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=PMC12724260&blobtype=pdf | # Cmss1 limits FMDV infection by enhancing antigen presentation and CD8 + T cell responses
Yang Wang, Lihong Zhang, Jieru Deng, Linlin Zheng, Zhihua Chen, Zhao Zhang, Han Zhang, Jingjing Pei, Haixue Zheng
## Abstract
The foot and mouth disease virus (FMDV) poses a heavy burden on the global swine industry, underscoring the urgent need for long-term protective vaccines to control FMDV infection. Dendritic cells (DCs) are critical in activating innate immune cells and initiating adaptive immune responses. However, the role of DCs in FMDV infection remains poorly understood. In this study, we developed a mouse model lacking the type I interferon receptor for infection, followed by single-cell RNA sequencing to generate a high-resolution map of DCs in the spleens of FMDV-infected mice. Next, we established an evaluation system to investigate the antigen-presenting capacities of gene-edited DCs and mice. The results reveal that FMDV infection alters the propor tions of DC subsets and significantly suppresses antigen processing and presentation. Additionally, we identified Cmss1 as a novel host factor that antagonizes the inhibition of the antigen-presenting process caused by FMDV infection, thereby limiting FMDV pathogenicity in mice. These findings provide valuable insights into potential antiviral strategies against FMDV infection.
IMPORTANCEIn this study, we first developed C57BL/6 mice lacking the type I interferon receptor as an infectious model, which was subjected to single-cell RNA sequencing analysis of DC features in response to FMDV infection. We identified VP41 9-30 as an immunodominant CD8 + T cell epitope of FMDV and established an in vitro system to evaluate the antigen-presenting capacities. Based on these findings, we validate Cmss1 as a novel host factor in antigen processing and presentation during FMDV infection. Since DCs play critical roles in mediating immune responses, our findings comprehensively characterize the immune features of dendritic cells for the first time and present a new mechanism through which the host defends against FMDV infection, suggesting Cmss1 as a novel potential target for antiviral therapies.
F oot and mouth disease (FMD) is a widespread disease that affects cloven-hoofed ruminants such as swine, sheep, and cattle. Since livestock is a predominant source of economic income in many countries and regions, the disease significantly burdens the global economy and food security. FMD virus (FMDV), also known as the aphthous fever virus, is a non-enveloped RNA virus belonging to the Picornaviridae family. Seven serotypes of FMDV (A, O, C, SAT1, SAT2, SAT3, and Asia1) exhibit diverse antigenicity. This diversity complicates cross-protection among serotypes, meaning each serotype requires a specific vaccine to provide adequate immunity (1)(2)(3)(4). Among all seven serotypes, type O is the most common and poses a significant challenge worldwide. It is responsible for approximately 70% of FMD outbreaks worldwide and is, therefore, considered the most serious FMD serotype (5)(6)(7). Upon FMDV infection, the host's humoral immune system rapidly produces specific antibodies targeting the virus. This antibody-mediated response is essential for neutralizing the virus and can provide protective immunity for months (8). However, recent studies have reported that antibody responses alone cannot entirely prevent FMDV infection (9). Although antibodies are essential for disease defense, they respond relatively late and often do not provide timely protection against infections, while cellular immune responses can elicit an immediate and effective reaction to viral infections. Therefore, clarifying the interaction between FMDV and host cellular immune responses will guide and assist in developing antiviral strategies based on cellular immunity (9,10). Furthermore, inactivated vaccines that induce antibody-mediated immune protection are widely used in clinical settings; however, latent infections and reinfections among immune animals are quite common, underscoring the limitations of currently available vaccines. Our limited understanding of the T-cell immune mechanisms triggered by FMDV infection remains a fundamental obstacle to the defects in current vaccine development (11)(12)(13).
Dendritic cells (DCs) play a crucial role in presenting pathogen antigens and activating the adaptive immune response. Typically, after FMDV infection, DCs internal ize the virus and process its antigens. These processed antigens are then presented on the surface of DCs, bound to major histocompatibility complex (MHC) molecules, which subsequently trigger the adaptive immune response to eliminate the infection by activating T cells (14). Studies have indicated that DCs comprise two major subsets: classical DCs (cDCs) and plasmacytoid DCs (pDCs), both of which are essential in sensing viral pathogens and inducing long-term antiviral immunity (15)(16)(17)(18), while cDCs are known for their pivotal function in stimulating naïve T cells and orchestrating acquired immune responses (19,20). According to Guilliams et al., cDCs are categorized into two major lineages: cDC1 and cDC2 (21). Specialized DC subsets exhibit diverse ontogeny, phenotypes, and functions, and many viruses disrupt the functions of DCs to suppress the immune response (22,23). However, the relationship and regulatory mechanisms between FMDV infection and host antigen presentation remain unclear. Therefore, a comprehensive understanding of the biological characteristics of DCs and their antigen presentation function during FMDV infection is essential for investigating the mecha nisms of FMDV immunosuppression and immune evasion and will significantly assist in vaccine improvement.
FMDV infection strongly suppresses the type I IFN response (24)(25)(26)(27) and causes disease in cloven-hoofed animals, but not in wild-type (WT) mice. Previous studies have used nude mice as models for FMDV infection (28)(29)(30), some of which have restored adaptive immunity by the adoptive transfer of immunocompetent cells (31). However, these models are not ideal for studying DCs. In our study, we established a model of FMDV infection using Ifnar knockout (Ifnar -/-) mice on the C57BL/6 background, which effectively replicates the clinical symptoms and pathological changes associated with FMD and serves as an effective animal model to investigate the effects of FMDV infection on antigen presentation function. Furthermore, by utilizing single-cell RNA sequencing (scRNA-seq), we conducted a high-resolution analysis of splenocytes from Ifnar -/-mice infected with FMDV at 5 days post-infection (dpi). This approach provided a comprehensive transcriptome landscape and enabled the deconvolution of the genomic profiles of DC subsets. Additionally, we mapped the T cell responses during FMDV infection in mouse spleens and established a system for assessing antigen presentation capacity. Concurrently, we identified Cms1 small ribosomal subunit homolog (Cmss1) as a previously unrecognized host factor that mediates DC responses and antigen processing and presentation during FMDV infection. These findings provide new insights into the pathogenesis of FMDV and aid in developing antiviral strategies.
## RESULTS
## A novel mouse model susceptible to FMDV infection
To establish a mouse model susceptible to FMDV infection, Ifnar -/-and WT C57BL/6 mice were subcutaneously injected with 10 4 plaque-forming units (PFUs) of FMDV, and the viral RNA loads in the heart, peripheral blood, liver, lymph node, spleen, lung, and kidney were evaluated subsequently. At 3, 5, and 7 dpi, the viral RNA was detectable in all Ifnar -/-tissues, while viral RNA loads remained extremely low and nearly undetectable in WT mice. It should be noted that viral loads in the tissues of Ifnar -/-mice were 10 to 10 5 times higher than those in WT mice, particularly in the spleen (Fig. 1A). Next, we investigated whether Ifnar -/-mice showed clinical symptoms after FMDV infection using a clinical criteria scale (Table 1) and weight monitoring. As expected, Ifnar -/-mice showed significant weight loss and related clinical symptoms (Fig. 1B andC). Notably, 100% of the Ifnar -/-mice died from FMDV infection (Fig. 1D). After that, we further determined the pathological effects of FMDV infection. To this end, heart and spleen tissues from Ifnar -/-and WT mice infected with FMDV were collected and subjected to clinicopathological analysis. Compared to WT mice, Ifnar -/-mouse hearts exhibited significantly more severe parenchymatous myocarditis, and their spleens showed more serious acute splenitis, effectively simulating the typical pathological characteristics and clinical symptoms of FMDV infection (Fig. 1E). Finally, flow cytometry was performed to analyze the immune cell composition in the spleen using the gating strategy outlined in Fig. S1A. No significant differences were observed in the proportions of T lymphocytes, B lymphocytes, DCs, and macrophages, indicating that the susceptibility of Ifnar -/-mice, compared with WT mice, was not due to the depletion of immune cells ( Fig. S1B). This further suggests that Ifnar -/-mice do not impact the adaptive immune response triggered by FMDV infection. Collectively, these results indicate that Ifnar -/-mice are a more suitable animal model for in vivo infection and immunological studies than WT mice.
## The global cell distribution in the mouse spleen infected with FMDV
The spleen, one of the peripheral lymphoid organs, plays a critical role in trapping antigen-bearing DCs and initiating adaptive immune responses. Pathogen antigens are transported to the spleen from infection sites, primarily by DCs. Considering the high levels of viral loads (Fig. 1A) and the lesions observed in spleens (Fig. 1E), this organ was selected for further investigation. We chose 5 dpi as the sampling timepoint based on our viral replication assay, which showed that viral loads peak at this stage, indicating active infection. To comprehensively characterize the molecular profile of DCs in the spleens of FMDV-infected mice, scRNA-seq was conducted on freshly collected spleens, capturing a total of 25,379 and 30,440 high-quality cells from FMDV-infected and mock-infected mice, respectively, at 5 dpi (Fig. 2A).
According to the expression of canonical markers, eight significant populations were identified via graph-based clustering of uniform manifold approximation and projection (UMAP) (Fig. 2B andC). These populations included T cells, natural killer cells (NKs), innate lymphoid cells (ILCs), B cells, DCs, macrophages (Macs), neutrophils, and fibroblasts (Fig. 2B through D). Among these cells, the DC populations exhibited high levels of Flt3, CD209a, Itgax, and Tap2 (Fig. 2E), suggesting the accuracy of the identification. Notably, there are no significant changes in the proportion of any cell types, including DCs (Fig. 2B andD). These findings were further validated by flow cytometric analysis (Fig. S2A andB). To further explore the potential shifts in the global proportion of immune cells during secondary FMDV infection, four experimental groups were established with initial infectious doses of 0, 0, 50, and 50 PFUs per mouse (n = 5) on day 0, followed by secondary infectious doses of 0, 5,000, 0, and 5,000 PFUs per mouse on day 30, respectively. All spleens were harvested at 35 dpi and subjected to flow cytometric analysis following the gating strategies (Fig. S1A). Consistent with previous results, the distribution of all cell types remained relatively stable at the global level (Fig. S2C). Based on the above results, we comprehensively defined the cellular distribution globally in the spleen of FMDV-infected mice.
## APC subset distribution in the mouse spleen after FMDV infection
Since different subsets of DCs play varied roles in pathogen recognition and immune activation, each DC subset may serve as the coordinator for a specific function mod ule. Therefore, we further characterized the distribution of DC subsets in the mouse spleen, and nine subsets were recovered and manually annotated based on signature markers (Fig. 3A; Fig. S3A). They were Clec10a -cDC2, CD24 low cDC1, Atf3 + DCs, CD24 high cDC1, Clec10a + cDC2, Ifitm + Ltb + DCs, pDCs, CD63 + DCs, and Mki67 + DCs. Notably, we observed decreases in DC1, Ifitm + Ltb + DCs, and CD63 + DCs and increases in DC2 and pDCs in infected samples (Fig. 3B). According to previous research, cDC1s primarily prime CD8 + T cells, while cDC2s are crucial for CD4 + T cell activation (32). The reduction in cDC1s indicates a potential inhibition of CD8 + T cell responses during FMDV infection. Moreover, RNA velocity analysis (33) showed three main directions from the center (CD24 high cDC1) to CD24 low cDC1, CD63 + DCs, and Atf3 + DCs (Fig. S3B through D Fig. 3C). Furthermore, pseudotime analysis was used to dissect the cell states of DC subsets during FMDV infection. We found that DCs were divided into two branches: CD24 low cDC1 was significantly enriched at the end of the upper branch, while CD24 high cDC1 and DC2s were distributed around the origin. Meanwhile, other DC subsets exhibited varied distributions: pDCs were present at the origin, Atf3 + DCs clustered at the end of the lower branch, and Ifitm + Ltb + DCs, along with CD63 + DCs, exhibited aggregation at the upper tail (Fig. S3E). These six DC subsets were categorized into seven different states (Fig. 3D), with terminal states 6 and 7 showing different function modules (Fig. 3E) and dynamic compositional changes (Fig. S4A). These two states of cells had similar compositions at the starting point (mainly composed of Mki67 + DCs, pDCs, CD24 high cDC1, and cDC2) and subsequently changed along the pseudotime (cells of state 6 were primarily composed of Atf3 + DCs, while state 7 contained four DC subsets, namely, Mki67 + DCs, CD63 + DCs, Ifitm + Ltb + DCs, and CD24 low cDC1, at the end) (Fig. S4A). Based on our findings, the numbers of CD63 + DCs, Ifitm + Ltb + DCs, and CD24 low cDC1 remarkably decreased during FMDV infection (Fig. 3B), indicating the reduction of cells of state 7. According to GO analysis, the genes in module 2, which were significantly upregulated in state 7, strongly correlated with ribosome-related functions (Fig. S4B). These findings indicate that ribosome-associated genes could play a crucial role in the interactions between FMDV and the host. Altogether, these data revealed the spectrum of DC lineage, their development, and the molecular profiles of individual DC subsets.
## FMDV infection suppresses MHC expression and reprograms ribosome-rela ted genes in DCs
DCs play a central role in antigen presentation. Our study identified differentially expressed genes (DEGs) in DCs during FMDV infection, revealing 342 downregulated and 234 upregulated genes (Fig. 4A). Among the DEGs, we identified significant downregula tion of MHC class I (H2-D1 and H2-K1) and MHC class II (H2-Eb2, H2-Aa, and H2-Ab1) in DCs (Fig. S5A), which is consistent with the known ability of FMDV to suppress MHC class I in swine cells (34). Moreover, Gene Ontology (GO) and Gene Set Enrichment Analysis (GSEA) analyses indicated that the antigen processing and presentation process was downregulated by FMDV infection (Fig. 4B andC). These findings suggest that FMDV actively inhibits antigen processing and presentation in DCs. Notably, the downregula ted genes were markedly enriched in ribosome-related biological processes, such as ribosomal large and small subunit assembly and ribosomal biogenesis (Fig. 4B; Fig. S5B). Interestingly, Cmss1, which plays a crucial role in ribosomal biology, was significantly upregulated in response to FMDV infection (Fig. 4A; Fig. S5C). Based on the scRNA-seq results, we further explored the potential regulation of Cmss1 in porcine lung epithelial cells infected with FMDV using Western blot analysis and observed similar results (Fig. 4D). Ribosomes typically act as protein factories and are essential in immune respon ses. Our findings revealed that FMDV infection results in the disruption of ribosomal function and the inhibition of antigen processing and presentation pathways. However, intriguingly, the expression level of the ribosomal functional protein CMSS1 significantly increased following virus infection. Next, we aim to further investigate whether CMSS1 plays a regulatory role in host antigen presentation, which is crucial for our comprehen sive understanding of FMDV immune regulation.
## Screening of immunodominant T cell epitopes derived from FMDV structural proteins
To better understand the impact of CMSS1 on antigen presentation, we first needed to establish methods that could accurately reflect the function of antigen presentation and T-cell responses to FMDV. First, to characterize the immunodominant T cell response induced by FMDV, we identified the epitopes from FMDV structural proteins (VP1, VP2, VP3, and VP4) that are recognized by T cells. Next, a total of 114 peptides covering all structural proteins of FMDV were chemically synthesized, and an IFNγ-(Enzyme-Linked ImmunoSpot) ELISpot assay was performed to individually screen the peptides using splenocytes from FMDV-infected mice. Notably, 26 peptides tested positive for FMDV, with seven of these peptides showing significant differences compared to DMSO controls (Fig. 5A). To further characterize the specificity of the MHC-restricted T cell response, we evaluated these seven identified peptides using the IFNγ-ELISPOT assay, focusing on CD8 + or CD4 T cells (Fig. 5E andG). Overall, all seven peptides identified through the IFNγ-ELISPOT assay elicited high frequencies of either CD8 + IFNγ + or CD4 + IFNγ + T cells. Finally, we used a bioinformatics program to examine the four FMDV structural proteins (VP1, VP2, VP3, and VP4) (35). Notably, the top-ranked core sequence, which is predicted to possess high binding affinities to MHC molecules, was included in VP1 61-78 and VP1 67-84 for H-2Kb, VP2 133-150 , and VP2 139-156 for H-2IAb, and VP4 13-30 and VP4 19-36 for H-2Db. This result further supports our findings described above (Table S1). In conclusion, we identified and screened seven peptides linked to four immunodominant T-cell epitopes found in the structural proteins of FMDV using the IFNγ-ELISPOT and ICS assays.
## An evaluation system for antigen-presenting capacities
Based on our results from scRNA-seq analysis, the distribution of cDC1 subpopulations was predominantly reduced after FMDV infection. cDC1 cells, being professional antigenpresenting cells, activate CD8 responses. In conjunction with screening for T-cell-specific epitope peptides, we discovered that VP4 13-30 and VP4 19-36 can significantly activate T cells, particularly CD8 + T cells. Therefore, we opted to synthesize the common region of the two to validate their antigen presentation effects. To this end, we synthesized the overlapped 15 amino acids between these two peptides and subsequently tested the new peptide VP4 19-30 via IFNγ-ELISPOT and ICS. Not surprisingly, VP4 19-30 showed intense activity in activating IFNγ-expressing T cells (Fig. 6A through C andE), with 465 ± 10 SFC/10 6 splenocytes (Fig. 6A andB) and approximately 20% of CD8 + T cells expressing IFNγ (Fig. 6C andE). Meanwhile, we also examined the properties of VP4 19-30 in WT C57 mice, which barely elicited T cell responses (Fig. 6A, B, D andE). The results indicate that VP413-30 is the most effective epitope peptide for activating CD8 + T cells. It will also be used in the antigen presentation ability evaluation system during FMDV infection to further assess the regulation of antigen presentation function of Cmss1, as screened in our study.
## Cmss1 antagonizes FMDV-induced MHC class I inhibition and enhances antigen presentation
Since the positive marker VP4 19-30 described above is associated with the MHC class Ipeptide and the inhibition of MHC class I observed during FMDV infection in scRNA-seq (Fig. S5A), we primarily focused on investigating the potential regulation of Cmss1 on MHC class I during FMDV infection. We first tested the MHC class I expression level of porcine lung epithelial cells during FMDV infection by flow cytometry. Notably, MHC class I was found to be downregulated by FMDV in a time-dependent manner (Fig. 7A), which is consistent with previous studies (34). In our scRNA-seq data, Cmss1 exhibited remarkable upregulation during FMDV infection. However, its role in the antigen Based on these results, we further investigated whether Cmss1 could improve the processes of antigen processing and presentation using the evaluation system men tioned above. CD8 + T cells from FMDV-infected Ifnar -/-mice were mixed with Cmss1 knockout and wild-type DC2.4, respectively, stimulated with the positive marker VP4 19-30 , and subsequently subjected to the IFNγ-ELISPOT assay. The Cmss1 knockout DC2.4 group demonstrated significantly fewer positive spots compared to the wild-type DC2.4 group (Fig. 7C andD). The results support the observation that Cmss1 upregulates MHC class I and indicate its ability to enhance antigen processing and presentation during FMDV infection. To further validate this conclusion, Cmss1 -/-Ifnar -/-doubleknockout mice were constructed using CRISPR/Cas9 technology (Fig. S6C), and the antigen-presenting capacities of VP4 19-30 were assessed using the evaluation system described above. Consistent with our expectations, compared to Cmss1 +/+ Ifnar -/-mice, Cmss1 -/-Ifnar -/-mice demonstrated reduced capabilities to produce IFNγ during FMDV infection (Fig. 7E andF). Therefore, these results indicate that FMDV inhibits MHC class Imediated CD8 + T cell responses, a process that Cmss1 antagonizes. This suggests that Cmss1 may function as a host molecule in antiviral immunity during FMDV infection, limiting viral infection by enhancing antigen presentation.
## Cmss1 deficiency enhances FMDV pathogenicity in mice.
As Cmss1 has been shown to enhance antigen processing and presentation by mediating MHC class I, we hypothesized that deleting Cmss1 would increase FMDV infection. To test this hypothesis, we evaluated the resistance of Cmss1 +/+ Ifnar -/-and Cmss1 -/-Ifnar -/-mice to FMDV infection by injecting 10 3 PFUs of FMDV subcutaneously into each mouse. The results indicated that Cmss1 -/-Ifnar -/-mice were significantly less resistant to FMDV infection than Cmss1 +/+ Ifnar -/-mice. Infection with 10 3 PFUs of FMDV caused death at 5 dpi in Cmss1 -/-Ifnar -/-mice, whereas it did not result in death in Cmss1 +/+ Ifnar -/-mice (Fig. 8A). Additionally, we analyzed the viral RNA from the organs of Cmss1 +/+ Ifnar -/-and Cmss1 -/-Ifnar -/-mice, discovering that knocking out Cmss1 led to an increase in viral RNA levels by approximately 20, 50, 7.5, 600, and 190 times in the liver, spleen, lung, kidney, and lymph node, respectively (Fig. 8B). These results further demonstrate that Cmss1 antagonizes FMDV infection and reduces viral pathogenicity by enhancing host antigen presentation function.
## DISCUSSION
FMD is a severe, highly contagious viral disease affecting livestock, which has a considerable economic impact. However, the mechanisms by which FMDV prolongs infection in vivo, and the responses of host cells to infection remain largely unclear. Since DCs coordinate both the innate and adaptive immune systems, gaining a deeper understanding of DC-related biological processes during FMD enhances our knowledge of the underlying interactions between the host and FMDV. In this study, scRNA-seq was conducted to comprehensively characterize the cell distributions and gene expression profiles of DCs derived from the immune cells in the spleens of FMDV-infected sub jects. In search of potentially novel host factors involved in antigen processing and presentation, we established an evaluation system by identifying immunodominant T cell epitopes derived from FMDV structural proteins. Finally, we conclude that FMDV infection alters the proportions of DC subsets and impairs their antigen processing and presentation functions. Importantly, we identified cmss1 as a novel host factor; its depletion in DC2.4 cells and mice resulted in a significant decrease in MHC class I, which inhibited the CD8 + T cell response and ultimately reduced resistance to FMDV infection. The interactions between viruses and DCs remain a vibrant area of research. Many viruses engage with the biological processes of DCs, thereby enhancing their replica tion. The Ebola virus VP35 and the gamma 1 34.5 protein of herpes simplex virus 1 impair dendritic cell maturation (36,37). The Epstein-Barr virus inhibits DC development by promoting the apoptosis of their monocyte precursors in the presence of granulo cyte-macrophage colony-stimulating factor and interleukin-4 (38). The measles virus suppresses cell-mediated immunity by interfering with the survival and functions of dendritic cells and T cells (39). In our scRNA-seq data, FMDV suppresses the MHC class I and II molecules of DCs during infection. These phenotypes have previously been observed in murine DCs (40) and PK-15 and ESK-4 cell lines (34). Our study also reveals the reprogramming of DC subsets during FMDV infection, which aligns with that of varicella zoster virus during its natural infection (41). Type 1 immune responses target intracellular pathogens, such as viruses, requiring IFN-γ-activated cytotoxic CD8 + T cells for clearance. Numerous studies have shown that cDC1s are crucial for type 1 responses due to their ability to activate CD8 + T cells (42)(43)(44)(45)(46). Based on our data, the decrease in the proportion of cDC1 indicates that FMDV may enhance its productive infection by impairing cDC1s, which are crucial for Type 1 immune responses.
Our data from murine spleens also indicate significant downregulation of ribosomerelated functions. Interestingly, a recent study has demonstrated high-efficiency induced ribosomal frameshifting by the West Nile Virus (47). Viruses regulate gene expression by manipulating programmed ribosomal frameshifting (PRF), which alters the mRNA reading frame of ribosomes during translation. Furthermore, a recent study on murine norovirus also suggests a downregulation of ribosome biogenesis in infected cells at single-cell resolution (48). These studies suggest that manipulating ribosomes may be crucial for virus and host infection. However, despite the overall reduction in ribosomerelated processes, Cmss1 displays a unique trend during FMDV infection. Cmss1 is an RNA-binding protein that acts as a component of the small ribosomal subunit, contributing to translation (49). Its expression has been observed in various tissues, including those involved in immune responses. Gene enrichment analyses indicate that Cmss1 primarily participates in ribonucleoprotein complex biogenesis, rRNA metabo lism, ncRNA processing, translation initiation, RNA localization, and RNA catabolism (50). While Cmss1 primarily functions in ribosomal activity and protein synthesis, its expression patterns and associations with immune-related tissues suggest it may play a potential role in immune responses. For instance, a recent study indicates that Cmss1 is a novel host factor in maintaining HIV-1 latency (51). However, the precise role of Cmss1 in influencing immunity has yet to be fully clarified. Considering the clinically latent FIG 8 Cmss1 -/-mice are less resistant to FMDV infection. Cmss1 -/-Ifnar -/-and Cmss1 +/+ Ifnar -/-mice at 4 weeks of age were infected with 10 3 PFUs of FMDV. Indicated organs were harvested at day 7 post-infection, and the levels of FMDV RNA were determined. (A) Survival curves of infected Cmss1 -/-Ifnar -/-(red triangles) and Cmss1 +/+ Ifnar -/-mice (black circles) (n = 5). (B) FMDV RNA loads in the heart, peripheral blood, liver, lymph node, spleen, lung, and kidney of Cmss1 -/-Ifnar -/-(red triangles) and Cmss1 +/+ Ifnar -/-mice (black circles) were quantified using RT-qPCR (n = 3). Statistical significance is represented by asterisks (****P < 0.0001).
infection of FMDV, Cmss1 may differ from other ribosome-related genes and play a vital role in antigen processing and presentation.
To explore the potential function of Cmss1 in antigen processing and presentation, we provide a validated map of the T cell response to FMDV structural proteins, identi fying four epitopes. Among these, the most effective epitope is the CD8 + T cell epit ope, indicating the critical role of the CD8 + T cell response in FMDV infection. These findings are supported by those of other research. For instance, studies have demon strated that enhancing antigen-specific CD8 + T cell responses can contribute to early protection against FMDV (52). Additionally, specific viral peptides displayed by MHC class I molecules stimulate the activation and proliferation of CD8 + T cells, resulting in the targeted destruction of infected cells (53). Since VP4 13-30 and VP4 19-36 activated IFNγ-secreting T cells most effectively, we established a system to evaluate the antigenpresenting capabilities of DCs by using overlapping 15 amino acids (VP4 19-30 ) as a positive marker for activating CD8 + T cells. Using this tool, we validate the auxo-action of Cmss1 in the MHC class I-dependent antigen processing and presentation processes. CD8 + T cell responses play a crucial role in antiviral immunity, acting rapidly before the humoral immune response is completely established. These cells secrete cytokines like IFN-γ, which enhance antiviral immune responses. They also exert cytotoxic effects by releasing perforin and granzymes, inducing apoptosis, and limiting viral replication. The CD8 + T cell response often precedes the antibody immune response, enabling a timely reaction to viral infections and helping resist them before neutralizing antibod ies are produced. Additionally, the CD8 + T cell-mediated immune response is crucial for the latency and persistence of infections as the body requires a continuous cellu lar immune response for adequate protection. Moreover, CD8 + T cells contribute to long-term immunity by differentiating into memory T cells, which facilitate a quicker and more vigorous response upon subsequent exposure to the same virus, thereby lessening disease severity. This memory function plays a crucial role in vaccine-induced protection. Consequently, CD8 + T cell responses are essential for both immediate viral clearance and long-term immune protection, forming a fundamental part of the host's defense against viral pathogens. Currently, FMDV-inactivated vaccines exhibit good efficacy but do not induce early immune responses in vaccinated animals. In our study, Cmss1 has been shown to regulate MHC-I-dependent CD8 + T cell antigen presentation during FMDV infection. This indicates that Cmss1 is crucial in guiding early immune activation and promoting long-term immunity in FMDV vaccine development. To further explore the underlying mechanism of the auxo-action of Cmss1 in MHC class I-dependent antigen processing and presentation, we constructed a Cmss1 -/-DC2.4 cell line and Cmss1 -/- mice. Using these, we proved that Cmss1 significantly upregulates MHC class I in DCs. Importantly, Cmss1 -/-mice are less resistant to FMDV infection than their Cmss1 +/+ littermates, exhibiting notable clinical symptoms and elevated virus titers in multiple organs. These findings present a valuable new gene not previously identified as involved in FMDV infection and the antigen presentation process.
In summary, our findings confirm the reprogramming of DC subsets and suppression of antigen processing and presentation in DCs during FMDV infection, as revealed by scRNA-seq. This method is a powerful means of assessing the antigen-presenting functions of cells concerning FMDV infection and highlights Cmss1 as a promising target for future defenses against FMD. It is important to note that identifying func tionally relevant molecular targets involved in antigen processing and presentation is a significant challenge in the field. Clearly, our understanding of the molecular mecha nisms involved in the interplay between FMDV and its host is still inadequate. There fore, further investigation of various infection samples from different periods, including immune tissue samples from both acute and chronic stages as well as from recovering animals, will comprehensively assess the immune characteristics of FMDV infection and identify valuable host molecules that may be crucial for developing a more robust strategy for the prevention and control of FMDV infection.
## MATERIALS AND METHODS
## Mice, virus, and cells
WT C57BL/6 mice were bred at the Lanzhou Veterinary Research Institute, Chinese Academy of Agricultural Sciences (LVRI, CAAS). Ifnar KO mice (Ifnar -/-) and Cmss1 KO mice (Cmss1 -/-, Ifnar -/-) were obtained from Cyagen Biosciences Inc. Animals were bred from heterozygous parents and validated using PCR and Western blot. The FMDV/O/GD/CHA/ 2010 /S/BF8 strain virus was stored at LVRI, CAAS. DC2.4, HEK293T cells, and porcine lung epithelial cells were stored in the laboratory.
## Generation of knockout DC2.4 by CRISPR/Cas9 technology
HEK293T cells were initially transfected with lentivirus plasmid and packaging assistant plasmids psPAX2 and pMD2.G for 48 h. After that, the viral supernatant was collected, and the virus was concentrated through ultracentrifugation. Subsequently, the wild-type DC2.4 cells (2 × 10 6 cells/well in 6-well plates) were transfected using lentivirus specific to the indicated target genes. After 24 hours of infection, the positive cells were screened by adding 5 µg/mL of puromycin before the cloning selection. All selected clones were amplified and sequenced for further use. Genomic DNA was extracted from the edited cells using the MiniBEST Universal Genomic DNA Extraction Kit (TaKaRa, 9765, Dalian, China) following the manufacturer's protocol, and 50 ng of genomic DNA from each clone was utilized to amplify the region surrounding the sgRNA cleavage site. The PCR was performed using a Bio-Rad C1000 thermal cycler. The final products were visualized on a 1% agarose gel and then sequenced. This research used the published mouse genome sequence and gene annotation information (Mus musculus) to design sgRNAs based on the reference standards (54).
## Single-cell RNA-seq data analysis
After undergoing the QC criteria, single splenocytes were subjected to downstream analysis. The Normalize Data function in Seurat (56) and the global-scaling normalization method "Log Normalize" were applied to adjust the library size and the gene expression measurements for each cell by the total expression, multiplied by a scaling factor-10,000 (the results were log-transformed). Moreover, top variable genes across single cells were identified using the previously described method (57), and the FindVariableGenes function (mean.function = Fast Exp Mean, dispersion.function = Fast Log VMR) in Seurat (56) was used for selection. To reduce dimensionality using the RunPCA func tion in Seurat (56) and cluster cells according to their gene expression profiles, princi pal component analysis (PCA) and graph-based clustering (by using the FindClusters function in Seurat ( 56)) were performed. Then, the 2-dimensional Uniform Manifold Approximation and Projection (UMAP) algorithm with the Run UMAP function in Seurat (56) was used to visualize cells, and marker genes of each cluster were identified via the FindAllMarkers function (test. use = presto) in Seurat (56). FindAllMarkers identified positive markers for a given cluster in comparison to all other cells. We then utilized the R package Single R (58) (version 1.4.1), a novel computational method for unbiased recognition of cell types in scRNA-seq, along with the reference transcriptomic data set "scmca" (59) to infer the originating cell of each single cell independently and identify their cell types. Differentially expressed genes (DEGs) were identified using the FindMarkers function (test. use = presto) in Seurat (56). A P value of < 0.05 and |log2foldchange| > 0.58 were established as the thresholds for significant differential expression. Gene Ontology (GO) enrichment and KEGG pathway enrichment analyses of DEGs were performed using R according to the hypergeometric distribution.
## Flow cytometry (FACS)
For flow cytometry analysis, cells were counted using the Countess 3 FL (Invitrogen, USA) and washed twice with 500 µL of flow cytometry staining buffer (Biosharp, BL1136A). Murine cells were labeled with anti-CD16/CD32 (BioLegend, 101302) and treated with the specified antibodies, while swine cells were treated with the specified antibodies immediately. The Zombie Aqua Fixable Viability Kit (BioLegend, 423102) was subsequently used according to the manufacturer's protocol. Finally, the stained cells were washed twice, resuspended in 200 µL of flow cytometry staining buffer, and collected for analysis using the CytoFlex LX (Beckman, Germany) and FlowJo software (BD Biosciences). The antibodies used for flow cytometry analysis in this study are listed as follows. FITC-conjugated SLA class I (Bio-Rad, MCA2261), PE-conjugated MHC class I (Invitrogen, 12-5958-82), PE594-conjugated CD45 (BioLegend, 103146), AF700conjugated CD3 (BioLegend, 100216), Pacific Blue-conjugated CD4 (BioLegend, 100531), PE-conjugated CD8a (BioLegend, 100708), PEcy7-conjugated CD19 (BioLegend, 115519), PerCP-conjugated I-A/I-E (BioLegend, 107626), AF488-conjugated CD11c (BioLegend, 117311), AF647-conjugated F4/80 (BioLegend, 123122), and BV605-conjugated CD11b (BioLegend, 101257).
## Immunohistochemistry (IHC)
The heart and spleen tissues were collected at the specified time and fixed in 4% paraformaldehyde. Following this, all tissues were dehydrated, paraffin-embedded, sectioned at 4 µm, and stained with hematoxylin and eosin (H&E).
## RNA extraction and reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
Total RNA was extracted from heart, peripheral blood, liver, lymph nodes, spleen, lung, and kidney tissues using TRIzol reagent (TaKaRa, 9766, Dalian, China). It was then reverse-transcribed into complementary DNA (cDNA) with a PrimeScript RT reagent kit (TaKaRa, RR036A, Dalian, China) according to the manufacturer's instructions. The cDNA underwent quantitative analysis using the Light Cycler 480 II System (Roche, Basel, Switzerland), utilizing Probe qPCR Mix with UNG (TaKaRa, RR391A, Dalian, China). The primers used in this study are listed in Table 2.
## Peptide synthesis
All peptides were synthesized by Nanjing TG peptide Biotechnology Co., Ltd on the 10 mg scale, and mass spectral analysis was performed for each peptide to validate the synthesis. All peptides were purified by reverse-phase HPLC to ≥ 95% purity and dissolved in DMSO.
## Western blot analysis
The cell precipitate was lysed in RIPA Lysis Buffer (Beyotime, P0013C) supplemented with 1% protease inhibitor cocktail (NCM Biotech, P001). A BCA Protein Assay kit (Thermo Fisher Scientific, A55864) was used to determine the total protein concentration. The samples were analyzed by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% non-fat milk and then incuba ted with primary antibodies, followed by incubation with HRP-conjugated secondary antibodies. Immunoreactive proteins were detected using an imaging analysis system (Amersham Imager 600; GE, USA). The antibodies used for Western blot analysis in this study are listed as follows. Cmss1 primary antibody (Abmart, PU758791S) and tubulin primary antibody (Proteintech, 66240-1-Ig).
## ICS
For ICS, splenocytes were counted using Countess 3 FL (Invitrogen, USA) and washed with 500 µL of flow cytometry staining buffer (Biosharp, BL1136A) twice. The splenocytes were then resuspended in 10% FBS/RPMI 1640 medium. After that, the splenocytes were seeded in 96-well plates at a density of 10 6 cells per well and cultured in 10% FBS/RPMI 1640 medium with the indicated individual peptide (10 µg/mL) for 1 hour. Positive (PMA-ionomycin) and negative (DMSO) controls were included in all experiments. Splenocytes were treated with 1 × brefeldin A (Biolegend, 420601, USA) for 5 hours and then labeled with anti-CD16/CD32 (BioLegend, 101302), PE594-conjugated CD45 (BioLegend, 103146), AF700-conjugated CD3 (BioLegend, 100216), Pacific Blue-conjuga ted CD4 (BioLegend, 100531), PE-conjugated CD8a (BioLegend, 100708), and Zombie AquaTM Fixable Viability Kit (BioLegend, 423102). Cells were subsequently fixed and permeabilized using the Fixation and Permeabilization kit (Invitrogen, GAS0003, USA) following the manufacturer's protocol, and PE-Cy7-conjugated anti-IFN gamma (MBL, 60-7311-U100) was employed to label IFNγ + cells. Cells were finally collected and analyzed using CytoFlex LX (Beckman, Germany) and FlowJo software (BD Biosciences).
## References
1. Brooksby (1982) "Portraits of viruses: foot-and-mouth disease virus" *Intervirology*
2. Cartwright, Chapman, Sharpe (1982) "Stimulation by heterotypic antigens of foot-and-mouth disease virus antibodies in vaccinated cattle" *Res Vet Sci*
3. Mattion, König, Seki et al. (2004) "Reintroduc tion of foot-and-mouth disease in Argentina: characterisation of the isolates and development of tools for the control and eradication of the disease" *Vaccine (Auckl)*
4. Paton, Valarcher, Bergmann et al. (2005) "Selection of foot and mouth disease vaccine strains-a review" *Rev Sci Tech*
5. Mahapatra, Upadhyaya, Paton et al. (2017) "Genetic and antigenic characterization of serotype O FMD viruses from East Africa for the selection of suitable vaccine strain" *Vaccine (Auckl)*
6. Aslam, Alkheraije (2023) "The prevalence of foot-and-mouth disease in" *Asia. Front Vet Sci*
7. Mahapatra, Yuvaraj, Madhanmohan et al. (2015) "Antigenic and genetic comparison of foot-and-mouth disease virus serotype O Indian vaccine strain, O/IND/R2/75 against currently circulating viruses" *Vaccine (Auckl)*
8. Windsor, Carr, Bankowski et al. (2011) "Cattle remain immunocompetent during the acute phase of foot-and-mouth disease virus infection" *Vet Res*
9. Lee, Shin, Kim et al. (2019) "Mincle and STING-stimulating adjuvants elicit robust cellular immunity and drive long-lasting memory responses in a foot-and-mouth disease vaccine" *Front Immunol*
10. Lee, Park, Ko et al. (2020) "Advanced foot-and-mouth disease vaccine platform for stimulation of simultane ous cellular and humoral immune responses" *Vaccines (Basel)*
11. Cox, Voyce, Parida et al. (2006) "Effect of emergency FMD vaccine antigen payload on protection, sub-clinical infection and persistence following direct contact challenge of cattle" *Vaccine (Auckl)*
12. Pacheco, Smoliga, Donnell et al. (2015) "Persistent foot-and-mouth disease virus infection in the nasopharynx of cattle; tissue-specific distribution and local cytokine expression" *PLoS One*
13. Stenfeldt, Diaz-San, Segundo et al. (2016) "The pathogenesis of foot-and-mouth disease in pigs" *Front Vet Sci*
14. Harwood, Gerber, Sobrino et al. (2008) "Dendritic cell internalization of foot-and-mouth disease virus: influence of heparan sulfate binding on virus uptake and induction of the immune response" *J Virol*
15. Perussia, Fanning, Trinchieri (1985) "A leukocyte subset bearing HLA-DR antigens is responsible for in vitro alpha interferon production in response to viruses" *Nat Immun Cell Growth Regul*
16. Heath, Carbone (2009) "Dendritic cell subsets in primary and secondary T cell responses at body surfaces" *Nat Immunol*
17. Cella, Jarrossay, Facchetti et al. (1999) "Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon" *Nat Med*
18. Siegal, Kadowaki, Shodell et al. (1999) "The nature of the principal type 1 interferonproducing cells in human blood" *Science*
19. Steinman, Witmer (1978) "Lymphoid dendritic cells are potent stimulators of the primary mixed leukocyte reaction in mice" *Proc Natl Acad Sci*
20. Nussenzweig, Steinman, Gutchinov et al. (1980) "Dendritic cells are accessory cells for the development of anti-trinitrophenyl cytotoxic T lymphocytes" *J Exp Med*
21. Guilliams, Ginhoux, Jakubzick et al. (2014) "Dendritic cells, monocytes and macrophages: a unified nomenclature based on ontogeny" *Nat Rev Immunol*
22. Saichi, Ladjemi, Korniotis et al. (2021) "Single-cell RNA sequencing of blood antigenpresenting cells in severe COVID-19 reveals multi-process defects in antiviral immunity" *Nat Cell Biol*
23. Sun, Hua, Chen et al. (2017) "Transcrip tional Changes during Naturally Acquired Zika Virus Infection Render Dendritic Cells Highly Conducive to Viral Replication" *Cell Rep*
24. Li, Wang, Liu et al. (2013) "Engagement of soluble resistance-related calcium binding protein (sorcin) with foot-and-mouth disease virus (FMDV) VP1 inhibits type I interferon response in cells" *Vet Microbiol*
25. Medina, Knudsen, Greninger et al. (2017) "Interaction between FMDV L pro and transcription factor ADNP is required for optimal viral replication" *Virology (Auckl)*
26. Li, Zhang, Yang et al. (2019) "Poly (rC) binding protein 2 interacts with VP0 and increases the replication of the foot-and-mouth disease virus" *Cell Death Dis*
27. Li, Yang, Yang et al. (2016) "The VP3 structural protein of foot-and-mouth disease virus inhibits the IFN-β signaling pathway" *FASEB J*
28. Fernández, Borca, Sadir et al. (1986) "Foot-and-mouth disease virus (FMDV) experimental infection: susceptibility and immune response of adult mice" *Vet Microbiol*
29. Borca, Fernández, Sadir et al. (1986) "Immune response to foot-and-mouth disease virus in a murine experimental model: effective thymus-independent primary and secondary reaction" *Immunology*
30. López, Sadir, Borca et al. (1990) "Immune response to foot-and-mouth disease virus in an experimental murine model. II. Basis of persistent antibody reaction" *Vet Immunol Immunopathol*
31. Piatti, Berinstein, Lopez et al. (1991) "Comparison of the immune response elicited by infectious and inactivated foot-and-mouth disease virus in mice" *J Gen Virol*
32. Dudziak, Kamphorst, Heidkamp et al. (2007) "Differential antigen processing by dendritic cell subsets in vivo" *Science*
33. Manno, Soldatov, Zeisel et al. (2025) *Full-Length Text Journal of Virology*
34. Van Bruggen, Guo, He et al. (2018) "RNA velocity of single cells" *Nature*
35. Sanz-Parra, Sobrino, Ley (1998) "Infection with foot-and-mouth disease virus results in a rapid reduction of MHC class I surface expression" *J Gen Virol*
36. Kim, Ponomarenko, Zhu et al. (2012) "Immune epitope database analysis resource" *Nucleic Acids Res*
37. Jin, Yan, Prabhakar et al. (2010) "The VP35 protein of Ebola virus impairs dendritic cell maturation induced by virus and lipopolysaccharide" *J Gen Virol*
38. Jin, Ma, Prabhakar et al. (2009) "The γ 1 34.5 protein of herpes simplex virus 1 is required to interfere with dendritic cell maturation during productive infection" *J Virol*
39. Li, Liu, Hutt-Fletcher et al. (2002) "Epstein-Barr virus inhibits the development of dendritic cells by promoting apoptosis of their monocyte precursors in the presence of granulocyte macrophage-colony-stimulating factor and interleukin-4" *Blood*
40. Fugier-Vivier, Servet-Delprat, Rivailler et al. "Rabourdin-Combe C. 1997. Measles virus suppresses cell-mediated immunity by interfering with the survival and functions of dendritic and T cells" *J Exp Med*
41. Ostrowski, Vermeulen, Zabal et al. (2005) "Impairment of Thymus-dependent responses by murine dendritic cells infected with foot-and-mouth disease virus" *J Immunol*
42. Huch, Cunningham, Arvin et al. (2010) "Impact of varicellazoster virus on dendritic cell subsets in human skin during natural infection" *J Virol*
43. Scharton-Kersten, Contursi, Masumi et al. (1997) "Interferon consensus sequence binding protein-deficient mice display impaired resistance to intracellular infection due to a primary defect in interleukin 12 p40 induction" *J Exp Med*
44. Liu, Fan, Dias et al. (2006) "Cutting edge: dendritic cells are essential for in vivo IL-12 production and development of resistance against Toxoplasma gondii infection in mice" *J Immunol*
45. Mashayekhi, Sandau, Dunay et al. (2011) "CD8α(+) dendritic cells are the critical source of interleukin-12 that controls acute infection by Toxoplasma gondii tachyzoites" *Immunity*
46. Alexandre, Ghilas, Sanchez et al. (2016) "XCR1+ dendritic cells promote memory CD8+ T cell recall upon secondary infections with Listeria monocytogenes or certain viruses" *J Exp Med*
47. Askenase, Han, Byrd et al. (2015) "Bone-marrow-resident NK cells prime monocytes for regulatory function during Infection" *Immunity*
48. Aleksashin, Langeberg, Shelke et al. (2024) "RNA elements required for the high efficiency of West Nile virus-induced ribosomal frameshifting"
49. Matsushima, Levenson, Chaimongkol et al. (2024) "Single-cell transcriptional analysis of murine norovirus infection in a human intestinal cell line" *J Virol*
50. Manouchehri, Salinas, Hussain et al. (2023) "Distinctive transcriptomic and epigenomic signatures of bone marrow-derived myeloid cells and microglia in CNS autoimmunity" *Proc Natl Acad Sci*
51. Chen, Wang, Liu et al. (2024) "Prognostic value and gene regulatory network of CMSS1 in hepatocellular carcinoma" *Cancer Biomark*
52. Röling, Sisakht, Ne et al. (2021) "A two-color haploid genetic screen identifies novel host factors involved in HIV-1 latency" *MBio*
53. Mu, Chen, Dong et al. (2024) "Enhanced antigen-specific CD8 T cells contribute to early protection against FMDV through swine DC vaccination" *J Virol*
54. Guzman, Taylor, Charleston et al. (2010) "Induction of a crossreactive CD8(+) T cell response following foot-and-mouth disease virus vaccination" *J Virol*
55. Sanjana, Shalem, Zhang (2014) "Improved vectors and genomewide libraries for CRISPR screening" *Nat Methods*
56. Savas, Virassamy, Ye et al. (2018) "Single-cell profiling of breast cancer T cells reveals a tissue-resident memory subset associated with improved prognosis" *Nat Med*
57. Butler, Hoffman, Smibert et al. (2018) "Integrating single-cell transcriptomic data across different conditions, technologies, and species" *Nat Biotechnol*
58. Macosko, Basu, Satija et al. (2015) "Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets" *Cell*
59. Aran, Looney, Liu et al. (2019) "Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage" *Nat Immunol*
60. Han, Wang, Zhou et al. (2018) "Mapping the mouse cell atlas by microwell-seq" *Cell*
61. Trapnell, Cacchiarelli, Grimsby et al. (2014) "The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells" *Nat Biotechnol*
62. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12624820&blobtype=pdf | # Co-Infection With Hepatitis E Virus and SARS-CoV-2 in a Lymphoma Patient
Yacine Boucetta, | Boschi, Erika Peroni, Rodolphe Jean, Erwan Bories, Patrick Borentain, Julien Andreani, Philippe Colson
## Abstract
Hepatitis E virus (HEV) and SARS-CoV-2 are both ubiquitous worldwide. We report here a rare case of HEV and SARS-CoV-2 coinfection. HEV and SARS-CoV-2 diagnoses were carried out by qPCR. Virus genotyping was performed by analyzing partial or near full-length genomes obtained by next-generation sequencing from plasma or nasopharyngeal specimens, respectively. The case was a lymphoma patient admitted to hospital for oxygen-requiring qPCR-documented Covid-19. SARS-CoV-2 was a Delta variant. Reevaluation of a liver cytolysis documented 1 year earlier revealed concurrent chronic hepatitis E at cirrhosis stage. HEV genotype was 3c. Ribavirin was administered, but to date only temporarily led to HEV RNA undetectability in plasma when combined with immunosuppressive therapy interruption. This case and the three previous cases of coinfection with HEV and SARS-CoV-2 point out the possible deleterious effect of SARS-CoV-2 on the liver and the increased risk of such concurrent infections in immunocompromized patients.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
Hepatitis E virus (HEV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are both single-stranded (+) RNA viruses. HEV was discovered in 1983 in Russia and SARS-CoV-2 in 2019 in China [1,2]. Both are ubiquitous worldwide. HEV is a major cause of acute hepatitis, and its epidemiology varies according to the geographical area and the genotype as it is mainly waterborne in developing countries and most often documented as associated with consumption of pig-derived products in developed countries [2]. It can cause chronic hepatitis and cirrhosis in immunocompromised people. It is endemic in the south of France [3]. SARS-CoV-2 became pandemic during early 2020 and essentially causes a respiratory syndrome, COVID-19, which can be clinically severe [4,5]. Due to the tremendous SARS-CoV-2 incidence at the global and country scales over the past 5 years, many concurrent infections occurred in SARS-CoV-2-positive patients. These included infections with other respiratory viruses [6] but also with a broad range of other viruses and microorganisms. Here we report a case of co-infection with HEV and SARS-CoV-2 in an immunocompromised patient with severe pneumonia. A man in his fifties with marginal zone lymphoma initially treated with Ibrutinib and Human immunoglobulins was hospitalized during fall 2021 for oxygen-dependent pneumonia due to SARS-CoV-2. First symptoms appeared 5 days before admission and included cough, asthenia, right ear pain and fever. The patient was diagnosed as infected by SARS-CoV-2 using a realtime PCR (qPCR) assay (BGI, Shenzhen, China), qPCR cycle threshold value (Ct) being 20. A Delta variant was identified using the Nextclade web application (https://clades.nextstrain.org/) [7] based on the viral genome (GenBank (https://www.ncbi.nlm.nih. gov/genbank/) Accession no. ON276954.1) obtained by nextgeneration sequencing (NGS) with the Illumina technology (Illumina Inc., San Diego, USA) on a NovaSeq 6000 Instrument as previously described [8]. The patient had received two doses of the Pfizer-BioNTech COVID-19 mRNA vaccine 3 and 5 months before this episode, as well as immunoglobulins (Polyvalent human immunoglobulins, 30 mg) due to hypogammaglobulinemia at 1.63 g/L. Because of hypoxia and initial pulmonary damage estimated at 50%-60% by computed tomography scan, he received oxygen therapy at 8 L/min. Probabilistic antibiotic therapy with ceftriaxone was implemented for 5 days and switched to piperacillin/tazobactam (4 g, three times a day) for 5 days combined with dexamethasone (6 mg/day), anakinra (300 mg), and an injection of the combination of casirivimab-imdevimab, two monoclonal antibodies. Microbial cultures of blood and urine returned negative. SARS-CoV-2 serology (Liaison XL DiaSorin, Saluggia, Italy) was weakly positive (67 binding antibody units (BAU)/mL) 5 days post-admission. SARS-CoV-2 qPCR turned negative on a nasopharyngeal specimen swab and weaning from oxygen therapy occurred 5 days later.
In parallel with his respiratory infection, the patient was presenting a perturbation of liver biochemical tests with a cytolysis and anicteric cholestasis the day of his admission (alanine aminotransferase (ALT), 98 IU/L; aspartate aminotransferase (AST), 138 IU/L; gammaglutamyltransferase (GGT), 376 IU/L; total bilirubinemia 23 µmol/L). As a matter of fact, liver disturbances were already present in 2020, and liver biopsy performed during summer 2021 showed fibrosis at METAVIR stage F3. Two months before admission, ALT, AST and GGT were 444, 269, and 293 IU/L, respectively, and shear wave elastography showed liver fibrosis at METAVIR stage F4 (13 kPa), indicating liver cirrhosis. Autoimmune hepatitis had been considered but auto-antibodies were negative, while liver toxicity of ibrutinib was also suspected. At the time SARS-CoV-2 qPCR turned negative, HEV qPCR was performed and found positive, viral load being 7.5 log 10 IU/mL in plasma and 8.3 log 10 IU/mL in feces (Altona diagnostics GmbH, Hamburg, Germany; detection threshold, 2.3 log 10 IU/mL). Anti-HEV IgG (Wantai, Beijing China) and IgM (Liaison XL, DiaSorin) testing were negative. HEV genotype was 3c, as determined by a BLAST search and a phylogenetic analysis based on the partial viral genome (GenBank no. PQ878621.1) obtained directly from the DNA/ RNA extract recovered from the plasma by NGS using the Illumina technology, as previously described [9] (Figure 1). No source or transmission route of HEV infection was documented.
One week after the HEV diagnosis, plasma HEV RNA load was still 7.7 log 10 IU/mL, but the patient had improved clinically and returned home. Due to the patient's immunocompromised status and the at least 3 month-duration of the liver cytolysis, treatment with ribavirin (800 mg per day) was introduced for 3 months. Plasma HEV RNA was still detectable (5.4 log 10 IU/mL) after this 3-month course of therapy but ribavirin was nonetheless discontinued for 3 months due to drug shortage. Resumption of ribavirin (800 mg per day) occurred at month 7, combined with the cessation of ibrutinib immunosuppressive therapy at month 9. This led to HEV RNA undetectability at month 10 of first ribavirin introduction. Ribavirin was discontinued, but a virological rebound occurred 4 months later. Ribavirin was reintroduced at month 18 of first ribavirin introduction, at dosages of 800 mg per day, then 1200 mg per day and 400 mg per day for 6, 6 and 10 months, respectively. From month 21 through month 32 of first ribavirin introduction, HEV RNA was tested on five occurrences in plasma and was detectable in all cases, with viral loads ranging between 1.9 and 2.4 log 10 IU/mL. Thereafter, the patient was lost to follow-up.
## 3 | Discussion
To our best knowledge, the present case is the fourth report of HEV and SARS-CoV-2 co-infection [11,12]. The two first cases (both men in their 60 s) were reported to result from nosocomial transmission, documented by sequence analysis, of a HEV of genotype 3 f in an intensive care unit in England in 2021 [11] (Table 1). This was possibly due to a failure in infection control practices in the context of the Covid-19 pandemics with sharing of a two-bed space. The deemed source patient, with a medical history of ischemic heart disease, hypertension and alphathalassemia, experienced chronic HEV infection requiring ribavirin therapy in the setting of immunomodulation by FIGURE 1 | Phylogenetic tree that incorporated the partial HEV genome of the index case. The HEV partial genome sequence obtained by nextgeneration sequencing directly from the blood sample through metagenomics from the case is indicated by a white bold font and a dark blue background. It spans positions 86-7,017 from genome GenBank Accession no. FJ705359. The 100 sequences with the highest BLAST (https://blas-t.ncbi.nlm.nih.gov/Blast.cgi) scores recovered from the NBCI GenBank nucleotide sequence databases (http://www.ncbi.nlm.nih.gov/nucleotide/), were incorporated in the phylogeny reconstruction, in addition to reference sequences for HEV genotypes, indicated by a light blue bold font [10]. Nucleotide alignments were performed using the Mafft web application (https://mafft.cbrc.jp/alignment/server/index.html). Avian orthohepevirus was used as outgroup sequence. Phylogeny reconstruction was carried out using the MEGA v11 program (https://www.megasoftware.net/). The evolutionary history was inferred using the Neighbor-Joining method. The evolutionary distances were computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates) are shown below the branches. Bootstrap values > 50% are labeled on the tree. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree; the scale bars indicate the number of nucleotide substitutions per site. Ribavirin (800 mg per day) for 3 months followed by a three-month interruption due to ribavirin shortage. Then ribavirin reintroduction at a dosage of 800 mg per day for three months, ribavirin cessation for four months, then ribavirin administration at a dosage of 800 mg per day for six months, at a dosage of 1200 mg per day for six months, and at a dosage of 400 mg per day for 10 months.
Thereafter the patient was lost to follow-up
Negative SARS-CoV-2 RNA on day 5. Weaning from oxygen therapy on day 10 and hospital discharge on day 15.
Negative HEV RNA in the blood on month 10 then positive again four months later. After ribavirin reintroduction for 22 months, HEV RNA remained weakly detectable in plasma, on five samples then the patient was lost to follow-up dexamethasone and tocilizumab for his persistent SARS-CoV-2 infection. The exposed patient, with a medical history of hypertension, asthma, obstructive sleep apnea and benign prostatic hyperplasia, experienced HEV infection that resolved spontaneously in approximately 1 month [11]. The third case, described in China, was a man in his 60 s who was immunocompromised due to chronic lymphocytic leukemia [12]. He was concomitantly diagnosed with SARS-CoV-2 and HEV infections by molecular testing, and SARS-CoV-2 replication and related symptoms persisted for approximately 60 days, with administration of methylprednisolone and lately, nirmatrelvir/ritonavir. HEV RNA testing was performed again 1 month post-discharge and was negative. Overall, these cases highlight the increased risk of chronic HEV infection in patients with a weakened immune status in the setting of anti-SARS-CoV-2 therapies, and include one of the rare cases of HEV human-to-human transmission.
As a matter of fact, there is evidence of the deleterious effect of SARS-CoV-2 infection on the liver [13][14][15]. Thus, a Chinese study conducted in 2020-2021 identified that 18% of 7,622 Covid-19 patients experienced a rise of ALT/AST, with two patterns observed including an early one within 14 days of symptom onset and a late one more than 14 days from symptom onset [16]. Such liver perturbations were significantly associated with a lower SARS-CoV-2 qPCR Ct and a longer duration of SARS-CoV-2 RNApositivity, and this suggested a potential direct virus injury to the liver. In another study conducted in 13 Asian countries, among SARS-CoV-2 infected patients, 43% of those with chronic liver disease developed acute liver injury and 20% of those with cirrhosis developed acute-on-chronic liver failure (11%) or acute decompensation (9%) [17]. In addition, liver injury was progressive in 57% of patients with decompensated cirrhosis, with 43% mortality. In an Italian multicenter retrospective study, of 50 SARS-CoV-2-infected patients with cirrhosis, acute-on-chronic liver failure and de novo acute liver injury occurred in 14 (28%) and 10 (20%) patients, respectively [18]. Also, in a German cohort of 72 SARS-CoV-2-infected patients, AST and ALT were increased in 63% and 39% of the cases, while these proportions were 57% and 37%, respectively in a US cohort of 1,219 SARS-CoV-2-infected patients [14]. In addition, Wanner et al also provided evidence by multiple approaches of SARS-CoV-2 tropism to the liver, with viral RNA detection in approximately two-thirds of autopsy liver specimens from patients who exhibited severe COVID-19 and isolation of infectious virus from liver tissue post-mortem. Moreover, they reported in some cases transcriptional and proteomic signatures in liver autopsy samples that were similar to the signatures associated with multiple other viral infections of the human liver, which included significant upregulation of type I and II interferon responses, IFN-related JAK-STAT signaling, and liver-specific metabolic modulation [14]. Recently, SARS-CoV-2 spike and nucleocapsid proteins were detected in liver tissues from SARS-CoV-2-infected patients with severe disease or deceased, and spatial transcriptomics revealed that SARS-CoV-2 RNA was also present in liver tissues [19]. Besides, SARS-CoV-2 was reported to bind to the ACE2 receptor on hepatocytes and cholangiocyte membrane [20]. It is also worthy to note that SARS-CoV, which caused an outbreak of severe acute respiratory diseases in 2003 in Asia, was reported to be associated with hepatitis and was detected in liver tissue samples [15,21]. In the present case, the patient presented a liver cytolysis at approximately 10 times the upper usual values 2 months before HEV diagnosis. Autoimmune hepatitis, which was associated with a bad clinical outcome in a large Covid-19 series [22], was suspected although not ascertained.
Overall, previous data highlight that protracted HEV and SARS-CoV-2 infections that occur in immunocompromised patients increase the risk of co-incidence of these infections that are both quite common worldwide and in Europe. Consequently, HEV infections, acute or chronic, may have been overlooked during the SARS-CoV-2 pandemic. Testing for HEV in case of liver cytolysis may be warranted regardless of another or other possible etiologies, including SARS-CoV-2.
## References
1. Zhu, Zhang, Wang (2019) "A Novel Coronavirus From Patients With Pneumonia in China"
2. Kamar, Izopet, Pavio (2017) "Hepatitis E Virus Infection" *Nature Reviews Disease Primers*
3. Kaba, Brouqui, Richet (2010) "Hepatitis E Virus Infection in Sheltered Homeless Persons, France" *Emerging Infectious Diseases*
4. Cucinotta, Vanelli (2020) "WHO Declares COVID-19 a Pandemic" *Acta bio-medica: Atenei Parmensis*
5. Xu, Li, Tian et al. (2020) "Full Spectrum of COVID-19 Severity Still Being Depicted" *Lancet*
6. Glass, Hoang, Boschi (2021) "Incidence and Outcome of Coinfections With SARS-CoV-2 and Rhinovirus" *Viruses*
7. Aksamentov, Roemer, Hodcroft et al. (2021) "Nextclade: Clade Assignment, Mutation Calling and Quality Control for Viral Genomes" *Journal of Open Source Software*
8. Colson, Fournier, Chaudet (2022) "Analysis of SARS-CoV-2 Variants From 24,181 Patients Exemplifies the Role of Globalization and Zoonosis in Pandemics" *Frontiers in Microbiology*
9. Colson, Borentain, Ravaux et al. (2020) "Hepatitis B Virus Genomics Knocking at the Door of Routine Diagnostic Laboratories" *Journal of Infectious Diseases*
10. Smith, Izopet, Nicot (2020) "Update: Proposed Reference Sequences for Subtypes of Hepatitis E Virus (Species Orthohepevirus A)" *Journal of General Virology*
11. Lampejo, Curtis, Ijaz (2022) "Nosocomial Transmission of Hepatitis E Virus and Development of Chronic Infection: The Wider Impact of COVID-19" *Journal of Clinical Virology*
12. Liu, Tang, Shi et al. (2024) "Hepatitis E Virus and SARS-CoV-2 Co-Infection in an Immunocompromised Patient: A Case Report"
13. Zhang, Shi, Wang (2020) "Liver Injury in COVID-19: Management and Challenges" *Lancet Gastroenterology & Hepatology*
14. Wanner, Andrieux, Badia-I-Mompel (2022) "Molecular Consequences of SARS-CoV-2 Liver Tropism" *Nature Metabolism*
15. Liptak, Nosakova, Rosolanka et al. (2023) "Acute-On-Chronic Liver Failure in Patients With Severe Acute Respiratory Syndrome Coronavirus 2 Infection" *World Journal of Hepatology*
16. Wong, Yip, Wong (2021) "SARS-CoV-2 Viral Persistence Based on Cycle Threshold Value and Liver Injury in Patients With COVID-19" *Open Forum Infectious Diseases*
17. Sarin, Choudhury, Lau (2020) "Pre-Existing Liver Disease is Associated With Poor Outcome in Patients With SARS CoV2 Infection; The APCOLIS Study (APASL COVID-19 Liver Injury Spectrum Study)" *Hepatology International*
18. Iavarone, D'ambrosio, Soria (2020) "High Rates of 30-day Mortality in Patients With Cirrhosis and COVID-19" *Journal of Hepatology*
19. Chen, Zhang, Ashuo (2025) "Combination of Spatial Transcriptomics Analysis and Retrospective Study Reveals Liver Infection of SARS-COV-2 is Associated With Clinical Outcomes of COVID-19" *EBioMedicine*
20. Ji, Zhang, Yang (2020) "Effect of COVID-19 on Patients With Compensated Chronic Liver Diseases" *Hepatology International*
21. Chau, Lee, Yao (2004) "SARS-Associated Viral Hepatitis Caused by a Novel Coronavirus: Report of Three Cases" *Hepatology*
22. Satapathy, Roth, Kvasnovsky (2021) "Risk Factors and Outcomes for Acute-on-Chronic Liver Failure in COVID-19: A Large Multi-Center Observational Cohort Study" *Hepatology International* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12617324&blobtype=pdf | # Contemporary Antiretroviral Therapy Regimens in People With HIV Who Initiated Treatment in the Pre-Antiretroviral Therapy Era
Eisuke Adachi, Shuhei Okuno, Yoshiaki Kanno, Michiko Koga, Hiroshi Yotsuyanagi
## Abstract
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.
To the Editor, The treatment of HIV has evolved dramatically since the preantiretroviral therapy (ART) era, when monotherapy with zidovudine (AZT) or dual nucleoside reverse transcriptase inhibitors (NRTIs) therapy was commonly used [1]. A major milestone was the introduction of combination ART in 1996, consisting of two NRTIs plus a key drug, typically a protease inhibitor (PI). Currently, integrase strand transfer inhibitorsbased regimen, such as dolutegravir/lamivudine (DTG/3TC) or bictegravir/emtricitabine/tenofovir alafenamide (B/F/TAF), as well as long-acting antiretroviral formulations like cabotegravir plus rilpivirine (CAB + RPV), can suppress HIV even in the presence of historical resistance mutations. Nevertheless, lamivudine (3TC) and emtricitabine (FTC) remain widely used, and resistance to these agents is still common [2]. Against this backdrop, we aimed to characterize current ART regimens among heavily treatment-experienced (HTE) people with HIV (PWH), focusing on individuals with over 30 years of continuous ART history.
We retrospectively analyzed PWH who initiated antiretroviral drugs (ARVs) at IMSUT Hospital, Institute of Medical Science, The University of Tokyo, before 1996, and continued from April 2019 onwards. Eligible individuals had available HIV RNA data from both early ARVs, and the initiation of an effective threedrug regimen (3DR), and with confirmed virologic outcomes and the most recent ART regimens since April 2019, when B/F/ TAF became available in Japan. Clinical data collected included age, the year of initial ARV initiation, HIV RNA levels at the start of effective 3DR, the presence of M184V/I mutations, history of ARV use, and the longitudinal history of ART regimens up to April 2025.
Among 202 individuals who initiated ARVs before 1996, 23 remained in follow-up after April 2019. The median current age in 2025 was 64 years (range, 50-82). Initial regimens consisted of NRTI monotherapy in 8 individuals (6 with AZT) and dual NRTI therapy in 14 individuals (Table 1). Seventeen individuals (74%) had detectable viremia at the time of switching to two NRTIs plus a key drug after 1996, with a median HIV RNA of 19,500 copies/mL (range, 740-170,000), while six achieved virologic suppression even before the ART era (Table 2). The M184V mutation had been detected in 7 individuals (30%), and nearly half had prior exposure to 3TC with subsequent virological failure.
Since 2019, the most recent regimens included TAF-based regimens in nine individuals and two-drug regimens (2DR) in 12 individuals. B/F/TAF was the most common TAF-based regimen, while DTG/3TC was the predominant 2DR, with one individual who received CAB + RPV (Table 1). Despite diverse treatment histories and prior resistance, all individuals maintained virologic suppression from April 2019 onward.
In our cohort, the M184V mutation was detected in 30% of individuals. This relatively low detection rate may reflect the limited availability of resistance testing at that time, as commercial assays were not routinely performed. In addition, most pre-ART era regimens involved AZT, which is not associated with M184V, and none received 3TC monotherapy. These This study has limitations. Of 202 individuals who initiated ARVs during the pre-ART era, only 23 individuals remained in followup, raising the possibility of survivor bias. HIV RNA quantification was not available until 1996, and the records of ARV use before that may be incomplete. Nevertheless, this cohort provides rare longitudinal data spanning three decades of continuous therapy in HTE individuals. Regarding treatment options for HTE, as of August 2025, lenacapavir is available in Japan, whereas ibalizumab and fostemsavir remained unapproved [7][8][9].
In summary, despite drug resistance and extensive treatment histories, these individuals are generally able to maintain virologic suppression with modern ART, underscoring the durability and effectiveness of contemporary regimens in HTE populations.
## References
1. Shafer, Vuitton (1999) "Highly Active Antiretroviral Therapy (HAART) for the Treatment of Infection With Human Immunodeficiency Virus Type 1" *Biomedicine & Pharmacotherapy*
2. Acosta, Willkom, Andreatta (2020) "Switching to Bictegravir/Emtricitabine/Tenofovir Alafenamide (B/F/TAF) From Dolutegravir (DTG)+F/TAF or DTG+F/Tenofovir Disoproxil Fumarate (Tdf) In the Presence of Pre-Existing NRTI Resistance" *JAIDS Journal of Acquired Immune Deficiency Syndromes*
3. García-Lerma, Macinnes, Bennett et al. (2004) "Transmitted Human Immunodeficiency Virus Type 1 Carrying the D67N or K219Q/E Mutation Evolves Rapidly to Zidovudine Resistance In Vitro and Shows a High Replicative Fitness In the Presence of Zidovudine" *Journal of Virology*
4. Ministry, Health, Welfare et al. (2025)
5. (2025) "Panel on Antiretroviral Guidelines for Adults and Adolescents, Guidelines for the Use of Antiretroviral Agents in Adults and Adolescents"
6. Sax, Andreatta, Molina (2022) "High Efficacy of Switching to Bictegravir/Emtricitabine/Tenofovir Alafenamide in People With Suppressed HIV and Preexisting M184V/I" *AIDS*
7. Riccardi, Berruti, Del Puente et al. (2019) "Ibalizumab and Fostemsavir in the Management of Heavily Pre-Treated HIV-Infected Patients" *Recent patents on antiinfective drug discovery*
8. Anderson, Van Doornewaard, Turner (2022) "Comparative Efficacy and Safety of Fostemsavir in Heavily Treatment-Experienced People With HIV-1" *Clinical Therapeutics*
9. Dvory-Sobol, Shaik, Callebaut et al. (2022) "Lenacapavir: A First-in-Class HIV-1 Capsid Inhibitor" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645984&blobtype=pdf | # Niacin, an active form of vitamin B3, exerts antiviral function by recruiting β-arrestin through GPR109A to activate the phosphorylation of ERK and STAT1 axis
Shunran Li, Jin Zhao, Ziwen Song, Xinyu Zhang, Bing Lang, Congcong Wang, Huanle Luo, Jun Qian, Caijun Sun
## Abstract
The emergence of viral infectious diseases poses a significant threat to public health, and thus the development of effective and safe antiviral drugs is becoming an urgent priority in the pandemic era. Niacin is an active form of vitamin B3 and has been used for hyperlipidemia treatment for decades, with well-established safety and pharmacological profiles. In this study, niacin was found to be a promising antiviral agent with both nutritional and therapeutic benefits. Further data showed that niacin could bind to its receptor GPR109A and then recruit β-arrestin to promote the ERK-STAT phosphorylation axis, subsequently leading to the activation of interferon signaling. Therefore, these findings provided evidence for repurposing this well-established drug as an antiviral agent and also highlighted the potential of GPR109A as a novel target in antiviral therapy.
IMPORTANCEThe frequent emergence of viral infections is a major concern for global health. Finding safe and effective antiviral treatments has become more urgent than ever. Niacin-a form of vitamin B3-has long been used to treat hyperlipidemia, and its safety is well-established in clinical applications. Notably, this research demonstrates niacin may also work as an antiviral agent, offering both nutritional and therapeutic benefits. Further studies show that niacin binds specifically to a receptor in host cells called GPR109A and then triggers a series of signals inside the cell that ultimately strengthen the host's antiviral interferon system. These findings suggest that this widely available drug could be repurposed to fight against viruses, and that the GPR109A receptor might be a promising new target for antiviral drug development.KEYWORDS niacin, vitamin B3, GPR109A, IFN-β, antiviral agent T he frequent emergence of viral infectious diseases poses a significant threat to public health (1-3). Consequently, developing novel antiviral drugs is becoming a priority in the post-COVID-19 pandemic era. Recent studies have highlighted close interactions between antiviral immunity and various metabolic processes (4, 5). Notably, several key rate-limiting enzymes and metabolites in the tryptophan-kynurenine pathway play crucial roles in immune regulation and exhibit antiviral functions (6). We recently identified kynurenine-3-monooxygenase (KMO) and its metabolite quinolinic acid (QUIN) in the tryptophan-kynurenine pathway as having broad-spectrum antiviral activity (7). To further explore KMO-related metabolites with novel antiviral functions, we screened various structural analogs of QUIN to optimize its antiviral efficacy, safety, and pharmaco kinetics.Niacin, also termed nicotinic acid, or 3-pyridine carboxylic acid, is a structural analog of QUIN and widely recognized as a kind of nutrient that is an active vitamin B3.
Niacin also serves as a precursor to NAD+ (nicotinamide adenine dinucleotide), which is essential to the plethora of cellular processes. Deficiency in niacin could lead to pellagra, with clinical symptoms such as dementia, diarrhea, and dermatitis, ultimately leading to death (8). Moreover, niacin has been extensively used as a medication for hyperlipidemia treatment for decades (9), with well-established safety and pharmacological profiles. However, the role of niacin in immune regulation and antiviral immunity remains poorly investigated.
In this study, we present the novel finding that niacin exerts antiviral functions and further elucidate the underlying mechanisms. Our findings demonstrate that niacin could be developed as a promising antiviral agent, combining both nutritional and therapeutic benefits. This study also enhanced our understanding of the complex relationship between cell metabolism and host antiviral immunity.
## MATERIALS AND METHODS
## Cell lines
293T cells (from the embryonic kidney of a female human fetus), Vero cells (from the kidney of a female normal adult African green monkey), A549 cells (from human alveolar adenocarcinoma basal epithelial cells), and Raw 264.7 cells (from macrophage of a male adult mouse) were cultured in complete Dulbecco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/strep tomycin (Gibco), at 37°C in an atmosphere of 5% of CO 2 . The above cells were preserved in our laboratory.
A549 GPR109A knockdown cells were constructed in our laboratory. In brief, A549 cells were infected with sgRNA-expressing lentivirus with polybrene and then added with puromycin. The cells after puromycin screening were selected and validated by Western blotting analysis. The sgRNA sequences were listed in Table S1.
## Viruses
Vesicular stomatitis virus (VSV)-GFP was generously provided by Dr. Tian Lan (Vec torBuilder, Guangzhou, China), while herpes simplex virus (HSV)-green fluorescent protein-luciferase is stored in our laboratory. These viruses were utilized in cell stud ies. The pathogenic HSV-1 McKrae strain, kindly gifted by Prof. Jumin Zhou (Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, China), was used for the animal study.
## Plasmid constructs
The full-length human GPR109A and β-arrestin1 and mouse GPR109A were cloned into the pcDNA3.1 vector, which is maintained in our laboratory and used as a mock transfection control in this study. The primers used for cloning were listed in Table S2.
## siRNA and cell transfection
Knockdown experiments were performed as previously described (10). In brief, cells at 80% confluence were transfected with various plasmids using Lipofectamine 2000 Transfection Reagent (Invitrogen), following the manufacturer's instructions. After 5 h of transfection, the medium was replaced with DMEM containing 5% FBS and 1% penicillin/streptomycin, and the cells were incubated for an additional 24 or 48 h. siRNA oligonucleotide duplexes targeting GPR109A and β-arrestin were synthe sized by Genepharma (Suzhou, China). According to the manufacturer's protocol, cells were transfected with 100 nM of the indicated siRNAs using Lipofectamine RNAiMax Transfection Reagent (Invitrogen) for 48-72 h. The knockdown efficacy of the target genes was assessed by quantitative real-time PCR (RT-qPCR) or Western blot analysis. The sequences of all siRNAs are provided in Table S3, and the primers for RT-qPCR are listed in Table S4.
## RNA-seq library preparation, sequencing, and data processing
RNA-seq experiments were performed as previously described (11). In brief, Raw264.7 cells were incubated with or without niacin at a dose of 1 mM, after which the cells were infected with HSV-1 for 24 h, total RNA was extracted from the collected samples according to the manufacturer's instructions, and RNA-seq libraries were generated using the TruSeq PE Cluster Kit v4-cBot-HS (Illumina, USA). The prepared libraries were sequenced on an Illumina platform by Sangon Biotech (Shanghai, China). Genes with P-values < 0.05 and |Log2 FC| > 1.5 were considered differentially expressed. Gene Ontology (GO) analysis was performed by using the GO knowledge base (https:// geneontology.org/), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed by using the KEGG database (https://www.kegg.jp/). The volcano plot was drawn by using the Volcano mapping tool in Hiplot (Tengyu Biotech, Shanghai, China).
## Luciferase assay
The Steady-Glo Renilla Luciferase detection system was utilized according to the manufacturer's instructions (Promega, Madison, WI, USA).
## Mice
BALB/c mice were obtained from the Laboratory Animal Resource Center of Sun Yat-sen University and were bred in the SPF animal facility of Sun Yat-sen University in individu ally ventilated cages.
## In vivo therapy
BALB/c mice were bred in the SPF animal facility of the Laboratory Animal Resource Center of Sun Yat-sen University. Six-to 8-week-old mice were anesthetized, and corneal epithelial debridement was performed using a 30-gage needle, followed by the inoculation of 10 5 plaque-forming units (PFU) HSV-1 (McKrae). Intraperitoneal injections of acyclovir (ACV) (5 mg/kg), niacin (5 mg/kg or 15mg/kg), or vehicle alone in 2% DMSO were administered daily for 1 week, and the ocular washes, disease scores, and corneal images (Carl Zeiss stereoscope) were acquired during the experiment. The corneal surface was washed with PBS (20 µL/eye) at various times post-infection, and then the virus titer of these samples was quantified by plaque assay.
## Challenge experiment in mice
BALB/c mice were infected with 1 × 10 5 PFU HSV-1 (McKrae) by corneal inoculation as described above. The disease symptoms of experimental mice were monitored, and samples were collected as described above for subsequent analyzes.
## Enzyme-linked immunosorbent assay (ELISA)
The samples were collected and analyzed with the Mouse Interferon (IFN) Beta ELISA Kit (Solarbio Life Science, Beijing, China) according to the manufacturer's instructions.
## ELISPOT
Enzyme-linked immunosorbent spot (ELISPOT) assays were performed as previously described (12). Briefly, 96-well plates (Millipore, Immobilon-P membrane) were coated with anti-IFN-γ monoclonal antibody (BD Pharmingen) overnight at 4°C and then blocked with 10% fetal bovine serum for 2 h at 37°C. Freshly isolated splenocytes were added in 4 × 10 5 cells/well, and the HSV-1 peptides (Genscript, Nanjing, China) listed in Table S5 were immediately added at a final concentration of 2 mg/mL. The cells were incubated for 24 h at 37°C, and the expression of IFN-γ was then detected using biotinylated polyclonal anti-mouse IFN-γ (BD Pharmingen) and NBT/BCIP reagent (Pierce). Finally, the numbers of spots were quantified using an ELISpot reader (Bio reader4000, BIOSYS, Germany). The data were reported as spot-forming cells (SFC) per million cells.
## Western blotting analysis
The Western Blotting assay was performed as previously described (13). The antibodies used are listed in Table S6 in supplemental material.
## Cytotoxicity assay
A549, J2-BMM, and Vero cells, suspended in DMEM or RPMI 1640 with 10% fetal calf serum, were seeded into 96-well microtiter plates at a density of 1 × 10 4 cells/well. Compounds to be screened were diluted with serum-free DMEM/ RPMI 1640 medium. These solutions were then added to the plates at various concentrations in a final volume of 100 µL. After 24 h, 10 µL of CCK8 reagent (Yeason, Shanghai) was added, and 4 h later, the value of OD450 was read by the microplate reader (Synergy HTX, Biotek). Compounds to be screened were listed in Table S7.
## Hematoxylin and eosin (HE) staining
Tissues from experimental mice were fully immersed in 4% paraformaldehyde, gradually dehydrated, embedded in paraffin, and cut into sections. HE staining was performed according to standard protocol by Wuhan Service Biotechnology CO, LTD.
## Statistical analysis
Statistical analyses were performed using GraphPad Prism software version 8 (GraphPad Software, Inc.). Statistical significance was calculated using Student's two-tailed unpaired t-test or analysis of variance (ANOVA) with Holm-Sidak's multiple comparisons test. *P < 0.05; **P < 0.01; ***P < 0.001.
## RESULTS
## Niacin is a potential antiviral agent
QUIN, also known as 2,3-pyridine carboxylic acid, has been identified in previous studies as a broad-spectrum antiviral agent (7). It is an essential metabolite in the tryptophan metabolism (14,15). Structurally, it comprises a pyridine ring scaffold, with carboxyl groups connecting to the carbon atoms. To find more effective and safer antiviral drugs, we screened a series of compounds that bear structural similarity to QUIN (Fig. S1).
To assess viral inhibition, HSV-1 containing a luciferase reporter gene (7) was used to infect Vero cells treated with the corresponding compounds. Subsequently, cell viability was assessed using the CCK-8 assay (Fig. 1A through K). Our results showed that niacin exhibited promising performance in these assays (Fig. 1B), so we further evaluated its cytotoxicity in Vero, BMM, and A549 cells (Fig. 1L through N). Our findings indicated that niacin as a pharmaceutical agent demonstrated both favorable safety profiles and effective antiviral properties.
## Niacin inhibits HSV-1 replication in various cell lines
To further verify the antiviral effect of niacin, experiments were performed to evaluate its activity across various cell lines. Niacin, at concentrations of 0, 1, 10, 100, 500, and 1,000 µM was co-incubated with the HSV-1 virus and the indicated cell lines for 24 h. Subsequently, the expression of HSV ICP0 or ICP27 was measured by Western blot, and results showed that niacin had the ability to inhibit HSV replication across multiple cell lines in a dose-dependent manner (Fig. 2A through C). Additionally, the anti-HSV-1 effect of niacin in Raw264.7 and Vero cells was confirmed by RT-qPCR (Fig. 2D through E) and plaque assay, respectively (Fig. S2). These findings collectively indicated the inhibitory effect of niacin against HSV-1.
## Antiviral efficacy of niacin in a highly pathogenic HSV-1 infection mouse model
To investigate the antiviral efficacy of niacin in vivo and assess its potential as a thera peutic agent, we employed a murine model infected with a highly pathogenic strain of HSV-1 McKrae. Adult mice aged 6-8 weeks underwent an incision in their eyes, and HSV-1 McKrae was inoculated at a dose of 10 5 PFU/mice. We administered treatments via intraperitoneal injection starting from Day 0, with daily doses thereafter. Acyclovir was utilized as a positive control to evaluate the protective effect of niacin against viral Full-Length Text infection while PBS was used as a negative control (Fig. 3A). Our data demonstrated that niacin treatment significantly increased the weight and survival rate of HSV-infected mice (Fig. 3B andC). Additionally, disease scores were assigned to assess symptomatic improvement in infected mice following niacin administration (Fig. 3D). Scores ranged from 0 to 4. Mice exhibiting no signs of infection and maintaining normal activity receive a score of 0, while mice displaying symptoms such as trembling, huddling, evident ocular infection, and significant weight loss receive a score of 4.
On the 2nd and 4th day post-infection, eye wash samples were collected from the mice to measure viral titers. We found that niacin treatment reduced the viral titers in these mice eyewash samples (Fig. 3E). Subsequently, the mice were euthanized on the 7th day post-infection, and serum and splenic lymphocytes were harvested for ELISA and ELISpot assays. Our findings indicated that niacin therapy increased the frequency of HSV-1 antigen-specific IFN-γ-secreting cells (Fig. 3F) and elevated IFN-β levels in the plasma of HSV-infected mice (Fig. 3G). Niacin treatment also mitigated pathological damage in the brain and spleen of HSV-infected mice (Fig. 3H). In addition, niacin treatment reduced the progressive corneal scarring and visual impairment (Fig. 3I).
Furthermore, we assessed the prophylactic efficacy of niacin against HSV-1 infection by administering daily injections for 7 days prior to viral challenge. Mice were moni tored for 15 days post-infection. Results demonstrated that pretreatment with niacin significantly enhanced survival rates and mitigated weight loss in infected mice (Fig. S3A andB), and reduced disease scores and viral titers at the primary infection site (Fig. S3C andD), achieving effects comparable to niacin treatment initiated concurrently with infection. However, its efficacy in suppressing viral replication within organs was weaker compared with the infection site (Fig. S3E), and no significant impact was detected on plasma IFN-β levels or the frequency of spleen-derived antigen-specific T cells (Fig. S3F andG). Therefore, niacin demonstrated significant therapeutic antiviral efficacy and moderate prophylactic protection.
## Niacin restricts virus replication via GPR109A
Next, we further explored the antiviral mechanism of niacin. Niacin significantly enhanced the expression of type I IFN (Fig. 4A andB). Results of transcriptome sequenc ing revealed that there were 1,091 upregulated genes and 1,617 downregulated genes in niacin-treated cells (Fig. 4C). GO enrichment analysis indicated significant enrichment in the Toll-like receptor signaling pathway, TNF signaling pathway, NF-kappa B signaling pathway, MAPK signaling pathway, and G protein-coupled receptors signaling (Fig. 4D).
Combined with the literature review, we found that GPR109A, a G protein-coupled receptor that could activate the MAPK pathway, is a high-affinity receptor of niacin (16)(17)(18).
Based on this observation, we hypothesized that GPR109A may play a role in the antiviral properties of niacin. Then, we treated Raw 264.7 cells with niacin for 12 h, followed by infection with HSV-1 for 12 h. The data demonstrated niacin treatment significantly upregulated the expression of GPR109A, especially a remarkably synergistic upregulation under HSV-1 infection status (Fig. 4E). Meanwhile, we observed that the expression of GPR109A was significantly increased in various cell lines during HSV-1 and VSV infections (Fig. 4F through I). We further knocked down GPR109A in A549 cells by the CRISPR-Cas9 system (Fig. S4) and found that the inhibition effect on HSV ICP27 expression by niacin was diminished in the GPR109A knockdown A549 cells (Fig. 4J), suggesting the involvement of GPR109A in niacin's antiviral function.
Previous studies have focused on the role of niacin as a hypolipidemic drug and a vitamin. Physiologically, niacin acts as a precursor to NAD + , which supplies protons for a variety of oxidation-reduction reactions within the cell, making NAD + essential for a wide range of metabolic functions. To clarify whether the antiviral function of niacin is also associated with its effects on NAD + and mitochondria activity, we used nicotinamide (NAM), which is the intermediate product of niacin generating NAD + , to treat cells, and our data indicated that NAM does not have antiviral activity (Fig. S5A andB) or the ability to enhance IFN or inflammatory factor secretion (Fig. S5C andD).
## GPR109A inhibits viral replication through upregulating type I IFN
We then cloned the GPR109A gene into plasmid pcDNA3.1 and found that GPR109A overexpression could effectively inhibit HSV-1 and VSV replication by applying RT-PCR and Western blot (Fig. 5A andB). This inhibition effect in GPR109A overexpressing or enhancement effect in GPR109A knockdown cells was also confirmed using the TCID50 assay (Fig. S6A andB). Furthermore, MK-6892, a GPR109A agonist, also significantly suppressed viral replication (Fig. 5C andD). Consistent with niacin treatment, overexpres sion of GPR109A showed comparable results of upregulating type I IFN and IFN-stimula ted genes (ISGs), mirroring the effects of niacin (Fig. 5E through G). To further elucidate the role of GPR109A, we designed siRNA to knock down GPR109A expression (Fig. 5H). Following knockdown, we observed the increased replication of HSV-1 and VSV (Fig. 5I andJ), concomitant with decreased expression of type I IFN and its downstream ISGs (Fig. 5K andL). These findings indicated that GPR109A can exert its antiviral effects through upregulation of type I IFN expression.
## Antiviral effect of GPR109A relies on the β-arrestin-ERK-STAT1 axis
It is well known that G protein-coupled receptors (GPCRs) can react to diverse extracel lular stimuli, and then transduce these signals to different cellular functional outputs mainly via two types of transducers, G proteins and arrestins (19)(20)(21).
A previous study showed that β-arrestin1 can serve as an intracellular adaptor protein for GPR109A, mediating its downstream signaling (22). Our study showed that overex pression of β-arrestin1 also exhibited antiviral effects and upregulated type I IFN and downstream ISGs, similar to GPR109A (Fig. S7A andB). RT-PCR analysis of A549 cells co-transfected with GPR109A and β-arrestin1, followed by HSV-1 infection, revealed a synergistic interaction between these two proteins in upregulating IFN and downstream ISGs (Fig. 6A andB). Furthermore, β-arrestin1 promoted the secretion of type I IFN in response to the stimulation by HSV-1, LPS, Poly(I:C) (Fig. 6C through E) and downstream ISGs (Fig. S7C through F).
Since studies have reported that GPR109A can recruit β-arrestin1 to the cell membrane and promote the phosphorylation of extracellular regulated protein kinases (ERK) (22,23), we therefore performed experiments to detect whether niacin treatment can affect the phosphorylation of ERK. As expected, niacin effectively promoted the level of ERK phosphorylation in GPR109A-transfected 293T cells (Fig. 6F). We further validated this effect using wild-type and GPR109A knockdown A549 cells treated with niacin (Fig. S8A andB). Additionally, knockdown of β-arrestin diminished niacin-induced ERK phosphorylation (Fig. S8C). Additionally, we found that GPR109A overexpression increased STAT1 expression (Fig. 6G), and inhibition of β-arrestin1 reduced the expression of STAT1 and enhanced the replication of HSV-1 (Fig. 6J). GPR109A also enhanced STAT1 phosphorylation induced by IFN γ (Fig. 6H), which was significantly reduced in GPR109A knockdown cells (Fig. 6I). Moreover, to clarify whether the phenomenon of increased STAT1 phosphorylation is mediated by ERK phosphorylation, we treated GPR109A-over expressing cells with the ERK inhibitor PD98059, followed by HSV infection for 12 h, and we found that inhibition of ERK phosphorylation resulted in a significant reduction in STAT1 phosphorylation (Fig. 6K). Together, these findings demonstrated that GPR109A can recruit β-arrestin to promote the phosphorylation of ERK and STAT1, subsequently leading to the activation of IFN signaling.
To further investigate how GPR109A exerts its antiviral function, we constructed mutant variants of GPR109A. It is reported that the phosphorylation pattern of GPCRs might mediate the receptor endocytosis or its downstream signaling (24). Specifically, there is a predicted phosphorylation site S328 in the intracellular domain of GPR109A. The residue S328 to alanine or aspartate was mutated to assess the impact of GPR109A phosphorylation on antiviral and IFN-upregulating functions. Overexpression of corresponding plasmids in A549 cells revealed that the level of phosphorylation at residue S328 mitigated the antiviral function of GPR109A (Fig. S9A andB).
To understand the role of different structural domains of GPR109A in its antiviral function, we generated truncation mutants lacking the extracellular domain, the third and fourth transmembrane domains, or specific intracellular segments. Our results indicated that the extracellular, transmembrane, and intracellular domains were all crucial for the antiviral and IFN-upregulating functions of GPR109A (Fig. S9A andB). However, given that the structural integrity of GPCRs is crucial for their function of signal transduction, studies utilizing truncated forms may be interpreted with caution. To further elucidate the role of GPR109A signaling in antiviral activity, we compared multiple agonists with distinct signaling biases (Fig. S9C). Among these, our data showed that niacin exhibited the most potent antiviral effect. MK-6892 is a potent dual-pathway agonist which activates both Gi and β-arrestin pathways but demon strates higher efficacy toward Gi protein activation. MK-0354 is a Gi-biased agonist with minimal β-arrestin recruitment . Our results demonstrate that agonists with stronger β-arrestin recruitment induce more potent antiviral effects, supporting our proposed mechanism that GPR109A-mediated antiviral activity depends on β-arrestin recruitment and subsequent ERK-STAT1 phosphorylation cascade.
## DISCUSSION
In the present study, we reported that niacin exerts a novel antiviral function in a highly pathogenic HSV-1 infection mouse model. RNA-seq analysis and literature review (25-29) gave hints that the antiviral function of niacin was likely mediated by its high-affinity receptor GPR109A. Our evidence confirmed that GPR109A overexpression suppressed the replication of HSV-1 and VSV, upregulated Type I IFN and downstream ISGs. Additionally, GPR109A agonists besides niacin also exhibited antiviral effects. These findings suggested that GPR109A could be a potential target for antiviral drug develop ment. Our further research demonstrated that the activation of GPR109A can upregulate ISG expression via the β-arrestin-ERK-STAT axis, thereby exerting antiviral effects (Fig. 7). The development of antiviral drugs currently follows two main strategies: targeting virus cellular machinery or targeting host cells/cellular mechanisms involved in antiviral responses (30). Antiviral drugs approved by the FDA are primarily based on the first strategy, which provides good specificity and antiviral efficacy (30,31). However, they often lack broad-spectrum antiviral function and are prone to resistance due to viral mutations under drug selection pressure. The host-targeting approach aims to enhance the host's antiviral immunity, with IFNs being a promising target due to their broadspectrum activity. However, host-targeting drugs often face challenges in drug ability and potential toxicity. As a kind of vitamin, niacin has a high safety threshold and has been used as an anti-hyperlipidemia drug for many years. Its pharmacokinetics are well-characterized. Our study demonstrated that niacin inhibited viral infection through interaction with its receptor GPR109A, providing valuable hints for expanding its indications of niacin as an antiviral drug candidate. More importantly, GPR109A is a member of the GPCR protein family. This suggests that by designing ligands for GPR109A, it may be possible to develop antiviral drugs with fewer side effects and better efficiency.
The therapeutic potential of niacin against viral infections remains poorly studied. While a Canadian clinical trial evaluated adjunctive niacin therapy in HIV infection (32), aiming to counteract tryptophan depletion and immune hyperactivation caused by HIV infection and thus improve CD4 + T-cell recovery, results showed that niacin significantly reduced plasma kynurenine levels and modestly decreased CD4 + T-cell activation (33). Critically, this approach utilized niacin primarily as a tryptophan supplement to enhance antiretroviral outcomes, not to probe its intrinsic antiviral properties. In addition, a screen of 1,500 compounds identified the niacin analog 6-aminonicotinamide (6-AN) as an inhibitor of hepatitis B virus replication (34). Notably, although 6-AN is structurally related to niacin, this study did not address niacin's antiviral potential. Here, we report the first evidence that niacin exhibits direct and intrinsic antiviral activity by inhibiting HSV-1 replication.
Studies have also indicated the involvement of GPR109A in inflammation and tumorigenesis (27,35,36). GPR109A is highly expressed not only in white and brown adipose tissues but also in the spleen and various immune cells including monocytes, macrophages, dendritic cells, and neutrophils (26,28,29,37,38), but its role in antiviral immunity remains poorly understood. Our study demonstrates that activation of GPR109A suppresses HSV-1 replication by upregulating type I IFN and ISGs, thereby uncovering a previously unrecognized function of this receptor in innate antiviral defense. Given that GPR109A signaling can exhibit biased agonism, future studies identifying the specific adaptor proteins responsible for its antiviral effects and quantifying their contributions will be essential for developing targeted therapeutic strategies. by HSV-1 infection for 12 h. pcDNA3.1-empty plasmid and negative control RNA were used as negative control. Then, total cell lysis was used for Western blot assay. The expression level of HSV ICP0 protein, STAT1, β-arrestin, and GAPDH was measured. NC: negative control. (K) pcDNA3.1-GPR109A and vector plasmid were transfected in A549 cells for 24 h, and then PD98059 at a dose of 0, 5, 10, 15, 25, 50 µM and HSV-1 were co-incubated with cells for 12 h. Total cell lysis was used for Western blot assay.
Final data were presented as the mean ± SD of triplicate experiments. ns: no significance. *P < 0.05, **P < 0.01, ***P < 0.001. This study has several limitations. First, the antiviral mechanism mediated by the GPR109A-β-arrestin-ERK-STAT1 signaling axis, although clearly supported by our in vitro data, requires further validation using appropriate in vivo models. Second, the rapid disease progression in our highly pathogenic HSV-1 murine infection model might limit the ability to fully evaluate the long-term therapeutic efficacy of niacin. Animal infection model with extended disease courses would be more suitable for long-term assessment of post-exposure therapeutic regimens in future study. Third, while we demonstrated efficacy against HSV-1, the antiviral activity of nicotinic acid and GPR109A engagement against other viruses remains unexplored. Further studies assessing the preventive or therapeutic potential of niacin against diverse viral pathogens would significantly advance its potential clinical application as a broad-spectrum antiviral drug.
In summary, our study identified niacin can exert antiviral function through its receptor GPR109A-recruiting β-arrestin to promote the phosphorylation of ERK and STAT1 axis. Our findings provided evidence for repurposing this well-established drug as an antiviral agent and also highlighted the potential of GPR109A as a novel target in antiviral therapy.
## References
1. Javanian, Barary, Ghebrehewet et al. (2021) "A brief review of influenza virus infection" *J Med Virol*
2. Wang, Wang, Zhang et al. (2022) "The epidemiology and disease burden of children hospitalized for viral infections within the family Flaviviridae in China: a national crosssectional study" *PLoS Negl Trop Dis*
3. Plourde, Bloch (2016) "A literature review of Zika virus" *Emerg Infect Dis*
4. Artyomov, Van Den Bossche (2020) "Immunometabolism in the single-cell era" *Cell Metab*
5. Zhao, Chen, Li et al. (2020) "Multifaceted functions of CH25H and 25HC to modulate the lipid metabolism, immune responses, and broadly antiviral activities" *Viruses*
6. Tanaka, Tóth, Polyák et al. (2021) "Immune influencers in action: metabolites and enzymes of the tryptophankynurenine metabolic pathway" *Biomedicines*
7. Zhao, Chen, Wang et al. (2022) "Kynurenine-3-monooxygenase (KMO) broadly inhibits viral infections via triggering NMDAR/Ca2+ influx and CaMKII/ IRF3-mediated IFN-β production" *PLoS Pathog*
8. Meyer-Ficca, Kirkland (2016) *Niacin. Adv Nutr*
9. Carlson (2005) "Nicotinic acid: the broad-spectrum lipid drug. A 50th anniversary review" *J Intern Med*
10. Wang, Liu, Liu et al. (2023) "IFNinducible SerpinA5 triggers antiviral immunity by regulating STAT1 phosphorylation and nuclear translocation" *Int J Mol Sci*
11. Feng, Hao, Zhao et al. (2021) "Shell-mediated phagocytosis to reshape viral-vectored vaccine-induced immunity" *Biomaterials*
12. Li, Yang, Wang et al. (2023) "Rapid induction of long-lasting systemic and mucosal immunity via thermostable microneedlemediated chitosan oligosaccharide-encapsulated DNA nanoparticles" *ACS Nano*
13. Zhao, Li, Li et al. (2021) "Broadly antiviral activities of TAP1 through activating the TBK1-IRF3-mediated type I interferon production" *Int J Mol Sci*
14. Musso, Gusella, Brooks et al. (1994) "Interleukin-4 inhibits indoleamine 2,3-dioxygenase expression in human monocytes" *Blood*
15. Platten, Nollen, Röhrig et al. (2019) "Tryptophan metabolism as a common therapeutic target in cancer, neurodegenera tion and beyond" *Nat Rev Drug Discov*
16. Tunaru, Kero, Schaub et al. (2003) "PUMA-G and HM74 are receptors for nicotinic acid and mediate its anti-lipolytic effect" *Nat Med*
17. Offermanns (2006) "The nicotinic acid receptor GPR109A (HM74A or PUMA-G) as a new therapeutic target" *Trends Pharmacol Sci*
18. Wise, Foord, Fraser et al. (2003) "Molecular identification of high and low affinity receptors for nicotinic acid" *J Biol Chem*
19. He, Huang, Jia et al. (2021) "Structural studies of phosphorylation-dependent interactions between the V2R receptor and arrestin-2" *Nat Commun*
20. Smith, Lefkowitz, Rajagopal (2018) "Biased signalling: from simple switches to allosteric microprocessors" *Nat Rev Drug Discov*
21. Reiter, Ahn, Shukla et al. (2012) "Molecular mechanism of β-arrestin-biased agonism at seven-transmembrane receptors" *Annu Rev Pharmacol Toxicol*
22. Walters, Shukla, Kovacs et al. (2009) "β-arrestin1 mediates nicotinic acid-induced flushing, but not its antilipolytic effect, in mice" *J Clin Invest*
23. Kim, Jadhav, Jeong et al. (2015) "Discovery of 4-(phenyl)thio-1H-pyrazole derivatives as agonists of GPR109A, a high affinity niacin receptor" *Arch Pharm Res*
24. Grimes, Koszegi, Lanoiselée et al. (2023) "Plasma membrane preassociation drives β-arrestin coupling to receptors and activation" *Cell*
25. Digby, Martinez, Jefferson et al. (2012) "Anti-inflammatory effects of nicotinic acid in human monocytes are mediated by GPR109A dependent mechanisms" *ATVB*
26. Zandi-Nejad, Takakura, Jurewicz et al. (2013) "The role of HCA2 (GPR109A) in regulating macrophage function" *FASEB J*
27. Singh, Gurav, Sivaprakasam et al. (2014) "Activation of Gpr109a, receptor for niacin and the commensal metabolite butyrate, suppresses colonic inflammation and carcinogene sis" *Immunity*
28. Kostylina, Simon, Fey et al. (2008) "Neutrophil apoptosis mediated by nicotinic acid receptors (GPR109A)" *Cell Death Differ*
29. Wu, Zhang, Teng et al. (2022) "Propionate and butyrate attenuate macrophage pyroptosis and Full-Length Text Journal of Virology November"
30. "CoCrMo alloy particles" *Mil Med Res*
31. Tompa, Immanuel, Srikanth et al. (2021) "Trends and strategies to combat viral infections: a review on FDA approved antiviral drugs" *Int J Biol Macromol*
32. Kausar, Khan, Mujeeb Ur Rehman et al. (2021) "A review: mechanism of action of antiviral drugs" *Int J Immunopathol Pharmacol*
33. Lebouché, Jenabian, Singer et al. (2014) "The role of extended-release niacin on immune activation and neurocognition in HIV-infected patients treated with antiretroviral therapy -CTN PT006: study protocol for a randomized controlled trial" *Trials*
34. Lebouché, Yero, Shi et al. (2020) "Impact of extended-release niacin on immune activation in HIV-infected immunological non-responders on effective antiretroviral therapy" *HIV Res Clin Pract*
35. Ren, Yang, Hu et al. (2019) "Niacin analogue, 6-Aminonicotina mide, a novel inhibitor of hepatitis B virus replication and HBsAg production" *EBioMedicine*
36. Chai, Digby, Choudhury (2013) "GPR109A and vascular inflammation" *Curr Atheroscler Rep*
37. Elangovan, Pathania, Ramachandran et al. (2014) "The niacin/butyrate receptor GPR109A suppresses mammary tumorigenesis by inhibiting cell survival" *Cancer Res*
38. Gille, Bodor, Ahmed et al. (2008) "Nicotinic acid: pharmacological effects and mechanisms of action" *Annu Rev Pharmacol Toxicol*
39. Maciejewski-Lenoir, Richman, Hakak et al. (2006) "Langerhans cells release prostaglandin D2 in response to nicotinic acid" *J Invest Dermatol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724387&blobtype=pdf | # The Shigella siphophage Sf11 tail structure and host attachment mechanism
Sundharraman Subramanian, John Dover, Kristin Parent
## Abstract
A paucity of reports is available describing the structures of Shigella phages, and these have focused to date on a few short, non-contractile podophages and one long contractile myophage. Here, we report the cryo-EM structure of a Shigella sipho phage, where we can visualize the capsid and surface decoration proteins and many components of the tail, including the Tail Tube Protein (TTP), Distal Tail Protein (DTP), Baseplate Hub Proteins (BHUB1 & BHUB2), and the Tape Measure Protein (TMP). The tail is also decorated with six copies of a trimeric tailspike protein that is similar to Sf6-like and P22-like phages. We used mass spectrometry to confirm the identity of the proteins in the mature virion and present atomic models for the majority of these phage proteins. In addition, host range studies show clearly that these tailspike appendages have a homologous function to those in the Sf6-like and P22-like phages in recognizing the O-antigen on the host lipopolysaccharide (LPS). IMPORTANCE Few Shigella phages have been studied structurally to date. By character izing phage Sf11, we see evidence for a tail adaptor domain that is used for decorating the siphophage tail tip with enzymatic, P22-like tailspike proteins. This is important for both understanding the evolutionary relationships among Shigella phages and also could be exploited as a type of protein scaffolding for creating designer phages for therapeutic and/or industrial purposes. KEYWORDS bacteriophage, cryo-EM structure, phage entry, tails, Shigella phage, siphophage V iruses that infect bacteria, called bacteriophages or phages, are highly abundant, with estimates on the order of ~10 31 particles in the biosphere. Most phages contain dsDNA genomes contained within protein shells called capsids and tails that are responsible for host recognition and attachment. Originally, bacteriophages were grouped into families based entirely on the tail morphology, and these included the siphophages (long, non-contractile tails) (1-3), myophages (long contractile tails) (4), and podophages (short, non-contractile tails) (5). Although the International Committee on the Taxonomy of Viruses (ICTV) has recently steered away from these older classifica tions, they are still useful descriptions as virus structure is often highly conserved, even in the absence of sequence similarity.Historically, siphophages have not been well described structurally. However, in the last few years, there has been a rapid increase in structural studies of siphophage tails and/or entire virions as cryo-electron microscopy has evolved. Some examples include lactococcal phage p2 (6), marine Roseobacter phage R4C (7), Salmonella phage χ (8), Staphylococcus phage 80α (9), and Escherichia coli phage T5 (10). Most siphophage tails have the same core building blocks (for review, see, [1][2][3]11]). These include the tail completion protein, tail tube protein, neck protein, and tape measure proteins. The distal tip of siphophage tails is responsible for host recognition and attachment. Tail tips can
be highly diverse in complexity, with many copies of the receptor binding proteins, and can vary drastically according to species. However, most, if not all, have the minimum components of Tal (tail-associated lysozyme) and DTP (distal tail protein).
There have been relatively few characterized Shigella phages (12) and studies on Shigella phages in general. The most well-studied Shigella phage is Sf6, a podophage with a comprehensive body of work regarding structure (13)(14)(15), biochemistry (16,17), and genetics (18,19). Research on Sf6 has been important in demonstrating a fundamen tal shift in the way we think of phage:host interactions, as it was shown that Sf6 has an innate ability to utilize more than one receptor type (12,16,17,20). By contrast, the old canon of thought was that phages are highly specific and only utilize single, highly specific receptors. We sought to investigate whether innate multiple receptor usage was common among Shigella phages or specific to Sf6. In over 100 years of phage biological studies, there were fewer than 40 Shigella phages reported in the literature. This expanded greatly beginning in 2016 when the Parent lab began "hunting" for Shigella phages from a wide variety of environmental samples to understand host range infection profiles among Shigella phages (21)(22)(23). As a result of this work, structures of a few other Shigella phages have been described, such as the N4-like podophages Moo19 and B2 (24), podophage HRP29 (25), and myophage Sf14 (26). However, no structures of Shigella siphophages have been published to date.
Here, we use cryo-electron microscopy (cryo-EM) and receptor binding studies to describe the tail complex of Shigella siphophage Sf11. We resolved the entire virion at high resolution, where we were able to fully model the capsid, including the major capsid and head decoration proteins, and the portal complex. In addition, we were able to model the tail apparatus and tail tip. We confirmed the presence of virion-associated proteins with mass spectrometry, including some that were not resolved in the cryo-EM reconstruction. The core proteins in the tail are conserved and consistent with siphoph ages observed in other species. However, we show that an adaptor domain on the DTP is utilized to decorate the long non-contractile tail with Sf6-like tailspikes at the distal tip. Receptor binding studies show that this phage utilizes the O-antigen of lipopolysac charide (LPS) as a primary receptor, similarly to its distant podophage relatives. Negative stain electron microscopy shows that infection is not localized to a specific feature (e.g., such as the cell poles, etc). The discovery of Sf6-like tailspike proteins spanning diverse podophages like Sf6, Moo19, B2, and the siphophage Sf11 suggests evolutionary relationships among tails of Shigella phages.
## RESULTS AND DISCUSSION
## Virion structure
Shigella phage Sf11 was isolated and briefly described previously using negative stain electron microscopy to confirm siphophage morphology (21). Here, we investigated the structure of the entire virion using high-resolution cryo-electron microscopy to visualize the entire virion in greater detail. See Fig. S1 and S2 for the cryo-EM processing workflow for all structures. The capsids have clear icosahedral symmetry and long, flexible tails, with decorated ends (Fig. 1A). We used mass spectrometry to identify proteins associated with a purified, high-titer Sf11 stock to determine which gene products were part of mature virions. Phage proteins listed in Table 1 were positively identified from trypsindigested fragments of an excised SDS-PAGE gel band with over 95% confidence using Michigan State University's core proteomics facility.
Next, we performed a reconstruction of the capsid imposing icosahedral (532) symmetry to 3.4 Å resolution. Sf11 has a ~50 kbp genome (Genbank accession number MF158038) encapsidated in an icosahedral shell formed by 415 copies of the major capsid protein in a typical T = 7 geometry (Fig. 1B andC). The major capsid protein (gp39) adopts the classical HK97-like fold adorned by most, if not all, dsDNA-tailed bacteriophages (28). The Sf11 major capsid protein model derived from our cryo-EM data contains many of the hallmarks of the HK97-like fold, including the P domain, E-loop, G-loop, and A domain (Fig. 1D). We were able to model residues 46-355 but could not resolve the full-length N-terminal "N-arm. " Mass spectrometry data of mature Sf11 virions (Table 1) showed nearly complete coverage of the mature major capsid protein spanning residues 40-356. Therefore, it is likely that the N-arm is cleaved during maturation, as is observed in phages such as HK97, T5, and others (29,30). The copy number of each protein per virion and residues modeled is listed in Table 1 for all Sf11 structures shown. Like many phages, the Sf11 capsid also has several copies of a surface-bound auxiliary protein (gp40) called a decoration, or "cement, " protein. Decorative proteins are often used to stabilize phage capsids and take on a few distinct folds (31). Visual inspection of the Sf11 decoration protein (gp40) shows that it forms a homotrimer (Fig. 1E) and is structurally similar to the β-tulip family of decoration proteins, such as phage λ's gpD (32) and gp87 of the hyperthermophilic phage P74-26 (33), despite no recogniza ble sequence homology. A DALI search (34) indicates the best structural equivalent is from marine phage TW1 (PDB 5WK1) at 26% sequence similarity, a root mean square deviation (R.M.S.D.) of 1.6 Å for Cα atoms, and a Z score of 20.0. See Fig. 1F for an alignment showing structural homology. Furthermore, Sf11's gp40 displays the same capsid binding occupancy as phage λ, as it binds the capsid at all of the icosahedral threefold symmetry sites. However, unlike phage λ, the N-terminus of gp40 does not appear to be the stabilizing binding force cementing the decoration protein to the capsid (35), as long extensions of the N-terminus of gp40 were not visible in binding to the capsid in the cryo-EM density map of Sf11.
## Portal and tail structures
We performed a focused reconstruction of the capsid at the unique vertex that binds to the tail. First, we aligned the capsid using EMD-8867, which is the portal and neck of marine phage TW1 (36). We then relaxed the imposed icosahedral symmetry, recen tered the particles to focus on the neck region, and applied C1, C3, and C6 symmetry, individually. Only C6 symmetry improved the refinement. We then further restricted the box size and performed 3D classification. We picked two classes that showed a clearly defined neck region, then applied C12 symmetry for an additional round of refinement, applying both local CTF correction and C12 symmetry. We were able to solve the portal complex to 2.35 Å (Fig. S2). ModelAngelo (37) was then used to refine the portal structure de novo (Fig. 2B). The portal of Sf11 is a dodecameric structure like the vast majority of phage portals (38) (Fig. 2A). The structure of the portal protein has domain features in common with many phages including obvious stalk, stem, and wing domains (Fig. 2B). It is unclear if the Sf11 portal has an extended C-terminal barrel such as phage P22 (39), as we could only model residues 29-410 of a 470 amino acid protein. However, prediction programs such as HHPred and AlphaFold 3 did not predict helical features in the C-terminus (data not shown). A DALI search shows that the closest structural homolog is PDB ID 4ZJN, a portal protein in bacteriophage G20 (40), with a Cα R.M.S.D. of 3.8 Å over 313 residues and a Z score of 20.2 (Fig. 2C).
Unfortunately, despite our best efforts, we could not resolve the rest of the neck region. We relaxed the C12 symmetry in the region adjoining our well-aligned portals and tried applying C1, C3, C5, and C6 symmetry individually. No improvement in the neck region was observed, and we believe this connector region is a bit flexible, inhibiting further refinement. Furthermore, we analyzed the remaining proteins detected in our mass spectrometry data that were not accounted for in our density maps (Table 1). This includes four proteins annotated as "hypothetical, " one annotated as a head morphogenesis protein, and one designated as an HK97 gp10 family protein. Despite using BlastP, AlphaFold 3, and DALI to analyze sequences and predicted protein folds, none had significant homology to any known neck or collar region in tailed phage. Therefore, we cannot determine which proteins comprise the neck region of Sf11 at this time.
We also could not perform a reconstruction of the entire virion, as the tails are very curved and not consistent from particle to particle. However, we performed a focused reconstruction of the tail, including a map where we used symmetrized reconstructions (C3) as well as a completely asymmetric (C1) reconstruction. We resolved the tip of the tail (Fig. 3A) as well as the tail repeating unit (gp24) to 3.2 Å resolution. See Fig. S1 for the cryo-EM processing workflow for tail tip structures. The tail repeating unit forms the length of the siphophage tail as a series of rings (Fig. 3B). Whether the rings stack with true helical symmetry is unclear, as we were unable to analyze the very curvy tails using helical reconstruction approaches. In general, the core of the tail tip of phage Sf11 is highly structurally equivalent to that of phage χ (8). Two proteins within the core are structurally very similar to those in phage χ, including the Baseplate Hub 1 Protein (BH1P, gp16) and Baseplate Hub 2 Protein (BH2P; gp18). See Fig. S3 and S4 for comparisons of these proteins. Additionally, the alpha helical C-terminus of the Tape Measure Protein (TMP; gp18) was observed embedded in the tail tip as has been seen in phages T5 (10), λ (41), 80α (42), χ (8), as well as Bxb1 (43).
A major difference when compared with other siphophages, such as phage χ, in the Sf11 tail tip assembly is the presence of six large appendages (Fig. 1A, arrows). The overall shape of the density map (Fig. 4) is reminiscent of a tailspike appendage, which is an enzymatic protein found in podophages such as Salmonella phage P22 (44) and Shigella phage Sf6 (45). In Sf6, this is the protein that hydrolyzes Shigella lipopolysaccharide (LPS) and determines host range by serotype (19,46). The presence of a structurally homologous P22-like tail spike in a siphophage has been seen before in a mature virion in the marine phage TW1 (36). How this tail spike appendage was assembled onto the tail machinery could not be determined due to the low resolution of the TW1 structure. In Salmonella phage 9NA, a tailspike gene was identified in the genome, and the protein was recombinantly expressed and subsequently crystallized, revealing a P22-like tailspike trimer (47). However, how this was assembled onto the 9NA virion was also not observed. Alternatively, here, we were able to clearly resolve the tailspike attachment to the siphophage tail and identify an adaptor region of DTP (gp17) that is crucial for this protein (gp13) to join onto the virion.
We could not fully model side chains de novo for the entire protein length for either DTP (gp17) or tailspike (gp13), as the resolution was a bit lower in this region, likely owing to flexibility (see Fig. S2 for a local resolution map). However, we successfully built atomic models de novo for the well-resolved regions of the cryo-EM density map in order to analyze the interaction sites between the two proteins (Fig. 4A). The N-terminal alpha helices of gp13 (residues 4-59) were modeled into the high-resolution tail density map, as well as residues 3-165 of the DTP gp17. The DTP protein has a single helical adaptor region that interacts with all three N-terminal helices from gp13 to create a helical bundle that appears to stabilize the tailspike trimer assembly onto the tail (region highlighted with the red box in Fig. 4A andB). There are several key salt bridges between the four helices. Residue Glu136 in DTP interacts with tailspike Lys19 at a distance of 4.0 Å. A second and third salt bridge forms between two DTP residues, Glu147 and Glu152, with tailspike residue Lys19 at distances of 2.3 Å and 2,7 Å (Fig. 4A). Several longer-range electrostatic interactions (~5-10 Å apart) are also observed along the length of the helical bundle that could stabilize the complex as well. For example, Glu129 on DTP interacts with Arg16 on the tail spike at a distance of around 5.2 Å (Fig. 4A).
We also used AlphaFold to model a monomer of the full-length DTP (gp17) and the full-length tailspike (gp13), as shown in Fig. 4B andC. Like typical P22-like tailspikes (5), the Sf11 gp13 tailspike is a homotrimer. In the Sf11 tail, the DTP protein forms a ring made of six monomer subunits (Fig. 4A andD). Six gp13 trimers join to the gp17 hexameric ring for a total of 18 tailspike protein subunits. The C-terminal region of tailspike trimers binds cell surface receptor(s) such as LPS (46) and OmpC (19) in phages P22 and Sf6. This C-terminal region in Sf11 fits into the density map well without the need for flexible fitting (Fig. 4D). However, the N-terminus of gp13 (residues 1-162) is rigid and elongated in the AlphaFold model (Fig. 4C) and does not dock into the cryo-EM density map of the virion without adjustment. Visual inspection of the AlphaFold models and the cryo-EM density shows that there is likely a flexible "hinge" region around residue 162 that must bend to allow gp13 attachment to the tail.
There is also a large extra appendage of gp17 in the C-terminal domain that extends away from the tail, of currently unknown function (Fig. 4B andD). In order to fully model the C-terminal region, we would need to use flexible fitting to adjust the AlphaFold model. Since the resolution of this region of the map was too poor for accurate flexible fitting, we did not attempt this. The biological purpose of this extra domain is not clear at this time; however, we note this feature is one of the largest DTP proteins resolved in any phage tail that we are aware of. A DALI search of gp17 showed little structural homology to known phage structures, with the top hit being 22% identity and an R.M.S.D. of 16.6 Å and a Z-score of 14.6 for the gp31 sheath protein from Pseudomonas phage E217 (48). Interestingly, this protein makes the sheath of a contractile myophage tail and is not functionally related to the DTPs that connect the tails to tips in any known siphophages. The tail tip protein domain is commonly found in tail tips and hubs (Tal and Dit proteins), as well as tail terminator proteins and major tail proteins, suggesting a common evolutionary origin (11,49).
## Sf11 has an Sf6-like tailspike that utilizes the O-antigen for entry
Based on the cryo-EM reconstruction of the Sf11 tail, it appeared that Sf11 has a tailspike-like protein. Sf6 utilizes lipopolysaccharide (LPS) as the initial and reversible primary receptor, and outer membrane proteins A or C (Omps) as the secondary, irreversible receptors (16,17). The enterobacterial common antigen ("ECA") has been shown to be important for defense against podophages with similar tailspikes, such as Salmonella phage P22 (50). Therefore, we tested whether Sf11 uses similar receptors by performing qualitative spot tests on single-gene deletion strains with knockouts of these various surface components (Fig. 5). Productive infections produce plaques, and non-productive infections do not. Deletions of genes wecC and wecD (required for ECA synthesis) had no obvious effect on Sf11 plaquing. Next, we tested deletions of waaC, waaL, or waaG, which are truncations to create progressively shorter versions of LPS. All three deletions completely inhibited Sf11 growth, demonstrating that LPS is essential for Sf11 attachment and infection (Fig. 5), and importantly, the full-length molecule is essential. Next, we tested Sf11 for dependence on outer membrane proteins (Omps). Previous work has shown that various Omps can act as receptors for Shigella phages, such as OmpA, OmpC, and/or OmpF. Individual ompF, ompA, or ompC knockouts had no effect on Sf11 plaque formation. Taken together, these results indicate that these outer membrane proteins are not essential for Sf11 infection. Lastly, the inner membrane protein YajC was found to be essential for Sf6 as previously shown ( 25), but was not essential for Sf11 entry.
## Sf11 entry can occur anywhere on the cell surface
Some siphophages are known to be flagellotropic, meaning they bind to the bacterial flagella as a primary receptor and then bind to LPS as a secondary receptor (e.g., phage χ) (51,52). In this model, the siphophages "walk" down the flagella in a directional manner. By contrast, others bind preferentially to very specific membrane transporters, which are localized to specific areas of the cell surface. For a classical example, phage λ binds to a sole proteinaceous receptor, LamB, which is preferentially localized to the cell poles (53). Since Shigella are nonflagellated (54), and the putative primary receptor is LPS based on our host range studies (Fig. 5), we hypothesized that Sf11 could bind anywhere on the Shigella flexneri cell surface and would not likely demonstrate preferential binding. To test this hypothesis, we imaged phage bound to host cells using electron microscopy of negatively stained samples (Fig. 6). After a relatively short incubation period of 30 min at 22°C, we saw phages bound all over the surface of S. flexneri PE577 cells, indicating no specific binding sites for Sf11 (Fig. 6). This is similar behavior to Shigella podophage Sf6, which has a similar tailspike protein and also does not demonstrate preferential binding to specific cell surface features (16).
## Summary
Traditionally, siphophages have relatively few types of tail tip morphologies, and the DTP often serves as an attachment site for critical components of the infection machinery. Typically, this protein forms a hexameric ring and represents a baseplate docking site (1, 3), with the C-terminal domain displaying variability in terms of structural moieties providing a means for different mechanisms of host adsorption (49). This has been observed in phage T5 (55), where the DTP serves to attach the side tail fibers, and in phage p2, it serves as an attachment site for the receptor binding protein (RBP) (56).
In a recent structure of phage χ, extensions on the DTP were proposed to serve an alternative function. Sonani et al. suggested that the ring formed by DTP serves to reduce the complexity of a symmetry mismatch between the C6 symmetry of the tail and the C3 symmetry of the tip (8). The extension of the C-terminal domain of gp17 serves both functions in Shigella phage Sf11, simultaneously bridging symmetry mismatches between the tail and tip and providing a docking site for the LPS-hydrolyzing tailspike gp13. Additionally, comparing the virion shape while in solution (Fig. 1A) with that attached to cells (Fig. 6) reveals that upon cell attachment, the virion tails stiffen and no longer appear flexible. It is likely that once the tail spike appendage docks to the LPS receptor and the tail tip interacts with the membrane, a signal is transmitted along the length of the tail, inducing conformational changes and subsequent rigidity to allow the dsDNA genome to transmit along the length of the tail into the host. Atomic details of this mechanism remain to be determined and will be the focus of future studies. Why would siphophage Sf11 have a podophage tail spike, and how did it acquire this appendage? It appears that many Shigella phages utilize tailspike proteins in their tails. This is well documented for phage Sf6 (46), which has a tail machine very similar to E. coli phage CUS-3 and Salmonella phage P22 (57). Tail spikes have also been reported in other Shigella podophages with drastically different morphologies. Examples include phage HRP29, which has a hybrid tail that is a mix between phages T7 and Sf6 (25), as well as phage Moo19, which is an N4-like phage with an entirely different tail baseplate (24). The tailspike is important for host recognition in these Shigella phages, and they can evolve to infect new hosts through point mutations in this protein (19). Since the vast majority of known Shigella phages come from phage hunting studies (12,(21)(22)(23) using similar "bait bacteria, " it is possible we are selecting for phages that preferentially use this receptor and have an overabundance of this structure in our library. However, the fact that this same structural unit is found in phages with such diverse morpholo gies suggests a strong evolutionary relationship between these phages, as conserved structures found in nature often highlight their biological importance. Proteins involved in the receptor binding process can be conserved across broad groups of phages. This phenomenon of shared receptor protein structures between siphophages and podoph ages has been observed in other systems, such as Staphylococcal phages (9,58,59). Since horizontal gene transfer is a common evolutionary mechanism among phages, tail component sharing may be facilitated by horizontal acquisition across highly diverse phages.
Adaptor domains that allow for tailspike protein addition to diverse tail structures may be a way to allow for a universal, and "generalist" strategy of using tailspikes to aid in Shigella phage cell entry. This could be exploited to create designer phages with different tail appendages for use in therapeutic and/or industrial applications.
## MATERIALS AND METHODS
## Phage isolation, purification, and amplification
Sf11 stocks were grown from a single plaque in a 30 mL preparation using our standard approaches and yielded a working stock at 4 × 10 11 plaque-forming units per mL (PFU/ mL). This high-titer working stock was used to seed a larger, 500 mL prep at a multiplicity of infection (MOI) of 0.1 in LB with S. flexneri PE577 (serotype Y) for 2 h, followed by spiking the culture with additional cells. The prep was grown for a total of 4 h at 37°C while shaking at 200 rpm. The lysate was centrifuged at 4°C for 30 min at 8,000 × g to remove debris, and the supernatant was spun at 4°C for 120 min at 20,000 × g. The phage pellet was then resuspended by overnight nutation at 4°C with 4 mL of phage dilution buffer (10 mM Tris, pH 7.6, 10 mM MgCl 2 ) for a final titer of ~1.5 × 10 12 PFU/mL.
## Transmission electron microscopy and three-dimensional image reconstruc tion
A small aliquot (~5 µL) of phage was added to Quantifoil R2/2 grids that were glow-dis charged using a Pelco EasiGlow glow discharge apparatus and frozen with a Vitrobot Mark IV with a blot force of 1, blot time of 5 s, at 4°C and 100% humidity. Cryo-EM data were collected at the University of Michigan Cryo-EM facility using a Titan Krios equipped with a K3 direct electron detector and operating at 300 keV with a post-column GIF (20 eV slit width). Micrographs were collected at 105,000× magnification (0.834 Å/pixel) by recording 50 frames over 1.93 s for a total dose of 50.45 e-/Å2. Data processing was carried out using Cryosparc v4.6.0 (60). The micrographs were first motion corrected using Patch Motion correction, followed by CTF estimation using Patch CTF estimation, and particles were picked using template picker. For capsid, a total of 43,242 particles were used for 3D refinement (Icosahedral symmetry), and for tail, a total of 35,748 particles were used for 3D refinement (C3 symmetry). The overall map resolutions were estimated based on the gold-standard Fourier shell correlation (FSC 0.143) (61). The final maps were deposited into the Electron Microscopy Data Bank (EMD-70309, EMD-70310, and EMD-72999; see Table 2). We generated initial models using ModelAngelo (57), using a combination of both sequence and non-sequence modes. Refinement was carried out using Phenix (58), and model adjustments were carried out in COOT (59). Models were deposited into the Protein Data Bank (PDB ID: 9OCB, 9OCC, and 9YIN; see Table 2)
## Host range and efficiency of plating
Initial host range results were performed by combining bacterial cells in a double agar overlay method. Receptor binding assays were done by spot tests. Once agar containing each bacterial host solidified, 2 µL of phage stock and subsequent serial dilutions were applied to the top of the agar and left to dry before incubating overnight. Hosts with a positive result showed a cleared spot the next day, whereas hosts that produced a negative result had no inhibited cell growth. All spot tests were done with biological replicates (at least three plates per strain).
## Negative staining of phage attachment to hosts
An equal volume of PE577 cells grown overnight in LB at roughly ~10 8 cells/mL and phage Sf11 stock at ~1 × 10 11 phage per mL were incubated for 30 min at RT (~22°C). A small aliquot of 5 µL was deposited onto continuous carbon copper grids (TedPella) and stained with 1% uranyl acetate. Micrographs were imaged using a Talos Arctica and a Ceta camera at a lower magnification (13,500×; pixel size of 1.07 nm) and at a higher magnification of 45,000× (pixel size of 3.2 Å)
## References
1. Goulet, Spinelli, Mahony et al. (2020) "Conserved and diverse traits of adhesion devices from Siphoviridae recognizing proteinaceous or saccharidic receptors" *Viruses*
2. (2025) *Full-Length Text Journal of Virology*
3. Linares, Arnaud, Degroux et al. (2020) "Structure, function and assembly of the long, flexible tail of siphophages" *Curr Opin Virol*
4. Davidson, Cardarelli, Pell et al. (2012) "Long noncontractile tail machines of bacteriophages" *Adv Exp Med Biol*
5. Taylor, Van Raaij, Leiman (2018) "Contractile injection systems of bacteriophages and related systems" *Mol Microbiol*
6. Casjens, Molineux (2012) "Short noncontractile tail machines: adsorption and DNA delivery by podoviruses"
7. Bebeacua, Tremblay, Farenc et al. (2013) "Structure, adsorption to host, and infection mechanism of virulent lactococcal phage p2" *J Virol*
8. Huang, Sun, Wei et al. (2023) "Structure and proposed DNA delivery mechanism of a marine roseophage" *Nat Commun*
9. Sonani, Esteves, Scharf et al. (2024) "Cryo-EM structure of flagellotropic bacteriophage Chi"
10. Kizziah, Manning, Dearborn et al. (2020) "Structure of the host cell recognition and penetration machinery of a Staphylococcus aureus bacteriophage" *PLoS Pathog*
11. Linares, Arnaud, Effantin et al. (2023) "Structural basis of bacteriophage T5 infection trigger and E. coli cell wall perforation" *Sci Adv*
12. Acapito, Decombe, Arnaud et al. (2025) "Comparative anatomy of siphophage tails before and after interaction with their receptor" *Curr Opin Struct Biol*
13. Subramanian, Parent, Doore (2020) "Ecology, structure, and evolution of Shigella phages" *Annu Rev Virol*
15. Bhardwaj, Molineux, Casjens et al. (2011) "Atomic structure of bacteriophage Sf6 tail needle knob" *J Biol Chem*
16. Li, Hou, Yang et al. (2022) "High-resolution cryo-EM structure of the Shigella virus Sf6 genome delivery tail machine" *Sci Adv*
17. Parent, Gilcrease, Casjens et al. (2012) "Structural evolution of the P22-like phages: comparison of Sf6 and P22 procapsid and virion architectures" *Virology (Auckl)*
18. Parent, Erb, Cardone et al. (2014) "OmpA and OmpC are critical host factors for bacteriophage Sf6 entry in Shigella" *Mol Microbiol*
19. Porcek, Parent (2015) "Key residues of S. flexneri OmpA mediate infection by bacteriophage Sf6" *J Mol Biol*
20. Casjens, Da, Gilcrease et al. (2004) "The chromosome of Shigella flexneri bacteriophage Sf6: complete nucleotide sequence, genetic mosaicism, and DNA packaging" *J Mol Biol*
21. Subramanian, Dover, Parent et al. (2022) "Host range expansion of Shigella phage Sf6 evolves through point mutations in the tailspike" *J Virol*
22. Doore, Parent, Subramanian et al. (2021) "Bacteriophage receptor proteins of gram-negative bacteria"
23. Doore, Schrad, Dean et al. (2018) "Shigella phages isolated during a dysentery outbreak reveal uncommon structures and broad species diversity" *J Virol*
24. Doore, Schrad, Perrett et al. (2019) "A cornucopia of Shigella phages from the Cornhusker state" *Virology (Auckl)*
25. Bittle, Brittain, Doore et al. (2023) "Phage hunting in the high school classroom: phage isolation and characteriza tion" *Am Biol Teach*
26. Subramanian, Drarvik, Tinney et al. (2024) "Moo19 and B2: structures of Schitoviridae podophages with T = 9 geometry and tailspikes with esterase activity" *Sci Adv*
27. Subramanian, Drarvik, Tinney et al. (2024) "Cryo-EM structure of a Shigella podophage reveals a hybrid tail and novel decoration proteins" *Structure*
28. Subramanian, Kerns, Braverman et al. (2025) "The structure of Shigella virus Sf14 reveals the presence of two decoration proteins and two long tail fibers" *Commun Biol*
29. Pettersen, Goddard, Huang et al. (2004) "UCSF Chimera-A visualization system for exploratory research and analysis" *J Comput Chem*
30. Duda, Teschke (2019) "The amazing HK97 fold: versatile results of modest differences" *Curr Opin Virol*
31. Huang, Khayat, Lee et al. (2011) "The prohead-I structure of bacteriophage HK97: implications for scaffoldmediated control of particle assembly and maturation" *J Mol Biol*
32. Huet, Duda, Hendrix et al. (2016) "Correct assembly of the bacteriophage T5 procapsid requires both the maturation protease and the portal complex" *J Mol Biol*
33. Dedeo, Teschke, Alexandrescu (2020) "Keeping it together: structures, functions, and applications of viral decoration proteins" *Viruses*
34. Yang, Forrer, Dauter et al. (2000) "Novel fold and capsid-binding properties of the λ-phage display platform protein gpD" *Nat Struct Biol*
35. Stone, Hilbert, Hidalgo et al. (2018) "A hyperthermophilic phage decoration protein suggests common evolutionary origin with herpesvirus triplex proteins and an anti-CRISPR protein" *Structure*
36. Holm, Laiho, Törönen et al. (2023) "DALI shines a light on remote homologs: one hundred discoveries" *Protein Sci*
37. Lander, Evilevitch, Jeembaeva et al. (2008) "Bacteriophage lambda stabilization by auxiliary protein gpD: timing, location, and mechanism of attachment determined by cryo-EM" *Structure*
38. Wang, Hardies, Fokine et al. (2018) "Structure of the marine siphovirus TW1: evolution of capsidstabilizing proteins and tail spikes" *Structure*
39. Jamali, Käll, Zhang et al. (2024) "Automated model building and protein identification in cryo-EM maps" *Nature*
40. Dedeo, Cingolani, Teschke (2019) "Portal protein: the orchestra tor of capsid assembly for the dsDNA tailed bacteriophages and herpesviruses" *Annu Rev Virol*
41. Tang, Lander, Olia et al. (2011) "Peering down the barrel of a bacteriophage portal: the genome packaging and release valve in p22" *Structure*
42. Williams, Levdikov, Minakhin et al. (2013) "12-Fold symmetry of the putative portal protein from the Thermus thermophilus bacteriophage G20C determined by X-ray analysis" *Acta Crystallogr Sect F Struct Biol Cryst Commun*
43. Wang, Duan, Gu et al. (2024) "Architecture of the bacteriophage lambda tail" *Structure*
44. (2025) *Full-Length Text Journal of Virology*
45. Kizziah, Mukherjee, Parker et al. (2025) "Structure of the Staphylococcus aureus bacteriophage 80α neck shows details of the DNA, tail completion protein, and tape measure protein" *Structure*
46. Freeman, Mondal, Macale et al. (2025) *Structure and infection dynamics of mycobacteriophage Bxb*
47. Steinbacher, Seckler, Miller et al. (1994) "Crystal structure of P22 tailspike protein: interdigitated subunits in a thermostable trimer" *Science*
48. Chua, Manning, Morona (1999) "The Shigella flexneri bacterio phage Sf6 tailspike protein (TSP)/endorhamnosidase is related to the bacteriophage P22 TSP and has a motif common to exo-and endoglyca nases, and C-5 epimerases" *Microbiology (Reading)*
49. Müller, Barbirz, Heinle et al. (2008) "An intersubunit active site between supercoiled parallel beta helices in the trimeric tailspike endorhamnosidase of Shigella flexneri phage Sf6" *Structure*
50. Andres, Roske, Doering et al. (2012) "Tail morphology controls DNA release in two Salmonella phages with one lipopolysaccharide receptor recognition system" *Mol Microbiol*
51. Li, Hou, Lokareddy et al. (2023) "High-resolution cryo-EM structure of the Pseudomonas bacteriophage E217" *Nat Commun*
52. Veesler, Cambillau (2011) "A common evolutionary origin for tailedbacteriophage functional modules and bacterial machineries" *Microbiol Mol Biol Rev*
53. Bohm, Porwollik, Chu et al. (2018) "Genes affecting progression of bacteriophage P22 infection in Salmonella identified by transposon and single gene deletion screens" *Mol Microbiol*
54. Esteves, Scharf (2022) "Flagellotropic bacteriophages: opportuni ties and challenges for antimicrobial applications" *Int J Mol Sci*
55. Gambino, Sørensen (2024) "Flagellotropic phages: common yet diverse host interaction strategies" *Curr Opin Microbiol*
56. Chatterjee, Rothenberg (2012) "Interaction of bacteriophage l with its E. coli receptor, LamB" *Viruses*
57. Schnupf, Sansonetti (2019) "Shigella pathogenesis: new insights through advanced methodologies" *Microbiol Spectr*
58. Flayhan, Vellieux, Lurz et al. (2014) "Crystal structure of pb9, the distal tail protein of bacteriophage T5: a conserved structural motif among all siphophages" *J Virol*
59. Sciara, Bebeacua, Bron et al. (2010) "Structure of lactococcal phage p2 baseplate and its mechanism of activation" *Proc Natl Acad Sci*
60. Parent, Tang, Cardone et al. (2014) "Three-dimensional reconstructions of the bacteriophage CUS-3 virion reveal a conserved coat protein I-domain but a distinct tailspike receptor-binding domain" *Virology (Auckl)*
61. Hawkins, Kizziah, Hatoum-Aslan et al. (2022) "Structure and host specificity of Staphylococcus epidermidis bacteriophage Andhra" *Sci Adv*
62. Hrebík, Štveráková, Škubník et al. (2019) "Structure and genome ejection mechanism of Staphylococcus aureus phage P68" *Sci Adv*
63. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
64. Henderson, Sali, Baker et al. (2012) "Outcome of the first electron microscopy validation task force meeting" *Structure*
65. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548478&blobtype=pdf | # Primary envelopment of Kaposi's sarcoma-associated herpesvirus at the nucleoplasmic reticulum
Alexa Wilson, Neale Ridgway, Craig Mccormick
## Abstract
Herpesvirus egress begins with primary envelopment of newly assembled capsids at the inner nuclear membrane (INM). Primary envelopment has been observed at the peripheral INM as well as nuclear infoldings. Nuclear infoldings from invagina tions of the INM are Type-I nucleoplasmic reticulum (NR), whereas infoldings of both INM and outer nuclear membrane (ONM) are Type-II NR. Here, we report that Kapo si's sarcoma-associated herpesvirus (KSHV) reactivation from latency and lytic cycle progression correlates with increases in both types of NR, but primary envelopment is restricted to peripheral INM and Type-I NR. These Type-I NR structures co-localized with puncta containing CTP:phosphocholine cytidylyltransferase (CCTα), the enzyme that catalyzes the rate-limiting step in phosphatidylcholine (PtdCho) synthesis that drives the de novo membrane biogenesis and membrane curvature required for NR expansion; CCTα recruitment may provide sufficient Type-I NR to facilitate nuclear egress. Despite the concurrent expansion of Type-II NR, primary envelopment involves a mechanism that specifically targets capsids to Type-I NR. Building upon our observation of capsids lacking envelopes in complex higher-order Type-I NR structures, we used polar lipid dyes, CLICK-labeled fluorescent viral genomes, and fluorescent KSHV capsids to track the fate of NR-associated capsids via live cell microscopy. These studies provide evidence for trafficking of NR-associated capsids toward the nuclear periphery and cytoplasm. Taken together, these findings suggest that nuclear egress occurs not only at the nuclear periphery but also at the Type-I NR.IMPORTANCE Herpesvirus capsids assemble in the cell nucleus but are too large to exit via nuclear pores. Instead, they bud into the inner nuclear membrane to acquire a provisional lipid envelope that is shed through fusion with the outer nuclear membrane, delivering the capsid to the cytoplasm for subsequent steps in assembly and egress. These nuclear membranes are dynamic, with the ability to fold into invaginations that access the nuclear interior. Here, we demonstrate that during Kaposi's sarcoma-associ ated herpesvirus (KSHV) replication, nuclear membrane infolding increases, coinciding with recruitment of a host enzyme required for PtdCho synthesis at these sites. We observed accumulation of KSHV capsids at infoldings of the inner nuclear membrane and tracked the association and trafficking of fluorescent viral particles through these structures by live cell microscopy. This complements a more well-established mecha nism of KSHV egress at the nuclear periphery and suggests versatility in nuclear egress mechanisms.
(MCP) arranged into pentons and hexons (3), small capsid protein (SCP) that caps the outer surface of hexons (4), triplex protein complexes that link adjacent capsomers (5), and a dodecameric portal protein complex (6). A-capsids are abortive, with an empty capsid shell devoid of internal scaffolding proteins or a viral genome. B-capsids are assembly intermediates or abortive capsids that contain an inner scaffold consisting of scaffold protein and scaffold protease but no viral genome (7). C-capsids are mature capsids that have ejected scaffolding proteins and replaced them with the linear viral genome. There is evidence that C-capsids gain priority access to the host cell cytoplasm for subsequent steps in assembly and egress, at least in part, through the capsid-vertexspecific component (CVSC), a multiprotein complex enriched on the penton vertices of C-capsids; however, molecular support for this mechanism remains incomplete. These newly assembled capsids are too large to exit the nucleus through nuclear pores and instead bud at the INM into the perinuclear membrane space, acquiring an envelope in the process. This event has been termed "primary envelopment. " Enveloped viruses in the perinuclear space fuse with the outer nuclear membrane (ONM), losing their envelope in the process of gaining access to the cytoplasm.
The generally accepted model of herpesvirus egress is primary envelopment at the peripheral INM, but there is growing evidence for an alternative mechanism involving envelopment at nuclear infoldings (NI) (8)(9)(10)(11). NI share structural similarities with the nucleoplasmic reticulum (NR); "NI" describes any membrane infoldings that extend into the nucleoplasm, whereas "NR" refers to a subset of these structures that meet the definition of nucleoplasmic reticulum (12). Type-I NR features single-membrane branched or unbranched invaginations of INM that extend into the nucleoplasm (Fig. 1). Type-II NR are double-membrane invaginations of both inner and outer NM. As such, Type-II NR contains nuclear pores and nuclear lamina, facilitating an extended cytoplasmic interface with the nucleoplasm, whereas the Type-I NR is lamin-poor and devoid of proteins associated with the ONM. Hybrid structures incorporating both Type-I and Type-II morphology have been reported (e.g., a Type-II invagination that gives rise to multiple Type-I extensions) (12). These hybrid structures contribute to the architectural complexity of the NR.
It has long been recognized that cytomegalovirus (CMV) egress involves primary envelopment at NI. As early as 1964, intranuclear structures resembling NI were reported during murine CMV (MCMV) infection (13). In the years that followed, intranuclear compartments were documented with heightened detail during MCMV infection, and their role in primary envelopment became evident (14). Eventually, intranuclear structures were documented for human CMV (HCMV) as well (15), which were later referred to as "pseudo-inclusions" of the nucleus (16). More recent reports demonstrate that although NI account for a small fraction of nuclear membrane area during CMV infection, most capsids bud at NI rather than the peripheral INM (9). Analysis of HCMV-infected cells using focused ion beam/scanning electron microscopy (FIB/SEM) tomography revealed complex networks of tubules and spherical compartments, with the latter assuming a variety of hierarchical structures consistent with the Type-I NR (10,12). It has been well documented that lytic herpesvirus replication causes the nucleus to increase in size (17). Villinger et al. proposed a "pushing membrane model" that considers changes in nuclear membrane size and the emergence of complex NR networks observed in HCMV-infected cells (10). In this model, disruption of the nuclear lamina is accompanied by the synthesis of new membrane, resulting in the invagination of the INM into the nucleoplasm to form complex membrane structures and providing numerous opportunity sites for HCMV budding. Similar to CMV, nuclear pseudorabies virus (PRV) capsids bud into membranous nuclear structures called "tegusomes" that resemble the NR (8). Tegusomes are sometimes connected with the NE and open into the cytoplasm, and most comprise a single membrane with occasional double-membrane structures. Inspecting TEM images of tegusomes reveals remarkable similarity to the Type-I NR and Type-II NR, as well as complex structures defined more recently by Buser et al. and Villanger et al., which we now know to be the Type-I NR (9,10,18). Together, these studies provide ample evidence for herpesvirus utilization of NI for budding.
Primary envelopment of herpesvirus capsids requires the viral nuclear egress complex (NEC), which consists of a type-II transmembrane protein and a soluble phosphopro tein that assemble into a hexameric lattice structure on the INM to promote local membrane curvature required for capsid budding into the perinuclear space (19). The PRV NEC induces local INM curvature and drives the formation of vesicles embed ded within the Type-I NR (20)(21)(22)(23). Similarly, the NEC of murine gammaherpesvirus-68 (MHV-68) promotes nuclear invaginations bearing lamin A/C (24). Ectopic expression of Kaposi's sarcoma-associated herpesvirus (KSHV) NEC proteins elicited formation of nuclear structures described as "vesicles wrapped by membranous structures in the nucleoplasm that appeared to have no connection to the nuclear membrane" (24). During herpes simplex virus type 1 (HSV-1) infection, the NEC proteins localize to the INM and induce vesicle formation (19). The HSV-1 NEC promotes the reorganization of the lamina, which acts as a structural barrier to capsids accessing the INM, causing lamin-positive protrusions in the intranuclear space that resemble the NR (25). During HSV-1 infection, the viral US3 kinase and a complex comprising the NEC, the viral protein γ 1 34.5, host scaffold protein p32, and protein kinase C, phosphorylate and destabilize the nuclear lamina (26)(27)(28)(29)(30). Similarly, viral serine/threonine kinases from herpes simplex virus type 2 (HSV-2) (31), Epstein-Barr virus (EBV) (32), and HCMV (33,34), all phosphorylate and destabilize the lamina. Thus, lamina disassembly is not only a prerequisite for NR formation but also a conserved feature of herpesvirus replication.
To date, the utilization of NI for gammaherpesvirus primary envelopment remains unresolved. One study of KSHV infection demonstrated the presence of an intranuclear double-membrane structure separated from the nuclear envelope, but the authors did not comment on the nature of this invagination (35). Another study of MHV-68-infec ted cells reported INM invaginations that contained enveloped capsids and secondary compartments within these invaginations that contained non-enveloped capsids (11).
Here, we report that KSHV reactivation from latency and lytic cycle progression correlates with increases in both Type-I NR and Type-II NR, but we only observed primary envelopment at Type-I NR and the peripheral INM. The selective utilization of Type-I NR for primary envelopment despite the concurrent expansion of Type-II NR suggests a mechanism to selectively target KSHV capsids to these membranes for budding. Over a time course of infection, we frequently observed DNA-containing KSHV C-capsids budding into nuclear infoldings contiguous with the INM and sparsely decorated with nuclear lamina, characteristic of Type-I NR. These Type-I NR structures co-localized with puncta containing CTP:phosphocholine cytidylyltransferase (CCTα), the enzyme that catalyzes the rate-limiting step in PtdCho synthesis that drives the de novo membrane biogenesis and membrane curvature required for NR expansion. Thus, CCTα activity may provide sufficient Type-I NR to match requirements for KSHV nuclear egress. Finally, we employed live-cell imaging approaches to track newly assembled KSHV capsids as they transited through NR structures to the cytoplasm. Together, these approaches provide a robust framework to investigate the role of the NR in KSHV nuclear egress.
## RESULTS
## Nuclear infolds increase during KSHV lytic replication
NI described in previous studies of herpesvirus infection have many of the characteris tics of the Type-I NR (9,10,12). To study NI during latent and lytic phases of KSHV infection, we used a well-established model system consisting of inducible SLK (iSLK) cells latently infected with KSHV BAC16 strain (iSLK-BAC16 cells); treatment of these cells with doxycycline (dox) and sodium butyrate stimulates ectopic expression of the viral replication and transcription activator (RTA) lytic switch protein and reactivation from latency (36). Using reverse transcription quantitative PCR (RT-qPCR), we characterized the progression of lytic replication in iSLK-BAC16 cells by measuring the accumulation of representative gene products from different stages of the lytic replication cycle, including transcripts from immediate early gene RTA, early gene ORF57, and late gene ORF65 (Fig. 2A). iSLK-BAC16 cells have two sources of RTA: the dox-inducible RTA from a transgene stably integrated into the cellular genome and native RTA (nRTA) derived from the viral genome; we used oligonucleotide primers specific for nRTA to selectively measure RTA transcript produced by the viral genome. nRTA transcript levels increased steadily throughout the 72 hour (h) time course, demonstrating reactivation from latency in this model (Fig. 2A). RTA promotes the transcription of early genes, including ORF57; we observed maximal ORF57 transcript levels at 24 h post-reactivation. As expected, the levels of the ORF65 transcript that encodes the small capsid protein peaked at 72 h post-reactivation, marking progression to late lytic replication (Fig. 2A). Viral genome replication is required to license late gene expression (37). We used a qPCR assay to demonstrate that genome replication sharply increased between 24 h and 48 h with little additional increase thereafter (Fig. 2B). The production of infectious KSHV virions was measured by collecting cell supernatants over time and titering them on naïve HEK 293 A cell monolayers and enumerating infected cells expressing the GFP reporter from the BAC16 genome. We observed that while infectious virions were produced at 24 h and 48 h post-reactivation, production peaked at 72 h post-reactivation and increased very little thereafter (Fig. 2C). Based on this finding, we selected the 72 h time point as the optimal window for analysis of KSHV nuclear egress. Transmission electron microscopy (TEM) of iSLK-BAC16 cells harvested at 72 h post-reactivation revealed the presence of NI that appear as large spheroid com partments containing enveloped C-capsids (Fig. 3A). We applied a previously estab lished hierarchical system of NI categorization (10) to our studies of KSHV infection, whereby first-order NI contain C-capsids within a single membrane compartment and second-order NI have non-enveloped C-capsid-containing periplasm-derived vesicles within a single membrane compartment. As previously reported (10), our findings suggest that the lumen of the first-order NI is continuous with the perinuclear space. The lumen of the second-order NI is reminiscent of the nucleoplasm and hypothesized to originate from invaginations of nucleoplasm into first-order NI (10). Third-order infoldings originate from separate compartments within second-order infoldings and have a lumen that has an electron density resembling that of the perinuclear space in our studies and others (10). However, the origin of these compartments remains speculative, and further confirmatory studies are required. We also describe a novel "convoluted membrane" NI, which appeared as a single membrane containing multiple encapsulated vesicles or membrane whorls.
To visualize nuclear envelope changes over the course of viral replication, iSLK-BAC16 cells were reactivated from latency and harvested for TEM at 24, 48, 72, 96, or 120 h post-reactivation (Fig. 3B). During latency, the nuclear envelope is smooth with evenly distributed lamin (Fig. 3B, first row), and only a few small NI are observed, indicating low basal NI activity, as previously reported in other human cell lines (38,39). By 24 h post-reactivation (early lytic replication), we observed increases in NI, appearing as short tubules radiating from the nuclear periphery that are sometimes linked to spherical 5). Second-order NI appear as a single membrane compartment that contains one or more additional single membrane compartments (example image reproduced from Fig. 3B). Third-order NI are second-order NI that contain additional compartments (example image reproduced from Fig. 6). Convoluted membranes often appear as a second-order infolds, but the second compartment is a large, convoluted membrane whorl (example image reproduced from Fig. 5). (B) Transmission electron microscopy (TEM) analysis of nuclear envelope remodeling in doxycycline-inducible iSLK cells infected with KSHV BAC16. Cells were reactivated with 1 µg/mL doxycycline and 1 mM sodium butyrate and harvested at 24, 48, 72, 96, and 120 h post-reactiva tion. Four representative images were selected per timepoint. White arrows mark NI. Black, blue, green, and yellow arrows specifically mark first, second, third, and convoluted membrane infolds, respectively. Nucleus (Nuc), cytoplasm (Cyto), and nuclear membranes (NM) are labeled. Representative images are from N = 3 biological replicates. The complete set of original electron micrographs is available via the Dryad open data publishing platform at the following link: https://doi.org/10.5061/dryad.qbkh18v9.
Full-Length Text compartments (Fig. 3B, second row). By 48 h post-reactivation (late lytic replication), NI expand into larger spherical vacuoles connected to the INM and often form networks (Fig. 3B, third row). C-capsids enter first-order NI, acquiring a primary envelope, whereas some naked capsids are located in second-order NI; more complex third-order NI are also evident. By 72 h post-reactivation, NI increase in abundance and form large compart ments containing numerous enveloped C-capsids (Fig. 3B, fourth row). At this time, NI appear as first, second, third order, and convoluted membrane morphologies. By 96 h post-reactivation, NI are largely devoid of viral capsids and have an altered elongated appearance featuring empty first-and second-order infolds (Fig. 3B, fifth row). A few residual C-capsids can be observed in the nucleus at 96 h post-reactivation, but most are found in the cytoplasm or at the cell surface. Similarly, at 120 h post-reactivation, empty NI continue to be observed, suggesting that these compartments persist long past their peak association with viral capsids (Fig. 3B, sixth row). Table 1 provides quantitative data from all electron micrographs included in this study, enumerating capsid types and their subcellular localization at different times post-reactivation, including information about their association with various types of NI.
During peak activity at 72 h post-reactivation, NI-containing enveloped KSHV C-capsids were often observed within single membrane compartments with first-order NI morphology (Fig. 3B, black arrows). However, C-capsids were also observed budding into single membrane structures surrounding a multi-membrane whorl (Fig. 3B, yellow arrows). In addition, a subset of NI appeared to have a second-order infold structure, and vesicles or tubules were seen inside a compartment akin to first-order NI (Fig. 3B, blue arrows). Finally, although we captured significantly more capsids budding into NI, we still observed "classical" budding at the peripheral NM at 72 h post-reactivation (Fig. 4, white arrows). For example, we observed a cell that contained three classical budding events at the peripheral NM, alongside a complex NI that contained multiple enveloped C-capsids within a first-order NI (Fig. 4 panel iv, black arrow). Taken together, these observations indicate that KSHV C-capsids are associated with diverse NI structures, and envelopment at the peripheral NM and envelopment at NI can take place concurrently in an infected cell.
## KSHV buds into Type-I nucleoplasmic reticulum
We observed accumulation of enveloped KSHV C-capsids in NI that resemble the Type-I NR. However, definitive identification of these compartments as Type-I NR requires evidence of their direct connection to the INM. At 72 h post-reactivation, we observed enveloped KSHV C-capsids within NI that were clearly connected to the INM, but not the ONM, by a tubule (Fig. 5, white arrows; blue lines trace continuity with the INM). An electron-dense layer resembling the nuclear lamina extends along the neck of the Type-I NR but appears largely absent around spheroid compartments containing enveloped C-capsids (Fig. 5, white dashed lines). In some cases, this dense layer is challenging to distinguish from peripheral heterochromatin, and we therefore acknowledge the possibility that, in certain images, the observed density may include contributions from both the lamina and adjacent chromatin. Our observations are consistent with estab lished literature describing Type-I NR, which typically feature a small amount of nuclear lamina limited to regions adjacent to the peripheral INM (12,40). These observations are also consistent with the lamin-poor Type-I NR observed during MCMV infection (9). By contrast, even though Type-II NR also expands during KSHV lytic replication, and we observed C-capsids adjacent to Type-II NR, we have never observed budding events at these structures.
During HCMV infection, Type-I NR is a complex consisting of a hierarchy of compart ments interconnected by tubules (10). Similarly, we documented numerous instances where spherical compartments containing enveloped C-capsids were connected to other spherical compartments by a tube-shaped "neck, " resulting in the formation of a network (Fig. 6). We frequently observed C-capsids budding into 1st order NI surrounding, and/or sometimes appearing to be contiguous with, convoluted membranes, but C-capsid access to the convoluted membrane interior was exceedingly rare, suggesting a potential barrier in capsid access to convoluted membrane structures.
Taken together, our observations indicate that NI associated with KSHV C-capsids during infection are lamin-poor and continuous with the INM, meeting the definition of the Type-I NR (12). Similar to HCMV infection (10), the Type-I NR is remodeled into membranous networks during KSHV infection, characterized by multiple spherical compartments connected by tubule necks to the INM.
## Nucleoplasmic reticulum expansion in KSHV-infected cells correlates with recruitment of CTP:phosphocholine cytidylyltransferase and increased intranuclear VAPA
We next conducted immunofluorescence microscopy studies on KSHV-infected cells to determine the subcellular localization of key proteins involved in NR expansion, lamin A/C, VAPA, and CCTα (41). VAPA engages oxysterol-binding protein-related protein 3 and Rab7 to form the VOR complex responsible for the transfer of extracellular vesicle CCTα is activated by insertion of a lipid-sensing amphipathic helix into the INM, where it catalyzes the rate-limiting step for PtdCho synthesis required for de novo membrane biogenesis (42). CCTα also aids in NR expansion by promoting membrane curvature in collaboration with a lamin A/B1 scaffold (Fig. 1). Since the ER is continuous with the ONM, VAPA marks Type-II NR, whereas CCTα is associated with both Type-I and II NR.
In latently infected cells, lamin A/C is uniformly distributed across the INM and intensified at the nuclear envelope edges (Fig. 7A andB). By 72 h post-reactivation, lamin A/C projections radiate from the INM, forming spherical compartments, consis tent with NR expansion reported in previous studies of the NR (43,44) and lamin reorganization during HCMV, EBV, HSV-1, and HSV-2 infection (31,32,45,46). By 96 h, lamin A/C-positive tubules emerged from the INM, exhibiting a branched, cylindrical morphology and linking multiple NR compartments. One limitation of this lamin A/C immunostaining procedure is that it does not detect lamin-poor NR compartments that were readily detected by TEM; instead, these immunostaining images show lamin A/C-positive tubules lacking a terminal compartment, which is likely undetectable due to lamin deficiency.
CCTα staining in latently KSHV-infected cells was uniformly distributed in the nucleoplasm, with only occasional puncta, suggesting that CCTα primarily remains in its inactive state (Fig. 7A). By 72 and 96 h post-lytic reactivation, CCTα puncta were prominent, indicating activation at membranes within the nucleoplasm. These CCTα puncta partially overlapped with lamin A/C-positive structures, suggesting that CCTα localizes to compartments with and without lamin A/C. VAPA staining in latent cells
## Selective budding of C-capsids into Type-I nucleoplasmic reticulum
Our frequent observation of KSHV C-capsids near and within Type-I NR motivated further analysis of selective capsid recruitment and budding at these structures. Indeed, we captured events of capsid budding into Type-I NR where the surrounding membrane appeared to conform tightly around this capsid, consistent with membrane curvature changes associated with capsid envelopment (Fig. 8A). Furthermore, in nuclei containing an assortment of capsid types, we observed that C-capsids were associated with Type-I NR structures, whereas B-capsids and A-capsids were largely restricted to the nucleo plasm (Fig. 8B). We also observed complex Type-I NR structures bearing enveloped C-capsids, as well as naked C-capsids in the interior of second-order Type-I NR. These observations suggest that C-capsids can accumulate at invaginations of the nucleoplasm into the Type-I NR. Capsids from all TEM images in this study were counted and their abundance in different regions of the infected cell was recorded as follows: (i) capsids in the NR (further divided into capsids present in first-order infolds and second-order infolds), (ii) capsids in the perinuclear space, (iii) total capsids in the nucleus (including all capsids associated with the NR and peripheral NE, and (iv) total capsids in the cytoplasm (Table 1). At 48 h post-reactivation, 18% of nuclear capsids were associated with the NR compartment, increasing to a peak of 30% by 72 h post-reactivation. Most NR-associated capsids at all timepoints were localized to 1st order infolds, suggesting preferential capsid accumulation at these structures. By 96 h post-reactivation, the proportion of NR-associated capsids declined to 3% and most C-capsids were found in the cytoplasm, leaving behind a greater proportion of nuclear A-and B-capsids. These findings provide evidence for a primary envelopment pathway that specifically selects for C-capsids, where the hierarchy of Type-I NR architecture plays a role in capsid transit. Importantly, the cells analyzed were enriched for prominent NR compartments and therefore do not represent their frequency across the total lytic population.
## Dynamic labeling of viral genomes using CLICK chemistry enables capsid tracking at the nucleoplasmic reticulum
Despite evidence that KSHV C-capsids can undergo primary envelopment and perhaps de-envelopment at Type-I NR and accumulate in the cytoplasm over time, it remains difficult to determine the fate of capsids by inspecting static electron micrographs. Nevertheless, we did observe capsids in the narrow "neck" of Type-I NR structures and congregating in the space between the ONM and expanded INM (Fig. 9A, black arrows). We reasoned that labeling of viral genomes prior to their loading into nascent capsids could enable tracking of capsid nuclear egress. To monitor capsid trafficking in NR compartments, we conducted live-cell imaging experiments using the green fluorescent dye 3,3′-Dihexyloxacarbocyanine Iodide (DiOC 6 ) to label polar lipids, including NR, and inverse electron demand Diels-Alder (IEDDA) CLICK chemistry to label viral genomes (47). We pulsed labeled newly synthesized viral DNA with the synthetic nucleotide 5-vinyl-2'deoxyuridine (VdU), followed by treatment with acridine tetrazine (PINK) (48); a similar approach was recently used to study replication compartments during adenovirus and HSV-1 infection (49). iSLK-BAC16 cells stably expressing blue fluorescent protein (BFP) were seeded on coverslips, reactivated from latency, and pulsed with VdU at 40 h post-reactivation. Cells were subsequently fixed, incubated with PINK, and processed for immunofluorescence imaging at 72 h or 96 h post-reactivation. In latently infected cells, nuclear DNA labeling was uniformly distributed, consistent with VdU incorporation into viral DNA as well as host DNA during S-phase. Because KSHV lytic replication triggers host cell cycle arrest (50), these cells exhibit reduced DNA labeling overall and the emergence of nuclear puncta representing newly synthesized viral DNA (Fig. 9B). These red fluorescent puncta co-localized with KSHV ORF65 (small capsid protein), confirming their identity as viral genomes that were subsequently packaged into capsids (Fig. 9C). At 72 h post-reactivation, puncta containing viral DNA and ORF65 were often clustered near DiOC 6 -labeled invaginations, confirming capsid association with the NR as we observed previously by electron microscopy. By 96 h post-reactivation, these capsid puncta largely accumulated in the cytosol, suggesting that newly assembled capsids containing labeled viral genomes were competent for nuclear egress. We next used live-cell microscopy to track the fate of NR-associated CLICK-labeled viral genomes in real time. iSLK-BAC16 cells were reactivated from latency and pulsed with VdU at 40 h post-reactivation, followed by a PINK pulse from 44 to 48 h post-reacti vation. DiOC 6 staining allowed us to visualize the expanded NR in live cells, highlighting brightly stained compartments connected to the INM by a hollow "neck" (Fig. 9D). In the cell nucleus, DiOC 6 -positive NR co-localized with large red fluorescent clusters of newly synthesized DNA (Fig. 9D, Movies S1 and S2). In real time, we observed smaller DNA foci appear to emerge from the densely packed compartment and transit an NR neck to the peripheral NE (Fig. 9D, 1: 32 and 2:19 minutes). In the next frames, we observed the positive curvature of the NE and arrival of viral DNA puncta in the cytoplasm (Fig. 9D,3:05 and 3:52 minutes). Thus, these images show that capsid transit from large NR compartments to the cytoplasm can occur in a few minutes. Increasing the contrast of the DiOC 6 -stained NR neck revealed intricate spherical patterns along the NR neck, resembling primary envelopes in both size (~300 nm) and shape. Based on this data, we suggest that capsids with labeled viral genomes can transit from NR compartments to the cytoplasm.
## Fluorescent KSHV capsid transit through nucleoplasmic reticulum
To further investigate the role of the NR in KSHV nuclear egress, we performed live-cell imaging using a dual-color reporter KSHV (51). This BAC16-ORF65_mScarlet-gM_mNeon virus (hereafter RG-BAC16) encodes fluorescent capsid (ORF65_mScarlet) and envelope (gM_mNeon) proteins, enabling real-time tracking of capsid assembly and nuclear egress (Fig. 10A, bottom right cartoon). At different times post-reactivation, iSLK cells infec ted with RG-BAC16 were stained with anti-lamin B receptor (LBR) antibody to mark both Type-I NR and Type-II NR structures (Fig. 10A). At 72 h post-reactivation, bright ORF65_mScarlet puncta appear in the nucleus, alongside more diffuse cytoplasmic staining, likely reflecting newly synthesized ORF65_mScarlet that has yet to be impor ted to the nucleus. ORF65_mScarlet puncta were often observed in association with LBR-positive NR that appeared as spherical compartments connected to the NE by hollow "necks. " By 96 h post-reactivation, cytoplasmic ORF65_mScarlet puncta increased and could be seen in proximity to gM_mNeon, although some ORF65_mScarlet signal remained associated with the NR and peripheral nuclear membrane. These images suggest that ORF65-mScarlet reporter protein can be used to monitor trafficking of newly assembled KSHV capsids.
To visualize trafficking of ORF65_mScarlet capsids in real time, iSLK cells infected with RG-BAC16 were seeded in chamber well slides and reactivated from latency by addition of 1 µg/mL doxycycline and 1 mM sodium butyrate. At 72 h post-reactivation, cells were stained with DiOC 6 to label NR structures, and Z-stack images were taken on a spinning disc microscope at 18 second intervals (Fig. 10B). Distinct DiOC 6 -positive intranuclear compartments were observed around the nuclear periphery, showcasing the expanded NR, in association with ORF65_mScarlet-positive puncta (Fig. 10B, blue boxes; Movies S3 and S4). These NR-associated ORF65_mScarlet puncta moved toward the nuclear periphery over time (Fig. 10B, time 0:00 and 0:18 minutes, blue box) and accessed the cytoplasm (Fig. 10, 0:36 minutes, blue box). Even though precise mechanisms of KSHV nuclear egress via NR structures remain obscure, these movies and static images support the idea that NR-associated capsids can transit to the cytoplasm.
## DISCUSSION
During herpesvirus primary envelopment, capsids bud into INM with the help of a two-component viral NEC, briefly acquiring an envelope that is shed via fusion with the ONM. Upon gaining access to the cytoplasm, these capsids bud into the trans Golgi network and travel to the cell surface in secretory vesicles that fuse with the plasma membrane to release progeny herpesviruses into the extracellular environment. This primary envelopment event at the INM has been observed at the nuclear periphery, but also at NI that provide a tantalizing route for acquiring newly assembled capsids from viral replication compartments. However, mechanisms governing herpesvirus access to NI remain poorly characterized. Here, we report that both Type-I and Type-II NR increase during KSHV lytic replication, correlating with the recruitment of CCTα to these membranes, which drives membrane proliferation via PtdCho synthesis. There is emerging evidence that herpesvirus nuclear egress operates with a quality-control mechanism wherein intact DNA-containing C-capsids are selectively exported, leaving behind immature or defective capsids. Our study reveals another aspect of selectivity in KSHV nuclear egress; even though both Type-I NR and Type-II NR increase during KSHV lytic replication, we only observe budding into Type-I NR. This supports a mechanism of KSHV primary envelopment that complements the canonical model, whereby laminpoor Type-I NR membranes that probe the nuclear interior provide a site for primary envelopment of newly synthesized KSHV capsids.
Inspecting a large collection of electron micrographs allowed us to observe KSHV budding at the Type-I NR and the peripheral INM in the same nucleus, suggesting that both mechanisms operate concurrently. These observations are consistent with previous studies of human betaherpesviruses and murine gammaherpesviruses and suggest versatility in herpesvirus nuclear egress pathways. However, we do not yet understand the factors that influence which route of primary envelopment a newly assembled KSHV capsid pursues, or their relative rates of usage. Furthermore, the factors that limit capsid budding into Type-II NR remain obscure. These undetermined mechanisms represent attractive targets for future research.
In addition to broadly promoting NR proliferation, KSHV lytic replication increased the formation of complex higher-order Type-I NR structures. We observed a striking accumulation of naked C-capsids in second-order infoldings of Type-I NR, suggesting a role for these complex structures in KSHV nuclear egress. However, the mechanism of capsid recruitment to second-order infoldings remains unknown. We also frequently observed greatly expanded Type-I NR-derived convoluted membranes, which may be the product of runaway INM expansion that causes the membrane to fold into a whorl. We speculate that CCTα drives INM expansion and convoluted membrane formation through its ability to dimerize in trans, linking and stabilizing opposing membranes while catalyzing the rate-limiting step in PtdCho synthesis (52). In this way, CCTα could promote unchecked INM expansion and formation of convoluted structures. After inspecting hundreds of electron micrographs, we were unable to identify KSHV capsids associated with these NR-derived convoluted membranes, suggesting that these membranes are not targets for primary envelopment.
To further advance understanding of KSHV nuclear egress mechanisms, we conduc ted live cell microscopy experiments to monitor capsid egress at the NR in real time. We employed fluorescently labeled KSHV genomes and capsids, and a fluorescent polar lipid dye, to monitor association of capsids with NR structures and subsequent trafficking events. After validating that fluorescent CLICK-labeled viral genomes co-localize with ORF65 small capsid protein in KSHV-capsid-sized puncta by immunofluorescence microscopy on fixed cells, we tracked their association with NR structures using the fluorescent polar lipid dye DiOC 6 by spinning disc fluorescence microscopy. In parallel, we performed live-cell imaging of ORF65_mScarlet capsids in DiOC 6 -stained cells. These orthogonal approaches provide preliminary evidence that KSHV capsids associated with NR compartments can rapidly transit to the cytoplasm in minutes. While substantially more work is required to fully elucidate KSHV nuclear egress mechanisms at the NR, our study provides a roadmap towards development of more sensitive assays, combining approaches in labeling of viral genomes and capsids, as well as NR membranes. Future work should investigate the molecular machinery involved in remodeling these NR compartments, including whether the KSHV NEC is recruited to NR budding sites.
## MATERIALS AND METHODS
## Cell culture and reagents
All doxycycline-inducible iSLK-BAC16 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 1% penicillin-streptomycin, and grown in 5% CO 2 at 37°C. To maintain stability of episomal KSHV DNA in iSLK cells, parental cell cultures were maintained in 1 µg/mL puromycin (ThermoFisher) and 1 mg/mL hygromycin B (Invitrogen). iSLK-BAC16 cells were reactivated from latency with the addition of 1 mM sodium butyrate (Sigma) and 1 µg/mL doxycycline (Sigma).
## Transmission electron microscopy
iSLK-BAC16 cells were seeded in 10 cm dishes (VWR) at a density of approximately 1.5 × 10 5 cells per dish (5% CO 2 at 37°C). The next day, cells were reactivated from latency by the addition of 1 mM sodium butyrate and 1 µg/mL doxycycline and harvested at 24, 48, 72, or 96 h. Cells were harvested by trypsin digestion and centrifuged at 250 × g for 5 minutes. Samples were fixed for a minimum of 2 hours in 2.5% glutaraldehyde diluted in 0.1 M sodium cacodylate buffer. Following fixation, samples were rinsed three times for at least 10 minutes each with 0.1 M sodium cacodylate buffer. Secondary fixation was performed using 1% osmium tetroxide for 2 hours, followed by a brief rinse with distilled water. Samples were then placed in 0.25% uranyl acetate at 4°C overnight. Dehydration was carried out using a graded acetone series: 50% acetone for 10 minutes, 70% acetone for two 10 minute incubations, 95% acetone for two 10 minute incubations, and 100% acetone for two 10 minute incubations, followed by a final 10 minute incubation in dried 100% acetone. Samples were then infiltrated with Epon Araldite resin using a stepwise approach: 3:1 (dried 100% acetone to resin) for 3 hours, 1:3 (dried 100% acetone to resin) overnight, and finally 100% resin for two 3 hour incubations. Samples were embedded in 100% Epon Araldite resin and cured at 60°C for 48 hours. Ultrathin sections (~100 nm thick) were obtained using a Reichert-Jung Ultracut E ultramicrotome equipped with a diamond knife and were placed on 300-mesh copper grids. Sections were stained with 2% aqueous uranyl acetate for 10 minutes, followed by two 5 minute rinses with distilled water. Lead citrate staining was performed for 4 minutes, followed by a quick rinse with distilled water, and the grids were air-dried. Prepared samples were imaged using a JEOL JEM 1230 transmission electron microscope operating at 80 kV. Images were acquired using a Hamamatsu ORCA-HR digital camera. All TEM experiments were performed in three biological replicates.
## Immunofluorescence and image processing
Doxycycline-inducible iSLK-BAC16 cells were seeded at 2 × 10 5 cells/mL on glass coverslips (Paul Marienfeld Gmb H & Co.; 0117580) in 12-well plates (VWR). The next day, cells were reactivated with 1 µg/mL doxycycline and 1 mM sodium butyrate and harvested at 72 h or 96 h. Coverslips were fixed in 4% paraformaldehyde for 15 min at room temperature, washed with PBS, and incubated in blocking/permeabilization (block/perm) buffer (Triton-X100 0.1% (vol/vol) (Sigma; 1002614889) and 1% human AB serum (Sigma) in PBS for 1 hour at RT. Coverslips were incubated with mouse anti-Lamin A/C (1:200, Sigma) and/or rabbit anti-CCTα (1:500) and/or rabbit anti-VAPA (1:200) antibodies overnight at 4°C in block/perm buffer. Experiments using iSLK-ORF65_mScar let-ORF39_mNeon cells received rabbit polyclonal lamin B receptor (Proteintech Group Inc.; 12398-1-AP) diluted at 1:300. The next day, coverslips were washed three times with PBS followed by a 1 h incubation at room temperature with secondary donkey anti-rab bit AlexaFluor555 (Invitrogen; A31572) and chicken anti-mouse AlexaFluor647 (Invitro gen; A21463) for CCTα and LBR immunofluorescence experiments, and goat anti-mouse AlexaFluor555 (Invitrogen; A21422) and goat anti-rabbit AlexaFluor647 (Invitrogen; A21244) for VAPA immunofluorescence. All secondary antibodies were diluted 1:1,000 in block/perm buffer. After three additional PBS washes, coverslips were counterstained with Hoechst 33342 (1:5,000, 5 min, RT) and mounted on microscope slides (Fisherbrand; 12-550-15) using ProLong Gold Antifade (ThermoFisher, P36930).
## CLICK chemistry
iSLK-BAC16 cells were seeded at a density of 2 × 10⁵ cells per well onto either four-well chamber slides (Nunc Lab-Tek chambered cover glass; 155383) for live-cell imaging or glass coverslips for fixed-cell experiments and reactivated from latency as described above. For both conditions, cells were pulsed with 100 µM 5-vinyl-2′-deoxyuridine (VdU; Lumiprobe B2540) at 40 hours post-reactivation. For live-cell imaging, PINK was added at 44 hours post-reactivation to a final concentration of 100 µM and removed at 48 hours, followed by three washes with warm complete DMEM. Chamber wells were imaged 72 hours post-reactivation as described below. For fixed-cell imaging, coverslips were harvested at 72-and 96 hours post-reactivation, washed with PBS, and fixed in 4% paraformaldehyde as described above. Fixed cells were incubated overnight with a mouse monoclonal antibody against ORF65 (gift from S.-J. Gao) diluted 1:600 in blocking buffer. The following day, cells were incubated with chicken anti-mouse Alexa Fluor 647 secondary antibody (Invitrogen; A21463) for 1 hour, then washed three times with PBS. Coverslips were then inverted onto 40 µL droplets of 10 µM PINK in PBS within a humidified plastic container (moistened with a wet paper towel and sealed with Parafilm) and incubated for 4 hours at 37 °C with 5% CO₂. Coverslips were washed three times with PBS and incubated with DiOC₆ (final concentration 0.2 µL/mL in PBS) for 5 minutes at room temperature, followed by three PBS washes. Coverslips were then mounted onto microscope slides as described above. Live-cell CLICK chemistry experiments were successful using both PINK 1.0 and PINK 2.0 probes with comparable results.
## Spinning disc live-cell imaging
iSLK-BAC16 cells were seeded in four-well chamber slides (Nunc Lab-Tek; 155383) at a density of 2 × 10⁵ cells per well. The following day, cells were reactivated from latency with 1 µg/mL doxycycline and 1 mM sodium butyrate. CLICK chemistry experiments were pulsed with VdU and PINK as described above. At 72 hours post-reactivation, cells were stained with 10 µM BDP 630/650 lipid dye (Lumiprobe 1233) in DMEM without phenol red for 15 minutes, washed three times, and imaged in DMEM with out phenol red, supplemented with 5% FBS. For dual-labeled virus experiments using iSLK-ORF65_mScarlet-ORF39_mNeon cells, reactivation and BDP 630/650 staining were performed as described above. Cells were imaged using a Zeiss Cell Observer spinningdisk confocal microscope with a 100 × oil immersion objective. Z-stacks ranged from 7 to 11 slices per stack, adjusted based on cell size and experiment duration. For extended imaging (>30 min), seven slices were used to minimize laser exposure and preserve cell health. Laser intensities were kept at the lowest settings to reduce phototoxicity. Imaging intervals varied depending on z-stack thickness, typically ranging from 40 seconds to 1.5 minutes (exact intervals noted in figure captions). Z-stacks were processed in Imaris for brightness and contrast enhancement, and individual timestamped images were exported using the Imaris capture tool for figures and videos.
## De novo infection and titering
iSLK-BAC16 cells were seeded in six-well plates (VWR) at a density of approximately 4 × 10 5 cells per well (5% CO 2 at 37°C). The next day, cells were reactivated from latency by the addition of 1 mM sodium butyrate and 1 µg/mL doxycycline and harvested at 24, 48, 72 h, 96 h, or 120 h. Virus-containing supernatant was harvested from iSLK-BAC16 cells at the indicated times by pelleting cellular debris at 3,300 × g for 5 minutes and then stored at -80°C until ready to titer the virus. To titer the virus, naïve 293 A cells were seeded at 150,000 cells/well in a 96-well plate. 5 µL of virus-containing supernatant was added to each well in technical duplicates. The plate was centrifuged at 800 × g for 2 h at 30°C. 24 h after infection, plates were fixed in 4%paraformaldehyde and counterstained with Hoechst 33342 (1:5,000, 5 min, RT) in block/perm buffer (see above). Infected cells were imaged using a Zeiss Axio Imager Z2 epifluorescence microscope equipped with a 20× objective. The microscope was calibrated to automatically image three randomly selected fields of view per well. Image processing and quantification were performed using Imaris software. Masks were generated for the Hoechst (nuclei) and GFP (infected cells) channels, and the number of nuclei overlapping with GFP-positive signals was quantified automatically. All images were reviewed manually to exclude artifacts such as monolayer tears or debris. For each biological replicate (n = 3), the average number of cells showing overlap between green and blue signals was calculated from technical duplicates.
## Quantitative reverse-transcription PCR
iSLK-BAC16 cells were seeded in six-well plates (VWR) at a density of approximately 400,000 cells per well (5% CO 2 at 37°C). The next day, cells were reactivated from latency by the addition of 1 mM sodium butyrate and 1 µg/mL doxycycline and harvested at 24, 48, or 72 h. RNA was isolated from cells with the RNeasy Plus Kit (Qiagen), and 500 ng total RNA was reverse transcribed with the Maxima H Minus First Strand cDNA Synthesis Kit (Thermo Cat: K1652). For optimal first-strand synthesis, primer volume was made up from half Oligo(dT)18 and half Random hexamer primers. A CFX96 Touch Real-Time PCR Detection System (Bio-Rad) and Luna Universal qPCR Master Mix (NEB cat: M3003L) were used to perform Real-Time PCR. Changes in mRNA levels were calculated by the ΔCt method and normalized using 18S rRNA as a reference gene. Four biological replicates were analyzed, each processed as two technical replicates. The following primer sets were used in this assay:
18S F: 5′-TTCGAACGTCTGCCCTATCAA-3′; R: 5′-GATGTGGTAGCCGTTTCTCAGG-3′ nRTA F: 5′-TCCAGTTTTGCTCCCCACTG -3′; R: 5′-TTCTGCCGTATTGTAGGCGG-3′ ORF57: 5′-TCCAGTTTTGCTCCCCACTG-3′ R: 5′-TTCTGCCGTATTGTAGGCGG-3′ SOX: 5′-ACCACGGAGTCTGACGTCTA-3′ R: 5′-ACGATCGAACTCTGCAGCAA-3′ vGPCR: 5′-GTACTGACATCCGCTGCACT-3′ R: 5′-TCATGTTTCCCGCGTTCTCA-3′ K2: F: 5′-TCTCTTGCTGGTCCGGTTCAC-3′ R: 5′-CGGTACGGTTAACAGAGGTCG-3′ ORF65: F: 5′-TGGCTCGCATGAATACCCTG-3′ R: 5′-CTGCAGATGATCCCCGCTTT-3′
Nuclear-associated episome quantification iSLK-BAC16 cells were seeded in six-well plates (VWR) at a density of approximately 400,000 cells per well (5% CO 2 at 37°C). The next day, cells were reactivated from latency by the addition of 1 mM sodium butyrate and 1 µg/mL doxycycline and harvested at 24 h, 48 h, 72 h, 96 h, or 120 h. DNA was harvested from cells using 200 µL/well of virus lysis buffer (20 mM Tris-HCl, pH 7.4, 300 mM NaCl, and 2.5% NP-40). Samples were sonicated 10 seconds to break nuclear membranes, diluted with 200 µL of dH 2 O to create a stock, and 25 µL of that stock was further diluted with 975 µL of dH 2 O to create a working stock for use in qPCR and stored at -80°C. Prior to qPCR, samples were spun down at 3,300 × g for 5 minutes, and only the top portion of the sample was used. qPCR (or qPCR) was carried out as described above with primers specific to KSHV ORF26 and β-actin.
Changes in KSHV genome copy number were calculated by the ∆∆Ct method relative to latent cells and normalized to β-actin. Three biological replicates were analyzed, each processed as two technical replicates. The following primer sets were used in this assay: ORF26: 5′-CAGTTGAGCGTCCCAGATGA-3′R: 5′-GGAATACCAACAGGAGGCCG-3′ β-actin: 5′-CTTCCAGCAGATGTGTGATCA-3′R: 5′-AAAGCCATGCCAATCTCATC-3′
## Graphing and statistical analysis
GraphPad Prism Version 10.4.1 was used to generate graphs and complete statistical analysis. One-way ANOVA comparing means between groups (i.e., latent vs. 24 h, latent vs. 48 h, etc.) with a Tukey multiple comparison test was used to compare differences in gene expression between groups. P-values < 0.05 were considered significant and denoted as the following: <0.05 (*), <0.01 (**), <0.001 (***), <0.0001 (****), and > 0.05 was denoted as not significant (ns).
## Imaging and data processing
Z-stacks were acquired using a Zeiss LSM 880 fluorescence microscope (100× oil objective) and processed into maximum intensity projections using Zeiss Black software. Images were further analyzed in FIJI (ImageJ), with a global contrast increase of 0.34% for VAPA experiments. Notably, 96 h samples exhibited intense CCTα staining, suggesting upregulation or aggregation. To compensate, the A568 laser power was reduced from 4 to ~ 2.6 when imaging 96 h samples. Figure 3A depicts the diversity of NI observed in KSHV-infected cells. Each cartoon is paired with a representative TEM image from the published data set to convey a clear understanding of the appearance of these structures; these include a first-order NI from Fig. 5, a second-order NI from Fig. 3B, a third-order NI from Fig. 6, and a convoluted membrane structure from Fig. 5. The legend associated with Fig. 5 is designed to help the reader inspect TEM images and identify key structures in the raw images and marked-up images, including envelope, genome, capsid, ONM, INM, lamina, and absence of lamina. The images associated with the legend were reproduced from our primary image data set and include a capsid presented in Fig. 8.
## References
1. Tandon, Mocarski, Conway (2015) "The A, B, Cs of herpesvirus capsids"
2. Nealon, Newcomb, Pray et al. (2001) "Lytic replication of Kaposi's sarcoma-associated herpesvirus results in the formation of multiple capsid species"
3. Costa, Cohen, Eisenberg et al. (1984) "Direct demonstration that the abundant 6-kilobase herpes simplex virus type 1 mRNA mapping between 0.23 and 0.27 map units encodes the major capsid protein VP5" *J Virol*
4. Booy, Trus, Newcomb et al. (1994) "Finding a needle in a haystack: detection of a small protein (the 12-kDa VP26) in a large complex (the 200-MDa capsid of herpes simplex virus)" *Proc Natl Acad Sci*
5. Okoye, Sexton, Huang et al. (2006) "Functional analysis of the triplex proteins (VP19C and VP23) of herpes simplex virus type 1" *J Virol*
6. Newcomb, Juhas, Thomsen et al. (2001) "The UL6 gene product forms the portal for entry of DNA into the herpes simplex virus capsid" *J Virol*
7. Preston, Kobaisi, Mcdougall et al. (1994) "The herpes simplex virus gene UL26 proteinase in the presence of the UL26.5 gene product promotes the formation of scaffold-like structures" *J Gen Virol*
8. Roffman, Albert, Goff et al. (1990) "Putative site for the acquisition of human herpesvirus 6 virion tegument" *J Virol*
9. Buser, Walther, Mertens et al. (2007) "Cytomegalovirus primary envelopment occurs at large infoldings of the inner nuclear membrane" *J Virol*
10. Villinger, Neusser, Kranz et al. (2015) "3D analysis of HCMV induced-nuclear membrane structures by FIB/SEM tomography: insight into an unprecedented membrane morphology" *Viruses*
11. Peng, Ryazantsev, Sun et al. (2010) "Three-dimensional visualization of gammaherpesvirus life cycle in host cells by electron tomography" *Structure*
12. Malhas, Goulbourne, Vaux (2011) "The nucleoplasmic reticulum: form and function" *Trends Cell Biol*
13. Bh, Miyai, Slusser et al. (1964) "Mouse cytomegalovirus infection. An electron microscopic study of hepatic parenchymal cells" *Am J Pathol*
14. Papadimitriou, Shellam, Robertson (1984) "An ultrastructural investigation of cytomegalovirus replication in murine hepatocytes" *J Gen Virol*
15. Nassiri, Gilloteaux, Taichman et al. (1998) "Ultrastructural aspects of cytomegalovirus-infected fibroblastic stromal cells of human bone marrow" *Tissue Cell*
16. Dal Monte, Pignatelli, Zini et al. (2002) "Analysis of intracellular and intraviral localization of the human cytomegalovirus UL53 protein" *J Gen Virol*
17. Sutter, De Oliveira, Tobler et al. (2012) "Herpes simplex virus 1 induces de novo phospholipid synthesis" *Virology (Auckl)*
18. Pignatelli, Dal Monte, Landini et al. (2007) "Cytomegalovirus primary envelopment at large nuclear membrane infoldings: what's new?" *J Virol*
19. Reynolds, Ryckman, Baines et al. (2001) "U(L)31 and U(L)34 proteins of herpes simplex virus type 1 form a complex that accumulates at the nuclear rim and is required for envelopment of nucleocapsids" *J Virol*
20. Lorenz, Vollmer, Unsay et al. (2015) "A single herpesvirus protein can mediate vesicle formation in the nuclear envelope" *J Biol Chem*
21. Klupp, Granzow, Fuchs et al. (2007) "Vesicle formation from the nuclear membrane is induced by coexpres sion of two conserved herpesvirus proteins" *Proc Natl Acad Sci*
22. Hagen, Guttmann, Klupp et al. (2012) "Correlative VIS-fluorescence and soft Xray cryo-microscopy/tomography of adherent cells" *J Struct Biol*
23. Hagen, Dent, Zeev-Ben-Mordehai et al. (2015) "Structural basis of vesicle formation at the inner nuclear membrane" *Cell*
24. Lv, Shen, Xiang et al. (2019) "Functional identification and characterization of the nuclear egress complex of a gammaherpesvirus" *J Virol*
25. Scott, O'hare (2001) "Fate of the inner nuclear membrane protein lamin B receptor and nuclear lamins in herpes simplex virus type 1 infection" *J Virol*
26. Mou, Forest, Baines (2007) "US3 of herpes simplex virus type 1 encodes a promiscuous protein kinase that phosphorylates and alters localization of lamin A/C in infected cells" *J Virol*
27. Leach, Roller (2010) "Significance of host cell kinases in herpes simplex virus type 1 egress and lamin-associated protein disassembly from the nuclear lamina" *Virology (Auckl)*
28. Park, Baines (2006) "Herpes simplex virus type 1 infection induces activation and recruitment of protein kinase C to the nuclear membrane and increased phosphorylation of lamin B" *J Virol*
29. Wu, Pan, Zhang et al. (2016) "Herpes simplex virus 1 induces phosphorylation and reorganization of lamin A/C through the γ134.5 protein that facilitates nuclear egress" *J Virol*
30. Bjerke, Roller (2005) "Roles for herpes simplex virus type 1 UL34 and US3 proteins in disrupting the nuclear lamina during herpes simplex virus type 1 egress" *Virology (Auckl)*
31. Cano-Monreal, Wylie, Cao et al. (2009) "Herpes simplex virus 2 UL13 protein kinase disrupts nuclear lamins" *Virology (Auckl)*
32. Lee, Huang, Lin et al. (2008) "Epstein-Barr virus BGLF4 kinase induces disassembly of the nuclear lamina to facilitate virion production" *J Virol*
33. Sharma, Kamil, Coughlin et al. (2014) "Human cytomegalovirus UL50 and UL53 recruit viral protein kinase UL97, not protein kinase C, for disruption of nuclear lamina and nuclear egress in infected cells" *J Virol*
34. Hamirally, Kamil, Ndassa-Colday et al. (2009) "Viral mimicry of Cdc2/cyclin-dependent kinase 1 mediates disruption of nuclear lamina during human cytomegalovirus nuclear egress" *PLoS Pathog*
35. Naniima, Naimo, Koch et al. (2021) "Assembly of infectious Kaposi's sarcoma-associated herpesvirus progeny requires Full-Length Text Journal of Virology October"
36. "formation of a pORF19 pentamer" *PLoS Biol*
37. Brulois, Chang, Lee et al. (2012) "Construction and manipulation of a new Kaposi's sarcoma-associated herpesvirus bacterial artificial chromosome clone" *J Virol*
38. Wang, Tang, Maul et al. (2006) "Kaposi's sarcoma-associated herpesvirus ori-Lyt-dependent DNA replication: dual role of replication and transcription activator" *J Virol*
39. Drozdz, Jiang, Pytowski et al. (2017) "Formation of a nucleoplasmic reticulum requires de novo assembly of nascent phospholipids and shows preferential incorporation of nascent lamins" *Sci Rep*
40. Pytowski, Drozdz, Jiang et al. (2019) "Nucleoplasmic Reticulum Formation in Human Endometrial Cells is Steroid Hormone Responsive and Recruits Nascent Components" *Int J Mol Sci*
41. Mcphee, Dellaire, Ridgway (2024) "Mechanisms for assembly of the nucleoplasmic reticulum" *Cell Mol Life Sci*
42. Santos, Rappa, Karbanová et al. (2018) "VAMP-associated protein-A and oxysterol-binding protein-related protein 3 promote the entry of late endosomes into the nucleoplasmic reticulum" *J Biol Chem*
43. Kent (1997) "CTP:phosphocholine cytidylyltransferase" *Biochimica et Biophysica Acta (BBA) -Lipids and Lipid Metabolism*
44. Gehrig, Cornell, Ridgway (2008) "Expansion of the nucleoplasmic reticulum requires the coordinated activity of lamins and CTP:phospho choline cytidylyltransferase alpha" *Mol Biol Cell*
45. Lagace, Ridgway (2005) "The rate-limiting enzyme in phosphatidyl choline synthesis regulates proliferation of the nucleoplasmic reticulum" *Mol Biol Cell*
46. Camozzi, Pignatelli, Valvo et al. (2008) "Remodelling of the nuclear lamina during human cytomegalovirus infection: role of the viral proteins pUL50 and pUL53" *J Gen Virol*
47. Simpson-Holley, Colgrove, Nalepa et al. (2005) "Identification and functional evaluation of cellular and viral factors involved in the alteration of nuclear architecture during herpes simplex virus 1 infection" *J Virol*
48. Greber (2025) "Clicking viruses-with chemistry toward mechanisms in infection" *J Virol*
49. Loehr, Luedtke (2022) "A Kinetic and fluorogenic enhancement strategy for labeling of nucleic acids" *Angew Chem Int Ed*
50. Gomez-Gonzalez, Burkhardt, Bauer et al. (2024) "Stepwise virus assembly in the cell nucleus revealed by spatiotemporal click chemistry of DNA replication" *Sci Adv*
51. Balistreri, Viiliäinen, Turunen et al. (2016) "Oncogenic herpesvirus utilizes stress-induced cell cycle checkpoints for efficient lytic replication" *PLoS Pathog*
52. Liu, Schlagowski, Großkopf et al. (2025) "Kaposi's sarcoma-associated herpesvirus (KSHV) gB dictates a low-pH endocytotic entry pathway as revealed by a dual-fluorescent virus system and a rhesus monkey rhadinovirus expressing KSHV gB" *PLoS Pathog*
53. Taneva, Patty, Frisken et al. (2005) "CTP:phosphocholine cytidylyltransferase binds anionic phospholipid vesicles in a crossbridging mode" *Biochemistry*
54. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12718280&blobtype=pdf | Jeesun Chun, Yo-Han Ko
## Abstract
We analyzed the dsRNA profiles of Trichoderma spp. and identified a novel mycovirus from Trichoderma harzianum strain NFCF419. Phylogenetic and genomic analysis suggested that this virus is a novel member of the family Alternaviridae. Next-generation sequencing (NGS) analysis revealed that this virus consisted of four genome segments. The complete genome sequences of the segments were determined by NGS, and the NGS results were confirmed by manual sequencing of RT-PCR amplicons using specific primer pairs. 5'-and 3'-RACE was performed to determine the terminal sequences of each segment. The genome segments were found to be 3,572 bp (dsRNA1), 2,552 bp (dsRNA2), 2,593 bp (dsRNA3), and 1,484 bp (dsRNA4) in size. The largest segment (dsRNA1) contains a single open reading frame (ORF) encoding a putative RNA-dependent RNA polymerase (RdRP). dsRNA2, 3, and 4 each contain a single ORF encoding a putative methyltransferase, a coat protein, and a hypothetical protein with unknown function, respectively. Evaluation of the genome organization, analysis of the deduced amino acid sequence of the RdRP, and phylogenetic analysis indicated that this virus is a new member of the genus Alternavirus in the family Alternaviridae. Accordingly, we designated this novel mycovirus "Trichoderma harzianum alternavirus 1" (ThAV1). This is the first report of a mycovirus of the family Alternaviridae that infects a member of the genus Trichoderma.
and Polymycoviridae (International Committee on Taxonomy of Viruses, http://ictv.global). Fungi belonging to the genus Trichoderma are widely studied because of their ecological significance and potential applications in agriculture and biotechnology [5,6]. In recent years, the occurrence and characterization of dsRNA mycoviruses have been reported in different Trichoderma isolates using molecular and genomic approaches, revealing diverse genomic organizations and evolutionary relationships [7][8][9][10][11][12][13][14][15][16][17][18]. So far, members of at least four virus families, including the established families Hypoviridae, Partitiviridae, Fusagraviridae, and the proposed family "Ambiguiviridae", have been shown to infect Trichoderma spp. Here, we report the identification of a novel mycovirus of the family Alternaviridae, which we have named "Trichoderma harzianum alternavirus 1" (ThAV1), in Trichoderma harzianum strain NFCF419.
To the best of our knowledge, this is the first report of the presence of an alternavirus in a member of the genus of Trichoderma.
The fungus T. harzianum NFCF419 was isolated from a sawdust-based cultivation bag of Lentinula edodes Fungal viruses, referred to as mycoviruses, play an important role in fungal biology and ecology. They occur in diverse fungal lineages and are considered to be ubiquitous throughout the fungal kingdom [1][2][3][4]. Of the diverse groups of mycoviruses, double-stranded RNA (dsRNA) viruses are among the most extensively studied. Taxonomically, dsRNA mycoviruses are classified into several families, including Partitiviridae, Chrysoviridae, Orthototiviridae, Quadriviridae, Megabirnaviridae, Spinareoviridae, Alternaviridae, showing typical green mold symptoms in Chungcheongbuk-do, Korea [8]. To detect mycoviruses in T. harzianum strain NFCF419, mycelia were cultivated for 5 days on potato dextrose agar (PDA) overlaid with sterilized cellophane, and dsRNA was extracted from fresh biomass using cellulose chromatography [11]. The extracted dsRNA was treated with DNase I and S1 nuclease and then subjected to agarose gel electrophoresis, which revealed a multipleband pattern indicative of infection by multiple viruses (Fig. 1A). To identify the viral genome sequences, nextgeneration sequencing (NGS) was performed using the purified dsRNA. Briefly, dsRNA samples were fragmented and reverse transcribed to generate cDNA libraries, which were sequenced on an Illumina HiSeq 2000 platform. Raw sequence data were quality-filtered and adapter-trimmed using Trimmomatic, and de novo assembly was performed using Trinity. This resulted in a total of 5,341 contigs with an average length of 383 nucleotides. The resulting contigs were used for BLASTn and BLASTx searches of the NCBI virus database, which revealed the presence of an alternavirus and a hypovirus, suggesting a mixed infection. The hypovirus was found to be very similar to a previously reported virus [13], but the alternavirus was found to be a novel virus and is the focus of this report. NGS sequencing produced four alternavirus contigs, 3,561, 2,529, 2,570, and 1,471 nt in size. RT-PCR amplification using sequence-specific primer pairs based on the assembled contigs yielded amplicons of the expected sizes, and manual sequencing of the amplicons verified the presence of the viral genome and confirmed the NGS sequence.
Northern blot analysis using RT-PCR amplicons representing each of the four contigs as probes confirmed the identity of the dsRNA bands at the expected positions in the gel (Fig. 1B), and the terminal sequence of each genome segment was determined using 5'-and 3'-RACE.
The complete nucleotide sequences of the four dsRNA segments were determined to be 3,572 bp (dsRNA1) with a GC content of 53%, 2,552 bp (dsRNA2) with a GC content of 53%, 2,593 bp (dsRNA3) with a GC content of 57%, and 1,484 bp (dsRNA4) with a GC content of 60%, which is consistent with the sizes of the bands observed in agarose gel electrophoresis (Fig. 1A). The nucleotide sequences of the genome segments of the novel alternavirus, which we have named "Trichoderma harzianum alternavirus 1" (ThAV1) were deposited in the GenBank database under the accession numbers PX441381, PX441382, PX441383, and PX441384, respectively. Sequence analysis revealed that each segment contains a single ORF and terminates with a poly(A) tail at the 3′ end (Fig. 1C).
The ORF of dsRNA1 (ORF1) of ThAV1 starts at nucleotide position 72 and ends at position 3,446, and it is predicted to encode a protein of 1,124 amino acids with a calculated molecular weight of 127.8 kDa. A BLASTp search indicated that the deduced amino acid sequence had the highest similarity (59.9% identity) to the RNA-dependent RNA polymerase (RdRP) of Dactylonectria torresensis alternavirus 1 (DtAV1). A multiple alignment of RdRP sequences showed that the RdRp of ThAV1 contains all of the conserved motifs (I-VIII) commonly found in RdRPs. A distinctive feature of alternaviruses, the substitution of glycine (G) by alanine (A) within the metal ion-binding triplet sequence (ADD) in motif VI, was also observed (Fig. 2A). The ADD motif is conserved among alternavirus isolates that are closely related to ThAV1.
The ORF of dsRNA2 (ORF2) of ThAV1 extends from nt 66 to nt 2,372, encoding a protein of 768 amino acids with a calculated molecular mass of approximately 84.4 kDa. BLASTp analysis revealed that this protein shared the highest amino acid sequence similarity (36.1% identity) with the putative methyltransferase (MTase) encoded by dsRNA2 of DtAV1 [19], whose counterparts in other alternaviruses have been classified as hypothetical proteins with unknown function [20]. A comparison with the putative MTases of other alternaviruses [19] showed that one highly conserved sequence (GDXPG[T/S][L/F][G/A/S]RXL) is well conserved as "GDHPGSLGRAL" from aa 262 to aa 272, while another conserved sequence (V[V/T]GXDP[K/R]N) is only partially conserved as "SIGIDPLN" from aa 279 to aa 286.
The ORF in dsRNA3 (ORF3) of ThAV1 extends from nt 92 to nt 2,386 and encodes a protein containing 764 aa with a calculated molecular weight of 83.1 kDa. BLASTp analysis revealed that this protein shared the highest amino acid sequence similarity (47.2% identity) with the coat protein encoded by dsRNA3 of DtAV1. Interestingly, the segment of ThAV1 encoding the coat protein is longer than the segment (dsRNA2) encoding the methyltransferase, which has also been observed in the case of Ilyonectria robusta alternavirus 1 (IrAV1) [19].
The ORF of dsRNA4 (ORF4) of ThAV1 extends from nt 218 to nt 1,270 and encodes a protein containing 350 aa with a calculated molecular weight of 37.7 kDa. BLASTp analysis revealed that this protein shared the highest amino acid sequence similarity (29.2% identity) with a hypothetical protein of a different virus, Alternaria alternata alternavirus 1 (AaV1) [21]. The relatively low sequence similarity of this segment to its closest relative reflects the fact that DtAV1, which showed the highest similarity to ThAV1 in the other three segments, consists of only three genome segments and lacks a dsRNA4.
A comparison of the 5'-terminal sequences of the four segments of the ThAV1 genome showed them to be highly diverse but to have the first seven nucleotides 5'-G C C C C G T-3' in common (Fig. 2B).
Phylogenetic analysis was performed by the maximumlikelihood (ML) method [22] to investigate the evolutionary relationships of ThAV1 to other viruses. The RdRP acid sequence of the putative RdRP of ThAV1 and those of 22 reference alternaviruses, constructed using the maximum-likelihood (ML) method with 1,000 bootstrap replicates. A representative member of the family Quadriviridae was used as an outgroup. Bootstrap values greater than 70% are shown at the nodes sequence of ThAV1 was aligned with sequences from reference viruses, using CLUSTALx2 [23], and the alignment was manually inspected and trimmed to remove ambiguous regions. A phylogenetic tree was constructed in MEGA 7 [24] by the ML method with the Jones-Taylor-Thornton (JTT) amino acid substitution model and validated by bootstrap analysis with 1,000 replicates. The RdRP sequence of a virus belonging to the family Quadriviridae was included as an outgroup to root the phylogenetic tree, which showed that ThAV1 clustered with other alternaviruses (Fig. 2C). The 23 alternaviruses included in the analysis formed three clades, and most of them were in clade I, within which ThAV1 and DtAV1 formed a separate subclade.
Based on its genome organization and phylogenetic relationships, ThAV1 can be considered a new member of the family Alternaviridae. Since the first report in 2009, more than 20 viruses have been identified as members of the family Alternaviridae, and many of them infect fungi of the genus Fusarium [20]. To the best of our knowledge, this is the first report of an alternavirus infecting a member of the genus Trichoderma.
## References
1. Ghabrial, Suzuki (2009) "Viruses of plant pathogenic fungi" *Annu Rev Phytopathol*
2. Pearson, Beever, Boine et al. (2009) "Mycoviruses of filamentous fungi and their relevance to plant pathology" *Mol Plant Pathol*
3. Ghabrial, Castón, Jiang et al. (2015) "50-plus years of fungal viruses" *Virology*
4. Kotta-Loizou, Coutts (2017) *Mycoviruses in Aspergilli: A comprehensive review. Front Microbiol*
5. Harman (2011) "Multifunctional fungal plant symbionts: new tools to enhance plant growth and productivity" *New Phytol*
6. Harman, Howell, Viterbo et al. (2004) "Trichoderma species-opportunistic, avirulent plant symbionts" *Nat Rev Microbiol*
7. Lee, Yun, Chun et al. (2017) "Characterization of a novel dsRNA mycovirus of Trichoderma atroviride NFCF028" *Arch Virol*
8. Yun, Lee, So et al. (2016) "Incidence of diverse dsRNA mycoviruses in Trichoderma spp. causing green mold disease of shiitake Lentinula edodes" *FEMS Microbiol Lett*
9. Chun, Yang, Kim (2018) "Identification and molecular characterization of a novel partitivirus from Trichoderma atroviride NFCF394" *Viruses*
10. Chun, Yang, Kim (2018) "Identification of a novel partitivirus of Trichoderma harzianum NFCF319 and evidence for the related antifungal activity" *Front Plant Sci*
11. Chun, Na, Kim (2020) "Characterization of a novel dsRNA mycovirus of Trichoderma atroviride NFCF377 reveals a member of "Fusagraviridae" with changes in antifungal activity of the host fungus" *J Microbiol*
12. Chun, Yoon, Lee et al. (2024) "Co-infection with two novel mycoviruses affects the biocontrol activity of Trichoderma polysporum" *Biol Control*
13. Chun, So, Ko et al. (2022) "Molecular characteristics of a novel hypovirus from Trichoderma harzianum" *Arch Virol*
14. Liu, Li, Redda et al. (2019) "Complete nucleotide sequence of a novel mycovirus from Trichoderma harzianum in China" *Arch Virol*
15. Liu, Li, Redda et al. (2019) "A novel double-stranded RNA mycovirus isolated from Trichoderma harzianum" *Virol J*
16. You, Hu, Li et al. (2023) "The effect of trichoderma harzianum hypovirus 1 (ThHV1) and its defective RNA ThHV1-S on the antifungal activity and metabolome of Trichoderma koningiopsis T-51" *J Fungi (Basel)*
17. Zhang, Zeng, Cai et al. (2018) "Molecular characterization of a novel double-stranded RNA mycovirus of Trichoderma asperellum strain JLM45-3" *Arch Virol*
18. Wang, Liu, Jiang et al. (2022) "The newly identified Trichoderma harzianum partitivirus (ThPV2) does not diminish spore production and biocontrol activity of its host" *Viruses*
19. Pielhop, Popp, Fricke et al. (2023) "Molecular characterization of two new alternaviruses identified in members of the fungal family Nectriaceae" *Arch Microbiol*
20. Thompson, Gibson, Plewniak et al. (1997) "The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools" *Nucleic Acids Res*
21. Kumar, Stecher, Tamura (2016) "MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets" *Mol Biol Evol*
22. "Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations"
23. Hua, Zhang, Liu et al. (2024) "A novel strain of Fusarium oxysporum alternavirus 1 isolated from Fusarium oxysporum f. sp. melonis strain T-BJ17 confers hypovirulence and increases the sensitivity of its host fungus to difenoconazole and pydiflumetofen" *Viruses*
24. Aoki, Moriyama, Kodama et al. (2009) "A novel mycovirus associated with four double-stranded RNAs affects host fungal growth in Alternaria alternata" *Virus Res*
25. Fitch (1971) "Toward defining the course of evolution: minimum change for a specific tree topology" *Syst Zool* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12819382&blobtype=pdf | # A non-spike nucleocapsid R204P mutation in SARS-CoV-2 Omicron XEC enhances inflammation and pathogenicity
Shuhei Tsujino, Masumi Tsuda, Sayaka Deguchi, Jumpei Ito, Taha Taha, Hesham Nasser, Lei Wang, Julia Rosecrans, Rigel Suzuki, Saori Suzuki, Kumiko Yoshimatsu
## Abstract
On behalf of The Genotype to Phenotype Japan (G2P-Japan) Consortium*The global circulation of SARS-CoV-2 in human populations has driven the emergence of Omicron subvariants, which have become highly diversified through recombination. In late 2024, SARS-CoV-2 Omicron XEC variant emerged from the recombination of two JN.1 progeny, KS.1.1 and KP.3.3, and became predominant worldwide. Here, we investigate virological features of the XEC variant. Epidemic dynamics modeling suggests that spike substitutions in XEC mainly contribute to its increased viral fitness. Additionally, four licensed antivirals are effective against XEC. Although the fusogenicity of XEC spike is comparable to that of the JN.1 spike, the intrinsic pathogenicity of XEC in male hamsters is significantly higher than that of JN.1. Notably, we find that the nucleocapsid R204P mutation of XEC enhances inflammation through NF-κB activation. Recent studies suggest that the evolutionary potential of spike protein is reaching its limit. Indeed, our findings highlight the critical role of non-spike mutations in the future evolution of SARS-CoV-2.RNA viruses are prone to high mutation rates. Environmental factors including host immune responses and antiviral treatment serve as selective pressures that favor the expansion of resistant variants. Genetic variation in RNA viruses typically occur through substitution, deletion and insertion. In addition, dramatic variation through recombination occurs in RNA viruses and is thought to have played a significant role in the recent evolutionary histories 1-4 . A notable example is severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 5,6 the causative agent of the highly contagious disease COVID-19.Because humans experience infection and/or vaccination, the circulating variants, namely Omicron have evolved to exhibit reduced intrinsic pathogenicity, increased transmissibility, and enhanced immune escape compared with ancestral variants [7][8][9][10][11][12][13][14][15][16] . The continuous circulation of Omicron has led to the emergence of "recombinant variants" through simultaneous infection with multiple variants and recombination in the host. Omicron XBB lineage was the most prevalent lineage worldwide, a recombinant between BA.2 subvariants BJ.1 and BM.1.1.1 (a descendant of BA.2.75). XBB is the first SARS-CoV-2 variant whose fitness increased through recombination rather than substitution 12 . Other recombinant variants, such as XBC.1.6 (Delta variant B.1.617.2 × BA.2) and XDD.1.1 (EG.5.1.1 × JN.1), have also been identified, though they did not become the predominant variant.Recent studies indicate that intracellular dynamics can modulate recombination frequency and thereby shuffle traits such as transmissibility and antigenicity 17 , and clinical reports confirm that recombination can occur in vivo and be transmitted onward 18 . In line with this,
newly emerging lineages such as NB.1.8.1 (JN.1 × XDE) and XFG (LF.7 × LP.8.1.2) have become globally predominant, underscoring that recombination remains an active force in SARS-CoV-2 evolution.
Omicron XEC variant emerged from the recombination of two JN.1 descendants, KS.1.1 (JN.13.1.1.1) and KP.3.3 (JN. 1.11.1.3.3). XEC was first identified in Germany in August 2024, has rapidly spread in several Western continents 19 . As of April 2025, XEC is predominantly circulating worldwide according to Nextstrain (clade 24F; https:// nextstrain.org/ncov/gisaid/global/6m). Because XEC is a chimera of KS.1.1 and KP.3.3 variants, spike (S) protein of XEC inherited key mutations enhanced binding affinity to the angiotensin-converting enzyme 2 (ACE2) receptor from KP.3.3 and acquired mutations for the immune-evasive properties from KS.1.1 [19][20][21][22][23] . This suggests that XEC S maintains a similar affinity to ACE2 and a higher capacity to evade immunity compared with the previously dominant variants, KP.3 and KP.3.1.1. In addition, XEC has the R204P mutation in the nucleocapsid (N) protein and the A419T mutation in the NSP2 protein, which it inherited from KP.3.3 and KS.1.1, respectively. In the past variants, the co-occurring mutations at positions 203 and 204 of N (R203K/G204R) are involved in the efficient assembly of viral particle, leading to increased viral replication and pathogenicity of SARS-CoV-2 [24][25][26] . The effect of the N:R204P mutation on the virological characteristics including replication and pathogenicity has not been reported.
In this study, we aimed to characterize the virological features of SARS-CoV-2 XEC variant and investigate how the R204P mutation in the nucleocapsid protein affects viral replication and pathogenicity.
## Results
Mutations contributing to the increased viral fitness of XEC Compared with JN.1, XEC has five amino acid substitutions (S:T22N, S:F59S, S:F456L, S:Q493E, and S:V1104L) in S and two substitutions in the non-spike protein genes (NSP2:A419T and N:R204P) 19 . To identify the mutations contributing to the rapid spread of XEC, we applied a hierarchical Bayesian multinomial logistic model established in our previous study to viral genome surveillance data from GISAID 11 . Here, fitness refers to the relative effective reproductive number (R e ) between variants, estimated under the assumption that the relative values of the R e among variants remain constant over time. This model predicts the effect of mutations by assuming that the fitness of a given variant is represented as the sum of the effects of its constituent mutations 11 . When multiple mutations are highly co-occurring, it becomes challenging to estimate their individual effects. To address this, these co-occurring mutations were grouped into mutation clusters, and their effects were estimated at the mutation cluster level. We applied this model to viral genome epidemiological surveillance data from the USA, covering the period from January 1, 2024, to December 31, 2024. This dataset includes 209 mutation clusters, composed of 277 distinct mutations (Supplementary data 1).
?A3B2 twb=.3w?>Among the 209 analyzed mutation clusters, 15 were estimated to contribute positively to increased fitness (Fig. 1A,B). The mutation cluster that exhibited the highest positive effect corresponded to mutations acquired on the stem branch of the BA.2.86 lineage or the JN.1 lineage. Notably, among the seven characteristic mutations of XEC, S:F456L, S:Q493E, S:F59S, and S:T22N were estimated to have a significant positive effect on fitness, ranking in the top 2, 3, 5 and top 8, respectively. In contrast, no statistically significant effect was detected for S:V1104L, NSP2:A419T and N:R(G)204 P (Fig. 1A). These results suggest that the rapid spread of XEC was driven by S:F456L, S:Q493E, S:T22N, and S:F59S rather than other substitutions.
## Fusogenicity of XEC S
The fusogenicity of the XEC S protein was measured by the SARS-CoV-2 S protein-mediated membrane fusion assay using Calu-3/DSP1-7 cells (Fig. S1A, B) 27 . The surface expression level of XEC S was comparable to that of the parental JN.1 S (Fig. S1C,D). As previously reported [27][28][29] , the Delta S protein exhibited the greatest fusogenicity, while the KP.3 S protein exhibited the weakest fusogenicity (Fig. 2A). In the case of XEC S, comparable fusogenicity to the JN.1 S protein was observed, suggesting the XEC S protein contributes similar viral pathogenicity to the JN.1 S protein.
## Antiviral effect of clinically available compounds against XEC
We assessed the sensitivity of XEC to four licensed antiviral drugs, EIDD-1931, nirmatrelvir (also known as PF-07321332), remdesivir, and ensitrelvir. Clinical isolate of JN.1 was used as a control. Both viruses were inoculated into human induced pluripotent stem cell (iPSC)-derived lung organoids, a physiologically relevant model, and treated with the four antiviral drugs. Among them, nirmatrelvir showed the strongest antiviral effects, with no significant differences in antiviral efficacy were observed between JN.1 and XEC (Figs. 2B andS2). Remdesivir and ensitrelvir demonstrated significant antiviral effects on the two isolates, whereas EIDD-1931 exhibited moderate antiviral effects on the two isolates.
## Viral replication and pathogenicity of XEC variant
To investigate the replication efficiency and intrinsic pathogenicity of XEC, we used the clinical isolates of JN.1 and XEC. First, we inoculated clinical isolates of JN.1 and XEC into VeroE6 cells expressing TMPRSS2 30 (Fig. 2C) and into human iPSC-derived lung organoids (Fig. S3). Quantifying viral infectious titers in supernatants showed that the replication efficiency of XEC was comparable to that of JN.1 in both cell culture systems. Next, clinical isolates of JN.1 and XEC were respectively intranasally inoculated into hamsters, the established animal model for evaluation [7][8][9][10][11][12][13][14][15][16]28 . To evaluate viral spread in infected hamsters, we routinely measured the viral RNA load in oral swabs. The viral RNA load of XEC-infected hamsters was comparable to that of JN.1-infected hamsters at 2 and 5 d.p.i. (Fig. 2D). Interestingly, the body weights of the hamsters infected with XEC were significantly lower than those of JN.1-infected hamsters (Fig. 2E), representing a transient weight loss. To further evaluate intrinsic pathogenicity, we performed histopathological analyses of lung tissues. At 2 d.p.i., JN.1-infected hamsters showed only limited bronchial and bronchiolar inflammation, whereas XECinfected hamsters exhibited more pronounced alveolar damage (Figs. 2F andS4). Together, these results indicate that, despite comparable replication efficiency, XEC exhibits higher pathogenicity than JN.1.
## Mutation dynamics of nucleocapsid protein
Because N protein functions as the basis for viral RNA genome packaging into ribonucleotide complex (RNP) and assembly into virus particles, functional changes lead to alter viral pathogenicity in vivo 25,[31][32][33] . To investigate the effects of mutation in N, the frequency of mutations in JN.1 and XEC was examined. P13L, E31del, R32del, S33del, G204R, and S413R mutations were acquired in a convergent manner during the evolution of the Omicron variant (Fig. 3A) 34 . In addition, Q229K mutation was acquired during evolution to BA.2.86 15 . The previous study showed R203K/G204R mutation contributed to enhance both viral replication and pathogenicity. Therefore, the R204P mutation in XEC may affect alteration of viral replication and pathogenicity.
## Effects of N:R204P on the viral assembly and replication
To evaluate the effect of the N:R204P mutation on the viral assembly, we performed a virus-like particle (VLP) assay. The formation of VLPs in 293T cells containing SARS-CoV-2 structural proteins (S, E, M, and N) and packaging RNA is detected through expression of a luciferase reporter in infected receiver cells 35,36 . When compared within the same backbone (JN.1N vs. XEC N), the N:R204P mutation did not affect the assembly of VLPs (Fig. 3B). Next, to examine the effect of N:R204P mutation on viral replication, we generated recombinant viruses carrying the single mutation in the N protein of XEC, rXEC/N:P204R, rXEC, and rJN.1. The recombinant viruses were inoculated into the cell cultures we examined as above. In VeroE6/TMPRSS2 cells (Fig. 3C), no significant differences in growth kinetics were observed between the viruses. These findings suggest that the N:R204P mutation has little effect on the viral replication.
To investigate the impact of XEC infection on the airway epithelial and endothelial barriers, we employed an airway-on-a-chip system (Fig. S5). Viral penetration from the upper channel to the lower channel serves as an indicator of the viral ability to breach these barriers. The proportion of viruses that penetrated the lower channel of the XECinfected airway-on-a-chip was higher compared with JN.1-and XEC/ N:P204R-infected airway-on-a-chip (Figs. 3D, E andS6). SARS-CoV-2 N activates human endothelial cells through Toll-like receptor 2 (TLR2)/ NF-κB and mitogen-activated protein kinase (MAPK) signaling pathways 37 . Therefore, N:R204P mutation may affect the airway epithelial and endothelial barriers through endothelial activation. In our previous study using airway-on-a-chips, we confirmed that SARS-CoV-2 infection causes paracellular gaps of endothelial cells to widen 38 . Thus, the N protein may facilitate the viral penetration to the lower channel by widening the gaps between endothelial cells.
## Effects of N:R204P on the viral pathogenicity
To further investigate the intrinsic pathogenicity of the three recombinant viruses, the viruses were respectively inoculated as above. Consistent with the result of clinical isolates, the body weights of the hamsters infected with rXEC were significantly lower than those of rJN.1-infected hamsters (Fig. 3F). Notably, the weight of hamsters infected with rXEC was significantly lower than those of hamsters infected with rXEC/N:P204R, indicating that N:R204P mutation contributes to enhanced pathogenicity.
To evaluate viral spread of the three recombinant viruses in hamsters, the viruses were respectively inoculated into hamsters. We routinely measured the viral RNA load in oral swabs. The viral RNA load of rJN.1-, rXEC/N:P204R-and rXEC-infected hamsters were comparable (Fig. 3G). We then compared viral spread in respiratory tissues. We collected the lungs of infected hamsters at 2 and 5 d.p.i., and the harvested tissues were separated into the hilum and periphery regions. The viral RNA loads of rXEC-infected hamsters were significantly lower than those of rJN.1-and rXEC/N:P204R-infected hamsters (Fig. 3H left and right). The red number in each panel indicates the fold difference between JN.1 and the derivative tested at 24 h post coculture. Assays were performed in quadruplicate. Statistically significant differences versus JN.1 across time points were determined by multiple regression. B Effect of antiviral drugs against XEC. Antiviral effects of the four drugs (EIDD-1931, nirmatrelvir [also known as PF-07321332], remdesivir, and ensitrelvir) in human iPSC-derived lung organoids. The assay of each antiviral drug was performed in triplicate, and the 50% effective concentration (EC 50 ) was calculated. The viral RNA amount without treatment with antiviral drugs was set as 100%. C JN.1 and XEC were inoculated into VeroE6/TMPRSS2 cells (MOI = 0.01). The 50% tissue culture infectious dose (TCID 50 ) of the culture supernatant were routinely quantified. (n = 3 independent experiments). D, E Syrian hamsters were intranasally inoculated with JN.1 and XEC. Six hamsters of the same age were intranasally inoculated with saline (uninfected). D Six hamsters per group were quantified viral RNA load in oral swab by RT-qPCR. E Six hamsters per group were used to routinely measure the body weight. Uninfected (saline) hamster data is also shown. The familywise error rates (FWERs) calculated using the Holm method are indicated in the figures. h.p.i: hours post-infection; d.p.i: days post-infection. F H&E staining of the lungs at 2 d.p.i. of infected hamsters. Representative figures and uninfected lung alveolar space are shown. The presented data are expressed as the average ± SD (A, B) or SEM (C-E). Scale bars, 250 µm.
## Effects of XEC N:R204P on immunopathogenic features of XEC
We further performed IHC analysis targeting the viral N protein in the respiratory tissues of infected hamsters. At 2 d.p.i, N-positive cells were more detected in the bronchi/bronchioles of rJN.1-infected hamsters than rXEC-and the N:P204R-infected hamsters (Figs. 4A andS7).
To investigate the intrinsic pathogenicity of recombinant viruses, histopathological analyses were performed. At 2 d.p.i, alveolar damage around the bronchi was prominent in rXEC-and the N:P204R-infected hamsters (Figs. 4B andS8). On the other hand, inflammation was limited in bronchi/bronchioles in the hamsters infected with rJN.1 (Fig. 4B). At 5 d.p.i, the alveolar architecture was more destroyed by alveolar damage and the expansion of type II pneumocytes in rXEC-infected hamsters (Fig. 4B). The low viral load in XEC-infected lungs may be due to viral clearance caused by an enhanced immune response, judging from the degree of inflammation (Fig. 4A,B). Notably, a strong inflammation in the acute phase was observed in the lungs of rXEC-infected hamsters even at 5 d.p.i. These results suggest that the N:R204P mutation contributes to increased inflammation in the lung.
## NF-κB activation of XEC N protein
SARS-CoV-2 accessory and non-structural proteins, including ORF3a, ORF7a, NSP5, NSP6 and NSP14, have been shown to activate the NF-κB pathway and induce downstream inflammatory cytokines and chemokines [39][40][41][42][43][44] . On the other hand, SARS-CoV-2 N protein has been reported to inhibit NF-κB activation and downstream signaling by inhibiting the formation of the TAK1-TAB2/3 complex 45 . In addition, N protein lacking interaction with the TAK1-TAB2/3 complex has been shown to induce strong inflammation. Since proline in the linker increases structural rigidity 46 , we hypothesized that the R204P mutation might alter the overall structure of N protein and impair this regulatory function.
Using AlphaFold3 47 , we found that the amino acid substitutions from arginine to proline at position 204 in XEC N altered the position of the α-helix and the linker (Fig. 4C). As this α-helix has been reported to be important for interaction with the TAK1-TAB2/3 complex, we investigated the effect of N:R204P mutation on NF-κB activation by NF-κB promoter-driven luciferase assay in HEK293/ACE2/TMPRSS2 cells. As expected, the luciferase assay indicated that rXEC-infection activated NF-κB promoter more than rJN.1-and rXEC/N:P204R-infection in HEK293/ACE2/TMPRSS2 cells 48 (Fig. 4D).
To assess NF-κB-induced inflammatory signaling provoked by viral infection in vivo, we extracted mRNA of the lung hilum and periphery areas at 2 d.p.i., and quantified the expression of four interferon-stimulated genes (ISGs) (Il-１β, Il-6, Il-8, and Ccl2) (Fig. 4E). In the lung hilum, rXEC-infection upregulated Il-6 and Il-8 expression more than rJN.1-and rXEC/N:P204R-infection, suggesting that XEC could induce severe inflammation, supporting the histopathological data. In the lung hilum, the expression levels of Il-１β and Ccl2 were also higher in rXEC-infected lungs than in rJN.1-infected lungs. Taken together, these results indicate that the N:R204P mutation enhances inflammation through NF-κB activation.
## Discussion
SARS-CoV-2 continues to circulate in human populations worldwide, and recombination between co-circulating variants remains a major driver of its diversification. SARS-CoV-2 Omicron XEC variant, which emerged through recombination between two JN.1 descendants, rapidly spread and became globally predominant. Our previous report demonstrated that XEC possesses an effective reproduction number (R e ) 1.13 times higher than that of the previously dominant KP.3.1.1 variant 19 . To the best of our knowledge, XEC is the second SARS-CoV-2 recombinant lineage to achieve global predominance, following the XBB lineage. In this study, we conducted a comprehensive multiscale investigation to elucidate the virological properties of XEC and identify the mutation(s) defining the characters.
Epidemic dynamics modeling showed that two mutations in the XEC lineage, S:T22N and S:F59S contribute significantly to its rapid spread (Fig. 1). Consistent with these findings, Li et al. recently reported that these mutations alter hydrophilic interactions with adjacent S protein residues, thereby affecting both structural stability and neutralization capacity 23 . We then evaluated the growth kinetics, sensitivity to clinically available antiviral compounds and fusogenicity of the XEC variant in cell culture and lung organoids. Our experimental results showed that sensitivity to licensed antivirals, fusogenicity and replication efficiency of XEC were comparable to those of JN.1 (Fig. 2A-D). Because the lung organoids used to evaluate the antiviral drugs were generated from a single donor, our findings may not fully reflect inter-individual differences. Therefore, future studies should employ organoid panels generated from multiple donors to better capture variability in drug responses.
The most striking difference between XEC and JN.1 lies in the enhanced intrinsic pathogenicity of XEC, as assessed in the hamster model that has been established and employed in our series of studies [7][8][9][10][11][12][13][14][15][16]28 . As shown in Fig. 2E, XEC infection resulted in a greater reduction in body weight compared to JN.1. In a recent study, Guo et al. reported that SARS-CoV-2 N protein inhibits NF-κB activation and that functional loss of N leads to enhanced inflammation 45 . Thus, we generated the single mutant virus, rXEC/N:P204R, which carries an N gene mutation reverted to the JN.1 sequence (Fig. 3F). rXEC/N:P204R infection led to less attenuation of weight loss compared to the parental rXEC infection. Histopathological analyses also indicated the diminished inflammation upon infection with rXEC/N:P204R (Fig. 4B). Structural modeling using AlphaFold3 suggested that the N:R204P substitution alters the position of the α-helix in the C-terminal domain of the N protein, potentially impairing its ability to inhibit NF-κB activation via interaction with the TAB2/3 complex. Of noted, these predictions will require biochemical validation in future studies, for example by Surface Plasmon Resonance (SPR) or Isothermal Titration Calorimetry (ITC). Consistent with this, the promoter assay revealed that the N:R204P mutation contributes to NF-κB activation (Fig. 4C,D). Furthermore, in vivo analysis showed significantly elevated expression of IL-6 and IL-8 in the lungs of rXEC-infected hamsters compared to those infected with rXEC/N:P204R, indicating enhanced NF-κBmediated inflammatory signaling (Fig. 4E). Taken these findings together, the N:R204P mutation might lead to disrupt the N protein's ability to inhibit NF-κB activation, thereby contributing to the enhanced pathogenicity of XEC in vivo. Although the N:R204P mutation did not affect viral replication in vitro (Fig. 3C), viral RNA levels and N-positive areas in the lungs of XEC-infected hamsters were lower than those in rJN.1 infection (Figs. 3H and4A). We interpret this discrepancy between in vitro and in vivo as the result of an exaggerated inflammatory response: accelerated viral clearance coinciding with immune-mediated tissue damage. These findings highlight a mechanism by which host inflammatory responses, rather than intrinsic replication capacity, may predominantly drive the severe pathogenicity of the XEC variant.
Thus far, Omicron subvariants have largely evolved through modifications in the S protein that enable escape from immune pressure in the human population [6][7][8][9][10][11][12][13][14][15] . While the N:R204P itself offers no apparent adaptive benefit and the S mutation remains the principal determinant of XEC dissemination, accumulating evidence indicates E The percentage of viral RNA load in the bottom channel per top channel at 6 d.p.i. (i.e., % invaded virus from the top channel to the bottom channel) is shown. F-H Syrian hamsters were intranasally inoculated with the recombinant viruses. Six hamsters per group were used to routinely measure the respective parameters. F Body weight of infected hamsters (n = 6 per infection group). Uninfected hamster data is also shown. G Viral RNA loads in the oral swab (n = 6 per infection group) at 2 and 5 d.p.i. H Viral RNA loads in the lung hilum (left) and lung periphery (right) of infected hamsters (n = 4 per infection group) at 2 and 5 d.p.i. The FWERs calculated using the Holm method are indicated in the figures. The presented data are expressed as the average ± SEM (B-H). Statistical significance was determined using Tukey's multiple comparison test ( * p < 0.05, * * p < 0.01, * * * p < 0.001, that the evolutionary plasticity of SARS-CoV-2 mediated by the S protein may be approaching saturation 16,[49][50][51][52] . For example, while the JN.1 subvariant demonstrates enhanced immune evasion conferred by the S:L455S mutation, this is accompanied by compromised replication efficiency and decreased pathogenicity in the lungs 16,49 . Our data indicate that the acquisition of the N:R204P mutation enhances pathogenicity of XEC compared to JN.1. This highlights that strategies to modulate certain virological properties through mutations in nonspike proteins may become more common in the future evolution of SARS-CoV-2. In addition, even non-spike mutations that do not provide an immediate selective advantage may persist at low frequency and reemerge under certain evolutionary contexts. This is consistent with the observation that although N:G204P had been reported in B.1.1.7 sublineage, it did not become dominant 53 . While this mutation has not been retained in currently circulating variants, frequent recombination among co-circulating lineages can reintroduce mutations with pathogenic potential, even if they are not maintained under strong positive selection.
Beyond spike-mediated adaptations, increasing evidence highlights the contribution of non-spike proteins to SARS-CoV-2 pathogenicity and transmission. ORF7b perturbs host defenses by disrupting epithelial integrity and modulating innate immune signaling 54,55 . ORF8 facilitates immune evasion through MHC-I downregulation while enhancing proinflammatory responses 56,57 . Deletions in ORF7b and ORF8 have been repeatedly detected in circulating variants, suggesting a role in viral transmissibility 58,59 . NSP6 remodels host membranes to form replication organelles and suppresses autophagy and interferon responses; recurrent mutations such as Δ106-108 and R252K fine-tune viral fitness by altering autophagy and replication dynamics [60][61][62][63] . In addition, the nucleocapsid (N) protein is particularly multifunctional, mediating RNA packaging through liquid-liquid phase separation, antagonizing interferon signaling by interfering with stress granules, and activating the NLRP3 inflammasome [64][65][66][67][68][69] . Given this pleiotropy, the impact of N is likely amplified through interactions with other non-spike proteins. For example, NSP2-which also carries mutations in the XEC lineage-regulates host translation and suppresses antiviral signaling, thereby modulating innate immune responses 70,71 . Although direct synergy between NSP2 and N has not yet been demonstrated, their convergence on innate immune regulation raises the possibility of cooperative effects that exacerbate immune-mediated pathology.
In summary, our comprehensive characterization of the XEC variant suggests an emerging shift in the evolutionary trajectory of SARS-CoV-2, from spike-driven adaptation to a combination of spike and non-spike modulation. These insights not only advance our understanding of SARS-CoV-2 biology but also inform more effective strategies for controlling COVID-19.
## Methods
## Virus-like particle assay
A virus-like particle (VLP) assay was employed as a physiological model to test the efficiency of packaging and assembly as a function of the mutations on SARS-CoV-2 N protein 35,36 . The assay was conducted as described previously 35,36 . Briefly, the VLPs were generated by coexpressing all four structural proteins of SARS-CoV-2 in HEK293T cells along with a construct containing ~1 kb viral packaging signal PS9 incorporated into the untranslated region of a firefly luciferase reporter. VLPs in the supernatant carrying luciferase reporter were added to receiver HEK293T cells stably expressing ACE2 and TMPRSS2 (HEK293T/ACE2/TMPRSS2 cells) in 96-well plates. Luminescence in receiver cells was measured at 12-16 h post infection using luciferase assay system (Promega).
## Cell culture
HEK293 cells (a human embryonic kidney cell line; ATCC, CRL-1573) and HEK293T cells (a human embryonic kidney cell line; ATCC, CRL-3216) were maintained in DMEM (high glucose) (Sigma-Aldrich, Cat# 6429-500 ML; Nacalai Tesque, Cat# 08458-16) containing 10% FBS and 1% PS. VeroE6/TMPRSS2 cells (an African green monkey kidney cell line stably expressing human TMPRSS2; JCRB Cell Bank, JCRB1819) 30 were maintained in DMEM (low glucose) (FUJIFILM Wako Chemicals, Cat# 041-29775) containing 10% FBS, G418 (1 mg/ml), and 1% PS. Calu-3/ DSP 1-7 cells (Calu-3 cells stably expressing DSP 1-7 ) 72 were maintained in EMEM (FUJIFILM Wako Chemicals, Cat# 056-08385) containing 20% FBS and 1% PS.
## NF-κB promoter assay
HEK293/ACE2/TMPRSS2 cells were seeded onto 24-well plates and transfected. To assess NF-κB signaling, 50 ng of the pNL3.2.NF-κB-RE vector (Promega, Cat# N1111) was utilized. Following transfection for 24 h, cells were infected for 1 h. For controls, cells were treated with DMEM (negative control) or stimulated with the recombinant TNF-α (positive control) (Thermo Fisher Scientific, Cat# 300-01 A) for 1 h. The cells were then lysed using a passive lysis buffer and subjected to luciferase activity measurements by the Luciferase Reporter Assay System using an AB-2270 Luminescencer Octa (Atto).
## Modeling the relationship between amino acid mutations and epidemic dynamics
To identify the mutations contributing to the rapid spread of XEC, we utilized a hierarchical Bayesian multinomial logistic model previously developed by our research group to estimate the effects of mutations on fitness 11 . Here, fitness refers to the relative R e between variants, estimated under the assumption that the relative values of the R e among variants remain constant over time. Unlike the conventional method of estimating R e using a multinomial logistic model 28 , this model does not directly estimate the fitness of variants. Instead, it introduces a hierarchical structure to estimate fitness as a linear combination of mutations. As a result, this model can estimate not only the fitness of each variant but also the effects of individual mutations on fitness. For highly co-occurring mutations (mutation clusters), our method does not permit the estimation of individual effects, and thus, effects are estimated at the mutation cluster level. For details on the model, please refer to Ito et al 11 . In this study, we estimated the effects of mutations not only in the spike protein but across all viral proteins.
The data used in this analysis were downloaded from the GISAID database (https://www.gisaid.org/) on January 26, 2025 73 . For quality control, we excluded the data of viral sequences with the following features from the analysis: (i) a lack of collection date information; (ii) sampling in animals other than humans; (iii) >1% undetermined nucleotide characters; or (iv) sampling by quarantine. Furthermore, in this analysis, we analyzed viral sequences collected in the USA from January 1, 2024, to December 31, 2024.
For the analysis, we selected mutations (including substitutions, insertions, and deletions) observed in ≥100 sequences in the dataset we used. We then excluded mutations commonly (≥95%) detected in sequences analyzed. According to the criteria above, 209 mutations were retrieved. Subsequently, we classified viral sequences into haplotypes, a group of viral sequences sharing the same set of mutations, according to the profile of the selected mutations. We excluded haplotypes with ≤30 sequences from the downstream analyses. According to the criterion above, 366 types of haplotypes, composed of 51,127 sequences, were retrieved. Then, we clustered highly cooccurring mutations (i.e., a pair of mutations with >0.9 Pearson's Images are from comparable lung lobes, not identical microscopic fields. Uninfected hamster data is also shown. B H&E staining of the lungs at 2 d.p.i. (left) and 5 d.p.i. (right) of infected hamsters. Representative figures and uninfected lung alveolar space are shown. C The structure of N (JN.1 N and XEC N) was predicted by Alphafold3. D HEK293/ACE2/TMPRSS2 cells were transfected with NF-κB reporter vector. At 24 h after transfection, cells were infected with rJN.1, rXEC/N:P204R, and rXEC for 1 h. Luciferase activity was measured at 12 and 24 h.p.i. Horizontal lines in figures represent the average value of the negative control group. (n = 3 independent experiments). E mRNA of the lung tissues obtained at 2 d.p.i. was used to measure expression levels of inflammatory genes (Il-1β, Il-6, Il-8, and Ccl2) with normalization using the housekeeping gene Rpl18. (n = 4 per infection group). The presented data are expressed as the average ± SEM (D, E). Statistical significance was determined using Tukey's multiple comparison test ( * p < 0.05, * * p < 0.01, correlation in the mutation profile matrix) into mutation clusters. Consequently, our dataset included the profiles of 209 mutation clusters for 366 haplotypes. For the reference haplotype, which is estimated to have a relative R e of 1, we selected the haplotype with the highest sequence count, corresponding to the major haplotype of JN.1.
Parameter estimation was performed via the MCMC approach implemented in CmdStan v2.30.1 (https://mc-stan.org) with CmdStanr v0.5.3 (https://mc-stan.org/cmdstanr/). Four independent MCMC chains were run with 500 and 2000 steps in the warmup and sampling iterations, respectively. We confirmed that all estimated parameters showed <1.01 R-hat convergence diagnostic values and >200 effective sampling size values, indicating that the MCMC runs were successfully convergent. The above analyses were performed in R v4.2.1 (https:// www.r-project.org/). Information on the estimated effect size of each mutation cluster on relative R e is summarized in Supplementary data 1.
## Plasmid construction
The nine pmW118 plasmids containing the partial gene of SARS-CoV-2 BA.2.86 were previously generated 74 . To generate the recombinant JN.1 and XEC viruses, the mutations were introduced into the corresponding plasmids encoding BA.2.86 gene by inverse fusion PCR cloning using the primers listed in Table S1. Sequences of all the plasmids used in this study were confirmed by a SeqStudio Genetic Analyzer (Thermo Fisher Scientific) and an outsourced service (Fasmac). Primer and plasmid information can be provided upon request.
## SARS-CoV-2 S-based fusion assay
A SARS-CoV-2 S-based fusion assay was performed as previously described 75 . Briefly, on day 1, effector cells (i.e., S-expressing cells) and target cells (Calu-3/DSP1-7 cells) were prepared at a density of 0.6-0.8 × 10 6 cells in a 6-well plate. On day 2, for the preparation of effector cells, HEK293 cells were cotransfected with the S expression plasmids and pDSP 8-11 76 using TransIT-LT1 (Mirus Bio, Cat# MIR2300). On day 3, 16,000 effector cells were detached and reseeded into a 96well black plate (PerkinElmer), and target cells were reseeded at a density of 1,000,000 cells/2 ml/well in 6-well plates. On day 4, target cells were incubated with EnduRen live cell substrate (Promega, Cat# E6481) for 3 h and then detached, and 32,000 target cells were added to a 96-well plate with effector cells. Renilla luciferase activity was measured at the indicated time points using Centro XS3 LB960 (Berthold Technologies). For measurement of the surface expression level of the S protein, effector cells were stained with rabbit anti-SARS-CoV-2 S S1/S2 polyclonal antibody (Thermo Fisher Scientific, Cat# PA5-112048, 1:100). Normal rabbit IgG (Southern Biotech, Cat# 0111-01, 1:100) was used as a negative control, and APC-conjugated goat antirabbit IgG polyclonal antibody (Jackson ImmunoResearch, Cat# 111-136-144, 1:50) was used as a secondary antibody. The surface expression level of S proteins was measured using CytoFLEX Flow Cytometer (Beckman Coulter) and the data were analyzed using FlowJo software v10.7.1 (BD Biosciences). For calculation of fusion activity, Renilla luciferase activity was normalized to the mean fluorescence intensity (MFI) of surface S proteins. The normalized value (i.e., Renilla luciferase activity per the surface S MFI) is shown as fusion activity.
## Antiviral drug assay using SARS-CoV-2 clinical isolates and human iPSC-derived lung organoids
The antiviral drug assay was performed as previously described 15 . Human iPSC-derived lung organoids were used to perform the antiviral drug assay. These lung organoids contain both airway and lung epithelial cells 77 . As we reported previously 15 , after recovering the lung organoids from Matrigel, we seeded them as clusters onto thin Matrigel-coated plates. This procedure enables SARS-CoV-2 to access the apical side of the lung organoids. The human iPSC-derived lung organoids were infected with either JN.1 or XEC isolate (100 TCID 50 ) at 37 °C for 2 h. Following infection, the cells were washed with DMEM and cultured in DMEM supplemented with 10% FCS, 1% PS, and the serially diluted EIDD-1931 (an active metabolite of Molnupiravir; Cell Signaling Technology, Cat# 81178S), Nirmatrelvir (MedChemExpress, Cat# HY-138687), Remdesivir (Clinisciences, Cat# A17170), or Ensitrelvir (MedChemExpress, Cat# HY-143216). At 72 h post-infection, the culture supernatants were collected, and viral RNA was quantified using RT-qPCR. The assay of each compound was performed in triplicate, and the 50% effective concentration (EC 50 ) was determined using Prism 9 software v9.1.1 (GraphPad Software). The viral RNA amount without treatment with antiviral drugs was set as 100%.
## SARS-CoV-2 preparation and titration
The working virus stocks of SARS-CoV-2 were prepared and titrated as previously described. In this study, clinical isolates of JN.1 (strain LG0688; GISAID ID: EPI_ISL_18771637) and XEC (strain TKYnat18145; GISAID ID: EPI_ISL_19512397) were used.
Recombinant viruses were generated by a circular polymerase extension reaction (CPER) 48 . The resultant CPER products were transfected into VeroE6/TMPRSS2 cells as described previously 13 . All the viruses were stored at -80 °C until use, and viral genome sequences were confirmed by SANGER sequencing as described above.
## Titration and growth kinetics
The infectious titers in culture supernatants obtained from the infected cells were determined by quantifying the 50% tissue culture infectious dose (TCID 50 ) 78 . For growth kinetics, the viruses were respectively inoculated into VeroE6/TMPRSS2 cells in 12-well plates at a multiplicity of infection (MOI) of 0.01. The infectious titers of the indicated timepoints were determined.
## Airway-on-a-chip
Airways-on-a-chip were prepared as previously described 38 . Human lung microvascular endothelial cells (HMVEC-L) were obtained from Lonza and cultured with EGM-2-MV medium (Lonza, Cat# CC-3202). For preparation of the airway-on-a-chip, the bottom channel of a polydimethylsiloxane (PDMS) device was first precoated with fibronectin (3 μg/ml, Sigma-Aldrich, Cat# F1141). The microfluidic device was generated according to our previous report 79 . HMVEC-L cells were suspended at 5,000,000 cells/ml in EGM2-MV medium. Then, 10 μl of suspension medium was injected into the fibronectin-coated bottom channel of the PDMS device. The PDMS device was turned upside down and incubated. After 1 h, the device was turned over, and EGM2-MV medium was added into the bottom channel. After 4 d, airway organoids (AO) were dissociated and seeded into the top channel. AOs were generated according to our previous report 80 . AOs were generated from normal human bronchial epithelial cells, not from human iPSCs. They contain ciliated, goblet, basal, and club cells. AOs were dissociated into single cells and then suspended at 5,000,000 cells/ml in the AO differentiation medium. Ten microliters of suspension medium were injected into the top channel. After 1 h, the AO differentiation medium was added to the top channel. In the infection experiments, the AO differentiation medium, containing either recombinant virus (500 TCID 50 ), was inoculated into the top channel. At 2 h.p.i., the top and bottom channels were washed and cultured with AO differentiation and EGM2-MV medium, respectively. The culture supernatants were collected, and viral RNA was quantified using RT-qPCR.
## Assessment of viral pathogenicity in hamsters
Animal experiments were performed as previously described [7][8][9][10][11][12][13][14][15][16] . In brief, Syrian hamsters (male, 4 weeks old) were purchased from Japan SLC Inc. (Shizuoka, Japan). For the virus infection experiments, hamsters were intranasally inoculated under anesthesia with the viruses (2000 TCID 50 in 100 μL for hamsters) or saline (100 μL). Body weight was recorded daily by 7 d.p.i. Lung tissues were anatomically collected at 2 and 5 d.p.i. The viral RNA load in the respiratory tissues was determined by RT-qPCR as described previously 81 . These tissues were also used for IHC and histopathologic analyses as previously described [7][8][9][10][11][12][13][14][15][16] . The viral proteins were visualized by anti-SARS-CoV-2 N monoclonal antibody (R&D Systems, Clone 1035111, Cat# MAB10474-SP, 1:400). Pathological features-including (i) bronchitis or bronchiolitis, (ii) hemorrhage with congestive edema, (iii) alveolar damage with epithelial apoptosis and macrophage infiltration, (iv) hyperplasia of type II pneumocytes, and (v) the area of hyperplasia of large type II pneumocytes-were evaluated by certified pathologists after H&E staining. Images were incorporated as virtual slides by NDP.scan software v3.2.4 (Hamamatsu Photonics). The area of N-protein positivity and inflammation was measured using Fiji software v2.2.0 (ImageJ).
To evaluate inflammation levels evoked by viral infection in hamsters, 500 μg of the lung RNA was used to synthesize cDNA with SuperScript IV VILO Master Mix (Thermo Fisher Scientific, Cat# 11756050). The resulting cDNA was used to quantify the expression of host genes 82,83 with a Power SYBR Green Master Mix (Thermo Fisher Scientific, Cat# 4367660) and QuantStudio Real-time PCR System (Thermo Fisher Scientific).
## Quantification and statistical analysis
Statistical significance was tested by one-way ANOVA with Tukey's multiple comparisons test using GraphPad Prism 9 unless otherwise noted. The values p < 0.05 were considered statistically significant ( * p < 0.05, * * p < 0.01, * * * p < 0.001, * * * * p < 0.0001). In the time-course experiments, a non-parametric permutation test was performed to evaluate the difference between experimental conditions through all timepoints. For each comparison, the area under the curve (AUC) was calculated as the sum of the values across timepoints. Group labels were randomly shuffled to generate the null distribution of AUC differences, and two-sided P values were calculated based on this distribution. Subsequently, familywise error rates (FWERs) were calculated by the Benjamini-Hochberg method. These analyses were performed in R v4.2.1 (https://www.r-project.org/). All assays were performed independently at least 3 times.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creativecommons.org/licenses/by-nc-nd/4.0/. © The Author(s) 2025 1 Department of Virology, Faculty of Medical Sciences, Kyushu University, Fukuoka, Japan. 2 Department of Microbiology and Immunology, Faculty of Medicine, Hokkaido University, Sapporo, Japan. 3 Department of Cancer Pathology, Faculty of Medicine, Hokkaido University, Sapporo, Japan. 4 Institute for Chemical Reaction Design and Discovery (WPI-ICReDD), Hokkaido University, Sapporo, Japan. 5 Department of Synthetic Human Body System, Medical Research Laboratory, Institute of Integrated Research, Institute of Science Tokyo, Tokyo, Japan. 6 Center for iPS Cell Research and Application (CiRA), Kyoto University, Kyoto, Japan. 7 Division of Systems Virology, Department of Microbiology and Immunology, The Institute of Medical Science, The University of Tokyo, Tokyo, Japan. 8 International Research Center for Infectious Diseases, The Institute of Medical Science, The University of Tokyo, Tokyo, Japan.
## References
1. Dudas, Rambaut (2016) "MERS-CoV recombination: implications about the reservoir and potential for adaptation" *Virus Evol*
2. Lau (2015) "Severe acute respiratory syndrome (SARS) coronavirus orf8 protein is acquired from SARS-related coronavirus from greater horseshoe bats through recombination" *J. Virol*
3. Holmes, Rambaut (2004) "Viral evolution and the emergence of SARS coronavirus" *Philos. Trans. R. Soc. Lond. B. Biol. Sci*
4. Bobay, O'donnell, Ochman (2020) "Recombination events are concentrated in the spike protein region of Betacoronaviruses" *PLoS Genet*
5. Turakhia (2022) "Pandemic-scale phylogenomics reveals the SARS-CoV-2 recombination landscape" *Nature*
6. Wells (2023) "The coronavirus recombination pathway" *Cell Host Microbe*
7. Suzuki (2022) "Attenuated fusogenicity and pathogenicity of SARS-CoV-2 Omicron variant" *Nature*
8. Kimura (2022) "Virological characteristics of the SARS-CoV-2 Omicron BA.2 subvariants, including BA.4 and BA" *Cell*
9. Tamura (2023) "Comparative pathogenicity of SARS-CoV-2 Omicron subvariants, including BA.1, BA.2, and BA" *Commun. Biol*
10. Saito (2022) "Virological characteristics of the SARS-CoV-2 Omicron BA.2.75 variant" *Cell Host Microbe*
11. Ito (2023) "Convergent evolution of SARS-CoV-2 Omicron subvariants leading to the emergence of BQ.1.1 variant" *Nat. Commun*
12. Tamura (2023) "Virological characteristics of the SARS-CoV-2 XBB variant derived from recombination of two Omicron subvariants" *Nat. Commun*
13. Tamura (2024) "Virological characteristics of the SARS-CoV-2 Omicron XBB.1.5 variant" *Nat. Commun*
14. Tsujino (2024) "Virological characteristics of the SARS-CoV-2 Omicron EG.5.1 variant" *Microbiol. Immunol*
15. Tamura (2024) "Virological characteristics of the SARS-CoV-2 BA.2.86 variant" *Cell Host Microbe*
16. Tsujino (2025) "Evolution of BA.2.86 to JN.1 reveals that functional changes in non-structural viral proteins are required for fitness of SARS-CoV-2" *J. Virol*
17. Bonavita, Wells, Anthony (2024) "Cellular dynamics shape recombination frequency in coronaviruses" *PLoS Pathog*
18. Dyrdak (2024) "A novel SARS-CoV-2 recombinant transmitted from a patient with an acute co-infection" *Lancet Microbe*
19. Kaku (2024) "Virological characteristics of the SARS-CoV-2 XEC variant" *Lancet Infect. Dis*
20. Wang (2024) "Recurrent SARS-CoV-2 spike mutations confer growth advantages to select JN.1 sublineages" *Emerg. Microbes Infect*
21. Arora (2024) "Impact of JN.1 booster vaccination on neutralisation of SARS-CoV-2 variants KP.3.1.1 and XEC" *Lancet Infect. Dis*
22. Liu (2024) "Enhanced immune evasion of SARS-CoV-2 variants KP.3.1.1 and XEC through N-terminal domain mutations" *Lancet Infect. Dis*
23. Li (2025) "Role of glycosylation mutations at the N-terminal domain of SARS-CoV-2 XEC variant in immune evasion, cell-cell fusion, and spike stability" *J Virol*
24. Wu (2021) "Nucleocapsid mutations R203K/G204R increase the infectivity, fitness, and virulence of SARS-CoV-2" *Cell Host Microbe*
25. Johnson (2022) "Nucleocapsid mutations in SARS-CoV-2 augment replication and pathogenesis" *PLoS Pathog*
26. Shuaib (2023) "Impact of the SARS-CoV-2 nucleocapsid 203K/ 204R mutations on the inflammatory immune response in COVID-19 severity" *Genome Med*
27. Begum (2024) "Virological characteristics correlating with SARS-CoV-2 spike protein fusogenicity" *Front. Virol*
28. Saito (2022) "Enhanced fusogenicity and pathogenicity of SARS-CoV-2 Delta P681R mutation" *Nature*
29. Yamasoba (2022) "Virological characteristics of the SARS-CoV-2 Omicron BA.2 spike" *Cell*
30. Matsuyama (2020) "Enhanced isolation of SARS-CoV-2 by TMPRSS2-expressing cells" *Proc Natl. Acad. Sci. USA*
31. El-Maradny (2024) "Unraveling the role of the nucleocapsid protein in SARS-CoV-2 pathogenesis: from viral life cycle to vaccine development" *Int. J. Biol. Macromol*
32. Wu, Cheng, Zhou et al. (2023) "The SARS-CoV-2 nucleocapsid protein: its role in the viral life cycle, structure and functions, and use as a potential target in the development of vaccines and diagnostics" *Virol. J*
33. Han (2024) "SARS-CoV-2 N protein coordinates viral particle assembly through multiple domains" *J. Virol*
34. Nguyen (2024) "Modulation of biophysical properties of nucleocapsid protein in the mutant spectrum of SARS-CoV-2" *Elife*
35. Syed (2021) "Rapid assessment of SARS-CoV-2-evolved variants using virus-like particles" *Science*
36. Syed (2022) "Omicron mutations enhance infectivity and reduce antibody neutralization of SARS-CoV-2 virus-like particles" *Proc. Natl. Acad. Sci. USA*
37. Qian (2021) "Direct activation of endothelial cells by SARS-CoV-2 nucleocapsid protein is blocked by simvastatin" *J. Virol*
38. Hashimoto (2022) "SARS-CoV-2 disrupts respiratory vascular barriers by suppressing Claudin-5 expression" *Sci. Adv*
39. Li (2021) "SARS-CoV-2 Nsp5 activates NF-κB pathway by upregulating SUMOylation of MAVS" *Front. Immunol*
40. Su, Wang, Yoo (2021) "Activation of NF-κB and induction of proinflammatory cytokine expressions mediated by ORF7a protein of SARS-CoV-2" *Sci. Rep*
41. Nishitsuji, Iwahori, Ohmori et al. (2022) "Ubiquitination of SARS-CoV-2 NSP6 and ORF7a facilitates NF-κB activation" *mBio*
42. Nie (2023) "SARS-CoV-2 ORF3a positively regulates NF-κB activity by enhancing IKKβ-NEMO interaction" *Virus Res*
43. Hua (2024) "Linear ubiquitination mediates coronavirus NSP14induced NF-κB activation" *Cell Commun. Signal*
44. Ito (2024) "Involvement of SARS-CoV-2 accessory proteins in immunopathogenesis" *Microbiol. Immunol*
45. Guo (2025) "SARS-CoV-2-specific adaptations in the N protein inhibit NF-κB activation and alter pathogenesis" *J. Cell Biol*
46. Reddy Chichili, Kumar, Sivaraman (2013) "Linkers in the structural biology of protein-protein interactions" *Protein Sci*
47. Abramson (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
48. Torii (2021) "Establishment of a reverse genetics system for SARS-CoV-2 using circular polymerase extension reaction" *Cell Rep*
49. Wickenhagen (2025) "Evolution of Omicron lineage towards increased fitness in the upper respiratory tract in the absence of severe lung pathology" *Nat. Commun*
50. Yang (2024) "Fast evolution of SARS-CoV-2 BA.2.86 to JN.1 under heavy immune pressure" *Lancet Infect. Dis*
51. Yang (2024) "Structural basis for the evolution and antibody evasion of SARS-CoV-2 BA.2.86 and JN.1 subvariants" *Nat. Commun*
52. Chakraborty, Bhattacharya, Abdelhameed (2025) "Recent SARS-CoV-2 evolution trajectories indicate the emergence of Omicron's several subvariants and the current rise of KP.3.1.1 and XEC" *Virology*
53. Stadtmüller (2022) "Emergence and spread of a sub-lineage of SARS-CoV-2 Alpha variant B.1.1.7 in Europe, and with further evolution of spike mutation accumulations shared with the Beta and Gamma variants" *Virus Evol*
54. Nguyen (2024) "Analysis of the structure and interactions of the SARS-CoV-2 ORF7b accessory protein" *Proc. Natl. Acad. Sci. USA*
55. Deshpande (2024) "SARS-CoV-2 Accessory Protein Orf7b Induces Lung Injury via c-Myc Mediated Apoptosis and Ferroptosis" *Int. J. Mol. Sci*
56. Zhang (2021) "The ORF8 protein of SARS-CoV-2 mediates immune evasion through down-regulating MHC-Ι"
57. Mcgrath (2024) "SARS-CoV-2 ORF8 modulates lung inflammation and clinical disease progression" *PLoS Pathog*
58. Su (2020) "Discovery and genomic characterization of a 382nucleotide deletion in ORF7b and ORF8 during the early evolution of SARS-CoV-2" *mBio*
59. Rogozin, Saura, Bykova et al. (2023) "Deletions across the SARS-CoV-2 genome: molecular mechanisms and putative functional consequences of deletions in accessory genes" *Microorganisms*
60. Ricciardi (2022) "The role of NSP6 in the biogenesis of the SARS-CoV-2 replication organelle" *Nature*
61. Zhang (2024) "SARS-CoV-2 NSP6 reduces autophagosome size and affects viral replication via sigma-1 receptor" *J. Virol*
62. Feng, O'brien, Chen et al. (2023) "SARS-CoV-2 nonstructural protein 6 from alpha to Omicron: evolution of a transmembrane protein" *mBio*
63. Taha (2025) "Enhanced RNA replication and pathogenesis in recent SARS-CoV-2 variants harboring the L260F mutation in NSP6" *PLoS Pathog*
64. Wang (2021) "Targeting liquid-liquid phase separation of SARS-CoV-2 nucleocapsid protein promotes innate antiviral immunity by elevating MAVS activity" *Nat. Cell Biol*
65. Long (2024) "SARS-CoV-2 N protein recruits G3BP to double membrane vesicles to promote translation of viral mRNAs" *Nat. Commun*
66. Liu (2022) "SARS-CoV-2 N protein antagonizes stress granule assembly and IFN production by interacting with G3BPS to facilitate viral replication" *J. Virol*
67. Cai (2023) "Phase-separated nucleocapsid protein of SARS-CoV-2 suppresses cGAS-DNA recognition by disrupting cGAS-G3BP1 complex" *Signal Transduct. Target. Ther*
68. Gutmann, Kuster, Hyman (2025) "SARS-CoV-2 nucleocapsid protein directly prevents cGAS-DNA recognition through competitive binding"
69. Pan (2021) "SARS-CoV-2 N protein promotes NLRP3 inflammasome activation to induce hyperinflammation" *Nat. Commun*
70. Xu (2022) "SARS-CoV-2 impairs interferon production via NSP2induced repression of mRNA translation" *Proc. Natl. Acad. Sci. USA*
71. Korneeva (2023) "SARS-CoV-2 viral protein Nsp2 stimulates translation under normal and hypoxic conditions" *Virol. J*
72. Yamamoto (2020) "The anticoagulant nafamostat potently inhibits SARS-CoV-2 S protein-mediated fusion in a cell fusion assay system and viral infection in vitro in a cell-type-dependent manner" *Viruses*
73. Khare (2021) "GISAID's role in pandemic response" *China CDC Wkly*
74. Kawashiro (2024) "Neutralizing antibody responses and cellular responses against SARS-CoV-2 Omicron subvariants after mRNA SARS-CoV-2 vaccination in kidney transplant recipients" *Sci. Rep*
75. Nasser (2022) "Monitoring fusion kinetics of viral and target cell membranes in living cells using a SARS-CoV-2 spike-proteinmediated membrane fusion assay" *STAR Protoc*
76. Kondo, Miyauchi, Matsuda (2011) "Monitoring viral-mediated membrane fusion using fluorescent reporter methods" *Curr. Protoc. Cell Biol. Chapter*
77. Hashimoto (2025) "Human iPS cell-derived respiratory organoids as a model for respiratory syncytial virus infection" *Life Sci. Alliance*
78. Reed, Muench (1938) "A simple method of estimating fifty percent endpoints" *Am.J.Epidemiol*
79. Deguchi (2021) "Usability of polydimethylsiloxane-based microfluidic devices in pharmaceutical research using human hepatocytes" *ACS Biomater. Sci. Eng*
80. Sano (2022) "Cell response analysis in SARS-CoV-2-infected bronchial organoids" *Commun. Biol*
81. Motozono (2021) "SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity" *Cell Host Microbe*
82. Bessière (2021) "Intranasal type I interferon treatment is beneficial only when administered before clinical signs onset in the SARS-CoV-2 hamster model" *PLoS Pathog*
83. Zhang (2016) "Developing a triple transgenic cell line for highefficiency porcine reproductive and respiratory syndrome virus infection" *PLoS ONE*
84. "Lin Pan 7 , Mai Suganami 7 , Mika Chiba 7 , Kyoko Yasuda 7 , Kazuhisa Yoshimura 28 , Kenji Sadamasu 28 , Mami Nagashima 28 , Hiroyuki Asakura 28 , Isao Yoshida 28" *Hiroki Futatsusako* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12866524&blobtype=pdf | # Functional characterization of a novel βglucosidase converting trillin to diosgenin from Enterococcus faecalis
Jing Li, Chen Zhou, Mengying Zhu, Qi Li, Wenjun Deng, Ningning Sun, Wenting Liu, Yawen Li, Xinyue Wang, Fan Zhang, Li Li
## Abstract
Background Diosgenin, known as "medicinal gold", is the main precursor for the synthesis of steroid hormone drugs. Currently, diosgenin is primarily produced industrially via the extraction of plant-derived steroidal saponins followed by acid hydrolysis, which generates substantial acidic wastewater.
ResultsTo address this environmental challenge, we sought to discover efficient glycosidases capable of converting steroidal saponins to diosgenin as a sustainable alternative to conventional acid hydrolysis. Through selective enrichment using dioscin as the sole carbon source, we isolated a strain Enterococcus faecalis D1 from rat gut microbiota. Subsequent genomic sequencing and functional annotation identified a potent steroid saponin β-glucosidase (designated EfD08) from E. faecalis D1, which represents the first reported bacterial β-glucosidase that efficiently converts trillin to diosgenin. EfD08 achieved soluble expression in Escherichia coli with an exceptionally high yield exceeding 200 mg/L, overcoming the low-expression bottleneck typical of reported steroid saponin glycosidases. Enzymatic characterization demonstrated that EfD08 exhibits robust activity with a half-life of approximately 100 h at its optimal temperature of 30 °C, which further extended to 355 h and 154 h in the presence of 5 mM Mg²⁺ or Mn²⁺, respectively. Furthermore, EfD08 displayed broad substrate specificity, efficiently hydrolyzing diverse steroid saponins (e.g., zingiberensis newsaponin, deltonin) and ginsenosides (e.g., Rb1, Rc).Conclusions Given its high protein yield, catalytic efficiency, and stability, EfD08 emerges as a promising biocatalyst for the environmentally sustainable industrial production of diosgenin, other secondary saponins or sapogenins from natural saponins.
## Background
Diosgenin, a pivotal precursor for steroid hormone synthesis, is clinically indispensable in treating cancer, neurodegenerative disorders, atherosclerosis, and asthma [1,2]. Diosgenin occurs naturally at trace concentrations in plants, primarily as glucose-and rhamnose-conjugated steroid saponins in species such as Dioscorea zingiberensis and fenugreek. Of note, the industrial production of diosgenin heavily relies on conventional acid hydrolysis, which generates about 600 tons of wastewater with high chemical oxygen demand (COD: 30,000-50,000 mg/L) per ton of diosgenin, causing severe environmental pollution [3]. Over 50% of China's diosgenin extraction facilities have been shut down due to noncompliance with environmental regulations. In order to maintain the healthy development of the diosgenin industry, it is urgent to develop a green and sustainable biocatalytic process for diosgenin production.
Although several glycosidases have been reported capable of hydrolyzing steroid saponins [4][5][6][7][8][9], research on these enzymes remains severely limited. To date, only a limited number of enzymes have been genetically characterized, all of which originate from eukaryotic organisms. The expression of these enzymes poses some bottlenecks for their industrial application, including the formation of inclusion bodies in prokaryotic systems (e.g., E. coli) and low expression levels in eukaryotic systems (e.g., Pichia pastoris) characterized by extended fermentation cycles and low yields [10][11][12]. The gut microbiota, comprising 10 13 -10 14 microorganisms with 1,000-1,500 species in humans alone [13], harbors unparalleled genetic diversity, particularly in carbohydrate-active enzymes (CAZymes). Notably, Bacteroides thetaiotaomicron and Bacteroides ovatus allocate an exceptionally high proportion (~ 6%) of their genes to glycoside hydrolases (GHs) and polysaccharide lyases, far exceeding the 1-3% typical of most other microorganisms [14]. GHs constitute the most abundant CAZyme class [15]. Consequently, gut microbiota also exhibits remarkable efficiency in hydrolyzing various natural saponins [16][17][18], making them a crucial genetic resource for mining steroid saponin glycosidases. For example, a β-glucosidase, BaBgl1A, was isolated and functionally characterized from Bifidobacterium adolescentis ATCC 15703, showing the ability to hydrolyze Rb1 and Rd to produce the rare ginsenosides Gyp XVII and F2 [19].
In this study, we first isolated and identified the trillinhydrolyzing bacteria from rat gut microbiota via selective culturing and 16S rRNA gene sequencing. Next, a β-glucosidase (EfD08) was annotated and confirmed to degrade trillin to diosgenin via whole-genome sequencing and enzymatic functional assays. Finally, the key biochemical properties of EfD08 were characterized. Our findings may provide a promising biocatalyst for the environmentally sustainable industrial production of diosgenin and other sapogenins from steroidal saponins.
## Materials and methods
## Materials
The following standard compounds of steroid saponins zingiberensis newsaponin (ZSA), deltonin (DL), dioscin (DC), prosapogenin A (PV), trillin (TR), diosgenin (DG), protopanaxadiol (PPD)-type ginsenosides (ginsengenin) Rb1, Rb2, Rc, Gyp XVII, Rd, F2, Rg3, Rh2, Compound O (C-O), Compound Mc-1(C-Mc1), Compound K (CK), PPD and protopanaxatriol (PPT)-type ginsenosides (ginsengenin) Rg1, Rf, Rh1, F1, PPT were purchased from Sichuan Weikeqi Biotechnology Co., Ltd (China). The CAS numbers of these compounds are listed in Table S1. Acetonitrile, methanol, and formic acid (all LC/MS grade) were purchased from Fisher Scientific (Pittsburgh, USA). All other chemicals and reagents were of analytical grade and purchased from Beijing Chemical Co., Ltd (China).
## Animals
Male Wistar-Kyoto (WKY) rats were obtained from Charles River Laboratories and had free access to food and water in a controlled environment (temperature: 23 ± 1 °C; relative humidity: 50% ± 10%). The rats were housed under the above conditions for a one-week acclimation period before experiments.
## Dioscin treatment in vivo and in vitro
For the in vivo experiments, dioscin was suspended in 0.5% (w/v) carboxymethyl cellulose sodium (CMC-Na) solution. The dosage associated with the effectiveness and safety of dioscin was based on a previous study [20]. The rats were randomly divided into two groups: (1) a control group receiving CMC-Na solution via oral gavage and (2) a treatment group administered dioscin (60 mg/ kg) via oral gavage. The animals were anesthetized using isoflurane (2.5%) after 24 h of treatment. Fresh cecal contents from the cecum of the rats were collected for 16S rRNA gene sequencing.
For the in vitro experiments, fresh cecal contents of three WKY rats were homogenized with 10 mL PBS. The 100 µL cecal suspension was added to 13 mL anaerobic medium containing 50 µM dioscin in triplicate. Then the mixtures were incubated in both aerobic and anaerobic conditions for 72 h at 37 °C. The control groups received an equivalent volume of DMSO to the treatment group. After the incubation, these samples were collected for 16S rRNA gene sequencing.
DNA extraction and sequencing were performed by BENAGEN (Wuhan, China) using the Illumina Nova-Seq6000 platform (Illumina, CA, USA). The V3-V4 hypervariable region of the 16S rRNA gene was amplified with primers 341 F and 805R for the subsequent sequencing analysis (Table S2).
## Isolation of diosgenin-producing bacteria
Fresh cecal contents from WKY rats were added to 10 mL of selective medium comprising basic medium and 50 µM dioscin. The basic medium composition was as follows: 19 [21]. After anaerobic incubation at 37 °C for 48 h, 100 µL of bacterial suspension was serially passaged seven times into fresh selective medium. Final cultures were plated on GAM agar for colony isolation. The strains were tested for their ability to degrade dioscin and trillin, and then those that degraded either compound were identified by 16S rRNA gene sequencing.
The complete 16S rRNA gene was amplified by PCR using the universal primers of 27 F and 1492R (Table S2). The 16S rRNA gene sequences of trillin-hydrolyzing bacteria were aligned using Clustal X with corresponding reference sequences from NCBI. The phylogenetic analysis was carried out with neighbor-joining criteria in MEGA 7.0. The bootstrap test was performed with 1000 repeated samples to analyze and evaluate the stability of the topological structure of the phylogenetic tree.
## Whole-genome sequencing of E. faecalis D1
The genomic DNA of E. faecalis D1 was initially extracted using E.Z.N.A®Bacteria DNA kit (OMEGA), following the Manufacturer's instructions. Whole-genome sequencing was performed using Illumina and PacBio platforms at LingEn Biotechnology Co., Ltd. (Shanghai, China). The raw paired-end reads were trimmed and quality controlled by Trimmomatic. Clean data obtained by the above quality control processes were used for further analysis. Raw PacBio reads were converted to FASTA format with Samtools Fasta ( h t t p : / / w w w . h t s l i b . o r g / d o c / s a m t o o l s . h t m l). The NGS data were used to evaluate the complexity of the genome and correct the PacBio long reads. Unicycler was used to perform genome assembly with default parameters, yielding optimal assembly results.
## Gene cloning and production of recombinant proteins
Plasmid pET28a was used as an overexpression vector to express the candidate genes in E. coli. The genomic DNA of E. faecalis D1 was used as the template for amplification, with the primers listed in Supplementary Table 1. The amplified fragments were cloned into the NcoI/ XhoI sites of pET28a using homologous recombination. The correct plasmids were introduced into E. coli BL21 (DE3), and the vector pET28a was introduced into E. coli BL21 (DE3) to construct E. coli BL21 (DE3)/pET28a as a control. The recombination strains were incubated in 100 mL LB medium with kanamycin (50 µg/mL) at 37 °C on a rotary shaker at 220 rpm. When the OD 600 reached 0.6-0.8, 0.5 mM IPTG was added to the medium to induce gene expression.
For crude enzyme extraction, cells were harvested by centrifugation and resuspended in PBS buffer, followed by ultrasonication and centrifugation to collect the supernatant. For protein purification, cells were collected and resuspended in 10 mL binding buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole, pH 7.5). Then the cells were broken by sonication, and the supernatant was collected after centrifugation at 10,000 rpm for 10 min at 4 °C. Subsequently, the supernatant was added to the balanced Ni-NTA column. Then, 15 mL solution A (50 mM NaH 2 PO 4 , 300 mM NaCl, 20 mM imidazole, pH 7.5) was added for the first wash, followed by 20 mL solution B (50 mM NaH 2 PO 4 , 300 mM NaCl, 50 mM imidazole, pH 7.5) for the second wash. Finally, 10 mL solution C (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole, pH 7.5) was added for elution, and the eluate was collected. The purified protein was concentrated using Amicon Ultra-15 ml centrifugal filter devices (Millipore, Billerica, MA, USA), and the buffer was replaced with storage buffer (10 mM phosphate buffer, pH 7.4). Then, the purified protein was detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The concentration of the purified protein was measured using the BCA method. Briefly, protein samples were mixed with BCA working reagent (1:50 ratio) and incubated at 37 °C for 30 min. Absorbance was measured at 562 nm, and concentration was calculated against a bovine serum albumin (BSA) standard curve (0-2000 µg/mL).
## Enzyme reactions and activity visualization
For testing the enzyme activity against 2 mM p-nitrophenyl-beta-D-glucopyranoside (pNPG), 10 µL of 0.02 mg/ mL recombinant enzyme was added to 70 µL Na 2 HPO 4citric acid buffer (pH 6) and incubated at 30 °C for 5 min. Enzyme activity was terminated by heat inactivation at 98 °C for 5 min. Then 100 µL 1 M Na 2 CO 3 was added, and absorbance at 405 nm was measured using a Biotek microplate spectrometer. One unit (U) of enzyme activity was defined as the amount of enzyme catalyzing the formation of 1.0 µmol p-nitrophenol per minute. The kinetic parameters of EfD08 were measured using pNPG as the substrate at concentrations ranging from 2.5 mM to 8 mM. K m , V max and k cat were calculated by fitting the activity data to a linear regression on Lineweaver-Burk double-reciprocal plots. To determine the optimal conditions and stability of enzyme activity, the effects of pH and temperature on EfD08 activity were evaluated using pNPG as the substrate. All assays were performed in triplicate.
Trillin is a monoglycosylated derivative of diosgenin, consisting of the diosgenin aglycone attached to a single glucose molecule at the C3 position. For testing enzyme activity against trillin, the recombinant enzyme sample was diluted to 5 mg/mL with 50 mM sodium phosphate buffer (pH 7). For trillin, 10 µl recombinant enzyme was added to 70 µL sodium phosphate buffer (pH 7) containing 100 µM trillin to give a total reaction volume of 100 µL, then reactions were incubated at 30 °C for 3 h. A control reaction was included and comprised 100 µL of 50 mM PBS buffer containing 100 µM trillin.
Ultra-high performance liquid chromatography with a quadrupole time-of-flight mass spectrometer (Agilent Technologies, 1290 Infinity II 6530 C) was used for qualitative analyses. Chromatographic separation was conducted on a Waters ACQUITY UPLC BEH C18 column (1.7 μm, 2.1 mm × 100 mm). The mobile phase consisted of 0.1% (v/v) formic acid in water (A) and acetonitrile (B). For the analysis of steroid saponins, the flow rate was 0.3 mL/min. The column temperature was set at 30 °C. A gradient elution program was used: 0-1 min, 5% B; 1-1.5 min, 5-25% B; 1.5-6 min, 25-75% B; 6-12 min, 75-90% B;12-16 min, 90% B. For glycosides, the mass scan was over the range of m/z 50-700 for negative ion mode with electrospray ionization (ESI) interface. For aglycones, the mass scan was over the range of m/z 100-1000 for positive ion mode with ESI interface.
## Enzyme characterization
Different pH buffers were prepared using a 50 mM Na 2 HPO 4 -citrate buffer. Enzyme activities were measured at pH 4.5-9.0 at 30 °C to investigate the effect of pH. Effects of different temperatures on enzymes were evaluated by measuring activity at pH 6 across temperatures from 4 °C to 45 °C. To evaluate the effects of metal ions on enzymes, the activities were measured in the presence of 1 mM or 5 mM Na + , K + , Zn 2+ , Ca 2+ , Cu 2+ , Mg 2+ , Mn 2+ , Ni 2+ , Fe 3+ , and ethylene diamine tetraacetic acid (EDTA). Enzymes were pre-incubated in the KH 2 PO 4 -Na 2 HPO 4 buffer at pH levels ranging from 5.0 to 9.0 at 30 °C at different times, and the residual activities were measured to study the pH stability of EfD08. Residual enzyme activities of EfD08 after pretreatment at 25-45 °C for different durations were measured to assess enzyme thermostability. Substrate specificity was studied by incubating the steroid saponins ZSA, DL, DC, PV, and the Ginsenoside Rb1, Rb2, Rc, Gyp XVII, Rd, F2, Rg3, Rh2, C-K, Rg1, Rf, Rh1, F1 with EfD08 at pH 7 and 25 °C. Ultra-high performance liquid chromatography with a quadrupole time-of-flight mass spectrometer (Agilent Technologies, 1290 Infinity II 6530 C) was used for qualitative analyses. Chromatographic separation was conducted on a Waters ACQUITY UPLC BEH C18 column (1.7 μm, 2.1 mm × 100 mm). The mobile phase consisted of 0.1% (v/v) formic acid in water (A) and acetonitrile (B). For the analysis of PPD-type ginsenosides, the flow rate was 0.2 mL/min. The column temperature was set at 30 °C. A gradient elution program was used: 0-10 min, 30%-55% B; 10-24 min, 55-90% B; 24-27 min, 90-100% B. For the analysis of PPT-type ginsenosides, the gradient elution program was used: 0-3 min, 25% B; 3-9 min, 25-55% B; 9-13 min, 55-100% B; 13-15 min, 100% B. The mass scan was over the range of m/z 50-1700 for positive ion mode with ESI interface. The results were analyzed by Agilent MassHunter Qualitative Analysis Navigator B.08.00 (Q-TOF).
## Results
## Isolation of an efficient diosgenin-producing strain from rat gut microbiota
Dioscin is composed of diosgenin linked to a trisaccharide, α-L-rhamnopyranosyl-(1→4)-[α-Lrhamnopyranosyl-(1→2)]-β-D-glucopyranoside, at its C3 oxygen (Fig. 1A). To assess the dioscin-converting efficiency of rat gut microbiota, we analyzed dioscin metabolites (Fig. 1A) in the intestinal tract of dioscinadministered rats. UPLC-Q-TOF-MS analysis of rat cecal contents (in vivo) revealed predominant accumulation of diosgenin with only trace levels of the intermediates prosapogenin A and trillin (Fig. 1B andC), which was consistent with the deglycosylation of dioscin catalyzed by cultured gut microbiota of rats in vitro (Fig. 1D andE). The predominant accumulation of diosgenin indicates that rat gut microbiota-derived hydrolases efficiently hydrolyze the glycosidic bonds of dioscin. Furthermore, 16S rRNA gene sequencing demonstrated that dioscin administration did not significantly alter Simpson diversity indices of gut microbiota in vivo and in vitro (Fig. 1F andH). At the phylum level, dioscin exerted no distinct impact on the microbial composition in vivo (Fig. 1G). However, in vitro, dioscin significantly reduced the abundance of Cyanobacteria, Proteobacteria, and Actinobacteriota, while significantly promoting the enrichment of Bacteroidota and Desulfobacterota (Fig. 1I).
To screen for dioscin-hydrolyzing bacterial strains from rat gut microbiota, we cultured rat cecal content samples in the selective medium containing dioscin as the sole carbon source. Twelve candidate strains were isolated and identified by 16S rRNA gene sequencing as Enterococcus faecalis, Enterococcus gallinarum, Escherichia fergusonii, and Escherichia coli (Fig. 2A andB). Biochemical analysis uncovered that the strains D1, D2, D8, D9, and D11 efficiently converted trillin to diosgenin, with E. faecalis D1 showing the highest catalytic activity (Fig. 2C, Figure S1). However, the corresponding metabolites were not detected when using dioscin as the substrate (Figure S2). Consequently, E. faecalis D1 was selected for the further gene mining of trillin-hydrolyzing glycosidases.
## Identification of glycosidase from E. faecalis D1
Whole-genome sequencing revealed E. faecalis D1 consists of a circular chromosome with a total length of 2,922,093 bp and four circular plasmids: D1-p1 (99,478 bp), D1-p2 (58,795 bp), D1-p3 (49,916 bp), and D1-p4 (29,628 bp), with an average G + C content of 37.6% (Fig. 3, Table S3). A total of 3,142 coding sequences (CDSs) were predicted and annotated (Table S3). Among these genes, 118 CDSs were classified as carbohydrateactive enzymes (CAZymes), including 58 glycoside hydrolases (GHs) (Table 1). Gene Ontology (GO) term analysis revealed that 715 genes could be assigned to three primary functional categories: biological processes (dominated by "cellular process" [487 genes] and "metabolic process" [412 genes]), cellular components (dominated by "cell anatomical entity" [507 genes]), and molecular functions (dominated by "catalytic activity" [363 genes] and "binding" [233 genes]) (Figure S3A). KEGG pathway analysis manifested that the majority of annotated genes were classified into metabolic pathways, particularly within global/overview maps (1028 genes) and carbohydrate metabolism (250 genes), followed by environmental information processing, genetic information processing, cellular processes, and organismal systems (Figure S3B). Additionally, Pathogen-Host Interactions (PHI) database annotation identified 88 virulence-associated genes, including 61 associated with reduced virulence, 8 with increased virulence, and 1 lethal gene (Figure S3C).
A few glycoside hydrolases hydrolyzing trillin to diosgenin have been genetically and functionally characterized. Representative examples include the β-Dglucosidase from Talaromyces stollii CLY-6 (GenBank accession no. QYF65035) [11], which belongs to the GH3 family, as well as the C3 β-glycosidase from Aspergillus fumigatus (GenBank accession no. QBZ28529) [10] and β-glucosidase FBG1 from Fusarium sp. CPCC 400,709 (GenBank accession no. QLY89363) [12], both of which are members of the GH1 family. Therefore, to identify the glycosidases responsible for trillin hydrolysis, all eight β-glycosidase genes from the GH1 and GH3 families of E. faecalis D1 were cloned and heterologously expressed in E. coli BL21 (DE3) (Figure S4, Table S4). Soluble expression at high yields was achieved for all proteins except EfD02 (Figure S4A-C, Fig. 4A). β-Glycosidase activity of crude lysates was assayed using trillin as the substrate. Among the tested enzymes, EfD03, EfD04, and EfD08 demonstrated catalytic conversion of trillin to diosgenin (Figure S4D). EfD08 exhibited the highest hydrolytic activity and was selected for further biochemical characterization (Figure S4D).
## High-level expression of EfD08 in E. coli
The EfD08 was heterologously expressed with a 6xHis tag at the N-terminus in E. coli BL21 (DE3) cells and purified using Ni-NTA resin. SDS-PAGE analysis revealed that the purified EfD08 migrated as a single band with a molecular mass of approximately 80 kDa (Fig. 4A), consistent with the theoretical value. Notably, EfD08 achieved a high soluble expression yield (238 mg/L) in 250 mL shake flasks after 16 h induction. Consistent with the crude extract, the purified protein EfD08 effectively catalyzed the conversion of trillin to diosgenin (Fig. 4B).
## Enzymatic characterization of EfD08
The optimal activity of EfD08 was observed at 30 °C (Fig. 5A), and the thermostability profile showed that EfD08 was highly stable at its optimal temperature with a half-life (t 1/2 ) of approximately 100 h (Fig. 5B). The optimum pH of EfD08 was around pH 5.5-6.0 (Fig. 5C), while EfD08 reached the maximum stability at pH 7.0. More than 90% of its maximal activity was retained after incubation at pH 7 for 12 h (Fig. 5D). Under the optimal reaction conditions (30 °C and pH 6.0), the specific activity of EfD08 was 42 U/mg. Kinetic parameters were determined from a Lineweaver-Burk plot. The analysis gave K m and V max values of 2.7 mM and 0.4 mM min -1 , respectively. The k cat was determined to be 106.1 s⁻¹, yielding a catalytic efficiency (k cat /K m ) of 39.6 s⁻¹ mM⁻¹.
Many enzymes require specific metal ions for full catalytic activity. EDTA is a crucial reagent for studying metal dependence and for maintaining enzyme stability during experiments by controlling metal ion availability. Thus, the effects of different metal ions and EDTA on the activity of EfD08 were also assayed (Table 2). The chelating agent EDTA did not affect on activity. Addition of 5 mM Ca 2+ , Na + , Mg 2+ , Mn 2+ , K + , and Ni 2+ significantly enhanced the activity of EfD08, with relative activity of 137.1%, 126.4%, 142.6%, 144.9%, 115.8% and 113.8%, respectively. Conversely, 5 mM Cu²⁺ almost completely inactivated EfD08, with a relative activity of 2.2 ± 0.1% (p < 0.0001). Fe³⁺ is also a strong inhibitor of EfD08 with a relative activity of 58% at 1 mM. Several metal ions displayed concentration-dependent biphasic effects on enzymatic activity, such as Zn²⁺ acting as an activator at 1 mM but an inhibitor at 5 mM (Table 2).
Considering that Mg²⁺ and Mn²⁺ exerted significantly positive effects on EfD08 activity, we further investigated the effects of Mg²⁺ and Mn²⁺ (5 mM) on the structural stability of EfD08. As shown in Fig. 5E and F, the thermal stability of EfD08 was significantly improved after the addition of Mg²⁺ and Mn²⁺; the half-life (t₁/₂) of EfD08 was extended from 100 h to 355 h (5 mM Mg 2+ ) and 154 h (5 mM Mn 2+ ). The pH stability of EfD08 was also enhanced by both Mg²⁺ and Mn²⁺ under alkaline conditions (pH 7-9) (Fig. 5G andH). Specifically, at pH 9.0, EfD08 retained 50% and nearly 100% residual activity after 128 h incubation with 5 mM Mn²⁺ or Mg²⁺, respectively, whereas it was completely inactivated within 24 h without these ions (Fig. 5G andH).
## EfD08 activity on the production of diosgenin
As Mg²⁺ and Mn²⁺ (5 mM) significantly enhanced EfD08 activity and stability, we systematically investigated their effects on trillin conversion efficiency at pH 7 and 30 °C by using 0.5 mg/ml enzyme with 200 µM trillin. After 72 h of reaction, Mg²⁺ and Mn²⁺ (5 mM) increased trillin conversion rates by 24.60% and 32.73% respectively, compared with the control group. Prolonging the reaction time to 108 h resulted in more pronounced improvements, with a conversion rate up to 78.31% and 52.04%, respectively (Fig. 6). Despite these marked enhancements, the absolute conversion rate remained relatively low (~ 2%). This phenomenon may be attributed to the intrinsically low hydrolytic efficiency of EfD08 toward the inner glucose moiety at the C3-position of trillin.
## Substrate specificity and biotransformation pathway
To assess the substrate specificity of the recombinant EfD08, the hydrolysis reactions were performed using steroidal saponins zingiberensis newsaponin (ZSA), deltonin (DL), dioscin (DC), and prosapogenin A (PV), protopanaxadiol-type (PPD-type) ginsenosides Rb1, Rb2, Rc, Rd, Rg3, F2, Rh2, Compound K, and Gyp XVII, protopanaxatriol-type (PPT-type) ginsenosides Rg1, Rf, Rh1, and F1 as substrates (Fig. 7A-C). EfD08 exhibits high hydrolytic activity on steroidal saponins ZSA, ginsenoside Rb1, Rb2, Rc, Rd, Rg3, and Gyp XVII, low activity on DL, Rf, and F2, and no detectable activity on DC, PV, CK, Rh2, Rg1, Rh1, and F1(Fig. 7D-F; Table 3). EfD08 specifically hydrolyzed glucosyl groups without the activity towards rhamnosyl or arabinosyl moieties (e.g., no hydrolysis of DC, PV, Rc, or Rb2) (Fig. 7D andE). EfD08 cleaved glucose moieties at C-3, C-6, and C-20 positions (e.g., using ZSA, DL, Rb1, Gyp XVII, Rd, Rg3, Rf as substrates) (Fig. 7D-F), with efficiency showing positional preference (highest activity at C-3). Notably, using F2 as substrate, only CK (hydrolysis at C-3-O-Glc) was detected, while Rh2 (expected from C-20-O-Glc hydrolysis) was absent (Fig. 7E). Additionally, EfD08 differentially cleaved Rg3 (C-3-O-diglucoside) and Rf (C-6-O-diglucoside), converting them to monoglucosides Rh2 (C-3-O-Glc) and Rh1 (C-6-O-Glc) at rates of 86.6% and 22.6%, respectively (Fig. 7E andF). Interestingly, the hydrolytic activity of EfD08 toward the C-3 glucosyl group was influenced by substituents at other positions.
For instance, the hydrolysis efficiency of the terminal C-3 glucosyl group of Rd was 100% when a glucosyl group is present at C-20. In contrast, the hydrolysis efficiency of the terminal C-3 glucosyl group of Rg3 decreased to 86.6% without a glucosyl group at the C-20 position (Fig. 7E). This allosteric modulation might suggest that the C-20 glycosyl group may influence substrate recognition at the enzyme's active site through conformational changes. Furthermore, EfD08 exhibits much higher hydrolytic activity toward terminal glucosidic bonds compared to inner glucosidic bonds. For example, the terminal glucose moieties of ZSA, Rb1, Rd, Rc, Rb1, and Gyp XVII were completely hydrolyzed, whereas hydrolysis of the inner glucose moiety was extremely inefficient (e.g., only 11.85% conversion from DL to PV and 3.16% from F2 to CK) (Fig. 7D andE). Moreover, no hydrolysis products were detected in reactions involving Rh2 (C-3-O-Glc), CK (C-20-O-Glc), or Rh1 (C-6-O-Glc), suggesting that EfD08 has limited catalytic activity toward inner glucosidic bonds for the tested substrates (Fig. 7E andF).
The UPLC-Q-TOF-MS/MS data in ESI mode, along with the fragmentation spectra of the metabolites and standards, are shown in Figure S5 and Table S5.
## Discussion
Steroids have garnered significant attention in the biopharmaceutical industry due to their diverse biological activities and physicochemical properties. Diosgenin, a crucial precursor for the synthesis of numerous steroid drugs, is traditionally produced through acid hydrolysis in industrial processes. However, this method generates substantial acid-containing wastewater and byproducts, causing severe environmental pollution that has severely hindered the development of the diosgenin industry [22,23]. With the elucidation of the diosgenin biosynthetic pathway, researchers have engineered heterologous synthesis pathways in chassis organisms such as yeast and tobacco, achieving a maximum reported yield of 2 g/L [24,25]. Nevertheless, this biosynthetic approach remains economically unfeasible for large-scale production. Enzyme-mediated conversion has emerged as a promising strategy for synthesizing bioactive compounds due to the advantages of biocatalysis, such as mild reaction conditions, high efficiency, and environmental friendliness [26]. Diosgenin predominantly accumulates in plants (e.g., Dioscorea spp.) as steroid saponins, with D. zingiberensis containing up to 16.15% diosgenin by dry weight. Coupled with cost-effective extraction techniques, enzymatic hydrolysis of saponins represents the most viable green production route. However, only a limited number of eukaryotic steroid saponin glycoside hydrolases have been characterized to date, many of which suffer from suboptimal catalytic activity and poor heterologous expression.
The gut microbiota harbors a diverse reservoir of enzymes, particularly glycoside hydrolases, which play a crucial role in the biotransformation of complex plant glycosides into bioactive sapogenins with enhanced bioavailability or specific bioactivities [27][28][29]. Although studies have reported mining novel glycoside hydrolases from gut microbiota to hydrolyze natural saponins (e.g., ginsenosides) into rare saponins and sapogenins, to the best of our knowledge, EfD08 is the first bacterial steroid saponin glycosidase cloned and functionally validated from the gut microbes. EfD08 was efficiently expressed as a soluble protein in E. coli, yielding 238 mg/L after only 16 h of induction in shake-flask culture. This expression level substantially exceeds that reported for other steroid saponin glycosidases, which often face challenges in heterologous expression. For instance, glycosidases from Trichoderma viride and Aspergillus fumigatus (C3-βGL) show negligible soluble expression in E. coli [10], while Fusarium sp. FBG1 reached only 30.1 mg/L in P. pastoris after 96 h induction in shake flasks, and even under optimized bioreactor conditions, its titer remained at 158.1 mg/L after 60 h of induction [12]. Given that highlevel soluble expression is a critical factor in reducing enzyme production costs, the robust and rapid expression of EfD08 in a common bacterial host makes it a highly promising biocatalyst for the economic conversion of trillin.
The enzymatic activity and stability of EfD08 were significantly potentiated by several metal ions, most notably Mg²⁺ and Mn²⁺. These ions markedly enhanced the enzyme's operational half-life from approximately 100 h to 355 and 154 h, respectively, at 30 °C. Correspondingly, they boosted the catalytic activity against trillin by 78.31% and 52.04%, respectively. Although there is a significant increase in activity and stability, it should be noted that the absolute trillin conversion rate remains low at 2%. The poor aqueous solubility of trillin is speculated to be one of the main reasons for the low conversion efficiency. For enzymatic reactions involving highly hydrophobic substrates, existing studies have significantly enhanced the conversion efficiency through strategies such as nonaqueous catalysis [30,31]. These approaches also merit consideration as potential directions for improving the trillin conversion of EfD08 in subsequent studies.
Intriguingly, Fe³⁺ exerted a biphasic regulatory effect on EfD08: enzymatic activity was markedly inhibited at low Fe³⁺ concentrations (1 mM), while this suppression was alleviated at elevated concentrations (5 mM). It has been reported that Ca 2+ /calmodulin-dependent protein phosphatase calcineurin (CN) exists in an inactive state on account of the occupancy of two high-affinity binding sites on the regulatory subunit calcineurin B (CNB) by the low level of Ca 2+ . Oppositely, in response to high calcium levels, the occupancy of the low-affinity sites on CNB by Ca 2+ causes dissociation of the CaM-binding domain from the B-subunit-binding helix, and the subsequent binding of calmodulin finally leads to displacement of the autoinhibitory domain from the active site, and then full CN activation [32]. However, the molecular mechanism underlying Fe³⁺'s regulation of EfD08 remains to be elucidated; future structural studies could determine whether similar autoinhibitory principles work. A variety of steroidal saponin glycoside hydrolases have been reported, exhibiting significant differences in hydrolysis specificity. For example, the glucoamylase from Curvularia lunata [4,33] and Rhase-TS from Talaromyces stollii CLY-6 [11] both demonstrate strict specificity for terminal α-1,2-linked rhamnosyl bonds, enabling them to hydrolyze dioscin into prosapogenin B rather than prosapogenin A. In contrast, protodioscin-glycosidase-1 (PGase-1) from Aspergillus oryzae shows strict specificity for α-1,4-linked rhamnosyl bonds, thereby hydrolyzing dioscin into prosapogenin A instead of prosapogenin B [34]. Compared to these, the dioscin-α-L-rhamnosidase from pig liver possesses a broader substrate range, capable of hydrolyzing both α-L-(1→2) and α-L-(1→4) rhamnosidic bonds, thus directly converting dioscin into trillin [9]. Furthermore, some enzymes exhibit multifunctional activities. For instance, the dioscin-glycosidase from Absidia sp. D38 can hydrolyze not only rhamnosidic bonds but also glucosidic bonds, thereby catalyzing the one-step conversion of dioscin into diosgenin [5]. Similarly, the β-glucosidase AfG from Aspergillus fumigatus can also hydrolyze various saponins such as dioscin, deltonin and trillin [7]. Additionally, GiGly from Gibberella intermedia WX12 exhibits high specificity for the C-3 position of multi-3-O-glycosides like dioscin and trillin, yet was completely inactive toward substrates such as ginsenoside Re and saikosaponin A possessing terminal rhamnosyl, glucosyl, or galactosyl groups [8]. In our study, EfD08 possesses a comprehensive substrate spectrum and remarkable catalytic plasticity. It exhibited hydrolytic activity toward terminal glucosidic linkages, as demonstrated by its ability to catalyze the stepwise conversion of zingiberensis newsaponin to deltonin via terminal glucose hydrolysis, deltonin to prosapogenin A, and trillin to diosgenin. Notably, D08 appeared to lack high selectivity regarding the type of glucosidic bond or the position of glucosyl attachment, as it was capable of hydrolyzing β-(1→2)-glucosidic linkages (such as those in ginsenoside Rd), β-(1→3)-glucosidic linkages (as in ZSA), β-(1→4)-glucosidic linkages (as in DL), and β-(1→6)-glucosidic linkages (as in Rb1), as well as the C3-O-glucosidic bond (as in trillin). Furthermore, EfD08 cleaved glucosyl attachments distributed across various aglycone carbon positions, such as C3 (e.g., ZSA, Rd), C6 (e.g., Rf ), and C20 (e.g., Gyp XVII). This broad specificity may stem from the enzyme's active-site structural flexibility, which enables it to accommodate diverse glycosidic configurations. This functional versatility positions EfD08 as a unique glycoside hydrolase, highly suitable for the targeted production of valuable secondary saponins. Furthermore, the broad specificity of EfD08 allows it to act synergistically with other enzymes, facilitating the efficient generation of specific saponin derivatives and their aglycones.
## Conclusion
In this study, we isolated and identified a trillin-hydrolyzing strain, E. faecalis D1, from rat cecal contents using a selective medium. Genomic sequencing and functional annotation of E. faecalis D1 identified a potent steroid saponin β-glucosidase, EfD08, which is capable of converting trillin to diosgenin. EfD08 was efficiently expressed in a soluble form in E. coli, yielding 238 mg/L, with a molecular mass of approximately 79.2 kDa. Enzymatic characterization revealed that EfD08 exhibits highly stable and robust ambient-temperature activity with a half-life of ~ 100 h at the optimal temperature of 30 °C, which extended to 355 h and 154 h in the presence of 5 mM Mg²⁺ or Mn²⁺, respectively. Furthermore, these metal ions enhanced EfD08's activity against trillin by 78% and 52%, respectively. However, despite this activation, the trillin conversion rate remained exceptionally low, below 2%. Additionally, EfD08 efficiently hydrolyzes a broad spectrum of steroid saponins (e.g., zingiberensis newsaponin, deltonin) and ginsenosides (e.g., Rb1, Rc). Taken together, EfD08 could be a promising biocatalyst for the environmentally sustainable industrial production of diosgenin or other sapogenins from steroidal saponins, owing to its remarkable yield, high stability, and long half-life.
## References
1. Arya, Munshi, Kumar (2023) "Diosgenin: chemistry, extraction, quantification and health benefits" *Food Chem Adv*
2. Semwal, Painuli, Abu-Izneid et al. (2022) "Diosgenin: an updated Pharmacological review and therapeutic perspectives" *Oxid Med Cell Longev*
3. Zhang, Tadesse, Zhan et al. (2018) "Methods to treat the industrial wastewater in Diosgenin enterprises produced from Diosorea zingiberensis C. H. Wright" *J Clean Prod*
4. Feng, Hu, Ma et al. (2007) "Purification, characterization, and substrate specificity of a glucoamylase with steroidal saponin-rhamnosidase activity from Curvularia Lunata" *Appl Microbiol Biotechnol*
5. Fu, Yu, Tang et al. (2010) "New Dioscin-glycosidase hydrolyzing multi-glycosides of Dioscin from Absidia strain" *J Microbiol Biotechnol*
6. Huang, Zhao, Lu et al. (2013) "Pathways of biotransformation of Zingiberen newsaponin from Dioscorea Zingiberensis C. H. Wright to Diosgenin" *J Mol Catal B-Enzym*
7. Lei, Niu, Li et al. (2011) "A novel β-glucosidase from Aspergillus fumigates releases Diosgenin from Spirostanosides of Dioscorea zingiberensis C. H. Wright (DZW)" *World J Microbiol Biotechnol*
8. Li, Wang, Ma et al. (2016) "Purification and characterization of a glycosidase with hydrolyzing multi-3-O-glycosides of Spirostanol saponin activity from Gibberella intermedia" *J Mol Catal B-Enzym*
9. Qian, Yu, Zhang et al. (2005) "Purification and characterization of dioscin-α-L-rhamnosidase from pig liver" *Chem Pharm Bull*
10. Huang, Wang, Fang et al. (2019) "Purification, molecular cloning and expression of three key saponin hydrolases from Trichoderma reesei, Trichoderma viride and Aspergillus fumigatus" *Nat Environ Pollut Technol*
11. Cheng, Zhang, Cui et al. (2021) "Efficient enzyme-catalyzed production of diosgenin: inspired by the biotransformation mechanisms of steroid saponins in Talaromyces stollii CLY-6" *Green Chem*
12. Liu, Xiang, Shiloach et al. (2021) "Efficient biocatalysis of trillin through Recombinant enzyme hydrolysis for clean Diosgenin production" *Process Saf Environ Protect*
13. He, Liu, Zhang et al. (2024) "Gut microbiome-derived hydrolases-an underrated target of natural product metabolism" *Front Cell Infect Microbiol*
14. Wardman, Bains, Rahfeld et al. (2022) "Carbohydrate-active enzymes (CAZymes) in the gut Microbiome" *Nat Rev Microbiol*
15. El Kaoutari, Armougom, Gordon et al. (2013) "The abundance and variety of carbohydrate-active enzymes in the human gut microbiota" *Nat Rev Microbiol*
16. Kim, Park, Kim et al. (2013) "Metabolite profiling of ginsenoside re in rat urine and faeces after oral administration" *Food Chem*
17. He, Li, Xu et al. (2015) "Pharmacokinetics, bioavailability, and metabolism of notoginsenoside Fc in rats by liquid chromatography/ electrospray ionization tandem mass spectrometry" *J Pharm Biomed Anal*
18. Dong, Yu, Pan et al. (2021) "Biotransformation of Timosaponin BII into seven characteristic metabolites by the gut microbiota" *Molecules*
19. Hu, Zhai, Hong et al. (2022) "Study on the biochemical characterization and selectivity of three β-glucosidases from Bifidobacterium adolescentis ATCC15703" *Front Microbiol*
20. Tang, Pang, He et al. (2015) "UPLC-QTOF-MS identification of metabolites in rat biosamples after oral administration of Dioscorea saponins: a comparative study" *J Ethnopharmacol*
21. Li, Liu, Wang et al. (2022) "3beta-Hydroxysteroid dehydrogenase expressed by gut microbes degrades testosterone and is linked to depression in males" *Cell Host Microbe*
22. Peng, Yang, Wang et al. (2011) "Pathways for the steroidal saponins conversion to Diosgenin during acid hydrolysis of Dioscorea zingiberensis C. H. Wright" *Chem Eng Res Des*
23. Yang, Yin, Wang et al. (2015) "In situ pressurized biphase acid hydrolysis, a promising approach to produce bioactive Diosgenin from the tubers of Dioscorea zingiberensis" *Pharmacogn Mag*
24. Xu, Wang, Chen et al. (2022) "Metabolic engineering of Saccharomyces cerevisiae for gram-scale Diosgenin production" *Metab Eng*
25. Yin, Liu, Kou et al. (2023) "Deciphering the network of cholesterol biosynthesis in Paris polyphylla laid a base for efficient Diosgenin production in plant chassis" *Metab Eng*
26. Kissman, Sosa, Millar et al. (2024) "Expanding chemistry through in vitro and in vivo biocatalysis" *Nature*
27. Niu, Smith, Yang et al. (2012) "Bioactivity and bioavailability of ginsenosides are dependent on the glycosidase activities of the A/J mouse intestinal Microbiome defined by pyrosequencing" *Pharm Res*
28. Wan, Zhang, Yuan et al. (2016) "Biotransformation and metabolic profile of Anemoside B4 with rat small and large intestine microflora by ultra-performance liquid chromatography-quadrupole time-of-flight tandem mass spectrometry" *Biomed Chromatogr*
29. Zhang, Chen, Duan et al. (2021) "Prebiotics enhance the biotransformation and bioavailability of ginsenosides in rats by modulating gut microbiota" *J Ginseng Res*
30. Li, Han, Wang et al. (2022) "Ionic liquids as a tunable solvent and modifier for biocatalysis" *Catal Rev*
31. Brogan, Bui-Le, Hallett (2018) "Non-aqueous homogenous biocatalytic conversion of polysaccharides in ionic liquids using chemically modified glucosidase" *Nat Chem*
32. Li, Wang, Ma et al. (2016) "Cooperative autoinhibition and multi-level activation mechanisms of calcineurin" *Cell Res*
33. Feng, Kang, Ma et al. (2007) "The substrate specificity of a glucoamylase with steroidal saponin-rhamnosidase activity from curvularia Lunata" *Tetrahedron*
34. Liu, Yu, Liu et al. (2013) "Protodioscinglycosidase-1 hydrolyzing 26-O-beta-D-glucoside and 3-O-(1→4)-alpha-L-rhamnoside of steroidal saponins from Aspergillus oryzae" *Appl Microbiol Biotechnol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12474201&blobtype=pdf | # Evaluation of Porcine Gastric Mucin-Based Method for Extraction of Noroviruses from Seaweed Salad
Philippe Raymond, Sylvianne Paul, Roxanne Blain, Neda Nasheri
## Abstract
Human noroviruses (HuNov) are the major cause of foodborne illness globally. Several HuNoV outbreaks have been linked to contaminated ready-to-eat seaweed products. Standard protocols such as the ISO 15216 show limited efficiency in extracting foodborne viruses from seaweed products. Therefore, we evaluated the efficiency of an extraction protocol based on porcine gastric mucin conjugated magnetic beads (PGM-MBs) to recover HuNoVs from Wakame seaweed salad. Compared to other HuNoV extraction methods, the PGM-MB method was more efficient. We then aimed to further improve this protocol by modifying several factors such as the buffers, pH, bead concentration, centrifugation and incubation time. The optimized PGM-MB method yielded 19 ± 3% and 17 ± 4% recovery, for HuNoV GI and GII, respectively. The limit of detection (LOD 95 ) for Wakame seaweed salad was 131 and 56 genomic equivalents per 25 g for HuNoV GI and GII. Although some variability in recovery efficiency was observed between the PGM sources, the optimized PGM-MB protocol effectively extracts HuNoVs from Wakame seaweed salads of various brands and other commodities such as dates, green onions, and salted seaweed. These results support the implementation of the optimized PGM-MB method as a viable alternative for HuNoV surveillance in complex food matrices.
## 1. Introduction
Human noroviruses (HuNoV) are the leading cause of foodborne outbreaks [1][2][3]. According to the Public Health Agency of Canada, 65% of known causes of foodborne illnesses in Canada are related to HuNoV [4]. In the United States, HuNoV infections account for 35% of single-etiology foodborne outbreaks and 46% of foodborne illnesses [5].
Noroviruses are small (27-40 nm), non-enveloped, single-stranded RNA viruses that belong to the Caliciviridae family. There are 10 distinct norovirus genetic groups [6], and most outbreaks are associated with HuNoV GI and GII. They are mainly transmitted via the fecal-oral route. These viruses can withstand a broad pH range (pH 2-9) and remain infectious for extended periods in the environment, water, on food, and surfaces (reviewed in Olaimat et al. [7], Cook et al. [8]). HuNoV can contaminate food either at the point of harvest or during food handling and processing [9].
HuNoVs are inactivated by cooking, but may remain infectious after chilling, desiccation, and drying [8,10]. While fresh and frozen produce are common vehicles for norovirus infections, HuNoVs can also be detected and remain infectious on low-moisture foods such as sun-dried tomatoes, palm dates, and dried seaweeds [11][12][13][14].
HuNoV-contaminated seaweeds have been implicated in multiple outbreaks, including an outbreak associated with seasoned green seaweeds involving 91 cases [15], and another large outbreak associated with shredded nori infecting 2094 patients [11]. Frozen Goma Wakame seaweed salads imported from China were recalled following an outbreak of HuNoV GI and GII in both Norway (100 cases) and Spain [16]. In 2024, five cases were reported in Italy, where the Ministry of Health reported the detection of HuNoV GII in Goma Wakame seaweed salad from Taiwan and issued a recall [17]. Nori is usually made from red seaweed species of the genus Pyropia, while Wakame is a large brown seaweed belonging to the order Laminariales. An FAO expert meeting on food safety concluded that HuNoVs should be considered potential microbiological hazards in seaweeds [18]. Seaweed farming in sewage-contaminated water and improper handling are likely the main routes of viral contamination [19].
The International Organization for Standardization (ISO) has established the ISO 15216 standard for the extraction and quantitative (15216-1) or qualitative (15216-2) detection of HuNoV from various soft fruits and leafy greens, stems, and bulb vegetables, based on polyethylene glycol (PEG) precipitation [20]. Since the probability of obtaining falsenegative results is higher when extraction methods have low recovery rates and/or high inhibition, the ISO 15216 standard requires a minimum extraction efficiency of 1% and a maximum RT-qPCR inhibition rate of 75%. The ISO 15216 methods have not been validated for seaweed. In preliminary assays performed using the ISO 15216-1 protocol to extract HuNoV from seaweed salad (Goma Wakame), our laboratory observed gelation of the extract following the addition of guanidine thiocyanate-based lysis buffer, which precluded the elution process (unpublished). Recently, a survey was conducted to assess the suitability of the ISO 15216 standard method for HuNoV detection in seaweeds and halophytes using samples collected along the western coast of Portugal [21]. Using the Mengo virus as a process control to evaluate the extraction efficiency, these authors were able to process only half of the brown seaweed extracts. Consequently, they recommended optimizing the HuNoV extraction protocol for brown seaweed.
A potential solution to improve the selective extraction of HuNoVs at low concentrations is the use of ligands. HuNoVs bind to the fucose residues of histo-blood group antigens (HBGAs) expressed on the human gastrointestinal mucosa [22][23][24]. They also bind to porcine gastric mucin (PGM), which contains fucose residues similar to those of A and H antigens [25,26]. A high norovirus GII.4 recovery rate (47 ± 8%) from fresh green seaweed (Enteromorpha spp.) was reported using an approach based on PGM conjugated magnetic beads (PGM-MBs) [27]. However, the reported PGM-MB protocol needs to be adapted for the larger food portions typically used in food surveillance laboratories, as well as being tested on different varieties of seaweed-containing commodities such as seaweed salads.
Here, we report on the performance of a PGM-MB extraction protocol for Wakame seaweed. We evaluated and adapted the PGM-MB methodology reported in Suresh, Harlow and Nasheri [27] to improve its sensitivity for 25 g analytical portions, and conducted a fit-for-purpose validation for the extraction of HuNoV from Wakame seaweed salads. We also evaluated the impact of different factors on the recovery rates. Finally, we explored the potential application of this optimized PGM-MB approach for other ready-to-eat seaweedbased products and food matrices. This method should improve future food surveillance efforts and help mitigate potential food outbreaks linked to HuNoV-contaminated seaweed salads.
## 2. Materials and Methods
## 2.1. Viruses
HuNoV-positive stool samples were provided by the British Columbia Center for Disease Control. Clarified 10% stool samples were prepared as described previously [28]. Unless otherwise specified, the HuNoV GII.4 (CFIA-FVR-020, MT754281.1) and GI.5 (CFIA-FVR-022, OL345567.1) were used in the method optimization and validation. The genomic equivalent (gEq) levels of the virus aliquots were estimated following their extraction using the QIAcube and the RNeasy mini kit (Qiagen, Mississauga, ON, Canada) as described before [28]. The RNA was eluted in 50 µL RNase-free water, 1 µL of RNasin™ Plus RNase Inhibitor (Thermo Fisher, Asheville, NC, USA) was added, and the RNA was stored at -80 • C until the RT-qPCR assays were performed as described below.
In preliminary experiments, to compare the RNA extraction efficiency, the virus in 50 µL ultrapure water was heated at 100 • C for 10 min, then cooled down to 0 • C on a thermocycler and stored as described above.
## 2.2. Seaweed Samples
Samples of Wakame seaweed salads were collected from various local stores in Quebec, Canada. Unless otherwise indicated, brand A was used for method development and validation. Subsamples were prepared with 25 g ± 1 g of seaweed salad. The ingredients of the different seaweed salad brands tested are listed in Supplementary Table S1.
## 2.3. Artificial Inoculation
Unless otherwise specified, virus preparation aliquots were vortexed for 2 s and diluted in Wisent-PBS (0.005 M Na 2 HPO 4 , 0.001M KH 2 PO 4 , 0.15 M NaCl, pH 7.4, Wisent, St-Bruno, QC, Canada) at various gEq levels. One hundred µL of the diluted aliquots were spiked throughout the surface of the seaweed salad in a Whirl-Pak ® filter bag (VWR, Mont-Royal, QC, Canada), then left to air dry for 30 min in a biosafety cabinet. Unspiked seaweed salad subsamples were included in each extraction batch as negative controls. The amount of virus inoculated was assessed in parallel using the same RNA extraction kit as for the spiked matrices.
## 2.4. PGM-MB Preparation
Unless otherwise specified, Type III mucin from porcine stomach (catalog no. M1778, MilliporeSigma, Oakville, ON, Canada) was cross-linked to MagnaBind carboxyl derivatized beads (catalog no. 21353, Thermo Fisher) using EDC (1-ethyl-3-[3dimethylaminopropyl] carbodiimide hydrochloride) (Thermo Fisher), according to the manufacturer's protocol. The PGM-MBs were resuspended in 1 mL of PBS (0.05 M Na 2 HPO 4 , 0.05 M NaH 2 PO 4 , 0.15 M NaCl, pH 7.2, Fisher Scientific, Nepean, ON, Canada) with 0.05% sodium azide (MilliporeSigma) and stored at 4 • C.
## 2.5. Extraction Methodologies 2.5.1. ISO 15216-1
The spiked viruses were extracted using the ISO 15216-1:2017 protocol for leafy greens and soft fruits [20]. The RNA was extracted with the NucliSens miniMAG kit (Biomérieux, Montréal, QC, Canada), following the manufacturer's instructions. One µL of RNasin Plus RNase Inhibitor (40 U/µL) (Promega, Madison, WI, USA) was added to the eluate before storage at -80 • C.
## 2.5.2. Magnetic Silica Beads
Viral RNAs were also extracted from Wakame seaweed salads using the magnetic silica beads (MSB) method as described before [28]. Briefly, the HuNoVs were eluted from the matrices using a Bis-Tris Propane buffer pH 8 (MilliporeSigma). The eluate was clarified by centrifugation, and pectinase was added. Magnetic silica AccuNanobeads™ (Bionneer, Oakland, CA, USA), ascorbic and malic acid (MilliporeSigma) were added. Next, HCl was added to lower the pH to 3 to bind the virus to the silica. The captured viruses were eluted from the beads by increasing the pH to 7-9. The total RNA was extracted using the RNeasy kit as described above.
## 2.5.3. Reference PGM-MB Protocol
The protocol described by Suresh, Harlow and Nasheri [27] was used as a reference protocol for the development of the PGM-MB methodology (Figure 1). Briefly, 25 g of the inoculated food matrices were incubated with 40 mL PBS pH 7.2 and incubated at room temperature (RT) for 30 min on a rocking plate. The viral eluate was removed, and either 1 or 40 mL was incubated with 0.1 mL of PGM-MBs on a rotary platform for 30 min at RT. The 1 mL aliquots were washed using a magnet as described before [27]. The 40 mL supernatant was discarded using a 50 mL magnet rack, and the PGM-MBs were washed three times with 1 mL PBS. The total RNA was extracted from the PGM-MBs using the RNeasy kit as described above.
## 2.5.4. PGM-MB Protocol Optimization
The impact of four different matrix elution buffers on the recovery rates was compared using PBS, TGBE buffer pH 9.5 (100 mM Tris base, Fisher scientific; 50 mM glycine, MilliporeSigma; 1% beef extract Thermo Fisher) with or without 0.05% Triton™ x-100 (Thermo Fisher), and a Tris/Glycine/pectinase buffer (100 mM Tris base; 50 mM glycine; 30 U Aspergillus niger pectinase, MilliporeSigma) with 0.05% Triton. Forty ml of elution buffer was added to the matrix at RT for 30 min on a rocking plate at 60 rpm. The viral eluate was removed and centrifuged for 30 min at 10,000× g at 5 • C. The pH was adjusted to 7.0 ± 0.5 with 5N HCl. The eluate was incubated with 0.1 or 0.2 mL of PGM-MBs on a rotatory platform at 8 rpm for 120 min at RT. Then, the eluate was incubated for 10 min on large magnetic racks, and the supernatant was discarded. The PGM-MBs were washed once with the respective elution buffer at pH 7. The total RNA was extracted from the PGM-MBs using the RNeasy kit as described above.
## 2.5.5. The Optimized PGM-MB Protocol
The inoculated food samples were incubated with 40 mL of TGBE buffer pH 9.5 with 0.05% Triton™ x-100, 30 min at RT, on a rocking plate at 60 rpm (Figure 1). The viral eluates were removed and centrifuged for 10 min at 10,000× g at 5 • C. The pH was adjusted to 7.0 ± 0.5 with 5 N HCl, and the eluates were incubated with 0.2 mL of PGM-MBs on a rotatory platform at 8 rpm for 120 min at RT. The eluates were then incubated for 10 min on large magnetic racks before discarding the supernatant. The PGM-MBs were washed once with 1 mL TGBE pH 7 plus 0.05% Triton™ x-100. To elute the RNA, the PGM-MBs were suspended in 500 µL RLT buffer (Qiagen) with 2% dithiothreitol (Thermo Fisher), incubated for 2 to 5 min on the magnet rack, and extracted using the RNeasy mini kit as described above.
## 2.6. RT-qPCR Detection
HuNoV GII and GI RT-qPCR assays were carried out using the TaqMan Fast Virus 1-Step Master Mix (Thermo Fisher) as described previously [28]. Briefly, HuNoV GII RT-qPCR was performed using QNIF2d and COG2R primers and the probe QNIFS. HuNoV GI RT-qPCR was performed using QNIF4 and NV1LCR primers with the TM9 probe.
## 2.7. Recovery Rates
The recovery rates associated with the virus elution and concentration steps were estimated using the cycle threshold (Ct). The virus recovery rate = 10 (∆Ct/m) × 100% where ∆Ct is the Ct value of extracted viral RNA from the matrix minus the Ct value of viral RNA extracted from the inoculum, and m is the slope of the virus RNA transcript standard curve [20].
## 2.8. Sensitivity
The proportion of positive observations for each concentration was used to assess the probability of detection (POD) and calculate the limit of detection (LOD 50 and LOD 95 ) and the confidence intervals (CI95%) with the PODLOD program (v9) [29]. Briefly, at each inoculum level, three to five 25 g seaweed salad subsamples were spiked with HuNoV GII.4 or GI.5 at concentrations ranging from ~10 1 to 10 3 gEq per 25 g. The spiked subsamples were extracted using the optimized PGM-MB protocol.
## 2.9. Inhibition Measurement
HuNoV GII RNA transcripts with insert prepared in our laboratory were used to estimate the level of inhibition as described before, by spiking a known concentration of those transcripts in the negative Wakame seaweed salad RNA extract [30].
## 2.10. Robustness 2.10.1. Analyst
The impact of different analysts performing the experiments on the optimized PGM-MB recovery rates was evaluated using various Wakame seaweed salad brand A samples spiked with HuNoV GI and/or GII at 10 3 gEq per 25 g.
## 2.10.2. PGM-MB Preparation
Different PGM lots and brands were compared using the optimized PGM-MB method. The Biovenic Native Porcine Stomach Mucin (MUC) (Biovenic, Hauppauge, NY, USA) was compared to the MilliporeSigma Type III mucin.
In addition, the impact of PGM-MBs aging on HuNoV GI and GII recovery rates was analyzed with the Wakame seaweed salad brand A, using a PGM-MB preparation aged over a period of 16 days for GI and 48 days for GII, respectively.
## 2.10.3. Competitor
The recovery rates of the optimized PGM-MB method when testing both HuNoV GI and HuNoV GII in competition were assessed by spiking these two viruses at various concentrations (10 3 to 10 4 gEq per 25 g) on the Wakame salad brand A.
## 2.10.4. Genotype
Because the PGM-MB method requires HuNoV capsid binding, the recovery rates of a collection of different HuNoV GI (4) and GII (3) genotypes were tested in triplicate at ~10 3 gEq with the optimized PGM-MB method.
## 2.10.5. Seaweed Salad
Wakame seaweed salads contain multiple ingredients, which could vary between different commercial brands (Supplementary Table S1). The impact of the Wakame salad brand on the optimized PGM-MB recovery rates was evaluated using five different brands spiked with HuNoV GI and GII at ~10 3 gEq per 25 g. Non-spiked control samples were tested in parallel.
## 2.11. Matrix Extension
A preliminary analysis of HuNoV GII.4 recovery from other Wakame seaweed types (dry Wakame, salted Wakame salad), and other food matrices (green onion, dates), was performed using 25 g of food matrices spiked at ~10 3 gEq per 25 g.
## 2.12. Statistical Analysis
One-Way ANOVA followed by a Tukey pairwise comparison was used to compare the extraction methodology recoveries (95% CI). The Spearman correlation statistic was used to evaluate the impact of the PGM-MBs aging (p < 0.05).
## 3. Results
## 3.1. Reference Protocol Recovery Rates
The recovery efficiency of several published viral extraction protocols was evaluated using Wakame seaweed salad (Figure 2). The recovery efficiency was calculated as the ratio of the total viral genome copies recovered to those initially inoculated. Preliminary tests, performed either with inoculum alone or in the absence of matrices, showed that RNA recovery rates using the RNeasy kit were higher than those obtained with the boiling technique (Supplementary Tables S2). Consequently, the RNeasy kit was selected for total RNA extraction from the inoculum and subsequent PGM-MB experiments. The tested HuNoV GII inoculum level ranged from 5 × 10 2 to 6 × 10 3 gEq. The HuNoV GII recovery rates with both the leafy green and soft fruit protocols of the ISO 15216-2017 method were low, with undetected samples and recovery rates of 1.2 ± 1.1% (2/3) and 0.1 ± 0.2% (4/9), respectively. The magnetic silica beads (MSBs) protocol, which was developed previously for leafy greens and berries, was similarly inefficient with a HuNoV GII recovery rate of 0.5 ± 0.3% (n = 6) [28,30]. In these preliminary assays, the reference PGM-MB protocol (1:40 mL) [27] was also applied to 25 g of seaweed salad. A 1 mL aliquot of eluate was extracted using 0.1 mL of PGM-MBs, yielding an estimated 45% recovery of the inoculated virus. Since this aliquot accounted for only 2.5% of the total eluate volume, the overall recovery rate from the inoculated matrix was 1.14 ± 0.04%.
## 3.2. PGM-MB Optimization
## 3.2.1. Buffer Selection
We next aimed to optimize the PGM-MB method by testing whether the ISO 15216 buffer (TGBE, pH 9.5) could enhance viral extraction compared to PBS (pH 7.2), as used in the reference PGM-MB protocol [27]. Recently, Bai et al. [31] also recommended using 0.05% Triton X-100 in a Tris/Glycine/pectinase elution buffer for virus recovery from strawberries. In this set of experiments, we compared the impact of the different elution buffer compositions, with or without 0.05% Triton X-100, on HuNoV GII recovery rates, using 200 µL of PGM-MBs (Figure 3). Using PBS, the HuNoV GII recovery rate was below the 1% threshold, at 0.3 ± 0.2% (n = 3). Both the traditional TGBE without Triton and the Tris/Glycine/pectinase buffer with Triton yielded recovery rates above the 1% threshold at 3 ± 2% (n = 3) and 2 ± 1% (n = 3), respectively, with no significant difference between them (p = 0.578). In contrast, the HuNoV GII recovery rate from the seaweed salad, using the TGBE buffer with Triton, reached 8 ± 1% (n = 3), which is significantly higher than other buffers (p = 0.000). Therefore, the TGBE buffer with Triton was selected for subsequent optimization steps.
## 3.2.2. The PGM-MB Ratio
We next evaluated the effect of the PGM-MB volume on the HuNoV recovery. When using the spiked TGBE buffer with Triton eluate from Wakame seaweed salad, the difference in recovery rates between 200 µL and 100 µL of beads (37% vs. 31%) was not statistically significant (p = 0.052), although a slight improvement was noted (Table 1). Therefore, 200 µL of beads was used in the remaining experiments.
Table 1. The effect of PGM-MB volumes on HuNoV recovery from spiked Wakame seaweed salad extract. The Wakame seaweed salad was incubated with TGBE with Triton buffer as described in the optimized PGM-MB reference protocol. After centrifugation, the extract was spiked with HuNoV.
## Beads Volume (µL)
HuNoV GII
## Inoculum
## 3.2.3. The pH Adjustment
The impact of adjusting the pH to improve the HuNoV binding to PGM-MBs was also evaluated based on the work by Tian, Brandl and Mandrell [25], who reported a significant increase in recovery at pH 3.5. For this reason, we examined whether lowering the pH to 3.5 would improve the recovery rates. As shown in Table 2, the recovery rate for HuNoV GII was significantly higher at pH 7.2 than at 3.5 (p = 0.0069).
## 3.2.4. Optimization of the Centrifugation Time
We next examined whether extending centrifugation time prior to pH adjustment and beads addition would enhance the recovery rate. Using the spiked Wakame seaweed salad, the HuNoV GII recovery rates were higher at 16 ± 3% (n = 5) with a 10 min centrifugation at 10,000× g compared to 11 ± 4% (n = 5) with a 30 min centrifugation (p = 0.040) (Table 3), which demonstrates that extending centrifugation time would negatively affect viral recovery.
## 3.2.5. Virus Incubation Time
No statistically significant difference in HuNoV GII recovery rates (17% vs. 19%) was observed when the spiking contact time and temperature (30 min at room temperature vs. 48 h at 4 • C) were compared (Table 4). On the other hand, a decrease in recovery was observed with HuNoV GI after 48 h incubation at 4 • C (p = 0.014). Therefore, a 30 min contact time at room temperature was selected for the remaining experiments.
## 3.2.6. PGM-MB Aging
The effect of PGM-MB preparation aging on method performance was analyzed by comparing the recovery rates for HuNoV GI and HuNoV GII from Wakame seaweed salads over several weeks (Supplementary Figure S1). HuNoV GII recovery was inversely correlated with PGM-MB aging, with an estimated half-life of 69 days (p = 0.000). The HuNoV GI recovery rate did not correlate with the PGM-MB preparation aging in the time frame tested (16 days, p = 0.364). Additional data covering a longer time frame are required for evaluating the PGM-MBs aging impact on HuNoV GI recovery.
Without considering the effect of aging on the tested PGM-MB preparations, the average recovery rates for HuNoV GI and GII from Wakame seaweed salads during the study were 15 ± 5% (n = 24) and 14 ± 5% (n = 38), respectively.
## 3.2.7. PGM Sources
The impact of the PGM lot and type on HuNoV GII recovery rates from seaweed salad using the optimized PGM-MB protocol was explored with two different PGM type III lots from MilliporeSigma and a native PGM from a second provider Biovenic (Table 5). Upon adding EDC to the mixture of PGM and magnetic beads, the Biovenic PGM appeared more viscous than the MilliporeSigma PGM. Nevertheless, for both PGM types, the recovery rates were above 5%. However, the recovery rate using PGM-MBs with type III mucin lot A was significantly higher than with lot B or native mucin (p = 0.006). The type III mucin lot A was used throughout this study.
## 3.3. Validation
## 3.3.1. LOD
The LOD 95 values for the optimized PGM-MB extraction method from Wakame seaweed salad for HuNoV GI and GII were estimated at 131 gEq per 25 g (CI95: 56-311) and 56 gEq per 25 g (CI95: , respectively (Supplementary Figure S2). The LOD 50 values for HuNoV GI and GII were 30 gEq per 25 g (CI95: 13-72) and 13 gEq per 25 g (CI95: 4-45), respectively.
## 3.3.2. Inhibition
Using the optimized PGM-MB negative seaweed salad control matrix (brand A) RNA extracts spiked with the HuNoV GII transcript as an external amplification control, the RT-qPCR inhibition level was estimated at 13 ± 13% (n = 12). The inhibition rates ranged from -1% to 40%.
## 3.3.3. Robustness Inter-Analyst Repeatability
When we compared the variation between two or three analysts in the HuNoV GI and GII recovery rates using the optimized PGM-MB method from Wakame seaweed salad, no statistically significant difference was observed (p = 0.09) (Table 6).
## Analyst
## Competition
The optimized PGM-MB HuNoV GI and GII recovery rates from Wakame seaweed salad remained similar to the average recovery rate, whether the two viruses were spiked together or individually (Table 7). The 1 log concentration difference in the inoculum during the competition experiment had no statistically significant impact on the recovery rates (p = 0.089).
Table 7. The effect of competition between HuNoV GI and GII on their recovery rates from Wakame seaweed salads using the optimized PGM-MB protocol.
## Inoculum Description
## HuNoV
## Genotype
Since the efficiency of the PGM-MB method is influenced by the capsid structure, we also evaluated its robustness with various HuNoV genotypes from genogroup I and genogroup II samples available in our collection (Table 8). In this assay, the within-genotype coefficient of variation was 27% for HuNoV GI and 70% for HuNoV GII. The average HuNoV GI recovery rate was 17 ± 5 gEq per 25 g. Except for GII.3, which was significantly lower, the average HuNoV GII recovery rate was 13 ± 5 gEq per 25 g (p = 0.000).
## Seaweed Salad Brands
The recovery efficiency was evaluated using various brands of Wakame seaweed salad (Table 9). The HuNoV GI and GII recovery rates using the optimized PGM-MB method ranged from 9 to 28% with an average of 16% ± 8%. A negative control from brand D (1/3) was also positive for GII with a Ct value of 43.8. However, the amplicon could not be sequenced as the contamination level was below the method LOD. The composition of seaweed salad might also impact the recovery rate as the ingredient lists of brands B and D were nearly identical, and both showed a lower recovery rate at 9% for HuNoV GII compared to the other tested brands (Supplementary Table S1) (p = 0.004). There was a statistically significant difference in HuNoV GI recovery between brands (p = 0.000).
## 3.4. Extension
The efficiency of the optimized PGM-MB method was tested on additional high-risk matrices. The HuNoV GII recovery rates from 25 g of whole Medjool dates, green onion, and salted seaweed salad were above 10% (Figure 4). The average inhibition was 15 ± 12% for green onion. The recovery from dry Wakame was around the 1% threshold, at 1.4 ± 0.7% (n = 5). In these samples, 25 g of dry Wakame absorbed most of the 40 mL elution buffer, and the inhibition was relatively high at 72 ± 21%, with three out of five samples showing inhibition rates above 75%.
## 4. Discussion
Edible seaweed is available on the market in various formats, including dried or dehydrated forms (e.g., Nori sheets or flakes), semi-dried, fresh (refrigerated), preserved in water or brine, frozen, ready-to-eat products (e.g., Wakame salad), and fermented preparations. The reference PGM-MB protocol from Suresh, Harlow and Nasheri [27] was originally developed for fresh green seaweed sold in water. Dried seaweed typically represents a single ingredient matrix consumed in small portions and tends to absorb large volumes of buffer during rehydration. In contrast, the Wakame seaweed salad is a complex matrix with multiple ingredients such as oil, sugar, gum, and seeds. Additional optimization of the reference PGM-MB protocol was required. First, the portions tested were increased to 25 g, which is a common analytical portion [20]. Second, we observed that twice as much virus could be detected using the BOOM method compared to the boiling approach (Supplementary Tables S2) [32]. While this might not have a major effect on the recovery rate, which is based on a ratio, it could impact the LOD estimates. Therefore, the RNeasy extraction kit was employed for RNA extraction. A third modification was to analyze the entire eluate using the PGM-MB method instead of testing 1 mL portions. Analyzing only a fraction of the food matrix reduces the method's sensitivity and may result in an overestimation of the total virus recovery rates [33]. When using the same ratio of the PGM-MBs to a 1 mL aliquot of 40 mL eluate from the Wakame seaweed salad, we obtained a recovery rate of 45%, which was similar to the 47 ± 7.8% values reported previously [27].
The other reference protocols tested also showed limited efficiency with the Wakame seaweed salad. The PEG-based precipitation method (ISO 15216) and silica-based approaches showed limited recovery efficiency, close to or below 1%. The high concentrations of polyphenolics and polysaccharides (30-50% alginate) in brown seaweeds may have interfered with the virus recovery [34]. Moreover, the Wakame seaweed salad is a complex matrix containing multiple ingredients. The ISO 15216 reference protocol might perform better with different types of seaweed matrix. For instance, the Mengo virus RNA recovery efficiency was validated (>1%) for the majority of green and red macroalgae samples tested using the ISO 15216 protocol [21].
Several factors that could impact the recovery yields were also evaluated. Although the buffer composition was not investigated in detail, the addition of Triton significantly improved the performance of the TGBE buffer (pH 9.5) with the Wakame seaweed salad. The addition of a non-ionic surfactant may facilitate virus elution and dispersion, thus enhancing virus recovery [31]. In contrast to our findings, Bai, Pu, Suo, Zhang, Qu, Feng, Huang, Shao and Dai [31] reported a negative impact of beef extract on virus recovery from strawberries, and a 40-fold higher recovery with Tris/glycine/pectinase buffer pH 9.5 with Triton compared to TGBE with Triton (pH 9.5). Other differences, such as bead preparation, RNA extraction, or the presence of inhibitors, might explain the discrepancy in recovery compared to our study. Another important parameter is pH. The pH could affect the virus conformation and its binding to PGM-MBs. When evaluating the binding of HuNoV GII to PGM-MB without matrix, Tian et al. [35] reported a 3-4-fold increase at pH 3.6 compared to pH 7.2. In contrast, using spiked Wakame seaweed salad, in our setting, the HuNoV recovery rates increased by 1.5 to 4-fold using TGBE with Triton at pH 7.2 compared to pH 3.5. Additional differences from our study, such as the absence of matrix, buffer composition (citrate and PBS), and RNA extraction protocols, may account for the variation in recovery.
Although the different types and lots of PGM tested with the optimized PGM-MB protocol allowed an efficient HuNoV recovery (>5%), i.e., above the 1% threshold, the mucin lot variability had a noticeable impact on recovery. A decrease of approximately 3-fold in HuNoV GII recovery rate was observed between two mucin Type III lots in the same experiment. Mucin lot A, used throughout the study, consistently yielded recovery rates close to 15% with the optimized PGM-MB protocol. However, the mucin type III lot B yielded a HuNoV GII recovery rate of 5%, comparable to that of a generic mucin product lacking sialic acid specification. Accordingly, this variability should be considered as a quality control factor when PGM-MB assays are implemented.
The optimized PGM-MB protocol provided a high recovery ranging from 5% to 23%, and a low LOD 95 , close to 100 gEq, with inhibition rates below 40%. There was no significant impact of competition when both HuNoV GI and GII genogroups were co-inoculated. The assay performance was consistent across different brands and analysts. The genotype impact on the recovery rates was limited. Only one of the eight genotypes tested had a low recovery close to the 1% threshold. Genotype GII.3 is also known to be more challenging to cultivate in human intestinal enteroids [36]. A different buffer composition or additional compounds such as bile acids and ceramides might improve the mucin binding. While the ISO 15216 standard protocol does not exhibit the same genotype limitation, its performance was not superior in our assays using HuNoV GII.4. On the other hand, Hepatitis A Virus (HAV), another important human foodborne virus that is frequently surveyed with norovirus, binds to a different type of receptor than HuNoV. It uses a cellular receptor called Hepatitis A Virus Cellular Receptor 1 (HAVCR1) to enter host cells [37]. PEG-based precipitation, ultracentrifugation, silica beads, and other approaches might be more suitable for HAV extraction [38,39]. However, their performance, regarding the HAV recovery rates from the Wakame seaweed salad, remains to be evaluated.
Other PGM-MB protocols have been used previously to extract norovirus from green seaweed, lettuce, berries, and green onions [27,31,33,40]. We tested the performance of the optimized PGM-MB protocol with a limited set of samples from dates and green onions, and other seaweed products spiked with HuNoV GII. The recovery from the dates was acceptable, but 3-fold lower than the recovery yields achieved with ISO 15216 leafy green protocols [14]. In contrast, with spiked green onion, the HuNoV GII recovery rates were approximately 4-fold higher than those reported using a PEG precipitation protocol or another PGM-MB approach based on a pH 3.6 citrate binding buffer [40]. The low inhibition observed with HuNoV extracted from green onion is promising. In our hands, the ISO 15216 protocol was frequently associated with high inhibition levels when HuNoV, or the process control MNV, was extracted from green onion (data not shown).
The food portion sizes used in standard virology and microbiology assays are designed to reflect typical consumer exposure and to facilitate method comparison and result interpretation. For some low-moisture foods that require rehydration, dried matrices or spices, the consumers are certainly not exposed to the same portion sizes. The appropriate food portion size, along with the quantity and composition of the elution buffer, still needs to be determined for matrices such as dried seaweeds.
## 5. Conclusions
The optimized PGM-MB protocol significantly improves the recovery of HuNoVs from Wakame seaweed salad, achieving good recovery rates and low limits of detection. The method is robust across genotypes, brands, and analysts, though mucin lot variability should be monitored. Its potential as a valuable tool for enhanced HuNoV surveillance in other high-risk foods should also be explored.
