, hospitalization incidence rates, and burden of disease.
Our overarching aim is to equip practicing clinicians and public-health decisionmakers with a structured and transparent overview of the available resources, including direct links to databases to facilitate access. By mapping the current landscape of RSV monitoring systems, we provide a foundation for tracking infection dynamics in real time, analyzing the course of past epidemic seasons, and developing more accurate predictions of future trends. Ultimately, this knowledge will support the efficient allocation of healthcare resources and inform strategies for the treatment, prevention, and long-term management of RSV and other respiratory infections.
## 2. Methods
Databases and surveillance systems covering RSV in Germany were identified by searching for publications reporting on RSV in Germany. For the compilation of publicly accessible databases/surveillance systems, we included publicly searchable databases with access to raw data as well as annually published surveillance reports. Excluded were datasets based on results from nonrecurring studies and those published in a single report or publication. Past surveillance systems that are no longer active at the time of our search (June 2024 and updated in November 2024) were also excluded. Further surveillance systems were identified by expert consultation.
For the identification of regional RSV surveillance systems, we performed searches by search engine and AI (Google, Bing, Copilot) in March 2025 to check every federal state in Germany using the search terms (ARE = akute respiratorische Erkrankung; acute respiratory disease), "ARE Surveillance", "ARE Erfassung" (capture), "Erfassung Atemwegserkrankungen" and "Surveillance Atemwegserkrankungen". When regular surveillance efforts were identified, we checked for the inclusion of RSV in the reported data.
In order to evaluate the scope of each database/surveillance system, we analyzed the databases/surveillance system for the following criteria (see Tables 1 and2):
## •
## Risk of bias/limitations of the database
To obtain an overview of the depth and level of detail with which the disease burden and epidemiology of RSV in Germany can be described based on publicly available databases, the relevant data were extracted in an exemplary manner.
Due to the large heterogeneity of the data from the different sources, a metanalytical approach was not feasible.
Incidence rates are reported as RSV cases per 100,000 persons per year, using the size of the respective population as the denominator.
## 2.1. Data Evaluation
## 2.1.1. Robert Koch Institute (RKI)
To gather information on RSV incidence rates and patient age, the SurvStat database [31] and the annual reports published by the RKI were analyzed. SurvStat collects cases from mandatory reporting, which was implemented for RSV only in Saxony between 2020 and July 2023. For the calculation of outpatient incidence rates, weekly reports across all age groups were used during the 2018/2019 seasons.
In July of 2023, Germany extended mandatory reporting of RSV cases across all German states. Cases are recorded in four case definition categories: (i) clinical and epidemiological criteria met, (ii) clinical and laboratory criteria met, (iii) laboratory criteria met, clinical criteria not met, and (iv) laboratory criteria met, clinical criteria undetermined, and reported in SurvStat [31]. In addition, case numbers of a reference definition consisting of the sum of all four case definition categories is reported in SurvStat and also in the RKI's annual reports. Data can be displayed by calendar year, season, calendar week, or season week and is available in close to real time after the reporting has reached the health authorities.
Data from the hospital-based surveillance program (ICOSARI) is available in the weekly reports (https://influenza.rki.de/Wochenberichte.aspx; accessed on 11 January 2025) and has been reported in detail by Cai et The annual incidence of RSV inpatient cases per 100,000 persons was calculated using the 2011 German standard population for all age categories, as well as for infants under one year of age. Both primary and secondary diagnoses for these groups in the respective years were recorded using the ICD10 codes J12.1 "Respiratory syncytial virus pneumonia", J20.5 "Acute bronchitis due to respiratory syncytial virus", J21.0 "Acute bronchiolitis due to RSV", and B97.4 "Respiratory syncytial virus as the cause of diseases classified elsewhere".
## 2.1.5. The German Institute for the Hospital Remuneration System (InEK)
As the InEK data browser sources from the same hospital ICD-10 codes as GBE-Bund, InEK data was not analyzed separately. A recently published study analyzed data from InEK to report the inpatient burden of RSV in children ≤2 years of age in Germany for 2019-2022 [18].
## 2.1.6. RespVir/Clinical Virology Network
Monthly data of submitted RSV-positive samples were extracted from the year 2015 through May 2022 from the RespVir dashboard [33]. As a centralized reporting system for notifiable diseases, SurvStat provides near real-time monitoring and thus supports outbreak detection. However, the data lacks clinical detail. Additionally, data availability depends on reporting compliance, making it vulnerable to underreporting. (2) Physicians across Germany report current diagnoses of respiratory diseases in the outpatient sector to the online database called ARE Sentinel Surveillance. This data source consists of approximately 700 primary care practices as a representative sample of the population. The data collected include information on the severity and frequency of current RSVIs, specifically acute respiratory diseases, based on clinical symptoms. Additionally, virological surveillance is conducted in approximately 100 sentinel practices that submit patient samples of symptomatic patients to the National Reference Center to identify currently circulating respiratory viruses. The collected data is evaluated on a weekly basis and presented in the form of weekly reports [21].
## 3. Surveillance Systems and Databases on RSV in Germany
Data on RSV consultation rates can be downloaded from an online repository [34] with an additional extraction step.