## References
1. Ahmed, Hall, Robinson et al. (2014) "Global prevalence of norovirus in cases of gastroenteritis: A systematic review and meta-analysis" *Lancet Infect. Dis*
2. Scallan, Hoekstra, Angulo et al. (2011) "Foodborne illness acquired in the United States-Major pathogens" *Emerg. Infect. Dis*
3. Ludwig-Begall, Mauroy, Thiry (1541) "Noroviruses-The State of the Art, Nearly Fifty Years after Their Initial Discovery" *Viruses*
4. Anonymous (2017) "National Enteric Surveillance Program Annual Summary"
5. Cdc (2017) "Surveillance for Foodborne Disease Outbreaks"
6. Chhabra, De Graaf, Parra et al. (2019) "Updated classification of norovirus genogroups and genotypes" *J. Gen. Virol*
7. Olaimat, Taybeh, Al-Nabulsi et al. "Common and Potential Emerging Foodborne Viruses: A Comprehensive Review" *Life*
8. Cook, Knight, Richards (2016) "Persistence and Elimination of Human Norovirus in Food and on Food Contact Surfaces: A Critical Review" *J. Food Prot*
9. Nasheri, Vester, Petronella (2019) "Foodborne viral outbreaks associated with frozen produce" *Epidemiology Infect*
10. Cook, Bertrand, Gantzer et al. (2018) "Persistence of Hepatitis A Virus in Fresh Produce and Production Environments, and the Effect of Disinfection Procedures: A Review" *Food Environ. Virol*
11. Sakon, Sadamasu, Shinkai et al. (2017) "Foodborne Outbreaks Caused by Human Norovirus GII.P17-GII.17-Contaminated Nori" *Emerg. Infect. Dis*
12. Donnan, Fielding, Gregory et al. (2009) "A Multistate Outbreak of Hepatitis A Associated With Semidried Tomatoes in Australia" *Clin. Infect. Dis*
13. Nasheri, Harlow, Chen et al. (2021) "Survival and Inactivation by Advanced Oxidative Process of Foodborne Viruses in Model Low-Moisture Foods" *Food Environ. Virol*
14. Raymond, Blain, Nasheri (2025) "Detection of Foodborne Viruses in Dates Using ISO 15216 Methodology" *Viruses*
15. Park, Jeong, Lee et al. (2014) "First norovirus outbreaks associated with consumption of green seaweed (Enteromorpha spp.) in South Korea" *Epidemiology Infect*
16. Whitworth (2019) "Norway Norovirus Outbreaks Linked to Seaweed Salad from China"
17. Anonymous (2024) "Modello di Richiamo Asiantrade Srl-Antipasto Giapponese di Alghe-Congelato-Goma Wakame; c17; Ministero della Saute: Comunicazione Istituzionale"
18. Fao (2021) "Report of the Expert Meeting on Food Safety for Seaweed-Current Status and Future Perspectives; Rome"
19. Løvdal, Lunestad, Myrmel et al. "Microbiological Food Safety of Seaweeds. Foods 2021"
20. (2017) "Microbiology of the Food Chain-Horizontal Method for Determination of Hepatitis A Virus and Norovirus Using Real-Time RT-PCR-Part 1: Method for Quantification"
21. Oliveira, Pardal, Pereira et al. (2024) "Portuguese macroalgae and halophytes for human consumption: Minimal risk of norovirus and Salmonella infection" *Food Control*
22. Tian, Engelbrektson, Mandrell (2008) "Two-Log Increase in Sensitivity for Detection of Norovirus in Complex Samples by Concentration with Porcine Gastric Mucin Conjugated to Magnetic Beads" *Appl. Environ. Microbiol*
23. Dancho, Chen, Kingsley (2012) "Discrimination between infectious and non-infectious human norovirus using porcine gastric mucin" *Int. J. Food Microbiol*
24. Cannon, Vinjé (2008) "Histo-Blood Group Antigen Assay for Detecting Noroviruses in Water" *Appl. Environ. Microbiol*
25. Tian, Brandl, Mandrell (2005) "Porcine gastric mucin binds to recombinant norovirus particles and competitively inhibits their binding to histo-blood group antigens and Caco-2 cells" *Lett. Appl. Microbiol*
26. Esseili, Wang, Saif (2012) "Binding of Human GII.4 Norovirus Virus-Like Particles to Carbohydrates of Romaine Lettuce Leaf Cell Wall Materials" *Appl. Environ. Microbiol*
27. Suresh, Harlow, Nasheri (2019) "Evaluation of porcine gastric mucin assay for detection and quantification of human norovirus in fresh herbs and leafy vegetables" *Food Microbiol*
28. Raymond, Paul, Perron et al. (2021) "Norovirus Extraction from Frozen Raspberries Using Magnetic Silica Beads" *Food Environ. Virol*
29. Wilrich, Wilrich (2009) "Estimation of the POD Function and the LOD of a Qualitative Microbiological Measurement Method" *J. AOAC Int*
30. Raymond, Paul, Perron et al. (2021) "Extraction of human noroviruses from leafy greens and fresh herbs using magnetic silica beads" *Food Microbiol*
31. Bai, Pu, Suo et al. (2023) "Exploring magnetic capture to improve the detection of human norovirus in strawberries" *Food Front*
32. Boom, Sol, Salimans et al. (1990) "Rapid and simple method for purification of nucleic acids" *J. Clin. Microbiol*
33. Plante, Barrera, Lord et al. (2023) "Examining the efficiency of porcine gastric mucin-coated magnetic beads in extraction of noroviruses from frozen berries" *Food Microbiol*
34. Sim, Ho, Phang (2013) "A simple and effective method for RNA isolation and cDNA library construction from the brown seaweed Sargassum polycystum (Fucales, Phaeophyceae)" *J. Appl. Phycol*
35. Tian, Yang, Jiang et al. (2010) "Specificity and kinetics of norovirus binding to magnetic bead-conjugated histo-blood group antigens" *J. Appl. Microbiol*
36. Murakami, Tenge, Karandikar et al. (2020) "Bile acids and ceramide overcome the entry restriction for GII.3 human norovirus replication in human intestinal enteroids" *Proc. Natl. Acad. Sci*
37. Feigelstock, Thompson, Mattoo et al. (1998) "The Human Homolog of HAVcr-1 Codes for a Hepatitis A Virus Cellular Receptor" *J. Virol*
38. Lowther, Bosch, Butot et al. (2019) "Validation of EN ISO method 15216-Part 1-Quantification of hepatitis A virus and norovirus in food matrices" *Int. J. Food Microbiol*
39. Williams-Woods, Rodriguez, Marchant et al. (2022) "Chapter 26-Concentration, Extraction and Detection of Enteric Viruses from Food"
40. Zhang, Pan, Li et al. (2012) "Comparative Detection of Human Noroviruses in Green Onion and Grape Using Porcine Gastric Mucin-Conjugated Magnetic Beads and Polyethylene Glycol Enrichment" *Food Sci*
41. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12455918&blobtype=pdf | # Correction for Oom et al., "The two-dose MVA-BN mpox vaccine induces a nondurable and low avidity MPXV-specific antibody response"
Aaron Oom, Kesi Wilson, Miilani Yonatan, Stephanie Rettig, Allison Youn, Michael Tuen, Yusra Shah, Ashley Dumont, Hayley Belli, Jane Zucker, Jennifer Rosen, Sedaghat Herati, Marie Samanovic, Ralf Duerr, Angelica Kottkamp, Mark Mulligan
## Abstract
Due to an equipment malfunction, the serum samples included in the published arti cle were improperly heat-inactivated at 43°C as opposed to the standard 56°C. To address this, tests of different heat inactivation (HI) conditions were conducted for both our neutralization and binding assays using a variety of samples from across our cohort. We found that proper heat inactivation significantly reduces MPXV neutraliza tion, while the addition of 5% guinea pig serum (GPS, Sigma-Aldrich G9774) restores signal (Fig. 7A). This is in line with the finding from Hubert et al. that most of the MPXV neutralizing capacity is dependent upon complement (M. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12737774&blobtype=pdf | # Regional, Age, and Sex Patterns of Hepatitis C Virus Infection in Russia: Insights from a 42,000-Participant Serosurvey
Victor Manuylov, Vladimir Gushchin, Vladimir Chulanov, Olga Isaeva, Denis Kleymenov, Andrei Pochtovyi, Elena Mazunina, Evgeniia Bykonia, Irina Tragira, Yana Simakova, Sergey Netesov, Artem Tkachuk, Tatyana Semenenko, Alexander Gintsburg, Karen Kyuregyan, Mikhail Mikhailov
## Abstract
Identifying population groups at greatest risk of hepatitis C virus (HCV) infection is essential for targeting screening and treatment. We analyzed the seroprevalence of antibodies
## 1. Introduction
Direct-acting antiviral (DAA) therapies can cure up to 95% of patients with hepatitis C virus (HCV) infection [1]. This makes the identification and timely treatment of infected individuals a cornerstone strategy not only to reduce mortality from cirrhosis and hepatocellular carcinoma, but also to interrupt onward transmission [2].
Chronic hepatitis C (CHC) is often asymptomatic, with up to 90% of infected individuals unaware of their status [3]. Many of them do not seek medical care, yet remain a source of potential transmission. Such individuals are typically identified incidentally, e.g., during medical visits for unrelated reasons, or through targeted screening of selected subpopulations within the conditionally healthy population.
Population-wide screening would be the most appropriate approach; however, it has never been conducted in Russia, likely due to administrative challenges and high costs. A practical alternative is to focus on identifying subgroups (defined by region, sex, age, or social characteristics) where the proportion of HCV carriers is highest. Data from such studies can be used for the development of targeted diagnostic and treatment programs.
The primary serological marker of HCV infection is the presence of antibodies to HCV (anti-HCV), representing both IgG and IgM against viral antigens [4]. Detection of HCV RNA by PCR serves two purposes: confirming chronic infection-an indication for initiating antiviral therapy according to clinical guidelines [5], and identifying active viral replication. Individuals with detectable HCV RNA derive the greatest benefit from treatment and should be prioritized for it [6].
HCV remains a significant public health problem in the Russian Federation. National surveillance data show that, over the past decade (2014-2024), the incidence of acute hepatitis C (AHC) has remained stable at 1-1.5 cases per 100,000 annually, while CHC has been reported at 31-40 cases per 100,000 [7,8]. Long-term incidence trends show no consistent decline (Figure 1); the transient drop during 2020-2021 likely reflects COVID-19related disruptions to healthcare [7].
Information on the main risk factors of HCV infection in modern Russia remains quite inconsistent, likely reflecting the absence of a single dominant risk factor. Authors of review papers on this topic (e.g., [9][10][11]) generally agree that most individuals currently living with CHC in Russia acquired the virus between the late 1990s and around 2010, when intravenous drug use was the principal and clearly predominant mode of the HCV transmission [12,13]. These days, however, this factor no longer plays such a decisive role (though it remains important), having been surpassed by medical procedures, including blood transfusions [9][10][11]. Other commonly cited risk factors include perinatal transmission, tattoos and piercings, imprisonment, and, notably, sexual contact with infected partners among young people [9]. The distribution and relative impact of these factors, however, appear to vary considerably across regions and even among different social groups within the same region [9][10][11]. Data are derived from official statistical reports [7,8] and regional health statistics provided by the Russian Ministry of Health. Some estimates suggest that ~1.8% of the Russian population are HCV carriers [14], that is a moderately high rate globally. For comparison, prevalence is estimated at 0.1-0.2% in Western Europe, 0.3-0.4% in China and India, and 0.7% in the United States; higher rates are reported in Ukraine (3.3%) and Pakistan (3.8%) [14]. Given its population size, Russia ranks among the top five countries worldwide in absolute number of carriers, estimated at 2.7 million [14]. Another estimates, based on the mathematical modeling [15] suggests even higher numbers, up to 4.25 million (2.9%) in 2020.
National averages, however, mask substantial heterogeneity across Russia's diverse geographic, economic, and demographic regions. Studies in conditionally healthy populations have reported anti-HCV prevalence both far below and well above the national average. Within the same Sakha Republic (Yakutia), for example, estimates have ranged from 0-1% to 10-13%, depending on climatic zone (Arctic or southern districts) and study period (1999-2002 vs. 2005-2006) [16,17].
In this context, we aimed to assess age-and sex-specific HCV prevalence in a conditionally healthy population across twelve Russian regions (Table 1), drawing on both original data and published studies. Our findings are intended to help public health authorities better identify high-risk groups for targeted screening and to ensure timely access to antiviral therapy. This study 1 Proportion of the study group relative to the officially reported population of the region in the year of sample collection. If the collection spanned two years, data for the first year are shown. 2 For age distribution, see Supplementary Table S1. 3 Abbreviated in the text as (groups of) "St. Petersburg" and "Moscow," respectively.
## 2. Materials and Methods
## 2.1. Study Design and Samples Collection
A total of 42,055 healthy volunteers from twelve regions across the Russian Federation, spanning from west to east, participated in the serosurvey conducted over multiple years (Table 1, Figure 2). Data from 30,724 of these participants are presented here for the first time. The data obtained from an additional 11,331 volunteers were previously published, primarily in Russian-language sources (see references in Table 1). To enable comparison with earlier findings, we also refer to previous studies by other authors, particularly those conducted in the 1990s and early 2000s (see Section 3.2 for details).
In the main text, as well as in the tables and figure captions, we use abbreviated names for the studied regions. Specifically, Kaliningrad, Belgorod, Rostov, Sverdlovsk, Novosibirsk, and Khabarovsk regions are referred to by the names of their capital cities. For the republics, we use the names Dagestan, Tatarstan, and Tyva. "Moscow" and "St. Petersburg" refer to the combined populations of Moscow city with Moscow Region, and St. Petersburg with Leningrad Region, respectively. In the Republic of Sakha (Yakutia), we identified two distinct groups. The group generally referred to as "Yakutia" represents the population of the so-called "agricultural zone" (southern districts, including the major cities of Yakutsk and Neryungri), studied in 2008 and 2018 (Table 1). The "Arctic zone of Yakutia" refers to the northern Momsky District, surveyed in 2022 (Table 1). This distinction is important, as the southern and northern regions of Yakutia have shown marked differences in the epidemiology and prevalence of parenteral hepatitis [16,17].
The study included male and female participants aged 0 to 95 years, divided into nine age groups: under 1 year, 1-9, 10-14, 15-19, 20-29, 30-39, 40-49, 50-59, and ≥60 years. The number of participants in each age group and region is shown in Supplementary Table S1. The male-to-female ratio ranged from 1:1 to 1:2.1, depending on the region (Table 1 and Table S1). Ages were grouped into relatively broad 10-year intervals to ensure that each group contained a sufficient number of participants for statistical reliability. However, within adolescence (10-20 years), two five-year subgroups were identified, as this age range was previously associated with HCV infection linked to drug use (particularly in the late 1990s and 2000s; see [12,13] for details). [12,16,17,25,26] and are included for comparison with our original results (see Section 3.2 for details).
All participants met the following inclusion criteria: apparently healthy, with no obvious symptoms of acute illness at the time of enrollment, and permanent residency in the study region. Exclusion criteria included a history of liver disease (infectious or non-infectious), any current acute illness/body temperature above 37.1 • C; as well as any surgery, blood transfusion, or treatment with blood products in the three months prior to enrollment (self-reported or, for participants under 15 years of age, reported by a parent or guardian).
All serum samples were coded, aliquoted, and stored at -70 • C until testing.
## 2.2. Ethics
The study was conducted in accordance with the principles of the World Medical Association Declaration of Helsinki for ethical medical research involving human subjects. Informed written consent was signed by all the participants or their legal guardians.
## 2.3. ELISA Testing
Serum samples from Belgorod region (Table 1) were tested for anti-HCV antibodies using the Architect Anti-HCV test (Abbott Laboratories, Abbott Park, IL, USA). Samples reactive in the screening test were confirmed by immunoblotting for antibodies to structural and non-structural HCV proteins (INNO-LIA HCV, Fujirebio Europe N.V., Gent, Belgium). All other serum samples listed in Table 1 were tested for anti-HCV (IgG + IgM) using commercial ELISA kits (Vector-Best, Novosibirsk, Russia). Samples that tested positive in the screening assay were subsequently analyzed using the anti-HCV confirmation test from the same manufacturer. A participant was classified as an anti-HCV-positive carrier if their sample tested positive in the confirmation ELISA.
## 2.4. PCR Testing
All samples that were found to be positive in the anti-HCV confirmation assay, were further examined using a commercial real-time PCR kit for qualitative detection of HCV RNA (AmpliSens, Moscow, Russia). This assay has a detection limit of 10 IU/mL when nucleic acids are extracted from a 1 mL sample. Samples collected in Moscow, St. Petersburg, Dagestan, Novosibirsk, and Khabarovsk (2018-2020) that were negative for anti-HCV by ELISA were also tested by PCR in pools of 10 samples using the same AmpliSens reagent kit. If a pooled sample tested positive for HCV RNA, each of the 10 individual samples in the pool was subsequently tested to identify the HCV RNA-positive sample.
## 2.5. Statistical Analysis
Statistical analysis included assessment of differences in proportions of anti-HCV carriers in groups/cohorts using the Chi-square test with Yates' correction, or Fisher's exact test for small samples. A p-value < 0.05 was considered statistically significant. Confidence intervals (95% CI) for proportions were calculated using the binomial distribution. Correlation coefficients (r) between quantitative variables, such as the incidence of hepatitis C and anti-HCV prevalence, were determined using the two-tailed Spearman's non-parametric test in Prism software, version 9.0.2 (GraphPad, Boston, MA, USA).
## 3. Results
## 3.1. Differences in Anti-HCV Prevalence Among Regional Groups
The prevalence of antibodies to the hepatitis C virus (anti-HCV prevalence) in the regional groups surveyed in 2018-2022 is summarized in Figure 2 and Table 2 (with separate data for men and women). Detailed age stratification data are provided in Supplementary Table S1. Table 3 presents the results of pairwise comparisons between regional groups from 2018 to 2022, based on the proportions of anti-HCV-positive individuals, using the chi-square test.
Based on these results, we considered it possible to broadly classify regional groups that were observed in 2018-2022 into three categories. The regions with a relatively low prevalence of anti-HCV (1.1-1.4%) included Belgorod Region, Moscow (meaning Moscow Region along with Moscow City), and Saint Petersburg (including Saint Petersburg City and Leningrad Region). Differences among these regions were not statistically significant (p > 0.05). Table 3. Pairwise comparison of anti-HCV prevalence between regional groups surveyed in 2018-2022. Each cell at the intersection of a row and a column presents the p-value for the difference in anti-HCV prevalence between the corresponding regional groups, calculated using pairwise chi-square tests. Color coding indicates the level of statistical significance: light green, p < 0.05; green, p < 0.01; "nd", no statistically significant difference detected. Diagonal cells (highlighted in gray) display the anti-HCV prevalence for that region together with the 95% confidence interval (95% CI). "Moderate" anti-HCV prevalence (1.8-2.1%) was found in Republic of Dagestan, Republic of Tatarstan, Novosibirsk Region, Republic of Tyva, and Republic of Sakha (Yakutia, southern districts). Differences within this group were also not statistically significant (p > 0.05).
Finally, relatively high prevalence (3.4-5.2%) was reported for Khabarovsk Region and Yakutia (Arctic zone). Differences between these two regions were not statistically significant (p > 0.05; Table 3).
Kaliningrad Region occupied an intermediate position, statistically similar to both regions with the "moderate" prevalence (Republic of Tatarstan and Republic of Tyva) and to the "high-prevalence" region of Khabarovsk. Notably, two "moderate-prevalence" regions (Republic of Tatarstan and Republic of Tyva) did not differ significantly from "lowprevalence" regions (Saint Petersburg and Moscow). Apart from these overlaps, differences between the three prevalence categories were statistically significant (Table 3).
Data for the 2008 survey are also presented in Figure 2, Table 2 and Supplementary Table S1. In 2008, both low-and high-prevalence groups were identifiable: Moscow (1.7%) represented the lower end, while the Republic of Tyva and Yakutia (3.3%), were at the higher end (p < 0.05, Table 2). Rostov Region and Sverdlovsk Region occupied intermediate positions (2.1-3.0%) and did not differ significantly from any of the other regions in 2008.
These findings indicate that, at least up to the recent 2018-2022 period, the prevalence of HCV infection in the Russian Federation has remained markedly heterogeneous across regions.
## 3.2. Dynamics of Anti-HCV Prevalence in Regional Groups over Time
The anti-HCV prevalence in the conditionally healthy population across the surveyed regions at two time points (2008 and 2018-2022; see Tables 2 and3) made it possible to evaluate temporal trends. Additional conclusions were drawn by comparing our findings with similar studies performed by other researchers in the same regions. All the data for available time points is presented in Figure 2.
In Moscow, the proportion of anti-HCV carriers decreased from 1.7% in 2008 to 1.2% in 2018-2019. However, this decrease was not statistically significant (see confidence intervals in Table 2).
For Novosibirsk region, data on the anti-HCV prevalence in the conditionally healthy population were previously reported for the period 1995-1999 [25]. In that study, a group of 1073 individuals were examined, including schoolchildren, medical students, and randomly selected adults, with a mean anti-HCV prevalence of 4.2 ± 1.2%. Another study [12], conducted in 2000-2002, reported a prevalence of 5.6 ± 2.1% among 500 patients attending a non-infectious outpatient clinic. The prevalence estimates from these two earlier studies did not differ significantly from each other, but both were significantly higher than the value obtained in our study for Novosibirsk Region in 2019-2020 (1.9 ± 0.3%, p-value < 0.01; Table 2).
In the Republic of Tyva, the proportion of anti-HCV-positive carriers also declined from 3.3% in 2008 to 2.0% in 2019; however, this difference was not statistically significant (p-value > 0.05).
The proportion of anti-HCV carriers among the conditionally healthy population of Yakutia (excluding the Arctic zone; see below), as observed in our study in 2018, was 2.0%, which was lower than the 3.3% recorded in 2008 (Table 2), although the difference was not statistically significant. At the same time, a comparison with earlier results reported from the same southern regions of Yakutia reveals a more pronounced downward trend. In particular, a study conducted between 1999 and 2002 [16], which included 1394 adolescents and adults from the southern "agricultural" districts of Yakutia (Namsky, Gorny, Vilyuisky), reported a high mean anti-HCV prevalence of 6.2 ± 1.3% (ranging from 2.7% to 13.4% depending on age). A subsequent study carried out in 2005-2006 in the city of Neryungri, also located in southern Yakutia, examined 329 conditionally healthy individuals and reported an even higher prevalence of 13.1 ± 3.9% [17]. Both of these values were significantly higher than those observed in our study for the same "agricultural" zone in 2008 (3.3%) and, especially, in 2018 (2.0%) (Table 2; p-value < 0.01 for all comparisons).
Thus, the available data suggest a decreasing trend in the prevalence of antibodies to the HCV in these regions from 2008 (or earlier) to more recent years (2018-2020), although this trend was not always statistically significant.
However, this decline was not observed in all study groups. For example, in the same study cited above [16], a very low prevalence of antibodies to the HCV was recorded in the Abyisky and Eveno-Byntaisky districts of the Arctic zone of Yakutia: 0% among 296 children and 162 adolescents, and 1.1% among 370 adults. It should be noted that these figures refer to the northern regions of Yakutia, in contrast to the southern areas discussed above.
This low prevalence was confirmed by another study conducted in 2005-2006 [17], which reported the anti-HCV prevalence of only 1.3 ± 1.3% among 301 representatives of the Evenki ethnic group, who predominantly inhabit the northern territories of Yakutia. In contrast, our 2022 study revealed a markedly higher prevalence in residents of the Momsky district, also located in the Arctic zone and bordering the Abyisky district, with anti-HCV detected in 5.2 ± 2.3% of the examined individuals, that was the highest rate among all groups included in Table 2. The reasons for this sharp increase remain unclear and require further investigation.
Similarly, the increase in anti-HCV prevalence over time was observed in the Republic of Dagestan (North Caucasus). In 2019-2020 our study found the anti-HCV prevalence of 1.8% in this region (Table 2), whereas an earlier investigation of 10,682 blood donor samples collected in 1994-1996 reported a significantly lower rate of 0.9 ± 0.2% (p-value < 0.01) [26]. One explanation for this difference may be that during the earlier period, the Republic of Dagestan had probably not yet experienced the surge in hepatitis C incidence that peaked in the Russian Federation in the late 1990s [12,13]. Additionally, donor cohorts (such as a studied in the cited investigation) typically have lower morbidity from parenterally transmitted infections than the general population [13].
Overall, if we assume that the volunteer groups examined in our study are at least partially representative of the conditionally healthy population of the Russian Federation (taking into account the limitations discussed further in the Section 4) it can be concluded that the proportion of anti-HCV carriers in the country has significantly declined over the past 10-12 years. In the combined group of volunteers studied in 2008 (4764 individuals from five regions differing in economic, geographic, and climatic characteristics), the mean prevalence of anti-HCV was 2.6 ± 0.5% (see bottom lines of Table 2). By 2018-2022 (37,291 participants from ten regions), this average prevalence had significantly decreased to 1.9 ± 0.1% (Table 2, p-value < 0.01). However, as illustrated by the examples of rising anti-HCV prevalence in some regions, it cannot yet be concluded that the overall downward trend is consistent or stable across the entire country. This uncertainty is also reflected in the slow and gradual decline in the official number of newly registered cases of hepatitis C over the years (Figure 1), that is discussed in the following section.
## 3.3. Correlation Between Hepatitis C Incidence and Anti-HCV Prevalence in Regional Groups
The pronounced regional heterogeneity described above is also reflected in the officially reported incidence rates of hepatitis C. Figure 1 illustrates the dynamics of long-term cumulative incidence, including both acute hepatitis C and chronic hepatitis C, while Supplementary Table S2 provides detailed data on the incidence in the surveyed regions for the year in which the samples were collected, presented separately for acute hepatitis C and chronic hepatitis C.
The data show that hepatitis C incidence rates do not always correspond to the prevalence of anti-HCV. For example, Saint Petersburg has consistently recorded the highest incidence of hepatitis C in the Russian Federation, 70 to 90 cases per 100,000 population annually (including both acute hepatitis C and chronic hepatitis C). However, the prevalence of anti-HCV in this region was among the lowest, at only 1.4% (Table 2). Similarly high incidence rates were reported in Novosibirsk Region (50-90 per 100,000 annually), despite a moderate anti-HCV prevalence of 1.9%. In Moscow, where the prevalence of anti-HCV was low (1.2%), the long-term incidence of hepatitis C remained at 35-55 cases per 100,000 annually, exceeding the national average of 30-40 per 100,000 (Figure 1).
In contrast, in the Republic of Dagestan, the incidence of hepatitis C has remained below 10 per 100,000 annually for more than a decade, yet the prevalence of anti-HCV was 1.8%-comparable to that of Novosibirsk Region, despite much higher incidence rate there.
Formal correlation analysis using the two-tailed Spearman rank test revealed no statistically significant association between anti-HCV prevalence and the regional incidence of either acute or chronic hepatitis C in the year of sample collection, or with their combined incidence. In all cases, correlation coefficients (r) ranged from -0.12 to -0.01, with corresponding p-values between 0.63 and 0.95 (see Supplementary Table S2).
## 3.4. Dynamics of Anti-HCV Prevalence in Age Groups
Supplementary Table S1 presents data on the anti-HCV prevalence among participants in the following age categories: under 1 year, 1-9 years, 10-14 years, 15-19 years, 20-29 years, 30-39 years, 40-49 years, 50-59 years, and 60 years or older, across the surveyed regions. Based on these data, age-dependent prevalence curves were constructed and are shown in Figure 3 (A: groups surveyed in 2018-2022; B: groups surveyed in 2008).
The graphs clearly demonstrate that anti-HCV prevalence increases with age. In the younger age groups, the proportion of positive carriers remains consistently low. Among participants aged 1 to 19 years, anti-HCV prevalence in almost all regions remained below 1%. Minor deviations from this trend, likely representing random fluctuations, were observed in the 1-9-year age group in Sverdlovsk Region (2008) and in the 10-14-year age group in Kaliningrad Region (2019) (Figure 3).
A marked increase in anti-HCV prevalence begins in adulthood. In the 2008 cohort, this increase occurred starting from the 20-29-year age group, whereas in the 2018-2022 cohort it was delayed until the 30-39-year age group. This shift is evident not only visually in the graphs but is also supported by statistical analysis. For example, in the entire 2008 group (depicted as the thick black line in Figure 3B), the differences in prevalence between adjacent younger age groups (for example, <1 year vs. 1-9 years, 1-9 years vs. 10-14 years, 10-14 years vs. 15-19 years) were not statistically significant. However, a significant difference was observed between the 15-19-year age group and the 20-29-year age group (1.2 ± 0.9% vs. 3.1 ± 1.5%, p-value < 0.05; Supplementary Table S1).
In contrast, for the 2018-2022 cohort, the first statistically significant increase occurred between the 20-29-year age group and the 30-39-year age group (0.8 ± 0.2% vs. 2.0 ± 0.3%, p-value < 0.01; Supplementary Table S1). No significant increases were observed in age groups younger than 20 years, with the exception of the <1-year group, which is discussed additionally below. Another statistically significant rise was observed after the age of 39 years: in the 40-49-year age group, anti-HCV prevalence reached 3.6 ± 0.6%, compared with 2.0 ± 0.3% in the 30-39-year age group (p-value < 0.01; Supplementary Table S1). Beyond the age of 49 years, changes in prevalence occurred more gradually (Figure 3A). S2.
As noted previously, in several regions surveyed in 2018-2022-including Kaliningrad Region, Moscow, Tatarstan, Yakutia (non-Arctic zone), and Khabarovsk-an unusually high proportion of anti-HCV carriers was identified among children younger than one year (Figure 3 and Table 2). Moreover, in the entire 2018-2022 cohort, the prevalence of anti-HCV among infants younger than one year was 1.5 ± 0.8% (Table 2, line 17), which was significantly higher than the prevalence among children aged 1-9 years (0.8 ± 0.2%; p-value < 0.05). Since no reports of hepatitis C outbreaks specifically affecting newborns in the Russian Federation were identified, we believe that this elevated prevalence of antibodies in infants is most likely the result of passive transfer of maternal antibodies from anti-HCV-positive mothers. Supporting this interpretation, the prevalence of anti-HCV among infants in the regions mentioned (ranging from 1.9% to 2.9%, and reaching as high as 4.9% in Tatarstan) closely mirrored the prevalence observed among women of childbearing age (30-39 years) in the same regions (0.7-3.7%; Supplementary Table S2).
The absence of a similar peak in the <1-year age group in the 2008 data may be explained by the fact that, at that time, the number of women infected with the HCV (or carrying anti-HCV antibodies) who had recently given birth was likely too low for their infants to be represented in the study sample. In any case, based on the trends illustrated in Figure 3, maternal antibodies appear to persist in newborns for approximately one year.
## 3.5. Anti-HCV Prevalence in Men and Women
Table 2 presents the anti-HCV prevalence separately for men and women in each surveyed regional group. More detailed information, including the number of male and female anti-HCV carriers in specific age groups, is provided in Supplementary Table S2.
In all regional groups surveyed in 2008, no statistically significant differences in anti-HCV prevalence between men and women were observed (Table 2). On average, in the combined 2008 cohort, anti-HCV prevalence was 2.6 ± 0.7% among men and 2.7 ± 0.6% among women.
In contrast, in many of the regional groups surveyed during 2018-2022, the prevalence of anti-HCV was noticeably higher among men compared with women. Statistically significant differences were observed in the following groups: Saint Petersburg (2.4% versus 0.9%, p-value < 0.01), Belgorod (1.6% versus 0.7%, p-value < 0.05), Moscow (2.2% versus 0.9%, p-value < 0.01), Dagestan (2.2% versus 1.3%, p-value < 0.05), and Khabarovsk (4.2% versus 2.4%, p-value < 0.01). In the remaining regions, differences between male and female cohorts were not statistically significant (Table 2).
In the entire 2018-2022 cohort, anti-HCV prevalence among men (2.5 ± 0.2%) was significantly higher than among women (1.5 ± 0.2%, p-value < 0.01; Table 2). Although the identification of gender-specific risk factors for anti-HCV carriage was beyond the scope of this study, the observed data support the conclusion that, in contemporary Russian society, men are at a higher risk of infection with the HCV compared with women.
## 3.6. Results of PCR Testing
Data on participants who tested positive for HCV RNA by PCR are presented in Table 2. In the vast majority of cases, individuals who were PCR-positive for HCV RNA were also positive for antibodies to the HCV. That was expected, since PCR testing in this study was primarily conducted for samples that were seropositive in anti-HCV screening and confirmation assays.
However, in a number of large regional groups surveyed between 2018 and 2020specifically Moscow, Saint Petersburg, Dagestan, Novosibirsk, and Khabarovsk-PCR testing was also performed on anti-HCV-negative samples using pooled sample analysis (ten samples per pool). In total, 30,232 anti-HCV-negative samples were tested in this manner, which resulted in the detection of 30 additional HCV RNA-positive sera (0.1%). These RNA-positive individuals are not distinguished in Table 2 from those identified through PCR testing of anti-HCV-positive samples.
Notably, all but one of these "seronegative" HCV RNA-positive samples were positive for antibodies to the hepatitis B core antigen (anti-HBcAg), a serological marker of chronic or past hepatitis B virus infection, and lacked any other serological markers of parenterally transmitted hepatitis viruses (that is, they were negative for hepatitis B surface antigen-HBsAg, negative for antibodies to hepatitis B surface antigen-anti-HBs IgG, and negative for antibodies to the HCV). Further details on the serological profiles of the studied groups with respect to hepatitis B virus infection can be found in our previous publication [27].
Among individuals who were positive for anti-HCV, the proportion of those who were also positive for HCV RNA ranged from 20% to 62% across the different regional groups (Table 2). When comparing the aggregated data from the 2008 and 2018-2022 cohorts, this proportion was similar: 38.5% and 32.6%, respectively. The relatively high number of individuals who were anti-HCV-positive but HCV RNA-negative is most likely explained by spontaneous clearance of the virus, which occurs in approximately 20-40% of individuals following an acute HCV infection [28].
## 3.7. Estimation of the Present Number of Individuals Carrying Anti-HCV in the Russian Federation
Attempts to estimate the number of individuals carrying markers of infectious diseases at the national level based on limited sample data are, by nature, speculative. The reasons for this are discussed in detail in the "Limitations of the Study" subsection at the end of the Discussion. Nevertheless, such estimates remain a favorite exercise among epidemiologists and can be valuable, particularly for planning the economic resources required for identifying and treating infected individuals. In this context, we present our own estimate, while acknowledging its inherent limitations.
In the combined group studied during 2018-2022 (37,291 participants; see Tables 2 and3), the overall prevalence of antibodies to HCV was found to be 1.9 ± 0.1%. However, as discussed in Sections 3.4 and 3.5, anti-HCV prevalence varies significantly depending on age and sex. Furthermore, Table S1 (Supplementary File) shows that the age and sex distribution of our surveyed groups differed considerably from that of the general population.
For instance, individuals aged 30-39 years constituted 18% of the actual population in the Moscow region (including the city of Moscow) in 2019 [29] (the population data from the last pre-COVID-19 year will be used as a base in this section) but accounted for 36% of our surveyed group in 2018-2019 (Tables S1 andS3). In Dagestan, children aged 1-9 years represented 23% of the sample, compared to only 15% of the republic's actual population. The overall male-to-female ratio in the 2018-2022 sample was 1:1.3 (Table 2), whereas the national ratio in 2019 was 1:1.15 [29], with similar discrepancies observed in other regions. Therefore, it was necessary to adjust the experimental sample to better reflect the true demographic structure of the Russian population.
To achieve this, we applied the age-and sex-specific prevalence rates of anti-HCV observed in our experimental sample to the actual demographic structure of each region. In practical terms, if the observed prevalence of anti-HCV in a given sex-age group was q%, and the size of that demographic stratum in the real population of a region was P, then the estimated number of anti-HCV carriers in that group would be P × q%. Summing the results across all sex-age strata yielded the total estimated number of anti-HCV carriers in the region (see Table S3 in Supplementary File for detailed calculations).
This approach, however, does not allow the calculation of meaningful aggregated confidence intervals, as their combination would produce an excessively wide range. Additionally, data from the Momsky District of Yakutia were excluded from this calculation due to its small population size.
Applying this method to all regions included in the 2019-2022 study-covering a total population of 42 million in 2019-produced an estimate of approximately 937,000 individuals carrying antibodies to HCV, corresponding to a prevalence of 2.2%. This value is slightly higher than the prevalence observed directly in the experimental sample (1.9 ± 0.1%; 713 carriers among 37,291 participants; see Table 2). We consider this higher estimate more appropriate for the purposes of planning national HCV control strategies, given the methodological validity of the demographic adjustment.
If we further assume that this 2.2% prevalence applies to the entire population of the Russian Federation (approximately 146 million in 2019), the estimated total number of individuals carrying anti-HCV nationwide would be about 3.23 million. This figure aligns with previously published estimates ranging from 2.7 to 4.5 million [14,15]. Of course, our estimate is based on a relatively small sample compared with the total population of Russia, and therefore may differ from results obtained in larger-scale studies. The Supplementary File provides the complete experimental dataset and corresponding demographic statistics, enabling other researchers to perform more refined calculations should new or additional information become available in the future.
## 4. Discussion
In this study, we investigated the prevalence of antibodies to the HCV by direct In this study, we examined the prevalence of antibodies to HCV through direct serological testing of blood samples from volunteers representing various population groups. Between 2018 and 2022, a total of 37,291 unique samples were collected from the conditionally healthy population of all ages across ten geographical regions of the Russian Federation. Importantly, approximately 82.5% of these samples were collected before the end of 2019, i.e., before the onset of the COVID-19 pandemic, which otherwise would have delayed this epidemiological study. Additionally, 4764 samples collected in 2008 were analyzed. Based on this dataset, the average prevalence of anti-HCV was estimated at 1.9 ± 0.1% for the 2018-2022 cohort and 2.8 ± 0.5% for the 2008 cohort (Table 2).
Notable regional differences in anti-HCV prevalence were found. In the 2018-2022 cohort, the lowest prevalence values were recorded in the Belgorod region, Moscow, and Saint Petersburg (1.1-1.4%), while the highest values were recorded in Khabarovsk region and the Arctic zone of Yakutia (3.4-5.2%). These differences were statistically significant (Table 3).
Overall, the results indicate a general downward trend in HCV prevalence over time. However, this trend is not universal: in the Dagestan and the Arctic zone of Yakutia, the proportion of anti-HCV-positive individuals in 2018-2020 exceeded the levels reported in earlier studies by other authors.
An important finding was the absence of correlation between anti-HCV prevalence and the officially registered incidence of hepatitis C (see Section 3.3). This suggests that official incidence rates are not reliable indicators of the true morbidity of infection in the population but instead reflect the extent of diagnostic testing coverage in different regions. Substantial differences were also observed in the prevalence of anti-HCV across age groups. In most younger age groups examined in 2018-2022, the prevalence did not exceed 1%. However, starting from approximately 30 years of age, prevalence increased markedly to 2-4%, and in older age groups, it reached 4-10% (Figure 3A, Table S1). In the 2008 cohort, this increase was already evident starting from the 20-29-year age group. It is plausible that individuals in this cohort were infected during adolescence or early adulthood in the late 1990s and early 2000s, when predominant risk factors (especially injection drug use) were more widespread [12,13]. Children and adolescents under 20 years of age examined in 2008, by contrast, likely had significantly lower lifetime exposure to these risk factors.
In the 2018-2022 cohort, the age-related increase in anti-HCV prevalence has shifted by approximately one decade compared to the 2008 data. This suggests that the same cohort of individuals likely infected in the late 1990s to early 2000s had aged by ten years and were now represented in older age groups. Meanwhile, individuals under 30 years of age in 2018-2022 continued to show relatively low infection rates, suggesting that high-risk factors prevalent among young people in the 1990s are no longer as influential for the younger generation. This shift may reflect positive changes in public health and behavioral patterns.
At that, in contemporary Russian society, certain as yet unidentified risk factors appear to be disproportionately associated with the male sex. In the entire 2018-2022 cohort, the prevalence of anti-HCV among men was significantly higher than among women (2.5% versus 1.5%; see Section 3.5). No such sex-related difference was detected in the 2008 cohort, suggesting that before 2008, there were no risk factors disproportionately affecting one sex.
Taken together, these findings underscore the pronounced heterogeneity in the HCV epidemiology across the Russian Federation. Seroepidemiological studies such as ours help to identify demographic and regional groups most affected by the infection, which should be prioritized in prevention and control programs. Based on the age-and sex-specific patterns identified, it can be concluded that in modern Russia (2018-2022), the majority of infected individuals are men over 30 years of age (Figure 3A, Table 2). Detailed numerical data are provided in Table S1 (Supplementary File), which can be regarded not only as reference material but also as a basis for designing targeted public health interventions.
In accordance with the aims outlined in the Introduction, diagnostic screening followed by antiviral treatment would be most effective if directed at groups with the highest concentration of seropositive individuals. Adjusting for the contribution of different age and sex cohorts in the study sample, we estimated that the total number of anti-HCV carriers in modern Russia is approximately 3.23 million individuals. As demonstrated in our study, only about one-third of individuals testing positive for anti-HCV also test positive for the HCV RNA and thus have a clinical indication for antiviral therapy [5,6].
The remaining anti-HCV-positive but RNA-negative individuals require only observation. According to current Russian guidelines, such individuals undergo two polymerase chain reaction tests six months apart, and if both results are negative, they are considered recovered and removed from the viral hepatitis register [30].
Based on our estimates, approximately 1.1 million people in the Russian Federation require antiviral treatment. At an average cost of 200,000-300,000 rubles per course (equivalent to USD 2000-3000), the total projected cost of therapy would amount to 200-300 billion rubles (USD 2-3 billion). For comparison, official estimates place the annual economic burden of chronic hepatitis C in Russia at 65-75 billion rubles [7,8]. Therefore, the economic benefits of implementing large-scale screening and treatment programs could become apparent within five to ten years. A significant step in this direction is the inclusion of anti-HCV screening in the adult preventive medical examination program beginning in 2024 [31] as well as the launch in 2024 of the federal project "Combating Hepatitis C" within the framework of the national program "Long and Active Life" [32], which includes a substantial expansion of access to antiviral treatment funded by the federal budget.
## Limitations of the Study
The present work-or more precisely, a set of studies conducted in different years but united by a common objective and presented here under a single title-has several methodological limitations. These limitations are characteristic of many seroepidemiological studies and are primarily associated with methodological simplifications that should be explicitly acknowledged:
1. Enrollment of participants in the study was based on inviting volunteers rather than employing a randomized sampling strategy that would statistically represent different social, economic, and behavioral strata of the population. Such an approach may have led to an underestimation of the anti-HCV prevalence in the surveyed groups. Volunteer-based recruitment usually does not adequately capture marginalized or high-risk groups, which, although relatively small in number, often have a disproportionately high prevalence of HCV infection [12]. Furthermore, the primary aim of the present work was to identify asymptomatic anti-HCV carriers "hidden" within the ostensibly healthy population. For this reason, individuals with a known diagnosis of HCV infection or with a history of other viral hepatitis or parenteral infections were deliberately excluded (self-reported or guardsreported hepatitis history served as the exclusion criterion for participants). Consequently, these individuals were not included in the statistical analyses of regional prevalence rates.
2. The regional cohorts presented in Tables 1 and2, although all derived from the "conditionally healthy" population, are not epidemiologically homogeneous with one another. While an effort was made to recruit a balanced distribution of participants across age and sex groups to ensure adequate statistical power for subgroup analysis, it was not feasible to perfectly match the demographic structure of the study samples to that of the actual population in each region. Moreover, it was not possible to refuse participation to volunteers who wished to be included, even if their demographic subgroup was already sufficiently represented. As a result, as shown in Table 1 and Table S1 (Supplementary File), the representation of specific age and sex categories varied between regions (see also Section 3.7). This demographic imbalance should be considered when interpreting and comparing prevalence estimates across regions.
3. Almost no epidemiological data were collected alongside the samples (apart from demographic information), which prevents drawing reliable conclusions about the causes of observed disparities in anti-HCV prevalence among different groups. While we were able to describe the proportions of individuals carrying the marker within the population, we can only hypothesize about the epidemiological processes that have led to the current situation.
4. In this study, the primary analytical focus was on the prevalence of antibodies to the HCV, with the assumption that this marker correlates proportionally with the overall prevalence of HCV infection in the population. The results of PCR testing for HCV RNA were used only as information to indicate the proportion of "active", that is, replicative, infections. As shown in Table 2, the proportion of anti-HCV-positive samples that were also positive for HCV RNA varied between 30% and 70% depending on the region. As discussed in Section 3.6, this discrepancy is most likely attributable to spontaneous clearance of the virus, which occurs in approximately 20-40% of individuals following acute infection [28]. However, it cannot be ruled out that some of the PCR-negative results could be explained by the possible hyper sensitivity of the ELISA assays used, or conversely, by the limited sensitivity of the PCR assays applied in certain cases. In any case, prevalence estimates based solely on anti-HCV seropositivity tend to overstate the proportion of individuals with active, chronic infection.
These limitations inevitably complicate the determination of the absolute number of individuals infected with HCV in the Russian Federation as a whole and in individual regions. Nevertheless, the consistent application of the same methodology across all study groups allows for the assessment of temporal trends within the same population (where multiple time points are available) and enables qualitative comparison between different regions. This approach has yielded several important and noteworthy findings, which are presented above and elaborated upon here.
5. One of the main limitations of this study is the absence of data on anti-HCV prevalence for the period between 2008 and 2018-2022. Therefore, the study does not provide a continuous 14-year trend but rather compares two distinct time points at the beginning and end of this interval. This gap may affect the interpretation of the observed trends: although a decline in anti-HCV prevalence is evident in many regions between 2008 and 2018-2022, it remains unclear whether this decrease was steady throughout the entire period or if fluctuations occurred in between.
## 5. Conclusions
In this large-scale seroepidemiological investigation, we examined the prevalence of antibodies to the HCV across diverse age, sex, and regional cohorts of the Russian Federation. The analysis was based on more than 42,000 serum samples collected over a fourteen-year period, from 2008 to 2022. This comprehensive dataset allowed for the assessment of both cross-sectional and temporal patterns in the HCV spread.
Our results demonstrated substantial variation in the prevalence of antibodies to the hepatitis C virus between different geographic regions and demographic groups. These findings confirm the pronounced heterogeneity of the HCV epidemiology within the Russian Federation. The data indicate a gradual overall decline in the prevalence of anti-HCV over time; however, this trend was not consistent across all surveyed regions, with some areas showing stable or even increasing prevalence rates.
By applying an adjustment for the actual age and sex distribution of the Russian population, we estimated that approximately 3.23 million individuals in the country are positive for antibodies to the HCV. Of these, roughly one-third are expected to have detectable HCV RNA and, therefore, meet clinical criteria for antiviral therapy. This proportion translates to an estimated 1.1 million individuals who require antiviral treatment.
These results highlight the importance of prioritizing targeted screening and treatment interventions for demographic groups with the highest prevalence rates. In particular, the findings indicate that men over 30 years of age constitute a key high-risk group in modern Russia. The integration of anti-HCV screening into the national program of preventive medical examinations for adults, along with the significant expansion of treatment programs beginning in 2024, represents a timely and strategically important step toward reducing the burden of HCV infection, improving population health outcomes, and ensuring a more efficient allocation of healthcare resources.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v17121529/s1, Supplementary File (Table S1: Prevalence of the HCV Markers Across Sex and Age Groups of the Studied Regions; Table S2: Correlations between the prevalence of anti-HCV in the study groups and the incidence of hepatitis C in the corresponding regions; E; Table S3: Estimated number of anti-HCV carriers by age and sex groups in the studied regions of Russia, adjusted for their contribution to the actual population structure).
Author Contributions: Conceptualization, V.A.M., V.A.G., A.P.T., and K.K.K.; methodology, V.A.M., V.A.G., V. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
## References
1. Falade-Nwulia, Suarez-Cuervo, Nelson et al. (2017) "Oral direct-acting agent therapy for hepatitis C virus infection: A systematic review" *Ann. Intern. Med*
2. Durham, Skrip, Bruce et al. (2016) "The impact of enhanced screening and treatment on hepatitis C in the United States" *Clin. Infect. Dis*
3. Thrift, El-Serag, Kanwal (2017) "Global epidemiology and burden of HCV infection and HCV-related disease" *Nat. Rev. Gastroenterol. Hepatol*
4. (2020) "European Association for the Study of the Liver. EASL recommendations on treatment of hepatitis C: Final update of the series" *J. Hepatol*
5. Gerlach, Diepolder, Zachoval et al. (2003) "Acute hepatitis C: High rate of both spontaneous and treatment-induced viral clearance" *Gastroenterology*
6. (2023) "Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing. State Report on the Sanitary and Epidemiological Welfare of the Population in the Russian Federation"
7. (2024) "Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing. State Report on the Sanitary and Epidemiological Welfare of the Population in the Russian Federation"
8. Mukomolov, Trifonova, Levakova et al. (2016) "Hepatitis C in the Russian Federation: Challenges and future directions" *Hepatic Med. Evid. Res*
9. Isakov, Nikityuk "Elimination of HCV in Russia: Barriers and Perspective" *Viruses*
10. Pimenov, Kostyushev, Komarova et al. (1482) "Epidemiology and Genotype Distribution of Hepatitis C Virus in Russia" *Pathogens*
11. Shustov, Kochneva, Sivolobova et al. (2005) "Molecular epidemiology of the hepatitis C virus in Western Siberia" *J. Med. Virol*
12. Shakhgildyan, Mikhailov, Onishchenko (2003) "Parenteral Viral Hepatitis: Epidemiology, Diagnosis, Prevention; State Educational Institution VUNMC of the Ministry of Health of the Russian Federation"
13. *The CDA Foundation. Hepatitis C*
14. (2024)
15. (2022) "Global change in hepatitis C virus prevalence and cascade of care between 2015 and 2020: A modelling study" *Lancet Gastroenterol. Hepatol*
16. Kuzin, Pavlov, Semenov et al. (2004) "The spread of viral hepatitis among various population groups in the Republic of Sakha (Yakutia)" *J. Microbiol. Epidemiol. Immunobiol*
17. Zotova (2010) "Parenteral Viral Hepatitis in South Yakutia"
18. Kyuregyan, Isaeva, Kichatova et al. (2020) "Prevalence of hepatitis B and C markers among the apparently healthy population of the Kaliningrad Region" *Epidemiol. Infect. Dis. Curr. Probl*
19. Kyuregyan, Malinnikova, Soboleva et al. (1038) "Community screening for hepatitis C virus infection in a low-prevalence population"
20. Soboleva, Karlsen, Kozhanova et al. (2017) "The prevalence of the hepatitis C virus among the conditionally healthy population of the Russian Federation" *J. Infectology*
21. Kichatova, Kyuregyan, Karlsen et al. "Frequency of detecting markers of hepatitis C among the conditionally healthy population of Tatarstan Republic" *Therapy*
22. Saryglar, Isaeva, Kichatova et al. (2023) "Dynamics of the prevalence of hepatitis C infection markers among the conditionally healthy population of the Republic of Tyva" *J. Infectology*
23. Kichatova, Lopatukhina, Potemkin et al. (2024) "Epidemiology of Viral Hepatitis in the Indigenous Populations of the Arctic Zone of the Republic of Sakha (Yakutia). Microorganisms"
24. Kyuregyan, Soboleva, Karlsen et al. (2019) "Dynamic changes in the prevalence of hepatitis C virus in the general population in the Republic of Sakha (Yakutia) over the last 10 years" *Infect. Dis. News Opin. Train*
25. Reshetnikov, Khryanin, Teinina et al. (2001) "Hepatitis B and C seroprevalence in Novosibirsk, western Siberia" *Sex. Transm. Infect*
26. Abdourakhmanov, Hasaev, Castro et al. (1998) "Epidemiological and clinical aspects of hepatitis C virus infection in the Russian Republic of Daghestan" *Eur. J. Epidemiol*
27. Asadi Mobarkhan, Manuylov, Karlsen et al. (2023) "Post-vaccination and post-infection immunity to the hepatitis B virus and circulation of immune-escape variants in the Russian Federation 20 years after the start of mass vaccination"
28. Aisyah, Shallcross, Hully et al. (2018) "Assessing hepatitis C spontaneous clearance and understanding associated factors-A systematic review and meta-analysis" *J. Viral Hepat*
29. (2019) "Population Size of the Russian Federation by Gender and Age as of"
30. (2021) "Federal Service for Surveillance on Consumer Rights Protection and Human Wellbeing. Sanitary Rules and Regulations 3.3686-21: Sanitary and Epidemiological Requirements for the Prevention of Infectious Diseases"
31. (2021) "Order No. 378n of July 19, 2024 on Amendments to the Procedure for Preventive Medical Examination and Dispensary Monitoring Approved by Order No. 404n of April 27"
32. (2017) "On Amendments to the Government Decree No. 1640 of 26 December"
33. "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=PMC12439802&blobtype=pdf | # Detection of clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus in a sheep in Great Britain, 2025
Ashley Banyard, Holly Coombes, Jacob Terrey, Natalie Mcginn, James Seekings, Benjamin Clifton, Benjamin Mollett, Cecilia Genova, Pia Sainz-Dominguez, Laura Worsley, Raquel Jorquera, Elizabeth Billington, Edward Fullick, Audra-Lynne Schlachter, David Jorge, Alejandro Núñez, Marco Falchieri, Joe James, Scott Reid
## Abstract
Clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus (HPAIV) continues to pose a significant global threat, affecting wild and domestic avian, and mammalian species. In early 2024, H5N1 HPAIV was detected in dairy cattle in the United States of America, where it has continued to circulate, with sporadic detections also reported in other ruminant species. The detection of high viral loads in milk from infected cattle, resulted in several human infections, underscoring the zoonotic potential of these viruses. In response, several countries have intensified surveillance in non-avian species to evaluate the potential for undetected viral circulation in captive mammals. In Great Britain, bulk milk tank testing of cattle and targeted surveillance of captive mammalian species on an infected premises is undertaken in accordance with the outcome of a rapid risk assessment. This assessment is undertaken to determine epidemiological links between the poultry and captive mammals. A result of this testing was the first recorded detection of clade 2.3.4.4b H5N1 HPAIV in a sheep in March 2025, identified on an infected poultry premises housing ducks, chickens, turkeys and geese in Great Britain. An initial seropositive result in a single ewe triggered further investigation, confirming serological positivity across repeated sampling and the presence of viral RNA in milk samples. This detection was confined to a single animal and is likely attributable to proximity to infected poultry and a presumed heavily contaminated environment. The implications of this detection in a ruminant host are discussed in the context of interspecies transmission and surveillance strategies.
## Introduction
Clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus (HPAIV) continues to cause significant infection and mortalities in both wild birds and poultry globally [1,2]. The impact of the H5N1 panzootic has been profound, with outbreaks decimating wild bird populations, the poultry sector, and populations of terrestrial and marine mammals where spillover has occurred [2,3]. High mortality rates are frequently observed among wild animals, significantly affecting avian species including those of conservation concern [2,4]. Wild bird mortalities increase environmental pressure which in turn increases the risk of infection for wild scavenging species, including terrestrial and aquatic mammals [3]. Alongside the infection of wildlife, infection of captive mammals has also been reported in zoological collections, household settings and farming sectors. Such outbreaks have included captive bush dogs [5], farmed mink and foxes in Europe [6,7], household cats [8], and most notably outbreaks in farmed dairy cattle in the United States of America (USA) [9,10].
From a ruminant perspective, H5N1 HPAIV (genotype B3.13) was first reported in dairy cattle in the USA on 25th March 2024 [11]; however, phylogenetic analysis indicates initial spillover from wild birds into cattle in late 2023. Since this detection, B3.13 has continued to spread in cattle, forming a discrete lineage, yet apparently becoming extinct in wild birds [12]. The widespread infection of dairy cattle has significantly elevated the zoonotic risk posed by H5N1 HPAIV, largely due to; (i) the observation of high viral titres present in milk from infected cattle, and (ii) concerns around the virus acquiring mammalian adaptations which may bestow increased fitness in humans [10,13,14]. While genotype B3.13 was still circulating in cattle, a second H5N1 HPAIV genotype was also reported in cattle in the state of Nevada, USA on the 31th January 2025 [15], and separately in the state of Arizona, USA on 13th February 2025 [16]. Both of these detections were with genotype D1.1 predominantly circulating in wild birds across the USA [17]. While the spillover dynamics of D1.1 remains to be elucidated, these detections clearly indicate multiple separate incursions from birds to cattle have occurred. To date, HPAIV (mainly of genotype B3. 13) has been detected in 17 states in over 1000 dairy herds across the USA. Zoonotic risk has increased with detections in dairy cattle, attributable to very high titres of virus in milk of infected cattle and it is of note that of the 70 cases of human infection reported in the USA with H5N1, 41 of these have been associated with exposure to infectious material in a dairy setting [11]. Alongside dairy cattle, H5N1 has also been sporadically reported in other farmed mammals including alpacas, pigs and goats [12,18], as well as captive mammals present on farms such as cats [19]. Currently there have been no previous reported detections of H5 HPAIV in sheep globally.
To date, only two H5N1 HPAIV genotypes (B3.13 and D.1.1) have been detected in dairy cattle globally, and both are exclusively found in wild birds within the Americas (mainly the USA) [18,20]. However, experimental studies have shown that the infection of cattle is not constrained by the genetic composition of these viruses; both American and European strains can induce similar infection and disease characteristics in cattle and other mammalian species [21]. These findings, alongside the scale of H5N1 HPAIV infections in USA cattle, have driven many countries to enhance surveillance for HPAIV in mammals, particularly where there is a higher risk of exposure to infectious material on infected poultry premises. However, to date, H5N1 HPAIV has not been detected in cattle outside of the USA [22][23][24][25]. In Great Britain (GB), national bulk milk tank sampling of a representative proportion of dairy cattle herds was undertaken. Over 500 bulk-milk samples, from 455 farms across GB, were all found negative for HPAIV [26]. In addition, targeted surveillance is being undertaken for captive mammals co-located with infected poultry where the risk of exposure to HPAIV is assessed as greater than the background risk posed by wild bird populations. Thus far in GB, the detection of influenza of avian origin in mammalian species has been limited to sporadic infections of wild mammals [27], including detections in foxes and otters alongside infection of marine mammals, predominantly seal species [27]. So far there has only been a single detection in captive mammals in GB, with H5N1 HPAIV being detected in a single group of captive bush dogs within a zoological collection in November 2022 [5,28]. However, as the clade 2.3.4.4b H5N1 panzootic continues to impact both wild and captive avian species, the risk to mammals co-located with infected poultry remains.
Here, the detection of H5N1 in a backyard flock including chickens, ducks, turkeys and geese is described whereby virus detection in avian species led to sampling of mammals with the resultant detection of infection with clade 2.3.4.4b H5N1 HPAIV in a single sheep. Factors leading to this detection are described.
## Materials and methods
## Clinical investigation, post-mortem examination, tissue sampling and histopathological analysis
Samples from birds were taken as described previously [29]. Samples from a proportion (38%; n = 10/26) of the sheep were initially taken to evaluate infection status with oral and rectal swabs being taken alongside blood sampling. Later sampling included individual animal composite milk samples where serological data had indicated infection (Supplementary Table 4). Following molecular and serological detection, a full postmortem (PM) examination and extensive tissue sampling of all major organ systems was undertaken on the affected ewe. Tissue samples were split for homogenization and histopathological assessment. For the former ∼0.5 grams of each tissue was processed into Precellys Hard tissue Homogenizing Tubes (CK28; Bertin Technologies, France) and 0.5 ml of phosphate-buffered saline (PBS) was added. Samples were homogenized for 2 min at 7000 RPM using Precellys24 homogenizer (Bertin Technologies, France). Supernatants were centrifuged at 10,000 xg for 30 s, and clarified supernatants were processed for extraction of viral RNA (vRNA) for molecular analyses [30].
For histopathology, samples were fixed in 10% (v/v) neutral buffered formalin for a minimum period of 5 days before being routinely processed and embedded in paraffin wax. Twenty-two different wax blocks were prepared from mammary gland (11 for each left and right mammary gland) Four-micron thick serial sections were either stained with haematoxylin and eosin (H&E), or for immunohistochemistry (IHC), using a mouse monoclonal anti-influenza A nucleoprotein (NP) antibody (Statens Serum Institute, Copenhagen, Denmark) as described previously [30]. The overall distribution of virus-specific staining in each tissue was assessed using a previously established criteria modified from [31]. Specificity of immunolabelling was assessed using positive control sections and by replacing the primary antibody with a matching mouse IgG isotype in test sections; no non-specific cross-linking was observed. Additionally, Gram Twort staining was performed in mammary gland sections.
## Virological investigation
## RNA extraction and molecular analysis
RNA was extracted from samples using the MagMAX CORE Nucleic Acid Purification Kit (ThermoFisher Scientific) as previously described [30] with tissues being homogenised in L-15 buffer prior to extraction. Extracted RNA from swabs and tissues was assessed for vRNA using Real-Time Reverse Transcription-Polymerase Chain Reaction (RT-PCR). RT-PCR assays used were matrix (M) gene specific [32], a H5 HPAIV specific [33] and/or NA specific [34,35] gene. RT-PCR Cq values < 36.00 were considered as AIV positive. Samples with Cq ≥36 were considered negative [30].
## Attempted virus isolation
For virus isolation, 100 µl of the sample material was diluted with 100 µl of PBS and inoculated into the allantoic cavity of three specific pathogen-free (SPF) embryonated chicken eggs (ECEs), following established protocols [36,37]. At two days post-inoculation (dpi), the allantoic fluid from one ECE was tested for the presence of virus using the haemagglutination assay as previously described [36]. Fluids exhibiting haemagglutination activity ≥16 were considered positive, whereas those with activity <16 were classified as negative. Allantoic fluids testing negative at 2 dpi were subjected to a second passage (passage 2) by inoculation into additional ECEs under the same conditions. At six dpi, allantoic fluid from all ECEs was harvested and tested by HA. Samples yielding negative haemagglutination activity following this protocol were deemed negative for infectious virus [45].
Attempts were also made to concentrate virus present in milk by the addition of chicken red blood cells (cRBCs), following a previously published protocol [38]. Briefly, 10 µl of packed cRBCs were added to 1 ml of milk collected from the positive sheep and mixed by gentle inversion. The samples were incubated at 4 °C for 1 hour, after which the cells were pelleted by centrifugation at 1500 rpm for 5 minutes at 4 ° C. The resulting cell pellet was resuspended in 200 µl of Leibovitz medium (LM) and used for virus isolation by inoculation into ECEs as described above.
## Serological assessment
Serum was separated from blood samples and heated at 56 °C for 30 min before being treated with receptor destroying enzyme (RDE) as described previously [39]. Hemagglutination inhibition assays were performed on processed serum samples as described previously using A/Teal/England/7394-2805/2005 (H5N3), A/chicken/Scotland/1959 (H5N1), A/ Chicken/Wales/053969/2021 (H5N1) (clade 2.3.4.4b H5N1) antigens all at 4 hemagglutination units [36]. In addition, batches of serum which included an HI positive sample were run on two commercial ELISA tests were used as per manufacturer's instructions. These included (i) ID Screen® Influenza A Antibody Competition Multi-species ELISA (ID Screen® Influenza A Antibody Competition Multi-species -Innovative Diagnostics), a competitive ELISA for the detection of antibodies against the NP of the Influenza A virus in multiple species and (ii) the ID Screen® Influenza H5 Antibody Competition 3.0 Multi-species ELISA (ID Screen® Influenza H5 Antibody Competition 3.0 Multi-species -Innovative Diagnostics), a multi-species competitive ELISA for the detection of antibodies against the H5 hemagglutinin of the Influenza A virus. All ELISAs were undertaken as per manufacturer's instructions. Sheep sera were diluted 1/5 through addition to receptor destroying enzyme (RDE).
## Genomic analysis
Viral genome sequences were generated from the composite milk sample using Oxford Nanopore Technology, by adapting a method previously described [40,41]. For all avian samples extracted vRNA was converted to double stranded cDNA and amplified using a one-step RT-PCR using SuperScript III One-Step RT-PCR kit (Thermo Fisher Scientific). Extracted RNA from the composite milk sample was converted to double stranded cDNA and amplified in triplicate. All three aliquots of amplified cDNA were pooled and concentrated using an ethanol precipitation. Briefly, 5 µL 5M NaCl and 1 µL glycogen was added to 50 µL cDNA and mixed by vortexing. Then 165 µL of 100% ethanol was added and the sample incubated at -20°C for a minimum of 30 minutes. The cDNA was pelleted by centrifugation at 1500 rpm for 5 minutes at 4 °C. The resulting pellet was washed with 70% ethanol and resuspended in sterile H 2 O.
Due to difficulty in generating complete sequences for the milk polymerase acidic (PA), polymerase basic 1 (PB1) and haemagglutinin (HA) segments from the initial sequencing attempts, these segments were converted to cDNA and amplified individually using segment specific primers (Supplementary Table 1).
Amplified cDNA was purified with Agencourt AMPure XP beads (Beckman Coultrer) prior to sequencing library preparation using the Native Barcoding Kit (Oxford Nanopore Technologies) and sequenced using a GridION Mk1 (Oxford Nanopore Technologies), according to manufacturer's instructions.
Assembly of the influenza A viral genomes was performed using a custom in-house pipeline publicly available here (APHA-VGBR/WGS_Pipelines/deno-voAssembly_ONT_Public.sh). Raw reads from the composite milk, PB1, PA and HA samples were combined prior to assembly. Comparison of the studyderived sequences and contemporary H5 sequences was undertaken against all avian H5 sequences available on GISAID between 1st January 2020 and 23rd April 2025. All sequences were aligned on a per segment basis using Mafft v7.520 [42] and trimmed against a reference using SeqKit v2.5.1 [4]. The trimmed alignments were used to a infer maximumlikelihood phylogenetic tree IQ-Tree version 2.2.3 [43] along with ModelFinder [40] and 1,000 ultrafast bootstraps [42]. Ancestral sequence reconstruction and inference of molecular-clock phylogenies were performed using TreeTime v0.10.1 [44]. Phylogenetic trees were visualized using R version 4.3.3 with ggplot2 [45] and ggtree version 3.14.0 [46]. Sequences were genotyped from phylogenetic trees by comparison to known reference sequences for all genotypes currently circulating in the UK.
Sequences derived from both sheep and associated poultry outbreak were assessed for the presence of adaptive mutations that may confer increased replication in mammals. All sequences were aligned on a per segment basis using MAFFT v7.520 [42] and manually trimmed to the open reading frame using Aliview version 1.28 [47]. Trimmed sequences were translated to amino acids and visually inspected for mutations. All influenza sequences generated in this study are available through the GISAID EpiFlu Database (https:// www.gisaid.org, Supplementary Table 4).
## Results
On 18 th February 2025 HPAI H5N1 was confirmed on a backyard mixed poultry holding in Yorkshire. The infected premises was a flock (n = 60 poultry) consisting of 34 chickens (Gallus gallus domesticus), 5 turkeys (Meleagris gallopavo), 19 ducks (Anas platyrhynchos domesticus) and 2 geese (Anser anser domesticus) (Figure 1a andb). The birds had been housed since the housing order came into effect on 23rd December 2024 as the infected premises lies within the Avian Influenza Prevention Zone (AIPZ) [48]. The poultry were housed in four groups; Group A comprised 31 chickens in a stable, Group B comprised 19 ducks and two geese in another stable, Group C comprised five turkeys in another shed, and Group D comprised 3 chickens in a wooden coop in the garden. Additionally, the premises had one household cat (Felis catus) and 26 sheep (Ovis aries) (16 breeding ewes, 9 lambs and 1 ram). At the time of the report case (Figure 1a), one group of sheep were housed next to the turkey shed (Group E), another group (which included the affected ewe) were housed next to the duck and goose shed (Group H) with the remainder in the field adjacent to the garden (Group I). There were no linked premises nor commercial activities reported for this premises. Eggs and meat from the premises were only consumed by the keeper within the window of suspected infection.
Although the birds were housed indoors, overall biosecurity at the premises was assessed as poor. Structural deficiencies such as gaps in roofs, doors, and windows were identified, and no disinfectant footbaths were in use either at the entrance to the premises or between different sheds. The poultry sheds were not secure against wildlife ingress; entry points such as damaged windows, holes in the roof, and visible gaps in gates allowed potential access. Evidence of rat activity was also present, and the keeper reported observing a rat feeding on the head of a deceased chicken, although no samples were collected from rats on site. The birds were bedded daily, and the interiors of the poultry sheds were clean and dry. Mains-supplied drinkers were available within each shed. Although wild bird activity on site was reported to be low, sightings of wild birds flying overhead had been noted and there were bird feeders in the garden area near the sheds housing the poultry and sheep. Following the onset of avian mortality, recently deceased birds were stored in a garden waste bin.
The infection event in avian species began with initial signs of inappetence observed on 10 th February 2025 in both the ducks and chickens, accompanied by a decrease in egg production among the chickens (Figure 1c). By 16 th February 2025, one chicken and one duck had died, and one of the two geese was exhibiting neurological signs consistent with clinical disease, including head tremors. On 17 th February 2025, the clinically affected goose was found dead, and a further two chickens and one duck had succumbed to the infection. The owner reported suspicion of disease on 17 th February 2025 (Figure 1c). An official veterinarian from the Animal and Plant Health Agency (APHA) attended the premises the same day. Upon inspection, two of the four bird groups (Groups A and B) were found to be affected (Figure 1a). Two ducks displayed clinical signs suggestive of a notifiable avian disease (NAD); one exhibited "snicking" (a respiratory sign analogous to sneezing in humans), while the other appeared lethargic with ocular discharge. Additionally, one chicken was noted to be lethargic and was euthanised. A PM examination of that chicken revealed no significant gross lesions. However, NAD could not be ruled out, and samples were collected accordingly. These included oropharyngeal and cloacal swabs from 8 chickens (Group A), with 2 whole heads submitted for laboratory brain sampling; oropharyngeal and cloacal swabs from 11 ducks, along with 2 blood samples (from duck 18 and 19); and oropharyngeal and cloacal swabs from 1 goose (Group B). One of the clinically affected ducks died later on 17 th February 2025, and by the following day, the remaining clinically affected duck and the affected chicken had also died (Figure 1c).
Swab samples tested positive by RT-PCR for the influenza A virus (M-gene), H5 HPAIV (HP-H5), and N1 vRNA in 50% (4/8) of the chickens, 100% (11/11) of the ducks, and the single goose (1/1) (Supplementary Table 2). All swabs tested negative for type 1 avian paramyxovirus (APMV-1; referred to as Newcastle disease [ND] when virulent strains are present in poultry), as determined by RT-PCR targeting the L gene. Serological testing of duck blood samples by HAI revealed one (duck 18) out of two samples to be positive for H5 antibodies against clade 2.3.4.4b H5N1 antigen, with am HI titre of 1/64. These laboratory findings, in conjunction with the observed clinical signs, led to the official confirmation of H5N1 HPAI in avian species on the premises on 18th February 2025 (Figure 1c). Subsequent genetic sequencing identified all poultry genomes (Supplementary Table 3) as a clade 2.3.4.4b H5N1 HPAIV, belonging to the DI.2 genotype, in accordance with the genotype classification system established by the European Union Reference Laboratory (EURL) for Avian Influenza [49].
In accordance with protocols established in response to detections of HPAIV in captive dairy cattle in the USA, an epidemiological assessment was performed for this premises to determine whether any captive mammalian species present were at increased risk of exposure to HPAIV beyond the background level of risk posed by wild birds. Following this assessment, the sheep at this premises were triaged and sampled for diagnostic testing for the presence of HPAIV (Figure 1b).
On the 7 th of March, ten of the 26 sheep were sampled with nasal and rectal swabs being taken as well as a blood sample; the remaining animals were not sampled due to welfare concerns, as they were heavily pregnant. Nasal and rectal swabs collected from the sampled sheep (n = 10) (across Group H, G and F; Figure 1b) tested negative for vRNA by RT-PCR using the M-gene, H5-HP and N1 assays. Notably, serological testing of blood samples revealed that one ewe in Group H (Figure 1b) had detectable H5specific antibodies, with a HI titre of 1:80. Two additional ELISAs also returned positive results from this serum, indicating the presence of antibodies reactive to NP and H5 antigens, respectively (see Supplementary Table 4). This initial positive serological detection led to repeat sampling of that sheep which was undertaken on the 14 th of March 2025 and although the animal was negative for vRNA from nasal and rectal swabs, the sheep was again positive for H5 reactive antibodies by HAIT (having a titre of 1/160), and the serum was positive by both ELISA tests. A composite milk sample (collected from both halves of the udder) tested positive for H5N1 HPAIV vRNA by M gene, HPH5 and N1 PCR assays. The milk sample was also assessed by competitive ELISA for NP and H5 and was positive in both tests.
Following the detection of H5-specific antibodies and HPAIV vRNA in the milk, the affected ewe was culled on the 19 th of March 2025, and a full PM examination was undertaken. At this time, the ewe remained lactating, and milk was again expressed from both teats. No abnormalities in milk consistency or presence of clots were observed. Gross PM examination noted the left mammary gland to be slightly firm and mildly reddened; however, these findings were not considered clinically significant. Milk was collected post-euthanasia and post-swabbing, with samples from both halves of the udder pooled into a single container. Notably, this milk sample tested positive for both anti-NP and anti-H5 antibodies. All tissue samples taken at PME (n = 28) and swab samples (n = 12) (Supplementary Table 5) tested negative for HPAIV RNA by M gene, H5-HP, and N1 RT-PCRs. However, another milk sample from the ewe tested positive for H5N1 vRNA across all three RT-PCR assays, consistent with previous findings (Supplementary Table 5). Subsequently, tissue samples were further homogenised and reassessed for vRNA. Borderline detection of vRNA was observed in the homogenised right mammary gland tissue, but this result could not be consistently reproduced upon repeat RNA extraction and testing. Histological examination was conducted on these tissues. No lesions consistent with active mastitis or viral replication within the glandular tissue were observed, nor were there any definitive histological features indicative of prior viral exposure. Histopathological examination revealed areas of multifocal lymphoplasmacytic infiltration around alveoli and ducts in one half (left teat) (suggestive of chronic or residual inflammation). In contrast, multifocal regions of the other half exhibited acute neutrophilic alveolar exudation (more consistent with a recent inflammatory insult) (right teat), but not showing intraluminal epithelial sloughing and cellular debris in mammary alveoli described in active infection in cattle. IHC for the presence of viral antigen was also performed on mammary tissue sections; however, all sections were negative for viral antigen. Gram staining (Twort counterstain) of mammary gland samples did not reveal the presence of bacterial colonies (no further bacteriological studies were conducted).
Virus isolation was attempted on both H5N1 HPAIV positive milk samples using ECEs. However, infectious virus could not be isolated. We also attempted to sequester and concentrate potential virus in the milk samples using chicken RBCs. However, no infectious virus could be isolated following this treatment.
Full sequences were obtained from the milk sample for all segments except the PB1, which contained a large gap between nucleotides 333 and 786, as well as some smaller gaps between 190 and 330. The resulting sequence was compared to those derived from the infected birds and identified as a genotype DI.2 H5N1 clade 2.3.4.4b virus, clustering with other DI.2 isolates detected in GB during the same period (SI Figure 1). Further analysis of positive material from all infected birds (n = 21) (Supplementary Table 2) on the premises revealed >99.9% sequence identity across all segments. The viral sequence from the sheep contained nine amino acid substitutions relative to the avian viral sequences (Supplementary Table 6). Two changes were identified in the HA protein (D171N and D277G); neither is associated with increased zoonotic potential, although D171N has been previously reported in genotype B3.13 H5N1 sequences from cattle in the USA and may represent a ruminantassociated adaptation (Figure 2) [12,50]. Additional substitutions distinguishing the sheep and avian sequences included PB2 N456D, NA L75F and V114M, PA L335F and PB1 M290 V, K577E, and Q688H. Notably, previously reported cattle-adaptive mutations, such as PB2 E362G, M631L, and E677G; PB1 N642S; and PA A448S [12,51], were absent from the sheep-derived sequence. No mutations were detected in either the NP, MP or NS genes when compared with the avian reference sequence (Figure 2). The viral sequence from the sheep was also compared to publicly available sequences isolated from US goats in 2024. None of the sheep-associated mutations were present in the goat sequences.
## Discussion
The detection of clade 2.3.4.4b H5N1 vRNA in a sheep represents the first reported case of infection with this virus in this species globally, and the first detection in a ruminant outside of the USA. Previous detections of clade 2.3.4.4b H5N1 in mammalian species have generally been attributed to elevated environmental infection pressure, typically associated with large-scale mortality events in wild bird populations [2,3]. In such scenarios, transmission pathways are often apparent, particularly among scavenging species, where these mammals are presumed to have acquired infection through close contact with infected carcasses, most plausibly via the ingestion of dead or sick wild birds [2,3].
However, this unusual detection raises the possibility of alternative routes of infection beyond the typical scavenging-behaviour related exposures. The ewe was co-located on an infected holding where chickens, ducks, and geese had tested positive for H5N1, and had been housed since December 2024 in a shed directly adjacent to the one housing both the ducks and geese (Group B). Initial avian testing revealed widespread vRNA shedding among the birds, via both cloacal and oropharyngeal routes. Importantly, the presence of infection in Anseriformes species (ducks and geese) increases the likelihood of a heavily contaminated environment, given their known capacity for extensive environmental viral shedding [30]. Multiple studies have shown that although chickens can efficiently transmit avian influenza viruses within dense commercial flocks, they typically shed lower quantities of infectious material when compared to ducks [30,52,53]. Experimental work consistently demonstrates that infected ducks release high viral loads into the environment, contaminating water and surfaces extensively. Based on this evidence, the lack of mitigating biosecurity measures in place between the poultry and the sheep managed by a single keeper who oversaw their daily management, and the common DI.2 sequence results isolated from the poultry and the ewe, it is reasonable to infer that the ewe may have acquired infection through indirect contact with the infected poultry shedding large amounts of HPAIV into the environment. However, another plausible hypothesis for source of HPAIV infection for the ewe considered was contact with infected wild birds or vermin. Indirect contact with wildlife was identified as the most likely source of infection for the poultry on this premises. This was supported by evidence including gaps in the shed doors and roofs which could have allowed for ingress by wild birds, associated faecal material or contaminated water, poor site biosecurity, evidence of rat activity on the premises, site features (including bird feeders and a pond) which may have attracted wild birds and the occurrence of both several other infected poultry premises and wild bird HPAIV detections in the surrounding area. It is possible that these same pathways and sources of infection are implicated for the ewe as well as the poultry. However, the weight of the above evidence reasonably supports the conclusion that poultry were more likely to be the source of infection for the ewe than wild birds.
Despite these findings, the precise route of infection in the ewe remains undetermined. However, several potential pathways are plausible, including (i) asymptomatic infection though the respiratory or oral routes via contaminated fomites, (ii) intramammary infection via contaminated material introduced into the teat by suckling lambs, or (iii) direct inoculation into the mammary gland as a result of the ewe lying in virus-contaminated material. Although none of these scenarios can be definitively ruled out, asymptomatic respiratory infection followed by systemic viral dissemination resulting in vRNA detection in milk appears unlikely. Recent studies have proposed that transfer of infectious virus between animals may occur during suckling with oral tissues of cattle supporting virus binding and replication [54], however, we did not detect infection of lambs during this study. Analysis of airborne virus present in commercial poultry sectors has demonstrated relatively low levels of virus present even within densely packed houses where assessed [55], but nevertheless various strains of H5N1 have been shown to transmit between avian species and/or between mammal species in experimental infections. While the infectious dose for sheep is unknown, considering the density of infected poultry present, the level of virus present in the air was likely low. In addition, if this had occurred, one would expect to observe evidence of systemic infection, such as seropositivity in other sheep or lambs; however, all animals tested were seronegative. HAIT was the frontline assay used to assess seroconversion and unfortunately, as samples were of a low volume, further assessment, such as virus neutralization tests, could not be undertaken. Of note, albeit anecdotal, the ewe with detectable vRNA in milk had a reported episode of mastitis on the 3 rd of March 2025. There was no evidence that prior unrelated infections may have reduced immune status in the affected ewe although with the report of mastitis in that animal in the period before this positive detection it is not possible to rule this out. No additional clinical signs or behavioural abnormalities were reported in the sheep following detection of infection in poultry. Interestingly, the detection of anti-NP and anti-H5 antibodies, indicating that the ewe was likely sampled during a phase of viral clearance, may explain the failure to isolate infectious virus from the collected samples.
For the second hypothesis (transmission via suckling lambs) to be plausible, one would expect some evidence of infection in the lambs, either through direct mucosal contact with infectious material or ingestion of contaminated milk. However, all lambs tested negative by both swab and serological assays, making this route of transmission less likely. Therefore, the third hypothesis appears most consistent with the available data. Experimental studies have shown that intramammary inoculation with both North American and European genotypes of H5N1 can lead to infection in cattle and is an efficient route of infection, albeit one that generally results in infection restricted to the udder [21]. In such cases, infection is frequently restricted to a single mammary quarter, without resulting in systemic dissemination. Unfortunately, in this case, milk samples were pooled from both teats, precluding any assessment of whether the infection was localized to one mammary gland. This limitation hinders the ability to definitively determine the anatomical extent of infection.
The histopathological findings were unable to provide clear evidence of active viral infection and where significantly different between both halves of the mammary gland, While in one of them subacute to chronic changes may have been consistent with a prior viral infection, the neutrophilic alveolar exudation in the other half would suggest a more recent insult or recent clearance, as no epithelial sloughing and cellular debris in the mammary alveoli associated with active infection [10] were seen. Although extensive sampling and examination of the mammary gland was conducted, some areas where virus was replicating may have been present elsewhere. Further data limitations arose due to logistical constraints associated with sampling pregnant animals. The detection of seropositivity in a single ewe highlighted the need for broader investigation within the flock. However, due to the flock's lambing status, comprehensive sampling of the entire flock could not be completed and only the initially seropositive ewe was subjected to further diagnostic evaluation at the time. All the sheep in the flock were ultimately sampled at least once, and all with negative results, although testing was protracted in the case of four ewes who did not lamb until late April and were only sampled on 26 th April 2025. Additionally, environmental sampling across the premises would have provided valuable insight into the degree of environmental contamination to which other mammals may have been exposed following the avian cull.
The viral sequence obtained from the ewe provides the only definitive evidence that replication occurred within this host. Comparative genomic analysis revealed several amino acid substitutions across multiple gene segments in the virus isolated from the sheep. These changes were absent in viral sequences derived from infected birds on the same premises, which exhibited a high degree of genetic identity among themselves. This divergence strongly suggests that the virus underwent replication and genetic adaptation within the ewe, consistent with host-specific evolutionary pressure. However, the functional significance of these mutations, particularly regarding viral fitness, host range, or transmissibility, remain unknown. Further virological and in vivo studies are required to assess the potential implications of these sequence changes for cross-species transmission and adaptation.
In conclusion, this study has demonstrated that it is possible for mammalian species co-located on an infected premises to become infected with these viruses through unusual infection routes and that appropriate risk assessments and both sampling and testing of mammals in such scenarios is important to understand where infection may have occurred. A recent report found that blood samples collected in 2024 from 220 sheep, which had grazed in areas affected by dead and diseased H5N1-infected pheasants in Norway during 2023, contained antibodies against H5 AIV [56]. Interestingly no clinical disease was reported in these sheep [56]. This supports the hypothesis that such spillover events may be more common than previously recognized. However, comprehensive surveillance for HPAIV is not routinely conducted in the ruminant sector. Should future cases arise whereby suspicion of infection of ruminant species is suspected then it would be critical to have better access to animals for sampling. On this occasion, the fact that sheep were in lamb significantly reduced our ability to sample the herd. Having access to a greater range of sample types, and volumes would allow confirmatory assessments to be made across the herd during the disease event. Certainly, with continued detections of high pathogenicity avian influenza across a broad range of wild and captive mammals globally it is important that a full characterization and assessment of spillover events is undertaken and shared with the global scientific community.
## References
1. Krammer, Hermann, Rasmussen (2025) "Highly pathogenic avian influenza H5N1: history, current situation, and outlook" *J Virol*
2. Bellido-Martín (2025) "Evolution, spread and impact of highly pathogenic H5 avian influenza A viruses" *Nat Rev Microbiol*
3. Peacock (2025) "The global H5N1 influenza panzootic in mammals" *Nature*
4. Tremlett, Morley, Wilson (2024) "UK seabird colony counts in 2023 following the 2021-22 outbreak of Highly Pathogenic Avian Influenza" *RSPB Res Rep*
5. Falchieri (2024) "Rapid mortality in captive bush dogs (Speothos venaticus) caused by influenza A of avian origin (H5N1) at a wildlife collection in the United Kingdom" *Emerg Microbes Infect*
6. Agüero (2022) "Highly pathogenic avian influenza A(H5N1) virus infection in farmed minks" *Eurosurveillance*
7. Kareinen (2023) "Highly pathogenic avian influenza A(H5N1) virus infections on fur farms connected to mass mortalities of black-headed gulls" *Euro Surveill*
8. Rabalski (2023) "Emergence and potential transmission route of avian influenza A (H5N1) virus in domestic cats in Poland" *Euro Surveill*
9. Garg (2024) "Outbreak of highly pathogenic avian influenza A(H5N1) viruses in U.S. dairy cattle and detection of Two human cases -United States" *MMWR Morb Mortal Wkly Rep*
10. Caserta (2024) "Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle" *Nature*
11. (2025) "CDC). H5 Bird Flu: Current Situation"
12. Nguyen (2025) "Emergence and interstate spread of highly pathogenic avian influenza A(H5N1) in dairy cattle in the United States" *Science*
13. Garg (2025) "Highly pathogenic avian influenza A(H5N1) virus infections in humans" *N Engl J Med*
14. Spackman (2024) "Characterization of highly pathogenic avian influenza virus in retail dairy products in the US" *J Virol*
15. (2025) "APHIS Confirms D1.1 Genotype in Dairy Cattle in Nevada"
16. (2025) "APHIS Identifies Third HPAI Spillover in Dairy Cattle"
17. (2025) "Confirmations of Highly Pathogenic Avian Influenza in Commercial and Backyard Flocks"
18. Burrough (2024) "Highly pathogenic avian influenza A(H5N1) clade 2.3.4.4b virus infection in domestic dairy cattle and cats, United States, 2024" *Emerg Infect Dis*
19. Naraharisetti (2024) "Highly pathogenic avian influenza A(H5N1) virus infection of indoor domestic cats within dairy industry worker households -Michigan" *MMWR Morb Mortal Wkly Rep*
20. (2025) "APHIS Confirms D1.1 Genotype in Dairy Cattle in Nevada"
21. Halwe (2025) "H5n1 clade 2.3.4.4b dynamics in experimentally infected calves and cows" *Nature*
22. Authority (2024) "Avian influenza overview" *EFSA J*
23. (2024) "Human Animal Infections and Risk Surveillance group (HAIRS)"
24. Fabri (2025) "No indication of highly pathogenic avian influenza infections in Dutch cows" *JDS Commun*
25. Friedrich-Loeffler-Institut, Fli) (2024) "Avian influenza: No evidence of H5N1 infection in dairy cows outside the USA"
26. (2024) "Diagnostic excellence in action: APHA's response to high pathogenicity avian influenza threats in cattle"
27. (2025) "Confirmed findings of influenza of avian origin in non-avian wildlife"
28. (2025) "Confirmed findings of influenza of avian origin in captive mammals"
29. Reid (2024) "A multi-species, multi-pathogen avian viral disease outbreak event: investigating potential for virus transmission at the wild bird -poultry interface" *Emerg Microb Infect*
30. James (2023) "Clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus (HPAIV) from the 2021/22 epizootic is highly duck adapted and poorly adapted to chickens" *J Gen Virol*
31. Löndt (2008) "Pathogenesis of highly pathogenic avian influenza A/turkey/Turkey/1/2005 H5N1 in Pekin ducks (Anas platyrhynchos) infected experimentally" *Avian Pathol*
32. Nagy (2021) "A universal RT-qPCR assay for "One Health" detection of influenza A viruses" *PLoS One*
33. James (2022) "Rapid and sensitive detection of high pathogenicity Eurasian clade 2.3.4.4b avian influenza viruses in wild birds and poultry" *J Virol Methods*
34. Slomka (2012) "Challenges for accurate and prompt molecular diagnosis of clades of highly pathogenic avian influenza H5N1 viruses emerging in Vietnam" *Avian Pathol*
35. James (2018) "Development and application of realtime PCR assays for specific detection of contemporary avian influenza virus subtypes N5" *Avian Dis*
36. (2019) "Terrestrial Manual: Avian influenza (infection with avian influenza viruses)"
37. Reid (2025) "Validation of a reduction in time for avian influenza virus isolation using specific pathogen-free embryonated chicken eggs"
38. Okuya (2015) "Isolation and characterization of influenza A viruses from environmental water at an overwintering site of migratory birds in Japan" *Arch Virol*
39. Bhat (2022) "Coinfection of chickens with H9N2 and H7N9 avian influenza viruses leads to emergence of reassortant H9N9 virus with increased fitness for poultry and a zoonotic potential" *J Virol*
40. Alexander (2023) "Investigating the genetic diversity of H5 avian influenza viruses in the United Kingdom from 2020-2022" *Microbiol Spectr*
41. Banyard (2024) "Detection and spread of high pathogenicity avian influenza virus H5N1 in the antarctic region" *Nat Commun*
42. Lean (2022) "Subclinical hepatitis E virus infection in laboratory ferrets in the UK" *J Gen Virol*
43. Minh (2020) "IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic Era" *Mol Biol Evolut*
44. Sagulenko, Puller, Neher (2018) "Treetime: maximum-likelihood phylodynamic analysis" *Virus Evolut*
45. Wilkinson, Wickham (2011) "Ggplot2: elegant graphics for data analysis by" *Biometrics*
46. Yu (2017) "Ggtree: an R package for visualization and annotation of phylogenetic trees with their covariates and other associated data" *Meth Ecol Evolut*
47. Larsson (2014) "Aliview: a fast and lightweight alignment viewer and editor for large datasets" *Bioinformatics*
48. (2024) "The Animal and Plant Health Agency (APHA) and The Department for Environment Food and Rural Affairs (DEFRA). Bird flu: avian influenza prevention zone (AIPZ) (England)"
49. (2025) "EURL, Avian Flu Data Portal"
50. Yang (2025) "The haemagglutinin gene of bovine-origin H5N1 influenza viruses currently retains receptor-binding and pH-fusion characteristics of avian host phenotype" *Emerg Microb Infect*
51. Dholakia (2025) "Polymerase mutations underlie early adaptation of H5N1 influenza virus to dairy cattle and other mammals" *bioRxiv*
52. Seekings (2021) "Highly pathogenic avian influenza virus H5N6 (clade 2.3.4.4b) has a preferable host tropism for waterfowl reflected in its inefficient transmission to terrestrial poultry" *Virology*
53. Seekings (2024) "Transmission dynamics and pathogenesis differ between pheasants and partridges infected with clade 2.3.4.4b H5N8 and H5N1 highpathogenicity avian influenza viruses" *J Gener Virol*
54. Shi (2025) "H5n1 virus invades the mammary glands of dairy cattle through "mouth-to-teat" transmission" *Natl Sci Rev*
55. James (2023) "The role of airborne particles in the epidemiology of clade 2.3.4.4b H5N1 high pathogenicity avian influenza virus in commercial poultry production units" *Viruses*
56. Veterinaerinstituttet (2025) "Forskningsprosjekt viser tidligere fugleinfluensasmitte hos en norsk sau (English translation: Research project shows previous bird flu infection in Norwegian sheep)" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12529347&blobtype=pdf | # Impact of IL10 polymorphism in chronic progression of hepatitis B virus infection
Yasmine Chelbi, Manel Hamdoun, Hamza Cherni, Hinda Triki, Ben Ahmed, Olfa Bahri
## Abstract
Chronic infection with the hepatitis B virus (HBV) remains a major public health issue. Its progression depends on several factors, including immunogenetic factors. The aim of this study was to investigate the association between interleukin 10 gene (IL10) polymorphism and the progression of this infection. This retrospective casecontrol study involved 156 chronic HBV carriers (CHBV-C) and 174 healthy HBV-negative controls (HBsAg-). The analysis of IL10 promoter polymorphism was carried out using the TaqMan allele discrimination technique at two single nucleotide polymorphisms (SNPs), -592A>C (rs1800872) and -1082A>G (rs1800896), of the IL10 promoter. IL10 levels were measured using an in-house Enzyme-Linked Immunosorbent Assay (ELISA) for all patients chronically infected by HBV who had not yet received treatment. Chronic HBV infection (CBI) was present in 32% (n = 43) of cases, 37% (n = 51) had active chronic hepatitis (ACH), and 31% had complicated hepatitis. The analysis of allele polymorphism identified six genotypes: AA (14%), AC (43%), and CC (43%) for SNP-592A>C, and AA (41%), AG (45%) and GG (14%) for SNP-1082A>G. The only genotype that was substantially more common in CHBV-C patients was -1082/GG (OR=1.9; CI95%=[1,3.62]; p = 0.046). When compared to controls, the IL10 level was significantly higher in CBI patients (3.27 vs. 2.56 pg/ml;p = 0.002). Significantly higher IL10 levels were also linked to the genotypes -1082 GG (6.02 pg/ml;p = 0.04) and -592CC (3.73 pg/ml;p = 0.039). With the -592 AA genotype, this level was noticeably lower (1.35 pg/ml;p = 0.014). These findings support the hypothesis that the development of chronicity in HBV infection is linked to elevated IL10 levels and the -1082 GG genotype in the gene's promoter.
## 1. Introduction
Hepatitis B Virus (HBV) infection is a major public health issue worldwide. Every year, about 1.23 million new cases of infection are reported, with 63 % of these infections occurring in Africa (World Health Organization, 2024). About 5 % of adult-aged HBV infections progress to chronicity and persist; in 2016, the estimated prevalence of hepatitis B surface antigen (HBsAg) was 291 million worldwide (Shi and Zheng, 2020). Chronic HBV infection represents a considerable challenge leading to severe complications, such as hepatocellular carcinoma (HCC) and cirrhosis, and contributing to substantial morbidity and mortality. Indeed, about 1.1 million deaths were reported globally in 2022, according to the World Health Organization (WHO) estimates (World Health Organization, 2024).
Several factors leading to progression to chronicity of HBV infection were studied, such as viral and immunogenetic factors. Indeed, it has been shown that an inadequate immune response is a major factor in the long-term development of infection (Shi and Zheng, 2020;Ribeiro et al., 2022). In individuals with a resolutive infection, memory T and B cells effectively maintain immune surveillance, reacting to the virus quickly and inhibiting replication. In a chronic HBV infection, long-term exposure to elevated viral antigens and an immunosuppressive liver milieu maintains T and B cell dysfunction (Shi and Zheng, 2020).
This environment is characterized by a specific cytokine pattern that controls the inflammatory response, which influences the evolution of HBV (Ribeiro et al., 2022). The literature examined a number of cytokines, including interferon gamma (IFN-γ), tumor necrosis factor alpha (TNF-α), interleukin 1β (IL1β), interleukin 2 (IL2), interleukin 4 (IL4), interleukin 6 (IL6), interleukin 10 (IL10), interleukin 17A (IL17A), interleukin 18 (IL18), and interleukin 28B (IL28B), that may be implicated in the pathophysiology of this infection (Ribeiro et al., 2022;Amirpour-Rostami and Kazemi Arababadi, 2019 ;Takahashi, 2014). The degree of cytokine production, based on variations in their promoter regions, appears to play a significant part in the progression of the infection. The primary processes that lead to the development of chronicity, however, remain unclear. In fact, various single nucleotide polymorphisms (SNPs) of several interleukin genes have been studied and were associated with HBV infection progression. IL10 is recognized to play a key role in regulating the immune response to HBV (Rybicka et al., 2020). SNPs located in the 5′ promoter region of the IL10 gene were demonstrated to be involved in the regulation of IL10 production. Among these SNPs, -1082 A > G (rs1800896) and -592 A > C (rs1800872) appear to be particularly relevant to HBV infection (Rybicka et al., 2020).
The aim of this study was to investigate whether the -1082 A > G and the -592 A > C SNPs in the promotor of IL10 gene is associated with the chronic progression of HBV infection in a Tunisian community.
## 2. Material and methods
In this retrospective case-control study, 330 patients were matched by age and sex and split into two groups: 156 chronically infected by HBV (CHBV-C) and 174 healthy HBV-negative controls (HBsAg negative). Chronicity was defined as the persistence of HBsAg for more than six months; patients were followed up at one of the three university hospital centers (UHCs): La Rabta UHC, the regional hospital of Bizerte and the regional hospital of Nabeul. A pre-established information form was filled out for each patient; the data primarily covered sociodemographic characteristics, clinical status, and biological data (plasma viral load and transaminase level). These patients were all previously tested negative for hepatitis C virus (HCV), hepatitis D virus (HDV), and human immunodeficiency virus (HIV) serological markers. Individuals from the control group were previously tested negative for HCV and HBV serological markers (HBsAg and anti-HCV antibodies). Blood samples were collected for the studied population in EDTA tubes; following centrifugation, plasma was kept at -20 • C for IL10 titration and pellets at -40 • C for IL10 polymorphism analysis.
## 1. IL10 titration
Patients with CBI who had not yet received therapy (n = 37) had their IL10 levels assessed. The immunology lab of the Pasteur Institute of Tunis has created and regularly uses an in-house sandwich Enzyme-Linked Immunosorbent Assay (ELISA) technology. The process involved adding an anti-IL10 antibody diluted in a carbonate/bicarbonate solution to the wells of a Maxisorb ELISA microplate. An overnight incubation at +4 • C was carried out to allow for the fixation of the antibodies. After a series of three washes to remove unbound substances, a saturation step was performed for 2 h at room temperature using PBS supplemented with 10 % fetal calf serum. After washing again, the standards and test sera were left at room temperature for 2 h so that IL10 could attach to the capture antibody. Then, the conjugate, which contained biotinylated specific antibodies along with avidin linked to peroxidase, is added. The reaction was subsequently revealed by introducing the enzyme substrate, tetramethylbenzidine (TMB), with the reaction being halted by the addition of sulfuric acid, and the resulting optical densities (ODs) were read using a spectrophotometer at 450/630 nm. Commercial standards from BD Biosciences were used for the titration of IL10 in the tested samples. The reaction was validated by calculating the positivity threshold according to the following formula: threshold value (TV) is equal to the negative controls' mean OD plus 3 times their OD standard deviation.
## 2. IL10 genetic polymorphism
Genomic deoxyribonucleic acid (DNA) was manually extracted from peripheral leukocytes using the QIAamp®DNA Blood Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's recommendations. The extracted DNA was quantified by spectrophotometry in a NanoDrop ® system (Thermo Fisher Scientific, Wilmington, DE, USA). Two SNPs, -592A>C and -1082A>G, were analyzed using the allelic discrimination method in order to investigate the allelic polymorphism of the IL10 promoter. In this process, the two alleles for each SNP were distinguished using the TaqMan® SNP Genotyping Assays kit (Thermo Fisher Scientific, Massachusetts, United States) and TaqMan probes tagged with either VIC or FAM fluorophores. The alleles A and C for SNP-592A>C (rs1800872) and A and G for SNP-1082A>G (rs1800896) were correspondingly represented by VIC and FAM fluorophores. The polymerase chain reactions (PCR) were conducted using the ABI 7500 thermocycler (Applied Biosystems, California, United States) in a final volume of 20 µl, which contained 20 ng of DNA, TaqMan® Universal PCR Master Mix [02X], and [10X] of each labeled probe. The following were the PCR conditions: 60 • C for one minute, then 95 • C for ten minutes, then 40 cycles of 95 • C for fifteen seconds and 60 • C for one minute, ending with one minute at 60 • C. Data was collected at 60 • C steps. For the corresponding SNP, homozygosity is indicated by an increase in either the VIC or FAM signal alone, whereas heterozygosity is shown by an increase in both VIC and FAM.
## 3. Statistical analysis
Statistical analysis was performed using IBM SPSS Statistics version 13.0 (IBM Corp., Armonk, NY, USA). Qualitative variables were compared using Chi-square test or Fisher exact test. Quantitative variables were compared using the Mann-Whitney U test and Kruskall-Wallis test. Similarly, Hardy-Weinberg equilibrium was evaluated for the SNPs -592A>C and -1082A>G which showed that the observed genotypic frequencies did not differ from the expected theoretical frequencies for -592A>C and -1082A>G. A p-value less than 0.05 was considered statistically significant.
## 4. Ethical considerations
This study was conducted in accordance with the principles of the Declaration of Helsinki. It was approved by the ethical committee of Aziza Othmana Hospital. All samples studied were collected within the framework of routine diagnostics. Patients' data were processed anonymously.
## 3. Results
The average age of the study population was 46 years for CHBV-C and 47 years for the control group; the sex ratio was 0.90 (74 women and 82 men) and 1.23 (96 women and 78 men) for the two groups, respectively. Age and gender of patients and controls were compatible. CBI: Chronic HBV infection; ACH: chronic HBV active hepatitis; HCC: hepatocellular carcinoma; F/M: female to male; xN: x times the normal value; ND: non determined.
Table 1 shows the CHBV-C group's characteristics; Clinical status was available for 135 cases; data were missing for the remaining patients.
Compared to other groups, patients with cirrhosis and HCC were significantly older; their mean viral load and transaminase levels were also significantly elevated.
1. IL10 -592 A > C, and -1082 A > G promoter polymorphism analysis
Genotyping was successfully performed for both SNPs in most of the study population. Amplification failed for the -1082 A > G SNP in 8 cases and for the -592 A > C SNP in 7 cases, resulting in successful genotyping for 322 cases for -1082 A > G SNP and 323 cases for -592 A > C. Regarding the -592 A > C SNP, allele A was observed in 35 % of cases (n = 227), while allele C was present in 65 % (n = 419). The genotype distribution was 14 % (n = 44) for AA, 43 % (n = 139) for AC, and 43 % (n = 140) for CC. Allele distribution for the -1082 A > G SNP revealed frequencies of 64 % for allele A (n = 410) and 36 % for allele G (n = 234). The genotypic distribution was as follows: AA in 41 % (n = 133), AG in 45 % (n = 144), and GG in 14 % (n = 45) of cases. Comparison of allelic and genotypic frequencies between CHBV-C and controls indicated that the GG genotype of the -1082 A > G polymorphism was associated with an approximately twofold increased risk of hepatitis B chronicity (OR=1.9; 95 % CI: 1.00-3.62) (Table 2). No significant association was observed for the other genotypes or for the -592 A > C SNP. Also, statistical analysis did not reveal any significant association between genotypic profiles in both SNPs and patients' clinical status (Table 3).
## 2. Association of IL10
-592 A > C, and -1082 A > G promoter polymorphism, IL10 levels with HBV infection IL10 titration was performed for 37 untreated patients and 74 controls; individuals were matched for age and gender. Overall, the mean IL10 level in the study population was 2.81 pg/ml; it was significantly higher in the HBV group compared to controls (3.27 vs 2.56 pg/ml, p = 0.002). For the -592 A > C SNP, the AA genotype was significantly associated with reduced IL10 levels (1.35 vs. 3.07 pg/ml; p = 0.014) (Fig. 1). However, higher levels of IL10 were linked to the CC genotype (p = 0.039). Regarding the -1082 A > G SNP, IL10 levels were three times higher in patients with the GG genotype (p = 0.04). For the other two genotypes, no significant variations were found.
IL10 levels unit is pg/ml;(-): negative for the corresponding genotype; (+): positive for the corresponding genotype
The comparison of IL10 levels between the two study groups showed that the elevated IL10 levels previously associated with the -1082 GG genotype were significantly higher in patients with CBI than in controls (9.45 pg/mL vs. 2.97 pg/mL; p = 0.005) (Table 4).
## 4. Discussion
The involvement of cytokines in various infectious or inflammatory diseases has been well documented in several studies (Saraiva et al., 2020). Their role in the pathogenesis of viral infections, particularly HBV infection, is complex. IL10 is a major anti-inflammatory cytokine produced by monocytes/macrophages, T cells, B cells, and dendritic cells (Saraiva et al., 2020). Its anti-inflammatory function primarily involves targeting antigen-presenting cells (APCs) by inhibiting the cytokine production and the expression of MHC class II molecules (Saraiva et al., 2020). This suppression of the adaptive immune response may compromise effective viral clearance, thereby contributing to HBV persistence. Elevated IL10 levels were substantially linked to CBI in our study, which may indicate a role in the development of chronicity. This finding is consistent with other studies that found that CHBV-C had higher levels of IL10 than healthy controls ( Özgüler et al., 2015). Some research, however, has found that patients with acute HBV infection had much higher levels of IL10 than those with CHBV-C. This suggests that IL10 plays a part in promoting viral removal and preventing chronicity (Ribeiro et al., 2022). The most significant factor in cytokine regulation has been shown to be host genetic variables, such as promoter polymorphisms. The three main SNPs among the IL10 promoter polymorphisms that have been thoroughly investigated for their impact on cytokine production and gene expression are SNPs -592A>C, -819C>T, and -1082A>G (Rybicka et al., 2020). We focused on both SNPs -592A>C and -1082A>G due to contradictory results in earlier research. The only genotype that was significantly associated with persistent infection was -1082 GG; it conferred a risk that was almost double that of healthy controls. Similar results were reported in an IL10: interleukin 10 gene; OR: odds ratio; CI: confidence interval. Iranian population, where the -1082 GG genotype was significantly associated with chronic HBV infection, while no association was found with the -592 A > C SNP (Moudi et al., 2016). Another study observed a significant association between the -1082 G allele and HBV infection outcomes, which was not replicated in our population. In contrast, Gao et al. suggested that the -1082 G allele may serve as a protective factor against HBV infection, although it might not influence recovery after infection (Gao et al., 2016). A Turkish study did not find any association between -592A>C and -1082A>G IL10 polymorphisms and HBV infection (Temel et al.,. 2023). These contradictory results may be attributed to differences in sample sizes, control group selection, population genetic backgrounds, or the influence of other immune-related genes not examined in this study. Other potentially involved SNPs have been explored. A Chinese study including 996 CHBV-C found that the +504T allele was associated with increased susceptibility to chronic HBV infection (Zhang et al., 2014). Possible associations between IL10 SNP alleles, genotypes, and haplotypes with therapeutic response and liver damage have also been investigated in literature. Indeed, -819T, -592A, and +504T alleles were associated with treatment-induced HBsAg seroclearance (Rybicka et al., 2020). IL10 GCCT haplotype (-1082G/-819C/-592C/-1353T) was associated with an increased risk of cirrhosis; -819TT, +954TT, -592AA, and +504TT genotypes were predictive of lower fibrosis scores (Rybicka et al., 2020). A recent systematic review showed that IL10 and IL18 polymorphisms had the most evidence of correlation with HBV associated liver cirrhosis (Heiat et al., 2025). Although our study did not examine the relationship between IL10 SNPs and liver injury or treatment response, these factors warrant further investigation. Given the role of IL10 in immune modulation, targeting its production holds potential as a therapeutic strategy in infectious and inflammatory diseases. However, its clinical application remains controversial due to the cytokine's variable effects and the potential for adverse effects (Saraiva et al., 2020).
The main limitations of our study include the small sample size for IL10 titration and the lack of a control group consisting of untreated HBV-infected patients in the HBsAg-negative phase. Still, to our knowledge, this is the first study to investigate the relationship between IL10 polymorphisms and the chronic progression of HBV infection in a Tunisian population.
In conclusion, this study highlights the potential role of IL10 in the persistence of HBV, with the -1082 GG genotype being significantly associated with chronic progression of this infection in our Tunisian cohort. These results highlight the importance of host genetic factors in HBV pathogenesis and may contribute to the identification of individuals at higher risk for chronic infection. Further studies with larger sample sizes and extended clinical data, including treatment response and liver damage evaluation, are warranted to confirm these associations and explore their potential therapeutic implications.
## References
1. Amirpour-Rostami, Kazemi Arababadi (2019) "IL-18 and IL-1β gene polymorphisms: the plausible risk factors for chronic hepatitis B" *Viral. Immunol*
2. Gao, Chen, Zhang et al. (2016) "Association of IL-10 polymorphisms with hepatitis B virus infection and outcome in Han population" *Eur. J. Med. Res*
3. Heiat, Javanbakht, Jafari et al. (2025) "Correlation of IL-10 and IL18 with the development of liver cirrhosis associated with hepatitis B virus infection: a systematic review" *Cytokine*
4. Moudi, Heidari, Mahmoudzadeh-Sagheb et al. (2016) "Association between IL-10 gene promoter polymorphisms (-592 A/C, -819 T/C, -1082 A/G) and susceptibility to HBV infection in an iranian population" *Hepat. Mon*
5. Özgüler, Akbulut, Akbulut (2015) "Evaluation of Interleukin-10 levels in patients diagnosed with chronic hepatitis" *West Indian Med. J*
6. Ribeiro, De, Beghini et al. (2022) "Cytokines profile in patients with acute and chronic hepatitis B infection" *Microbiol. Immunol*
7. Rybicka, Woziwodzka, Sznarkowska et al. (2020) "Genetic variation in IL-10 influences the progression of hepatitis B infection" *Int. J. Infect. Dis*
8. Saraiva, Vieira, O'garra (2020) "Biology and therapeutic potential of interleukin-10" *J. Exp. Med*
9. Shi, Zheng (2020) "Hepatitis B virus persistence and reactivation" *BMJ*
10. Takahashi (2014) "Interleukin 28B genetic polymorphism and hepatitis B virus infection" *World J. Gastroenterol*
11. Temel, Akcam, Caner et al. (2023) "Relationship between IL-17, TNF-α, IL-10, IFN-γ, and IL-18 polymorphisms with the outcome of hepatitis B virus infection in the Turkish population" *Rev. Assoc. Med. Bras*
12. (2024) "Global hepatitis report 2024: action for access in lowand middle-income countries"
13. Zhang, Zhang, Zhao et al. (2014) "Gene variation in IL10 and susceptibility to chronic hepatitis B" *J. Infect* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12435045&blobtype=pdf | # Retraction notice to "Advancements in SARS-CoV-2 detection: Navigating the molecular landscape and diagnostic technologies" [Heliyon 10 (2024) e29909]
Nuha Almulla, Raya Soltane, Ahlam Alasiri, Abdou Kamal Allayeh, Taha Alqadi, Fatma Alshehri, Ahlam Alrokban, Sameh Zaghlool, Abdallah Zayan, Karam Abdalla, Ahmed Sayed
## Abstract
Post-publication, an investigation conducted by Elsevier's Research Integrity & Publishing Ethics team on behalf of the journal identified references that are irrelevant to the article. The authors were asked to comment upon the presence of these references in their work but were unable to satisfactorily address the reason for the references.Moreover, the same investigation found phrases that make some passages in the article difficult to parse. The authors were requested to explain the use of these passages of text but were unable to do so.Consequently, the editor no longer has confidence in the integrity and the findings of the article and has decided to retract it. The scientific community takes a very strong view on this matter and apologies are offered to readers of the journal that this was not detected during the submission process.The authors disagree with retraction and dispute the grounds for it. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12775147&blobtype=pdf | Lisa Övermöhle, Maximilian Baum, Katharina Scherer
## Visualising viral interactions and mechanisms at the nanoscale with expansion microscopy
Check for updates Lisa Övermöhle 1,3 , Maximilian Baum 1,3 , Rohan Bhatia 2 , Karin Byskata 1 & Katharina M. Scherer 1 Expansion microscopy is a groundbreaking technique that enables nanoscale imaging of biological specimens using standard optical microscopes. By embedding specimens in swellable hydrogels, it achieves sub-diffraction resolution. Compatible with various tissue types, it offers 3D, multi-colour visualisation of cellular and sub-cellular structures. While challenges remain, like sample isotropy and preservation of molecular integrity, expansion microscopy is a transformative tool for cell and neurobiology. Here, we discuss its potential for virology.
## Principles of expansion microscopy
The standard expansion microscopy protocol involves several key steps (Fig. 1): (1) fixation of the specimen, (2) labelling of biomolecules of interest, (3) anchoring chemical linkers to biomolecules, (4) embedding in a hydrogel, (5) homogenisation through enzymatic digestion or denaturation of the tissue, and (6) subsequent expansion by immersion in water or an expansion buffer. These steps result in the isotropic enlargement of the sample, preserving the native tissue architecture while enabling visualisation at a significantly higher resolution -typically on the order of 70 nm and less. The exact resolution attained depends on the specific expansion protocol, as different methods yield varying expansion factors (commonly 4x, but protocols achieving 10x and more have been developed) and may also involve iterative rounds of expansion.
Several practical considerations must be made regarding each individual step. Based on the specific structure of interest, the most suitable reagent should be chosen for fixation. These are usually the same as commonly used for immunocytochemistry, such as methanol and aldehydes (e.g., paraformaldehyde, glutaraldehyde, glyoxal). Cryo-fixation provides the most accurate preservation of biological samples in their native state 1 ; however, it requires more time and skilled handling than conventional chemical fixation methods.
The selection of fluorescent labels is crucial: some free-radical-induced dye degradation is inevitable during the polymerisation step of forming the hydrogel backbone, but some dyes are more photo-and chemically stable than others. For example, rhodamine-based dyes (e.g., Atto647N) show much better stability than cyanine dyes (e.g., Alexa Fluor 647) 2 which should best be avoided for expansion microscopy. Autofluorescent and selflabelling proteins can stay intact if digestion is not too prolonged, so the digestion time must be adjusted for each sample type like single cells or tissue sections [3][4][5][6] . Although enzymatic homogenisation generally improves expansion, excessive digestion can degrade these proteins, especially autofluorescent proteins and SNAP-tags. There are further strategies to preserve and enhance fluorescence signals in expanded samples. Label retention might be improved by using small, inert trifunctional probes that bind the target, anchor to the hydrogel and carry a fluorophore 5 . Signal amplification can be achieved using fluorescent nanobodies or tyramide signal amplification 7 to boost fluorescence intensity. Post-expansion labelling can improve labelling efficiency due to better epitope accessibility 8 , often using treatment with sodium dodecyl sulfate (SDS) for long durations and/or high temperature treatment 9,10 instead of enzymatic digestion.
The anchoring step requires thorough incubation of the sample with the linker solution to ensure uniform distribution within the sample. Slow polymerisation at lower temperatures during embedding promotes uniform reactions and homogeneity. For expansion microscopy, labelling and anchoring strategies must be compatible with the functional groups present on the target molecules. Proteins are typically anchored to the hydrogel matrix using N-hydroxysuccinimidyl (NHS) ester-functionalised linkers like 6-((acryloyl)amino)hexanoic acid (acryloyl-X) 4 or methacrylic acid 3 . The NHS ester moiety reacts with the primary amines in the lysine side chains whereas the polymerizable handle of the linker covalently binds to the hydrogel during polymerisation. Protein lLabels -such as antibodies, streptavidin, or genetically encoded tags like self-labelling or fluorescent proteins -are covalently anchored to the matrix along with their targets. Non-specific lysine staining with NHS-activated fluorescent dyes enables visualisation of protein densities (ultrastructure) similar to electron microscopy 11 . RNA and DNA can be visualised using fluorescence in situ hybridisation (FISH), which employs short fluorescent oligonucleotides that hybridise to specific RNA or DNA sequences. The guanine-reactive Label-IT reagent can anchor nucleic acid targets as well as their probes and through functionalisation with a polymerizable handle covalently attach them to the hydrogel 12 . DNA can be labelled non-specifically with highaffinity intercalating dyes (e.g., DAPI, SYTO dyes) for high-resolution imaging of chromatin and sub-nuclear structures. Lipid labelling and anchoring is more challenging due to membrane disruption by detergents which are needed to permeabilise samples for monomer infiltration. However, it can be achieved using mild detergents combined with amine- 13,14 , azide- 15,16 or alkyne- 17 (for click-chemistry) as well as biotin 16 -functionalised lipid probes that allow coupling to the gel matrix. Universal linkers such as methacrolein 9 or glycidyl methacrylate 18 are designed to work with a broad range of targets (proteins, RNA, lipids) with minimal or no need for target-specific chemistry. More specialised multifunctional linkers integrate targeting, anchoring, and detection in one molecule to improve multiplexing capacities 14,19 . Another strategy uses various clickable probes in combination with biotin click-labelling and fluorescent streptavidin staining to target a range of biomolecules (lipids, glycans, proteins, DNA, RNA) using conventional protein retention hydrogels 17 .
Assuming sufficient labelling density, the resolution improvement in expansion microscopy depends mainly on the expansion factor, which is defined by the gel recipe and describes the size ratio between the expanded and original sample. Since its original description 20 , many variations of expansion microscopy protocols have been developed to further increase the expansion factor and hence the effective image resolution 21 . Two major strategies have been developed to extend the original protocol. The first approach focuses on modifying the polymer network to enable greater expansion. By reducing cross-linker concentration in bis-acrylamide-crosslinked hydrogels typically used in the standard 4-5x expansion protocols, the expansion factor can be increased to around 10x 9,13 . Achieving such high expansion factors necessitates enhancing structural integrity. Therefore, mechanically robust and elastic hydrogels composed of N,N-dimethylacrylamide and sodium acrylate have been developed. By removing oxygen from the gelation solution, performing gelation in an oxygen-depleted environment, and optimising both gelation time and initiator concentration, expansion factors between 10x and 20x were achieved with this monomer chemistry [22][23][24][25] . The second approach increases the expansion factor by using multiple rounds of expansion, a method called iterative expansion microscopy 26 . The first gel(s) are coupled to a final gel, which is expanded after disrupting the first gel(s). Several iterative approaches have demonstrated electron microscopy-like resolution using measured expansion factors ranging from 15 to 20x 11,26,27 .
While expansion microscopy can achieve substantial magnification, practical limitations might arise due to sample distortion. Local inhomogeneities and deformations in the polyacrylate or other hydrogel matrix might lead to non-isotropic expansion, especially beyond expansion factors of 4-5x.
Heat treatment 24 or alternative enzymes for digestion 7 might improve homogenisation. Isotropic expansion can be verified using biological rulers such as microtubules, nuclear pore complexes (NPCs) and centrioles, which are also commonly measured to determine the expansion factor. Isotropy of expansion might e.g. be measured as the ratio between centriole length and diameter 28 or as a localisation error from comparing the same structure before and after expansion, most often microtubules. By adapting open-source software 29 , post-expansion images can first be aligned to pre-expansion ones using translation, rotation, and scaling corrections, followed by non-rigid registration to account for deformations like stretching and bending. The deformation field generated can then be used to calculate localisation errors 3,4,30 . Artificial molecular rulers such as a gel-embedded reference grid have also been developed for quality control and to intrinsically calibrate the polymer matrix 31 . This approach allows distortions to be corrected without pre-expansion reference images through software tools using landmarkbased deformable image alignment like BigWarp 32 and simultaneously embeds a coordinate system that facilitates sample navigation.
Swollen gels can be stored short-term (up to about a week) in pure water at 4 °C in the dark, which maintains full expansion but leaves them prone to drying and degradation. For longer-term storage, gels are usually shrunk in buffers with sodium azide at 4 °C; HEPES buffer is preferred to preserve fluorescence. Before imaging, gels are re-expanded in water. Less commonly, gels can be stored in protective agents and frozen (e.g., 50% glycerol for a sample of algae 33 ), but freeze-thaw cycles and residual glycerol may hinder re-expansion and increase background.
A wide range of expansion microscopy protocols exists, differing in their polymer chemistry, anchoring strategies, digestion conditions, and achievable expansion factors. This diversity can make the methodological landscape difficult to navigate, especially for new users. For this reason, it is useful to begin with the original, basic expansion microscopy workflow 20,34,35 to become familiar with the fundamental steps -anchoring, gelation, digestion, and expansion -before moving on to specialised variants. To help readers orient themselves and choose the most appropriate protocol for their application, we recommend consulting recent comprehensive reviews on expansion microscopy 21,36 . These reviews provide clear overviews of available variants (e.g., Magnify 9 , X10 22 , TREx 13 , BOOST 37 ), discuss their strengths and limitations, summarise advances and offer a solid foundation for selecting and optimising an expansion microscopy protocol for specific experimental needs.
## Imaging strategies for expanded samples
A central parameter in expansion microscopy is the expansion factor. In first approximation, the optical resolution improves linearly with the inverse of the expansion factor. With standard fourfold expansion microscopy protocols in combination with conventional optical microscopes such as a confocal microscope (diffraction limit ~300 nm), a resolution improvement of down to ~75 nm is achievable. More advanced protocols with expansion factors of 10-20 should attain a resolution improvement in the range of 15-30 nm in theory, achieving resolution capabilities closer to those offered by electron microscopy (Fig. 2).
While this theoretical calculation provides a simple and intuitive relationship, in practice, several factors can significantly affect the actual resolution using expansion microscopy. To achieve optimal imaging performance and reach the theoretical optical resolution limit when imaging expanded samples, several factors must be carefully considered.
First, the theoretical gain in resolution assumes that the fluorescent label is small relative to the structure of interest. However, commonly used labels (e.g., primary and secondary antibodies) are ~10-20 nm in size, which becomes a significant source of localisation uncertainty when attempting to resolve structures at tens of nanometres. Here, post-expansion instead of pre-expansion labelling leads to an improvement in accuracy. For postexpansion labelling, the linkage error (i.e. the physical separation between the target and the fluorophore) is reduced proportional to the expansion factor since the affinity reagent (e.g., antibodies) does not undergo expansion so that the fluorophore is closer to the target compared to preexpansion labelling 8,9 .
Furthermore, sufficient labelling density is required to satisfy the Nyquist sampling criterion (i.e. the minimum sampling that preserves all available resolution). Sparse or uneven labelling may result in incomplete or misleading reconstructions, regardless of the expansion factor. Postlabelling can also help in this case as it improves labelling efficiency.
Optimal resolution and image quality in light microscopy also depend on minimising refractive index mismatch between the sample and the immersion medium. Expansion microscopy reduces local refractive index heterogeneity and light scattering through its inherent sample clearing 38 and decrowding 39 properties. This also improves sample transparency and hence imaging depth, especially important for big samples. Due to the water-based environment of expanded samples, it is usually recommended to use waterimmersion objectives. High numerical aperture (NA > 1.0) objectives increase the resolution but also improve fluorescence light gathering due to their large collection angles. Efficient light detection is essential, as the expansion process lowers fluorophore density by decrowding the samples molecules, thereby decreasing the overall brightness. A beneficial side effect though can be an improvement of the signal-to-noise ratio due to reduced crowding and background fluorescence 39 . The limited working distance of typical high-NA objectives however constrains the maximum imaging depth -an important consideration for expanded samples -and can introduce mechanical interference or compromise focus precision.
Deciding which microscope is most appropriate for different samples involves not only assessing resolution and imaging depth, but also imaging speed, optical sectioning capability, sample compatibility, user-friendliness, photobleaching and field of view (Fig. 3a). Standard microscopy techniques, such as widefield and confocal microscopy, are commonly used for imaging expanded samples due to their accessibility (Fig. 3b andc). Typical inverted configurations however limit imaging depth, restricting observations mainly to expanded single cell and tissue layers.
Widefield microscopy offers rapid, multi-channel imaging with minimal excitation power requirements. However, its lack of optical sectioning leads to the detection of out-of-focus fluorescence, resulting in low contrast and signal-to-noise ratio. Deconvolution algorithms are often applied post-acquisition to enhance contrast and resolution 40 . These algorithms require user-informed application based on the relevant image acquisition parameters. Deconvolution accuracy depends critically on how well the point spread function (PSF) of the imaging system is known or modelled. If the PSF is mismatched, the algorithm can produce distortions or artifacts. Over-iteration or improper settings can also generate artificial features and mislead quantitative interpretation, while incorrect use may amplify noise and obscure true structures 41 .
Confocal microscopy, on the other hand, achieves optical sectioning through the use of spatial pinholes that significantly reduce out-of-focus light 42 . In conventional point-scanning systems, the inherently low acquisition speed can limit throughput. Slow imaging is particularly problematic for expanded samples, as their increased size together with multi-colour imaging requirements and potential need for volumetric acquisition can substantially prolong total acquisition time. Spinning disk confocal microscopy offers a higher-speed alternative, but it typically exhibits less efficient out-of-focus light rejection, increased background noise, and lower sensitivity. These factors might be limiting when imaging the inherently dim and diluted fluorescent signals characteristic of expanded samples.
In comparison to these techniques, light sheet and oblique plane microscopy offer some distinct advantages (Fig. 3d ande). They combine the speed of widefield microscopy with the optical sectioning capability of confocal microscopes. Optical sections are generated by illuminating the sample with a thin sheet of light. Low photo-bleaching, highly light-efficient detection and fast volumetric imaging make them in principle ideally suited for imaging expanded gels [43][44][45] . Light sheet microscopy has typically been employed for volumetric imaging of large samples like organoids or organisms such as zebrafish 46 . The features of interest in these samples lie in the micrometer range and therefore require only micrometer-level resolution, which conventional light-sheet setups typically provide. However, the axial resolution of light-sheet microscopes can be tuned through sheet thickness and beam profile to reach values as low as ~250 nm 47 . This makes light sheet microscopy a versatile technique capable of spanning imaging scales from the mesoscopic to the nanoscale.
For light sheet microscopy, illumination and detection light path are separated geometrically (Fig. 3d), usually by mounting two objectives perpendicular to each other around the sample in different geometries dependent on the type of specimen 48 . Integration with expansion microscopy has proven particularly powerful for mapping of neuron architecture in organoids 49 as well as in animal brains 43,[50][51][52][53][54] . A challenge of the optical pathway geometry apart from sample mounting is that either stage or objectives must be repositioned during imaging to maintain focus throughout the entire depth of the sample.
Oblique plane microscopy (OPM) is a variant of light-sheet fluorescence microscopy designed to image adherent cells and samples on flat substrates such as coverslips 55,56 . It has been demonstrated suitable for expanded samples by imaging a ~4x expanded brain organoid 57 . The light sheet is delivered at an angle through the same high-NA objective used for detection, creating a tilted illumination plane near the coverslip (Fig. 3e). OPM works with conventionally mounted specimens on inverted microscopes and allows high-speed acquisition without moving parts. Although technically challenging, its adaptability to various sample types and sizes, Theoretical resolution limit calculated for a confocal microscope setup equipped with a 1.2 NA water objective lens, with NA = numerical aperture. The Rayleigh resolution limit is defined as 0.61*λ/NA, with λ = emission wavelength. The resolution is dependent on the emission wavelength and, in first approximation, on the expansion factor. along with advanced variants offering improved resolution, field of view, and modalities, makes it a promising tool to enhance expansion microscopy.
Sample expansion may furthermore be integrated with superresolution microscopes 58,59 to further extend the achievable resolution in biological imaging 60,61 (Fig. 4), though the intrinsic restrictions of each method must be taken into account. Through technically advanced setups, often paired with complex data processing, super-resolution microscopes modulate fluorescence emission in space or time to surpass the diffraction limit of optical microscopes. It is possible to reveal the molecular architecture of sub-cellular ultrastructure with near electron microscopic precision (ca. 10-20 nm) by coupling standard expansion microscopy (4-5x) with e.g. stimulated emission depletion (STED) 28 or direct stochastic optical reconstruction microscopy (dSTORM) 8 (Fig. 4). Recently, it was shown, that protein shapes could be revealed improving the resolution below 10 nm 39 by coupling 10x expansion with super-resolution radial fluctuation (SRFF) analysis of the fluorescence signal 62,63 . Limitations depending on the technique include the need for compatible fluorophores and specialised buffers rather than water -which is required for maximal expansion -as well as motion artifacts and photobleaching caused by long imaging times and the often-high laser powers required. Despite these challenges, the advantages are significant: by combining different methods, it becomes attainable to resolve the structure of viruses whose diameters lie well below the traditional diffraction limit of light microscopes (Fig. 4).
## Single virus expansion microscopy
Directly visualising the structure of viral particles is essential for advancing our understanding of the viral lifecycle. It informs about how viruses attach to and penetrate host cells, copy and process their genomes and assemble proteins into virions. Identifying structural features of viral particles can help to reveal neutralizing epitopes, guide the design of antigen-based tests and structure-based drug design.
Electron microscopy was and still is the benchmark technique for imaging and analysing virus particles. Its principal strength lies in its exceptionally high resolution at near-atomic level, which surpasses that of any other currently available imaging method. The extremely good resolution, however, comes at a cost: it has low contrast, sample preparation is time-consuming and requires a high level of skill, and there is little, or no, molecular identification of specific proteins. Furthermore, labelling and detection of specific proteins inside large assemblies and cells is challenging with electron microscopy. Fluorescence optical microscopy provides several significant advantages for the study of virus structure and morphology, including high-contrast images, multi-colour acquisition and high specificity through the use of specific fluorescent labels for viral components (capsid and envelope proteins as well as nucleic acids).
Virus particles vary in size across different virus species (Fig. 4). The size of most virions and their sub-structures is below the resolution limit of conventional light microscopes. However, super-resolution microscopes with resolution capabilities of up to 20 nm, using techniques such as stimulated emission depletion (STED) and single molecule localisation microscopy (SMLM), can provide nanoscale insights into virus structure 64 . The direct imaging of virions with these advanced light microscope methodologies has expanded our scientific knowledge into various human pathogens [65][66][67][68][69] .
The potential of expansion microscopy for imaging single virions has only just started to be explored. Expansion microscopy has already been employed to examine single virions of herpes simplex virus-1 (HSV-1), Epstein-Barr virus (EBV) and human immunodeficiency virus-1 (HIV). It is possible to visualise the capsid layer around the double-stranded DNA of HSV-1 particles by use of a standard 4x expansion protocol in combination with AiryScan confocal microscopy (an advanced form of confocal microscopy that utilises a specialised detector to improve spatial resolution beyond the diffraction limit) as shown in Fig. 5. Measurement of the tip-totip distance of isotropically expanded, antibody-labelled HSV-1 capsids within infected cells yielded an average capsid size of 125.4 nm 70 , consistent with electron microscopy data 71 . Using 12x expansion microscopy, the viral architecture of EBV could be resolved and the inner and outer diameter of EBV virions was determined (162 and 76 nm respectively) 72 , in line with previous electron microscopy data 73 . Applying the same expansion protocol, sites of HIV-1 genome integration could be visualised by distinguishing monomers and dimers of unspliced viral RNA and co-localisation with the viral Gag structural protein 72 .
These examples show that expansion microscopy is ready to be used for detailed examination of virion layers and architecture. The method has a huge advantage since virion architecture can be readily observed within the context of infected cells which is not easy to achieve with electron microscopy. It might be applied to reveal key information about all stages in the Of course, high enough resolution and localisation accuracy are essential for single virion imaging. For rigid macromolecular assemblies with highly ordered and symmetric architectures -such as viral capsids -it is especially critical that all expansion steps are efficient. This ensures isotropic swelling of the sample and a sufficiently high labelling density to accurately retain structural features. By using an optimised polymer matrix and iterative ~10x expansion, the spherical shape of HSV-1 virions was better preserved and the median spatial error for localising the viral envelope layer was improved from 14.3 to 9.2 nm compared to standard protocols 74 .
Testing the limits of expansion microscopy, the combination of 20x one-step expansion microscopy and computational super-resolution using videos of temporal fluorophore fluctuations has recently proven that resolution of protein shapes is possible 39 . Achieving single-protein resolution with standard confocal microscopy would require an expansion greater than 20-fold, ideally in the range of at least 50-fold. By iterating protocols with high expansion factors, it is feasible that expansion factors of 50 or more can be achieved in the near future.
## Nanoscale insights into viral infection through expansion microscopy
The rapidly advancing field of viral optical imaging -which encompasses investigations into virus-host interactions, viral transmission, replication, assembly, and egress, as well as the characterisation of viral vaccines and antiviral strategies -continues to evolve and expand (see reviews [75][76][77][78] ). Within this field, expansion microscopy is emerging as a powerful tool to study complex viral processes such as viral replication and host immune evasion. Offering greater accessibility than super-resolution microscopes and fewer limitations related to sample preparation and imaging conditions, expansion microscopy holds significant promise for broad adoption in laboratories.
Expansion microscopy enables high-precision sub-cellular imaging, allowing for detailed quantitative analysis of viral processes. It facilitates for example the measurement of viral particle numbers and co-localisation with host proteins. This enables quantitative measurements with substantially higher confidence compared to conventional light microscopy. Due to its compatibility with multi-spectral imaging, expansion microscopy also allows for an increase in the number of distinguishable labels -from the typical three or four to up to eight and more. This could allow a more detailed investigation into complex biomolecular interaction patterns in the future. Furthermore, the spatial distribution of viral proteins within the host cell and its compartments can be mapped in great detail, which aids e.g. in deciphering mechanisms of viral trafficking and genome delivery. Such capabilities are crucial for advancing our understanding of virus-host interactions at the molecular level.
For several of the most virulent and extensively studied human pathogens, including high-priority viruses of global health concern, expansion microscopy has already been successfully applied to elucidate key aspects of their replication and host interactions.
## Hepatitis virus
Expansion microscopy was used to study the exact role of the multifunctional hepatitis C virus (HCV) phosphoprotein NS5A which is essential for HCV RNA replication and virion assembly 79 . The mechanism by which its phosphorylation controls these functions has been poorly understood. The study uncovered that phosphorylation at serine 225 orchestrates the nanoscale structure and distribution of NS5A assemblies. Applying ~4-5x sample expansion ( ~70 nm spatial resolution), puncta size, shape, and cell surface localisation of serine 225-phosphorylated NS5A was visualised and quantified (Fig. 6a). The spatial control of NS5A is essential for forming effective HCV replication complexes and sustaining viral replicationhighlighting the value of expansion microscopy in revealing spatial regulation of viral proteins within infected cells.
## Herpesvirus
Herpesviruses are particularly adept at remodelling their host cell environment to support their replication mechanisms. Studying these processes is fundamental for understanding herpesvirus assembly and egress. During infection with the prototypic herpesvirus herpes simplex virus 1 (HSV-1), especially the nucleus as the site of viral DNA replication, gene expression and capsid assembly undergoes extensive structural reorganisation. Using expansion microscopy, it was precisely mapped how HSV-1 forces chromatin rearrangement to create egress routes for viral capsids 80 . In a stepwise, time-dependent mechanism, chromatin is first compacted before channels are formed. These low-density interchromatin channels emerge at late infection stage within compacted peripheral chromatin, likely providing pathways for capsid movement. HSV-1 not only remodels chromatin architecture, but also organisation of nucleoli 81 . High-resolution 3D mapping by expansion microscopy showed furthermore how chromatin redistributes around displaced nucleoli and that key nucleolar proteins are delocalised during infection before relocating together with marginalised chromatin. These rearrangements coincide with impaired rRNA synthesis and processing, suggesting that viral infection actively hijacks nucleolar structure and function. But structural rearrangements also occur in the cytoplasm during HSV-1 infection. Mitochondria, which normally show a reticular network, were observed via expansion microscopy to undergo progressive condensation and collapse toward the perinuclear region during late infection 82 . This redistribution may support energy demands or help remodel membranes during virus assembly. A recent study uncovered distinct morphologies of HIV-1 capsid assemblies inside nuclei through ~4x isotropic expansion in multiple cell types: HeLa-derived lines, primary CD4⁺ T-cells, and monocyte-derived macrophages (key reservoirs for HIV-1) 83 . These structural features were previously obscured by standard immunofluorescence. The findings suggest a wider diversity in nuclear capsid structures than previously observed. This provides fresh insight into post-entry processes such as uncoating and nuclear import -key to deciphering HIV-1 replication strategy. A new mechanism involving lipid composition in HIV-1 assembly was uncovered by showing that the incorporation of host proteins depends on phosphatidylinositol 4,5bisphosphate (PIP 2 )-dependent co-clustering with HIV-1 Gag at the plasma membrane 84 . The acidic phospholipid PIP 2 is crucial for recruiting the host transmembrane proteins CD43, PSGL-1, and CD44 into budding virions, likely through electrostatic clustering. When PIP 2 is depleted, these proteins are no longer incorporated into HIV particleseven though Gag is still expressed -disrupting virion composition and potentially reducing infectivity. Critically, only through ~4-5x sample expansion could these nanoscale co-clustering events in the range of ~10-100 nm be directly visualised and quantified (Fig. 6b). Expansion microscopy was also employed to develop tools for investigating HIV-1 latency 85 . By engineering a hybrid sensor, the PromA sequence within the U3 region of the HIV-1 long terminal repeat (LTR) was targeted and validated as a specific and conserved biomarker of infected cells. PromA transcripts are short non-coding RNAs and are transcribed even when HIV is transcriptionally silent. By ~4.5x isotropic expansion, 3D spatial mapping of PromA captured the sub-cellular distribution of HIV-1 RNA. This information may aid in identifying cells that harbour latent HIV, which typically evade immune detection and antiretroviral therapy. Latent reservoirs can be distinguished from actively replicating virus based on their distinct localisation patterns -punctate nuclear clusters versus diffuse signals, respectively.
## Influenza virus
Expansion microscopy has enabled high-resolution mapping of host structures involved in influenza virus replication relative to viral components. Detailed imaging using super-resolution microscopes and expansion microscopy revealed thin, F-actin-rich nanotubes (50-200 nm in diameter) connecting influenza A virus (IAV)-infected lung epithelial cells with naïve cells 86 . IAV does not only use theses nanotubes to transfer viral proteins (e.g., NP, HA) but also viral genomes (vRNPs) to recipient cells, resulting in productive viral replication. These observations indicate that influenza viruses use these intercellular networks to bypass immune defences. By applying ∼4x expansion, the fine-scale localisation of viral proteins to cytoskeletal structures was monitored 44 . This provided insights into how viral components may associate with microtubules or other cellular networks -information not resolved by conventional confocal microscopy. To understand how IAV enters the host cytoplasm after endocytosis, expansion microscopy was used to explore virus-endosome interactions with enhanced resolution by mapping the spatial separation between virus and membrane 87 . This quantitative approach demonstrated that in the absence of the host proteins light chain 3 proteins (LC3s) or pericentrin, virus particles remained further from the endosome membrane -signalling a failure to uncoat and enter the cytoplasm. As a response of IAV infection, the formation of PANoptosomes is triggered. PANoptosomes are macromolecular complexes that regulate a unique innate immune inflammatory cell death pathway. By applying expansion microscopy, it was demonstrated how IAV infection promotes the assembly and co-localisation of PANoptosome components, including inflammasome adaptor protein ASC, apoptotic caspase-8 (CASP8), and necroptotic RIPK3 88 . The results support PANoptosome role in inflammatory responses and offering insight into the cellular mechanisms underlying viral-induced immune activation.
## SARS-CoV-2
Replication organelles and viral assembly sites represent dynamic and densely packed hubs of viral and host protein activity during SARS-CoV-2 infection. These compartments support viral RNA synthesis and coordinate the recruitment of structural proteins for virion assembly. However, the close apposition of membranes, proteins, and RNA within these confined regions presents a major challenge for ultrastructural characterisation by electron microscopy alone, particularly when attempting to resolve the spatial organisation of specific viral proteins. Recent work employing expansion microscopy has begun to overcome these limitations. Tenfold expansion microscopy allowed the quantification of spatial relationships among viral structural proteins, organelles and epithelial surface protrusions in an intact three-dimensional context 89 . Imaging of SARS-CoV-2-infected human airway cells revealed Golgi fragmentation, virus-laden multivesicular bodies and apical membrane as well as ciliary remodelling. By physically enlarging fixed SARS-CoV-2-infected cells 4x and applying light-sheet imaging, it was shown that the nucleocapsid protein accumulates in layered, vesicle-like compartments closely associated with replication organelles 90 . Viral RNA replication foci (dsRNA) were found embedded within the outer layers of the N-positive compartments, confirming these as active sites of viral RNA synthesis. Stress granule formation is an effect of SARS-CoV-2 infection. However, the stress granule proteins G3BP1 and G3BP2 are not passive bystanders but are actively recruited into SARS-CoV-2 particles during viral assembly. These host proteins localise within toroidal ERGIC-derived vesicles, which serve as key sites of virion formation. Expansion microscopy provided the spatial resolution necessary to detect and characterise these highly packed, multicomponent organelles 91 (Fig. 6c). Depletion of G3BPs disrupted assembly site formation, underscoring their proviral role and highlighting them as potential antiviral targets.
## Detection of pathogenic features in viral infections with expansion microscopy
Pathology plays a critical role in virology by providing the structural and functional context in which viral infections manifest within host tissues. Through histological, molecular, and imaging techniques, pathology enables the identification of virus-induced cellular damage, inflammation, and tissue remodelling. This helps to link specific viral mechanisms -such as immune evasion, latency, or cytopathic effects -with clinical outcomes. Conversely, virology informs pathology by investigating the life cycles, tropisms, and host interactions of viruses, guiding the interpretation of disease patterns and aiding in the development of diagnostic markers and targeted therapies.
Expansion microscopy can enhance diagnostic precision due to a better resolution of pathological features. Specialised protocols have been developed to accommodate various sample types, including the reuse of formalin-fixed, paraffin-embedded (FFPE) tissue slices commonly archived in clinical settings 92,93 . Also, recent protocols have improved both the speed and expansion factor of the technique, including the development of an optimised, broadly compatible biomolecular anchor 37,94,95 . In addition to the expansion factor itself, effective optical clearing in expanded samples plays a key role in reducing light scattering. Thus, besides the resolution improvement, expansion microscopy can also enhance imaging depth in dense or heterogeneous tissues. A further advantage is the improved labelling efficiency achieved when staining in the expanded state, as epitopes become more accessible. In some applications, resolution may be less critical, allowing samples to be shrunk back afterward for faster imaging -an important consideration for diagnostic workflows. This approach can enhance the visibility of disease or infection markers, supporting more reliable assessments.
A pathological case study demonstrated the value of expansion microscopy by revealing uncontrolled mpox virus (MPXV) infection in the gastrocnemius muscle of an immunocompromised mpox patient 96 . The patient, who had severe HIV and a low CD4⁺ count, showed a lack of effective immune response in the infected tissue. Threefold expansion of the formalin-fixed tissue enabled high-resolution immunofluorescence imaging, even within the densely packed, necrotic regions. This approach revealed viral structures and infected cells that would have been difficult to detect using conventional microscopy, allowing precise localisation of MPXV antigens and exposing the lack of effective immune cell engagement with infected cells.
## Outlook
Expansion microscopy has already demonstrated significant promise for studying virus biology as well as virion morphology by enabling highresolution imaging using standard diffraction-limited microscopes. It offers a more accessible and user-friendly alternative to optical super-resolution microscopes, without compromising spatial detail. Its low equipment demands and compatibility with common optical setups like confocal microscopes, make it straightforward to implement. In most current studies, the straightforward fourfold expansion is employed, which already allows detailed analysis of viral structures in complex tissue environments. However, newly developed protocols that further increase the expansion factor and shorten preparation time -along with more robust and universally compatible biomolecular anchors -could unlock the full potential of expansion microscopy for virology. Beyond the increase in resolution, additional advantages of expansion protocols -such as tissue clearing and enhanced labelling efficiency -can be especially valuable for analysing pathological samples. If these advances prove consistent and reliable in sample handling and dimensional control, they could provide unprecedented access to nanoscale features of virus biology, including capsid structure, host-virus interactions, and viral replication sites, across a broad range of sample types.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/bync-nd/4.0/.
## References
1. Laporte, Klena, Hamel et al. (2022) "Visualizing the native cellular organization by coupling cryofixation with expansion microscopy (Cryo-ExM)" *Nat. Methods*
2. Wen, Leen, Jia et al. (2022) "Improved dye survival in expansion microscopy through stabilizer-conjugated linkers" *Chem. A Eur. J*
3. Chozinski (2016) "Expansion microscopy with conventional antibodies and fluorescent proteins" *Nat. Methods*
4. Tillberg (2016) "Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies" *Nat. Biotechnol*
5. Shi (2021) "Label-retention expansion microscopy" *J. Cell Biol*
6. Campbell, Pannoni, Savory et al. (2021) "Protein-retention expansion microscopy for visualizing subcellular organelles in fixed brain tissue" *J. Neurosci. Methods*
7. Wang (2023) "Protein and lipid expansion microscopy with trypsin and tyramide signal amplification for 3D imaging" *Sci. Rep*
8. Zwettler (2020) "Molecular resolution imaging by post-labeling expansion single-molecule localization microscopy (Ex-SMLM)" *Nat. Commun*
9. Klimas (2023) "Magnify is a universal molecular anchoring strategy for expansion microscopy" *Nat. Biotechnol*
10. Valdes (2024) "Improved immunostaining of nanostructures and cells in human brain specimens through expansion-mediated protein decrowding" *Sci. Transl. Med*
11. Saad, Bewersdorf (2020) "Light microscopy of proteins in their ultrastructural context" *Nat. Commun*
12. Chen (2016) "Nanoscale imaging of RNA with expansion microscopy" *Nat. Methods*
13. Damstra (2022) "Visualizing cellular and tissue ultrastructure using Ten-fold Robust Expansion Microscopy (TREx). eLife 11"
14. Wen (2020) "Evaluation of Direct Grafting Strategies via Trivalent Anchoring for Enabling Lipid Membrane and Cytoskeleton Staining in Expansion Microscopy" *ACS Nano*
15. Götz (2020) "Nanoscale imaging of bacterial infections by sphingolipid expansion microscopy" *Nat. Commun*
16. Shin (2025) "Dense, continuous membrane labeling and expansion microscopy visualization of ultrastructure in tissues" *Nat. Commun*
17. Sun (2021) "Click-ExM enables expansion microscopy for all biomolecules" *Nat. Methods*
18. Cui (2023) "Expansion microscopy using a single anchor molecule for high-yield multiplexed imaging of proteins and RNAs" *PLoS ONE*
19. Wen (2021) "A Universal Labeling Strategy for Nucleic Acids in Expansion Microscopy" *J. Am. Chem. Soc*
20. Chen, Tillberg, Boyden (2015) "Expansion microscopy" *Science*
21. Hümpfer, Thielhorn, Ewers (2024) "Expanding boundariesa cell biologist's guide to expansion microscopy" *J. Cell Sci*
22. Truckenbrodt (2018) "X10 expansion microscopy enables 25-nm resolution on conventional microscopes" *EMBO Rep*
23. Truckenbrodt, Sommer, Rizzoli et al. (2019) "A practical guide to optimization in X10 expansion microscopy" *Nat. Protoc*
24. Saal (2023) "Heat denaturation enables multicolor X10-STED microscopy" *Sci. Rep*
25. Wang (2024) "Single-shot 20-fold expansion microscopy" *Nat. Methods*
26. Chang (2017) "Iterative expansion microscopy" *Nat. Methods*
27. Louvel (2023) "iU-ExM: nanoscopy of organelles and tissues with iterative ultrastructure expansion microscopy" *Nat. Commun*
28. Gambarotto (2019) "Imaging cellular ultrastructures using expansion microscopy (U-ExM)" *Nat. Methods*
29. Klein, Staring, Murphy et al. (2010) "A Toolbox for Intensity-Based Medical Image Registration" *IEEE Trans. Med. Imaging*
30. Gao (2018) "Expansion Stimulated Emission Depletion Microscopy (ExSTED)" *ACS Nano*
31. Damstra (2023) "GelMap: intrinsic calibration and deformation mapping for expansion microscopy" *Nat. Methods*
32. Bogovic, Hanslovsky, Wong et al. (2016) "Robust registration of calcium images by learned contrast synthesis"
33. Klena (2023) "An In-depth Guide to the Ultrastructural Expansion Microscopy (U-ExM) of Chlamydomonas reinhardtii" *BIO-PROTOCOL*
34. Gao, Asano, Boyden et al. (2017) "Expansion microscopy"
35. Wassie, Zhao, Boyden (2019) "Expansion microscopy: principles and uses in biological research" *Nat. Methods*
36. Jia (2025) "Derivative Technologies of Expansion Microscopy and Applications in Biomedicine" *ChemBioChem*
37. Guo (2025) "BOOST: a robust ten-fold expansion method on hourscale" *Nat Commun*
38. Chuang (2024) "Super-Resolution Imaging in Collagen-Abundant Thick Tissues" *Small Struct*
39. Shaib (2024) "One-step nanoscale expansion microscopy reveals individual protein shapes" *Nat Biotechnol*
40. "Quantitative Fluorescence Microscopy and Image Deconvolution"
41. Wallace, Schaefer, Swedlow (2001) "A Workingperson's Guide to Deconvolution in Light Microscopy" *BioTechniques*
42. Elliott (2020) "Confocal Microscopy: Principles and Modern Practices" *Curr. Protoc. Cytom*
43. Bürgers (2019) "Light-sheet fluorescence expansion microscopy: fast mapping of neural circuits at super resolution" *Neurophoton*
44. Mascheroni (2020) "Combining sample expansion and light sheet microscopy for the volumetric imaging of virus-infected cells with super-resolution" *Biomed. Opt. Express*
45. Pesce, Ricci, Sportelli et al. (2024) "Expansion and Light-Sheet Microscopy for Nanoscale 3D Imaging" *Small Methods*
46. Keller, Schmidt, Wittbrodt et al. (2008) "Reconstruction of Zebrafish Early Embryonic Development by Scanned Light Sheet Microscopy" *Science*
47. Chen (2014) "Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution" *Science*
48. Stelzer (2021) "Light sheet fluorescence microscopy" *Nat Rev Methods Primers*
49. Rodriguez-Gatica (2022) "Imaging three-dimensional brain organoid architecture from meso-to nanoscale across development" *Development*
50. Düring, Rocha, Dittrich et al. (2019) "Expansion Light Sheet Microscopy Resolves Subcellular Structures in Large Portions of the Songbird Brain" *Front. Neuroanat*
51. Tian (2024) "Rapid lightsheet fluorescence imaging of whole Drosophila brains at nanoscale resolution by potassium acrylatebased expansion microscopy" *Nat. Commun*
52. Scardigli (2021) "Comparison of Different Tissue Clearing Methods for Three-Dimensional Reconstruction of Human Brain Cellular Anatomy Using Advanced Imaging Techniques" *Front. Neuroanat*
53. Gao (2019) "Cortical column and whole-brain imaging with molecular contrast and nanoscale resolution" *Science*
54. Lillvis (2022) "Rapid reconstruction of neural circuits using tissue expansion and light sheet microscopy"
55. Dunsby (2008) "Optically sectioned imaging by oblique plane microscopy" *Opt. Express*
56. Kumar (2011) "High-speed 2D and 3D fluorescence microscopy of cardiac myocytes" *Opt. Express*
57. Lamb, Cardoso Mestre, Lancaster et al. (2025) "Directview oblique plane microscopy" *Optica*
58. Jacquemet, Carisey, Hamidi et al. (2020) "The cell biologist's guide to super-resolution microscopy" *J. Cell Sci*
59. Schermelleh (2019) "Super-resolution microscopy demystified" *Nat. Cell Biol*
60. Aristova (2025) "Nanoscale imaging of biological systems via expansion and super-resolution microscopy" *Appl. Phys. Rev*
61. Woo, Brown (2025) "Review of expansion microscopy combined with advanced imaging modalities" *Journal of Microscopy jmi*
62. Gustafsson (2016) "Fast live-cell conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations" *Nat. Commun*
63. Laine (2019) "NanoJ: a high-performance open-source superresolution microscopy toolbox" *J. Phys. D: Appl. Phys*
64. Robb (2022) "Virus morphology: Insights from super-resolution fluorescence microscopy" *Biochimica et. Biophysica Acta (BBA) -Mol. Basis Dis*
65. Chojnacki (2012) "Maturation-Dependent HIV-1 Surface Protein Redistribution Revealed by Fluorescence Nanoscopy" *Science*
66. Laine (2015) "Structural analysis of herpes simplex virus by optical super-resolution imaging" *Nat. Commun*
67. Mehedi (2017) "Multicolor Stimulated Emission Depletion (STED) Microscopy to Generate High-resolution Images of Respiratory Syncytial Virus Particles and Infected Cells" *BIO-PROTOCOL*
68. Liu, Chen, Aguilar et al. (2018) "A stochastic assembly model for Nipah virus revealed by super-resolution microscopy" *Nat. Commun*
69. Kummer, Avinoam, Kräusslich (2019) "IFITM3 Clusters on Virus Containing Endosomes and Lysosomes Early in the Influenza A Infection of Human Airway Epithelial Cells" *Viruses*
70. Mäntylä (2023) "Iterative immunostaining combined with expansion microscopy and image processing reveals nanoscopic network organization of nuclear lamina" *MBoC*
71. Trus (1996) "The herpes simplex virus procapsid: structure, conformational changes upon maturation, and roles of the triplex proteins VP19c and VP23 in assembly" *J. Mol. Biol*
72. Norman (2025) "One step 4× and 12× 3D-ExM enables robust super-resolution microscopy of nanoscale cellular structures" *J. Cell Biol*
73. Nanbo, Noda, Ohba et al. (2018) *Virus Acquires Its Final Envelope on Intracellular Compartments With Golgi Markers. Front. Microbiol*
74. Gao (2021) "A highly homogeneous polymer composed of tetrahedron-like monomers for high-isotropy expansion microscopy" *Nat. Nanotechnol*
75. Grove (2014) "Super-Resolution Microscopy: A Virus" *Eye View of the Cell. Viruses*
76. Witte, Andriasyan, Georgi et al. (2018) "Concepts in Light Microscopy of Viruses"
77. Castelletto, Boretti (2021) "Viral particle imaging by super-resolution fluorescence microscopy" *Chem. Phys. Impact*
78. Arista-Romero, Pujals, Albertazzi (2021) "Towards a Quantitative Single Particle Characterization by Super Resolution Microscopy: From Virus Structures to Antivirals Design" *Front. Bioeng. Biotechnol*
79. Goonawardane, Yin, Roberts et al. (2025) "A key role for hepatitis C virus NS5A serine 225 phosphorylation revealed by super-resolution microscopy" *Sci Rep*
80. Aho (2019) "Quantitative Microscopy Reveals Stepwise Alteration of Chromatin Structure during Herpesvirus Infection" *Viruses*
81. Leclerc (2024) "Herpesvirus-Induced Manipulation of the Nucleolus" *Microscopy and Microanalysis*
82. Scherer (2021) "A fluorescent reporter system enables spatiotemporal analysis of host cell modification during herpes simplex virus-1 replication" *J. Biol. Chem*
83. Petrich (2024) "Expanding Insights: Harnessing Expansion Microscopy for Super-Resolution Analysis of HIV-1-Cell Interactions" *Viruses*
84. Souza Cardoso, Murakami, Jacobovitz et al. (2025) "PIP 2 promotes the incorporation of CD43, PSGL-1, and CD44 into nascent HIV-1 particles" *Sci. Adv*
85. Amodio (2022) "Nanoscale probing and imaging of HIV-1 RNA in cells with a chimeric LNA-DNA sensor" *Nanoscale*
86. Kumar (2017) "Influenza virus exploits tunneling nanotubes for cellto-cell spread" *Sci Rep*
87. Cong (2025) "Influenza A virus subverts the LC3-pericentrin dynein adaptor complex for host cytoplasm entry" *Sci. Adv*
88. Wang (2022) "Single cell analysis of PANoptosome cell death complexes through an expansion microscopy method" *Cell. Mol. Life Sci*
89. Nijenhuis (2021) "Optical nanoscopy reveals SARS-CoV-2-induced remodeling of human airway cells"
90. Scherer (2022) "SARS-CoV-2 nucleocapsid protein adheres to replication organelles before viral assembly at the Golgi/ERGIC and lysosome-mediated egress" *Sci. Adv*
91. Murigneux (2024) "Proteomic analysis of SARS-CoV-2 particles unveils a key role of G3BP proteins in viral assembly" *Nat Commun*
92. Zhao (2017) "Nanoscale imaging of clinical specimens using pathology-optimized expansion microscopy" *Nat. Biotechnol*
93. Bucur (2020) "Nanoscale imaging of clinical specimens using conventional and rapid-expansion pathology" *Nat. Protoc*
94. Park (2019) "Scalable and isotropic expansion of tissues with simply tunable expansion ratio" *Adv. Sci*
95. Cheng (2023) "MicroMagnify: a multiplexed expansion microscopy method for pathogens and infected tissues" *Adv. Sci*
96. Matschke (2024) "Inefficient tissue immune response against MPXV in an immunocompromised mpox patient" *J. Med. Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12691614&blobtype=pdf | # | Bacteriophages | Perspective mGem: Immune recognition and clearance of bacteriophagesimplications for phage therapy
H Le, G Ahlenstiel, C Venturini, S Read
## Abstract
Bacteriophages (phages) hold significant promise as targeted antibacterial therapies in the era of rising multidrug-resistant infections. Despite their therapeutic potential, the clinical application of phages for human infections has been significantly hindered by the rapid and robust immune response to phages in blood. The rapid clearance of >99% of phages from circulation within hours of injection is the result of innate and adaptive immune responses that target therapeutic phage for clearance and destruction. Methodologies must be developed to isolate and/or modify phages that are not only therapeutically potent but also immunologically camouflaged. The resulting second-and third-generation phage therapies will be more effective by evading host immune responses, enabling more efficient targeting of bacterial pathogens.
and in vivo (12,13), indicating that phage antigens can produce potentially harmful responses. A current focus of study, therefore, is the modification of phage particles to dampen antiphage immune responses and prolong the half-life of phages in blood. In doing so, phage-based therapeutics will be not only safer but also more effective by promoting phage-mediated bacterial killing through improved longevity in circulation.
## IMMUNE-MEDIATED PHAGE CLEARANCE
IV phage administration enables systemic delivery and achieves higher bioavailability compared to oral, nasal, or topical routes (14). Once in circulation, however, rapid clearance of phage is coupled with accumulation primarily in the liver (15,16), followed by the spleen (15,17,18), both organs containing abundant phagocytic immune cell populations. In addition to circulating phagocytes such as monocytes and neutrophils (10), splenic macrophages and Kupffer cells (liver macrophages) are primary drivers of phage engulfment (17,19). As one might expect, macrophage depletion can significantly increase phage duration in circulation (20). Intriguingly, however, B-cell depletion also significantly impairs clearance following IV T7 phage administration (21), suggesting that both innate and adaptive immune systems work in concert to clear phages from blood. These data indicate that opsonization (the coating of foreign particles with molecules to increase immune recognition and phagocytosis) is essential to immune-mediated phage clearance. Opsonins include antibodies, complement proteins, and lesser-known pattern recognition molecules such as mannose-binding lectin (MBL) and pentraxins (Fig. 1). Phagocytes, such as neutrophils, macrophages, and dendritic cells, express receptors to recognize opsonins (e.g., antibody Fc receptor CD16 and complement receptors) that promote the capture and internalization of opsonized material (22).
## FIG 1
Innate and adaptive immune recognition of phage particles. Upon IV administration of phages, soluble innate immune proteins, including natural antibodies, mannose-binding lectin (MBL), and acute-phase response proteins such as complement and pentraxins may bind phage particles to opsonize and impair phage infectivity. This has not been demonstrated experimentally for MBL and pentraxins but is a strong possibility due to the heteromeric composition of innate immune proteins and repetitive capsid structure of phages. Combined with the production of adaptive antigen-specific antibodies, phagocytic cell populations in the blood and tissue recognize and engulf opsonized phage via complement and antibody Fc receptors.
## Antibodies
In both mouse and human models, phage administration stimulates antibody produc tion (23,24) that promotes phage clearance (25), neutralizes phage infection in vitro (26,27), and impacts therapeutic efficacy (28). The antigenic targets of these antibodies are diverse, polyclonal, and directed against phage structural proteins, including the major capsid and receptor-binding proteins, as well as decorative proteins like Hoc and Soc (29)(30)(31). In addition to phage therapy-driven antibody production, phage-neutralizing antibodies are present in up to 40% of individuals who have not undergone phage therapy (32). This finding supports the presence of either (i) cross-reactive antibod ies generated against environmental phages or (ii) phage neutralization by natural antibodies. Natural antibodies are low-affinity, polyreactive antibodies generated by B cells without a simulating antigen (33) and bind evolutionarily conserved epitopes occurring in microbes such as phospholipid phosphorylcholine (34) or glycans (35). Foreseeable improvements to current phage therapy pipelines may therefore include the quantification of baseline and treatment-induced neutralizing antibodies (at a patientspecific and community level). While these data may provide important insights into phage immunogenicity, the relationship between antibodies and therapeutic efficacy remains uncertain, as the development of antiphage antibodies does not exclude positive clinical outcomes (36).
## Complement
In addition to antibodies, complement deposition on phage has been well documented, inhibiting phage infectivity (37,38) and acting as an opsonin to increase phagocyte engulfment (10). Importantly, complement deposition initiated by both natural IgM antibodies (39) and acquired phage-specific antibodies (30) via the classical pathway of complement activation has been documented. Phage-mediated complement pathway activation by other innate signaling molecules has not been shown but is possible. Soluble heteromeric innate sensors such as IgM, MBL, and pentraxins function to recognize repetitive protein motifs such as those found in phage capsids (40) (Fig. 1) and may stimulate the deposition of complement. Pentameric C-reactive protein and serum amyloid P are prime examples of pentraxins secreted by the liver as part of the acute-phase response that are deposited on microbes and dying cells, facilitating the binding of complement factors (41).
In summary, it may be just as important to routinely measure antiphage immune responses (e.g., natural and acquired antibodies), in addition to circulating phage concentration during phage therapy. These data will provide insights into the mecha nisms and antigens that trigger phage recognition, aiding the development of secondand third-generation phage therapies based on modified phages and/or immunomodu lation.
## PHAGE CLEARANCE VIA NON-IMMUNE MECHANISMS
Phage uptake is not restricted to immune cells. Upon intravenous injection, circulating phages also interact with the blood vessel endothelium, which must be transversed to enter tissue. This may be desirable to reach a site of infection but requires transcytosis of phage as opposed to engulfment and intracellular degradation. T4 phage transcytosis has been documented across endothelial and epithelial cell lines (42) but has also been shown to accumulate within cell monolayers where they become largely inactivated (43).
The liver is the primary site of phage accumulation in vivo, where both Kupffer cells and, to a lesser degree, primary liver sinusoidal endothelial cells (LSECs) filter phage from blood (19). LSEC internalization and degradation of K1F and T4 phage have been demonstrated in vivo (19) and ex vivo (44), respectively, in a manner that appears independent of known LSEC endocytosis receptors stabilin 1/2, mannose receptor, or the FcγRIIb2. In line with immune-mediated internalization, understanding the mechanisms of non-immune internalization will be important to prevent phage clearance and, if needed, promote transcytosis into tissues. Indeed, interactions with cell surface receptors may be key to this process. Studies using cancer cell lines have shown that phages can bind sialic acid (SA) glycans (45) and integrins (46). Interestingly, this represents a shared mechanism of internalization among prokaryotic and eukaryotic viruses (47). Multiple low-affinity interactions between repeating icosahedral capsid proteins and highly expressed outer membrane SA/integrin molecules may mediate binding and internalization, which can be inhibited by blocking the interaction (45,46).
## PHAGE MODIFICATION TO ESCAPE IMMUNE CLEARANCE
The success of phage therapy will require the selection or production of phages that are immunologically camouflaged, extending their duration in circulation to enable bacterial infection and lysis. Unfortunately, as phages are incredibly diverse genetically with minimal characterization of their protein encoding genes, we still know very little regarding phage immunogenicity across the spectrum of phage species. It remains to be determined which structural components of phages lead to their immunogenic properties and whether phage immunogenicity might be predictable via genome characterization.
Phage display may prove to be a particularly useful tool for improving our under standing of the interaction between phage components and the immune system. This method, based on the construction of a polypeptide library fused to phage capsid proteins and affinity selection of binders, has been employed to identify peptide-recep tor binding for a variety of applications (48). Phage display can be used to characterize the phage sites that are recognized by the immune system to be targeted for modification, e.g., sites to which antibodies bind (49), and peptides prolonging blood circulation (50) or involved in phage internalization by eukaryotic cells (51) or translocation to body sites (52).
Both genetic engineering (e.g., recombineering or CRISPR-Cas [53,54]) and natural techniques (e.g., serial co-incubation and mutant selection [55,56]) have been explored to reduce phage immunogenicity. Generated phages possess modifications in immu nogenic epitopes (e.g., capsid proteins) to prolong phage persistence in blood and reduce the endotoxin load released by bacterial cell lysis (e.g., lethal lysis-deficient phage variants [57,58]), thus limiting immune activation. In principle, capsid genes encoding proteins that mimic human proteins could also be used to reduce the likelihood of immune recognition. Whether this is feasible is not yet known. Indeed, more work is needed to understand phage epitopes and immunogenic properties in bacterial hosts and humans before any of these approaches can be fully realized.
Cloaking approaches such as liposomes (59) offer immune protection and can lead to controlled phage release into tissues to target intracellular pathogens. Phages target ing extracellular bacteria, however, will likely require modifications to their structural proteins to prevent uptake. While natural capsid glycosylation can impair phage-spe cific antibody recognition (60), the addition of glycans can drastically increase cellular uptake in a glycan-and organ-specific manner, suggesting that glycosidase treatments may enhance circulation time (61). Additionally, while increasing the positive charge of phage surface proteins via nanocapping has been shown to increase phage engulfment (62), reducing the positive charge can effectively increase the circulating half-life of phage particles (63). This is a particularly enticing avenue for research as it aligns with long-circulating lambda phage mutants that possess an acidic glutamic acid residue in place of the wild-type basic lysine residue (56). Lastly, coating of phage capsids with polyethylene glycol has also shown to prolong phage circulation and reduce immune recognition, representing another phage modification option (64).
## CONCLUSION
While phages offer advantages such as host specificity and natural abundance, their clinical utility appears to be limited by rapid immune clearance and immunogenicity. Genetic engineering, encapsulation, and phage immunogenicity screening approaches have shown potential to enhance phage stability and prolong circulation. Whether these approaches will truly improve the therapeutic effect of phage therapies in vivo remains to be seen. Nonetheless, the first successful use of an engineered phage created via bacteriophage recombineering of electroporated DNA against drug-resistant Mycobacte rium abscessus highlights the translational potential of such strategies (65). To fully realize the potential of phage therapy, future work must integrate immunological insights, synthetic biology, and targeted delivery to create durable, effective, and personalized treatments.
## References
1. Herelle (1917) "Sur un microbe invisible antagoniste des bacillus dysentérique" *Comptes Rendus Acad Sci*
2. Twort (1915) "An investigation on the nature of ultra-microscopic viruses" *Lancet*
3. Eskenazi, Lood, Wubbolts et al. (2022) "Combination of pre-adapted bacterio phage therapy and antibiotics for treatment of fracture-related infection due to pandrug-resistant Klebsiella pneumoniae" *Nat Commun*
4. Van Nieuwenhuyse, Van Der Linden, Chatzis et al. (2022) "Bacteriophage-antibiotic combination therapy against extensively drug-resistant Pseudomonas aeruginosa infection to allow liver transplantation in a toddler" *Nat Commun*
6. Kellerr, Jr (1958) "Fate of bacteriophage particles introduced into mice by various routes" *Proc Soc Exp Biol Med*
7. Echterhof, Dharmaraj, Khosravi et al. (2024) "The contribution of neutrophils to bacteriophage clearance and pharmacokinetics in vivo" *JCI Insight*
8. Tan, Chen, Jiang et al. (2024) "Evaluation of the impact of repeated intravenous phage doses on mammalian hostphage interactions" *J Virol*
9. Le, Nang, Zhao et al. (2023) "Therapeu tic potential of intravenous phage as standalone therapy for recurrent drug-resistant urinary tract infections" *Antimicrob Agents Chemother*
10. Roach, Leung, Henry et al. (2017) "Synergy between the host immune system and bacteriophage is essential for successful phage therapy against an acute respiratory pathogen" *Cell Host Microbe*
11. Le, Venturini, Lubian et al. (2025) "Differences in phage recognition and immunogenicity contribute to divergent human immune responses to Escherichia coli and Klebsiella pneumoniae phages" *Eur J Immunol*
12. Van Belleghem, Clement, Merabishvili et al. (2017) "Pro-and anti-inflammatory responses of peripheral blood mononuclear cells induced by Staphylococcus aureus and Pseudomonas aeruginosa phages" *Sci Rep*
13. Gogokhia, Buhrke, Bell et al. (2019) "Expansion of bacteriophages is linked to aggravated intestinal inflammation and colitis" *Cell Host Microbe*
14. Sweere, Van Belleghem, Ishak et al. (2025) "Perspective mBio December"
15. Birukova, Katznelson, Lazzareschi et al. (2019) "Bacteriophage trigger antiviral immunity and prevent clearance of bacterial infection" *Science*
16. Nang, Lin, Petrovic Fabijan et al. (2023) "Pharmacokinetics/pharmacodynamics of phage therapy: a major hurdle to clinical translation" *Clin Microbiol Infect*
17. Nungester, Watrous (1934) "Accumulation of bacteriophage in spleen and liver following its intravenous inoculation" *Exp Biol Med (Maywood)*
18. Rusckowski, Gupta, Liu et al. (2004) "Investigations of a 99m Tc-labeled bacteriophage as a potential infection-specific imaging agent" *J Nucl Med*
19. Inchley (1969) "The actvity of mouse Kupffer cells following intrave nous injection of T4 bacteriophage" *Clin Exp Immunol*
20. Yip, Hawkins, Smith et al. (1999) "Biodistribution of filamentous phage-Fab in nude mice" *J Immunol Methods*
21. Øie, Wolfson, Yasunori et al. (2020) "Liver sinusoidal endothelial cells contribute to the uptake and degradation of entero bacterial viruses" *Sci Rep*
22. Zborowsky, Seurat, Balacheff et al. (2025) "Macrophage-induced reduction of bacteriophage density limits the efficacy of in vivo pulmonary phage therapy" *Nat Commun*
23. Srivastava, Kaido, Carrier (2004) "Immunological factors that affect the in vivo fate of T7 phage in the mouse" *J Virol Methods*
24. Mancardi, Daëron (2014) "Fc receptors in immune responses"
25. Bernabéu-Gimeno, Pardo-Freire, Chan et al. (2024) "Neutralizing antibodies after nebulized phage therapy in cystic fibrosis patients" *Med*
26. Nick, Dedrick, Gray et al. (2022) "Host and pathogen response to bacteriophage engineered against Mycobacterium abscessus lung infection" *Cell*
27. Zhong, Xu, He et al. (2025) "Preexposure to phage particles reduces their antibacterial therapeutic efficacy both in vitro and in vivo" *Int J Med Microbiol*
28. Hodyra-Stefaniak, Miernikiewicz, Drapała et al. (2015) "Mammalian Host-Versus-Phage immune response determines phage fate in vivo" *Sci Rep*
29. Majewska, Kaźmierczak, Lahutta et al. (2019) "Induction of phage-specific antibodies by two therapeutic staphylococcal bacteriophages administered per os" *Front Immunol*
30. Berkson, Wate, Allen et al. (2024) "Phage-specific immunity impairs efficacy of bacteriophage targeting Vancomycin Resistant Enterococcus in a murine model" *Nat Commun*
31. Gembara, Dąbrowska (2021) "Phage-specific antibodies" *Curr Opin Biotechnol*
32. Dąbrowska, Miernikiewicz, Piotrowicz et al. (2014) "Immunogenicity studies of proteins forming the T4 phage head surface" *J Virol*
33. Ledeboer, Bezemer, De Hiaard et al. (2002) "Preventing phage lysis of Lactococcus lactis in cheese production using a neutraliz ing heavy-chain antibody fragment from llama" *J Dairy Sci*
34. Hodyra-Stefaniak, Kaźmierczak, Majewska et al. (2020) "Natural and induced antibodies against phages in humans: induction kinetics and immunogenicity for structural proteins of PB1related phages" *Phage (New Rochelle)*
35. Maddur, Lacroix-Desmazes, Dimitrov et al. (2020) "Natural antibodies: from first-line defense against pathogens to perpetual immune homeostasis" *Clin Rev Allergy Immunol*
36. Wardemann, Yurasov, Schaefer et al. (2003) "Predominant autoantibody production by early human B cell precursors" *Science*
37. Bello-Gil, Audebert, Olivera-Ardid et al. (2019) "The formation of glycan-specific natural antibodies repertoire in GalT-KO mice is determined by gut microbiota" *Front Immunol*
38. Łusiak-Szelachowska, Żaczek, Weber-Dąbrowska et al. (2017) "Antiphage activity of sera during phage therapy in relation to its outcome" *Future Microbiol*
39. Hodyra-Stefaniak, Lahutta, Majewska et al. (2019) "Bacteriophages engineered to display foreign peptides may become short-circulating phages" *Microb Biotechnol*
40. Egido, Dekker, Toner-Bartelds et al. (2023) "Human complement inhibits myophages against Pseudomonas aeruginosa" *Viruses*
41. Sokoloff, Bock, Zhang et al. (2001) "Specific recognition of protein carboxy-terminal sequences by natural IgM antibodies in normal serum" *Mol Ther*
42. Zabel, Kündig, Bachmann (2013) "Virus-induced humoral immunity: on how B cell responses are initiated" *Curr Opin Virol*
43. Ma, Parente, Zhong et al. (2023) "Comple ment-pentraxins synergy: navigating the immune battlefield and beyond" *Biomed Pharmacother*
44. Nguyen, Baker, Padman et al. (2017) "Bacteriophage transcytosis provides a mechanism to cross epithelial cell layers" *mBio*
45. Bichet, Chin, Richards et al. (2021) "Bacteriophage uptake by mamma lian cell layers represents a potential sink that may impact phage therapy"
46. Romano, Simón-Santamaría, Mccourt et al. (2024) "Liver sinusoidal cells eliminate blood-borne phage K1F"
47. Lehti, Pajunen, Skog et al. (1915) "Internalization of a polysialic acid-binding Escherichia coli bacteriophage into eukaryotic neuroblastoma cells" *Nat Commun*
48. Dabrowska, Opolski, Wietrzyk et al. (2004) "Antitumor activity of bacteriophages in murine experimental cancer models caused possibly by inhibition of beta3 integrin signaling pathway" *Perspective mBio*
49. Maginnis (2018) "Virus-receptor interactions: the key to cellular invasion" *J Mol Biol*
50. Huang, Bishop-Hurley, Cooper (2012) "Development of antiinfectives using phage display: biological agents against bacteria, viruses, and parasites" *Antimicrob Agents Chemother*
51. Tremblay, Tegoni, Spinelli et al. (2006) "Receptor-binding protein of Lactococcus lactis phages: identification and characterization of the saccharide receptor-binding site" *J Bacteriol*
52. Jin, Wang, Sha et al. (2021) "A blood circulation-prolonging peptide anchored biomimetic phage-platelet hybrid nanoparticle system for prolonged blood circulation and optimized anti-bacterial performance" *Theranos tics*
53. Duerr, White, Schluesener (2003) "Identification of peptide sequences that induce the transport of phage across the gastrointestinal mucosal barrier" *J Virol Methods*
54. Asar, Newton-Northup, Soendergaard (2024) "Improving pharmacokinetics of peptides using phage display" *Viruses*
55. Carmody, Goddard, Nugen (2021) "Bacteriophage capsid modification by genetic and chemical methods" *Bioconjug Chem*
56. Matsuda, Freeman, Hilbert et al. (2005) "Lysis-deficient bacteriophage therapy decreases endotoxin and inflammatory mediator release and improves survival in a murine peritonitis model" *Surgery*
57. Merril, Biswas, Carlton et al. (1996) "Long-circulating bacteriophage as antibacterial agents" *Proc Natl Acad Sci*
58. Vitiello, Merril, Adhya (2005) "An amino acid substitution in a capsid protein enhances phage survival in mouse circulatory system more than a 1000-fold" *Virus Res*
59. Jia, Jia, Yin et al. (2023) "Engineering bacterioph ages for enhanced host range and efficacy: insights from bacteriophagebacteria interactions" *Front Microbiol*
60. Paul, Sundarrajan, Rajagopalan et al. (2011) "Lysis-deficient phages as novel therapeutic agents for controlling bacterial infection" *BMC Microbiol*
61. Singla, Harjai, Katare et al. (2016) "Encapsulation of bacteriophage in liposome accentuates its entry in to macrophage and shields it from neutralizing antibodies" *PLoS One*
62. Freeman, Robotham, Parks et al. (2023) "Virion glycosylation influences mycobacteriophage immune recognition" *Cell Host Microbe*
63. Lin, Sojitra, Carpenter et al. (2023) "Chemoenzymatic synthesis of genetically-encoded multivalent liquid Nglycan arrays" *Nat Commun*
64. Meng, Yang, Pang et al. (2005) "Nanocappingenabled charge reversal generates cell-enterable endosomal-escapable bacteriophages for intracellular pathogen inhibition" *Sci Adv*
65. Prasuhn, Singh, Strable et al. (2008) "Plasma clearance of bacteriophage Qβ particles as a function of surface charge" *J Am Chem Soc*
66. Kim, Cha, Jang et al. (2008) "PEGylation of bacteriophages increases blood circulation time and reduces T-helper type 1 immune response" *Microb Biotechnol*
67. Dedrick, Guerrero-Bustamante, Garlena et al. (2019) "Engineered bacteriophages for treatment of a patient with a disseminated drug-resistant Mycobacterium abscessus" *Nat Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12268033&blobtype=pdf | ## Transcriptomic skin niches in systemic sclerosis underpin a role for mitochondrial dysfunction
Dear Editor, Systemic sclerosis (SSc), or scleroderma, is a chronic, multiorgan inflammatory disease characterized by vasculopathy, fibrosis of the skin and visceral organs, and immune dysregulation. As the pathogenesis of SSc remains elusive, there have been only modest clinical improvements from treatments targeting fibrosis and/or inflammatory responses. 1 Based on our previous studies analysing spatial patterning of inflammation in the skin, 2 we hypothesized that the pathogenic processes in SSc are not diffusely uniform throughout the dermis, but represent a network of functionally distinctive topographic areas (niches) that share one or more common and overlapping molecular mechanisms. Therefore, we performed spatial transcriptomic analysis of 11 biopsies from lesional SSc skin. Detailed methods are available from the corresponding author on request.
Transcriptomes corresponding to 55-μm diameter circular barcodes overlapping vertical skin sections were pooled from all biopsies and analysed with Uniform Manifold Approximation and Projection (UMAP) using previously established bioinformatics pipelines (https://satijalab.org/ seurat/articles/spatial_vignette) (Figure 1a). These transcriptomes were then visualized with Loupe Browser (10x Genomics, Pleasanton, CA, USA) (Figure 1b). We identified six distinct transcriptomic clusters, one of which (cluster 2) encompassed the epidermis, while the remaining clusters represented irregular, interwoven zones spanning the entire dermis. Clusters 3 and 5 were associated with hair follicles, and cluster 4 represented pericyte-enriched perivascular areas. Although potentially significant, the small number of barcodes in these clusters limited their interpretability within the scope of our analysis.
Cell type annotation showed that cluster 1 was predominantly comprised of SSc fibroblast subtypes (SFRP2, CRABP1/ASPN, CCL19/C7 ), complemented by CXCL9/SPP1 macrophages and smooth muscle cells (Figure 1c). 3 Query of the differentially expressed genes in each cluster against the Gene Ontology (GO) (geneontology.org) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (genome.jp/ kegg/) databases confirmed that this cluster represented the primary fibrotic zone (extracellular matrix organization GO:0030198, collagen fibril organization GO:0030199) with a component of an immune response signature (antigen processing and presentation GO:0019882).
Cluster 0 occupied predominantly the upper dermis and encompassed diverse keratinocyte subpopulations along with CRABP1/ASPN fibroblasts (Figure 1c), supporting previous observations that CRABP1/ASPN + fibroblasts may represent fibroblasts found in dermal papillae. 3 Surprisingly, a prominent fibrotic signature was not observed in this cluster.
Instead, its defining characteristic was a strong association with mitochondrial metabolism, marked by an abundance of transcripts linked to the respiratory electron transport chain (GO:0022904) and oxidative phosphorylation (GO:0006119). Additionally, a robust humoral immune response signature (GO:0006959) was evident, though the antigen presentation signature (GO:0019882) observed in cluster 1 was absent.
To further define the mitochondrial changes associated with SSc, we analysed changes in the relative expression levels of mitochondrial DNA (mtDNA)-encoded and nuclear DNA (nDNA)-encoded mitochondrial genes across transcriptomic clusters (Figure 1d). Cluster 0 (and cluster 5) exhibited downregulation of several nDNA-encoded mitochondrial genes, such as COX4I1, COX7C and COX6A1, alongside a compensatory upregulation of mtDNA-encoded mitochondrial genes like MT-ND4 and MT-ND5. This opposing trend between nDNA-and mtDNA-encoded mitochondrial gene expression has been linked with the mitochondrial unfolded protein response (UPR mt ) 4 and the integrated stress response (ISR), 5 and is suggested to be a compensatory mechanism that is activated to maintain mitochondrial function under cellular stress conditions. In contrast, both nDNA-encoded and mtDNA-encoded mitochondrial genes were consistently downregulated in the fibrotic areas (cluster 1), including those encoding key components of aerobic respiration such as COX4I1, COX7C and MT-ND4, indicating total mitochondrial dysfunction and impaired ability to activate UPR mt or engage ISR.
Together, our data using an unbiased, transcriptomic map of SSc skin revealed spatial zonation partitioning the pathogenic inflammatory and fibrotic processes into a zone of inflammation (cluster 0) and fibrosis (cluster 1). Mitochondrial impairment was the most striking difference between these zones and potentially represented an increased mitochondrial stress response in inflammatory zones. Hence, these clusters may represent distinct patterning reflecting sequential temporal phases during the pathogenesis of SSc.
Although the cause of SSc remains unknown, it is likely to be multifactorial. Several environmental and occupational toxins have been linked to an increased risk of SSc, and interestingly, they all share the ability to promote genotoxic stress and toxicity towards mitochondria. 6 Mitochondrial dysfunction can promote oxidative stress, mutagenesis of nuclear DNA and cellular senescence -all of which are present in SSc. 7,8 Furthermore, dysfunctional mitochondria may leak DNA into the cytosol, thereby activating the cGAS-STING pathway and instigating type I interferon production, contributing to local inflammation. 9 Finally, persistent mitochondrial impairment may foster a senescence-associated secretory phenotype, 8 thereby exacerbating oxidative stress and promoting fibrosis in SSc. We acknowledge that the modest sample size, limited healthy control data and absence of comprehensive functional validation warrant further investigation to fully elucidate these mechanisms.
Although the mechanisms leading to mitochondrial disturbances are undoubtedly heterogeneous and multifactorial, our results indicate that adjunct pharmacological approaches targeting mitochondria may be relevant even in established SSc.
Junqin Wang ,foot_0 Dylan Hennessey, 1 Aishwarya Iyer, 1 Sandra O'Keefe, 1 Lamia Khan,foot_1 Desiree Redmond, 2 Mohammed Osman 2 and Robert Gniadecki 1
## References
1. Papa, Pignataro, Zaccara (2018) "Autologous hematopoietic stem cell transplantation for treatment of systemic sclerosis" *Front Immunolog*
2. Ringham, Prusinkiewicz, Gniadecki (2019) "Skin patterning in psoriasis by spatial interactions between pathogenic cytokines" *iScience*
3. Tabib, Huang, Morse (2021) "Myofibroblast transcriptome indicates SFRP2 hi fibroblast progenitors in systemic sclerosis skin" *Nat Commun*
4. Shpilka, Haynes (2017) "The mitochondrial UPR: mechanisms, physiological functions and implications in ageing" *Nat Rev Mol Cell Biol*
5. Pakos-Zebrucka, Koryga, Mnich (2016) "The integrated stress response" *EMBO Rep*
6. Dorado, Jeljeli, Chêne et al. (2019) "Implication of oxidative stress in the pathogenesis of systemic sclerosis via inflammation, autoimmunity and fibrosis" *Redox Biol*
7. Gniadecki, Iyer, Hennessey (2022) "Genomic instability in early systemic sclerosis" *J Autoimmun*
8. Bueno, Papazoglou, Valenzi (2020) "Mitochondria, aging, and cellular senescence: implications for scleroderma" *Curr Rheumatol Rep*
9. Steadman, Reilly (2025) "Aberrant fumarate metabolism links interferon release in diffuse systemic sclerosis" *J Dermatol Sci* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12772400&blobtype=pdf | # Cathepsin S contributes to influenza-induced lung injury by driving inflammation, promoting apoptosis, and disrupting epithelial barrier integrity
Tianxin Ma, Chunguang Yang, Yang Wang, Chuanmeizi Tu, Jiawei Zhang, Kailin Mai, Shengzhen Wu, Hongxuan Zhou, Shengfeng Li, Sulan Ye, Lixi Liang, Qingsheng Huang, Zhenhui Zhang, Zhengshi Lin, Weiqi Pan, Zifeng Yang
## Abstract
Influenza virus infection causes significant morbidity and mortality worldwide, with severe cases often driven by excessive host inflammatory responses and disruption of epithelial barrier function. Here, we identified cathepsin S (CTSS), a lysosomal cysteine protease, as a key mediator of influenza-induced lung injury. Influenza virus infection upregulated CTSS in a time-dependent and dose-dependent manner, leading to lysosomal membrane permeabilization and cytoplasmic release of CTSS, which correlated with increased apoptosis and loss of epithelial barrier integrity. Knockdown of CTSS reduced proinflammatory cytokine production, apoptosis, and barrier disruption in A549 cells. Air-liquid interface airway epithelial cultures further validated the essential role of CTSS in preserving epithelial barrier integrity. In vivo, pharmacological inhibition of CTSS alleviated lung inflammation and disease severity in infected mice without reducing viral titers. Tumor necrosis factor-alpha (TNF-α) strongly induced CTSS activation, but failed to restore apoptosis and inflammatory responses in CTSS-knockdown cells, suggesting that CTSS acts downstream of cytokine signaling. These findings reveal a central role for CTSS in linking viral infection, cytokine storm, and epithelial damage and highlight CTSS as a promising target for host-directed therapy against severe influenza. IMPORTANCE Severe influenza is often driven by excessive host inflammation and epithelial barrier disruption, yet the molecular mediators connecting viral infection to these pathological processes remain poorly defined. This study identifies cathepsin S (CTSS) as a central driver of influenza-induced epithelial injury, acting downstream of viral replication and cytokine signaling to promote apoptosis, inflammation, and barrier disruption. Using both in vitro and in vivo models, we demonstrate that inhibiting CTSS preserves barrier integrity and attenuates inflammation despite having no beneficial effect on viral clearance. These findings provide mechanistic insight into influenza pathogenesis and support CTSS as a promising target for host-directed interventions to mitigate cytokine storm-driven lung injury in severe respiratory viral infections. KEYWORDS cathepsin S, inflammatory response, apoptosis, epithelial barrier I nfluenza is an acute, highly contagious respiratory disease caused by influenza viruses that occur annually and occasionally cause pandemics (1). Globally, influenza is estimated to cause approximately 1 billion infections each year, including 3-5 million cases of severe illness and 290,000-650,000 respiratory-related deaths (2, 3). Typical clinical features include the sudden onset of fever, cough, sore throat, myalgia, and fatigue (4, 5). While most cases are self-limiting, the disease can rapidly progress to severe complications in vulnerable populations, including acute lung injury and, in
critical cases, acute respiratory distress syndrome with multi-organ dysfunction, often leading to death (6,7).
Influenza virus primarily infects the respiratory tract and replicates in epithelial cells. Alveolar epithelial cells are the key components of the air-blood barrier, which consists of alveolar epithelium, basement membrane, and endothelial cells. During acute lung injury, this barrier is disrupted by damage to intercellular junctions as well as injury and death of epithelial and endothelial cells (8). The extracellular matrix (ECM), a major component of the cellular microenvironment, also contributes to influenza-induced lung injury through extensive remodeling and degradation (9,10).
Most mechanistic studies have focused on immune cell responses and inflammatory signaling, but there remains a significant gap in understanding the direct pathogene sis of influenza-induced acute lung injury. Cathepsin S (CTSS) has recently garnered attention for its role in inflammation and infection (11)(12)(13). CTSS is a lysosomal cysteine protease with structural features that distinguish it from other family members such as cathepsin B (CTSB) and cathepsin L (CTSL) (14). Importantly, CTSS maintains enzymatic activity across a broad pH range, with high activity in both acidic and neutral environ ments. Its substrates include multiple ECM components-laminin, fibronectin, elastin, and collagen-as well as basement membrane constituents such as chondroitin sulfate, heparan sulfate, and proteoglycans (15). Researchers have enriched their understanding of this protease and found that it is involved in the regulation of a series of key physio logical processes, including cell cycle regulation, tissue remodeling, inflammation, and immune response (16). Pathologically, while the role of CTSS in antigen presentation has long been recognized, CTSS has been implicated in the pathogenesis of some viral infections (17). CTSS is primarily produced by antigen-presenting cells such as macrophages but can also be expressed and secreted by airway epithelial cells, alveolar epithelial cells, and vascular endothelial cells (18,19).
This study aimed to investigate the role of CTSS in the pathogenesis of influenza infection, with a particular focus on acute lung injury. Understanding the contribution of CTSS to influenza pathology may provide new insights and support the development of host-directed therapeutic strategies.
## MATERIALS AND METHODS
## Cells and virus
MDCK (NBL-2) cells (American Type Culture Collection [ATCC], CCL-34) were cultured in minimum essential medium (Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco) at 37°C with 5% CO 2 . A549 cells (ATCC, CCL-185) were cultured in Dulbecco's modified Eagle medium/F12 (1:1) (DMEM/F12, Gibco) supplemented with 10% FBS, under the same conditions. The Influenza A/Puerto Rico/8/1934 (H1N1) (PR8) virus (ATCC, VR-95PQ) was propagated in 9-11-day-old embryonated chicken eggs and titrated in MDCK cells using either plaque assays or 50% tissue culture infective dose (TCID 50 ) assays (20,21).
## Animal studies
Specific pathogen-free, 6-8-week-old female C57BL/6J mice were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. and housed in individually ventilated cages. In accordance with ethical guidelines, any animal that lost more than 25% of its body weight was humanely euthanized.
For influenza infection, mice were anesthetized with isoflurane and intranasally inoculated with various doses of PR8 virus. As a control, one group of mice received an equal volume of phosphate buffered saline (PBS). The body weight and survival rate were monitored for 14 days post-infection (dpi). Lungs and bronchoalveolar lavage fluid (BALF) were collected at designated time points for viral titration, pathological examination, CTSS activity, and transcriptomic analysis.
For CTSS inhibition, mice received intraperitoneal injections of 30 mg/kg LY3000328 (MedChem Express, USA, HY-15533) or vehicle twice daily for 5 days. Mice were inoculated intranasally with 100 PFU of PR8 virus 4 h after the first dose of LY3000328. The body weight and survival rate were monitored for 14 dpi, and lungs from infec ted mice were harvested at 5 dpi for viral titration, histopathology, CTSS activity, and cytokine expression analysis.
## Transcriptome analysis
C57BL/6J mice were infected with specified doses of the virus, with mock-infected mice serving as controls. Lung tissues were harvested at designated time points, and total RNA was isolated using TRIzol (Invitrogen, USA, 15596026CN) according to the manufac turer's instructions. The RNA was quantified, and libraries were prepared at Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). Libraries were sequenced on the NovaSeq 6000 platform (Illumina, USA) using 150 bp paired-end reads. Raw reads were processed and trimmed to ensure high-quality, reliable data, which were then mapped to the mouse genome using HISAT2 (v2.0.5). Gene expression levels were quantified as fragments per kilobase of transcript per million mapped reads (FPKM) using Feature Counts (v1.5.0-p3). Differentially expressed genes (DEGs) were identified using DESeq2 (v1.20.0), with a significance threshold of P adj ≤ 0.05 and |log ₂ (fold change)| ≥ 1.
Functional enrichment analysis of DEGs was performed in R (v4.2.2) using cluster Profiler, enrichplot, GOplot, ggplot2, and the org.Mm.eg.db database. GO and KEGG pathway analyses used P value and q value cutoffs of 0.05. ECM-related genes were curated based on the Gene Ontology database.
Weighted gene co-expression network analysis (WGCNA) identified clusters of co-regulated genes, or "modules, " associated with specific infection parameters. Using unsupervised hierarchical clustering of FPKM data for genes expressed above a threshold of 0.5 across all time points, a weighted adjacency matrix was constructed with a soft threshold power of 14 to approximate a scale-free topology. Thirty modules were identified by average linkage hierarchical clustering. Module-trait relationships, membership, and gene significance were assessed using Pearson correlation. Modules significantly correlated with traits such as time post-infection, infection status, weight change, and viral titer (P value < 0.05) were further analyzed. Hub genes were identified using Cytoscape (v3.9.1) based on the topological overlap matrix of the relevant module.
## CTSS knockdown in A549 cells
CTSS-targeting siRNAs [siCTSS #1: 5′-CAUGUUCAAAGUACACUGA (dT)(dT)-3′ and siCTSS #2: 5′-CAAUGGGAAUGCACUCAUA (dT)(dT)-3′] and a scramble control siRNA [siSCR: 5′-UUCUCCGAACGUGUCACGU (dT)(dT)3′] were purchased from Tsingke Biotechnology (Beijing, China). Transfections were performed with Lipofectamine RNAiMAX (Thermo Fisher Scientific, USA, 13778075) following the manufacturer's instructions. Final siRNA concentration was 100 nM in six-well plates.
## Influenza infection in A549 cells
A549 cells were seeded 12 h before transfection and infected with the influenza virus at the indicated multiplicity of infection (MOI) 24 hours post-transfection. After a 1.5 h absorption, the viral inoculum was removed, and the wells were replenished with DMEM/F12 supplemented with 1 µg/mL tosyl phenylalanyl chloromethyl ketone-treated trypsin (Sigma, USA, 4370285).
For the extraction of lysosomes, supernatant, lysosomes, and cytoplasm were isolated at the indicated time points post-infection using the Lysosome Protein Isolation Kit (BestBio, China, BB-31452), following the manufacturer's instructions.
For the recombinant protein treatment experiment, human-CTSS (MedChem Express, USA, HY-P7756) was added at the indicated doses 1.5 hours post-infection (hpi), cells were harvested and lysed 24 hpi for subsequent experiments.
For TNF-α treatment, human TNF-α protein (MedChem Express, USA HY-P7058) was added at 100 ng/mL 1.5 hpi. Cells were harvested and lysed at 24 hpi for subsequent analyses.
Gene expression was measured by real-time quantitative polymerase chain reaction (RT-qPCR), protein levels by western blotting, and enzymatic activity by CTSS activity assays at designated time points.
## Real-time qPCR
Total RNA from tissues or cells was extracted with TRIzol reagent (Invitrogen, USA, 15596026CN). cDNA was synthesized using HiScript IV RT SuperMix (Vazyme, China, R423). Real-time qPCR was performed with Universal SYBR qPCR Master Mix (Vazyme, China, Q712) according to the manufacturer's instructions. Primer sequences are listed in Table S1.
## Western blotting
Lung or cell lysates were prepared in radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime, China, P0013) supplemented with protease inhibitor phenylmetha nesulfonyl fluoride (PMSF). The lysates were mixed with 5× SDS loading buffer and denatured at 95°C for 10 min. Samples were then separated by SDS-PAGE and transferred to hydrophobic PVDF membranes (Millipore, USA, IPVH00010). The membranes were blocked with 5% skim milk for 1 h at room temperature, then incubated with pri mary and horseradish peroxidase (HRP)-conjugated secondary antibodies. The primary antibodies used included anti-human CTSS rabbit polyclonal antibody (Proteintech, China, 27538-1-AP), anti-mouse CTSS rabbit monoclonal antibody (Affinity, UK, DF8246), anti-β-actin rabbit monoclonal antibody (Abcam, UK, ab8227), anti-PARP1 monoclonal antibody (Gene Tex, USA, GTX100573), anti-GSDMD-N monoclonal antibody (Abcam, UK, ab215203), anti-pMLKL monoclonal antibody (Abcam, UK, ab187091), anti-E-cad herin antibody (Abcam, UK, ab40772), anti-cleaved-PARP1 rabbit monoclonal antibody (Zenbio, China, R380374), anti-β-tubulin rabbit monoclonal antibody (Zenbio, China, R380628), anti-influenza A virus NP monoclonal antibody (Sino Biological, USA, 11675-T62), and anti-LAMP1 monoclonal antibody (CST, USA, 9091S). Detection was performed using HRP-conjugated goat anti-rabbit IgG (H&L) (Abcam, UK, ab6721).
Band intensities were quantified using ImageJ. Target proteins were normalized to loading controls, and data are presented as fold change relative to controls.
## CTSS activity assay
CTSS activity was measured with the CTSS Activity Assay Kit (Abcam, UK, ab65307) following the manufacturer's protocol. Briefly, 50-200 μg of tissue extracts, cell lysates, or BALF were prepared in lysis buffer, and the supernatants were collected. Protein concentrations were determined using the BCA Protein Assay Kit (Beyotime, China, P0009). For activity measurements, 50 µL sample, 50 µL of reaction buffer, and 2 µL of substrate were mixed in a 96-well white plate and incubated at 37°C for 1-2 h. Fluorescence was recorded using 400 nm excitation and 505 nm emission.
## Cell death detection
A549 cells were seeded in 96-well plates, then the siRNA transfection and influenza virus infection were carried out in sequence according to the method described earlier. At 24 hpi, supernatants were collected and used for cell death detection using the Cytotoxicity LDH Assay Kit (Roche, USA, 91963). Monolayers were fixed in 4% paraformaldehyde for 30 min at room temperature, permeabilized with 0.5% Triton X-100 for 30 min, washed with PBST, and stained using the TUNEL Detection Kit (Beyotime, China, C1090). Images were acquired by fluorescence microscopy.
## Trans-epithelial electrical resistance measurement
Differentiated, mature human airway organoids grown on Transwell inserts under air-liquid interface (ALI) conditions were kindly provided by Dr. Guanghui Jin (The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China). siRNA transfection followed the A549 protocol. At 24 h post-transfection, trans-epithelial electrical resistance (TEER) was measured using a Millicell ERS2 (Millipore, USA) equipped with chopstick electrodes, according to the manufacturer's instructions (22,23). The measured resistance (Ω) was corrected by subtracting the blank insert and converted to TEER (Ω cm²) (24):
$$TEER = (measured resistance -blank resistance) × membrane area (cm 2 ) .$$
## Statistical analysis
Statistical analyses were performed in Prism (v9.0.0). Comparisons between two groups were performed using the unpaired two-tailed Student's t-test. For analyses involving multiple groups, one-way ANOVA followed by Tukey's post hoc test was utilized to assess significant differences. Survival curves were compared using the log-rank (Mantel-Cox) test, and P < 0.05 was considered statistically significant.
## RESULTS
## Influenza virus infection upregulates CTSS levels in the mouse lung
To determine the effects of influenza infection on the dynamic host responses in mouse lungs, 6-8-week-old female C57BL/6J mice were intranasally infected with serial doses (10-10,000 PFU) of A/Puerto Rico/8/1934 (H1N1) (PR8) or PBS. A dose-depend ent increase in body weight loss was observed, with higher doses resulting in more significant weight loss (Fig. S1A). Survival rates were dose dependent: 83.3% (5/6) at 10 PFU, 16.7% (1/6) at 100 PFU, and 0% at 1,000 or 10,000 PFU (Fig. S1B). Consistently, viral loads in lungs reached a peak at 3 dpi, with higher viral loads correlating with increased infection doses (Fig. S1C). Hematoxylin and eosin (H&E) staining revealed progressive lung lesions and immune cell infiltration with increasing doses and over time from 1 to 7 dpi, with slight recovery noted at 14 dpi (Fig. S1D).
To identify key host factors in acute lung injury caused by influenza A virus (IAV) infection, we performed lung transcriptomics across doses and time points. Principal component analysis (PCA) is shown in Fig. 1A. DEGs (log ₂ -centered FPKM) versus PBS controls are summarized in volcano plots (Fig. 1B). As ECM-associated protease activity is important for lung function after severe respiratory infection, we evaluated ECM-related gene regulation; heat maps of the top 30 ECM-related genes across dose and time are shown in Fig. 1C andD. Among these ECM-related proteases, CTSS-a lysosomal cysteine proteinase-displayed the most prominent upregulation, correlating with dose and duration of infection. In order to further validate the key function of DEGs involved in lung injury caused by influenza infection in mice, DEGs were grouped into 13 modules using WGCNA (Fig. 1E). Of note, ME-darkred was positively correlated with infection status, time post-infection, and viral titer, and negatively correlated with weight change, indicating a potential role in disease progression. GO enrichment analysis and protein interaction network analyses showed that this module was enriched for antiviral immune responses and apoptosis-related pathways (Fig. S2). Hub genes included CTSS, Myd88, Caspase8, Sifn2, and Mthfd2 (Fig. 1F).
We validated CTSS upregulation and activation in the lungs of mice infected with 100 PFU PR8. RT-qPCR showed significantly increased CTSS mRNA (Fig. 2A). Western blotting showed decreased pro-CTSS and increased activated CTSS (Fig. 2B). Moreover, CTSS activity in both BALF and lung tissue homogenates was significantly higher than in PBS controls (Fig. 2C). Activated CTSS further increased with dose and time (Fig. 2D andE), consistent with transcriptomic patterns.
## CTSS inhibition alleviates influenza virus-induced lung injury in mice
To evaluate whether CTSS inhibition attenuates influenza virus-induced pulmonary injury, mice were infected with 100 PFU PR8 and treated with the CTSS inhibitor LY3000328 (30 mg/kg, intraperitoneally, twice daily) or vehicle, mock mice received the vehicle and PBS (Fig. 3A). CTSS activity in lung tissues was assessed at 5 dpi. Compared with the mock group, CTSS activity was significantly elevated in the placebo group following infection, whereas LY3000328 treatment significantly suppressed CTSS activity relative to the placebo group (Fig. 3B). Despite similar lung viral loads at 5 dpi across both infected groups (Fig. 3C), notable differences in disease progression were observed. Placebo-treated mice exhibited progressive body weight loss from 3 dpi and began dying at 7 dpi, reaching 100% mortality by 9 dpi. While LY3000328-treated mice showed milder body weight loss, with partial weight recovery by 10 dpi, and 37.5% mice survived to 14 dpi (Fig. 3D andE).
The lung lesions in the LY300328-treated mice were obviously milder than in the placebo group (Fig. 3F), consistent with the differences in body weight loss between the two groups. Additionally, the mRNA expression level of pro-inflammatory cytokines and chemokines (i.e., Myd88, IL-6, TNF-α, CCL-5/RANTES) in lung tissues collected at 5 dpi was determined (Fig. 3G). Results showed that, compared to the placebo group, influenzainfected mice treated with LY3000328 had significantly lower mRNA expressions of cytokine and chemokine in the lungs.
## CTSS does not affect PR8 virus replication but reduces cytokine expression in A549 cells
To determine the role of CTSS in influenza infection-induced lung lesions, we first determined the expression and activity of CTSS in influenza virus-infected human lung epithelial (A549) cells. CTSS mRNA levels in A549 cells infected with different MOI of influenza virus were assessed at 12, 18, and 24 hpi (Fig. 4A). The results showed a significant increase in CTSS mRNA expression in a dose-dependent and time-dependent manner, with higher MOIs (0.5 and 1) showing greater increases in CTSS expression compared to mock and lower MOIs (0.1). Similarly, compared to the mock control, both pro-CTSS and activated CTSS protein levels were elevated in the influenza-infected A549 cells, with a trend of increasing CTSS protein levels corresponding to higher MOIs, which aligned with higher viral NP protein levels (Fig. 4B). Furthermore, CTSS enzymatic activity in A549 cell lysates increased in a dose-dependent manner after infection, and the activity of CTSS in the cell supernatant was also slightly increased (Fig. 4C).
Treatment of A549 cells with different concentrations of recombinant human CTSS protein during influenza virus infection significantly enhanced the mRNA expression of virus-induced inflammatory genes (Fig. S3A). To test effects on viral replication, CTSS was knocked down by siRNAs (siCTSS #1 and #2), both of which reduced CTSS protein (Fig. 5A). However, neither viral mRNA (M gene), viral proteins (NP and HA), nor viral titers were altered by CTSS knockdown (Fig. 5A through C). In contrast, the mRNA expression of TNF-α, RANTES, and IP-10 was downregulated after CTSS knockdown, consistent with the CTSS inhibition experiment in vivo (Fig. 5D).
## CTSS is released into the cytoplasm during influenza infection and promotes apoptosis
Given the observed correlation between programmed cell death and CTSS upregulation from WGCNA (Fig. 1F), we evaluated the cell death in A549 cells after CTSS knockdown. Our findings revealed that CTSS knockdown significantly reduced cell death caused by influenza virus infection through detecting lactate dehydrogenase (LDH) released in the cell supernatant (Fig. 6A). Similarly, TUNEL assay confirmed that CTSS knockdown significantly reduced TUNEL-positive cells caused by influenza infection (Fig. 6B).
Because influenza virus infection can induce lysosomal membrane permeabilization (LMP) (25), lysosomal contents may be released into the cytoplasm, causing cell stress and damage. To determine whether CTSS-a lysosome-resident cathepsin markedly upregulated after influenza infection-undergoes subcellular redistribution, we examined subcellular CTSS distribution and activity over time. At the 6 hpi, lysosomal CTSS activity was significantly elevated, whereas cytoplasmic activity remained low. From 12-24 hpi, lysosomal CTSS expression and activity declined, whereas cytoplasmic CTSS protein and activity increased, coinciding with elevated cleaved-PARP1, indicating the induction of apoptosis. Extracellular CTSS also rose modestly. These data suggest that activated CTSS is progressively released from lysosomes into the cytoplasm during infection, where it contributes to apoptosis (Fig. 6C andD).
To identify the predominant form of cell death, A549 cells were infected with increasing MOIs and analyzed for apoptosis (cleaved-PARP1), necroptosis (p-MLKL), and pyroptosis (cleaved-GSDMD). Cleaved-PARP1 increased dose-dependently, while p-MLKL and cleaved-GSDMD showed no consistent trends, indicating that apoptosis was the major pathway triggered by influenza infection in A549 cells (Fig. S4). Accordingly, siRNAmediated CTSS knockdown significantly reduced PARP1 cleavage without altering MLKL phosphorylation or GSDMD cleavage in A549 cells (Fig. 6E). Moreover, treatment with recombinant CTSS protein enhanced influenza-induced PARP1 cleavage in a dosedependent manner (Fig. S3B), further supporting a pro-apoptotic role for CTSS during infection.
## CTSS disrupts epithelial barrier integrity in human airway organoids
Because CTSS activity increased in supernatants after infection (Fig. 4C) and CTSS targets ECM and junctional components, we tested effects on epithelial barrier function. Western blotting analysis in A549 cells revealed that influenza virus infection led to a dose-dependent reduction in E-cadherin expression (Fig. S4), indicating disruption of cell-cell adhesion and compromise of barrier integrity. It is worth noting that siRNAmediated knockdown of CTSS significantly increased the expression level of E-cadherin and reversed the decrease in E-cadherin levels induced by influenza virus (Fig. 6E).
To further assess the functional impact of CTSS on epithelial barrier properties, differentiated airway organoids were transfected with CTSS-targeting siRNA. Twenty-four hours after transfection, TEER was measured, and organoids were lysed for western blotting analysis. Compared with the non-targeting control, CTSS knockdown produced a modest but significant increase in TEER (Fig. 7C). Consistently, western blotting confirmed reduced levels of activated-CTSS and cleaved-PARP1, accompanied by a slight enhancement of E-cadherin expression in the airway organoids (Fig. 7A andB). Because siRNA-mediated knockdown efficiency in organoids was limited, recombi nant CTSS (rCTSS) protein was added to differentiated airway organoids to further evaluate the role of CTSS in epithelial barrier function. Exogenous rCTSS at 10 ng and 100 ng caused a dose-dependent decrease in TEER (Fig. 7F) and significantly increased activated-CTSS and cleaved-PARP1, while reducing E-cadherin levels (Fig. 7D andE). These results demonstrate that CTSS compromises tight-junction integrity and epithelial barrier function, and they provide functional confirmation of the siRNA data.
Collectively, these data indicate that lowering CTSS activity helps preserve epithelial barrier integrity and intercellular junctions during influenza infection.
## DISCUSSION
In this study, we demonstrated that CTSS is markedly upregulated and activated during influenza virus infection and plays a critical role in modulating host inflammatory responses and epithelial barrier integrity rather than directly affecting viral replication.
In addition, we found that high-dose infection further increased the accumulation of activated-CTSS (Fig. 2E and3A). We propose that stronger viral replication and inflammation at higher doses enhance lysosomal activity, driving the conversion of pro-CTSS to its active form. Moreover, elevated proinflammatory cytokines such as TNF-α and interferon (IFN)-γ may additionally promote CTSS processing and activation, changes that are less likely to occur during low-dose infection. Activity inhibition of CTSS in mice did not alter pulmonary viral titers but significantly alleviated lung inflammation, reduced disease severity, and improved survival. Similarly, siRNA-mediated knockdown of CTSS in A549 cells led to decreased produc tion of proinflammatory cytokines, reduced cleavage of the apoptotic marker PARP1, and preserved E-cadherin expression, thereby maintaining tight junctions and epithe lial barrier function during infection. Conversely, treatment with recombinant CTSS protein exacerbated IAV-induced PARP1 cleavage and cytokine expression in A549 cells, indicating that CTSS acts as an amplifier of influenza-induced inflammatory and pathological injury process.
ECM-associated proteases are increasingly recognized as key regulators of host responses and viral pathogenesis (26). The ECM comprises structural proteins and several proteases that remodel these proteins (27). We previously reported that the ECM protease ADAMTS4 mediates lung injury during influenza infection (10). Here, transcrip tion analysis highlighted CTSS among eight major human cathepsins: only cathepsin W (CTSW) and CTSS were significantly upregulated, whereas others were downregulated or unchanged (Fig. S5), suggesting unique functions. CTSW is known to facilitate influenza virus release from late endosome (28), although CTSS is an interferon-stimulated gene upregulated by IFN-γ (29), its role in influenza-induced tissue injury had not been elucidated.
Cathepsins can affect viral infection either directly-by targeting viral proteins and altering the infection cycle-or indirectly-by modulating host immune responses. For example, CTSB and CTSL cleave viral glycoproteins to facilitate entry of Ebola virus (30,31); CTSB participates in human papillomavirus type 16 entry (32); CTSL and CTSS support reovirus infection in an acid-independent manner (33,34); and CTSL mediates spike protein cleavage in coronaviruses, including SARS-CoV, MERS-CoV, and SARS-CoV-2 (35,36). Therefore, we explored the mechanism of CTSS in influenza infection.
Mechanistically, we observed subcellular redistribution of CTSS from lysosomes into the cytoplasm during the late stages of infection, consistent with LMP (37). Lysoso mal leakage is a well-established trigger of apoptosis and has been reported for dengue virus, Newcastle disease virus, and coronaviruses including SARS-CoV, MERS-CoV, and SARS-CoV-2 (38)(39)(40). Our data suggest that accumulated CTSS in the cyto plasm contributes to apoptotic activation, leading to extensive epithelial cell death and barrier disruption. Knockdown of CTSS significantly reduced inflammation and apoptosis, preserved epithelial barrier function, and mitigated lung injury. Importantly, TNF-α robustly induced CTSS expression and activation; however, in CTSS-knockdown cells, TNF-α failed to restore apoptosis or inflammatory responses (Fig. S6), indicating that CTSS is a critical downstream effector of TNF-α-driven immunopathology. The role of CTSS in epithelial barrier disruption was further validated in air-liq uid interface (ALI) differentiated human airway organoids, which recapitulate airway architecture and barrier properties and have been widely used to study viral infec tion and epithelial dysfunction (41,42). CTSS knockdown significantly increased TEER and restored E-cadherin expression, indicating improved tight junction integrity and epithelial cohesion during infection. These findings provide strong experimental evidence that CTSS directly contributes to epithelial barrier breakdown in the context of viral infection.
Given the contribution of cytokine storm to severe influenza outcomes (43,44), our findings position CTSS as a host-derived pathogenic mediator integrating viral and cytokine-driven injury pathways. Targeting CTSS may therefore offer a promising therapeutic strategy to mitigate immunopathology. Indeed, CTSS has been proposed as a biomarker and therapeutic target in diverse pulmonary diseases (45), and preclin ical studies in cystic fibrosis and COPD models show that CTSS inhibition alleviates mucus hypersecretion, inflammation, and functional decline (46)(47)(48). These observations support the translational potential of CTSS-targeted interventions for influenza-induced lung injury.
Nevertheless, while in vitro models such as A549 cells and air-liquid interface (ALI) differentiated airway epithelial cultures provide valuable insight, they cannot fully recapitulate the complex lung microenvironment, including immune cell crosstalk and extracellular matrix interactions. Future investigations addressing these limitations will be critical to validate CTSS as a viable therapeutic target and to deepen our understand ing of its role in respiratory viral pathogenesis.
In summary, our findings identify CTSS as a pivotal mediator linking influenza virus infection to excessive inflammation, apoptosis, and epithelial barrier dysfunction. Targeting CTSS holds promise as a host-directed therapeutic approach to improve clinical outcomes in severe influenza, warranting further investigation and development.
## References
1. Krammer, Smith, Fouchier et al. (2018) *Influenza. Nat Rev Dis Primers*
2. Macias, Mcelhaney, Chaves et al. (2021) "The disease burden of influenza beyond respiratory illness" *Vaccine (Auckland)*
3. Iuliano, Roguski, Chang et al. (2018) "Estimates of global seasonal influenza-associated respiratory mortality: a modelling study" *Lancet*
4. Guan, Qu, Shen et al. (2024) "Baloxavir marboxil use for critical human infection of avian influenza A H5N6 virus" *Med*
5. Nicholson, Wood, Zambon (2003) *Influenza. Lancet*
6. Sarda, Palma, Rello (2019) "Severe influenza: overview in critically ill patients" *Curr Opin Crit Care*
7. Kamidani, Garg, Rolfes et al. (2022) "Epidemiology, clinical characteristics, and outcomes of influenza-associated hospitalizations in US children over 9 seasons following the 2009 H1N1 pandemic"
8. Hook, Bhattacharya (2024) "The pathogenesis of influenza in intact alveoli: virion endocytosis and its effects on the lung's air-blood barrier" *Front Immunol*
9. Rothan, Mostafa, Bayoumi et al. (2025) "Emerging highly pathogenic H5N1 influenza triggers fibrotic remodeling in human airway organoids" *Emerg Microbes Infect*
10. Boyd, Allen, Randolph et al. (2020) "Exuberant fibroblast activity compromises lung function via ADAMTS4" *Nature*
11. Shen, Sigal, Boes et al. (2004) "Important role of cathepsin S in generating peptides for TAP-independent MHC class I crosspresentation in vivo" *Immunity*
12. Gao, Zhang, Deng et al. (2025) "Cathepsin S: molecular mechanisms in inflammatory and immunological processes" *Front Immunol*
13. Geetha, Skaria (2025) "Cathepsin S: a key drug target and signalling hub in immune system diseases" *Int Immunopharmacol*
14. Wei, Shao, Peng et al. (2021) "Inhibition of cathepsin S restores TGF-β-induced epithelial-to-mesenchymal transition and tight junction turnover in glioblastoma cells" *J Cancer*
15. Smyth, Sasiwachirangkul, Williams et al. (2022) "Cathepsin S (CTSS) activity in health and disease -a treasure trove of untapped clinical potential" *Mol Aspects Med*
16. Taggart, Greene, Smith et al. (2003) "Inactivation of human beta-defensins 2 and 3 by elastolytic cathepsins" *J Immunol*
17. Scarcella, Angelo, Ciampa et al. (2022) "The key role of lysosomal protease cathepsins in viral infections" *Int J Mol Sci*
18. Chen, Zhou, Bai et al. (2025) "Macrophage-derived CTSS drives the agedependent disruption of the blood-CSF barrier" *Neuron*
19. Zhang, Li, Lu et al. (2022) "Imbalanced IL-37/TNF-α/CTSS signaling disrupts corneal epithelial barrier in a dry eye model in vitro" *Ocul Surf*
20. Pan, Dong, Li et al. (2013) "Visualizing influenza virus infection in living mice" *Nat Commun*
21. Wang, Lv, Niu et al. (2020) "L226Q mutation on influenza H7N9 virus hemagglutinin increases receptor-binding avidity and leads to biased antigenicity evaluation" *J Virol*
22. Malik, Steele, Long et al. (2025) "Trans-epithelial/ endothelial electrical resistance (TEER): current state of integrated TEER measurements in organ-on-a-chip devices" *Curr Opinion Biomed Eng*
23. Chiu, Li, Liu et al. (2022) "Human nasal organoids model SARS-CoV-2 upper respiratory infection and recapitulate the differential infectivity of emerging variants" *mBio*
24. Woodall, Cujba, Worlock et al. (2024) "Agespecific nasal epithelial responses to SARS-CoV-2 infection" *Nat Microbiol*
25. Ju, Yan, Liu et al. (2015) "Neuraminidase of influenza A virus binds lysosome-associated membrane proteins directly and induces lysosome rupture" *J Virol*
26. Bonnans, Chou, Werb (2014) "Remodelling the extracellular matrix in development and disease" *Nat Rev Mol Cell Biol*
27. Hynes, Naba (2012) "Overview of the matrisome--an inventory of extracellular matrix constituents and functions" *Cold Spring Harb Perspect Biol*
28. Edinger, Pohl, Yángüez et al. (2015) "Cathepsin W is required for escape of influenza A virus from late endosomes" *mBio*
29. Storm Van's Gravesande, Layne, Ye et al. (2002) "IFN regulatory factor-1 regulates IFN-gamma-dependent cathepsin S expression" *J Immunol*
30. Kaletsky, Simmons, Bates (2007) "Proteolysis of the Ebola virus glycoproteins enhances virus binding and infectivity" *J Virol*
31. Schornberg, Matsuyama, Kabsch et al. (2006) "Role of endosomal cathepsins in entry mediated by the Ebola virus glycoprotein" *J Virol*
32. Dabydeen, Meneses (2009) "The role of NH4Cl and cysteine proteases in human papillomavirus type 16 infection" *Virol J*
33. Golden, Bahe, Lucas et al. (2004) "Cathepsin S supports acid-independent infection by some reoviruses" *J Biol Chem*
34. Ebert, Deussing, Peters et al. (2002) "Cathepsin L and cathepsin B mediate reovirus disassembly in murine fibroblast cells" *J Biol Chem*
35. Jackson, Farzan, Chen et al. (2022) "Mechanisms of SARS-CoV-2 entry into cells" *Nat Rev Mol Cell Biol*
36. Zhao, Yang, Yang et al. (2021) "Cathepsin L plays a key role in SARS-CoV-2 infection in humans and humanized mice and is a promising target for new drug development" *Signal Transduct Target Ther*
37. Mahapatra, Mishra, Behera et al. (2021) "The lysosome as an imperative regulator of autophagy and cell death" *Cell Mol Life Sci*
38. Chen, Zhu, Liao et al. (2024) "The HN protein of Newcastle disease virus induces cell apoptosis through the induction of lysosomal membrane permeabilization" *PLoS Pathog*
39. Fung, Yuen, Ye et al. (2020) "A tug-of-war between severe acute respiratory syndrome coronavirus 2 and host antiviral defence: lessons from other pathogenic viruses" *Emerg Microbes Infect*
40. Xiong, Liu, Cao et al. (2020) "Transcriptomic characteristics of bronchoalveolar lavage fluid and peripheral blood mononuclear cells in COVID-19 patients" *Emerg Microbes Infect*
41. 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*
42. Zhou, Li, Sachs et al. (2018) "Differentiated human airway organoids to assess infectivity of emerging influenza virus" *Proc Natl Acad Sci*
43. Wang, Gao, Ma et al. (2025) "N460S in PB2 and I163T in nucleoprotein synergistically enhance the viral replication and pathogenicity of influenza B virus" *PLoS Pathog*
44. Liu, Zhou, Yang (2016) "The cytokine storm of severe influenza and development of immunomodulatory therapy" *Cell Mol Immunol*
45. Brown, Nath, Samaha et al. (2020) "Cathepsin S: investigating an old player in lung disease pathogenesis, comorbidities, and potential therapeutics" *Respir Res*
46. Small, Brown, Doherty et al. (2019) "Targeting of cathepsin S reduces cystic fibrosis-like lung disease" *Eur Respir J*
47. Cimerman, Brguljan, Krasovec et al. (2001) "Circadian and concentration profile of cathepsin S in sera from healthy subjects and asthmatic patients" *Pflugers Arch*
48. Andrault, Schamberger, Chazeirat et al. (2019) "Cigarette smoke induces overexpression of active human cathepsin S in lungs from current smokers with or without COPD" *Am J Physiol Lung Cell Mol Physiol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12777813&blobtype=pdf | # Association of inflammatory scoring tools with spirometry indices in COVID-19 patients: a single center cohort study
Mohsen Farrokhpour, Fahimeh Safarnezhad Tameshkel, Niloufar Kalaki, Azra Asghari Marzidareh, Aliarash Anoushirvani, Neda Rahimian, Mohammad Karbalaie Niya, Mohammad Hadi, Karbalaie Niya
## Abstract
Background and Objectives: Patients with COVID-19 have spirometry parameters linked to various biological markers, including CRP, NLR, MPV, RDW, and APACHE II score. The objective of this study was to investigate the association of inflammatory scoring tools with spirometry indices in a three-month follow-up of COVID-19 patients. Materials and Methods: Spirometry records of 369 COVID-19 cases with complications were analyzed at baseline and three months after discharge. Generalized linear models and logistic regression analysis were performed to compare the variables using SPSS version 25 software. Results: The baseline NLR was 3.20 (95% CI: 2.96, 3.43); PCT was 0.26 (95% CI: 0.25, 0.27); and MPV was 7.23 (95% CI: 7.10, 7.35). We found that the effects of NLR, CRP, and APACHE II score on the respiratory indices FEV1 and FEV1/ FVC three months after discharge had an inverse relationship. Patients with asthma had significantly lower FEV1 and FEV1/ FVC values, and the level of FVC did not have any significant difference between people with asthma and COPD patients. Conclusion: CRP, NLR, and APACHE II score are among the main factors that are directly related to respiratory indices and they are considered to be appropriate indicators of prognosis for these conditions in COVID-19 patients.
## INTRODUCTION
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), as a coronavirus that appeared in China in late 2019 and quickly spread across the globe, has posed a serious threat to global health, http://ijm.tums.ac.ir claiming the lives of millions of people worldwide (1). Coronavirus disease of 2019 (COVID- 19), named for SARS-CoV-2 infection, primarily affects the lungs, but it can also harm other organs (2). Organ damage can lead to long-term health risks. Predicting the morbidity and mortality rates of COVID-19 patients has increasingly posed a challenge for healthcare systems globally since the onset of the pandemic. Research reveals the potential usefulness of certain analytical markers, proteins, and blood parameters (3). The alterations in the aforementioned factors offer diagnostic insights regarding the existence of a COVID-19 infection within the body. They also provide vital information for predicting complications and developing treatment plans for managing COVID-19 infection (4).
The ratio of lymphocytes to neutrophils (NLR) is a biological factor that changes in various conditions such as tumors, pancreatic cancer, chronic obstructive pulmonary disease (COPD), and cardiovascular diseases. Higher NLR values at intensive care unit (ICU) admission are linked to worse clinical symptoms in COVID-19 involved individuals (5). Additionally, the mean platelet volume (MPV) is a measurable complete blood count (CBC) parameter in which recent studies represent a connection between the severity of COVID-19 and the presence of larger and younger platelets, making this observation a powerful indicator for evaluating platelet activity (6). Furthermore, high levels of C-reactive protein (CRP) can be found in patients with severe pneumonia. Therefore, CRP is not only a key parameter for diagnosing and assessing severe lung infections, but also a valuable prognostic marker. High CRP levels indicate a higher risk of disease worsening and a longer hospital stay (7). Lactate dehydrogenase (LDH) is another important biomarker, as initial data from COVID-19 patients have shown a significant difference in LDH levels between those with severe disease and those without severe symptoms. These results suggest that high LDH levels are associated with a higher chance of severe disease and COVID-19 mortality (8). Additionally, elevated procalcitonin (PCT) levels have been shown to correlate positively with COVID-19 severity (9). Moreover, the acute physiology and chronic health evaluation (APACHE) II score is a valid and practical scoring system used to evaluate and differentiate patients with a high risk of severe medical conditions. It allows the classification of acute patients based on their physiological parameters, enabling the selection of optimal treatment strategies. This not only gives patients hope for effective future treatment but also highlights the importance of starting the recovery process as soon as possible (10).
The impact of SARS-CoV-2 on global health necessitates further investigation of inflammatory markers and lung function. Comprehending these indicators is essential for enhancing patient outcomes and directing effective treatment strategies. In this study, we demonstrate that different biological markers, such as CRP, NLR, MPV, red cell distribution width (RDW), and APACHE II score, are associated with spirometry parameters in a follow-up for three months in patients with COVID-19. We hope that by finding a significant association, we will be able to apply the appropriate treatment strategies in the areas of prognosis and diagnosis to cope more effectively with COVID-19.
## MATERIALS AND METHODS
Patients and study design. In this prospective cohort study, we included 369 COVID-19 patients admitted to Firoozgar Hospital, Tehran, Iran. These patients were confirmed cases, verified by positive RT-PCR tests and supported by positive CT scan results. During hospitalization and the subsequent three-month follow-up, a thorough assessment, including spirometry and blood tests, was performed. In addition, vital demographic and clinical data such as age, clinical symptoms, baseline laboratory indicators, clinical measurements, and underlying medical conditions were carefully recorded. Importantly, we deliberately excluded individuals with blood malignancies from the study group, ensuring the homogeneity and relevance of our patient sample. A checklist was filled out for each participant at the baseline and follow-up step. The checklist data included all demographic, clinical and paraclinical data of each participant, which was used for further analysis. We obtained data from each patient's records at the hospital repository and used it for the analysis.
Measurements. Data were collected from the hospital repository at the patients' baseline admission and at their three-month follow-up after discharge. In this study, patients who met the inclusion criteria first had their baseline spirometry indexes and disease complications recorded, and were then contacted after three months for their follow-up. Their complications and spirometry indexes were assessed again and used for further analysis. The primary objective of this research is to examine the correlation between spirometry scores and a range of inflammatory markers, which include NLR, PCT, MPV, CRP, LDH, ferritin, in addition to the APACHE II score. We used Spearman's correlation to evaluate associations between inflammatory markers and pulmonary function tests. Additionally, we used generalized linear models to evaluate the combined effect of inflammatory markers and the APACHE II score on respiratory indices (FEV1/FVC, FVC, and FEV1) three months after patient discharge.
The forced vital capacity (FVC) is a crucial test of lung function. It measures how much air the patient can breathe out forcefully after taking a deep breath. The FEV1 (Forced expiratory volume) is the amount of air that comes out in the first second of the FVC test. The FEV1/FVC ratio shows what percentage of the FVC is achieved in one second. The actual ratio, not the predicted one, is used to interpret the results (11)(12)(13). Interpretation of spirometry parameters was done according to guidelines (14), and the correlation coefficient of their scores with inflammatory markers was calculated.
## Statistical analysis.
We performed statistical analysis on the collected data using generalized linear models to enable comparative evaluations. Moreover, we used logistic regression analysis to measure the impact of the dataset on the final outcome. We used SPSS version 25 software for data analysis, ensuring rigorous and systematic analysis of the obtained information. P-values < 0.05 were considered as statistically significant.
Ethics Statement. The study protocol was approved by the ethics committee of the Iran University of Medical Sciences, located in Tehran, Iran, under the code: IR.IUMS.FMD.REC.1400.278.
## RESULTS
## Patients demographics.
A comprehensive investigation analyzed data from 369 hospitalized patients diagnosed with COVID-19. Of them, 238 cases af-ter three-month follow-up met the inclusion criteria and had sufficient data for analysis. The study cohort (n=238) had an average age of 56.11 years (95% CI: 54.35, 57.87), with 55% of patients being male (n=130). The average body mass index (BMI) was 26.33 kg/m² (95% CI: 25.81, 26.85). Among the patients, 76% (n=180) were nonsmokers, while 24% (n=58) were identified as smokers. Notably, 26% of participants had comorbidities, with hypertension and ischemic heart disease/congestive heart failure (IHD/ CHF) each accounting for 14% (n=33), and chronic obstructive pulmonary disease (COPD) at 12% (n=28). Baseline mean values for key inflammatory markers were as follows: NLR (neutrophil-to-lymphocyte ratio), 3.20 (95% CI: 2.96, 3.43); PCT (procalcitonin), 0.26 (95% CI: 0.25, 0.27); and MPV (mean platelet volume), 7.23 (95% CI: 7.10, 7.35). Refer to (Table 1) for a comprehensive summary of baseline values for other relevant laboratory indicators and clinical measurements in the patient cohort. 2 shows the relationship between baseline inflammatory factors, the APACHE II score, and respiratory indices (FEV1/FVC, FVC, and FEV1) three months post-hospital discharge. Moreover, it shows that the FEV1 index has a negative linear relationship with the baseline levels of NLR, CRP, and the APACHE II score. This means that lower FEV1 values, indicating more severe obstructive disease, are related to higher levels of NLR, CRP, and the APACHE II score three months later. Also, there was a significant negative linear relationship between the baseline CRP levels and the FVC index three months after discharge (P = 0.044). This implies that higher baseline CRP levels, indicating more severe restrictive disease, are related to lower FVC values at the same time. On the other hand, PCT and MPV had a positive correlation, meaning that higher levels of these factors were related to higher FVC values. Moreover, patients with higher baseline levels of NLR, CRP, and the APACHE II score had a negative linear relationship with the FEV1/FVC index. This suggests that lower FEV1/FVC values, indicating more severe obstructive disease, are related to higher levels of NLR, CRP, and the APACHE II score three months later (Table 2).
## Spirometry parameters analysis. Table
We used a generalized linear model to examine how NLR, CRP, APACHE II score, and a history of respiratory diseases affected FEV1, FVC, and FEV1/FVC 3). Only a history of asthma had a significant impact on FEV1 levels. Patients with an asthma history had an average FEV1 level 29 units lower than those without such a history. FVC levels had significant relationships with several factors; patients with a history of ILD had an average FVC level 14.97 units lower than those without such a history (P<0.001). Patients with a history of asthma had an average FVC level 5 units higher than those without such a history (P=0.03). Also, for every unit increase in CRP, the average FVC level three months after discharge decreased by 0.05 units (P=0.021). In contrast, for every unit increase in Procalcitonin (PCT) or Mean Platelet Volume (MPV), the average FVC level three months after discharge increased by 16.98 units (P=0.039) and 1.75 units (P=0.002), respectively.
There was no significant relationship between FEV1/ FVC levels and asthma three months after discharge in patients with a history of asthma. Therefore, the average FEV1/FVC level at this time in patients with an asthma history was only 0.21 units lower than in patients without such a history (Table 3).
## DISCUSSION
While most COVID-19 patients recover and resume their normal health after treatment, some may experience symptoms that last for weeks or months. People with mild illness who do not need hospitalization can also have persistent or late symptoms (15). Furthermore, more than 75% of COVID-19 patients reported at least one long-term effect six months after the onset of the disease (16). COVID-19 is mainly characterized by pneumonia, but its most serious and fatal outcomes are related to cardiovascular complications. The nervous system is also severely impacted by COVID-19. A study showed that about 40% of COVID-19 patients had neurological symptoms due to the infection (17). A study found that the combination of the APACHE II score, NLR, and expired tidal volume was more effective in predicting the failure of Non-Invasive Ventilation (NIV) than the APACHE II score, NLR, or expired tidal volume alone (18). Moreover, ferritin and LDH levels increased significantly in the moderate and severe stages of the disease (19). The APACHE II score has also been shown to be independently associated with inpatient mortality (20). We examined the association between baseline inflammatory markers and the APACHE II score and post-discharge respiratory measures, namely FEV1/ FVC, FVC, and FEV1, three months after discharge. We found a negative relationship between the levels of NLR, CRP, and the APACHE II score and the FEV1 measure, meaning that lower FEV1 values were related to higher levels of NLR, CRP, or APACHE II score. On the other hand, the baseline CRP level showed a negative linear relationship with the FVC measure, while a positive relationship was seen for PCT and MPV. Higher CRP, NLR, and APACHE II scores were associated with lower FVC values, indicating that systemic inflammation may contribute to reduced pulmonary capacity. Moreover, the baseline levels of NLR, CRP, and the APACHE II score had a negative relationship with the FEV1/ FVC measure. A higher FEV1/FVC measure was related to lower levels of NLR, CRP, and the APACHE II score. Additionally, we explored the combined effects of NLR, CRP, the APACHE II score, and a history of respiratory disease on FEV1, FVC, and FEV1/FVC values three months after discharge. We found that only asthma significantly affected FEV1 and FEV1/FVC values. The FVC value was significant for all factors in the model except for COPD, which had no significant effect. These results agree with other previous studies. A study showed a moderate connection between the ICU stay length and the APACHE II scores, especially for FVC, FEV1, and changes in peripheral oxygen saturation (SpO2) values in COVID-19 patients (21). Another study showed that the main abnormality observed was a reduction in DLCO (diffusing capacity of the lungs for carbon monoxide), followed by lower FEV1 and FVC (22). Sun et al.'s study (2021) showed significant positive relationships between NLR and white blood cell counts, NLR and CRP levels, and NLR and PCT levels, as shown by the statistical analysis (23). Concerning MPV, a recent study has indicated its potential in offering valuable insights for the diagnosis and prognosis of sepsis and COVID-19. Additionally, MPV has been proposed as an inflammatory marker for serious conditions such as pneumonia and coronary heart diseases (24). Another study showed a correlation between standard platelet measures (PLT, MPV, PCT, PDW) and CRP and different symptoms of depression (25). As Yuksel et al. found in their study, high levels of MPV, NLR, and MPV-NLR can be used as simple markers for predicting mortality and mental decline in COVID-19 (26). Our study results support the association between inflammatory markers, the APACHE II score, and post-discharge respiratory measures as reported by another research. This suggests that these markers could be useful in predicting the severity and outcomes of COVID-19 in hospitalized patients.
This study has some limitations, mainly its prospective design. Also, the number of patients with COVID-19-related complications was relatively small. Future studies ought to focus on incorporating a broader and more varied patient demographic to enable a more reliable and accurate statistical analysis. It is important to recognize that this study did not consider various other inflammatory, hematological, and clinical variables. Therefore, there is an urgent need for more studies that examine the effect of these additional factors on the occurrence of complications from COVID-19. Such comprehensive studies are crucial for improving our knowledge of the disease and its management.
In conclusion, this thorough study has offered useful insights into the connection between laboratory markers and clinical outcomes. The main goal of this study was to investigate the relationship among factors affecting COVID-19-related complications. The results showed several significant associations between different biomarkers and health outcomes. When evaluating the long-term impact on respiratory function, our study showed a negative relationship between NLR, CRP, APACHE II score, and post-discharge FEV1 and FEV1/FVC values, especially in asthmatic patients. This indicates the lasting effect of inflammatory factors and the APACHE II score on respiratory health in these individuals. However, the level of FVC did not have a significant impact on patients with COPD. Our study provides essential data for clinicians and researchers, revealing the complex relationship between various laboratory markers and clinical outcomes in COVID-19 patients. These findings have the potential to improve risk assessment, patient management, and treatment strategies in the ongoing fight against this pandemic.
## References
1. Al-Rohaimi, Otaibi (2020) "Novel SARS-CoV-2 outbreak and COVID19 disease; a systemic review on the global pandemic" *Genes Dis*
2. Shang, Wang, Yuan et al. (2022) "Global Excess Mortality during COVID-19 Pandemic: A Systematic Review and Meta-Analysis" *Vaccines (Basel)*
3. Tian, Jiang, Yao et al. (2020) "Predictors of mortality in hospitalized COVID-19 patients: A systematic review and meta-analysis" *J Med Virol*
4. Covid-Icu (2021) "Group on behalf of the REVA Network and the COVID-ICU Investigators.Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study" *Intensive Care Med*
5. Aly, Meshref, Abdelhameid et al. (2021) "Can hematological ratios predict outcome of COVID-19 patients? a multicentric study" *J Blood Med*
6. Denorme, Ajanel, Campbell (2022) "Shining a light on platelet activation in COVID-19" *J Thromb Haemost*
7. Aydınyılmaz, Aksakal, Pamukcu et al. (2021) "Significance of MPV, RDW and PDW with the Severity and Mortality of COVID-19 and Effects of Acetylsalicylic Acid Use" *Clin Appl Thromb Hemost*
8. Henry, Aggarwal, Wong et al. (2020) "Lactate dehydrogenase levels predict coronavirus disease 2019 (COVID-19) severity and mortality: A pooled analysis" *Am J Emerg Med*
9. Aon, Alsaeedi, Alzafiri et al. (2022) "The association between admission procalcitonin level and the Severity of COVID-19 Pneumonia: A Retrospective Cohort study"
10. Beigmohammadi, Amoozadeh, Motlagh et al. (2022) "Mortality predictive value of APACHE II and SOFA scores in COVID-19 patients in the intensive care unit" *Can Respir J*
11. Barreiro, Perillo (2004) "An approach to interpreting spirometry" *Am Fam Physician*
12. Langan, Goodbred (2020) "Office spirometry: indications and interpretation" *Am Fam Physician*
13. Burrill, Mcardle, Davies (2021) "Lung function in children: a simple guide to performing and interpreting spirometry" *J Paediatr Child Health*
14. Patil, Patil, Gondhali (2023) "Pulmonary functions assessment in post-COVID-19 pneumonia cases by spirometry: Study of 600 cases in tertiary care setting in India" *J Appl Sci Clin Pract*
15. Brojakowska, Eskandari, Bisserier et al. (2021) "Comorbidities, sequelae, blood biomarkers and their associated clinical outcomes in the Mount Sinai Health System COVID-19 patients" *PLoS One*
16. Huang, Huang, Wang et al. (2023) "6-month consequences of COVID-19 in patients discharged from hospital: a cohort study" *Lancet*
17. Wu, Li, Chen et al. (2020) "Inhibition of Sema4D/PlexinB1 signaling alleviates vascular dysfunction in diabetic retinopathy" *EMBO Mol Med*
18. Sun, Luo, Cao et al. (2022) "A combination of the APACHE II score, neutrophil/lymphocyte ratio, and expired tidal volume could predict non-invasive ventilation failure in pneumonia-induced mild to moderate acute respiratory distress syndrome patients" *Ann Transl Med*
19. Bakry, Sayed (2021) "Chest CT manifestations with emphasis on the role of CT scoring and serum ferritin/lactate dehydrogenase in prognosis of coronavirus disease 2019 (COVID-19)" *Egypt J Radiol Nucl Med*
20. Zou, Li, Fang et al. (2020) "Acute physiology and chronic health evaluation II score as a predictor of hospital mortality in patients of coronavirus disease 2019" *Crit Care Med*
21. Sirayder, Inal-Ince, Kepenek-Varol et al. (2022) "Long-Term Characteristics of Severe COVID-19: Respiratory Function, Functional Capacity, and Quality of Life" *Int J Environ Res Public Health*
22. Sibila, Albacar, Perea et al. (2021) "Lung function sequelae in COVID-19 Patients 3 months after hospital discharge" *Arch Bronconeumol*
23. Sun, Luo, Cao et al. (2021) "The Neutrophil/ Lymphocyte ratio could predict noninvasive mechanical ventilation failure in patients with acute exacerbation of chronic obstructive pulmonary disease: A Retrospective observational study" *Int J Chron Obstruct Pulmon Dis*
24. Hlapčić, Somborac-Bačura, Popović-Grle et al. (2020) "Platelet indices in stable chronic obstructive pulmonary disease -association with inflammatory markers, comorbidities and therapy" *Biochem Med (Zagreb)*
25. Wang, Yang, Wu et al. (2022) "Platelet Parameters, C-Reactive Protein, and Depression: An Association Study" *Int J Gen Med*
26. Yuksel, Dirik, Gursoy et al. (2021) "A simple scoring system in COVID-19 patients with neurological manifestations" *Neurol Asia* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12281961&blobtype=pdf | # Correction: EGHRIN conclusions on pandemic preparedness: no whole-of-society approach without society
Marie Stockman, Antonio Plasència, Heidi Larson, Leesa Lin, Ana Antic, Janharmen Drost, Guenter Froeschl, Jolene Skordis, Anne-Mieke Vandamme, Stockman Marie, Plasència Antonio, Larson Heidi, Lin Leesa, Antic Ana, Drost Janharmen, Froeschl Guenter |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12540899&blobtype=pdf | # Correction: The Activation of G Protein-Coupled Receptor 30 (GPR30) Inhibits Proliferation of Estrogen Receptor Negative Breast Cancer Cells in vitro and in vivo
W Wei, Z-J Chen, K-S Zhang, X-L Yang, Y-M Wu, X-H Chen, H-B Huang, H-L Liu, S-H Cai, J Du, H-S Wang
Our internal review identified one duplication within our manuscript, specifically where the p-p38 band in Figure 5a overlaps with the cyclin E band in Figure 5d. This issue likely arose during the figure assembly process due to a copy-paste error, particularly when using the template of Figure 5d to prepare Figure 5a, resulting in the cyclin E band from Figure 5d overlaying the p-p38 band in Figure 5a. We have thoroughly reviewed our original data. However, due to the significant time elapsed since the experiment and the graduation of the involved students, we were able to locate the original data for the cyclin E band in Figure 5d, but not the original p-p38 band data for Figure 5a.
To rigorously address this, we repeated the experiments related to Figure 5a three times. Antibodies used: p-p38 (Santa Cruz, sc-166182), p38 (Santa Cruz, sc-81621), GAPDH (CST, 5174S). Experiments were repeated under identical conditions as the original study (SkBr3 cells were treated with 1 μM G-1 for the indicated time periods, and then the phosphorylation and total protein levels of p-38 were detected by Western-blotting)."
The new results (see attached raw data) consistently demonstrate that G-1 does not affect p-p38 levels, which aligns with the original conclusion. Nonetheless, to uphold the highest standards of data accuracy, we propose the following correction: |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12832576&blobtype=pdf | # Metagenomic screening of the virome of symptomatic tomato plants from La Réunion Island uncovers a complex of viruses including a newly identified whitefly-transmitted polerovirus
Jean-Michel Lett, Sarah Scussel, Sélim Chéhida, Murielle Hoareau, Denis Filloux, Emmanuel Fernandez, Philippe Roumagnac, Evelyne Parvedy, Elise Quirin, Clarisse Clain, Janice Minatchy, Estelle Roux, Pierre-Yves Teycheney, Pierre Lefeuvre
## Abstract
Using unbiased high-throughput sequencing for metagenomic screening of viruses in diseased tomato plants, we identified a viral complex that includes viruses previously reported in tomato crops on La Réunion Island as well as a novel polerovirus, tentatively named "tomato necrotic yellowing virus" (ToNYV, proposed species, "Polerovirus ToNYV"). Molecular characterization and phylogenetic analysis revealed that ToNYV is closely related to two recently described poleroviruses from Africa and the Middle East, one of which is transmitted by the whitefly Bemisia tabaci, a trait uncommon among poleroviruses. Our transmission experiments demonstrated that ToNYV is also transmitted by B. tabaci and is prevalent across major tomato-growing regions of La Réunion. These findings highlight the value of metagenomic virome analysis in diseased plants for identifying novel viruses potentially involved in emerging plant diseases, either individually or as components of viral complexes.
Recent advances in high-throughput sequencing (HTS) technologies and bioinformatics have transformed the field of plant virology, particularly in the detection and diagnosis of viral pathogens of plants [1]. Viral metagenomics, referred to as viromics, enables comprehensive profiling of viral communities in plant samples by detecting both known and novel viral sequences. Viromics has matured significantly, with various enrichment strategies now available to enhance the recovery of viral nucleic acids. These include total RNA sequencing with ribosomal RNA (rRNA) depletion [2], extraction of virion-associated nucleic acids (VANAs) from viral particles [3], isolation of double-stranded RNA (dsRNA) [4], and sequencing of virus-derived small interfering RNAs (siRNAs) [5]. As this methodology has gained widespread acceptance by the scientific community, it has proven highly effective in identifying plant viruses across both wild and cultivated ecosystems [4][5][6][7]. Viromics is now increasingly regarded as the method of choice in complex epidemiological contexts, particularly when the etiological agents are unknown or poorly characterized.
One such epidemiological scenario involved viral diseases affecting tomato crops on La Réunion Island. Since the 1990s, these crops have faced a series of successive biological invasions involving various viruses and their insect vectors. Notable examples include tomato spotted wilt virus (TSWV, species Orthotospovirus tomatomaculae, genus Orthotospovirus, family Tospoviridae) and its thrips vector Frankliniella occidentalis [8]; potato virus Y (PVY, species Potyvirus yituberosi, genus Potyvirus, family Potyviridae) and its aphid vectors Myzus persicae, Aphis gossypii, and A. fabae (PRPV database, https://db.e-prpv. org/) [9]; and both the Mild (TYLCV-Mld [10]) and Israel (TYLCV-IL [11,12]) strains of tomato yellow leaf curl virus (TYLCV, species Begomovirus coheni, genus Begomovirus, family Geminiviridae) and their whitefly vector Bemisia tabaci MEAM1 (formerly known as biotype B) [13]. Additionally, tomato chlorosis virus (ToCV, species Crinivirus tomatichlorosis, genus Crinivirus, family Closteroviridae) [13], transmitted by both B. tabaci and Trialeurodes vaporariorum [13], has been reported, as well as southern tomato virus (STV, species Amalgavirus lycopersici, genus Amalgavirus, family Amalgaviridae) [14] for which no evidence of horizontal transmission has been reported [15].
In 2015, two tomato leaf samples displaying a mixture of symptoms characteristic of TYLCV and ToCV infections were collected on Reunion Island, one from a fieldgrown plant (C15-7) and one from a greenhouse-grown plant (C15-10). The viral diversity in these samples was investigated using a VANA-based metagenomic approach [3]. Libraries were prepared from purified amplicons and sequenced by Genewiz (USA) on an Illumina HiSeq device using a 2x250 bp configuration. Reads were demultiplexed using cutadapt 1.18 [16] with a minimum overlap of 10 nt and other parameters set to default. After quality control of demultiplexed reads using Trimmomatic v0.35 [17] (parameters, SLIDINGWINDOW:5:20 and MINLEN:80), a total of 59,182 and 92,526 quality-controlled reads were obtained for sample C15-7 and sample C15-11, respectively (Supplementary Table S1). Reads were then assembled de novo into contigs using SPAdes v3.13.0 [18]. BLASTn and BLASTx comparisons of the resulting contigs against viral sequences in the GenBank database revealed that nine contigs (>500 nt in length) from sample C15-7 and 16 from sample C15-11 were potentially derived from plant RNA or DNA viruses. These contigs showed similarity to genome sequences from viruses from several families, including Amalgaviridae, Closteroviridae, Geminiviridae, Potyviridae, and the former family Luteoviridae (Supplementary Table S1).
In plant C15-7, two contigs (Contig-1395, 1,935 nt, and Contig-1396, 1,418 nt) shared the highest nucleotide sequence similarity with TYLCV-IL, showing 99.1% and 98% identity to sequences AM409201 and FJ012359, respectively (Supplementary Table S2). Additionally, three contigs (Contig-1401, 1332, and 900, ranging from 1,077 to 3,066 nt) showed 90.5% to 91.5% nucleotide sequence identity to PVY (family Potyviridae, genus Potyvirus) sequences (AM409201 and FJ012359).
From plant C15-11, five contigs (Contig-5040, 5146, 3288, 5051, and 3286, ranging in size from 1,253 to 2,485 nt) displayed 98% to 99% nucleotide sequence identity with RNA1 of ToCV (family Closteroviridae, genus Closterovirus) sequences (KJ740256 and KY471129). Another five contigs (Contig-5139, 5035, 3282, 5031, and 5145, ranging in size from 2,224 to 3,788 nt) shared 96.7% to 97.5% identity with ToCV RNA2 sequences (KP137101, KY810787, and KY471130). Two additional contigs (Contig-5036, 968 nt, and Contig-3280, 536 nt) showed 98% and 99.8% identity, respectively, to a southern tomato virus (STV, family Amalgaviridae, genus Amalgavirus) sequence (MN216389).
The nucleotide and amino acid sequence identity of these contigs to reference sequences in the GenBank database exceed the species demarcation thresholds recommended for their respective genera: 65-70% amino acid sequence identity in the RNA-dependent RNA polymerase (RdRp) for the genus Amalgavirus [19], h t t p s : / / i c t v . g l o b a l / r e p o r t / c h a p t e r / a m a l g a v i r i d a e / a m a l g a v i r i d a e; 75% amino acid sequence identity in the RdRp protein, heat shock protein 70 homologue (HSP70h), and coat protein (CP) for the genus Closterovirus [20]; 91% nucleotide sequence identity in the complete DNA-A component for the genus Begomovirus [21]; and 80% amino acid sequence identity in the CP for the genus Potyvirus [22]. These results indicate that the viral isolates from which these partial genomic sequences were derived are members of the established virus species Amalgavirus lycopersici (STV), Crinivirus tomatichlorosis (ToCV), Begomovirus coheni (TYLCV), and Potyvirus yituberosi (PVY).
In contrast, analysis of plant C15-7 revealed four contigs (Contig-894, 1392, 1390, and 897) ranging in size from 1,755 to 4,218 nt, that shared the highest nucleotide sequence identity (82.3% to 86.6%) with sequences of African eggplant yellowing virus (AeYV; accession no. KX856972), a member of the genus Polerovirus (family Solemoviridae) (Supplementary Table S2). Similarly, from plant C15-11, four contigs (Contig-5034, 5142, 5140 and 3283) ranging in size from 1,990 to 4,703 nt, exhibited 75.5 to 80.8% nucleotide sequence identity to another AeYV sequence (accession no. KX856971). Within each sample, the AeYV-related contigs exhibited large overlaps with one another, with minimum pairwise identity values of 97.5% and 99.1% for contigs obtained from plant C15-7 and plant C15-11, respectively.
To obtain the complete genome sequence of the AeYVrelated virus, a tomato sample (C20- 19) showing symptoms of necrosis, yellowing, and reddening of the leaves (Fig. 1) that tested positive for the AeYV-related virus and negative for ToCV and PVY by RT-PCR, and negative for TYLCV by 1 3 PCR, was subjected to Illumina RNA sequencing (2x150-bp paired-end reads on an Illumina HiSeq platform) following ribosomal RNA depletion (GeneWiz, Leipzig, Germany). Quality control of the raw reads using Trimmomatic v0.35 [17] (parameters, SLIDINGWINDOW:5:20 and MIN-LEN:80) resulted in ~11.4 million paired-end reads. Among the contigs obtained after de novo assembly using MEGA-HIT v1.2.9 [23], a single viral contig of 5,953 nt with an average coverage of 5004-fold was identified through searches against the NCBI gbvrl viral database using DIA-MOND 0.9.22 with an e-value cutoff of <10 -5 [24]. No other viral sequences were detected in sample C20-19.
The low-coverage region corresponding to the readthrough domain between ORF3 and ORF5 was confirmed by Sanger sequencing of RT-PCR amplicons obtained using primers AeYV-Like-F3988 and AeYV-Like-R4795 (Supplementary Table S3). The sequence of the 5'-terminal region was verified using both (1) the MinION sequencing strategy with the strand-switching method described by Filloux et al. [25] and (2) the rapid amplification of cDNA ends strategy with viral sequence-specific reverse primers (Supplementary Table S3) followed by direct Sanger sequencing in both directions (Macrogen Europe, The Netherlands), as described in Orfanidou et al. [26]. Likewise, the sequence of the 3'-terminal region was verified using MinION sequencing after poly(A)-tailing, following the method of Wongsurawat et al. [27].
Integration of Illumina and MinION data resulted in a final consensus sequence of 5,995 nt for the RE-C20-19-20 isolate of the AeYV-related virus, which was deposited in the GenBank database under the accession number PV289033. BLASTn analysis revealed that the AeYV-related virus genome sequence shared the highest nucleotide sequence similarity (86.9% identity with 88% query coverage) with the eMA4 isolate of AeYV (KX856972). The AeYV-related virus genome exhibits the typical organization of poleroviruses, containing six open reading frames (ORFs), as predicted using the ORFfinder tool available on the NCBI website (Fig. 2A). The 5' and 3' termini of the genome begin with the nucleotides ACAAA and end with GT, respectively. A predicted -1 ribosomal frameshift for the P1-P2 fusion protein was identified at position 1,682, involving a conserved "slippery heptamer" sequence (GGGAAAC-GGGAAA). The conserved sequence flanking the leaky stop codon separating the coat protein (P3, ORF3) from the readthrough domain (RTD) (P5, ORF5) (CCCAAATAGG-TAGA, with the stop codon in bold) was also present.
To clarify the phylogenetic relationship between the AeYV-related virus and other poleroviruses, a maximumlikelihood (ML) phylogenetic tree was constructed using MEGA X software [28], based on a multiple sequence alignment performed using MUSCLE and the automatic selection of the best-fit nucleotide substitution model (GTR+G). Pairwise nucleotide sequence identity comparisons were performed using SDT v1.2 with pairwise deletion of gaps [29]. The resulting ML phylogenetic tree and the pairwise nucleotide identity matrix (Fig. 3) revealed that the complete genome sequence of the RE-C20-19-20 isolate clustered closely with the eMA4 isolate of AeYV from Mali (83% nucleotide sequence identity). Together, AeYVrelated virus-RE-C20-19-20 and AeYV-eMA4 isolates form a distinct cluster positioned outside the pepper vein yellows virus (PeVYV) clade.
Amino acid (aa) sequence comparisons of proteins encoded in the genome of isolate RE-C20-19-20 with those of closely related poleroviruses (Fig. 2B) were performed using MEGA X [28]. Proteins encoded in the 3' region of the genome (P3, coat protein,P4, movement protein; and the P3-P5 readthrough fusion protein) shared the highest sequence similarity (85.3% to 93% identity) with their counterparts in AeYV and pepper whitefly-borne vein yellows virus (PeWBVYV). In contrast, proteins encoded in the 5' half of the genome (P0, RNA silencing suppressor; P1, serine protease; and P1-P2 fusion, RNA-dependant S4 1 3
To further investigate the distribution of ToNYV, a virus survey was conducted in 2018 and 2020 in tomato greenhouses located in the main tomato-producing regions of southern and south-eastern La Réunion (Supplementary Table S5). Leaf samples exhibiting virus-like symptoms, including leaf yellowing, reddening, necrosis, deformation, and stunting, were collected from multiple commercial cultivars. The presence of ToNYV genomic RNA was assessed using RT-PCR on total RNA extracted from symptomatic leaf tissues using an RNeasy Plant Mini Kit (QIAGEN, France), following the manufacturer's protocol. Amplicons were sequenced directly in both directions (Macrogen Europe, The Netherlands). RT-PCR was performed using the degenerate primers Polero-FD3249 and Polero-RD4339 (Supplementary Table S3), which were designed in this study to cover the ORF2 and ORF3 regions, encoding the polerovirus RNA-dependent RNA polymerase (RdRp) and the coat protein (CP), respectively. These regions were identified in contigs obtained through the VANA-based metagenomics approach, as well as sequences from emerging poleroviruses such as AeYV [35], PeVYV-1 to -6 [21,36], and PeWBVYV [33]. Reverse transcription was carried out using a RevertAid RT Reverse Transcription Kit (Thermo Fisher Scientific, France) with 1 µM downstream primer and 1 µM random hexamers, following the manufacturer's instructions. Briefly, 5 μl of extracted RNA was added to a reaction mixture consisting of 1 μl of random hexamer primers and 6 μl of diethylpyrocarbonate (DEPC)-treated water. This mixture was incubated at 70°C for 5 min for RNA denaturation and then placed on ice for an additional 5 min. A second reaction mix containing 4 μl of reaction buffer, 1 μl of Ribolock RNase Inhibitor, 2 μl of 10 mM dNTP mix, and 1 μl of reverse transcriptase (RevertAid H Minus M-MuLV Reverse Transcriptase) at a concentration of 200 U/μl was prepared and added to the first reaction mix. Negative controls were included in each set of RT reactions in order to detect potential contamination. PCR amplification was performed using a GoTaq G2 DNA Polymerase kit (Promega, France), following the manufacturer's instructions. Reactions were prepared in a final volume of 25 µL: 5 µL of GoTaq Flexi Buffer, 2.5 µL of 10 mM dNTP mix, 1.5 µL of 25 mM MgCl 2 , 1 µL of forward primer, 1 µL of reverse primer, 0.2 µL of GoTaq G2 Flexi DNA Polymerase (5 U/µL), 11.8 µL ultrapure water, and 2 µL of DNA sample. The cycling conditions used were 94 °C for 5 min, followed by 35 cycles of 94 °C for 1 min, 55 °C for 1 min, 72 °C for 30 s and a final extension at 72 °C for 5 min. The amplified PCR products were analysed on a 1% agarose gel to confirm the presence of amplicons of the expected size and sequenced in both directions by the Sanger method at Macrogen Europe (The Netherlands).
RNA polymerase) displayed lower sequence identity values, ranging from 70.3% to 85.6%, with the highest similarity observed with AeYV. The highest aa sequence similarity was found in the P3 protein, showing 93% and 92% identity to AeYV (KX856971) and PeWBVYV (MK333461), respectively (Fig. 2B). The lowest similarity (70.3% sequence identity) was observed for the P1 protein when compared to its counterpart in AeYV (KX856971). Except for the P3 protein, all of the AeYV-related viral proteins exhibited more than 10% aa sequence divergence relative to known poleroviruses, thus exceeding the species demarcation threshold defined for the genus Polerovirus [30]. These findings support the identification of a putative novel polerovirus, which we have tentatively named "tomato necrotic yellowing virus" (ToNYV).
Recombination analysis using RDP4 software [31], with default parameters and a full-genome sequence alignment of ToNYV and other poleroviruses did not reveal any evidence of recombination in the ToNYV genome.
Poleroviruses were traditionally thought to be transmitted exclusively by aphids in a persistent and circulative manner [32]. However, recent discoveries have identified two poleroviruses, PeWBVYV [33] and the Brazilian recombinant isolates of cucurbit aphid-borne yellows virus (CABYV) [34], that are transmitted by whiteflies. Given the near absence of aphids and the widespread presence of whiteflies in tomato greenhouses affected by ToNYV, we investigated the vector transmission capacity of ToNYV by the aphid Myzus persicae and the whitefly Bemisia tabaci MEAM1. For transmission assays, approximately 400 aphids and 1,200 whiteflies, reared on tobacco and cabbage, respectively, were given a 48-hour acquisition period on ToNYV-infected tomato leaves exhibiting strong ToNYVD symptoms. Leaves were collected from a 5-month-old greenhouse tomato plant showing severe symptoms of ToNYVD. The presence of ToNYV genomic RNA was confirmed by RT-PCR, and the absence of ToCV and PVY genomic RNA and TYLCV genomic DNA was confirmed by specific RT-PCR and PCR, respectively. Subsequently, these aphids and whiteflies were transferred to separate rearing cages and allowed a 48-hour inoculation access period on 10 and 20 15-day-old tomato seedlings (cv. Roma), respectively. No symptoms of ToNYVD were observed, and no ToNYV genomic RNA was detected by RT-PCR in any of the 10 tomato plants 30 days after they were exposed to viruliferous M. persicae, indicating that this aphid species may not transmit ToNYV. In contrast, 35% (7/20) of the tomato plants exposed to viruliferous B. tabaci developed yellowing symptoms 30 days post-exposure, five of which tested positive for ToNYV by RT-PCR followed by Sanger sequencing. These findings confirm that ToNYV is transmissible by B. tabaci MEAM1. essential to identify their dissemination routes and to evaluate their potential threat to global tomato production.
ToNYV genomic RNA was detected in 21% (8/38) of the samples collected in 2018 and 14.6% (14/96) of those collected in 2020. Positive samples were distributed across four localities (Saint-Louis, Saint-Pierre, Saint-Joseph, and Saint-Philippe; Supplementary Table S5), covering the southern and south-eastern parts of the island's main vegetable production zone. These finding indicate a widespread presence of ToNYV within the tomato-growing production areas of La Réunion.
In conclusion, we demonstrate the utility of highthroughput-sequencing-based metagenomic screening for the detection and characterisation of plant viruses in symptomatic tomato plants from fields and greenhouses in La Réunion. Alongside previously reported tomato-infecting viruses in La Réunion (TSWV, PVY, TYLCV, ToCV, and STV), we identified and characterised a novel polerovirus associated with necrotic reddening and yellowing of leaves. The complete genome sequence of this virus was obtained and analysed. Based on its distinct genomic organisation, amino acid sequence divergence, phylogenetic placement, and confirmed transmission by B. tabaci, we propose that this virus represents a new species within the genus Polerovirus, for which we suggest the species name "Polerovirus ToNYV".
As noted by Ghosh et al. [33], the emergence of whiteflytransmitted poleroviruses poses a significant threat to global agriculture, particularly because B. tabaci is regarded as a "supervector" of plant viruses. Some of its cryptic species, such as MEAM1 and MED, exhibit extreme polyphagy and high levels of insecticide resistance [37]. The specificity of polerovirus transmission by aphid vectors is largely determined by the N-terminal region of the readthrough domain (RTD) [32,38]. While the vector of AeYV remains unconfirmed, the consistent association of AeYV-infected tomato crops in Côte d'Ivoire with whitefly populations [39], together with the high degree of sequence similarity in the CP and RTD regions between AeYV, ToNYV, and PeWB-VYV, strongly suggests that AeYV may also be transmitted by B. tabaci. The recent emergence of these novel poleroviruses across geographically distant regions, including the Middle East, South America, West Africa, and the Mascarene Archipelago, raises important questions regarding their evolutionary trajectories and possible common origins. However, the apparent lack of homology in factors that are likely to determine transmission specificity (CP and RTD) between African poleroviruses (AeYV, ToNYV and PeW-BVYV) and the South American recombinant isolates of CABYV, which are also transmitted by B. tabaci, suggests distinct evolutionary lineages and a convergent acquisition of this transmission ability. Clarifying these patterns will require in-depth studies on the ecology, vector interactions, and phylogeography of these viruses, as such knowledge is
## References
1. Massart, Olmos, Jijakli et al. (2014) "Current impact and future directions of high throughput sequencing in plant virus diagnostics" *Virus Res*
2. Adams, Fox (2016) "Diagnosis of plant viruses using next-generation sequencing and metagenomic analysis. Current research topics in plant virology"
3. François, Filloux, Fernandez et al. (2018) "Viral metagenomics approaches for high-resolution screening of multiplexed arthropod and plant viral communities. Viral metagenomics : . methods and protocols"
4. Roossinck, Saha, Wiley et al. (2010) "Ecogenomics: using massively parallel pyrosequencing to understand virus ecology" *Mol Ecol*
5. Kreuze, Perez, Untiveros et al. (2009) "Complete viral genome sequence and discovery of novel viruses by deep sequencing of small RNAs: a generic method for diagnosis, discovery and sequencing of viruses" *Virology*
6. Filloux, Fernandez, Loire et al. (2018) "Nanopore-based detection and characterization of yam viruses" *Sci Rep*
7. Orfanidou, Efthimiou, Katis et al. (2022) "Elucidating the sweet potato virome in Greece with the aid of highthroughput sequencing technology" *Plant Pathol*
8. Wongsurawat, Jenjaroenpun, Taylor et al. (2019) "Rapid sequencing of multiple RNA viruses in their native form" *Front Microbiol*
9. Kumar, Stecher, Li et al. (2018) "MEGA X: molecular evolutionary genetics analysis across computing platforms" *Mol Biol Evol*
10. Muhire, Varsani, Martin (2014) "SDT: a virus classification tool based on pairwise sequence alignment and identity calculation" *Plos One*
11. Sõmera, Fargette, Hébrard et al. (2021) "ICTV report consortium. ICTV virus taxonomy profile: Solemoviridae" *J Gen Virol*
12. Martin, Murrell, Golden et al. (2015) "RDP4: detection and analysis of recombination patterns in virus genomes" *Virus Evol*
13. Brault, Périgon, Reinbold et al. (2005) "The polerovirus minor capsid protein determines vector specificity and intestinal tropism in the aphid" *J Virol*
14. Ghosh, Kanakala, Lebedev et al. (2019) "Transmission of a new polerovirus infecting pepper by the whitefly Bemisia tabaci" *J Virol*
15. Costa, Inoue-Nagata, Vidal et al. (2020) "The recombinant isolate of cucurbit aphid-borne yellows virus from Brazil is a polerovirus transmitted by whiteflies" *Plant Pathol*
16. Afouda, Kone, Zinsou et al. (2017) "Virus surveys of Capsicum spp. in the Republic of Benin reveal the prevalence of pepper vein yellows virus and the identification of a previously uncharacterised polerovirus species" *Adv Virol*
17. Lotos, Olmos, Orfanidou et al. (2017) "Insights into the etiology of polerovirusinduced pepper yellows disease" *Phytopathology*
18. Gilbertson, Batuman, Webster et al. (2015) "Role of the insect supervectors Bemisia tabaci and Frankliniella occidentalis in the emergence and global spread of plant viruses" *Ann Rev Virol*
19. Schiltz, Wilson, Hosford et al. (2022) "Polerovirus N-terminal readthrough domain structures reveal molecular strategies for mitigating virus transmission by aphids" *Nat Commun*
20. Bele, Diallo, Martin et al. (2023) "First report of African eggplant yellowing virus on tomato exhibiting necrotic yellowing symptoms in northern Côte d'Ivoire" *Plant Dis*
21. "Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations"
22. Adams, Miano, Kinyua et al. (2013) "Use of next-generation sequencing for the identification and characterization of Maize chlorotic mottle virus and Sugarcane mosaic virus causing maize lethal necrosis in Kenya" *Plant Pathol*
23. Bernardo, Charles-Dominique, Barakat et al. (2017) "Geometagenomics illuminates the impact of agriculture on the distribution and prevalence of plant viruses at the ecosystem scale" *ISME J*
24. Reynaud, Chabriere (1991) "Virology of market vegetable cultivars" *Annual Report of Cirad*
25. Grisoni (1996) "Quelques données sur les Potyvirus des cultures horticoles à la Réunion. CIRAD-FLHOR"
26. Peterschmitt, Granier, Mekdoud et al. (1999) "First report of tomato yellow leaf curl virus in Réunion Island" *Plant Dis*
27. Delatte, Holota, Naze et al. (2005) "The presence of both recombinant and nonrecombinant strains of Tomato yellow leaf curl virus on tomato in Reunion Island" *Plant Pathol*
28. Delatte, Reynaud, Granier et al. (2005) "A new silverleaf-inducing biotype Ms of Bemisia tabaci (Hemiptera: Aleyrodidae) indigenous to the islands of the south-west Indian Ocean" *Bull Entomol Res*
29. Delatte, Naze, Cottineau et al. (2006) "Occurrence of tomato chlorosis virus on tomato in reunion Island" *Plant Pathol*
30. Roux (2020) "Compte rendu du groupe de travail phytosanitaire"
31. Sabanadzovic, Valverde, Brown et al. (2009) "Southern tomato virus: the link between the families Totiviridae and Partitiviridae" *Virus Res*
32. Martin (2011) "Cutadapt removes adapter sequences from highthroughput sequencing reads" *EMBnet J*
33. Bolger, Lohse, Usadel (2014) "Trimmomatic: a flexible trimmer for Illumina sequence data" *Bioinformatics*
34. Prjibelski, Antipov, Meleshko et al. (2020) "Using SPAdes de novo assembler" *Curr Protoc Bioinformatics*
35. Nibert, Pyle, Firth (2016) "A +1 ribosomal frameshifting motif prevalent among plant amalgaviruses" *Virology*
36. Fuchs, Bar-Joseph, Candresse et al. (2020) "Report consortium I. ICTV virus taxonomy profile: Closteroviridae" *J Gen Virol*
37. Fiallo-Olivé, Navas-Hermosilla, Ferro et al. (2018) "Evidence for a complex of emergent poleroviruses affecting pepper worldwide" *Adv Virol*
38. Adams, Antoniw, Fauquet (2005) "Molecular criteria for genus and species discrimination within the family Potyviridae" *Adv Virol*
39. Li, Liu, Luo et al. (2015) "MEGA-HIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph" *Bioinformatics*
40. Buchfink, Xie, Huson (2015) "Fast and sensitive protein alignment using DIAMOND" *Nat Methods* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12477023&blobtype=pdf | # A healthcare-associated outbreak of hepatitis C virus infections attributable to tampering injectable anaesthetic opioids, South Germany, 2017-2018
Katarzyna Schmidt, Stefanie Böhm, Raffaella Hesse, C.-Thomas Bock, Sebastian Haller, Katharina Katz, Stefan Ross, Jörg Timm, Ruth Zimmermann, Sandra Niendorf, Marc Struelens, Eustachio Cuscianna
## Abstract
Introduction:In October 2018, an outbreak of hepatitis C virus (HCV) in southern Germany was communicated to the Robert Koch Institute (RKI). Healthcareassociated transmission during invasive procedures involving a specific anaesthetist at a Bavarian hospital was suspected. The aim was to conduct a retrospective molecular outbreak investigation in order to elucidate the course of the outbreak. Methods: An exposed patient was defined as a person who underwent a surgical procedure involving the anaesthetist in the Bavarian hospital from May 2016 to April 2018. A probable case was defined as an exposed patient with a positive HCV antibody test result and unknown HCV genotype. A confirmed case represented a probable case with hepatitis C genotype 3 (3a) infection. Descriptive epidemiological and phylogenetic analyses (using four HCV regions: Core, HVR1, NS5A and NS5B) were conducted. Results: Of the 1,714 exposed patients, to whom HCV testing was recommended, 1,558 (90.9%) responded and were tested, 63 met the definition of a probable case, and 51 of those were confirmed cases. Sequencing data were available for 39 of the 51 confirmed cases. A sample from the anaesthetist was unavailable for further analysis. Phylogenetic analysis revealed close genetic relatedness of all 39 confirmed cases with identified HCV genotype 3a. Phylogenetic results indicated a common source of infection. Discussion: To prevent healthcare-associated HCV transmission during anaesthetic procedures, protocols must document the amount of medication used and discarded. Regular staff testing and storing of clinical samples are also crucial for timely outbreak analysis and response.
## 1 Introduction
Hepatitis C is a liver inflammation caused by hepatitis C virus (HCV) predominantly spreading through the parenteral route. An acute infection leads to a chronic course in 50-85% of individuals, potentially progressing to liver cirrhosis and hepatocellular carcinoma (1). Treatment options have improved since the introduction of direct acting antivirals with cure rates of 90% and more (2,3).
The HCV genome is an enveloped, positive-strand RNA of about 9.6 kb with a single open reading frame encoding 3,000 amino acids and cleaved into structural (core, E1, E2) and non-structural (p7, NS2, NS3, NS4A, NS4B, NS5A, NS5B) proteins (4). Seven HCV genotypes and numerous subtypes have already been described (5), but only a few of them are currently circulating (1a, 1b, 2a, 3a, 4) in high income countries (5)(6)(7).
Healthcare-associated HCV transmission usually involves contaminated equipment (8)(9)(10) or unsafe injection practices (11,12). Transmission can also occur via healthcare providers diverting drugs through unsafe injections (13)(14)(15). Typically, drug diversion involves a healthcare worker (HCW) misusing narcotic drugs intended for patients, also called "tampering" (16). HCV transmission could occur if the HCW is infected with HCV.
In June 2018, a local public health authority (LPHA) in Bavaria received a report of hepatitis C infection in an anaesthetist, whose employment at a local hospital had ended in April 2018 due to termination agreement. In October 2018, the LPHA was notified about a cluster of three new HCV infections of unknown origin in patients without clear risk behaviours as reported by a local general practitioner (GP). The LPHA investigation revealed that the affected patients had surgeries at the local hospital where the anaethetist was previously employed, and that these operations involved the anaesthetist in question. Therefore, the suspected transmission of HCV was a link between the patients and the anaesthetist. When an epidemiological link to the anaesthetist emerged, the LPHA, together with the local hospital, launched an extensive case finding and testing campaign to find further infected patients and determine the extent of the outbreak.
Phylogenetic analysis of viral sequences is often used to determine which samples are part of an outbreak and -when samples from suspected index persons are available -to identify the source of healthcare-associated HCV outbreaks (17)(18)(19)(20). Determining the degree of genetic relatedness between viral isolates from the probable source of infection and from known cases provides valuable evidence to support and/or rule out hypotheses about possible transmission pathways. Here, we applied this phylogenetic method combined with epidemiological contact tracing data to demonstrate its utility in a retrospective analysis of a large hospital-acquired HCV outbreak associated with anaesthetic procedures in Germany.
## 2 Methods
## 2.1 Case finding
The case-finding period was delineated based on the anaesthetist's previous HCV tests and the dates of anaesthetist's employment at the local hospital. The anaesthetist's last negative HCV test result during occupational health examinations, was in November 2016. Anti-HCV seroconversion occurs on average eight to 11 weeks after infection (21), but may take up to six months (22). Therefore, the start of the casefinding period was set as May 2016, which was six months before the last negative test. The end of the case-finding period was set as the April 24th 2018, which was the anaesthetist's last working day at the local hospital. An acute HCV genotype 3 infection in the anaesthetist was confirmed in June 2018, with antiviral therapy initiated one month later. A negative PCR result was recorded in October 2018. Unfortunately, none of the anaethetist's samples were available for this investigation, as they were discarded in the laboratory and not retained for further analysis.
Hospital staff reviewed over 10,000 surgical protocols from the defined period to determinate the involvement of the anaesthetist in question. Certain procedures, including major surgeries, were linked to potential exposure, while others such as central lines insertions, were excluded based on interviews with anaesthetists, clarifying the scope of procedures considered. A total of 1,714 patients, whose surgical procedures involved the anaesthetist during the case-finding period, were contacted via letter by the hospital and LPHA to inform them about the outbreak and to invite them for hepatitis C screening. Testing for other bloodborne pathogens including hepatitis A, hepatitis B and HIV was not offered, as these infections had been ruled out based on prior examinations of the anaesthetist, which yielded negative results. Among patients contacted anti-HCV status was determined. None of the individuals had a previously known hepatitis C infection. Additionally, all employees (doctors, nurses as well as operating theatre and cleaning staff) working in the hospital's operating theatre were tested for an HCV infection, with the exception of 6-8 former employees who could not be located or who refused to provide consent. No additional hepatitis C-positive staff member was identified.
## 2.2 Case definitions
An exposed patient was defined as a person who underwent a surgical procedure involving the anaesthetist in the local Bavarian hospital from May 2016 to April 24th 2018. A probable case was defined as an exposed patient with a positive HCV antibody test result and unknown HCV genotype. A confirmed case was defined as a probable case where infection with hepatitis C genotype 3 (3a) was confirmed via genotyping. Phylogenetic analyses were conducted on confirmed cases with available serum samples.
## 2.3 Descriptive analysis
Detailed demographic and clinical information were collected in a linelist. An epicurve was constructed using the exposure date (date of surgery) (Figure 1). Descriptive epidemiological analysis was performed for basic demographics (gender and age) using R (version 4.1.3). Due to the retrospective nature of the analysis, age and age groups were calculated based on the time of HCV testing.
## 2.4 HCV testing
Exposed patients had HCV antibody tests performed at the local hospital or at a private laboratory initiated by the patients' GPs. Serum samples of 44 out of the 51 confirmed cases were available and were sent to the RKI laboratory for further investigation. The RNA was extracted from 140 μL of the serum using the Qiagen viral RNA extraction kit (Qiagen Hilden, German) according to the manufacturer's instructions. Reverse transcription was performed using SuperScript™ IV Reverse Transcriptase (200 U/μL) (Thermo Fisher, Waltham, United States) at the following conditions: denaturation at 65 °C for 5 min., annealing at 23 °C for 10 min., cDNA synthesis at 50 °C for 6 min., 55 °C for 6 min., 60 °C for 6 min., and termination at 80 °C for 10 min. Hepatitis C genotype confirmation was performed using SuperScript III Platinum One-Step qRT-PCR Kit (Thermo Fisher) with primers and TaqMan probe specific for HCV 3 genotype (Table 1). PCR conditions were as followed: 50 °C for 5 min., 55 °C for 10 min., 95 °C for 1 min., and then 45 cycles of PCR amplification at 95 °C for 15 s., and at 58 °C for 45 s. Prior to the Sanger sequencing, four independent nested PCR tests were performed using primers specific for four HCV genotype 3a regions (core, hypervariable-HVR1, nonstructural 5A-NS5A and nonstructural 5B-NS5B) described in Table 1. PCR products were visualized on a 1x TAE 1.5% agarose gel using 100 base pairs (bp) DNA Ladder (Thermo Fisher). PCR products (core: 230 bp, HVR1: 523 bp, NS5A: 679 bp and NS5B: 674 bp) were cleaned by ExoSap-IT (Thermo Fisher) and sequenced with corresponding PCR primers using BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher) by the Method development, research infrastructure and information technology department at the RKI.
## 2.5 Phylogenetic analysis
For 39 confirmed cases, isolates were available for sequencing. Phylogenetic analysis was performed on 36 core sequences (225 bp), 39 HVR1 sequences (504 bp), 38 NS5A sequences (657 bp) and 38 NS5B sequences (653 bp). The alignments for each of the regions were performed using MUSCLE 3.8.425 in Geneious software (v2021.2.2) with default parameters (maximum of 8 interactions). Phylogenetic trees were constructed by the maximum likelihood (ML) method with bootstrap 1,000 using the software MEGA11 (v11.0.11) and 13 reference sequences (Figure 2). For each region, the model based on the lowest Bayesian Information Criterion scores was selected. The Kimura 2-parameter model with invariant sites (K2 + I) was used for HCV core region, and the Tamura 3-parameter model with gamma distribution and invariant sites (T92 + G + I) was used for HVR1, NS5A and NS5B regions. The remaining parameters were set as defaults. Bootstrap values higher than 70 were shown at the nodes of the phylogenetic trees.
Given that no serum sample from the anaesthetist was available, a most recent common ancestor (MRCA) sequence for each genomic region was reconstructed using Arning approach (23). This method reconstructs a probable ancestral sequence by aligning viral genomes from confirmed cases and selecting the sequence with the fewest mutations relative to the outbreak samples. The MRCA was then used as a reference to calculate the number of single nucleotide polymorphisms (SNPs) distances in the 39 confirmed cases with available sequencing data. SNP distances were interpreted in terms of genetic relatedness. Generally, a low number of SNPs indicates a higher degree of similarity between the MRCA and a confirmed case suggesting a closer genetic relationship within the outbreak. To assess whether the number of cumulative SNPs increased with time, correlation coefficient and the R squared values for the relationship between the SNPs count and the exposure and the testing dates were calculated.
## 3 Results
## 3.1 Descriptive findings
Of the 1,714 eligible patients contacted, 1,558 were tested for HCV antibody screening (response rate = 90.9%). Among them, 63 probable cases were identified, with 51 cases confirmed by genotyping. The remaining twelve probable cases were HCV-RNA negative and genotyping was unable to be performed. Among the 51 confirmed cases, the epidemiological investigation revealed no alternative sources of hepatitis C exposure besides surgeries involving the anaesthetist in question. Thus, the timeframe from the first to the last exposure dates of probable and confirmed cases ranged from February 2017 until April 2018 (Figure 1). In the two cases with more than one surgery, only the first date of surgery was reported and used in further analysis. Nine cases had exposure dates between February 2017 and July 2017. After October 2017, there was a sharp uptick in the probable and confirmed cases, peaking at 13 cases in February 2018. Twelve cases had exposure dates in March 2018 and eight cases in April 2018. Confirmed cases were equally distributed among males and females (25:26) and were mostly adults in older age groups (51-90 years old) (Table 2).
## 3.2 Laboratory findings
Of the 51 confirmed cases, six were identified with genotype 3, and 45 with genotype 3a. Serum samples of 44 of the 51 confirmed cases were available for sequence analysis, with sequence and epidemiological data concordant in 39 cases. Phylogenetic analyses revealed that all 39 confirmed cases were closely related and belonged to the HCV genotype 3a (Figure 2), which is the same genotype previously identified in the anaesthetist (written communication from the local health authority). The examined samples clustered with a high level of confidence, supported by high bootstrap values (greater than 90%) for each region, indicating close genetic relatedness between the samples. In the remaining 12 confirmed cases without sequence data, it was not possible to determine with certainty that they were highly related, as their classification was based solely on hepatitis C genotype 3 (3a) infection and not on phylogenetic analysis. The Arning's MRCA sequence reconstruction approach revealed that all viral sequences were closely related, thus almost all of them could be used as an ancestor in further analyses. Therefore, the sequences that were the closest to the MRCA (21_HCV_2018 for core region and 11_HCV_2018 for the remaining regions) were used as reference to calculate the number of SNPs distances in the HCV isolates. In total, we identified 1,073 SNPs across four genomic regions (Core: 43, HVR1: 354, NS5A: 435, NS5B: 241) with an average 27 SNPs per sample (range 3-36 SNPs per sample) in the HCV isolates from the 39 confirmed cases (Figure 3). The range of nucleotide genetic distances between the PCR amplicons to the genomic regions were as follows for core: 0-3, HVR1: 0-16, NS5A: 1-15, and NS5B: 0-10. The timeframe from exposure to testing dates spanned from February 2017 to November 2018 (Figure 3), with exposure-to-testing periods ranging from three to 20 months, and averaging nine months. The statistical test showed no correlation between the number of SNPs and the exposure date and testing date (Table 3). Frontiers in Public Health 06 frontiersin.org
## 4 Discussion
This report describes the largest known hospital-acquired HCV outbreak in Germany, linked to drug diversion by an infected HCW. The findings indicate that at least 51 patients were likely infected by the same anaesthetist during anaesthetic procedures. This investigation retrospectively describes the transmission cluster of HCV-affected patients, and ascertains the genetic relatedness of the HCV isolates from the confirmed cases. Genotyping, together with epidemiological data, identified 51 confirmed cases with HCV genotype 3 (3a) infection from 63 probable cases. In-depth phylogenetic analysis of four HCV regions in a sub-sample of confirmed cases suggested that all were infected by the same single viral strain of HCV genotype 3a, indicating close genetic relatedness, supported by high bootstrap values. While the anaesthetist's sample was unavailable for analysis, no evidence of other HCV-positive HCWs or epidemiological links between the confirmed cases were found. Based on these findings, we hypothesize that the probable source of infection was an HCV-positive anaesthetist involved in all the confirmed cases; and the suspected transmission route was the contamination of syringes and/or needles through misuse by the infected anaesthetist prior to administration to the patient. However, the study's limitations prevent us from conclusively verifying these hypotheses.
The outbreak was detected in October 2018, when a Bavarian LPHA was notified of three newly diagnosed cases of hepatitis C in patients with unknown exposure, and an epidemiological link to one local hospital was identified. An outbreak investigation was launched using epidemiological analysis and anti-HCV screening for patients who underwent surgery between May 2016 and April 2018. The earliest surgery date was February 2017. Knowing that spontaneous clearance of acute HCV associated with genotype 3 is common, at least in the first 3 months after infection (24, 25), the delay for casesearching may have led to undiagnosed and/or missed cases. Here, the exposure-to-testing period for probable cases ranged from 1 to 21 months (average 9 months), which means that some patients might have developed a chronic condition before the implementation of treatment. However, no clinical information regarding the HCV infections, symptoms and treatment were available. Due to HCV's long incubation and often asymptomatic nature, healthcare-associated HCV outbreaks are frequently identified with a significant delay. Consequently, determining the exposure period and transmission route may be challenging or even impossible, potentially resulting in undiagnosed and therefore untreated cases. Despite challenges in distinguishing acute from chronic HCV infection, implementing recommendations for managing HCV-infected HCWs in clinical settings remains essential. For instance, regular mandatory testing of HCWs by occupational health can support early detection of infected HCWs, and by that, reduce risk of transmission during exposureprone procedures to patients. The German Association for the Control of Viral Diseases (DVV) prohibits high-risk activities for HCWs with HCV viral loads >25,000 IU/mL, and allows those with viral loads <250 IU/mL to perform such activities with precautions (e.g., double gloves) (26). In addition, early antiviral therapy is recommended for infected HCWs and patients (25) to prevent liver damage and transmission.
We hypothesized that a longer exposure-to-testing interval would result in a higher number of cumulative SNPs per sample due to the accumulation of mutations over time. However, data from the 39 confirmed cases despite varying intervals showed no consistent pattern (Figure 3). For example, patient number 8_HCV_2018 (Figure 3), with the shortest period (3 months) between exposure and testing dates, had 27 cumulative SNPs identified similar to the number found in patient 49_HCV_2018 (26 SNPs) despite the longer interval (20 months). In contrast, two patients (11_HCV_2018 and 63_HCV_2018), with similar interval between exposure and testing dates (19 and 20 months) showed substantial differences in SNPs counts (3 versus 29 SNPs respectively). The reconstructed sequence of patient number 11_HCV_2018 demonstrated the closest similarity to the MRCA (exposure date: February 2017), thus serving as the common ancestor for most of the HCV regions. Despite the expectation that cumulative SNPs would increase over time, the linear regression revealed no significant correlation between the total number of SNPs and time (measured by the exposure date and the testing date). However, the lack of the viral sequence from the suspected anaesthetist's sample prevented a direct comparison with confirmed cases, limiting our ability to reconstruct transmission chains with higher certainty. Furthermore, the minimal genetic variation among available sequences restricted temporal inferences, making it difficult to determine precise transmission timelines.
In western European countries, the distribution of HCV genotypes is relatively uniform. A recent study showed that the predominant HCV genotypes in Germany are genotype 1 (47%) and genotype 3 (46%) (27). HCV genotype 3a is commonly found among people who inject drugs (28)(29)(30)(31). Healthcare-associated HCV outbreaks attributable to narcotic diversion or unsafe injection practices have been reported in the United States (14,32,33), and in several European countries, including France (11) and Spain (19,20). These cases reveal common vulnerabilities such as delayed outbreak recognition, insufficient monitoring of anaesthetic drug access, and institutional failure to detect diversion behaviours. Hatia et al. (16) conducted a meta-analysis of 46 studies published between 1990 and 2012 describing nosocomial HCV outbreaks caused by HCWs who diverted injectable opioids. Their findings showed that the HCV transmission risk from drug diversion was substantially higher compared to surgical exposure, underscoring the importance of targeted prevention strategies.
Key aspects to prevent drug diversion outbreak involve proactive measures and protocols in healthcare settings to minimize implementation of a centralized national reporting system to support healthcare institutions in fulfilling their ethical obligation to protect patient from such preventable harm. Similarly, our investigation relied on collaboration with local, regional, and state health authorities, underscoring the importance of transparency, reporting structures, and system-wide learning in responding to and preventing future outbreaks.
The molecular identification of HCV sequence clusters can speed up public health response (18). Two prominent molecular approaches have been proposed to investigate the HCV transmission clusters in outbreaks settings (18,36). While many uses next-generation sequencing to track transmission chain, we used Sanger sequencing and phylogenetic analysis of four HCV regions, which proved effective for this investigation.
The study had four sample-related limitations. First, the anaesthetist's sample was unavailable. Thus, we could not include it in the phylogenetic analysis to definitively confirm that the anaesthetist was a source of the outbreak. The lack of this sequence data limits our ability to establish a direct genetic link between the aneasthetist and other cases, which affects the strength of causal conclusions regarding transmission pathways. However, despite this limitation, our study provides strong epidemiological and genomic evidence supporting transmission events. The available viral sequences from confirmed cases show high genetic relatedness, indicating a common transmission source. Although, the missing sample limits a complete phylogenetic reconstruction, the clustering of cases -considering their genetic similarity as well as temporal and spatial proximity through surgeries performed at the same hospital by the anaesthetist in question -supports the common source of transmission. To enhance genomic investigations in future outbreaks, the systematic collection and long-term storage of all relevant clinical samples should be prioritized. Additionally, further analysis (e.g., molecular clock) to determinate the HCV genome's evolution rate and estimate the exact date of infection, was not possible. This would have helped to narrow the exposure period and reduce the number of patients needing testing. However, the lack of the viral sequence data from the anaethetist prevented the establishment of reliable priors for our analysis. Furthermore, the genetic similarity among the available sequences from confirmed cases resulted in minimal sequence variation, limiting the resolution of temporal inferences needed to accurately estimate infection timing. Second, sequencing data were missing for 12 confirmed cases, all of which had the same exposure period as the other confirmed cases falling within the case-finding period, and underwent surgery at the same hospital involving the suspected anaethetist. Although, the lack of sequence data unable their direct inclusion in the phylogenetic analysis, their strong epidemiological link suggests they would have clustered with the outbreak cases, already supported by a high bootstrap value (greater than 90%). This limitation highlights the need for improved communication and/or cooperation between parties involved, as well as the importance of storing HCV-positive samples for extended period to facilitate future genomic investigations. Third, clinical information, including symptomatology, liver function outcomes, and antiviral treatment, was not available for the confirmed cases. While such data would have strengthened the clinical interpretation of our findings, their absence does not affect the genetic and epidemiological conclusions of the study. Future investigations would benefit from integrating clinical and genomic data to provide a more comprehensive understanding of transmission dynamics and disease outcomes. Lastly, the sampling dates (day of sample collection) were missing for almost 50% of the sequenced HCV isolates. In order to disclose the number of SNPs that evolved within a timespan (exposure-to-sampling date) we used a proxy for the sampling date. This likely introduced little change to our findings, as the testing date (the date when the samples were tested positive) is the closest date to the sample collection date to the best of our knowledge.
This study describes a large hospital-acquired HCV outbreak in Germany likely caused by drug diversion. Molecular analyses indicated that all the confirmed cases were closely genetically related and likely stemmed from the same infection source. Early outbreak detection of chronic infections like HCV is challenging, making preventive measures crucial. This includes maintaining staff restrictions to accessing controlled substances, the monitoring of access to dispensing systems, and establishing drug diversion prevention teams. Furthermore, we recommend regular testing for HCV and other blood-borne infections of all HCWs, in particular of those involved in anaesthetic procedures and opioids handling. Preventing the transmission of infections from infected staff to patients, whether HCV or others-is crucial. Therefore, clear guidelines, monitoring and control systems of drug diversion should be implemented in hospitals. Importantly, positive samples should be stored for further testing to support rapid and effective outbreak investigations, if needed.
## Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.
## References
1. Zaltron, Spinetti, Biasi et al. (2012) "Chronic HCV infection: epidemiological and clinical relevance" *BMC Infect Dis*
2. Cornberg, Stoehr, Naumann et al. (2022) "Realworld safety, effectiveness, and patient-reported outcomes in patients with chronic hepatitis C virus infection treated with Glecaprevir/Pibrentasvir: updated data from the German hepatitis C-registry (DHC-R)" *Viruses*
3. Krüger, Rossol, Krauth et al. (2023) "Real-world experience for the outcomes and costs of treating hepatitis C patients: results from the German hepatitis C-registry (DHC-R)" *Z Gastroenterol*
4. Kato (2000) "Genome of human hepatitis C virus (HCV): gene organization, sequence diversity, and variation" *Microb Comp Genomics*
5. Messina, Humphreys, Flaxman et al. (2015) "Global distribution and prevalence of hepatitis C virus genotypes" *Hepatology*
6. (2022) "Global change in hepatitis C virus prevalence and cascade of care between 2015 and 2020: a modelling study" *Lancet. Gastroenterol Hepatol*
7. Petruzziello, Loquercio, Sabatino et al. (2019) "Prevalence of hepatitis C virus genotypes in nine selected European countries: a systematic review" *J Clin Lab Anal*
8. Furusyo, Kubo, Nakashima et al. (2004) "Confirmation of nosocomial hepatitis C virus infection in a hemodialysis unit" *Infect Control Hosp Epidemiol*
9. Savey, Simon, Izopet et al. (2005) "A large nosocomial outbreak of hepatitis C virus infections at a hemodialysis center" *Infect Control Hosp Epidemiol*
10. Singh, Stoitsova, Zakrzewska et al. (2006) "Healthcareassociated hepatitis B and C transmission to patients in the EU/EEA and UK: a systematic review of reported outbreaks between"
11. Germain, Carbonne, Thiers et al. (2005) "Patientto-patient transmission of hepatitis C virus through the use of multidose vials during general anesthesia" *Infect Control Hosp Epidemiol*
12. Comstock, Mallonee, Fox et al. (2004) "A large nosocomial outbreak of hepatitis C and hepatitis B among patients receiving pain remediation treatments" *Infect Control Hosp Epidemiol*
13. Warner, Schaefer, Patel et al. (2015) "Outbreak of hepatitis C virus infection associated with narcotics diversion by an hepatitis C virus-infected surgical technician" *Am J Infect Control*
14. Schaefer, Perz (2014) "Outbreaks of infections associated with drug diversion by US health care personnel" *Mayo Clin Proc*
15. Shemer-Avni, Cohen, Naus et al. (2007) "Iatrogenic transmission of hepatitis C virus (HCV) by an anesthesiologist: comparative molecular analysis of the HCV-E1 and HCV-E2 hypervariable regions" *Clin Infect Dis*
16. Hatia, Dimitrova, Skums et al. (2015) "Nosocomial hepatitis C virus transmission from tampering with injectable anesthetic opioids" *Hepatology*
17. Spada, Sagliocca, Sourdis et al. (2004) "Use of the minimum spanning tree model for molecular epidemiological investigation of a nosocomial outbreak of hepatitis C virus infection" *J Clin Microbiol*
18. Campo, Xia, Dimitrova et al. (2016) "Accurate genetic detection of hepatitis C virus transmissions in outbreak settings" *J Infect Dis*
19. Bruguera, Saiz, Franco et al. (2002) "Outbreak of nosocomial hepatitis C virus infection resolved by genetic analysis of HCV RNA" *J Clin Microbiol*
20. González-Candelas, Bracho, Wróbel et al. (2013) "Molecular evolution in court: analysis of a large hepatitis C virus outbreak from an evolving source" *BMC Biol*
21. Barrera, Bruguera, Ercilla et al. (1995) "Persistent hepatitis C viremia after acute self-limiting posttransfusion hepatitis C" *Hepatology*
22. Laperche, Le Marrec, Girault et al. (2005) "Simultaneous detection of hepatitis C virus (HCV) core antigen and anti-HCV antibodies improves the early detection of HCV infection" *J Clin Microbiol*
23. Arning, Lehmann, Meyer et al. (2004) "High rate of spontaneous clearance of acute hepatitis C virus genotype 3 infection" *J Med Virol*
24. Ch, Zimmermann, Berg et al.
25. Prophylaxis (2018) "diagnosis and therapy of hepatitis-C-virus (HCV) infection: the German guidelines on the management of HCV infection" *Gastroenterol*
26. Glebe, Van Bömmel, Dudareva et al. (2020) "Prävention der nosokomialen Übertragung von Hepatitis-B-Virus (HBV) und Hepatitis-C-Virus (HCV) durch im Gesundheitswesen Tätige: Empfehlungen der Deutschen Vereinigung zur Bekämpfung der Viruskrankheiten (DVV) e. V" *Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz*
27. Rki (2016) "Abschlussbericht der Studie "Drogen und chronischen Infektionskrankheiten in Deutschland" (DRUCK-Studie)"
28. Berg, Hopf, Stark et al. (1997) "Distribution of hepatitis C virus genotypes in German patients with chronic hepatitis C: correlation with clinical and virological parameters" *J Hepatol*
29. Schröter, Zöllner, Schäfer et al. (2002) "Epidemiological dynamics of hepatitis C virus among 747 German individuals: new subtypes on the advance" *J Clin Microbiol*
30. Driesel, Wirth, Stark et al. (1994) "Hepatitis C virus (HCV) genotype distribution in German isolates: studies on the sequence variability in the E2 and NS5 region" *Arch Virol*
31. Zimmermann, Krings, Steffen et al. (2018) "-Pilotierung eines Surveillancesystems zu durch Blut und sexuell übertragenen Infektionen bei Drogengebrauchenden" *Clin Infect Dis*
32. Hellinger, Bacalis, Kay et al. (2012) "Health care-associated hepatitis C virus infections attributed to narcotic diversion" *Ann Intern Med*
33. Knight, May, Tyson et al. (2022) "Detecting drug diversion in health-system data using machine learning and advanced analytics" *Am J Health Syst Pharm*
34. Lahey (2015) "A proposed nationwide reporting system to satisfy the ethical obligation to prevent drug diversion-related transmission of hepatitis C in healthcare facilities" *Clin Infect Dis*
35. Longmire, Sims, Rytsareva et al. (2017) "GHOST: global hepatitis outbreak and surveillance technology" *BMC Genomics* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12607861&blobtype=pdf | # From zoonotic spillover to endemicity: the broad determinants of human coronavirus tropism
| Virology, | Minireview, Saskia Westhoven, Luca Bertzbach, Mara Kloehn, Cedric Mahncke, Natalie Heinen, Richard Brown, Stephanie Pfaender
## Abstract
Given the recurring threat of coronavirus outbreaks, understanding the specificity of coronaviruses in terms of their host, tissue, and cell tropism is crucial. This review consolidates and critically assesses the current literature on the tropism of endemic, epidemic, and pandemic coronaviruses. We explore different levels of tropism, including species tropism (virus preference for specific host species), host cell tropism (virus specificity for particular cell types), and tissue tropism (specificity for certain tissues or organs). This review compiles extensive basic research, particularly from recent years, to provide critical insights into the viral mechanisms that are key to improving future pandemic preparedness.
## SPECIES TROPISM
CoVs exhibit diverse host species tropism, which plays a crucial role in their pathogenesis and transmission (Fig. 1B). CoV tropism is defined by both susceptibility, determined by the presence of specific host receptors required for viral entry, and permissiveness, which refers to the capacity of a host cell to support productive infection, including the expression and activity of necessary host dependency factors. Conversely, the presence of cell intrinsic restriction factors may limit productive infection even in susceptible cells, underlining the complex interplay of host determinants that define cellular tropism (26). Different CoVs use distinct entry receptors, which vary not only in their expression across host species but also in their structural and biochemical properties. These differences, including variations in receptor-binding affinities, significantly influence host tropism and cross-species transmission potential (Table 1).
Bats are suspected as the origin for most HCoVs, with the exception of lineage A beta-CoVs, which may have reservoirs in rodents (27). Through cross-species transmis sion and adaptation, these viruses have acquired the ability to infect different hosts, most notably humans, emphasizing the complexity of CoV evolution and their ability to cross species barriers.
Recent estimates indicate that the endemic HCoVs have emerged from zoonotic reservoirs within the last 1,000 years but have since undergone evolutionary adapta tions, establishing humans as their primary hosts. Phylogenetic analyses suggest that HCoV-229E originated from bat CoVs (28), with transmission to humans occurring via an intermediate host, most likely camelids (29). Analysis of circulating HCoV-229E S and N genes highlights signatures of genetic drift, positive selection, and increasing divergence over time in human populations (30). Mice are resistant to HCoV-229E infection, although experimental studies suggest rodents can be made susceptible upon genetic modifica tion (31).
The beta-CoV HCoV-OC43 is suspected to have originated from rodents (10), with bovines proposed as a likely intermediate host. A zoonotic transmission to humans may have occurred around 1890, coinciding with the "Russian flu" pandemic, although the exact causative agent-whether HCoV-OC43, influenza A virus, or another patho gen-remains uncertain to this day (32). Interestingly, a recent study revealed a close evolutionary relationship and genomic homology between HCoV-OC43 and a porcine CoV, indicating a shared evolutionary origin: this was further supported by the sus ceptibility of porcine intestinal organoids to HCoV-OC43 infection (33). Experimental studies have shown that HCoV-OC43 can infect mice, rapidly gaining virulence in the murine brain (34,35). Together, these findings suggest that either pigs or rodents could potentially serve as natural hosts for ancestral HCoV-OC43-like viruses, prior to its spillover and adaptation to humans.
The animal reservoir of the alpha-CoV HCoV-NL63 remains elusive, although phylogenetic evidence suggests divergence from an HCoV-229E ancestor approximately 1,000 years ago and continued circulation in humans for centuries (36). Closely related viruses have been identified in various bat species, suggesting a potential zoonotic origin from bats (37,38), which is further supported by permissiveness of bat cells to HCoV-NL63 infection (37). Similar to HCoV-229E, mice can be engineered to be susceptible to HCoV-NL63 infection via genetic modification (31). The most recently identified endemic circulating HCoV, HKU1, is thought to have originated from rodents (39), possibly via direct transmission to humans (40). A recent study has identified transmembrane protease serine 2 (TMPRSS2) as host receptor for HCoV-HKU1 (41) and revealed that TMPRSS2 orthologs from at least five mammalian orders including rodents support HCoV-HKU1 entry into cells (42,43). These data support a possible rodent origin of HKU1 and implicate various species as potential reservoirs or intermediate hosts.
The initial spillover of SARS-CoV to humans highlighted the potential of CoVs to cause severe disease and initiated efforts to understand their emergence, host range, and transmission dynamics. The first SARS-CoV infections were linked to animal wet markets in China, so the hypothesis quickly arose that the virus may have crossed the species barrier, transmitting from an animal host to humans (44). Sampling of various animals sold at Chinese live animal markets for the presence of virus and neutralizing antibod ies demonstrated a broad host range. Indeed, susceptibility of Himalayan palm civets, raccoon dogs, Chinese ferret badgers, hog-badgers, domestic cats, beavers, Chinese hares, and Chinese muntjacs to SARS-CoV infection was confirmed (45,46). Studies have confirmed that the SARS-CoV epidemic lineage was likely introduced to humans via masked palm civets due to highly homologous (99.8%) viral genome identities from nasal swabs from palm civets (47), with rhinolophid bats suspected to be the zoonotic reservoir species (48)(49)(50)(51). Since then, studies have identified multiple SARS-like CoVs in a range of bat species (48,(52)(53)(54). While bats are the likely natural source of SARS-CoV, there is still a genetic gap concerning the amplification host, in which likely recombina tion occurred that facilitated the species-jump toward humans. It is currently unclear where, when, and in which animal species this recombination could have occurred (55).
The first infection events of SARS-CoV-2 were epidemiologically traced to the Wuhan Huanan Seafood Wholesale Market, where various live animals were sold (56). This is further supported by a recent study tracing the genetic signature of market wildlife and SARS-CoV-2 positivity, identifying the presence of SARS-CoV-2 in stalls containing wildlife DNA from various animal species, including civets, bamboo rats, and raccoon dogs (57). Phylogenetically related viruses to SARS-CoV-2 were identified in various species, including BANAL-20-52 from Malayan horseshoe bats, RaTG13 derived from the intermediate horseshoe bat, and pangolin-CoV identified in Malayan pangolins (58)(59)(60). A recent study implied that the closest-related bat virus ancestors of SARS-CoV and SARS-CoV-2 existed less than a decade prior to their emergence in humans (61). SARS-CoV-2 can infect a wide range of hosts, including dogs, mink, ferrets, otters, hamsters, voles, deer, deer mice, bats, felines, mice, and several nonhuman primates, while the virus cannot replicate in pigs, chickens, and ducks (62)(63)(64). The golden Syrian hamster is now considered the gold-standard animal model to study pathogenesis and for antiviral testing, as it is highly susceptible to infection and recapitulates clinical disease symptoms seen in humans (65)(66)(67). Similar to SARS-CoV, mice are not naturally susceptible to SARS-CoV-2 infection but can be genetically modified to support viral replication (68,69). However, some SARS-CoV-2 variants of concern have acquired mutations that enhance binding to mouse angiotensin-converting enzyme 2 (ACE2), rendering wild-type mice partially or fully susceptible to infection (70). Of note, there are significant genetic and phenotypic differences between pre-omicron and omicron SARS-CoV-2 variants, which reflect their evolving virological and clinical characteristics.
For MERS-CoV, continuous circulation of this virus within dromedary camel popula tion likely facilitates zoonotic cross-species transmissions to humans. A comprehensive study reported MERS-CoV seropositivity in dromedary camels in Africa, the Middle East, and Asia (71-74), while horses, sheep, and goats were all seronegative (75). Addition ally, the virus can readily replicate in primary camelid airway cultures (76). Although dromedary camels are suspected as the primary zoonotic reservoir for MERS-CoV, several lines of evidence implicate bats as ancestral reservoir hosts. A number of phyloge netically related viral isolates, including BtCoV-HKU4 and BtCoV-HKU5, NeoCoV, and PDF 2180, have been identified in various Vespertilionidae bat species (77). Mice are not naturally susceptible to MERS-CoV infection but can be genetically engineered to support replication (78).
In summary, HCoVs exhibit distinct host ranges: epidemic and pandemic HCoVs are generalists that can infect a variety of different mammals, while endemic seasonal HCoVs are restricted to humans. SARS-CoV and MERS-CoV are not specifically adapted to humans due to recent cross-species transmission events and inefficient human-tohuman transmission. In contrast, highly efficient human-to-human transmission allows SARS-CoV-2 to continue to circulate in humans, with a marked reduction in pathoge nicity observed as the virus becomes endemic (79). Phylogenetic evidence highlights a likely zoonotic origin for all HCoVs, suggesting potential circulation back and forth between different species, facilitating recombination, adaptation, and potentially driving emergence in novel hosts. Indeed, altered pathogenesis or transmission characteris tics could facilitate cross-species transmission events associated with more efficient replication in humans or other mammals. Thus, it is crucial to understand mechanisms both limiting and driving CoV host-switching to prepare for future zoonotic spillover events.
## HOST CELL TROPISM
Successful CoV attachment and subsequent internalization into host cells represent the first stage of the infection process, the first layer of virus-host interactions and a determinant of host, tissue, and cellular tropism. On the virus side, this process is mediated by the envelope-anchored spike glycoproteins. For all CoVs, these glycopro teins are presented as trimeric complexes, which decorate the virion surface and give virions their characteristic "corona" appearance when visualized by electron microscopy (80). These proteins are heavily glycosylated and decorated with N-linked carbohydrate moieties, which are important for entry and immune evasion. CoV spike proteins are type I fusion proteins that form homotrimers (81). Each monomer is typically 1,200-1,400 amino acids in length and consists of three segments-a short intracellular tail, a transmembrane anchor, and a large ectodomain. This ectodomain consists of the receptor-binding S1 subunit and the membrane-fusion-promoting S2 subunit. In addition, the S1 subunit can be further subdivided and contains two independent domains, an N-terminal domain and a C-terminal domain (S1-CTD), both of which can function as receptor-binding domains (RBD) that recognize cell surface molecules (82). Structural studies have revealed that spike proteins undergo extensive conformational changes and structural rearrangements, as well as cleavage, during binding to their cognate receptors. These events precede internalization or virus-host membrane fusion (83)(84)(85).
## Attachment factors
Diverse host factors mediate viral attachment to the cell membrane prior to primary receptor engagement. These factors localize the virus near the plasma membrane to facilitate receptor binding and can also induce spike conformational changes. Both HCoV-OC43 and HCoV-HKU1 bind to acidic carbohydrate 9-O-acetylated sialic acid (86,87). Indeed, sialoglycan binding to the HCoV-HKU1 spike S1 domain triggers confor mational changes in spike, which are required for subsequent receptor binding (88,89). Similarly, MERS-CoV spike binds to sialic acid (90), and HCoV-NL63 spike binds to heparan sulfate proteoglycans (91). In contrast, SARS-CoV-2 attaches to cells via C-type lectin receptors L-SIGN, DC-SIGN, and SIGLEC1 (92). Omicron variants exhibit significantly stronger attachment to host cell membranes, primarily due to increased affinity for the co-receptor heparan sulfate (93). In all cases, attachment factor engagement enhances receptor-mediated entry.
## Primary entry factors
To date, four distinct plasma membrane-anchored enzymes have been identified to act as primary receptors for human endemic, epidemic, and pandemic CoVs (Fig. 2A). Of note, all four described CoV receptors are proteolytically active. However, this protease activity is not required for successful receptor engagement. Indeed, CoV spike proteins bind to proximal regions of these proteases with minimal disruption of enzymatic activity.
## Angiotensin-converting enzyme 2
HCoVs -NL63, SARS-CoV, and SARS-CoV-2 all engage ACE2 as their primary receptor to facilitate cell entry (60,(104)(105)(106). ACE2 is a heavily glycosylated type I transmembrane protein and is part of the renin-angiotensin system, regulating blood pressure and electrolyte balance through coordinated effects on the heart, blood vessels, and kidneys (42,(94)(95)(96)(97)(98). Structural representations of host receptors were created with AlphaFold and UCSF ChimeraX (99,100). Spike protein colors match the viruses from Fig. 1A andB, and receptor colors match the following panels. Interacting residues in the binding of spike protein and receptor are depicted in spherical style. (B) Illustration of the abundance of receptors for HCoVs within the nasal respiratory tract, trachea, intermediate, and distal bronchia. The anatomical overview highlights primary infection sites, with magnified insets depicting the normalized transcript expression (101,102). Representations of the human receptors at the cellular level are shown in the top inset and retrieved as described for panel A (99,100). Correctly and incorrectly predicted folding of transmembrane domains are colored yellow and red, respectively. For detailed information on the different receptors and entry factors, as well as a mapping of the individual HCoVs to their respective receptors, see Host Cell Tropism. (C) Normalized transcript expression for secondary infection sites is presented as described in panel B (103). Brain data are comprised from hippocampal formation, amygdala, basal ganglia, midbrain, spinal cord, cerebral cortex, cerebellum, hypothalamus, and choroid plexus, and gastrointestinal data from stomach, small intestine, colon, rectum, and duodenum. The figure was created with BioRender.com. (107). The ACE2 ectodomain has a zinc-dependent carboxypeptidase, which hydrolyzes angiotensin II to form Ang-(1-7), promoting vasodilation and inhibiting fibrosis and inflammation (108).
Cryo-EM studies of authentic SARS-CoV-2 virions indicate that spike proteins form trimers with two flexible hinges in the stalk domain, enabling cell surface scanning to locate receptor complexes and initiate cell entry (80). A cryo-EM structure for the SARS-CoV-2 spike trimer in a prefusion conformation confirms the three RBDs are located at the head of the globular trimer and show a single RBD in an "up" conformation, ready for receptor engagement (109). The crystal structure of the SARS-CoV RBD bound to the peptidase domain of human ACE2 highlights binding to the N-terminal lobe of the peptidase (94). Interestingly, trimer structures and RBD engagement of ACE2 for SARS-CoV and SARS-CoV-2 are similar, although SARS-CoV-2's binding affinity for ACE2 is reportedly 10-20-fold higher than for SARS-CoV (109). A distinguishing feature of the omicron variant is its increased binding affinity to the human ACE2 receptor compared to the wild-type virus, which is attributed to multiple mutations in the RBD of the spike protein (110). The SARS-CoV-2 RBD, located in the S1 domain of spike, is composed of five stranded antiparallel betasheets, stabilized by four pairs of disulfide bonds. Crystallographic comparison of SARS-CoV and SARS-CoV-2 RBDs: ACE2-binding interfaces revealed nonidentical but overlapping binding profiles. Of the 20 ACE2 residues that interact with either the SARS-CoV or SARS-CoV-2 RBDs, 17 residues are shared between both interactions (94). These similar binding profiles likely reflect shared inheritance from a common ancestor rather than convergent evolution, as both viruses are genetically divergent, but likely ACE2 receptor usage is a conserved property among their distinct progenitor lineages (94). HCoV-NL63 similarly utilizes ACE2 as its cellular receptor. While the S1 domains of HCoV-NL63 and SARS-CoV are quite different, they both associate with a region of human ACE2 that includes a key loop formed by beta strands 4 and 5. However, the S protein interaction of HCoV-NL63 with ACE2 is uniquely sensitive to residue 354, unlike the S protein of SARS-CoV, which is only modestly affected. This suggests that while both viruses bind overlapping regions of ACE2, HCoV-NL63 relies more heavily on the specific conformation or interactions for efficient receptor binding (111).
High sequence homology of SARS-CoV and SARS-CoV-2 has been described for several bat isolates. SARS-CoV-like viruses, RsSHC014 and Rs3367, were recovered from Chinese horseshoe bats, and bat virus SL-CoV-WIV1, isolated from bat fecal samples and propagated on VeroE6 cells, was able to utilize bat, civet, and human ACE2 for cell entry (54) (please refer to Species Tropism). ACE2 orthologs from 46 bat species exhibit different levels of interaction with the spike proteins of SARS-CoV and SARS-CoV-2. ACE2 orthologs from 24, 21, and 16 bat species were resistant to infection with either SARS-CoV, SARS-CoV-2, or both viruses, respectively, indicating not all bat species can act as hosts for these viruses (112). ACE2 orthologs from a further 48 mammalian species, including domestic animals, pets, and zoo animals, were also investigated for their ability to bind SARS-CoV-2 spike (beta variant [B.1.351]), with 44 species supporting viral entry (113). Of note, mutations at residues H41 and E42 in ACE2 orthologs from New World monkeys disrupt their ability to bind the viral spike protein and explain their resistance to SARS-CoV-2. Viruses closely related to SARS-CoV-2 are also reported to circulate in wild pangolins, which are predicted to bind to human and pangolin ACE2 (114). These animals are heavily trafficked and therefore represent a risk for zoonotic spillover events. Indeed, subsequent studies have confirmed that pangolin CoVs can utilize human ACE2 as a receptor and can cause severe disease in K18-hACE2 transgenic mice (115,116). A novel mink respiratory CoV also utilizes ACE2 as a receptor and can additionally utilize bat, monkey, and human orthologs to enter cells, binding to the same interface as SARS-CoV-2 (117).
Initially, merbecoviruses were considered to exclusively utilize dipeptidyl-peptidase 4 (DPP4) to initiate cell infection (see below). However, multiple independent studies have identified a range of merbecoviruses that can use ACE2 for cell uptake. Bat CoVs PDF-2180 and NeoCoV, close relatives of MERS-CoV, reportedly bind efficiently to bat ACE2 (and less efficiently, human ACE2), and cryo-EM identified distinct noncanon ical ACE2:RBD binding interfaces which involve protein-glycan interactions (118). Two European bat CoVs related to MERS-CoV, MOW15-22 and PnNL2018bb, engage ACE2 through distinct surface-binding regions but cannot bind human ACE2 and exhibit a narrow species range (119). The related merbecovirus HKU5 also enters cells of Pipistrellus bats and other mammals via ACE2 engagement (119), and a novel lineage 2 isolate, HKU5-CoV-2, was also able to infect human-ACE2 expressing cells (120). These studies highlight ACE2 utilization evolved independently on multiple occasions among both sarbecoviruses and merbecoviruses.
Together, these investigations highlight high promiscuity and broad ACE2 receptor usage for divergent endemic, epidemic, pandemic, and nonhuman CoVs. These data also underscore the requirement for continued virological surveillance-including systematic sampling, sequencing, and characterization-of ACE2 utilizing CoVs in wild bats, farmed mink, and heavily trafficked species such as pangolins. While such efforts may not directly lead to vaccines being stockpiled in advance, they enable early risk assessment, guide targeted virological and ecological studies, inform policies on wildlife trade and farming practices, and allow for the rapid development of diagnostics, antivirals, and vaccines based on pre-existing knowledge.
## Aminopeptidase N
Aminopeptidase N (APN; a.k.a. CD13) is the primary receptor used by HCoV-229E to enter cells (121). APN represents an ~150 kDa type II transmembrane protein, which exhibits extensive N-linked glycosylation and contains a zinc-containing aminopeptidase ectodomain. APN plays a role in a range of physiological processes, including pain sensation, blood pressure regulation, tumor angiogenesis and metastasis, immune cell chemotaxis, sperm motility, and cell adhesion (122).
X-ray crystallographic and cryo-EM studies indicate that HCoV-229E spike RBD:APN receptor binding is mediated by the interaction of three RBD loops that are located in the S1-CTD (95,123,124). The RBD must flip to an "up" position to engage APN (95). Phylogenetic analysis of RBD loops from deposited HCoV-229E database sequences identified six RBD classes, with RBD sequences representing the most variable regions of the viral genome. Indeed, conformational plasticity in these regions drives differences in APN-binding affinity and antibody recognition observed between circulating strains (123).
APN orthologs from multiple species also serve as receptors for pathogenic CoVs infecting nonhuman species, including transmissible gastroenteritis virus (TGEV -infecting pigs) (125), feline infectious peritonitis virus, and feline enteric CoV (126). Studies of chimeric human, porcine, and feline APN glycoproteins reveal APN receptor and species specificity is determined by two crucial regions (127-129), with differences in N-linked glycosylation of APN also representing important determinants of species range (123,127). These studies confirm broad APN receptor usage by distinct alpha-CoVs with different routes of transmission and associated pathology-HCoV-229E has relatively mild respiratory symptoms in humans, while TGEV infects the gastrointestinal tract and causes fatal diarrhea in piglets. Combined, these studies highlight the potential for cross-species transmission of highly pathogenic animal CoVs to humans based on shared APN receptor usage, although barriers and incompatibilities exist.
## Dipeptidyl-peptidase 4
DPP4 (a.k.a. CD26) is the primary receptor used by MERS-CoV for cell entry (130). DPP4 is a type II transmembrane glycoprotein with a short cytoplasmic tail and a large extracellu lar region composed of a short flexible segment, separate glycan-rich and cysteine-rich domains, and a C-terminal serine-protease domain with catalytic activity that mediates cleavage and inactivation of a range of circulating regulatory peptides (131). Crystallographic and cryo-EM studies have provided structural insights into MERS-CoV spike RBD:DPP4 binding, identifying domains, interfaces, and critical residues involved in this interaction (94,132). A 3.0 Å-resolution crystal structure of the MERS-CoV RBD in complex with soluble DPP4 reveals that the RBD directly interacts with the beta-propeller domain of DPP4 through two major patches, which do not overlap with the C-terminal protease domain (96). Cryo-EM structures of prefusion trimeric MERS-CoV spike in complex with DPP4 reveal the RBD can occur in two states: "standing" or "lying. " This dynamic and flexible RDB contributes to efficient DPP4 recognition. Only "standing" RDBs on monomeric spike trimer subunits can bind to DPP4 (132).
MERS-CoV likely originates from bats, with DPP4 orthologs from 16 bat species supporting MERS-CoV infection to varying degrees (133). Serial passage of MERS-CoV on cells expressing a suboptimal bat DPP4 variant resulted in the accumulation of spike mutations that boost entry, highlighting the rapid adaptability of MERS-CoV to improve virus-host receptor interactions which can occur by multiple mutational pathways (133). MERS-CoV cannot utilize murine, hamster, or ferret DPP4 for cell entry (134,135). Despite its zoonotic origin, MERS-CoV spike preferentially binds to human DPP4 over its bat ortholog to enter cells, while bat CoV HKU4-CoV can only efficiently utilize bat DPP4 (136). A bat MERS-like CoV similar to HKU4-CoV circulates in Malayan pangolins and binds to human, bat, and pangolin DPP4. This MjHKU4r-CoV-1 virus is infectious in human airway and intestinal organoids, and pathogenic in hDPP4 transgenic mouse lungs (137). Furthermore, Ty-BatCoV HKU4 was isolated from lesser bamboo bats and was shown to utilize DPP4, with replication and cytopathology reported in human cells and hDPP4 transgenic mice (138). Together, these studies demonstrate the rapid adaptability of MERS-CoV to novel host DPP4 utilization and the circulation of multiple DPP4-utilizing viruses in both bats and rodents, highlighting the potential for future spillover events to humans and the requirement for continued surveillance.
## Transmembrane protease serine 2
TMPRSS2 serves as a functional receptor for HCoV-HKU1, required for both viral entry and spike-mediated cell-cell fusion (41). TMPRSS2 is a type II transmembrane protein with a cytoplasmic tail domain. The large extracellular portion contains low-density lipoprotein receptor class A domain, a scavenger receptor cysteine-rich domain, and a C-terminally encoded serine protease domain (139). The serine protease activity is not required for HCoV-HKU1 entry, and catalytically inactive TMPRSS2 can still facilitate HCoV-HKU1 entry via an endosomal route (41). In addition to its role as an entry receptor for HCoV-HKU1, TMPRSS2 is also involved in proteolytic priming of diverse CoV spike glycoproteins to enhance viral uptake.
Crystallographic and cryo-EM studies have determined the virus:host molecular interaction interface at high resolution (42,89,140,141). Zymogenic TMPRSS2 under goes autocleavage to become fully active, inducing conformational changes in three activation loops that increase HCoV-HKU1 RBD-binding affinity (141). The crystal structure of the HCoV-HKU1 RBD-TMPRSS2 complex highlights that in trimeric spike, at least one RBD must be in a "up" confirmation to allow binding to the periphery of the catalytic groove of the TMPRSS2 serine protease domain (141). Cryo-EM studies further confirm sialoglycan binding induces conformational changes that promote RBD opening, enabling spike binding to TMPRSS2 via multiple key residues and interfaces (42,89). HCoV-HKU1 utilizes glycan shielding and conformational masking to evade host humoral responses while maintaining TMPRSS2 engagement, illustrating general immune evasion strategies that can complicate vaccine design for HCoVs (42).
HCoV-HKU1 genotypes A and C share a highly conserved mechanism for sequen tial binding of sialoglycans and TMPRSS2, suggesting a universal mode of receptor recognition across different circulating strains (89). TMPRSS2 residues D417 and Y469 are reported as critical for human specificity and determinants of the narrow host range of HCoV-HKU1 (141,142). Despite this restricted tropism, the HCoV-HKU1 RBD footprint of TMPRSS2 contact residues is largely conserved among diverse mammals. Indeed, HCoV-HKU1 spike-mediated cell entry is facilitated by TMPRSS2 orthologs from distinct mammalian orders, including primates, rodents, artiodactyls, lagomorphs, and bats (42) (please refer to Species Tropism). This RBD footprint is only partially conserved in both reptiles and birds and minimally conserved in amphibians and other vertebrates. These observations support recent findings indicating cell entry is not the major barrier which limits viral cross-species transmissions, with downstream incompatibilities in novel host cells representing a major obstacle (143,144).
## Accessory proteases
## Furin
Furin is a ubiquitously expressed calcium-dependent serine protease, which localizes in the Golgi apparatus where it cleaves precursor proteins at their basic amino acid processing site into mature or active forms. A polybasic (PRARR) insertion in the SARS-CoV-2 S1/S2 spike junction allows proteolytic processing by furin during virion egress from infected cells (105). Furin cleavage at S1/S2 exposes the C-terminus of S1 and facilitates subsequent spike binding to neurophilin-1, which ultimately primes the spike protein for enhanced host cell infection. This is specific for SARS-CoV-2 (145,146). Indeed, the presence of the furin cleavage site in SARS-CoV-2 confers a selective advantage in lung cells, human airway epithelial cultures, and is required for efficient ferret-to-ferret transmission (84), while furin inhibition has been shown to suppress spike-mediated syncytia formation in SARS-CoV-2-infected cells (147). Still, evolutionary analysis of diverse CoV spike proteins suggested that furin cleavage motifs have arisen independently on multiple occasions in the family Coronaviridae (148). Omicron carries three mutations close to the furin cleavage site (P681H, H655Y, and N679K), which reduce the efficiency of spike protein cleavage at the S1/S2 junction (149). Furin is also co-opted by unrelated viruses, including mosquito-transmitted orthoflaviviruses, where the chaperone protein prM is cleaved by furin at a conserved polybasic motif to facilitate envelope dimerization and virion maturation during particle morphogenesis (150).
## TMPRSS2 and cathepsin L
After receptor engagement, which induces conformational changes in the spike protein, there are two distinct cellular locations at which CoV-host membrane fusion can potentially occur: at the cell surface or in endosomal compartments. TMPRSS2 is localized to the plasma membrane, and co-opting of TMPRSS2 to enhance virus infection was initially described for influenza, where hemagglutinin cleavage for proteolytic activation was demonstrated (151). Subsequently, TMPRSS2 was shown to be a broad enhancer of CoV infection, augmenting uptake into permissive cells of SARS-CoV (152), HCoV-229E (153), MERS-CoV, and SARS-CoV-2 (154)(155)(156)(157). The abundant expression of TMPRSS2 in small intestinal enterocytes and hepatocytes contributes to SARS-CoV-2 tropism for intestinal cells and the liver, while the ability to enhance SARS-CoV-2 entry is conserved among TMPRSS2 orthologs from diverse mammalian orders (154,156,157).
TMPRSS2's enzymatic activity cleaves spike at S2′ to expose the fusion peptide and facilitate membrane fusion. TMPRSS2 infection enhancement is dependent on its serine protease activity, which can be blocked pharmacologically or by mutational deletion of the HDS catalytic triad (157,158). In the presence of TMPRSS2, the SARS-CoV-2 spike protein is proteolytically cleaved at S2′ at the cellular surface. Under these conditions, the virus enters the cell within 10 minutes in a pH-independent manner, bypassing the endosomal route (159). If the cell expresses insufficient amounts of TMPRSS2, the entire virus-receptor complex is internalized via clathrin-mediated endocytosis (160) into endolysosomes, where spike cleavage is mediated by CTSL, a cysteine protease that functions in protein catabolism and requires an acidic environment for proteolytic activity (159,161). This step takes up to 60 minutes post infection. Therefore, TMPRSS2 expression levels are proposed to determine which route the virus utilizes to enter ACE2-expressing cells, with TMPRSS2-primed entry more efficient than endosomal CTSL priming. More recently, EM visualization and quantification of early SARS-CoV-2 entry steps showed TMPRSS2-mediated enrichment of internalized virions into endosomal compartments, which was unexpected and requires further investigation (157). Together, the dual functions of TMPRSS2, serving as both a primary receptor and entry enhanc ing co-factor, highlight its importance for CoV entry, in general, and its potential as a therapeutic target.
While numerous post-entry factors can influence the efficiency of CoV replication, these alone are generally insufficient to determine cellular susceptibility. For many viruses, cellular entry is not the primary barrier to productive infection. However, this does not appear to apply to CoVs, for which the presence and availability of the appropriate entry receptor remain the dominant determinants of host cell permissive ness (124). This underscores the unique reliance of CoVs on receptor-mediated entry as a critical gatekeeper of infection. Follow-up studies will likely reinforce these findings, providing further insights into the essential role of entry receptors in CoV infectivity.
## Suspected additional receptors
While bona fide receptors for human CoVs are well characterized and undisputed, multiple additional proteins have been identified as potential receptors or entry co-factors, mostly for SARS-CoV-2. Examples include glucose-regulated protein 78, a molecular chaperone, which was proposed as an auxiliary factor that facilitates SARS-CoV-2 entry by forming a complex with the spike protein and ACE2 (162,163). Addi tionally, the receptor tyrosine kinase AXL was identified as a candidate receptor that promotes infection of pulmonary and bronchial epithelial cells, while the glutamyl-ami nopeptidase was proposed as a co-receptor due to its co-expression with ACE2 in various tissues (164,165). Angiotensin-II receptor type 2 (AGTR2, a G protein-coupled receptor) and basigin/CD147 were initially described but are now considered unlikely candidates for SARS-CoV-2 receptors. Confirmation of AGTR2 as an entry factor was not supported by additional studies (166), and the role of CD147 remains unclear, with conflicting data on its ability to mediate viral entry (167)(168)(169). Similarly, TMEM106B was initially described as a proviral host factor for SARS-CoV-2 (170) and subsequently shown to bind directly to the spike RBD and mediate ACE2-independent entry into cells (171). However, more recently, while TMEM106B-mediated infection was confirmed to mediate ACE2-independent entry in vitro, this phenotype could not be recapitulated in vivo (172). In summary, these findings highlight the potential complexity of the CoV-host interac tome during the entry process but also underscore the need for further basic research to validate these proteins in the CoV infection process before definitive roles can be assigned. In addition, CoV utilization of nonhuman orthologs in susceptible reservoir species should also be confirmed.
## TISSUE TROPISM
HCoVs are primarily recognized as respiratory pathogens and mainly transmitted via respiratory droplets and aerosols, making the respiratory tract the first and most common site of viral entry (173). Virions are inhaled into the upper and lower respira tory tracts, encountering epithelial cells lining the nasal passages, throat, and lungs. Consequently, the respiratory system serves as the primary gateway for HCoV infection, facilitating viral replication and the subsequent spread to other tissues and organs (174). Of note, pathogenic SARS-CoV, MERS-CoV, and SARS-CoV-2 can be disseminated systemically, affecting multiple organs beyond the lungs (Fig. 2B andC). The predomi nance of apical-only release for SARS-CoV and SARS-CoV-2 suggests these viruses are unlikely to breach the epithelial barrier and enter circulation. In contrast, basolateral release, as seen with MERS-CoV (and certain animal CoVs like mouse hepatitis virus, canine CoV, and feline CoV), is associated with systemic spread. Incorporating these distinctions can clarify why some CoVs remain localized while others may disseminate (175). It is important to note that the evidence of coronavirus presence in extrapulmo nary tissues primarily comes from in vitro studies or post-mortem analyses of patients with severe disease, often complicated by underlying conditions. Thus, such findings may be artificial or reflect late-stage dissemination facilitated by tissue damage and may not represent typical viral tropism during average infections, underscoring the need for cautious interpretation.
## Respiratory tract
Most HCoVs begin their infection in the nasal epithelium, where the virus first establishes itself. Figure 2 illustrates the expression profiles of host receptors that mediate binding of the seven HCoVs within tissues of the nasal, pharyngeal, and distal respiratory tract. Endemic HCoV-229E primarily infects non-ciliated cells, whereas HCoV-OC43, -NL63, and -HKU1 preferentially infect ciliated cells in the respiratory tract (176)(177)(178). HCoV-HKU1 has been additionally shown to infect primary human alveolar type II cells (13). HKU1 recognizes TMPRSS2, facilitating viral entry, which is highly expressed in small airway epithelium and nasal epithelium in contrast to a lower expression in masticatory mucosa (179), as illustrated in Fig. 2B, potentially contributing toward a preferential cellular susceptibility within the nasal mucosa. Interestingly, HCoV-NL63 utilizes ACE2 for cell entry, like the more pathogenic SARS-CoV and SARS-CoV-2. Thus, it shares some of the cellular tropism; however, HCoV-NL63 usually induces only mild or moderate respiratory disease. Clinical data indicate that immune control could partly determine pathogenicity, as infants and immunocompromised individuals experience more severe disease with HCoV-NL63 (37,177,180). Additionally, differences in ACE2 interactions could contribute to differential pathogenicity (106). ACE2 expression is higher in the nasal epithelium, gradually decreasing toward the lung, influencing tropism (181, 182) (Fig. 2B). Infection with SARS-CoV and SARS-CoV-2 can cause significant damage to the lung epithelium. This, combined with a heightened inflammatory response, often leads to alveolar injury and the development of acute respiratory distress syndrome, which can progress to respiratory failure (183,184). Within the respiratory system, SARS-CoV and SARS-CoV-2 have been shown to infect ciliated epithelial cells, goblet cells, and alveolar type II cells (113,185,186). MERS-CoV similarly primarily infects tissues of the respiratory tract. The virus enters human cells via interaction with the DPP4 receptor, which is widely expressed on the surfaces of respiratory epithelial cells, particularly in the lungs (187), as illustrated in Fig. 2B. On a cellular level within the respiratory tract, specifically goblet cells have been shown to support viral replication (188). The clinical spectrum can range from asymptomatic or mild respiratory disease to severe pneumonia including ARDS but also multiorgan failure (189,190).
## Central nervous system
All seven human-infecting CoVs have been associated with central nervous system (CNS) dysfunctions (191). Reports of encephalitis associated with seasonal hCoVs are rare and largely anecdotal and are often limited to immunocompromised or pediatric patients. For example, HCoV-229E has been described to have the capacity to infect neuroblastoma, neuroglioma, and oligodendrocytic cells, which is likely related to the high expression of its receptor aminopeptidase N in the brain (192) (Fig. 2C). HCoV-OC43 is a neurotropic virus that can invade the central nervous system and primarily infects neurons, as has been demonstrated in both human and mouse models. Infection of neurons can lead to neurodegenerative effects that contribute to cell stress and apoptosis. This neuroinvasive ability has been associated with neurological complica tions, such as encephalitis and neurodegenerative diseases (34,35). HCoV-NL63 also appears to be neurotropic in very rare cases. The virus is able to infect mononuclear circular cells, and few independent cases of encephalopathy are reported (193,194). ACE2 is expressed by neuronal and glial cells in the brainstem. Expression has also been detected in the amygdala and cerebral cortex, with the highest levels observed in the pons and medulla oblongata (195,196). The more pathogenic hCoVs have been more clearly associated with neurological complications; however, even for these viruses, encephalitis remains a rare complication. For SARS-CoV-2, neurological manifestations were reported, including encephalopathy and anosmia (loss of smell), suggesting that SARS-CoV-2 can affect the CNS (197,198). Compared to SARS-CoV, SARS-CoV-2 is associated with a broader spectrum and greater frequency of neurological symptoms. This may be due to its higher affinity for ACE2, as described in Host Cell Tropism, ' and the additional use of co-receptors such as neuropilin-1, which is highly expressed in nervous tissue (199). SARS-CoV-2 also has a greater neuroinvasive potential, possibly entering the CNS via the olfactory nerve, hematogenous spread, or infected immune cells (200,201). In addition, it triggers greater systemic inflammation and endothelial dysfunction, contributing to cerebrovascular complications (202). High levels of DPP4 protein have been observed in the fetal and perinatal human brain, particularly within neuroblasts, neurons, brain capillaries, the ependymal lining, and the choroid plexus. In adulthood, however, DPP4 mRNA expression in the brain is markedly lower, especially when compared to organs such as the placenta, kidneys, lungs, and liver (191). Neurolog ical involvement in MERS-CoV infections is rare but has been reported in a small number of cases, with symptoms ranging from confusion to encephalitis. However, the precise mechanism underlying these pathways remains incompletely understood.
## Liver, kidney, and gastrointestinal tract
HCoV-229E replicates in Huh7 cells, an immortalized human liver cell line commonly used as the standard for infection experiments with this virus (203). In vivo, however, there is no evidence of liver tropism. The virus is also associated with gastrointestinal symptoms, including abdominal pain, diarrhea, and vomiting. These symptoms are thought to be due to the ability of the virus to infect next to respiratory epithelial cells and also intestinal epithelial cells. It should be mentioned that the presence of viral RNA in stool could reflect passive transit rather than local replication, unless supported by additional evidence such as detection of viral proteins, infectious virus, or histopatho logical findings from gastrointestinal tissue. Virus particles resembling HCoV-OC43 have been detected in stool samples from infected patients, suggesting active replication in the intestinal mucosa (204). HCoV-HKU1 has also been detected in stool samples, and the virus is associated with symptoms of the gastrointestinal tract (205). In cell culture, HCoV-HKU1 is very different from the other HCoVs. It has not yet been possible to grow the virus in immortalized cell lines; only cultivation in primary human respiratory epithelial cell cultures has been successful (206). For HCoV-NL63, viral RNA has also been detected in a few stool samples from children with acute gastroenteritis (204,207). In the laboratory, LLCMK2 cells, an immortalized cell line derived from epithelial cells of primate kidney, are used to propagate this virus (208). A recent study demonstrated differential susceptibility and replication kinetics for HCoV-229E, -NL63, and -OC43 in various cell lines, with HCoV-OC43 alone being able to replicate in extra-pulmonary tissues, including human colon cancer cells and African green monkey kidney cells (209). It has been shown that SARS-CoV RNA is not only found in the lungs, bronchi, and trachea but also in the stomach, small intestine, sweat glands, pancreas, liver, and adrenal glands (210,211). Indeed, a high density of ACE2 receptors has been described within the kidneys and gastrointestinal tract (212) (Fig. 2C). In addition, in situ hybridization implicated the presence of the virus in various tissues, specific in the epithelial cells of mucosa of the small and large intestine, in the stomach and in the esophagus of the digestive tract, in the distal tubular epithelium within the kidney, and hepatocytes in the liver (185). SARS-CoV-2 has been shown to affect the kidneys, which can lead to an acute kidney injury. ACE2 is expressed on proximal tubules, parietal epithelial cells, mesangial cells, and podocytes (213). Also, gastrointestinal involvement is a recognized feature of SARS-CoV-2 infection. The virus can infect enterocytes via ACE2, which is highly expressed in the small intestine. Possible symptoms include diarrhea, nausea, vomiting, and abdominal pain, which may occur even in the absence of respiratory manifestations. Viral RNA has been detected in fecal samples, suggesting active replication in the gastrointestinal tract and possible fecal-oral transmission, although the latter remains under investigation (214,215). Liver dysfunction, including elevated liver enzymes, has been observed specifically after SARS-CoV-2 infection, especially those with severe COVID-19 disease. Although ACE2 is modestly expressed in this tissue, it is sufficient to enable productive replication (154,216,217) (Fig. 2C). MERS-CoV has also been found to affect other organs, although less frequently reported compared to SARS-CoV-2. The presence of DPP4 receptors in tissues, such as the kidneys, gastrointestinal tract, and liver, allows the virus to spread beyond the lungs (218,219). Renal failure has been observed in some severe cases, particularly in patients with pre-existing conditions (220). MERS-CoV has also been associated with mild gastrointestinal symptoms, including diarrhea, although this is less common (221), and human intestinal cells have been experimentally shown to be susceptible to infection, prompting speculation about the human intestinal tract as an alternative infection route (222).
## Circulatory system and heart
The human heart is characterized by particularly high ACE2 expression, especially in cardiomyocytes, endothelial cells, and pericytes, highlighting its potential susceptibility. In patients with or without a history of cardiovascular disease, cardiac function may be impaired in the context of SARS-CoV-2 infection (223, 224) (Fig. 2C). SARS-CoV-2 has been found in myocardial tissue, which can possibly lead to cardiovascular manifestations, such as myocarditis, pericarditis, acute coronary syndrome, thromboembolic events, and heart failure (223,225). Cardiovascular complications were occasionally reported in SARS-CoV infection, but data remain limited and largely anecdotal due to the lack of systematic studies. Case reports and small cohorts described events such as acute myocardial infarction, transient diastolic dysfunction, tachycardia, hypotension, and rare arrhythmias, most of which were self-limiting and occurred in otherwise asymptomatic patients (226). Autopsy findings from a small study revealed thromboembolic events and myocardial infarction, though the relevance of these findings remains unclear due to small sample sizes and lack of confirmatory studies (227). Although DPP4 is widely expressed in the vascular system, the available data on cardiac tropism of MERS-CoV are very limited (228). To date, there is little direct evidence supporting significant involvement of the heart in MERS-CoV infection, and dedicated studies addressing this aspect are lacking (229). For endemic HCoVs, cardiovascular effects are rare and typically limited to exacerbation of pre-existing heart conditions in vulnerable individuals (227).
The exact mechanisms by which SARS-CoV-2 and other HCoVs spread to different organs are still unclear. It is suspected that the virus may reach distant tissues through the bloodstream or could be transported on immune cells. Another theory suggests that nerve pathways could facilitate viral spread (200,(230)(231)(232).
In conclusion, experimental and clinical evidence suggests a broad tissue and cell tropism associated with both epidemic and pandemic HCoVs, affecting the hematolog ical, cardiovascular, renal, gastrointestinal and hepatobiliary, endocrinological, neuro logical, ophthalmological, and dermatological systems (233). In addition to receptor availability, additional factors, including co-opted host factors, immune modulation, and the role of specific cellular signaling pathways, may contribute to HCoV spread and tissue tropism, highlighting the need for further research to fully determine the profiles of permissive and susceptible cells (234,235).
## CONCLUSIONS
It is estimated that more than 60% of emerging viruses causing disease in humans originate from zoonotic transmission (236). CoVs represent a large family of viruses associated with a wide range of diseases, including respiratory, enteric, hepatic, and neurological manifestations and have been detected in a wide range of vertebrate species (4). It is important to note that the interpretation of CoV tropism is highly dependent on the experimental model used, with each system (ranging from immor talized cell lines to organoids and animal models) offering distinct advantages and limitations. Careful selection and contextual understanding of these models are essential to accurately assess viral behavior and improve the translation of findings to human physiology and disease.
Despite significant advances in our understanding of HCoVs, critical gaps in knowledge remain. The precise mechanisms underlying viral persistence, particularly in asymptomatic or mildly symptomatic individuals, are not fully elucidated. Similarly, host restriction and dependency factors that govern species specificity and transmission dynamics across seasonal HCoVs are understudied and still poorly defined. Moreover, a deeper understanding of cross-reactive immunity-especially how prior exposure to seasonal HCoVs may influence responses to emerging pathogens like SARS-CoV-2-is urgently needed.
Progress in these areas is hampered by several technical limitations. The lack of robust in vitro models for certain HCoVs continues to constrain experimental validation of host-virus interactions. Moreover, primary human airway cultures and organoids, which are essential for physiologically relevant studies, remain limited in availability and standardization and lack important immune factors. Research tools and genomic databases also exhibit a notable underrepresentation of non-SARS HCoVs, limiting cross-comparative analyses and mechanistic insight. Overcoming these barriers will require interdisciplinary collaborations and coordinated infrastructure to support both basic and translational research. Together, these efforts will be crucial for anticipating and mitigating future coronavirus emergence events.
Many factors are predicted to increase zoonotic spillovers from wildlife to humans in the future, driven by increasing human encroachment into wildlife habitats and accelerating climate change (237)(238)(239). The likely zoonotic origins of HCoVs underscore that cross-species spillover events have been central to the emergence of human pathogens. HCoVs typically cause respiratory infections ranging from mild common cold symptoms (as seen with seasonal HCoVs) to severe disease, such as pneumonia and acute respiratory distress syndrome, exemplified by SARS-CoV, MERS-CoV, and SARS-CoV-2. The rapid adaptation of SARS-CoV-2 to humans, including its efficient utilization of the ACE2 receptor and TMPRSS2 protease for entry, highlights how quickly CoVs can evolve to establish sustained human-to-human transmission. Given these dynamics, monitoring zoonotic reservoirs remains critical-not only to detect viruses already using key human receptors such as ACE2, APN, DPP4, or TMPRSS2, but also to identify emerging variants with pandemic potential before widespread human infection occurs. For example, CoVs closely related to SARS-CoV and MERS-CoV that utilize ACE2 or DPP4 receptors pose the highest risk of spillover and adaptation. Targeted surveil lance in animal hosts, combined with functional studies of viral tropism and receptor usage, can inform risk assessment and guide early intervention strategies. By deepen ing our understanding of CoV host adaptation, tropism, and receptor interactions, we can improve pandemic preparedness and response efforts. This proactive approach will help mitigate the global impact of emerging CoVs and reduce the likelihood of future pandemics by enabling timely identification and containment of novel threats. Importantly, addressing these gaps will require sustained efforts in field surveillance, viral genome cataloging, and functional characterization of novel coronaviruses in both known reservoirs (e.g., bats) and high-risk interfaces such as wildlife trade and wet markets. Bats, for example, may serve as key reservoirs for coronavirus evolution, similar to the role of birds and pigs in influenza (240). Evidence of spike gene recombination among bat coronaviruses supports their potential to generate viruses with altered host tropism (241).