The ARE Sentinel Surveillance offers broad insights into respiratory illness trends across age groups and regions in the outpatient sector by capturing syndromic as well as virological data. Although representative sampling is aimed for, the system is limited by its sample size (>1% coverage of primary care physicians in Germany) and regional differences in coverage. (3) Nationwide, "citizen scientists" can self-report cases of acute respiratory infections (ARIs) at the population level to the online database known as "GrippeWeb". Individuals aged 16 years and above, residing primarily in Germany, can participate through the web portal in a population-based approach. They can self-report once a week whether they experienced a new respiratory illness in the preceding week. This process tracks the percentage of the entire population that has developed an acute respiratory infection on a weekly basis, including visits to their general practitioners (GPs) [35].
A strength of GrippeWeb is the estimation of ARI incidence even independent of healthcare utilization by collecting self-reported data. However, the data may be biased or incomplete due to self-reporting, and it lacks clinical and virological specificity. The latter limitation is addressed by "GrippeWeb-Plus". (4) Since 2020, the RKI has been conducting an additional virological surveillance program as part of GrippeWeb (influenza web), called "GrippeWeb-Plus." In this program, a randomly selected sample of regularly reporting GrippeWeb participants receive swab materials. In the event of an acute respiratory infection, they take a sample from their frontal nasal area and subsequently send it to the RKI for testing for 24 different respiratory pathogens, including influenza viruses, SARS-CoV-2, and RSV. Currently, around 800 children and adults from approximately 480 different households participate in GrippeWeb-Plus. Since multiple individuals from one household participate in GrippeWeb-Plus, a household-adjusted positive rate is calculated [36]. (5) The syndromic surveillance of severe acute respiratory infections (SARIs) in the inpatient environment involves the use of ICD-10 codes to monitor cases in sentinel hospitals (referred to as ICOSARI, [23]). This surveillance is conducted in approximately 70 selected hospitals, covering about 5-6% of all hospitalized patients in Germany, and is considered representative. As ICOSARI is dependent on ICD-10 codes, it may be affected by coding inaccuracies. (6) As of week 7, 2025, data on RSV detection in wastewater has been reported [37].
For the first report in February 2025, data from 25 wastewater treatment plants was analyzed and reported as viral load, stratified by RSV subgroup A and B [38]. Data provided by further wastewater treatment plants will be included subsequently. A key challenge lies in establishing standardized sampling frequencies and laboratory methodologies to ensure data comparability and reliability across surveillance sites.
## Frequency of updates
Current case numbers provided on weekly basis.
Current case numbers and positivity rates provided on weekly basis.
The billing data in the InEK data browser is provided three times a year, on June 15th, October 15th, and January 15th. These data include discharges from January 1st to May 31st, January 1st to September 30th, and January 1st to December 31st of the current calendar year.
Weekly reports on current season.
## Limitations/Risk of bias Voluntary participation of hospitals
Voluntary participation of laboratories. In most settings, only patients with respiratory symptoms are tested.
Only ICD-10 coded diagnoses documented.
Usually preschool children as sentinels for virus activity.
Abbreviations: ARE = acute respiratory disease, RSV = respiratory syncytial virus, DGPI = German Society for Pediatric Infectiology, ICU = intensive care unit, ICD-10 = international statistical classification of diseases V.10, PCR = polymerase chain reaction.
## 3.2. Surveillance of ARE ("Akute Respiratorische Erkrankungen", Acute Respiratory Illnesses) Surveillance by Federal States in Germany
The State of Saxony Ministry of Social Affairs was the first state to establish an ordinance that mandates the regional reporting of laboratory-confirmed RSV cases (both hospitalized and non-hospitalized) under the State Infection Protection Act [42] in 2002. The reporting is carried out by the testing laboratories.
The reported data is transmitted to the RKI by the Landesuntersuchungsanstalt (LUA; Saxonian Health Institute). Based on the infection epidemiological data, weekly and monthly epidemiological reports are prepared. Additional analyses and trend assessments are included in an annual report.
Several additional states have implemented surveillance systems to monitor Acute Respiratory Diseases via voluntary reporting. Some states have recently established ARE surveillance systems during the COVID-19 pandemic; these systems are now also covering RSV (see Table 3).
In summary, most states activate their ARE surveillance during the typical cold season when the risk of contracting cold-related illnesses is highest; approx. from the 40th to the 15th calendar week of the following year. In most cases, the surveillance comprises two components: a reporting system for ARE incidence in preschool day-care facilities and participation of (pediatric) sentinel practices. Regional ARE reporting complements national systems and supports localized outbreak detection and preparedness. Notably, substantial regional disparities exist, as no data on ARE reporting could be identified for six of Germany's sixteen federal states (see Table 3). Federal Health Monitoring (Gesundheitsberichterstattung des Bundes, GBE-Bund).
The federal health reporting online database (GBE-Bund) is a service run by the German Statistical Office (DESTATIS) and serves as a central hub for consolidating healthrelated data and information from a wide array of more than 100 sources [24]. All hospitals that bill according to the DRG reimbursement system and fall under the scope of § 1 of the Hospital Remuneration Act (KHEntgG) are required to submit data. Main and secondary diagnoses originating from all reporting hospitals in Germany are recorded in an annual survey in the GBE-Bund database and compiled with administrative data, which is predominantly derived from coded discharge diagnoses based on the 10th International Statistical Classification of Diseases and Related Health Problems (ICD-10). Apart from information on hospital primary diagnoses, the database provides data categorized by the place of treatment, the patient's age, gender, residence, duration of stay, and data on cause of death [43]. Additionally, for all codes, secondary diagnoses can be obtained through the GBE-Bund database on request. Based on administrative data from German hospitals, the GBE-Bund database constitutes one of the most comprehensive sources for estimating the burden of RSV and associated healthcare utilization across large population segments. However, the availability of data is subject to a considerable time delay.
## 3.3. German Institute for the Hospital Remuneration System (InEK)
InEK is the organization responsible for developing and maintaining the Diagnosis-Related Groups (DRG) system in Germany. The DRG system is used for hospital reimbursement, where patients are grouped based on their diagnoses and treatments. As required by law, case-related data are to be reported to InEK by every hospital in Germany. Data can be extracted via an online data access tool using primary and secondary diagnoses (ICD-10-codes), age groups, diagnosis-related groups, discharge reason, length of stay, ICU admissions, ventilation hours, and dates of admission and discharge as search categories [25]. Similar to GBE-Bund, InEK provides a comprehensive dataset covering the majority of German hospitals. However, the InEK online system offers enhanced filtering capabilities and greater data granularity. As with GBE-Bund, a key limitation is the delay in data availability.
## 3.4. German Society for Pediatric Infectious Diseases (DGPI)
As Germany experienced a surge of RSV cases in the post COVID-19 pandemic season in 2021/22, the DGPI initiated an ad hoc nationwide surveillance program with 156 participating German pediatric hospitals that monitored hospitalizations due to RSV among children [26]. For the RSV seasons 2022/23, 2023/24, and 2024/25, a registration system for respiratory tract infections that allows pediatric hospitals across Germany to record clinically significant respiratory infections that resulted in hospital admissions was introduced. The number of participating children's hospitals and pediatric departments increased from on average 37 in 2022/23 to 51 in 2023/24 and to 61 (18.3% of pediatric hospitals in Germany) in 2024/25. The survey tracks the percentage of admissions to the hospital and to intensive care units attributed to RSV, SARS-CoV2, and influenza, providing fine-grained age stratification. The data is visualized online on the DGPI homepage [27][28][29]. The DGPI's 'AWI-Erfassung' system monitors hospitalized pediatric respiratory cases with high data quality and timely updates. The scope of data collection is determined annually based on current needs. Initially limited to RSV cases in 2021/22, it was later expanded to include all respiratory illnesses. While data on both new admissions and intensive care were available for 2022/23 and 2023/24, only admission data is currently provided for the 2024/25 season. Tracking for 2024/25 additionally includes the proportion of children aged 0-2 years with RSV immunization or maternal RSV vaccination during pregnancy.
## 3.5. RespVir, Clinical Virology Network
RespVir was established in 2007 as an initiative originating from a clinical virology working group (now the Clinical Virology Network) is part of "Deutsche Vereinigung zur Bekämpfung der Viruserkrankungen" (German Society for Infectious Disease Control) and affiliated with the German Society for Virology (GfV), and the "Paul-Ehrlich-Gesellschaft für Infektionstherapie" (Society for Infectious Diseases Therapy). Its primary objective is to maintain an online database for the comprehensive documentation of respiratory infections, offering healthcare professionals close to real-time information about currently circulating pathogens.
The Clinical Virology Network (RespVir, CVN) database contains diagnostic data sourced from around 55 laboratories, mainly including university hospitals and private laboratories in Germany, Austria, and Switzerland.
The database analyzes 26 different respiratory pathogens, comprising 18 viruses and eight bacteria. RespVir is designed to incorporate data from samples collected from all patients exhibiting respiratory symptoms and submitted by physicians for diagnostic purposes. Most samples are from hospitalized patients. Subjecting the samples to monoand multiplex PCR testing methods reduces the bias that may result from testing only for selected pathogens [44]. A distinctive feature of the CVN is that it collects not only positive RSV test results but also negative ones. This allows for the calculation of positivity rates, which provide a more accurate reflection of epidemiological trends, as absolute case numbers are heavily influenced by overall testing volume. An increase in the positivity rate can indicate rising infection levels, even when the absolute number of cases declines.
Reports on test results in the form of a dashboard regarding respiratory infections can be accessed through an online database interface [33]. Data can be visualized stratified by age and gender. A key strength is the ability to show results either as absolute counts or as positivity rate, which enables comparisons between periods of low test numbers and periods of high test numbers. Limitations include the non-population-based design, absence of detailed clinical data, and variable site participation.
Table 4 provides hyperlinks to publicly accessible surveillance systems and databases. Abbreviations: RSV = respiratory syncytial virus, RKI = Robert Koch Institute, ARE = acute respiratory disease, SARI = severe acute respiratory infections, DGPI = German Society for Pediatric Infectiology.