A particular focus on receptor-dependent and -independent mechanisms of viral tropism, combined with intensive wildlife surveillance, is required to identify potential host shifts to humans before they occur. Understanding these molecular interactions at the organismal level will help predict which CoVs are most likely to cross species barriers and establish sustained human-to-human transmission.
Finally, improved understanding of CoV tropism has significant implications for the development of vaccines and therapeutics. By elucidating the mechanisms governing tissue and species specificity, such insights can inform the design of broadly protective or universal interventions that target conserved viral entry pathways. This knowledge could ultimately enhance preparedness against both current and emerging CoV threats by enabling more effective cross-reactive immune responses.
## FUNDING
## References
1. Chen, Liu, Guo (2020) "Emerging coronaviruses: genome structure, replication, and pathogenesis" *J Med Virol*
2. Letko, Marzi, Munster (2020) "Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronavi ruses" *Nat Microbiol*
3. Hoenigsperger, Sivarajan, Sparrer (2024) "Differences and similarities between innate immune evasion strategies of human coronaviruses" *Curr Opin Microbiol*
4. Steiner, Kratzel, Barut et al. (2024) "SARS-CoV-2 biology and host interactions" *Nat Rev Microbiol*
5. Naqvi, Mohammad, Fatima et al. (2020) "Insights into SARS-CoV-2 genome, structure, evolution, pathogenesis and therapies: structural genomics approach" *Biochim Biophys Acta Mol Basis Dis*
6. Madeira, Madhusoodanan, Lee et al. (2024) "The EMBL-EBI Job Dispatcher sequence analysis tools framework in 2024" *Nucleic Acids Res*
7. Poelen, Simons, Mungall (2014) "Global biotic interactions: an open infrastructure to share and analyze species-interaction datasets" *Ecol Inform*
8. Sayers, Beck, Bolton et al. (2025) "Database resources of the National Center for Biotechnology Information in 2025" *Nucleic Acids Res*
9. Hamre, Procknow (1966) "A new virus isolated from the human respiratory tract" *Exp Biol Med (Maywood)*
10. Corman, Muth, Niemeyer et al. (2018) "Hosts and sources of endemic human coronaviruses" *Adv Virus Res*
11. Lau, Lee, Tsang et al. (2011) "Molecular epidemiology of human coronavirus OC43 reveals evolution of different genotypes over time and recent emergence of a novel genotype due to natural recombina tion" *J Virol*
12. Sung, Lee, Eun et al. (2010) "Role of human coronavirus NL63 in hospitalized children with croup" *Pediatr Infect Dis J*
13. Dominguez, Travanty, Qian et al. (2013) "Human coronavi rus HKU1 infection of primary human type II alveolar epithelial cells: cytopathic effects and innate immune response" *PLoS One*
14. Woo, Lau, Chu et al. (2005) "Characterization and complete genome sequence of a novel coronavirus, coronavirus HKU1, from patients with pneumonia" *J Virol*
15. Jara, Santos, Reyes et al. (2025) "Endemic coronavirus in children and adults with acute respiratory infection before the COVID-19 pandemic" *Rev Argent Microbiol*
16. Rodriguez-Nava, Egoryan, Dong et al. (2022) "Comparison of clinical characteristics and outcomes of hospitalized patients with seasonal coronavirus infection and COVID-19: a retrospective cohort study" *BMC Infect Dis*
17. Roussel, Gatineau, Jimeno et al. (2020) "SARS-CoV-2: fear versus data" *Int J Antimicrob Agents*
18. Cimolai (2021) "Complicating infections associated with common endemic human respiratory coronaviruses" *Health Secur*
19. Chung, Hong, Huh et al. (2022) "Clinical features and outcomes of severe pneumonia caused by endemic human coronavirus in adults" *Am J Respir Crit Care Med*
20. Pormohammad, Ghorbani, Khatami et al. (2020) "Comparison of confirmed COVID-19 with SARS and MERS cases -Clinical characteristics, laboratory findings, radiographic signs and outcomes: a systematic review and meta-analysis" *Rev Med Virol*
21. Bolles, Donaldson, Baric (2011) "SARS-CoV and emergent coronaviruses: viral determinants of interspecies transmission" *Curr Opin Virol*
22. Zaki, Van Boheemen, Bestebroer et al. (2012) "Isolation of a novel coronavirus from a man with pneumo nia in Saudi Arabia" *N Engl J Med*
23. Raj, Osterhaus, Fouchier et al. (2014) "MERS: emergence of a novel human coronavirus" *Curr Opin Virol*
24. Baggen, Vanstreels, Jansen et al. (2021) "Cellular host factors for SARS-CoV-2 infection" *Nat Microbiol*
25. Maurya, Swaminathan, Shamim et al. (2023) "Co-evolution of SARS-CoV-2 variants and host immune response trajectories underlie COVID-19 pandemic to epidemic transition"
26. Martin-Sancho, Lewinski, Pache et al. (2021) "Functional landscape of SARS-CoV-2 cellular restriction" *Mol Cell*
27. Forni, Cagliani, Clerici et al. (2017) "Molecular evolution of human coronavirus genomes" *Trends Microbiol*
28. Pfefferle, Oppong, Drexler et al. (2009) "Distant relatives of severe acute respiratory syndrome coronavi rus and close relatives of human coronavirus 229E in bats" *Ghana. Emerg Infect Dis*
29. Sabir, Lam, Ahmed et al. (2016) "Co-circulation of three camel coronavirus species and recombination of MERS-CoVs in Saudi Arabia" *Science*
30. Chibo, Birch (2006) "Analysis of human coronavirus 229E spike and nucleoprotein genes demonstrates genetic drift between chronologi cally distinct strains" *J Gen Virol*
31. Liu, Chen, Chen et al. (2023) "Mouse models susceptible to HCoV-229E and HCoV-NL63 and cross protection from challenge with SARS-CoV-2" *Proc Natl Acad Sci*
32. Vijgen, Keyaerts, Moës et al. (2005) "Complete genomic sequence of human coronavirus OC43: molecular clock analysis suggests a relatively recent zoonotic coronavirus transmission event" *J Virol*
33. Xu, Qiao, Schraauwen et al. (2024) "Evidence for cross-species transmission of human coronavirus OC43 through bioinformatics and modeling infections in porcine intestinal organoids" *Vet Microbiol*
34. Jacomy, Fragoso, Almazan et al. (2006) "Human coronavirus OC43 infection induces chronic encephalitis leading to disabilities in BALB/C mice" *Virology (Auckland)*
35. Butler, Pewe, Trandem et al. (2006) "Murine encephalitis caused by HCoV-OC43, a human coronavirus with broad species specificity, is partly immune-mediated" *Virology (Auckland)*
36. Pyrc, Dijkman, Deng et al. (2006) "Mosaic structure of human coronavirus NL63, one thousand years of evolution" *J Mol Biol*
37. Huynh, Li, Yount et al. (2012) "Evidence supporting a zoonotic origin of human coronavirus strain NL63" *J Virol*
38. Tao, Shi, Chommanard et al. (2017) "Surveillance of bat coronaviruses in Kenya identifies relatives of human coronaviruses NL63 and 229E and their recombination history" *J Virol*
39. Stout, Millet, Stanhope et al. (2021) "Furin cleavage sites in the spike proteins of bat and rodent coronaviruses: implications for virus evolution and zoonotic transfer from rodent species" *One Health*
40. Wang, Lin, Zhang et al. (2020) "Extensive genetic diversity and host range of rodent-borne coronaviruses" *Virus Evol*
41. Saunders, Fernandez, Planchais et al. (2023) "TMPRSS2 is a functional receptor for human coronavirus HKU1" *Nature*
42. Mccallum, Park, Stewart et al. (2024) "Human coronavirus HKU1 recognition of the TMPRSS2 host receptor" *Cell*
43. Catanzaro, Wu, Fan et al. (2025) "ACE2 from Pipistrellus abramus bats is a receptor for HKU5 coronaviruses" *Nat Commun*
44. Cheever, Daniels, Pappenheimer et al. (1949) "A murine virus (JHM) causing disseminated encephalomyelitis with extensive destruction of myelin" *J Exp Med*
45. Cyranoski, Abbott (2003) "Virus detectives seek source of SARS in China's wild animals" *Nature*
46. Shi, Hu (2008) "A review of studies on animal reservoirs of the SARS coronavirus" *Virus Res*
47. Guan, Zheng, He et al. (2003) "Isolation and characterization of viruses related to the SARS coronavirus from animals in southern China" *Science*
48. Poon, Chu, Chan et al. (2001) "Identification of a novel coronavirus in bats" *J Virol*
49. Lau, Li, Huang et al. (2010) "Ecoepidemiology and complete genome comparison of different strains of severe acute respiratory syndrome-related Rhinolophus bat coronavirus in China reveal bats as a reservoir for acute, self-limiting infection that allows recombination events" *J Virol*
50. Lau, Fan, Luk et al. (2018) "Replication of MERS and SARS coronaviruses in bat cells offers insights to their ancestral origins" *Emerg Microbes Infect*
51. Balboni, Battilani, Prosperi (2012) "The SARS-like coronaviruses: the role of bats and evolutionary relationships with SARS coronavirus" *New Microbiol*
52. Lau, Woo, Li et al. (2005) "Severe acute respiratory syndrome coronavirus-like virus in Chinese horseshoe bats" *Proc Natl Acad Sci*
53. Drexler, Gloza-Rausch, Glende et al. (2010) "Genomic characterization of severe acute respiratory syndrome-related coronavirus in European bats and classification of coronaviruses based on partial RNA-dependent RNA polymerase gene sequences" *J Virol*
54. Ge, Li, Yang et al. (2013) "Isolation and characteriza tion of a bat SARS-like coronavirus that uses the ACE2 receptor" *Nature*
55. Chan, Chan (2013) "Tracing the SARS-coronavirus" *J Thorac Dis*
56. Worobey, Levy, Serrano et al. (2022) "The Huanan Seafood Wholesale Market in Wuhan was the early epicenter of the COVID-19 pandemic" *Science*
57. Crits-Christoph, Levy, Pekar et al. (2024) "Genetic tracing of market wildlife and viruses at the epicenter of the COVID-19 pandemic" *Cell*
58. Zhang, Wu, Zhang (2020) "Probable pangolin origin of SARS-CoV-2 associated with the COVID-19 outbreak" *Curr Biol*
59. Gupta, Minocha, Thapa et al. (2022) "Role of the pangolin in origin of SARS-CoV-2: an evolutionary perspective" *Int J Mol Sci*
60. Zhou, Yang, Wang et al. (2020) "A pneumonia outbreak associated with a new coronavirus of probable bat origin" *Nature*
61. Pekar, Lytras, Ghafari et al. (2025) "The recency and geographical origins of the bat viruses ancestral to SARS-CoV and SARS-CoV-2" *Cell*
62. Shi, Wen, Zhong et al. (2020) "Susceptibility of ferrets, cats, dogs, and other domesticated animals to SARS-coronavirus 2" *Science*
63. Zhou, Shi (2021) "SARS-CoV-2 spillover events" *Science*
64. Sit, Brackman, Ip et al. (2020) "Infection Minireview mBio November"
65. "of dogs with SARS-CoV-2" *Nature*
66. Bertzbach, Vladimirova, Dietert et al. (2021) "SARS-CoV-2 infection of Chinese hamsters (Cricetulus griseus) reproduces COVID-19 pneumonia in a wellestablished small animal model" *Transbound Emerg Dis*
67. Trimpert, Vladimirova, Dietert et al. (2020) "The Roborovski dwarf hamster is a highly susceptible model for a rapid and fatal course of SARS-CoV-2 infection" *Cell Rep*
68. Sia, Yan, Chin et al. (2020) "Pathogenesis and transmission of SARS-CoV-2 in golden hamsters" *Nature*
69. Hassan, Case, Winkler et al. (2020) "A SARS-CoV-2 infection model in mice demonstrates protection by neutralizing antibodies" *Cell*
70. Sun, Zhuang, Zheng et al. (2020) "Generation of a broadly useful model for COVID-19 pathogenesis, vaccination, and treatment" *Cell*
71. Liu, Selvaraj, Sangare et al. (2022) "Spike proteinindependent attenuation of SARS-CoV-2 Omicron variant in laboratory mice" *Cell Rep*
72. Reusken, Haagmans, Müller et al. (2013) "Middle East respiratory syndrome coronavirus neutralising serum antibodies in dromedary camels: a comparative serological study" *Lancet Infect Dis*
73. Chu, Poon, Gomaa et al. (2014) "MERS coronaviruses in dromedary camels" *Egypt. Emerg Infect Dis*
74. Corman, Jores, Meyer et al. (1992) "Antibodies against MERS coronavirus in dromedary camels" *Emerg Infect Dis*
75. Müller, Corman, Jores et al. (1983) "MERS coronavirus neutralizing antibodies in camels, Eastern Africa"
76. Reusken, Ababneh, Raj et al. (2013) "Middle East respiratory syndrome coronavirus (MERS-CoV) serology in major livestock species in an affected region in Jordan" *Euro Surveill*
77. Gultom, Kratzel, Portmann et al. (2022) "Establishment of well differentiated camelid airway cultures to study Middle East respiratory syndrome coronavirus" *Sci Rep*
78. Mulemena, Sichamba, Muleya et al. (2025) "Global distribution of coronaviruses among bat populations detected using molecular techniques, a systematic review"
79. Leists, Cockrell (2020) "Genetically engineering a susceptible mouse model for MERS-CoV-induced acute respiratory distress syndrome" *Methods Mol Biol*
80. Townsend, Hassler, Lamb et al. (2023) "Seasonality of endemic COVID-19" *mBio*
81. Turoňová, Sikora, Schürmann et al. (2020) "In situ structural analysis of SARS-CoV-2 spike reveals flexibility mediated by three hinges" *Science*
82. Zhao, Praissman, Grant et al. (2020) "Virus-receptor interactions of glycosylated SARS-CoV-2 spike and human ACE2 receptor"
83. Wu, Yin, Jiang et al. (2022) "Structure genomics of SARS-CoV-2 and its Omicron variant: drug design templates for COVID-19" *Acta Pharmacol Sin*
84. Díaz-Salinas, Li, Ejemel et al. (2022) "Conformational dynamics and allosteric modulation of the SARS-CoV-2 spike" *Elife*
85. Peacock, Goldhill, Zhou et al. (2021) "The furin cleavage site in the SARS-CoV-2 spike protein is required for transmission in ferrets" *Nat Microbiol*
86. Zhang, Zheng, Niu et al. (2021) "SARS-CoV-2 spike protein dictates syncytium-mediated lymphocyte elimination" *Cell Death Differ*
87. Hulswit, Lang, Bakkers et al. (2019) "Human coronaviruses OC43 and HKU1 bind to 9-O-acetylated sialic acids via a conserved receptor-binding site in spike protein domain A" *Proc Natl Acad Sci*
88. Tortorici, Walls, Lang et al. (2019) "Structural basis for human coronavirus attachment to sialic acid receptors" *Nat Struct Mol Biol*
89. Pronker, Creutznacher, Drulyte et al. (2023) "Sialoglycan binding triggers spike opening in a human coronavirus" *Nature*
90. Wang, Liu, Zhang et al. (2024) "TMPRSS2 and glycan receptors synergistically facilitate coronavirus entry" *Cell*
91. Li, Hulswit, Widjaja et al. (2017) "Identification of sialic acid-binding function for the Middle East respiratory syndrome coronavirus spike glycoprotein" *Proc Natl Acad Sci*
92. Milewska, Zarebski, Nowak et al. (2014) "Human coronavirus NL63 utilizes heparan sulfate proteoglycans for attachment to target cells" *J Virol*
93. Lempp, Soriaga, Montiel-Ruiz et al. (2021) "Lectins enhance SARS-CoV-2 infection and influence neutralizing antibodies" *Nature*
94. Conca, Bano, Graul et al. (2025) "Variantspecific interactions at the plasma membrane: heparan sulfate's impact on SARS-CoV-2 binding kinetics" *Anal Chem*
95. Li, Li, Farzan et al. (2005) "Structure of SARS coronavirus spike receptor-binding domain complexed with receptor" *Science*
96. Li, Tomlinson, Wong et al. (2019) "The human coronavirus HCoV-229E Sprotein structure and receptor binding. Elife 8:e51230"
97. Wang, Shi, Jiang et al. (2013) "Structure of MERS-CoV spike receptor-binding domain complexed with human receptor DPP4" *Cell Res*
98. Wang, Zhang, Wu et al. (2020) "Structural and functional basis of SARS-CoV-2 entry by using human ACE2" *Cell*
99. Wu, Li, Peng et al. (2009) "Crystal structure of NL63 respiratory coronavirus receptor-binding domain complexed with its human receptor" *Proc Natl Acad Sci*
100. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
101. Meng, Goddard, Pettersen et al. (2023) "UCSF ChimeraX: tools for structure building and analysis" *Protein Sci*
102. Han, Zhou, Fei et al. (2020) "Construction of a human cell landscape at single-cell level" *Nature*
103. Deprez, Zaragosi, Truchi et al. (2020) "A single-cell Atlas of the human healthy airways" *Am J Respir Crit Care Med*
104. Karlsson, Zhang, Méar et al. (2021) "A single-cell type transcriptomics map of human tissues" *Sci Adv*
105. Li, Moore, Vasilieva et al. (2003) "Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus" *Nature*
106. 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*
107. Hofmann, Pyrc, Van Der Hoek et al. (2005) "Human coronavirus NL63 employs the severe acute respiratory syndrome coronavirus receptor for cellular entry" *Proc Natl Acad Sci*
108. Boehm, Nabel (2002) "Angiotensin-converting enzyme 2 -a new cardiac regulator" *N Engl J Med*
109. Oudit, Wang, Viveiros et al. (2023) "Angiotensin-converting enzyme 2-at the heart of the COVID-19 pandemic" *Cell*
110. Wrapp, Wang, Corbett et al. (2020) "Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation" *Science*
111. Lupala, Ye, Chen et al. (2022) "Mutations on RBD of SARS-CoV-2 Omicron variant result in stronger binding to human ACE2 receptor" *Biochem Biophys Res Commun*
112. Li, Sui, Huang et al. (2007) "The S proteins of human coronavirus NL63 and severe acute respiratory syndrome coronavirus bind overlapping regions of ACE2" *Virology (Auckl)*
113. Yan, Jiao, Liu et al. (2021) "ACE2 receptor usage reveals variation in susceptibility to SARS-CoV and SARS-CoV-2 infection among bat species" *Nat Ecol Evol*
114. Liu, Hu, Wang et al. (2021) "Functional and genetic analysis of viral receptor ACE2 orthologs reveals a broad potential host range of SARS-CoV-2" *Proc Natl Acad Sci*
115. Xiao, Zhai, Feng et al. (2020) "Isolation of SARS-CoV-2-related coronavirus from Malayan pangolins" *Nature*
116. Huang, Sun, Zhou et al. (2023) "A pangolin-origin SARS-CoV-2-related coronavirus: infectivity, pathogenicity, and cross-protection by preexisting immunity" *Cell Discov*
117. Hou, Chiba, Leist et al. (2023) "Host range, transmissibility and antigenicity of a pangolin coronavirus" *Nat Microbiol*
118. Wang, Jiao, Veit et al. (2025) "A MERS-CoV-like mink coronavirus uses ACE2 as an entry receptor" *Nature*
119. Xiong, Cao, Ma et al. (2022) "Close relatives of MERS-CoV in bats use ACE2 as their functional receptors" *Nature*
120. Ma, Liu, Park et al. (2025) "Multiple independent acquisitions of ACE2 usage in MERS-related coronaviruses" *Cell*
121. Chen, Zhang, Li et al. (2025) "Bat-infecting merbecovirus HKU5-CoV lineage 2 can use human ACE2 as a cell entry receptor" *Cell*
122. Yeager, Ashmun, Williams et al. (1992) "Human aminopeptidase N is a receptor for human coronavirus 229E" *Nature*
123. Chen, Lin, Peng et al. (2012) "Structural basis for multifunctional roles of mammalian aminopeptidase N" *Proc Natl Acad Sci*
124. Wong, Tomlinson, Zhou et al. (2017) "Receptor-binding loops in alphacoronavirus adaptation and evolution" *Nat Commun*
125. Tsai, Chien, Hsu et al. (2025) "Molecular basis of host recognition of human coronavirus 229E" *Nat Commun*
126. Delmas, Gelfi, 'haridon et al. (1992) "Aminopeptidase N is a major receptor for the enteropathogenic coronavirus TGEV" *Nature*
127. Tresnan, Levis, Holmes (1996) "Feline aminopeptidase N serves as a receptor for feline, canine, porcine, and human coronaviruses in serogroup I" *J Virol*
128. Wentworth, Holmes (2001) "Molecular determinants of species specificity in the coronavirus receptor aminopeptidase N (CD13): influence of N-linked glycosylation" *J Virol*
129. Siddell, Hegyi, Kolb (1997) "Identification of residues critical for the human coronavirus 229E receptor function of human aminopepti dase N" *J Gen Virol*
130. Kolb, Heister, Siddell (1996) "Characterization of functional domains in the human coronavirus HCV 229E receptor" *J Gen Virol*
131. Raj, Mou, Smits et al. (2013) "Dipeptidyl peptidase 4 is a functional receptor for the emerging human coronavirus-EMC" *Nature*
132. Mentlein (1999) "Dipeptidyl-peptidase IV (CD26)-role in the inactiva tion of regulatory peptides" *Regul Pept*
133. Yuan, Cao, Zhang et al. (2017) "Cryo-EM structures of MERS-CoV and SARS-CoV spike glycoproteins reveal the dynamic receptor binding domains" *Nat Commun*
134. Letko, Miazgowicz, Mcminn et al. (2018) "Adaptive evolution of MERS-CoV to species variation in DPP4" *Cell Rep*
135. Cockrell, Peck, Yount et al. (2014) "Mouse dipeptidyl peptidase 4 is not a functional receptor for Middle East respiratory syndrome coronavirus infection" *J Virol*
136. Van Doremalen, Miazgowicz, Milne-Price et al. (2014) "Host species restriction of Middle East respiratory syndrome coronavirus through its receptor, dipeptidyl peptidase 4" *J Virol*
137. Yang, Du, Liu et al. (2014) "Receptor usage and cell entry of bat coronavirus HKU4 provide insight into bat-to-human transmission of MERS coronavirus" *Proc Natl Acad Sci*
138. 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*
139. Lau, Fan, Zhu et al. (2021) "Isolation of MERS-related coronavirus from lesser bamboo bats that uses DPP4 and infects human-DPP4transgenic mice" *Nat Commun*
140. Paoloni-Giacobino, Chen, Peitsch et al. (1997) "Cloning of the TMPRSS2 gene, which encodes a novel serine protease with transmembrane, LDLRA, and SRCR domains and maps to 21q22" *Genomics*
141. Gao, Zhu, Wang et al. (2024) "Structural basis for the interaction between human coronavirus HKU1 spike receptor binding domain and its receptor TMPRSS2" *Cell Discov*
142. Fernández, Saunders, Duquerroy et al. (2024) "Structural basis of TMPRSS2 zymogen activation and recognition by the HKU1 seasonal coronavirus" *Cell*
143. Chen, Ou, Li et al. (2024) "Identification of the critical residues of TMPRSS2 for entry and host range of human coronavirus HKU1" *J Virol*
144. Dufloo, Andreu-Moreno, Moreno-García et al. (2025) "Receptor-binding proteins from animal viruses are broadly compatible with human cell entry factors" *Nat Microbiol*
145. Palakurty, Diamond (2025) "Viral entry as a low barrier to zoonosis" *Nat Microbiol*
146. Daly, Simonetti, Klein et al. (2020) "Neuropilin-1 is a host factor for SARS-CoV-2 infection" *Science*
147. Kielian (2020) "Enhancing host cell infection by SARS-CoV-2" *Science*
148. Cheng, Chao, Li et al. (2020) "Furin inhibitors block SARS-CoV-2 spike protein cleavage to suppress virus production and cytopathic effects" *Cell Rep*
149. Wu, Zhao (2021) "Furin cleavage sites naturally occur in coronavi ruses" *Stem Cell Res*
150. Vu, Alvarado, Morris et al. (2023) "Loss-of-function mutation in Omicron variants reduces spike protein expression and attenuates SARS-CoV-2 infection"
151. Pierson, Diamond (2020) "The continued threat of emerging flaviviruses" *Nat Microbiol*
152. Böttcher, Matrosovich, Beyerle et al. (2006) "Proteolytic activation of influenza viruses by serine proteases TMPRSS2 and HAT from human airway epithelium" *J Virol*
153. Shulla, Heald-Sargent, Subramanya et al. (2011) "A transmembrane serine protease is linked to the severe acute respiratory syndrome coronavirus receptor and activates virus entry" *J Virol*
154. Bertram, Dijkman, Habjan et al. (2013) "TMPRSS2 activates the human coronavirus 229E for cathepsinindependent host cell entry and is expressed in viral target cells in the respiratory epithelium" *J Virol*
155. Heinen, Khanal, Westhoven et al. (2024) "Productive infection of primary human hepatocytes with SARS-CoV-2 induces antiviral and proinflam matory responses" *Gut*
156. Gierer, Bertram, Kaup et al. (2013) "The spike protein of the emerging betacoronavirus EMC uses a novel coronavirus receptor for entry, can be activated by TMPRSS2, and is targeted by neutralizing antibodies" *J Virol*
157. Zang, Castro, Mccune et al. (2020) "TMPRSS2 and TMPRSS4 promote SARS-CoV-2 infection of human small intestinal enterocytes" *Sci Immunol*
158. Qu, Miskey, Gömer et al. (2024) "TMPRSS2-mediated SARS-CoV-2 uptake boosts innate immune activation, enhances cytopathology, and drives convergent virus evolution" *Proc Natl Acad Sci*
159. Hoffmann, Hofmann-Winkler, Smith et al. (2021) "Camostat mesylate inhibits SARS-CoV-2 activation by TMPRSS2-related proteases and its metabolite GBPA exerts antiviral activity" *EBioMedi cine*
160. Koch, Uckeley, Doldan et al. (2021) "TMPRSS2 expression dictates the entry route used by SARS-CoV-2 to infect host cells" *EMBO J*
161. Bayati, Kumar, Francis et al. (2021) "SARS-CoV-2 infects cells after viral entry via clathrin-mediated endocytosis" *J Biol Chem*
162. Zhao, Yang, Yang et al. (2021) "Cathepsin L plays a key role in SARS-CoV-2 infection in humans and humanized mice and is a promising target for new drug development" *Sig Transduct Target Ther*
163. Shin, Toyoda, Nishitani et al. (2021) "Possible involvement of adipose tissue in patients with older age, obesity, and diabetes with SARS-CoV-2 infection (COVID-19) via grp78 (BIP/HSPA5): significance of hyperinsulinemia management in COVID-19" *Diabetes*
164. Carlos, Ha, Yeh et al. (2021) "The chaperone GRP78 is a host auxiliary factor for SARS-CoV-2 and GRP78 depleting antibody blocks viral entry and infection" *J Biol Chem*
165. Wang, Qiu, Hou et al. (2021) "AXL is a candidate receptor for SARS-CoV-2 that promotes infection of pulmonary and bronchial epithelial cells" *Cell Res*
166. Qi, Qian, Zhang et al. (2020) "Single cell RNA sequencing of 13 human tissues identify cell types and receptors of human coronavi ruses" *Biochem Biophys Res Commun*
167. Cui, Huang, Zhou et al. (2021) "AGTR2, one possible novel key gene for the entry of SARS-CoV-2 into human cells" *IEEE/ACM Trans Comput Biol and Bioinf*
168. Shilts, Crozier, Greenwood et al. (2021) "No evidence for basigin/CD147 as a direct SARS-CoV-2 spike binding receptor" *Sci Rep*
169. Wang, Chen, Zhang et al. (2020) "CD147-spike protein is a novel route for SARS-CoV-2 infection to host cells" *Sig Transduct Target Ther*
170. Ragotte, Pulido, Donnellan et al. (2021) "Human basigin (CD147) does not directly interact with SARS-CoV-2 spike glycoprotein" *mSphere*
171. Baggen, Persoons, Vanstreels et al. (2021) "Genome-wide CRISPR screening identifies TMEM106B as a proviral host factor for SARS-CoV-2" *Nat Genet*
172. Baggen, Jacquemyn, Persoons et al. (2023) "TMEM106B is a receptor mediating ACE2-independent SARS-CoV-2 cell entry" *Cell*
173. Yan, Dumenil, Stewart et al. (2024) "TMEM106B-mediated SARS-CoV-2 infection allows for robust ACE2-independent infection in vitro but not in vivo" *Cell Rep*
174. Jayaweera, Perera, Gunawardana et al. (2020) "Transmis sion of COVID-19 virus by droplets and aerosols: a critical review on the unresolved dichotomy" *Environ Res*
175. Shafaghi, Talabazar, Koşar et al. (2020) "On the effect of the respiratory droplet generation condition on COVID-19 transmission" *Fluids*
176. Zhu, Wang, Liu et al. (2020) "Morphogenesis and cytopathic effect of SARS-CoV-2 infection in human airway epithelial cells" *Nat Commun*
177. Loo, Wark, Esneau et al. (2020) "Human coronaviruses 229E and OC43 replicate and induce distinct antiviral responses in differentiated primary human bronchial epithelial cells" *Am J Physiol Lung Cell Mol Physiol*
178. Pyrc, Berkhout, Van Der Hoek (2007) "The novel human coronavi ruses NL63 and HKU1" *J Virol*
179. Wang, Deering, Macke et al. (2000) "Human coronavirus 229E infects polarized airway epithelia from the apical surface" *J Virol*
180. Liu, Qu, Qu et al. (2020) "Expression pattern of the SARS-CoV-2 entry genes ACE2 and TMPRSS2 in the respiratory tract" *Viruses*
181. Fielding (2011) "Human coronavirus NL63: a clinically important virus?" *Future Microbiol*
182. Sungnak, Huang, Bécavin et al. (2020) "SARS-CoV-2 entry factors are highly expressed in nasal epithelial cells together with innate immune genes" *Nat Med*
183. Hou, Okuda, Edwards et al. (2020) "SARS-CoV-2 reverse genetics reveals a variable infection gradient in the respiratory tract" *Cell*
184. Xu, Shi, Wang et al. (2020) "Pathological findings of COVID-19 associated with acute respiratory distress syndrome" *Lancet Respir Med*
185. Hu, Yen, Singh et al. (2012) "SARS-CoV regulates immune function-related gene expression in human monocytic cells" *Viral Immunol*
186. Gu, Gong, Zhang et al. (2005) "Multiple organ infection and the pathogenesis of SARS" *J Exp Med*
187. To, Lo (2004) "Exploring the pathogenesis of severe acute respiratory syndrome (SARS): the tissue distribution of the coronavirus (SARS-CoV) and its putative receptor, angiotensin-converting enzyme 2 (ACE2)" *J Pathol*
188. Chan, Hemida, Kayali et al. (2014) "Tropism and replication of Middle East respiratory syndrome coronavirus from dromedary camels in the human respiratory tract: an in-vitro and ex-vivo study" *Lancet Respir Med*
189. Otter, Tan, Khosla et al. (2023) "Infection of primary nasal epithelial cells differentiates among lethal and seasonal human coronaviruses" *Proc Natl Acad Sci*
190. Cockrell, Yount, Scobey et al. (2016) "A mouse model for MERS coronavirus-induced acute respiratory distress syndrome" *Nat Microbiol*
191. Alenazi, Arabi (2022) "Severe Middle East respiratory syndrome (MERS) pneumonia"
192. Morgello (2020) "Coronaviruses and the central nervous system" *J Neurovirol*
193. Talbot, Ékandé, Cashman et al. (1994) "Neurotropism of human coronavirus 229E"
194. Saeed, Helali, Alhammadi (2024) "An interesting case of coronavirus NL63 encephalitis diagnosed in a 14-year-old immunocom petent female: a case report and literature review"
195. Yuan, Chunjie (2021) "Acute necrotizing encephalitis caused by respiratory coronavirus-NL63: a case report" *Chin J Neurol*
196. Lukiw, Pogue, Hill (2022) "SARS-CoV-2 infectivity and neurological targets in the brain" *Cell Mol Neurobiol*
197. Emmi, Tushevski, Sinigaglia et al. (2023) "ACE2 receptor and TMPRSS2 protein expression patterns in the human brainstem reveal anatomical regions potentially vulnerable to SARS-CoV-2 infection" *ACS Chem Neurosci*
198. Pang, Tang, He et al. (2024) "Neurological complications caused by SARS-CoV-2" *Clin Microbiol Rev*
199. Lopes, Sitton, Pillat et al. (2025) "SARS-CoV-2 and neurotropism: evidence, gaps and reflections" *J Med Microbiol*
200. Kong, Montano, Corley et al. (2022) "Neuropilin-1 mediates SARS-CoV-2 infection of astrocytes in brain organoids, inducing inflammation leading to dysfunction and death of neurons" *mBio*
201. Burks, Rosas-Hernandez, Ramirez-Lee et al. (2021) "Can SARS-CoV-2 infect the central nervous system via the olfactory bulb or the blood-brain barrier?" *Brain Behav Immun*
202. Wellford, Moseman (2024) "Olfactory immune response to SARS-CoV-2" *Cell Mol Immunol*
203. Xu, Ilyas, Weng (2023) "Endothelial dysfunction in COVID-19: an overview of evidence, biomarkers, mechanisms and potential therapies" *Acta Pharmacol Sin*
204. Andreu, Ripa, López-Guerrero et al. (2024) "Human coronavirus 229E uses clathrin-mediated endocytosis as a route of entry in Huh-7 cells" *Biomolecules*
205. Principi, Bosis, Esposito (2010) "Effects of coronavirus infections in children" *Emerg Infect Dis*
206. Vabret, Dina, Gouarin et al. (2006) "Detection of the new human coronavirus HKU1: a report of 6 cases" *Clin Infect Dis*
207. Pyrc, Sims, Dijkman et al. (2010) "Culturing the unculturable: human coronavirus HKU1 infects, replicates, and produces progeny virions in human ciliated airway epithelial cell cultures" *J Virol*
208. Risku, Lappalainen, Räsänen et al. (2010) "Detection of human coronaviruses in children with acute gastroenteritis" *J Clin Virol*
209. Orenstein, Banach, Baker (2008) "Morphogenesis of coronavi rus HCoV-NL63 in cell culture: a transmission electron microscopic study" *Open Infect Dis*
210. Siragam, Maltseva, Castonguay et al. (2024) "Seasonal human coronaviruses OC43, 229E, and NL63 induce cell surface modulation of entry receptors and display host cellspecific viral replication kinetics" *Microbiol Spectr*
211. Ding, He, Zhang et al. (2004) "Organ distribution of severe acute respiratory syndrome (SARS) associated coronavirus (SARS-CoV) in SARS patients: implications for pathogenesis and virus transmission pathways" *J Pathol*
212. Lee, Hui, Wu et al. (2003) "A major outbreak of severe acute respiratory syndrome in Hong Kong" *N Engl J Med*
213. Rabaan, Smajlović, Tombuloglu et al. (2022) "SARS-CoV-2 infection and multi-organ system damage: a review" *Bosn J Basic Med Sci*
214. Fabrizi, Nardelli, Regalia et al. (2024) "Are kidneys affected by SARS-CoV-2 infection? An updated review on COVID-19associated AKI" *Pathogens*
215. Heneghan, Spencer, Brassey et al. (2021) "SARS-CoV-2 and the role of orofecal transmission: a systematic review" *F1000Res*
216. Durairajan, Singh, Saravanan et al. (2023) "Gastrointestinal manifestations of SARS-CoV-2: transmission, pathogenesis, immunomodulation, microflora dysbiosis, and clinical implications" *Viruses*
217. Heinen, Klöhn, Westhoven et al. (2024) "Host determinants and responses underlying SARS-CoV-2 liver tropism" *Curr Opin Microbiol*
218. Wanner, Andrieux, Badia-I-Mompel et al. (2022) "Molecular consequences of SARS-CoV-2 liver tropism" *Nat Metab*
219. Meyerholz, Lambertz, Mccray (2016) "Dipeptidyl peptidase 4 distribution in the human respiratory tract" *Am J Pathol*
220. Panchapakesan, Pollock (2015) "The role of dipeptidyl peptidase -4 inhibitors in diabetic kidney disease" *Front Immunol*
221. Alghamdi, Mushtaq, Awn et al. (2015) "MERS CoV infection in two renal transplant recipients: case report" *Am J Transplant*
222. Ramadan, Shaib (1155) "Middle East respiratory syndrome coronavirus (MERS-CoV): a review" *Germs*
223. Zhou, Li, Zhao et al. (2017) "Human intestinal tract serves as an alternative infection route for Middle East respiratory syndrome coronavirus" *Sci Adv*
224. Chen, Li, Chen et al. (2020) "The ACE2 expression in human heart indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2" *Cardiovasc Res*
225. Pal, Ahirwar, Sakarde et al. (2021) "COVID-19 and cardiovascular disease: a review of current knowledge" *Horm Mol Biol Clin Investig*
226. Palazzuoli, Giustozzi, Ruocco et al. (2021) "Thromboembolic complications in Covid-19: from clinical scenario to laboratory evidence" *Life*
227. Peretto, Limite, Cianflone (2014) "Postoperative arrhythmias after cardiac surgery: incidence, risk factors, and therapeutic management" *Cardiol Res Pract*
228. Madjid, Safavi-Naeini, Solomon et al. (2020) "Potential effects of coronaviruses on the cardiovascular system: a review" *JAMA Cardiol*
229. Chen, Kong, Zhang et al. (2022) "DPP4 as a potential candidate in cardiovascular disease" *J Inflamm Res*
230. Alhogbani (2016) "Acute myocarditis associated with novel Middle East respiratory syndrome coronavirus" *Ann Saudi Med*
231. Saá, Fink, Bakkour et al. (2022) "Frequent detection but lack of infectivity of SARS-CoV-2 RNA in presymptomatic, infected blood donor plasma" *J Clin Invest*
232. De Melo, Perraud, Alvarez et al. (2023) "Neuroinvasion and anosmia are independent phenomena upon infection with SARS-CoV-2 and its variants" *Nat Commun*
233. Zeng, Evans, King et al. (2021) "SARS-CoV-2 spreads through cell-to-cell transmission" *bioRxiv*
234. Gupta, Madhavan, Sehgal et al. (2020) "Extrapulmonary manifestations of COVID-19" *Nat Med*
235. Li, Wohlford-Lenane, Perlman et al. (2016) "Middle East Minireview mBio November"
236. "respiratory syndrome coronavirus causes multiple organ damage and lethal disease in mice transgenic for human dipeptidyl peptidase 4" *J Infect Dis*
237. Zhao, Jiang, Qiu et al. (2015) "Multi-organ damage in human dipeptidyl peptidase 4 transgenic mice infected with Middle East respiratory syndrome-coronavirus" *PLoS One*
238. Jones, Patel, Levy et al. (2008) "Global trends in emerging infectious diseases" *Nature*
239. Glidden, Nova, Kain et al. (2021) "Human-mediated impacts on biodiversity and the consequences for zoonotic disease spillover" *Curr Biol*
240. Plowright, Reaser, Locke et al. (2021) "Land use-induced spillover: a call to action to safeguard environmental, animal, and human health" *Lancet Planet Health*
241. Keesing, Ostfeld (2021) "Impacts of biodiversity and biodiversity loss on zoonotic diseases" *Proc Natl Acad Sci*
242. Banerjee, Kulcsar, Misra et al. (2019) "Bats and coronaviruses" *Viruses*
243. Hu, Zeng, Yang et al. (2017) "Discovery of a rich gene pool of bat SARS-related coronaviruses provides new insights into the origin of SARS coronavi rus" *PLoS Pathog* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12702985&blobtype=pdf | # Integrated PRRSV prevention and control strategy based on the One Health concept: across the boundaries of virology, ecology and public health
Makoto Ujike, Ayako Miyazaki, Xiaobing Li, Hongbo Chen, Chengzhen Weng, Xinxin Huang, Dianning Duan
## Abstract
Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) poses a major threat to global swine production, with substantial economic losses and serious animal welfare concerns. Although PRRSV is not considered a zoonotic agent, its control exemplifies the necessity of a One Health approach, incorporating virological, ecological, immunological, and agricultural dimensions. This article contends that the impact of PRRSV extends beyond porcine populations, significantly contributing to the emergence of antimicrobial resistance (AMR) via secondary bacterial infections and the consequent misuse of antibiotics. Moreover, the environmental persistence of the virus and its potential for indirect transmission raise critical ecological questions that remain unresolved. By synthesizing current evidence, this review delineates the complex interrelationships among PRRSV outbreaks, patterns of antimicrobial use, and environmental contamination. This study propose an integrated One Health framework for PRRSV surveillance and control, emphasizing the implementation of genomic tools, systematic environmental monitoring, and enhanced collaboration among public health, veterinary, and environmental sectors. Integrating these disciplines is crucial to alleviating the multidimensional challenges posed by PRRSV, thereby protecting animal welfare, supporting sustainable agriculture, and strengthening global public health.
## 1 Introduction
Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) remains one of the most economically significant pathogens in global swine production. The virus primarily targets macrophages in pigs, weakening their immune systems and making infected animals highly susceptible to secondary infections (Sun et al., 2023). Clinical manifestations include reproductive disorders in sows (with abortion rates exceeding 30%) and respiratory diseases across all age groups, particularly acute respiratory symptoms in piglets that can result in mortality rates as high as 80%-100% (Cui et al., 2022;Yim-Im et al., 2023).
Despite decades of intensive eorts by global scientific and industrial communities, controlling Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) remains a formidable challenge. This stems from the virus's high mutability, persistent infection capabilities, and complex pathogenic mechanisms. Current vaccines demonstrate limitations in providing protection, failing to stimulate potent neutralizing antibody responses while also raising safety concerns. Compounding these issues, PRRSV transmission extends beyond pig populations through environmental vectors like air and contaminated water, as well as human activities such as transportation and trade. Traditional PRRSV control strategies primarily focus on herd management measures such as vaccination, biosecurity protocols, and quarantine systems. While these approaches are crucial, they often overlook the ecological dimensions of viral transmission and human-related factors (Gao and Wen, 2025). The emergence of the One Health concept has led us to recognize that human health, animal health, and environmental health form an interconnected whole (Desvars-Larrive et al., 2024). This holistic perspective oers a fresh approach to PRRSV control, requiring interdisciplinary and cross-sectoral collaboration to address this complex challenge. This study aims to explore the application of One Health framework in PRRSV prevention and control, analyze the current gaps in monitoring and control, and propose an integrated framework integrating virology, ecology, epidemiology and public policy to provide a scientific basis for more eective and sustainable PRRSV management strategies.
2 PRRSV in a One Health context: beyond the pig
## 2.1 Viral characteristics and multidirectional transmission mechanisms
Porcine Reproductive and Respiratory Syndrome Virus is a membrane-coated single-stranded positive-sense RNA virus with high genetic diversity, classified into two genotypes: European (Type 1) and American (Type 2). While primarily transmitted through direct contact, studies reveal its transmission pathways extend far beyond this. The virus can spread over short to medium distances via aerosols, a transmission route confirmed by molecular evidence in field studies (Hu et al., 2023;Zhang et al., 2024). A recent study systematically summarized the stability of PRRSV under various environmental conditions, particularly highlighting temperature as a critical factor influencing its survival outside a host (Mesa et al., 2024). More importantly, PRRSV demonstrates strong environmental persistence, with studies indicating survival in feces for up to several weeks and in wastewater for several days, potentially spreading indirectly through contaminated objects, water sources, and human activities (Arruda et al., 2019;Fan et al., 2024). Studies indicate that PRRSV-infected pigs not only shed the virus through respiratory secretions but also via fecal matter, with fecal shedding posing a significant risk for environmental contamination and between-farm transmission through sewage systems, feed supplies, and transport vehicles (Mesa et al., 2024).
Serological testing shows positive results for PRRSV in infected piglets from day 3 onward, with fecal-positive detection emerging by day 5. By day 7, fecal viral shedding reaches peak levels (10 3.9 copies/0.1 g) (Arruda et al., 2019). This multi-channel shedding mechanism heightens environmental contamination risks, potentially allowing virus transmission between farms through sewage systems, feed supplies, and transport vehicles. PRRSV exhibits high mutation rates (4.7-9.8 × 10 2 /site/year) and frequent recombination events, leading to diverse lineages and sublineages. In China, lineages 1, 3, 5, and 8 co-circulate, with lineage 1 (NADC30-like) currently dominant. Recombination between sublineages (e.g., 1.8 and 8.7) further complicates control eorts (Zhang et al., 2024;Zhou et al., 2024).
## 2.2 The between PRRSV infection and antibiotic
Porcine Reproductive and Respiratory Syndrome Virusinduced immunosuppression often leads to bacterial secondary infections such as Streptococcus suis infection, Haemophilus parasuis disease, and Actinobacillus pleuropneumoniae pneumonia. Clinically, this has resulted in increased antibiotic reliance within the pig farming industry, with usage potentially rising by 30%-50% during the nursery phase. This practice not only raises production costs but also accelerates the development of antimicrobial resistance (AMR), becoming a critical concern in the One Health framework (Trevisi et al., 2022;Machado et al., 2024). It is worth noting that the abuse of antibiotics may further disrupt the intestinal microbial balance of pigs, aect immune function, and form a vicious circle. Therefore, eective control of PRRSV infection itself is one of the important strategies to reduce the use of antibiotics in pig industry, which is of great significance for alleviating the global AMR crisis (Figure 1).
## 2.3 Potential zoonotic risks and cross species transmission possibilities
Although there is currently no documented evidence of direct human infection by PRRSV, its structural similarities to human receptors and high mutation rate suggest a theoretical potential for zoonotic transmission (Gorp et al., 2010;Liu et al., 2025). Research has found that PRRSV can enter cells using receptors in pigs (such as CD163), while human cells also have similar receptors, which theoretically may have a molecular basis for cross species transmission (Gorp et al., 2010;Liu et al., 2025). On the other hand, pigs play a crucial role as "mixers" for influenza viruses in virus reassortment. The immune suppression caused by PRRSV infection may accelerate the evolution and transmission of other viruses (such as influenza virus), indirectly increasing the risk of zoonotic diseases, a concern also noted in other reviews (Lagumdzic et al., 2023). Moreover, PRRSV's immunosuppressive eects in pigs could accelerate the evolution of co-circulating viruses, such as influenza, indirectly increasing zoonotic risks (Rajeev et al., 2020). The inherent high mutation rate and recombination capability of RNA viruses like PRRSV provide a theoretical basis for host The transmission path, monitoring nodes, and intervention strategies of Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) in the One Health framework.
adaptability, which is a key factor in assessing potential crossspecies transmission risks from a One Health perspective (Russell et al., 2017;Ajuwon et al., 2022). In addition, PRRSV induced immunosuppression in pigs can trigger a surge of secondary bacterial infections, many of which are zoonotic, thereby posing a potential zoonotic risk from a One Health perspective (Huong et al., 2016). This indirect impact is often overlooked by traditional prevention and control strategies, but it deserves high attention under the One Health framework.
3 Current gaps in PRRSV surveillance and control
## 3.1 Systematic lack of environmental monitoring
The current PRRSV monitoring system mainly focuses on clinical case reports and pig herd testing, and almost completely ignores virus monitoring in the environment. Research has shown that PRRSV can survive for a long time in feces, sewage, and soil (Table 1), but the role of these environmental reservoirs in virus transmission has not been fully evaluated and monitored (Alvarez-Norambuena et al., 2025). The lack of systematic environmental sampling schemes and standardized detection methods hinders our comprehensive understanding of the transmission dynamics and environmental residual risks of PRRSV. Especially in high-density aquaculture areas, the viral load in the environment may continue to be high, leading to the risk of reinfection even on farms that implement strict biosecurity measures (Makau et al., 2021). The neglect of this environmental transmission pathway is one of the important limitations of current PRRSV control.
## 3.2 Shortcomings and limitations of genome monitoring applications
Although next-generation sequencing technology (NGS) has been widely used for monitoring many infectious diseases, it has not been fully utilized in PRRSV control. At present, whole genome sequencing (WGS) is mainly used in research scenarios and local epidemic investigations, lacking a systematic and largescale genome monitoring network (Kim et al., 2022;Xing et al., 2022). This limitation prevents us from fully understanding the evolutionary dynamics and transmission pathways of PRRSV. Especially for the insuÿcient monitoring of recombinant strains, it is diÿcult to cope with the challenge of increasing virus diversity. The diversity of PRRSV strains has increased the diÿculty of PRRS control in China, and the recombination between dierent strains is very serious (Xing et al., 2022). Establishing a national or even global PRRSV genome database and sharing platform is crucial for real-time tracking of virus mutations and transmission.
## 3.3 Lack of cross departmental collaboration mechanism
Porcine Reproductive and Respiratory Syndrome Virus control has traditionally been seen as the responsibility of aquaculture and veterinary departments, lacking eective collaboration with public health, environmental protection, and wildlife management departments. This departmental gap leads to a lack of overall prevention and control strategies, making it diÿcult to cope with complex environmental transmission and transmission risks mediated by human activities (Osemeke et al., 2025). For example, farm wastewater treatment and discharge may involve environmental protection departments, while wildlife (such as wild boars) as potential vectors of transmission may involve wildlife management departments. Lack of collaborative participation from these departments often results in blind spots and loopholes in prevention and control measures. The One Health approach emphasizes the importance of cross departmental collaboration, but in practice, this collaboration is still limited.
## 4 A proposed One Health framework for PRRSV
## 4.1 Novel integration of genomic and environmental monitoring
Establishing a comprehensive PRRSV genome monitoring network is a core component of the One Health framework. This network should integrate: clinical isolates WGS: whole genome sequencing of PRRSV in clinical samples to monitor virus variation and evolutionary dynamics; Environmental sample testing: Regularly collect environmental samples (sewage, soil, air, etc.) for virus testing and sequencing; Data sharing platform: Establish standardized data formats and sharing mechanisms to promote data exchange between dierent regions and institutions (Xie et al., 2021). In recent years, the development and cost reduction of metagenomics technology have provided feasible tools for environmental monitoring. Research has shown that monitoring based on metagenomics can identify epidemic signals in advance, providing valuable time windows for intervention measures. Meanwhile, integrating genomic data with epidemiological data can better track transmission pathways and identify transmission hotspots (Barton and Colijn, 2023). Our proposed framework introduces a novel integration of genomic surveillance with systematic environmental monitoring, addressing a critical gap in current PRRSV control strategies. Unlike previous approaches, our framework emphasizes real-time data sharing and metagenomic analysis to enhance early detection capabilities. The novelty of our proposal lies not in these individual technologies, but in their systematic integration across the One Health domains and the creation of a feedback loop to inform interventions, as visualized in Figure 1.
## 4.2 A novel three-tiered collaborative governance model
An eective One Health response requires the establishment of institutionalized collaboration mechanisms, that connect traditionally independent departments and disciplines. We suggest establishing a three-level collaboration framework.
Level 1: Local collaborative network, connecting farms, local veterinarians, environmental protection departments, and public health institutions, responsible for daily monitoring and information sharing. Level 2: Regional Expert Committee, composed of virologists, epidemiologists, veterinarians, ecologists, and public health experts, responsible for data analysis and policy recommendations. Level 3: National and international coordination agencies responsible for overall coordination, standard setting, and resource allocation. This hierarchical structure is a key conceptual contribution of this study, designed to translate the principle of One Health into a practical, actionable governance chain for PRRSV control.
## 4.3 Differentiating existing strategies from our Integrated, risk-based approach
In situations where resources are limited, adopting riskbased precision prevention and control strategies can improve eÿciency. Elements like biosecurity, vaccination, and monitoring are standard practice. However, the core of our proposed framework is the dynamic, data-driven integration of these elements based on a continuous risk assessment informed by the surveillance system (Figure 1). We propose a stratified approach where:
High risk areas (such as high breeding density and high prevalence): Implement strengthened monitoring (including environmental monitoring), strict biosafety measures, and regional vaccination plans.
Medium risk areas: Implement routine monitoring and standard prevention and control measures, with a focus on input risks.
Low risk areas: focusing on monitoring, emphasizing early detection and rapid response capacity building.
It is worth noting that the behavior of farmers is a key factor aecting the eectiveness of prevention and control. Research shows that the willingness to take action of "non-inheritor type" breeders is significantly lower than that of other types of breeders (Haile et al., 2025). Adopting dierentiated communication and education strategies for farmers with dierent characteristics can improve the compliance and eectiveness of prevention and control measures. This represents a shift from a one-size-fits-all application of measures to a precision-guided deployment, which is a central tenet of our proposed framework.
5 Challenges and future directions
## 5.1 Overcoming technical challenges: the authors' proposed solutions
The implementation of the One Health framework faces multiple challenges. Technical challenges include low virus concentrations in environmental samples, high detection sensitivity requirements, and the complexity of large-scale genomic data analysis. This requires the development of more sensitive and economical detection methods, as well as standardized bioinformatics processes (Kumblathan et al., 2021;Chen et al., 2024). We propose the development of standardized bioinformatics pipelines to manage large-scale genomic data, a solution not previously addressed in PRRSV literature.
## 5.2 Future research priorities stemming from the proposed framework
The challenges of resources and policies cannot be ignored, especially in low -and middle-income countries. Establishing a cross departmental collaboration mechanism requires policy support and financial investment, as well as overcoming cultural dierences and communication barriers between departments. On a global scale, it is necessary to strengthen international cooperation and establish a PRRSV global monitoring network and data sharing platform, similar to the human infectious disease monitoring system.
Future research directions should include several critical areas that are essential for operationalizing and validating our proposed framework. Validation of the Integrated Surveillance System: Research is needed to assess the cost-eectiveness and practical implementation of the combined genomic and environmental surveillance network we propose. Operational Research on the Governance Model: The eectiveness of the three-tiered collaborative model requires empirical testing and refinement in dierent regional contexts. Refining the Risk Assessment Model: Future work should focus on developing and validating quantitative risk models that integrate the multi-source data (genomic, environmental, AMR) outlined in our framework to automate and improve the precision of the risk-based interventions. Virus ecology research: exploring the survival ability and transmission eÿciency of PRRSV under dierent environmental conditions. Cross species transmission risk assessment: Evaluate the potential risks of PRRSV cross species transmission through experimental research and molecular simulations. Eectiveness evaluation of intervention measures: Compare the cost-eectiveness of dierent prevention and control strategies to provide scientific basis for policy formulation. Of particular concern is that climate change may aect the transmission dynamics and geographic distribution of PRRSV. Temperature, humidity, and extreme weather events may alter the survival time and transmission patterns of viruses in the environment, while also potentially aecting the distribution and abundance of vector organisms. Incorporating climate change factors into PRRSV risk models is an important direction for future research.
## 6 Final considerations
The control of Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) necessitates a paradigm shift from the traditional single-pathogen-single-host model toward a holistic One Health approach. This transition requires not only technological innovation but also a fundamental evolution in mindset and collaborative mechanisms. By integrating perspectives from human, animal, and environmental health, we can achieve a more comprehensive understanding of PRRSV transmission dynamics and develop more eective intervention strategies. The One Health framework proposed in this studythe integrative feedback between surveillance and intervention, operationalized through a structured three-tiered governance model-provides a novel pathway for PRRSV management. While its implementation may face practical challenges, the potential benefits are substantial, including reduced economic impact on the swine industry, decreased antibiotic usage, mitigated antimicrobial resistance (AMR), and enhanced sustainability in agriculture and public health.
Given the evolving landscape of global swine production and the dynamic interfaces between humans, animals, and the environment, adopting a One Health approach to address animal disease threats has become increasingly imperative. PRRSV control can serve as a model for tackling complex health challenges through interdisciplinary collaboration, oering valuable insights and reference experience for the prevention and control of other emerging and endemic animal diseases.
## References
1. Ajuwon, Roper, Richardson et al. (2022) "One health approach: A data-driven priority for mitigating outbreaks of emerging and reemerging zoonotic infectious diseases" *Trop. Med. Infect. Dis*
2. Alvarez-Norambuena, Quinonez-Munoz, Corzo et al. (2025) "Comparative adsorption of porcine reproductive and respiratory syndrome virus strains to minnesota soils" *Viruses*
3. Arruda, Tousignant, Sanhueza et al.
4. (2019) "Aerosol detection and transmission of porcine reproductive and respiratory syndrome virus (PRRSV): What is the evidence, and what are the knowledge gaps?" *Viruses*
5. Barton, Colijn (2023) "Genomic, clinical and immunity data join forces for public health" *Nat. Rev. Microbiol*
6. Chen, Li, Xu et al. (2024) "Characterising global antimicrobial resistance research explains why One Health solutions are slow in development: An application of AI-based gap analysis" *Environ. Int*
7. Cui, Xia, Huang et al. (2022) "Analysis of recombinant characteristics based on 949 PRRSV-2 genomic sequences obtained from 1991 to 2021 shows that viral multiplication ability contributes to dominant recombination" *Microbiol. Spect*
8. Desvars-Larrive, Vogl, Puspitarani et al. (2024) "Oropharyngeal swab sampling for PRRSV detection in large-scale pig farms: A convenient and reliable method for mass sampling" *Porcine Health Manage*
9. Gao, Wen, Gorp et al. (2010) "Identification of the CD163 protein domains involved in infection of the porcine reproductive and respiratory syndrome virus" *Virol. J*
10. Haile, Liu, Carrai et al. (2025) "Characterization of biosecurity practices and viral infections on pig farms in Hong Kong" *Prev. Vet. Med*
11. Hu, Tian, Lai et al. (2023) "Airborne transmission of common swine viruses" *Porcine Health Manage*
12. Huong, Thanh, Phu et al. (2016) "Temporal and spatial association of Streptococcus suis infection in humans and porcine reproductive and respiratory syndrome outbreaks in pigs in northern Vietnam" *Epidemiol. Infect*
13. Kim, Moon, Jeong et al. (2022) "Whole-genome sequencing and genetic characteristics of representative porcine reproductive and respiratory syndrome virus (PRRSV) isolates in Korea" *Virol. J*
14. Kumblathan, Liu, Uppal et al. (2021)
15. "Wastewater-Based epidemiology for community monitoring of SARS-CoV-2: Progress and challenges" *ACS Environ. Au*
16. Lagumdzic, Pernold, Ertl et al. (2023) "Gene expression of peripheral blood mononuclear cells and CD8+ T cells from gilts after PRRSV infection" *Front. Immunol*
17. Liu, Huang, Yang et al. (2025) "GP2a I118 and GP4 D43 play critical roles in the attachment of PRRSV to the CD163 receptor: Implications for anti-PRRSV infection targets" *J. Virol*
18. Machado, Petznick, Poeta et al. (2024) "Assessment of changes in antibiotic use in grow-finish pigs after the introduction of PRRSV in a naïve farrow-to-finish system" *Prev. Vet. Med*
19. Makau, Alkhamis, Paploski et al. (2021) "Integrating animal movements with phylogeography to model the spread of PRRSV in the USA" *Virus Evol*
20. Mesa, Munoz, Sobhy et al. (2024) "Survival of porcine reproductive and respiratory syndrome virus (PRRSV) in the environment" *Vet. Sci*
21. Osemeke, Silva, Corzo et al. (2016) "Economic impact of productivity losses attributable to porcine reproductive and respiratory syndrome virus in United States pork production" *Prev. Vet. Med*
22. Rajeev, Prathiviraj, Kiran et al. (2020) "Zoonotic evolution and implications of microbiome in viral transmission and infection" *Virus Res*
23. Russell, Katz, Richgels et al. (2017)
24. "A framework for modeling emerging diseases to inform management" *Emerg. Infect. Dis*
25. Sun, Chen, Liu et al.
26. "PRRSV-induced inflammation in pulmonary intravascular macrophages (PIMs) and pulmonary alveolar macrophages (PAMs) contributes to endothelial barrier function injury" *Vet. Microbiol*
27. Trevisi, Amatucci, Ruggeri et al. (2022) "Pattern of antibiotic consumption in two italian production chains diering by the endemic status for porcine reproductive and respiratory syndrome" *Front. Vet. Sci*
28. Xie, Duan, Zeng et al. (2021) "Clinical metagenomics assessments improve diagnosis and outcomes in community-acquired pneumonia" *BMC Infect. Dis*
29. Xing, Zheng, Cao et al. (2022) "Whole genome sequencing of clinical specimens reveals the genomic diversity of porcine reproductive and respiratory syndrome viruses emerging in China"
30. Yim-Im, Anderson, Paploski et al. (2023) "Refining PRRSV-2 genetic classification based on global ORF5 sequences and investigation of their geographic distributions and temporal changes" *Microbiol. Spect*
31. Zhang, Li, Xie et al. (2024) "Comparing the molecular evolution and recombination patterns of predominant PRRSV-2 lineages co-circulating in China" *Front. Microbiol*
32. Zhou, Han, Yang (2024) "The evolution and diversity of porcine reproductive and respiratory syndrome virus in China" *Vet. Microbiol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12683980&blobtype=pdf | # High-Throughput Targeted Sequencing Identifies an HPV Methylation Panel for Detecting Cervical Lesion Progression
Hui Liu, Jie Zhou, Yunxia Xiao, Yuming Zheng, | Liyao, Dirong Dong, Yanqing Shen, Wen Zhang, Wei Guo, Rui Tian, Xun Tian, Xia Huang, | Hu, Lili Sun, Chen Cao, Liyao Yu, Zheng Hu
## Abstract
High-risk human papillomavirus (hrHPV) infection is the primary cause of cervical cancer. However, hrHPV testing lacks specificity in detecting neoplastic changes. This study explored the utility of quantitative methylated HPV DNA markers for precise detection of cervical lesions. Using hybridization capture-based bisulfite sequencing, we analyzed genome-wide HPV methylation patterns. The study included a training cohort of 60 cervical exfoliated cell samples and a validation cohort of 29 samples. Analysis of 112 CpG sites across the HPV genome revealed that genome-wide HPV16 methylation levels correlated with disease progression. Squamous cell carcinoma (SCC) showed 1.4-fold higher methylation levels compared to normal tissue (p = 0.0032). Progressive methylation increases in the E5-α and L2 genes were observed across the spectrum of cervical lesion severity, from normal tissue through high-grade squamous intraepithelial lesion (HSIL) to SCC. Intersection analysis of differentially methylated CpG sites between HSIL vs Normal and SCC vs Normal identified 16 consistently hypermethylated CpG sites in the E5-α, E7, L2, and L1 genes, distinguishing both HSIL and SCC from normal tissue. This pilot study identifies a five-CpG methylation panel (E5-α_3887, E5-α_3941, E7_701, L2_4441, and L2_5128) as promising triage biomarkers for HPV16positive women, achieving high discriminatory performance (AUC = 0.919 in a validation cohort) for detecting cervical lesions.This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.
This genome-wide capture sequencing identified novel HPV16 methylation markers that distinguish cervical lesions from normal tissue, supporting the feasibility of HPV methylation-based triage for HPV-positive women in cervical cancer screening.
## 1 | Introduction
Persistent infections with high-risk HPV (hrHPV) is the essential etiologic factor of cervical cancer [1]. hrHPV testing has replaced cytology as the primary screening method for cervical cancer prevention [2]. However, most hrHPV infections are transient [3,4]. Effective triage of HPV-positive samples is therefore needed to inprove the specificity of HPV testing, minimize unnecessary referrals, and optimize detection of clinically significant lesions [5].
Several biomarkers have shown considerable potential in addressing this critical clinical need [6]. Triage testing options, including dual immunostaining for p16/Ki-67 [7,8], have proven valuable despite requiring for additional sample processing and microscopic evaluation. In addition, DNA methylation-based assays, including the S5 [9,10] and WID-qCIN classifiers [11], have demonstrated strong clinical performance. Nevertheless, current methods generally rely on targeted PCR-based designs and separate workflows. The development of additional biomarkers therefore remains necessary, particularly those can be integrated into a single assay to simultaneously provide HPV genotyping and triage biomarker information [12].
Advances in next-generation sequencing technologies have created new opportunities for comprehensive molecular characterization of HPV infections [13]. In this pilot study, we employed capture-based bisulfite sequencing to generate genome-wide methylation profiles of HPV16 in a cohort of HPV16-positive women. We aimed to identify informative methylation markers capable of distinguishing high-grade cervical lesions and carcinoma from normal epithelium.
## 2 | Materials and Methods
## 2.1 | Patients and Specimens
The study was approved by the Ethics Committee of the Guangdong North People's Hospital (SUMC-IRB-2018) and the Ethics Committee of the Wuhan Central Hospital (WHZXKYL2022-047). Women tested positive for HPV16 single infection by Roche Cobas 4800 HPV test in our outpatient department from December 2020 to December 2023 were recruited. Written informed consent was obtained from each recruited patient. Cervical biopsies under colposcopy were interpreted by experienced pathologists for histological diagnosis. Recruited patients were divided into three groups based on pathological results: normal cervical epithelium or cervicitis without atypical hyperplasia (Normal), high-grade squamous intraepithelial lesion (HSIL), and squamous cell carcinoma (SCC). In China, clinical practice follows the WHO classification guidelines [14,15], where both CIN2 and CIN3 are classified as HSIL requiring similar management approaches. This guided our decision to group CIN2 and CIN3 cases together as HSIL group. Given that the majority of low-grade squamous intraepithelial lesions (LSILs) are likely to regress, patients with LSIL were not included in this study. All cervical exfoliated cell samples were collected and stored at -80°C at the time of diagnosis before any treatment. Genomic DNA was extracted from cervical exfoliated cells using the TIANamp Genomic DNA Kit (TIANGEN) according to the manufacturer's protocol.
## 2.2 | Bisulfite Conversion and Library Construction
For each sample, after DNA quantification and purity assessment, 1 ug of genomic DNA, spiked with 1% unmethylated standard DNA, was sheared into 600 bp fragments using Bioruptor Pico (Diagenode). The EpiArt Ultrafast DNA Methylation Bisulfite Kit (Vazyme) was employed to convert unmethylated cytosine of 600 ng of sheared DNA according to the instructions. Methylation libraries were subsequently constructed using the EpiArt DNA Methylation Bibrary Kit (Vazyme) following instruction. After purification using the Agencourt AMPure XP beads (Beckman), the library concentration was quantified using Qubit 4.0 Fluorometer, and the fragment size was assessed by a Qsep100 bioanalyzer (BIOptic).
## 2.3 | HPV Target Enrichment, Hybridization Capture and Sequencing
The HPV16 reference genome was obtained from the Papillomavirus Episteme (PaVE) database (http://pave.niaid. nih.gov). A custom biotinylated DNA probe panel targeting HPV16 genome (Generulor Company, Zhuhai, China) was employed for targeted capture hybridization. The captured sequence data were analyzed using the custom Virus Integration and Presence Analysis (VIPA) algorithm [16], which simultaneously enables complete viral genome assemblies, HPV genotyping, HPV SNP profiling, and viral-host integration mapping.
To confirm probe specificity in the present study, ten clinically confirmed HPV16-negative cervical specimens were processed as negative controls using identical bisulfite conversion, library preparation, and hybridization capture procedures. These samples generated negative HPV16 sequencing results (depth < 10 ×, coverage < 0.5, and HPV16 reads < 100), confirming the specificity of our probe design.
Hybridization capture was performed as follows: the libraries were hybridized with custom HPV16-specific biotinylated DNA probes at 65°C for 16 h. After hybridization, the captured DNA fragments were isolated using streptavidin magnetic beads (Generulor). Following the capture and washing steps, the DNA was amplified through 10-12 cycles of polymerase chain reaction. Library concentrations and quality were analyzed using Qubit and Qsep100 bioanalyzer. Library sequencing was performed on MGISEQ-T7 sequencer (MGI), generating pairedend 150 bp reads.
## 2.4 | HPV DNA Methylation Analysis
Raw sequencing data were processed to trim the adapter sequences and filter out the low-quality reads by FastQC The HPV methylation rate of a specific cytosine site was calculated as the proportion of the number of reads supporting methylation to the total number of reads covering the cytosine. Median methylation rates of all CpGs in each gene produced HPV methylation levels by gene. Median methylation rates of all genes in each sample produced the genome-wide HPV methylation levels by sample.
## 2.5 | Statistical Methods
Analysis was performed using the R software package. The Wilcoxon rank-sum test was used to determine the differences in methylation level between two categories. Receiver operating characteristic (ROC) curve analysis, including calculation of the area under the ROC curve (AUC) using the R package "pROC," was used to assess the performance of different combinations of CpG loci to distinguish women with different severity of cervical lesions. Least absolute shrinkage and selection operator (LASSO)-penalized binomial regression analysis was applied to further reduce candidate methylation sites by using the R package "glmnet." The p values from multiple comparisons were adjusted using the Benjamini-Hochberg method. All significant differences were assessed using a two-tailed p < 0.05. 1. No statistically significant age differences were observed among the three groups (Supporting Information S1: Figure 1A). quality control measures (Figure 1B), the final validation cohort consisted of 29 participants: 15 in the Normal group, seven with HSIL, and seven with cervical cancer. Baseline characteristics of the validation cohort were presented in Table 1 and Supporting Information S1: Figure 1B. 1.
## 3 | Results
## 3.1 | Study Population and Characteristics
## 3.2 | Methylation
Genome-wide HPV DNA methylation levels demonstrated a positive association with cervical lesion severity (Figure 2A). Median methylation rose steadily from normal cervical tissue (30.2%) to HSIL (35.3%) and SCC (42.3%). Compared with the Normal group, SCC group displayed significantly higher methylation (1.4-fold; p = 0.0032). The HSIL group also exhibited elevated methylation relative to the Normal group (p = 0.04), whereas the difference between SCC and HSIL was not statistically significant (p = 0.19). Since approximately 50% of CIN2/HSIL lesions regress spontaneously, the distinct methylation signatures between HSIL and SCC groups may reflect the heterogenous nature of the HSIL cases.
Gene-specific methylation analysis, derived from median CpG methylation within each HPV gene region, revealed distinct upregulation patterns across disease severity groups. HSIL versus Normal comparison revealed significantly elevated methylation in E1/E2 (p = 0.042), E5-α (p = 0.001), L1 (p = 0.027), L2 (p = 0.001), and URR (p = 0.009; Figure 2B). In SCC versus Normal comparison, all HPV genes except E6 showed elevated methylation in cancer, with significant increases in six genes: E1/E2 (p = 0.016), E2 (p = 0.025), E5-α (p < 0.001), E7 (p = 0.002), L1 (p = 0.002), and L2 (p < 0.001; Figure 2C). Comparison of SCC versus HSIL groups demonstrated elevated methylation levels in all HPV genes except URR in the SCC group, with statistically significant elevation observed in E1, E5-α, and L2 (p = 0.016, 0.006, and 0.035, respectively; Figure 2D). Collectively, these findings demonstrate progressive increases in both genome-wide HPV methylation levels and gene-specific methylation, particularly in E5-α and L2, correlating with cervical lesion severity.
At the individual CpG methylation level, stage-specific patterns were evident. The HSIL versus Normal comparison identified 17 significantly higher methylated sites (AUC: 0.753-0.934; Figure 3A). Between SCC and Normal groups, 55 CpGs showed significant methylation elevation (AUC: 0.715-0.98; Figure 3B). Comparison of SCC versus HSIL samples identified 24 CpGs with significantly higher methylation in SCC (AUC: 0.748-0.909; Figure 3C). Supporting Information S1: Table 2 presents the adjusted p-values and cutoff values for all significantly differentiated CpG sites.
To evaluate the potential impact of inflammatory conditions on HPV methylation patterns, we performed a comparison within the control group between normal cervical epithelium (n = 10) and cervicitis samples (n = 10). No significant differences in genome-wide HPV16 methylation levels were observed between normal epithelium and cervicitis groups (median: 29.9% vs 30.7%, p = 0.91, Supporting Information S1: Figure 2A). Genelevel and individual CpG site methylation analyses also revealed no significant differences between these two subgroups (Supporting Information S1: Figure 2B).
## 3.3 | Candidate CpGs Showing Potential as Methylation Markers to Differentiate High-Grade Precancer/Cancer From Normal Cervix Was Identified
Since patients with HSIL and SCC require colposcopic evaluation and management, we focused on the comparison of HSIL & SCC versus normal controls. The intersection analysis of differentially methylated CpG sites between HSIL vs Normal (17 sites) and SCC vs Normal (55 sites) yielded 16 overlapping differentially methylated CpG sites (Figure 4A): two in E5-α (E5-α_3887 and E5-α_3941), one in E7 (E7_701), nine in L2 (L2_5617, L2_5608, L2_5128, L2_4240, L2_4441, L2_4437, L2_4906, L2_4894, and L2_5611), and four in L1 (L1_7034, L1_6665, L1_6389, and L1_6367). These CpG sites were consistently altered in both HSIL and SCC when compared to normal controls, demonstrating effective stratification of samples by cytological grade and cervical cancer status (Figure 4B). These differentially methylated CpGs located within E5-α, L1, and L2 gene regions were previously identified as potential diagnostic biomarkers [17][18][19][20][21][22]. The novel identified differentially methylated CpG sites was E7_701.
To develop a diagnostic classifier based on minimal number of CpGs, LASSO-penalized binomial regression analysis was applied to the 16 differentially methylated CpGs in the training cohort. This analysis identified a five-CpG classifier, consisting of E5-α_3887, E5-α_3941, E7_701, L2_4441, and L2_5128, which demonstrated optimal performance across multiple comparisons. 0.0056 × L2_5128) -3.6103, achieving an AUC of 0.872 (cutoff: 0.461; Figure 5C).
To further evaluate the clinical utility of the five-CpG classifier, we combined the SCC and HSIL groups into cervical lesion group and compared it against the Normal group. ROC analysis of the classifier yielded the following risk score formula: Risk score = (-0.0061 × E5-α_3887) + (0.0140 × E5-α_3941) + (0.0167 × E7_701) + (0.0784 × L2_4441) + (0.0207 × L2_5128) -3.7577, achieving an AUC of 0.9 for distinguishing cervical lesions from normal controls (cutoff: 0.684; Figure 5D). This finding demonstrated that the five-CpG methylation panel represents a promising triage biomarker for risk stratification of HPV16-positive patients.
## 3.4 | Validation in an Independent Clinical Cohort
HPV16 DNA methylation analysis was performed on a validation cohort comprising 29 patients. The five-CpG classifier of E5-α_3887, E5-α_3941, E7_701, L2_4441, and L2_5128 showed promising discriminatory performance with AUC values of 0.905 for HSIL versus Normal (Figure 6A), 0.971 for SCC versus Normal (Figure 6B), and 0.776 for SCC versus HSIL (Figure 6C).
For the discrimination of cervical lesions (SCC and HSIL) from normal epithelium, the five-CpG panel achieved an AUC of 0.919 (Figure 6D).
## 4 | Discussion
The shift from cytology-based to hrHPV testing for cervical cancer primary screening demonstrated improved sensitivity and longer screening intervals [23]. However, effective molecular triage tools are still necessarily required to stratify hrHPVpositive women. Our study addresses this need by providing a comprehensive genome-wide analysis of HPV16 methylation patterns and identifying novel potential viral methylation biomarkers for risk stratification. Using hybridization capturebased bisulfite sequencing, we demonstrated the progressive increase in methylation levels of specific sites correlating with current disease severity-from normal through HSIL to cancer. Our five-CpG classifier demonstrated optimal performance with an AUC of 0.9 in the training cohort and 0.919 in the validation cohort. While the sensitivity for cervical cancer detection was not 100%, the high AUC and the underlying biology-showing markedly increased methylation in all cancer samples versus normal samples-are promising for a pilot study. Importantly, the observed hypermethylation is not a general feature of HPV infection but specifically associates with HSIL and SCC, which represent the targets for clinical intervention. This strong correlation with current disease severity represents a foundation for risk stratification and a necessary first step toward developing a prognostic marker.
When compared to established triage tools like the S5 classifier [9,10], our method provides the ability to assess a broader range of methylation sites across the HPV16 genome. The S5 classifier, a PCR-based assay targeting EPB41L3 and four HPV types (16, 18, 31, and 33), is effective for detecting mixed HPV infections but focuses specifically on methylation in the L1 and L2 regions of the HPV genome. In contrast, our genome-wide viral methylation profiling not only corroborates known methylated regions such as L2 [24,25] but also identifies novel sites such as E5-α and E7 missed by targeted methods [26,27]. Furthermore, we selected capture sequencing for HPV methylation testing because next-generation sequencing (NGS) has been shown in our previous work [28] and others [29] as a robust and accurate method for detecting HPV. While NGSbased HPV genotyping remains under investigation and has not yet received clinical approval, our approach for HPV methylation detection provides a foundation for integration with NGS-based hrHPV workflows, potentially reducing the need for additional sampling and separate triage assays. Future studies should explore how this approach can provide deeper molecular insights, such as HPV genotyping, viral variation and integration status, enabling more precise risk assessment.
Several limitations of this study warrant careful consideration. First, the limited sample size in both training (n = 60) and validation (n = 29) cohorts provides proof-of-concept data while highlighting the need for larger studies to fully characterize the viral methylome landscape. Second, our analysis only focused on HPV16, the most clinically significant hrHPV type accounting for approximately 50%-60% of cervical cancers. Clinical translation would require expansion to comprehensive multitype hrHPV panels, integration with host gene markers, and combination with existing screening modalities to achieve optimized sensitivity for cervical cancer screening.
## 5 | Conclusions
This genome-wide HPV methylation profiling identified a five-CpG markers that effectively distinguish cervical lesions from normal tissue, including novel CpG sites previously unreported by targeted approaches. The capture sequencing approach holds promise for integrating methylation-based triage into HPV testing workflows, as NGS-based viral detection platforms progress from research to clinical implementation. Future expansion to multiple hrHPV types and host genome markers offers a promising strategy to further enhance cervical cancer screening.
## References
1. Tainio, Athanasiou, Tikkinen (2018) "Clinical Course of Untreated Cervical Intraepithelial Neoplasia Grade 2 Under Active Surveillance: Systematic Review and Meta-Analysis" *BMJ*
2. Ronco, Dillner, Elfström (2014) "Efficacy of HPV-Based Screening for Prevention of Invasive Cervical Cancer: Follow-up of Four European Randomised Controlled Trials" *Lancet*
3. Mccredie, Sharples, Paul (2008) "Natural History of Cervical Neoplasia and Risk of Invasive Cancer in Women With Cervical Intraepithelial Neoplasia 3: A Retrospective Cohort Study" *Lancet Oncology*
4. Tainio, Athanasiou, Tikkinen (2018) "Clinical Course of Untreated Cervical Intraepithelial Neoplasia Grade 2 Under Active Surveillance: Systematic Review and Meta-Analysis" *BMJ*
5. Bouvard, Wentzensen, Mackie (2021) "The IARC Perspective on Cervical Cancer Screening" *New England Journal of Medicine*
6. Thrall, Mccarthy, Mito et al. (2025) "Triage Options for Positive High-Risk HPV Results From HPV-Based Cervical Cancer Screening: A Review of the Potential Alternatives to Papanicolaou Test Cytology" *Journal of the American Society of Cytopathology*
7. Yung-Taek Ouh, Kyong, Wook (2024) "p16/Ki-67 Dual Staining as a Predictive Value for Cervical Cancer Compared to Other Conventional Triage Tools: A Descriptive Literature Review" *European Journal of Gynaecological Oncology*
8. Ouh, Kim, Yi et al. (2024) "Enhancing Cervical Cancer Screening: Review of p16/Ki-67 Dual Staining as a Promising Triage Strategy" *Diagnostics*
9. Brentnall, Vasiljevic, Scibior-Bentkowska (2015) "HPV33 DNA Methylation Measurement Improves Cervical Pre-Cancer Risk Estimation of an HPV16, HPV18, HPV31 and\textitEPB41L3 Methylation Classifier" *Cancer Biomarkers*
10. Lorincz, Brentnall, Scibior-Bentkowska (2016) "Validation of a Dna Methylation HPV Triage Classifier in a Screening Sample" *International Journal of Cancer*
11. Schreiberhuber, Barrett, Wang (2024) "Cervical Cancer Screening Using DNA Methylation Triage in a Real-World Population" *Nature Medicine*
12. Gradíssimo, Burk (2017) "Molecular Tests Potentially Improving HPV Screening and Genotyping for Cervical Cancer Prevention" *Expert Review of Molecular Diagnostics*
13. Groves, Coleman (2018) "Human Papillomavirus Genome Integration in Squamous Carcinogenesis: What Have Next-Generation Sequencing Studies Taught us?" *Journal of Pathology*
14. Waxman, Chelmow, Darragh et al. (2012) "Revised Terminology for Cervical Histopathology and Its Implications for Management of High-Grade Squamous Intraepithelial Lesions of the Cervix" *Obstetrics & Gynecology*
15. "WHO Classification of Tumours"
16. Tian, Cui, He (2019) "Risk Stratification of Cervical Lesions Using Capture Sequencing and Machine Learning Method Based on HPV and Human Integrated Genomic Profiles" *Carcinogenesis*
17. Lorincz, Brentnall, Scibior-Bentkowska (2016) "Validation of a DNA Methylation HPV Triage Classifier in a Screening Sample" *International Journal of Cancer*
18. Frimer, Sun, Mcandrew (2015) "HPV16 CpG Methyl-Haplotypes Are Associated With Cervix Precancer and Cancer in the Guanacaste Natural History Study" *Gynecologic Oncology*
19. Mirabello, Frimer, Harari (2015) "HPV16 Methyl-Haplotypes Determined by a Novel Next-Generation Sequencing Method Are Associated With Cervical Precancer" *International Journal of Cancer*
20. Mirabello, Schiffman, Ghosh (2013) "Elevated Methylation of HPV16 DNA Is Associated With the Development of High Grade Cervical Intraepithelial Neoplasia" *International Journal of Cancer*
21. Brandsma, Harigopal, Kiviat (2014) "Methylation of Twelve CpGs in Human Papillomavirus Type 16 (HPV16) as an Informative Biomarker for the Triage of Women Positive for HPV16 Infection" *Cancer Prevention Research*
22. Anderson, Banister, Kassler (2016) "Human Papillomavirus Type 16 L2 DNA Methylation in Exfoliated Cervical Cells From College-Age Women" *Journal of lower genital tract disease*
23. Sankaranarayanan, Nene, Shastri (2009) "HPV Screening for Cervical Cancer in Rural India" *New England Journal of Medicine*
24. Louvanto, Franco, Ramanakumar (2015) "Methylation of Viral and Host Genes and Severity of Cervical Lesions Associated With Human Papillomavirus Type 16" *International Journal of Cancer*
25. Lorincz, Brentnall, Vasiljević (2013) "HPV16 L1 and L2 DNA Methylation Predicts High-Grade Cervical Intraepithelial Neoplasia in Women With Mildly Abnormal Cervical Cytology" *International Journal of Cancer*
26. Liu, Iden, Fye (2017) "Targeted, Deep Sequencing Reveals Full Methylation Profiles of Multiple HPV Types and Potential Biomarkers for Cervical Cancer Progression" *Biomarkers & Prevention*
27. Hillyar, Kanabar, Pufal (2022) "A Systematic Review and Meta-Analysis of the Diagnostic Effectiveness of Human Papillomavirus Methylation Biomarkers for Detection of Cervical Cancer" *Epigenomics*
28. Zhang, Tian, Chen (2021) "Feasibility and Accuracy of Menstrual Blood Testing for High-Risk Human Papillomavirus Detection With Capture Sequencing" *JAMA Network Open*
29. Andersen, Holm, Tranberg (2022) "Targeted Next Generation Sequencing for Human Papillomavirus Genotyping in Cervical Liquid-Based Cytology Samples" *Cancers* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12674535&blobtype=pdf | # Genetic characterization of equine arteritis virus associated with outbreaks in the UK, 2019
Sushant Bhat, Siva Karunakaran, Jean-Pierre Frossard, Bhudipa Choudhury, Falko Steinbach
## Abstract
InTRoDuCTIonEquine arteritis virus (EAV) is a major equine pathogen responsible for a contagious respiratory and reproductive disease known as equine viral arteritis (EVA) [1]. EAV (formally classified as species Alphaarterivirus equid) belongs to the genus Alphaarterivirus, family Arteriviridae and order Nidovirales and is an enveloped, linear, single-stranded, positive-sense RNA virus [2]. The EAV genome comprises at least ten ORFs that encode different structural and non-structural proteins (nsp) [3]. EVA can have a significant economic impact on the equine industry due to abortions in pregnant mares and through the establishment of a carrier/persistent state in stallions [4,5]. Despite neutralizing antibodies (nAbs) in their blood, carrier stallions continuously shed virus in their semen over a variable period [6], resulting in strain diversity, with corresponding differences in their genotype and phenotype [7,8]. EVA usually manifests as a subclinical disease; however, clinical signs, including severe respiratory illness, fever and reproductive complications, may be seen in cases, which vary depending on the genotype of the virus, dose of the virus and route of infection [9]. EAV infection evokes virus-specific antibodies; thus, pre-breeding screening of stallions for the presence of EAV antibodies is recommended by the UK Horserace Betting Levy Board [10].
EVA was first identified in the mid-20th century and has since been recognized as a disease of global concern, with sporadic outbreaks occurring globally [11] except for Iceland, Japan and New Zealand, which have declared themselves free from EAV. During the last decade, EAV has continued to be reported, for example, in Algeria [12], Germany [13], Serbia [14] and South America [15], with some reports of seropositive horses in Costa Rica, France, Serbia and Spain [16][17][18][19]. The UK experienced its first incursion of EAV in 1993 from a non-thoroughbred stud from Poland [20]. The disease was made notifiable in the UK in 1995, and the detection of EAV in a horse is classed as an outbreak [21]. Subsequently, there were reports of EAV-seropositive equines [22], and EAV infections were identified in 2004 and 2010 in stallions imported from the Netherlands [23,24]. Prior to the 2019 outbreak, the last recorded case in the UK was a genotypically divergent EAV identified in an imported Spanish stallion in 2012 [25].
In early 2019, following routine pre-breeding serological testing, four apparently healthy stallions from the southern counties of Devon and Dorset tested positive for EAV [26]. Epidemiological investigations indicated no spread to the wider UK equine population [27], although another horse in the northern county of Shropshire also tested positive in July 2019 [28]. In this study, virological characterization of all 2019 outbreak samples was undertaken.
## METHoDS
## Clinical samples
Semen and blood samples from apparently healthy stallions in Devon, Dorset and Shropshire were submitted to the Animal and Plant Health Agency, UK. Semen samples were centrifuged at 117 g at room temperature for 5 min to collect seminal plasma. Blood samples were left overnight for the separation of serum. Sera were harvested and stored at -80 °C until further use.
## nucleic acid extraction
RNA was extracted from the seminal plasma using TRIzol reagent (Thermo Fisher Scientific) and QIAamp Viral RNA Mini kit (Qiagen) using protocols provided by the manufacturers. The samples were first treated with TRIzol, following which the aqueous phase was removed and then processed using the QIAamp Viral RNA Mini kit.
## oRF5 RT-PCR and oRF7 qRT-PCR
Reverse transcription PCR (RT-PCR) targeting the ORF5 gene and quantitative reverse transcription PCR (qRT-PCR) targeting the ORF7 gene of equine arteritis virus (EAV) were carried out using gene-specific primers, using previously described methods [29,30]. ß actin was used as an internal control for RNA extraction [31].
## oRF5 sequencing
The ORF5 sequencing was carried out by the Sanger dideoxy chain termination method as described previously [25]. To determine the homology between the ORF5 sequences, the pairwise distance between ORF5 sequences was calculated using the maximum composite likelihood in MEGA X [32]. Percentage nucleotide identity (PNI) was calculated using the formula PNI (%)=(1-pairwise distance)×100. The data were plotted, and a heatmap matrix was generated using the webserver 'matrix2png' [33].
## next-generation sequencing
Total RNA was extracted from semen samples, as described above. The RNA was treated with TURBO ™ DNAse (Invitrogen), and the library was prepared using Nextera XT DNA Library Prep kit (Illumina) and sequenced using the MiSeq System (Illumina).
EAV reference sequence (RefSeq) Bucyrus (GenBank accession no. NC_002532.2) was used as a reference genome for mapping paired-end Illumina reads using Geneious mapper in Geneious Prime 2023.2 (https://www. geneious. com/). The consensus full-length genomes were identified by setting a threshold of 70%.
## Phylogenetic analysis
Phylogenetic analysis was carried out using ORF5 and whole-genome nucleotide sequences. For ORF5 phylogenetic analysis, partial ORF5 sequences belonging to different phylogroups [25], previously reported from other EAV outbreaks around the world (and listed in Fig. 1), were aligned using clustalw in MEGA X [32]. For whole-genome phylogenetic analysis, the genome sequences of 2019 EAV strains were aligned with the full genome sequences available in GenBank using clustalw. The Bayesian estimation of phylogeny was carried out using MrBayes [34] plugin in the Geneious Prime version 2025.0.3. The GTR substitution model with gamma rate variation was selected. Markov Chain Monte Carlo (MCMC) run was performed for 1,100,000 generations, sampling every 200 generations. The first 100,000 trees were discarded as burn-in, and the remaining trees were used to generate a 50% consensus tree. The analysis was performed using a relaxed molecular model by selecting the unconstrained branch length. The consensus tree was exported for downstream visualization and annotated in iTOL V7 [35].
The final Bayesian tree was visualized, with branch support indicated by posterior probabilities [36,37].
## virus isolation
Virus isolation in Rabbit Kidney 13 (RK13) cells was attempted on qRT-PCR-positive semen samples following a WOAHrecommended protocol described in the Terrestrial Manual 2018 [1]. The presence of the virus was confirmed by virus titration and immunoperoxidase staining of the infected cells. Briefly, RK13 cells were infected with 10-fold serial dilutions of cell supernatant showing positive cytopathic effects (CPE) and incubated for 1 h at 37 °C before being washed twice with Eagle's Minimal Essential Medium (EMEM) and incubated at 37 °C (5% CO 2 ) for 24 h. Cells were fixed in a PBS solution containing 20% acetone for 10 min. The cells were washed twice with wash buffer (PBS; 0.5% Tween) followed by the addition of EAV-specific antiserum (1:100) diluted in PBS; 1% Tween; 2.1% sodium chloride; serum diluent and incubated at 37 °C for 1 h. Cells were washed thrice with wash buffer, followed by the addition of Rabbit Anti-Horse IgG (whole molecule)-HRP (Sigma-Aldrich) (1:200 dilution in serum diluent) and incubated at 37 °C for 1 h. The EAV-infected cells were detected using ethyl carbazole (0.089% glacial acetic acid, 0.289% sodium acetate).
## Titration of the bucyrus strain
The prototype EAV Bucyrus strain was 10-fold serially diluted in EMEM (containing 1 mM sodium pyruvate, 20 mM HEPES and 1X Penicillin Streptomycin), and 100 µl of each virus dilution was added to the wells of the microplate containing RK13 cells; each virus dilution was added to a separate column containing eight wells. Two columns were kept as non-infected cell controls.
The plate was incubated at 37 °C for 72 h. To calculate the virus dose for the serum neutralization test (SNT), wells exhibiting CPE over 25% in any field were identified under a microscope and marked as positive. The virus titre (TCID 50 ) was determined by the Reed and Muench method [38].
## neutralization test
SNTs were carried out to determine the presence of nAbs in serum. For the SNT, the serum samples, including known EAV-positive antiserum and EAV-negative antiserum, were heat treated at 56 °C for 30 min to inactivate the complement. Twofold serial dilutions of the antisera were made (from 1/32 to 1/65,536) in EMEM (containing 1 mM sodium pyruvate, 20 mM HEPES and 1X Penicillin Streptomycin). Twenty-five µl per well of each serum sample was added to two rows of a microtitre plate, and each sample was tested in duplicate. Fifty µl of RK13 cell suspension containing 3×10 5 cells in EMEM (with 20% FBS, HEPES and 1 X Penicillin Streptomycin) was added to each well containing antiserum. Twenty-five µl of EAV Bucyrus strain (200TCID 50 ) was added to each well, and plates were incubated at 37 °C for 72 h with 5% CO 2 . The wells showing CPE were identified by observing the plates under an inverted microscope. Positive wells were marked when CPE was present in more than 25% of the field. The reciprocal of the highest dilution of antiserum, which could prevent virus-induced CPE, was taken as a serum nAb titre.
## Glycosylation prediction and analysis of neutralization sites
The potential N-linked glycosylation sites were identified using the NetNGlyc web server [39] to compare glycosylation sites in glycoprotein (GP) 3 and GP5. The Asn in the Asn-Xaa-Ser/Thr sequons (Asn -Asparagine, Xaa -Any amino acid, Ser -Serine, Thr -Threonine) predicted to be N-glycosylated were identified using a 0.5 cut-off. The neutralization sites were analysed for conserved amino acids and represented using Weblogo [40].
## Selection analysis
Selection pressure analysis was done on GP3 and GP5 proteins using single-likelihood ancestor counting (SLAC) [41], fixed effects likelihood (FEL) [41], mixed effects model of evolution (MEME) [42] and fast unconstrained Bayesian approximation (FUBAR) [43] methods in the datamonkey web application (https://www. datamonkey. org). For SLAC analysis, non-synonymous (dN) and synonymous (dS) substitution rates on a per-site basis were identified using a combination of maximum likelihood (ML) and counting approaches. Sites with a dN/dS >1 were considered to undergo positive selection, using a posterior probability value threshold of 0.1. For FEL analysis, codons where the relative non-synonymous substitution rate for tested branches (β) exceeded the relative non-synonymous substitution rate (α), showing β>α at a P value <0.1, were considered to undergo positive selection.
For MEME analysis, sites where the non-synonymous rate under positive selection (β+) exceeded α with a P value <0.1 were considered to be under positive selection. For FUBAR analysis, codons where the mean posterior non-synonymous substitution rate (β) exceeds the mean posterior synonymous substitution rate (α) were identified. A high Bayes factor was used as evidence for positive selection at a given site. The sites which showed evidence of positive selection using all four methods were considered to undergo positive or diversifying selection.
## Analysis of the CXCL16 gene variants in stallions
Total DNA was extracted from semen samples collected from each stallion using QIAamp DNA Mini kit (Qiagen) as per the manufacturer's instructions. PCR reactions were performed using PfuII Ultra Hotstart Master Mix (Agilent). Each reaction mix consisted of 0.4 µM of each primer (CXCL16-F and CXCL16-R11) [44], 1×master mix and 2 µl of DNA in a total volume of 50 µl. The 280 bp CXCL16 amplicons were gel extracted and sequenced using the Sanger dideoxy chain termination method. The sequenced amplicons were aligned to the reference equine CXCL16 sequence (GenBank ref: XM_001504756.6) and genotyped based on the presence of specific amino acids at positions 40, 49, 50 and 52. The presence of homozygous alleles (Y, D, F and E) at specified positions was marked as resistant genotype 'EqCXCL16R' , while the presence of heterozygous alleles (Y/F, D/H, F/I, E/K) associated with the long-term carrier status was designated 'EqCXCL16S' . All analyses were performed using Snapgene.
## RESuLTS
## Disease identification: qRT-PCR and virus isolation
The infections were detected as a consequence of pre-breeding testing in clinically healthy non-thoroughbred stallions. Semen and blood samples were tested for the presence of viral RNA by qRT-PCR and antibodies by SNT, respectively (Table 1). Paired semen samples from the infected horses tested positive using the qRT-PCR targeting ORF7. Samples from Dorset Horse 1 were weakly positive, with Ct values ranging from 35.9 to 38.3; the remaining six samples (from Dorset Horses 2 and 3, Devon Horse 4 and Shropshire Horse 5) were strongly positive, showing Ct values in the range of 21.2-25.6 (Table 1). All samples were then tested by virus isolation, but the virus could only be isolated from two semen samples belonging to Dorset Horses 2 and 3.
## Phylogenetic analysis of the viruses
Initial phylogenetic analysis was carried out using ORF5 nucleotide sequences derived from the EAV outbreak sequences (1 per animal) and 62 reference genomes belonging to phylogroups A-G [25]. Bayesian phylogenetic analysis carried out by the MCMC method showed that four out of five 2019 outbreak strains clustered together, showing a posterior probability of 1, while the GB_024365_2019 strain sampled from Shropshire Horse 5 was genetically distinct (Figs 1 and S1, available in the online Supplementary Material). All 2019 outbreak strains belong to phylogroup D [25] forming a monophyletic group of sequences identified in the UK between 2004 and 2011 and clustered together with Polish and Hungarian isolates. It should be noted that this analysis resulted in a new genotype H that arose from the additional EAV isolates since 2015 from Serbia [25]. Posterior probability confirmed the reliability of trees, which was also corroborated by ML analysis (Fig. S2).
Reference sequences clustered into different phylogroups with posterior probability ranging from 0.96 to 1 and bootstrap values ranging from 60 to 100% (Fig. S3).
Bayesian phylogenetic analysis was also undertaken with whole-genome sequences of strains identified in 2019 and 36 complete genomes which were representative of all the EAV sequences available on GenBank. This complete genome-based phylogenetic analysis confirmed that the Shropshire strain was phylogenetically distinct (outgroup) compared to other 2019 UK strains (Figs 2 andS4). Here, the 2019 UK outbreak strains clustered with the EAVs identified in the USA in the 2006-2007 outbreak and with Polish EAV strains identified in Hucul horses [45]. Reference genomes from donkeys appeared to be the most distant in all analyses.
Bayesian estimation combining whole-genome phylogeny and number of nucleotide substitutions per site indicated that within the Dorset/Devon cluster, Horse 2 was first infected and transmitted the virus to Horses 1, 3 and 4 (Figs 1b and2b). The precise order of transmission cannot be resolved through this analysis, but it seems likely that transmission occurred on at least two separate occasions, with Dorset Horse 1 being an intermediate.
## Comparison of genome sequences
Whole-genome sequences could be recovered from four of the five horses. The comparison showed that Dorset strains (GB_009362_2019 and GB_009363_2019) had a genome length of 12,704, while the Shropshire strain (GB_024365_2019) was 12,700nt in length. The Devon strain (GB_015276_2019) showed an incomplete EAV leader sequence with a 12,697 nt genome length. The sequences were compared with the GB_Glos_2012 genome, which was 12,702 nt in length.
Among the 2019 strains, the Shropshire strain showed less similarity to the Devon and Dorset strains, with the most variability found in ORF3 (GP3) and ORF1ab, followed by ORF5 (GP5), ORF4 (GP4), ORF2b (GP2b), ORF7 (N), ORF6 (M) and ORF2a (E) (Fig. 3). Within ORF1ab, maximum variability was seen in nsp11, nsp2 and nsp12, followed by nsp4, nsp5, nsp6, nsp3, nsp7, nsp9, nsp10, nsp8 and nsp1 (Fig. S5). The per cent similarity between the Dorset strain (GB_015276_2019) and the two Devon strains (GB_009362_2019 and GB_009363_2019) was 99.72 and 99.89%, respectively, with maximum divergence in GP3 followed by GP5. The GB_009362_2019 and GB_010228_2019 genomes (identified in paired semen samples from Horse 2) differed from one another due to nine synonymous mutations (data not shown), and importantly a non-synonymous mutation (P194S) was observed in the consensus sequence of nsp2. The whole-genome sequences of 2019 EAV strains did not show any indels in the ORF1ab polypeptide, unlike the one amino acid deletion observed in nsp2 of GB_Glos_2012 [25]. Phylogenetic analysis confirms that the EAV from Horse 2 has an ancestral linkage to the viruses in Horses 3 and 4.
The GP4 mature peptide of the Shropshire strain showed a three amino acid truncation due to a premature stop-codon incorporated due to C/T transition in the codon at position 149, while the strains from Devon and Dorset had full-length GP4. The C/T transition at position 149 was observed in more than 70% of sequencing reads.
A percentage homology comparison between different proteins of the 2019 EAV strains with reference to the GB_Glos_2012 genome revealed nsp6 protein was most conserved with 100% homology, followed by the nucleoprotein (N), while the GP3 protein showed the highest variability (Fig. S6).
## Glycosylation and selection analysis in GP5 and GP3
All available EAV ORF5 (encoding GP5) and ORF3 (encoding GP3) sequences were retrieved from GenBank and aligned using muscle. Sequences having ambiguous bases were removed. A total of 359 GP5 and 300 GP3 sequences were subsequently analysed to identify potential N-linked glycosylation sites and to assess selection pressure. For GP5, the glycosylation site identified at position 56 was conserved in all sequences (Fig. 4). A second site at position 81 was relatively conserved, present in 312 sequences.
Other potential sites at positions 73, 82, 233 and 240 were non-conserved and observed in only a few sequences. Notably, the GP5 of 2019 EAVs had only two glycosylation sites, namely at amino acid positions 56 and 81.
Sequence alignment of GP5 ORFs showed that 2019 EAV outbreak strains differed among themselves (Fig. S7A) and previous UK EAV outbreak viruses (Fig. S7B) at neutralization sites C and D. The sites C and D harbour the major neutralizing epitopes of EAV by forming a conformational dependent neutralization domain as proposed in some studies [46,47]. Selection pressure analysis of full-length GP5 using SLAC, FEL, MEME and FUBAR (Table S1) showed that site 61 in neutralization site B, site 82 in neutralization site C and sites 101 and 104 in neutralization site D showed positive or diversifying selection. This suggests that the neutralization (but not glycosylation) sites in GP5 are under evolutionary pressure from the host immune response, leading to mutations that help the virus persist or spread.
For GP3, glycosylation was identified at positions 28,29,39,49,96,106,115,118,119 and 120 (Fig. 5). Among these sites, glycosylation at positions 29 (296/300), 49 (300/300), 96 (299/300) and 106 (300/300) was conserved or relatively conserved across the dataset. Positive selection was identified at least at sites 3-6, 9-10, 16, 18-25, 27, 120 and 123 (Table S2). Compared to GP5 sequences, GP3 sequences exhibited a higher proportion of glycosylation sites (5.95% vs. 2.35%) and a significantly greater number of codons under positive selection (11.9% vs. 3.52%) which implies that EAV GP3 is subject to a stronger evolutionary pressure and is evolving more rapidly than GP5.
## Antibody response
All horses showed the presence of nAbs when tested in SNT with titres ranging from 1,536 to 3,072 (Table 1).
## EqCXCL16 genotyping of EAv-infected stallions
Among the horses sampled in Dorset, Horse 1 was found to be homozygous for the EqCXCL16R allele. This allele is associated with resistance of CD3+T lymphocytes to EAV infection, thereby reducing the likelihood of the horse developing a carrier status.
In contrast, all other horses were heterozygous for the EqCXCL16 allele (EqCXCL16S/EqCXCL16R), which has been linked to a susceptible phenotype [44]. These horses were therefore at an increased risk of establishing a carrier stage following EAV infection (Table 2).
## DISCuSSIon
EVA is a notifiable disease in equines, potentially causing significant economic loss [1,9]. In 2019, EAV outbreaks in the UK, specifically in Devon, Dorset and Shropshire, were identified through pre-breeding tests, illustrating the importance of proactive surveillance systems in detecting and managing infectious diseases.
Analysing EAV's genetic diversity is crucial for understanding the molecular epidemiology of outbreaks and limiting disease transmission within equine populations. ORF5 of EAV is commonly used for phylogenetic analysis [29,48,49], although GP3 [7,50] and ORF1b have also been used [51][52][53]. Phylogenetic analysis of ORF5 showed that the 2019 outbreak strains from Devon and Dorset were from the same epidemiological site and belonged to phylogroup D that also contains EAV strains identified in the UK between 2004 and 2011 [25]. Continued efforts to detect EAV resulted in another case identified in Shropshire in July 2019. Although the strain identified from Shropshire also belongs to phylogroup D, it formed a sub-clade with the Polish 'PL5' strain [29], which is phylogenetically distinct from the Devon and Dorset strains, as supported by very high posterior probability and bootstrap values. There was no history of mating or transport of the Shropshire horse to events within or outside the UK, demonstrating a missing link between the source of infection and identification of carrier status. This emphasizes the need for vigilance to better understand the dynamics of EAV transmission.
As more variability was noted in GP3 and ORF1ab (particularly in nsp5 and nsp9), compared to GP5 (Figs 3 and S5), GP5-based phylogeny may overestimate the reliability of tracing the order of infection. A comparison of evolutionary divergence among ORF3, ORF4 and ORF5 sequences of all available EAV strains also showed GP3 with more variability than GP5 and GP4 (Fig. S8). Whole-genome-based phylogenetic analysis is so far less common [54] for EAV. Moreover, commercial diagnostic testing does not mandate sequencing, and both cost considerations and low viral loads in some semen samples often lead to a preference for partial PCR-based sequencing. As a result, the number of complete EAV genomes available in the GenBank database remains limited.
Both the ORF5-based analysis and the whole-genome phylogenetic analysis showed fewer nucleotide substitutions in sequences from Dorset Horse 2 compared to Devon Horse 4 and Dorset Horse 3. Hence, in both analyses, Dorset Horse 2 was more likely to be the index horse, which was infected from an unconfirmed source, followed by the spread of infection to Devon Horse 4 and Dorset Horse 3. The role of horse 1 from Devon in the infection chain remains more uncertain due to the absence of whole genomic data, and there is no evidence supporting this horse to be the index case as assumed by the clinical epidemiological investigation. The absence of whole-genome sequences from recent EAV outbreaks makes it challenging to map the exact source of the outbreak.
The amino acid analysis of different proteins showed a unique truncation in GP4 of the Shropshire strain, which could have implications for virus assembly and host-cell attachment [55]. While deletions at the 3′ end of ORF3 are common in arteriviruses [56] and affect the 5′ end of ORF4 (GP4) due to overlapping reading frames, truncations at the carboxy-terminal of GP4 leading to a shorter GP had not been identified before. Whether the GP4 truncation has any phenotypic implication on EAV needs to be analysed further. GP3 encoded by ORF3 was found to be the most variable, as seen in other arteriviruses, like Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) [57], and has been reported to encounter strong selection pressure during persistent EAV infection in a thoroughbred carrier stallion [58]. This was also corroborated by our GP3 selection analysis, which showed an increased number of sites undergoing positive selection compared to the GP5. Glycosylation of viral proteins can have a pronounced effect on antigenicity, affect virus virulence and antibody-based neutralization [59,60]. While no change in glycosylation sites was identified in GP5 of 2019 outbreak strains compared to the previous EAV strains identified in the UK, the 2019 EAV outbreak strains differed amongst themselves at the previously proposed antibody neutralization sites, which had been identified as mutational 'hot spots' responsible for the emergence of phenotypic variants with altered neutralization potential [58,61]. EAV infection triggers a humoral antibody response [62]. Currently, there is only one serotype associated with GP5 of EAV, and selection analysis indicates that four sites (61, 82, 101 and 104) in neutralization sites B, C and D are undergoing positive selection. Changes in the neutralization sites reflect adaptive changes that allow the virus to evade antibody recognition. The positively selected sites in neutralization sites B, C and D may therefore represent an immune-driven hotspot of variation that warrants continued monitoring as more EAV outbreaks emerge.
EAV can be spread via fomites, close contact or sexual transmission [5]. Widespread EVA epizootics are uncommon, and most of the EAV infections occur without clinical manifestation and are confirmed by genome detection and seroconversion [5]. While EAV is present in various EU Member States, imported horses could act as a 'trojan horse' due to the absence of statutory testing in many countries. If stallions are not used for breeding, asymptomatic horses may go unnoticed and may spread the disease via equine events of various kinds. The initial sources of infection leading to the EAV outbreaks in the UK in 2019 have not been determined, but the Shropshire case related best to an infection spread from outside the UK. Phylogenetic analysis showed a within-outbreak spread from Horse 2 to the other horses in the Dorset/Devon cluster (illustrated in Fig. 6). Since there is a usual practice of stabling horses together at show events, close contact can occur, thereby enabling the origin of an outbreak and spread of infection. Analyses of ORF5 sequences show that the 2019 Dorset/Devon cases could be linked to previous (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011) EAV outbreaks in the UK. Thus, the presence of carrier stallions who were subclinically infected between 2004 and 2011 and transmitted the virus to other susceptible horses in 2019 cannot be negated. Equally, further circulation and re-introduction of the virus from other countries is possible. Genotyping of the CXCL16 gene in horses affected by the 2019 outbreak revealed that the majority carried the 'heterozygous susceptible' genotype, which is significantly associated with an increased risk of becoming long-term carriers of EAV transmission [6,44]. Detection of EAV during routine pre-breeding testing is formally classified as an outbreak, while not necessarily leading to active disease, and can provide important epidemiological insight into the presence of infected carrier stallions spreading virus within the local horse population. Such carriers play a critical role in virus maintenance and dissemination not only during breeding activities. This highlights ongoing virus circulation and underlines the importance of continuous surveillance to prevent potential transmission events that could result in future outbreaks.
In conclusion, the 2019 EAV outbreak reinforces the importance of surveillance, rapid response and effective implementation of control measures to mitigate the spread of EAV. It also underpins the importance of a multifaceted approach involving veterinary surgeons, diagnostic laboratories and horse premises in effectively controlling and managing such outbreaks. Screening breeding stallions alone is insufficient to prevent the transmission of EAV. To effectively control the spread of the virus, particularly in subclinical forms, it is essential to extend testing to non-breeding stallions. A robust DIVA test would further assist in differentiating infected horses from vaccinated ones.
## References
1. Oie (2018) "OIE terrestrial manual, Chapter 3.5.10. -Equine viral arteritis (infection with equine arteritis virus)"
2. Ictv, Family (2011) "Arteriviridae Chapter Version: ICTV Ninth Report; 2009 Taxonomy Release"
3. Snijder, Kikkert, Fang (2013) "Arterivirus molecular biology and pathogenesis" *J Gen Virol*
4. Timoney, Mccollum, Roberts et al. (1986) "Demonstration of the carrier state in naturally acquired equine arteritis virus infection in the stallion" *Res Vet Sci*
5. Timoney, Mccollum (1993) "Equine viral arteritis" *Vet Clin North Am Equine Pract*
6. Carossino, Dini, Kalbfleisch et al. (2019) "Equine arteritis virus long-term persistence is orchestrated by CD8+ T lymphocyte transcription factors, inhibitory receptors, and the CXCL16/CXCR6 axis" *PLoS Pathog*
7. Balasuriya, Hedges, Smalley et al. (2004) "Genetic characterization of equine arteritis virus during persistent infection of stallions" *J Gen Virol*
8. Miszczak, Legrand, Balasuriya et al. (2012) "Emergence of novel equine arteritis virus (EAV) variants during persistent infection in the stallion: origin of the 2007 French EAV outbreak was linked to an EAV strain present in the semen of a persistently infected carrier stallion" *Virology*
9. Balasuriya, Carossino, Timoney (2018) "Equine viral arteritis: a respiratory and reproductive disease of significant economic importance to the equine industry" *Equ Vet Ed*
10. Hblb (2023) "International codes of practice 2023 -equine viral arteritis"
11. Bryans, Crowe, Doll et al. (1957) "Isolation of a filterable agent causing arteritis of horses and abortion by mares; its differentiation from the equine abortion (influenza) virus" *Cornell Vet*
12. Laabassi, Amelot, Laugier et al. (2014) "Prevalence of equine viral arteritis in Algeria" *Rev Sci Tech*
13. Otzdorff, Beckmann, Goehring (2021) "Equine Arteritis Virus (EAV) outbreak in a show stallion population" *Viruses*
14. Gaudaire, Lazić, Lupulović et al. (2019) "Complete genome sequence of an equine arteritis virus strain isolated from a Lipizzaner stallion in 2015 in Serbia" *Microbiol Resour Announc*
15. Rivas, Neira, Mena et al. (2017) "Identification of a divergent genotype of equine arteritis virus from South American donkeys" *Transbound Emerg Dis*
16. Cruz, Fores, Mughini-Gras et al. (2016) "Seroprevalence and factors associated with seropositivity to equine arteritis virus in Spanish Purebred horses in Spain" *Equine Vet J*
17. Lazić, Lupulović, Gaudaire et al. (2017) "Serological evidence of equine arteritis virus infection and phylogenetic analysis of viral isolates in semen of stallions from Serbia" *BMC Vet Res*
18. Amat, Vergne, Tapprest et al. (2016) "Estimating the incidence of equine viral arteritis and the sensitivity of its surveillance in the French breeding stock" *Vet Microbiol*
19. Jiménez, Romero-Zuñiga, Dolz (2014) "Serosurveillance of infectious agents in equines of the Central Valley of Costa Rica" *Open Vet J*
20. Wood, Chirnside, Mumford et al. (1995) "First recorded outbreak of equine viral arteritis in the United Kingdom" *Vet Rec*
21. (1995) "The equine viral arteritis order"
22. Newton, Wood, Castillo-Olivares et al. (1999) "Serological surveillance of equine viral arteritis in the United Kingdom since the outbreak in 1993" *Vet Rec*
23. Manser, Westcott (2005) "Equine viral arteritis in a stallion" *Vet Rec*
24. Crabtree, Newton (2020) "Equine viral arteritis (EVA): a potential trapdoor for the practicing veterinary surgeon in the United Kingdom" *Equ Vet Ed*
25. Steinbach, Westcott, Mcgowan et al. (2015) "Re-emergence of a genetic outlier strain of equine arteritis virus: impact on phylogeny" *Virus Res*
26. Apha (2019) "Equine viral arteritis: epidemiology reports"
27. Lattimer, Roberts, Barnard et al. (2020) "Investigating an outbreak of equine viral arteritis at two connected premises" *Vet Rec*
28. Apha (2019) "Epidemiology report detailing the investigation of EVA virus disclosed on a premises in Shropshire"
29. Stadejek, Rklund, Ana et al. (1999) "Genetic diversity of equine arteritis virus" *J Gen Virol*
30. Balasuriya, Leutenegger, Topol et al. (2002) "Detection of equine arteritis virus by realtime TaqMan reverse transcription-PCR assay" *J Virol Methods*
31. Toussaint, Sailleau, Breard et al. (2007) "Bluetongue virus detection by two real-time RT-qPCRs targeting two different genomic segments" *J Virol Methods*
32. Kumar, Stecher, Li et al. (2018) "MEGA X: molecular evolutionary genetics analysis across computing platforms" *Mol Biol Evol*
33. Pavlidis, Noble (2003) "Matrix2png: a utility for visualizing matrix data" *Bioinformatics*
34. Ronquist, Teslenko, Van Der Mark et al. (2012) "MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space" *Syst Biol*
35. Letunic, Bork (2021) "Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation" *Nucleic Acids Res*
36. Huelsenbeck, Ronquist, Nielsen et al. (2001) "Bayesian inference of phylogeny and its impact on evolutionary biology" *Science*
37. Huelsenbeck, Ronquist (2001) "MRBAYES: Bayesian inference of phylogenetic trees" *Bioinformatics*
38. Reed, Muench (1938) "A simple method of estimating fifty per cent endpoints12" *Am J Hyg*
39. (2023) "N-linked glycosylation sites in human proteins"
40. Crooks, Hon, Chandonia et al. (2004) "WebLogo: a sequence logo generator" *Genome Res*
41. Pond, Frost (2005) "Not so different after all: a comparison of methods for detecting amino acid sites under selection" *Mol Biol Evol*
42. Murrell, Wertheim, Moola et al. (2012) "Detecting individual sites subject to episodic diversifying selection" *PLoS Genet*
43. Murrell, Moola, Mabona et al. (2013) "FUBAR: a fast, unconstrained bayesian approximation for inferring selection" *Mol Biol Evol*
44. Sarkar, Bailey, Go et al. (2016) "Allelic variation in CXCL16 determines CD3+ T lymphocyte susceptibility to equine arteritis virus infection and establishment of long-term carrier state in the stallion" *PLoS Genet*
45. Rola, Larska, Rola et al. (2011) "Epizotiology and phylogeny of equine arteritis virus in hucul horses" *Vet Microbiol*
46. Balasuriya, Dobbe, Heidner et al. (2004) "Characterization of the neutralization determinants of equine arteritis virus using recombinant chimeric viruses and site-specific mutagenesis of an infectious cDNA clone" *Virology*
47. Balasuriya, Patton, Rossitto et al. (1997) "Neutralization determinants of laboratory strains and field isolates of equine arteritis virus: identification of four neutralization sites in the amino-terminal ectodomain of the GL envelope glycoprotein" *Virology*
48. Balasuriya, Timoney, Mccollum et al. (1995) "Phylogenetic analysis of open reading frame 5 of field isolates of equine arteritis virus and identification of conserved and nonconserved regions in the GL envelope glycoprotein" *Virology*
49. Metz, Ocampos, Serena et al. (2011) "Extended phylogeny of the equine arteritis virus sequences including South American sequences" *Intervirology*
50. Hedges, Balasuriya, Topol et al. (2001) "Genetic variation of ORFs 3 and 4 of equine arteritis virus" *Adv Exp Med Biol*
51. Lauck, Alkhovsky, Bào et al. (2015) "Historical outbreaks of simian hemorrhagic fever in captive macaques were caused by distinct arteriviruses" *J Virol*
52. Oğuzoğlu, Karapınar, Bilge-Dagalp (2013) "Partial sequence of the orf1b gene fragment of equine arteritis viruses detected in turkey and phylogenic analysis" *Revue de Medecine Veterinaire*
53. Mittelholzer, Stadejek, Johansson et al. (2006) "Extended phylogeny of equine arteritis virus: division into new subgroups" *J Vet Med B Infect Dis Vet Public Health*
54. Vanmechelen, Vergote, Laenen et al. (2018) "Expanding the arterivirus host spectrum: Olivier's shrew virus 1, a novel arterivirus discovered in African giant shrews" *Sci Rep*
55. Snijder, Meulenberg (1998) "The molecular biology of arteriviruses" *J Gen Virol*
56. Dortmans, Buter, Dijkman et al. (2016) "Molecular characterization of type 1 porcine reproductive and respiratory syndrome viruses (PRRSV) isolated in the Netherlands from 2014 to" *PLoS One*
57. Kvisgaard, Hjulsager, Kristensen et al. (2013) "Genetic and antigenic characterization of complete genomes of type 1 Porcine Reproductive and Respiratory Syndrome viruses (PRRSV) isolated in Denmark over a period of 10 years" *Virus Res*
58. Hedges, Balasuriya, Timoney et al. (1999) "Genetic divergence with emergence of novel phenotypic variants of equine arteritis virus during persistent infection of stallions" *J Virol*
59. Vigerust, Shepherd (2007) "Virus glycosylation: role in virulence and immune interactions" *Trends Microbiol*
60. Feng, Zhang, Chen et al. (2022) "Glycosylation of viral proteins: implication in virus-host interaction and virulence" *Virulence*
61. Balasuriya, Patton, Rossitto et al. (1997) "Neutralization determinants of laboratory strains and field isolates of equine arteritis virus: identification of four neutralization sites in the amino-terminal ectodomain of the G(L) envelope glycoprotein" *Virology*
62. Wagner, Balasuriya, Maclachlan (2003) "The serologic response of horses to equine arteritis virus as determined by competitive enzyme-linked immunosorbent assays (c-ELISAs) to structural and non-structural viral proteins" *Comp Immunol Microbiol Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12622759&blobtype=pdf | # Herpes simplex virus type 2 implicated in a case of acute disseminated encephalomyelitis
Stuart Booth, Igor Starinskij, Stuart Gallacher, Peter Garmany
## Abstract
Data for cerebrospinal fluid multiplex PCR result comparison were extracted from the laboratory information management system using Dedalus Telepath version 1.9 by a data manager independent from the project. The following parameters were retrieved: patient demographic information, location at time of request, sample collection date, result authorization date, sample type, test interpretation results and cycle threshold values. The data were exported in XLSX format for processing and analysis. Samples obtained from the specific patient described in the report, as well as samples that did not complete testing, were excluded. The final dataset comprised 3,270 samples collected between 27 February 2019 and 30 October 2021. All information except for specimen type, collection date and testing results was removed to preserve patient confidentiality.
## INTRODUCTION
Acute disseminated encephalomyelitis (ADEM) is a well-described neurological disorder that can complicate acute infection, vaccination and organ transplantation and often leads to acute urinary retention [1][2][3][4][5][6][7]. Preceding infection is not part of a formal definition of ADEM, which is considered an immune-mediated condition, although it is almost certainly one of the major triggers [8].
There are case reports of ADEM with evidence of recent or current herpes simplex virus (HSV) type 1 (HSV-1) infection as well as a survey reporting HSV genome detection in cerebrospinal fluid (CSF) in one out of three ADEM cases [9][10][11][12]. In contrast, there is no literature, to the best of the authors' knowledge, describing an association between herpes simplex virus type 2 (HSV-2) and presentations of ADEM.
## CASE PRESENTATION
A 20-year-old man presented to the emergency department of a large inner city teaching hospital complaining of weakness and sensory disturbance affecting his lower limbs which had gradually progressed over the preceding 7 days.
The patient described initial altered sensation in his left foot with associated limp which, over a period of 4 days, developed to painful paraesthesia of his entire left leg. On the morning of attending hospital, he reported a complete loss of sensation to both legs and an inability to stand or walk. At this time, there was associated urinary incontinence. He required being carried into the emergency department due to bilateral limb weakness.
He was fit and well prior to this, except for a single episode of pharyngitis and tonsillar exudate 14 days preceding initial neurological symptoms, which fully resolved with 7 days of oral phenoxymethylpenicillin (500 mg four times daily).
There was no history of recent travel or unwell contacts. He had been immunized fully in line with the standard UK vaccination schedule including coronavirus disease 2019 vaccination. There was no family history of inherited conditions including no previous familial neurological condition.
He had one female sexual partner of around 2 months, was heterosexual and reported no unprotected sex. There was no evidence of genital blistering or ulceration, and he did not report any genitourinary symptoms.
On examination, he was noted to have Medical Research Council (MRC) grade 0/5 power in both lower limbs, with complete sensory loss (pin prick, light touch, vibration and proprioception) from the T10 through S5 dermatomes. Both legs were areflexic with flaccid muscle tone. He was found to be in urinary retention, requiring urinary catheterization and was incontinent of faeces. He was febrile with a temperature of 38.8 °C, and observations were otherwise within normal parameters.
Computed tomography head scanning was performed with intravenous contrast and was unremarkable. Blood laboratory results, including full blood count, C-reactive protein, liver function and renal function, were essentially within normal limits, while CSF results can be found in Table 1, although viral PCR results were not available at this point.
Empirical treatment was started (once daily unless stated otherwise):
• intravenous ceftriaxone 2 g twice daily,
• intravenous aciclovir 10 mg kg -1 three times daily, • oral dexamethasone 10 mg twice daily, • oral rifampicin 720 mg, • oral isoniazid 300 mg, • oral pyrazinamide 1,800 mg, • oral ethambutol 1,200 mg, • oral pyridoxine 10 mg.
At this point, the working diagnosis was a central nervous system (CNS) infection, such as CNS tuberculosis, but other common bacterial and viral pathogens were also covered with the empirical regimen.
At day 10 of admission, he reported new paraesthesia to both feet which progressed over the following 2 days to involve his entire lower limbs. On day 12, he was noted to have light touch and pinprick sensation to the soles of both feet. On day 13, it was noted that motor function was recovering with hip adduction power bilaterally at MRC grade 1/5.
Direct Gram staining of CSF was negative, and bacterial culture was performed with no growth of bacterial pathogens by 48-h incubation. Bacterial PCR was negative for Haemophilus influenzae, Neisseria meningitidis and Streptococcus pneumoniae. A total of 10 ml of CSF was sent for mycobacterial culture with no growth at 42-day incubation. The CSF samples underwent PCR testing using an in-house assay for HSV-1, HSV-2, varicella zoster virus, enteroviruses and parechoviruses. HSV-2 was detected in the first CSF sample at a cycle threshold (Ct) of 22.2. The patient was extensively investigated for infectious and autoimmune conditions, and the results can be found in Table 2.
Electromyography showed mild abnormalities in lower limb motor studies with preserved sensory studies but no clear motor activity in the lower limbs in keeping with a central pathology. T1, T2, diffusion weighted, short tau inversion recovery magnetic resonance imaging (MRI) of the head and spine with gadolinium enhancement was performed with findings largely confined to within the brain stem and spinal cord. There were multiple intramedullary cord lesions. The appearances were in keeping with a diagnosis of ADEM (Fig. 1). Predominant spinal involvement and enhancement of the spinal meninges are uncommon features but previously described [13,14].
A repeat lumbar puncture was performed on day 26 of admission, prior to completion of intravenous antiviral therapy. CSF testing results across the admission are summarized in Table 1. CSF microscopy with Gram staining, bacterial culture and viral PCR screen, excluding HSV-2, remained negative throughout.
MRI at this time showed interval improvement in intracranial and spinal appearances, with no new evidence of leptomeningeal enhancement and previous findings less appreciable in keeping with response to treatment.
Treatment was rationalized as culture results became available, with tuberculosis treatment continuing until negative mycobacterial culture. Intravenous aciclovir was continued for a total of 28 days before being switched to 500 mg valaciclovir once daily for a total of 6 months. Dexamethasone was discontinued at day 10 after liaison with the clinical neurology service.
He underwent extensive physiotherapy and had a gradual progressive improvement in motor and sensory function. On day 90 of admission, the patient was transferred to the physical disability rehabilitation unit (PDRU). At this time, he had MRC grade 2/5 bilateral lower limb weakness with normal sensation to pinprick and light touch, although he remained areflexic with flaccid muscle tone in T10-S5 myotomes. He remained catheterized due to multiple failed trials without catheter and was intermittently incontinent of faeces.
The patient underwent a further 45 days of physical rehabilitation in the PDRU. By the time of discharge, power had improved to MRC grade 4/5 on his right and 3/5 on his left lower limb. He was mobilizing with a wheelchair and was independent with all transfers and self-care. Bowel function normalized; however, the patient continued to require intermittent self-urinary catheterization 2-3 times weekly due to incomplete bladder emptying.
The patient made continued improvement over the next 3 years. His leg power had progressed to MRC grade 5/5 on the right and 4+/5 on the left and he had become independently mobile with elbow crutches. He was able to drive, work in retail for around 5 h per day and attend a gym. However, neurological sequelae, in the form of lower limb spasticity and chronic urinary retention, remain.
## DISCUSSION
This is a case of ADEM with strong molecular and serological evidence of a concurrent HSV-2 CNS infection.
Acute haemorrhagic leucoencephalitis is the most severe and rapidly progressive form of ADEM [15,16]. Nonetheless, the absence of evidence of cerebral bleeding in this patient and non-hyperacute course both make this diagnosis unlikely.
CSF PCR was used as the primary aetiological diagnostic tool during the acute phase of illness. PCR does not directly measure the number of replicating virions, but the Ct values obtained using this method correlate with both disease severity and duration of illness [17,18]. To put the results into context, data from CSF PCR assays performed in the local virology laboratory over a 20-month period between 2019 and 2021 were reviewed (Table 3). In this case, the Ct of 22.2 is lower than ever obtained for any pathogen in the examined timeframe, compatible with an early phase of a severe infection.
few alternative explanations should be considered. Firstly, HSV-2 encephalitis, although uncommon in general, is in keeping with the finding of a high amount of HSV-2 DNA in the CSF [19]. On the other hand, although the neurological features of both ADEM and HSV encephalitis can be diverse, there were no clinical signs of generalized cerebral inflammation, such as seizures and reduced consciousness level. In addition, widespread MRI findings were in keeping with ADEM rather than HSV-2 encephalitis, which is often confined to the temporal and frontal lobes [4,19]. Secondly, postinfectious immune-mediated encephalitis is often described as one of the forms of ADEM but can also present differently [5,20]. It can be distinguished from ADEM by an absence of new lesions compared to the areas affected by HSV during the first episode of overt encephalitis with negative viral PCR during the second presentation [20]. Here, there was no preceding illness suggestive of HSV-2 encephalitis and the viral PCR is positive. Finally, acute HSV-2 CNS infection could have coincided with ADEM with distinct aetiology. In this case, pharyngitis antecedent to neurological presentation was indeed diagnosed and theoretically could have been the original cause of autoimmune neurological disease. However, its causative agent had never been identified; hence, it is impossible to conclusively link this presentation to the past event. *Ganglioside screen includes GM1 IgG, GM2 IgG, GD1a IgG, GD1b IgG, GQ1b IgG, GM1 IgM, GM2 IgM, GD1a IgM, GD1b IgM and GQ1b IgM. †Neuronal antibodies include anti-HU, anti-Yo, anti-RI, anti-CV2, anti-MA2TA, anti-amphiphysin, anti-SOX1, anti-recoverin and anti-titin. IgG, immunoglobulin G; IgM, immunoglobulin M; rDNA, ribosomal deoxyribonucleic acid.
Multiple sclerosis (MS), the first attack which can be highly similar to ADEM, should be thought of when evaluating the long-term prognosis of a patient with ADEM [21,22]. All three proposed differentiation criteria (atypical sign -paraplegia, absence of oligoclonal bands and grey matter involvement) present in this case were supportive of ADEM rather than MS [23]. To date, our patient had not experienced recurrent or new symptoms suggestive of MS.
One limitation to the primary diagnosis is that a biopsy with relevant histological investigations is required for a final confirmation of ADEM. However, a close temporal association, robust detection by repeated PCR assays as well as the known role of a closely related HSV-1 in ADEM pathogenesis strongly support HSV-2 as the neurological disease-causing agent here. Additionally, we demonstrated HSV-2 type-specific immunoglobulin G seroconversion (Table 2), further solidifying the case for an acute HSV-2 CNS infection, while imaging was most in keeping with ADEM rather than alternative neurological conditions.
In summary, we describe a case of ADEM affecting primarily the spinal cord with strong molecular and serological evidence of HSV-2 CNS infection. The case was treated with a variety of antimicrobials, notably aciclovir, and a 10-day course of dexamethasone. The patient had a slow but steady improvement in neurological function over a period of months, but years later, neurological sequelae remain.
The Microbiology Society is a membership charity and not-for-profit publisher.
Your submissions to our titles support the community -ensuring that we continue to provide events, grants and professional development for microbiologists at all career stages.
Find out more and submit your article at microbiologyresearch.org
## References
1. "Axial T1 post-gadolinium contrast of the brain. Focal leptomeningeal enhancement high left frontal lobe (arrow)"
2. "Axial T2 fluid-attenuated inversion recovery of the brain. A focal high signal lesion at the parafalcine cortex right frontal lobe (arrow). (c) Sagittal T2-weighted images of the craniocervical junction. A high signal lesion in the anterior medulla oblongata (arrow). (d) Axial diffusion-weighted imaging of the craniocervical junction. Restricted diffusion within the lesion (arrow). (e) Complementary low signal on the apparent diffusion coefficient map (arrow). (f) Sagittal T2-weighted images of the cervicothoracic spine"
3. Garg (2003) "Acute disseminated encephalomyelitis" *Postgrad Med J*
4. Leake, Albani, Kao et al. (2004) "Acute disseminated encephalomyelitis in childhood: epidemiologic, clinical and laboratory features" *Pediatr Infect Dis J*
5. Silvia, Licht (2005) "Pediatric central nervous system infections and inflammatory white matter disease" *Pediatr Clin North Am*
6. Stone, Hawkins (2007) "A medical overview of encephalitis" *Neuropsychol Rehabil*
7. Greenlee (2012) "Encephalitis and postinfectious encephalitis" *Continuum (Minneap Minn)*
8. Sakakibara, Hattori, Yasuda et al. (1996) "Micturitional disturbance in acute disseminated encephalomyelitis (ADEM)" *J Auton Nerv Syst*
9. Sasaki, Ohara, Hayashi et al. (2006) "Aseptic meningo-radiculo-encephalitis presenting initially with urinary retention: a variant of acute disseminated encephalomyelitis" *J Neurol*
10. Young, Weinshenker, Lucchinetti (2008) "Acute disseminated encephalomyelitis: current understanding and controversies" *Semin Neurol*
11. Bocos-Portillo, Sánchez-Menoyo, Beldarrain et al. (2018) "Acute disseminated encephalomyelitis: a rare autoimmune complication of herpes simplex encephalitis in the adult" *Clin Neurol Neurosurg*
12. Ito, Watanabe, Akabane (2000) "Acute disseminated encephalomyelitis developed after acute herpetic gingivostomatitis" *Tohoku J Exp Med*
13. Kaji, Kusuhara, Ayabe et al. (1996) "Survey of herpes simplex virus infections of the central nervous system, including acute disseminated encephalomyelitis, in the Kyushu and Okinawa regions of Japan" *Mult Scler*
14. Sarioglu, Kose, Saritas et al. (2014) "Severe acute disseminated encephalomyelitis with clinical findings of transverse myelitis after herpes simplex virus infection" *J Child Neurol*
15. Monden, Yamagata, Kuroiwa et al. (2012) "A case of ADEM with atypical MRI findings of a centrally-located long spinal cord lesion" *Brain Dev*
16. Roshanisefat, Henderson, Press (2016) "An unusual Guillain-Barré Syndrome mimic: a case of Spinal ADEM and review of the literature" *Clin Case Rep Rev*
17. Groves, Martinian, Kuo et al. (2002) "Detection of infectious agents in brain of patients with acute hemorrhagic leukoencephalitis" *J Neurovirol*
18. Martins, Teixeira, Jr et al. (2004) "Brazilian Committee for Treatment and Research in Multiple Sclerosis. Acute hemorrhagic leukoencephalitis mimicking herpes simplex encephalitis: case report" *Arq Neuropsiquiatr*
19. Ziyaeyan, Alborzi, Haghighi et al. (2011) "Diagnosis and quantitative detection of HSV DNA in samples from patients with suspected herpes simplex encephalitis" *Braz J Infect Dis*
20. Bhullar, Chandak, Purohit et al. (2014) "Determination of viral load by quantitative real-time PCR in herpes simplex encephalitis patients" *Intervirology*
21. Aurelius, Johansson, Sköldenberg et al. (1993) "Encephalitis in immunocompetent patients due to herpes simplex virus type 1 or 2 as determined by type-specific polymerase chain reaction and antibody assays of cerebrospinal fluid" *J Med Virol*
22. De Tiège, Laet, Mazoin et al. (2005) "Postinfectious immune-mediated encephalitis after pediatric herpes simplex encephalitis" *Brain Dev*
23. Schwarz, Mohr, Knauth et al. (2001) "Acute disseminated encephalomyelitis: a follow-up study of 40 adult patients" *Neurology*
24. Skripchenko, Zheleznikova, Alekseeva et al. (2021) "Herpesviruses and biomarkers in disseminated encephalomyelitis and multiple sclerosis in children" *Zh Nevrol Psikhiatr Im S S Korsakova*
25. De Seze, Debouverie, Zephir et al. (2007) "Acute fulminant demyelinating disease: a descriptive study of 60 patients" *Arch Neurol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12557823&blobtype=pdf | # Efficient replication of avian, porcine and human influenza A viruses in well-differentiated bovine airway epithelial cells
Ang Su, Miaomiao Yan, Georg Herrler, Paul Becher
## Abstract
We report that differentiated bovine airway epithelial cells are susceptible to avian, swine, and human influenza A viruses. This indicates that cows are a potential host of avian and mammalian influenza A viruses and may be implicated in the transmission of influenza viruses to other animal species and humans.
In recent decades, the highly pathogenic avian influenza virus (HPAIV) H5N1 has caused an increasing number of high-mortality outbreaks in domestic and wild birds resulting in mass culling of poultry worldwide [1]. Moreover, sporadic spillover of H5N1 to a growing number of mammalian species including humans has been reported [1,2]. Recently, several independent spillover events of HPAIV H5N1 viruses from birds to dairy cows in the US, resulting in subsequent spread to over 1,000 dairy herds in 17 states [3] and 41 confirmed bovine associated human cases [4] have led to increasing concern that cattle may play an important role in the zoonotic transmission of influenza viruses [1,2].
It is known that both HPAIV and low pathogenic avian influenza viruses (LPAIV) can infect humans and mammalian species [5]. Moreover, the swine-origin influenza pandemic of 2009 continues to circulate in swine and humans [5,6]. The host respiratory epithelium is usually the primary target of influenza viruses [7]. While several recent studies have demonstrated efficient replication of HPAIV H5N1 in bovine respiratory cells, only little is known about the susceptibility of the bovine airway epithelium to LPAIV and mammalian influenza A viruses (IAV) [8,9].
The major aim of this study was to investigate the susceptibility of differentiated bovine airway epithelial cells (BAEC) to avian and mammalian IAVs, and to analyze their effects on BAEC. Two LPAIV strains, H7N7 (A/duck/Potsdam/15/80) and H9N2 (A/ chicken/Saudi Arabia/CP7/98), swine influenza virus (SIV) strain H1N1 (A/sw/Bad Griesbach/IDT5604/ 2006), and the human pandemic 2009 IAV strain H1N1 A/Germany/1580/2009 (HA H1N1(pdm09)) were analyzed [10,11].
Bovine air-liquid interface (ALI) cultures derived from bronchi and tracheas were grown under ALI conditions for four weeks and infected with the four influenza virus strains at a multiplicity of infection (MOI) of 0.25, respectively. To determine the cell tropism of the four different viruses, the cells were subjected to immunofluorescence staining. At 24 h post infection (h.p.i.), the cells were analyzed for the presence of viral nucleoprotein (NP) and cellular β-tubulin, the major component of cilia. All influenza viruses used in this study infected ciliated cells (Figure 1A). The titer of released SIV H1N1 at 24 h and later time points p.i. was significantly higher than the titers of the other three viruses (Figure 1B). Compared to AIV H9N2, AIV H7N7 released almost one hundred times more virus into the supernatant. Similar differences in the titers of released virus were observed for the four influenza viruses after infection of bovine tracheal ALI cultures (see Suppl. Material). The obtained results provide evidence that cattle are a potential host of avian and mammalian influenza viruses and can be involved in the transmission of IAV to humans and animals.
To analyze the detrimental effects of infection by the avian and mammalian IAVs, we determined the loss of ciliated cells by monitoring changes in β-tubulin staining at 8 days p.i. (d.p.i.). Infection by any of the viruses resulted in a partial reduction in cilia coverage. The loss of ciliated cells was most pronounced after infection by SIV H1N1 (33.8%) and AIV H7N7 (38.5%) (Figure 1C). The other avian influenza virus (AIV H9N2) and HA H1N1(pdm09) only slightly decreased the number of ciliated cells compared to mock-infected samples (Figure 1C). In vertical sections, the thickness of the pseudostratified cell layer infected by AIV H9N2 and HA H1N1(pdm09) was slightly reduced compared to the mock-infected samples (Figure 1D). A pronounced reduction (∼40%) was observed on ALI cultures infected by SIV H1N1 and AIV H7N7 (Figure 1E). During the whole infection period analyzed, the transepithelial electrical resistance was not affected by any of the four influenza viruses (values ranged between 205 and 256 Ω•cm²), indicating that the barrier function was maintained in bovine bronchial ALI cultures infected by the avian, swine and human influenza viruses used in this study.
## Discussion
The analysis of the amount of released infectious virus and the number of foci formed on ciliated cell layers demonstrated significant differences in susceptibility to the individual influenza viruses used in this study. Avian influenza viruses in general prefer the α2,3linked sialic acid (SA) as their receptor determinant [12]. However, the AIV H9N2 virus used here had been shown by glycan array analysis to recognize both α2,3-and α2,6-linked SAs as receptor determinant [11]. It might be surprising that AIV H9N2 replicated significantly less efficient (100-fold) compared with AIV H7N7, as both α2,3-and α2,6-linked SAs can be detected in the bovine respiratory tract [13]. Mucins, which may hamper the spread of infection and have to be inactivated by appropriate viral neuraminidases [14], post-entry steps [15], and differences in the response of the innate immune system to different virus strains may also result in different replication efficiencies of AIV H7N7 and AIV H9N2. Only mild respiratory signs have been reported in many affected dairy cattle herds in the US [9]. These characteristics can be explained by limited spread of influenza A viruses in the bovine respiratory tract, which correlated with the rapid recovery ability of bovine airway cells. Future investigations will extend the application of this model to other influenza viruses, including avian and bovine HPAIV H5N1, or assess whether human-derived airway epithelial cells display comparable strain-specific differences in virulence.
## References
1. Krammer, Hermann, Rasmussen (2025) "Highly pathogenic avian influenza H5N1: history, current situation, and outlook" *J Virol*
2. Peacock, Moncla, Dudas (2025) "The global H5N1 influenza panzootic in mammals" *Nature*
3. Abdelwhab, Mettenleiter (2023) "Zoonotic animal influenza virus and potential mixing vessel hosts" *Viruses*
4. Zell, Groth, Krumbholz (2020) "Cocirculation of swine H1N1 influenza A virus lineages in Germany" *Viruses*
5. Denney, Ho (2018) "The role of respiratory epithelium in host defence against influenza virus infection" *Biomed J*
6. Bordes, Gerhards, Peters (2024) "H5N1 clade 2.3.4.4b avian influenza viruses replicate in differentiated bovine airway epithelial cells cultured at airliquid interface" *J Gen Virol*
7. Halwe, Cool, Breithaupt (2025) "H5N1 clade 2.3.4.4b dynamics in experimentally infected calves and cows" *Nature*
8. Fu, Dürrwald, Meng (2019) "Infection studies in pigs and porcine airway epithelial cells reveal an evolution of A(H1N1)pdm09 influenza A viruses toward lower virulence" *J Infect Dis*
9. Punyadarsaniya, Liang, Winter (2011) "Infection of differentiated porcine airway epithelial cells by influenza virus: differential susceptibility to infection by porcine and avian viruses" *PLoS One*
10. Liu, Huang, Smits (2022) "Human-type sialic acid receptors contribute to avian influenza A virus binding and entry by hetero-multivalent interactions" *Nat Commun*
11. Nelli, Harm, Siepker (2024) "Sialic acid receptor specificity in mammary gland of dairy cattle infected with highly pathogenic avian influenza A(H5N1) virus. Emerg Infectious Dis"
12. Matrosovich, Matrosovich, Gray (2004) "Neuraminidase is important for the initiation of influenza virus infection in human airway epithelium" *J Virol*
13. Borau, Stertz (2021) "Entry of influenza A virus into host cells -recent progress and remaining challenges" *Curr Opin Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12163611&blobtype=pdf | # EDITED AND REVIEWED BY
Michael Kogut, Yan Zhou, Hang Yang, Shaohui Wang
## Introduction
In the field of bacteriology, the rapid rise of drug-resistant bacterial infections and the lack of effective antimicrobial agents have necessitated the development of alternative therapies. Bacteriophages, or phages, are bacterial viruses that are abundant and diverse, with lytic phages possessing the ability to invade bacterial cells, disrupt their metabolism, and cause cell lysis. These properties make lytic phages potential biocontrol agents for treating bacterial infections in humans and animals, particularly against drug-resistant bacteria. Phages have, therefore, global attention. Phage-derived proteins, such as endolysins, holins, polysaccharide depolymerases, and peptidoglycan hydrolases, have also demonstrated antibacterial activity. However, despite the increasing interest in phage-based biocontrol strategies, there is a need for more comprehensive studies to better understand the interplay between phages or phage-derived proteins and bacteria or biofilms. Thus, in this Research Topic, four original research articles as well as one review article were published.
## Organization of the Research Topic
This Research Topic aims to publish original research and review articles that explore the application of phages or phage-derived proteins as antimicrobial agents against animal bacterial infections or biofilms. Given the escalating global challenge of drug-resistant bacteria, they highlight the growing importance and potential of phages and phage-derived proteins in human health, aquaculture, and agriculture. Five articles focused on phages and phage-derived proteins research across various pathogenic bacteria species, including Aeromonas hydrophila, Proteus mirabilis, Salmonella abortus equi, Staphylococcus aureus, and Streptococcus bovis/Streptococcus equinus complex (SBSEC).
Aeromonas hydrophila, a Gram-negative facultative anaerobic bacterium, is a common pathogen of freshwater farmed animals. It is pathogenic to a wide range of fish, amphibians, and reptiles and can cause systemic and ulcerative infections, including septicemia, gill rot, and kidney disease. Wang et al. characterized a novel virulent Aeromonas hydrophila phage phiA051 isolated from aquaculture water. They found the genome of phage phiA051 has high similarity to many prophages of Aeromonas spp., suggesting its prophage origin. Proteus mirabilis is a Gram-negative, rod-shaped bacterium widely found in natural environments. It is known for causing a range of severe illnesses in mammals, particularly urinary tract infections (UTIs). Wu et al. evaluated the therapeutic efficacy of phage P2-71 against Proteus mirabilis in vivo and in vitro environments. Results revealed that in vivo, phage treatment significantly lowered bacterial concentrations in the urine on Days 1 and 3 (p < 0.0001), achieving a maximum reduction of 4.602 log 10 CFU/mL. Their findings demonstrated that phage P2-71 is a promising alternative therapy for UTIs caused by MDR Proteus mirabilis. Salmonella abortus equi is a prominent pathogen known to cause abortion in equidae (horses, donkeys, and mules). Phage depolymerase breaks the bacterial polysaccharide structure, repeating polymer units and facilitating phage infection of bacterial cells. Zhao et al. investigated a Salmonella abortus equi phage 4FS1 and its depolymerase, Dpo36. Their findings confirmed that Dpo36 effectively disrupts biofilms and exhibits potent antimicrobial activity against S. abortus equi in both in vitro and in vivo settings. Staphylococcus aureus is one of the most important zoonotic pathogens and can be transmitted to humans through the meat diet routes, causing necrotizing pneumonia. Phage lysin is a cell wall hydrolytic enzyme synthesized by the phage gene coding in the later stage of phage infection of bacteria. Lysin can target peptidoglycan in bacterial cell walls, causing peptidoglycan lysis and resulting in bacterial cell wall rupture, leading to bacterial death. Zhang et al. investigated the therapeutic efficacy of phage lysin LysGH15 against necrotizing pneumonia caused by Staphylococcus aureus in a rabbit model. The study revealed that LysGH15 treatment effectively reduced number of bacteria in infected rabbit lungs, inhibited the production of bacterial toxins, reduced the production of cytokines, significantly improved the pathological manifestations of lung tissues, and ultimately increased the survival rate. Their results suggest that LysGH15 has the potential to be used as a novel antimicrobial agent for the treatment of necrotizing pneumonia caused by Staphylococcus aureus. Streptococcus bovis/Streptococcus equinus complex (SBSEC) comprises eight (sub)species, with several opportunistic pathogenic members. These SBSEC species are associated with metabolic disorders in ruminants, resulting in economic losses to the global livestock industry. Moreover, the emergence of antimicrobial resistance (AMR) in SBSEC strains, particularly against commonly used antibiotics, poses serious concerns to the livestock industry. Park et al. reviewed SBSEC and their phages in ruminants. In the review, authors discussed the taxonomy, AMR characteristics, and diversity of SBSEC, and the potential of SBSEC-specific phages, focusing on recent advances in basic research and biotechnological applications in ruminants. They pointed out the potential and limitations of phage therapy and highlighted that developing phage cocktails, screening for strictly lytic phages, and exploring strategies to minimize resistance development, such as combination therapies with antibiotics or phage-derived enzymes, could enhance the efficacy of phage therapy.
## Conclusion
In conclusion, these articles underscore the significance and essential role of phage and phage-derived protein-based biocontrol strategies as novel and effective therapeutic agents in eliminating and reducing pathogenic bacteria. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12282087&blobtype=pdf | # Metatranscriptomic detection of rabbit hemorrhagic disease virus 2 in karoro (southern black-backed gulls)
Stephanie Waller, Chris Niebuhr, Jessica Darnley, Kate Mcinnes, David Winter, Edward Holmes, Jemma Geoghegan
## Abstract
R abbit hemorrhagic disease virus 2 (RHDV2; species Lagovirus europaeus) is a highly pathogenic lagovirus (Caliciviridae) responsible for a lethal disease in rabbits and hares (1). First identified in Europe in 2010 (2), RHDV2 has spread to over 35 countries (3). RHDV2 is believed to have arrived in New Zealand's North Island in 2016 and has become established in both wild and domestic rabbit populations (3). In contrast, rabbit hemorrhagic disease virus 1 has been present in New Zealand since its illegal release as a biological control agent in 1997 (4).The RHDV2 strain circulating in New Zealand belongs to the GI.3P-GI.2 variant-a recombinant of the non-structural protein of a benign GI.3P lagovirus with the structural protein of GI.2 RHDV2 (3, 5) (Fig. 1). This variant has been detected in Europe (3, 5-7), North America (3, 8, 9), and China (3, 10), but notably not in Australia where RHDV2 has circulated since 2014 (3,11). This suggests that the New Zealand incursion likely originated from outside Australia, although its precise source remains unknown (3).Since its introduction, RHDV2 has spread within New Zealand, including at least two transmission events from the North Island to the South Island (3). The Cook Strait, which separates the islands, is approximately 23 km wide at its narrowest point. Although the exact mechanism of inter-island transmission is unknown, potential pathways include human-mediated movement, contaminated fomites, wind-assisted dispersal of fly vectors, or scavenging birds (3,17,18).Through metatranscriptomic sequencing, we recovered one complete and one nearcomplete RHDV2 GI.3P-GI.2 genome (PV602081 and PV602082) from two RNA pools derived from oral and cloacal swabs collected in 2024 from eight southern black-backed gulls (karoro, Larus dominicanus) on the South Island. Oral and cloacal swabs were stored separately in DNA/RNA Shield (Zymo Research) immediately upon collection and subsequently at -80°C until total RNA was extracted using the ZymoBIOMICS MagBead RNA kit (Zymo Research). These viral sequences, with abundances of 30 (oral) and 1,159 (cloacal) reads per million, exhibited over 98% nucleotide sequence similarity in both structural and non-structural regions to the RHDV2 previously detected in European rabbits (Oryctolagus cuniculus) from the Otago region (South Island, New Zealand) in 2019 (3) (Fig. 1). Southern black-backed gulls are opportunistic feeders that consume a variety of terrestrial and marine organisms, as well as organic waste from farms and landfills (19). The presence of RHDV2 in these gulls is therefore likely dietary in origin, which is supported by the presence of rabbit reads in the metatranscriptomic data. In addition, RHDV2 is not believed to replicate in avian hosts.The observation that RHDV2 is present in avian samples, albeit likely of dietary origin, makes it theoretically possible that birds act as mechanical vectors for virus transmission, perhaps through contaminated feces. This may even explain the spread of RHDV2 between New Zealand's islands. While southern black-backed gulls are largely resident
and not known for long-distance migration, there is strong evidence for their dispersal between the North and South Islands (20). The detection of complete RHDV2 genomes in these scavenger birds warrants further investigation of their potential role in viral dissemination. More broadly, these birds may influence the long-range spread of pathogens, shape viral dynamics, and present challenges for disease outbreak contain ment.
## References
1. Hall, King, Connor et al. (2021) "Age and infectious dose significantly affect disease progression after RHDV2 infection in naïve domestic rabbits" *Viruses*
2. Rouco, Aguayo-Adán, Santoro et al. (2019) "Worldwide rapid spread of the novel rabbit haemorrhagic disease virus (GI.2/RHDV2/b)" *Transbound Emerg Dis*
3. Hall, Trought, Strive et al. (2024) "First detection and circulation of RHDV2 in New Zealand" *Viruses*
4. O'keefe, Tempero, Motha et al. (1999) "Serology of rabbit haemorrhagic disease virus in wild rabbits before and after release of the virus in New Zealand" *Vet Microbiol*
5. Abrantes, Droillard, Lopes et al. (2020) "Recombination at the emergence of the pathogenic rabbit haemorrhagic disease virus Lagovirus europaeus" *GI.2. Sci Rep*
6. Miao, Qi, Veldkamp et al. (2019) "Immunogenicity in rabbits of virus-like particles from a contemporary rabbit Haemorrhagic disease virus type 2 (GI.2/RHDV2/b) isolated in the Netherlands" *Viruses*
7. Fitzner, Kesy, Bulenger et al. (2021) "Evidence of independ ent introductions of RHDV2 strains in Poland based on the genome analysis of viral isolates from 2016-2018" *Acta Biochim Pol*
8. Ambagala, Schwantje, Laurendeau et al. (2021) "Incursions of rabbit haemorrhagic disease virus 2 in Canada-Clinical, molecular and epidemiological investigation" *Transbound Emerg Dis*
9. O'donnell, Xu, Moran et al. (2021) "Coding-complete genome sequences of emerging rabbit hemorrhagic disease virus type 2 isolates detected in 2020 in the United States" *Microbiol Resour Announc*
10. Hu, Wei, Song et al. (2021) "Emergence of rabbit haemorrhagic disease virus 2 in China in 2020" *Vet Med Sci*
11. Ramsey, Cox, Strive et al. (2020) "Emerging RHDV2 suppresses the impact of endemic New-Data Letter Journal of Virology July"
12. "and novel strains of RHDV on wild rabbit populations" *J Appl Ecol*
13. Katoh, Misawa, Kuma et al. (2002) "MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform" *Nucleic Acids Res*
14. Nguyen, Schmidt, Haeseler et al. (2015) "IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies" *Mol Biol Evol*
15. Kalyaanamoorthy, Minh, Wong et al. (2017) "ModelFinder: fast model selection for accurate phylogenetic estimates" *Nat Methods*
16. Paradis, Strimmer (2004) "APE: analyses of phylogenetics and evolution in R language" *Bioinformatics*
17. Galili (2015) "dendextend: an R package for visualizing, adjusting and comparing trees of hierarchical clustering" *Bioinformatics*
18. Abrantes, Van Der Loo, Pendu et al. (2012) "Rabbit haemorrhagic disease (RHD) and rabbit haemorrhagic disease virus (RHDV): a review" *Vet Res*
19. Neimanis, Ahola, Pettersson et al. (2018) "Overcoming species barriers: an outbreak of Lagovirus europaeus GI.2/RHDV2 in an isolated population of mountain hares (Lepus timidus)" *BMC Vet Res*
20. Heather, Robertson (2015) "The field guide to the birds of New Zealand"
21. Rowe (2013) "Dispersal of southern black-backed gulls (Larus dominicanus dominicanus) banded in Canterbury" *Notornis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12607894&blobtype=pdf | # Climate change and neurotropic vector-borne viruses: addressing emerging threats through a One Health approach
Kamalika Roy, Rajyashree Basu, Anirban Basu
## Abstract
Vector-borne diseases are mainly transmitted through the bites of infected arthropods. They are a major public health concern as they account for more than 700,000 deaths annually. Among many vector-borne pathogens, the neurotropic viruses have been contributing to the increased number of deaths across the globe due to severe neurological complications. Despite the advancement of vector control strategies, the prevalence and severity of neurotropic viral infections have not been alleviated till date. Anthropogenic activities cause persistent fluctuations in temperature and weather trends. This plays a major part in shaping the fate of transmission dynamics and pathogenesis of such diseases. Changes in climatic factors, such as global warming and delayed withdrawal of monsoon, have had huge impacts on stretching the window of disease transmission worldwide. The abundance, survival, feeding activity, and vectorial competence of the arthropods are expected to increase with rising temperatures. This review aims to discuss how climate change affects ecosystems, thereby influencing vectors and the associated neurotropic viruses. It also highlights the urgent need for the "One Health" strategy. It is a concept that recognizes that humans and animals do not exist in isolation and are part of a larger ecosystem where their activity and health are interconnected to one another. This holistic approach is essential in addressing the emerging threats posed by climate change, rising rates of infection, and epidemics across the globe.
T he pathology of the central nervous system (CNS) is a public health concern, and the arboviral neurotropic infections are a major contributor to the problem. Neurotropic viruses are potentially capable of invading the CNS, manipulating various cell popula tions, and promoting neuropathogenesis (1). Among many, the viruses belonging to the family Flaviviridae, such as Japanese encephalitis virus (JEV), dengue virus (DENV), West Nile virus (WNV), Zika virus (ZIKV), and Rhabdoviridae, such as Chandipura virus (CHPV), have been contributing to the increasing numbers of deaths due to severe neurologi cal complications (Table 1). These neurotropic viruses are mainly transmitted by the mosquito and sand fly vectors (2,3). JEV, a principal causative agent of viral encephalitis, is transmitted by the bite of the Vishnui group of Culex mosquitoes like Culex tritaenio rhynchus, Culex vishnui, etc. (4). Despite the advancement of vector control strategies, the annual clinical incidences of JEV have not been alleviated below 100,000. Among the patients with JEV, the case-fatality rate can be as high as 30% (4). Approximately 3 billion people are exposed to the risk of JEV infections in the endemic countries of Southeast Asia and the Western Pacific Regions (4). Dengue fever is also raising concerns day by day due to an eightfold increase in dengue cases globally, with approximately 100-400 million active infections (5,6). DENV, transmitted by the bites of female Aedes aegypti and Aedes albopictus mosquitoes (5), leads to a broad spectrum of neurological manifestations like encephalopathy, encephalitis, meningitis, extensive neuroinflamma tion, and neuronal cell damage (2). The severity and prevalence of dengue are more alarming, as nearly 96 million of the global dengue infections manifest clinically, posing major risks to almost 3.9 billion people worldwide (5). Culex-borne neuroinvasive WNV, which was initially thought to be endemic to West Asia, the Middle East, Africa, and Europe, has been causing recurrent and fatal outbreaks in the United States in recent times (7). Cases of Aedes-transmitted Zika virus infections still persist in 89 endemic countries, including America (8). CHPV infection, transmitted primarily by sand flies like Phlebotomus papatasi, Phlebotomus argentipes, and Sergentomyia sp., also fairly accounts for a significant proportion of acute encephalitis syndrome (AES) in an endemic country like India. Sixty-four cases of CHPV-mediated AES have been reported in India between early June and mid-August 2024 (9). India has seen the largest outbreak of CHPV in the past 20 years (9). The high case-fatality ratio (56%-75%) from CHPV infection is alarming (9) and needs global attention. Several factors, like increasing urbanization, population growth, migratory popula tions, and changes in climatic factors such as global warming and delayed withdrawal of monsoon, have a huge impact on stretching the window of transmission of the neurotropic viruses across the globe (28). Climate can directly affect the disease-causing pathogen, the vector, the human and non-human hosts, and the environment (Fig. 1). These, in turn, determine the fate of the geographic outbreaks, transmission dynamics, emergence, and re-emergence of the neurotropic viruses (29).
Ambient temperature is critically important as the insect vectors carrying arboviruses are ectothermic and immensely susceptible to the slightest changes in temperature (29,30). It determines the survival, feeding activity, and abundance of the vectors, as well as the rate of development of the pathogen within the vectors (28,29). The extrinsic incubation period (EIP), the time between the intake of a pathogen by the vector through a blood meal from the infected host and the vector becoming infectious after the amplification and circulation of the pathogen inside the vector's body, can be altered by temperature fluctuations (29,31). The vector microbiome has a profound effect on the development of pathogens inside the vector. Climatic and environmental disruptions can also alter this microbial composition, thereby modulating the pathogen transmission (32). It has been reported that 49% of the emerging viruses are neurotropic, leading to encephalitis or other serious neurological clinical symptoms. More importantly, their emergences are triggered by environmental, ecological, or human demographic changes (33). Several reports have also suggested the expansion of brackish water bodies in coastal zones due to global warming-driven sea-level rise, followed by increased abundances of salinity-tolerant JEV-carrying mosquito vectors such as Culex sitiens and C. tritaeniorhynchus, as well as the adaptation of the DENV-carrying fresh-water mosquito vectors like A. aegypti and A. albopictus to salinity (34).
Therefore, the relationship between climate change, especially the warming of the earth due to anthropogenic activities, and the epidemiology of vector-borne diseases has grown a considerable amount of interest among researchers recently. However, there are contradictory reports on the correlation of ambient temperature with vector competence (30). A few studies have associated increased vectorial capacity with lower temperature (22,(35)(36)(37). On the other hand, a couple of studies on Aedes mosquitoes infected with DENV-2 (38,39) and Culex mosquitoes infected with WNV (40) have shown that the rise in temperature is positively correlated with the infection rate and inversely correlated with the EIP. The current global expansion of some major arthropod vectors into temperate regions is adding to the complexity of this conflicting evidence (41). For example, the spread of A. albopictus from Southeast Asia to North America and Europe has led to the subsequent outbreaks of dengue fever in France and Hawaii (42)(43)(44). In the past years, DENV, WNV, and many other viruses have emerged and re-emerged in Asia, Europe, the Americas, and the Middle East (45). The Zika virus outbreak also caused an epidemic in South America (46). All of these neurotropic virus emergences are posing a threat to mankind and resulting in a global health crisis. Over the past few decades, consequences of the earth's climatic shift have been drastically affecting the poor and developing countries that are already facing the worst infectious disease scenarios. The emergence of new transmission zones, even in the developed countries, is a matter of great concern. Since vector-borne diseases, including neurotropic viral ones, are influenced by an ecosystem that comprises a vector-pathogen-host tripartite relation ship linked to environmental parameters, a multidisciplinary "One Health" approach is urgently needed to address this crisis. In this review, we focus on the current knowl edge of the paradigm shifts in neurotropic arboviral transmissions driven by climatic deformations and a unified holistic approach as the method of prevention.
## IMPACT OF CLIMATE CHANGE ON THE VECTOR-BORNE NEUROTROPIC VIRAL INFECTIONS
Climate change is the persistent alteration in weather trends and temperature, which may arise naturally, like fluctuations in the solar cycle, or through anthropogenic activities. Since the 19th century, the primary cause of climate change has been mainly attributed to the combustion of non-renewable resources such as fossil fuels (like coal, oil, and gas) and other human activities (47). Climate change poses the most serious present-day challenge for humanity. The current unsustainable pattern of development is increasing the exposure of people and different ecosystems to climate hazards. The Intergovernmental Panel on Climate Change (IPCC) has reported that 3.3 billion of the world's population is currently highly vulnerable to climate change, with the situation growing worse (48,49). The rise of the mean global temperature by 1°C above preindustrial levels, due to anthropogenic activities, has had a profound impact on the climate (48). This includes an increase in the number of warm days and nights, extreme heat events, an accelerating rise in sea levels, and a decrease in snow cover. Extreme precipitation events have also been reported in both wet and dry regions (29). According to the IPCC in 2021, global temperature could rise by 2°C-4°C due to increasing emissions of carbon dioxide in the atmosphere. This rise is expected to be higher in the continental areas of the Northern Hemisphere (32). The poorest countries have been the worst hit and face the most due to the spread of infectious diseases caused by this climate crisis, the neurotropic arboviral infection being a prominent one among them (31,49). It has been reported that climatic alterations can facilitate spillover between previously geographically isolated species. The global movement of pathogens and vectors and range shifts of wildlife will be amplified by climate, and these can put larger populations at higher risk of viral spillover events. For example, in the recent past, pandemics and epidemics like COVID, Ebola, and SARS have all been caused by viruses carried by rodents, bats, and other animals (50). The Zika virus, initially endemic in primates, began to be transmitted to humans by A. aegypti, which has now emerged in temperate countries due to climate change (51). Climate-induced biodiversity loss will also make ecosystems more susceptible to invasive species (50). Rapid thawing of the permafrost due to global warming increases the chances of novel disease outbreaks due to the release of unknown bacteria and viruses. Temperatures in the Arctic region have been reported to rise twice as fast as other regions across the globe, and two-thirds of the near-surface permafrost is projected to melt by 2100 (50). Scientists have predic ted the emergence of new combinations of species at high elevations, in biodiversity hotspots, and areas with high human population density in Africa and Asia, which may enhance cross-species viral transmission by almost 4,000 times (52).
## Climate change and JEV infections
Studies from India, China, Vietnam, Taiwan, and Nepal have shown a significant association of temperature, rainfall, relative humidity, and solar radiation with JE cases (53). In the Gorakhpur district of India from 2001 to 2016, the use of generalized additive models to verify this association predicted that with every one-unit rise in mean temperature, relative humidity, and rainfall, the average JE admissions would increase by 22.23%, 5.22%, and 0.66%, respectively, and the JE mortality rate would increase by 13.27%, 3.27%, and 0.94%, respectively. In contrast, a decrease in wind speed and solar radiation by every unit increased JE admission by 11.42% and 17%, respectively, and JE mortality rate by 4.88% and 9.37%, respectively (53). The upsurge in JE cases has been linked to the increasing temperature and rainfall in China as well (54). Within a certain range, an increase in temperature results in the rapid development of larvae, shorter time between blood meals, and less incubation time for viral infection in mosquito vectors carrying JEV (55). Higher humidity allows longer survival and farther dispersal of mosquitoes. This allows them to feed on an infected animal and survive long enough to transmit the virus to humans or other animals (55). A study conducted in Chongqing in Southwest China showed that the number of JE cases began to increase when the temperature was greater than 16°C (55). In Taiwan, the number of JE cases increased when the temperature reached 22°C (56). Temperature has been a dominant factor influencing the monthly JEV outbreak in the Gansu Province of China (57). In India, the average temperature ranging between 22.8°C and 34.5°C facilitates JEV transmission (58), whereas the temperature ranging from 21.0°C to 25.2°C in China has been found optimal for the same (59). Thus, a JE outbreak may be facilitated by two factors: adaptations in agriculture (such as the use of pesticides, adoption of paddy cultivation, and creation of modern pig farms) and global warming. Temperature and precipitation have been reported to contribute to variations in the density of mosquitoes (59,60) and, according to the Konno transmission model, are also highly related to the occurrence of JE (61,62).
## Climate change and DENV infections
Around 40% of the world's population is at higher risk of dengue fever as they live in tropical and subtropical climates (63,64). It was initially thought that the disease was endemic in tropical regions; however, travelers in southern Europe also started getting infected (65). The northwest global expansion of dengue was driven by several factors, with climate change being the most crucial one among them (65). The Aedes mosquito population is blooming as cities expand, human population increases, and more people move to urban areas. This is especially true in overpopulated regions that have poor waste management and sanitation (66). Aedes mosquitoes may now spread the dengue virus for much longer into the year due to increased longevity and activity of the species at higher temperatures (66). Therefore, one profound effect of global warming is a longer transmission season. Clearly, climate change, caused by anthropogenic activities, is considered to be a major contributing factor in the rapid spread of this pandemic disease (63,66), and it is estimated that around 60% of the world's population will be affected by dengue fever by 2080 (63).
Dengue incidence has been shown to respond nonlinearly to temperature, peaking at 27.8°C and then declining at higher temperatures, with predicted maximum transmission at 29°C for A. aegypti and 26°C for A. albopictus (67). Some low-elevation equatorial areas are projected to see small declines in dengue incidence due to climate change. However, the majority of locations, including some cooler regions of Mexico, Peru, Brazil, Bolivia, and many of the largest cities in America, are predicted to see a 150% increase in incidence (68). Among the 21 countries included in a study, 15 countries were projected to see an increase under all scenarios (68). The National Center for Vector Borne Diseases Control (NCVBDC) states that dengue cases have quadrupled from 2015 to 2023 in India. The optimal temperature range for increased dengue transmission is a mean range of 27°C-35°C, although it mostly varies based on local climatic conditions (69). In Singapore, a 14% increase in the incidence rates of dengue within a 40-year span has been attributed to rising temperatures (70). In Brazil, the areas that were more urbanized and exposed to high temperatures for a prolonged time between 2014 and 2020 had higher incidence rates of dengue (71). On the other hand, the optimal temperature for disease propagation can vary depending on the local and regional climates, too. In India, the rise in temperature, even up to 39°C, has facilitated dengue transmission in Delhi, whereas temperatures ranging between 27°C and 35°C were optimal for the same in Pune (69). A study conducted in Pune between 2004 and 2015 has shown that moderate rainfall over a prolonged period can increase the dengue mortality rates, whereas heavy rains can reduce it by flushing out the vector's habitat. A relative humidity of 60%-78% has also been reported to be optimal for dengue transmission in this region (69).
## Climate change and WNV infections
WNV has been experimentally shown to replicate across a broad range of temperatures, starting from 14°C in ectothermic mosquitoes (72) to 45°C in febrile avian hosts (73). The temperature threshold for the survival of C. tarsalis is demonstrated to be between 14°C and 35°C, where temperature positively correlates with the development of the vector (74,75). The replication cycle was shown to be completed more quickly in mosquitoes at higher temperatures (76). A clear link was present between the intensity of the outbreak in humans and extreme heat (77,78). However, extremely high temperatures have been shown to cause a decline in mosquito activities, including reducing larval survival and virus growth (79). Warmer conditions enable the spread of WNV in new areas by expanding the seasonal abundance and range of the vector species, as well as increasing their transmission competence (79). High spring and summer tempera tures, drier winters, and water scarcity, along with changes in land, such as frequently irrigated crops and highly fragmented forests, were found to be the major determinants for increased WNV incidence across Europe (80). Researchers studied the association between heatwaves and WNV incidence in humans in Israel and showed that an extreme temperature rise early in the summer is a good indication of rising populations of vectors (81). Platonov et al. (82) studied the outbreaks in the Volgograd province in Russia and showed that years with milder winters and hotter summers led to an increase in Culex mosquitoes during the epidemic season. The year 1999 marked the first appearance of WNV in the Western Hemisphere, starting from New York City (83), then reaching Canada and Central America by 2002, thereafter spreading to the Pacific coast (California) by 2003 (84), and then to Argentina by 2005 (85,86), where it infected numerous species of birds, along with humans and other mammals. It became endemic across most North American temperate regions (87), and the initial outbreak in the USA is thought to be due to a prior drought (87,88). WNV is a notable concern in the Canadian Prairies, where it has been observed that warmer and drier summers have led to an increased scarcity in water levels (89,90). Observations made in Florida note that spatial and temporal differences in periods of droughts and rain events are associated with the increase in WNV infection in humans and sentinel chickens (91)(92)(93). Drought conditions lead to an increase in the number of larval breeding sites with fewer competitors, as well as less mosquito predators, with closer proximity of birds and mosquito vectors that enhance virus transmission, hence increasing the prevalence of vector populations in semi-permanent wetlands. Springtime drought, followed by a wet summer, was reported to be a reliable predictor of the incidence of WNV infections in humans (91)(92)(93). WNV is known to affect multiple species of mosquitoes and birds and currently has a widespread distribution across Africa, southern and eastern Europe, western Asia, the Middle East, and Australia. Biodiversity loss is also thought to promote patterns of transmission of disease as vector-host encounter rate increases with the decrease in host community diversity (94). In Missouri, a negative correlation has been found between WNV infection in vectors and bird diversity at the regional scale, and at a national scale with humans in the USA (95). However, one report suggests that avian biodiversity loss can also lead to a decrease in mosquito infection rates and avian seroprevalence in Atlanta (96). In the USA alone, between 1999 and 2021, more than 55,000 cases of WNV have been reported, out of which more than 27,500 cases developed neuroinvasive disease and more than 2,500 deaths (80,97).
There are cases of WNV emerging or re-emerging at the edges of current endemic zones and in high-latitude regions (98,99). Similarly, by 2050 and 2080, the suitable range for WNV in North America is projected to extend northward and to higher altitudes, potentially leading to both native and non-native species getting infected (98). Expansion of outbreak locations and high-risk areas is predicted for the future in Europe and South America as well, with more pronounced changes expected under high greenhouse gas emission scenarios (100).
## Climate change and ZIKV infections
Since 2013, the Zika virus has spread to at least 49 countries and territories (101), reporting an estimated 150,000-500,000 cases, with Brazil alone facing 3,000 cases (102). There has been some research linking the 2015 Zika outbreak to El Niño in South America (103). Research done during the 2016 Zika outbreak suggests that transmission may be restricted to warmer and less seasonally variable parts of the world as compared to dengue (104), and the minimum temperature for ZIKV transmission by A. aegypti is around 5°C higher than that of DENV (105,106). It is predicted that if climate change is not mitigated, as many as 1.33 billion new people (1.17 billion outside the Line of Actual Control or LAC) could be pushed into areas that would become suitable for ZIKV transmission with respect to temperature, even though it is only confined to the tropics currently (106). It is estimated that a total of 737 million people worldwide (635.8 million outside LAC) are at risk of exposure to year-round climate suitability that aids in ZIKV transmission, especially in South and East Asia, and sub-Saharan Africa (106). Some of the regions with populations of 100 million or more people, which are projected to experience higher rates of ZIKV transmission due to climate suitability, are East Africa, North Africa, high-income North America, East Asia, the Middle East, and Western Europe (considering regions designated by the Global Burden of Disease or GBD study) (106). Usually, a temperature ranging from 18°C to 34°C is ideal for ZIKV transmission, with a peak at 29°C (107). The warming of winter temperatures enhances the overwinter egg survival of A. aegypti, which in turn expands the geographic range of this vector mosquito (107). Similarly, higher temperatures in seasons like spring, summer, and autumn actually prolong the transmission season, even in temperate regions (107). Other than the temperature, the precipitation rate also influences the transmission dynamics of ZIKV. It has been reported that the ZIKV outbreak in America during 2015-2017 was inversely correlated with precipitation, whereas drought actually enhanced the viral transmission (106).
## Climate change and CHPV infections
While endemic to India, CHPV is also likely to be present in other countries in Asia and Africa (108). The virus has been reported to persist in parts of Gujarat, Maharashtra, and Telangana, and its recurring outbreaks have become a health concern, affecting mostly children below the age of 15 (108,109). It is yet to be explained why these parts of India are affected by the CHPV virus despite Phlebotomine sand flies being prevalent in all parts of the country. The virus remains infective in sand flies and cell culture supernatant at 4°C for 8 weeks, and it remains viable at 37°C for 18 days in infected cell supernatants. However, the virus loses virulence within a week when infected sand flies are stored at 37°C (109). Climate change may expand the geographic footprint of epidemics caused by CHPV through shifts in temperature, humidity, and seasonal monsoon patterns. It can accelerate CHPV transmission into newer regions, thus amplifying the risk of widespread and more frequent epidemic outbreaks (110). Despite the decrease in the case numbers of CHPV-mediated AES, further outbreaks can be triggered by climate change and drawn-out monsoons (108).
## ONE HEALTH STRATEGY TO COMBAT THE CRISIS
The alarming expansion of the geographic ranges of neurotropic arboviral infections caused by JEV, DENV, WNV, Zika, or CHPV is triggered by certain environmental fluctuations. These fluctuations affect three main components: the virus, the arthropod vector, and the vertebrate host. The transmission of the neurotropic arboviruses happens through the following cycles, such as the enzootic or sylvatic cycle, the humanamplified or urban cycle, and the epizootic or rural cycle (111). In the sylvatic cycle, the virus is maintained in the ecosystem through interactions between the birds or wild animals, such as non-human primates, rodents, etc., and the associated arthropod vectors (23). For many neurotropic arboviruses, there are dead-end hosts like domestic animals or humans. These hosts can stop the transmission cycle of the virus as the viral load in their blood is too low to be transmitted by the biting vectors, but the clinical symptoms of the disease can still persist in them (112). For example, the transmission of JEV and WNV involves sylvatic cycles where the vector Culex mosquitoes spread the viruses among birds, which are the amplifying hosts. They can harbor the viruses in them for several days, and their migration can carry the viruses over long distances (113). Pigs also act as the amplifying hosts for JEV (113). The vector-borne transmission of both of these viruses infects various other vertebrates, including dead-end hosts such as humans and equines (113). Rhabdoviruses are also known to infect a wide range of hosts, such as vertebrates (like mammals, fish, birds, and reptiles), invertebrates, protozoans, plants, and fungi. One such virus is CHPV. The transmission cycle of CHPV in nature involves sand flies, rodents, and small mammals (114) and has been isolated from hedgehogs and sand flies in Nigeria and Senegal (115,116). It has also been isolated from humans (117)(118)(119). Anti-CHPV antibodies have been found in humans (117,119,120), toque macaques (121), and cattle, pigs, goats, sheep, and other domestic animals (122), and a broader geographical distribution was indicated when antibodies against CHPV were reportedly found in monkeys in Sri Lanka (116). Infected domestic animals only developed ulcers at the inoculation site, and no other clinical symptoms were shown in experiments. Isolating the virus from tissues other than the inoculation site was unsuccessful. Thus, further research about the disease and how animals respond to it under natural conditions is lacking (114). Contrastingly, several other neurotropic arboviruses can be transmitted among humans by vector bites. Viruses like Zika and DENV were initially transmitted by forest mosquitoes among the African and Asian primates, respectively. However, the vectorial capacity of the anthropophilic A. aegypti mosquitoes started the urban cycle of these viral transmissions, making humans a new host (123,124). Therefore, entry of a virus from the sylvatic to the urban cycle can result in the rapid spread of the virus by anthropophilic mosquitoes. On the other hand, the rural cycle involves domestic animals serving as amplifying hosts, leading to the enhanced spillover of viruses to humans in agricultural settings (125). In the context of human medicine, the urban cycle is of utmost importance because the widespread urban population is prone to massive propagation of the infection. Similarly, the areas with dense livestock populations are at risk of the rural cycle of viral transmission. So, the viruses that follow these cycles should be given attention from a consolidated perspec tive. To successfully restrict the emergence of such viruses by a One Health approach, surveillance is the foremost step, which will help to monitor the viral spread, identify livestock and humans at risk, introduce vaccination programs in areas with high rates of viral transmission, and ultimately offer means to control the disease (126). In new areas, which were previously free from emerging viral threats, the increased prevalence of vector-borne neurotropic viral infections is being facilitated by climatic alterations such as changes in average rainfall, flood situations, and rising temperatures. An integrated approach is always welcome to understand the interconnectivity between the climate, zoonoses, and public health and mitigate the dramatic resurgences of these viruses.
## One Health approach to control JEV transmission
Determining the abundance and seasonal prevalence of vector mosquitoes and their infection rates can indicate the possibility of JE outbreaks in any particular region. The distribution of the vectors is reported to be proportionally related to the JE outbreaks in India and Korea (127)(128)(129). Weather events can also regulate the population dynamics of the vectors, thereby being potential predictors in forecasting JE upsurges. Vector surveillance has been given more importance as a tool to monitor JEV emergence in Australia (130). The genetic heterogeneity among the JEV isolates and re-emergence of JEV in the temperate regions can be attributed to the migratory viremic birds as well. Migratory birds from China were reported to result in the shift in the widespread genetic variant of JEV in Japan, Korea, and Vietnam (126,131). Therefore, serological surveys among sentinel birds and wild or domestic animals may be useful in forecasting JE epidemics. In Australia, routine mosquito and reservoir host surveillance, prior to the JEV transmission to humans, has provided early warnings about the extent of the disease outbreak (132). Surveillance is followed by control measures like mass vaccina tion among sentinel animals and humans. It is thought to be effective as the number of JE cases in horses, pigs, and humans has been reduced with the advancement in the production and scale-up of JE vaccines (126). However, the cost, requirement of booster doses, and long-term follow-up to maintain immunity, along with a few reported adverse side effects, including allergic reactions, unavailability of the vaccination program during severe outbreaks, especially in poor or developing countries, are the major constraints (126,133). Hence, the non-vaccine control has to play an equally important role here. The changes in land cover and use, followed by the environmental changes in habitats due to extensive anthropogenic activities, are shaping the risk pattern of humans getting infected with JEV (134). The tropical regions having a large cover of rice fields are more prone to JEV dissemination (135). These rice fields offer excellent breeding sites for Culex mosquitoes and foraging and resting sites for birds (134). Moreover, increased domestic and industrial pig farming is providing a potential blood meal source for the mosqui toes. So, rearing pigs away from humans, the use of barn fans to protect animals from mosquito bites, and biocontrol of the JEV-carrying mosquitoes by using plant-based insecticides or larvivorous fishes may presumably hamper the transmission dynamics of JEV (126). Intermittent drainage of the rice fields can also limit the vector prevalence and subsequent disease occurrences (126,136). Besides the vaccination strategy, the implementation of modern agricultural practices and moving human populations away from the animal farms were crucial factors that facilitated the drastic reduction in JE cases in an endemic country like Japan (126). In the Gorakhpur district of India, where several climatic factors like excessive rainfall, flood situations, cultivation of paddy, low topography, and accumulation of silt in riverbeds favored mosquito breeding leading to the JEV transmission, a multidisciplinary One Health strategy had been able to limit the disease burden by reducing the morbidity and mortality rate by 90% (137). Thus, elaborate research on the various aspects of the JEV transmission cycle is required to eradicate the chances of sudden and rapid outbreaks.
## One Health approach to control DENV transmission
The severe impact of climate change on dengue transmission, the global threat it poses, and fragile vaccination strategies emphasize the requirement of a unified One Health strategy for disease management. In addition to the vector control programs, recent reports suggest that the application of omics science and biotechnology to identify new vaccine candidates and prophylactic approaches to counter the viral threat (138). Dengue control program in China, by implementation of the One Health strategy, can be an excellent case study for its efficacy in minimizing disease fatalities (139). The rise in density of A. aegypti and A. albopictus populations due to increased precipitation and temperature in China led to dengue outbreaks that could only be handled by multi-sec toral interventions. China's swift response in real-time data collection and analysis of the dengue cases at the onset of outbreaks offers a compelling framework that can be adapted by the dengue-endemic countries. In 2018, the monitoring of meteorological updates, vector density, and study of the case numbers helped predict and prevent a potential dengue outbreak in Guangdong Province (139,140). Vector surveillance in 23 Chinese provinces identified the prevalence of Aedes mosquitoes. In addition, sentinel surveillance, serological monitoring, and epidemiological investigative procedures across 16 districts of Hainan, Guangdong, Yunnan, Fujian, etc., have been proven fruitful in resisting this public health challenge (139,141). The collaboration between Yunnan Province and its neighboring Southeast Asian countries, through data sharing, enhan ces the chances of regional prevention and restricts cross-border dengue transmission (142). Also, the Wolbachia-based population replacement of Aedes mosquitoes by the successful release of mosquitoes having novel Wolbachia infections is now being carried out in China to effectively control the vector-borne DENV spread by biological means (143)(144)(145). Thus, China has achieved a significant decline in dengue cases (19,451 cases) in 2023 compared to the previous five years (43,095 cases) (139). Moreover, certain provinces like Guangdong and Yunnan have reported remarkably lower incidence rates in comparison to their neighboring countries like Sri Lanka, Malaysia, and Laos in 2023 (146,147). Through the effective application of an integrated One Health program, China has been able to reduce the mortality rates due to dengue fever to 0.01 per 100,000 between 2005 and 2023 (139). The One Health program is being adopted as a tool to fight dengue and other vector-borne zoonotic diseases not only in China but also in Brazil (148).
## One Health approach to control WNV transmission
Studies in the recent past have taken genomic surveillance of the propagating WNV into account to develop the framework of One Health surveillance. This strategy has helped in acquiring detailed knowledge of the viral evolution pattern and the influence of viral genetics on the morbidity and mortality rates of West Nile fever in the endemic regions of Italy during 2023 (149). Whole-genome sequencing, along with the epidemiology and clinical diagnosis, aided in characterizing the indigenous and imported WNV circulating at the vector-animal-human interface in Romagna (149). In the WNV-endemic Veneto region of northeastern Italy, a One Health-based surveillance plan effectively brought down the WNV infection rates among mosquitoes and birds within a year. The active human WNV infections had also been reduced to 56 in 2023 from 531 in 2022 (150). All these findings collectively show a way forward to eliminate WNV.
## Predictive models for global projections of the neurotropic viral diseases under changing climate trends
Various observational and experimental studies can be performed to understand the climate-disease linkage. Observational studies include retrospective and prospective analysis of the fluctuations in climate variables and disease patterns, along with interregional comparisons. Experimental studies range from laboratory-based reduction ist studies at the molecular and organism levels to fieldbased manipulation studies at the population level (151). There are certain analytical and advanced modeling approaches that offer quantitative global projections of disease transmission risk under various climate change scenarios. Mechanistic or process-based models use the theoretical knowledge of the biophysical mechanisms to simulate the health impacts of climate change, whereas empirical-statistical models use the empirical data on the past trend of variations to predict the future pattern of changes in the studied variables (151). A detailed dynamic simulation model of the abundance of vector mosquitoes such as Anopheles, Aedes, and Culex (91), or a matrix population model of the dynamics of Culex quinquefasciatus under fluctuating weather trends, made valuable contributions to the field (152). Predictive models were also used to project JEV outbreaks in Aus tralia (153) and China (154). A joint spatiotemporal modeling revealed the seasonal variations in the abundance of JEV vectors across India (155). The maximum entropy model projected the distribution of A. aegypti and A. albopictus across mainland China for 2041-2100, based on the impact of annual mean temperature, seasonality of temperature, and precipitation. The results predicted the expansion and emergence of suitable habitats for both species of mosquitoes (156). A predictive model based on the temperature and precipitation data of Reunion Island in France showed that decreasing precipitation is going to negatively impact the A. albopictus abundance in low-elevation areas. Contrarily, at mid and high elevations, a significant warming will counterbalance the decreasing precipitation, leading to increased abundance of this dengue vector in 2070-2100 (70). Interestingly, the different thermal niches of these two dengue vectors, A. aegypti and A. albopictus, show different shift patterns under climatic variations. The temperature-dependent, month-wise global transmission risk by these vectors was predicted by using a model of viral transmission that was empirically parameterized. The prediction was also compared with the projected risk in 2050 and 2080 based on general circulation models. The results predicted that severe climate change scenarios will expose a larger population to transmission by heat-tolerant A. aegypti, not by heat-limited A. albopictus. In South Asia and sub-Saharan Africa, the transmission potential of A. aegypti throughout the year is likely to expand, whereas the transmission potential of A. albopictus is expected to decrease, as a result of tropical warming (157). On the other hand, a few reports suggest that A. albopictus is a more invasive vector than A. aegypti, as it has a greater niche and range expansion over its shorter invasion history. Therefore, it is important to pay more attention to A. albopictus if the future climate changes promote their invasiveness (158). Another model also predicted that A. albopictus may have a greater capacity to transmit DENV than A. aegypti due to temperature and other contributing factors (13). In the recent past, a group of researchers assessed the risk of WNV outbreaks in Europe using an ensemble climate model and a multi-scenario approach. The projections of areaspecific outbreaks and populations at risk were estimated, which predicted up to a fivefold increase in WNV infections in Europe by 2040-2060, due to the climatic alterations, compared to 2000-2020. The WNVaffected land areas could increase from 15% to 23%-30%, putting 161 to 244 million European people at risk (159). Predictive spatial models were also used to estimate the risk of WNV exposure in Colorado (160). In Eastern Croatia and Northwestern and Northeastern Turkey, the climate change predictions for 2025 revealed a higher probability of WNV infection, which is likely to expand more in 2050, along with a significant increase in the prevalence of infection in blood donor populations in the WNVaffected areas (99). Therefore, such modeling-based projections can draw attention to targeted public health responses to manage the crises.
## Gaps and challenges
In recent times, increasing numbers of imported cases and severe epidemics across the globe have intensified the magnitude of the challenges faced in this domain. For example, at least 10 million cases and 6,000 deaths due to dengue have been reported in more than 80 countries as of 2023 (139). This global surge is a major factor behind the increase in the number of non-native dengue cases, even in a country like China, where the dengue control program has been proven effective (139). Population movement and urbanization are also impacting the neurotropic virus transmission and vector prolifera tion, respectively. In Malaysia, the dengue case numbers are much higher in residential urban areas than in industrial or commercial zones due to the urbanization-mediated changes in patterns of land use (139). Several preventive measures are falling short, too, such as periodic irrigation of the paddy fields to limit JEV transmission, which can be impractical in many rural areas (126). During vector monitoring, the requirement of skilled and adequate manpower for operating the equipment and identifying the species is often not meeting the standard. The emergence of insecticide-resistant vector species is another serious concern. In addition, the lack of effective therapeutics or vaccines, impaired public awareness, and inadequate community participation are hurdles that need to be overcome.
## Road ahead
Despite the existing challenges, certain agendas such as improving the risk assess ment team's early detection abilities, organizing rapid response teams, deploying modern information technologies, clarifying and distributing tasks and responsibilities among multiple sectors, sufficient information sharing, advancement of diagnostic techniques and eco-friendly control strategies, strengthening collaborations, focusing on the translational research, elimination of the vector breeding sites, betterment of the community sanitation, environmental governance, infrastructural development, government funding, and public health awareness campaigns can be adopted for the abatement of the crisis. Joint ventures of several sectors and disciplines are progressing toward the elimination or at least reduction of vector-borne neurotropic viral infections caused by JEV, WNV, or DENV in endemic countries all over the world. The model can also be executed in India to diminish fatal CHPV outbreaks. To ensure optimal protection from Zika fever, endemic countries like Singapore are also relying on the One Health concept (161).
## CONCLUDING REMARKS
This study systematically summarizes the potential threat of rapid climatic altera tions, their profound implications on transmission dynamics, and the unprecedented geographical spread of neurotropic arboviruses. It reviews a comprehensive One Health approach that works at the regional, national, and global levels to obliterate the mentioned threat. As the exposure to these viruses leads to a wide spectrum of deleterious effects on human health, the progress toward restraining the viral propaga tion needs to be escalated, and this goal can only be achieved by integrated surveillance, interdisciplinary research, and intersectoral coordination.
## References
1. Uribe, González, Kalergis et al. (2024) "Understanding the neurotrophic virus mechanisms and their potential effect on systemic lupus erythematosus development" *Brain Sci*
2. Li, Ning, Liu et al. (2017) "Neurological manifestations of dengue infection" *Front Cell Infect Microbiol*
3. Ghosh, Basu (2017) "Neuropathogenesis by Chandipura virus: an acute encephalitis syndrome in India" *Natl Med J India*
4. (2024) "Dengue and severe dengue"
5. Baskey, Verma, Mondal et al. (2024) "Geographic information system-aided evaluation of epidemiological trends of dengue serotypes in West Bengal" *India. Indian J Med Res*
6. Naeem, Naeem, Tabassum et al. (2023) "Recurrent West Nile virus outbreak in the United States in 2022: current challenges and recommendations" *J Biosaf Biosecur*
7. (2024) "Acute encephalitis syndrome due to Chandipura virus -India"
8. Cdc (2025)
9. Mackenzie, Gubler, Petersen (2004) "Emerging flaviviruses: the spread and resurgence of Japanese encephalitis, West Nile and dengue viruses" *Nat Med*
10. Misra, Kalita (2010) "Overview: Japanese encephalitis" *Prog Neurobiol*
11. Brady, Golding, Pigott et al. (2014) "Global temperature constraints on Aedes aegypti and Ae. albopictus persistence and competence for dengue virus transmission" *Parasit Vectors*
12. Cdc (2025) "Areas with risk of dengue" *Dengue*
13. Haider, Hasan, Onyango et al. (2024) "Global landmark: 2023 marks the worst year for dengue cases with millions infected and thousands of deaths reported" *IJID Reg*
14. Cdc (2025) "Current dengue outbreak. Dengue"
15. (2025) "West Nile virus: education, public health, mosquito management"
16. Campbell, Marfin, Lanciotti et al. (2002) "West Nile virus" *Lancet Infect Dis*
17. Bunning, Bowen, Cropp et al. (2002) "Experimental infection of horses with West Nile virus" *Emerg Infect Dis*
18. Usepao (2016) "Climate change indicators: West Nile virus"
19. Cdc (2024) "West Nile: symptoms, diagnosis, & treatment"
20. Kay, Jennings (2002) "Enhancement or modulation of the vector competence of Ochlerotatus vigilax (Diptera: Culicidae) for ross river virus by temperature" *J Med Entomol*
21. Socha, Kwasnik, Larska et al. (2022) "Vector-borne viral diseases as a current threat for human and animal health-one health perspective" *J Clin Med*
22. Cdc (2025) "About Zika. Zika virus"
23. Zika Outbreaks (2025)
24. Cdc (2025) "Zika symptoms and complications"
25. Sapkal, Sawant, Mourya (2018) "Chandipura viral encephalitis: a brief review" *Open Virol J*
26. De Souza, Weaver (2024) "Effects of climate change and human activities on vector-borne diseases" *Nat Rev Microbiol*
27. Rocklöv, Dubrow (2020) "Climate change: an enduring challenge for vector-borne disease prevention and control" *Nat Immunol*
28. Samuel, Adelman, Myles (2016) "Temperature-dependent effects on the replication and transmission of arthropod-borne viruses in their insect hosts" *Curr Opin Insect Sci*
29. Fouque, Reeder (2019) "Impact of past and on-going changes on climate and weather on vector-borne diseases transmission: a look at the evidence" *Infect Dis Poverty*
30. De Angeli Dutra, Salloum, Poulin (2023) "Vector microbiome: will global climate change affect vector competence and pathogen transmission?" *Parasitol Res*
31. Olival, Daszak (2005) "The ecology of emerging neurotropic viruses" *J Neurovirol*
32. Ramasamy, Surendran (0198) "Global climate change and its potential impact on disease transmission by salinity-tolerant mosquito vectors in coastal zones" *Front Physio*
33. Kramer, Hardy, Presser (1983) "Effect of temperature of extrinsic incubation on the vector competence of Culex tarsalis for western equine encephalomyelitis virus" *Am J Trop Med Hyg*
34. Turell (1993) "Effect of environmental temperature on the vector competence of Aedes taeniorhynchus for rift valley fever and venezue lan equine encephalitis viruses" *Am J Trop Med Hyg*
35. (2025) "Minireview mBio November"
36. Westbrook, Reiskind, Pesko et al. (2010) "Larval environmental temperature and the susceptibility of Aedes albopictus Skuse (Diptera: Culicidae) to Chikungunya virus" *Vector Borne Zoonotic Dis*
37. Watts, Burke, Harrison et al. (1987) "Effect of temperature on the vector efficiency of Aedes aegypti for dengue 2 virus" *Am J Trop Med Hyg*
38. Mclean, Clarke, Coleman et al. (1974) "Vector capability of Aedes aegypti mosquitoes for California encephalitis and dengue viruses at various temperatures" *Can J Microbiol*
39. Dohm, 'guinn, Turell (2002) "Effect of environmental temperature on the ability of Culex pipiens (Diptera: Culicidae) to transmit West Nile virus" *J Med Entomol*
40. Juliano, Lounibos (2005) "Ecology of invasive mosquitoes: effects on resident species and on human health" *Ecol Lett*
41. Enserink (2008) "A mosquito goes global" *Science*
42. Gould, Gallian, De Lamballerie et al. (2010) "First cases of autochthonous dengue fever and chikungunya fever in France: from bad dream to reality" *Clin Microbiol Infect*
43. Effler, Pang, Kitsutani et al. (2001) "Hawaii Dengue Outbreak Investigation Team" *Emerg Infect Dis*
44. Gubler (2002) "The global emergence/resurgence of arboviral diseases as public health problems" *Arch Med Res*
45. Who (2016) "Zika-strategic response framework & joint operations plan"
46. Rio, Caldarelli, Gasbarrini et al. (2024) "The impact of climate change on immunity and gut microbiota in the development of disease" *Diseases*
47. (2009) "Global Change Research Program"
48. Patt, Tadross, Nussbaumer et al. (2010) "Estimating least-developed countries' vulnerability to climate-related extreme events over the next 50 years" *Proc Natl Acad Sci*
49. Wilson (2025) "Spillover: from climate change to pandemics. Field actions science reports. The journal of field actions"
50. Dash, Dipankar, Burange et al. (2021) "Climate change: how it impacts the emergence, transmission, resistance and consequences of viral infections in animals and plants" *Crit Rev Microbiol*
51. Carlson, Albery, Merow et al. (2022) "Climate change increases cross-species viral transmission risk" *Nature*
52. Singh, Singh, Mall (2020) "Japanese encephalitis and associated environmental risk factors in eastern Uttar Pradesh: a time series analysis from 2001 to 2016" *Acta Trop*
53. Liu, Zhang, Tong et al. (2020) "Nonlinear and threshold effect of meteorological factors on Japanese encephalitis transmission in southwestern China" *Am J Trop Med Hyg*
54. Tu, Xu, Xu et al. (2021) "Association between meteorological factors and the prevalence dynamics of Japanese encephalitis" *PLoS One*
55. Lin, Chang, Lin et al. (2017) "Seasonal patterns of Japanese encephalitis and associated meteorological factors in Taiwan" *Int J Environ Res Public Health*
56. Li, Zhao, Tian et al. (2005) "Different responses of Japanese encephalitis to weather variables among eight climate subtypes in Gansu" *BMC Infect Dis*
57. Murty, Rao, Arunachalam (2010) "The effects of climatic factors on the distribution and abundance of Japanese encephalitis vectors in Kurnool district of Andhra Pradesh" *India. J Vector Borne Dis*
58. Bi, Zhang, Parton (2007) "Weather variables and Japanese encephalitis in the metropolitan area of Jinan city" *China. J Infect*
59. Gingrich, Nisalak, Latendresse et al. (1992) "Japanese encephalitis virus in Bangkok: factors influencing vector infections in three suburban communities" *J Med Entomol*
60. Konno, Endo, Agatsuma et al. (1966) "Cyclic outbreaks of Japanese encephalitis among pigs and humans 1" *Am J Epidemiol*
61. Hsu, Yen, Chen (2008) "The impact of climate on Japanese encephalitis" *Epidemiol Infect*
62. Bhatia, Bansal, Patil et al. (2022) "A retrospective study of climate change affecting dengue: evidences, challenges and future direction" *Front Public Health*
63. Foote (1960) "Aedes Aegypti (L.), the Yellow Fever Mosquito. Its life history, bionomics, and structure. Sir S. Rickard Christophers"
64. Parums (2023) "Editorial: climate change and the spread of vectorborne diseases, including dengue, malaria, lyme disease, and West Nile virus infection" *Med Sci Monit*
65. Naji (2023) "Dengue fever and global warming: an epidemiological analysis" *Eur J Med Health Sci*
66. Mordecai, Cohen, Evans et al. (2017) "Detecting the impact of temperature on transmission of Zika, dengue, and chikungunya using mechanistic models" *PLoS Negl Trop Dis*
67. Childs, Lyberger, Harris et al. (2024) "Climate warming is expanding dengue burden in the" *Americas and Asia. medRxiv*
68. Sophia, Roxy, Murtugudde et al. (2025) "Dengue dynamics, predictions, and future increase under changing monsoon climate in India" *Sci Rep*
69. Lamy, Tran, Portafaix et al. (2023) "Impact of regional climate change on the mosquito vector Aedes albopictus in a tropical island environment: La Réunion" *Sci Total Environ*
70. Barcellos, Matos, Lana et al. (2024) "Climate change, thermal anomalies, and the recent progression of dengue in Brazil" *Sci Rep*
71. Cornel, Jupp, Blackburn (1993) "Environmental temperature on the vector competence of Culex univittatus (Diptera: Culicidae) for West Nile virus" *J Med Entomol*
72. Kinney, Huang, Whiteman et al. (2006) "Avian virulence and thermostable replication of the North American strain of West Nile virus" *J Gen Virol*
73. Reisen (1995) "Effect of temperature on Culex tarsalis (Diptera: Culicidae) from the Coachella and San Joaquin valleys of California" *J Med Entomol*
74. Henn, Metzger, Kwan et al. (2008) "Development time of Culex mosquitoes in stormwater management structures in California" *J Am Mosq Control Assoc*
75. Jia, Moudy, Dupuis et al. (2007) "Characterization of a small plaque variant of West Nile virus isolated in New York in 2000" *Virology (Auckl)*
76. Paz, Malkinson, Green et al. (2010) *PLoS One*
77. Paz, Albersheim (2008) "Influence of warming tendency on Culex pipiens population abundance and on the probability of West Nile fever outbreaks" *Ecohealth*
78. Paz (2015) "Climate change impacts on West Nile virus transmission in a global context" *Phil Trans R Soc B*
79. Erazo, Grant, Ghisbain et al. (1196) "Contribution of climate change to the spatial expansion of West Nile virus in Europe" *Nat Commun*
80. Paz (2006) "The west nile virus outbreak in Israel (2000) from a new perspective: the regional impact of climate change" *Int J Environ Health Res*
81. Platonov, Fedorova, Karan et al. (2008) "Epidemiology of West Nile infection in Volgograd, Russia, in relation to climate change and mosquito (Diptera: Culicidae) bionomics" *Parasitol Res*
82. Nash, Mostashari, Fine et al. (1999) "The outbreak of West Nile virus infection in the New York City area in 1999" *N Engl J Med*
83. Hayes, Gubler (2006) "West Nile virus: epidemiology and clinical features of an emerging epidemic in the United States" *Annu Rev Med*
84. Petersen, Hayes (2008) "West Nile virus in the Americas" *Med Clin North Am*
85. Lindsey, Staples, Lehman et al. (1999) "Surveillance for human West Nile virus disease -United States" *MMWR Surveill Summ*
86. May, Davis, Tesh et al. (2011) "Phylogeography of West Nile virus: from the cradle of evolution in Africa to Eurasia, Australia, and the Americas"
87. Degroote, Sugumaran, Brend et al. (2008) "Landscape, demographic, entomological, and climatic associations with human disease incidence of West Nile virus in the state of Iowa, USA" *Int J Health Geogr*
88. (2012) "climate-change/climate-change-adaptation/impacts-adaptation-ca nada-changing-climate"
89. Ford (2009) "Climate change and health in Canada" *Mcgill J Med*
90. Shaman, Stieglitz, Stark et al. (2002) "Using a dynamic hydrology model to predict mosquito abundances in flood and swamp water" *Emerg Infect Dis*
91. Paull, Horton, Ashfaq et al. (2017) "Drought and immunity determine the intensity of West Nile virus epidemics and climate change impacts" *Proc R Soc B*
92. Day, Shaman (2008) "Using hydrologic conditions to forecast the risk of focal and epidemic arboviral transmission in peninsular Florida" *J Me Entomol*
93. Keesing, Holt, Ostfeld (2006) "Effects of species diversity on disease risk" *Ecol Lett*
94. Allan, Langerhans, Ryberg et al. (2009) "Ecological correlates of risk and incidence of West Nile virus in the United States" *Oecologia*
95. Levine, Hedeen, Hedeen et al. (2017) "Avian species diversity and transmission of West Nile virus in Atlanta" *Georgia. Parasites Vectors*
96. Cdc (2024)
97. Harrigan, Thomassen, Buermann et al. (2014) "A continen tal risk assessment of West Nile virus under climate change" *Glob Chang Biol*
98. Semenza, Tran, Espinosa et al. (2016) "Climate change projections of West Nile virus infections in Europe: implications for blood safety practices" *Environ Health*
99. Wang, Liu, Gao et al. (2024) "Impact of climate change on the global circulation of West Nile virus and adaptation responses: a scoping review" *Infect Dis Poverty*
100. O'reilly, Lowe, Edmunds et al. (2018) "Projecting the end of the Zika virus epidemic in Latin America: a modelling analysis" *BMC Med*
101. Chan, Choi, Yip et al. (2016) "Zika fever and congenital Zika syndrome: an unexpected emerging arboviral disease" *J Infect*
102. Caminade, Turner, Metelmann et al. (2015) "Global risk model for vector-borne transmission of Zika virus reveals the role of El Niño" *Proc Natl Acad Sci*
103. Carlson, Dougherty, Getz (2016) "An ecological assessment of the pandemic threat of Zika virus" *PLoS Negl Trop Dis*
104. Tesla, Demakovsky, Mordecai et al. (2018) "Temperature drives Zika virus transmission: evidence from empirical and mathematical models" *Proc Biol Sci*
105. Ryan, Carlson, Tesla et al. (2021) "Warming temperatures could expose more than 1.3 billion new people to Zika virus risk by 2050" *Glob Chang Biol*
106. Ali, Gugliemini, Harber et al. (2017) "Environ mental and social change drive the explosive emergence of Zika virus in the Americas" *PLoS Negl Trop Dis*
107. (2024) "India facing largest Chandipura virus outbreak in 20 years" *Lancet*
108. Sudeep, Gunjikar, Ghodke et al. (2019) "Temperature sensitivity and environmental stability of Chandipura virus" *VirusDis*
109. Mallick, Yadav, Gupta et al. (2025) "The evolving landscape of Chandipura virus: a comprehensive account of outbreaks to recent advances" *Virology (Auckl)*
110. Weaver, Forrester, Liu et al. (2021) "Population bottlenecks and founder effects: implications for mosquito-borne arboviral emergence" *Nat Rev Microbiol*
111. Gubler (2001) "Human arbovirus infections worldwide" *Ann N Y Acad Sci*
112. Weaver, Barrett (2004) "Transmission cycles, host range, evolution and emergence of arboviral disease" *Nat Rev Microbiol*
113. Jancarova, Polanska, Volf et al. (2023) "The role of sand flies as vectors of viruses other than phleboviruses" *J Gen Virol*
114. Kemp (1975) "Viruses other than arenaviruses from West African wild mammals" *Bull World Health Organ*
115. Pandey, Singh (2025) "Biological and pathogenic blueprint of Chandipura virus" *Rev Med Virol*
116. Rao, Basu, Wairagkar et al. (2004) "A large outbreak of acute encephalitis with high fatality rate in children in Andhra Pradesh, India, in 2003, associated with Chandipura virus" *Lancet*
117. Fontenille, Traore-Lamizana, Trouillet et al. (1994) "First isolations of arboviruses from phlebotomine sand flies in West Africa" *Am J Trop Med Hyg*
118. Gurav, Tandale, Jadi et al. (2010) "Chandipura virus encephalitis outbreak among children in Nagpur division, Maharashtra, 2007" *Indian J Med Res*
119. Dwibedi, Sabat, Hazra et al. (2015) "Chandipura virus infection causing encephalitis in a tribal population of Odisha in eastern India" *Natl Med J India*
120. Peiris, Dittus, Ratnayake (1993) "Seroepidemiology of dengue and other arboviruses in a natural population of toque macaques (Macaca sinica) at Polonnaruwa, Sri Lanka" *J Med Primatol*
121. Joshi, Patil, Tupe et al. (2005) "Incidence of neutralizing antibodies to Chandipura virus in domestic animals from Karimnagar and Warangal Districts of Andhra Pradesh" *Acta Virol*
122. Hanley, Monath, Weaver et al. (2013) "Fever versus fever: the role of host and vector susceptibility and interspecific competition in shaping the current and future distribu tions of the sylvatic cycles of dengue virus and yellow fever virus" *Infect Genet Evol*
123. Guth, Hanley, Althouse et al. (2020) "Ecological processes underlying the emergence of novel enzootic cycles: arboviruses in the neotropics as a case study" *PLoS Negl Trop Dis*
124. Weaver (2018) "Prediction and prevention of urban arbovirus epidemics: a challenge for the global virology community" *Antiviral Res*
125. Impoinvil, Baylis, Solomon (2012) "In One Health: the human-animalenvironment interfaces in emerging infectious diseases: the concept and examples of a One Health approach"
126. Kanojia, Shetty, Geevarghese (2003) "A long-term study on vector abundance & seasonal prevalence in relation to the occurrence of Japanese encephalitis in Gorakhpur district" *Indian J Med Res*
127. Mani, Rao, Rajendran et al. (1991) "Surveillance for Japanese encephalitis in villages near Madurai" *Trans R Soc Trop Med Hyg*
128. Masuoka, Klein, Kim et al. (2010) "Modeling the distribution of Culex tritaeniorhynchus to predict Japanese encephalitis distribution in the Republic of Korea" *Geospat Health*
129. Johansen, Hall, Van Den Hurk et al. (2002) "Detection and stability of Japanese encephalitis virus RNA and virus viability in dead infected mosquitoes under different storage conditions" *Am J Trop Med Hyg*
130. Nga, Parquet, Cuong et al. (2004) "Shift in Japanese encephalitis virus (JEV) genotype circulating in northern Vietnam: implications for frequent introductions of JEV from Southeast Asia to East Asia" *J Gen Virol*
131. Mcguinness, Lau, Leder (2023) "The evolving Japanese encephalitis situation in Australia and implications for travel medicine" *J Travel Med*
132. Plesner (2003) "Allergic reactions to Japanese encephalitis vaccine" *Immunol Allergy Clin North Am*
133. Flohic, Porphyre, Barbazan et al. (2013) "Review of climate, landscape, and viral genetics as drivers of the Japanese encephalitis virus ecology" *PLoS Negl Trop Dis*
134. Xiao, Boles, Frolking et al. (2006) "Mapping paddy rice agriculture in South and Southeast Asia using multi-temporal MODIS images" *Remote Sens Environ*
135. Rajendran, Reuben, Purushothaman et al. (1995) "Prospects and problems of intermittent irrigation for control of vector breeding in rice fields in southern India" *Ann Trop Med Parasitol*
136. Dwivedi, Mishra, Singh et al. (2025) "Gorakhpur experience of Japanese encephalitis management: a successful One Health approach" *J Vector Borne Dis*
137. Procopio, Colletta, Laratta et al. (2024) "Integrated One Health strategies in dengue" *One Health*
138. Feng, Jiang, Zheng et al. (2024) "Advancing knowledge of One Health in China: lessons for One Health from China's dengue control and prevention programs" *Sci One Health*
139. Wu, Ren, Li (2019) "Neglected urban villages in current vector surveillance system: evidences in Guangzhou, China" *Int J Environ Res Public Health*
140. Sun, Wu, Zhou et al. (2014) "The epidemiological characteristics and genetic diversity of dengue virus during the third largest historical outbreak of dengue in Guangdong" *J Infect*
141. Wang, Liang, Yang et al. (2017) "Co-circulation of 4 dengue virus serotypes among travelers entering china from Myanmar" *Emerg Infect Dis*
142. Wei, Wang, Wei et al. (2021) "Vector competence for DENV-2 among Aedes albopictus (Diptera: Culicidae) populations in China" *Front Cell Infect Microbiol*
143. Hu, Xi, Liu et al. (2020) "Identification and molecular characterization of Wolbachia strains in natural populations of Aedes albopictus in China" *Parasites Vectors*
144. Zhang, Zhan, Wu et al. (2014) "A field survey for Wolbchia and phage WO infections of Aedes albopictus in Guangzhou City" *Parasitol Res*
145. Zhao, Zhou, Chen et al. (2024) "Dengue incidence trends and spatio-temporal distribution characteristics in China" *Chin J Trop Med*
146. Abdalgader, Zheng, Banerjee et al. (2024) "The timeline of overseas imported cases acts as a strong indicator of dengue outbreak in mainland China" *Chaos*
147. De, Lopes, Martins et al. (2021) "The adoption of the One Health approach to improve surveillance of venomous animal injury, vector-borne and zoonotic diseases in Foz do Iguaçu, Brazil" *PLoS Negl Trop Dis*
148. Brandolini, Pascali, Zaghi et al. (2024) "Advancing West Nile virus monitoring through whole genome sequencing: insights from a One Health genomic surveillance study in Romagna (Italy)" *One Health*
149. Gobbo, Chiarello, Sgubin et al. (2022) "Integrated One Health surveillance of West Nile virus and Usutu virus in the Veneto Region, northeastern Italy" *Pathogens*
150. (2001) "Analytical approaches to studying climate/disease linkages"
151. "ecosystems, and infectious disease"
152. Ahumada, Laoointe, Samuel (2004) "Modeling the population dynamics of Culex quinquefasciatus (Diptera: Culicidae), along an elevational gradient in Hawaii" *J Med Entomol*
153. Lima, Cotton, Marais et al. (2024) "Modelling the risk of Japanese encephalitis virus in Victoria, Australia, using an expertsystems approach" *BMC Infect Dis*
154. Wang, He, Zhang et al. (2023) "Japanese encephalitis transmission trends in Gansu, China: a time series predictive model based on spatial dispersion" *One Health*
155. Franklinos, Redding, Lucas et al. (2022) "Joint spatiotemporal modelling reveals seasonally dynamic patterns of Japanese encephalitis vector abundance across India" *PLoS Negl Trop Dis*
156. Tong, Xu, Long et al. (2024) "Modeling the future distribution of Aedes aegypti and Ae. albopictus in China: implications of climate change" *Res Square*
157. Ryan, Carlson, Mordecai et al. (2019) "Global expansion and redistribution of Aedes-borne virus transmission risk with climate change" *PLoS Negl Trop Dis*
158. Nie, Feng (2023) "Niche and range shifts of Aedes aegypti and Ae. albopictus suggest that the latecomer shows a greater invasiveness" *Insects*
159. Farooq, Sjödin, Semenza et al. (2023) "European projections of West Nile virus transmission under climate change scenarios" *One Health*
160. Winters, Eisen, Lozano-Fuentes et al. (2008) "Predictive spatial models for risk of West Nile virus exposure in eastern and western Colorado" *Am J Trop Med Hyg*
161. Lysaght, Lee, Watson et al. (2016) "Zika in Singapore: insights from One Health and social medicine" *Singapore Med J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12172412&blobtype=pdf | # Correction for Ligios et al., "Sheep with Scrapie and Mastitis Transmit Infectious Prions through the Milk"
Ciriaco Ligios, Maria Cancedda, Antonello Carta, Cinzia Santucciu, Caterina Maestrale, Francesca Demontis, Mariangela Saba, Cristiana Patta, James Demartini, Adriano Aguzzi, Christina Sigurdson
V olume 85, no. 2, p. 1136-1139, 2011, https://doi.org/10.1128/jvi.02022-10. Page 1138, Fig. 2: Since publication, we discovered irregularities in the Western blot images in Fig. 2B ("Scrapie" group). The blots contained samples from a comparison sheep group that had ingested milk from mastitis-free, scrapie-infected sheep and were blotted in the Ligios lab. Due to the age of the study, the original sheep samples are no longer available; thus, it is not possible to repeat the experiment. Replicate Western blot experiments were performed at the time of the original study, which upheld the conclusion that there was no PrP Sc in sheep ingesting milk from mastitis-free sheep. The conclusion that there was no PrP Sc in the brain or tonsil of this sheep group is also supported by the negative IHC shown in panel A. The conclusions of the paper are not impacted; sheep with lentiviral mastitis and scrapie secrete prions into milk to infect suckling lambs. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12269117&blobtype=pdf | # Interferon-induced immune signatures are associated with suppression of HEV infection in porcine cell culture models
Sarah Schlienkamp, Olinda Pinto Veiga, André Gömer, Luca Nörthemann, Leyla Sirkinti, Nicola Frericks, Kathrin Sutter, Florian Vondran, Axel Hamprecht, Daniel Todt, Eike Steinmann, Volker Kinast
## Abstract
Hepatitis E virus genotype 3 (HEV-3) is a zoonotic pathogen with pigs representing the natural host. Although HEV-3 infections in humans are often self-limiting, severe or chronic cases can occur. In contrast, HEV-3 infections in pigs, the primary reservoir, remain asymptomatic. To assess the initial transcriptional response in porcine cells during HEV-3 infection and pave the way for mechanistic studies of species-specific virus-host interactions, we aimed to establish porcine cell culture models, including primary porcine hepatocytes (PPHs) and porcine cell lines. PPHs supported the full HEV-3 replication cycle while intrinsic immunity, driven by the interferon-stimulated gene (ISG) system, played a central role in restricting viral replication. JAK inhibition enhanced viral replication and suppressed ISG expression, highlighting the importance of IFN signalling in antiviral defense. Transcriptional profiling revealed a global modulation of host responses upon HEV infection, including pathways linked to immunity, inflammation, and metabolism. Porcine cell lines were permissive to HEV infection and treatment with recombinant porcine IFN-α subtypes induced a robust ISG response and effectively inhibited HEV replication in a dose-dependent manner. These findings establish porcine hepatocytes and cell lines as valuable tools to study HEV-host interactions, demonstrating the critical role of IFN-mediated intrinsic immunity in HEV restriction and highlighting subtype-specific antiviral effects of porcine IFN-α.
## Introduction
Hepatitis E virus (HEV, species Paslahepevirus balayani [1]) infects approximately 20 million humans every year, including 3.3 million acute symptomatic cases as well as 70,000 deaths and is therefore the most common cause of acute viral hepatitis worldwide [2]. HEV is a quasi-enveloped, positive-sense singlestranded RNA virus belonging to the Hepeviridae family. It comprises eight genotypes, of which HEV-1-4 are of major relevance in humans. While HEV-1 and -2 are primarily limited to waterborne infections, HEV-3 and HEV-4 are zoonotic, with their reservoirs in domestic pigs, but also wild boar and deer. HEV-3 is widely spread, being the predominant genotype in Europe, Australia, major parts of North and South America as well as developed parts of Asia [3]. It spills over from reservoir hosts to humans via direct contact to infected animals and consumption of contaminated animal food products [4,3].
Of note, there is a discrepancy in disease progression between humans and pigs, the primary reservoir of HEV relevant to cross-species transmission. In humans, symptoms of acute infection are usually selflimiting, but HEV infections can occasionally result in severe outcomes, including fulminant hepatitis, cirrhosis and liver failure [5,6]. In contrast, there are no reports of symptomatic HEV infections in pigs.
Since species-specific virus-host interactions may explain why HEV infections cause symptoms in humans but not in pigs, it is essential to study host responses to HEV in its primary reservoir. While the host response has been studied in human cell culture models [7], comparable porcine cell culture models are limited [8] and subsequent analyses of the host response upon HEV encounter are lacking.
To expand the experimental systems and address these limitations, we have made efforts to establish protocols enabling the study of HEV in additional porcine cell culture models. Based on these systems, we not only assess the susceptibility of porcine cells to HEV but also provide an initial characterization of the porcine innate immune response to HEV infection. In addition, we make use of a unique set of recombinant porcine interferon-α (pIFN-α) subtypes to characterize the IFN-induced immune response in porcine cell lines and investigate the effect on HEV replication.
## Materials and methods
All materials and methods utilized in this study are described in the Supplementary Information.
## Results
## Intrinsic immunity suppresses HEV-3 replication in primary porcine hepatocytes
To enable the study of HEV-3 infection in an authentic model system of its natural host, we first aimed to establish primary porcine hepatocyte (PPH) cell culture models. Following perfusion, isolation and cultivation, we infected PPHs with cell culture-derived HEV-Kernow-C1/p6 (HEVcc p6 , Figure 1(A)) and performed immunofluorescence microscopy analysis at 3 days post infection (d.p.i.) using an antibody targeting the HEV ORF2 capsid protein (Figure 1(B)). Ribavirin (RBV) served as a control to inhibit viral replication, while a JAK-inhibitor (JAK-I, baricitinib) was applied to examine the impact of innate immune suppression during HEV infection in PPHs. We detected a modest amount of individual ORF2+ cells in untreated HEVinfected cell cultures, suggesting that HEV-3 can replicate in PPHs. We confirmed these results using the HEV-3 wild boar isolate 83-2-27 (HEVcc 83-2 ) to infect PPHs, where we detected HEV ORF2 protein by immunofluorescence microscopy as well as HEV RNA in cell culture supernatants at 3 d.p.i. (Fig. S1). Importantly, both the number of ORF2+ cells and intensity of the ORF2 signal were strongly increased in JAK-I-treated HEV-infected PPHs, suggesting a key role of the IFN-ISG system in controlling HEV infection in PPHs. Consistent with this hypothesis, we observed increased relative light unit (RLU) counts as a surrogate for HEV RNA replication in JAK-I treated PPHs following electroporation with a HEV-Kernow-C1/p6 subgenomic replicon (Fig. S2). In contrast, the lack of an increase in RLU counts upon electroporation of PPHs with a subgenomic replicon of the anthropic HEV-1 Sar55/S17 suggested that PPHs do not support HEV-1 Sar55/S17 replication (Fig. S2). To quantify virus propagation, we collected intracellular HEV-Kernow-C1/p6 progeny virus at 3, 5 and 7 d.p.i. and performed retitration assays on HepG2/C3a cells (Figure 1(C,D)). Progeny virus titres of untreated PPHs reached 2.7 × 10 2 FFU/mL, whereas JAK inhibition resulted in a more than 400-fold increase. In contrast, porcine IFN (pIFN) treatment reduced viral titres. These data mirror the trend observed in the immunofluorescence analysis and emphasize the effectiveness of the porcine IFN-ISG system in controlling HEV-3 infection.
Primary cells provide an excellent model for studying host responses; however, their utility is often limited by their resistance to efficient genetic manipulation, such as lentiviral transduction, which is essential for investigating host and restriction factors. To overcome this challenge, we developed protocols to transduce hepatocytes with lentiviral vectors, using red fluorescent protein (RFP) as a proof of concept. Notably, we successfully transduced PPHs with lentivirus mediating the ectopic expression of porcine IRF7 and thereby inducing the expression of ISGs (Fig. S3). Thereby, we provide a foundation for studying potential HEV host and restriction factors in primary hepatocytes. Together, we established PPHs as a cell culture system for HEV-3 supporting the full viral replication cycle and identified a crucial role for intrinsic immunity in controlling HEV infection.
Next, we aimed to characterize the transcriptomic profile of cultivated PPHs by Illumina sequencing. After mapping of the sequencing reads to a porcine reference genome (GenBank accession no. GCA_000003025.6), we detected high transcript abundances of hepatocyte-specific genes and minimal or absent expression of neuronal-or lung-cell-specific transcripts, confirming the identity of the hepatocyte cultures (Figure 1(E)). Further, we revealed abundant expression of cathepsin L (CTSL), Epidermal Growth Factor Receptor (EGFR), integrin β1 (ITGB1), integrin α3 (ITGA3) and Src-family kinase Yes1 (Yes1) as reported HEV host factors across all analysed donors ([9-13], Figure 1(F)).
## PPHs activate antiviral defenses upon HEV-3 infection
To elucidate the cellular response of PPHs upon HEV infection, we incubated PPHs with conditioned media or infected them with HEVcc p6 , followed by isolation of cellular RNA at 4 or 72 h.p.i. for RNA sequencing. Principal component analysis (PCA, Figure 2(A)) revealed donor-specific differences. At 72 h.p.i., HEV-infected samples formed a distinct cluster, suggesting a global modulation of gene expression induced by HEV infection in PPHs. To verify the presence of HEV RNA in PPHs, we mapped the sequencing reads to the HEV Kernow-C1/p6 reference genome (Figure 2(B), GenBank accession no. JQ679013.1). We detected no HEV RNA in mockinfected cell cultures, while the number of mapped reads increased over time in the HEV-infected cultures in all three donors, indicating active viral RNA replication in PPHs. Together, these data highlight a donor-to-donor variability, yet uniform responses during HEV RNA replication.
We next sought to characterize the modulation of the host transcriptome during HEV infection in more detail by analysing the gene ontology (GO) of differentially expressed genes (DEGs) at 4 and 72 h.p.i. compared to mock-infected cultures (Figure 2(C)). We observed significant changes in GO-terms associated with viral infection, metabolic processes, signalling and inflammation, as well as immune response. Consistent with this, we identified "response to virus" and "cytokine-mediated signaling pathway" by filtering of the most represented GO-terms among DEGs with the highest fold-change at 72 h.p.i. (Figure 2(D)).
Both GO-terms included up-as well as downregulated genes. Interestingly, "ATP synthesis coupled electron transport" was also among the most deregulated GOterms, with only downregulated genes involved. Based on our earlier findings that IFN signalling is crucial to control HEV infection in PPHs, we analysed the induction of type I, II and III IFNs in mock-and HEV-infected PPHs at 72 h.p.i, notably observing increased IFN-β1 and IFN-λ3 mRNA expression (Fig. S4). We aimed to further deepen the analysis of DEGs related to innate immune pathways by focusing on a highly conserved set of interferon-stimulated genes (core-ISGs [14]). Although individual core-ISGs were expressed in a constitutive manner, the upregulation of a majority of core-ISGs at 72 h.p.i. was evident in all three PPH donors (Figure 2(E,F), Fig. S5A). Together, these data indicate a general induction of an IFN response in the host cells upon HEV infection. Notably, we compared ISG induction upon HEV-3 infection in our study to findings from other models, including in vitro, ex vivo and in vivo systems [7,15]. Our analysis showed that several ISGs were upregulated in both PPH and PHH, highlighting a conserved antiviral response across species and experimental settings. Further, all tested ISGs that were upregulated in vivo in pigs were also induced in our PPH system, suggesting a high relevance of our system for studying HEV-3 host interactions (Fig. S6).
## JAK inhibition during HEV infection suppresses upregulation of ISGs
Since we detected a strong increase in HEV progeny virus titres upon JAK inhibition, we hypothesized that JAK-dependent signalling is critical for controlling HEV infection in PPHs. PPH cultures were either treated with JAK-I, infected with HEV or both JAK-Itreated and HEV-infected for 72 h, followed by RNA sequencing analysis. JAK inhibition of mock-infected PPHs led to a broad gene downregulation in the investigated GO-terms including immune response, inflammation and signalling (Figure 3(A)). In HEVinfected cells treated with JAK-I, the previously observed significant enrichment of GO-terms related to viral infection, immune response and inflammation was absent in the three donors. Furthermore, the upregulation of ISGs upon HEV infection was abolished in the presence of JAK-I (Figure 3(B,C), Fig. S5B). Together with our results from Figure 1(C,D), these data support the hypothesis that the IFN response is a central parameter for the fate of HEV infection in PPH cultures.
In summary, we provided a snapshot of HEVinduced transcriptional changes in PPHs and revealed a broad induction of ISGs contributing to the porcine antiviral response. Our data support the hypothesis that a functional IFN response is a central determinant of the outcome of HEV infection in PPH cultures.
## Porcine cell lines are susceptible to HEV infection, support HEV replication and progeny production
Primary hepatocytes are considered the gold standard for studying hepatitis virus infection, as they provide the most authentic cellular background. However, cell lines offer advantages, including phenotypic consistency, cost-effectiveness, time efficiency, and simplified handling. Therefore, we aimed to broaden the scope of porcine HEV research by establishing the option to study HEV infection in porcine cell lines. Given the lack of a porcine hepatoma cell line we selected two kidney-derived cell lines, PK15 and NSK, along with a tracheal cell line, NPTr, for this study.
To assess the susceptibility of these cell lines to HEV infection, we inoculated them with HEVcc p6 for 5 days and quantified infection events via immunofluorescence microscopy, reporting infectivity as FFU/mL (Figure 4(A,B)). Infection levels were comparable across all cell lines, ranging from 1.8 × 10 3 to 3.6 × 10 4 FFU/mL. To investigate the potential role of intrinsic immunity on HEV infection in these porcine cell lines, we treated the cells with JAK-I. JAK inhibition did not significantly alter HEV infection, suggesting that these cell lines do not mount an efficient IFN response upon HEV infection.
To evaluate whether the porcine cell lines can respond to activators of antiviral signalling pathways, we transfected cells with poly(I:C) or infected them with HEV for 24 h and assessed ISG induction by RT-qPCR targeting IFIT1 and MX1 (Fig. S7). While HEV infection did not affect IFIT1 and MX1 transcript levels, poly(I:C) treatment triggered a strong ISG response. These data suggest that the porcine cell lines are capable of responding to innate immune stimuli, although HEV does not appear to trigger an IFN response at the timepoint tested.
To focus on HEV RNA replication in the porcine cell lines, we employed a subgenomic replicon of the HEV Kernow-C1/p6 strain. HepG2 cells served as a positive control for a highly permissive cell line (Fig. S7). We electroporated the cell lines with HEV subgenomic RNA and included RBV as a control. Cell viability was assessed at 72 h.p.e. and remained mostly unchanged when RBV and JAK-I were added (Fig. S8A). Viral replication was quantified at 4, 24, 48 and 72 h.p.e. (Figure 4(C)), revealing an approximately 3 log fold increase across all three cell lines relative to the 4 h.p.e. baseline, thereby confirming the capacity of these cell lines to support HEV RNA replication. Similar to the HEV infection assay, JAK-I treatment did not affect HEV RNA replication rates.
To evaluate the capability of the cell lines to produce infectious HEV progeny particles, we electroporated the cells with full-length HEV Kernow-C1/p6 RNA. JAK-I was added to investigate whether suppression of innate immune pathways influences HEV progeny titres in porcine cell lines. Successful transfection was validated at 3, 5 and 7 days post electroporation (d.p.e.) by immunofluorescence microscopy (Figure 4(D)). Interestingly, only a slight increase of HEV ORF2 signal intensity was apparent in all tested cell lines upon JAK-I treatment. At 3, 5 and 7 d.p.e., we harvested both intracellular and extracellular progeny virus and quantified viral titres by retitration on HepG2/C3a cells (Figure 4(E)). We observed a time-dependent increase in intracellular viral titres reaching up to 3.4 × 10 2 , 4.4 × 10 3 and 7.4 × 10 2 FFU/mL in PK15, NPTr and NSK, respectively. JAK inhibition resulted in viral titres of 1 × 10 3 FFU/mL for PK15, 1.6 × 10 4 FFU/mL for NPTr and 2.3 × 10 3 FFU/mL for NSK. Extracellular titres of progeny HEV were lower compared to intracellular titres with up to 1.2 × 10 2 , 4.5 × 10 2 and 4.4 × 10 2 FFU/mL for PK15, NPTr and NSK, respectively. At 7 d.p.e., JAK-I treatment resulted in a 2.3-fold (2.6 × 10 2 FFU/mL), 2.7-fold (1.2 × 10 3 FFU/mL) and 2-fold (8.7 × 10 2 FFU/mL) increase in enveloped viral titres for PK15, NPTr and NSK, respectively. RT-qPCR targeting HEV RNA showed similar results with a timedependent increase of viral RNA as well as higher rates of HEV RNA upon JAK-I treatment (Fig. S8B). These data confirm that the three cell lines support the complete replication cycle of HEV-3.
Together, our results from porcine cell lines (PK15, NPTr, and NSK) demonstrate their ability to support HEV infection, RNA replication, and progeny virus production. In contrast to PPHs, these cells do not mount a JAK-dependent immune response to potently restrict HEV infection.
## Porcine IFN-α subtypes induce innate immune response and differentially inhibit HEV replication in porcine cell lines
Given that JAK-I treatment did not lead to a strong increase in HEV replication, we next evaluated whether the porcine cell lines could respond to exogenously applied pIFN. Pigs express 17 functional pIFN-α subtypes with distinct antiviral activity and expression profiles [16]. In the current study we tested the subtypes pIFN-α1, pIFN-α7, pIFN-α8 and pIFN-α14. To identify their underlying immune signatures in an HEV-dependent context, we electroporated NSK cells with subgenomic HEV RNA (Kernow-C1/ p6) and exposed them to high and low concentrations of pIFN-α subtypes (1000, 8 and 0.32 ng/mL). The IFN response was assessed by RT-qPCR at 48 h.p.e., targeting the ISGs IFIT1 and MX1. All four pIFN-α subtypes were able to induce the expression of both ISGs in a dose-dependent manner. We observed distinct magnitudes of the immune responses for each pIFN-α subtype, with pIFN-α8 eliciting the strongest ISG induction. While pIFN-α8 triggered an over 100-fold upregulation of IFIT1 at 8 ng/mL, pIFN-α7 only induced a 1.5-fold upregulation of IFIT1 at the same concentration. For pIFN-α1 and pIFN-α14 we observed similar induction of ISG expression, with a modest ISG induction at the lowest concentration, but an up to 100-fold upregulation at 1000 ng/µL (Figure 5(A)).
In order to determine the antiviral potency of the porcine IFN-α subtypes against HEV, we titrated pIFN-α subtypes on PK15, NPTr and NSK cells and assessed their effect on HEV replication. We transfected the cells with a subgenomic HEV replicon (Kernow-C1/ p6) to monitor HEV RNA replication. Cells were exposed to pIFN-α immediately after HEV electroporation and incubated for 48 h. To determine the inhibitory concentration 50 (IC 50 ) of the pIFN-α subtypes, we performed dose-response analyses covering concentrations from 25.6 pg/mL to 5000 ng/mL. Our results demonstrate that treatment with different pIFN-α subtypes potently restricts HEV replication (Figure 5(B)). We observed a differential antiviral pattern for the individual IFN-α subtypes, with pIFN-α7 (IC 50 > 1000 ng/mL) showing the lowest antiviral activity against HEV in all tested cell lines. Subtypes pIFN-α1, pIFN-α8 and pIFN-α14 showed stronger antiviral potency against HEV (IC 50 < 100 ng/mL), except for pIFN-α14 in PK15 cells (IC 50 = 328.2 ng/mL). In conclusion, our experiments with pIFN treatment revealed that the pIFNs with the highest activity of ISG expression were also the most effective in inhibiting HEV replication, highlighting the critical role of ISGs in controlling HEV infection.
## Discussion
Pigs represent the major natural reservoir of HEV-3, and their food products facilitate a continuous spillover of the virus into the human population. HEV remains largely asymptomatic in pigs, while humans potentially experience symptomatic infections with the risk of acute or chronic hepatitis. This contrast in pathogenesis, coupled with frequent spillover events, underscores the importance of studying HEV-host interactions within its natural host.
Since physiologically relevant cell culture models are demanded, we established a novel porcine model based on primary porcine hepatocytes (PPHs) to study the host response to HEV infection in its natural reservoir. We focused our studies on the zoonotic HEV-3 genotype, using the well characterized and cell culture adapted Kernow-C1/p6 strain, yielding high HEV infection efficiency [17,7].
The importance of the innate immune response in the control of HEV infections in primary human hepatocytes (PHHs) was previously described [18,7], but the immune signatures associated with the antiviral response in pigs are poorly understood. To elucidate the porcine host response during HEV infections, we performed transcriptomic analysis of PPHs to determine the transcriptional landscape induced by HEV. Despite donor-specific differences, we observed a global gene modulation upon HEV infection. We were able to identify an upregulation of multiple cellular pathways involved in the host's cell defense response. The strong activation of the type I IFN signalling pathway in all donors demonstrated their ability to sense HEV infections and trigger the IFN/ISG response.
JAK-STAT inhibition attenuated the upregulation of multiple ISGs, concomitant with an increase of HEV replication and progeny virus production, confirming the crucial role of the IFN-ISG axis in PPHs to control HEV infection. This is in line with a recent study by Meyer et al. [15] which provided initial insights on several ISGs involved in the porcine immune response against HEV. However, the importance of individual ISGs in controlling HEV and their mechanism of action remain to be clarified. In this context, a major advancement of our cell culture system is its capacity to genetically manipulate PPHs, enabling the ectopic expression of genes of interest. This will be particularly valuable for studying hostspecific and evolutionary conserved factors (e.g. ISGs) that modulate HEV infection.
In addition to the PPH model, we demonstrated that different porcine cell lines support the complete HEV replication cycle. Despite the lack of a commercially available porcine hepatoma cell line, we presented a new system to study the HEV replication cycle in pigs, offering a tool for comparative studies in addition to the PPHs. Observed differences in the HEV replication efficiency can be accounted to cellspecific variances. In contrast to the PPHs, the tested cell lines did not seem to mount robust IFN responses upon HEV infection, which might be explained by impaired innate immune cascades [19]. Thus, while PPHs are ideal for analysing endogenous responses, we demonstrated that porcine cell lines complement PPHs by providing a valuable tool for studying responses to exogenous IFNs.
The multigenic family of pIFN-α comprises 17 subtypes that share a high sequence identity suggesting functional conservation among them, but several studies show very diverse antiviral activities depending on the IFN-α subtype [20][21][22]. The tested set of pIFN-α subtypes were all able to restrict HEV replication in our experiments. pIFN-α7 showed the lowest effect against HEV replication, which is in accordance with its low antiviral activity against several other viruses, including porcine reproductive and respiratory virus (PRRSV), vesicular stomatitis virus (VSV) and pseudorabies virus (PRV) [20,22,16]. Antiviral activity was significantly higher for the other subtypes, with differences noted between the cell types. Thus, our results confirm the previous standing, that the antiviral effects depend on the tested target cellvirus system [16]. This is in line with the functional antiviral diversity of human IFN-α subtypes, depending on the pathogen and the cellular background [23,24]. Previous studies showed that IFN-α subtypes differ in their affinity to bind to the IFNAR receptor [25], potentially explaining the differential antiviral activities we observed in our study. The positive correlation between ISG induction and antiviral effect in the porcine cell lines supports the role of ISGs as key mediators of HEV restriction. While endogenous IFN signalling seems to be insufficient in the tested porcine cell lines, they responded to exogenously applied pIFN-α, demonstrating their use for the evaluation of different immune modulators.
Previous studies have primarily focused on analysing the immune response in immunocompromised pigs, as HEV infection in porcine models is often studied in a liver transplant setting. These studies demonstrated an overall downregulation of the type I IFN responses during persistent HEV infections [26] and focused on the cellular response upon viral infection [27,28]. Our study depicts for the first time the host transcriptional response with particular emphasis on the innate immune response and thereby sets the foundation for the detailed investigation of the virus-host interplay upon HEV infection in pigs, including molecular mechanisms used by HEV to evade and antagonize innate immunity. The high degree of similarity between humans and pigs in their viral immune response and interferome [14,29,30] demonstrates the importance of our model, offering a unique opportunity to further characterize the antiviral response in human and porcine cells, which could yield valuable insights into the species-specific pathomechanisms and viral adaptation strategies. Of note, despite the substantial levels of similarity of the antiviral response between humans and pigs, they have distinct evolutionary relationships with HEV-3 [30]. Pigs are considered as the host to which HEV-3 is adapted, while humans are the host into which HEV-3 spills over. Therefore, studying virushost interactions in human and porcine cellular contexts is crucial to gain knowledge about the evolution of HEV immune evasion strategies and its ability to cause disease.
To conclude, our findings provide a first global transcriptional overview over the host innate immune response upon HEV infection in primary porcine hepatocytes and provide a novel model to study the HEV replication cycle in its natural host.
## References
1. (1990) "Global burden of disease study 2021 (GBD 2021) burden and strength of evidence by risk factor"
2. Rein, Stevens, Theaker (2012) "The global burden of hepatitis E virus genotypes 1 and 2 in 2005" *Hepatology*
3. Kamar, Izopet, Pavio (2017) "Hepatitis E virus infection" *Nat Rev Dis Primers*
4. Nimgaonkar, Ding, Schwartz (2018) "Hepatitis E virus: advances and challenges" *Nat Rev Gastroenterol Hepatol*
5. Dalton, Bendall, Keane (2009) "Persistent carriage of hepatitis E virus in patients with HIV infection" *N Engl J Med*
6. Wedemeyer, Pischke, Manns (2012) "Pathogenesis and treatment of hepatitis e virus infection" *Gastroenterology*
7. Todt, Friesland, Moeller (2020) "Robust hepatitis E virus infection and transcriptional response in human hepatocytes" *Proc Natl Acad Sci U S A*
8. Rogée, Talbot, Caperna (2013) "New models of hepatitis E virus replication in human and porcine hepatocyte cell lines" *J Gen Virol*
9. Klöhn, Burkard, Janzen (2024) "Targeting cellular cathepsins inhibits hepatitis E virus entry" *Hepatology*
10. Schrader, Burkard, Brüggemann (2023) "EGF receptor modulates HEV entry in human hepatocytes" *Hepatology*
11. Fu, Engels, Weihs (2023) "A high-content RNA-based imaging assay reveals integrin beta 1 as a cofactor for cell entry of non-enveloped hepatitis E virus"
12. Shiota, Li, Nishimura (2019) "Integrin α3 is involved in non-enveloped hepatitis E virus infection" *Virology*
13. Haase, Baheerathan, Zhang (2024) "The tyrosine kinase Yes1 is a druggable host factor of HEV" *Hepatol Commun*
14. Shaw, Hughes, Gu (2017) "Fundamental properties of the mammalian innate immune system revealed by multispecies comparison of type I interferon responses" *PLoS Biol*
15. Meyer, Duquénois, Gellenoncourt (2023) "Identification of interferon-stimulated genes with modulated expression during hepatitis E virus infection in pig liver tissues and human HepaRG cells" *Front Immunol*
16. Sang, Rowland, Hesse (2010) "Differential expression and activity of the porcine type I interferon family" *Physiol Genomics*
17. Shukla, Nguyen, Faulk (2012) "Adaptation of a genotype 3 hepatitis E virus to efficient growth in cell culture depends on an inserted human gene segment acquired by recombination" *J Virol*
18. Kinast, Andreica, Ahrenstorf (2023) "Janus kinaseinhibition modulates hepatitis E virus infection" *Antiviral Res*
19. Hare, Collins, Cuddington (2016) "The importance of physiologically relevant cell lines for studying virus-host interactions" *Viruses*
20. Fang, Zhang, Xi (2023) "Analysis of the differential expression and antiviral activity of porcine interferonα in vitro" *Int J Pept Res Ther*
21. Sang, Bergkamp, Blecha (2014) "Molecular evolution of the porcine type I interferon family: subtype-specific expression and antiviral activity" *PLoS One*
22. Zanotti, Razzuoli, Crooke (2015) "Differential biological activities of swine interferon-α subtypes" *J Interferon Cytokine Res*
23. Moll, Maier, Zommer (2011) "The differential activity of interferon-α subtypes is consistent among distinct target genes and cell types" *Cytokine*
24. Schuhenn, Meister, Todt (2022) "Differential interferon-α subtype induced immune signatures are associated with suppression of SARS-CoV-2 infection" *Proc Natl Acad Sci U S A*
25. Schreiber (2017) "The molecular basis for differential type I interferon signaling" *J Biol Chem*
26. León-Janampa, Caballero-Posadas, Barc (2023) "A pig model of chronic hepatitis E displaying persistent viremia and a downregulation of innate immune responses in the liver" *Hepatol Commun*
27. Cao, Cao, Subramaniam (2017) "Pig model mimicking chronic hepatitis E virus infection in immunocompromised patients to assess immune correlates during chronicity" *Proc Natl Acad Sci U S A*
28. Krishna, Kim, Yang (2020) "Immune responses to porcine epidemic diarrhea virus (PEDV) in swine and protection against subsequent infection" *PLoS One*
29. Dawson, Smith, Chen (2017) "An in-depth comparison of the porcine, murine and human inflammasomes; lessons from the porcine genome and transcriptome" *Vet Microbiol*
30. Starbaek, Brogaard, Dawson (2018) "Animal models for influenza A virus infection incorporating the involvement of innate host defenses: enhanced translational value of the porcine model" *ILAR J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548416&blobtype=pdf | # GCN2 enhances host survival and drives eIF2α phosphorylation during mouse adenovirus type 1 infection
Luiza Castro, Daniel Edwards, Rosario Labastida, Danielle Goodman, Estela Pereira, Oded Foreman, Katherine Spindler
## Abstract
The integrated stress response (ISR) is a cellular signaling pathway that reduces protein synthesis in the face of cellular stress, including viral infection. Two eukaryotic initiation factor 2α (eIF2α) kinases, protein kinase R (PKR) and general control nonderepressible 2 (GCN2), are commonly activated during viral infections. Mouse adenovirus type 1 (MAV-1) infection leads to a steep reduction of PKR levels by proteasomal degradation. We assayed whether GCN2, a sensor of amino acid starvation and UV damage, plays a role in the ISR to MAV-1 infection. There was more phos phorylated GCN2 in MAV-1-infected cells, and its activation was dependent on virus replication since UV-inactivated virus was not able to increase the phosphorylation of GCN2. Infected Eif2ak4 tm1.2Dron mice (designated here Gcn2 -/-mice) had lower survival than wild-type (WT) mice, but results indicated that this was not due to increased viral replication. Both Gcn2 -/-and WT mice developed multifocal brain parenchymal microhemorrhages during infection. While Gcn2 -/-animals had more lesions, their higher mortality is likely not due to the microhemorrhages alone. Cytokine RNA and protein assays of WT and Gcn2 -/-mice only showed a difference for IL-β levels, which were higher in Gcn2 -/-mice. Our results also indicate that of the two eIF2α kinases, PKR and GCN2, GCN2 is the primary inducer of phosphorylated-eIF2α during MAV-1 infection. GCN2 is thus antiviral and contributes to the host response to MAV-1 infection. IMPORTANCE Cells often respond to viral infection by activation of the host protein kinase R (PKR), part of the integrated stress response (ISR). We show that a second host protein kinase, general control nonderepressible 2 (GCN2), is activated by phosphoryla tion in response to mouse adenovirus type 1 (MAV-1) infection. Our results indicate GCN2 is antiviral: without it, the mortality in MAV-1-infected mouse is higher. Furthermore, the data show that GCN2, rather than PKR, is the main inducer of eIf2α phosphorylation (and thus the ISR) upon MAV-1 infection. This is consistent with PKR exerting antiviral effects in MAV-1 infections through a pathway independent of eIf2α phosphorylation. KEYWORDS integrated stress response, PKR, protein kinase R, general control nonderepressible 2V irus infection relies on the host translation machinery to produce viral proteins necessary for virus replication. Therefore, the reduction of protein synthesis is an effective mechanism employed by the host to resist viral infection. Host cells have several ways to reduce protein synthesis. One involves the integrated stress response (ISR), which is an intricate signaling pathway that can be induced by four different eIF2α kinases: heme-regulated inhibitor of translation (HRI and EIF2AK1), protein kinase R (PKR and EIF2AK2), PKR-like endoplasmic reticulum kinase (PEK/PERK and EIF2AK3), and general control nonderepressible 2 (GCN2 and EIF2AK4) (1, 2). Each of these eIF2α-kina ses responds to distinct environmental stresses. HRI monitors changes in hemoglobin
levels (3); PKR senses dsRNA during viral infection (4); PERK is activated by endoplasmic reticulum (ER) stress (5); and GCN2 detects amino acid starvation, UV damage, and viral infection (6,7). However, they all converge on eIF2α phosphorylation, which broadly reduces translation but allows selective translation of genes with upstream ORFs in their 5′ untranslated region (8).
Many viruses have mechanisms that counteract the action of eIF2α kinases, and PKR has been one of the most studied eIF2α kinases in this context. Some viruses block PKR activation by encoding proteins that sequester dsRNA, blocking its interaction with PKR, such as influenza virus NS1 (9), vaccinia virus E3L (10)(11)(12), and Ebola virus VP35 (13). Other viruses encode proteins or RNAs that bind directly to PKR, inhibiting its activation, such as HIV-1 Tat protein (14), herpes simplex virus US11 (15,16), and human adenovirus (HAdV) VA RNAs (17). Another mechanism is PKR degradation, employed by poliovirus (18), Rift Valley fever virus (19,20), Toscana virus (21), foot-and-mouth-disease virus (22), and enterovirus A71 (23). We reported that mouse adenovirus type 1 (MAV-1; also known as MAdV-1) can also degrade PKR, being the first DNA virus identified that does so (24,25). Recently, fowl adenovirus was identified to also degrade PKR (26). Together, these results indicate that ISR signaling is important during adenovirus infection.
MAV-1 is in the Adenoviridae family, and it causes both acute and persistent infection in mice, leading to a dose-dependent encephalitis in susceptible mice (27,28). MAV-1 enables the investigation of adenovirus pathogenesis in a small animal host; the study of HAdV pathogenesis in an animal model is difficult because adenoviruses are species-spe cific (29). However, MAV-1 pathogenicity and tropism differ from known HAdVs: MAV-1 infects endothelial cells and monocytes, causing encephalitis and myocarditis, while HAdVs infect epithelial cells, leading to upper respiratory and GI tract infections and conjunctivitis (27,30,31). We and others have been investigating innate and adaptive immune responses to MAV-1. The ISR likely plays an important role in MAV-1 pathogene sis not only through PKR (24,25,32) but also through GCN2. A loss-of-function mutation in Eif2ak4, encoding GCN2, the Atchoum (Atc) mutation, led to an increased susceptibility to infection in peritoneal macrophages by human adenovirus (33).
GCN2's role in viral infections is less well appreciated or understood than that of PKR. However, several studies have shown that GCN2 also inhibits viral infections (34). Sindbis virus infection of mouse embryonic fibroblasts (MEFs) activates GCN2, and infection of Gcn2 -/-mice results in more viral replication than in control mice (35). Similarly, Atc mice show lower survival rates when infected with mouse cytomegalovirus (MCMV) compared with wild-type (WT) mice (33). HIV-1 RNA increases GCN2 kinase activity in cell-free extracts and transfected cells; however, HIV-1 infection of a human T cell line leads to cleavage of GCN2 by HIV-1 protease, showing that HIV-1 not only induces GCN2, but it also antagonizes its function (36). GCN2 silencing by siRNA increases HIV-1 infectivity, concomitant with an increase in new protein synthesis (37), consistent with GCN2 being antiviral through inhibition of translation. During viral infection, GCN2 may be activated by two main mechanisms: amino acid depletion with accumulation of uncharged tRNAs or ribosome collisions stimulated by ribosome stalling (34,38). Infection of human dendritic cells by live attenuated yellow fever vaccine virus (YF-17D) leads to a reduction in arginine and an increased phosphorylation of GCN2 (39). Another flavivirus, Zika virus (ZIKV), also leads to GCN2 phosphorylation; in ZIKV infections, GCN2 activation is dependent on viperin-induced translation inhibition triggered by colliding ribosomes (40). GCN2 activation has a broad impact on cell survival and on immune responses during infection, regulating a fine-tuned balance between autophagy, apoptosis, and cell cycle progression (41)(42)(43).
Because MAV-1 infection leads to a reduction of the ISR PKR protein kinase levels by proteasomal degradation (24,25), we hypothesized that GCN2 also plays a role in MAV-1 pathogenesis. We found that GCN2 is more highly phosphorylated in MAV-1-infected cells and GCN2 activation is dependent on viral replication. Gcn2 -/-mice infected with MAV-1 had lower survival than wild-type mice, indicating that GCN2 has an antiviral role during MAV-1 infection. However, infection of the Gcn2 -/-mice did not lead to higher viral loads in brain and spleen than WT mice. This suggests that the difference in survival between Gcn2 -/-and WT mice was not due to increased viral replication in Gcn2 -/-mice but rather a more complicated interplay between viral infection and the host response. Histopathologic findings of multifocal parenchymal microhemorrhages were seen in both Gcn2 -/-and WT mice, and they were more abundant in the Gcn2 -/- animals. However, the microhemorrhages alone could not explain the higher mortality of Gcn2 -/-mice. Therefore, we evaluated cytokine levels by both RNA and protein assays. Of all the cytokines analyzed, only IL-1β levels differed between the strains. Because GCN2 signaling can also regulate the balance between autophagy induction and inflammasome activation (41), we investigated the levels of IL-1β production in vitro and observed higher levels in Gcn2 -/-cells compared to WT. Our results also suggest that GCN2 (and not PKR) is the primary inducer of phosphorylated-eIF2α (peIF2α) during MAV-1 infection.
## RESULTS
## MAV-1 infection leads to GCN2 phosphorylation in cultured mouse embryo fibroblasts (MEFs)
To evaluate whether MAV-1 infection induces phosphorylation of GCN2, we infected C57BL/6J MEFs with MAV-1 at an MOI of 5 and harvested cell lysates at 6, 12, 24, and 48 hours post-infection (hpi). As a positive control for GCN2 activation by phosphorylation, we UV-treated the cells with 50,000 μJ/cm 2 and allowed 30 minutes for recovery. We used a detection method to optimize visualizing phosphorylated GCN2 (pGCN2) and GCN2 levels that involves treatment with hydrogen peroxide after the first (pGCN2) antibody probing (instead of harsh stripping [44,45]). Due to steric hindrance of the anti-pGCN2 antibody persisting after the hydrogen peroxide treatment, it is likely that only nonphos phorylated GCN2 will be detected by the anti-GCN2 antibody. We assayed pGCN2 levels by first immunoblotting with an antibody specific for pGCN2. The levels of pGCN2 were higher at 6 hpi compared to uninfected cells and remained elevated up to 48 hours (Fig. 1A). At the same time, the levels of nonphosphorylated GCN2, detected by an antibody that should recognize both pGCN2 and GCN2, were lower in infected cells compared to uninfected cells. We used E1A protein detection to confirm viral infection. Very low levels of E1A were detected at 6 hpi and increased as infection progressed, reaching a very high level at 48 hours (Fig. 1A). Using UV-inactivated virus, we also evaluated whether the virus particles could activate GCN2 or whether the virus needed to be actively replicating for GCN2 to become phosphorylated. We infected C57BL/6J MEFs with MAV-1 MOI of 5 or treated these cells with an equivalent amount of UV-inactivated virus (Fig. 1B). Only inoculation with the non-UV-treated MAV-1 resulted in phosphorylated GCN2 (lane 3); the levels of GCN2 phosphorylation were similar between cells treated with the UV-inactivated virus and uninfected cells (lanes 5 and 4, respectively). These results demonstrate that MAV-1 infection leads to GCN2 phosphorylation, and GCN2 phosphor ylation is dependent on virus replication.
## Absence of GCN2 results in increased MAV-1 yield in cultured cells
To determine whether the GCN2 (Atc) mutation leads to an increased yield of virus following infection with MAV-1, we infected peritoneal macrophages isolated from Atchoum (Atc) mice and WT (C57BL/6J) mice. We infected cells at an MOI of 1 and harvested cells and supernatants at 24, 48, and 72 hpi for DNA extraction. MAV-1 DNA levels were evaluated by qPCR. No appreciable difference in DNA levels was detected between peritoneal macrophages of the two strains at 24 hpi (Fig. 2A). However, at both 48 and 72 hpi, the viral DNA yields from Atc macrophages were higher than those from WT macrophages. This suggests that GCN2 is antiviral during MAV-1 infection.
We evaluated GCN2 phosphorylation levels in Atc MEFs infected with MAV-1 at MOI of 5 (Fig. 2B, lanes 7-9). We observed a faint pGCN2 band in the immunoblots from cellular extracts (lane 8) and GCN2 bands in lanes 7-9. The Atc mutation corresponds to a thymine-to-cytosine transition of the sixth nucleotide of intron 2 of Eif2ak4, encoding GCN2, leading to exon 2 skipping in peritoneal macrophages, and in some cases skipping of exons 3 and 4 (33). This was reported to produce no GCN2 in peritoneal macrophages. However, in addition to the Atc MEF results here, we detected the expression of GCN2 protein from additional Atc cell types (data not shown). Mice with a different GCN2 mutation, Eif2ak4 tm1.2Dron mice (46), referred to here as Gcn2 -/-mice, have a deletion of exon 12 of the Eif2ak4 (GCN2) gene. MEFs from the Gcn2 -/-mice did not have GCN2 or pGCN2 protein bands (Fig. 2B, lanes 4-6). Because the mutation of GCN2 appeared to be leaky in Atc cells, we continued our subsequent experiments with the Eif2ak4 tm1.2Dron (Gcn2 -/-) mice (46). In both MEFs and bone marrow-derived macrophages (BMDMs) isolated from Gcn2 -/-mice, MAV-1 replicated to higher levels than in WT cells both at 48 and 72 hpi, as determined by plaque assay (Fig. 2C). These results support the earlier data with Atc mice suggesting that GCN2 has an antiviral role during infection because the virus replicates to higher levels in cells lacking GCN2.
## Mice deficient in GCN2 production have lower survival than WT mice after sublethal MAV-1 challenge but no difference in viral load or blood-brain barrier disruption
The role of GCN2 during MAV1 viral encephalitis is unknown. To characterize the in vivo physiological role of GCN2 in protection from MAV-1 infection, we compared survival of WT mice and Gcn2 -/-mice after intraperitoneal (i.p.) infection with 10 2 PFU of MAV-1.
There was a statistically significant difference in survival: ~80% of WT mice survived compared to ~55% of the mutant mice (Fig. 3A). For subsequent experiments, we used a dose of 10 2 PFU/mouse and assayed at 8 days post-infection (dpi) (when significant mortality was observed) to examine parameters that might differ between WT and Gcn2 -/-mice.
The highest levels of MAV-1 in infections are found in brains and spleens (27,(47)(48)(49). We quantitated MAV-1 virus present in the brains and spleens of WT and Gcn2 -/-mice 8 dpi with 10 2 PFU MAV1 by measuring viral DNA levels by qPCR. We did not observe any difference in viral DNA levels between WT and GCN2 mouse brains (Fig. 3B) or spleens (data not shown). Additionally, we evaluated infectious virus levels by plaque assay from randomly selected samples. There was a good correlation between plaque assay titers and MAV-1 genome copies measured by qPCR, and there was also no difference in virus titers between WT and Gcn2 -/-mice (data not shown). We also measured MAV-1 viral DNA levels in brains of Atc and WT mice at 3, 5, and 7 dpi and did not observe any differences between the strains (data not shown).
MAV-1 infection in mice leads to the disruption of the blood-brain barrier (BBB) (50). To determine whether the difference in mouse susceptibility seen in Fig. 3A correlated with any difference in BBB disruption, we assayed BBB permeability to sodium fluorescein. Sodium fluorescein is a small molecule (376 Da) that can only access and stain brain tissue when the BBB is compromised. We administered sodium fluorescein i.p. at 8 dpi to WT and Gcn2 -/-mice, and we quantitated the sodium fluorescein present in the right brain hemispheres. There was no statistical difference in sodium fluorescein uptake in brains from WT and Gcn2 -/-mice (Fig. 3C). Although GCN2 plays a role in the survival of mice infected with MAV-1, it does not affect the viral replication levels nor the bloodbrain barrier disruption in these mice.
MAV-1 pathogenesis in WT and Gcn2 -/-mice One possible explanation for the higher mortality in Gcn2 -/-mice compared to WT mice without accompanying higher viral load or higher BBB disruption is that the absence of GCN2 could lead to an altered immune response that contributes to the disease severity. We analyzed the levels of chemokines and cytokines in brains of WT and Gcn2 -/-mice infected with 10 2 PFU of MAV-1 and harvested at 8 dpi. We evaluated RNA levels by qPCR and protein levels by ELISA. We observed a large difference between the levels of IL-6, TNFα, IL-10, CCL5, and CXCL10 in mock and infected mice for both strains; however, there was no difference between the strains (Fig. 4A through L). In contrast, IL-1β levels were higher in Gcn2 -/-mouse brains compared to WT mouse brains, as measured by both qPCR and ELISA (Fig. 4M andN). Because GCN2 signaling can regulate the balance between inflammasome activation and autophagy induction (41), we investigated IL-1β production in BMDMs isolated from WT and Gcn2 -/-mice and observed higher levels of IL-1β in Gcn2 -/-BMDMs (Fig. 4O).
To better understand the difference in mouse survival and the higher IL-1β present in the brains of Gcn2 -/-mice, we evaluated the histopathological findings in WT and Gcn2 -/- mice infected with 10 2 MAV-1 PFU. Microscopic findings at 8 dpi in the brains of MAV-1infected mice consisted of multifocal encephalitis and meningitis with vasculitis and perivascular edema (Fig. 5A through D). Lesions contained infiltrates of neutrophils with few macrophages. The severity and distribution of the lesions were quantified by blinded scoring of histological sections of brains (Fig. 5E). Although Gcn2 -/-mice had worse histopathological scores than WT mice, the overall pathology in both strains was mild, suggesting mortality was unlikely due to these lesions alone. No differences were seen between the strains for other organs examined (thymus, lung, heart, brain, liver, kidney, and spleen). While both strains showed mild histopathological changes and similar cytokine profiles for most markers, the distinct increase in IL-1β in Gcn2 -/-mice suggests a unique inflammatory response contributing to their increased disease severity.
## eIF2α phosphorylation during MAV-1 infection is predominantly due to GCN2 activation rather than PKR activation
To examine whether eIF2α is responsible for the increased susceptibility of GCN2 -/-mice to MAV-1 infection when compared to WT, we used cells that do not express PKR, PKR-TKO MEFs (32). We evaluated eIF2α phosphorylation levels in WT, Gcn2 -/-and PKR-TKO MEFs infected with MAV-1 at an MOI of 5 (Fig. 6A) and found little to no eIF2α phosphor ylation in cells lacking GCN2. We observed only a faint peIF2α band in immunoblots from UV-stimulated positive control Gcn2 -/-cellular extracts (lane 4) and almost no band in MAV-1-infected Gcn2 -/-MEFs (lane 6), confirmed by peIF2α densitometry analysis (Fig. 6B). In contrast, WT and PKR-TKO MEFs exhibited peIF2α bands in positive controls (lanes 1 and 7) and MAV-1-infected MEFs (lanes 3 and 9) (Fig. 6), with higher peIF2α levels in infected cells compared to mock. We have shown that MAV-1 very effectively degrades PKR, such that there is almost no PKR detectable during infection (24). However, eIF2α phosphorylation still occurred at high levels in the absence of PKR, but not in the absence of GCN2 (Fig. 6). Because the lack of GCN2 nearly eliminated eIF2α phosphoryla tion in MAV-1-infected MEFs, whereas the lack of PKR did not alter eIF2α phosphorylation levels, these data suggest that GCN2 is more responsible for eIF2α phosphorylation than PKR during MAV-1 infection.
## DISCUSSION
We sought to determine the role of GCN2 during MAV-1 infection. We evaluated whether MAV-1 activates GCN2 and the effects of this antiviral pathway during infection. We demonstrated that MAV-1 induces the phosphorylation of GCN2 (Fig. 1A andB), which has been implicated in the pathogenesis of many viruses (33)(34)(35)(36). GCN2 is antiviral in MAV-1 infections of cultured cells, and lack of GCN2 leads to a decrease in mouse survival during MAV-1 infection (Fig. 2 and3). This higher mortality was not due to higher viral replication or an increase of BBB disruption in infected mouse brains. Instead, we showed that GCN2 deficiency leads to higher IL-1β levels in infected mouse brains (Fig. 4), suggesting that there could be an imbalance of autophagy and inflammasome activa tion upon virus infection, as described for other virus infections (39). We also showed that GCN2 is the primary inducer of eIF2α phosphorylation during MAV-1 infection and that eIF2α phosphorylation occurs even in the absence of PKR (Fig. 6).
GCN2 phosphorylation can be induced by amino acid starvation and a decrease in uncharged tRNAs or by ribosome stalling (38,40). We have shown that during MAV-1 infection, GCN2 is phosphorylated, and viral replication is required for its activation. Exposing MAV-1 to UV-irradiation renders it biologically inactive (confirmed by plaque assay of UV-treated MAV-1 and the absence of E1A gene expression in Fig. 1B, lane 5). UVinactivated virus was unable to induce phosphorylation of GCN2 (Fig. 1B). The same is observed for yellow fever virus in human dendritic cells, in which only actively replicating virus causes a depletion in amino acid pools, specifically in arginine levels (39). Importantly, we showed that GCN2 signaling has an antiviral role during MAV-1 infection. Gcn2 -/--infected MEFs and BMDMs produced more infectious virus than did WT cells (Fig. 2C). This correlates with the fact that GCN2 inhibits virus replication of HIV-1 and VSV (37,51). In addition, overexpression of GCN2 in a MEF cell line severely reduced Sindbis virus (SINV) replication (35).
When investigating the impact of GCN2 activation during MAV-1 infection in vivo, we observed that GCN2 also played a protective role in mice. There was an increase in mortality of almost 30% after MAV-1 infection in the absence of GCN2. Similarly, mice infected with MCMV show around 20% higher mortality when they lack functional GCN2, compared to mice with functional GCN2 (33). However, there were no differences in MAV-1 viral replication in the brains or spleens of WT and Gcn2 -/-mice. While other organs were not examined for viral loads, we did not see differences in histopathology in organs other than the brain. During SINV infection, GCN2 is important for viral defense during the early stages of infection (35). At 3-4 dpi, Gcn2 -/-mice have significantly higher SINV titers in their brains compared to WT animals, while at 5 dpi, there were no differences in viral titers in infected Gcn2 -/-and wild-type mouse brains (35). We measured MAV-1 viral loads in Atc and wild-type mice brains at 3-and 5-days post infection and did not observe any differences between the strains (data not shown).
We hypothesized that the lack of GCN2 could elicit a different inflammatory response after MAV-1 infection, leading to the higher mortality rate observed in Gcn2 -/-mice. Examination of histopathology induced by MAV-1 infection only revealed differences in brains and not other organs. Brains from both WT and Gcn2 -/-mice indicated the presence of multifocal encephalitis and meningitis with vasculitis and perivascular edema. However, even though GCN2 -/-mice had worse histopathological scores than WT mice overall, the lesions were generally mild, and therefore mortality seems unlikely to be due to these lesions alone. We also evaluated the brain uptake of sodium fluorescein after MAV-1 infection in WT and Gcn2 -/-mice since it can be used to detect BBB disrup tion and presence of microhemorrhages. We did not observe any difference in brain sodium fluorescein uptake between WT and Gcn2 -/-mice infected with MAV-1 (Fig. 3C). This contrasts with our observation of slightly more brain pathology in the Gcn2 -/-mice. However, a caveat is that we evaluated global fluorescein uptake by homogenizing the whole brain and assaying fluorescence. Detecting sodium fluorescein uptake by imaging would be more sensitive to detect small, localized changes since there can be a variation of uptake levels depending on the brain area (52)(53)(54).
GCN2 is involved in the development of the remission phase of experimental autoimmune encephalomyelitis in C57BL/6 (WT) mice (55). When WT and GCN2 KO mice were immunized with myelin oligodendrocyte glycoprotein peptide, GCN2 KO mice did not develop the remission phase of the disease, and this was associated with higher levels of CNS inflammation and increased presence of effector T cells (Th1/Th17). When evaluating histological sections of the lumbar spinal cord, the presence of the inflammatory cells was more evident in GCN2 KO mice, especially at the remission phase of the disease at day 21, compared to the peak phase at day 15 post-immunization. Upon examination of the cytokines involved in MAV-1 pathogenesis, both by qPCR and ELISA in mouse brains, we only noted altered levels of one: we detected elevated levels of only IL-1β in brains of Gcn2 -/-mice compared to WT mice after MAV-1 infection. The small difference in the cytokines and histopathological alterations in these mice indicate that there are other factors affecting mouse survival that should be explored in this system.
GCN2 is critical to the maintenance of cellular and organismal homeostasis and plays multiple roles in immune cell functions during viral infection (34,43). Virus-induced GCN2 activation has a key role in programming dendritic cells to initiate autophagy and enhance antigen presentation to both CD4 and CD8 T cells in the context of yellow fever 17D vaccine (39). The yeast eIF2α kinase GCN2 and the eIF2α-regulated transcriptional transactivator GCN4 are essential for starvation-induced autophagy, and notably, PKR can rescue GCN2-disrupted yeast from starvation-induced autophagy (56). Divergent stress stimuli such as nutrient deprivation and herpes simplex virus infection stimulate eIF2α kinase-dependent translational arrest, and they also stimulate eIF2α kinase-dependent autophagy. GCN2 is also involved in controlling intestinal inflammation by suppressing inflammasome activation in a mouse model of acute colitis (41). Genetic deletion of GCN2 in CD11c + antigen-presenting cells or intestinal epithelial cells results in enhanced intestinal inflammation and T helper 17 cell (TH17) responses and a reduction in autophagy, leading to increased reactive oxygen species (ROS), with enhanced inflammasome activation and IL-1β production (41). During MAV-1 infection, GCN2 may help balance inflammasome activation and autophagy since IL-1β levels were higher by both qPCR and ELISA in Gcn2 -/-mouse brains compared to WT mouse brains (Fig. 4M andN). Further investigation is necessary to further determine whether the higher mortality observed in the absence of GCN2 is connected to a decrease in autophagy and increase in inflammasome activation.
We determined that the presence of GCN2 in MEFs was required to increase eIF2α phosphorylation during MAV-1 infection (Fig. 6, lanes 5 and 6). In contrast, the presence of PKR did not affect eIF2α phosphorylation (Fig. 6, lanes 8 and 9). This indicates that of these two eIF2α-kinases, GCN2 is the primary one phosphorylating eIF2α during MAV-1 infection. Similar findings are observed when MEFs are infected with murine norovi rus, and virus-induced eIF2α phosphorylation is impaired in GCN2 -/-MEFs compared to WT (57). The extent of eIF2α phosphorylation will influence the degree to which global protein synthesis is reduced (58). Strong eIF2α phosphorylation can halt global translation, while weak eIF2α phosphorylation may not affect all translation. Considering the effects of a viral infection on cellular homeostasis from the cells' standpoint, the impairment of protein synthesis would be desired to prevent or reduce viral replica tion. We hypothesize that during MAV-1 infection, GCN2 is the primary eIF2α-kinase phosphorylating eIF2α, while PKR does not affect eIF2α phosphorylation levels. We have demonstrated that there is substantial viral degradation of PKR induced by MAV-1 infection (24), and we show here that a lack of PKR does not affect total eIF2α phos phorylation. This indicates that other functions of PKR are important during infection (32) and that GCN2 is the main eIF2α-kinase involved in ISR activation in the context of MAV-1 infection. It will be necessary to study this mechanism further to determine whether GCN2 activation leads to an alteration in the balance between inflammasome and autophagy.
Primary peritoneal macrophages were isolated from 6-to-10-week-old C57BL/6J mice, as described previously (61). Primary BMDMs were prepared by flushing DMEM into mouse tibia and femurs of male or female mice aged between 8 and 12 weeks. Cells were differentiated by incubation with DMEM supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, 30% L929 cell-conditioned medium, 20% heat-inactivated FBS, 1 mM penicillin-streptomycin (GIBCO 15140-122), and 2 mM non-essential amino acids (NEAA, GIBCO 11140-050). After 7 days in culture, BMDMs were harvested and seeded at the required density for each experiment. L-929 cells were cultured in DMEM supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, 1 mM NEAA, 10 mM HEPES, and 10% heat-inactivated FBS. All cells were incubated at 37°C in 5% CO 2 .
## Mice
Wild-type C57BL/6J (cat. no. 000664) and B6.129S6-Eif2ak4 tm1.2Dron /J (cat. no. 008240, Gcn2 -/-) were purchased from Jackson Laboratory (Bar Harbor, ME). The Gcn2 -/-mice have a deletion of exon 12 of the Eif2ak4 (GCN2) gene (46). Eif2ak4 m1Btlr (Atc) mice on the C57BL/6 background obtained from Bruce Beutler (UT Southwestern Medical Center) have an ENU-induced T to C transition at position 12,038 (GenBank NC_000068); the mutation affects the donor splice site of intron 2. The Atc, Gcn2 -/-, and C57BL/6J mice were bred in-house, and both sexes were used in experiments. No differences based on sex in any assayed phenotypes were noted. Atc mutations were confirmed by genotyping mice as described (33). Gcn2 -/-mutations were confirmed by Transnetyx (Memphis, TN) based on the reported genotype (46). All animals were housed in specific pathogen-free facilities at the University of Michigan Medical School Unit for Laboratory Animal Medicine (ULAM). Animals were housed in microisolator cages and provided food and water ad libitum, and health checks were performed daily. Male and female mice were infected i.p. with the indicated virus dose diluted in endotoxin-free PBS in 0.1 mL, between the ages of 4 to 5 weeks. Mock-infected mice were infected with conditioned media prepared in parallel with the virus stock by collection of media from uninfected cells. Infected mice were housed in biosafety level 2 containment and treated in accordance with an IACUC-approved protocol. All animal work complied with relevant federal and University of Michigan policies.
## Quantitation of virus titers
MAV-1 titers in cells were determined either by qPCR or by plaque assay. When comparing Atc and WT peritoneal macrophages, cells were infected with MAV-1 at an MOI of 1. After 24, 48, and 72 hpi, cells were washed twice with room temperature PBS and harvested by scraping into PBS, centrifuging at 100 × g for 4 min at 4°C, and resuspending in PBS. Total cellular DNA was purified using an Invitrogen PureLink DNA purification kit (Thermo Scientific catalog no. K1820-02) and quantitated using a NanoDrop spectrophotometer. Total cellular DNA (10 ng) was analyzed by qPCR using custom primers specific to MAV-1 E1A (mE1Agenomic Fwd [5′ GCA CTC CAT GGC AGG ATT CT 3′] and mE1Agenomic Rev [5′ GGT CGA AGC AGA CGG TTC TTC 3′]), and the results were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which was analyzed using a GAPDH-specific primer/probe set (Thermo Fisher Scientific Mm99999915_g1; catalog no. 4331182). When comparing Gcn2 -/-and WT MEFs and BMDMs, cells were infected with MAV-1 at an MOI of 5. At 48 and 72 hpi, cells and supernatants went through three freeze-thaw cycles, were clarified by centrifugation at 500 × g for 10 min, and were stored at -80°C until titration in 3T6 cells by plaque assay (60).
MAV-1 viral DNA loads in brains and spleens were determined by qPCR with MAV-1 E1A genomic primers (62). Organs were harvested at 8 dpi, frozen, and DNA was extracted using the Invitrogen PureLink Genomic DNA Kit (K182002). DNA (10 ng) was analyzed by qPCR in 10 µL reactions. Real-time PCR was performed on an ABI Prism 7500 Fast Real-Time PCR System (Applied Biosystems), and the results were compared to a standard curve of plasmid containing known amounts of an E1A gene-containing plasmid to convert cycle threshold values to E1A DNA copy numbers. Each sample was assayed in triplicate.
## Antibodies
Blots for pGCN2 detection were first probed with the rabbit monoclonal Anti-GCN2 (phospho T899), which does not bind to nonphosphorylated GCN2 (Abcam, ab75836). To avoid removing GCN protein from the membrane by harsh stripping, the blots were briefly washed with 95% ethanol and then treated with 15% hydrogen peroxide for 30 min, as described (44). Then, to detect total GCN2, we used the rabbit monoclo nal Anti-GCN2 (Cell Signaling, 3302). Before other primary antibody incubations, the ethanol/hydrogen peroxide treatment was repeated. To detect PKR, we used mouse monoclonal anti-PKR B-10 (Santa Cruz Biotechnology, sc-6282). To detect eIF2α, we used the rabbit polyclonal Anti-eIF2α (Invitrogen, AHO1182), and to detect phosphorylated eIF2α, we used the rabbit polyclonal Anti-eIF2α [pS52] (Invitrogen, 44728G). The primary antibody to MAV-1 E1A was 10B10, a mouse monoclonal described previously (63). Secondary antibodies for immunoblot were goat anti-rabbit-HRP and goat anti-mouse HRP (Kindle Biosciences R1006, and 1005, respectively). Before using the actin antibody, a harsh stripping method was used (rather than ethanol/hydrogen peroxide), which was a 15 min treatment at 56°C with 10% SDS, 62.5 mM Tris pH 6.8, and 0.8% 2-mercaptoe thanol. We used a mouse monoclonal antibody to β-actin, (Santa Cruz Biotechnology, sc-47778) as a loading control, following all other antibody treatments.
## Preparation of cell lysates for immunoblot
Cell lysates for analysis of GCN2 were prepared as described (44), with minor modifications. Briefly, MEFs and BMDMs were plated in 12-well plates, with 4 × 10 5 cells per well for MEFS and 1 × 10 6 cells per well for BMDMs. Cells were infected with MAV-1 MOI of 5, and the virus inoculum was not removed. When ready to harvest, cells were washed once with DPBS, and then 80 µL of denaturing protein sample buffer (0.625M Tris-HCl pH 6.8; 10% glycerol; 3% SDS; 0.5 mM EDTA, 5% (vol/vol) 2-mercaptoetanol, 0.1% (wt/vol) bromophenol blue sample buffer) was added dropwise evenly across each well. The lysates were scraped and collected in 1.5 mL tubes and immediately snap-frozen. Before loading on gels, samples were incubated in a Thermomixer for 10 min at 99°C with 1,400 rpm of agitation.
## Immunoblot analysis
Lysates for immunoblots were prepared as described above and were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 4%-15% gradient gels (BioRad 4561085). Gels were immunoblotted and blots visualized as described (25). Protein standards (Bio-Rad 1610374) were included in the gel as a size marker. ImageJ analysis for densitometry was performed using the Gel Analysis method (64).
## BBB permeability assay
Male and female WT and Gcn2 -/-mice, 4-5 weeks old, were injected i.p. with 10 2 MAV-1. After 8 days of infection, 100 µL of 10% sodium fluorescein (Sigma) diluted in DPBS was injected i.p. 10 minutes prior to euthanasia. Cardiac blood was collected, and mice were transcardially perfused with 30 mL of DPBS. Brains were snap-frozen until used for quantitation. Sodium fluorescein levels in brain and serum were determined as previously described using the right brain hemisphere (65). Fluorescence levels were measured on a Bio-Tek microplate reader with a 485 nm excitation and a 530 nm emission. Sodium fluorescein standards were prepared in DPBS and used to calculate the sodium fluorescein content of brain and serum samples. Brain values were normalized to their respective serum dye values to allow comparisons among mice.
## Cytokine quantitation by qPCR and ELISA
For in vivo cytokine determination, WT and Gcn2 -/-mice, 4 to 5 weeks old, were infected with 10 2 PFU of MAV-1, and brains were collected at 8 dpi. Approximately 50 mg of each brain was homogenized using sterile glass beads in a Mini-Beadbeater (Biospec Products) for 30 s in 1 mL of TRIzol (Invitrogen). RNA was then isolated from the homogenates according to the manufacturer's protocol and stored at -70°C until use. Then, 1 µg of RNA was reverse-transcribed using the High-Capacity cDNA reverse transcription kit (Thermo Fisher Scientific) according to the manufacturer's instructions. cDNA corresponding to 35 ng of RNA equivalent was used in each qPCR, and each sample was analyzed in triplicate. Quantitation was performed by normalizing target gene mRNA levels to β-actin levels, and infected sample values were expressed relative to the mean of mock values, set to 1 for each gene. To calculate the statistical significance of between-group differences, we used ΔCT and Log (2-ΔΔCT) values. Primers and probes used were as described (65).
For protein quantitation by ELISA in mouse brains, approximately 80 mg of each brain was processed as described (65). For in vitro cytokine determination, BMDMs were seeded overnight at a density of 2 × 10 5 cells/well in 48-well plates and prestimula ted with 0.2 µg/mL of bacterial lipopolysaccharide (LPS, Sigma-Aldrich L2630) for 4 h and subsequently infected with conditioned media or MAV-1 MOI of 10. The levels of cytokines and chemokines in mouse brains and cell culture supernatants were measured by ELISA at the University of Michigan Cancer Center Immunology Core. We measured IFN-γ, TNF-α, IL-1β, IL-6, IL-10, IP-10 (CXCL10), and RANTES (CCL5) proteins. Samples were stored at -70°C until use. Before use, samples were thawed on ice and centrifuged at 20,000 × g for 5 min at 4°C. Cytokine measurement (R&D and Peprotech) and protein quantitation by Pierce BCA protein assay kit (Thermo Scientific) were performed according to the manufacturer's instructions.
## Histological analysis
WT and Gcn2 -/-mice were either mock-infected or infected with 10 2 PFU of MAV-1, and at 8 dpi, organs were processed for histology. After euthanasia, mice were perfused with 30 mL 10% formalin (3.7% formaldehyde in PBS), and organs (thymus, lung, heart, brain, liver, kidney, and spleen) were immersion fixed in 10% neutral buffered formalin at 4°C. After 24 hours, organs were transferred to 70% ethanol and embedded in paraffin and sectioned at 4 µm. Sections were stained with hematoxylin and eosin. The University of Michigan Comprehensive Cancer Center Research Histology and Immunoperoxidase Laboratory performed sectioning and staining. Slides were randomized and blinded for evaluation by a board-certified pathologist. The sections were evaluated and scored as described (65).
## References
1. Pakos-Zebrucka, Koryga, Mnich et al. (2016) "The integrated stress response" *EMBO Rep*
2. Harding, Zhang, Zeng et al. (2003) "An integrated stress response regulates amino acid metabolism and resistance to oxidative stress" *Mol Cell*
3. (1016)
4. Igarashi, Murase, Iizuka et al. (2008) "Elucidation of the heme binding site of heme-regulated eukaryotic initiation factor 2alpha kinase and the role of the regulatory motif in heme sensing by spectroscopic and catalytic studies of mutant proteins" *J Biol Chem*
5. Clemens, Hershey, Hovanessian et al. (1993) "PKR: proposed nomenclature for the RNA-dependent protein kinase induced by interferon" *J Interferon Res*
6. Harding, Zhang (1999) "Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase" *Nature*
7. Wek, Jackson, Hinnebusch (1989) "Juxtaposition of domains homologous to protein kinases and histidyl-tRNA synthetases in GCN2 protein suggests a mechanism for coupling GCN4 expression to amino acid availability" *Proc Natl Acad Sci*
8. Deng, Harding, Raught et al. (2002) "Activation of GCN2 in UVirradiated cells inhibits translation" *Curr Biol*
9. Lee, Cevallos (2009) "An upstream open reading frame regulates translation of GADD34 during cellular stresses that induce Full-Length Text Journal of Virology October"
10. "eIF2alpha phosphorylation" *J Biol Chem*
11. Lu, Wambach, Katze et al. (1995) "Binding of the influenza virus NS1 protein to double-stranded RNA inhibits the activation of the protein kinase that phosphorylates the elF-2 translation initiation factor" *Virology (Auckl)*
12. Chang, Jacobs (1993) "Identification of a conserved motif that is necessary for binding of the vaccinia virus E3L gene products to doublestranded RNA" *Virology (Auckl)*
13. Rice, Turner, Embury et al. (2011) "Roles of vaccinia virus genes E3L and K3L and host genes PKR and RNase L during intratracheal infection of C57BL/6 mice" *J Virol*
14. Carroll, Stein, Moss et al. (1993) "Recombinant vaccinia virus K3L gene product prevents activation of double-stranded RNAdependent, initiation factor 2 alpha-specific protein kinase" *J Biol Chem*
15. Cárdenas, Loo, Jr et al. (2006) "Ebola virus VP35 protein binds double-stranded RNA and inhibits alpha/beta interferon production induced by RIG-I signaling" *J Virol*
16. Cai, Carpick, Chun et al. (2000) "HIV-I TAT inhibits PKR activity by both RNA-dependent and RNA-independent mecha nisms" *Arch Biochem Biophys*
17. Cassady, Gross, Roizman (1998) "The herpes simplex virus US11 protein effectively compensates for the gamma1(34.5) gene if present before activation of protein kinase R by precluding its phosphorylation and that of the alpha subunit of eukaryotic translation initiation factor 2" *J Virol*
18. Poppers, Mulvey, Khoo et al. (2000) "Inhibition of PKR activation by the proline-rich RNA binding domain of the herpes simplex virus type 1 Us11 protein" *J Virol*
19. Kitajewski, Schneider, Safer et al. (1986) "Adenovirus VAI RNA antagonizes the antiviral action of interferon by preventing activation of the interferoninduced eIF-2 alpha kinase" *Cell*
20. Black, Safer, Hovanessian et al. (1989) "The cellular 68,000-Mr protein kinase is highly autophosphorylated and activated yet significantly degraded during poliovirus infection: implications for translational regulation" *J Virol*
21. Habjan, Pichlmair, Elliott et al. (2009) "NSs protein of rift valley fever virus induces the specific degradation of the double-stranded RNAdependent protein kinase" *J Virol*
22. Ikegami, Narayanan, Won et al. (2009) "Rift Valley fever virus NSs protein promotes post-transcriptional downregulation of protein kinase PKR and inhibits eIF2alpha phosphor ylation" *PLoS Pathog*
23. Kalveram, Ikegami (2013) "Toscana virus NSs protein promotes degradation of double-stranded RNA-dependent protein kinase" *J Virol*
24. Li, Zhu, Du et al. (2017) "Foot-and-mouth disease virus induces lysosomal degradation of host protein kinase PKR by 3C proteinase to facilitate virus replication" *Virology (Auckl)*
25. Chang, Lau, Kuo et al. (2017) "dsRNA binding domain of PKR is proteolytically released by enterovirus A71 to facilitate viral replication" *Front Cell Infect Microbiol*
26. Goodman, Pretto, Krepostman et al. (2019) "Enhanced replication of mouse adenovirus type 1 following virusinduced degradation of protein kinase R (PKR)" *mBio*
27. Tejera-Hernández, Goodman, Nevarez et al. (2022) "Mouse adenovirus type 1 E4orf6 induces PKR degradation" *J Virol*
28. He, Lu, Lin et al. (2023) "Fowl adenovirus serotype 4 52/55k protein triggers PKR degradation by ubiquitin-proteasome system to evade effective innate immunity" *Vet Microbiol*
29. Guida, Fejer, Pirofski et al. (1995) "Mouse adenovirus type 1 causes a fatal hemorrhagic encephalomyelitis in adult C57BL/6 but not BALB/c mice" *J Virol*
30. Spindler, Fang, Moore et al. (2001) "SJL/J mice are highly susceptible to infection by mouse adenovirus type 1" *J Virol*
31. Spindler, Moore, Cauthen et al. (2007) "Mouse adenoviruses"
32. Blailock, Rabin, Melnick (1968) "Adenovirus myocarditis in mice. An electron microscopic study" *Exp Mol Pathol*
34. Mccarthy, Procario, Twisselmann et al. (2015) "Proinflammatory effects of interferon gamma in mouse adenovirus 1 myocarditis" *J Virol*
35. Iii, Pereira, Castro-Jorge et al. (2025) "Role of mouse adenovirus type 1 E4orf6-induced degradation of protein kinase R in pathogenesis" *J Virol*
36. Won, Eidenschenk, Arnold et al. (2012) "Increased susceptibility to DNA virus infection in mice with a GCN2 mutation" *J Virol*
37. Gibbs, Lin, Ghuge et al. (2024) "GCN2 in viral defence and the subversive tactics employed by viruses" *J Mol Biol*
38. Berlanga, Ventoso, Harding et al. (2006) "Antiviral effect of the mammalian translation initiation factor 2alpha kinase GCN2 against RNA viruses" *EMBO J*
39. Del Pino, Jiménez, Ventoso et al. (2012) "GCN2 has inhibitory effect on human immunodeficiency virus-1 protein synthesis and is cleaved upon viral infection" *PLoS One*
40. Cosnefroy, Jaspart, Calmels et al. (2013) "Activation of GCN2 upon HIV-1 infection and inhibition of translation" *Cell Mol Life Sci*
41. Misra, Carlson, Spandau et al. (2024) "Multiple mechanisms activate GCN2 eIF2 kinase in response to diverse stress conditions" *Nucleic Acids Res*
42. Ravindran, Khan, Nakaya et al. (2014) "Vaccine activation of the nutrient sensor GCN2 in dendritic cells enhances antigen presentation" *Science*
43. Hsu, Laurent-Rolle, Pawlak et al. (2022) "Viperin triggers ribosome collision-dependent translation inhibition to restrict viral replication" *Mol Cell*
44. Ravindran, Loebbermann, Nakaya et al. (2016) "The amino acid sensor GCN2 controls gut inflammation by inhibiting inflammasome activation" *Nature*
45. Gauthier-Coles, Rahimi, Bröer et al. (2023) "Inhibition of GCN2 reveals synergy with cell-cycle regulation and proteostasis" *Metabolites*
46. (2025) *Full-Length Text Journal of Virology*
47. Zhao, Guo, Hou et al. (2023) "Multiple roles of the stress sensor GCN2 in immune cells" *IJMS*
48. Silva, Castilho, Sattlegger (2018) "A rapid extraction method for mammalian cell cultures, suitable for quantitative immunoblotting analysis of proteins, including phosphorylated GCN2 and eIF2α" *MethodsX*
49. Sennepin, Charpentier, Normand et al. (2009) "Multiple reprobing of Western blots after inactivation of peroxidase activity by its substrate, hydrogen peroxide" *Anal Biochem*
50. Maurin, Jousse, Averous et al. (2005) "The GCN2 kinase biases feeding behavior to maintain amino acid homeostasis in omnivores" *Cell Metab*
51. Charles, Guida, Brosnan et al. (1998) "Mouse adenovirus type-1 replication is restricted to vascular endothelium in the CNS of susceptible strains of mice" *Virology (Auckl)*
52. Kajon, Brown, Spindler (1998) "Distribution of mouse adenovi rus type 1 in intraperitoneally and intranasally infected adult outbred mice" *J Virol*
53. Kring, King, Spindler (1995) "Susceptibility and signs associated with mouse adenovirus type 1 infection of adult outbred Swiss mice" *J Virol*
54. Gralinski, Ashley, Dixon et al. (2009) "Mouse adenovirus type 1-induced breakdown of the blood-brain barrier" *J Virol*
55. Neznanov, Dragunsky, Chumakov et al. (2008) "Different effect of proteasome inhibition on vesicular stomatitis virus and poliovirus replication" *PLoS One*
56. Hawkins, Egleton (2006) "Fluorescence imaging of blood-brain barrier disruption" *J Neurosci Methods*
57. Anastasiadis, Gandhi, Guo et al. (2021) "Localized blood-brain barrier opening in infiltrating gliomas with MRI-guided acoustic emissions-controlled focused ultrasound" *Proc Natl Acad Sci*
58. Xu, On, Preul (2025) "Fluorescein sodium as a marker for focused ultrasound-induced blood-brain barrier disruption: a case report in a porcine model" *Front Surg*
59. Orsini, Araujo, Maricato et al. (2014) "GCN2 kinase plays an important role triggering the remission phase of experimental autoimmune encephalomyelitis (EAE) in mice" *Brain Behav Immun*
60. Tallóczy, Jiang, Virgin et al. (2002) "Regulation of starvation-and virusinduced autophagy by the eIF2alpha kinase signaling pathway" *Proc Natl Acad Sci*
61. Brocard, Iadevaia, Klein et al. (2020) "Norovirus infection results in eIF2α independent host translation shut-off and remodels the G3BP1 interactome evading stress granule formation" *PLoS Pathog*
62. Dever (2002) "Gene-specific regulation by general translation factors" *Cell*
63. Ball, Beard, Villegas et al. (1991) "Early region 4 sequence and biological comparison of two isolates of mouse adenovirus type 1" *Virology (Auckl)*
64. Cauthen, Welton, Spindler (2007) "Construction of mouse adenovirus type 1 mutants" *Methods Mol Med*
65. Ashley, Welton, Harwood et al. (2009) "Mouse adenovirus type 1 infection of macrophages" *Virology (Auckl)*
66. Nguyen, Mcguffie, Anderson et al. (2008) "Gammaher pesvirus modulation of mouse adenovirus type 1 pathogenesis" *Virology (Auckl)*
67. Fang, Stevens, Berk et al. (2004) "Requirement of Sur2 for efficient replication of mouse adenovirus type 1" *J Virol*
68. Stael, Miller, Fernández-Fernández et al. (2022) "Detection of damage-activated metacaspase activity by Western blot in plants" *Methods Mol Biol*
69. Castro-Jorge, Pretto, Smith et al. (2017) "A protective role for interleukin-1 signaling during mouse adenovirus type 1-induced encephalitis" *J Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12403207&blobtype=pdf | # Orthohantavirus seoulense as a cause of acute, dengue-negative febrile illness in southern Vietnam
Mai Thi, Quynh Le, Kumiko Yoshimatsu, Haruka Abe, Thuy Nguyen, Hang Le, Khanh Nguye, Trang Thi, Hong Ung, Phuong Vu, Mai Hoang, Nobuo Koizumi, Futoshi Hasebe, Kozue Miura
## Abstract
Background: Dengue fever has surged in Vietnam since 2021; however, the aetiology of non-dengue undifferentiated febrile illnesses remains poorly understood.
Methods:Fifty whole blood samples that tested negative in rapid tests for anti-dengue virus antibodies (IgM and IgG) and NS1 viral antigen at Vung Tau General Hospital, southern Vietnam, were subjected to nucleic acid amplification tests for flaviviruses, hantaviruses, Leptospira spp. and Orientia tsutsugamushi , followed by DNA sequencing. The plasma samples were also tested for anti-hantavirus IgM and IgG antibodies using ELISA.Results: Of the 50 samples, eight were PCR-positive for flaviviruses and two were positive for hantaviruses. Sequencing analysis revealed that three and five of the eight flavivirus-positive samples were dengue virus type 1 and dengue virus type 2, respectively. The hantavirus species was identified as Orthohantavirus seoulense (SEOV). None of the patients tested positive for Leptospira spp. or O. tsutsugamushi . Anti-hantavirus IgM and IgG antibodies were detected in five and four patients, respectively.
Conclusions:This study suggests that SEOV is a notable contributor to dengue-negative febrile illnesses in southern Vietnam.
## Introduction
In tropical and subtropical regions, acute febrile illness is a common reason for seeking healthcare. 1 Although malaria has been a common aetiology of tropical acute febrile illness in the last decade, its incidence has substantially declined due to the success of malaria eradication programmes. 2 By contrast, dengue incidence has surged globally, increasing 30-fold over the past 50 y, with an estimated 100-400 million infections occurring annually across > 100 endemic countries. 3 In addition to dengue, other common causes of febrile illness in South and Southeast Asia include leptospirosis, typhoid, scrub typhus, rickettsioses, influenza, Japanese encephalitis and 'conventional' bacteremia, as identified through active surveillance and studies conducted in recent years. 4 Typically, patients with febrile illness who test negative for major suspected pathogens via rapid diagnostics are classified as having undifferentiated febrile illnesses. Dengue fever, caused by dengue virus (DENV), is primarily endemic to tropical and subtropical regions of the world and is the most common mosquito-borne viral infection. 5 DENV consists of four antigenically distinct serotypes (DENV1-4). The disease manifests as a sudden onset of fever accompanied by headache, ocular pain, myalgia and osteoarticular and gastrointestinal symptoms, typically lasting for up to 7 d. 6 Heteroserotype DENV secondary infection is a high-risk factor for severe dengue, which can lead to organ failure and death. 7 In Vietnam, the number of suspected dengue fever cases in 2021 rose to approximately five times the number reported in 2020 and a large outbreak due to a new DENV-2 lineage occurred in 2022. 8 Hantaviruses are primarily rodent-borne and can cause haemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS) in humans who inhale aerosolised excreta from chronically infected reservoir rodents. 9 The clinical features of HFRS include mainly fever, and then haemorrhage and varying degrees of renal and hepatic dysfunction. More than 90% of HFRS cases occur in Russia and East Asian countries, including China and Korea. 10 , 11 12 -15 However, little is known about the occurrence of hantavirus infections in humans in Vietnam. The most recently reported human case was a case of HFRS due to Orthohantavirus seoulense (SEOV) infection in southern Vietnam in 2008. 16 In a serological study of patients with unidentified febrile illnesses, anti-hantavirus antibodies were detected in 2.3% (5/220) of patients in northern Vietnam. 12 , 16 Because the clinical symptoms of hantavirus infection are similar to those of other tropical febrile illnesses, such as dengue, it is possible that cases of hantavirus infection remain undiagnosed among patients with unidentified febrile illness in this country.
The current study aimed to elucidate the causative agents other than DENV for undifferentiated febrile illness in Vietnam. We performed nucleic acid amplification tests for flaviviruses, hantaviruses, Leptospira spp. and Orientia tsutsugamushi , which cause clinical manifestations similar to dengue, in plasma samples collected from patients with febrile illness who tested negative for dengue.
## Materials and methods
## Sample collection
This was a pilot study carried out retrospectively at Vung Tau General Hospital, the main hospital in Vung Tau Province, southern Vietnam. Whole blood samples were collected with EDTA in 2024 from febrile patients ( > 38°C body temperature) with headache who were empirically suspected of having dengue by clinicians at the hospital. The whole blood samples were subjected to rapid tests for anti-dengue virus antibodies (IgG and IgM) and NS1 viral antigen using the SD BIOLINE Dengue Duo NS1 Ag + Ab Combo (Abbott, IL, USA). The most recent 50 samples that were collected on and before 27 January 2024 and that were tested negative in all rapid tests were separated into buffy coat and plasma components, which were used in this study.
## Detection of pathogen genes
RNA was extracted from the buffy coat using the ISOSPIN Cell and Tissue RNA Kit (Nippon Gene, Tokyo, Japan), following the manufacturer's instructions. cDNA was synthesised from the total RNA using the PrimeScript RT Master reagent kit (Takara Bio, Japan). To detect the dengue virus gene, the NS5 gene of flaviviruses was amplified using the primers cFD2 (5 -GTGTCCCAGCCGGCGGTGTCATCAGC-3 ) and MAMD (5 -AACATGATGGGRAARAGRGARAA-3 ) in the first reaction. Semi-nested amplification was then performed using the cFD2 and FS778 (5 -AARGGHAGYMCDGCHATHTGGT-3 ) primers. 17 For hantavirus detection, a conserved domain of the L genome segment of hantavirus was amplified using the HAN-L-F2 (5 -TGCWGATGCHACIAARTGGTC-3 ) and HAN-L-R1 (5 -AACCADTCWGTYCCRTCATC-3 ) primers for the first reaction. Semi-nested amplification was conducted using the HAN-L-F2 and HAN-L-R2 (5 -GCRTCRTCWGARTGRTGDGCAA-3 ) primers. 18 To detect Leptospira , the flaB gene was amplified from the cDNAs prepared as above by nested PCR using L-flaB F1 (5 -CTCACCGTTCTCTAAAGTTCAAC-3 ) and L-flaB R1 (5 -TGAATTCGGTTTCATATTTGCC-3 ) primers for the first reaction, followed by amplification with M-L-flaB F2 (5 -TGTGGACAAGACGATGAAAGC-3 ) and M-L-flaB R2 (5 -AACATTGCCGATCCACTCTG-3 ) primers for the second reaction. 19 For the detection of O. tsutsugamushi , the gene encoding the 56 kDa type specific antigen was amplified from the cDNAs prepared as above by nested PCR using 34 (5 -TCAAGCTTATTGCTAGTGCAATGTCTGC-3 ) and 55 (5 -AGGGATCCCTGCTGCTGTGCTTGCTGCG-3 ) primers for the first reaction, and 10 (5 -GATCAAGCTTCCTCAGCCTACTATAATGCC-3 ) and 11 (5 -CTAGGGATCCCGSCAGATGCACTATTAGGC-3 ) primers for the second reaction. 20 The PCR products were subjected to agarose gel electrophoresis, and amplicons of the expected size were purified using the NucleoSpin Gel and PCR Clean-up (Takara Bio, Japan). These amplicons were then analysed by Sanger sequencing to confirm the pathogen species. The nucleotide sequence obtained from the current study has been deposited in the DDBJ database [accession no. LC822654].
## Antibody detection for hantavirus
Anti-hantavirus IgG and IgM antibodies were detected in plasma samples using the Hantavirus IgG Dx Select and Hantavirus IgM Dx Select kits (Diasorin, Cypress, CA, USA), respectively. For IgG detection, 10 µl of each sample was diluted with 1 mL of the sample diluent provided. After equilibrating the antigen wells with the wash buffer, 100 µl of each diluted specimen was added to the antigen wells. The plate was incubated for 60 min at room temperature then washed three times with 300 µL of the wash buffer. Then 100 µl of the IgG conjugate was added to the wells. After incubation at room temperature for 30 min and washing three times, 100 µl of the substrate reagent was added to the wells and incubated for 10 min at room temperature. The optical density (OD) values were measured at a wavelength of 450 nm using a Multiskan FC microplate spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The ELISA procedure for IgM detection was nearly identical to that for IgG, except that the plasma samples were mixed with the sample diluent and incubated for 10 min. The OD values of > 1.10 were defined as positive for both IgG and IgM according to the manufacturer's instructions.
## Phylogenetic analysis
The partial sequence of the L genome segment, amplified using the HAN-L-F2 and HAN-L-R2 primers, was aligned with representative sequences from other rodent-borne and Vietnamese hantaviruses using MUSCLE as implemented in GENETYX-MAC
## Results
## Detection of pathogen genes in patients with febrile illness negative for dengue rapid tests
We performed gene detection of flavivirus, hantavirus, Leptospira spp. and O. tsutsugamushi in 50 febrile patients who tested negative for dengue. Among the 50 samples tested, eight were positive for the flavivirus gene by PCR (Table 1 ). Sequencing analysis confirmed that three of these cases were caused by dengue virus type 1, while the remaining five were due to dengue virus type 2. Of the 42 samples that tested negative for flavivirus by PCR, two (4.8%) were positive for hantavirus, which was identified as Orthohantavirus seoulense through sequencing analysis, with two sequences found to be identical. None of the patients tested positive for Leptospira spp. or O. tsutsugamushi .
## Detection of IgM and IgG antibodies for hantaviruses in patients with febrile illness negative for dengue rapid tests
Anti-hantavirus IgM and IgG antibodies were detected in five and four febrile patients, respectively (Table 2 ). Four of the five IgMpositive patients also tested positive for anti-hantavirus IgG. One of the two patients in whom the O. seoulense gene was detected tested positive for both anti-hantavirus IgM and IgG. No anti-
## Phylogenetic analysis of hantaviruses
Phylogenetic analysis of the partial L genome segment revealed that the sequence derived from patient plasma samples clustered with Seoul viruses (Figure 1 ). This sequence was highly divergent from that of other hantaviruses derived from a bat and a Niviventer rodent in Vietnam. Notably, in the phylogenetic trees, the sequence formed the basal clade of strain LC822653, a Seoul virus isolated from R. norvegicus captured in Saigon harbour, Vietnam.
## Discussion
In the current study, we detected an O. seoulense genome fragment in two of the 42 dengue-negative, undifferentiated febrile patients (4.8%) in southern Vietnam. In addition, anti-hantavirus IgM was detected in five patients (11.9%), with both IgM and IgG present in four of these patients. In European and Asian hantavirus infections the presence of IgM is typically accompanied by IgG in > 90% of cases. 21 , 22 Although IgM can disappear after 12 mo, IgG remains elevated, 21 , 22 suggesting that IgM-positive patients were probably admitted to the hospital due to acute hantavirus infection. These findings suggest that O. seoulense may account for a substantial proportion of dengue-negative febrile illnesses, at least in southern Vietnam.
Hantavirus distribution in rodents generally aligns with the geographic distribution of their hosts. For instance, hantavirus associated with HCPS, transmitted by New World rodents, has only been reported in the Americas. 23 However, SEOV, which is transmitted by brown rats, is distributed globally, with cases reported in the USA and Brazil, as well as various Asian and European countries. 24 -30 It is believed that rodent-borne hantaviruses co-evolve with their hosts over geological timescales. 31 By contrast, the global distribution of SEOV is thought to have expanded rapidly in tandem with the worldwide migration of rats within the last few hundred years. SEOV may have spread globally with rats through shipping. Rattus norvegicus originated in Asia, spread to Europe in the seventeenth century and arrived in North America around 1755. 32 Due to the highly migratory nature of brown rats carrying SEOV, pinpointing the origin and spread of SEOV is challenging. However, genetic analysis has revealed little diversity between SEOV strains from China and Europe, 26 , 33 while unique lineages have been found to exist in Southeast Asia, including Singapore, Vietnam and Cambodia. 12 Most HFRS cases caused by SEOV are reported in East Asian countries, including Korea, Japan and China. 33 By contrast, few have been reported in Southeast Asian countries, aside for one case imported from Indonesia to Germany. 34 In Vietnam, rats harbouring SEOV are common in urban and port areas. 13 , 14 , 35 In a previous study, SEOV was detected in rats near the homes of febrile patients who tested positive for anti-hantaviral antibodies, 16 although no viral genes were detected in patients. This study, however, showed that the SEOV sequence direc tly detec ted in patients clustered closely with the SEOV sequence found in a brown rat in Saigon harbour, suggesting that the endemic SEOV strain in Vietnam can indeed cause disease in humans.
The current study has some limitations. It focused on a single hospital in Vung Tau Province, southern Vietnam, and examined only 50 patients. Therefore, it remains unclear whether SEOV causes febrile illnesses in humans in other parts of Vietnam. Anti-SEOV antibodies were detected in 2.9% (7/245) of individuals living in farming communities in Dong Thap Province, 15 although their clinical histories were not obtained. Moreover, hantaviral genome fragments, whose sequences clustered with those from southern Vietnam, have been detected in rats in Hai Phong city, northern Vietnam. 14 Further prospective investigations are required to clarify the significance of circulating SEOV in the country. In the current study, the buffy coat samples were collected from febrile patients, who were considered to be in the acute phase of infection, and we only performed PCR. However, PCR alone may underestimate the prevalence of Japanese encephalitis virus, Leptospira spp. and O. tsutsugamushi , because IgM against hantavirus was detected in 11.9% of the patients in this study. In addition, we were unable to obtain information on antibiotic treatment prior to testing and days of fever at the time of testing, which can potentially influence PCR results, especially for Leptospira spp. and O. tsutsugamushi . We were only able to obtain age and gender data, and were unable to obtain information on the clinical manifestations and laboratory test results of patients with the SEOV genome and/or anti-hantavirus antibodies. Further epidemiological and clinical studies are required to identify the risk factors and the clinical and laboratory features associated with hantavirus infection. In the current study, we determined the sequence of only the L genome segment. Sequencing the complete genome of SEOV would offer deep insights into the virulence and pathogenesis of SEOV circulating in Vietnam. Finally, dengue virus RNA was detected in 16% of febrile patients who tested negative on the dengue rapid tests in this study, highlighting the need to improve current rapid test kits for accurate diagnosis.
## References
1. Shrestha, Dahal, Ogbonnaa-Njoku (1980) "Non-malarial febrile illness: A systematic review of published aetiological studies and case reports from Southern Asia and South-eastern Asia" *BMC Med*
2. Who (2022) "World malaria report 2022"
3. Who, Dengue, Dengue (2025)
4. Shrestha, Roberts, Homsana (2018) "Febrile illness in Asia: gaps in epidemiology, diagnosis and management for informing health policy" *Clin Microbiol Infect*
5. Näslund, Ahlm, Islam (2021) "Emerging mosquito-borne viruses linked to Aedes aegypti and Aedes albopictus : global status and preventive strategies" *Vector Borne Zoonotic Dis*
6. Who (2025) "Dengue guidelines, for diagnosis, treatment, prevention and control"
7. Katzelnick, Gresh, Halloran (2017) "Antibody-dependent enhancement of severe dengue disease in humans" *Science*
8. Nabeshima, Tun, Thuy (2023) "An outbreak of a novel lineage of dengue virus 2 in Vietnam in 2022" *J Med Virol*
9. Noack, Goeijenbier, Reusken (2020) "Orthohantavirus Pathogenesis and Cell Tropism" *Front Cell Infect Microbiol*
10. Sehgal, Mehta, Sahay (2023) "Hemorrhagic fever with renal syndrome in Asia: history, pathogenesis, diagnosis, treatment, and prevention" *Viruses*
11. Tkachenko, Kurashova, Balkina (2023) "Cases of Hemorrhagic Fever with Renal Syndrome in Russia during 2000-2022" *Viruses*
12. Truong, Yoshimatsu, Araki (2009) "Molecular epidemiological and serological studies of hantavirus infection in northern Vietnam" *J Vet Med Sci*
13. Luan, Yoshimatsu, Endo (2012) "Studies on hantavirus infection in small mammals captured in southern and central highland area of Vietnam" *J Vet Med Sci*
14. Koma, Yoshimatsu, Yasuda (2013) "A survey of rodent-borne pathogens carried by wild Rattus spp. in Northern Vietnam" *Epidemiol Infect*
15. Van Cuong, Carrique-Mas, Be (2015) "Rodents and risk in the Mekong Delta of Vietnam: seroprevalence of selected zoonotic viruses in rodents and humans" *Vector Borne Zoonotic Dis*
16. Huong, Yoshimatsu, Luan (2010) "Hemorrhagic fever with renal syndrome" *Vietnam. Emerg Infect Dis*
17. Scaramozzino, Crance, Jouan (2001) "Comparison of flavivirus universal primer pairs and development of a rapid, highly sensitive heminested reverse transcription-PCR assay for detection of flaviviruses targeted to a conserved region of the NS5 gene sequences" *J Clin Microbiol*
18. Muthusinghe, Shimizu, Lokupathirage (1984) "Identification of novel rodent-borne Orthohantaviruses in an endemic area of chronic kidney disease of unknown etiology (CKDu) in Sri Lanka" *Viruses*
19. Koizumi, Muto, Akachi (2013) "Molecular and serological investigation of Leptospira and leptospirosis in dogs in Japan" *J Med Microbiol*
20. Furuya, Yoshida, Katayama (1993) "Serotype-specific amplification of Rickettsia tsutsugamushi DNA by nested polymerase chain reaction" *J Clin Microbiol*
21. Pattamadilok, Lee, Kumperasart (2006) "Geographical distribution of hantaviruses in Thailand and potential human health significance of Thailand virus" *Am J Trop Med Hyg*
22. Lundkvist, Hörling, Niklasson (1993) "The humoral response to Puumala virus infection (nephropathia epidemica) investigated by viral protein specific immunoassays" *Arch Virol*
23. Young, Mills, Enria (1998) "New World hantaviruses" *Br Med Bull*
24. Childs, Glass, Korch (1988) "Evidence of human infection with a rat-associated Hantavirus" *Am J Epidemiol*
25. Iversson, Da Rosa, Rosa (1994) "Human infection by Hantavirus in southern and southeastern Brazil]" *Rev Assoc Med Bras*
26. Wang, Yue, Yao (2020) "Epidemic Trend and Molecular Evolution of HV Family in the Main Hantavirus Epidemic Areas From 2004 to 2016" *P.R. China. Front Cell Infect Microbiol*
27. Verner-Carlsson, Lõhmus, Sundström (2015) "First evidence of Seoul hantavirus in the wild rat population in the Netherlands" *Infect Ecol Epidemiol*
28. Heuser, Drewes, Trimpert (2023) "Pet rats as the likely reservoir for human Seoul Orthohantavirus infection" *Viruses*
29. Alburkat, Smura, Bouilloud (2020) "Evolution and genetic characterization of Seoul virus in wild rats Rattus norvegicus from an urban park in Lyon" *PLoS Negl Trop Dis*
30. Fitte, Brignone, Sen (2023) "First study of Seoul virus (SEOV) in urban rodents from newly urbanized areas of Gran La Plata" *Braz J Infect Dis*
31. Plyusnin, Vapalahti, Vaheri (1996) "Hantaviruses: genome structure, expression and evolution" *J Gen Virol*
32. Suckow, Wilson, Foley (2019) "The laboratory rat"
33. Wang, Yoshimatsu, Ebihara (2000) "Genetic diversity of hantaviruses isolated in china and characterization of novel hantaviruses isolated from Niviventer confucianus and Rattus rattus" *Virology*
34. Hofmann, Weiss, Kuhns (2018) "Importation of Human Seoul Virus Infection to Germany from Indonesia" *Emerg Infect Dis*
35. Yasuda, Shimizu, Koma (2021) "Immunological Responses to Seoul Orthohantavirus in Experimentally and Naturally Infected Brown Rats" *Viruses* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12282185&blobtype=pdf | # Correction for Gurda et al., "Mapping a Neutralizing Epitope onto the Capsid of Adeno-Associated Virus Serotype 8"
Brittney Gurda, Christina Raupp, Ruth Popa-Wagner, Matthias Naumer, Norman Olson, Robert Ng, Robert Mckenna, Timothy Baker, Jürgen Kleinschmidt, Mavis Agbandje-Mckenna
## Abstract
The middle panel of Fig. 7A, which depicts the rAAV2 dot blots, appears to show a duplicated image for the trypsin-treated (-and +) samples. Because of the long passage of time since this work was published, we no longer have access to the original records. It is very possible that the rAAV2 dot blots were unintentionally duplicated during the assembly of Fig. 7, particularly as the original images were most likely visually similar.Despite this apparent error, we believe the overall conclusion drawn from Fig. 7 remains valid. Figures 7B andC present the numerical quantification of the dot blot data and do show subtle but discernible differences between the -and +trypsin treatments. This suggests that the original blot data varied and the error likely occurred during figure preparation rather than experimentation. This pattern of slight variation is also observed in the top and bottom panels of Fig. 7A (rAAV8 and rAAV2-RGNRQ/QQNTA), and these trends are consistently reflected in the accompanying quantitative analysis (Fig. 7B andC).In conclusion, while we agree that the rAAV2 dot blot image (Fig. 7A, middle panel) is flawed, we do not believe this affects the central findings or conclusions of the study. Nonetheless, we appreciate the thorough review that led to the finding and investigation of this error. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12816068&blobtype=pdf | # Assessing a Syndemic of Discrimination, Material Insecurity, Depression, Substance Use, and Violence Among Sexual and Gender Minorities in Nigeria Using Mixed Methods
Rodman Turpin, Megan Mansfield, Typhanye Dyer, Andrew Mitchell, Chama John, Ruxton Adebiyi, Uchenna Ononaku, Christiana Katu, Jumoke Aigoro, Abayomi Aka-Bashorun, Sylvia Adebajo, Manhattan Charurat, Rachel Robinson
## Abstract
Sexual and gender minority people (SGM) in Nigeria experience disproportionate HIV burden, with an HIV prevalence four to ten times higher than the national average. Better understanding the factors that create HIV vulnerability in this population is important for designing effective interventions, particularly in a context largely hostile to SGM. We assessed a conceptual model describing a syndemic of discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other violence among SGM in Abuja, Nigeria. As part of a larger, longitudinal study examining noncommunicable disease outcomes within this population, we conducted a mixed methods analysis using both quantitative intake data (n=515) as well as data from three focus groups (n=36), collected from July 2023 through May 2024. We tested for intercorrelations among syndemic components, and associations between a cumulative syndemic index and HIV status using modified Poisson regression. We also conducted a convergent qualitative assessment of the conceptual model in three focus group discussions. Finally, we examined co-prevalence of syndemic components highlighted in our qualitative findings. There were consistent intercorrelations among syndemic components, supporting the presence of a syndemic. After adjustment for sociodemographic factors, every quartile-unit increase in the syndemic index was associated with an 18% increase in prevalence of HIV (aPR=1.18, 95% CI 1.07, 1.29). Additionally, our qualitative findings highlighted relationships between discrimination, material insecurity, and depression as especially relevant among this population. When using our quantitative data to examine the co-prevalence of pairs of syndemic components identified as particularly salient in our qualitative analyses, nearly every relationship was significantly stronger than expected. We found strong evidence of a syndemic of discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other violence among SGM in Abuja, Nigeria as salient to the health outcomes of SGM in Nigeria. Overall, our findings highlight the presence of a multilevel syndemic that informs multilevel intervention targets. Interventions must target not simply the individual level, but also incorporate larger scale social and structural change efforts.
## Introduction
HIV in Nigeria remains a critical area for public health efforts. Based on both national surveillance and predictive modeling, the estimated national HIV prevalence in Nigeria among adults aged 15-49 years ranges from 1.3 to 2.1%, equivalent to about 1.5 to 2 million people living with HIV [1,2]. Notably, sexual and gender minority people (SGM) in Nigeria are disproportionately vulnerable to HIV; this includes men who have sex with men (MSM), gay, bisexual, and other people with sexually minoritized identities, and transgender women [3,4]. HIV prevalence among Nigerian MSM is estimated to be 23%, ten times higher than the national average [1]. Among Nigerian transgender women (TGW) and other SGM sampled through community-based health centers offering SGM-friendly services, HIV prevalence ranged from 44 to 66% [5]. Factors that have been associated with HIV among SGM populations include depression, discrimination, violence, and substance use [6][7][8], which often co-occur with one another, substantially affecting health outcomes for this population. These disproportionately high prevalence rates, as well as the high prevalence of co-occurring conditions, highlight the need for better understanding the co-occurrence of factors that impact HIV vulnerability in this population.
Syndemic theory posits that risk factors cluster because of common social and structural drivers, which amplify the negative impact of these risk factors on health outcomes, such as HIV prevalence, diabetes, and others [9]. This theory has been utilized to more effectively elucidate synergistic HIV risk factors among marginalized communities of SGM, including Black SGM in the United States [6][7][8]. Our conceptual framework (Fig. 1) illustrates how six different individual-level factors particularly relevant to the lived experiences of SGM (sex and gender-based discrimination, depression, intimate partner violence, police and other violence, material insecurity, and polysubstance use) interact to form a syndemic associated with adverse health outcomes in this population, particularly related to HIV. Importantly, structural factors impact and affect the clustering and interactions of individual-level factors. Structural factors identified in our conceptual framework include discriminatory polices (e.g., the Same-Sex Marriage (Prohibition) Act, the Nigerian Criminal Code) and social norms regarding marriage, childbearing, and gender. These structural factors facilitate discrimination by health care providers, the media, religious leaders, law enforcement, family members, and the overall community and there is substantial literature supporting the relationships between these factors in related populations [10][11][12][13]. For example, several studies among Black MSM have demonstrated relationships between discrimination and depression [10][11][12][13][14], intimate partner violence and substance use [15], stigma and mental health [13,16], and socioeconomic insecurity and depression [17,18]. The specific legal, social, and economic context of Nigeria related to the criminalization of same-sex sexuality, strong social norms around childbearing, and the high poverty rates, makes this syndemic uniquely relevant to the health outcomes of the Nigerian SGM population [11].
The purpose of this study is to assess the presence of a syndemic of discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other forms of violence among SGM in Nigeria using a convergent mixed methods approach. We tested for quantitative evidence of interrelationships of the components identified in our framework and associations with HIV status. We also qualitatively explored relationships between these components using focus group discussions (FGD). Finally, we assessed with the quantitative data intercorrelations of pairwise components highlighted by our qualitative findings. We hypothesized that this syndemic characterized by our conceptual model (see Fig. 1) would be associated with positive HIV status among SGM in Nigeria, and that qualitative findings would provide insight into how the different elements of this syndemic interrelate to impact the health of the SGM community.
## Methods
## Study Setting
This convergent mixed method study is part of a prospective longitudinal cohort study (5R01HL165686-02) examining noncommunicable disease (NCD) outcomes among SGM in Abuja, Nigeria. The overall aims of the parent study are to (1) examine associations between a multilevel syndemic and both HIV and NCD outcomes, (2) assess the influence of structural-level components on HIV and NCDs, and (3) determine the extent to which syndemic components increase risk of disengaging from care for HIV and NCDs. This study was conducted at the TRUST Clinic in Abuja, Nigeria, established in 2012 through a tripartite partnership among the nonprofit community-based organization, International Center for Advocacy and Rights to Health (ICARH), the Institute of Human Virology Nigeria, and the University of Maryland Baltimore [19]. As a result of this well-established collaboration, ICARH has recruited a cohort of nearly 2,800 SGM and rigorously evaluated the benefits of HIV test and treat within the secure environment of the community center. The TRUST Clinic provides broad HIV care and prevention as well as other health services to the SGM community and is embedded within the community-based organization ICARH. ICARH conducts advocacy programs, research, and human rights activism. The clinic has an on-site laboratory for HIV viral load monitoring and an in-house pharmacy which dispenses antiretroviral therapy, pre-exposure prophylaxis, and post-exposure prophylaxis as well as other routine medicines. These HIV prevention and management services are provided to clients by a team of SGM-friendly nurses, a pharmacist, a social worker, the HIV testing service team, and community extension workers. We collected all data from July 2023 through May 2024.
## Quantitative Measures
Quantitative data come from intake surveys. Syndemic exposures included depression using the PHQ-9 (continuous) [20], any cigarette use (yes, no), any stimulant use (yes, no), a 17-item index of material insecurity (continuous) [21], any intimate partner violence (yes, no), any other interpersonal violence (yes, no), a three-item index of experienced sexual/gender discrimination from family (yes, no), a fiveitem index experienced sexual/gender discrimination in healthcare settings (yes, no), a three-item index experienced sexual/gender discrimination or harassment from police (yes, no), and a nine-item index HIV stigma (continuous).
## Regression Analyses
We used modified Poisson regression with robust standard errors to test associations between our syndemic index and HIV status. This method is useful for generating prevalence ratios for binary outcomes and allows for inclusion of more confounders than log-binomial modeling. We generated unadjusted models and models adjusted for age, education, and personal income. We did not include sexual identity in models due to stability difficulties related to the unusually small "other" category. For all models we generated ratio estimates and 95% confidence intervals for associations between quartile-unit increases in the index and HIV status. The total range of the index was 0 to 10, and a quartile unit increase was 3 units, as this was half the interquartile range rounded to the nearest whole number. So each estimate in our regression models reflects a 3-unit syndemic index increase. We use quartiles as this reflects a more meaningful change in the index rather than a 1-unit increase, which would be overly small given the numerous constructs included. The quartile increase also roughly approximates a standard deviation, a commonly used unit measure for index differences, but without relying on assumptions of normality likely to be violated. Fit and variance metrics for models are also included. Additionally, we utilized a syndemic index with the HIV stigma component removed, to determine if there was a substantial difference in the association with HIV status without this component.
## Quality Assurance
Missingness for all items was low (less than 10%), with most items having less than 5% missingness. Given the low level of missingness, we employed maximum likelihood imputation to impute missing data. Post-imputation, we retained all observations (n = 515). Analysis of Cook's distances and leverages revealed no significant outliers. Additionally, variance inflation factors of less than 5 for all models indicated no evidence of intercollinearity. All quantitative analyses were conducted in SAS 9.4 [23].
## Qualitative Analyses
In addition to the analysis of intake survey data, we also conducted a convergent qualitative assessment of the conceptual framework, and its six components. Participants from the Syndemics Study who were enrolled as of September 2023 (n = 70) were randomly selected for participation in the first two FGD (n = 24). A third focus group (n = 12) was conducted with purposeful sampling of TGW resulting in three FGD with 36 participants across them. Each of the three FGD included 8-13 participants. Participants All discrimination and stigma items were adapted from the American Men's Internet Survey of Men Who Have Sex With Men in the United States [22]. Our 17-item measure of material insecurity is a composite of a nine-item measure of financial insecurity and an eight-item measure of nutritional insecurity [21]. We combined these two sets of measures due to both strong conceptual similarities (e.g., some of the financial insecurity items also reflect nutritional insecurity) but also very strong covariance between the two indices. Additionally, we kept the three constructs for SGM discrimination distinct, as these had varied relationships with other syndemic components. All components were scaled to range between 0 and 1 (i.e., scaled in percentage) to allow for equivalent weighting of items. For all scales, we subtracted the original scale minimum, then divided by the original scale maximum minus the original scale minimum. This transforms all scales to range between 0 and 1. For example, the PHQ-9 ranges from 0 to 27, so we subtracted 0, then divided by 27 minus 0, which simplified to dividing all values by 27. This resulted in a transformed scale that ranged from 0 to 1. Percentage scaling allows for items to be compared visually with more ease, with means for continuous items and proportions for binary items having the same range. Our primary outcome for assessing the syndemic was HIV status (negative, positive) confirmed through biological testing. Covariates included age (18-24, 25-29, 30-34, 35+), highest education level (less than secondary, senior secondary school, more than senior secondary), personal income (Naira:) in the last month (less than 20,000, 20,000 to 59,999, 60,000 or more), and sexual identity (bisexual, gay, other). We selected covariates based on associations with HIV status and vulnerability identified in the literature.
## Bivariate Analyses
We tested for associations between our syndemic index and ordinal covariates (age, education, income) using Spearman rank-sum correlation. Similarly, we tested for associations between our syndemic index and our binary outcome (HIV status) using a point-biserial Spearman rank-sum correlation. We also tested for associations between sexual identity and our syndemic index using a Kruskal-Wallis test. Finally, we tested for intercorrelations between all of our syndemic components using both Spearman rank-sum correlation (between two continuous/ordinal components), pointbiserial Spearman rank-sum correlation (between binary and continuous/ordinal components), and Phi correlations (between two binary components). We also present syndemic components associated with HIV status individually.
## Co-occurrence and Sensitivity Analyses
Following both initial quantitative and qualitative analyses, we assessed with the quantitative data the co-occurrence of syndemic components highlighted in the qualitative findings. Additionally, we conducted sensitivity analyses collapsing substance use into one factor, and discrimination into a single factor, as these were the only two constructs with more than one component in the index. We compared results for this modified index to the unmodified index.
## Ethics
This study received ethics approval from the institutional review boards of the University of Maryland, Baltimore, USA and the National and Federal Capital Territory Health Research Ethics Committees, Abuja, Nigeria. Separate written consent was provided for the quantitative and qualitative components.
## Results
## Sample and Bivariate Results
Our final sample consisted of 515 participants (Table 1). The median age was 25-29, and the median highest education level was senior secondary school. Two-thirds of the sample reported a personal income of less than 60,000 in the last month. Just over half of the sample identified as bisexual (57.3%), and just over a third identified as gay (37.5%). Nearly two-thirds of the sample was HIV-positive (61.9%). A greater syndemic index was associated with younger age, "other" sexual identity, and positive HIV status. Additionally, consistent intercorrelations among syndemic components existed, supporting the use of a syndemic index (Table 2). Intimate partner violence, other personal violence, and HIV stigma were all associated with positive HIV status (Fig. 2).
## Regression Results
In both unadjusted and adjusted regression models, our syndemic index was significantly associated with positive HIV status (Table 3). After adjustment, every quartile-unit increase in the syndemic index was associated with an 18% increase in prevalence of HIV (aPR = 1.18, 95% CI 1.07, 1.29). Notably, estimates were largely similar between adjusted and unadjusted models. Older age was also associated with HIV prevalence. Our socioeconomic covariates were not, though it should be noted that our syndemic index also captures socioeconomic deprivation. Finally, removing provided informed consent individually prior to participating in the FGD and received transportation funds as well as lunch after the FGD was completed. During the FGD, we prompted participants to discuss their conceptualizations of health; structural, community, and individual influences on health; and experiences and observations related to NCDs. Each FGD lasted about two hours. The FGD were audio recorded, transcribed verbatim and -as needed -simultaneously translated into English (from either Hausa or Pidgin). Two members of the research team independently coded the transcripts (MM and RSR) using a hybrid deductive-inductive coding method and the qualitative analysis software NVivo. We began with the creation of an initial codebook based predominately on deductively identified codes from the FGD guide and syndemic framework. We coded all three FGD using the initial codebook and then discussed points of disagreement and agreement in coding. This resulted in a refined codebook that also took into consideration patterns that emerged inductively from the data. We then independently coded the three transcripts again and discussed disagreements until agreement was reached. After coding was completed, themes were identified and organized using axial coding.
## Qualitative Results
Participants' reported ages ranged from 19 to 40 years and the majority (n = 22, 61%) identified as men, with the remainder self-identifying as either women (n = 12, 33%) or transgender (n = 2, 6%). Additionally, of the participants the HIV stigma component from the index only slightly attenuated results, though the syndemic index remained both a statistically significant predictor of HIV status and largely interpreted in the same way (aPR = 1.15, 95% CI 1.05, 1.28). pairs of syndemic factors, with more frequent references to the links between material insecurity and three other factors (depression and substance use in particular, but also discrimination), as well as to connections between depression and two other factors (substance use and discrimination).
## Material Insecurity
Many Nigerians face material insecurity, including members of the SGM community. Approximately 40% of Nigerians live below the national poverty line, and two thirds are multidimensionally poor [24]. As in most of the world, economic hardships followed Nigeria's COVID lockdown, as did spiraling inflation. Nigeria has also uniquely faced currency devaluation, and everyday life has become significantly more expensive in recent years, particularly with the removal of a fuel subsidy following the change in presidential administration in 2023 [25]. SGM are at additional risk of material insecurity given discrimination faced in securing employment and housing [26].
Participants understood material insecurity as preceding depression, often with discrimination at the root of material insecurity. For example, "I have a friend, who lost his job because his boss got to know that he is queer. Now, this person lost his job, lost his source of income, and lost everything. It got to the stage; this guy was depressed" (FGD1, unidentified participant). Or, as someone else described, "When you are discriminated, you might not be able to work. Like you cannot go and apply for jobs in an organization because when you get there, they say, 'This person is gay. You can't employ a gay person to work for us.' So you won't really have the boldness to go and seek for jobs. When you are discriminated, you get depressed" (FGD2, participant #8). These experiences were particularly acute for SGM who did not conform to gender expectations. For example, "You cannot go to a place and look for a job if you as a feminine person as a guy.. there are so many people, they cannot look for job because their papers are still reading their biological birth" (FGD3, participant #2).
Others noted social norms at the root of material insecurity, with SGM pressured into marriage and having children, not being able to afford the associated responsibilities, and falling into depression. "Nigeria, [these] days of course, who responded, a plurality identified as gay (n = 15, 42%) with the remainder identifying as bisexual (n = 9, 25%), queer (n = 3, 8%), or something else (n = 4, 11%). The FGD participants thus roughly mirrored the broader study population included in the quantitative analysis, although with a lower percentage identifying as bisexual and a higher percentage identifying as women due to our purposive sample of TGW for one FGD.
To parallel the bivariate analysis described above, we conducted a frequency analysis of the co-occurrence of syndemic factors (Table 4). When prompted with the six factors constituting the syndemic, participants discussed depression and material insecurity the most, followed by substance use and discrimination. They mentioned violence (interpersonal or at the hands of the police) somewhat less. This analysis revealed that participants identified connections between all you get this injustice, be it from the clinic center, be it from the police station, be it from your church or be it in any peer group. That is when you go back home and you enter into the house of depression" (FGD2, participant #10).
## Substance Use
Participants described high rates of substance use in the SGM community. In particular they identified substance use as a maladaptive coping mechanism for depression, discrimination, and material insecurity. They observed that in the long-term, substance use makes most people's situation worse. As one participant indicated, "Substance use is very common in the community. Some people think it is a way to cure depression, but it's killing us faster than the depression will even kill us" (FGD1, participant #9). Another participant explained a link with police violence: "When it comes to the brutality from the policemen, many people have been, they have been shaken out of their life. So, when they come out from jail, they now start taking drugs to be showing they are masculine, that they are no longer disturbed" (FGD1, participant #8). Another participant made the explicit link between different syndemic factors: "When you are discriminated, you get depressed. Then, when you are discriminated, you fall down to using drugs to make yourself feel OK" (FGD2, participant #8).
Not only did participants readily make connections between syndemic factors through direct pathways described above, but they also sometimes saw connections that reversed the order in which syndemic factors connected. For example, substance use heightening material insecurity, or material insecurity facilitating discrimination. One participant recounted how material insecurity necessitated he and his partner share a single room, which made it easier for the neighbors to hear them when they fought, and in turn to call the police on them (FGD1, unidentified participant). In particular, multiple participants indicated that money allowed freedom of expression and freedom from the scrutiny of police and society more generally. Participants from the TGW FGD were more likely to highlight the freedom associated with material security. Money in particular made it possible to pay bribes requested by police: "Most of the police that accept bribe, when you don't have money, you will not be able to cover everything up" (FGD3, participant #1). Money also provided recourse in the event of familial repudiation: "When my family members find out who I am, what covers me up [protects me] was that I am not depending on them.. If tomorrow, I leave the family and go to another environment, I can feed and take care of myself. I can defend myself and I will focus my life on the way I want to live" (FGD3, unidentified participant). As another participant summarized the relationship, "For our nobody's willing to increase your pay -you work and not get paid, and that's how it is. But.. then children come in.. and things are not there in place to take care of everything, you start breaking down, mentally" (FGD1, participant #9). Someone else noted, "For the Naira and depression, for somebody like me, it fucked with my mental health. If I'm broke, like it.. gives me this boredom of staying indoors.. I don't want to see anybody because I don't have money to go out" (FGD2, participant #10). Or, as another participant put it, "Without having finance, you cannot be able to live the life you want to live, and then will lead into depression" (FGD3, participant #2).
## Discrimination
Participants' accounts demonstrated multi-faceted experiences with discrimination related to both sexual orientation and gender identity, and associated discriminatory experiences with depression and other mental health issues, including anxiety. Social norms writ large, often enacted by family members as well as religious leaders, allowed for the articulation of discrimination. Participants particularly noted the challenge of meeting expectations around marriage to a woman. "'Who is your girlfriend?' is a social norm in our society. Like, when you get to a certain age, and the next question is that, 'Ah, what's up? We have not seen you with any girl!' And before you know it, you will be forced to have a girlfriend.. you're starting to force yourself to do those things you do not want to do, and breaking down already" (FGD1, participant #6). As someone else described it, "Growing up in a very homophobic environment, where you are judged for basically everything you do, even just for breathing, you will begin to question your life, question everything that you do at that point in time, you will not have anyone to talk to because you're trying to put up this facade of being straight.. Such situations can lead one into depression" (FGD1, participant #5). Focusing on Nigeria's laws criminalizing same-sex marriage, one participant noted, "You can't, like, be with your partner outside. It's not allowed in this country.. And it has caused fear, I mean scared, like not being free to express my own self" (FGD2, participant #4). Another stated, "With that law, anxiety comes in. Because the fear of going to jail, the fear of being arrested, the fear of being molested by police" (FGD2, participant #8). Participants identified specifically how stigma and discrimination stemming from religion could lead to depression. For example, "When you feel you are not really doing the will of God, you just want to be on yourself, you isolate yourself from friends of the [SGM] community. I think it affects us mentally and it gets us into depression" (FGD2, participant #8). As someone else put it, "Society causes depression because when you're not treated fairly, partner violence in the identified syndemic, which identified positive relationships between intimate partner violence and the other components of the syndemic.
Although discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other violence can independently result in greater vulnerability to HIV, these factors co-occur in ways that amplify one another and subsequent behaviors that increase the risk of HIV acquisition [16,27,28]. In our focus groups, participants described how material insecurity can lead to depressive symptoms, which leads to various forms of substance use as a means of coping. Participants noted discrimination based on SGM identity as especially impactful, particularly when expressed by family members, church communities, and law enforcement. This discrimination led to feeling hopeless and depressed, with subsequent substance use. Many of these findings highlight the importance of structural factors on individual-level components of the syndemic. Community and societal-level systems, such as churches and police, have a direct impact on individual-level behaviors and experiences through systematic discrimination [11]. The range of discrimination experiences from many sources, which functioned differently in their day-to-day lives, supports keeping these multiple sources of discrimination as distinct components of the syndemic. These factors are especially relevant in the context of Nigeria, where discrimination against SGM people is culturally acceptable and explicitly coded into law [11,29]. Thus, addressing this syndemic, and the vulnerability to HIV acquisition and other health issues it creates, requires systemic changes to both Nigerian cultural norms and laws, as well as other efforts to reduce discriminatory actions by religious leaders, healthcare workers, and the police.
Overall, our study highlights the importance of developing structural and individual interventions that target syndemic components that co-occur, such as material insecurity and depression, or depression and substance use. For example, an intervention addressing material insecurity and depression could include depression screening and support groups [30]
## Co-Prevalence and Sensitivity Analyses
Using the quantitative data to examine co-prevalence of pairwise syndemic components identified in our qualitative analyses, nearly every single pairwise co-prevalence was significantly greater than expected (Table 5). However, compared to the quantitative findings, qualitative discussions did not mention intimate partner violence as much as may have been expected. Additionally, our sensitivity analyses collapsing substance use into one factor, and discrimination into a single factor, did not change results in a substantial way (less than a 5% difference in regression estimates, with associations remaining statistically significant).
## Discussion
We identified a syndemic characterized by discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other violence among SGM in Abuja, Nigeria. Our quantitative analyses found strong relationships between this syndemic and positive HIV status, both before and after adjustment for sociodemographic characteristics. Our qualitative findings provided further evidence for relationships between components of the syndemic, with participants describing how these components interconnected in their lived experiences to impact their health and well-being. Overall, our study aligns with much of the literature demonstrating the relevance of a syndemic framework to understanding HIV-related health outcomes among SGM [6,8,16,27,28]. A unique exception was the limited discussions of intimate partner violence in the FGD. This absence does not necessarily indicate the absence of such violence, as it is likely due to the heavily vulnerable and stigmatized nature of this topic, making it extremely difficult to raise in the presence of others in a FGD. But, the quantitative analysis supports the relevance of intimate sensitive nature of the topics discussed, social desirability is likely to have resulted in some underreporting, particularly in the FGD, where experiences are shared among peers. Despite the risk of social desirability, participants shared many vulnerable experiences, in part due to a strong rapport with one another, the data collectors, and the TRUST clinic.
## Conclusion
We found strong evidence of a syndemic of discrimination, material insecurity, depression, substance use, intimate partner violence, and police and other violence among SGM in Abuja, Nigeria. Using a mixed-methods approach, we found statistically significant associations between our syndemic index and positive HIV status, and provided nuanced descriptions of how the component factors of the syndemic interrelate to lead to adverse health outcomes among this population. Discrimination, material insecurity, and depression are especially salient to the lived experiences of SGM in Abuja, Nigeria. This multilevel syndemic has a significant impact on the physical, mental, and emotional health of this community. We recommend future research assessing this syndemic overall and temporal relationships with additional health outcomes in this population. Similarly, future research exploring this syndemic among TGW in Nigeria will further help inform health equity efforts for this community. Overall, our findings highlight important needs and disparities that must be addressed not simply at the individual level, but through larger scale social and structural change efforts.
imperative to continue resourcing SGM-friendly health services for both sexual health and broader health as well as other mechanisms to improve their wellbeing. Changing Nigeria's criminal code to decriminalize same-sex sex and/or repealing the 2013 Same-Sex Marriage (Prohibition) Act would beneficially alter the social and legal structure, removing symbolic and legal support for discrimination. Nigerian groups have attempted such legal efforts, albeit with minimal success [31]. Training to reduce discriminatory behavior among police officers would also be beneficial [32], as would increased pay for police officers to reduce extortion and bribery of SGM. Our study has several strengths that bolster the significance of our findings. We utilized a mixed methods approach that triangulated strong quantitative relationships among our syndemic components and HIV status with qualitative themes that described these social and structural interrelationships. The use of a mixed methods approach provides additional qualitative nuance and description that our quantitative findings alone would not achieve. We also utilized a large range of factors, most of which include multidimensional items, capturing a broad set of relevant syndemic components. Finally, our study fills an important gap in the literature on how a multi-level syndemic is associated with HIV status and health seeking behavior among SGM in Abuja, Nigeria, one of the largest cities in Nigeria.