## 4. RSV Burden of Disease in Germany
## 4.1. RSV Epidemiology in Germany
Seasonality of RSV As they do in many countries in the Northern hemisphere, RSVIs display a typical seasonal pattern in Germany, with interruptions of the established seasonality observed thus far only during the COVID-19 pandemic (Figure 1). The RKI AG Influenza's weekly reports for the years 2016-2019 reveal RSV seasons that span on average 14-18 weeks and typically peak during the winter months of January and February, usually starting between November and January [45]. This dynamic is very similar to the situation in hospitalized patients as reported through the RespVir program of the Clinical Virology Net (Figure 1A,B). During the RSV season, hospitalized children up to six years of age were more likely, compared to all age groups, to have a positive RSV test result. At the height of the 2018/19 season, 51% of all tests among children aged 0-6 years were positive for RSV, while the positivity rate for all age groups was substantially lower at 29%. Positivity rates of the tested samples aggregated across the months of the year show regular offseason/on-season patterns from 2015 to the beginning of 2020 and thus underscore the atypical course of the 2020/2021 and 2021/2022 seasons (Figure 1B). Most likely due to the increased non-pharmaceutical interventions (NPIs) during the SARS-CoV2 pandemic, RSVI (as well as respiratory infections due to other pathogens) was virtually absent from July 2020 to June 2021 [46]. In 2021, RSVI returned unexpectedly early with unusually high numbers of positives, which peaked in August [26,27].
## 4.2. RSV Incidence
To determine absolute numbers of clinically relevant RSVI cases and estimate their impact on the German health care system, data on the diagnoses of hospitalized patients of the German Federal Statistical Office for the years 2010 to 2021 were analyzed (Figure 2). Between 2010 and 2021, documented RSVI cases fluctuated between 16,169 (2011) and 42,857 (2021) per year, resulting in incidence rates from 20 to 51 cases per 100,000 per year (Figure 2C) across all age groups. Most of these cases occurred in children within the first year of life, with case numbers between 12,575 (2011) and 24,637 (2021), yielding incidence rates between 1898 and 3097 per 100,000 (Figure 2D). Bronchiolitis (ICD10-J21.0), bronchitis (ICD10-J20.5), and pneumonia (ICD10-J.12.1) had a significant share among RSVI. Most of the RSVIs in German hospitals were recorded as the primary diagnosis.
The incidence rate of outpatient visits to doctors' offices for RSVI can be estimated from the sentinel program of the RKI. During the 2018/2019 season, between 800 and 2000 consultations per 100,000 were recorded in association with respiratory infections [47]. Differentiation by age showed that consultations for RSVI were strongly skewed toward young children: in infants up to 1 year of age, the average consultation incidence was 12,400/100,000 per year and thus about twice as high as the consultation incidence for influenza. In children aged 2 to 4 years, this value was still at 7700/100,000 per year. For children aged 5 to 14 years (1100/100,000), adolescents and young adults (15 to 34 years, 800/100,000), adults (35 to 59 years, 600/100,000), and the elderly (≥60 years, 700/100,000), yearly consultation incidences were considerably lower [47]. From 2016 to 2021, the incidence of RSVI notifications in Saxony among all age groups rose from 61/100,000 to ca. 144/100,000 (with a dip in 2020 due to non-pharmaceutical interventions during the COVID-19 pandemic) [32]. During the same period, the RSV hospitalization incidence among all age groups in Saxony, as recorded by the German Statistical Office, increased from 22/100,000 to 52/100,000 per year.
Nationwide, SurvStat recorded during the 2024/25 season (as of 20 May 2025) 68,265 cases, resulting in an incidence rate of approx. 78/100,000 across all age groups and an incidence of approx. 88/100,000 in >60-year-olds. Incidence rates were considerably higher in infants (1014/100,000 in <1-year-olds), small children (710/100,000 in <5-year-olds), and persons of old age (208/100,000 in >79-year-olds).
## 4.3. Severity and Hospitalization Burden of RSVI
The DGPI's ad hoc RSVI surveillance for the 2021/22 season provides data on the average number of daily RSV cases reported from both regular hospital wards and intensive care units. Approximately 10% of hospitalized RSV cases required intensive care. The stratification by age provided for the seasons 2022/23, 2023/24 and 2024/25 again revealed a prominent overrepresentation of small children among patients hospitalized for RSVI, especially during the years following the end of the COVID-19 pandemic: between 58% and 69% of newly admitted patients and between 68% and 78% of patients in intensive care units were under the age of one year. In 2024/25, a similar number of infants in their first year of life and toddlers in their second year were admitted, accounting for 33% and 36% of all new pediatric admissions, respectively (Figure 3). The ICD-10-code-based hospital surveillance study for severe acute respiratory infections (ICOSARI) of 8761 RSVI cases between 2009 and 2018 largely confirmed the high burden of severe RSVI among infants, with 57% of hospitalized RSVI patients < 1 year of age [22]. About 5.6% of admitted patients received intensive care and 38% of these required ventilator support. Twenty-five patients died, with almost half of them over the age of 65 years. The case fatality rate (CFR) of hospitalized RSVI cases varied significantly depending on the age of the patient. Of 122 severe RSV cases among the elderly aged 65 and older, 12 patients passed away, yielding a high CFR of 9.5%. Among the 7996 infants with severe RSV, there were 8 fatal cases, yielding a CFR of 0.1%. The examination of potential risk factors identified several underlying conditions that were more frequently found in severe cases. Notably, respiratory and cardiovascular disorders specific to the perinatal period as well as cardiovascular diseases were highly prevalent in patients receiving intensive care (13% and 14%) and even more so in patients that needed ventilator support (19% and 25%) [22].
## 5. Discussion
Information on RSVI in Germany is collected, analyzed, and presented from various sources and in various formats, with each of the different surveillance systems coming with specific strengths and weaknesses. RSV episodes may be captured in distinct surveillance systems contingent upon clinical severity (Figure 4). Disease progression can result in the same episode being sequentially documented in multiple data repositories. Surveillance efforts by the DGPI and Clinical Virology Net (RespVir) focus on recording the number and percentage of positive tests and on promptly disseminating results. This approach offers valuable, near real-time insights into current epidemiological dynamics across a spectrum of pathogens, supporting healthcare professionals in allocating shortterm resources.
Sample-based surveys with a high degree of representativity, such as the RKI ARE Sentinel Surveillance and ICOSARI programs, can estimate national case numbers.
RSV incidence rates are of greatest interest to determine the clinical, but also the economic burden of disease. The GBE-Bund system and InEK, which compile RSVI hospitalization cases from administrative data, stand out for their automated recording during the processing of DRG-based reimbursement claims. Their strength lies in their coverage, resulting from the comprehensive inclusion of almost all German hospitals, thus minimizing potential bias associated with systems that rely on smaller population samples that may be less representative. However, Cai et al. showed a high specificity of RSV-specific ICD-10 codes (99.8%), but very poor sensitivity (6%) for Germany in a comparison of ICD-10 codes with virological data [23]. Thus, data on RSV incidence rates from public databases relying on RSV-specific ICD-10 codes must be assumed to be an underestimation of the actual burden of disease-for adults to a much greater extent than for children [48,49]. This coding bias is likely due to diagnostic testing for RSV not being regularly conducted, especially in adult and senior patients presenting at hospitals with respiratory symptoms. Reasons are likely a lack of disease awareness and the absence of an efficient and specific antiviral therapy [50]. Improvement in the RSV testing of adults has been observed since the introduction of influenza/RSV combination PCRs, which revealed a substantial number of previously unrecognized RSV infections in adults.
PCR testing is considered the most sensitive test in both pediatric and adult patients. Adults and seniors, however, often exhibit lower viral loads during RSV infection compared to children [51], resulting in reduced test positivity rates and underreporting in older age groups [52]. Notably, no consistent correlation has been observed between viral load and disease severity [53]. Of importance for diagnostic practice, sensitivity can be substantially improved by testing multiple specimen types-such as saliva or sputum-in addition to nasopharyngeal swabs [52,54,55]. A recent modeling study from Spain confirms that the use of RSV-specific ICD-10 codes alone leads to an underestimation of the real burden of disease. Rates estimated through modeling were 1.33 times higher than those based on RSV-specific codes for infants. For adults aged 18-79 years, the estimated rates were 6-8 times higher than those based on RSV-specific codes, and for the elderly ≥80 years, underestimation based on ICD-10 codes alone was as high as 16 times [56]. The authors of a prospective surveillance study conducted between 2021 and 2023 in Thuringia also concluded that RSV-specific ICD-10 codes were not suitable for estimating the burden of RSV-pneumonia, as only a third of RSV-pneumonia cases were documented as such with RSV-specific ICD-10 codes [57].
A recent study analyzed nationwide data on RSV-specific ICD-10-coded hospitalizations from Germany and showed RSV hospitalization incidence rates between 0.1 and 11.08 per 100,000 persons per year between 2010 and 2018 in adults ≥60 years [16]. Polkowska-Kramek et al. model incidence rates that are at a minimum 21 times higher. In their analysis of statutory health insurance data from Germany, incidence rates of RSV-associated respiratory, cardiovascular, and cardiorespiratory hospitalizations in adults were modeled for the years 2015 to 2019 [17]. They report RSV-attributable respiratory hospitalization rates between 236.4 and 362.8 per 100,000 person-years in adults ≥60 years of age. According to this modeling study, incidence rates of RSV-attributable hospitalizations may actually be as high as 911.5 per 100,000 person-years when taking into account that in addition to respiratory disease, RSV can cause exacerbation of cardiological conditions [8,9,11], even though it may not have been diagnosed as such [17]. The above-mentioned prospective surveillance study from Thuringia showed an RSV-related ARI hospitalization incidence among those ≥60 years of 401.6/100,000 [58]. In a recent systematic literature review and meta-analysis of 21 studies, acute RSV-related respiratory infections in adults of 60 years and older in high-income countries were estimated at a rate of 1620 per 100,000 and the hospitalization incidence for RSVI was estimated at 150 per 100,000. The in-hospital case fatality rate was high at 7% [59].