Our findings also have important limitations to consider. Our study population is SGM living in Abuja, Nigeria, so our findings may not be fully generalizable outside of this context. This focus is justified however, given both the disproportionate vulnerability to HIV among SGM, and the greater relevance of some of our syndemic components in Nigeria, particularly discrimination towards SGM, which is both legally and culturally enforced in Nigeria. The relatively small number of TGW in our sample limits our understanding of the identified syndemic in this population, indicating an important area for future research. Some factors, such as stigma and discrimination, had much more granular information than our single-item measures, such as intimate partner violence. There are additional factors of interest related to HIV acquisition in this population, such as transactional sex, that we were unable to include in our analyses but are important for future research. Finally, given the
## References
1. Hall, Newcomb, Dyar et al. (2022) "Patterns of polyvictimization predict stimulant use, alcohol and marijuana problems in a large cohort of sexual minority and gender minority youth assigned male at birth" *Psychol Addict Behav*
2. Turpin, Dyer, Dangerfield et al. (2020) "Syndemic latent transition analysis in the HPTN 061 cohort: prospective interactions between trauma, mental health, social support, and substance use" *Drug Alcohol Depend*
3. Leblanc, Crean, Dyer (2021) "Ecological and syndemic predictors of drug use during sex and transactional sex among U.S. Black men who have sex with men: a secondary data analysis from the HPTN 061 study" *Arch Sex Behav*
4. Ramadhani, Ndembi, Nowak (2018) "Individual and network factors associated with HIV care continuum outcomes among Nigerian MSM accessing health care services" *J Acquir Immune Defic Syndr*
5. Charurat, Emmanuel, Akolo (2015) "Uptake of treatment as prevention for HIV and continuum of care among HIV-positive men who have sex with men in Nigeria" *J Acquir Immune Defic Syndr*
6. Kroenke, Spitzer, Williams (2001) "The PHQ-9: validity of a brief depression severity measure" *J Gen Intern Med*
7. Sheikomar, Dean, Ghattas et al. (2021) "Validity of the food insecurity experience scale (FIES) for use in league of Arab States (LAS) and characteristics of food insecure individuals by the human development index (HDI)" *Curr Dev Nutr*
8. Wiatrek, Zlotorzynska, Rai et al. (2018) "The annual American men's internet survey of behaviors of men who have sex with men in the United States: key indicators report"
9. (2023) "SAS/STAT® 15.3 user's guide"
10. Nigeria, Nbos (2022) "The 2022 Multidimensional Poverty Index" *Statistics NBo*
11. Okereke, Onyeneke, Nnamani et al. (2024) "Nigeria's fossil fuel subsidy reforms: the welfare effects on households"
12. Robinson, Ndembi, Aka et al. (2021) "Financial Insecurity and HIV Outcomes among Men who have Sex with Men and Transgender Women in Nigeria. Population Association of America Annual Meeting"
13. Walters, Braksmajer, Coston (2020) "A syndemic model of exchange sex among HIV-positive men who have sex with men" *Arch Sex Behav*
14. Godley, Adimora (2020) "Syndemic theory, structural violence and HIV among African-Americans" *Curr Opin HIV AIDS*
15. Kalu, Ross, Taegtmeyer et al. (2024) "Association of same-sex criminalisation laws and national HIV policies with HIV testing in African MSM: an ecological single-level and multilevel cross-sectional study of sub-Saharan African countries" *Sex Transm Infect*
16. Barry, Threats, Blackburn (2018) "Stay strong! Keep Ya head up! Move on! It gets better!!!! Resilience processes in the healthmpowerment online intervention of young black gay, bisexual and other men who have sex with men" *AIDS Care*
17. Jjuuko (2020) "Strategic litigation and the struggle for lesbian, gay and bisexual equality in Africa"
18. Giwa, Logie, Karki et al. "Police violence targeting LGBTIQ + people in Nigeria: advancing References"
19. Onovo, Adeyemi, Onime (2023) "Estimation of HIV prevalence and burden in Nigeria: a bayesian predictive modelling study" *EClinicalMedicine*
20. Bassey, Miteu (2023) "A review of current trends in HIV epidemiology, surveillance, and control in Nigeria" *Ann Med Surg*
21. Onovo, Kalaiwo, Katbi et al. (2021) "Geographical disparities in HIV seroprevalence among men who have sex with men and people who inject drugs in Nigeria: exploratory spatial data analysis"
22. Eluwa, Adebajo, Eluwa et al. (2019) "Rising HIV prevalence among men who have sex with men in Nigeria: a trend analysis" *BMC Public Health*
23. Tiamiyu, Lawlor, Hu (2020) "HIV status disclosure by Nigerian men who have sex with men and transgender women living with HIV: a cross-sectional analysis at enrollment into an observational cohort"
24. Houang, Kafka, Choi (2023) "Co-occurring epidemic conditions among Southern U.S. Black men who have sex with men in an online eHealth intervention" *AIDS Behav*
25. Chandler, Bukowski, Matthews (2020) "Examining the impact of a psychosocial syndemic on past six-month HIV screening behavior of black men who have sex with men in the United States: results from the POWER study" *AIDS Behav*
26. Boyd, Martinez, Mammadli et al. (1007) "Intersecting epidemics: examining the impact of internalized homophobia and depression symptoms on HIV testing through a suicide syndemic among young black men who have sex with men. J Racial Ethn Health Disparities"
27. Singer, Bulled, Ostrach et al. (2017) "Syndemics and the biosocial conception of health" *Lancet*
28. Ochonye, Emmanuel, Abang (2024) "Prevalence and factors associated with psychological distress among key populations in Nigeria" *PLoS One*
29. Lyons, Rwema, Makofane (2023) "Associations between punitive policies and legal barriers to consensual samesex sexual acts and HIV among gay men and other men who have sex with men in sub-Saharan africa: a multicountry, respondentdriven sampling survey" *Lancet HIV*
30. Leevan, Hu, Mitchell (2022) "Associations of gender identity with sexual behaviours, social stigma and sexually transmitted infections among adults who have sex with men in Abuja and Lagos" *Nigeria. J Int AIDS Soc*
31. Dirisu, Eluwa, Callens (2024) "I take the drugs… to make the sickness to move out of me': key populations' and service provider perspectives about facilitators and barriers to ART adherence and retention in care in Nigeria"
32. Quinn, Walsh, Difranceisco (2024) "The inherent violence of anti-Black racism and its effects on HIV care for black sexually minoritized men" *J Urban Health* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846366&blobtype=pdf | # High-Accuracy Serodiagnosis of African Swine Fever Using P72 and P30-Based Lateral Flow Assays: A Validation Study with Field Samples in Thailand
Nitipon Srionrod, Supphathat Wutthiwitthayaphong, Teera Nipakornpun, Sakchai Ruenphet
## Abstract
African Swine Fever is a highly destructive disease devastating pig populations and causing severe economic damage globally. A major challenge in controlling its spread is the reliance on slow, expensive laboratory tests, which delay critical containment efforts. This study aimed to develop and validate simple, rapid "strip tests" (lateral flow assays) that could quickly detect antibodies to the virus at the farm. We created three test prototypes, each targeting a different viral protein (P72, P30, or P54), and tested them against 143 pig serum samples from Thailand. Our results showed that the rapid test targeting P72 was perfectly accurate, matching the complex laboratory test in every case. The P30 test was also found to be highly reliable, while the P54 test proved unsuitable due to a high number of false positive results. We conclude that the P72 and P30 rapid tests are excellent, lowcost tools for surveillance. Their value lies in allowing veterinarians and farmers to obtain accurate results "pen-side" in under 20 min to identify animals with past exposure, enabling immediate action to control outbreaks and protect the pork industry.
## 1. Introduction
African Swine Fever (ASF) is one of the most formidable and economically devastating transboundary diseases affecting the global swine industry. Its causative agent, African Swine Fever Virus (ASFV), is a large, complex, enveloped double-stranded DNA virus and the sole member of the Asfivirus genus within the Asfarviridae family [1][2][3]. Historically confined to sub-Saharan Africa-where the disease was first described in Kenya in 1921 and maintained within a sylvatic cycle involving warthogs and Ornithodoros soft ticks-the global distribution of ASF shifted dramatically in 2007 following the virus's introduction into Georgia. This event marked the onset of an inexorable pan-continental spread [4]. Since 2018, the world has experienced an unprecedented epizootic characterized by rapid dissemination across Asia, particularly in China, Vietnam, and the Philippines, alongside persistent establishment in wild boar populations in Eastern and Central Europe [5]. This expansive spread has caused catastrophic economic losses, leading to the death or culling of hundreds of millions of pigs and posing serious threats to global pork production and food security [6]. The recent re-emergence of ASFV in the Americas-specifically in the Dominican Republic and Haiti in 2021 after nearly four decades of absence-further underscores the virus's relentless transboundary potential and its ongoing threat to all swine-producing nations [7]. Consequently, ASF is designated as a notifiable disease by the World Organisation for Animal Health (WOAH), and its control remains an urgent global priority [8].
The control of ASF is remains exceptionally challenging, largely due to the lack of a globally available, safe, and effective vaccine [9]. In addition, the virus is remarkably resilient, capable of surviving for prolonged periods in the environment, on contaminated fomites, and in various pork products. Transmission is multifactorial, occurring through direct contact with infected domestic pigs, ingestion of contaminated feed, interaction with wild boar reservoirs, and persistence in survivor or carrier animals [5]. In the current absence of effective prophylactic or therapeutic tools, ASF control strategies depend almost entirely on "stamping-out" policies. These measures require rapid and early detection, strict on-farm biosecurity, movement restrictions, and the culling of affected and at-risk herds [10]. Consequently, timely and accurate diagnostics constitute the most critical pillar in the global response to ASF. Diagnostic approaches fall into two primary categories. The first is viral nucleic acid detection-most commonly performed using real-time quantitative PCR (qPCR), which remains the gold standard for identifying acutely infected animals during the viremic phase [11,12]. The second category includes WOAH-prescribed serological assays, such as ELISA and the immunoperoxidase test (IPT), which are used to identify animals that have mounted an antibody response [13].
While molecular detection by qPCR is essential for confirming acute outbreaks, serology remains indispensable for broader surveillance objectives, including proof-of-freedom certification, detection of chronic carriers or survivors, and monitoring low-virulence strains that may not produce overt clinical signs [14]. Despite their diagnostic value, both qPCR and conventional ELISA/IPT share a major operational limitation: they are laboratory-bound and time-consuming. Specifically, they require centralized facilities, expensive analytical equipment (such as thermal cyclers, plate readers, and washers), stable cold-chain logistics for reagent storage, and personnel with specialized technical training. Dependence on this infrastructure creates a diagnostic lag that often results in delays of several hours to days before results become available from the field. Such delays impede rapid containment efforts and are especially problematic in remote rural regions or in countries with limited veterinary diagnostic capacity [15]. Collectively, these limitations underscore the need for point-of-care (POC) diagnostic tools that are rapid, simple, cost-effective, and sufficiently robust for field or "pen-side" use without laboratory support. Lateral flow assays (LFAs), also referred to as immunochromatographic assays, meet these criteria by delivering results in under 20 min, requiring minimal operator training, and remaining stable at ambient temperatures [16].
In response to this need, numerous rapid LFAs for ASFV have been developed and commercialized [17,18]. Among the factors that influence their diagnostic performance, the most critical is the choice of recombinant antigen used to capture antibodies from the sample, as this directly determines test sensitivity and specificity. The ASFV proteome contains several highly immunodominant proteins that are widely targeted for serological assays. The most commonly used is the major capsid protein p72 (B646L), which is highly conserved, strongly immunogenic, and elicits a robust and persistent antibody response, making it a benchmark antigen in ASF serological assays [19]. In addition to P72, the p30 protein (CP204L) represents an important antigenic target; antibodies against p30 develop 2-4 days earlier than those against p72, positioning it as a valuable marker for early seroconversion [20,21]. A third widely recognized antigen is the p54 protein (E248R), an external envelope protein that is likewise highly immunogenic [22]. Although these antigens have been incorporated into duplex LFAs [23,24] and multi-antigen chimeric ELISAs [25], rigorous side-by-side comparisons of single-antigen LFAs remain limited. Consequently, it is still uncertain whether p54 contributes additional sensitivity or is susceptible to non-specific reactions in rapid assays, and whether a p30-based LFA performs as reliable as a p72-based LFA in detecting established infections. Clarifying the individual diagnostic contribution of each antigen is therefore critical for designing an optimal LFA. To address this gap, the present study developed and systematically evaluated three inhouse single-antigen LFAs targeting p30, p54, and p72. By comparing their diagnostic sensitivity, specificity, and overall accuracy against a widely used commercial multi-antigen ELISA, this study aims to identify the most reliable antigen target for rapid, high-accuracy serological surveillance of ASFV.
## 2. Materials and Methods
## 2.1. Ethical Approval
The animal procedures were approved by the Animal Research Ethics Committee of the Faculty of Veterinary Medicine, Mahanakorn University of Technology, Thailand (Approval code: ACUC-MUT-2024/004).
## 2.2. Sample Collection and Characterization of the Study Population
A total of 143 swine serum samples were utilized as the testing panel for this study. To ensure the assessment of the LFAs was relevant to real-world surveillance scenarios in endemic regions, samples were collected from commercial swine farms located in high-risk areas of Thailand that had experienced sporadic ASF outbreaks. The study population was strictly limited to sows and replacement gilts (aged >6 months). This age group was selected to eliminate the potential interference of maternally derived (colostral) antibodies, which can confound serological results in piglets or weaners. Fattening pigs were excluded to focus on the breeding herd, which represents the long-term reservoir potential on farms.
At the time of sampling, all 143 animals were clinically healthy and showed no signs suggestive of ASF, such as fever, hemorrhage, or anorexia. To determine infection status, all samples were tested for ASFV DNA using a qPCR assay, and all were confirmed to be PCR-negative (Ct > 40). Following qPCR screening, the samples were further analyzed using a commercially available ELISA kit to determine their antibody status. Consequently, the seropositive animals identified in this study (n = 64) likely represent a population with prior exposure to ASFV. Although definitive classification would require longitudinal follow-up, this serological profile is consistent with convalescent animals or those that have recovered from infection, rather than acutely infected carriers. This completes a set of 143 well-characterized samples that served as the reference panel. All samples were obtained from the Virology and Molecular Diagnostic Center, Faculty of Veterinary Medicine, Mahanakorn University of Technology, Thailand, and were stored at-30 • C until testing (Figure 1). Schematic overview of the study design and diagnostic pipeline. One hundred forty-three swine serum samples, all confirmed negative for African swine fever virus (ASFV) by quantitative polymerase chain reaction (qPCR), were subjected to parallel comparative analysis. Samples were tested using the ID Screen ® African Swine Fever Indirect Antibody Test (ID Screen ® , ID Vet, Grabels, France) and in-house lateral flow assays (LFAs) specific for p30, p54, and p72 antibody detection. The diagnostic performance of each LFA was statistically compared against the ELISA results to evaluate sensitivity, specificity, accuracy, and concordance.
## 2.3. Assessment of Analytical Specificity and Cross-Reactivity
To evaluate the analytical specificity of the developed LFAs, the assays were tested against serum samples positive for antibodies to other economically important swine pathogens that could potentially cause cross-reactivity or confound clinical diagnosis. The panel included sera positive for antibodies to Classical swine fever virus (CSFV), Porcine reproductive and respiratory syndrome virus (PRRSV), Foot-and-mouth disease virus (FMDV) serotype A, Porcine circovirus type 2 (PCV-2), and Pseudorabies virus (PRV-gB). These samples were obtained from the archive of the Virology and Molecular Diagnostic Center, Faculty of Veterinary Medicine, Mahanakorn University of Technology.
## 2.4. Quantitative Polymerase Chain Reaction
This protocol adhered to the methodological framework described in our previous study [26]. Briefly, viral DNA was extracted from all samples using the TAN Bead ® Nucleic Acid Extraction Kit (Taiwan Advanced Nanotech, Taoyuan, Taiwan) in conjunction with the Automated Nucleic Acid Extractor (Smart LabAssist SLA-E13200, Taoyuan, Taiwan). Following DNA extraction, quantitative detection of ASFV DNA was performed using the Virotype ® ASFV 2.0 PCR Kit (Indical Bioscience, Leipzig, Germany) on a C1000 Touch Thermal Cycler (Bio-Rad, Hercules, CA, USA). The assay incorporated two internal controls-an endogenous β-actin control and an exogenous control introduced during DNA purification-to ensure both extraction efficiency and amplification reliability. The qPCR cycling conditions were as follows: an initial denaturation at 95 • C for 2 min, followed by 40 cycles of 95 • C for 5 s and 60 • C for 30 s. Sample results were interpreted according to the manufacturer's recommended cycle threshold (Ct) cutoffs: positive when Ct < 35, suspect when 35-40, and negative when >40.
## 2.5. Enzyme-Linked Immunosorbent Assay
All 143 swine serum samples were analyzed using a commercial indirect ELISA kit (ID Screen ® , ID Vet, Grabels, France), which served as the reference method for this study. The kit's microplate wells are coated with recombinant ASFV proteins p32, p62, and p72. The assay was performed following the manufacturer's instructions. Briefly, all reagents and samples were equilibrated to room temperature (21 • C ± 5 • C). A total of 190 µL of Dilution Buffer was added to each well, followed by 10 µL of the Negative Control, Positive Control, or each test serum sample. The plate was then covered and incubated for 45 ± 4 min at 21
After incubation, the wells were emptied and washed three times with at least 300 µL of 1× Wash Solution. Subsequently, 100 µL of 1× anti-multi-species HRP conjugateprepared by diluting the 10× concentrate 1:10 in Dilution Buffer-was added to each well, and the plate was incubated for an additional 30 ± 3 min at 21 • C (±5 • C). A second washing step identical to the first was then performed.
To initiate the colorimetric reaction, 100 µL of Substrate Solution was added to each well, and the plate was incubated in the dark for 15 ± 2 min at 21 • C (±5 • C). The reaction was stopped by adding 100 µL of Stop Solution (0.5 M acid), and the optical density (OD) of each well was immediately measured at 450 nm using a microplate reader (EUROIMMUN Analyzer I-2P, EUROIMMUN US, Inc., Mountain Lakes, NJ, USA).
The assay run was considered valid when the mean Positive Control OD (ODpc) exceeded 0.350 and the ODpc/ODnc ratio was greater than 3. For each sample, a sampleto-positive (S/P) percentage was calculated using the following formula:
Samples with S/P% ≥ 40% were classified as positive, S/P% ≤ 30% as negative, and values between 30% and 40% were considered doubtful.
$$• C (±5 • C).$$
$$S/P% = [(ODsample -ODnc)/(ODpc-ODnc)] × 100$$
## 2.6. Lateral Flow Assay
The in-house LFA for ASFV antibody detection was developed and optimized based on a double recognition (indirect sandwich) assay principle. The finalized prototypes were manufactured by Pacific Biotech Co., Ltd. (Petchaboon, Thailand) under a controlled manufacturing process to ensure reproducibility. Briefly, the strip components were prepared as follows. For the Test line (T line), recombinant p30, p54, or p72 antigen was diluted in Tris-HCl (pH 8.5) containing sucrose and dispensed onto a nitrocellulose membrane. For the Control line (C line), a monoclonal antibody (MAb) specific to a non-relevant control protein was diluted in 20 mM Tris-HCl (pH 7.5) with sucrose and dispensed parallel to the T line on the same membrane. Detector reagents were prepared by coupling proteins to colloidal gold nanoparticles. Two separate conjugations were performed to allow differential detection: recombinant p30, p54, or p72 antigen was conjugated to one batch of nanoparticles, while the control protein recognized by the C-line MAb was conjugated to a second batch.
The LFA strips were assembled by affixing the nitrocellulose membrane, conjugate pad, absorbent pad, and sample pad onto an adhesive backing card with appropriate overlaps (Figure 2). For the test procedure, 10 µL of serum or 20 µL of whole blood was applied to the sample window. After absorption, five drops (approximately 150 µL) of running buffer were added to the sample window, and the results were read after 10 min. A valid test was indicated by a visible signal at the C line, confirming proper sample migration and correct reagent functionality, as the gold-conjugated control protein was captured by the immobilized Mab (Figure 3).
## 2.7. Statistic Analysis
The diagnostic performance of the three in-house LFAs targeting P30, P54, and P72 was evaluated against a commercial indirect ELISA, which was considered the reference standard. The dichotomous outcome (positive/negative) from each LFA and the ELISA for all 143 serum samples were organized into 2 × 2 contingency tables, from which the numbers of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) were obtained for each LFA prototype (Table 1).
Table 1. Performance of antibody detection against P30, P54, and P72 of African swine fever virus using lateral flow assays (LFAs) compared to an enzyme-linked immunosorbent assay (ELISA). In addition, the positive predictive value (PPV), also referred to as precision, was calculated using the formula [TP / (TP + FP) × 100] to estimate the likelihood that a positive LFA result represented a true positive. The 95% confidence intervals (CIs) for all diagnostic estimates were also computed.
## ELISA
To further assess the agreement between the LFAs and the reference method, two statistical tests were performed. Inter-rater concordance was evaluated using Cohen's Kappa (κ), which measures the degree of agreement beyond chance. κ values were interpreted using standard criteria, where κ = 1.0 indicates perfect agreement and values > 0.81 indicate almost perfect agreement. McNemar's test for paired nominal data was used to analyze discordant classifications (FP and FN) and to determine whether a significant difference or systematic bias existed between the LFAs and the ELISA. A p-value < 0.05 was considered statistically significant.
## 3. Results
## 3.1. Overall Sample Analysis
A total of 143 swine serum samples were included in the comparative analysis. All samples were confirmed negative for ASFV genetic material by qPCR prior to serological testing. The reference indirect ELISA was then used to determine the antibody status of these samples. Among them, 64 samples tested positive and 79 tested negatives for ASFV antibodies. Based on the ELISA results, the samples were subsequently tested using three prototype LFAs targeting the ASFV P30, P54, and P72 proteins, respectively.
## 3.2. Assay Performance and Contingency Analysis
The comparative performance of the three LFAs against the reference ELISA is summarized in the 2 × 2 contingency tables (Table 1). Detailed individual sample results are provided in the Supplementary Materials (Tables S1-S3).
The P72-based LFA demonstrated perfect concordance with the ELISA, correctly identifying all 64 positive samples (true positives, TP) and all 79 negative samples (true negatives, TN), with no false positives (FP = 0) or false negatives (FN = 0).
The P30-based LFA correctly identified all 64 positive samples (TP) and 78 of the 79 negative samples (TN). A single serum sample (Sample 13) produced a false-positive result (FP = 1), while no false negatives were observed (FN = 0).
The P54-based LFA correctly identified all 64 positive samples (TP); however, nine false positives were recorded (FP = 9), and 70 of the 79 negative samples were correctly classified (TN). No false negatives were observed (FN = 0).
## 3.3. Diagnostic Parameters and Statistical Agreement
The diagnostic sensitivity, specificity, accuracy, and precision (positive predictive value, PPV) for each assay were calculated from the contingency tables, and statistical agreement was further evaluated using Cohen's Kappa (κ) and McNemar's test (Table 2). The LFA targeting P72 exhibited perfect performance, with 100% diagnostic sensitivity (95% CI: 94.4-100), 100% specificity (95% CI: 95.4-100), and 100% accuracy (95% CI: 97.5-100). Agreement with ELISA was perfect, with a Kappa value of 1.0.
The LFA targeting P30 also showed high performance, with 100% diagnostic sensitivity (95% CI: 94.4-100), 98.7% specificity (95% CI: 93.2-99.9), and 99.3% accuracy (95% CI: 96.4-99.9). Agreement with ELISA was classified as 'Almost Perfect' (κ = 0.9859), and McNemar's test indicated no statistically significant difference between the P30 LFA and ELISA results (p > 0.999).
In contrast, the LFA targeting P54 achieved 100% diagnostic sensitivity (95% CI: 94.4-100) but had a lower specificity of 88.6% (95% CI: 79.4-94.7), resulting in an overall accuracy of 93.7% (95% CI: 88.5-96.9). Although the Kappa value indicated 'Almost Perfect' agreement (κ = 0.8745), McNemar's test revealed a statistically significant difference (p = 0.0039) compared with ELISA, primarily due to the high rate of false positive results.
## 3.4. Cross-Reactivity Analysis
The analytical specificity of the P30, P54, and P72 LFAs was assessed using serum samples positive for antibodies to other common swine pathogens, including CSFV, PRRSV, FMDV serotype A, PCV-2, and PRV-gB. No cross-reactivity was detected, as all three LFA prototypes yielded negative results for every heterologous antibody tested. These findings demonstrate that the developed LFAs exhibit high analytical specificity for ASFV antibodies and do not cross-react with antibodies to these major swine viral pathogens.
## 4. Discussion
The global persistence of ASF underscores the need for rapid, accurate, and fielddeployable serological diagnostics, which are crucial for effective disease management. Such tools are indispensable for surveillance, monitoring control measures, and certifying disease-free status, and complement early molecular detection methods [1]. Beyond antigenspecific performance, LFAs, as demonstrated by our P72 and P30 prototypes, offer key operational advantages over conventional serological methods such as ELISA or IPT. While laboratory-based assays require specialized equipment (e.g., plate readers, washers), coldchain logistics, and trained personnel, LFAs are designed for field deployment. They are rapid, require minimal training for "pen-side" application, and remain stable at ambient temperatures. Another practical benefit of LFAs is their single-sample format. Unlike ELISA, which is most economical when run in batches (e.g., 96-well plates), an LFA can test a single, high-priority animal immediately and cost-effectively. This capability is crucial for rapid screening in smallholder settings, outbreak investigations, or movement control, providing actionable results in under 20 min rather than hours or days. In this study, we evaluated the diagnostic performance of three single-antigen LFAs targeting ASFV P30, P54, and P72, comparing them against a commercial multi-antigen indirect ELISA that targets recombinant P32, P62, and P72 proteins.
The LFA targeting P72 demonstrated perfect sensitivity and specificity (100%) and achieved complete agreement (κ = 1.0) with the reference ELISA, highlighting its exceptional diagnostic performance. This high performance was, to some extent, anticipated given the biological characteristics of the P72 protein. P72, encoded by the B646L gene, is the major, highly conserved capsid protein and the primary immunodominant antigen in ASFV infection [25,30,31]. Consequently, it serves as the benchmark target for many recommended serological tests, including the OIE-recognized immunoperoxidase test (IPT) [13]. Importantly, our single-antigen P72 LFA performed identically to the multiantigen (P32/P62/P72) commercial ELISA. This indicates that, for the 143 samples in this panel, the anti-P72 antibody response was sufficiently robust and persistent to correctly identify all positive cases. Notably, the inclusion of P32 and P62 antigens in the ELISA, also known immunogenic proteins [32,33], provided no additional sensitivity in this dataset.
Furthermore, our single-antigen P30 LFA demonstrated excellent performance, with 100% sensitivity and 98.7% specificity. The single false positive (Sample 13) represents an acceptable discrepancy for a rapid screening test. These results are supported by the 'Almost Perfect' Kappa agreement (κ = 0.9859) and a non-significant McNemar's test (p > 0.999). Biologically, the P30 protein (CP204L gene) is a well-established early diagnostic marker, with antibodies often appearing 2-4 days earlier than those against P72 [18,23,34]. Notably, our P30 LFA detected all 64 positive samples that were also identified by the P32/P62/P72-based ELISA. This concordance suggests that the samples were not in the very early seroconversion window (e.g., 7-10 days post-infection), a period when P30 might be the only detectable antibody [35]. Therefore, the dataset likely represents animals in the mid-to-late or chronic stages of infection, during which robust antibody populations against both P30 and P72 co-exist [36].
Conversely, the P54-based LFA was determined to be unsuitable for diagnostic use. Despite achieving 100% sensitivity, its specificity was only 88.6%. This resulted in nine false positives, which are unacceptable for any reliable surveillance program. This significant discrepancy (McNemar's p = 0.0039) underscores a critical limitation of the assay. Although the P54 protein (E248R gene) is immunogenic [37], it may be prone to non-specific binding or cross-reactivity when used as a standalone target in a rapid lateral flow format, a limitation also noted in other studies [34]. While P54 has been used effectively in multiantigen or chimeric ELISA formats [25,38], these results demonstrate its unreliability as a single target in an LFA.
A critical observation in this study was the high seroprevalence (44.7%, 64/143) identified within a PCR-negative, clinically healthy population. The presence of PCRnegative but seropositive animals in endemic regions represents a complex epidemiological scenario. Although such animals are often referred to in the field as "survivors," this classification should be interpreted with caution. These animals show evidence of prior exposure and have mounted an immune response, yet they lacked detectable viral DNA in serum at the time of sampling. This profile clearly differs from that of acutely infected animals, which are typically qPCR-positive. Although seropositive animals may not contribute substantially to viral spread during their non-viremic phase, they serve as important sentinels indicating prior herd exposure to the virus.
It is crucial to consider the infection timeline when interpreting these results. The LFAs validated in this study are designed to detect antibodies, which typically become detectable 7-10 days post-infection. Accordingly, these assays are not suitable for identifying early, acute infections during the viremic phase prior to seroconversion. Consequently, these LFAs should not replace qPCR but should instead be used as complementary diagnostic tools. In a comprehensive surveillance program, qPCR is essential for detecting early outbreaks, whereas LFAs are valuable for retrospective screening and assessing herd immune status.
In designing the study, we specifically excluded piglets and weaners to eliminate the confounding effects of maternal antibodies, which can persist for several weeks. By restricting the sample set to sows and replacement gilts, we ensured that detected antibodies reflected active immune responses to natural field exposure.
The primary limitation of this study is the exclusion of PCR-positive (viremic) animals, fattening pigs, and time-course sera from experimentally infected individuals. Because the study's objective was to validate the LFA's sensitivity in detecting non-viremic survivorsthe most diagnostically challenging group-the data could not conclusively confirm the expected early-phase detection advantage of P30 over P72. Controlled infection studies are therefore needed to precisely define the temporal diagnostic sensitivity of these assays.
Importantly, a distinction must be drawn between the analytical validation performed in this study and full field deployment. While the LFAs are designed for pen-side use, testing in the present study was conducted under controlled laboratory conditions using stored serum samples. Although the assay format is rapid and requires no equipment, performance under field conditions may be influenced by factors such as extreme environmental temperatures, dust exposure, and interpretation by non-technical personnel. Therefore, additional field testing across diverse environmental conditions is recommended to confirm the practical robustness of these LFAs.
Although the laboratory validation performed here was rigorous, additional field testing across diverse environmental conditions is recommended to confirm the practical robustness of these LFAs. As PCR remains the gold standard for detecting acute or viremic infections, the main utility of LFAs lies in identifying chronic or convalescent cases in which PCR results are negative. Future research should include PCR-positive animals and fattening pigs to correlate LFA reactivity with viral load dynamics and clinical progression.
Finally, the analytical specificity of the LFAs was clearly demonstrated by the absence of cross-reactivity with major swine antibodies, including CSFV, PRRSV, FMDV serotype A, PCV-2, and PRV-gB. This high specificity is particularly critical for differentiating ASF from CSFV, which presents with clinically indistinguishable hemorrhagic signs, ensuring that positive LFA results reliably indicate ASFV exposure.
Based on these findings, these results strongly support the development of a duplex LFA. Such a test, combining P30 and P72 on a single strip as previously described by others [24,39], would leverage the strengths of both markers. Specifically, it would utilize the P30 antigen to detect antibodies in the critical early serological window and the P72 antigen to ensure robust detection of persistent antibodies in mid-to-late stages. This tool would theoretically provide diagnostic coverage superior to our single-antigen prototypes and could match or exceed the utility of the multi-antigen (P32/P62/P72) ELISA in a rapid, field-deployable format.
Finally, the demonstrated diagnostic performance of these assays has important implications for global ASF control strategies. According to the Global Framework for the Progressive Control of Transboundary Animal Diseases (GF-TADs), rapid and decentralized diagnostic tools are essential for effective disease management, particularly in resourcelimited settings [26]. By offering a reliable and low-cost alternative to laboratory-bound assays, the P72 and P30 LFAs described here can strengthen surveillance capacity, facilitate faster field-level response by local veterinarians, and help maintain disease-free zones in endemic regions.
## 5. Conclusions
This study demonstrated high diagnostic performance of single-antigen LFAs targeting ASFV P72 and P30. The P72 LFA exhibited perfect statistical agreement, while the P30 LFA showed almost perfect agreement with a commercial multi-antigen ELISA reference. The P72 LFA proved robust for detecting established antibody responses, whereas the P30 LFA served as a reliable serological marker. In contrast, the P54 LFA was deemed unsuitable for diagnostic use due to unacceptably low specificity. Taken together, these findings indicate that the P72 and P30 LFAs are highly accurate candidates for use in serosurveillance. However, they should be interpreted as assays for past exposure and applied as complementary tools alongside molecular methods for acute infection detection.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vetsci13010004/s1, Table S1. Individual results of antibody detection against African swine fever virus P30 obtained using a lateral flow assay (LFA) and compared with an enzyme-linked immunosorbent assay (ELISA). Table S2. Individual results of antibody detection against African swine fever virus P54 obtained using a lateral flow assay (LFA) and compared with an enzyme-linked immunosorbent assay (ELISA). Table S3. Individual results of antibody detection against African swine fever virus P72 obtained using a lateral flow assay (LFA) and compared with an enzyme-linked immunosorbent assay (ELISA).
Author Contributions: Conceptualization, N.S., S.W., T.N. and S.R.; Formal analysis, N.S., S.W. and S.R.; Funding acquisition, S.W., T.N. and S.R.; Investigation, N.S., S.W. and S.R.; Methodology, N.S., S.W. and S.R.; Project administration, N.S., S.W. and S.R.; Supervision, S.R.; Writing-original draft, N.S., S.W. and S.R.; Writing-review and editing, N.S., S.W., T.N. and S.R. All authors have read and agreed to the published version of the manuscript.
Funding: The present study received partial financial backing from Mahanakorn University of Technology (Thailand) and Bangkok R.I.A. Group (Thailand), through the allocation of funds provided by the Veterinary Research Grant under Contract No. Immune-001-2025.
## Institutional Review Board Statement:
The guidelines concerning the appropriate care and utilization of animals received approval from the Animal Research Ethics Committee of the Faculty of Veterinary Medicine at Mahanakorn University of Technology, Thailand. To substantiate their validation, these guidelines were identified with the specific approval code ACUC-MUT-2024/004. Informed Consent Statement: Signed informed consent was provided by the owner for all procedures related to the case.
## References
1. Sánchez-Vizcaíno, Mur, Gomez-Villamandos et al. (2015) "An Update on the Epidemiology and Pathology of African Swine Fever" *J. Comp. Pathol*
2. Taesuji, Rattanamas, Punyadarsaniya et al. (2021) "In vitro primary porcine alveolar macrophage cell toxicity and African swine fever virus inactivation using five commercially supply compound disinfectants under various condition" *J. Vet. Med. Sci*
3. Sovijit, Taesuji, Rattanamas et al. (2021) "In vitro cytotoxicity and virucidal efficacy against African swine fever using two potassium hydrogen peroxymonosulfate compared to a quaternary ammonium compound under various concentrations, exposure times and temperatures" *Vet. World*
4. Blome, Franzke, Beer (2020) "African Swine Fever-A Review of Current Knowledge" *Virus Res*
5. Sauter-Louis, Conraths, Probst et al. (1717) "African Swine fever in wild boar in Europe-A review" *Viruses*
6. Nguyen-Thi, Pham-Thi-Ngoc, Nguyen-Ngoc et al. (2021) "An assessment of the economic impacts of the 2019 African swine fever outbreaks in Vietnam" *Front. Vet. Sci*
7. Ramirez-Medina, O'donnell, Silva et al. (1090) "Experimental infection of domestic pigs with an African swine fever virus field strain isolated in 2021 from the Dominican Republic" *Viruses*
8. (2016) "Chapter 15.1. Infection with African Swine Fever Virus"
9. (2026) *Vet. Sci*
10. Dixon, Stahl, Jori et al. (2020) "African Swine Fever Epidemiology and Control" *Annu. Rev. Anim. Biosci*
11. Beltrán-Alcrudo, Arias, Gallardo et al. (2017) "African Swine Fever: Detection and Diagnosis-A Manual for Veterinarians"
12. King, Reid, Hutchings et al. (2003) "Development of a TaqMan ® PCR assay with internal amplification control for the detection of African swine fever virus" *J. Virol. Methods*
13. Fernaández-Pinero, Gallardo, Elizalde et al. (2013) "Molecular diagnosis of African swine fever by a new real-time PCR using universal probe library" *Transbound Emerg. Dis*
14. Hong, Moon, Cho et al. (2019) "First Serologic Analysis of Antibodies Against African Swine Fever Virus Detected in Domestic Pig Farms in South Korea from"
15. Gallardo, Soler, Nurmoja et al. (2021) "Dynamics of African swine fever virus (ASFV) infection in domestic pigs infected with virulent, moderate virulent and attenuated genotype II ASFV European isolates" *Transbound Emerg. Dis*
16. Inui, Gallardo, Portugal et al. (2022) "The OIE ASF Reference Laboratory Network's Overview of African Swine Fever Diagnostic Tests for Field Application; World Organisation for Animal Health"
17. Koczula, Gallotta (2016) "Lateral flow assays" *Essays Biochem*
18. Aira, González-García, Martínez-Cano et al. "Simultaneous Detection of Antigen and Antibodies of African Swine Fever in a Novel Combo Lateral Flow Assay. Vaccines 2024"
19. Vu, Le, Jeong et al. (2023) "Establishment of a p30-based lateral flow assay for African swine fever virus detection" *J. Virol. Methods*
20. Kim, Park, Kim et al. (2019) "Complete genome analysis of the African swine fever virus isolated from a wild boar responsible for the first viral outbreak in Korea" *Front. Vet. Sci*
21. Tian, Sun, Wang et al. (2024) "Identification of a novel linear B-cell epitope on the p30 protein of African swine fever virus using monoclonal antibodies" *Virus Res*
22. Neilan, Zsak, Lu et al. (2004) "Neutralizing antibodies to African swine fever virus proteins p30, p54, and p72 are not sufficient for antibody-mediated protection" *Virology*
23. Yang, Miao, Liu et al. "Structure and function of African swine fever virus proteins: Current understanding"
24. Wang, Kim, Kang et al. "Development and evaluation of two rapid lateral flow assays for on-site detection of African swine fever virus"
25. Sastre, Pérez, Costa et al. (2016) "Development of a duplex lateral flow assay for simultaneous detection of antibodies against African and Classical swine fever viruses" *J. Vet. Diagn. Investig*
26. Li, Zhang, Liu et al. (2660) "Recombinant Proteins to Detect Antibodies against African Swine Fever Virus in Pigs" *Viruses*
27. Ruenphet, Srionrod, Nipakornpun et al. (1068) "RNase Hybridization-Assisted Amplification (RHAM) Technology: A High-Sensitivity, Field-Deployable Alternative to Quantitative Polymerase Chain Reaction for the Rapid Detection of African Swine Fever Virus" *Vet. Sci*
28. Suwannachote, Prasitsuwan, Sumalai et al. (1484) "A Comparison of Diagnostic Methods for Feline Leukemia Virus and Feline Immunodeficiency Virus: Immunochromatographic Assay and RNases Hybridization-Assisted Amplification Test Kit Compared to Reverse Transcription Quantitative Polymerase Chain Reaction" *Animals*
29. Prasitsuwan, Suwannachote, Sumalai et al. (2025) "A comparison of diagnostic methods for canine Ehrlichiosis: Microscopy and RNases hybridization-assisted amplification technology compared with the quantitative polymerase chain reaction" *Vet. World*
30. Punyadarsaniya, Taesuji, Rattanamas et al. (1433) "Establishment of an In-House Indirect Enzyme-Linked Immunosorbent Assay to Detect Antibodies Against African Horse Sickness Based on Monovalent and Polyvalent Live Attenuated Vaccines During the First Outbreak in Thailand" *Animals*
31. Yin, Geng, Shao et al. (2022) "Identification of novel linear epitopes in P72 protein of African swine fever virus recognized by monoclonal antibodies" *Front. Microbiol*
32. (2026) *Vet. Sci*
33. Heimerman, Murgia, Wu et al. (2018) "Linear epitopes in African swine fever virus p72 recognized by monoclonal antibodies prepared against baculovirus-expressed antigen" *J. Vet. Diag. Investig*
34. Wu, Lu, Zhu et al. (1939) "Developing an Indirect ELISA for the Detection of African Swine Fever Virus Antibodies Using a Tag-Free p15 Protein Antigen" *Viruses*
35. Shen, Qiu, Luan et al. (2023) "I329L protein-based indirect ELISA for detecting antibodies specific to African swine fever virus" *Front. Cell. Infect. Microbiol*
36. Hu, Xia, Mo et al. (2025) "Rapid detection of African swine fever virus by a blue latex microsphere immunochromatographic strip" *AMB Expr*
37. Onyilagha, Nguyen, Luka et al. "Evaluation of a Lateral Flow Assay for Rapid Detection of African Swine Fever Virus in Multiple Sample Types" *Pathogens*
38. Liu, Liu, Hu et al. "Development of high-concentration labeled colloidal gold immunochromatographic test strips for detecting African swine fever virus p30 protein antibodies"
39. Yang, Li, Wang et al. (2025) "Fc-Labeled Gold Nanoparticle-Based Lateral Flow Strip-Assisted Portable Devices for Rapid and Quantitative Point-of-Care Detection of ASFV Antibodies"
40. Huang, Cao, Xu et al. (2023) "A blocking ELISA based on virus-like nanoparticles chimerized with an antigenic epitope of ASFV P54 for detecting ASFV antibodies" *Sci. Rep*
41. Laurent, Hinnant, Talbott et al. (2024) "Automation for lateral flow rapid tests: Protocol for an open-source fluid handler and applications to dengue and African swine fever tests" *PLoS Glob. Public Health*
42. "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=PMC12645999&blobtype=pdf | # A hepatitis B virus RNA-sensing and RNA-editing-dependent reporter system
Liren Sun, Andrew Snedeker, Liudi Tang
## Abstract
The compact genomic organization of hepatitis B virus (HBV) has long hindered the development of reporter viruses that do not compromise viral gene expression and replication. Leveraging the advantage of an RNA-sensing and ADAR-edit ing-dependent cellular reporter system termed reprogrammable adenosine deaminase acting on RNA sensors (RADARS), we developed an HBV-RADARS reporter. In this system, the expression of the reporter gene is activated in trans by HBV RNA-guided, cellular adenosine deaminase acting on RNA 1 (ADAR1)-dependent reporter RNA editing. Using a luciferase reporter, we systematically scanned all ADAR1 targetable sites present in HBV RNAs and selected an optimal sensor sequence. As anticipated, the activation of reporter mRNA translation is HBV RNA sequence-specific and quantitatively correlates with the abundance of HBV RNA and cellular ADAR1 deaminase activity. The optimized HBV-RADARS system can be used with versatile reporter proteins in multiple HBV cell culture settings, including HBV replicon plasmid-transfected cells, cell lines expressing HBV RNAs from integrated transgenes, and cells infected with HBV. Particularly, using an antibiotic resistance gene as the reporter for HBV-RADARS allows for the phenotypic selection of HBV-infected HepG2 cells expressing human sodium taurocholate cotrans porting polypeptide. Therefore, this HBV RNA-sensing reporter system is compatible with authentic HBV infection and can serve as a versatile platform for high-throughput screening of compounds that inhibit HBV infection and genome-wide genetic screens to identify cellular factors required for HBV infection of hepatocytes. IMPORTANCE Traditional recombinant hepatitis B virus (HBV) reporter viruses are compromised in replication fitness and restricted to a single round of infection. HBV-RADARS, in principle, does not interfere with the viral life cycle as it acts through sensing the presence of HBV RNA and thus reports de novo HBV infection as well as the repli cation of transfected HBV replicons. The HBV-RADARS system represents a significant advancement in HBV research tool development and offers a replication-competent and highly adaptable reporter platform in live cells for genome-wide genetic and chemical screens. Hence, it opens new avenues for dissecting HBV-hepatocyte interactions and holds promises for the identification of host-encoded antiviral targets, thereby advanc ing efforts toward a functional cure for chronic HBV infections. KEYWORDS HBV, reporter, RADARS, RNA editing E ven with an effective vaccine to prevent new infections, 254 million people worldwide still suffer from chronic hepatitis B virus (HBV) infections. HBV-infected patients are at high risk for developing end-stage liver disease, causing an estimated 1.1 million deaths per year (1, 2). Although the production of virus is suppressed by nucleos(t)ide analogs (NUCs), covalently closed circular (ccc) DNA, the template for the transcription of viral RNAs, persists in the nuclei of infected hepatocytes. Because of this limitation, a functional cure for the disease that yields sustained suppression of viremia
and HBsAg loss and further mitigates the incidence of HBV-associated hepatocellular carcinoma is rarely achieved (3,4). Hence, there is a pressing need to develop novel and curative anti-HBV therapy.
HBV enters hepatocytes by engaging its receptor, sodium taurocholate cotransport ing polypeptide (NTCP) (5). Following viral entry, HBV nucleocapsid docks at the nuclear pore complex, and the viral relaxed circular DNA (rcDNA) genome is released into the nucleus upon nucleocapsid uncoating. Inside the nucleus, rcDNA can be recog nized by cellular DNA repair machinery and converted into chromatinized cccDNA (6,7). Transcription of cccDNA minichromosomes gives rise to four major RNA species, including the 3.5 kb pregenomic RNA (pgRNA) encoding HBV core protein (HBc) and viral polymerase (Pol), the 2.4 kb and 2.1 kb RNAs encoding HBV envelope proteins (HBs), and the 0.7 kb RNA encoding HBV X protein (HBx) with regulatory functions. Production of progeny virions begins with the packaging and reverse transcription of pregenomic RNA inside the nucleocapsid in the cytoplasm, followed by the synthesis of genomic rcDNA (6). Nucleocapsids enter the secretory pathway to acquire viral envelope proteins and eventually exit the infected hepatocytes as virions (8). Of particular significance for HBV persistence is the so-called retrograde transport of a portion of mature nucleocapsids to the nucleus early in an infection, which leads to the amplification of the cccDNA pool (6,9).
The development of reporter-expressing systems for HBV infection has met limited success (10) because HBV has a small viral genome (3.2 kb) and does not tolerate large insertions (11). Moreover, the genome is tightly packed with four overlapping open reading frames (ORF), two enhancers, and four promoters. Therefore, inserting a reporter gene into the HBV genome almost inevitably disrupts viral gene expression and genome replication. Over the years, several studies reported different designs of recombinant HBV reporter viruses, such as replacing part of core ORF with NanoLuc, inserting DsRed fluorescent gene into pol ORF, duplicating the overlapping region of core ORF and pol ORF to allow insertion of blasticidin resistance gene, or adding a small tag into pol ORF (12)(13)(14)(15). A common concept used in producing these reporter viruses is to modify HBV core or the spacer region of the pol gene so that it harbors a reporter gene. However, the yield of recombinant reporter viruses is often limited by sub-optimal replication efficiency (11,16,17). Moreover, the diminished HBV replication following initial infection results in a "one-round trip" for these recombinant reporter viruses. This has constituted a major obstacle to the implementation of high-throughput chemical and genome-wide genetic screens targeting HBV.
To overcome this limitation, we exploited reprogrammable adenosine deaminase acting on RNA sensors (RADARS) as a tool for the development of a live cell reporter system, which reports the presence and abundance of HBV RNA without disrupting HBV genome or its viral life cycle. RADARS technology was recently invented to selectively translate payload proteins based on sensing specific RNA species in living cells (18)(19)(20). Mechanistically, this approach harnesses the cellular double-stranded RNA (dsRNA) specific, A-to-I RNA editing enzyme adenosine deaminase acting on RNA (ADAR), to edit a UAG amber stop codon in a reporter messenger RNA into a UIG tryptophan codon. Protein translation is therefore made contingent on the hybridization of a specific target RNA to a corresponding sensor sequence, which then triggers ADAR-mediated A-to-I editing. Taking advantage of this system, we developed an HBV-RADARS reporter, which expresses reporter RNA with a UAG stop codon that is translated after sensing/base pairing with HBV RNA sequences. The HBV-RADARS reporter system bypasses the need to modify the viral genome and does not compromise viral fitness, allowing versatile reporter protein design.
## RESULTS
## Target site selection on the HBV genome and length optimization for RADARS
To test the efficiency of the RADARS reporter system upon sensing HBV RNA, we initially designed a plasmid construct that contains an upstream mCherry ORF and a down stream Gaussia luciferase (Gluc) reporter ORF separated by ribosomal skipping sequen ces (T2A and P2A) and a 147-nt anti-HBV sensor sequence flanking a TAG stop codon (Fig. 1A). This construct allows constitutive mCherry expression, while it conditionally expresses a downstream Gluc reporter. Gluc expression requires that HBV RNA hybridizes to the sensor sequence triggering ADAR-dependent RNA A-to-I editing, converting the TAG stop codon into a TIG, which results in translation readthrough (Fig. 1A). Since an A:C mismatch flanked by long dsRNAs is the preferred ADAR target (21), we searched for the presence of 5′-CCA-3′ in HBV RNAs to design the 5′-TAG-3′ RADARS target and found 66 sites on HBV genotype D (GenBank: MF967563.1). Since RNA secondary structure, protein binding, and other sequence properties can influence ADAR editing efficiency (22), we systematically tested all ADAR targetable sites using co-transfection with a plasmid encoding HBV RNAs (pHBV1.3) or green fluorescent protein (GFP) RNA (pCMV-GFP) as a negative control, and calculated the Gluc fold activation to rank these target sites. The most efficient target sites were mapped to regions overlapping three of the four major HBV RNAs and to the 3′ end of all four major RNAs (Fig. 1B). The overall profile agrees with a study suggesting the RADARS sensor has a higher dynamic range when targeting the coding sequences of secreted proteins or the 3′ UTRs of transcripts (23). Individual validation of the most efficient constructs led to the selection of 1570-1716 (C1643) as the ideal target site, which targets an overlapping region of HBV RNA within the HBx ORF. Because the length of the RNA sensor has been shown to impact the downstream reporter activation likely through altering the hybridization strength with target RNA and ADAR recruitment (18,21,22), we further varied the anti-HBV sensor RNA lengths from 51 nts to 171 nts at 6-nt intervals and found that the HBV-RADARS construct harboring a 141-nt anti-HBV sensor sequence complementary to HBV 1573-1713 yielded the highest and most consistent Gluc activation, 22.4-fold (Fig. 1C). We therefore selected a prototype HBV-RADARS that senses the 1573-1713 (C1643) region of HBV RNA for further study.
Transfection of the replication-competent HBV plasmid (pHBV1.3) produces HBx that is known to elevate episomal DNA transcription by degrading the SMC5/6 complex (24-26). To further verify the contribution of RNA editing to the activation of HBV-RADARS, a plasmid that expresses only the HBV 2.1 kb RNA (pCMV-HBV2.1) encoding HBV small envelope protein (HBs) was co-transfected with the HBV-RADARS plasmid. It resulted in peak activation of 24.6-fold and 75.8-fold at 3 days post-transfection in HepG2-NTCP-C4 and Huh7-NTCP cells, respectively (Fig. 2A andB). Furthermore, replacing the down stream Gluc reporter with a GFP reporter gives rise to green cells when co-transfected with pCMV-HBV2.1, whereas co-transfection with pCMV-HBc that lacks its target RNA region showed only basal GFP expression (Fig. 2C andD). Thus, efficient activation of the reporter can occur following expression of an HBV target transcript independent 0f HBx expression.
## Activation of HBV-RADARS is target RNA specific and correlates with the A-to-I RNA editing ratio
To further verify the specificity controlling HBV-RADARS activation, we constructed a negative control plasmid designated as HBV-RADARS-Rev in which the HBV RNA sensor sequence was reversed and thus does not recognize HBV 2.1 kb RNA (Fig. 3A). Another control, the HBV-RADARS-On construct, has the TAG stop codon replaced with TGG so that the translation of the downstream reporter is constitutively active and does not depend on HBV RNA sensing and ADAR editing (Fig. 3A). As expected, only the HBV-RADARS construct, but not the other two control constructs, exhibited significant Gluc activation (15.2-fold) in response to comparable levels of HBV plasmid transfection in HepG2-NTCP-C4 cells (Fig. 3B andC). The ADAR deamination-dependent mechanism is further supported by a 30-fold increase in the A-to-I editing of HBV-RADARS RNA Plasmid 064 targeting CCA1643 (highlighted in purple) was selected for further optimization. (C) HepG2-NTCP-C4 cells were transfected with the HBV-RADARS plasmids (plasmid 131 through 151) targeting CCA1643, varying the length of their sensor sequences (51-171 nts). Cells were either co-transfected with the pHBV1.3 plasmid or the pCMV-GFP plasmid as a negative control. Five days after transfection, Gluc luminescence was measured, and the fold activation of HBV-RADARS was calculated as the relative ratio of Gluc levels in the presence of HBV plasmid versus control plasmid. The HBV-RADARS plasmid 136 targeting CCA1643 with a 141-nt sensor was selected as prototype HBV-RADARS and is highlighted in purple. Mean ± SD is shown with three biological replicates. transcripts after co-transfection with pCMV-HBV2.1 plasmid encoding HBV RNA (Fig. 3D). To investigate the efficiency of the prototype HBV-RADARS construct in sensing different HBV genotypes, we generated additional HBV2.1 kb RNA expressing plasmids derived from HBV genotype A, B, and C (GenBank MN645903.1, MN645904.1, and MN645905.1). Interestingly, despite lower levels of HBsAg produced by pCMV-HBV2.1-Genotype D compared with genotypes A, B, and C, it nevertheless led to the highest Gluc induction, presumably due to the best sequence complementarity to the HBV-RADARS (Fig. 4). In contrast, HBV genotype A, B, and C yielded relatively less Gluc induction likely due to various degrees of additional mismatches, especially near the center TAG site that may cause misalignment of the dsRNA structure and thus impair the subsequent ADAR editing (Fig. 4 and Fig. S1). Taken together, these results demonstrated that HBV-RADARS function depends on specific HBV-derived target RNA sensing.
## Activation of HBV-RADARS is ADAR deaminase activity dependent
ADAR1 is ubiquitously expressed across human tissues and cell types, and it has been implicated in RNA editing of hepatitis D virus in hepatocytes (27)(28)(29). Therefore, we next assessed the contribution of ADAR1 in the translational readthrough of the HBV-RADARS RNA. Notably, HepG2-NTCP-C4 cells showed detectable levels of the ADAR1 p110 isoform, which is constitutively expressed and localized in the nucleus (Fig. 5A), but not the interferon-inducible N-terminal extended ADAR1 p150 isoform (compare to Fig. 5D), which primarily localizes in the cytoplasm (30). Co-transfection of siRNA targeting ADAR1, which modestly reduced the amount of ADAR1 protein, proportionally attenu ated the HBV RNA-triggered Gluc activation of HBV-RADARS (Fig. 5A through C). Conversely, the activation of HBV-RADARS is further enhanced when ADAR1 was overexpressed ectopically, which increases both ADAR1 p110 and ADAR1 p150 levels (Fig. 5D through F). However, the expression of ADAR1 mutant (ADAR1-E912A), which can bind dsRNA but lacks deaminase activity, resulted in significantly reduced enhance ment (Fig. 5D through F) (31). Collectively, these data demonstrate that the prototype HBV-RADARS functions in an HBV RNA-sensing and ADAR1 RNA-editing-dependent manner.
## Activity of the HBV-RADARS correlates with HBV RNA expression levels
To determine whether Gluc expression from HBV-RADARS correlates with the level of its target HBV RNA, we next co-transfected cells with various amounts of plasmid expressing HBV 2.1 kb RNA (pCMV-HBV2.1), matched with the same amounts of GFP plasmid (pCMV-GFP) as controls. As expected, increasing the amount of pCMV-HBV2.1 plasmid during co-transfection generated more HBV 2.1 kb RNA in a dose-dependent fashion (Fig. 6B) and thereby higher levels of HBsAg (Fig. 6C), which led to an increasing Gluc activation from comparable levels of HBV-RADARS mRNA (Fig. 6A through D). The correlation between HBV-RADARS activation and its target HBV RNA abundance was further tested in HepG2.2.15 cells that harbor a genotype D HBV transgene and thus constitutively express all major HBV RNA species (32). Treating cells with the HBV RNA destabilizing compound RG7834 (except for HBx mRNA) (33)(34)(35) significantly reduced total HBV RNA and antigen expression, corresponding with a decrease in HBV RNA-mediated Gluc activation from HBV-RADARS reporter (Fig. S2).
Endogenous dsRNA and virus-derived dsRNA can potentially activate cellular dsRNA sensors, leading to ISG expression, such as IFN-β (36). Since the mechanism of HBV-RADARS reporter involves dsRNA formation, which is a potential trigger for innate immune activation, we tested IFN-β RNA levels in HBV-RADARS and pCMV-HBV2.1 cotransfected cells. The lack of significant IFN-β induction suggests that at least in these cells, the HBV-RADARS activation is not accompanied by dsRNA-mediated innate immune responses (Fig. S3A). This result aligns with the original RADARS study, which showed no activation of the dsRNA-sensing pathway across a panel of human cell lines, including HepG2 (18).
## HBV-RADARS responds to authentic HBV infection of HepG2-NTCP
To determine whether HBV-RADARS can also sense HBV RNA derived from infection of permissive cells with HBV, we first transfected HepG2-NTCP-C4 with the HBV-RADARS plasmid, followed by infection with 5,000 genome equivalents (GE) of HBV per cell. Nine days post-infection, luciferase activity was determined. HBV infection-mediated Gluc expression from HBV-RADARS was approximately 16-fold greater than that in mockinfected cells (Fig. 7A). Blocking the HBV receptor with the entry inhibitor MyrB efficiently blocked the reporter luciferase activation, whereas reduction of HBV RNA (except for HBx mRNA) by the HBV RNA destabilizing compound RG7834 partially ablated the reporter activation. As expected, the HBV RNA reverse transcription inhibitor ETV did not alter the reporter activation because it did not reduce HBV RNA (Fig. 7A through D). Importantly, HBV infection led to a twofold (0.36% over 0.18%) increase in the A-to-I editing of HBV-RADARS RNA transcripts (Fig. 7E). Consistent with the results from HBV plasmid-transfec ted cells, no significant IFN-β induction was detected with HBV-RADARS activation in HBV-infected cells (Fig. S3B). In agreement with HBV-RADARS luciferase reporter, the HBV-RADARS-GFP reporter also exhibited elevated GFP expression specific to HBV infection versus mock (Fig. S4). Notably, the overall GFP induction and the increase in Ato-I editing ratio by HBV infection of HepG2-NTCP-C4 cells are not as strong as expected (see Discussion); we therefore attempted overexpression of ADAR1 protein. However, while excessive amounts of ADAR1 increased the GFP-positive cell ratio, it also led to significant non-specific expression of GFP in mock-infected cells, likely due to promiscu ous ADAR1 off-target editing events as reported by other ADAR1 overexpression-based approaches too (Fig. S4) (18,37). Taken together, the results demonstrated the expected relationship between HBV RNA expression levels and reporter activity in the HepG2-NTCP HBV infection system.
## HBV-RADARS expressing an antibiotic-resistant gene allows for positive selection of HBV-infected HepG2-NTCP cells
Next, we tested whether an HBV-RADARS reporter conditionally expressing an antibi otic-resistant gene permits the selection of HBV-infected cells. This phenotypic assay could enable genetic screens to identify elusive host factors controlling HBV replication. Accordingly, we replaced the Gluc gene with a puromycin resistance gene and designa ted it as HBV-RADARS-Puro r (Fig. 8A). Transfection of the plasmid into HepG2-NTCP-C4 cells gave rise to a comparable level of mCherry expression from the upstream of the reporter at 1 day post-HBV infection versus mock infection (Fig. 8B). Notably, the mCherry level was slightly higher in HBV-infected cells after 4 dpi, presumably due to HBx and its transactivation function to promote the HBV-RADARS-Puro r plasmid expression (24, 25). After HBV infection establishment at 4 dpi, cells were transiently exposed to 2.5 μg/mL puromycin for 3 days and cultured to 13 dpi, which led to the elimination of the vast majority (>99%) of mock-infected cells (Fig. 8B). Importantly, however, a substantial number (about 5%-10%) of HBV-infected cells survived puromy cin selection and stained positive for HBc, indicating that they had a reasonably high level of HBV RNA as well as HBV-RADARS-Puro r to activate the puromycin-resistant gene during the selection phase (Fig. 8B). Therefore, these results demonstrated the feasibility of using this live cell HBV reporter system for phenotypic selection of HBV RNA-positive cells, promising to make it a versatile molecular tool for chemical and genetic screens, as well as selective targeting of HBV-infected cells.
## DISCUSSION
RADARS are based on the conditional protein translation activated by the sensing of specific RNA sequences in living cells (18)(19)(20). In this study, we adapted RADARS to HBV in vitro culture systems by developing an HBV-RADARS reporter through scanning different target regions on HBV RNA and optimizing the length of HBV RNA sensor (Fig. 1). Our results indicate that the HBV-RADARS reporter activation is target sequence specific (Fig. 3 and4), and the level of reporter activation correlates with the amount of target RNA present in cells (Fig. 6 and7, and Fig. S2). Comparing with conventional HBV reporter systems, which are usually based on replacing part of the HBV genome by a reporter gene and then supplementing the disrupted viral gene in-trans to generate recombinant viruses (13,15), HBV-RADARS reporter has several key advantages. First, it can be used with wild-type HBV, supporting unperturbed viral replication and infectious progeny production. Second, although HBV has a small viral genome (3.2 kb) and does not tolerate large insertions, HBV-RADARS can be designed to express versatile reporter proteins such as Gluc, GFP, puromycin-resistant gene, etc. Third, while recombinant HBV reporter viruses are difficult to produce due to a strong cis-preference and genome size limitation during HBV replication (11,16,17), HBV-RADARS reporter is relatively straightforward to generate, as it can be expressed from a plasmid and, in principle, other expression platforms as well, thereby offering a versatile and adaptable tool for diverse applications. Sensing of dsRNA by host pattern recognition receptors constitutes a major cell innate immune defense mechanism. Activation of dsRNA sensors can lead to global translational shutdown, IFN responses, and inflammatory cytokine responses (36,38). Although HBV-RADARS reporters involve the formation of dsRNA between reporter RNA and HBV RNA, we did not detect discernible protein expression changes, such as HBsAg, Gluc (Fig. 3 and5), and ADAR1 (Fig. 5), nor did we find significant induction of IFN-β RNA (Fig. S3). This is potentially due to the fact that ADAR-mediated RNA editing is known as a general suppressive mechanism to attenuate cellular dsRNA responses, in addition to the intrinsically weakened dsRNA responses in hepatoma cells (39)(40)(41). Further studies are needed to fully understand the dynamics of dsRNA formation and its immunostimu latory potential for a more comprehensive assessment.
As anticipated, the activation of HBV-RADARS reporter is ADAR deaminase depend ent (Fig. 5). Therefore, it is conceivable that the amount and functionality of ADAR in target cells will ultimately impact the degree of HBV-RADARS activation. This probably contributes to the higher Gluc-activation observed in Huh7-NTCP over HepG2-NTCP-C4 cells (Fig. 2). Of note, the HBV-RADARS reporter is effective even with endogenous ADAR levels. This is important because while overexpression of ADAR1 enhances the reporter activation, it also leads to elevated non-specific reporter expression in the absence of target RNA (Fig. S4). Such an observation is consistent with the higher RADARS off-target rate upon ADAR overexpression indicated by whole-genome sequencing (18).
Interestingly, when HBV-RADARS is introduced as a plasmid reporter, it undergoes HBx-mediated episomal DNA transcriptional activation upon HBV infection (24, 25), which further aids in the HBV-specific reporter expression. The time-dependent, elevated mCherry expression (upstream of the stop codon) in HBV-infected cells can reflect the contribution of HBx (Fig. 8 andS4). However, the RNA editing mechanism still contributes significantly to reporter activation based on the increased A-to-I editing ratio after HBV infection (Fig. 7E) as well as the reduced Gluc activation following RG7834 treatment that reduces all major HBV RNA species but not HBx mRNA (Fig. 7) (35,42).
Notably, it appears that our prototype HBV-RADARS can more efficiently detect HBV RNA produced by plasmid transient transfection than by HBV infection in HepG2-NTCP-C4 cells (Fig. 6 and7). In fact, a very high amount of viral input (5,000 GEs/cell) is needed to achieve ideal reporter activation, which is a limitation of this approach. This is to best overcome the lower HBV RNA levels produced by only a few copies of cccDNA present per infected cell. Another possibility might be the sub-optimal transfection and infection rate. It is conceivable that a significant proportion of HBV-RADARS plasmid-transfected cells are not infected by HBV. Engineering of HepG2-NTCP-C4 cells that constitutively express HBV-RADARS RNA through lentiviral transduction was not successful either due to low basal reporter expression when compared to transient transfection (Fig. S5). Alternatively, our Western blot results indicate that the nuclear form of ADAR1 (ADAR1 p110) is the major isoform in HepG2-NTCP-C4 cells (Fig. 5); therefore, RNA editing that occurs co-transcriptionally in the nucleus could be the most efficient way to activate reporter expression in our cell cultures. Since co-transfection of reporter and HBV plasmids likely delivers high amounts of these two plasmid populations into the nuclei within close distance, it is intrinsically more efficient than HBV infection, where HBV RNA is derived from cccDNA, which could localize and transcribe at very specific nuclear regions (43,44). The possible temporal and spatial factors that impact HBV-RADARS reporter in the HBV infection system warrant further investigation.
It should be acknowledged that the full utility of the HBV RADAR reporter system depends on increasing sensitivity and specificity in infection systems. The HBV sensor sequence in the current prototype HBV-RADARS design may theoretically be improved. This is based on the incomplete A-to-I editing of HBV-RADARS RNA (Fig. 3 and7), and the HBV-RADARS-On readthrough construct had about 100-fold higher Gluc expression than the activated HBV-RADARS (Fig. 3). In the meantime, although minimal, both Gluc and GFP-expressing HBV-RADARS reporters exhibit leaky expression even in the absence of target RNA species (Fig. 2A andD), presumably due to non-specific ADAR editing or RNA splicing that overcomes the stop codon. Furthermore, another recent technical advancement in RADARS design, which utilizes the folding of tertiary RNA structures on the reporter RNA to bypass the need for the CCA site on the target RNA, results in broadened target site selection (21). Therefore, it is entirely possible that future designs leveraging better structural prediction of specific dsRNA between sensor and target RNA, as well as inserting additional stop codons as translational gatekeepers, may further improve the sensitivity and specificity of HBV-RADARS, which is essential when implementing HBV-RADARS as a therapeutic platform to selectively target HBV RNA-positive cells.
Finally, our proof-of-concept study demonstrated that the HBV-RADARS reporter expressing an antibiotic-resistant gene marker allows phenotypic selection of HBV-infec ted cells (Fig. 8). This selection will potentially enable unbiased genome-wide genetic screens and large-scale small molecule chemical screens, which have been challenging for HBV research in the past.
In summary, this live cell HBV-RADARS reporter has the potential to fill important gaps in knowledge concerning host-virus interactions critical for the HBV life cycle and ultimately provide new targets for antiviral therapies to achieve a functional cure for patients suffering from chronic hepatitis B infection.
## MATERIALS AND METHODS
## Cell culture
Human hepatoblastoma cell line HepG2-NTCP-C4 was kindly gifted from Dr. Koichi Watashi (45) at Japan National Institute of Infectious Diseases and cultured in DMEM/F12 (1:1) supplemented with 10% fetal bovine serum (FBS) (Gibco), 100 U/mL of penicil lin, 100 µg/mL of streptomycin, and 500 µg/mL G418 (Gibco). HepG2.2.15 (32) was maintained the same way as HepG2-NTCP-C4. Huh7-NTCP was engineered by transduc ing Huh7 (gifted from Dr. Ju-Tao Guo, Blumberg Institute) with human NTCP expressing retroviruses (packaged from pCX4bsr-NTCP and pVSVG transfected GP2-293 cells) and selecting under 10 µg/mL Blasticidin (Gibco) for 2 weeks. Huh7-NTCP was cultured in DMEM medium supplemented with 10% FBS (Gibco), 100 U/mL of penicillin, 100 µg/mL of streptomycin, and 5 µg/mL Blasticidin. HepAD38 cells were obtained from Dr. Christoph Seeger from Fox Chase Cancer Center, Philadelphia, USA (46), and cultured in DMEM/F12 (1:1) supplemented with 10% FBS (Gibco), 100 U/mL of penicillin, 100 µg/mL of streptomycin, and 1 µg/mL tetracycline (Sigma). Tetracycline was removed from HepAD38 culture media to initiate pgRNA transcription and HBV production as needed. All cells were maintained in 5% CO 2 incubators at 37˚C. All cell culture experiments were performed in 50 µg/mL rat tail collagen (Corning) coated plates.
## Plasmids
The HBV-RADARS plasmid was generated as follows: First, an intermediate plasmid containing mCherry-P2A-Sensor-T2A-Gluc was generated by digesting the backbone plasmid pDY1243-Fluorescent RADARS (Addgene #182540) with BsrGI and NotI, followed by assembly with a synthesized DNA fragment containing P2A-T2A-Gluc sequences (designated as F RD000K, IDT, Table S2) using the NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs). Next, the RADARS expression cassette was PCR amplified with primers F1 and R1 (Table S1), and cloned into the pENTR4 backbone derived from a pEN-H133A plasmid (47) digested with PspXI and HindIII, to yield the second intermedi ate plasmid pEN-RADARS. Finally, a series of HBV RADARS plasmids (HBV-RADARS-001 to -066) containing various HBV sensor sequences targeting different CCA sites present in HBV genotype D RNA (GenBank: MF967563.1) were constructed by amplifying pEN-RADARS with primers F2 and R2 (Table S1) to obtain the linear plasmid backbone, which was assembled with different 147-nt synthesized sensor fragments (Table S2, IDT eBlock). Similarly, HBV-RADARS-131 to -151 were designed to screen for the optimal sensor length, using the pEN-RADARS backbone, PCR amplified with F3 and R3, and assembled with various sensor sequences (Table S2, IDT eBlock).
Plasmids HBV-RADARS-Rev with reversed sensor sequence to the prototype HBV-RADARS, and plasmid HBV-RADARS-On containing TGG readthrough sensor sequence were derived from backbone PCR of HBV-RADARS with primers F1 and R1 (Table S1), and fusion with synthetic sensor sequences (Rev Sensor and On sensor, IDT), respectively, using NEBuilder HiFi DNA Assembly. Plasmids HBV-RADARS-GFP and HBV-RADARS-Puro r have the Gluc sequence replaced with the eGFP sequence and puromycin-resistant gene sequence, respectively. Plasmids were constructed by PCR amplification of the HBV-RADARS backbone with primers F4 and R4, insertion of eGFP synthesized from IDT (Table S2), or insertion of Puro r that is PCR amplified from pLVX-Puro (Takara #632164) using primer F5 and R5 (Table S1).
Several HBV expression constructs were used in this study. The pHBV1.3 plasmid and pCMV-HBc plasmid are kind gifts from Ju-Tao Guo (Blumberg Institute) and were previously described (48). pCMV-HBV2.1 (genotype D) was a kind gift from Tianlun Zhou (Blumberg Institute). To generate plasmids expressing HBV genotype A, B, and C RNA transcripts, a set of intermediate plasmids containing full-length HBV was constructed by PCR amplifying pTRE-HBV plasmid backbone (49) with primers F6 and R6 (Table S1), and assembling the linear backbone with synthesized fragments A1/A2, B1/B2, or C1/C2 (Table S2, IDT). Then, the pCMV-HBV2.1-GA, -GB, and -GC plasmids were generated by PCR amplifying the pCMV-HBV2.1-GD backbone with primers F7 and R7 (Table S1), and assembling with the HBV 2.1 kb regions derived from pTRE-HBV-GA,-GB, and GC (PCR by primers F8 and R8, F9 and R9, and F10 and R10, respectively).
To express ADAR1 and its mutant form, HBV-RADARS backbone (pEN) was PCR amplified with F11 and R11 to introduce an HA-tag. The coding sequences for ADAR1 p150 were PCR cloned from pmGFP-ADAR1-p150 (Addgene #117927) using primers F12 and R12, and ADAR1 p150 E912A mutant was cloned using the same backbone to assemble with two overlapping fragments, which are generated by pmGFP-ADAR-p150 plasmid PCR with F12 and R13, and with F13 and R12 (Table S1), to introduce the E912A mutation. The resulting plasmids are designated as pEN-HA-ADAR1 and pEN-HA-ADAR1-E912A.
All plasmid constructs were assembled using the NEBuilder HiFi DNA Assembly Master Mix. Primers used for construction of these plasmids are listed in Table S1. All self-constructed plasmids were confirmed by sequencing.
## Antibodies, chemicals, and siRNAs
Anti-ADAR1 antibody and anti-α-Tubulin antibody were purchased from Cell Signaling Technology. All antibodies were used at 1:1,000 dilutions for Western blot assays. Anti-HBc antibody (C1-5) was purchased from Santa Cruz Biotechnology and used at 1:100 dilution for the immunofluorescence assay. MyrB, ETV, and RG7834 were all purchased from MedChemExpress. Puromycin and Blasticidin were purchased from Gibco. Hoechst 33342 was purchased from Thermo Fisher. DAPI was purchased from Cell Signaling Technology. ADAR1 siRNA (ON-TARGETplus siRNA SMARTpool) and scrambled control siRNAs (ON-TARGETplus non-targeting control pool) used in this study were purchased from Dharmacon.
## Plasmid transfection and siRNA transfection
Plasmid transfections were carried out using Lipofectamine-3000 (Invitrogen) based on the manufacturer's instructions. For experiments involving HBV plasmid transfection (unless otherwise indicated), 200 ng DNA of HBsAg-expressing plasmid and 100 ng DNA of HBV-RADARS plasmid were resuspended in 30 µL Opti-MEM reduced serum medium (Gibco) containing 0.9 µL P3000. The DNA mixture was then added to 30 µL Opti-MEM containing 0.9 µL Lipo-3000, mixed, and incubated at room temperature for 15 min. The DNA-liposome suspension was added to 70%-80% confluent cells cultured in 500 µL Opti-MEM medium in a 24-well plate. Transfection medium was removed 6 h after and refreshed with regular DMEM/F12 growth medium. For experiments involving HBV infection, 200 ng HBV-RADARS plasmid was used to transfect cells in 24-well plates. For experiments involving ectopic expression of ADAR1, 200 ng ADAR1 or control plasmid was used to transfect cells in a 24-well plate.
siRNA transfections were carried out using Lipofectamine RNAiMAX (Invitrogen) based on the manufacturer's instructions. For one well of a 24-well plate, 5 pmol siRNA oligonucleotides and 1 µL Lipofectamine RNAiMAX were used to generate siRNA-lipo some suspension in Opti-MEM medium at room temperature for 15 min, followed by transfecting cells seeded at 70%-80% confluency to have a final siRNA concentration at 10 nM. Transfection medium was removed 6 h after and refreshed with regular DMEM/F12 growth medium.
## Gaussia luciferase assay
Medium containing secreted luciferase was collected 96 h after HBV-RADARS plasmid transfection, unless otherwise noted. 10 µL of culture medium was used to measure luciferase activity using Gaussia Luciferase Assay Kits (Thermo Fisher #16161) on a Tecan plate reader following the manufacturer's protocol. The absolute Gluc values were directly read out from the Tecan plate reader.
## HBsAg and HBeAg chemiluminescent immunoassays
The HBsAg and HBeAg that are secreted into the culture supernatant following HBV infection or plasmid transfection were collected, diluted threefold, and subsequently used for quantification using HBsAg and HBeAg CLIA kits (Ig Biotechnology) following the manufacturer's protocol. For HepG2.2.15, HBsAg and HBeAg levels were measured directly without dilution. The final concentrations were calculated by correlating to the standard concentration curves and multiplying by the dilution factors.
## Western blot assay
Cells grown in a 24-well plate were lysed with 100 µL 1× NuPAGE LDS sample buffer supplemented with 2.5% 2-mercaptoethanol. Cell lysate was boiled on a heat block at 100°C for 20 min, briefly spun, and then resolved by running in NuPAGE 4-12% Bis-Tris Gel (Thermo Fisher) with MOPS-SDS Running Buffer. Proteins were transferred from the gel onto a PVDF membrane using the iBlot 2 Dry Blotting System. Membranes were blocked in Intercept Blocking Buffer (LI-COR) at room temperature for 1 h and then incubated with indicated antibodies in Intercept T20 Antibody Diluent (LI-COR) at 4°C overnight. After washing with TBST (TBS + 0.1% Tween 20), the membrane was incubated with LI-COR IRDye secondary antibodies in Intercept T20 Antibody Diluent (LI-COR) at room temperature for 1 h. Membranes were again washed with TBST and imaged with the LI-COR Odyssey system.
## RNA extraction and qPCR analysis
Total RNA was extracted with NucleoZOL (Takara) according to the manufacturer's instructions. The potential plasmid DNA contaminants were digested by incubating samples with 2.5 μL DNaseI (New England Biolabs) at 37°C for 1 h, followed by heating at 75°C for 10 min to inactivate DNaseI. RNA sample quality was measured by UV absorbance at 260, 280, and 230 nm with a NanoDrop 2000 Spectrophotometer (Thermo Scientific). Detection of HBV-RADARS RNA, HBV RNA, and cellular RNAs was conducted using SuperScript III Platinum One-Step qRT-PCR kit (Invitrogen). Real-time PCR assays were performed using a LightCycler 480 II (Roche) with primers listed in Table S1. The qRT-PCR program was performed following the manufacturer's protocol.
## A-to-I editing ratio analysis
After total RNA extraction, residual HBV-RADARS plasmid DNA contamination was digested by incubating with 5 µL DNaseI (New England Biolabs) at 37°C for 1 h, followed by heating at 75°C for 10 min to inactivate DNaseI. Then, the Sensor region of the HBV-RADARS RNA was PCR amplified using a cDNA One-step kit (NEB) with the primers F14 and R14 (Table S1), during which the adapter sequences for next-gen eration sequencing (NGS) amplicon seq were also ligated to the DNA fragment. The intended DNA amplicon was retrieved by 1.5% agarose gel electrophoresis based on the calculated size, purified by QIAquick Gel Extraction Kit (Qiagen), and submitted to NGS amplicon sequencing services to count reads that contain TAG or TGG (after ADAR editing) in the middle of the sensor region.
## HBV virion production, titer determination, and HBV infection assay
Genome type D HBV virion was generated from collecting the culture media (DMEM containing 3% FBS, 1× NEAA, and 100 U/mL of penicillin, 100 µg/mL of streptomycin) of HepAD38 cells for about 8 weeks after removing tetracycline and subsequently concentrated for 200-fold (volume) by precipitating with 8% PEG-8000 (Sigma) at 2,000 × g for 15 min and resuspending in Opti-MEM medium. Genome equivalents were determined by subjecting a small aliquot of virus stock to DNA extraction using the HBV core DNA extraction method followed by HBV DNA qPCR quantification (50). HBV plasmid pHBV1.3 with serial dilutions was used for generating the standard curve and calculating the genome equivalents. For HBV infection, HepG2-NTCP-C4 cells that were seeded confluently in collagen-coated plates were pretreated with DMEM medium supplemented with 3% FBS, 1× NEAA, 2% DMSO, and 100 U/mL of penicillin, 100 µg/mL of streptomycin for 24 h. Cells were then infected with HBV at 5,000 genome equivalents per cell in pretreatment media added with 4% PEG-8000 (Sigma). The inoculums were removed at 24 h post-infection and cells were washed with PBS 3-5 times and kept in DMEM medium supplemented with 3% FBS, 1× NEAA, 2% DMSO, and 100 U/mL of penicillin, 100 µg/mL of streptomycin.
## Immunofluorescence assay
HepG2-NTCP-C4 cells were fixed with PBS pH 7.4 containing 4% paraformaldehyde (Thermo Fisher), followed by 10 min incubation with 0.25% Triton X-100. Cells were then blocked by a 30 min incubation at room temperature with 1% BSA, 22.52 mg/mL glycine in PBST (PBS + 0.1% Tween 20). Next, cells were incubated with 1:100 diluted HBc antibody in 1% BSA in PBST (PBS + 0.1% Tween 20) overnight at 4°C. Bound primary antibody was visualized by using Alexa Fluor 488-conjugated secondary antibodies in 1% BSA in PBST at room temperature for 1 h. Cell nuclei were stained with 1 µg/mL DAPI during the secondary antibody incubation period.
## Engineering of HepG2-NTCP-C4-HBV-RADARS stable cell line
To generate HepG2-NTCP-C4 stably expressing HBV-RADARS reporter, the reporter segment of HBV-RADARS plasmid was PCR amplified using primers ATCGGAATTCTAG CGTTTAAACTTAAGCTT and TATGGCTGATTATGATCTCTAGTCAGGTTTAAACGGGCCCTCTAG , followed by EcoRI and XbaI restriction digestion before ligating into the pLVX-Puro (Takara) lentiviral vector by T4 DNA ligase (Takara). Pseudotyped lentiviruses were packaged by Lenti-X Packaging Single Shots (VSV-G) (Takara) in Lenti-X 293T cells (Takara). Produced lentiviruses were applied to HepG2-NTCP-C4 cells, followed by puromycin (2 µg/mL) selection for 2 weeks. The puromycin-resistant cells were expanded and designated as HepG2-NTCP-C4-HBV-RADARS.
## Statistical analysis
Data shown in the bar graphs indicate mean ± standard deviation. Data were analyzed using two-way ANOVA analysis for multiple group analysis, or two-tailed Student's t-test for comparing two groups. All bar graphs and statistics were generated by Prism GraphPad 10. The level of significance was set at P < 0.05.
## References
1. Hsu, Huang, Nguyen (2023) "Global burden of hepatitis B virus: current status, missed opportunities and a call for action" *Nat Rev Gastroenterol Hepatol*
2. Anonymous, Who (2025)
3. Papatheodoridis, Lampertico, Manolakopoulos et al. (2010) "Incidence of hepatocellular carcinoma in chronic hepatitis B patients receiving nucleos(t)ide therapy: a systematic review" *J Hepatol*
4. (2025) *Full-Length Text Journal of Virology*
5. Tang, Zhao, Wu et al. (2017) "The current status and future directions of hepatitis B antiviral drug discovery" *Expert Opin Drug Discov*
6. Yan, Zhong, Xu et al. (2012) "Sodium taurocholate cotransporting polypeptide is a functional receptor for human hepatitis B and D virus"
7. Seeger, Mason (2015) "Molecular biology of hepatitis B virus infection" *Virology (Auckl)*
8. Wei, Ploss (2021) "Mechanism of Hepatitis B Virus cccDNA Formation" *Viruses*
9. Jiang, Hildt (2020) "Intracellular trafficking of HBV particles" *Cells*
10. Tuttleman, Pourcel, Summers (1986) "Formation of the pool of covalently closed circular viral DNA in hepadnavirus-infected cells" *Cell*
11. Bai, Cui, Xie et al. (2016) "Engineering hepadnaviruses as reporterexpressing vectors: recent progress and future perspectives" *Viruses*
12. Protzer, Nassal, Chiang et al. (1999) "Interferon gene transfer by a hepatitis B virus vector efficiently suppresses wildtype virus infection" *Proc Natl Acad Sci*
13. Morita, Wada, Ohsaki et al. (2025) "Generation of replication-competent hepatitis B virus harboring tagged polymerase for visualization and quantification of the infection" *Microbiol Immunol*
14. Nishitsuji, Ujino, Shimizu et al. (2015) "Novel reporter system to monitor early stages of the hepatitis B virus life cycle" *Cancer Sci*
15. Wang, Wu, Cheng et al. (2013) "Replication-competent infectious hepatitis B virus vectors carrying substantially sized transgenes by redesigned viral polymerase translation" *PLoS One*
16. Hong, Bai, Zhai et al. (2013) "Novel recombinant hepatitis B virus vectors efficiently deliver protein and RNA encoding genes into primary hepatocytes" *J Virol*
17. Bartenschlager, Junker-Niepmann, Schaller (1990) "The P gene product of hepatitis B virus is required as a structural component for genomic RNA encapsidation" *J Virol*
18. Yu, Kass, Zhang et al. (2024) "Deep mutational scanning of hepatitis B virus reveals a mechanism for cis-preferential reverse transcription" *Cell*
19. Jiang, Koob, Chen et al. (2023) "Programma ble eukaryotic protein synthesis with RNA sensors by harnessing ADAR" *Nat Biotechnol*
20. Kaseniit, Katz, Kolber et al. (2023) "Modular, programmable RNA sensing using ADAR editing in living cells" *Nat Biotechnol*
21. Qian, Li, Zhao et al. (2022) "Programmable RNA sensing for cell monitoring and manipulation" *Nature*
22. Qin, Chen, Tan et al. (2025) "Programming ADARrecruiting hairpin RNA sensor to detect endogenous molecules" *Nucleic Acids Res*
23. Nishikura (2010) "Functions and regulation of RNA editing by ADAR deaminases" *Annu Rev Biochem*
24. Gayet, Ilia, Razavi et al. (2012) "Hepatitis B virus X protein stimulates gene expression selectively from extrachromosomal DNA templates" *Nat Commun*
25. Abdul, Diman, Baechler et al. (2022) "Smc5/6 silences episomal transcription by a three-step function" *Nat Struct Mol Biol*
26. Decorsière, Mueller, Van Breugel et al. (2016) "Hepatitis B virus X protein identifies the Smc5/6 complex as a host restriction factor" *Nature*
27. Gabay, Shoshan, Kopel et al. (2022) "Landscape of adenosine-to-inosine RNA recoding across human tissues" *Nat Commun*
28. Polson, Bass, Casey (1996) "RNA editing of hepatitis delta virus antigenome by dsRNA-adenosine deaminase" *Nature*
29. Wong, Lazinski (2002) "Replicating hepatitis delta virus RNA is edited in the nucleus by the small form of ADAR1" *Proc Natl Acad Sci*
30. George, Samuel (1999) "Human RNA-specific adenosine deaminase ADAR1 transcripts possess alternative exon 1 structures that initiate from different promoters, one constitutively active and the other interferon inducible" *Proc Natl Acad Sci*
31. Lai, Drakas, Nishikura (1995) "Mutagenic analysis of double-stranded RNA adenosine deaminase, a candidate enzyme for RNA editing of glutamate-gated ion channel transcripts" *J Biol Chem*
32. Sells, Chen, Acs (1987) "Production of hepatitis B virus particles in Hep G2 cells transfected with cloned hepatitis B virus DNA" *Proc Natl Acad Sci*
33. Han, Zhou, Jiang et al. (2018) "Discovery of RG7834: the first-in-class selective and orally available small molecule hepatitis B virus expression inhibitor with novel mechanism of action" *J Med Chem*
34. Mueller, Lopez, Tropberger et al. (2019) "PAPD5/7 are host factors that are required for hepatitis B virus RNA stabilization" *Hepatology*
35. Sun, Zhang, Guo et al. (2020) "The dihydroquinolizinone compound RG7834 inhibits the polyadenylase function of PAPD5 and PAPD7 and accelerates the degradation of matured hepatitis B virus surface protein mRNA" *Antimicrob Agents Chemother*
36. Cottrell, Andrews, Bass (2024) "The competitive landscape of the dsRNA world" *Mol Cell*
37. Vallecillo-Viejo, Brauer, Montiel-Gonzalez et al. (2018) "Abundant off-target edits from site-directed RNA editing can be reduced by nuclear localization of the editing enzyme" *RNA Biol*
38. Chung, Calis, Wu et al. (2018) "Human ADAR1 prevents endogenous RNA from triggering translational shutdown" *Cell*
39. Hu, Heraud-Farlow, Sun et al. (2023) "ADAR1p150 prevents MDA5 and PKR activation via distinct mechanisms to avert fatal autoinflammation" *Mol Cell*
40. Li, Banerjee, Goldstein et al. (2017) "Ribonuclease L mediates the cell-lethal phenotype of double-stranded RNA editing enzyme ADAR1 deficiency in a human cell line" *Elife*
41. Nicolay, Moeller, Kahl et al. (2021) "Characterization of RNA sensing pathways in hepatoma cell lines and primary human hepatocytes" *Cells*
42. (2025) *Full-Length Text Journal of Virology*
43. Mouzannar, Schauer, Liang (2024) "The post-transcriptional regulatory element of hepatitis B virus: from discovery to therapy" *Viruses*
44. Tang, Zhao, Wu et al. (2021) "Transcriptionally inactive hepatitis B virus episome DNA preferentially resides in the vicinity of chromo some 19 in 3D host genome upon infection" *Cell Rep*
45. Xia, Cheng, Nilsson et al. (2023) "Nucleolin binds to and regulates transcription of hepatitis B virus covalently closed circular DNA minichromosome" *Proc Natl Acad Sci*
46. Iwamoto, Watashi, Tsukuda et al. (2014) "Evaluation and identification of hepatitis B virus entry inhibitors using HepG2 cells overexpressing a membrane transporter NTCP" *Biochem Biophys Res Commun*
47. Ladner, Otto, Barker et al. (1997) "Inducible expression of human hepatitis B virus (HBV) in stably transfected hepatoblastoma cells: a novel system for screening potential inhibitors of HBV replication" *Antimicrob Agents Chemother*
48. Zhou, Block, Liu et al. (2018) "HBsAg mRNA degradation induced by a dihydroquinolizinone compound depends on the HBV posttranscriptional regulatory element" *Antiviral Res*
49. Liu, Cheng, Viswanathan et al. (2021) "Amino acid residues at core protein dimer-dimer interface modulate multiple steps of hepatitis B virus replication and HBeAg biogenesis" *PLoS Pathog*
50. Guo, Jiang, Zhou et al. (2007) "Characteri zation of the intracellular deproteinized relaxed circular DNA of hepatitis B virus: an intermediate of covalently closed circular DNA formation" *J Virol*
51. Guo, Tang, Shu et al. (2017) "Activation of stimulator of interferon genes in hepatocytes suppresses the replication of hepatitis B virus" *Antimicrob Agents Chemother*
52. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12605410&blobtype=pdf | # Targeting the LPS-STING axis: neomycin restores STING-mediated anti-tumor immune suppression and inhibits tumor growth
Hong Fan, Dongjie Fu, Mingfu Tian, Zhiqiang Li, Siyu Liu, Chenglin Ye, Kailang Wu, Chengliang Zhu, Lishan Su, Namrata Anand
## Abstract
Introduction:The interplay between microbial metabolites and host immunity within the tumor microenvironment (TME) critically modulates anti-tumor immune responses. The role of Gram-negative bacteria and their cell wall component lipopolysaccharide (LPS) in this context warrants further investigation. Methods: We assessed the impact of low-dose LPS pretreatment on macrophage function by measuring type I interferon (IFN-b) secretion in response to tumor cell debris. Mechanistic insights were gained by analyzing endogenous signaling pathways in macrophages. The therapeutic potential of targeting LPS was evaluated in melanoma-bearing mice treated with neomycin, alone or in combination with STING agonists. Results: Low-dose LPS pretreatment significantly suppressed IFN-b secretion by macrophages, indicating LPS-mediated immunosuppression. Mechanistically, LPS disrupted endogenous signaling pathways, blunting the ability of macrophages to sense tumor-derived damage signals. In vivo, neomycin treatment markedly inhibited melanoma growth and synergized with STING agonists. Discussion: Our findings demonstrate that elevated LPS in the TME inhibits antitumor innate immunity by impairing macrophage function. The combination of LPS modulation via neomycin with innate immune activation via STING agonists presents a potential strategy to enhance tumor immunotherapy.
## 1 Introduction
Tumor development is a complex biological process involving multiple factors. It is not solely driven by genetic and epigenetic alterations in tumor cells but is also profoundly shaped by the tumor microenvironment (TME) (1)(2)(3). Recent advances in microbiomics have revealed that the microbiota is a critical component of the TME, highlighting its significance and the need for further investigation (4,5). This progress stems largely from high-throughput sequencing technology, which has identified characteristic microbial colonization in tumor tissues-previously considered sterile (6). These microbes have been observed to coexist with their hosts for extended periods, thereby establishing intricate interactions that have the potential to exert a profound influence on the biological behaviour of tumors (7,8).
Conventionally, cancer research and microbiology have been regarded as discrete domains. However, the identification of bacteria in tumor tissue has prompted scientists to re-evaluate the potential association between bacteria and tumours (5). Recent research findings indicate a potential role for bacteria in the development, progression and metastasis of tumors (9)(10)(11)(12). These microorganisms that colonize tumor tissues are collectively termed the 'Intratumoral Microbiota' (IMM) (13). Despite the fact that a significant number of studies have investigated the effects of bacteria on tumor progression and treatment response, the precise mechanism of action remains to be elucidated, particularly with regard to the role of microbes in shaping the tumor microenvironment through metabolites or immunomodulation. This aspect requires further investigation (14).
Bacteria, as key components of the TME, produce a wide range of genotoxic and metabolically active substances, which promote tumorigenesis and progression through diverse molecular mechanisms (15,16). Among these bacterial components, 16S rRNA and LPS exhibit distinctive biological properties and are detectable in nearly all tumor types, showing highly conserved spatial distribution patterns. This widespread presence suggests that LPS may play a fundamental role in tumorigenesis. In contrast, lipoteichoic acid (LTA), a surface component of Gram-positive bacteria, is rarely detected in most tumor tissues (5,17,18). The selective enrichment of LPS (or Gram-negative bacteria) further highlights their unique function in tumor development (19).
Under normal conditions, LPS activates the NF-kB signaling pathway in immune cells such as macrophages and dendritic cells (DCs), stimulating their secretion of pro-inflammatory cytokines and immunomodulatory factors to promote anti-tumor immune responses (20). However, within the tumor microenvironment (TME), the sustained NF-kB activation driven by chronic LPS exposure becomes a hallmark of cancer-associated inflammation and exerts an opposing effect. This persistent signaling reprograms macrophages towards a pro-tumoral, M2-like state, which in turn enhances tumor cell proliferation, survival, migration, and invasion (21,22). This suggests that LPS-induced macrophage immune tolerance may underlie this paradoxical effect.
A thorough investigation of LPS mechanisms in the TME may yield dual benefits: advancing our understanding of tumormicrobiome interactions at the molecular level and informing the design of innovative anticancer therapies. This article explores the emerging role of intratumoral bacteria in oncogenesis, with a focus on deciphering how LPS modulates immunity within the TME. Our findings aim to provide mechanistic insights for developing nextgeneration microbiome-targeted anticancer strategies.
## 2.2 Isolation of peritoneal macrophages from mice
Four days before the experiment, 1 mL of 3% sodium thioglycolate (108191, Millipore) was injected(i.p.) to each C57BL/6J WT mice to induce the aggregation of peritoneal macrophages. The mice were euthanised by cervical dislocation and subsequently fixed in the supine position on an anatomical plate. The abdominal skin was sterilised with 75% alcohol. Using sterile scissors and forceps to lift the abdominal skin, a small incision was made to avoid damage to the abdominal viscera, and 7 mL of phosphate-buffered saline (PBS) was injected into the abdominal cavity. The abdomen of the mice was gently massaged for 2-3 minutes to allow the intraperitoneal fluid to be thoroughly mixed, and then the peritoneal lavage fluid was withdrawn with a sterile syringe and collected into a centrifuge tube. The collected fluid was centrifuged at 300×g for 5 min, and the supernatant was discarded to obtain peritoneal macrophage precipitates.
## 2.3 Cell culture
Mouse colon cancer cell line MC38 and mouse melanoma cell line B16 were obtained from the American Type Culture Collection (ATCC). Cells were cultured in DMEM high glucose medium (Gibco-ThermoFisher) supplemented with 10% FBS and 1% penicillin-streptomycin solution and grown at 37°C in a humidified 5% CO2 incubator.
## 2.4 RNA isolation and quantitative realtime PCR
The cells was first homogenized with 500 mL TRIzol ™ Reagent (Invitrogen), 100ul chloroform is introduced and vortexed, then centrifuged at 12000 rpm for 5 min causing the homogenate to segregate into distinct layers: a clear upper aqueous layer that harbors RNA, an intermediate interphase, and a red lower organic layer which contains DNA and proteins. To obtain RNA, 200ul isopropanol is utilized to precipitate it from the aqueous layer and the precipitate was finally washed with 500ul 70% ethanol. RNasefree EP tubes are required for the entire procedure. Using Hiscript II Q RT SuperMix (Vazyme Biotech, China) for cDNA synthesis in the 96-well Veriti Thermal Cycler (Thermo Scientific, China).
## 2.5 Western blotting
Protein samples were mixed with 2×SDS loading buffer, heated at 95°C for 10 min, and then separated by 10% SDS-PAGE. Proteins were transferred onto PVDF membranes, which were blocked with 5% nonfat milk and incubated overnight at 4°C with primary antibodies. The following antibodies were used: anti-phospho-TBK1 (ab109272, Abcam), anti-phospho-TAK1 (ab109404, Abcam), anti-TBK1 (ab40676, Abcam), anti-TAK1 (ab09526, Abcam), and anti-GAPDH (G9295, Sigma). The next day, membranes were incubated with HRPconjugated secondary antibodies (Jackson ImmunoResearch, USA) for 1 h. After washing, protein bands were visualized using an ECL kit (E423-02, Vazyme Biotech).
## 2.6 Measurement of serum LPS level
Whole blood samples collected from mice were left at room temperature for 2 h and centrifuged at 1000×g for 20 min to separate the serum. LPS levels in mouse serum were measured using a commercial ELISA kit (JL20691, Jianglai Biotechnology).
## 2.7 Cell viability
A cell suspension was prepared and 10 4 cells were seeded into each well of a 96-well plate. After cell adhesion, cells were treated with LPS at varying concentrations for 24 h, after which 10 ml of CCK-8 solution (CCK004, Biolight Biotechnology) was added to each well. Plates were incubated for 1-4 h, and absorbance was measured at 450 nm.
## 2.8 DNA extraction
After cell collection, 20 μL proteinase K and 300 μL PK buffer (10 mM Tris-HCl, pH 8.0) were added to the samples, followed by incubation at 56°C for 1 h in a water bath with gentle inversion every 10 min. When the solution turned clear, 100 μL phenolchloroform (p1013; Polarbio) was added, and the mixture was centrifuged at 12, 000 ×g for 5 min at room temperature. The upper aqueous phase (200 μL) was carefully transferred and mixed with an equal volume of isopropanol. After repeating the centrifugation, the pellet was washed with 500 μL of 75% ethanol.
## 2.9 Immunofluorescence
Mouse tissues were collected, washed with PBS, and fixed in 4% paraformaldehyde. After paraffin embedding, tissue sections were prepared and subjected to immunohistochemical staining using the following primary antibodies: anti-phospho-STING (Ser366; 19851-1-AP, Proteintech) and anti-LPS (ab35654, Abcam). Nuclei were counterstained with DAPI.
## 2.10 Statistical analysis
All statistical analyses in this study were performed using GraphPad Prism (version 10.1.2). The results were displayed as mean ± SEM. The statistical significance of the differences between the groups was evaluated using the Student's t-test, and p-value <0.05 was considered statistically significant.
## 2.11 Establishment of tumor-implantation mice model
WT C57BL/6J mice were subcutaneously injected with 200 mL of B16-F10 cells (1.0×10 5 cells/mouse) suspended in sterile PBS. Three days later, tumor-bearing mice were randomly divided into three groups (n=4/group): B16 (PBS control), LPS (1 mg/kg, L2630, Sigma-Aldrich), and E. coli (1×10 8 CFU, CD201, TransGen Biotech). Starting on day 4, the LPS and E. coli groups received intraperitoneal injections of their respective treatments on alternating days. Tumor length (a) and width (b) were measured every 3 days, and volume (V) was calculated as V = ½ × a × b². When V reached 1500 mm³, tumors were excised entirely. One portion was fixed in 4% paraformaldehyde for 24 h, and the remainder was snap-frozen in liquid nitrogen for molecular analysis (23)(24)(25).
## 2.12 In vivo assays for neomycin
WT C57BL/6J mice were subcutaneously inoculated with B16-F10 melanoma cells (1.0×10 5 cells/mouse) to establish the tumor model. Three days post-inoculation, tumor-bearing mice were randomly assigned into four groups (n=4/group) (1): PBS control (2); Neomycin (30 mg/kg) (3); CMA (Cridanimod; 1.5 mg/kg, T5317, TargetMol); and (4) Neomycin + CMA combination. Treatments were administered intraperitoneally on alternating days.
## 3 Results
## 3.1 LPS enrichment in the tumor microenvironment
To explore the distribution characteristics and potential functions of LPS in the tumor microenvironment, we first established a subcutaneous xenograft model of B16 melanoma in wild-type C57BL/6J mice. When the tumors reached 500 mm³, tumor tissues, adjacent muscle tissues, and major solid organs (including the liver, spleen, and lungs) were systematically collected. Through immunofluorescence staining combined with quantitative image analysis, we found that the signal intensity of LPS in tumor tissues was significantly higher than that in other normal tissues (Figures 1A,B), suggesting that there might be a unique LPS enrichment mechanism in the tumor microenvironment. Notably, in vitro experiments showed that the conditioned medium of B16 cells could significantly up-regulate the expression level of Cd14 mRNA in mice peritoneal macrophages (Figure 1C). Based on these findings, we speculate that the LPS enriched in the tumor might affect tumor exerts its influence on tumor progression through specific mechanisms.
## 3.2 Gram-negative bacteria promote tumor growth through LPS
We sought to determine the contribution of Escherichia coli (E. coli) to tumor progression. In mice bearing B16F10 melanoma, administration of inactivated E. coli or its purified LPS component both enhanced tumor growth relative to controls. A key observation was that the LPS-treated group displayed a steeper tumor growth curve (Figure 2A), indicating a faster growth rate, despite the lack of a significant difference in final tumor mass between the E. coli and LPS groups (Figure 2B). This kinetic profile implies that the potent immunomodulatory molecule LPS is a primary driver of this effect. The findings support a model wherein the tumor-promoting capacity of Gram-negative bacteria can be largely attributed to LPS, suggesting that live bacteria are not obligatory for this process.
The bioactive components derived from bacteria, such as lipopolysaccharide or other metabolites, seem sufficient to sustain the process of promoting tumor growth. Immunofluorescence analysis revealed substantial LPS accumulation in tumors after E. coli or LPS administration (Figure 2C). While LPS did not affect tumor cell proliferation in vitro (Figure 2D), its in vivo effects likely occurby modulating tumor-associated macrophages (TAMs) and inducing immunosuppressive cytokines, suggesting microenvironmentmediated promotion. These findings establish LPS as a key mediator of bacteria-driven tumor progression, acting primarily through microenvironmental regulation rather than direct tumor cell stimulation.
## 3.3 LPS suppresses STING-mediated antitumor immunity
Subsequently, we extracted the DNA from a variety of tumor cells, including LLC, MC38, and B16. Macrophages were pretreated with LPS, and then stimulated with the tumor cell DNA. Intriguingly, it was observed that upon LPS pretreatment, the macrophages failed to produce a high level of IFN-b in response to the tumor DNA stimulation (Figure 3A). This finding strongly suggests that LPS is capable of suppressing the anti-tumor immune response. In our study, macrophages were pretreated with LPS, and subsequently, the stimulant was replaced with nucleic acid analogues. These nucleic acid analogues are capable of activating macrophages to produce IFN-b through different receptors. Among the two nucleic acid analogues examined, LPS exhibited an inhibitory effect on the IFN-b production in macrophages induced by the double-stranded DNA analogue poly(dA:dT) (Figures 3B,C). Consistent with the suppression of IFN-b, the production of key immune mediators CXCL10 (Figure 3F) and TNF-a (Figure 3G) was also significantly attenuated by LPS pretreatment, indicating a broad suppression of innate immune activation.
Within cells, the major DNA sensor cGAS activates the crucial cGAS-STING pathway upon DNA recognition. Notably, cGAS can acutely sense tumor-derived DNA, thereby triggering an anti-tumor immune response. This characteristic provides an important theoretical basis for subsequent research. To further investigate this process, we separately used the second messenger cGAMP and the STING-specific stimulator diABZI to stimulate macrophages. As a result, it was found that the production of IFN-b in macrophages was also inhibited (Figures 3D,E). Similarly, the induction of both CXCL10 and TNF-a by these STING agonists was markedly suppressed in LPS-pretreated macrophages (Figures 3F,G). This phenomenon fully indicates that the regulation of IFN-b production in macrophages by lipopolysaccharide is not isolated but shows obvious pathway dependence, that is, it depends on specific cell signal transduction pathways to achieve its regulatory effect.
To elucidate the molecular mechanism underlying this suppression, we examined the activation of key signaling components. Western Blot analysis revealed that LPS pretreatment significantly inhibited the phosphorylation of TANK-binding kinase 1 (TBK1) at Ser172 in macrophages stimulated with poly(dA:dT), cGAMP, or diABZI, while total TBK1 levels remained unchanged (Figures 3H). This suppression of TBK1 phosphorylation provides mechanistic evidence for the observed inhibition of downstream cytokine production. To further verify the actual situation of this conclusion in vivo, we established a melanoma mouse model. In this model, we compared the data of the E. coli group/LPS group with that of the control group. The experimental results showed that compared with the control group, the phosphorylation level of STING in the E. coli group/LPS group was significantly reduced (Figure 3I). This in vivo experimental result is highly consistent with our previous findings at the cellular level, further confirming the close connection between the regulation of IFN-b production in macrophages by LPS and the STING pathway.
## 3.4 Neomycin suppresses tumor growth by targeting LPS
Next, we focused our attention on the potential role of neomycin, a drug targeting Gram-negative bacteria, in tumor treatment. For this purpose, we selected two representative mouse tumor models, melanoma and colon cancer. During the experiment, the mice in the treatment group received precise neomycin treatment intervention, while the control group did not receive any special drug treatment (Figure 4A). In the mice treated with neomycin, the growth rate of the (f&g) Mouse peritoneal macrophages (PMs) were pretreated with lipopolysaccharide for 24 hours and then washed twice with PBS. PM cells were then stimulated with poly(dA:dT) or diABZI, respectively. The mRNA levels of (F) CXCL10 and (G) TNF-a were detected by qRT-PCR. (H) Under the conditions with or without LPS pretreatment, mouse peritoneal macrophages were stimulated with poly(dA:dT) (6-8 h) or cGAMP (4-6 h) via polyethyleneimine (PEI) transfection at a 2:1 mass ratio of PEI to nucleic acids, or with the STING agonist diABZI (4-6 h). The phosphorylation and total protein levels of TANK-binding kinase 1 (TBK1), including phospho-TBK1 (Ser172), were analyzed by Western Blot with GAPDH as the loading control (I) Immunofluorescence Detection of Phosphorylated STING (indicated by green) in Tumor Tissues, with DAPI (in blue) serving as the nuclear counterstain. The magnification is denoted by a 100mm scale bar. Data are expressed as mean ± SD of three independent experiments, two-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.
tumor was significantly slowed down, the increase in tumor size was significantly lower than that in the control group, and the tumor weight was also relatively lighter (Figures 4B-E). This significant difference indicates that neomycin has played an active and effective role in inhibiting the development of the tumor, strongly proving its potential anti-tumor activity. In order to further explore the internal mechanism of neomycin's anti-tumor effect, we further carried out detailed analysis and detection on the serum and the components inside the tumor of the mice. The results showed that in the mice treated with neomycin, the LPS content inside the tumor tissue showed an obvious decreasing trend (Figure 4F). At the same time, the LPS content in the serum was also significantly reduced compared with that in the control group (Figure 4G). Notably, this anti-tumor effect was not due to a direct cytotoxic effect of neomycin on cancer cells, as in vitro CCK-8 assays demonstrated that neomycin treatment did not directly promote tumor cell proliferation (Figure 4H). Importantly, immunofluorescence analysis revealed that neomycin treatment substantially reduced the infiltration of CD206 + M2-like macrophages in the tumor microenvironment (Figure 4I), indicating a reversal of the immunosuppressive state. These findings collectively reveal the mechanistic pathway of neomycin's anti-tumor action: by reducing LPS content, neomycin remodels the tumor microenvironment and modulates immune regulatory pathways, particularly by shifting macrophage polarization away from the M2-like phenotype, ultimately achieving effective suppression of tumor development.
## 4 Discussion
Abnormal accumulation of LPS in the tumor microenvironment (TME) has become an important phenomenon studied in recent years, and our study further validates this important finding with experimental data (26,27). This accumulation may originate from two main pathways. Firstly, there is the concept of gut microbiota translocation, whereby the integrity of the intestinal barrier is compromised, resulting in the intestinal microbiota breaching the intestinal mucosal barrier and migrating to distal organs via the portal vein system or the lymphatic circulation. Secondly, the specific colonization and proliferation of tumor-associated bacteria (such as Fusobacterium nucleatum or Escherichia coli) in tumor tissues, which continuously release LPS to form a local high-concentration microenvironment (28,29). To elucidate the functional impact of this LPS enrichment, we employed both E. coli and purified LPS in our experimental systems. While E. coli modeled the complex biological scenario of bacterial encounter within the TME, LPS allowed us to precisely attribute the observed immunosuppressive effectsspecifically the suppression of STING signaling and reduction in IFN-b production-to this key. Future studies directly comparing the effects of whole bacteria versus purified PAMPs on this pathway will provide further mechanistic insights. However, the aberrant accumulation of such LPS in the tumour microenvironment, caused by either approach, has a promoting effect on tumour growth, and the mechanisms behind it are extremely complex and multifaceted (5,(30)(31)(32).
Extensive research has established that LPS can directly enhance tumor cell proliferation and migration through activation of canonical signaling pathways such as TLR4/NF-kB (33)(34)(35). Importantly, the pro-tumorigenic role of LPS/TLR4/NF-kB signaling extends beyond colorectal cancer, with demonstrated significance in bone cancer, hepatocellular carcinoma, and ovarian cancer, highlighting its broad relevance as a therapeutic target across malignancies (36-38). However, our study uncovers a unique indirect pro-tumorigenic mechanism of LPS in colorectal cancer MC38 cell lines. Contrary to conventional understanding, our experimental findings demonstrate that LPS does not directly stimulate MC38 cell proliferation, but indirectly drives tumor progression by remodeling the phenotype and function of TAMs in the TME. This discovery is highly consistent with recent studies on the pancreatic cancer microenvironment, which elucidated that LPS regulates macrophage polarization in a time-dependent manner through the TLR4/NF-kB pathway, culminating in a deeply immunosuppressive tumor microenvironment (26). Our findings, together with emerging evidence from other cancer types, establish the LPS/TLR4/NF-kB axis as a conserved pathway that shapes the immunosuppressive landscape across diverse tumors (36-38). Our study not only further validates the applicability of this mechanism in colorectal cancer models, but also provides new experimental evidence supporting the critical role of LPS-mediated TAM functional modulation in driving tumor progression.
The human anti-tumor immune response represents a precisely orchestrated, multi-tiered defense mechanism that integrates both innate and adaptive immune components through coordinated interactions (39). Within this sophisticated immunological network, the cGAS-STING pathway serves as a central regulatory hub (40). When macrophages phagocytose tumor-derived DNA within the tumor microenvironment, this triggers activation of the cGAS-STING signaling cascade, inducing robust type I interferon production (41)(42)(43). These interferons subsequently drive macrophage polarization toward an immunostimulatory M1 phenotype characterized by enhanced antigen presentation capabilities. Furthermore, this cascade facilitates the recruitment and activation of cytotoxic CD8+ T lymphocytes and natural killer (NK) cells, collectively establishing a potent anti-tumor immune surveillance system.
Based on this pivotal mechanism, STING agonists have been regarded as a highly promising therapeutic strategy for cancer treatment. However, clinical studies have demonstrated limited efficacy of STING agonist monotherapy, which may be attributed to their activation of negative feedback regulation (44). Our study elucidates a crucial underlying mechanism: LPS perturbs the precisely coordinated STING-mediated anti-tumor immune response by reducing tumor cell sensitivity to cytosolic DNA and elevating the activation threshold of the cGAS-STING pathway. These alterations collectively result in suppression of macrophage-mediated antitumor functions and induction of an immunosuppressive tumor microenvironment. These findings provide a compelling explanation for the suboptimal therapeutic efficacy observed with STING agonists in clinical settings.
More importantly, our study provides the first experimental evidence that the aminoglycoside antibiotic neomycin possesses dual anti-tumor mechanisms: specifically, it neutralizes LPSmediated immunosuppression while synergizing with the STING agonist Cridanimod (CMA, 10-Carboxymethyl-9-acridanone) to significantly potentiate STING pathway activation (45).This pivotal finding not only clarifies how bacterial components regulate tumor immunity, but more critically, proposes an innovative combination strategy to address the clinical challenges of STING-targeted therapies, with substantial translational potential.
## References
1. Hanahan, Robert (2011) "Hallmarks of cancer: the next generation" *Cell*
2. Gajewski, Schreiber, Fu (2013) "Innate and adaptive immune cells in the tumor microenvironment" *Nat Immunol*
3. Hanahan, Lisa (2012) "Accessories to the crime: functions of cells recruited to the tumor microenvironment" *Cancer Cell*
4. Sholl, Sepich-Poore, Knight et al. (2022) "Redrawing therapeutic boundaries: microbiota and cancer" *Trends Cancer*
5. Nejman, Livyatan, Fuks et al. (2020) "The human tumor microbiome is composed of tumor type-specific intracellular bacteria" *Science*
6. Bertocchi, Carloni, Ravenda et al. (2021) "Gut vascular barrier impairment leads to intestinal bacteria dissemination and colorectal cancer metastasis to liver" *Cancer Cell*
7. Natalini, Singh, Segal (2022) "The dynamic lung microbiome in health and disease" *Nat Rev Microbiol*
8. Goto (2022) "Microbiota and lung cancer" *Semin Cancer Biol*
9. Dai, Tan, Qiao et al. (2024) "Emerging clinical relevance of microbiome in cancer: promising biomarkers and therapeutic targets" *Protein Cell*
10. Battaglia, Mimpen, Traets et al. (2024) "A pan-cancer analysis of the microbiome in metastatic cancer" *Cell*
11. Fu, Yao, Dong et al. (2023) "Emerging roles of intratumor microbiota in cancer metastasis" *Trends Cell Biol*
12. Fu, Yao, Dong et al. (2022) "Tumor-resident intracellular microbiota promotes metastatic colonization in breast cancer" *Cell*
13. Stevens, Benidovskaya, Llorens-Rico et al. (2024) "Bacteria in metastatic sites: unveiling hidden players in cancer progression" *Cancer Cell*
14. Yin, Pu, Chen et al. (2021) "Gut-derived lipopolysaccharide remodels tumoral microenvironment and synergizes with pd-L1 checkpoint blockade via tlr4/myd88/akt/nf-Kb pathway in pancreatic cancer" *Cell Death Dis*
15. Sepich-Poore, Zitvogel, Straussman et al. (2021) "The microbiome and human cancer" *Science*
16. Lu, Xu, Chen et al. (2025) "Intrahepatic microbial heterogeneity in multifocal hepatocellular carcinoma and its association with host genomic and transcriptomic alterations" *Cancer Discov*
17. Xue, Chu, Zheng et al. (2023) "Current understanding of the intratumoral microbiome in various tumors" *Cell Rep Med*
18. Zhao, He, Lai et al. (2022) "Comprehensive histological imaging of native microbiota in human glioma" *J Biophotonics*
19. Li, Zhao, Peng et al. (2025) "Multi-omics analysis reveals the interplay between intratumoral bacteria and glioma" *Msystems*
20. Kawai, Akira (2010) "The role of pattern-recognition receptors in innate immunity: update on toll-like receptors" *Nat Immunol*
21. Guo, Chen, Ye et al. (2024) "Nf-Kb in biology and targeted therapy: new insights and translational implications" *Signal Transduction Targeted Ther*
22. Zhang, Ma, Zhang et al. (2021) "Nf-Kb signaling in inflammation and cancer" *MedComm*
23. Gu, Lin, Dong et al. (2023) "Pcsk9 facilitates melanoma pathogenesis via a network regulating tumor immunity" *J Exp Clin Cancer Res*
24. Scatozza, Moschella, 'arcangelo et al. (2020) "Nicotinamide inhibits melanoma in vitro and in vivo" *J Exp Clin Cancer Res*
25. Sun, Hu, Liu et al. (2023) "Macrophage sting signaling promotes nk cell to suppress colorectal cancer liver metastasis via 4-1bbl/4-1bb costimulation" *J Immunother Cancer*
26. Han, Fu, Wang et al. (2024) "Probiotics functionalized with a gallium-polyphenol network modulate the intratumor microbiota and promote anti-tumor immune responses in pancreatic cancer" *Nat Commun*
27. Xie, Xie, Zhou et al. (2022) "Microbiota in tumors: from understanding to application" *Adv Sci*
28. Derosa, Iebba, Silva et al. (2024) "Custom scoring based on ecological topology of gut microbiota associated with cancer immunotherapy outcome" *Cell*
29. Sulit, Daigneault, Allen-Vercoe et al. (2023) "Bacterial lipopolysaccharide modulates immune response in the colorectal tumor microenvironment" *NPJ Biofilms Microbiomes*
30. Liu, Xu, Wang et al. (2014) "Lps induced mir-181a promotes pancreatic cancer cell migration via targeting pten and map2k4" *Digestive Dis Sci*
31. Thibodeau, Bourgeois-Daigneault, Lapointe (2014) "Targeting the mhc class ii antigen presentation pathway in cancer immunotherapy" *OncoImmunology*
32. El Tekle, Andreeva, Garrett (2024) "The role of the microbiome in the etiopathogenesis of colon cancer" *Annu Rev Physiol*
33. Sun, Wu, Ma et al. (2016) "Toll-like receptor 4 promotes angiogenesis in pancreatic cancer via pi3k/akt signaling" *Exp Cell Res*
34. Dianne, Mencin, Pradere et al. (2012) "Promotion of hepatocellular carcinoma by the intestinal microbiota and tlr4" *Cancer Cell*
35. Jiang, Yuan, Dou et al. (2021) "Lipopolysaccharide affects the proliferation and glucose metabolism of cervical cancer cells through the fra1/ mdm2/P53 pathway" *Int J Med Sci*
36. Schweer, Anand, Anderson et al. (2023) "Human macrophage-engineered vesicles for utilization in ovarian cancer treatment" *Front Oncol*
37. Wu, Anand, Guo et al. (2025) "Bridging immune evasion and vascular dynamics for novel therapeutic frontiers in hepatocellular carcinoma" *Cancers*
38. Martino, Rathmell, Galluzzi (2024) "Vanpouille-Box C. Cancer cell metabolism and antitumour immunity" *Nat Rev Immunol*
39. Motwani, Pesiridis, Fitzgerald (2019) "DNA sensing by the cgas-sting pathway in health and disease" *Nat Rev Genet*
40. Kwon, Bakhoum (2020) "The cytosolic DNA-sensing cgas-sting pathway in cancer" *Cancer Discov*
41. Lanng, Lauridsen, Jakobsen (2024) "The balance of sting signaling orchestrates immunity in cancer" *Nat Immunol*
42. Zhang, Zhang (2025) "Regulation of cgas-sting signalling and its diversity of cellular outcomes" *Nat Rev Immunol*
43. Song, Chen, Pan et al. (2025) "Targeting tumor monocyte-intrinsic pd-L1 by rewiring sting signaling and enhancing sting agonist therapy" *Cancer Cell*
44. Hu, Sańchez-Rivera, Wang et al. (2023) "Sting inhibits the reactivation of dormant metastasis in lung adenocarcinoma" *Nature* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12560573&blobtype=pdf | # Senescent cells promote viral infectionassociated inflammation and tissue damage through a robust NF-κB pathway
Chengliang Zhu, Mingfu Tian, June Ma, Zhiqiang Li, Chenglin Ye, Xianghua Cui, Guolei Wang, Siyu Liu, Muhammad Suhaib Qudus, Uzair Afaq, Hong Fan, Jiali Xiong, Guangli Li, Chuanjin Luo, Zhen Wang, Zhengjiang Jin, Ming Guo, Xin Wang, Zhixiang Huang, Kailang Wu
## Abstract
Respiratory virus infections have been presenting significant global public health challenges. The virulence of SARS-CoV-2 and seasonal influenza largely relies on triggering abnormal host immune responses, particularly the production of a cytokine storm, which is notably increased in elderly patients. However, as the mechanisms underlying this age-associated exacerbation remain unclear, we investigated the role of the aging tissue microenvironment in promoting inflammation associated with viral infection. Our research, based on clinical samples, cellular experiments, and mouse models, provides evidence that the aging lung microenvironment induces severe inflammatory responses and leads to tissue damage, with senescent cells playing a crucial role in this process. Further mechanistic insights reveal that elevated levels of downstream inflammatory factors result from a significant and robust activation of the NF-κB pathway. This increase is attributed to the accumulation of reactive oxygen species in senescent cells and subsequent reduced expression of PDLIM2, an E3 ubiquitin ligase regulating P65 degradation. Finally, restoring PDLIM2 significantly inhibits viral infection-mediated inflammatory responses and organ damage in the aging body. Therefore, this study offers a novel perspective by elucidating the molecular mechanism and exploring the therapeutic potential behind viral infection-related inflammatory responses, particularly the mechanism accelerating inflammatory storms in elderly patients post-infection.
## Introduction
Senescent cells cease to divide but remain metabolically active, representing an inevitable process in the natural aging of humans. It is characterized by a blockade of the cell cycle in dividing cells, induced by various forms of cell injury or stress. Accumulated senescent cells drive chronic inflammation through the senescence-associated secretory phenotype (SASP), contributing to various chronic diseases and age-related dysfunction [1]. Immunosenescence, a term encompassing age-related alterations in the immune system, manifests primarily as overall immune response deficiency and chronic systemic inflammation [2]. Viral infections may prematurely stimulate cellular senescence, referred to as virus-induced senescence (VIS) [3,4]. The excessive inflammatory response resulting from the body's natural senescence, combined with virus-induced senescence, is closely associated with heightened inflammatory reactions and severe tissue damage in elderly individuals infected with SARS-CoV-2 [5]. In patients with COVID-19 pneumonia, the leading causes of death include metabolic acidosis, septic shock, pulmonary edema, multi-organ failure, and deep vein thrombosis. These outcomes are linked to severe pathological damage in vital organs like the lungs and kidneys due to inflammatory storms. The disease's occurrence, progression, and deterioration are intricately connected to immune imbalance mechanisms, such as lymphocyte subset disorder and cytokine storm production in the body. Cellular immunity plays a crucial role, with a significant increase in the expression of many inflammatory factors observed in the blood of critically ill patients [6,7]. Therefore, we hypothesize that senescence is intricately linked to viral infections and the subsequent inflammatory outbreaks. However, the role of senescence in the post-viral infection increase in inflammation and its specific underlying mechanism remains unclear.
NF-κB, a nuclear transcription factor, plays a crucial role in regulating inflammatory factor levels through the classical and alternative pathways. The NF-κB dimer, typically composed of the P65/P50 subunits, serves as the key protein in this process, and NF-κB activation is commonly assessed by measuring the protein levels of P65 and phosphorylated P65 [8,9]. NF-κB is a transcription factor that plays a critical role in a variety of biological processes, including immune responses, inflammation, cell growth and survival, and development [10]. While in an inactive state, NF-κB resides in the cytoplasm. Upon activation, the active NF-κB translocates to the nucleus and promotes transcription of NF-κB-dependent genes, such as TNF-α, IL-1β, and IL-6, collectively contributing to inflammatory responses and tissue injuries [11,12]. The activation of NF-κB serves as a pivotal step in the immune system's response to external threats and is tightly regulated to maintain homeostasis under physiological conditions [13]. Many studies have shown that NF-κB is closely related to aging. Senescent cells activate NF-κB signaling, which upregulates senescence-associated secretory phenotype (SASP) factors. A predominant mechanism involves NF-κB-mediated induction of DNA damage and mitochondrial dysfunction during cellular senescence. This leads to cytoplasmic leakage of nuclear and mitochondrial DNA, subsequently activating pattern recognition receptors (PRRs) that further amplify NF-κB activation and pro-inflammatory cytokine secretion [14][15][16]. Targeting NF-κB has been shown to inhibit inflammation and age-related tissue degeneration [16]. However, whether the NF-κB pathway is involved in the inflammatory response and subsequent infection-related diseases during virus infection of senescent cells still needs to be further studied.
This study reveals an intensified inflammatory response in elderly patients post-viral infection, a phenomenon consistently observed in aged mouse models. Furthermore, at the cellular level, the research demonstrates that senescent cells exhibit enhanced viral replication efficiency, amplified virus-induced inflammatory cascades, and aggravated acute lung injury upon SARS-CoV-2 or other virus infections. Mechanistically, senescenceassociated oxidative stress downregulates PDLIM2 expression, impairing ubiquitin-mediated proteasomal degradation of P65. This leads to sustained NF-κB pathway hyperactivation and consequent upregulation of pro-inflammatory cytokines. Critically, PDLIM2 reconstitution attenuates viral-triggered inflammatory pathology and mitigates multiorgan dysfunction in aged organisms. In summary, this research identifies novel potential targets for intervention in diseases arising from viral infections in the geriatric population.
## Results
## Ageing promotes SARS-CoV-2 infection-associated inflammation and lung injury
The cytokine storm, characterized by the rapid release of various cytokines [17], plays a critical role in the pathophysiological progression from mild to severe pneumonia in SARS-CoV-2-infected patients. This phenomenon, particularly pronounced in the elderly, is consistent with established findings that aging exacerbates infectiontriggered inflammatory responses. To investigate the role of senescence in virus-induced inflammation, we analyzed serum IL-6 and TNF-α levels in young and elderly COVID-19 patients (Fig. 1 A) (Supplementary Tables 123). The results revealed a significant elevation in IL-6 and TNF-α levels in the elderly group, consistent with previous reports. Furthermore, analysis of the age distribution of infected patients who died during the same period showed that elderly patients with SARS-CoV-2 infection had a higher mortality rate, also in line with previous findings (Fig. 1B) [18,19]. To evaluate senescence-associated vulnerability to SARS-CoV-2 infection, we established an aged mouse model using 55-week-old mice. The lung tissues of these aged mice exhibited a notable increase in senescence markers P16 and P21 (Supplementary Fig. 1A). Infection experiments with young (8 weeks) and old (55 weeks) h-ACE2 mice revealed a significant upregulation of SARS-CoV-2, IL-1β, and IL-6 mRNA levels in the lungs of old mice (Fig. 1C,D). Immunohistochemistry confirmed severe lung injury, elevated virus replication, and a heightened inflammatory response in aged mice infected with SARS-CoV-2, which was mitigated by pre-treatment with the NF-κB inhibitor SC75741 (Fig. 1E-H). The accumulation of senescent cells is a key factor in aging [20,21]. Therefore, we further evaluated the function of senescent cells in viral infection. Administration of the senolytic drug ABT-263 [22] in aged mice reduced lung senescence levels, alleviated lung injury, inflammation, and virus replication after SARS-CoV-2 infection (Fig. 1I-M). These findings underscore the close association between the accumulation of senescent cells and the occurrence of inflammatory responses and lung damage in SARS-CoV-2 infection.
## Ageing promotes influenza virus and HSV-1 infectionassociated inflammation and lung injury
Since aging can promote inflammation and lung damage associated with SARS-CoV-2 infection, we explored whether this characteristic extends to other viruses, such as seasonal influenza. To investigate, we analyzed clinical data from patients infected with the influenza virus and measured serum C-Reactive Protein (CRP), Serum Amyloid A (SAA), and IL-6 levels in both young and elderly patients (Supplementary Tables 456). The results revealed a significant elevation in CRP, SAA and IL-6 levels in the elderly group (Fig. 2A). To further assess the impact of senescence in influenza virus infection, we intranasally infected C57BL/6 young and aged mice with the influenza virus (H1N1-PR8) (Supplementary Fig. 1B). Seven days post-infection, more severe lung damage and higher H1N1 replication levels were observed in aged mice (Fig. 2B-E). Immunohistochemistry revealed increased immune cell infiltration and higher IL-1β levels in their lungs (Fig. 2D,E). Additionally, protein levels of IL-1β and TNF-α in bronchoalveolar lavage fluid were elevated in aged mice (Fig. 2F). Considering that both COVID-19 and influenza are RNA viruses, we further evaluated whether aging promotes inflammation mediated by other DNA virus infections. Young and aged mice were infected with the DNA virus herpes simplex virus type 1 (HSV-1) for 12 h (Supplementary Fig. 1C). The results showed higher virus replication and increased IL-6 mRNA levels in the lung tissues of old mice (Fig. 2G). Serum IL-6 protein concentration was significantly increased in old mice after HSV-1 infection compared to young mice (Fig. 2H). H&E staining revealed exacerbated organ damage in the lungs of old mice after HSV-1 infection (Fig. 2I,J). Moreover, older mice exhibited significantly reduced motor abilities after viral infection compared to younger mice (Supplementary Videos 1-2). These results collectively suggest that aging promotes inflammation and lung injury associated with SARS-CoV-2, H1N1, and HSV-1 infections.
## Senescent cells promote infection-related inflammation
Data demonstrated that removal of senescent cells was associated with the reduction in inflammatory responses and exacerbation of organ damage induced by SARS-CoV-2 infection (Fig. 1H-J). Therefore, we hypothesized that senescent cells promote the inflammatory response mediated by virus infection. In patients with severe or critical COVID-19, abnormal immune system activation often leads to the release of excessive cytokines by immune cells, triggering a severe systemic inflammatory response and resulting in inflammatory damage. Studies indicate that SARS-CoV-2 infection can affect monocytes and macrophages, leading to programmed cell death [23]. To investigate this phenomenon, we utilized peritoneal macrophages (PMs) from mice. PMs from old mice (16-20 months old) constituted the old group, while PMs from young mice (6-8 weeks old) represented the young group, with clear signs of senescence observed, including increased expression of senescence-associated factors P16 and P21, and elevated inflammation (Supplementary Fig. 2A). Both groups of PMs were infected with H1N1, HSV-1, and vesicular stomatitis virus (VSV) as model viruses. The expression of inflammatory factors IL-1β and IL-6 was analyzed by RT-PCR. Our results revealed a more robust inflammatory response in the old group compared to the young group (Fig. 3A,B). Western blotting further confirmed elevated IL-1β protein levels in the old group (Fig. 3C). Additionally, ELISA demonstrated (See figure on previous page.) Fig. 1 Ageing promotes SARS-CoV-2 infection-associated inflammation and lung injury. A Serum levels of IL-6 and TNF-α in the young group (< 40 years old) and ageing group (> 60 years old) of COVID-19 patients were analyzed. B Age distribution of COVID-19 death patients. C, D Viral replication, IL-1β, and IL-6 levels in lung tissues of young (8-week-old) and old (55-week-old) K18 mice infected with SARS-CoV-2 were detected by RT-PCR. E, F Representative images of H&E staining of lung tissues in young, old, and the inhibitor-treated group of mice infected with SARS-CoV-2, and quantitative analysis of the pathological injury values was performed for each group. G, H Representative IHC images and quantitative data of IL-6 and N-protein in lung tissue of two groups and the inhibitor group of mice infected with the SARS-CoV-2. I-M Young (8-week-old) and aged (55-week-old) K18 mice were infected with SARS-CoV-2 and treated with ABT-263. Lung tissue P16 and P21 mRNA levels (I), representative H&E staining images and histological scores (J, K), representative P16, IL-1β, and N protein IHC images and their quantification plots (L, M) increased IL-6 levels in the supernatant of the old group (Fig. 3D). Having confirmed this phenomenon in naturally senescent PMs, we extended our investigation to stimulation-induced senescent PMs. Senescence can be triggered by various damage-related stimuli, including telomere shortening (replicative senescence), DNA damage (DNA damage-induced senescence), and oncogenic signal transduction (oncogene-induced senescence) [24][25][26]. Doxorubicin (DOX), an antitumor antibiotic that inhibits RNA and DNA synthesis [27], was used to induce damage-induced senescence in cells. After 7 days of DOX treatment, evident signs of senescence were observed, including increased expression of senescenceassociated factors P16 and P21, elevated inflammation, and senescence-associated characteristics detected by β-Gal staining (Supplementary Fig. 2B,C). Subsequently, PMs isolated from young mice were induced to undergo senescence using DOX stimulation and were then infected with HSV-1 and VSV. The levels of inflammatory factors were measured by RT-PCR, while IL-1β protein levels were analyzed by Western blotting, and IL-6 levels in the supernatant were assessed using ELISA. Similar to the results observed in naturally senescent cells, the DOX-induced senescent group exhibited a robust inflammatory response compared to the young group (Fig. 3E-G). These findings suggest that macrophages derived from aged mice significantly promote viral infectioninduced inflammatory responses.
In addition, aged lung tissue also contains other senescent non-immune cells. To explore whether this phenomenon exists in non-immune cells, we also established damage-induced senescence models in Human non-small cell lung cancer cells (A549) and mouse lung fibroblast cells (MLF), with senescence-associated characteristics observed following β-Gal staining (Supplementary Fig. 2D). Subsequently, young and damage-induced senescence A549 and MLF cells were separately infected with HSV-1, VSV, and H1N1 for 16 h. RT-PCR results indicated higher levels of IL-1β and IL-6 in senescent cells (Fig. 3H,I). Meanwhile, testing the level of HSV-1, VSV, or H1N1 replication revealed higher viral replication in old PM cells, compared to the young group (Supplementary Fig. 3A-C). Senescent A549 cells exhibited lower replication of HSV-1 and VSV, while H1N1 replication was higher (Supplementary Fig. 3D). These results suggest that both cell type and virus type may influence the viral replication in senescent cells. Furthermore, it indicates that the excessively high inflammation associated with viral infection in an aging environment is not exclusively driven by viral replication. In summary, these data demonstrate that senescent cells exacerbate virusinduced inflammation across different cell types and viral pathogens.
## Senescent cells promote inflammation through activation of the NF-κB pathway
The results presented above in the research consistently indicate that aging promotes virus-induced inflammation across various viral infections, independent of cell type, suggesting the existence of a common mechanism in this process. According to literature reports, various signals, such as bacterial and multiple virus infections, can activate NF-κB and induce inflammatory responses [28]. Activated NF-κB translocates and binds to its related DNA motif in the nucleus, inducing the transcription of target genes, controlling cytokine production, and ultimately leading to inflammation. Therefore, we further investigated whether the NF-κB pathway mediates inflammatory responses during the aging process. After viral infection, we quantified the expression of total P65 and phosphorylated P65 in cells from two distinct senescence models. Notably, senescent macrophages exhibited significantly increased P65 and phospho-P65 protein levels post-infection compared to the young control group (Fig. 4A,B). Additionally, viral infection itself induced P65 expression (Supplementary Fig. 4A-D). Furthermore, the protein levels of P65 were significantly elevated in lung tissues of old mice (Fig. 4C). To verify whether increased P65 activation promotes infectionrelated inflammation, we treated the cells with the NF-κB inhibitor SC75741 prior to viral infection. Surprisingly, the expression of inflammatory factors significantly decreased in both groups after inhibitor treatment, with almost no difference observed between old and young PMs (Fig. 4D-F). Therefore, the excessive activation of the NF-κB pathway drives an abnormally elevated level of inflammation in senescent cells after viral infection.
## Senescent cells have a robust NF-κB pathway through target P65
As demonstrated in the results above, treatment of PMs with an NF-κB pathway inhibitor led to a significant reduction in the expression of inflammatory cytokines. Thus, abnormal accumulation and activation of P65 (See figure on previous page.) Fig. 3 Senescent cells promote viral replication and inflammatory responses. A, B Natural senescent PM cells were infected with H1N1 (MOI = 1), HSV-1 (MOI = 1), and VSV (MOI = 1), and the relative expression levels of IL-1β and IL-6 mRNA were detected by RT-PCR. C, D Natural senescent PM cells were infected with HSV-1 and VSV, and the expression levels of IL-1β were detected by Western blot, and the IL-6 levels in the cellular supernatant were measured by ELISA. E-G DOX (0.5 µM) induced senescent PM cells infected with HSV-1 and VSV. The expression level of IL-1β was detected by Western blot (E), the relative expression levels of IL-1β and IL-6 mRNA were detected by RT-PCR (F), and the IL-6 level of cellular supernatant was measured by ELISA (G). H, I DOX (0.5 µM) induced senescent A549 (H) or MLF (I) cells infected with H1N1 (MOI = 1), HSV-1 (MOI = 1), and VSV (MOI = 1), and the relative expression levels of IL-1β and IL-6 mRNA were detected by RT-PCR promote viral infection-mediated inflammatory response in senescent cells. Therefore, we further explored the mechanism of P65 hyperactivation in senescent cells. Initially, we assessed the relative expression of P65 at the RNA level in young and old PMs, and found no statistically significant difference between the two groups (Fig. 5A). Given that excessive and prolonged NF-κB activation can disrupt tissue homeostasis, leading to inflammation-related diseases, we focused on NF-κB signaling termination, where protein degradation serves as a key regulatory mechanism [29]. CHX treatment indicates delayed P65 protein degradation in senescent cells (Fig. 5B, Supplementary Fig. 5A). Eukaryotic cells primarily use three pathways for protein degradation: the lysosomal pathway, the ubiquitin-proteasome pathway, and the apoptosis pathway. By inhibiting the lysosomal and ubiquitin-proteasome pathways using CQ (lysosomal inhibitor) and MG132 (proteasome inhibitor), respectively, we determined that P65 degradation primarily occurred through the ubiquitin-proteasome pathway (Fig. 5C,D). Interestingly, the total ubiquitination level was comparable between the two groups (Fig. 5E). Subsequently, we assessed P65 ubiquitination through co-immunoprecipitation (Co-IP), which revealed a decreased level of P65 ubiquitination in the old group (Fig. 5F). These results indicate impaired P65 ubiquitination and consequent protein accumulation in senescent cells. In the ubiquitination process, three enzymes-ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2), and ubiquitin ligase (E3)-are involved. Notably, E3 ubiquitin ligase determines the specific recognition of target proteins, playing pivotal roles in the ubiquitin pathway and participating in various cellular physiological processes by regulating the ubiquitination of regulatory proteins. In our investigation, we screened E3 ligases known to regulate the ubiquitination and degradation of P65, including PPARγ, RBCK1, TRIM7, PDLIM2, and PDLIM7. Among these, the expression of PDLIM2 and PDLIM7 was found to be reduced in old PM cells (Supplementary Fig. 5B), with a significant decrease observed in the protein level of PDLIM2 in senescent PM cells (Supplementary Fig. 5C). Further analysis revealed a significant reduction in both the mRNA and protein levels of PDLIM2 in the lung tissues of old mice (Supplementary Fig. 5D,E). Additionally, induced-senescent MLF cells with VSV and H1N1 infection lead to a notable reduction in PDLIM2 expression (Supplementary Fig. 5F). These results indicate that PDLIM2 is downregulated in senescent cells and aged mouse tissues. Overexpression of PDLIM2 promoted P65 ubiquitination (Fig. 5G). Thus, we propose that PDLIM2 is the primary E3 ligase mediating the ubiquitination and degradation of P65 in this study.
Furthermore, we conducted a preliminary exploration of the factors contributing to decreased PDLIM2 in senescent cells. As aging progresses, aging-associated DNA damage accumulates in the body, and the body struggles to clear the excessive production of reactive oxygen species (ROS), which in turn induces DNA damage, exacerbates aging-related DNA damage, and accelerates cell senescence, forming a vicious cycle [30,31]. We observed a significant increase in ROS levels in senescent macrophages and A549 cells (Fig. 5H,I). Building on this observation, we aimed to investigate the link between ROS accumulation and the decline in PDLIM2 expression in senescent cells. We induced a high ROS environment by stimulating cells with H 2 O 2 . In comparison to untreated controls, we found that the expression of PDLIM2 was reduced in H 2 O 2 -stimulated cells (Supplementary Fig. 5G). Subsequently, by using a ROS inhibitor (N-acetylcysteine, NAC) in the context of H 2 O 2 -induced high ROS levels, we observed a significant increase in PDLIM2 levels after suppressing ROS generation both in control and senescence cells (Fig. 5J-L). Correspondingly, the levels of the inflammatory factors IL-1β and IL-6 showed a significant increase in H 2 O 2 -stimulated senescent cells but a reduction after NAC treatment (Fig. 5M). In summary, these data demonstrate that ROS accumulation decreases PDLIM2 expression in senescent cells, inhibiting the ubiquitination and degradation of P65, thereby leading to increased P65 levels and elevated inflammatory factor levels.
## Recovery of PDLIM2 alleviates virus infection-induced inflammation
Based on the current findings, we proposed that the downregulation of PDLIM2 in the aging environment inhibits the ubiquitination and degradation pathway of P65, thereby mediating the production of more inflammatory factors in senescent cells during virus infection. To further validate this conclusion, we generated a lentivirus expressing PDLIM2 (Supplementary Fig. 5H), infected PM cells for 48 h to overexpress PDLIM2, and then infected them with HSV-1 and VSV model viruses. The results revealed that overexpression of PDLIM2 could reverse the higher levels of inflammation in senescent cells (Fig. 6 A). Upon infecting C57BL/6 young and old mice with H1N1 for 7 days, we found a significant decrease in the protein level of PDLIM2 in lung tissues of old mice compared to the young group (Fig. 6B). Furthermore, we constructed an adeno-associated virus (AAV) expressing PDLIM2 (Fig. 6 C), injected it into the old mice for 21 days to overexpress PDLIM2, and then infected them with H1N1. The results showed that PDLIM2 was overexpressed in the lungs and liver, and reduced the mRNA levels of the inflammatory factor IL-1β in the mouse tissues after virus infection (Fig. 6D,E). Compared to the control group, old mice with PDLIM2 overexpression exhibited milder damage in the lungs and liver after H1N1 infection (Fig. 6F-H), with fewer immune cell infiltrations and lower expression of the inflammatory factor IL-1β (Fig. 6I-K). These findings establish PDLIM2 deficiency as a key driver of excessive inflammation during viral infection in aging, and highlight PDLIM2 restoration as a potential therapeutic strategy.
## Discussion
Cellular senescence, an irreversible state of growth arrest triggered by diverse stress conditions, has been recognized as a fundamental mechanism driving aging and age-related diseases. This state is characterized by profound functional alterations mediated through metabolic, chromatin structural, and transcriptional reprogramming, and is distinctly marked by the senescence-associated secretory phenotype (SASP) [32][33][34][35]. The SASP entails a robust secretion of pro-inflammatory factors, chemokines, and other signaling molecules that perpetuate chronic inflammation and promote the (See figure on previous page.) Fig. 5 Senescent cells have a robust NF-κB pathway through target P65. A The RNA level of P65 was detected in the two groups by RT-PCR. B The protein level of P65 was detected in the two groups after CHX (10 µM) treatment by Western blot. C, D MG132 (20 µM, 4 h) and CQ (10 µM, 6 h) were added to the two groups of cells, respectively, to detect the protein expression level of P65 (C: natural senescent cells, D: DOX-induced senescent cells). E Two groups of cells were infected with HSV-1 and VSV, and the protein ubiquitination level was detected by Western blot. F The ubiquitination level of P65 was detected by Co-IP. G HEK293T cells were co-transfected with P65, PDLIM2, and Ub plasmids, and Co-IP was used to verify the regulatory effect of PDLIM2 on P65 ubiquitination. H, I ROS staining was performed to compare the two groups in PM and A549 cells, and their quantification plots were analyzed. J-M Two groups of cells were treated with NAC (5 mM) and H 2 O 2 , and ROS staining was performed (J, K). The relative expression levels of PDLIM2, IL-1β, and IL-6 were then detected by RT-PCR (L, M) pathogenesis of age-related disorders [36][37][38][39]. Concurrently, age-related decline in immune function, termed immunosenescence, is associated with impaired immune responses, diminished vaccine efficacy, and heightened susceptibility to infections [40,41]. Although recent investigations, particularly in the outbreak of the COVID-19 pandemic, have underscored associations between cellular senescence, viral infections, and inflammatory processes-including exacerbated inflammatory responses in senescent cells, animal models, and elderly populations-the precise mechanistic pathways linking immunosenescence to increased viral susceptibility remain elusive [42,43]. In light of the severe inflammatory pathology and tissue damage observed in aged patients with COVID-19 and influenza, this study utilizes both immune and non-immune cell models to decipher the complex interplay and molecular mechanisms underlying senescence, viral infection, and immune dysregulation, with the ultimate aim of identifying novel therapeutic targets.
Our key findings elucidate the molecular mechanisms underlying exacerbated inflammation in senescent cells after viral infection. The study demonstrates that the nuclear factor-kappa B (NF-κB) signaling pathway is hyperactivated in aged cells and tissues. The NF-κB transcription factor family plays a central role in orchestrating inflammatory and immune responses by translating external inflammatory signals into precise transcriptional instructions that regulate the timely expression of various cytokines and chemokines [44][45][46]. However, persistent and excessive NF-κB activation can lead to severe inflammatory damage, making the timely termination of its activation crucial for preventing such pathology. This study reveals significantly enhanced NF-κB pathway activity accompanied by elevated inflammation in senescent cells and tissues; administration of a specific inhibitor markedly alleviated this inflammatory response. To further investigate the cause of aberrant NF-κB activation in senescent macrophages, we observed a substantial increase in P65 protein levels in virus-infected senescent macrophages, where degradation processes were inhibited, leading to P65 accumulation and triggering a cascade of downstream inflammatory reactions. Notably, we identified reduced expression of the E3 ubiquitin ligase PDLIM2 in senescent cells and mouse tissues, which impedes P65 ubiquitination and degradation, thereby exacerbating the inflammatory response [47]. Moreover, for the first time, we experimentally demonstrated that diminished PDLIM2 levels in senescent cells are associated with excessive ROS production, offering a novel perspective on the molecular intricacies linking viral infection and inflammation.
Another member of the LIM protein family, PDLIM7, shares a similar role as an E3 ubiquitin ligase, inhibiting the NF-κB-mediated inflammatory response by promoting the proteasome degradation of P65 [48]. Our observation of reduced PDLIM2 levels in senescent cells naturally prompts further inquiry into the expression and role of PDLIM7. It would be of significant interest to examine whether PDLIM7 undergoes analogous alterations in senescence and, importantly, how the cooperative regulation of P65 by both PDLIM7 and PDLIM2 is modulated during viral infection in senescent cells. Elucidating these interactions may yield a more integrated understanding of the regulatory network governing NF-κB signaling in senescent cells.
Building upon clinical observations of markedly enhanced inflammation and severe tissue damage in elderly patients with COVID-19 or influenza, this study aimed to thoroughly investigate the interplay among viral infection, cellular senescence, and inflammatory responses. After validating these findings in mouse models and cells infected with SARS-CoV-2, influenza virus, or other viruses, we systematically examined the underlying molecular mechanisms driving accelerated inflammatory storms in aged individuals following viral infection. Our results elucidate how senescent cells exacerbate inflammatory reactions, offering a novel perspective on the molecular processes through which viral infections trigger severe inflammation. Furthermore, we demonstrated that targeted suppression of inflammatory responses in senescent cells can mitigate infection-induced inflammatory storms, highlighting the therapeutic potential of tissue rejuvenation strategies in preventing widespread tissue damage and preserving organ function.
Despite these significant insights, this study has several limitations. First, the research relied primarily on in vitro cell cultures and mouse models; inherent species-specific differences may constrain the translational relevance to human physiology. Second, although we elucidated the role of PDLIM2, the collaborative regulation among multiple E3 ubiquitin ligases-such as PDLIM7-and other regulatory components within the NF-κB pathway requires further comprehensive investigation. Third, while the study focused on viral infections, the effects of other stressors, including bacterial infections or oxidative stress, on NF-κB activation and PDLIM2 function in senescent cells remain to be explored.
Looking ahead, future research could pursue several promising directions. First, it is crucial to develop humanized animal models or initiate clinical studies to validate our findings within the context of human aging and viral infection. Second, a deeper mechanistic investigation into the regulatory networks involving PDLIM2, PDLIM7, and other associated proteins within the NF-κB pathway is warranted. Third, exploring the therapeutic potential of enhancing PDLIM2 or PDLIM7 expression or activity in senescent cells could pave the way for novel interventions in age-related inflammatory diseases and reduce the severity of viral infections in the elderly. Furthermore, examining the crosstalk between diverse stress-induced senescence pathways and their collective influence on immune responses during viral challenges would contribute to a more integrated understanding of the complex interplay among cellular senescence, immunity, and viral pathogenesis.
In conclusion, this study elucidates the molecular mechanisms driving enhanced inflammatory responses in senescent cells during viral infection. The identification of PDLIM2 as a key regulator in this process establishes a foundation for future research directed toward developing novel therapeutic interventions against agerelated inflammatory diseases and viral infections.
## Materials and methods
## Detection of inflammation-related indicators in clinical laboratories
IL-6 and TNF-α testing: venous blood samples were collected in EDTA-K 2 anticoagulant tubes, centrifuged, and measured using the flow cytometer (BD FACSCanto, BD, USA) with supporting reagents. CRP and SAA testing: venous blood samples were collected in EDTA-K 2 anticoagulant tubes and measured using the Specific Protein Analyzer (H780, Mindray, China) with supporting reagents.
## Mice
C57BL/6 wild-type (WT) mice were purchased from Hubei Laboratory Animal Research Center (Wuhan, Hubei, China). All mice were bred under specific pathogen-free conditions at the College of Life Sciences, Wuhan University. K18-hACE2 (B6/JGpt-H11 emICmK18-ACE2 )/Gpt mice were obtained from the Animal Experiment Center of Wuhan University for each experimental procedure.
## Preparation of mice peritoneal macrophages
Peritoneal macrophages (PM) were collected 3 days after broth injection into the mice and cultured in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin.
Specific steps: mice of different ages (young mice, 6-8 weeks old; elderly mice, 16-20 months old) were immersed in 75% ethanol after cervical dislocation. Their skin and subcutaneous tissue were cut along the medioventral line to expose the unilateral peritoneum. The abdominal cavity was rinsed by injecting 6-8 mL of 1 × PBS using a 10 mL syringe. Then, peritoneal fluid was sucked out with a syringe and centrifuged at 1,500 rpm for 5 min. After discarding the supernatant, the pellet was resuspended in DMEM containing 10% FBS and 1% penicillin-streptomycin and cultured at 37 °C in a humidified atmosphere of 5% CO 2 . After 24 h, the cells were replaced with fluid, washed once with PBS, and the floating cells were removed, followed by follow-up experiments. The experiments were performed the next day when the cells were attached.
## Cell lines and cell culture
Human embryonic kidney 293 T (HEK293T) and A549 cell lines were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA) and cultured in DMEM (GIBCO, Grand Island, NY, USA) supplemented with 10% FBS and 1% penicillin-streptomycin at 37 °C in a humidified atmosphere of 5% CO 2 . MLF cells were isolated from the lungs of 6-10-week-old mice. Briefly, lungs were aseptically dissected, minced, and digested with type I collagenase for 30-60 min. After centrifugation, the pellet was resuspended in DMEM with 10% FBS and 1% penicillin-streptomycin and cultured at 37 °C in a humidified atmosphere of 5% CO 2 . The following day, cells were washed with PBS to remove debris and passaged after 3-5 days.
## Reagents
RPMI-1640 and DMEM were obtained from Gibco (Grand Island, NY, USA). Mouse IL-6 ELISA kit was purchased from MABTECH. Senescence β-Galactosidase Staining Kit and Reactive Oxygen Species Assay Kit were purchased from Beyotime (Shanghai, China).The NF-κB inhibitor SC75741 was purchased from MedChemExpress (New Jersey, USA). Antibody against Flag (F3165) (1:2500), HA (H6908) (1:2500) and monoclonal mouse anti-GAPDH (G9295) (1:2000) were purchased from Sigma (St Louis, MO, USA). Monoclonal rabbit anti-P65 (8242 S) (1:2500), monoclonal rabbit anti-p-P65 (3033 S) (1:2500), monoclonal rabbit anti-Ubiquitin (3936) (1:2500) and monoclonal rabbit anti-PDLIM2 (8144 S) (1:1000) were purchased from Cell Signaling Technology (Beverly, MA, USA). IL-1β mouse mAb (AF-401-NA) (1:2500) was purchased from R&D (Minnesota, USA).
(See figure on previous page.) Fig. 6 Recovery of PDLIM2 alleviates virus infection-induced inflammation. A Two groups of cells were infected with lentivirus plenti-PDLIM2 to overexpress PDLIM2. After 48 h, cells were stimulated with HSV-1 and VSV, and the relative expression level of IL-1β was detected by RT-PCR. B Two groups of mice were infected with H1N1 for 7 days, and the expression of PDLIM2, P65, and p-P65 proteins in lung tissue was detected by Western blot. C HEK293T cells were infected with AAV-PDLIM2, and the expression of PDLIM2 was detected by RT-PCR. D, E AAV-PDLIM2 was injected into old mice (16-20 months old), followed by infection with HSV-1. The expression of PDLIM2 and the IL-1β level were then detected using RT-PCR. F-K AAV-PDLIM2 was injected into old mice (16-20 months old), then infected with HSV-1, and their representative H&E staining images and histological scores (F-H), along with representative F4/80 and IL-1β IHC images and their quantification plots (I-K)
Lipofectamine 2000 was purchased from Invitrogen Corporation (Carlsbad, CA, USA).
## RNA extraction and quantitative RT-PCR
Trizol reagent (Invitrogen) was used to extract the total RNA of tissues or cells. Briefly, 500 µL of Trizol reagent per well was added and then lysed for 3 min at room temperature. The lysate was transferred into an RNAase-free EP tube, followed by adding 100 µL chloroform, which was left to react for 3 min after being vortexed for 15 s and centrifuged at 12,000 rpm at 4 ℃ for 5 min. The supernatant was transferred into a new RNAase-free EP tube, and an equal volume of isopropanol was added. The mixture was left to react again for 3 min after being vortexed for 15 s and centrifuged at 12,000 rpm at 4 ℃ for 5 min. Then, the supernatant was removed, and the pellets were washed with 500 µL of 70% ethanol and centrifuged at 12,000 rpm at 4 ℃ for 5 min. Then, the supernatant was removed, and the pellets were air-dried for 5 min, followed by adding 20 µL of ddH₂O for dissolution. The RNA concentration was measured.
The Reverse transcription for cDNA synthesis was performed using 1 µg of RNA by reverse transcriptase (Vazyme Biotech Co., Ltd., China).
The reverse transcription products were used as a template for amplification, and real-time PCR was conducted using the ChamQ SYBR qPCR Master Mix and primers as per the experimental needs.
GAPDH was used as the internal control for target gene expression levels, including IL-6, IL-1β, HSV-1, and VSV. The cycle conditions were: 42 ℃ for 5 min, 95 °C for 10 s, and 40 cycles of 95 °C for 5 s and 60 °C for 30 s.
Primer sequences were as follows:
$$Gene Forward (5'→3') Reverse (5'→3') M-IL-1β CCAGCTTCAAATCTCACAGCAG CTTCTTTGGGTATT- GCTTGG GATC M-IL-6 TTCCATCCAGTTGCCTTCTTG AATTAAGCCTCC- GACTTGTGAA M-P65 TGCGATTCCGCTATAAATGCG ACGGCCAAATC- CGTTCACACC M-PDLIM2 TGGGGCTTCCGAATTAGCG TCCGCGTGTAGCAT- GTTCTC M-PDLIM7 GACCTCTGACAAACAGTTGCT AGTCCCGGAGAC- GTGTATAGC M-PPARγ GGAAGACCACTCGCATTCCTT GTAATCAGCAAC- CATTGGGTCA M-TRIM7 ACAGAAACAGAATGAGAACCTGG GCTCAGTGT- GCTTTTGAACTCC M-RBCK1 GGAGATGAACAGGCTGCTATC GGGAGACCTC- GGGTTTTACTT M-P16 CCCAACGCCCCGAACT GCAGAAGAGCTGC- TACGTGAA M-P21 GTCAGGCTGGTCTGCCTCCG CGGTCCCGTGGA- CAGTGAGCAG Gene Forward (5'→3') Reverse (5'→3') M-GAPDH ACGGCCGCATCTTCTTGTGCA ACGGCCAAATC- CGTTCACACC VSV-N TGATAGTACCGGAGGATTGACGAC CCTTGCAGTGACAT- GACTGCTCTT HSV-1-UL30 CATCACCGACCCGGAGAGGGAC GGGCCAGGCGCTT- GTTGGTGTA H1N1-M AAGACCAATCCTGTCACCTCTGA CAAAGCGTCTAC- GCTGCAGTCC SARS-CoV- 2-N ACCCGCAATCCTGCTAACAA ACGAGAAGAG- GCTTGACTGC$$
## Lentivirus packaging and plasmid transfection
The pcDNA3.1-3×Flag-PDLIM2 plasmid was constructed via standard molecular biotechnology. The murine PDLIM2 sequence was amplified by PCR and cloned into the plenti plasmids to generate the expression plasmid with 3 × FLAG. The plenti-3×Flag-PDLIM2 plasmid was co-transfected with packaged plasmids psPAX2 and pMD2.G into HEK293T cells using Lipofectamine 2000 for lentivirus packaging and production. Then, 48 h later, the medium containing lentiviral particles was collected and used for PM infection. The Flag-PDLIM2, Flag-P65, and HA-Ub plasmids were transfected using PEI at a 1:3 ratio.
## Adeno-associated virus (AAV) overexpression system
The pscAAV-PDLIM2-GFP overexpression system was constructed by inserting the PDLIM2 gene sequence into the pscAAV-GFP vector (Addgene, #32396). Using Lipofectamine 2000, the pscAAV8-PDLIM2-GFP plasmid was co-transfected into HEK293T cells with the packaging plasmids pAdDeltaF6 (Addgene, #112867) and pAAV2/9n (Addgene, #112865) at a ratio of 6:12:7. Cells and culture medium were harvested 48-60 h post-transfection. The cells were subjected to freeze-thaw cycles to release AAV particles, and the supernatant was collected by centrifugation. The supernatant was combined with the culture medium and concentrated using PEG8000.
The concentrated AAV was resuspended in PBS and administered to mice via intraperitoneal injection for 21 days to achieve PDLIM2 overexpression.
## Western blotting
Samples were diluted with 2 × loading buffer and boiled for 10 min, followed by electrophoresis with 8%-12% SDS-PAGE gel (Bio-Rad). Then, the proteins were transferred to a PVDF membrane and blocked with 5% nonfat milk in TBST containing 0.05% Tween 20 for 1 h at room temperature. The membrane was washed three times with TBST for 10 min each and then incubated with primary antibodies diluted in TBST containing 5% BSA at 4 °C overnight. This was followed by incubation with secondary antibodies diluted in TBST containing 3% non-fat milk for 1 h at room temperature, after which the membrane was washed three times with TBST. The images were captured using the LAS-4000 imager.
## Co-immunoprecipitation
Peritoneal macrophages were collected in 1 mL of 1 × PBS and lysed with 500 µL of lysis buffer (50 mM Tris-HCl, pH 7.4-7.5, 150 mM NaCl, 5 mM EDTA, and 1% NP-40) containing phosphatase and protease inhibitors for 30 min. The supernatant was collected, incubated with P65 antibodies, and subsequently incubated with protein G beads at 4 ℃ overnight. The pellet of immunoprecipitation was washed 3 times with a lysis buffer, and the expression of the target proteins was determined by Western blotting.
## Immunohistochemistry (IHC)
Mouse tissues (lung and liver) were preserved in 4% paraformaldehyde and embedded in paraffin to prepare tissue sections, which were stained with staining agents or antibodies, and then observed under a microscope.
## Establishment of mouse infection models
WT mice with matched sex and weight were selected and divided into a young group (6-8 weeks old) and an elderly group (16-20 months old). Mice in the two groups were intraperitoneally injected with HSV-1 (1 × 10 7 PFU/ mouse). After 12 h, orbital blood was collected and centrifuged to get the serum. The concentrations of IL-6 were measured by ELISA. After 24 h, the lungs, spleen, and liver of the mice were isolated to analyze the relative expression of HSV-1, IL-1β, and IL-6. Western blotting was performed to measure P65 expression, and pathological lesions were analyzed by histological staining. For influenza virus infection, all mice were inoculated with the H1N1-PR8 (2 × 10 5 PFU/mouse) via nose drip. After 7 days, the lungs of the mice were isolated to analyze the relative expression of H1N1, IL-1β, and IL-6, and pathological lesions were analyzed by histological staining.
## SARS-CoV-2 infection
K18-hACE2 (B6/JGpt-H11 emICmK18-ACE2 )/Gpt mice with matched sex and weight were divided into a young group (8-week-old) and an elderly group (55-week-old). All mice were nose-dripping infected with the Wuhan strain of SARS-CoV-2 (250-500 PFU/mouse). After 1-7d, the lungs of the mice were isolated to analyze the relative expression of SARS-CoV-2, IL-1β, and IL-6, and pathological lesions were examined by histological staining. For SC75741 treatment, mice were administered 20 mg/kg of the drug intraperitoneally on day 1. Lung tissues were then collected to analyze the relative expression of SARS-CoV-2, IL-1β, and IL-6, and pathological lesions were assessed by histological staining. For ABT-263 treatment, mice wereintranasally infected with the Wuhan strain of SARS-CoV-2 (50 PFU/mouse). Mice were gavaged with ABT-263 (60 mg/kg) on days 2, 4, and 6. After 7 days, the lungs of the mice were isolated to analyze the relative expression of SARS-CoV-2, IL-1β, and IL-6. Pathological lesions were analyzed by histological staining.
## Statistics and data analysis
All data were analyzed using comparable results from at least three independent experiments. Data are represented as mean ± SD. Analyses between two groups were performed using Student's t-test or Welch's t-test (Fig. 1A), and one-way ANOVA was performed to compare multiple groups. ImageJ software was employed for quantitative image analysis. The Smith score was referenced to assess tissue lesion severity based on the lesion area percentage：<25% = 1, 25-50% = 2, 50-75% = 3, >75% = 4. A score of 0 indicates normal. GraphPad Prism 9 software was used for statistical analyses. P-value < 0.05 was considered statistically significant. Statistically significant differences are expressed as follows: *P < 0.05, **P < 0.01, and ***P < 0.001.
## References
1. Camell, Yousefzadeh, Zhu et al. (2021) "Senolytics reduce coronavirus-related mortality in old mice" *Science*
2. Fulop, Larbi, Pawelec et al. (2023) "Immunology of aging: the birth of inflammaging" *Clin Rev Allergy Immunol*
3. Lipskaia, Delval, Sencio et al. (2025) "Virusinduced cellular senescence causes pulmonary sequelae post-Influenza infection" *Aging Cell*
4. Schulz, Hornung, Häder et al. (2023) "Influenza virus-induced paracrine cellular senescence of the lung contributes to enhanced viral load" *Aging Dis*
5. Tsuji, Minami, Hashimoto et al. (2022) "SARS-CoV-2 infection triggers paracrine senescence and leads to a sustained senescenceassociated inflammatory response" *Nat Aging*
6. Ramasamy, Subbian (0299) "Critical determinants of cytokine storm and type I interferon response in COVID-19 pathogenesis" *Clin Microbiol Rev*
7. Malireddi, Sharma, Kanneganti (2024) "Innate immunity in protection and pathogenesis during coronavirus infections and COVID-19" *Annu Rev Immunol*
8. Li, Verma (2002) "NF-kappaB regulation in the immune system" *Nat Rev Immunol*
9. Oeckinghaus, Hayden, Ghosh (2011) "Crosstalk in NF-κB signaling pathways" *Nat Immunol*
10. Napetschnig, Wu (2013) "Molecular basis of NF-κB signaling" *Annu Rev Biophys*
11. Liu, Zhang, Joo et al. (2017) "NF-κB signaling in inflammation"
12. Tang, Xu, Zeng et al. (2021) "Effect of gut microbiota on LPS-induced acute lung injury by regulating the TLR4/NF-kB signaling pathway" *Int Immunopharmacol*
13. Pasparakis (2009) "Regulation of tissue homeostasis by NF-kappaB signalling: implications for inflammatory diseases" *Nat Rev Immunol*
14. Dou, Ghosh, Vizioli et al. (2017) "Cytoplasmic chromatin triggers inflammation in senescence and cancer" *Nature*
15. Taniguchi, Karin (2018) "NF-κB, inflammation, immunity and cancer: coming of age" *Nat Rev Immunol*
16. Victorelli, Salmonowicz, Chapman et al. (2023) "Apoptotic stress causes MtDNA release during senescence and drives the SASP" *Nature*
17. Channappanavar, Perlman (2017) "Pathogenic human coronavirus infections: causes and consequences of cytokine storm and immunopathology" *Semin Immunopathol*
18. Williamson, Walker, Bhaskaran et al. (2020) "Factors associated with COVID-19-related death using opensafely" *Nature*
19. Wu, Leung, Bushman et al. (2020) "Estimating clinical severity of COVID-19 from the transmission dynamics in Wuhan" *Nat Med*
20. Baker, Childs, Durik et al. (2016) "Naturally occurring p16(Ink4a)-positive cells shorten healthy lifespan" *Nature*
21. Childs, Baker, Kirkland et al. (2014) "Senescence and apoptosis: dueling or complementary cell fates" *EMBO Rep*
22. Chang, Wang, Shao et al. (2016) "Clearance of senescent cells by ABT263 rejuvenates aged hematopoietic stem cells in mice" *Nat Med*
23. Junqueira, Crespo, Ranjbar et al. (2022) "FcγR-mediated SARS-CoV-2 infection of monocytes activates inflammation" *Nature*
24. Hayflick, Moorhead (1961) "The serial cultivation of human diploid cell strains" *Exp Cell Res*
25. Di D' Adda, Reaper, Clay-Farrace et al. (2003) "A DNA damage checkpoint response in telomere-initiated senescence" *Nature*
26. Zhu, Blake, Kusuma et al. (2020) "Oncogene-induced senescence: from biology to therapy" *Mech Ageing Dev*
27. Sultana, Domenico, Tseng et al. (2010) "Doxorubicin-induced thymus senescence" *J Proteome Res*
28. Baker, Hayden, Ghosh (2011) "NF-κB, inflammation, and metabolic disease" *Cell Metab*
29. Ghosh (2002) "Missing pieces in the NF-kappaB puzzle" *Cell*
30. Guo, Wang, Wu et al. (2020) "DCAF1 regulates Treg senescence via the ROS axis during immunological aging" *J Clin Invest*
31. Gusarov, Shamovsky, Pani et al. (2021) "Dietary thiols accelerate aging of C. elegans" *Nat Commun*
32. Herbig, Ferreira, Condel et al. (2006) "Cellular senescence in aging primates" *Science*
33. Campisi (2013) "Aging, cellular senescence, and cancer" *Annu Rev Physiol*
34. Van Deursen (2014) "The role of senescent cells in ageing" *Nature*
35. Childs, Durik, Baker et al. (2015) "Cellular senescence in aging and age-related disease: from mechanisms to therapy" *Nat Med*
36. Wajapeyee, Serra, Zhu et al. (2008) "Oncogenic BRAF induces senescence and apoptosis through pathways mediated by the secreted protein IGFBP7" *Cell*
37. Acosta, Banito, Wuestefeld et al. (2013) "A complex secretory program orchestrated by the inflammasome controls paracrine senescence" *Nat Cell Biol*
38. Krtolica, Parrinello, Lockett et al. (2001) "Senescent fibroblasts promote epithelial cell growth and tumorigenesis: a link between cancer and aging" *Proc Natl Acad Sci U S A*
39. Krizhanovsky, Dickins, Hearn et al. (2008) "Senescence of activated stellate cells limits liver fibrosis" *Cell*
40. Mcelhaney, Effros (2009) "Immunosenescence: what does it mean to health outcomes in older adults?" *Curr Opin Immunol*
41. Söderberg-Nauclér, Fornara, Rahbar (2016) "Cytomegalovirus driven immunosenescence-an immune phenotype with or without clinical impact?" *Mech Ageing Dev*
42. Zheng, Liu, Le et al. (2020) "A human circulating immune cell landscape in aging and COVID-19" *Protein Cell*
43. Angel, Pham, Du et al. (2021) "Signatures of immune dysfunction in HIV and HCV infection share features with chronic inflammation in aging and persist after viral reduction or elimination" *Proc Natl Acad Sci*
44. Hayden, Ghosh (2008) "Shared principles in NF-kappaB signaling" *Cell*
45. Natoli, Ostuni (2019) "Adaptation and memory in immune responses" *Nat Immunol*
46. Kizilirmak, Bianchi, Zambrano (2022) "Insights on the NF-κB system using live cell imaging: recent developments and future perspectives" *Front Immunol*
47. Tanaka, Grusby, Kaisho (2007) "PDLIM2-mediated termination of transcription factor NF-kappaB activation by intranuclear sequestration and degradation of the p65 subunit" *Nat Immunol*
48. Jodo, Shibazaki, Onuma et al. (2020) "PDLIM7 synergizes with PDLIM2 and p62/Sqstm1 to inhibit inflammatory signaling by promoting degradation of the p65 subunit of NF-κB" *Front Immunol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12506040&blobtype=pdf | # Naturally occurring mutations in envelope mediate virulence of Usutu virus
Megan Vogt, Jeffrey Marano, William Hanrahan, Seth Hawks, Anne Brown, Sheryl Coutermarsh-Ott, James Weger-Lucarelli, Nisha Duggal
## Abstract
Usutu virus (USUV) is a mosquito-transmitted flavivirus that is closely related to West Nile virus. Recently, USUV emerged in Europe, where it has caused multiple bird die-off events and neuroinvasive disease in humans. Previously, we showed that USUV isolates from Africa cause significantly more severe disease than European isolates in mice. Sequence analysis revealed that the most virulent isolate (Uganda 2012) and the least virulent isolate (Netherlands 2016) differed by 21 amino acids across the viral polyprotein. Here, we sought to identify the viral determinants of and mechanisms for differential virulence. To accomplish this, we used our USUV reverse genetics system and bacteria-free cloning to generate chimeric viruses between Uganda 2012 and Netherlands 2016. Ifnar1 -/-mice infected with a Netherlands 2016 chimera containing all of the structural genes, or just the envelope gene, from Uganda 2012 had significantly higher mortality rates and viremia than those infected with wild-type Netherlands 2016. We were unable to identify a single amino acid in the envelope protein that resulted in significantly increased virulence compared to wild-type Netherlands 2016. These results indicate that multiple mutations in USUV envelope protein contribute to differential virulence between isolates. Through in vitro assays, we discovered that envelope mediates replication kinetics, with fusion occurring more slowly for wild-type Netherlands 2016 compared to viruses containing the envelope from Uganda 2012, suggesting a mechanism for envelope-mediated differential virulence of USUV. These studies provide insights into USUV pathogenic mechanisms, which could be used to evaluate the disease potential of related emerging viruses.
IMPORTANCE Usutu virus (USUV) is currently emerging in Europe, where it has caused numerous mass bird die-off events and neuroinvasive disease in humans. Multiple strains of USUV are circulating throughout Europe, but only some of them have been associated with severe disease in humans. The USUV proteins responsible for and the mechanisms through which they cause severe disease are unknown; however, this information could be invaluable in evaluating disease potential of specific strains and the creation of anti-viral therapies. Here, we swapped genes between USUV strains that cause mild and severe disease and were able to identify a viral protein that mediates virulence. We also discovered that the mild strain of USUV takes significantly longer to complete fusion during viral entry into host cells than the severe strain. This delayed fusion could have impacts on cellular tropism, viral kinetics, susceptibility of the virus to immune responses, and, ultimately, disease severity. KEYWORDS Usutu virus, flavivirus, chimeric viruses, viral determinants, virulence U sutu virus (USUV; Flaviviridae) is a neuroinvasive, mosquito-transmitted flavivirus that is closely related to West Nile virus (WNV). Passerine birds and Culex spp. mosquitoes are the reservoir hosts and vectors for USUV, respectively, with humans
being dead-end hosts (1)(2)(3)(4). Like WNV, USUV can cause neuroinvasive disease in both birds and humans, resulting in mass die-off events in birds and encephalitis or meningitis in humans (5). USUV has followed a similar pattern of global emergence to WNV. Originally isolated in South Africa in 1959, USUV has been spreading throughout Europe since the late 1990s (6)(7)(8)(9), resulting in several mass die-offs of the Eurasian blackbird (Turdus merula) and neuroinvasive disease in humans (5,7,(10)(11)(12)(13)(14)(15)(16)(17)(18).
USUV isolates have been grouped into eight lineages (Africa 1-3 and Europe 1-5) based on genetic similarity of the viral polymerase gene (19). Previously, we compared virulence in the Ifnar1 -/-mouse model of contemporary USUV isolates from multiple lineages: Africa 2 (Senegal 2003 and Spain 2009), Africa 3 (Netherlands [NE] 2016) and Europe 5 (Uganda [UG] 2012) (20). While all isolates grew similarly in vitro, there were significant differences in mortality and viremia (serum virus concentration), with UG2012 and NE2016 exhibiting the most and least virulence, respectively. These two isolates differ by 21 amino acids across the entire polyprotein; specific viral proteins that differ between the two isolates include capsid, envelope, non-structural (NS)1, NS2a, NS3, NS4b, and NS5.
Multiple flavivirus proteins have been implicated in virulence. The envelope protein, a structural protein found on the outer surface of the virion, binds receptors on host cells, mediates fusion of the virus to endosomal membranes, and is a target of host antibody responses. Mutations in envelope protein can significantly impact the virulence of WNV, Murray Valley encephalitis virus, tick-borne encephalitis virus, Japanese encephalitis virus, and others (21)(22)(23)(24)(25)(26)(27)(28). Extracellular NS1 can modulate vascular leakage in tissue-spe cific and viral-specific manners in dengue, Zika, Japanese encephalitis, and yellow fever viruses (29,30). Additionally, mutations in either NS3 (viral protease and helicase) or NS5 (viral polymerase and methyltransferase) can significantly impact the virulence of yellow fever, Japanese encephalitis, dengue, Kunjin, and tick-borne encephalitis viruses (31)(32)(33). For USUV, viral determinants of disease are unknown.
Here, we investigated which viral proteins are responsible for the differential virulence observed between the USUV isolates UG2012 and NE2016. Using bacteria-free cloning techniques, we created chimeric viruses in which multiple genes and single genes were swapped between UG2012 and NE2016. Virulence of these newly created chimeric viruses was evaluated in the Ifnar1 -/-mouse model. We discovered that the envelope protein mediates USUV virulence, and multiple mutations are required for envelopemediated virulence. Finally, we demonstrated that viruses containing envelope protein from UG2012 have faster fusion rates, highlighting a possible mechanism for the differential virulence observed between UG2012 and NE2016. These studies are crucial to understanding the virulence of emerging USUV in Europe and for understanding how the evolution of flaviviruses impacts virulence in mammalian hosts.
## MATERIALS AND METHODS
Additional method details can be found in Supplemental methods.
## Cells and mice used
Vero cells (ATCC; Cercopithecus aethiops) were grown in Dulbecco's modified Eagle medium (DMEM; Corning) supplemented with 5% fetal bovine serum (FBS; Avantor) and 1% penicillin/streptomycin mix (Gibco) at 37°C in a humidified incubator. BHK-21 (ATCC; Mesocricetus auratus) cells were grown in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin mix. Ifnar1 -/-mice were originally purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Additional mice were bred at Virginia Tech. All mice used in studies were 6-11 weeks old. Both male and female mice were used in these studies and were spread evenly across the study groups.
## Generation of infectious clones, chimeric viruses, and point mutants
Infectious clones, chimeric viruses, and point mutants were made using bacteria-free cloning. Briefly, viruses were designed in SnapGene (v.6.0.2 through v.8.0.01). Viral genome fragments were PCR amplified out of previously generated infectious clones or synthesized DNA (Twist Biosciences). Genome fragments were assembled and amplified using the OriCiro kit following the manufacturer's instructions. Additional amplification occurred via rolling circle amplification using the Equiphi polymerase (ThermoFisher). Viruses were rescued in Vero or BHK-1 cells. Viruses were sequence-verified and validated via in vitro growth curves.
## Mouse experiments
Ifnar1 -/-mice were infected in groups of eight with 1,000 plaque-forming units (PFUs) in a 40 µL dose via subcutaneous injection in the rear footpad (20). Mice were monitored daily to assess for signs of disease. Blood was collected on days 3, 4, and 5 post-inocula tion and immediately prior to euthanasia via submandibular bleed. Serum was separated from whole blood using serum separator tubes (Sarstedt or BD Biosciences) via the manufacturer's instructions. Serum samples were stored at -80°C until titration via plaque assay. Mice were euthanized if they lost more than 15% of their starting weight, appeared moribund (hunched back, ruffled fur, refusal to move), or at the termination of the study (day 13). Upon euthanasia, brains were collected from each mouse. Half of each brain was stored at -80°C until titration via plaque assay, while the other half was fixed in 10% buffered formalin for histology.
## Structure modeling methods
Structural models of USUV envelope proteins of NE2016 and UG2012 were generated using Robetta Comparative Modeling (34). NetNGlyc was used to identify glycosylation sites on both the Uganda and Netherlands strains (35).
## Time-to-fusion assay
Virus (1,000 PFUs) was allowed to adsorb for 1 h at 4°C on to Vero cells at 90% confluency. Unbound virus was removed by washing with ice-cold PBS. Then, the media were added, and the cells were incubated at 37°C for 1 h. At 15-min intervals, the media were removed from specified wells and replaced with media containing 200 mM ammonium chloride (NH 4 Cl) solution. At the conclusion of the incubation, NH 4 Cl was removed by washing with PBS. From here, plates were treated as in plaque assay.
## Plaque reduction neutralization titer (PRNT) assay
Neutralizing antibody titers in mouse serum were quantified using a PRNT assay as described in reference 36. Briefly, serum collected from infected mice upon euthanasia was heat-inactivated, and a twofold dilution series was made. An equal volume of USUV was added to serum dilutions. Virus diluted 1:2 in media served as a no-antibody control. Virus and antibody were incubated at 37°C for 1 hour. From here on, plates were treated as in plaque assay.
## Phylogenetic analyses
USUV sequences for phylogenetic analysis were selected from GenBank. Isolation dates for sequenced viruses ranged from 1959 to 2021. Sequences were aligned via Clustal Omega and trimmed to include only the coding region (37). Identical sequences were identified and removed using sRNAtoolbox Helper Tools (38). Phylogenetic trees of the remaining 74 sequences were constructed using PhyML (v.3.0) with 1,000 boot straps (39). The resulting tree was visualized using FigTree (v. 1.4.4). USUV lineages were determined by referencing previous phylogenetic analyses (40).
## Statistical analyses
Statistical analyses were performed using GraphPad Prism (v.10.3.1). Data from in vitro multi-step growth curves, mouse weights, mouse viremia, and fusion timing assay were analyzed using two-way ANOVA followed by multiple comparisons t-tests using Dunnett's adjustment. Survival data were analyzed using the Mantel-Cox test and adjusted for multiple comparisons via the Holm-Sidak method. Brain titers and single-step growth curves were analyzed using one-way ANOVA followed by multiple comparisons t-tests using Dunnett's adjustment. Histology scores were analyzed via the Mann-Whitney test and adjusted for multiple comparisons using the Holm-Sidak method. Correlation analyses were performed using the Spearman correlation coefficient.
## RESULTS
## UG2012ic is more virulent than NE2016ic in Ifnar1 -/-mice
We previously created infectious clones (ic) of the UG2012 (GenBank MN813491.1) and NE2016 (GenBank MN813490.1) isolates (41). The amino acid sequences of these clones each differ by one amino acid in envelope protein from the isolates: NE2016ic, E-K292E; and UG2012, E-I213V. Further analysis of the isolates revealed polymorphisms at those sites with a mixed population of the original sequence and the clone sequence. The clones performed similarly in in vitro and in vivo models compared to parental isolates (41). Here, we adapted the infectious clones (ic) for both UG2012 and NE2016 to use a cell-free method of DNA amplification reliant on replication cycle reaction, which yields supercoiled, plasmid, icDNA that is easier to manipulate and transfect into eukaryotic cells than that generated from traditional bacteria-free cloning methods (42)(43)(44)(45). Following rescue, sequence verification, and validation of growth kinetics (Fig. S1), we assessed the pathogenicity of UG2012ic and NE2016ic in adult Ifnar1 -/-mice, a commonly used mouse model of flavivirus infection (46). We observed significantly lower survival rates in animals infected with UG2012ic (0% survival) compared to NE2016ic (63% survival; P = 0.0001; Fig. 1A; Table S1). To assess the morbidity of mice, we measured weight daily (Fig. 1B; Fig. S2). Overall, the pattern of weight loss between UG2012ic and NE2016ic differed (P < 0.0001). Specifically, animals infected with UG2012ic lost significantly more weight than those infected with NE2016ic on day 4 post-inoculation (P < 0.0001) and day 5 post-inoculation (P < 0.0001). Finally, we assessed viremia in the infected animals (Fig. 1C). Mice infected with UG2012ic had significantly higher titers than those infected with NE2016ic at day 3 (P < 0.0001), day 4 (P < 0.0001), and day 5 post-inoculation (P < 0.0001). Taken together, these data indicate that UG2012ic is more virulent than NE2016ic in Ifnar1 -/-mice. These results are consistent with our previous study in which differential virulence was observed between UG2012 and NE2016 isolates in Ifnar1 -/-mice (20).
## USUV structural proteins mediate virulence
Next, we sought to identify the viral determinants of the differential virulence phenotype that we observed in Ifnar1 -/-mice. NE2016ic and UG2012ic have 97.4% nucleotide similarity and 99.4% amino acid similarity. There are 21 amino acid differences between NE2016ic and UG2012ic polyproteins (Table 1). To narrow down the specific proteins for differential virulence, we generated chimeric viruses in which multi-gene segments of the UG2012ic genome were inserted into NE2016ic. The chimeras made were NE2016 UG2012 C-prM-E , NE2016 UG2012 NS1-NS2a-NS2b , and NE2016 UG2012 NS3-NS5 (Fig. 2A).
To determine which of these gene portions impacts USUV virulence, we infected Ifnar1 -/-mice with the aforementioned chimeras and compared morbidity, mortality, and viral titers to mice infected with NE2016ic (Fig. 2). We observed significantly lower survival in animals infected with NE2016 UG2012 C-prM-E chimera compared to those infected with NE2016ic (P = 0.025; Fig. 2B; Table S1). Animals infected with NE2016 UG2012 C-prM-E experienced significantly more weight loss than those infected with NE2016ic on day 4 (P = 0.011), day 5 (P = 0.0023), day 6 (P = 0.0005), and day 7 (P = 0.016) post-inoculation (Fig. 2C; Fig. S3). Animals infected with the NE2016 UG2012 NS3-NS5 chimera lost significantly less weight than the NE2016ic group on day 5 (P = 0.0007) and day 6 (P = 0.0001) postinoculation. Viremia was assessed on days 3, 4, and 5 post-inoculation (Fig. 2D). The mean viremia of animals infected with NE2016 UG2012 C-prM-E was significantly higher than those infected with NE2016ic on day 3 (P < 0.0001), day 4 (P < 0.0001), and day 5 (P < 0.0001) post-inoculation. Viremia was only detected in the NE2016 UG2012 NS3-NS5 group on day 5 post-inoculation. The mean titer at this time was significantly lower than that of the NE2016ic group (P < 0.0001). Taken together, these results indicate that the structural proteins mediate USUV virulence, as indicated by lower survival rates and higher viremia titers in animals infected with the NE2016 UG2012 C-prM-E chimera compared to NE2016ic.
$$Y 891 R K NS2A 1,268 F L 1,287 A V 1,322 I V 1,334 A V NS3 1,549 L F 1,602 I V 1,983 N S 2,059 I V NS4B 2,460 L F NS5 2,803 A S 3,060 K R$$
## USUV envelope protein mediates differential virulence
Next, we sought to determine which of the structural proteins mediate differential virulence. To do this, we generated chimeras in which the capsid (C) or envelope (E) genes from UG2012ic were inserted into the NE2016ic background (Fig. 3A). We did not make a chimera swapping prM, because no amino acid differences exist between UG2012ic and NE2016ic in prM. To assess virulence of these chimeric viruses, we infected adult Ifnar1 -/-mice with them and compared morbidity, mortality, and viral load to mice infected with NE2016ic. We observed a significantly lower survival rate in animals infected with NE2016 UG2012 E compared to those infected with NE2016ic (P = 0.019; Fig. 3B; Table S1). The survival rate of the NE2016 UG2012 C group was the same as the NE2016ic group. There were no significant differences in weight loss between mice infected with NE2016ic versus NE2016 UG2012 C (Fig. 3C, Fig. S4). In contrast, animals infected with NE2016 UG2012 E lost significantly more weight than those infected with NE2016ic at day 4 (P = 0.023), day 7 (P = 0.0014), day 8 (P = 0.0003), and day 9 (P = 0.038) post-inoculation.
Viremia was assessed on days 3, 4, and 5 post-inoculation (Fig. 3D). The mean titer of the NE2016 UG2012 C group did not significantly differ from the NE2016ic group at any time point. The mean titer of the NE2016 UG2012 E group was significantly higher than that of the NE2016ic group at day 4 (P = 0.0035) and day 5 (P = 0.021) post-inoculation. Taken together, these data indicate that USUV envelope protein plays a key role in mediating virulence, as indicated by lower survival rates and higher viremia in animals infected with NE2016 UG2012 E compared to those in animals infected with NE2016ic.
## Envelope-mediated differential virulence requires multiple amino acids
Next, we sought to identify which amino acid(s) in USUV envelope protein mediate virulence. The envelope proteins in UG2012ic and NE2016ic differ by six amino acids (Table 1). We generated six point mutants of NE2016ic envelope, each with a different amino acid change that matched UG2012ic at that position: E-S52N, E-N88D, E-E179K, E-L231S, E-T238I, and E-T344S. To assess virulence of these mutants, we infected adult Ifnar1 -/-mice and assessed morbidity, mortality, and viremia (Fig. 4). Using spot sequencing on serum samples, we confirmed that the viruses did not revert to wild-type during infection. None of the groups had survival rates that significantly differed from NE2016ic (88%; Fig. 4A; Table S1). Overall, the weight loss curves are similar between groups. Significant differences occur between NE2016ic and E-N88D (P = 0.023), E-T238I (P = 0.029), and E-T344S (P = 0.0075) on day 1 and NE2016ic and E-T344S on day 4 (Fig. 4B; Table S5). However, given the small magnitude of the differences in weight loss, it is unclear whether these are biologically significant. Viremia was assessed on days 3, 4, and 5 post-inoculation (Fig. 4C). Viral titers from animals infected with any of the point mutants were not significantly different from those in the NE2016ic group at any time point; however, on day 4 post-inoculation, there was a trend toward higher viremia in animals infected with the E-T344S mutant compared to NE2016ic (P = 0.08). Taken together, these results show that a single amino acid change in the envelope protein of NE2016 is not sufficient to augment virulence as indicated by the lack of significant changes in mortality and viremia between mice infected with any of the point mutants versus wild-type NE2016ic.
## USUV and pathologies were detected in the brain of animals that succumbed to infection
Since severe USUV disease in humans results in neurological complications, we investiga ted whether severe USUV disease in mice was also accompanied by signs of neuroinvasive disease. To that end, we evaluated viral load and pathology of brains from the mice in Fig. 1 through 4 that succumbed after infection with NE2016 backbone viruses (N = 30). USUV was detected in the brains of 97% of animals that succumbed to infection. The median USUV titer in the brain was 4.8 log 10 PFU/g, with titers from individual animals ranging from 2.6 to 7.8 log 10 PFU/g (Fig. 5A). Because mice succum bed at different days post-infection, we evaluated whether viral titer was associated with day of euthanasia (Fig. 5B). Linear regression (slope = -0.02; P = 0.93) and correlation (Spearman r = -0.007; P = 0.97) analyses revealed that there was no relationship between brain titer and day of euthanasia. Median brain titers were not different between animals that succumbed to NE2016ic versus NE2016 UG2012 C-prM-E (P = 0.72) or NE2016 UG2012 E (P = 0.99) (Fig. 5C).
Next, we evaluated brain pathology on a subset of mice that succumbed to USUV infection (n = 11). Brain sections were evaluated based on evidence of inflammation, necrosis, vascular lesion, and leptomeningitis. Most brain samples exhibited inflammation and/or necrosis. Many samples exhibited some degree of cell death characterized by individual or aggregates of shrunken, hypereosinophilic (intensely pink staining) cells with fragmented nuclei. Parenchymal inflammation most consistently appeared as gliosis, or expansion of glial cells throughout the sections. Occasionally, gitter cells (macrophages in the brain), lymphocytes, plasma cells, or neutrophils were also observed. Similarly, perivascular inflammation and inflammation within the leptome ninges (leptomeningitis) were dominated by mononuclear cells but sporadically included neutrophils. These changes were graded semi-quantitatively to produce a total histologic score. The median total histology score for succumbed animals was 3 (maxi mum score = 12), with scores ranging from 0 to 11 (Fig. 5D). Additionally, the median leptomeningitis score for succumbed animals was 1 (maximum score = 3), with scores ranging from 0 to 3 (Fig. 5E). Inflamed meninges can be visualized in Fig. 5G, a brain slice of an animal infected with NE2016 UG2012 E that had a leptomeningitis score of 3. As a comparison, normal meninges can be visualized in Fig. 5F, a brain slice of an animal infected with NE2016ic that had a leptomeningitis score of 0.
Taken together, these results show that severe USUV infection is associated with neuroinvasive disease in mice, as evidenced by the presence of USUV and pathologies, including leptomeningitis, in brains of mice that succumbed to USUV infection.
## USUV envelope protein impacts viral fusion kinetics but does not impact neutralizing antibody titers
To gain insights into the mechanism through which USUV envelope mediates differential virulence, we modeled and analyzed the structures of envelope protein from both NE2016 and UG2012. With only six amino acid substitutions distinguishing the NE2016 and UG2012 envelope proteins, the secondary and tertiary structural differences between the two strains are minimal (Fig. 6A). These substitutions are not predicted to alter hydrophobicity of the protein or impact the two previously described N-linked glycosylation sites at residues E-118 and E-154 (47). Two substitutions (E-S52N and E-N88D) are predicted to affect other putative N-linked glycosylation sites, potentially enabling a glycosylation site at residue E-52 in UG2012 or at residue E-88 in NE2016. These glycosylation sites have yet to be experimentally validated and are not predicted to cause any observable secondary or tertiary structural changes in the proteins.
Because four out of six envelope mutations are located in DII, which contains the fusion domain, we sought to investigate whether the USUV envelope impacts fusion kinetics by performing a time-to-fusion assay in Vero cells with NE2016ic, UG2012ic, and NE2016 UG2012 E using ammonium chloride (NH 4 Cl) as a fusion inhibitor (48,49). When we compared the results of individual viruses with each other, we found that NE2016 UG2012 E -infected cells treated with NH 4 Cl at 15 min post adsorption had a significantly higher infection rate than NE2016-infected cells treated at the same time point (Fig. 6B). Additionally, both NE2016 UG2012 E and UG2012ic infected cells treated with NH 4 Cl at 30 (purple diamonds) were quantified using PRNT assay, using the same viruses with which mice were originally infected (i.e., NE2016ic or NE2016 UG2012 E ). Mice that were still viremic at the time of euthanasia were excluded. PRNT 50 titers are expressed as the log 10 reciprocal dilution. Horizontal lines represent the mean of each group.
Data were analyzed via student's t-test. min, minute; NH 4 Cl, ammonium chloride; **P < 0.01; ****P < 0.0001; n.s., not significant. minutes post adsorption had significantly higher infection rates than NE2016-infected cells (Fig. 6B). These data indicate that fusion begins earlier in cells infected with UG2012ic and NE2016 UG2012 E , both of which contain the envelope protein from UG2012, than in cells infected with NE2016ic.
Next, we investigated whether fusion kinetics impacted replication kinetics early in infection. To test this, we performed single-step growth curves of NE2016, UG2012, and NE2016 UG2012 E at an MOI of 10 in Vero cells. At twelve hours post-inoculation, we observed significantly higher viral titers in the supernatant of cells infected with UG2012ic (P < 0.0001) or NE2016 UG2012 E (P = 0.01) compared to cells infected with NE2016ic (Fig. 6C). These results indicate that USUV envelope protein impacts singlecycle replication kinetics in vitro.
Envelope protein is also the main target of neutralizing antibodies, and even a single mutation in envelope protein can impact the production and function of neutralizing antibodies (21,50,51). Therefore, we investigated whether the virulence following NE2016 UG2012 E infection was a result of lower neutralizing antibody titers than those generated during NE2016ic infection. We quantified neutralizing antibodies in serum collected at euthanasia from NE2016ic and NE2016 UG2012 E -infected mice. The mean PRNT 50 titers were similar between NE2016ic and NE2016 UG2012 E -infected mice (Fig. 6D). These results indicate that the differential virulence between NE2016ic and NE2016 UG2012 E is not due to differences in neutralizing antibody titers.
## NE2016 envelope mutations are found in nearly all Africa 3 lineage viruses
Finally, we sought to identify whether the six amino acid differences in the envelope protein of NE2016 compared to that of UG2012 were unique to NE2016. We performed phylogenetic analysis on over 70 full genome sequences publicly available in GenBank (Table S2) and found multiple USUV isolates that matched NE2016 at all six envelope residues evaluated here. All of these sequences cluster together (Fig. 7) and were identified to be within the Africa 3 lineage of USUV. A small subset of Africa 3 lineage isolates matches NE2016 at five out of six envelope residues; the residue that differs in this subset is residue E-231, and it is a serine (matching UG2012) as opposed to a leucine (matching NE2016). All other sequences that were analyzed in Africa 2, Europe 1, Europe 2, Europe 3, and Europe 5 lineages matched UG2012 at the six envelope residues. This suggests that the decreased virulence of NE2016 is likely a shared feature across many isolates of the Africa 3 lineage.
## DISCUSSION
USUV is a flavivirus that can cause neuroinvasive disease in humans. Previously, we showed that two USUV strains, NE2016 and UG2012, have different virulence phenotypes in Ifnar1 -/-mice (20). Here, we sought to determine the viral determinants of this differential virulence phenotype. We discovered that envelope protein mediates USUV virulence, potentially through differential fusion kinetics. Additionally, we determined that USUV can cause neuroinvasive disease in a mouse model.
We used the Ifnar1 -/-mouse model for these studies, which is a common mouse model for studying flavivirus pathogenesis and has been used by us and others to study USUV infection (46). Wild-type mice have an inherent resistance to several flaviviruses, including USUV, and antagonism of the interferon response is necessary to infect mice with USUV (52,53). USUV infection in Ifnar1 -/-mice tends to be lethal, with the exception of infection with NE2016. Recently, we have described a non-lethal mouse model of USUV infection in which wild-type animals were transiently immuno suppressed via treatment with an antibody that temporarily blocks the type I interferon receptors; however, NE2016 causes little to no viremia in this model (41). It should be noted, however, that humans who develop severe USUV disease tend to be older, immunosuppressed, and/or have comorbidities, all of which can have significant impacts on anti-viral immune responses (5). Additionally, anti-interferon antibodies have been detected in individuals experiencing severe disease following infection with USUV or other flaviviruses (54,55). This evidence supports the use of an immunocompromised mouse model to study USUV infection.
USUV can cause significant neurological complications, including meningitis and encephalitis, in humans. In the study presented here, we observed clear signs of brain pathology, including leptomeningitis, indicating that the Ifnar1 -/-mouse model can be used to study USUV neuroinvasive disease. To our knowledge, ours is the first study to visualize USUV-associated meningitis in mice that are infected peripherally. At this point, it is unknown whether USUV neuroinvasion or meningitis occurs in mice without severe clinical disease. Because our studies were designed to assess survival and not neuroinvasion, we were unable to determine whether USUV envelope protein mediates neuroinvasive disease. Future studies will incorporate planned euthanasia between days 6 and 11 post-infection to determine whether USUV can be detected in the brains of mice that experience milder USUV disease.
Using a chimeric virus approach, we identified envelope protein as a key mediator of USUV virulence; however, envelope is likely not the only viral determinant of severe USUV disease. In our studies, the mortality rates and viremia observed in mice infected with either NE2016 UG2012 C-prM-E or NE2016 UG2012 E were significantly higher than those of mice infected with NE2016ic, but neither chimera completely recapitulates the disease phenotype of UG2012ic. Epistatic interactions between envelope protein and another viral protein may be necessary to fully recapitulate the UG2012ic phenotype. The envelope proteins of NE2016 and UG2012 differ by six amino acids. Using point mutants, we were unable to identify a single amino acid responsible for differential virulence between NE2016 and UG2012; however, we did identify two residues, E-231 and E-344, which trended toward having significant impacts on survival and viremia, respectively. Future studies will utilize double mutants to determine the minimal changes necessary in USUV envelope protein to impact virulence.
Since four out of six amino acid residues that differ between UG2012 and NE2016 envelope proteins are located in the domain that contains the fusion loop (DII), we hypothesized that envelope protein mediates altered fusion kinetics. Using a time-tofusion assay, we determined that NE2016 took longer to fuse than either UG2012 or NE2016 UG2012 E . Since fusion rates are dependent on pH of the endosome and endosomal membrane composition (56), the difference in fusion rates between USUV viruses may be more profound in cells with slower endosomal maturation or altered membrane composition. Slower fusion may render a virus more susceptible to intracellular innate immune factors, particularly IFITM3, an interferon-inducible transmembrane protein that localizes to the membranes of late endosomes and can inhibit viral fusion of Influenza A virus, WNV, and dengue virus (57,58). Additionally, in vivo, delayed fusion and lower initial viral output may limit NE2016's ability to effectively escape the inoculation site and disseminate through the host before induction of innate immune responses, ultimately leading to reduced viral titers and less virulence. Future studies will address the mechanism of delayed fusion by NE2016 and its impact on cellular tropism and innate immune responses in vitro and in vivo.
We modeled the envelope protein structures of both UG2012ic and NE2016. The overall structures are highly similar, with the six amino acid differences having no discernible impact on the secondary or tertiary structures. A limitation of these structural studies is that we only modeled a single, static dimer of envelope protein. Like most flaviviruses, the USUV virion contains 180 copies of envelope protein that arrange themselves in 90 dimers in the prefusion form (47). During fusion, these dimers rearrange to form trimers (47). Since overall fusion kinetics differ between UG2012 and NE2016, we hypothesize that the conformation change from dimer to trimer occurs at a slower rate with NE2016 than with UG2012. Future structural studies will include dynamic modeling, which will allow us to ascertain whether the amino acid changes between NE2016 and UG2012 envelope proteins impact interactions between envelope dimers or the transition between the dimeric pre-fusion state and the trimeric fusion state. The two putative glycosylation sites in USUV envelope protein (E-154 and E-118; [47]) are not predicted to be impacted by the amino acid changes between NE2016 and UG2012. However, our structural analyses predicted additional glycosylation sites in both NE2016 (at E-88) and UG2012 (at E-52). These sites have yet to be validated experimen tally. Modulation of envelope glycosylation sites has been shown to alter virulence in other flaviviruses, including Kunjin and Tembusu viruses (26,59). Therefore, it would be pertinent to study envelope glycosylation in relation to USUV virulence.
In this study, we focused on the 21 amino acid differences between NE2016 and UG2012; however, there are 268 nucleotide differences between the NE2016 and UG2012 genomes that do not result in amino acid changes: 1 in the 5′ untranslated region (UTR), 15 in the 3′UTR, and 252 silent mutations throughout the coding sequence. Because the secondary (hairpins, pseudoknots, etc.) and tertiary (long-range interac tions) structures of flavivirus RNA are crucial for viral replication and antagonizing host immune responses, even silent mutations may impact flavivirus virulence (60, 61). Indeed, altering codon optimization of flavivirus genomes, which introduces numerous silent mutations, results in viral attenuation and is being explored as a method of creating vaccine strains of flaviviruses (61)(62)(63). Thus, it is possible that silent mutations between NE2016 and UG2012 could contribute to differential virulence. Future studies could address whether NE2016 and UG2012 differ in RNA structure and whether the UTRs impact differential virulence.
Phylogenetic analysis revealed that the six amino acids in the envelope protein of NE2016 that differ from UG2012 (E52, E88, E179, E231, E238, and E344) are consistently found in USUV strains from the Africa 3 lineage. USUV sequences from all other lineages (Africa 2, Europe 1, Europe 2, Europe 3, and Europe 5) matched UG2012 at these six residues in envelope. Here and previously, we have shown that NE2016 is less virulent in a mouse model of disease than UG2012 and other USUV isolates (Spain 2009, Senegal 2003, and South Africa 1959[SAAR-1776]) (20). However, we have previously shown that NE2016 is not attenuated in two bird models-2-day-old chickens (Gallus gallus) and wild-caught house sparrows (Passer domesticus)-and in Culex quinquefasciatus mosquitoes (64,65). Thus, while the amino acid changes in envelope between NE2016 and UG2012 may contribute to differential virulence in mammalian models, they appear to have little effect on the ability of NE2016 to replicate in bird models and mosqui toes. Humans and other mammals are dead-end hosts for USUV. Therefore, there is no evolutionary pressure for USUV to replicate efficiently or modulate virulence in mammals.
USUV is a flavivirus capable of causing severe disease in humans. The mechanisms through which USUV causes disease in humans are not understood. Through investiga tion of USUV strains that cause differing levels of virulence in a mouse model, we identified envelope as a viral factor that contributes to virulence and cell fusion. Future studies will investigate whether these specific components of USUV alter neuroinvasion. This information is key to (i) being able to predict the likelihood of a given USUV isolate to cause neuroinvasive disease and (ii) the development of specific anti-viral treatments to prevent or lessen USUV neuroinvasive disease. Furthermore, information gained from studying USUV may be applied to other closely related flaviviruses, like WNV or St. Louis encephalitis virus.
## References
1. Fros, Miesen, Vogels et al. (2015) "Comparative Usutu and West Nile virus transmission potential by local Culex pipiens mosquitoes in north-western" *Europe. One Health*
2. Nikolay (2015) "A review of West Nile and Usutu virus co-circulation in Europe: how much do transmission cycles overlap?" *Trans R Soc Trop Med Hyg*
3. Durand, Haskouri, Lowenski et al. (2016) "Seroprevalence of West Nile and Usutu viruses in military working horses and dogs, Morocco, 2012: dog as an alternative WNV sentinel species?" *Epidemiol Infect*
4. Diagne, Ndione, Paola et al. (2019) "Usutu virus isolated from rodents in Senegal" *Viruses*
5. Clé, Beck, Salinas et al. (2019) "Usutu virus: a new threat?" *Epidemiol Infect*
6. Williams, Simpson, Haddow et al. (1964) "The Isolation of West Nile virus from man and of Usutu virus from the bird-biting mosquito mansonia aurites (Theobald) in the Entebbe Area of Uganda" *Ann Trop Med Parasitol*
7. Weissenböck, Kolodziejek, Url et al. (2002) "Emergence of Usutu virus, an African mosquito-borne flavivirus of the Japanese encephalitis virus group" *Europe. Emerg Infect Dis*
8. Engel, Jöst, Wink et al. (2016) "Reconstruction of the evolutionary history and dispersal of Usutu virus, a neglected emerging Arbovirus in" *Europe and Africa. mBio*
9. Roesch, Fajardo, Moratorio et al. (2019) "Usutu virus: an Arbovirus on the rise" *Viruses*
10. Lühken, Jöst, Cadar et al. (2017) "Distribution of Usutu virus in Germany and its effect on breeding bird populations" *Emerg Infect Dis*
11. Caracciolo, Mora-Cardenas, Carletti et al. (2020) "Comprehensive response to Usutu virus following first isolation in blood donors in the Friuli Venezia Giulia region of Italy: development of recombinant NS1-based serology and sensitivity to antiviral drugs" *PLoS Negl Trop Dis*
12. Carletti, Colavita, Rovida et al. (2017) "Expanding Usutu virus circulation in Italy: detection in the Lazio region, central Italy" *Euro Surveill*
13. Nagy, Mezei, Nagy et al. (2018) "Extraordinary increase in West Nile virus cases and first confirmed human Usutu virus infection in Hungary" *Euro Surveill*
14. Pacenti, Sinigaglia, Martello et al. (2018) "Clinical and virological findings in patients with Usutu virus infection, northern Italy" *Euro Surveill*
15. Percivalle, Cassaniti, Sarasini et al. (2020) "West Nile or Usutu virus? a three-year follow-up of humoral and cellular response in a group of asymptomatic blood donors" *Viruses*
16. Zaaijer, Slot, Molier et al. (2019) "Usutu virus infection in Dutch blood donors" *Transfusion*
17. Pecorari, Longo, Gennari et al. (2009) "First human case of Usutu virus neuroinvasive infection"
18. Santini, Vilibic-Cavlek, Barsic et al. (2013) "clinical and laboratory features"
19. Cadar, Lühken, Van Der Jeugd et al. (2016) "Widespread activity of multiple lineages of Usutu virus" *Euro Surveill*
20. Kuchinsky, Hawks, Mossel et al. (2020) "Differential pathogenesis of Usutu virus isolates in mice" *PLoS Negl Trop Dis*
21. Goo, Vanblargan, Dowd et al. (2017) "A single mutation in the envelope protein modulates flavivirus antigenicity, stability, and pathogenesis" *PLoS Pathog*
22. Zhang, Li, Woodson et al. (2006) "A mutation in the envelope protein fusion loop attenuates mouse neuroinvasiveness of the NY99 strain of West Nile virus" *Virology*
23. Yang, Yang, Li et al. (2017) "Envelope protein mutations L107F and E138K are important for neurovirulence attenuation for Japanese Encephalitis virus SA14-14-2 strain" *Viruses*
24. Monath, Arroyo, Levenbook et al. (2002) "Single mutation in the flavivirus envelope protein hinge region increases neurovirulence for mice and monkeys but decreases viscerotropism for monkeys: relevance to development and safety testing of live, attenuated vaccines" *J Virol*
25. Huang, He, Zhang et al. (2024) "The mutation of Japanese encephalitis virus envelope protein residue 389 attenuates viral neuroinvasiveness" *Virol J*
26. Yan, Wang, Shi et al. (2022) "A single mutation at position 120 in the envelope protein attenuates Tembusu virus in ducks" *Viruses*
27. Lee, Lobigs (2000) "Substitutions at the putative receptor-binding site of an encephalitic flavivirus alter virulence and host cell tropism and reveal a role for glycosaminoglycans in entry" *J Virol*
28. Mcminn (1997) "The molecular basis of virulence of the encephalito genic flaviviruses" *J Gen Virol*
29. Puerta-Guardo, Glasner, Espinosa et al. (2019) "Flavivirus NS1 triggers tissue-specific vascular endothelial dysfunction reflecting disease tropism" *Cell Rep*
30. Lo, Roodsari, Tin et al. (2022) "Molecular determinants of tissue specificity of flavivirus nonstructural protein 1 interaction with endothelial cells" *J Virol*
31. Bondaryuk, Kulakova, Potapova et al. (2022) "Genomic determinants potentially associated with clinical manifestations of human-pathogenic tick-borne flaviviruses" *Int J Mol Sci*
32. Yoshii, Sunden, Yokozawa et al. (2014) "A critical determinant of neurological disease associated with highly pathogenic tick-borne flavivirus in mice" *J Virol*
33. Hurrelbrink, Mcminn (2003) "Molecular determinants of virulence: the structural and functional basis for flavivirus attenuation" *Adv Virus Res*
34. Baek, Dimaio, Anishchenko et al. (2021) "Accurate prediction of protein structures and interactions using a three-track neural network" *Science*
35. Gupta, Brunak (2002) "Prediction of glycosylation across the human proteome and the correlation to protein function"
36. Salgado, Hawks, Frere et al. (2021) "West Nile virus vaccination protects against Usutu virus disease in mice" *Viruses*
37. Madeira, Madhusoodanan, Lee et al. (2024) "The EMBL-EBI job dispatcher sequence analysis tools framework in 2024" *Nucleic Acids Res*
38. Aparicio-Puerta, Lebrón, Rueda et al. (2019) "sRNAbench and sRNAtoolbox 2019: intuitive fast small RNA profiling and differential expression" *Nucleic Acids Res*
39. Guindon, Dufayard, Lefort et al. (2010) "New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0" *Syst Biol*
40. Zhou, Xing, Zhou et al. (2022) "A comprehensive analysis of Usutu virus (USUV) genomes revealed lineage-specific codon usage patterns and host adaptations" *Front Microbiol*
41. Bates, Chuong, Hawks et al. (2021) "Development and characterization of infectious clones of two strains of Usutu virus" *Virology*
42. Marano, Cereghino, Finkielstein et al. (2023) "An in vitro workflow to create and modify infectious clones using replication cycle reaction" *Virology*
43. Mukai, Yoneji, Yamada et al. (2020) "Overcoming the challenges of megabase-sized plasmid construction in Escherichia coli" *ACS Synth Biol*
44. Hasebe, Narita, Hidaka et al. (2018) "Efficient arrangement of the replication fork trap for in vitro propagation of monomeric circular DNA in the chromosome-replication cycle reaction" *Life (Basel)*
45. Su'etsugu, Takada, Katayama et al. (2017) "Exponential propagation of large circular DNA by reconstitution of a chromosomereplication cycle" *Nucleic Acids Res*
46. Marín-Lopez, Calvo-Pinilla, Moreno et al. (2019) "Modeling arboviral infection in mice lacking the interferon Alpha/Beta receptor" *Viruses*
47. Khare, Klose, Fang et al. (2021) "Structure of Usutu virus SAAR-1776 displays fusion loop asymmetry" *Proc Natl Acad Sci*
48. Noval, Rodriguez-Rodriguez, Rangel et al. (2019) "Evolution-driven attenuation of Alphaviruses highlights key glycopro tein determinants regulating viral infectivity and dissemination" *Cell Rep*
49. Gollins, Porterfield (1986) "The uncoating and infectivity of the flavivirus West Nile on interaction with cells: effects of pH and ammonium chloride" *Journal of General Virology*
50. Roy (2020) "Structural and molecular analyses of functional epitopes and escape mutants in Japanese encephalitis virus envelope protein domain III" *Immunol Res*
51. Gromowski, Roehrig, Diamond et al. (2010) "Mutations of an antibody binding energy hot spot on domain III of the dengue 2 envelope glycoprotein exploited for neutralization escape" *Virology*
52. Blázquez, Escribano-Romero, Martín-Acebes et al. (2015) "Limited susceptibility of mice to Usutu virus (USUV) infection and induction of flavivirus cross-protective immunity" *Virology*
53. Martín-Acebes, Blázquez, Cañas-Arranz et al. (2016) "A recombinant DNA vaccine protects mice deficient in the alpha/beta interferon receptor against lethal challenge with Usutu virus" *Vaccine*
54. Gervais, Bastard, Bizien et al. (2024) "Auto-Abs neutralizing type I IFNs in patients with severe Powassan, Usutu, or Ross River virus disease" *J Exp Med*
55. Gervais, Marchal, Fortova et al. (2024) "Autoantibodies neutralizing type I IFNs underlie severe tick-borne encephalitis in approximately 10% of patients" *J Exp Med*
56. Moesker, Rodenhuis-Zybert, Meijerhof et al. (2010) "Characterization of the functional requirements of West Nile virus membrane fusion" *J Gen Virol*
57. Brass, Huang, John et al. (2009) "The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus" *Cell*
58. Li, Markosyan, Zheng et al. (2013) "IFITM proteins restrict viral membrane hemifusion" *PLoS Pathog*
59. Alsaleh, Khou, Frenkiel et al. (2016) "The E glycoprotein plays an essential role in the high pathogenicity of European-Mediterranean IS98 strain of West Nile virus" *Virology*
60. Abram, Landry, Wang et al. (2024) "The myriad roles of RNA structure in the flavivirus life cycle" *RNA Biol*
61. Fernández-Sanlés, Ríos-Marco, Romero-López et al. (2017) "Functional information stored in the conserved structural RNA domains of flavivirus genomes" *Front Microbiol*
62. Chin, Kong, Zhu et al. (2023) "Flavivirus genome recoding by codon optimisation confers genetically stable in vivo attenuation in both mice and mosquitoes" *PLoS Pathog*
63. Stauft, Song, Gorbatsevych et al. (2019) "Extensive genomic recoding by codon-pair deoptimization selective for mammals is a flexible tool to generate attenuated vaccine candidates for dengue virus 2" *Virology*
64. Kuchinsky, Frere, Heitzman-Breen et al. (2021) "Pathogenesis and shedding of Usutu virus in juvenile chickens" *Emerg Microbes Infect*
65. Kuchinsky, Marano, Hawks et al. (2022) "North American house sparrows are competent for Usutu virus transmission" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12646008&blobtype=pdf | # CD36 is required for human sapovirus propagation
Tomoichiro Oka, Yuko Okemoto-Nakamura, Hirotaka Takagi
## Abstract
Human sapoviruses (HuSaVs), which cause acute gastroenteritis, are highly diverse. Fifteen HuSaV genotype strains (GI.1-7, GII.1-5, -8, and GV.1-2) were efficiently propagated in the human duodenum-derived cell line HuTu80 when supplemented with conjugated bile acid (T.
major capsid protein (VP1), which constitutes the surface structure of viral particles (10)(11)(12)(13)(14). Recently, we successfully propagated and serially passaged HuSaV strains of 15 genotypes using the high-passage-number human duodenal cell line HuTu80 supple mented with conjugated bile acids (12,15). HuSaV propagation in Caco-2 cells and their derivative was recently reported (16). HuSaV replication has also been reported in human small intestine enteroids (17) and in induced pluripotent stem (iPS)-derived human intestinal epithelial cells (18). However, the essential cellular factor for HuSaV propagation remains unknown.
In this study, we identified a cellular protein factor that was critical for HuSaV propagation.
## RESULTS
## Candidate host factors for HuSaV propagation in HuTu80 cells
We identified candidate cellular factors that might be essential for HuSaV propagation by the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR associated protein 9 (Cas9) based genome-wide knockout screening of a clone of HuTu80 cells (HuTu80-3C3) that supported the propagation of 15 HuSaV genotypes (GI.1-7, GII.1-5, -8, GV.1, and GV.2), when supplemented with conjugated bile acids, sodium glycocholate (GlyCA) or sodium taurocholate (TauCA) (Fig. 1A), similar to that in HuTu80 cells (Fig. S1A) (12). In contrast to HuTu80 cells, the HuSaV GII.4 strain could propagate in HuTu80-3C3 when supplemented with GlyCA (Fig. 1A; Fig. S1A). Most HuSaV strains tested, other than GV strains, had a marked cytopathic effect (CPE) when supplemented with bile acids (Fig. 1B). Among them, two GI strains, GI.1 and GI.6, and two GII strains, GII.2 and GII.3, were selected as representative strains. Although GlyCA and TauCA supported these HuSaV propagation (Fig. 1A), TauCA was selected for its lower cost (1/5).
Then, the number of reads of single guide (sg)RNAs for knockout genes from cells that survived repeated inoculation with these HuSaVs was counted. Upset plot analysis of the top 100 genes revealed guide RNAs targeting three genes were consistently detected in four HuSaV strains (GI.1, GI.6, GII.2, or GII.3) resistant cells (Fig. 2A). These guide RNAs targeted CD36, myeloid-associated differentiation marker (MYADM), and a non-targeting control (Fig. 2B).
The number of sgRNA reads in HuSaV-resistant HuTu80 cells was higher for the CD36 gene than for the MYADM gene except for HuSaV GII.3 (Fig. 2B). In this study, we mainly focused on CD36 (see Discussion).
## Knockout of CD36 abolishes HuSaV propagation ability in HuTu80 cells
To evaluate whether CD36 affects the propagation ability of HuSaV, we knocked out (KO) the CD36 gene in HuTu80-3C3, as well as in a parent HuTu80 cells. GI.1, GI.6, GII.2, and GII. 3 HuSaV strains used in the CRISPR screening propagated in HuTu80 and HuTu80-3C3 cells when supplemented with GlyCA or TauCA, but not in HuTu80-CD36-KO and HuTu80-3C3-CD36-KO (Fig. 3). Gene modifications in HuTu80-3C3 CD36 knockout clone 11C cells (HuTu80-3C3-CD36-KO-11C) were confirmed by genome sequencing (Fig. 4A andB). There was no increase in HuSaV VP1 protein levels in the culture supernatant and no obvious CPE was observed for any of the 15 HuSaV genotypes used to infect the HuTu80-3C3-CD36-KO-11C cells with or without bile acid supplementation (Fig. 4C andD).
Therefore, we concluded that CD36 is required for HuSaV propagation.
## Restoration of HuSaV propagation capacity by the re-expression of human CD36 in knockout HuTu80 cells
To eliminate the possibility of the off-target effects of CRISPR/Cas9, we confirmed that the reintroduction of the human CD36 gene restored HuSaV propagation ability. To achieve this, a synthetic CRISPR/Cas9 recognition sequence-synonymous mutated CD36 isoform 1 gene with a tag coding sequence added to the C-terminus of the expressed protein was introduced into HuTu80-3C3-CD36-KO-11C cells. Although endogenous CD36 expression in HuTu80-3C3 cells could not be detected with the anti-CD36 antibody used in this study (Fig. 5A). The expression of the introduced CD36 was detectable by anti-FLAG and anti-CD36 antibodies in human CD36 isoform 1 gene-introduced HuTu80-3C3-CD36-11C cells (HuTu80-3C3-CD36-KO-11C/Re1) (Fig. 5A). CD36 gene expression was similarly confirmed in HuTu80 3C3 cells with or without bile acid supplementation by reverse transcription-PCR (RT-PCR), but was not detected in HuTu80-3C3-CD36-KO-11C cells (Fig. S2). This suggests the undetectability of endoge nous CD36 in HuTu80 3C3 cells at the protein level is likely due to the insufficient sensitivity of the anti-CD36 antibody to detect endogenous levels of CD36, because a protein band corresponding to CD36 was detected in HuTu80-3C3-CD36-KO-11C/Re1 cells (Fig. 5A). Viral propagation, determined by an increase in HuSaV VP1 protein levels in the culture supernatant, in HuTu80-3C3-CD36-KO-11C/Re1 cells was observed for the 15 genotypes tested under bile acid supplemented conditions except for GI.6 and GII.3 (Fig. 5B). VP1 signal for GI.1 increased only with GlyCA, and for GII.2, the mean optical density (OD) of VP1 was GlyCA > TauCA. For GV.1, the mean OD value was TauCA > GlyCA (Fig. 5B), showing HuSaV propagation varied by bile acid type. Re-expression of human CD36 in knockout HuTu80 cells restored HuSaV propagation capacity in 13 HuSaV genotypes. tested with or without bile acid supplementation (Fig. S1C andD). Based on these results, they were used as representative HuSaV-insensitive non-human origin cells for human CD36 complementation.
CHO and Vero cells were transduced with the human CD36 isoform 1 gene and the production of CD36 was confirmed by western blotting using anti-FLAG and anti-human CD36 antibodies (Fig. 6A). In CHO cells transduced with the human CD36 gene (CD36-CHO), 10 of the 15 HuSaV genotypes (GI.3, GI.5, GI.6, GII.1, GII.2, GII.3, GII.5, GII.8, GV.1, GV.2) showed HuSaV propagation under bile acid-supplemented conditions, as demon strated by an increase in HuSaV VP1 protein in the culture supernatant. The VP1 OD value of HuSaV GI.3 was GlyCA > TauCA (Fig. 6B, upper panel).
In contrast, in Vero cells transduced with the human CD36 isoform 1 gene (CD36-Vero), only the six HuSaV GII strains (GII.1, GII.2, GII.3, GII.4, GII.5, and GII.8) showed viral propagation under bile acid-supplemented conditions, as demonstrated by an increase in HuSaV VP1 protein in the culture supernatant. HuSaV GII.4 propagated only with TauCA supplementation. VP1 OD value of HuSaV GII.3 and GII.5 was TauCA > GlyCA supplementation (Fig. 6B, lower panel). In western blot analysis using Vero cells, the production of the introduced human CD36 was detectable with the anti-FLAG antibody, but no clear signal was observed with the anti-CD36 antibody even after increasing the amounts of cell lysate (Fig. 6A). The difference in transduced CD36 expression between CHO and Vero cells may be partially related to the different puromycin-resistant concen trations (50 and 3.2 µg/mL, respectively). We hypothesize that the observed difference in HuSaV propagation ability between human CD36-introduced CHO and Vero cells may be partially related to the differences in human CD36 production levels between these two cell lines. Expression of human CD36 results in the acquisition of HuSaV propagation ability in 11 HuSaV genotypes to HuSaV-insensitive non-human origin cells.
## Full-Length Text
## Different effects on distinct genotypes of HuSaV propagation between human CD36 gene knockout and expression
The knockout of the human CD36 gene prevented the propagation of all 15 HuSaV genotypes in HuTu80-3C3-CD36-KO-11C cells (Fig. 4C). Furthermore, HuSaV propagation was confirmed under the conjugated bile acid supplementation condition in at least one of the following (i) human CD36 isoform 1 gene-introduced HuTu80-3C3-CD36 gene KO clone 11C cells (HuTu80-3C3-CD36-KO-11C/Re1) (Fig. 5B), (ii) CHO cells with the intro duced human isoform 1 CD36 gene (CD36-CHO), or (iii) Vero cells with the introduced human isoform 1 CD36 gene (CD36-Vero) (Fig. 6B).
Of note, the HuSaV GI.6 and GII.3 strains did not re-acquire viral propagation potential in HuTu80-3C3-CD36-11C-KO/Re1 (Fig. 5B). However, these two HuSaV genotype strains were able to propagate in CD36-CHO (Fig. 6B, upper panel). The HuSaV GII.4 strain, which did not propagate in CD36-CHO cells, was propagated in CD36-Vero cells and HuTu80-3C3-CD36-KO-11C/Re1 (Fig. 6B and5B).
Based on these results, we conclude that human CD36 is an essential cellular factor common to all 15 HuSaV genotypes that propagate in HuTu80 cells. Conjugated bile acids, GlyCA and/or TauCA, were necessary supplements for HuSaV propagation under all of the tested conditions.
## DISCUSSION
We previously reported HuSaV and human parechovirus propagation in HuTu80 cells (15,20) and identified MYADM as an essential factor for human parechovirus propagation (21). The MYADM knockout did not eliminate HuSaV propagation ability using 15 genotypes in HuTu80 cells (Fig. S1B) than in the HuTu80 cells (Fig. S1A). Based on these data, we searched for other cellular factors essential for HuSaV propagation by CRISPR screening, with HuTu80 3C3 clones showing clear CPE upon HuSaV propagation (Fig. 1B) and CD36 emerging as the top candidate genes for the four HuSaV genotypes (Fig. 2B). CD36 is expressed in various human tissues, including the gastrointestinal tract (https://www.proteinatlas.org/ENSG00000135218-CD36/tissue). CD36 RNA expression has been reported in HuTu80 cells (https://www.proteinatlas.org/ENSG00000135218-CD36/cell+line). We confirmed CD36 gene expression in HuTu80-3C3 cells using RT-PCR (Fig. S2).
In our previous studies, we used enzyme-linked immunosorbent assay (ELISA) and reverse transcription-quantitative PCR (RT-qPCR) as HuSaV propagation indicators (12,15) and demonstrated that the results of these methods were consistent. In this study, we used Ag-ELISA for all experiments because of its simplicity. The inoculated HuSaV had a low ELISA OD value (approximately 0.1 or lower). After virus inoculation, we maintained the culture for 10 days post-inoculation (dpi) without washing and measured VP1 levels in the supernatant to determine increases compared to the input levels. This protocol was previously used (12) and has advantages over RT-qPCR, which detects input virus RNA and requires washing steps that leave substantial signals, necessitating comparisons between 1 and 10 dpi (12,15).
After CRISPR screening with HuTu80-3C3, we introduced CD36 targeting guide RNA into parent HuTu80 and HuTu80-3C3 cells. In both lines, the four HuSaV genotype strains showed no increase in the VP1 signal under bile acid conditions, unlike HuTu80 and HuTu80-3C3 cells (Fig. 3). Loss of HuSaV propagation was further confirmed in four HuTu80-3C3-CD36-KO clones using HuSaV GII.2 inoculation (Fig. S3). Then, we confirmed that the 15 HuSaV genotypes did not propagate in HuTu80-3C3-CD36-KO-11C cells (Fig. 4C).
Lentiviral constructs and CRISPR machinery for MYADM and CD36 knockout were identical, except for the guide RNA sequence, and HuSaV propagation was lost only in CD36 KO cells. We concluded that CD36 was essential for HuSaV propagation in HuTu80 and HuTu80-3C3 cells.
Restoration of HuSaV propagation capacity was confirmed in 13 among 15 genotypes by the re-expression of human CD36 in HuTu80-CD36-KO-11C cells (Fig. 5B). Introduc ing the CD36 isoform 1 into four HuTu80-3C3-CD36-KO clones also enabled HuSaV propagation (Fig. S3). GII.2 strain was used because it had the highest VP1 signal in the HuTu80-3C3-CD36-KO-11C/Re1 among the four HuSaV genotypes used for CRISPR screening (Fig. 5B). Although the HuSaV GII.2 VP1 OD values in 3C3 were similar to those of GlyCA or TauCA (Fig. 1; Fig. S3), all HuTu80 3C3-CD36-KO-clone cell lines with CD36 isoform 1 expression showed greater HuSaV VP1 OD values for GlyCA than for TauCA (Fig. 5B; Fig. S3). This indicates that, unlike HuTu80-3C3 cells with endogenous CD36, the bile acid type in HuTu80-3C3-CD36-KO clone cells with CD36 isoform 1 expression affected HuSaV propagation.
To investigate whether CD36 alone is sufficient to confer HuSaV propagation ability, CHO and Vero cells were used as representative HuSaV-insensitive cells of non-human origin (Fig. S1C andD), which were used in our previous recombinant protein expression studies (22,23). Acquisition of HuSaV propagation capacity was confirmed in 11 of the 15 HuSaV genotype strains (Fig. 6B).
Our data support that human CD36 is a key cellular factor enabling HuSaV propaga tion in the tested cells (Fig. S3; Fig. 5B and6B).
Human CD36 (isoform 1), a membrane glycoprotein with two transmembrane domains of 472 amino acids, is expressed in various tissues and cells, involved in immunity, metabolism, and angiogenesis. Its ligands include thrombospondin, collagen, and malaria-infected erythrocytes, diverse ligands without structural similarity. It functions in recognition and phagocytosis of foreign substances by monocytes and macrophages, and in the uptake of long-chain fatty acids. The binding of CD36 to these ligands activates intracellular signaling pathways, including Src family kinases, MAP kinases, and PI3K/Akt pathways, altering cellular functions and responses (24)(25)(26)(27).
In addition to the 472 amino acid CD36 isoform 1, which is mainly expressed, five other isoforms lacking certain regions are present in the NCBI database (https:// www.ncbi.nlm.nih.gov/datasets/gene/id/948/products/) (27). We conducted overexpres sion experiments for CD36 isoform 2 (433 aa in length, ∆ 234-272 aa) and isoform 3 (412 aa in length, ∆ 144-203 aa) in HuTu80-3C3-KO-11C using the same lentiviral backbone for CD36 isoform 1. Using the GII.2 strain, HuTu80-3C3-CD36KO-11C cells expressing these two CD36 isoforms showed no increase in HuSaV VP1 signals with or without bile acids (Fig. S3). The CD36 isoform protein expression in HuTu80-3C3-CD36-KO-11C, and reduction in protein size following deglycosylation treatment were confirmed by western blotting (Fig. S4). These results show that the expression of CD36 genes other than isoform 1 via the lentiviral backbone does not restore HuSaV propagation ability.
The difference in results when using HuTu80-3C3-CD36-KO, CHO, and Vero cells for CD36 isoform 1 overexpression with distinct genotypes of HuSaVs with different bile acids, GlyCA, and TauCA (Fig. 5B and6B; Fig. S3) is puzzling and suggests a complex mechanism and pathway for HuSaVs propagation, which may be influenced by the expression of CD36 and activity of bile acids, as well as cell origin and cellular factors other than human CD36, depending on the genogroup or genotype strain.
CRISPR/Cas9 screening also identified other sgRNA sequences of genes other than human CD36 in the HuSaV-infected CPE-resistant cell population (Fig. 2B). The same guide RNA was detected in HuSaV GI.1 and 6 (n = 52) and GII.2 and 3 (n = 57) resistant cells, respectively, in the top 100 (Fig. 2A; Table S1). Specifically, activating transcription factor 2, keratin-associated protein 3-3, and transmembrane protein 60 ranked higher in GI.1-and GI.6-resistant cells, and dynein axonemal heavy chain 14, protocadherin 19, and caspase 14 ranked higher in GII.2-and GII.3-resistant cells in the top 20 (Fig. 2B). Therefore, evaluating the effects of these cellular factors on HuSaV propagation is warranted in the future.
Consistent with previous reports using enteroids or iPS-derived intestinal cells, fucosyltransferase 2, which is essential for the replication of various human norovirus strains and which had no effect on the replication of HuSaVs (17,18), was not a candidate in our screening. Furthermore, occludin, previously reported to confer sensitivity to porcine sapovirus in non-permissive CHO cells (28), was also not a candidate in our study.
The non-targeting control guide RNA was also ranked among the top three common genes by CRISPR screening, alongside CD36 and MYADM (Fig. 2). We confirmed the knockout effect of these specific protein-coding genes, especially CD36; however, we did not generate HuTu80 cells using the non-targeting control guide RNA. One possible reason for this is that we introduced the lentivirus at a higher multiplicity of infection than anticipated, introduced gene-specific guide RNA and non-targeting control guide RNA into the same cell; however, this has not been confirmed experimentally.
This study had some limitations. First, HuSaV propagation was verified throughout the study by increased VP1 protein levels via Ag-ELISA. Second, we did not resolve the mechanism by which CD36 is involved in HuSaV propagation. Third, the mechanism by which conjugated bile acids, GlyCA, and TauCA are involved in HuSaV propagation, and why they promoted different effects on the tested human CD36-expressed cells and HuSaV genotypes, was not determined.
To partially resolve the first limitation, we performed a median tissue culture infectious dose (TCID 50 ) assay with representative samples using a cell-based ELISA, as described in Materials and Methods. Virus titer ranges for HuTu80-3C3 with GlyCA were 4.0-5.5 log 10 TCID 50 /50 µL for HuSaV GI.1, GI.6, GII.2, GII.3, and 3.75-5.25 log 10 TCID 50 /50 µL for GI.1, GI.6, GII.2, GII.3 for TauCA, compared with ≤0.5 log 10 TCID 50 /50 µL for HuTu80-3C3 without bile acid supplementation and HuTu80-3C3-CD36-KO with/ without bile acid supplementation, for all four HuSaV genotypes (Fig. S5).
Virus titer ranges of GI.6, GII.2, and GII.3 in CD36-CHO with GlyCA and TauCA were 4.75-5.0; 4.0-4.5 log 10 TCID 50 /50 µL, compared with GI.1 ≤0.5 log 10 TCID 50 /50 µL (Fig. S5). CD36-Vero with GlyCA and TauCA showed higher average virus titers (3.0, 4.75) for GII.2 and were ≤0.5-1.5 log 10 TCID 50 /50 µL for GI.1, GI.6, and GII.3. Virus titers were ≤0.5 log 10 TCID 50 /50 µL in cell supernatants without bile acid supplementation (Fig. S5). Thus, increased VP1 protein levels measured by Ag-ELISA reflected infectious virus production in the tested samples (Fig. 3C, D, and6B; Fig. S5). SaV-virion production in non-suscepti ble cells might be variable, depending on the genotype or bile acid type.
Additional experiments to determine how these two bile acids impact the localization of CD36 and whether CD36 knockout impacts the binding/entry of HuSaVs to cells, as well as whether the CD36 knockout blockade of HuSaV propagation can be bypassed by transfecting HuSaV RNA into cells, might reveal the potential role of CD36 as an entry factor.
Nevertheless, our findings are an important first step toward understanding HuSaV propagation mechanisms and the establishment of methods to control this highly contagious virus by developing anti-HuSaV and HuSaV-susceptible animal models in the future.
## MATERIALS AND METHODS
## Cell lines
Three human cell lines were used for HuSaV culture, including (i) a human duodenum carcinoma-derived cell line (HuTu80 cells; ATCC #HTB-40) and their clone or geneti cally modified cells; (ii) a Chinese hamster ovary-derived cell line (CHO-K1 cells; JCRB IFO50414); and (iii) a monkey kidney-derived cell line (Vero cells; JCRB0111).
## Cell culture conditions
HuTu80 cells that had undergone a high number of passages (>150) and selected as a clone were designated HuTu80-3C3 and grown in Iscove's Modified Dulbecco's medium (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 3% heat-inactivated fetal bovine serum (FBS) (Biosera, Kansas City, MO, USA), and antibiotics (100 U/mL penicillin and 100 µg/mL streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) as previously described (12). Human CD36 gene KO HuTu80-3C3, or their clones, were grown under the above conditions supplemented with puromy cin (3.1 µg/mL) (Invivogen, San Diego, CA, USA), and CD36 gene KO HuTu80-3C3 clones with the re-expression of the human CD36 gene were grown under the above conditions supplemented with blasticidin (25 µg/mL) (FUJIFILM Wako Pure Chemical Corporation). CHO-K1 cells were grown in Ham's F12 medium (FUJIFILM Wako Pure Chemical Corporation) supplemented with 5% FBS and antibiotics. Vero cells were grown in Eagle's minimal essential medium (FUJIFILM Wako Pure Chemical Corporation), supplemented with 5% FBS and antibiotics. mCherry or human CD36-expressing CHO and Vero cells were grown under the above conditions supplemented with puromycin (50 and 3.1 µg/mL, respectively).
## Viruses
HuSaV passage 1 stock (cell culture supernatant) of one of the 15 genotypes (GI.1-AK20, GI.2-FS25, GI.3-D1736, GI.4-Chiba000496, GI.5-D1729, GI.6-Chiba000764, GI.7-D1714, GII.1-Kuma129, GII.2-Kuma130, GII.3-AK11, GII.4-D1739, GII.5-Kashiwa1, GII.8-AK764, GV.1-D3302, and GV.2-NGY1) (12,14,15) were used for viral propagation studies.
## Bile acids
GlyCA or TauCA, both purchased from Nacalai Tesque (Kyoto, Japan), were prepared as a 100 mM stock solution in 20% ethanol, filtered through a 0.22 µm filter, and added as supplements to the HuSaV culture medium as described previously (12).
## CRISPR/Cas9 screening to identify candidate genes
HuTu80-3C3 cells cultured in 6-well plates were infected with commercial ready-touse all-in-one CRISPR/Cas9 based genome-wide human gene-knockout pooled library lentiviruses (Addgene, Watertown, MA, USA, Catalog #73179) based on the Human Brunello CRISPR knockout pooled library (29), which targets 19,114 human genes, with four different target sequences per gene and 1,000 non-targeting control gRNA sequence that do not recognize any sequence in the human genome with a calculated multiplicity of infection of ~1, using TransDux Max Lentivirus Transduction Reagents (System Biosciences, LLC, Palo Alto, CA, USA) at a culture volume of 1.3 mL per well. One milliliter of culture medium was added the following day, and 4 days later, the medium was replaced with 2 mL of medium supplemented with 2 mM TauCA. Then, the lentivirus-transduced HuTu80-3C3 cells were further infected with HuSaV GI.1-AK20, GI.6-Chiba000764, GII.2-Kuma130, or GII.3-AK11 (1 × 10 6 viral genomic copies/well) and maintained for 10 days. Six plates (a total of 36 wells each) were used for each genotype.
To select HuSaV propagation-resistant HuTu80-3C3 cells, the cells in each well were trypsinized and then inoculated with HuSaV in medium supplemented with TauCA with the same HuSaV genotype strain combination (1 × 10 7-8 viral genomic copies/well, higher than the initial inoculation) and maintained for 7-10 days. These steps were repeated four times. Genomic DNA was extracted from the surviving cells in each well using the KANEKA Easy DNA Extraction Kit version 2 (KANEKA Corporation, Tokyo, Japan, KN-T110005). Using this genomic DNA, a 160 bp DNA fragment corresponding to the gRNA regions in surviving cells was amplified by PCR using the forward primer (P5) 5′-T TGTGGAAAGGACGAAACACCG-3′ and reverse primer (P7) 5′-CCAATTCCCACTCCTTTCAAG ACCT-3′ according to the Addgene protocol using MightyAmp DNA polymerase Ver. 3 (Takara Bio Inc., Shiga, Japan, R076A) under the following conditions: 98°C for 2 min, followed by 40 cycles of a three-step PCR (98°C for 10 s, 60°C for 15 s, and 68°C for 30 s) as previously described (21). PCR products of 160 bp were extracted and pooled along with the infected HuSaV genotypes. The library preparation of HuSaV-resistant cells was performed using the Illumina TruSeq Nano DNA Sample Prep Kit (Illumina, San Diego, CA, USA). NGS sequencing using a NovaSeq 6000 sequencer (Illumina), and data processing, including read count guide RNA sequencing corresponding to the human genes list provided by Addgene, was performed at Macrogen Incorporated (Tokyo, Japan).
The top 100 guide RNA reads of each of the four HuSaV genotypes were analyzed and visualized common candidate genes using Intervene's Upset plot (https://asntech.shi nyapps.io/intervene/) (30,31).
## Generation of candidate gene knockout HuTu80 cells
Confluent HuTu80 or HuTu80-3C3 cells cultured in 12-well plates were infected with Ready-to-use Human MYADM or CD36 sgRNA CRISPR All-in-One Lentivirus (Human) (Cat No. K1370716 or 155341110603, Applied Biological Materials Inc., Richmond, Canada), using TransDux Max Lentivirus Transduction Reagents (System Biosciences, LLC) with a volume of 1.3 mL per well. One milliliter of culture medium was added the following day, and 4 days later, the cells were trypsinized, spread onto 24-well plates with medium supplemented with serial dilutions of puromycin, and resistant cells with the highest puromycin concentration (3.2 µg/mL) were expanded.
## Confirmation of target gene knockout by genomic sequencing
HuTu80-3C3 CD36 KO cells were subjected to clonal selection to isolate single-cell clones following CRISPR/Cas9-mediated genome editing. Genomic DNA was extracted from the cells using the NucleoSpin Tissue Kit (Takara Bio Inc.) according to the manufacturer's instructions. The targeted genomic region of human CD36 was amplified by PCR using locus-specific primers (forward: 5′-A CCA GAG CTT GTA GAA ACC ACT-3′; reverse: 5′-A CAT GCA TAC CTG TAG ACA GCT-3′). Similarly, the targeted genomic region of human MYADM was amplified by PCR using locus-specific primers (forward: 5′-A AGA AAA GAA AAC CGA AAG CCC-3′; reverse: 5′-G TAA GCC AC A CAC GCG ATG-3′) from the HuTu80-MYADM KO cells. The PCR products were subsequently purified using the NucleoSpin Gel and PCR Clean-up Kit (Takara Bio Inc.) and analyzed by Sanger sequencing.
## Confirmation of gene expressions by RT-PCR
Confluent HuTu80-3C3 cells or HuTu80-3C3-CD36-KO-11C cells cultured in 12-well plates were treated with 1 mL of medium supplemented with 1 mM GlyCA, 2 mM TauCA, or without bile acids for 24 h. RNA was extracted from the pooled cells from the three wells for each cell and treatment using High-Pure RNA Isolation Kits (Roche Diagnostics, Mannheim, Germany) following the manufacturer's instructions. cDNA was synthesized as follows: 10 µL of DNase-treated RNA samples were mixed with a 10 µL mixture, including 20 pmol oligo dT 30 primer, 1 mM dNTPs, 20 units of recombinant RNase inhibitor (Takara Bio Inc.), and 60 units of ReverTra Ace (Toyobo, Osaka, Japan). Reactions were performed at 30°C for 10 min, 42°C for 30 min, and 95°C for 5 min. PCR was performed using 2 µL of cDNA with primers targeting the partial human CD36 gene, forward primer (Human CD36 F1:5′-CTGTGTTTGGAGGTATTCTAATGCCAG-3′), reverse primer (Human CD36 R1: 5′-CCTGTGGATTTTGCACATCAAAGATCCA-3′), or primer set targeting the partial glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene, forward primer (:5′-CAATGACCCCTTCATTGACC-3′), reverse primer (5′-TTGATTTTGGAGG GATCTCG-3′), using 10 µL of 2G Fast HotStart ReadyMix with dye (KAPA Biosystems, Wilmington, MA, USA) under the following conditions: 95°C for 5 min, followed by 50 cycles of 95°C for 20 s, 58°C for 20 s, and 72°C for 5 s, and a final extension at 72°C for 1 min. Each PCR reaction mixture of 10 µL was directly applied and analyzed by 20 mM Tris-acetate-0.5 mM EDTA buffer-based 2% agarose gel electrophoresis in the presence of FastGene Midori Green Direct (Nippon Genetics), and the gel image was captured with a Gel Doc Go Imaging System (BioRad, Tokyo, Japan).
## Complemented CD36 gene expression in cloned knockout HuTu80-3C3 cells
Confluent HuTu80-3C3-CD36 KO (HuTu80-3C3-CD36-KO) clone 1C, 1E, 10E, or 11C cells cultured in 12-well plates were infected with ready-to-use custom ordered lentivirus, pLV[Exp]-CMV > hCD36-mut(ns)/Myc/FLAG:P2A:Bsd, vector ID VB220531-1120gfv (Vector Builder, Chicago, IL, USA), which carry a CRISPR/Cas9 Guide RNA resistant human CD36 (isoform 1 encoding 472 amino acids (aa) in length: GenBank accession number NM_000072.3) with C-terminus Myc (EQKLISEEDL) and FLAG-tag (DYKDDDDK) coding sequences. To achieve this, the CRISPR Cas9/guide RNA target sequence of human CD36 (NM_001001547) was modified (67-GGtGGaATctTgATGCCtGTcGG-89) (synonymous substitutions are shown in lower case). Then, the cells were treated as described in the section above, selected with blasticidin (25 µg/mL), and designated as human CD36 isoform 1 gene-introduced HuTu80-3C3-CD36 gene KO clone 11C cells (HuTu80-3C3-CD36-KO-1C, -1E, 10E, or 11C/Re1).
Additional experiments expressing CD36 isoforms 2 and 3 (encoding 433 aa [GenBank accession number NM_001289908.1], and 412 aa [NM_001289909.1] in length, respectively) (27) were performed using the ready-to-use custom ordered lentivirus, vector ID VB220821-1074yvq, -1075csd (Vector Builder), with the same vector, includ ing tag, as well as selection marker to CD36 isoform 1 as described above using
## References
1. "and -Re3, respectively. Expression of the human CD36 gene in non-susceptible cells Ready-to-use lentivirus expression of mCherry pLV[Exp]-Puro-EF1A > mCherry, vector ID VB010000-9497fvv (Vector Builder), or human CD36 isoform 1 with C-terminus Myc and FLAG-tag coding sequences pLV[Exp]-CMV > hCD36"
2. Oka, Wang, Katayama et al. (2015) "Comprehensive review of human sapoviruses" *Clin Microbiol Rev*
3. Kauppinen, Pitkänen, Al-Hello et al. (2019) "Two drinking water outbreaks caused by wastewater intrusion including sapovirus in Finland" *Int J Environ Res Public Health*
4. Hergens, Öhd, Alm et al. (2016) "Investigation of a food-borne outbreak of gastroenteritis in a school canteen revealed a variant of sapovirus genogroup V not detected by standard PCR" *Euro Surveill*
5. Shibata, Sekizuka, Kodaira et al. (2015) "Complete genome sequence of a novel GV.2 sapovirus strain, NGY-1, detected from a suspected foodborne gastroenteritis outbreak" *Genome Announc*
6. Oka, Doan, Haga et al. (2017) "Genetic characterization of rare genotype GII.5 sapovirus strain detected from a suspected food-borne gastroenteritis outbreak among adults in Japan in 2010" *Jpn J Infect Dis*
7. Zou, Li, Zhou et al. (2021) "A large acute gastroenteritis outbreak associated with both Campylobacter coli and human sapovirus -Beijing Municipality" *China CDC Wkly*
8. Kobayashi, Fujiwara, Yasui et al. (2012) "A foodborne outbreak of sapovirus linked to catered box lunches in Japan" *Arch Virol*
9. Wang, Zhang, Shen (2015) "The impact of calicivirus mixed infection in an oyster-associated outbreak during a food festival" *J Clin Virol*
10. Shirai, Motooka, Ushikai et al. (2025) "Molecular epidemiology of human sapovirus based on the surveillance of wastewater and patients with acute gastroenteritis in Osaka" *Japan. Sci Total Environ*
11. Diez-Valcarce, Castro, Marine et al. (2018) "Genetic diversity of human sapovirus across the Americas" *J Clin Virol*
12. Miyazaki, Song, Oka et al. (2022) "Atomic structure of the human sapovirus capsid reveals a unique capsid protein conformation in caliciviruses" *J Virol*
13. Oka, Li, Yonemitsu et al. (2024) "Propagating and banking genetically diverse human sapovirus strains using a human duodenal cell line: investigating antigenic differences between strains" *J Virol*
14. Yokoyama, Doan, Motomura et al. (2024) "Strong evolutionary constraints against amino acid changes in the P2 subdomain of sapovirus GI.1 capsid protein VP1" *Biochem Biophys Res Commun*
15. Li, Kataoka, Doan et al. (2022) "Characterization of a human sapovirus genotype GII.3 strain generated by a reverse genetics system: VP2 is a minor structural protein of the virion" *Viruses*
16. Takagi, Oka, Shimoike et al. (2020) "Human sapovirus propaga tion in human cell lines supplemented with bile acids" *Proc Natl Acad Sci*
17. Fukuda, Ishikawa, Ishiyama et al. (2025) "Establishment of a novel caco-2based cell culture system for human sapovirus propagation" *Genes Cells*
18. Euller-Nicolas, Mennec, Schaeffer et al. (2023) "Human sapovirus replication in human intestinal enteroids" *J Virol*
19. Matsumoto, Kurokawa, Tamiya et al. (2023) "Replication of human sapovirus in human-induced pluripotent stem cell-derived intestinal epithelial cells" *Viruses*
20. Armesilla, Vega (1994) "Structural organization of the gene for human CD36 glycoprotein" *J Biol Chem*
21. Takagi, Oka, Ami et al. (2022) "A human intestinal cell line suitable for the propagation of human parechovirus type 1 to 6 with a clear cytopathic effect" *Jpn J Infect Dis*
22. Watanabe, Oka, Takagi et al. (2023) "Myeloidassociated differentiation marker is an essential host factor for human parechovirus PeV-A3 entry" *Nat Commun*
23. Jauhiainen, Huuskonen, Baumann et al. (1999) "Phospholipid transfer protein (PLTP) causes proteolytic cleavage of apolipoprotein A-I" *J Lipid Res*
24. Someya, Okemoto-Nakamura, Kurata et al. (2023) "Establishment of measles virus receptor-expressing Vero cells lacking functional poliovirus receptors" *Microbiol Immunol*
25. Silverstein, Febbraio (2009) "CD36, a scavenger receptor involved in immunity, metabolism, angiogenesis, and behavior" *Sci Signal*
26. Bachmann, Metwally, Allweier et al. (2022) "CD36-a host receptor necessary for malaria parasites to establish and maintain infection" *Microorganisms*
27. Chen, Zhang, Cui et al. (2022) "CD36, a signaling receptor and fatty acid transporter that regulates immune cell metabolism and fate" *J Exp Med*
28. Rac (2025) "Human CD36: gene regulation, protein function, and its role in atherosclerosis pathogenesis" *Genes (Basel)*
29. Alfajaro, Cho, Kim et al. (2019) "Early porcine sapovirus infection disrupts tight junctions and uses occludin as a coreceptor" *J Virol*
30. Doench, Fusi, Sullender et al. (2016) "Optimized sgRNA design to maximize activity and minimize offtarget effects of CRISPR-Cas9" *Nat Biotechnol*
31. Lex, Gehlenborg, Strobelt et al. (2014) "UpSet: visualization of Intersecting Sets" *IEEE Trans Vis Comput Graph*
32. Khan, Mathelier (2017) "Intervene: a tool for intersection and visualization of multiple gene or genomic region sets" *BMC Bioinformat ics*
33. Okemoto-Nakamura, Someya, Yamaji et al. (2021) "Poliovirus-nonsusceptible vero cell line for the World Health Organization global action plan" *Sci Rep*
34. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12557038&blobtype=pdf | # presentations highlight how surveillance is being reshaped by the availability of new data, computational capacity, and methodological innovation. A final roundtable, moderated by Professors Harvala and Ganczak, renowned experts in public health surveillance and policy, will invite discussion with the audience to reflect on priorities, ethical challenges, and pathways to interoperability. By connecting historical insight with contemporary evidence, this session aims to reframe infectious disease surveillance as a strategic, anticipatory, and adaptive public health function. Key messages: • Surveillance must integrate clinical, environmental, behavioural and digital data to face 21st-century infectious disease threats. • Coordinated, adaptive and ethically grounded surveillance is key to strengthening public health preparedness and response
Viviani Luca, Barbati, L Viviani, L Gentile, L Prisciano, L Sacchi, R Bellazzi, F Baldanti, R Bruno, A Muzzi, A Odone
## Abstract
citation ID: ckaf161.740 From smallpox to dengue: 200 years of integrated surveillance duced in parallel with the first mass vaccination campaigns. Archival sources from the city of Pavia, Northern Italy, document systematic case reporting for smallpox, measles, rubella, and epidemic typhus, revealing how public health authorities began to use data not only to control contagion but also to assess the effectiveness of anti-smallpox immunisation efforts. These early records offer compelling insights into the rationale and structure of disease surveillance during a period of major biomedical and social transition. Notably, notifications were not limited to human disease. Veterinary registers from the same archives report episodes of animal mortality, including mass deaths among ducks in the 1820s, which suggests an early awareness of zoonotic threats and possibly the circulation of avian diseases. This proto-One Health approach underscores how disease monitoring was already conceived across species boundaries. Fastforward to the present, the same region has recently reported one of the first locally transmitted dengue outbreaks in continental Europe, with vector presence and climate conditions favouring endemic transmission. This geographical continuity reinforces the need to The tems. |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645978&blobtype=pdf | # Mammalian innate antiviral defenses: beyond interferon
Emily Rex, Joy Shaffer, Daniel Deng, Don Gammon
## Abstract
Mammalian cells employ a wide array of antiviral defense mechanisms to restrict viral replication at virtually all steps of the viral life cycle. Notably, the interferon (IFN) response has been shown to play a central role in restricting the replication of disparate viral pathogens in mammals. Consequently, since its discovery in 1957, the IFN response has dominated antiviral immunity research, leaving IFN-independent pathways relatively understudied. Exploring these alternative host defenses is crucial for under standing the complete arsenal that mammalian hosts deploy to combat viral disease, as IFN responses undoubtedly work in concert with other antiviral defenses to achieve virus restriction. Here, we discuss selected examples of antiviral factors and pathways in mammals that are not classically associated with the IFN response. These defenses range from constitutively expressed host restriction factors that directly inhibit specific steps of the viral life cycle to signaling pathways that invoke IFN-independent antiviral gene expression programs to cell death mechanisms that sacrifice the infected cell to prevent viral spread. Ultimately, our goal is to highlight the diversity of IFN-independent antiviral defenses that mammalian hosts utilize to block viral infection. KEYWORDS interferon-independent antiviral immunity, innate immunity, TRIM E3 ubiquitin ligases, FEAR pathway, autophagy, septins, programmed cell death, viral immune evasion I nterferons (IFNs) are classified into three groups based on the different receptors they bind and have been reviewed in detail elsewhere (1-4). Here, we will focus on discussing the type I IFN (IFN-I) response as it has long been regarded as the most vital antiviral defense system in mammals, where it induces a plethora of IFN-stimula ted genes (ISGs) that encode factors which restrict virus replication through distinct mechanisms (1,5,6). The importance of the IFN-I response is underscored by the strong association of animals and humans with deficiencies in IFN-I responses with increased susceptibility to viral infection (6-9). Moreover, virtually all mammalian viruses encode antagonists of the IFN-I response, further highlighting the central role of the IFN-I response to the evolutionary battle between viral pathogens and their mammalian hosts (10).Accumulating evidence suggests that mammals also employ a diverse set of IFN-I-independent antiviral defenses that likely operate alongside IFN-I responses. Many of these alternative defenses are ancient innate responses that predate the rise of IFNs in jawed vertebrates during eukaryotic evolution but have been retained in mamma lian hosts to combat infection (11). We define IFN-I-independent antiviral immunity as defenses that do not rely on canonical IFN or ISG production or those that induce ISGs but through non-canonical signaling pathways (1, 2). Such responses are likely critical to the overall control of viral infections, particularly when viruses deploy effective IFN-I antagonism mechanisms (2, 12). Moreover, IFN-I-independent responses can function in the absence of parallel IFN-I responses, which can be especially critical in stem cells where IFN-I signaling is incongruent with the maintenance of stem cell pluripotency (13). IFN-I-independent defenses include both constitutive and inducible cell-intrinsic
restriction mechanisms that attempt to block viral replication at various stages of the viral life cycle or that activate programmed cell death to sacrifice infected cells to prevent further viral spread. In vivo, IFN-I-independent responses can also promote the recruit ment of immune cells to the sites of infection and thus can function to activate other arms of the immune system (14). Understanding IFN-I-independent antiviral responses is not only critical for ascertaining a more complete view of mammalian innate immunity but also may offer novel therapeutic strategies for the treatment of viral diseases, which is important given the adverse side effects of IFN therapy (15,16).
Here, we discuss selected examples of IFN-I-independent antiviral defenses in mammals, focusing on the molecular mechanisms by which they restrict viral replication, the cellular and viral contexts in which these defenses operate, and their contribution to the outcome of viral infection in vivo. While not the focus of this article, we also provide examples of viral evasion strategies to overcome the discussed IFN-I-independent responses to illustrate their physiological relevance to the evolutionary struggle between viral pathogens and their mammalian hosts. Importantly, due to space limitations, many additional IFN-I-independent antiviral defenses that also combat viral disease (see Table 1 for additional examples) are unable to be discussed here in detail. Thus, the examples discussed below are meant to illustrate the diversity of IFN-I-independent defenses employed by mammals rather than provide a comprehensive overview of all known IFN-I-independent antiviral mechanisms.
## EXAMPLES OF IFN-I-INDEPENDENT ANTIVIRAL DEFENSES IN MAMMALS ISG-mediated antiviral defense independent of canonical IFN-I signaling
ISGs are classically thought of as being induced by canonical IFN-I signaling involv ing IFN-α/β binding to the IFN-α/β receptor (IFNAR), activation of intracellular kina ses (JAK1, JAK2, and TYK2) that phosphorylate STAT1/STAT2 transcription factors, and the translocation of STAT1/STAT2/IRF9 (ISGF3) complexes into the nucleus to induce transcription of ISGs containing IFN-stimulated response elements (ISREs) (1,37). However, there is substantial evidence that basal levels of ISGs or induction of ISGs through mechanisms independent of JAK-STAT signaling can provide initial viral containment, and such early ISG responses may be less susceptible to viral evasion strategies (see more comprehensive reviews on this topic reviewed in references 2, 38).
One mechanism for IFN-I-independent ISG induction involves recognition of pathogen-associated molecular patterns (PAMPs) by host-encoded pattern recognition receptors (PRRs) such as RIG-I, MDA5, and cGAS, which leads to the activation of IFN regulatory factors IRF3 and IRF1 (39,40). These transcription factors can then bind to ISREs in the promoters of ISGs and initiate their expression without requiring IFN secretion or STAT activation (2). For example, cGAS can activate a STING-IRF3 signal ing axis to induce ISGs in STAT1 -/-fibroblasts (39,41), and human cytomegalovirus (HCMV; Herpesviridae) infection can induce robust, IRF-3-dependent ISG expression when JAK-STAT signaling is abrogated by viral STAT1 antagonists (2,41). Interestingly, MAVS, which is known to associate with mitochondria to mediate canonical, IRF-3-dependent IFN-I responses downstream of RIG-I and MDA5, can also associate with peroxisomes to activate IRF-1-dependent ISG induction that does not require JAK-STAT signaling (42). Additionally, experiments using viral RNA mimetics or UV-inactivated viruses demon strate that cells can mount rapid ISG responses via IRF1 or IRF3, independently of IFN cytokine production or JAK-STAT signaling (43)(44)(45). Moreover, MDA5, a sensor of viral dsRNA, was implicated in restricting coxsackievirus B3 (CVB3; Picornaviridae) replication in murine fibroblasts independently of IFNAR signaling (46). These examples illustrate how upstream components of IFN-I signaling pathways (e.g., PRRs and IRF3/IRF1) can play critical roles in mounting ISG induction without the requirement for JAK-STAT signaling.
Additionally, non-canonical cytokine pathways can also induce ISGs independently of IFN-α/β binding to IFNAR. The cytokines TNF-α, IL-1, and IL-27 can drive ISG expression through the action of various transcription factors, including NF-κB, STAT1, and IRF-3 in the absence of canonical IFN-I signaling (47)(48)(49)(50). The physiological relevance of NF-κBmediated antiviral responses to controlling viral infection is well illustrated by the finding that vaccinia virus (VACV; Poxviridae) encodes at least 18 independent inhibitors of NF-κB signaling (51). Importantly, ISGs induced through these IFN-independent mechanisms can play critical, early roles in restricting viral replication. For example, in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (Coronaviridae)-infected hamsters, lung epithelial cells significantly upregulate ISGs (e.g., MX1, IFIT2, and IFIT3) prior to detecta ble IFN production, and this corresponds with early containment of viral replication (52).
In summary, IFN-I-independent induction of ISGs via IRF3/IRF1 activation and alternative cytokine pathways can be crucial in mounting early innate antiviral responses. These mechanisms enable cells to rapidly deploy antiviral defenses that can be sufficient to control viral infection under low multiplicity of infection conditions (53). Moreover, these alternative ISG induction mechanisms both bide time for the host to robustly induce ISGs through canonical IFN-I signaling and provide new routes for ISG induction that viral factors may not fully antagonize.
## Direct ubiquitination of viral proteins by tripartite motif-containing 7
Tripartite motif-containing 7 (TRIM7) (also known as GNIP or RNF90) is a member of the TRIM family of E3 ubiquitin ligases that has received considerable attention in recent years due to emerging evidence suggesting it plays diverse, context-dependent antiviral and proviral roles during viral infection. Here, we focus on discussing the direct antiviral role of TRIM7 in mediating viral protein degradation; however, TRIM7 has also been implicated in both negative and positive regulation of innate immune responses, including IFN-I and NF-κB pathways (reviewed in references 54,55).
Early evidence for an antiviral role of TRIM7 emerged from CRISPR activation screening that identified TRIM7 as a restriction factor for murine norovirus (MNoV; Caliciviridae) in human and murine cells (56). TRIM7 proteins encode a conserved N-terminal "RBCC" motif (composed of RING, B-box, and coiled-coil domains) and a C-terminal PRYSPRY domain that confers substrate specificity (55). Importantly, TRIM7 is not induced by IFN-I signaling and thus is not an ISG (57). The TRIM7 gene enco des four isoforms, with isoform 1 (the canonical and longest isoform that encodes a C-terminal PRYSPRY domain) and isoform 4 (a shorter isoform lacking the PRYSPRY domain) being the most well-studied isoforms and the only isoforms that encode the RING domain required for E3 ubiquitin ligase activity (56). Interestingly, overexpression of TRIM7 isoform 1, but not isoform 4, restricted MNoV replication in murine and human cells (56), suggesting an important role for the PRYSPRY domain. Transfection of MNoV genomic RNA into TRIM7-overexpressing cells was unable to overcome the block to MNoV replication, suggesting that TRIM7 inhibited MNoV replication post-entry (56). Shortly after this report, an independent study screened a library of 118 RING-type E3 ubiquitin ligases for antiviral activity and identified TRIM7 as a potent cell-intrinsic restriction factor for a wide array of human enteroviruses, such as CVB3 (58). However, TRIM7 did not affect flaviviruses, alphaviruses, or paramyxoviruses (58). These observa tions indicated that TRIM7 has selective antiviral activity against members of Caliciviridae and Picornaviridae. Subsequent structural and biochemical studies revealed that TRIM7 utilizes its PRYSPRY domain to specifically recognize proteins terminating in C-terminal glutamine residues as substrates (59)(60)(61). Importantly, C-terminal glutamine-terminated proteins are often generated during the replication of several positive-sense ssRNA viruses, such as caliciviruses and enteroviruses, that use viral 3C-like proteases to cleave their polyproteins to generate viral proteins ending in glutamine (54,55,59). Once bound, TRIM7 catalyzes K48-linked polyubiquitination via its RING domain, leading to the proteasomal degradation of targeted viral components (54,55). For example, TRIM7 targets the CVB3 2BC protein, a membrane remodeling factor, for ubiquitination and degradation in the proteasome (58). Interestingly, TRIM7 could bind to both viral 2C and its precursor 2BC, but TRIM7 only mediated degradation of the latter, revealing an unappreciated vulnerability of viruses to host restriction via targeting of their precursor proteins (58). Two recent studies have demonstrated that MNoV NS6 proteins, which terminate at a glutamine after viral protease cleavage, were targeted by TRIM7 (59,62). Interestingly, in contrast to the finding with enteroviruses, TRIM7 could only bind to NS6, but not the NS6-NS7 precursor protein (62).
Given the potent antiviral activity of TRIM7, several viruses have evolved counter measures to evade TRIM7-mediated restriction. For example, the 3C proteases from CVB3 and poliovirus (Picornaviridae) cleave TRIM7 at a conserved glutamine site (Q24), which dampens its E3 ubiquitin ligase and antiviral activities (63). However, these viral evasion strategies are clearly insufficient to completely inactivate TRIM7 function because serial passage of CVB3 in TRIM7-overexpressing cells gave rise to variants encoding a T323A point mutation in 2C that resulted in further TRIM7 resistance (58). Notably, these CVB3 variants exhibited enhanced replication and virulence in mice (58), suggesting TRIM7 evasion has important consequences for enterovirus pathogenesis. Similar passaging studies using TRIM7-overexpressing cells with MNoV identified a TRIM7-resistant variant with a single F182C substitution that reduced cleavage of the NS6-NS7 precursor protein (62). This suggests that this mutant evades TRIM7 restriction by reducing the production of free NS6 that TRIM7 can target (62). However, in contrast to the CVB3 escape mutant, the F182C-encoding MNoV strain displayed highly attenuated replica tion and pathogenesis in mice (62). Other recent MNoV passaging studies selected for variants that overcome a post-entry restriction in human cells, and these mutants also displayed attenuated virulence in mice (64). These examples illustrate that there can be consequential evolutionary tradeoffs for overcoming host restrictions and highlight the importance of examining virus-host interactions in relevant in vivo models. This is further underscored by a recent study with TRIM7 -/-mice that has called into question whether TRIM7 plays a substantial role in antiviral defense against MNoV. Using two independently derived Trim7-deficient mouse strains, these authors found no differences in viral burden or tissue distribution of MNoV during either acute or persistent infections (65).
A recent study suggests that TRIM7 may also target SARS-CoV-2 proteins but through a non-degradative mechanism (66). TRIM7 ubiquitinated the SARS-CoV-2 membrane (M) protein at lysine-14 (K14), which suppressed M protein-mediated cell death induction that would typically promote viral spread (66). TRIM7 -/-mice exhibited greater weight loss and higher SARS-CoV-2 titers, suggesting an overall antiviral function for TRIM7 in mice (66). However, others have not found TRIM7 overexpression to affect SARS-CoV-2 replication in human HeLa cells (59); therefore, the physiological role of TRIM7 in SARS-CoV-2 infection remains unclear.
Interestingly, when TRIM7 engages with viral proteins, it does not always act as an antiviral factor. For example, during Zika virus (ZIKV; Flaviviridae) infection, TRIM7 was found to catalyze the K63-linked ubiquitination of the viral envelope protein, which enhanced virion infectivity and viral dissemination (67). Consequently, ZIKV replicated less efficiently in the brain and reproductive tissues of TRIM7 -/-mice (67). Thus, as with other TRIM family members (68), TRIM7 can play very different roles during viral infection, depending on the context.
## The FACT-ETS1-antiviral response pathway
The FACT-ETS1-antiviral response (FEAR) pathway is a newly discovered innate antiviral response that requires the "facilitates chromatin transcription" (FACT) complex, a histone chaperone that is highly conserved among invertebrate and vertebrate eukaryotes (69). The human FACT complex is a heterodimer composed of human suppressor of Ty 16 homolog (hSpt16) and structure-specific recognition protein-1 (SSRP1) subunits (70,71). FACT regulates cellular gene transcription by remodeling histones in chromatin that would otherwise impede transcription complexes (72). The antiviral role of FACT was initially identified through studies with VACV. Upon VACV infection, early viral gene expression triggers a SUMOylated form of hSpt16 to translocate from the cytoplasm into the nucleus, where it forms specialized, antiviral FACT complexes with SSRP1. These antiviral FACT complexes subsequently bind chromatin to induce the expression of the transcription factor E26 transformation-specific sequence-1 (ETS-1), which is thought to invoke antiviral gene expression programs (73, 74) (Fig. 1). Though first identified to restrict VACV, a DNA virus, growing evidence suggests the FEAR pathway has a broader antiviral role against disparate RNA viruses, including vesicular stomatitis virus (VSV; Rhabdoviridae), influenza A virus (IAV; Orthomyxoviridae), yellow fever virus (Flaviviridae), and the Paramyxoviridae members Sendai virus (SeV) and human parainfluenza virus type 1 (HPIV-1) (73,74).
ETS-1 is the founding member of the ETS transcription factor family that shares a common DNA-binding domain termed the "ETS domain" (75). Roles for ETS-1 in immune cell development have been reported previously (75)(76)(77), but ETS-1 had not been implicated in virus restriction. However, ELF1, another ETS transcription factor family member, was previously shown to induce an antiviral gene expression program that impedes DNA and RNA virus replication independently of the IFN-I response (78). Consistently, the FEAR pathway was activated in cells lacking key IFN-I signaling components, such as IRF3, IFNAR, or STAT1 (73,74). Moreover, ISG induction was normal in ETS-1-deficient cells, suggesting that the FEAR pathway does not require or regulate the IFN-I response (74). Given that ETS transcription factors arose ~600 million years ago during invertebrate metazoan evolution, these transcription factors may activate ancient antiviral gene expression programs that predate IFN-I responses. This is consistent with the finding that the transcriptional programs induced by ELF-1 and the FEAR pathway were largely distinct from IFN-I transcriptional signatures (74,78). However, key questions still remain regarding the identity of the host sensor(s) involved in activating the FEAR pathway and the ETS-1-induced factors that restrict viral replication (79).
The importance of the FEAR pathway to antiviral defense in mammals was further underscored by the identification of viral antagonists of this pathway. Poxvirus-enco ded A51R proteins were the first FEAR pathway inhibitors described. A51R proteins outcompete SSRP1 to directly bind to cytoplasmic SUMOylated hSpt16 subunits and, using C-terminal microtubule-binding domains (34,74), tether SUMOylated hSpt16 to microtubules to prevent FACT-dependent ETS-1 expression in the nucleus (74) (Fig. 1). Unlike wild-type VACV, strains lacking A51R strongly induced ETS-1 expression during infection and displayed attenuated replication in cell culture that could be rescued by hSpt16 RNA interference (RNAi)-mediated depletion (74,80). Electron microscopy studies of VACV A51R knockout virus-infected cells suggested that the FEAR pathway restricts VACV at the stage of virion morphogenesis, as these mutant viruses produced reduced numbers of mature VACV particles, and this phenotype could be reversed by depletion of hSpt16 levels (74). Furthermore, mice inoculated with VACV strains lacking A51R or encoding A51R mutants that cannot bind SUMOylated hSpt16 displayed increased survival rates compared to mice infected with wild-type VACV, suggesting FEAR antagonism promotes poxvirus pathogenesis (34,74,80). within an alpha helix conserved between SeV and HPIV-1 C proteins and thus may be important for promoting SUMOylated hSpt16 depletion (73). Notably, the C-terminus of paramyxovirus C proteins mediates interaction with STAT1 (81). Thus, paramyxovirus C proteins use N-terminal domains for FEAR pathway inhibition, while their C-terminal domains antagonize IFN-I signaling.
FACT and ETS-1 upregulation are associated with oncogenesis, transformation, and tumor invasion (82)(83)(84)(85), leading to the development of small molecule FACT inhibitors called "curaxins" currently in clinical trials to treat human malignancies (86). Coinciden tally, VSV mutant strains are similarly being pursued as an oncolytic agent in the clinic. However, some cancer cell types are refractory to oncolytic VSV strain infection, and the FEAR pathway has been recently implicated in the restriction of oncolytic VSV strains in some of these cancer cell types (73,(87)(88)(89). Curaxin treatment together with oncolytic VSV infection increased cancer cell death in multiple resistant cell types, suggesting that FEAR inhibition through curaxin treatment can sensitize refractory cancer cells to oncolytic VSV infection (73). Thus, this combinatorial therapeutic approach may broaden the use of oncolytic VSV therapy against cancer cell types that would otherwise be refractory to such virotherapy.
Interestingly, FACT has been reported to have additional roles outside of the FEAR pathway during infection that can either help or hinder viral replication. For example, RNAi screens identified both hSpt16 and SSRP1 subunits as suppressors of transcription by integrated retroviruses, such as human immunodeficiency virus-1 (HIV-1; Retroviridae) (90). In contrast, FACT is recruited by members of the Herpesviridae family to promote viral replication. The herpes simplex virus-1 (HSV-1; Herpesviridae)-encoded ICP22 protein was found to interact with both FACT subunits, which resulted in the redistribution of FACT to viral genomes, facilitation of viral transcription, and suppression of FACTdependent cellular gene transcription (91,92). HCMV was also shown to recruit FACT to drive viral reactivation by helping to activate the major immediate early promoter, which controls the expression of immediate early genes encoding viral transactivators required for lytic replication (93). Curaxins reduced HCMV reactivation and lytic infection, suggesting that such drug treatments may be capable of maintaining HCMV latency (93)(94)(95). Together, these observations, along with the conservation of FACT from yeast to humans, suggest that the FACT complex has likely played many diverse roles in the regulation of viral replication throughout eukaryotic evolution. Moreover, FACT inhibitor drugs may be useful therapeutics for either promoting the replication of oncolytic viruses antagonized by FACT (i.e., to improve oncolytic virotherapy efficacy) or for inhibiting the active replication of viruses that require FACT for their life cycle, as in the case of herpesviruses.
## Degradation of virus particles and components: autophagy in antiviral defense
Autophagy is a highly conserved catabolic pathway in eukaryotic cells that supports cellular homeostasis by degrading and recycling superfluous or damaged organelles, proteins, lipids, and nucleic acids through the use of lysosomes (77)(78)(79)(80). When cells experience stressors, such as pathogen invasion, autophagy aids in the maintenance of cellular integrity and clearance of pathogens through the action of autophagosomes, which are double-membrane vesicles that deliver cellular components to lysosomes for degradation (96)(97)(98)(99). Over 30 autophagy-related (ATG) proteins orchestrate the different steps of autophagy, including initiation, phagophore membrane elongation and closure, autophagosome maturation and fusion with lysosomes, and degradation and recycling (Fig. 2A, Steps 1-5) (100,101). Autophagy initiation can be activated through inhibi tion of the PI3K-AKT-mTOR signaling pathway or activation of the PI3K-VPS34-Beclin1 pathway (Fig. 2A, Step 1) (98,(102)(103)(104). Initiation involves the ULK1 complex activation of the type III phosphoinositide 3-kinase (PI3K) complex I (including Beclin1 and Vps34), generating phosphatidylinositol-3-phosphate (PtdIns3P) on membranes and forming the phagophore (Fig. 2A, Step 1) (105). Elongation and closure of the phagophore to form the autophagosome are supported by both the ATG5-ATG12-ATG16L1 complex, an E3-like ligase, promoting LC3 oligomerization on the phagophore membrane and the LC3-phosphatidylethanolamine system converting LC3 to LC3-II (Fig. 2A, Step 2) (106-110). LC3-II is attached to membranes during elongation, serving as docking sites for host receptors that deliver interaction-specific cargo (Fig. 2A, Step 2) (111,112). Subsequent fusion of autophagosomes with lysosomes to generate mature autolyso somes involves the coordination of many proteins, including SNARE complexes, Rab GTPases, and HOPS, which facilitate membrane apposition (Fig. 2A, Steps 3 and 4) (112)(113)(114)(115)(116). Rab7 and UVRAG promote fusion (112)(113)(114)(115), while Rubicon negatively regulates autophagosome maturation (Fig. 2A, Steps 3 and 4) (117,118). These autolysosomes then degrade their contents with hydrolases, and degradation products are subse quently recycled (Fig. 2A, Step 5) (117,118).
Autophagy plays a crucial role in antiviral defense by degrading viral components, stimulating immune responses, and modulating inflammation (97)(98)(99)120). Through a process called "virophagy, " autophagy selectively targets viral particles and components, enclosing them in autophagosomes for degradation via lysosomes (119,121,122). Host selective autophagy receptor proteins, such as p62/SQSTM1, TAX1BP1, NDP52/ CALCOCO2, NBR1, and OPTN, can recognize specific viral cargo, tethering them to the phagophore through LC3-interacting regions and ubiquitin-binding domains (Fig. 2B) (97,(123)(124)(125)). One early discovered example of virophagy involves p62-directed degradation of the Sindbis virus (SINV; Togaviridae) capsid in the autophagosome (126,127). More recently, host cyclin-dependent kinase-like 5 (CDKL5) was shown to promote p62 phosphorylation, which promoted interaction with the SINV capsid (Fig. 2B) (119). CDKL5 knockout mice exhibited increased viral antigen levels and neuronal cell death after SINV infection (119). Furthermore, challenges with several neurotropic viruses resulted in reduced survival rates in these knockout mice, suggesting CDKL5-mediated virophagy is a critical antiviral defense in vivo (119). p62 is also involved in targeting VP1 and VP3 of Seneca Valley virus (Picornaviridae) (128), HSV-1 (129), and capsid proteins of Chikungunya virus (Togaviridae) (130). Besides ubiquitin tags, galectin decoration of viral capsid proteins can also facilitate p62-mediated virophagy (131,132). Another cargo receptor, OPTN, targets HSV-1 proteins for degradation in the central nervous system (133). A diverse array of additional host factors can function as cargo receptors, which may explain how host cells can utilize autophagy to recognize and eliminate disparate viral pathogens (134)(135)(136)(137)(138).
In addition to canonical IFN pathways (120,(139)(140)(141)(142), cGAS-STING (143,144), NF-κB (145)(146)(147), and eIF2α kinase signaling (141,148,149) pathways influence autophagy activation in an IFN-I-independent manner (Fig. 2C). The cGAS-STING pathway, known to recognize cytosolic DNA and trigger IFN production (150), has two recent reports suggesting STING has IFN-I-independent antiviral functions (Fig. 2C, left) (143,144). In the presence of cGAMP, poly(dA:dT), or HSV-1 infection, STING activated a non-can onical autophagy response that is dependent on ATG5 but independent of canonical autophagy factors, such as ULK1 and p62 (Fig. 2C, left) (151,152). Once activated, STING supported LC3 lipidation, which is essential for autophagy, through WIPI2 and ATG5 (Fig. 2C, left) (152). The pathway is also tightly regulated by selective autophagy; p62 targets ubiquitinated cGAS and STING for degradation, preventing overactivation of the pathway (Fig. 2C, left) (153). Given that cGAS-STING signaling is conserved in some invertebrates (which typically lack IFNs), the activation of autophagy may be the primordial function of the cGAS-STING signaling pathway, predating its role in activating IFN-I signaling that arose during the evolution of vertebrates (152). Interest ingly, TRIM23, an E3 ubiquitin ligase and GTPase, was recently found to be crucial for a cGAS-STING-TBK1-TRIM23 antiviral autophagy pathway (154). HSV-1 infection or cGAS-STING stimulation was found to induce TBK1-mediated TRIM23 phosphorylation, which, in turn, triggered TRIM23 autoubiquitination and GTPase activity, leading to autophagy and suppression of HSV-1 replication (154). Cross-talk between autophagy and other immunity pathways can also regulate autophagic antiviral responses (Fig. 2C). For example, inhibition of NF-κB pathway signaling can promote autophagy through upregulation of ATG protein expression (Beclin1) and induction of autophagosome formation; this inhibition also reduced p-AKT and p-mTOR levels, further driving autophagy activation (Fig. 2C, middle) (145)(146)(147). During infection, OPTN not only helped target ubiquitinated cargo for degradation but also inhibited NF-κB activation to promote autophagy (Fig. 2C, middle) (145,155). eIF2α kinase activation by various stress stimuli, including starvation and viral infection, leads to eIF2α phosphorylation and also induces ATG gene expression (Fig. 2C, right) (148). During HSV-1 infection, the dsRNA-sensing eIF2α kinase PKR phosphorylates eIF2α and induces antiviral autophagy (Fig. 2C, right) (141). Interestingly, during infection with pseudorabies virus (Herpesviridae), the host USP14 deubiquitinase becomes inactivated, which serves to activate the PERK eIF2α kinase and the unfolded protein response (156). Together, this promotes the degradation of the viral transcriptional regulator VP16 via p62-mediated autophagy, thereby suppressing viral replication (156).
The importance of autophagy in antiviral defense is underscored by the discovery of many virally encoded inhibitors of autophagy. Viruses often manipulate the PI3K-AKT-mTOR or PI3K-VPS34-Beclin1 pathways to exploit autophagic membranes or machinery while also avoiding degradation to benefit their replication (99,104,120). These viral antagonists employ diverse strategies to block autophagy-mediated degradation, ranging from the direct inhibition or degradation of ATG proteins to the prevention of lysosomal acidification (Table 2) (97,129,142,.
Aberrant regulation of autophagy has been tied to various pathological conditions, including neurodegenerative diseases (192)(193)(194)(195)(196), cancers (197,198), inflammation (199), and infections (200,201). Importantly, drugs targeting autophagy have shown therapeu tic potential in a wide array of viral infections using cell culture and mouse model systems (122). These include autophagy activators, rapamycin (202)(203)(204), metformin (205,206), and CSC27 (207), as well as autophagy inhibitors, wortmannin (208) and corticosteroids (209)(210)(211)(212). Clearly, there may be utility in developing therapeutic strategies to manipulate autophagic pathways for the treatment of viral disease.
## Stem cells and the endogenous RTase/RNase H-mediated antiviral system
Even though IFN-I responses are often the most critical antiviral defense induced during viral infection of most differentiated cell types, pluripotent stem cells do not elicit a productive IFN response after infection with viruses, exposure to PAMPs, or when treated with recombinant IFN (13,213). Interestingly, evidence suggests that the maintenance of stem cell pluripotency is compromised when these cells are engineered to express a constitutively active form of IRF7 (which induces IFN-I responses) (13). These findings suggest that the IFN-I response and stem cell pluripotency may be incompatible with one another, possibly explaining why stem cells do not employ the IFN-I response for antiviral defense (13). However, this surprising finding has led to the suggestion that stem cells may either be resistant to viral infections because they lack expres sion of receptors or co-receptors required for viral entry (214), or because these cells utilize alternative, IFN-I-independent antiviral responses to block infection. For example, antiviral RNAi responses have recently been reported to restrict ZIKV and SARS-CoV-2 replication in mammalian stem cells through the action of a shortened Dicer isoform termed antiviral Dicer, which is particularly active in dicing long dsRNAs (215). However, the role of RNAi in mammalian antiviral defense remains controversial and has been recently reviewed elsewhere (28).
Recently, a nucleic acid-based antiviral system, distinct from RNAi, was identified in mouse embryonic stem cells (ESCs) that utilizes the action of endogenous reverse transcriptases (RTases) to mediate RNA virus restriction. Mammalian genomes contain a considerable number of residual endogenous retrovirus (ERV) sequences-an outcome of historical retroviral infection, reverse transcription, and integration events. While most ERVs are defective, some encode retroviral proteins, such as RTases, that are highly active in ESCs compared to differentiated, somatic cells (216)(217)(218)(219). Mechanistically, when mouse ESCs are infected with RNA viruses, RTases function to generate viral DNA copies from viral RNA templates, creating viral DNA/RNA duplexes (219). This unique hybrid structure is recognized and cleaved by RNase H, which specifically degrades the RNA strand, ultimately leading to viral inhibition (219). This newly identified endogenous RTase/RNase H-mediated antiviral system (termed "ERASE" for short) helps to explain how stem cells can resist viral infection without inducing IFN-I responses.
Currently, there are no known viral inhibitors of ERASE. Although mouse ESCs pre-treated with azidothymidine (AZT), a mammalian RTase inhibitor, increased viral
## Impeding viral release: septins
Septins are a conserved family of GTP-binding proteins that were initially identified as ~10 nm nonpolar filaments forming ring-like structures at the bud necks of Saccharo myces cerevisiae (221). Subsequent studies revealed that septins function at the plasma membrane to regulate a range of cellular processes, including cytokinesis, ciliogenesis, phagocytosis, and innate immunity (222,223). Septins assemble into hetero-oligomeric complexes that can further polymerize into filaments, rings, and cage-like structures (224). These higher-order assemblies serve as scaffolds for intracellular protein localiza tion and as barriers that compartmentalize membrane domains (223,225). In humans, the 13 septin genes are grouped into four subfamilies based on sequence homology: SEPT2 (SEPT1, 2, 4, 5), SEPT3 (SEPT3, 9, 12), SEPT6 (SEPT6, 8, 10, 11, 14), and SEPT7 (SEPT7) (225)(226)(227)(228). Members from each subfamily co-assemble into complexes capable of forming functional septin structures (223,229,230). During bacterial infection, septins mediate cell-intrinsic immunity by forming filamentous cages that entrap motile pathogens, such as Shigella flexneri (231,232). Sensing the micron-scale membrane curvature as S. flexneri invades cells and escapes from its phagosomal vacuole into the host cytoplasm, septins recognize the poles of growing S. flexneri and build filament cages around the bacterium during its early stages of replication (231,232). This physical septin cage inhibits actin tail formation, a mechanism used by S. flexneri for intracellular propulsion, as well as targets the bacteria for degradation via autophagy (232,233). Notably, septin entrapment is enhanced against growing bacteria, as non-replicating bacteria partially evaded cage formation (231,234). The direct mechanisms behind this escape strategy and the evolutionary tradeoffs for having this escape mechanism have yet to be fully elucidated.
The best-characterized antiviral roles of septins have been described in the context of VACV, which produces intracellular enveloped virions that egress from infected cells using both microtubule-dependent transport and actin-based motility (235)(236)(237). Septins were originally implicated in VACV restriction by two independent genome-wide RNAi screens, which identified several septins (e.g., SEPT7, SEPT11, etc.) as antiviral factors (238,239). RNAi-mediated knockdown of septins resulted in enhanced VACV replication and increased plaque size, indicating that septins restrict VACV spread (240). Without septins, VACV cell-associated enveloped viruses (CEVs; i.e., virions that exit the cell but remain attached to the membrane where they facilitate Arp2/3 complexdependent actin polymerization) (237,241) exhibited more frequent and longer-lived actin tails, indicating that septins suppress actin tail formation required for VACV egress (240). Similar to their interactions with S. flexneri, septins were found to assemble into cage-like structures at the plasma membrane that trap CEVs (Fig. 3A) (226,240). However, this restriction appears to be transient as septin cages are eventually dismantled by localized actin remodeling events facilitated by VACV A36R proteins that engage with the host SH2/SH3 domain-containing adaptor Nck (Fig. 3B) (240). Nck then recruits dynamin to promote septin disassembly, which is followed by formin-mediated actin polymerization and displacement of septin structures (Fig. 3B) (240). This subsequently permits actin tail formation and facilitates VACV egress (Fig. 3C) (240).
While VACV represents the primary model for studying septin-mediated antiviral defense, emerging evidence suggests potential interactions between septins and other viral pathogens. For example, the ZIKV-encoded NS2B3 protease was shown to target SEPT2 and cause delayed cytokinesis in neural stem cells (242). Although it is unclear how this SEPT2 targeting impacts the viral life cycle, it has been suggested that this may contribute to infection-associated neurotoxicity (242). Interestingly, although septins do not appear to be ISGs (243), there have been reports of differential regulation of septins at both the transcriptional and protein levels during viral infection (244). For example, infection with hepatitis C virus (HCV) was associated with transcriptional upregulation of septins that, in turn, underwent filament assembly to regulate lipid droplet growth (245), and SEPT2 and SEPT6 have been implicated in supporting HCV replication (245), suggesting that septins can play antiviral or proviral roles (244). In contrast to infectioninduced septin filamentation observed with HCV, the human immunodeficiency virus Tat protein was shown to promote disruption of SEPT7 filaments (246). Thus, there are clearly complex relationships between septins and viral pathogens that remain poorly understood.
## Inhibition of viral replication and spread through programmed cell death
While cell intrinsic factors attempt to limit viral infection, PAMPs and damage-associ ated molecular patterns (DAMPs) activate programmed cell death pathways to restrict viral spread, representing a "last resort" innate response (247). Cell death occurs via non-lytic or lytic programs (248,249). Apoptosis is non-lytic and immunologically silent, maintaining membrane integrity while inducing morphological changes, such as nuclear fragmentation and membrane blebbing, that signal clearance by phagocytic cells (250,251). In contrast, pyroptosis and necroptosis are lytic and pro-inflammatory, wherein cellular membranes are ruptured and release cellular contents that trigger strong immune responses (252). Although programmed cell death mechanisms can be potent antiviral defenses, viruses have often evolved strategies to prolong viral replication or promote viral spread by inhibiting or usurping key components of these cell death pathways (253,254). Table 3 provides examples of how viruses regulate three key antiviral programmed cell death pathways: apoptosis, pyroptosis, and necroptosis.
Apoptosis is mediated by intrinsic (mitochondrial) or extrinsic (death receptor) pathways, both converging to activate a cascade of cysteine aspartyl proteases, or caspases (CASPs), which cleave host substrates, resulting in cellular degradation and death (344,345). Apoptotic initiation typically involves (i) extracellular ligand-receptor binding, including TNF family of ligands such as TNF-α, TRAIL, and FasL to TNFR1, DR4 or 5, and FasR, respectively; (ii) virus-host receptor interaction during attachment prior to entry; (iii) host sensing of viral replication intermediates; or (iv) virus-induced activation of the unfolded protein response (254,346,347).
The intrinsic pathway responds to intracellular physiological stresses that increase mitochondrial membrane permeability (348). Normally, B-cell lymphoma-2 (Bcl-2)-like proteins inhibit mitochondrial pore-forming Bcl-2-associated X protein (BAX) and Bcl-2 homologous antagonist/killer (BAK) from oligomerizing. However, oligomerization occurs when transcriptional or post-translational stress signals release Bcl-2-like proteins from BAX/BAK, or when pro-apoptotic BH3-only proteins directly activate BAX/BAK, causing mitochondrial outer membrane permeabilization and the release of apopto genic molecules, like Cytochrome C (349)(350)(351)(352). Cytochrome C unleashes active CASP9 to initiate CASP3 and CASP7 proteolytic activity (353)(354)(355)(356)(357). The extrinsic pathway is triggered when transmembrane death receptors bind to extracellular ligands, induc ing oligomerization and formation of the death-inducing signaling complexes (DISC) through death domain interactions (358)(359)(360)(361)(362). DISC activates CASP8, which simulta neously cleaves pro-apoptotic Bid protein that feeds into the intrinsic pathway by activating BAX/BAK (363), and initiates CASP3 and CASP7 activation (Fig. 4A) (364). These executioner caspases cleave cellular components, exposing intracellular phos pholipid phosphatidylserine on the cell surface, signaling for phagocytic clearance without provoking inflammation (365)(366)(367). Overall, apoptosis is a key antiviral defense, eliminating infected cells while minimizing immune activation.
Viruses have evolved an array of strategies to regulate apoptosis (Table 3). For example, many viruses encode viral Bcl-2-like proteins, which are structurally homol ogous to cellular Bcl-2 proteins (368). Herpesvirus saimiri (Herpesviridae) proteins (ORF16) and adenovirus (Adenoviridae) (E1B19K) express Bcl-2 homologs that inhibit BAX/BAK-mediated mitochondrial permeabilization (284,285,296). Poxviruses encode multiple Bcl-2 proteins that inhibit apoptosis at several junctures. For example, VACV F1 proteins prevent BAX/BAK oligomerization and inhibit CASP9 function, while VACV B13 suppresses initiator caspases in both intrinsic (CASP8) and extrinsic (CASP9) pathways (261,263,(369)(370)(371). In contrast, reovirus (Reoviridae) μ1 proteins promote apoptosis to enhance viral spread while avoiding triggering an inflammatory response (306,307). West Nile virus (WNV, Flaviviridae) NS3 proteins also induce CASP3-triggered apoptosis in neuronal cells, resulting in severe neuropathogenesis (372,373). However, synthetic tetrapeptide aldehyde molecules that reversibly inhibit CASP3 and CASP7 specifically have promising therapeutic potential, as they significantly reduced WNV viral titers, as well as WNV-induced meningitis and encephalitis in murine brain tissues (372,373). These diverse viral strategies underscore the central role of apoptosis in host antiviral defense.
Pyroptosis is a lytic programmed cell death pathway also driven by CASP activa tion but results in membrane rupture and inflammation (374)(375)(376). PRRs, like Toll-like receptors (TLRs) (377), AIM2 (378), RIG-I-like receptors (379), and NOD-like receptors (B) Initiation of viral escape and actin nucleation. The viral envelope protein A36R is phosphorylated, enabling the recruitment of the host adaptor protein Nck (green) (240). Nck, in turn, recruits dynamin (orange spirals), a GTPase that contributes to local septin disassembly (240). As septins are removed, formins (blue) are recruited, initiating the formation of linear actin filaments (red) beneath the virion (240). (C) Actin tail formation and outward propulsion. The viral particle then engages the Arp2/3 complex via N-WASP (240), leading to the formation of branched actin filaments. These actin tails push the virion away from the membrane, producing a cell-associated enveloped virus poised for transmission to neighboring cells (240) (NLRs) (380), engage with PAMPs (e.g., viral nucleic acids [381][382][383]) or endogenously generated DAMPs (oxidized lipids [384], uric acid [385], reactive oxygen species [386], ATPs [387], etc.) during infection (388)(389)(390). PRR-ligand interactions result in the recruit ment of adaptor proteins that organize a cytosolic multiprotein inflammasome complex, where pro-caspases are hydrolyzed into mature products to cleave gasdermin D (GSDMD)-the executioner protein (293,(391)(392)(393). The N-terminal fragment of GSDMD oligomerizes to non-selectively perforate the cell membrane, causing osmotic swelling and lysis (394)(395)(396). Concurrently, CASP1 processes pro-inflammatory cytokines pro-IL-1β and pro-IL-18 into their active forms (IL-1β and IL-18), which leak through the GSDMD pore to signal neighboring cells to initiate immune defenses and activate adaptive responses (397-400) (Fig. 4B). Thus, pyroptosis eliminates infected cells and triggers a strong inflammatory response to further limit pathogen spread. As with apoptosis, viruses have evolved diverse mechanisms to disrupt inflammasome function to prevent cytokine production and cell death (Table 3). Orthopoxviruses, such as Mpox virus and variola virus, express serpin protease inhibitors that antagonize CASP1, preventing pyroptosis and IL-1β maturation (267,401,402). HSV-1 ICP0 proteins target NLRP1 and function to stabilize the inhibitory N-terminal NLRP1 fragment in a manner that depends upon the cytoplasmic localization and E3 ubiquitin ligase activity of ICP0 (314,403). Coronaviruses, such as SARS-CoV-2, encode two factors that directly impede pyroptosis: the nucleocapsid protein binds to GSDMD, prevent ing CASP1-mediated GSDMD N-terminal processing, while NSP5 cleaves GSDMD to remove critical pore-forming N-terminal residues, preventing pyroptotic initiation (316,404,405). In contrast, SARS-CoV-2 proteins ORF3a, E, and M trigger NLRP3 inflammasome-activated pyroptosis in certain cell types, such as macrophages and pulmonary epithelial cells (406). Endothelial cells infected with picornaviruses, such as coxsackievirus and enterovirus A71, undergo NLRP3-mediated pyroptosis that elevates inflammatory responses. However, NLRP3 inhibitor (MCC950 sodium) treatment improved cell viability, reduced CASP1 activity, and attenuated IL-1β cytokine production during infection, suggesting that NLRP3 inhibitors may be a promising therapy against pyroptotic inflammation (407,408). Together, these viral mechanisms aimed to subvert cellular pyroptosis demonstrate the critical role of inflammasome signaling in antiviral immunity.
Necroptosis is a caspase-independent, pro-inflammatory cell death program that morphologically resembles pyroptosis and serves as a fail-safe when apoptotic caspase activation is inhibited during infection (409)(410)(411). Necroptosis shares many initiating stimuli with apoptosis, for example, the TNF family of ligands, endoplasmic stress, and viral nucleic acids (412). A common feature of necroptotic receptors is that they contain a receptor-interacting protein kinase (RIPK) homology interaction motif (RHIM) that recruits receptor-interacting protein kinase 3 (RIPK3), activating mixed lineage kinase domain-like (MLKL) proteins (412)(413)(414).
TNF-α is the most well-characterized necroptotic trigger. When TNF-α is bound to its receptor, TNFR1, multiple proteins, including receptor-interacting protein kinase 1 (RIPK1), dock to the cytosolically protruding TNFR1 interface to assemble complex I (415)(416)(417). RIPK1 ubiquitination at complex I promotes cell survival, whereas deubiquitinated RIPK1 causes RIPK3 phosphorylation (418)(419)(420)(421). Alternatively, Z-DNA binding protein 1 (ZBP1) is an intracellular PRR, sensing Z-form nucleic acids generated during viral infection, which operates independently of RIPK1 to trigger necroptosis (422)(423)(424). ZBP1 detection of these nucleic acid forms allows its RHIM domain to efficiently recruit and directly interact with RIPK3, resulting in RIPK3 autophosphorylation. TLR engagement with viral PAMPs also results in RIPK3 activation (250,251). Regardless of RIPK1 involve ment, active RIPK3 phosphorylates MLKL, induces a conformational change, and allows pore-forming oligomeric MLKL complexes to translocate to the cell membrane (425,426). Although the precise mechanism and structural requirements governing MLKL assembly remain unknown, MLKL permeabilizes the cell membrane, resulting in NF-κB and mitogen-activated protein kinase signaling and decreased membrane integrity, causing a Ca 2+ ion influx, promoting cell swelling and membrane rupture (Fig. 4C) (427,428). Thus, necroptosis represents a vital backup cell death pathway, providing a robust immune response against pathogens upon apoptotic inhibition.
Though necroptosis likely evolved as a "back-up" cell death program, many viruses manipulate this pathway (Table 3). Epstein-Barr virus (EBV)-encoded latent membrane protein 1 suppresses necroptosis by interfering with RIPK1 ubiquitination, limiting the recruitment/assembly of complex I proteins and downstream RIPK3 phosphorylation, thereby maintaining cell survival (332). EBV also induces RIPK3 promoter hypermethyla tion to suppress expression (339). Pseudorabies virus VP22 proteins interact with ZBP1, inhibiting RIPK3 recruitment (341). Poxviruses encode E3L proteins that sequester Z-form RNA, preventing ZBP1 activation, and deploy decoy MLKL proteins lacking the cell-mem brane binding domains that sequester RIPK3, preventing cellular MLKL phosphorylation, oligomerization, and permeabilization (329)(330)(331). In contrast, norovirus NS3 proteins have repurposed the MLKL executioner domain to actively induce cell death, facilitating viral spread (429). Similarly, IAV NS1 invokes necroptosis by enhancing interactions with RIPK3 phosphorylators, facilitating MLKL oligomerization to regulate viral egress (328,430,431). Interestingly, a selective RIPK3 inhibitor, UH15-38, has shown therapeutic promise against IAV in vivo, as it blocked IAV-triggered necroptosis, prevented inflammation, and improved animal survival (432). This example illustrates how our understanding of virus-cell death pathway interactions may form the basis for the rational design of novel antiviral therapeutic strategies.
Although IFN-I is unnecessary to initiate programmed cell death in most cell types, IFN-I signaling can synergize with other inflammatory pathways to promote cell death. For example, intracellular PRR sensing of DNA and RNA leads to the activation of RIPK3 in bone marrow-derived macrophages, though this necroptosis response in these macrophages requires cooperative signaling between IFN-I and TNF-α (433). Furthermore, IFN-I signaling induce expression of cell death-associated factors and sensitize cells to the activation of infection-triggered cell death pathways, illustrating the importance of cross-talk between these antiviral defenses (434,435).
## CONCLUSIONS AND FUTURE DIRECTIONS
Since its discovery over 60 years ago (436), the IFN-I response has been the focus of antiviral immunity studies in mammals. However, the examples above clearly illustrate that mammals also employ IFN-I-independent antiviral responses that play crucial, and often complementary, roles in controlling viral replication, particularly during early stages of infection or in cell types with limited IFN-signaling capacity. These defenses include cell intrinsic restriction pathways such as autophagy that can degrade incoming virions (119), as well as constitutive factors and systems such as TRIM7 (58,59,62,66) and ERASE (219) that target viral proteins or nucleic acids, respectively, for destruction. Even if viruses can complete their replication, additional IFN-I-independent host defense mechanisms exist to prevent their spread to the next cell, such as the entrapment of exiting viral particles by septins (239,240) and the induction of cell death pathways (247). It is important to reiterate that many other known IFN-I-independent respon ses that could not be discussed in detail here, such as various metabolic reprogram ming-mediated defenses that create an unfavorable intracellular environment for viral replication, also play critical roles in antiviral immunity (437) (see also Table 1). Thus, together, these IFN-I-independent defenses provide a multilayered response to restrict viral replication, helping to control viral infection when IFN signaling is compromised by viral antagonism or kinetically delayed. However, many key questions regarding the molecular mechanisms governing the activation, regulation, and cell/tissue specificity of these IFN-I defenses and pathways remain. Moreover, a deeper understanding of the cross-talk (if any) between these pathways and the canonical IFN-I response, whether synergistic, redundant, or antagonistic, requires further investigation. Evidence suggests that low-level, constitutive (or "tonic") IFN-I signaling may help to maintain immune homeostasis by regulating other immune pathways (438)(439)(440). Therefore, if, for example, IFN-I signaling maintains basal expression of key components of IFN-I-inde pendent pathways, this could result in important cross-talk between IFN-I-dependent and -independent responses. New approaches will be needed to specifically identify IFN-I-independent antiviral defenses. High-throughput approaches such as genome-wide CRISPR activation screens have already proven useful in the identification of novel IFN-I-independent antiviral factors, including TRIM7, as well as unappreciated host factors regulating basal and inducible ISG expression. However, such screens on their own do not per se identify only IFN-I-independent responses and, in fact, are often swamped by hits related to the IFN-I response. However, combining such screens with IFNAR -/-or STAT1 -/-cells may be a powerful strategy to more specifically identify antiviral defense factors that function independently of IFN-I signaling. Moreover, given the fact that viral IFN-I antagonists have revealed key facets of IFN-I response components and regulation (10), viral antagonists might also be useful tools for identifying and studying IFN-I-inde pendent responses. For example, by screening for mammalian poxvirus-encoded factors that retain immunosuppressive function in insect cells (which lack IFNs), poxvirus A51R proteins were discovered as novel immune evasion proteins that rescue the restricted replication of heterologous viruses (80). Follow-up studies aimed at understanding A51R-mediated immune suppression led to the discovery of the FEAR pathway (74), which appears to be conserved between mammals and insects (73). Thus, IFN-I-defective mammalian cells, invertebrate host models, and viral antagonist screens may provide complementary approaches to uncovering new IFN-I-independent defenses.
The recent finding that TRIM7 -/-mice are equally susceptible to MNoV infection as wild-type animals underscores the importance of studying IFN-I-independent antiviral factors and pathways identified from cell culture studies in animal models to assess their physiological relevance to controlling viral replication in vivo. Studies with TRIM7 -/- mice will be needed to determine the in vivo role of this host factor in the restriction of other viruses identified as being restricted by TRIM7 in cell culture studies, such as enteroviruses (58,63). Although several of these recently described IFN-I-independent mechanisms may depend upon host factors that are essential for mouse development (e.g., septins), the use of alternative approaches, such as conditional knockout/knock down mice (441), may provide strategies for examining their role in antiviral defense in animals.
In conclusion, IFN-I-independent antiviral defenses represent vital, though historically underappreciated, components of the mammalian innate immune response. A deeper understanding of how these defenses are regulated, and how viral pathogens engage with them, will not only lead to a more complete understanding of innate immunity but may also provide new therapeutic strategies to augment their activity for the treatment of viral disease. Development of such approaches could especially benefit individuals with genetic deficiencies in IFN responses, who are predisposed to viral infection (6,9). Furthermore, understanding IFN-I-independent responses in mammals may also provide critical insights into how lower eukaryotes that lack IFNs combat viral infection. Such knowledge may be useful for the development of intervention strategies to block viral replication in insect vectors that transmit viral pathogens to mammalian hosts. Thus, with the growing appreciation for, and interest in, IFN-I-independent responses, this is an exciting time to explore the additional layers of innate defenses employed by mammals that go beyond IFNs.
## References
1. Schoggins (2019) "Interferon-stimulated genes: what do they all do?" *Annu Rev Virol*
2. Swaraj, Tripathi (2024) "Interference without interferon: interferonindependent induction of interferon-stimulated genes and its role in cellular innate immunity"
3. Lazear, Schoggins, Diamond (2019) "Shared and distinct functions of Type I and Type III interferons" *Immunity*
4. Mihaescu, Chifiriuc, Bleotu et al. (2023) "Role of interferons in the antiviral battle: from virus-host crosstalk to prophylactic and therapeutic potential in SARS-CoV-2 infection" *Front Immunol*
5. Mcdougal, Maria, Ohlson et al. (2023) "Interferon inhibits a model RNA virus via a limited set of inducible effector genes"
6. Stine, Cahill, Humphries (2025) "Interferons in human inborn errors of disease" *mBio*
7. Honda, Yanai, Negishi et al. (2005) "IRF-7 is the master regulator of type-I interferon-dependent immune responses" *Nature*
8. Wong, Qiu (2018) "Type I interferon receptor knockout mice as models for infection of highly pathogenic viruses with outbreak potential" *Zool Res*
9. Naesens, Callens, Kerre et al. (2025) "Inborn errors of nucleic acid sensing and type I interferon signaling determine viral susceptibility in humans" *Eur J Immunol*
10. Zhu, Chiang, Gack (2023) "Viral evasion of the interferon response at a glance" *J Cell Sci*
11. Secombes, Zou (2017) "Evolution of interferons and interferon receptors" *Front Immunol*
12. Yu, Bruneau, Brennan et al. (2021) "Battle royale: innate recognition of poxviruses and viral immune evasion" *Biomedicines*
13. Eggenberger, Blanco-Melo, Panis et al. (2019) "Type I interferon response impairs differentiation potential of pluripotent stem cells" *Proc Natl Acad Sci*
14. Iversen, Reinert, Thomsen et al. (2016) "An innate antiviral pathway acting before interferons at epithelial surfaces" *Nat Immunol*
15. Sleijfer, Bannink, Van Gool et al. (2005) "Side effects of interferon-alpha therapy" *Pharm World Sci*
16. Jung, Mckenna, Vijayamahantesh et al. (1916) "Protective versus pathogenic type I interferon responses during virus infections" *Viruses*
17. Castilha, Biondo, Trugilo et al. (2025) "APOBEC3 proteins: from antiviral immunity to oncogenic drivers in HPV-positive cancers" *Viruses*
18. Sadeghpour, Khodaee, Rahnama et al. (2021) "Human APOBEC3 variations and viral infection" *Viruses*
19. Cheung, Yang, Wu (2025) "dNTP depletion and beyond: the multifaceted nature of SAMHD1-mediated viral restriction" *J Virol*
20. St Gelais, De Silva, Amie et al. (2012) "SAMHD1 restricts HIV-1 infection in dendritic cells (DCs) by dNTP depletion, but its expression in DCs and primary CD4+Tlymphocytes cannot be upregulated by interferons" *Retrovirology (Auckl)*
21. Guo, Yang, Li et al. (2024) "The SAMHD1-MX2 axis restricts HIV-1 infection at postviral DNA synthesis"
22. An, Ge, Shao et al. (2022) "Interferon-inducible SAMHD1 restricts viral replication through downregulation of lipid synthesis" *Front Immunol*
23. Peng, Meng, Zhang et al. (2022) "Structure and function of an effector domain in antiviral factors and tumor suppressors SAMD9 and SAMD9L" *Proc Natl Acad Sci*
24. Meng, Zhang, Yan et al. (2018) "A paralogous pair of mammalian host restriction factors form a critical host barrier against poxvirus infection" *PLoS Pathog*
25. Zhang, Ji, Chaturvedi et al. (2023) "Human SAMD9 is a poxvirus-activatable anticodon nuclease inhibiting codon-specific protein synthesis" *Sci Adv*
26. Embry, Gammon (2024) "Abortive infection of animal cells: what goes wrong" *Annu Rev Virol*
27. Wu, Zhang, Li et al. (2021) "The regulation of integrated stress response signaling pathway on viral infection and viral antagonism" *Front Microbiol*
28. Wang, Li (2024) "Current advances in antiviral RNA interference in mammals" *FEBS J*
29. Li, Lu, Han et al. (2013) "RNA interference functions as an antiviral immunity mechanism in mammals" *Science*
30. Maillard, Ciaudo, Marchais et al. (2013) "Antiviral RNA interference in mammalian cells" *Science*
31. Tada, Zhang, Koyama et al. (2015) "MARCH8 inhibits HIV-1 infection by reducing virion incorporation of envelope glycoproteins" *Nat Med*
32. Zhang, Lu, Liu (2018) "MARCH2 is upregulated in HIV-1 infection and inhibits HIV-1 production through envelope protein translocation or degradation" *Virology (Auckl)*
33. Umthong, Lynch, Timilsina et al. (2021) "Elucidating the antiviral mechanism of different MARCH factors" *mBio*
34. Seo, Oliveira, Ye et al. (2024) "Poxvirus A51R proteins regulate microtubule stability and antagonize a cell-intrinsic antiviral response" *Cell Rep*
35. Irwan, Cullen (2023) "The SMC5/6 complex: an emerging antiviral restriction factor that can silence episomal DNA" *PLoS Pathog*
36. Xu, Ma, Zhang et al. (2018) "PJA1 coordinates with the SMC5/6 complex to restrict DNA viruses and episomal genes in an interferon-independent manner" *J Virol*
37. Platanitis, Decker (2018) "Regulatory networks involving STATs, IRFs, and NFκB in inflammation" *Front Immunol*
38. Wang, Xu, Su et al. (2017) "Transcriptional regulation of antiviral interferon-stimulated genes" *Trends Microbiol*
39. Schoggins, Macduff, Imanaka et al. (2014) "Panviral specificity of IFN-induced genes reveals new roles for cGAS in innate immunity" *Nature*
40. Dalskov, Gad, Hartmann (2023) "Viral recognition and the antiviral interferon response" *EMBO J*
41. Ashley, Abendroth, Mcsharry et al. (2019) "Interferonindependent upregulation of interferon-stimulated genes during human cytomegalovirus infection is dependent on IRF3 expression" *Viruses*
42. Dixit, Boulant, Zhang et al. (2010) "Peroxisomes are signaling platforms for antiviral innate immunity" *Cell*
43. Kim, Latham, Krug (2002) "Human influenza viruses activate an interferon-independent transcription of cellular antiviral genes: outcome with influenza A virus is unique" *Proc Natl Acad Sci*
44. Noyce, Collins, Mossman (2006) "Identification of a novel pathway essential for the immediate-early, interferon-independent antiviral response to enveloped virions" *J Virol*
45. Xu, Li, Zhou et al. (2017) "RIG-I is a key antiviral interferon-stimulated gene against hepatitis E virus regardless of interferon production" *Hepatology*
46. Francisco, Suthar, Gale et al. (2019) "Celltype specificity and functional redundancy of RIG-I-like receptors in innate immune sensing of Coxsackievirus B3 and encephalomyocarditis virus" *Virology (Auckl)*
47. Rubio, Xu, Remakus et al. (2013) "Crosstalk between the type 1 interferon and nuclear factor kappa B pathways confers resistance to a lethal virus infection" *Cell Host Microbe*
48. Amsden, Kourko, Roth et al. (2022) "Antiviral activities of interleukin-27: a partner for interferons?"
49. Kwock, Handfield, Suwanpradid et al. (2020) "IL-27 signaling activates skin cells to induce innate antiviral proteins and protects against Zika virus infection" *Sci Adv*
50. Wang, Xu, Brandsma et al. (2016) "Convergent transcription of interferon-stimulated genes by TNF-α and IFN-α augments antiviral activity against HCV and HEV" *Sci Rep*
51. Reus, Rex, Gammon (2022) "How to inhibit nuclear factorkappa B signaling: lessons from poxviruses" *Pathogens*
52. Cho, Shin, Kim et al. (2023) "Insights on interferon-independ ent induction of interferon-stimulated genes shaping the lung's response in early SARS-CoV-2 infection"
53. Paladino, Cummings, Noyce et al. (2006) "The IFNindependent response to virus particle entry provides a first line of antiviral defense that is independent of TLRs and retinoic acidinducible gene I" *J Immunol*
54. Liu, Jiang, Sun et al. (2023) "Interplay between TRIM7 and antiviral immunity" *Front Cell Infect Microbiol*
55. Fan, Hu, Hu et al. (2025) *Cell Signal*
56. Orchard, Sullender, Dunlap et al. (2019) "Identification of antinorovirus genes in human cells using genome-wide CRISPR activation screening" *J Virol*
57. Carthagena, Bergamaschi, Luna et al. (2009) "Human TRIM gene expression in response to interferons" *PLoS One*
58. Fan, Mar, Sari et al. (2021) "TRIM7 inhibits enterovirus replication and promotes emergence of a viral variant with increased pathogenicity" *Cell*
59. Luptak, Mallery, Jahun et al. (2022) "TRIM7 restricts coxsackievirus and norovirus infection by detecting the C-terminal glutamine generated by 3C protease processing" *Viruses*
60. Liang, Li, Liu et al. (2022) "A C-terminal glutamine recognition mechanism revealed by E3 ligase TRIM7 structures" *Nat Chem Biol*
61. Ru, Yan, Zhang et al. (2022) "C-terminal glutamine acts as a C-degron targeted by E3 ubiquitin ligase TRIM7" *Proc Natl Acad Sci*
62. Sullender, Pierce, Srinivas et al. (2022) "Selective polyprotein processing determines norovirus sensitivity to Trim7" *J Virol*
63. Fan, Mcdougal, Schoggins (2022) "Enterovirus 3C protease cleaves TRIM7 to dampen its antiviral activity" *J Virol*
64. Budicini, Rodriguez-Irizarry, Maples et al. (2024) "Murine norovirus mutants adapted to replicate in human cells reveal a postentry restriction" *J Virol*
65. Srinivas, Pierce, Olson et al. (2025) "Trim7 does not have a role in the restriction of murine norovirus infection in vivo" *J Virol*
66. Gonzalez-Orozco, Tseng, Hage et al. (2024) "TRIM7 ubiquitinates SARS-CoV-2 membrane protein to limit apoptosis and viral replication" *Nat Commun*
67. Giraldo, Xia, Aguilera-Aguirre et al. (2020) "Envelope protein ubiquitina tion drives entry and pathogenesis of Zika virus" *Nature New Biol*
68. Chabot, Durantel, Lucifora (2025) "TRIM proteins: a "swiss army knife" of antiviral immunity" *PLoS Pathog*
69. (2025) *Minireview Journal of Virology*
70. Rex, Seo, Gammon (2018) "Arbovirus infections as screening tools for the identification of viral immunomodulators and host antiviral factors" *J Vis Exp*
71. Belotserkovskaya, Oh, Bondarenko et al. (2003) "FACT facilitates transcription-dependent nucleosome alteration" *Science*
72. Orphanides, Leroy, Chang et al. (1998) "FACT, a factor that facilitates transcript elongation through nucleosomes" *Cell*
73. Winkler, Muthurajan, Hieb et al. (2011) "Histone chaperone FACT coordinates nucleosome interaction through multiple synergistic binding events" *J Biol Chem*
74. Rex, Seo, Embry et al. (2025) "Activation and evasion of the FEAR pathway by RNA viruses"
75. Rex, Seo, Chappidi et al. (2024) "FEAR antiviral response pathway is independ ent of interferons and countered by poxvirus proteins" *Nat Microbiol*
76. Garrett-Sinha (2013) "Review of Ets1 structure, function, and roles in immunity" *Cell Mol Life Sci*
77. Mouly, Chemin, Nguyen et al. (2010) "The Ets-1 transcription factor controls the development and function of natural regulatory T cells" *J Exp Med*
78. Taveirne, Wahlen, Van Loocke et al. (2020) "The transcription factor ETS1 is an important regulator of human NK cell development and terminal differentiation" *Blood*
79. Seifert, Si, Saha et al. (2019) "The ETS transcription factor ELF1 regulates a broadly antiviral program distinct from the type I interferon response" *PLoS Pathog*
80. Walsh (2024) "Primal FEAR protects against infection" *Nat Microbiol*
81. Gammon, Duraffour, Rozelle et al. (2014) "A single vertebrate DNA virus protein disarms invertebrate immunity to RNA virus infection"
82. Oda, Matoba, Irie et al. (2015) "Structural basis of the inhibition of STAT1 activity by sendai virus C protein" *J Virol*
83. Oda, Abe, Sato (1999) "ETS-1 converts endothelial cells to the angiogenic phenotype by inducing the expression of matrix metallo proteinases and integrin beta3" *J Cell Physiol*
84. Garcia, Miecznikowski, Safina et al. (2013) "Facilitates chromatin transcription complex is an "accelerator" of tumor transformation and potential marker and target of aggressive cancers" *Cell Rep*
85. Dittmer (2015) "The role of the transcription factor Ets1 in carcinoma" *Semin Cancer Biol*
86. Park, Jung, Ahn et al. (2008) "Ets-1 upregulates HER2-induced MMP-1 expression in breast cancer cells" *Biochem Biophys Res Commun*
87. Gasparian, Burkhart, Purmal et al. (2011) "Curaxins: anticancer compounds that simultaneously suppress NF-κB and activate p53 by targeting FACT" *Sci Transl Med*
88. Holbrook, Goad, Grdzelishvili (2021) "Expanding the spectrum of pancreatic cancers responsive to vesicular stomatitis virusbased oncolytic virotherapy: challenges and solutions" *Cancers (Basel)*
89. Felt, Grdzelishvili (2017) "Recent advances in vesicular stomatitis virus-based oncolytic virotherapy: a 5-year update" *J Gen Virol*
90. Arulanandam, Batenchuk, Varette et al. (2015) "Microtubule disruption synergizes with oncolytic virotherapy by inhibiting interferon translation and potentiating bystander killing" *Nat Commun*
91. Huang, Santoso, Power et al. (2015) "FACT proteins, SUPT16H and SSRP1, are transcriptional suppressors of HIV-1 and HTLV-1 that facilitate viral latency" *J Biol Chem*
92. Fox, Dembowski, Deluca (2017) "A herpesviral immediate early protein promotes transcription elongation of viral transcripts" *mBio*
93. Isa, Bensaude, Aziz et al. (2021) "HSV-1 ICP22 is a selective viral repressor of cellular rna polymerase II-mediated transcription elongation" *Vaccines (Basel)*
94. O'connor, Nukui, Gurova et al. (2016) "Inhibition of the FACT complex reduces transcription from the human cytomegalovirus major immediate early promoter in models of lytic and latent replication" *J Virol*
95. Volokh, Sivkina, Moiseenko et al. (2022) "Mechanism of curaxin-dependent nucleosome unfolding by FACT" *Front Mol Biosci*
96. Bhakat, Ray (2022) "The FAcilitates Chromatin Transcription (FACT) complex: its roles in DNA repair and implications for cancer therapy" *DNA Repair (Amst)*
97. De Duve, Wattiaux (1966) "Functions of lysosomes" *Annu Rev Physiol*
98. Liang, Wu, Li et al. (2021) "Autophagy in viral infection and pathogenesis" *Front Cell Dev Biol*
99. Chen, Tu, Ding et al. (2023) "The role of autophagy in viral infections" *J Biomed Sci*
100. Jassey, Jackson (2024) "Viruses and autophagy: bend, but don't break" *Nat Rev Microbiol*
101. Sakamoto, Nakada-Tsukui, Besteiro (2021) "The autophagy machinery in human-parasitic protists; diverse functions for universally conserved proteins" *Cells*
102. Iriondo, Etxaniz, Varela et al. (2023) "Effect of ATG12-ATG5-ATG16L1 autophagy E3-like complex on the ability of LC3/ GABARAP proteins to induce vesicle tethering and fusion" *Cell Mol Life Sci*
103. Mizushima, Yoshimori, Ohsumi (2011) "The role of Atg proteins in autophagosome formation" *Annu Rev Cell Dev Biol*
104. Fujita, Itoh, Omori et al. (2008) "The Atg16L complex specifies the site of LC3 lipidation for membrane biogenesis in autophagy" *Mol Biol Cell*
105. Velazquez, Jackson (2018) "So many roads: the multiface ted regulation of autophagy induction" *Mol Cell Biol*
106. Menon, Dhamija (2018) "Beclin 1 phosphorylation -at the center of autophagy regulation" *Front Cell Dev Biol*
107. Tanida (2011) "Autophagy basics" *Microbiol Immunol*
108. Tanida, Ueno, Kominami (2004) "LC3 conjugation system in mammalian autophagy" *Int J Biochem Cell Biol*
109. Nakatogawa (2013) "Two ubiquitin-like conjugation systems that mediate membrane formation during autophagy" *Essays Biochem*
110. Lystad, Carlsson, Simonsen (2019) "Toward the function of mammalian ATG12-ATG5-ATG16L1 complex in autophagy and related processes" *Autophagy*
111. (2019)
112. Dooley, Razi, Polson et al. (2014) "WIPI2 links LC3 conjugation with PI3P, autophagosome formation, and pathogen clearance by recruiting Atg12-5-16L1" *Mol Cell*
113. Shpilka, Weidberg, Pietrokovski et al. (2011) "Atg8: an autophagy-related ubiquitin-like protein family" *Genome Biol*
114. Mcewan, Popovic, Gubas et al. (2015) "PLEKHM1 regulates autophagosome-lysosome fusion through HOPS complex and LC3/GABARAP proteins" *Mol Cell*
115. Yu, Chen, Tooze (2018) "Autophagy pathway: cellular and molecular mechanisms" *Autophagy*
116. Shvarev, Schoppe, König et al. (2022) "Structure of the HOPS tethering complex, a lysosomal membrane fusion machinery"
117. Song, Orr, Lee et al. (2020) "HOPS recognizes each SNARE, assembling ternary trans-complexes for rapid fusion upon engagement with the 4th SNARE" *Elife*
118. Zhao, Codogno, Zhang (2021) "Machinery, regulation and pathophysiological implications of autophagosome maturation" *Nat Rev Mol Cell Biol*
119. Chen, Huang, Lin et al. (1322) "VPS34 K29/K48 branched ubiquitination governed by UBE3C and TRABID regulates autophagy, proteostasis and liver metabolism" *Nat Commun*
120. Sun, Zhang, Wong et al. (2011) "The RUN domain of rubicon is important for hVps34 binding, lipid kinase inhibition, and autophagy suppression" *J Biol Chem*
121. Thinwa, Zou, Parks et al. (2024) "CDKL5 regulates p62-mediated selective autophagy and confers protection against neurotropic viruses" *J Clin Invest*
122. Zhai, Wang, Liu et al. (2023) "Autophagy as a dualfaced host response to viral infections" *Front Cell Infect Microbiol*
123. Choi, Bowman, Jung (2018) "Autophagy during viral infection -a double-edged sword" *Nat Rev Microbiol*
124. Liu, Zhou, Hu et al. (2022) "Targeting selective autophagy as a therapeutic strategy for viral infectious diseases" *Front Microbiol*
125. Viret, Duclaux-Loras, Rozières et al. (2021) "Selective autophagy receptors in antiviral defense" *Trends Microbiol*
126. Johansen, Lamark (2020) "Selective autophagy: ATG8 family proteins, LIR motifs and cargo receptors" *J Mol Biol*
127. Kirkin, Rogov (2019) "A diversity of selective autophagy receptors determines the specificity of the autophagy pathway" *Mol Cell*
128. Orvedahl, Macpherson, Sumpter R Jr et al. (2010) "Autophagy protects against Sindbis virus infection of the central nervous system" *Cell Host Microbe*
129. Liang, Kleeman, Jiang et al. (1998) "Protection against fatal Sindbis virus encephalitis by beclin, a novel Bcl-2-interacting protein" *J Virol*
130. Wen, Li, Yin et al. (2021) "Selective autophagy receptor SQSTM1/ p62 inhibits Seneca Valley virus replication by targeting viral VP1 and VP3" *Autophagy*
131. Pino-Belmar, Aguilar, Valenzuela-Nieto et al. (2024) "An intrinsic host defense against HSV-1 relies on the activation of xenophagy with the active clearance of autophagic receptors" *Cells*
132. Judith, Mostowy, Bourai et al. (2013) "Species-specific impact of the autophagy machinery on Chikungunya virus infection" *EMBO Rep*
133. Miyakawa, Nishi, Ogawa et al. (2022) "Galectin-9 restricts hepatitis B virus replication via p62/SQSTM1mediated selective autophagy of viral core proteins" *Nat Commun*
134. Montespan, Marvin, Austin et al. (2017) "Multi-layered control of Galectin-8 mediated autophagy during adenovirus cell entry through a conserved PPxY motif in the viral capsid" *PLoS Pathog*
135. Ames, Yadavalli, Suryawanshi et al. (2021) "OPTN is a host intrinsic restriction factor against neuroinvasive HSV-1 infection" *Nat Commun*
136. Minton (2016) "Autophagy: inflammatory pathology of Fanconi anaemia" *Nat Rev Mol Cell Biol*
137. Sumpter R Jr, Sirasanagandla, Fernández et al. (2016) "Fanconi anemia proteins function in mitophagy and immunity" *Cell*
138. Orvedahl, Sumpter, Xiao et al. (2011) "Image-based genome-wide siRNA screen identifies selective autophagy factors" *Nature*
139. Kim, Kim, Sung et al. (2016) "Interferoninducible protein SCOTIN interferes with HCV replication through the autolysosomal degradation of NS5A" *Nat Commun*
140. Mandell, Jain, Arko-Mensah et al. (2014) "TRIM proteins regulate autophagy and can target autophagic substrates by direct recognition" *Dev Cell*
141. Schmid, Fischer, Engl et al. (2024) "The interplay between autophagy and cGAS-STING signaling and its implications for cancer" *Front Immunol*
142. Schmeisser, Bekisz, Zoon (2014) "New function of type I IFN: induction of autophagy" *J Interferon Cytokine Res*
143. Tallóczy, Jiang, Virgin et al. (2002) "Regulation of starvation-and virusinduced autophagy by the eIF2α kinase signaling pathway" *Proc Natl Acad Sci*
144. Orvedahl, Alexander, Tallóczy et al. (2007) "HSV-1 ICP34.5 confers neurovirulence by targeting the Beclin 1 autophagy protein" *Cell Host Microbe*
145. (2025) *Minireview Journal of Virology*
146. Wu, Dobbs, Yang et al. (2020) "Interferon-independent activities of mammalian STING mediate antiviral response and tumor immune evasion" *Immunity*
147. Yamashiro, Wilson, Morrison et al. (2020) "Interferonindependent STING signaling promotes resistance to HSV-1 in vivo" *Nat Commun*
148. Yi, Wen, Tan et al. (2019) "Impact of NF-κB pathway on the apoptosis-inflammation-autophagy crosstalk in human degenerative nucleus pulposus cells" *Aging (Milano)*
149. Zhu, Wu, Zhao et al. (2007) "Optineurin negatively regulates TNFα-induced NF-κB activation by competing with NEMO for ubiquitinated RIP" *Curr Biol*
150. Richter, Sliter, Herhaus et al. (2016) "Phosphorylation of OPTN by TBK1 enhances its binding to Ub chains and promotes selective autophagy of damaged mitochondria" *Proc Natl Acad Sci*
151. B'chir, Maurin, Carraro et al. (2013) "The eIF2α/ATF4 pathway is essential for stress-induced autophagy gene expression" *Nucleic Acids Res*
152. Bullido, Martínez-García, Tenorio et al. (2008) "Double stranded RNA activated EIF2 alpha kinase (EIF2AK2; PKR) is associated with Alzheimer's disease" *Neurobiol Aging*
153. Decout, Katz, Venkatraman et al. (2021) "The cGAS-STING pathway as a therapeutic target in inflammatory diseases" *Nat Rev Immunol*
154. Liu, Wu, Wang et al. (2019) "STING directly activates autophagy to tune the innate immune response" *Cell Death Differ*
155. Gui, Yang, Li et al. (2019) "Autophagy induction via STING trafficking is a primordial function of the cGAS pathway" *Nature*
156. Chen, Meng, Qin et al. (2016) "TRIM14 inhibits cGAS degradation mediated by selective autophagy receptor p62 to promote innate immune responses" *Mol Cell*
157. Acharya, Sayyad, Hoenigsperger et al. (2025) "TRIM23 mediates cGAS-induced autophagy in anti-HSV defense" *Nat Commun*
158. Slowicka, Vereecke, Van Loo (2016) "Cellular functions of optineurin in health and disease" *Trends Immunol*
159. Ming, Zhang, Wang et al. (2022) "Inhibition of USP14 influences alphaherpesvirus proliferation by degrading viral VP16 protein via ER stress-triggered selective autophagy" *Autophagy*
160. Rubio, Mohr (2019) "Inhibition of ULK1 and Beclin1 by an αherpesvirus Akt-like Ser/Thr kinase limits autophagy to stimulate virus replication" *Proc Natl Acad Sci*
161. Vinogradskaya, Ivanov, Kushch (2022) "Mechanisms of survival of cytomegalovirus-infected tumor cells" *Mol Biol*
162. Mouna, Hernandez, Bonte et al. (2016) "Analysis of the role of autophagy inhibition by two complementary human cytomegalovirus BECN1/ Beclin 1-binding proteins" *Autophagy*
163. Gonnella, Dimarco, Farina et al. (2020) "BFRF1 protein is involved in EBVmediated autophagy manipulation" *Microbes Infect*
164. Cirone (2018) "EBV and KSHV infection dysregulates autophagy to optimize viral replication, prevent immune recognition and promote tumorigenesis" *Viruses*
165. Lee, Li, Lee et al. (2009) "FLIP-mediated autophagy regulation in cell death control" *Nat Cell Biol*
166. Zhang, Dong, Liang et al. (2015) "G-proteincoupled receptors regulate autophagy by ZBTB16-mediated ubiquitination and proteasomal degradation of Atg14"
167. Bhatt, Damania (2012) "AKTivation of PI3K/AKT/mTOR signaling pathway by KSHV" *Front Immunol*
168. Liu, Zhuang, Chen et al. (2019) "Enterovirus 71 VP1 protein regulates viral replication in SH-SY5Y cells via the mTOR autophagy signaling pathway" *Viruses*
169. Wang, Yang, Wan et al. (2020) "Baicalin inhibits coxsackievirus B3 replication by reducing cellular lipid synthesis" *Am J Chin Med*
170. Mohamud, Shi, Qu et al. (2018) "Enteroviral infection inhibits autophagic flux via disruption of the SNARE complex to enhance viral replication" *Cell Rep*
171. Zhang, Wang, Qi et al. (2014) "The regulation of autophagy by influenza A virus" *Biomed Res Int*
172. Gannagé, Dormann, Albrecht et al. (2009) "Matrix protein 2 of influenza A virus blocks autopha gosome fusion with lysosomes" *Cell Host Microbe*
173. Liu, Fang, Hu et al. (2014) "Hepatitis B virus X protein inhibits autophagic degradation by impairing lysosomal maturation" *Autophagy*
174. Castro-Gonzalez, Shi, Colomer-Lluch et al. (2021) "HIV-1 Nef counteracts autophagy restriction by enhancing the association between BECN1 and its inhibitor BCL2 in a PRKNdependent manner" *Autophagy*
175. Blanchet, Moris, Nikolic et al. (2010) "Human immunodeficiency virus-1 inhibition of immunoamphisomes in dendritic cells impairs early innate and adaptive immune responses" *Immunity*
176. Zhou, Zhang, Yang et al. (2024) "Cleavage of SQSTM1/p62 by the Zika virus protease NS2B3 prevents autophagic degradation of viral NS3 and NS5 proteins" *Autophagy*
177. Khan, Ling, Li (2024) "Is autophagy a friend or foe in SARS-CoV-2 infection?" *Viruses*
178. Tiwari, Dang, Lin et al. (2020) "Zika virus depletes neural stem cells and evades selective autophagy by suppressing the Fanconi anemia protein FANCC" *EMBO Rep*
179. Aydin, Stephens, Chava et al. (2018) "Chaperone-mediated autophagy promotes beclin1 degradation in persistently infected hepatitis C virus cell culture" *Am J Pathol*
180. Bader, Cooney, Pellegrini et al. (2022) "Programmed cell death: the pathways to severe COVID-19?" *Biochem J*
181. Ke (2022) "Autophagy and antiviral defense" *IUBMB Life*
182. Hayn, Hirschenberger, Koepke et al. (2025) *Minireview Journal of Virology*
183. Conzelmann, Müller, Badarinarayan et al. (2021) "Systematic functional analysis of SARS-CoV-2 proteins uncovers viral innate immune antagonists and remaining vulnerabilities" *Cell Rep*
184. Miao, Zhao, Li et al. (2021) "ORF3a of the COVID-19 virus SARS-CoV-2 blocks HOPS complexmediated assembly of the SNARE complex required for autolysosome formation" *Dev Cell*
185. Qu, Wang, Zhu et al. (2021) "ORF3amediated incomplete autophagy facilitates severe acute respiratory syndrome coronavirus-2 replication" *Front Cell Dev Biol*
186. Zhang, Sun, Pei et al. (2021) "The SARS-CoV-2 protein ORF3a inhibits fusion of autophagosomes with lysosomes" *Cell Discov*
187. Koepke, Hirschenberger, Hayn et al. (2021) "Manipulation of autophagy by SARS-CoV-2 proteins" *Autophagy*
188. Shroff, Nazarko (2021) "The molecular interplay between human coronaviruses and autophagy"
189. Su, Shen, Hu et al. (2023) "SARS-CoV-2 ORF3a inhibits cGAS-STING-mediated autophagy flux and antiviral function" *J Med Virol*
190. Zhang, Chen, Li et al. (2021) "The ORF8 protein of SARS-CoV-2 mediates immune evasion through down-regulating MHC-Ι" *Proc Natl Acad Sci*
191. Li, Li, Wang et al. (2021) "SARS-CoV-2 spike promotes inflammation and apoptosis through autophagy by ROS-suppressed PI3K/AKT/mTOR signaling" *Biochim Biophys Acta*
192. Corona, Saulsbery, Velazquez et al. (2018) "Enteroviruses remodel autophagic trafficking through regulation of host SNARE proteins to promote virus replication and cell exit" *Cell Rep*
193. Gladue, Donnell, Baker-Branstetter et al. (2012) "Foot-and-mouth disease virus nonstructural protein 2C interacts with Beclin1, modulating virus replication" *J Virol*
194. Wang, Tian, Ou (2015) "HCV induces the expression of Rubicon and UVRAG to temporally regulate the maturation of autophagosomes and viral replication" *PLoS Pathog*
195. Hernaez, Cabezas, Muñoz-Moreno et al. (2013) "A179L, a new viral Bcl2 homolog targeting Beclin 1 autophagy related protein" *Curr Mol Med*
196. Choi, Wang, Yoo et al. (2023) "Autophagy enables microglia to engage amyloid plaques and prevents microglial senescence" *Nat Cell Biol*
197. Guan, Deng, Liu et al. (2024) "Corynoxine promotes TFEB/TFE3-mediated autophagy and alleviates Aβ pathology in Alzheimer's disease models" *Acta Pharmacol Sin*
198. Naia, Shimozawa, Bereczki et al. (2023) "Mitochondrial hypermetabolism precedes impaired autophagy and synaptic disorganization in App knock-in Alzheimer mouse models" *Mol Psychiatry*
200. Tang, Walter, Wohleb et al. (2024) "ATG5 (autophagy related 5) in microglia controls hippocampal neurogenesis in Alzheimer disease" *Autophagy*
201. Hegdekar, Sarkar, Bustos et al. (2023) "Inhibition of autophagy in microglia and macrophages exacerbates innate immune responses and worsens brain injury outcomes" *Autophagy*
202. White, Lattime, Guo (2021) "Autophagy regulates stress responses, metabolism, and anticancer immunity" *Trends Cancer*
203. Barthet, Brucoli, Ladds et al. (2021) "Autophagy suppresses the formation of hepatocyte-derived cancer-initiating ductular progenitor cells in the liver" *Sci Adv*
204. Ge, Guo, Xiao et al. (2024) "Qingfei Tongluo mixture attenuates bleomycin-induced pulmonary inflammation and fibrosis through mTOR-dependent autophagy in rats" *Mediators Inflamm*
205. 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*
206. Zou, Chen, Liu et al. (2023) "Autophagy promotes jasmonatemediated defense against nematodes" *Nat Commun*
207. Keating, Hertz, Wehenkel et al. (2013) "The kinase mTOR modulates the antibody response to provide cross-protective immunity to lethal infection with influenza virus" *Nat Immunol*
208. Kindrachuk, Ork, Hart et al. (2015) "Antiviral potential of ERK/MAPK and PI3K/AKT/mTOR signaling modulation for Middle East respiratory syndrome coronavirus infection as identified by temporal kinome analysis" *Antimicrob Agents Chemother*
209. Husain, Byrareddy (2020) "Rapamycin as a potential repurpose drug candidate for the treatment of COVID-19" *Chem Biol Interact*
210. Chen, Gu, Guan (2018) "Metformin might inhibit virus through increasing insulin sensitivity" *Chin Med J*
211. Sharma, Ray, Sadasivam (2020) "Metformin in COVID-19: a possible role beyond diabetes" *Diabetes Res Clin Pract*
212. Wang, Zhu, Hu et al. (2020) "The mTOR inhibitor manassantin B reveals a crucial role of mTORC2 signaling in Epstein-Barr virus reactivation" *J Biol Chem*
213. Peng, Liu, He et al. (2018) "Zika virus induces autophagy in human umbilical vein endothe lial cells"
214. Kyrmizi, Gresnigt, Akoumianaki et al. (2013) "Corticosteroids block autophagy protein recruitment in Aspergillus fumigatus phagosomes via targeting dectin-1/Syk kinase signaling" *J Immunol*
215. Auyeung, Lee, Lai et al. (2005) "The use of corticosteroid as treatment in SARS was associated with adverse outcomes: a retrospec tive cohort study" *J Infect*
216. Arabi, Mandourah, Al-Hameed et al. (2018) "Corticosteroid therapy for critically ill patients with Middle East respiratory syndrome" *Am J Respir Crit Care Med*
217. Lipworth, Kuo, Lipworth et al. (2020) "Inhaled corticosteroids and COVID-19" *Am J Respir Crit Care Med*
218. Wang, Wang, Paul et al. (2013) "Mouse embryonic stem cells are deficient in type I interferon expression in response to viral infections and double-stranded RNA" *J Biol Chem*
219. Mueller, Witteveldt, Macias (2024) "Antiviral defence mechanisms during early mammalian development" *Viruses*
220. Poirier, Buck, Chakravarty et al. (2021) "An isoform of Dicer protects mammalian stem cells against multiple RNA viruses" *Science*
221. Wang, Xie, Singh et al. (2014) "Primate-specific endogenous retrovirus-driven transcription defines naive-like stem cells" *Nature*
222. Blanco-Melo, Gifford, Bieniasz (2017) "Co-option of an endogenous retrovirus envelope for host defense in hominid ancestors" *Elife*
223. Heidmann, Heidmann, Nicolas (1988) "An indicator gene to demonstrate intracellular transposition of defective retroviruses" *Proc Natl Acad Sci*
225. Wu, Wu, Xing et al. (2021) "Endogenous reverse transcriptase and RNase H-mediated antiviral mechanism in embryonic stem cells" *Cell Res*
226. Goo, Cho (2013) "The expansion and functional diversification of the mammalian ribonuclease a superfamily epitomizes the efficiency of multigene families at generating biological novelty" *Genome Biol Evol*
227. Hartwell (1971) "Genetic control of the cell division cycle in yeast. IV. Genes controlling bud emergence and cytokinesis" *Exp Cell Res*
228. Cavini, Rosa, Castro et al. (2021) "The structural biology of septins and their filaments: an update" *Front Cell Dev Biol*
229. Mostowy, Cossart (2012) "Septins: the fourth component of the cytoskeleton" *Nat Rev Mol Cell Biol*
230. Van Ngo, Mostowy (2019) "Role of septins in microbial infection" *J Cell Sci*
231. Marquardt, Chen, Bi (2019) "Architecture, remodeling, and functions of the septin cytoskeleton" *Cytoskeleton (Hoboken)*
232. Khairat, Hatta, Abdullah et al. (2024) "Unearthing the role of septins in viral infections" *Biosci Rep*
233. Wang, Fei, Qu et al. (2018) "The role of septin 7 in physiology and pathological disease: a systematic review of current status" *J Cellular Molecular Medi*
234. Nguyen, Sawa, Okano et al. (2000) "The C. elegans septin genes, unc-59 and unc-61, are required for normal postembryonic cytokineses and morphogenesis but have no essential function in embryogenesis" *J Cell Sci*
235. Saarikangas, Barral (2011) "The emerging functions of septins in metazoans" *EMBO Rep*
236. Kinoshita (2003) "Assembly of mammalian septins" *J Biochem*
237. Krokowski, Lobato-Márquez, Chastanet et al. (2018) "Septins recognize and entrap dividing bacterial cells for delivery to lysosomes" *Cell Host Microbe*
238. Lobato-Márquez, Xu, Güler et al. (2021) "Mechanistic insight into bacterial entrapment by septin cage reconstitution" *Nat Commun*
239. Welch, Way (2013) "Arp2/3-mediated actin-based motility: a tail of pathogen abuse" *Cell Host Microbe*
240. Krokowski, Mostowy (2019) "Bacterial cell division is recognized by the septin cytoskeleton for restriction by autophagy" *Autophagy*
241. Chou, Ngo, Gershon (2012) "An overview of the vaccinia virus infectome: a survey of the proteins of the poxvirus-infected cell" *J Virol*
242. Moss (2013) "Poxvirus DNA replication" *Cold Spring Harb Perspect Biol*
243. Doceul, Hollinshead, Van Der Linden et al. (2010) "Repulsion of superinfecting virions: a mechanism for rapid virus spread" *Science*
244. Sivan, Martin, Myers et al. (2013) "Human genome-wide RNAi screen reveals a role for nuclear pore proteins in poxvirus morphogenesis" *Proc Natl Acad Sci*
245. Beard, Griffiths, Gonzalez et al. (2014) "A loss of function analysis of host factors influencing vaccinia virus replication by RNA interference" *PLoS One*
246. Pfanzelter, Mostowy, Way (2018) "Septins suppress the release of vaccinia virus from infected cells" *J Cell Biol*
247. Humphries, Dodding, Barry et al. (2012) "Clathrin potentiates vaccinia-induced actin polymerization to facilitate viral spread" *Cell Host Microbe*
248. Li, Saucedo-Cuevas, Yuan et al. (2019) "Zika virus protease cleavage of host protein septin-2 mediates mitotic defects in neural progenitors" *Neuron*
249. Hertzog, Forster, Samarajiwa (2010) "Systems biology of interferon responses" *J Interferon Cytokine Res*
250. Henzi, Lannes, Filgueira (2021) "Septins in infections: focus on viruses"
251. Kim, Seol, Song et al. (2007) "An RNA-binding protein, hnRNP A1, and a scaffold protein, septin 6, facilitate hepatitis C virus replication" *J Virol*
252. Wu, Frank, Derby et al. (2021) "HIV-1 establishes a sanctuary site in the testis by permeating the BTB through changes in cytoskeletal organization" *Endocrinology*
253. Nourazarian, Yousefi, Biray Avci et al. (2025) "The interplay between viral infection and cell death: a ping-pong effect" *Adv Virol*
254. Lee, Song, Bae et al. (2023) "Regulated cell death pathways and their roles in homeostasis, infection, inflammation, and tumorigenesis" *Exp Mol Med*
255. Vicar, Raudenska, Gumulec et al. (2020) "The quantitativephase dynamics of apoptosis and lytic cell death" *Sci Rep*
256. Kerr, Wyllie, Currie (1972) "Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics" *Br J Cancer*
257. Shen, Shao, Li (2023) "Different types of cell death and their shift in shaping disease" *Cell Death Discov*
258. Bertheloot, Latz, Franklin (2021) "Necroptosis, pyroptosis and apoptosis: an intricate game of cell death" *Cell Mol Immunol*
259. Kvansakul, Hinds (2013) "Structural biology of the Bcl-2 family and its mimicry by viral proteins" *Cell Death Dis*
260. Verburg, Lelievre, Westerveld et al. (2022) "Viral-mediated activation and inhibition of programmed cell death" *PLoS Pathog*
261. Stewart, Wasilenko, Barry (2005) "Vaccinia virus F1L protein is a tail-anchored protein that functions at the mitochondria to inhibit apoptosis" *J Virol*
262. Pelin, Foloppe, Petryk et al. (2019) "Deletion of apoptosis inhibitor F1L in vaccinia virus increases safety and oncolysis for cancer therapy" *Mol Ther Oncolytics*
263. Wasilenko, Banadyga, Bond et al. (2005) "The vaccinia virus F1L protein interacts with the proapoptotic protein Bak and inhibits Bak activation" *J Virol*
264. Kvansakul, Yang, Fairlie et al. (2008) "Vaccinia virus anti-apoptotic F1L is a novel Bcl-2-like domain-swapped dimer that binds a highly selective subset of BH3-containing death ligands" *Cell Death Differ*
265. Cheltsov, Aoyagi, Aleshin et al. (2010) "Vaccinia virus virulence factor N1L is a novel promising target for antiviral therapeutic intervention" *J Med Chem*
266. Aoyagi, Zhai, Aleshin et al. (2007) "Vaccinia virus N1L protein resembles a B cell lymphoma-2 (Bcl-2) family protein" *Protein Sci*
267. Kettle, Blake, Law et al. (1995) "Vaccinia virus serpins B13R (SPI-2) and B22R (SPI-1) encode M(r) 38.5 and 40K, intracellular polypeptides that do not affect virus virulence in a murine intranasal model" *Virology (Auckl)*
268. Shisler, Isaacs, Moss (1999) "Vaccinia virus serpin-1 deletion mutant exhibits a host range defect characterized by low levels of intermediate and late mRNAs" *Virology (Auckl)*
269. Kettle, Alcamí, Khanna et al. (1997) "Vaccinia virus serpin B13R (SPI-2) inhibits interleukin-1beta-converting enzyme and protects virus-infected cells from TNF-and Fas-mediated apoptosis, but does not prevent IL-1beta-induced fever" *J Gen Virol*
270. Dobbelstein, Shenk (1996) "Protection against apoptosis by the vaccinia virus SPI-2 (B13R) gene product" *J Virol*
271. Ryerson, Richards, Kvansakul et al. (2017) "Vaccinia virus encodes a novel inhibitor of apoptosis that associates with the apoptosome" *J Virol*
272. Marshall, Puthalakath, Caria et al. (2015) "Variola virus F1L is a Bcl-2-like protein that unlike its vaccinia virus counterpart inhibits apoptosis independent of Bim" *Cell Death Dis*
273. Ray, Black, Kronheim et al. (1992) "Viral inhibition of inflammation: Cowpox virus encodes an inhibitor of the interleukin-1β converting enzyme" *Cell*
274. Zhou, Snipas, Orth et al. (1997) "Target protease specificity of the viral serpin CrmA. Analysis of five caspases" *J Biol Chem*
275. Wang, Barrett, Nazarian et al. (2004) "Myxoma virus M11L prevents apoptosis through constitutive interaction with Bak" *J Virol*
276. Su, Wang, Barrett et al. (2006) "Myxoma virus M11L blocks apoptosis through inhibition of conformational activation of Bax at the mitochondria" *J Virol*
277. Douglas, Corbett, Berger et al. (2007) "Structure of M11L: a myxoma virus structural homolog of the apoptosis inhibitor"
278. Kvansakul, Van Delft, Lee et al. (2007) "A structural viral mimic of prosurvival Bcl-2: a pivotal role for sequestering proapoptotic Bax and Bak" *Mol Cell*
279. Bertin, Armstrong, Martin et al. (1997) "Death effector domain-containing herpesvirus and poxvirus proteins inhibit both Fas-and TNFR1-induced apoptosis" *Proc Natl Acad Sci*
280. Mehta, Taylor, Quilty et al. (2015) "Ectromelia virus encodes an anti-apoptotic protein that regulates cell death" *Virology (Auckl)*
281. Suraweera, Burton, Hinds et al. (2020) "Crystal structures of the sheeppox virus encoded inhibitor of apoptosis SPPV14 bound to the proapoptotic BH3 peptides Hrk and Bax" *FEBS Lett*
282. Okamoto, Campbell, Mehta et al. (2012) "Sheeppox virus SPPV14 encodes a Bcl-2-like cell death inhibitor that counters a distinct set of mammalian proapoptotic proteins" *J Virol*
283. Banadyga, Lam, Okamoto et al. (2011) "Deerpox virus encodes an inhibitor of apoptosis that regulates Bak and Bax" *J Virol*
284. Westphal, Ledgerwood, Hibma et al. (2007) "A novel Bcl-2-like inhibitor of apoptosis is encoded by the parapoxvirus ORF virus" *J Virol*
285. Suraweera, Hinds, Kvansakul (2020) "Crystal structures of ORFV125 provide insight into orf virus-mediated inhibition of apoptosis" *Biochem J*
286. Suraweera, Hinds, Kvansakul (2021) "Structural investigation of Orf virus Bcl-2 homolog ORFV125 interactions with BH3-motifs from BH3-only proteins Puma and Hrk" *Viruses*
287. Suraweera, Anasir, Chugh et al. (2020) "Structural insight into tanapoxvirus-mediated inhibition of apoptosis" *FEBS J*
288. Wang, Garvey, Cohen (1999) "The murine gammaherpesvirus-68 M11 protein inhibits Fas-and TNF-induced apoptosis" *J Gen Virol*
289. Chaudhry, Kasmapour, Plaza-Sirvent et al. (2017) "UL36 rescues apoptosis inhibition and in vivo replication of a chimeric MCMV lacking the M36 gene" *Front Cell Infect Microbiol*
290. Kvansakul, Wei, Fletcher et al. (2010) "Structural basis for apoptosis inhibition by Epstein-Barr virus BHRF1" *PLoS Pathog*
291. Vilmen, Glon, Siracusano et al. (2021) "BHRF1, a BCL2 viral homolog, disturbs mitochondrial dynamics and stimulates mitophagy to dampen type I IFN induction" *Autophagy*
292. Marshall, Yim, Gustafson et al. (1999) "Epstein-Barr virus encodes a novel homolog of the bcl-2 oncogene that inhibits apoptosis and associates with Bax and Bak" *J Virol*
293. Gallo, Lampe, Günther et al. (2017) "The viral Bcl-2 homologs of Kaposi's sarcoma-associated herpesvirus and rhesus rhadinovirus share an essential role for viral replication" *J Virol*
294. Sarid, Sato, Bohenzky et al. (1997) "Kaposi's sarcomaassociated herpesvirus encodes a functional bcl-2 homologue" *Nat Med*
295. Derfuss, Fickenscher, Kraft et al. (1998) "Antiapoptotic activity of the herpesvirus saimiri-encoded Bcl-2 homolog: stabilization of mitochondria and Minireview Journal of Virology November"
296. "inhibition of caspase-3-like activity" *J Virol*
297. Nava, Cheng, Veliuona et al. (1997) "Herpesvirus saimiri encodes a functional homolog of the human bcl-2 oncogene" *J Virol*
298. Thome, Schneider, Hofmann et al. (1997) "Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors" *Nature*
299. Hagglund, Munger, Poon et al. (2002) "U(S)3 protein kinase of herpes simplex virus 1 blocks caspase 3 activation induced by the products of U(S)1.5 and U(L)13 genes and modulates expression of transduced U(S)1.5 open reading frame in a cell type-specific manner" *J Virol*
300. Shi, Liu, Zhang et al. (2021) "The A179L gene of African swine fever virus suppresses virus-induced apoptosis but enhances necroptosis" *Viruses*
301. Reis, Rathakrishnan, Goulding et al. (2023) "Deletion of the gene for the African swine fever virus BCL-2 family member A179L increases virus uptake and apoptosis but decreases virus spread in macrophages and reduces virulence in pigs" *J Virol*
302. Brun, Rivas, Esteban et al. (1996) "African swine fever virus gene A179L, a viral homologue of bcl-2, protects cells from programmed cell death" *Virology (Auckl)*
303. Chiou, Tseng, Rao et al. (1994) "Functional complementation of the adenovirus E1B 19-kilodalton protein with Bcl-2 in the inhibition of apoptosis in infected cells" *J Virol*
304. Tarakanova, Wold (2009) "Adenovirus E1A and E1B-19K proteins protect human hepatoma cells from transforming growth factor beta1induced apoptosis" *Virus Res*
305. White (2001) "Regulation of the cell cycle and apoptosis by the oncogenes of adenovirus" *Oncogene*
306. Subramanian, Vijayalingam, Chinnadurai (2006) "Genetic identification of adenovirus type 5 genes that influence viral spread" *J Virol*
307. Salako, Carter, Kass (2006) "Coxsackievirus protein 2BC blocks host cell apoptosis by inhibiting caspase-3" *J Biol Chem*
308. Chau, Yuan, Zhang et al. (2007) "Coxsackievirus B3 proteases 2A and 3C induce apoptotic cell death through mitochondrial injury and cleavage of eIF4GI but not DAP5/p97/NAT1" *Apoptosis*
309. Ilkow, Goping, Hobman (2011) "The Rubella virus capsid is an anti-apoptotic protein that attenuates the pore-forming ability of Bax" *PLoS Pathog*
310. Han, Wang, Yang et al. (2021) "Zika virus infection induced apoptosis by modulating the recruitment and activation of pro-apoptotic protein Bax" *J Virol*
311. Li, Huang, Liao et al. (2012) "Dengue virus utilizes calcium modulating cyclophilin-binding ligand to subvert apoptosis" *Biochem Biophys Res Commun*
312. Prikhod'ko Gg, Ea, Pletnev et al. (2002) "Langat flavivirus protease NS3 binds caspase-8 and induces apoptosis" *J Virol*
313. Danthi, Kobayashi, Holm et al. (2008) "Reovirus apoptosis and virulence are regulated by host cell membrane penetration efficiency" *J Virol*
314. Wisniewski, Werner, Hom et al. (2011) "Reovirus infection or ectopic expression of outer capsid protein micro1 induces apoptosis independently of the cellular proapoptotic proteins Bax and Bak" *J Virol*
315. Gerlic, Faustin, Postigo et al. (2013) "Vaccinia virus F1L protein promotes virulence by inhibiting inflammasome activation" *Proc Natl Acad Sci*
316. Veyer, De Motes, Sumner et al. (2014) "Analysis of the anti-apoptotic activity of four vaccinia virus proteins demonstrates that B13 is the most potent inhibitor in isolation and during viral infection" *J Gen Virol*
317. Boys, Johnson, Quinlan et al. (2023) "Structural homology screens reveal host-derived poxvirus protein families impacting inflammasome activity" *Cell Rep*
318. Garg, Jackson, Rahman et al. (2019) "Myxoma virus M013 protein antagonizes NF-κB and inflammasome pathways via distinct structural motifs" *J Biol Chem*
319. Dorfleutner, Talbott, Bryan et al. (2007) "A Shope Fibroma virus PYRINonly protein modulates the host immune response" *Virus Genes*
320. Gregory, Davis, West et al. (2011) "Discovery of a viral NLR homolog that inhibits the inflammasome" *Science*
321. Parameswaran, Payne, Powers et al. (2024) "A viral E3 ubiquitin ligase produced by herpes simplex virus 1 inhibits the NLRP1 inflammasome" *J Exp Med*
322. Deng, Ostermann, Brune (2024) "A cytomegalovirus inflammasome inhibitor reduces proinflammatory cytokine release and pyroptosis" *Nat Commun*
323. Ma, Zhu, Zhao et al. (2021) "SARS-CoV-2 nucleocapsid suppresses host pyroptosis by blocking Gasdermin D cleavage" *EMBO J*
324. Shi, Lv, Wang et al. (2022) "Coronaviruses Nsp5 antagonizes porcine gasdermin D-mediated pyroptosis by cleaving pore-forming p30"
325. Planès, Pinilla, Santoni et al. (2022) "Human NLRP1 is a sensor of pathogenic coronavirus 3CL proteases in lung epithelial cells" *Mol Cell*
326. Zhang, Ji, Huang et al. (2025) "The SARS-CoV-2 3CL protease inhibits pyroptosis through the cleavage of gasdermin D" *Virol Sin*
327. Lei, Zhang, Xiao et al. (2017) "Enterovirus 71 inhibits pyroptosis through cleavage of gasdermin D" *J Virol*
328. Wen, Li, Wang et al. (2021) "Seneca Valley virus 3C protease induces pyroptosis by directly cleaving porcine gasdermin D" *J Immunol*
329. Song, Wu, Xu et al. (2020) "HPV E7 inhibits cell pyroptosis by promoting TRIM21-mediated degradation and ubiquitination of the IFI16 inflammasome" *Int J Biol Sci*
330. Komune, Ichinohe, Ito et al. (2011) "Measles virus V protein inhibits NLRP3 inflammasome-mediated interleukin-1β secretion" *J Virol*
331. Komatsu, Tanaka, Kitagawa et al. (2018) "Sendai virus V protein inhibits the secretion of interleukin-1β by preventing NLRP3 inflammasome assembly" *J Virol*
332. Moriyama, Chen, Kawaguchi et al. (2016) "The RNA-and TRIM25-binding domains of influenza virus NS1 protein are essential for suppression of NLRP3 inflammasome-mediated interleukin-1β secretion" *J Virol*
333. (2025) *Minireview Journal of Virology*
334. Park, Liu, Raman et al. (2018) "NS1 protein of 2009 pandemic influenza A virus inhibits porcine NLRP3 inflammasome-mediated interleukin-1 beta production by suppressing ASC ubiquitination" *J Virol*
335. Chung, Kang, Yoon et al. (2015) "Influenza A virus NS1 protein inhibits the NLRP3 inflammasome"
336. Gaba, Xu, Lu et al. (2019) "The NS1 protein of influenza A virus participates in necroptosis by interacting with MLKL and increasing its oligomerization and membrane translocation" *J Virol*
337. Koehler, Cotsmire, Zhang et al. (2021) "Vaccinia virus E3 prevents sensing of Z-RNA to block ZBP1-dependent necroptosis" *Cell Host Microbe*
338. Liu, Nailwal, Rector et al. (2021) "A class of viral inducer of degradation of the necroptosis adaptor RIPK3 regulates virus-induced inflammation" *Immunity*
339. Petrie, Sandow, Lehmann et al. (2019) "Viral MLKL homologs subvert necroptotic cell death by sequestering cellular RIPK3" *Cell Rep*
340. Muscolino, Schmitz, Loroch et al. (2020) "Herpesviruses induce aggregation and selective autophagy of host signalling proteins NEMO and RIPK1 as an immuneevasion mechanism" *Nat Microbiol*
341. Upton, Kaiser, Mocarski (2010) "Virus inhibition of RIP3dependent necrosis" *Cell Host Microbe*
342. Muscolino, Castiglioni, Brixel et al. (2021) "Speciesspecific inhibition of necroptosis by HCMV UL36" *Viruses*
343. Fletcher-Etherington, Nobre, Nightingale et al. (2020) "Human cytomegalovirus protein pUL36: a dual cell death pathway inhibitor" *Proc Natl Acad Sci*
344. Heusel, Rapp, Stamminger et al. (2024) "IE1 of human cytomegalovirus inhibits necroptotic cell death via direct and indirect modulation of the necrosome complex" *Viruses*
345. Guo, Omoto, Harris et al. (2015) "Herpes simplex virus suppresses necroptosis in human cells" *Cell Host Microbe*
346. Wang, Li, Liu et al. (2014) "Direct activation of RIP3/MLKL-dependent necrosis by herpes simplex virus 1 (HSV-1) protein ICP6 triggers host antiviral defense" *Proc Natl Acad Sci*
347. Shi, Zhou, Shang et al. (2019) "EBV(LMP1)-induced metabolic reprogramming inhibits necroptosis through the hypermethylation of the RIP3 promoter" *Theranostics*
348. Ma, Tummers, Van Esch et al. (2016) "Human papillomavirus downregulates the expression of IFITM1 and RIPK3 to escape from IFNγ -and TNFα -mediated antiproliferative effects and necroptosis" *Front Immunol*
349. Ma, Liu, Cao et al. (2024) "Suppression of ZBP1-mediated NLRP3 inflammasome by the tegument protein VP22 facilitates pseudorabies virus infection"
350. Schock, Chandra, Sun et al. (2017) "Induction of necroptotic cell death by viral activation of the RIG-I or STING pathway" *Cell Death Differ*
351. Mukhopadhyay, Patra, Chandra et al. (2022) "Rotavirus activates MLKL-mediated host cellular necroptosis concomitantly with apoptosis to facilitate dissemination of viral progeny" *Mol Microbiol*
352. Thomson (2001) "Viruses and apoptosis" *Int J Exp Pathol*
353. Connolly, Ho (2017) "Viral hijacking of host caspases: an emerging category of pathogen-host interactions" *Cell Death Differ*
354. Wyżewski, Gregorczyk-Zboroch, Mielcarska et al. (2025) "Bid protein: a participant in the apoptotic network with roles in viral infections" *Int J Mol Sci*
355. Mustafa, Ahmad, Tantry et al. (1838) "Apoptosis: a comprehen sive overview of signaling pathways, morphological changes, and physiological significance and therapeutic implications" *Cells*
356. Neumann, Maadidi, Faletti et al. (2015) "How do viruses control mitochondriamediated apoptosis?" *Virus Res*
357. Gavathiotis, Reyna, Davis et al. (2010) "BH3triggered structural reorganization drives the activation of proapop totic BAX" *Mol Cell*
358. Pang, Dai, Smith et al. (2012) "Bak conformational changes induced by ligand binding: insight into BH3 domain binding and Bak homo-oligomerization"
359. Zhang, Zhu, Lapolla et al. (2010) "Bax forms an oligomer via separate, yet interdependent, surfaces" *J Biol Chem*
360. Dewson, Ma, Frederick et al. (2012) "Bax dimerizes via a symmetric BH3:groove interface during apoptosis" *Cell Death Differ*
361. Acehan, Jiang, Morgan et al. (2002) "Three-dimensional structure of the apoptosome: implications for assembly, procaspase-9 binding, and activation"
362. Kim, Du, Fang et al. (2005) "Formation of apoptosome is initiated by cytochrome c-induced dATP hydrolysis and subsequent nucleotide exchange on Apaf-1" *Proc Natl Acad Sci*
363. Qin, Srinivasula, Wu et al. (1999) "Structural basis of procaspase-9 recruitment by the apoptotic protease-activating factor 1" *Nature*
364. Brentnall, Rodriguez-Menocal, Guevara et al. (2013) "Caspase-9, caspase-3 and caspase-7 have distinct roles during intrinsic apoptosis" *BMC Cell Biol*
365. Mccomb, Chan, Guinot et al. (2019) "Efficient apoptosis requires feedback amplification of upstream apoptotic signals by effector caspase-3 or -7" *Sci Adv*
366. Banner, 'arcy, Janes et al. (1993) "Crystal structure of the soluble human 55 kd TNF receptor-human TNF beta complex: implications for TNF receptor activation" *Cell*
367. Kischkel, Hellbardt, Behrmann et al. (1995) "Cytotoxicity-dependent APO-1 (Fas/CD95)associated proteins form a death-inducing signaling complex (DISC) with the receptor" *EMBO J*
368. Walczak (2013) "Death receptor-ligand systems in cancer, cell death, and inflammation" *Cold Spring Harb Perspect Biol*
369. (2025) *Minireview Journal of Virology*
370. Lin, Tann, Goh et al. (2015) "Structural basis of death domain signaling in the p75 neurotrophin receptor"
371. Green (2022) "The death receptor pathway of apoptosis" *Cold Spring Harb Perspect Biol*
372. Schug, Gonzalvez, Houtkooper et al. (2011) "BID is cleaved by caspase-8 within a native complex on the mitochondrial membrane" *Cell Death Differ*
373. Dickens, Boyd, Jukes-Jones et al. (2012) "A death effector domain chain DISC model reveals a crucial role for caspase-8 chain assembly in mediating apoptotic cell death" *Mol Cell*
374. Slee, Adrain, Martin (2001) "Executioner caspase-3, -6, and -7 perform distinct, non-redundant roles during the demolition phase of apoptosis" *J Biol Chem*
375. Nagata, Suzuki, Segawa et al. (2016) "Exposure of phosphatidyl serine on the cell surface" *Cell Death Differ*
376. Arandjelovic, Ravichandran (2015) "Phagocytosis of apoptotic cells in homeostasis" *Nat Immunol*
377. Suraweera, Espinoza, Hinds et al. (2024) "Mastering death: the roles of viral Bcl-2 in dsDNA viruses" *Viruses*
378. Zhai, Yu, Welsh et al. (2010) "Vaccinia virus protein F1L is a caspase-9 inhibitor" *J Biol Chem*
379. Yu, Zhai, Gerlic et al. (2011) "Structural determinants of caspase-9 inhibition by the vaccinia virus protein, F1L" *J Biol Chem*
381. Taylor, Quilty, Banadyga et al. (2006) "The vaccinia virus protein F1L interacts with Bim and inhibits activation of the proapoptotic protein Bax" *J Biol Chem*
382. Samuel, Morrey, Diamond (2007) "Caspase 3-dependent cell death of neurons contributes to the pathogenesis of West Nile virus encephalitis" *J Virol*
383. Ramanathan, Chambers, Pankhong et al. (2006) "Host cell killing by the West Nile Virus NS2B-NS3 proteolytic complex: NS3 alone is sufficient to recruit caspase-8-based apoptotic pathway" *Virology (Auckl)*
384. Cookson, Brennan (2001) "Pro-inflammatory programmed cell death" *Trends Microbiol*
385. Bergsbaken, Fink, Cookson (2009) "Pyroptosis: host cell death and inflammation" *Nat Rev Microbiol*
386. Kovacs, Miao (2017) "Gasdermins: effectors of pyroptosis" *Trends Cell Biol*
387. (2015) "Toll-like receptors: activation, signalling and transcriptional modulation" *Cytokine*
388. Rathinam, Jiang, Waggoner et al. (2010) "The AIM2 inflammasome is essential for host defense against cytosolic bacteria and DNA viruses" *Nat Immunol*
389. Onomoto, Onoguchi, Yoneyama (2021) "Regulation of RIG-I-like receptor-mediated signaling: interaction between host and viral factors" *Cell Mol Immunol*
390. Platnich, Muruve (2019) "NOD-like receptors and inflammasomes: a review of their canonical and non-canonical signaling pathways" *Arch Biochem Biophys*
391. Allen, Scull, Moore et al. (2009) "The NLRP3 inflammasome mediates in vivo innate immunity to influenza A virus through recognition of viral RNA" *Immunity*
392. Kanneganti, Body-Malapel, Amer et al. (2006) "Critical role for Cryopyrin/Nalp3 in activation of caspase-1 in response to viral infection and double-stranded RNA" *J Biol Chem*
393. Bauernfried, Scherr, Pichlmair et al. (2021) "Human NLRP1 is a sensor for double-stranded RNA" *Science*
394. Kang, Zeng, Zhu et al. (2018) "Lipid peroxidation drives gasdermin D-mediated pyroptosis in lethal polymicrobial sepsis" *Cell Host Microbe*
395. Martinon, Pétrilli, Mayor et al. (2006) "Goutassociated uric acid crystals activate the NALP3 inflammasome" *Nature*
396. Devant, Boršić, Ngwa et al. (2023) "Gasdermin D poreforming activity is redox-sensitive" *Cell Rep*
397. (1016)
398. Mariathasan, Weiss, Newton et al. (2006) "Cryopyrin activates the inflammasome in response to toxins and ATP" *Nature*
399. Shi, Chen, Wu et al. (2023) "Inflammasome activation by viral infection: mechanisms of activation and regulation" *Front Microbiol*
400. Pan, Cai, Huang et al. (2022) "Pyroptosis in development, inflammation and disease"
401. Ernandes, Kagan (2021) "Interferon-independent restriction of RNA virus entry and replication by a class of damage-associated molecular patterns" *mBio*
402. Dai, Liu, Chen et al. (2023) "Gasdermin Dmediated pyroptosis: mechanisms, diseases, and inhibitors" *Front Immunol*
403. Liu, Zhang, Ruan et al. (2016) "Inflammasome-activated gasdermin D causes pyroptosis by forming membrane pores" *Nature*
404. Liu, Wang, Yang et al. (2020) "Caspase-1 engages full-length gasdermin D through two distinct interfaces that mediate caspase recruitment and substrate cleavage" *Immunity*
405. Rühl, Broz (2015) "Caspase-11 activates a canonical NLRP3 inflammasome by promoting K + efflux" *Eur J Immunol*
406. De Vasconcelos, Van Opdenbosch, Van Gorp et al. (2019) "Single-cell analysis of pyroptosis dynamics reveals conserved GSDMD-mediated subcellular events that precede plasma membrane rupture" *Cell Death Differ*
407. Shi, Zhao, Wang et al. (2015) "Cleavage of GSDMD by inflammatory caspases determines pyroptotic cell death" *Nature*
408. He, Wan, Hu et al. (2015) "Gasdermin D is an executor of pyroptosis and required for interleukin-1β secretion" *Cell Res*
409. Exconde, Hernandez-Chavez, Bourne et al. (2023) "The tetrapeptide sequence of IL-18 and IL-1β regulates their recruitment and activation by inflammatory caspases" *Cell Rep*
410. Kayagaki, Stowe, Lee et al. (2015) "Caspase-11 cleaves gasdermin D for non-canonical inflammasome signalling"
411. Xia, Zhang, Magupalli et al. (2021) "Gasdermin D pore structure reveals preferential release of mature interleukin-1" *Nature*
412. Bloomer, Kitevska-Ilioski, Pantaki-Eimany et al. (2019) "CrmA orthologs from diverse poxviruses potently inhibit caspases-1 and -8, yet cleavage site mutagenesis frequently produces caspase-1-specific variants" *Biochem J*
413. Kumar, Singh, Anvikar et al. (2024) "Monkeypox virus: insights into pathogenesis and laboratory testing methods"
414. Johnson, Chikoti, Chandran (2013) "Herpes simplex virus 1 infection induces activation and subsequent inhibition of the IFI16 and NLRP3 inflammasomes" *J Virol*
415. Grin, Baid, De Jesus et al. (2024) "SARS-CoV-2 3CL pro (main protease) regulates caspase activation of gasdermin-D/E pores leading to secretion and extracellular activity of 3CL pro" *Cell Rep*
416. Moustaqil, Ollivier, Chiu et al. (2021) "SARS-CoV-2 proteases PLpro and 3CLpro cleave IRF3 and critical modulators of inflammatory pathways (NLRP12 and TAB1): implications for disease presentation across species" *Emerg Microbes Infect*
417. Ambrożek-Latecka, Kozlowski, Hoser et al. (2024) "SARS-CoV-2 and its ORF3a, E and M viroporins activate inflammasome in human macrophages and induce of IL-1α in pulmonary epithelial and endothelial cells" *Cell Death Discov*
418. Hu, Zhao, Lv et al. (2024) "NLRP3-dependent pyroptosis exacerbates coxsackievirus A16 and coxsackievirus A10-induced inflammatory response and viral replication in SH-SY5Y cells" *Virus Res*
419. Wang, Yu, Zhuang et al. (2022) "Cell pyroptosis in picornavirus and its potential for treating viral infection" *J Med Virol*
420. Grootjans, Berghe, Vandenabeele (2017) "Initiation and execution mechanisms of necroptosis: an overview" *Cell Death Differ*
421. Han, Zhong, Zhang (2011) "Programmed necrosis: backup to and competitor with apoptosis in the immune system" *Nat Immunol*
422. Palmer, Chappidi, Pinkham et al. (2021) "Evolutionary profile for (Host and Viral) MLKL indicates its activities as a battlefront for extensive counteradaptation" *Mol Biol Evol*
423. Koehler, Titus, Lawson (2024) "Cell-type dependence of necropto sis pathways triggered by viral infection" *FEBS J*
424. Morgan, Kim (2022) "Roles of RIPK3 in necroptosis, cell signaling, and disease" *Exp Mol Med*
425. Cho, Challa, Moquin et al. (2009) "Phosphorylation-driven assembly of the RIP1-RIP3 complex regulates programmed necrosis and virus-induced inflammation" *Cell*
426. Kawahara, Ohsawa, Matsumura et al. (1998) "Caspase-independent cell killing by Fas-associated protein with death domain" *J Cell Biol*
427. Chen, Zhu, Zhong et al. (2022) "Mosaic composition of RIP1-RIP3 signalling hub and its role in regulating cell death" *Nat Cell Biol*
428. Haas, Emmerich, Gerlach et al. (2009) "Recruitment of the linear ubiquitin chain assembly complex stabilizes the TNF-R1 signaling complex and is required for TNF-mediated gene induction" *Mol Cell*
429. Li, Zhang, Huang et al. (2020) "Ubiquitination of RIPK1 regulates its activation mediated by TNFR1 and TLRs signaling in distinct manners" *Nat Commun*
430. Annibaldi, John, Berghe et al. (2018) "Ubiquitinmediated regulation of RIPK1 kinase activity independent of IKK and MK2" *Mol Cell*
431. Najjar, Saleh, Zelic et al. (2016) "RIPK1 and RIPK3 kinases promote celldeath-independent inflammation by toll-like receptor 4" *Immunity*
432. Du, Xiang, Liu et al. (2021) "RIPK1 dephosphorylation and kinase activation by PPP1R3G/PP1γ promote apoptosis and necroptosis" *Nat Commun*
433. Takaoka, Wang, Choi et al. (2007) "DAI (DLM-1/ZBP1) is a cytosolic DNA sensor and an activator of innate immune response" *Nature*
434. Wang, Choi, Ban et al. (2008) "Regulation of innate immune responses by DAI (DLM-1/ZBP1) and other DNA-sensing molecules" *Proc Natl Acad Sci*
435. Basavaraju, Mishra, Jindal et al. (2022) "Emerging role of ZBP1 in Z-RNA sensing, influenza virus-induced cell death, and pulmonary inflammation" *mBio*
436. Su, Quade, Wang et al. (2014) "A plug release mechanism for membrane permeation by MLKL" *Structure*
437. Chen, Li, Ren et al. (2014) "Translocation of mixed lineage kinase domain-like protein to plasma membrane leads to necrotic cell death" *Cell Res*
438. Ousingsawat, Cabrita, Wanitchakool et al. (2017) "Ca 2+ signals, cell membrane disintegration, and activation of TMEM16F during necroptosis" *Cell Mol Life Sci*
439. Taslimi, Fields, Dahl et al. (2022) "Spatiotemporal control of necroptotic cell death and plasma membrane recruitment using engineered MLKL domains" *Cell Death Discov*
440. Wang, Zhang, Orchard et al. (2023) "Norovirus MLKL-like protein initiates cell death to induce viral egress" *Nature*
441. Sun, Ji, Liu et al. (2024) "Influenza virus infection activates TAK1 to suppress RIPK3-independent apoptosis and RIPK1-dependent necroptosis" *Cell Commun Signal*
442. Boyd, Jordan, Balachandran (2025) "ZBP1-driven cell death in severe influenza" *Trends Microbiol*
443. Gautam, Boyd, Nikhar et al. (2024) "Necroptosis blockade prevents lung injury in severe influenza" *Nature*
444. Brault, Olsen, Martinez et al. (2018) "Intracellular nucleic acid sensing triggers necroptosis through synergistic type I IFN and TNF signaling" *J Immunol*
445. Malireddi, Kanneganti (2013) "Role of type I interferons in inflammasome activation, cell death, and disease during microbial infection" *Front Cell Infect Microbiol*
446. Stolzer, Ruder, Neurath et al. (2021) "Interferons at the crossroad of cell death pathways during gastrointestinal inflammation Minireview Journal of Virology November"
447. *Int J Med Microbiol*
448. Isaacs, Lindenmann (1957) "Virus interference. I. The interferon" *Proc R Soc Lond B Biol Sci*
449. Palmer (2022) "Innate metabolic responses against viral infections" *Nat Metab*
450. Gough, Messina, Clarke et al. (2012) "Constitutive type I interferon modulates homeostatic balance through tonic signaling" *Immunity*
451. Marié, Brambilla, Azzouz et al. (2021) "Tonic interferon restricts pathogenic IL-17-driven inflammatory disease via balancing the microbiome"
452. Cao (2024) "In sickness and in health-Type I interferon and the brain" *Front Aging Neurosci*
453. Gönczi, Ráduly, Szabó et al. (2022) "Septin7 is indispensable for proper skeletal muscle architecture and function" *Elife* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12657343&blobtype=pdf | # Negative feedback regulation of CHK2 in VIN18 raft cultures naturally infected with HPV18
Ming Wu, Jun Liu, Hui Li, Rui Yang
## Abstract
The life cycle of human papillomavirus (HPV) is intricate, and a lack of appropriate in vitro models for natural HPV infection has led to a dearth of effective treatments for infection and related tumors. The HPV life cycle is strictly dependent on the differentiation of epithelial cells. Therefore, in our previous study, we used conditional reprogramming(CR) technology to establish human vaginal intraepithelial neoplasia cell infected with HPV18 naturally (VIN18) and verified that HPV18 completes its viral life cycle in these cells. This article refers to it as VIN18. In this study, we utilize VIN18 to establish a 3D differentiation model that facilitates a time gradient of infection within raft-like organotypic cultures.Our findings clarify previous understanding of the interaction between HPV18 and the host during the viral physiological cycle. We observed that under undifferentiated conditions, ATM-CHK2 is not essential for the genomic stability of HPV18. However, the differentiation environment primarily activates HPV18 amplification via ATM/CHK2 signaling. CHK2 exerts negative feedback regulation on the activity of upstream ATM. Moreover, upregulation of p53 and p21 leads to a reduction in cyclin D1. Consequently, increased HPV18 E7 expression induces the re-entry of VIN18 cells into S phase, resulting in elevated expression of cyclin A2 and cyclin B1, which causes the cell cycle arrest in S phase/G2 phase, thereby supporting viral genome amplification. This study provides a valuable new model for HPV biology research and offers insights into the regulation of the HPV life cycle through the differentiation process.