Several studies suggest that a significant degree of under-ascertainment may occur in many epidemiological studies on RSVI [60,61]. Li et al. [60] examined the RSV-associated acute respiratory infection hospitalization burden in older adults in high-income countries. Using a two-step framework that incorporated empirical data on the RSV detection proportion of different clinical specimens and testing approaches as well as their statistical uncertainty, they estimated that the pooled hospitalization rate more than doubled from 157 per 100,000 (95%CI 98-252) for adults aged ≥ 65 years to 347 per 100,000 (95%CI 203-595) after accounting for under-ascertainment. The in-hospital case fatality rate of RSV was estimated at 6.1% (3.3-11.0).
Obviously, to unequivocally diagnose RSVI, laboratory tests must be done in the first place. Rozenbaum [62] showed that in a sample of 937 US hospitals, in adults of 65 years or more hospitalized for LRTI, testing for RSVI was conducted only infrequently, suggesting that a significant number of RSVIs remain undiagnosed. However, one positive side effect of the COVID-19 pandemic is the notable increase in the use of multiplex assays screening for respiratory viruses such as SARS-CoV-2, Influenza A and B, and RSV simultaneously. A recent study comparing German hospitalized RSV cases among adults 60+ between preand post-pandemic seasons (2019/20 vs. 2022/23) suggests an underdetection of RSV of up to sevenfold, which was revealed through the increase in testing in adults [63].
## 6. Conclusions and Outlook
The limited capacities of hospitals in Germany demand a critical ascertainment of the RSV burden of disease to allow the efficient management of medical care. Multiple initiatives contribute in different ways to RSVI surveillance. The recent implementation of a nationwide reporting mandate for RSVI applying standardized and comparable methods of assessment and analysis is anticipated to further enhance the quality of RSVI surveillance. Due to limitations regarding testing strategies and high-sensitivity diagnostic tools, particularly in the outpatient setting, infection rates deduced by the present surveillance systems in Germany may not completely reflect the real infection dynamics and may introduce bias by underestimation. Especially in adults, both early awareness of RSV-associated symptoms of the upper and lower respiratory tract and consistent testing must be strengthened, as they are key to accurately assessing the individual risk and societal burden of RSVI. It remains to be evaluated whether greater standardization of diagnostics and consistent demographic stratification could enhance the accuracy of national surveillance systems. Accurate and up-to-date epidemiological data on RSVI are imperative for effectively allocating healthcare resources and estimating the clinical, societal, and economic burden of the disease.
For clinicians, our findings highlight the importance of understanding the scope and limitations of different RSV monitoring systems. Awareness of the variability in timeliness and accuracy among data sources can improve the interpretation of local infection dynamics and help contextualize clinical decision-making. Reliable access to epidemiological information is particularly valuable for anticipating seasonal surges, optimizing patient management strategies, and informing preventive measures such as prophylaxis or vaccination.
From a research perspective, the review underscores the need for the harmonization and integration of existing monitoring systems. Enhanced surveillance frameworks could enable more robust epidemiological modeling, facilitate the evaluation of preventive and therapeutic interventions, and ultimately contribute to a deeper understanding of RSV's impact within the broader spectrum of respiratory infections.
## References
1. Janssen, Biontech ; Roche, Novartis et al. "MSD, Novartis, Pfizer and SOBI. HS received consultation and speaker fees from Pfizer. MW received consultation and speaker fees from GSK, Exeltis, MSD, Novartis, Pfizer, Sanofi-Pasteur and Viatris/Mylan. CB and BN have been contracted as consultant/employee of AMS Advanced Medical Sciences GmbH by Pfizer to assist in the preparation of this and previous manuscripts.BN is currently an employee of ZEG Berlin GmbH. CvE, GB and LB are Pfizer employees and may own Pfizer stock"
2. Shang, Tan, Ma (2021) "Respiratory syncytial virus: From pathogenesis to potential therapeutic strategies" *Int. J. Biol. Sci*
3. Ciarlitto, Vittucci, Antilici et al. (2019) "Respiratory Syncityal Virus A and B: Three bronchiolitis seasons in a third level hospital in Italy" *Ital. J. Pediatr*
4. Redlberger-Fritz, Springer, Aberle et al. (2023) "Respiratory syncytial virus surge in 2022 caused by lineages already present before the COVID-19 pandemic" *J. Med. Virol*
5. Ng, Cox, Atkins et al. (2001) "Encephalopathy associated with respiratory syncytial virus bronchiolitis" *J. Child Neurol*
6. Robinson, Busl (2020) "Neurologic Manifestations of Severe Respiratory Viral Contagions" *Crit. Care Explor*
7. Koch (2025) "RSV-Infektionen. Available online"
8. Rosas-Salazar, Chirkova, Gebretsadik et al. (2023) "Respiratory syncytial virus infection during infancy and asthma during childhood in the USA (INSPIRE): A population-based, prospective birth cohort study" *Lancet*
9. Eisenhut (2006) "Extrapulmonary manifestations of severe RSV bronchiolitis" *Lancet*
10. Ivey, Edwards, Talbot (2018) "Respiratory Syncytial Virus and Associations With Cardiovascular Disease in Adults" *J. Am. Coll. Cardiol*
11. Kwong, Schwartz, Campitelli (2018) "Acute Myocardial Infarction after Laboratory-Confirmed Influenza Infection" *N. Engl. J. Med*
12. Ramaswamy, Groskreutz, Look (2009) "Recognizing the importance of respiratory syncytial virus in chronic obstructive pulmonary disease" *COPD J. Chronic Obstr. Pulm. Dis*
13. Falsey (2007) "Respiratory syncytial virus infection in adults" *Semin. Respir. Crit. Care Med*
14. Falsey, Hennessey, Formica et al. (2005) "Respiratory syncytial virus infection in elderly and high-risk adults" *N. Engl. J. Med*
15. Falsey, Walsh (2005) "Respiratory syncytial virus infection in elderly adults" *Drugs Aging*
16. Njue, Nuabor, Lyall et al. (2023) "Systematic Literature Review of Risk Factors for Poor Outcomes Among Adults With Respiratory Syncytial Virus Infection in High-Income Countries" *Open Forum Infect. Dis*
17. Niekler, Goettler, Liese et al. (2010) "Hospitalizations due to respiratory syncytial virus (RSV) infections in Germany: A nationwide clinical and direct cost data analysis" *Infection*
18. Polkowska-Kramek, Begier, Bruyndonckx et al. (2015) "Estimated Incidence of Hospitalizations and Deaths Attributable to Respiratory Syncytial Virus Infections Among Adults in Germany Between" *Infect. Dis. Ther*
19. Wick, Poshtiban, Kramer et al. (2019) "Inpatient burden of respiratory syncytial virus in children ≤2 years of age in Germany: A retrospective analysis of nationwide hospitalization data" *Influenza Other Respir. Viruses*
20. Impfkomission (2024) "Beschluss und wissenschaftliche Begründung zur Empfehlung der STIKO zur spezifischen Prophylaxe von RSV-Erkrankungen mit Nirsevimab bei Neugeborenen und Säuglingen in ihrer 1. RSV-Saison" *Epidemiol. Bull*
21. Impfkomission (2024) "Beschluss zur Empfehlung der STIKO für eine Standardimpfung gegen Erkrankungen durch Respiratorische Synzytial-Viren (RSV) für Personen ≥ 75 Jahre sowie zur Indikationsimpfung von Personen im Alter von 60 bis 74 Jahren mit Risikofaktoren" *Epidemiol. Bull*
22. Goerlitz, Cai, Tolksdorf et al. "ICD-10-Code-basierte syndromische Surveillance akuter Atemwegserkrankungen mit COVID-19 im ambulanten Bereich" *Epidemiol. Bull. 2021*
23. Cai, Buda, Schuler et al. (2020) "Risk factors for hospitalized respiratory syncytial virus disease and its severe outcomes. Influenza Other Respir" *Viruses*
24. Cai, Tolksdorf, Hirve et al. (2020) "Evaluation of using ICD-10 code data for respiratory syncytial virus surveillance. Influenza Other Respir" *Viruses*
25. Bundesamt (2025) "Das Informationssystem der Gesundheitsberichterstattung des Bundes. Available online"
26. (2024) "Institut für das Entgeltsystem im Krankenhaus (InEK). InEK DatenBrowser"
27. Gesellschaft Für Pädiatrische Infektiologie (2025) "Ergebnisse der Adhoc-Erfassung der RSV-Assoziierten Krankheitslast bei Stationär Behandelten Kindern und Jugendlichen in Deutschland im Winter 2021-2022"
28. (2022) "Deutsche Gesellschaft für Pädiatrische Infektiologie. Ad hoc Atemwegsinfektions-Erfassung der Krankheitslast bei Stationär Behandelten Kindern und Jugendlichen in Deutschland-Ergebnisse der Ad hoc Atemwegsinfektions-Erfassung"
29. (2023) "Deutsche Gesellschaft für Pädiatrische Infektiologie. Ad hoc Atemwegsinfektions-Erfassung der Krankheitslast bei Stationär Behandelten Kindern und Jugendlichen in Deutschland"
30. (2025) "Deutsche Gesellschaft für Pädiatrische Infektiologie. Ad hoc-Erfassung Aller stationären Aufnahmen Aufgrund Akuter Respiratorischer Infektionen (ARI) bei Kindern und Jugendlichen in Deutschland"
31. (2025) "Clinical Virology Network. Clinical Virology Network. Available online"
32. Koch (2025) "SurvStat@RKI 2.0-Web-Basierte Abfrage der Meldedaten Gemäß Infektionsschutzgesetz (IfSG). Available online"
33. (2025) "b6250b64beea12bf"
34. Goerlitz, Tolksdorf, Prahm et al. (2024)
35. Koch, Grippeweb (2025)
36. Buchholz, Buda, Eberle et al. (2024) *GrippeWeb-Wochenbericht KW*