## 1. Introduction
The life cycle of human papillomavirus (HPV) is strictly dependent on the differentiation of epithelial cells (Doorbar et al., 2012). Therefore, establishing a differentiation model is essential for studying HPV. Raft cultures, also known as air-liquid interface cultures (ALI), are three-dimensional (3D) cultures that enable epithelial cells to differentiate into polarized stratified structures, thus simulating the physiological structure and function of epithelial tissues (Shamir and Ewald 2014). This in vitro model circumvents species differences associated with animal models (Habanjar et al. 2021) and can be developed to mimic the epithelial differentiation microenvironment for HPV genome amplification (Egawa et al. 2017). Researchers have used immortalized male foreskin keratinocytes and cervical epithelial cells to establish raft-like organotypic cultures in vitro to study the replication and proliferation cycle of HPV (Banerjee et al. 2018). However, the currently available raft cultures do not support natural HPV infection. HPV DNA-transduced human foreskin keratinocytes (HFKs) cannot effectively initiate proliferative HPV infection, and the genetic background of the epithelial cells undergoes significant alterations (Chapman et al. 2010). Due to the lack of an in vitro system that supports the complete life cycle of HPV, the mechanisms of the HPV infection and HPV-induced tumorigenesis remain largely uncharacterized.
In the life cycle of HPV, the levels of the viral E1^E4 protein have been shown to increase as the initiation of genome amplification occurs, leading to the accumulation of E4 protein in the upper layers of differentiating epithelial cells, thereby supporting viral genome amplification (Bhattacharjee et al. 2022;Xu et al. 2024). E1 can bind to the upstream regulatory sequences of HPV, promoting the expression of the HPV E6 and E7 oncogenes (Wang et al. 2023). The two late genes of HPV (L1 and L2) encode the viral capsid structural proteins, with L1 being the major capsid protein (Xiao et al. 2025) .
The fidelity of normal cell replication is controlled by signaling pathways that prevent the propagation of damaged DNA (Sebastian et al. 2025) . Central to these repair pathways are the ATM and ATR kinases (Yates et al. 2025). The DNA damage response (DDR) pathway is essential for maintaining replication fork stability and the smooth progression of the Sphase (Studstill et al. 2023;Fielden et al. 2025). In previous studies, HPV E7 has been reported to disrupt RB/E2F complexes and induce S-phase re-entry and DDR proteins such as Ataxia Telangiectasia Mutated (ATM) and ATR (ATM and Rad3-related) kinases, leading to G2 prolongation and viral DNA amplification (Moody 2017;Prasad Tharanga Jayasooriya et al. 2018). Moreover, HPV E4 has been shown to induce G2 arrest via interaction with cyclin B/cdk1 complexes, while E6 in high-risk HPV types targets and inactivates p53, with downstream effects that differ depending on the study. However, most of the latter studies have been conducted using overexpressed proteins and/or culture systems that do not support the natural HPV infection cycle. Furthermore, the current understanding of the interaction between the virus and host under the physiological cycle of HPV18 remains limited.
Recently, we used conditional reprogramming(CR) technology to establish human VIN18 cells (vaginal intraepithelial neoplasia cell infected with HPV18 naturally), thus closely mimicking the viral survival state in humans (Zhu et al. 2017;Wu et al. 2022).VIN18 cells contain both episomal HPV18 (Wu et al. 2022) and exhibit HPV18 gene integration(figure S1) .In this study, we further advanced the VIN18 model by establishing a 3D time gradient system of ALI to evaluate different states of differentiation. We observed that stratified epithelium forms through in vitro differentiation, resembling the polarized structure of the vaginal epithelium in vivo, including the basal layer, spinous layer, granular layer, and stratum corneum, thereby replicating the epithelial differentiation conditions required for HPV replication and proliferation. Most importantly, we found that in ALI 3D culture, HPV18 DNA is effectively amplified, the L1 capsid protein is expressed in the stratum corneum, and the viral particles complete their assembly during the VIN18 differentiation process. The establishment of the VIN18 ALI 3D raft organoid model is of significance for understanding the viral life cycle, virus-host interactions, precancerous lesions, and the development of cancer under conditions that closely mimic human pathophysiological conditions.
Using VIN18 ALI 3D rafts, we analyzed the expression changes of host cell differentiation proteins, including involucrin and filaggrin, as well as cell cycle regulatory proteins such as cyclin A2, pRB, phosphorylated ppRB, p53, and HPV18 E7. Furthermore, we used specific small molecule inhibitors to explore the molecular mechanisms by which HPV18 regulates host cells to facilitate the completion of the viral life cycle, focusing on the ATM-CHK2 signaling pathway. The results reveal that normal epithelial cells exit the cell cycle after differentiation, but HPV-infected cells can bypass cell cycle checkpoint controls and remain active. The CHK2 signaling pathway downstream of ATM exerts a negative feedback regulation on the activity of upstream ATM, whereas activated pATM upregulates p53 and p21, leading to a decrease in cyclin D1. Furthermore, elevated expression of HPV18 E7 induces reentry of VIN18 cells into the S phase, with increased expression of cyclin A2 and cyclin B1, resulting in S phase/G2 phase arrest to support viral genome replication and amplification. These findings offer a valuable model for HPV biological research and provide new insights into the study of the differentiation-regulated HPV life cycle, with implications for furthering understanding of the occurrence and development of HPV-related cancers and anti-HPV drug screening.
## 2. Materials and methods
## 2.1. 2D culture of undifferentiated cells
VIN18 (vaginal intraepithelial neoplasia cell infected with HPV18 naturally) were co-cultured with feeder cells (irradiated mouse fibroblast 3T3 cells) in medium supplemented with ROCK inhibitor as described previously (Liu et al. 2017).And then cultured with epithelial cell growth medium to 50 % density. Medium was prepared with designated concentrations of DMSO (control), ATM inhibitor KU55933, or CHK2 inhibitor BML277. The cells were rinsed three times with DPBS, fresh culture medium was applied, and culture was continued for 24 h. Human normal vulvar epithelial cells (NVECs) were established and cultured in a previous study (Zhu et al. 2017).
## 2.2. Air-liquid interface (ALI) 3D culture
VIN18 and NVECs ALI 3D cultures were established as described previously (Wu et al. 2022;Xia et al. 2022). cells(2.5 × 10 5 ) were suspended in 400 µL growth medium (CELLnTEC Advanced Cell Systems AG, Stauffacherstrasse, Bern, Switzerland) and seeded into the Millicell PCF inserts (12 mm size, Millipore). Cells were cultured at 37 • C and 5 % CO2 for 72 h. The growth medium was replaced with differentiation medium (CELLnTEC Advanced Cell Systems AG) for 30 h, and then ALI 3D cultures were cultured for another 8 to 20 days, with the medium changed every two days. On the 11th day of gas-liquid differentiation, inhibitor was diluted at the required concentration with 3D differentiation medium, and incubation was continued for 7 days.
## 2.3. H&E staining
ALI 3D cultures were fixed in 4 % paraformaldehyde, dehydrated, paraffin embedded, and sectioned using standard histological procedures. The sections were stained with hematoxylin and eosin (H&E) (Zhongshan Golden Bridge Company, Beijing, China) and photographed under an EVOS microscope (ThermoFisher Scientific).
## 2.4. DAB staining and immunofluorescence assay
ALI or matrigel 3D cultures were fixed in 4 % paraformaldehyde, embedded in paraffin, and cut into 4 µm sections. DAB staining was performed using the EliVision Super DAB kit (Maixin biotech company, Fuzhou, China) as described previously (Xia et al. 2022). Immunofluorescence assays were performed as described previously (Wu et al. 2022). Slides were probed with the primary antibodies listed in Table 1.
## 2.5. BrdU incorporation assays
Eight hours prior to sample collection at each time point in the ALI 3D culture, 100 µg/mL of BrdU (5-bromo-2′-deoxyuridine) was added to the differentiation medium to label newly synthesized DNA in living cells (Banerjee et al. 2018). BrdU incorporation into VIN18 ALI cultures was visualized by immunofluorescence microscopy.
## 2.6. Analysis of the drug toxicity
Cells in growth phase were seeded in a 96-well plate with 100 μL of culture medium per well. Drugs were diluted in triplicate in culture
## Table 1
List of PCR primers used in this study.
## Primer Names
Sequences (5'-to-3') Amplicon size (bp)
ICTCCTTAATGTCACGCACGATT medium in a geometric series with 4-7 different concentrations, a DMSO control group and a blank group. After 24 h of treatment, culture medium containing 1/10 vol of CCK-8 reagent was applied. The absorbance was measured using a microplate reader, and the cell viability corresponding to each drug concentration was calculated according to the reagent instructions. GraphPad Prism 8.0 was used to plot the data and calculate the IC50.
$$HPV18 L1-F GCCGCCACGTCTAATGTTTC 149 HPV18 L1-R CCCTGTGATAAAGGACGCGA HPV18 E6-F CGCTTTGAGGATCCAACACG 439 HPV18 E6-R GTTCCTGTCGTGCTCGGTTG HPV18 E7-F GTCACGAGCAATTAAGCGAC 212 HPV18 E7-R CACAAAGGACAGGGTGTTCA HPV18 E1^E4-F ACTCTATGTGCAGTACCAGTG 266 HPV18 E1^E4-R TATTATAGGCGTAGTGTTAC β-actin-F GACTACCTCATGAAGATCCTCACC 102 β-actin-R$$
## 2.7. Western blot analysis
VIN cells or ALI 3D raft cultures were collected in RIPA buffer (Beyotime, Beijing, China). Protein concentrations were measured with a BCA Protein Assay Kit (Beyotime, Beijing, China). Western blotting analysis was carried out with the primary antibody listed in Table 1.
## 2.8. PCR and quantitative RT-PCR
DNA was extracted from VIN cells or ALI 3D raft cultures using the TIANamp Genomic DNA Kit (Tiangen). 1).Std curve plotting with Bio-Rad software: ① Construct pUC19-HPV18 E2/E7 plasmid as standard via homologous recombination. Calculate its copy number using the formula (copies/mL = 6.02 × 10²³ × C/MW, MW=base count × 660), then perform 5-step 10-fold dilution and compute copy numbers of each diluted standard. ② Quantify HPV18 E7 region of each standard by QPCR; plot std curve (X: Log initial copies, Y: Ct value) via Bio-Rad software to get linear equation (primers in Table 1).2).Cell total viral DNA QPCR: Use DNA from early/mid/late-passage VIN18 cells as templates to quantify HPV18 E7 region.3).Substitute Ct values into the std curve equation to obtain total initial HPV18 DNA copies in VIN18 cells at different passages.4).Per-cell HPV18 DNA relative copy number:Record DNA concentration/volume during extraction, calculate total cells (6.6 pg DNA per diploid cell), then divide total copies by cell number to get per-cell copies as described previously (Banerjee et al. 2017).
For quantitative RT-PCR, RNA was extracted using TRIZOL (Ther-moFisher Scientific, Waltham, MA, USA), and cDNA was generated using PrimeScript™ RT Master Mix (Takara Bio Inc., Dalian, China) according to the manufacturer's instructions. Quantitative PCR was performed using the Premix Ex Taq kit (Takara Bio Inc., Dalian, China) with the specific primers listed in Table 2. The average Ct was normalized to GADPH mRNA levels.
## 2.9. Statistical analysis
Statistical analysis and data plotting were conducted using GraphPad Prism8.0. All experiments were performed at least three times. Data are presented as mean ± SEM, and statistical significance was considered at p < 0.05. (* p < 0.05, ** p < 0.01, *** p < 0.001)
## 3. Results
## 3.1. Establishment of a 3D time gradient model of ALI based on VIN18 cells
In our previous research, we utilized the Conditional Reprogramming (CR) technique to establish the first human cell line from vulvar lesions, and for the first time, we observed patient-derived natural infection of HPV18 completing its full life cycle in vitro (Wu et al. 2022). Building on this foundation, we employed VIN18 cells for ALI 3D culture. The cells were collected on days 14, 16, 18, and 20 of differentiation for evaluation. HE staining revealed a typical stratified squamous epithelial structure (Fig. 1A left). As the differentiation process progresses, the organoids become thicker (Fig. 1 right), and the expression of the differentiation marker involucrin continues to increase (Fig. 1B), which is conducive to the completion of the differentiation-dependent HPV18 life cycle.Additionally, immunofluorescence results indicated that Ki67, a cell proliferation marker, was predominantly expressed in the nuclei of the basal and spinous layers of the ALI 3D cultures (Fig. 1C). The late differentiation marker Filaggrin was primarily localized in the granular and cornified layers, while the early differentiation marker Involucrin was expressed in the spinous layer and partially in the granular layer cells (Fig. 1D). There was overlap in the expression of these two differentiation markers within the granular layer, and the expression of both markers gradually increased from day 14 to day 20 of differentiation. These experimental results verify that the ALI 3D raft organoids arising from VIN18 cells exhibit expression patterns of differentiation, characteristic of human stratified squamous epithelium, thereby simulating vaginal epithelial tissue organoid formation in vitro.
## 3.2. HPV18 E7 promotes host DNA replication in VIN18 ALI 3D rafts
Next, we utilized the newly established pathophysiological model to investigate the molecular mechanisms underlying the interaction between naturally infected HPV18 and the host. Initially, we analyzed the expression of HPV18 E7 through immunohistochemical experiments. Our findings revealed that E7 was predominantly localized in the cytoplasm and nuclei of the granular layer and the underlying differentiated layers. Furthermore, from day 14 to day 20 of differentiation, the number of positive nuclei progressively increased (Fig. 2A). Consequently, we explored the impact of HPV18 E7 on host DNA replication. Immunofluorescence staining of Cyclin A2 (which is expressed in the S phase (Hochegger et al. 2008;Bertoli et al. 2013)) and BrdU (which is incorporated in S phase and passed to daughter cells) revealed co-localized cells that were positive for both Cyclin A2 and BrdU, suggesting that these cells had entered the S phase. Additionally, cells were observed that were positive for Cyclin A2 but negative for BrdU, representing the very early S phase; as well as a small subset that were BrdU-positive but Cyclin A2-negative, representing the S/G2 transition or G2 phase (Fig. 2B). The number of co-localized nuclei increased from day 14 to day 20, indicating an increase in the number of cells re-entering the S phase. Notably, BrdU-positive cells were observed in the stratum spinosum and stratum granulosum on day 20 of differentiation, with co-localization of BrdU and Cyclin A2. This suggests that viral DNA replication increases with prolonged differentiation time, reflecting the differentiation-dependent amplification of viral DNA. To validate this hypothesis, we extracted total DNA from ALI 3D cultures and performed quantitative PCR. The results indicated that the viral copy number gradually increased with prolonged differentiation time (Fig. 2C), consistent with the increase in the number of cells re-entering the S phase. Collectively, these results indicate that the virus initiates replication and proliferation as the cells differentiate. This is the first demonstration of the replication of naturally infected HPV18 virus in an in vitro 3D model.
## 3.3. Key viral and host regulatory proteins in the ALI 3D VIN18 cell model display distinct regulatory patterns
To evaluate the temporal expression of viral genes in the ALI 3D
VIN18 cell model, we collected samples at several time points during differentiation and extracted mRNA for RT-qPCR quantitative analysis.
From day 14 to day 18 of differentiation, the transcription levels of critical viral genes, including E1^E4, E6, E7 and L1, increased with the progression of differentiation, while the capsid gene L1 increased and peaked on day 20 (Fig. 3A). Notably, E1^E4 has been established as a marker of viral amplification (Nakahara et al. 2005), while L1 is a late assembly protein that is critical for viral packaging (Pinidis et al. 2016).
These results suggest that we captured both the initiation and late assembly phases of the viral life cycle. For additional evidence of viral-host interaction during HPV18 infection in the VIN18 differentiation model, we performed western blot analysis of the early differentiation protein Involucrin, the late differentiation protein Filaggrin, and HPV18 E7 in the ALI 3D cultures. All of these proteins were gradually upregulated over the differentiation period, although the level of HPV18 E7 slightly decreased from day 18 to day 20, which is consistent with the RT-qPCR results (Fig. 3B, first three rows). Additionally, the S-phase regulatory protein cyclinA2 showed a gradual increase from day 14 to day 18 and remained stable from day 18 to day 20, closely following the trend of E7, which is consistent with previous findings indicating that E7 can induce S-phase re-entry (Shin et al. 2012) (Fig. 3B, fourth row). However, the pattern of changes in ppRB and p53 were consistent with those of E7 (Fig. 3B, last four rows), rather than the expected targeted degradation (Boyer et al. 1996;Tomita et al. 2020). Therefore, these results suggest that the expression patterns of key regulatory genes of host cells during the physiological cycle in the VIN18 cell differentiation model differ from those in previous studies.
## 3.4. HPV18 activates the DDR (DNA damage response, DDR) pathway in the VIN18 cells ALI 3D differentiation model
Viral replication stress has been shown to impact DDR pathways, The total DNA in ALI 3D cultures was collected at 14th, 16th, 18th and 20th days, and the recombinant plasmid pUC19-HPV18E2/E7 was used as a standard. The DNA copy number of HPV18 in VIN18 cells was quantitatively detected by qPCR. including ATM, CHK2, and associated downstream molecules such as p53 (Sharma and Munger 2020). Therefore, we sought to analyze the effect of naturally infected HPV18 on the ATM-CHK2 pathway in VIN18 cell rafts. The phosphorylation levels of both pATM (S1981) and its downstream targets pCHK2 (T68) and p53 increased with prolonged differentiation time, with an upward trend consistent with the expression level changes of HPV18 E7 (Fig. 4A). This suggests that the ATM pathway is gradually activated as VIN18 cells differentiate.
To distinguish whether this activation is caused by HPV18 or cell differentiation, we utilized NVEC as a normal cell control for ALI 3D culture. The early differentiation marker protein involucrin significantly increased from day 8 to day 10 and remained stable from day 10 to day 14 in NVEC rafts. Furthermore, the phosphorylation level of pATM (S1981) showed no significant increase from day 8 to day 12 of differentiation, with only a slight increase by day 14 (Fig. 4B, top half). Moreover, the pCHK2 (T68) exhibited only a brief increase on the 10th day of differentiation, followed by a decline, and did not sustain continuous phosphorylation; whereas p53 increased and maintained a higher and stable level (Fig. 4B, bottom half). Next,we want to know whether this corresponds with downregulation of ATR signaling.In the upstream of ATR, the protein expression of TopBP1 increases, and the phosphorylation level of CHK1 (S296) downstream of ATR gradually rises (Figure S2A). Meanwhile, when normal NVEC cells are subjected to 3D ALI culture, the opposite results are observed: as normal cells differentiate, the expression of TopBP1 gradually decreases, and the phosphorylation level of CHK1 (S296) also gradually decreases (Figure S2B). This indicates that the phosphorylation and activation of CHK1 ( S296) is mainly regulated by the virus. These results are consistent with the possibility that HPV18 mildly activates the ATM-CHK2 signaling pathway, with the virus playing the primary activating role, while p53, as a downstream molecule of ATM, is likely to be mainly upregulated through phosphorylation-regulated ATM.
## 3.5. Host ATM-CHK2 interaction with HPV18 is dependent on the VIN18 differentiation status
To further investigate the host ATM-CHK2 interaction with HPV18, we sought to identify drugs that inhibit naturally infected HPV18 at the 2D undifferentiated level. Through literature review and preliminary screening, we identified KU55933 (Stakyte et al. 2021) targeting ATM; and BML277 (Tian et al. 2020) targeting CHK2. The toxicity and inhibition rates of these drugs on NVEC and VIN18 cells were determined (Table 3), and final working concentrations of 1 µM and 2.5 µM were selected. Upon specific inhibition of the ATM-CHK2 targets, both phosphorylated pATM (S1981) and CHK2 (T68) showed a concentration-dependent decrease. However, there was no effect on the expression of HPV18 E7 (Fig. 5A andB) or the HPV18 DNA copy number (Fig. 5C andD). The results indicate that under undifferentiated conditions, the ATM-CHK2 pathway does not regulate viral maintenance and replication in VIN18 cells. This is consistent with the conclusion of a previous study (Moody and Laimins 2009).
Next, we investigated the regulatory role of the ATM-CHK2 signaling pathway on HPV18 under 3D differentiation conditions. The results showed that ATM inhibitor treatment significantly suppressed ATM and CHK2 phosphorylation in differentiated VIN18 cells. Furthermore, a concentration-dependent reduction was observed in the expression of HPV18 E7 protein (Fig. 5E). In contrast, after CHK2 inhibition, ATM was activated and the HPV18 E7 protein was increased (Fig. 5F). Consistently, the HPV18 copy number was decreased in response to ATM inhibitor and increased in response to CHK2 inhibitor in the differentiated cells (Fig. 5G andH). Our results suggest that the activity of downstream CHK2 in differentiated VIN18 cells exerts a negative feedback regulation on the activity of upstream ATM, thereby influencing the replication and Fig. 4. Activation of ATM-CHK2 signaling pathway in VIN18 cells under ALI 3D culture. VIN18 cells were inoculated at a density of 3 × 10 5 cells per insert, and total proteins were collected on the 14th, 16th, 18th and 20th days. The changes in phosphorylated ATM, phosphorylated CHK2 and related total proteins were detected by western blotting. (B) Normal cervical epithelial cells (NVECs) were inoculated at 6 × 10 5 cell density, and total proteins were collected at 8, 10, 12, and 14 days. The changes in phosphorylated ATM, phosphorylated CHK2 and related total proteins were detected by western blotting. GAPDH was assessed as an internal control. The band grayscale values of phosphorylated ATM were analyzed using ImageJ.
## Table 3
Half of the inhibitory concentration of the inhibitor. After the cells were treated for 24 h according to the drug concentration gradient set in advance, the cell survival rate was measured using CCK-8. The graphs were drawn by GraphPad Prism 8.0, and the IC50 of each drug in the two kinds of cells was calculated by software analysis.
## 3.6. ATM and CHK2 inhibitors have differential effects on the proliferation and differentiation of VIN18 cells
To provide additional insight into the effects of ATM and CHK2 inhibition on VIN18 host cells, we evaluated the morphology and proliferation status of 3D cultures after drug treatment. HE staining as shown in Fig. 6A, we find that after treatment with the CHK2 (T68)-specific inhibitor BML-277, the basal to granular layers of the ALI 3D structure gradually thickened, whereas the ATM inhibitor KU55933 had the opposite effect(Fig. 6B). Next, we evaluated the effect of pathwayspecific inhibition on Notch, which is a strong positive regulator of keratinocyte differentiation (Rangarajan et al. 2001). Because Notch signaling involves nuclear translocation of the Notch intracellular domain (NICD) to initiate the transcription of Notch target genes (Zhou et al. 2022;Dilawar et al. 2025), we examined the protein expression of NICD and the cell proliferation marker (proliferating cell nuclear antigen, PCNA). ATM inhibition in ALI 3D differentiation cultures led to a decrease in the protein levels of both NICD and PCNA (Fig. 6C), indicating a reduction in both the proliferative and differentiation capacities of the 3D cultures. However, inhibition of CHK2 promoted a gradual increase in NICD, and the protein level of PCNA did not decrease in response to CHK inhibition (Fig. 6D). These results are consistent with the observed effects of these inhibitors on the HPV18 E7 protein levels and viral copy numbers (Fig. 5), further supporting the negative feedback regulation by CHK2 of upstream ATM activity.
## 3.7. Regulation of the VIN18 cell cycle by CHK2
While p53 primarily mediates DDR (Bunz et al. 1998), the p53-p21 and RB signaling pathways play crucial roles in cell cycle regulation (Engeland 2022). Therefore, we evaluated these pathways to analyze the potential regulatory role of ATM and CHK2 on the cell cycle. Upon ATM inhibition, the expression of both p53 and p21 decreased, while the level of cyclinD1 remained relatively stable. The S-phase cyclin A2 also decreased, leading to a reduction in RB phosphorylation and cyclinB1 expression (Fig. 7A). Because CyclinB1 is the primary cyclin for progression from S to G2 phase, these results are indicative of a short maintenance period of the G2 phase, which may explain the observed decrease in HPV18 DNA copy number upon ATM inhibition (Fig. 5G).
Upon inhibition of CHK2, p21 was elevated, the expression of cyclinD1 was decreased, and the expression of cyclin A2 was increased (Fig. 7B). This was accompanied by increased levels of pRB phosphorylation and CyclinB1 expression (Fig. 7B), thus providing evidence of Sphase re-entry and G2 arrest to facilitate viral self-amplification. Notably, CHK2 inhibitor also promoted increases in cell division cycle 25A (CDC25A) and cell division cycle 25C (CDC25C), which could contribute to cell cycle progression at the G2 phase (Prasad Tharanga Jayasooriya et al. 2018), thereby promoting viral genome amplification. Collectively, these findings support a mechanism in which activated CHK2 exerts negative feedback regulation on upstream ATM. Under the action of inhibitor, CHK2 is suppressed, leading to upregulation of p53 and p21, and decreased cyclinD1. Elevated expression of HPV18 E7 induces re-entry of VIN18 cells into the S phase, with increased expression of cyclinA2 and cyclinB1, and the cells are arrested in the S phase/G2 phase to support viral genome amplification (Fig. 8). The effect of CHK2 inhibitor on cell cycle progression of VIN18 ALI 3D culture. VIN18 cells were inoculated at a density of 3 × 10 5 cells per insert, and inhibitors were applied from day 11 to 18 after air-liquid differentiation. DMSO was applied as the control group. On the 18th day of differentiation, ALI cultures were collected to extract total proteins, and the changes in the protein levels of cycle-related proteins were analyzed by western blotting. GAPDH and β-actin were used as internal controls.
## 4. Discussion
high-risk HPV18 differed from those reported in previous studies (Sima et al. 2008;Nicolò et al. 2023). Previous conclusions have been primarily based on cervical cancer cell lines or immortalized cell lines with exogenous viral gene transduction. Therefore, the results from our study highlight the utility of the VIN18 3D ALI model for identifying physiologically relevant regulatory mechanisms in precancerous cells naturally infected with HPV18.
The fidelity of normal cellular replication is governed by signaling pathways that prevent the propagation of damaged DNA (Drew et al. 2025). Central to these repair pathways are the ATM and ATR kinases (Zhou et al. 2023). Previous studies have indicated the ATM double-strand DNA damage activation response is essential for differentiation-dependent HPV genome amplification (Vats and Laimins 2025). Importantly, the DDR prolongs the G2 phase, during which host replication can support viral DNA replication and amplification (Moody 2017). In HPV-positive cells, ATM-dependent DDR activation is essential for differentiation-dependent HPV genome amplification. This activation occurs in the absence of external DNA damaging agents, but the specific regulatory mechanism of ATM activation remains unclear (Drew et al. 2025). Our results demonstrate that in HPV18-positive VIN18 cells, differentiation only mildly activates the ATM-CHK2 signaling pathway, with HPV18 playing the primary activating role in VIN18 cells (Fig. 8), thus clarifying the molecular mechanisms of viral/host interactions associated with natural infection.
For additional insight, we used specific inhibitors to suppress the upstream and downstream of ATM-CHK2 pathway. Our results indicate that this pathway is unnecessary for the stable maintenance of the virus under undifferentiated conditions but is essential for differentiationdependent HPV18 genome amplification. We also discovered that the phosphorylation of downstream CHK2 T68 has a negative feedback regulatory effect on the phosphorylation of upstream ATM S1981. Under this negative feedback response, the expression of the NICD protein in VIN18 cells increases, while the expression of PCNA remains stable. This supports a mechanism that may ensure balance between proliferation and differentiation in host VIN18 cells to support viral amplification. We demonstrated that upon inhibition of CHK2 T68 phosphorylation, the phosphorylation of the upstream ATM is activated, which subsequently upregulates the expression of p53. Although p53 is regulated by viral oncoproteins in HPV-positive cells, these results suggest that the phosphorylation of the activated upstream molecule ATM (S1981) modulates downstream p53, maintaining stable protein levels of p53 under the joint regulation of the host and the virus. The host regulates the cell cycle through P53-P21, leading to a decrease in cyclinD1. Increased expression of HPV18 E7 induces the re-entry of VIN18 cells into the S phase, and the expression of cyclinA2 and cyclinB1 increases, resulting in cell cycle arrest in the S phase/G2 phase to support viral genome amplification. It has been reported in the literature that E4 may enhance the efficiency of viral genome amplification by maintaining an environment where G2 is suppressed during differentiation and by activating the MAPK signaling pathway (p38, ERK1/2, pJNK) (Raj et al. 2022). Therefore, future efforts to examine the changes in the early protein E4 at the transcriptional level are likely to yield additional insights.
HPV16 is the most prevalent genotype in clinical vulvar intraepithelial neoplasia (VIN); thus, extending our HPV18-based VIN model to HPV16 and delineating genotype-specific host-virus interactions are critical for boosting translational relevance. To address this, future research will first enroll HPV16-positive VIN patients and isolate primary VIN16 cells using our standardized VIN18 protocol, followed by conducting systematic comparative analyses between VIN16 and VIN18 cells. The comparisons will focus on three core aspects: HPV E6/E7 expression and interaction with host p53/pRb, genotype-specific host immune responses (including TLR signaling and cytokine secretion), and disparities in cell proliferation, apoptosis, and epithelial-mesenchymal transition.
## 5. Conclusions
In conclusion, our established VIN18 cell-derived ALI 3D culture system provides a valuable model for HPV biology studies. Based on the ALI 3D differentiation model, we analyzed the DDR pathway in VIN18 cells, which revealed that the regulatory mechanisms of ATM differ from those described in previous studies using more primitive culture systems. The ALI 3D organoid model established with VIN18 cells provides a valuable model for studying HPV biology, including the occurrence and development of HPV-related cancers, and may serve as a drug screening platform for assessment of therapeutic potential.
& editing, Software, Funding acquisition.
## Declaration of competing interest
We dedlare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no Professional or other personal interest of any nature or kind inany product, senvice and/or company that could be construed as influencing the position presented in or the review of the manuscript entitied.
## References
1. Banerjee, Moore, Broker et al. (2018) "Vorinostat, a pan-HDAC inhibitor, abrogates productive HPV18 DNA amplification" *Proc. Natl. Acad. Sci. U S A*
2. Banerjee, Wang, Beadle et al. (2017) "Evaluation of ODE-bn-PMEG, an acyclic nucleoside phosphonate prodrug, as an antiviral against productive HPV infection in 3D organotypic epithelial cultures" *Antivir. Res*
3. Bertoli, Skotheim, De Bruin (2013) "Control of cell cycle transcription during G1 and S phases" *Nat. Rev. Mol. Cell Biol*
4. Bhattacharjee, Das, Biswal et al. (2022) "Mechanistic role of HPVassociated early proteins in cervical cancer: molecular pathways and targeted therapeutic strategies" *Crit. Rev. Oncol. Hematol*
5. Boyer, Wazer, Band (1996) "E7 protein of human papilloma virus-16 induces degradation of retinoblastoma protein through the ubiquitin-proteasome pathway" *Cancer Res*
6. Bunz, Dutriaux, Lengauer et al. (1998) "Requirement for p53 and p21 to sustain G2 arrest after DNA damage" *Science*
7. Cardona-Mendoza, Fonseca-Benitez, Buitrago et al. (2023) "Down-regulation of human papillomavirus E6 oncogene and antiproliferative effect of Schisandra chinensis and Pueraria lobata natural extracts on Hela cell line" *J. Ethnopharmacol*
8. Chapman, Liu, Meyers et al. (2010) "Human keratinocytes are efficiently immortalized by a Rho kinase inhibitor" *J. Clin. Invest*
9. Dilawar, Yu, Jin et al. (2025) "Notch signaling pathway in osteogenesis, bone development, metabolism, and diseases" *FASEB J.: Off. Publ. Fed. Am. Soc. Exp. Biol*
10. Doorbar, Quint, Banks et al. (2012) "The biology and life-cycle of human papillomaviruses" *Vaccine*
11. Drew, Zenke, Curtin (2025) "DNA damage response inhibitors in cancer therapy: lessons from the past, current status and future implications" *Nat. Rev. Drug Discov*
12. Egawa, Wang, Griffin et al. (2017) "HPV16 and 18 genome amplification show different E4-dependence, with 16E4 enhancing E1 nuclear accumulation and replicative efficiency via its cell cycle arrest and kinase activation functions" *PLoS. Pathog*
13. Engeland (2022) "Cell cycle regulation: p53-p21-RB signaling" *Cell Death. Differ*
14. Fielden, Siegner, Gallagher et al. (2025) "Comprehensive interrogation of synthetic lethality in the DNA damage response" *Nature*
15. Habanjar, Diab-Assaf, Caldefie-Chezet et al. "2021. 3D Cell culture systems: tumor application, advantages, and disadvantages" *Int. J. Mol. Sci*
16. Hochegger, Takeda, Hunt (2008) "Cyclin-dependent kinases and cell-cycle transitions: does one fit all?" *Nat. Rev. Mol. Cell Biol*
17. Hoppe-Seyler, Bossler, Braun et al. (2018) "The HPV E6/E7 oncogenes: key factors for viral carcinogenesis and therapeutic targets" *Trends. Microbiol*
18. Liu, Krawczyk, Suprynowicz et al. (2017) "Conditional reprogramming and long-term expansion of normal and tumor cells from human biospecimens" *Nat. Protoc*
19. Mckinney, Hussmann, Mcbride (2015) "The role of the DNA damage response throughout the papillomavirus life cycle" *Viruses*
20. Moody (2017) "Mechanisms by which HPV induces a replication competent environment in differentiating keratinocytes" *Viruses*
21. Moody, Laimins (2009) "Human papillomaviruses activate the ATM DNA damage pathway for viral genome amplification upon differentiation" *PLoS. Pathog*
22. Nakahara, Peh, Doorbar et al. (2005) "Human papillomavirus type 16 E1circumflexE4 contributes to multiple facets of the papillomavirus life cycle" *J. Virol*
23. Nicolò, Antonelli, Tanturli et al. (2023) "Bacterial species from vaginal microbiota differently affect the production of the E6 and E7 oncoproteins and of p53 and p-Rb oncosuppressors in HPV16-infected cells" *Int. J. Mol. Sci*
24. Pinidis, Tsikouras, Iatrakis et al. (2016) "Human papilloma Virus" *Life cycle and carcinogenesis. Maedica (Bucur)*
25. Tharanga Jayasooriya, Dilshara, Neelaka Molagoda et al. (2018) "Camptothecin induces G(2)/M phase arrest through the ATM-Chk2-Cdc25C axis as a result of autophagy-induced cytoprotection: implications of reactive oxygen species" *Oncotarget*
26. Raj, Kesari, Kumar et al. (2001) "Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation" *Mol. Cancer*
27. Sebastian, Sun, Fedkenheuer et al. (2025) "Mechanism for local attenuation of DNA replication at double-strand breaks" *Nature*
28. Shamir, Ewald (2014) "Three-dimensional organotypic culture: experimental models of mammalian biology and disease" *Nat. Rev. Mol. Cell Biol*
29. Sharma, Munger (2020) "Expression of the long noncoding RNA DINO in Human papillomavirus-positive cervical cancer cells reactivates the dormant TP53 tumor suppressor through ATM/CHK2 signaling" *mBio*
30. Shin, Pitot, Lambert (2012) "Pocket proteins suppress head and neck cancer" *Cancer Res*
31. Sima, Wang, Kong et al. (2008) "RNA interference against HPV16 E7 oncogene leads to viral E6 and E7 suppression in cervical cancer cells and apoptosis via upregulation of Rb and p53" *Apoptosis*
32. Stakyte, Rotheneder, Lammens et al. (2021) "Molecular basis of human ATM kinase inhibition" *Nat. Struct. Mol. Biol*
33. Stanley (2021) "Host defence and persistent human papillomavirus infection" *Curr. Opin. Virol*
34. Studstill, Mac, Moody (2023) "Interplay between the DNA damage response and the life cycle of DNA tumor viruses" *Tumour Virus Res*
35. Tian, Wang, Xu et al. (2020) "The expression and therapeutic potential of checkpoint kinase 2 in laryngeal squamous cell carcinoma" *Drug Des. Devel. Ther*
36. Tomita, Huibregtse, Matouschek (2019) "A masked initiation region in retinoblastoma protein regulates its proteasomal degradation" *Nat. Commun*
37. Vats, Laimins (2025) "How human papillomavirus (HPV) targets DNA repair pathways for viral replication: from guardian to accomplice" *Microbiol. Mol. Biol. Rev.: MMBR*
38. Wang, Guan, Cai et al. (2023) "Human papillomavirus E1 protein regulates gene expression in cells involved in immune response" *Appl. Biochem. Biotechnol*
39. Warburton, Della Fera, Mcbride (2021) "Dangerous liaisons: long-term replication with an extrachromosomal HPV genome" *Viruses*
40. Wu, Zhang, Kang et al. (2022) "The first Human vulvar intraepithelial neoplasia cell line with naturally infected episomal HPV18 genome" *Viruses*
41. Xia, Wu, Zhou et al. (2022) "Treating intrauterine adhesion using conditionally reprogrammed physiological endometrial epithelial cells" *Stem Cell Res. Ther*
42. Xiao, Liu, Li et al. (2025) "Viral oncogenesis in cancer: from mechanisms to therapeutics" *Signal Transduct. Target. Ther*
43. Xu, Shi, Zhou et al. (2024) "Mapping the landscape of HPV integration and characterising virus and host genome interactions in HPV-positive oropharyngeal squamous cell carcinoma" *Clin. Transl. Med*
44. Yates, Zhang, Burgers (2025) "DNA damage and replication stress checkpoints" *Annu. Rev. Biochem*
45. Zhou, Lin, Long et al. (2022) "Notch signaling pathway: architecture, disease, and therapeutics" *Signal Transduct. Target. Ther*
46. Zhou, Börcsök, Adib et al. (2023) "ATM deficiency confers specific therapeutic vulnerabilities in bladder" *cancer. Sci. Adv*
47. Zhu, Yang, Guo et al. (2017) "Ex vivo 2D and 3D HSV-2 infection model using human normal vaginal epithelial cells" *Oncotarget*
48. Zur Hausen (2002) "Papillomaviruses and cancer: from basic studies to clinical application" *Nat. Rev. Cancer* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12793645&blobtype=pdf | # Abstract citation ID: ofaf695.2344 P-2181. Post-Exposure Prophylaxis (PEP) of Respiratory Syncytial Virus (RSV) Infection After High-Inoculum RSV Human Challenge: Analysis of a Randomized Double-Blind, Placebo-Controlled Trial of EDP-323, an Oral, Non-Nucleoside Polymerase Inhibitor Antiviral
John Devincenzo, Alaa Ahmad, Shijie Chen, Scott Rottinghaus, Enanta Pharmaceuticals, Watertown
Background. RSV causes high secondary transmission rates in families, hospitals, nursing homes and other settings. Breakthrough RSV infections occur despite RSV vaccination or monoclonal antibody injection. Certain patients at great risk of severe RSV do not respond to RSV vaccination. No effective RSV treatments exist. EDP-323, a first in class, potent, oral, non-nucleoside small molecule RSV polymerase (L-protein) inhibitor rapidly reduces viral load and disease severity when treatment is started after an active RSV infection is identifiedfoot_0 . The efficacy of RSV antivirals used for post-exposure prophylaxis (PEP) is unknown.
Methods. A randomized, double-blind, placebo (PBO)-controlled study (NCT06170242) evaluated the efficacy, antiviral activity and safety, of EDP-323. Healthy volunteers received large intranasal exposures of a low-passage clinical strain of RSV-A (Memphis-37, 4 Log 10 plaque forming units) 2 on Day (D)0. RSV RT-PCR was performed on nasal washes collected twice daily on D2-12. If PCR-confirmed RSV infection had not occurred by D5am after RSV exposure, participants were randomized to receive daily oral EDP-323 high dose (600mg), low dose (200mg with 600mg loading dose) or PBO for 5D. PEP efficacy was evaluated in this pre-specified population using Fisher's Exact test (two-tailed) of RSV-uninfected vs infected (pre-defined as PCR-positive on 2 consecutive specimens).
Results. In this population, 68 RSV-exposed, susceptible subjects were randomized to receive EDP-323 (low dose N=24, high dose N=21) and 23 PBO. A 2x2 table shows infection outcomes (Table 1). 26% (6/23) of PBO recipients became infected vs 0% (0/45) of EDP-323 recipients (P< 0.001). Evaluated separately, the 2 EDP-323 dosing groups' PEP effects were statistically significant (low dose P= 0.009, high dose P= 0.022) vs PBO. No serious TEAEs, severe AEs, or AEs leading to treatment discontinuation or study withdrawal occurred. Frequencies of treatment-emergent adverse events (TEAEs) were similar across EDP-323 and PBO groups.
Conclusion. EDP-323 appears highly effective in preventing RSV infections when initiated up through 5 days after high-inoculum intranasal RSV exposure. These findings support further evaluation of EDP-323 for prophylaxis.
Disclosures |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548224&blobtype=pdf | # BMC Infectious Diseases
Birendra Gupta, Birendra Prasad Gupta, Ajay Sah, Abhay Kumar Sah, Vivek Gupta, Dinesh Ghimire, Nikita Acharya, Chandramani Wagle
## Abstract
Introduction Hepatitis E virus (HEV) is a leading cause of acute viral hepatitis in South Asia, frequently causing waterborne outbreaks. Despite the recognized past and potential future epidemic burden in Nepal, data on the serological and molecular characteristics of, and co-infections with, Hepatitis A, B, and C viruses remain scarce. This study aimed to perform a detailed virological and serological characterization of a 2014 outbreak in Biratnagar, Nepal, to define the etiology and evaluate diagnostic patterns.
MethodsIn this cross-sectional study, 211 patients with suspected acute viral hepatitis were enrolled from three hospitals in Biratnagar, Nepal, during the outbreak peak (April-May 2014). Serum samples were tested for HEV RNA (qRT-PCR), antigen, IgM, and IgG (ELISA), alongside HAV IgM, HBV core antibody, and HCV antibody. Concordance between HEV markers was assessed using Cohen's κ, associations with co-infections via Fisher's exact tests, and clinical correlates with Mann-Whitney U tests.Results HEV marker positivity among the 211 patients was as follows: HEV RNA was detected in 12 patients (5.7%), HEV antigen in 29 patients (13.7%), and HEV IgM in 36 patients (17.1%), indicating acute or recent infection. HEV IgG, reflecting past or recent exposure, was detected in 62 patients (29.4%). All successfully sequenced HEV RNApositive isolates were identified as genotype 1a, consistent with the predominant HEV genotype reported in Nepal. In contrast, HAV, HBV, and HCV infections were infrequent, each accounting for less than 3% of cases, and no significant co-infections were observed. Concordance between HEV markers was variable, supporting the utility of a combined molecular and serological diagnostic approach to capture infections at different stages.
ConclusionThis study provides a detailed characterization of HEV markers during the 2014 Biratnagar outbreak and confirms the circulation of HEV genotype 1a. HAV, HBV, and HCV infections were uncommon, with no significant co-infections observed. These findings offer valuable insights into HEV marker distribution and co-infection patterns, which can inform public health strategies, including enhanced surveillance, improved access to safe water, and consideration of targeted vaccination to mitigate future outbreaks in endemic settings.
## Introduction
Viral hepatitis is one of the few communicable diseases with rising mortality, with deaths increasing from 1.1 million in 2019 to 1.3 million in 2022. Of these deaths, 83% were due to hepatitis B and 17% to hepatitis C [1]. While Hepatitis E Virus (HEV) causes relatively low mortality in the general population (0.2-1%), the fatality rate rises sharply in pregnant women (10-40%) and in individuals with severe underlying conditions [2]. HEV remains a leading cause of acute viral hepatitis in low-and middleincome countries due to limited access to clean water and the higher risk of contaminated food, making it a continuing public health concern [3]. Four genotypes cause human disease, genotype 1 (HEV-1) is predominant across South Asia and is primarily transmitted through the fecal-oral route, often via contaminated water [4].
In South Asia, HEV genotypes 1 and 2 drive both epidemic and sporadic infections, with outbreaks amplifying disease burden in urban and rural settings [5].
Nepal and neighboring South Asian countries have experienced recurrent HEV outbreaks, with notable epidemics in Kathmandu (1987Kathmandu ( , 2012) ) and northern India (1955,2004), often linked to contaminated water sources [6,7]. Despite these events, the role of co-circulating hepatitis viruses namely hepatitis A (HAV), hepatitis B (HBV), and hepatitis C (HCV) in HEV outbreaks is poorly characterized [8]. Although documented co-infection rates with HAV and HBV are low (0-10%) in South Asian HEV outbreaks, they underscore the importance of systematic profiling [9][10][11]. Co-infections may alter clinical severity, diagnostic accuracy, and transmission patterns, yet systematic assessments of their prevalence during HEV epidemics are scarce [8]. Comprehensive serological and molecular profiling of HEV and coinfections is essential to elucidate viral interactions and inform effective outbreak control strategies in resourceconstrained settings [12]. Existing data from Nepal are limited by incomplete marker panels, absence of quantitative viral load assessment, and scarce evaluation of co-infections. Consequently, true disease burden attributable to HEV versus other hepatitis viruses is unknown, and the extent of diagnostic discordance among antigen, antibody, and RNA positivity remains unquantified [13].
Outbreak studies provide critical data on transmission patterns (e.g., waterborne routes) that guide surveillance and prevention strategies for endemic HEV in South Asia [14,15].
The 2014 HEV outbreak in Biratnagar, Nepal, which unfolded between April and July 2014, provides a valuable opportunity to address critical knowledge gaps. During this epidemic. 2,789 jaundice cases were reported to the District Public Health Office (DPHO), Morang from diverse geographic areas. Notably, the majority of cases (n = 2,250; 80.7%) were from Biratnagar itself, reflecting the epidemic epicenter [16][17][18]. Contemporary investigations linked the outbreak to sewerage pipeline damage during urban construction, leading to water contamination, a common driver of HEV epidemics in Nepal [17]. Analyzing this dataset in 2025 remains critical, given HEV's persistent endemicity in Nepal, recurring outbreaks (with Kathmandu as a hotspot), and the lack of widespread vaccination, all of which highlight ongoing preparedness gaps [19][20][21]. The primary objective of this study was to perform a comprehensive serological and molecular characterization of HEV, evaluate co-infections with HAV, HBV, and HCV, and elucidate infection patterns to guide outbreak control strategies in 2025. Using data from a significant past outbreak, this study aims to enhance understanding of HEV epidemiology and co-infection dynamics, providing actionable insights for public health interventions in endemic settings.
## Methods
## Study design and setting
This was a cross-sectional, hospital-based study conducted in Biratnagar (Morang District) during the 2014 HEV outbreak, which spanned April to July 2014. Serum specimens were collected between 27 April -3 May, 2014 from three health facilities chosen to capture the public, private, and teaching-hospital sectors: Koshi Zonal Hospital (Government), Nobel Medical College Teaching Hospital (non-profit), and Biratnagar Hospital (private). This one-week collection period was selected based on the operational availability of investigators and resources.
## Participants and sampling frame
Consecutive ambulatory patients who had liver-function tests (LFTs) ordered by attending clinicians for suspected acute viral hepatitis were screened. After obtaining consent, residual serum samples were retrieved from hospital laboratories following routine clinical testing.
Inclusion criteria included clinical suspicion of acute viral hepatitis (AVH) (e.g., jaundice, anorexia, hepatomegaly, abdominal pain, nausea, vomiting, or fever) within the first 9 days of symptom onset. This extended window captures the acute-phase viremia period for HEV while accommodating real-world healthcare-seeking delays. Patients with pre-existing chronic liver disease or hepatotoxic drug exposure were excluded, and these conditions were verified using available clinical records.
## Data collection
Demographics and symptoms duration (days since onset) were abstracted from medical records onto a standardized case-report form. Point-of-care LFTs (total/direct bilirubin, aspartate aminotransferase, alanine aminotransferase and alkaline phosphatase) were performed as part of routine care; results were transcribed to the study database after de-identification. Written informed consent was obtained from patients or their guardians. Ethical approval was obtained from the Nepal Health Research Council, and data were anonymized using sample codes to preserve participant confidentiality.
## Serological testing
Venous blood (3-5 mL) was collected in serum-separator tubes at each hospital's clinical laboratory. After centrifugation (1 500 × g, 15 min, 4 °C), sera were aliquoted into 2 mL cryovials within 4 h of collection and stored at -80 °C. All 211 epidemic-period serum samples were tested in duplicate for hepatitis E virus (HEV) IgM, IgG, and antigen (Ag) using Wantai® ELISA kits (Beijing, China) following manufacturer instructions. Tests were conducted on an automated ELx50 plate washer and read using a VERSAmax microplate reader (Molecular Devices, USA). Optical density (OD) values were measured at 450 nm. Positive controls consisted of 1:16 and 1:32 dilutions of manufacturer-provided standards, and assays showed intra-assay coefficients of variation (CV) below 8%.
Manufacturer-defined cut-offs were applied for interpretation: OD values ≥ cut-off were considered positive, < cut-off negative, and those falling within 0.9-1.1× cut-off were considered equivocal and retested. The cut-off values were as follows: IgM = 0.26, IgG = 0.19, and Ag = 0.1. All ELISA assays used a standardized 50 µL serum volume per well. The assay limit of detection (LOD) is reported in copies/mL to remain consistent with the validated reference protocol and to facilitate comparison with existing HEV literature.
In addition to HEV testing, each serum was tested for evidence of hepatitis A virus (HAV) IgM, hepatitis B core antibody (HBV-cAb), and hepatitis C virus (HCV) antibodies using commercial ELISA kits (Wantai Biological Pharmacy Enterprise, Beijing, China) which was developed from the reported pair of antibodies [22]. HAV-IgM and HCV-Ab assays used 50 µL of serum; HBV-cAb detection was performed on 1:100 diluted serum, with OD-based interpretation using manufacturer cut-offs (HAV-IgM: 0.105; HBV-cAb: 0.1; HCV-Ab: 0.13). All serological tests were performed in duplicate to ensure reliability.
## Molecular detection and quantification
Total RNA was extracted from 200 µL of patient serum using the MagNA Pure LC 2.0 automated platform (Roche Diagnostics, Indianapolis, Indiana, USA), employing the Total Nucleic Acid Kit -High Performance protocol [23]. To monitor for extraction efficiency and PCR inhibition, an internal control bacteriophage MS2 RNA was added to each sample prior to extraction.
Detection and quantification of HEV RNA were performed using a laboratory-developed quantitative reverse-transcription PCR (qRT-PCR) assay targeting the ORF3 region as previously described (Wang et al., 2013), which provides full primer/probe sequences, reaction mixture, and cycling conditions [24]. The reaction utilized TaqMan chemistry, and the limit of detection (LOD) for the assay was 50 copies/mL (95% CI: 40-62). The qRT-PCR assay was validated using a standard curve generated from in vitro-transcribed RNA, and all samples were run in duplicate. Quantification followed the standard curve method described by Wang et al. [24], which includes PCR efficiency assessment during curve generation; therefore, no additional independent efficiency testing was conducted.
Samples positive for HEV RNA were further characterized through genotyping. Nested reverse-transcription PCR (RT-PCR) was conducted to amplify two genomic fragments: A ~ 900-nucleotide (nt) segment of the ORF1 region using primers from Wang et al. and Dong et al., and a 348-nt segment of the ORF2 region using primers from Wang et al. and Ticehurst [25,26].
Amplicons were purified and Sanger sequenced bidirectionally using an Applied Biosystems 3730xl DNA Analyzer. Consensus genome sequences were generated by reference-based assembly. Forward and reverse Sanger reads were quality-trimmed and aligned to the hepatitis E virus genotype 1 reference genome (GenBank accession M73218, the same reference used by Wang et al. [24] using CodonCode Aligner v8.0. Ambiguous bases were resolved by manual inspection of chromatograms. Final consensus sequences were compared with publicly available HEV sequences in GenBank to confirm genotype assignment.
## Data management and analysis
De-identified data were entered into Microsoft Excel (v 2507) and analyzed using Python (v3.11) with libraries including pandas (v 2.2.2), SciPy (v 1. 16.1), and statsmodels (v 0.14.5). Continuous variables, including liver function tests (SGOT, SGPT, total and direct bilirubin, ALP) and post-onset days, were assessed for normality using the Shapiro-Wilk test. Non-normally distributed variables were summarized using medians and interquartile ranges (IQR), while normally distributed variables were presented as means ± standard deviation (SD). Categorical variables, such as HEV, HAV, HBV, and HCV positivity, sex, and co-infection patterns, were summarized as frequencies and percentages.
Comparative analyses between HEV-positive and HEVnegative patients were performed using Mann-Whitney U tests for non-parametric continuous variables and independent t-tests for parametric variables. Associations between categorical variables were evaluated using the Fisher's exact test. Effect sizes were calculated using rank-biserial correlation for continuous variables and Cramér's V for categorical variables. Concordance between different HEV assays (RNA, antigen, IgM) was assessed using Cohen's kappa statistic with 95% confidence intervals, and paired proportions were compared using McNemar's test. Single and co-infections with HEV, HAV, HBV, and HCV, were summarized and visualized using bar plots and correlation heatmaps. Due to the small number of RNA-positive cases, we did not perform formal statistical correlation between viral load and serologic markers or clinical parameters. P-values < 0.05 were considered statistically significant.
## Results
## Study population and baseline characteristics
A total of 211 patients presenting with suspected acute viral hepatitis during the peak of the 2014 HEV outbreak were enrolled across three healthcare facilities in Biratnagar, Nepal. The cohort was predominantly young adults (median age 26 years, range 5-90), with the largest proportion in the 15-24-year group (34.1%). Female patients comprised 55.0% of the cohort. Patients typically presented early in the clinical course, with median symptom duration of 4 days at enrollment (range 1-9 days). Liver function tests showed elevations consistent with acute hepatitis, including median SGOT of 45 U/L and SGPT of 40 U/L. Baseline demographic and laboratory characteristics are presented in Table 1. These baseline findings provided the clinical context for subsequent serological and co-infection analyses.
## HEV serological and molecular findings
All 211 serum samples were tested for HEV markers, including RNA, antigen (Ag), IgM, and IgG. HEV IgG positivity was most common, detected in 62 patients (29.4%), followed by IgM in 36 patients (17.1%) and antigen in 29 patients (13.7%). HEV RNA, the marker of active viremia, was detected in 12 patients (5.7%).
For these 12 RNA-positive patients, quantitative HEV RNA measurements derived from the standard curve ranged from 367 to 28,400 copies/mL (median = 5,370; IQR = 1,280-16,800). Overall, 24.2% (n = 51) of patients were positive for acute HEV infection (defined as detection of RNA, antigen, or IgM), while IgG positivity provided evidence of past or recent exposure beyond acute cases. These findings confirm HEV as the predominant etiology of acute viral hepatitis during the 2014 outbreak. Figure 1 presents a horizontal lollipop plot of HEV marker positivity, showing both absolute counts and percentages.
## Concordance between HEV assays
The concordance between HEV diagnostic markers varied notably across assay pairs. Cohen's κ analysis indicated no agreement between HEV RNA and HEV antigen (κ = 0.00, 95% CI: 0.00-0.00) as well as between HEV RNA and HEV IgM (κ = 0.00, 95% CI: 0.00-0.00), reflecting that RNA detection occurred independently of these serologic markers. In contrast, moderate agreement was observed between HEV antigen and IgM (κ = 0.36, 95% CI: 0.20-0.52). Paired comparisons using McNemar's test showed no statistically significant differences in positivity rates between any of the markers (all p ≥ 0.311).
## HEV genotyping and sequencing
Of the 12 patients who were HEV RNA-positive, nine had sufficient residual serum volume and RNA quality for downstream amplification. Nested RT-PCR targeting the ORF2 region yielded three high-quality amplicons that were successfully Sanger sequenced. All three sequences were identical across the amplified region and clustered within genotype 1a. These findings confirm that the 2014 Biratnagar outbreak was caused by a genotype 1a HEV strain, consistent with the predominant lineage circulating in South Asia.
## Co circulating hepatitis viruses and infection patterns
Among the 211 patients, non-HEV hepatitis viruses were infrequently detected. HAV IgM positivity was observed in 4 patients (1.9%), HBV core antibody in 4 patients (1.9%), and HCV antibody in 1 patient (0.5%). Most participants 125 (59.2%) were negative for all tested hepatitis viruses. HEV alone was detected in 50 patients (23.7%), while co-infections were uncommon: HEV + HAV in 3 patients (1.4%) and HEV + HCV in 1 patient (0.5%). These patterns of single and co-infections are visualized in Fig. 2.
These findings indicate that HEV was the predominant cause of acute viral hepatitis during the 2014 outbreak in Biratnagar, with limited co-circulation of HAV and HBV and negligible HCV involvement. Statistical associations between HEV positivity and other hepatitis viruses were One patient with HEV-HCV co-infection (0.5%) was excluded from these analyses due to the singular occurrence (Table 2).
Pairwise correlations between HEV, HAV, and HBV positivity were generally weak, consistent with the low frequency of co-infections. The correlation coefficients were r = 0.50 for HEV-HAV, r = -0.11 for HEV-HBV, and r = 0.10 for HAV-HBV (Fig. 3), indicating minimal concurrent circulation of these viruses during the 2014 outbreak in Biratnagar.
## Comparison of clinical and biochemical characteristics by HEV status
We evaluated associations between HEV infection and baseline clinical and biochemical parameters among 211 patients. Continuous variables -including liver function tests (SGOT, SGPT, total and direct bilirubin, ALP) and post-onset day (POD), were compared using the Mann-Whitney U test due to non-normal distributions. No statistically significant differences were observed between HEV-positive and HEV-negative patients. Median SGOT levels were 48.0 U/L (IQR: 30.0-73.0) in HEV-negative patients versus 38.0 U/L (IQR: 32.0-67.5) in HEV-positive patients (p = 0.426, rank-biserial r = -0.07). Similarly, SGPT, total bilirubin, direct bilirubin, ALP, and POD showed minimal differences between groups (effect sizes ranged from -0.15 to 0.02) (Table 3).
Sex distribution was also comparable between groups, with 42.2% females in HEV-negative versus 12.8% in HEV-positive patients (p = 0.488, Cramér's V = 0.05). Overall, HEV infection was not significantly associated with any measured baseline clinical or biochemical variables in this cohort, and effect sizes were uniformly small, indicating minimal variation between groups (Table 3).
## Table 2 Distribution of HAV and HBV co-infection among HEVpositive and HEV-negative patients
## Infection
## Discussion
This study characterizes HEV markers among patients during the 2014 outbreak in Biratnagar, Nepal. Among the patients, 12 patients (5.7%) tested positive for HEV RNA, 29 patients (13.7%) were positive for HEV antigen, and 36 patients (17.1%) were positive for HEV IgM.
Additionally, 62 patients (29.4%) exhibited IgG positivity, indicating past or recent exposure to HEV. The affected cohort was primarily composed of young adults, with a median age of 26 years. Most patients presented early, with a median of 4 days after the onset of symptoms, underscoring the acute nature of the outbreak and the need for timely intervention. The detection of HEV IgG, IgM, antigen, and RNA in varying proportions among patients reflects the diverse stages of infection and underscores the utility of multiple diagnostic markers in identifying active and past infections. Notably, the absence of significant concordance between HEV RNA and other markers suggests that HEV RNA detection may not always correlate with serological evidence, which highlights the complexity of HEV diagnostics [27].
Our serological and molecular profiling revealed a gradient in HEV marker positivity, with HEV RNA detected in only 5.7% of patients reflecting the brief viremic phase while HEV antigen (13.7%) and IgM (17.1%) were more prevalent, consistent with their utility for identifying acute infection during outbreaks [28]. The very low concordance between HEV RNA and serological markers (κ = 0.00 for RNA vs. antigen/IgM) highlights limitations in assay sensitivity and the narrow timing of RNA detection, as viremia often precedes or wanes before a measurable antibody response in endemic settings [29]. In contrast, the moderate agreement between antigen and IgM (κ = 0.36) supports their combined use for acute diagnosis and aligns with studies reporting variable RNA-antigen concordance (e.g., κ ≈ 0.6 in some contexts) [30]. International guidelines therefore recommend a combined molecular-serological approach to improve diagnostic yield, a strategy our findings clearly support [31]. Moreover, host immunocompetency may contribute to this discordance: individuals with subclinical immune deficiencies or nutritional compromise can show delayed or blunted antibody responses, reducing overlap between RNA detection and serology even in otherwise healthy populations. Although our cohort consisted mainly of immunocompetent adults, such variability could partly explain the observed marker mismatch. Collectively, these results underscore the need for multi-assay diagnostic strategies in resource-limited outbreak settings [32,33].
Co-infections with other hepatitis viruses were infrequent, with HEV alone was detected in 23.7% of patients. The low prevalence of HAV, HBV, and HCV co-infections aligns with previous studies in Nepal, where HEV circulation has been frequently reported during acute viral hepatitis outbreaks [21,27,34,35]. The minimal cocirculation observed here likely reflects the outbreak's waterborne nature. The limited co-infection with other hepatitis viruses observed in this cohort underscores the importance of monitoring HEV markers to understand infection patterns during outbreaks. Our analysis revealed no significant differences in liver function tests or clinical parameters between HEV-positive and HEVnegative patients. This finding suggests that HEV infection may not present with distinct clinical or biochemical features, making it challenging to differentiate from other causes of acute hepatitis based solely on these parameters. The absence of significant associations underscores the importance of laboratory diagnostics in confirming HEV infection, especially in regions where multiple hepatitis viruses co-circulate.
In our study, all successfully sequenced isolates from the 2014 Biratnagar outbreak were identified as genotype 1a, consistent with the predominant HEV genotype reported across Nepal. Earlier work from Kathmandu during a 1997 outbreak found all 48 HEV RNA-positive samples to be genotype 1, with subtypes 1a and 1c circulating and a marked rise of subtype 1c during the rainy season [28]. Similarly, a previous investigation of the Biratnagar outbreak confirmed subtype 1a as the main circulating strain [24]. These findings reinforce that genotype 1a remains the dominant cause of HEV epidemics in Nepal and help explain the persistence of large waterborne outbreaks in settings with comparable Collectively, these data reinforce HEV as the primary driver of acute viral hepatitis during this outbreak and offer valuable historical context for strengthening current outbreak preparedness in eastern Nepal, particularly regarding early case detection and integration of molecular and serological diagnostics.
Despite being based on a 2014 outbreak, the study provides valuable insights into HEV diagnostic strategies and epidemiological patterns, which remain informative for laboratory and clinical preparedness in regions at risk of similar outbreaks. Our study underscores the predominance of HEV as a cause of acute viral hepatitis and the limited role of HAV, HBV, and HCV co-infections, providing crucial evidence to guide targeted public health interventions, including prioritization of water, sanitation, and hygiene (WASH) programs. A major strength of this study is the inclusion of a well-characterized cohort of 211 patients from multiple healthcare facilities, reflecting diverse geographic and demographic backgrounds and enhancing the generalizability of the findings. Additionally, the use of multiple diagnostic markers (RNA, antigen, IgM, IgG) allowed for a nuanced understanding of acute versus past infections and the concordance between assays, strengthening the epidemiological insights.
However, these findings must be interpreted within the study's limitations. First, the retrospective nature of the study and reliance on a dataset collected during a single outbreak limits temporal generalizability. Second, the inclusion of cases from only one week of the epidemic period may not fully capture the complete spectrum of outbreak dynamics. Third, the absence of information on risk or exposure factors associated with the participants limits the ability to identify determinants of infection. Fourth, the low number of co-infections restricted statistical power for some analyses, such as evaluating associations between HEV and other hepatitis viruses or clinical severity in subgroups. Fifth, clinical and biochemical data were limited to baseline measurements, preventing assessment of longitudinal disease progression or outcomes. Furthermore, the inability to identify an etiological agent in a large proportion of patients (59.2%) may reflect the sensitivity limitations of available serological assays for hepatitis A-E viruses and the restricted use of molecular testing, underscoring the potential contribution of non-ABCDE hepatitis agents or other non-infectious causes a critical area for future research. Finally, the dataset lacked information on pregnancy status and other potential risk factors for many participants, limiting evaluation of HEV-associated maternal risk and preventing assessment of additional exposures that may influence infection.
This study underscores the substantial role of HEV in the 2014 Biratnagar outbreak and its limited overlap with other hepatitis viruses. Future research should focus on longitudinal follow-up of patients, genomic characterization of circulating HEV strains, and evaluation of vaccine feasibility in high-risk populations to better guide prevention and control strategies in endemic regions.
## Conclusion
In conclusion, the 2014 Biratnagar outbreak was overwhelmingly driven by hepatitis E virus, while revealing a critical complexity in its diagnosis. The integration of serological and molecular markers provided a comprehensive understanding of infection dynamics and underscored the value of combined diagnostics for outbreak investigations. These findings emphasize the need for strengthened surveillance, timely laboratory confirmation, and targeted prevention strategies, including improved sanitation and consideration of HEV vaccination in high-risk groups. While the study was limited by its outbreak-specific and cross-sectional design, it provides valuable insights into the epidemiology of HEV in Nepal and informs priorities for future research and public health interventions.
## References
1. (2024) "Action for Access in Low-and Middle-Income Countries"
2. Pérez-Gracia, Suay-García, Ml (1929) "Hepatitis E and pregnancy: current state" *Rev Med Virol*
3. Satapathy, Gaidhane, Bishoyi et al. (2025) "Burden of acute hepatitis E virus in South asia: insights from global burden of disease study 2021" *Diagn Microbiol Infect Dis*
4. Nelson, Labrique, Kmush (2019) "Epidemiology of genotype 1 and 2 hepatitis E virus infections"
5. Pallerla, Harms, Johne et al. (2020) "Hepatitis E virus infection: Circulation, molecular Epidemiology, and impact on global health" *Pathogens*
6. Shrestha, ; / J P -J O U R N A L S - (1128) "Epidemiology of viral hepatitis and liver diseases in Nepal" *Euroasian J Hepatogastroenterol*
7. Nasir, Wu (2020) "HEV, Dual Infection HBV. A review" *J Clin Transl Hepatol*
8. (1930)
9. Qadri, Fomda, Wani et al. (2025) "Hepatitis A and E: Prevalence, epidemiology and Co-infection among patients with acute viral hepatitis" *Int J Translational Med Res Public Health*
10. (1999)
11. Sayed (2023) "Dual infection of hepatitis A virus and hepatitis E virus-what is known?" *Viruses*
12. Daheriya, Gajbhiye, Shrikhande (2025) "Prevalence and co-infection of acute hepatitis A virus and hepatitis E virus infections in patients with acute viral hepatitis" *Int J Res Med Sci*
13. Rencken, Kirkwood, Kumar et al. (2023) "Systematic review of global hepatitis E outbreaks to inform response and coordination initiatives"
14. Tene, Diouara, Sané et al. (2025) "Virus (HEV) Infection in the Context of the One Health Approach"
15. Samaddar, Taklikar, Kale et al. (2019) "Infectious hepatitis: A 3-year retrospective study at a tertiary care hospital in India" *Indian J Med Microbiol*
16. Arora, Jindal, Shukla et al. (2013) "Water borne hepatitis a and hepatitis e in Malwa region of punjab, India" *J Clin Diagn Res*
17. Shrestha, Lama, Karki et al. (2014) *Emerging infectious diseases journal -CDC. Emerg Infect Dis*
18. Subba (2015) "Managing Hepatitis Outbreak in Biratnagar Nepal" *Science Journal of Public Health*
19. (2012)
20. Shrestha, Lama, Karki et al. (2014) "1 / E I D" *Emerg Infect Dis*
21. Katuwal, Thapa, Shrestha et al. (2024) "Hepatitis E virus in the Kathmandu valley: insights from a representative longitudinal serosurvey" *PLoS Negl Trop Dis*
22. U R N A L . P N T D
23. Shrestha, Lama, Gupta et al. (2016) "Hepatitis E virus outbreak in postearthquake nepal: is a vaccine really needed?" *J Viral Hepat*
24. Gupta, Adhikari, Chaudhary (2018) "Hepatitis viruses in Kathmandu, nepal: Hospital-based study" *BMC Res Notes*
25. Wen, Tang, Yang et al. (2014) "A valuable antigen detection method for diagnosis of acute hepatitis E" *J Clin Microbiol*
26. Diagnostics (2025) "MagNA Pure 24 Total NA Isolation Kit"
27. Wang, Gupta, Ji et al. (2018) "Application of hepatitis E Virus-Related markers on samples from a developing country" *Clin Lab*
28. Wang, Ling, Erker et al. (0169) "A divergent genotype of hepatitis E virus in Chinese patients with acute hepatitis" *J Gen Virol*
29. Sanger, Coulson (1975) "A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase" *J Mol Biol*
30. Shrestha, Flower, Seed et al. (2016) "Hepatitis E virus seroepidemiology: A post-earthquake study among blood donors in Nepal" *BMC Infect Dis*
31. Shrestha, Shrestha, Tsuda et al. (2003) "Molecular investigation of hepatitis E virus infection in patients with acute hepatitis in Kathmandu, Nepal" *J Med Virol*
32. Vollmer, Knabbe, Dreier (2014) "Comparison of real-time PCR and antigen assays for detection of hepatitis e virus in blood donors" *J Clin Microbiol*
33. Gupta, Pandey, Pandey et al. (2013) "Role of hepatitis E virus antigen in confirming active viral replication in patients with acute viral hepatitis E infection" *J Clin Virol*
34. Dalton, Kamar, Baylis et al. (2018) "EASL clinical practice guidelines on hepatitis E virus infection" *J Hepatol*
35. Talapko, Meštrović, Pustijanac et al. (2021) "Towards the improved accuracy of hepatitis E diagnosis in vulnerable and target groups: A global perspective on the current state of knowledge and the implications for practice" *Healthcare*
36. Lu, Huang, Wang et al. (2021) "Dynamics of hepatitis E virus (HEV) antibodies and development of a multifactorial model to improve the diagnosis of HEV infection in resource-limited settings" *J Clin Microbiol*
37. Wagle, Baniya, Aryal et al. (2025) "HEV Seroprevalence and associated risk factors among HIV-positive individuals in postearthquake kathmandu: a 2016 cross-sectional study" *BMC Infect Dis*
39. (1925)
40. Shrestha, Adhikari, Bhattarai et al. (2017) "Prevalence and risk of hepatitis E virus infection in the HIV population of Nepal" *Virol J*
42. 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* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12844545&blobtype=pdf | # Comparative evaluation of the antibacterial effect of ultraviolet radiation on alginate and condensation silicone impressions compared to hypochlorite
Farshad Bajoghli, Farzaneh Rostami, Hossein Gugunani, Arezoo Nazarifar
## Abstract
Background:Dental impressions are a known potential vector for cross-contamination between patients and the dental laboratory. Effective disinfection is, therefore, a critical step in infection control protocols. This in vitro study aimed to evaluate and compare the antibacterial efficacy of ultraviolet (UV) radiation and 0.525% sodium hypochlorite solution for disinfecting two common impression materials: condensation silicone and alginate. Materials and Methods: This in vitro study employed a comparative experimental design to evaluate disinfection efficacy. A total of 195 samples were utilized, comprising 90 discs each of condensation silicone and alginate, alongside positive and negative controls (n = 9 and n = 6, respectively). All samples were experimentally contaminated with standardized suspensions of three pathogenic species: Streptococcus pyogenes (beta-hemolytic Group A), Pseudomonas aeruginosa, and Staphylococcus aureus. The disinfection protocols consisted of either exposure to 0.525% sodium hypochlorite spray for 10 min or treatment with UV radiation using a dedicated device ("Fast Steril"). Antibacterial efficacy was quantitatively assessed by enumerating the mean colony-forming units (CFUs) postdisinfection. Statistical analysis was performed using the Kruskal-Wallis and Mann-Whitney U-tests, with the significance level defined at α = 0.05. Results: The analysis revealed a statistically significant difference in bacterial reduction based on the microbial species (P < 0.001). UV radiation demonstrated superior efficacy compared to sodium hypochlorite in disinfecting condensation silicone impressions (P < 0.05). Conversely, no significant difference was observed between the two disinfection methods for alginate impressions. Regarding bacterial susceptibility, the mean reduction in CFUs for S. pyogenes was significantly greater than for S. aureus and P. aeruginosa (P < 0.001), between which no significant difference was found (P = 1.0).
Conclusion:Within the limitations of this study, UV radiation proved to be a more effective disinfection method for condensation silicone impressions than sodium hypochlorite spray. For alginate impressions, both methods were equally effective. Given its efficacy and the superior dimensional stability of UV-treated impressions reported in the literature, the adoption of UV radiation is recommended as a viable and efficient method for disinfecting both
## INTRODUCTION
Dental impression making is a critical step in restorative treatment, providing a precise three-dimensional replica of the oral structures. [1] This replica allows for the fabrication of restorations that reconstruct tooth form and function and enables laboratory work to proceed in the patient's absence. [2,3] Dental impressions are a potential source of infection in prosthetic workflow and can lead to the transmission of infection, especially in individuals with weakened immune systems. [4,5] For this reason, the American Dental Association (ADA) and the Centers for Disease Control and Prevention have published guidelines for the disinfection of dental impressions. [6] All patients should be considered potential carriers, and their impressions should be handled similarly to those of a high-risk patient. [7] Rinsing under water cannot completely remove saliva and blood from the impression surface because salivary mucins and adhesive salivary proteins interfere with simple washing. [8] Therefore, a suitable method for disinfecting dental impressions is essential. Since impressions cannot be sterilized by heat, chemical disinfection is the most common disinfection method. Chemical disinfection is divided into two methods: immersion and spraying. [7] The hydrophilic nature of the materials, the presence or absence of surfactants, and their tolerance to immersion in water or other fluids are key elements in selecting the appropriate chemical protocol for impression materials. [9,10] To date, a global standard method for disinfecting impression materials has not been achieved. [11,12] An alternative disinfection strategy employs ultraviolet (UV) radiation. The efficacy of UV light is contingent upon several factors, including exposure duration, intensity, ambient humidity, and the requirement for direct line-of-sight to the microbial organisms. Furthermore, its application is constrained by significant limitations: the need for multiangular exposure to ensure comprehensive surface coverage and the imperative to remove organic debris from the impression before treatment to achieve optimal efficacy. However, the nonchemical nature of this method, the lack of dimensional changes in the impression, and its broad spectrum of effectiveness are advantages of this method. [13] Therefore, given the critical role of impression disinfection in preventing cross-contamination and the potential advantages of UV radiation such as avoiding dimensional change and chemical residue over conventional chemical disinfectants, this in vitro study aimed to comparatively evaluate the antibacterial efficacy of a specific UV radiation device ("Fast Steril") against the standard chemical agent, 0.525% sodium hypochlorite, on two widely used impression materials: Condensation silicone and alginate.
## MATERIALS AND METHODS
## Study design
This in vitro investigation utilized a comparative experimental design to assess disinfection efficacy across two impression materials and three bacterial species.
## Ethical approval and study design
## Sample preparation and experimental groups
The experimental design incorporated 195 specimens distributed across four categories: 90 alginate disks, 90 condensation silicone disks, 9 positive control disks (allocated equally among three bacterial species), and 6 negative control disks (assigned proportionally to assess both impression material types). This configuration enabled comprehensive evaluation of both material-specific characteristics and disinfection efficacy across experimental conditions. Group A beta-hemolytic Streptococcus pyogenes ATCC 19615 were prepared for contamination of the impression material disks. Each bacterial strain was initially streaked onto blood agar plates using sterile swabs and incubated aerobically at 37°C for 24-48 h. Following incubation, isolated colonies were transferred to test tubes containing Tryptic Soy Broth and subjected to secondary incubation at 37°C until achieving a turbidity equivalent to the 0.5 McFarland standard, indicating a concentration of approximately 1.5 × 10⁸ colony-forming units (CFUs)/mL. The alginate and condensation silicone disks were systematically contaminated by immersion in these standardized bacterial suspensions before comparative evaluation of sodium hypochlorite and UV irradiation disinfection protocols. [13,14]
## Sample fabrication and sterilization protocol
All instrumentation utilized in this study underwent sterilization through autoclave treatment at 121°C and 15 PSI for 20 min before sample preparation. Alginate (Iralgin, Golchai, Iran) and condensation silicone (Sildent, Lascod S. P. A., Florence, Italy) were manipulated in strict accordance with manufacturer specifications. The materials were subsequently cast into specialized metal molds to generate standardized disks measuring 30 mm in diameter and 7 mm in height.
A total of 180 experimental disks were fabricated (90 per material type). These specimens were systematically randomized into six experimental groups (n = 15 per group) for each bacterial species, with additional allocations for positive and negative control groups to ensure methodological rigor. This allocation strategy enabled precise comparison of disinfection efficacy across both material types and microbial challenges.
## Contamination and disinfection protocol
The experimental disks were subjected to controlled contamination by immersion in individual containers housing standardized bacterial suspensions (0.5 McFarland standard) of S. aureus ATCC 25923, P. aeruginosa ATCC 27853, or Group A beta-hemolytic S. pyogenes ATCC 19615 for a duration of 2 min. Following contamination, each disk was aseptically retrieved using sterile forceps and underwent an initial rinsing procedure consisting of 60 mL sterile distilled water applied for 30 s to remove nonadherent bacteria. The disinfection phase employed two distinct methodologies: one group of disks received chemical disinfection through complete surface spraying with 0.525% sodium hypochlorite solution followed by a 10-min contact time, while the second group underwent physical disinfection using a GermGuardian Portable UVC Wand ("Fast Steril") (Guardian Technologies LLC, Euclid, Ohio, USA) maintained at a standardized distance of 1 inch (2.54 cm) from the surface for 10 s of continuous exposure. [13]
## Microbiological assessment and quality control
All disks received a final rinse with 60 mL sterile distilled water for 30 s following disinfection procedures. Microbial sampling was performed by systematically swabbing the entire surface of each disk with a sterile dry swab, which was subsequently streaked in a linear pattern onto blood agar plates. All plates underwent aerobic incubation at 37°C for 48 h. CFUs were enumerated manually following the incubation period. To eliminate observational bias, the microbiologist performing colony counts was blinded to group assignments throughout the enumeration process. The UV irradiation disinfection methodology is visually documented in Figure 1.
## Statistical analysis
Quantitative data analysis was performed using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean ± standard deviation. The Kolmogorov-Smirnov test confirmed nonnormal distribution of the data, and Levene's test indicated violation of homogeneity of variances. In addition, a significant interaction effect was observed between the independent variables. Consequently, nonparametric analyses were conducted using the Kruskal-Wallis test for overall group comparisons, followed by pairwise Mann-Whitney U-tests with Bonferroni adjustment for multiple comparisons. A significance level of α = 0.05 was applied for all statistical tests.
## RESULTS
The validation of experimental conditions was confirmed by the control groups. All positive control samples demonstrated 100% microbial growth, while all negative control samples maintained 100% sterility throughout the study. The quantitative assessment of disinfection efficacy is presented in Figure 2, which illustrates the mean CFU counts and corresponding standard deviations for three microbial species on both alginate and condensation silicone impression materials following application of two disinfection protocols: UV irradiation and sodium hypochlorite treatment. The bar chart provides a comparative visualization of the bacterial reduction achieved by each disinfection method across both material types.
## Statistical analysis of microbial reduction
The Kruskal-Wallis test demonstrated a statistically significant difference in colony counts among the three microbial species (P < 0.001). Post hoc analysis using Bonferroni-corrected Mann-Whitney U-tests revealed that S. pyogenes (Group A beta-hemolytic) showed significantly different susceptibility compared to both P. aeruginosa (P < 0.001) and S. aureus (P < 0.001). However, no significant difference was observed between S. aureus and P. aeruginosa (P = 1.000).
Regarding material characteristics, statistical analysis indicated a significant overall difference in disinfection efficacy between the two impression materials (P = 0.002), with alginate demonstrating greater resistance to disinfection protocols compared to condensation silicone.
Furthermore, a significant difference was observed between the two disinfection methods (P < 0.001), with UV radiation demonstrating superior antimicrobial efficacy compared to 0.525% sodium hypochlorite solution across all experimental conditions.
## DISCUSSION
Dental impressions frequently come into contact with blood and saliva, which can harbor pathogenic microorganisms capable of transmitting infectious diseases. This risk of cross-contamination underscores the need for stringent infection control measures throughout impression-making and subsequent laboratory processing. [15] The present study evaluated the antibacterial efficacy of UV irradiation on condensation silicone and alginate impression materials in comparison with 0.525% sodium hypochlorite. The selection of test microorganisms was guided by their clinical relevance, high pathogenicity, and documented resistance to disinfectants. Pseudomonas aeruginosa presents a significant cross-infection risk in dental environments due to its intrinsic antibiotic resistance and potential to cause nosocomial infections. Staphylococcus aureus was included as a benchmark organism for disinfectant efficacy testing owing to its widespread antibiotic resistance. S. pyogenes (Group A beta-hemolytic) was selected for its established pathogenic role in oral and systemic infections. [12] Although this study focused on highly resistant pathogenic strains, future research should incorporate representative members of the normal oral microbiota to enhance clinical generalizability.
Two disinfection methods were evaluated for impression materials: UV irradiation using a fast-sterilizer device and spray application of 0.525% sodium hypochlorite with a 10-min contact time. Although the ADA recommends immersion in 0.5% sodium hypochlorite (5000 ppm free chlorine) for 10 min, [6] and manufacturers report 99.8% tuberculocidal efficacy with this protocol, [16] the spray method was selected for this investigation due to concerns regarding dimensional instability associated with immersion techniques. [17] While immersion remains the gold standard for disinfectant reliability, [12,18,19] spraying represents a clinically acceptable alternative that minimizes potential material distortion.
The present study also evaluated UV-C (UVC) irradiation as a disinfection alternative using a fast-sterilization device. The biocidal mechanism of UVC radiation primarily involves induction of genomic damage through thymine dimer formation in microbial DNA, leading to irreversible inactivation of pathogens. [20] Notable advantages of UVC over chemical disinfectants include exceptional preservation of impression dimensional accuracy and complete avoidance of chemical residue on material surfaces. [21] While this technology demonstrates well-documented efficacy in surface disinfection, it also shows promising applications in endodontic therapy, including root canal disinfection and management of periapical inflammatory conditions. These findings are consistent with previous research by Ishida et al., [13] who reported complete eradication of Candida species on silicone impression materials following 5 min of UVC exposure, with no statistically significant alterations in dimensional stability or surface characteristics. In a 2019 investigation, Nimunkar et al. [22] evaluated the dimensional stability of polyvinyl siloxane impressions following disinfection using 2% glutaraldehyde, 1% sodium hypochlorite, and UV irradiation. Their findings indicated that UV irradiation, in contrast to chemical disinfectants, produced no measurable dimensional alterations. This is consistent with a body of research documenting dimensional changes resulting from chemical disinfection of dental impressions, [23,24] although some studies have reported no significant effects on dimensional stability from disinfection procedures. [25,26] Alginate, a representative irreversible hydrocolloid, remains one of the most frequently utilized dental impression materials owing to its user-friendly application, procedural simplicity, and cost-effectiveness. [27] Nevertheless, its inherent hydrophilicity increases its susceptibility to microbial retention, while its dimensional accuracy and stability are notably compromised upon exposure to liquid disinfectants. [28] Condensation silicones, similarly employed in routine prosthetic impression procedures, represent another mainstream material in clinical dentistry. [27] Based on their prevalence and distinct material characteristics, these two impression materials were selected for the current investigation. Results demonstrated that UV irradiation yielded superior disinfection efficacy compared to sodium hypochlorite solution when applied to condensation silicone impressions. Furthermore, a statistically significant difference was observed between the two materials, with alginate exhibiting reduced susceptibility to disinfection, a phenomenon likely attributable to its heightened porosity and consequent increased potential for microbial entrapment.
In a 2010 comparative study, Samra et al. evaluated the disinfection efficacy of UV irradiation versus sodium hypochlorite on alginate and silicone impression materials. Their findings indicated that UV chamber disinfection yielded superior results compared to hypochlorite a conclusion consistent with the present study regarding silicone materials, though not observed with alginate. Further supporting the utility of UV irradiation, Aran et al. demonstrated its potential for significantly reducing colony counts of oral pathogens on various patient-derived impression materials, including alginate, addition silicone, and polyether. However, as their study utilized clinical impressions, the precise microbial composition remained uncharacterized. Notably, Aran et al. also reported that impression material type did not influence the efficacy of UV disinfection. [29] The observed differential efficacy of disinfection between alginate and silicone impression materials in the present study may be attributed to alginate's characteristically porous microstructure and its reported capacity for two to three times greater microbial absorption compared to silicone. This inherent property may necessitate extended disinfection durations beyond the 10-s UV exposure protocol employed herein. Furthermore, the use of a multidirectional UV chamber as opposed to the single-angle portable device utilized in this study may provide more uniform irradiation and enhance disinfection outcomes. It is also noteworthy that discrepancies between our results and those of earlier studies may stem from differences in microbial strains; prior investigations predominantly used normal oral flora, whereas the present study employed standardized ATCC strains with well-defined profiles. These findings are nevertheless consistent with recent work by Wezgowiec et al., [30] who demonstrated the effectiveness of UV irradiation in disinfecting both condensation and addition silicones of varying consistencies against P. aeruginosa, S. aureus, and Candida albicans, further supporting the utility of UV-based disinfection in dental practice.
One notable finding of this study was the significantly higher colony counts observed for S. aureus and P. aeruginosa compared to Group A beta-hemolytic Streptococcus (GAS). This differential efficacy may be attributed to the higher intrinsic resistance of S. aureus and P. aeruginosa both recognized as resilient nosocomial pathogens to various disinfection methods and antibiotics when compared to GAS. [31,32] The effective elimination of resistant nosocomial pathogens such as S. aureus and P. aeruginosa suggests potential applications of the UV disinfection device beyond dental settings, including hospital environments where such pathogens pose significant challenges to infection control. However, this extrapolation requires further validation through targeted clinical studies. In conjunction with the established advantage of superior dimensional stability reported in literature when using UV irradiation compared to chemical alternatives, the findings of this study support the conclusion that UV irradiation presents a superior alternative to chemical disinfectants for both condensation silicone and alginate impression materials.
## CONCLUSION
Based on the results of this study, UVC irradiation demonstrated superior disinfection efficacy compared to sodium hypochlorite for condensation silicone impressions, while both methods showed comparable results for alginate. The differential efficacy between materials highlights the influence of material composition and porosity on disinfection outcomes. Given its minimal impact on dimensional stability and clinical practicality, UV irradiation is recommended as a preferable disinfection method for dental impression materials.
## References
1. Punj, Bompolaki, Garaicoa (2017) "Dental impression materials and techniques. Dental Clinics"
2. Vázquez-Rodríguez, Estany-Gestal, Seoane-Romero et al. (2018) "Quality of cross-infection control in dental laboratories. A critical systematic review" *International Journal for Quality in Health Care*
3. Sedky (2019) "A comparative study of practicing cross-infection control of dental prostheses and implant components among prosthodontists and dental technicians in Qassim province, Saudi Arabia" *International Journal of Infection Control*
4. Napoli, Galle, Montagna et al. (2007) "Guidelines for infection control practices for dentistry" *Ann Igiene Med Prev Comunita*
5. Haralur, Os, Gana et al. (2012) "Effect of alginate chemical disinfection on bacterial count over gypsum cast" *J Adv Prosthodont*
6. Kohn, Collins, Cleveland et al. (2003) "Guidelines for infection control in dental health-care settings-2003" *MMWR Recomm Rep*
7. Gounder, Vikas (2016) "Comparison of disinfectants by immersion and spray atomization techniques on the linear dimensional stability of different interocclusal recording materials: An in vitro study" *European journal of dentistry*
8. Hardan, Bourgi, Cuevas-Suárez et al. (2022) "Disinfection procedures and their effect on the microorganism colonization of dental impression materials: a systematic review and meta-analysis of in vitro studies" *Bioengineering*
9. Al-Jabrah, Al-Shumailan, Al-Rashdan (2007) "Antimicrobial effect of 4 disinfectants on alginate, polyether, and polyvinyl siloxane impression materials" *International Journal of Prosthodontics*
10. Donovan, Chee (2004) "A review of contemporary impression materials and techniques. Dental Clinics"
11. Devine, Keech, Wood et al. (2001) "Ultraviolet disinfection with a novel microwave-powered device" *Journal of Applied Microbiology*
12. Memarian, Fazeli, Jamalifar et al. (2007) "Disinfection efficiency of irreversible hydrocolloid impressions using different concentrations of sodium hypochlorite: A pilot study" *J Contemp Dent Pract*
13. Ishida, Nahara, Tamamoto et al. (1991) "The fungicidal effect of ultraviolet light on impression materials" *J Prosthet Dent*
14. Shahraki, Mohammadzadeh-Rostami, Haddadi-Feishani et al. (2015) "Evaluation of quinolone-resistant strains of Klebsiella pneumoniae in clinical specimens obtained from patients referred to Zahedan educational hospitals" *Zahedan J Res Med Sci*
15. Godbole, Dahane, Patidar et al. (2014) "Evaluation of the Effect of Ultraviolet Disinfection on Dimensional Stability of the Polyvinyl Silioxane Impressions" *Journal of Clinical and Diagnostic Research: JCDR*
16. Ahila, Thulasingam (2014) "Effect of disinfection on gypsum casts retrieved from addition and condensation silicone impressions disinfected by immersion and spray methods" *SRM J Res Dent Sci*
17. Hamedi Rad, Ghaffari, Safavi (2010) "In vitro evaluation of dimensional stability of alginate impressions after disinfection by spray and immersion methods" *J Dent Res Dent Clin Dent Prospects*
18. Samra, Bhide (2010) "Efficacy of different disinfectant systems on alginate and addition silicone impression materials of Indian and international origin: A comparative evaluation" *J Indian Prosthodont Soc*
19. Bustos, Herrera, González et al. (2010) "Effect of inmersion desinfection with 0.5% sodium hypochlorite and 2% glutaraldehyde on alginate and silicone: Microbiology and SEM study" *Int J Odontostomat*
20. Malpartida-Carrillo, Tinedo-López, Salas-Quispe et al. (2024) "Effect of ultraviolet C light disinfection on the dimensional stability of dental impression materials: A scoping review of the literature" *J Clin Exp Dent*
21. Tseng, Li (2007) "Inactivation of viruses on surfaces by ultraviolet germicidal irradiation" *J Occup Environ Hyg*
22. Nimonkar, Belkhode, Godbole et al. (2019) "Comparative evaluation of the effect of chemical disinfectants and ultraviolet disinfection on dimensional stability of the polyvinyl siloxane impressions" *J Int Soc Prev Community Dent*
23. Michalakis, Bakopoulou, Hirayama et al. (2007) "Pre-and post-set hydrophilicity of elastomeric impression materials" *Journal of Prosthodontics: Implant, Esthetic and Reconstructive Dentistry*
24. Carvalhal, Mello, Sobrinho et al. (2011) "Dimensional change of elastomeric materials after immersion in disinfectant solutions for different times" *J Contemp Dent Pract*
25. Sabanov, Dostinova, Elencevski et al. (2024) "Methods of assessment of dimensional stability of elastomeric impression materials after disinfection: A literature review" *Acta Medica Medianae*
26. Cohn, Owiredu, Taylor et al. (2021) "Eliminating mother-to-child transmission of human immunodeficiency virus, syphilis and hepatitis B in sub-Saharan Africa" *Bulletin of the World Health Organization*
27. Grădinaru, Ciubotaru, Dascălu et al. (2022) "Alginate Dental Impression Materials with Allantoin Enrichment: A Morphology, Dynamic Vapor Sorption And Swelling Evaluation"
28. Shinde, Borle, Dahane et al. (2021) "Comparative Evaluation of Antimicrobial Activity of Irreversible Hydrocolloid Impression Material Incorporated with Various Disinfectants-An In-vitro Study" *Journal of Pharmaceutical Research International*
29. Aeran, Sharma, Kumar et al. (2015) "Use of clinical UV chamber to disinfect dental impressions: A comparative study" *Journal of clinical and diagnostic research*
30. Wezgowiec, Wieczynska, Wieckiewicz et al. (2022) "Evaluation of antimicrobial efficacy of UVC radiation, gaseous ozone, and liquid chemicals used for disinfection of silicone dental impression materials" *Materials*
31. Einabadi, Abdolerahmani, Mashoof et al. (2018) "Study of drug resistance of Staphylococcus aurous and Pseudomonas aeruginosa strains isolated from environmental samples of Hamadan educational hospitals in 2017 using disk diffusion and minimum inhibitory concentration" *Feyz Med Sci J*
32. Rostami, Shalibeik, Mousavi (2020) "Molecular Characterization and Antibiotic Resistance Pattern of Nosocomial Clinical Isolates in Southeast of Iran" *Medical Laboratory Journal* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12505971&blobtype=pdf | # | Molecular and Cellular Biology | Perspective mGem: Guides or triggers? Extracellular RNAs beyond vesicular miRNAs
Juan Tosar, Amy Buck
## Abstract
Despite a huge expansion in the last decades, several assumptions have directed, and perhaps pigeonholed, the evolution of the extracellular RNA (exRNA) field. For example, extracellular vesicles (EVs) have been assumed to be the main carriers of RNA molecules between cells. In parallel, microRNAs (miRNAs) have been assumed to be the main EV RNA cargo. However, from mammals to microbes, these assumptions do not seem to fall out of the data. In addition, miRNAs need to localize to the cytosol to be active but are likely to start in endosomes in most EV entry pathways. The mecha nisms for their endosomal escape and the quantities of imported miRNAs required for their functions are not always considered. Without questioning the empirical evidence supporting EV-miRNA-mediated intercellular communication, we would like to shed light on the overlooked aspects of the exRNA biology that may bear important insights into how cells and organisms interact and sense one another.
report of membraneless extracellular nanoparticles, such as exomeres and supermeres (10,30). Nonvesicular Argonautes complexed with siRNAs are also released from parasitic nematodes and can enter host cells (31). Other large ribonucleoprotein complexes, such as extracellular ribosomes (8), can also give place to stable protein/RNA complexes upon extracellular fragmentation (32). In addition, some RNAs might achieve extracel lular stability and abundance, thanks to their compact three-dimensional structures stabilized mostly by RNA:RNA interactions (33,34). These stable "naked" RNAs can also be internalized and sensed by immune cells, playing a role in intercellular communica tion, especially in the context of inflammation (35). In retrospect, the assumption that all extracellular samples are characterized by a potent RNase activity might be a generali zation of experiments done in either blood or in cell culture, which usually contains serum (e.g., FBS) as an additive. When naked extracellular RNA is added to physiological compartments with a low RNase content, such as the peritoneal cavity, its capacity to trigger inflammatory responses becomes evident (35).
Thus, the popularity of EVs and miRNAs is better explained by the chronology of the exRNA field and the large focus on mammalian systems, rather than emerging from the data itself. For example, the discovery of EV-miRNAs occurred at a time when the interest in miRNAs as gene expression regulators was at its peak, bolstered by a mechanistic framework for how miRNAs can regulate gene expression. A few years afterward, next-generation sequencing became widely accessible, and, consequently, many studies started to explore the small RNA content of EVs. However, small RNA sequencing was designed for miRNA identification and is highly biased toward this small RNA class (36). Without enzymatic treatment of RNAs, or modifications to protocols to read through structured RNAs, many of the exRNAs present in a sample are never detected.
What is, then, the picture that emerges from the data? While we may not yet have a definitive answer to this question, we would like to propose four additional questions that might illuminate the path forward (Fig. 1).
## What RNA types are most abundant in the extracellular space both within and outside EVs?
To date, most analyses have focused on guide-acting small RNAs (miRNAs, siRNAs, short tRNA fragments) or messenger RNAs (mRNAs) where there is a mechanistic framework for how they could function and existing methodologies for validation, for example, with reporter assays. However, the data across diverse organisms show that other classes of RNA (tRNAs, Y RNAs, SRP RNA, vault RNAs, rRNAs) are often more abundant both inside and outside of EVs, and there remain few investigations into their transmission or function within recipient cells (Fig. 1A).
## What are we missing due to technical or conceptual limitations but that could be relevant to understanding exRNA functionality?
The most "unbiased" method to answer this question is sequencing, but the library generation protocols and methods of analysis will direct the answer. An example is the focus in the literature on fragments of tRNAs or Y RNAs in EVs based on sequencing, rather than a focus on full-length forms that are excluded when making the library and only become revealed by northern blots or specialized sequencing techniques (e.g., ARM-seq, TGIRT-seq, and hydro-tRNA-seq) (8,15). RNA modifications and the balance between RNases and RNase inhibitors can also strongly influence what gets sequenced and what is not. For a recent review on biases in small RNA sequencing and their impact on exRNA profiling, see reference 36. Finally, the analysis methods define what is kept as useful information or discarded. For example, often, sequences that map to repetitive regions in the genome are thrown out, even though this is a ubiquitous and poorly understood source of exRNA across eukaryotes (11,37,38) that is also prevalent in host-pathogen exRNA interactions (17). We are still stuck in many ways looking for what we already understand and expect (Fig. 1B).
## What exRNAs are most likely to play a role in intercellular communication based on their proposed mechanism of action, stability, and abundance?
While the focus has been on miRNAs due to the existing mechanistic framework for how they function in the cytoplasm of cells, the framework for how they make it to the cytoplasm after entering the cell is lacking. If EVs are internalized by endocytosis, EV-miRNAs could be released into the cytosol of a recipient cell after the fusion of the EV lipid bilayer with the endosomal membrane (39). However, this seems to be quite an inefficient process in mammalian studies (12,(40)(41)(42), and more work is needed to understand when/where/how escape can occur in different contexts and how small RNAs subsequently end up in the right complexes (43). A simpler mechanism for some exRNAs could occur within endosomes (Fig. 2), where exRNAs can activate RNA-specific Toll-like receptors (TLRs) (44). This has been shown for vesicular tRNA-derived fragments (45,46) that are prevalent and abundant across EVs from Bacteria, Archaea, and Eukarya. It is even tempting to speculate that microbiome-derived exRNAs might have shaped the evolution of mammalian endosomal RNA sensors. For example, a highly stable bacterial rRNA-derived fragment can be spontaneously internalized by murine immune cells, even when present in culture media as a naked RNA, and is a potent trigger of endosomal TLR13 (35) (Fig. 1C).
## Could there be important non-cell autonomous functions of some non-cod ing RNAs that have defined their evolution?
Over the last 50+ years, the field of RNA biology has built quantitative and mechanistic data on how diverse RNAs function inside cells. The field of exRNA has evolved after (and largely separate from) this. Based on this chronology, any role of an exRNA might be expected to be known already; the only details we have to work out are how it gets from donor to recipient. But, what if the functions of some RNAs can only be well understood if we account for their roles outside the cell? What have we been missing? For example, Y RNAs are abundant in extracellular samples (7,11,47), including in human biofluids (48)(49)(50), but their intracellular roles are still not fully understood despite being discovered more than 40 years ago. In fact, these RNAs were originally discovered in the extracellular space as the RNA component of a ribonucleoprotein particle targeted by self-reactive antibodies (51) (Fig. 1D).
## FIG 2
Guides or triggers? Subcellular localization restricts or enables functional possibilities. Both vesicular and nonvesicular exRNAs can be internalized by endocytosis if not degraded by extracellular RNases. Once inside endosomes, vesicular RNAs need to escape into the cytosol to recognize specific targets in a sequence-dependent manner ("guides"; e.g., miRNAs and siRNAs that function in gene silencing with Argonaute proteins). This is thought to occur by fusion of the EVs with the endosomal membrane. However, vesicular RNAs can still be functional in the absence of efficient endosomal escape, for example, by activating RNA sensors localized inside endosomes ("triggers"). The same considerations apply to nonvesicular exRNAs, which are directly exposed in the endosomal lumen. Note that exRNAs can also act as triggers in the cytosol if they are recognized by a protein or by another nucleic acid in a structure-dependent but sequence-independent manner (e.g., RIG-I). Thus, "guides vs triggers" is a distinction based on the mechanism of action rather than in subcellular localization, but sequence-dependent mRNA recognition is thought to occur exclusively in the cytosol.
Why is it worth pushing this field forward? The RNA world drove the evolution of life, and it would be bizarre for its innovative power to be restricted to within the cell membrane. Indeed, most aspects of life require interaction outside the cell, yet our understanding of the different roles of exRNAs in living systems is still incredibly limited. The exRNA field remains ripe for discovery if we can evolve our technologies and minds to build a foundation of knowledge that includes things we may not already expect.
## References
1. Ratajczak, Miekus, Kucia et al. (2006) "Embryonic stem cell-derived microvesicles reprogram hemato poietic progenitors: evidence for horizontal transfer of mRNA and protein delivery" *Leukemia*
2. Skog, Würdinger, Van Rijn et al. (2008) "Glioblas toma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers" *Nat Cell Biol*
3. Valadi, Ekström, Bossios et al. (2007) "Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells" *Nat Cell Biol*
4. Arroyo, Chevillet, Kroh et al. (2011) "Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma" *Proc Natl Acad Sci*
5. Geekiyanage, Rayatpisheh, Wohlschlegel et al. (2020) "Extracellular microRNAs in human circulation are associated with miRISC complexes that are accessible to anti-AGO2 antibody and can bind target mimic oligonucleotides" *Proc Natl Acad Sci*
6. Sork, Conceicao, Corso et al. (2021) "Profiling of extracellular small RNAs highlights a strong bias towards non-vesicular secretion" *Cells*
7. Tosar, Gámbaro, Sanguinetti et al. (2015) "Assessment of small RNA sorting into different extracellular fractions revealed by high-throughput sequencing of breast cell lines" *Nucleic Acids Res*
8. Tosar, Segovia, Castellano et al. (2020) "Fragmenta tion of extracellular ribosomes and tRNAs shapes the extracellular RNAome" *Nucleic Acids Res*
9. Turchinovich, Weiz, Langheinz et al. (2011) "Characterization of extracellular circulating microRNA" *Nucleic Acids Res*
10. Zhang, Jeppesen, Higginbotham et al. (2021) "Supermeres are functional extracellular nanoparticles replete with disease biomarkers and therapeutic targets" *Nat Cell Biol*
11. Nolte-'t Hoen, Buermans, Waasdorp et al. (2012) "Deep sequencing of RNA from immune cellderived vesicles uncovers the selective incorporation of small noncoding RNA biotypes with potential regulatory functions" *Nucleic Acids Res*
12. Albanese, Chen, Hüls et al. (2021) "MicroRNAs are minor constituents of extracellular vesicles that are rarely delivered to target cells" *PLoS Genet*
13. Chevillet, Kang, Ruf et al. (2014) "Quantitative and stoichiometric analysis of the microRNA content of exosomes" *Proc Natl Acad Sci*
14. Wei, Batagov, Schinelli et al. (2017) "Coding and noncoding landscape of extracellular RNA released by human glioma stem cells" *Nat Commun*
15. Shurtleff, Yao, Qin et al. (2017) "Broad role for YBX1 in defining the small noncoding RNA composition of exosomes" *Proc Natl Acad Sci*
16. Buck, Coakley, Simbari et al. (2014) "Exosomes secreted by nematode parasites transfer small RNAs to mammalian cells and modulate innate immunity" *Nat Commun*
17. Chow, Koutsovoulos, Ovando-Vázquez et al. (2019) "Secretion of an argonaute protein by a parasitic nematode and the evolution of its siRNA guides" *Nucleic Acids Res*
18. Quintana, Kumar, Ivens et al. (2019) "Comparative analysis of small RNAs released by the filarial nematode Litomosoides sigmodontis in vitro and in vivo" *PLoS Negl Trop Dis*
19. White, Kumar, Chow et al. (2020) "Extracellular vesicles from Heligmoso moides bakeri and Trichuris muris contain distinct microRNA families and small RNAs that could underpin different functions in the host" *Int J Parasitol*
20. Cai, Qiao, Wang et al. (2018) "Plants send small RNAs in extracellular vesicles to fungal pathogen to silence virulence genes" *Science*
21. Borniego, Singla-Rastogi, Baldrich et al. (2025) "Diverse plant RNAs coat Arabidopsis leaves and are distinct from apoplastic RNAs" *Proc Natl Acad Sci*
22. Ravet, Zervudacki, Singla-Rastogi et al. (2024) "Vesicular and non-vesicular extracellular small RNAs direct gene silencing in a plant-interacting bacterium" *bioRxiv. htt*
23. Karimi, Baldrich, Rutter et al. (2022) "Arabidopsis apoplastic fluid contains sRNA-and circular RNA-protein complexes that are located outside extracellular vesicles" *Plant Cell*
24. Alves, Da Silva, Sanchez et al. (2019) "Extracellular vesiclemediated RNA release in Histoplasma capsulatum" *mSphere*
25. Bayer-Santos, Lima, Ruiz et al. (2014) "Characterization of the small RNA content of Trypanosoma cruzi extracellular vesicles" *Mol Biochem Parasitol*
27. Fernandez-Calero, Garcia-Silva, Pena et al. (2015) "Profiling of small RNA cargo of extracellular vesicles shed by Trypanosoma cruzi reveals a specific extracellular signature" *Mol Biochem Parasitol*
28. Ghosal, Upadhyaya, Fritz et al. (2015) "The extracellular RNA complement of Escherichia coli" *Microbiologyopen*
29. Koeppen, Hampton, Jarek et al. (2016) "A novel mechanism of host-pathogen interaction through sRNA in bacterial outer membrane vesicles" *PLoS Pathog*
30. Mills, Gebhard, Schubotz et al. (2024) "Extracellular vesicle formation in Euryarchaeota is driven by a small GTPase" *Proc Natl Acad Sci*
31. Zhang, Freitas, Kim et al. (2018) "Identification of distinct nanoparticles and subsets of extracellular vesicles by asymmetric flow field-flow fractionation" *Nat Cell Biol*
32. Neophytou, Martínez-Ugalde, Fenton et al. (2025) "A non-vesicular argonaute protein is transmitted from nematode to mouse and is important for parasite survival"
33. Tosar, Cayota, Witwer (2022) "Exomeres and supermeres: monolithic or diverse?" *J Extracell Biol*
34. Tosar, Gámbaro, Darré et al. (2018) "Dimerization confers increased stability to nucleases in 5' halves from glycine and glutamic acid tRNAs" *Nucleic Acids Res*
35. Costa, Calzi, Castellano et al. (2023) "Nicked tRNAs are stable reservoirs of tRNA halves in cells and biofluids" *Proc Natl Acad Sci*
36. Castellano, Blanco, Calzi et al. (2025) "Ribonuclease activity undermines immune sensing of naked extracellular RNA" *Cell Genom*
37. Tosar, Castellano, Costa et al. (2024) "Small RNA structural biochemistry in a post-sequencing era" *Nat Protoc*
38. Reggiardo, Maroli, Halasz et al. (2022) "Mutant KRAS regulates transposable element RNA and innate immunity via KRAB zinc-finger genes" *Cell Rep*
39. Reggiardo, Maroli, Peddu et al. (2023) "Profiling of repetitive RNA sequences in the blood plasma of patients with cancer" *Nat Biomed Eng*
40. O'brien, Breyne, Ughetto et al. (2020) "RNA delivery by extracellular vesicles in mammalian cells and its applications" *Nat Rev Mol Cell Biol*
41. Somiya, Kuroda (2021) "Reporter gene assay for membrane fusion of extracellular vesicles" *J Extracell Vesicles*
42. Somiya, Kuroda (2021) "Real-time luminescence assay for cytoplasmic cargo delivery of extracellular vesicles" *Anal Chem*
43. De Jong, Murphy, Mäger et al. (2020) "A CRISPR-Cas9based reporter system for single-cell detection of extracellular vesiclemediated functional transfer of RNA" *Nat Commun*
44. Buck (2022) "Cells choose their words wisely" *Cell*
45. Xiao, Driedonks, Witwer et al. (2020) "How does an RNA selfie work? EV-associated RNA in innate immunity as self or danger" *J Extracell Vesicles*
46. Pawar, Kawamura, Kirino (2024) "The tRNA Val half: a strong endogenous Toll-like receptor 7 ligand with a 5′-terminal universal sequence signature" *Proc Natl Acad Sci*
47. Pawar, Shigematsu, Sharbati et al. (2020) "Infection-induced 5′half molecules of tRNA HisGUG activate Toll-like receptor 7" *PLoS Biol*
48. Driedonks, Ressel, Ngoc Minh et al. "Nolte-'t Hoen ENM. 2024. Intracellular localisation and extracellular release of Y RNA and Y RNA binding proteins" *J Extracell Biol*
49. Hulstaert, Morlion, Avila Cobos et al. (2020) "Charting extracellular transcriptomes in the human biofluid RNA Atlas" *Cell Rep*
50. Godoy, Bhakta, Barczak et al. (2018) "Large differences in small RNA composition between human biofluids" *Cell Rep*
51. Dhahbi, Spindler, Atamna et al. (2013) "′-YRNA fragments derived by processing of transcripts from specific YRNA genes and pseudogenes are abundant in human serum and plasma" *Physiol Genomics*
52. Lerner, Boyle, Hardin et al. (1981) "Two novel classes of small ribonucleoproteins detected by antibodies associated with lupus erythematosus" *Science* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12299351&blobtype=pdf | # The Frontier of Entomo-Virology: Applications and Tools for Virus and Vector Surveillance
P Lemos, Morais Pacheco, B Nascimento, Franco Coelho, L Filho, D Dias, Leonardo Sena, S Silva, Mureb Sallum, Poliana Da, Silva Lemos, Mayron Mielly, Bruna Laís, Sena Do Nascimento, Mônica Coelho, Luciano Chaves, Franco Filho, Daniel Damous Dias, Sandro Patroca Da Silva, Maria Anice
## Abstract
The term 'entomo-virology' arose because of the confluence of entomology and virology, focused on deepening the knowledge about the interactions between vectors and viruses and the aspects that involve hosts and the environment. Based on this, entomo-virological surveillance was proposed, aiming to develop tools that strengthen prevention for arboviral disease and vector control strategies. This review aims to present a narrative synthesis regarding the component elements of the concept of entomo-virology. In addition, the applications and tools for the surveillance of viruses and vectors, their implementation challenges, and perspectives are discussed.
## 1. Introduction
Vector-borne diseases pose a significant burden on global health. More than 80% of the world's population lives in regions where at least one vector-borne disease is endemic. Prominent examples of vector-borne diseases include malaria, dengue, lymphatic filariasis, schistosomiasis, chikungunya, onchocerciasis, Chagas disease, leishmaniasis, Zika virus disease, yellow fever, and Japanese encephalitis [1].
Arboviral diseases are a subset of vector-borne diseases caused by viral pathogens transmitted by arthropod vectors to vertebrate hosts. These viruses belong to diverse taxonomic families, including Togaviridae, Flaviviridae, Peribunyaviridae, and Rhabdoviridae [2]. Hematophagous arthropods such as mosquitoes, ticks, sand flies, and midges are vectors of several arboviruses, which can be transmitted to susceptible vertebrate hosts and cause diseases. Arboviral diseases are acute febrile illnesses with a wide range of clinical manifestations, from mild fever to severe complications such as neurological disorders, shock, congenital anomalies, and hemorrhagic fever. Fatal outcomes are not uncommon [3]. These diseases pose a significant global health challenge, with dengue alone accounting for an estimated annual economic burden of USD 8.9 billion [4]. The lack of effective vaccines against most arboviruses and specific treatments highlights the critical importance of robust surveillance and vector control strategies as the primary tools for preventing and mitigating arbovirus outbreaks [3,5].
Despite the implementation of control measures in numerous countries, the desired outcomes have often fallen short [1]. This shortfall is largely due to the lack of comprehensive policies that account for the territorial, demographic, and ecological factors shaping host-pathogen-vector interactions [1,6]. Furthermore, gaps in understanding the biology, diversity, and ecological dynamics of vector species undermine predictive modeling efforts and complicate the design and implementation of effective control strategies.
This narrative review provides an overview of entomo-virological surveillance, especially focused on mosquitoes, encompassing prevention and vector control strategies for arboviruses, in the context of advancements in virus and vector surveillance tools. We included articles published between 1 January 1994 and 30 December 2024. The literature search encompassed English, Spanish, and Portuguese publications, including peer-reviewed articles, systematic reviews, official documents, randomized controlled trials, and observational studies that utilized entomo-virological approaches for arbovirus surveillance, diagnosis, and prevention. Scientific articles not published in English, Spanish, and Portuguese, case studies, and articles that do not discuss the relationship between arboviruses and mosquitoes were excluded. This review aims to inform a diverse audience, including entomologists, virologists, public health managers, academics, and health professionals.
## 2. What Is Entomo-Virological Surveillance?
The molecular detection of arboviruses in arthropod vectors, including mosquitoes, biting midges, and ticks, is a critical component of entomo-virological surveillance. This approach involves collecting vector samples from the field and analyzing them to identify the circulation of arboviruses in the environment, enabling the early detection of potential outbreaks. It also plays a critical role in monitoring the geographic distribution and diversity of arboviruses, which is essential for designing targeted control strategies. A notable example is a study carried out in Greece by Tsioka et al. [7], which demonstrated the effectiveness of entomo-virological surveillance detecting West Nile virus (Orthoflavivirus nilense-WNV) in mosquitoes two weeks prior to the first reported human case. In regions where arboviral transmission is sporadic or produces mild symptoms, the importance of early detection becomes even more pronounced. Research conducted in India highlighted this utility when Zika virus (Orthoflavivirus zikaense-ZIKV) was detected in Aedes aegypti mosquitoes, emphasizing the vector's pivotal role in the 2016-2021 ZIKV epidemic [8], as well as in infections detected in Brazil [9]. Chikungunya virus (Alphavirus chikungunya-CHIKV) and dengue virus (Orthoflavivirus denguei-DENV) were also detected using the entomo-virological approach, reinforcing the importance of monitoring Ae. aegypti as the main arbovirus vector in urban areas of Brazil [10][11][12][13]. Studies demonstrated the importance of the entomo-virological approach by analyzing data from the largest outbreak of sylvatic yellow fever virus (Orthoflavivirus flavi-YFV) in Brazil between 2016 and 2018, demonstrating that not only the primary vectors Haemagogus janthinomys and Hg. leucocelaenus were involved in the transmission cycle, but several other species recorded YFV detection and thus must be closely monitored, such as Sabethes chloropterus, Ae. albopictus, Ae. scapularis, and Ae. serratus [14][15][16][17][18][19].
Entomo-virological data can incorporate mathematical models to predict arboviral outbreak risks, identify key drivers of viral transmission, and evaluate the environmental factors influencing viral dissemination. This integrative approach has been validated through several studies, further illustrating its potential to enhance public health preparedness and response [20][21][22][23]. Entomo-virological surveillance is also a fundamental component of integrated vector management (IVM), a comprehensive strategy that employs a combination of interventions to optimize urban vector control. The selection of specific interventions within IVM is guided by factors such as the nature of the vector-borne disease problem, cost-effectiveness, ecological sustainability, and available resources [24]. The urgent need for implementing IVM is underscored by the emergence of insecticide resistance, as well as the globalization-driven introduction of novel vectors and pathogens. Expanding the application of molecular protocols in entomo-virological surveillance could, in due course, bolster the monitoring of insecticide resistance. Once laboratory infrastructure and workflows are established, these protocols could facilitate the mapping of resistance mechanisms within vector populations. This includes, for instance, the identification of kdr mutations, the detection of P450 gene overexpression, and/or the analysis of chitin synthase I1043F mutation, which is associated to diflubenzuron resistance [12,25,26].
## 3. How Do Mosquitoes Participate in the Maintenance and Transmission of Arboviruses in the Environment?
Arbovirus transmission involves a complex mechanism comprising several factors, including a virulent viral strain capable of replicating in both vertebrate and invertebrate hosts, susceptible vertebrate hosts, and competent arthropod vectors. For an arthropod to serve as a vector, it must acquire the pathogen through a blood meal, allow for its replication within its body, and subsequently transmit it to a susceptible host [27]. It is worth noting that pathogen acquisition in vectors can also occur through alternative routes, such as horizontal and sexual transmission, which will be discussed later in this article. A critical aspect of this process is the extrinsic incubation period (EIP), which is the time interval between the arthropod acquiring the pathogen and its ability to transmit it via its saliva. The EIP is influenced by environmental factors such as temperature and humidity, as well as vector-intrinsic factors, including susceptibility to infection and interactions with its microbiota. During a viral infection in mosquitoes, the virus must overcome several barriers. These include successful infection and replication in midgut epithelial cells after a blood meal (midgut infection barrier-MIB), subsequent traversal of the midgut basal lamina to access the mosquito hemocoel (midgut escape barrier-MEB), survival and transportation within the hemocoel to the target tissues, primarily the salivary glands, infection of the salivary gland epithelial cells (salivary gland infection barrier-SGIB), and eventual release into the salivary ducts for dissemination during feeding (salivary gland escape barrier-SGEB) [28,29].
The transmission dynamic of a mosquito arbovirus is primarily shaped by ecological factors, as the cohabitation of mosquitoes and hosts in shared environments enhances transmission opportunities through an increased host-mosquito contact rate [30]. Also, vector longevity is critical, since viruses must complete their replication cycle within the mosquito before being transmitted to a new host [31]. Vector competence, which consists in the intrinsic capacity of a vector to transmit a pathogen considering complex processes and barriers specific to the vector [32], varies significantly among mosquito species and even between populations of the same species [33][34][35][36]. These variations underscore the need to characterize local mosquito populations for accurate arbovirus transmission risk assessments. Furthermore, the appetitive behavior of mosquito populations has been shown to be an important factor in defining their vectorial competence. Recent studies have shown that multiple sequential blood feedings can increase the dissemination of arboviruses in several mosquito genera. This phenomenon may be attributed to several factors, including damage to the midgut basal lamina caused by blood feeding, a reduction in the EIP [37][38][39], or acceleration of parasite development [40,41].
The biological interaction between viruses and their vectors arises from a continuous coevolutionary process. This process is shaped by the vector's intrinsic factors and environmental interactions, including the microbiota and the presence of insect-specific viruses (ISVs). ISVs are a diverse group of viruses that replicate in insects (in vivo and in vitro) but are incapable of replicating in vertebrates and their cells, being transmitted vertically to progeny in insects [42,43]. These elements, as well as the microbiota, can exert a direct influence on the vectorial capacity of a vector population, defined as the ability of a vector population to transmit pathogens to a susceptible host, providing a more robust framework for understanding its role in pathogen transmission dynamics [44]. One example of the modulation promoted by ISVs is a phenomenon called 'superinfection exclusion', whereby a primary ISV infection in an insect inhibits infections by genetically similar or closely related viruses [45]. Another example of interaction occurs in Ae. aegypti infected with symbiotic bacteria of the genus Wolbachia. The presence of these bacteria inhibits the replication of arboviruses, such as DENV, in Aedes sp. mosquitoes, in addition to impacting population density, reducing the longevity of females, and reducing the number of offspring due to cytoplasmic incompatibility between reproductive cells [26,32,46].
Mosquitoes exhibit diverse adaptive strategies to cope with environmental changes, including insecticide resistance, ecological plasticity, dietary shifts, and physiological adjustments. Insecticide resistance has become a growing concern in recent years, with studies highlighting the rapid evolution of resistance mechanisms in mosquito populations, such as target-site mutations and increased detoxification-enzyme activity, which compromise the efficacy of chemical control measures [47,48]. Ecological plasticity enables mosquitoes to exploit a wide range of habitats, from natural to urban environments, allowing species to thrive in densely populated areas with abundant artificial breeding sites [49,50]. Dietary shifts, including increased reliance on human blood meals, have been observed in some mosquito populations, further enhancing their ability to transmit pathogens [51]. Physiological adaptations, such as thermal tolerance, enable mosquitoes to survive in a broader range of climatic conditions, extending their geographic distribution and seasonal activity [52,53].
During adverse conditions such as drought or low temperatures, certain mosquito species, such as those of the tribe Aedini, may enter reproductive diapause, a state of suspended development that enhances survival. For example, observations made with Hg. leucocelaenus indicated that eggs were 1.5 times more likely to hatch during the rainy season than during the dry season. This phenomenon may represent an evolutionary strategy that increases the survival rates, since the rainy season provides mosquitoes with greater access to essential resources, including breeding sites, high relative humidity, and greater plant cover and host abundance [54]. Sim and Denlinger [55] identified the molecular and hormonal pathways involved in diapause, offering insights into how mosquitoes synchronize their reproductive cycles with environmental cues. Mosquito eggs also demonstrate remarkable resilience; for example, Ae. aegypti eggs can remain viable for over a year in dry conditions, allowing populations to persist and rebound quickly when conditions improve [56,57]. Emerging studies have also reported that these mosquitoes can withstand varying levels of salinity, suggesting an even greater adaptive capacity in response to environmental stressors [58][59][60]. These findings underscore the remarkable adaptability of mosquitoes and the challenges this poses for controlling mosquito-borne diseases in a rapidly changing world.
## 4. How Does the Prolonged Viability of Mosquito Eggs Influence Viral Transmission Dynamics?
Arthropods are the true reservoir of arboviruses. Once infected, arthropods remain infected and may be capable of transmitting the virus to susceptible vertebrates throughout their lives [61]. Arthropods are unable to effectively clear a virus from their bodies, and in some instances, can vertically transmit the virus to their offspring, which results in infected progeny [62].
Vertical transmission can occur via two primary mechanisms: transovarial transmission, where the virus infects the female germline, and trans-egg transmission, where the virus infects the developing egg during oviposition (Figure 1) [5,62]. The hypothesis of vertical transmission has been explored in the context of pathogen persistence under adverse environmental conditions, such as drought, winter, interepidemic periods, and intensive vector control measures [62,63]. In this scenario, viruses (arboviruses and ISVs) can persist within mosquito eggs, immature mosquito stages, and adult females, including those entering diapause, without requiring a vertebrate host [5]. The detection of arboviruses in male mosquitoes supports the role of vertical transmission in maintaining arbovirus persistence in the environment. A study conducted in Mexico revealed that 6.7% of the male mosquitoes collected during a post-epidemic period were positive for arboviruses, with CHIKV being the most prevalent (5.7%), followed by DENV (0.9%) and ZIKV (0.1%) [64]. Alencar et al. [65] identified positivity for ZIKV and YFV in samples of male and female Ae. albopictus and Hg. leucocelaenus, collected through ovitraps, in Rio de Janeiro, Brazil. Although Culex eggs exhibit reduced environmental resilience compared to Aedes sp. eggs, vertical transmission of arboviruses has been observed in Culex species [66]. There is limited information about how these horizontal infection events could trigger new transmission cycles leading to outbreaks or even how this persistence affects the virulence of viruses. Two studies [67,68] demonstrated the horizontal (venereal) transmission of ZIKV in Ae. aegypti. In these research works, male mosquitoes (infected through intrathoracic microinjection) transmitted ZIKV to females during copulation. Similarly, female mosquitoes (orally infected) transmitted the virus to males during copulation. Studies on horizontal transmission need to be encouraged, and this information considered, as outbreak forecasting based on traditional epidemiological models that do not incorporate this mechanism may lead to inaccurate predictions.
## 5. Methodologies Applied to Entomo-Virological Surveillance
Vector surveillance methodologies are, in general, a combination of the techniques employed in field entomology, such as the collection of specimens in the field, identification, and sometimes, their maintenance in the laboratory, and virology techniques that, in part, are based on molecular biology and genomics. Studies focused on the analysis of vector competence in mosquitoes may also use techniques of cell culture and viral isolation [34,36,69].
The choice of mosquito collection strategies depends on the surveillance objective. The methodological design should incorporate the biological and behavioral characteristics of the vector species involved in arbovirus transmission. Both active and passive methods can be used to achieve this, including CDC light traps (LT), CDC LT with CO 2 attractants, human landing catches, ovitraps, BG-Pro sentinels, and larval collection. Comprehensive surveillance or outbreak investigations can benefit from a combination of complementary collection methods to facilitate an extensive sampling of the local vector population. For example, a study conducted in West Africa using nets suspended from helium balloons at altitudes of 120-290 m allowed for the tracking of 61 mosquito species, with some mosquitoes positive for arbovirus, Plasmodium sp., and filarial infections [70]. A list of materials, equipment, and biosafety requirements needed to conduct entomological investigations can be found in the Supplementary Materials section of this article.
It is necessary to note that the primary objective of entomo-virological surveillance is the detection of viruses. Consequently, all aspects of field operations, from specimen collection and identification to nucleic acid extraction and subsequent analyses, must be optimized to ensure successful viral detection. For instance, mosquitoes collected using traps should be processed promptly and stored under conditions that preserve the viral genetic material, such as ultra-refrigeration (-80 • C), in liquid nitrogen dewars (-196 • C), or in appropriate preservation media (ethanol, propylene glycol, and nucleic acid preservation reagent) [71,72].
In entomo-virological studies, specimens in a range of life stages can be collected, including eggs [34,69], larvae, pupae [8,35,73], and adults [7,14,66,[74][75][76][77][78][79]. Adults are preferentially collected due to the higher likelihood that they offer of detecting viral infections, considering both vertical and horizontal transmission routes. The combination of vertical and horizontal arbovirus transmission in mosquito populations represents a highly effective strategy that promotes the persistence of these pathogens in the environment. These modes of transmission diversify the pathways through which pathogens can spread and endure within a vector population [62]. Furthermore, morphological identification keys are primarily based on adult mosquitoes, particularly females, as many diagnostic morphological characteristics are not fully developed in early larval instars [80]. Collecting immature life stages (eggs, larvae, and pupae) is generally considered to pose lower risks to researchers compared to adult collection methods like human landing catches, Shannon traps, and manual aspiration. Nevertheless, collecting immature stages demands a comprehensive understanding of vector ecology and appropriate facilities to rear specimens to later developmental stages [81]. However, molecular methods that will be mentioned below in this review can be used for the identification of immature specimens, eliminating the need for laboratory rearing. Another interesting biological specimen processed in some studies is the adult mosquito's excreta. An adaptation of BG-Sentinel traps, developed by Manzi et al. [82], utilizes FTA cards soaked with a solution made of honey and a hydroxycellulose hydrogel. This medium not only ensures the preservation of the viral nucleic acids expelled by mosquitoes onto the card but also remarkably keeps the mosquitoes alive for days, thereby maximizing the collection efficiency. L'Ambert et al. [83] were able to detect WNV circulation in Camargue, France, using a xenomonitoring method based on the molecular detection of the virus in excreta from trapped mosquitoes. This strategy shed light on how viral surveillance can complement standard surveillance methods.
Accurate taxonomic identification is essential for comprehending the dynamics and factors driving arbovirus transmission. Morphological identification remains the gold standard [80]. However, it necessitates the expertise of skilled taxonomists, particularly when rapid viral detection is the primary goal. Identification on refrigerated tables or on chemical ice packs improves the sample preservation conditions, aiming at viral detection or isolation. Innovative approaches have been developed to supplement or improve the accuracy of mosquito species identification. Machine learning algorithms offer a promising avenue for automated morphological identification, although large datasets are required for training the algorithms [84]. Over the past two decades, mitochondrial and ribosomal genes have become established tools for species-level taxonomic identification in entomological research. Commonly employed markers include cytochrome oxidase 1 (COI), cytochrome oxidase B (CytB), ITS2, D2, and 28S and 12S rRNA [80,[85][86][87]. DNA barcoding can be integrated with viral detection protocols [85,[88][89][90][91], providing a valuable tool for situations where expert taxonomic expertise is limited. However, DNA barcoding-based identification still faces challenges in resolving recent intraspecific divergences [86,92].
Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) has become a cornerstone technique for the direct detection of viruses in mosquito samples within the field of entomo-virology [7,8,14,66,74,[77][78][79]89]. While RT-qPCR is often used as a primary detection tool, it is frequently complemented by whole-genome sequencing of positive samples to characterize viral genomes and infer phylogenetic relationships [7,66,74,76,78,89,93]. A critical aspect of RT-qPCR-based arbovirus detection in mosquitoes is the establishment of a standardized threshold for positivity. The cycle threshold (Ct) value, representing the number of cycles required for the fluorescent signal to cross a defined threshold, is commonly used to determine positivity. However, the optimal Ct cutoff can vary depending on factors such as the target virus, sample quality, and assay sensitivity. While some studies have adopted Ct cutoffs of 35 [36], 37 [14], 38 [64,77,93], or even 40 [79], the choice of the cutoff remains a subject of ongoing discussion. For samples with high or indeterminate Ct values, it is recommended to perform replicate RT-qPCR assays using separate aliquots of the extracted RNA to minimize the risk of compromising RNA integrity due to repeated freeze-thaw cycles.
Viral metagenomics offers a novel approach to explore the diverse viral communities associated with mosquito hosts. Unlike targeted RT-qPCR methods, virome analysis provides an unbiased exploration of the entire viral community, including ISVs, viruses associated with the microbiota, and arboviruses [94]. This approach can facilitate the discovery of novel arboviruses with potential public health significance [76]. Virome analysis typically involves next-generation sequencing (NGS) [95] and often leverages high-throughput sequencing platforms [90,96]. While NGS-based virome analysis offers an unparalleled depth of viral discovery, it is a resource-intensive methodology that generates vast amounts of data requiring sophisticated bioinformatics pipelines. Factors such as sample type, collection methods, and sequencing depth can significantly influence the complexity and interpretation of virome data [96]. Although the routine integration of virome analysis into entomo-virological surveillance may be challenging due to technical and logistical constraints, it remains a promising complementary tool for addressing specific research questions and public health concerns.
The positivity of mosquito samples collected from field sites exhibits variability that may be influenced by factors such as vector species/lineage, viral species, and geographic location. For instance, studies investigating WNV surveillance have reported an average positivity rate of 4.71% among mosquito pools comprising Culex sp. and Aedes sp. in Israel [74]. A comparable positivity rate of 4.4% for WNV-positive pools was documented in Greece [76].
To better assess the transmission risk, entomo-virological studies often employ additional metrics such as the minimum infection rate (MIR) and maximum likelihood estimate (MLE), beyond simple positivity rates [7,8,20,63,66,73,[77][78][79][97][98][99][100]. The MIR is determined by dividing the number of mosquito pools positive for arboviruses by the total number of mosquitoes tested and multiplying the result by 1000. The MLE is calculated using the formula [97]:
In this equation:
• n = total number of pools tested; • x = number of positive pools;
The MIR provides a conservative estimate of the infection rate, while the MLE offers a more precise estimate of the true infection rate, since it normalizes for the pool size. Under low-intensity transmission scenarios, the MIR and MLE values may be comparable. However, in high-transmission or epidemic settings, these two estimates can diverge significantly [20,101].
However, entomo-virological analyses, even without the taxonomic identification step, can still have important informative value in the context of vector-borne disease surveillance. Although it does not identify the presence of species known to be important vectors, the analysis of large pools with few resources can still inform the presence of the pathogen of interest, serving as a predictor for emergency alert of an outbreak.
Resource limitations are a common challenge for municipalities, making the development of more accessible diagnostic tools crucial. Isothermal amplification technologies, applied to arbovirus detection, offer a timely solution. These methodologies utilize a specific restriction enzyme that functions without thermal cycling. While isothermal amplification may exhibit lower sensitivity compared to diagnostic methods considered gold standards for arboviruses, such as RT-qPCR, its utility as a screening tool in surveillance is significant. When a sample tests positive via this method, it can then be forwarded for more specific and sensitive analyses, optimizing resource allocation and accelerating the response times in public health initiatives [102].
$$MLE = [1 -(n -χ/n) 1/m] × 1000$$
$$• m = pool size.$$
## 6. The Role of Entomo-Virological Surveillance in Understanding the Multi-Vector Transmission of Arboviruses
Upon identifying an arbovirus outbreak or epidemic, immediate preventive actions must be implemented to protect the exposed population. These measures may include vaccination (if available), establishing outpatient and inpatient care services, intensifying laboratory testing, and executing vector control interventions. However, developing a comprehensive prevention and control strategy becomes particularly challenging in urban and rural contexts when multiple vector species are involved in transmitting and maintaining a specific arbovirus. Zoonotic arboviruses that thrive in wild, rural, or suburban environments often rely on multiple vector species and infect a wide range of vertebrate hosts [19,103]. The emergence of such arboviruses at epidemic levels can be triggered by disruptions to natural ecosystems caused by changes in viral genetics, host or vector population dynamics, or anthropogenic environmental modifications [104]. A notable example is the transmission of Oropouche fever in Brazil.
Oropouche fever is a febrile illness in humans caused by the tri-segmented Orthobunyavirus oropoucheense (OROV). Its clinical presentation, which includes myalgia, arthralgia, and headache, often overlaps with that of other arboviral infections like dengue. This overlap, coupled with limited diagnostic testing and the co-circulation of multiple arboviruses, contributes to its underreporting [105]. Before the emergence of CHIKV and ZIKV, OROV was the second most prevalent arbovirus in Brazil after dengue. However, a surge in OROV transmission was detected in Brazil starting in 2023, following the implementation of enhanced laboratory testing for cases negative for dengue, chikungunya, and Zika. This heightened surveillance revealed significant OROV circulation, initially concentrated in the Brazilian Amazon (Acre, Amazonas, Rondônia, and Roraima) and later spreading to other regions of the country [106,107].
While Culicoides paraensis (Diptera: Ceratopogonidae) is the primary vector for OROV transmission, other Diptera species, such as Culex quinquefasciatus, Aedes scapularis, Aedes serratus, Coquillettidia venezuelensis, Psorophora cingulata, and Haemagogus tropicalis, may also play a role [77,105,[107][108][109][110][111]. Additionally, other Culicoides species cannot be ruled out as contributors to OROV transmission [112]. Effective prevention strategies for vector-borne diseases, especially those without available vaccines, often begin with vector control measures using insecticides. However, uncertainty surrounding the specific roles of multiple vectors with diverse ecological characteristics increases the risk of implementing ineffective control strategies.
A thorough understanding of vector diversity, competence, and biology is essential to designing targeted control strategies that account for technical, economic, environmental, and public health considerations [113]. Entomo-virological surveillance serves as a critical tool for comprehending the natural history of arboviral diseases and identifying the key factors contributing to the establishment and maintenance of arbovirus transmission. This knowledge is crucial for developing effective preventive measures tailored to the specific epidemiological complexities of each situation.
## 7. Entomology-Virology for Strengthening Border and Port Surveillance
Virological surveillance for arboviruses necessitates prioritizing strategic locations such as ports, borders, and urban centers. These areas function as critical hubs for the movement of goods, people, and potential vectors, creating an environment conducive to the introduction and dissemination of arboviruses. The challenges associated with border regions are further compounded by disparities in healthcare systems, which can impede the prevention and control of emerging health threats [114]. The absence of robust surveillance tools in these strategic locations can compromise preparedness and response efforts in the face of emerging imported pathogens. For instance, DENV and YFV, and likely ZIKV, were introduced into Brazil via ports and cross-border human movement. Airports also play a significant role in the introduction of arboviruses [115][116][117][118]. Travelers can bring new pathogens into a territory. Although the importation of infected vectors is less frequent, it still poses a non-negligible risk. A surveillance program conducted at a major international airport in Central Europe underscored the potential for this route of introduction, revealing high levels of WNV and Usutu virus (Orthoflavivirus usutuense-USUV) transmission [66].
## 8. Challenges in Implementing Entomo-Virological Surveillance for Public Health
Implementing a comprehensive public health surveillance strategy necessitates a multifaceted approach that encompasses management, service organization, and robust data infrastructure. While the collection of relevant indicators is a fundamental component, it alone is insufficient for establishing effective entomo-virological surveillance. A systematic framework for data collection, analysis, and response is essential to ensure timely and appropriate interventions.
The COVID-19 pandemic has accelerated the development of genomic surveillance capabilities, particularly within public health laboratories, in many countries, including Brazil [119,120]. Anticipating that climate change may alter the transmission dynamics of arboviruses and increase their impact on human and animal health [121], it is opportune and strategic to leverage these expanded laboratory capacities for integrated arbovirus surveillance [119]. Entomo-virological surveillance represents a strategic approach to expanding the genomic surveillance of arboviruses with public health significance. This approach offers several advantages, including the ability to anticipate outbreaks, understand the epidemic potential of emerging and circulating viral strains, and assess the impact of antiviral drugs and novel vector control strategies [120].
The implementation of entomo-virological surveillance, however, is often hindered by various challenges, including limited human resources, inadequate infrastructure, and difficulties in data collection and sharing. To address these limitations, a comprehensive surveillance strategy should be developed, fostering collaboration among stakeholders and providing opportunities for training and capacity building. Effective health management is essential for consolidating integrated surveillance systems for arboviruses. This involves coordinating efforts with sectors responsible for solid waste management, urbanization, water supply, and environmental education, as well as engaging with the community. By investing in surveillance infrastructure and promoting data sharing, it is possible to strengthen the integration of environmental, epidemiological, and laboratory surveillance. And with this understanding, it is possible to reinforce the importance of directing government investments towards the training of entomologists and the construction of structures that enable the implementation of entomo-virological surveillance as a tool for preventing and responding to the threats that vector-borne diseases represent for public health.
Community communication, education, and engagement are fundamental pillars that support the structure of communicable disease surveillance, with special emphasis on entomo-virological surveillance. This interface is crucial for the effective execution and success of surveillance activities. A participatory community process significantly increases the receptiveness of the population to public health interventions. This direct collaboration facilitates the identification of vector breeding sites, the application of vector prevention and control measures, the detection of suspected human cases, and epizootics [122]. This strategy helps to direct actions and optimize epidemiological and entomological investigations, thus allowing for a faster and more efficient response to public health threats.
## 9. The Arbovirus Diagnostic Laboratory Network of the Americas (RELDA)
The Pan American Health Organization (PAHO/WHO), in partnership with its member states, established the Americas Arbovirus Diagnostic Network (RELDA) to enhance the capacity for arbovirus diagnosis and surveillance. Building upon the successful Dengue Laboratory Network of the Americas, which was launched in 2008, RELDA aims to strengthen laboratory infrastructure, harmonize diagnostic protocols, and promote col-laboration among laboratories in the region (https://www.paho.org/en/topics/dengue/ arbovirus-diagnosis-laboratory-network-americas-relda (accessed on 27 December 2024)).
The emergence of CHIKV and ZIKV viruses in the Americas highlighted the need for a more robust and coordinated regional response to arbovirus outbreaks. As a result, RELDA expanded its scope to include the Arbovirus Genomic Surveillance Platform (Vi-GenDa) and the Entomo-Virological Laboratory Network (RELEVA). Currently, RELDA comprises 40 laboratories from across the Americas, including Argentina, Brazil, Canada, and the United States. These laboratories collaborate to improve diagnostic capabilities, share data, and respond to emerging threats. Additionally, five collaborating centers provide technical support and guidance to the network. These centers include the National Institute of Human Viral Diseases in Argentina, the Evandro Chagas Institute in Brazil, and the Centers for Disease Control and Prevention in the United States.
## 10. Conclusions
Entomo-virological surveillance is a necessary tool for assessing risks and identifying the factors that drive arbovirus outbreaks and epidemics. In public health, this type of surveillance can be implemented at various levels, ranging from localized monitoring in specific areas such as ports, airports, and sentinel municipalities to broader, nationwide efforts, depending on the operational resources available.
The transmission dynamics of arboviruses depends on specific environmental conditions, characteristics of the arboviruses itself and its vectors (especially the capacity of being maintained for several months in eggs), and previous contact of human populations with that pathogen, which will affect their susceptibility to disease and symptoms. All factors will influence the risk assessment in a local population.
Several methods for mosquito and arbovirus identification using high-throughput genetic data may contribute significantly to assessing the transmission risk and comparing outbreaks in different areas. However, the conditions needed to achieve the whole potential of those strategies are challenging due to the complex logistics of their routine use.
## References
1. (2017) "/WHO Special Programme for Research and Training in Tropical Diseases"
2. Donalisio, Freitas, Zuben (2017) "Arboviruses Emerging in Brazil: Challenges for Clinic and Implications for Public Health" *Rev. Saúde Pública*
3. Chauhan, Matthews, Piquet et al. "Nervous System Manifestations of Arboviral Infections" *Curr. Trop. Med. Rep*
4. Shepard, Undurraga, Halasa et al. (2016) "The Global Economic Burden of Dengue: A Systematic Analysis" *Lancet Infect. Dis*
5. Lequime, Lambrechts (2014) "Vertical Transmission of Arboviruses in Mosquitoes: A Historical Perspective" *Infect. Genet. Evol*
6. Weetman, Kamgang, Badolo et al. (2018) "Aedes Mosquitoes and Aedes-Borne Arboviruses in Africa: Current and Future Threats" *Int. J. Environ. Res. Public Health*
7. Tsioka, Gewehr, Pappa et al. (2022) "West Nile Virus in Culex Mosquitoes in Central Macedonia" *Viruses*
8. Akhtar, Gupta, Singh (2022) "Surveillance of Zika and Dengue Viruses in Field-Collected Aedes aegypti Mosquitoes from Different States of India" *Virology*
9. Ferreira-De-Brito, Ribeiro, Miranda et al. (2016) "First Detection of Natural Infection of Aedes aegypti with Zika Virus in Brazil and throughout South America" *Mem. Inst. Oswaldo Cruz*
10. Aragão, Pinheiro, Nunes Neto et al. (1126) "Natural Infection of Aedes aegypti by Chikungunya and Dengue Type 2 Virus in a Transition Area of North-Northeast Brazil" *Viruses*
11. Teixeira, De Brito, Correia et al. (2021) "Simultaneous Circulation of Zika, Dengue, and Chikungunya Viruses and Their Vertical Co-Transmission among Aedes aegypti" *Acta Trop*
12. Almeida-Souza, Oliveira, Brito et al. (2024) "High Frequencies of Kdr Mutation and Chikungunya Infection in Aedes aegypti Population from Minas Gerais"
13. Reis, Gibson, Ayllón et al. (2019) "Entomo-Virological Surveillance Strategy for Dengue, Zika and Chikungunya Arboviruses in Field-Caught Aedes Mosquitoes in an Endemic Urban Area of the Northeast of Brazil" *Acta Trop*
14. Cruz, Hernández, Aragão et al. "The Importance of Entomo-Virological Investigation of Yellow Fever Virus to Strengthen Surveillance in Brazil" *Trop. Med*
15. Pinheiro, Rocha, De Oliveira et al. (2019) "Detection of Yellow Fever Virus in Sylvatic Mosquitoes during Disease Outbreaks of 2017-2018 in Minas Gerais State" *Brazil. Insects*
16. Abreu, Ribeiro, Ferreira-De-Brito et al. (2016) "Haemagogus leucocelaenus and Haemagogus janthinomys Are the Primary Vectors in the Major Yellow Fever Outbreak in Brazil" *Emerg. Microbes Infect*
17. Cunha, Faria, Caleiro et al. (2016) "Genomic Evidence of Yellow Fever Virus in Aedes scapularis" *Southeastern Brazil*
18. Stanzani, Motta, Erbisti et al. (2017) "Back to Where It Was First Described: Vectors of Sylvatic Yellow Fever Transmission in the"
19. De Oliveira, Andrade, Campos et al. (2021) "Yellow Fever Virus Maintained by Sabethes Mosquitoes during the Dry Season in Cerrado, a Semiarid Region of Brazil" *Viruses*
20. Cevallos, Ponce, Waggoner et al. (2018) "Zika and Chikungunya Virus Detection in Naturally Infected Aedes aegypti in Ecuador" *Acta Trop*
21. Wang, Fan, Ji et al. "Mapping the Distributions of Mosquitoes and Mosquito-Borne Arboviruses in China" *Viruses*
22. Fairbanks, Daly, Tildesley (1221) "Modelling the Influence of Climate and Vector Control Interventions on Arbovirus Transmission" *Viruses*
23. Medeiros-Sousa, Lange, Mucci et al. (2024) "Modelling the Transmission and Spread of Yellow Fever in Forest Landscapes with Different Spatial Configurations" *Ecol. Model*
24. (2012) "World Health Organization. Handbook for Integrated Vector Management; World Health Organization"
25. Fotakis, Mavridis, Kampouraki et al. "Mosquito Population Structure, Pathogen Surveillance and Insecticide Resistance Monitoring in Urban Regions of Crete" *PLoS Negl*
26. Ateutchia-Ngouanet, Nanfack-Minkeu, Mavridis et al. (2024) "Monitoring Aedes Populations for Arboviruses, Wolbachia, Insecticide Resistance and Its Mechanisms in Various Agroecosystems in Benin" *Acta Trop*
27. Salazar, Richardson, Sánchez-Vargas et al. (2007) "Dengue Virus Type 2: Replication and Tropisms in Orally Infected Aedes aegypti Mosquitoes" *BMC Microbiol*
28. Franz, Kantor, Passarelli et al. (2015) "Tissue Barriers to Arbovirus Infection in Mosquitoes" *Viruses*
29. Lewis, Gallichotte, Randall et al. (2023) "Intrinsic Factors Driving Mosquito Vector Competence and Viral Evolution: A Review" *Front. Cell. Infect. Microbiol*
30. Tabachnick (2016) "Ecological Effects on Arbovirus-Mosquito Cycles of Transmission" *Curr. Opin. Virol*
31. Wu, Yu, Wang et al. (2019) "Arbovirus Lifecycle in Mosquito: Acquisition, Propagation and Transmission" *Expert Rev. Mol. Med*
32. Gesto, Ribeiro, Rocha et al. "Reduced Competence to Arboviruses Following the Sustainable Invasion of Wolbachia into Native Aedes aegypti from Southeastern Brazil" *Sci. Rep*
33. Black, Bennett, Gorrochótegui-Escalante et al. (2002) "Flavivirus Susceptibility in Aedes aegypti" *Arch. Med. Res*
34. Gutiérrez-Bugallo, Boullis, Martinez et al. "Vector Competence of Aedes aegypti from Havana, Cuba, for Dengue Virus Type 1, Chikungunya, and Zika Viruses" *PLoS Negl. Trop*
35. Morales-Vargas, Missé, Chavez et al. "Vector Competence for Dengue-2 Viruses Isolated from Patients with Different Disease Severity" *Pathogens*
36. Amoa-Bosompem, Kobayashi, Itokawa et al. "Determining Vector Competence of Aedes aegypti from Ghana in Transmitting Dengue Virus Serotypes 1 and 2. Parasit. Vectors 2021"
37. Armstrong, Ehrlich, Magalhaes et al. (2019) "Successive Blood Meals Enhance Virus Dissemination within Mosquitoes and Increase Transmission Potential" *Nat. Microbiol*
38. Johnson, Cozens, Ferdous et al. (2023) "Increased Blood Meal Size and Feeding Frequency Compromise Aedes aegypti Midgut Integrity and Enhance Dengue Virus Dissemination" *PLoS Negl. Trop. Dis*
39. Ferdous, Dieme, Sproch et al. (2024) "Multiple Bloodmeals Enhance Dissemination of Arboviruses in Three Medically Relevant Mosquito Genera" *Parasit. Vectors*
40. Shaw, Holmdahl, Itoe et al. (2020) "Multiple Blood Feeding in Mosquitoes Shortens the Plasmodium falciparum Incubation Period and Increases Malaria Transmission Potential" *PLoS Pathog*
41. Brackney, Lareau, Smith (2021) "Frequency Matters: How Successive Feeding Episodes by Blood-Feeding Insect Vectors Influences Disease Transmission" *PLoS Pathog*
42. Vasilakis, Tesh (2015) "Insect-Specific Viruses and Their Potential Impact on Arbovirus Transmission" *Curr. Opin. Virol*
43. Carvalho, Long (2021) "Insect-Specific Viruses: An Overview and Their Relationship to Arboviruses of Concern to Humans and Animals" *Virology*
44. Cansado-Utrilla, Zhao, Mccall et al. "The Microbiome and Mosquito Vectorial Capacity: Rich Potential for Discovery and Translation"
45. Laureti, Paradkar, Fazakerley et al. (1259) "Superinfection Exclusion in Mosquitoes and Its Potential as an Arbovirus Control Strategy" *Viruses*
46. Minwuyelet, Petronio, Yewhalaw et al. "Symbiotic Wolbachia in Mosquitoes and Its Role in Reducing the Transmission of Mosquito-Borne Diseases: Updates and Prospects"
47. Lima, Paiva, De Araújo et al. "Insecticide Resistance in Aedes aegypti Populations from Ceará"
48. Moyes, Vontas, Martins et al. "Contemporary Status of Insecticide Resistance in the Major Aedes Vectors of Arboviruses Infecting Humans" *PLoS Negl*
49. Ndenga, Mutuku, Ngugi et al. (2017) "Characteristics of Aedes aegypti Adult Mosquitoes in Rural and Urban Areas of Western and Coastal Kenya" *PLoS ONE*
50. Chandrasegaran, Lahondère, Escobar et al. (2020) "Linking Mosquito Ecology, Traits, Behavior, and Disease Transmission" *Trends Parasitol*
51. Harrington, Edman, Scott (2001) "Why Do Female Aedes aegypti (Diptera: Culicidae) Feed Preferentially and Frequently on Human Blood?" *J. Med. Entomol*
52. Mordecai, Caldwell, Grossman et al. (2019) *Thermal Biology of Mosquito-borne Disease. Ecol. Lett*
53. Lahondère, Bonizzoni (2022) "Thermal Biology of Invasive Aedes Mosquitoes in the Context of Climate Change" *Curr. Opin. Insect Sci*
54. Alencar, De Mello, Leite et al. "Oviposition Activity of Haemagogus leucocelaenus (Diptera: Culicidae) during the Rainy and Dry Seasons"
55. Sim, Denlinger (2013) "Insulin Signaling and the Regulation of Insect Diapause" *Front. Physiol*
56. Silva, Silva (1999) "Influência do período de quiescência dos ovos sobre o ciclo de vida de Aedes aegypti (Linnaeus, 1762) (Diptera, Culicidae) em condições de laboratório" *Rev. Soc. Bras. Med. Trop*
57. Prasad, Sreedharan, Bakthavachalu et al. (2023) "Eggs of the Mosquito Aedes aegypti Survive Desiccation by Rewiring Their Polyamine and Lipid Metabolism" *PLoS Biol*
58. Multini, Oliveira-Christe, Medeiros-Sousa et al. (2021) "The Influence of the pH and Salinity of Water in Breeding Sites on the Occurrence and Community Composition of Immature Mosquitoes in the Green Belt of the City of São Paulo"
59. De Brito Arduino, Mucci, Serpa et al. (2015) "Effect of Salinity on the Behavior of Aedes aegypti Populations from the Coast and Plateau of Southeastern Brazil" *J. Vector Borne Dis*
60. Ratnasari, Jabal, Syahribulan et al. (2021) "Salinity Tolerance of Larvae Aedes aegypti Inland and Coastal Habitats in Pasangkayu" *Biodiversitas J. Biol. Divers*
61. Consoli, Oliveira (1994) "Principais Mosquitos de Importância Sanitária No Brasil; Editora FIOCRUZ: Rio de Janeiro" *Brazil*
62. Lequime, Paul, Lambrechts (2016) "Determinants of Arbovirus Vertical Transmission in Mosquitoes" *PLoS Pathog*
63. Dahiya, Yadav, Yadav et al. (2022) "Zika Virus Vertical Transmission in Mosquitoes: A Less Understood Mechanism" *J. Vector Borne Dis*
64. Kirstein, Talavera, Wei et al. (2022) "Natural Aedes-Borne Virus Infection Detected in Male Adult Aedes aegypti (Diptera: Culicidae) Collected From Urban Settings in"
65. Alencar, Ferreira De Mello, Brisola Marcondes et al. (2021) "Natural Infection and Vertical Transmission of Zika Virus in Sylvatic Mosquitoes Aedes albopictus and Haemagogus leucocelaenus from Rio de Janeiro" *Brazil. Trop. Med. Infect. Dis*
66. Bakran-Lebl, Camp, Kolodziejek et al. (2022) "Diversity of West Nile and Usutu Virus Strains in Mosquitoes at an International Airport in Austria"
67. Pereira-Silva, Nascimento, Belchior et al. (2017) "First Evidence of Zika Virus Venereal Transmission in Aedes aegypti Mosquitoes" *Mem. Inst. Oswaldo Cruz*
68. Campos, Fernandes, Dos Santos et al. (2017) "Zika Virus Can Be Venereally Transmitted between Aedes aegypti Mosquitoes" *Parasit. Vectors*
69. Chen, Bozic, Mathias et al. (2023) "Immune-Related Transcripts, Microbiota and Vector Competence Differ in Dengue-2 Virus-Infected Geographically Distinct Aedes aegypti Populations" *Parasit. Vectors*
70. Bamou, Dao, Yaro et al. (2024) "Pathogens Spread by High-Altitude Windborne Mosquitoes"
71. Torres, Weakley, Hibbert et al. (1431) "Ethanol as a Potential Mosquito Sample Storage Medium for RNA Preservation" *F1000Research*
72. Kai, Kobayashi, Itokawa et al. (2024) "Evaluation of Long-Term Preservation Methods for Viral RNA in Mosquitoes at Room Temperature" *J. Virol. Methods*
73. Jain, Kushwah, Singh et al. (2016) "Evidence for Natural Vertical Transmission of Chikungunya Viruses in Field Populations of Aedes aegypti in Delhi and Haryana States in India-A Preliminary Report" *Acta Trop*
74. Lustig, Hindiyeh, Orshan et al. (2016) "Mosquito Surveillance for 15 Years Reveals High Genetic Diversity Among West Nile Viruses in Israel" *J. Infect. Dis*
75. Sadeghi, Altan, Deng et al. (2018) "Virome of >12 Thousand Culex Mosquitoes from throughout California" *Virology*
76. Papa, Gewehr, Tsioka et al. (2018) "Detection of Flaviviruses and Alphaviruses in Mosquitoes in Central Macedonia" *Acta Trop*
77. Pereira-Silva, Ríos-Velásquez, Lima et al. "Distribution and Diversity of Mosquitoes and Oropouche-like Virus Infection Rates in an Amazonian Rural Settlement" *PLoS ONE*
78. Rothman, Jones, Ladeau et al. (2021) "Higher West Nile Virus Infection in Aedes albopictus (Diptera: Culicidae) and Culex (Diptera: Culicidae) Mosquitoes From Lower Income Neighborhoods in Urban Baltimore, MD" *J. Med. Entomol*
79. Seo, Lee, Yang et al. (2085) "National Monitoring of Mosquito Populations and Molecular Analysis of Flavivirus in the Republic of Korea in 2020" *Microorganisms*
80. Chan, Chiang, Hapuarachchi et al. "Complementing Morphological Identification of Mosquito Species in Singapore"
81. Mcdermott, Lysyk (2020) "Sampling Considerations for Adult and Immature Culicoides (Diptera: Ceratopogonidae)" *J. Insect Sci*
82. Manzi, Nelli, Fortuna et al. "A Modified BG-Sentinel Trap Equipped with FTA Card as a Novel Tool for Mosquito-Borne Disease Surveillance: A Field Test for Flavivirus Detection. Sci"
83. L'ambert, Gendrot, Briolant et al. (2023) "Analysis of Trapped Mosquito Excreta as a Noninvasive Method to Reveal Biodiversity and Arbovirus Circulation" *Mol. Ecol. Resour*
84. Kittichai, Kaewthamasorn, Samung et al. (2023) "Automatic Identification of Medically Important Mosquitoes Using Embedded Learning Approach-Based Image-Retrieval System" *Sci. Rep*
85. Batovska, Lynch, Cogan et al. (2018) "Effective Mosquito and Arbovirus Surveillance Using Metabarcoding" *Mol. Ecol. Resour*
86. Moraes Zenker, Portella, Pessoa et al. (2024) "Low Coverage of Species Constrains the Use of DNA Barcoding to" *Assess Mosquito Biodiversity. Sci. Rep*
87. Oliveira, Saraiva, Da Silva et al. "Molecular Identification of Mosquitoes (Diptera: Culicidae) Using COI Barcode and D2 Expansion of 28S Gene" *DNA*
88. Hoyos-López, Suaza-Vasco, Rúa-Uribe et al. (2016) "Molecular Detection of Flaviviruses and Alphaviruses in Mosquitoes (Diptera: Culicidae) from Coastal Ecosystems in the Colombian Caribbean" *Mem. Inst. Oswaldo Cruz*
89. Guarido, Govender, Riddin et al. (2021) "Detection of Insect-Specific Flaviviruses in Mosquitoes (Diptera: Culicidae) in Northeastern Regions of South Africa" *Viruses*
90. Hameed, Wahaab, Shan et al. "A Metagenomic Analysis of Mosquito Virome Collected From Different Animal Farms at Yunnan-Myanmar Border of China"
91. Hernández-Triana, Garza-Hernández, Ortega Morales et al. "An Integrated Molecular Approach to Untangling Host-Vector-Pathogen Interactions in Mosquitoes (Diptera: Culicidae) From Sylvan Communities in Mexico"
92. Beebe (2018) "DNA Barcoding Mosquitoes: Advice for Potential Prospectors" *Parasitology*
93. Main, Nicholson, Winokur et al. (2018) "Vector Competence of Aedes aegypti, Culex tarsalis, and Culex quinquefasciatus from California for Zika Virus" *PLoS Negl. Trop. Dis*
94. Moonen, Schinkel, Van Der Most et al. (2023) "Composition and Global Distribution of the Mosquito Virome-A Comprehensive Database of Insect-Specific Viruses" *One Health*
95. Gómez, Martínez, Páez-Triana et al. (2023) "Characterizing Viral Species in Mosquitoes (Culicidae) in the Colombian Orinoco: Insights from a Preliminary Metagenomic Study" *Sci. Rep*
96. Liu, Cui, Liu et al. "Association of Virome Dynamics with Mosquito Species and Environmental Factors" *Microbiome*
97. Condotta, Hunter, Bidochka et al. (2004) "Virus Infection Rates in Pooled and Individual Mosquito Samples" *Vector-Borne Zoonotic Dis*
98. Balingit, Carvajal, Saito-Obata et al. (2020) "Surveillance of Dengue Virus in Individual Aedes aegypti Mosquitoes Collected Concurrently with Suspected Human Cases in Tarlac City" *Philippines. Parasit. Vectors*
99. Heath, Grossi-Soyster, Ndenga et al. "Evidence of Transovarial Transmission of Chikungunya and Dengue Viruses in Field-Caught Mosquitoes in Kenya" *PLoS Negl*
100. Fish, Tesh, Guzman et al. "Emergence Potential of Mosquito-Borne Arboviruses from the Florida Everglades" *PLoS ONE*
101. Gu, Lampman, Novak (2003) "Problems in Estimating Mosquito Infection Rates Using Minimum Infection Rate" *J. Med. Entomol*
102. Varghese, De Silva, Millar (1159) "Latest Advances in Arbovirus Diagnostics" *Microorganisms*
103. Gyawali, Bradbury, Aaskov et al. (2017) "Neglected Australian Arboviruses: Quam Gravis? Microbes Infect"
104. Weaver, Reisen (2010) "Present and Future Arboviral Threats" *Antivir. Res*
105. Riccò, Corrado, Bottazzoli et al. (1498) "Emergence of Oropouche Virus (OROV) Infections: Systematic Review and Meta-Analysis of Observational Studies" *Viruses*
106. Brazil, Ministry, Health et al. (2024)
107. Naveca, Almeida, Souza et al. (2024) "Human Outbreaks of a Novel Reassortant Oropouche Virus in the Brazilian Amazon Region" *Nat. Med*
108. Romero-Alvarez, Escobar, Oropouche Fever (2018) *Microbes Infect*
109. De Mendonça, Rocha, Ferreira et al. (2021) "Evaluation of Aedes aegypti, Aedes albopictus, and Culex quinquefasciatus Mosquitoes Competence to Oropouche Virus Infection" *Viruses*
110. Bonifay, Le Turnier, Epelboin et al. (1268) "Review on Main Arboviruses Circulating on French Guiana, An Ultra-Peripheric European Region in South America" *Viruses*
111. Zhang, Liu, Wu et al. (2024) "Oropouche Virus: A Neglected Global Arboviral Threat" *Virus Res*
112. Requena-Zúñiga, Palomino-Salcedo, García-Mendoza et al. (2024) "First Detection of Oropouche Virus in Culicoides"
113. Carpenter, Mellor, Torr (2008) "Control Techniques for Culicoides Biting Midges and Their Application in the U.K. and Northwestern Palaearctic" *Med. Vet. Entomol*
114. Santos-Melo, Andrade, Rocha et al. "Importância e desafios da vigilância em saúde em uma região de fronteira internacional: Um estudo de caso"
115. Benchimol (1994) "História Da Febre Amarela No Brasil" *História Ciênc. Saúde-Manguinhos*
116. Braga, Valle (2007) "Aedes aegypti: Histórico do controle no Brasil" *Epidemiol. Serviços Saúde*
117. Faria, Quick, Claro et al. (2017) "Establishment and Cryptic Transmission of Zika Virus in Brazil and the Americas" *Nature*
118. Naveca, Claro, Giovanetti et al. (2019) "Epidemiological and Digital Surveillance of Chikungunya Virus in the Brazilian Amazon" *PLoS Negl. Trop. Dis*
119. Carter, Yu, Sacks et al. (2022) "Global Genomic Surveillance Strategy for Pathogens with Pandemic and Epidemic Potential 2022-2032" *Bull. World Health Organ*
120. Wallau, Abanda, Abbud et al. (2023) "Arbovirus Researchers Unite: Expanding Genomic Surveillance for an Urgent Global Need" *Lancet Glob. Health*
121. De Souza, Weaver (2024) "Effects of Climate Change and Human Activities on Vector-Borne Diseases" *Nat. Rev. Microbiol*
122. (2024) "Pan American Health Organization. Effective Communication Strategies and Practices for Dengue and Other Arboviral Diseases: Systematic Review" *Eff. Commun*
123. "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=PMC12696956&blobtype=pdf | # The top 100 most cited articles on pediatric respiratory syncytial virus pneumonia over the last 30 years: a bibliometric analysis
Fei Luo, Chanchan Hu, Qian Liu, Naixu Liu, Kang Lian, Demei Wu, Zijian Shao, Yuanyuan Wang, Mingchen Jiang, Bin Yuan
## Abstract
Background Respiratory syncytial virus (RSV) is a leading cause of acute lower respiratory infections in humans, contributing to a substantial burden on both families and society. To date, no bibliometric studies have specifically addressed RSV pneumonia. We therefore employed bibliometric methods to analyze the top 100 most-cited articles in this field, aiming to construct a visual knowledge map and quantitatively identify current research hotspots and emerging trends.
MethodsWe retrieved relevant publications from the Web of Science Core Collection (WoSCC) database. Using Microsoft Excel 2019, CiteSpace 6.2.R4, and VOSviewer 1.6.18, we performed a visual analysis of annual publication trends, countries, institutions, authors, journals, and keywords.
ResultsThe 100 most-cited articles received a total of 15,949 citations, with individual citation counts ranging from 46 to 2846 and a median of 74. The United States contributed the most publications, and the Centers for Disease Control and Prevention (CDC) was the most productive institution. The most prolific authors were Cohen, Cheryl; Graham, Barney S; Anderson, LJ; and Ramilo, O. The Pediatric Infectious Disease Journal published and received the most citations in this domain. "Bronchiolitis" was identified as the keyword with the strongest citation burst.
ConclusionCurrent research on RSV pneumonia remains focused on pathogenesis, treatment, and prognosis. The development of new antiviral drugs and immunoprophylaxis strategies continues to be a central direction for future studies.
## Introduction
Respiratory syncytial virus (RSV) is a major pathogen responsible for acute lower respiratory tract infections, hospitalizations, and deaths in humans [1]. Initially isolated from chimpanzees in 1956, it is named for its characteristic induction of cell fusion and syncytia formation in cultured cells. Based on host specificity, RSV is classified into human respiratory syncytial virus (HRSV, first isolated from an infant in 1957), bovine respiratory syncytial virus (BRSV), and mouse respiratory syncytial virus (MRSV). Virologically, RSV was formerly placed within the genus Pneumovirus of the † Fei Luo and Chanchan Hu have contributed equally to this work and share first authorship.
family Paramyxoviridae. However, the International Committee on Taxonomy of Viruses (ICTV) reclassified it into the genus Orthopneumovirus within the family Pneumoviridae in 2015, and in 2016, HRSV was formally renamed Human Orthopneumovirus [2]. RSV primarily infects infants, the elderly, and immunocompromised individuals, with an estimated 33 million new global cases annually [3], resulting in 118,000 deaths each year among children under five [4]. The burden is especially high in low-and middle-income countries. Furthermore, RSV is a significant risk factor for asthma, recurrent wheezing, and impaired lung function. In children under 3 years of age, the incidence of recurrent wheezing after RSV infection ranges from 4 to 47%, while that of asthma ranges from 8 to 76% [5]. Impaired lung function is associated with an increased risk of cardiovascular disease and premature mortality in adulthood [6]. Clinical symptoms include nasal congestion, rhinorrhea, cough, dyspnea, and fever, with severe cases progressing to respiratory failure [7]. Current preventive measures rely on monoclonal antibodies such as Palivizumab and Nirsevimab, which target the viral fusion (F) protein to block host cell entry, substantially reducing hospitalization rates in high-risk pediatric populations.
Bibliometrics provides a quantitative approach to reveal disciplinary trends and frontiers by analyzing the distribution and relationships among knowledge carriers [8]. Commonly used tools include CiteSpace [9], which visualizes knowledge structures and detects emerging themes, and VOSviewer [10], which is particularly suited for co-citation network analysis. Among citation databases, Web of Science (WOS) is the most authoritative and covers the broadest range of disciplines, with an emphasis on natural sciences [11]. Although Scopus includes a wide selection of journals, it suffers from delayed indexing; Google Scholar often provides incomplete citation data, and PubMed lacks a native citation analysis module [12].
To date, no bibliometric study has specifically addressed RSV pneumonia. This work presents the first bibliometric analysis of the top 100 most-cited articles on RSV pneumonia from the Web of Science Core Collection (WoSCC). Using methods such as cocitation clustering, keyword emergence analysis, and knowledge mapping, we systematically examine the evolution, core topics, and research frontiers in this field. Our objective is to offer a quantitative foundation and conceptual framework to inform future investigative priorities.
## Materials and methods
## Data selection
On December 24, 2024, we searched the WoSCC for RSV pneumonia-related literature, ranking the results by citation count in descending order and selecting the 100 most highly cited articles. Our search criteria included the terms "respiratory syncytial virus pneumonia" or "RSV pneumonia" in combination with "Child, " "Children, " "Childhood, " or "Pediatric. " The study focused on English-language articles and reviews, excluding abstracts, editorials, conference papers, book chapters, and retracted publications. The search encompassed a 30-year period from January 1, 1994, to December 31, 2023, with the results obtained promptly on the same day. Study selection and data extraction were performed independently by two authors. Any discrepancies were resolved by discussion or by consulting a third author to reach a consensus. Figure 1 illustrates the literature search process.
## Data analysis
Extract key information from the eligible articles, such as title, authors, countries, institutions, keywords, and publication year. The collected data was then processed utilizing Microsoft Excel 2019, along with CiteSpace 6.1.R6 and VOSviewer 1.6.18 software.
## Quality control
To ensure methodological rigor and reporting transparency, this study adhered strictly to the Preliminary Guideline for Reporting Bibliometric Reviews of the Biomedical Literature (BIBLIO) [13]. This checklist offers a comprehensive framework for bibliometric research, and its recommendations guided all stages of data collection, processing, analysis, and interpretation.
## Result
## Characteristics of the included articles
This study systematically searched the WoSCC database for literature related to RSV pneumonia in children from 1994 to 2023, yielding a total of 1699 research records. Through descending citation frequency, the 100 most influential articles were selected for in-depth analysis, including 86 research papers and 14 review articles.
Table 1 lists the top 10 most-cited articles in the field of pediatric RSV pneumonia research. Figure 2 illustrates the annual distribution of the top 100 most-cited articles from 1994 to 2023, where the x-axis represents the year, and the y-axis corresponds to the annual publication volume. Analysis revealed that these 100 highly cited articles have been cited a total of 15,949 times, with individual articles being cited between 46 and 2846 times, and a median citation count of 74. Regarding annual distribution, the publication volume fluctuated between 1 and 7 articles, peaking in 2011 (n = 7). Notably, only 2 articles were cited over 2000 times, and 3 articles were cited over 1000 times.
This study involved 643 researchers from 42 countries, spanning 243 research institutions, and the research outcomes were published in 60 academic journals, demonstrating a broad trend of international collaboration and exchange.
## Analysis of countries
As shown in Table 2 and Fig. 3, a total of 41 countries worldwide have participated in highly cited research in the field of RSV pneumonia in children. Analyzing the number of publications by country, the United States ranks first with 63 articles, followed by the United Kingdom (16 articles) and the Netherlands (14 articles) in second and third places, respectively. In terms of total citation frequency by country, the United States (11,631 citations), the United Kingdom (3710 citations), and Kenya (2699 citations) form the core knowledge production group. Notably, in terms of average citations per article, Kenya (386 citations/article), Japan (373 citations/ article), and France (356 citations/article) demonstrate significant academic influence.
Figure 3, as an analysis diagram of the national cooperation network for RSV pneumonia, provides a visual representation of the global research landscape. Figure 3a clearly illustrates the geographical distribution of RSV pneumonia publications by country. Evidently,
## Analysis of institutions
According to Table 3 and Fig. 4, 243 research institutions worldwide have participated in the knowledge production of highly cited studies in the field of pediatric RSV pneumonia. The top two institutions with the most published papers are the Centers for Disease Control and Prevention (CDC) in the United States (9 papers) and the University of Edinburgh (6 papers). In terms of citation counts, the top two institutions are the CDC (3776 citations) and the University of the Witwatersrand (2371 citations) (Table 3). Betweenness centrality is an indicator that measures the importance of a node in a network. A higher centrality value indicates that the node plays a bridging role in communicating with other nodes [14]. Nodes with a centrality value > 0.1 are considered relatively important. Empirical analysis shows that the University of Edinburgh (centrality = 0.16), as the only node breaking the threshold (> 0.1) (marked in purple), plays a key bridging function in cross-regional scientific research collaboration. Additionally, despite having both a high number of publications and citations, the CDC's centrality value (0.09) did not reach the threshold, suggesting that its research relies more on internal resources rather than international collaboration.
## Analysis of authors
The highly cited literature in the field of pediatric RSV pneumonia involves 519 researchers from around the globe. Table 4 showcases the top 10 authors who have contributed the most to these 100 articles. Cohen, Cheryl, Graham, Barney S, Anderson, LJ, and Ramilo, O are tied for the first place as the most prolific authors, each with 4 articles published. Graham, Barney S, Anderson, LJ, and Ramilo, O are all from the United States, while Cohen, Cheryl is from South Africa. As illustrated in Fig. 5, the author collaboration network exhibits a multi-center discrete distribution. Despite the presence of cross-institutional collaborations, none of the nodes' betweenness centrality exceeded the significance threshold of 0.1, indicating no high-centrality authors were observed.
## Analysis of Journals
Based on Fig. 6 and Table 5, the top 100 highly cited articles in the field of pediatric RSV pneumonia were published across 60 academic journals. The Pediatric Infectious Disease Journal published the highest number of articles (n = 10), followed by the Journal of Infectious Diseases (n = 5) and the Journal of Virology (n = 4). Among the top 10 journals, 5 belong to the JCR Q1 quartile, 3 belong to the Q2 quartile, and 2 belong to the Q3 quartile.
## Analysis of keywords
The top 100 most-cited articles in this study contained 453 keywords. Table 6 lists the top 15 high-frequency keywords, with "respiratory syncytial virus" (appearing 48 times), "Child" (appearing 42 times), "infant" (appearing 32 times), and "infection" (appearing 31 times) being the keywords that appeared more than 30 times.
From the keyword co-occurrence graph in Fig. 7a, it can be seen that the red phrase group involves RSV infection and immune response, such as "activation", "immune prophylaxis", "cytokine", "T-cells", "viral infection", etc. These keywords indicate that this cluster mainly focuses on the immune response triggered by RSV infection, cytokine release, and immune prophylaxis mechanisms, especially the body's immune response after RSV infection and the virus's immune evasion mechanisms. The yellow phrase group includes risks and pathologies related to pneumonia and bacterial infections, such as "bacteremia", "bacterial infection", "community-acquired Fig. 3 Analysis of the countries. A World the cooperation intensity map. B A circle dia-gram that evaluates international collaboration between clusters pneumonia", "risk", "epidemiology", "burden", etc. These keywords show that this cluster mainly concerns the pathophysiology of pneumonia caused by bacterial infections, community-acquired pneumonia, pathological burden, and epidemiological data, with an emphasis on co-infections of RSV and bacteria, as well as complications such as pneumonia. The blue phrase group covers the clinical manifestations and treatment of RSV infection in children, including "child", "infants", "high-risk children", "hospitalization", "therapy", "prevention", "monoclonal-antibody", etc. These keywords suggest that this cluster primarily focuses on the clinical features, treatment strategies, and preventive measures of RSV infection in children, particularly the incidence of RSV in high-risk children, hospitalization situations, and the effectiveness of monoclonal antibody therapy. Figure 7b illustrates the top 25 keywords based on burst analysis intensity. The top five keywords with the highest burst intensities are "bronchiolitis" (3.43), "g protein" (3.07), "pneumonia" (2.95), "tract infection" (2.87), and "cell" (2.23). The earliest emerging keywords are "Washington" and "epidemiology", while the most recent emerging keyword is "hospitalization".
## Discussion
Employing bibliometric methods, this study identified and rigorously analyzed 100 highly influential articles in the field of RSV pneumonia. By systematically examining research trends, core themes, and emerging frontiers, it integrates foundational studies with recent advances. The resulting academic landscape reveals the evolution of knowledge in this domain, offering both quantitative evidence and theoretical support for understanding the developmental trajectory of RSV pneumonia research.
## General information
Our research results indicate that the median citation count for the top 100 articles on RSV pneumonia is 74.
In terms of document type, over 86% of the top 100 cited key evidence-based support for the European Union to develop RSV monitoring guidelines, but also directly promote the scientific decision-making process of incorporating monoclonal antibody prevention strategies into the immunization programs of member states. The Centers for Disease Control and Prevention (CDC) in the United States has conducted systematic epidemiological research on Respiratory Syncytial Virus (RSV) infection in adults, building a quantitative database covering hospitalization rates, outpatient visit rates, and the economic burden of the disease. The research findings provide profound evidence-based support for developing intervention strategies for high-risk groups with RSV [17].
In the field of RSV pneumonia research, Professor Cohen, Cheryl focuses on studying key pathogens, including influenza, RSV, SARS-CoV-2, and pneumococcus [18,19]. Professor Graham, Barney S. from NIAID is a prominent scholar in the RSV field. His innovative work on the RSV prefusion F structure has established a "game-changing" approach for structurebased vaccine design, which is currently being applied to the development of vaccines for coronaviruses and other important pathogens, including Ebola and influenza [20]. Professor Anderson, LJ's team mainly studies the pathogenic mechanism of RSV to guide vaccine development [21]. Professor Ramilo, O is also dedicated to the development of an RSV vaccine [22].
Regarding journals, the Pediatric Infectious Disease Journal is the most prolific publication, holding significant academic importance in the field of RSV pneumonia. Moreover, most of the listed journals fall into the Q1 or Q2 categories, indicating that these journals hold a high academic status in RSV pneumonia research and are widely esteemed by scholars.
## Research hotspots
Combining keyword co-occurrence and cluster analysis, the current research hotspots of Respiratory Syncytial Virus (RSV) pneumonia can be summarized into three aspects: pathophysiology, treatment, and prognosis.
## Pathophysiology
RSV infection primarily affects the respiratory system, with the main mechanisms being airway obstruction, bronchial smooth muscle spasm, and subsequent airway hyperresponsiveness [23].
Mechanical airway obstruction: RSV invades the epithelial cells of the trachea, bronchioles, and alveoli, triggering the shedding of ciliated epithelial cells. This process creates a vicious cycle: shed cells accumulate in the airways along with inflammatory cells such as neutrophils and lymphocytes, contributing to luminal narrowing due to mucus hypersecretion and mucosal edema. It's worth noting that neutrophils exacerbate pathological changes through a dual role: ① releasing oxygen free radicals and elastase, which directly damage the epithelial barrier; ② upregulating the expression of TNF-α and IL-13, driving excessive mucus gland secretion.
Neurogenic bronchial spasm: After the epithelial barrier is disrupted, exposed sensory nerve endings release neuropeptides such as substance P, triggering bronchial smooth muscle contraction through the following mechanisms: ① directly activating l-type calcium channels in the ASMC cell membrane, elevating intracellular Ca 2 ⁺ concentration; ② stimulating mast cells to release bronchoconstrictor mediators like histamine and leukotrienes; ③ inducing increased cholinergic nerve excitability, promoting acetylcholine release. It's particularly noteworthy that neurogenic inflammation induced by substance P can produce a continuous amplification effect.
Persistent airway hyperresponsiveness: Airway remodeling after RSV infection involves a complex regulatory network: ① imbalance between β-adrenergic receptor function inhibition and M receptor activation; ② NGF-mediated sensory nerve sensitization; ③ a chronic inflammatory state driven by the Th17/IL-17 axis.
## Treatment
Current treatment methods include supportive care (such as maintaining electrolyte balance, providing nutrition, and ensuring airway patency), symptomatic treatment (such as antipyretic and analgesic drugs, bronchodilators), antiviral therapy (such as the monoclonal antibody nirsevimab), and immunotherapy (such as monoclonal antibody prophylaxis for high-risk groups) [24]. The use of glucocorticoids and leukotriene receptor antagonists remains controversial.
## Prognosis
RSV infection is a self-limiting disease, and the vast majority of infected children have a good prognosis without any lasting sequelae. A very small number of children may experience respiratory failure, neurological complications, or even death. In a minority of children, recurrent wheezing and bronchial asthma may occur in the later stages of infection, with some cases resulting in post-infectious bronchiolitis obliterans. Severe RSV infection is associated with an increased risk of recurrent wheezing and asthma in preschool children [24]. Research on RSV is undergoing a major paradigm shift, propelled by innovations in prevention strategies and the influence of the COVID-19 pandemic. The field is gradually moving away from a historically reactive posture, centered on post-infection treatment, toward a new era defined by proactive intervention. This transition is supported by breakthroughs in long-standing research challenges. Recent advances, drawing on structural insights into the pre-fusion F protein, enabled the approval of long-acting monoclonal antibodies like nirsevimab and maternal vaccines in 2023, fundamentally altering the prevention landscape [25,26]. Investigative priorities have now broadened beyond traditional clinical trials to include four key real-world concerns: real-world effectiveness (RWE), implementation science focused on accessibility and equity, genomic surveillance of viral escape mutations, and mechanisms through which maternal antibodies shape infant immune responses [27][28][29]. At the same time, the COVID-19 pandemic has substantially reshaped RSV epidemiology, disrupting seasonal trends, introducing the concept of "immunity debt" [30,31], and stimulating deeper inquiry into the clinical effects of viral co-infection [32]. The success of COVID-19 vaccines, especially mRNA platforms, has also established new technological foundations for RSV vaccine development [33]. In summary, the convergence of preventive advances and pandemic-related changes is driving RSV research into an era of interdisciplinary integration. Future efforts will connect clinical medicine, public health, and viral immunology, as novel and forward-looking scientific questions continue to expand and redefine conventional research priorities.
## Limitation
This study has several limitations. First, methodological biases may have arisen from the inclusion of self-citations, potentially inflating citation counts, and from using citation volume as a screening criterion, which could favor earlier publications, review articles, and studies in high-impact journals while underrepresenting recent original research and specialized journals. Second, limitations pertain to the analytical tools and search strategy: the performance of the bibliometric software depends on data quality and the stability of keyword extraction algorithms, and although the search terms encompassed core pediatric concepts, it is possible that a small number of studies using terms such as "infant" alone were overlooked. Third, systemic biases exist in the data sources due to the restriction to English-language articles within the Web of Science Core Collection, which excludes non-English publications and regional databases and may limit the global representativeness and generalizability of the results.
## Conclusion
This systematic review analyzes the 100 most-cited research articles on pediatric RSV pneumonia from the past three decades, constructing a knowledge map and tracing the evolution of the field. The findings indicate that research on RSV pneumonia centers primarily on pathophysiology, treatment, and prognosis. Developing novel antiviral agents and immunoprophylaxis strategies continues to represent a major direction for future investigation.
## References
1. Borchers, Chang, Gershwin (2013) "Respiratory syncytial virus-a comprehensive review" *Clin Rev Allergy Immunol*
2. Duan, Liu, Zang (2024) "Landscape of respiratory syncytial virus" *Chin Med J*
3. Lozano, Naghavi, Foreman (2010) "Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study" *Lancet*
4. Openshaw, Chiu, Culley (2017) "Protective and harmful immunity to RSV infection" *Annu Rev Immunol*
5. Fauroux, Simões, Checchia (2017) "The burden and long-term respiratory morbidity associated with respiratory syncytial virus infection in early childhood" *Infect Dis Ther*
6. Zar, Cacho, Kootbodien (2024) "Early-life respiratory syncytial virus disease and long-term respiratory health" *Lancet Respir Med*
7. Colosia, Costello, Mcquarrie (2023) "Systematic literature review of the signs and symptoms of respiratory syncytial virus. Influenza Other Respir Viruses"
8. Ninkov, Frank, Maggio (2022) "Bibliometrics: methods for studying academic publishing" *Perspect Med Educ*
9. Chen (2006) "CiteSpace II: detecting and visualizing emerging trends and transient patterns in scientific literature" *J Am Soc Inf Sci Technol*
10. Van Eck, Waltman (2010) "Software survey: VOSviewer, a computer program for bibliometric mapping" *Scientometrics*
11. Luo, Li, Zhang (2022) "Bibliometric analysis of IgA vasculitis nephritis in children from 2000 to 2022" *Front Public Health*
12. Falagas, Pitsouni, Malietzis (2008) "Comparison of PubMed, Scopus, Web of Science, and Google Scholar: strengths and weaknesses" *FASEB J*
13. Montazeri, Mohammadi, Hesari (2023) "Preliminary guideline for reporting bibliometric reviews of the biomedical literature (BIBLIO): a minimum requirements" *Syst Rev*
14. Luo, Ding, Zhang (2024) "Frontier and hotspot evolution in cerebrotendinous xanthomatosis: a bibliometric analysis from 1993 to 2023" *Front Neurol*
15. Li, Reeves, Wang (2019) "Global patterns in monthly activity of influenza virus, respiratory syncytial virus, parainfluenza virus, and metapneumovirus: a systematic analysis" *Lancet Glob Health*
16. Chatzis, Darbre, Pasquier (2018) "Burden of severe RSV disease among immunocompromised children and adults: a 10 year retrospective study" *BMC Infect Dis*
17. Zheng, Warren, Shapiro (2022) "Estimated incidence of respiratory hospitalizations attributable to RSV infections across age and socioeconomic groups" *Pneumonia*
18. Cohen, Zar (2024) "Early life respiratory syncytial virus disease-a preventable burden" *Lancet Infect Dis*
19. Cohen, Kleynhans, Moyes (2017) "Incidence and transmission of respiratory syncytial virus in urban and rural South Africa" *Nat Commun*
20. Graham (2011) "Biological challenges and technological opportunities for respiratory syncytial virus vaccine development" *Immunol Rev*
21. Anderson, Jadhao, Hussaini (2023) "Development and comparison of immunologic assays to detect primary RSV infections in infants" *Front Immunol*
22. Glowinski, Mejias, Ramilo (2021) "New preventive strategies for respiratory syncytial virus infection in children" *Curr Opin Virol*
23. Mazur, Caballero, Nunes (2024) "Severe respiratory syncytial virus infection in children: burden, management, and emerging therapies" *Lancet*
24. Penders, Brusselle, Falsey (2025) "Burden of respiratory syncytial virus disease in adults with asthma and chronic obstructive pulmonary disease: a systematic literature review" *Curr Allergy Asthma Rep*
25. Hammitt, Dagan, Yuan (2022) "Nirsevimab for prevention of RSV in healthy late-preterm and term infants" *N Engl J Med*
26. Kampmann, Madhi, Munjal (2023) "Bivalent prefusion F vaccine in pregnancy to prevent RSV illness in infants" *N Engl J Med*
27. Sumsuzzman, Wang, Langley (2025) "Real-world effectiveness of nirsevimab against respiratory syncytial virus disease in infants: a systematic review and meta-analysis" *Lancet Child Adolesc Health*
28. Moline, Tannis, Toepfer (2023) "Early estimate of Nirsevimab effectiveness for prevention of respiratory syncytial virus-associated hospitalization among infants entering their first respiratory syncytial virus season-new Vaccine Surveillance Network" *MMWR Morb Mortal Wkly Rep*
29. Yunker, Fall, Norton (2024) "Genomic evolution and surveillance of respiratory syncytial virus during the 2023-2024 season" *Viruses*
30. Olsen, Winn, Budd (2021) "Changes in influenza and other respiratory virus activity during the COVID-19 pandemic-United States, 2020-2021" *MMWR Morb Mortal Wkly Rep*
31. Baker, Park, Yang (2020) "The impact of COVID-19 nonpharmaceutical interventions on the future dynamics of endemic infections" *Proc Natl Acad Sci USA*
32. Agathis, Patel, Milucky (2023) "Codetections of other respiratory viruses among children hospitalized with COVID-19" *Pediatrics*
33. Walsh, Perez, Zareba (2023) "Efficacy and Safety of a Bivalent RSV Prefusion F Vaccine in Older Adults" *Journal Article; Randomized Controlled Trial* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12671185&blobtype=pdf | # Multiplex detection and application of MALDI-TOF NAMS for porcine diarrheal pathogens
Jiangbing Shuai, Shiqi Song, Xiao Han, Ya Zhao, Kexin Chen, Youran Guo, Huimin Guo, Nan Zhang, Xiaofeng Zhang
## Abstract
Porcine diarrheal diseases caused by mixed viral and bacterial infections pose significant challenges to swine health and production. Rapid and accurate identification of key pathogens, including porcine epidemic diarrhea virus (PEDV), transmissible gastroenteritis virus (TGEV), porcine deltacoronavirus (PDCoV), porcine rotavirus (PoRV), swine acute diarrhea syndrome coronavirus (SADS-CoV), porcine bocavirus (PBoV), hepatitis E virus (HEV), and Salmonella, remains a challenge due to overlapping clinical symptoms and co-infections. The zoonotic potential of HEV and PBoV also highlights the need for reliable, high-throughput detection methods. We developed a multiplex Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Nucleic Acid Mass Spectrometry (MALDI-TOF NAMS) assay for the simultaneous detection of eight major porcine gastrointestinal pathogens. The method utilizes a multiplex PCR approach and a single-base extension strategy to improve sensitivity and specificity, and the extension products are displayed by mass spectrometry. The analytical performance of the assay was assessed by evaluating the limits of detection (LoD), analytical specificity, and repeatability. Additionally, we validated its diagnostic accuracy using 242 clinical samples, comparing the results with real-time quantitative PCR (qPCR) as the reference method. The MALDI-TOF NAMS assay demonstrated high analytical specificity, with no cross-reactivity to non-target pathogens. The limit of detection (LoD) ranged from 12.20 to 33.59 copies/μL. The intra-and inter-batch reproducibility assessments showed a 100% detection rate across high, medium, and low concentrations (20/20 per concentration, 60/60 total). Validation using 242 clinical samples demonstrated a 98.3% sensitivity and 99.5% specificity compared to qPCR, with an overall concordance rate of 96.2%. The MALDI-TOF NAMS assay provides a sensitive and high-throughput method for the detection and genotyping of major swine diarrhea pathogens. This method performs well in mixed infections and allows for expansion of pathogen species, making it an important addition to traditional diagnostic methods. IMPORTANCE Porcine diarrheal diseases are a major threat to the swine industry, often caused by multiple viruses and bacteria infecting animals at the same time. Fast and accurate detection of these pathogens is crucial to prevent outbreaks, reduce economic losses, and protect public health, especially given the potential of some pathogens to infect humans. This study introduces a new method that uses nucleic acid mass spectrometry technology to quickly and accurately detect eight major pathogens from pig samples, such as feces, blood, or tissue. Unlike traditional tests that often detect one pathogen at a time, this method can screen for many at once, even in complex cases of mixed infections. It is sensitive, reliable, and can handle large numbers of samples efficiently. This tool offers farmers, veterinarians, and disease control agencies a faster and more effective way to monitor pig health and respond to outbreaks before they spread.
P orcine gastrointestinal diseases are among the most common health issues in swine production and carry significant economic consequences. These diarrheal diseases are particularly common in piglets, where viral and bacterial diarrhea can lead directly to growth retardation and even high mortality due to the incomplete development of their digestive system (1). Notable among these are porcine epidemic diarrhea virus (PEDV), transmissible gastroenteritis virus (TGEV), porcine delta coronavirus (PDCoV), porcine rotavirus (PoRV), and swine acute diarrhea syndrome coronavirus (SADS-CoV). These gastrointestinal viruses are highly detrimental, and their predominantly clini cal symptoms, diarrhea, are often indistinguishable, posing significant challenges to differential diagnosis in the field (2,3). Porcine bocavirus (PBoV), a virus frequently detected in pigs, has recently emerged as a potential contributor to gastrointestinal diseases (4). Increasing evidence suggests that PBoV is often present in mixed infections with other pathogens, such as PEDV, PCV2, and CSFV, where it can exacerbate clinical severity and increase herd morbidity and mortality (5). Although genotypes G1, G2, and G3 have been identified, studies investigating their prevalence and significance in diarrheic pigs remain limited (6). Beyond swine health, PBoV has also been sporadically detected in humans, raising concern about its zoonotic potential (7). These findings underscore the importance of including PBoV in multiplex diagnostic assays to improve the etiological diagnosis of porcine diarrheal diseases. Additionally, fecal-oral pathogens, such as Salmonella and hepatitis E virus (HEV), infect pigs and pose serious public health concerns due to their zoonotic potential (8,9). HEV and Salmonella excreted in the feces of infected pigs can contaminate the environment, leading to cross-species transmission to humans and amplifying the risk of foodborne or waterborne outbreaks.
Fast and accurate detection of these pathogens is crucial to prevent outbreaks, reduce economic losses, and protect public health, especially given the potential of some pathogens to infect humans. Importantly, epidemiological studies have shown that porcine coronaviruses (PEDV, PDCoV, TGEV, and SADS-CoV) and rotavirus (PoRV) frequently co-circulate, with mixed or secondary infections significantly increasing herd morbidity and mortality (10). Because these pathogens are serologically distinct, diagnosing them individually increases both cost and turnaround time. Multiplex detection can improve diagnostic efficiency, identify the major causative agents in co-infections, and facilitate the timely removal or management of infected animals, thereby reducing herd-level transmission and economic loss (11). From a broader One Health perspective, comprehensive pathogen detection is also critical for surveillance, outbreak control, and informing vaccination strategies, while also reducing the risk of zoonotic spread (12).
Currently, diagnostic methods for porcine gastrointestinal pathogens include ELISA, real-time PCR (qPCR), and next-generation sequencing (NGS). ELISA is suitable for serological surveillance but may fail in early-stage infections and is less reliable in the presence of antigenic cross-reactivity. While qPCR is widely used and offers high sensitivity and specificity, it is often limited in multiplexing capacity and requires separate reactions to detect multiple pathogens, increasing cost and time. NGS provides a broad pathogen overview but is costly, labor-intensive, and less practical for rou tine diagnostics. In contrast, Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Nucleic Acid Mass Spectrometry (MALDI-TOF NAMS) offers the advantage of simultane ously detecting multiple pathogens in a single reaction, with high sensitivity, excellent specificity, and robust performance in complex samples, including those with mixed infections.
MALDI-TOF NAMS is a nucleic acid detection technique that combines multiplex PCR with mass spectrometry (13,14). It operates by ionizing nucleic acid fragments and measuring their unique time-of-flight in a vacuum to differentiate target DNA sequences based on their molecular weights. By identifying distinct unextended probes (UEPs) and single-base extended products (SEPs), this technology can simultaneously detect and analyze several or even dozens of target sequences in the same sample. This effectively improves the ability to detect multiple pathogens in a single reaction. MALDI-TOF NAMS, based on the sensitivity of PCR and utilizing the accuracy of mass spectrometry technology, can intuitively and rapidly present the results of multi-tar get detection. Originally, MALDI-TOF NAMS was employed in clinical diagnostics to detect disease-or drug resistance-related genes and served as a validation platform when discrepancies arose between first-and second-generation sequencing results (15,16). The scalability and speed of MALDI-TOF NAMS make it particularly suitable for high-throughput screening in clinical and surveillance settings. It has been successfully applied to multiplex detection of various infectious agents, such as human coronaviruses (17), sexually transmitted pathogens (18), Mycobacterium tuberculosis (19), and swine viruses (20).
This study aims to address this gap by developing a MALDI-TOF NAMS assay for the simultaneous detection of eight major pathogens associated with porcine diarrheal syndromes: PDCoV, PEDV, TGEV, SADS-CoV, HEV, PoRV, PBoV, and Sal. By enabling the rapid, accurate, and high-throughput detection of these pathogens, this assay not only supports the timely prevention and control of porcine infectious diarrheal diseases but also provides an effective tool for pathogen surveillance in swine populations. Further more, from the perspective of cross-border animal trade and biosecurity, such a method holds promise for enhancing the diagnostic capabilities of customs and quarantine agencies, ensuring the early detection and monitoring of pathogens that may pose risks to animal health, food safety, and public health.
## MATERIALS AND METHODS
## Viruses and positive templates
The strains, as well as bacterial and viral nucleic acid samples, used in this study were primarily employed for specificity evaluation. Inactivated viral strains, including Porcine epidemic diarrhea virus (PEDV, CV777 strain), Porcine transmissible gastroenteritis virus (TGEV, WH-1R strain), Porcine reproductive and respiratory syndrome virus (PRRSV, HuN4 strain), Pseudorabies virus (PRV, Bartha strain), and Porcine circovirus type 2 (PCV-2, JH-SRJ strain), were kindly provided by Zhejiang University. Positive nucleic acid samples of porcine deltacoronavirus (PDCoV, P25 GH7DQ0301 strain), swine acute diarrhea syndrome coronavirus (SADS-CoV), porcine rotavirus (PRoV), porcine bocavirus (PBoV), porcine hepatitis E virus (HEV), classical swine fever virus (CSFV, C strain), foot-and-mouth disease virus (FMDV, OHM/02 and AKT-111 strains), Japanese encephalitis virus (JEV, strain SA14-14-2), and the PEDV+TGEV bivalent live vaccine (WH-1R + AJ1102R), as well as nucleic acid samples of Salmonella enterica subsp. enterica (ATCC 9842, ATCC 13076, and ATCC 35640), were collected and preserved by our laboratory.
## Reagents and kits
Viral nucleic acids were extracted and purified using the TaKaRa MiniBEST Viral RNA/DNA Extraction Kit (Takara, Japan; Cat#9766). Multiplex PCR was performed using the Hifair V Multiplex One Step RT-PCR Kit (Yeasen, China; Cat#13089). dNTP digestion and extension reactions were performed using the IPLEX Universal Kit (Agena, USA). qPCR was performed using TB Green Premix Ex Taq (2×) (Takara, Japan; Cat#RR420). Clinical samples were subjected to nucleic acid extraction and purification using the MagNA Pure 24 Instrument and the corresponding Total NA Isolation Kit (Roche, Switzerland; Cat#07658036001).
## Primers and UEPs
Based on the genome sequences of each of the above pathogens retrieved from the NCBI database, primers for the conserved regions of the genes and corresponding unextended probes (UEPs) were designed. Sequence analysis and multiple sequence alignment were conducted using CLC Genomics Workbench 23 (Qiagen, Germany) and MEGA-X (Auckland, New Zealand). Primer and UEP designs were performed using Primer3 Plus (https://www.primer3plus.com) and MassARRAY Assay Design Suite (Agena, USA). For TGEV, primers were specifically designed to target the S gene, ensuring specificity by avoiding its PRCV variant. For HEV, a single-base extend site was designed to differentiate between HEV-3 (A, 5560.6 Da) and HEV-4 (C, 5536.6 Da) based on single nucleotide polymorphisms (SNPs) at the 3′ end (Fig. S1). PBoV genotypes (G1-G3) were determined from complete genome phylogenetic analysis (neighbor-joining, 1,000 bootstraps, Fig. S2). Genotype-specific primers and UEPs were then designed from conserved regions in either NS1 or VP1 to maximize inclusivity and specificity. Addition ally, the porcine RPL4 gene (Accession number: XM_005659862) was used as the internal controls in this study.
To prevent potential peak interference caused by multiple primers in mass spectra, a 10-base tag (ACGTTGGGATG, highlighted in bold in Table 1) was appended to the 5′ end of each PCR primer. The primers and probes sequences were designed by the Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast) and MFEprimer 4 (https://m4.igenetech.com). The sequences of all primers and UEPs are listed in Table 1. All oligonucleotides were synthesized and purified by Sangon Biotech (China).
## Construction of standard plasmids
The target gene sequences corresponding to the primer design regions for each virus were synthesized and cloned into the pUC-57 plasmid vector (Sangon Biotech, China). These recombinant plasmids served as standard templates to evaluate the performance of the detection system in this study. Details of the standard plasmids are listed in Table 2.
The copy numbers of each plasmid were quantified using Qubit 4 (Thermo Fisher, USA). Each standard plasmid was initially diluted to 10 8 copies/μL and then serially diluted tenfold to create a range of concentrations from 10 8 to 10 0 Copies/μL. For subsequent experiments, mixed plasmid solutions were prepared at each gradient concentration (10 7 to 10 0 Copies/μL) to serve as positive controls. Aliquots of these plasmid mixtures were stored for use in downstream assays.
## Optimization of reaction system and conditions
To optimize the reaction system, 10 4 copies/μL plasmid templates of each target were tested with the primers and UEPs listed in Table 1. Subsequently, a mixed plasmid sample containing equal concentrations of 11 targets at 10 4 copies/μL was used to optimize multiplex conditions. Initial concentrations of all primers and UEPs were set to 5 μmol/L, and adjustments were made to achieve uniform UEP peak intensities and an E SEP/UEP ≥ 0.8. Optimization included fine-tuning primer and UEP concentrations, annealing temperatures (55°C-60°C), and annealing times (20-35 s) in the multiplex PCR system. Negative controls using ddH 2 O were included in all experiments, and the presence of correct mass spectrum peaks was used to evaluate the results.
## PCR setup and reaction conditions
Each reaction mixture contained 2 μL of recombinant plasmid (10 4 copies/μL), 2.5 µL of 2 × Multiplex PCR buffer, 0.3 µL of enzyme mix, and 0.2 µL of each target-specific primer.
The thermal cycling conditions were as follows: reverse transcription was performed at 50° for 10 min, initial denaturation at 95°C for 5 min, then 45 cycles of 95°C for 15 s, 60°C for 30 s, and finally 72°C for 5 min. After amplification, 2 µL of shrimp alkaline phosphatase (SAP) and reaction buffer were added to the PCR products for dNTP dephosphorylation. The mixture was incubated at 37°C for 40 min, followed by enzyme inactivation at 85°C for 5 min. For the extension reaction, the dephosphorylated products were mixed with 1 µL of UEP mix, 0.04 µL of Iplex enzyme, 0.2 µL of termination mix (containing ddNTP), 0.2 µL of Iplex buffer, and 0.56 µL of ddH 2 O. The extension program consisted of 40 cycles of 30 s at 95°C, 5 s at 95°C, 5 s at 52°C, 5 s at 80°C, and a final extension of 3 s at 72°C. The UEP will specifically bind to the target sequence in the PCR product, and the chain termination reaction using ddNTPs produced the SEP. The SEPs were transferred to a 384-well plate, diluted to 25 µL with ddH 2 O, and centrifuged at 8,000 rpm for 2 min. The prepared plate, along with the inert matrix chip, was loaded into the DP-TOF mass spectrometer (Digena, China). Pre-edited assay files containing molecular weight information for each UEP and SEP were imported into the instrument software. After completing sample and plate setup, the resin was programmed to purify the products for mass spectrometry analysis. The results were evaluated based on the correct positioning of peaks in the mass spectrum. Detection of the corresponding UEP and SEP peaks for each target was used to confirm successful amplification and extension.
## Analytical specificity and sensitivity
Analytical specificity of NAMS was assessed by testing the assay against a panel of non-target pathogens to confirm absence of cross-reactivity. Non-target nucleic acids from inactivated CSFV, FMDV, PRRSV, PRV, and PCV-2 were used as negative controls, while a plasmid mixture at 10 4 copies/μL served as the positive control, and ddH 2 O was used as the blank control. The analytical sensitivity of NAMS was assessed by preparing a twofold serial dilution of the plasmid mixture (initial concentration: 100 copies/μL) to create a series of concentrations, including 50, 25, 12.5, 6.2, 3.1, and 1.6 copies/μL. The limit of detection (LoD) was defined as the lowest concentration at which all replicates (n = 10) tested positive.
## Repeatability and reproducibility
For repeatability evaluation, plasmid mixtures at high, medium, and low concentrations (10 6 , 10 4 , and 10 2 copies/μL, respectively) were tested in 20 replicates under the same conditions. For reproducibility assessment, two additional independent experiments were conducted seven days apart, resulting in three independent experimental batches. Results from these batches were compared to evaluate consistency across different runs and time points.
## Evaluation of clinical sample results by MALDI-TOF NAMS
To validate the ability of the MALDI-TOF NAMS method to detect multiple pathogens in clinical samples, a total of 242 diarrhea samples of various types were evaluated using both the MALDI-TOF NAMS and qPCR methods. The samples included 97 fecal samples, 132 tissue samples, and 13 serum samples, all of which were collected from the Zhejiang Institute of Inspection and Quarantine Science and Technology between 2023 and 2024. All discordant results between the two methods were further analyzed using digital PCR (dPCR) for confirmation.
For sample preparation, blood samples (1-2 mL) were directly used for nucleic acid extraction. Fecal samples (1-2 g) were tenfold diluted with PBS, vortexed thoroughly, and centrifuged at 10,000 rpm for 5 min, after which the supernatant was collected. Tissue samples (1-2 g) were homogenized in 5-10 mL of 10 mM phosphate-buffered solution (PBS, pH 7.2) at a mass-to-volume ratio of 1:5. The homogenate was processed using a tissue grinder, followed by centrifugation at 10,000 rpm for 5 min, and the supernatant was collected.
Nucleic acid extraction and purification were performed using the MagNA Pure 24 Instrument and the corresponding kit. Subsequently, the optimized conditions for multiplex reverse transcription PCR, SAP digestion, and UEP extension were applied. Each sample was tested in triplicate, with ddH 2 O and a 10 4 copies/μL plasmid mixture serving as negative and positive controls, respectively.
## qPCR reference assay
For comparison, commercial qPCR kits (Vipotion, China; Cat#SD-R-0411, SD-R-0511, SD-R-0611, SD-R-0911, MQ-R-5411) were used for the detection of PEDV, TGEV, PoRV, PDCoV, and HEV, according to the manufacturer's instructions. For Salmonella and PBoV, published primers (Table S1) were used. qPCR was performed using TB Green Premix Ex Taq for single-cycle amplification, with a reaction volume of 20 µL containing 2 µL template, 10 µL master mix, and 0.2 µM primers. Cycling conditions were as follows: 95°C for 2 min; 40 cycles of 95°C for 10 s and 60°C for 30 s. Samples with Ct values below 38 were considered positive.
## RESULTS
## Single-target MALDI-TOF NAMS assay
The mass spectrometry profiles using single plasmids as templates (Fig. 1) demonstrated that each tested standard plasmid produced a distinct SEP peak at the predicted mass position (black), while the corresponding blank controls detected only UEP peaks (red). The UEP and SEP peaks can be clearly distinguished for the specific mass of each target. The positions of both peaks were consistent with those listed in Table 1, confirming the specificity and functionality of the designed primers and UEPs.
## Multiplex detection
Using a mixed plasmid template (10 4 copies/μL for each target), the system's ability to detect multiple targets simultaneously was evaluated. Figure 2 illustrates the mass spectra obtained under optimized reaction conditions. All 11 SEP peaks were detected at their expected mass positions, with minimal overlap or interference, demonstrat ing the successful multiplex amplification and extension of the UEPs. Additionally, no extraneous peaks or significant variations in peak height were observed, affirming the assay's compatibility in a multiplex setting. To ensure uniform signal detection, the initial concentrations of primers and UEPs were adjusted during optimization (Table 3).
## Analytical specificity
Using genomic DNA or cDNA from 10 target pathogens and five non-target controls as templates, the extension reaction products were analyzed via mass spectrometry after amplification. As shown in Fig. 3, SEP peaks corresponding to PDCoV, PEDV, TGEV, SADS-CoV, HEV, PoRV, PBoV-G1/G2/G3, and Sal are exclusively detected in reactions using their respective target pathogens or mixed plasmids as templates. In addition,
## FIG 2
The black signal peak used 10 3 copies/μL of the mixed plasmid as a template to simulate the presence of all targets. The red signal peaks were templated with ddH 2 O. The UEP and SEP of each target were labeled using their respective colors, where SEP was indicated using the bases (A, C, or T) extended by the target. The SEP peaks and UEP peaks can be clearly distinguished in the NAMS images. UEP, un-extended probe; SEP, single-base extended products; PDCoV, porcine deltacoronavirus; PEDV, porcine epidemic diarrhea virus; TGEV, transmissible gastroenteritis virus; SADS-CoV, swine acute diarrhea syndrome coronavirus; PoRV, porcine rotavirus; HEV-3/4, hepatitis E virus genotype 3/4; PBoV-G1/G2/G3, porcine bocavirus group 1/2/3; Sal, Salmonella. depending on the extended bases, HEV-3 and HEV-4 are able to be distinguished by a single UEP (Fig. 1A). In contrast, no amplification occurred for non-target pathogens (including CSFV, FMDV, PRRSV, PRV, and PCV 2) or blank controls. The mass spectra of these non-target samples displayed only UEP peaks, confirming the high specificity of the MALDI-TOF NAMS assay for the selected target pathogens. This specificity ensures that the assay can reliably differentiate target pathogens within complex biological samples without cross-reactivity from non-target microorganisms, highlight ing its robustness and accuracy in diagnostic applications.
## Analytical sensitivity
The analytical sensitivity of the MALDI-TOF NAMS assay was evaluated using serially diluted mixed plasmids at concentrations ranging from 100 copies/μL to 1.6 copies/μL. All targets were reliably detected at 12.5 copies/μL. Probit regression analysis was then performed to calculate the LoD (99% detection probability) and its 95% confidence intervals based on 10 replicates at each concentration. The results, shown in Table 4, indicate that the LoD values for all targets ranged from 12.20 copies/μL to 33.59 FMDV, foot-and-mouth disease virus; JEV, Japanese encephalitis virus; PCV-2, porcine circovirus 2; PRRSV, porcine reproductive and respiratory syndrome virus;
PRV, pseudorabies virus.
copies/μL, highlighting the assay's high sensitivity and reliable performance at low nucleic acid concentrations.
## Repeatability and reproducibility
The repeatability and reproducibility of the MALDI-TOF NAMS assay were evaluated using mixed plasmid templates at high, medium, and low concentrations. At each concentration, the detection rate for all targets was 100% (20/20 per concentration, totaling 60/60). In two subsequent independent experiments, all targets were also detected with 100% positivity (60/60 per batch).
## Evaluation of the MALDI-TOF NAMS on three types of samples
The performance of the MALDI-TOF NAMS method was further validated using 242 clinical samples, including feces (n = 97), tissue (n = 132), and serum (n = 13) specimens. Overall, 192 samples tested positive for at least one target pathogen. Notably, while TGEV was not detected in any sample type, all other targets were identified to varying degrees (Fig. 4; Table 5). The results of clinical samples showed the distinct SEP peaks corresponding to detected pathogens (Fig. 5).
To assess the accuracy of MALDI-TOF NAMS, qPCR was used as a reference method, and the concordance rates between the two methods were analyzed (Table 5). The statistics for HEV include HEV-3 and HEV-4, and PBoV includes the combined results for all genotypes. The results revealed that the total concordance rates for individual targets ranged from 92.6% to 100.0% (except TGEV). When data for all targets were combined, the total concordance rates for different sample types ranged from 93.2% to 98.3%, with an overall concordance rate of 96.2% for all 242 samples. These findings demonstrate a high level of agreement between MALDI-TOF NAMS and qPCR, supporting the reliability and accuracy of the newly developed assay for detecting multiple pathogens in clinical diarrhea samples.
## DISCUSSION
Porcine diarrhea syndrome (PDS) disease poses a serious threat to the health of pig herds and the development of the pig industry worldwide, leading to significant economic losses, especially in the import and export trade (21). The complex landscape of diarrheal pathogens in pigs, coupled with the emergence of novel pathogens, has heightened the challenge of disease prevention and control. Thus, there is an urgent need for new diagnostic techniques capable of addressing mixed infections and detecting a wide range of pathogens efficiently. This study aimed to evaluate the performance of the nucleic acid mass spectrometry (NAMS) method in detecting pathogens associated with PDS. Our findings demonstrated the capability of NAMS to achieve high-throughput, multiplex pathogen detection with high sensitivity and specificity, addressing some of e The total compliance rate is calculated for PBoV by the results of G1, G2 and G3 typing tests.
the limitations of existing diagnostic methods, such as ELISA, qPCR, and NGS. These results underscore the potential of NAMS as a robust tool for large-scale pathogen surveillance and disease control. Our analysis of 242 samples revealed an overall positivity rate of 81.4%, 77.3%, and 84.6% in fecal, tissue, and serum samples, respectively (Table 5). Notably, mixed infections were prevalent, with 39.2% of positive samples exhibiting co-detection of multiple pathogens. The proportion of multiple detections reached 61.9% (60/97) in fecal samples, 27.3% (36/132) in tissue, and 69.2% (9/13) in serum. Specifically, our findings revealed that HEV and Salmonella were exclusively detected in mixed infec tions, suggesting their potential role in exacerbating disease severity. Moreover, PBoV, frequently detected in co-infections with PDCoV (41 cases), PoRV (36 cases), and PEDV (29 cases), emerged as a key player in mixed infections, consistent with its known role as an auxiliary pathogen (6). These findings emphasize the complexity and diversity of pathogen interactions in the context of porcine diarrheal syndromes.
From the results of individual viruses, PEDV, the main pathogen of porcine diarrhea outbreaks, had a significantly higher positive rate than other pathogens. In this study, the positivity rate for PEDV was as high as 40.9% (99/242), with single-pathogen infections predominantly observed in tissue samples (90.7%, 49/54). This aligns with reports of PEDV's high virulence and association with severe outbreaks in piglets (22). In addition, studies have shown that the use of commercial vaccines may accelerate the evolution of PEDV, resulting in multiple PEDV infections in the same pig farm (23). This highlights the severity of PEDV epidemics and the urgency for improved immunization and prevention strategies. As the other major causative virus of porcine diarrhea, PDCoV was primarily detected in fecal samples (95.7%, 44/46). Clinical symptoms of PDCoV infection resemble those of PEDV and TGEV but are less severe in piglets (24). Therefore, PDCoV often co-infects with other pathogens, such as PEDV and PoRV (25). This was confirmed in the present study, where the co-infection rates of PDCoV with PEDV and PoRV were 21.7% and 23.9%, respectively. PoRV infections, unlike PEDV and TGEV, generally cause sporadic diarrhea with lower mortality (26). Recent investigations (2022-2023) reported that the positivity rate of PoRV in Chinese pig farms reached 86.5%, with a sample positivity rate of 51.2% (27). In this study, PoRV was detected across all three sample types, with positivity rates of 25.8% (25/97) in fecal samples and 25.0% (33/132) in tissues, both lower than previously reported data. However, PoRV positivity rose to 53.8% (7/13) in sera, potentially due to the small sample size of serum tested.
The SADS-CoV was detected at a low frequency in this instance (5.8%, 140/242), mainly as a single pathogen (71.4%, 10/14). This may be related to its sporadic regional distribution, and as a diarrheal pathogen with a high mortality rate, further surveillance is needed (28,29). TGEV was once a major pathogen causing diarrhea in pigs, lead ing to significant economic losses in the global swine industry. The absence of TGEV in this study corroborates reports of its reduced prevalence due to the widespread circulation of its natural deletion mutant strain, PRCV (30,31). To minimize interference from PRCV in detection, the TGEV target region in this study was designed within the PEDV S gene region where PRCV exhibits a deletion. This strategy will effectively enhance detection specificity and reduce the impact of PRCV on the diagnosis of diarrhea-associated pathogens. It should be noted that in qPCR validation, one fecal sample tested positive for TGEV/SADS-CoV (Ct = 36.2/18.7), with only SADS-CoV detected by the NAMS method. A later digital droplet PCR (ddPCR) confirmed TGEV/SADS-CoV positivity (positive/total droplet count = 29/262,400). Earlier specificity testing with TGEV inactivated strains (WH-1R) and bivalent PEDV/TGEV live vaccines (WH-1R + AJ1102R) consistently yielded expected positive results. In addition, all the remaining samples were verified as TGEV-free using qPCR. Thus, the absence of TGEV detection in this sample is likely due to its proximity to the NAMS detection limit (10.71-45.40, 95% confidence level). This limitation does not compromise the reliability of TGEV detection in most cases of porcine diarrhea. All Salmonella-positive samples were associated with co-detection of multiple pathogens, a scenario frequently encountered in fecal samples from complex composite infections (8).
Beyond the major diarrheal pathogens, PBoV exhibited a high overall detection rate (39.7%) across all sample types. Among the positive cases, the G1 genotype of PBoV was the most frequently detected (23.5%-55.7%), consistent with previous findings (4). While the pathogenic mechanism of PBoV remains unclear, studies indicate frequent co-infec tions with other pathogens, such as PRRSV, PCV2, and CSFV, with rising prevalence in recent years (6, 32). PBoV has also been detected in both clinically healthy pigs and those with respiratory or enteric diseases worldwide, leading some researchers to propose that PBoV may act as an auxiliary virus rather than a primary pathogen (6). In this study, co-infection rates of PBoV with other diarrhea-related pathogens were 38.0%, significantly higher than the 1.2% in healthy pigs, suggesting a potential role for PBoV in porcine diarrhea. Additionally, a substantial proportion of samples contained two or more PBoV genotypes, indicating widespread cross-infections among PBoV subtypes in hosts. Previous studies have also demonstrated that multiple PBoV genotypes frequently co-circulate, with individuals often simultaneously infected with several PBoV strains (32,33). The potential of PBoV for cross-species transmission to humans further underscores its significance as a potential public health concern (7). The detection rate of HEV (9.1%) was much lower than that of PBoV for the same severe mixed infections. HEV-3 was exclusively detected in fecal samples, while HEV-4 was identified across all sample types, with detection rates ranging from 3.8% to 10.3%. As a zoonotic pathogen with a broad host range, the correlation between HEV prevalence and multiviral co-infections emphasizes the need for targeted surveillance on diarrhea-endemic farms (9, 34).
PEDV, TGEV, SADS-CoV, and PDCoV all belong to the Coronaviridae family. Coronavi ruses have garnered significant attention due to their broad host range and potential for cross-species transmission (2,35). The emergence of coronaviruses, such as MERS-CoV, SARS-CoV, and SARS-CoV-2, serves as a stark reminder of the risks associated with inter-species spillover (36,37). For instance, PDCoV, first identified in 2012, is thought to have originated from an interspecies transmission event involving avian and mammal coronaviruses, as suggested by phylogenetic analyses. Research has shown that PDCoV utilizes its spike (S) protein to bind with host aminopeptidase N (APN), enabling it to infect cells from pigs, chickens, and humans. The highly conserved nature of APN across species likely plays a critical role in facilitating PDCoV's cross-species infectivity (35). Notably, PDCoV is not the only enteric coronavirus that leverages APN. PEDV and TGEV also depend on this receptor, suggesting a potential commonality in the mechanisms underlying host switching in coronaviruses (37,38). Moreover, studies indicate that SADS-CoV can effectively replicate in various mammalian cell lines, including primary human lung and intestinal cells, further underscoring the potential for interspecies transmission (28). Similar to HEV and PBoV, the widespread distribution and zoonotic potential of swine-origin coronaviruses necessitate comprehensive investigations into their epidemiology and transmission dynamics. Such efforts are critical for assessing and mitigating potential public health risks associated with these pathogens.
In this study, the advantages of NAMS were utilized for SNP genotyping of HEV-3 and HEV-4 genes (Fig. 6). Similarly, SNP genotyping can be further employed to distinguish between pathogen variants, such as virulent versus attenuated strains and wild-type versus vaccine strains. On the other hand, the successful detection of the Salmonella target demonstrates that the NAMS method can simultaneously detect RNA viruses, DNA viruses, and bacteria. This significantly broadens the scope of pathogen detection and its application scenarios, particularly for diagnosing syndromes with unknown etiologies. This capability relies on selecting an appropriate nucleic acid extraction method. In future research, leveraging the scalability of targets in the NAMS method, we can further expand its application to include bacterial pathogens causing porcine diarrhea, aiming to establish a more comprehensive diagnostic system for porcine digestive tract diseases. It should be noted that, due to the relatively small number of clinical samples collec ted, the distribution of certain pathogens may only reflect regional epidemiological characteristics and farm management practices and may not fully reflect the diversity of cases in the real world (e.g., Salmonella and hepatitis E virus).
Although portable mass spectrometry devices have been proposed, current MALDI-TOF MS platforms remain relatively large and laboratory-based, making this method better suited to centralized surveillance rather than true field deployment. From a cost-benefit perspective, while the instrumentation is relatively costly and may not outperform conventional qPCR in small-scale testing, the MALDI-TOF NAMS approach provides clear advantages in high-throughput, multiplex applications. The platform can analyze up to ~40 targets simultaneously (39), and the use of unmodified UEPs enables low-cost panel expansion, improving the economics per target as panel size and sample volume increase. These features make the method particularly advantageous for routine monitoring, mixed-infection detection, and comprehensive pathogen surveillance in centralized laboratory settings. Future advances in miniaturized MS technology may further enhance its applicability in resource-limited or field environments.
## Conclusion
In conclusion, this study demonstrates the potential of NAMS as a promising tool for the detection and genotyping of swine diarrhea pathogens. While the method shows strong sensitivity and specificity, it remains an emerging technology that complements traditional diagnostic approaches, such as qPCR and NGS. The ability to perform multiplexed detection of these pathogens in a single assay not only enhances diagnostic accuracy but also allows for a comprehensive understanding of co-infections, which are common in clinical settings. This is particularly useful for complex mixed infections, where traditional single-target approaches tend to complicate diagnosis. The broader potential of NAMS lies in its application to pathogen monitoring, especially in epide miological surveillance, early detection of emerging infectious diseases across animal populations, and its possible integration into entry-exit quarantine processes. It provides a new means for more efficient and accurate pathogen screening, which is vital for controlling the spread of animal diseases across borders.
## References
1. Jacobson (2022) "On the infectious causes of neonatal piglet diarrhoea -a review" *Vet Sci*
2. Liu, Wang (2021) "Porcine enteric coronaviruses: an updated overview of the pathogenesis, prevalence, and diagnosis" *Vet Res Commun*
3. Curry, Schwartz, Yoon et al. (2017) "Effects of porcine epidemic diarrhea virus infection on nursery pig intestinal function and barrier integrity" *Vet Microbiol*
5. Zheng, Cui, Qiao et al. (2021) "Detection and genetic characteristics of porcine bocavirus in central China" *Arch Virol*
6. Prpić, Keros, Božiković et al. (2024) "Current Insights into porcine bocavirus (PBoV) and its impact on the economy and public health" *Vet Sci*
7. Aryal, Liu (2021) "Porcine bocavirus: a 10-year history since its discovery" *Virol Sin*
8. Safamanesh, Azimian, Shakeri et al. (2018) "Detection of porcine bocavirus from a child with acute respiratory tract infection" *Pediatr Infect Dis J*
9. Soliani, Rugna, Prosperi et al. (2023) "Salmonella infection in pigs: disease, prevalence, and a link between swine and human health" *Pathogens*
10. Aslan, Balaban (2020) "Hepatitis E virus: epidemiology, diagnosis, clinical manifestations, and treatment" *World J Gastroenterol*
11. Zhang, Luo, Lin et al. (2024) "Epidemio logical monitoring and genetic variation analysis of pathogens associated with porcine viral diarrhea in southern China from 2021 to 2023" *Front Microbiol*
12. Mainquist-Whigham, Mauch-Swinford, Stephenson et al. (2025) "Risk factors associated with prolonged infection of porcine reproductive and respiratory syndrome virus determined by whole-herd sampling methods" *JSHAP*
13. Ellwanger, Veiga, Da et al. (2021) "Control and prevention of infectious diseases from a one health perspective" *Genet Mol Biol*
14. Singhal, Kumar, Kanaujia et al. (2015) "MALDI-TOF mass spectrometry: an emerging technology for microbial identification and diagnosis" *Front Microbiol*
15. Gao, Tan, Sugrue et al. (2013) "MALDI mass spectrometry for nucleic acid analysis" *Top Curr Chem*
16. Arbefeville, Timbrook, Garner (2024) "Evolving strategies in microbe identification-a comprehensive review of biochemical, MALDI-TOF MS and molecular testing methods" *J Antimicrob Chemother*
17. Patel (2015) "MALDI-TOF MS for the diagnosis of infectious diseases" *Clin Chem*
18. Liu, Kang, Li et al. (2022) "Simultaneous detection of seven human coronaviruses by multiplex PCR and MALDI-TOF MS" *Covid*
19. Xiu, Zhang, Li et al. (2019) "Simultaneous detection of eleven sexually transmitted agents using multiplexed PCR coupled with MALDI-TOF analysis" *Infect Drug Resist*
20. Yang, Li, Dang et al. (2023) "A rapid, accurate, and low-cost method for detecting Mycobacte rium tuberculosis and its drug-resistant genes in pulmonary tuberculosis: applications of massARRAY DNA mass spectrometry" *Front Microbiol*
21. Shuai, Song, Wang et al. (2024) "MALDI-TOF nucleic acid mass spectrometry for simultaneously detection of fourteen porcine viruses and its application" *J Virol Methods*
22. Vanderwaal, Deen (2018) "Global trends in infectious diseases of swine" *Proc Natl Acad Sci*
23. Zhang, Zou, Peng et al. (2023) "Global dynamics of porcine enteric coronavirus PEDV epidemiology, evolution, and transmission" *Mol Biol Evol*
24. Zhang, Qing, Yan et al. (2023) "Investigation and analysis of porcine epidemic diarrhea cases and evaluation of different immunization strategies in the large-scale swine farming system" *Porcine Health Manag*
25. Lorsirigool, Saeng-Chuto, Temeeyasen et al. (2016) "The first detection and full-length genome sequence of porcine deltacoronavirus isolated in Lao PDR" *Arch Virol*
26. Saeng-Chuto, Madapong, Kaeoket et al. (2021) "Coinfection of porcine deltacoronavirus and porcine epidemic diarrhea virus increases disease severity, cell trophism and earlier upregulation of IFN-α and IL12" *Sci Rep*
28. Gao, Shen, Zhao et al. (2024) "Isolation and pathogenicity analysis of a G5P[23] porcine rotavirus strain" *Viruses*
29. Qiao, Li, Li et al. (2024) "Recent molecular characterization of porcine rotaviruses detected in china and their phylogenetic relation ships with human rotaviruses" *Viruses*
30. Edwards, Yount, Graham et al. (2020) "Swine acute diarrhea syndrome coronavirus replication in primary human cells reveals potential susceptibility to infection" *Proc Natl Acad Sci*
31. Gong, Li, Zhou et al. (2017) "A new bat-HKU2-like coronavirus in swine" *Emerg Infect Dis*
32. Bedsted, Rasmussen, Martinenghi et al. (2024) "Porcine respiratory coronavirus genome sequences; comparisons and relationships to transmissible gastroenteritis viruses" *Virology (Auckl)*
33. Chen, Zhang, Chu et al. (2023) "Prevalence of transmissible gastroenteritis among swine populations in China during 1983-2022: a systematic review and meta-analysis" *Microb Pathog*
34. Zhang, Huang, Liu et al. (2011) "Porcine bocaviruses: genetic analysis and prevalence in Chinese swine population" *Epidemiol Infect*
35. Jiang, Xiao, Yin et al. (2014) "High prevalence and genetic diversity of porcine bocaviruses in pigs in the USA, and identification of multiple novel porcine bocaviruses" *J Gen Virol*
36. Salines, Dumarest, Andraud et al. (2019) "Natural viral co-infections in pig herds affect hepatitis E virus (HEV) infection dynamics and increase the risk of contaminated livers at slaughter" *Transbound Emerg Dis*
37. Li, Hulswit, Kenney et al. (2018) "Broad receptor engagement of an emerging global coronavirus may potentiate its diverse cross-species transmissibility" *Proc Natl Acad Sci*
38. Cui, Li, Shi (2019) "Origin and evolution of pathogenic coronavi ruses" *Nat Rev Microbiol*
39. Wang, Liu, Ji et al. (2018) "Porcine deltacoronavirus engages the transmissible gastroenteritis virus functional receptor porcine aminopeptidase n for infectious cellular entry" *J Virol*
40. Li, Li, De Esesarte et al. (2017) "Cell attachment domains of the porcine epidemic diarrhea virus spike protein are key targets of neutralizing antibodies" *J Virol*
41. Anon (2018) "Chinese expert consensus group on the application of MALDI-TOF MS. Chinese expert consensus on the application of MALDI-TOF MS" *National Medical Journal of China* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12008908&blobtype=pdf | # Retraction Note: Reduced expression of Jak-1 and Tyk-2 proteins leads to interferon resistance in hepatitis C virus replicon
Salima Luftig, Robert Haque, Sander Garry, Flor, Lizeth Taylor, Sidhartha Hazari, Srikanta Dash, Salima Haque, Robert Garry, Sander Florman, Ronald Luftig, Frederic Regenstein |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12254312&blobtype=pdf | # Author Correction: A comprehensive overview of the burden, prevention, and therapeutic aspects of arboviral diseases in India Check for updates
Himanshu Gupta, Pradip Barde, Pal Mrigendra, Praveen Singh, Nitika Bharti, Nitika, Medicine |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12874598&blobtype=pdf | # A brief review of chikungunya fever: From molecular virology to countermeasures
Xin Zhang, Xiaoxi Li, Tianjun Jiang, Junliang Fu
## Abstract
Chikungunya fever (CHIKF), resulting from the chikungunya virus infection, has become a major global health issue in recent years. This review summarizes the epidemiology, virology, pathogenesis, clinical manifestations, diagnosis, vaccine development and treatment strategies of the CHIKF. And by integrating the findings from various studies, an attempt is made to propose future research directions and intervention strategies.
IntroductionChikungunya virus (CHIKV), a mosquito-transmitted alphavirus, is the etiological agent of chikungunya fever (CHIKF) and constitutes an increasingly prominent global public health concern. CHIKF is clinically marked by a sudden high fever, maculopapular rash, headache, myalgia, and severe, often debilitating, polyarthralgia. Although the acute phase is generally self-limiting, 30% to over 60% of patients in certain cohorts develop chronic inflammatory rheumatological conditions that may last for months or years. This chronic phase, marked by ongoing joint pain, stiffness, and swelling, presents a major public health issue, leading to considerable declines in quality of life and economic productivity. 1 The virus mainly spreads to humans via bites from infected female mosquitoes, especially Aedes aegypti and Aedes albopictus . The Aedes albopictus , also known as "Asian tiger mosquito ", with its extensive geographic range and adaptability to cooler climates, has played a crucial role in the global spread of CHIKV beyond tropical regions. Since the early 2000s, CHIKV has undergone a dramatic global resurgence. In the first half of 2025, more than 220,000 cases and 80 fatalities were recorded worldwide across 14 nations, recent local transmission in subtropical urban regions of China, notably in Guangdong Province, reported 16,452 confirmed cases as of 27 September 2025, emphasizes the virus's adaptability to new environments 2-3 . This review offers an in-depth analysis of recent CHIKV research, focusing on its evolving epi-
demiology and viral evolution, the molecular mechanisms of replication and host-virus interaction, the pathogenesis, advancements in diagnostics, vaccines, and antiviral treatments.
## 2. Epidemiology
## 2.1. Global distribution and outbreak patterns
Historically, CHIKV was mainly found in Africa and Asia, causing sporadic outbreaks. The global epidemiology of CHIKV has been characterized by its rapid geographic expansion and the occurrence of largescale, explosive outbreaks over the past two decades ( Fig. 1 ).
In Africa, the continent of origin for CHIKV, the virus continues to circulate, causing both epidemic and endemic disease. A study in the Republic of Senegal has provided evidence for the re-emergence of CHIKV as an endemic, locally sustained transmission cycle, rather than a series of re-introductions from elsewhere. 4 Asia has also experienced major CHIKV outbreaks. The virus is also being detected in new regions; a study identified the presence of the West African genotype of CHIKV in China, highlighting the risk of viral importation and establishment in non-endemic areas through international travel. 5 In the People's Republic of Bangladesh (Bangladesh), seroprevalence surveys have been conducted to estimate the true burden of infection in the population, often revealing that a much larger pro-Fig. 1. Geographical distribution of CHIKV disease cases, data from January to September 2025. 3 * : excluding Hong Kong SAR, Macao SAR, and Taiwan, China. Abbreviations : SAR, special administrative region; Comoros, Union of the Comoros; Kenya, the Republic of Kenya; Mauritius, the Republic of Mauritius; Senegal, the Republic of Senegal; Pakistan, the Islamic Republic of Pakistan; Somalia, the Federal Republic of Somalia; France, the French Republic; Italy, Repubblica Italiana; Bolivia, Plurinational State of Bolivia; Brazil, the Federative Republic of Brazil; Cuba, the Republic of Cuba; India, the Republic of India; Bangladesh, the People's Republic of Bangladesh; Sri Lanka, the Democratic Socialist Republic of Sri Lanka; Thailand, the Kingdom of Thailand; Indonesia, Republic of Indonesia; Philippines, Republic of the Philippines; Singapore, Republic of Singapore. portion of people have been infected than is captured by official case reports. 6 In South America, the virus has become firmly established. Investigations in urban settings in the Federative Republic of Brazil (Brazil) have detailed the intense transmission dynamics of the East/Central/South African (ECSA) lineage of CHIKV. 7 Recent outbreaks, such as a major epidemic in Republic of Paraguay, underscore the virus's capacity for rapid and widespread transmission in immunologically naive populations. 8 Co-infections between CHIKV and Mayaro virus, O'nyong-nyong virus or Dengue virus have been reported, which can complicate clinical diagnosis and management. 9-13
## 2.2. Vector ecology and zoonotic cycle
A key driver of CHIKV's global success is its ability to adapt to different mosquito vectors. Specific variants of CHIKV have been shown to affect the transmission competence of Aedes aegypti . 14 At the vector level, the regulation of oxidative stress within the mosquito has been shown to be a key modulator of its antiviral immune response. 15 The long-term persistence of CHIKV is believed to depend on a sylvatic, or zoonotic, cycle involving non-human hosts. For example, natural infection with CHIKV has been documented in wild lion tamarins ( Leontopithecus species) in Brazil, confirming their role as hosts in the neotropical sylvatic cycle. 16 Beyond primates, studies have detected neutralizing antibodies against CHIKV in bats and opossums, suggesting that these mammals may also participate in the virus's maintenance cycle in nature. 17
## 2.3. Risk factors and population susceptibility
Various risk factors contribute to CHIKV infection and the severity of the disease. Increased risk of severe disease and mortality is associated with older age, male sex, and specific comorbidities like diabetes mellitus, hypertension, and chronic kidney disease. 18 CHIKV infection in pregnant women is linked to a higher likelihood of obstetric complications. Research conducted in the United Mexican States (Mexico) identified chikungunya infection during pregnancy as an independent risk factor for obstetric complications, with an adjusted odds ratio of 1.6. 19 Smoking, particularly in males, is a recognized risk factor for severe arthralgia during both the acute and chronic phases of CHIKV infection. 20 While primarily mosquito-borne, evidence for other transmission routes has emerged. A notable finding is the prolonged presence of CHIKV in human semen. Studies have detected infectious virus in semen for up to 56 days post-infection, indicating the potential for sexual transmission of CHIKV and adding another dimension to its epidemiology and control. 21
## 3. Virology
## 3.1. Virus structure and genome
CHIKV is an alphavirus from the Togaviridae family, characterized by a spherical, enveloped structure and a positive single-stranded RNA genome. 22 The replication of the genome takes place in membranous replication structures known as "spherule ". The 5 ′ region of the genome encodes four non-structural proteins (nsP1-4), initially synthesized as the polyprotein P1234, assembled into viral replication complex (VRC), which is essential for viral RNA replication. 23 In the VRC, nsP4 functions as the RNA-dependent RNA polymerase, while nsP2 exhibits helicase activity and cleaves the nsPs from a viral polyprotein precursor to form mature VRC, nsP3 plays a part in bringing host factors to the spherule, nsP1 acts as the membrane anchor for the complex, creating dodecameric pores that connect with the membrane at the spherule necks to control their entry. 24 Mutations within nsP1 have been shown to attenuate viral virulence. 25 By suppressing NF-𝜅B activation, nsP3 helps the virus evade early host defenses and establish a foothold in the host. 26 The nsP4 protein serving as the central catalytic enzyme of the VRC. Mutations in nsP4 can have profound effects on viral replication and can also confer resistance to antiviral drugs, such as the nucleoside analog 4 ′ -fluorouridine. 27 The nsP2 has emerged as the most promising and intensely studied target for antiviral drug development. A wealth of research has focused on identifying small molecule inhibitors of the nsP2 protease. 28 One-third of the genome encodes structural proteins, such as the capsid protein and envelope glycoproteins E1 and E2. Host proteins SPCS3 and eIF3k interact with E1/E2 glycoproteins, regulating the production of new infectious viral particles. This indicates that the virus hijacks specific components of the host's protein synthesis and trafficking machinery to ensure efficient virion morphogenesis ( Fig. 2 ). 29 Fig. 2. The CHIKV genome organization, virion structure, and viral replication cycle. CHIKV attaches to host receptors (Step 1) and subsequently enters host cells via receptor-mediated endocytosis (Step 2). The acidic environment within endosomes triggers the fusion of the endosomal membrane with the viral envelope, releasing the nucleocapsid into the cytoplasm (Step 3). Disassembly of the nucleocapsid leads to the liberation of the viral genome, which then utilizes the host cell's translational machinery to translate the non-structural polyprotein (nsP1-4) (Step 4). The polyprotein is cleaved into separate nonstructural proteins by viral proteases, predominantly nsP2 (Step 5). Nonstructural proteins form VRCs on intracellular membranes and replicate negative-strand RNA (Step 6). Subgenomic RNAs translate polyprotein precursors (C-E3-E2-6K-E1), which is subsequently cleaved into individual structural proteins (Step 7). Capsid proteins assemble with positive-strand RNA to form nucleocapsids (Step 8). E1 and E2 are co-transported into the ER and undergo glycosylation (Step 9). Viral particles are assembled and released by budding through the plasma membrane, where they acquire an envelope embedded with viral glycoproteins (Step 10). The figure was adapted from References 107-108. [107][108] Abbreviations : CHIKV, Chikungunya virus; VRC, viral replication complex; ER, endoplasmic reticulum.
## 3.2. Host-virus interactions
The establishment of a successful infection relies on the virus's ability to co-opt a multitude of host cellular factors and pathways. For example, Catenin-𝛼-1 has been demonstrated to be critically important for a productive CHIKV infection. Co-immunoprecipitation shown that Catenin-𝛼-1 can interact with nsP2, and silencing its gene via small interfering RNA (siRNA) can significantly reduce viral particle formation and enhance the survival rate of infected cells. 30 The host protein BCL2 interacting protein 3 (BNIP3) has been shown to be a key regulator of the early stages of CHIKV replication. BNIP3 regulates CHIKV infection following virus entry and membrane hemifusion. Depletion of BNIP3 increases the expression of CHIKV proteins translated from the viral genomic and subgenomic RNAs. However, this effect is independent of the protein's functions in autophagy and cell death. 31 The matrix-remodeling associated 8 (MXRA8) protein has been confirmed as a key receptor that mediates the internalization of CHIKV into susceptible cells. 32 MXRA8 is predominantly expressed in dermal fibroblasts, bone marrow, and synovium. 33 In post-exposure treatment experiments, administration of anti-MXRA8 monoclonal antibody at 8 or 24 hours post viral inoculation reduced CHIKV infection in the ankles and muscles. 34 Conversely, host cells also possess intrinsic defense mechanisms that can restrict viral infection. The T-cell immunoglobulin and mucin domain-containing protein 1 has been found to inhibit the release of newly formed CHIKV particles from infected cells. 35 The host protein nucleophosmin 1 has been identified as an antiviral factor that suppresses CHIKV replication through its role in modulating the expression of interferon-stimulated genes (ISGs). 36
## 3.3. Viral expansion and adaptation
CHIKV has transitioned from its ancestral sylvatic foci in Sub-Saharan Africa to urban transmission cycles. Phylogenetic research has identified three primary genotypes of CHIKV: West African, Asian, and ECSA. The ECSA genotype strains are linked to higher viremia and more severe symptoms compared to the Asian genotype. 37 In 2004, CHIKV re-emerged in East Africa and subsequently spread worldwide, leading to epidemics in the Indian Ocean islands, Asia, Europe, and Oceania. The epidemic strains were mainly of the ECSA genotype. In 2013, the Asian genotype strains spread to the Americas, causing widespread outbreaks. 38 Acquired key mutations enhanced its fitness in the Aedes albopictus mosquito vector. 39 The E1-A226V mutation, which facilitates viral dissemination from the mosquito midgut. 40 The cumu-lative effect of these genetic changes has been the transformation of CHIKV into a more efficient and widespread pathogen, capable of causing explosive outbreaks far beyond its historical geographical limits. 41
## 4. Immune response and pathogenesis
The immune response to CHIKV infection is intricate, involving both innate and adaptive components. The innate immune system serves as the initial defense mechanism against viral infections. Type I interferons (IFNs) are central to this response. Interferon-induced transmembrane proteins (IFITMs), a subset of ISGs, are essential in limiting CHIKV infection. IFITMs, whose expression is triggered by toll-like receptor signaling, inhibit viral entry, illustrating a crucial mechanism of innate immune control over viruses. 42 In the adaptive immune response, antibodies play a crucial role. IgG enhances the neutralizing ability of human immune sera, with a positive baseline CHIKV plaque reduction neutralization test (PRNT) titer linked to protection against symptomatic CHIKV infection, as observed in a study conducted in Republic of the Philippines. 43 CHIKV's strategies for immune evasion, with an emphasis on type I IFN responses. The cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes pathway is crucial for initiating interferon gene expression in response to tissue damage, cellular stress, and infections. CHIKV infection leads to a significant decrease in cGAS expression. CHIKV nsP2 and E1/E2 suppress the activation of the IFN 𝛽-promoter that is triggered by the melanoma differentiation-associated protein 5/retinoic acid-inducible gene I receptor signaling pathway. 44 The nsP3 protein plays a direct role in suppressing the NF-𝜅B pathway, which is critical for producing pro-inflammatory cytokines and antiviral molecules. 26 Another key strategy for immune evasion involves the disruption of antigen presentation. The nsP2 disrupts the major histocompatibility complex class I pathway. The virus evades cytotoxic T lymphocyte detection and destruction by inhibiting the presentation of viral antigens on infected cell surfaces, thereby enhancing its persistence and replication. 45 CHIKV can create stable tunnels between cells, shielding it from neutralizing antibodies and promoting effective intercellular transmission both in vitro and in vivo . 46 The inflammatory response in the joints is driven by a massive infiltration of immune cells. Macrophages expressing high levels of CD64 (Fc 𝛾RI) are key players in the inflammatory cascade within affected joints. 47 Although chronic arthritis caused by CHIKV shares similarities with rheumatoid arthritis in terms of clinical manifestations and disease course, the two differ in pathogenesis and prognosis ( Fig. 3 ). 48 CHIKV infection can induce lasting epigenetic changes in mesenchymal stromal cells, altering their function and promoting a persistent proinflammatory state within the joint, even in the absence of replicating virus. 49 A marked decrease in regulatory T cells (Tregs) is closely linked to the onset of CHIKV-induced arthritis. The reduction of Tregs probably results in unregulated activation of effector T cells and other immune cells within the joints, promoting chronic inflammation. 50 Recent studies have confirmed that CHIKV can infect human articular chondrocytes, this infection leads to cell death and the production of inflammatory mediators, contributing directly to cartilage damage and the development of arthritis. 51 CHIKV infection can also impair the function of osteogenic cells. Bone marrow-derived mesenchymal stem cells and osteogenic cells are susceptible to CHIKV infection. This infection leads to reduced expression of RUNX2, a key regulator of bone cell differentiation. Consequently, the functional abilities of osteogenic cells are impaired, as seen in lower alkaline phosphatase production and diminished matrix mineralization. 52 The neurological complications of CHIKV, though less common, can be severe. The virus can penetrate the blood-brain barrier and infect central nervous system cells. Experimental model research indicates that CHIKV infection triggers apoptosis in microglia, the brain's resident immune cells. The loss of these crucial cells can disrupt central nervous system homeostasis and trigger neuroinflammation. This pathophysio-logical process may explain some of the neurological symptoms, such as encephalitis and myelopathy. 53 The severity of this neurotoxicity appears to be dependent on both the age of the host and the specific strain of the virus, with younger individuals and certain viral genotypes being associated with more severe neurological outcomes. 54 In rare but devastating cases, CHIKV has been linked to severe outcomes such as acute lower limb flaccid paralysis. 55
## 5. Clinical manifestations
The acute phase of CHIKV infection usually presents with a sudden high fever exceeding 39°C, accompanied by symptoms such as arthralgia, back pain, headache, nausea, vomiting, arthritis, rash, and conjunctivitis. In a 2011 rural Bangladesh outbreak, the attack rate was 29%, with 76% of confirmed cases experiencing rash alongside fever and joint pain. 56 A study of patients in the Caribbean region of the Republic of Colombia identified the primary signs and symptoms as lower limb arthralgia (96%), fever (91%), upper limb arthralgia (85%), rash (64%), and headache (57%). 57 Atypical presentations can also occur, especially in children. In the Republic of Suriname, children with CHIKV infection presented with symptoms such as acute-onset fever, painful knees, rash, hypotension, tachycardia, and meningism. 58 Ocular complications, including uveitis, have been reported, which may present at the time of systemic manifestations or as a delayed presentation. 59 Although most CHIKV infections are self-limiting, some patients may develop complications. A study conducted in Martinique and Guadeloupe during the 2013-2014 outbreak reported severe cases, including Guillain-Barré syndrome, encephalitis, and severe sepsis, among patients in the intensive care unit. 60 Chronic symptoms, such as persistent or relapsing-remitting polyarthralgias, polyarthritis, and myalgias, can endure for months to years, significantly impacting patients' quality of life. 61 A systematic review and meta-analysis determined that the combined prevalence of post-chikungunya chronic inflammatory rheumatism was 40.22% among 5,702 patients. 62 The intrapartum vertical transmission of CHIKV from an infected mother to her newborn poses a significant risk. Follow-up studies of children infected during the perinatal period have revealed long-term cognitive and behavioral impairments, indicating that early-life infection can have lasting consequences on neurological function. 63 Maternal infection with CHIKV and other arboviruses, such as the Zika virus, has been linked to a heightened risk of miscarriage, underscoring the potential dangers during pregnancy. 64 However, studies in animal models have shown that maternal immunity conferred through breast milk can enhance the resistance of offspring to subsequent CHIKV infection, underscoring the immunological benefits of breastfeeding in endemic areas. 65 Furthermore, children are not immune to the chronic consequences of the disease. Recent studies have demonstrated that children contracting CHIKV can also experience chronic musculoskeletal symptoms lasting months or years, challenging the previous belief that it was mainly an adult issue. This chronicization of the disease in pediatric populations represents a major, and perhaps under-recognized, public health issue. 66
## 6. Diagnostics
Differentiating CHIKV infection from other febrile illnesses can be challenging due to the overlapping clinical symptoms. Diseases like dengue fever, Zika virus infection, and rheumatoid arthritis can exhibit similar symptoms, including fever, rash, and joint pain. 67 During the initial week of illness, when viremia is present, reverse transcription-polymerase chain reaction demonstrates high sensitivity and specificity. 68 Isothermal methods like reverse transcription loopmediated isothermal amplification have been created for CHIKV detection. These methods enable quick and sensitive viral RNA detec-Fig. 3. Similarities and differences between chikungunya virus-induced chronic arthritis and rheumatoid arthritis. tion without the need for advanced thermal cycling equipment, making them ideal for low-resource environments. [69][70] Furthermore, research has validated the use of alternative sample types for diagnosis. Detecting CHIKV RNA in saliva and urine provides a less invasive alternative to blood draws, beneficial for large-scale screening and pediatric populations. 71 An evaluation of the VIDAS® automated immunoassay system for detecting anti-CHIKV IgM and IgG antibodies confirmed its utility for routine diagnostics. 72 The PRNT is another serological method used to detect neutralizing antibodies, crucial for evaluating immunity. 43
## 7. Prevention and therapy
## 7.1. Prevention strategies
Novel biotechnological approaches are also being explored to combat CHIKV at the vector level. One innovative strategy involves the use of engineered ribozymes -catalytic RNA molecules -designed to specifically target and cleave the CHIKV RNA genome. When expressed in mosquitoes, these ribozymes can inhibit viral replication within the vector, thereby reducing or blocking its transmission to humans. This represents a form of genetic vector control aimed at making mosquito populations resistant to the virus. 73 Transgenic mosquitoes have been genetically modified to produce anti-CHIKV single-domain antibodies in their midgut. This strategy aims to create a transmission-blocking mosquito population that is incapable of sustaining and transmitting the virus. 74 Atovaquone, an antimalarial drug, has been shown to inhibit CHIKV transmission in Aedes aegypti , indicating its potential for repurposing as a transmission-blocking agent in humans. 75 The polyclonal antibody has been shown to be effective in preventing CHIKV infection in animal models when administered prophylactically. 76 It has been demonstrated that the interaction of anti-CHIKV IgG antibodies with the Fc-gamma receptor IIIa on immune cells, such as natural killer cells, significantly enhances their protective efficacy. 77 Furthermore, understanding the targets of protective antibodies is crucial. The B domain of the E2 envelope protein has been identified as a target for cross-neutralizing antibodies that can recognize multiple alphaviruses. 78 Socioeconomic factors significantly influence the risk of arboviral infections. Given the increasing severity of deltamethrin resistance in Aedes albopictus , it is of great importance to control vectors through comprehensive measures. 79 Studies have shown that factors such as housing quality, access to piped water, and waste management practices can influence the abundance of mosquito breeding sites and, consequently, the risk of infection to viruses like CHIKV and Dengue. 80 After the outbreak in Guangdong Province, China in 2025, the health authorities implemented a series of response measures: (1) CHIKV PCR testing was universally conducted among high-risk populations and febrile patients to improve the detection rate of patients; (2) Confirmed cases were isolated in designated hospitals equipped with mosquito-proof facilities, including insecticide-treated window screens and bed nets; (3) Comprehensive vector management measures were implemented, encompassing targeted insecticide spraying, elimination of breeding sites, application of biological control, and real-time vector monitoring. These measures led to the timely containment of the outbreak. 81 These findings highlight that disease control cannot rely solely on biomedical interventions but must also incorporate improvements in social and environmental conditions.
## 7.2. Vaccine development
Different vaccine platforms have been investigated to prevent CHIKV infection, such as inactivation, live-attenuated strains, virus-like particles (VLPs), viral vectors, and mRNA technologies. In the United States, two chikungunya vaccines have received licensing and approval: IXCHIQ (VLA1553), a live-attenuated vaccine 82 , and VIMKUNYA (PXVX0317), a VLP vaccine. 83 A crucial phase 2 trial of the VIMKUNYA vaccine confirmed its safety and ability to provoke an immune response in adults without prior CHIKV exposure, showing a lasting serum neutralizing antibody response for up to two years after vaccination. 84 This trial also highlighted the vaccine's favorable safety profile, with no serious adverse events reported, which is consistent with findings from other studies on VLP-based vaccines. The VIMKUNYA vaccine demonstrated consistent immunogenicity across various age groups in a phase 3 trial with participants aged 12-64 years. The trial demonstrated a strong and swift immune response, with a seroresponse rate of 97.8% in the vaccine group, significantly higher than the 1.2% observed in the placebo group. 85 Another phase 3 trial has confirmed the safety and ef- Abbreviations: CI, confidence interval; SD, standard deviation; VLP, virus-like particle vaccine. a ID of ClinicalTrials.gov, data available: https://clinicaltrials.gov/ .
ficacy of the VIMKUNYA vaccine in adults aged 65 and older, a group at higher risk for severe CHIKF complications. The trial showed that the vaccine was well-tolerated, offering significant protection within two weeks after administration, with no serious adverse events or deaths linked to the vaccine. 86 BBV87, a live-attenuated vaccine, has demonstrated safety and efficacy in protecting non-human primates from CHIKV infection, endorsing its continued clinical development. 87 For specific populations, such as travelers, inactivated vaccines are also being developed, which may offer a different safety profile compared to live-attenuated versions. 88 A significant advancement has been the development of a vaccine candidate based on E2-B nanoparticles. These nanoparticles display a portion of the CHIKV E2 glycoprotein and have been shown to induce potent and broadly neutralizing antibody responses in preclinical models. 89 A novel vaccine candidate using a non-replicating, attenuated CHIKV RNA packaged within liposomes has been developed. When administered to mice, this liposome-formulated RNA vaccine induced potent protective immunity against a lethal CHIKV challenge. 90 Other candidate vaccines with available clinical trial results are shown in Table 1 .
## 7.3. Investigational antiviral agents
Numerous investigational antiviral agents have been studied for their potential against CHIKV. As discussed, the nsP2 protease is a prime target. Synthetic small molecules like acrylamide derivatives have been developed to specifically block its function. 92 The nsP3 protein, with its role in VRC assembly, has also been targeted. The antiviral effectiveness of Ag/NiO and Ag 2 O/NiO/ZnO mixed metal oxide nanocomposites against CHIKV was assessed using in vitro tests., by disrupting function of nsP3 and inhibiting viral replication, this presents high engineering potential for broad-spectrum antiviral activity. 93 Beyond targeting individual viral proteins, researchers are exploring compounds with broader mechanisms of action. Research indicates that L-cysteine can inhibit CHIKV infection in vitro , implying that altering cellular metabolic or redox states may serve as an effective antiviral strategy. 94 The search for novel chemical scaffolds has also been fruitful. Benzothiazole (BTA) derivatives can reduce viral replication by 98%. Molecular docking indicated that there are strong binding affinities to CHIKV's non-structural proteins and envelope glycoproteins. Infrared spectroscopy confirmed it's interaction with the glycoprotein complex and lipids, highlight BTA derivatives as promising CHIKV inhibitors. 95 Mechanistic tests showed that sulfur-and selenium-containing benzotriazoles disrupted viral adsorption, had virucidal properties, and blocked several stages of the replication cycle, emphasize these compounds as potential dual-action antiviral agents with wide-ranging effects against Zika virus (ZIKV) and CHIKV. 96 Thiocyanate compounds 97 and chlorinated biscoumarin derivatives 98 have been identified as potent inhibitors of CHIKV replication in cell culture models. Additional preclinical studies are needed to assess their therapeutic potential.
Drug repurposing offers a faster path to clinical use. For instance, the HIV reverse transcriptase inhibitor Efavirenz has been shown to possess anti-CHIKV activity. 99 Similarly, certain drugs approved for treating hepatitis C virus have demonstrated inhibitory effects against CHIKV, suggesting they could be repurposed for this new indication. 100 Etravirine, an approved non-nucleoside reverse transcriptase inhibitor for HIV treatment, exhibits inhibitory effects on West Nile Virus and CHIKV, indicating its potential as a broad-spectrum antiviral agent. 101 Nifuroxazide, an oral nitrofuran antibiotic, exhibits broad-spectrum antiviral activity by inhibiting the replication of various insect-borne viruses, including CHIKV. 102 Furthermore, derivatives of artesunate, an antimalarial drug, such as Anthrone-Spirolactam, have been synthesized and found to have potent anti-CHIKV activity. 103 RNA interference offers a highly specific way to silence viral genes. A novel system using zeolitic imidazolate framework-coated carbon nanodots (ZIF-C) has been developed to effectively deliver siRNA molecules into infected cells. ZIF-C confers maximal protection to the loaded nucleic acids against nuclease-mediated degradation. By targeting the E2 and nsP1 genes, viral replication and infectivity are reduced. 104 Peptidebased inhibitors represent another promising class of biologics. A hybrid peptide named GA-Hecate was designed and shown to have potent inhibitory activity against CHIKV. 105 Similarly, certain dipeptides have been identified that exhibit antiviral activity against both CHIKV and Zika virus, suggesting they may target a conserved process in flavivirus and alphavirus replication. 106
## 8. Conclusion
Due to the long-term and severe consequences brought about by the CHIKV infection, CHIKF has clearly become a persistent and highly threatening global health issue. There is an urgent need for the development of novel molecular and antigen detection assays with high sensitivity, rapid recognition, accessibility. Future fundamental research on CHIKV should prioritize elucidating its genetic variation, host-pathogen interactions, and mechanisms of vector adaptation. In the development of antiviral agents, significant challenges include the virus's capacity to mutate and develop resistance to antiviral therapies. Therefore, establishing efficient antiviral screening platforms for drug discovery is imperative. Promising targets for breakthroughs include the inhibition of the capping, macrodomain, and capsid protease functions of nsPs. Other therapeutic strategies under investigation involve targeting host cell pathways that either facilitate or inhibit viral replication, such as fatty acid synthesis and cholesterol trafficking pathways; dysregulating endosome acidification to inhibit viral entry; inhibiting nucleobase biosynthesis; or using immunomodulatory therapies to stimulate an interferon response. Given the absence of specific treatments for CHIKV, the developing protective monoclonal antibody therapy may temporarily address the disease's deterioration.
In vaccine development, it is imperative to ensure long-term protection and safety, particularly among high-risk populations such as children, the elderly, and individuals with comorbidities. A comprehensive understanding of humoral immunity and a detailed examination of antibody-dependent enhancement mechanisms are critical for the design of effective and safe vaccines. An optimal vaccine should possess thermostability and be facile in production, transportation, and storage, especially in low-and middle-income countries that are disproportionately affected by CHIKV.
Collectively, the recent advances in virology, immunology, clinical medicine, and drug development provide a solid and expanding foundation of knowledge and tools to mitigate the impact of CHIKV on human populations worldwide. The advancement of this field necessitates interdisciplinary research.
## CRediT authorship contribution statement
## Informed consent
Not applicable.
## Organ donation
Not applicable.
## Ethical statement
Not applicable.
## References
1. Acosta-Reyes, Tuesca, Navarro-Lechuga (2025) "Chronicity and quality of life in Chikungunya virus infection: a cross-sectional study in Barranquilla, Colombia" *J Microbiol Immunol Infect*
2. Feng, Chang, Yang (2025) "From dengue to chikungunya: Guangdong as a sentinel for arboviral threats in East Asia" *Biosci Trends*
3. (2025) "Disease Outbreak News; Chikungunya virus disease-Global situation. World Health Organization"
4. De Souza, Ndiaye (2020) "Serosurvey of chikungunya virus in old world fruit bats" *Emerg Infect Dis*
5. Li, Peng, Yuan (2023) "A new cluster of chikungunya virus West Africa genotype isolated from Aedes albopictus in China" *J Infect*
6. Allen, Santos, Paul (2024) "Results of a nationally representative seroprevalence survey of chikungunya virus in Bangladesh" *J Infect Dis*
7. Anjos, Portilho, Lc (2023) "Dynamics of chikungunya virus transmission in the first year after its introduction in Brazil: a cohort study in an urban community" *PLoS Negl Trop Dis*
8. Giovanetti, Vazquez, Lima (2023) "Rapid epidemic expansion of chikungunya virus east/central/south African lineage" *Paraguay. Emerg Infect Dis*
9. Parker, Haileselassie, Hailemariam (2025) "High seroprevalence of antibodies to Dengue, Chikungunya, and Zika viruses in Dire Dawa , Ethiopia: a cross-sectional survey in 2024" *PLoS Negl Trop Dis*
10. Visser, Wang, Abbo (2025) "Effect of chikungunya, Mayaro and Una virus coinfection on vector competence of Aedes aegypti mosquitoes" *One Health*
11. Wesselmann, Baronti, Nougairède (2025) "Development and evaluation of a duplex RT-qPCR assay for the detection and identification of Mayaro and chikungunya viruses" *J Clin Microbiol*
12. Tinto, Bicaba, Kagoné (2024) "Co-circulation of two Alphaviruses in Burkina Faso: Chikungunya and O'nyong nyong viruses" *PLoS Negl Trop Dis*
13. Wesselmann, Luciani, Thirion (2024) "Analytical and clinical evaluation of a duplex RT-qPCR assay for the detection and identification of o'nyong-nyong and chikungunya virus" *Emerg Microbes Infect*
14. Azman, Chan, Chua (2024) "A change in circulating chikungunya virus variant impacts Aedes aegypti vector competence and spatiotemporal distribution of disease in Malaysia" *PLoS Negl Trop Dis*
15. Mehta, Chaudhary, Sunil "Oxidative stress governs mosquito innate immune signalling to reduce chikungunya virus infection in Aedes -derived cells" *J Gen Virol*
16. Bernal-Valle, De, Mares-Guia et al. (2025) "Natural exposure to Chikungunya virus in golden-headed lion tamarin ( Leontopithecus chrysomelas , Kuhl, 1820) from non-protected areas in southern Bahia , Brazil: Implications and significance" *PLoS Negl Trop Dis*
17. Aranda-Coello, Machain-Williams, Weber (2025) "Serologic surveillance for orthoflaviviruses and chikungunya virus in bats and opossums in Chiapas" *Viruses*
18. Micheleto, Melo, Veloso (2025) "Risk factors for mortality in patients with chikungunya: a systematic review and meta-analysis" *Trop Med Int Health*
19. De Jesús, Ortiz-Mesina, Caballero-Hoyos et al. (2019) "Obstetric complications of dengue and chikungunya in the pregnant patient: case-control study" *Rev Med Inst Mex Seguro Soc*
20. Delgado-Enciso, Michel, Melnikov (2018) "Smoking and female sex as key risk factors associated with severe arthralgia in acute and chronic phases of Chikungunya virus infection" *Exp Ther Med*
21. Martins, De Bruycker-Nogueira, Rodrigues (2022) "Chikungunya virus shedding in Semen : a case series" *Viruses*
22. Manzoor, Javed, Ejaz (2022) "The global emergence of Chikungunya infection: an integrated view" *Rev Med Virol*
23. Pietilä, Hellström, Ahola (2017) "Alphavirus polymerase and RNA replication" *Virus Res*
24. Jones, Hons, Rabah (2023) "Structural basis and dynamics of Chikungunya alphavirus RNA capping by nsP1 capping pores" *Proc Natl Acad Sci USA*
25. Chamberlain, Dowall, Smith (2025) "Attenuation of chikungunya virus by a single amino acid substitution in the nsP1 component of a non-structural polyprotein" *Viruses*
26. Roberts, Stonehouse, Harris (2025) "The chikungunya virus nsP3 macro domain inhibits activation of the NF-𝜅B pathway" *Viruses*
27. Yin, Sobolik, May (2025) "Mutations in chikungunya virus nsP4 decrease viral fitness and sensitivity to the broad-spectrum antiviral 4'-Fluorouridine" *PLoS Pathog*
28. Ghoshal, Tse, Hossain (2025) "A covalent chemical probe for Chikungunya nsP2 cysteine protease with antialphaviral activity and proteome-wide selectivity" *Sci Rep*
29. Yao, Ramachandran, Huang (2024) "Interaction of chikungunya virus glycoproteins with macrophage factors controls virion production" *EMBO J*
30. Chatterjee, Subudhi, Chattopadhyay (2023) "A hidden gem Catenin-𝛼-1 is essential for Chikungunya virus infection" *Microbiol Spectr*
31. Echavarria-Consuegra, Kumar, Van Der Laan (2023) "Mitochondrial protein BNIP3 regulates Chikungunya virus replication in the early stages of infection" *PLoS Negl Trop Dis*
32. Feng, Bouma, Hu (2023) "Colocalization of chikungunya virus with its receptor MXRA8 during cell attachment, internalization, and membrane fusion" *J Virol*
33. Zhang, Earnest, Kim (2019) "Expression of the Mxra8 receptor promotes alphavirus infection and pathogenesis in mice and Drosophila" *Cell Rep*
34. Zhang, Kim, Fox (2018) "Mxra8 is a receptor for multiple arthritogenic alphaviruses" *Nature*
35. Ballista, Hoover, Noble (2024) "Chikungunya virus release is reduced by TIM-1 receptors through binding of envelope phosphatidylserine" *J Virol*
36. Pradeep, Sivakumar, Sreekumar (2023) "Host factor nucleophosmin 1 (NPM1/B23) exerts antiviral effects against chikungunya virus by its interaction with viral nonstructural protein 3" *Microbiol Spectr*
37. Lau, Chua, Chan (2023) "Replication and innate immune responses of two chikungunya virus genotypes in human monocyte-derived macrophages" *J Gen Virol*
38. Sam, Kümmerer, Chan (2015) "Updates on chikungunya epidemiology, clinical disease, and diagnostics. Vector Borne Zoonotic Dis"
39. Frumence, Piorkowski, Traversier (2024) "Genomic insights into the re-emergence of chikungunya virus on Réunion Island" *Eurosurveillance*
40. Romano, Pavesi, Ferrari (2025) "Chikungunya cases with vector-adaptive mutations detected in Italy imported from Madagascar" *J Travel Med*
41. Liu, Shen, Gu (2025) "Chikungunya virus in Europe: a retrospective epidemiology study from 2007 to 2023" *PLoS Negl Trop Dis*
42. Gumpangseth, Villarroel, Diack (2025) "IFITMs exhibit antiviral activity against Chikungunya and Zika virus infection via the alteration of TLRs and RLRs signaling pathways" *Sci Rep*
43. Yoon, Alera, Lago (2015) "High rate of subclinical chikungunya virus infection and association of neutralizing antibody with protection in a prospective cohort in the Philippines" *PLoS Negl Trop Dis*
44. De Oliveira Souza, Júnior, Casa (2024) "Unraveling the complex interplay: immunopathology and immune evasion strategies of alphaviruses with emphasis on neurological implications" *Front Cell Infect Microbiol*
45. Ware, Parks, Da Silva (2024) "Chikungunya virus infection disrupts MHC-I antigen presentation via nonstructural protein 2" *PLoS Pathog*
46. Yin, Davenport, Wan (2023) "Chikungunya virus cell-to-cell transmission is mediated by intercellular extensions in vitro and in vivo" *Nat Microbiol*
47. Lum, Chan, Teo (2024) "Crosstalk between CD64 + MHCII + macrophages and CD4 + T cells drives joint pathology during chikungunya" *EMBO Mol Med*
48. Kumar, Ahmed, Parray (2021) "Chikungunya and arthritis: an overview" *Travel Med Infect Dis*
49. Amaral, Bingham, Taylor (2023) "Pathogenesis of chronic chikungunya arthritis: Resemblances and links with rheumatoid arthritis" *Travel Med Infect Dis*
50. Dobbs, Tritsch, Encinales (2022) "Regulatory T-cells and GARP expression are decreased in exercise-associated chikungunya viral arthritis flares" *Front Immunol*
51. Legros, Belarbi (2025) "Use of recombinant chikungunya virus expressing nanoluciferase to identify chondrocytes as target cells in an immunocompetent mouse model" *J Infect Dis*
52. Roy, Shi, Duan (2020) "Chikungunya virus infection impairs the function of osteogenic cells. mSphere"
53. Kumar, Santhoshkumar, Venkataswamy (2024) "Chikungunya virus infection in human microglial C20 cells induces mitochondria-mediated apoptosis" *Front Cell Infect Microbiol*
54. Anderson, Knight, Heise (2023) "Effect of viral strain and host age on clinical disease and viral replication in immunocompetent mouse models of chikungunya encephalomyelitis" *Viruses*
55. Matungala-Pafubel, Bulabula-Penge, Matondo-Kuamfumu (2024) "Lower limb paralysis associated with chikungunya in Kinshasa, the democratic republic of the Congo: survey report" *Pathogens*
56. Khatun, Chakraborty, Rahman (2011) "An Outbreak of Chikungunya in Rural Bangladesh" *PLoS Negl Trop Dis*
57. Mattar, Miranda, Pinzon (2015) "Outbreak of Chikungunya virus in the north Caribbean area of Colombia: clinical presentation and phylogenetic analysis" *J Infect Dev Ctries*
58. Donald-Ottevanger, Gravenberch-Ramnandanlall, Zijlmans
59. (2015) *Ned Tijdschr Geneeskd*
60. Mahendradas, Patil, Kawali (2024) "Systemic and ophthalmic manifestations of chikungunya fever" *Ocul Immunol Inflamm*
61. Crosby, Perreau, Madeux (2016) "Severe manifestations of chikungunya virus in critically ill patients during the 2013-2014 Caribbean outbreak" *Int J Infect Dis*
62. Arroyo-Ávila, Vilá (2015) "Rheumatic manifestations in patients with chikungunya infection" *P R Health Sci J*
63. Rodríguez-Morales, Cardona-Ospina, Urbano-Garzón (2016) "Prevalence of post-chikungunya infection chronic inflammatory arthritis: a systematic review and meta-analysis" *Arthritis Care Res*
64. Sarton, Carbonnier, Robin (2025) "Perinatal mother-to-child chikungunya virus infection: screening of cognitive and learning difficulties in a followup study of the chimere cohort on Reunion Island" *Viruses*
65. De Carvalho, Cruz, Quaresma (2025) "Impact of zika and chikungunya viruses on spontaneous abortions: insights from a reference maternity hospital. Microorganisms"
66. De, Souza, De Jesus et al. (2023) "Breastfeeding by chikungunya virus-infected dams confers resistance to challenge in the offspring" *Transl Res*
67. De, Pereira, Brasil et al. (2025) "Chikungunya in a pediatric cohort: Asymptomatic infection, seroconversion, and chronicity rates" *PLoS Negl Trop Dis*
68. Miner, Aw-Yeang, Fox (2015) "Chikungunya viral arthritis in the United States: a mimic of seronegative rheumatoid arthritis" *Arthritis Rheumatol*
69. Sajith, Iyengar, Varamballi (2025) "Diagnostic utility of real-time RT-PCR for chikungunya virus detection in the acute phase of infection: a retrospective study" *Ann Med*
70. Wu, Liu, Chang (2024) "Rapid and sensitive detection of chikungunya virus using one-tube, reverse transcription, semi-nested multi-enzyme isothermal rapid amplification, and lateral flow dipstick assays" *J Clin Microbiol*
71. Da Silva, De Magalhães, Matthews (2024) "Development and field validation of a reverse transcription loop-mediated isothermal amplification assay (RT-LAMP) for the rapid detection of chikungunya virus in patient and mosquito samples" *Clin Microbiol Infect*
72. Portilho, Anjos (2024) "Detection of chikungunya virus RNA in oral fluid and urine: an alternative approach to diagnosis Viruses"
73. Pereira, Manuli, Coulon (2023) "Performance evaluation of VIDAS® diagnostic assays detecting anti-chikungunya virus IgM and IgG antibodies: an international study" *Diagnostics*
74. Mishra, Balaraman, Jr (2023) "Maxizyme-mediated suppression of chikungunya virus replication and transmission in transgenic Aedes aegypti mosquitoes" *Front Microbiol*
75. Webb, Compton, Rai (2023) "Expression of anti-chikungunya singledomain antibodies in transgenic Aedes aegypti reduces vector competence for chikungunya virus and Mayaro virus" *Front Microbiol*
76. Wang, Sanon, Khoiriyah (2023) "Tarsal exposure to atovaquone inhibits chikungunya virus transmission by Aedes aegypti mosquitoes, but not the transmission of Zika virus" *Antiviral Res*
77. Barker, Han, Wang (2023) "Equine polyclonal antibodies prevent acute chikungunya virus infection in mice" *Viruses*
78. Fox, Roy, Gunn (2023) "Enhancing the therapeutic activity of hyperimmune IgG against chikungunya virus using Fc 𝛾RIIIa affinity chromatography" *Front Immunol*
79. Powers, Lyski, Weber (2023) "Infection with chikungunya virus confers heterotypic cross-neutralizing antibodies and memory B-cells against other arthritogenic alphaviruses predominantly through the B domain of the E2 glycoprotein" *PLoS Negl Trop Dis*
80. Mohapatra, Bhattacharjee, Desai (2024) "Global health concern on the rising dengue and chikungunya cases in the American regions: Countermeasures and preparedness" *Health Sci Rep*
81. Tariq, Khan, Mutuku (2024) "Understanding the factors contributing to dengue virus and chikungunya virus seropositivity and seroconversion among children in Kenya" *PLoS Negl Trop Dis*
82. Tee, Mu, Xia (2025) "Explosive chikungunya virus outbreak in China" *Int J Infect Dis*
83. Maurer, Buerger, Larcher-Senn (2025) "Comprehensive assessment of reactogenicity and safety of the live-attenuated chikungunya vaccine (IXCHIQ®)"
84. Raju, Adams, Earnest (2023) "A chikungunya virus-like particle vaccine induces broadly neutralizing and protective antibodies against alphaviruses in humans" *Sci Transl Med*
85. Bennett, Mccarty, Ramanathan (2022) "Safety and immunogenicity of PXVX0317, an aluminium hydroxide-adjuvanted chikungunya virus-like particle vaccine: a randomised, double-blind, parallel-group, phase 2 trial" *Lancet Infect Dis*
86. Richardson, Anderson, Mendy (2025) "Chikungunya virus virus-like particle vaccine safety and immunogenicity in adolescents and adults in the USA: a phase 3, randomised, double-blind, placebo-controlled trial" *Lancet*
87. Tindale, Richardson, Anderson (2025) "Chikungunya virus virus-like particle vaccine safety and immunogenicity in adults older than 65 years: a phase 3, randomised, double-blind, placebo-controlled trial" *Lancet*
88. Kempster, Ferguson, Ham (2025) "Inactivated viral vaccine BBV87 protects against chikungunya virus challenge in a non-human primate model" *Viruses*
89. Freedman (2025) "A new non-live chikungunya vaccine for travellers" *J Travel Med*
90. Tong, Hernandez, Basore (2024) "Chikungunya virus E2 B domain nanoparticle immunogen elicits homotypic neutralizing antibody in mice" *Vaccine*
91. Rao, Abeyratne, Freitas (2023) "A booster regime of liposome-delivered live-attenuated CHIKV vaccine RNA genome protects against chikungunya virus disease in mice" *Vaccine*
92. De Souza, Choudhary, Vilela (2023) "Design, synthesis, antiviral evaluation, and in silico studies of acrylamides targeting nsP2 from Chikungunya virus" *Eur J Med Chem*
93. Bhatia, Singh, Rani (2023) "Cellular uptake of metal oxide-based nanocomposites and targeting of chikungunya virus replication protein nsP3" *J Trace Elem Med Biol*
94. Kumar, Shrinet, Sunil (2023) "Chikungunya virus infection in Aedes aegypti is modulated by L-cysteine, taurine, hypotaurine and glutathione metabolism" *PLoS Negl Trop Dis*
95. Feferbaum-Leite, Cassani, Ruiz (2025) "Benzothiazole derivatives as inhibitors of chikungunya virus replicative cycle" *Future Med Chem*
96. Gomes, Cc Cirne-Santos (2025) "Sulfur/selenium-functionalized benzotriazoles as multifunctional antivirals targeting Zika & Chikungunya" *Future Med Chem*
97. Alagarasu, Dhote, Bagad (2025) "Effectiveness of 3-amino-2-thiocyanato-𝛼, 𝛽-unsaturated carbonyl compounds against chikungunya virus" *Future Med Chem*
98. Orji, Loeanurit, Pham (2025) "Chlorinated biscoumarins inhibit chikungunya virus replication in cell-based and animal models" *Emerg Microbes Infect*
99. Nehul, Rani, Walia (2025) "Repurposing efavirenz, the HIV antiretroviral drug for chikungunya virus infection" *ACS Infect Dis*
100. Kalam, Ali, Balasubramaniam (2025) "Exploring the potential of direct-acting antivirals against Chikungunya virus through structure-based drug repositioning and molecular dynamic simulations" *Comput Biol Med*
101. Zheng, He, Xia (2024) "Etravirine prevents west Nile virus and chikungunya virus infection both in vitro and in vivo by inhibiting viral replication" *Pharmaceutics*
102. Liu, Xu, Xia (2024) "Nifuroxazide prevents chikungunya virus infection both in vitro and in vivo via suppressing viral replication" *Viruses*
103. Darole, Bagad, Gonnade (2023) "Synthesis of novel rhodamine type Anthrone Spiro-lactam (ASL) analogues and evaluation of antiviral activity against dengue and chikungunya viruses" *Eur J Med Chem*
104. Tagore, Alagarasu, Patil (2022) "Targeted in vitro gene silencing of E2 and nsP1 genes of chikungunya virus by biocompatible zeolitic imidazolate framework" *Front Bioeng Biotechnol*
105. Ayusso, Da Silva, Sanches et al. (2023) "The synthetic peptide GA-Hecate and its analogs inhibit multiple steps of the chikungunya virus infection cycle in vitro" *Pharmaceuticals*
106. Ayusso, Lima, Da et al. (2023) "The dimeric peptide (KKYRYHLKPF) 2 K shows broad-spectrum antiviral activity by inhibiting different steps of chikungunya and zika virus infection" *Viruses*
107. Wang, Sun, Tang (2025) "Re-emergence of chikungunya virus in China by 2025: What we know and what to do" *PLoS Pathog*
108. Freppel, Silva, Stapleford (2024) "Pathogenicity and virulence of chikungunya virus" *Virulence* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12532353&blobtype=pdf | # Dysbiosis of the oropharyngeal microbiota in COVID-19: distinct profiles in patients with severe respiratory symptoms
Sunitha Kumari, Varsha Potdar, Manohar Shinde, Deepti Parashar, Kalichamy Alagarasu, Sarah Cherian, Mallika Lavania
## Abstract
Background: COVID-19 has been strongly associated with alterations in the oropharyngeal microbiota, yet the microbial features linked to disease severity remain unclear.Objective: This study aimed to elucidate the microbial signatures associated with COVID-19 disease severity. Design: 16S rRNA gene sequencing was employed to profile the oropharyngeal microbiota of patients with varying degrees of COVID-19 severity. Results: A significant reduction in alpha diversity suggests a major microbial dysbiosis in critically ill patients compared to less severe cases and healthy individuals, whereas beta diversity analysis revealed a broadly conserved community structure across different groups. Comparative analysis showed significant depletion of the phylum Fusobacteriota and enrichment of bacterial families, including Corynebacteriaceae, Methylobacteriaceae, Acetobacteraceae, Bradyrhizobiaceae, Lactobacillaceae, Staphylococcaceae, Propionibacteriaceae, and Moraxellaceae. Rothia mucilaginosa was notably enriched in patients with severe respiratory symptoms, and many of the enriched taxa are known opportunistic pathogens associated with respiratory infections.
Conclusion:The marked dysbiosis and enrichment of opportunistic pathogens in the oropharyngeal microbiota of critically ill patients indicate their possible role in respiratory complications. The identified microbial patterns highlight the potential of microbiome profiling as a tool for disease prognosis and guide further research into the role of microbes in COVID-19 pathogenesis and implications for treatment protocols.
## Introduction
SARS-CoV-2 has caused an unprecedented global health crisis, with over 778.45 million cases and 7.1 million deaths reported as of 24 August 2025 [1]. COVID-19 presents with symptoms ranging from mild illness to severe acute respiratory distress syndrome (ARDS) [2], with disease severity shaped by both host immune responses and viral factors [3]. The oropharyngeal microbiota plays a vital role in respiratory health by supporting mucosal immunity and acting as a first line of defence against pathogens [4]. In COVID-19, increasing evidence indicates that SARS-CoV-2 infection can disturb this microbial equilibrium, referred to as dysbiosis, which may be linked to increased disease severity [5]. Individuals with severe respiratory symptoms frequently show significant alterations in their upper airway microbiota, which could impact viral replication, susceptibility to secondary infections and overall clinical outcomes [6]. Investigating these microbial changes is crucial for identifying potential biomarkers and therapeutic strategies, especially as COVID-19, which is now endemic, continues to cause periodic outbreaks.
Previous studies have shown that the oral, oropharyngeal and lung microbiota share common bacterial species, with the oral microbiome exhibiting greater diversity than the lung microbiome [7]. The respiratory viral infections such as influenza have been linked to oral microbial imbalances, including
## Materials and methods
## Study design and sample collection
To investigate the link between the oropharyngeal microbiome profiles and the severity of SARS-CoV-2 infection, a total of 87 participants were enrolled and categorised into four distinct groups. Among them, 66 individuals were confirmed to be positive for SARS-CoV-2 through RT-PCR testing. Based on clinical evaluation, these infected individuals were grouped into 25 asymptomatic (patients with no symptoms), 19 with mild symptoms (patients who did not require hospitalization) and 22 critically ill patients (patients requiring hospitalization and ICU support). In addition, 21 RT-PCR-negative individuals who showed no signs of respiratory illness were included as healthy controls. The asymptomatic cases included in our study were identified as close contacts of confirmed COVID-19 patients who tested positive for SARS-CoV-2 by RT-PCR but did not develop clinical symptoms during the observation period. The selection of healthy controls was based on individuals with no history of COVID-19 exposure or symptoms and who tested negative for SARS-CoV-2 by RT-PCR. The samples were collected over a period from 1 July 2022 to 1 May 2023 (Table S1). All participants were residents of Pune, Maharashtra, India, and the hospitalised cases were admitted to tertiary care hospitals in Pune, which were designated for COVID-19 complicated cases. The comorbidities and symptoms of the patients were recorded and documented for those in both critically ill and mild conditions.
## Oropharyngeal swab collection and RT-PCR
The throat swab of COVID-19 patients and healthy controls were collected by trained laboratory personnel using disposable sterile cotton swabs. The swabbing procedure involved gently rotating the swab three to five times across the posterior pharyngeal wall, lateral sides and tonsillar crypts. Following collection, the swabs were immediately transferred into specialised preservation tubes designed for oral sample storage. Total RNA was subsequently extracted from all 87 swab samples using the Qiagen Viral RNA Kit in accordance with the manufacturer's guidelines. RT-PCR was performed to detect SARS-CoV-2 in freshly extracted RNA, which targeted the E gene and Orf1b, using assays as per the ICMR-NIV protocol [17]. All the experimental procedures were performed in accordance with the guidelines and regulations. Ethical approval for the study was granted by the institutional human ethics committee (Ethics reference ID: NIV/IEC/Oct/2021/D-1/20-3-16R). Written informed consent was obtained from all participants involved in the study.
## DNA extraction and 16S rRNA targeted metagenome sequencing
Total microbial DNA was extracted from oropharyngeal swabs obtained from participants using QIAamp DNA Microbiome Kit following the manufacturer's instructions. The 16S rRNA hypervariable regions were amplified with two separate primer pools targeting the V2-4-8 and V3-6,7-6 regions, using the Ion 16S™ Metagenomics Kit (Thermo Fisher Scientific, USA). Amplification was carried out separately for each primer pool in individual reaction tubes for each sample. These tubes were then combined for library preparation using Ion Plus Fragment Library Kit (Thermo Fisher Scientific, USA). The resulting libraries were purified with Agencourt® AMPure® XP Beads and quantified using the Agilent® 2000 Bioanalyzer® instrument. The amplified fragments were sequenced using the Ion Personal Genome Machine™ (PGM™) platform, and the data were analysed with the Ion 16S™ Metagenomic Kit analysis module in Ion Reporter™ software.
## 16S rRNA sequence data analysis
The paired raw sequences were analysed using the metagenomics workflow in Ion Reporter™ Software. This platform processed the raw reads by performing demultiplexing, denoising, quality filtering, alignment to a 16S rRNA gene reference database and clustering into operational taxonomic units (OTUs). Default primers were applied for curation, with a primer search of 15 base pairs and a maximum of 3 errors. A read length filter of 150 and a minimum alignment coverage of 90 were used to eliminate low-quality reads. Additionally, low-copy reads were excluded with an abundance cutoff of 10. The OTUs were taxonomically assigned using Megablast against both curated Greengenes (v13.5) and the Curated MicroSEQ™ ID 16S reference database (v2013.1). A genus cutoff of 97% and a species cutoff of 99% were applied for the taxonomic assignment. The alpha and beta diversity measures were calculated using Quantitative Insights into Microbial Ecology (QIIME) available in Ion Reporter™.
## Microbial diversity analyses
Alpha diversity was assessed using the Chao1 and Shannon indices, which were calculated through the QIIME pipeline [18]. Boxplots for both indices were constructed using the ggplot2 package [19] in R v4.4.3 [20]. Statistical comparisons were carried out using the Kruskal-Wallis test and pairwise Wilcoxon test, implemented via the ggsignif [21] and rstatix [22] packages in R v4.4.3 [20]. To evaluate microbial diversity across groups, beta diversity was estimated using the Bray-Curtis dissimilarity metric within QIIME [18]. Principal coordinates analysis (PCoA) was then applied to the distance matrices to generate twodimensional plots using ggplot2 [19] in R v4.4.3 [20]. The spread of the PCoA clusters was analysed using the betadisper function from the Vegan package [23] in R v4.4.3 [20]. The permutational analysis of variance (PERMANOVA) test, calculated using the function adonis in the Vegan package [23], was performed to determine whether there was a significant separation between different sample groups (statistical significance level p-value = 0.05), and the plot was generated using ggplot2 [19] in R v4.4.3 [20]. Additionally, PCoA was conducted based on the variables 'age' and 'sex' to explore their potential impact on the overall microbiome structure.
## Microbiome composition and differential analyses
To examine the microbiota composition at the phylum and family levels, bar plots with relative abundance percentages were generated using ggplot2 [19] in R v4.4.3 [20]. Differences among the critical, mild, asymptomatic and control groups were evaluated at both taxonomic levels using the Kruskal-Wallis test, followed by Dwass-Steel-Critchlow-Fligner pairwise comparisons conducted in Jamovi v2.6.13 [24]. Boxplots for significantly different phyla (p-value < 0.05) were generated using ggplot2 [19], and the statistical analysis was performed using the ggsignif [21] and rstatix [22] packages in R v4.4.3 [20]. To visualize similarities and differences in microbial composition at the family level across samples, a heatmap was created using the heatmap function from the clustVis webserver [25].
To quantify the effect size of different microbial populations in the oropharyngeal microbiome of COVID-19 patients across severity groups and healthy controls, a linear discriminant analysis effect size (LEfSe) [26] available in ImageGP2 [27] was used. LEfSe employs the Kruskal-Wallis non-parametric test and multiple pairwise Wilcoxon tests (both with a significance threshold of p-value < 0.05), followed by linear discriminant analysis (LDA) to identify microbial features (genera/species) that differ between groups. The cutoff value for the linear discriminant analysis was set as the LDA score (log 10) = 3. The genera with LDA scores above this cutoff were visualised using bar plots, and species with high LDA scores were represented using boxplots and both plots were generated with ggplot2 [19] in R v4.4.3 [20].
## Prevalence analysis
The prevalence of the species was calculated for different severities of COVID-19 patients (critical, mild and asymptomatic) and healthy controls separately using prevalence matrix generated using ggplot2 [19] in R v4.4.3 [20]. A cutoff of 50 was applied, ensuring that only the top 50 most prevalent species were displayed in the prevalence profile. The shared, accessory, and unique species across the different groups were visualised using the InteractiVenn webserver [28].
## Prediction of functions
Functional profiling of microbial communities from 16S rRNA was performed using PICRUSt [29] to predict Kyoto Encyclopaedia of Genes and Genomes (KEGG) metabolic pathways, which are available in ImageGP2 [27]. Statistical differences in KEGG level 2 functional pathways between critical COVID-19 patients and non-critical individuals (including healthy controls) were assessed using pairwise Wilcoxon tests implemented with the dplyr package [30] in R v4.4.3 [20]. At KEGG level 3, differential analysis was done using the LEfSe method [26], and features with p-value < 0.05 and LDA scores (log10) ≥ 2 were considered significant and visualised using ggplot2 [19] in R v4.4.3 [20].
## Results
## Study participant characteristics
A total of 87 individuals participated in the study, which were distributed across four groups based on COVID-19 status and disease severity. Twenty-two participants were included in the critical group, 19 in the mild group, 25 in the asymptomatic group and 21 people were included as healthy controls. The cohort included 34 females (39%) and 53 males (61%), indicating a male predominance within the study cohort (Table 1). The ages of the participants ranged from 1 to 86 years (Figure S1), with a median age of 39 years. In terms of comorbidities, 7 patients in the critical (ICU-admitted) group had underlying health conditions. These included one patient each with heart disease, hypertension and chronic kidney disease; two with diabetes mellitus; one with both heart disease and hypertension and one with coexisting diabetes and hypertension (Table S1). Additionally, four ICU patients presented with respiratory distress requiring supportive care. These demographic and clinical characteristics were carefully considered during downstream metagenomic analyses to assess potential confounding factors in microbiome variation.
## Microbial diversity analysis across COVID-19 severity categories
To investigate the variation in oropharyngeal microbiota composition associated with COVID-19 disease severity, alpha and beta diversity analyses were conducted across four groups such as critical, mild, asymptomatic and COVID-19-negative (control) individuals. Alpha diversity was evaluated using the Chao1 and Shannon indices, where Chao1 indicates species richness, while the Shannon index accounts for both richness and evenness. The Chao1 index did not reveal any statistically significant differences in species richness among the four groups (Figure S2). However, the Shannon diversity index demonstrated a significant reduction in microbial diversity at the species level in critically ill patients compared to all other groups (Kruskal-Wallis test, p-value < 0.001) (Figure 1a). This reduction in alpha diversity in the critical group suggests a disruption of the oropharyngeal microbial ecosystem during severe SARS-CoV-2 infection.
To further evaluate differences in microbial composition between groups, beta diversity was analysed using the Bray-Curtis dissimilarity metric and visualised through PCoA plot (Figure 1b). The results showed no statistically significant clustering of microbial communities by disease category (PERMAN-OVA, R² = 0.0439, p-value = 0.161). This finding indicates that although species richness is markedly reduced in critically ill patients, the overall community composition remains largely consistent across different COVID-19 severity groups. Furthermore, additional PCoA also revealed that neither age nor gender influenced microbial clustering patterns (Figures S3 &S4).
## Bacterial composition dysbiosis at the phylum and family levels
To gain initial insights into the microbial community structure, the bacterial composition at the phylum level was analysed across individuals with different severities of COVID-19, such as critical, mild and asymptomatic individuals, as well as the COVID-19-negative control group (Figure 1c). The oropharyngeal microbiota across all groups was predominantly comprised of Bacillota (33.22%), followed by Pseudomonadota (30.61%), Bacteroidota (25.05%), Actinomycetota (9.52%) and Fusobacteriota (1.42%) (Table S2). To determine whether these microbial distributions varied significantly across the clinical groups, a non-parametric Kruskal-Wallis test was applied. Among the identified phyla, Fusobacteriota emerged as the only group exhibiting a statistically significant difference in relative abundance (pvalue < 0.001). Subsequent pairwise Wilcoxon tests revealed that the Fusobacteriota was significantly more abundant in the mild, asymptomatic and control groups compared to the critical group (Figure S5).
To further elucidate the structure of oropharyngeal bacterial communities across varying COVID-19 severity levels, a comprehensive analysis was conducted at the family taxonomic level. Among all the study participants, Prevotellaceae emerged as the most prevalent family, comprising 20.34% of the total bacterial abundance (Table S3). This was followed by Streptococcaceae (13.12%), Pasteurellaceae (8.77%) and Enterobacteriaceae (6.91%). Additionally, some specific bacterial families were more prominent within particular clinical categories. In critically ill patients, Lachnospiraceae (6.2%), Ruminococcaceae (6.25%) and Corynebacteriaceae (8.51%) were found to exceed the 5% abundance threshold in addition to the most dominant families, suggesting their possible involvement in disease progression or secondary infections. The mild COVID-19 group was characterised by notable representation of Micrococcaceae (5.23%) and Sphingobacteriaceae (5.33%). On the other hand, the COVID-19-negative control group was dominated by Veillonellaceae (6.94%) and Pseudomonadaceae (6.35%), indicating their possible role in maintaining health in the absence of infection. To identify taxonomic differences with statistical confidence, a non-parametric Kruskal-Wallis test was conducted across the four groups. This analysis revealed 36 bacterial families exhibiting significant variation in abundance (p-value < 0.05), as visualised in the heatmap (Figure 2). The heatmap delineated four distinct clustering patterns among the microbial families (Clusters I-IV). Notably, two Clusters (II and III) exhibited similar abundance patterns across the mild, asymptomatic and control groups, while showing clear differences from those found in the critical category. The Cluster II included 8 bacterial families, including Moraxellaceae, Corynebacteriaceae, Staphylococcaceae, Propionibacteriaceae, Bradyrhizobiaceae, Methylobacteriaceae, Lactobacillaceae and Acetobacteraceae, which presented markedly higher relative abundance in critically ill patients. Conversely, the Cluster III comprised 10 families, including Neisseriaceae, Fusobacteriaceae, Bacillales incertae sedis, Veillonellaceae, Leptotrichiaceae, Streptococcaceae, Micrococcaceae, Actinomycetaceae, Coriobacteriaceae and Clostridiales Family XIII Incertae Sedis, which were significantly depleted in the critical group compared to the other categories.
Interestingly, within the critical group (n = 22), a further stratification was observed based on the presence or absence of respiratory distress. Specifically, four patients (Sample IDs: 2221933, 2221557, 2221589 and 2221490) exhibited notable breathing distress. The microbial profiling highlighted clear differences between patients with and without respiratory distress, especially within Clusters II and III, indicating that respiratory issues may contribute to additional changes in the microbial community (Figure 2). To ensure accurate interpretation and reduce confounding effects, these subgroup-specific microbial variations were considered in the downstream analyses.
## Genus-level taxonomic profiling and differential abundance analysis
To further explore the differences in the composition of oropharyngeal microbiota, a genus-level analysis was performed by comparing COVID-19 patients across different severity levels with healthy controls. This analysis was performed using the LEfSe method, which is well-suited for detecting biomarkers across multiple groups. A stringent LDA score threshold of >3 was applied to ensure high-confidence to identify the discriminative microbial features. The overall taxonomic structure is represented in an LEfSe cladogram (Figure 3a), which shows hierarchical relationships along with the key taxa enriched in each group. A total of 84 microbial taxa (species, genus, family, order, class or phylum) showed differential abundance across all groups. Among these, 38 genera were significantly altered among the study groups. The bar chart in Figure 3b illustrates these results, emphasising the genera that most strongly discriminate between the critical, mild, asymptomatic and control groups. Notably, 17 genera were enriched in critically ill patients, primarily belonging to the orders Bacillales, Eubacteriales, Rhodospirillales and Enterobacterales. Additionally, 8 genera were found to be enriched in the mild category and 7 genera were associated with asymptomatic individuals. The healthy control group showed enrichment of only 6 genera, primarily from the orders Bacteroidales and Campylobacterales, indicating the presence of a more balanced oropharyngeal microbiota in healthy control and less severe COVID-19 infection.
To further delineate differences between COVID-19 patients and healthy individuals, an integrated LEfSe analysis was carried out using the control group and combined COVID-19 samples (critical, mild and asymptomatic). This comparison revealed 20 microbial taxa with discriminatory potential (Figure S6). Specifically, Rothia, Catenibacterium and Acetobacter were significantly enriched in COVID-19 patients, whereas Megasphaera, Veillonella, Fusobacterium, Brucella, Zooshikella and Moraxella were more enriched in healthy controls. Moreover, given our earlier observations of taxonomic shifts at the family level, particularly among patients with respiratory distress, the critical group was further stratified into a subcategory termed 'criticalB' (critical with breathing distress) and conducted a separate LEfSe analysis comparing critically ill patients with and without respiratory distress. This analysis identified 59 microbial taxa with differential abundance between these subgroups (Figure S7). Among these, 8 genera, including Actinomyces, Rothia, Alloscardovia, Atopobium, Streptococcus, Afipia, Nitratireductor and Rhodopseudomonas, were significantly enriched in critical B patients exhibiting respiratory distress.
## Species-level bacterial prevalence across COVID-19 severity groups and controls
The prevalence of microbial taxa at the species level was analysed across four distinct groups: critical, mild, asymptomatic COVID-19 patients and healthy controls. The top 50 most abundant species identified within each group are shown in Figure 4. In the critical group, the top 3 most dominant species were Prevotella copri, Faecalibacterium prausnitzii and Prevotella stercorea. The mild group was characterised by Rothia mucilaginosa, Streptococcus salivarius and Prevotella melaninogenica, while the asymptomatic group exhibited a different profile dominated by Prevotella copri, Haemophilus parainfluenzae and Rothia mucilaginosa. The control group, on the other hand, showed a prevalence of Prevotella copri, Haemophilus parainfluenzae and Prevotella melaninogenica. Notably, no single species was consistently dominant across all four groups, although Prevotella copri appeared recurrently in most of the categories. The core microbiome analysis revealed 17 shared species, suggesting a conserved microbial baseline, despite differences in disease severity (Figure S8). The analysis of group-specific species within the top 50 revealed distinct microbial signatures associated with disease severity. The critical group had the highest number of unique species (n = 17), followed by the mild group (n = 9), the control group (n = 3) and the asymptomatic group (n = 2). Additionally, 13 accessory species were identified as being uniquely shared among the mild, asymptomatic and control groups, while being absent in the top 50 species of the critical group.
## Bacterial composition dysbiosis at the species level
To gain deeper insight into microbiome alterations associated with varying COVID-19 severity, a species-level differential abundance analysis was conducted across all study groups. Using LEfSe, a total of 63 species were identified as differentially enriched among the critical, mild, asymptomatic and control groups. The LEfSe-derived cladogram depicting taxonomic differences down to the species level is presented in Figure S9. Among these differentially enriched species, 20 were significantly elevated in the critical group and 13 in the control group. The boxplots (Figures S10 &S11) depict these findings, highlighting distinct bacterial species between healthy individuals and those with severe disease. A separate LEfSe analysis was also conducted by considering the criticalB subgroup. This targeted analysis revealed a distinct pattern of microbial enrichment, with 14 species differentially abundant in the criticalB group, as shown in the boxplot (Figure 5). Interestingly, the majority of enriched species differed between the critical and criticalB groups when compared with the less severe groups, with Rothia dentocariosa being the sole species common to both. This finding suggests that, despite sharing similar disease severity, these subgroups possess markedly different microbial composition. The species enriched in the criticalB group consists of Alloscardovia omnicolens, Actinomyces odontolyticus, Actinomyces graevenitzii, Actinomyces viscosus, Atopobium parvulum, Streptococcus thermophilus, Granulicatella adiacens, Granulicatella elegans, Lactobacillus fermentum, Rothia dentocariosa, Rothia mucilaginosa, Stomatobaculum longum, Streptococcus peroris and Rothia aeria. These genera represent key microbial signatures associated with the criticalB phenotype. The majority of these species belong to specific genera, comprising three from Rothia, three from Actinomyces, two from Streptococcus and two from Granulicatella. Among these species, Rothia mucilaginosa stood out as the predominant species in the criticalB group, with an average relative abundance of approximately 20%.
## Predicted metabolic functions in COVID-19 critical and non-critical conditions
To determine whether the differences in Shannon diversity are reflected in the functional potential, PICRUSt analysis was performed on the 16S rRNA profiles of critical and non-critical groups. The comparison of predicted metabolic functions at KEGG level-2 indicated a clear difference between the critical and non-critical groups, with 10 categories showing significant differences (Figure S12). In particular, critical samples exhibited enrichment of carbohydrate metabolism, transcription, enzyme families and secondary metabolite biosynthesis, whereas non-critical samples were enriched in replication and repair, translation, genetic information processing, folding/sorting/degradation, glycan biosynthesis and cell growth and death. To gain deeper insights, the functional profiles were analysed at KEGG level 3 using LEfSe, which identified 48 pathways that significantly differed between these groups (Figure 6). Specifically, critical patients showed higher abundance of carbohydrate metabolism (starch and sucrose, fructose and mannose, galactose, pentose metabolism and glucuronate interconversions) and amino acid metabolism (arginine, aspartate, glutamate, cysteine, methionine, histidine, lysine, proline, alanine and cyanoamino acid metabolism). They also exhibited enrichment of stress/adaptation-associated functions (sporulation, two-component systems, signal transduction, protein kinases and cytoskeletal proteins) and pathways linked to cofactor and secondary metabolite biosynthesis (thiamine, porphyrin and chlorophyll metabolism, phenylpropanoid biosynthesis, streptomycin biosynthesis, sphingolipid metabolism and methane metabolism), as well as transcription factors and ion-coupled transporters. In contrast, noncritical patients presented enrichment of pathways including core cellular processes (translation factors/ proteins, ribosome biogenesis, protein folding/processing), DNA repair and recombination (base excision repair, mismatch repair, homologous recombination) and metabolic and biosynthetic functions (biosynthesis of valine, leucine, isoleucine, folate, fatty acid, lipid, metabolism of pyruvate and glycerophospholipid). Additional enrichment was observed in cell structure and interaction-related functions (peptidoglycan biosynthesis, lipopolysaccharide biosynthesis membrane and intracellular structural molecules and pores/ion channels), microbial competition/stress tolerance mechanisms (tetracycline biosynthesis, glutathione metabolism, cell motility and secretion) and protein export.
## Discussion
This study highlights the significant impact of SARS-CoV-2 infection on the oropharyngeal microbiome, with marked microbial dysbiosis observed in critically ill patients. A notable reduction in alpha diversity was detected in this group, reflecting a disrupted microbial community, while mild and asymptomatic cases retained diversity levels similar to the healthy controls. These findings are consistent with previous reports that demonstrated reduced alpha diversity in severe COVID-19 cases [31,32]. However, studies that did not stratify patients by disease severity often reported non-significant changes in diversity [33][34][35], indicating the importance of severity-based subgrouping in COVID-19 microbiome studies. Despite changes in richness, beta diversity analysis revealed no significant clustering by severity, age, or gender, indicating that the overall community structure remains relatively conserved, a trend also observed in earlier reports [31]. The participants were recruited across a wide range of ages to examine possible agerelated influences on microbial composition, and individuals with comorbidities were also included to explore their possible associations. However, no significant effects of these factors were observed.
At the phylum level, a significant depletion of Fusobacteriota in critically ill patients emerged as a key feature of microbial dysbiosis. This finding is consistent with previous studies that showed a reduced abundance of Fusobacterium periodonticum across multiple taxonomic levels, including phylum (Fusobacteriota), family (Fusobacteriaceae), and genus (Fusobacterium), and they are potentially linked to altered sialic acid metabolism in COVID-19 [36]. However, other studies have shown elevated levels of Fusobacterium nucleatum in COVID-19 patients [37,38], suggesting that species-level variation may influence disease outcomes. Further, the microbial community shifts at the family level underscored disease-specific signatures, with critically ill COVID-19 patients showing significant enrichment of Cluster II families, including Moraxellaceae, Corynebacteriaceae, Staphylococcaceae, Propionibacteriaceae, Bradyrhizobiaceae, Methylobacteriaceae, Lactobacillaceae and Acetobacteraceae. These findings are consistent with earlier reports identifying Moraxellaceae, Staphylococcaceae and Corynebacteriaceae were more abundant in COVID-19 patients [35,39]. Notably, Corynebacterium accolens has been linked to ventilator-associated pneumonia in severe COVID-19 cases [40]. Several of these taxa are opportunistic or nosocomial pathogens that are potentially affected by hospital-related factors, medical treatments and antibiotic usage [41,42]. A recent study [43] reported that bacterial and fungal infections acquired in hospital settings were common complications among COVID-19 patients admitted to the intensive care unit. This finding indicates that hospital-related factors may compound COVID-19-related dysbiosis associated with COVID-19 [41,42]. The inclusion of the criticalB subgroup further demonstrated that respiratory distress is associated with alterations in the oropharyngeal microbiota, providing a more detailed understanding of dysbiosis in the context of severe disease and respiratory complications.
Further, genus-level microbial analysis identified unique profiles in patients experiencing respiratory distress, with the criticalB group showing significant enrichment of genera such as Actinomyces, Rothia, Alloscardovia, Atopobium, Streptococcus, Afipia, Nitratireductor and Rhodopseudomonas. Many of these genera are known to be associated with respiratory diseases. For instance, Streptococcus has been shown to exacerbate pulmonary dysfunction [44], Rothia has been linked to cases of pneumonia in immunocompetent individuals [45], and Actinomyces is known to cause pulmonary actinomycosis [46]. These findings suggest that shifts in the oropharyngeal microbiota may influence the development of severe respiratory conditions, highlighting the potential role of microbial composition in the pathogenesis of SARS-CoV-2. Species-level analysis further highlights the disruption of the microbiome in relation to disease severity. The analysis of the top 50 most prevalent taxa showed that the critical group contained the highest number of unique species [17], while the mild, asymptomatic and control groups shared 13 species that were absent in the top 50 species of critically ill patients, highlighting a distinct microbiome profile in the critical cases compared to the less severe cases. Notably, Rothia mucilaginosa was found to be enriched in the criticalB group, comprising 20% of the microbial community, and has been associated with respiratory infections in immunocompromised hosts, indicating its potential role in COVID-19-related pulmonary complications [47].
Furthermore, the observed reduction in Shannon diversity in critical COVID-19 patients was also evident in their functional capacity. The predicted metabolic pathways of critically ill patients showed an increased representation of stress adaptation and survival, including two-component regulatory systems, signal transduction processes, sporulation, sphingolipid metabolism and diverse nutrientutilization functions. Notably, sphingolipid metabolism is recognised as microbial signatures of pathogenesis [48]. In contrast, non-critical patients exhibited enrichment of biosynthetic and translational functions, such as pyruvate metabolism, fatty acid and branched-chain amino acid biosynthesis, ribosome biogenesis, translation-associated factors and protein folding pathways. These functions point toward a more metabolically versatile and diverse microbiome capable of sustaining growth, protein synthesis and structural maintenance. Whereas, severe disease is accompanied by functional reprogramming of the oral microbiome toward survival and stress resilience at the expense of biosynthetic and growth-related processes. The lower representation of these functions in critical patients may result from the decline in microbial diversity and overrepresentation of a narrower set of taxa.
Taken together, these findings highlight notable shifts in the microbiota associated with severe COVID-19, suggesting that the changes in the microbiota may function as both biomarkers and modulators of disease outcomes. In particular, the significant enrichment of Rothia mucilaginosa in critically ill patients implicates it as a potential microbial marker of severe respiratory outcomes, underscoring its role in disease progression and positioning it as a promising prognostic indicator.
Despite the promising findings, this study has several limitations. The reliance solely on 16S rRNA gene sequencing restricts taxonomic resolution to the genus/species level and provides only predicted rather than direct functional insights, limiting the depth of microbial characterization. The relatively small sample size, particularly for the criticalB subgroup, reduces the statistical power and limits generalizability. In addition, only limited clinical information beyond disease severity and comorbidities was available, constraining associations between microbiome changes and patient outcomes. The cross-sectional design captures the microbiota at a single time point and does not reflect the temporal dynamics of microbial shifts during the course of infection. Furthermore, confounding factors such as prior antibiotic or medication use were not fully controlled for, which may have influenced the results. These limitations highlight the need for future studies with larger, longitudinal cohorts, comprehensive clinical metadata and multi-omics approaches such as shotgun metagenomics, metatranscriptomics and metabolomics to better understand the intricate interplay between the microbiome, host immunity and SARS-CoV-2 pathophysiology.
## Conclusions
In conclusion, this study underscores the significant impact of COVID-19 severity on the oropharyngeal microbiome, with critically ill patients exhibiting marked dysbiosis, including reduced alpha diversity and enrichment of opportunistic pathogens. Notably, Rothia mucilaginosa was identified as a key species associated with severe respiratory complications, highlighting the potential role of microbial composition in the pathogenesis of SARS-CoV-2. The identification of microbial signatures specific to disease severity highlights the potential of microbiome profiling as a tool for diagnosing COVID-19 severity, predicting clinical outcomes and suggesting appropriate therapeutic interventions.
## Ethical statement
This study was reviewed and approved by the Ethics Committee of the ICMR-National Institute of Virology, Pune (NIV/EC/Oct/2021/D-1; No. 20-3-16R), in accordance with established ethical guidelines for biomedical research involving human samples.
## References
1. Mathieu, Ritchie, Rodés-Guirao (2020) "Coronavirus pandemic (COVID-19)"
2. Guan, Ni, Hu (2020) "Clinical characteristics of coronavirus disease 2019 in China" *NEJM*
3. Rovito, Augello, Ben-Haim (2022) "Hallmarks of severe COVID-19 pathogenesis: a pas de deux between viral and host factors" *Front Immunol*
4. Pathak, Yan, Zhang (2021) "The role of oral microbiome in respiratory health and diseases" *Respir Med*
5. Gupta, Bhanushali, Sanap (2022) "Oral dysbiosis and its linkage with SARS-CoV-2 infection" *Microbiol Res*
6. Hanada, Pirzadeh, Carver (2018) "Respiratory viral infection-induced microbiome alterations and secondary bacterial pneumonia" *Front Immunol*
7. Radaic, Kapila (2021) "The oralome and its dysbiosis: new insights into oral microbiome-host interactions" *Comput Struct Biotechnol J*
8. Ramos-Sevillano, Wade, Mann (2019) "The effect of influenza virus on the human oropharyngeal microbiome" *Clin Infect Dis*
9. Manohar, Loh, Nachimuthu (2020) "Secondary bacterial infections in patients with viral pneumonia" *Front Med*
10. Ganesan, Peter, Withanage (2024) "COVID-19 associated oral and oropharyngeal microbiome: systematic review and meta-analysis" *Periodontology 2000*
11. Soffritti, 'accolti, Fabbri (2021) "Oral microbiome dysbiosis is associated with symptoms severity and local immune/inflammatory response in COVID-19 patients: a cross-sectional study" *Front Microbiol*
12. Lin, Yang, Wen (2021) "Crosstalk between the oral microbiota, mucosal immunity, and the epithelial barrier regulates oral mucosal disease pathogenesis" *Mucosal Immunol*
13. Jiang, Yang, Qian (2023) "Tongue coating microbiome composition reflects disease severity in patients with COVID-19 in Nanjing" *China. J Oral Microbiol*
14. Haran, Bradley, Zeamer (2021) "Inflammation-type dysbiosis of the oral microbiome associates with the duration of COVID-19 symptoms and long COVID" *JCI Insight*
15. Paine, Choudhury, Alam (2024) "Multi-faceted dysregulated immune response for COVID-19 infection explaining clinical heterogeneity" *Cytokine*
16. Bhanu, Buchke, Hemandhar-Kumar (2025) "Comparative metagenomic analysis of the oral microbiome in COVID-19 patients and healthy individuals" *Sci Rep*
17. Choudhary, Vipat, Jadhav (2020) "Development of in vitro transcribed RNA as positive control for laboratory diagnosis of SARS-CoV-2 in India" *Indian J Med Res*
18. Caporaso, Kuczynski, Stombaugh (2010) "QIIME allows analysis of high-throughput community sequencing data" *Nat Methods*
19. Gómez-Rubio (2017) "ggplot2-elegant graphics for data analysis (2nd Edition)" *J Stat Softw*
20. Core "R: A language and environment for statistical computing"
21. Ahlmann-Eltze, Patil, Ggsignif "R package for displaying significance brackets for "ggplot2"
22. Kassambara (2019) "rstatix: pipe-friendly framework for basic statistical tests. CRAN: contributed packages"
23. Dixon (2003) "VEGAN, a package of R functions for community ecology" *J Veg Sci*
24. Şahin, Aybek (2020) "Jamovi: an easy to use statistical software for the social scientists" *Int J Assess Tools Educ*
25. Metsalu, Vilo (2015) "ClustVis: a web tool for visualizing clustering of multivariate data using principal component analysis and heatmap" *Nucleic Acids Res*
26. Segata, Izard, Waldron (2011) "Metagenomic biomarker discovery and explanation" *Genome Biol*
27. Chen, Liu, Chen (2024) "ImageGP 2 for enhanced data visualization and reproducible analysis in biomedical research" *iMeta*
28. Heberle, Meirelles, Da Silva (2015) "InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams" *BMC Bioinform*
29. Langille, Zaneveld, Caporaso (2013) "Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences" *Nat Biotechnol*
30. Hadley, Maintainer (2019) "Package 'plyr'. A grammar of data manipulation. R package version"
31. Ma, Zhang, Zhou (2021) "Metagenomic analysis reveals oropharyngeal microbiota alterations in patients with COVID-19" *Signal Transduct Target Ther*
32. Wu, Cheng, Jiang (2021) "Altered oral and gut microbiota and its association with SARS-CoV-2 viral load in COVID-19 patients during hospitalization" *NPJ Biofilms Microbiomes*
33. Bourumeau, Tremblay, Jourdan (2023) "Bacterial biomarkers of the oropharyngeal and oral cavity during SARS-CoV-2 infection" *Microorganisms*
34. Lai, Cheung, Lui (2022) "Limited impact of SARS-CoV-2 on the human naso-oropharyngeal microbiota in hospitalized patients" *Microbiol Spectr*
35. Thissen, Morrison, Mulakken (2022) "Evaluation of co-circulating pathogens and microbiome from COVID-19 infections" *PLoS One*
36. Nardelli, Gentile, Setaro (2021) "Nasopharyngeal microbiome signature in COVID-19 positive patients: can we definitively get a role to fusobacterium periodonticum?" *Front Cell Infect Microbiol*
37. Bao, Zhang, Lyu (2021) "Beware of pharyngeal fusobacterium nucleatum in COVID-19" *BMC Microbiol*
38. Wolff, Martiny, Deyi (2021) "COVID-19-associated fusobacterium nucleatum bacteremia" *Belgium. Emerg Infect Dis*
39. Aljabr, Dandachi, Abbas (2024) "Metagenomic next-generation sequencing of nasopharyngeal microbiota in COVID-19 patients with different disease severities" *Microbiol Spectr*
40. Liu, Beck, Fisher (2021) "The brief case: ventilator-associated Corynebacterium accolens pneumonia in a patient with respiratory failure due to COVID-19" *J Clin Microbiol*
41. Markovskaya, Gavioli, Cusumano (2022) "Coronavirus disease 2019 (COVID-19): secondary bacterial infections and the impact on antimicrobial resistance during the COVID-19 pandemic" *Antimicrob Steward Healthc Epidemiol*
42. Saha, Dubourg, Yacouba (2022) "Profile of the nasopharyngeal microbiota affecting the clinical course in COVID-19 patients" *Front Microbiol*
43. Søgaard, Baettig, Osthoff (2021) "Community-acquired and hospital-acquired respiratory tract infection and bloodstream infection in patients hospitalized with COVID-19 pneumonia" *J Intensive Care*
44. Dickson, Erb-Downward, Huffnagle (2014) "Towards an ecology of the lung: new conceptual models of pulmonary microbiology and pneumonia pathogenesis" *Lancet Respir Med*
45. Martínez, Molina, Sevila (2014) "Rothia mucilaginosa pneumonia in an immunocompetent patient" *Arch Bronconeumol*
46. Yuan, Hou, Peng (2022) "Pulmonary actinomyces graevenitzii infection: case report and review of the literature" *Front Med*
47. Bilici (2023) "Rothia mucilaginosa pneumonia developing after COVID-19" *J Clin Images Med Case Rep*
48. Wang, Chen, Li (2021) "Functions of sphingolipids in pathogenesis during host-pathogen interactions" *Front Microbiol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12835043&blobtype=pdf | # Handling editor Nicolas Bejerman
Tamara Collum, Andrew Stone, Elizabeth Rogers
## Abstract
Plum pox virus (PPV) is a serious viral threat to stone fruit trees worldwide. Wild Prunus species including American plum (Prunus americana) can serve as sources of inoculum. High-throughput sequencing was used to characterize PPV populations in American plum and peach after aphid inoculation and after two cycles of cold-induced dormancy (CID). A significant decrease in the number of sequence variations in the PPV genome was observed after CID in American plum, but not in peach. Seventeen were identified as unique to American plum, while eight were unique to peach. These findings provide insight into the genetic diversity of PPV in a potential reservoir host.
6]. However, the native North American P. americana, commonly called American plum, has the potential to act as a reservoir host for Pennsylvania isolates of PPV-D and may potentially serve as a source of infection for important commercial crops such as peach, but little is known about PPV population dynamics in this alternative host [7]. In 2019, the U.S. Department of Agriculture declared that PPV had been eradicated in the United States [8]. PPV is also considered eradicated in Nova Scotia, Canada, but it is still present in the Niagara Region in Ontario, Canada [9].
PPV is a member of the genus Potyvirus in the family Potyviridae. The PPV genome is typical for a potyvirus and is comprised of a positive-sense single-stranded RNA containing a large open reading frame (ORF) that encodes 10 proteins (P1, HC-Pro, P3, 6K1, CI, 6K2, VPg, NIa-Pro, NIb, and CP) and an additional small overlapping ORF called PIPO that is translated as fusion product, P3N-PIPO, by slippage of the viral RNA-dependent RNA polymerase [2,10].
Viral genetic diversity plays a significant role in virus-host interactions and adaptation to new environments [11][12][13]. RNA virus populations are considered to exist as 'quasi-species' or groups of closely related genomic sequences [14]. Sequence diversity is increased by the error-prone nature of viral RNA polymerases, which lack proofreading activity, and the large population sizes in infected hosts. The longlived nature of perennial fruit trees can also lead to the evolution of complex viral populations [15]. Diversity can be reduced through selection and bottlenecks, which can occur during transmission by aphid vectors or movement within Plum pox virus (PPV), the causative agent of the disease sharka, is the most serious viral threat to stone fruit trees worldwide. Since its first detection on plum in Bulgaria around 1917, the global cost of PPV has been estimated to exceed $15 billion [1]. The name sharka is the Bulgarian word for pox. In addition to the chlorotic rings and spots that gave plum pox its name, symptoms can include vein clearing, leaf distortion, fruit deformation and discoloration, reduced fruit quality, and premature fruit drop. Ten PPV strains are currently recognized based on sequencing and phylogenetic analysis. The host range of PPV is overlapping and only partially strain-specific. PPV-D is the prevalent strain of PPV and is globally widespread [1]. Historically, PPV-D has been associated with plum infections, but it can also infect peach, nectarine, apricot, and almond [1][2][3][4].
In the United States, PPV-D was first found in 1999 in Pennsylvania. Although the Pennsylvania isolates of PPV-D have a large host range under experimental conditions, most PPV infections in the United States have been found in cultivated peach (Prunus persica) and plum (P. domestica) [5, a host [16,17]. High-throughput sequencing technology has improved our ability to characterize genetic variation within virus populations and detect viral variants that occur at low frequencies [18 -20]. To date, research examining the genetic diversity of PPV-D has primarily relied on analysis of consensus sequences and has not been performed in the reservoir host American plum.
In this study, we used high-throughput sequencing to identify variations in the PPV-D genome in American plum and peach (P. persica cv. GF305). PPV-D Penn39 was maintained both by grafting and aphid transmission to healthy Prunus persica cv. GF305 peach seedlings for approximately 20 years in the U.S. Department of Agriculture (USDA)-Agricultural Research Service plant BSL3 containment facility at Ft. Detrick, Maryland. All trees were grown under standard greenhouse conditions (14 h daylight, temperatures 24 to 26 °C daytime, 20 °C nighttime) and vernalized every 3 to 6 months for a 60to 90-day cold-induced dormancy (CID) period in a 4 °C dark cold box. Healthy GF305 peach and American plum Fig. 1 Maximum-likelihood phylogenetic tree of selected PPV-D isolates. The tree is rooted on two M isolates (at the top of the tree). Lines connecting the D and M clades have been shortened to 10% of their original length to facilitate display on a single page. Each isolate is labeled with its country of origin and GenBank accession number. Consensus sequences from this study are labeled in orange text 1 3 trees, 1 to 2 years old, were inoculated using green peach aphids (Myzus persicae) placed on detached symptomatic leaves from a Penn39-positive GF305 peach tree as described previously [7]. Leaf punches were taken from systemic leaves from four individual trees of each species at 60 days after initial aphid infection and at 60 days post CID2. Each tree served as a biological replicate with 10 systemic leaf punches pooled per tree. Total RNA was isolated using a Plant RNeasy Kit according to the manufacturer's specifications (QIAGEN, Germantown, MD) and sent to Azenta Life Sciences (South Plainfield, NJ) for poly-A selection and cDNA library preparation.
Libraries were barcoded and sequenced on an Illumina HiSeq platform, yielding between 12 and 129 million 150-bp paired-end reads per sample (Supplementary Table S1). Sequencing data are available under NCBI bioproject accession no. PRJNA881753 ( h t t p s : / / w w w . n c b i . n l m . n i h . g o v / b i o p r o j e c t / 8 8 1 7 5 3).
The total reads obtained from each sample were trimmed and filtered based on their quality score and mapped to the sequence of the inoculated strain, PPV-D Penn39 (PX208212) using CLC Genomics Workbench. Analysis of variants was carried out using the CLC Genomics Workbench low-frequency variant detection tool with the following parameters: Required significance = 1.0%, Min coverage = 10, Min count = 2, Min frequency = 1.0, Ignore non-specific matches = Reads. The Illumina sequencing error rate has been estimated to be approximately 0.1% [21], so a cutoff of 1.0% was applied for identifying PPV variants, which is tenfold higher than what would be expected from sequencing errors. After filtering, a total of 813 unique variations were identified from all samples and time points, 74% of which were nonsynonymous. Read counts and mapping statistics for all samples are available in Supplementary Table S1, and all of the variations identified in this study can be found in Supplementary Table S2.
Nearly complete genome sequences could be assembled from each sample, and these were 99.69-99.83.69.83% identical to the Penn39 source genome. A maximum-likelihood phylogenetic tree (Fig. 1) was constructed from these consensus sequences and other complete PPV-D genome sequences using Geneious Prime v. 2025.2.1 and IQTree v. 2.4.0 as described previously [5]. All 16 consensus sequences formed a well-supported clade that was sister to the Penn39 source and within a larger clade of sequences from Pennsylvania. It is notable that the four American plum post-CID2 samples grouped together, which reflects the genome changes that occurred over two growth and dormancy cycles in the novel American plum host.
The number of variations observed per sample ranged from 36 to 185 for peach and 37 to 183 for American plum (Supplementary Table S1). In peach, there was no significant difference between the total number of variations observed in post-inoculation samples and samples collected after CID2. However, in American plum, the total number of variations observed decreased after CID2 (Fig. 2). We did not observe a relationship between the total number of PPV reads and the number of variations observed.
Of the 813 total variations observed, 267 (33%) were unique to peach samples and 437 (54%) were unique to American plum samples. Ninety and 93% of the host-specific variations (239/267 for peach and 406/437 for American plum) were observed only in one sample (Supplementary Table S2). To further analyze differences in PPV populations between hosts, we only considered variations that were observed in at least four of the 16 samples. This included 103 variations, 43 of which (42%) were nonsynonymous Fig. 2 Comparison of the number of nucleotide variations in the PPV genome in infected peach and American plum leaves. Samples were collected after inoculation and after the second round of cold-induced dormancy (CID2). The observed variations include nucleotide polymorphisms, indels, and SNPs. Bars represent the mean ± SE of four biological replicates. Statistical analysis was performed using a twotailed Student's t-test and E6276L in VPg (Fig. 3B). SNVs that were only observed in American plum mapped to the P1, HC-Pro, P3, 6K1, CI, NIa-Pro, NIb, and CP coding regions (Fig. 3A). Only six of the SNVs unique to American plum were nonsynonymous. These include A228T and N744D in P1, M3870V in CI, D8535N in NIb, and E8585D and P8665L in CP (Fig. 3B). With the exception of P8665L, these SNVs have been observed in other PPV-D isolates.
In conclusion, the total number of PPV variants observed in American plum decreased significantly post-CID2, but the total number of PPV variants observed in peach did not change. In contrast, there were more host-specific PPV variations in American plum than in peach (17 vs. eight). While further experiments will be needed to determine the role the identified variations may play in viral biology, this study provides new information on the genetic variation of PPV and its population dynamics in the alternative host American plum. and 101 of which were single-nucleotide variations (SNVs). An insertion at position 2907 was detected in 15 of the 16 samples. This insertion, an 'A' at position 2907, is part of the slippage site preceding the PIPO ORF and was observed at a frequency of 4.9% in peach and 5.6% in American plum (Supplementary Table S3). This is higher than the 1.6% slippage rate reported previously for P. domestica 'Jojo' infected with PPV-D and PPV-Rec [22] but similar to the 4.0% rate observed in P. domestica 'Stanley' and 'President' infected with PPV-D [19]. A deletion at position 7786 was observed in two American plum samples and three peach samples at a frequency of 1.01-5.31.01.31%, and this deletion results in a premature stop codon in the NIb coding region.
Of the 101 SNVs, 14 occurred in all 16 trees at a frequency of >67%. These mapped to the P1, HC-Pro, P3, CI, 6K2, VPg, NIb, and CP coding regions (Fig. 3A). A3296T in P3 and A7400G in NIb were the only two SNVs not observed in other PPV-D isolates based on the analysis of consensus sequences of complete PPV genomes retrieved from GenBank (Supplementary Table S4). Only three SNVs (A3296T in P3, A6279G in VPg, and A8892G in CP) were nonsynonymous; these three, along with most other SNVs, were observed in other PPV-D isolates.
Eight of the 101 SNVs were observed only in peach and 17/101 SNVs were observed only in American plum (Fig. 3A). Fisher's exact test was used to identify variations that were significantly more common in American plum than in peach samples (FDR p-value < 0.05). These included 12 of the 17 SNVs identified as unique to American plum and one shared SNV at position 7892 that was observed in all eight American plum samples and in only one peach sample (Fig. 3). No variations were significantly more common in peach than in American plum, as all of the variations that were unique to peach occurred only in four or five of the eight peach samples.
SNVs observed only in peach mapped to the P1, HC-Pro, CI, VPg, and NIb coding regions. Two SNVs that were unique to peach were nonsynonymous: I1803F in HC-Pro
## References
1. (2019) "USDA Declares United States Free from Plum Pox Virus"
2. Cfia (2022) "Update to plum pox virus pest fact sheet" *Plant Health Risk Assessment Unit*
3. Yang, Li, Wang (2021) "Research advances in potyviruses: from the laboratory bench to the field" *Annu Rev Phytopathol*
4. Schneider, Roossinck (1128) "Genetic diversity in RNA virus quasispecies is controlled by host-virus interactions" *J Virol*
5. Nigam, Latourrette, Souza et al. (2019) "Genome-wide variation in potyviruses" *Front Plant Sci*
7. Rubio, Galipienso, Ferriol (1092) "Detection of plant viruses and disease management: Relevance of genetic diversity and evolution" *Front Plant Sci*
8. Domingo, Sheldon, Perales (2012) "Viral quasispecies evolution" *Microbiol Mol Biol Rev*
9. Predajňa, Šubr, Candresse et al. (1016) "Evaluation of the genetic diversity of Plum pox virus in a single plum tree" *Virus Res*
10. Ali, Li, Schneider (2006) "Analysis of genetic bottlenecks during horizontal transmission of Cucumber mosaic virus" *J Virol*
11. Li, Roossinck, Da Silva et al. (1128) "Genetic bottlenecks reduce population variation in an experimental RNA virus population"
12. Tamukong, Collum, Stone (2020) "Dynamic changes impact the plum pox virus population structure during leaf and bud development" *Virology*
13. Beerenwinkel, Zagordi (2011) "Ultra-deep sequencing for the analysis of viral populations" *Curr Opin Virol*
14. Goodwin, Mcpherson, Mccombie (2016) "Coming of age: ten years of next-generation sequencing technologies" *Nat Rev Genet*
15. Rodamilans, Valli, Mingot (2015) "RNA polymerase slippage as a mechanism for the production of frameshift gene products in plant viruses of the Potyviridae family" *J Virol*
16. "Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References"
17. García, Rodamilans, Martínez-Turiño (2025) "Plum pox virus: An overview of the potyvirus behind sharka, a harmful stone fruit disease" *Annals Appl Biol*
18. Rodamilans, Valli, García (2020) "Molecular plant-plum pox virus interactions" *Mol Plant Microbe Interact*
19. Sihelská, Glasa, Šubr (2017) "Host preference of the major strains of Plum pox virus-Opinions based on regional and world-wide sequence data" *J Integr Agric*
20. Rogers, Stone, Burchard (2024) "Almond can be infected by Plum pox virus-D isolate Penn4 and is a transmissioncompetent host" *Plant Dis*
21. (1910)
22. Rogers, Stone, Sherman (2025) "Phylogenetic reconstruction from sequences of plum pox virus samples collected in the United States points to multiple, independent introductions. PhytoFrontiers"
23. Schneider, Damsteegt, Gildow (0256) "Molecular, ultrastructural, and biological characterization of Pennsylvania isolates of Plum pox virus" *Phytopathology*
24. Collum, Stone, Sherman (2022) "Viral reservoir capacity of wild prunus alternative hosts of plum pox virus through multiple cycles of transmission and dormancy" *Plant Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12411050&blobtype=pdf | # Testing a Susceptible Population Density Among Other Explanatory Factors of African Swine Fever Spread in Wild Boar Using the Russian Federation Data, 2007-2023
O Zakharova, E Liskova, N Gladkova, I Razheva, I Iashin, A Blokhin, D Kolbasov, F Korennoy
## Abstract
This study aims to identify the role of various natural, socioeconomic, and demographic factors in the development of the African swine fever (ASF) epizootic among wild boar in the Russian Federation (RF) from 2007 to 2023. In this study, particular emphasis was placed on testing the significance of wild boar population density as a key factor contributing to the spread of ASF within this population. During the study period, 1711 outbreaks in wild boars were reported in the RF, accounting for 41.7% of all ASF outbreaks in the country. We tested two regression approaches to model the dependance of the total number of ASF outbreaks in second-level municipal units (districts) on a range of potential explanatory factors, including the dynamically changing annual population density of wild boar. We employed negative binomial regression (NBR) and, as an alternative approach, classification and regression trees (CARTs). The predictive capabilities of both models were evaluated using 10-fold cross-validation. One of the most significant identified factors was the number of ASF outbreaks in domestic populations, which may indicate a close coexistence of both domestic and wild ASF cycles. Population density showed limited significance in the negative binomial model (p ¼ 0:05). The CART model demonstrated high significance for this factor in the Far Eastern regions of the country, where the highest number of outbreaks occurred at density values above 0.120 individuals/km 2 . For the European part of the RF, the threshold density value was 0.026 individuals/km 2 , which closely corresponds to the threshold established by country's authorities for managing wild boar populations to prevent the spread of ASF. The results demonstrated a complex and nonlinear influence of wild boar population density and ASF outbreaks among domestic pigs on the likelihood of new infection foci emerging in the wild fauna. The modeling results indicated that although both types of models had comparable predictive capabilities, the CART approach provided better visualization and understanding of the analysis results. These findings can be used to optimize population management activities to regulate wild boar numbers in infection hotspots across different geographical areas delineated by the risk level of infection spread.
## 1. Introduction
African swine fever (ASF) is a transboundary viral disease that affects both domestic pigs and wild boar and causes significant damage to the pig farming industry in many countries. ASF can manifest in both large-scale and localized epidemics, and it is associated with various risk factors that must be considered when selecting appropriate surveillance and control strategies [1][2][3].
Current scientific discussions are actively addressing the role played by domestic pigs and wild boars in the emergence and spread of ASF in areas that have not yet been reported.
Research and experiments aim to identify the contribution of each of these animal species to the chains of virus transmission, as well as to assess the potential and conditions for controlling and preventing the spread of infection [4,5].
The spread of ASF occurs not only through direct contact between susceptible animals but also indirectly through contact with infected carcasses of wild boars or contaminated objects in the surrounding environment [6,7]. The spread of ASF can be enabled by the movement of contaminated items, such as mammals or birds scavenging carcasses, or biting flies/ticks [8].
Understanding the various mechanisms of ASF transmission in terms of the wild boar-domestic pig-environment ecosystem will help specialists develop effective disease control strategies and minimize the risk of emerging new epidemics [9,10].
Despite the efforts made by many countries to develop an effective vaccine against ASF, the current strategy for combating the disease is based on assessing the risk factors for the spread of the infection, strictly adhering to biosecurity measures in the management of domestic pigs and wild boar and applying culling and cleansing/disinfection in response to a confirmed outbreak. Biosecurity includes controlling the movement of animals, maintaining hygiene standards, disinfecting equipment and care items, as well as zoning areas. Additionally, an effective fight against ASF involves informational campaigns aimed at pig farmers, agricultural workers, hunters, and the general public regarding preventive measures and control of infection spread [11][12][13].
Risk factors contributing to the spread of the ASF virus among wild boar encompass a wide and diverse range of predictors that have not yet been thoroughly studied and are represented in various scenarios of epizootic development. These factors may include the dynamics of wild boar population trends, changes in habitat conditions, contact with other animal species, and the impact of human economic activities on the spread of the infection. A comprehensive study of these factors will enhance our understanding of ASF transmission mechanisms and help us develop optimal strategies for controlling the virus among wild boar, which is crucial for effective management of the disease [14][15][16].
Studying the theoretical foundations for determining threshold values of susceptible animals in the emergence of ASF is crucial for planning measures to reduce wild boar populations around infection foci and subsequent campaigns for population management in disease-free areas to prevent the spread of the infection [17].
The aim of this study was to assess the significance of wild boar population density and to evaluate other risk factors for the spread of ASF among wild boar in regions of the Russian Federation (RF) using regression analysis methods. In our work, we analyzed a wide range of independent variables, including environmental factors such as forest cover percentage, the density of major roads, and the percentage of water bodies, as well as demographic characteristics expressed through wild boar population density, human population density, density of settlements, and the number of hunted wild boars, carcasses, and remains of infected animals found, which have potential significance in the ASF epizootic in the RF. Of particular interest in our study was the factor of wild boar population density, as the confirmation of ASF in wildlife necessitates wild boar depopulation measures, which are implemented in infection hotspots to regulate animal numbers to the threshold value recommended by government authorities of 0.025 individuals/km 2 .
## 2. Materials and Methods
## 2.1. Study Area.
For the study, model subjects (first-level administrative divisions) of the RF were selected based on the following criteria: (a) ASF outbreaks among wild boars were regularly registered during the analyzed period in the same districts; (b) the presence and availability of annual data on wild boar population numbers at the district level (second-level administrative divisions). The model region presented for regression analysis consisted of 39 subjects of the RF and was divided into two territories based on geographic criteria: the European territory, comprising 35 subjects, and the Far Eastern territory, including four subjects. The model subjects included 2440 districts, which were the units of analysis in our study.
## 2.2. ASF Data.
Data on ASF outbreaks in the RF were obtained from the World Organization for Animal Health's animal disease notification database (WOAH WAHIS) for the period from 2007 (a year when the disease was first introduced to Russia) to 2023. During the study period, a total of 1711 ASF outbreaks in wild boars were registered in the model districts, with 1308 outbreaks noted in the European zone and 403 outbreaks in the Far Eastern region. In this study, we considered an "outbreak" as a case of ASF in wild boars confirmed by laboratory methods and notified to WOAH, defined by geographical coordinates and the date of occurrence [18] (Figure 1).
## 2.3. Explanatory Factors.
The most significant factors that play a role in the spread of ASF among wild boar were selected as independent variables based on a literature review [18][19][20][21][22].
Data on wild boar population numbers in the studied districts from 2007 to 2023, as well as information on carcasses of animals that died from various reasons and their remains found during routine monitoring activity, were obtained from statistical reports of the regional ministries of natural resources and ecology of the RF (https://www.mnr.gov.ru/ about/).
Environmental variables were gathered from the vector and raster GIS layers of Open Street Maps (OSMs) (https://www. openstreetmap.org/#map=3/69.62/-74.90). The landscape variables included the percentage of water bodies (rivers and lakes), the area and share of vegetation cover, and the length and density of roads. All variables were extracted and summed by district, and median values were calculated using GIS zonal statistics tools. Data on population density and the number of settlements in the districts was obtained from the Federal State Statistics Service website (https://rosstat.gov.ru/). All shares were calculated based on the total area of the districts. The explanatory variables included in the analysis are presented in Table 1.
All variables were initially analyzed for mutual correlation using the nonparametric Spearman correlation test with a threshold value of r s = 0.7 to avoid multicollinearity. In each pair of correlated variables, the one demonstrating the lowest correlation with other variables was retained for analysis [23].
## 2.4. Regression Analysis.
In our study, we tested two regression approaches to examine the possible relationship between the cumulative intensity of ASF outbreaks in the model districts and various explanatory factors, including wild boar population density. In both cases, the response variable was the total annual number of ASF outbreaks in wild boar in the district [24].
One of the tested regression models was the negative binomial regression (NBR) model (NBRM), which is traditionally used to analyze count data with overdispersion [25,26]. Additionally, this regression was used to investigate the significance of only one factor, namely the density of the wild boar population itself, in each model region.
As an alternative, we used the classification and regression tree (CART) model [27]. This approach is more flexible regarding variables of different scales and the potential presence of nonlinear relationships between the explanatory variables and the response variable [23].
Regression modeling was conducted simultaneously for two territories: the European part of Russia and the Far Eastern region.
2.4.1. NBRM. The NBRM is a specific type of regression used for count data when the variation of the response variable exceeds its mean (i.e., when overdispersion is observed) [28,29]. The choice of NBR in our case was justified by the distribution of the number of outbreaks in wild boars across municipal districts, where the mean is 1.84 and the variance is 38.41.
To ensure a stable and reliable selection of variables for the regression model, we applied the Lasso method. Lasso regression is a type of linear regression that introduces a regularization penalty to the loss function during training. This penalty is proportional to the absolute value of the coefficients, encouraging the model not only to fit the data but also to minimize the magnitude of the model weights. This characteristic makes Lasso regression particularly effective for feature selection, as it can reduce the number of features by setting the coefficients of less important variables to zero. All variables with zero coefficients were excluded from further analysis. The analysis was conducted in the R programing environment. The significance of the variables was assessed using Student's t-test with the corresponding p-value (a p-value ≤ 0.05 indicates sufficient statistical significance of the variable as a predictor in the regression model). The overall model fit quality was evaluated using the coefficient of determination R 2 , which represents the proportion of variance of the response variable accounted for by the model. The spatial distributions of both model residuals were assessed using Moran's I spatial autocorrelation test, which demonstrates the correspondence between the observed spatial distribution of the analyzed variable and a hypothetical random distribution. Moran's I value close to zero, corresponding to low z-scores (p >0:05), indicate a near-normal distribution. The presence of spatial autocorrelation in the residuals indicate an unexplained clustering of the phenomenon under study that is not accounted for by the explanatory factors [30].
The negative binomial model with all selected factors was applied to the European and Far Eastern model regions.
## 2.4.2. CART Analysis.
CART analysis is a nonlinear nonparametric model constructed by binary partitioning a multidimensional set of covariates [26,31].
At each step, the CART model divides the observations using a simple decision rule (e.g., if the wild boar density is less than 0.025 individuals/km 2 , then other variables need to be considered; if the density is above 0.025 individuals/km 2 , the risk of an outbreak increases and the presence of other predictors has negligible weight). This rule was chosen to minimize the diversity (regarding the binary outcome or classification) in the right and left "child nodes." Branches and nodes are added until a stopping criterion is reached, and the tree is completed with "leaves" or "bins," which contain the proportions of correctly and incorrectly classified observations [32].
The Gini index was used as the variable splitting method, and 10-fold cross-validation was applied to assess the predictive power of the resulting trees. The CART automatically performs cross-validation by growing the maximum number of branches on subsets of the data and then calculating error rates based on the unused portions of the dataset.
The completed CART analysis resulted in a "tree" with multiple partitions or branches depicted as branches. The independent variables and their splitting points are chosen to optimize a given suitability criterion, such as minimizing the residual sum of squares (MSE) applied to continuous data [33].
Regression and classification trees have advantages over other types of regression analysis in that they can handle various types of explanatory variables and do not require any data transformations [34,35]. Compared to approaches based on linear regression, an advantage of CART analysis is that it can account for nonlinear relationships between the dependent variable and the set of explanatory factors. Missing or unaccounted values of independent variables have little impact on Length and density of the road network km and km/km 2 A dense road network increases the accessibility of an area by hunters and can increase the detection of infected animals, and is also an indirect indicator of economic activity of the population.
[4, 5, 9, 23]
## Number of small-scale pig farms Unit
An increase in the number of pig farms, especially small ones, is associated with an increased frequency of infected domestic animals. [4,5] ASF outbreaks in domestic pigs Unit Proximity to outbreaks in domestic increases the likelihood of between-population contacts. [4,5] Wild boar population density Animals/km 2 High boar density is directly related to the likelihood of disease occurrence. [4,5,9,23] the outcome of the analysis. Thus, the method utilizes the best available information in the absence of variable values. In datasets of acceptable quality, this allows for the inclusion of all observations. CART methods provide a visual representation of the decision tree, which is intuitive and likely to be more acceptable to those unfamiliar with statistical analysis. The tree diagrams generated from the CART analysis can help structure explanations of the predictions.
The CART analysis was conducted for the final variables by region of the subjects of the RF using the rpart package [36], implemented in the R programing environment.
## 2.4.3. Comparison of Regression Methods: Indicators for
Evaluating the Performance of Models. The quality assessment of the constructed predictive regression models, examining the dependance of ASF outbreak intensity among wild boars on a set of risk factors, was conducted using k-fold cross-validation. This method is based on partitioning the data into a training set, used for model training, and a validation set, used to evaluate the prediction error.
To evaluate the predictive capability of the models, statistical indicators such as R 2 (coefficient of determination), root mean square error (RMSE), and mean absolute error (MAE) were used [36][37][38]. In our study, we applied 10-fold cross-validation to evaluate the quality of the constructed models [39]. The performance quality assessment of the regression models was conducted using the "caret" package in the R software.
2.5. Software. Preliminary data processing and evaluation were conducted using Microsoft Office Excel (Microsoft Corporation, Redmond, Washington, USA). Visualization of the ASF epizootic situation and the wild boar population density was performed using ArcMap 10.8.2 (Esri, Redlands, California, USA). The statistically-oriented programing environment R (R Core Team, 2023) was used for regressionmodeling. This accounted for 41.7% of all ASF outbreaks in the country during these years. The highest number of outbreaks among wild boars was recorded in the years 2013 (116 outbreaks), 2016 (118 outbreaks), 2020 (170 outbreaks), and 2021 (104 outbreaks) (Figure 2).
## 3. Results
Geographically, massive outbreaks of ASF among wild boars were concentrated in the following subjects: Ryazan, Moscow, Tula, Tver, Vladimir, Smolensk, and Samara regions, as well as in the Pskov and Leningrad regions, which are adjacent to the border with Estonia. In the Far East, long-term persistence of the ASF virus has been noted in Primorsky Krai and in border regions-local areas where the population density of wild boars remains relatively high (Figure 1).
There are epidemiological features of the infection's manifestation in different regions. In some areas, the ASF outbreaks in wild boar were sporadic [38] (e.g., in Nizhny Novgorod Oblast) [19], while in other regions, it has been characterized by short-term but large-scale epizootics with widespread distribution across a significant part of the affected area. For instance, in Samara Oblast, such a massive epizootic occurred in 2020, with 60 ASF outbreaks in wild boars recorded over the year [39].
## 3.2. Regression Modeling Results
## 3.2.1. NBRM.
The NBRM applied to the regions of the European part of the RF (Zone 1 at Figure 1) identified the following significant factors associated with the intensity of ASF outbreaks among wild boars: the proportion of forest cover, the number of ASF outbreaks in domestic pigs, and the density of the wild boar population (the latter with a marginal significance level of p ¼ 0:05) (Table 2). The spatial autocorrelation test on the model residuals returns Moran's I coefficient of -0.265 (p ¼ 0:532) suggesting near-normal distribution of the residuals.
In the regions of the Far Eastern model zone (Zone II at Figure 1), the following significant factors were identified: the number of wild boars found dead, the number of ASF outbreaks among domestic pigs, and the density of the wild boar population (the latter with a marginal significance level of p ¼ 0:05) (Table 3). The spatial autocorrelation test on the model residuals returns Moran's I coefficient of -0.027 (p ¼ 0:432) suggesting near-normal distribution of the residuals.
The regression analysis with only a single factor (the density of the wild boar population) conducted for each of the model regions showed that for 17 out of 39 regions, the density of the wild boar population was a statistically significant predictor of the intensity of outbreak occurrences (Table 4).
## 3.2.2. CART Analysis.
The classification tree diagrams constructed using CART for the study of ASF outbreaks in wild boar intensity in both the European and the Far Eastern parts of Russia are presented in Figures 3 and4, respectively.
For the model pertaining to the European part of the RF, the significant factors were (in order of significant descend): the presence of outbreaks among domestic pigs, the density of the wild boar population, and the proportion of forest cover. The most significant node is represented by the occurrence of outbreaks among domestic pigs, suggesting that ASF in the wild population primarily arises in the presence of a certain number of outbreaks among domestic pigs. It should be noted that the significance of the threshold wild boar population density for this area is indicated by a cutoff point of 0.026 animals/km 2 . The third most significant factor in the model for the European part of Russia was the indicator of forest cover in the region, suggesting that with more than 30% forest coverage, there were 16 cases (13%) where the ASF outbreaks among wild boars persisted even with a low number of outbreaks among domestic pigs.
In the model for the Far Eastern part of the RF, the most significant factor is the density of the wild boar population, indicating that at densities above 0.12 individuals/km 2 , the highest proportion of outbreaks occurred without additional conditions. At densities lower than 0.12 individuals/km 2 , additional factors contributing to the occurrence of outbreaks in wild boars included the presence of ASF in the domestic pig population as well as a considerable number of found carcasses or remains of wild boars from which genetic material of the ASF virus was isolated.
When more than two ASF outbreaks are registered in the domestic pig population, the model, determined by the outcome variable of ASF outbreak intensity among wild boars, establishes a dependency of the epizootic on the number of found dead carcasses or remains of animals. This factor is represented by a final cutoff point of 138 individuals and was the concluding factor in determining significant elements of the epizootic identified through the modeling process. Table 5 presents the predictive significance of risk factors for ASF among wild boars, determined by the significance weights (%), identified in the CART models for the two studied geographical territories.
## 3.2.3. Comparative Metrics of Predictive Ability of Negative
Binomial and CART Regression Models. The quality assessment indicators for the fit of the NBR and CART models, applied to reveal the risk factors for ASF outbreaks among wild boars in the European part and the Far Eastern region of the RF, are presented in Table 6, where, for each model, training and validation performance metrics are provided.
## 4. Discussion
Despite the efforts of scientists from many countries to develop a safe and effective vaccine against ASF, the current strategy for eradicating this disease relies on assessing the risks posed by identified factors that contribute to the spread of the infection, as well as strict adherence to biosecurity measures in animal husbandry. Most actions for the eradication and prevention of ASF are based on traditional principles of disease control, including epidemiological surveillance, investigation and destruction of infected herds, establishment of disease control zones, and movement restrictions and control of wild boar population, which may include fencing, depopulation, and passive and active monitoring measures.
Particular interest lies in analyzing the risk factors that facilitate the spread of the disease among wild boars, including the potential introduction of ASF into areas that are currently free of the virus [6]. The ecological cycle involving wild boars and the presence of the ASF virus in the environment is a major issue in contemporary ASF epidemiology, as not all mechanisms for the persistence of the pathogen in affected areas have yet been uncovered [7,8].
Currently, there are extensive discussions regarding the significance of wild boar population density in the spread of ASF during localized introductions of the pathogen. Based on the experiences of European countries studying wild boar population density, this relationship is present but does not always hold primary importance in the occurrence of ASF outbreaks [5,19,20].
The key objective of this study was to utilize existing information on the population size and density of animals, as well as recorded ASF outbreaks among wild boars, available to professional epidemiologists, to identify the main risk factors contributing to the further expansion of the virus. A number of potential explanatory variables were tested using various analytical approaches, including linear NBR and CART analysis.
The comparative analysis of prediction errors from models employing different methodological approaches demonstrated a slight advantage for the CART model. This approach facilitates the development of clearer and more explainable regression structures for forecasting the significance of risk factors for ASF outbreaks among wild boars in affected regions of Russia.
Modeling results indicated that the occurrence of outbreaks among domestic pigs is the primary factor associated with ongoing local ASF epizootics in wild boars. The failure to confirm the reverse hypothesis-that wild boar cases influence outbreaks in domestic pigs, as previously tested [40]-may suggest that ASF in wild fauna plays a secondary role in the transmission dynamics. The factor of the number of wild boar remains found dead from various reasons emerged as significant in the models for the Far Eastern territory. On one hand, this relationship demonstrates a natural pattern in the frequent finding of remains in areas affected by ASF epizootics. On the other hand, it underscores the importance of enhanced passive monitoring measures based on the search for and removal of fallen boar carcasses. Indirectly, this may indicate that the remains of wild boars serve as a natural reservoir for the ASF virus, posing a threat in terms of maintaining the circulation of the virus. It should also be noted that the identified relationships are influenced by the quality of the raw data on reported cases in both domestic and wild pigs, which may be subject to underestimation.
The factor of forest cover proportion, which demonstrated significance in the model for the European territory of the RF, indicates a natural tendency for the development of ASF epizootics in areas with a larger habitat areas suitable for wild boars. In contrast, for the Far Eastern territory, this factor was not among the significant variables due to the generally denser forest cover in this region. Specifically, the range of forest cover proportion for the European territory is 33% AE 29%, whereas for the Far Eastern territory, it is 64% AE 27%.
This fact can also explain the higher threshold density value for wild boar as a risk factor in the Far Eastern territory. The identified threshold value for the European territory (0.026 animals/km 2 ) almost exactly coincides with the density threshold recommended by the ASF control strategy in Russia (0.025 individuals/km 2 ).
Based on the modeling results for individual subjects of the RF experiencing prolonged ASF outbreaks among wild boars, the subjects listed in Table 3 were identified, where a significant while wild boar population density plays an important role in the spread of the infection, it cannot be considered a primary factor, and control of ASF in this population should not be limited to depopulation measures only [13].
It should also be noted that the population density of wild boars is a value characterized by a high degree of uncertainty. Several reasons for this uncertainty may include: (1) the unreliability of primary data on wild boar population counts due to the inadequate reliability of the accounting methods used [13];
(2) the use of the total area of a region as a denominator in calculations can lead to inaccuracies, as the actual distribution area of wild boars may be much more compact, resulting in locations with localized high densities. Seasonal migrations of wild boars also play a significant role, leading to constant fluctuations in areas of increased local animal density.
The identified relationship somewhat accounts for the patterns of the epizootic process that determine the role of the susceptible population in the transmission of infection. The higher the level of the existing density of the susceptible population in the infection focus, the faster and more extensive the spread of the infection occurs, making it more challenging to control the process.
In cases of reduced wild boar populations, there is a possibility of infectious material persisting in the environment, creating conditions for a sustained epizootic process of ASF in specific, geographically constrained areas, leading to the formation of endemic regions. These areas are also difficult to combat and control with the standard preventive measures typically employed for ASF eradication [4,5,41].
## 5. Conclusion
A comparative analysis of the results from the constructed regression models leads to the following key conclusions:
1. The presence and number of ASF outbreaks among domestic pigs are determinants of the presence of ASF among wild boars in the study area. 2. The density of wild boar populations is a significant, albeit not decisive, factor in the development of epizootics in wild boar. 3. The regression approach using the CART method is a reliable modeling tool that provides more easily interpretable results. 4. To enhance the reliability of predictive models, efforts are needed to develop and implement more advanced methods for estimating wild boar population numbers.
## References
1. Dixon, Stahl, Jori et al. (2020) "African Swine Fever Epidemiology and Control" *Annual Review of Animal Biosciences*
2. Dixon, Sun, Roberts (2019) "African Swine Fever" *Antiviral Research*
3. Sur (2019) "How Far Can African Swine Fever Spread?" *Journal of Veterinary Science*
4. Baños, Boklund, Gogin (2022) "Epidemiological Analyses of African Swine Fever in the European Union" *EFSA Journal*
5. More, Miranda, Bicout (2018) "African Swine Fever in Wild Boar" *EFSA Journal*
6. Gervasi, Marcon, Guberti (2022) "Estimating the Risk of Environmental Contamination by Forest Users in African Swine Fever Endemic Areas" *Acta Veterinaria Scandinavica*
7. Carlson, Fischer, Zani (2020) "Stability of African Swine Fever Virus in Soil and Options to Mitigate the Potential Transmission Risk" *Pathogens*
8. Rietz, Ischebeck, Conraths (2024) "Scavenger-Induced Scattering of Wild Boar Carcasses Over Large Distances and Its Implications for Disease Management" *Journal of Environmental Management*
9. Depner, Gortazar, Guberti (2017) "Epidemiological Analyses of African Swine Fever in the Baltic States and Poland" *EFSA Journal*
10. Loi, Cappai, Coccollone et al. (2019) "Standardized Risk Analysis Approach Aimed to Evaluate the Last African Swine Fever Eradication Program Performance, in Sardinia" *Frontiers in Veterinary Science*
11. Danzetta, Marenzoni, Iannetti et al. (2020) "African Swine Fever: Lessons to Learn From Past Eradication Experiences. A Systematic Review" *Frontiers in Veterinary Science*
12. Zakharova, Korennoy, Yashin (2023) "Spatiotemporal Patterns of African Swine Fever in Wild Boar in Revealing High-Risk Areas" *Animals*
13. Zakharova, Blokhin, Toropova et al. (2022) "Density of Wild Boar Population and Spread of African Swine Fever in the Russian Federation" *Veterinary Science Today*
14. Schulz, Staubach, Blome (2019) "Analysis of Estonian Surveillance in Wild Boar Suggests a Decline in the Incidence of African Swine Fever" *Scientific Reports*
15. Bergmann, Schulz, Conraths et al. (2021) "A Review of Environmental Risk Factors for African Swine Fever in European Wild Boar" *Animals*
16. Lim, Andraud, Kim et al. (2019) "Three Years of African Swine Fever in South Korea"
17. Pepin, Borowik, Frant et al. (2023) "Risk of African Swine Fever Virus Transmission Among Wild Boar and Domestic Pigs in Poland"
18. Zani, Masiulis, Bušauskas (2020) "African Swine Fever Virus Survival in Buried Wild Boar Carcasses" *Transboundary and Emerging Diseases*
19. Boklund, Dhollander, Vasile (2019) "Risk Factors for African Swine Fever Incursion in Romanian Domestic Farms During"
20. Boklund, Ståhl, Miranda Chueca (2024) "Risk and Protective Factors for ASF in Domestic Pigs and Wild Boar in the EU, and Mitigation Measures for Managing the Disease in Wild Boar" *EFSA Journal*
21. Masiulis, Bušauskas, Jonušaitis et al. (2019) "Potential Role of Domestic Pig Carcasses Disposed in the Forest for the Transmission of African Swine Fever" *Berliner und Münchener tierärztliche Wochenschrift*
22. Cukor, Linda, Václavek (2020) "Wild Boar Deathbed Choice in Relation to ASF: Are There Any Differences Between Positive and Negative Carcasses?" *Preventive Veterinary Medicine*
23. Podgórski, Borowik, Łyjak et al. (2020) "Spatial Epidemiology of African Swine Fever: Host, Landscape and Anthropogenic Drivers of Disease Occurrence in Wild Boar" *Preventive Veterinary Medicine*
24. Wahis "WAHIS: World Animal Health Information System"
25. Bellini, Casadei, Lorenzi et al. (2021) "A Review of Risk Factors of African Swine Fever Incursion in Pig Farming Within the European Union Scenario" *Pathogens*
26. O'neill, White, Ruiz-Fons et al. (2020) "Modelling the Transmission and Persistence of African Swine Fever in Wild Boar in Contrasting European Scenarios" *Scientific Reports*
27. Hilbe (2007) "Negative Binomial Regression Second Edition-Negative Binomial Regression: Second Edition Joseph M (Hilbe Frontmatter More information"
28. Li, Wong, Lamoureux et al. (2012) "Are Linear Regression Techniques Appropriate for Analysis When the Dependent (Outcome) Variable Is Not Normally Distributed?" *Investigative Opthalmology & Visual Science*
29. Sroka, Nagaraja (2018) "Odds Ratios From Logistic, Geometric, Poisson, and Negative Binomial Regression Models" *BMC Medical Research Methodology*
30. Breiman, Friedman, Olshen et al. (2017) "Classification and Regression Trees"
31. Porter, Leblond, Lecollinet (2011) "Clinical Diagnosis of West Nile Fever in Equids by Classification and Regression Tree (CART) Analysis and Comparative Study of Clinical Appearance in Three European Countries"
32. Baker (1993) "Classification and Regression Tree Analysis for Assessing Hazard of Pine Mortality Caused by Heterobasidion annosum" *Plant Disease*
33. Megahed, Kandeel, Alshaya (2022) "A Comparison of Logistic Regression and Classification Tree to Assess Brucellosis Associated Risk Factors in Dairy Cattle" *Preventive Veterinary Medicine*
34. Pfeiffer, Stevenson, Firestone et al. (2021) "Using Farmer Observations for Animal Health Syndromic Surveillance: Participation and Performance of an Online Enhanced Passive Surveillance System" *Preventive Veterinary Medicine*
35. Therneau, Atkinson, Ripley (2013) "Rpart: Recursive Partitioning"
36. Chicco, Warrens, Jurman (2021) "The Coefficient of Determination R-Squared Is More Informative Than SMAPE, MAE, MAPE, MSE and RMSE in Regression Analysis Evaluation" *PeerJ Computer Science*
37. Picard, Cook (1984) "Cross-Validation of Regression Models" *Journal of the American Statistical Association*
38. Liland, Skogholt, Indahl (2024) "A New Formula for Faster Computation of the K-Fold Cross-Validation and Good Regularisation Parameter Values in Ridge Regression" *IEEE Access*
39. Zakharova, Blokhin, Burova et al. (2007) "Spatiotemporal Analysis of African Swine Fever Spread in Wild Boar Population in Russian Federation"
40. Zakharova, Blokhin, Yashin et al. (2023) "Investigation of Risk Factors Associated With the African Swine Fever Outbreaks in the Nizhny Novgorod Region of Russia, 2011-2022" *Transboundary and Emerging Diseases*
41. Glazunova, Korennoy, Sevskikh et al. (2021) "Risk Factors of African Swine Fever in Domestic Pigs of the Samara Region" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12315483&blobtype=pdf | # Design and Analysis of Novel HEV Vaccine Variants and Evaluation of Two Selected Candidates in a Porcine Infection Model
Isabella Hrabal, Elmira Aliabadi, Saskia Weber, | George, Liam Ssebyatika, Thomas Krey, Cora Holicki, Laura Schmid, Katja Dinkelborg, Charlotte Schröder, Christine Fast, | Patrick Behrendt, Martin Groschup, Martin Eiden
## Abstract
Background and Aims: Hepatitis E virus (HEV) poses a significant global health concern, with millions of annual infections and a notable impact on public health. Although HEV is the leading cause of acute viral hepatitis worldwide, there is a substantial lack of approved and licensed vaccines. In this study, we evaluated the efficacy of several protein-and DNA-based vaccine candidates against HEV using a combined in vitro/in vivo workflow. Methods: Corresponding vaccine candidates were produced, biochemically analysed and characterised. The general immunogenicity of suitable vaccine candidates was initially evaluated in a rabbit model. Resulting antibodies were assessed for their reactivity and neutralising efficiency. Finally, the most effective candidates were tested in a pig infection model using a primeboost vaccination regimen. Results: Using this approach, we analysed a total of seven vaccine candidates and demonstrated that the two most promising candidates significantly reduced virus shedding in swine faecal samples after infection. However, no sterile immunity was achieved.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.
Conclusions: This study conducted a comprehensive analysis to establish a rational approach for post-vaccination immune responses in pigs. The insights gained from this research are expected to significantly contribute to the development and evaluation of future vaccine candidates for pig herds, ultimately reducing viral dissemination among pigs and preventing HEV transmission from pigs to humans. These findings hold important translational value, offering a foundation for both improving animal health and safeguarding public health.
## 1 | Background and Aims
The hepatitis E virus (HEV) is a leading cause of acute viral hepatitis, particularly prevalent in developing countries. It is known to cause acute liver inflammation, ranging from a selflimiting illness to a severe, potentially life-threatening condition [1].
HEV belongs to the Hepeviridae family, which is divided into two subfamilies: Orthohepevirinae and Parahepevirinae. Orthohepevirinae encompasses four genera: Avihepevirus, Chirohepevirus, Rocahepevirus and Paslahepevirus. Within the genus Paslahepevirus, there are a total of eight genotypes, with genotypes 1-4 (HEV-1-4) being mainly associated with human infections [2]. HEV leads to frequent outbreaks in Africa and Asia, with considerable impact on public health, often connected to contaminated water sources [3].
The zoonotic genotypes HEV-3 and HEV-4 primarily infect pigs and wild boars as their main reservoir and are particularly widespread in industrialised countries [4][5][6]. In Germany, HEV-3 has the highest prevalence, resulting in an estimated 417 000 seroconversions per year [7]. Transmission to humans primarily occurs through direct contact with the faeces of infected pigs or the consumption of undercooked pork products [8][9][10]. Although most HEV infections usually lead to self-limiting diseases, chronic infections with HEV-3 have been reported, especially in immunocompromised individuals, representing an additional burden for disease control [1].
The development of vaccines plays a crucial role in reducing the spread of the virus. In recent years, various vaccines against different HEV genotypes have been developed [11]. Notably, Hecolin (Xiamen Innovax Biotech Co., Xiamen, China), utilised in humans, has demonstrated a protective effect against HEV-4 [12]. It has now been approved and licensed for humans in China for over a decade [12,13] and has recently also been approved in Pakistan [14]. However, a recent publication has shown that the vaccine offers only partial protection against HEV-3 in a pig model [15]. Further efforts are therefore necessary to develop an effective vaccine against all genotypes, including HEV-3.
In this study, we focused on establishing a standardised workflow to evaluate seven new vaccine candidates. This included comprehensive in vitro analyses to assess the candidates, as well as an in vivo approach to determine their effectiveness in a pig infection model. We aimed for the development of a vaccine against HEV-3 in pig herds to reduce viral dissemination between pigs, thereby mitigating the risk of infection in humans through a One Health approach.
## 2 | Material and Methods
## 2.1 | Animals
In total, seven rabbits were obtained from an internal Friedrich-Loeffler-Institut (FLI; Greifswald-Riems, Germany) breeding programme and kept in the local husbandry. The competent authority of the Federal State of Mecklenburg-Vorpommern, Germany, was notified based on national legislation (LALLF MV 7221.3-2-042/17).
Twenty-two 11-week-old mast hybrid piglets were purchased from a commercial HEV-negative herd from a local breeder (Landboden Glasin, Glasin, Germany). The experiment was approved by the State Office for Agriculture, Food Safety and Fishery in the Federal State of Mecklenburg-Western Pomerania, Germany, based on national and European legislation, EURL 63/2010 for the protection of laboratory animals (LALLF M-V 7221.3-1-010/22).
All possible steps were taken to improve animal wellbeing and to keep the number of animals in the experiment to the required minimum.
## 2.2 | Vaccine Design
The design of the majority of the analysed vaccine candidates is based on a partial sequence of the capsid protein from a German HEV-3 isolate (GenBank Acc. Number KP294371.1 [16]). This includes the two bacterial expressed open reading frame (ORF) 2 constructs p429 and p429-ORF3, which encompass 429 amino acids of the central region of the ORF2, with or without fusion to ORF3, connected via a helix-forming peptide linker. In addition, three DNA-based vaccines, pVax1-ub-HEV-SMP, pVax1-ub-HEV-SMP-ORF3 and pVax1-ub-HEV-ORF3-SMP, were prepared. They code for the 3 functional (S, M, P) capsid domains either solely or fused to ORF3 protein at the 3′ or 5′ end of the capsid sequence. The sequences were cloned into a modified pVAX1-Ub universal fusion vector which enables expression of a 5′-ubiquitin antigen fusion protein [17]. Corresponding sequences and details are compiled in the Sequences S1-S5.
In addition, non-secreted and secreted forms of HEV-3 P domain (pGS99/100, respectively) were cloned into a pMT vector and expressed in Drosophila S2 cells as described previously [18]. The gene encodes the HEV-3 P domain residues 456-660 (Kernow-C1 clone p6, genotype 3, GenBank Acc. Number JQ679013). It was designed with and without a BiP signal sequence to represent the glycosylated (secreted: pGS100) and non-glycosylated (nonsecreted: pGS99) form of the P domain. All vaccine variants are summarised in Table 1.
## 2.3 | Expression and Purification of Recombinant Proteins and Plasmids
Bacterial proteins and pVax1 constructs were expressed in E. coli, while eukaryotic proteins were expressed in Drosophila S2 cells. All proteins and constructs were purified, diluted and stored at appropriate temperatures, with detailed cloning and purification methods provided in Method S1.
## 2.4 | Rabbit Immunisation
The protein vaccines were administered by mixing 600 μL of the vaccine candidate [0.5 mg/mL] with 600 μL of Gerbu adjuvant MM (Biotechnik Gerbu, Heidelberg, Germany). This mixture was applied subcutaneously to one rabbit per vaccine candidate. Three successive subcutaneous boosts followed at intervals of 3 weeks. Blood samples were taken from the lateral saphenous vein to determine the development of antibody titers. The obtained serum samples were stored at -20°C for further analysis. The experiment design is shown in Figure 2A.
Plasmid vaccine pVax1-ub-HEV-SMP was initially injected using the needle-free PharmaJet injection system Stratis IM/SC (Colorado, USA) for all plasmid-based vaccines. Using this system, 800 μL of plasmids was administered with the PharmaJet device into the semitendinosus muscle followed by 250 μL of the Gerbu adjuvant MM at the same application site while releasing the adjuvant when the needle was withdrawn. The vaccination cycle of 3 weeks continued until a plateau of serum antibodies was observed.
## 2.5 | ELISA (Enzyme-Linked Immunosorbent Assay)
An indirect antigen ELISA was performed according to standard protocol [19] using recombinant HEV antigens p239, p429 or 2xORF3 for coating (Sequences S1 and S6). A detailed description can be found in the Method S2. Additionally, a commercial ELISA (ID Screen Hepatitis E Indirect Multi-species, ID.vet Innovative Diagnostics, Grabels, France) was used following the ID Screen manual.
## 2.6 | Neutralisation Assay
Naked and pseudo-enveloped HEV-3 viral strain Kernow C1p6 G1634R was generated following the previously described method [20]. The analysis of serum samples as well as the counting of focus forming units (FFU) was performed as described in a recent publication [15]. Additional details can be found in the Method S3.
## 2.7 | Immunofluorescent Staining
Human hepatoma HepG2/C3A cells were transfected with two HEV-3 constructs ('Kernow-C1 p6 clone' and 'HEV83-2-27-clone') and treated with rabbit sera overnight, followed by staining and Mean Fluorescence Intensity (MFI) calculation as detailed in Method S6 and S7.
## 2.8 | Inoculum
A 25% (w/v) liver inoculum of a HEV positive liver sample in phosphate buffered saline (PBS) was obtained from an
## Summary
• In this study, a comprehensive method was developed to select and evaluate new vaccine candidates against HEV-3.
• These candidates underwent rigorous in vitro characterisation followed by thorough testing in two animal models. experimentally HEV-3 infected wild boar [16] (GenBank Acc. Number KP294371.1). The liver was homogenised in PBS using the TissueLyser II (Qiagen, Hilden, Germany), centrifuged at 7459×g for 5 min, pooled and filtered twice through syringe filters (0.22 μL Millex-GP 33 mm filter unit, Carrigtwohill, Ireland). The obtained infectious homogenates were aliquoted into 2 mL portions and stored at -80°C. The corresponding inocula were thawed overnight at 4°C and brought to room temperature shortly before infection.
## 2.9 | Molecular Analysis
RNA was extracted from various matrices using the NucleoVet Mag kit (Macherey Nagel, Düren, Germany) and KingFisher Flex robot (ThermoScientific, Darmstadt, Germany), then quantified by qRT-PCR targeting a conserved ORF2/3 region [16], with β-actin as a control. Standards for quantification were prepared using digital droplet PCR (ddPCR). A detailed description is found in Method S8.
## 2.10 | Experimental Design of the Pig Vaccination Study
The experiment was performed under biosafety level (BSL)-2 conditions in the corresponding animal facilities at the FLI, Germany. Twenty-two 11-week-old HEV-negative male and female mast hybrid pigs were divided into five experimental groups: Uninfected control group (n = 2), infected adjuvant control group (n = 2), infection control group (n = 6) and two vaccine groups (n = 6 each) (Table 2). The pigs were housed in separate stable units, each with 2-3 piglets of the same experimental group and gender. Vaccine groups received 300 μg of vaccine protein, mixed 1:1 (v/v) with Gerbu adjuvant F (Biotechnik Gerbu, Heidelberg, Germany). The experimental design is shown in Figure 2A and followed a recently established protocol [15]. Further details adhering to the ARRIVE guidelines including information on the pigs, health evaluation, termination criteria, acclimation period, vaccination regimen, sampling, euthanasia and necropsy are provided in Method S9 and Data S1.
## 2.11 | Statistical Evaluation
The statistical analysis was performed using R statistical software (R Core
## 3 | Results
## 3.1 | Rabbits Immunisation Elicits Neutralisation IgG With High Avidity Against HEV Genotype 3
Following the immunisation of rabbits with seven vaccine candidates, serum samples were tested for HEV-capsid-directed antibodies using ELISA, with partial HEV capsid proteins p429 and p239 as the coating antigens. Additionally, ORF3-directed antibodies were assessed using ELISA with 2xORF3 as the coating antigen. High optical densities (OD) in the ELISA were indicative of elevated levels of HEV-directed antibodies, with an upper detection limit of an OD of 4. All vaccinations were administered subcutaneously. However, four subcutaneous vaccinations with the DNA-based vaccine candidate pVax1-ub HEV-SMP was unsuccessful in producing detectable antibody formation; consequently, the administration route for DNA-based vaccines was changed to intramuscular injection.
Immunisation (Figure 1A) resulted in an increase in p429directed antibodies across all seven vaccine candidates, as evidenced by higher optical density readings. However, the immunogenicity varied among the individual vaccines. Protein-based vaccines induced capsid-directed antibody production after a single vaccination, while DNA-based vaccines showed increased antibody levels only after the second vaccination, with a slower rise of the OD compared to the protein vaccines. An antibody plateau in p429-ELISA was reached after four subcutaneous vaccinations for all protein vaccines, six intramuscular vaccinations for pVax1-ub HEV-SMP-ORF3 and pVax1-ub HEV-ORF3-SMP, and seven intramuscular vaccinations for pVax1-ub HEV-SMP (Figure 1B), in addition to the four previous ineffective subcutaneous applications with this vaccine candidate (Figure S1A). The highest antibody titers were induced by the vaccines p429, p429-ORF3 and pGS99, in which antibodies could still be detected at the highest serum dilution of 1:2 500 000. Overall, ODs between three and four were achieved in ELISA with p429 coating for all vaccine candidates except pVax1-ub HEV-ORF3-SMP, which also did not lead to an increase in anti-ORF3 antibodies (Figure 1B). These results were confirmed with a similar ELISA using the p239 partial capsid protein as the coating antigen (Figure S1B).
The binding specificity of rabbit antibodies to HEV-3 capsid proteins was confirmed by western blotting against bacterially expressed capsid proteins p239, p429 and 2xORF3 (Method S5, Figures S2B andS3). To assess whether the antibodies developed in the immunised rabbits could bind to HEV-3 subtypes in a more authentic capsid conformation, we performed immunofluorescence staining of HEV-transfected hepatoma (HepG2) cells. The DNA-based vaccine candidate pVax1-ub HEV-ORF3-SMP was not included in this assay, since it performed poorly compared to the other vaccine candidates in previous analyses. The results demonstrated that antibodies in serum samples from immunised rabbits with all protein-based and DNA-based vaccine candidates could bind to both HEV Kernow C1p6 G1634R and HEV pUC83-2 (Figure 1C). Moreover, the mean fluorescent intensity (MFI) was stronger for the samples obtained from rabbits immunised with the protein-based vaccine, compared to the DNA-based vaccine, which is consistent with the ELISA results (Figure S4).
We performed a comprehensive analysis to assess the neutralisation capacity of the sera from the rabbits immunised with different vaccine candidates against HEV after the last immunisation (Figure 1D). In our assay, all vaccinated rabbits developed neutralising activity against HEV-3. Notably, rabbits inoculated with the protein-based vaccines p429 and p429-ORF3 appeared to exhibit enhanced neutralisation efficacy compared to the other vaccine candidates, as observed in the graphical data. The dilution of the serum at which the antibodies were able to neutralise 50% of the naked and pseudo-enveloped viral infection (ID-50) is summarised in Figure S5.
To evaluate the avidity of the rabbit sera antibodies towards HEV, we utilised a urea-based avidity ELISA, employing different HEV epitopes as coating antigens including pGS99, p429, or p239 (Method S4). The titration curves of the rabbit serum samples from the various animals immunised with distinct vaccines are shown in Figure S6A. All rabbit sera showed strong binding towards pGS99 in the ELISA without urea incubation. Notably, the signal intensity did not decrease after incubation of the sera-antigen mixture with an 8 M urea solution, pointing to a high avidity of the antibody-antigen complex. The avidity indices are displayed in Figure S6B, revealing that the p429 vaccine induced antibodies with the highest avidity. Similar results were determined for the binding of rabbit sera to p239 and p429 antigens. A summary of avidity indices in ELISA using p239 and p429 is provided in the table in Figure S6B.
## 3.2 | Anti-HEV Antibody Increase in Pigs After Initial Vaccination With p429 and p429-ORF3, Followed by Decline Post-Booster and Infection
All protein vaccines elicited a similarly rapid increase in antibody response in rabbits compared to the DNA-based vaccines and showed comparable performance in p239-ELISA, Western blot and immunofluorescent staining. Overall, these results emphasise the highest potential of p429 and p429-ORF3 to elicit protective immunity against HEV-3: they were thus selected as the most promising vaccine candidates for further investigation in a pig infection model. They were freshly expressed and dialysed in carbonate bicarbonate buffer pH 10.3 (Figure S2A). The vaccination regimen in the pigs involved two intramuscular administrations, 28 days apart, with an infection 4 weeks after the second vaccination (Figure 2A).
Antibody titers against the vaccine candidates were evaluated using the p239 ELISA (Figure 2B) and the commercial ID Screen Hepatitis E Indirect Multi-species ELISA (Figure S7A) between vaccinations and after infection. Both ELISAs showed that the homologous prime-boost vaccination strategy initially led to an increase in antibody titers in all six vaccinated pigs across both vaccine groups. However, following this initial increase, antibody titers began to decrease after the booster vaccination and continued to decline after the infection on Day 56. Two pigs in the p429 vaccine group (pig 12b and pig 16), as well as one pig in the p429-ORF3 vaccine group (pig 82), exhibited a detectable rise in antibody titers at the end of the observation period. Antibody titers also increased in the infection control group as well as in the adjuvant control group at the end of the observation period, with the rise occurring earlier in the adjuvant control group.
In addition, an ELISA was performed with 2xORF3 as the coating antigen (Figure S7B). None of the pigs in the infection control group exhibited detectable antibodies against 2xORF3. In contrast, two pigs from the P429-ORF3 vaccinated group (pig 95 and pig 98) showed a slight increase in ORF3 antibodies after vaccination, which subsequently decreased in the further course of the trial.
## 3.3 | Immunisation With p429 and p429-ORF3 Vaccines Induces Neutralising Antibodies in Pigs
To evaluate the humoral immune responses caused by the p429 and p429-ORF3 vaccines in pigs, we assessed the neutralisation capacity of IgGs isolated from pig sera at 49 days postvaccination (dpv) before exposure to HEV-3. The results indicate that IgGs from all vaccinated pigs, with the exception of pig 17b, were capable of neutralising the virus in a dose-dependent manner. The data were normalised to the adjuvant control at 49 dpv to eliminate any nonspecific neutralisation caused by the adjuvants (Figure 3A). The neutralisation capacity of IgGs from each pig is combined in Figure 3B, where each dot represents one pig.
The findings demonstrate that both vaccine candidates induced neutralising antibodies, with no apparent differences between the two vaccines, as observed in the graphical data.
Subsequently, we investigated whether HEV antibodies present in the pig sera after the second vaccination and subsequent infection exhibited neutralisation activity. For this analysis, the data for the vaccine groups were normalised to the infected adjuvant control at 81 dpv to account for any nonspecific neutralisation attributed to antibodies induced by the infection. At this time point, no obvious differences were observed between the vaccine groups and the infected adjuvant control. This suggests that although the vaccines induced neutralising antibodies, these antibodies did not enhance the overall neutralisation capacity following infection (Figure 3C,D).
Additionally, the avidity of antibodies in pig serum samples after vaccination was determined using a urea-based avidity ELISA. The titration curves obtained from the average of the avidity ELISA experiments per vaccine group are depicted in Figure S8, with individual titrations from each pig shown in Figure S9. The data indicate that antibodies in the pig sera from both vaccine groups were capable of binding to all HEV antigens, with the strongest signal observed for the p239 coating protein. However, the avidity indices were only half those observed in immunised rabbits, suggesting that the binding strength of vaccine-induced antibodies was higher in rabbits than in pigs.
## 3.4 | p429 and p429-ORF3 Vaccination Reduces Viral Load Post-Infection
Viral shedding in faeces and serum was monitored throughout the experiment using RT-qPCR (Figure 2C). Three animals in the p429-ORF3 vaccine group exhibited mild diarrhoea on Day 61 (Table S3, Data S1). Viral RNA in faecal samples was detectable in all pigs (n = 6) of the non-vaccinated infection control group and all pigs (n = 2) from the adjuvant control group between Days 58 and 84. In contrast, only three out of six p429vaccinated pigs and four out of six p429-ORF3-vaccinated pigs displayed faecal virus shedding. Moreover, the onset of virus shedding was delayed 2-16 days compared to the nonvaccinated infection control group. Viremia was detectable only on one day in one pig in the infection control group (pig number 6) and one pig in the p429-ORF3 vaccine group (pig number 82, Figure S7C).
Examination of the viral load in the faeces showed a significant reduction of viral RNA in the vaccinated pigs of both vaccination groups compared to the unvaccinated control group in both tests (p < 0.01). Statistics of faecal samples remained significant after exclusion of pigs with no HEV-derived RNA in faeces, serum or organs (pigs 7 and 20 from the p429 vaccine group and pig 97 from the p429-ORF3 vaccine group). A summary of the pvalues of all statistical tests performed is presented in Table S2.
Differences in faecal HEV RNA between infected adjuvant control and vaccinated pigs were also found (t-test: p429-vaccinated pigs p < 0.1, p429-ORF3 vaccinated pigs p < 0.05; Wilcoxon test: p429-vaccinated pigs p < 0.1). However, it should be noted that exact food intake and stool volume were not monitored during the observation period. Therefore, a dilution effect cannot be excluded.
After necropsy, the viral load of organ samples (brain, kidney, liver, cranial mesenteric lymph node, spleen, gallbladder wall and bile) was determined (Method S9, Table S1). HEV-RNA was detected in the bile of three out of six animals in the infection control group, one (p429 vaccination group) and three pigs (p429-ORF3 vaccination group) in the vaccinated groups.
No viral RNA was detected in the bile of the infected adjuvant control group. However, one pig in this group showed a low viral load in the liver.
The Wilcoxon Rank Sum Test revealed significant differences in viral load in liver (p < 0.1) and bile (p < 0.05) between the infected adjuvant control and both vaccine groups.
## 4 | Discussion
Although hepatitis E is a globally prevalent virus and one of the most common causes of acute hepatitis, there is currently no globally licensed vaccine on the market. Hecolin, the only registered vaccine, is available only in China and Pakistan with proven efficiency against HEV-4 [12,21]. Hecolin was evaluated in numerous studies, demonstrating efficacy against HEV-1 and rabbit HEV strains [22][23][24]. However, data on its effectiveness against HEV-3 are limited [14]. A recent publication demonstrated that Hecolin and a similar HEV-3-based vaccine (p239 Riems) showed only limited protection against HEV-3 in pigs. Notably, both vaccines induced antibodies capable of neutralising HEV-3 isolates in vitro [15,25].
Therefore, there is a significant gap in the availability of an effective vaccine against HEV-3. The aim of this study was to develop new vaccine candidates against HEV-3 by investigating alternative vaccine designs using a coordinated approach involving cell culture assays and animal models. We assessed these new candidates in a small animal model (rabbits), characterised resulting antibodies in vitro and evaluated selected candidates in a pig infection model.
In total, seven new vaccine candidates were produced and evaluated using this approach. These candidates included four recombinant protein-based vaccines and three DNA-based vaccines. Two protein-based candidates, p429 and p429-ORF3, consisted of bacterially expressed central domains of the ORF2 capsid protein, either as a single capsid protein or fused with the ORF3 protein. The ORF3 protein has been proposed as a potentially neutralising target for HEV, as this protein associates with the lipid layer in quasi-enveloped virions [26], and vaccination with the ORF3 antigen has been shown to induce partial protection against HEV-1 in rhesus macaques in a previous study [27]. Two further candidates were expressed in insect cells, which consisted of the capsid P-domain and were either non-glycosylated (pGS99) or glycosylated (pGS100). A third approach dealt with three DNA-based vaccines, which consisted of plasmid-expressed S-, M-and P-capsid domains, presented either solely or fused with the ORF3 encoding protein at the 5′ or 3′ end. PVax1-Ub universal fusion vector was used to produce 5′-ubiquitin-antigen fusion constructs, which are subjected to enhanced intracellular degradation and improved entry of its epitope peptides into the class I Major histocompatibility complex (MHC) signalling pathway [17,28]. In addition, these vaccine variants undergo host posttranslational modifications, which can potentially alter the structure and immunological efficacy of the antigen [29]. Our approach could not confirm an enhancement of the immune response, as the humoral response occurred later and antibody levels were lower in the DNA-based vaccines compared to the protein-based vaccines. However, Rodriguez et al. found that vaccination with ubiquitin-coupled DNA vaccines led to a complete abrogation of the humoral immune response in a mouse model [30].
All seven vaccine candidates were administered to individual rabbits and resulting antibodies were examined by testing the corresponding sera. The results showed a faster increase in antibody levels in the four rabbits that were immunised with a protein-based vaccine, compared to the three rabbits who received a DNA-based vaccine candidate. Antibodies raised against all vaccine candidates demonstrated binding to HEV-3 strains in the immunofluorescence staining assay against 'Kernow C1p6 G1634R' and 'pUC83-2' (Figure 1C, Figure S4) and exhibited specific, dose dependent neutralising activity against enveloped and naked HEV-3 in vitro (Figure 1D, Figure S5). Moreover, the binding strength to various epitopes derived from capsid fragments such as pGS99, p239 and p429 was assessed using an avidity ELISA (Figure S6). The resulting data demonstrated that sera from all vaccinated animals exhibited a strong binding affinity, underscoring the efficacy of the vaccines. Based on these tests, vaccines p429 and p429-ORF3 were identified as the most promising candidates (particularly due to their strong neutralising activity) and subsequently evaluated in a pig infection model. Pigs serve as one of the primary models for HEV infection [31] and are used for evaluating vaccines against HEV-3 due to their high susceptibility to hepatitis E, with successful infection already upon exposure to low viral doses of 6.5 virus particles [15,32]. The experiment design including infection procedure and sampling was based on a previously established protocol [15]. Therefore, pigs received two vaccinations with a 4-week interval and were infected with a HEV-3 strain 56 days post first vaccination. Both p429 and p429-ORF3 vaccines elicited a specific antibody response, resulting in significantly lower virus excretion in faeces and delayed onset of viral shedding compared to the non-vaccinated infection control group (p < 0.01 for both vaccines, Figure 2). However, bile samples from a few of the vaccinated pigs tested positive for the virus, indicating that both vaccines offer only partial protection against an HEV-3 infection and do not confer sterile immunity (Table S1). This is supported by the lack of a consistent increase in serum antibodies or stable antibody titers in either group. Instead, antibody titers declined after the second vaccination and during the infection. This is in contrast to a previous vaccination study with Hecolin and bacterial p239 vaccine in pigs where high antibody titers were maintained until the end of the study on Day 71 [15].
Of particular interest is the first application of ORF2/ORF3 fusion vaccines: The role of ORF3 as a target for neutralising antibodies remains uncertain. Notably, only one out of six piglets in the p429-ORF3 group developed specific antibodies against ORF3, and this response was also very weak while the immunised rabbit showed a strong antibody response. Additionally, the viral load in the faeces and the virus distribution in the tissues showed no significant difference between the two vaccinated groups. Similarly, no significant differences were observed in viremia across the groups. This might indicate a limited role for ORF3 in the immune response against HEV-3 in pigs. Therefore, the data imply that the ORF2/ORF3 fusion vaccine might not be as immunogenic in pigs as it is in rabbits, indicating possible species-dependent differences in immune responses.
Generally, this study focused exclusively on the humoral immune response, which is the major driver of protection by Hecolin as demonstrated in a large clinical Phase III trial [12,21]. Future examinations of the cellular immune response will provide valuable insights into specific T cell responses triggered by the vaccine. T cells are a critical component of the adaptive immune system and play a central role in recognising and eliminating infected cells.
This study offers a comprehensive framework for selecting new vaccine candidates against HEV-3, particular for the use in pigs. It includes both extensive in vitro characterisation and thorough in vivo testing in pig models. Vaccinating pigs, the reservoir hosts of HEV, is a promising strategy to prevent transmission within pig populations and especially from pigs to humans. As summarised by Salines et al. (2017), previous studies have reported HEV RNA prevalence rates of 0.8%-10% in pig liver samples, indicating ongoing infection in pigs at the time of slaughter [33]. Similarly, viral RNA was detected in bile samples in this study. Salines et al. identified two key factors associated with a high prevalence of HEV in slaughtered pigs: (1) the presence of intra-farm circulation of HEV and (2) the timing of infection, with later infections increasing the likelihood that pigs will remain infectious at slaughter [33]. An ideal vaccine would either reduce virus shedding below the infectious threshold to limit herd transmission, or minimise viral loads in edible tissues at slaughter; however, given the high susceptibility of pigs to HEV [2] and the fact that even a relatively low dose of HEV genomes can lead to a substantial risk of infection [3,32] the ultimate goal is to achieve sterile immunity by completely suppressing viral replication. Therefore, future studies should incorporate additional tissue samples, such as muscle tissue, to ensure HEV clearance in all edible tissues. This targeted approach aligns with the One Health concept and aims to reduce the global burden of hepatitis E.
## References
1. Horvatits, Zur Schulze Wiesch, Lütgehetmann et al. (2019) "The Clinical Perspective on Hepatitis E" *Viruses*
2. Purdy, Drexler, Meng (2022) "ICTV Virus Taxonomy Profile: Hepeviridae 2022" *Journal of General Virology*
3. Khuroo, Khuroo (2016) "Hepatitis E: An Emerging Global Disease -From Discovery Towards Control and Cure" *Journal of Viral Hepatitis*
4. Van Der Poel, Verschoor, Van Der Heide (2001) "Hepatitis E Virus Sequences in Swine Related to Sequences in Humans, the Netherlands" *Emerging Infectious Diseases*
5. Martelli, Caprioli, Zengarini (2008) "Detection of Hepatitis E Virus (HEV) in a Demographic Managed Wild Boar (Sus scrofa scrofa) Population in Italy" *Veterinary Microbiology*
6. Zahmanova, Takova, Tonova (2023) "The Re-Emergence of Hepatitis E Virus in Europe and Vaccine Development" *Viruses*
7. Faber, Willrich, Schemmerer (2018) "Hepatitis E Virus Seroprevalence, Seroincidence and Seroreversion in the German Adult Population" *Journal of Viral Hepatitis*
8. Emerson, Arankalle, Purcell (2005) "Thermal Stability of Hepatitis E Virus" *Journal of Infectious Diseases*
9. Colson, Borentain, Queyriaux (2010) "Pig Liver Sausage as a Source of Hepatitis E Virus Transmission to Humans" *Journal of Infectious Diseases*
10. Thippornchai, Leaungwutiwong, Kosoltanapiwat (2022) "Survey of Hepatitis E Virus in Pork Products and Pig Stools in Nakhon Pathom Province, Thailand" *Veterinary Medicine and Science*
11. Behrendt, Wedemeyer (2022) "Impfstoffe Gegen Hepatitis E: Wo Stehen Wir?" *Bundesgesundheitsblatt, Gesundheitsforschung, Gesundheitsschutz*
12. Zhu, Zhang, Zhang (2010) "Efficacy and Safety of a Recombinant Hepatitis E Vaccine in Healthy Adults: A Large-Scale, Randomised, Double-Blind Placebo-Controlled, Phase 3 Trial"
13. Zhang, Shih, Wu et al. (2013) "Development of the Hepatitis E Vaccine: From Bench to Field" *Seminars in Liver Disease*
14. (2024) "itis_e_ backg round_ paper_ sage_ mar24"
15. Dähnert, Aliabadi, Fast (2024) "Immunisation of Pigs With Recombinant HEV Vaccines Does Not Protect From Infection With HEV Genotype 3" *One Health*
16. Schlosser, Eiden, Vina-Rodriguez (2014) "Natural and Experimental Hepatitis E Virus Genotype 3 -Infection in European Wild Boar is Transmissible to Domestic Pigs" *Veterinary Research*
17. Hawman, Ahlén, Appelberg (2021) "A DNA-Based Vaccine Protects Against Crimean-Congo Haemorrhagic Fever Virus Disease in a Cynomolgus Macaque Model" *Nature Microbiology*
18. Krey, Alayer, Kikuti (2010) "The Disulfide Bonds in Glycoprotein E2 of Hepatitis C Virus Reveal the Tertiary Organization of the Molecule" *PLoS Pathogens*
19. Jäckel, Eiden, Balkema-Buschmann (2013) "A Novel Indirect ELISA Based on glycoProtein Gn for the Detection of IgG Antibodies Against Rift Valley Fever Virus in Small Ruminants" *Research in Veterinary Science*
20. Todt, Friesland, Moeller (2020) "Robust Hepatitis E Virus Infection and Transcriptional Response in Human Hepatocytes" *Proceedings of the National Academy of Sciences of the United States of America*
21. Huang, Zhang, Su (2024) "Long-Term Efficacy of a Recombinant Hepatitis E Vaccine in Adults: 10-Year Results From a Randomised, Double-Blind, Placebo-Controlled, Phase 3 Trial" *Lancet*
22. Li, Zhang, Li (2005) "A Bacterially Expressed Particulate Hepatitis E Vaccine: Antigenicity, Immunogenicity and Protectivity on Primates" *Vaccine*
23. Liu, Du, Wang (2014) "Management of Hepatitis E Virus (HEV) Zoonotic Transmission: Protection of Rabbits Against HEV Challenge Following Immunization With HEV 239 Vaccine" *PLoS One*
24. Zhang, Zeng, Liu (2015) "Hepatitis E Vaccine Immunization for Rabbits to Prevent Animal HEV Infection and Zoonotic Transmission" *Vaccine*
25. Wen, He, Tang (2020) "Quantitative Evaluation of Protective Antibody Response Induced by Hepatitis E Vaccine in Humans" *Nature Communications*
26. Yang, Nan (2021) "Open Reading Frame 3 Protein of Hepatitis E Virus: Multi-Function Protein With Endless Potential" *World Journal of Gastroenterology*
27. Ma, Song, Harrison (2009) "Immunogenicity and Efficacy of a Bacterially Expressed HEV ORF3 Peptide, Assessed by Experimental Infection of Primates" *Archives of Virology*
28. Boshra, Lorenzo, Rodriguez et al. (2011) "A DNA Vaccine Encoding Ubiquitinated Rift Valley Fever Virus Nucleoprotein Provides Consistent Immunity and Protects IFNAR(-/-) Mice Upon Lethal Virus Challenge" *Vaccine*
29. Ozdilek, Avci (2022) "Glycosylation as a Key Parameter in the Design of Nucleic Acid Vaccines" *Current Opinion in Structural Biology*
30. Rodriguez, Zhang, Whitton (1997) "DNA Immunization: Ubiquitination of a Viral Protein Enhances Cytotoxic T-Lymphocyte Induction and Antiviral Protection but Abrogates Antibody Induction" *Journal of Virology*
31. Kenney, Meng (2019) "Hepatitis E Virus: Animal Models and Zoonosis"
32. Dähnert, Eiden, Schlosser (2018) "High Sensitivity of Domestic Pigs to Intravenous Infection With HEV" *BMC Veterinary Research*
33. Salines, Andraud, Rose (2017) "From the Epidemiology of Hepatitis E Virus (HEV) Within the Swine Reservoir to Public Health Risk Mitigation Strategies: A Comprehensive Review" *Veterinary Research* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548436&blobtype=pdf | # A novel accessory gene product of tick-borne Dhori-Orthomyxovirus, encoded by overlooked spliced transcripts of RNA segment 6
E Bendl, G Lampo, P Chlanda, E Schnettler, G Kochs, J Dengjel, L Graf
## Abstract
Dhori virus (DHOV) (Orthomyxoviridae, genus: Thogotovirus) is a tick-borne virus with a segmented, negative-sense RNA genome. Thogotoviruses can be divided into two clades, Thogoto-like and Dhori-like viruses, which differ in their coding strategy of the segment 6 encoded matrix protein (M): Thogoto viruses translate M from spliced transcripts and a long M (ML) from unspliced transcripts. Dhori-like viruses encode their M using unspliced transcripts, whereas no splicing or additional coding capacity has been described, yet. Here, we identified spliced transcripts of segment 6 in DHOV-infec ted cells by RT-PCR and sequencing. The gene product M2-248 codes for a truncated, 186 amino acids-long N-terminal moiety of M fused to 62 C-terminal amino acids from a -1 shifted reading frame. The splicing sites and the amino acid sequences of the C-terminal M2 part are conserved in Dhori-like viruses. Expression of M2-248 was confirmed in infected cells by mass spectrometry and western blot analysis and was detected in purified virions. Recombinant DHOV(ΔM2) lacking M2-248 showed reduced virulence compared with recombinant wild-type in mice but no attenuation in cell culture and no effect on virion assembly and morphology. An interferon antagonistic function of M2-248, similar to THOV ML, was excluded. Thus, our data suggest that DHOV segment 6 splicing and expression of M2-248 protein affect virulence in mammals by a mechanism distinct from THOV-ML. It remains to be analyzed how M2-248 might affect viral host switch and replication in ticks. IMPORTANCE Dhori and Thogoto viruses are tick-transmitted orthomyxoviruses comprising two clades differing in the coding strategy of their matrix protein. The recent detection of Dhori-like Bourbon virus (BRBV) highlighted their zoonotic potential. M of Dhori-like viruses is translated from a collinear transcript of the smallest of the six genomic ssRNA segments. Thogoto-like viruses express M from spliced transcripts, and unspliced transcripts encode an extended ML protein. No analogous splicing event or additional gene product has been characterized for Dhori-like viruses, yet. Here, we describe the hitherto overlooked splicing of segment 6 transcripts of Dhori virus. The 5'-part of the processed transcript is collinear with the M-ORF, whereas splicing shifts it into a -1 frame. The predicted product M2-248 was detected in infected cells. Recombi nant Dhori virus lacking M2 was attenuated in vivo, whereas replication in mammalian cells was not impaired, suggesting a modulatory function of M2 in in vivo-specific, cellular immunity-related processes. KEYWORDS orthomyxoviruses, thogotoviruses, Dhori virus, Bourbon virus, tick-borne, arbovirus, matrix protein, new accessory viral gene product T hogotoviruses with their segmented, negative sense-ssRNA genome form a separate genus within the family of Orthomyxoviridae. Each of the six segments codes for
one structural protein (1): the three subunits of the viral polymerase, PB2, PB1, PA, the viral glycoprotein (GP) inserted into the viral envelope, the nucleoprotein (NP), and the matrix protein (M). A characteristic feature of thogotoviruses is their transmission by ticks and their capacity to replicate in ticks as well as in mammals (2)(3)(4). Accordingly, the viral glycoprotein (GP) structurally resembles that of insect baculoviruses (5), enabling thogotoviruses to infect tick as well as mammalian cells. During the last years, several thogotoviruses were isolated, mostly from infected ticks collected in different parts of the world, and serological studies showed thogotovirus infections in various wild and domestic mammalian species including humans (1). Phylogenetic and serological analyses led to the classification of thogotoviruses into two clades (1,6): The Dhori-like and the Thogoto-like viruses, with the prototype species DHOV/India/1313/61 (7) and THOV/SiAr/126/72 (8).
Historic reports described five laboratory human infections with DHOV in the former Soviet Union provoking febrile illness and encephalitis (9). More recently, several tick-borne zoonotic infections with the Dhori-like Bourbon virus (BRBV) were reported in the United States with occasionally severe outcomes (10)(11)(12). The pathogenesis of DHOV was experimentally studied in mice. Intraperitoneal or subcutaneous infections of laboratory animals led to severe clinical outcomes with high viral loads in the spleen, lungs, and liver, exceeding cytokine production and death from lung and liver damage within 4-5 days (6,(13)(14)(15).
Members of the Orthomyxoviridae expand their coding capacity by utilizing host splicing machinery to produce multiple transcripts generated from certain genome segments. This process has been well described for the influenza A virus (IAV) M and NS segments (16). Accordingly, it has been shown that the M protein of Thogoto-like viruses is translated from a spliced transcript of segment 6, and the splicing event creates a stop codon to terminate the M open reading frame (ORF) (Fig. 1a). The unspliced transcripts of THOV segment 6 encode the M-ORF C-terminally elongated by 38 amino acids, called ML (17,18). The ML gene product, and especially its C-terminal extension, acts as a viral interferon (IFN) antagonist (17,19). However, previous studies established that Dhori-like viruses express their M from an unspliced transcript of segment 6 (20,21) and that DHOV segment 6 does not undergo splicing. Segment 6 of DHOV contains 961 nucleotides (nt) and codes for an M protein of 270 amino acids, called M-270. In DHOV/India/1313/61 and Oz virus (OzV)-infected cells, only a single transcript species corresponding in size to the genomic RNA of segment 6 was previously described (20,21). However, in their characterization of the DHOV matrix protein, Clay and Fuller described subgenomic transcripts of segment 6 in low abundance detected in overexposed Northern blot analyses and proposed a second ORF in segment 6. This 327 nucleotides-long ORF in a -1 frame partly overlapped with the M-ORF and had a coding capacity of 141 amino acids. The expression of this putative second gene product could not be verified (20).
This prompted us to reevaluate the coding capacity of DHOV segment 6 by analyzing viral transcripts from DHOV/India/1313/61 infected cells. Interestingly, we detected mRNA processing by two independent splicing events. The distal splicing resulted in a -1-frame shift compared with the M-ORF. Accordingly, the removal of the second intron truncated the M-ORF to its N-terminal 186 amino acids before shifting into the -1 frame that elongated the polypeptide by a unique C-terminal extension of 62 amino acids. This second gene product of segment 6, referred to as M2-248 from here on, is dispensable for viral replication and lacks IFN-antagonistic activity in vitro. However, recombinant DHOV(ΔM2) lacking M2-248 expression showed an attenuated phenotype in mice, indicating that M2 is a significant virulence factor.
## RESULTS
## Detection of subgenomic transcripts of segment 6 in DHOV-infected cells
We re-evaluated the transcripts of segment 6 isolated from DHOV-infected human hepatoma Huh7 cells by RT-PCR using random hexamers and an oligo dT primer mix for cDNA synthesis and a segment 6 specific primer pair amplifying the near full-length segment 6 in an amplicon of 957 nt. The agarose gel analysis of the PCR products showed the expected prominent band corresponding to the full-length transcript (mRNA1), but some additional smaller bands (mRNA2-4) down to about 300 nt in length (Fig. 1b), indicating spliced products of mRNA1. Visible bands were cut out from the gel and subjected to Sanger sequencing, which returned four cDNAs showing similarities to DHOV segment 6 predominantly in the 5´-and 3´-end sequences (Fig. 1c). Additional bands visible in the agarose gel returned inconclusive sequencing data and might represent splicing intermediates or unspecific amplification of other viral RNAs due to the high similarity of the 3'-and 5'-non-coding regions to which the primers partly bind. mRNA1 was verified as the full-length transcript of segment 6, and the sequences of mRNA2, 3, and 4 partially overlapped with mRNA1 but were reduced in length by two distinct deletions: mRNA3 and mRNA4 showed a 465 nt deletion from nts 53 to 517, and mRNA2 and mRNA4 a 161 nt deletion from nts 590 to 750 (Fig. 1c andd). Accordingly, the predicted ORF of mRNA1 codes for the 270 amino acids (aa) of the matrix protein (from here on also referred to as M-270) and mRNA3 for a truncated version of M lacking 155 aa (M-115). Removal of the second intron resulted in a -1 shift of the M-ORF into the new, 189 nt-long M2 reading frame coding for 63 codons (Fig. 1c). Single spliced mRNA2 codes for a truncated M protein of 186 N-terminal aa continued by 62 M2-specific aa (hereafter M2-248). The double-spliced mRNA4 codes for a predicted peptide of 31 N-terminal aa of the M-ORF continued by the 62 M2-specific aa (M2-93).
An alignment of the segment 6 nucleotide sequences publicly available for Dhori-like viruses showed an overall high similarity of >63%, including a nearly complete conserva tion around the two putative splice donor and acceptor sites (Fig. 1d).
To investigate whether splicing of DHOV segment 6 also occurs not only in mam malian but also in tick cells, we chose two tick cell lines from Hyalomma anatolicum (HAE/CTVM9) and Rhipicephalus appendiculatus (RAE/CTVM1) because both Ixodid tick genera had been described to host Dhori-like viruses (22, 23; reviewed in reference 1). We infected both cell cultures with DHOV/1313/16 at a moi of 0.1 and harvested RNA from the infected cells at 8 dpi. A conventional RT-PCR analysis targeting segment 5 demonstrated the presence of viral RNA in the cells (Fig. S1a). A segment 6 spanning RT-PCR (as in Fig. 1b) demonstrated the accumulation of all expected transcripts of segment 6 comparable with the transcripts in Huh7 cells infected for 16 h: unspliced mRNA1, intron 1-spliced mRNA3, intron 2 spliced mRNA2, and double spliced mRNA4. Finally, the splicing of segment 6 transcripts in the infected HAE cells was confirmed by next-generation deep sequencing of the viral RNAs (Fig. S1b).
## Detection of M2 protein expression in DHOV-infected cells
To get the first evidence for the expression of the splice products containing the M2-specific region, we analyzed lysates of DHOV-infected human lung epithelial A549 cells by mass spectrometry (MS). The MS/MS analysis detected peptides spanning the entire M-270 protein but also peptides corresponding to the unique, C-terminal part of the M2-specific region (Fig. 2a, yellow and red small bars, respectively). Intriguingly, the MS analysis also detected two peptides that consist of amino acid sequences from exon 1 and exon 2 (gray small bars), suggesting the translation of spliced intron 1 transcripts. We quantified the relative levels of peptides derived from host and viral proteins based on iBAQ (intensity-based absolute quantification) values, confirming an abundant presence of viral matrix protein (M-270) as well as M2-containing viral gene products in infected cells (Fig. 2b).
Subsequently, we intended to prove M2 expression in infected cells using western blot analysis. Thus, polyclonal rabbit antisera were raised against the matrix protein purified from extracellular DHOV/India/1313/61 particles (anti-M) or against a 14 aa-long synthetic peptide, corresponding to the unique, C-terminal part of M2 (Fig. 2c, anti-M2 epitope). Different mammalian cell lines were infected with DHOV/India/1313/61, and the cell lysates were analyzed using gel electrophoresis and western blot. The polyclonal anti-M antiserum recognized a protein of about 29 kDa in the lysates of DHOV-infected, but not of mock-treated, cells (Fig. 2d). This protein was also observed in the lysates of 293T cells transfected with M-270 and M2-248 encoding expression plasmids (Fig. 2e). The size of about 29 kDa corresponds to the predicted molecular weight of M-270 and coincides with the molecular weight determined previously for purified DHOV M (20). Higher bands of about ~10 kDa above the predicted molecular weights of M and M2 were especially observed in transfected cells (Fig. 2e, two red asterisks). We speculate that the upper band of about 39 kDa corresponds to post-translationally modified versions of the M or M2 protein. Furthermore, the polyclonal M-specific antiserum recognized M2-248 (Fig. 2e), most likely by binding to epitopes localized in its N-terminal part that corresponds to the N-terminus of M-270 (see scheme in Fig. 1c). Accordingly, the M2-specific 63 aa of a GST fusion protein (GST-M2-63) was not recognized by the anti-M antiserum (Fig. 2e). Although the MS/MS analysis confirmed the presence of peptides derived from intron 1-spliced transcripts, we did not detect protein bands corresponding to the predicted molecular weights of M-115 or M2-93 in western blot analyses (Fig. 2d).
The M2-specific antiserum recognized M2-248 but not M-270 (Fig. 2d, lower panel). Again, the western blot analysis (Fig. 2e, lower panel) of transfected cells showed two bands of about 28 and 38 kDa, corresponding to the predicted molecular weight of M2-248 and its possibly modified version (Fig. 2e, two red asterisks). Additionally, the two antibodies recognized M-270 and M2-248 in lysates of different mammalian cells infected with three different isolates of DHOV (Fig. 2f andg). However, an M2-248 protein was not detected in the lysates of Dhori-like OzV and BRBV-infected cells, although the expression of M-270 could be detected by the anti-M serum. The dissimilarity of the M2 amino acid sequences of the two Dhori-like viruses when compared with the DHOV sequences in the antigenic M2 peptide region (Fig. 2c) most likely accounts for the lack of M2-specific antibody recognition. This antigenic heterogeneity between DHOV and Dhori-like viruses was also evident for the viral NP protein (Fig. 2g).
## Structure prediction and analysis of M2-248
AlphaFold 3 was used (24) to predict the structures of DHOV M-270 and M2-248 as well as THOV M and ML (Fig. S2a). Overall, the predicted DHOV M-270 as well as the THOV M structures show a two-domain architecture similar to the M1 protein of influenza A virus (25,26). The N-terminal domains (NTD) and a more variable fold of the C-terminal domain (CTD) were predicted to be formed by alpha-helices linked by a disordered region. It is noteworthy that the predicted structure of the N-terminal domain of THOV M matches the structure determined using X-ray crystallography (PDB: 5I5N) (27). Prediction confidence of the NTD (aa 13-140) of DHOV M-270 and M2-248 and CTD of M-270 (aa 148-270) was high, whereas low prediction confidence was reported for the disordered "linker" region (aa 141-167) (Fig. S2b). However, the C-terminal domain of M2-248 (aa 148-248) was predicted with only medium-to-low confidence that increases for the predicted C-terminal α-helix (Fig. S2b). A hydrophobicity plot (Fig. S2c) of DHOV M-270 and M2-248 predicts similar biophysical properties for the N-terminal part and the linker region with a single hydrophobic stretch (aa 115-141) already described by Clay and Fuller (20). Interestingly, the C-terminal regions showed different profiles (Fig. S2c). None of the two proteins contained predicted transmembrane structures according to the DeepTMHMM v1.0 (28) prediction tool (not shown). However, both M-270 and M2-248 contain a predicted amphipathic helix at their C-termini (Fig. S2d, in gray) in addition to two putative late domain motifs (YQIL and YQLL) in M-270. These late domain motifs are conserved in the M protein sequences of all DHOV and Dhori-like sequences (not shown) and are commonly associated with the budding and release of enveloped RNA viruses (29). inactivating mutations in the splice acceptor site of intron 1 and the splice donor and acceptor sites of intron 2, and for the M2-248 coding cDNA with an inactivating mutation in the splice acceptor site of intron 1. Furthermore, expression constructs encoding GST-fusion constructs of M-270, M2-248, and M2-63 were transfected, the latter containing only the C-terminal 62 amino acids of the M2-specific region fused to GST. (panels d and e) The western blots were incubated with polyclonal rabbit antisera raised against the matrix protein purified from extracellular DHOV/India/1313/61 virions (anti-M, yellow arrowheads) or against a synthetic M2 peptide (see panel 2 c, anti-M2, red arrowheads). The red asterisk in panel (d) marks an upper band of about 45 kDa observed also in uninfected A549 cells. Two red asterisks in panel (e) mark additional upper bands of M and M2 in transfected cells. A viral NP-specific antiserum and a β-actin antibody were used as an infection control and as a loading control, respectively. (f ) Western blot analysis of mammalian cells infected with DHOV strains: India/1313/61 (India), PoTi-461, and Batken virus (BTKV). Cells were infected with moi 1 and lysed 24 hpi. (g) Western blot analysis of Vero cells infected and processed as in panel f with strains of DHOV as well as Dhori-like OzV and BRBV. mw = molecular wt marker. In summary, our molecular analyses of DHOV-infected cells confirmed the synthesis of a new viral gene product, called M2-248, that consists of the N-terminal part of the matrix protein fused to a unique M2 structure translated from a shifted reading frame caused by a splicing event.
## Generation of a recombinant DHOV lacking M2 expression (ΔM2)
To evaluate the effect of the new gene product, M2-248, on DHOV replication, we established a reverse genetic system to rescue recombinant rDHOV. For this, the six segments of DHOV strain India/1313/61 were amplified by RT-PCR, and the cDNAs are cloned into the pHW2000 ambisense vector (30). Generation of rDHOV was performed as previously described for rTHOV (19). Virions harvested from the transfected cell cultures were plaque purified, and virus stocks were prepared on BHK-21 cells.
The segment 6 coding capacity and design of three generated recombinant DHOVs is summarized in Fig. 3a. Segment 6 of rDHOV(wt) corresponds to the sequence of the parental DHOV/India/1313/61 encoding the entire M and the various M2 gene products by functional splice sites of intron 1 and intron 2. Segment 6 of rDHOV(M2stop) also encodes functional intron 1 and intron 2 splice sites, but a truncated M2-ORF by introducing a premature stop codon that terminates the M2-ORF 30 codons before its authentic stop codon, resulting in a shortened, 218 aa-long M2-218 protein, without affecting the sequence of the M protein. Segment 6 of rDHOV(ΔM2) lacks the expression of spliced mRNAs and, therefore, the expression of the protein products M2-93 and M2-248, containing the M2 region, by mutating the splice acceptor site of intron 1 as well as the splice donor and splice acceptor sites of intron 2. The splice donor site of intron 1 could not be incapacitated without resulting in amino acid changes affecting the M-ORF. To verify the expected coding capacities of the three recombinant viruses, Huh7 cells were infected with the recombinant viruses, and the expression patterns of segment 6 were analyzed using RT-PCR of the viral transcripts (Fig. 3b) and using western blot to detect the viral proteins (Fig. 3c). The gel electrophoresis pattern of the smaller, spliced transcripts appeared identical for rDHOV(wt) and rDHOV(M2stop). In contrast, rDHOV(ΔM2) lacked amplicons of intron 2 spliced transcripts (Fig. 3b). However, in rDHOV(ΔM2)-infected cells, we detected two cDNA amplicons of about 500 and 800 bp (Fig. 3b, red asterisks). Sanger sequencing of these cDNAs confirmed the lack of intron 2 splicing but indicated sporadic splice events in intron 1. This suggests that inactivation of the intron 1 splice acceptor site did not totally suppress splicing and instead led to the use of alternative splice-acceptor sites. As a control, amplicons of segment 5 of about 1,500 nt were amplified from the isolated RNAs (lower panel), confirming comparable infection efficiency. The western blot analyses of viral proteins accumulated in infected cells 24 h post-infection showed equal production of M-270 and NP proteins (Fig. 3c). However, the M2-248 band detected in the rDHOV(wt)-infected cells shifted from about 29 kDa to a reduced molecular weight of about 25 kDa in rDHOV(M2stop) infected cells and was absent in the rDHOV(ΔM2)-infected cell lysates (Fig. 3c), confirming the expected M2 expression patterns of the recombinant viruses.
Next, we analyzed the capacity of the three recombinant viruses to replicate in mammalian cell cultures. Infection of the cells with a low moi of 0.0005 for 72 h led to progeny release of about 6 × 10 7 pfu/mL for all three rDHOVs (Fig. 3d). Infection with this low moi revealed growth kinetics comparable for all three rDHOV over 72 h in Huh7, Vero, and A549 cells (Fig. 3d). To examine the intracellular localization of the viral proteins, Huh7 cells were infected with the rDHOVs at a higher moi of 2 that resulted in almost complete infection efficiency. Immunofluorescence microscopy of the fixed cell cultures at 24 hpi using specific antibodies showed nuclear staining of NP as well as M protein staining in the nuclear and the cytoplasmic compartments in almost all cells (Fig. 3e). Interestingly, the staining with the M2-specific antiserum showed a lower number of positive cells with some strong signals in the nucleus and a distinct web-like, cytoplasmic localization for the wild-type as well as for the truncated M2stop. Both M and M2 were also detected at the plasma membrane of the infected cells. Overall, our comparative Segment 5 (S5) and β-actin-specific primer sets were included as controls. (c) Western blot analysis of rDHOV-infected Huh7 cells. Cells were infected with a moi of 0.1, and whole cell lysates were obtained at 36 hpi. Viral proteins were detected using M-, M2-, and NP-specific antisera. Detection of β-actin was used as a loading control. (d) Growth kinetics of rDHOV in mammalian cell lines. Huh7 cells were infected with different MOIs for 2 h. Afterward, the cells were washed with PBS, and the inoculum was replaced by a fresh medium. Progeny virus in the supernatants was quantified by plaque assay at 72 hpi.
(Continued on next page) analysis of rDHOV(ΔM2) with rDHOV(wt) suggests that M2-248 expression does not grossly affect viral replication in mammalian cell cultures.
## M2-248 interacts with M and is incorporated into released virions
Because of the structural similarity of M2-248 and M-270 (Fig. S2), we asked whether the former might be incorporated into viral particles as a structural protein. To investigate this, extracellular virus particles from the supernatants of DHOV/India/1313/61-infected BHK-21 cells were concentrated through a glycerol cushion and analyzed using western blot after a sucrose gradient fractionation. The analysis confirmed the presence of NP and M-270 as canonical structural proteins and, interestingly, also of the M2-248 protein as detected by the M2-specific antibody (Fig. 4a). As expected, extracellular virions of rDHOV(ΔM2)-infected cells lacked the M2-248 signal (Fig. 4a). As a control for contamina tion of our virion preparation with cellular components, we stained the western blot for cellular tubulin that was present in the mock-infected whole cell lysate (WCL) but not in the purified virions of the parental DHOV(wt). Because the virion preparations of rDHOV contained residual cellular debris as indicated by the detection of tubulin, the parental DHOV(wt) preparation that contained no detectable amounts of tubulin was used for the subsequent experiment.
For further analysis, the purified parental DHOV(wt) virions were lysed using deter gent, and the structural components of the particles were separated by discontinuous 5%-73% glycerol density gradient centrifugation. Ten fractions were collected and analyzed for their protein content by SDS-PAGE and western blot. The analysis revealed sedimentation of NP (55 kDa) to the bottom of the gradient, most likely in the form of vRNA-associated protein forming dense vRNP structures. In the middle of the gradient, a signal around 68 kDa might reflect the sedimentation of the viral glycoprotein, and in the most upper fractions, around 34 kDa, the viral M-270 protein (Fig. 4b). Interestingly, the western blot demonstrated that M2-248 is also present in these M-270-positive fractions (Fig. 4b), suggesting that M2 is a structural viral component and might co-sediment in the glycerol gradient due to its interaction with the M protein. As mentioned before, we were not able to detect protein products of intron 1-spliced transcripts of segment 6 in this analysis.
A common feature of orthomyxoviral matrix proteins is their capacity to oligomerize, a requirement for the formation of the matrix layer beneath the envelope of orthomyxo viruses (25,27). Therefore, we used co-immunoprecipitation to check whether M-248 can interact with M-270 in transfected cells. For this, the cDNAs coding for M-270 and M2-248 were C-terminally fused to the cDNA encoding a FLAG-or an HA-tag, respectively. Transfected 293T cells co-expressing these constructs were used for co-precipitation using anti-FLAG IgG-coated agarose beads. Accordingly, FLAG-tagged M-270 and M2-248 proteins were precipitated from the transfected cell lysates together with HA-tagged M-270 and M2-248 (Fig. 4c). As specificity controls, we used FLAG-tagged viral NP and chloramphenicol acetyltransferase (CAT) as well as HA-tagged CAT. These control proteins showed no precipitation of HA-tagged M-270 or M2-248, confirming a specific interaction between M-270 and M2-248.
The presence of M2-248 in extracellular particles implies that it might be involved in virion assembly and budding. To test this hypothesis, we established a system to reconstitute infectious virus-like particles (VLPs). For the expression of the recombinant proteins, the ORFs coding for the structural viral proteins, PA, PB1, PB2, GP, NP, and M(wt), were cloned into eukaryotic expression vectors. The M(wt) cDNA encodes the ORF of the authentic M gene with functional splice sites of intron 1 and 2, resulting in M-270 as well as M2-248 expression. Splicing of transcripts of the M-270 and M2-248-ORFs was prevented by the inactivation of the respective splice sites as mentioned above for the construction of the rDHOV rescue plasmids (Fig. 3a). In addition, the cDNA of an artificial genomic segment encoding firefly luciferase flanked by the non-coding regions of segment 5 was cloned into an RNA-polymerase I expression plasmid, resulting in the expression of a genomic vRNA segment of about 1,700 nt that can be transcribed by the reconstituted viral polymerase complex. Co-transfection of the complete set of viral structural proteins and the firefly luciferase vRNA into 293T donor cells resulted in the production of firefly luciferase, indicating the reconstitution of vRNPs with a functional viral polymerase complex (Fig. 4d, donor cells). Expression of the recombinant proteins in the transfected donor cells was confirmed by western blot analysis with M-, M2-, and NPspecific antibodies (Fig. 4d). The supernatants of these transfected donor cells were transferred to BHK-21 indicator cells to test for infectivity of the released VLPs by detecting the expression of firefly luciferase (Fig. 4d, upper panel). Omission of either the GP or the M(wt) expression plasmids did not grossly affect the expression of firefly luciferase in the donor cells but did not support the formation of VLPs and firefly luciferase activity in the indicator cells (Fig. 4d). Replacing M(wt) by the M-270 expression plasmid also supported VLP formation, resulting in firefly luciferase activity in the indicator cells, whereas M2-248 did not (Fig. 4d). Co-transfection of the M-270 plasmid with increasing amounts of M2-248-encoding plasmid had no effect on the reconstituted polymerase activity in the donor cells or on VLP formation detected in the indicator cells. The experiment depicted in Fig. 4e confirmed these results by showing the close dependency of VLP formation on GP and M-270 but not M2-248 expression. Likewise, attempts to rescue rDHOV with an M2-248 encoding bidirectional vector lacking intron 2, and therefore, M-270 expression were not successful (not shown).
## Structural analysis of budding and released DHOVs
To investigate whether M2-248, as a structural protein, might affect the budding phenotype and particle morphology, we performed scanning electron microscopy (SEM) and cryo-electron tomography (cryo-ET). Huh7 cells were infected with recombinant rDHOV(wt), rDHOV(M2stop), and rDHOV(ΔM2) for 24 h, fixed, and subjected to SEM sample preparation. The SEM images from the infected cells showed elongated budding profiles for all three viruses, indicating that M2 expression does not alter the budding process (Fig. 5a). The budding profiles can be distinguished from other physiological protrusions also observed in mock-treated controls. Those have a wider diameter and are shorter than budding virions and likely represent filopodia and microvilli (Fig. 5a). Furthermore, supernatants of infected BHK-21 cells were harvested and analyzed by cryo-ET. As reported previously for parental DHOV (6), the recombinant rDHOV(wt) forms spherical (120 nm in diameter) and long filamentous virions of up to 1,300 nm in length (Fig. 5b andc). The viral envelope with the glycoprotein spikes and the matrix layer beneath the viral membrane are clearly visible in the images. However, no major differences in the virions' morphology could be detected when comparing rDHOV(wt) with rDHOV(M2stop) and rDHOV(ΔM2) (Fig. 5b andc). Based on this limited data set (n = 40 for each recombinant virus), an axis ratio analysis suggests a marginally more elongated median axis ratio for rDHOV(ΔM2) when compared with rDHOV(wt) and rDHOV(M2stop). However, this difference was not significant (Fig. 5d). In summary, our data show that the newly described M2-248 viral protein can interact with the viral M-270 protein and is a structural component present in extracellular virions. However, despite sharing a large, 185 aa long portion of identical sequence with M-270, M2-248 does not undertake the function of a viral matrix protein. It neither affected viral polymerase activity, assembly, or budding of virions nor the morphology of extracellular particles.
## rDHOV(ΔM2) is attenuated in vivo
To determine the impact of M2 expression on the virulence of rDHOV, we used labora tory mice as an in vivo model. Previous experiments with DHOV/India/1313/62 showed high virulence of this virus strain upon intraperitoneal (i.p.) infection with high virus loads in the liver, lung, and spleen and a fatal outcome within 5-6 days post-infection (dpi) (6,14). Therefore, we infected C57BL/6 mice i.p. with a dose of 40 pfu of the rDHOVs (Fig. 6a through c) and determined the body weight and clinical score (Fig. S3j) daily, over 14 days. In case of severe disease symptoms, the animals were euthanized. Upon infection with rDHOV(wt) and rDHOV(M2stop), the animals developed severe symptoms including a decline in their body weight starting at 7-8 dpi when three and two of the 10 infected animals had to be euthanized, respectively (Fig. 6a through c). In parallel, animals were infected with 4 or 400 pfu (Fig. S3a to c, g to i). Infections with the reduced dose of rDHOV(wt) and DHOV(M2stop) resulted in the transient manifestation of disease symptoms, but only two out of 10 animals had to be euthanized around 8 dpi. However, infections with a high dose of 400 pfu resulted in higher mortality: six out of 10 infected animals for rDHOV(wt) as well as rDHOV(M2stop) succumbed to the infection. Using these results, we calculated a mouse lethal dose 50 (LD 50 ) for rDHOV(wt) of ~186 pfu and for rDHOV(M2stop) of ~241 pfu. In contrast, the rDHOV(ΔM2)-infected animals showed only mild clinical symptoms around 8 dpi, and all animals survived the infection even with the high dose of 400 pfu (Fig. 6a through c and Fig. S3a through i), indicating an LD 50 for rDHOV(ΔM2) far above 400 pfu.
In a second set of in vivo experiments, C57BL/6 mice were infected i.p. with a higher dose of 1,000 pfu but only for 4 days, to determine virus replication in different organs. As described before (6,14), we detected a systemic infection of the animals with rDHOV(wt), detecting progeny virus and viral RNA in the lung, spleen, and liver (Fig. 6d and e and Fig. S4). Infection with rDHOV(M2stop) resulted in comparable titers like wildtype in the lung, spleen, and liver. However, for rDHOV(ΔM2), we detected slightly reduced viral replication (Fig. 6d ande). Of note, intron 2-spliced transcripts were not detected in rDHOV(ΔM2)-infected animals by RT-qPCR (Fig. 6f). Viral replication was not detected in the kidney and brain of the infected animals (data not shown).
Li and colleagues reported the elevated expression of inflammatory cytokines during experimental infection of mice with a lethal dose of DHOV/India/1313/62 that reflects the fulminant, systemic disease of the animals (15). Therefore, we also monitored cytokine induction in the organs of the rDHOV-infected animals at 4 dpi by RT-qPCR analysis of the RNAs isolated from the spleen, lung, and liver. rDHOV(wt) and rDHOV(M2stop) infections induced a clear upregulation of proinflammatory cytokine expression including IFNα2, IFNβ, and IL6 in the spleens of infected animals when compared with the mock-treated animals (Fig. 6g andh). These expression levels were only slightly reduced in the organs of rDHOV(ΔM2)-infected animals (Fig. 6g andh). Interestingly, the induction of RIGI, as a type I IFN-induced gene, and of IL-10 in the spleen remained almost unchanged. Analyses of RNAs extracted from the infected lungs and livers showed only low levels of cytokine induction and no differences between the three rDHOVs (Fig. S4a andb), most likely reflecting reduced levels of viral replication in these organs. Overall, the small differences in viral replication and cytokine induction between the three rDHOVs do not reflect the strong differences in inducing severe disease in the infected animals.
To confirm the successful infection of the rDHOV(ΔM2)-inoculated animals, we harvested sera from the surviving animals at 14 dpi and tested for neutralization and general seroconversion. Plaque reduction neutralization assays showed robust neutralization of DHOV infectivity by the reconvalescent sera of rDHOV(wt) and rDHOV(ΔM2)-infected mice, with mean plaque reduction neutralization titers (PRNT 50 ) of 1:40 or 1:28, respectively (Fig. 7a, upper and right panels). Expectedly, neutralization was specific for DHOV and did not impact THOV infectivity, as quantified for a subset of sera (Fig. 7a, lower panel). DHOV-specific IgG was found in the sera of all surviving animals infected with the three viruses at each dose, confirming successful infection. IgG in the sera predominantly recognized bands corresponding in size to NP (~52 kDa) and M-270 or M2-248 (~28-30 kDa) proteins in the western blot analysis (Fig. 7b). Finally, an immunofluorescence assay of DHOV India/1313/61-infected cells showed detection of viral antigens at high dilutions (1:4,096 to 1:16,384) of the post-infectious sera (Fig. 7c).
## Lack of IFN-antagonistic activity of DHOV M2-248
Like DHOV M2-248, THOV encodes an additional gene product on the viral segment 6. This THOV ML protein was characterized as a potent type I IFN antagonist (17,19), and recombinant rTHOV-ML-, lacking this antagonistic activity, was found to be attenuated in vivo (31). Therefore, we hypothesized that the M2-248 gene product encoded by DHOV might also interfere with type I IFN induction or signaling. To get an idea of the potential IFN-antagonistic activity of DHOV M2-248, IFNβ promoter activation was determined in a luciferase reporter assay. To this end, 293T cells were co-transfected with a plasmid coding for firefly luciferase under the control of the human IFNβ promoter and expres sion constructs for DHOV NP, M-270, and M2-248 as well as the THOV M or ML protein. To activate the IFNβ promoter, the cells were infected with Sendai virus (SeV) or cotransfected with a constitutively active N-terminal fragment of RIG-I or MDA5, two sensors of intracellular viral RNA, leading to type I IFN induction. Upon stimulation, the IFNβ promoter was highly activated, resulting in elevated expression of firefly luciferase in the cells transfected with a CAT plasmid as an internal control (Fig. 8a andb). Cotransfection of DHOV NP reduced IFNβ promoter activation by 40%-50%, whereas THOV ML showed a strong effect as reported previously (17). However, the expression of DHOV M2-248 reduced the reporter gene expression by only 40%-50% and was not specifically different in comparison to the co-expressed M-270 or NP (Fig. 8a andb). Furthermore, we evaluated the effect of M2-248 on type I IFN signaling by using an IFN-sensitive, murine Mx1 promoter reporter construct. Upon treatment with IFNα, the reporter expression was stimulated in the presence of the CAT control and was suppressed by co-expressed THOV ML protein (Fig. 8c). However, the increase of reporter activity upon IFNα treat ment was not reduced by co-expressed M2-248 or the M-270 control (Fig. 8c).
To evaluate the effect of M2-248 on IFN induction in infected cells, A549 cells were infected with the recombinant DHOV either expressing M2-248, that is, rDHOV(wt), or lacking M2-248 expression, that is, rDHOV(ΔM2) for 24 h. Detection of viral NP expression confirmed comparable replication of the two viruses (Fig. 8d). rDHOV(wt) infection led to the upregulation of IFNβ transcripts, as reported previously (32). Infection with rDHOV(ΔM2) did only slightly increase endogenous IFNβ expression (Fig. 8d). As a control, we used recombinant THOV either expressing or lacking ML (17). As expected, A549 cells infected with rTHOV(ML-) showed elevated induction of IFNβ expression when compared with rTHOV(wt) infection (Fig. 8d). In summary, M2-248 lacks a significant effect on IFN induction or signaling in transfected or infected cell cultures matching the lack of in vivo effects of M2-248 deletion on the cytokine induction in rDHOV-infected animals (Fig. 6g andh; Fig. S4).
## DISCUSSION
Most animal viruses encode accessory proteins that support intracellular viral replication or act as viral antagonists of the innate or adaptive host defense (33). An example of such an accessory viral protein within the Thogotovirus genus is the ML protein of the Thogoto-like virus clade. THOV and also Jos virus were shown to express viral M protein from spliced transcripts of segment 6, whereas unprocessed transcripts code for an elongated M protein called ML (17,18). In-depth characterization of the M and ML gene products of THOV revealed the matrix protein function of the M protein for virus assembly and budding and the IFN-antagonistic function of the ML protein (19,34). However, Dhori-like thogotoviruses seemed to possess a fundamentally different coding strategy for segment 6 in that the M protein is expressed from the collinear, unprocessed transcripts of segment 6, and splicing of these transcripts had not been reported (1). In detail, Clay and Fuller reported the expression of a 270 aa long M protein encoded by the unprocessed transcripts of the prototype strain DHOV/India/1313/61 (20). In the present study, we show that the DHOV M protein functions as an orthomyxoviral matrix protein in that it is found as a prevalent, structural component of purified extracellular particles of DHOV-infected cells and supports the viral budding process when co-expressed with the viral glycoprotein in a cell culture-based VLP system. Thereby, the DHOV M protein functions like the M protein of THOV and the M1 protein of IAV. Despite only low sequence similarities between 15% and 25% with the matrix proteins of IAV and THOV, a structural prediction of the DHOV M protein shows a two-domain structure very similar to the other orthomyxoviral M proteins (25,27).
Clay and Fuller already described an additional ORF encoded by segment 6, overlap ping with the M-ORF and spanning 141 amino acids in the -1 frame (20). However, in their study, neither subgenomic transcripts of segment 6 nor a second gene product could be detected in DHOV-infected cells. The results of the present study show that transcripts of DHOV segment 6 are spliced in a distinct double-intron pattern, that is, to our knowledge, unique among orthomyxoviruses, in a process that occurs in mammalian as well as tick cells. Furthermore, we demonstrate that splicing of intron 2 on segment 6 transcripts causes the shift into the -1 frame, leading to the synthesis of a novel viral protein, M2-248. The presence of this protein, consisting of the unique M2 region fused to the N-terminal M protein moiety, was confirmed in a range of infected mammalian cells. A putative gene product of double-spliced transcripts (M2-93) was not observed despite the relative abundance of this mRNA species on the amplicon level. A polypep tide of this size might be rapidly degraded in infected cells and thus not detectable in western blot analyses.
Our newly described splicing of segment 6 explains why Clay and Fuller failed to detect the expression of an M2-containing protein because the antigenic peptide sequence used to produce M2-specific antibodies (20) was localized within the intron 2 sequence that is absent in M2-248. A search of public databases for sequences related to the unique M2 region of M2-248 revealed no significant amino acid homology to other viral or cellular polypeptides. However, the M2 sequence is quite conserved among members of the Dhori-like clade, arguing for a significant role in DHOV replication.
Purification of DHOV particles and co-sedimentation of the virion structural proteins revealed that M2 is a structural component of extracellular particles and can interact with the M protein. However, our VLP assay system showed that M2-248 alone or in combina tion with M-270 was not able to support the budding of infectious particles, although M2-248 has the first 187 amino acids in common with the M-270 protein and shows a similar 3D fold in its N-terminal part. This is reminiscent of the ML protein of THOV that consists of the complete viral M protein fused to additional, C-terminal 38 amino acids, unique for ML (17). Similarly, ML was found in viral particles together with M but did not support virus budding in a VLP assay (34). Furthermore, the growth of a recombinant DHOV lacking M2-248, rDHOV(ΔM2), was not attenuated in cell culture, and virion morphology was not altered when compared with rDHOV(wt). Overall, this argues against a crucial function of M2-248 for viral assembly, contrasting the function of the M2 protein of IAV, which assembles a membrane-spanning homotetrameric ion channel critical for virus entry and release (35). Accordingly, M2-248 does not show any sequence similarity to the IAV M2 protein, and the AlphaFold prediction of the M2-248 structure did not show a hydrophobic pattern that could serve as a transmembrane domain.
The function of M2-248 could also be discussed in its property as an M-270-like protein, which is missing some of the properties of the canonical matrix protein, such as the putative viral late domains that were predicted for the M-270 C-terminal region for the recruitment of components of the ESCRT-complex. Thereby, M2-248 could serve as a "decoy" matrix protein that shares some functions with M-270 but does not partake in the recruitment of the ESCRT-complex, thereby fine-tuning the budding process.
The strong attenuation of the rDHOV(ΔM2) virulence in vivo was unexpected because viral polymerase reconstitution assays, VLP assays analyzing the production of infec tious viral particles, and growth kinetics of M2-defective rDHOV(ΔM2) in cell culture revealed that M2-248, and canonical splicing of DHOV segment 6, is without a signif icant influence on viral replication. Furthermore, infection of mice with rDHOV(ΔM2) revealed only slight differences in tissue tropism and viral replication in different organs when compared with rDHOV(wt)-infected animals. Interestingly, the virulence of rDHOV(M2stop), a variant encoding a C-terminally truncated M2 due to an artificial, premature stop codon in the M2-ORF, without affecting splicing of segment 6, was unaltered in virulence when compared with rDHOV(wt). This strong virulence of DHOV(M2stop) suggests that either the truncated M2 sequence is sufficient for the function of M2 or that the intron 2 splicing event itself is crucial for DHOV to cause severe disease symptoms.
It has been shown that infection with DHOV causes a systemic spread of the virus and a severe fulminant fatal illness in mice (6,13,14) with strong cytokine induction, resembling a "cytokine storm" known from infections with highly pathogenic IAVs (15). Therefore, we expected the reason for the attenuated virulence of rDHOV(ΔM2) to be due to a reduced induction of inflammatory cytokines. However, analysis of cytokine expression in different organs of the infected animals did not indicate obvious differences that could account for the differences in the course of the infections. Additionally, we detected no clear effect of M2-248 expression on IFN induction or IFN signaling in transfected cells. Only in A549 cells did we detect slightly increased levels of IFNβ induction by rDHOV(ΔM2) infection; however, these differences were not significant and not matching the effect observed for THOV ML-.
In the light of the recent emergence of zoonotic Dhori-like thogotoviruses, such as Bourbon virus (BRBV) in the USA and Oz virus in Japan, in vivo-attenuated viruses analogous to rDHOV(ΔM2) might serve as a blueprint for the development of live-atte nuated vaccine candidates since the mutant virus replicated readily in cell culture and evoked a robust production of neutralizing antibodies in mice.
In summary, we identified a new gene product of DHOVs, called M2-248, that is encoded by a spliced transcript of segment 6. The splicing process causes a shift of the M-ORF into a -1 frame coding for a new C-terminal appendix to the truncated M sequence.
Despite animal experiments indicating a role of M2 in DHOV virulence in vivo, our assay systems did not provide conclusive evidence for the function of this newly identified gene product. However, given the evolutionary pressure to conserve only functional proteins, especially within DHOV's compact genome and considering the presence of a sophisticated splicing mechanism, future studies will likely uncover the biological significance of M2-248 for DHOV replication and for the switch between mammalian and tick hosts.
## MATERIALS AND METHODS
## Biosafety
All work with thogotoviruses was performed under biosafety level (BSL) 2 conditions, except for the human isolate of BRBV that was handled under BSL3 conditions. Generation of recombinant rDHOV and the manipulation of segment 6 of the viral genome were approved by the authorities of Baden-Wuertemberg, Germany (Regierung spraesidium Tuebingen, permit UNI.FRK.05.22-89/05.16-26/05. .
## Cell lines, tick cells, and IFN treatment
Human lung epithelial A549 cells (ATCC CCL-185), human embryonic kidney HEK-293T cells (ATCC CRL-3216), human hepatoma Huh7 cells (36), Syrian golden hamster kidney cells BHK-21 (ATCC CCL-10), and African green monkey kidney Vero cells (ATCC CCL-81) were cultivated in Dulbecco's Modified Eagle Medium (DMEM, Gibco; 41966-029) supplemented with 5%-10% fetal calf serum (FCS) and antibiotics (100 units/mL of penicillin and 100 µg/mL of streptomycin) at 37 ˚C and 5% CO 2 . 293T cells were treated with recombinant human IFN-α2a (PBL assay science). Tick cell cultures. HAE/CTVM9 from Hyalomma anatolicum anatolicum (37) and RAE/ CTVM1 from Rhipicephalus appendiculatus (38) (kindly provided by L. Bell-Sakyi, Tick cell biobank, University of Liverpool, UK) were propagated either in L-15/MEM or L15 medium, respectively, each supplemented with 20% FCS and 10% Tryptose phosphate broth. Cells were grown at 33°C (HAE) and 28°C (RAE) without additional CO 2 as described previously (38). For infection, about 10 5 cells were seeded in flat-sided tubes. The medium was almost completely removed, and the virus inoculum of 10 4 pfu was supplied in a small volume of used medium for 2 h. Then, the infection medium was replaced by a mixture of conditioned and fresh medium. At 8 dpi, the cells and culture supernatants were collected and used for RNA isolation.
## Viruses
For the present study, we used DHOV/India/1313/61, kindly provided by Fred J. Fuller (7), Batken virus (BTKV, strain LEIV306K), kindly provided by Robert E. Shope (39), and PoTi461, kindly provided by Armindo R. Filipe (40), as well as the Dhori-like Bourbon virus (BRBV strain Kansas, NR-50132, ATCC VR-1842), kindly provided by Amy J. Lambert and Brandy Russell (11), and Oz virus (OzV, isolate number 264.1), kindly provided by Kyoko Sawabe (21). Sendai virus (SeV) strain Cantell (41) was used for IFN induction experiments. Virus stocks were produced on Vero or BHK-21 cells, and viral titers were determined by plaque assay on Vero cells as described previously (6).
For virus infection experiments including growth kinetics, the cells were seeded to 90% confluency overnight in 6-well cavities and infected with the respective virus diluted in 500 µL OptiMEM (Gibco, 11058-021) for 2 h at 37°C and 5% CO 2 . Afterward, the cells were washed three times with PBS and incubated with DMEM containing 2% FCS, 20 mM HEPES, and 0,1% NaHCO 3 . Viral titers were determined by plaque assay on Vero cells as described previously (6).
Inactivation of DHOV was performed by UV irradiation (two times 900 µJ/cm 2 ; Ultra.LUM, UVC-515 Multilinker) of infectious cell supernatants on ice. Virus inactivation was confirmed by direct plaque assay of the treated supernatants or by incubation of BHK-21 cell cultures with the treated material for 6 days followed by plaque assay on Vero cells.
## Purification of extracellular DHOV virions
BHK-21 cells were infected with a moi of 0.001 of DHOV/India/1313/61 for 48 h. The rDHOV virions were propagated on Huh7 cells for 60 h. The cell supernatants were collected and centrifuged at 2,000 × g for 30 min at 8°C. Then, the supernatants were subjected to ultracentrifugation through a 30% glycerol cushion in PBS at 100,000 × g (SW32, Beckman Coulter) for 90 min at 8°C. The pellets were resuspended in a small volume of PBS and subjected to a discontinuous 30%-60% sucrose gradient in PBS with centrifugation (SW41) at 100,000 × g for 90 min at 8°C. The virion-containing band at the interface between 30 and 40% sucrose was aspirated, diluted with PBS, and sedimented through a 30% glycerol cushion in PBS (TLA55) at 100,000 × g for 40 min at 8°C. The resulting pellet was resuspended in a small volume of PBS and analyzed using SDS-PAGE and western blot.
## Virion lysis and glycerol gradient centrifugation
The resuspended pellet from the TLA55 centrifugation, about 3 × 10 7 pfu in 500 µL, was mixed 1:1 with 2× lysis buffer: 100 mM Tris (7.5); 200 mM NaCl; 10% glycerol; 10 mM MgCl 2 ; 2 mM DTT; 1% NP40; 2% Triton X-100; 20 mg/mL Lysolecithin (Sigma-Aldrich L5254); and protease inhibitor cocktail (cOmplete, Merck, Darmstadt, Germany) and incubated for 20 min at 30°C. Then, the suspension was cleared by centrifugation at 5,000 × g for 20 min at 8°C. The supernatant was applied on top of a discontinuous 33%-73% glycerol gradient: 1 mL steps of 33%, 43%, 52%, and 73% glycerol in 50 mM Tris (7.5), 150 mM NaCl, 1 mM MgCl 2 , and 1 mM DT and centrifuged at 135,000 × g (SW55) for 5 h at 8°C. Then, the tube was dropped out in 10 fractions of 500 µL, with fraction number one at the bottom (73%) and fraction number 10 at the top (5%). The single fractions of the gradient were analyzed by SDS-PAGE and western blot using DHOV-NP-, M-, and M2-specific antisera and a monoclonal anti-β-tubulin antibody as a marker of cellular protein contaminations.
## Cloning of viral cDNAs and expression of recombinant proteins
To clone the cDNAs of the viral genome segments, Vero cells were infected with DHOV/ India/1313/61 (moi of 1) for 24 h. RNA was isolated using the NucleoSpin RNA Kit (Macherey-Nagel, 740955.50) according to the manufacturer's protocol. Total RNA (1 µg) was reverse transcribed using the QuantiTect Reverse Transcription kit (Qiagen). The ORFs of DHOV PB1, PB2, PA, GP, NP, and M were amplified by PCR using KOD hot start polymerase (Sigma-Aldrich) and specific primer pairs (Table S1). For cloning the different splice variants of segment 6, the amplicons were separated by agarose gel electrophoresis, and the bands were purified according to their different sizes using the Zymoclean™ Gel DNA Recovery Kit (Zymo Research). The isolated amplicons were digested and ligated into the digested and dephosphorylated pCAGGS expression vector (42) via T4 DNA ligase (Thermo Fisher Scientific). The sequences of all cloned cDNAs were confirmed by Sanger sequencing.
For cloning of the rescue plasmids of DHOV/India/1313/61, the cDNAs encoding the six genomic segments were amplified using primers complementary to the individual non-coding regions of the segments (Table S1) and cloned into ambisense pHW2000 vector, kindly provided by R. G. Webster (30).
For site-directed in vitro mutagenesis, a two-step PCR was performed, in which the target gene was first amplified in two parts with internal primers harboring the desired nucleotide substitutions. These two parts of the insert were used as templates in a subsequent second PCR to generate the full-length target sequences. The inactivating mutations of the splice donor and acceptor sites of segment 6 were introduced using specific primer pairs (Table S1): Splice acceptor site of intron 1 (genomic nt position A516C; primer pair 2724/2725), the splice donor site of intron 2 (genomic nt position C591A + G594A; primer pair 2718/2719), and splice acceptor site of intron 2 (genomic nt position G750A; primer pair 2720/2721), yielding pHW-Seg6(ΔM2). A premature stop codon was introduced into the M2-ORF at genomic position CAC846-848TAG (corresponding to the aa change in M2-248 of E218*) of the viral cDNA to yield pHW-Seg6(M2stop), using primer pair 2716/2717. For the corresponding protein expression vectors, the mutated ORFs were amplified and cloned into the pCAGGS vector.
For the expression of recombinant viral proteins, adherent, nearly confluent 293T cells were transfected with plasmid DNA and jetPEI (Polyplus) transfection reagent according to the manufacturer's instructions. For western blot analysis, the cells were lysed at 24-48 h post-transfection, as indicated.
## Generation of recombinant DHOV
rDHOV(wt), rDHOV(M2stop), and rDHOV(ΔM2) were generated as described previously for rTHOV (19). The pHW2000 plasmids encoding PB1-, PB2-, PA-, GP-, and NP-segments (1-5) and the respective wild-type or mutated M-segment (segment 6) of DHOV/India/ 1313/61 were transfected (500 ng/each) (Lipofectamine 2000, Thermo Fisher Scientific) into a 2:1 co-culture of 293T and Vero cells (~1 × 10 6 cells per 6-well). At 72 h post-trans fection, the supernatant was harvested, and rDHOV was purified by plaque assay on Vero cells. Each recombinant virus was independently rescued twice. Virus stocks of rDHOV were produced by transferring single plaques to fresh BHK-21 cell cultures. The sequence of segment 6 in progeny viruses of the second cell culture passage and the presence of the introduced mutations were verified by RT-PCR and sequencing of the cDNAs.
## Viral polymerase reconstitution system and virus-like particles (VLPs)
To reconstitute the polymerase activity of DHOV/India/1313/61, 293T cells (~4 × 10 5 cells per 12-well) were co-transfected (JetPEI; Polyplus) with 10 ng of pCAGGS expression plasmids encoding the polymerase subunits PB2, PB1, PA, and 50 ng of NP plasmids as previously described for THOV (43). In addition, 50 ng of an artificial viral minigenome encoding firefly luciferase (FF) in negative-sense orientation flanked by the 5'-and 3′-NTRs from DHOV segment 5 (pPolI-FF) and 10 ng of a plasmid coding for a Renilla luciferase (RL) under the constitutive SV40 promoter (SV40p-RL) were added. At 24 h post-transfection, FF and RL luciferase activities were measured (Dual-luciferase reporter kit; Promega). FF luciferase activity was normalized to RL luciferase activity (FF/RL).
For the production of replication-incompetent virus-like particles (VLPs), 293T cells (~5 × 10 5 cells per 6-well) were transfected with pCAGGS expression plasmids encoding the structural proteins of DHOV/India/1313/61: 20 ng of PB2, PB1, PA, 100 ng of NP, 75 ng of pPolI-FF for the vRNA minigenome, and 5 ng of SV40p-RL as a transfection control. In addition to the components of the polymerase reconstitution system, expression plasmids encoding the viral glycoprotein, GP (50 ng), and the viral matrix proteins, M-270 (50 ng), and M2-248 (100 ng), were co-transfected. At 48 h post-transfection, 293T cell culture supernatants were harvested and cleared by centrifugation at 2,000 × g for 20 min at 4°C. The expression of the viral proteins in the transfected cell cultures was monitored using western blot analysis of the lysed 293T cells using NP-, M-, and M2-specific antibodies and anti-β-actin as a loading control. Formation of infectious VLPs was determined by transferring the cleared supernatants onto BHK-21 indicator cells. VLP formation was monitored by the detection of FF luciferase activity in the BHK-21 cells 48 h post-transfer.
## Immunoblotting and antibodies
Infected cells were lysed with a 1:1 mix of T-PER tissue protein extraction reagent and SDS sample buffer (Thermo Fisher). Cell lysates of the polymerase reconstitution assays were mixed with a 2-fold SDS sample buffer. Following full denaturation at 95°C for 5 min, the samples were separated by SDS polyacrylamide gel electrophoresis (SDS-PAGE, 10% acrylamide). The separated proteins were transferred onto a PVDF membrane (Merck). The membranes were first blocked with blocking buffer (0.1% Tween-20, 5% milk powder in PBS) for 1 h and then stained with the primary antibody for 1 h at RT, followed by incubation with secondary, fluorescent-labeled antibodies (LI-COR) for 1 h at RT as well. The antibodies were diluted in blocking buffer, and in between the staining steps, the membranes were washed three times for 10 min with washing buffer (0.1% Tween-20 in PBS). In the case of the anti-M2 staining for western blot analyses of viral particles (Fig. 4a andb), an HRP-coupled, secondary antibody (Agilent/Dako, ref. P0448) was used for higher sensitivity. Finally, the membranes were washed 4 times for 5 min with a washing buffer, and fluorescent or chemiluminescence signals were detected using the LI-COR Odyssey Imaging System (LI-COR, Lincoln, NE, USA).
Primary antibodies used were as follows: anti-SiAr126 NP (rabbit, polyclonal [44]), anti-JOSV (mouse, polyclonal [6]), anti-FLAG M2 (mouse, monoclonal, Sigma-Aldrich), anti-β-actin (rabbit, polyclonal, Abcam), and anti-β-tubulin (mouse, monoclo nal, Sigma-Aldrich). Anti-DHOV NP antiserum (rabbit, polyclonal) was raised against the purified His-tagged NP produced in Escherichia coli as described previously (44). Rabbit, polyclonal antiserum detecting THOV M and ML was raised against the purified His-tagged M protein produced in E. coli (45). To generate DHOV anti-M antiserum, the rabbits were immunized with purified viral matrix protein, isolated from extracellular viral particles. Viral proteins of lysed particles were subjected to SDS-PAGE, and upon Coomassie staining, the M protein band was cut out from the gel, resuspended in PBS, and used for immunization of rabbits (Davids Biotechnologie, Regensburg, Germany). A polyclonal anti-M2 specific rabbit serum was raised by immunization with an LPH-con jugated peptide corresponding to M2-248(194-207) (BioGenes, Berlin, Germany). The specificity of the DHOV-specific anti-NP, -M, and -M2 antisera was confirmed using western blot analysis of lysates from transfected cells expressing the recombinant viral proteins.
## Fluorescence microscopy
For immunofluorescence microscopy analysis, Huh7 cells were seeded onto coverslips and infected. The cells were fixed in paraformaldehyde (4% in PBS) for 15 min at RT and washed with PBS. Afterwards, the cells were permeabilized with Triton X-100 (0.5% in PBS) and washed again with PBS. After blocking for 1 h with blocking buffer (PBS with 1% BSA and 0,1% Tween 20), the coverslips were incubated with the primary antibody for 1 h at RT, followed by washing five times with PBS. The secondary fluorescence-labeled antibody was incubated in the dark at RT for 1 h. After washing once with PBS, the cells were stained with DAPI (4' ,6-diamidino-2-phenylindole) (0.3 mM in PBS) for 10 min, washed again three times with PBS, and mounted onto microscope slides using FluorSave (Millipore). Pictures were taken with an LSM 880 AiryScan (Carl Zeiss, Jena, Germany). Polyclonal rabbit antisera directed against DHOV NP, M, and M2 were preadsorbed to naive, fixed and permeabilized Huh7 cells for 24 h at 8°C prior to using them as primary antibodies for immunofluorescence.
For the investigation of seroconversion in reconvalescent mice, confluent Vero cells in 96-well plates were infected with DHOV/India/1313/61 at moi 5. After 20 h, the cells were fixed with 4% PFA, and staining was conducted as above with 2-fold serial dilutions of the respective mouse sera and an Alexa Fluor 488-coupled donkey-anti-mouse-IgG secondary antibody (A21202; Invitrogen).
## Scanning electron microscopy (SEM)
Huh7 cells were grown to 60% confluency on 18 × 18 mm indium tin oxide-coated coverslips (SPI supplies, #06465-AB, 8-12 Ω) and treated with BHK-21 mock-supernatant or rDHOVs at a MOI of 4. At 24 hpi, the cells were washed once with PBS and fixed with 4% paraformaldehyde (PFA) and 0.5% glutaraldehyde (GA) in PBS for 1 h at RT. Coverslips were then processed for scanning electron microscopy (SEM). Coverslips were washed with 0.1 M Cacodylate buffer. Next, the cells were incubated with 1% OsO4 at 4°C for 30 min, followed by washing with Cacodylate buffer. Dehydration was done in acetone solutions with increasing concentrations of acetone (25%, 50%, 75%, 95%, and 100%) and incubation for 10 min. Critical point drying was done on a Leica CPD300 at 17°C and 63.5 bar followed by sputter coating with a 5 nm thick layer of Au/Pd (80/20) using the Leica ACE600. Samples were mapped by SEM using an Aquilos 2 dual-beam cryo-focused ion beam-scanning electron microscope (ThermoFisher Scientific) operated at room temperature at magnifications between 10,000× and 25,000× and five keV, using OptiPlan with a working distance of 3 mm and in-column T2 secondary electron detector. SEM images were acquired in the MAPS software (ThermoFisher Scientific).
## Cryo-electron tomography (cryo-ET)
BHK-21 cells were infected with the respective viruses at moi 0.001 and incubated for 72 h with DMEM supplemented with 2% FCS, 20 mM HEPES, and 0.1% NaHCO 3 . Supernatants containing >1 × 10 6 pfu/mL of infectious particles were UV-inactivated (two times 900 µJ/cm 2 ) on ice. Successful inactivation was confirmed by plaque assay on Vero cells. The virus-containing supernatant was mixed with 10 nm protein A-coated colloidal gold (Aurion). The mixture (3-4 µL) was applied onto 200 mesh, copper R2/1 grids (Quantifoil), which were plasma cleaned using H 2 /O 2 mix for 10 seconds using Solarus plasma cleaner (Gatan). Plunge-freezing into liquid ethane was performed using an automatic EM GP2 plunge-freezing device (Leica) under the following conditions: chamber temperature: 25°C, humidity: 80%, back-side blotting: 3 s. Grids were stored in liquid nitrogen until cryo-transmission electron microscopy.
Cryo-ET data were collected using a Titan Krios transmission electron microscope (ThermoFisher Scientific) operated at 300 keV and equipped with a Quanta Imaging Filter (Gatan) with an energy filter slit set to 20 eV and a K3 direct electron detector (Gatan). Grids were mapped at 8,700× magnification (pixel spacing: 10.64 Å) to localize virions, and tilt series were acquired at 33,000× magnification (pixel spacing: 2.67 Å) in SerialEM (46) using a dose-symmetric tilting scheme (47), nominal tilt range from 60° to -60° and 3° increments, target defocus -3 µm, electron dose per record 3 e -/Å 2 . Tomograms were reconstructed in Etomo using weighted back projection with simultaneous iterative reconstruction technique (SIRT)-like filter equivalent to seven iterations, dose-weighting, and 2D contrast transfer function (CTF) correction. The length and diameter of virions from the outer membrane to the outer membrane were measured in IMOD (48).
## IFN induction and signaling reporter assays
To measure the influence of DHOV M-270 and M2-248 on IFN induction, 293T cells (~3 × 10 4 cells per 96-well cavity) were co-transfected with the following plasmids using FuGeneHD (Promega): 10 ng of p125-Luc (IFNβp-FF) encoding firefly luciferase (FF) under the control of the IFNβ promoter (49) and 4 ng of SV40p-RL, constitutive expression of Renilla luciferase (RL)(Promega), as well as 20 ng of each pCAGGS expression plasmid encoding FLAG-tagged bacterial chloramphenicol acetyltransferase (CAT), as a nega tive control, DHOV-NP (pCAGGS-DHOV-NP), DHOV M-270, DHOV M2-248, HA-tagged THOV-M (pCAGGS-M), or THOV-ML (pCAGGS-M5xTΔSA [17]), as a positive control. At 6 h post-transfection, the cells were infected with moi 1.0 of Sendai virus (SeV) strain Cantell (19) to activate the IFNβ promoter. At 18 hpi, the cells were lysed, and FF as well as RL luciferase activities were measured (Dual-luciferase reporter kit; Promega). FF luciferase was normalized to RL luciferase activity (FF/RL) and is indicated as a relative activity of 1.0 for the CAT control. Expression of the recombinant DHOV-NP, M-270 and M2-248, and THOV-ML were validated by western blot analysis using specific antibodies. Detection of β-actin was used as a loading control.
In a second set of experiments, the IFNβ promoter was activated by co-transfection of 100 ng per 12-well cavity of expression plasmids pCAGGS-FLAG-RIG-I-N, encoding the constitutively active N-terminal domain of the intracellular RNA sensors RIG-I, or plasmid pEF-MDA5-Myc encoding Myc-tagged MDA5 (kindly provided by Rick Randall) (50). Expression of reporter gene activity was determined at 24 h post-transfection as described above.
To measure the effect of M-270 and M2-248 on IFNα signaling, the 293T cells were co-transfected with 10 ng per 96-well cavity of a reporter plasmid encoding FF luciferase under the control of the IFN-stimulated murine Mx1 promoter, pGL3-Mx1p-FF (51), and 20 ng of the plasmids encoding the viral M proteins. At 24 h post-transfection, the cells were treated with 10 ng/mL of human IFN-α2a (PBL Assay Science), and activation of the Mx1 promoter was determined by measuring luciferase activities in the cell lysates at 24 h post-treatment as described above.
## Co-immunoprecipitation
293T cells (~1 × 10 6 cells per 6-well) were transfected with 1 µg of each pCAGGS expression construct coding for C-terminally FLAG-or HA-tagged M-270 and M2-248. As a negative control, chloramphenicol-acetyltransferase (CAT) N-terminally FLAG-or HA-tagged was used. At 48 h post-transfection, the cells were lysed for 10 min on ice in lysis buffer (50 mM Tris-HCl, pH 8.0, 20 mM NaCl, 0.2% NP-40, 1 mM DTT, and protease inhibitor cocktail [Merck, 11873580001]), adjusted to 150 mM NaCl and cleared by centrifugation for 10 min at 12,000 rpm at 4°C. Cleared supernatants were incubated with 13 µL of anti-FLAG-M2-affinity agarose beads (Sigma-Aldrich, A2220) for 3 h at 6°C under rotation. Whole cell lysates and FLAG-agarose precipitated proteins were denatured in SDS-sample buffer for 5 min at 95°C and analyzed by SDS-PAGE and western blot using DHOV-NP-, M-, and M2-specific antisera.
## Mass spectrometry (MS) analysis
To prove the expression of M2 on protein level in DHOV-infected cells, triplicates of A549 cells (~1 × 10 6 cells per condition) were infected with DHOV/India/1313/61 at moi 1 or mock-treated. The cells were lysed in 0.5 mL of 1% SDS in PBS at 6 and 24 hpi. Lysates were separated by SDS-PAGE, and bands in the range of 0-35 kDa were excised from the gel and digested with trypsin. Resulting peptides were analyzed by LC-MS/MS on a Qexactive HF-X mass spectrometer coupled to an EasyLC 1200 nanoflow-HPLC (Thermo Scientific). MS raw files were analyzed using MaxQuant (version 1.6.2.10) (52). MaxQuant results were analyzed using Perseus (v.1.5.5.3) (53) and Instant Clue (v.0.10.10.20210316) (54). iBAQ (intensity-based absolute quantification) values for DHOV M-270 and M2 as well as cellular GAPDH and β-actin were median normalized and log-transformed (55).
## RNA isolation and RT-PCR
RNA was isolated from cultured A549 cells or mouse organs using the NucleoSpin RNA mini kit (Macherey-Nagel). RNA concentration was determined using a NanoDrop photometer (Thermo Scientific).
Reverse transcription (RT) was performed using 10 ng of the purified RNA and the Qiagen RT-kit (Qiagen, Germany; cat. 205313) with random hexamer primers. Upon inactivation of the RT reaction, the cDNA was used for PCR analysis using KOD-Polymer ase (Sigma-Aldrich, cat. 71085) and the following primer pairs (Table S1) according to the manufacturer's instructions (tm 58°C): DHOV/M-segment primer pair 1553/1554; DHOV/NP-segment 3011/1952; and human β-actin 560/561. PCR products were analyzed in a 1.5% agarose gel electrophoresis and stained with ethidium bromide (1:10,000). Bands were excised from agarose gels and purified with a gel purification kit (Zymo; cat. D4002 according to the manufacturer's instructions). Gel-purified PCR products were used for cloning or for Sanger sequencing (Microsynth, CH).
## Quantitative RT-PCR (RT-qPCR)
For the quantitative analysis of viral and host gene expression, qPCR was performed with the PowerUp SYBR Green master mix (Thermo Fisher Scientific, Waltham, MA, USA; Cat. A25918) on a Quant Studio 5 system (ThermoF.). For the detection of DHOV transcripts,
## ETHICS APPROVAL
Animal experiments were performed in compliance with the German animal protec tion law and approved by the local animal welfare committee (Regierungspraesidium Freiburg, permit 35-9185.81 /G-21/075). The animals were handled in accordance with the guidelines of the Federation for Laboratory Animal Science Associations and the national animal welfare body.
## References
1. Bendl, Fuchs, Kochs (2023) "Bourbon virus, a newly discovered zoonotic thogotovirus" *J Gen Virol*
2. Godsey, Rose, Burkhalter et al. (2021) "Experimental Infection of Amblyomma americanum (Acari: Ixodidae) With Bourbon Virus (Orthomyxoviridae: Thogotovirus)" *J Med Entomol*
3. Talactac, Yoshii, Hernandez et al. (2018) "Vector competence of Haemaphysalis longicornis ticks for a Japanese isolate of the Thogoto virus" *Sci Rep*
4. Jones, Davies, Green et al. (1987) "Reassortment of Thogoto virus (a tick-borne influenza-like virus) in a vertebrate host" *J Gen Virol*
5. Peng, Zhang, Cui et al. (2017) "Structures of humaninfecting Thogotovirus fusogens support a common ancestor with insect baculovirus" *Proc Natl Acad Sci*
6. Fuchs, Lamkiewicz, Kolesnikova et al. (2022) "Comparative study of ten Thogotovirus isolates and their distinct in vivo characteristics" *J Virol*
7. Fuller, Faulstich, Barnes (1987) "Complete nucleotide sequence of the tick-borne, orthomyxo-like Dhori/Indian/1313/61 virus nucleoprotein gene" *Virology (Auckl)*
8. Albanese, Bruno-Smiraglia, Di Cuonzo et al. (1972) "Isolation of Thogoto virus from Rhipicephalus bursa ticks in western Sicily" *Acta Virol*
9. Butenko, Leshchinskaia, Semashko et al. (1987) "Dhori virus--a causative agent of human disease. 5 cases of laboratory infection" *Vopr Virusol*
10. Bricker (2019) "Therapeutic efficacy of favipiravir against Bourbon virus in mice" *PLoS Pathog*
11. Kosoy, Lambert, Hawkinson et al. (2014) "Novel Thogotovirus associated with febrile illness and death" *Emerg Infect Dis*
12. Cdc (2018) "Bourbon Virus"
13. Filipe, Peleteiro, De Andrade Hr (1990) "Dhori virus induced lesions in mice" *Acta Virol*
14. Mateo, Xiao, Lei et al. (2007) "Dhori virus (Orthomyxoviridae: Thogotovirus) infection in mice: a model of the pathogenesis of severe orthomyxovirus infection" *Am J Trop Med Hyg*
15. Li, Wang, Guzman et al. (2008) "Dhori virus (Orthomyxoviridae: Thogotovirus) infection of mice produces a disease and cytokine response pattern similar to that of highly virulent influenza A (H5N1) virus infection in humans" *Am J Trop Med Hyg*
16. Dubois, Terrier, Rosa-Calatrava (2014) "Influenza viruses and mRNA splicing: doing more with less" *mBio*
17. Hagmaier, Jennings, Buse et al. (2003) "Novel gene product of Thogoto virus segment 6 codes for an interferon antagonist" *J Virol*
18. Bussetti, Palacios, Da Rosa et al. (2012) "Genomic and antigenic characterization of Jos virus" *J Gen Virol*
19. Vogt, Preuss, Mayer et al. (2008) "The interferon antagonist ML protein of thogoto virus targets general transcription factor IIB" *J Virol*
20. Clay, Fuller (1992) "Nucleotide sequence of the tick-borne orthomyxo-like Dhori/India/1313/61 virus membrane protein gene" *J Gen Virol*
21. Ejiri, Lim, Isawa et al. (2018) "Characteriza tion of a novel thogotovirus isolated from Amblyomma testudinarium ticks in Ehime, Japan: a significant phylogenetic relationship to Bourbon virus" *Virus Res*
22. Anderson, Casals (1973) "Dhori virus, a new agent isolated from Hyalomma dromedarii in India" *Indian J Med Res*
23. Sang, Onyango, Gachoya et al. (2006) "Tickborne arbovirus surveillance in market livestock" *Emerg Infect Dis*
24. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature New Biol*
25. Peukes, Xiong, Erlendsson et al. (2020) "The native structure of the assembled matrix protein 1 of influenza A virus" *Nature New Biol*
26. Selzer, Su, Pintilie et al. (2020) "Full-length threedimensional structure of the influenza A virus M1 protein and its organization into a matrix layer" *PLoS Biol*
28. Yang, Feng, Liu et al. (2016) "pHdependent conformational changes of a Thogoto virus matrix protein reveal mechanisms of viral assembly and uncoating" *J Gen Virol*
29. Hallgren, Tsirigos, Pedersen et al. (2022) "DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks" *Bioinformat ics*
30. (2025) *Full-Length Text Journal of Virology*
31. Welker, Paillart, Bernacchi (2021) "Importance of viral late domains in budding and release of enveloped RNA viruses" *Viruses*
32. Hoffmann, Neumann, Kawaoka et al. (2000) "A DNA transfection system for generation of influenza A virus from eight plasmids" *Proc Natl Acad Sci*
33. Pichlmair, Buse, Jennings et al. (2004) "Thogoto virus lacking interferon-antagonistic protein ML is strongly attenuated in newborn Mx1-positive but not Mx1-negative mice" *J Virol*
34. Fuchs, Straub, Seidl et al. (2019) "Essential role of interferon response in containing human pathogenic Bourbon Virus" *Emerg Infect Dis*
35. García-Sastre (2017) "Ten strategies of interferon evasion by viruses" *Cell Host & Microbe*
36. Hagmaier, Gelderblom, Kochs (2004) "Functional comparison of the two gene products of Thogoto virus segment 6" *J Gen Virol*
37. Rossman, Jing, Leser et al. (2010) "Influenza virus M2 protein mediates ESCRT-independent membrane scission" *Cell*
38. Nakabayashi, Taketa, Miyano et al. (1982) "Growth of human hepatoma cells lines with differentiated functions in chemically defined medium" *Cancer Res*
39. Bell-Sakyi (1991) "Continuous cell lines from the tick Hyalomma anatolicum anatolicum" *J Parasitol*
40. Bell-Sakyi (2004) "Ehrlichia ruminantium grows in cell lines from four Ixodid Tick Genera" *J Comp Pathol*
41. Frese, Weeber, Weber et al. (1997) "Mx1 sensitivity: batken virus is an orthomyxovirus closely related to Dhori virus" *J Gen Virol*
42. Filipe, Casals (1979) "Isolation of Dhori Virus from Hyalomma marginatum Ticks in Portugal" *Intervirology*
43. Basler, Wang, Mühlberger et al. (2000) "The Ebola virus VP35 protein functions as a type I IFN antagonist" *Proc Natl Acad Sci U S A*
44. Niwa, Yamamura, Miyazaki (1991) "Efficient selection for highexpression transfectants with a novel eukaryotic vector" *Gene*
45. Patzina, Haller, Kochs (2014) "Structural requirements for the antiviral activity of the human MxA protein against Thogoto and influenza A virus" *J Biol Chem*
46. Kochs, Haller (1999) "GTP-bound human MxA protein interacts with the nucleocapsids of Thogoto virus (Orthomyxoviridae)" *J Biol Chem*
47. Kochs, Weber, Gruber et al. (2000) "Thogoto virus matrix protein is encoded by a spliced mRNA" *J Virol*
48. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
49. Hagen, Briggs (2017) "Implementation of a cryoelectron tomography tilt-scheme optimized for high resolution subtomogram averaging" *J Struct Biol*
50. Kremer, Mastronarde, Mcintosh (1996) "Computer visualization of three-dimensional image data using IMOD" *J Struct Biol*
51. Yoneyama, Suhara, Fukuhara et al. (1998) "Direct triggering of the type I interferon system by virus infection: activation of a transcription factor complex containing IRF-3 and CBP/ p300" *EMBO J*
52. Andrejeva, Childs, Young et al. (2004) "The V proteins of paramyxoviruses bind the IFNinducible RNA helicase, mda-5, and inhibit its activation of the IFN-beta promoter" *Proc Natl Acad Sci*
53. Jorns, Holzinger, Thimme et al. (2006) "Rapid and simple detection of IFN-neutralizing antibodies in chronic hepatitis C nonresponsive to IFN-alpha" *J Med Virol*
54. Cox, Mann (2008) "MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification" *Nat Biotechnol*
55. Tyanova, Temu, Sinitcyn et al. (2016) "The Perseus computational platform for comprehensive analysis of (prote)omics data" *Nat Methods*
56. Nolte, Macvicar, Tellkamp et al. (2018) "Instant Clue: a software suite for interactive data visualization and analysis" *Sci Rep*
57. Schwanhäusser, Busse, Li et al. (2013) "Corrigendum: global quantification of mammalian gene expression control" *Nature New Biol*
58. Pettersen (2021) "UCSF ChimeraX: structure visualization for researchers, educators, and developers" *Protein Sci*
59. Kyte, Doolittle (1982) "A simple method for displaying the hydro pathic character of a protein" *J Mol Biol*
60. Duvaud, Gabella, Lisacek et al. (2021) "Expasy, the Swiss bioinformatics resource portal, as designed by its users" *Nucleic Acids Res*
61. Jones (2014) "InterProScan 5: genome-scale protein function classification" *Bioinformatics*
62. Madeira (2024) "The EMBL-EBI Job Dispatcher sequence analysis tools framework in 2024" *Nucleic Acids Res*
63. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12502766&blobtype=pdf | # Molecular epidemiology and pretreatment drug resistance of HIV-1 among newly diagnosed individuals in Nanning City, Guangxi, China
Ting Huang, Jinfeng He, Qiuqian Su, Liangjia Wei, Jiao Qin, Xinju Huang, Chunxing Tao, Fei Zhang, Li Ye, Ping Cen, Hao Liang, Bingyu Liang
## Abstract
Emerging pretreatment drug resistance (PDR) significantly reduces the effectiveness of HIV antiretroviral therapy (ART). This study assessed the prevalence, associated factors, and transmission networks of PDR among newly diagnosed, ARTnaïve individuals in Guangxi, China, from 2019 to 2022. A cross-sectional study was conducted between 2019 and 2022 in Nanning, Guangxi, involving 1,260 newly diagnosed, ART-naïve individuals with HIV. PDR levels and mutations were identified using the Stanford HIV Drug Resistance Database. Multivariable logistic regression models were employed to identify factors associated with PDR and to cluster the molecular network. A total of 1,048 eligible pol sequences were analyzed. The over all prevalence of PDR was 8.4%, with non-nucleoside reverse transcriptase inhibitors (NNRTIs) being the most commonly affected drug class. The most prevalent NNRTIassociated mutation was E138A (2.4%). High-level resistance was primarily observed to efavirenz and nevirapine. The CRF08_BC subtype exhibited significant clustering of PDR-related sequences. Those individuals diagnosed in 2022 were more likely to have PDR. Transmission networks clustering was significantly associated with CRF01_AE and CRF08_BC subtypes, older age, and heterosexual transmission. This study identi fied a moderate prevalence of PDR among newly diagnosed HIV patients in Guangxi, primarily driven by NNRTI-associated resistance mutations. The pronounced clustering of PDR in the CRF08_BC subtype highlights the need for subtype-specific surveillance and intervention strategies. To improve treatment outcomes and constrain the spread of resistance, targeted educational programs on ART adherence and drug resistance awareness should be prioritized, especially among older adults and individuals infected through heterosexual contacts, alongside enhanced molecular monitoring. IMPORTANCE This study highlights the growing challenge of pretreatment drug resistance (PDR) among newly diagnosed individuals with HIV in Nanning City. As antiretroviral therapy (ART) coverage expands, the persistence and transmission of drug-resistant strains pose a significant barrier to long-term treatment success. By documenting trends in PDR and identifying associated factors within a large repre sentative sample, this study offers timely and actionable insights for clinicians and public health policymakers. The identification of key resistance mutations and clustering patterns, particularly in the CRF08_BC subtype, provides a critical foundation for tailored intervention strategies. Overall, these findings address a significant regional data gap and contribute to the optimization of HIV treatment and prevention efforts in China.
A ntiretroviral therapy (ART) has been highly effective in suppressing HIV viral loads to undetectable levels, significantly reducing transmission and transforming acquired immunodeficiency syndrome (AIDS) into a manageable chronic disease (1). However, the emergence of drug resistance during ART remains a significant challenge for both HIV treatment and prevention. Pretreatment drug resistance (PDR) refers to resistance detected in individuals who are ART-naïve or have prior exposure to antiretroviral drugs. PDR can compromise the effectiveness of ART, accelerate disease progression, increase mortality, and facilitate secondary HIV transmission. The rising prevalence of resistance, particularly to non-nucleoside reverse transcriptase inhibitors (NNRTIs), has been widely documented (2,3). Notably, efavirenz (EFV) and nevirapine (NVP) are associated with higher levels of resistance compared to other NNRTIs (4,5), prompting the World Health Organization (WHO) to recommend prioritizing PDR surveillance in countries that use these drugs as part of their first-line ART regimens (6). Therefore, ongoing surveillance of PDR prevalence, resistance patterns, and transmission dynamics is critical for optimizing HIV preventive strategies and ensuring the long-term effectiveness of ART programs.
In 2003, China launched its National Free Antiretroviral Treatment Program (7). By 2016, the "Treat-all" policy was implemented, extending ART access to all consenting people living with HIV (PLWH). ART implementation in China has significantly reduced all-cause mortality among PLWH, from 5.4% in 2013 to 2.7% in 2022 (8). However, a substantial number of PLWH continue to die from AIDS-related complications, underscor ing the importance of investigating contributing factors, including drug resistance, that compromise treatment effectiveness. Previous studies have shown that PDR is increas ingly prevalent in several regions of China, with reported rates of 23.1% in Shenzhen City (9), 18.3% in Xi'an City (10), 17.4% in Shanghai City (11), and 10.0% in Hunan Province (12). These findings highlight the necessity for continuous surveillance of PDR trends to inform and enhance HIV prevention and treatment strategies.
Molecular transmission networks provide critical insights into the dissemination pathways of drug-resistant HIV strains within populations. Studies from Croatia (13) and Mexico (14) have shown that resistant strains frequently circulate within tightly connected transmission networks. Similarly, the emergence of drug resistance clusters in certain epidemic regions of China indicates ongoing transmission of resistant HIV strains among PLWH (15,16).
In Guangxi, molecular network studies on PDR have been conducted in several regions, including Qinzhou and Baise (2). This study revealed that individuals over the age of 50 were more likely to be part of transmission clusters (TCs) and that the CRF08_BC subtype was particularly prone to sharing drug resistance mutations (DRMs) (2). These findings highlight the urgent need to strengthen surveillance and implement targeted interventions. Understanding regional differences in the dynamics of drug resistance is essential for optimizing treatment strategies and guiding the development of new antiretroviral drugs. However, existing studies have not comprehensively covered all areas, leaving significant gaps that may distort the overall picture of drug resistance in the province.
Therefore, this study aims to investigate the prevalence of primary PDR and to explore the transmission factors of DRM within molecular networks by recruiting newly diagnosed, ART-naive individuals, thereby providing essential evidence to inform and enhance HIV control strategies.
## RESULTS
## Sociodemographic and clinical characteristics of the participants
In this study, we enrolled 1,260 newly diagnosed HIV-infected individuals. Among them, 1,159 pol gene sequences were successfully amplified. Of these, 11 sequences contained mixed bases >5%, 41 contained stop codons, and 39 contained both mixed bases >5% and stop codons. After applying quality control criteria, 1,048 sequences were included in the final analysis (Fig. 1). As shown in Table 1, the majority of participants were aged 50-69 years. Most were male (73.3%) and of Zhuang ethnicity (65.9%). In terms of education, 52.8% had completed only primary school or below. Regarding marital status, 51.2% of participants were married. Most participants were farmers (76.5%). Heterosex ual transmission was the predominant route of HIV transmission (92.5%). Additionally, 57.4% had CD4 + T-cell counts of ≥200 cells/mm 3 before initiating ART, and 39.2% were infected with the CRF01_AE subtype.
## Trends in HIV PDR prevalence and associated mutations
The prevalence of PDR increased steadily from 4.6% in 2019 to 6.3% in 2020, 8.3% in 2021, and peaked at 13.5% in 2022, as shown in Fig. 2A. Regarding the types of drug resistance, NNRTI-associated PDR predominated throughout the study period, consis tently exceeding resistance to NRTIs and PIs. Notably, the prevalence of NNRTI resistance reached its peak in 2022 (Fig. 2B).
Among PI-associated mutations, Q58E was the most prevalent (0.6%), followed by M46V (0.3%) and V82VF (0.3%) (Fig. 3A). For NRTI-associated mutations, Y115YF was the most prevalent (0.6%), followed by M184MV (0.3%) (Fig. 3B). The most NNRTI-associated mutation was E138A, with a prevalence of 2.4% (Fig. 3C).
Figure 3D illustrates the prevalence of resistance to these drugs, categorized into low, intermediate, and high levels of resistance. High-level resistance was predominantly observed for EFV (1.6%) and NVP (1.9%). From 2019 to 2022, the prevalence of PDR to EFV increased from 0.6% to 4.5%, and resistance to NVP rose from 1.1% to 4.9%.
Over the same period, NNRTI-associated PDR increased significantly (β = 2.00, P = 0.005). Although EFV (β = 1.14) and NVP (β = 1.11) exhibited upward trends, their slopes did not differ significantly from that of overall NNRTIs (P for interaction = 0.237 and 0.223) (Fig. S2).
## Characteristics of PDR TCs in molecular networks
In the CRF01_AE network, three PDR TCs were observed. One cluster exhibited two Y115YF mutations, another displayed three M46V mutations, and the third contained two sequences harboring both V106I and V179D mutations. These clusters were primarily transmitted through heterosexual transmission (Fig. 4A).
In the CRF07_BC network, two PDR TCs were identified. One cluster included sequences with both K103S and G190A mutations, while another showed M184MV, M46MI, and M184MI mutations. These clusters were primarily transmitted through heterosexual contact (Fig. 4B).
In the CRF08_BC network, five PDR TCs were identified. The E138A mutation predominantly characterized four clusters, while one cluster showed Q58E and Q58QE mutations. Similar to the other subtypes, these clusters were primarily transmitted through heterosexual transmission (Fig. 4C).
## Factors associated with PDR
As shown in Table 2, multivariate logistic regression analysis revealed that the year of diagnosis was a significant factor associated with PDR. Individuals diagnosed in 2022 had a significantly higher risk of PDR compared to those diagnosed in 2019 (adjusted odds ratio [aOR] = 2.86, 95% CI: 1.32-6.90). In addition, there was weak but statistically nonsignificant evidence suggesting that individuals infected with the CRF08_BC subtype were more likely to exhibit PDR (aOR = 1.64, 95% CI: 0.96-2.84).
## Factors associated with transmission network clusters
Regarding molecular transmission network clustering, multivariate logistic regression analysis (Table 3) identified HIV subtype, age, ethnicity, and HIV transmission route as significant factors associated with the network. Individuals aged 50 years or older were more likely to be included in TCs (aOR = 1.75, 95% CI: 1.05-2.93). In contrast, individuals with heterosexual transmission were more likely to cluster compared to those with homosexual transmission (aOR = 3.08, 95% CI: 1.50-6.81). Moreover, infections with the CRF01_AE (aOR = 2.12, 95% CI: 1.54-2.94) and CRF08_BC (aOR = 2.54, 95% CI: 1.79-3.62) subtypes were significantly associated with higher clustering probability.
## Factors associated with PDR TCs
As shown in Table 4, infection with the CRF_08BC subtype was significantly associated with inclusion in PDR TCs (aOR = 2.88, 95% CI: 1.17-7.81). In addition, there was weak but statistically non-significant evidence that individuals diagnosed in 2022 were more likely to form PDR TCs (aOR = 4.49, 95% CI: 1.17-29.58).
## DISCUSSION
This cross-sectional study provided comprehensive insights into PDR trends and molecular transmission networks in Nanning City, Guangxi Province. Our findings revealed an increasing trend of PDR from 2019 to 2022, with resistance against NNRTIs being the most prevalent and peaking in 2022. Notably, molecular transmis sion network analysis identified fully resistant clusters in which all sequences shared identical resistance mutations, indicating ongoing transmission of resistant strains. This phenomenon was particularly pronounced in the CRF08_BC subtype, which exhibited both higher rates of clustering and a greater potential for resistance transmission. These findings highlighted the rising burden of PDR and the urgent need for sustained surveillance and targeted intervention strategies. By addressing critical data gaps and advancing understanding of regional resistance dynamics, this study provided essential evidence to inform the design of targeted and effective treatment and prevention programs in Guangxi and other areas.
A previous national HIV molecular epidemiology survey reported a national average prevalence of 4.4% for drug resistance (3). In contrast, our study identified a higher prevalence of PDR, at 8.4%, among newly diagnosed individuals in Nanning, exceeding the previously reported 6.7% in 2017 (17). Although this prevalence remained moderate and below the WHO's 10% threshold for public health concern (18), a steady increase in resistance was observed from 2019 to 2022. Notably, individuals newly diagnosed and ART-naïve in 2022 had a higher risk of PDR, suggesting a possible acceleration in the transmission of resistant HIV strains. Several factors may have contributed to this trend. First, the relaxation of COVID-19-related restrictions may have facilitated increased HIV transmission, including transmission of drug-resistant strains (19). Second, disruptions in HIV care during the pandemic, such as reduced access to routine viral load monitoring and delayed switching of failing ART regimens, may have contributed to the accumula tion of resistance (20). Finally, the cumulative effect of transmitted resistance over the years may have reached a tipping point, resulting in a noticeable rise in 2022. In addition, local spread of drug resistance might have been driven by individuals experiencing ART failure or those with transmitted DRMs. Given that PDR could lead to virological failure (21), the accumulation of DRMs (22), and reduced efficacy of first-line regimens (23), ongoing surveillance of PDR in Guangxi is crucial. It is essential to improve the monitoring of resistance trends, promote adherence to ART, and use molecular tools to track the spread of resistant strains. These strategies are critical for mitigating drug resistance and ensuring the continued effectiveness of HIV treatment programs.
The highest prevalence of drug resistance in this study was observed to NNRTIs, particularly EFV and NVP, consistent with previous studies (24,25). NNRTIs are known for their significant treatment efficacy (26), low pill burden, and favorable tolerability, which collectively improve adherence to ART. However, their widespread use has facilitated the emergence and persistence of drug-resistant mutations, which are easily detected due to their long-term stability in the viral population. HIV-1 is characterized by a low genetic barrier to NNRTI resistance, whereby a single mutation at the binding site can lead to drug resistance (27). A previous study has indicated that mutations in the HIV reverse transcriptase (RT) gene, beyond the currently defined resistance-associated mutations, may also contribute to NNRTI resistance in vivo (28). Long-standing HIV epidemics may further compound the accumulation of resistant mutations over time. According to WHO guidelines, when the prevalence of NNRTI-related PDR reaches or exceeds 10%, switching to a non-NNRTI-based first-line regimen is advised. These findings underscore the importance of adaptive ART strategies as resistance thresholds are approached, thereby ensuring sustained treatment efficacy and preventing the transmission of further resistance.
Consistent with previous studies in Guangxi (2), Sichuan (29), and Anhui (30) provinces, our study found that individuals aged 50 years or older were more likely to be part of molecular TCs. This pattern may reflect the relatively stable geographic locations and limited mobility of older adults, which restricts the broader dissemination of HIV-1 within this subgroup. Furthermore, older individuals often have a lower awareness of HIV-related risks (31) and are more likely to engage in commercial sex, exacerbating local transmission. Geographic hotspots arising from commercial sexual behaviors between older men and female sex workers play a significant role in driving the local HIV-1 epidemic (32). Given these observations, urgent efforts are required to curb ongoing transmission among older adults. Specifically, public health responses should prioritize the identification of TCs and the implementation of localized, targeted interventions aimed at disrupting transmission pathways in this high-risk and often underserved population.
We identified three PDR clusters in CRF01_AE, two in CRF07_BC, and five in CRF08_BC. These findings suggested that individuals within the same cluster are closely linked in terms of transmission dynamics. In this study, CRF08_BC exhibited a higher likelihood of PDR clustering compared to CRF01_AE. CRF08_BC has become a predominant strain among heterosexuals and injection drug users in Southern China. Notably, the transmis sion route for five clusters within the CRF08_BC subtype was primarily heterosexual contact. A previous study found that heterosexual transmission has surpassed injection drug use as the leading mode of CRF08_BC infection nationwide (33). We also observed a high prevalence of the E138A mutation, particularly within CRF08_BC strains, which aligns with findings from an earlier study in China (34), suggesting that E138A may be a signature mutation of this subtype. The identification of PDR clusters in this study underscores the importance of ongoing molecular network surveillance among newly diagnosed individuals, enabling precise and targeted interventions aimed at containing PDR dissemination.
This study had several limitations. First, the cross-sectional design limited our ability to assess the temporal dynamics of HIV transmission. Second, molecular network analysis does not establish direct transmission links between genetically connected individuals, and self-reported data, such as modes of sexual contact, are subject to information bias. Third, although all participants were newly diagnosed and ART-naïve, we were unable to verify whether any had previously used antiretroviral drugs for non-HIV indications. The limitation could have resulted in an incomplete representation of the transmission network and the exclusion of relevant individuals. Future studies involving larger and more diverse populations, with available contact and behavioral information, are needed to design effective interventions for high-risk groups. Regardless, given the significant risk of PDR, it is essential to integrate these findings into both treatment and prevention strategies. to NNRTIs, especially involving EFV and NVP. The E138A mutation was frequently detected in the CRF08_BC subtype, which showed increased PDR clustering within transmission networks. These findings underscore the utility of molecular network analysis for monitoring PDR and identifying targeted opportunities for intervention. Continued surveillance, particularly of NNRTI resistance and key mutations such as E138A, along with strengthened control measures, is essential to limit the emergence and transmission of PDR. These will be critical for guiding the selection and optimization of ART regimens in Guangxi.
## MATERIALS AND METHODS
## Study setting and population
A cross-sectional survey using a convenience sampling method was conducted across the nine counties and districts of Nanning City, Guangxi Province, from 1 January 2019 to 31 December 2022. During each survey period, staff at the recruitment sites recruited individuals newly diagnosed with HIV. To construct the molecular transmission network, we recruited more than 60% newly diagnosed HIV patients each year.
The participant inclusion criteria were as follows: (i) aged 18 years or older; (ii) newly diagnosed with HIV and ART naive; (iii) residence in Nanning for more than three months before diagnosis; and (iv) ability to provide verbal and written informed consent in Mandarin.
Venous blood samples (10 mL) were collected from each participant. Plasma was separated by centrifugation and stored in aliquots at -80°C for further analysis. Demographic and epidemiological information, including sex, ethnicity, age, education, occupation, marital status, year of diagnosis, and transmission route, was obtained through a structured questionnaire administered by trained personnel.
## Laboratory testing
HIV-1 RNA was extracted from plasma samples using the High Pure Viral RNA Kit (Roche, Germany). An in-house RT nested PCR was used to amplify a 1,061 bp fragment of the pol gene, covering the full-length protease (PR, 99 codons) and the first 299 codons of the RT gene. Reverse transcription and the first-round PCR were performed using a Prime Script One Step RT-PCR Kit (Takara, Dalian, China) under the following thermal cycling conditions: 50°C for 30 minutes, 94°C for 5 minutes, followed by 30 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 2 minutes, with a final extension at 72°C for 10 minutes.
For the first-round PCR, two forward primers were used: F1a (5′-TGAARGAITGYACT GARAGRCAGGCTAAT-3′, HXB2 positions 2,057-2,085) and F1b (5′-ACTGARAGRCAGGC TAATTTTTTAG-3′, HXB2 positions 2,068-2,092), along with the reverse primer RT-R1 (5′-A TCCCTGCATAAATCTGACTTGC-3′, HXB2 positions 3,370-3,348).
Nested PCR was performed in a 50 µL reaction volume under the following condi tions: 94°C for 5 minutes; 30 cycles of 94°C for 30 seconds, 63°C for 30 seconds, and 72°C for 2.5 minutes, and a final extension at 72°C for 10 minutes. The nested PCR used the forward primer PRT-F2 (5′-CTTTARCTTCCCTCARATCACTCT-3′, corresponding to HXB2 positions 2,243-2,266) and the reverse primer RT-R2 (5′-CTTCTGTATGTCATTGACAGTCC -3′, corresponding to HXB2 positions 3,326-3,304).
PCR products were confirmed by agarose gel electrophoresis, and positive amplicons were sequenced by Sangon Biotech (Shanghai, China) using the Applied Biosystems 3730XL Genetic Analyzer. Raw chromatogram data were assembled and cleaned using Sequencher 5.4.6.
## Sequence processing and quality control
Quality control was performed using the quality control tool from the Los Alamos National Laboratory HIV Sequence Database (https://www.hiv.lanl.gov) to identify and exclude sequences with mixed bases >5% or premature stop codons. All nucleotide sequences were aligned using the HIV Align tool (https://www.hiv.lanl.gov) and manually edited in BioEdit (version 7.0.9.0).
## Genotypic resistance analysis
DRMs and PDR were performed using the Genotypic Resistance Interpretation tool provided by the Stanford University HIV Drug Resistance Database (version 8.9; https:// hivdb.stanford.edu). DRMs were classified based on their ability to confer resistance to nucleoside RT inhibitors (NRTIs), NNRTIs, and protease inhibitors (PIs). PDR was defined based on resistance to one or more of the following antiretroviral drugs: seven NRTIs
## Molecular transmission network inference
Genetic distances were calculated for three predominant HIV-1 subtypes (CRF01_AE, CRF07_BC, and CRF08_BC) using the HIV TRACE tool (https://github.com/veg/hivtrace). To achieve a high-resolution molecular network, genetic distance thresholds for all sequences and the three major subtypes were optimized to maximize the number of molecular clusters, prevent the formation of oversized clusters, and more accurately identify potential transmission relationships (35). The optimal genetic distance threshold was defined as the value that yielded the most significant number of TCs. In this study, the optimal genetic distance thresholds were 0.015 for CRF01_AE, 0.008 for CRF07_BC, and 0.013 for CRF08_BC, ensuring both resolution and interpretability of the transmis sion networks (Fig. S1). A PDR cluster was defined as a network containing two or more identical DRMs. The HIV-1 genetic transmission network was visualized and analyzed using Cytoscape version 3.10.0.
## Statistical analysis
Statistical analysis was performed using R version 4.3.1. Quantitative data were presented as mean ± standard deviation for normally distributed variables and as median (interquartile range) for non-normally distributed variables. Frequencies and percentages were used to describe categorical variables. Univariate and multivariate logistic regression analyses were used to identify factors associated with PDR, transmis sion network clustering, and clustering within PDR transmission networks. A backward stepwise selection method was employed to determine the final model. P-values <0.05 were considered statistically significant.
To examine whether the temporal trends of PDR differed among NNRTIs overall and the specific drugs EFV and NVP, we fitted a linear regression model that included an interaction term between year and drug group. A statistically significant interaction term indicates that the temporal trend (i.e., slope of change) differs among drug groups. This approach allows for direct assessment of whether the rate of change in PDR differs between specific drug subgroups. The observed and fitted trends were visualized using the ggplot2 package. B.L., H.L., and P.C. conceived and designed the study. T.H., J.H., and Q.S analyzed the data and wrote the manuscript. L.W., J.Q., X.H., C.T., F.Z., and L.Y. were responsible for data collection. B.L., H.L., and P.C. further edited the manuscript and gave final approval. T.H., J.H., and Q.S. contributed equally to the manuscript. All authors have critically reviewed the paper.
## References
1. Volberding, Deeks (2010) "Antiretroviral therapy and management of HIV infection" *Lancet*
2. Zhang, Liang, Liang et al. (2021) "Using molecular transmission networks to reveal the epidemic of pretreatment HIV-1 drug resistance in Guangxi" *China. Front Genet*
3. Hao, Zheng, Gan et al. (2015) "Changing proportions of HIV-1 subtypes and transmitted drug resistance among newly diagnosed HIV/ AIDS individuals -China" *China CDC Wkly*
4. Fokam, Chenwi, Tala et al. (2023) "Pre-treatment HIV drug resistance and genetic diversity in Cameroon: implications for firstline regimens" *Viruses*
5. Cao, Wu, Liu et al. (2024) "Molecular transmission network and drug resistance in treatment-naive HIV-1-infected patients in the Liangshan District" *AIDS Res Hum Retroviruses*
6. Who (2021) "WHO Releases HIV Drug Resistance Report 2021"
7. Zhang, Haberer, Wang et al. (2007) "The Chinese free antiretroviral treatment program: challenges and responses" *AIDS*
8. Zhao, Wei, Dou et al. (2023) "Changing mortality and patterns of death causes in HIV-infected patients -China, 2013-2022" *China CDC Wkly*
9. Li, Zhou, Zhang et al. (2023) "Characteristics of genotype, drug resistance, and molecular transmission network among newly diagnosed HIV-1 infections in Shenzhen" *China. J Med Virol*
10. Xia, Ba, Zhang et al. (2023) "Genetic diversity and characteristics of drug resistance among treatment-naive people living with HIV in Xi'an, China" *Drug Des Devel Ther*
11. Wang, Zhang, Zhang et al. (2019) "Diversity of HIV-1 genotypes and high prevalence of pretreatment drug resistance in newly diagnosed HIV-infected patients in Shanghai, China" *BMC Infect Dis*
12. Cao, Cao, Qi et al. (2023) "Prevalence of primary drug resistance among newly diagnosed HIV-1-infected individuals in Hunan Province" *AIDS Res Hum Retroviruses*
13. Oroz, Begovac, Planinić et al. (2019) "Analysis of HIV-1 diversity, primary drug resistance and transmission networks in Croatia" *Sci Rep*
14. Matías-Florentino, Chaillon, Ávila-Ríos et al. (2020) "Pretreatment HIV drug resistance spread within transmission clusters in Mexico City" *J Antimicrob Chemother*
15. Zhao, Song, Kang et al. (2021) "Molecular network analysis reveals transmission of HIV-1 drug-resistant strains among newly diagnosed HIV-1 infections in a moderately HIV endemic city in China" *Front Microbiol*
16. Liu, Dong, Liao et al. (2020) "Survey of pretreatment HIV drug resistance and genetic transmission network analysis among HIV patients in a high drug-use area of Southwest China" *CHR*
17. Kang, Liang, Ma et al. (2017) "Pretreatment HIV drug resistance in adults initiating antiretroviral therapy in China"
18. Bertagnolio, Beanland, Jordan et al. (2017) "The World Health Organization's response to emerging human immunodeficiency virus drug resistance and a call for global action" *J Infect Dis*
19. Yang, Yi, Qian et al. (2022) "Post-lockdown rebounding high-risk behaviors and HIV testing among MSM in China in the era of the COVID-19 pandemic" *Curr HIV Res*
20. Thu, Schemelev, Ostankova et al. (2025) "Experience in diagnostic of HIV drug resistance in the Mekong Delta region, Vietnam: a comparative analysis before and after the COVID-19 pandemic" *Diagnostics (Basel)*
22. Li, Song, Dong et al. (2018) "Impact of HIV pretreatment drug resistance on virological failure after one-year antiretroviral therapy -China" *China CDC Wkly*
23. Kityo, Boerma, Sigaloff et al. (2017) "Pretreatment HIV drug resistance results in virological failure and accumulation of additional resistance mutations in Ugandan children" *J Antimicrob Chemother*
24. Hamers, Schuurman, Sigaloff et al. (2012) "Effect of pretreatment HIV-1 drug resistance on immunological, virological, and drug-resistance outcomes of first-line antiretroviral treatment in sub-Saharan Africa: a multicentre cohort study"
26. Hui, Chen, Li et al. (2022) "Factors associated with newly diagnosed HIV infection and transmitted drug resistance among men who have sex with men in Harbin" *P.R. China. Front Public Health*
27. Zheng, Wu, Hao et al. (2022) "Epidemic characteristics of HIV drug resistance in Hefei"
28. Peng (2006) "A comparison of three highly active antiretroviral treatment strategies consisting of non-nucleoside reverse transcriptase inhibitors, protease inhibitors, or both in the presence of nucleoside reverse transcriptase inhibitors as initial therapy (CPCRA 058 FIRST Study): a long-term randomised trial" *Lancet*
29. Clutter, Jordan, Bertagnolio et al. (2016) "HIV-1 drug resistance and resistance testing" *Infect Genet Evol*
30. Ceccherini-Silberstein, Svicher, Sing et al. (2007) "Characterization and structural analysis of novel mutations in human immunodeficiency virus type 1 reverse transcriptase involved in the regulation of resistance to nonnucleoside inhibitors" *J Virol*
31. Yf, Liu, Hao (2021) "Epidemiological and spatiotemporal analyses of HIV/AIDS prevalence among older adults in Sichuan, China between 2008 and 2019: a population-based study" *Int J Infect Dis*
32. Wu, Zhang, Shen et al. (2019) "Phylogenetic analysis highlights the role of older people in the transmission of HIV-1 in Fuyang" *BMC Infect Dis*
33. Liao, Lin, Wang (2023) "AIDS awareness and HIV testing of elderly population in a rural area of Nanning City and their influencing factors" *Chin J Health Educ*
34. Jiang, Fan, Zhang et al. (2020) "A geographic hotspot and emerging transmission cluster of the HIV-1 epidemic among older adults in a rural area of eastern China" *AIDS Res Hum Retroviruses*
35. Li, Liu, Chen et al. (2021) "Using molecular transmission networks to understand the epidemic characteristics of HIV-1 CRF08_BC across China" *Emerg Microbes Infect*
36. Zhang, Hu, Song (2023) "Analysis of pretreatment drug resistance and polymorphic sites in CRF08_BC strains among HIV-1 patients" *J Microbiol Immunol Infect*
37. Wertheim, Pond, Forgione et al. (2017) "Social and genetic networks of HIV-1 transmission in New York City" *PLoS Pathog* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12424556&blobtype=pdf | # Proteomic analysis of rice mutant pir1 reveals molecular mechanisms triggering PCD and conferring high resistance to bacterial blight
Xinyu Chen, Yujie Zhou, Weifang Liang, Yuhang Zhou, Liujie Xie, Fan Hou, Bingsong Zheng, Jianzhong Li, Qibin Wu, Fatma Salem, Chengyu Liu
## Abstract
Most rice mutants exhibit some level of resistance to bacterial blight. This study demonstrates that the rice lesion mimic mutant (LMM) pir1 possesses enhanced resistance to bacterial leaf blight and triggers the upregulation of multiple pathogenesis-related (PR) proteins. Concurrently, photosynthetic parameter measurements revealed a significant impairment in the photosynthetic electron transport chain and photosynthetic capacity in pir1. Assessments of various stress factors and electron microscopy observations indicated that accumulated reactive oxygen species (ROS) caused severe damage to plant organelles. Utilizing proteomic approaches, we analyzed differentially expressed proteins (DEPs) between pir1 and its wild-type counterpart. Two-dimensional fluorescence difference gel electrophoresis (2D-DIGE) combined with mass spectrometry (MS) analysis of different leaf positions from both materials identified a total of 321 DEPs, comprising 87 upregulated and 234 downregulated proteins. Bioinformatics analysis of these DEPs revealed their involvement in diverse biological processes, including photosynthesis, carbohydrate metabolism, defense responses, redox homeostasis, and energy metabolism. Analysis of the regulatory network suggests that the mutation pir1 participates in programmed cell death (PCD), thereby triggering disease resistance responses.
## 1 Introduction
Rice bacterial blight, caused by Xanthomonas oryzae pv. oryzae (Xoo), is one of the three major traditional bacterial diseases of rice in China (Mew, 1987;Shasmita et al., 2023). Epidemiological surveys revealed severe outbreaks of bacterial blight (BB) in areas such as Quzhou and Shaoxing in Zhejiang Province, China, in 2014. These outbreaks resulted in extensive yellowing, chlorosis, and lodging of rice plants, with estimated yield losses exceeding 80%. Since then, both the geographical range and severity of the disease have continued to increase annually (Huadi et al., 2017). Long-term agricultural practice has demonstrated that utilizing resistant cultivars is the most economical, effective, and environmentally friendly strategy for BB control. However, under the long-term selection pressure from the same resistance sources, the pathotypes of Xoo readily evolve (Liang et al., 2024). This leads to the emergence and spread of new virulent strains, rendering existing resistance genes ineffective. Consequently, the continuous mining and functional characterization of novel resistance gene resources is imperative.
In nature, plants have evolved sophisticated immune mechanisms to defend against pathogen attacks. PCD is a ubiquitous process in organismal development and serves as a defense mechanism against pathogen infection and abiotic stress by selectively eliminating cells (Dangl et al., 1996;Greenberg, 1997). Most incompatible plant-pathogen interactions trigger an oxidative burst, during which PCD is initiated in infected and adjacent cells to restrict pathogen spread (Apostol et al., 1989;Draper, 1997). Upon sensing pathogen infection, Hypersensitive Response (HR) is rapidly activated at the infection site and surrounding tissues, leading to the sacrifice of plant cells to restrict pathogen invasion (Pankaj et al., 2018;Pitsili et al., 2020).
Rice LMMs are a class of mutants that spontaneously develop necrotic lesions resembling HR symptoms in the absence of pathogen infection; these lesions occur on leaves or leaf sheaths. Studies have shown that most rice LMMs confer durable and broadspectrum disease resistance through mechanisms involving cell death, ROS accumulation, and activation of defense-related genes (Sha et al., 2023). At present, many rice LMMs have been characterized, with at least 61 causal genes cloned and characterized. Among these, 40 were identified via map-based cloning (Chen et al., 2025;Ruan et al., 2024;Wang et al., 2025). These LMM genes encode diverse proteins with various biological functions and can be classified into functional groups such as transcription and protein translation, the ubiquitin-proteasome pathway, protein phosphorylation, vesicle trafficking, metabolic pathways, and plant hormone signaling. This diversity indicates that multiple pathways are involved in lesion mimic formation. OsNPR1 is a key regulator of Systemic Acquired Resistance (SAR) and confers durable, broad-spectrum resistance in rice. OsNPR1-OE plants develop lesion mimic spots on leaves and exhibit enhanced resistance to BB (Chern et al., 2005). The rice spl33 gene encodes a eukaryotic translation elongation factor 1 alpha protein, comprising a non-functional zinc finger domain and three functional EF-Tu domains. Research suggests that ROS accumulation may cause cell death in the spl33 mutant, and the activation of multiple defenserelated genes, such as PR1a and PBZ1, likely contributes to its enhanced disease resistance (Wang et al., 2017). The rice OsRLR1 gene encodes a CC-NB-LRR protein showing high homology to RPM1 in Arabidopsis thaliana, a canonical NLR gene in the disease resistance pathway. In OsRLR1-OE plants, the expression levels of OsNPR1, OsPR1a, and OsPR10 is significantly upregulated. This indicates that OsRLR1 regulates OsNPR1 expression in rice, thereby activating OsNPR1-mediated defense responses. Furthermore, OsRLR1 interacts with the transcription factor OsWRKY19 via its CC domain, suggesting OsWRKY19 may participate in the OsRLR1induced disease resistance response (Du et al., 2021). Other LMM genes participate in the ubiquitin-proteasome pathway (Ren et al., 2023). For instance, Protein SPL11 possesses E3 ubiquitin ligase activity and acts as a negative regulator of plant cell death and defense responses. SDS2 encodes an S-domain receptor-like kinase, interacts with SPL11 and causes its phosphorylation. Subsequently, SPL11 ubiquitinates SDS2 to regulate its stability. Mutation of SDS2 partially suppresses the LM phenotype and disease resistance in the spl11 mutant. In addition, SDS2 interacts with two positive regulatory factors of rice immunity, OsRLCK118 and OsRLCK176. Subsequently, OsRLCK118 phosphorylates the NADPH oxidase OsRbohB, inducing ROS outbreaks in plants during pathogen infection, resulting in a disease-resistant response (Fan et al., 2018). In the ebr1 mutant, accumulation of OsBAG4 triggers PCD, while reducing OsBAG4 expression leads to the inhibition of cell death and disease resistance. EBR1 encodes an E3 ubiquitin ligase and serves as a negative regulator of PCD and immunity in rice. EBR1 interacts with OsBAG4 and targets it for ubiquitination and degradation (You et al., 2016).
Despite significant progress in identifying rice LMM genes and characterizing some related pathways, substantial gaps remain. Firstly, while transcriptomic studies of mutants exist, comprehensive proteomic analyses of LMMs, particularly those exhibiting strong disease resistance, are lacking. Proteomics is essential for revealing post-transcriptional regulation, protein modifications, and functional protein networks directly driving PCD and resistance. Furthermore, the connection between proteomic changes in LMMs and physiological alterations in stress markers, including antioxidant enzymes and ROS levels, has not been fully explored. Understanding the protein network after transcriptional regulation in LMM is crucial for clarifying the functional mechanism of disease resistance. On the other hand, the diversity of pathogen races and the highly mutable nature of pathogens themselves lead to the rapid breakdown of resistance in bred varieties by newly emerging races, resulting in resistance gene failure.
Therefore, to address these gaps and expedite the cloning of the PIR1 gene, we conducted an integrated study on the mutant pir1, which exhibits high resistance to Xoo. This mutant displays not only dwarfism, reduced fresh and dry weight, and pollen sterility, but also develops distinct reddish-brown lesions. This indicates that the function of the target PIR1 gene may be pleiotropic, as its deletion leads to alterations in multiple phenotypic traits. Therefore, studying the mutant pir1 is significant for elucidating lesion mimic genes and the interdependent regulatory mechanisms between closely associated functional genes.
Previous transcriptomic analyses revealed that lignin biosynthesis and plant hormones are involved in triggering PCD (Chen et al., 2020). Here, we performed comparative proteomic analysis between the mutant pir1 and its wild type to identify core proteins associated with PCD and disease immunity. Additionally, by measuring key stress physiological indicators, including superoxide dismutase (SOD), catalase (CAT), peroxidase (POD) activities, and hydrogen peroxide (H 2 O 2 ) levels, we linked the proteomic changes to functional physiological responses. This work provides a proteomic network of the mutant pir1, integrated with physiological data, offering novel insights into the molecular mechanisms behind its robust disease resistance. Simultaneously, leveraging proteomics can significantly accelerate the cloning process of the pir1 LMM gene.
## 2 Materials and methods
## 2.1 Plant materials
Mutant pir1 is a lesion mimic mutant from an ethyl methyl sulfone (EMS) mutant library that is derived from the ZJ22 (Oryza sativa ssp. Japonica cultivar ZJ22). The rice materials and cultivation management methods for proteomics are de-scribed by Chen et al (Chen et al., 2020).
## 2.2 Pathogen inoculation and evaluation
Xoo Philippine race 6 (P6, strain PXO99A) was used for pathogen inoculation. Xoo was subcultured at 28 °C on potato semisynthetic agar (PSA) medium (potato, 300 g/L; Ca(NO 3 ) 2 •4H 2 O, 0.5 g/L; Na 2 HPO 4 •12H 2 O, 2.0 g/L; sugar, 15 g/L; agar, 15 g/L) for 48 h. Inoculates were prepared by suspending bacterial cells in sterile water and adjusting the concentration to OD600 = 0.5-0.8. Immerse the scissors in the bacterial suspension, and the leaves were infected with P6 by using scissors dipped in bacterial suspensions to clip leaves 1-2 cm down from the tip of the leaf blade at the heading stage. A total of at least 30 leaves of each of 10 pir1 and wild-type rice plants were inoculated with Xoo. Approximately three weeks after inoculation, the length of the lesion was measured from the cut surface at the tip to the distal-most position on the leaf to assess the resistance. To quantify bacterial populations, three inoculated leaves per plant were homogenized and resuspended in 10 ml of sterile H 2 O, with bacteria harvested individually. Diluted suspensions were plated on PSA medium, and bacterial growth was determined via colony counting of colony-forming units (CFUs).
## 2.3 Extraction and purification of total protein from rice leaves
Three biological replicates were set up in the experiment, and 2.5 g of leaves were taken from each replicate. The flag leaf was named leaf a, he 2nd leaf (from the top) was named leaf b, and the 3rd leaf (from the top) was named leaf c. The total protein of rice leaves was extracted by the previously described phenol-methanol method (Deng et al., 2007). Protein purification was performed using the 2-D clean up Kit (GE Healthcare, UK). According to the manufacturer's instructions, the protein concentration was measured using the 2d quantitative kit (GE Healthcare, UK). The absorbance of the sample and the standard solution at 480 nm was measured using a spectrophotometer (Ultrospec1100 pro UV), and the concentration of the tested protein was calculated based on the standard curve.
## 2.4 Fluorescent labeling and 2D-DIGE
The fluorescent dye was equilibrated at room temperature for 5 minutes. Then, 25 mL of anhydrous N,N-dimethylformamide (DMF; moisture content <0.005%, purity >99.8%) was added. The mixture was vortexed for 30 sec until complete dissolution of the dye, followed by brief centrifugation at 1200 g. After centrifugation, a working solution was prepared by mixing the fluorescent dye stock solution and DMF at a 1:1.5 ratio. This mixture was vortexed for 30s and centrifuged briefly to yield the fluorescent dye working solution. Separately, the protein solution pH was adjusted to 8.5 using 100 mM NaOH and/or 1M HCl. 20 mL aliquot of the protein solution was transferred to a new PCR tube, and its concentration was adjusted to 5mg/mL. Subsequently, 10 mL of the protein sample was labeled by adding 1 mL of Cy3/Cy5 fluorescent dye working solution and incubating on ice for 30 min. The reaction was quenched with 1 mL of 10 mM lysine, followed by brief vortexing, centrifugation (1 min), and 10 min incubation on ice. Finally, 3 mL of the Cy2 fluorescent dye working solution was added to the mixture, followed by brief vortexing, centrifugation 30 sec, and incubation on ice for 30 minutes. Labeling was terminated by adding 3 mL of 10 mM lysine, vortexing briefly, centrifuging 1 min, and incubating on ice for 10 min. The labeled sample was divided equally into three aliquots for immediate use in subsequent experiments. 2D-DIGE was done according to the method of Dong et al (Dong et al., 2017).
## 2.5 Gel scanning and image analysis
Scanning was performed using a Typhoon 8600 scanner (Ettan DIGE fluorescence scanner). Following initial testing, regions containing high-abundance proteins were selectively scanned, followed by scanning of the entire gel surface. Exposure levels were set, with pixel resolution selected between 30,000 and 55,000 dpi, and the scan resolution was set to 100 mm. Scanner settings were as follows: Cy3 dye: Resulting images appear blue. Emission filter wavelength: 595/25 nm; Excitation filter wavelength: 540/25 nm. Cy5 dye: Resulting images appear red. Emission filter wavelength: 680/30 nm; Excitation filter wavelength: 635/30 nm. Cy2 dye: Resulting images appear green. Emission filter wavelength: 530/40 nm; Excitation filter wavelength: 480/30 nm.
The scanned 2D-DIGE images were analyzed using DeCyder 2D software (version 7.0). Differential In-Gel Analysis (DIA) and Biological Variation Analysis (BVA) modules were employed for the analysis. Protein spots exhibiting differential expression were identified using a threshold of ≥1.5-fold change in abundance value and a statistically significant difference (p < 0.05) as determined by Student's t-test.
## 2.6 MS identification of DEPs
Protein samples (500 mg) were mixed with an equivalent volume of 2× rehydration buffer and 2 mL of 1% bromophenol blue. The volume was adjusted to 150 mL using 1× rehydration buffer. Following electrophoresis, the gel was transferred into fixing solution and incubated with gentle agitation on an orbital shaker for 1 hour. After fixation, the fixing solution was discarded and replaced with Coomassie Brilliant Blue working solution. The gel was stained on the orbital shaker for 12 hours. Upon completion of staining, the dye solution was discarded and the gel was rinsed with deionized water (ddH 2 O). The gel image was subsequently captured using a UMAX Power Look 2100XL scanner at a scanning resolution of 300 dpi. MS identification was performed by Applied Protein Technology (APT, Shanghai) and Beijing Genomics institution (BGI, Beijing).
## 2.7 Quantification of gene expression by quantitative real-time PCR
Total RNA was extracted from different leaf positions using Trizol. Genomic DNA was removed by treatment with RNase-free DNase. RNA was reverse transcribed into cDNA using the Hifair ® III 1st Strand cDNA Synthesis Kit. qRT-PCR was performed using the 2×Super EvaGreen ® Master Mix for HRM under the following conditions: 45 cycles of 95°C for 30 s, 60°C for 45 s, and 72°C for 30 s. Actin was used as the internal control. Three replicates were performed per sample. The relative expression levels of the selected DEPs normalized to the expression level of the internal reference control were calculated using the 2 -△△Ct method. The primers were designed with Primer Premier Software and are listed in Supplementary Table S1.
## 2.8 Determination of stress factors
Nine stress-related indicators -SOD, CAT, POD, Soluble sugar, Soluble protein, H 2 O 2 , O 2 • -, Malondialdehyde (MDA), and Proline -were determined by Wuhan ProNets Testing Technology Co., Ltd, WuHan, China.
## 2.9 Determination of photosynthetic parameters
Measurements were conducted during the rice booting stage, with six biological replicates per treatment group. Leaf physiological indices were measured under clear and windless weather conditions. Soil and Plant Analyzer Development (SPAD) values were measured daily at 12:00 using a CL-01 chlorophyll meter on the flag leaf. The middle region of the leaf blade was selected as the measurement site to record the relative chlorophyll content. Photosynthetic performance was measured using the LI-6400XT portable photosynthesis system (LI-COR Biosciences, USA). Measurements were taken on the middle region of the flag leaf (avoiding the midrib by one-third) for the following parameters: Photosynthetic rate (Pn), Stomatal conductance (Gs), and Intercellular CO 2 concentration (Ci). All data were collected continuously for three days, and the daily mean values were used for statistical analysis.
## 2.10 Transmission electron microscope
Rice leaf segments from ZJ22 and pir1 plants were cut to dimensions of 1 mm × 3 mm. Samples were fixed overnight at 4°C in 2.5% glutaraldehyde, followed by rinsing with phosphate-buffered saline (PBS; 0.1 M, pH 7.0). Subsequently, samples were fixed in 0.1 M osmium tetroxide solution for 2 hours and rinsed again with PBS. Dehydration was performed using a graded ethanol series (50%, 70%, 80%, 90%, and 95% ethanol), followed by multiple rinses in absolute ethanol (2-3 times) and multiple rinses in absolute acetone (2-3 times). Subsequently, the embedding agent was permeated. Acetone was mixed with the embedding agent in a 1:1 solution and permeated for 1 hour. Prepare a solution of acetone and embedding agent at a ratio of 1:3 and permeate it for 3 hours. The sample blotting solution was placed into a new tube and penetrated with pure embedding agent overnight. Samples were embedded in pure embedding medium and polymerized at 70°C for 48 hours. The embedded blocks were sectioned, and ultrathin sections were stained with uranyl acetate and lead citrate prior to observation using the TEM.
## 3 Results
## 3.1 Resistance analysis of mutant pir1
In this study, pir1 and wild-type plants were inoculated with P6. Studies have shown that the lesion length of pir1 was significantly shorter than that of the wild type, and the bacterial count was significantly lower than that of the wild type (Figures 1a-c), indicating that the resistance of this mutant to white leaf blight is significantly increased compared with its wild type. The four PRs detected by qRT-PCR, including OsPR1a, OsPR1b, OsPR5, and OsPR10, were significantly upregulated in pir1 compared with wild-type plants (Figure 1d). This indicates that pir1 has typical autoimmune characteristics.
## 3.2 Proteomic analysis
Since the resistance of pir1 to Xoo is closely related to its degree of lesion mimic, we compared the protein profiles of pir1 in three different leaf positions, as described by chen et al (Chen et al., 2020), Leaf a (a large numbere of lesion mimic spots), leaf b (a small numbere of lesion mimic spots), leaf c (no lesion mimic spots), The leaves at the corresponding positions on wild-type plants were used as controls. 2D-DIGE analysis was used to separate DEPs, and two-dimensional fluorescence electrophoresis images with good repeatability were obtained after scanning (Supplementary Figure S1). The 2D-DIGe images of total leaf proteins from three leaf sites of mutant pir1 and wild-type ZJ22 were analyzed respectively by DeCyder 2D 7.0 software. Protein points with both up-regulation and down-regulation multifactorial of protein expression richness values greater than 1.5 times and p < 0.05 after Student's t-test were selected. Finally, 325 protein points with significant expression differences were obtained.
After comparing the Coomas brilliant blue staining gel images with the 2D-DIGE images, the differential protein spots were extracted from the Coomas staining gel, and finally 321 protein spots (Supplementary Table S2) were identified by MS. Among them, there were 87 protein spots (denoted as M) that were significantly upregulated by the mutant pir1 compared to the wild type. There were 234 significantly downregulated protein spots (denoted as W) (Supplementary Table S2). The three-dimensional stereoscopic images of the expression abundances of some protein spots of ZJ22 and pir1 identified are shown in the Figure 2.
Among these, leaf a exhibited 20 differentially expressed protein sites, while leaf b and leaf c showed 255 and 205 sites, respectively (Supplementary Figure S2). This disparity may arise because the mutant's first true leaf had not yet developed lesion mimic phenotype, maintaining a morphology and physiological profile similar to the wild-type counterpart. Consequently, the differential protein expression at this stage was minimal, accounting for only 9.3% of the total quantified proteins. In contrast, The 2nd and 3rd leaves-stages at which lesion mimic phenotypes emergedisplayed the highest differential expression, representing 79.4% and 63.8% of total proteins, respectively. Notably, 139 proteins were exclusively differentially expressed in both the second and third true leaves but absent in the first true leaf. These proteins likely represent key regulators through which pir1 triggers PCD and induces disease resistance.
## 3.3 Function classification of DEPs
We conducted GO enrichment and KEGG pathway enrichment analyses on DEPs respectively. GO enrichment analysis revealed that within Biological Processes, the terms "photorespiration" (GO: 0009853), "reductive pentose phosphate cycle" (GO: 0019253), and "proton transport coupled ATP synthesis" (GO: 0015986) were the most significantly enriched, encompassing 84, 73, and 50 proteins, respectively (Figure 3a). In the Molecular Function analysis, "ATP binding" (GO: 0005524), "magnesium ion binding" (GO: 0000287), and "ribulose-bisphosphate carboxylase activity" (GO: 0016984) were the most enriched GO terms, followed by "monooxygenase activity" (GO: 0004497) and "nucleotide binding" (GO: 0000166) (Figure 3a). For Cellular Component, the majority of proteins were localized to terms including "plastid" (GO: 0009536), "chloroplast" (GO: 0009507), and "chloroplast thylakoid membrane" (GO: 0009535) (Figure 3a).
KEGG enrichment analysis revealed that the DEPs were primarily enriched in the following pathways: metabolic pathways (KO01100), carbon metabolism (KO01200), carbon fixation in photosynthetic organisms (KO00710), biosynthesis of secondary metabolites (KO01110), and glyoxylate and dicarboxylate metabolism (KO00630) (Figure 3b).
The functional analysis of the DEPs by using bioinformatics databases such as NCBI, Uniprot, and RGAP, revealed their distribution across various biological processes, including photosynthesis, energy metabolism, defense response, sugar metabolism, and other functions (Figures 3c-e). The functional analysis indicated that the largest proportion of DEPs was associated with photosynthesis (38.01%). This was followed by proteins involved in carbohydrate metabolism and energy metabolism (15.89%), while redox-related proteins constituted the smallest group (4.98%) (Figure 3c). Furthermore, separate functional analyses of upregulated and downregulated proteins showed that both categories were predominantly enriched in the photosynthetic pathway, accounting for 45.98% and 35.04% of their respective groups (Figures 3d,e). This demonstrates that photosynthesis is strongly impaired in the mutant pir1. Additionally, defense response-related proteins constituted a significant proportion (21.84%) of the upregulated proteins (Figure 3d), suggesting that disease resistance and defense responses are activated in the mutant pir1. Conversely, among all downregulated proteins, those associated with carbohydrate metabolism and energy metabolism were relatively abundant, comprising 20.09% and 18.80%, respectively (Figure 3e). This pattern is consistent with the observed agronomic traits and characteristics such as the poor fertility in the mutant pir1.
## 3.4 Correlation analysis between mRNA and protein expression by qRT-PCR
Changes in DEPs were analyzed at the mRNA level using qRT-PCR (Figure 4). Figure 4 provides a visual overview of the correlation between mRNA and protein levels. To select genes for qRT-PCR validation, particular attention was paid to DEPs identified in key pathways highlighted in this study, such as Redox, Carbohydrate metabolism, Energy metabolism, Photosynthesis, and Defensive response. Furthermore, we noted that these DEPs were also differentially expressed at the transcriptional level in our prior transcriptome comparison of the mutant pir1 versus wild-type (Chen et al., 2020), indicating strong consistency in the expression of these genes at both transcriptional and post-translational levels (Supplementary Table S3). This correlation analysis approach allowed us to investigate the concordance between transcriptomic and proteomic changes in pir1. Total mRNA was isolated from leaf a, leaf b, and leaf c of both ZJ22 and pir1. The results demonstrate that the vast majority of DEPs exhibit a strong correlation with their corresponding mRNA levels. Specifically, the mRNA levels of the qRT-PCR analysis of differentially expressed genes in wild-type and mutant leaves. qRT-PCR was performed to determine the relative mRNA levels of selected differentially expressed genes (DEGs) in wild-type and mutant leaves. Data represent the mean relative expression level ± standard error (n = 3 biological replicates). In the bar chart, the green dotted line indicates the relative expression level for the wild-type control, while the purple solid line represents the relative expression level observed in the mutant. The bars depict the expression ratio of the mutant relative to the wild type for each DEP.
following proteins were highly correlated with changes in their protein abundance: Glutathione S-transferase (MA1B1C1), Ascorbate peroxidase (MB18C18), Manganese superoxide dismutase (MB20C20), Glycolate oxidase (WB86C86), Isoflavone reductase (MC1), ATP synthase subunit beta (MB9C9), Nucleoside diphosphate kinase (MB22C22), Translation elongation factor (WB43), Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit (Rubisco LSU, WA1B1C1), Photosystem II oxygen-evolving complex protein (WB45), Oxygen-evolving enhancer protein (WB39C39), PR protein (MA2B2C2), 14-3-3-like protein (MB11), PR protein (MB17C17), Chloroplast chaperonin (WB3), Inorganic pyrophosphatase (WB111C111), Ferritin (MC24), and Glyceraldehyde-3-phosphate dehydrogenase (MB1C1). In contrast, 14-3-3-like protein (MB33) and Peroxidase (MA1) exhibited a weak correlation with their respective protein levels in pir1.
## 3.5 Analysis of stress factors
Peroxidases and other enzymes involved in high metabolic turnover and self-replication can generate ROS as byproducts of fatty acid b-oxidation. ROS function not only as antimicrobial agents but also as signaling molecules regulating disease resistance responses. In our study, we observed significant enrichment of redox-related proteins. We therefore speculated that the levels of key antioxidant enzymes, such as SOD, ascorbate peroxidase (APX), and POD, might be substantially altered in the mutant pir1, leading to ROS accumulation. Consequently, we quantified the levels of relevant stress markers in both WT and mutant pir1.
As shown in Figure 5, the pir1 mutant exhibited significantly elevated levels of H 2 O 2 , O 2 • -(superoxide anion), MDA, and proline compared to the wild type. This accumulation of ROS appears to trigger PCD, thereby playing a positive regulatory role in disease resistance. Both SOD and POD activities were significantly increased in pir1. These findings are largely consistent with data obtained at the transcriptional and translational levels.
SOD catalyzes the dismutation of O 2 • -to H 2 O 2 . Consequently, the observed increase in SOD activity likely contributes to the higher H 2 O 2 accumulation detected in pir1. In contrast, CAT activity was markedly and significantly reduced in the mutant compared to the wild type. This indicates a pronounced impairment in the mutant's ability to scavenge H 2 O 2 , further exacerbating ROS accumulation.
Additionally, soluble protein and soluble sugar contents were higher in the wild type than in the mutant pir1. This suggests that the mutant experiences ROS-induced stress, potentially resulting in insufficient carbohydrate catabolism. This deficit may impair energy metabolism essential for vital cellular processes, contributing to the observed mutant phenotypes of dwarfism and sterility.
## 3.6 Analysis of photosynthetic indexes
Proteomic analysis revealed the down-regulation of several key photosynthetic electron transport proteins, specifically PsbO, PsbP, PsaE and LFNR, in both leaf b and leaf c of the mutant pir1. Concurrently, multiple other proteins associated with photosynthesis were also down-regulated (Supplementary Table S2). Consequently, we measured SPAD values and photosynthetic parameters in both WT plants and the mutant pir1. The results demonstrated that SPAD values were significantly lower in the mutant pir1 compared to the WT (Figure 6a). Similarly, Pn and Gs were also significantly reduced in pir1 (Figures 6b,c). In contrast, Ci was higher in the WT than in the mutant (Figure 6d). These findings collectively indicate that photosynthetic pathways are substantially impaired in the mutant pir1, resulting in a severe reduction in photosynthetic capacity.
## 3.7 Observation of TEM
At the tillering stage, transverse sections were taken from the mid-section of leaves of both wild-type and mutant plants, and cellular structures were examined using TEM (Figure 7). The results revealed that in the wild-type, the cytoplasm was uniformly distributed, the grana lamellae of chloroplasts were well-organized, and the overall cellular architecture was intact. In contrast, the cellular structure of the mutant pir1 was severely compromised. Cells exhibited significant shrinkage and underwent intense plasmolysis. Although the cell wall remained relatively intact, it displayed localized thickening and a loose structure. The plasma membrane detached from the cell wall, exhibiting invagination and vesiculation. The cytoplasmic matrix was thin and watery, containing numerous irregular vacuoles. The chloroplast envelope was largely ruptured, the thylakoid system was completely disintegrated into fragments, the stroma was vacuolated, and the grana thylakoids were disorganized and fragmented. Mitochondrial cristae appeared blurred and reduced. The central vacuole was reduced in volume, with its membrane structure appearing wrinkled. In proteomics studies of the mutant pir1, downregulation of numerous proteins related to photosynthesis and carbohydrate metabolism, alongside upregulation of defense-related proteins, was observed. These changes correlate with PCD phenotype observed in pir1. Based on these observations and existing literature, we propose a potential functional model (Figure 8). Downregulation of photosynthetic proteins may lead to reduced photosynthetic efficiency. Unutilized light energy potentially promote the accumulation of ROS, a key signal triggering PCD. Concurrently, the upregulation of defense-related proteins may participate in enhancing disease resistance responses. These proteins themselves might generate ROS or participate in ROS signaling, thereby further regulating the progression of PCD. The upregulation of some redox-related proteins may help mitigate ROS accumulation, limiting the spread of lesions and thus offering some protection to the plant. The ensuing PCD process likely impacts carbohydrate metabolism in the mutant pir1, potentially causing significant disruption to cellular metabolism, including carbohydrate metabolism and energy production. It is hypothesized that this may prioritize the allocation of energy and carbon resources towards defense responses.
## 4.1 Photosynthetic related proteins
Research indicates that under strong light conditions, the closure of stomata in plant leaves leads to insufficient CO 2 supply, resulting in reduced carbon assimilation efficiency (Keyse, 1992). The excess light energy causes a depletion of the electron acceptor NADP + . This deficiency forces Photosystem I (PSI) to transfer electrons to molecular oxygen (O 2 ), reducing it to the superoxide radical (O 2 • -) (Mehta et al., 1992). Concurrently, excitation energy transfers the excess energy to ground-state oxygen, converting it into highly reactive singlet oxygen (¹O 2 ). The O 2 • -can react with compounds such as plastocyanin or cytochromes, generating more ROS molecules like H 2 O 2 (Henkes, 2001;Scandalios, 1993). These ROS intensify peroxidation processes and further attack other cellular components in the plant, damaging protein structures and ultimately causing cellular injury (Kang et al., 2007).
In the mutant pir1, downregulation of multiple photosynthesisrelated proteins was detected, including Ferredoxin-NADP + Reductase (FNR), the large subunit of Ribulose-1,5-bisphosphate Carboxylase/ Oxygenase (Rubisco-L), components of the Photosystem II Oxygen-Evolving Complex (OEC) and Oxygen-Evolving Enhancer (OEE) proteins, as well as Chlorophyll A-B Binding Proteins (CAB). This downregulation correlates with the PCD phenotype, suggesting impaired photosynthetic capacity in pir1. Considering the role of ROS in PCD, we speculate that the downregulation of photosynthetic proteins may reduce light capture and utilization efficiency, potentially increasing the risk of excess oxygen radical generation during photosynthetic electron transport. These accumulated ROS could be a key factor triggering PCD in pir1 and simultaneously contributing to the observed enhanced disease resistance. However, it is equally reasonable that the occurrence of PCD may also lead to the down-regulation of these photosynthetic proteins. However, it is equally plausible that the occurrence of PCD causes the downregulation of these photosynthetic proteins. Validating this requires direct functional assays to establish causality for the identified proteins.
The mutant pir1 exhibits phenotypes such as reduced photosynthetic efficiency and decreased chlorophyll content, indicating that pir1 is essential for maintaining normal photosynthetic function. This suggests that pir1 may influence the photosynthetic process through some mechanism. However, the specific molecular mechanisms by which pir1 affects or participates in the photosynthetic pathway, such as its direct targets or regulatory modifications, will be a major focus for future research.
## 4.2 Defense related protein
Compared to the wild type, pir1 exhibits differential abundance of several defense-related proteins, including: PRs, 14-3-3 proteins, thaumatin-like protein (TLP), wound-induced protein (WIP), Tcomplex protein, and AAA-type ATPase. TLP is considered a member of the pathogenesis-related protein family 5 (PR-5), involved in defense responses and exhibiting antifungal activity against numerous plant pathogens. Studies report that the TaTLP1 gene isolated from wheat participates in wheat's defense response against leaf rust (Puccinia triticina). Furthermore, through the interaction of TaTLP1 with TaPR1, which triggers ROS accumulation, TaTLP1 positively regulates wheat resistance to leaf rust (Wang et al., 2020). Haru Kirino et al. found that three TLP proteins secreted by the pinewood nematode (Bursaphelenchus xylophilus) significantly induced cell death in tobacco leaves (Kirino et al., 2020).
14-3-3 proteins constitute a class of important phospho-serine/ phospho-threonine binding molecules that play crucial roles in regulating plant development and defense response (Oh, 2010). Chang-Sik Oh et al. discovered that during the incompatible interaction between tomato (Solanum lycopersicum) and the fungal pathogen Verticillium dahliae carrying the Avr9 effector, three 14-3-3 genes (TFT1, TFT4, and TFT6) were induced (Oh and Martin, 2011;Roberts and Bowles, 1999). These genes may play a role in regulating HR.
Therefore, based on the known functions of these proteins in defense and cell death, we speculate that the increased expression of these defense-related proteins in the mutant pir1 may play a role in its enhanced defense response and could be involved in inducing or regulating the observed PCD process, potentially correlating with the increased resistance exhibited by pir1.
Notably, although R genes play crucial roles in disease resistance pathways, our proteomic analysis did not detect significant differential expression between wild-type and mutant plants. This observation could be attributed to several factors: Inherently Low Abundance: R proteins are typically expressed at very low levels. Even upon induction, their absolute abundance might fall below the reliable detection limit of mass spectrometry.Mechanism Independent of Abundance Changes: The core activation mechanism of R proteins primarily relies on post-translational modifications (PTMs) and conformational changes, rather than large-scale changes in total protein abundance. The key event is their activation triggered by effector recognition, leading to conformational shifts and PTMs. Consequently, the total protein levels may not exhibit significant upregulation within short timeframes. Function as Signaling Hubs: Activated R proteins function primarily as molecular switches and signaling platforms that recruit downstream components. "Sentry" Mode: Many R proteins are constitutively expressed and act as preformed "sentinels" surveilling the intracellular environment. Following activation, their primary responses involve modifications and oligomerization, rather than a significant increase. These hypotheses require further validation in future studies.
## 4.3 Redox-related proteins
SOD, an antioxidant enzyme, catalyzes the dismutation of the superoxide anion radical into oxygen and H 2 O 2 . Subsequently, enzymes including APX, glutathione transferase (GST), POD, and dehydroascorbate reductase (DHAR) can decompose H 2 O 2 into water and oxygen (Hiraga et al., 2000). Upregulation of SOD was observed in the pir1 mutant, which may promote the conversion of O 2
-to H 2 O 2 (Figure 5). Simultaneously, the upregulation of GST, POD, and APX suggests that pir1 likely activated its antioxidant defense system. This may help scavenge excess H 2 O 2 and other ROS, alleviating oxidative damage. This activation of the antioxidant system could be induced by PCD or photosynthetic imbalance, thereby offering some protection to the plant against death from severe oxidative stress.
## 4.4 Proteins related to energy metabolism
ATP synthase synthesizes ATP from Pi and ADP using the transmembrane proton gradient generated by photosynthesis (Boyer, 2000;Weber and Senior, 2000). In the mutant pir1, the downregulation of multiple ATP synthases results in insufficient energy supply within the plant. It is hypothesized that this energy deficiency contributes to the dwarfism of the mutant plants. The downregulation of multiple ATP synthase subunits indicates potentially limited ATP synthesis capacity in pir1.
ABC transporters are a class of ATP-driven pumps composed of two transmembrane domains and two ATP-binding domains (Jasinski et al., 2003). By binding and hydrolyzing ATP to release energy, ABC transporters translocate substrates-such as peptides, amino acids, sugars, alkaloids, glutathione, vitamins, and cellular metabolites-across membranes. This process facilitates critical physiological functions in plants, including the maintenance of cellular osmotic homeostasis and antigen presentation (Banasiak et al., 2020;Rinaldo et al., 2017;Verrier et al., 2008;Zhang et al., 2013). The ClpP protease is an ATP-dependent proteolytic enzyme that typically forms a proteolytic complex with AAA+ proteins. Members of the AAA+ protein family utilize energy derived from ATP hydrolysis to translocate protein substrates into the ClpP protease for degradation. This mechanism is essential for maintaining cellular homeostasis and energy regulation (Lozano-Durań and Robatzek, 2015;Xia et al., 2013). Sucher et al. demonstrated that introducing the wheat Lr34 gene into maize conferred enhanced resistance to northern leaf blight. Similarly, Lr34 expression in barley conferred resistance to leaf rust and powdery mildew. Krattinger et al. further identified abscisic acid (ABA) as a substrate for the Lr34 transporter. Lr34-expressing rice lines exhibited increased ABA accumulation, indicating Lr34 mediates ABA redistribution to enhance biotic stress tolerance. Silencing genes encoding ABCG/PDR-type transporters increases disease susceptibility (Boni et al., 2018;Krattinger and Kang, 2019;Sucher et al., 2017). For example, silencing both NaPDR1 and NaPDR1-like in wild tobacco following Alternaria alternata infection compromised resistance, impaired growth, and increased foliar disease severity, confirming PDR transporters function in biotic stress responses (Xu et al., 2018).
Therefore, we speculate that the downregulation of ABC transporters and ClpP protease disrupts cellular homeostasis and energy balance, potentially triggering cell death within the plant. This may contribute to the induction of PCD and the dwarfism and sterility phenotypes of the mutant. The downregulation of ABC transporters and ClpP protease likely interferes with intracellular homeostasis maintenance and ATP-dependent metabolic processes. Collectively, the downregulation of these energy metabolismrelated proteins may lead to insufficient cellular energy supply and metabolic dysregulation in pir1. This state of energy and metabolic imbalance, along with potentially compromised protein homeostasis, could be a factor promoting PCD initiation. However, while potential energy limitation might be one factor contributing to dwarfism and poor fertility, other mechanisms such as hormonal imbalance or cell death in meristematic/reproductive tissues could equally play significant roles.
## 4.5 Proteins related to carbohydrate metabolism
The primary physiological function of carbohydrates is to provide energy required for biological activities. Glycolysis is an essential stage for glucose catabolism in all organisms. In strawberry plants infected with C. fragariae, most glycolysis-related enzymes showed a significant decrease by 72 hours post-inoculation (HPI). However, aldolase (ALD) and phosphoglycerate kinase (PGK) exhibited a transient increase at 24 and 48 HPI before declining, while alcohol dehydrogenase (ADH) progressively declined throughout the infection period. This downregulation of ADH may enhance plant defense responses, suggesting that glycolytic inhibition could limit the pathogen's energy supply and thereby promote its fungal glycolysis (Fang et al., 2012;Pathuri et al., 2011). In the mutant pir1, proteins involved in this pathway, such as phosphoglycerate kinase (PGK), transketolase (TK), fructose-1, 6bisphosphate aldolase (FBA), and phosphoribulokinase (PRK), were down-regulated. This indicates that carbohydrate metabolism may be broadly suppressed. Metabolic suppression could further exacerbate the energy supply shortage.
The pentose phosphate pathway (PPP) represents an alternative route for glucose oxidation, providing reducing power for various cellular reactions and maintaining the reduced state of glutathione to prevent membrane lipid peroxidation (Fang et al., 2012). Therefore, it is hypothesized that the downregulation of PPP enzymes, such as ribulose-5-phosphate 3-epimerase (RPE), in pir1 suppresses this pathway. This could disrupt the glutathione redox balance, potentially causing membrane lipid peroxidation, insufficient carbohydrate breakdown, and energy deficiency. We hypothesize that this may contribute to phenotypic abnormalities.
The Tricarboxylic Acid (TCA) cycle is the final metabolic pathway for carbohydrates, lipids, and amino acids, and serves as the central hub connecting these three metabolic streams. Within this cycle, downregulation of enzymes like Malate dehydrogenase (MDH) and Succinyl-CoA synthetase (SuSc) also reduces the production of energy required for vital activities and may contribute to the dwarfism and sterility of the mutant plant. However, it is worth noting that these phenotypes could also stem directly from hormonal changes or cell death events in meristematic or reproductive tissues. Determining the primary causative factor for dwarfism, sterility, and other phenotypes requires further experimental evidence.
## 5 Conclusions
Our study demonstrates that the mutant pir1 exhibits strong resistance to Xoo. Proteomic analysis revealed downregulation of proteins associated with photosynthesis, carbohydrate metabolism, and energy metabolism in pir1, alongside upregulation of defenserelated proteins. Concurrently, key photosynthetic parameters were significantly reduced compared to the wild type, indicating diminished photosynthetic efficiency. This impairment in light energy utilization likely leads to excess excitation energy, triggering ROS accumulation and potentially inducing PCD. Furthermore, the upregulation of defense-related proteins in pir1 may contribute to enhanced apoptosis, ultimately resulting in significantly improved resistance to bacterial blight.
for Young Scholars (31101208), and the Key Program of Zhejiang Provincial Foundation for Natural Science (LZ16C130002).
## References
1. Apostol, Heinstein, Low (1989) "Rapid stimulation of an oxidative burst during elicitation of cultured plant cells: role in defense and signal transduction" *Plant Physiol*
2. Banasiak, Borghi, Stec et al. (2020) "The full-size ABCG transporter of medicago truncatula is involved in strigolactone secretion, affecting arbuscular mycorrhiza" *Front. Plant Sci*
3. Boni, Chauhan, Hensel et al. (2018)
4. "Pathogen-inducible Ta-Lr34res expression in heterologous barley confers disease resistance without negative pleiotropic effects" *Plant Biotechnol. J*
5. Boyer (2000) "Catalytic site forms and controls in ATP synthase catalysis" *Biochim. Biophys. Acta*
6. Chen, Liu, Zhou et al. (2025) "Lesion mimic mutant: an ideal genetic material for deciphering the balance between plant immunity and growth" *Rice*
7. Chen, Mei, Liang et al. (2020) "Gene mapping, genome-wide transcriptome analysis, and WGCNA reveals the molecular mechanism for triggering programmed cell death in rice mutant pir1" *Plants (Basel)*
8. Chern, Fitzgerald, Canlas et al. (2005) "Overexpression of a rice NPR1 homolog leads to constitutive activation of defense response and hypersensitivity to light" *Mol. Plant Microbe Interact*
9. Dangl, Dietrich, Richberg (1996) "Death don't have no mercy: cell death programs in plant-microbe interactions" *Plant Cell*
10. Deng, Zhang, Tang et al.
11. "A proteomics study of brassinosteroid response in Arabidopsis" *Mol. Cell Proteomics*
12. Dong, Fang, Yang et al. (2017) "Comparative proteomic analysis of susceptible and resistant rice plants during early infestation by small brown planthopper" *Front. Plant Sci*
13. Draper (1997) "Salicylate, superoxide synthesis and cell suicide in plant defence" *Trends Plant Sci*
14. Du, Zhang, Xing et al. (2021) "The CC-NB-LRR OsRLR1 mediates rice disease resistance through interaction with OsWRKY19" *Plant Biotechnol. J*
15. Fan, Bai, Ning et al. (2018) "The monocotspecific receptor-like kinase SDS2 controls cell death and immunity in rice" *Cell host & microbe*
16. Fang, Chen, Xin et al. (2012) "Proteomic analysis of strawberry leaves infected with Colletotrichum fragariae" *J. Proteomics*
17. Greenberg (1997) "Programmed cell death in plant-pathogen interactions" *Annu. Rev. Plant Physiol. Plant Mol. Biol*
18. Henkes (2001) "A small decrease of plastid transketolase activity in antisense tobacco transformants has dramatic effects on photosynthesis and phenylpropanoid metabolism" *THE Plant Cell Online*
19. Hiraga, Yamamoto, Ito et al. (2000) "Diverse expression profiles of 21 rice peroxidase genes" *FEBS Lett*
20. Huadi, Jianping, Chengqi et al. (2017) "Occurrence,epidemics dynamic,green prevention and control technology of rice bacterial leaf blight in Southern China" *Acta Agriculturae Zhejiangensis*
21. Jasinski, Ducos, Martinoia et al. (2003) "The ATP-binding cassette transporters: structure, function, and gene family comparison between rice and Arabidopsis" *Plant Physiol*
22. Kang, Matin, Bae et al. (2007) "Proteome analysis and characterization of phenotypes of lesion mimic mutant spotted leaf 6 in rice" *PROTEOMICS*
23. Keyse (1992) "Molecular biology of free radical scavenging systems" *Genet. Res*
24. Kirino, Yoshimoto, Shinya (2020) "Thaumatin-like proteins and a cysteine protease inhibitor secreted by the pine wood nematode Bursaphelenchus xylophilus induce cell death in Nicotiana benthamiana" *PloS One*
25. Krattinger, Kang (2019) "Abscisic acid is a substrate of the ABC transporter encoded by the durable wheat disease resistance gene Lr34" *New Phytol*
26. Liang, Zhou, Xu et al. (2024) "Identification and Genome Sequencing of Novel Virulent Strains of Xanthomonas oryzae pv. oryzae Causing Rice Bacterial Blight in Zhejiang"
27. Lozano-Durań, Robatzek, Mehta et al. (1992) "Oxidative stress causes rapid membrane translocation and in vivo degradation of ribulose-1,5-bisphosphate carboxylase/oxygenase" *Mol. Plant Microbe In*
28. Mew (1987) "Current status and future prospects of research on bacterial blight of rice" *Annu. Rev. Phytopathol*
29. Oh (2010) "Characteristics of 14-3-3 proteins and their role in plant immunity" *Plant Pathol. J*
30. Oh, Martin (2011) "Tomato 14-3-3 protein TFT7 interacts with a MAP kinase kinase to regulate immunity-associated programmed cell death mediated by diverse disease resistance proteins" *J. Biol. Chem*
31. Pankaj, Akshay, Preeti et al. (2018) "Prospects of understanding the molecular biology of disease resistance in rice" *Int. J. Mol. Sci*
32. Pathuri, Reitberger, Hückelhoven et al. (2011) "Alcohol dehydrogenase 1 of barley modulates susceptibility to the parasitic fungus Blumeria graminis f.sp. hordei" *J. Exp. Bot*
33. Pitsili, Phukan, Coll (2020) "Cell death in plant immunity" *Cold Spring Harb. Perspect. Biol*
34. Ren, Ding, Qian (2023) "Molecular bases of rice grain size and quality for optimized productivity" *Sci. Bull*
35. Rinaldo, Gilbert, Boni et al. (2017)
36. "The Lr34 adult plant rust resistance gene provides seedling resistance in durum wheat without senescence" *Plant Biotechnol. J*
37. Roberts, Bowles (1999) "Fusicoccin, 14-3-3 proteins, and defense responses in tomato plants" *Plant Physiol*
38. Ruan, Wu, Jiang et al. (1993) "SPL50 regulates cell death and resistance to magnaporthe oryzae in rice" *Plant Physiol*
39. Sha, Sun, Kong et al. (2023) "Genome editing of a rice CDP-DAG synthase confers multipathogen resistance" *Nature*
40. Shasmita, Mishra, Mohapatra et al. (2023) "Chemopriming for induction of disease resistance against pathogens in rice" *Plant Sci*
41. Sucher, Boni, Yang et al. (2017) "The durable wheat disease resistance gene Lr34 confers common rust and northern corn leaf blight resistance in maize" *Plant Biotechnol. J*
42. Verrier, Bird, Burla et al. (2008)
43. "Plant ABC proteins-a unified nomenclature and updated inventory" *Trends Plant Sci*
44. Wang, Huang, Ruan et al. (2025) "A MACPF protein osCAD1 balances plant growth and immunity through regulating salicylic acid homeostasis in rice" *Plant Cell Environ*
45. Wang, Lei, Wang et al. (2017) "SPL33, encoding an eEF1A-like protein, negatively regulates cell death and defense responses in rice" *J. Exp. Bot*
46. Wang, Yuan, Wu et al. (2020) "TaTLP1 interacts with TaPR1 to contribute to wheat defense responses to leaf rust fungus" *PloS Genet*
47. Weber, Senior (2000) "ATP synthase: what we know about ATP hydrolysis and what we do not know about ATP synthesis" *Biochim. Biophys. Acta*
48. Xia, Wei, Sun et al. (2013) "The maize AAA-type protein SKD1 confers enhanced salt and drought stress tolerance in transgenic tobacco by interacting with Lyst-interacting protein 5" *PloS One*
49. Xu, Song, Ma et al. (2018) "NaPDR1 and NaPDR1-like are essential for the resistance of Nicotiana attenuata against fungal pathogen Alternaria alternata. Plant Divers"
50. You, Zhai, Yang et al. (2016) "An E3Ubiquitin ligase-BAG protein module controls plant innate immunity and broad-spectrum disease resistance" *Cell Host Microbe*
51. Zhang, Huang, Zhu et al. (2013) "Isolation and characterization of a novel PDR-type ABC transporter gene PgPDR3 from Panax ginseng C.A. Meyer induced by methyl jasmonate" *Mol. Biol. Rep* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12614000&blobtype=pdf | # Complete genome sequence of bacteriophages Merry and Sunny infecting Microbacterium chocolatum strain GAI20246-6 isolated from an outdoor commercial algal pond
Alice Levesque, Ariel Rabines, Entesar Alrubaiaan, Aaron Oliver, Eric Allen, Dave Hazlebeck, Aga Pinowska, Jesse Traller, Lisa Allen, Lisa Zeigler
## Abstract
We report the isolation of two virulent phages from an outdoor algal pond infecting the bacteria Microbacterium chocolatum strain GAI20246-6. Their genomes are both 53.6kb with a GC content of 67.8%. Some genomic features are described as well as morphological characteristics based on transmission electron microscopy (TEM) micrographs.
KEYWORDS microbacteriophage, viral diversity, algae aquacultureO utdoor algal ponds are a sustainable alternative to produce various biomolecules (1). At Global Algae Innovations (21.996N 159.375W, Lihue, HI, USA), Nannochlo ris sp. (Chlorophyta) is used and develops an associated microbiome. Microbacterium chocolatum strain GAI20246-6 was isolated from these ponds (2) and demonstrated algicidal properties in laboratory conditions when co-cultivated with Nannochloris sp. We sought to use phage therapy as a strategy to regain algal productivity, thus demonstrat ing the use of microbacteriophage within commercial applications. Using M. chocolatum strain GAI20246-6 as host for all assays, two phages were isolated.Using the double-layer method, 50× concentrated (Vivaflow 50R, Sartorius) surface pond water (July 2018-January 2019) was spotted onto a host lawn (30°C Tryptic Soy Agar (TSA) 48 h). Clear viral plaques, indicative of a virulent phage, were isolated from two different samples, and a single clonal population was obtained following three rounds of purification (3). Liquid phage propagations were conducted in Tryptic Soy Broth (TSB) and incubated overnight at 30°C (200 rpm). For TEM, phage particles were stained (2% uranyl acetate) on a grid (Formvar/Carbon 200 mesh TH Copper, Ted Pella) and observed on a JEOL 1400 instrument. DNA was extracted from a phage lysate using the NucleoMag Virus kit (Machery-Nagel) and treated with RNase A (Thermo Scientific, 10 mg/mL). DNA was fragmented for size selection (Covaris S220), and libraries were constructed using Accel-NGS 2S PCR-free DNA library kit (Swift Biosciences), including eight PCR cycles (Kapa HiFi PCR, Kapa Biosystems) due to low DNA input (<10 ng). Sequencing was performed on Illumina MiSeq v2 500 cycles (2 × 150 bp). Reads were trimmed (Trimmomatic v1.2.15 [4]) for quality (minimum score 33) and to remove Illumina adapters (minimum internal and terminal hit length of 10). Viral genomes were assembled using CLC Genomics Workbench (clc_assembler) followed by VirSorter2 (5) and CheckV 1.0.1 (6). Annotation was conducted with Pharokka 1.7.1 (7), where CDS are predicted using Pyrodigal-gv 0.3.1 (8, 9) and PHANOTATE 1.5.1 (10), and tRNAs using tRNAscan-SE 2.0 (11) and Aragorn 1.2.41 (12). Each CDS function was assigned using the PHROG (13), VFDB (14) and CARD (15) databases within MMseq2 (16) and PyHMMER (17). Final annotation was manually curated, and five pairs of primers were designed to confirm genomic termini regions with PCR (2× Taq RED Master Mix, Apex). A repeated
motif within both genomes was detected using Geneious Prime 2024.0.4 and aligned with MUSCLE 5.1 (18). Default parameters were used except where otherwise noted.
TEM revealed siphovirus morphology (Fig. 1A andB). BLASTn (19) showed similarity to the genera Quhwahvirus and Metamorphoovirus (Table 1). Genomic features are summarized in Table 1. The genomes differ (nucleotide identity 89%) in a 1 kb region comprising a putative head-tail adaptor gene. Both phages are complete with a circularly permuted genome confirmed through PCR. Phages were assigned to Microbacteriophage Cluster EC based on genome similarity (Table 1) (https://phagesdb.org, accessed on May 2024 [20,21]) and shared features, such as possessing phage-encoded glycosyltransferase and UDP-glucose dehydrogenase genes, and a conserved 33 bp motif that is repeated 13 times throughout their genome in intergenic regions (Fig. 1C).
## ACKNOWLEDGMENTS
## References
1. Hachicha, Elleuch, Hlima et al. (2022) "Biomole cules from microalgae and Cyanobacteria: applications and market survey" *Appl Sci (Basel)*
2. Bd, Frits, Pham et al. (1993) "Viability and isolation of marine bacteria by dilution culture: theory, procedures, and initial results" *Appl Environ Microbiol*
3. Poxleitner, Pope, Jacobs-Sera et al. (2012) "Phage discovery guide. Howard hughes medical institute: Chevy Chase, MD (USA) and the NGRI in silico phage resource guide"
4. Bolger, Lohse, Usadel (2014) "Trimmomatic: a flexible trimmer for Illumina sequence data" *Bioinformatics*
5. Guo, Bolduc, Zayed et al. (2021) "VirSorter2: a multi-classifier, expert-guided approach to detect diverse DNA and RNA viruses" *Microbiome*
6. Nayfach, Camargo, Schulz et al. (2021) "CheckV assesses the quality and completeness of metagenomeassembled viral genomes" *Nat Biotechnol*
7. Bouras, Nepal, Houtak et al. (2023) "Pharokka: a fast scalable bacteriophage annotation tool" *Bioinformatics*
8. Larralde (2022) "Pyrodigal: python bindings and interface to Prodigal, an efficient method for gene prediction in prokaryotes" *JOSS*
9. Hyatt, Chen, Locascio et al. (2010) "Prodigal: prokaryotic gene recognition and translation initiation site identification" *BMC Bioinformatics*
10. Mcnair, Zhou, Dinsdale et al. (2019) "PHANOTATE: a novel approach to gene identification in phage genomes"
11. Chan, Lin, Mak et al. (2021) "tRNAscan-SE 2.0: improved detection and functional classification of transfer RNA genes" *Nucleic Acids Res*
12. Laslett, Canback (2004) "ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences" *Nucleic Acids Res*
13. Terzian, Ndela, Galiez et al. (2021) "PHROG: families of prokaryotic virus proteins clustered using remote homology" *NAR Genomics and Bioinformatics*
14. Chen, Yang, Yu et al. (2005) "VFDB: a reference database for bacterial virulence factors" *Nucleic Acids Res*
15. Alcock, Raphenya, Lau et al. (2020) "CARD 2020: antibiotic resistome surveillance with the comprehensive antibiotic resistance database" *Nucleic Acids Res*
16. Steinegger, Söding (2017) "MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets" *Nat Biotechnol*
17. Larralde, Zeller (2023) "PyHMMER: a Python library binding to HMMER for efficient sequence analysis" *Bioinformatics*
18. Edgar (2004) "MUSCLE: multiple sequence alignment with high accuracy and high throughput" *Nucleic Acids Res*
19. Altschul, Gish, Miller et al. (1990) "Basic local alignment search tool" *J Mol Biol*
20. Kim, Paul, Diaz (2022) "Characterization of phages YuuY, KaiHaiDragon, and OneinaGillian isolated from Microbacterium foliorum" *Int J Mol Sci*
21. Jacobs-Sera, Abad, Alvey et al. (2020) "Genomic diversity of bacteriophages infecting Microbacterium spp" *PLoS One* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12580268&blobtype=pdf | # Acceptance and impact of Nirsevimab and the RSVpreF vaccine following implementation in Austria
Michaela Höck, Wegene Borena, Jürgen Brunner, Karina Wechselberger, Johanna Scheiring, Elisabeth Ralser, Ulrike Peglow, Peter Wöckinger, Elisabeth D'costa, Verena Kaiser, Klaus Kapelari, Gisa Gerold, Thomas Müller, Ursula Kiechl-Kohlendorfer, Elke Griesmaier, Dorota Zarębska-Michaluk, Herbert Kurz, Clinic Donaustadt, Antoine Austria, Brault, Wöckinger P, D'costa
## Abstract
Background: Since summer 2024, passive immunization with nirsevimab (Beyfortus ® ) has been recommended for all infants in Austria to prevent severe respiratory syncytial virus (RSV) infection. Maternal vaccination with RSVpreF (Abrysvo ® ), which provides transplacental protection, became available in autumn 2023. The expected public health benefits of these preventive strategies depend largely on widespread acceptance; however, real-world data from Austria are unavailable.Objective: This study aimed to assess the acceptance and impact of RSV immunization strategies during the 2024/2025 season in Tyrol, Austria. Methods: A retrospective study was conducted analyzing all live births at three Tyrolean maternity wards (Innsbruck, Hall, and Schwaz) from 5 December 2024 to 15 April 2025. Immunization rates were analyzed, and RSV-related hospitalization frequency and duration were compared to pre-pandemic seasons. Results: Of 1,156 newborns, 57% received nirsevimab and 12% were protected by maternal RSVpreF protection, resulting in an overall coverage of almost 70%. RSV-related hospitalizations for infants under 1 year of age significantly decreased from 151 in pre-pandemic seasons to 47 in the post-nirsevimab season (p = 0.018). During the post-nirsevimab season, the median age at hospital admission was significantly higher (p < 0.001), and the length of stay was shorter (p = 0.031). Importantly, none of the hospitalized infants received nirsevimab, and only one was born to a vaccinated mother. Conclusion: Our findings highlight the positive impact of both RSV immunization strategies-nirsevimab and RSVpreF vaccine-while underscoring the need to enhance public awareness and education to improve immunization rates. Future immunization programs must be strengthened to provide better protection for the pediatric population and reduce RSV-associated morbidity in early life.
## Introduction
Respiratory syncytial virus (RSV) is the leading cause of bronchiolitis and hospitalizations in infants (1), with peak incidence between the ages of 3 and 6 months and seasonal activity from November to April (2). Despite limited antigenic variation, immunity against RSV is short-lived, and reinfections can occur throughout life, with the first infection often being the most severe (3). In Austria, RSV places a significant seasonal burden on pediatric healthcare, but treatment options are limited to supportive care, underscoring the need for effective preventive measures (4). For over 25 years, palivizumab (Synagis ® , AstraZeneca), a monoclonal antibody targeting the F-glycoprotein of the virus, was the only prophylaxis of RSV-related lower respiratory tract infections (LRTIs), limited to high-risk infants due to monthly dosing and high costs (5,6). In October 2022, the European Medicine Agency approved nirsevimab (Beyfortus ® , Sanofi-Aventis), a single-dose, long-acting monoclonal antibody, and demonstrated its effectiveness in preventing RSV-related LRTIs and reducing hospital admissions by at least 70 to 90% in clinical trials and real-world settings (7,8). This new monoclonal antibody targets the prefusion conformation of the RSV fusion (F) protein at a highly conserved antigen site, neutralizing a broad panel of RSV A and B viral subgroups (9).
A national campaign launched in December 2024 has provided it free of charge, prioritizing administration before maternity ward discharge with a recommended dose of 50 mg for those under 5 kg and 100 mg for those weighing more than 5 kg at the time of immunization. In parallel, seasonal administration of the bivalent recombinant stabilized prefusion F protein subunit vaccine (RSVpreF, Abrysvo ® ; Pfizer) at a single dose of 120 μg was available for pregnant women between 32 and 36 weeks of gestation to enhance neonatal protection through transplacental antibody transfer (10). These new RSV-preventing strategies are expected to have a strong impact on the burden of medically attended RSV-associated acute respiratory illness (ARI) among young children in Austria. While clinical trials have shown promising efficacy for nirsevimab and the RSVpreF vaccine (11)(12)(13), real-world data from Austria are lacking. This study, therefore, aims to assess the acceptance and impact of these RSV immunization strategies during the 2024/2025 season in Tyrol, Austria.
## Materials and methods
## Study design
The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies. The analysis was conducted at the maternal ward in the University Hospital Innsbruck-the tertiary care centre in Western Austria-and two peripheral primary care maternity wards in Hall and Schwaz. Immunization data were systematically collected between 5 December 2024 and 15 April 2025. Demographic and clinical data from electronic medical records were used to compare the epidemiology and burden of medically attended RSV-associated ARI in children during the 2024/2025 RSV season with previous seasons, which included two pre-pandemic seasons (2018-2019), the pandemic season (2020-2022), and the transmission seasons (2023-2024). The typical RSV season lasts from October to March. To ensure comprehensive capture of all cases within a given season, the observation period was extended to include 3 months prior to and 3 months after the defined season, resulting in a study period from July to June. Because RSV transmission patterns were disrupted during and after the COVID-19 pandemic, the RSV season after implementation of nirsevimab was compared to pre-pandemic seasons.
## Acceptance
Acceptance was assessed based on documented immunization rates for both nirsevimab and the RSVpreF vaccine across all three maternity wards (University Hospital Innsbruck, state Hospital Hall, and district Hospital Schwaz-both care level I) from 5 December 2024 to 15 April 2025. This period corresponds to the RSV season, which officially ended on April 15, along with the conclusion of the free immunization program. Maternal RSV vaccination status was determined from birth records and prenatal care documentation.
## Impact
The impact was evaluated based on the numbers and length of hospitalization due to RSV infection. During the study period, ICD10 diagnosis codes were used to identify cases of RSV-associated hospitalizations (J12.1; J21.0; J20.5; B97.4). The University Hospital of Innsbruck has the largest pediatric department in Tyrol and serves as the referral center for the most advanced neonatal and pediatric care. It manages a large proportion of all births in southwestern Austria, with most children requiring rehospitalization being admitted here. Inclusion criteria comprised hospitalization due to respiratory symptoms caused by RSV infection, confirmed by a positive antigen or polymerase chain reaction (PCR) test, and documented in the medical chart.
## Ethics approval
The study was conducted in compliance with the Declaration of Helsinki and with the approval of the Ethics Committee of the Medical University of Innsbruck (EC No. 1013/2023).
## Laboratory methods
Among children presenting with symptoms of respiratory tract infection, RSV status was determined using mid-turbinate nasal or oropharyngeal specimens, analyzed by either rapid antigen testing or institution-specific in-house PCR assays. Patient samples were processed within 24 h using an automated RNA extraction system (EMAG, Biomerieux), followed by RSV genotyping via RealStar RSV-RT-PCR to distinguish RSV A and B (Altona, Germany). Samples with a CT value below 30 were selected for further analysis. An F protein targeted Sanger sequencing was conducted. Amplified F-gene products were sequenced and aligned with reference strains and previous Austrian sequences (GISAID) to detect mutations in nirsevimabrelevant epitopes.
## Statistical analysis
A retrospective data analysis was performed using SPSS, version 29.0 for Windows (IBM Corp., Chicago, IL, USA). Descriptive statistics were used to characterize the individual variables and to determine the distribution of data, using the Shapiro-Wilk test. Values are expressed as numbers (frequencies, %), mean with standard deviation (±SD), and median with interquartile range (IQR). The Mann-Whitney U-test, Student's t-test, and the χ2 test were used where appropriate. A p-value of <0.05 was considered statistically significant.
## Results
During the study period (5 December 2024 to 15 April 2025), a total of 1,156 neonates were born across the three participating maternity wards (Innsbruck n = 706, Hall n = 279, and Schwaz n = 171), of them 559 were male newborns (48%).
## Acceptance
At the University Hospital of Innsbruck, 368 out of 706 newborns (55%) received nirsevimab, and 60 mothers (8%) were vaccinated with the RSVpreF vaccine during pregnancy. In the district hospital Schwaz, 105 out of 171 infants (61%) were given nirsevimab, and 23 mothers (13%) were vaccinated with the RSVpreF vaccine. In the state hospital Hall, 279 births were recorded, with 101 newborns (36%) receiving nirsevimab and 32 mothers (11%) vaccinated during pregnancy.
A total of 143 infants (12%) were admitted to the neonatal intensive care unit (NICU) due to prematurity or neonatal conditions such as congenital malformations, respiratory distress syndrome, pulmonary hypertension, perinatal asphyxia, sepsis, or hypoglycemia before nirsevimab immunization. Transfers to the NICU were documented to minimize potential bias and are illustrated in detail in Figure 1.
In total, 88 infants (57 preterm and 31 term-born) were immunized with nirsevimab in the NICU, notably including 23 born before 5th December. These patients had a mean birth weight of 2,381 g (range: 730-4,200 g) and a mean gestational age of 35 + 2 weeks (range: 24 + 2 to 42 + 0 weeks). Additionally, 27 infants were born to mothers who had received the RSVpreF vaccine. Three preterm infants received palivizumab.
In the study cohort, 660 newborns (57%) were discharged with protection against RSV via nirsevimab and 139 (12%) via maternal RSV vaccination, yielding a total prophylactic coverage of almost 70%. Immunization rates varied significantly between the three hospitals (p < 0.001), with nirsevimab rates ranging from 36% in Hall to 61% in Schwaz, and RSVpreF vaccination rates ranging from 8% in Innsbruck to 13% in Schwaz. An overview of the immunization rates is provided in Figure 1.
The timing of immunizations showed significant variation, with peaks in January (n = 139, 23%) and March (n = 153, 25%), and the lowest rate recorded in April (n = 55, 9%), as illustrated in Figure 2. Similarly, the highest number of NICU immunizations also occurred during these 2 months, with January accounting for 23 (34%) and March for 25 (37%) immunizations.
## Impact
Between 2018 and 2025, 882 patients under 18 years of age were hospitalized due to RSV infection, of whom 572 (65%) were infants below age 1 year. The 2024/2025 RSV season showed a similar seasonal trend to the previous year, with cases beginning to rise in December and peaking in January. Compared to pre-pandemic seasons, there was an apparent decline in RSV-related hospitalizations during the 2024/2025 season-similar to the reductions observed during the COVID-19 pandemic (Figure 3). This decline was particularly significant among infants under 1 year of age (151 vs. 47; p = 0.018).
Table 1 summarizes hospitalizations across RSV seasons since 2018, detailing patient age, gender, and length of hospital stay.
## RSV-associated deaths
Since 2018, five RSV-associated deaths have been reported in the pediatric department of the university hospital Innsbruck, all involving children with severe underlying conditions. In 2021, a 16-month-old patient with mumps parotitis experienced complications from acute liver failure. In 2023, a 16-month-old patient with a genetic syndrome involving a deletion in the 1q23.3 to q25.1 region. In 2024, a 14-year-old boy with an SMARCB1-deficient malignant tumor of the left orbit, staged T4N0M0. Additionally, in 2024, a 3-year-old boy had myotubular myopathy caused by an MTM1 gene mutation. There was also a 15-year-old boy with developmental and epileptic encephalopathy due to terminal deletion of Xp22.33 and terminal gain of Xq28.
## Hospitalizations
Between 5 December 2024 and 9 June 2025, a total of 85 patients (including 9 preterm infants with a gestational age range of 24 + 0 to 36 + 4 weeks, and 76 term-born infants) were hospitalized due to severe lower respiratory tract infections (LRTI) caused by RSV. Diagnosis of RSV was primarily confirmed by PCR or rapid antigen testing. The median length of hospital stay was 2 days (1; 5), with a median patient age of 269 (116; 964) days. Seven patients (8%) required admission to the pediatric intensive care unit. Supplemental oxygen was administered to 42 patients. Seven patients required non-invasive respiratory support, and two were intubated.
In the post-nirsevimab RSV season, the median age at admission was significantly higher (p < 0.001), while the median length of hospital stay was significantly shorter (p = 0.031) compared to the pre-pandemic period, as shown in Figure 4.
## Immunization status
Among all patients hospitalized since the implementation of the new RSV prevention strategies, one term-born infant was hospitalized at the age of 2.5 months for 2 days with RSV bronchiolitis, although no respiratory support was needed. The infant's mother had received Abrysvo ® 14 days before delivery.
A term-born infant with congenital diaphragmatic hernia who had received nirsevimab was diagnosed with bronchiolitis at 4 months
Immunization rates of nirsevimab on the three maternal wards (Innsbruck, Hall, and Schwaz). of age, requiring a 3-day hospitalization. At the time of discharge, the results of the respiratory panel were still pending. Subsequent testing did not detect RSV; instead, nucleic acids specific to rhinovirus were identified in the respiratory specimen.
Three former preterm infants had been immunized with palivizumab; however, their RSV infections occurred at 4.5, 1.6, and 2.9 years of age, respectively-well beyond the period during which protection from the antibody would be expected. The analysis of F protein sequences from 14 samples collected during the 2024/2025 RSV season did not identify any relevant mutations at the Nirsevimab binding site (previously unreported). The subgroup analyses for children hospitalized with RSV-associated ARI since the implementation of nirsevimab are presented in Table 2.
## Discussion
RSV remains one of the leading causes of LRTIs in infants and young children, placing a considerable burden on healthcare systems each winter season (14)(15)(16). The introduction of two new RSV immunization strategies-nirsevimab for all neonates and infants, and Median age at admission and median hospital stay, with p < 0.001 marked with *** and p < 0.05 marked with *.
## Variable Nirsevimab 2025
Total hospitalizations, n 85
Re-hospitalizations, n (%) 5 (6) Admissions to the pediatric intensive care unit, n (%) 7 ( 8) maternal vaccination with the RSVpreF vaccine-marks a major advancement in pediatric infectious disease prevention (11,12). This study provides the first real-world report in Austria following the implementation of these new RSV immunization strategies. In the 2024/2025 RSV season, 680 newborns (58%)including 23 preterm infants born before 5th December 5-were discharged with protection against severe RSV infection via nirsevimab, and 142 (12%) via maternal RSVpreF vaccination. Nonetheless, the uptake of both immunization strategies differed markedly among the three maternal wards. Nirsevimab coverage ranged from 36% in Hall to 55% in Innsbruck and 61% in Schwaz. A similar pattern was observed for the maternal RSVpreF vaccine, with uptake rates of 11% in Hall, 8% in Innsbruck, and 13% in Schwaz. These differences likely reflect variations in local implementation logistics, levels of parental education, and the degree of healthcare provider engagement. We found low uptake of the maternal RSV vaccination, although it became available in September 2023, which highlights the need to better integrate this intervention into routine prenatal care and to strengthen awareness among pregnant women and obstetric healthcare providers. Moreover, nirsevimab is included in the publicly funded national immunization program, whereas the RSVpreF vaccine is not covered and must be purchased privately. Martin et al. found that parents of immunized infants were more likely to have a university education (60.2% vs. 36.1%) and to be vaccinated against influenza (49.9% vs. 21.5%) compared to those of non-immunized infants. Awareness of RSV was higher among parents of immunized infants (73.8%), with 70.1% feeling well-informed, whereas 59.8% of non-immunized parents were unaware of RSV and reported feeling poorly informed. The majority of parents identified pregnancy as the optimal time to receive immunization information. Preferences for immunization strategies differed: 32.4% of immunized parents favored infant vaccination, while 60.3% of non-immunized parents preferred maternal vaccination. Despite some safety-related hesitancy, high parental acceptance and satisfaction with nirsevimab were reported during the 2023-2024 season in the Murcia region (17). Consistent with previous data, no serious adverse events were observed in our study; the most common side effects reported were mild, such as rash and injection-site irritation. These findings support the favorable safety profile of nirsevimab up to 360 days post-administration, regardless of gestational age or comorbidities (18)(19)(20). The effective reduction of RSV-associated hospitalizations and the economic burden was shown by several studies (16,21).
The introduction of passive immunization with nirsevimab for all neonates and infants, combined with maternal vaccination using RSVpreF, has led to a significant reduction in RSV-related hospitalizations among patients under 1 year of age. This decline parallels the decrease observed during the COVID-19 pandemic, when public health measures such as social distancing disrupted RSV transmission. While RSV circulation was markedly altered during the pandemic, our data indicate a return to typical seasonal patterns, with infection rising in December and peaking in January, consistent with other reports (22). This re-emergence of predictable patterns reinforces the importance of timely and widespread immunization prior to peak RSV activity. Importantly, this reduction in hospital burden occurred despite immunization rates of 57% for nirsevimab and even lower rates of maternal vaccination with the RSVpreF vaccine (12%). In our study, we observed a shift toward older age at the time of infection and a reduced duration of hospitalization. This positive real-world impact observed in our study aligns with post-licensure data from other European countries, which have reported risk reductions exceeding 80% (8,(23)(24)(25)(26)(27). Notably, a term-born infant aged 2.5 months was hospitalized with confirmed RSV bronchiolitis, diagnosed by rapid antigen testing. The infant's mother had received RSVpreF vaccination 14 days prior to delivery, indicating that while maternal immunization may attenuate disease severity, breakthrough infections can still occur, though in this case without requiring respiratory support. Conversely, another termborn infant with congenital diaphragmatic hernia who had received nirsevimab developed bronchiolitis at 4 months of age, necessitating a brief hospital stay. Interestingly, RSV was not detected in this case; instead, rhinovirus-specific nucleic acids were identified, suggesting that despite prophylaxis, infants remain susceptible to other viral pathogens causing similar clinical symptoms. These findings highlight the strong protective effect of the new RSV immunization strategies, while also underscoring the complexity of RSV prevention and the need for ongoing surveillance to better understand the spectrum of viral etiologies and the effectiveness of current prophylactic measures. Ernst et al. compared two consecutive seasons (2022/23 and 2023/24) and also showed the impact of nirsevimab in mitigating severe RSV-disease, following a national implementation in Luxembourg, with an immunisation coverage of up to 84% among neonates (26). This study is supported by a recent model simulation by Du et al., demonstrating the potential of RSVpreF vaccines for public health, but assuming high vaccination uptake rates (28).
Sustained substantial protection provided by nirsevimab will also depend on the ongoing genetic stability of the targeted epitope. Prior molecular studies have demonstrated that the RSV F protein, particularly the epitope targeted by nirsevimab (site Ø), exhibits low genetic diversity (29). Nevertheless, with the widespread and increasing use of monoclonal antibodies in the current and forthcoming seasons, there is a plausible concern that it may exert selective pressure on the circulating virus, potentially promoting the emergence and spread of escape variants. Although our retrospective analysis of F protein sequences from 14 samples collected during the 2024/25 RSV season did not identify any mutations associated with nirsevimab resistance, ongoing comprehensive genomic surveillance of circulating RSV remains crucial, particularly among hospitalized pediatric populations (30).
$$Death, n (%) 1(1)$$
## Limitation
Several limitations should be noted. The availability of nirsevimab only after the onset of the RSV season limited the window of prophylactic protection for some infants, potentially reducing the overall impact observed. Additionally, incomplete vaccine uptake and delayed immunization may have constrained the full impact of the campaign. Moreover, while hospital data were robust, cases managed in other healthcare settings or those with milder disease may be underrepresented, as data from smaller Tyrolean hospitals were not included.
Frontiers in Public Health 08 frontiersin.org
## Conclusion
The introduction of nirsevimab and maternal RSVpreF vaccination in Tyrol, Austria, represents a major step forward in preventing severe RSV disease in infants. This real-world evaluation confirms their positive impact on reducing RSV-related hospitalizations. However, limited product availability and suboptimal coverage during the first season hindered their full potential. To maximize public health impact, efforts must focus on improving awareness among parents and healthcare providers, overcoming logistical challenges, and integrating maternal vaccination into routine prenatal care. Strengthened collaboration between hospitals, pediatricians, and obstetric services will be crucial to ensure timely, broad, and equitable protection for all infants.
## References
1. Mansbach, Piedra, Teach et al. (2012) "Prospective multicenter study of viral etiology and hospital length of stay in children with severe bronchiolitis" *Arch Pediatr Adolesc Med*
2. Ralston, Lieberthal, Meissner et al. (2014) "Clinical practice guideline: the diagnosis, management, and prevention of bronchiolitis" *Pediatrics*
3. Hall, Weinberg, Iwane et al. (2009) "The burden of respiratory syncytial virus infection in young children" *N Engl J Med*
4. Resch, Puchas, Resch et al. (2020) "Epidemiology of respiratory syncytial virus-related hospitalizations and the influence of viral coinfections in southern Austria in a 7-year period" *Pediatr Infect Dis J*
5. Höck
6. "Frontiers in Public Health 09 frontiersin"
7. Resch, Muller (1999) "Prophylaxis of respiratory syncytial virus (RSV) in preterm infants with/without bronchopulmonary dysplasia: hyperimmune globulin (RSV-IGIV) and palivizumab (MEDI-493)" *Klin Padiatr*
8. Wegzyn, Toh, Notario et al. (2014) "Safety and effectiveness of Palivizumab in children at high risk of serious disease due to respiratory syncytial virus infection: a systematic review" *Infect Dis Ther*
9. Muller, Madhi, Nunez et al. (2023) "Nirsevimab for prevention of RSV in term and late-preterm infants" *N Engl J Med*
10. Drysdale, Flamein, Knuf et al. (2023) "Nirsevimab for prevention of hospitalizations due to RSV in infants" *N Engl J Med*
11. Zhu, Mclellan, Kallewaard et al. (2017) "A highly potent extended half-life antibody as a potential RSV vaccine surrogate for all infants" *Sci Transl Med*
12. Kampmann, Madhi, Munjal et al. (2023) "Bivalent Prefusion F vaccine in pregnancy to prevent RSV illness in infants" *N Engl J Med*
13. Sumsuzzman, Wang, Langley et al. (2025) "Real-world effectiveness of nirsevimab against respiratory syncytial virus disease in infants: a systematic review and meta-analysis" *Lancet Child Adolesc Health*
14. Marc, Vizzotti, Fell et al. (2025) "Real-world effectiveness of RSVpreF vaccination during pregnancy against RSVassociated 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*
15. Simoes, Madhi, Muller et al. (2023) "Efficacy of nirsevimab against respiratory syncytial virus lower respiratory tract infections in preterm and term infants, and pharmacokinetic extrapolation to infants with congenital heart disease and chronic lung disease: a pooled analysis of randomised controlled trials" *Lancet Child Adolesc Health*
16. Shi, Mcallister, Brien et al. (2017) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study" *Lancet*
17. Kurz, Hoffmann, Oeser et al. (2025) "Burden of disease and seasonal data of children hospitalized due to RSV and influenza infection before, during and after the COVID-19 pandemic" *Eur J Pediatr*
18. Yildiz, Resch, Strenger et al. (2024) "Evaluating the economic and epidemiological impact of RSV hospitalizations in southern Austria [southern Austria respiratory syncytial virus INpatient investigation (ARNI study)]. Influenza Other Respir Viruses"
19. Perez Martin, Gomez Moreno, Manresa et al. (2025) "Respiratory syncytial virus immunization with nirsevimab: acceptance and satisfaction assessment in infants and risk groups in the region of Murcia (Spain)" *Hum Vaccin Immunother*
20. Mankad, Leach, Chang et al. (2024) "Comprehensive summary of safety data on nirsevimab in infants and children from all pivotal randomized clinical trials" *Pathogens*
21. Hammitt, Dagan, Yuan et al. (2022) "Nirsevimab for prevention of RSV in healthy late-preterm and term infants" *N Engl J Med*
22. Griffin, Yuan, Takas et al. (2020) "Singledose Nirsevimab for prevention of RSV in preterm infants" *N Engl J Med*
23. Brault, Pontais, Enouf et al. (2024) "Effect of nirsevimab on hospitalisations for respiratory syncytial virus bronchiolitis in France, 2023-24: a modelling study" *Lancet Child Adolesc Health*
24. Hamid, Winn, Parikh et al. (2023) "Seasonality of respiratory syncytial virus-United States, 2017-2023" *MMWR Morb Mortal Wkly Rep*
25. Moline, Tannis, Toepfer et al. (2023) "Early estimate of Nirsevimab effectiveness for prevention of respiratory syncytial virusassociated hospitalization among infants entering their first respiratory syncytial virus season -new vaccine surveillance network" *MMWR Morb Mortal Wkly Rep*
26. Ares-Gomez, Mallah, Santiago-Perez et al. (2024) "Effectiveness and impact of universal prophylaxis with nirsevimab in infants against hospitalisation for respiratory syncytial virus in Galicia, Spain: initial results of a population-based longitudinal study" *Lancet Infect Dis*
27. Lopez-Lacort, Munoz-Quiles, Iglesias et al. (2023) "Early estimates of nirsevimab immunoprophylaxis effectiveness against hospital admission for respiratory syncytial virus lower respiratory tract infections in infants" *Euro Surveill*
28. Ernst, Bejko, Gaasch et al. (2024) "Impact of nirsevimab prophylaxis on paediatric respiratory syncytial virus (RSV)-related hospitalisations during the initial 2023/24 season in Luxembourg" *Euro Surveill*
29. Assad, Romain, Aupiais et al. (2024) "Nirsevimab and hospitalization for RSV bronchiolitis" *N Engl J Med*
30. Du, Pandey, Moghadas et al. (2025) "Impact of RSVpreF vaccination on reducing the burden of respiratory syncytial virus in infants and older adults" *Nat Med*
31. Peeples, Thongpan (2023) "Nirsevimab-resistant respiratory syncytial virus strains are rare but there" *Lancet Infect Dis*
32. Fourati, Reslan, Bourret et al. (2025) "Genotypic and phenotypic characterisation of respiratory syncytial virus after nirsevimab breakthrough infections: a large, multicentre, observational, real-world study" *Lancet Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645998&blobtype=pdf | # Optimization of VE607 to generate analogs with improved neutralization activities against SARS-CoV-2 variants
Shilei Ding, Derek Yang, Irfan Ullah, Ling Niu, Matthew Unger, Marco Díaz-Salinas, Monika Chandravanshi, Fei Zhou, Guillaume Beaudoin-Bussières, Mehdi Benlarbi, William Tolbert, Keon-Woong Yoon, Ruixue Xu, Geneviève Laroche, Fleur Gaudette, Abraham Morton, Zabrina Lang, Anna Son, Cameron Abrams, Marceline Côté, Amos Smith, Rick Huang, Doreen Matthies, James Munro, Marzena Pazgier, Pradeep Uchil, Andrés Finzi
## Abstract
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection remains a threat to human health, particularly among immunocompromised and elderly individuals, given their heightened vulnerability to coronavirus disease 2019 (COVID-19)associated morbidity and mortality. Recently, omicron subvariants such as KP.3.1.1 and XEC have emerged with an enhanced ability to evade humoral immunity. The devel opment of new strategies against these variants of concern remains an intense area of research. The small molecule VE607 is an entry inhibitor that targets the Spike glycoprotein and delays virus spread in vivo. To improve the potency of this new class of SARS-CoV-2 entry inhibitors, we generated and characterized VE607 analogs and identified candidates with enhanced activity against variants, including KP.3.1.1 and XEC. Promising analogs exhibited higher inhibitory potency than the original compound and stabilized the receptor-binding domain in its "up" conformation. Among these, DY-III-281 also reduced viral burden and delayed death in SARS-CoV-2-challenged K18-hACE2 transgenic mice. Furthermore, combining DY-III-281 with a non-neutralizing antibody engineered for Fc-enhanced functions exhibited an additive effect in reducing SARS-CoV-2-induced disease burden in mice. Our findings support the continued develop ment of small-molecule entry inhibitors, alone or in combination with antibody-based therapies, as a promising strategy to counteract emerging SARS-CoV-2 variants. IMPORTANCE Mutations in the Spike glycoprotein drive viral evolution and confer resistance to current vaccines and some therapeutic interventions against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Here, we report new analogs of the SARS-CoV-2 small-molecule entry inhibitor VE607. These analogs exhibited improved potency against emerging SARS-CoV-2 variants, including KP.3.1.1 and XEC. One analog, DY-III-281, delayed viral replication in SARS-CoV-2 WA1 -challenged K18-hACE2 transgenic mice, suggesting that small-molecule compounds targeting viral entry might be useful in fighting evolving SARS-CoV-2 variants.
mutations found in recent Omicron subvariants, have enabled the virus to better escape protective antibodies generated by infection and vaccination (7,8). Repeated breakthrough infections compounded by vaccine hesitancy further heighten the risk of long coronavirus disease (COVID) and contribute to a growing public health burden.
Most neutralizing Abs generated either from vaccination or infection target the receptor-binding domain (RBD) or the N-terminal domain (NTD) of the S1 subunit of S, but some also target the S2 subunit, which is the most conserved region of the SARS-CoV-2 S (9)(10)(11). Therapeutic monoclonal antibodies (mAbs) were developed for the treatment of infection by preventing virus spread through neutralization (12,13) and Fc-mediated effector functions (14)(15)(16)(17)(18)(19). The potential of these antibodies in therapeutic or prophylaxis settings was first tested in animal models (14,(20)(21)(22)(23)(24)(25)(26)(27)(28). These models have included non-human primates, hamsters, B6, and K18-hACE2 mice (29). Challenge studies in K18-hACE2 transgenic mice with protective Abs have revealed that Abs require both neutralization and Fc-effector functions to treat or prevent SARS-CoV-2 infections (14,20).
Small-molecule inhibitors targeting different viral proteins have been developed and approved to treat SARS-CoV-2 infections (30)(31)(32). At the beginning of the COVID-19 pandemic, the Food and Drug Administration approved Remdesivir to treat SARS-CoV-2 infection (33,34), which was originally discovered for the treatment of hepatitis C virus and Ebola virus (35,36). New drugs such as Paxlovid (a combination of a protease inhibitor, nirmatrelvir, and the Cyp3A inhibitor ritonavir) were approved later (37)(38)(39). Because of the relatively short phylogenetic distance between SARS-CoV-1 and SARS-CoV-2, we also thought of repurposing small-molecule inhibitors previously developed against SARS-CoV-1. During the SARS pandemic, an entry inhibitor targeting the viral Spike, VE607, was reported (40,41). We previously reported that VE607 inhibits SARS-CoV-2 entry by stabilizing the Spike in the RBD "up" conformation, thereby limiting infection by multiple variants, including several Omicron VOCs and reduced viral burden in K18-hACE2 transgenic mice (42). VE607 is a mixture of three stereochemical isomers composed of (S,S)-VE607, (R,S)-VE607, and (R,R)-VE607 in a ratio of 1:2:1, respectively. (R,R)-VE607 presented a slightly better neutralization (42) and was therefore selected as the basis of our medicinal chemistry efforts.
To advance this class of SARS-CoV-2 entry inhibitors, we synthesized and profiled a panel of VE607-derived analogs. This led to the identification of compounds with improved potency against circulating variants, such as KP.3.1.1 and XEC. Select ana logs demonstrated superior inhibitory activity compared to the parent molecule and promoted the "up" conformation of the spike RBD. Notably, a VE607 analog, DY-III-281, not only decreased virus burden but also extended survival in SARS-CoV-2-infected K18-hACE2 mice. Co-administration of DY-III-281 with a non-neutralizing antibody CV3-13 engineered for enhanced Fc effector function further mitigated disease severity, suggesting an additive treatment benefit. These results underscore the potential of combining optimized small-molecule inhibitors with antibody-based approaches to address ongoing viral evolution.
## RESULTS
## VE607 analogs
Our medicinal chemistry design, focused on the (R,R)-VE607 stereoisomer, sought to understand the structural requirements for activity and to improve potaency and pharmaceutical properties. From our previous in silico studies (42), we identified a potential binding site for VE607 where it interacted with RBD residue Tyr 505 (Fig. 1A). Therefore, in one series of analogs 15-25, we explored the chemical space around the proposed binding site by modifying region III of (R,R)-VE607 to understand what types of functional groups and substituents were tolerated at this position (Fig. 1B). Other analogs prepared included those with addition of a bromine atom on the aromatic ring 26-41 and truncated analogs 42-65 (Table 1).
To access the various VE607 analogs, we started with aryl alcohols 1-6 (Fig. 1C). To these alcohols, we added (S)-epichlorohydrin and an ethanolic solution of NaOH/EtOH in appropriate equivalents to form either the mono-or di-addition epoxides, 7-12 (Fig. 1C). The enantioselectivity of the synthesis of these epoxides was confirmed by chiral supercritical fluid chromatography. Additionally, the absolute stereochemistry of 8 was determined by single-crystal X-ray diffraction analysis (Fig. 1D). The corresponding epoxides were opened by a series of amines to yield the final VE607 analogs, 13-65 (Table 1; Supplemental material).
We tested all 64 VE607 analogs (Table 1) to verify the neutralizing capacity against pseudoviral particles bearing the Spike glycoproteins from SARS-CoV-1 and SARS-CoV-2. We used pseudotypes with the G glycoprotein of VSV (VSV-G) as a specificity control. Of the compounds that were synthesized, the identity of the amino function had the most impact on non-specificity, as measured by inhibition of VSV-G pseudotyped viruses, and less so on the inhibition of infectivity with SARS-CoV-2 or SARS-CoV-1 Spike pseudovi ruses. Particularly, lipophilic secondary amines (region III in Fig. 1B) that resulted in tertiary amine analogs with a calculated log of the distribution coefficient (cLogD) at pH 7.4 of greater than 1.0 resulted in non-specific inhibition of VSV-G. Additionally, while the tertiary amine analogs showed mostly specific neutralization to coronavirus envelopes, secondary amines exhibited broad non-specificity regardless of cLogD. For example, the dimethylamine 15 (IC 50 VSV-G: 70.7 µM) showed specific inhibition of coronaviruses; however, when one of the terminal methyl groups was removed to yield a methylamine in compound 25 (IC 50 VSV-G: 2.69 µM), the compound showed non-specific antiviral properties.
A derivative of (R,R)-VE607 (obtained from 2) with a brominated benzene at region I 26 was synthesized to obtain single-crystal X-ray diffraction data as well as facilitate co-crystallization with the viral Spike using cryo-EM. This compound showed better neutralization capacity (IC 50 SARS-CoV-2: 0.44 µM) than its parent compound, (R,R)-VE607, (IC 50 SARS-CoV-2: 0.98 µM). As such, a series of analogs bearing a 3-bromo substituent on the central aromatic core (region I) was synthesized as congeners of the previous resorcinol scaffold, 26-41. Similar trends that were observed in the resorcinol backbone were mirrored in the 5-bromoresorcinol backbone. Notably, secondary amines still showed broad non-specific inhibition of VSV-G pseudotypes, and highly lipophilic amine R groups similarly resulted in non-specificity. However, by modulating the amine groups to be smaller in size, we could take advantage of the increase in binding affinity afforded by the bromine while also specifically targeting coronavirus spike proteins. This series of modifications resulted in compounds like 27 (DY-III-281) with a superior neutralization capability against both SARS-CoV-1-S (IC 50 : 0.62 µM) and SARS-CoV-2-S (IC 50 : 0.44 µM) pseudoviruses.
Given the C2 symmetry of these analogs, we were then curious to see whether both arms were essential to the neutralization capabilities of these compounds. As such, we started from phenol 3 and synthesized compound 42. The antiviral assays showed that this compound was active (IC 50 SARS-CoV-2: 1.35 µM) and had a remarkable ligand efficiency (LE) of 0.47 (Table 1). Therefore, we pursued a series of these compounds starting from phenolic derivatives 9-12 (Supplemental material). Unfortunately, many of these compounds showed no activity against SARS-CoV-2, and therefore attention was turned back toward the symmetric two-armed derivatives, indicating that the dual-func tional motive is critical for viral neutralization.
From this brief structure-activity relationship (SAR) study of VE607 analogs, three compounds emerged as potential lead compounds, 27 (DY-III-281, IC 50 SARS-CoV-1: 0.625 µM, IC 50 SARS-CoV-2: 0.437 µM, LE: 0.36), 30 (DY-III-287, IC 50 SARS-CoV-1: 1.98 µM, IC 50 SARS-CoV-2: 1.48 µM, LE: 0.25), and 33 (DY-IV-048, IC 50 SARS-CoV-1: 1.44 µM, IC 50 SARS-CoV-2: 1.71 µM, LE: 0.30), and were selected to undergo further in vitro and in vivo testing. Of note, these compounds showed no toxicity in 293T-ACE2 and Vero-E6 cells at concentrations up to 100 µM (Fig. 2). or from SARS-CoV-1 were produced and tested in a standard pseudoviral neutralization assay (43). Pseudoviral particles carrying the VSV-G were used as a specificity control. As shown in Fig. 3,(R,R)-VE607 (Fig. 3A) and the three new analogs: DY-III-281 (Fig. 3B), DY-III-287 (Fig. 3C), and DY-IV-048 (Fig. 3D) neutralized pseudoviral particles carrying SARS-CoV-2 or SARS-CoV-1 Spikes. This neutralization was specific as they did not neutralize pseudoviral particles carrying the G glycoprotein of VSV (VSV-G). Overall, compound DY-III-281 was more potent against the different VOCs Spikes than the lead (R,R)-VE607 inhibitor (Table 2). This could be highlighted when comparing the IC 50 of the four compounds against the different pseudoviruses tested (Fig. 3E).
## Inhibition of viral infection
We next validated the neutralizing activity of DY-III-281 and DY-IV-048 using authentic SARS-CoV-2 D614G and XBB.1 viruses (44). Briefly, serially diluted (half-Log dilutions) compounds were pre-incubated with the authentic viruses before the addition to Full-Length Text authentic virus, we included a neutralizing antibody in both assays, CV3-25, which inhibits the infection of all current SARS-CoV-2 variants (45). As shown in Fig. S2, IC 50 of CV3-25 against SARS-CoV-2 D614G Spike pseudotyped virus is lower than that against SARS-CoV-2 D614G authentic virus, indicating the variability of different neutralization assays.
## VE607 analogs promote the RBD-up conformation
Previous work on the viral entry inhibitor (R,R)-VE607 demonstrated that it decreased viral replication in the lungs of K18-hACE2 mice. The decreased replication may have resulted from allosterically blocking ACE2-mediated Spike conformational changes necessary for membrane fusion (42). Here, we sought to determine how the optimized analogs impact Spike conformation. To that end, we implemented a previously described smFRET assay to probe the RBD dynamics of the Spike glycoprotein ectodomain in the absence or presence of ACE2, DY-III-281, DY-IV-048, a non-neutralizing NTD-specific antibody CV3-13 mAb ( 14), both CV3-13 mAb and DY-III-281, or both CV3-13 mAb and DY-IV-048. A site-specific enzymatic labeling approach was implemented for Spike such that fluorophores were positioned in the NTD and RBD (Fig. 5A). Under all conditions tested, hidden Markov modeling (HMM) of the smFRET trajectories revealed transitions between high-(~0.65) and low-(~0.35) FRET states, which reflect the RBD-down and RBD-up conformations, respectively (Fig. 5B) (46). The impact of DY-III-281 or DY-IV-048 on the fraction of time (occupancy) spent in the RBD-up conformation was assessed by incubating 0.1 µM fluorescently labeled Spike with 0.6 µM ACE2, 50 µM DY-III-281 or DY-IV-048, 0.3 µM CV3-13, or both 50 µM DY-III-281 or DY-IV-048 and 0.3 µM CV3-13, followed by smFRET analysis (Fig. 5C). In the absence of bound ligand, the occupancy of the RBD-up conformation was 40% ± 2%. Incubation with ACE2 increased the RBD-up occupancy to 67% ± 2% (P < 10 -4 ), consistent with previous work (46). Similarly, DY-III-281 increased the RBD-up occupancy to 55% ± 2% (P < 10 -4 ), DY-IV-048 increased the RBD-up occupancy to 61% ± 2% (P < 10 -4 , Fig. 5C). The increase in RBD-up occupancy upon binding to DY-III-281 or DY-IV-048 is consistent with previous work on the precursor molecule, (R,R)-VE607 (42). The NTD-targeting antibody CV3-13 moderately increased the RBD-up occupancy of Spike to 50% ± 2% (P = 0.002), consistent with previous studies of similar antibodies (46,47). Lastly, the combination of CV3-13 with DY-III-281 or DY-IV-048 increased the RBD-up to 59% ± 2% (P < 10 -4 ) or 56% ± 2% (P < 10 -4 ), respectively, which was not significantly different than the RBD-up occupancy in the presence of either ligand independently. These data demonstrate that the new analog shifts the conformational equilibrium of Spike in favor of the RBD-up conformation to an extent comparable to a NTD-targeting antibody.
## A VE607 analog modifies the conformational state of the RBD
To better understand the molecular details of (R,R)-VE607 analogs on the interaction with the SARS-CoV-2 spike, we employed single-particle cryo-electron microscopy (cryo-EM).
For our structural studies, we used a soluble preparation of a stabilized, uncleaved SARS-CoV-2 spike trimer, expressed in GnTI -293F cells. This spike construct, referred to as HPM7, is based on the original Wuhan strain and includes six proline mutations as well as an engineered interprotomer disulfide bond (48). To assess conformational changes in the spike glycoprotein induced specifically by the (R,R)-VE607 analog DY-III-281, we collected two cryo-EM data sets: one for the SARS-CoV-2 spike treated with DY-III-281, and a control ("apo") data set in which the spike was preincubated with dimethyl sulfoxide (DMSO) at the same concentration used to solubilize DY-III-281, but without the compound itself. Figure 6 schematically illustrates how both data sets were processed to obtain final maps representing the major conformational states adopted by the SARS-CoV-2 spike in each experimental sample. We also quantified the abundance of each confor mational class in the data set, calculated as the percentage of particles assigned to each class. In both the apo and DY-III-281-treated data sets, we identified three distinct SARS-CoV-2 conformations: closed (with all three RBDs down), open (with one RBD up), and intermediate (characterized by weak or undefined density for the RBD of one S1/S2 protomer). We proceeded with fitting and refinement of the closed and open conformations from each data set, obtaining structures at 3.14 Å and 3.09 Å for the apo SARS-CoV-2 spike, and 3.56 Å and 3.30 Å for the DY-III-281-treated spike, respectively (Table S1 and Fig. S3).
First, we conducted a thorough inspection of the densities for both the open and closed conformations of the DY-III-281-treated sample to determine whether the DY-III-281 compound could be detected bound to the SARS-CoV-2 spike. However, we were unable to identify any distinct additional density that could be confidently attributed to the compound. This suggests DY-III-281 binding is likely transient, or that the local resolution of the RBDs is insufficient to unambiguously resolve its binding site (Fig. S3). Given the absence of detectable compound-bound density, we next investiga ted the distribution of conformational states in our DY-III-281-treated and untreated (apo) samples. Although both data sets exhibited the same conformational classes, their relative abundances differed. In the apo spike data set, 28.5% of particles adop ted the closed conformation, compared to only 16.9% in the DY-III-281-treated data set. Conversely, the one-RBD-up open conformation accounted for 52.3% of particles in the apo sample (Fig. 6A) and 68.0% of particles in the DY-III-281-treated sample (Fig. 6B). Based on these observations, we propose that DY-III-281 either promotes the transition of the SARS-CoV-2 spike into the one-RBD-up conformation or stabilizes this conformation once adopted. These findings are also consistent with smFRET results in which a different (R,R)-VE607 analog shifted the conformational landscape to the RBD up conformation (Fig. 5), which is considered as energetically favorable upon substrate binding.
Finally, we superimposed the closed conformation (three RBD down) structures of the apo and DY-III-281-treated spike proteins to evaluate structural differences (Fig. 7). The analysis revealed no significant conformational deviation between the two states, with a root-mean-square deviation (RMSD) of 1.63 Å (Fig. 7C). Similarly, superposition of the open conformation (one RBD up) structures from the apo and DY-III-281-treated data sets yielded a Cα RMSD of 2.12 Å (Fig. 7D). To further investigate domain-specific dynam ics, we calculated domain-wise RMSD values. While the S2 domain remained relatively stable in both the closed (RMSD: 1.17 Å) and open state (RMSD: 1.26 Å), there was a notable increase in RMSDs corresponding to conformational change in the NTD and RBD domains. Next, we plotted the RMSD plot for each residue to highlight this change in the NTD and RBD regions (Fig. 7E andF). Although the S2 domain remained structurally rigid, the NTD and RBD domains exhibited significant conformational heterogeneity between samples.
## DY-III-281 delays virus spread and decreases viral load in organs from SARS-CoV-2 WA1 -challenged mice
Pharmacokinetic analysis was performed to determine DY-III-281 tissue distribution and concentration decay over time (Fig. S4). K18-hACE2 mice received DY-III-281 or vehicle (DMSO) intraperitoneally at doses of 12.5, 25, and 50 mg/kg, followed by blood collection at 0.5, 2, 8, and 24 hours of intervals and concentrations evaluated by mass spectrometry. In addition, DY-III-281 concentrations in organs, bronchoalveolar lavage Full-Length Text fluids (BALF), and nasal wash were evaluated after necropsy at 24 h. Thirty minutes after administering 25 mg/kg, we detected ~1.5 µM of DY-III-281 in plasma, but the levels waned at later time points. DY-III-281 levels were lower in lung, liver, kidney, and spleen compared to plasma, with minimal penetration in the brain (Fig. S4). Because DY-III-281 displayed short plasma retention in PK studies, we evaluated the in vivo efficacy by treating K18-hACE2 mice at 25 mg/kg once daily from day -1 to day 4 post-infection with SARS-CoV-2 WA1-nLuc (Fig. 8A). While all the mice in this DY-III-281-treated group succumbed to virus-induced mortality, we observed a significant delay in body weight (48). In contrast, in our apo spike data set, the majority of particles adopt the one-RBD-up open conformation, rather than the closed state observed by others. We speculate that this difference arises from our use of GnTI -293F cells instead of standard 293F cells, which may influence glycosylation and, consequently, conformational preferences. loss and survival was extended by 1 day compared to the control group (Fig. 8B). As is true for many viral infections, combination regimens targeting different steps of virus replication are required for efficient virus clearance. In our previous study, we had demonstrated that a combination of virus-directed (molnupiravir) and host-augmented (convalescent plasma with Fc-effector functions) therapy was successful when the monotherapies on their own failed in preventing SARS-CoV-2-induced mortality (49). In addition, we previously demonstrated that neutralization and Fc-effector functions can be uncoupled and still confer protection against SARS-CoV-2-induced mortality (14). We therefore investigated whether the entry inhibtion provided by DY-III-281 can be modulated by the enhanced Fc-effector functions (GASDALIE mutations) provided by a NTD-binding non-neutralizing antibody CV3-13. K18-hACE2 mice were prophylactically treated with engineered CV3-13 mAb variants carrying LALA (diminished Fc-effector function) or GASDALIE (enhanced Fc-effector function) mutations either alone or in combination with DY-III-281. As expected (14), prophylactic administration of CV3-13 LALA failed to confer protection, as body weight loss and mortality were comparable to those in isotype control-treated mice (Fig. 8B). Consistent with these findings, combining DY-III-281 with CV3-13 LALA did not improve survival or mitigate weight loss beyond the effect of DY-III-281 alone. Prophylaxis with CV3-13 GASDALIE significantly delayed body weight loss and death by 1-2 days compared to control groups of mice (Fig. 8B) (14). Furthermore, the combination of DY-III-281 with CV3-13 GASDALIE significantly reduced body weight loss and extended survival to 10 dpi and was more effective than either intervention alone. Viral load (flux quantification and N mRNA; Fig. 9A through D), inflammatory cytokine, and lung pathology marker mRNA expression at the time of death (Fig. 10A through C) corroborated these findings, demonstrating the superior efficacy of the DY-III-281 combination regimen with CV3-13 GASDALIE to other tested prophylactic administrations. While we observed significantly reduced viral titers in both the brain and lungs of treated mice, the reduction in brain viral loads was not sufficient to achieve complete protection from mortality in this susceptible model. Under these conditions, we were able to observe only delayed death for various treatment regimens. This limited improvement can be attributed to several factors. Based on our pharmacoki netic study results (Fig. S4), relatively low levels of DY-III-281 were detected in the brain compared to other organs. Additionally, viral persistence in the brain was still observed after treatment (Fig. 9), likely due to insufficient drug penetration across the blood-brain barrier. Since K18-hACE2 mice succumb primarily to brain infection, inadequate levels of drug in the brain may explain the modest protective effect despite a significant reduc tion in viral titers in lungs. Given these limitations of using survival as the sole endpoint, we performed disease burden calculations incorporating seven parameters and dimen sionality reduction using t-SNE plots (see Materials and Methods) to comprehensively assess treatment efficacy (Fig. 10C andD). This multi-parameter approach provides a more nuanced evaluation of treatment efficacy beyond the binary survival endpoint. Indeed, the calculation of the Bliss Index yielded a score of 5.8, indicating that the combined regimen of DY-III-281 and CV3-13-GASDALIE provided an additive benefit. The t-SNE plots corroborated this observation. Overall, our results demonstrate the potential of combining the antiviral activities of DY-III-281 with other prophylactic interventions to enhance in vivo efficacy.
## DISCUSSION
Small molecules have been used for years to fight viral infections. However, until now, no SARS-CoV-2 small-molecule drugs that directly target the SARS-CoV-2 Spike glycopro tein have been approved for therapeutic or prophylactic use in humans. VE607 is one small-molecule inhibitor that was originally developed to specifically target SARS-CoV-1 (40,41). We previously repurposed VE607 and reported its capacity to inhibit SARS-CoV-2 entry (42). VE607 inhibits the infection of SARS-CoV-2 original Wuhan strain and Omicron sublineages by stabilizing the "up" conformation of the RBD and preventing viral entry (42). In this study, we aimed at improving potency and breadth of VE607 and developed a series of (R,R)-VE607 derivatives that were next tested for neutralizing efficacy against representative SARS-CoV-2 variants. Three analogs, DY-III-281, DY-III-287, and DY-IV-048, showed similar or superior capacity to inhibit entry of pseudoviral particles carrying Spike glycoproteins from Omicron subvariants, including KP.3.1.1 and XEC, as well as authentic viruses. As seen previously for VE607, these analogs stabilized the Spike in the RBD "up" conformation as measured by smFRET (DY-IV-048, DY-III-281) and cryo-EM (DY-III-281). SARS-CoV-2 is an enveloped, positive-sense single-stranded RNA virus that enters host cells through a two-step process: initial attachment to the angiotensin-converting enzyme 2 (ACE2) receptor, followed by membrane fusion. The interaction between the RBD of the viral Spike glycoprotein and ACE2 triggers significant conformational rearrangements in the Spike, ultimately exposing the fusion peptide necessary for membrane fusion. Our smFRET and single-particle cryo-EM studies demonstrate that VE607 analogs stabilize the Spike in a conformation with one RBD in the "up" position.
While our cryo-EM data suggest that the interaction between VE607 analogs and the Spike is likely transient, we observe a clear increase in the proportion of Spike trimers adopting the one RBD-up conformation. A limitation of our study is that we were unable to localize the compound within the Spike density, likely due to both the transient nature of the interaction and the reduced resolution in the RBD and NTD regions of the cryo-EM maps. Despite this transient binding, we postulate that VE607 analogs act by stabilizing the SARS-CoV-2 Spike in a conformation that interferes with the large-scale structural rearrangements required for viral entry. K18-hACE2 transgenic mice are a highly susceptible model to test prophylactic or therapeutic interventions targeted against SARS-CoV-2 (29,50). We have previously used this model to demonstrate that neutralizing antibodies required both neutraliza tion and Fc-effector functions for effective SARS-CoV-2 clearance in vivo (20). We also observed that virus neutralization and Fc-effector functions could be separated, with each provided by a different antibody. When combined, these antibodies completely protected SARS-CoV-2-challenged mice even though neither was protective on their own (14). Here, we investigated the possibility of combining DY-III-281, a virus entry inhibitor, with CV3-13, a non-neutralizing antibody with enhanced Fc-effector functions. When combined, neither the entry inhibition by DY-III-281 nor the antibody binding of CV3-13 was altered (Fig. S1). Nevertheless, our multiparametric in vivo analyses indicated an additive benefit for DY-III-281 when combined with CV3-13 GASDALIE, supporting the potential of combination strategies to enhance the in vivo efficacy of next-generation DY-III-281 analogs.
Current SARS-CoV-2 Omicron subvariants harbor more than 60 mutations in their Spike compared to the original strain (51,52). These viruses remain a challenge to protect against re-infection (53-56). Several clinically approved mAbs against SARS-CoV-2 have now become obsolete due to resistant mutations in Spike (Imdevimab, Casirivimab, Evusheld [57,58]). Therefore, multiple tools of intervention are needed to fight the evolving SARS-CoV-2. With broad neutralizing capacity, VE607 analogs represent an additional tool of intervention against emerging SARS-CoV-2 VOCs.
## MATERIALS AND METHODS
Materials and methods have been previously reported (16,23,42,43,59) and are summarized below.
## Viruses
Authentic SARS-CoV-2 and XBB.1 were isolated, sequenced, and amplified from clinical samples obtained from infected patients by the Laboratoire de Santé Publique du Québec (LSPQ) and were previously described (60). The virus was sequenced by MinION technology (Oxford Nanopore Technologies, Oxford, UK). All work with the infectious SARS-CoV-2 authentic virus was performed in Biosafety Level 3 (BSL3) facilities at CRCHUM using appropriate positive-pressure air respirators and personal protective equipment. For in vivo experiments, we utilized SARS-CoV-2-WA1 expressing nanolucifer ase (SARS-CoV-2 WA1-nLuc ), which was obtained from Craig B Wilen, Yale University, and generously provided by K. Plante and Pei-Yong Shi, World Reference Center for Emerging Viruses and Arboviruses, University of Texas Medical Branch) (61,62). We propagated the virus by infecting Vero-E6-TMPRSS2 at a multiplicity of infection of 0.1. The culture supernatants were harvested after 72 h when cytopathic effects were clearly visible. To generate viral stocks, the supernatant was first cleared of cell debris by centrifuga tion and passed through a 0.45 micron filter. The viruses were then concentrated by mixing three parts of filtered supernatant with one part of cold (4°C) 4× PEG-it Virus Precipitation Solution (System Biosciences) and incubating the mixture overnight at 4°C. The precipitated virus was harvested by centrifugation at 1,500 × g for 60 min at 4°C and the resulting pellet resuspended in phosphate-buffered saline (PBS). Final stocks were aliquoted for storage at -80°C and estimation of virus titers by plaque assay. All work involving infectious SARS-CoV-2 was conducted within BSL3 and A-BSL3 facilities at Yale University School of Medicine following protocols approved by the IBSC and using appropriate protective equipment, including positive pressure air respirators.
## Plaque-forming assay
To determine the concentration of viral stocks, a standard plaque assay was performed. The procedure began by seeding Vero-E6 cells into 12-well plates at a density of 4 × 10 5 cells per well and allowing them to adhere for 24 hours. Following this incubation, the cell monolayers were infected with serial dilutions of the virus. An overlay consisting of 1 mL of 0.6% Avicel (RC-581 FMC BioPolymer) in pre-warmed complete RPMI medium (Thermo Fisher Scientific) was then added to each well. After 48 hours, the cells were fixed for 15 minutes with 10% paraformaldehyde. To visualize the plaques, a staining solution of 0.2% crystal violet in 20% ethanol (both from Sigma Aldrich) was applied for 1 hour. Finally, the plates were rinsed gently with water to reveal the plaques for quantification.
## Chemical synthesis: general information
All solvents were reagent or high-performance liquid chromatography (HPLC) grade. Anhydrous CH2Cl2 and tetrahydrofuran were obtained from the Pure SolveTM PS-400 system under argon atmosphere. All reagents were purchased from commercially available sources and used as received. Reactions were magnetically stirred under a nitrogen or argon atmosphere, unless otherwise noted, and reactions were monitored by thin-layer chromatography performed on pre-coated silica gel 60 F-254 plates (40-55 micron, 230-400 mesh) and visualized by UV light. Yields refer to chromatographically and spectroscopically pure compounds. Optical rotations were measured on a JASCO P-2000 polarimeter. Proton (1H) and carbon (13C) nuclear magnetic resonance (NMR) spectra were recorded on a Bruker Avance III 500-MHz spectrometer or a Bruker NEO600 600-MHz spectrometer. Chemical shifts (δ) are reported in parts per million (ppm) relative to chloroform (δ 7.26), or methanol (δ 3.31) for 1H NMR, and chloroform (δ 77.2) or methanol (δ 49.15) for 13C NMR. High-resolution mass spectra (HRMS) were recorded at the University of Pennsylvania Mass Spectroscopy Service Center on either a VG Micromass 70/70H or VG ZAB-E spectrometer. Analytical HPLC was performed with a Waters HPLC-MS system consisting of a 515 pump and Sunfire C18 reverse phase column (20 µL injection volume, 5 µm packing material, and 4.5 × 50 mm column dimensions) with detection accomplished by a Micromass ZQ mass spectrometer and 2996 PDA detector. Preparative-scale HPLC was carried out on a Waters AutoPurification system (Milford, MA) equipped with a 3100 mass detector, a 2767 sample manager, and a 2489 UV/visible detector. Purification was done on a 19 × 100 mm SunFire Prep C18 OBD 5 µm column using a binary solvent gradient with mobile phase A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile). HPLC-grade water and acetonitrile, as well as Optima liquid chromatography-mass spectrometry (LC/MS)-grade formic acid, were purchased from Fisher Scientific and used without further purification. Fraction collection was triggered using the mass detector. X-ray intensity data were collected on a Rigaku XtaLAB Synergy-S diffractometer equipped with an HPC area detector (Dectris Pilatus3 R 200K) and employing confocal multilayer optic-monochro mated Mo-Kα radiation (λ = 0.71073 Å) at a temperature of 100K. Preliminary indexing was performed from a series of thirty 0.5° rotation frames with exposures of 15 seconds. A total of 1,212 frames (nine runs) were collected employing ω scans with a crystal to detector distance of 34.0 mm, rotation widths of 0.5°, and exposures of 50 seconds. The purity of new compounds was judged to be >95% pure by NMR and LC-MS analysis, unless otherwise noted. Chemical synthesis is detailed in the supplemental material.
## X-ray crystallography
X-ray intensity data were collected on a Rigaku XtaLAB Synergy-S diffractometer equipped with an HPC area detector (Dectris Pilatus3 R 200K) and employing confocal multilayer optic-monochromated Mo-Kα radiation (λ = 0.71073 Å) at a temperature of 100K. Preliminary indexing was performed from a series of thirty 0.5° rotation frames with exposures of 15 seconds. A total of 1,212 frames (nine runs) were collected employing ω scans with a crystal to detector distance of 34.0 mm, rotation widths of 0.5°, and exposures of 50 seconds
## Cell viability test
To measure the cytotoxicity of VE607 and the derived compounds on 293T-ACE2 or Vero-E6 cells, a cell viability assay using CellTiter-Glo One Solution Assay (Promega) was performed. Briefly, 293T-ACE2 or Vero-E6 cells were seeded at a density of 10 4 cells/well in 96-well luminometer-compatible tissue culture plates (Perkin Elmer). After 24 h, indicated concentrations of (R,R)-VE607, DY-III-281, DY-III-287, or DY-IV-048 up to concentrations of 100 µM were added to the cells, followed by incubation for 48 h at 37°C, and the same volume of its vehicle, DMSO, was added as a control. Then a volume of CellTiter-Glo One Solution buffer equal to the volume of cell culture medium present in each well was added, followed by 2 minutes mixing on a shaker and 10 minutes incubation at room temperature. An LB941 TriStar luminometer (Berthold Technologies) was used to measure the luciferase activity of each well.
## Neutralization assay using pseudoviral particles
Target cells were infected with single-round luciferase-expressing lentiviral particles as described previously (43). Briefly, 293T cells were transfected by the calcium phos phate method with the lentiviral vector pNL4. were produced, as previously reported (45). 293T-ACE2 target cells were seeded at a density of 1 × 10 4 cells/well in 96-well luminometer-compatible tissue culture plates (Perkin Elmer) 24 h before infection. Pseudoviral particles in a final volume of 100 µL were incubated with the indicated concentrations of small molecules (R,R-VE607 or its derived compounds) up to concentrations of 100 µM for 1 h at 37°C and were then added to the target cells, followed by incubation for 48 h at 37°C. Cells were lysed by the addition of 30 µL of passive lysis buffer (Promega) followed by one freeze-thaw cycle. An LB941 TriStar luminometer (Berthold Technologies) was used to measure the luciferase activity of each well after the addition of 100 µL of luciferin buffer (15 mM MgSO4, 15 mM KPO4 [pH 7.8], 1 mM ATP, and 1 mM dithiothreitol) and 50 µL of 1 mM d-luciferin potassium salt (Prolume). The neutralization half-maximal inhibitory concentration (IC 50 ) represents the concentration to inhibit 50% of the virus infection of 293T-ACE2 cells.
## Microneutralization with authentic viruses
One day prior to infection, 2 × 10 4 Vero-E6 cells were seeded per well in the 96well flat-bottom plate and incubated overnight to permit cell adherence. Compounds dilutions ranged from 0, 0.316, 1, 3.16, 10, 31.6, and 100 µM were performed in a separate 96-well culture plate using Dulbecco's modified Eagle medium (DMEM) supplemented with penicillin (100 U/mL), streptomycin (100 µg/mL), HEPES, 0.12% sodium bicarbon ate, 2% fetal bovine serum (FBS), and 0.24% BSA. 10 4 TCID50/mL of SARS-CoV-2 virus was prepared in DMEM +2% FBS and combined with an equivalent volume of diluted compounds for 1 hour. After this incubation, all media was removed from the 96-well plate seeded with Vero-E6 cells, and a virus-compound mixture was added to each respective well at a volume corresponding to 600 TCID50 per well and incubated for one hour further at 37°C. Both virus-only and media-only (MEM +2% FBS) conditions were included in this assay. All virus-compound supernatant was removed from wells without disrupting the Vero-E6 monolayer. Each diluted compound (100 µL) was added to its respective Vero-E6 seeded well in addition to an equivalent volume of MEM +2% FBS and was then incubated for 48-72 hours. The media was then discarded and replaced with 10% formaldehyde for 24 hours to cross-link the Vero-E6 monolayer. Formaldehyde was removed from wells and subsequently washed with PBS. Cell monolayers were permeabilized for 15 minutes at room temperature with PBS + 0.1% Triton X-100, washed with PBS, and then incubated for 1 hour at room temperature with PBS + 3% non-fat milk. An anti-mouse SARS-CoV-2 nucleocapsid protein (Clone 1C7, Bioss Antibodies) primary antibody solution was prepared at 1 µg/mL in PBS + 1% non-fat milk and added to all wells for one hour at room temperature. Following extensive washing (3×) with PBS, an anti-mouse IgG HRP secondary antibody solution was formulated in PBS + 1% non-fat milk. One hour post-room temperature incubation, wells were washed with 3 × PBS, substrate (ECL) was added, and an LB941 TriStar luminometer (Berthold Technologies) was used to measure the signal of each well.
## Preparation of proteins for smFRET assays
Expression, purification, and fluorescent labeling of SARS-CoV-2 (Wuhan strain) ectodomain SΔTM trimers (SΔTM spikes) for smFRET experiments have been repor ted previously (46). Briefly, SΔTM hetero-trimers were expressed by co-transfection of ExpiCHO-S cells (Thermo Scientific, Waltham, MA, USA) with both the untagged SΔTM and A4-peptide-tagged SΔTM (amino acid positions 161 and 345) plasmids at a 2:1 molar ratio. SΔTM hetero-trimers were purified by affinity chromatography using nickel-nitrilo triacetic acid (Ni-NTA) agarose beads (Invitrogen, Waltham, MA, USA) and size exclusion chromatography (SEC), before being labeled by overnight incubation at room temper ature with coenzyme A (CoA)-conjugated LD550 and LD650 fluorophores (Lumidyne Technologies, New York, NY, USA) and Acyl carrier protein synthase (AcpS). SΔTM was purified away from unbound dye and AcpS by a second round of SEC. Aliquots were stored at -80°C until use. Expression and preparation of soluble human monomeric ACE2 (hACE2) for smFRET experiments have been described previously (46).
## smFRET imaging and data analysis
Fluorescent-labeled SARS-CoV-2 SDTM spikes prepared as described above were immobilized on streptavidin-coated quartz microscope slides by way of Ni-NTA-biotin (Sigma-Aldrich, St. Louis, MO, USA) and imaged using wide-field prism-based total internal reflection fluorescence microscopy as described as well (46,47,63,64). Imaging was performed in PBS containing 1 mM cyclooctatetraene (Sigma-Aldrich, St. Louis, MO, USA), 1 mM 4-nitrobenzyl alcohol (Sigma-Aldrich, St. Louis, MO, USA), 1 mM trolox (Sigma-Aldrich, St. Louis, MO, USA), 2 mM protocatechuic acid (Sigma-Aldrich, St. Louis, MO, USA), and 8 nM protocatechuate 3,4-deoxygenase (Sigma-Aldrich, St. Louis, MO, USA) to remove molecular oxygen and stabilize fluorescence. When indicated, 100 nM labeled SΔTM spikes were incubated with 600 nM of purified monomeric soluble human ACE2 (shACE2) as described (46), 312.5 nM (50 µg/mL) CV3-13 mAb, 50 µM DY-III-281, 50 µM DY-IV-048, or the indicated combination of them for 60 minutes before imaging. Concentrations of shACE2, CV3-13, DY-IV-048, DY-IV-048, and the indicated combina tions of them were maintained during imaging. smFRET data were collected using Micromanager v2.0 (65) at 25 frames/s. All smFRET data were processed and ana lyzed using the SPARTAN software package (https://github.com/stjude-smc/SPARTAN) in Matlab (Mathworks, Natick, MA) (66). smFRET traces were identified according to criteria previously described (46,47,63). Traces meeting those criteria were then verified manually. Traces from each of three technical replicates were then compiled into FRET histograms, and the mean probability per histogram bin ±standard error was calculated. Traces were idealized to a three-state HMM (two nonzero-FRET states and a 0-FRET state) using the maximum point likelihood algorithm (67) implemented in SPARTAN. The three-state model was previously selected by comparing the Akaike information criterion across multiple different models with a range of state numbers and topologies as described (46). The idealizations were used to determine the occupancies (fraction of time until photobleaching) in each FRET state and construct Gaussian distributions, which were overlaid on the FRET histograms to visualize the results of the HMM analysis. The distributions in occupancies were used to construct violin plots in Matlab, as well as calculate mean occupancy and standard errors, as displayed in Fig. 5. Statistical significance measures (P-values) of FRET state occupancies were determined by one-way ANOVA in Matlab (The MathWorks, Waltham, MA, USA). The analysis displays the full breadth of dynamic behavior across the total population of traces analyzed. The total number of traces analyzed was sufficient to ensure minimally 85% statistical power during comparison of occupancy data from unbound to ligand-bound SΔTM (46). P-values < 0.05 were considered to indicate statistical significance.
## Protein expression and purification for Cryo-EM
SARS-CoV-2 spike construct HPM7 (48) was kindly provided by Dr. Andrew Ward (The Scripps Research Institute). The stabilized, uncleaved Spike trimer (HPM7) is based on the Wuhan strain with six additional proline (hexaproline or HP) mutations (F817P, A892P, A899P, A942P, K986P, and V987P) (68) and an engineered interprotomer disulfide (mut7 or M7) between residues 705 and 883 of the S2 subunit. SARS-CoV-2 spike HPM7 was transiently expressed in 293F GNTI-cells (Thermo Fisher Scientific) using transfection reagent FectoPRO (Polyplus 116-010). Briefly, diluted 50 µg HPM7 plasmid and 75 µL FectoPRO were mixed and incubated at room temperature for 10 min. Then the transfection mixture was added to 90 mL of 293F GNTI -cells with cell viability of 95% and at a density of ~1 million cells/mL. 4 days after the transfection, superna tant was harvested and filtered with a 0.22 µm membrane. HPM7 protein was initially purified using Strep-Tactin XT resin (IBA Lifesciences) according to the manufacturer's instructions. Then it was purified further by SEC using a Superdex 200 10/300 Gl (GE Healthcare) column. Protein purity was checked by SDS-PAGE, and protein concentration was determined using a NanoDrop by measuring the absorbance at 280 nm.
## Cryo-EM sample preparation and data collection
A stock solution of the small-molecule DY-III-281 was prepared in 100% DMSO to a final concentration of 100 mM. For complex preparation, DY-III-281 was serially diluted into PBS and mixed with 1.5 mg/mL of SEC-purified SARS-CoV-2 spike in a 10-fold excess to a final concentration of 500 µM. The mixture was incubated at 4 degrees overnight before cryo-EM grids preparation. Grids (Cu R1.2/1.3, 400 mesh, Quantifoil) were glow-discharged for 60 s at 15 mA. 3 µL of the sample was applied onto the grid, the grid was blotted for 5 s, and then it was plunge-frozen into liquid ethane using a Leica EM GP2 plunge freezer with its chamber set to 4°C and 95% humidity. As a control, an apo SARS-CoV-2 spike sample was prepared by incubation with an equivalent concentration and volume of DMSO without the compound. The cryo-EM grid for the apo sample was prepared using the same parameters. Cryo-EM data were acquired on a FEI Titan Krios electron (G1) microscope operating at 300 keV equipped with Gatan Bioquantum Image filter-K3 direct electron detector (Gatan Inc.) with 20 eV energy slit. 50-frame dose-fractionated movies in super resolution mode were collected at a nominal magnification of 105,000 corresponding to a calibrated physical pixel size of 0.832 Å/px (0.416 Å/px super resolution), with a total exposure dose of 54.2 e -/ Å2 at a dose rate of 15 e -/px/s and a defocus range of -0.5 to -2.7 µm. Automated data acquisition was done in SerialEM version 4.0.27 (69). More details can be found in Table S1.
## Cryo-EM data processing
Cryo-EM data sets of DMSO and DY-III-281-treated samples were processed using CryoSPARC v4.6.0 (70). Drift correction of raw micrographs was performed using Patch Motion Correction, and the contrast transfer function (CTF) was estimated by Patch CTF Estimation. Particle picking was carried out using the template picker, and two rounds of 2D classification were applied to eliminate junk particles. High-quality 2D class averages were used for ab initio reconstruction with three classes followed by heterogeneous refinement. To distinguish the different states of SARS-CoV-2 spike protein, particle alignment was constrained during subsequent 3D classification and 3D refinement steps. The initial round of 3D classification was conducted using three classes with a target resolution of 6 Å. Particles contributing to structurally distinct conformations were analyzed using UCSF Chimera (71) and grouped based on conformation. Each subset, representing open, closed, or intermediate conformations, was processed independently through iterative rounds of 3D classification until well-segregated classes were obtained. 3D classes of particles with the highest quality and resolution were selected and subjected to 3D refinement. A similar workflow, incorporating reference-based iterative 3D classification, was applied for the DMSO-treated spike data set. Final maps were generated using Non-uniform Refinement, and C1 symmetry was applied throughout all processing steps. The overall resolution of each map was assessed using the gold-stand ard Fourier shell correlation 0.143 criterion. Final density maps used for model building were derived from the non-uniform refinement results.
## Atomic model building and refinement
Final density maps obtained from non-uniform refinement were used to build atomic models of the open and closed conformational states of the SARS-CoV-2 spike protein.
An initial fit to the open and closed state of Cryo-EM reconstruction was done using the cryo-EM structure of SARS-CoV-2-6P-Mut7 S protein complexed with CC6.33 IgG as a starting model (PDB ID 7RU3). To generate initial models, the CC6.33 IgG component was removed from the complex, retaining only the spike protein. Then only SARS-CoV-2-6P-Mut7 S protein was fitted into the corresponding cryo-EM maps using UCSF ChimeraX (72). To get the initial model of the closed state, the up-positioned RBD in the 7RU3 structure was truncated and reoriented to match the density corresponding to the down conformation, then merged back with the rest of the spike trimer to complete the closed-state model for refinement.
While the open and closed conformations maps obtained from DMSO-treated data sets were similar to those in the DY-III-281-treated data sets, the models derived from the DY-III-281 reconstructions did not dock well into the DMSO-treated cryo-EM maps. Therefore, for the DMSO-treated samples, initial models of the open and closed states were generated using the cryo-EM structure of the SARS-CoV-2 spike protein (PDB ID: 7N0G). After generating the initial models, these structures were then manually fitted into the reconstruction using COOT (73) and then refined with real-space refinement in PHENIX (74). Several rounds of refinement and model building were done to finalize the model for each of the conformational states. Data collection and refinement statistics are located in Table S1.
## Mouse experiments
All experiments were approved by the Institutional Animal Care and Use Committees and the Institutional Biosafety Committee (IBSC) of Yale University. All the animals were housed under specific pathogen-free conditions in the facilities provided and supported by Yale Animal Resources Center (YARC). hACE2 transgenic B6 mice (heterozygous) were obtained from Jackson Laboratory. 6-to 8-week-old male and female mice were used for all the experiments. The heterozygous mice were crossed and genotyped to select heterozygous mice for experiments using the primer sets recommended by Jackson Laboratory.
## Pharmacokinetics of DY-III-281
DY-III-281 or vehicle (DMSO) was administered intraperitoneally (i.p.) to male and female K18-hACE2 mice aged 6-8 weeks at doses of 50 mg/kg, 25 mg/kg, and 12.5 mg/kg, and the concentration in blood was measured 30 min, 2, 8, and 24 h after adminis tration. The mice were sacrificed at 24 h, and the indicated organs or washes (lung, brain, liver, spleen, kidney, nasal wash, or BALF) were isolated for mass spectrometry. Tissue was weighed, resuspended in serum-free RPMI, and homogenized using 1.5 mm Zirconium beads with BeadBug 6 homogenizer (Benchmark Scientific, TEquipment Inc.). To determine drug concentrations, the homogenized tissue was briefly centrifuged at 13,000 rpm at 4°C, and the clarified supernatant containing free drug was analyzed by mass spectrometry, and data were extrapolated to reflect the amount of drug per organ based on weight.
## Mass spectrometry
DY-III-281 mouse plasma and tissue levels were determined by ultra-HPLC (UHPLC) HRMS. Briefly, DY-III-281 was extracted from mouse plasma and tissue homogenate samples using protein precipitation. Five hundred microliter of internal standard (IS) solution (2.0 ng/mL of VE607 in methanol) was added to an aliquot of 10 µL of sample. The sample was vortexed for approximately 5 s and let stand for a period of 10 min, then centrifuged at 13,000 × g for 10 min. The supernatant was transferred to 13 × 100 mm borosilicate tubes and evaporated to dryness at 50°C under a gentle stream of nitrogen. The dried extract was re-suspended with 200 µL of 10:90 methanol: 10 mM ammonium formate pH 3 solution and transferred to an injection vial for analysis.
The analysis was performed using a Thermo Scientific Q-Exactive Plus Orbitrap Mass Spectrometer interfaced with the Thermo Scientific UltiMate 3000 XRS UHPLC system using a pneumatic-assisted heated electrospray ion source. MS detection was performed in positive ion mode operating in scan mode at high resolution and accurate mass (HR/AM). Nitrogen was used for the sheath and auxiliary gases and was set at 35 and 15 arbitrary units. The HESI electrode was set to 3,500 V. The capillary temperature was set at 350°C, and the vaporizer temperature was set at 400°C. The scan range was set to m/z 200-600. Data were acquired at a resolving power of 70,000 (FWHM), resulting in a scanning rate of <0.75 scans/s when using an automatic gain control target of 3.0 × 10 6 and a maximum ion injection time of 200 ms. Post-acquisition high-reso lution extracted ion chromatograms were generated using exact masses of targeted compounds ± 5 ppm.
Chromatographic separation was achieved using gradient elution with a Phenom enex Phenyl-Hexyl analytical column (150 × 2.0 mm I.D., 5 µm) operating at 50°C. The initial mobile phase condition consisted of methanol and 10 mM ammonium formate, pH 3, at a ratio of 10:90, respectively, and this ratio was maintained for 1 min. From 1 to 4.5 min a linear gradient was applied up to a ratio of 90:10 and maintained for 0.6 min. At 5.1 min, the mobile phase composition was reverted to 10:90 and the column was allowed to equilibrate for 4.9 min for a total run time of 10 min. The flow rate was fixed at 0.35 mL/min, and DY-III-281 and IS were eluted at 3.7 and 3.8 min, respectively.
Data acquisition and processing were performed using Thermo Scientific Xcalibur 4.2.47. Calibration curves were generated using weighted (1/x) linear regression of DY-III-281/IS peak-area ratios, with an analytical range of 10-20,000 ng/mL. Sample concentrations were interpolated from the standard curve. The method demonstrated acceptable precision and accuracy per bioanalytical standards.
## Mouse studies
All animal studies were conducted using 6-to 8-week-old male and female hetero zygous hACE2 transgenic B6 mice (K18-hACE2), which were originally sourced from The Jackson Laboratory (Bar Harbor, ME, USA). We established our own breeding colony and confirmed genotypes using the recommended primer sets. Experimental cohorts, composed of 4-8 animals, were assembled from randomly selected, sex-and age-matched littermates. The sample size for each group was justified by a priori power analysis using data from our previous work and pilot experiments (14,20,49). We made every effort to balance the number of male and female mice within experimental groups to mitigate sex as a confounding variable. No animals were excluded from the study post-procedure due to illness. All animals were maintained in the specific pathogen-free (SPF) barrier facility at the YARC with a 14:10 light-dark cycle. Animals infected with SARS-CoV-2 were housed in a dedicated BSL3 containment room, separate from the main breeding populations. All procedures involving infectious agents were performed under ABSL3 conditions, with personnel equipped with full protective gear, including pressurized air-purifying respirators, disposable gowns, and double gloves with shoe and sleeve covers. Decontamination and disposal of all animal-related materials followed the guidelines set by Yale University Environmental Health Services. The infection was initiated by intranasally delivering 1 × 10⁵ PFU of SARS-CoV-2 WA1-nLuc to anesthetized mice in a total volume of 25-30 µL. Anesthesia was administered with a precision Dräger vaporizer supplying 0.5%-5% isoflurane in oxygen. DY-III-281 was administered intraperitoneally (i.p. 25 mg/kg), 24 h prior to infection and daily until day 4 follow ing infection. Isotype, CV3-13 LALA, or CV3-13 GASDALIE antibody was administered once (i.p., 12.5 mg/kg), 24 h before infection. Post-infection monitoring involved daily tracking of body weight relative to the initial measurement (100%). For mortality analysis beginning on day 6, we performed welfare checks every 8-12 hours. Humane endpoints were established, requiring euthanasia for any mouse that lost over 20% of its body weight or displayed signs of severe illness or lethargy. These euthanized animals were included as mortalities in Kaplan-Meier survival analyses.
## Ex vivo bioluminescence analysis
The imaging workflow was conducted entirely post-mortem. Each animal at the time of euthanasia (humane endpoints; see above) was anesthetized using isoflurane and received a systemic dose of 100 µL furimazine substrate (diluted 1:40 in sterile PBS) through a retro-orbital route. Following substrate dissemination, the animals were euthanized, and the carcass was first imaged in its entirety (whole body) using an IVIS Spectrum instrument housed within an XIC-3 biocontainment chamber (PerkinElmer, Inc.). Subsequently, the lungs and brain were carefully dissected. These explanted organs then had 200 µL of the substrate solution applied directly to their surface and were incubated for 1-2 minutes before undergoing a second round of imaging. All images were captured and analyzed with Living Image software (v4.7.3). Acquisition settings were automated, with a luminescent f/stop of 2, a photographic f/stop of 8, and medium binning. Photon flux was quantified as luminescent radiance (photons/sec/cm²/sr), and a uniform luminescent scale was applied across all images for accurate comparison. Signal thresholds were established to exclude background radiance measured from areas devoid of tissue.
## Measurement of viral burden in tissues
To assess viral loads, brain and lung tissues were harvested from both infected (at the time of death, post necropsy) and uninfected mice and weighed. The collected organs were homogenized in 1 mL of serum-free RPMI with penicillin-streptomycin using a BeadBug 6 homogenizer (Benchmark Scientific) with 1.5 mm Zirconium beads. Total RNA was isolated from tissue homogenates with the RNeasy Plus Mini kit (Qiagen, Cat # 74136) and subsequently reverse transcribed into cDNA using the iScript advanced cDNA kit (Bio-Rad, Cat #1725036). The abundance of the SARS-CoV-2 N gene was then quantified via SYBR Green Real-time PCR with the following primers: N-Forward (5′-ATGC TGCAATCGTGCTACAA-3′) and N-Reverse (5′-GACTGCCGCCTCTGCTC-3′). The specificity of the PCR product was confirmed for all reactions using melt-curve analysis.
## mRNA expression analyses of signature inflammatory cytokines and lung injury/repair genes
To analyze gene expression changes in response to infection, brain and lung tissues were sampled at the time of death post-necropsy. Approximately 20 mg of tissue from each organ was placed in 500 µL of RLT lysis buffer for RNA isolation with the RNeasy Plus Mini kit (Qiagen, Cat # 74136). The purified RNA was converted to cDNA with the iScript advanced cDNA kit (Bio-Rad, Cat #1725036). We then quantified the mRNA transcripts of key inflammatory cytokines and lung pathology markers using multiplex qPCR. This was performed with the iQ Multiplex Powermix (Bio-Rad, Cat # 1725848) and PrimePCR Probe Assays for murine Gapdh, Il6, Ccl2, Cxcl10, Ifnɣ, Il1b, Krt8 (injury/repair marker), Krt5 (dysfunctional repair marker), Adamts4 (fibrosis marker), and Itga5 (damage-responsive fibroblasts marker). Reactions were run on a CFX96 touch system (Bio-Rad) under the following thermal cycling conditions: an initial denaturation at 95°C for 2 min, followed by 40 cycles of 95°C for 10 s and 60°C for 45 s. A final melt-curve analysis verified the amplification of a single product per primer set. Transcript levels of target genes were normalized to Gapdh mRNA expression (ΔCt), and the fold change relative to uninfected controls was determined using the 2⁻ ΔΔCt method.
## Estimation of disease burden, Bliss Index Scores, and visualization
We calculated a "disease burden" score to provide a holistic measure of pathology, as previously described (49,75). This composite score integrated measurements from brain and lung tissues, including viral loads (N mRNA copies), inflammatory cytokine, and pathology mRNA expression (Ccl2, Ifng, Cxcl10, Krt8), in the lung, and the extent of delayed death was also included, totaling seven parameters.
To calculate the score, each parameter was first normalized by setting the value from the control group (vehicle or isotype) to 100. A composite disease burden score was calculated by taking the mean of the seven normalized parameter values as shown in the equation below:
Total disease burden = % reduction N mRNA lung + N mRNA brain + Ccl2 mRNA lung + Cxcl10 mRNA lung + Ifng mRNA lung + Krt8 mRNA lung + % Mortality + (100 -% Delay in death)
To assess the nature of DY-III-281 and CV3-13 GASDALIE combination, we calculated Bliss index scores from the overall disease burden data as previously described (49,75). The score was calculated using the following equation to quantify the nature of the interaction:
Bliss index (DY-III-281 + CV3-13GASDALIE) = % Inhibition in disease burden (DY-III-281 + CV3-13GASDALIE)
-100 1 -1 -% Inhibition in disease burden (DY-III-281) 100
1 -% Inhibition in disease burden (CV3-13GASDALIE) 100
Based on the resulting Bliss index, interactions were classified as synergistic (score >10), additive (-10 ≤ score ≤ 10), or antagonistic (score < -10).
We also used the t-distributed stochastic neighbor embedding (t-SNE) algorithm to reduce the 7-dimensional data set to two dimensions for better visualization (Fig. 10D). Each parameter was z-score normalized across animals prior to dimensionality reduction. The analysis was performed in MATLAB R2024a using the tsne function with a perplexity of 8, a learning rate of 500, and random initialization.