37. Koch (2025) "AMELAG: Abwasser Monitoring für die epi Demio Logische Lage Bewertung-Überwachung von SARS-CoV-2, Influenzaviren, RSV und Anderen Erregern im Abwasser"
38. Koch (2025) *ARE-Wochenbericht KW*
39. Koch (2024) "Diagramme Virologischer Surveillance-Deutschland (Gesamt) Saison"
40. (2023) "Ad hoc-Erfassung der Aufgrund von Atemwegsinfektionen Stationär Aufgenommenen und Intensivmedizinisch Behandelten Kinder und Jugendlichen in Deutschland; Klinik und Poliklinik für Kinder-und Jugendmedizin"
41. (2025) "Deutsche Gesellschaft für Pädiatrische Infektiologie. FAQs: Häufig Gestellter Fragen zu den DGPI-Surveys; 2025. Available online"
42. "Sächsisches Staatsministerium für Soziales und Verbraucherschutz. Verordnung des Sächsischen Staatsministeriums für Soziales und Verbraucherschutz über die Erweiterung der Meldepflicht für übertragbare Krankheiten und Krankheitserreger nach dem Infektionsschutzgesetz"
43. Bundesamt, Respiratory-Syncytial-Viren (2025) "?p_uid=gast&p_aid=52372343&p_knoten=VR&p_sprache=D&p_suchstring=respiratory%20syncytial"
44. Horemheb-Rubio, Eggeling, Schmeibetaer et al. (1167) "Respiratory viruses dynamics and interactions: Ten years of surveillance in central Europe" *BMC Public Health*
45. Cai, Durrwald, Biere et al. (2011) "Determination of respiratory syncytial virus epidemic seasons by using 95% confidence interval of positivity rates" *Viruses*
46. Shaw, Abad, Amin-Chowdhury et al. "Trends in invasive bacterial diseases during the first 2 years of the COVID-19 pandemic: Analyses of prospective surveillance data from 30 countries and territories in the IRIS Consortium"
47. Koch (2018) *Bericht zur Epidemiologie der Influenza in Deutschland*
48. "Jahresbericht 2019 der Landesuntersuchungsanstalt für das Gesundheits-und Veterinärwesen" *LUA*
49. (2020) "Landesuntersuchungsanstalt für das Gesundheits-und Veterinärwesen Sachsen: Dresden"
50. (2023) "Sächsische Impfkommission (SIKO). 7. Positionspapier der SIKO-1. November 2023-RSV-Impfung; Sächsische Impfkommission; Sächsische Landesärztekammer: Dresden"
51. Drews, Branche, Falsey et al. (2019) "What is the role of rapid molecular testing for seniors and other at-risk adults with respiratory syncytial virus infections?" *J. Clin. Virol*
52. Li, Bont, Weinberger et al. (2024) "Relating in vivo RSV infection kinetics to host infectiousness in different age groups"
53. Onwuchekwa, Moreo, Menon et al. (2023) "Underascertainment of Respiratory Syncytial Virus Infection in Adults Due to Diagnostic Testing Limitations: A Systematic Literature Review and Meta-analysis" *J. Infect. Dis*
54. Garcia-Maurino, Moore-Clingenpeel, Thomas et al. (2019) "Viral Load Dynamics and Clinical Disease Severity in Infants With Respiratory Syncytial Virus Infection" *J. Infect. Dis*
55. Onwuchekwa, Atwell, Moreo et al. (2023) "Pediatric Respiratory Syncytial Virus Diagnostic Testing Performance: A Systematic Review and Meta-analysis" *J. Infect. Dis*
56. Ramirez, Carrico, Wilde et al. (2023) "Diagnosis of Respiratory Syncytial Virus in Adults Substantially Increases When Adding Sputum, Saliva, and Serology Testing to Nasopharyngeal Swab RT-PCR" *Infect. Dis. Ther*
57. Haeberer, Bruyndonckx, Polkowska-Kramek et al. (2016) "Estimated Respiratory Syncytial Virus-Related Hospitalizations and Deaths Among Children and Adults in Spain" *Infect. Dis. Ther*
58. Pletz, Ankert, Schwarz et al. (2024) "Primary International Classification of Diseases Codes Are Only Modestly Sensitive for Detection of Community-Acquired Pneumonia Hospitalization"
59. Liang, Begier, Hagel et al. (2025) "Incidence of RSV-related hospitalizations for ARIs, including CAP: Data from the German prospective ThEpiCAP study" *J. Infect*
60. Savic, Penders, Shi et al. (2023) "Respiratory syncytial virus disease burden in adults aged 60 years and older in high-income countries: A systematic literature review and meta-analysis. Influenza Other Respir" *Viruses*
61. Li, Kulkarni, Begier et al. (2023) "Adjusting for Case Under-Ascertainment in Estimating RSV Hospitalisation Burden of Older Adults in High-Income Countries: A Systematic Review and Modelling Study" *Infect. Dis. Ther*
62. Rozenbaum, Begier, Kurosky et al. (2023) "Incidence of Respiratory Syncytial Virus Infection in Older Adults: Limitations of Current Data" *Infect. Dis. Ther*
63. Rozenbaum, Judy, Tran et al. (2023) "Low Levels of RSV Testing Among Adults Hospitalized for Lower Respiratory Tract Infection in the United States" *Infect. Dis. Ther*
64. Scholz, Dobrindt, Tufts et al. (2024) "The Burden of Respiratory Syncytial Virus (RSV) in Germany: A Comprehensive Data Analysis Suggests Underdetection of Hospitalisations and Deaths in Adults 60 Years and Older" *Infect. Dis. Ther*
65. "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" |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.