## Quantification and statistical analysis
Statistics were analyzed using GraphPad Prism version 8.4 (GraphPad, San Diego, CA, USA). Every data set was tested for statistical normality, and this information was used to apply the appropriate (parametric or nonparametric) statistical test. Statistical details of experiments are indicated in the figure legends. P values < 0.05 were considered significant; significance values are indicated as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
## References
1. Dewolf, Laracy, Perales et al. (2022) "SARS-CoV-2 in immunocompromised individuals" *Immunity*
2. Tenforde, Self, Adams et al. (2021) "Association between mRNA vaccination and COVID-19 hospitalization and disease severity" *JAMA*
3. Hogan, Duerr, Dimartino et al. (2019) "Remdesivir resistance in transplant recipients with persistent coronavirus disease" *Clin Infect Dis*
4. Baang, Smith, Mirabelli et al. "Lauring AS. 2021. Prolonged severe acute respiratory syndrome coronavirus 2 replication in an immunocompromised patient" *J Infect Dis*
5. Choi, Choudhary, Regan et al. (2020) "Persistence and evolution of SARS-CoV-2 in an immunocompromised host" *N Engl J Med*
6. Who (2025) "COVID-19 Circulation"
7. Roemer, Sheward, Hisner et al. (2023) "SARS-CoV-2 evolution in the Omicron era" *Nat Microbiol*
8. Li, Faraone, Hsu et al. (2024) "Neutralization escape, infectivity, and membrane fusion of JN.1-derived SARS-CoV-2 SLip, FLiRT, and KP.2 variants" *Cell Rep*
9. Mccallum, Walls, Sprouse et al. (2021) "Molecular basis of immune evasion by the Delta and Kappa SARS-CoV-2 variants" *Science*
10. Shah, Canziani, Carter et al. (2021) "The case for S2: the potential benefits of the S2 subunit of the SARS-CoV-2 spike protein as an immunogen in fighting the COVID-19 pandemic" *Front Immunol*
11. Zhou, Song, Liu et al. (2023) "Broadly neutralizing anti-S2 antibodies protect against all three human betacoronaviruses that cause deadly disease" *Immunity*
12. Butler, Crowley, Natarajan et al. (2020) "Distinct features and functions of systemic and mucosal humoral immunity among SARS-CoV-2 convalescent individuals" *Front Immunol*
13. Gaebler, Wang, Lorenzi et al. (2021) "Evolution of Full-Length Text Journal of Virology November"
14. "antibody immunity to SARS-CoV-2" *Nature*
15. Beaudoin-Bussières, Chen, Ullah et al. (2022) "A Fc-enhanced NTDbinding non-neutralizing antibody delays virus spread and synergizes with a nAb to protect mice from lethal SARS-CoV-2 infection" *Cell Rep*
16. Beaudoin-Bussières, Finzi (2024) "Deciphering Fc-effector functions against SARS-CoV-2" *Trends Microbiol*
17. Ullah, Beaudoin-Bussières, Symmes et al. (2023) "The Fc-effector function of COVID-19 convalescent plasma contributes to SARS-CoV-2 treatment efficacy in mice" *Cell Rep Med*
18. Tauzin, Nayrac, Benlarbi et al. (2021) "A single dose of the SARS-CoV-2 vaccine BNT162b2 elicits Fc-mediated antibody effector functions and T cell responses" *Cell Host Microbe*
19. Winkler, Gilchuk, Yu et al. (2021) "Human neutralizing antibodies against SARS-CoV-2 require intact Fc effector functions for optimal therapeutic protection" *Cell*
20. Mackin, Sariol, Diamond (2024) "Antibody-mediated control mechanisms of viral infections" *Immunol Rev*
21. Ullah, Prévost, Ladinsky et al. (2021) "Live imaging of SARS-CoV-2 infection in mice reveals that neutralizing antibodies require Fc function for optimal efficacy" *Immunity*
22. Cao, Su, Guo et al. (2020) "Potent neutralizing antibodies against SARS-CoV-2 identified by high-throughput single-cell sequencing of convalescent patients' B Cells" *Cell*
23. Jennewein, Maccamy, Akins et al. (2021) "Isolation and characterization of cross-neutralizing coronavirus antibodies from COVID-19+ subjects" *Cell Rep*
24. Li, Chen, Prévost et al. (2022) "Structural basis and mode of action for two broadly neutralizing antibodies against SARS-CoV-2 emerging variants of concern" *Cell Rep*
25. Liu, Wang, Nair et al. (2020) "Potent neutralizing antibodies against multiple epitopes on SARS-CoV-2 spike" *Nature*
26. Rappazzo, Tse, Kaku et al. (2021) "Broad and potent activity against SARS-like viruses by an engineered human monoclonal antibody" *Science*
27. Suryadevara, Shrihari, Gilchuk et al. (2021) "Neutralizing and protective human monoclonal antibodies recognizing the N-terminal domain of the SARS-CoV-2 spike protein" *Cell*
28. Voss, Hou, Johnson et al. "2021. Prevalent, protective, and convergent IgG recognition of SARS-CoV-2 non-RBD spike epitopes" *Science*
29. Wu, Wang, Shen et al. (2020) "A noncompeting pair of human neutralizing antibodies block COVID-19 virus binding to its receptor ACE2" *Science*
30. Mccray, Pewe, Wohlford-Lenane et al. (2007) "Lethal infection of K18-hACE2 mice infected with severe acute respiratory syndrome coronavirus" *J Virol*
31. Dai, Zhang, Jiang et al. (2020) "Structure-based design of antiviral drug candidates targeting the SARS-CoV-2 main protease" *Science*
32. Kokic, Hillen, Tegunov et al. (2021) "Mechanism of SARS-CoV-2 polymerase stalling by remdesivir" *Nat Commun*
33. Batool, Chokkakula, Jeong et al. (2025) "SARS-CoV-2 drug resistance and therapeutic approaches" *Heliyon*
34. Beigel, Tomashek, Dodd (2020) "Remdesivir for the treatment of Covid-19 -preliminary report. reply" *N Engl J Med*
35. Grein, Myers, Brainard (2020) "Compassionate use of remdesivir in Covid-19. reply" *N Engl J Med*
36. Warren, Jordan, Lo et al. (2016) "Therapeutic efficacy of the small molecule GS-5734 against Ebola virus in rhesus monkeys" *Nature*
37. Cho, Saunders, Butler et al. (2012) "Synthesis and antiviral activity of a series of 1′-substituted 4aza-7,9-dideazaadenosine C-nucleosides" *Bioorganic & Medicinal Chemistry Letters*
38. Hammond, Leister-Tebbe, Gardner et al. "EPIC-HR Investigators. 2022. Oral nirmatrelvir for high-risk, nonhospitalized adults with Covid-19" *N Engl J Med*
39. Rawson, Donaldson, 'rear et al. (2025) "Independ ent FDA analyses of nirmatrelvir/ritonavir resistance in the phase 2/3 trials EPIC-HR and EPIC-SR" *Clin Infect Dis*
40. Mahase (2021) "Covid-19: Pfizer's paxlovid is 89% effective in patients at risk of serious illness, company reports" *BMJ*
41. Adedeji, Severson, Jonsson et al. (2013) "Novel inhibitors of severe acute respiratory syndrome coronavirus entry that act by three distinct mechanisms" *J Virol*
42. Kao, Tsui, Lee et al. (2004) "Identification of novel smallmolecule inhibitors of severe acute respiratory syndrome-associated coronavirus by chemical genetics" *Chem Biol*
43. Ding, Ullah, Gong et al. (2022) "VE607 stabilizes SARS-CoV-2 Spike in the "RBD-up" conformation and inhibits viral entry. iScience 25:104528"
44. Prévost, Gasser, Beaudoin-Bussières et al. (2020) "Cross-sectional evaluation of humoral responses against SARS-CoV-2 spike" *Cell Rep Med*
45. Amanat, White, Miorin et al. (2020) "An in vitro microneutralization assay for SARS-CoV-2 serology and drug screening" *CP Microbiology*
46. Benlarbi, Ding, Bélanger et al. (2024) "Temperature-dependent Spike-ACE2 interaction of Full-Length Text Journal of Virology November"
47. "Omicron subvariants is associated with viral transmission"
48. Díaz-Salinas, Li, Ejemel et al. (2022) "Conformational dynamics and allosteric modulation of the SARS-CoV-2 spike" *Elife*
49. Díaz-Salinas, Jain, Durham et al. (2024) "Single-molecule imaging reveals allosteric stimulation of SARS-CoV-2 spike receptor binding domain by host sialic acid" *Sci Adv*
50. Zhao, Keating, Ozorowski et al. (2022) "Engineering SARS-CoV-2 neutralizing antibodies for increased potency and reduced viral escape pathways"
51. Ullah, Escudie, Scandale et al. (2024) "Bioluminescence imaging reveals enhanced SARS-CoV-2 clearance in mice with combinatorial regimens"
52. Winkler, Bailey, Kafai et al. (2020) "SARS-CoV-2 infection of human ACE2transgenic mice causes severe lung inflammation and impaired function" *Nat Immunol*
53. Yuan, Wilson (2025) "Structural Immunology of SARS-CoV-2" *Immunol Rev*
54. Outbreak (2025) "SARS-CoV-2 (hCoV-19) Mutation Reports"
55. Cao, Wang, Jian et al. (2022) "Omicron escapes the majority of existing SARS-CoV-2 neutralizing antibodies" *Nature*
56. Cele, Jackson, Khoury et al. (2022) "Omicron extensively but incompletely escapes Pfizer BNT162b2 neutralization" *Nature*
57. Dejnirattisai, Huo, Zhou et al. (2022) "SARS-CoV-2 Omicron-B.1.1.529 leads to widespread escape from neutralizing antibody responses" *Cell*
58. Halfmann, Iida, Iwatsuki-Horimoto et al. (2022) "SARS-CoV-2 Omicron virus causes attenuated disease in mice and hamsters" *Nature*
59. Jaki, Weigang, Kern et al. (1999) "Total escape of SARS-CoV-2 from dual monoclonal antibody therapy in an immunocompromised patient" *Nat Commun*
60. Planas, Bruel, Staropoli et al. (2023) "Resistance of Omicron subvariants BA.2.75.2, BA.4.6, and BQ.1.1 to neutralizing antibodies" *Nat Commun*
61. Beaudoin-Bussières, Laumaea, Anand et al. (2020) "Decline of humoral responses against SARS-CoV-2 spike in convalescent individuals" *mBio*
62. Prévost, Richard, Gasser et al. (2021) "Impact of temperature on the affinity of SARS-CoV-2 Spike glycoprotein for host ACE2" *J Biol Chem*
63. Xie, Muruato, Lokugamage et al. (2020) "An infectious cDNA clone of SARS-CoV-2" *Cell Host Microbe*
64. Xuping, Muruato, Zhang et al. (2020) "A nanoluciferase SARS-CoV-2 for rapid neutralization testing and screening of anti-infective drugs for COVID-19" *Nat Commun*
65. Egri, Wang, Díaz-Salinas et al. (2023) "Detergent modulates the conformational equilibrium of SARS-CoV-2 Spike during cryo-EM structural determination" *Nat Commun*
66. Blakemore, Burnett, Swanson et al. (2021) "Stability and conformation of the dimeric HIV-1 genomic RNA 5'UTR" *Biophys J*
67. Edelstein, Tsuchida, Amodaj et al. (2014) "Advanced methods of microscope control using μManager software" *J Biol Methods*
68. Juette, Terry, Wasserman et al. (2016) "Single-molecule imaging of non-equilibrium molecular ensembles on the millisecond timescale" *Nat Methods*
69. Qin, Auerbach, Sachs (2000) "A direct optimization approach to hidden Markov modeling for single channel kinetics" *Biophys J*
70. Hsieh, Goldsmith, Schaub et al. (2020) "Structurebased design of prefusion-stabilized SARS-CoV-2 spikes" *Science*
71. Mastronarde (2005) "Automated electron microscope tomography using robust prediction of specimen movements" *J Struct Biol*
72. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
73. Pettersen, Goddard, Huang et al. (2004) "UCSF Chimera--a visualization system for exploratory research and analysis" *J Comput Chem*
74. Meng, Goddard, Pettersen et al. (2023) "UCSF ChimeraX: Tools for structure building and analysis"
75. Emsley, Lohkamp, Scott et al. (2010) "Features and development of Coot" *Acta Crystallogr D Biol Crystallogr*
76. Adams, Afonine, Bunkóczi et al. (2010) "PHENIX: a comprehensive Python-based system for macromo lecular structure solution" *Acta Crystallogr D Biol Crystallogr*
77. Ullah, Symmes, Keita et al. (2024) "Beta spike-presenting SARS-CoV-2 virus-like particle vaccine confers broad protection against other VOCs in mice" *Vaccines (Basel)* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584765&blobtype=pdf | # Epidemiology of respiratory syncytial virus in young, hospitalized children in Jordan: a prospective viral surveillance study
Justin Amarin, Haya Hayek, Olla Hamdan, Yasmeen Qwaider, Tala Khraise, Ahmad Khader, Qusai Odeh, Rami Salim, Hadeel Shalabi, Ahmad Alhajajra, Yousef Khader, Basim Al-Zoubi, Najwa Khuri-Bulos, Andrew Spieker, Leigh Howard, James Chappell, Natasha Halasa
## Abstract
Respiratory syncytial virus (RSV) is a leading cause of hospitalization in young children. Understanding RSV burden and seasonality is crucial for implementing effective preventive strategies, especially in the wake of disruptions related to the coronavirus disease 2019 (COVID-19) pandemic. We aimed to determine RSV burden and seasonality among young children hospitalized in Jordan. We conducted a prospective viral surveillance study at Al-Bashir Hospital (1 November 2023 to 4 April 2024). Children <5 years old hospitalized with fever or respiratory symptoms were eligible. Nasal (and optional throat) swabs were collected and tested for common respiratory viruses using real-time polymerase chain reaction. We compared characteristics and outcomes of hospitalizations by RSV detection status and assessed RSV seasonality. Of 2,610 children, RSV was detected in 713 (27.3%), making it the second most common virus overall and the most common in children <2 years old (n = 680 [30.0%]). Children with RSV were more likely than those without RSV to receive low-flow oxygen (74.9% vs 23.2%; P < 0.001) and high-flow nasal cannulation (3.2% vs 1.2%; P < 0.001) and were more likely to be admitted to the intensive care unit (13.2% vs 8.2%; P < 0.001). At least one other respiratory virus was co-detected with RSV in 244 children (34.2%). During the 2023-2024 season, RSV circulation exhibited a clear winter seasonality, consistent with historical patterns. In conclusion, the burden of RSV in children in Jordan remains substantial following the COVID-19 pandemic. The return to historical winter seasonal ity has important implications for the timing of preventive interventions. Continued surveillance is crucial for monitoring RSV epidemiology in this region. IMPORTANCE This study confirms the persistent and significant burden of respiratory syncytial virus (RSV) among young, hospitalized children in Jordan. Crucially, our data reveal the normalization of RSV circulation patterns in 2024 following disruptions related to the coronavirus disease 2019 pandemic. This finding has important implications for optimizing the timing of preventive interventions, such as monoclonal antibodies and maternal vaccination, particularly in a resource-limited setting where they are costly and limited in availability. By providing these contemporary surveillance data from the Eastern Mediterranean-where sentinel surveillance platforms are lacking-this work has the potential to inform public health strategies directly and emphasizes the critical need for sustained monitoring to guide effective RSV prevention and control efforts.
R espiratory syncytial virus (RSV) is the most frequent and one of the most virulent respiratory pathogens during early childhood. Within the first 24 months of life, most children will have been infected with RSV, half of them twice (1). In 2019, RSV accounted for an estimated 33 million episodes of acute lower respiratory tract infection, 95% of which occurred in low-and middle-income countries (2). The virus drives a substantial proportion of pediatric hospitalizations worldwide; for example, in the United States, acute bronchiolitis due to RSV is the leading cause of infant hospitalizations (3). While most infants who develop severe RSV disease are previously healthy, risk factors for severe disease include prematurity, young infancy, and underlying cardiopul monary, neurologic, and immunocompromising conditions (4,5). Therefore, preventive measures are necessary to safeguard young children-especially infants-from RSV and its complications.
In addition to everyday prevention (e.g., handwashing, breastfeeding), several pharmaceutical interventions are currently available to protect young children from severe RSV. These options include monoclonal antibodies (e.g., palivizumab, nirsevi mab) and maternal vaccination (e.g., bivalent RSV prefusion F subunit vaccine) (6)(7)(8). Palivizumab is a once-monthly monoclonal antibody recommended during the RSV season for infants at higher risk of severe disease but not healthy term infants or otherwise healthy preterm infants. Palivizumab prophylaxis is not considered high-value healthcare for any group of infants because its high cost is associated with minimal health benefits (6). Nirsevimab is a single-dose monoclonal antibody that is a much more potent inhibitor of RSV than palivizumab in vitro, has a substantially longer half-life, and is relatively less costly (6,7). Due to its favorable properties, nirsevimab prophylaxis is used more broadly in high-income countries: in the United States, for example, nirsevimab prophylaxis is recommended for infants <8 months old born during or entering their first RSV season and for children 8-19 months old at higher risk of severe disease entering their second season (7). Another approach, seasonal maternal vaccination, may also be used broadly; in the United States, it is recommen ded for pregnant persons during 32 through 36 weeks of gestation to prevent severe RSV disease in infants <6 months old (8). The effectiveness of these pharmaceutical interventions relies heavily on properly aligning the timing of administration with the seasonality of RSV circulation.
Despite their demonstrated efficacy, RSV monoclonal antibodies and maternal vaccines are costly and limited in availability, particularly in low-and middle-income countries, where they are needed most. As of April 2025, only palivizumab is available in Jordan. In a modeling study, Li et al. showed that seasonal dosing approaches in these supply-constrained settings may be more cost-effective and feasible (9). Therefore, in anticipation of broader access to these interventions and the approval and release of additional products, robust data on RSV epidemiology across diverse geographic settings are crucial, especially because the seasonality of RSV circulation varies by geographic and meteorologic factors (10). Building on our group's previous viral surveillance (2007, 2010-2013, and 2020) at the largest public hospital in Amman, Jordan-Al-Bashir Hospital-we launched a contemporary viral surveillance study in January 2023 for a duration of 16 months to determine the burden of RSV in young, hospitalized children, with the hypothesis that RSV remains the most frequent and virulent respiratory virus (11)(12)(13). In addition, we investigated potential disruptions to the seasonality of RSV following the coronavirus disease 2019 (COVID-19) pandemic compared with historical patterns.
## MATERIALS AND METHODS
## Study design and population
Between 11 January 2023 and 30 April 2024, we conducted a prospective viral surveil lance study of children <5 years old hospitalized in Al-Bashir Hospital, a large public hospital that provides care for at least 50%-60% of children in Amman, the capital of Jordan and its most populous city (12). According to 2023 estimates, Amman was home to 42.0% of Jordan's 11.5 million residents-more than double that of any of Jordan's governorates (12 in total) (14). Eligible children were those <5 years old at hospital admission who developed a fever or at least one respiratory symptom within the preceding 14 days and were screened within 72 h of admission. Respiratory symptoms qualifying for eligibility included cough, earache, runny nose or nasal congestion, sore throat, posttussive vomiting, wheezing, rapid or shallow breathing, chest retractions or abdominal breathing, stridor, apnea (including brief resolved unexplained events), and myalgia. We excluded newborns never discharged home after birth, children previously enrolled within the 14 days preceding the day of hospital admission, and those with a known nonrespiratory cause for hospitalization. Research staff screened and enrolled eligible children 6 days per week (Saturday through Thursday), excluding public holidays and observances.
## Study procedures
After obtaining written informed consent from a parent or legal guardian, bilingual research staff administered an English-language standardized questionnaire to the parent or legal guardian. The staff was trained to convey questionnaire content verbally in Arabic, ensuring an accurate representation of the original English items. Following discharge, research staff abstracted additional data from electronic health records using a standardized case report form that included sections on health status at admission (e.g., results of chest radiography), review of underlying medical conditions, clinical course (e.g., intensive care unit admission, respiratory support, antibiotic use, and in-hospital mortality), and bacterial cultures and viral testing.
In addition, research staff collected an anterior nasal flocked swab from the child with or without a throat swab. The nasal swab, combined with the throat swab (if collected), was placed into a single vial containing PrimeStore Molecular Transport Medium or UTM Universal Transport Medium, stored at 4°C for a maximum of 28 days (PrimeStore) or -80°C (UTM), and shipped to Vanderbilt University Medical Center on frozen cold packs (PrimeStore) or dry ice (UTM). For a single participant, a throat swab was collected without an accompanying nasal swab. This specimen was processed and analyzed in the same manner as other specimens.
Laboratory staff at Vanderbilt University Medical Center tested specimens by singleplex reverse transcription real-time polymerase chain reaction assays for the following respiratory viruses: RSV; influenza virus types A, B, and C; parainfluenza virus (PIV) types 1-4; human metapneumovirus (HMPV); human rhinovirus (HRV); enterovirus D68 (EV-D68); adenovirus (AdV); common cold coronavirus (ccCoV) species 229E, HKU, NL63, and OC43; and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Notably, the polymerase chain reaction assay for HRV is cross-reactive with the genet ically similar enteroviruses. Therefore, while we use the shorthand "HRV" to denote positive results, we acknowledge that some detections may represent enteroviruses.
## Statistical methods
We summarized categorical variables using absolute and relative frequencies and continuous variables using median and interquartile range. We used logistic regres sion for binary outcomes and linear regression for continuous outcomes to compare characteristics of children by RSV detection status, with RSV detection status as the predictor. Cluster-robust standard errors were used to account for children who enrolled more than once.
To assess the associations between RSV detection status and in-hospital outcomes, we used generalized estimating equations with a working independence structure, adjusting for well-established risk factors for severe RSV illness: namely, age at admission (fitted using restricted cubic splines with three knots), presence of at least one underly ing medical condition, and prematurity status.
Statistical significance was determined to be achieved based on a nominal threshold of α = 0.05 (two-sided, where applicable). All analyses were performed using R (version 4.4.1).
## RESULTS
## Study participants
Between 8 January 2023 and 30 April 2024, 9,252 children <18 years old were admitted to Al-Bashir Hospital, averaging 19.3 daily admissions (median, 19; range, 4-43) over 479 days. Of all children, 6,918 were <5 years old, representing 74.8% of the pedia tric caseload. Our research staff began screening on 11 January 2023 and conducted screening activities on 393 (82.6%) of the 476 possible days. Research staff did not conduct screening on scheduled days off: Fridays (67 days), national holidays (15 days), and a general strike (1 day). The flow of study participants is shown in Fig. 1. Briefly, 3,128 children were provisionally eligible, 2,622 (83.8%) were enrolled, and 2,615 (99.7%) comprised the study population. To construct our analytic sample, we excluded children with no respiratory specimen available for molecular diagnostics, yielding an analytic sample size of 2,610. The sample included 2,408 children, with 2,238 enrolled once (n = 2,238 observations; 85.7%), 142 enrolled twice (n = 284 observations; 10.9%), 24 enrolled three times (n = 72 observations; 2.8%), and 4 enrolled four times (n = 16 observations; 0.6%). Throughout the article, we use the term "children" as shorthand to refer to discrete hospitalizations, while accounting for clustered observations in all statistical comparisons.
## Respiratory virus detections
Of the 2,610 children, 2,121 (81.3%) were positive for at least one respiratory virus (Table S1). The most common virus was HRV (n = 733 [28.1%]), followed by RSV (n = 713
## Seasonality of RSV
During the study period, RSV was detected in a seasonal pattern, peaking in the winter of 2024 (Fig. 2a). RSV detections during the midwinter and late winter of 2023 appeared to have been trailing off, suggesting that the peak had occurred in late 2022. Indeed, a plot of the monthly circulation of RSV per annum in the same setting (using historical surveillance data for reference; Fig. 2b) showed that RSV circulation in the first quarter of 2023 was atypical, whereas the circulation of RSV during 2024 was consistent with historical patterns (11)(12)(13).
## Baseline characteristics
We compared baseline characteristics of children with or without RSV (Table 1). Briefly, children with RSV tended to be younger (median, 2.9 months vs 5.7 months; P < 0.001) and were less likely to be male (52.2% vs 56.9%; P = 0.028), less likely to have at least one underlying medical condition (21.2% vs 30.7%; P < 0.001), more likely to be breastfed at the time of enrollment (61.0% vs 47.5%; P < 0.001), less likely to attend daycare or preschool (2.4% vs 4.0%; P = 0.047), less likely to be exposed to any tobacco-related smoke (80.8% vs 84.3%; P = 0.030), and more likely to receive antibiotics prior to hospitalization (38.0% vs 33.6%; P = 0.037). Overall, 5 out of 2,594 children (0.2%) received palivizumab prior to hospitalization.
## Symptoms and diagnostics
The burden of symptoms at or prior to hospital admission tended to be higher among children with RSV, with some exceptions. Of the 21 signs and symptoms we evaluated, the frequencies of 10 were significantly higher among children with RSV, while the frequencies of 5 were significantly lower (Fig. 3). Notably, 13 out of 2,610 children (0.5%) were clinically tested for RSV prior to or within the first 72 h of admission (Table 1).
## In-hospital outcomes
Children with RSV were more likely than those without RSV to receive antibiotics during hospitalization (97.6% vs 94.3%; P < 0.001), low-flow oxygen (74.9% vs 23.2%; P < 0.001), and high-flow nasal cannulation (3.2% vs 1.2%; P < 0.001). They were also more likely to be admitted to the intensive care unit (13.2% vs 8.2%; P < 0.001; Table 1). We present select in-hospital outcomes among subgroups of children defined more granularly by detection status in Table 2, and we present the adjusted associations between RSV detection status and in-hospital outcomes in Table 3. The results of these adjusted analyses were consistent with those of the unadjusted analyses. The odds of antibiotic use during hospitalization were 2.03 higher (95% confidence interval [95% CI], 1.20-3.41) among children with RSV than those without. The odds of low-flow oxygen were also significantly higher (adjusted odds ratio [aOR], 10.0; 95% CI, 8.2-12.4) among children with RSV than those without, as were the odds of high-flow nasal cannula (aOR, 2.60; 95% CI, 1.44-4.69) and intensive care unit admission (aOR, 1.65; 95% CI, 1.25-2.17).
## DISCUSSION
In this prospective viral surveillance study conducted between 11 January 2023 and 30 April 2024, at Jordan's largest public hospital, RSV was the second most common viral pathogen among children <5 years old hospitalized with fever or respiratory symptoms and the most common among those <2 years old, especially during infancy. Children with RSV, who represented nearly one-third of the study population, generally had a more severe course of illness than children without RSV, as indicated by higher proportions of respiratory support and intensive care utilization. Co-detection of RSV with other respiratory viruses-particularly HRV-was common, occurring in roughly one-third of children with RSV. Importantly, our surveillance uniquely captured the apparent normalization of RSV circulation patterns in 2024 following disruptions related to the COVID-19 pandemic, with winter peaks aligning with historical seasonal trends observed in this setting in the past.
The substantial burden of RSV we observed among hospitalized, young children in Amman, Jordan, concords with global estimates highlighting RSV as the leading cause of acute lower respiratory infections in this age group (2). Our finding that children with RSV required respiratory support and intensive care at higher proportions than their counterparts without RSV echoes findings from other settings, including our prior viral surveillance studies in Amman, Jordan (11,12,(15)(16)(17)(18)(19). In a viral surveillance study conducted in Nashville, TN, Haddadin et al. evaluated the severity of RSV in children <5 years old with fever or respiratory symptoms and found that 61.5% and 16.9% required supplemental oxygen and intensive care, respectively, consistent with our results (74.9% and 13.2%, respectively) (15). In both cohorts, the clinical presentation of RSV was characterized by higher proportions of lower respiratory symptoms-particularly rapid or shallow breathing, wheezing, and chest retractions-consistent with RSV's predilec tion for the lower airways (17). Despite this characteristic presentation and severity profile, we found that clinical testing for RSV was rarely performed (only 0.5% of children were tested). This lack of routine testing likely contributes to diagnostic uncertainty and may explain, at least partially, the near-ubiquitous use of in-hospital antibiotics. In complementary analyses of this and prior surveillance cohorts at Al-Bashir Hospital, we showed that over 90% of children received antibiotics despite viral etiologies predomi nating (20). In low-resource settings, the downstream impact of such diagnostic gaps is amplified, fueling widespread antibiotic use and contributing to antimicrobial resist ance. Enhancing rapid diagnostic capacity is, therefore, essential not only for optimiz ing individual patient management and promoting antimicrobial stewardship but also for robustly monitoring the ongoing effectiveness of early preventive interventions. Moreover, the young age distribution of children in whom RSV was detected in our study, with most in infancy, is congruent with a well-established vulnerability window and supports current recommendations for early preventive interventions, including maternal vaccination during pregnancy and nirsevimab administration during early life (6)(7)(8).
Our observation that RSV circulation in Amman has returned to its historical winter seasonality following pandemic-related disruptions mirrors emerging global patterns and has important implications for the timing of preventive interventions. A recent multinational analysis demonstrated that while initial RSV epidemics following the COVID-19 pandemic exhibited unusual temporality, subsequent waves progressively resembled typical RSV seasonality across geographic settings, albeit with little represen tation from the Eastern Mediterranean region, where sentinel surveillance platforms are lacking (21). In Jordan-and other resource-limited settings-understanding these consistent seasonal patterns is crucial for optimizing the timing of preventive measures (9). The close monitoring we performed of circulation onset, peak, and offset will be particularly valuable for informing the delivery of preventive interventions as global RSV dynamics continue to stabilize. Beyond seasonality, we also observed a high proportion of viral co-detection, mirroring findings from other recent studies. For instance, our (16). While the high proportion of viral co-detection raises important considerations about the interpretation of positive respiratory viral tests during peak RSV season, the clinical significance of RSV co-detec tion remains to be fully elucidated (22). Nevertheless, these cases highlight the complex viral ecology that characterizes pediatric respiratory infections during winter months and emphasize the need for ongoing research. While global data highlight the evolving dynamics of RSV, studies conducted within Jordan by other research groups consistently reinforce the substantial burden and characteristic seasonality observed in our investigation. Notably, investigations of RSV in Jordan date back as early as 1993 (23), providing a long-term perspective on the epidemiology of the respiratory virus in the country. Earlier studies at hospitals in Amman, including Al-Bashir Hospital, prior to the COVID-19 pandemic, reported similar RSV detection proportions among hospitalized children with respiratory tract infections, as well as a comparable winter seasonality (24,25). These findings are further groun ded by similar results from studies performed in northern Jordan and another governo rate in central Jordan, suggesting a high degree of consistency in RSV epidemiology across different regions of the country (23,26,27). Studies encompassing periods of COVID-19-related disruption in circulation-with some relying on clinical testing rather than systematic surveillance-provide additional context (28,29), and a more recent multicenter surveillance study conducted between November 2022 and April 2023 using a geographically representative sample supports the general burden of RSV characterized herein (30). Importantly, many of these studies, particularly those utilizing clinical data, help to fill temporal gaps in our continual surveillance efforts (28)(29)(30)(31). The consistency of findings across multiple studies, spanning various periods (from 1993 to 2023) and geographic settings (from the north to the south of Jordan), strengthens the validity of our findings and underscores the persistent public health challenge posed by RSV in Jordan. Collectively, these findings highlight the urgent need to implement effective RSV prevention measures to reduce the burden of RSV-related hospitalizations, especially given that medical care for children <6 years old in Jordan is government-funded. Our findings provide inputs for evaluating the cost-effectiveness of preventive measures such as nirsevimab and maternal vaccination. Although formal economic analyses are beyond the scope of this study, modeling work has shown that aligning administration with local seasonality can improve feasibility and value in resource-limited settings (9). Companion cost-effectiveness studies are warranted to fully evaluate the benefits of such measures.
Our study has several strengths. First, we provide robust viral surveillance data from the Eastern Mediterranean region, where sentinel surveillance platforms are lacking and contemporary RSV circulation dynamics are poorly understood. Second, our research group's history of conducting similarly designed studies at Al-Bashir Hospital since 2007 provides valuable context to assess current RSV circulation patterns in Jordan, fortifying our ability to identify true deviations from typical patterns. Our study also has several limitations. First, as a single-center study, our findings may not fully represent RSV circulation dynamics throughout Jordan, though Al-Bashir's large catchment area and Jordan's meteorologic homogeneity reasonably permit generalization to urban areas (as corroborated by other studies in diverse settings across Jordan). Nevertheless, RSV circulation in rural regions might differ. Second, while our historical data span multi ple years, they are discontinuous, potentially missing important transitional periods. However, the consistency we observed in RSV seasonality across discrete study periods lends credibility to our characterization of "typical" dynamics in Jordan and supports our interpretation of post-pandemic normalization.
In conclusion, our findings highlight the importance of sustained viral surveillance in regions bearing a disproportionate public health burden from RSV, particularly as new preventive interventions become available. While the normalization of RSV circulation following pandemic-related disruptions suggests a return to predictable seasonality, continued monitoring remains crucial for optimizing the timing of preventive strategies.
| Leigh M. Howard, Writing -review and editing | James D. Chappell, Project administra tion, Validation, Writing -review and editing | Natasha B. Halasa, Conceptualization, Funding acquisition, Project administration, Supervision, Writing -review and editing
## ETHICS APPROVAL
The Vanderbilt University Institutional Review Board and the Jordanian Ministry of Health Institutional Review Board reviewed and approved the study protocol. Parents or legal guardians provided written informed consent.
## References
1. Glezen, Taber, Frank et al. (1986) "Risk of primary infection and reinfection with respiratory syncytial virus" *Am J Dis Child*
2. 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*
3. Suh, Movva, Jiang et al. (2022) "Respiratory syncytial virus is the leading cause of united states infant hospitalizations, 2009-2019: a study of the national (nationwide) inpatient sample" *J Infect Dis*
4. Hall, Weinberg, Blumkin et al. (2013) "Respiratory syncytial virus-associated hospitaliza tions among children less than 24 months of age" *Pediatrics*
5. Halasa, Zambrano, Amarin et al. (2023) "Infants admitted to US intensive care units for RSV infection during the 2022 seasonal peak" *JAMA Netw Open*
6. Caserta, Leary, Munoz et al. "Committee on Infectious Diseases. 2023. Palivizumab prophylaxis in infants and young children at increased risk of hospitalization for respiratory syncytial virus infection" *Pediatrics*
7. Jones, Ke, Prill et al. (2023) "Use of nirsevimab for the prevention of respiratory syncytial virus disease among infants and young children: recommendations of the advisory committee on immunization practices -United States" *Morbid Mortal Wkly Rep*
8. 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" *Morbid Mortal Wkly Rep*
9. Li, Hodgson, Wang et al. (2021) "Respiratory syncytial virus seasonality and prevention strategy planning for passive immunisation of infants in low-income and middle-income countries: a modelling study" *Lancet Infect Dis*
10. Li, Wang, Broberg et al. (2022) "Seasonality of respiratory syncytial virus and its association with meteorological factors in 13 European countries" *Euro Surveill*
11. Khuri-Bulos, Williams, Shehabi et al. (2010) "Burden of respiratory syncytial virus in hospitalized infants and young children in Amman" *Jordan. Scand J Infect Dis*
12. Halasa, Williams, Faouri et al. (2015) "Natural history and epidemiology of respiratory syncytial virus infection in the Middle East: hospital surveillance for children under age two in Jordan" *Vaccine (Auckland)*
13. Hamdan, Stopczynski, Stahl et al. (2023) "Patterns of antibiotic use in young children hospitalized with RSV in Amman" *Jordan. J Pediatric Infect Dis Soc*
14. Haddadin, Beveridge, Fernandez et al. (2021) "Respiratory syncytial virus disease severity in young children" *Clin Infect Dis*
15. Hayek, Amarin, Qwaider et al. (2023) "Codetection of respiratory syncytial virus with other respiratory viruses across all age groups before and during the COVID-19 pandemic" *Front Virol*
16. García, Bhore, Soriano-Fallas et al. (2010) "Risk factors in children hospitalized with RSV bronchiolitis versus non-RSV bronchiolitis" *Pediatrics*
17. Amini, Gilca, Boucher et al. (2019) "Respiratory syncytial virus contributes to more severe respiratory morbidity than influenza in children < 2 years during seasonal influenza peaks" *Infection*
18. Mathisen, Strand, Sharma et al. (2010) "Clinical presentation and severity of viral community-acquired pneumonia in young Nepalese children" *Pediatr Infect Dis J*
19. Hayek, Amarin, Hamdan et al. (2010) "Jordan Viral Surveillance Studies Group"
20. Thindwa, Li, Cooper-Wootton et al. (2024) "Global patterns of rebound to normal RSV dynamics following COVID-19 suppression" *BMC Infect Dis*
21. Amarin, Halasa, Rebeiro (2024) "Selection bias may compromise our understanding of the clinical significance of the co-detection of respiratory viruses" *Microbiol Spectr*
22. Meqdam, Rawashdeh, Masaadeh et al. (1998) "Respiratory syncytial virus infection in infants hospitalized with respiratory illness in northern Jordan" *J Trop Pediatr*
23. Al-Toum, Bdour, Ayyash (2006) "Epidemiology and clinical characteristics of respiratory syncytial virus infections in Jordan" *J Trop Pediatr*
24. Biggs, Simões, Khader et al. (2023) "Respiratory syncytial virus infection among hospitalized infants in four middleincome countries" *J Pediatric Infect Dis Soc*
26. Awad, Khader, Mansi et al. (2016) "Viral surveillance of children with acute respiratory infection in two main hospitals in northern Jordan" *J Pediatr Infect Dis*
27. Bdour (2001) "Respiratory syncytial virus subgroup A in hospitalized children in Zarqa" *Jordan. Ann Trop Paediatr*
28. Al-Zayadneh, Khraisat, Musa et al. (2024) "Exploring the epidemiolog ical burden of RSV pre-and post-COVID-19 pandemic: a Jordanian tertiary hospital experience" *J Int Med Res*
29. Al-Iede, Alhouri, Marwa et al. (2024) "Respiratory syncytial virus in pediatric patients admitted to a tertiary center in Amman: clinical characteristics, and age-related patterns" *BMC Pediatr*
30. Abu-Helalah, Sf, Lubad et al. (2024) "The epidemiology, clinical, and economic burdens of respiratory syncytial virus infections amongst hospitalized children under 5 years of age in Jordan: a national multi-center cross-sectional study" *Viruses*
31. Khasawneh, Himsawi, Sammour et al. (2024) "Molecular characteri zation of human respiratory syncytial virus strains circulating among hospitalized children in Jordan" *BMC Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12548383&blobtype=pdf | # High seroprevalence and high risk: why are older adults more prone to respiratory syncytial virus?
Piotr Rzymski, Barbara Poniedziałek, Dorota Zarębska-Michaluk, Krzysztof Tomasiewicz, Robert Flisiak
## Abstract
Despite widespread seropositivity, respiratory syncytial virus (RSV) remains a major cause of severe illness in adults aged 60 years and older. This review exam ines why infection-acquired immunity fails to protect this group, focusing on four key factors: structural lung decline, comorbidities, immunosenescence, and impaired antibody responses. Age-related changes weaken mechanical defenses and antiviral immunity, while chronic diseases amplify RSV risk. Critically, repeated RSV infections may preferentially boost non-neutralizing antibodies targeting the postfusion F protein, limiting protection and possibly enhancing disease. The review also highlights how newly approved vaccines, based on stabilized prefusion F protein, can overcome these barriers by inducing strong neutralizing responses, offering a targeted strategy to reduce RSV burden in older adults.
R espiratory syncytial virus (RSV), classified in the family Pneumoviridae, is an enveloped, negative-sense, single-stranded RNA virus first isolated in 1956 from chimpanzees and, 1 year later, from children with bronchiolitis (1). Its characteristic ability to induce syncytium formation via the viral fusion (F) protein gave the pathogen its name. RSV circulates globally in predictable winter epidemics, infecting nearly all children by 2 years of age and reinfecting individuals throughout life. Antigenically, RSV is divided into two principal subgroups, A and B, defined by their G glycoprotein. Each consists of multiple, continuously evolving clades that co-circulate within a given season (2,3). Although subgroup A strains generally predominate, both subgroups can circulate during outbreaks and contribute substantially to the overall disease burden, with current evidence showing no consistent differences in clinical severity between them (4-7).
RSV is among the leading causes of acute lower respiratory tract infection (LRTI) across the lifespan. Moreover, its reinfections are common and can occur within the same epidemic season (8)(9)(10)(11). In 2019, the virus was estimated to cause more than 30 million LRTI episodes in children younger than 5 years worldwide, resulting in approximately 60,000 deaths (12,13). A second, often under-recognized, burden peak is observed in adults aged 60 years and older. In the United States alone, RSV accounts for 123,000 to 193,000 hospitalizations and more than 10,000 deaths annually in this age group (14)(15)(16)(17), with comparable age-specific morbidity documented in Europe (18) and Asia (19). The total annual costs of healthcare related to RSV in the United States were estimated to exceed $1 billion (20). Significantly, RSV disproportionatelyy affects individuals in low-to middle-income countries due to limited access to healthcare services, numerous clinical, socioeconomic, and environmental risk factors, and increased morbidity of HIV infection, which constitutes an additional severity risk (21)(22)(23)(24).
High-risk groups for severe RSV disease include preterm infants, children with chronic lung or congenital heart disease, and immunocompromised individuals (25).
Contemporary surveillance, however, identifies older adults, especially those with underlying health conditions, as an equally vulnerable population (26,27). The disease burden in hospitalized older adults can exceed or be similar to that reported for nonpandemic influenza A, particularly among residents of long-term care facilities or individuals with chronic cardiopulmonary disease (28)(29)(30).
Although RSV was identified as a pathogen in 1957, progress toward an effective vaccine has been hindered by multiple setbacks (31,32). The most notable failure occurred in the 1960s with a formalin-inactivated RSV vaccine, which unexpectedly caused enhanced respiratory disease in vaccinated children upon natural infection (33,34). Approximately 80% of vaccine recipients required hospitalization, and two fatalities were reported. Subsequent studies demonstrated that formalin inactivation converted the F protein from its native prefusion to a postfusion conformation, exposing non-neu tralizing epitopes (35). Therefore, the vaccine elicited antibodies with poor neutralizing capacity that contributed to antibody-dependent enhancement (ADE) of disease (36,37). This experience delayed RSV vaccine research for decades.
Consequently, RSV management relied primarily on supportive care. Passive immunoprophylaxis with the monoclonal antibody palivizumab (approved in 1998) offered seasonal protection for a narrow subset of high-risk infants. It was impractical for widespread use because of its cost and monthly dosing schedule (38,39). A significant advance came with the regulatory approval of nirsevimab in 2023 and clesrovimab in 2025 for infants and young children. These recombinant monoclonal antibodies target a conserved epitope on the F protein, have an extended serum half-life, and are adminis tered intramuscularly as a single dose to protect for an entire RSV season (40)(41)(42)(43)(44).
For active immunization in adults, including older people, a pivotal breakthrough occurred in 2013 when stabilized prefusion structures were resolved, revealing neutralization-sensitive epitopes previously obscured in postfusion antigens that had dominated earlier vaccine efforts (45). Antigenic sites I-IV are present in both the pre-and postfusion F structures. In contrast, antigenic sites Ø and V (also known as site VIII) only occur in the prefusion conformation (45)(46)(47). Less than a decade after crystal structures of the F protein had been elucidated, two subunit vaccines based on recombinant prefusion F protein, AS01E-adjuvanted RSVPreF3 OA (Arexvy, GSK) and non-adjuvanted bivalent RSVpreF (Abrysvo, Pfizer), were authorized in 2023 for adults aged 60 years and older, each demonstrating greater than 80% efficacy against severe RSV-LRTI (48,49). This was followed in 2024 by approval of an mRNA vaccine (mRNA-1345; mRESVIA, Moderna), encoding stabilized prefusion F protein, for older adults (50). A maternal formulation of Abrysvo has also been authorized to protect newborns via transplacental antibody transfer (51). Together with nirsevimab and clesrovimab, these products constitute the first comprehensive RSV prevention toolkit spanning infancy to late adulthood (39).
Serologic surveys, most of which were based on detecting IgG anti-RSV antibodies using enzyme-linked immunosorbent assay (ELISA), show that most adults older than 60 years possess RSV-specific antibodies, reflecting decades of repeated exposure (Table 1). Moreover, some studies demonstrate that their serum concentrations increase stead ily with age (52). Nevertheless, incidence, hospitalization, and mortality continue to rise in this population (53)(54)(55)(56). This narrative review explores why infection-induced adaptive immune responses are often insufficient to prevent severe RSV disease in older adults and evaluates how vaccination, particularly with prefusion-focused platforms, may reduce the substantial burden of the virus in this rapidly growing demographic.
## WHY ARE OLDER PEOPLE MORE PRONE TO RSV INFECTION?
## Age-related changes in the respiratory system
With advancing age, the pulmonary system undergoes progressive anatomic and physiologic remodeling that diminishes respiratory reserve. Senile emphysema, characterized by loss of alveolar elastic recoil and airspace enlargement, lowers maximal expiratory flow and impairs oxygen diffusion, increasing vulnerability to lower respiratory tract infections (LRTIs), such as RSV (62,63). These changes are compounded by calcification of thoracic joints and kyphotic curvature, which stiffen the chest wall and raise functional residual capacity (64). These structural alterations restrict the ability to compensate for ventilation-perfusion mismatch during acute viral infection (65,66).
Aging also impairs mucociliary clearance, the first mechanical barrier to inhaled pathogens. Studies have demonstrated that ciliary beat frequency decreases significantly in older adults, and ultrastructural abnormalities become more common with age, reducing mucus transport efficacy (67,68).
In parallel, expiratory muscle strength declines with age, impairing cough effectiveness. Age-related loss of airway surface liquid, attributable to epithelial dysfunction, further reduces the mobility of mucus (62,69). This can impede the expulsion of RSV particles from the upper airway and synergize to allow deeper RSV penetration into the bronchioles and alveoli.
## Chronic diseases that amplify RSV risk
Aging is accompanied by a rising prevalence of chronic cardiopulmonary, vascular, metabolic, and oncologic disorders that synergistically increase RSV morbidity and mortality. These comorbidities impair host defenses, reduce physiologic reserve, and promote systemic inflammation, each contributing to a heightened risk of severe RSV disease in older adults.
Chronic obstructive pulmonary disease (COPD) and asthma are major risk factors for severe RSV infection. A recent systematic review estimated that their prevalence among RSV-infected adults in inpatient settings is 19% and 31%, respectively. Hospital ized COPD patients with RSV experience higher rates of respiratory failure and pro longed illness compared to non-infected counterparts (70)(71)(72)(73)(74)(75)(76). Moreover, the virus can exacerbate COPD by increasing airway inflammation, accelerating lung function decline, and contributing to disease progression (72,75).
Congestive heart failure (CHF), ischemic heart disease, and a history of stroke also significantly raise the risk of severe RSV infection and related complications such as myocardial infarction (71,72,(75)(76)(77)(78). Data from RSV-NET (2015-2017) showed that adults ≥18 years with CHF had an adjusted hospitalization rate of 26.7 per 10,000 (compared with 3.3 per 10,000 in those without CHF), increasing to 40.5 per 10,000 in those ≥65 years, compared with 3.3 per 10,000 in those without CHF (79). RSV exacer bates cardiac stress by inducing hypoxia and systemic inflammation, worsening heart failure symptoms, and increasing the likelihood of ICU admission (80,81). Importantly, the increased risk of cardiovascular events can persist for up to 6 months post-infection, especially for stroke and heart failure, though that risk gradually decreases over time (77,82,83). Chronic kidney disease is associated with some of the highest RSV hospitalization rates among comorbidities (70,78). Moreover, diabetes mellitus is frequently observed in elderly RSV patients and is associated with higher rates of severe disease and poor outcomes (70,74). Both of these conditions can impair immune responses, increasing susceptibility to severe RSV-related complications, such as secondary bacterial pneumo nia (81).
## Innate and adaptive immune senescence
With age, the human immune system undergoes a multidimensional functional decline known as immunosenescence, which impairs both antiviral defense and immunoregula tion during RSV infection. This senescence affects the balance and activity of innate immune cells, reshapes adaptive T-and B-cell compartments, and fosters a state of low-grade chronic inflammation, often termed "inflammaging" (84)(85)(86).
Innate immune recognition is weakened by reduced expression of pattern recogni tion receptors (PRRs), such as Toll-like receptors (TLRs) and RIG-I-like receptors (RLRs), diminishing early detection of RSV and delaying the initiation of antiviral signaling cascades (87). Monocytes and macrophages in aged individuals exhibit reduced phagocytic activity and a shift toward pro-inflammatory phenotypes, producing elevated levels of IL-6, IL-1β, and TNF-α even in the absence of infection (87)(88)(89). This base line hyperinflammation paradoxically coexists with impaired pathogen clearance and contributes to the immunopathology seen in older adults with RSV (90,91).
Aging also impairs macrophage function, which in elderly individuals exhibits an altered response to stimuli, a declined ability to migrate, and remains in chronic activation, releasing increased levels of inflammatory cytokines, e.g., IL-1β and SASP cytokines (92,93). At the same time, the expansion of CD16 + monocytes, which are more proinflammatory than classical monocytes, is observed in older adults (94). This is accompanied by the diminished production of antiviral type I interferons (93,95). These changes can also contribute to increased susceptibility of the elderly to RSV infection and less effective viral clearance.
Importantly, T-cell immunity is also significantly compromised in the elderly (96). Thymic involution restricts the production of naïve CD4 + and CD8 + T cells, leading to a reliance on clonally expanded, often senescent memory populations with limited responsiveness to novel antigenic variants (97). The CD4 + /CD8 + T cell ratio becomes skewed, and cytotoxic CD8 + T cells exhibit reduced proliferation and effector molecule secretion in response to RSV (98).
Regulatory T cells, whose frequency and suppressive activity increase with age (99), can also dampen effector responses to RSV antigens, potentially limiting viral clear ance but exacerbating immunosuppression (100). This shift contributes to inadequate resolution of infection and prolonged viral shedding, which is known to be a marker of disease severity (101,102). Moreover, the elderly show a significantly lower number of RSV F protein-specific IFN-γ-producing T cells and a reduced frequency of activated CD8 + T cells. Additionally, they exhibit higher IL-13 levels, which may indicate a skewed or less effective immune response (103). These findings suggest that impaired T-cell-mediated immunity also contributes to increased vulnerability to severe RSV disease in older adults.
Aging leads to a decline in naïve B cell output and reduced somatic hypermutation (97,99), potentially impairing antibody diversity and affinity maturation against RSV. Memory B cell function is also compromised, resulting in shorter-lived and less effective humoral responses (104,105).
Together, these changes culminate in a paradoxical immune landscape: increased baseline inflammation, defective early pathogen sensing, diminished cytotoxicity and release of antiviral molecules, and poor generation of protective antibody responses. All of these contribute to higher RSV susceptibility, prolonged disease, and poorer outcomes in older adults.
## Antibody quality, antigenic bias, and the postfusion trap
Although seropositivity for RSV is nearly universal among adults aged 60 years and older (Table 1), it is now well established that antibody quality, not quantity, is the critical determinant of protection. Laboratory adsorption experiments using stabilized prefusion F vs postfusion F proteins have shown that only the former can elicit high neutralizing activity in human sera (106). It unequivocally emphasizes that the most potent neutralizing antibodies target prefusion F-specific epitopes.
However, the gradual loss of neutralizing activity of anti-RSV antibodies was shown to occur with age despite an increase in seropositivity, measured by a higher percent age of individuals with detectable antibodies (60). Therefore, repeated RSV expo sure may drive an imbalanced immune response, favoring recall of non-neutralizing postfusion F-specific B cells over de novo responses to prefusion F. This possibility requires further investigatio,n though it would result from immunological imprinting, whereby the immune system preferentially recalls memory B cells targeting the more stable, immunodominant postfusion conformation of F protein encountered during initial childhood infections. Such imprinting may limit the formation of new antibody responses directed against neutralization-sensitive prefusion-specific epitopes in later life. Importantly, both prefusion F-and postfusion F-directed antibodies may bind with high affinity, but antibodies targeting the former conformation are far more effective at neutralization because they block the F protein refolding process required for viral membrane fusion.
In addition, immunosenescence impairs germinal center function, thereby reducing the overall capacity of aged individuals to generate diverse class-switched antibody repertoires (107)(108)(109). This impacts all antigenic targets, but the consequences are particularly important for RSV: if prefusion F-specific antibodies are not effectively induced, the neutralization capacity is markedly compromised, because antibodies that target postfusion F, although they may reveal high affinity, generally recognize epitopes less relevant for blocking viral entry. Thus, the issue is not that prefusion F antibodies are preferentially diminished, but that their absence or insufficiency leaves the host relying on functionally weaker postfusion F responses. Antigen presentation dynamics also play a role, as the prefusion F protein is structurally unstable and often transitions into the postfusion form before being processed by antigen-presenting cells, leading to a biased display of less protective epitopes (110). Consequently, repeated RSV infections may reinforce this postfusion-biased memory through clonal expansion of B cells targeting accessible but non-neutralizing regions. This immunological skewing results in elevated IgG binding titers that are functionally weak and poorly neutralizing. Supporting this, a prospective study in frail elderly persons showed that individuals who later developed symptomatic RSV infection had significantly lower preseason serum IgG titers to RSV F protein and lower neutralizing antibody titers to both RSV-A and RSV-B compared with matched controls who remained uninfected (111).
Importantly, animal studies have mechanistically linked low-quality post-vaccination antibodies to ADE, which had been observed historically with the formalin-inactivated RSV vaccine (36,37). Furthermore, an in vitro study demonstrated that RSV co-incubated with suboptimal concentrations of neutralizing antibodies led to ADE and increased lung viral loads (112). Possibly, the similar suboptimal levels that occur in the elderly contribute to disease severity following subsequent RSV reinfections.
These findings highlight a central paradox in RSV immunity among older adults: despite near-universal seropositivity and sometimes elevated RSV-specific IgG titers, protection against infection and severe disease remains limited. This may not be due to a lack of antibodies per se, but rather to a qualitative decline in their functional capacity, marked by a shift toward non-neutralizing, postfusion F-biased responses that may even potentiate disease through mechanisms like ADE, as well as one or more of the physiological changes in older people. Table 2 summarizes mechanisms that may contribute to the predominance of weakly neutralizing immune responses against RSV in older adults. As such, these age-associated alterations in B cell memory and antibody quality underscore the urgent need for immunization strategies that selectively boost prefusion F-specific responses. Targeting these highly neutralizing epitopes may offer a more effective path to protective immunity in the elderly, a population especially vulnerable to RSV-related morbidity and mortality.
## THE BENEFIT OF RSV VACCINATION IN THE ELDERLY
All three vaccines currently available (Table 3) focus the immune responses on the prefusion F protein, which displays the most potent neutralizing epitopes (Table 3). Although natural RSV infection predominantly induces memory B cells specific for prefusion F, a substantial fraction of cross-reactive antibodies show higher apparent affinity for the postfusion form (113); these postfusion F-biased responses are generally less efficient at neutralization and, under certain conditions, may contribute to ADE (36,37,114) (Fig. 1). A seroprevalence survey that also focused on neutralization aspects clearly shows that despite nearly universal anti-RSV IgG seropositivity in older people, the serum of many of them may not exhibit any neutralizing activity against the virus (60). This leaves these individuals vulnerable not only to infection but also to its severe clinical course due to the accumulation of other factors, including underlying diseases and immunosenescence. In contrast, vaccines focusing on prefusion F protein as a target elicit a higher ratio of neutralizing to binding antibodies, effectively blocking viral fusion without enhancing disease severity (45,(115)(116)(117). Importantly, all authorized vaccines are based on a stabilized form of this protein, which is conformationally locked in the prefusion form and engineered to resist potential shifts to postfusion form and has preserved neutralizing epitopes (50,118,119). Importantly, the half-life of weakly neutralizing antibodies is shorter than that of antibodies with high potencies (120). Altogether, this explains the superiority of RSV vaccination in providing protection in the elderly. In addition, all authorized RSV vaccines also activate the cellular immune response, stimulating memory B cells and polyfunctional CD4 + T cells, with evidence that mRNA-1345 additionally stimulates specific CD8 + T cell responses (121)(122)(123). This cellular activation is important because it broadens the immune repertoire, supports more durable protection, and may enhance the response upon subsequent exposure to RSV.
Notably, inter-individual variability in vaccine-induced immunity is substantial in older and immunocompromised adults, with some mounting robust neutralizing antibody responses, while others respond only weakly. A small exploratory study in immunocompromised individuals suggested that AS01E-adjuvanted formulation Arexvy may elicit higher neutralizing antibody titers than the non-adjuvanted vaccine (128). However, these preliminary findings are based on limited sample sizes, and any firm conclusions or preference for a specific product would require further investigations. It should also be noted that aging-related immunosenescence may diminish TLR signaling, but this is unlikely to compromise the action of the AS01E adjuvant. As a TLR4 agonist, (129)(130)(131). Clinical trials have shown substantial efficacy across all three vaccines. In the phase 3 AReSVi-006 trial, vaccination of adults aged 60 years and older with Arexvy demon strated 82.6% efficacy against RSV-associated lower respiratory tract disease (RSV-LRTD; defined as two or more lower respiratory symptoms/signs, including ≥1 lower respiratory sign, or three or more lower respiratory symptoms, lasting 24 h or longer) in the first RSV season and 56.1% in the second season, with a pooled efficacy of 74.5% over two seasons (126). Efficacy against medically attended LRTD was 87.5%. Efficacy over two seasons of a first dose followed by revaccination was 67.1% against RSV-LRTD and 78.8% against severe RSV-LRTD (126). The phase 3 RENOIR trial of Abrysvo, vaccination of individuals ≥60 years, resulted in 88.9% and 77.8% efficacy against RSV-LRTD with more than three symptoms over the first and second seasons, respectively. The efficacy against acute respiratory illness was 62.2% after the first season and 36.9% during the second season (127). In the case of mRESVIA, the efficacy against the disease with at least three symptoms in the ConquerRSV trial of adults aged ≥60 years was 82.4% during the first season, while the efficacy against acute respiratory disease was 68.4% (50). Direct head-to-head comparisons among these vaccines are not available, as the clinical trials used different outcome definitions and study designs (125); nonetheless, all authorized RSV vaccines have demonstrated substantial efficacy and should be considered effective, despite differences in technological platforms and composition.
Real-world observational studies further support these findings, with vaccine effectiveness (VE) estimates ranging from 75% to 80% against RSV-related acute respiratory infections, emergency department visits, and hospitalizations in older adults (132)(133)(134). Immunocompromised individuals aged over 60 years showed reduced but still clinically meaningful protection. VE against RSV-associated urgent care visits or hospitalizations was 67.0% for the group aged 60-74 years and 73.1% for those aged 75 years and older, with the lowest VE observed in the subgroup of patients who received stem cell transplants (133). Since immunocompromised individuals, including those with age-related immune deficiencies, may mount diminished responses to RSV vaccines while remaining at increased risk of severe disease, additional vaccination strategies in this group, such as revaccination or higher antigen formulations, could be considered in the future.
Both clinical trials and post-authorization surveillance indicated a generally favorable safety profile of RSV vaccines authorized for older people, with the most common side effects being mild to moderate, such as injection-site pain, fatigue, and headache (135). Some studies have reported a small but statistically significant increase in Guillain-Barré syndrome (GBS) cases (ranging from 5.2 to 18.2 per million doses in one study to 6.5-9.0 per million in another), although the overall risk remains low (133,136). The potential mechanisms and contributing factors underlying this association are not yet fully understood and warrant further investigation. Importantly, and in line with expectations, the existing clinical and real-world studies indicate no evidence of ADE in adults vaccinated against RSV with prefusion-based formulations.
It is currently unknown how RSV vaccination may influence imprinting or the immune response to subsequent infections with circulating RSV strains. Concepts, such as original antigenic sin, in which the initial exposure shapes future immune responses, suggest that the age and prior infection history of the host could modulate vaccine-induced immunity, but this remains largely theoretical in the context of RSV (137).
In summary, RSV vaccination in the elderly provides robust protection against severe disease, with real-world data reinforcing clinical trial results. From a public health perspective, modeling studies suggest that RSV vaccination in older adults could reduce hospitalizations by 35%-64% in high-income countries, with substantial cost savings (138). Ongoing monitoring is necessary to assess long-term durability and rare adverse events, but current evidence strongly supports its use in this vulnerable population. Vaccination leads to the generation of neutralizing anti-prefusion F antibodies (colored in green). At the same time, natural infection induces a broader response that includes neutralizing antibodies to prefusion F as well as cross-reactive antibodies with higher apparent affinity for postfusion F (colored in red). The latter contributes less effectively to viral neutralization and, in some contexts, may even facilitate antibody-dependent enhancement, which in elderly individuals could exacerbate disease severity due to comorbidities and immunosenescence.
## CONCLUSIONS
Older adults are disproportionately vulnerable to severe RSV disease due to the combined effects of age-related lung degeneration, multimorbidity, immunosenescence, and a skewed antibody repertoire dominated by non-neutralizing responses. Together, these mechanisms explain why naturally acquired immunity, despite repeated exposures and high seroprevalence, often fails to confer protection in later life. Understanding these interacting factors underscores the clinical and immunologic value of prefusion F-based vaccination. Such vaccines elicit high-quality neutralizing antibodies and show strong efficacy in both clinical trials and real-world studies. To reduce global RSV-related morbidity and mortality in older populations, wider vaccine adoption, coupled with continued monitoring of safety and real-world effectiveness, is now a public health priority.
## References
1. Griffiths, Drews, Marchant (2017) "Respiratory syncytial virus: infection, detection, and new options for prevention and treatment" *Clin Microbiol Rev*
2. Wei, Wang, Li et al. (2024) "Novel imported clades accelerated the RSV surge in Beijing" *J Infect*
3. Yunker, Fall, Norton et al. (2024) "Genomic evolution and surveillance of respiratory syncytial virus during the 2023-2024 season" *Viruses*
4. Walsh, Mcconnochie, Long et al. (1997) "Severity of respiratory syncytial virus infection is related to virus strain" *J Infect Dis*
5. Nuttens, Moyersoen, Curcio et al. (2024) "Differences between RSV A and RSV B subgroups and implications for pharmaceutical preventive measures" *Infect Dis Ther*
6. Staadegaard, Caini, Wangchuk et al. (2021) "The global epidemiology of RSV in community and hospitalized care: findings from 15 countries" *Open Forum Infect Dis*
7. Osei-Yeboah, Amankwah, Begier et al. (2024) "Burden of respiratory syncytial virus (RSV) infection among adults in nursing and care homes: a systematic review" *Influenza Resp Viruses*
8. Walsh (2011) "Respiratory syncytial virus infection in adults" *Semin Respir Crit Care Med*
9. Branche, Falsey (2015) "Respiratory syncytial virus infection in older adults: an under-recognized problem" *Drugs Aging*
10. Nduaguba, Tran, Choi et al. (2023) "Respiratory syncytial virus reinfections among infants and young children in the Minireview Journal of Virology October"
11. (2011) *PLoS One*
12. Pasittungkul, Thongpan, Vichaiwattana et al. (2024) "Prevalence and genetic diversity of respiratory syncytial virus reinfections in young Thai children, 2016-2023" *J Med Virol*
13. Shi, Mcallister, Brien et al. (2017) "Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study" *Lancet*
14. 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*
15. Falsey, Mcelhaney, Beran et al. (2014) "Respiratory syncytial virus and other respiratory viral infections in older adults with moderate to severe influenza-like illness" *J Infect Dis*
16. Shi, Denouel, Tietjen et al. (2020) "Global disease burden estimates of respiratory syncytial virus-associated acute respiratory infection in older adults in 2015: a systematic review and meta-analysis" *J Infect Dis*
17. Balasubramani, Nowalk, Eng et al. (2022) "Estimating the burden of adult hospitalized RSV infection using local and state data -methodology" *Hum Vaccin Immunother*
18. Havers, Whitaker, Melgar et al. (2016) "Burden of respiratory syncytial virus-associated hospitalizations in US adults" *JAMA Netw Open*
19. Osei-Yeboah, Spreeuwenberg, Riccio et al. "Respiratory Syncytial Virus Consortium in Europe (RESCEU) Investigators. 2023. Estimation of the number of respiratory syncytial virus-associated hospitalizations in adults in the European Union" *J Infect Dis*
20. Lai, Hsueh (2024) "An overview on disease burden and management of respiratory syncytial virus infections in older adults in the Asia-Pacific region" *Rev Med Virol*
21. Choi, Hill-Ricciuti, Branche et al. (2017) "Cost determinants among adults hospitalized with respiratory syncytial virus in the United States" *Influenza Other Respir Viruses*
22. Geoghegan, Erviti, Caballero et al. (2017) "Mortality due to respiratory syncytial virus. Burden and risk factors" *Am J Respir Crit Care Med*
23. Young, Smitherman (2021) "Socioeconomic impact of RSV hospitalization" *Infect Dis Ther*
24. King (1997) "Community respiratory viruses in individuals with human immunodeficiency virus infection" *Am J Med*
25. 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 metaanalysis" *Influenza Other Respir Viruses*
26. Mazela, Jackowska, Czech et al. (2024) "Epidemiology of respiratory syncytial virus hospitalizations in Poland: an analysis from 2015 to 2023 covering the entire polish population of children aged under five years" *Viruses*
27. Alfano, Bigoni, Caggiano et al. (2024) "Respiratory syncytial virus infection in older adults: an update" *Drugs Aging*
28. Tanriover, Azap, Edis et al. (2025) "Respiratory syncytial virus (RSV) infections in adults: current trends and recommendations for prevention -a global challenge from a local perspective" *Hum Vaccin Immunother*
29. Falsey, Hennessey, Formica et al. (2005) "Respiratory syncytial virus infection in elderly and high-risk adults" *N Engl J Med*
30. Wee, Lim, Ho et al. (2025) "Severity of respiratory syncytial virus versus SARS-CoV-2 Omicron and influenza infection amongst hospitalized Singaporean adults: a national cohort study" *Lancet Reg Health West Pac*
31. Surie, Yuengling, Decuir et al. (2024) "Severity of respiratory syncytial virus vs COVID-19 and influenza among hospitalized US adults" *JAMA Netw Open*
32. Ruckwardt (2023) "The road to approved vaccines for respiratory syncytial virus" *NPJ Vaccines*
33. Rzymski, Szuster-Ciesielska, Dzieciątkowski et al. (2023) "mRNA vaccines: the future of prevention of viral infections" *J Med Virol*
34. Kim, Canchola, Brandt et al. (1969) "Respiratory syncytial virus disease in infants despite prior administration of antigenic inactivated vaccine" *Am J Epidemiol*
35. Kapikian, Mitchell, Chanock et al. (1969) "An epidemiologic study of altered clinical reactivity to respiratory syncytial (RS) virus infection in children previously vaccinated with an inactivated RS virus vaccine" *Am J Epidemiol*
36. Killikelly, Kanekiyo, Graham (2016) "Pre-fusion F is absent on the surface of formalin-inactivated respiratory syncytial virus" *Sci Rep*
37. Ponnuraj, Springer, Hayward et al. (2003) "Antibody-dependent enhancement, a possible mechanism in augmented pulmonary disease of respiratory syncytial virus in the Bonnet monkey model" *J Infect Dis*
38. Schneider-Ohrum, Cayatte, Bennett et al. (2017) "Immunization with low doses of recombinant postfusion or prefusion respiratory syncytial virus F primes for vaccine-enhanced disease in the cotton rat model independently of the presence of a Th1biasing (GLA-SE) or Th2-biasing (alum) adjuvant" *J Virol*
39. Ambrose, Chen, Kumar (2014) "A population-weighted, condition-adjusted estimate of palivizumab efficacy in preventing RSVrelated hospitalizations among US high-risk children" *Hum Vaccin Immunother*
40. Rzymski, Gwenzi (2024) "Respiratory syncytial virus immunoprophy laxis: novel opportunities and a call for equity" *J Med Virol*
41. Zhu, Mclellan, Kallewaard et al. (1928) "A highly potent extended half-life antibody as a potential RSV vaccine surrogate for all infants" *Sci Transl Med*
42. Muller, Madhi, Nuñez et al. (2023) "Nirsevimab for prevention of RSV in term and latepreterm infants" *N Engl J Med*
43. (2025) *Minireview Journal of Virology*
44. Domachowske, Chang, Atanasova et al. (2023) "Safety of re-dosing nirsevimab prior to RSV season 2 in children with heart or lung disease" *J Pediatric Infect Dis Soc*
45. Zar, Simoes, Madhi et al. (2025) "A phase 2b/3 study to evaluate the efficacy and safety of an investigational respiratory syncytial virus (RSV) antibody, clesrovimab, in healthy preterm and full-term infants" *Open Forum Infect Dis*
46. Anderer (2025) "FDA approves new RSV monoclonal antibody to protect infants" *JAMA*
47. Mclellan, Chen, Joyce et al. (2013) "Structure-based design of a fusion glycoprotein vaccine for respiratory syncytial virus" *Science*
48. Mclellan, Yang, Graham et al. (2011) "Structure of respiratory syncytial virus fusion glycoprotein in the postfusion conformation reveals preservation of neutralizing epitopes" *J Virol*
49. Mousa, Kose, Matta et al. (2017) "A novel prefusion conformation-specific neutralizing epitope on the respiratory syncytial virus fusion protein" *Nat Microbiol*
50. Walsh, Marc, Zareba et al. (2023) "Efficacy and safety of a bivalent RSV prefusion F vaccine in older adults" *N Engl J Med*
51. Papi, Ison, Langley et al. "AReSVi-006 Study Group. 2023. Respiratory syncytial virus prefusion F protein vaccine in older adults" *N Engl J Med*
52. Wilson, Goswami, Baqui et al. (2023) "Efficacy and safety of an mRNA-based RSV PreF vaccine in older adults" *N Engl J Med*
53. Kampmann, Madhi, Munjal et al. (2023) "Bivalent prefusion F vaccine in pregnancy to prevent RSV illness in infants" *N Engl J Med*
54. Poniedziałek, Majewska, Kondratiuk et al. (2025) "Seroprevalence of RSV IgG antibodies across age groups in Poland after the COVID-19 pandemic: data from the 2023/2024 epidemic season" *Vaccines (Basel)*
55. Cong, Dighero, Zhang et al. (2023) "Understanding the age spectrum of respiratory syncytial virus associated hospitalisa tion and mortality burden based on statistical modelling methods: a systematic analysis" *BMC Med*
56. Méroc, Liang, Iantomasi et al. (2015) "A model-based estimation of RSVattributable incidence of hospitalizations and deaths in Italy between" *Infect Dis Ther*
57. Scholz, Dobrindt, Tufts et al. (2024) "The burden of respiratory syncytial virus (RSV) in Germany: a comprehensive data analysis suggests underdetection of hospitalisa tions and deaths in adults 60 years and older" *Infect Dis Ther*
58. Inoue, Nagai, Fushimi (2025) "Severity and outcomes of adult respiratory syncytial virus inpatient compared with influenza: observational study from Japan" *Infect Dis (Lond)*
59. Teodoro, Ovsyannikova, Grill et al. (2022) "Seroprevalence of RSV antibodies in a contemporary"
60. "cohort of adults" *Int J Infect Dis*
61. Nham, Jang, Hyun et al. (2024) "Age-stratified seroprevalence of respiratory syncytial virus: analysis using prefusion F and G protein antibodies" *Vaccines (Basel)*
62. Arankalle, Kulkarni, Malshe et al. (2019) "Seroepidemiology of respiratory syncytial virus in western India with special reference to appropriate age for infant vaccination" *J Med Virol*
63. Terrosi, Genova, Martorelli et al. (2009) "Humoral immunity to respiratory syncytial virus in young and elderly adults" *Epidemiol Infect*
64. Sastre, Ruiz, Schildgen et al. (2012) "Seroprevalence of human respiratory syncytial virus and human metapneumovirus in healthy population analyzed by recombinant fusion protein-based enzyme linked immunosorbent assay" *Virol J*
65. Janssens, Pache, Nicod (1999) "Physiological changes in respiratory function associated with ageing" *Eur Respir J*
66. Sharma, Goodwin (2006) "Effect of aging on respiratory system physiology and immunology" *Clin Interv Aging*
67. Verbeken, Cauberghs, Lauweryns et al. (1994) "Anatomy of membranous bronchioles in normal, senile and emphy sematous human lungs" *J Appl Physiol*
68. Tzelepis (2018) "Chest wall diseases: respiratory pathophysiology" *Clin Chest Med*
69. Wang, Huang, Luo et al. (2024) "The aging lung: microenvironment, mechanisms, and diseases" *Front Immunol*
70. Ho, Chan, Hu et al. (2001) "The effect of aging on nasal mucociliary clearance, beat frequency, and ultrastructure of respiratory cilia" *Am J Respir Crit Care Med*
71. Bailey (2022) "Aging diminishes mucociliary clearance of the lung" *Adv Geriatr Med Res*
72. Freitas, Ibiapina, Alvim et al. (2010) "Relation ship between cough strength and functional level in elderly" *Rev Bras Fisioter*
73. Osei-Yeboah, Johannesen, Egeskov-Cavling et al. (2024) "Respiratory syncytial virus-associated hospitalization in adults with comorbidities in 2 European countries: a modeling study" *J Infect Dis*
74. Ackerson, Tseng, Sy et al. (2019) "Severe morbidity and mortality associated with respiratory syncytial virus versus influenza infection in hospitalized older adults" *Clin Infect Dis*
75. Singer, Wang, Wu et al. (2025) "P-686. Risk of COPD exacerbations, asthma exacerbations, and hospitalizations for heart failure after RSV hospitalization among US adults aged at least 50 years" *Open Forum Infect Dis*
76. Kurai, Song, Huang et al. (2023) "Targeted literature review of the burden of respiratory syncytial infection among high-risk and elderly patients in Asia Pacific region" *Infect Dis Ther*
77. Fu, Lai, Su et al. (2025) "High bacterial coinfection rates and associated mortality among hospitalized older adults with laboratory-confirmed respiratory syncytial virus infection" *J Microbiol Immunol Infect*
78. (2025) *Minireview Journal of Virology*
79. Korsten, Welkers, Van De Laar et al. (2025) "Unveiling the spectrum of respiratory syncytial virus disease in adults: from community to hospital" *Influenza Other Respi Viruses*
80. Bhargava, Szpunar, Sharma et al. (2025) "P-676. Risk factors associated with longer hospital stay in elderly patients with respiratory syncytial virus infections" *Open Forum Infect Dis*
81. Singer, Wang, Wu et al. (2024) "Abstract 4123849: respiratory syncytial virus (RSV) cases involving hospitaliza tion are associated with an increased risk of myocardial infarction and All-Cause mortality among adults aged 50 years and older" *Circulation*
82. Wyffels, Kariburyo, Gavart et al. (2020) "A realworld analysis of patient characteristics and predictors of hospitaliza tion among US Medicare beneficiaries with respiratory syncytial virus infection" *Adv Ther*
83. Kujawski, Whitaker, Ritchey et al. (2015) "Rates of respiratory syncytial virus (RSV)-associated hospitalization among adults with congestive heart failure-United States" *PLoS One*
84. Sikkel, Quint, Mallia et al. (2008) "Respiratory syncytial virus persistence in chronic obstructive pulmonary disease" *Pediatr Infect Dis J*
85. Nguyen-Van-Tam, Leary, Martin et al. (2022) "Burden of respiratory syncytial virus infection in older and high-risk adults: a systematic review and meta-analysis of the evidence from developed countries" *Eur Respir Rev*
86. Liang, Aliabadi, Sato et al. (2025) "P-682. The risk of cardiorespiratory events for up to 180 days following respiratory syncytial virus (RSV) infection hospitalization: a self-controlled case series analysis" *Open Forum Infect Dis*
87. Sudnik, Walsh, Branche et al. (2025) "P-718. Getting to the heart of respiratory syncytial virus disease burden" *Open Forum Infect Dis*
88. Franceschi, Garagnani, Parini et al. (2018) "Inflammaging: a new immune-metabolic viewpoint for age-related diseases" *Nat Rev Endocrinol*
89. Santoro, Bientinesi, Monti (2021) "Immunosenescence and inflammaging in the aging process: age-related diseases or longevity?" *Ageing Res Rev*
90. Fulop, Witkowski, Olivieri et al. (2018) "The integration of inflammaging in age-related diseases" *Semin Immunol*
91. Quiros-Roldan, Sottini, Natali et al. (2024) "The impact of immune system aging on infectious diseases" *Microorganisms*
92. Palacios-Pedrero, Osterhaus, Becker et al. (2021) "Aging and options to halt declining immunity to virus infections" *Front Immunol*
93. Haynes (2020) "Aging of the immune system: research challenges to enhance the health span of older adults" *Front Aging*
94. Del Giudice, Goronzy, Grubeck-Loebenstein et al. (2018) "Fighting against a protean enemy: immunosenescence, vaccines, and healthy aging" *NPJ Aging Mech Dis*
95. Smits, Jochems (2024) "Diverging patterns in innate immunity against respiratory viruses during a lifetime: lessons from the young and the old" *Eur Respir Rev*
96. Duong, Radley, Lee et al. (2021) "Macrophage function in the elderly and impact on injury repair and cancer" *Immun Ageing*
97. Maeyer, Chambers (2021) "The impact of ageing on monocytes and macrophages" *Immunol Lett*
98. Nyugen, Agrawal, Gollapudi et al. (2010) "Impaired functions of peripheral blood monocyte subpopulations in aged humans" *J Clin Immunol*
99. Molony, Nguyen, Kong et al. (2017) "Aging impairs both primary and secondary RIG-I signaling for interferon induction in human monocytes" *Sci Signal*
100. Han, Li, Wang et al. (2025) "Infection of nonclassic monocytes by respiratory syncytial virus induces an imbalance in the CD4 + T-cell subset response" *Microbiol Spectr*
101. Garrido-Rodríguez, Herrero-Fernández, Castro et al. (2021) "Immunological features beyond CD4/CD8 ratio values in older individuals"
102. Chen, Kelley, Goldstein (2020) "Role of aging and the immune response to respiratory viral infections: potential implications for COVID-19" *J Immunol*
103. Palatella, Guillaume, Linterman et al. (2022) "The dark side of Tregs during aging" *Front Immunol*
104. Mangodt, Van Herck, Nullens et al. (2015) "The role of Th17 and Treg responses in the pathogenesis of RSV infection" *Pediatr Res*
105. Walsh, Peterson, Kalkanoglu et al. (2013) "Viral shedding and immune responses to respiratory syncytial virus infection in older adults" *J Infect Dis*
106. Watanabe, Nishiyama, Lyra et al. (2024) "Prolonged viral shedding as a marker of severity in respiratory syncytial virus bronchiolitis"
107. Cherukuri, Patton, Gasser et al. (2013) "Adults 65 years old and older have reduced numbers of functional memory T cells to respiratory syncytial virus fusion protein" *Clin Vaccine Immunol*
108. Frasca, Blomberg (2011) "Aging affects human B cell responses" *J Clin Immunol*
109. Yu, Lu, Yu et al. (2024) "B cells dynamic in aging and the implications of nutritional regulation" *Nutrients*
110. Ngwuta, Chen, Modjarrad et al. (2015) "Prefusion F-specific antibodies determine the magnitude of RSV neutralizing activity in human sera" *Sci Transl Med*
111. Foster, Marcial-Juárez, Linterman (2025) "The cellular factors that impair the germinal center in advanced age" *J Immunol*
112. Rodas, Maul, Gearhart (2023) "Pre-activation of follicular B cells in old mice generates a hyper-response during germinal center reaction" *J Immunol*
113. Finney, Yeh, Kelsoe et al. (2018) "Germinal center responses to complex antigens" *Immunol Rev*
114. Patel, Tian, Flores et al. (2020) "Flexible RSV prefusogenic fusion glycoprotein exposes multiple neutralizing epitopes that may collectively contribute to protective immunity" *Vaccines (Basel)*
115. (2025) *Minireview Journal of Virology*
116. Falsey, Walsh (1998) "Relationship of serum antibody to risk of respiratory syncytial virus infection in elderly adults" *J Infect Dis*
117. Van Erp, Feyaerts, Duijst et al. (2019) "Respiratory syncytial virus infects primary neonatal and adult natural killer cells and affects their antiviral effector function" *J Infect Dis*
118. Gilman, Castellanos, Chen et al. (2016) "Rapid profiling of RSV antibody repertoires from the memory B cells of naturally infected adult donors"
119. Luo, Zhao, Lu et al. (2025) "Sub-neutralizing levels of antibodies against RSV F protein enhance RSV infection via Fc-FcγR interactions" *Front Immunol*
120. Goswami, Baqui, Doreski et al. (2024) "Humoral immunogenicity of mRNA-1345 RSV vaccine in older adults" *J Infect Dis*
121. Simões, Center, Tita et al. (2022) "Prefusion F protein-based respiratory syncytial virus immunization in pregnancy" *N Engl J Med*
122. Bouzya, Rouxel, Sacconnay et al. (2023) "Immunogenicity of an AS01-adjuvanted respiratory syncytial virus prefusion F (RSVPreF3) vaccine in animal models" *NPJ Vaccines*
123. Schmoele-Thoma, Falsey, Walsh et al. (2019) "Phase 1/2, first-in-human study of the safety, tolerability, and immunogenicity of an RSV prefusion F-based subunit vaccine candidate" *Open Forum Infect Dis*
124. Ison, Papi, Langley et al. (2022) "Respiratory syncytial virus (RSV) prefusion F protein candidate vaccine (RSVPreF3 OA) is efficacious in adults ≥ 60 years of age (YOA)"
125. Taleb, Al-Ansari, Nasrallah et al. (2021) "Level of maternal respiratory syncytial virus (RSV) F antibodies in hospitalized children and correlates of protection" *Int J Infect Dis*
126. Schwarz, Hwang, Ylisastigui et al. (2024) "Immunogenicity and safety following 1 dose of AS01 E -adjuvanted respiratory syncytial virus prefusion F protein vaccine in older adults: a phase 3 trial" *J Infect Dis*
127. Baber, Arya, Moodley et al. (2022) "A phase 1/2 study of a respiratory syncytial virus prefusion F vaccine with and without adjuvant in healthy older adults" *J Infect Dis*
128. Espeseth, Cejas, Citron et al. (2020) "Modified mRNA/ lipid nanoparticle-based vaccines expressing respiratory syncytial virus F protein variants are immunogenic and protective in rodent models of RSV infection" *NPJ Vaccines*
129. Acip (2014) "Overview of Moderna's investigational RSV vaccine (mRNA-1345) in adults ≥ 60 years of age"
130. Williams, Gessner, Begier et al. (2025) "Is a consensus case definition for viral associated lower respiratory tract disease (LRTD) in clinical trials possible?" *Infect Dis Ther*
131. Ison, Papi, Athan et al. "AReSVi-006 Study Group. 2024. Efficacy and safety of respiratory syncytial virus (RSV) prefusion F protein vaccine (RSVPreF3 OA) in older adults over 2 RSV seasons" *Clin Infect Dis*
132. Walsh, Marc, Falsey et al. (2024) "RENOIR trial -RSVpreF vaccine efficacy over two seasons" *N Engl J Med*
133. Karaba, Hage, Sengsouk et al. (2025) "Antibody response to respiratory syncytial virus vaccination in immunocompromised persons" *JAMA*
134. Stylianou, Bertram, Vo et al. (2024) "Innate immune cell activation by adjuvant AS01 in human lymph node explants is age independent" *J Clin Invest*
135. Nanishi, Angelidou, Rotman et al. (2022) "Precision vaccine adjuvants for older adults: a scoping review" *Clin Infect Dis*
136. Roman, Burny, Ceregido et al. (2024) "Adjuvant system AS01: from mode of action to effective vaccines" *Expert Rev Vaccines*
137. Bajema, Yan, Li et al. (2023) "Respiratory syncytial virus vaccine effectiveness among US veterans" *Lancet Infect Dis*
138. Fry, Terebuh, Kaelber et al. (2025) "Effectiveness and safety of respiratory syncytial virus vaccine for US adults aged 60 years or older" *JAMA Netw Open*
139. Surie, Self, Zhu et al. "Investigating Respiratory Viruses in the Acutely Ill (IVY) Network. 2024. RSV vaccine effectiveness against hospitalization among US adults 60 years and older" *JAMA*
140. Pang, Lu, Cao et al. (2024) "Efficacy, immunogenicity and safety of respiratory syncytial virus prefusion F vaccine: systematic review and meta-analysis" *BMC Public Health*
141. Lloyd, Shah, Zhang et al. (2025) "Evaluation of Guillain-Barré syndrome following respiratory syncytial virus vaccination among Medicare beneficiaries 65 years and older"
142. Tripp, Power (2019) "Original antigenic sin and respiratory syncytial virus vaccines" *Vaccines (Basel)*
143. Du, Pandey, Moghadas et al. (2025) "Impact of RSVpreF vaccination on reducing the burden of respiratory syncytial virus in infants and older adults" *Nat Med* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12009664&blobtype=pdf | # Survey of clinical microbiology and infectious disease testing capabilities among institutions in Africa
Jeremy Jacobs, Brian Adkins, Danny Milner, Evan Bloch, Quentin Eichbaum
## Abstract
Objectives: Inadequate laboratory infrastructure and testing capabilities are a major impediment to addressing the infectious disease burden in Africa. Therefore, the aims of this study were to characterize the clinical microbiology/infectious disease laboratory capabilities among countries in Africa.
Methods:A survey to assess the microbiological testing capabilities at hospitals, government laboratories, and free-standing public and private laboratories in African countries was developed by subject matter experts. Questions included institutional demographics and microbiology services in the broad categories of bacteriology, virology, mycology, parasitology, and rapid diagnostics/point-of-care testing. The survey was distributed using the American Society of Clinical Pathology email listserv between June and August 2022.Results: In total, 131 unique institutions in 28 countries endorsed at least 1 type of microbiology service, with parasitology (80.9%, 106/131) and bacteriology (77.9%, 102/131) being most common, while mycology (45.0%, 59/131) and virology (45.8%, 60/131) laboratories were less prevalent. The most frequently performed bacteriology test was bacterial identification (90.2%, 92/102), followed by aerobic bacterial cultures and antimicrobial susceptibility testing (both 89.2%, 91/102). Among all clinical microbiology/infectious disease laboratories, the most commonly tested agents were HIV (90.8%, 119/131), Treponema pallidum (78.6%, 103/131), Plasmodium falciparum (76.3%, 100/131), Mycobacterium tuberculosis (76.3%, 100/131), and hepatitis C virus (74.8%, 98/131).Conclusions: These findings provide contemporary data regarding the availability of critical infectious disease testing capabilities among institutions in Africa. These results and future additional studies will be crucial for understanding where strategic investment in the laboratory and public health infrastructure is warranted.
i n t r O D U c t i O nInfectious diseases are associated with substantial morbidity and mortality in lowresource settings, particularly in sub-Saharan Africa. 1,2 A variety of factors contribute to the
## K e Y P O i n t S
• Understanding the current availability of clinical microbiology/ infectious disease laboratory capabilities among countries in Africa is critical to reducing the infectious disease burden.
• The availability of contemporary infectious disease diagnostic testing is variable across Africa; improving access to virology (non-HIV) and mycology testing warrants prioritization.
• The most frequent agents tested for were HIV (91%), Treponema pallidum (79%), Plasmodium falciparum (76%), Mycobacterium tuberculosis (76%), and hepatitis C virus (75%). prevalence of infectious disease in Africa, including low socioeconomic status, limited access to health care, lack of trained health care workers and personal protective equipment, and inadequate laboratory infrastructure and testing capabilities. [2][3][4] Laboratory medicine capacity, inclusive of clinical microbiology and infectious disease laboratory diagnostic testing, is crucial to a well-functioning medical and public health system. These capabilities enable infections to be diagnosed accurately and guide appropriate therapy. In the absence of quality laboratory testing, clinical diagnosis is utilized (ie, relying on clinical signs and symptoms); however, this may be unreliable and has been associated with increased mortality. 5 Laboratory diagnosis is also critical for disease surveillance. In the absence of epidemiologic analysis, the impact of specific infectious diseases and targeted interventions on morbidity or mortality cannot be ascertained. 6 Finally, given the clinically significant burden of antimicrobial resistance, antimicrobial susceptibility testing (AST) capabilities are pivotal to addressing this public health threat. 7,8 Given these considerations, understanding the clinical microbiology/infectious disease laboratory testing capacity is crucial for strategic investment in this key component of medical care. Thus, we aimed to characterize the clinical microbiology/infectious disease laboratory capabilities among countries in Africa.
## K e Y W O r D S
## M at e r i a l S a n D M e t H O D S
This cross-sectional study aimed to assess the microbiological testing capabilities at hospitals, government laboratories, and free-standing public and private laboratories in African countries. The microbiology survey was developed as part of a larger clinical pathology (CP) survey (available in the Supplement; all supplementary material is available at American Journal of Clinical Pathology online) by specialists in anatomic pathology, CP, microbiology, infectious disease, and global health and administered to individuals (eg, pathologists, microbiologists, biomedical laboratory scientists) via the American Society for Clinical Pathology (ASCP) email listserv, as described previously. 9 This listserv is composed of ASCP member emails and requires individual and/or institutional membership in ASCP. This listserv comprises more than 38,000 email addresses worldwide, although it was not known at the time of survey distribution how many were inactive or not monitored, nor was the number of email addresses for individuals at a particular institution or in a particular country known. The initial survey invitation was distributed in June 2022, with 2 reminder emails 4 weeks apart.
Survey questions included institutional demographics and microbiology services in the broad categories of bacteriology, virology, mycology, parasitology, and rapid diagnostics/point-of-care (POC) testing. Microbiology testing that was related specifically to blood donors, blood products, and transfusion-transmitted infections has previously been reported elsewhere. 10 Survey respondents were asked to identify their institution and position to ensure duplicate responses were not included. If more than 1 individual from the same laboratory responded (n = 8), we analyzed the response from the individual who indicated they were a microbiologist and/or worked in the clinical microbiology/infectious disease laboratory (n = 6). If no respondent indicated that they specifically worked in the clinical microbiology/infectious disease laboratory (n = 2), we analyzed the response from the most senior individual. All questions were optional, and the respondents could opt out throughout the survey.
Descriptive statistical analyses were performed with Prism version 10.2.3 (GraphPad Software). The figure was created with Microsoft Excel version 16.88. As no protected information was collected, and all responses were analyzed in aggregate for deidentification purposes, the institutional review board at Vanderbilt University Medical Center deemed the survey exempt from review.
## r e S U lt S
A total of 137 unique institutions in 29 countries responded to the primary CP survey. Among these, 95.6% (131/137) of institutions representing 28 countries reported that they had at least 1 type of microbiology service, with parasitology (80.9%, 106/131) and bacteriology (77.9%, 102/131) being the most common laboratories available, while mycology (45.0%, 59/131) and virology (45.8%, 60/131) laboratories were less prevalent TABLE 1 . Approximately two-thirds of respondents (67.9%, 89/131) reported the ability to perform at least 1 rapid diagnostic/POC test. Thirty-five sites in 12 countries provided all 5 types of clinical microbiology/infectious disease testing queried in the survey (bacteriology, virology, mycology, parasitology, rapid diagnostic/POC): Côte d'Ivoire (1), Egypt (1), Ethiopia (2), Gabon (1), Ghana (2), Kenya (3), Malawi (1), Nigeria (17), South Africa (3), Sudan (1), Uganda (2), and Zambia (1). Among institutions that reported performing any type of microbiology testing, the number of institutions per country ranged from 1 (multiple countries) to 37 (Nigeria) FIGURE 1 .
Most clinical microbiology/infectious disease laboratories were located at public institutions (74.8%, 98/131), while 25.2% (33/131) were private laboratories. There was no difference in the proportion of public and private laboratories that provided bacteriology testing (77.6%, 76/98 vs 78.8%, 26/33; P > .99) or parasitology testing (81.6%, 80/98 vs 78.8%, 26/33; P = .80). A larger proportion of public laboratories provided mycology testing compared to private laboratories, but this difference was not statistically significant (49.0%, 48/98 vs 33.3%, 11/33; P = .16). However, a statistically significantly larger proportion of public laboratories performed virology testing compared to private laboratories (52.0%, 51/98 vs 27.3%, 9/33; P = .016).
The most frequently performed bacteriology test among the 102 sites that had a bacteriology laboratory was bacterial identification (90.2%, 92/102), followed by aerobic bacterial cultures and AST (both 89.2%, 91/102). At least 1 site in each of the 28 countries performed AST except the sites in Burundi (0/1), Equatorial Guinea (0/1), Lesotho (0/2), Liberia (0/2), and Sudan (0/1). The least frequently performed bacteriology test was the purified protein derivative (PPD) skin test (32.4%, 33/102) and mycobacterial cultures (42.2%, 43/102). Seventeen laboratories in 8 countries provided all 7
| O r i g i n a l a rt i c l e bacteriology services queried: Benin (1), Côte d'Ivoire (1), Egypt (1), Kenya (2), Nigeria (9), Rwanda (1), South Africa (1), and Zimbabwe (1). There was no statistically significant difference in the proportion of public and private laboratories that performed specific bacteriology testing.
Among the 60 sites that provided virology testing services, 86.7% (52/60) performed viral serology, 81.7% (49/60) performed nucleic acid testing via polymerase chain reaction, 33.3% (20/60) evaluated for the presence of viral proteins, and 16.7% (10/60) performed viral cultures. Six laboratories in 3 countries reported the ability to perform all 4 virology services: Côte d'Ivoire (1), Nigeria (3), and South Africa (2). The availability of specific viral testing did not differ significantly between public and private laboratories.
Among the 59 sites with mycology testing capabilities, 84.7% (50/59) performed fungal identification testing by any method (eg, antigen, antibody, molecular testing), 78.0% (46/59) performed fungal cultures, and 55.9% (33/59) performed fungal AST.
Thirty-one laboratories in 14 countries reported the ability to perform all mycology testing services: Cameroon (1), Côte d'Ivoire (1), Egypt (1), Ethiopia (1), Gabon (1), Ghana (1), Kenya (4), Malawi (1), Mozambique (1), Nigeria (12), South Africa (4), Sudan (1), Uganda (1), and Zimbabwe (1). No differences were observed in the proportion of public and private laboratories regarding specific mycology testing.
Institutions were asked if they had dedicated testing for 17 specific agents, including via rapid diagnostic and/or POC testing (Supplementary Table 1). The single most commonly tested agent was HIV, as 90.8% (119/131) of respondents performed at least 1 assay, with HIV antibody testing the most frequently employed methodology (89.9%, 107/119). Other commonly evaluated pathogens included Treponema pallidum (78.6%, 103/131), Plasmodium falciparum (76.3%, 100/131), Mycobacterium tuberculosis (76.3%, 100/131), and hepatitis C virus (74.8%, 98/131).
## D i S c U S S i O n
These findings demonstrate that the availability of clinical microbiology/infectious disease laboratory diagnostic testing is variable across Africa. The most commonly available microbiology services among these 131 sites were parasitology and bacteriology testing, while fewer than half of respondents reported virology (non-HIV) and mycology testing capabilities. Another notable finding was that approximately 11% of sites reported that they do not have the capability to perform bacterial AST. This included 5 countries with no sites that perform bacterial AST. This is an important finding given the increasing prevalence and clinically significant burden of bacterial antimicrobial resistance in Africa. 11 Another important finding was the relatively robust diagnostic capacity for certain infectious agents, including HIV and P falciparum. Availability of these tests is likely due to the high prevalence of these infections across Africa and their statistically significant contribution to mortality. 12 It is important to note that HIV testing, including viral load, is often performed by a public health laboratory in each country, but the logistical nuances of this testing (eg, centrally vs decentralized) were not included in this survey; thus, additional studies assessing these details are warranted. Finally, while 76% of sites reported that they can test for M tuberculosis, 32% and 42% of sites performed the PPD skin test and mycobacterial cultures, respectively. This is likely due to the use of molecular methods (eg, GeneXpert) that were not specifically queried in this survey, as the PPD skin test is not reliable given the high prevalence of Bacillus Calmette-Guérin vaccination and mycobacterial culture is impractical. 13 Nevertheless, given that the World Health Organization African Region accounts for 25% of new tuberculosis cases and 33% of deaths worldwide, 14,15 elucidating the actual M tuberculosis testing capacity is critical.
Pathology and laboratory medicine infrastructure has historically been neglected in low-and lower-middle-income countries and other low-resource regions. [16][17][18][19] However, the past 2 decades have witnessed greater attention to global health challenges, which has spurred increased funding, including for HIV/AIDS, tuberculosis, malaria, and neglected tropical diseases. 16,20 This is reflected in the increasing capacity of these diagnostic clinical microbiology/ infectious disease laboratory services identified in this study. Nevertheless, the finding that fewer than half of respondents provide virology services (other than HIV) is concerning, highlighting an area of diagnostic testing with room for improvement to ensure the ability to respond to the next pandemic, which is likely to be viral. Challenges that should be addressed to continue improving the availability and quality of these services include education, training, and retention of qualified laboratory staff and laboratory directors; procurement and maintenance of equipment and reagents; and quality assurance and quality control processes. 21 Given the importance of a well-functioning clinical microbiology/infectious disease laboratory to the health care system, continued investment is integral to reducing morbidity and mortality.
Limitations to this study include that survey respondents comprised primarily large, tertiary care centers. This likely reflects that these institutions are more likely to have an individual included in the ASCP email listserv; however, these institutions are also more likely to have the capabilities to perform many of the more advanced laboratory assays queried in this survey. As such, we believe the findings from this sample provide important insight into the clinical microbiology/infectious disease laboratory diagnostic capabilities across Africa. Nevertheless, no responses were received from multiple countries, which may be partially attributable to the survey only being distributed in English. Further, there is not a comprehensive database of all laboratories in Africa, and a subset of laboratories are small, unaccredited establishments; however, the number of internationally accredited laboratories (defined as those that have received ISO 15189 accreditation) was reported to be 668 as of 2020. 22,23 If this number is accurate, and if all of the laboratories that responded to our survey were "internationallyaccredited," we received responses from approximately 20% of all accredited laboratories in Africa. However, as these are only assumptions, and it is not known how many of these laboratories provide clinical microbiology/infectious disease testing, we are unable to accurately assess the comprehensiveness of our survey. Finally, while we attempted to include many laboratory assays performed in clinical microbiology/infectious disease laboratories, this survey does not include all possible testing modalities or | O r i g i n a l a rt i c l e pathogens. Therefore, investment in external, on-site observation that aims to characterize more nuanced testing capabilities would be informative.
In summary, despite the limitations of this survey, the findings presented here provide contemporary data regarding the availability of critical clinical microbiology/infectious disease laboratory testing capabilities among institutions in Africa. These results and future additional studies will be crucial for understanding where strategic investment in the laboratory and public health infrastructure is warranted.
## References
1. Baron (2019) "Clinical microbiology in underresourced settings" *Clin Lab Med*
2. Ombelet, Ronat, Walsh (2018) "Clinical bacteriology in low-resource settings: today's solutions" *Lancet Infect Dis*
3. Moyo, Mhango, Moyo et al. (2023) "Emerging infectious disease outbreaks in sub-Saharan Africa: learning from the past and present to be better prepared for future outbreaks"
4. Nkengasong, Tessema (2020) "Africa needs a new public health order to tackle infectious disease threats" *Cell*
5. Petti, Polage, Quinn et al. (2006) "Laboratory medicine in Africa: a barrier to effective health care" *Clin Infect Dis*
6. Hunsperger, Juma, Onyango (2019) "Building laboratory capacity to detect and characterize pathogens of public and global health security concern in Kenya"
7. Murray, Ikuta, Sharara (2022) "Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis" *Lancet*
8. Musa, Okoliegbe, Abdalaziz et al. (2023) "Laboratory surveillance, quality management, and its role in addressing antimicrobial resistance in Africa: a narrative review" *Antibiotics (Basel)*
9. Jacobs, Milner, Shibemba et al. (2024) "Anatomic and clinical pathology services and infrastructure in Zambia" *Am J Clin Pathol*
10. Jacobs, Stephens, Milner Da Jr (2023) "Survey of blood collection and transfusion practices among institutions in Africa" *Transfusion*
11. Resistance (2024) "The burden of bacterial antimicrobial resistance in the WHO African region in 2019: a crosscountry systematic analysis" *Lancet Glob Health*
12. Niohuru "Healthcare and Disease Burden in Africa: The Impact of Socioeconomic Factors on Public Health"
13. Williams, Calnan, Edem (2022) "GeneXpert rollout in three high-burden tuberculosis countries in Africa: a review of pulmonary tuberculosis diagnosis and outcomes from 2001 to 2019" *Afr J Lab Med*
14. (2024) "A spotlight on the tuberculosis epidemic in South Africa" *Nat Commun*
15. Nkengasong, Nsubuga, Nwanyanwu (2010) "Laboratory systems and services are critical in global health: time to end the neglect?" *Am J Clin Pathol*
16. Nkengasong, Mesele, Orloff (2009) "Critical role of developing national strategic plans as a guide to strengthen laboratory health systems in resource-poor settings" *Am J Clin Pathol*
17. Birx, De Souza, Nkengasong (2009) "Laboratory challenges in the scaling up of HIV, TB, and malaria programs: the interaction of health and laboratory systems, clinical research, and service delivery" *Am J Clin Pathol*
18. Wilson, Fleming, Kuti et al. (2018) "Access to pathology and laboratory medicine services: a crucial gap" *Lancet*
19. (2024) "The Maputo declaration on strengthening of laboratory systems"
20. Masanza, Nqobile, Mukanga et al. (2010) "Laboratory capacity building for the International Health Regulations (IHR[2005]) in resource-poor countries: the experience of the African Field Epidemiology Network (AFENET)" *BMC Public Health*
21. (2014) "Proportion of internationally accredited medical/clinical laboratories in Africa from"
22. Odhiambo, Van Der Puije, Maina (2023) "Examining 7 years of implementing quality management systems in medical laboratories in sub-Saharan Africa" *Trop Med Int Health* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12820486&blobtype=pdf | # Urinary tract infection in outpatients in Germany: a cross-sectional study of diagnostics and susceptibility testing in medical laboratories
Hannah Bender, Kathrin Jobski, Guido Schmiemann, Axel Hamprecht, Falk Hoffmann
## Abstract
Background: Urinary tract infections (UTIs) are common, representing a frequent cause of antibiotic prescription in primary care worldwide. Selection of antibiotics for antimicrobial susceptibility testing and the reporting of test results by laboratories can directly impact antibiotic prescribing and guideline adherence.Aim: To assess the current practice of susceptibility testing by laboratories for outpatient UTIs in Germany.Design & setting: A cross-sectional study was conducted including all laboratories identified by searching for specialists in laboratory medicine and microbiology on the websites of the 17 German associations of statutory health insurance physicians.Method: Between January and April 2024, a survey using a standardised questionnaire was conducted across identified laboratories.Results: Of the 396 laboratories identified, 65.2% (n = 258) replied. Of these, 106 laboratories performed susceptibility testing and on average tested for 13.1 (standard deviation [SD] 3.6) different antibiotics in a urine culture positive for Escherichia coli. The most commonly tested antibiotics were ciprofloxacin (98.1%), co-trimoxazole (97.2%), cefuroxime, and nitrofurantoin (both 91.5%). On average, laboratories tested 3.8 of the five antibiotics recommended in the German guidelines on uncomplicated UTI, with 26.4% testing for all five. Laboratories received clinical information on previous treatments and comorbidities in an estimated one-fifth (on average 21.3% and 21.5%, respectively) of the urine samples, and information on the type of the urine sample in an estimated three-fifths (63.7%) of samples.
Conclusion:Laboratories should test and report as many first-line antibiotics as possible. Further, a more detailed and standardised transfer of clinical information to laboratories could enhance the quality of antibiotic prescribing.
How this fits inThis study provides novel insights into the current practice of antimicrobial susceptibility testing for outpatient urinary tract infections (UTIs) in Germany. Current susceptibility testing for outpatient UTIs in Germany varies, with 84.9% of laboratories testing for ≥3 first-line antibiotics while 15.1% test for Research ≤2 antibiotics. Laboratories receive clinical information about previous treatments and comorbidities in only about 20% of the urine samples. Standardising clinical information transfer between outpatient physicians and laboratories could improve guideline adherence and antibiotic selection.
## Introduction
Urinary tract infections (UTIs) are among the most common infections and represent a frequent cause of medical consultations in primary care. 1 More than 150 million people worldwide are affected by community-acquired UTIs each year. 2 Although UTIs can be self-limiting, they are a common reason for antibiotic prescribing. 1,3,4 As antibiotic resistance is increasing worldwide, particularly among gram-negative bacteria, it is important to avoid its inappropriate use. 5,6 However, several studies have shown that treatment of UTIs in the outpatient setting often does not follow guideline recommendations, which can lead to inappropriate antibiotic use. 4,7,8 Although Malmros et al identified differences in the guidelines across 15 European countries, 12 of the 15 guidelines recommended nitrofurantoin as the first-line treatment for uncomplicated UTIs, followed by pivmecillinam and fosfomycin, which were listed second and third most frequently. 9 The two previous German guidelines, as well as the latest interdisciplinary guideline, recommend as first-line treatment for uncomplicated UTIs fosfomycin trometamol, nitrofurantoin, nitroxoline, and pivmecillinam. In case of local resistance rates of <20% trimethoprim can be equally considered. 10,11 Moreover, the guidelines recommend against the use of fluoroquinolones and cephalosporins as first-line treatment for outpatient UTIs. [10][11][12] In selecting an appropriate antibiotic, a urine culture can be a useful tool. The German guideline on uncomplicated UTI recommends urine cultures in cases of ambiguous symptoms, or recurrent or complicated UTIs. 12 The interpretation of a urine culture result not only depends on bacterial count and the pathogen detected, but also on the type of urine collection, the presence or absence of leukocyturia, and other clinical information from the patient. 13 In instances where a urine culture is conducted and typical uropathogens are identified, the laboratory performs susceptibility testing for a pre-defined range and number of antibiotics. 14 Studies show that the reporting of susceptibility testing varies, which can directly impact antibiotic prescribing and guideline adherence in the treatment of UTIs. [15][16][17][18] An earlier German study found that microbiological tests often did not include the first-line antibiotics recommended in the guidelines. 14 However, this study was conducted >10 years ago and covered only northern Germany. 14 Furthermore, the exchange between laboratories and primary care is less well studied.
Therefore, the aim of this cross-sectional study was two-fold, to describe a) the current practice of susceptibility testing by laboratories for outpatient UTIs in Germany, including the range of antibiotics tested; and b) the communication and exchange of information between outpatient physicians and laboratories.
## Method Study design and study population
A cross-sectional study was conducted among medical laboratories in Germany using a standardised questionnaire (data collected January-April 2024). As no list of all microbiology laboratories in Germany was available, laboratories were identified by searching the websites of all 17 German associations of statutory health insurance (SHI) physicians for all registered specialists in the two relevant medical specialties (microbiology, virology, and infectious epidemiology, as well as laboratory medicine). Using their registered addresses, we identified the corresponding laboratories and determined contact details via internet research. In two of the 17 associations (that is, Bavaria and Rhineland-Palatinate) no search via their websites was offered. In the case of Bavaria, the laboratories were identified via the Bavarian Medical Association, and in the case of Rhineland-Palatinate, via the website of the Federal Ministry of Health (https://gesund.bund.de). In addition, the list of laboratories was compared with the list of accredited laboratories (Deutsche Akkreditierungsstelle) and supplemented where necessary. Participation in the survey was possible by telephone, email, or fax. The questionnaire could be completed by medical or laboratory staff. In a first step, laboratories were contacted by telephone. During the initial telephone call, the interviewer provided a brief explanation of the survey's content and requested to be connected with medical staff or other appropriately qualified laboratory staff. It was possible to take part in the survey directly, to arrange a telephone appointment, or to receive further information and the questionnaire by email or fax. When participating by telephone, the interviewer filled out the questionnaire. If a laboratory could not be reached after five attempts by telephone at different times, it was contacted by email. If an email was sent successfully, no further contact was made.
The methodological approach (identifying laboratories, contacting them, and piloting of the questionnaire) was pre-tested in two federal German states (that is, Bremen and Lower Saxony). The pre-test led to a change in the method of initial contact (that is, from fax to telephone).
## Instrument and included variables
The questionnaire consisted of eight items. Initially, besides the federal state, the laboratory was asked if it performed susceptibility testing for outpatient UTI. If yes, the laboratory was asked which testing standards were used. Answers included the Clinical and Laboratory Standards Institute, the European Committee on Antimicrobial Susceptibility Testing (EUCAST), and the National Antimicrobial Susceptibility Testing Committee. Responders could provide a free-text response and multiple responses were allowed. As Escherichia coli (E. coli) is clearly the most common pathogen causing uncomplicated and complicated UTIs, laboratories were asked to report on their practice of susceptibility testing for this species. 19 Additionally, the bacterial count (in colony-forming units [CFU]/ ml), which prompts testing of E. coli in pure culture, was requested (options '10² CFU/ml', '10³ CFU/ ml', '10⁴ CFU/ml', and 'other'). 10,11 Laboratories were also asked to list all antibiotics that are routinely tested for susceptibility for E. coli. The questionnaire listed 14 antibiotics (amoxicillin, amoxicillin and clavulanic acid, cefpodoxime, cefuroxime, ciprofloxacin, co-trimoxazole, fosfomycin, levofloxacin, nitrofurantoin, nitroxoline, norfloxacin, ofloxacin, pivmecillinam, and trimethoprim) in alphabetical order, including those recommended in the previous German guidelines, 10,11 in the quality standards (Mikrobiologisch-infektiologische Qualitätsstandards [MIQ]), 13 and a group of frequently used antibiotics. 4 Laboratories could also list any additional substances tested. Subsequently, situations were asked in which susceptibility tests are conducted for antibiotics not previously mentioned or for a different number of substances, and it was possible to explain these situations in more detail. In addition, participants were asked which information (besides age and sex) they would like to receive with the urine samples. Lastly, the proportion of urine samples with accompanying information (type of urine collection, antibiotic pre-treatment, and comorbidities) should be estimated.
## Statistical analysis
Descriptive statistics (mean, standard deviation [SD], median, and percentages) were used. Inductive categorisation was performed to summarise the open-ended questions. Participants' responses were categorised and quantified by two independent persons (HB and KJ). If there was a disagreement, a third person was consulted. Denominators vary depending on the question. SPSS Statistics for iOS (version 28.0.1.0) was used for all statistical analyses. A total of 396 laboratories were identified (Figure 1). Of these, 65.2% (n = 258 laboratories) provided information on whether or not they perform susceptibility testing in outpatient UTIs, with 76.7% (n = 198) of these responding via telephone. A total of 106 laboratories performed susceptibility testing for outpatient UTI (Table 1).
## Results
## Baseline characteristics of the study population
## Test standards and bacterial count limit used
Overall, 101 (96.2%) laboratories performed the antibiotic susceptibility testing according to EUCAST (Table 2). A total of 50.9% (n = 54) of laboratories performed susceptibility tests for E. coli in pure culture with a bacterial count of 10 3 CFU/ml. A total of 19.8% (n = 21) and 4.7% (n = 5) of the laboratories carried out susceptibility tests with a bacterial count of 10 4 CFU/ml and ≥10 5 CFU/ml, respectively. In total, 8.5% of laboratories reported that factors such as inhibitor test results and the presence of leukocytes influence the bacterial count limit used to decide when to perform antimicrobial susceptibility testing.
## Antibiotics tested
On average, laboratories performed susceptibility tests for 13.1 (SD 3.6) antibiotics (Table 3).
Laboratories most commonly tested for ciprofloxacin (98.1%), co-trimoxazole (97.2%), cefuroxime, and nitrofurantoin (both 91.5%), while 72.6% tested for trimethoprim and 34.0% for nitroxoline (Figure 2). Susceptibility testing of substances not previously listed in the questionnaire was reported by 67.9% of laboratories, with meropenem (50.9%), piperacillin and tazobactam (49.1%), ceftazidime (41.5%), and cefotaxime (38.7%) being the most common antibiotics. The majority of laboratories (80.8%, n = 84) reported performing susceptibility testing for different antibiotics in specific cases. In particular, 68.3% (n = 56) reported performing such tests for (multidrug)resistant pathogens. In addition, 31.7% (n = 26) reported performing susceptibility testing at the request of the sender, 13.4% (n = 11) reported doing so for specific patient groups (for example, hospitalised patients or children), while 9.8% (n = 8) reported performing these tests to detect other pathogens (Table 2).
On average, laboratories tested for 3.8 (SD 1.1) of the five antibiotics recommended in the recent guidelines, with only 26.4% (n = 28) of laboratories testing for all (Table 3). A total of 84.9% (n = 90) of laboratories tested for ≥3 first-line antibiotics, while 15.1% (n = 16) tested ≤2.
## Information accompanying the urine sample
A total of 101 laboratories provided information on the desired accompanying information for urine samples (Table 2). Of those, 56.4% (n = 57) desired detailed clinical information, particularly information about pregnancy (12.9%, n = 13) or immunosuppression (7.9%, n = 8). In addition, 55.4% (n = 56) stated that they needed information about current or previous antibiotic therapy. Further, 45.5% considered precise information on the type of urine sample, particularly information on catheter urine (19.8%, n = 20), to be particularly relevant. In addition, 28.7% (n = 29) considered it necessary to provide a diagnosis or suspected diagnosis and 18.8% (n = 19) to indicate results of any urine dipstick examination and evidence of leukocyturia.
The participants estimated that, on average, they received information on the exact type of urine sample in 63.7% (SD 37.0) of the cases. With regard to possible previous treatment or existing concomitant diseases, information was estimated to be available for a mean proportion of only 21.3% (SD 23.4) and 21.5% (SD 24.5) of the urine samples, respectively.
## Discussion Summary
In our cross-sectional study, we found that only 26.4% of German laboratories tested for all five antibiotics recommended in the guidelines for uncomplicated UTI, with trimethoprim (72.6%) and nitroxoline (34.0%) being tested the least. The most frequently tested antibiotic was ciprofloxacin (98.1%), followed by co-trimoxazole (97.2%), cefuroxime (91.5%), and nitrofurantoin (91.5%). Furthermore, clinical information received about possible previous treatment and the presence of concomitant diseases was estimated to be available in about one-fifth of urine samples only (on average 21.3% and 21.5%, respectively).
## Strengths and limitations
The high response rate of 65.2% is a notable strength of our study. However, out of the 258 participating laboratories, only 106 performed susceptibility testing for outpatient UTIs. Since no publicly available lists of all outpatient laboratories in Germany exists, it remains unclear whether all laboratories primarily serving outpatients could be reached in the survey, despite using various sources to identify them. Another limitation is that only estimates of the proportion of information accompanying urine samples were requested and no samples were analysed. Beyond that, no further follow-up questions were asked, such as why (recommended) antibiotics were not tested or how the susceptibility tests were technically performed, as we aimed to keep the questionnaire concise to encourage participation. Additionally, information exchange (for example, data transmission and the space allowed for clinical details) and the communication between the laboratory and the outpatient
## Table 2 Continued
Table 3 Antibiotic testing for outpatient urinary tract infections in German laboratories (N = 106) 10,11 Category n (%) a a Unless otherwise stated. b The two valid guidelines 10,11 at the time of data collection recommend as firstline treatment for uncomplicated urinary tract infections fosfomycin trometamol, nitrofurantoin, nitroxoline, pivmecillinam, and trimethoprim, with local resistance rates of <20%. IQR = interquartile range. SD = standard deviation.
physicians was not explored. Furthermore, the information provided relates only to the testing of E. coli from urine cultures.
## Comparison with existing literature
While most laboratories tested for nitrofurantoin, fosfomycin, and pivmecillinam, considerably fewer included trimethoprim and, above all, nitroxoline. The three most frequently tested antibiotics include a fluoroquinolone (ciprofloxacin) and a cephalosporin (cefuroxime), which should only be used as second-line antibiotics or for complicated infections owing to side effects and the increased risk of antibiotic resistance. 12,20 Our results are comparable with those of Schmiemann et al for northern Germany in the study from 2013, when ciprofloxacin (95%) and cefuroxime (93%) were also among the most frequently tested antibiotics, while only 58% of laboratories tested for trimethoprim. No comparative data are available for pivmecillinam, which has been available in Germany since 2016. 14 Reasons why laboratories may not perform susceptibility testing for trimethoprim may be the testing for co-trimoxazole, the lack of inclusion of trimethoprim in susceptibility testing cards, or the local resistance rates. 21 Nevertheless, studies suggest that trimethoprim resistance rates in community-acquired UTIs may be overestimated, as routinely collected data typically reflect resistance rates in complicated UTIs, whereas no further diagnostic measures are usually performed in uncomplicated UTIs. 10,11,22 Only 34% of laboratories performed standard susceptibility testing for nitroxoline, although it has also been included in the German guidelines since 2017 and can be an alternative as it has a low in vitro resistance rate in E. coli and retains a good activity even in multidrug-resistant isolates. 23,24 One main reason for this may be that many laboratories are using automated systems to perform susceptibility testing, which can deliver results in less time and at lower cost. 25 However, nitroxoline is not included in most standard test panels of automated systems and therefore needs to be tested in addition, at extra costs. 26 Nitroxoline is also rarely used in other European countries and is not included in the updated European Society of Urology guideline for the treatment of UTIs. 9,27 Another reason why laboratories may not test for all the antibiotics listed in the guidelines is the fee schedule of the National Association of Statutory Health Insurance Funds, which pays a flat rate for testing. In addition, clinical guidelines do not recommend routine urine culture, so most samples tested in laboratories are likely to be from complicated cases. 12,28 However, most laboratories do not have the information if the urine is from an uncomplicated or complicated infection, and will therefore likely test antibiotics for both indications. Nevertheless, some first-line antibiotics often can be used in complicated UTIs when susceptibility data are available, and the kidneys are not involved. This may help to reduce reliance on second-line antibiotics, such as ciprofloxacin, to combat antimicrobial resistance. 12,20 With an average of 13.1 (SD 3.6) different antibiotics, the laboratories in our study tested a wide range of substances, which increases the risk for reporting a large number of susceptible antibiotics to the outpatient physicians without limiting the choice of antibiotic therapies. This might hamper optimal treatment as the performance of susceptibility testing and the reporting have a direct influence on the prescription of antibiotics for UTIs. 15,16,29 Even more, selective reporting of susceptibility test results can be a powerful tool in antibiotic stewardship and can lead to greater guideline adherence when prescribing antibiotics for UTIs. 16,29 The laboratories received clinical information on previous treatments and comorbidities in only an estimated one-fifth (on average 21.3% and 21.5%) of urine samples, and information on the type of the urine sample in an estimated three-fifths (63.7%) of samples. However, results of a urine culture need be interpreted in light of the type of urine sample and patients' clinical presentation, as bacteriuria alone does not constitute a UTI and justify antibiotic therapy. 12,27 Our study indicates that microbiological laboratories often lack information to select the optimal antibiotics for testing and therapy. The German quality standards MIQ emphasises that meaningful assessment of a urine culture requires clinical information about the patient, and discusses the significance of leukocyturia and the type of urine collection. 13 Neither the recent German guideline nor the European Association of Urology's guideline contain any recommendations regarding the specific information that should be provided to a laboratory accompanying a urine sample. 12,28
## Implications for research and practice
In conclusion, laboratories in Germany tested 13.1 different antibiotics for isolates from urines, but only 26.4% tested for all five antibiotics recommended by German guidelines for uncomplicated UTI. Ideally, laboratories should test all first-line antibiotics recommended in the guidelines. The inclusion of pivmecillinam and nitroxoline in more automated systems could increase the testing frequency of first-line antibiotics in the future. Furthermore, it would be beneficial to prioritise the reporting of first-line antibiotics. Such an approach provides the outpatient physician with an overview of the local resistance situation and could support clinical decision making for future patients. Further highlighting first-line antibiotics in laboratory reports may help to align prescribing practices more closely with current recommendations, reducing the unnecessary use of broad-spectrum antibiotics and contributing to improved antibiotic stewardship.
In addition, laboratories often seem to lack clinical information, which may influence the selection of antibiotics for susceptibility testing. The optimal interpretation of laboratory results also depends on relevant clinical information. Therefore, a more detailed and standardised transfer of clinical information from outpatient physicians to laboratories could improve the quality of antibiotic prescribing and compliance with antimicrobial stewardship.
However, a concise and clearly defined list of specific recommendations on how and what relevant information should be shared with laboratories is still lacking and should be included in the updated guidelines.
## Funding
This study received no external funding.
## Ethical approval
The article does not include studies on humans or animals. A corresponding waiver (reference: 2023-043, dated 1 February 2023) has been received from the ethics committee of the Carl von Ossietzky Universität Oldenburg.
## Provenance
Freely submitted; externally peer reviewed.
## Data
The dataset relied on in this article is available from the corresponding author on reasonable request.
## References
1. Butler, Hawking, Quigley et al. (2015) "Incidence, severity, help seeking, and management of uncomplicated urinary tract infection: a population-based survey" *Br J Gen Pract*
2. Yang, Chen, Zheng (2022) "Disease burden and long-term trends of urinary tract infections: a worldwide report" *Front Public Health*
3. Hoffmann, Peiris, Mar (2020) "Natural history of uncomplicated urinary tract infection without antibiotics: a systematic review" *Br J Gen Pract*
4. Schmiemann, Hoffmann, Hamprecht et al. (2022) "Patterns and trends of antibacterial treatment in patients with urinary tract infections, 2015-2019: an analysis of health insurance data" *BMC Prim Care*
5. Murray, Ikuta, Sharara (2022) "Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis" *Lancet*
6. Schneidewind, Stangl, Dräger (2022) "What is the proportion of infectiology in the specialization urology?: a pilot study to underline the significance of antibiotic stewardship in urology]" *Urologie*
7. Kranz, Schlager, Mühlstädt (2019) "Barriers to guideline adherence: identification of barriers to guideline adherence using a survey on the AWMF S3 guideline epidemiology, diagnosis, treatment, and management of uncomplicated bacterial, community-acquired urinary tract infections in adult patients]" *Urologe A*
8. Goebel, Trautner, Grigoryan (2021) "The five Ds of outpatient antibiotic stewardship for urinary tract infections" *Clin Microbiol Rev*
9. Malmros, Huttner, Mcnulty (2019) "Comparison of antibiotic treatment guidelines for urinary tract infections in 15 European countries: results of an online survey" *Int J Antimicrob Agents*
10. Schmiemann, Gebhardt, Hummers (2018) "DEGAM S3 guideline: burning sensation when urinating]"
11. Fünfstück, Helbig, Hofmann (2017) "DGU guideline programme interdisciplinary S3 guideline: epidemiology, diagnosis, treatment, prevention and management of uncomplicated, bacterial, community-acquired urinary tract infections in adult patients]"
12. (2024) "3 guideline: epidemiology, diagnosis, treatment, prevention and management of uncomplicated, bacterial, community-acquired urinary tract infections in adult patients, update]"
13. Schubert, Fünfstück, Gatermann (2020) "MiQ 02: urinary tract infections: quality standards in microbiological and infectious disease diagnostics]"
14. Schmiemann, Noll, Hoffmann (2016) "Resistance testing for urinary tract infections. a barrier to guideline implementation]" *Urologe A*
15. Langford, Daneman, Diong (2021) "Antibiotic susceptibility reporting and association with antibiotic prescribing: a cohort study" *Clin Microbiol Infect*
16. Simon, Fougnot, Monchy (2023) "Impact of selective reporting of antibiotic susceptibility testing results for urinary tract infections in the outpatient setting: a prospective controlled before-after intervention study" *Clin Microbiol Infect*
17. Schuster, Tigges, Grune (2023) "GPs' perspective on a multimodal intervention to enhance guidelineadherence in uncomplicated urinary tract infections: a qualitative process evaluation of the multicentric RedAres cluster-randomised controlled trial" *Antibiotics (Basel)*
18. Schmiemann, Greser, Maun (2023) "Effects of a multimodal intervention in primary care to reduce second line antibiotic prescriptions for urinary tract infections in women: parallel, cluster randomised, controlled trial" *BMJ*
19. Flores-Mireles, Walker, Caparon et al. (2015) "Urinary tract infections: epidemiology, mechanisms of infection and treatment options" *Nat Rev Microbiol*
20. Bender (2025) *BJGP Open*
21. Stapleton, Wagenlehner, Mulgirigama et al. (2020) "Escherichia coli resistance to fluoroquinolones in community-acquired uncomplicated urinary tract infection in women: a systematic review" *Antimicrob Agents Chemother*
22. Stoltidis-Claus, Rosenberger, Mandraka (2016) "Antimicrobial resistance of clinical Enterobacterales isolates from urine samples" *Euro Surveill*
23. Klingeberg, Noll, Willrich (2018) "Antibiotic-resistant E. coli in uncomplicated community-acquired urinary tract infection" *Dtsch Arztebl Int*
24. Fuchs, Hamprecht (2019) "Susceptibility of carbapenemase-producing Enterobacterales (CPE) to nitroxoline" *J Antimicrob Chemother*
25. Plambeck, Fuchs, Sattler et al. (2022) "In vitro activity of mecillinam, temocillin and nitroxoline against MDR Enterobacterales" *JAC Antimicrob Resist*
26. Gajic, Kabic, Kekic (2022) "Antimicrobial susceptibility testing: a comprehensive review of currently used methods" *Antibiotics (Basel)*
27. Schaumburg, Gatermann, Becker (2018) "Guidelines for interpretation required" *Dtsch Arztebl Int*
28. Kranz, Bartoletti, Bruyère (2024) "European Association of Urology guidelines on urological infections: summary of the 2024 guidelines" *Eur Urol*
29. Bonkat, Bartoletti, Bruyère (2024) "EAU guidelines on urological infections"
30. Bourdellon, Thilly, Fougnot (2017) "Impact of selective reporting of antibiotic susceptibility test results on the appropriateness of antibiotics chosen by French general practitioners in urinary tract infections: a randomised controlled case-vignette study" *Int J Antimicrob Agents* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12482263&blobtype=pdf | # Epidemiology and genetic diversity of norovirus GII genogroups among pediatric patients in Beijing, China, during 2023-2024
Junhong Ai, Qiliang Li, Ke Xu, Yuxuan Li, Ying Liu, Luci Huang, Wenqi Song, Zhengde Xie
## Abstract
Background Norovirus is an important cause of viral acute gastroenteritis (AGE) worldwide.
MethodsIn order to characterize the molecular epidemiology and genetic diversity of norovirus in children in Beijing, 3634 anal swab samples of AGE patients from January 2023 to December 2024 were analyzed. Norovirus was detected using RT-PCR and genotyped by sequencing the partial RdRp and VP1 region.
ResultsDuring the two-year period, norovirus was detected in 19.6% of AGE cases, with the highest detection rate in children under 3 years of age. GII.4 and GII.P16 were the dominant genotypes of VP1 and RdRp, with a detection rate of 36.39% and 44.59%, respectively. According to the dual-typing system combined the RdRp and VP1, the dominant genotypes of norovirus changed between 2023 and 2024. In 2023, the most common genotype was GII.3[P12] (39.15%), followed by GII.4 Sydney[P16] (32.34%) and GII.4 Sydney[P31] (15.32%). However, in 2024, the dominant genotype was GII.17[P17] (41.43%), followed by GII.4 Sydney[P16] (34.29%) and GII.3[P12] (20.0%). The GII.17 variants in this study were divided into two clusters: cluster IIIa and IIIb, which shared high nucleotide identity with GII.17 variant emerged in 2014/2015. Significantly, GII.4 Sydney[P31] and novel GII.4 Sydney[P16] variants co-circulating in this region from 2023 to 2024.
ConclusionThe data provided useful information on the molecular epidemiology of norovirus in sporadic AGE among children and highlighted the necessary to continuously monitor the epidemiological characteristics of norovirus associated AGE.
## Background
Acute gastroenteritis (AGE) is a major public health problem worldwide. More than 1 million deaths due to AGE occur globally each year [1]. AGE can be caused by bacteria, viruses or parasites. Norovirus is one of the leading viral pathogen cause of AGE in all-age patients, responsible for over 200 000 deaths annually in lowincome and middle-income countries [2][3][4]. Norovirus is an enteric non-enveloped single-stranded RNA virus belongs to the Caliciviridae family and Norovirus genus. The whole genome of norovirus is approximately 7.5-7.7 kb encoding three open reading frames (ORFs). ORF1 encodes six nonstructural proteins, including RNA-dependent RNA polymerase (RdRp), which is responsible for viral RNA replication. ORF2 and ORF3 encode for the structural proteins, viral protein 1 (VP1) and VP2, respectively. The recombination of norovirus genome often occurs between overlapping regions of ORF1 and ORF2, leading to the emergence of new strains containing different combinations of RdRp and VP1 [5].
Norovirus can be classified based on the complete amino acid (aa) diversity of VP1 (VP1 genotyping) and partial of nucleotide diversity of RdRp (RdRp P-typing) according to the dual-typing system [6]. At present, there were 10 genogroups (GI-GX) and 49 genotypes based on the VP1 genotyping [7,8]. Among them, GII is responsible for the majority of norovirus AGE in humans and GII.4 genotypes are the most prevalent worldwide. Approximately every 3-5 years, new GII.4 variants emerged and replaced previous circulating strains due to aa substitution on major neutralization epitopes, which allow the new virus to effectively evade immune attacks [9]. Since the first pandemics of GII.4 95/96-US variant in mid-1990s, five new GII.4 variants have emerged and caused pandemic of norovirus globally [4].
Although the GII.4 variants are the most prevalent genotype circulating worldwide, new emerged non-GII.4 variants, such as GII.17 and GII.2 have been reported in both outbreak and sporadic epidemic cases since 2014 [10][11][12][13][14][15][16][17][18]. Furthermore, during 2023-2024 an increased GII.17 norovirus outbreaks and sporadic infection were reported in six European countries and the United States in AGE patients [19].
Noroviruses are genetically diverse and evolving through rapid antigenic drift in common genotypes. Therefore, it is essential to monitor the variation, genetic diversity, and evolution of norovirus overtime. However, there were few studies about the epidemiology and genetic diversity of norovirus in sporadic AGE cases in China, recently. Here, we describe the molecular epidemiology of norovirus in sporadic AGE patients in Beijing, China.
## Methods
## Sample collection
From January 2023 to December 2024, anal swab samples of AGE patients less than 18 years old from Beijing Children's Hospital were collected. AGE patients were characterized as who suffered from vomiting or ≥ 3 liquid or semi liquid fecal within 24 h. All the samples were stored at -80℃.
## Viral RNA extraction and detection
For each sample, 1 mL of saline was added and vibrated violently three times. Approximately 200 uL of supernatant was used for RNA extraction. The nucleic acid was extracted using NucliSens easyMAG system and Viral Nucleic Acid Extraction Kit (BioPerfectus, China) according to the manufacturer's instructions and stored at -80℃ immediately. Norovirus GI, GII, and GIV RNA was detected with a commercial reverse transcription PCR kit (LAND MEDICAL, China) employing universal primers. The reverse transcription PCR was performed in a 50 µL reaction mixture containing 35.8 µL of RT buffer, 4.2 µL of enzyme mix, and 10 µL of RNA. The thermal cycling protocol consisted of cDNA synthesis at 42 °C for 30 min, followed by reverse transcriptase inactivation and initial denaturation at 95 °C for 3 min, and 40 cycles of amplification (95 °C for 10 s, 60 °C for 1 min). Samples testing positive for norovirus with a threshold cycle value below 36 were subjected to genotyping.
## Norovirus genotyping
For genotyping, partial RdRp sequences in ORF1 and VP1 sequences in ORF2 were amplified with conventional one-step RT-PCR (Takara, China). In this study, we mainly focus on GII norovirus. The classical primers MON431 and G2SKR were used for genotype [20]. For each 25 uL reaction system, 5 uL RNA template was added. The thermal cycling condition were carried out as follows: reverse transcribed for 30 min at 50℃, followed by a denaturation step at 94℃ for 15 min and 40 cycles of amplification at 94℃ for 15 s, 50℃for 30 s, and 68℃ for 1 min. The PCR products was 570 bp and sequenced directly by Sanger sequencing (SinoGeno Max, China).
The genotypes of sequences obtained in this study were confirmed in terms of closest homology with GenBank reference sequence, using Basic Local Alignment Search Tool (blastn).
## Phylogenetic analysis
Phylogenetic trees of partial RdRp or VP1 were constructed using the neighbor-joining method with 1000 bootstrap replicates in MEGA 7. The norovirus reference sequences were obtained from the National Center for Biotechnology Information (NCBI) database. Nucleotide sequences obtained in this study were submitted to the GenBank database, the accession numbers are: PV759851-PV760155.
## Definition of seasons
The definition of seasons as follow in this study: Spring spans from March to May, Summer lasts from June to August, Autumn covers September to November, Winter extends from December to February of the following year.
## Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics 19.0. Non-normally distributed continuous data were presented as median (minimum-maximum). Categorical data were presented as frequency and percentage.
The Chi-square test was used for group comparisons. The P-value < 0.05 was considered statistically significant.
## Results
## Epidemiological features of norovirus
A total of 3634 patients with AGE were enrolled in this study from January 2023 to December 2024. Norovirus GI, GII, and GIV RNA was detected in 712 samples with a positive rate of 19.6%. There was a significant difference in the positive rate of norovirus between years (χ 2 = 14.799, P = 0.000) (Table 1).
The sex ratio of male to female is 1.51:1. There was no statistical difference in the positive rate of norovirus between gender. The median age of patients was 5.0 years (range from 0.01 to 17.0 years). Four age groups were divided in this study. The highest detection rate of norovirus was observed in group from 1 to 3 years of age, with a positive rate of 27.8%, followed by the group less than 1 year of age (27.5%) and the group from 3 to 5 years (20.7%). There was a statistical difference among age groups (χ 2 = 100.685, P = 0.000) (Table 1).
The positive rate of norovirus in different months among AGE patients is shown in Fig. 1. In 2023, there was one epidemic peak from March to May. While in 2024, norovirus had two distinct peaks, from January to March and from October to December.
## Genotype distribution of norovirus GII genogroup
Norovirus can be divided into 10 genogroups (G1-GX), of which GII is the main genogroup causing AGE in humans. Therefore, this study will focus on the analysis of norovirus GII. Total of 305 positive samples were genotyped successfully. The genotypes based on the partial RdRp (249 bp) and VP1 (285 bp) sequences are shown in Fig. 2. The dominant genotypes were GII.P16 for RdRp A wider variety of recombinant variants were detected in the age group of more than 5 years and in Spring (Fig. 3).
## The variant of GII.17 norovirus
Compared to the GII.4 genotype, GII.17 has a lower detection rate worldwide. In 2014/2015, norovirus GII.17[P17] variant emerged and caused a series of AGE outbreaks in China and Japan [21]. In this study, we found that GII.17[P17] became the dominant variant (41.43%) of norovirus circulating in pediatric AGE patients in 2024. It has been reported that GII.17 can be divided into cluster I (1976-2002), cluster II (2005-2007) and cluster III (2013III ( -2015)). Cluster III can be further divided into sub-cluster IIIa (which first found in Japan and Korea) and cluster IIIb (which first found in countries other than Japan and Korea) [22]. According to the phylogenetic analysis, the GII.17 genotype in this study could be divided into two sub-clusters, cluster IIIa and cluster IIIb (Fig. 4). The GII.17 variants in cluster IIIa shared 97.1%-98.5% nucleotide (nt) identity with the reference strain Kawasaki323 (AB983218). By contrast, all the GII.17 variants clustered with IIIb shared 100% nt identity with the reference strain Kawasaki308 (LC037415).
## The variant of GII.4 norovirus
Since the emergence of GII.4 95/96-US variant, six more GII.4 variants such as Farmington Hills, Hunter, Den Haag, New Orleans, Sydney and San Francisco have been reported worldwide [4,23]. Phylogenetic analysis revealed that the GII.4 genotype identified in this study were GII.4 Sydney variants and divided into two clusters (Fig. 5). All of the GII.
## Discussion
In this study, we characterized the epidemiology of norovirus and analyzed the genetic diversity of GII variants in pediatric patients with sporadic AGE in Beijing, China, from 2023 to 2024. The norovirus detection rate in this study was 19.6%, which was slightly higher than the 17.7% positive rate observed in pediatric patients in Beijing during 2020 [24], yet substantially lower than the 30.7% detection rate reported among Beijing pediatric cases from 2018 to 2020 [25]. From 2012 to 2023, the detection rate of norovirus in sporadic pediatric AGE cases demonstrated significant variability, range from 10.8% to 39.1% in different studies in China [26][27][28][29][30][31][32]. Geographic location appears to be a key determinant of norovirus prevalence. During 2018 to 2020, the detection rate of norovirus was 26.4% in Tianjin AGE patients [33], while the positive rate was only 9.0% in Tibet in the same period [34]. Notably, substantial variations can occur even within the same city, as evidenced by the 14.9% detection rate in Shanghai pediatric AGE cases from 2014 to 2018 versus the remarkably high 65.88% positivity rate reported in Shanghai's Pudong New Area between 2014 and 2016 [27,35]. Beyond geographic factors, socioeconomic conditions may also influence norovirus epidemiology. For instance, norovirus detection rates tend to be higher in developing countries than in developed ones. Previous studies reported a positivity rate of 37.2% in Brazil, compared to only 8.1% in Leipzig, Germany [36,37]. The implementation of non-pharmaceutical interventions (NPIs) during the COVID-19 pandemic may have an impact on norovirus detection rates. Enhanced public health measures, improved hygiene practices, and heightened health awareness during this period have been associated with a reduction in norovirus transmission [32,37]. In the present study, the lower detection rate of norovirus compared to that reported in a previous study from the same region (2018-2020) may reflect sustained public adherence to these protective behaviors developed during the pandemic [25]. In addition, the detection rate of norovirus may be influenced by the genogroup specificity of the assay. Unlike most previous studies that targeted only GI and GII [29,35,36], the present investigation also included detection for GIV, which may have contributed to differences in the observed prevalence. Additional variables including patient age distribution, study sample size and specimen type may also influence the positive rate of norovirus. Norovirus associated AGE primarily affected children under 5 years of age, with the highest detection rate Norovirus strains obtained in this study are indicated by solid red triangles. Trees were constructed using MEGA 7.0 with the neighbor-joining method and the Kimura 2-parameter model. Bootstrap values more than 90% (1000 replicates) are shown in the phylogenetic tree observed in those under 2 years [28,29,38]. In our study, most norovirus-positive cases were infants under 3 years old, which is consistent with the previous studies.
Since 1987, GII.4 variants have been the predominant norovirus genotype circulating globally, accounting for more than six pandemics [4]. In China, GII.4 Sydney[P31] was the dominant strain responsible for norovirus outbreak between 2007 and 2013 [39]. However, GII.17 In 2022, laboratory-based surveillance data from Cali-ciNet China revealed that GII.3[P12] emerged as the primary genotype responsible for norovirus associated AGE outbreaks [43]. Our findings identified GII.3[P12] as the dominant norovirus genotype in 2023, consistent with the variant circulating in China during 2022.
In 2024 the dominant genotype changed to GII.17[P17] in this study. Compared to GII.4, the detection rate of GII.17 has been very low. However, during the 2014/2015 winter season, novel GII.17[P17] variant contributed to most of the norovirus outbreaks in Southern China, Japan, South Korea, Argentina and Europe [11-14, 44, 45]. Phylogenetic analysis demonstrated that, GII.17
## Conclusion
In conclusion, this study provided the molecular epidemiology features of norovirus in sporadic pediatric AGE patients in Beijing, China, during 2023 to 2024. We found that norovirus GII are genetically diverse and a wider variety of recombinant variants were detected in children P17] were the dominant genotypes circulated in AGE cases in 2023 and 2024, respectively in Beijing. Meanwhile, we characterized the molecular evolution of GII.17[P17] and GII.4 Sydney variants in this study. Although significant progress in sanitation status and public health awareness, norovirus associated AGE remains one of the important disease burden in China [52]. Therefore, it is necessary to actively monitor the epidemiological characteristics of norovirus AGE and provide basic data for disease prevention and control.
## References
1. Collaborators (2025) "Global, regional, and national age-sex-specific burden of diarrhoeal diseases, their risk factors, and aetiologies, 1990-2021, for 204 countries and territories: a systematic analysis for the global burden of disease study 2021" *Lancet Infect Dis*
2. Ahmed, Hall, Robinson (2014) "Global prevalence of norovirus in cases of gastroenteritis: a systematic review and meta-analysis" *Lancet Infect Dis*
3. Mans (1104) "Norovirus infections and disease in lower-middleandlow-income countries, 1997(-)2018. Viruses"
4. Winder, Gohar, Muthana (2022) "Norovirus: An Overview of Virology and Preventative Measures" *Viruses*
5. White (2014) "Evolution of norovirus" *Clin Microbiol Infect*
6. Kroneman, Vega, Vennema (2013) "Proposal for a unified norovirus nomenclature and genotyping" *Arch Virol*
7. Vinje (2015) "Advances in laboratory methods for detection and typing of norovirus" *J Clin Microbiol*
8. Chhabra, De Graaf, Parra (2019) "Updated classification of norovirus genogroups and genotypes" *J Gen Virol*
9. Parra (2019) "Emergence of norovirus strains: a tale of two genes" *Virus Evol*
10. Tohma, Lepore, Siltz (2018) "Evolutionary dynamics of non-GII genotype 4 (GII.4) noroviruses reveal limited and independent diversification of variants" *J Gen Virol*
11. Lu, Sun, Fang (2014) "Gastroenteritis outbreaks caused by norovirus GII" *Emerg Infect Dis*
12. Fu (2014) "Emergence of a new GII.17 norovirus variant in patients with acute gastroenteritis in Jiangsu" *Euro Surveill*
13. Matsushima, Ishikawa, Shimizu (2015) "Genetic analyses of GII.17 norovirus strains in diarrheal disease outbreaks from December 2014 to March 2015 in Japan reveal a novel polymerase sequence and amino acid substitutions in the capsid region" *Euro Surveill*
14. Degiuseppe, Gomes, Hadad (2015) "Detection of novel GII.17 norovirus in Argentina" *Infect Genet Evol*
15. Giammanco, Grazia, Bonura (2017) "Norovirus GII.17 as major epidemic strain in Italy, winter 2015-16" *Emerg Infect Dis*
16. Niendorf, Jacobsen, Faber (2016) "Steep rise in norovirus cases and emergence of a new recombinant" *Euro Surveill*
17. Jin, Wu, Kong (2016) "Norovirus outbreak surveillance" *Emerg Infect Dis*
18. Wei, Ge, Tan (2021) "Epidemiology and evolution of Norovirus in China" *Hum Vaccin Immunother*
19. Chhabra, Wong, Niendorf (2024) "Increased circulation of GII.17 noroviruses, six European countries and the United States, 2023 to 2024" *Euro Surveill*
20. Cannon, Barclay, Collins (2017) "Genetic and epidemiologic trends of Norovirus outbreaks in the United States from 2013 to 2016 demonstrated emergence of novel GII.4 recombinant viruses" *J Clin Microbiol*
21. De Graaf, Van Beek, Vennema (2015) "Emergence of a novel GII.17 norovirus -End of the GII.4 era?" *Euro Surveill*
22. Kim, Won, Kang (2019) "Complete sequence analysis of human norovirus GII.17 detected in South Korea" *Epidemiol Infect*
23. Chhabra, Tully, Mans (2024) "Emergence of novel Norovirus GII.4 variant" *Emerg Infect Dis*
24. Jia, Zhao, Zhou (2021) "Molecular epidemiology of norovirus associated with pediatric acute gastroenteritis in Beijing in 2020" *Zhonghua Er Ke Za Zhi*
25. Ai, Zhang (2018) "Recombinant GII.4[P31] was predominant norovirus circulating in Beijing area" *Virol Sin*
26. Fang, Zhang, Wang (2022) "Molecular epidemiology of norovirus infections in children with acute gastroenteritis in 2017-2019 in Tianjin" *China. J Med Virol*
27. Wang, Wei, Guo (2019) "Norovirus activity and genotypes in sporadic acute diarrhea in children in Shanghai during 2014-2018" *Pediatr Infect Dis J*
28. Lu, Zhong, Xu (2019) "Genetic diversity and epidemiology of genogroup II noroviruses in children with acute sporadic gastroenteritis in Shanghai" *BMC Infect Dis*
29. Li, Zhang, Zou (2023) "Epidemiology and genetic diversity of norovirus GII genogroups among children in Hubei" *Virol Sin*
30. Tian, Du, Gao (2025) "Norovirus molecular trends in Harbin preschoolers post-NPI easing" *J Infect Public Health*
31. Duan, Hu, Tang (2021) "A molecular epidemiological study of pediatric norovirus gastroenteritis" *Zhongguo Dang Dai Er Ke Za Zhi*
32. Chen, Shao, Ru (2024) "Epidemiological and genetic characteristics of norovirus in Hangzhou, China, in the postepidemic era" *J Clin Virol*
33. Fang, Dong, Liu (2018) "Molecular epidemiology and genetic diversity of norovirus among hospitalized children with acute gastroenteritis in Tianjin" *BMC Infect Dis*
34. Mao, Yang, Shi (2022) "Molecular epidemiological characteristics of the virus in 96 children with acute diarrhea in Changdu of Tibet" *Zhongguo Dang Dai Er Ke Za Zhi*
35. Xue, Pan, Zhu (2018) "Molecular epidemiology of genogroup II norovirus infections in acute gastroenteritis patients during 2014-2016 in Pudong New Area" *Gut Pathog*
36. Ennuschat, Hartel, Pietsch "Norovirus epidemiology and genetic diversity in Leipzig, Germany during 2013-2017"
37. Sarmento, De Andrade, Malta (2019) "Norovirus Epidemiology and Genotype Circulation during the COVID-19 Pandemic in Brazil" *Pathogens*
38. Zou, Cui, Wang (2015) "Clinical characteristics and molecular epidemiology of noroviruses in outpatient children with acute gastroenteritis in Huzhou of China" *PLoS ONE*
39. Yu, Jiang, Guo (2022) "Norovirus outbreaks in China, 2000-2018: a systematic review" *Rev Med Virol*
40. Qi, Dong, Cheng (2023) "Epidemiological characteristics of Norovirus outbreaks in Shenyang from 2017 to 2021" *J Microbiol*
41. Sun, Yuan, Mao (2021) "Molecular Epidemiology of Human Norovirus Variants from Outbreaks in Zhejiang Province" *Adv Virol*
42. Cao, Ma, Li (2015) "Epidemiology of norovirus gastroenteritis in hospitalized children under five years old in western China" *J Microbiol Immunol Infect*
43. Yanhui, Qing, Lijiao et al. (2022) "Epidemiological characteristics of outbreaks of norovirus GII.17[P17] acute gastroenteritis in China" *Chinese J Exp Clin Virol*
44. Dang Thanh, Than, Nguyen (2016) "Emergence of Norovirus GII.17 Variants among Children with Acute Gastroenteritis in South Korea" *PLoS One*
45. Dinu, Nagy, Negru (2015) "Molecular identification of emergent GII.P17-GII.17 norovirus genotype" *Euro Surveill*
46. Van Beek, Ambert-Balay, Botteldoorn (2012) "Indications for worldwide increased norovirus activity associated with emergence of a new variant of genotype II.4, late" *Euro Surveill*
47. Matsushima, Shimizu, Ishikawa (2016) "Complete genome sequence of a recombinant GII.P16-GII.4 norovirus detected in Kawasaki City" *Genome Announc*
48. Barreira, Fumian, Tonini (2017) "Detection and molecular characterization of the novel recombinant norovirus GII.P16-GII.4 Sydney in southeastern Brazil in 2016" *PLoS ONE*
49. Bidalot, Thery, Kaplon (2016) "Emergence of new recombinant noroviruses GII" *Euro Surveill*
50. Medici, Tummolo, Martella (2016) "Emergence of novel recombinant GII.P16_GII.2 and GII. P16_GII.4 Sydney 2012 norovirus strains in Italy" *New Microbiol*
51. Zheng, Zhu, Cui (2022) "Evolutionary analyses of emerging GII.2[P16] and GII" *Virus Evol*
52. Walker, Rudan, Liu (2013) "Global burden of childhood pneumonia and diarrhoea" *Lancet* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12746535&blobtype=pdf | # Unraveling Cytomegalovirus Drug Resistance in Transplant Patients by Targeting Deep Sequencing
Salvador Alemán, | Camacho, Vanessa Recio, Estrella Ruiz, | Zamarrón, Jorge Anel, | Montserrat Enjuto, David Tarragó
## Abstract
Drug-resistant cytomegalovirus (CMV) poses a major clinical challenge in transplant recipients, leading to treatment failure and increased morbidity. This study applied a next-generation sequencing (NGS) approach to identify antiviral resistance mutations (ARMs) in 71 samples from 68 CMV-positive patients who had undergone hematopoietic stem cell transplantation (HSCT) or solid organ transplantation (SOT) between 2018 and 2024. A custom nested-PCR protocol targeting six CMV genes (UL27, UL51, UL54, UL56, UL89, and UL97) was developed for enrichment prior to NGS. ARMs were detected in 23% of patients without clinical suspicion of resistance and in 62% of those with suspected resistance, most frequently affecting UL97. The most common UL97 mutations were A594V (24.4%), C603W (20.0%), and L595S (15.6%), while D301N (50%) predominated in UL54. Mutations associated with foscarnet and maribavir resistance were found in five and eight patients, respectively. NGS identified ARMs in 29 patients not detected by Sanger sequencing (p < 0.00001), while no additional ARMs were identified by Sanger alone. Importantly, these minority variants, revealed by NGS, are clinically relevant, as they may expand under antiviral pressure and contribute to virological failure. ARM presence was not significantly associated with viral load or mortality, though recurrent CMV reactivation showed a trend toward association (p = 0.0504). Survival was significantly lower in HSCT versus SOT recipients (p = 0.027). These findings support the routine clinical use of NGS for CMV resistance testing, particularly in complex cases and in the context of expanding antiviral options such as maribavir and letermovir.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
Transplant recipients face significant challenges in managing cytomegalovirus (CMV) infection due to immunosuppression and the emergence of antiviral resistance caused by prolonged antiviral exposure. Genotypic resistance assays are critical for guiding therapy in hematopoietic stem cell transplant (HSCT) and solid organ transplant (SOT) recipients, who are highly susceptible to severe CMV-related complications. Resistance to antiviral drugs should be suspected when CMV viral load increases by more than 1 log 10 IU/mL or fails to decline after at least 2 weeks of antiviral therapy. CMV resistance is identified by detecting specific mutations in genes encoding antiviral targets; the development of resistance is often closely related to the antiviral agent administered in each clinical case [1,2]. Antiviral resistance mutation (ARM) in the UL97 gene are a major mechanism of CMV resistance to ganciclovir (GCV), valganciclovir (VGCV), and ARM conferring varying levels of resistance to MBV have also been described [3]. The majority of reported ARM in clinical series involve the UL97 kinase, with substitutions commonly occurring at positions M460V/I, H520Q, C592G, A594V, L595S, and C603W [4]. Less frequent ARM, clustered between codons 590 and 607, have also been implicated in GCV resistance [5].
Additionally, ARM in the CMV DNA polymerase gene UL54 are associated with resistance to polymerase inhibitors, including GCV, VGCV, and nondependent viral kinases as foscarnet (FOS), and cidofovir (CDV). ARM in UL54 typically clustered in functional domains of the DNA polymerase and exhibited distinct resistance phenotypes. New ARM continues to be identified, some of which confer cross-resistance to multiple antivirals, including FOS [5]. These ARM often emerge following prolonged antiviral exposure. The coexistence of ARM in both UL97 and UL54 can result in higher levels of GCV resistance [5].
Recently, two new CMV antivirals have been approved by the FDA. MBV is a UL97 kinase inhibitor, which is indicated for the treatment of posttransplant CMV infection or disease that is refractory (with or without genotypic resistance) to GCV, VGCV, CDV, or FOS. In addition to UL97, resistance to MBV has also been associated with mutations in the UL27 gene, which encodes a viral nuclear protein [5]. Letermovir (LET), a CMV DNA terminase complex inhibitor, is approved for prophylaxis of CMV infection in CMV-seropositive adult HSCT recipients and highrisk kidney transplant recipients. The viral terminase complex, comprising the UL56, UL89, and UL51 genes, plays a crucial role in the cleavage and packaging of viral genomes following DNA replication via rolling-circle amplification [6]. LET resistance is primarily associated with mutations in UL56, although mutations in UL89 and UL51 have also been reported. Notably, UL51 mutations may enhance resistance conferred by UL56 mutations due to their low fitness cost [6]. In vitro studies have identified LET resistance mutations between codons 231 and 369 of UL56, indicating a low genetic barrier to resistance [1,2,7].
In clinical practice, laboratory testing for antiviral-resistant CMV is essential, as many cases of persistent viremia are not related to antiviral resistance [1]. Due to the declining use of viral culture in routine diagnostics, genotypic testing has become the primary method for detecting antiviral resistance. For routine testing, Sanger sequencing of UL97 and UL54 (for GCV/VGCV/FOS resistance) and UL56 (for LET resistance) remains the most utilized [8]. However, in research and clinical trial settings, extended sequencing may include UL27 for MBV resistance and UL51, UL56, and UL89 for LET resistance.
In recent years, next-generation sequencing (NGS) has emerged as a promising technique for the detection of CMV ARMs [9][10][11], with several studies evaluating its clinical utility and costeffectiveness compared to Sanger sequencing. NGS offers several advantages: (1) higher sensitivity, capable of detecting resistant variants even at viral loads as low as 500 IU/mL [9]; (2) detection of minor variants representing 5%-15% of the viral population, compared to ~20% detection limits for Sanger sequencing [8,9]; and (3) potential for greater efficiency and cost-effectiveness through multiplexing of genes and samples [8]. Nevertheless, most studies to date have focused only on UL97 and UL54, while resistance mutations in UL89 and UL51-although currently less frequent-may become increasingly relevant with the growing use of LET [7][8][9]12]. Despite its advantages, the routine implementation of NGS in clinical laboratories remains limited due to technical complexity, longer turnaround times, the need for bioinformatics expertise, and challenges in data interpretation. Furthermore, only a few studies have applied NGS to CMV resistance testing in larger patient cohorts, and most have been restricted to analysis of UL97 and UL54 [13].
The primary aim of this study was to develop and apply a methodology for detecting ARMs in six target genes (UL27, UL51, UL54, UL56, UL89, and UL97) by NGS in transplant recipients. This approach seeks to expand the current knowledge of CMV resistance and highlight the relevance of NGSbased ARM testing in clinical practice. Additionally, secondary objectives included exploring potential associations between ARM detection and clinical variables such as viral load, transplant type, treatment history, and patient outcomes.
## 2 | Materials and Methods
## 2.1 | Patients and Clinical Samples
In this retrospective study, 71 clinical samples from 68 CMV DNA-positive transplant patients were analyzed by NGS. All samples had previously been submitted to the National Centre for Microbiology (CNM) for genotypic testing by Sanger sequencing, originating from hospitals across Spain. The samples were collected between January 2018 and April 2024. In some patients, multiple sequential clinical samples were available for analysis. We explicitly clarify that this study comprised (i) a single-center cohort from Hospital Puerta de Hierro (n = 35) with detailed clinical data and (ii) a multicenter cohort from 33 patients across multiple Spanish hospitals, without associated clinical/demographic data.
The patients were divided into three groups. The first group included patients from Hospital Puerta de Hierro without clinical suspicion of CMV resistance (treatment responders; n = 13). The second group consisted of patients from Hospital Puerta de Hierro with clinical suspicion of antiviral resistance (n = 22). For all patients from Hospital Puerta de Hierro (n = 35), clinical records were available, allowing for the collection of risk factors, clinical conditions, and outcome data. The third group included patients with clinical suspicion of resistance whose samples had been submitted to the CNM for resistance genotyping from hospitals throughout Spain (n = 33); for this group, clinical or demographic data were not available due to data protection regulations under the CNM portfolio. All samples consisted of residual plasma or blood, which had been stored at -80°C prior to processing.
The median age of patients from Hospital Puerta de Hierro was 57 years. Among these, 16 patients had undergone solid organ transplantation (SOT)-including kidney (n = 7), lung (n = 5), heart (n = 2), and liver (n = 2) transplantation-and 19 patients had received HSCT. Definitions for resistant and refractory CMV infection were applied according to standardized criteria [1]. Individual treatment regimens, viral loads, demographic data, and clinical variables for Groups 1 and 2 are summarized in Table S6.
For positive controls, CMV DNA extracted from in vitro cultured virus previously used for diagnostic purposes was included in all experiments. Negative controls consisted of plasma and blood samples that had previously tested negative for CMV by real-time PCR.
This study was approved by the Ethics Committee of the Instituto de Salud Carlos III (CEI PI 11_2021-v3).
## 2.2 | DNA Extraction From Clinical Samples
Extraction was performed with an automatic extractor (QIAsymphony, Qiagen) with a commercial kit (QIAsymphony Virus/Bacteria Midi Kit (96), Qiagen) as per the manufacturer's instructions.
## 2.3 | Enrichment of CMV DNA Through Amplification With Specific Nested-PCR
Nested-PCRs were designed to amplify each whole target gene based on consensus CMV sequences from GenBank. Six pairs of external primers were used in the first round of the PCR and other six internal pairs were used in the second round. Reactions were performed in Biorad C1000 Touch Thermal Cycler using Platinum SuperFi II DNA Polymerase (Thermo Fisher Scientific, Invitrogen), according to the manufacturer′s instructions with the addition of Q5 High GC Enhancer (New England Biolabs) to all reactions, in a final volume of 50 µL. PCR products were stored at -20°C until their use in the preparation of DNA library for NGS. Primer sets used for the amplification of UL27, UL51, UL54, UL56, UL89, and UL97 genes, amplicon sizes and nested-PCR conditions, are shown in Tables S1-S3.
## 2.4 | DNA Quantification and Quality Control Throughout the Workflow
CMV viral load was quantified in all clinical samples using a previously validated real-time PCR assay routinely used at the Reference Laboratory for Immune Preventable Diseases (National Centre for Microbiology) using WHO standards [4].
Throughout the study, DNA was quantified at multiple stages, including postextraction, postnested PCR amplification, and after library preparation. Quantification was carried out using the QuantiFluor dsDNA System with a Quantus Fluorometer (Promega), following manufacturer instructions and expressed in ng/μL. Amplicon integrity was verified via agarose gel electrophoresis (1% in TAE buffer), using the FastRuler Middle Range DNA ladder (Thermo Fisher Scientific). Electrophoresis was performed at 90 V for 45 min.
Final library quality and fragment size were assessed at the Genomics Department (National Centre for Microbiology) using the Agilent 4150 TapeStation system. DNA ScreenTape or High Sensitivity ScreenTape kits (Agilent) were selected based on library concentration to determine size distribution, quality, and DNA concentration.
## 2.5 | Library Pooling and DNA Library Preparation
Following the second round of nested PCR, six amplicons corresponding to the target genes were generated per clinical sample. To prepare the DNA libraries for NGS, the resulting PCR products were pooled into a single tube per clinical sample. Each pool was adjusted to a final volume of 26 µL containing 100 ng of total DNA, following the recommendations of the NEBNext Multiplex Oligos for Illumina protocol (New England Biolabs). A total of 73 libraries were generated, corresponding to the 71 clinical samples and two additional controls (positive and negative). Library construction was then carried out using the NEBNext Multiplex Oligos for Illumina (Dual Index Primers Set 1 and 2), which allows multiplexing of up to 96 samples. Library preparation was performed according to the manufacturer's instructions. Final quantification of the DNA libraries was determined prior to sequencing.
## 2.6 | Sanger Dideoxy Sequencing
Briefly, Sanger dideoxy sequencing was performed as detailed in Tarragó et al. [14] and Gómez et al. [15]. Afterwards, all sequences were downloaded, aligned, and translated by MEGA 7 software using the UL54 and UL97 gene sequences from Human Herpesvirus 5 (Merlin strain) as reference sequence (NCBI Reference Sequence: NC_006273.2).
## 2.7 | NGS and Bioinformatic Analysis of Raw Data
Sequencing is carried out by the Genomics Unit of the ISCIII using NextSeq 500 Illumina platform with NextSeq 500 cartridges, which have a capacity up to 130 million reads, performing 2 reads per 150 base pair fragments. The limit of detection to avoid false positive results will be defined at 5% according to a previous study [16].
The raw FASTQ sequences obtained were sent to the Bioinformatics Unit of the ISCIII for processing and analysis by the nf-core/viralrecon bioinformatics pipeline. NGS data were processed using the viralrecon pipeline (nf-core/BU-ISCIII, implemented in Nextflow DSL2 with Docker/Singularity containers) to ensure full reproducibility and traceability of computational steps.
The workflow included the following stages:
(1) Quality control of raw reads with FastQC and trimming of adapters and low-quality bases with fastp; (2) Removal of host reads (Illumina data only) using Kraken2; (3) Read alignment against the CMV reference genome (GenBank accession NC_006273.2) using Bowtie2, followed by sorting and indexing with SAMtools; (4) For amplicon data, primer trimming with iVar; duplicate marking with Picard; and calculation of coverage metrics with mosdepth; (5) Variant calling with iVar variants (amplicon data) or BCFTools (metagenomic data), using a minimum variant frequency threshold of 5%, minimum base quality of 20, and minimum coverage of 10×. Variants below these thresholds were excluded. ( 6) Consensus sequence generation with iVar consensus, requiring a minimum allele frequency of 50% to call a base, otherwise reporting an "N"; (7) Functional annotation of variants with SnpEff and SnpSift to determine amino acid changes and potential impact on antiviral resistance genes; (8) Genome assembly quality assessment with QUAST, and coverage plots across the genome for visualization of sequencing depth; (9) Summary reports integrating all QC and analysis steps were produced with MultiQC. All steps were executed using containerized environments with fixed tool versions to guarantee reproducibility. This pipeline is publicly available at https://github.com/BU-ISCIII/viralrecon.
## 2.8 | Characterization of ARM
Characterization of ARM versus mutation not associated to antiviral resistance was performed according to previously published data [4].
## 2.9 | Statistical Analysis
All statistical analyses were performed using SPSS software version 25 (IBM Corporation Inc.). Associations among clinical conditions, risk factors, and the presence of ARM were tested using the χ 2 test or Student's t-test, depending on the variable type. Fisher's exact test was used to estimate the effect of categorical variables and their interactions when sample sizes were small. Differences in viral load between groups with and without mutations were assessed using the nonparametric Mann-Whitney U test due to the non-normal distribution of the data. Concordance between NGS and Sanger sequencing was evaluated by McNemar's test for paired categorical data. Unless otherwise indicated, statistical significance was established at p < 0.05 (95% confidence level).
## 3 | Results
## 3.1 | Sequencing Clinical Samples and ARM Detected by NGS
All target genes were successfully sequenced in full by NGS in 34 of the 71 clinical samples. Details for each gene are shown in Table 1.
The most frequent ARM identified by NGS were A594V (n = 11), C603W (n = 9), L595S (n = 7), T409M (n = 7), and D301N (n = 3). These substitutions are located in the UL97 protein kinase, a key target of GCV, and have been consistently associated with reduced susceptibility or resistance to this drug. In particular, A594V and L595S are among the most prevalent resistance mutations described worldwide and are strongly linked to clinical failure of GCV therapy in transplant recipients. C603W also confers significant GCV resistance, although it is less frequent in clinical cohorts. T409M has been reported to reduce GCV susceptibility, often appearing in combination with other UL97 mutations and contributing to cumulative resistance. D301N has been described as a resistance-associated mutation with moderate impact, potentially contributing to reduced GCV efficacy when present alongside other substitutions. The detection of these well-characterized ARM underscores the clinical significance of NGS in identifying both dominant and minority resistant variants that may compromise antiviral treatment in immunosuppressed patients. ARM cohort are summarized in Table 2. ARM detected by NGS and involved drugs are detailed in Figure 1A for UL97, Figure 1B for UL54, and Figure 1C for global drug totals.
## 3.2 | ARM in Transplant Recipients Without Clinical Suspicion of Resistance
We analyzed 16 samples from 13 patients without a priori suspicion of resistance (Hospital Puerta de Hierro). At the patient level, 3/13 (23.1%) carried UL97 ARM by NGS while no UL54 ARM were detected (Table 3). Importantly, in this cohort, Sanger reported S/S in all cases, including the three UL97-ARM patients, underscoring the added yield of NGS for unsuspected resistance. Full sample-level results are provided in Table S4. Several samples showed indeterminate calls in UL97 due to low/ambiguous coverage as shown in Table S5.
## 3.3 | ARM in Transplant Recipients With Clinical Suspicion of Resistance
We sequenced 55 clinical samples (55 patients) with suspected resistance. Any genotypic resistance by NGS (UL97 and/or UL54) was found in 34/55 (61.8%) samples. UL97 ARM were detected in 33/55 (60.0%), and UL54 ARM in 5/55 (9.1%). UL54 ARM cooccurred with UL97 ARM in all but one sample (the exception carried UL54 D301N with UL97 susceptible). The most frequent UL97 ARM were A594V (n = 9), C603W (n = 8), and L595S (n = 7). For UL54, D301N was observed in three samples. ARM counts by gene in the suspicion cohort are shown in Table 4. Full sample-level results are provided in Table S5.
## 3.4 | ARM Detected by NGS Versus Sanger Sequencing
All ARM detected by Sanger sequencing were also identified by NGS. In contrast, NGS detected a total of 51 ARM in 37 patients (54.4%), whereas Sanger sequencing detected only 15 ARM in 8 patients (11.8%). In 31 patients, no ARM were detected by either However, the analysis of raw Sanger sequencing data showed that minor subpopulations of ARM were not detected as for the ARM M460I, which is exemplified in the following Sample 72. Sanger chromatogram analysis ruled out minor populations considered as background or noise as shown in Figure 2.
## 3.5 | Clinical Conditions of Patients and Virological Findings
Among (n = 2), and liver (n = 2) (Figure 3A). Based on clinical criteria, 22/35 (63%) were suspected of antiviral resistance; ARM were detected in 16 patients (UL97, n = 16; UL54, n = 2), including two carrying mutations in both genes. Of the 13 patients without initial suspicion, 3 harbored ARM (all in UL97).
Overall survival was 43%, with substantially lower survival in HSCT than SOT recipients (32% vs. 56%; OR = 0.37, p = 0.027; Fisher's exact; Figure 3B). Survival by age group was 33% (40-49 years), 66% (50-59 years), and 60% (60-69 years) (Figure 3C).
Antiviral exposure patterns differed by resistance status. GCV use was significantly less frequent among ARM-positive than ARM-negative patients (21% [3/14] vs. 60% [12/20]; OR = 0.18, p = 0.038), consistent with early discontinuation/switching in nonresponders (Figure 3D). VGCV showed a nonsignificant trend toward less frequent use in ARM-positive patients (36% vs. 65%; p = 0.163). No significant associations were observed for other agents (FOS, CDV, LET; all p > 0.5). Maribavir (MBV) was only used in ARM-positive patients (2/14).
Viral load analyses included patients with available NGS and viral load data, excluding indeterminate results. ARM-positive vs. WT had similar distributions (median 4.92 log 10 IU/mL vs. 5.49 log 10 IU/mL; Mann-Whitney U = 156.5, p = 0.745), with medians and test statistics displayed in the summary box of Figure 3E. By transplant type, HSCT versus SOT viral loads were also comparable (median 4.28 log 10 IU/mL vs. 4.38 log 10 IU/mL; Mann-Whitney U = 166.0, p = 0.632; see summary box in Figure 3F).
Recurrent CMV infection was more frequent in ARM-positive than ARM-negative patients (93% [13/14] vs. 60% [12/20]; OR = 8.67, p = 0.0504), suggesting a relationship between resistance development and recurrence (Figure 3G). Full patient-level demographics, treatments, and virological results are shown in Table S6.
## 4 | Discussion
The emergence of ARM in CMV poses a significant threat to transplant recipients, particularly those undergoing HSCT or SOT, where immunosuppression increases susceptibility to viral reactivation and complicates treatment [1,2]. Traditionally, ARM detection has relied on Sanger sequencing of the UL54 and UL97 genes. However, this approach must be reconsidered with the expanded use of MBV and LET, which target additional viral genes beyond UL54 and UL97.
Moreover, Sanger sequencing is limited in sensitivity, particularly for detecting minority resistant subpopulations, prompting the integration of NGS into clinical practice. NGS provides broader genomic coverage and enhanced detection capacity [17]. In this study, the ability of NGS to detect minority ARM populations was exemplified by the identification of the M460I mutation in one sample. This mutation was undetectable by Sanger sequencing due to its frequency falling below the detection threshold (< 30%) and being masked by the dominant wild-type variant. This finding underscores the limitations of conventional genotyping methods and highlights the clinical utility of deep sequencing for early resistance detection and appropriate antiviral selection, particularly in complex cases where minor variants may expand under drug pressure [9].
Thanks to the capabilities of NGS, the targeted PCR approach developed in this study allowed for the sequencing of entire genes, in contrast to the fragment-based sequencing typical of Sanger methods. This enabled the creation of a comprehensive database to support improved surveillance of ARM and polymorphisms across complete genes. Our findings demonstrated high NGS coverage across all six targeted CMV genes, ranging from 77.5% for UL27 to 97.2% for UL56. This high coverage affirms the feasibility of applying NGS routinely, consistent with previous studies in which deep sequencing enabled robust genotypic surveillance even at relatively low viral loads [5].
For UL54 and UL97 specifically, NGS enabled the sequencing of nine and six additional patients, respectively, compared to Sanger sequencing. Overall, NGS detected ARM in 29 patients that were missed by Sanger sequencing. Conversely, Sanger did not identify any additional ARM undetected by NGS. In total, NGS detected ARMs in 54.4% of patients, compared to 11.8% by Sanger sequencing. These results align with previous studies demonstrating the superior sensitivity of NGS, particularly for mutations present at frequencies below 30% [8,9], which are often missed by Sanger's known limitations in detecting minority variants [11].
The mutational landscape observed was consistent with prior literature. In the UL97 gene, A594V, C603W, and L595S were the most frequently identified mutations, all associated with resistance to GCV and VGCV, as previously reported [4,13].
Less common mutations such as T409M, M460I, and C592G were also detected, though at lower frequencies compared to earlier studies [9,18]. A notable concentration of mutations was found between codons 590-607 in UL97, a well-established hotspot for resistance that preserves viral fitness [19]. In UL54, D301N was the most commonly detected ARM, consistent with prior reports from transplant cohorts treated with DNA polymerase inhibitors [10].
A particularly significant finding was the detection of ARM in 23% of transplant recipients without clinical suspicion of resistance, identified exclusively by NGS. Notably, C603W and A594V were found in three patients who were responding to therapy at the time of sampling. These findings suggest that resistance mutations can be present even in the absence of overt treatment failure, reinforcing the need for broader implementation of resistance surveillance strategies [11].
Among patients with clinical suspicion of resistance, virological resistance was confirmed by NGS in 34 patients (61.8%), predominantly involving UL97. ARMs were significantly more prevalent in UL97 than in UL54. When UL54 mutations were present, they frequently co-occurred with UL97 mutations, consistent with previous studies [9,20]. Since FOS is typically used as second-line therapy following GCV failure, the detection of FOS resistance in at least five patients significantly limited available treatment options in those cases. Additionally, ARMs associated with MBV resistance, such as T409M, M411Y, and C480F, were identified in eight patients, reflecting the growing clinical use of this newer antiviral agent.
From a clinical standpoint, survival outcomes were affected by transplant type. As reported in prior studies, HSCT recipients exhibited lower survival rates compared to SOT recipients, highlighting the vulnerability of this population to severe CMV disease [21]. However, ARM presence was not statistically associated with mortality, viral load, or patient age in our cohort, consistent with observations that ARM detection alone does not independently predict clinical outcomes without considering additional host and viral factors [18].
Interestingly, GCV use at the time of sampling was significantly less frequent among patients with ARMs (OR = 0.18, p = 0.038), likely reflecting early discontinuation due to clinical nonresponse and switching to alternative antivirals before resistance testing. A similar trend was observed for VGCV (OR = 0.30, p = 0.163), though it did not reach statistical significance [22]. MBV was exclusively used in ARM-positive patients, consistent with its indication as a salvage therapy. Although not statistically significant (OR≈6.7, p = 0.162), this finding aligns with its targeted use in resistant infections.
Recurrent CMV infection was more frequent among ARMpositive patients (93% vs. 60%), with a borderline significant association (OR = 8.67, p = 0.0504), supporting the hypothesis that prolonged viral replication increases the risk of resistance development [22].
Finally, NGS's principal practical advantage over Sanger is highthroughput multiplexing: in our protocol, six CMV genes were amplified and pooled per sample-logistically impractical and cost-prohibitive with conventional methods. In urgent single-case scenarios, Sanger (UL97/UL54) can return results in ~1-2 working days, whereas targeted NGS typically requires ~2-4 working days including library preparation, sequencing, and reporting; with batching and streamlined bioinformatics, this gap narrows. When multiplexed across genes and samples, per-sample reagent costs for NGS become comparable to Sanger, which grows less efficient as the number of loci or repeat tests increases. Key barriers to routine NGS adoption include the need for batching to sustain cost-effectiveness, preanalytical constraints (adequate viral load and amplicon performance), specialized bioinformatics/reporting, and requirements for assay validation, external quality assessment, and reimbursement. Despite these hurdles, NGS offers broader gene coverage (beyond UL97/ UL54, e.g., UL27/UL51/UL56/UL89) and detection of minority variants relevant for refractory disease and newer antivirals. We therefore support a hybrid workflow: prioritize targeted NGS for high-risk contexts (e.g., HSCT, recurrent/refractory viremia, prolonged exposure to GCV/LET/MBV, rising/persistent viral load) to capture multidrug resistance and low-frequency variants, and use Sanger for rapid single-gene confirmation of actionable highfrequency variants, orthogonal verification near decision thresholds, or when NGS batching is not feasible in urgent turnarounds.
## 5 | Limitations
As a retrospective analysis based on residual clinical samples, certain constraints arose, particularly regarding sample volume and quality. In some cases, limited material prevented repeat sequencing or confirmation of minor variants. Although many target genes were successfully sequenced by NGS, variable genome coverage and occasional low viral loads in certain samples may have influenced detection sensitivity. Additionally, repeated cycles of freezing and thawing and the elapsed time between sample collection and analysis could affect DNA quality. An important limitation is the relatively small number of patients treated with newer antivirals such as MBV and LET, which restricted the ability to draw definitive conclusions regarding resistance to these drugs. Finally, despite the comprehensive clinical data collection from patients at Hospital Puerta de Hierro, the retrospective design inherently limits the ability to establish causality between the presence of ARM and clinical data. Moreover, patients with clinical suspicion of resistance whose samples had been submitted to the CNM for resistance genotyping from hospitals throughout Spain (excluding Hospital Puerta de Hierro), clinical or demographic data were not available due to data protection regulations under the CNM portfolio.
## References
1. Chemaly, Chou, Einsele (2019) "Definitions of Resistant and Refractory Cytomegalovirus Infection and Disease in Transplant Recipients for Use in Clinical Trials" *Clinical Infectious Diseases*
2. Razonable (2018) "Drug-Resistant Cytomegalovirus: Clinical Implications of Specific Mutations" *Current Opinion in Organ Transplantation*
3. Chou (2008) "Cytomegalovirus UL97 Mutations in the Era of Ganciclovir and Maribavir" *Reviews in Medical Virology*
4. Recio, González, Tarragó (2023) "Cytomegalovirus Drug Resistance Mutations in Transplant Recipients With Suspected Resistance" *Virology Journal*
5. Chou (2020) "Advances in the Genotypic Diagnosis of Cytomegalovirus Antiviral Drug Resistance" *Antiviral Research*
6. Chou (2017) "Comparison of Cytomegalovirus Terminase Gene Mutations Selected After Exposure to Three Distinct Inhibitor Compounds" *Antimicrobial Agents and Chemotherapy*
7. Muller, Tilloy, Frobert (2022) "First Clinical Description of Letermovir Resistance Mutation in Cytomegalovirus UL51 Gene and Potential Impact on the Terminase Complex Structure" *Antiviral Research*
8. Mostafa (2023) "Next-Generation Sequencing for Cytomegalovirus Genotypic Antiviral Resistance Testing" *Journal of Clinical Microbiology*
9. Streck, Espy, Ferber (2023) "Use of Next-Generation Sequencing to Detect Mutations Associated With Antiviral Drug Resistance in Cytomegalovirus" *Journal of Clinical Microbiology*
10. Mallory, Hymas, Simmon (2023) "Development and Validation of a Next-Generation Sequencing Assay With Open-Access Analysis Software for Detecting Resistance-Associated Mutations in CMV" *Journal of Clinical Microbiology*
11. Bredow, Caldera, Cerón (2023) "Clinical Next-Generation Sequencing Assay Combining Full-Length Gene Amplification and Shotgun Sequencing for the Detection of CMV Drug Resistance Mutations" *Journal of Clinical Virology*
12. Kleiboeker (2023) "Prevalence of Cytomegalovirus Antiviral Drug Resistance in Transplant Recipients" *Antiviral Research*
13. López-Aladid, Guiu, Mosquera (2019) "Improvement in Detecting Cytomegalovirus Drug Resistance Mutations in Solid Organ Transplant Recipients With Suspected Resistance Using Next Generation Sequencing" *PLoS One*
14. Tarragó, González, González-Escribano (2022) "HLA-E Restricted Cytomegalovirus UL40 Peptide Polymorphism May Represent a Risk Factor Following Congenital Infection" *BMC Genomics*
15. Gómez, Pérez-Vázquez, Tarragó (2022) "Molecular Epidemiology of Kaposi Sarcoma Virus in Spain" *PLoS One*
16. Sahoo, Lefterova, Yamamoto (2013) "Detection of Cytomegalovirus Drug Resistance Mutations by Next-Generation Sequencing" *Journal of Clinical Microbiology*
17. Chorlton, Ritchie, Lawson (2021) "Next-Generation Sequencing for Cytomegalovirus Antiviral Resistance Genotyping in a Clinical Virology Laboratory" *Antiviral Research*
18. Khawaja, Spallone, Kotton et al. (2023) "Cytomegalovirus Infection in Transplant Recipients: Newly Approved Additions to Our Armamentarium" *Clinical Microbiology and Infection*
19. Booker (2020) "Inferring Parameters of the Distribution of Fitness Effects of New Mutations When Beneficial Mutations Are Strongly Advantageous and Rare" *G3 Genes|Genomes|Genetics*
20. Chou, Kleiboeker (2022) "Relative Frequency of Cytomegalovirus UL56 Gene Mutations Detected in Genotypic Letermovir Resistance Testing" *Antiviral Research*
21. Douglas, Barnard, Holder (2020) "Letermovir Resistance Analysis in a Clinical Trial of Cytomegalovirus Prophylaxis for Hematopoietic Stem Cell Transplant Recipients" *Journal of Infectious Diseases*
22. Piret, Boivin (2019) "Clinical Development of Letermovir and Maribavir: Overview of Human Cytomegalovirus Drug Resistance" *Antiviral Research* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12583693&blobtype=pdf | # Viro3D: a comprehensive database of virus protein structure predictions
Ulad Litvin, Spyros Lytras, Alexander Jack, David Robertson, Joseph Hughes, Joe Grove
## Abstract
Viruses are genetic parasites of cellular life. Tolerance to genetic change, high mutation rates, adaptations to hosts, and immune escape have driven extensive sequence divergence of viral genes, hampering phylogenetic inference and functional annotation. Protein structure, however, is more conserved, allowing searches for distant homologs and revealing otherwise obscured evolutionary histories. Viruses are underrepresented in current protein structure databases, but this can be addressed by recent advances in machine learning. Using AlphaFold2-ColabFold and ESMFold, we predicted structures for >85,000 proteins from >4400 viruses, expanding viral coverage 30 times compared to experimental structures. Using this data, we map form and function across the human and animal virosphere and examine the evolutionary history of viral class-I fusion glycoproteins, revealing the potential origins of coronavirus spike glycoprotein. Our database, Viro3D (https:// viro3d.cvr.gla.ac.uk/), will allow the virology community to fully benefit from the structure prediction revolution, facilitating fundamental molecular virology and structure-informed design of therapies and vaccines.
## Introduction
Viruses are obligate intracellular parasites whose replication depends on the metabolism and translational machinery of cellular organisms. Viruses have the capacity to evolve rapidly and infect organisms from all domains of life. They play crucial roles in ocean biogeochemical cycles (Breitbart et al, 2018;Suttle, 2007) and control of prokaryotic populations in the human gut microbiome (Shkoporov and Hill, 2019) but also infect and cause disease in crops, livestock, and humans. Virus particles are the most abundant biological entities on our planet (Güemes et al, 2016), with metagenomic and metatranscriptomic studies only starting to sample the staggering genetic diversity of viral communities (Gregory et al, 2019;Hou et al, 2024;Zhang et al, 2019).
Genetic parasites, like viruses and mobile genetic elements, seem to be an inherent property of any replicating system (Iranzo et al, 2016b). Viruses likely emerged on multiple independent occasions, with the origin of the most ancient lineages probably coinciding with the origin of life and preceding the appearance of the last common ancestor of cellular organisms (LUCA) (Krupovic et al, 2019). Indeed, the genome of LUCA likely already carried an early antiviral defense system (Moody et al, 2024), implying that cellular organisms are inseparable from viruses and have been locked in a continuous arms race for the last 4 billion years. Independent evolutionary origins of viruses are reflected in the modern virus taxonomy by a specific taxonomic rank, the realms, which brings together viruses that share a set of conserved genes usually involved in genome replication or virion morphogenesis (Gorbalenya et al, 2020;Koonin et al, 2020).
The diversity of virus genome architectures with their frequent modular organization (Iranzo et al, 2016a), high mutation rates (Peck and Lauring, 2018), positive selection (Daugherty and Malik, 2012), and dependency on host cells, drive viruses to evolve faster than cellular organisms. Viruses exchange genetic material with their hosts and other viruses, turning their genomes into a mosaic of protein-coding and non-coding elements, with interweaved evolutionary histories (Mavrich and Hatfull, 2017). Countless examples of gene exchange between cellular organisms and viruses (Irwin et al, 2022) and co-opting of proteins that evolve to fulfill a new function (called exaptations) (Johnson, 2019;Koonin et al, 2022) highlight the evolutionary importance of genetic exchange between viruses and their hosts.
However, frequent genome reorganizations and high levels of divergence make the identification of gene function, investigation of deep phylogenetic relationships, and taxonomic assignments at higher ranks particularly challenging. In cases like these, comparison of protein structures for inference of evolutionary relatedness tends to be more reliable (Ravantti et al, 2020). Protein function is defined by tertiary structure, which is, as a result, more conserved than nucleotide or amino acid sequence (Ingles-Prieto et al, 2013;Chothia and Lesk, 1986;Illergård et al, 2009). Despite the striking diversity of viral sequences, viral protein structures are under-represented in public databases. Experimental protein structures from a viral source constitute less than 10% of the Protein Data Bank (PDB) (Berman et al, 2000).
Recent advances in machine learning have made it possible to predict protein structures from sequence alone, achieving accuracy similar to that of experimental structure determination (Akdel et al, 2022;Jumper et al, 2021;Lin et al, 2023). These state-of-the-art approaches have been applied at scale to produce comprehensive databases of predicted protein structures. The AlphaFold Structural Database (AFDB) (Varadi et al, 2024) contains more than 214 million models for proteins from UniProtKB (The UniProt Consortium, 2023) predicted using AlphaFold2 (Jumper et al, 2021). However, viral proteins were excluded from the prediction efforts. The Evolutionary Scale Modelling (ESM) Metagenomic Atlas became the largest structural database with more than 770 million models for proteins from metagenomic samples (MGnify database (Richardson et al, 2023)) predicted using ESMFold (Lin et al, 2023). Although this database contains some viral structures, they come predominantly from viruses of prokaryotes and unicellular eukaryotes.
Systematic exclusion of viral proteins from structure prediction efforts and their underrepresentation in public databases created a gap that is currently being addressed by the scientific community (Kim et al, 2025;Nomburg et al, 2024;Soh et al, 2024). In our study, we have generated 170,000 viral protein structure predictions from 4400 human and animal viruses using AlphaFold2-ColabFold (herein referred to as ColabFold) and ESMFold. We assessed the model quality produced by both methods and performed structural analysis of the proteins to expand their functional annotation. We also demonstrate that protein structure can guide the inference of deep phylogenetic relationships between viruses, using class-I membrane fusion glycoproteins as an example. To meet the needs of the virology community, we created Viro3D, a fully searchable and browsable database, allowing users to visualize and download proteome-level structural models for a virus of interest and explore similar structures present in other virus species (https:// viro3d.cvr.gla.ac.uk/). We expect that this resource will find broad utility, accelerating fundamental molecular virology, enabling studies of virus evolution, and facilitating structure-informed development of therapies and vaccines.
## Results
## Systematic viral protein structure prediction with ColabFold and ESMFold
Our initial focus has been on predicting protein structures for human and animal viruses (Fig. 1A). We relied on data from the International Committee on Taxonomy of Viruses (ICTV) Virus Metadata Resource (VMR) (Lefkowitz et al, 2018;Data ref: ICTV VMR MSL38v2), which provides a comprehensive list of virus species and representative isolates, along with their GenBank accession numbers and host associations. At the time of our analysis, the latest version of the ICTV VMR included 3173 virus species infecting vertebrate and/or invertebrate hosts, represented by 4407 virus isolates/genotypes (see Dataset EV1). These viruses encoded a total of 71,274 proteins, spanning over 29.2 million amino acid residues (Fig. 1B). To simplify the analysis and protein structure prediction procedure, we excluded large polyproteins (≥2000 residues) from the dataset, replacing them with their constituent matured cleaved proteins (annotated on GenBank as "mature peptides") and protein regions; this increased the total number of analyzed protein records to 85,162 (see Dataset EV2).
For protein structure prediction, we applied two state-of-the-art approaches: ColabFold (Mirdita et al, 2022), a method based on AlphaFold2 (Jumper et al, 2021) and dependent on multiple sequence alignments (MSAs), and ESMFold (Lin et al, 2023), a method that uses ESM-2, a transformer protein language model, and infers structure from input protein sequence alone. With ColabFold, we successfully predicted structures for all 85,162 records (27.2 million residues, covering 93.1% of total amino acid residues). Due to compute limitations in predicting longer proteins, ESMFold yielded slightly fewer predictions-84,964 protein records (27.0 million residues, covering 92.3% of amino acid residues). Structural coverage varied across viral realms, with most of them achieving coverage between 95% and 100% by both methods (Fig. EV1A). However, because of the lack of mature peptide and region annotation for many large polyproteins in the Riboviria, the coverage of this realm did not exceed 79.5% of total amino acid residues when we used ColabFold and 78.7% when we applied ESMFold. Nonetheless, since experimental structures in the PDB cover less than 3.3% of amino acid residues present in proteins of human and animal viruses (~890,000 amino acids), we have expanded the structural coverage for viral proteins by more than 30 times.
Overall, ColabFold models showed higher accuracy than those produced by ESMFold (Figs. 1C and EV1B). A total of 17.2 million residues predicted with ColabFold (63.3% of the modeled residues) were assigned high or very high quality, based on predicted localdistance difference test (pLDDT) score with the median pTM score of ColabFold predictions almost reaching 0.6 (Fig. EV1C). The number of high-quality residues increased to 87.4% (5.7 million residues) when evaluating models where a sequence homolog (defined as ≥30% identity) was available in the PDB at the time of AlphaFold2 training, dropping to 55.7% (11.5 million residues) for models without homologs (Fig. 1D). In contrast, only 31.6% of residues predicted by ESMFold (8.5 million residues) achieved high or very high quality with the mean pTM score of the models being around 0.3. Nonetheless, ESMFold models followed the same trend: with higher quality predictions for structures where a PDB sequence homolog was available at the time of training (47.6%, or 3.5 million residues) and lower quality for those without a homolog (25.5%, or 5 million residues). This suggests that training data may account for some higher accuracy models but, nonetheless, high-confidence predictions can still be achieved for sequences that were not well represented on the PDB (this may be particularly important for viral proteins, which are underrepresented in experimental structural data).
For almost 16% of the records, ESMFold models demonstrated higher pTM scores than ColabFold models (Fig. EV1D) with 9% of the models (7769 proteins) also having higher average pLDDT score (Fig. 1E). Although these improvements were typically minor, for 2% of the records (1753 proteins), the difference in pLDDT scores exceeded 10 points (Fig. 1F). One contributing factor to lower performance of ColabFold for these proteins was low MSA depth (Fig. EV1E), confirming the importance of having sufficient similar sequences available in public databases. The protein records where ESMFold outperformed ColabFold also tended to have longer sequences (Fig. EV1F).
We sought to test the accuracy of predicted models through comparison to experimental structures solved after the AlphaFold2 and ESMFold training dates, and for which there were no sequence homologs available on the PDB at the time of training. Very few targets met these criteria; nonetheless, we found excellent agreement between predicted and experimental structures (Fig. 1G), with the only major source of divergence being the positioning of, otherwise, well-folded domains (Fig. EV1G).
For all subsequent analyses, we created a non-redundant set of 85,162 models (Fig. 1A) representing each protein record with either ColabFold or ESMFold model, depending on which one has the highest confidence (average pLDDT score).
In parallel with our work, two other efforts have recently performed systematic structure prediction for viral proteins. Nomburg et al predicted structures for 67,715 proteins from 4463 eukaryotic viruses derived from RefSeq entries in the NCBI Viruses portal (Nomburg et al, 2024). The Big Fantastic Virus Database (BFVD) did not consider viral taxonomy but used viral sequences from UniRef30 clusters to generate 351,242 protein structures (Kim et al, 2025). We used Foldseek (van Kempen et al, 2023) to compare Viro3D to these alternative resources (Appendix Fig. S1A). There was moderate overlap with Nomburg et al's dataset with 47,306 structures being unique to Viro3D; whilst only 12,894 BFVD structures were shared with Viro3D (72,268 being unique to Viro3D). To better understand the overlaps in data, we performed proteome-level comparisons for exemplar human pathogens. Here, the majority of entries had a match in the Nomburg et al dataset, but almost no proteins had a complete counterpart in BFVD, with many appearing only as truncated proteins (Appendix Fig. S1B). Indeed, comparison of protein lengths for all entries in BFVD and Viro3D, demonstrated that short sequences predominate in BFVD (Appendix Fig. S1C). This likely reflects the composition of the UniRef30 clustered sequences that underly BFVD (Kim et al, 2025). The high frequency of matches between Viro3D and the Nomburg et al dataset permitted side-by-side comparison of model quality, assessed by pLDDT prediction confidence. Here, Viro3D models outperformed their counterparts in Nomburg et al (Appendix Fig. S1D,E). This is explained by the MSA generation method (drawing only on RefSeq virus sequences) and structure prediction workflow of Nomburg et al (Nomburg et al, 2024). The benefits of Viro3D are well illustrated by comparisons of proteome structures from the PR8 isolate of influenza A virus (a commonly used model system for experimental virology), where Viro3D provides complete highconfidence models for all proteins (Appendix Fig. S1F). Therefore, whilst the Nomburg et al dataset covers a wider diversity of virology (including viruses of unicellular eukaryotes) and BFVD represents an excellent survey of structural diversity across the entirety of virology, Viro3D is the most comprehensive database of highquality protein structure predictions for human and animal viruses.
## Exploration of the viral protein structure space by clustering and network analysis
We started with clustering the non-redundant set based on 90% bidirectional sequence overlap and 50% sequence identity that resulted in 33,151 sequence clusters. Then we selected a representative for each cluster, taking the highest confidence model (by average pLDDT score), and clustered this set of structural representatives based on 90% bi-directional structure overlap and 1e-5 e-value of the structural alignment (see "Methods"). This produced 19,067 structural clusters, 64.35% of which contained only one member (singleton clusters). Finally, we performed a Foldseek search among the representatives of these clusters to generate a structure similarity network of the viral proteins.
By applying restrictive clustering parameters, we ensured high structural homogeneity and consistency of functional annotation within each cluster (Appendix Fig. S2) but allowed homologous viral proteins to form multiple structural clusters. For instance, one of the most abundant proteins in our dataset, RNA-dependent RNA polymerase (RdRp), formed at least 110 structural clusters. The structure similarity network allowed us to address this issue by capturing communities of clusters that possess the same general protein fold or share a protein domain.
To demonstrate the power of structural network analysis, we examined the distribution of various common and hallmark viral proteins, many of which are used to define viral realms (Simmonds et al, 2023). This includes proteins involved in genome replication: RdRp, reverse transcriptase (RT), and DNA polymerase B (PolB); virion morphogenesis: single jelly roll (SJR), double jelly roll (DJR), and HK97 major capsid proteins (MCP); and membrane fusion: class-I, class-II, and class-III fusion glycoproteins (FG). We used Foldseek to identify communities of hallmark proteins by querying our structural network with a single experimental structure representative of each hallmark group (Fig. 2A). Protein communities located in the center of the network (e.g., RdRp, RT and SJR MCP) tend to be strongly interconnected, likely because of the clusters of polyproteins that possess multiple functional regions and therefore bring different communities together. This networkbased structural search enabled the identification of significantly more proteins than comparable searches using either a standalone Foldseek structural search on a non-redundant set of 85,162 structures or hidden Markov model (HMM) profiles via HHblits (Remmert et al, 2012) (Fig. 2B). For example, it resulted in a 90% increase in identified RdRp structures-an improvement not achieved by HMM-profile searches, even after five iterations (Fig. EV2C). While HHblits identified distant RdRp homologs with 12.5% amino acid identity to the probe, the structural network approach extended this further, revealing homologs with as little as 6.5% identity (Fig. EV2B).
We plotted the distribution of hallmark proteins across viral families (Appendix Fig. S3) and realms (Fig. 2C). Here, the distribution of proteins correlates well with ICTV realm classification (Simmonds et al, 2023). For example, the HK97 capsid fold is a defining feature of the Duplodnaviria and, in our analysis, is confined to this realm (Fig. 2C). A similar relationship is seen for the double jelly roll capsid fold, which is a defining feature of the Varidnaviria.
RdRp or RT are hallmarks of the Riboviria, and whilst we detect these proteins in the majority of expected species, ~9% lacked Foldseek network hits for RdRp or RT. A detailed comparison to Pfam annotations (Fig. EV2) showed that most of the missing records were short regions and simply did not meet the requirement of 90% coverage with the RdRp reference structure (Fig. EV2E). Moreover, an additional barrier to identification of RdRps/RTs is partial genome coverage and incomplete gene annotation (Fig. EV2D); this serves as a further reminder that any effort in systematic structure prediction is limited by the quality, curation, and annotation of underlying sequence data. We also showed that five members of the Varidnaviria realm (Poxviridae and Iridoviridae families) possess an RT in addition to PolB. Four of these RTs are in a community of RdRp clusters (reflecting their shared ancestry) and therefore can be found using an RdRp probe, while one is clustered with other RTs.
PolB is abundant in the realms Varidnaviria, Duplodnaviria, and several viral families that currently do not belong to any realm (e.g., Baculoviridae). We also found a PolB in the Bidnaviridae family which belongs to the Monodnaviria realm. As expected, Single jelly roll MCP is prevalent in the realms Riboviria, Monodnaviria, and family Baculoviridae. All three classes of fusion
## Molecular Systems Biology
Ulad Litvin et al glycoprotein are abundant in Riboviria, while class-III is the dominant glycoprotein in Duplodnaviria. Class-I and III fusion glycoproteins are also common among members of the Baculoviridae family (which is examined further, below).
Interestingly, PolB structural search also identified hits against other, non-PolB, proteins such as poxvirus F12L (AAL73754.1) and herpesvirus DNA helicase/primase (QBM10893.1) that possess shared domains or demonstrate similar fold architecture. Whilst these "off-target" homologs are rare, this suggests there is a sensitivity/accuracy trade-off that could be balanced through changes in clustering and search parameters (e.g., higher or lower e-value thresholds). Nonetheless, by leveraging the fundamentally conserved nature of structure over sequence, combined with clustering and network analysis, we have achieved extremely sensitive detection of deep evolutionary relatedness, beyond even high-sensitivity sequence-based approaches. This permits efficient and accurate navigation of viral protein structure space and allows mapping of protein form and function across diverse species, as demonstrated by the consistent identification of hallmark proteins across viral realms.
## R i b o v i r i a M o n o d n a v i r i a V a r i d n a v i r i a D u p l o d n a v i r i a U n c l a s s i fi e d
We also used this approach to expand functional annotation, propagating sequence-based annotation using structural clusters and network. Out of 85,162 protein records, 65.6% have at least partial Pfam annotation (Fig. 2D). By propagating these annotations to unannotated cluster members, we expanded the functional coverage by 3.99% (3395 records). The propagation of annotations to clusters that do not have any annotated members using the structural network expanded the functional coverage by an additional 3.37% (2870 records, Fig. EV3). The consistency of propagated annotation is 100% for the majority (70.3%) of protein records (4403 out of 6265 records, Appendix Fig. S4). We also expanded Gene3D (Lewis et al, 2018) and Superfamily (Pandurangan et al, 2019) annotation, which gave an increase of 11.1% and 10.7%, respectively. Expansion of gene ontology (GO) annotation (The Gene Ontology Consortium, 2019) for molecular function, biological process, and cellular component increased the number of annotated records by 10.9%, 9.2% and 4.4%, respectively. Therefore, clustering of structures and network analysis permits the identification and/or functional annotation of as-yet unclassified viral proteins. Nonetheless, it is important to note that despite propagation, there is still a high proportion of proteins with no known function (Fig. 2D).
To estimate the number of protein structures shared between viruses and cellular organisms, we performed a structure similarity search between cluster representatives and the AlphaFold Structural Database (AFDB, Fig. 2E). Consistent with the notion that viruses are a source of novel protein folds (Nomburg et al, 2024), the majority of viral protein clusters do not share detectable homology with cellular life, with only 17.8% of clusters (3393 out of 19,067) having significant structural similarity to proteins in the AFDB. Of these homologs, 71.5% are coming from Eukaryota, 15.9% from Bacteria, 1.7% from Archaea, and 10.9% from metagenomic and environmental samples. However, this number is likely to be an overestimation because in many instances, hits are coming from endogenous or symbiotic viruses. For instance, many Homo sapiens hits with high sequence identity are proteins from the integrated Human betaherpesvirus 6A (Fig. 2F), while hits from Cotesia congregata and Fopius artisans, two species of parasitoid wasps, are proteins from symbiotic viruses of genera Bracoviriform and Alphanudivirus, respectively.
In all, 14.4% of the protein records form singleton clusters and potentially represent structural novelty. Our species-focussed approach captured the genomic context of all predicted structures and, therefore, allowed us to investigate the genome positions of singleton and non-singleton clusters (Fig. 2G). Interestingly, unique protein structures are more frequently found at the start or end of linear viral genomes, while common protein structures (which are likely to be associated with critical functions) tend to have a more uniform distribution across the genome length. The picture is different for some viral realms, some of which have predominantly circular genomes (e.g., Monodnaviria; Appendix Fig. S5), nonetheless, this suggests that genomic termini are hotspots for evolutionary innovation, which may drive the emergence of novel protein functions and adaptations to hosts.
## The deep evolutionary history of class-I fusion glycoproteins
Class-I fusion glycoproteins can be found in a wide range of important human pathogens, including SARS-CoV-2, HIV, Influenza, and Ebola, and have been exapted by mammals to mediate placental morphogenesis (Mi et al, 2000). Mechanistic knowledge of these proteins informs rational vaccine design (Sanders and Moore, 2021). It is thought that class-I fusion glycoproteins share a common origin, however, extensive sequence divergence has all but erased this deep ancestry at the amino acid level. Therefore, classification of fusion glycoproteins is typically achieved through structural and functional characterization. Indeed, recent experimental structures of retroviral Env class-I glycoproteins suggest shared ancestry with fusion glycoproteins of negative-sense RNA viruses (Calcraft et al, 2024;Fernández et al, 2024); this evolutionary relationship is also supported by analysis of the viral "fossil-record" provided by endogenous viral elements (Nino Barreat and Katzourakis, 2024). Nonetheless, traditional structural biology and/or sequence analyses provide only a fragmentary picture of glycoprotein ancestry. We reasoned that homology detection, achieved through structural similarity and network analysis (Fig. 2A,C), would permit a virosphere-wide survey of class-I fusion glycoproteins and provide a clear view of their deep evolutionary history.
First, to gain a global perspective, we generated a structureinformed map of the human and animal virosphere. This allows simultaneous visualization of all viruses represented in Viro3D, onto which the distribution of proteins can be projected. This was achieved by systematic structural comparison of each virus' proteome (see "Methods"), resulting in a scatter plot, with viruses segregated by structural similarity (Fig. 3A), which broadly recapitulates viral taxonomy (Appendix Fig. S6).
Mapping the viruses that possess class-I fusion glycoproteins (identified by structural homology to RSV-F, Fig. 2C) demonstrates a distribution across viral realms, indicative of extensive genetic exchange (Fig. 3B). Alongside expected instances of class-I fusion glycoproteins in negative-sense RNA viruses, reverse-transcribing viruses, and positive-sense Nidoviruses, we discovered previously unknown class-I fusion glycoproteins in the Herpesvirales and Baculoviridae. Note that detection was achieved using a single structural reference (RSV F), we expect further instances may be found (e.g., in the Retroviridae) using alternative reference structures.
The presence of class-I fusion glycoproteins in the Herpesvirales and Baculoviridae was not expected, as both of these viral taxa are more commonly known to possess class-III fusogens; gB in Herpesvirales (Heldwein et al, 2006) and gp64 in Baculoviridae (Kadlec et al, 2008). To examine this further, we surveyed the proteomes of all Herpesvirales and Baculoviridae species represented on Viro3D, using Foldseek to search for structural homology
## Molecular Systems Biology
Ulad Litvin et al to a range of experimentally determined class-I and class-III fusion protein structures. Mapping this information against DNA polymerase phylogeny reveals the distribution of fusion mechanisms within either taxonomic group (Fig. EV4). The vast majority of viruses within the Herpesvirales exhibit strong Foldseek hits against class-III references (particularly vesicular stomatitis virus G protein and human herpesvirus-5 gB). However, a minor, highly divergent, clade containing aquatic herpesviruses (e.g., Channel catfish virus), possess no class-III homologs, instead exhibiting Foldseek hits against class-I proteins (particularly the SARS-CoV-2 spike). Class-I glycoproteins predominate in the Baculoviridae, with only a subset possessing class-III fusion glycoproteins; consistent with the confinement of gp64 to group I alphabaculoviruses (and a single species of betabaculoviruses (Ardisson-Araújo et al, 2016)).
To gain further insights into the evolutionary history of the class-I fusion mechanism, we performed structure-guided sequence alignment, permitting phylogenetic inference using both structure and sequence information (Fig. 3C; Appendix Fig. S7). This reveals four major clades (arbitrarily designated I-IV). The first clade is comprised of proteins from the Baculoviridae. The second contains reverse-transcribing viruses, from the Metaviridae (a family of LTR retrotransposons (Llorens et al, 2020)), which share insect hosts with the Baculoviridae, providing a feasible route of genetic exchange (Ozers and Friesen, 1996). The third clade is monophyletic for the Mononegavirales, this may suggest a single genetic acquisition by an ancestral negative-sense RNA virus, from which class-I fusion glycoproteins propagated throughout the order. The final clade (IV) contains a mixture of viral taxonomies, including negative-sense Aliusviridae, positive-sense Coronaviridae, and dsDNA Herpesvirales. These clade IV glycoproteins are particularly long when compared to the rest of the phylogeny (most being >1100 residues; Fig. 3C), suggesting adaptive diversification.
Comparison of structures from across the phylogeny reveals a central structurally conserved architecture common to all of the identified class-I fusion glycoproteins (Fig. 3D), consistent with recent experimental studies (Calcraft et al, 2024;Fernández et al, 2024). Representative clade IV glycoproteins from an aquatic herpesvirus and Coronaviridae (infectious bronchitis virus), however, possess very long N-terminal extensions; this includes the N-terminal domain (NTD) and receptor-binding domain (RBD) that constitute the majority of the S1 subunit in coronavirus spike. Moreover, the region conserved across all species broadly corresponds to the fusogenic S2 subunit of spike. Notably, the Coronaviridae form a high-confidence branch with the Alloherpesviridae (Fig. 3C). This is consistent with the Foldseek homology for SARS-CoV-2 spike apparent in the aquatic herpesviruses (Fig. EV4). This suggests that the coronavirus spike glycoprotein may have originated from genetic exchange with an ancestral herpesvirus.
## Discussion
In this work, we expanded the structural coverage for viral proteins by 30 times compared to that of experimental structures by generating structural models for 85,000 proteins from 4400 human and animal viruses. Approximately 64% of the produced protein models are confident predictions with an average pLDDT score above 70. Moreover, 65% of residues in the predicted dataset have high or very high confidence. We confirmed that, in general, ColabFold has a higher chance of producing a confident structure than ESMFold. However, since ESMFold models have higher confidence for almost 10% of protein records, do not require an MSA, and are usually much faster to generate, the combination of both approaches is highly beneficial.
Parallel efforts to ours have produced alternative repositories of viral structure predictions. While the Nomburg et al dataset covers a greater diversity of eukaryotic viruses, and BFVD surveys a wider diversity of viral sequence and structure space, we demonstrated that Viro3D contains the highest-quality and most complete structural models for human and animal viruses. We anticipate that Viro3D will be particularly valuable for informing experimental molecular virology and evolutionary studies of functional diversity across the animal virosphere.
By performing clustering of protein structures encoded by human and animal viruses, the diversity of viral proteins was reduced to ~19,000 distinct protein structures; almost 65% of these structures are unique within our dataset. Interestingly, in viruses with linear genomes (realms Varidnaviria, Duplodnaviria, and, to some extent, Riboviria), these unique proteins are usually found closer to the ends of the genomes, which probably represent hot spots of gene acquisition or de novo gene origination.
More than 82% of distinct viral structures are unique to viruses and do not have apparent structural homologs in cellular organisms. This high percentage of unique viral structures may seem surprising given the extensive horizontal gene transfer between viruses and their hosts, and may suggest that even acquired proteins undergo extensive remodeling and diversification in the context of a viral genome. Therefore, we acknowledge that the number of viral proteins retaining more distant structural similarity to cellular components may be higher than that demonstrated in our analysis; indeed, parallel studies have identified slightly higher numbers of cellular homologs in viruses (Nomburg et al, 2024).
We showed that structure similarity searches combined with a structural network allowed us to substantially accelerate the search process by increasing the number of homologous structures identified in a single run. For instance, using single RdRp and RT structures, we identified RNA-directed polymerases for 90% of viruses in the Riboviria realm. This demonstrates that structural searches and network analysis are an extremely efficient means of traversing large evolutionary distances.
Classification of viral realms (the highest taxonomic rank for viruses) is achieved through the identification of hallmark genes. These are, semi-arbitrarily chosen, proteins involved in genome replication and capsid morphogenesis. For example, RdRp and RT are hallmarks of the Riboviria, while DJR-MCP and HK97 MCP are specific to Varidnaviria and Duplodnaviria, respectively (Krupovic and Koonin, 2017). PolB is distributed more broadly: realms Varidnaviria, Duplodnaviria, and some unclassified DNA viruses (e.g., Baculoviridae). Our analyses are largely consistent with, and verify, these taxonomic relationships, with rare exceptions. Most families in the Monodnaviria rely on rolling-circle replication endonuclease (Kazlauskas et al, 2019), however, we found PolB in one family of this realm: Bidnaviridae (Krupovic and Koonin, 2014). Similarly, we found multiple instances of RT/RdRP in members of the Varidnaviria (Poxviridae and Iridoviridae). These edge cases likely reflect inter-realm horizontal gene transfer and serve as reminders that the mosaic nature of viral genomes will often present challenges for taxonomic classification. As an additional example, membrane fusion glycoproteins are common but not considered hallmark proteins, and are spread across multiple realms. Class-III fusion glycoproteins seem to be the most widely distributed among human and animal viruses, being absent only in the realms Monodnaviria and Ribozyviria.
Fusion glycoproteins are fundamentally important for virus transmission, pathogenesis, and spillover, and are major targets for host immunity. Viro3D has provided a new opportunity to comprehensively survey the distribution of glycoproteins and, using structure-guided approaches, infer their evolutionary past. Using just a single structural reference to query Viro3D (RSV F glycoprotein), we identified 251 highly diverse class-I fusion glycoproteins, many sharing <10% sequence identity. This includes both known instances (e.g., in Mononegavirales) and unexpected hits within the Herpesvirales and Baculoviridae. Systematic Foldseek surveys across herpes-and baculoviruses taxa suggest a mixture of both class-III and class-I fusion mechanisms. In fact, class-I fusion appears to predominate in the Baculoviridae (including in those species with gp64) and homology frequently maps to known/suspected fusogens such as the F protein of betabaculoviruses (Rohrmann, 2019). This suggests that class-I glycoproteins are the ancestral fusion mechanism in Baculoviruses, with the class-III gp64 being acquired as an alternative virus entry system (Rohrmann, 2019). Indeed, gp64 may have originated through a horizontal gene transfer from thogotoviruses (Milhomem Pilati Rodrigues et al, 2025).
Combined structure and sequence-based phylogenetics provided the first view of the deep evolutionary history of class-I fusion glycoproteins. This is consistent with multiple independent horizontal gene transfers from an, as yet unidentified, ancestral source. Whilst all of the identified glycoproteins share a structurally conserved central fold, one clade in particular (including coronaviruses and aquatic herpesviruses) shows signs of extensive adaptive diversification. These glycoproteins have doubled in size by gaining novel regulatory elements, such as the NTD and RBD in the S1 subunit of spike. This provides an evolutionary perspective on membrane fusion in coronaviruses: the S1 subunit is proteolytically separated from spike and is shed following receptor engagement, leaving the structurally conserved S2 to mediate fusion (Grunst et al, 2024). Thus, to achieve cellular entry, coronaviruses undergo a regulated unmasking of an ancestral fusogen, which is fundamentally conserved across all class-I fusion glycoproteins. The phylogenetic topology of this clade suggests that coronaviruses gained their spike glycoprotein by genetic exchange with an ancestor of the known aquatic herpesviruses; albeit, we cannot eliminate the involvement of other unsampled, or extinct, taxa in this exchange. In summary, using class-I fusion glycoproteins as an exemplar, we have demonstrated that structure-guided discovery, enabled by Viro3D, is likely to provide unprecedented insights on the origins and evolution of viruses and their defining proteins.
Whether for lab-based investigators wishing to bring threedimensional context to their molecular virology experiments, researchers performing structure-guided design of therapies, or computational biologists studying deep virus evolution, we expect the rich structural dataset presented here to be a valuable resource for the virology community. Viro3D is fully searchable and browsable here: https://viro3d.cvr.gla.ac.uk/.
## Methods
## Reagents and tools table
## Protein structure prediction with ColabFold
Protein models for all 85,162 records were predicted using LocalColabFold v.1.5.2 with default settings (Mirdita et al, 2022). Multiple sequence alignments (MSAs) were constructed for each record using MMseqs2 v.15.6f452 (Steinegger and Söding, 2017), and colabfold_envdb_202108, pdb100_230517, uniref30_2302 databases installed locally. For each protein record, ColabFold produced five models using three recycles and no PDB templates. All models were ranked based on the mean predicted local-distance difference test (pLDDT) score. The top-ranked model was subjected to constrained relaxation by gradient descent in the Amber force field (Pearlman et al, 1995;Hornak et al, 2006;Mirdita et al, 2022) with the following settings: max_iterations 2000; tolerance 2.39; stiffness 10.0. The relaxed models were used for structural analysis.
## Protein structure prediction with ESMFold
We also predicted 84,964 protein models for 69,043 proteins without peptide or region annotation, 4070 mature peptides, 11,767 protein regions, and 84 proteins with peptide and/or region annotation using ESMFold module of ESM-2 v.1.0.3 (Lin et al, 2023) with default settings. For each record ESMFold produced 1 model using four recycles. For records with protein length greater than 1236 aa, the following settings were used:
--max-tokens-per-batch 1 --chunk-size 128. Due to memory constraints on the GPU, we were unable to predict models for records with protein length greater than 2840 aa. ESMFold models were subjected to relaxation in the Amber force field as explained above.
## Estimation of structural coverage expansion
To estimate the expansion of structural coverage relative to experimentally determined structures available in the PDB (Berman et al, 2000), and to consider the influence of PDB training data on prediction accuracy, we performed a sequence similarity search of 71,269 protein records against the PDB 2024-02-20 using MMseqs2 v.15.6f452 (Steinegger and Söding, 2017) easy-search command with the following sensitivity settings: -s 7.5 --max-seqs 100,000 -e 1e-3. To calculate the percentage of residues covered by PDB structures, only PDB hits with sequence identity greater or equal to 95% were retained. To calculate the percentage of residues covered by homologs from the PDB, PDB entries released after 2018-04-30 or 2020-05-01, training data cut-offs for AlphaFold2 (Jumper et al, 2021) and for ESMFold (Lin et al, 2023), respectively, were filtered out, only PDB hits with sequence identity of 30% or greater were retained.
## Clustering procedure
Overall, 79,036 ColabFold and 6126 ESMFold models with the highest pLDDT score per protein record were combined into a non-redundant dataset for the downstream structural analysis. ESMFold models were used as representatives of protein records if they had mean pLDDT score greater than 50, and this score was greater than the mean pLDDT score of a corresponding ColabFold structure. First, we clustered 85,162 protein records based on their sequence similarity using MMseqs2 v.15.6f452 (Steinegger and Söding, 2017) easy-cluster command with minimum sequence identity of 50% and bi-directional coverage of 90% (--min-seq-id 0.5 --cov-mode 0 -c 0.9 -e 1e-3 --cluster-mode 0). Second, we defined structural representatives for 33,151 MMseqs2 clusters by choosing the member with the highest model pLDDT score. We clustered 33,151 structure representatives based on structural similarity using Foldseek v.9.427df8a (van Kempen et al, 2023) easy-cluster command with no minimum sequence identity but bidirectional coverage of 90% (--min-seq-id 0 --cov-mode 0 -c 0.9 -e 1e-5), producing 19,067 structural clusters (Dataset EV3). Foldseek e-value cutoff (1e-5) is an arbitrary threshold chosen to achieve high-sensitivity for detecting structural homologs; however, this can result in some "off-target" hits such as the herpesvirus DNA helicase/primase proteins identified by homology to DNA polymerase B.
## Structural and functional homogeneity of clusters
To estimate structural homogeneity within each non-singleton cluster, we extracted LDDT and alignment TM scores from a reciprocal Foldseek v.9.427df8a (van Kempen et al, 2023) easysearch of the non-redundant dataset with next settings: --exhaustive-search -e 0.1. For each cluster, we calculated the median LDDT and TM score of the alignments between the cluster representative and other cluster members. To estimate the consistency of function annotation within non-singleton clusters, we relied on a sequencebased Pfam annotation acquired using InterProScan v.5.69-101.0 (Jones et al, 2014) with default settings. For each protein record, we retained a functional annotation with the highest sequence coverage and calculated the percentage of annotated records that share the predominant Pfam annotation for each cluster containing at least two annotated members.
## Structure similarity network analysis
Similar to MMseqs2 clusters, we defined structural representatives for 19,067 Foldseek clusters by choosing the model with the highest pLDDT score from each cluster. We performed a structural comparison of the representatives using Foldseek v.9.427df8a (van Kempen et al, 2023) easy-search command with default settings and applied the results to construct a structure similarity network using NetworkX v.3.3 (Hagberg et al, 2008). Only structural representatives with pLDDT score above 50 and at least one connection with Foldseek e-value below 1e-3 were retained, producing a graph with 7812 nodes and 37,751 edges (Dataset EV4). The graph was visualized using NetworkX spring_layout() function with k of 0.3 and 500 iterations.
## Identification of hallmark proteins and fusion glycoproteins
We started with one iteration of Foldseek v.9.427df8a (van Kempen et al, 2023) easy-search against the non-redundant structural dataset with the following settings -e 1e-5 --max-seqs 10,000 and using a set of experimentally determined structures as probes: poliovirus RNA-dependent RNA polymerase (PDB ID: 4R0E (Moustafa et al, 2014)), human immunodeficiency virus reverse transcriptase (1HMV (Rodgers et al, 1995)), Phi29 DNA polymerase B (2PY5 (Berman et al, 2007)), circovirus rolling-circle replication endonuclease (8H56 (Guan et al, 2023)), papillomavirus hexameric superfamily 3 helicase (5A9K (Chaban et al, 2015)), poliovirus VP3 protein (8E8R (Charnesky et al, 2023)) as single jelly roll capsid protein, human adenovirus 5 hexon protein (6B1T (Dai et al, 2017)) as double jelly roll capsid protein, varicella zoster virus HK97 capsid protein (6LGL (Wang et al, 2020)), respiratory syncytial virus F protein (6APB (Goodwin et al, 2018)) as a class-I fusion glycoprotein, spondweni virus E protein (6ZQI (Renner et al, 2021)) as a class-II fusion glycoprotein, human cytomegalovirus gB protein (7KDP (Liu et al, 2021)) as a class-III fusion glycoprotein.
To test if querying the structure similarity network increases the number of hits, we performed a separate Foldseek easy-search using the same list of probes and search parameters, but this time only against a set of 19,067 structural representatives of Foldseek clusters. We then expanded the number of structurally similar clusters by adding Foldseek clusters that have a significant network connection (e-value 1e-5 and 90% coverage of the Foldseek hit). The final list of the hits contained all members of the Foldseek clusters identified using this approach. For comparison we also performed up to five iterations of a profile-profile HHsuite v3.3.0 HHblits search (Remmert et al, 2012) against protein sequences in the non-redundant dataset using sequences from the hallmark PDBs as probes with following settings -e 1e-3 -Z 5000 -B 5000.
## Expansion of functional annotations
To demonstrate how structural information can be used to expand functional annotation, first we performed a sequence-based annotation of 85,162 proteins using InterProScan v.5.69-101.0 (Jones et al, 2014) with default settings. We focused on annotations from Pfam (Mistry et al, 2021), Superfamily (Pandurangan et al, 2019), and Gene3D (Lewis et al, 2018)
## Structural comparison to the AlphaFold structural database
We compared 19,067 structural representatives of Foldseek clusters to protein models in the AlphaFold Structural Database (Varadi et al, 2024) using Foldseek v.9.427df8a (van Kempen et al, 2023) easy-search option with e-value of 1e-3. For each structural representative, we retained the hit with the lowest e-value.
## Comparison to alternative repositories of viral protein structure predictions
We systematically compared Viro3D to the dataset generated by Nomburg et al and to the Big Fantastic Virus Database (BFVD) (Kim et al, 2025;Nomburg et al, 2024). We first used Foldseek searches to assess overlap with Viro3D. Here, protein models were considered shared if they had a Viro3D Foldseek hit with an e-value lower than 1e-5, sequence identity equal or greater than 95% and query coverage equal or greater than 95%. Next, we examined structure predictions from a panel of priority human viral pathogens. Using taxonomic identifiers to find the relevant models, we compiled structure sets for each pathogen from Viro3D, Nomburg et al and BFVD. Using protein sequence to identify matching models (95% sequence similarity), we evaluated coverage of proteins (relative to Viro3D) and the confidence of structure prediction (assessed using pLDDT values).
## Building a structural-similarity map displaying all viral species
We first performed all-vs-all Foldseek comparison of the entire non-redundant structure set (85,162 models), therefore surveying global structural-similarity. Using this, for any given virus' proteome (i.e., collection of structures), we extracted the lowest evalue against each of the 85,162 models. If no e-value was present for a given virus-model pair, we substituted an arbitrary high value of 10. This resulted in a matrix where each of the 4407 viruses is represented by 85,162 structural similarity scores. We then used principal component dimensionality reduction to group viruses based on the structural similarity of their respective proteomes.
## Phylogenetic analysis of the class I fusion glycoprotein
Class-I fusion proteins were identified using Foldseek homology search (see above, identification of hallmark proteins and fusion glycoproteins), resulting in 259 structural models. These were filtered to remove those with extremely low confidence (pLDDT ≤40), resulting in a final set of 251 structures with mean pLDDT=70.4( + /-7.3) and mean pTM=0.58( + /-0.08); a list of these proteins is provided in Dataset EV6. We note that confidence metric filtering may be important to remove erroneous homologs. These structures were aligned using two approaches: (i) FoldMason (Gilchrist et al, 2024) (easy-msa -report-mode 1) and (ii) the famsa3di method described in Puente-Lelievre et al ( 2024) (Deorowicz et al, 2016;Puente-Lelievre et al, 2024), producing corresponding 3Di and amino acid alignments. Subsequently, the phylogenetic relationships were determined using IQ-TREE v.2.3.6 (Minh et al, 2020) using the two 3Di alignments (FoldMason, famsa3di) (Appendix Fig. S7). Suitable substitution models were tested by BIC using ModelFinder (Kalyaanamoorthy et al, 2017) and the custom 3DI substitution matrix (Puente-Lelievre et al, 2024) with empirically counted frequencies from the alignment and accounting for rate heterogeneity under a FreeRate model was the best (for both 3Di alignments (five FreeRate categories for famsa3di, 3DI + F + R5, and six categories for FoldMason, 3DI + F + R6). To inform phylogenetic reconstruction with amino acid homology, we also performed partitioned model inferences where one partition corresponds to the 3Di alignment, using the 3DI + F + R6 substitution model and the other to the amino acid alignment, using the BIC-selected most suitable substitution model for each corresponding amino acid alignment (famsa3di: WAG + F + R9, FoldMason: WAG + F + R8). Branch support was assessed using 10,000 replicates of the ultrafast bootstrap approximation method (Hoang et al, 2018). Phylogenetic trees were visualized and prepared for publication with the ggtree R package v3.10.1 (Yu et al, 2017) (Appendix Fig. S7). The phylogeny that had a topology most consistent with taxonomy (famsa3di alignment, 3DI + AA) is presented in Fig. 3C. All accompanying structure models were visualized and prepared for publication with UCSF ChimeraX (Meng et al, 2023).
## Phylogenetic analysis of the DNA polymerase from Herpesvirales and Baculoviridae
In order to survey class-I and III fusion proteins across the Herpesvirales and Baculoviridae, we reconstructed polymerasebased phylogenies that recapitulate continuous underlying evolutionary histories, onto which we can map glycoprotein distribution. We retrieved DNA polymerases from each species in the two families based on our hallmark gene Foldseek search, using the Phi29 DNA polymerase B (PDB ID: 2PY5 (Berman et al, 2007)) as a query. One match with the highest similarity was used for each species, resulting in 119 Herpesvirales and 94 Baculoviridae PolB amino acid sequences. Each set of sequences was aligned using mafft v7.525 (Katoh and Standley, 2013) (--localpair option) and phylogenies were inferred using IQ-TREE v.2.3.6 (Minh et al, 2020) under the most suitable substitution model selected by ModelFinder (Kalyaanamoorthy et al, 2017).
Foldseek homology searches against class-I and III fusion proteins in the Herpesvirales and Baculoviridae
We performed Foldseek searches against the entire Viro3D database using experimentally determined protein structures for class-I and III fusion glycoproteins (RSV F pre-fusion PDB ID:5EA3 (Battles et al, 2016), RSV F post-fusion PDB ID:6APB (Goodwin et al, 2018), hMPV F PDB ID:7TJQ (Rush et al, 2022), SARS-CoV-2 S PDB ID:6VXX (Walls et al, 2020), baculovirus gp64 PDB:8YG6 (Guo et al, 2024), VSV G PDB ID:5I2F (Roche et al, 2007), HHV-5 gB PDB ID:7KDP (Liu et al, 2021)). We examined the results of this for the presence of proteins from each of the species represented in our Herpesvirales and Baculoviridae phylogenies, extracting Foldseek e-values or assigning an arbitrarily large e-value (10) when no protein was detected. E-values were plotted against corresponding phylogenies using iTOL (Letunic and Bork, 2024).
## Development of the Viro3D web-resource
The frontend of Viro3D was built using React and Typescript, providing a dynamic user interface. To enhance specific features, we integrated PDBE-Molstar Viewer to provide an interactive 3D rendering of protein structures molstar); Sodaviz (https://github.com/sodaviz) to allow for browsing protein structures across the virus genome; KonvaJS to construct the interactive map of viruses based on structure similarities (https:// github.com/konvajs/konva). The Viro3D backend was developed using FastAPI to allow for programmatic access to the data. Biopython was integrated for leveraging the NCBI BLASTp command-line wrapper to enable searching by protein sequence. The MongoDB Community Edition was used to store the data. The source code for the web resource can be found at https://github.com/centre-for-virusresearch/viro3d-frontend and at https://github.com/centre-for-virusresearch/viro3d-backend.
## Expanded View Figures
Residue occupation in Whole proteomes for all species within either taxonomic group were surveyed, using Foldseek, for structural homology against class-I and class-III fusion proteins. Foldseek structural homology scores were mapped against the underlying DNA polymerase amino acid sequence phylogeny to reveal the distribution of fusion mechanisms. Presented phylogenies are midpoint rooted. Heatmaps display Foldseek log transformed e-values (as indicated in the key) for the stated references, including RSV F protein in both pre-and post-fusion states (see Methods for details). For orientation in the Herpesvirales phylogeny, human herpesviruses have tips colored blue with species labels, whereas the aquatic herpesviruses identified in our structural cluster search (Fig. 3B) are colored yellow. Scale bar represents amino acid substitutions per site. hMPV human metapneumovirus, VSV vesicular stomatitis virus, HHV-5 human herpesvirus-5.
## References
1. Akdel, Pires, Pardo et al. (2022) "A structural biology community assessment of AlphaFold2 applications"
2. Ardisson-Araújo, Melo, Clem et al. (2016) "A betabaculovirus-encoded gp64 homolog codes for a functional envelope fusion protein" *J Virol*
3. Battles, Langedijk, Furmanova-Hollenstein et al. (2016) "Molecular mechanism of respiratory syncytial virus fusion inhibitors" *Nat Chem Biol*
4. Berman, Kamtekar, Goodman et al. (2007) "Structures of phi29 DNA polymerase complexed with substrate: the mechanism of translocation in B-family polymerases" *EMBO J*
5. Berman, Westbrook, Feng et al. (2000) "The Protein Data Bank" *Nucleic Acids Res*
6. Breitbart, Bonnain, Malki et al. (2018) "Phage puppet masters of the marine microbial realm" *Nat Microbiol*
7. Calcraft, Stanke-Scheffler, Nans et al. (2024) "Integrated cryoEM structure of a spumaretrovirus reveals crosskingdom evolutionary relationships and the molecular basis for assembly and virus entry" *Cell*
8. Chaban, Stead, Ryzhenkova et al. (2015) "Structural basis for DNA strand separation by a hexameric replicative helicase" *Nucleic Acids Res*
9. Charnesky, Faust, Lee et al. (2023) "A human monoclonal antibody binds within the poliovirus receptor-binding site to neutralize all three serotypes" *Nat Commun*
10. Chothia, Lesk (1986) "The relation between the divergence of sequence and structure in proteins" *EMBO J*
11. Dai, Wu, Sun et al. (2017) "Atomic structures of minor proteins VI and VII in human adenovirus" *J Virol*
12. Daugherty, Malik (2012) "Rules of engagement: molecular insights from host-virus arms races" *Annu Rev Genet*
13. Deorowicz, Debudaj-Grabysz, Gudyś (2016) "FAMSA: fast and accurate multiple sequence alignment of huge protein families" *Sci Rep*
14. Fernández, Bontems, Brun et al. (2024) "Structures of the Foamy virus fusion protein reveal an unexpected link with the F protein of paramyxo-and pneumoviruses" *Sci Adv*
15. Gilchrist, Mirdita, Steinegger (2024) "Multiple protein structure alignment at scale with FoldMason"
16. Goodwin, Gilman, Wrapp et al. (2018) "Infants infected with respiratory syncytial virus generate potent neutralizing antibodies that lack somatic hypermutation" *Immunity*
17. Gorbalenya, Krupovic, Mushegian et al. (2020) "The new scope of virus taxonomy: partitioning the virosphere into 15 hierarchical ranks" *Nat Microbiol*
18. Gregory, Zayed, Conceição-Neto et al. (2019) "Marine DNA viral macro-and microdiversity from pole to pole" *Cell*
19. Grunst, Qin, Dodero-Rojas et al. (2024) "Structure and inhibition of SARS-CoV-2 spike refolding in membranes" *Science*
20. Tian, Jing, Yuan et al. (2023) "Crystal structure of the dimerized of porcine circovirus type II replication-related protein Rep′" *Proteins Struct Funct Bioinforma*
21. Güemes, Youle, Cantú et al. (2016) "Viruses as winners in the game of life" *Annu Rev Virol*
22. Guo, Li, Bai et al. (2024) "Structural transition of GP64 triggered by a pH-sensitive multihistidine switch" *Nat Commun*
23. Hagberg, Swart, Schult (2008) "Exploring network structure, dynamics, and function using NetworkX"
24. Heldwein, Lou, Bender et al. (2006) "Crystal structure of glycoprotein b from herpes simplex virus 1" *Science*
25. Hoang, Chernomor, Haeseler et al. (2018) "UFBoot2: improving the ultrafast bootstrap approximation" *Mol Biol Evol*
26. Hornak, Abel, Okur et al. (2006) "Comparison of multiple Amber force fields and development of improved protein backbone parameters" *Proteins Struct Funct Bioinforma*
27. Hou, He, Fang et al. (2024) "Using artificial intelligence to document the hidden RNA virosphere" *Cell*
28. Illergård, Ardell, Elofsson (2009) "Structure is three to ten times more conserved than sequence-a study of structural response in protein cores" *Proteins Struct Funct Bioinforma*
29. Ingles-Prieto, Ibarra-Molero, Delgado-Delgado et al. (2013) "Conservation of protein structure over four billion years"
30. "International Committee on Taxonomy of Viruses (ICTV) (2023) Virus Metadata Resource (VMR) 38v2"
31. Iranzo, Krupovic, Koonin (2016) "The double-stranded DNA virosphere as a modular hierarchical network of gene sharing" *mBio*
32. Iranzo, Puigbò, Lobkovsky et al. "Koonin EV (2016b) Inevitability of genetic parasites" *Genome Biol Evol*
33. Irwin, Pittis, Richards et al. (2022) "Systematic evaluation of horizontal gene transfer between eukaryotes and viruses" *Nat Microbiol*
34. (2019) "Origins and evolutionary consequences of ancient endogenous retroviruses" *Nat Rev Microbiol*
35. Jones, Binns, Chang et al. (2014) "InterProScan 5: genome-scale protein function classification" *Bioinformatics*
36. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
37. Kadlec, Loureiro, Abrescia et al. (2008) "The postfusion structure of baculovirus gp64 supports a unified view of viral fusion machines" *Nat Struct Mol Biol*
38. Kalyaanamoorthy, Minh, Wong et al. (2017) "ModelFinder: fast model selection for accurate phylogenetic estimates" *Nat Methods*
39. Katoh, Standley (2013) "MAFFT multiple sequence alignment software version 7: improvements in performance and usability" *Mol Biol Evol*
40. Kazlauskas, Varsani, Koonin et al. (2019) "Multiple origins of prokaryotic and eukaryotic single-stranded DNA viruses from bacterial and archaeal plasmids" *Nat Commun*
41. Kim, Karin, Mirdita et al. (2025) "BFVD-a large repository of predicted viral protein structures" *Nucleic Acids Res*
42. Koonin, Dolja, Krupovic (2022) "The logic of virus evolution" *Cell Host Microbe*
43. Koonin, Dolja, Krupovic et al. (2020) "Global organization and proposed megataxonomy of the virus world" *Microbiol Mol Biol Rev*
44. Krupovic, Dolja, Koonin (2019) "Origin of viruses: primordial replicators recruiting capsids from hosts" *Nat Rev Microbiol*
45. Krupovic, Koonin (2014) "Evolution of eukaryotic single-stranded DNA viruses of the Bidnaviridae family from genes of four other groups of widely different viruses" *Sci Rep*
46. Krupovic, Koonin (2017) "Multiple origins of viral capsid proteins from cellular ancestors" *Proc Natl Acad Sci*
47. Lefkowitz, Dempsey, Hendrickson et al. (2018) "Virus taxonomy: the database of the International Committee on Taxonomy of Viruses (ICTV)" *Nucleic Acids Res*
48. Letunic, Bork (2024) "Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool" *Nucleic Acids Res*
49. Lewis, Sillitoe, Dawson et al. (2018) "Gene3D: extensive prediction of globular domains in proteins" *Nucleic Acids Res*
50. Lin, Akin, Rao et al. (2023) "Evolutionary-scale prediction of atomic-level protein structure with a language model" *Science*
51. Liu, Heim, Chi et al. (2021) "Prefusion structure of human cytomegalovirus glycoprotein B and structural basis for membrane fusion" *Sci Adv*
52. Llorens, Soriano, Krupovic (2020) "ICTV virus taxonomy profile: Metaviridae" *J Gen Virol*
53. Mavrich, Hatfull (2017) "Bacteriophage evolution differs by host, lifestyle and genome" *Nat Microbiol*
54. Meng, Goddard, Pettersen et al. (2023) "UCSF ChimeraX: tools for structure building and analysis" *Protein Sci*
55. Mi, Lee, Li et al. (2000) "Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis" *Nature*
56. Pilati Rodrigues, Janssen, Da Silva et al. (2025) "Experimental and evolutionary evidence for horizontal transfer of an envelope fusion protein gene between thogotoviruses and baculoviruses" *J Virol*
57. Minh, Schmidt, Chernomor et al. (2020) "IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era" *Mol Biol Evol*
58. Mirdita, Schütze, Moriwaki et al. (2022) "ColabFold: making protein folding accessible to all" *Nat Methods*
59. Mistry, Chuguransky, Williams et al. (2021) "Pfam: the protein families database in 2021" *Nucleic Acids Res*
60. Moody, Álvarez-Carretero, Mahendrarajah et al. (2024) "The nature of the last universal common ancestor and its impact on the early Earth system" *Nat Ecol Evol*
61. Moustafa, Korboukh, Arnold et al. (2014) "Structural dynamics as a contributor to error-prone replication by an RNA-dependent RNA polymerase*" *J Biol Chem*
62. Barreat, Katzourakis (2024) "Deep mining reveals the diversity of endogenous viral elements in vertebrate genomes" *Nat Microbiol*
63. Nomburg, Doherty, Price et al. (2024) "Birth of protein folds and functions in the virome" *Nature*
64. Ozers, Friesen (1996) "TheEnv-like open reading frame of the baculovirusintegrated retrotransposon TED encodes a retrovirus-like envelope protein" *Virology*
65. Pandurangan, Stahlhacke, Oates et al. (2019) "The SUPERFAMILY 2.0 database: a significant proteome update and a new webserver" *Nucleic Acids Res*
66. Pearlman, Caldwell, Ross et al. (1995) "AMBER, a package of computer programs for applying molecular mechanics, normal mode analysis, molecular dynamics and free energy calculations to simulate the structural and energetic properties of molecules" *Comput Phys Commun*
67. Peck, Lauring (2018) "Complexities of viral mutation rates" *J Virol*
68. Puente-Lelievre, Malik, Douglas et al. (2024) "Tertiary-interaction characters enable fast, model-based structural phylogenetics beyond the twilight zone"
69. Ravantti, Martinez-Castillo, Abrescia (2020) "Superimposition of viral protein structures: a means to decipher the phylogenies of viruses" *Viruses*
70. Remmert, Biegert, Hauser et al. (2012) "HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment" *Nat Methods*
71. Renner, Dejnirattisai, Carrique et al. (2021) "Flavivirus maturation leads to the formation of an occupied lipid pocket in the surface glycoproteins" *Nat Commun*
72. Richardson, Allen, Baldi et al. (2023) "MGnify: the microbiome sequence data analysis resource in 2023" *Nucleic Acids Res*
73. Roche, Rey, Gaudin et al. (2007) "Structure of the prefusion form of the vesicular stomatitis virus glycoprotein G" *Science*
74. Rodgers, Gamblin, Harris et al. (1995) "The structure of unliganded reverse transcriptase from the human immunodeficiency virus type 1" *Proc Natl Acad Sci*
75. Rohrmann (2019) "Structural proteins of baculovirus occlusion bodies and virions"
76. Rush, Brar, Hsieh et al. (2022) "Characterization of prefusion-Fspecific antibodies elicited by natural infection with human metapneumovirus" *Cell Rep*
77. Sanders, Moore (2021) "Virus vaccines: proteins prefer prolines" *Cell Host Microbe*
78. Shkoporov, Hill (2019) "Bacteriophages of the human gut: the "known unknown" of the microbiome" *Cell Host Microbe*
79. Simmonds, Adriaenssens, Zerbini et al. (2023) "Four principles to establish a universal virus taxonomy" *PLoS Biol*
80. Soh, Ognibene, Sanders et al. (2024) "A proteome-wide structural systems approach reveals insights into protein families of all human herpesviruses" *Nat Commun*
81. Steinegger, Söding (2017) "MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets" *Nat Biotechnol*
82. Suttle (2007) "Marine virusesmajor players in the global ecosystem" *Nat Rev Microbiol*
83. Gene, Consortium (2019) "The Gene Ontology Resource: 20 years and still GOing strong" *Nucleic Acids Res*
84. (2023) "UniProt: the Universal Protein Knowledgebase in 2023" *Nucleic Acids Res*
85. Van Kempen, Kim, Tumescheit et al. (2023) "Fast and accurate protein structure search with Foldseek" *Nat Biotechnol*
86. Varadi, Bertoni, Magana et al. (2024) "AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences" *Nucleic Acids Res*
87. Walls, Park, Tortorici et al. (2020) "Structure, function, and antigenicity of the SARS-CoV-2 spike glycoprotein" *Cell*
88. Wang, Zheng, Pan et al. (2020) "Near-atomic cryo-electron microscopy structures of varicella-zoster virus capsids" *Nat Microbiol*
89. Yu, Smith, Zhu et al. (2017) "ggtree: an R package for visualization and annotation of phylogenetic trees with their covariates and other associated data" *Methods Ecol Evol*
90. Zhang, Chen, Qin et al. (2019) "Expanding the RNA virosphere by unbiased metagenomics" *Annu Rev Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12730501&blobtype=pdf | # Use of the Split Luciferase Complementation Assay to Identify Novel Small Molecules That Disrupt Essential Protein-Protein Interactions of Viruses
Tisa Biswas, Richard Sutton
## Abstract
Protein-protein interactions (PPIs) are fundamental to viral replication, regulating transcription, assembly, and genome packaging. Despite their biological importance, few FDA-approved therapeutics directly target these complexes. The split luciferase complementation assay (SLCA) is a quantitative bioluminescence system to measure proteinprotein interactions in vitro after the proteins in question have been fused in-frame to N and C luciferase fragments. The SLCA can be performed both in vitro using purified protein components and in live cells, as the luciferase substrate luciferin is cell-permeable, allowing detection of protein interactions in intact cells. Assay performance, however, depends on the expression level and stability of the fusion proteins used. SLCA has been successfully applied to target Rev-Rev interactions in human immunodeficiency virus type 1 (HIV-1) for high-throughput small-molecule screening, establishing a proof-of-concept to target other parts of the viral life cycle. The system can be extended to other pathogens that currently do not have specific antiviral therapies such as HIV-1 Tat-cyclin T1, Capsid dimerization in Dengue virus, capsid interactions in equine encephalitis viruses, capsid assembly in Epstein-Barr virus, and nucleoprotein oligomerization in rabies virus. These applications demonstrate how the assay's ability to quantify multimeric structural interactions is essential to viral replication, providing an avenue to identify small-molecule inhibitors that prevent viral replication and spread. Although there are challenges to protein stability and assay optimization, the sensitivity and adaptability of the SLCA has broader implications in virology to accelerate antiviral drug development.
## 1. Introduction
Over the past century, outbreaks of the human immunodeficiency virus-1 (HIV-1), Dengue virus, Ebola virus, and, most recently, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have demonstrated both the speed and scale at which viruses can spread, becoming pandemics and affecting hundreds of millions of people. These viruses are borderless; they are not exclusive to developing countries but affect human populations worldwide, raising concerns for global health security. Dengue virus alone infects about 400 million people annually, and HIV-1 continues to impact more than 39 million individuals across the globe [1,2]. Although specific treatments exist for some of these infectious diseases such as HIV and influenza, they are often limited in efficacy and accessibility, leading to high morbidity rates. Importantly, many of these viruses lack specific therapies that directly target the pathogen. These challenges highlight the urgent need to identify new therapeutic strategies that can directly impact protein-protein interactions to prevent viral replication and prevent the virus from spreading.
Viruses rely heavily on the molecular machinery of host cells to replicate and spread throughout the body. This process, also known as obligate intracellular parasitism, is often mediated by protein-protein interactions between the viral protein and host factors. Once the virus enters the host cell, the protein-protein interactions drive viral transcription and replication. For instance, HIV-1 Tat binds to human protein cyclin T1 to initiate transcriptional elongation, while flaviviruses such as Dengue rely on capsid-capsid interactions to form a higher-order, multimeric structure [3,4]. These interactions are specific and critical to viral replication, highlighting the associated proteins as potential antiviral targets.
A proven method to target these interactions can be accomplished using the split luciferase complementation assay (SLCA), which is designed based on bioluminescence by reconstituting a functional firefly luciferase enzyme (~62 kDa in size) [5]. Luciferase, originally found in Photinus pyralis fireflies, is a class of enzymes that catalyze bioluminescent reactions across various organisms-including click beetles, lantern fish, and marine ostracod Vargula hilgendorfii, also known as the sea firefly, where the enzyme functions in processes such as mating, communication, defense, camouflage, and predation [6]. The SLCA specifically uses firefly luciferase, which is divided into two inactive fragments: the N-terminal (Nluc) and the C-terminal (Cluc). The firefly luciferase enzyme can be successfully bisected as follows: Nluc (1-416 amino acids) and Cluc (398-500 amino acids) [7]. The NLuc component is fused in-frame to one protein of interest, while the Cluc is fused to the other. When the two proteins interact in the assay, it brings the luciferase fragments in close enough proximity to form an active enzyme (Figure 1). This reconstituted enzyme then generates a quantifiable bioluminescent signal in the presence of ATP and luciferin; ATP is poorly cell-permeable due to its hydrophilic nature, whereas luciferin is cell-permeable [7]. This system is used to detect interactions between structural proteins that multimerize, precisely quantifying protein-protein interactions that are critical to viral replication.
The SLCA can be performed both in vivo and in vitro, allowing quantitative detection of protein-protein interactions in live cells, cell lysates, or purified proteins [8][9][10]. In mammalian cell systems, constructs encoding NLuc-and CLuc-tagged proteins can be introduced either by co-transfection or by separate transfection, followed by the subsequent mixing of lysates. HEK293T cells are commonly used due to their robust protein expression and high signal-to-noise ratios, and the assay can be applied to other cell lines with adequate protein expression [10].
Adaptations of this system have extended its application to animal models. Specific fragment combinations of firefly luciferase-Nfluc(1-475)/Cfluc(245-550), Nfluc(1-475)/ Cfluc(265-550), and Nfluc(1-475)/Cfluc(300-550)-produced detectable luminescence in subcutaneous 293T cell implants in mice upon addition of d-luciferin, recovering 0.01-4% of the activity observed with intact luciferase [11]. Delivery of one fragment (TAT-Cfluc(265-550)) to cells stably expressing the other fragment (Nfluc(1-475)) restored luminescence, demonstrating the assay's potential for monitoring molecular interactions in vivo [11].
In addition to the SLCA, other luciferase complementation systems have been developed, such as Renilla luciferase assay, that can detect protein-protein interactions by the reconstitution of the bioluminescent enzyme derived from the sea pansy (Renilla reniformis). This method has been commonly used to measure transient protein-protein interactions inside mammalian cells and plant protoplasts [12]. Unlike firefly luciferase, Renilla luciferase is a smaller enzyme (~37 kDa in size) that catalyzes the oxidation of coelenterazine (cell-permeable) to generate bioluminescence [13]. Beyond the luciferasebased systems, non-luciferase complementation methods include the split green fluorescent protein (GFP), which is often used to study stable protein-protein interactions or for subcellular imaging. In split GFP assays, a similar process occurs as two non-fluorescent proteins are fused to different parts of GFP, and GFP fluorescence is restored when the proteins come into close proximity by the protein-protein interactions. While the SLCA can detect real-time readouts of the bioluminescence signal that helps to track when interactions form and break, the inactive GFP fragments in the split GFP assay can irreversibly fold into a complete beta barrel upon interaction and produce a fluorescent signal, even if the proteins later separate [14,15]. Similarly to the split GFP system, the split alkaline phosphatase or AP assay produces an irreversible signal after protein-protein interactions reconstitute a functional AP that cleaves phosphate groups from the added substrates (such as the hydrophilic p-nitrophenyl phosphate or hydroquinone diphosphate) to generate the visible signal in the form of color change or fluorescence [16,17]. This is particularly useful when studying protein interactions that occur on the cell surface or even extracellularly. Specifically, the SLCA has several advantages over biochemical assays currently used to study protein-protein interactions, such as co-immunoprecipitation (co-IP), mass spectrometry (MS), and surface plasmon resonance (SPR). The SLCA can be used to study protein-protein interactions both in vivo and in vitro, which provides flexibility for the types of experiments being conducted. Additionally, when SLCA is performed in live cells, the two proteins of interest are fused to luciferase fragments within the cellular environment, allowing detection of low-affinity interactions that are often lost during lysis and immobilization processes required for Co-IP and MS [18,19]. This system is also well-suited for high-throughput screening that can rapidly test hundreds of thousands of small molecules or compounds to identify new therapeutic candidates that either activate or inhibit the protein-protein interaction.
The potential applications of the SLCA in virology are broad. Viruses such as HIV-1, Dengue, Eastern equine encephalitis virus (EEEV), Western equine encephalitis virus (WEEV), Epstein-Barr virus (EBV), and Rabies virus rely on protein-protein interactions for viral replication and spread throughout the infected organism. The SLCA has been used to quantify Rev-Rev interactions in HIV-1 both in vitro and in vivo to optimize the assay for high-throughput screening to identify inhibitors of the interaction to prevent the export of intron-containing viral RNA from the nucleus and thus viral replication [20]. Beyond HIV-1, the utility of split luciferase-based assays for discovering modulators of viral protein interactions has been demonstrated in hepatitis B virus (HBV), where a cell-based split luciferase complementation (SLC) assay was used to monitor core protein dimerization. Using this system, Arbidol and 20-deoxyingenol were identified from a 672-compound library as regulators of HBV core dimerization, which correspondingly altered HBV DNA replication in vitro [21]. Although this method differs from SLCA used in HIV-1 and other viruses, it provides proof-of-concept that split luciferase complementation can be adapted to screen for inhibitors or modulators of critical viral protein-protein interactions in diverse viral systems. Many HIV-1 protein-protein interactions remain untargeted, such as Tatcyclin T1 and nucleocapsid. Nucleocapsid interactions are necessary for viral transcription and replication. Together, these findings serve as a proof-of concept, demonstrating SLCA's potential as a highly adaptable and impactful tool for the discovery of antiviral agents and for mechanistic studies of viral proteins to advance therapeutic development across multiple viral pathogens.
In this article, we outline how the SLCA method can be used to target viral proteinprotein interactions across six globally widespread viruses that currently lack specific antiviral therapies. We present the successful application of the assay against HIV-1, where the quantification of the Rev-Rev interaction serves as a proof-of-concept for targeting additional interactions in the virus, as well as Dengue, Eastern/Western equine encephalitis, Epstein-Barr, and rabies viruses. By extending applications to other clinically relevant viruses, we discuss the potential of SLCA as a sensitive and adaptable platform, applicable to any virus with multimerizing protein-protein interactions, to conduct high-throughput screening to identify new antiviral drug candidates. To provide a framework for understanding the therapeutic potential of targeting viral protein-protein interactions, we include a brief overview of FDA-approved small molecules that modulate such interactions. These examples demonstrate that protein-protein interfaces are targets for drugs in both infectious diseases and other pathologies, providing clinical proof-of-concept for approaches like the SLCA to identify novel antiviral agents.
## 2. SLCA Applications in Viruses
## 2.1. Human Immunodeficiency Virus Type-1 (HIV-1)
HIV-1 is an ongoing pandemic, affecting nearly 40 million people worldwide [22]. While sub-Saharan Africa remains the most heavily impacted region, as it accounts for almost two-thirds of all global infections, HIV-1 persists in every region of the world, including high-income countries such as the United States, parts of Europe, and East Asia [23]. The persistence of HIV-1 as a global epidemic is attributable to its ability to integrate its genetic material within the host genome, primarily within CD4+ T cells and macrophages, and establish transcriptionally silent latent reservoirs that remain unaffected by current antiretroviral therapies (ART). As a result, the virus is never eradicated from an infected person, and treatments must be lifelong, as brief interruptions can lead to viral rebound and clinical disease [24]. Therefore, identifying new mechanisms to reverse HIV-1 latency or control viral replication is critical for advancing therapeutics and may ultimately impact the cure of infected individuals.
One approach to targeting viral protein-protein interactions is the SLCA that has been successfully used to study Rev-Rev interaction in HIV-1. Rev is a viral regulatory protein that multimerizes on the Rev Response Element (RRE) within the major intron to facilitate the nuclear export of the unspliced viral RNAs to the cytosol [20]. The Rev-Rev protein interaction occurs to form dimers and oligomers for efficient RRE binding [25]. If the Rev-Rev interaction is inhibited, then intron-containing viral RNA cannot be translated, and the virus cannot replicate. Application of the SLCA to quantify the Rev-Rev interaction both in vitro and in vivo demonstrated a high Z ′ factor of 0.85, suggesting that the cell-free assay can be further used for the high-throughput screening of inhibitory small molecules [20]. This study serves as a proof-of-concept to detect and quantify HIV-1 regulatory protein and structural proteins to potentially inhibit HIV from transcribing and replicating in cells.
Capsid-capsid interactions can also be studied using the SLCA. The HIV-1 capsid protein (approximately 231 amino acids) multimerizes to form a protective shell around the diploid viral RNA genome, protecting it from degradation and facilitating reverse transcription to convert the viral RNA into DNA [26]. After reverse transcription, the nucleocapsid and capsid structure maintains the integrity of the viral core and interacts with host proteins to deliver the viral DNA into the nucleus and integrate the duplex viral DNA into the host genome. Inhibition of capsid-capsid interaction disrupts its multimerization process, preventing proper assembly and disassembly of the viral core and thereby blocking viral replication [26].
The FDA-approved capsid inhibitor Lenacapavir (LEN) targets capsid multimerization to suppress viral replication (Figure 2). Although LEN, an FDA-approved HIV-1 capsid inhibitor, was not discovered using the SLCA, it provides clinical proof-of-concept that disrupting viral protein-protein interactions is a viable antiviral strategy. Its mechanism of action-interfering with capsid assembly (when the capsid proteins produce a conical structure in newly formed virions within the cell) and disassembly (when the capsid disassembles and releases the viral genome within the nucleus of the host cell)-demonstrates the therapeutic potential of identifying small molecules that interfere with viral proteinprotein interactions (Figure 3) [27]. SLCA-based high-throughput screening therefore provides a complementary strategy to identify small molecules capable of modulating critical viral interactions similar to those targeted by LEN. LEN is highly potent, with a mean half-maximal inhibitory concentration (IC-50) value of 200 pM for HIV-1 [28]. LEN binds at a conserved pocket formed between capsid subunits at the interface of the N-terminal domain of one monomer and the C-terminal domain of an adjacent subunit. The allosteric binding stabilizes the hexameric capsid lattice, inducing a hyperstable conformation that disrupts the normal metastability required for the HIV-1 life cycle [29]. During early infection, LEN prevents uncoating by hyper-stabilizing the capsid lattice, thereby blocking reverse transcription and nuclear import [29,30]. In later stages, it interferes with the proper assembly of Gag-derived capsid proteins, leading to the formation of morphologically defective virions [31]. These multifaceted actions collectively result in potent inhibition of viral replication at multiple steps of the life cycle. Although LEN is highly effective, it does not fully inhibit all stages of capsid/nucleocapsid assembly and resistance to the drug can develop over time due to specific mutations in capsid, notably M66I. The M66I substitution is a single amino acid change in the capsid that interferes with LEN binding, reducing its potency by more than 3000-fold [32]. This mutation alters the drug-binding pocket without disrupting the capsid's overall structure or ability to assemble, allowing the virus to remain infectious despite reduced drug susceptibility (Figure 4) [33]. This highlights the importance of discovering new antiretroviral drug candidates to inhibit capsid-capsid, nucleocapsid-nucleocapsid, or HIV-1 Tat-cyclin T1 interactions. A major barrier to curing HIV-1 is viral latency, allowing the virus to remain transcriptionally quiescent in infected immune cells and reactivate if treatment is discontinued. The viral regulatory protein Tat (Trans-Activator of Transcription) plays a central role in this reactivation by binding to the host transcription factor cyclin T1 to stimulate transcriptional elongation of the viral genome [34]. Disrupting this interaction would prevent RNA polymerase II from efficiently transcribing the provirus and prevent viral replication from spreading further. Despite its critical role in viral tran-scription, no FDA-approved therapies currently target Tat-cyclin T1 binding, and how the Tat-cyclin T1 interaction could be directly inhibited remains unexplored. As a first step in understanding this interaction and identifying potential therapeutic avenues, an SLCA could be used to quantitatively assess the stability and behavior of HIV-1 Tat-cyclin T1 interaction in vitro in order to then test known inhibitors that have been shown to inhibit Tat from binding to the trans-activation response or TAR RNA, such as Roscovitine and Gemcitabine, to validate the assay and conduct a high-throughput screen [35].
## 2.2. Dengue
Dengue virus is one of the most common mosquito-borne viruses in the world, yet there are no specific antiviral drugs that target this virus. When the virus infects human host cells, primarily macrophages and dendritic cells, it releases its positive-sense singlestranded RNA genome directly into the cytoplasm to function as an mRNA. This allows the host ribosomes to recognize the viral RNA and initiate translation into a single, nonfunctional polyprotein [36]. However, subsequent polypeptic enzymatic cleavage by viral NS3 protease, in conjunction with its cofactor NS2B, generates functionality, including structural proteins (capsid, prM/M, and E) and nonstructural proteins (NS1-NS5) [37]. Among the structural proteins, the capsid (C) protein is responsible for packaging the viral genome and forming a nucleocapsid core. The C protein is small (~100 amino acids) and has a high concentration of positively charged basic residues, which forms strong electrostatic interactions with the negatively charged phosphate backbone of the viral RNA [38]. For assembly, two C proteins first dimerize (C-C interaction), which serves as a building block for nucleocapsid formation. These C-C dimers then oligomerize around the viral RNA, resulting in a protective, multimeric nucleocapsid that is later enveloped by the prM/M and E proteins during virus maturation [3]. This physically shields the viral RNA from degradation by host nucleases and establishes a structural foundation for the production of infectious virions. If the C-C interaction can be quantified by the SLCA to subsequently identify potential inhibitors of that interaction, then the proteins cannot oligomerize to form the nucleocapsid and the viral RNA will be vulnerable to degradation, preventing the assembly of new virions to block viral replication [39,40].
## 2.3. Eastern Equine Encephalitis Virus (EEEV)
EEEV is a mosquito-borne alphavirus, mainly transmitted by Culiseta melanura, with no current medications that specifically target the virus. Upon entry in host cells-primarily dendritic cells, macrophages, and fibroblasts-the viral positive-sense single-stranded RNA genome is directly released into the cytoplasm, where it functions as mRNA for immediate translation [41]. The genome encodes structural proteins (C, E1, E2, E3, and 6K) that assemble into viral particles at the host cell's plasma membrane [41]. Compared to other Western Hemisphere alphaviruses, such as Venezuelan equine encephalitis virus (VEEV), the capsid protein of EEEV can interact with the host transport machinery to exit the nucleus by binding to Importin α/β and CRM1 (exportin 1) to suppress RNA polymerase II-mediated transcription that the host cell uses to make mRNA for antiviral proteins [42]. By blocking the nuclear core complex, EEEV capsid prevents the nuclear import of transcription factors (e.g., STAT1, NF-κB) responsible for activating antiviral genes and the export of host mR-NAs necessary for immune signaling and cellular function. Selective inhibitors of nuclear export (SINE) compounds such as KPT-185, KPT-335/verdinexor, and KPT-350 were found to bind to CRM-1 in VEEV, preventing the nuclear export of capsid [43]. This causes the capsid proteins to accumulate in the nucleus, which then prevents them from entering the cytoplasm to multimerize around the viral RNA to form the multimeric structure that protects the viral RNA (Figure 5) [9]. It is still unclear, however, whether SINE compounds directly inhibit the EEEV capsid-capsid interactions. This question can be addressed using the SLCA to validate previous findings and to identify more potent analogs of these compounds that can specifically target this protein-protein interaction.
## 2.4. Western Equine Encephalitis Virus (WEEV)
WEEV is another mosquito-borne alphavirus, generally less severe than EEEV, with no current medications that can target the virus. Similarly to EEEV, the virus is a positive-sense, single-stranded RNA virus. Its genome is released into the host cytoplasm where it directly acts as mRNA [44]. Viral entry depends on interactions between the envelope glycoproteins E1 and E2, which form heterodimers on the virion surface. For wild-type WEEV strains, these E2-E1 heterodimers create binding sites for host receptors, with PCDH10 binding in the cleft [45]. However, a single amino acid change in E2 (V265F) can cause a conformational change in the binding site, thereby broadening the range of cell types the virus can infect. In principle, inhibitors that target the E1-E2 interaction or the receptor interfaces could block viral entry into cells, and the SLCA system can be adapted to quantify the E2-E1 interactions to screen for new potential therapeutic compounds.
In addition to the E1 and E2 structural proteins, the genome also encodes capsid proteins that multimerize around the viral RNA in the cytoplasm to form nucleocapsids (Figure 5). Thieno [3,2-b] pyrrole-based inhibitors exhibit antiviral activity against neurotropic alphaviruses like WEEV in cultured cells [46]. The mechanisms by which the inhibitors block viral replication, however, are unclear at present. Based on the experimental approach with a replicon-based assay, it indicates that the drug interferes with the viral replication machinery rather than viral entry into cells [47]. Thus, the SLCA could be used to determine whether the thieno [3,2-b] pyrrole-based inhibitors can directly disrupt capsid-capsid interactions and to quantitatively assess their antiviral potency.
## 2.5. Epstein-Barr Virus (EBV)
EBV is a human herpesvirus that infects more than 90% of the world population [48]. The virus primarily targets B lymphocytes and oropharyngeal epithelial cells [48]. During infection, the virus binds to the CD21 receptor on B cells via its glycoprotein gp350/220 or enters epithelial cells through gH/gL-and gB-mediated endocytosis, releasing its viral double-stranded DNA genome into the nucleus where early transcription takes place [49]. The genome encodes structural proteins such as the major capsid protein BcLF1, which binds to other BcLF1 monomers to form the hexons and pentons of the icosahedral lattice (Figure 5). The capsid is further stabilized by triplexes that are formed by minor capsid proteins, BORF1 (Tri1), and BDLF1 (Tri2) that bridge BcLF1 units [50]. Together, these interactions assemble a stable nucleocapsid that protects the viral genome from degradation. If the BcLF1-BcLF1 or BcLF1-triplex interaction can be targeted to prevent capsid assembly, it could inhibit virion maturation, which could then prevent the production of infectious EBV particles [50]. While there are no clinically approved inhibitors targeting capsid assembly or viral DNA synthesis, spironolactone, a mineralocorticoid receptor antagonist, has been shown to inhibit the function of the EBV SM protein, which is critical for late lytic gene expression and capsid antigen production [51]. This demonstrates that small molecules can interfere with capsid formation and virion maturation. The SLCA can provide a platform to test whether spironolactone can also target the BcLF1-BcLF1 or BcLF1-triplex interactions and screen additional inhibitors of the interactions.
## 2.6. Rabies Virus
Rabies is a negative-sense single-stranded RNA virus primarily transmitted to humans through saliva from an infected animal with no specific antiviral treatments currently available, and death is invariable. When rabies virus infects host neurons, the viral RNA genome is released into the cytoplasm, where it serves as a template for transcription and replication by the viral RNA polymerase [52]. The genome encodes structural proteins including nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and the large polymerase (L) protein. The N protein (~450 amino acids) multimerizes around the viral RNA to form a ribonucleoprotein (RNP) complex that not only protects the viral genome from the host RNAses but also provides the stability and structure for the microtubule-based retrograde transport (Figure 5) [53]. After initial viral replication near the bite site, rabies virions are transported from the peripheral nerve endings to the neural cell body in the spinal cord and brain. The ribonucleoprotein or RNP complexes and virions hijack the host's microtubule network and engage dynein motor proteins to directionally travel toward the minus end of microtubules, which transports the viral genome to the neuronal nucleus-proximal region, where transcription, replication, and assembly occur [54]. Without this directional transport, the virus would remain trapped at distal nerve terminals and would not be able to invade the CNS. The N-N protein interaction is critical for the virus to spread, and the SLCA can be used to identify smallmolecule inhibitors that disrupt N oligomerization to destabilize the RNP complex and prevent viral replication and the assembly of infectious virions [55].
## 3. SLCA
To identify inhibitors of the specific protein-protein interaction for the above applications, the protein nucleic acid sequences are fused in-frame to complementary luciferase fragments. Homotypic interactions, where identical proteins interact (e.g., capsid-capsid or Rev-Rev), often produce strong and consistent bioluminescence [18]. However, heterotypic interactions, which involve two different proteins, may require optimization, as one protein is typically fused to the NLuc fragment and the other to the CLuc fragment, or vice versa [18]. The fusion orientation of target proteins to luciferase fragments can influence assay performance by affecting protein folding or the accessibility of interaction interfaces. In such cases, empirical testing may be required to identify optimal configurations. The optimal amino acid length for each protein in the assay is not fully established; however, based on prior studies, proteins of approximately 200-300 amino acids generally perform well in the SLCA.
The reconstitution of luciferase activity occurs only when the two proteins interact, providing a quantitative readout in relative light units (RLUs). The assay is first optimized and validated using positive and negative controls, with a Z ′ or Z factor of between 0.5 and 1.0, indicating sufficient robustness for high-throughput screening [56]. The presence of a wellcharacterized inhibitor produces a ≥5-fold reduction in RLU. It is always quite helpful to have an already-identified inhibitor to reduce RLU by at least 5-fold (termed an S/B value), which can improve the reliability of the assay during optimization (Figure 6). Once validated with acceptable Z ′ and S/B values, the system can be used to test small-molecule inhibitors to target the protein-protein interface, scaled to a 384-well format for high-throughput screening using an exceptionally complex library of compounds [57]. A reduction in RLU indicates a potential disruption of the protein-protein interaction, and the extent of inhibition can be quantified through dose-response analysis by testing different concentrations of the compounds.
Compounds identified as preliminary hits should be tested against the full-length firefly luciferase (FFLUC) to eliminate factors that can directly inhibit the luciferase enzyme. Verified hits can be further evaluated using biochemical assays, and confirmed hits can be optimized chemically to increase potency and selectivity. Only compounds that demonstrate specific and reproducible inhibition of the protein-protein interaction and functional antiviral activity advance to preclinical development. Subsequent cell-and animal-based assays are used to quantify antiviral activity and verify that the compounds are functional, prior to any clinical studies.
## 4. FDA-Approved Inhibitors
Several FDA-approved small molecules have established that protein-protein interactions are viable therapeutical targets, providing a strong rationale for developing novel antivirals using the SLCA. In virology, compounds such as Fostemsavir and Maraviroc exemplify this approach by inhibiting interactions between the HIV-1 envelope glycoprotein gp120 and host receptors CD4 and CCR5, thereby blocking viral entry into host cells [58]. LEN extends this concept by targeting the viral capsid-capsid interaction, demonstrating that the disruption of structural protein multimerization impedes capsid assembly and consequently inhibits viral replication [28]. LEN is now FDA-approved for both HIV-1 treatment and prophylaxis [32]. Beyond virology, the success of PPI-targeting agents in oncology and immunotherapy further underscores the translational potential of this approach (Table 1). Venetoclax, for example, disrupts BCL2-BCL2 interactions to induce apoptosis in malignant cells [59], whereas Motixafortide antagonizes the CXCL12-CXCR4 axis to mobilize hematopoietic stem cells for the treatment of multiple myeloma [60]. These FDA-approved compounds span diverse mechanisms, demonstrating that protein-protein interactions can be targeted at multiple stages, including viral entry, replication, assembly, and immune modulation. However, gaps remain regarding the type of diseases these drugs target (mostly cancer), potency, and resistance to the medication over time, highlighting the importance of continued research and utilizing the SLCA to identify new drugs developed for deadly infectious diseases that are currently not targeted.
## 5. Conclusions
Despite advances in antiviral research, many clinically important viruses such as HIV-1, Dengue, equine encephalitis viruses, Epstein-Barr virus, and rabies remain major global health threats. Although there are current antiretroviral therapies for HIV-1, there is no cure for those infected with HIV-1, mainly due to viral latency. Other viruses such as Dengue, equine encephalitis viruses, EBV, and rabies do not have specific antiviral therapies, which can lead to severe acute disease and persistent virologic presence, depending on the viral infection. These limitations highlight the importance of identifying novel therapeutic candidates by using known methods such as the SLCA to target viral processes at the molecular level. The SLCA provides quantitative results both cell-free and within cells, with high sensitivity and low background that is adaptable to studying a number of viral protein-protein interactions. The system is best suited for detecting intracellular proteinprotein interactions, and its applications to extracellular or membrane-bound targets has not yet been demonstrated in the literature. By demonstrating its applications across multiple viruses, including HIV-1 as a proof-of-concept, this article provides a framework for future research aimed at developing antiviral therapies for viruses that currently lack specific therapies. A poignant example of this is LEN, which inhibits both virus disassembly and assembly in cells, and is now FDA-approved for treatment and prophylaxis of HIV-1.
## 6. Future Perspective
With many millions of individuals worldwide affected by HIV-1, Dengue, eastern and western equine encephalitis, Epstein-Barr, and rabies viruses, the next frontier in antiviral research may lie in targeting viral multimeric structural protein interactions that drive viral replication and cause human disease. Protein-protein interactions, such as HIV Tat-cyclin T1, Dengue capsid dimers, and rabies nucleocapsid oligomers, are highly specific and can be targeted with small-molecule inhibitors. To identify such inhibitors, the SLCA or a related assay can be utilized to conduct high-throughput screening of hundreds of thousands or millions of small molecules. With emerging viruses on the rise, SLCA reliably and reproducibly quantifies protein-protein interactions at various stages within the viral cycle and should thus accelerate treatment and possibly cure of acute or chronic viral infections. Importantly, therapies developed through this approach could provide affordable treatment options for patients throughout the world, in low-and middle-income countries in South America, Africa, and Asia.
## References
1. Ulgheri, Bernardes, Lancellotti (2025) "Decoding Dengue: A Global Perspective, History, Role, and Challenges. Pathogens"
2. Byk, Gamarnik (2016) "Properties and Functions of the Dengue Virus Capsid Protein" *Annu. Rev. Virol*
3. Zhang, Tamilarasu, Hwang et al. (2000) "HIV-1 TAR RNA Enhances the Interaction between Tat and Cyclin T1" *J. Biol. Chem*
4. Zhang "Id protein-firefly luciferase N-fragment & firefly luciferase C-fragment-myod protein" *Molecular Imaging and Contrast Agent Database (MICAD*
5. Misawa, Kafi, Hattori et al. (2010) "Rapid and high-sensitivity cell-based assays of proteinprotein interactions using split click beetle luciferase complementation: An approach to the study of G-protein-coupled receptors" *Anal. Chem*
6. Lang, Li, Li (2019) "Analysis of Protein-Protein Interactions by Split Luciferase Complementation Assay" *Curr. Protoc. Toxicol*
7. Kawamura, Ozawa (2025) "Luciferase complementation for cellular assays beyond protein-protein interactions" *Anal. Sci*
8. Liang, Li (2022) "Split-Luciferase Complementation for Analysis of Virus-Host Protein Interactions" *Methods Mol. Biol*
9. Villalobos, Naik, Piwnica-Worms (2008) "Detection of protein-protein interactions in live cells and animals with split firefly luciferase protein fragment complementation" *Methods Mol. Biol*
10. Paulmurugan, Gambhir (2005) "Firefly Luciferase Enzyme Fragment Complementation for Imaging in Cells and Living Animals" *Anal. Chem*
11. Kato, Jones (2010) "The split luciferase complementation assay" *Methods Mol. Biol*
12. Woo, Howell, Von Arnim (2008) "Structure-function studies on the active site of the coelenterazine-dependent luciferase from Renilla" *Protein Sci*
13. Bignon, Gruet, Longhi et al. (2022) "Strengths and Caveats from a Multiparametric Analysis" *Int. J. Mol. Sci*
14. Magliery, Wilson, Pan et al. (2005) "Detecting Protein-Protein Interactions with a Green Fluorescent Protein Fragment Reassembly Trap: Scope and Mechanism" *J. Am. Chem. Soc*
15. Niu, Ye, Wang et al. (2019) "A review on emerging principles and strategies for colorimetric and fluorescent detection of alkaline phosphatase activity" *Anal. Chim. Acta*
16. Shen, Wang, Ning et al. (2023) "Ultrasensitive alkaline phosphatase activity assay based on controllable signal probe production coupled with the cathodic photoelectrochemical analysis" *Food Chem*
17. Azad, Tashakor, Hosseinkhani (2014) "Split-luciferase complementary assay: Applications, recent developments, and future perspectives" *Anal. Bioanal. Chem*
18. Chen, Yan, Qin et al. (2022) "Near-Infrared Luciferase Complementation Assay with Enhanced Bioluminescence for Studying Protein-Protein Interactions and Drug Evaluation Under Physiological Conditions" *Anal. Chem*
19. Hansen, Baris, Zhao et al. "Cell-based and cell-free firefly luciferase complementation assay to quantify Human Immunodeficiency Virus type 1 Rev-Rev interaction" *Virology*
20. Wei, Gan, Cui et al. (2018) "Identification of Compounds Targeting Hepatitis B Virus Core Protein Dimerization through a Split Luciferase Complementation Assay" *Antimicrob. Agents Chemother*
21. Moyo, Moyo, Murewanhema et al. (2023) "Key populations and Sub-Saharan Africa's HIV response" *Front. Public Health*
22. Mbonye, Karn "The cell biology of HIV-1 latency and rebound" *Retrovirology*
23. Daugherty, Booth, Jayaraman et al. (2010) "HIV Rev response element (RRE) directs assembly of the Rev homooligomer into discrete asymmetric complexes" *Proc. Natl. Acad. Sci*
24. Campbell, Hope (2015) "HIV-1 capsid: The multifaceted key player in HIV-1 infection" *Nat. Rev. Microbiol*
25. Canales, Tse, Schroeder et al. (2025) "Discovery of Lenacapavir: First-in-Class Twice-Yearly Capsid Inhibitor for HIV-1 Treatment and Pre-exposure Prophylaxis" *J. Med. Chem*
26. Neverette, Dumond, Mcmahon et al. (2024) "Playing the Long Game in the New Era of Antiretrovirals" *Clin. Pharmacol. Ther*
27. Li, Burdick, Siddiqui et al. "Lenacapavir disrupts HIV-1 core integrity while stabilizing the capsid lattice"
28. Eschbach, Puray-Chavez, Mohammed et al. "HIV-1 capsid stability and reverse transcription are finely balanced to minimize sensing of reverse transcription products via the cGAS-STING pathway"
29. Huang, Briganti, Annamalai et al. (2025) "The primary mechanism for highly potent inhibition of HIV-1 maturation by lenacapavir" *PLoS Pathog*
30. (2025) "DailyMed-SUNLENCA-Lenacapavir Sodium Tablet, Film Coated SUNLENCA-Lenacapavir Sodium Kit"
31. Briganti, Annamalai, Bester et al. "Structural and mechanistic bases for resistance of the M66I capsid variant to lenacapavir"
32. Wei, Garber, Fang et al. (1998) "A Novel CDK9-Associated C-Type Cyclin Interacts Directly with HIV-1 Tat and Mediates Its High-Affinity, Loop-Specific Binding to TAR RNA" *Cell*
33. Shin, Kim, Park et al. (2020) "Identification of novel compounds against Tat-mediated human immunodeficiency virus-1 transcription by high-throughput functional screening assay" *Biochem. Biophys. Res. Commun*
34. Diamond, Pierson (2015) "Molecular Insight into Dengue Virus Pathogenesis and Its Implications for Disease Control" *Cell*
35. Tay, Saw, Zhao et al. (2015) "The C-terminal 50 amino acid residues of dengue NS3 protein are important for NS3-NS5 interaction and viral replication" *J. Biol. Chem*
36. Jablunovsky, Jose (2024) "The Dynamic Landscape of Capsid Proteins and Viral RNA Interactions in Flavivirus Genome Packaging and Virus Assembly. Pathogens"
37. Figueira-Mansur, Aguilera, Stoque et al. (2019) "Mutations in the dimer interfaces of the dengue virus capsid protein affect structural stability and impair RNA-capsid interaction" *Sci. Rep*
38. Sangiambut, Promphet, Chaiyaloom et al. (1635) "Increased capsid oligomerization is deleterious to dengue virus particle production" *J. Gen. Virol*
39. Hasan, Dey, Singh et al. "The Structural Biology of Eastern Equine Encephalitis Virus, an Emerging Viral Threat. Pathogens 2021"
40. Carey, Bakovic, Callahan et al. (2019) "New World alphavirus protein interactomes from a therapeutic perspective" *Antivir. Res*
41. Lundberg, Pinkham, De La Fuente et al. (2016) "Selective Inhibitor of Nuclear Export (SINE) Compounds Alter New World Alphavirus Capsid Localization and Reduce Viral Replication in Mammalian Cells" *PLoS Negl. Trop. Dis*
42. Gauci, Wu, Rayner et al. (2010) "Identification of Western equine encephalitis virus structural proteins that confer protection after DNA vaccination" *Clin. Vaccine Immunol*
43. Ma, Cao, Ding et al. (2025) "Structural basis for the recognition of two different types of receptors by Western equine encephalitis virus" *Cell Rep*
44. Peng, Peltier, Larsen et al. (2009) "Identification of thieno[3,2-b]pyrrole derivatives as novel small molecule inhibitors of neurotropic alphaviruses" *J. Infect. Dis*
45. Delekta, Dobry, Sindac et al. (2014) "Novel Indole-2-Carboxamide Compounds Are Potent Broad-Spectrum Antivirals Active against Western Equine Encephalitis Virus" *J. Virol*
46. Chakravorty, Afzali, Kazemian (2022) "EBV-associated diseases: Current therapeutics and emerging technologies" *Front. Immunol*
47. Turk, Jiang, Chesnokova et al. (2006) "Antibodies to gp350/220 enhance the ability of Epstein-Barr virus to infect epithelial cells" *J. Virol*
48. Henson Brandon, Perkins Edward, Cothran Jonathan et al. (2009) "Self-Assembly of Epstein-Barr Virus Capsids" *J. Virol*
49. Verma, Thompson, Swaminathan (2016) "Spironolactone blocks Epstein-Barr virus production by inhibiting EBV SM protein function"
50. Kiflu (1774) "The Immune Escape Strategy of Rabies Virus and Its Pathogenicity Mechanisms" *Viruses*
51. Schoehn, Iseni, Mavrakis et al. (2001) "Structure of Recombinant Rabies Virus Nucleoprotein-RNA Complex and Identification of the Phosphoprotein Binding site" *J. Virol*
52. Nevers, Scrima, Glon et al. (2022) "Properties of rabies virus phosphoprotein and nucleoprotein biocondensates formed in vitro and in cellulo" *PLoS Pathog*
53. Zheng, Zhu, Zhu et al. (2025) "A Novel Protein NLRP12-119aa that Prevents Rhabdovirus Replication by Disrupting the RNP Complex Formation" *Adv. Sci*
54. Dobritsa, Kuok, Nguyen et al. (2013) "Development of a high-throughput cell-based assay for identification of IL-17 inhibitors" *J. Biomol. Screen*
55. Fernandez, Hassen-Khodja, Georget et al. "Measuring the subcellular compartmentalization of viral infections by protein complementation assay"
56. Orkin, Cahn, Castagna et al. (2022) "Opening the door on entry inhibitors in HIV: Redefining the use of entry inhibitors in heavily treatment experienced and treatment-limited individuals living with HIV" *HIV Med*
57. Uchida, Isobe, Asano et al. (2019) "Targeting BCL2 with venetoclax is a promising therapeutic strategy for "double-proteinexpression" lymphoma with MYC and BCL2 rearrangements" *Haematologica*
58. Rebolledo-Bustillo, Garcia-Gomez, Dávila et al. (2023) "Structural Basis of the Binding Mode of the Antineoplastic Compound Motixafortide (BL-8040) in the CXCR4 Chemokine Receptor"
59. "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=PMC12815380&blobtype=pdf | # CXCR3/CXCL10 Axis-Mediated T Cell Infiltration in the Lungs of Patients With HTLV-1-Associated Diseases: Implications for Subclinical Pulmonary Involvement
Kanako Tsuchimoto, Ayasa Mori, Masakazu Tanaka, Shiho Arishima, Daisuke Kodama, Mika Dozono, | Nozuma, Hiroshi Takashima, Ryuji Kubota
## Abstract
HTLV-1 is a retrovirus associated with adult T cell leukemia/lymphoma (ATL) and inflammatory diseases, including HTLV-1-associated myelopathy (HAM) and HTLV-1-associated bronchopneumonopathy (HAB). Although pulmonary complications are common in HTLV-1-associated diseases, the underlying mechanisms remain unclear. We compared HTLV-1 proviral load (PVL) and chemokine receptor expression in bronchoalveolar lavage (BAL) cells and peripheral blood mononuclear cells (PBMCs) from asymptomatic carriers (ACs) and patients with HAB, HAM, or ATL. T cell subsets were analyzed by flow cytometry, and the expression of CXCR3 and CXCL10 in lung tissue was assessed by immunohistochemistry. HTLV-1 proviral DNA was detectable in BAL cells not only from HAB and ATL cases with pulmonary involvement, but also from some ACs and HAM cases without clinical respiratory symptoms, suggesting subclinical pulmonary infiltration. BAL samples from HAB and ATL showed increased CD8 + T cell frequency. Both CD4+ and CD8 + T cells in BAL expressed higher CXCR3 than their PBMC counterparts, whereas CCR4 and CXCR5 were not elevated. CADM1 + CD4 + T cells in BAL also exhibited higher CXCR3 than in PBMCs, and CXCR3+ frequencies were similar between CADM1+ and CADM1-CD4 + T cells within BAL, indicating that enhanced CXCR3 expression was largely independent of infection status. Histology revealed CD3+ mononuclear cell infiltration in alveolar septa and peribronchiolar regions in HAB and HAM, consistent with T cell-mediated alveolitis and bronchiolitis, and CXCL10 expression was elevated in infiltrated lesions. Collectively, these findings implicate the CXCR3/ CXCL10 axis as a common pathway for pulmonary T cell recruitment in HTLV-1-associated diseases.
| IntroductionHuman T cell leukemia virus type 1 (HTLV-1) is a retrovirus that preferentially infects CD4 + T cells in humans, affecting an estimated 10-20 million individuals worldwide [1]. The virus is the causative agent of adult T cell leukemia (ATL), a hematological malignancy; HTLV-1-associated myelopathy (HAM), an inflammatory neurological disorder; and HTLV-1-associated uveitis (HAU), an inflammatory eye condition [2][3][4][5]. In addition to these diseases, HTLV-1 has been implicated in various inflammatory conditions affecting the lungs, joints, and salivary glands [6][7][8].
ATL patients frequently present with lung involvement, including direct infiltration by malignant cells as well asThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.
secondary infections caused by bacterial, fungal, or viral pathogens [9]. HAM patients exhibit inflammatory lung diseases characterized by pronounced inflammatory cell infiltration [10,11]. Among HTLV-1 carriers, infection has been associated with an increased risk of tuberculosis, and community-acquired pneumonia [12,13]. Interestingly, pulmonary involvement has also been reported in HTLV-1-infected individuals without ATL, HAM, or microbial infections. A radiological study showed a higher incidence of pulmonary abnormalities in HTLV-1 carriers compared to non-infected individuals [14]. Additionally, bronchoalveolar lavage (BAL) fluid analyses in these patients revealed marked lymphocytosis, suggesting a contribution of HTLV-1 infection to inflammatory lung diseases [6,15]. Collectively, these lung-related conditions are referred to as HTLV-1-associated bronchopneumonopathy (HAB) or HTLV-1-associated bronchoalveolar disorder (HABA), and have gained attention as clinically significant entities [6,16,17]. Recently, a significantly high rate of pulmonary involvement was reported among HTLV-1-infected individuals in an indigenous tribe in Australia, further underscoring the global importance of HTLV-1-associated lung diseases [18].
HAB is characterized by a spectrum of pulmonary manifestations, including bronchiolitis, alveolitis, diffuse panbronchiolitis, and interstitial pneumonia [19][20][21][22]. Several studies have elucidated specific features of HAB. Notably, the HTLV-1 proviral load (PVL) in peripheral blood mononuclear cells (PBMCs) correlates with the degree of bronchoalveolar lymphocytosis [10], and the expression of HTLV-1 mRNA is significantly upregulated in BAL cells compared to PBMCs [23]. Furthermore, patients with HTLV-1-associated lung disease exhibit elevated levels of soluble adhesion molecules and soluble interleukin-2 receptor α in BAL fluid, as well as increased mRNA expression of cytokines and chemokines in BAL cells [24][25][26]. The frequency of HTLV-1-specific CD8 + T cells in BAL cells has been reported to be 5.1 times higher than in PBMCs, suggesting a localized immune response to HTLV-1 within the lungs [11]. These findings strongly support the hypothesis that HTLV-1 infection induces pulmonary inflammation. However, the distinction between HAB and pulmonary involvement in HAM, as well as the mechanisms driving lymphocyte infiltration into the lungs in HTLV-1-infected individuals, remains incompletely understood.
In this study, we analyzed the HTLV-1 PVL in BAL cells and PBMCs from HTLV-1 carriers, HAB, HAM, and ATL patients. We further investigated the role of chemokines and chemokine receptors in the lungs of individuals with HTLV-1-associated lung diseases. Our findings demonstrate a preferential accumulation of CD8 + T cells in the lungs of HAB patients, as well as subclinical pulmonary involvement of HTLV-1-infected cells in asymptomatic HTLV-1 carriers (ACs). Moreover, we identified the CXCR3/CXCL10 axis as a critical pathway mediating T cell infiltration into the lungs, thereby providing new insights into the pathogenesis of HTLV-1-associated lung conditions.
## 2 | Materials and Methods
## 2.1 | Subjects
Peripheral blood samples and BAL were collected from three healthy controls (HC; volunteers), 30 ACs, 20 patients with HAB, 24 patients with HAM, and 29 patients with ATL. HTLV-1 infection was diagnosed by detecting anti-HTLV-1 antibodies in serum. HAM was diagnosed based on WHO criteria, while HAB was diagnosed by the presence of abnormal shadows on chest X-ray in the absence of other respiratory diseases [27]. Clinical characteristics of the patients are summarized in Table 1. PBMCs were isolated and cryopreserved in liquid nitrogen until use. BAL cells were washed and cryopreserved in liquid nitrogen. Paired BAL cells were obtained via bronchoscopy from three HCs, 9 ACs, 10 HAB patients, 8 HAM patients, and 9 ATL patients. This cohort included one HAM patient with pneumonia and five ACs with interstitial pneumonia; however, the remaining seven HAM patients and five ACs showed no clinical signs of lung disease and had normal chest X-ray findings. A subset of samples was used for HTLV-1 PVL quantification, flow cytometry, or both, depending on cell viability and availability. For immunohistochemistry, lung tissue specimens were obtained from 2 HAB patients, 4 HAM patients, and 1 ATL patient by autopsy or biopsy. Samples used for immunohistochemistry were distinct from those used for flow cytometry. The study was approved by the Ethics Committee of Kagoshima University and conducted in compliance with the Declaration of Helsinki.
## 2.2 | Quantification of HTLV-1 PVL
DNA was extracted from PBMCs and BAL cells and quantitative real-time PCR was performed using a TaqMan-based assay as previously described [28]. Briefly, 10 ng of DNA and a standard template were subjected to 45 cycles of PCR using a OneStep-Plus Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed in triplicate. The copy number of HTLV-1 tax and β-actin genes was calculated based on standard curves. The PVL was expressed as copies per 100 cells, calculated using the following formula: PVL (copies/ 100 cells) = (HTLV-1 tax copy number)/(β-actin copy number x 1/2) x 100.
## 2.3 | Flow Cytometry
After thawing the cryopreserved cells, cells were stained with fluorescence-conjugated antibodies at room temperature for 10 min. Fluorescence signals were acquired using a CytoFLEX flow cytometer and analyzed with CytExpert software (Beckman Coulter, Tokyo, Japan). Lymphocytes were initially gated based on forward and side scatter properties (Figure 2A). Single cells were then gated, followed by exclusion of dead cells using a viability dye. CD3 + T cells were gated, and subsequently divided into CD4+ and CD8 + T cell subsets. Expression of chemokine receptors (CXCR3, CCR4, and CXCR5) and CADM1 (a specific marker of HTLV-1-infected cells) was assessed within these subsets [29]. Samples with at least 100 cells positive for each marker were analyzed. A list of antibodies used is provided in Supplementary Table 1.
## 2.4 | Immunohistochemistry
Lung tissues fixed in formalin or paraformaldehyde were paraffin-embedded and sectioned at 6 μm, followed by hematoxylin and eosin (H&E) staining. After deparaffinization and peroxidase blocking, antigen retrieval was performed. Tissue sections were then blocked with 5% normal goat serum and incubated overnight at 4°C with primary antibodies against lymphocyte surface markers (CD3, CD4, CD8, CD14, CD19, CD56) as well as CXCR3 and its ligands (CXCL9, CXCL10, CXCL11) [30]. A polymer-based secondary antibody (Nichirei Biosciences, Tokyo, Japan) was applied for 30 min at room temperature. Chromogenic detection was performed using 3,3′diaminobenzidine (DAB; Agilent Technologies, Santa Clara, CA, USA) to visualize chemokines, yielding a brown precipitate, and 3-amino-9-ethylcarbazole (AEC; Vector Laboratories, Newark, CA, USA) to detect CXCR3 and cell surface markers, producing a red coloration. Sections were evaluated for lymphocytic infiltration, lymphocytic markers, and chemokine expression. Staining intensity was scored as none (-), weak (±), mild (+), moderate (++), or strong (+++). Antibodies used are listed in Supplementary Table 2.
## 2.5 | Statistics
Multiple group comparisons were performed using the Tukey-Kramer method following one-way ANOVA to account for potential variability among groups. For comparisons between two groups, a two-tailed paired Student's t-test was employed.
To assess relationships between variables, Spearman's rank correlation test was utilized. A p-value of less than 0.05 was considered statistically significant.
## 3 | Results
## 3.1 | Presence of HTLV-1 in BAL From ACs and Increased PVL in BAL From HAM Patients
We compared the PVL in PBMCs among ACs, HAB, HAM, and ATL patients. As expected, ATL patients exhibited the highest PVL levels, followed by HAM, HAB, and ACs (Figure 1A). Notably, HTLV-1 provirus was detectable in BAL samples from ACs, with PVL levels comparable to those observed in HAB patients (Figure 1B). This finding suggests that HTLV-1-infected cells can infiltrate the lungs even in the absence of clinical symptoms. In HAM patients, PVL levels in BAL were the highest among all groups, despite 7 of 8 patients showing no clinical signs of pulmonary disease and having normal chest radiographs. The ratio of PVL in BAL to that in PBMCs was highest in ACs, followed by HAM, HAB, and ATL patients (Figure 1C). These results collectively indicate that pulmonary infiltration of HTLV-1-infected cells occurs even in asymptomatic individuals and is more pronounced in patients with HAM.
## 3.2 | Decreased CD8 + T Cells in PBMCs and Increased in BAL in HAB Patients
To characterize the phenotypic features of lung-infiltrating cells, we conducted flow cytometric analysis of BAL samples. As shown in Figure 2A, the gating strategy allowed for the identification of CD4 + T cells, CD8 + T cells, and CADM1 + CD4+ cells (HTLV-1-infected cells). Compared to HCs, the frequency of CD8 + T cells among PBMC-derived T cells was significantly lower in HAB and ATL patients (Figure 2B). In BAL cells, the frequency of CD8 + T cells in HAB patients was relatively high compared with those in AC, HAM, and ATL patients, although the difference was not statistically significant (Figure 2C). The BAL-to-PBMC ratio of CD8 + T cell frequency was highest in HAB patients (Figure 2D). Additionally, a significant increase in the frequency of CD8 + T cells was observed in BAL compared to PBMCs in HAB patients, with 9 out of 10 HAB patients exhibiting elevated frequencies (Figure 2E). In contrast, frequency of CADM1+ cells in CD4 + T cells was not different between BAL cells and PBMCs in all groups (Figure 2F), and the BAL-to-PBMC ratios of CADM1 + CD4+ cells were comparable across all groups (Figure 2G). These results suggest a depletion of CD8 + T cells in the blood of HAB patients, with a concurrent accumulation of these cells in the lungs.
## 3.3 | Increased CXCR3+ Cells in BAL Across All Groups
To investigate the chemokine molecules involved in T cell infiltration into the lungs, we analyzed the expression of chemokine receptors on BAL cells as shown in Figure 2A. We focused on three chemokine receptors: CXCR3, CCR4, and CXCR5. CXCR3, a proinflammatory Th1 marker, plays a critical role in T cell migration to the central nervous system (CNS) in HAM patients [31]. CCR4 is highly expressed on Th2-type cells and HTLV-1-infected cells, and associated with organ infiltration, including the lungs and skin [32]. CXCR5 is primarily expressed on lymph node-residing cells but can also be detected on inflammatory cells, such as those observed in rheumatoid arthritis [33]. Flow cytometric analysis revealed a significant increase in the frequency of CXCR3+ cells in CD4+ and CD8 + T cells within BAL compared to PBMCs across all patient groups, including ACs, HAB, HAM, and ATL patients (Figure 3A). Interestingly, a similar trend was observed in HCs; however, statistical analysis could not be performed due to the limited number of samples. Further analysis comparing CXCR3 expression between CADM1-and CADM1 + CD3 + CD4+ cells showed no significant differences in either PBMCs or BAL cells (Figure 3B). These findings suggest that CXCR3 is a key mediator of T cell migration into the lungs, irrespective of HTLV-1 infection status. In contrast, CCR4 expression was markedly higher in CD3 + CD4 + CADM1+ cells compared to their CADM1-counterparts in PBMCs across all groups, with levels exceeding 74.7% (Supplementary Figure 1B). This finding is consistent with the established role of CCR4 as a marker preferentially expressed on HTLV-1-infected cells [34]. However, CCR4 expression in CD3 + CD4 + CADM1+ cells from HAB and ATL patients was reduced in BAL (48.8 ± 18.5% and 77.5 ± 22.0%, respectively; mean ± SD) compared to PBMCs (74.7 ± 21.1% and 92.6 ± 11.5%, respectively) (Supplementary Figure 1A). Similarly, CCR4 expression in CD3 + CD8+ cells was lower in BAL (7.1 ± 4.7%) than in PBMCs (14.8 ± 10.6%) (Supplementary Figure 1A). Regarding CXCR5, its expression in CD3 + CD4+ cells was significantly lower in BAL than in PBMCs among ACs, HAB, and HAM patients, whereas no such trend was observed in CD3 + CD8+ cells (Supplementary Figure 2A). Among CD3 + CD4 + CADM1+ cells in BAL, CXCR5 expression was slightly higher than in CADM1-cells, although the difference was modest (Supplementary Figure 2B).
## 3.4 | Correlation Analysis of CXCR3 Expression, CADM1+ Cells, and HTLV-1 PVL
To investigate the relationships among frequency of CXCR3+ cells, CADM1+ cells, and HTLV-1 PVL, we performed correlation analyses using PBMC and BAL samples from all subjects. PVL and the frequency of CADM1+ cells in CD3 + CD4+ cell population were significantly correlated between PBMCs and BAL (Figure 4A,B). Furthermore, the frequency of CADM1+ cells was positively correlated with PVL in both PBMCs (Figure 4C) and BAL (Figure 4D). In contrast, the frequency of CXCR3+ cells within CD3 + CD4+ cell population show a negative correlation with PVL in PBMCs (Figure 4E) but not in BAL (Figure 4F). These findings support the notion that CADM1 preferentially expressed on HTLV-1-infected cells and that the burden of HTLV-1-infected cells in the lung reflects that in the peripheral blood. We also examined associations between PVL and immune composition in BAL, including CD3+ %, CD4 + /CD3+ %, CD8 + /CD3+ %, and the CD4/CD8 ratio. None of these relationships reached statistical significance (Supplementary Figure 3).
## 3.5 | Chemokine/Chemokine Receptor Expression in Lung Tissues
H&E staining revealed predominant infiltration of mononuclear cells in alveolar septa and peribronchiolar regions, with minimal or negligible presence of polymorphonuclear cells in two HAB, four HAM, and one ATL patient. Representative images are shown in Figure 5A. The extent of cellular infiltration was greater in HAM patients compared to those with HAB (Figure 5A and Table 2). The ATL patient (#9523) demonstrated extensive destruction of alveolar architecture, accompanied by fibrosis, and massive infiltration of ATL cells. CXCR3-positive lymphocytes were detected in alveolar septa and peribronchiolar regions across all groups, with enhanced infiltration in HAM (Figure 5B and Table 2). To investigate ligands associated with the CXCR3 receptor, we analyzed the expression of chemokines, including CXCL9, CXCL10, and CXCL11, in lung tissue samples (Figure 5C). Preliminary experiments revealed (G) BAL-to-PBMC ratios of CADM1+ cells within CD3 + CD4+ cell populations were consistent across groups. Ratios were calculated as the percentage of CADM1+ cells among CD3 + CD4 + BAL-derived T cells divided by the corresponding percentage in PBMCs. Statistical analyses were conducted using the Tukey-Kramer method, with no significant differences observed. NS: not significant.
no detectable staining for CXCL11 under various experimental conditions; therefore, subsequent analyses focused on CXCL9 and CXCL10. CXCL9 expression in the lungs was modest across the three patient groups. In contrast, immunohistochemical staining for CXCL10 revealed expression in relatively large cells suggestive of alveolar macrophages, as well as small round mononuclear cells located within the alveolar septa and spaces. Notable clustering of CXCL10-expressing cells was observed in both HAB and HAM patients (#9507 and #8838), although the expression was less pronounced in HAB patient (Figure 5C). In the ATL patient (#9523), CXCL10-positive cells exhibited a scattered distribution across the tissue. Table 2 summarizes the immunohistochemical findings for CXCR3 and its ligands and the degree of cellular infiltration. CXCL10 was consistently expressed across all samples, while CXCL9 expression was either undetectable or weak. Notably, strong CXCL10 expression was identified in two HAM patients (#8838 and #9754) and one ATL patient (#9523). Among these three patients, two exhibited moderate to massive cellular infiltration (#8838 and #9523). These findings indicate that CXCL10 is preferentially expressed by infiltrating cells within the alveolar septa and spaces. Furthermore, the CXCR3/CXCL10 axis appears to play a pivotal role in T cell infiltration into the lungs of patients with HTLV-1-associated lung pathologies.
## 3.6 | Lymphocyte Surface Marker Profiles in Pulmonary Infiltrates
Immunohistochemical analysis revealed that CD3+ lymphocytes were the predominant infiltrating population across all patient groups, with CD4+ and CD8+ subsets present at comparable levels (Figure 6). CD14+ cells, indicative of alveolar macrophages, were observed in both alveolar spaces and septa across all groups, with increased abundance in HAM patients. CD19+ cell (B cell) clusters were localized to the alveolar septa in HAB and HAM patients, whereas CD56+ cells (NK cells) were rarely detected in any group. These findings indicate that CD3 + T cells are the principal infiltrating population in the lungs of HAB and HAM patients, consistent with T cellmediated alveolitis and bronchiolitis. The immunophenotypic profiles of infiltrating cells were comparable between HAB and HAM patients, except for the greater extent of infiltration observed in HAM patients.
## 4 | Discussion
Our study demonstrated that the PVL in BAL cells was highest in HAM patients, followed by ATL, HAB, and AC groups (Figure 1B). Flow cytometric analysis revealed a preferential accumulation of CD8 + T cells in BAL from HAB patients (Figure 2E). Furthermore, CD4+ and CD8 + T cells in BAL exhibited elevated CXCR3 expression across all patient groups compared to their PBMC counterparts (Figure 3A). In contrast, CCR4 expression in BAL was reduced in CD8 + T cells and HTLV-1-infected cells from HAB patients (Supplementary Figure 1), while CXCR5 expression was significantly decreased in BAL-derived CD4 + T cells from ACs, HAB, and HAM patients (Supplementary Figure 2). Histological examination demonstrated infiltration of CD3+ mononuclear cells within the alveolar septa and peribronchiolar regions in both HAB and HAM patients, consistent with T cell-mediated alveolitis and bronchiolitis (Figures 5A and6). Immunohistochemistry further confirmed increased CXCR3 expression in infiltrating lymphocytes (Figure 5B and Table 2). CXCL10, a ligand of CXCR3, was predominantly expressed by alveolar macrophages in patients with HAB, HAM, and ATL, whereas CXCL9 and CXCL11 were scarcely detectable by staining (Figure 5C and Table 2).
In our cohort, CXCR3 expression was consistently higher in BAL cells than in PBMCs across all clinical groups (Figure 3A). Notably, CADM1 + CD4 + T cells in BAL exhibited levels of CXCR3 expression comparable to those of CADM1 -CD4 + T cells (Figure 3B), indicating that this elevation was largely independent of HTLV-1 infection status. Among CXCR3 ligands, CXCL10 was predominantly expressed in lung tissues (Figure 5C). Taken together, these findings indicate that the CXCR3/CXCL10 axis plays a pivotal role in promoting T cell migration into the lungs in HTLV-1-associated pulmonary diseases. A previous study demonstrating high CXCL10 levels in the BAL of patients with HTLV-1-related diffuse panbronchiolitis further supports the pivotal role of CXCL10 in lung inflammation [35]. Consistent with this interpretation, similar CXCR3/CXCL10-dependent T cell recruitment has been reported in other pulmonary inflammatory diseases, including idiopathic pulmonary fibrosis and hypersensitivity pneumonitis [36,37]. Collectively, these findings, together with our data, indicate that the CXCR3/CXCL10 axis functions as a common mediator of T cell infiltration into the lungs beyond HTLV-1-associated disease. While CXCL10 expression could not be assessed in lung tissue from ACs, the consistent presence of CXCR3+ cells across all groups underscores its potential relevance, though further studies are required to confirm its role in ACs. The association between lung-infiltrating HTLV-1-infected cells and CXCR3/CXCL10 expression remains an important question. However, reliable detection of HTLV-1-infected cells in formalin-fixed paraffin-embedded lung tissue-via immunostaining for viral proteins or CADM1-has not been achieved, including in our own attempts, precluding a direct spatial assessment of this relationship.
In addition to the CXCR3-dependent recruitment of T cells into the lungs, CADM1+ infected CD4 + T cells were not enriched in BAL compared with PBMCs (Figure 2F), whereas CD8 + T cells were preferentially increased in BAL from HAB patients (Figure 2E). Furthermore, the proportion of CXCR3 + CD4 + T cells in PBMCs showed an inverse correlation with PVL (Figure 4E). These findings suggest that lung infiltrates consist predominantly of CD8 + T cells and non-infected CD4 + T cells, rather than infected CD4 + T cells. We therefore suggest that chronic HTLV-1 infection promotes CD8 + T cell migration into the lungs by intensifying CXCL10/CXCR3-driven recruitment through inflammatory processes, rather than by imparting a distinct lung-homing phenotype to infected cells. Mechanistically, HTLV-1-infected CD4 + T cells and HTLV-1-specific CTLs may produce interferon-γ that induces CXCL10 in lungresident cells such as alveolar macrophages, thereby creating a chemokine gradient that attracts CXCR3 + Th1-type CD4+ and CD8 + T cells irrespective of infection status. In contrast, CCR4, which is preferentially expressed on Th2 and regulatory CD4 + T cells and enriched on infected cells in PBMCs, is relatively underrepresented in BAL (Supplementary Figure 1A). This Th1-skewed, CXCR3-dominant pattern helps reconcile the CCR4-biased phenotype of circulating infected cells with the CXCR3-high T cell infiltrates observed in the lungs.
The CXCR3/CXCL10 axis may also operate beyond the lung.
In HAM, cerebrospinal fluid contains elevated CXCL10 levels, and infiltrating cells display high CXCR3 expression [31]. Similarly, in multiple sclerosis, this axis drives T cell accumulation in the CNS [38,39]. These parallels suggest a shared mechanism of T cell migration across pulmonary and neurological inflammatory conditions. Nonetheless, direct causality remains to be established. Functional experiments-such as in vivo blockade of CXCR3 or CXCL10-are required to clarify their roles in T cell migration and disease progression. Therapeutically, targeting the CXCR3/CXCL10 pathway (e.g., with small-molecule inhibitors or monoclonal antibodies) may offer new strategies for HTLV-1-associated pulmonary diseases and other inflammatory disorders, provided interventions selectively modulate this pathway without compromising immunity. Regarding CCR4, the proportion of CCR4+ cells within the CADM1 + T cell compartment was lower in BAL than in PBMCs (Supplementary Figure 1A). This observation may reflect one or more of the following: preferential pulmonary recruitment of CCR4-cells, in situ downregulation of CCR4, or preferential homing of CCR4+ cells to non-pulmonary sites. Discriminating among these possibilities will require future studies employing matched tissue sampling and longitudinal analysis. CCR4 is typically expressed on Th2-type CD4 + T cells and regulatory T cells, and is preferentially upregulated on HTLV-1-infected cells, as shown in PBMCs (Supplementary Figure 1B). Accordingly, interpretation of CCR4 expression in the context of HTLV-1 infection warrants caution, as CCR4 may reflect the trafficking of both infected cells and Th2-type subsets. In contrast, our data indicate a CXCR3-biased axis in the lung, consistent with Th1-skewed recruitment. We therefore propose that disease context and local chemokine landscapes differentially shape the selection of CCR4+ versus CXCR3 + T cell subsets.
HTLV-1 proviral DNA (PVL) was detected in BAL cells from AC, HAB, HAM, and ATL patients (Figure 1B). PVL in BAL from ACs was comparable to that in HAB, indicating that infected cells infiltrate the lung even in asymptomatic carriers. This is consistent with reports of subclinical pulmonary involvement with bronchoalveolar lymphocytosis in a subset of ACs [10] Across all groups, however, the BAL-to-PBMC PVL ratio remained < 1 (Figure 1C), suggesting that a substantial fraction of lung-infiltrating cells comprises non-infected immune or inflammatory cells. Among the groups, ACs showed the highest BAL-to-PBMC PVL ratio, potentially reflecting limited bystander infiltration or an early stage of immune recruitment. Notably, HAM patients-despite lacking clinical pulmonary symptoms or abnormal chest radiographs in 7 of 8 cases-had the highest BAL PVL (Figure 1B), consistent with active recruitment of infected cells and possible subclinical pulmonary involvement. The discrepancy between high PVL in BAL and the absence of overt disease underscores the need to investigate localized immune responses and their contribution to disease progression. Collectively, these findings support subclinical pulmonary involvement in both ACs and HAM, albeit via distinct mechanisms: early, low-grade activation in ACs versus more pronounced inflammatory responses involving infected cells in HAM. Further work is needed to resolve the cellular composition of BAL and the molecular pathways driving lung infiltration in these settings.
Flow cytometry revealed a lower frequency of CD8 + T cells in BAL from HTLV-1-infected groups compared with healthy controls (Figure 2B). Nevertheless, the BAL-to-PBMC ratio of CD8 + T cell frequency was highest in HAB (Figure 2D), and CD8 + T cells were significantly enriched in BAL versus PBMCs in HAB but not in HAM (Figure 2E). In line with prior work in HAM showing alveolar lymphocytosis with CD8 predominance in many cases [15], our data demonstrated elevated CD8+ frequencies in BAL relative to PBMCs in 9 of 10 HAB patients (Figure 2E). Studies in HAM indicate that HTLV-1-specific CD8+ cytotoxic T lymphocytes (CTLs) frequently migrate into the CNS, induce apoptosis in adjacent cells, and contribute to spinal cord inflammation [40]. Consistently, our previous study showed a 5.1-fold enrichment of HTLV-1-specific CD8+ CTLs in the lung compared with PBMCs [11]. The presence of infected cells, predominant CD8+ infiltration, and accumulation of virus-specific CTLs together support HTLV-1-specific inflammation in the lungs of individuals with HTLV-1-associated pulmonary conditions. Further studies are required to elucidate the mechanisms that drive HTLV-1-specific lung inflammation.
Immunohistochemistry revealed prominent infiltration of CD3+ mononuclear cells within alveolar septa and peribronchiolar regions in both HAB and HAM, with minimal polymorphonuclear involvement-findings consistent with T cell-mediated alveolitis and bronchiolitis (Figures 5A and6) [6]. Pulmonary immune profiles were broadly similar between the two groups, with comparable proportions of CD4+ and CD8+ lymphocytes, whereas PVL in both PBMCs and BAL cells was higher in HAM. Notably, HAB showed greater accumulation of CD8+ cells in BAL than in PBMCs, a pattern not observed in HAM. Together with reports of increased HTLV-1-specific CD8 + T cells in BAL from HAB patients [11], these data suggest virus-specific immune responses in HAB and highlight the need to determine whether pulmonary pathology in HAM converges with or diverges from that in HAB.
In conclusion, HTLV-1-associated pulmonary diseases represent a spectrum of conditions ranging from subclinical inflammation in ACs and HAM patients to overt inflammatory pulmonary diseases in HAB patients. This study highlights the role of the CXCR3/ CXCL10 axis in mediating lymphocyte accumulation in the lungs and provide new insight into the immunopathogenesis of HTLV-1-associated pulmonary diseases.
## References
1. Gessain, Cassar (2012) "Epidemiological Aspects and World Distribution of HTLV-1 Infection" *Frontiers in Microbiology*
2. Uchiyama, Yodoi, Sagawa et al. (1977) "Adult T Cell Leukemia: Clinical and Hematologic Features of 16 Cases" *Blood*
3. Gessain, Vernant, Maurs (1985) "Antibodies to Human T-lymphotropic Virus Type-I in Patients With Tropical Spastic Paraparesis" *Lancet*
4. Osame, Usuku, Izumo (1986) "HTLV-I Associated Myelopathy, a New Clinical Entity" *Lancet*
5. Nakao, Ohba, Matsumoto (1989) "Noninfectious Anterior Uveitis in Patients Infected With Human T-Lymphotropic Virus Type I" *Japanese Journal of Ophthalmology*
6. Sugimoto, Nakashima, Watanabe (1987) "T-Lymphocyte Alveolitis in HTLV-I-Associated Myelopathy" *Lancet*
7. Vernant, Buisson, Magdeleine (1988) "T-Lymphocyte Alveolitis, Tropical Spastic Paresis, and Sjogren Syndrome" *Lancet*
8. Nishioka, Maruyama, Sato et al. (1989) "Chronic Inflammatory Arthropathy Associated With HTLV-I" *Lancet*
9. Yoshioka, Yamaguchi, Yoshinaga et al. (1985) "Pulmonary Complications in Patients With Adult T Cell Leukemia" *Cancer*
10. Mori (2005) "Bronchoalveolar Lymphocytosis Correlates With Human T Lymphotropic Virus Type I (HTLV-I) Proviral DNA Load in HTLV-I Carriers" *Thorax*
11. Kawabata, Higashimoto, Takashima et al. (2012) "Human T-Lymphotropic Virus Type I (HTLV-I)-Specific CD8+ Cells Accumulate in the Lungs of Patients Infected With HTLV-I With Pulmonary Involvement" *Journal of Medical Virology*
12. Marinho, Galvão-Castro, Rodrigues et al. (2005) "Increased Risk of Tuberculosis With Human T-Lymphotropic Virus-1 Infection: A Case-Control Study" *JAIDS Journal of Acquired Immune Deficiency Syndromes*
13. Atsumi, Yara, Higa (2009) "Influence of Human T Lymphotropic Virus Type I Infection on the Etiology of Communityacquired Pneumonia" *Internal Medicine*
14. Okada, Ando, Yoshitake (2006) "Pulmonary CT Findings in 320 Carriers of Human T-Lymphotropic Virus Type 1" *Radiology*
15. Couderc, Caubarrere, Venet (1988) "Bronchoalveolar Lymphocytosis in Patients With Tropical Spastic Paraparesis Associated With Human T Cell Lymphotropic Virus Type 1 (HTLV-1). Clinical, Immunologic, and Cytologic Studies" *Annals of Internal Medicine*
16. Maruyama, Mori, Kawabata et al. (1992) "Bronchopneumonopathy in HTLV-1 Associated Myelopathy (HAM) and Non-HAM HTLV-1 Carriers]" *Nihon Kyōbu Shikkan Gakkai zasshi*
17. Kimura "HABA (HTLV-I Associated Bronchiolo-alveolar Dis"
18. (1992) "Nihon Kyōbu Shikkan Gakkai zasshi"
19. Einsiedel, Pham, Wilson "Human T-Lymphotropic Virus Type 1c Subtype Proviral Loads, Chronic Lung Disease and Survival in a Prospective Cohort of Indigenous Australians" *PLoS Neglected Tropical Diseases*
20. Setoguchi, Takahashi, Nukiwa et al. (1991) "Detection of Human T Cell Lymphotropic Virus Type I-related Antibodies in Patients With Lymphocytic Interstitial Pneumonia" *American Review of Respiratory Disease*
21. Kikuchi, Saijo, Sakai (1996) "Human T Cell Lymphotropic Virus Type I (HTLV-I) Carrier With Clinical Manifestations Characteristic of Diffuse Panbronchiolitis" *Internal Medicine*
22. Sugimoto, Kitaichi, Ikeda et al. (1998) "Chronic Bronchioloalveolitis Associated With Human T cell Lymphotrophic Virus Type I Infection" *Current Opinion in Pulmonary Medicine*
23. Matsuyama, Kawabata, Mizoguchi et al. (2003) "Influence of Human T Lymphotrophic Virus Type I on Cryptogenic Fibrosing Alveolitis -HTLV-I Associated Fibrosing Alveolitis: Proposal of a New Clinical Entity" *Clinical and Experimental Immunology*
24. Higashiyama, Katamine, Kohno (1994) "Expression of Human T Lymphotropic Virus Type 1 (HTLV-1) Tax/Rex Gene in Fresh Bronchoalveolar Lavage Cells of HTLV-1-Infected Individuals" *Clinical and Experimental Immunology*
25. Sugimoto, Nakashima, Matsumoto et al. (1989) "Pulmonary Involvement in Patients With HTLV-I-Associated Myelopathy: Increased Soluble IL-2 Receptors in Bronchoalveolar Lavage Fluid" *American Review of Respiratory Disease*
26. Seki, Kadota, Higashiyama (1999) "Elevated Levels of Beta-Chemokines in Bronchoalveolar Lavage Fluid (BALF) of Individuals Infected With Human T Lymphotropic Virus Type-1 (HTLV-1)" *Clinical and Experimental Immunology*
27. Yamazato, Miyazato, Kawakami et al. (2003) "High Expression of p40(tax) and Pro-Inflammatory Cytokines and Chemokines in the Lungs of Human T-Lymphotropic Virus Type 1-Related Bronchopulmonary Disorders" *Chest*
28. Osame (1990) "Review of WHO Kagoshima Meeting and Diagnostic Guidelines for HAM/TSP"
29. Nagai, Usuku, Matsumoto (1998) "Analysis of HTLV-I Proviral Load in 202 HAM/TSP Patients and 243 Asymptomatic HTLV-I Carriers: High Proviral Load Strongly Predisposes to HAM/TSP" *Journal of Neurovirology*
30. Sasaki (2004) "Overexpression of a Cell Adhesion Molecule, TSLC1, as a Possible Molecular Marker for Acute-Type Adult T Cell Leukemia" *Blood*
31. Rubinstein, Kudryavtsev, Arsentieva et al. (2024) "CXCR3-Expressing T Cells in Infections and Autoimmunity" *Frontiers in Bioscience-Landmark*
32. Ando, Sato, Tomaru (2013) "Positive Feedback Loop via Astrocytes Causes Chronic Inflammation in Virus-Associated Myelopathy" *Brain*
33. Yoshie (2021) "CCR4 as a Therapeutic Target for Cancer Immunotherapy" *Cancers*
34. Moschovakis, Bubke, Friedrichsen et al. (2017) "T Cell Specific Cxcr5 Deficiency Prevents Rheumatoid Arthritis"
35. Yoshie, Fujisawa, Nakayama (2002) "Frequent Expression of CCR4 in Adult T Cell Leukemia and Human T Cell Leukemia Virus Type 1-Transformed T Cells" *Blood*
36. Yamamoto, Matsuyama, Oonakahara (2004) "Influence of Human T Lymphotrophic Virus Type I on Diffuse Pan-Bronchiolitis" *Clinical and Experimental Immunology*
37. Jiang, Liang, Hodge (2004) "Regulation of Pulmonary Fibrosis by Chemokine Receptor CXCR3" *Journal of Clinical Investigation*
38. Agostini, Calabrese, Poletti (2005) "CXCR3/CXCL10 Interactions in the Development of Hypersensitivity Pneumonitis" *Respiratory Research*
39. Balashov, Rottman, Weiner et al. (1999) "CCR5(+) and CXCR3(+) T Cells are Increased in Multiple Sclerosis and Their Ligands MIP-1alpha and IP-10 are Expressed in Demyelinating Brain Lesions" *Proceedings of the National Academy of Sciences*
40. Sørensen, Trebst, Kivisäkk (2002) "Multiple Sclerosis: A Study of CXCL10 and CXCR3 Co-localization in the Inflamed Central Nervous System" *Journal of Neuroimmunology*
41. Matsuura, Kubota, Tanaka et al. (2015) "Visualization of HTLV-1-Specific Cytotoxic T Lymphocytes in the Spinal Cords of Patients With HTLV-1-Associated Myelopathy/Tropical Spastic Paraparesis" *Journal of Neuropathology & Experimental Neurology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12332328&blobtype=pdf | # The Intracerebral Haemorrhage in Patients With Dengue Fever: A Systematic Review and Meta-Analysis
Mingxia Xu, Ming Dong
## Abstract
Dengue virus is a neurotropic virus capable of infecting the supporting cells of the central nervous system. One of the most severe neurological consequences of this infection is intracerebral haemorrhage, which is a leading cause of death worldwide. This study aimed to systematically review and analyse the existing literature on this topic, providing insights into the potential neurological consequences for patients with dengue fever. A comprehensive search was conducted across the PubMed, Scopus, Web of Science, and Embase databases to extract relevant published data up to February 2025. This meta-analysis included articles that were designed as cohort studies. A critical appraisal was conducted using the Newcastle-Ottawa Scale (NOS) score. Risk was employed as a measure of pooled effect size based on a random-effects model. Heterogeneity was assessed using the Q test and the I 2 index. This meta-analysis included 6 studies involving a total of 2861 individuals who directly assessed the risk of intracerebral haemorrhage. The reported risk of intracerebral haemorrhage was 14 cases per 1000 dengue fever patients [0.014 (95% CI: 0.002, 0.026), p = 0.020, I 2 = 94.64%]. Notably, prospective studies with low methodological quality indicate a higher risk of intracerebral haemorrhage compared to retrospective studies and those of high quality. Given the high risk of intracerebral haemorrhage in patients with dengue fever, it is essential for physicians to evaluate affected individuals for the potential occurrence of cerebral haemorrhage.However, dengue haemorrhagic fever (DHF), also known as severe dengue, presents a wide spectrum of bleeding manifestations and carries a mortality rate of 26% [7]. Neurological involvement, which may include encephalopathy, intracerebral haemorrhage (ICH), or infarction, is a rare but potentially fatal complication of dengue [8,9]. The incidence of ICH in dengue infections is approximately 1.1% [10].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
Dengue virus infection poses a significant health threat, affecting at least 3.6 billion people across more than 125 countries in tropical and subtropical regions who are at high risk [1,2]. Dengue is the most rapidly spreading mosquito-borne disease in worldwide and recent study estimated the global burden of dengue to be approximately 390 million infections per annually [3]. Dengue is a febrile illness with clinical manifestations ranging from asymptomatic infection to severe infection with multi-organ dysfunction. A majority of patients with dengue fever (DF) recover without complication, but the mortality rate ranges from 1% to 5% without treatment, but it decreases to less than 1% when adequate treatment is administered [4][5][6].
ICH, a condition characterised by bleeding within the brain, is a leading cause of morbidity and mortality worldwide [11]. Dengue virus is a neurotropic virus with the ability to infect the supporting cells of the central nervous system (CNS). Neural injury during the acute stage of the infection results from direct neuro-invasion and/or the phenomenon of antibody-dependent enhancement, resulting in plasma leakage and coagulopathy [12].
Dengue fever poses a significant public health challenge due to the potentially fatal outcomes associated with severe infections, including ICH. Consequently, assessing the risk of ICH and implementing appropriate clinical management strategies are crucial for preventing mortality related to dengue fever. The objective of this article is to systematically review and analyse the existing literature on this subject, offer insights into the risk of potential neurological complications of dengue, particularly ICH, and identify gaps in research that could inform clinical practice and public health strategies.
## 2 | Method
The PRISMA statement was followed in conducting this metaanalysis [13]. We performed a comprehensive literature search using the online databases PubMed, Scopus, Web of Science, and Embase, covering publications up to April 2025. The search terms included 'Intracerebral Haemorrhage OR Cerebral Haemorrhage OR Cerebrum Haemorrhage OR Cerebral Brain Haemorrhage OR Cerebral Parenchymal Haemorrhage' and 'Dengue Fever OR Classical Dengue OR Classical Dengue Fever OR Break Bone Fever' which the used to identify literature.
## 2.1 | Inclusion and Exclusion Criteria
Studies were included for further evaluation if they met the following criteria [1]: publications in English [2]; investigation of ICH, including subarachnoid haemorrhage, intraparenchymal haemorrhage (IPH), subdural haemorrhage (SDH), or haemorrhage related to strokes and venous thrombosis [3]; determination of the association between ICH and dengue fever; and [4] evaluation of risk or incidence, along with their corresponding 95% confidence intervals [14], or sufficient data was provided to assess these associations. Studies were excluded based on the following criteria: non-English articles; other types of publications such as conference proceedings, abstracts, reviews, or meta-analyses; and insufficient data to calculate risk and 95% CIs.
## 2.2 | Data Extraction and Quality Evaluation
All included studies were meticulously identified by two investigators, and any uncertain data were reviewed by a third author. The following information was collected: the first author's name, publication year, nationality, sample type, assay used to evaluate the expression of ICH, sample size, type of ICH, and the associated risk along with their 95% confidence intervals [14].
## 2.3 | Quality Appraisal
The quality assessment of the included studies was conducted using the Newcastle-Ottawa Scale (NOS). The NOS employs a 'star system' to evaluate three criteria: selection, comparability, and exposure. Studies could receive a maximum of four stars for selection, three stars for exposure, and two stars for comparability. In total, the methodological quality of each study was rated on a scale from 0 to 9 stars, with a higher star count indicating superior methodological quality. Furthermore, the quality assessment was carried out by two independent researchers, and any disagreements were resolved by a third party [15].
## 2.4 | Statistical Analysis
The association between dengue fever and ICH was evaluated using Stata software version 17, utilising a random effects model (StataCorp, 2024, Stata Statistical Software: Release 17, College Station, TX). The risk of ICH and the corresponding 95% confidence intervals [14] were collected to evaluate the risk of ICH in individuals infected with dengue fever. Heterogeneity was assessed using the Q test and the I 2 index. Studies with an I 2 index of less than 25%, between 25% and 75%, and greater than 75% were categorised as having low, moderate, and high heterogeneity, respectively. In cases where the I 2 index exceeded 75%, a subgroup and meta-regression analysis was conducted based on the types of study design and level of quality. Forest plots were utilised to visualise the risk in each study, along with the estimated values and their corresponding 95% confidence intervals (95% CI) in both the main analysis and the subgroup analysis. Publication bias of the meta-analysis was determined by Begg's test and Egger's test.
## 3 | Result
## 3.1 | Search Results
A total of 520 articles were retrieved from the databases PubMed, Scopus, Web of Science, and Embase during the initial search. We identified 192 papers from Medline/PubMed, 143 papers from Scopus, 133 papers from Web of Science, and 52 papers from Embase. After removing duplicates, case reports, case series, review articles, and meta-analyses, as well as studies that were not written in English or were unrelated, 70 records were deemed eligible for full-text review. Only 6 papers reported the risk of ICH in patients with dengue fever. These 6 papers underwent quality assessment and were included in our meta-analysis (Figure 1). The overall sample size comprised 2861 patients, with individual studies ranging from 92 to 1148 participants.
## 3.2 | Main Characteristics
The six selected articles, which involve a sample size of 2861 participants, have been published across four continents: Taiwan, Pakistan, India, and Nepal. The individuals involved in the studies were people infected with dengue fever who were referred to hospitals or other medical and research centres for treatment. Necessary tests were conducted to diagnose any brain and neurological damage. Table 1 presents a comprehensive overview of the key characteristics of the included studies. All articles were designed as cohort studies, incorporating both retrospective and prospective approaches. The lowest and highest risks of ICH were 1 and 45 cases per 1000 dengue fever patients, respectively.
## 3.3 | Meta-Analysis
A random-effects model was employed to calculate the pooled risk ICH. The overall risk of ICH was determined to be 14 cases per 1000 dengue fever patients [0.014 (95% CI: 0.002, 0.026), p = 0.020]. Significant heterogeneity was observed among the included studies (Q-value: 18.67, df = 5, z-value = 2.32, I 2 = 94.64%, p-value ˂ 0.001). However, a subgroup analysis was conducted to explore the research in greater detail based on the methodological quality and study design (Figure 2).
## 3.4 | Subgroup Analysis
Subgroup analyses were conducted based on the methodological quality (6, 7, and 8 stars) and study design (retrospective and prospective cohort) for further investigation. Regarding the significant subgroups of the included studies, lower-quality articles reported a higher risk of ICH. Studies that collected data retrospectively found a greater incidence of ICH in patients with dengue fever (19 cases vs. 11 cases per 1000 patients). The findings of the subgroup analysis are presented in Table 2 and Figures 3 and4.
## 3.5 | Meta-Regression
Meta-regression analysis revealed a statistically significant association between the risk of ICH with methodological quality and study design. Specifically, for each one-level increase in the quality of the article, the risk of reporting ICH decreased by 8 cases per 1000 individuals [-0.008 (95% CI: -0.032, 0.016), p = 0.508], as well as prospective studies indicate a higher risk of ICH in 8 out of 10,000 individuals with dengue fever when compared to retrospective studies [0.0005 (95% CI: -0.037, 0.038), p = 0.976]. Following the meta-regression analysis that included significant variables, the heterogeneity among the studies was generally reduced (Q-value: 5.73, df = 3, I 2 = 82.53%, p-value = 0.001). Detailed results of the metaregression are presented in Table 3.
## 3.6 | Publication Bias
Publication bias was not detected in the meta-analysis using Begg's test (p = 0.259), but significant publication bias was observed in the Egger's test (p < 0.001). Significant heterogeneity was observed in the meta-analysis (Q (5) = 18.67, p < 0.001).
## 4 | Discussion
A total of 2861 patients with dengue fever patients were included in this systematic review and meta-analysis, which comprised 6 articles that directly assessed the risk of ICH. All included studies employed a cohort design. The results indicated that the risk of ICH in patients with dengue fever patients was elevated, and this risk was found to be statistically significant. The methodological quality and the study design were identified as variables affecting the risk of ICH.
ICH in patients with dengue has rarely been reported. Reported on 799 patients with dengue fever, 21 of whom (0.5%) presented with central nervous system involvement, but only one of which had ICH [21]. Reported 9 cases of intracranial bleeding in patients with dengue, however, no incidence rate was reported [22]. The reported risk of ICH in this study was 14 cases per 1000 dengue patients, with a range of 1-45 cases. There was variability in the risk of ICH among individuals with dengue fever across the included studies. This variability may reflect differences in the patient populations studied; the highest risks were reported in cohorts of critically ill patients and in studies examining the neurological complications of dengue fever.
Most studies on ICH in patients with dengue fever are limited to examining individuals with ICH and are primarily conducted as case reports or case series. The objective of these studies was not to estimate the risk or incidence of ICH in this patient population [10,16]. However, the studies included in this systematic review and meta-analysis reported the number of individuals with dengue fever who developed ICH. Two prospective [20,21] and four retrospective [16][17][18][19] studies were reviewed. The results of the meta-regression model indicated that prospective studies, which monitor patients more closely for research purposes and are less likely to record events with bias, reported a higher risk of ICH.
The results presented in Table 2 indicate that lower-quality studies and retrospective studies reported a higher risk of intracranial haemorrhage (ICH) in individuals with dengue fever. However, in the meta-regression model shown in Table 3, where both study quality and study design were included simultaneously, studies with prospective designs reported a higher risk of ICH. This discrepancy may be attributed to the potential for measurement, confounding, and selection biases, which are more prevalent in retrospective studies [23][24][25].
The current study underscores the risk of ICH on dengue patients and the limited information available to medical practitioners. Timely diagnosis and intervention can save lives. Clinicians should remain vigilant for the possibility of ICH in dengue patients, particularly when they present with altered consciousness. Dengue infection in 2014 was identified as an independent risk factor for ICH, suggesting that viral pathogenesis may also contribute to the development of intracranial haemorrhage in these patients. The threshold for recommending a diagnostic brain CT scan should be lowered for dengue patients exhibiting altered consciousness, especially during years when a higher frequency of infarction and ICH has been observed [16].
## 5 | Conclusion
This systematic review and meta-analysis confirms that patients with dengue fever are at an increased risk of developing intracranial haemorrhage (ICH), a finding that is statistically significant. Although the incidence of ICH among dengue patients has historically been low, the results underscore the importance of recognising the potential for ICH, particularly in critically ill patients or those presenting with altered levels of consciousness.
The study further emphasises that prospective studies, which are
## References
1. Gibbons, Vaughn (2002) "Dengue: An Escalating Problem" *Bmj*
2. Deen, Harris, Wills (2006) "The WHO Dengue Classification and Case Definitions: Time for a Reassessment" *Lancet*
3. Bhatt, Gething, Brady (2013) "The Global Distribution and Burden of Dengue" *Nature*
4. Health Organization, Dengue (2009) "Guidelines for Diagnosis, Treatment, Prevention and Control"
5. Tsheten, Clements, Gray et al. (2021) "Clinical Predictors of Severe Dengue: A Systematic Review and Meta-Analysis" *Infectious Diseases of Poverty*
6. Kularatne, Dalugama (2022) "Dengue Infection: Global Importance, Immunopathology and Management" *Clinical Medicine*
7. Ranjit, Kissoon (2011) "Dengue Hemorrhagic Fever and Shock Syndromes" *Pediatric Critical Care Medicine*
8. Hendarto, Hadinegoro (1992) "Dengue Encephalopathy" *Pediatrics International*
9. Cam, Fonsmark, Hue et al. (2001) "Prospective Case-Control Study of Encephalopathy in Children With Dengue Hemorrhagic Fever" *American Journal of Tropical Medicine and Hygiene*
10. Gautam, Meena, Meena et al. (2016) "Retrospective Analysis of Prognostic Factors in Dengue Infected Patients With Intracranial Bleed"
11. Johansson, Mohamed, Moulin et al. (2021) "Neurological Manifestations of COVID-19: A Comprehensive Literature Review and Discussion of Mechanisms" *Journal of Neuroimmunology*
12. Trivedi, Chakravarty (2022) "Neurological Complications of Dengue Fever" *Current Neurology and Neuroscience Reports*
13. Moher, Liberati, Tetzlaff et al. (2009) "Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement" *Bmj*
14. Munhoz, Pedroso, Nascimento (2020) "Neurological Complications in Patients With SARS-CoV-2 Infection: A Systematic Review" *Arquivos de Neuro-Psiquiatria*
15. Stang (2010) "Critical Evaluation of the Newcastle-Ottawa Scale for the Assessment of the Quality of Nonrandomized Studies in Meta-Analyses" *European Journal of Epidemiology*
16. Chang, Huang, Chen et al. (2021) "Clinical Characteristics and Risk Factors for Intracranial Hemorrhage or Infarction in Patients With Dengue" *Journal of Microbiology, Immunology, and Infection*
17. Assir, Ahmad, Masood et al. (2014) "Deaths Due to Dengue Fever at a Tertiary Care Hospital in Lahore, Pakistan" *Scandinavian Journal of Infectious Diseases*
18. Kulkarni, Pujari, Gupta (2021) "Neurological Manifestations of Dengue Fever" *Annals of Indian Academy of Neurology*
19. Mathew, Pandian (2010) "Stroke in Patients With Dengue" *Journal of Stroke and Cerebrovascular Diseases*
20. Sahu, Verma, Jain (2014) "Neurologic Complications in Dengue Virus Infection: A Prospective Cohort Study" *Neurology*
21. Koshy, Joseph, John (2012) "Spectrum of Neurological Manifestations in Dengue Virus Infection in Northwest India" *Tropical Doctor*
22. Sam, Gee, Nasser (2016) "Deadly Intracranial Bleed in Patients With Dengue Fever: A Series of Nine Patients and Review of Literature" *Journal of Neurosciences in Rural Practice*
23. Euser, Zoccali, Jager et al. (2009) "Cohort Studies: Prospective Versus Retrospective" *Nephron Clinical Practice*
24. Sessler, Imrey (2015) *Observational Clinical Research*
25. Sessler, Imrey (2015) "Clinical Research Methodology 1: Study Designs and Methodologic Sources of Error" *Anesthesia & Analgesia* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12858440&blobtype=pdf | # Research Priorities for Human Immunodeficiency Virus and Acquired Immune Deficiency Syndrome in Iran: A Mixed-Methods Study Combining a Systematic Review and Delphi Consensus Approach
Forugh Aleebrahim, Marzieh Mahboobi, Fatem Rezaee, Maryam Nasirian
## Abstract
Background: In 2022, human immunodeficiency virus (HIV) affected 39 million globally, with Iran experiencing a concentrated epidemic. This study aimed to identify and prioritize critical HIV/ AIDS research areas to address existing gaps in the Iranian context. Methods: Researchers at Isfahan University of Medical Sciences employed a three-phase approach between May 2018 and March 2023. The first phase involved developing an HIV research matrix tailored to international guidelines and the Iranian context. In the second phase, a systematic review of HIV-related studies in Iran was conducted following the PRISMA protocol to assess quality and relevance. The third phase utilized a three-round Delphi method to collect expert feedback, which was used to prioritize research areas based on their importance and feasibility. Results: A systematic review of 6,310 sources yielded 745 documents for full-text analysis. Key research domains identified included prevention, diagnosis, care and treatment, epidemiology, and co-infections. Quantitative analysis indicated that 23% of studies focused on co-infections, while ethical issues were addressed in only 0.04%. Qualitative assessment revealed that 54% of reviewed studies were of low quality. Subsequent expert consultation and multi-criteria decision analysis ranked laboratory research and co-infections as the highest priorities. Specific sub-areas identified included molecular epidemiology and interventions addressing stigma and discrimination. Despite being highly cost-effective and urgently needed, research on stigma-related interventions exhibited a significant gap.
Conclusion:The study identified 11 main research areas and 60 sub-areas as priorities for HIV/AIDS research in Iran, highlighting the need for strategic funding and research planning. Ethical considerations and laboratory improvements are essential for enhancing public health outcomes and addressing emerging challenges in the HIV epidemic.
## •
HIV/AIDS continues to be a major public health challenge, particularly in lowand middle-income countries.
## •
Iran is experiencing a concentrated epidemic, requiring specifically tailored research and interventions.
## •
While key research areas, including prevention, treatment, epidemiology, and co-infections, are being studied, a comprehensive assessment of their quality and relevance is lacking.
## What's New
## •
This study identified 11 key research areas and 60 specific sub-areas to guide targeted HIV/AIDS research in Iran.
## •
It emphasized the critical need to enhance the quality of studies, particularly in laboratory research and co-infections.
## •
HIV/AIDS research priorities in Iran were identified using a systematic review, a tailored research matrix, and a rigorous three-round Delphi consensus process involving national experts.
## Introduction
Human immunodeficiency virus (HIV), the causative agent of acquired immune deficiency syndrome (AIDS), has remained one of the most critical global health challenges for the past four decades since its emergence. 1 According to the World Health Organization (WHO), approximately 39 million people were living with HIV in 2022, with 1.3 million new infections reported that year. 2 The infection has profoundly impacted millions of lives, contributing to poverty, homelessness, school dropout rates, discrimination, and loss of opportunities, with low-and middle-income countries being disproportionately affected. HIV has also impeded critical development goals, including poverty reduction, lower child mortality, improved maternal health, increasing access to primary education, and promoting equal rights for all. 3 According to the latest available report from the Iranian Ministry of Health, as of June 2023, the national HIV electronic data system had registered a total of 46,320 HIV cases (82% men, 18% women) and 22,415 documented deaths. 4 Current evidence indicates that the HIV epidemic in Iran remains concentrated, with a prevalence of less than 1% in the general population. 5 However, without timely and effective interventions, the conditions are highly favorable for a broader spread of the virus. 6 Despite the relatively low general prevalence, Iran faces unique challenges in HIV control. These challenges include persistent social stigmas surrounding HIV, inadequate access to harm reduction programs for people who inject drugs (PWID), a shift in transmission patterns from injection drug use to sexual transmission, and limited care availability for marginalized populations. [7][8][9] These challenges underscore the need for nuanced, contextualized research and targeted interventions to address the epidemic effectively.
As HIV infection spans various dimensions, from molecular to social and political, its control necessitates a comprehensive approach involving active collaboration among policymakers, healthcare providers, and researchers. Effective prevention and care for individuals living with HIV and AIDS requires an in-depth understanding of the epidemic's profile in each region, 10 which can only be obtained through rigorous and standardized research. 11 In Iran, this need is particularly critical due to unique challenges, including stigma, shifting transmission patterns, and inadequate access to harm reduction programs. Despite these pressing issues, a significant lack of data in the current literature hinders a comprehensive response, underscoring the urgent need for targeted research to inform evidence-based interventions. 12 Health research is a cornerstone of public health advancement and the pursuit of equity, especially in low-and middle-income countries. 13 However, conducting research requires skilled personnel, substantial financial resources, and time. Given the scarcity of resources alongside extensive health needs, prioritizing research activities is crucial to ensure their optimal allocation. 14 Thus, prioritization is a key management process for enhancing the effectiveness and impact of health research.
The Medical Commission conducted the first national health research prioritization in Iran between 1992 and 1994. Since 1995, this responsibility has been held by the Deputy of Research at the Ministry of Health and Medical Education. 15 A notable subsequent effort was a study led by Owlia and colleagues, which was conducted in collaboration with all medical universities and key stakeholders using the enhanced national health research strategy (ENHRS). This initiative identified national health research priorities through needs assessments at the university level, identifying HIV as a priority communicable disease. 16 Another study by Haghdoust and colleagues in 2008 aimed to determine research priorities for HIV and AIDS. It highlighted key areas, including education, national-level management, estimating the size and prevalence of HIV among high-risk groups and the general population, and exploring innovative prevention methods. 17 Similarly, a 2014 systematic review of HIV-related studies in Iran, which searched national and international databases, emphasized prioritizing epidemiological research on high-risk groups. 18 While these studies have provided valuable insights, they do not fully address the evolving dynamics of the HIV epidemic in Iran, such as the shift in transmission patterns and the needs of marginalized populations. Updated and targeted research prioritization is urgently required to address these gaps and to develop contextually relevant and effective interventions. 19 This study aimed to identify and prioritize research needs in the field of HIV/AIDS in Iran. Despite the recognized importance of such research, prior studies have been limited, highlighting the urgent need to address gaps and establish priorities, particularly within the context of constrained resources. The main innovation of this study lay in its combined application of a systematic review and a threestage Delphi method, aligned with a locally contextualized research matrix. Unlike previous studies that relied solely on expert opinions or single-method approaches, this study integrated quantitative evidence mapping with expert consensus to identify research gaps and local priorities tailored to the Iranian context.
## Materials and Methods
This study, conducted by epidemiologists at Isfahan University of Medical Sciences between May 2018 and March 2023, employed a multimethod approach to identify and prioritize HIV/AIDS research priorities in Iran. The methodology integrated a systematic review of peer-reviewed literature, a three-round Delphi technique to establish expert consensus, and expert consultations to gather practical insights from field practitioners.
## Phase 1: Preparation of the HIV Research Matrix
The preparation of the HIV research matrix was conducted in two stages to ensure a comprehensive and contextualized framework for identifying research priorities in Iran. In the first stage, a structured search was performed in international databases to identify global and regional HIV research areas and sub-areas, including guidelines, protocols, and international reports. Recognized global resources were included to ensure alignment with evidencebased practices and internationally accepted standards. These resources included WHO guidelines, which provide a globally recognized framework for health policies and practices; UNAIDS reports, which serve as key references for coordinating the global HIV response; guidelines from the International AIDS Society (IAS) and the International Union against Sexually Transmitted Infections (IUSTI) for insights into clinical management. Furthermore, official resources from the Centers for Disease Control and Prevention (CDC) and various government portals were selected to reflect diverse national and regional HIV policies.
A broad search of peer-reviewed literature was conducted using major academic databases, including Medline, PubMed Central, Embase, the Cochrane Library, Scopus, Web of Science, Ovid, NICE, and SIGN. This comprehensive search ensured coverage of medical, pharmacological, clinical, and interdisciplinary studies, as well as evidence-based guidelines. A structured search strategy employed relevant keywords related to HIV/AIDS, interventions, services, and prioritization. The search process involved two components: a broad search of all areas of HIV research globally and a targeted search focused specifically on HIV research in Iran. The global search encompassed interventions, services, and prioritization, while the Iran-specific search concentrated on high-risk groups, behaviors, research methods, and surveillance.
The search process was conducted without time constraints to include both historical and recent studies. Two independent epidemiologists screened and reviewed the documents based on pre-defined inclusion and exclusion criteria. Any disagreements regarding document inclusion were resolved through discussion. If a consensus could not be reached, a third senior epidemiologist was consulted to make the final decision.
The inclusion criteria encompassed documents directly related to HIV/AIDS research, including guidelines, protocols, or strategic reports providing data on HIV research priorities in global settings. Study designs spanning clinical, social, and epidemiological research were included. Exclusion criteria involved removing duplicate records, inaccessible fulltext documents, and publications unrelated to HIV research priorities. Non-English or non-Persian documents were also excluded after the initial screening.
Based on the extracted information, a multidimensional HIV research matrix was developed to detail global areas and sub-areas of HIV research.
In the second stage, the draft HIV research matrix was reviewed and refined during three expert meetings with HIV specialists and researchers. These sessions involved a detailed analysis of national-level documents, such as the national HIV program, strategic HIV reports, international expectations for Iran, priorities identified by the AIDS department, and inputs required for specific estimation models. Various aspects of the HIV research matrix were modified and contextualized to align with Iran's unique needs and challenges. This phase, which lasted from May 2018 to February 2020, resulted in a tailored HIV research matrix that integrated global best practices with national priorities and served as the foundation for subsequent phases of the study.
## Phase 2: Mapping of Studies Conducted in Iran
In this phase, a structured review was conducted following the PRISMA protocol to examine all documents and studies related to HIV and high-risk groups in Iran, published up to March 2023. The review was performed by two HIV research experts who searched both domestic (SID, IranMedex, and Magiran) and international databases and search engines (PubMed, Scopus, Google Scholar, and Web of Science). Keywords were selected using MeSH and Emtree terms, and a comprehensive search strategy was developed (table 1). Iran J Med Sci January 2026; Vol 51 No 1
A systematic approach was implemented to address potential disagreements during the review process. Following the initial screening and data extraction, any discrepancies regarding article selection or relevance were addressed through group discussions. If consensus could not be reached, an additional senior expert was consulted to make the final decision. This collaborative approach ensured the consistent application of inclusion criteria and upheld rigorous quality standards for the selected studies.
The inclusion criteria encompassed all epidemiological studies conducted in Iran after 1996, including cross-sectional, case-control, retrospective cohort, prospective cohort, clinical trials, systematic reviews, and molecular epidemiology studies. For prevalence studies, a clearly defined population was required; analytical studies were required to have defined exposures and outcomes. The abstracts and full texts of articles or reports, collected in the EndNote database, were reviewed for relevance to the study's purpose. Irrelevant records were excluded, while relevant ones proceeded to the subsequent quality assessment stage.
In the second stage, the quality of the selected studies was systematically evaluated. A scoring system was developed based on specific criteria. Studies were scored primarily according to their design and methodological rigor, with meta-analyses receiving the highest scores (up to 25 points) and case reports the lowest. The journal index was assessed using the impact factor (IF) and Quartile (Q) ranking, with the formula IF+(1/Q). Key methodological factors included adequate sample size, random sampling methods, unbiased data collection, and detailed descriptions of study location, timeframe, and analysis methods. Each study's total score was calculated as the sum of these categories. Based on their total scores, studies were categorized into one of three quality groups: low (scores from 1 to 9), moderate (scores from 9 to 15), or high (scores from 15 to 25). For example, a meta-analysis conducted with rigorous methodology and transparent reporting would typically score above 20 and be categorized as "high quality." In contrast, a small cross-sectional study with methodological limitations would likely score below 9 and fall into the "low quality" category. As per the reviewers' recommendations, we have entirely removed IF from our scoring method. The results were recalculated using a revised methodology that excludes IF. This adjustment did not significantly impact the ranking or prioritization outcomes in this study.
Critical metadata was extracted for each study, including title, year, geographical location, study type, demographic characteristics (age and sex), sample size, and sampling methods. The data were used to prepare a summary matrix displaying the distribution of studies across different quality categories, study types, and population characteristics. The summary matrix provided an analytical overview and facilitated the identification of gaps in HIVrelated research across various regions and key populations in Iran.
## Phase 3: Prioritization of Areas and Sub-Areas of HIV Research in Iran Using a Three-Stage Delphi Method
In this phase, a three-round Delphi method was employed to prioritize HIV research areas and sub-areas in Iran. A panel of 20 specialists with extensive experience in HIV research, planning, policymaking, and related fields was engaged. Initially, the specialists were presented with a research matrix-developed based on a mapping of studies in Iran-which included the number of studies in each sub-area, their quality scores, and the identified informational gaps.
In the first round of the Delphi process, participants were asked to score each research sub-area from 1 to 5 based on its importance, necessity, feasibility, cost-effectiveness, and alignment with Iran's fourth strategic HIV/AIDS plan. They also assessed the existing mapping results for each sub-area based on the given criteria. Additionally, participants provided suggestions on the proposed timeframe, scale (university-based, multi-provincial, or national), study type, and urgency for research in each sub-area. The total score for each main research area was derived from the aggregate scores of its sub-areas, which were then used to develop a prioritization chart.
In the second round, the results of the first round, along with a scoring checklist, were emailed to the participants. They were asked to review the findings, reassign scores (1 to 5) for each sub-area, and submit their responses within two weeks. In the third and final round, the updated results and priorities from the second round were shared with the participants, who were then invited to provide final feedback within one week. Feedback from all three rounds was subsequently combined and analyzed to establish the final research priorities.
Experts were selected based on specific inclusion criteria, which required at least five years of active engagement in HIV-related research or policymaking. This experience was demonstrated through contributions to This multi-tool approach ensured a thorough and diverse data collection process, thereby enhancing the validity and reliability of the study's outcomes. By integrating qualitative feedback, quantitative scoring, and structured analysis, this phase established a comprehensive framework for identifying and prioritizing HIV research areas in Iran, aligning them with national objectives and addressing critical gaps.
Ethical approval for this study was obtained from Isfahan University of Medical Sciences (Approval code: IR.ARI.MUI.REC.1401.258.296140). All necessary permissions were secured and presented to the relevant authorities and participating experts before the commencement of the study. Researchers thoroughly explained the study's objectives to all participants to ensure a complete understanding of its purpose and scope. Informed consent was obtained from all experts, and participants were assured of their freedom to withdraw from the study at any stage without obligation.
## Results
A total of 6,310 information sources related to HIV/AIDS in Iran and worldwide were initially identified and systematically screened. After excluding irrelevant and duplicate records, 745 sources were selected for full-text review (figure 1). This process identified 11 main research areas, including prevention, diagnosis, laboratory, care and treatment, care systems, epidemiology, co-infections, policy, ethics, social support, and stigma and discrimination.
Notably, co-infections represented the most frequent research topic (23%), while ethical issues were the least addressed (0.04%). In terms of quality, more than half of the studies were classified as low quality, with only a small proportion achieving high-quality status. Laboratory and social support studies contained the highest share of high-quality articles (table 2).
Based on expert consultations and prioritization criteria, laboratory research, especially in reference laboratory enhancement and molecular epidemiology, was identified as the highest priority. Co-infection management and diagnostic improvements were also highly ranked, whereas stigma and discrimination received the lowest priority across multiple criteria (table 2, figure 2).
Additional sub-areas of importance were identified within each main area. Key topics included behavioral interventions for prevention, blood screening in diagnostics, and quality of life in care and treatment. Other priorities included resource allocation in policy, the role of human rights in ethics, and patient support during prevention in the social support domain (table 3).
The scoring and prioritization of areas and sub-areas were based on six key criteria: importance, necessity, feasibility of implementation, cost-effectiveness, alignment with the strategic objectives of the HIV/ AIDS program, and the mapping of existing studies. Epidemiology research received the highest score for importance. Social support research scored highest for necessity and feasibility, while social support and stigma, and discrimination received the highest scores for cost-effectiveness. Experts also identified social support as the area with the highest alignment with the strategic objectives of the HIV/AIDS program, despite it having the lowest number of studies conducted in Iran (figure 3). This analysis highlighted the need for targeted research in laboratory sciences and co-infections as top priorities, while also revealing significant gaps in ethics, social support, and stigmarelated studies. The findings emphasized the importance of aligning research efforts with strategic objectives to address critical gaps in HIV/AIDS management effectively.
## Discussion
Based on the results, 11 main areas and 60 sub-areas were identified as research priorities for HIV/AIDS in Iran. Analysis revealed a disproportionate focus, with HIVrelated co-infections representing the majority of existing studies, while ethical issues were markedly neglected. A qualitative assessment found that over half (54%) of the reviewed studies were of low quality, with only 5% classified as high-quality. The identified research priorities encompassed laboratory studies, co-infections, diagnosis, policy, ethics, epidemiology, prevention, social support, care, surveillance, stigma, and discrimination. Key high-priority sub-areas included molecular epidemiology, strengthening reference laboratories, and research in virology and basic immunology. The identification of these priorities is critical for the strategic allocation of limited resources and for guiding future scientific inquiry. As in other regions, defining these key research areas is essential to inform evidence-based public health policies and clinical interventions in Iran. 20 Previous studies on HIV/AIDS research prioritization in Iran are limited. A seminal study by Haghdoost and colleagues identified four primary priority areas: preventive activities (2/43), national and provincial planning (4/25), HIV burden estimation (9/20), and fundamental research (5/10). Key sub-priorities included education (5/52), national-level macro management (8/31), assessing prevalence among high-risk groups (5/59), and exploring new prevention methods (7/66). 17 Notable differences exist between these earlier findings and those of the present study. For instance, while prevention research was the second most documented topic after co-infections in our analysis, it was ranked only seventh in overall priority. This discrepancy might be attributed to methodological differences. The present study employed a mixed-methods approach incorporating systematic reviews and expert consensus, whereas the prior study relied solely on expert opinions gathered through a limited questionnaire.
The field of epidemiology plays a crucial role in evaluating various aspects of HIV/AIDS and provides valuable insights for research, development, management, and strategic planning. Research in this area is significant for identifying both current and future risk factors. Among the 60 sub-disciplines, molecular epidemiology is one of the most important, although previous studies emphasized the need for estimating the incidence and prevalence of HIV/AIDS. 17,21 Within the laboratory field, "strengthening reference laboratories" has been identified as a key priority requiring increased attention. Enhancing laboratory capacity could improve diagnostic accuracy and establish a robust laboratory infrastructure, which is fundamental for the health sector. Furthermore, advanced laboratory studies that employed genetic analyses, immunological assessments, and viral load testing can elucidate molecular processes in HIV-infected patients, facilitating better predictions of treatment responses. Techniques, such as flow cytometry and RT-PCR, provide critical insights into the effects of HIV on the immune system and the progression of the disease in individuals with HIV and co-infections. 19 Conversely, some research highlighted the need to improve laboratory processes and reduce errors. Addressing these challenges in greater detail could lead to better strategies for enhancing laboratory efficiency, which would ultimately improve health services.
In the field of co-infections, cardiovascular diseases and other complex conditions associated with HIV/AIDS were prioritized, followed by infectious diseases such as hepatitis and tuberculosis. A study by Dousti and others in 2016, which used a structured review method to determine HIV/AIDS research priorities in Iran, revealed a limited number of studies on co-infections such as hepatitis B, hepatitis C, and tuberculosis. Research on drug user groups was also less common than studies on other highrisk populations. Consequently, co-infections, particularly hepatitis and tuberculosis, were identified as specific research priorities. 12 Given the increasing prevalence of cardiovascular diseases in Iran, a trend that mirrors global patterns, innovative strategies are required to prevent them in individuals with HIV/AIDS. As experts have emphasized, further research is essential to confirm or refute hypotheses in this area. Such studies could lead to the identification of more precise and compatible treatment strategies for managing simultaneous infections. Furthermore, laboratory research on immune cells and immune responses in patients with HIV and co-infections could deepen our understanding of immune-viral interactions and contribute to developing more targeted treatments for these individuals.
The results of this study indicated a shift in prevention efforts from solely focusing on education to emphasizing the interaction between behavior and biomedical prevention strategies. 22 While previous studies highlighted the importance of education in HIV prevention, the focus on cultural change and social behavior modification, as shown in this study, offered the potential to enhance the effectiveness of prevention programs. This integrated approach could have a more significant impact on preventing HIV transmission and improving public health outcomes. 23 In the area of HIV care and treatment, the study highlighted that issues related to aging and disability had the most significant impact on the quality of life of people living with HIV. This finding underscored the need for programs and policies that address quality of life, aging, and disability management. However, most existing research in this area concentrated on the biomedical effects of HIV, with less attention paid to these psychosocial and quality-of-life issues. 24 In Iran, economic challenges and sanctions significantly limit the resources, making strategic financial allocation for HIV research critical. 20 Proper funding is essential to upgrade research infrastructure and enable advanced studies through improved laboratory technologies, data analytics, and clinical trial resources. Such advancements are necessary to generate reliable data for evidence-based policymaking. With sufficient financial support, Iranian researchers could innovate in diagnostics, drug development, and prevention strategies, fostering self-reliance in public health. A transparent and equitable allocation process is essential to prevent resource wastage and ensure provincial and local institutions have equal access to funding, thereby promoting regional expertise. Wellfunded research not only improves the quality of life for people living with HIV but also reduces long-term healthcare costs through effective prevention and treatment strategies.
The field of ethics, particularly concerning human rights, remains crucial for a comprehensive HIV response. A review of over 31,000 HIVrelated articles and reports published between 2003 and 2015 found that 83% demonstrated a positive impact from human rights interventions. These studies often incorporated principles such as non-discrimination and accountability to improve access and acceptability concerning the right to health. 25 This study is the first in Iran to combine a comprehensive systematic review with a nationally adapted multi-criteria research matrix and a rigorous multi-round Delphi process. This novel approach prioritized HIV/AIDS research in alignment with policy objectives, ensuring the results were both evidence-based and context-specific.
This study contributed significantly to HIV/ AIDS research in Iran by employing a systematic review of extensive existing literature, expert consultations, and the Delphi method to prioritize research areas, all while carefully considering Iran's specific context. The findings highlighted the necessity for a comprehensive and transparent protocol to guide research activities, ensuring they align with the country's needs. Policymakers and stakeholders must focus on managing HIV programs by ensuring adequate research budgets and rigorous methodological evaluations. Collaboration with UNAIDS and the Ministry of Health is essential to define research priorities, including study frequency, target populations, and province-specific focus areas. A nationwide program led by the Ministry of Health should be developed and implemented to achieve these goals.
This study aimed to inform strategic planning, resource allocation, and policymaking by identifying critical areas requiring immediate attention and providing evidence-based recommendations. It also highlighted the importance of revisiting and updating research priorities every 2-3 years to ensure their continued relevance. The resulting framework provided a clear roadmap for targeted research and effective interventions to improve HIV/AIDS management in Iran.
However, the findings of this study should be interpreted within the context of several limitations. The reliance on published literature and excluding gray literature or documents in languages other than English and Persian might have introduced bias.
Additionally, the inclusion of studies with varying levels of quality-54% of which were categorized as low quality-could affect the comprehensiveness of the findings. Despite efforts to evaluate the adequacy of sample sizes in the included studies, the lack of detailed reporting and transparency in some articles limited our ability to reproduce and validate comprehensive statistical calculations for sample size sufficiency. Addressing this limitation in future research by employing more explicit criteria and consulting with statistical experts is essential for improving this process.
An additional limitation was the initial use of a journal's impact factor (IF) as part of the quality scoring framework, as IF reflects a journal's subject area rather than the intrinsic quality of an individual study. Although additional metrics such as quartile rankings (Q) were incorporated to balance this effect, valuable reviewer feedback prompted us to exclude IF from our scoring system. Notably, this change did not significantly affect the principal findings, the classification of study quality, or the overall research priorities identified, and it ultimately improved the methodological rigor and validity of our recommendations.
Finally, the postponement of planned focus group discussions with HIV experts due to the COVID-19 pandemic limited the methodological scope of the study. The reliance on the chosen prioritization framework might have also introduced publication and selection biases, which could have been minimized by employing a validated and universally recognized framework.
## Conclusion
This study addressed critical gaps in HIV/AIDS research in Iran by providing actionable insights for policymakers and researchers. It highlighted the urgent need to prioritize neglected areas, such as ethical considerations, stigma reduction, and social support mechanisms, which have historically received minimal focus. Policymakers are encouraged to establish dedicated funding streams to address these priorities, promote multidisciplinary research collaborations that include community stakeholders, and implement regular reassessments of research priorities to adapt to evolving epidemiological trends.
Moreover, translating research findings into evidence-based public health policies can significantly enhance prevention efforts, improve treatment equity, and optimize the management of co-infections. Such targeted interventions will not only bridge existing gaps but also establish a sustainable and adaptive framework for HIV research and policymaking in Iran, ultimately improving health outcomes and advancing the national response to HIV/AIDS.
## Acknowledgment
The authors would like to extend their gratitude to Isfahan University of Medical Sciences, particularly the Department of Epidemiology, for their invaluable research design and methodology expertise. We were also grateful to the AIDS and Infectious Diseases Office of the Ministry of Health and Medical Education for their critical insights and valuable data contributions. Special appreciation is extended to the panel of experts, researchers, practitioners, and community stakeholders who participated in the Delphi process, providing enriching discussions and practical feedback. Their collective efforts were instrumental in advancing this study and addressing key priorities in HIV/AIDS research in Iran. communicable disease management, Iran. Regional Knowledge Hub for HIV/AIDS Surveillance at Kerman University of Medical Sciences; 2010. 7 Al-Ghafri Q, Radcliffe P, Gilchrist G. Barri
## References
1. Aliyu, Varkey, Salihu et al. (2010) "The HIV/AIDS epidemic in Nigeria: progress, problems and prospects" *Afr J Med Med Sci*
2. Raeisi, Tabrizi, Gouya (2020) "IR of Iran National Mobilization against COVID-19 Epidemic" *Arch Iran Med*
3. Kimera, Vindevogel, Reynaert et al. (2020) "Experiences and effects of HIV-related stigma among youth living with HIV/AIDS in Western Uganda: A photovoice study" *PLoS One*
4. (2023) "Report of recorded cases of HIV infection in the Islamic Republic of Iran until the end of"
5. Seyedalinaghi, Taj, Mazaheri-Tehrani et al. (2021) "HIV in Iran: onset, responses, and future directions" *AIDS*
6. Haghdoost, Mirzazadeh, Nedjat et al. (2022) "Bio-behavioral surveillance of women sex workers project report: Center for Rep"
7. Uneke, Ezeoha, Ndukwe et al. (2013) "Research priority setting for health policy and health systems strengthening in Nigeria: the policymakers and stakeholders perspective and involvement" *Pan Afr Med J*
8. Owlia, Eftekhari, Forouzan et al. (2011) "Health research priority setting in Iran: Introduction to a bottom up approach" *J Res Med Sci*
9. Pubmed
10. Haghdoost, Sadeghi, Nasirian et al. (2012) "Research priorities in the field of HIV and AIDS in Iran" *J Res Med Sci*
11. Feizzadeh, Nedjat, Asghari et al. (2010) "Evidencebased approach to HIV/AIDS policy and research prioritization in the Islamic Republic of Iran" *East Mediterr Health J*
12. Demetriou, Abu-Shah, Valvo et al. (2020) "A dynamic CD2-rich compartment at the outer edge of the immunological synapse boosts and integrates signals" *Nat Immunol*
13. Pubmed
14. Kapiriri, Martin (2010) "Successful priority setting in low and middle income countries: a framework for evaluation" *Health Care Anal*
15. Sahu, Kumar, Chandra et al. (2020) "Findings from the 2017 HIV estimation round & trend analysis of key indicators 2010-2017: Evidence for prioritising HIV/AIDS programme in India" *Indian J Med Res*
16. Xu, Han, Jiang et al. (2021) "Prevention and control of HIV/ AIDS in China: lessons from the past three decades" *Chin Med J (Engl)*
17. Threats, Brawner, Montgomery et al. (2021) "A Review of Recent HIV Prevention Interventions and Future Considerations for Nursing Science" *J Assoc Nurses AIDS Care*
18. Beer, Tie, Crim et al. (2017) "Progress Toward Achieving National HIV/AIDS Strategy Goals for Quality of Life Among Persons Aged >/=50 Years with Diagnosed HIV -Medical Monitoring Project" *Journal Editors form for disclosure of potential conflicts of interest*
19. Stangl, Singh, Windle et al. (2019) "A systematic review of selected human rights programs to improve HIV-related outcomes from 2003 to 2015: what do we know?" *BMC Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12730492&blobtype=pdf | # Milk Modulates the Gastrointestinal Stability of Tick-Borne Encephalitis Virus: Implications for Alimentary Transmission
Martin Machacek, Michaela Berankova, Jiri Salat, Daniel Ruzek
## Abstract
Tick-borne encephalitis virus (TBEV) can be transmitted alimentarily through contaminated dairy products, yet the mechanisms by which the virus survives the digestive tract remain poorly understood. In this study, we investigated the stability of TBEV in milk under simulated gastrointestinal conditions. While milk is known to preserve viral infectivity at low temperatures, our results demonstrate that in the gastric environment and at physiological temperature, it exerts a destabilizing effect, significantly reducing TBEV viability. All major milk fractions-whey, casein, and lipids-contribute to this effect. This highlights the necessity for rapid transit of virus-containing milk through the stomach to avoid inactivation. Conversely, in the intestinal environment, milk protects TBEV from bile salt-mediated inactivation, allowing viral persistence in the upper small intestine. Casein was identified as the primary protective component counteracting bile salt disruption. These findings offer new insights into how milk can simultaneously act as a transmission vehicle and modulator of TBEV stability, advancing our understanding of alimentary infection routes and their implications for public health.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
Tick-borne encephalitis virus (TBEV; Orthoflavivirus encephalitidis) is an enveloped, positive-sense single-stranded RNA virus belonging to the Orthoflavivirus genus within the Flaviviridae family [1,2]. It is the causative agent of tick-borne encephalitis (TBE), a viral disease that primarily affects the central nervous system in humans and is endemic to large parts of Europe and Asia [1].
While the primary route of TBEV transmission to humans is through the bite of infected Ixodes ticks, an alternative transmission pathway through the alimentary route-specifically, the consumption of unpasteurized milk and dairy products from sheep, goats, or cows-has been well documented and remains a public health concern in endemic areas [3][4][5]. The first known alimentary outbreak of TBEV occurred in 1951 in Rožňava, Czechoslovakia, linked to the ingestion of contaminated goat milk [6]. Since then, multiple outbreaks have been reported in various countries, including Hungary (2011) [7], Croatia (2019) [8], and France (2020) [9], with the majority of alimentary cases being reported in Slovakia, where incidence has shown a rising trend in recent years [4,10].
Infected milk and dairy products from goats, sheep, and cows represent a recognized risk of alimentary TBEV transmission. Pasteurization has been shown to inactivate the virus effectively [11], mitigating the risk. One unconfirmed historical report also suggested possible TBEV transmission through human breast milk [12], although this has not been corroborated by further studies [13].
Although alimentary infection is not the main transmission route, it is epidemiologically important in rural areas where raw dairy consumption is common. Slovakia reports the highest incidence of alimentary TBE in Europe, accounting for up to 17% of national cases. In contrast, only about 0.9% of TBE cases in the Czech Republic are attributed to this route [14]. The mechanisms underlying TBEV infection via the gastrointestinal (GI) tract are not fully elucidated. Experimental studies have shown that TBEV can replicate in Caco-2 cells-an established model of human intestinal epithelium [15]. However, conflicting findings exist regarding the virus's stability in the acidic gastric environment. Earlier studies reported TBEV persistence in gastric fluid for up to 2 h even at very low pH (1.49-1.80) [16], whereas more recent in vitro experiments demonstrated complete viral inactivation at pH 2.0 after just 20 min [17]. This discrepancy supports the hypothesis that TBEV likely bypasses gastric degradation under specific conditions and gains access to the host via intestinal epithelium [18].
TBEV exhibits remarkable stability in milk, particularly at refrigeration temperatures (4°C-8°C), with no significant difference in viral stability between milk and cell culture media such as Dulbecco's Modified Eagle Medium (DMEM) or phosphate-buffered saline (PBS) [19,20]. However, at body temperature (37°C), viral stability is significantly reduced in milk compared to laboratory media [19], possibly due to enzymatic and compositional factors.
The physiological environment of the GI tract plays a critical role in the success of alimentary infection. Gastric pH is typically between 1.4 and 2.1 in a fasted state but can rise to near-neutral levels (6-7) shortly after feeding, remaining elevated for up to 2 h [21,22]. In the duodenum, pH ranges from ~5.0-6.0 in the fed state and 6.5 in the fasted state [23]. Similarly, bile acid concentrations vary from 0.1 mmol/L in the stomach to 12-16 mmol/L in the fed small intestine [24,25]. These pH shifts and bile acid concentrations likely influence viral stability, protein digestion, and ultimately mucosal absorption.
In terms of gastric transit, caloric content influences gastric emptying time. For example, 500 mL of water exits the stomach within 30-60 min, whereas an equivalent volume of milk may remain for over 2 h [26,27]. During digestion, milk proteins undergo enzymatic cleavage; casein, which comprises around 80% of milk protein, is rapidly digested by pepsin and is no longer detectable after 20 min in the stomach [28]. Other proteins vary in resistance-lactoferrin is quickly cleaved [29], while βlactoglobulins are notably resistant to gastric proteolysis [30,31].
In this study, we explored the mechanisms by which TBEV may survive and pass through the human GI tract during alimentary infection. Specifically, we examined the virus's stability in milk and its behavior under simulated gastric and intestinal conditions. The study also investigates how different milk fractions affect TBEV stability in the acidic stomach environment and how milk-particularly casein-can protect the virus against bile salts in the intestine. These findings help to clarify how TBEV may remain infectious after oral ingestion, contributing to our understanding of alimentary transmission pathways.
## 2 | Materials and Methods
## 2.1 | Cells and Virus
A549 cells (ATCC CCL-185) were maintained in DMEM supplemented with 10% fetal bovine serum (FBS), 100 μg/mL penicillin, 100 μg/mL streptomycin, and 1% L-glutamine. Cultures were incubated at 37°C in a humidified atmosphere containing 5% CO 2 .
The low-passage TBEV strain Hypr was obtained from the Collection of Arboviruses at the Institute of Parasitology, Biology Centre of the Czech Academy of Sciences (České Budějovice, Czech Republic). This strain was originally isolated in 1953 from the blood of a 10-year-old patient in Brno, Czechoslovakia [32].
## 2.2 | Processing of Milk Samples
Commercial cow milk (pasteurized) sourced from Czech producers was bought at a local grocery store. Unpasteurized cow and goat milk were sourced from South Moravian producers, while sheep milk came from a producer in South Bohemia. Pasteurization of goat milk and unpasteurized cow milk from local producer was performed by heating the milk to 63°C for 30 min in a water bath. Pasteurized sheep milk was obtained from a local grocery store. Milk used in experiments was not stored for longer than 2 weeks and its pH was always checked right before conducting the experiment.
Milk samples were fractionated into fat and aqueous components (skim milk) by centrifugation (4000g, 30 min, 5°C) as described in Blans et al. [33]. The aqueous phase was further separated into casein and whey fractions following a modified method in Jensen et al. [34]. Briefly, defatted milk was adjusted to pH 4.6 with 1 M acetic acid, incubated at 4°C for 10 min, and then neutralized with 1 M sodium acetate. After centrifugation (1500g, 10 min, 5°C), the supernatant (whey) was filtered through a 0.45 µm filter to remove residual casein micelles.
To assess hydrolyzed casein, we compared bovine casein sodium salt (Merck, USA, Cat. No. C8654) with casein peptone (Serva, Germany, Cat. No. 48600.04), both in concentration corresponding to casein concentration in milk (2.56 g/L). For hydrolyzed whey, whey protein (obtained from milk and diluted to its original concentration in milk) was treated with QIAGEN protease (Cat. No. 19157) (5 μL/10 mL whey) at 55°C for 60 min, then the enzyme was inactivated at 70°C for 30 min. Oleic acid (ROTH, Germany, Cat. No. 48600.04) and triolein (Merck, USA, Cat. No. Y0001113) were each diluted in water to 1.7 g/L, reflecting half the oleic acid concentration found in milk (3.4 g/L) [35].
## 2.3 | Biorelevant Media
To simulate GI conditions, biorelevant media fasted state simulated gastric fluid (FaSSG), FaSSIF (fasted state simulated intestinal fluid) and fed state simulated intestinal fluid (FeSSIF) were obtained from Biorelevant.com (London, UK). Biorelevant media containing pH buffer and 3F powder (bile salts and lecithin) were prepared as instructed by the manufacturer. For simulations of fed gastric pH, FaSSG was adjusted to pH values of 3.0, 4.5, and 6.0. To assess the effect of milk and its fractions, each solution was mixed with milk or milk fractions at a 1:1 volume ratio and its pH adjusted to 1.6, 3.0, 4.5, and 6.0 if necessary.
## 2.4 | Viral Stability Assays
To assess viral stability, TBEV (1 × 10 8 PFU/mL) was diluted 1:10 in PBS and then mixed with each test solution at a 1:9 ratio. Samples were incubated in a shaking heat block (37°C, 270 rpm). At time points 1, 10, 30, 60, and 120 min, 30 μL aliquots were collected, immediately diluted 1:10 in 10% FBS in PBS (to neutralize acidic pH), and subjected to plaque assay.
For long-term stability experiments, TBEV (1 × 10 8 PFU/mL) was diluted 1:25 in PBS and then combined with test media 1:9. Samples were incubated at 8°C, and aliquots were collected on Days 0, 1, 2, 3, 5, and 7. Each aliquot (25 μL) was transferred into fresh DMEM (225 μL) and stored at -80°C until all time points were collected for titration.
## 2.5 | Viral Affinity to Milk Fractions
To evaluate the potential binding affinity of TBEV to milk components, the virus (1 × 10 8 PFU/mL) diluted 1:15 in PBS was mixed with whole milk in ratio 1:20. After centrifugation (14 300g, 15 min), 20 μL of each fraction (milk fat, skim milk, and pellet) was subjected to plaque assay. A control sample (whole milk with virus, not centrifuged) was processed identically.
## 2.6 | Plaque Assay
Plaque assays were performed using A549 cells as described previously [36,37]. Briefly, serial 10-fold dilutions of viral suspension were prepared in 24-well plates and cells (1 × 10 5 cells/well) were added directly to the viral suspension in each well. After a 3-4-h adsorption period at 37°C, cells were overlaid with DMEM containing 1.5% carboxymethylcellulose and incubated for 4 days. Then, the cell monolayers were washed with PBS and stained with naphthalene black. Titers were expressed as plaque-forming units per milliliter (PFU/mL).
## 2.7 | ELISA
The presence of specific anti-TBEV antibodies in sheep, goat, and locally produced cow milk was assessed using ELISA IM-MUNOZYM FSME (TBEV) IgG All Species Kit (Progen, Cat. No. 7701075). The ELISA kit was used according to manufacturer's instructions. Samples with antibody concentrations below 63 VIEU/mL were considered negative.
## 2.8 | Statistical Analysis
All experiments were performed in hexaplicates (typically two independent experiments in biological triplicates) unless stated otherwise. Statistical comparisons of viral titers were performed using the nonparametric Mann-Whitney U test. Analyses were conducted using GraphPad Prism v7.0 (GraphPad Software, USA). A p value below 0.05 indicated statistical significance. The graphs show mean values; error bars denote standard error, and the dashed line marks plaque assay detection limits. We used the Kruskal-Wallis test to compare milk fractions after fractionalization.
## 3 | Results
## 3.1 | Milk Supports TBEV Stability at Low Temperature, but Reduces Viability During Short-Term Incubation at Body Temperature
To evaluate the role of milk in the persistence of TBEV under conditions relevant to alimentary transmission, we first investigated how the virus is affected by storage in milk over time (Figure 1A). Specifically, we compared the viral stability of TBEV in milk and PBS at refrigeration temperature (8°C) (Figure 1B) and at physiological temperature (37°C) (Figure 1C). At 8°C, a statistically significant reduction in viral titer was observed in both milk and PBS after just 1 day of incubation (Supporting Information S2: Table 1) when compared to the initial titer. However, infective particles were still detected in both samples after 7 days. A small but statistically significant reduction in viral titer was consistently observed in milk compared to PBS. After 2 days, titers were 4.363 ± 0.387 log 10 PFU/ mL in milk versus 4.674 ± 0.210 in PBS (p = 0.0022). This trend continued at 5 days (4.070 ± 0.233 vs. 4.577 ± 0.221; p = 0.0087) and 7 days (3.960 ± 0.208 vs. 4.422 ± 0.150; p = 0.0065), indicating a gradual loss of infectivity in milk under cold storage conditions (Figure 1B).
To simulate short-term exposure to body temperature, we incubated TBEV in milk and PBS at 37°C for 120 min (Figure 1C). The virus again exhibited a significantly greater decrease in titer in milk after 120 min (5.107 ± 0.300 log 10 PFU/ mL) than in PBS (5.642 ± 0.301; p = 0.0087), suggesting that components of milk may negatively affect TBEV stability under conditions resembling the upper GI environment.
Notably, a more pronounced destabilizing effect on TBEV was observed in goat milk, sheep milk, the raw cow milk, and its pasteurized variant used in the pasteurization experiments when incubated at 37°C. These differences in viral viability among different milk types are provided in the Supporting Information S2: Figure 1.
Prior to our experiments, samples of sheep, goat, and locally produced cow milk were tested by ELISA and found negative for TBEV-specific antibodies. As the experimental results were consistent across different batches of milk, there was no evidence that batch variability influenced the outcomes of the short-or long-term stability assessments or any other experiments.
## 3.2 | Milk Accelerates TBEV Inactivation Under Simulated Gastric Conditions
To explore how gastric conditions influence the fate of TBEV during the early stages of digestion, we examined viral stability in biorelevant solutions simulating fasted and fed gastric environments at various pH levels (1.6, 3, 4.5, and 6). We also assessed whether milk modulates viral survival under these acidic conditions (Figure 2A). In all tested gastric solutions, TBEV was rapidly inactivated compared to the PBS control. In the highly acidic fasted-state gastric solution (pH 1.6), no infectious viral particles were detected even after just 1 min of incubation. In the pH 3 solution, simulating a fed stomach with lower acidity, a significant reduction in viral titer was observed after 1 min (4.176 ± 0.260 vs. 5.431 ± 0.331 log 10 PFU/mL in PBS; p = 0.0022), and complete viral inactivation occurred within 120 min. Under moderately acidic conditions (pH 4.5 and 6), viral particles remained detectable after 120 min; however, the titers were still significantly lower than in PBS (pH 4.5: 2.926 ± 0.627 vs. 5.470 ± 0.346; p = 0.0022; pH 6: 3.754 ± 0.632 vs. 5.470 ± 0.346; p = 0.0022) (Figure 2B,C).
When milk was added to these gastric solutions (1:1 ratio), viral inactivation occurred more rapidly. In pH 1.6 solution containing milk, no infectious virus was detected after 1 min, identical to the milk-free condition. However, the virus was cleared even faster in less acidic conditions. For instance, in pH 3 with milk, no virus was detected after just 60 min (vs. 120 min without milk). Notably, a significant reduction in titer was already present after 1 min in milk-containing solutions: pH 3 (1.313 ± 1.439 vs. 5.834 ± 0.259 in PBS; p = 0.0022), pH 4.5 (2.608 ± 0.156 vs. 5.834 ± 0.259; p = 0.0022), and pH 6 (4.665 ± 0.197 vs. 5.834 ± 0.259; p = 0.0022) (Figure 2B,C).
When directly compared to solutions without milk, viral titers after 1 min were significantly lower in milk-containing solutions at all pH levels tested: pH 3 (1.313 ± 1.439 vs. 4.176 ± 0.260; p = 0.0022), pH 4.5 (2.608 ± 0.156 vs. 5.197 ± 0.354; p = 0.0022), and pH 6 (4.665 ± 0.197 vs. 5.281 ± 0.275; p = 0.0022) (Figure 2B,C).
After 120 min, milk still exhibited a suppressive effect, most notably at pH 4.5 (0.632 ± 0.984 with milk vs. 2.926 ± 0.627 without; p = 0.0022). At pH 6, the difference was smaller and not statistically significant (3.501 ± 0.377 vs. 3.754 ± 0.632; p = 0.5152) (Figure 2B,C). Similar trend confirming negative effect of milk in the stomach was observed in the solutions with 1:3 and 1:7 ratio of milk to gastric fluid mimicking solution (Supporting Information S2: Figure 2) and in solutions with goat milk, sheep milk, and cow milk from local producer both before and after pasteurization (Supporting Information S2: Figure 3).
Together, these results demonstrate that the gastric environment is highly unfavorable for TBEV survival, and surprisingly, the presence of milk further enhances viral inactivation across a range of gastric pH values. This suggests that milk does not protect the virus in the stomach and may even facilitate its degradation under acidic conditions.
## 3.3 | All Milk Fractions Contribute to TBEV Inactivation in Simulated Gastric Conditions
To assess viral affinity to individual components, TBEV was mixed with milk and samples were subsequently centrifuged. Comparison of viral titers in the separated fractions (milk fat, skim milk, and pellet) revealed no significant differences (p = 0.3272), indicating no preferential association of the virus with any particular milk component (Figure 3B).
To investigate the role of different milk components in the inactivation of TBEV in gastric-like environments, milk was fractionated into fat and skim milk by centrifugation. The skim milk was further separated into whey and casein fractions via acetic acid (CH 3 COOH) precipitation and centrifugation (Figure 3A). TBEV was added to each fraction and subsequently cultivated for 2 h at 37°C.
When the antiviral activity of the milk fractions was evaluated across different pH conditions (3.0, 4.5, and 6.0), all tested fractions (fat, skim milk, whey, and casein) significantly reduced viral titers within 1 min of exposure compared to PBS controls (p < 0.05). No infectious virus was detected after 30 min at pH 3.0 in any milk fraction; notably, in skim milk at pH 3.0, viral inactivation occurred within 10 min (Figure 4A,B). At pH 4.5, complete inactivation was observed in skim milk after 60 min, while in other fractions the virus remained detectable up to 120 min, albeit with significantly reduced titers compared to PBS. After 120 min at pH 4.5, viral titers in milk fat (2.804 ± 0.217 log 10 PFU/mL) were similar to PBS (2.926 ± 0.627; p = 0.8485), while titers in whey (0.291 ± 0.714; p = 0.0022) and casein (1.885 ± 0.931; p = 0.0498) were significantly lower (Figure 4A,B).
At pH 6.0, infectious virus remained detectable in all fractions after 120 min. However, titers in skim milk (2.188 ± 1.138; p = 0.0065), whey (2.217 ± 1.147; p = 0.0087), and casein (2.918 ± 0.280; p = 0.0195) were significantly reduced compared to PBS controls (3.754 ± 0.632). In contrast, titers in the milk fat fraction (3.772 ± 0.495) remained similar to controls (p = 0.7473) (Figure 4A,B). Samples treated with hydrolyzed casein and whey proteins showed reduced viral stability (see Supporting Information S2: Figure 4).
Additionally, treatment of TBEV with oleic acid at concentrations equivalent to those naturally present in milk fat [35], but in its fully dissociated free form (i.e., not incorporated in triacylglycerols), resulted in complete viral inactivation within 1 min of exposure, when the viral particles were treated with triolein to compare with dissociated oleic acid, the virus remained stable (Supporting Information S2: Figure 5).
## 3.4 | Milk Protects TBEV Against Bile Salt-Mediated Inactivation
To assess the effect of intestinal conditions on TBEV stability, we replaced full simulated intestinal fluid with 3F powder-cat physiologically relevant concentrations for fasting (2.24 μg/mL) and fed (11.2 μg/mL) states (Figure 5A). These were dissolved in PBS at a 1:1 ratio. Full simulated intestinal fluid would contain also FaSSIF and FeSSIF solution together with 3F powder, slightly adjusting pH of the biorelevant media. This substitution was based on comparable inactivation patterns with media containing only 3F powder or full simulated intestinal fluid (Supporting Information S2: Figure 6) and our prior detailed analysis of pH effects in gastric conditions.
In the high-concentration bile salt solution (fed-state equivalent), viral titers dropped significantly after 1 min in both PBS (0.662 ± 1.036 log 10 PFU/mL) and milk (01.762 ± 1.180) compared to PBS control (5.408 ± 0.278; p = 0.0022 for both). No infectious virus was detectable after 10 min in either condition (Figure 5B,C).
In the low-concentration bile salt solution (fasted-state equivalent), no virus was detected in the PBS mixture after 120 min. A significant reduction was already observed after 1 min (4.167 ± 0.779 vs. 5.408 ± 0.278; p = 0.0238). However, when the virus was exposed to the same concentration of bile salts in milk, it remained stable. Viral titers after 120 min (5.006 ± 0.528) were not significantly different from the control without bile salts (5.195 ± 0.422; p = 0.6234), indicating a protective effect of milk against bile salt-mediated inactivation (Figure 5B,C).
Similar trend confirming positive effect of milk in the intestine was observed in the solutions with two and three times higher concentration of bile salts (Supporting Information S2: Figure 7) and in solutions with goat milk, sheep milk, and cow milk from local producer both before and after pasteurization (Supporting Information S2: Figure 8).
## 3.5 | Casein Confers Protection Against Bile Salt-Mediated Inactivation
To identify the milk component responsible for protection against bile salts, we tested individual milk fractions-milk fat, skim milk, whey, and casein-each mixed with lowconcentration bile salts (fasted-state equivalent).
In the milk fat fraction, viral titers were significantly reduced after 1 min (2.982 ± 0.272 vs. 5.690 ± 0.384; p = 0.0022), and no virus was detectable after 120 min. Similarly, in the whey fraction, titers decreased significantly after 1 min (4.512 ± 0.910 vs. 5.875 ± 0.475 log 10 PFU/mL; p = 0.0430), with complete viral inactivation observed by 30 min (Figure 6A,B).
In contrast, skim milk showed moderate protective capacity. A statistically significant reduction in viral titer compared to PBS was observed at 60 min (5.065 ± 0.188 vs. 5.418 ± 0.187; p = 0.0195) and 120 min (5.063 ± 0.398 vs. 5.536 ± 0.150; p = 0.0152), though the absolute differences were small (Figure 5A,B).
Importantly, the casein fraction provided the highest level of protection. Viral titers remained stable over 120 min, with no statistically significant difference from PBS control (5.374 ± 0.559 vs. 5.804 ± 0.312; p = 0.1515), indicating that casein is likely the key component mediating the protective effect of milk against bile salts (Figure 5A,B). Viral stability disappeared when bile salts solutions were combined with hydrolyzed casein (Supporting Information S2: Figure 9).
## 4 | Discussion
TBEV is a neurotropic flavivirus typically transmitted via tick bites but can also be acquired alimentarily, most commonly through consumption of unpasteurized dairy products [4,5,10,14]. While this route of infection is well-documented, the mechanisms by which TBEV survives the hostile conditions of the GI tract remain poorly understood. In this study, we investigated the stability of TBEV in milk and its individual components under conditions simulating the gastric and intestinal environments, with the aim of identifying the factors contributing to viral survival or inactivation along the digestive route.
Our results demonstrate that TBEV remains relatively stable in milk under cold storage, with a moderate reduction in viral titer observed over 7 days at 8°C. This suggests that a considerable portion of viral particles remains infectious even after prolonged refrigeration. These findings are consistent with previous reports of TBEV stability at low temperatures in dairy products [19]. Additionally, when incubated at 37°C to simulate body temperature, no statistically significant decrease in viral titer was observed over 120 min, indicating that TBEV remains stable in milk for at least 2 h under physiological conditions. Despite historical reports suggesting TBEV can remain infectious in the stomach for up to 2 h at low pH (pH 1.5-1.8) [16], our findings suggest rapid inactivation at such acidic conditions. At pH 1.5, no viral particles were detected after even 1 min, confirming the virus's high sensitivity to very low pH. This is in agreement with recent findings [17], though we observed a higher tolerance at pH 3, with viable virus still detectable after 1 h. These differences may stem from experimental conditions or strain variation.
When milk was added to low-pH environments mimicking gastric conditions (pH 3, 4.5, and 6), a further reduction in viral stability was observed across all tested pH values. This suggests that milk components exacerbate TBEV destabilization under acidic conditions. Given that postprandial gastric pH drops from ~6 to ~3 within the first hour and to ~1.6 within 2 h [21], it is unlikely that TBEV remains infectious in the stomach for prolonged periods. Instead, we hypothesize that alimentary infection results from the rapid passage of infectious viral particles through the stomach before they are inactivated, consistent with evidence that portions of liquid meals can exit the stomach within minutes [38,39].
To identify which milk components contribute to viral destabilization under acidic conditions, we fractionated milk and tested individual components. Initial separation and plaque assay analysis showed no preferential affinity of TBEV for any particular milk fraction (fat, skim milk, or pellet), and virus could be detected in all. This confirmed the suitability of the fractionation approach for assessing antiviral effects.
When testing viral stability in each milk fraction under acidic conditions, we found that the fat fraction had a destabilizing effect at pH 3 but not at pH 4.5 or 6. We attribute this effect to the release of free fatty acids from triacylglycerols under acidic conditions, as these free fatty acids-particularly oleic acidhave been reported to disrupt viral membranes [40,41]. Our findings demonstrate that dissociated oleic acid, at physiological concentrations similar to those found in bovine milk [35], fully inactivated TBEV after just 1 min of exposure. In contrast, when oleic acid was present as part of triolein at comparable concentrations, TBEV exhibited stability. These results provide confirmation of our initial hypothesis. The skim milk fraction also exerted a destabilizing effect on TBEV across all tested pH conditions. Further separation revealed that both whey and casein contributed to this activity. Whey proteins, including lactoferrin, have well-documented antiviral properties [42][43][44], while some studies have also described antiviral effects of casein [45,46]. Notably, we did not observe significant antiviral effects in milk at neutral pH, suggesting that acidic conditions are required for these milk proteins to exert their activity against TBEV. Even mildly acidic pH (pH 6) was sufficient to trigger this effect. Both hydrolyzed casein and whey proteins showed destabilizing effects, indicating that antiviral activity may result from peptides in acidic conditions rather than the proteins themselves.
We next investigated TBEV stability in simulated intestinal conditions, focusing on the impact of bile salts. Exposure to bile salt concentrations mimicking fed-state intestines led to rapid viral inactivation, with no virus detectable after 10 min. Even at lower concentrations corresponding to fasting conditions, TBEV was inactivated within 120 min. These findings align with previous studies reporting antiviral activity of bile salts [47,48], although in contrast, bile salts may enhance replication in hepatotropic viruses [49][50][51]. When PBS was replaced with milk in the bile salt solution, a strong protective effect was observed at the lower bile salt concentration: TBEV remained stable for the full 120-min incubation. This protective effect was not seen at the higher bile salt concentration. To determine the responsible milk component, we tested individual fractions. The fat fraction failed to protect the virus, whereas the skim milk fraction conferred partial protection. Further separation showed that intact casein best preserved viral stability, with no significant titer loss over 120 min. When replaced by hydrolyzed casein, this protective effect was eliminated, indicating the importance of casein's integrity. This aligns with prior findings that casein can bind bile salts [52,53], offering a plausible mechanism for the observed protection.
In summary, our results suggest that while milk may initially protect TBEV in the intestine-likely through casein-mediated sequestration of bile salts-it contributes to viral destabilization in the acidic environment of the stomach. The presence of membranedisrupting fatty acids and antiviral milk proteins (whey and casein) in acidic conditions accelerates viral inactivation. These findings help explain how alimentary infection may occur despite the harsh conditions of the digestive tract and identify specific milk components that modulate viral stability in a compartment-specific manner.
## References
1. Ruzek, Županc, Borde (2019) "Tick-Borne Encephalitis in Europe and Russia: Review of Pathogenesis, Clinical Features, Therapy, and Vaccines"
2. Chiffi, Grandgirard, Leib et al. (2023) "Tick-Borne Encephalitis: A Comprehensive Review of the Epidemiology, Virology, and Clinical Picture" *Reviews in Medical Virology*
3. Elbaz, Gadoth, Shepshelovich et al. (1980) "Systematic Review and Meta-Analysis of Foodborne Tick-Borne Encephalitis" *Emerging Infectious Diseases*
4. Ličková, Fumačová, Havlíková et al. (2021) "Alimentary Infections by Tick-Borne Encephalitis Virus" *Viruses*
5. Buczek, Buczek, Buczek et al. (2022) "Food-Borne Transmission of Tick-Borne Encephalitis Virus-Spread, Consequences, and Prophylaxis" *International Journal of Environmental Research and Public Health*
6. Ruzek, Kaucka (2024) "A Brief Tale of Two Pioneering Moments: Europe's First Discovery of Tick-Borne Encephalitis (TBE) Virus Beyond the Soviet Union and the Largest Alimentary TBE Outbreak in History" *Ticks and Tick-Borne Diseases*
7. Caini, Szomor, Ferenczi (2011) "Tick-Borne Encephalitis Transmitted by Unpasteurised Cow Milk in Western Hungary" *Eurosurveillance*
8. Ilic, Barbic, Bogdanic (2019) "Tick-Borne Encephalitis Outbreak Following Raw Goat Milk Consumption in a New Micro-Location" *Ticks and Tick-Borne Diseases*
9. Gonzalez, Bournez, Moraes (2022) "A One-Health Approach to Investigating an Outbreak of Alimentary Tick-Borne Encephalitis in a Non-Endemic Area in France (Ain, Eastern France): A Longitudinal Serological Study in Livestock, Detection in Ticks, and the First Tick-Borne Encephalitis Virus Isolation and Molecular Characterisation" *Frontiers in Microbiology*
10. Kerlik, Avdičová, Štefkovičová (2018) "Slovakia Reports Highest Occurrence of Alimentary Tick-Borne Encephalitis in Europe: Analysis of Tick-Borne Encephalitis Outbreaks in Slovakia During 2007-2016" *Travel Medicine and Infectious Disease*
11. Rónai, Egyed (2020) "Survival of Tick-Borne Encephalitis Virus in Goat Cheese and Milk" *Food and Environmental Virology*
12. Dobler, Gniel, Petermann et al. (2012) "Epidemiology and Distribution of Tick-Borne Encephalitis" *Wiener Medizinische Wochenschrift*
13. Martello, Gillingham, Phalkey (2022) "Systematic Review on the Non-Vectorial Transmission of Tick-Borne Encephalitis Virus (TBEV)" *Ticks and Tick-Borne Diseases*
14. Kríz, Benes, Daniel (1997) "Alimentary Transmission of Tick-Borne Encephalitis in the Czech Republic" *Epidemiologie, Mikrobiologie, Imunologie*
15. Yu, Achazi, Möller et al. (2014) "Tick-Borne Encephalitis Virus Replication, Intracellular Trafficking, and Pathogenicity in Human Intestinal Caco-2 Cell Monolayers" *PLoS One*
16. Pogodina (1958) "Resistance of Tick-Borne Encephalitis Virus to Gastric Juice" *Voprosy Virusologii*
17. Wiesner, Schmutte, Steffen (2021) "Susceptibility of Tick-Borne Encephalitis Virus to Inactivation by Heat, Acidic pH, Chemical, or UV Treatment" *Journal of Infectious Diseases*
18. Balogh, Ferenczi, Szeles (2010) "Tick-Borne Encephalitis Outbreak in Hungary Due to Consumption of Raw Goat Milk" *Journal of Virological Methods*
19. Offerdahl, Clancy, Bloom (2016) "Stability of a Tick-Borne Flavivirus in Milk" *Frontiers in Bioengineering and Biotechnology*
20. Saier, Maier, Atamer et al. (2015) "Thermal Inactivation of Tickborne Encephalitis Virus in Milk" *International Journal of Dairy Technology*
21. Dressman, Berardi, Dermentzoglou (1990) "Upper Gastrointestinal (GI) pH in Young, Healthy Men and Women" *Pharmaceutical Research*
22. Mclauchlan, Fullarton, Crean et al. (1989) "Comparison of Gastric Body and Antral pH: A 24 Hour Ambulatory Study in Healthy Volunteers" *Gut*
23. Bergström, Holm, Jørgensen (2014) "Early Pharmaceutical Profiling to Predict Oral Drug Absorption: Current Status and Unmet Needs" *European Journal of Pharmaceutical Sciences*
24. Poquet, Wooster (2016) "Infant Digestion Physiology and the Relevance of In Vitro Biochemical Models to Test Infant Formula Lipid Digestion" *Molecular Nutrition & Food Research*
25. Riethorst, Mols, Duchateau et al. (2016) "Characterization of Human Duodenal Fluids in Fasted and Fed State Conditions" *Journal of Pharmaceutical Sciences*
26. Bateman (1982) "Effects of Meal Temperature and Volume on the Emptying of Liquid From the Human Stomach" *Journal of Physiology*
27. Kunz, Feinle, Schwizer et al. (1999) "Assessment of Gastric Motor Function During the Emptying of Solid and Liquid Meals in Humans by MRI" *Journal of Magnetic Resonance Imaging*
28. Barbé, Ménard, Gouar (2013) "The Heat Treatment and the Gelation Are Strong Determinants of the Kinetics of Milk Proteins Digestion and of the Peripheral Availability of Amino Acids" *Food Chemistry*
29. Furlund, Ulleberg, Devold (2013) "Identification of Lactoferrin Peptides Generated by Digestion With Human Gastrointestinal Enzymes" *Journal of Dairy Science*
30. Mandalari, Mackie, Rigby et al. (2009) "Physiological Phosphatidylcholine Protects Bovine Beta-Lactoglobulin From Simulated Gastrointestinal Proteolysis" *Molecular Nutrition & Food Research*
31. Inglingstad, Devold, Eriksen (2010) "Comparison of the Digestion of Caseins and Whey Proteins in Equine, Bovine, Caprine and Human Milks by Human Gastrointestinal Enzymes" *Dairy Science & Technology*
32. Pospisil, Jandasek, Pesek (1954) "Isolation of New Strains of Meningoencephalitis Virus in the Brno Region During the Summer of 1953" *Lekarske Listy*
33. Blans, Hansen, Sørensen (2017) "Pellet-Free Isolation of Human and Bovine Milk Extracellular Vesicles by Size-Exclusion Chromatography" *Journal of Extracellular Vesicles*
34. Jensen, Poulsen, Møller et al. (2012) "Comparative Proteomic Analysis of Casein and Whey as Prepared by Chymosin-Induced Separation, Isoelectric Precipitation or Ultracentrifugation" *Journal of Dairy Research*
35. Månsson (2008) "Fatty Acids in Bovine Milk Fat" *Food & Nutrition Research*
36. De Madrid, Porterfield (1969) "A Simple Micro-Culture Method for the Study of Group B Arboviruses" *Bulletin of the World Health Organization*
37. Růžek, Gritsun, Forrester (2008) "Mutations in the NS2B and NS3 Genes Affect Mouse Neuroinvasiveness of a Western European Field Strain of Tick-Borne Encephalitis Virus" *Virology*
38. Goyal, Guo, Mashimo (2019) "Advances in the Physiology of Gastric Emptying" *Neurogastroenterology & Motility*
39. Hellström, Grybäck, Jacobsson (2006) "The Physiology of Gastric Emptying" *Best Practice & Research Clinical Anaesthesiology*
40. Thormar, Isaacs, Brown et al. (1987) "Inactivation of Enveloped Viruses and Killing of Cells by Fatty Acids and Monoglycerides" *Antimicrobial Agents and Chemotherapy*
41. Leu, Lin, Hsu (2004) "Anti-HCV Activities of Selective Polyunsaturated Fatty Acids" *Biochemical and Biophysical Research Communications*
42. Bojsen, Buesa, Montava (2007) "Inhibitory Activities of Bovine Macromolecular Whey Proteins on Rotavirus Infections In Vitro and In Vivo" *Journal of Dairy Science*
43. Gallo, Giansanti, Arienzo et al. (2022) "Antiviral Properties of Whey Proteins and Their Activity Against SARS-CoV-2 Infection" *Journal of Functional Foods*
44. Alves, Azevedo, Dias (2023) "Inhibition of SARS-CoV-2 Infection in Vero Cells by Bovine Lactoferrin Under Different Iron-Saturation States" *Pharmaceuticals*
45. Inagaki, Muranishi, Yamada (2014) "Bovine κ-Casein Inhibits Human Rotavirus (HRV) Infection via Direct Binding of Glycans to HRV" *Journal of Dairy Science*
46. Saint-Jean, Prieto, López-Expósito et al. (2012) "Antiviral Activity of Dairy Proteins and Hydrolysates on Salmonid Fish Viruses" *International Dairy Journal*
47. Herold, Kirkpatrick, Marcellino (1999) "Bile Salts: Natural Detergents for the Prevention of Sexually Transmitted Diseases" *Antimicrobial Agents and Chemotherapy*
48. Kim, Lee, Kim (1999) "Inhibition of Initiation of Simian Virus 40 DNA Replication In Vitro by the Ursodeoxycholic Acid and Its Derivatives" *Cancer Letters*
49. Chhatwal, Bankwitz, Gentzsch (2012) "Bile Acids Specifically Increase Hepatitis C Virus RNA-Replication" *PLoS One*
50. Patton, George, Chang (2011) "Bile Acids Promote HCV Replication Through the EGFR/ERK Pathway in Replicon-Harboring Cells" *Intervirology*
51. Reese, Moore, Mclachlan (2012) "Limited Effects of Bile Acids and Small Heterodimer Partner on Hepatitis B Virus Biosynthesis In Vivo" *Journal of Virology*
52. Euston, Baird, Campbell et al. (2013) "Competitive Adsorption of Dihydroxy and Trihydroxy Bile Salts With Whey Protein and Casein in Oil-in-Water Emulsions" *Biomacromolecules*
53. Lanzini, Fitzpatrick, Pigozzi et al. (1987) "Bile Acid Binding to Dietary Casein: A Study In Vitro and In Vivo" *Clinical Science* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12727887&blobtype=pdf | # Maternal and neonatal outcomes in obstetric antiphospholipid syndrome: a retrospective case-control study
Lin Rao, Jia Lu, Hong Li, Liang Xu, Dongjian Yang, Wendong Han, Li Chen, A Seval Ozgu-Erdinc, Alban Deroux
## Abstract
Objective: The combination of low-dose aspirin (LDA) and low-molecularweight heparin (LMWH) is the standard of care for obstetric antiphospholipid syndrome (OAPS), significantly improving live birth rates. However, whether this regimen fully normalizes the pregnancy course and mitigates risks for both the mother and the neonate remains unclear. This study aimed to systematically evaluate whether significant maternal and neonatal morbidity persists in OAPS patients despite successful treatment and live birth. Methods: This retrospective cohort study included 256 OAPS patients, including 166 criteria OAPS patients (C-OAPS-patients who fulfilled both the clinical and laboratory criteria of the Sydney criteria) and 90 non-criteria OAPS patients (NC-OAPS-patients who fulfilled only the clinical or only the laboratory criteria of the Sydney criteria) who achieved live birth, along with 768 matched healthy controls. We compared basic characteristics, laboratory parameters, and perinatal outcomes between the groups. Results: Compared to healthy controls (n = 768), treated OAPS patients (n = 256) exhibited a persistent hypercoagulable state (elevated D-dimer and fibrin degradation product (FDP), p < 0.01) and a higher incidence of anemia (p < 0.001). Their neonates had significantly lower birth weight (p = 0.006) and elevated risks of neonatal infection (adjusted OR = 3.12, p = 0.004) and hyperbilirubinemia (adjusted OR = 2.06, p = 0.024), with the infection risk remaining significant in full-term infants. A subgroup analysis revealed no significant differences in obstetric history, maternal complications, comorbidities, and outcomes between the C-OAPS and NC-OAPS groups. Conclusion: Despite standard treatment, OAPS patients who deliver successfully remain at an increased risk for persistent maternal hypercoagulability and adverse neonatal outcomes. These findings underscore the need for a paradigm shift in management-from merely ensuring live birth to safeguarding neonatal health through proactive, multidisciplinary perinatal care.
## 1 Introduction
Antiphospholipid syndrome (APS) is a systemic autoimmune disorder characterized by clinical manifestations, including thrombotic events and/or obstetric complications, and it is accompanied by the presence of antiphospholipid antibodies (aPLs) (1). The 2006 Sydney International Consensus provides an extensive overview of adverse obstetric outcomes linked to recurrent firsttrimester miscarriage, fetal loss, stillbirth, early and severe pre-eclampsia, or preterm birth (before 34 weeks of gestation) (2). In patients with a history of thrombosis, these complications are classified as obstetric APS (OAPS) and are primarily attributed to placental dysfunction (3). This is partly due to the detrimental effects of aPLs throughout the stages from implantation and placentation to delivery (4).
Recent research has indicated that placental inflammatory responses, including complement activation and subsequent endothelial damage in patients with OAPS, may impair the invasive function of placental trophoblasts (5). Additionally, the formation of neutrophil extracellular traps (6), release of interleukin-8 (7), upregulation of the target of rapamycin complex on endothelial cells (8), and an imbalance in angiogenic factors (9) do not culminate in thrombosis. While the administration of glucocorticoids appears promising in mitigating inflammatory response (10), the current gold standard in clinical treatment remains the combination of low-dose aspirin (LDA) and low-molecular-weight heparin (LMWH) (11). This treatment regimen has significantly increased the rate of successful deliveries in patients with OAPS; however, potential maternal placental pathology and effects on the offspring cannot be ruled out, due to the persistence of aPLs.
However, whether this treatment fully normalizes the course of pregnancy and eliminates risks for both the mother and neonate remains a subject of inquiry. A focus solely on live birth may overlook significant maternal and neonatal morbidity that persists despite treatment. For instance, some reports suggest an association between OAPS and adverse outcomes such as fetal growth restriction, preterm birth, and placental pathology even in treated pregnancies (9). Furthermore, a notable proportion of patients with clinical features highly suggestive of OAPS tested negative for the criteria OAPS (C-OAPS), a category often referred to as seronegative or non-criteria OAPS (NC-OAPS) (12), presenting a diagnostic and management challenge. According to the 2023 American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) classification criteria for antiphospholipid syndrome, these patients are likely to be classified into an underrecognized group. Although the new standards significantly enhance specificity and introduce a more detailed clinical and laboratory stratification weighting system, their strict inclusion criteria-requiring a cumulative score of at least 3 in both clinical and laboratory domains-may result in the exclusion of some patients with typical obstetric clinical manifestations whose laboratory tests do not meet the threshold from the classification (13).
Given these considerations, this study was conducted to systematically evaluate the pregnancy characteristics and outcomes in a cohort of OAPS patients who achieved a live birth following standard treatment. We aimed to retrospectively analyze and compare basic maternal characteristics, laboratory parameters, and perinatal outcomes between these treated OAPS patients and matched healthy controls. The objective of this study was to assess the potential persistent effects of OAPS on pregnancy and to provide a detailed clinical profile that may inform future management strategies and research directions.
## 2 Materials and methods
## 2.1 Study population
All medical histories for this study were recorded at the International Peace Maternity and Child Health Hospital of the China Welfare Institute from 2018 to 2022. All research participants had delivered successfully (see Figure 1).
## 2.2 The inclusion and exclusion criteria of the patients with OAPS
## 2.2.1 Clinical inclusion criteria for C-OAPS and NC-OAPS
Patients with OAPS are classified as either the C-OAPS group or the NC-OAPS group according to the Sydney criteria.
C-OAPS required at least one clinical criterion plus one laboratory criterion. Conversely, those who fulfilled the criteria in only one aspectthat is, patients who fulfilled the clinical criteria but did not fulfill the laboratory criteria or those who fulfilled the laboratory criteria but did not fulfill the clinical criteria-were classified as having non-criteria OAPS (NC-OAPS) (14). In accordance with the prevailing clinical guidelines, all enrolled OAPS patients were managed with a combination of low-dose aspirin and prophylactic low-molecular-weight heparin throughout pregnancy (from the first confirmation of pregnancy until at least 6 weeks postpartum) to improve obstetric outcomes (15).
(1) The clinical criteria were unexplained fetal death ≥10 weeks, premature birth <34 weeks due to placental dysfunction, or ≥3 consecutive miscarriages <10 weeks. (2) The laboratory criteria were lupus anticoagulant (LA), medium-/high-titer immunoglobulin G (IgG)/ immunoglobulin M (IgM) anticardiolipin (aCL), or anti-β2glycoprotein-I (anti-β2GPI) antibodies-mandated positivity on two occasions at least 12 weeks apart.
## 2.2.2 Exclusion criteria
The clinical exclusion criteria were as follows:
(1) Patients with active hepatitis B virus, hepatitis C virus, human immunodeficiency virus, syphilis, or tuberculosis infection; (2) Those with a history of smoking or drinking;
(3) Those who use drugs during pregnancy which can seriously affect maternal and infant outcomes; (4) Those who underwent induced abortion due to family planning or personal request; (5) Those with a history of significant diseases, such as severe lesions of vital organs, during pregnancy; (6) Those with a history of previous thrombosis or pregnancy complicated by thrombosis;
Frontiers in Medicine 03 frontiersin.org (7) Those without routine prenatal examinations and whose data were incomplete; and (8) Those with multiple gestations.
## 2.3 The inclusion and exclusion criteria of the healthy pregnant women
Case and control groups were matched at a 1:3 ratio according to the time of delivery and age to reduce confounding factors and selection bias.
## 2.3.1 Inclusion criteria for healthy pregnant women
The inclusion criteria for healthy pregnant women were as follows:
(1) Previously healthy pregnant women who gave birth during the same period as the case group; (2) Women who did not meet any of the OAPS and non-criteria OAPS diagnostic criteria;
(3) Those with no history or family history of autoimmune disease; and (4) Those who have not used drugs such as corticosteroids and immunosuppressants.
## 2.3.2 Exclusion criteria for healthy pregnant women
The exclusion criteria were the same as the OAPS patients.
## 2.4 Clinical data collection
Our study collected three types of information: basic information, laboratory examinations, and clinical characteristics during the perinatal period.
## 2.4.1 Basic information
Basic information included age, pre-pregnancy body mass index (BMI), weight gain during pregnancy, days of latest pregnancy, number of deliveries, number of pregnancies, history of preterm labor, history of stillbirth, and number of miscarriages.
## 2.4.2 Laboratory examination
Upon admission to the hospital for delivery, blood samples were collected. Relevant experimental indices, including international normalized ratio (INR), activated partial thromboplastin time (APTT), thromboplastin time (TT), prothrombin time (PT), D-dimer, FDP, total bile acid (TBA), red blood cells (RBCs), hemoglobin (HB), and platelets (PLTs), were tested.
## 2.4.3 Detection of antiphospholipid antibodies: main instruments and reagents
Our laboratory used the enzyme-linked immunosorbent assay (ELISA) as the core detection technology, utilizing the fully automated EUROIMMUN analyzer and corresponding test kits to perform quantitative detection of antiphospholipid antibodies in clinical serum samples.
## 2.4.3.1 Main instruments
The main instruments were the Fully Automated Fluorescence Immunoassay Analyzer SPRINTER XL from EUROIMMUN.
## 2.4.3.2 Reagent kits (all from EUROIMMUN)
The reagent kits included the Anti-Cardiolipin Antibody IgA Test Kit (product code: EA 1621-9601A), Anti-Cardiolipin Antibody IgG Test Kit (product code: EA 1621-9601G), Anti-Cardiolipin Antibody IgM Test Kit (product code: EA 1621-9601 M), Anti-β2-Glycoprotein I Antibody IgG Test Kit (product code: EA 1632-9601G), and Anti-β2-Glycoprotein I Antibody IgM Test Kit (product code: EA 1632-9601 M).
## 2.4.4 Clinical characteristics of the perinatal period
To accurately gauge the severity of obstetric complications, the study adhered to the Sydney International Consensus, which categorizes complications into three primary groups:
(1) Adverse pregnancy outcomes: intrauterine distress, fetal growth restriction (FGR), and postpartum hemorrhage; (2) Maternal complications and comorbidities: diabetes mellitus in pregnancy (pre-pregnancy diabetes mellitus and gestational diabetes mellitus [GDM]), hypertensive disorders of pregnancy (gestational hypertension [GH] and pre-eclampsia/eclampsia), thyroid disorders of pregnancy, intrahepatic cholestasis in pregnancy, hematologic disorders of pregnancy (anemia), placental abnormalities (placental adhesion, placental lakes, placental abruption, and placental implantation), abnormalities (abnormal amount and color of amniotic fluid), pregnancyrelated sexually transmitted diseases (mycoplasma infection), and pregnancy-related streptococcal infection; (3) Neonatal complications: preterm labor, low birth weight, neonatal asphyxia, neonatal infections, and neonatal hyperbilirubinemia.
## 2.5 Statistical methods
Values are expressed as mean (±S.D.) and numbers and percentages for qualitative variables. Student's t-test was used to compare continuous variable data following a normal distribution, while Mann-Whitney Wilcoxon's test was used for continuous variable data not following a normal distribution. The chi-squared test, Yates's correction for continuity, and Fisher's exact test were applied to compare categorical variables. For comparisons of continuous variables that were not normally distributed across the three independent groups, the Kruskal-Wallis H test was used. Post-hoc pairwise comparisons were then conducted using Dunn's test, applying a Bonferroni correction for multiple comparisons. A difference was considered statistically significant at a p-value of <0.05. To further evaluate the independent association between obstetric antiphospholipid syndrome (OAPS) and maternal/neonatal outcomes, binary logistic regression analyses were performed. Outcomes that showed significant differences (p < 0.05) in the univariate analyses were included in the multivariate models. The regression models were adjusted for potential confounding factors identified from clinical relevance and univariate results; the results are expressed as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). The statistical software SPSS 23.0 (IBM, United States) was used for dataset analyses.
## 3 Results
A total of 1,024 women were included in this study: 256 OAPS cases and 768 healthy controls (HCs). The OAPS group was divided into 166 patients in the typical OAPS group and 90 patients in the atypical NC-OAPS group, according to the Sydney Classification Criteria. All of the participating women had delivered successfully following treatment during the most recent pregnancy in the hospital. We also randomly selected 768 healthy pregnant women admitted to the obstetrics department during the same period, who were matched at a 1:3 ratio to the OAPS group based on the time of delivery and age.
## 3.1 Basic information and reproductive history of the OAPS and healthy control groups
In this study, significant differences were observed in gestational weight gain and gestational duration between the OAPS and HC groups. The OAPS group exhibited a higher mean weight gain (12.35 ± 5.69 kg) than the HC group (10.35 ± 7.10 kg, p < 0.001), as well as a shorter mean gestational period (266.38 ± 9.85 days versus 270.44 ± 10.98 days, p < 0.05). Additionally, marked disparities in gravidity and parity were evident between the groups. Compared to the HC group (32.81%), the OAPS group had a higher incidence of pregnancies with three or more instances of gravidity (64.06%) and a lower incidence of two or more instances of parity (21.87% versus 29.30%). Furthermore, a significant difference was observed in adverse maternal history: the incidence of preterm deliveries was substantially higher in the OAPS group (10.55%) than in the HC group (0.26%), and the rate of stillbirths was higher in the OAPS group (6.64%) than in the HC group (1.04%). Moreover, the OAPS group exhibited a higher frequency of miscarriages, with two and three or more occurrences corresponding to percentages of 37.11% versus 16.15 and 24.61% versus 8.20%, respectively. There were no differences between the C-OAPS and NC-OAPS groups (refer to Table 4).
Frontiers in Medicine 05 frontiersin.org
## 3.2 Results of laboratory test parameters
The incidence of APTT and PT abnormalities was significantly greater in the OAPS group than in the HC group (p < 0.001). Conversely, the incidence of PTA abnormalities was lower in the OAPS group than in the HC group (p = 0.023). Additionally, patients with OAPS exhibited higher rates of abnormalities in D-dimer, FDP, and TBA than the HC group (p < 0.001), with abnormality rates of 14.51% vs. 7.42, 13.67% vs. 3.91, and 3.52% vs. 0.39%, respectively. Furthermore, the incidence of RBC abnormalities was higher in the OAPS group than in the HC group (p = 0.02), consistent with the trend observed in the HB test (p = 0.001) (see Table 1).
## 3.3 Maternal complications, comorbidities, and outcomes
The prevalence of anemia was significantly greater in the OAPS group than in the HC group (p < 0.001). Furthermore, no significant difference was observed in the incidence of abnormal placental invasion and morphological anomalies between the OAPS group and the HC group (p > 0.05). Additionally, the rate of cesarean delivery was higher in the OAPS group than in the HC group (p = 0.048). Among the pregnancy outcomes, neonatal weight was significantly lower in the OAPS group than in the HC group (p = 0.006). Conversely, the incidence of intrauterine distress was significantly higher in the HC group than in the OAPS group (p = 0.001). There were no differences between the C-OAPS and NC-OAPS groups (Table 2).
The analysis demonstrated that there was no statistically significant difference in the percentage of preterm neonates with low birth weight between the two groups (p = 0.893). However, the incidence of low birth weight was significantly higher among preterm neonates in the OAPS group than in the HC group (p = 0.004), while it was significantly lower among full-term neonates (p < 0.001). Regarding neonatal complications, the findings indicated that both the incidence of neonatal infections and the prevalence of neonatal hyperbilirubinemia were significantly higher in the OAPS group than in the HC group (p = 0.004, p = 0.024). For full-term infants, the incidence of neonatal infection and neonatal asphyxia was significantly greater in the OAPS group than in the HC group (p = 0.013, p = 0.011). Additionally, the occurrence of complications, including infection, asphyxia, and hyperbilirubinemia, was markedly higher in preterm infants in the OAPS group than in those in the HC group (p < 0.001) (refer to Table 3). There were no differences between the C-OAPS and NC-OAPS groups (Table 2).
## 4 Discussion
Compared to the HC group, the most salient clinical manifestation in the OAPS group was pathological pregnancy, particularly a history of preterm labor, stillbirth, and miscarriages, all of which were significantly more frequent than in the HC group. The underlying causes of pathological pregnancy may involve activating endothelial cells, monocytes, and platelets by aPLs, leading to a procoagulant state. This process may be further exacerbated by the direct action of the placental trophoblast, resulting in trophoblast cell destruction and apoptosis. This process can lead to a cascade of adverse outcomes, including reduction of hormones such as human chorionic gonadotropin, diminished capacity of trophoblast cells to invade and implant in the uterus, and suppression of trophoblast cell proliferation. These alterations may hinder the embryo's attachment to the uterus and progression of the pregnancy (16). Although the standard treatment regimen of low-dose aspirin combined with low-molecular-weight heparin significantly improved the live birth rate in patients with OAPS (17), our research found that, even after standard treatment, OAPS mothers and fetuses still face a series of significant residual risks, including a hypercoagulable state of the mother, unique placental pathological changes, and adverse neonatal outcomes. This finding indicates that the current therapy has limitations in correcting the fundamental pathophysiological mechanisms of OAPS, highlighting the urgency of assessing its impact on pregnancy quality beyond the live birth rate. Key laboratory findings from this study delineate distinct hematological and biochemical profiles in OAPS patients compared to healthy controls. The OAPS group exhibited a significantly higher incidence of abnormalities in activated partial thromboplastin time (APTT) and prothrombin time (PT), yet a lower incidence of abnormal prothrombin activity (PTA). This apparent prolonged conventional coagulation time alongside a lower rate of reduced PTA likely reflects the therapeutic effect of low-molecular-weight heparin (LMWH) (18). The observed alterations in these coagulation parameters are consistent with the documented pharmacologic profile of LMWH calcium, as previously reported (19). LMWH, primarily through the inhibition of factor Xa, can induce a modest prolongation of APTT and PT, representing a predictable and intended anticoagulant state rather than a marker of bleeding risk (20). Concurrently, the significant elevations in the mean values and abnormality rates of D-dimer and fibrin degradation products (FDP) in the OAPS group provide direct laboratory evidence of ongoing fibrin formation and degradation, indicating a persistent hypercoagulable state with secondary fibrinolysis. Furthermore, the observed increase in total bile acid (TBA) levels suggests the presence of intrahepatic cholestasis in a subset of patients, which may compound obstetric risks. The higher incidence of red blood cell (RBC) and hemoglobin (HB) abnormalities is consistent with the greater prevalence of anemia observed in the OAPS cohort, implying a potential impact of the disease on hematopoiesis or erythrocyte stability. Collectively, these laboratory findings define a distinct hematological profile in OAPS, characterized by a treated prothrombotic phenotype and associated systemic involvement, underscoring the need for comprehensive monitoring.
The observed lower incidence of intrauterine distress in the OAPS group can be logically explained by its standardized clinical management. The implementation of intensified antenatal surveillance in these high-risk pregnancies facilitated timely detection of complications, leading to earlier elective delivery and thereby reducing the incidence of frank distress. In addition, a comprehensive review of the extant literature indicates that the effects of OAPS on fetal and neonatal outcomes primarily encompass birth weight, fetal death, preterm delivery, FGR, and fetal acidosis (21). In this study, the birth weight of neonates in the OAPS group was significantly lower than that in the HC group. However, there was no statistically significant difference in the proportion of low birth weight. Additionally, the OAPS group exhibited a higher proportion of preterm births than the HC group. Previous research found that 127 patients with OAPS were at risk of preterm labor, and the incidence of "spontaneous preterm labor" was 16.5% (22). Consequently, the occurrence of preterm labor in pregnancies affected by OAPS encompasses two distinct categories: (1) medically indicated preterm labor associated with pre-eclampsia or fetal factors and (2) a latent predisposition to preterm labor. The underlying predisposition to preterm labor is well-documented. Furthermore, the study noted that neonatal infections and hyperbilirubinemia were more prevalent in the OAPS group than in the HC group. Previous research has indicated that preterm infants are more susceptible to diseases and hyperbilirubinemia compared to term infants (23). To further investigate this observation, a subgroup analysis was conducted. Crucially, the subgroup analysis confirmed that this elevated risk remained significant even after accounting for prematurity, as the infection rate was also considerably higher in term neonates from the OAPS group. This suggests that the predisposition to infection is not solely a consequence of preterm birth but is intrinsically linked to the OAPS condition itself. We propose that the altered intrauterine environment in OAPS-characterized by a persistent placental inflammatory state and immune dysregulation (24)-may impair fetal immune system programming or function, increasing susceptibility to postnatal infection. These findings underscore that the ramifications of OAPS transcend the endpoint of live birth, warranting a shift in clinical focus from merely "securing fetal survival" to "safeguarding neonatal health. " This necessitates closer collaboration between obstetricians and neonatologists to optimize perinatal management from pregnancy into the postpartum period. Given these persistent challenges in management, the recent introduction of the ACR/EULAR 2023 classification criteria for APS marks a significant step forward. These updated criteria refine the original Sydney standards by incorporating a weighted scoring system and expanded serological markers, thereby enhancing specificity for research (13). However, it is important to distinguish between classification criteria, designed to ensure cohort homogeneity, and the broader clinical spectrum of the disease. The stringent nature of such criteria may inevitably exclude some patients with highly suggestive clinical presentations-a group often identified as NC-OAPS. Our study found that there was no significant difference between the NC-OAPS group and the C-OAPS group, not only in obstetric history but also in terms of maternal complications, comorbidities, and outcomes (Tables 234). This key finding demonstrates that patients in the NC-OAPS group experience a clinical burden of disease comparable to that of patients who meet the formal criteria. By intentionally including NC-OAPS patients, our study aims to characterize the outcomes and pathophysiological features across the entire clinical phenotype managed as OAPS, thus providing insights that may inform future diagnostic and therapeutic strategies.
For patients diagnosed with OAPS, particularly those with recurrent pregnancy loss, combination therapy with low-dose aspirin and low-molecular-weight heparin remains the cornerstone of management and is effective in improving live birth rates (25). However, our findings underscore that this standard regimen does not fully eliminate significant residual risks, including maternal hypercoagulability, placental pathology, and adverse neonatal outcomes. Therefore, meticulous monitoring for maternal and neonatal complications is imperative, and management must be highly individualized. Future care for OAPS necessitates a paradigm shift, extending beyond anticoagulation. A proactive, multidisciplinary approach-involving obstetricians, rheumatologists, and often neonatologists-is essential throughout the pre-pregnancy, antenatal, and postpartum periods. The overarching goal must transition from merely "ensuring survival" to comprehensively "optimizing health. " Achieving this requires a dual strategy: advancing interventions to target fundamental pathophysiological mechanisms earlier in the disease course and deepening clinical vigilance toward placental function and long-term neonatal wellbeing. This redefined focus should guide the next generation of research and clinical innovation in OAPS.
## 5 Conclusion
This study, focusing on OAPS patients who achieved successful delivery, shows that despite standard treatment with low-dose aspirin and LMWH, these women continue to exhibit a distinct and complex clinical profile. Key findings include a persistent maternal hypercoagulable state, evidenced by elevated D-dimer and FDP and an increased risk of adverse neonatal outcomes, such as lower birth weight, preterm delivery, and a heightened susceptibility to neonatal infection and hyperbilirubinemia. These findings collectively indicate that the central clinical challenge in OAPS is shifting from achieving live birth to safeguarding the quality of pregnancy and neonatal health. Consequently, a paradigm shift in management is warranted-extending beyond current anticoagulation strategies to embrace a proactive, multidisciplinary approach focused on pre-pregnancy counseling, intensified placental and fetal surveillance, and long-term neonatal care, with the ultimate goal of optimizing overall maternal and offspring health outcomes.
## 6 Limitations
Several limitations of this study should be acknowledged. First, its retrospective and single-center design may introduce selection bias and limit the generalizability of the findings. The sample size, though substantial, may still be insufficient to detect significant differences in rarer complications. Second, the study population consisted of patients who ultimately achieved a successful delivery, which may not fully represent the entire spectrum of OAPS severity, particularly those with the majority of refractory cases. Furthermore, the absence of detailed placental histopathological analysis for all participants limits our ability to correlate clinical outcomes with specific underlying placental lesions. Third, the subgroup analyses based on antibody profiles were limited by small sample sizes, which may have obscured statistically significant differences and precluded a more robust comparison. Finally, potential unmeasured confounding factors, despite our matching efforts, could have influenced the observed associations. Future prospective, multicenter studies with larger cohorts and systematic placental examination are needed to validate these findings and further elucidate the pathophysiological mechanisms linking OAPS to adverse outcomes.
## References
1. Ruiz-Irastorza, Crowther, Branch et al. (2006) "International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS)" *J Thromb Haemost*
2. Alijotas-Reig, Ferrer-Oliveras, Grp (2012) "The European registry on obstetric antiphospholipid syndrome (EUROAPS): a preliminary first year report" *Lupus*
3. Out, Kooijman, Cd et al. "Histopathological findings in placentae from patients with intra-uterine fetal death and anti-phospholipid antibodies"
4. Agostinis, Biffi, Garrovo et al. (1991) "In vivo distribution of beta2 glycoprotein I under various pathophysiologic conditions" *Eur J Obstet Gynecol Reprod Biol*
5. Marder, Knight, Kaplan et al. (2016) "Placental histology and neutrophil extracellular traps in lupus and pre-eclampsia pregnancies" *Lupus Sci Med*
6. Gladigau, Haselmayer, Scharrer et al. (2012) "A role for toll-like receptor mediated signals in neutrophils in the pathogenesis of the anti-phospholipid syndrome" *PLoS One*
7. Canaud, Bienaimé, Tabarin et al. (2014) "Inhibition of the mTORC pathway in the antiphospholipid syndrome" *N Engl J Med*
8. Girardi, Yarilin, Thurman et al. (2006) "Complement activation induces dysregulation of angiogenic factors and causes fetal rejection and growth restriction" *J Exp Med*
9. Hoppe, Burmester, Dorner (2011) "Heparin or aspirin or both in the treatment of recurrent abortions in women with antiphospholipid antibody (syndrome)" *Curr Opin Rheumatol*
10. Alijotas-Reig, Esteve-Valverde, Ferrer-Oliveras et al. (2020) "Comparative study of obstetric antiphospholipid syndrome (OAPS) and non-criteria obstetric APS (NC-OAPS): report of 1640 cases from the EUROAPS registry" *Rheumatology*
11. Barbhaiya, Zuily, Naden et al. (2023) "The 2023 ACR/EULAR antiphospholipid syndrome classification criteria" *Arthritis Rheumatol*
12. Alijotas-Reig, Esteve-Valverde, Ferrer-Oliveras et al. (2019) "The European registry on obstetric antiphospholipid syndrome (EUROAPS): a survey of 1000 consecutive cases" *Autoimmun Rev*
13. Tektonidou, Andreoli, Limper et al. (2019) "EULAR recommendations for the management of antiphospholipid syndrome in adults" *Ann Rheum Dis*
14. Bouvier, Cochery-Nouvellon, Lavigne-Lissalde et al. (2014) "Comparative incidence of pregnancy outcomes in treated obstetric 388 antiphospholipid syndrome: the NOH-APS observational study" *Blood*
15. Deng, Liao, Bs et al. (2022) "Recent advances in treatment of recurrent spontaneous abortion" *Obstet Gynecol Surv*
16. Qi, Li, Han et al. (2025) "Diagnostic value of blood coagulation index in obstetric antiphospholipid syndrome and its subtypes" *Med Clin (Barc)*
17. Qiuling, Hongli (2023) "Clinical efficacy of low molecular weight heparin combined with expectant therapy in the treatment of preeclampsia" *Clin Res*
18. Hirsh, Dalen, Je, Deykin et al. (1992) "Heparin: mechanism of action, pharmacokinetics, dosing considerations, monitoring, efficacy, and safety" *Chest*
19. Gutierrez, As, Figueras et al. (2023) "Correlation of placental lesions in patients with systemic lupus erythematosus, antiphospholipid syndrome and non-criteria obstetric antiphospholipid syndrome and adverse perinatal outcomes" *Placenta*
20. Wenghui, Zi, Hongxia et al. (2016) "Effect of different intervention time and intervention method for preventing preterm birth in APS with pregnancy" *Chin J Pract Gynecol Obstet*
21. Sifeng, Hongliang, Dijin et al. (2023) "Efficacy and adverse effects of blue light irradiation supplemented by breastfeeding with breast milk fortification in the treatment of jaundice in preterm infants" *Jilin Yixue*
22. Raschi, Borghi, Mo, Tedesco et al. (2024) "Antiphospholipid syndrome pathogenesis in 2023: an update of new mechanisms or just a reconsideration of the old ones? Rheumatology (Oxford)"
23. Soto-Peleteiro, Gonzalez-Echavarri, Ruiz-Irastorza (2024) "Obstetric antiphospholipid syndrome" *Med Clin (Barc)* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12686977&blobtype=pdf | # Factors Associated with HIV Stigma and Depression Among People Living with HIV and Food Insecurity in the Dominican Republic
Gabriela Armenta, Bing Han, Kartika Palar, Amarilis Then- Paulino, Glenn Wagner, Kathryn Derose, Middle Menlo Park
## Abstract
Food insecurity, depression, and HIV stigma adversely affect people with HIV and women face heightened vulnerability. Limited evidence exists regarding the factors associated with HIV internalized and experienced stigmas and depression for people with HIV and food insecurity in the Dominican Republic (DR). Using an HIV clinic-based sample of people with food insecurity in the DR (n=115), we found that depressive symptoms and internalized and experienced stigmas were highly prevalent. A cross-sectional, multivariate linear regression analysis found that women, Haitians, and those with a detectable viral load had higher internalized stigma; those reporting intimate partner violence had higher internalized and experienced stigmas, while those reporting higher social support had lower stigma scores for both types; finally, those with an undetectable viral load, better physical health, and better antiretroviral therapy adherence had reduced depression symptom severity. Addressing inequities related to gender, nationality and/or ethnicity, and HIV disease progression may mitigate internalized HIV stigma, and addressing the correlates of HIV stigmas and depression may improve HIV outcomes.
## INTRODUCTION
Psychosocial issues, including HIV stigma and depression, undermine the health of people living with HIV (PLHIV). 1 Internalized stigma involves self-application of negative stereotypes, and decreasing self-esteem and self-efficacy, while experienced stigma refers to devaluation by others due to a specific condition. 2,3 Nearly 90% of PLHIV experience internalized stigma, 4 and many encounter discrimination or are denied health services. 5 Internalized stigma hampers access to care and antiretroviral therapy (ART) adherence, 6 as well as social interactions and access to food and other resources. 7,8 Experienced stigma can lead to isolation, economic or food hardship, and lower ART adherence. [7][8][9][10] In addition to HIV-related stigmas, PLHIV experience a disproportionate burden of depression. Depression is characterized by diminished self-esteem, lack of vitality, social isolation, and disruption of everyday activities. 11 In general, PLHIV experience depressive disorders at a rate that is nearly twice that of people without HIV. 12 But the prevalence of elevated depressive symptoms among PLHIV ranges widely across different populations, with a meta-analysis of studies from low-, middle-, and high-income countries finding estimates ranging from approximately 13% to 78%. 13 Given the wide range of depression prevalence and its known impact on worsening health outcomes, reducing life expectancy, hindering ART adherence, and accelerating HIV disease progression, 14,15 it is important to examine factors associated with depression among PLHIV within specific countries.
The interplay between depression and HIV stigma, as highlighted by Rueda et al. (2016), 16 is linked to food insecurity, and this connection seems particularly pronounced among women. 17,18 This relationship is bidirectional, indicating a complex interaction where each factor can influence and exacerbate each other. Stigma among PLHIV can lead or exacerbate social exclusion and psychosocial stressors which contribute to limited food access and poor dietary intake. 17,19 Social exclusion and stressors, such as blame, can trigger isolation and harmful coping strategies, including withdrawal from social support, disengagement from health services, and inconsistent ART adherence, all of which may worsen depression. 14 Concurrently, food insecurity increases vulnerability to HIV acquisition through increased risk taking, 17 while heightening the risk for mental health problems and exacerbating depression and stigma by increasing shame, stress, and worry. 17,20 Food insecurity also intensifies disease progression by reducing ART adherence and viral suppression. 21,22 In 2019, the Caribbean had the second highest prevalence of PLHIV after Sub-Saharan Africa, with the DR having the second highest PLHIV number regionally. 23 However, limited research exists on stigma and depression among PLHIV in the Caribbean, warranting sociocultural and region-specific studies due to factors' variation across countries and regions. 24,25 Additionally, few studies have examined HIV stigma and depression among food insecure PLHIV in the region. Existing literature on HIV stigma or depression in the DR has been qualitative or focused on specific subpopulations such as women or female sex workers living with HIV. [26][27][28][29] Theoretical frameworks suggest that HIV stigmas, food insecurity, and mental health outcomes are interrelated. [30][31][32] This is supported by studies from the DR which have explored the relationship between being a woman and experiencing higher HIV stigmas and depression, since PLHIV face gender-specific challenges while they struggle to manage the effects of their diagnosis on their mental health and their lives more generally. [26][27][28]33,34 Additionally, existing literature indicates that greater HIV-related stigmas and depression are associated with demographic, social, and economic factors, along with systemic factors such as racial inequalities and intimate partner violence. 3,17,28,29,33,[35][36][37][38] Exploring factors related to depression, and internalized and experienced stigmas among PLHIV can inform strategies to enhance HIV care and outcomes, particularly in foodinsecure settings. We investigated these factors in the Dominican Republic (DR) in a sample of men and women living with HIV and who had moderate or severe food insecurity. The analysis was aimed at answering: i) what are the factors associated with internalized stigma, experienced stigma, and depression among PLHIV and food insecurity in the DR? and ii) Is gender associated with HIV stigmas and depression among PLHIV and food insecurity in the DR?
## METHODS
## Conceptual framework
Figure 1 presents the conceptual framework that guided our analysis. The framework illustrates the pathways from HIV and food insecurity to their impacts on morbidity, quality of life, and other outcomes. HIV and food insecurity reinforce each other, and they are related to inequalities or characteristics that are drivers and facilitators of HIV stigmas and depression, including economic (i.e., poverty), demographic (e.g., Haitian background), social (e.g., social support), and clinical or health-related factors (e.g., detectable viral load). HIV-related stigmas and depression further affect people's health and social integration.
## Participants and study setting
This cross-sectional correlational analysis was conducted with baseline data from a pilot cluster randomized controlled trial of an urban garden intervention called ProMeSA, which is an acronym that translates as "Project to Improve Food Security" (Clinical Trials Identifier: NCT03568682). The team was led by researchers at RAND, in collaboration with the Universidad Autónoma de Santo Domingo (UASD) and University of California, San Francisco, and local partners from the Dominican Ministry of Public Health, Ministry of Agriculture (Division of Urban Gardens), the National HIV/AIDS Council (CONAVIHSIDA). Data were collected from patients receiving care at two HIV clinics. The study team and CONAVIHSIDA partners selected the two government-operated clinics with similar staff composition and in comparable regions with high HIV prevalence, but sufficiently distant to prevent cross-contamination. The clinics had between 500 and 800 patients on ART and approximately half of their patients were women.
Eligibility criteria included: i) being at least 18 years old; ii) registered at the clinic for 6 months or more; iii) detectable HIV viral load; iv) sole household member joining ProMeSA; v) urban or peri-urban residence within clinic provinces; vi) moderate to severe household food insecurity screened via the Latin American and Caribbean Food Security Scale; 39,40 vii) on ART for at least six months; viii) Spanish fluency.
Clinic staff screened and referred eligible and interested patients to study coordinators, who enrolled participants and collected most study data. The baseline survey was administered after obtaining written consent, excluding CD4 count and viral load, which were collected by phlebotomists at the HIV clinics (per usual procedures) and assessed by the DR's National Laboratory of Public Health.
## Measures
Dependent variables: Depression severity was assessed with the 9-item Patient Health Questionnaire (Kroenke, Spitzer, and Williams 2001a and 2001b), based on the Diagnostic Statistical Manual of Mental Disorders, 4 th Edition. 41 Each item measured symptom frequency in the past two weeks on a 0 to 3 scale (not at all to nearly every day). Total scores ranged from 0-27 (Cronbach's alpha=0.81). Higher values indicated higher depression severity.
Internalized HIV stigma was measured using an 8-item scale adapted for the DR. 27,42 Items such as "being HIV positive makes you feel a bad person" were rated on a 0 (strongly disagree) to 3 (strongly agree) scale. The sum score (range 0-32; Cronbach's alpha =0.84) represented the level of internalized stigma, with higher scores indicating higher stigma.
Experienced stigma was measured using 11 items adapted from Donastorg, et al., (2014) such as "have you felt excluded from family reunions because you are HIV positive?" using a yes/no response format (Cronbach's alpha = 0.81). 27 Items were summed and total scores ranged from 0 to 11, with higher scores indicating more experienced stigma.
Independent variables: Intimate partner violence (IPV) was assessed using two questions adapted from Mendoza et al. (2017), 43 and the World Health Organization Violence Against Women. 44 These questions, applicable to all participants regardless of gender identity, assessed experience of physical violence by a sexual partner, and experience of other various types of abuse by a sexual partner (e.g., mental, emotional, verbal, financial) in the past six months. Responses to these questions were either yes or no, and a response of yes to either question was used to represent the presence of IPV (victim of violence or mistreatment = 1).
Institutional support was measured by asking participants if they received various types of support, including a conditional cash transfer program, gas or electricity vouchers, subsidized national health insurance, and other forms of support. Responses were either yes or no, and yes responses were added. The resulting sum reflected the quantity of support types received, with a possible range from one to four.
## Social support was assessed using a modified version of the Medical Outcomes Study Social
Support Survey (MOS-SS). 45 Six items were derived from the MOS-SS (items 1 through 4, 11, and 13), while three items were based on our formative research and existing DR studies, including financial assistance, support with picking up HIV medications, and satisfaction with support received. 7,27 Items were rescaled and the summative score divided by the nine items ranged from 0 (no support) to 20 (maximum level of support possible) (Cronbach's alpha = 0.76) HIV viral load and CD4. Blood was drawn for the study to assess both HIV viral load and CD4 cell count and processed at the Dominican National Laboratory of Public Health using the Roche Cobas HIV-1 assay (Roche Molecular Diagnostics, Branchburg NJ, USA). Detectable viral load was defined as ≥ 20 copies of HIV per mL of blood.
Physical health was evaluated using the Global Physical Health subscale from the Patient Reported Outcome Measurement Information System. 46,47 The subscale consists of three items on a 5-pt Likert scale, along with one item rescaled to a 5-pt Likert scale. The four items were summed, and higher scores indicated better physical health (range 4-20; Cronbach's alpha = 0.43).
HIV medication adherence was assessed by asking how rigorously the participant adhered to the HIV medication in the prior seven days. A binary variable was then established, with perfect adherence being coded as 1 and 0 otherwise. 48,49 Demographic factors assessed included age, Haitian background (1=Haitian ancestry; 0=Dominican only), poverty (household income <5000 pesos/month), 50 and gender identity (man, woman, transgender man, transgender woman, other). All individuals self-identified as women or men, thus a binary gender variable was created.
## Data analysis
The analysis was carried out in STATA 14.2 and 15. Descriptive statistics, including means, frequencies, and standard deviations, were reported for the entire sample and by gender. Bivariate relationships among study variables were examined using Pearson correlations. Multivariate linear regression was performed to estimate the independent associations between each factor and the three outcomes (i.e., internalized HIV stigma, experienced HIV stigma, depression). Robust standard errors were employed for sensitivity analysis. Statistical significance was set at p<0.05, and no missing responses were observed from the analysis.
## RESULTS
## Sample characteristics
The study screened 345 patients between May 2018 and January 2019-213 in Mao, 132 in La Vega. Of these, 138 were deemed eligible and 115 patients enrolled and completed baseline assessments-52 in Mao, 63 in La Vega. Table 1 shows that at least half of study participants met the clinical criteria for moderate to severe depression, with some gender differences. Women reported higher average levels of internalized stigma (20.2 vs 16.8; p<0.01), more experienced stigma events (70% of women vs 51% of men endorsed two or more types of experienced stigma, p<0.05), and worse average physical health (11.8 vs 13.8, p<0.001), compared to men.
## Internalized stigma
In bivariate analysis (Table 2), higher internalized HIV stigma was related to higher experienced stigma (ρ=0.41; p<0.001) and depression severity (ρ=0.33; p<0.001). Higher internalized HIV stigma was also associated with being a woman (ρ=0.3; p<0.01), experiencing IPV (ρ=0.3; p<0.01), having a Haitian background (ρ=0.24; p<0.05), having a detectable viral load (ρ=0.19; p<0.05), and higher CD4 count (ρ=0.15; p<0.1). Negative relationships were identified between internalized stigma and social support (ρ=-0.18; p<0.1), institutional support (ρ=-0.21; p<0.05), and physical health (ρ=-0.21; p<0.05).
In multivariate analysis including all potential correlates (Table 3), self-identifying as a woman was associated with a 2.1-point increase in the internalized stigma scale compared to men (β = 2.13, p<0.05). Having experienced intimate partner violence or mistreatment (β = 3.40, p< 0.01), having a detectable viral load (β = 3.38, p<0.01), and Haitian background (β = 3.01, p<0.05) were associated with more internalized stigma, while institutional support was associated with less internalized stigma (β = -1.32, p<0.05). The association between CD4 count and internalized stigma was positive but small: an increase in CD4 count of 100 cells/mm 3 was associated with a 0.6-point increase in the internalized stigma scale (p<0.01).
## Experienced stigma
In bivariate analysis (Table 2), besides being positively correlated with internalized stigma (ρ=0.41; p<0.001), greater levels of experienced stigma were related to higher depression severity score (ρ=0.46; p<0.001). Regarding correlations with predictors in bivariate analyses, greater experienced stigma was related to being a victim of violence or mistreatment (ρ=0.32; p<0.001), less social support (ρ=-0.25; p<0.01), worse physical health (ρ=-0.2; p<0.05), and younger age (ρ=-0.16; p<0.1).
In multivariate analysis, only three covariates remained statistically significant. Violence or mistreatment from a sexual partner was associated with a 1.6-point increase in the experienced stigma scale (β = 1.55, p<0.05). Social support had a small negative association with experienced stigma; a 10-point increase in social support was associated with a 1.6point reduction in the experienced stigma scale (β = -0.16, p<0.05). CD4 count had a small positive association with experienced stigma; an increase of 100 CD4 count cells/mm 3 was associated with 0.2-point increase in the experienced stigma scale (p<0.05).
## Depression
In bivariate analysis (Table 2), depression severity was correlated with the other outcome variables and higher depression severity was associated with having experienced violence or mistreatment (ρ=0.30; p<0.01) and having a detectable viral load (ρ=0.18; p<0.1). Higher depression symptom severity was related to less institutional support (ρ=-0.21; p<0.05), worse physical health (ρ=-0.41; p<0.001), inconsistent ART Adherence (ρ=-0.20; p<0.05), and younger age (ρ=-0.3; p<0.01). However, in the multivariate analysis only three variables remained statistically significant. Having a detectable viral load had an associated 3-point increase in the depression score (β = 2.97, p<0.05), while better physical health (β = -0.71, p<0.001), and taking ART perfectly in the prior seven days (β = -2.71, p<0.05) were associated with less depression.
Regression diagnostics did not show notable multicollinearity, influential points, or outliers in all regression fittings. Results from sensitivity analysis using robust standard errors yielded very similar results with slightly weakened statistical significance levels, which were expected due to the lack of power by the robust standard error technique.
## DISCUSSION
The findings from this study increase our understanding of the factors associated with internalized stigma, experienced stigma, and depression among PLHIV and food insecurity in the understudied Caribbean region. Our study, which included both men and women, expands on prior research that focused exclusively on women in the DR and enables a clearer understanding of the differences by gender. We found that HIV stigmas and depression were highly prevalent in our sample, and they were intercorrelated, while characteristics reflecting greater social vulnerability, including self-identifying as a woman and Haitian background, were associated with higher internalized stigma. We also found that higher internalized stigma was associated with HIV disease progression, while IPV and lack of supports were important factors for HIV stigmas but not for depression. Moreover, depression severity was heightened by HIV disease progression, poorer physical health, and inconsistent ART adherence.
In this study, most participants met the clinical criteria for moderate to severe depression and nearly all endorsed at least one experienced HIV stigma item, which aligns with previous ranges identified for depression in PLHIV, 12 and the documented discrimination toward PLHIV in the DR. 5 Multivariate analysis also showed that being a woman was significantly associated with higher internalized HIV stigma, which aligns with previous literature describing how stigmas disproportionately affect women living with HIV and food insecurity in the DR. 7,27,33,34 Cultural norms in the DR, which often perceive women as submissive and dependent on men, contribute to discrimination against women living with HIV. 33 This discrimination, experienced in employment, healthcare, and communities, may lead to internalized stigma among these women. 7,27 Further research is needed to understand why experienced stigma and depression did not have a significant association with gender in multivariate analysis. Qualitative research suggests that both men and women living with HIV in the DR experience discrimination due to their HIV status, 51 future research should explore how discrimination experiences vary by gender.
The finding that having a Haitian background was significantly associated with higher internalized stigma aligns with literature documenting anti-Haitianism in the DR, 33 where Haitian immigrants are often employed in low-wage or informal jobs and subjected to systemic discrimination and rights violations. The social devaluation attached to Haitian identity may therefore compound the stigma of living with HIV in the country. 33 The multivariate analysis also showed a positive association between violence or mistreatment from a sexual partner and with internalized and experienced stigmas, which is consistent with literature on HIV-related violence. 35 Previous studies also highlighted IPV as a risk factor for HIV, 52 while research from the DR has suggested that such violence is commonly experienced by women with HIV, 7 and that it can affect HIV treatment outcomes, including care retention and ART adherence. 43 The connection between IPV and HIV stigmas results since IPV increases self-blame, shame, and embarrassment-elements of internalized stigma. Additionally, people who experience violence may face external stigma and judgement since they are often blamed for staying in abusive relationships. 53 Further investigation of IPV among PLHIV in contexts like the DR is crucial to inform interventions reducing violence and stigma and improve HIV outcomes. This analysis also found that institutional and social supports were negatively associated with internalized and experienced stigmas, respectively, which is consistent with literature on institutional support in the form of cash transfers addressing the drivers of HIV, such as poverty 54 and on social support improving the mental health and decreasing stigma. 3,18 Previous studies from the DR highlighted limitations in HIV clinic support and emphasized the importance of psychosocial support for PLHIV. 26,55 In practice, interventions may involve referring patients for support during routine healthcare visits and offering counseling to patients and their networks. This counseling could tackle the factors linked to stigma and depression by this study, such as racial discrimination, gender inequalities, intimate violence, physical health, and ART adherence.
Interventions such as mass communication campaigns or working with community leaders could also be used to disseminate HIV information, sensitize the public, and raise awareness about HIV-related stigmas and depression. Communication campaigns have proven effective at improving depression knowledge and decreasing stigma, 56 while engaging faith leaders has helped reduce HIV stigma, mistrust, and increased knowledge. 33,57 Working with faith leaders, could be useful for this population given the high religious affiliation of people in the DR-at least 75% of the population had a religious affiliation in 2020. 58
## Limitations
Limitations of this study included self-reported outcomes, predictors and controls, except for viral load and CD4 count which were obtained through blood samples. Self-reported variables are prone to social desirability biases and inaccurate recall, though we did use established and validated measures. Despite the limited sample size, we identified statistically significant associations between explanatory factors and outcomes. Causality could not be established due to the cross-sectional dataset; thus, we relied on associations between variables.
While gender was included as a covariate and bivariate comparisons by gender were presented, the sample size (n = 115) limited the power to conduct gender-stratified multivariate analyses. This constrained our ability to assess whether the associations between IPV, stigma, and depression differed by gender identity. Future studies with larger samples are needed to explore gender-specific pathways in greater depth.
## CONCLUSIONS
Our study showed a strong prevalence of HIV stigmas and depression among food insecure PLHIV in the DR. Internalized stigma was associated with self-identifying as a woman, Haitian background, and HIV progression. IPV and lack of supports influenced both HIV stigmas, while depression was aggravated by disease progression, poor physical health, and inconsistent ART adherence. Future studies could examine the ways in which experienced stigma and depression manifestations differ by gender. HIV stigmas and depression are obstacles to HIV prevention and mitigation efforts. 33 Therefore, interventions addressing stigma and depression in people living with HIV and food insecurity are needed to improve mental health, HIV-related health outcomes, and overall well-being. This study, alongside existing research, suggests the importance of multilevel interventions that address the cultural and social factors that create stigmatizing conditions for PLHIV and food insecurity. 32
## Bivariate correlations among all outcomes and independent variables
## References
1. "HIV Stigma and Discrimination"
2. Quinn, Earnshaw (2013) "Concealable stigmatized identities and psychological well-being" *Soc Personal Psychol Compass*
3. Takada, Weiser, Kumbakumba (2014) "The dynamic relationship between social support and HIV-related stigma in rural Uganda" *Ann Behav Med*
4. Hasan, Nath, Khan et al. (2012) "Internalized HIV/AIDS-related stigma in a sample of HIV-positive people in Bangladesh" *J Health Popul Nutr*
5. (2015) "On the Fast-Track to End AIDS by 2030, Focus on Location and Population"
6. Katz, Ryu, Onuegbu (2013) "Impact of HIV-related stigma on treatment adherence: systematic review and meta-synthesis" *J Int AIDS Soc*
7. Derose, Payán, Fulcar (2017) "Factors contributing to food insecurity among women living with HIV in the Dominican Republic: A qualitative study" *PloS One*
8. Palar, Martin, Camacho et al. (2013) "Livelihood experiences and adherence to HIV antiretroviral therapy among participants in a food assistance pilot in Bolivia: a qualitative study" *PloS One*
9. Katz, Ryu, Onuegbu (2013) "Impact of HIV-related stigma on treatment adherence: systematic review and meta-synthesis" *J Int AIDS Soc*
10. (2020) "The Well Project. Stigma and Discrimination Against Women Living with HIV. Published online February"
11. Tran, Ho, Ho (2019) "Depression among patients with HIV/AIDS: research development and effective interventions (GAPRESEARCH)" *Int J Environ Res Public Health*
12. Ciesla, Roberts (2001) "Meta-analysis of the relationship between HIV infection and risk for depressive disorders" *Am J Psychiatry*
13. Uthman, Magidson, Safren et al. (2014) "Depression and adherence to antiretroviral therapy in low-, middle-and high-income countries: a systematic review and meta-analysis" *Curr HIV/AIDS Rep*
14. Blashill, Perry, Safren (2011) "Mental health: A focus on stress, coping, and mental illness as it relates to treatment retention, adherence, and other health outcomes" *Curr HIV/AIDS Rep*
15. Parker, Aggleton (2003) "HIV and AIDS-related stigma and discrimination: a conceptual framework and implications for action" *Soc Sci Med*
16. Rueda, Chen (2016) "Examining the associations between HIV-related stigma and health outcomes in people living with HIV/AIDS: a series of meta-analyses" *BMJ Open*
17. Palar, Frongillo, Escobar (2018) "Food insecurity, internalized stigma, and depressive symptoms among women living with HIV in the United States" *AIDS Behav*
18. Tsai, Bangsberg, Frongillo (2012) "Food insecurity, depression and the modifying role of social support among people living with HIV/AIDS in rural Uganda" *Soc Sci Med*
19. Isaac, Jacobson, Wanke et al. (2008) "Declines in dietary macronutrient intake in persons with HIV infection who develop depression" *Public Health Nutr*
20. Maynard, Andrade, Packull-Mccormick et al. (2018) "Food insecurity and mental health among females in high-income countries" *Int J Environ Res Public Health*
21. Aibibula, Cox, Hamelin et al. (2017) "Association Between Food Insecurity and HIV Viral Suppression: A Systematic Review and Meta-Analysis" *AIDS Behav*
22. Au, Kayitenkore, Shutes (2006) "Access to adequate nutrition is a major potential obstacle to antiretroviral adherence among HIV-infected individuals in Rwanda" *AIDS*
23. Unaids, Aidsinfo (2019)
24. Kalichman, Simbayi, Cloete et al. (2009) "Measuring AIDS stigmas in people living with HIV/AIDS: the Internalized AIDS-Related Stigma Scale" *AIDS Care*
25. Logie, Gadalla (2009) "Meta-analysis of health and demographic correlates of stigma towards people living with HIV" *AIDS Care*
26. Barrington, Kerrigan, Ureña et al. (2018) "La vida normal: living with HIV" *Cult Health Sex*
27. Donastorg, Barrington, Perez et al. (2014) "Abriendo Puertas: baseline findings from an integrated intervention to promote prevention, treatment and care among FSW living with HIV in the Dominican Republic" *PloS One*
28. Rael, Davis (2017) "Depression and key associated factors in female sex workers and women living with HIV/AIDS in the Dominican Republic" *Int J STD AIDS*
29. Rael, Hampanda (2016) "Understanding internalized HIV/AIDS-related stigmas in the Dominican Republic: a short report" *AIDS Care*
30. Alaimo, Chilton, Jones "Chapter 17 -Food insecurity, hunger, and malnutrition"
31. Kinyanda, Salisbury, Muyingo (2020) "Major Depressive Disorder Among HIV Infected Youth in Uganda: Incidence, Persistence and Their Predictors" *AIDS Behav*
32. Stangl, Earnshaw, Logie (2019) "The Health Stigma and Discrimination Framework: a global, crosscutting framework to inform research, intervention development, and policy on healthrelated stigmas" *BMC Med*
33. Arregui (2007) "Living with HIV in the Dominican Republic" *Rev Interam Psicol J Psychol*
34. Payán, Derose, Fulcar et al. (2019) "It Was as Though My Spirit Left, Like They Killed Me": The Disruptive Impact of an HIV-Positive Diagnosis among Women in the Dominican Republic" *J Int Assoc Provid AIDS Care JIAPAC*
35. Deering, Braschel, Logie (2020) "Exploring pathways from violence and HIV disclosure without consent to depression, social support, and HIV medication self-efficacy among women living with HIV in Metro Vancouver" *Canada. Health Psychol Open*
36. Kelso-Chichetto, Okafor, Cook et al. (2018) "Association Between Depressive Symptom Patterns and Clinical Profiles Among Persons Living with HIV" *AIDS Behav*
37. Kerrigan, Vazzano, Bertoni et al. (2017) "Stigma, discrimination and HIV outcomes among people living with HIV in Rio de Janeiro, Brazil: the intersection of multiple social inequalities"
38. Seth, Kidder, Pals (2014) "Psychosocial functioning and depressive symptoms among HIV-positive persons receiving care and treatment in Kenya, Namibia, and Tanzania" *Prev Sci*
39. Melgar-Quiñonez, Uribe, Centeno (2010) "Características psicométricas de la escala de seguridad alimentaria ELCSA aplicada en Colombia" *Guatemala y México. Segur Aliment E Nutr*
40. Científico De La Elcsa (2012) "Escala Latinoamericana y Caribeña de Seguridad Alimentaria (ELCSA): Manual de Uso y Aplicaciones"
41. (1994) "Diagnostic Statistical Manual of Mental Disorders"
42. Berger, Ferrans, Lashley (2001) "Measuring stigma in people with HIV: Psychometric assessment of the HIV stigma scale" *Res Nurs Health*
43. Mendoza, Barrington, Donastorg (2017) "Violence From a Sexual Partner is Significantly Associated With Poor HIV Care and Treatment Outcomes Among Female Sex Workers in the Dominican Republic" *JAIDS J Acquir Immune Defic Syndr*
44. Schraiber, Latorre, Rdo et al. (2010) "Validity of the WHO VAW study instrument for estimating gender-based violence against women" *Rev Saude Publica*
45. Sherbourne, Stewart (1991) "The MOS social support survey" *Soc Sci Med*
46. Hays, Bjorner, Revicki et al. (2009) "Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items" *Qual Life Res*
47. Orthotoolkit (2019) "PROMIS Global-10 Score Sheet. General Health"
48. Chesney, Ickovics, Chambers (2000) "Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: The AACTG Adherence Instruments" *AIDS Care*
49. Reynolds, Sun, Nagaraja et al. (1999) "Optimizing Measurement of Self-Reported Adherence With the ACTG Adherence Questionnaire" *J Acquir Immune Defic Syndr*
50. (2020) "Poverty and Equity Brief: Dominican Republic"
51. Payan, Armenta (2023) "Exploring gender differences in HIV-related stigma and social support in a low-resource setting: A qualitative study in the Dominican Republic" *Plos One*
52. Campbell, Baty, Ghandour et al. (2008) "The intersection of intimate partner violence against women and HIV/AIDS: a review" *Int J Inj Contr Saf Promot*
53. Overstreet, Quinn (2013) "The Intimate Partner Violence Stigmatization Model and Barriers to Help Seeking" *Basic Appl Soc Psychol*
54. "Discussion paper: cash transfers and HIV-prevention"
55. Program (2014)
56. Brooks, Laws, Wilson (2014) "Allocating HIV treatment to the adherent: a qualitative study of patient perceptions of their HIV care in the Dominican Republic" *J Health Care Poor Underserved*
57. Unger, Cabassa, Molina et al. (2013) "Evaluation of a Fotonovela to Increase Depression Knowledge and Reduce Stigma Among Hispanic Adults" *J Immigr Minor Health Cent Minor Public Health*
58. Derose, Griffin, Kanouse (2016) "Effects of a pilot church-based intervention to reduce HIV stigma and promote HIV testing among African Americans and Latinos" *AIDS Behav*
59. Pública (2020) "Latinobarómetro 2020 Dominican Republic, Documento de Resultados" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12846516&blobtype=pdf | # The Role of Prior HBV Infection on the Efficacy of 3TC/DTG as a Maintenance Therapy
Tommaso Matucci, Sara Occhineri, Alessandra Palomba, Maria Vatteroni, Laura Bono, Marina Polidori, Riccardo Iapoce, Alberto Borghetti, Marco Falcone
## Abstract
Lamivudine/dolutegravir (3TC/DTG) is an effective and well-tolerated antiretroviral regimen for most people with HIV (PWH) who are virologically suppressed; however, specific clinical characteristics, such as prior hepatitis B virus (HBV) exposure or archived resistanceassociated mutations (RAMs), may influence the risk of virological failure (VF). We conducted a retrospective, monocentric cohort study to evaluate the incidence and predictors of VF among PWH who switched to 3TC/DTG after achieving virological suppression (HIV-RNA < 50 copies/mL). A total of 188 PWH were included. Over 5082 patient-years of follow-up (PYFU), 8 individuals (4.3%) experienced VF, corresponding to an incidence rate of 1.45 per 1000 PYFU. The cumulative probabilities of VF at 1, 2, 3, 4, and 5 years were 0.6%, 2.7%, 2.7%, 4.2%, and 22.3%, respectively. In exploratory multivariable analyses, anti-HBc positivity was associated with an increased risk of VF (adjusted hazard ratio [aHR] 4.80, 95% CI 1.03-22.43; p = 0.046). After adjustment for age and sex, individuals with anti-HBc positivity who had switched from a tenofovir-containing regimen showed the highest risk of VF compared with anti-HBc-negative individuals without prior tenofovir exposure (aHR 15.06,; p = 0.025). Given the limited number of virological events, these findings should be interpreted with caution. Nevertheless, they suggest that prior HBV exposure, particularly in the context of tenofovir discontinuation, may represent a clinically relevant factor when considering simplification to 3TC/DTG.
## 1. Introduction
Dual antiretroviral therapy (ART) with dolutegravir (DTG) and lamivudine (3TC) represents an effective and well-tolerated treatment option both as first-line therapy and as a simplification strategy for People living with HIV (PWH) [1]. Nevertheless, several limitations have been identified by international treatment guidelines [1,2]. In ARTnaïve individuals, contraindications include high baseline viral load and prior failure of tenofovir/emtricitabine-based pre-exposure prophylaxis (PrEP). In treatment-experienced patients, the use of 3TC/DTG is generally recommended in the absence of resistance to nucleoside reverse transcriptase inhibitors (NRTIs) or integrase strand transfer inhibitors (INSTIs) [1,2]. Recent updates to the European AIDS Clinical Society (EACS) guidelines have expanded the potential use of this regimen to individuals with archived M184V mutations, provided that virological suppression is maintained [1]. Conversely, current European and North American guidelines for the management of HIV [1,2] and hepatitis B virus (HBV) infection [3][4][5] advise against the use of dual therapy in the presence of active HBV infection, defined by HBsAg positivity. The interaction between HIV and HBV is well documented and is associated with poorer immunological recovery, increased immune activation, and higher rates of virological failure and liver-related complications [6][7][8][9]. Although isolated anti-HBc positivity is not considered an absolute contraindication to dual therapy, emerging evidence suggests that this serological profile may also be associated with inferior virological outcomes with 3TC/DTG [6]. The aim of the present study was therefore to evaluate the incidence and predictors of virological failure in a real-world cohort of PWH treated with 3TC/DTG, with particular focus on the role of prior HBV exposure as assessed by anti-HBc serostatus.
## 2. Materials and Methods
We conducted a retrospective cohort study at a tertiary referral university hospital in Pisa, Italy. Adult (≥18 years) PWH were eligible if they were receiving stable ART and switched to 3TC/DTG after achieving virological suppression, defined as HIV-RNA < 50 copies/mL in at least one determination prior to the switch. Eligible switches occurring between 1 January 2015 and 31 December 2023 were included. Exclusion criteria were positive HBsAg serostatus at baseline, prior exposure to 3TC/DTG, and the absence of at least one post-switch HIV-RNA determination. Baseline was defined as the date of switch to 3TC/DTG.
The primary endpoint was time to virological failure (VF), defined as the earliest occurrence of one of the following events: (a) the first of two consecutive HIV-RNA values > 50 copies/mL; (b) a single HIV-RNA ≥ 200 copies/mL, or (c) a single HIV-RNA between 50 and 200 copies/mL followed by treatment discontinuation or intensification. Participants were censored at treatment discontinuation (TD) not meeting VF criteria, at the date of last available viral load measurement or loss-to-follow-up (defined as the absence of any HIV-RNA determination for more than 12 months). HIV-RNA determinations were performed according to routine clinical practice.
Probability of time to VF was assessed using Kaplan-Meier estimates. Predictors of VF were evaluated using Cox proportional hazards models: variables associated with the outcome at univariable analysis (p < 0.05) were included in a multivariable Cox regression model. Potential predictors of VF included demographic characteristics (age, gender, risk factor for HIV acquisition), HIV-related variables (time since HIV diagnosis, time of continuous virological suppression at BL, zenith HIV-RNA, nadir and BL CD4 count, residual viremia at BL, genotypic susceptibility score for lamivudine based on historical genotype and calculated with the Stanford algorithm, version 9.8) and antiretroviral therapy prior to switch. The role of previous HBV infection, defined by the presence of anti-HBc antibodies in the absence of HBsAg positivity, was also investigated. Missing data were limited for most variables, except for resistance-associated mutations, which were unavailable for approximately 27% of participants due to the absence of historical genotypic resistance testing. Given the non-random nature of this missingness, multiple imputation was not performed. Instead, for the multivariable Cox regression, a sensitivity analysis was conducted excluding individuals with missing genotypic data or with documented lamivudine resistance mutations (n = 5).
A post hoc analysis was conducted to assess whether occult HBV infection had a causal role in time to VF. Differences in the demographic and viro-immunological characteristics of PWH with and without anti-HBcAg positivity were assessed using the Chi-square test for categorical variables and Student's t-test for continuous variables, to explore potential confounders of anti-HBcAg serostatus. A multivariable Cox regression model was fitted to assess the independent role of anti-HBc positivity on VF (potential confounders were chosen among variables that showed a statistically different distribution in the population with positive and negative anti-HBcAg serostatus, at a p-value < 0.05).
Clinical and treatment-related information was retrospectively collected from paperbased medical records. Laboratory data, including HIV-RNA measurements and immunological parameters, were obtained from the institutional electronic laboratory database. All analyses were performed with Stata Statistical Software (StataCorp, 2017. Stata Statistical Software: Release 15. College Station, TX, USA: StataCorp LLC).
## 3. Results
A total of 188 patients were included in our study: most were male (141, 75%), Caucasian (174, 92.5%), and with a median age of 54 years (IQR 44-61 years). Sexual transmission was the most common route of HIV acquisition, with similar proportions of individuals reporting heterosexual (HET) (70, 37.2%) and same-sex sexual contacts (84, 44.7%). The median time since HIV diagnosis was 11 years (IQR 5-17 years), with a median of 9 years since ART initiation (IQR 5-16 years), and 5 years of virological suppression (IQR 3-9). Thirty-five PWH (18.6%) had a history of a previous AIDS-defining condition. Complete characteristics of the study population are summarized in Table 1.
The reason for the shift to dual therapy was a proactive switch in all cases, with 129 (68.6%) patients switching from a 2NRTIs + INSTI regimen. One hundred and twenty-eight PWH (68.1%) had experienced at least one previous VF, and 127 (67.6%) were previously exposed to five or fewer regimens.
Among participants with at least one genotypic resistance test available before BL (137, 72.9%), 14 (10.1%) had at least one RAM to NRTIs; RAMs to 3TC were detected in 5 (3.6%) participants, including one case (0.7%) with an M184V mutation.
Concerning HBV serostatus, most patients were HBV-seronegative (74, 39.4%). Fiftyone (27.1%) had isolated anti-HBsAg positivity, and 35 (18.6%) were positive for both anti-HBsAg and anti-HBcAg. Ten patients (5.3%) had isolated positivity for anti-HBcAg, and 18 (9.6%) had an unknown HBV serostatus.
Nineteen (10.1%) patients had a positive HCV serostatus. Sixteen patients discontinued treatment with 3TC/DTG after a median time of 27 months (2.87 per 1000 patient-years of follow-up or PYFU). Reasons for TD included the following: VF (6/16, 37.5%), switch to a long-acting regimen (2/16, 12.5%), drug-related toxicity (5/16, 31.3%), and other/unknown causes (3/16, 18.7%).
VF occurred in eight patients (1.45 per 1000 PYFU) (Figure 1). The estimated probability of VF was 0.6% (95% CI 0.1-0.4) at 12 months, 2.7% (95% CI 1.0-7.1) at 24 and 36 months, 4.2% (95% CI 1.7-10.7) at 48 months, and 22.3% (95% CI 7.9-54.3) at 60 months (Figure 1).
Factors associated with VF at univariable analysis were as follows: baseline HIV-RNA between 20 and 49 copies/mL (versus < 20 copies/mL, HR 5.67, 95% CI 1.10-9.39, p = 0.039), a higher genotypic susceptibility score (based on Stanford algorithm, version 9.7) for 3TC (per 10-point increase, HR 1.74, 95% CI 1.23-2.48, p = 0.002), and positive anti-HBcAg serostatus (versus negative, HR 5.76, 95% CI 1.26-26.24; p = 0.024). Given the low number of events, a multivariable Cox model was fitted only including anti-HBc serostatus (positive versus negative, aHR 4.80, 95% CI 1.03-22.43, p = 0.046) and baseline HIV-RNA (20 and 49 copies/mL versus < 20 copies/mL, aHR 5.27, 95% CI 0.81-34.37, p = 0.082). As a sensitivity analysis, we excluded people with resistance mutations to 3TC (5 persons, with only one case of previously detected M184V) and with unknown resistance history (129 people were included in the final model, with 5 VFs): anti-HBcAg positivity was still associated with increased risk of VF at multivariable analysis (aHR 13.21, 95% CI 1.35-129.37; p = 0.027). Table 2 summarizes the associations among other potential predictors and the virological outcome (Table 2).
Among PWH with and without anti-HBcAg-positive serostatus, gender and age were significantly different, with a higher proportion of older men in the anti-HBcAgpositive group. Interestingly, fewer PWH in the anti-HBcAg-positive group switched from a tenofovir-containing strategy (see Table 3). After stratifying anti-HBcAg serostatus by prior tenofovir exposure, we found a 15-fold higher risk of VF in PWH with anti-HBcAg positivity and previous exposure to tenofovir (versus negative anti-HBcAg and no prior tenofovir use, aHR 15.06, 95% CI 1.40-161.38; p-value = 0.025) (Table 4). The effect of anti-HBcAg positivity was markedly reduced in those not switching from a tenofovir-based therapy and did not reach statistical significance (Table 4).
## 4. Discussion
The dual ART regimen with 3TC/DTG proved highly effective in our cohort of virologically suppressed PWH undergoing ART optimization. The most recent European guidelines [1] highlight the possibility of switching to 3TC/DTG in individuals with prior VF and/or in the presence of M184I/V mutations. Regarding HBV serological status, switching is recommended in the presence of anti-HBs antibodies, although no absolute contraindications are currently provided for isolated anti-HBcAg positivity [1,2].
In our study, anti-HBcAg positivity was associated with a higher risk of VF. Moreover, we found a 15-fold higher risk for VF among those anti-HBcAg seropositive who discontinued tenofovir.
To date, no clear evidence has demonstrated an increased risk of VF with 3TC/DTG in the context of prior HBV infection. However, a large study from the Italian ICONA cohort by Malagnino et al. reported an excess of risk of VF in individuals with isolated anti-HBcAg positivity, regardless of ART regimen, although this risk remained lower than in those with active HBV infection [7]. In another analysis by the same group, significantly fewer anti-HBcAg-positive PWH achieved target undetected HIV-RNA levels, compared with anti-HBcAg-negative individuals, after switching to 3TC/DTG; moreover, anti-HBcAg positivity was the only factor associated with suboptimal HIV suppression [10]. Conversely, a similar study conducted in China involving 601 PWH switching to 3TC/DTG found no differences in the proportion of PWH with undetectable viremia after 24 months [11]. Finally, in another Italian cohort of 606 virologically suppressed PWH switching to 3TC/DTG, no significant differences by HBV serostatus were observed in the risk of VF or viral blips; however, the effect of HBV serology was not adjusted for potential confounders [12].
Another unresolved issue concerns the lack of a confirmed biological mechanism explaining the reduced efficacy of ART in the setting of prior HBV exposure. Recently, cryptic serum HBV-DNA replication was demonstrated in a cohort of anti-HBc-positive/HBsAgnegative PWH despite ongoing tenofovir exposure [13]. After switching to a tenofovirsparing regimen (mostly 3TC-based), the proportion of PWH with HBV-DNA>10 IU/mL increased from 12.9% at T1 to 42.6% at T2 and was predicted by a lower nadir CD4 count and the presence of cryptic HBV-DNA at baseline. Although HIV virological failure was not specifically reported during follow-up, these findings support a previously proposed hypothesis [10,14] that suggested a synergistic viral interplay between HIV and HBV coinfection, whereby reduced antiviral drug pressure or the development of HBV resistance to 3TC may allow HBV rebound or low-level replication which in turn could promote HIV transcription through HBx activity [14].
The relevance of archived resistance associated mutations (RAMs) to 3TC in influencing the effectiveness of 3TC/DTG is still debated due to conflicting data [15][16][17][18][19][20][21][22][23]. In the present study, a higher risk of VF was also predicted by the presence of 3TC-associated RAMs in previous genotypic tests, although such mutations were overall rare, with M184V being reported in only one case. This topic has been the subject of extensive debate in recent years, with no definite evidence of a causal effect of the main 3TC resistance mutation, M184I/V, on virological outcomes. In the pilot ART-PRO clinical trial, no increased risk of VF was observed at week 144 in PWH with prior M184I/V, provided that the mutation was absent in the HIV-DNA genotypic test at screening [24]. More recently, the SOLAR-3D clinical trial, which included participants with and without a history of M184I/V (some with detectable mutations at baseline HIV-DNA testing), confirmed the absence of differences in rates of virological suppression and viral rebound at week 144 [25]. In contrast, one retrospective study reported that M184I/V increased the risk of viral rebound, particularly when combined with at least one thymidine analogue mutation (TAM), regardless of the duration of prior virological suppression [26]. Moreover, an emulated trial from the Italian ARCA cohort found a higher risk of failure with the dual regimen when the switch occurred within six months of virological suppression and in the presence of historical RAMs (either TAMs or isolated M184I/V), with a non-significant trend towards superior efficacy of triple therapy in this context [20]. Despite these discrepancies, which may partly reflect differences in study populations and methodologies, it is not possible to completely rule out an effect of historical RAMs on the efficacy of dual therapy. Caution therefore remains advisable when considering this strategy in individuals with a history of virological failure, even though the overall risk of failure, especially with resistance development, remains very low [27].
Our study has several limitations, primarily related to the low incidence of virological outcomes that makes the predictive model potentially overfitted, even if a parsimonious multivariable Cox model was chosen. The retrospective design also limited the possibility to capture relevant data, such as adherence, and to assess HBV-DNA at baseline and at failure, which could have clarified the role of occult HBV infection as the cause of VF. Finally, the monocentric design limits the generalizability of study results that should therefore be taken with caution.
Despite these limitations, our study partly confirms previous findings regarding the potential impact of prior HBV infection when switching to tenofovir-sparing regimens. Considering that most next-generation treatment strategies will lack HBV activity, further research on this topic is warranted.
## References
1. (2025) "Guidelines Version"
2. (2025) "Panel on Antiretroviral Guidelines for Adults and Adolescents"
3. Ali, Nguyen, Hernaez et al. (2025) "AGA Clinical Practice Guideline on the Prevention and Treatment of Hepatitis B Virus Reactivation in At-Risk Individuals" *Gastroenterology*
4. (2025) "European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatitis B virus infection" *J. Hepatol*
5. (2024) "Guidelines for the Prevention, Diagnosis, Care and Treatment for People with Chronic Hepatitis B Infection; World Health Organization"
6. Sarmati, Malagnino (1077) "HBV Infection in HIV-Driven Immune Suppression" *Viruses*
7. Malagnino, Cozzi-Lepri, Svicher et al. (2024) "Association between markers of hepatitis B virus infection and risk of virological rebound in people with HIV receiving antiretroviral therapy" *HIV Med*
8. Thornton, Jose, Bhagani et al. (2017) "Hepatitis B, hepatitis C, and mortality among HIV-positive individuals" *AIDS*
9. Rajbhandari, Jun, Khalili et al. (2016) "HBV/HIV coinfection is associated with poorer outcomes in hospitalized patients with HBV or HIV" *J. Viral Hepat*
10. Malagnino, Mulas, Teti et al. "HbcAb Positivity as a Risk Factor for Missing HIV RNA Undetectability After the 3TC+DTG Switch" *Viruses*
11. Fu, Biao, Liu et al. (2025) "24-month outcomes after switching to Dolutegravir/Lamivudine in people living with HIV and HbcAb positivity at the Beijing Ditan Hospital in China" *Ann Med*
12. Salvo, Ciccullo, Visconti et al. (2025) "Impact of HBV serological status on HIV virological efficacy of two-drug antiretroviral regimens: A retrospective observational study on virologically suppressed people with HIV switching to lamivudine/dolutegravir" *HIV Med*
13. Salpini, D'anna, Alkhatib et al. (2025) "Kinetics of hepatitis B virus replication in anti-HBc positive/HbsAg-negative people with HIV switching to tenofovir sparing therapy" *Int. J. Infect. Dis*
14. Gómez-Gonzalo, Carretero, Rullas et al. (2001) "The hepatitis B virus X protein induces HIV-1 replication and transcription in synergy with T-cell activation signals: Functional roles of NF-kappaB/NF-AT and SP1-binding sites in the HIV-1 long terminal repeat promoter" *J. Biol. Chem*
15. Rhee, Gonzales, Kantor et al. (2003) "Human immunodeficiency virus reverse transcriptase and protease sequence database" *Nucleic Acids Res*
16. Osiyemi, De Wit, Ajana et al. (2022) "Efficacy and Safety of Switching to Dolutegravir/Lamivudine Versus Continuing a Tenofovir Alafenamide-Based 3-or 4-Drug Regimen for Maintenance of Virologic Suppression in Adults Living with Human Immunodeficiency Virus Type 1: Results Through Week 144 from the Phase 3, Noninferiority TANGO Randomized Trial" *Clin. Infect. Dis*
17. (2026) *Viruses*
18. Underwood, Osiyemi, Rubio et al. (2022) "Archived resistance and response to <40 c/mL & TND-DTG/3TC FDC at week 48 in SALSA"
19. Blick, Cerreta-Dial, Mancini et al. (2024) "SS0403LB No confirmed virological failures (CVF) for 144 weeks when switching 2-/3-/4-drug ART to DTG/3TC in heavily treatment-experienced PLWHA with prior M184V/I and virological failures (VF) in the prospective SOLAR-3D study"
20. Kabra, Barber, Allavena et al. (2023) "Virologic Response to Dolutegravir Plus Lamivudine in People with Suppressed Human Immunodeficiency Virus Type 1 and Historical M184V/I: A Systematic Literature Review and Meta-analysis" *Open Forum Infect. Dis*
21. Borghetti, Ciccullo, Lombardi et al. "Efficacy of Lamivudine Plus Dolutegravir vs. Dolutegravir-Based 3-Drug Regimens in People with HIV Who Are Virologically Suppressed"
22. Gagliardini, Baccini, Modica et al. (2022) "Impact of resistance mutations on efficacy of dolutegravir plus rilpivirine or plus lamivudine as maintenance regimens: A cohort study" *J. Glob. Antimicrob. Resist*
23. Borghetti, Alkhatib, Dusina et al. (2022) "Virological outcomes with dolutegravir plus either lamivudine or two NRTIs as switch strategies: A multi-cohort study" *J. Antimicrob. Chemother*
24. Santoro, Armenia, Teyssou et al. (2022) "Virological efficacy of switch to DTG plus 3TC in a retrospective observational cohort of suppressed HIV-1 patients with or without past M184V: The LAMRES study" *J. Glob. Antimicrob. Resist*
25. De Miguel Buckley, Rial-Crestelo, Montejano et al. "Long-term Evaluation of Residual Viremia in a Clinical Trial of Dolutegravir Plus Lamivudine as Maintenance Treatment for Participants with and Without Prior Lamivudine Resistance" *Open Forum Infect. Dis*
26. Blick, Cerreta-Dial, Mancini et al. (2024) "No confirmed virological failures (CVF) for 144 weeks when switching 2-/3-/4-drug ART to DTG/3TC in heavily treatment-experienced PLWHA with prior M184V/I and multiple virological failures in the prospective SOLAR-3D study"
27. Borghetti, Giacomelli, Borghi et al. (2021) "Nucleoside Reverse-Transcriptase Inhibitor Resistance Mutations Predict Virological Failure in Human Immunodeficiency Virus-Positive Patients During Lamivudine Plus Dolutegravir Maintenance Therapy in Clinical Practice" *Open Forum Infect. Dis*
28. Marcelin, Soulie, Wirden et al. (2025) "Emergent resistanceassociated mutations at first-or second-line HIV-1 virologic failure with second-generation InSTIs in two-and three-drug regimens: The Virostar-1 study" *J. Antimicrob. Chemother*
29. "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=PMC12197373&blobtype=pdf | # Navigating Virology's Frontiers in Africa: Global Virus Network 2024 Durban Meeting
Maggie Bartlett, Rubeshan Perumal, Sten Vermund, Salim Karim
## Abstract
The Global Virus Network (GVN) is a voluntary consortium of virology laboratories and affiliated scientists that seek to prevent and control global viral threats. The meetings of the GVN are characterized by academic, health center, government, and industry participation, sharing information that is designed to further the mutual mission. In September 2024, the meeting in Durban, South Africa, highlighted diseases and investigators from Africa, and paid special attention to pandemic preparedness. Selected highlights from the meeting are presented here, along with a call-to-action in defense of global partnerships for research in the origins of human and animal viruses, the risk to humans from other animal sources, the pathogenesis of given viruses, and their prevention and treatment. Discussions of laboratory discovery science are juxtaposed with development of vaccines, antiviral drugs, immunotherapies, and innovative field strategies for control of viral diseases.
## 1. Introduction
With the goal of amplifying expertise and collaboration globally in virology, the Global Virus Network (GVN) was formed in 2011. The consortium sought to provide virologists with the opportunity to connect, collaborate, and elevate the discipline globally. The GVN hosts a global meeting at one of its ≈85 affiliated Centers of Excellence or Affiliates, selecting the Centre for the AIDS Programme of Research in South Africa (CAPRISA) in Durban as the meeting host on 16 to 18 September 2024 (Figure 1).
The GVN Durban meeting highlighted critical advancements in pathogen detection, immune responses, and pandemic preparedness, while underscoring the necessity of fostering the next generation of virologists and fighting misinformation. Held in Africa, the event emphasized the unique challenges and opportunities that the continent faces, from understanding regional viral variants to addressing disparities in healthcare access. The meeting showcased Africa's growing role in global health, aiming to enhance collaboration, build local research capacity, and ensure that pandemic solutions are inclusive and equitable. Here, we summarize highlights of selected presentations.
## 2. Evading Viral Evasion
Dr. Alan Landay (University of Texas Medical Branch) emphasized the growing necessity for more reliable, integrated, and objective methods to detect pathogens and address the complexity of host responses. He discussed the critical role of bioinformatics and analytic tools in pathogen discovery, particularly in the context of HIV, SARS-CoV-2, and Crimean-Congo hemorrhagic fever virus. Dr. Landay highlighted the uniqueness of biological aging across different organs, pointing to accelerated aging markers in blood from individuals on antiretroviral therapy, where oxidative and glycolytic pathway imbalances play a central role [1,2].
Dr. Sarah Londrigan (The Peter Doherty Institute for Infection and Immunity) delved into the role of macrophages within the airway immune system, particularly in relation to influenza A virus (IAV) infections. Her findings reveal that IAV's infection in macrophages is abortive, where viral ribonucleoproteins (vRNPs) are crucial, yet the macrophages act as a "dead end" for viral propagation [3]. Londgrin's research showed that red blood cells (RBCs) could effectively deplete viral loads due to their glycan-binding capabilities. Her insights on the airway microenvironment offer a novel understanding of how macrophages regulate immune responses to seasonal IAV.
Dr. Vineet Menachery (University of Texas Medical Branch) discussed defective interfering RNA and its often-overlooked role in viral recombination. His team found SARS-CoV-2 to be significantly more recombinogenic than the common cold or MERS. Notably, they discovered microdeletions near UUAU sites that enhance viral recombination, driven by the action of viral non-structural protein 15 (NSP15) [4]. Blocking NSP15 led to reduced infection but did not mitigate the associated pathology. Instead, recombination increased, suggesting that NSP15 may serve as a complex regulatory target in SARS-CoV-2 infections.
## 3. Translational Virology and Complex Co-Infections
Dr. Kizzmekia Corbett-Helaire (Harvard T.H. Chan School of Public Health) discussed how the immune landscape of the coronavirus spike is shaped by the receptor binding domain (RBD), the primary target for neutralizing antibodies, while the conserved N-terminal domain (NTD) provides protection when RBD immunity wanes. Immunized mice survived lethal MERS-CoV exposure despite low neutralizing antibody levels. Novel nanoparticle designs aim to enhance B-cell engagement and responses to hidden neutralizing epitopes using mRNA or DNA delivery. Cross-reactivity studies indicated that MERS nanoparticles elicited stronger binding responses to SARS-CoV-1 and WIV-1 spikes, informing universal coronavirus vaccine design [5].
Dr. Peter Quashie (University of Ghana) presented an analysis of the COVID-19 pandemic in Ghana, where over 171,000 cases resulted in a 1% death rate. His research challenged assumptions about the mildness of COVID-19 in West Africa and investigated whether cross-protection from malaria or other endemic infections played a role in the region's immune response [6]. Previous work in 2022 suggested that 68% of viral genomes in Ghana were variants of concern (VOC) [7]. The presentation explored the immune implications of prior viral exposures in shaping the region's response to COVID-19 and how to address systemic errors in viral phylogenies [8].
Dr. Susan Weiss (University of Pennsylvania) reported on the zoonotic transmission of MERS-CoV, which begins with Neoromicia capensis bats and spreads to camels before infecting humans. While camels experience only mild symptoms, human infections can be severe and result in limited human-to-human transmission. Dr. Weiss investigated viral replication in human nasal and bronchial cells, finding that MERS replication is restricted in nasal cells, likely due to receptor limitations, providing insights into the viral behavior at early infection stages [9].
## 4. Acute and Post-Viral Diseases
Dr. Maggie Bartlett (Johns Hopkins Bloomberg School of Public Health) gave an overview of dysautonomia, and neurotropic viruses included many viruses that cause post-acute infectious syndrome (PAIS), focusing on transcriptomics of neurons and how viruses alter messaging within infected cells leading to noncytolytic clearance. Preliminary clinical data were shared on viral recrudescence following treatments for PAIS and plans for continued work elucidating differences in those with hereditary connective tissue issues and their predisposition to PAIS [10].
Dr. Marc Lecuit (Institut Pasteur) discussed those pathogens with the ability to cause maternal-fetal infections and infections of the central nervous system. Dr. Lecuit shared data that compared the gut microbiota of children with Japanese encephalitis virus (JEV) in Southeast Asia. Lower gut microbiota diversity was observed in JEV cases, which could lead to reduced interferon signaling and immature immune responses.
Dr. Alfredo Garzino-Demo (University of Maryland, Baltimore) examined the mi-croRNA differences in olfactory cells to probe the anosmia caused by SARS-CoV-2. RNA-seq data showed clear separation between the groups. Between those with persistent olfactory symptoms and those with persistent non-olfactory symptoms, there were differentially expressed genes involved in neutrophil activation, which aligns with the current knowledge of SARS-CoV-2 pathogenesis. In the cohort with persistent olfactory symptoms, genes involved in inflammatory signaling were upregulated compared to controls. Between those with olfactory symptoms and those with no olfactory symptoms, there was enrichment in metallothionein genes. Metallothionein binds zinc, sequestering it, which may lead to lower available zinc levels.
## 5. Bats, Rats, and Other Vectors
Dr. Linfa Wang (Duke-NUS Medical School) opened the session by discussing the unique biology of bats and their ability to serve as reservoirs for a variety of pathogens. He emphasized the concept of "anti-disease" vs. "anti-pathogen" approaches, focusing on host immunity adaptation rather than solely targeting pathogens. Bats' unique biology allows them to tolerate viral infections without succumbing to disease, making them a key focus for future biomedical research. One of his groundbreaking discoveries involved the ASC2 gene in bats, which plays a critical role in inflammation control. Dr. Wang's team developed transgenic mice expressing bat ASC2, which demonstrated protection against influenza A virus, reducing lung inflammation without affecting viral load [11].
Dr. Jonathan Towner (US CDC) presented his work on tracking the movements of bat reservoirs, specifically focusing on the Marburg virus, a filovirus that causes hemorrhagic fever. Dr. Towner and his team utilized micro-GPS devices to monitor the nocturnal movements of Marburg virus-positive bats in Uganda. His research revealed the large distances bats can travel-up to 49.5 km in a single night-and their preference for specific habitats, such as fruit-bearing trees near human settlements.
Dr. Adeola Fowotade (University of Ibadan, Nigeria) provided a clinical perspective from Nigeria, Africa's most populous country. Her two case studies of viral infections and genomic analyses show that viruses belonged to clade IIb but represented three distinct lineages. Fowotade highlighted that lineage A, endemic to Nigeria, has been circulating between 2017 and 2022 through sustained human-to-human transmission, while lineage B is the strain circulating globally. Her research underscores the importance of genomic surveillance in tracking viral evolution and transmission patterns.
## 6. Genetic Sequences and Epidemiological Insights
Dr. Laura Dickson (University of Texas Medical Branch) highlighted how climate change is reshaping arbovirus transmission, particularly through Aedes aegypti adaptation. Urbanization and drier environments increase Zika virus (ZIKV) susceptibility in A. aegypti, yet most temperature-based models fail to incorporate humidity effects. Lower humidity reduces mosquito survival but paradoxically increases blood-feeding rates, potentially enhancing virus transmission [12].
Dr. William de Souza (University of Kentucky) reported on the Oropouche virus, a midge-borne orthobunyavirus responsible for febrile illness, CNS infections, and pregnancy complications [13]. Its increasing spread in South America underscores the need for enhanced surveillance and expanded development and deployment of affordable diagnostics.
Dr. Scott Weaver (University of Texas Medical Branch) discussed recent findings on viral evolution and cross-immunity. While Oropouche virus traditionally has a low case fatality rate, new data suggest recent increases, surpassing dengue in some regions. He also explored Western equine encephalitis virus (WEEV) evolution in North America. Notably, chikungunya virus (CHIKV) immunity appears to reduce susceptibility to Mayaro virus, suggesting that CHIKV circulation could help prevent future Mayaro outbreaks [13,14].
## 7. Cutting-Edge Diagnostics and Therapeutics
Dr. Marc Bonneville (Instititut Merieux) emphasized the need for advancing multiplex diagnostic platforms to simplify pathogen detection for clinicians. He discussed the importance of real-time surveillance, particularly in high-burden areas like sub-Saharan Africa, where automated diagnostic platforms could significantly improve public health outcomes by allowing rapid response to emerging pathogens [15].
Dr. Anne Wyllie (Yale University) emphasized saliva as a cost-effective sample type for outbreak control. Early saliva collection devices ranged from USD 7 to USD 28, with higher costs linked to stabilizing buffers, which her research found unnecessary. Studies showed stable SARS-CoV-2 detection in saliva under various conditions, including storage at 30-40 • C and mail transit for 56 h [16]. Similar results were found for influenza A/B, RSV, and Mpox, supporting saliva's potential for remote, cold chain-free testing. Saliva can detect SARS-CoV-2 and Mpox earlier and at higher viral loads than nasal swabs, making it ideal for pre-and asymptomatic screening.
Dr. Nokukhanya Msomi (CAPRISA) discussed current and emerging strategies for hepatitis B treatment, emphasizing viral suppression and the shift toward a functional cure-sustained HBV DNA suppression off treatment, HBsAg loss, and seroconversion. A major challenge is the persistence of covalently closed circular DNA (cccDNA), which drives chronic infection. New therapies aim to silence cccDNA, prevent transcription, and enhance viral decay. Diagnostic tools are evolving to measure ultra-low levels of HBsAg and HBV DNA, with additional biomarkers like core-related antigen and HBV RNA under investigation. A multi-target approach, similar to HIV treatment, may improve outcomes by combining nucleotide analogs, interferons, and novel antivirals.
## 8. Pandemic Preparedness
Dr. Jana Broadhurst (University of Nebraska Medical Center) underscored the need for community trust to be at the center of pandemic response. Pandemics continually face challenges in diagnostic testing, beginning with limited tools and struggling to rapidly develop novel solutions. A critical issue is the lack of coordination among clinical, public health, and research sectors in defining prioritized test use cases. While there is a strong effort in generating target profiles for diagnostics, improved collaboration is essential. Scaling up diagnostic responses remains slow, as seen in COVID-19, and requires community trust and engagement to ensure accessibility. Community-based approaches, such as saliva sampling, have shown success in remote Nebraska, where self-collected specimens maintained integrity and enabled variant sequencing of nearly 80% of positive SARS-CoV-2 cases.
Dr. Rachel Roper (East Carolina University) discussed the history of poxviruses and provided an overview of the status today. Mpox often presents with a single lesion, leading to missed diagnoses. The 2003 outbreak was linked to infected prairie dogs that were co-housed with imported African rodents (African giant pouched rats, dormice, and rope squirrels) for the pet trade. In 2023-2024, cases surged globally, with WHO reporting 97,000 cases and 200 deaths by May 2024. Underreporting due to inadequate surveillance and testing suggests higher true case numbers. The emergence of Clade 1, a more virulent strain with human-to-human transmission, raises concerns. While Clade 1 cases have declined in Europe and the Americas, they are rising in Africa and the Western Pacific. Safe poxvirus-based vaccines like MVA remain crucial in mitigating Mpox's impact.
Dr. Alash'le Abimiku (Institute of Human Virology, Nigeria) highlighted the role that data science plays in pandemic preparedness. Genomic analyses from GISAID revealed that early pandemic waves were underreported due to limited sequencing capacity. The Social Vulnerability Index (SVI) quantifies a community's resilience, with findings indicating that high-income individuals reduced travel distances but not frequency during lockdowns. Excess COVID-19 deaths disproportionately affected HIV-positive individuals and those with metabolic diseases; however, vaccines provided substantial protection, even among immunocompromised populations, underscoring the need for equitable vaccine distribution.
## 9. Scientific Misinformation and Public Trust
The session on scientific misinformation, led by Dr. Salim Abdool Karim, focused on the global challenge of combating misinformation during pandemics and outbreaks [17]. The discussion addressed the complexities of conveying accurate scientific information to the public in an age dominated by social media, where misinformation spreads rapidly. Dr. Abdool Karim emphasized the importance of clear communication strategies, transparency from public health officials, and robust fact-checking mechanisms. Panelists discussed the psychological factors that make individuals more susceptible to misinformation and highlighted the importance of training scientists and public health professionals in effective communication. The session underscored that restoring public trust in science requires sustained efforts from both the scientific community and media outlets to counteract misinformation and promote health literacy. This is especially compelling when arguing for donor nation assistance to tackle emerging infections at their sources in low-and middle-income countries [18,19].
## 10. Conclusions
The 2024 GVN Annual Meeting in Durban served as a testament to the power of global scientific collaboration in virology. By bringing together leading experts, emerging researchers, and key stakeholders, the conference fostered critical discussions on pathogen surveillance, vaccine development, and innovative strategies for pandemic preparedness. The emphasis on Africa's role in global virology highlighted the region's scientific contributions and underscored the importance of equitable access to research funding, infrastructure, and training. As misinformation continues to challenge public health responses, the meeting reinforced the need for science-driven policies and community engagement. Looking ahead, the GVN remains committed to strengthening international partnerships, advancing cutting-edge virology research, and ensuring that solutions to global health threats are both inclusive and sustainable.
## References
1. Sivanandham, Sivanandham, Xu et al. (2025) "Plasma lipidomic alterations during pathogenic SIV infection with and without antiretroviral therapy" *Front. Immunol*
2. Giron, Liu, Adeniji et al. (2024) "Immunoglobulin G N-glycan markers of accelerated biological aging during chronic HIV infection" *Nat. Commun*
3. Tang, Flavel, Londrigan et al. (2025) "Polyphenol rich sugarcane extract restricts select respiratory viruses depending on their mode of entry" *Virology*
4. Zhou, Ahearn, Lokugamage et al. (2024) "SARS-CoV-2 EndoU-ribonuclease regulates RNA recombination and impacts viral fitness"
5. Abiona, Wang, Leist et al. "MERS-CoV spike vaccine-induced N-terminal domain-specific antibodies are more protective than receptor binding domain-specific antibodies"
6. Tapela, Prah, Tetteh et al. (2024) "Cellular immune response to SARS-CoV-2 and clinical presentation in individuals exposed to endemic malaria" *Cell Rep*
7. Oduro-Mensah, Oduro-Mensah, Quashie et al. (1000) "Explaining the unexpected COVID-19 trends and potential impact across Africa"
8. Hunt, Hinrichs, Anderson et al. (2024) "Addressing pandemic-wide systematic errors in the SARS-CoV-2 phylogeny"
9. Renner, Parenti, Bracci et al. (2025) "Betacoronaviruses Differentially Activate the Integrated Stress Response to Optimize Viral Replication in Lung-Derived Cell Lines" *Viruses*
10. Bartlett, Sova, Jain (2024) "Hereditary Connective Tissue Diseases and Risk of Post-Acute SARS-CoV-2. Viruses"
11. Ahn, Chen, Rozario et al. (2023) "Bat ASC2 suppresses inflammasomes and ameliorates inflammatory diseases" *Cell*
12. Abu, Becker, Accoti et al. "Low humidity enhances Zika virus infection and dissemination in Aedes aegypti mosquitoes" *bioRxiv*
13. Scachetti, Forato, Claro et al. (2025) "Re-emergence of Oropouche virus between 2023 and 2024 in Brazil: An observational epidemiological study" *Lancet Infect. Dis*
14. Cansado-Utrilla, Saldaña, Golovko et al. (2025) "Mosquito host background influences microbiome-ZIKV interactions in field and laboratory-reared Aedes aegypti"
15. Llitjos, Carrol, Osuchowski et al. (2024) "Enhancing sepsis biomarker development: Key considerations from public and private perspectives" *Crit. Care*
16. Pan, Martin, Nazareth et al. (2024) "Defining within-host SARS-CoV-2 RNA viral load kinetics during acute COVID-19 infection within different respiratory compartments and their respective associations with host infectiousness: A protocol for a systematic review and meta-analysis" *BMJ Open*
17. Karim "Public understanding of science: Communicating in the midst of a pandemic" *Public Underst. Sci*
18. Boyce, Attal-Juncqua, Lin et al. "Global Fund contributions to health security in ten countries, 2014-2020: Mapping synergies between vertical disease programmes and capacities for preventing, detecting, and responding to public health emergencies" *Lancet Glob. Health*
19. Vermund (2017) "The Vital Case for Global Health Investments by the US Government" *Clin. Infect. Dis*
20. "The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12607583&blobtype=pdf | # Mutational dissection of HCMV gB and gH cytoplasmic tails highlights conserved and divergent features of fusion regulation
Chanyoung Lee, Hannah Dibella, Ekaterina Heldwein
## Abstract
In herpesviruses, the fusogenic activity of the conserved glycoproteins gB and gH is regulated by their cytoplasmic (or intraviral) tails. However, prior to this work, their regulatory mechanisms had only been investigated in Alphaand Gammabut not in Betaherpesvirinae. Here, we developed a plasmid-transfection-based split-luciferase assay as a quantitative platform for measuring cell-cell fusion mediated by glycoproteins from human cytomegalovirus (HCMV), a member of Betaherpesvirinae. Using this assay, we confirmed that glycoproteins gB, gH, gL, and gO, along with the PDGFRα receptor, are necessary and sufficient for efficient cell-cell fusion in two distinct HCMV strains, TR and AD169. To investigate the roles of the cytoplasmic tails of HCMV TR gB and gH in membrane fusion, we generated a series of truncation and point mutants analogous to those that have hyper-or hypo-fusogenic phenotypes in herpes simplex virus 1 (HSV-1). We found that, similarly to HSV-1, the C-terminal amphipathic helix in the HCMV gB cytoplasmic tail restricts fusion, whereas the entire HCMV gH cytoplasmic tail is required for fusion activation. However, the structure of the HCMV gB cytoplasmic tail and its interactions with the cytoplasmic tail of gH might be different from HSV-1. We hypothe size that while in HCMV-as in HSV-1-the cytoplasmic tails of HCMV gB and gH function as an inhibitory clamp and an activating wedge, their structures and interactions differ from HSV-1, implicating potential mechanistic differences in fusion regulation between herpesvirus subfamilies. IMPORTANCE Herpesviruses promote membrane fusion during infection using a complex multi-component membrane fusion machinery. Proper spatiotemporal deployment of such machinery is subject to precise regulatory inputs. One of these is the regulation of the fusogenic activity of the conserved herpesviral glycoproteins gB and gH by their cytoplasmic (or intraviral) domains. This regulatory mechanism has been investigated in Alphaand Gammabut not yet in Betaherpesvirinae. Here, we combined mutagenesis targeting the cytoplasmic tails of HCMV gB and gH with a plasmid-transfection-based split-luciferase assay for measuring cell-cell fusion, which we developed here. By testing a panel of truncation mutants, we showed that inhibitory and activating regulatory regions in gB and gH, respectively, are conserved in HCMV. However, none of the single point mutants of gB had expected phenotypes, suggesting that the cytoplasmic tails of HCMV gB and gH have distinct structures and interactions. This study introduces a robust and sensitive in vitro cell-cell fusion assay for probing HCMV fusion mechanism and shows that despite sequence conservation, the cytoplas mic domains of gB and gH may regulate fusion by distinct mechanisms. This knowledge may inform future efforts in rational vaccine design and antiviral development.
M embrane fusion is a fundamental biological process that enables enveloped viruses to enter host cells by merging the viral envelope with the host membrane. This process is mediated by viral fusogens that refold from a metastable prefusion conformation to a stable postfusion conformation while interacting with the opposing membranes (reviewed in references [1,2]). These major conformational changes are believed to provide the necessary energy for fusion. The activity of fusogens is tightly regulated to ensure proper spatiotemporal deployment.
Herpesviridae-a family of large double-stranded DNA viruses that infect mammals, birds, and reptiles-employ a particularly complex membrane fusion mechanism. Unlike most enveloped viruses that rely on a single fusogen for membrane merger (reviewed in references [1,2]), herpesviruses utilize a set of surface glycoproteins. The key players in this process are glycoprotein B (gB) and the glycoprotein H/glycoprotein L (gH/gL) complex, which are highly conserved across the Herpesviridae family. gB is composed of an N-terminal ectodomain (ecto), a membrane-proximal region (MPR), a hydrophobic transmembrane domain (TMD), and an intraviral, C-terminal cytoplasmic domain (CTD). gB is a homotrimeric class III membrane fusogen (reviewed in references [1,3]). Unlike many class III fusogens that are activated by exposure to low pH, gB is activated by the gH/gL complex (reviewed in references [4,5]). The gH/gL complex is a heterodimer containing an N-terminal ectodomain (ecto), a transmembrane domain (TMD), and a short C-terminal cytotail (CT) (reviewed in reference [4]). gL lacks a TMD and co-folds with the gH N terminus (reviewed in reference [4]). Depending on a herpesvirus, gH/gL triggers gB activation either upon binding a host receptor directly or as a larger complex with additional viral proteins that bind host receptors, conferring host tropism (reviewed in reference [5]).
Human cytomegalovirus (HCMV) is a member of the Betaherpesvirinae subfamily of Herpesviridae. It is the primary cause of congenital abnormalities, such as hearing loss, blindness, epilepsy, and microcephaly, in approximately 1% of newborns (reviewed in reference [6]). In addition, HCMV can cause disease in immunocompromised individuals, such as solid organ transplant patients and patients with AIDS, leading to conditions such as gastrointestinal ulceration, hepatitis, pneumonitis, and retinitis (reviewed in reference [7]).
In HCMV, cell entry requires fusogen gB and two gH/gL complexes: gH/gL/gO (trimer) and gH/gL/UL128-131 (pentamer), which have distinct functions. The trimer is required for entry into all cell types, including fibroblasts, epithelial, endothelial, and myeloid cells, and binds platelet-derived growth factor receptor alpha (PDGFRα) (8). Along with gB, the trimer is also necessary for efficient cell-cell fusion (9) and efficient fusion during HCMV entry (10). By contrast, the pentamer is required for entry into epithelial and endothelial cells and binds neuropilin-2 (Nrp2) (reviewed in reference [11]).
While the ectodomains of gB and gH are directly involved in membrane fusion, their cytoplasmic (intraviral) tails have regulatory roles (12,13). The cytoplasmic domain of gB (gB CTD ) inhibits the fusogenic activity of gB because point mutations, truncations, or insertions in the gB CTD increase the extent of cell-cell fusion (14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28). Such mutations are thus referred to as hyper-fusogenic. This phenomenon has been extensively studied in the Alphaherpesvirinae herpes simplex viruses 1 and 2 (HSV-1 and HSV-2) and in the Gammaherpesvirinae Epstein-Barr virus (EBV). Infection with HSV-1 bearing hyper-fuso genic mutations in the gB CTD causes the formation of multinucleated cells (syncytia) (14,17,29,30). Similarly, transient expression of mutant gB proteins, along with other core entry glycoproteins, in uninfected cells increases cell-cell fusion in HSV-1 (31,32), HSV-2 (15,19,24), and EBV (27,33,34). Based on mutational, structural, and biophysical studies in HSV-1, gB CTD has been proposed to form a membrane-dependent inhibitory clamp that stabilizes the gB ectodomain in its prefusion conformation and prevents its premature activation (12). By contrast, the cytoplasmic tail of gH (gH CT ) has been proposed to have an activating role. Progressive truncations of gH CT lead to a greater reduction in cell-cell fusion in HSV-1 (13,25) and EBV (35). Finally, interaction between gB CTD and gH CT has been proposed to trigger the fusogenic refolding of gB (13).
To dissect the roles of the cytoplasmic domains in the HCMV homologs of gB and gH, here, we developed an HCMV-specific dual split-luciferase assay (SLA), adapted from an HSV-specific SLA (13,36), which allowed real-time kinetic monitoring of fusion events. By combining the SLA with mutagenesis, we found that just as in HSV-1, the putative C-terminal amphipathic helix in HCMV gB CTD has an inhibitory effect on fusion, whereas the entire HCMV gH CT is indispensable for fusion and may have an activating role. Surprisingly, point mutations targeting conserved gB CTD residues that have hyperor hypo-fusogenic phenotypes in HSV failed to reproduce comparable phenotypes in HCMV. This suggests that while gB and gH cytoplasmic tails have analogous roles in HSV-1 and HCMV, their structures and interactions differ from HSV-1, implicating potential mechanistic differences in fusion regulation between herpesvirus subfamilies. Our study uncovers regulatory regions in HCMV entry glycoproteins and identifies phenotypic differences in cytoplasmic domain regulation between HCMV and HSV-1, laying the foundation for future studies of regulatory mechanisms controlling HCMV entry.
## RESULTS
## The optimization of the HSV-1-specific split-luciferase assay
Cell-cell fusion assays are commonly used as surrogate models to measure fusion mediated by herpesvirus entry glycoproteins. These range from microscopic counting of multinucleated cells (i.e., syncytia) (28,37,38) to detection of expression of a luciferase reporter (15,27,32,34,39). More recently, the split-luciferase assay, initially developed for studying HIV-mediated fusion (40), has been subsequently adapted to measure HSV fusion kinetics (36). This assay quantitatively measures cell-cell fusion over time between two distinct cell types: effector cells and target cells. Effector cells co-express viral entry glycoproteins (gD, gH, gL, gB) along with a portion of a split luciferase reporter (RLuc1-7), whereas target cells co-express the receptor and the complementary portion of the split luciferase reporter (RLuc8-11) (40). Upon cell fusion, the dual Renilla luciferase enzyme is reconstituted and produces a luminescent signal that allows for real-time measurement of fusion kinetics in intact cells with high sensitivity (36) (Fig. S1A).
The use of the split-luciferase assay (SLA) to measure cell-cell fusion mediated by HCMV entry glycoproteins has been reported (9). That study employed recombinant adenoviruses. Here, we developed an alternative, plasmid-transfection-based SLA by adapting the existing HSV-specific SLA protocol (13,36). Given the reported low surface levels of HCMV glycoproteins, notably, gB (41,42), and slower HCMV fusion kinetics (43) compared to HSV, we first optimized the SLA protocol for HSV-1 to enhance its absolute luminescence signal. These multiple optimization steps include the use of the fresh cell, single use of plasmids and EnduRen substrate aliquots, use of a plate reader with proper CO 2 level, and seeding cells with proper confluency for optimal transfection. When effector and target cells were transfected as previously published (effector cells, gB:gH:gL:gD (BHLD):RLuc1-7 = 3:1:1:1:1; target cells, nectin-1:RLuc8-11 = 1:1) (13), we achieved an absolute luminescence signal of approximately 500,000 raw luminescence units (RLU) in the HSV-1-specific SLA at 8 hours post-co-cultivation, nearly doubling previously reported values (Fig. S1B andC). Importantly, despite this increase, the assay preserved the expected trend of higher fusion extent of the hyper-fusogenic HSV-1 mutant gB868, ~150% fusion extent compared to WT gB (Fig. S1D). These optimized conditions served as a foundation for adapting the SLA to HCMV, allowing for a more robust and sensitive analysis of HCMV glycoprotein-mediated fusion.
## The development of HCMV-specific SLA
For the effector cells, we used human retinal pigment epithelial (ARPE-19) cells, which are susceptible to HCMV infection but have low endogenous expression of platelet-derived growth factor receptor alpha (PDGFRα) (44), which ensures low background fusion in the absence of target cells. HCMV entry glycoproteins gB, gH, gL, and gO from the HCMV TR or AD169 strain, along with RLuc1-7, were transiently expressed in the effector cells. The gH/gL/gO trimer was used because it is essential for HCMV entry across all cell types (8) and, along with gB, is necessary for efficient cell-cell fusion (9) and efficient fusion during HCMV entry (10) (Fig. 1A). Given broad genetic variation across HCMV glycoproteins, we chose two genetically distinct HCMV strains, the lab-adapted AD169 (reviewed in reference [45]) and the minimally passaged clinical isolate TR (46), which encode distinct gB, gH, and gO alleles (47,48). To render target ARPE-19 cells more susceptible to fusion in the presence of gH/gL/gO, PDGFRα V242K* was transiently expressed along with RLuc8-11 (PDGFRα V242K* :RLuc8-11 = 1:1) (Fig. 1A). PDGFRα V242K* is a variant of human PDGFRα isoform 1 that contains a deletion of residues 1-23 and a V242K substitution. These modifications preserve the wild-type PDGFRα affinity for HCMV gH/gL/gO while abrogating binding to the native ligand PDGF (49), which increases the surface levels of PDGFRα by reducing its internalization in the presence of PDGF.
We defined as "Standard" the condition where the effector cells received a total of 140 ng/well of plasmid DNA at a ratio of gB:gH:gL:gO (BHLO):RLuc1-7 = 3:1:1:1:1 whereas target cells received a total of 2 µg/well of plasmid DNA at a ratio of PDGFRα V242K* :RLuc8-11 = 1:1. This yielded a strong luminescence signal with TR gB, gH, gL, and gO (Fig. 1B) and a somewhat weaker signal with the AD169 gB, gH, gL, and gO (Fig. 1C). The TR glycoproteins promoted more efficient fusion (Fig. 1B Vs. C) and were, therefore, used in subsequent experiments. Nonetheless, the SLA can be used to detect fusion mediated by glycoproteins from two genetically distinct strains, AD169 and TR.
To confirm that the cell-cell fusion in our system was dependent on HCMV glyco proteins and receptor, we conducted control experiments either in the absence of HCMV glycoproteins in effector cells ("No gps") or in the absence of PDGFRα V242K* in target cells ("No PDGFRα"). Both controls exhibited minimal fusion activity (less than 5% relative fusion extent) when normalized to the "Standard" condition (Fig. 1D). These two conditions were henceforth used as negative controls in all SLA experiments. In addition, we showed that each of the four glycoproteins-gB, gH, gL, and gO-was required for cell-cell fusion and that in the absence of any one of these, fusion activity was minimal (less than 5% relative fusion extent) (Fig. 1D).
To optimize fusion extent, we modified transfection conditions for effector and target cells. First, transfecting only half the DNA amount for HCMV glycoproteins and reducing the total amount of transfected DNA from 140 ng/well to 100 ng/well in effector cells while maintaining the same relative glycoprotein ratios did not significantly change the overall extent of fusion ("1/2 gps"; total 100 ng/well, BHLO = 3:1:1:1) (Fig. 1B, C andE). This is consistent with a prior HCMV SLA study (9). Next, we examined variations in glycoprotein ratios by testing "Less gB": BHLO = 2:1:1:1; "More gB": BHLO = 4:1:1:1; or "Less gO"; BHLO = 3:1:1:0.5, while keeping the total DNA amount constant at 100 ng/well. Varying the relative amount of gB had little effect on fusion extent (Fig. 1E). However, reducing the amount of gO slightly increased the fusion extent (Fig. 1B andE). In addition, this condition induced a significant increase in absolute fusion curves in the AD169 strain compared to the "1/2 gps" condition (Fig. 1C), suggesting that excessive gO expression may inhibit fusion. This effect could be attributed to gO being a substrate of ER-associated degradation (ERAD) (11). Therefore, we defined as "Optimal" the condition where the effector cells received reduced gO amount ("Less gO": BHLO = 3:1:1:0.5).
Reducing the amount of PDGFRα V242K* ("0.5X PDGFRα") had little effect on fusion extent, whereas increasing the amount of PDGFRα V242K* ("1.5X PDGFRα") decreased fusion in a statistically significant manner (Fig. 1F). The observed reduction in fusion could be due to the difficulty in detaching cells following overexpression of PDGFRα V242K* even after a prolonged treatment with a chelating agent. In the final, optimized protocol used in all subsequent SLA experiments, the effector cells were transfected with 100 ng/well of total DNA (BHLO = 3:1:1:0.5), whereas the target cells were transfected with 2 µg/well of total DNA (PDGFRα V242K* :RLuc8-11 = 1:1).
Relative to HSV-1 (Fig. S1B), fusion mediated by HCMV glycoproteins had a lower extent (Fig. 1B andC) and reduced early and late fusion rates (Fig. S1E andF). Hence, we chose 8 hours and 20 hours as early and late time points, respectively, for measurement of HCMV fusion extents and rates. In addition, HCMV glycoproteins delayed the initiation of fusion around 40 min, whereas HSV-1 glycoproteins initiated the fusion around 10 min (Fig. S1G). As fusion mediated by HCMV glycoproteins started post 40-60 min of co-cultivation of effector and target cells, we defined the early rate of fusion as the slope of the fusion curve between 1 and 8 hours post-co-cultivation of effector and target cells in HCMV SLA.
## Similarly to HSV-1 gB CTD , the putative helix h3 in HCMV gB CTD has an inhibitory role whereas the rest of the gB CTD is essential for fusion
The HSV-1 gB CTD is composed of helices h1a, h1b, and h2 that form the folded tri meric core, resolved in the crystals of the nearly full-length HSV-1 gB (12) (Fig. 2A). An additional C-terminal amphipathic helix h3, unresolved in the crystal structure, was shown to interact with membrane by using electron spin resonance (12) (Fig. 2A). Both the C terminus of helix h2 and helix h3 are important inhibitory elements in gB homologs from HSV (15,19,24,31,32), PRV (50), and EBV gB (27,33,34), because their removal increases fusion. Larger C-terminal truncations of gB CTD that remove portions of the pedestal reduce gB surface levels and abolish cell-cell fusion in HSV (18,19,30,32) and EBV (27,33,34), and impaired ability to complement gB-null mutant in PRV ( 50) is likely due to protein misfolding.
To investigate the role of HCMV gB CTD in fusion and identify regulatory elements, we first generated an in silico model of HCMV gB using SWISS-MODEL (Fig. 2B). SWISS-MODEL predicted four helices within the gB CTD , h1a, h1b, h2, and h3 (Fig. 2B), in accordance with the crystal structure of the HSV-1 gB (12) (Fig. 2A). To test the roles of the individual helices within HCMV gB CTD , we designed five gB CTD mutants with partial or complete C-terminal truncations of predicted helices (Fig. 2C).
To assess their fusogenic potential, we measured the late fusion extent (20 hours) for each gB CTD truncation mutant. Truncation of the unstructured C terminus (gB895) reduced fusion extent to ~70% (Fig. 2D andE). Larger truncations that eliminated h3 (gB884) or the unstructured linker between h2 and h3 (gB868) enhanced fusion extent to ~170% and ~140% of WT gB, respectively (Fig. 2D andE). These two truncations also induced earlier fusion initiation (Fig. 2F). Truncations that eliminated the C-terminal portion of h2 (gB856) or the entire CTD (gB782) reduced fusion to background levels (Fig. 2D andE). In addition, these two truncations prevented fusion initiation within 20 hours (Fig. 2F).
To rule out defects in mutant gB expression, we assessed total cellular expression levels in ARPE-19 cells using Western blot (WB) (Fig. 2G; Fig. S2A). The four shorter gB CTD mutants (gB884, gB868, gB856, and gB782) were expressed at total levels comparable to WT gB (Fig. 2G), whereas the longer gB895 was expressed at a ~50% relative to WT gB (Fig. 2G; Fig. S2A). To check cell surface localization, we then measured surface expres sion levels by flow cytometry with mAb 1G2, which recognizes a conformational epitope in antigenic domain 5 (AD-5) of the gB ectodomain (Fig. 2H; Fig. S2B) (51). Mutations within the CTD are unlikely to affect the binding of 1G2 because its epitope retains its conformation in structures in both pre-and post-fusion forms of gB (52)(53)(54)(55). The surface levels of WT HCMV gB and the four longer gB CTD mutants (gB895, gB884, gB868, and gB856) were low (~5%-14% fluorescence-positive cells) (Fig. 2H; Fig. S2B) relative to HSV-1 WT gB (~70% fluorescence-positive cells) (Fig. S2C). However, the shortest mutant, gB782, which lacks the entire CTD, had significantly higher surface level, ~30% (Fig. 2H; Fig. S2B), likely due to the loss of a predicted tyrosine-based endocytic motif within helix h2 (YQML; residues 847-850), identified via Eukaryotic Linear Motif (ELM) analysis (http:// elm.eu.org/search.html). Other studies have also reported low surface expression levels of HCMV gB (41,42), presumably due to efficient endocytosis. Deletion of the entire HCMV gB CTD led to a substantial increase in cell surface expression (41,42). Analogous observations have been made for EBV gB (56,57). The reason for the reduced fusogenicity of the gB895 mutant (Fig. 2E) was unclear because its surface expression levels were comparable to WT gB (Fig. 2H). It could, perhaps, be less stable, which could also explain a 50% reduction in total cellular expression (Fig. 2G). The hyper-fusogenic phenotype of gB884 mutant (Fig. 2E) was not due to higher surface expression (Fig. 2H). Although the increased fusogenicity of gB868 could be partially due to higher surface expression (Fig. 2H), gB868 exhibited an earlier fusion initiation that was comparable to gB884 (Fig. 2F). Therefore, both gB868 and gB884 had hyper-fusogenic phenotypes. These results demonstrate that the putative amphipathic helix h3 in HCMV gB CTD has an inhibitory role. Lastly, the fusion-null mutant gB856 was expressed at a reduced surface level yet above background, whereas the other fusion-null mutant gB782 was expressed at an increased level relative to WT gB (Fig. 2H). Since fusion-null phenotypes of these two mutants are not due to the lack of cell surface expression, we conclude that the HCMV gB CTD , especially the putative helix h2, is essential for fusion.
## Mutations of residues that have key regulatory roles in HSV-1 gB CTD have divergent effects on function and cell surface expression in HCMV gB
To identify additional regulatory residues within gB CTD , we mapped previously characterized HSV-1 or HSV-2 gB CTD point mutations associated with hyper-fusogenic (summarized in reference [12]); hypo-fusogenic, that is, reduced fusion (13); or low-sur face level (15,32) phenotypes onto the sequence alignments of gB CTD from HSV-1 KOS and HCMV TR strains (Fig. S3).
Point mutations targeting residues at protein/membrane or trimeric interfaces within the gB CTD core in HSV-1 and HSV-2 are associated with hyper-fusogenic phenotypes (summarized in reference [12]). Five such residues in HSV-1 gB (R800, P805, V853, A855, and R858) (Fig. 3A) are conserved in HCMV gB (R778, P783, L856, A858, and R861) (Fig. S3) and also localize to the protein/membrane interface (R778, P783, and R861) or trimeric interfaces (L856 and A858) in the HCMV gB CTD homology model (Fig. 3B). We generated six single-point mutants in HCMV gB CTD by altering side chain size and/or polarity (gB R778W, P783A, L856A, A858V, R861H, and R861C). Unexpectedly, instead of hyper-fusogenic phenotypes, HCMV gB CTD single-point mutants had very low fusion levels, <20% of WT gB (gB L856A and A858V) or background (gB R778W, P783A, R861H, and R861C) (Fig. 3C andD). Whereas gB L856A and A858V initiated the fusion much later than the WT gB, after 10 or 6 hours post-co-cultivation, respectively, the rest showed no fusion within 20 hours (Fig. 3E). None of the single-point mutants showed any defect in total cellular expression (Fig. 3F; Fig. S4A). However, only gB R778W and P783A were expressed on the cell surface at the above-background levels (Fig. 3G; Fig. S4C). Based on these results, we classified gB R778W and P783A mutants as hypo-fusogenic and gB L856A, A858V, R861H, and R861C mutants as low-surface level. These results indicate that, despite sequence conservation, some residues might not be functionally conserved, whereas others might, instead, be important for protein folding.
## The functional pocket of HSV-1 gB CTD is not conserved in HCMV gB CTD
In HSV-1 gB CTD , T814L and A851V mutants have a rare hypo-fusogenic phenotype (13). Within the HSV-1 gB CTD structure (12), T814 and A851 are positioned at the bottom of a surface-exposed pocket (Fig. 4A), which has been proposed to bind the gH CT (13) according to the "clamp-and-wedge" model of gB activation by gH (12,13).
Analogous pockets in structural models of HCMV gB CTD generated by SWISS-MODEL (Fig. 4B) and AlphaFold (Fig. 4C) are lined by residues V792, L851, and V854. V792 and V854 in HCMV gB CTD align with T814 and A851 in HSV-1 gB CTD in sequence alignments (Fig. S3). To probe the role of the pocket residues, we generated five pocket-filling (V792L, L851F, and V854F) or pocket-emptying (L851V and V854A) mutants.
All mutants targeting residues L851 or V854 of HCMV gB CTD had very low fusion levels, <20% of WT gB (Fig. 4D andE) and initiated fusion later than the WT gB, after 2-3 hours post-co-cultivation (Fig. 4F). These mutants showed no defect in total cellular expression (Fig. 4G; Fig. S4B) but had significantly reduced surface expression, compara ble to background level (Fig. 4H; Fig. S4D). Therefore, we classified them as low-surface level. By contrast, gB V792L had background fusion levels (Fig. 4D andE) and had very low total cellular expression levels (Fig. 4G). These results show that despite sequence conservation, these residues in HCMV gB might be important for protein folding rather than function. Collectively, functional phenotypes of the mutations targeting putative interface and pocket residues point to fundamental differences in the predicted structures and/or regulatory mechanisms between HSV and HCMV.
## HCMV gH CT is required for cell-cell fusion
In HSV-1, fusion requires not only the gH/gL ectodomain but also the 14-residue gH cytotail (gH CT ) (Fig. 5A, left) (25). Truncation studies revealed that an eight-residue gH CT was sufficient to maintain wild-type fusion levels, whereas further truncations led to a progressive decline in fusion, culminating in a full loss of function upon complete deletion (13).
HCMV gH CT is significantly shorter across both clinical and laboratory-adapted strains, consisting of only six highly conserved residues (Fig. 5A, right and Fig. S5). To test its role in fusion regulation, we designed a panel of truncation mutants (gH742, gH741, gH740, gH739, gH738, and gH737) (Fig. 5B). Fusion efficiency decreased proportionally with the length of the remaining gH CT , culminating in a complete loss of fusion in gH737 (Fig. 5C andD), similarly to HSV-1 gH CT . In terms of fusion kinetics, gH742, gH741, and gH740 exhibited 40% of the early rate of fusion (Fig. 5E) and WT-level initiation of fusion (Fig. 5F). However, gH739 initiated fusion only after ~5 hours post co-cultivation (Fig. 5F), whereas gH738 and gH737 showed no fusion even after 20 hours (Fig. 5F).
To rule out defects in surface expression of the gH CT truncations, we performed flow cytometry analysis, which showed that, unlike HCMV gB, HCMV gH consistently displayed high surface expression levels (Fig. 5G; Fig. S6A). Notably, a progressive reduction in surface expression was observed across the gH CT truncations (Fig. 5G ; Fig. S6A), which is different from HSV-1 gH CT , in which truncations had no effect on cell surface expression (13). Reduction in cell surface expression did not account for the decrease in fusion, however, as can be seen from the progressive reduction in relative fusogenicity (Fig. S6B). Thus, efficient fusion requires all six gH CT residues.
## DISCUSSION
In herpesviruses, the cytoplasmic (or intraviral) tails of the conserved glycoproteins gB and gH regulate their fusogenic activity. While this regulatory role has been character ized in Alphaand Gammaherpesvirinae, it had not been investigated in Betaherpesvirinae prior to this study. Here, we generated a series of truncation and single point mutants in HCMV gB and gH cytoplasmic tails analogous to those that have hyper-or hypo-fuso genic phenotypes in HSV-1. By measuring the effect of mutations on cell-cell fusion using a plasmid-transfection-based split-luciferase assay, we showed that HCMV gB and gH have inhibitory and activating regulatory regions, respectively, that are conserved in other herpesviruses. We also uncovered phenotypic differences between analogous gB CTD mutants in HCMV and HSV-1, likely due to structural differences.
The first question addressed in this study was whether the HCMV gB CTD was as important for fusion and whether it restricted fusion. In HSV-1, the gB CTD has been proposed to form a membrane-dependent inhibitory clamp that stabilizes the gB ectodomain in its prefusion conformation (12). This clamp is composed of the folded trimeric core, resolved in the crystals of the nearly full-length HSV-1 gB, and the C-terminal amphipathic helix that interacts with the membrane (12). Point muta tions targeting protein/protein or protein/membrane interfaces within the gB CTD core cause hyperfusogenic phenotypes in HSV-1 and HSV-2 (summarized in reference [12]). Truncations of the amphipathic helix cause hyperfusogenic phenotypes in HSV-1 (30)(31)(32), HSV-2 (19,24), EBV (27,33,34), and larger plaques in PRV (50).
We found that truncations of the putative amphipathic helix h3 at the HCMV gB C terminus caused a hyperfusogenic phenotype (Fig. 2E), similarly to what has been observed in HSV-1 (30)(31)(32), HSV-2 (19,24), PRV (50), and EBV (27,33,34). In HSV-1 gB, this helix has been shown to interact with the membrane and is proposed to act as a stabilizing membrane bilayer anchor for the inhibitory gB CTD clamp (12). Larger truncations that eliminated a portion of helix h2 or the entire gB CTD (gB856 and gB782, respectively) completely abolished fusion activity (Fig. 2E), suggesting that HCMV gB CTD , especially the putative helix h2, is essential for fusion. Surprisingly, removal of the last 13 residues (gB895) reduced fusion (Fig. 2E), which could be due to problems in protein synthesis and/or stability, as suggested by a reduced total expression (Fig. 2G). This contrasts with analogous deletions in HSV and EBV that retained wild-type fusion levels (19,31). Taken together, these results show for the first time that HCMV gB CTD is not only important for membrane fusion induced by HCMV glycoproteins but that its putative C-terminal amphipathic helix has an inhibitory role.
The second question addressed in this study was whether the HCMV gH CT was as important for fusion. In HSV-1, the first 8 residues of the 14-residue gH CT are essential for membrane fusion (13,25,58,59). Truncations of HSV-1 gH CT reduce cell-cell fusion, with fusion levels being proportionate to the length of the remaining gH CT (13,25). In EBV, the 8-residue gH CT is also important for membrane fusion (35). Here, we found that the entire 6-residue HCMV gH CT is important for cell-cell fusion. Truncations reduced fusion proportionately to the length of the remaining gH CT to the point of abolishing it altogether (Fig. 5D).
While both the gB CTD and the gH CT are indispensable for HCMV-glycoproteininduced fusion, our mutational analysis suggests that the gB CTD structure and its interactions with the gH CT differ from their HSV-1 counterparts. First, mutations targeting conserved residues at putative protein/membrane or trimeric interfaces in HCMV gB CTD were either hypo-fusogenic or low-surface level (Fig. 3D andG) instead of hyper-fuso genic as in HSV-1 gB CTD . Moreover, mutations targeting the putative surface pocket in HCMV gB CTD were low-surface level (Fig. 4H) instead of hypo-fusogenic or hyper-fuso genic as in HSV-1 gB CTD . The most likely explanation for this discrepancy is that the point mutations, which were designed based on the homology-based structural model (Fig. 3B and4B) and Alphafold-predicted model (Fig. 4C), disrupt the actual HCMV gB CTD structure because it is different from the model. This also suggests that HCMV gB CTD and gH CT interact differently from what has been proposed for HSV-1. In HSV-1, the gH CT acts as a "wedge" that disrupts the gB CTD clamp by inserting residue V831 into the surface pocket (13). Our mutational analysis suggests that this surface pocket is not conserved in the HCMV gB CTD . In addition, the 6-residue gH CT lacks the counterpart of V831, which is the 7th residue in HSV-1 gH CT . We hypothesize that while in HCMV, the gB CTD may function as an inhibitory clamp and gH CT as an activating wedge, their structures and interactions differ from HSV-1 (Fig. 6). Our results implicate potential mechanistic differences in fusion regulation between herpesvirus subfamilies. Future studies will Clamp-and-wedge interaction triggers fusogenic refolding of gB from the prefusion into the postfusion conformation. Structures were rendered in ChimeraX using HSV-1 gH/gL ectodomain (PDB ID 3M1C), HSV-1 gB ectodomain (prefusion) (PDB ID 6Z9M), and HSV-1 gB MPR-TMD-CTD (PDB ID 5V2S). Residues of HSV-1 gH CT were presented using the AlphaFold-predicted gH/gL complex. (B) A model of HCMV fusion triggering based on this study. In HCMV, gH CT and gB CTD also act as a wedge and a clamp, respectively. However, HCMV gH CT interacts with different regions of the gB CTD , possibly due to different structures of the gB CTD clamp and the shorter gH CT wedge. Structures were rendered in ChimeraX using HCMV gH/gL/gO ectodomain (PDB ID 7LBE), HCMV gB ectodomain (prefusion) (PDB ID 7KDP), and HCMV gB MPR-TMD-CTD (homology modeling). Residues of HCMV gH CT were presented using the AlphaFold-predicted gH/gL/gO complex.
resolve the structure of the HCMV gB CTD and visualize gB CTD /gH CT interactions in HCMV and HSV-1.
Investigation of the fusion phenotypes reported here was made possible by the robust quantitative cell-cell fusion assay. HCMV-induced membrane fusion has traditionally been measured by visualizing syncytia, which is a labor-intensive, semiquantitative, endpoint method. To overcome these drawbacks, a kinetic SLA used to measure cell-cell fusion induced by HSV-1 glycoproteins was recently adapted to HCMV glycoproteins with recombinant adenoviruses (9). Here, we established an alternative, plasmid-transfection-based SLA and, for the first time, assessed membrane fusion mediated by HCMV glycoproteins from two genetically distinct TR and AD169 strains. By optimizing transfection conditions, we achieved robust signals (200,000-400,000 RLUs over 20 hours) (Fig. 1B andC), representing a fivefold increase in signal intensity over the adenoviral-based HCMV SLA. This enhanced sensitivity enabled accurate analysis of fusion kinetics, revealing that HCMV glycoproteins initiate the fusion approximately 4-6 times later and at slower rates than HSV-1 glycoproteins (Fig. S1E through G). These observations are consistent with the prolonged replication cycle of HCMV, which is 4-6 times longer than that of HSV-1 (43). One limitation of our study is the exclusive use of ARPE-19 cells and mutational analysis in the HCMV TR strain background. In addition, while AD169 and TR have distinct genotypes of gB, gH, and gO glycoproteins, the two strains do not capture full allelic diversity observed across clinical isolates, with nine gB (48), at least two gH (reviewed in reference [60]), and eight gO genotypes (47) currently recognized. Therefore, adapting the HCMV SLA assay to diverse cell types and additional HCMV strains will be important for evaluating the relevance of the proposed fusion mechanism more broadly.
## MATERIALS AND METHODS
## Cells and plasmids
Human ARPE-19 cells (arising retinal pigment epithelial cell line, ATCC CRL-2302) were cultured in Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F-12; Thermo Scientific) containing L-glutamine, HEPES, and Phenol Red and supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS; R&D systems) and 1× penicillinstreptomycin (pen-strep) solution (Corning). Chinese hamster ovary (CHO) cells (a gift from Dr. John M. Coffin, Tufts University) were grown in Ham's F12 medium (Corning) supplemented with 10% HI-FBS and 1× pen-strep solution. Both ARPE-19 and CHO cells were incubated at 37°C in the presence of 5% CO 2 , detached with trypsin, and subcultured 2-3 times per week.
Plasmids encoding codon-optimized, full-length HCMV (strain TR, Genbank: KF021605.1) genes for gB, gH, gL, and gO in a pDC316(io) vector background were a gift from Dr. Brent Ryckman, University of Montana. These plasmids were derived from replication-defective (E1-negative) adenovirus (Ad) vectors containing the packaging signal, a murine CMV promoter, and SV40 poly A sequences followed by a loxP site (61). Plasmid encoding mutant human PDGFRα V242K* , which contains a deletion of residues 1-23 and a V242K substitution within human PDGFRα isoform 1 (GenBank NM_006206.4) in a pDC316(io) vector background was a gift from Dr. Brent Ryckman, University of Montana. PDGFRα V242K* has a wild-type affinity for HCMV gH/gL/gO but does not bind its native ligand PDGF (49), which increases its surface levels due to blocked internalization in the presence of PDGF.
Plasmids RLuc1-7 and RLuc8-11 (carrying the Renilla split luciferase genes) (40) were gifts from Dr. Zene Matsuda (University of Tokyo). Plasmid pBG38 carrying the human nectin-1 gene (62) was a gift from Drs. Gary H. Cohen and Roselyn J. Eisenberg (Univer sity of Pennsylvania). pPEP98, pPEP99, pPEP100, and pPEP101 encode the full-length HSV-1 (strain KOS, Genbank: JQ780693.1) genes for gB, gD, gH, and gL, respectively, in a pCAGGS vector background and were gifts from Dr. Patricia G. Spear (Northwest ern University). Plasmid pJLS11 (encoding HSV-1 gB868 mutant) (32) was previously generated in our laboratory.
## Antibodies
Mouse anti-HCMV-gH monoclonal antibody 14-4b and mouse anti-HCMV-gB monoclonal antibody 27-156 were gifts from Dr. William J. Britt (University of Alabama). Human monoclonal anti-HCMV-gB antibody 1G2 (IgG3) was produced by Genscript. Rabbit polyclonal anti-HSV-gB antibody R69 was a gift from Dr. G. H. Cohen and R. J. Eisenberg (University of Pennsylvania).
## gB CTD mutagenesis
Truncations and single-point mutations in the cytoplasmic domain of the full-length HCMV (strain TR) gB gene were generated in pDC316(io)::HCMV TR gB background by using Phusion HF PCR kit (cat# M0530L) and either Gibson assembly master mix (cat# E2611L) or Quick-Change method. For all gB CTD truncation mutations (gB895, gB884, gB868, gB856, and gB782), two stop codons were introduced after the amino acid of interest by PCR and assembled using the Gibson Assembly method. For gB CTD single-point mutations, 1-3 nucleotide substitutions were introduced by PCR and cloned using either Gibson Assembly method (gB V792L, L851V, L851F, V854A, and V854F) or Quick-Change mutagenesis (gB R788W, P783A, L856A, A858V, R861H, and R861C). Primers used in gB CTD mutagenesis are listed in Table S1. All constructs were confirmed by plasmid sequencing. The resulting plasmids were pLL7 (gB895), pLL8 (gB884), pLL9 (gB868), pLL10 (gB856), pLL11 (gB782), pLL12 (V792L), pLL13 (L851V), pLL14 (L851F), pLL15 (V854A), pLL17 (V854F), pLL25 (R778W), pLL26 (P783A), pLL28 (L856A), pLL29 (A858V), pLL30 (R861H), and pLL31 (R861C).
## gH CT mutagenesis
Truncations in the cytoplasmic tail of the full-length HCMV (strain TR) gH gene were generated in pDC316(io)::HCMV TR gH background by using Phusion HF PCR kit (cat# M0530L) and either Gibson assembly master mix (cat# E2611L) or Quick-Change method. For gH CT truncations, two stop codons were introduced after the amino acid of interest by PCR and cloned using either the Gibson Assembly method (gH737, gH740) or the Quick-Change mutagenesis (gH738, gH739, gH741, and gH742). Primers used in gH CT mutagenesis are listed in Table S2. All constructs were confirmed by plasmid sequenc ing. The resulting plasmids were pLL19 (gH737), pLL20 (gH738) pLL21 (gH739), pLL22 (gH740), pLL23 (gH741), and pLL24 (gH742).
## Cell-cell fusion assay
Cell-cell fusion mediated by HSV-1 glycoproteins was measured using a SLA following the published protocol (13) with modifications, such as single use of plasmids and EnduRen substrate aliquots, use of a plate reader with proper CO2 level, and seeding cells with proper confluency. Cell-cell fusion mediated by HCMV glycoproteins was measured using a SLA adapted from reference (13). ARPE-19 cells were seeded into 3 wells per condition in a 96-well plate at 10,000 cells per well with 100 µL volume for effector cells and in 6-well plates at 200,000 cells per well with 2 mL volume for target cells.
The next day, when cell confluency is 60%-80%, a total of 100 ng DNA per well was transfected with 30 ng of gB (pDC316(io)::HCMV TR gB or gB CTD mutants), 10 ng of gH (pDC316(io):: HCMV TR gH or gH CT mutants), 10 ng of gL (pDC316(io)::HCMV TR gL), 5 ng of gO (pDC316(io)::HCMV TR gO), 10 ng of split luciferase (pCAGGS::RLuc1-7), and 35 ng of pCAGGS in effector cells using jetPRIME transfection reagent (0.3 µL per well in 10 µL jetPRIME buffer, cat# 89129-924). Target cells were transected with plasmids of a total 2 µg DNA per well with 1 µg of the mutated receptor PDGFRα V242K* (d 1-23 AA, V242K) (pDC316(io)::PDGFRα V242K* ) and 1 µg of the complementary part of split luciferase (pCAGGS::RLuc8-11) using jetPRIME transfection reagent (4 µg per well in 200 µL jetPRIME buffer) in a dropwise manner. In addition, "No gps" and "No PDGFRɑ, " which stand for deletion of all glycoproteins and PDGFRɑ V242K* , respectively, were included as negative controls.
On day 3, at 24 hours post-transfection, the medium of effector cells was replaced with 50 µL per well of fusion medium (DMEM/F-12 with 10% FBS, 1× Pen/Strep, 50 mM HEPES), containing 1:500 EnduRen live cell substrate (Promega, cat# E6482) added. After target cells were detached using 1 mL of Versene (Fisher Scientific, cat# 226-126) per well for 25 min, they were collected, centrifuged, and resuspended in fusion medium of 500 µL per well. Next, 50 µL of resuspended target cells was added to each 96-well of effector cells, and the plate was immediately placed in a BioTek plate reader at 37°C with 5% CO2. Luminescence was recorded every 2 min for 20 hours. Luminescence values were then averaged for the three wells in each condition, normalized to the WT signal at 20 hours, and expressed as a percentage of late fusion extent. For the initiation of fusion, the first time point where luminescence reached a value twofold above the "No gps" control was monitored. Data presented represent the mean with standard error of the mean (SEM) from at least three biological replicates unless otherwise noted.
## Flow cytometry
Cell surface expression of gB CTD and gH CT mutants was measured using flow cytometry. ARPE-19 cells were seeded at 2.5 × 10 5 cells per well in six-well plates (2 mL per well). The next day, ARPE-19 cells were transfected with a total of 2 µg of DNA using 4 µg per well of jetPRIME transfection reagent in 200 µL jetPRIME buffer (VWR, cat# 89129-924). For measurement of surface expressions of gB CTD mutants, 2 µg of WT gB or gB CTD mutants were transfected per well. For measurement of surface expressions of gH CT truncation mutants, 2 µg of gB, gH, gL, and gO (3:1:1:0.5) were transfected per well. Two wells were left for mock and empty vector (pCAGGS) transfections as negative controls.
On day 3, at 24 hours post-transfection, cells were incubated with 1 mL of Versene per well for 25 min at 37°C, collected in ice-cold FACS medium (PBS with 3% FBS), and washed with FACS media. For gB detection, additionally, cells were blocked with 100 µL of human Fc Block diluted 1:50 in PBS (BD, cat#564220) for 30 minutes at 4°C to prevent non-specific binding of anti-HCMV-gB antibody to human Fc receptor. For primary antibody incubation, cells were incubated with human monoclonal anti-HCMV-gB IgG 1G2 (1:1,000, 100 µL per well) or mouse monoclonal anti-HCMV-gH IgG 14-4b (1:200, 100 µL per well) for 1 hour at 4°C. Cells were then fixed with 4% paraformalde hyde (PFA) in PBS for 20 min at room temperature, washed three times with FACS media, and incubated with Alexa-488-conjugated secondary antibodies (1:250, 100 µL per well) (Alexa488-conjugated goat anti-human IgG for gB and Alexa488-conjugated goat anti-mouse IgG for gH) for 1 hour at 4°C in the dark. Cells were washed three times before resuspending in 250 µL of FACS media. gB-and gH-positive cells were measured by the fluorescence-activated cell sorting method using the pCAGGS-transfected cells with a cutoff of around 2% Alexa488-positive cells to capture most true positives while minimizing false positives. Flow cytometry scatter plots and histograms were generated by plotting single cells and cell count over fluorescence intensity, respectively.
HSV-1 gB surface expressions were measured following the above protocol with modifications. CHO cells were transfected with 2 µg of HSV-1 gB and rabbit polyclonal anti-HSV-gB IgG R69 (1:500, 100 µL per well) and FITC-conjugated goat anti-rabbit IgG (1:250, 100 µL per well) were used for primary and secondary antibodies, respectively.
## Western blots
Total cellular expressions of gB CTD truncation and point mutation constructs were tested using Western blotting. ARPE-19 cells were seeded at 2.5 × 10 5 cells in 2 mL per well in a 6-well plate. The next day, cells were transfected with 2 µg of WT gB or gB CTD mutant plasmids per well using 4 µg per well of jetPRIME transfection reagent in 200 µL jetPRIME buffer (VWR, cat# 89129-924). On day 3, at 24 hours post-transfection, cells were washed once with ice-cold PBS and collected using a cell scraper (Celltreat, cat# 229310) in a 100 µL of ice-cold radioimmunoprecipitation assay (RIPA) buffer supplemen ted with 1× Complete protease inhibitor (Sigma-Aldrich, cat# 05056489001) per well. After a 30 minute incubation on ice, cells were centrifuged, supernatants were collected, and protein concentration of each condition was normalized by mixing with SDS-PAGE loading dye after a BCA assay (Thermo Scientific cat# 23227).
Proteins were denatured by heating at 95°C for 5 minutes, separated by SDS-PAGE at 200 V for 30 minutes, and transferred onto nitrocellulose membranes (GE Healthcare, cat # 10600002) using the Trans-Blot Turbo Transfer System (Bio-Rad) at 25 V for 30 minutes. Membranes were then blocked with 3% BSA (Fisher Scientific cat# BP1600100) in TBST for 1 hour at room temperature with gentle rocking. For antibody staining, strips of membranes were incubated overnight at 4°C with the anti-HCMV-gB IgG 27-156 (mAb, mouse IgG2b, 1:1,000) or anti-β-actin IgG (pAb, 1:1,000,000) in 3% BSA/TBST. The next day, membranes were washed three times with TBST and incubated with a fluorescent secondary antibody IRDye 800cw goat anti-mouse IgG (LI-COR Bioscien ces, cat# 926-32210) or IRDye 800cw goat anti-rabbit IgG (LI-COR Biosciences, cat# 926-32211) (1:5,000) for 1 hour at room temperature with gentle rocking. Membranes were imaged using a Bio-Rad GelDoc system. For quantification of protein expressions, band intensities were adjusted by dividing by the band intensity of actin in ImageJ. Relative cellular expression was measured by normalizing its adjusted band intensity to WT gB intensity.
## Structural analysis
Structural models of the HCMV (TR) gB trimer were generated either using the Alpha Fold v.3 online server (https://alphafoldserver.com/) (63) or the SWISS-MODEL homology modeling program (64) using the crystal structure of the HSV-1 gB (PDB: 5V2S) (12) and visualized in ChimeraX (65). The structural models were aligned relative to the membrane using TMHMM 2.0 (https://services.healthtech.dtu.dk/services/TMHMM-2.0). In AlphaFold v.3, the accuracy of prediction for each residue is shown as the local-dis tance difference test (pLDDT), on a scale from 0 to 100, with higher scores indicating higher quality prediction. In SWISS-MODEL, the accuracy of prediction for each residue is shown as QMEANDisCO (Q mean distance constraint), which represents energies obtained statistically calculated relative to all known experimental 3D structures in the database, on a scale from 0 to 1, with higher scores indicating higher confidence.
## Sequence alignments
gB CTD sequences between HSV-1 (strain KOS) and HCMV (strain TR) were aligned in Clustal Omega and rendered using ESPript 3.0 (https://espript.ibcp.fr) (66). The secon dary structures of HSV and HCMV gB CTD were predicted in PSIPRED v.4 (http://bio inf.cs.ucl.ac.uk/psipred) (67). Locations of hyper-and hypo-fusogenic point mutations found in HSV-1 or HSV-2 were indicated above the residue with green and red asterisks, respectively. Locations of point mutations that resulted in limited surface expression in HSV-1 were indicated with purple asterisks. In all sequence alignments, conserved and similar residues of HCMV gB are highlighted in white letter within a red box and in red letter within a blue box, respectively. gH CT sequences from selected herpesviruses, including three alpha-herpesviruses (HSV-1 KOS, HSV-2 333, PRV Kaplan) and eight beta-herpesviruses (HCMV clinical and laboratory adapted strains AD169, JHC, JP, TB40, Toledo, Towne, Merlin, and TR) were aligned using Clustal Omega (68) and rendered with ESPript 3.0 (66) (https://espript.ibcp.fr).
## Statistics
Statistical analysis was performed on the normalized values using Graph-Pad PRISM 9 software. Unpaired t-test with Welch's correction was used to compare conditions as indicated.
## References
1. White, Ward, Odongo et al. (2023) "Viral membrane fusion: a dance between proteins and lipids" *Annu Rev Virol*
2. Harrison (2015) "Viral membrane fusion" *Virology*
3. Backovic, Jardetzky (2009) "Class III viral membrane fusion proteins" *Curr Opin Struct Biol*
4. Connolly, Jardetzky, Longnecker (2021) "The structural basis of herpesvirus entry" *Nat Rev Microbiol*
5. Pino, Heldwein (2022) "Well put together-a guide to accessorizing with the Herpesvirus gH/gL complexes" *Viruses*
6. Schleiss (2018) "Congenital cytomegalovirus: impact on child health" *Contemp Pediatr*
7. Griffiths, Baraniak, Reeves (2015) "The pathogenesis of human cytomegalovirus" *J Pathol*
8. Zhou, Lanchy, Ryckman (2015) "Human cytomegalovirus gH/gL/gO promotes the fusion step of entry into all cell types, whereas gH/gL/UL128-131 broadens virus tropism through a distinct mechanism" *J Virol*
9. Schultz, Ponsness, Lanchy et al. (2025) "Human cytomegalovirus gH/gL/gO binding to PDGFRα provides a regulatory signal activating the fusion protein gB that can be blocked by neutralizing antibodies" *bioRxiv*
10. Schultz, Ponsness, Lanchy et al. (2025) "Human cytomegalovirus gH/gL/gO binding to PDGFRα provides a regulatory signal activating the fusion protein gB that can be blocked by neutralizing antibodies" *J Virol*
11. Nguyen, Kamil (2018) "Pathogen at the gates: human cytomegalo virus entry and cell tropism" *Viruses*
12. Cooper, Georgieva, Borbat et al. (2018) "Structural basis for membrane anchoring and fusion regulation of the herpes simplex virus fusogen gB" *Nat Struct Mol Biol*
13. Pataki, Sanders, Heldwein (2022) "A surface pocket in the cytoplasmic domain of the herpes simplex virus fusogen gB controls membrane fusion" *PLoS Pathog*
14. Gage, Levine, Glorioso (1993) "Syncytium-inducing mutations localize to two discrete regions within the cytoplasmic domain of herpes simplex virus type 1 glycoprotein B" *J Virol*
15. Ruel, Zago, Spear (2006) "Alanine substitution of conserved residues in the cytoplasmic tail of herpes simplex virus gB can enhance or abolish cell fusion activity and viral entry" *Virology (Auckl)*
16. Engel, Boyer, Goodman (1993) "Two novel single amino acid syncytial mutations in the carboxy terminus of glycoprotein B of herpes simplex virus type 1 confer a unique pathogenic phenotype" *Virology (Auckl)*
17. Diakidi-Kosta, Michailidou, Kontogounis et al. (2003) "A single amino acid substitution in the cytoplasmic tail of the glycoprotein B of herpes simplex virus 1 affects both syncytium formation and binding to intracellular heparan sulfate" *Virus Res*
18. Cai, Gu, Person (1988) "Role of glycoprotein B of herpes simplex virus type 1 in viral entry and cell fusion" *J Virol*
19. Fan, Grantham, Smith et al. (2002) "Truncation of herpes simplex virus type 2 glycoprotein B increases its cell surface expression and activity in cell-cell fusion, but these properties are unrelated" *J Virol*
20. Walev, Lingen, Lazzaro et al. (1994) "Cyclosporin A resistance of herpes simplex virus-induced "fusion from within" as a phenotypical marker of mutations in the Syn 3 locus of the glycoprotein B gene" *Virus Genes*
21. Haanes, Nelson, Soule et al. (1994) "The UL45 gene product is required for herpes simplex virus type 1 glycoprotein Binduced fusion" *J Virol*
22. Muggeridge (2000) "Characterization of cell-cell fusion mediated by herpes simplex virus 2 glycoproteins gB, gD, gH and gL in transfected cells" *J Gen Virol*
23. Bzik, Fox, Deluca et al. (1984) "Nucleotide sequence of a region of the herpes simplex virus type 1 gB glycoprotein gene: mutations affecting rate of virus entry and cell fusion" *Virology*
24. Muggeridge, Grantham, Johnson (2004) "Identification of syncytial mutations in a clinical isolate of herpes simplex virus 2" *Virology*
25. Rogalin, Heldwein (2015) "Interplay between the Herpes Simplex virus 1 gB cytodomain and the gH cytotail during cell-cell fusion" *J Virol*
26. Chowdary, Heldwein (2010) "Syncytial phenotype of C-terminally truncated herpes simplex virus type 1 gB is associated with diminished membrane interactions" *J Virol*
27. Garcia, Chen, Longnecker (2013) "Modulation of Epstein-Barr virus glycoprotein B (gB) fusion activity by the gB cytoplasmic tail domain" *mBio*
28. Klupp, Nixdorf, Mettenleiter (2000) "Pseudorabies virus glycoprotein M inhibits membrane fusion" *J Virol*
29. Ejercito, Kieff, Roizman (1968) "Characterization of herpes simplex virus strains differing in their effects on social behaviour of infected cells" *J Gen Virol*
30. Baghian, Huang, Newman et al. (1993) "Truncation of the carboxy-terminal 28 amino acids of glycoprotein B specified by herpes simplex virus type 1 mutant amb1511-7 causes extensive cell fusion" *J Virol*
31. Foster, Melancon, Kousoulas (2001) "An alpha-helical domain within the carboxyl terminus of herpes simplex virus type 1 (HSV-1) glycoprotein B (gB) is associated with cell fusion and resistance to heparin inhibition of cell fusion" *Virology (Auckl)*
32. Silverman, Greene, King et al. (2012) "Membrane requirement for folding of the herpes simplex virus 1 gB cytodomain suggests a unique mechanism of fusion regulation" *J Virol*
33. Haan, Lee, Longnecker (2001) "Different functional domains in the cytoplasmic tail of glycoprotein B are involved in Epstein-Barr virusinduced membrane fusion" *Virology*
34. Chen, Zhang, Jardetzky et al. (2014) "The Epstein-Barr virus (EBV) glycoprotein B cytoplasmic C-terminal tail domain regulates the energy requirement for EBV-induced membrane fusion" *J Virol*
35. Chen, Jardetzky, Longnecker (2016) "The cytoplasmic tail domain of Epstein-Barr Virus gH regulates membrane fusion activity through altering gH binding to gp42 and epithelial cell attachment" *mBio*
36. Saw, Matsuda, Eisenberg et al. (2015) "Using a split luciferase assay (SLA) to measure the kinetics of cell-cell fusion mediated by herpes simplex virus glycoproteins" *Methods*
37. Browne, Bruun, Minson (2001) "Plasma membrane requirements for cell fusion induced by herpes simplex virus type 1 glycoproteins gB, gD, gH and gL" *J Gen Virol*
38. Atanasiu, Saw, Cohen et al. (2010) "Cascade of events governing cell-cell fusion induced by herpes simplex virus glycoproteins gD, gH/gL, and gB" *J Virol*
39. Pertel, Fridberg, Parish et al. (2001) "Cell fusion induced by herpes simplex virus glycoproteins gB, gD, and gH-gL requires a gD receptor but not necessarily heparan sulfate" *Virology (Auckl)*
40. Kondo, Miyauchi, Meng et al. (2010) "Conforma tional changes of the HIV-1 envelope protein during membrane fusion are inhibited by the replacement of its membrane-spanning domain" *J Biol Chem*
41. Reuter, Kropff, Chen et al. (2024) "The autonomous fusion activity of human cytomegalovirus glycoprotein B is regulated by its carboxy-terminal domain" *Viruses*
42. Hegde, Dunn, Lewinsohn et al. (2005) "Endogenous human cytomegalovirus gB is presented efficiently by MHC class II molecules to CD4+ CTL" *J Exp Med*
43. Mocarski (1988) "Biology and replication of cytomegalovirus" *Transfus Med Rev*
44. Wu, Oberstein, Wang et al. (2018) "Role of PDGF receptor-α during human cytomegalovirus entry into fibroblasts" *Proc Natl Acad Sci*
45. Prichard, Penfold, Duke et al. (2001) "A review of genetic differences between limited and extensively passaged human cytomegalovirus strains" *Rev Med Virol*
46. Murphy, Yu, Grimwood et al. (2003) "Coding potential of laboratory and clinical strains of human cytomegalovirus" *Proc Natl Acad Sci*
47. Day, Stegmann, Schultz et al. (2020) "Polymorphisms in Human Cytomegalovirus glycoprotein O (gO) exert epistatic influences on cell-free and cell-to-cell spread and antibody neutralization on gH epitopes" *J Virol*
48. Suárez, Wilkie, Hage et al. (2019) "Human cytomegalovi rus genomes sequenced directly from clinical material: variation, multiple-strain infection, recombination, and gene loss" *J Infect Dis*
49. Park, Gill, Aghajani et al. (2020) "Engineered receptors for human cytomegalovirus that are orthogonal to normal biology" *PLoS Pathog*
50. Nixdorf, Klupp, Karger et al. (2000) "Effects of truncation of the carboxy terminus of pseudorabies virus glycoprotein B on infectivity" *J Virol*
51. Pötzsch, Spindler, Wiegers et al. (2011) "B cell repertoire analysis identifies new antigenic domains on glycoprotein B of human cytomegalovirus which are target of neutralizing antibodies" *PLoS Pathog*
52. Chandramouli, Ciferri, Nikitin et al. (2015) "Structure of HCMV glycoprotein B in the postfusion conformation bound to a neutralizing human antibody" *Nat Commun*
53. Sponholtz, Byrne, Lee et al. (2024) "Structure-based design of a soluble human cytomegalovirus glycoprotein B antigen stabilized in a prefusion-like conformation" *Proc Natl Acad Sci*
54. Liu, Heim, Chi et al. (2021) "Prefusion structure of human cytomegalovirus glycoprotein B and structural basis for membrane fusion" *Sci Adv*
55. Burke, Heldwein (2015) "Crystal structure of the human cytomega lovirus glycoprotein B" *PLoS Pathog*
56. Lee, Longnecker (1997) "The Epstein-Barr virus glycoprotein 110 carboxy-terminal tail domain is essential for lytic virus replication" *J Virol*
57. Mcshane, Longnecker (2004) "Cell-surface expression of a mutated Epstein-Barr virus glycoprotein B allows fusion independent of other viral proteins" *Proc Natl Acad Sci*
58. Harman, Browne, Minson (2002) "The transmembrane domain and cytoplasmic tail of herpes simplex virus type 1 glycoprotein H play a role in membrane fusion" *J Virol*
59. Browne, Bruun, Minson (1996) "Characterization of herpes simplex virus type 1 recombinants with mutations in the cytoplasmic tail of glycoprotein H" *J Gen Virol*
60. Pignatelli, Dal Monte, Rossini et al. (2004) "Genetic polymorphisms among human cytomegalovirus (HCMV) wild-type strains" *Rev Med Virol*
61. Bangari, Mittal (2004) "Porcine adenoviral vectors evade preexisting humoral immunity to adenoviruses and efficiently infect both human and murine cells in culture" *Virus Res*
62. Krummenacher, Baribaud, Sanzo et al. (2002) "Effects of herpes simplex virus on structure and function of nectin-1/ HveC" *J Virol*
63. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
64. Waterhouse, Bertoni, Bienert et al. (2018) "SWISS-MODEL: homology modelling of protein structures and complexes" *Nucleic Acids Res*
65. Meng, Goddard, Pettersen et al. (2023) "UCSF ChimeraX: tools for structure building and analysis" *Protein Sci*
66. Robert, Gouet (2014) "Deciphering key features in protein structures with the new ENDscript server" *Nucleic Acids Res*
67. Mcguffin, Bryson, Jones (2000) "The PSIPRED protein structure prediction server" *Bioinformatics*
68. Madeira, Madhusoodanan, Lee et al. (2024) "Using EMBL-EBI services via web interface and programmatically via web services" *Current Protocols* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724371&blobtype=pdf | # Human antibody targeting of coronavirus spike S2 subunit is associated with protection mediated by Fc effector functions
Krithika Muthuraman, Matthew Jackman, Yu Liang, Meghan Garrett, Hong Cui, Loan Vu, Hong Nguyen, Danton Ivanochko, Chengjin Ye, Paula Pino, Amberlee Hicks, Billie Maingot, Erik Yusko, Sharon Benzeno, Luis Martínez-Sobrido, Jordi Torrelles, Amy Gilbert, Benjamin Evan, Russell Rubin, Gladys Keitany, Arif Jetha, Jean-Philippe Julien
## Abstract
Over the past two decades, betacoronaviruses (β-CoVs) have caused two epidemics and a pandemic and remain a high risk for future outbreaks through zoonotic transmissions, highlighting the need for broad biomedical countermeasures. Here, we describe a convalescent human monoclonal antibody (mAb 1871) that targets the S2 subunit of the coronavirus spike protein, with broad β-CoVs binding and sarbecovirus neutralization. Cryo-electron microscopy analysis revealed that mAb 1871 binds the upstream helix of the S2 subunit, interacting with partially conserved residues, providing a molecular basis for its cross-reactivity. Though less potent than receptor-binding domain-directed antibodies-approximately 500-fold lower neutralization potency than the emergency use authorized receptor-binding domain (RBD)-directed Pemgarda mAb against wild-type SARS-CoV-2-mAb 1871 provides protective efficacy in a mouse model. Notably, Fc effector functions are critical for its in vivo protection. This study further highlights the Fc dependence of S2-directed antibodies for in vivo protection and identifies a conserved epitope in the S2 subunit as a potential target of broad-β-CoVs countermeasures.IMPORTANCE Bats and pangolins are natural reservoirs of betacoronaviruses (β-CoVs) and continue to pose a significant risk for future outbreaks through zoonotic transmis sions. This highlights the need for effective countermeasures to prevent future pandem ics. While neutralizing antibodies targeting the receptor-binding domain of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) received emergency use authorization, many have lost efficacy as the virus evolved, and authorizations have been revoked. In contrast to the S1 subunit, the spike protein S2 subunit is more conserved across β-CoVs, making it an attractive target for the development of broadly neutralizing antibodies. Here, we describe a human mAb that targets a conserved epitope in the S2 subunit, demonstrating broad β-CoV binding, sarbecovirus neutralization, and in vivo protection mediated by Fc effector functions in a mouse model. These findings have important implications for pan-β-CoVs therapeutics and vaccine development.KEYWORDS cryo-EM structure, in vivo protection, S2 subunit, coronavirus spike, antibody S evere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused the COVID-19 pandemic, which presented an enormous economic and public health challenge. SARS-CoV-2 belongs to the family of Coronaviridae and the genus Betacoronavirus. Betacoronaviruses (β-CoVs) are one of the four genera of coronavirus (alpha, beta, gamma, and delta), with a large 30 kbp single-stranded (ss)RNA, that usually cause mild to moderate upper-respiratory tract infections. Previous coronavirus zoonotic transmis sions have caused widespread infections and led to the circulation of four endemic
human coronaviruses e.g., HCoV-229E, HCoV-NL63 (belonging to alphacoronavirus genus), HCoV-OC43, and HCoV-HKU1 (belonging to β-CoVs genus) that cause non-severe seasonal respiratory infections (1). However, in the last two decades, β-CoVs have caused two epidemics and a pandemic, namely SARS (2002), Middle Eastern respiratory syndrome (MERS) (2012), and COVID-19 (2019), with high lethality. Bats and rodents are natural reservoirs of β-CoVs, posing a continued high risk for future outbreaks through zoonotic transmissions (2,3). Therefore, to ensure pandemic preparedness, the swift development of broad biomedical countermeasures against emerging β-CoVs is paramount.
Pathogenesis of β-CoVs begins with receptor binding via the trimeric spike surface glycoprotein, followed by enzymatic cleavage and a conformational change in the spike, leading to fusion of viral and cell membranes. Spike protein (SP) consists of two subunits, namely, the S1-subunit consisting of the receptor-binding domain (RBD) and the S2-subunit containing the fusion machinery. SP is an immunodominant target for antibodies (Abs). RBD-directed Abs can block receptor binding and are the most potent; hence, these have been the major focus of therapeutic Abs and vaccine discovery efforts. However, the emergence of SARS-CoV-2 variants of concern (VOCs) (4) quickly rendered SARS-CoV-2 RBD-directed Ab interventions, such as the Bamlanivimab and Etesevimab cocktail (5), or the REGN-COV cocktail (6), ineffective to treat COVID-19, consequently losing their emergency use authorization. The loss in potency is due to a constellation of mutations within the SARS-CoV-2 SP, especially within the RBD (7). The S1 subunit exhibits less conservation among β-CoVs due to different use of entry receptors and because it is subject to significant immune pressure, which leads to increased mutations aimed at evading immune responses (8). In contrast, the S2 subunit shows a relatively higher conservation of 63%-98% (9) across different β-CoVs, which makes it an attractive target for the development of broad β-CoV-neutralizing Abs.
Several Abs targeting two conserved sites in the SARS-CoV-2 SP S2 subunit-the stem helix (10)(11)(12)(13)(14) and fusion peptide (15,16)-isolated from humans and vaccinated animals display broad β-CoV neutralization profiles, although at lower potencies than RBD-directed Abs. Furthermore, these Abs are shown to provide in vivo protection despite relatively low in vitro neutralization potency, with this effect attributed to their dependence on Fc effector functions (10,11). Identification of pan-β-CoVs Abs could aid in pandemic preparedness and rapid deployment in the event of an outbreak. Further more, information from these Abs that target novel conserved epitopes could be used for structure-based design of pan-β-CoV vaccines (17)(18)(19).
ADPT01871 (herein referred to as mAb 1871), a S2 subunit-directed Ab isolated from a COVID-19 convalescent donor, displayed broad β-CoV binding, SARS-CoV and SARS-CoV-2 neutralization, and blocked in vitro cell membrane fusion (20). Despite relatively low in vitro potency, mAb 1871 provided protection against SARS-CoV-2 in a mouse challenge study (20). Here, we sought to expand our understanding of the molecu lar basis of binding for mAb 1871. Indeed, previous efforts in epitope identification had proved challenging due to differences in prefusion and postfusion conformations, flexibility of target regions, and low affinity of this mAb. Here, we used cryo-electron microscopy (cryo-EM) for the full molecular characterization of mAb 1871 and delineated its epitope on the S2 subunit. We also further describe the importance of effector function for this SP S2-subunit directed mAb in providing in vivo protection.
## RESULTS mAb 1871 possesses broad betacoronavirus binding
mAb 1871 was isolated from Ab-secreting cells of a 30-year-old male infected with SARS-CoV-2 in the acute phase of illness, 9 days post-symptom onset, using pairSEQ technology (20). mAb 1871 can bind a broad panel of β-CoV full-length SP (SARS-CoV, SARS-CoV-2, MERS-CoV, HCoV-HKU1, and HCoV-OC43) and SP S2 subunits, and can neutralize SARS-CoV, SARS-CoV-2 wild-type USA/WA1/2020, alpha and beta VOC pseudoviruses with IC 50 s of 30, 0.44, 0.11, and 0.21 nM, respectively (20). This broad binding and neutralization profile suggests that mAb 1871 may be directed against one of the conserved epitopes in the S2 subunit domain, where a high sequence conserva tion is observed among β-CoVs (Fig. 1A) (9). To investigate whether mAb 1871 binding is dependent on SP conformation, we interrogated binding to full-length SP (SARS-CoV, SARS-CoV-2, RatG13, MERS, HKU1, OC43), full-length SP trimer stabilized in the prefusion conformation (SARS-CoV-2 SP full length with a polybasic cleavage site deletion and stabilizing mutations K986P and V987P, wild-type numbering [21]), and S2 subunit-only (SARS-CoV, SARS-CoV-2, MERS, OC43) proteins by biolayer interferometry. mAb 1871 bound to the full-length SP β-CoVs panel with an apparent K D in the range of 0.1 to 7 nM (Fig. 1B andC). mAb 1871 also bound to S2 subunit domains (Fig. S1). However, binding was not observed to the prefusion stabilized full-length SARS-CoV-2 SP trimer (Fig. 1B; Fig. S1), thereby suggesting that the recognized epitope may not be available in the prefusion state and might only become exposed after a conformational change occurs.
Although mAb 1871 was isolated only 9 days post-symptom onset, a high occur rence of somatic hypermutations (SHM) was observed (20). Interestingly, the fragment antigen-binding (Fab) 1871 displayed high binding affinity to OC43 S2 subunit (Fig. 2A andB) compared to MERS, SARS-CoV, and SARS-CoV-2 (Fig. S2), suggesting that this mAb might have initially developed during an infection by this commonly encountered β-CoV.
## OC43 S2 subunit adopts an elongated postfusion conformation
We used cryo-EM to understand the molecular basis of S2 subunit recognition by Fab 1871. For this purpose, we focused on the high-affinity OC43 S2 subunit construct comprising residues 757-1,291 along with a C-terminal T4 foldon trimerization domain (Fig. 2A). Analysis of cryo-EM micrographs and 2D classes revealed three Fabs bound to an elongated structure reminiscent of the post-fusion SP of SARS-CoV-2 (22), SARS-CoV (29), and MHV(32) (Fig. S3). Since most particles observed tended to have an orientation bias (Fig. S3J), data were collected at a 40° tilt to improve the distribution of orientations (Fig. S4). When C3 symmetry was applied, the single particle reconstructions of the Fab-S2 subunit complex generated a map at a resolution of 2.6 Å (Fig. 2C; Fig. S3 and S4 Table S1).
## Full-Length Text
The cryo-EM reconstruction revealed that the OC43 S2 subunit adopts a similar structure to the postfusion conformation of both SARS-CoV-2(30) (PDB ID: 8FDW, rmsd = 1.1 Å) and SARS-CoV (29) (PDB ID: 6M3W, rmsd = 1.2 Å) (Fig. 2D), corroborating high sequence similarity among these β-CoV strains (Fig. S5). The regions resolved in the OC43 structure include almost all domains; linker-1 (L1; 783-808), upstream helix (UH; 809-850), heptad repeat 1 (HR1; 995-1,067), central helix (CH; 1,068-1,118), beta hairpin (BH; 1,119-1,155), sub-domain 3 (SD3; 11,56-1,207), linker-2 (L2; 1,208-1,246), but not the fusion peptide (FP; 903-917), connecting region (CR; 928-995), and heptad repeat 2 (HR2; 1,247-1,281) (Fig. 2E).
As previously reported for other postfusion SP structures (29,30), the core of OC43 S2 subunit forms a conical shape that is approximately 135 Å long and 50 Å wide, having a long three-helix bundle formed by CH and HR1, with a three-stranded β-sheet wrapping around the C-terminal end of the CH bundle (Fig. 2E). In the prefusion conformation, the S2 subunit is short and then undergoes a large conformational change during which the HR1 and connecting loops re-orient with the central helix to form a single long continuous alpha-helix (Fig. 2F, Fig. S6A). This forms the stable triple helix bundle of the post-fusion S2 subunit core. Four disulfide bonds in the structure, two within the UH (C816-C838, C821-C827), one in BH (C1116-C1126), and one between the SD3 and L2 (C1166-C1211) are conserved both in the pre-fusion and post-fusion conformations (Fig. 2F andG; Fig. S6B). These disulfide bonds help the UH, BH, and SD3 domains retain their tertiary structure between the two conformational states.
## mAb 1871 binds to the upstream helix in the S2 subunit
To gain further confidence in the mAb 1871 structure, the unliganded 1871 Fab was solved by X-ray crystallography to 2.5 Å resolution (Table S2) and used as a starting model to build in the cryo-EM map. The low rsmd value of unliganded and liganded Fab structures for the variable region heavy (0.3 Å) and light chain (0.3 Å) suggests that the antigen and Ab largely interact via a lock and key mechanism. Each 1871 Fab interacts with two OC43 S2 subunit protomers via a conformational epitope consisting of UH, BH, and L2 (Fig. 3A; Fig. S6D), with UH contributing the majority of the interface with 25 interacting residues (Fig. 3B). The buried surface area (BSA) of the S2 subunit in the interface is 1,024 Å 2 , with 541 Å 2 conferred by the Ab heavy chain and 483 Å 2 conferred by the light chain. Residues from heavy chain complementarity-determining regions (CDRs) H2 and H3, light chain CDRs K1, K2, and K3, along with four residues in the kappa chain framework, form part of the interface (Fig. 3C). The four framework residues of kappa chain-Y49, I52, G64, and S65-interact with BH. Of these, residue I52 is mutated from the germline IGKV3-15*01 residue (T52I). Eight residues in the UH interact with seven residues in the heavy chain CDRs H2 and H3, and five residues in the light chain CDRs K1 and K3 via an extensive network of 18 hydrogen bonds (Fig. 3C; Table S3). Interfacing residues in L2 and BH do not contribute any hydrogen bond or salt bridge interactions. To further validate the structure interface, we performed binding to UH peptides of varying lengths. IgG bound to all three peptides of 12, 32, and 43 amino acids, whereas the Fab showed very weak binding only to the 43 amino acid peptide at the highest Fab concentration (Fig. 3D). Notably, the peptides contain different interface residues involved in the interaction, with the 12-aa peptide containing 5 hydrogen bonding residues, the 32-aa peptide containing 6 hydrogen bonding residues, and the 43-aa peptide containing all relevant residues. This suggests that, due to avidity, the IgG can bind to an incomplete epitope (shorter peptides), whereas the Fab, lacking avidity, displays weak binding only to a near-complete epitope (42-aa peptide). However, when the full epitope is presented, in the context of the S2 subunit, as observed in the structure, the Fab binds with high affinity.
Furthermore, the CDRH3 of 1871 Fab forms an anti-parallel β-sheet with the UH residues 811-815 via interaction through three hydrogen bonds (Fig. 3E). In the prefusion conformation (PDB ID: 7PNM) (31), hydrogen bond interactions between antiparallel β-strands in the CR and UH stabilize the region. Upon conformational change, the CR, along with the fusion peptide, moves toward the C-terminus of the postfusion structure. During this conformational change, the epitope on UH becomes available for 1871 binding, in a way that stabilizes the UH region in a similar way to the CR interaction (Fig. 3E). These structural insights help to better understand why 1871 does not bind to the prefusion stabilized SARS-CoV-2 trimer but binds to the S2 subunit (Fig. S1), as the stabilizing mutations lock the SP in its prefusion conformation, preventing movement in the CR. Furthermore, the UH region in the prefusion and postfusion conformations of OC43 has an rmsd value greater than 1.5 Å, indicating that structural differences may inhibit binding (Fig. S6C). S2-directed antibodies like S2P6 have been shown to bind both prefusion and postfusion S conformations, and their likely mechanism of action involves impeding fusogenic rearrangements (10). On the other hand, 1871 does not bind prefusion stabilized S (Fig. 1B; Fig. S1), and the resolved structure shows 1871 bound to a postfusion conformation (Fig. 2C). Together with functional assays demonstrating the ability of 1871 to block cell fusion ( 20), we propose Full-Length Text the possibility that binding by 1871 to its UH epitope may also impede fusogenic rearrangements.
As structural analysis of SARS-CoV-2 spike in complex with 1871 Fab would provide the most relevant information pertaining to functional data, we performed Alphafold 3 (AF3) modeling (33). Modeling of the SARS-CoV-2 spike UH region with 1871 Fab revealed a binding profile similar to OC43 UH with 1871 Fab (Fig. S7A). Alignment of a single protomer from the AF3 SARS-CoV-2 model with the OC-43 upstream helix region provided an rmsd of 0.76 Å (Fig. S7B). Interface analysis using PDB PISA showed that the predicted interfacing residues between CoV-2 and 1871 were largely consistent with those seen in the OC43 complex, including several key hydrogen bonds (Fig. S7C).
## Upstream helix epitope is partially conserved across β-CoVs
Known β-CoV-neutralizing Abs directed against the SARS-CoV-2 SP S2 subunit target two major epitope bins: the stem helix (1,140-1,160) and the fusion peptide (840-852) (Fig. 4A) (11,12,15,16). Here, we show that the upstream helix is another conserved epitope in the S2 subunit. Sequence analyses across β-CoV lineages (Fig. 4B) showed that there are 6/21 fully conserved and 5/21 partially conserved residues in the stem helix, 8/13 fully conserved and 2/13 partially conserved residues in the fusion peptide, and 8/42 fully conserved and 11/42 partially conserved residues in the upstream helix (Fig. 4B). Here, partially conserved refers to residues in bins 6-8 and fully conserved refers to residues in bin 9 as per the consurf database scheme (23)(24)(25)(26)(27)(28), where all bins 6-9 indicate slowly evolving conserved sites.
Structural analysis of postfusion structures of SARS-CoV-2 (PDB ID: 8FDW) ( 21), SARS-CoV (PDB ID: 6M3W) (30), and OC-43 SP S2 subunits showed that the UH region is exposed and available for binding by mAb 1871 (Fig. 2D). Furthermore, the UH region has a rmsd of 0.6 Å and 0.7 Å for SARS-CoV-2 and SARS-CoV, respectively (Fig. S8A), indicative of structural similarity and further explaining binding of mAb 1871 to all three S2 subunits.
Although mAb 1871 bound to a broad range of β-CoVs, it failed to neutralize the SARS-CoV-2 Omicron BA.1 and BA.5 VOCs (Fig. 4C; Fig. S8B). In this context, a single point mutation, N842K, occurs in the UH epitope of both Omicron BA.1 and BA.5 VOC, which we posit, based on structural modeling, may lead to steric clashes with the mAb 1871 light chain and presumably abrogate neutralization potency. This point mutation persists in more recent Omicron subvariants, which we predict would not be neutralized by 1871, considering the emerging trend of decreased activity of 1871 against viruses carrying a lysine at position 842 (Fig. 4D andE).
Like Ab S2P6 (11), which binds β-CoVs but not alphacoronaviruses, mAb 1871 also shows binding to β-CoVs (SARS-CoV, SARS-CoV-2, MERS, OC-43, and HKU-1) but does not bind to alphacoronavirus HCoV-229E (20). Molecular modeling suggests that three residues that are involved in hydrogen bonding with Lys 811, Thr 813, and Asp 839 are mutated in HCoV-229E, which could lead to changes in affinity and may explain the inability of mAb 1871 to bind to this alphacoronavirus (20) (Fig. 4D andF). Overall, our molecular understanding of the partially conserved mAb 1871 epitope in the UH aligns well with its ability to bind and neutralize sarbecoviruses. mAb 1871 is dependent on effector functions to confer in vivo protection SP S2 subunit-directed Abs have been shown to confer in vivo protection despite weaker in vitro neutralization potency compared to RBD-directed Abs (10)(11)(12). This effect has been attributed to leveraging Fc effector functions through viral clearance and promoting antiviral immune responses (10,11). Here, to assess the role of effector functions for the potency of mAb 1871, we generated mAb 1871 with a wild-type (WT) human Immunoglobulin 1 (IgG1) Fc domain and one with a human IgG1 Fc containing L234A, L235A, and P329G mutations to ablate binding to Fcγ receptors (herein referred to as mAb 1871 IgG1 EL). Mutations in the Fc of IgG1 EL did not affect binding to SARS-CoV-2 full-length antigen compared to the WT IgG1 and maintained the expected, unaltered high-affinity binding to human FcRn at pH 5.6 (Fig. 5A). However, mAb 1871 IgG1 EL displayed the expected, ablated binding to other mouse Fc receptors compared to mAb 1871 IgG1 (Fig. 5A). To investigate the effector function impact of this loss of Fc receptor binding, we assessed in vitro Ab-dependent cellular phagocytosis (ADCP) using a multiplexed bead-based assay. mAb 1871 IgG displayed high induction of ADCP, whereas, as expected, mAb 1871 IgG EL did not display any ADCP function (Fig. 5B).
To assess the impact of ablated Fc effector functions on in vivo protection against SARS-CoV-2, K18 human angiotensin-converting enzyme 2 (hACE2) and human FcRn double-transgenic mice were intraperitoneally treated with 1.5 mg/kg of the mAb 1871 IgG1 and mAb 1871 IgG1 EL and then challenged intranasally with a lethal dose of 1 × 10 4 PFUs of SARS-CoV-2 USA-WA1/2020 (Fig. 5C). Despite a relatively moderate neutralization potency (IC 50 0.051), though both tests trend towards improved survival with the 1871 IgG1 (Fig. 5D; Fig. S9). These results suggest that Abs targeting the UH epitope in the SP S2 subunit rely on effector functions to provide protection against SARS-CoV-2 challenge.
## DISCUSSION
Currently, two β-CoV-based diseases, MERS and COVID-19, are on the World Health Organization's (WHO) priority disease list (36). During the recent COVID-19 pandemic, the global toll reported reached over 7 million deaths (37). Repeated emergence of β-CoV-based epidemics caused by zoonotic transmissions warrants the development of therapeutics or prophylactics to combat future outbreaks (38). Several Abs that received emergency use authorization during the COVID-19 pandemic were revoked due to rapid mutations in their targeted epitopes, and SARS-CoV-2 vaccines are being consistently modified to cater to the new VOCs (4, 7). Therefore, the development of pan-β-CoV therapeutics and prophylactics is paramount to support readiness and global stability in the event of a future coronavirus pandemic.
The SP S2 subunit, comprising the fusion machinery, is more conserved than the SP S1 subunit, which is subjected to high immunological pressure. Hence, it is hypothe sized that Abs directed against the S2 subunit may provide an opportunity to develop pan-β-CoV therapeutics. Abs directed against two conserved S2 subunit epitope bins, namely the stem helix and fusion peptide, have been described (10-12, 15, 16). In addition, a few antibodies targeting other conserved epitopes in the S2 subunit, such as the HR2 epitope recognized by murine hMab5.17 (39) and the S2 apex epitope (between HR1 and CH) recognized by mAb 54043-5 (40), have been identified. Murine hMab5.17 displayed neutralization across multiple SARS-CoV-2 VOCs and conferred in vivo protective efficacy in a hamster challenge model (39). Although 54034-5 is non-neutralizing, this antibody-isolated from a human convalescent donor-displayed broad β-CoV binding and provided 40% protection in vivo in an Fc-attenuated format at a high dose (12 mg/kg) (38). These findings suggest that antibodies directed against conserved epitopes in S2 may act through various mechanisms to provide broad protection against β-CoVs. Further characterization of such antibodies may help better understand their modes of action and support vaccine design strategies to target conserved epitopes.
Here, we have characterized functional and molecular details of mAb 1871, which has broad reactivity to β-CoVs. Using cryo-EM, we elucidated that mAb 1871 targets the upstream helix, a partially conserved epitope in the S2 subunit. The upstream helix consists of 43 amino acids and retains its tertiary structure in the prefusion and postfusion SP conformations. The epitope consists of highly conserved disulfide bonds that help the S2 subunit to retain its structural integrity during conformational changes. We identified that mAb 1871 does not bind to the prefusion-stabilized SP. Indeed, the upstream helix, as recognized by mAb 1871, is occluded in the prefusion conformation due to the presence of the S1 subunit. Presumably, mAb 1871 binds to the upstream helix epitope when it becomes exposed during an intermediate stage when the SP undergoes conformational changes and interferes with the fusion machinery's ability to mediate membrane fusion as a mechanism to neutralize the virus.
Although we show that mAb 1871 has broad reactivity to a β-CoV panel and targets a partially conserved epitope in the S2 subunit, our study has some limitations. Indeed, this Ab loses neutralization against the SARS-CoV-2 Omicron BA.1 VOC likely due to an Asn842 to Lys842 mutation in the binding site, which presumably leads to steric clashes and loss of binding. Since this is one of the first mAbs described against the UH site, there may be further opportunities to identify other Abs that target more conserved residues in the region for improved breadth.
RBD-directed mAbs are potent neutralizers; for example, mAbs such as ADPT03019, ADPT0980, and ADPT03995 have in vitro neutralization potency in the range of 1-4 ng/mL (20), whereas mAb 1871, which is S2 directed, is a weak neutralizer with a potency of 1,500 ng/mL against SARS-CoV-2 US/WA-1. In a K18 hACE2 mouse model challenge study, when challenged intranasally with 1 × 10 5 PFU of SARS-CoV-2/US WA1/2020, these RBD-directed Abs had greater than 80% survival when administered at 1.5 mg/kg and mice treated with mAb 1871 at 5 mg/kg had 70% survival and presented minimal morbidity signs (weight loss) at 10 days post-infection (20). Here, we show that K18-hACE2 hFcRn double transgenic mice treated at the lower dose of mAb 1871 (1.5 mg/kg) and challenged with a lethal dose of 1 × 10 4 PFU of SARS-CoV-2/WA1/2020 provided 60% survival. Although we observed continued protection at a lower dose compared to the 5 mg/kg dose in the initial study, a variation in the mouse model and challenge dose between the studies is a criterion that could also play a role in the mAb circulation, exposure, and protection efficacy. Furthermore, in a similar transgenic hACE2 mouse model, S2-directed mAb CC40.8 displays dose-dependent protection (300, 100, 50, and 10 µg per animal) with minimal weight loss, and mAb ADPT01823 demonstrates 60% survival when administered at 5 mg/kg. These examples illustrate the protective efficacy of S2-directed antibodies in a transgenic hACE2 mouse model. However, it is important to note that these antibodies have not been evaluated in a double transgenic hACE2/hFcRn model, limiting direct comparison. Future benchmarking experiments of S2-directed mAbs in the same challenge model and experimental setup would enable direct activity relationships to be derived.
We hypothesized that even though mAb 1871 has weak neutralization compared to RBD-directed Abs, its in vivo protection might be due to Fc effector functions. To evaluate this possibility, we generated mAb 1871 IgG1 and mAb 1871 IgG1 EL with Fc silencing mutations L234A, L235A, and P329G to ablate binding to Fcγ receptor. Our results confirmed that mAb 1871 IgG1 EL exhibited reduced binding to Fc gamma receptors and a loss of ADCP activity. Interestingly, mAb 1871 IgG1 exhibited high ADCP activity at lower concentrations, but this activity diminished at higher concentrations. This effect may be attributed to the prozone effect or increased monovalent binding to Fcγ receptors, which could reduce receptor clustering and downstream signaling or potential antibody aggregation at higher concentrations. In vivo studies demonstrated that, when treated at the same dose, mAb 1871 IgG1 EL provided only 20% survival when compared to 60% survival with mAb 1871 IgG1. Notably, mice treated with mAb 1871 exhibited greater initial weight loss followed by recovery, whereas the group treated with RBD-directed antibody showed minimal weight loss. This may be attributed to the difference in the mechanism of action of these antibodies, and that mAb 1871 may rely more heavily on Fc effector function than on neutralization, which may allow some early viral replication. This result underscores the importance of Fc effector function for this S2 subunit-directed mAb. Our data contribute to expanding evidence that SP S2-subunit-directed mAbs benefit from effector function.
Other examples include CC40.8 (10), 76E1 (13), CC99.103 (12), CC25.106 (12), CC95.108 (12), and CC68.109 (12) that show improved in vivo protection despite relatively low neutralization potency compared to RBD-directed Abs. Fc effector functions have also been reported to be significant in therapeutic settings when compared to pro phylactic settings for RBD-directed Abs (41). In prophylactic studies in mouse models, Fc-silenced RBD-directed Abs have shown mixed results. COV2-2050 (42) failed to prevent mouse weight loss or reduce viral load, CV3-1 (43) showed a 100% mortality rate, SC31 (44) had 50% worse survival rates, and WRAIR-2123 (45) had 70% survival. These findings emphasize the critical role of neutralization in providing protection, especially in the absence of effector functions for RBD-directed Abs. Further studies could explore Fc engineering strategies to enhance effector functions, potentially improving the protective efficacy of SP S2 subunit-directed Abs. Additionally, combining RBD-directed and S2-directed antibodies in antibody cocktails or multispecific formats (46) could enhance both potency and breadth of neutralization. Identifying antibodies targeting distinct S2 epitopes may facilitate such approaches, contributing to the development of more effective pan-β-CoVs therapeutic strategies.
Here, we elucidate the binding of mAb 1871 to the upstream helix in the postfusion state, providing a glimpse into this conformation recognized by a mAb recovered from natural infection. Further studies to understand the events that lead to the unmasking of this epitope could provide detailed insight into the mechanism of fusion inhibition. The epitope uncovered for mAb 1871 also provides molecular details to be leveraged for structure-guided vaccine design efforts that specifically target the SP S2 subunit.
## MATERIALS AND METHODS
## Expression and purification of Fab and IgG
Genes encoding the heavy and light chains of Fab and IgG were synthesized and cloned by Geneart (Life Technologies) into a pcDNA3.4 expression vector. Fab and IgG were transiently expressed in HEK-293F cells (Thermo Fisher Scientific). Cells were seeded at a density of 0.8 × 10 6 cells/mL and were incubated for 24 h at 37°C, 8% CO 2 at 125 rpm in a Multitron Pro Shaker (InforS HT). Within 24 h, cells were co-transfected with 90 µg of DNA consisting of heavy chain and light chain in a 2:1 ratio, preincubated with polyethylenimine (PEI) (Polysciences) at room temperature for 10 min. After 6-7 days, cell suspensions were harvested by spinning at 6,000 rpm for 20 min, and supernatants were filtered using a 0.2 µm Steritop filter (EMD Millipore). Fab was purified using KappaSelect affinity (Cytiva), and IgGs were purified using protein A affinity chromatography (Cytiva); eluted using 100 mM glycine (pH 2.2), and neutralized with 1 M Tris-HCl (pH 9.0). Fab fractions were further purified using cation exchange chromatography (MonoS, Cytiva), and IgG fractions were purified using size exclusion chromatography (Superdex200, Cytiva). All IgGs for in vivo experiments were tested to ensure endotoxin concentrations were below 3.5 EU/mL at 1 mg/mL concentration.
## Design, expression, and purification of antigens for biolayer interferometry
Gene encoding the OC43 antigen, consisting of the extracellular S2 subunit domain (753-1240 GenBank: UOP57224.1) with a downstream TEV protease cleavage site, T4 foldon domain, and Histidine tag, was synthesized and cloned by Geneart (Life Technologies) into a pcDNA3.4 expression vector. HEK-293S (GnT I -/-; Thermo Fisher Scientific) cells (200 mL) were seeded at a density of 0.8 × 10 6 cells/mL and were incubated for 24 h at 37°C, 8% CO 2 at 125 rpm in a Multitron Pro Shaker (InforS HT). Cells were transfected with 50 µg of DNA preincubated with polyethylenimine (PEI) (Polysciences) at room temperature for 10 min. After 6-7 days, cell suspensions were harvested by spinning at 6,000 rpm for 20 min, and supernatants were filtered using a 0.2 µm Steritop filter (EMD Millipore). OC43 S2 subunit was purified using HisTrap affinity chromatography column (Cytiva) followed by size exclusion chromatography (Superose 6, Cytiva). Full-length prefusion stabilized SARS-CoV-2 spike ectodomain (BEI NR52394) was purified as described (47) using HisTrap-NiNTA column (Cytiva) followed by a Superose 6 column (Cytiva) in a 20 mM phosphate pH 8.0, 150 mM NaCl buffer. Recombinant mouse Fc receptors (FcγRI, FcγRIIb, FcγRIIIa, and FcγRIV) and human FcRn were expressed and purified as described (34,48) using a HisTrap-NiNTA (Cytiva) column followed by Superdex 200 increase column (Cytiva).
## Biolayer interferometry
An Octet Red96 Biolayer Interferometer instrument (Sartorius ForteBio) was used to assess binding kinetics. His-tagged full-length spike proteins SARS-CoV (Sino Biological, Cat#40634-V08B), SARS-CoV-2, MERS (Sino Biological, Cat#40069-V08B), HKU-1 (Sino Biological, Cat#40606-V08B), OC-43 (Sino Biological, Cat#40607-V08B), HCoV-229E (Sino Biological, Cat#40605-V08B) or S2 subunit SARS-CoV (Sino Biological, Cat#40150-V08B), SARS-CoV-2 (Sino Biological, Cat#40590-V08B), MERS (Sino Biological, Cat#40070-V08B), OC-43 or SARS-CoV-2 full-length prefusion stabilized trimer was loaded onto Ni-NTA biosensors (Sartorius ForteBio) until it achieved a 0.5-0.8 nm signal response or reached a maximum loading time of 800 s. The antigen loading time varied from 50 s to 800 s, resulting in 0.5-0.8 nm signal response depending on antigen (Fig. S10). Association was measured by transferring sensors to wells containing serial dilutions of the IgGs or Fab (500, 250, 125, 62.5, 31.2, and 15.6 nM) for 180 s, and dissociation rates were measured by transferring the sensors to buffer-containing wells for 180 s. Statistics are derived from concentrations across serial dilutions calculated from Sartorius Fortebio software. His-tagged mouse FcγRI, FcγRIIb, FcγRIIIa, FcγRIV, or human FcRn were used to assess the binding of the antibody to Fc receptors. The following biotinylated pepti des of the upstream helix were synthesized at GenScript: 12aa-SSPKVTIDCAAFGSGSG, 32aa-SSPKVTIDCAAFVCGDYAA CKLQLVEYGSFCDGSGSG, and 43aa-SPKVTIDCAAFVCG DYAACKLQLVEYGSFCDNINAILT EVNEGSGSG. Peptides were loaded on SAX biosensors (Sartorius ForteBio) until a 0.8 nm signal response was reached, and association and dissociation were measured as described for Ni-NTA sensors.
## Crystallization and structure determination of 1871 Fab
Purified 1871 Fab was concentrated to 5 mg/mL, and crystallization trials were set up by sitting drop vapor diffusion using the MCSG1 sparse screen matrix in a 1:1 protein:reser voir ratio. Crystals grew in a condition containing 0.1 M sodium acetate, pH 4.6, and 3.5 M sodium formate. Crystals were cryo-protected in 15% (vol/vol) ethylene glycol and flash frozen. X-ray diffraction data were collected at the Canadian Light Source on beamline (CMCF-ID). The data set was processed using XDS (49) and XPREP (50). Phaser (51) was used to determine phases with a 1871 Fab search model predicted by Abodybuilder (52). Further refinement was performed using Phenix (53), and structure building was done using Coot (54). All software was accessed through SBGrid (55). Pymol (56) was used to generate figures. The structure has been deposited in the Protein Data Bank under the accession number PDB ID 9NQ3.
## Expression and purification of OC43 S2 construct for antibody-antigen structure determination
OC43 S2 extracellular domain was purified as described above. For structure deter mination, the fractions collected from HisTrap affinity chromatography were further processed. Fractions collected were then treated with Endoglycosidase H (New England Biolabs) for 30 min at 4°C and purified via size exclusion chromatography (Superose6, Cytiva).
## Cryo-EM structure determination of the OC43 S2-1871 Fab co-complex
A complex of the OC43 S2 and 1871 Fab was obtained by mixing excess Fab:antigen to ensure full occupancy of the binding sites. After 30 min of incubation at 4°C, the complex was purified using size exclusion chromatography (Superose6, Cytiva). The fractions of interest were then concentrated to 250 µg/mL. Two gold grids made in-house were prepared as previously described (57) and were glow-discharged for 15 s at a current of 25 mA. Using a Leica EM GP2 plunge freezer, two grids were prepared using 3 s preblot time, 3 s blot time, and 3 µL of sample volume.
A Titan Krios equipped with a Falcon 4 camera was used to collect a total of 3,325 and 1,470 movies with 0° and 40° tilt, respectively (54 e -/Å 2 , 1.03 Å/pix). All single particle analysis was performed using CryoSPARC v4.3.1 (58). Patch motion correction, patch CTF estimation, exposure curation, blob picking, template picking, and particle stack cleaning using multiple 2D classifications (and the rebalance classes job for the 0° tilt data set) were done separately for the two datasets, with a total of 324,859 and 119,216 particles Fourier-cropped by a factor of 2 for the 0° and 40° tilt data sets, respectively. Combining the two particle stacks and running 2D classification to remove junk particles resulted in a total of 444,075 particles. Ten thousand particles from the combined particle stack were used for ab initio reconstruction. Templates were created from the resulting map for a second template picking from the 40° tilt micrographs, to maximize the number of rare views. The resulting particle stack and the 0° tilt particle stack prior to 2D classification with the original 40° tilt particles underwent exposure curation to remove low-quality particles by exclusion of low-quality micrographs; a combined total of 367,895 particles was kept. This particle stack and the map from ab initio reconstruction were used as input for non-uniform (NU) refinement with C1 symmetry, reaching a resolution of 4.48 Å. The 3D classification with three classes was performed using a mask that excluded density from the conserved domain of the 1871 Fab particles, followed by heterogene ous refinement with C3 symmetry. One class (109,888 particles, 4.41 Å) contained most of the density for OC43 and was used for three rounds of NU-refinement with alternating symmetry settings (C1, C3, C1) to aid in correct particle angular assignment; the resulting map reached 4.48 Å. The particle stack was re-extracted without Fourier cropping and used for NU-refinement with C1 symmetry, reaching a resolution of 3.09 Å with 109,156 particles. The particles were split into ten exposure groups, underwent local and global CTF refinement (trefoil and tilt only), NU-refinement (C1 symmetry), then another round of local and global CTF refinement (all corrections including magnification anisotropy), and two final rounds of NU-refinement (C1 symmetry, then C3 symmetry); this map (map A) reached a final resolution of 2.60 Å. To resolve more of the density present in the triple helix bundle, three NU-refinements were performed on map A: C3 symmetry refinement, C3 symmetry with a mask excluding the triple helix bundle (to ensure the angular assignment of the particles is correct), and C3 symmetry with a mask excluding the 1871 Fab particles; this map (map B) reached a resolution of 2.74 Å. The crystal structure of 1871 Fab and OC43 S2 subunit model obtained from Alphafold-2 (59) was manually docked into the cryo-EM map using UCSF Chimera (60). Manual adjustment based on this model was done using COOT (54), followed by iterative rounds of refinement using COOT (54) and Phenix (53). Interface residues were identified using PDB PISA (61). Figures were made using UCSF Chimera (60) and Pymol (56). The structure has been deposited in the Protein Data Bank under the accession number PDB ID 9NQZ and the Electron Microscopy Data Bank under the accession number EMD-49708. AF3 modeling (33) was performed with the SARS-CoV-2 spike sequence (Uniprot P0DTC2) and 1871 Fab.
## Pseudovirus neutralization
SARS-CoV-2 pseudotyped viruses (PsV) were generated through transient transfection of 293T cells (ATCC) with lenti-viral backbone (BEI NR52516), structural and regula tory genes (BEI NR52518, NR52517, NR52519), and SARS-CoV-2 spike of interest (SARS-CoV-2/WIV04/2019 wildtype (BEI NR52516), Omicron/BA.1 (Scripps Research) (34,47) or Omicron/BA.5. Gene encoding Omicron BA.5 spike protein was synthesized and cloned by GeneArt (Thermo Fisher Scientific) in pcDNA3.4 backbone. Neutralization was assessed by single-cycle neutralization assays using 293-ACE2 cells (BEI NR52511), as previously described (34). Two biological replicates with two technical replicates were performed for each molecule.
## Antibody-dependent cellular phagocytosis (ADCP)
ADCP was measured using the FcγRIIa-H ADCP Bioassay kit (Promega), with CHO-K1/ Spike stable cell line (GenScript) as the target cells. An assay was performed according to the manufacturer's protocol. In brief, 6,000 target cells in 25 µL of assay buffer were plated into each well of a white, clear-bottom flat 96-well plate, and 25 µL of diluted antibody at 2X final concentration was added. After incubating target cells and antibody at 37°C for 15 min, 30,000 of freshly thawed FcγRIIa-H Effector Cells in 25 µL were added to each well for an E:T ratio of 5:1, and the plate was incubated for 6 h at 37°C. To read out the plate, 75 µL of prepared Bio-Glo Reagent was added to each well and incubated at room temperature for at least 5 min. Luminescence was measured on a plate reader (BioTEK). Data were normalized and then plotted in GraphPad Prism (GraphPad Software, Inc.).
## References
1. Su, Wong, Shi et al. (2016) "Epidemiology, genetic recombination, and pathogenesis of coronavi ruses" *Trends Microbiol*
2. Li, Shah, Wang et al. (2022) "Crossspecies transmission, evolution and zoonotic potential of coronaviruses" *Front Cell Infect Microbiol*
3. Frutos, Serra-Cobo, Pinault et al. (2021) "Emergence of bat-related betacoronaviruses: hazard and risks" *Front Microbiol*
4. Markov, Ghafari, Beer et al. (2023) "The evolution of SARS-CoV-2" *Nat Rev Microbiol*
5. Dougan, Nirula, Azizad et al. (2021) "Bamlanivimab plus etesevimab in mild or moderate covid-19" *N Engl J Med*
6. (2025) *Full-Length Text Journal of Virology*
7. Weinreich, Sivapalasingam, Norton et al. (2021) "REGEN-COV antibody combination and outcomes in outpatients with covid-19" *N Engl J Med*
8. Liu, Iketani, Guo et al. (2022) "Striking antibody evasion manifested by the omicron variant of SARS-CoV-2" *Nature*
9. Walls, Park, Tortorici et al. (2020) "Structure, function, and antigenicity of the SARS-CoV-2 spike glycopro tein" *Cell*
10. Silva, Huang, Nguyen et al. (2023) "Identification of a conserved S2 epitope present on spike proteins from all highly pathogenic coronaviruses" *Elife*
11. Zhou, Yuan, Song et al. (2022) "A human antibody reveals a conserved site on beta-coronavirus spike proteins and confers protection against SARS-CoV-2 infection" *Sci Transl Med*
12. Pinto, Sauer, Czudnochowski et al. (2021) "Broad betacoronavi rus neutralization by a stem helix-specific human antibody" *Science*
13. Zhou, Song, He et al. (2022) "Broadly neutralizing anti-S2 antibodies protect against all three human betacoronaviruses that cause severe disease" *bioRxiv*
14. Sun, Yi, Zhu et al. (2022) "Neutralization mechanism of a human antibody with pancoronavirus reactivity including SARS-CoV-2" *Nat Microbiol*
15. Li, Chen, Prévost et al. (2022) "Structural basis and mode of action for two broadly neutralizing antibodies against SARS-CoV-2 emerging variants of concern" *Cell Rep*
16. Dacon, Tucker, Peng et al. (2022) "Broadly neutralizing antibodies target the coronavirus fusion peptide" *Science*
17. Low, Jerak, Tortorici et al. (2022) "ACE2-binding exposes the SARS-CoV-2 fusion peptide to broadly neutralizing coronavirus antibodies" *Science*
18. Ng, Faulkner, Finsterbusch et al. (2022) "SARS-CoV-2 S2-targeted vaccination elicits broadly neutralizing antibodies" *Sci Transl Med*
19. Halfmann, Frey, Loeffler et al. (2022) "Multivalent S2based vaccines provide broad protection against SARS-CoV-2 variants of concern and pangolin coronaviruses" *EBioMedicine*
20. Hsieh, Leist, Miller et al. (2024) "Prefusion-stabilized SARS-CoV-2 S2-only antigen provides protection against SARS-CoV-2 challenge" *Nat Commun*
21. Keitany, Rubin, Garrett et al. (2023) "Multimodal, broadly neutralizing antibodies against SARS-CoV-2 identified by high-throughput native pairing of BCRs from bulk B cells" *Cell Chem Biol*
22. Amanat, Stadlbauer, Strohmeier et al. (2020) "A serological assay to detect SARS-CoV-2 seroconversion in humans" *Nat Med*
23. Cai, Zhang, Xiao et al. (2020) "Distinct conformational states of SARS-CoV-2 spike protein" *Science*
24. Landau, Mayrose, Rosenberg et al. (2005) "ConSurf 2005: the projection of evolutionary conservation scores of residues on protein structures" *Nucleic Acids Res*
25. Ashkenazy, Erez, Martz et al. (2010) "ConSurf 2010: calculating evolutionary conservation in sequence and structure of proteins and nucleic acids" *Nucleic Acids Res*
26. Haim, Abadi, Martz et al. (2016) "ConSurf 2016: an improved methodology to estimate and visualize evolutionary conservation in macromolecules" *Nucleic Acids Res*
27. Ben Chorin, Masrati, Kessel et al. (2020) "ConSurf-DB: an accessible repository for the evolutionary conservation patterns of the majority of PDB proteins" *Protein Sci*
28. Celniker, Nimrod, Ashkenazy et al. (2013) "ConSurf: using evolutionary data to raise testable hypotheses about protein function" *Isr J Chem*
29. Goldenberg, Erez, Nimrod et al. (2009) "The ConSurf-DB: precalculated evolutionary conservation profiles of protein structures" *Nucleic Acids Res*
30. Fan, Cao, Kong et al. (2020) "Cryo-EM analysis of the post-fusion structure of the SARS-CoV spike glycoprotein" *Nat Commun*
31. Shi, Cai, Zhu et al. (2023) "Cryo-EM structure of SARS-CoV-2 postfusion spike in membrane" *Nature*
32. Wang, Hesketh, Shamorkina et al. (2022) "Antigenic structure of the human coronavirus OC43 spike reveals exposed and occluded neutralizing epitopes" *Nat Commun*
33. Walls, Tortorici, Snijder et al. (2017) "Tectonic conformational changes of a coronavirus spike glycoprotein promote membrane fusion" *Proc Natl Acad Sci*
34. Abramson, Adler, Dunger et al. (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" *Nature*
35. Aschner, Muthuraman, Kucharska et al. (2023) "A multi-specific, multi-affinity antibody platform neutralizes sarbecoviruses and confers protection against SARS-CoV-2 in vivo" *Sci Transl Med*
36. Liu, Iketani, Guo et al. (2022) "An antibody class with a common CDRH3 motif broadly neutralizes sarbecoviruses" *Sci Transl Med*
37. Who (2025) "Prioritizing Diseases for Research and Development in Emergency Contexts"
38. Who (2024) "COVID-19 Deaths | WHO COVID-19 Dashboard"
39. Tse, Hou, Mcfadden et al. (2023) "A MERS-CoV antibody neutralizes a pre-emerging group 2c bat coronavirus" *Sci Transl Med*
40. Wu, Chiang, Lai et al. (2022) "Monoclonal antibody targeting the conserved region Full-Length Text Journal of Virology December"
41. "SARS-CoV-2 spike protein to overcome viral variants" *JCI Insight*
42. Johnson, Wall, Kramer et al. (2024) "Discovery and characterization of a pan-betacoronavirus S2-binding antibody" *Structure*
43. Izadi, Nordenfelt (2024) "Protective non-neutralizing SARS-CoV-2 monoclonal antibodies" *Trends Immunol*
44. Winkler, Gilchuk, Yu et al. (2021) "Human neutralizing antibodies against SARS-CoV-2 require intact Fc effector functions for optimal therapeutic protection" *Cell*
45. Ullah, Prévost, Ladinsky et al. (2021) "Live imaging of SARS-CoV-2 infection in mice reveals that neutralizing antibodies require Fc function for optimal efficacy" *Immunity*
46. Chan, Seah, Chye et al. (2021) "The Fc-mediated effector functions of a potent SARS-CoV-2 neutralizing antibody, SC31, isolated from an early convalescent COVID-19 patient, are essential for the optimal therapeutic efficacy of the antibody" *PLoS One*
47. Dussupt, Sankhala, Mendez-Rivera et al. (2021) "Lowdose in vivo protection and neutralization across SARS-CoV-2 variants by monoclonal antibody combinations" *Nat Immunol*
48. Inoue, Yamamoto, Sato et al. (2024) "Overcoming antibody-resistant SARS-CoV-2 variants with bispecific antibodies constructed using non-neutralizing antibodies"
49. Rujas, Kucharska, Tan et al. (2021) "Multivalency transforms SARS-CoV-2 antibodies into ultrapotent neutralizers" *Nat Commun*
50. Rujas, Cui, Burnie et al. (2022) "Engineering pan-HIV-1 neutralization potency through multispecific antibody avidity" *Proc Natl Acad Sci*
51. (2024) "Phaser Crystallographic Software"
52. (2014) "X-ray data preparation and reciprocal space exploration program"
53. Mccoy, Grosse-Kunstleve, Adams et al. (2007) "Phaser crystallographic software" *J Appl Crystallogr*
54. Leem, Dunbar, Georges et al. (2016) "ABodyBuilder: automated antibody structure prediction with data-driven accuracy estimation" *MAbs*
55. Adams, Afonine, Bunkóczi et al. (2010) "PHENIX: a comprehensive python-based system for macromo lecular structure solution" *Acta Crystallogr D Biol Crystallogr*
56. Emsley, Lohkamp, Scott et al. (2010) "Features and development of coot" *Acta Cryst D*
57. Morin, Eisenbraun, Key et al. (2013) "Collaboration gets the most out of software. eLife 2:e01456"
58. Llc (2021) "The PyMOL Molecular Graphics System"
59. Marr, Benlekbir, Rubinstein (2014) "Fabrication of carbon films with ∼ 500nm holes for cryo-EM with a direct detector device" *J Struct Biol*
60. Punjani, Rubinstein, Fleet et al. (2017) "cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination" *Nat Methods*
61. Jumper, Evans, Pritzel et al. (2021) "Highly accurate protein structure prediction with AlphaFold" *Nature*
62. Pettersen, Goddard, Huang et al. (2004) "UCSF chimera--a visualization system for exploratory research and analysis" *J Comput Chem*
63. Krissinel, Henrick (2007) "Inference of macromolecular assemblies from crystalline state" *J Mol Biol*
64. Oladunni, Park, Pino et al. (2020) "Lethality of SARS-CoV-2 infection in K18 human angiotensin-converting enzyme 2 transgenic mice" *Nat Commun* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12685548&blobtype=pdf | # Beyond spillover: Leveraging zoonotic disease research to advance biodiversity conservation
Marinda De Vries, Teresa Kearney, Ara Monadjem, Peter Taylor, Wanda Markotter
## 1. Introduction
Biodiversity and conservation are at the forefront of global wellbeing and should form an integral part of One Health collaborative research in efforts aimed at achieving optimal health for people, animals and ecosystems [1]. To address biodiversity conservation, international policy frameworks such as the United Nations Sustainable Development Goals (SDG 15) [2], the Convention on Biological Diversity (CBD) and the more recent Kunming-Montreal Global Biodiversity Framework (KMGBF), have been developed. While many local, national, and international initiatives continuously strive to define, monitor, and protect in-country biodiversity, these efforts are often constrained by financial limitations and/or insufficient technical expertise and resources.
A recent perspective on strengthening biodiversity conservation across the African continent presents the collective opinion of 27 conservationists and researchers in Africa [3]. The authors propose five key recommendations: (i) accelerating data collection, sharing and analytics; (ii) innovate education and capacity building; (iii) enhance and expand protected areas, ecological networks, and foundational legal frameworks; (iv) identifying and unlocking creative funding channels for conservation initiatives; and (v) integrate indigenous and local knowledge into conservation strategies. Without adequate investment, efforts to generate and share biodiversity data will be severely constrained or delayed.
Mammalian biodiversity, specifically, is not only important from a global biological perspective but also has links to disease ecology. Strict biosafety regulations in countries like South Africa govern the capturing and handling of certain species like bats due to their potential role as reservoir species for zoonotic diseases. As such, researchers in the field of conservation, biodiversity and taxonomy frequently lack all the necessary permits, legislative approvals, biocontainment facilities, biosafety and biosecurity expertise or institutional capacity to work with these mammals. This poses yet another constraint on the progress towards defining in-country mammalian biodiversity.
The emergence and re-emergence of zoonotic diseases from wildlife such as Marburg, Ebola and Nipah viruses have been a main driver of increased efforts towards wildlife disease surveillance [4]. The more recent emergence of SARS-CoV-2 has further driven global attention towards prediction, prevention and response of future pandemics [5]. Resources have been directed towards identifying wildlife reservoirs, understanding infection dynamics and ecological contexts that facilitate spillover. As such, there has been a marked increase in field sampling, pathogen detection, and identification of wildlife species through molecular detection of pathogensparticularly of bats and other small mammals like rodents. While these surveillance studies strengthen global health preparedness, a secondary benefit with cross-disciplinary relevance is the collection of valuable biological material and data that could contribute towards species collections, inventories, and monitoringultimately supporting biodiversity conservation. However, despite the natural overlap between disease research and biodiversity conservation, a clear disconnect remains -likely stemming from the historical mistrust shaped by narratives framing wildlife as disease threats rather than essential ecosystem contributors [6]. This gap can be addressed through open dialogue, strategic alignment of collaborations, and more effective resource utilisation.
## 2. Zoonotic disease research -An underutilised source of biodiversity data
Field-based pathogen surveillance research typically involves the trapping, handling, and sampling of wildlife species -often in remote, understudied regions, or among species that are otherwise difficult to access. These activities generate valuable foundational resources, including voucher specimens, tissue samples and genetic barcodes that could be of considerable value to taxonomists, conservationists, and other biologists. Yet, the conservation value of these resources is seldom recognised or fully leveraged.
Tissue samples and associated metadata (e.g., spatiotemporal and morphological information) collected during pathogen surveillance are often archived or discarded once pathogen testing is complete. While host identification using molecular DNA barcoding is sometimes conducted, it is typically limited to individuals that test positive for a target pathogen. These data are published along with the pathogen findings in scientific outputs, but due to a lack of collaboration with national museums and governmental strategic frameworks, these data get overlooked and are not routinely integrated. Furthermore, voucher specimens collected as part of biosurveillance studies are rarely prepared and deposited in national museum collections [7,8], which prevents the linkage of DNA barcodes to morphological specimensespecially for undocumented species. Consequently, host species barcodes are submitted to public databases like the National Center for Biotechnology Information's (NCBI) Genbank with limited metadata and without physical specimens (museum vouchers) for morphological species confirmation of DNA barcodes. Physical specimens are key in updating taxonomic classification, where reference vouchers need to be revisited [9]. This is particularly relevant with DNA sequencing, resolving more species than previously identified by morphology alone. Specimens also provide a reference for conservation assessments and further research.
While the primary value of surveillance studies lies in strengthening preparedness for zoonotic disease emergence [5], these efforts could be more fully leveraged to support biodiversity research and conservation. In doing so, it would enhance the optimal use of collected materials and generate broader ecological insights to reinforce the One Health framework. Accurate species identification is an important component of disease surveillance, especially with the emergence of zoonotic pathogens from wildlife. It would thus be in the best interest of disease researchers to ensure accuracy of their reports and findings with regard to host identification, especially in instances where a new host is identified. This will have implications for pathogen monitoring, prevention, and intervention strategies. A collaborative approach between disease researchers, conservationists, taxonomists, other biologists and museums will ensure data accuracy and accelerate data collection (for genetic DNA databases and distribution records to inform the conservation status of species) by leveraging the funding, expertise and access provided by disease research to improve our taxonomic and systematic understanding of the mammalian biodiversity and local and regional species inventories.
## 3. A Southern African perspective
Recognizing and acknowledging shared challenges and mutual benefits across both biodiversity conservation and disease research have been major drivers in our adoption of a more integrative and collaborative research approach. Regulatory restrictions have been among the key drivers of convergence for our research teams. The distinction between high-and low-risk wildlife research remains a subject of ongoing debate, and both are currently governed under the same biosafety and permitting frameworks, lacking a more practical differentiationparticularly in South Africa [10]. This has historically created a significant bottleneck for biodiversity science. Activities such as specimen collection or tissue transfer for purely taxonomic purposes are subject to the same regulatory authorization as high-risk infectious disease research and that has hampered collection and sharing of samples for disease surveillance when only non-disease researchers are in the field. In parallel, the time-consuming process of permit acquisition, biosafety clearance and ethical approvals significantly slows research progress, particularly in academic settings where post-graduate projects are timebound. Collaboration has enabled us to address these challenges far more effectively than would have been possible through induvial or isolated efforts. Our collaboration is an example of the integration described above that has yielded bidirectional benefits for both disease research and mammalian biodiversity documentation. Over the past 15 years, the viral zoonoses research team has actively sought to undertake fieldwork together with small mammal researchers and to deposit voucher specimens in museums and biodiversity repositories across Southern Africa. This coordinated sample and data collection has optimised the use of human resources, funding sources, and biological material while minimising unnecessary duplication. The systematic collection of metadata and deposition of voucher specimens for a wide range of bat and rodent species into museum collections has enabled morphological identification of host species linked to viral pathogens [11,12].
Tissue samples of historical voucher specimens available in our sample biobanks were subjected to genetic barcoding to establish a local reference library of DNA barcodes for these vouchered specimensspecifically for bat species in South Africa and beyond. We aimed to generate at least one barcode from each of the three mitochondrial regions (Cytochrome c Oxidase Subunit I, 12S ribosomal RNA and Cytochrome b) for >60 % of South African bat species linked to voucher specimens from tissues available in our biobank (de Vries et al. in preparation). These data are being deposited in the Barcode of Life Data Systems (BOLD) in collaboration with the South African National Biodiversity Institute (SANBI), in alignment with South Africa's National Biodiversity Strategy and Action Plan. This barcode library will allow rapid identification of bat species without the need for unnecessary destructive sampling, especially for cryptic species that are difficult to distinguish morphologically in the field. Beyond improving species identification, this integrative approach has additionally contributed to the formal description of at least three novel bat species from South Africa (Kearney et al. in preparation), Angola and Eswatini [13], and generated valuable biodiversity data.
Through open and transparent dialogue at individual level -with regards to project goals, resource allocation, data requirement needs, and expected outputs, we were able to establish a mutually beneficially relationship that has strengthened institutional collaborations towards a shared vision. By leveraging the complementary strengths of each discipline, our team has developed a wholistic One Health approach to our research and adopted it as standard practice. Sustaining this collaboration has required a continuous dialogue beyond the project's initial phases which has been essential to building trust and aligning research prioritiesultimately allowing the team to remain actively involved as the research directions evolve.
This collaborative effort reinforces the interdependence of disease research and biodiversity conservation, recognizing the overlapping goals and logistical needs. An additional benefit has been the reciprocal transfer of skills and knowledge -training of biologists in biosafety principles and conversely, equipping disease researchers with fieldbased and taxonomic expertise. Our integrative strategy clearly highlights the broader taxonomic and conservation value of disease-focused research.
## 4. Reframing disease research
To optimally leverage wildlife disease research in advancing global biodiversity conservation, more deliberate and structured approaches will be needed. We propose the following as part of a framework for integration: i) Establishing formal institutional partnerships between surveillance teams, conservation bodies and museums. ii) Improved One Health education to address misconceptions and bridge the distrust between disease experts and conservationists. iii) Inclusion of a structured biodiversity component in disease research from the project planning stage. iv) Mandate or incentivise biodiversity contributions within research grants that support disease surveillance, ensuring the integration of biodiversity objectives as a minimum requirement. v) Investment in capacity building of non-disease research staff, enabling them to aid in data collection under biocontainment conditions. vi) Development of clear regulatory guidelines, supported by national permitting agencies, that distinguishes between high-risk work requiring biosafety certification and low-risk biodiversity sampling to enable safe and efficient collection and transfer of material such as voucher specimens between biosafety-certified institutions and biodiversity repositories. vii) Streamline international transfer of specimens and genetic material through coordinated frameworks to reduce administrative bottlenecks and strengthen local scientific capacity.
## 5. Conclusion
The global increase in zoonotic disease surveillance presents a unique opportunity to leverage alternative resources to secure foundation data to support biodiversity conservation and offers a pathway to access and document wildlife diversity. However, without a more deliberate effort to integrate biodiversity goals, this opportunity will remain largely underutilised. This paper argues that zoonotic disease research can, and should, do more to meaningfully contribute towards biodiversity conservation, particularly in countries where access to live wildlife is constrained by biodiversity and biosafety legislation or where underinvestment in natural history has hindered scientific progress. Our collaboration between research institutions and museums in South Africa demonstrates how disease research can actively contribute to mammalian biodiversity knowledge and may serve as an encouraging precedent for future integrative research efforts. Most importantly, a unified voice from disease experts and taxonomists and conservationists is critical to address the negative public paranoia about bats and wildlife as disease vectors.
## Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work the author(s) used ChatGPT (OpenAI, August 2025 version) in order to improve language clarity and structure. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article.
## References
1. Ohhlep), Adisasmito, Almuhairi et al. "One Health: a new definition for a sustainable and healthy future"
2. (2015) "Transforming Our World: The 2030 Agenda for Sustainable Development"
3. Bezeng, Ameka, Angui et al. (1917) "An African perspective to biodiversity conservation in the twenty-first century"
4. Letko, Seifert, Olival et al. (2020) "Bat-borne virus diversity, spillover and emergence" *Nat. Rev. Microbiol*
5. Ohhlep), Markotter, Mettenleiter et al. "Prevention of zoonotic spillover: from relying on response to reducing the risk at source" *PLoS Pathog*
6. Olival (2016) "To cull, or not to cull, bat is the question" *EcoHealth*
7. Tsang, Low, Wiantoro et al. (2021) "Detection of Tioman virus in Pteropus vampyrus near Flores"
8. Geldenhuys, Mortlock, Weyer et al. (2018) "A metagenomic viral discovery approach identifies potential zoonotic and novel mammalian viruses in Neoromicia bats within South Africa" *PloS One*
9. Monadjem, Demos, Dalton et al. (2021) "A revision of pipistrelle-like bats (Mammalia: Chiroptera: Vespertilionidae) in East Africa with the description of new genera and species" *Zool. J. Linn. Soc*
10. Alexander, Tolley, Maritz et al. (2021) "Excessive red tape is strangling biodiversity research in South Africa"
11. Dietrich, Tjale, Weyer et al. (2016) "Diversity of Bartonella and Rickettsia spp. in bats and their blood-feeding ectoparasites from South Africa and Swaziland" *PloS One*
12. Mortlock, Geldenhuys, Keith et al. "Paramyxo-and coronavirus diversity and host associations in non-volant small mammals: evidence of viral sharing" *Virus Evol*
13. Taylor, Strydom, Richards et al. (2022) "Integrative taxonomic analysis of new collections from the central Angolan highlands resolves the taxonomy of African pipistrelloid bats on a continental scale" *Zool. J. Linn. Soc* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12584610&blobtype=pdf | # Earlier and prolonged respiratory syncytial virus (RSV) seasons in young children compared to adults: implications for prevention in infants
Yuan Chao Xue, Natalie Williams-Bouyer, Ping Ren, Janak Patel, Sophia Georghiou
## Abstract
Respiratory syncytial virus (RSV) poses a major health concern, particularly for young children and older adults. In this 10-year, single-center retrospective study (15 May 2015 to 14 May 2025), we analyzed RSV positivity rates to characterize local epidemiologic trends. When stratified by age, children ≤2 years consistently exhibited the highest positivity rates. Using the Centers for Disease Control and Prevention's 3% positivity threshold, we found that the RSV season in the general population in our region began in early July and ended in late December, approximately 3 months earlier than the national average. Among infants ≤2 years, the season lasted 5 weeks longer on average than in the general population. These findings highlight regional variability in RSV seasonality and the earlier onset and extended duration of RSV activity in young children. This age group appeared particularly susceptible and likely played a key role in initiating and sustaining community transmission. Monitoring RSV positivity specifically in children ≤2 years, separate from the general population, may enhance local surveil lance accuracy and support more timely decisions regarding RSV immunoprophylaxis and vaccination strategies. IMPORTANCE Respiratory syncytial virus (RSV) is a major health concern, especially for young children and older adults. In this study, we explored whether different age groups changed the timing of RSV season. This is important because public health guidelines, hospital preparedness, and preventive strategies like antibody prophylaxis and vaccination rely on accurate RSV season timing. By showing how RSV trends differ by age, our findings can help improve seasonality responses and ensure that preventive measures reach high-risk groups, especially infants, at the right time.
R espiratory syncytial virus (RSV) is a leading cause of lower respiratory tract infections, posing a significant public health threat, particularly to young children and older adults. Among children under 5 years of age, RSV is a primary cause of bronchiolitis and pneumonia, often necessitating hospitalization (1). In older adults and individuals with underlying conditions, RSV infections contribute to substantial morbidity and mortality, further emphasizing its clinical significance (2). Understanding the dynamics of RSV transmission and seasonality is critical to mitigating its healthcare burden, specifically in these high-risk populations (1).
In the United States, the RSV season typically spans from October to April, with regional and local variations (3). The Centers for Disease Control and Prevention (CDC) defines the start and end of the RSV season as the first and last occurrence of two consecutive weeks with positivity rates of ≥3% based on molecular diagnostics among all tested individuals (3). However, this threshold may miss important age-related trends, particularly the disproportionate burden in younger children seen in both outpatient and inpatient settings (4). Monitoring RSV patterns separately in high-risk pediatric populations is crucial to uncover more nuanced patterns and improve the timing of prevention strategies, including passive immunization in infants and vaccination of pregnant women and older adults.
This 10-year, single-center retrospective study aimed to characterize local RSV testing and positivity rates, stratified by age, to better understand RSV seasonality in a regional context.
We extracted de-identified data from the University of Texas Medical Branch (UTMB) Epic electronic medical record system for patients receiving care at UTMB clinics, emergency rooms, urgent care centers, and hospitals. The data set included results, testing dates, patient gender, and age at the time of testing.
Data cleaning (excluding entries missing dates or age) and analysis were performed in RStudio (Version 2024.12.1 + 467), resulting in 92,921 eligible entries. Data consistency and integrity were also reviewed.
For analysis and visualization, we used the following R packages: ggplot2 (Version 3.5.1) for data visualization, dplyr (Version 1.1.4) and tidyverse (Version 2.0.0) for data manipulation, tidyr (Version 1.3.1) for data tidying, lubridate (Version 1.9.4) for date handling, and zoo (Version 1.8-12) for time-series analysis. To evaluate seasonal patterns, t-tests were used for comparison of group means, and Z-tests with Bonferroni correction were used for time-series analysis. This study was approved by the UTMB Institutional Review Board (#24-0390).
From 15 May 2015 to 14 May 2025, the overall number of RSV tests at UTMB steadily increased, but the distribution of testing across age groups did not change significantly (P >0.05), and weekly positivity rates remained relatively stable across each annual cycle (data not shown). To explore age-specific trends, data were stratified into three age groups: ≤2 years, 3-18 years, and ≥19 years.
Adults (≥19 years) accounted for 48.9% of all RSV tests during the study period, totaling 45,468 tests. In comparison, the pediatric population underwent a combined number of 47,453 tests. This included 33,106 tests (35.7%) for children ≤2 years and 14,347 tests (15.4%) for those aged 3-18 years. Despite lower testing volumes in younger age groups, children consistently exhibited higher positivity rates than adults (Fig. 1). Notably, children ≤2 years had significantly higher positivity rates than all other age groups and the overall population (Z-test: P <0.001).
According to the CDC definition, an RSV season spans from the first to the last occurrence of two consecutive weeks with positivity rates of ≥3% using molecular testing methods. In our region, RSV seasons typically began in early July and ended in late December, with the exception of atypical patterns observed from 2021 to 2023 due to the COVID-19 pandemic. During these years, a major summer outbreak occurred in 2021, followed by a distinct two-peak season in 2022-2023. Then, the 2023-2024 season reverted to the traditional fall-to-late-winter/early-spring pattern. Importantly, children ≤2 years consistently showed an earlier season onset by 1 week (28 June vs 8 July; t-test: P = 0.043) and a later end date by 4 weeks (22 January vs 26 December; t-test: P = 0.0019) compared to the overall population (Table 1). As a result, their RSV season was significantly longer by 5.1 weeks (33.5 weeks vs 28.4 weeks; t-test: P = 0.024), excluding the atypical 2021-2023 seasons (Table 1). To assess the resilience of RSV seasonality in children ≤2 years, we compared the impact of applying a higher positivity threshold of 5%. Relative to the 3% threshold, the 5% threshold shortened the RSV season by 4.3 weeks in children ≤2 years but by 8.7 weeks in the overall population (Table 1).
In this study, we confirmed regional variability in RSV seasonality, consistent with data from the CDC's National Respiratory and Enteric Virus Surveillance System (https:// www.cdc.gov/nrevss/php/dashboard/index.html). Notably, the RSV season in our region of southern Texas appeared to be approximately 2 weeks longer than that reported for Region 6, which encompassed 16-21 laboratories' data from New Mexico, Texas, Oklahoma, Louisiana, and Arkansas (3). Age-stratified analysis revealed that adults accounted for nearly half of the total RSV tests; however, the highest positivity rates were consistently observed in the pediatric population, particularly among children ≤2 years of age. These findings are aligned with those reported by Tran et al., who also showed high positivity rates in children under 4 years (5). Furthermore, we observed that children ≤2 years consistently experienced an earlier onset and longer duration of RSV seasons compared to the general population, suggesting that they may be more susceptible to RSV infection and at greater risk for severe disease than older children and adults. When the positivity threshold increased from 3% to 5%, the RSV season had less reduction in children ≤2 years than in the general population. This highlights the resiliency of RSV seasonality in young children, even when stricter criteria were applied. In conclusion, our observations indicate that RSV surveillance in children ≤2 years should guide the timing of immunoprophylaxis administration in infants or maternal RSV vaccination, rather than relying solely on data from the general population. However, this study's single-center design may limit the generalizability of the findings, as they reflect a specific regional experience.
## 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. Falsey, Hennessey, Formica et al. (2005) "Respiratory syncytial virus infection in elderly and high-risk adults" *N Engl J Med*
3. Hamid, Winn, Parikh et al. (2023) "Seasonality of sespiratory syncytial virus -United States, 2017-2023" *MMWR Morb Mortal Wkly Rep*
4. Hall, Weinberg, Iwane et al. (2009) "The burden of respiratory syncytial virus infection in young children" *N Engl J Med*
5. Tran, Nduaguba, Diaby et al. (2022) "RSV testing practice and positivity by patient demographics in the United States: integrated analyses of MarketScan and NREVSS databases" *BMC Infect Dis* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12239721&blobtype=pdf | # Mismatch-corrected CHIKV CDC Trioplex assay oligos restore sensitive pangenotype viral detection
Aminata Lika, Diouf, Mignane Ndiaye, Diamilatou Balde, Agathe Efire, Moussa Dia, Fatou Thiam, Manfred Weidmann, Oumar Faye, Idrissa Dieng
## Abstract
C hikungunya virus (CHIKV) is an arbovirus primarily transmitted by Aedes mosquitoes, predominantly in tropical and subtropical regions (1, 2). Due to global warming, the geographic spread of arboviruses, including CHIKV, is expanding, increasing their public health impact (3).Genetically, CHIKV exists in three major genotypes: Asian, East-Central-South African (ECSA), and West African (WA) (1).Although CHIKV originated in Africa, knowledge about its burden on the continent remains limited (2,4). This is linked to low awareness, lack of effective surveillance programs, and reliable diagnostic tools (2).Recently, Senegal experienced a significant CHIKV epidemic caused by the West African genotype, with up to 300 confirmed cases in late 2024 (5,6). Additionally, a cluster of 12 imported cases belonging to this genotype was reported in France from Côte d'Ivoire in 2023 and 2024, respectively (4). These detections highlight an intensive and cryptic circulation of the virus in West Africa (4, 7).To effectively monitor CHIKV spread in Africa and mitigate the risk of importation, reliable diagnostic tools are essential (4). For years, RT-qPCR has been the primary method for CHIKV detection (8). However, co-circulation of arboviruses such as Zika (ZIKV), Dengue (DENV), and CHIKV, which also share similar clinical pictures, poses diagnostic challenges (9). In response to this, the CDC Trioplex assay was developed for the simultaneous detection and differentiation of these viruses during the Zika outbreak in Latin America (9). Despite its broad utility, the Trioplex assay was reported to exhibit reduced sensitivity for the WA genotype, limiting its effectiveness in West Africa (6). To address this limitation, we developed AltoDesign, a novel RT-qPCR assay with optimized primers and probes designed to restore pangenotypic CHIKV detection (Table S1). This new oligonucleotide set was designed based on CHIKV sequences from GenBank, with a focus on recent Senegalese strains belonging to the WA genotype.The AltoDesign assay's ability to detect various CHIKV genotypes was evaluated in comparison to the in-house RT-qPCR assay and the CDC Trioplex assays using a panel of CHIKV strains available at the WHO Collaborating Center for Arboviruses at Institut Pasteur de Dakar (IPD) (Table S2). The performance of the AltoDesign, including the limit of detection (LOD), was assessed using a WA CHIKV strain (SH274640) with known viral titers in four technical replicates for probit analysis at 95% confidence.Obtained standard curves from a 6-log serial dilution series showed that the assays exhibit amplification efficiencies of 110.17% and 92.35% for the CHIKV Trioplex and the AltoDesign, respectively, and that both assays show R 2 values greater than 0.99.The AltoDesign assays yield consistently lower Cq values (6.715 ± 0.88) than the CHIKV Trioplex assay, indicating increased efficiency and suggesting that mismatches
described by Ndiaye and collaborators led to at least a 100-fold underestimation of the WA genotype template amount per reaction. Indeed, the limit of detection of the AltoDesign assay determined by using the PFU dilution series is about 5 Cq lower than the Trioplex (Fig. 1A), which roughly equates to an approximately 1-1.5 log₁₀ increased analytical sensitivity. In terms of PFU/reaction (log₁₀ 2), the AltoDesign assay detected 321 PFU/reaction (log₁₀ 2) with a mean Cq value of 34.97 ± 0.96, whereas the Trioplex assay had a detection limit of 32,100 PFU/reaction with a Cq value of 34.65 ± 0.36 (Table S3; Fig. 1A), which corresponds to an increased sensitivity of 2 log₁₀ steps. At 95% confidence, AltoDesign achieved a lower LOD (238.88 PFU/reaction) compared with the Trioplex assay (21858.02 PFU/reaction) (Fig. 1B). When testing CHIKV strain supernatants (Table S2), the AltoDesign for CHIKV WA genotype thus improves the lag Cq values previously described by Ndiaye and colleagues in comparison to the in-house assay and the Trioplex assay (Fig. 1C).
For RT-qPCR, previous work highlighted that mismatches on oligonucleotide binding sites can impair assay performances (10). Mismatches on target regions can lead to target failure, causing false negatives (11) or a drastic reduction in assay performances (6). In our study, we restored the sensitivity of the CHIKV CDC trioplex assay to efficiently detect the CHIKV WA genotype. The need for mismatch-corrected oligonucleotides to improve assay performance for emerging viral strains is a critical consideration in molecular S1) diagnostics, particularly for pathogens like MPOX and influenza viruses, which exhibit high genetic variability (12,13).
Additionally, the AltoDesign primers showed similar performance to both the IPD in-house assay (6) and the CDC Trioplex assay, with comparable Cq values observed across the tested samples of ECSA and Asian genotype strains (Fig. 1C; Table S2).
Our results show that the newly designed mismatches corrected oligonucleotides provide a more sensitive, accurate, and pangenotype detection of all available CHIKV genotypes.
The AltoDesign assay significantly improves CHIKV detection with a higher 95% limit of detection, particularly for the West African genotype, overcoming the limitations of the native CHIKV CDC Trioplex assay. Additionally, the updated oligos can still be combined with DENV and ZIKV oligos for differential viral identification while allowing pangenotype CHIKV detection with similar Cq values (Fig. S1). This optimized RT-qPCR system offers a more reliable diagnostic tool for enhanced surveillance and outbreak response in regions where CHIKV and/or flaviviruses sharing the same clinical presenta tion are endemic.
## References
1. Ramphal, Tegally, San et al. (2024) "Understand ing the transmission dynamics of the chikungunya virus in Africa" *Pathogens*
2. Russo, Subissi, Rezza (2020) "Chikungunya fever in Africa: a systematic review" *Pathog Glob Health*
3. Delrieu, Martinet, Connor et al. (2023) "Temperature and transmission of chikungunya, dengue, and Zika viruses: a systematic review of experimental studies on Aedes aegypti and Aedes albopictus" *Curr Res Parasitol Vector-Borne Dis*
4. Pezzi, Modenesi, Ayhan et al. (2025) "Cryptic circulation of chikungunya virus in Côte d'Ivoire revealed by sentinel travellers, 2023-2024" *J Travel Med*
5. Dieng, Sadio, Sagne et al. (2024) "Genomic characterization of a reemerging chikungunya outbreak in Kedougou" *Southeastern Senegal, New-Data Letter Journal of Clinical Microbiology*
6. (2023) *Emerg Microbes Infect*
7. Ndiaye, Kane, Balde et al. (2024) "CDC Trioplex diagnostic assay underperforms in detection of circulating chikungunya West African genotype" *J Clin Microbiol*
8. Jallow, Dieng, Sanneh et al. (2024) "Detection of chikungunya virus in the Gambia through a newly implemented sentinel surveillance program"
9. Thirion, Pezzi, Corcostegui et al. (2019) "Development and evaluation of a duo chikungunya virus real-time RT-PCR assay targeting two regions within the genome" *Viruses*
10. Santiago, Vergne, Quiles et al. (2013) "Analytical and clinical performance of the CDC real time RT-PCR assay for detection and typing of dengue virus" *PLoS Negl Trop Dis*
11. Stadhouders, Pas, Anber et al. (2010) "The effect of primer-template mismatches on the detection and quantification of nucleic acids using the 5' nuclease assay" *J Mol Diagn*
12. Ledeker, Long (2013) "The effect of multiple primer-template mismatches on quantitative PCR accuracy and development of a multiprimer set assay for accurate quantification of pcrA gene sequence variants" *J Microbiol Methods*
13. Wu, Oghuan, Gitter et al. (2023) "Wide mismatches in the sequences of primers and probes for monkeypox virus diagnostic assays" *J Med Virol*
14. Klungthong, Chinnawirotpisan, Hussem et al. (2010) "The impact of primer and probe-template mismatches on the sensitivity of pandemic influenza A/H1N1/2009 virus detection by real-time RT-PCR" *J Clin Virol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12565690&blobtype=pdf | # Advances in Fungal Infection Research: From Novel Diagnostics to Innovative Therapeutics
Célia Rodrigues, Lucia Černáková
Invasive and superficial fungal infections continue to impose a significant global health burden, with rising morbidity and mortality rates particularly affecting immunocompromised populations [1,2]. The emergence of antifungal resistance (AFR), coupled with the rise in the prevalence of opportunistic fungal pathogens, underscores the urgent need for innovative diagnostic approaches and alternative therapeutic strategies [3,4]. Indeed, fungal infections represent one of the most complex challenges in modern medicine, affecting millions of patients worldwide [5,6]. The clinical spectrum ranges from superficial mucocutaneous infections to life-threatening invasive diseases, with opportunistic pathogens such as Candida, Aspergillus, and Pneumocystis species leading the epidemiological burden [7,8], particularly in immunocompromised patients, including those undergoing chemotherapy, organ transplantation, or living with HIV/AIDS, who comprise an expanding population at risk for severe mycoses [9,10].
The diagnostic landscape for fungal infections has undergone significant transformation in recent decades [11,12]. Traditional culture-based methods, while remaining the gold standard, are often time-consuming and may lack sensitivity for certain pathogens [13]. The development of molecular diagnostic techniques, including real-time PCR assays and next-generation sequencing, has revolutionized pathogen detection and identification, including in fungal species [14,15]. These methods enable rapid, accurate diagnosis that can be decisive for patient outcomes, especially in critically ill populations where an early intervention is crucial [16,17].
While the antifungal therapeutic arsenal has expanded considerably, challenges persist [18,19]. The limited number of antifungal drug classes, combined with the emergence of AFR, poses major obstacles to effective treatments [20,21]. Azole resistance in Aspergillus fumigatus, echinocandin resistance in Candida species, and the global spread of multidrugresistant Candida auris exemplify the evolving threat landscape [22,23]. Novel therapeutic approaches are being explored across multiple fronts [24,25]. Structure-based drug design has yielded promising compounds targeting specific fungal pathways, while drug repurposing strategies have identified unexpected antifungal properties in existing medications [26,27]. Additionally, natural products and antimicrobial peptides represent other opportunities for therapeutic development, offering potential alternatives to conventional drugs [28,29].
The importance of systematic surveillance in healthcare settings cannot be overstated [30,31]. Effective monitoring programs enable early detection of outbreaks, facilitate antimicrobial stewardship, and guide infection prevention strategies [32]. The implementation of screening protocols for high-risk patients has proven valuable in predicting disease progression and optimizing therapeutic interventions [33,34]. Indeed, healthcare-associated fungal infections require comprehensive infection control measures [35]. The emergence of environmental pathogens and the challenges posed by biofilm formation (on medical devices) require multidimensional prevention policies [36,37]. Understanding the epidemiology of healthcare-associated mycoses is imperative for developing targeted interventions and reducing transmission risks [38].
Recent advances in understanding fungal pathogenesis have revealed the complexity of host-pathogen interactions [39,40]. The role of host immunity, particularly in immunocompromised patients, determines disease susceptibility and progression [41,42]. Fungal virulence factors, including morphological transitions, biofilm formation, and immune evasion mechanisms, contribute to pathogenic success [43,44]. In addition, the concept of polymicrobial infections has gained attention, with evidence suggesting that inter-microbial interactions can significantly influence disease outcomes [45,46]. These interactions may involve competitive or synergistic relationships between different fungal species or between fungi and bacteria, potentially affecting treatment strategies and patient prognosis [47].
Epidemiological studies continue to identify key risk factors for fungal infections across diverse populations [48,49]. Diabetes mellitus, advanced age, immunosuppressive therapy, and prolonged hospitalization represent well-established predisposing factors [50,51]. Understanding these risk profiles enables targeted screening and prevention strategies for high-risk populations [52]. Geographic variation in fungal disease burden reflects environmental factors, endemic species distribution, and healthcare infrastructure differences [53,54]. Regional studies offer basic insights for public health planning and resource allocation, particularly in areas with limited diagnostic capabilities or restricted access to antifungal therapy [55,56].
The integration of artificial intelligence and machine learning into diagnostic platforms promises to enhance disease detection and prediction capabilities [57,58]. Advanced imaging techniques, biomarker discovery, and point-of-care testing exemplify emerging frontiers in fungal diagnostics [59,60]. Precision medicine approaches, incorporating host genetic factors, pathogen characteristics, and environmental variables, may enable individualized treatment strategies [61,62]. The development of immunotherapeutic approaches, including vaccine strategies and immune modulators, offers additional therapeutic possibilities [63,64].
This Special Issue encompasses diverse research contributions that advance our understanding of fungal infections from diagnostic innovation and clinical assessment, surveillance and risk stratification, therapeutic innovations and drug development, clinical insights and disease mechanisms, epidemiological perspectives and risk factors, through to future directions and clinical implications of AMR and mycology. The multidisciplinary nature of this research, spanning microbiology, immunology, pharmacology, and epidemiology, reflects the complexity of medical mycology as a field.
The collaborative efforts of researchers, clinicians, and public health professionals represented in this collection underscore the global commitment to addressing fungal disease challenges. As we continue to face evolving threats from emerging pathogens and drug resistance, such collaborative research initiatives remain essential for improving patient outcomes and advancing the field of medical mycology. We extend our sincere gratitude to all contributing authors, reviewers, and editorial staff who made this Special
## References
1. Brown, Denning, Gow et al. (2012) "Hidden killers: Human fungal infections" *Sci. Transl. Med*
2. Bongomin, Gago, Oladele et al. (2017) "Global and multi-national prevalence of fungal diseases-Estimate precision" *J. Fungi*
3. Fisher, Hawkins, Sanglard et al. (2018) "Worldwide emergence of resistance to antifungal drugs challenges human health and food security" *Science*
4. Denning, Bromley (2015) "Infectious Disease. How to bolster the antifungal pipeline" *Science*
5. Rodrigues "The multifunctional fungal ergosterol"
6. Casadevall, Coelho, Cordero et al. (2019) "The capsule of Cryptococcus neoformans" *Virulence*
7. Kullberg, Arendrup (2015) "Invasive candidiasis" *N. Engl. J. Med*
8. Patterson, Thompson, Denning et al. (2016) "Practice guidelines for the diagnosis and management of aspergillosis: 2016 update by the Infectious Diseases Society of America" *Clin. Infect. Dis*
9. Pappas, Lionakis, Arendrup et al. (2018) "Invasive candidiasis" *Nat. Rev. Dis. Primers*
10. Singh, Perfect (2007) "Immune reconstitution syndrome and exacerbation of infections after pregnancy" *Clin. Infect. Dis*
11. Lass-Flörl (2009) "The changing face of epidemiology of invasive fungal disease in Europe" *Mycoses*
12. White, Wingard, Bretagne et al. (2015) "Aspergillus polymerase chain reaction: Systematic review of evidence for clinical use in comparison with antigen testing" *Clin. Infect. Dis*
13. Lionakis, Kontoyiannis (2003) "Glucocorticoids and invasive fungal infections" *Lancet*
14. Mcmullan, Halliday, Sorrell et al. (2016) "Clinical utility of the FilmArray meningitis/encephalitis panel for the diagnosis of suspected central nervous system infections" *Diagn. Microbiol. Infect. Dis*
15. Arvanitis, Anagnostou, Fuchs et al. (2014) "Molecular and nonmolecular diagnostic methods for invasive fungal infections" *Clin. Microbiol. Rev*
16. Cornely, Arikan-Akdagli, Dannaoui et al. (2013) "ESCMID and ECMM joint clinical guidelines for the diagnosis and management of mucormycosis" *Clin. Microbiol. Infect*
17. Morrell, Fraser, Kollef (2005) "Delaying the empirical treatment of candida bloodstream infection until positive blood culture results are obtained: A potential risk factor for hospital mortality" *Antimicrob. Agents Chemother*
18. Perlin (2015) "Mechanisms of echinocandin antifungal drug resistance" *Ann. N. Y. Acad. Sci*
19. Sanglard (2016) "Emerging threats in antifungal-resistant fungal pathogens" *Front. Med*
20. Cowen, Sanglard, Howard et al. (2015) "Mechanisms of antifungal drug resistance. Cold Spring Harb" *Perspect. Med*
21. Pfaller, Diekema, Turnidge et al. (2019) "Twenty years of the SENTRY antifungal surveillance program: Results for Candida species from 1997-2016" *Open Forum Infect. Dis*
22. Lockhart, Etienne, Vallabhaneni et al. (2017) "Simultaneous emergence of multidrug-resistant Candida auris on 3 continents confirmed by whole-genome sequencing and epidemiological analyses" *Clin. Infect. Dis*
23. Van Der Linden, Arendrup, Warris et al. (2015) "Prospective multicenter international surveillance of azole resistance in Aspergillus fumigatus" *Emerg. Infect. Dis*
24. Roemer, Krysan (2014) "Antifungal drug development: Challenges, unmet clinical needs, and new approaches. Cold Spring Harb" *Perspect. Med*
25. Perfect (2017) "The antifungal pipeline: A reality check" *Nat. Rev. Drug Discov*
26. Zhai, Wu, Wang et al. (2012) "The antidepressant sertraline provides a promising therapeutic option for neurotropic cryptococcal infections" *Antimicrob. Agents Chemother*
27. Butts, Koselny, Chabrier-Roselló et al. (2014) "Estrogen receptor antagonists are anti-cryptococcal agents that directly bind EF hand proteins and synergize with fluconazole in vivo"
28. Villar-Vidal, Coronado-Aceves, Shibayama et al. (2021) "Chitosan and its derivatives: Properties and applications in biotechnology" *Polymers*
29. Mookherjee, Singh, Maiti (2019) "Quorum sensing inhibitors: Can endophytes be prospective sources?" *Arch. Microbiol*
30. Pfaller, Castanheira (2016) "Nosocomial candidiasis: Antifungal stewardship and the importance of rapid diagnosis" *Med. Mycol*
31. Cleveland, Farley, Harrison et al. (2008) "Changes in incidence and antifungal drug resistance in candidemia: Results from population-based laboratory surveillance in Atlanta and Baltimore" *Clin. Infect. Dis*
32. Andes, Safdar, Baddley et al. (2012) "Impact of treatment strategy on outcomes in patients with candidemia and other forms of invasive candidiasis: A patient-level quantitative review of randomized trials" *Clin. Infect. Dis*
33. Garey, Rege, Pai et al. (2006) "Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: A multi-institutional study" *Clin. Infect. Dis*
34. Clancy, Nguyen (2013) "Finding the "missing 50%" of invasive candidiasis: How nonculture diagnostics will improve understanding of disease spectrum and transform patient care" *Clin. Infect. Dis*
35. Salmanton-García, Sprute, Stemler et al. (2020) "COVID-19-associated pulmonary aspergillosis" *Emerg. Infect. Dis*
36. Ramage, Rajendran, Gutierrez-Correa et al. (2011) "Aspergillus biofilms: Clinical and industrial significance" *FEMS Microbiol. Lett*
37. Nett, Andes (1920) "Candida albicans biofilm development, modeling, and measurement" *Curr. Protoc. Microbiol*
38. Lortholary, Renaudat, Sitbon et al. (2014) "Worrisome trends in incidence and mortality of candidemia in intensive care units (Paris area, 2002-2010)" *Intensive Care Med*
39. Brown, Denning, Levitz (2012) "Tackling human fungal infections" *Science*
40. Cowen, Anderson, Kohn (2002) "Evolution of drug resistance in Candida albicans" *Annu. Rev. Microbiol*
41. Lionakis (2014) "New insights into innate immune control of systemic candidiasis" *Med. Mycol*
42. Romani (2011) "Immunity to fungal infections" *Nat. Rev. Immunol*
43. Sudbery (2011) "Growth of Candida albicans hyphae" *Nat. Rev. Microbiol*
44. Mayer, Wilson, Hube (2013) "Candida albicans pathogenicity mechanisms" *Virulence*
45. Harriott, Noverr (2009) "Candida albicans and Staphylococcus aureus form polymicrobial biofilms: Effects on antimicrobial resistance" *Antimicrob. Agents Chemother*
46. Shirtliff, Peters, Jabra-Rizk (2009) "Cross-kingdom interactions: Candida albicans and bacteria" *FEMS Microbiol. Lett*
47. Peleg, Hogan, Mylonakis (2010) "Medically important bacterial-fungal interactions" *Nat. Rev. Microbiol*
48. Yapar (2014) "Epidemiology and risk factors for invasive candidiasis" *Ther. Clin. Risk Manag*
49. Guinea (2014) "Global trends in the distribution of Candida species causing candidemia" *Clin. Microbiol. Infect*
50. Odds, Hanson, Davidson et al. (2007) "One year prospective survey of Candida bloodstream infections in Scotland" *J. Med. Microbiol*
51. Wisplinghoff, Bischoff, Tallent et al. (2004) "Nosocomial bloodstream infections in US hospitals: Analysis of 24,179 cases from a prospective nationwide surveillance study" *Clin. Infect. Dis*
52. Bassetti, Merelli, Righi et al. (2013) "Epidemiology, species distribution, antifungal susceptibility, and outcome of candidemia across five sites in Italy and Spain" *J. Clin. Microbiol*
53. Denning, Pleuvry, Cole (2011) "Global burden of chronic pulmonary aspergillosis as a sequel to pulmonary tuberculosis" *Bull. World Health Organ*
54. Chowdhary, Sharma, Meis (2017) "Candida auris: A rapidly emerging cause of hospital-acquired multidrug-resistant fungal infections globally" *PLoS Pathog*
55. Oladele, Bongomin, Gago et al. (2017) "HIV-associated cryptococcal disease in resource-limited settings: A case for "prevention is better than cure" *J. Fungi*
56. Rajasingham, Smith, Park et al. (2017) "Global burden of disease of HIV-associated cryptococcal meningitis: An updated analysis" *Lancet Infect. Dis*
57. Liao, Wang, Feng et al. (2021) "Machine learning methods applied to predict ventilator-associated pneumonia with Pseudomonas aeruginosa infection via sensor array of electronic nose in intensive care unit" *Biosens. Bioelectron*
58. Segal, Frenkel, Vessman et al. "Machine learning algorithm for early detection of end-stage renal disease" *NPJ Digit*
59. Cruciani, Mengoli, Loeffler et al. (2015) "Polymerase chain reaction blood tests for the diagnosis of invasive aspergillosis in immunocompromised people" *Cochrane Database Syst. Rev*
60. Posteraro, Posteraro, Sanguinetti (2011) "Diagnosis of invasive aspergillosis: The role of PCR" *Expert Rev. Anti Infect. Ther*
61. Lionakis, Levitz (2018) "Host control of fungal infections: Lessons from basic studies and human cohorts" *Annu. Rev. Immunol*
62. Johnson, Perfect (2007) "Use of antifungal combination therapy: What laboratory and clinical studies tell us so far" *Expert Rev. Anti Infect. Ther*
63. Ashman, Papadimitriou (1995) "Production and function of cytokines in natural and acquired immunity to Candida albicans infection" *Microbiol. Rev*
64. Cassone, Casadevall (2012) "Recent progress in vaccines against fungal diseases" *Curr. Opin. Microbiol*
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" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC11931761&blobtype=pdf | Samy Kasem
## Virology Journal
Correction: Full-length genome reveals genetic diversity and extensive recombination patterns of Saudi GI-1 and GI-23 genotypes of infectious bronchitis virus In this article [1], the affiliation indicator of the first author was inadvertently disrupted during the correction process. There was a change in first author affiliation during the correction process; Ali N. Alhafufi was originally indicated with 1, the author replaced 1 with 7, and hence the affiliation indicator 7 should have been renumbered as 1. This oversight has been rectified, and all author affiliations have been renumbered accordingly.
The original article has been corrected.
## References
1. Alhafufi, Kasem, Almajhdi (2025) "Full-length genome reveals genetic diversity and extensive recombination patterns of Saudi GI-1 and GI-23 genotypes of infectious bronchitis virus" *Virol J* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12784427&blobtype=pdf | # Association Between Preoperative Negative Emotional States and Eye Movement During Photorefractive Keratectomy
Hesam Hashem, Hooman Ahmadzad, Alire Razav, Hassan Asadignadomani, Zahra Montazeriani, Mehdi Khodaparast
## Abstract
Purpose: To describe the link between negative emotions-depression, anxiety, and stressand eye movement during photorefractive keratectomy (PRK). Methods: This comparative case series was conducted on 53 PRK candidates and involved completing the Depression, Anxiety, and Stress Scale (DASS-21) before surgery. Eye movement, measured as the radial distance between the pupil and laser center during each shot, was analyzed. Average distance indicated centralization accuracy, while standard deviation (SD) indicated precision. Stress, depression, anxiety, and their relationship with eye movement during PRK were studied.
Results:The mean and SD of eye movements were not significantly correlated with depression, anxiety, stress, or the total DASS-21 score. A weak negative correlation was observed between the preoperative spherical equivalent (SE) and stress (r = -0.305, P = 0.004), anxiety (r = -0.401, P < 0.001), and total DASS-21 scores (r = -0.324, P = 0.002). Lastly, a weak positive correlation was found between ablation time and the SD of eye movement (r = 0.473, P < 0.001).
Conclusion:The DASS-21 questionnaire showed no link between negative emotions and eye movement. Additionally, longer ablation times correlated with greater SD of eye movement.
## INTRODUCTION
Uncorrected refractive errors are the most important cause of visual impairment worldwide. [1] As an alternative to glasses or contact lenses for optical correction of refractive errors, laser corneal refractive surgery has grown in popularity. [2] Despite years of continuous advancements in refractive surgery techniques, photorefractive keratectomy (PRK) remains a highly common refractive procedure worldwide. [3,4] The disadvantages of PRK consist of a slower healing rate, risk of haziness, and longer healing process. At the same time, its advantages include a more residual stromal bed and a reduced risk of ectasia. [3] Before surgery, most patients experience some degree of anxiety and stress, which is a natural reaction to unpredictable and risky conditions. Nevertheless, excessive stress and anxiety levels can potentially have a negative impact on treatment outcomes. [5] Since surgery is a complex procedure, we are unable to review the moderators of intraoperative stress and how they affect performance and outcomes. [6] Accordingly, some studies investigated the patient's cooperation during the PRK surgery as a crucial factor for its success. [7] This is because stress and anxiety can affect the patient's cooperation during surgery. It is possible to experience eye and head movements during the procedure as a result of stress, which can manifest differently in each eye. [7] In this context, we decided to investigate the effect of depression, anxiety, and stress on the amount of eye movement during PRK using the Depression, Anxiety, and Stress Scale (DASS-21).
## METHODS
## Study Design and Patients
This comparative case series enrolled patients from the Cornea Clinic at Farabi Eye Hospital. Participants signed informed consent and completed eye examinations from January to March 2023. The examinations included uncorrected visual acuity (UCVA) and bestcorrected visual acuity (BCVA) using E-chart (converted to LogMAR), manifest and cycloplegic refraction, intraocular pressure examination, slit lamp evaluation, complete posterior segment assessment, and Pentacam tomography (Oculus Optikgeräte GmbH, Wetzlar, Germany). Patients filled out the DASS-21 questionnaire to assess depression, anxiety, and stress 15-30 minutes before PRK. A single skilled surgeon performed all procedures, including PRK, in which the epithelium was removed with an alcohol application for 20 seconds. Eye tracking was centered on the entrance pupil for laser ablation using Amaris 1050RS (SCHWIND eye-tech-solutions, Kleinostheim, Germany). The Schwind excimer laser uses real-time eye-tracking technology to guarantee accurate ablation. [8][9][10] Before the commencement of the ablation procedure, the patient is instructed to focus on a designated target. At this point, the Schwind eye tracker captures the positions of critical ocular landmarks, such as the center of the pupil and Purkinje reflections. Throughout the ablation process, a high-speed camera, along with an analysis processor, continuously monitors the eye, identifying any movements or deviations from the initial calibration position. Surgery data were extracted directly from the device upon completion. The study aimed to investigate the correlation between preoperative emotional states (as measured by the DASS-21) and eye movements. The sample size was calculated to be at least 39 patients, considering a 15% effect size, 95% confidence level, and 80% power using Cohen's formula. This study received approval from the Ethics Committee (ID: IR.TUMS.FARABIH.REC.1401.047) and the Institutional Review Board of Tehran University of Medical Sciences. The study was conducted in accordance with the principles outlined in the Declaration of Helsinki. All patients provided signed informed consent.
## Eligibility Criteria
The study included patients over 18 years of age who had a maximum keratometry of <47 diopters (ranging from 37.00 to 47.00 diopters based on the study by Alsheri et al [11] ) and a regular Pentacam reading. Individuals with any type of eye disease, including keratoconus, corneal scar, history of hydrops; those with a previous eye surgery, including corneal cross-linking, corneal inlay, and keratorefractive surgery; those not completing the follow-up; and those with a corneal thickness <470 micrometers (to minimize the possibility of developing post-refractive surgery ectasia) were excluded from the study. We also excluded patients with poor cooperation or recognized difficulties before surgery, such as narrow palpebral fissures or suspicious topography.
## Data Collection Instruments
## Depression, Anxiety, and Stress Scale (DASS-21)
DASS-21 is a self-assessment tool with three subscales to measure negative emotional states: depression, stress, and anxiety. Each subscale has seven items. The depression subscale assesses dysphoria, hopelessness, and lack of interest. The stress subscale evaluates autonomic arousal and situational anxiety. The anxiety subscale measures difficulty relaxing and nervous arousal. Examples include: "I felt like I had nothing to look forward to" (Depression), "I found it difficult to wind down" (Stress), and "I experienced trembling in the hands" (Anxiety). [12] Sahebi et al translated this questionnaire into Persian, and its reliability and validity were confirmed. [13] The Persian version of the DASS-21 has demonstrated excellent psychometric properties, making it an effective and reliable assessment tool. [16] This makes it an invaluable resource for accurately evaluating depression, anxiety, and stress levels among the Iranian population. [14] In our study, participants rated their experience of each condition over the past week on a 4-point scale from 0 (not at all) to 3 (very much). The scores for depression, stress, and anxiety were obtained by summing the respective scores. [15,16] The responses to the component items were added up to determine the scores for each subscale. Participants were given a clear explanation of the study's purpose and instructions at the beginning. Only those who confirmed their willingness to participate were given access to the questionnaire. Responses were structured with fixed choices, requiring participants to complete all questions to avoid missing data.
## Mean and standard deviation (SD) of eye movement
In our work, we used the data provided by the laser eye tracker to measure eye movement, defined as the radial distance between the pupil and laser centers, based on the location of the pupil center in each laser shot during the operation. The mean of these values can be considered an indicator of accuracy in centralization. In other words, a lower mean value represents less eye movement during ablation. Moreover, the SD of pupil center locations data, which indicates how spread-out the data is compared to the average value, was determined by taking the square root of the sum of eigenvalues as a measure of centering accuracy. It means the lower the SD value, the more precise the centralization is.
## Statistical Analysis
Continuous variables were assessed for normality by the Kolmogorov-Smirnov test and reported as mean ± SD when normally distributed, or as median ± interquartile range otherwise. Qualitative values were reported as frequency (percentage). Pearson's correlation coefficient (r) was used to measure the statistical association between variables if normality could be assumed, and Spearman's rank correlation was applied otherwise. Paired t-test was used to compare two related samples to identify significant differences in case of normal distribution, whereas a Wilcoxon signed-rank test was used if the data lacked normal distribution. For data analysis, SPSS version 22.0 (IBM Inc., Chicago, IL, USA) was used at a significance level of 0.05.
## RESULTS
This study analyzed data from 103 eyes of 53 patients (34 Postoperative visual acuity (VA) showed significant improvement (P < 0.001). Eye movement metrics were not correlated with depression, anxiety, stress, or total DASS-21 scores (P > 0.1 for all). Subgroup analyses based on DASS-21 severity levels (normal to extremely severe) confirmed no significant correlations between the mean or SD of eye movement and psychological scores. Similarly, no correlations were observed between SD of eye movement and depression, anxiety, or stress scores (P > 0.1 for all). A weak negative correlation was observed between anxiety/stress scores and sphere [Table 2].
Ablation time showed a positive correlation with the SD of eye movement (r = 0.473, P < 0.001) [Table 3]. UCVA showed no significant correlation with either the mean or SD of eye movement (P = 0.388 and P = 0.848, respectively). Gender also did not correlate with these eye movement measures (P = 0.878 and P = 0.815, respectively). Age exhibited a weak negative correlation with the SD of eye movement (r = -0.211, P = 0.032), but it was not associated with the mean eye movement (P = 0.611). A weak negative correlation was noted between anxiety/stress scores and SE [Table 2]. The correlations between sphere and cylinder variables and both the mean and SD of eye movement were weak and not statistically significant [Table 3].
## DISCUSSION
This study was designed and implemented to investigate the association between the level of negative emotional states recorded in the DASS-21 questionnaire, as well as its subscales, with eye movement in patients who were candidates for PRK. The results of this study indicate that patients with lower SE exhibit higher anxiety scores, and a weak negative association exists between these two variables. As a result of variance in pupil center position, the mean of eye movement may not entirely represent the actual amount of eye movement during PRK. As part of our investigation, we calculated the SD of the distance between the pupil center location and the mean pupil movement at each laser pulse for each patient during ablation. In this study, a moderately positive correlation was found between ablation time and the SD of eye movement. Our findings suggest that the increase in the ablation time may lead to a significant increase in the SD of eye movement. Indeed, a large SD of eye movement indicates that patients experience greater and more dispersed eye movements during ablation than their mean pupil movement. A previous study has shown that patients with lower VA generally indicate greater decentration during surgery. It was also postulated that these patients cannot fixate on the distant blinking light. [8] In our sample, SE and UCVA were not significantly correlated with eye movement measures. However, higher eye movement was observed in patients with higher ablation time. Longer ablation times with more laser shots provide more opportunities for patients' eye movement, resulting in higher decentration. [17] Research has demonstrated that elevated anxiety levels in patients can adversely affect the outcomes of refractive surgeries such as PRK. [18,19] Anxiety often leads to increased involuntary eye movements during the procedure, which can prolong the ablation process and diminish the precision of the laser treatment. [18,19] To address these challenges, techniques such as offering a fixation target, utilizing a soft bandage contact lens, and administering mild sedatives can be effective in minimizing eye movement and enhancing surgical outcomes. [19] Our results demonstrated a notable improvement in VA following the procedure. Research indicates that patients generally experience enhanced VA after undergoing PRK, with many achieving 20/20 vision or better. [20] Another study highlighted that PRK delivered excellent refractive outcomes and long-term stability, with the majority of patients achieving 20/20 vision or better 12 months after surgery. [20] Some reports have shown that body position can affect eye movements during PRK. A study with customized ablation showed that most rotational movements (in 46 eyes) were caused by positional changes, and these movements were more significant between upright and supine positions. [21] The supine position may show higher cyclotorsions due to vestibular input variations. [9] As a result of these cyclotorsions, laser ablation can be misplaced, resulting in irregular astigmatism and poor VA correction without recovery. [22] In the present study, we investigated the effect emotional states on eye movements during the PRK procedure. We did not find a significant association between DASS-21 emotional states and eye movement measures. However, it is worth noting that most of our patients had low preoperative scores, which may have obscured any existing effects. Eye movement during PRK can negatively affect the aberration and contrast sensitivity outcomes and result in decreased VA. Therefore, it is essential to identify patient groups with a high propensity for eye movement. Studying specific populations with high levels of anxiety and stress would be beneficial in this respect.
There are key differences between conventional PRK and transepithelial PRK (trans-PRK). Trans-PRK typically results in faster visual recovery, with patients achieving better VA sooner and experiencing less post-surgical pain. [19,23] The epithelial layer heals more quickly in trans-PRK. [19,23] Both procedures effectively correct low to moderate myopia and astigmatism and offer comparable refractive outcomes. [23] Conventional PRK involves manually removing the epithelial layer before reshaping the cornea, while trans-PRK combines both steps using a laser, which may prolong the procedure. [24] This extended treatment time can lead to potential complications, such as loss of centration, although modern techniques have reduced these risks. [24] Emotional factors play a crucial role in the outcomes of refractive procedures such as SMILE (small incision lenticule extraction) and femto-LASIK (femtosecond laser-assisted in situ keratomileusis). [25] Research has shown that patients' psychological well-being can significantly influence their perception of surgical outcomes and overall satisfaction. [25][26][27] For instance, preoperative concerns such as anxiety, depression, and unrealistic expectations can affect how individuals view their outcomes and their quality of life following surgery. [26,28] Conversely, patients who maintain a positive emotional state and possess realistic expectations tend to report higher levels of satisfaction and better results. [29] Additionally, emotional support and a strong doctor-patient relationship can greatly enhance both the surgical experience and recovery process. [30] By addressing emotional concerns and providing comprehensive counseling before and after the procedure, we can significantly improve patient outcomes and overall satisfaction.
Uncooperative patients undergoing SMILE or femto-LASIK procedures can experience intraoperative complications. A major concern is suction loss, which occurs when a patient moves or fails to adhere to instructions. [31] This issue is particularly pronounced in SMILE due to longer laser cutting time and lower suction pressure, sometimes necessitating the abandonment of the procedure. Complications may include incomplete lenticule dissection, decentration, or interface abnormalities, all of which can compromise visual outcomes. Excessive eye movement may also lead to difficulties in laser centration, increasing the risk of postoperative aberrations and suboptimal refractive results. [32,33] With faster ablation speeds and advanced tracking systems, modern excimer laser platforms have significantly improved refractive surgery, particularly for patients with less compliant eyes. These developments reduce the time patients must stay motionless during treatments, which is particularly important for individuals who may have fixation issues. [34] Reduced anxiety and greater patient satisfaction have been linked to shorter treatment durations. Furthermore, real-time eye-tracking systems enable constant laser monitoring and modification, reducing the impact of eye movements and guaranteeing more precise corneal reshaping. [34] Older excimer lasers, on the other hand, had less sophisticated tracking and slower ablation speeds, which resulted in longer surgery times and a higher chance of misalignment. [35,36] This could have a detrimental effect on safety and visual results in less cooperative patients. [36] This study had some limitations. First, some data might be missing because this is a retrospective study. The small sample size limited the analysis. Additionally, the study did not include an evaluation of the accuracy of the laser center and pupil center alignment by the surgeon prior to commencing the surgical operation. It is imperative to take into account the correlation between the data collected from both eyes while merging them. Disregarding this correlation can lead to flawed interpretations that can have serious consequences. High skewness in ablation time warrants caution in interpreting results. Acknowledging skewness as a limitation enhances the rigor of research. It is crucial to note that this study did not evaluate the quality of vision, specifically factors such as contrast sensitivity and higher-order aberrations. This oversight may significantly impact the overall assessment. To better understand the progression of VA and refractive stability, we must extend the follow-up period for a more comprehensive evaluation. In this regard, it is essential to explore the distinct patterns of eye movements observed between the first and second eye. Analyzing these differences can provide valuable insights into visual processing and coordination. It would be intriguing to investigate the potential correlation between patients' motivation and their SE. It is rewarding to conduct larger, well-designed studies with longer follow-up periods to evaluate the effect of negative emotional states on eye movement before PRK.
In summary, no association was found between eye movements and negative emotional states, as measured by the DASS-21 questionnaire. Lower SE in PRK candidates slightly increased anxiety in these patients. An increase in ablation time was associated with an increase in the SD of eye movement. Future studies on patients with specifically high DASS-21 scores (with high stress, anxiety, and depression) would help identify patient groups with an increased risk of eye movement before PRK.
## References
1. Sharma, Lepcha, Lhamo et al. (2020) "Visual impairment and refractive error in school children in Bhutan: The findings from the Bhutan School Sight Survey (BSSS 2019)" *PLoS One*
2. Jabbour, Bower (2021) "Refractive Surgery in the US in 2021" *JAMA*
3. Ozulken, Gokce (2020) "Evaluation of the effect of optic zone diameter selection on high-order aberrations in photorefractive keratectomy excimer laser treatment" *Lasers Med Sci*
4. Somani, Moshirfar, Patel (2022) "Photorefractive keratectomy"
5. Kassahun, Mehdorn, Wagner et al. (2022) "The effect of preoperative patient-reported anxiety on morbidity and mortality outcomes in patients undergoing major general surgery" *Sci Rep*
6. Chrouser, Xu, Hallbeck et al. (2018) "The influence of stress responses on surgical performance and outcomes: Literature review and the development of the surgical stress effects (SSE) framework" *Am J Surg*
7. Zarei-Ghanavati, Eslampour, Shokouhirad et al. (2019) "The effect of eye dominancy on patients' cooperation and perceived pain during photorefractive keratectomy" *J Curr Ophthalmol*
8. Adib-Moghaddam, Soleyman-Jahi, Tofighi et al. (2018) "Factors associated with ocular cyclotorsion detected by high-speed dualdetection eye tracker during single-step transepithelial photorefractive keratectomy" *J Refract Surg*
9. Fea, Sciandra, Musso et al. (2006) "Cyclotorsional eye movements during a simulated PRK procedure" *Eye*
10. Fahd, Jabbour, Fahed (2014) "Static cyclotorsion measurements using the Schwind Amaris laser" *Arq Bras Oftalmol*
11. Alshehri, Abdelaal, Abudawood et al. (2022) "Normative values for corneal tomography and comparison of both eyes in young Saudi males with 20/20 vision using Pentacam-HR Scheimpflug Imaging" *Clin Ophthalmol*
12. Mellor, Vinet, Xu et al. (2015) "Factorial invariance of the DASS-21 among adolescents in four countries" *Eur J Psychol Assess*
13. Sahebi, Asghari, Salari (2005) "Validation of depression anxiety and stress scale (DASS-21) for an Iranian population" *J Develop Psychol*
14. Kakemam, Navvabi, Albelbeisi et al. (2022) "Psychometric properties of the Persian version of Depression Anxiety Stress Scale-21 Items (DASS-21) in a sample of health professionals: A cross-sectional study" *BMC Health Serv Res*
15. Jiang, Yan, Jin et al. (2020) "The depression anxiety stress Scale-21 in Chinese hospital workers: Reliability, latent structure, and measurement invariance across genders" *Front Psychol*
16. Beaufort, Weert-Van Oene, Buwalda et al. (2017) "The depression, anxiety and Preoperative Negative Emotional States in PRK; Hashemian et al stress scale (DASS-21) as a screener for depression in substance use disorder inpatients: A pilot study" *Eur Addict Res*
17. Porter, Yoon, Macrae et al. (2005) "Surgeon offsets and dynamic eye movements in laser refractive surgery" *J Cataract Refract Surg*
18. Steinert, Mccolgin (2013) "Surface ablation: Photorefractive keratectomy, LASEK, Epi-LASIK, and Epi-LASEK"
19. Gaeckle (2021) "Early clinical outcomes and comparison between trans-PRK and PRK, regarding refractive outcome, wound healing, pain intensity and visual recovery time in a real-world setup" *BMC Ophthalmol*
20. Tananuvat, Winaikosol, Niparugs et al. (2021) "Twelve-month outcomes of the wavefront-optimized photorefractive keratectomy for high myopic correction compared with low-to-moderate myopia" *Clin Ophthalmol*
21. Taylor, Teiwes (2004) "Eye tracking and alignment in refractive surgery: Requirements for customized ablation"
22. Fahd, Jabbour, Fahed (2014) "Static cyclotorsion measurements using the Schwind Amaris laser" *Arq Bras Oftalmol*
23. Hashemi, Alvani, Aghamirsalim et al. (2022) "Comparison of transepithelial and conventional photorefractive keratectomy in myopic and myopic astigmatism patients: A randomized contralateral trial" *BMC Ophthalmol*
24. Way, Elghobaier, Nanavaty (2024) "Transepithelial photorefractive keratectomy-Review" *Vision*
25. Matsuguma, Negishi, Kawashima et al. (2018) "Patients' satisfaction and subjective happiness after refractive surgery for myopia" *Patient Prefer Adherence*
26. Klokova, Sakhnov, Geydenrikh et al. (2019) "Quality of life after refractive surgery: ReLEx SMILE vs Femto-LASIK" *Clin Ophthalmol*
27. Levett, Grimmett (2019) "Psychological factors, prehabilitation and surgical outcomes: Evidence and future directions" *Anaesthesia*
28. Geoffrion, Koenig, Zheng et al. (2021) "Preoperative depression and anxiety impact on inpatient surgery outcomes: A prospective cohort study" *Ann Surg Open*
29. Wang, Liu, Wang (2023) "Patient satisfaction impact indicators from a psychosocial perspective" *Front Public Health*
30. Kaba, Sooriakumaran (2007) "The evolution of the doctorpatient relationship" *Int J Surg*
31. Moshirfar, Mccaughey, Reinstein et al. (2015) "Small-incision lenticule extraction" *J Cataract Refract Surg*
32. Alió, Toprak, Alrabiah "Albert and Jakobiec's principles and practice of ophthalmology"
33. Jacob "Complications of smile surgery"
34. Ang, Gatinel, Reinstein et al. (2021) "Refractive surgery beyond 2020" *Eye*
35. El Bahrawy, Alió (2015) "Excimer laser 6(th) generation: State of the art and refractive surgical outcomes" *Eye Vis*
36. Mrochen, Eldine, Kaemmerer et al. (2001) "Improvement in photorefractive corneal laser surgery results using an active eye-tracking system" *J Cataract Refract Surg* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12646010&blobtype=pdf | # Interferon alpha-inducible protein 27 (IFI27) inhibits hepatitis B virus (HBV) transcription through downregulating cellular transcription factor C/EBPα
Xiaoyang Yu, Cheng-Der Liu, Sheng Shen, Elena Kim, Zhentao Liu, Hu Zhang, Ning Sun, Yuanjie Liu, Pia Martensen, Yufei Huang, Haitao Guo
## Abstract
Interferon alpha (IFNα) is the only approved immunomodulatory drug for chronic hepatitis B treatment, exerting its antiviral effects through the induction of interferon-stimulated genes (ISGs). To identify key antiviral ISGs that inhibit hepatitis B virus (HBV) replication, we performed transcriptome analysis of IFNα-treated HepG2-NTCP cells and found that IFI27 was among the most differentially expressed genes. The high inducibility of IFI27 by IFNα was further validated in primary human hepatocytes. Overexpression of IFI27 significantly suppressed HBV replication in both HBV-transfec ted and -infected cells, primarily by reducing HBV RNA transcription. Conversely, IFI27 knockdown markedly diminished the antiviral effect of IFNα in HBV-infected cells. IFI27 is predominantly localized in the cytoplasm, and RNA-seq analysis revealed that IFI27 inhibits HBV transcription without drastically altering the host transcriptome, indicating that IFI27 does not inhibit HBV transcription directly or through altering the transcription of cellular transcription factors or inducing antiviral signaling pathways. Instead, we found that IFI27 suppresses HBV transcription by promoting the ubiquitination-depend ent proteasomal degradation of C/EBPα in the cytoplasm, a cellular transcription factor critical for HBV RNA transcription. Further investigation identified the E3 ubiquitin ligase SKP2 as a key mediator of this process, facilitating IFI27-induced C/EBPα ubiquiti nation and degradation. Notably, SKP2 knockdown abrogated IFI27's antiviral activity against HBV. Taken together, our findings reveal that IFI27 contributes to IFNα-mediated antiviral activity against HBV by targeting C/EBPα for SKP2-dependent ubiquitination and proteasomal degradation. This study thus sheds new light on the antiviral mechanism of IFNα-based therapy for chronic hepatitis B.
IMPORTANCE Chronic hepatitis B virus (HBV) infection affects approximately 250 millionpeople worldwide with limited treatment options. Interferon alpha (IFNα) remains the only approved immunomodulatory treatment for chronic hepatitis B, working in HBV-infected liver cells through inducing antiviral genes. To identify key interferoninducible genes involved in HBV suppression, we performed transcriptome analysis of IFNα-treated liver cells and identified IFI27 as one of the most upregulated genes. Functional studies demonstrated that IFI27 inhibits HBV replication by reducing viral RNA transcription, and its knockdown significantly impaired the antiviral effect of IFNα. Mechanistically, IFI27 suppresses HBV transcription by promoting the ubiquitin-protea some-mediated degradation of C/EBPα, a transcription factor critical for HBV RNA synthesis. This process is dependent on the E3 ubiquitin ligase SKP2, as SKP2 knock down abolished IFI27-mediated antiviral activity. These findings reveal IFI27 as a critical mediator of IFNα-induced antiviral responses against HBV and provide new insights into host-directed antiviral mechanisms with potential therapeutic implications.
## KEYWORDS HBV, IFI27
D espite the existence of a highly effective prophylactic vaccine, chronic hepatitis B virus (HBV) infection remains a formidable global health challenge affecting over 257 million individuals worldwide (1)(2)(3). HBV is a noncytopathic, hepatotropic virus belonging to the Hepadnaviridae family (4). The virion harbors a 3.2 kilobase (kb), partially double-stranded (ds), relaxed circular (rc) DNA genome (5). Upon infection by exploiting the hepatocyte-specific receptor sodium taurocholate cotransporting polypeptide (NTCP) (6), viral rcDNA is delivered into the nucleus to form an episomal covalently closed circular DNA (cccDNA) via DNA repair process (7)(8)(9). By utilizing cccDNA as the genuine template for transcription, five mRNAs with overlapping 3´ ends are transcribed, including the 3.5 kb pregenomic (pg) and precore (pC) mRNA, 2.4/2.1 kb preS/S mRNA, and 0.7 kb X mRNA (5,10). HBV replicates its DNA genome via viral polymerase-primed and -catalyzed reverse transcription of pgRNA in the cytoplasmic nucleocapsid (11,12). The newly synthesized rcDNA-containing nucleocapsid is either enveloped and secreted through multivesicular bodies as a virion, or it can be recycled back to the nucleus to replenish the cccDNA pool (4,13).
Interferons (IFNs) are a group of signaling proteins produced and released by host cells in response to diverse pathogens, such as bacteria, viruses, and parasites (14). Interferon alpha (IFNα) belongs to type I IFN, which is the first remedy approved for the treatment of chronic hepatitis B and can achieve sustained virological response in a minority of patients, with significant side effects (15). Upon binding of IFNα to its cognate receptor on cell surface, it initiates a JAK/STAT signaling cascade, which leads to the induction of more than 300 IFN-stimulated genes (ISGs) (16,17). Besides the immunomo dulatory functions of some ISGs induced in immune cells, certain ISGs play antiviral activities directly in the virally infected cells (16). Various ISGs have been reported to inhibit HBV replication at different steps, primarily through both transcriptional and post-transcriptional mechanisms (18)(19)(20). For example, APOBEC3A induces cccDNA deamination and degradation (21); STAT1, SMCHD1, and PML bind to cccDNA minichro mosome and shape a suppressive epigenetic status of cccDNA (22); TRIM5γ inhibits HBV transcription by promoting the degradation of HBx (23); zinc finger antiviral protein (ZAP) and ISG20 bind to HBV RNA and accelerate RNA decay (24,25); indoleamine 2,3-dioxygenase can induce tryptophan deprivation and block HBV protein translation (26); tetherin inhibits HBV virion egress (27), etc. Identifying additional anti-HBV ISGs and elucidating their antiviral mechanisms will enhance our understanding of IFNα therapy and potentially optimize its use for treating HBV infection.
In search of ISGs that inhibit HBV replication, we have previously found that the 12 kDa ISG product, referred to as ISG12a (also known as interferon alpha-inducible protein 27 [IFI27]), inhibited HBV replication primarily through reducing HBV RNA transcript levels in cell cultures (26). This protein belongs to the FAM14 gene family, whose members encode small hydrophobic proteins that contain at least one copy of an approximately 80 amino acid conserved ISG12 motif (28). The hydrophobic nature of these proteins suggests a potential association with cellular membranes. To date, IFI27 has been reported to exert its antiviral effects against hepatitis C virus (HCV), West Nile virus, and Newcastle disease virus, utilizing different mechanisms (29)(30)(31)(32). Furthermore, IFI27 has shown potential correlation with respiratory syncytial virus (RSV) infection due to its significant expression in clinical cases during RSV infection (33). Recently, IFI27 has also emerged as an early predictor of coronavirus disease 2019 . It has been reported that the expression of IFI27 was found in the respiratory tract of COVID-19 patients, and an elevated IFI27 expression in the lower respiratory tract was associated with higher viral load (34).
In this study, we aimed to further characterize the antiviral function and mechanism of IFI27 in the innate control of HBV. We report herein that (i) endogenous IFI27 acts as a restriction factor for HBV replication and is highly upregulated by IFNα to enhance its antiviral effect; (ii) IFI27 inhibits HBV replication primarily by suppressing viral promoter activity; (iii) the antiviral mechanism of IFI27 involves promoting the degradation of a cellular transcription factor (TF) C/EBPα (CCAAT-enhancer-binding protein alpha) via the ubiquitin-proteasome pathway; and (iv) the E3 ubiquitin ligase SKP2 is required for IFI27-mediated C/EBPα ubiquitination and degradation and plays a key role in IFI27-induced suppression of HBV transcription. Our findings provide new insights into IFI27 biology and its role in IFNα-based treatment of HBV infection, offering a poten tial opportunity for optimizing IFNα therapy and developing novel antiviral strategies against HBV.
## RESULTS
## IFI27 is highly induced in hepatocytes by IFNα treatment
In our previous study to identify anti-HBV ISGs by screening a collection of over 30 ISG expression plasmids in pHBV1.3-cotransfected HepG2 cells, IFI27 emerged as a hit exhibiting potent antiviral activity (26). In the current study, to further characterize the antiviral effect of IFI27 on HBV infection and elucidate its mechanism of action, we first assessed the inducibility of IFI27 in hepatocyte-derived cells. The RNA-seq transcrip tomic analysis demonstrated that IFI27 was one of the most highly induced ISGs in HepG2-NTCP cells upon IFNα treatment (Fig. 1A). Next, the inducibility of IFI27 mRNA and protein expression by IFNα was further analyzed in HepG2 and/or PHH cells, which demonstrate that IFNα induces IFI27 expression in a time-and dose-dependent manner (Fig. 1B through D). The observed less efficient IFI27 induction in PHH cells than that in HepG2 cells might be attributable to the specific batch of PHH cells used in this experiment.
## IFI27 inhibits HBV replication primarily through suppressing viral transcrip tion
To assess the antiviral effect of IFI27 on HBV replication in hepatocyte-derived cells, we co-transfected FLAG-IFI27 or a control vector along with pHBV1.3 into either HepG2 cells or Huh7 cells, followed by analyses of HBV total RNA and cytoplasmic capsid-asso ciated (core) DNA replicative intermediates. As shown in Fig. 2A andB, the overexpres sion of IFI27 markedly decreased the levels of both HBV core DNA and total RNA, including the 3.5 kb precore mRNA and pgRNA, the latter being the template for HBV reverse transcription (upper panels). Moreover, the quantitative analysis revealed a proportional ~50% reduction in both 3.5 kb RNA and core DNA (lower panels), indicating that IFI27 inhibits HBV DNA replication primarily through reducing viral pgRNA. Given the basal expression of IFI27 detected in both HepG2 and PHH cells (Fig. 1B andC), we next examined whether endogenously expressed IFI27 functions as a restriction factor against HBV transcription. To this end, pHBV1.3 was transfected into HepG2 cells with or without small interfering RNA (siRNA) knockdown (KD) of IFI27. The results revealed that knockdown of basal IFI27 resulted in a modest increase in HBV RNA levels (Fig. 2C), suggesting that IFI27 functions as an endogenous restriction factor for HBV under basal conditions.
Next, we assessed the antiviral effect of IFI27 in HBV-infected HepG2-NTCP cells. Upon HBV infection, the overexpressed IFI27 markedly reduced the levels of HBV DNA and 3.5 kb viral RNA levels as revealed by Southern and Northern blots, respectively (Fig. 2D). Interestingly, Southern blot analysis showed that only the ssDNA intermediate, but not the rcDNA, was reduced by IFI27, suggesting that the majority of rcDNA in HBV-infected HepG2-NTCP cells is derived from the inoculated rcDNA in HBV virions, while the ssDNA indicates de novo HBV DNA replication, which is consistent with our previous studies (35)(36)(37).
Previous studies have shown that IFI27/ISG12a can sensitize cytokine-or DNA damage-induced cell apoptosis through the formation of pores, inducing permeability in the inner mitochondrial membrane and disrupting the mitochondrial membrane potential (38,39). Consistent with the reported mitochondrial localization of IFI27, our confocal immunofluorescence assay confirmed that the ectopically expressed FLAG-IFI27 was predominantly and widely localized in cytoplasm, including the perinuclear areas of HepG2 cells, and it was largely associated with mitochondria in the cytoplasm, as evidenced by the colocalization of FLAG-IFI27 and MitoTracker (Fig. 3A). Therefore, we then examined whether the observed anti-HBV effects of IFI27 are related to cell death. We transfected HepG2 cells with either a control vector, pHBV1.3 plus control [vge]/cell) for 24 h, followed by transfection with control vector or FLAG-IFI27. Cells continued to be cultured for an additional 5 days. The harvested cells were subjected to HBV total RNA Northern blot, HBV core DNA Southern blot, and FLAG-IFI27 Western blot analyses as described above. β-Actin served as the protein loading control. The relative levels of HBV 3.5 kb RNA and ssDNA are presented as percentages of the control groups. The results shown are representative of at least two experimental trials. vector, or pHBV1.3 plus FLAG-IFI27. As anticipated, HBV transfection did not reduce cell viability in the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay due to the noncytopathic nature of HBV replication (40), and the cell viability remained unchanged in cells transfected with both HBV and FLAG-IFI27, suggesting that the observed anti-HBV effect of IFI27 does not involve cytotoxicity (Fig. 3B).
Given that IFI27 inhibits HBV replication primarily by reducing the steady-state level of HBV RNA, this process could be potentially attributed to either a transcriptional or post-transcriptional mechanism. To this end, we first compared the antiviral effect of IFI27 between two HBV replicons, pHBV1.3 and pCMVHBV, in which the transcription of pgRNA is governed by HBV core promoter and human cytomegalovirus immediate-early (CMV-IE) promoter, respectively (41) (Fig. 4A). As shown in Fig. 4B, IFI27 overexpression only resulted in a significant reduction of HBV RNA transcribed from pHBV1.3 but not pCMVHBV, indicating a viral promoter-specific antiviral activity of IFI27 and an absence of IFI27-mediated post-transcriptional HBV RNA destabilization. Next, to evaluate the effect of IFI27 on HBV promoter activities, a cell-based luciferase (Luc) reporter assay was conducted. The results revealed that IFI27 overexpression did not reduce the Renilla luciferase (RL) signal controlled by CMV-IE promoter on plasmid pRL-CMV (Fig. 4C), which is consistent with the pCMVHBV result (Fig. 4B). By using pRL-CMV as a control to normalize transfection efficiency, IFI27 exhibited significant suppression of the activity of HBV promoters tested, including the Enhancer II and Core promoter (EnII/Cp), S1 promoter (S1p), and S2 promoter (S2p) (Fig. 4D). Collectively, the above results suggest that IFI27 reduces HBV RNA transcription by inhibiting viral promoter activity. and the subcellular localization of FLAG-IFI27 was detected by immunofluorescence (green) using anti-FLAG antibodies. The cell nuclei were stained with 4´,6-diamidino-2-phenylindole (DAPI) (blue), and mitochondria were stained with MitoTracker (red). (B) HepG2 cells were transfected with control vector, control vector plus pHBV1.3, or pHBV1.3 plus FLAG-IFI27 for 4 days, followed by cell viability measurement by the MTT assay. The relative cell survival rate was normalized to the control vector group (mean ± SD, n = 3; ns: not significant).
## Assessment of the antiviral effect of endogenous IFI27 on HBV infection
Given that IFI27 has a detectable basal level of expression in hepatocytes, and it can be further induced by IFNα (Fig. 1), it is of interest to assess the role of endogenous IFI27 in IFNα-elicited antiviral response in the context of HBV infection. Hence, we performed IFI27 KD by siRNA in HepG2-NTCP cells. The IFI27 KD cells and control KD cells were then infected by HBV and treated with or without IFNα. As shown in Fig. 5A, IFNα treatment significantly upregulated IFI27 at the mRNA level in control KD cells, but the induction of IFI27 by IFNα was dramatically abolished upon IFI27 knockdown. Next, the outcomes of HBV infection were analyzed. The Northern blot and qPCR analyses demonstrated that IFNα profoundly inhibited HBV infection by reducing viral RNA levels in control KD cells; however, knockdown of IFI27 to 87.2% of its basal level partially, but markedly, abrogated the antiviral activity of IFNα in HBV-infected cells (Fig. 5B andC). The data above suggest that IFI27 functions as a host restriction factor and contributes to IFNα-mediated anti-HBV response, alongside many other antiviral ISGs known to reduce HBV RNA, including ZAP, ISG20, MX2, PML, etc. (20,22,24,25,42).
## IFI27 inhibits HBV transcription without drastically altering the host tran scriptome
HBV transcription relies on both the ubiquitous and liver-enriched TFs, including SP-1, CREB, C/EBPα, HNF4α, HNF1, FXR, etc. (43)(44)(45). To investigate the antiviral mechanism of IFI27 against HBV transcription, we first examined the effect of IFI27 on a list of TFs required for HBV transcription. The results showed no significant changes in the mRNA levels of tested TFs required for HBV transcription after IFI27 overexpression (Fig. 6A).
Next, to identify potential host factors/pathways involved in IFI27-mediated anti-HBV effect, we conducted comparative transcriptomic analysis between HepG2 cells with and without IFI27 overexpression. As depicted in Fig. 6B, IFI27 did not induce a significant change in the host gene transcriptome, and only a few genes showed a fold change greater than 1×log 2 . Among them, the gene with the highest level of upregulation was CXCL8, encoding interleukin-8 (IL-8). To validate the expression of the most upregulated genes in the RNA-seq results, an RT-qPCR analysis was performed (Fig. 6C). As observed, the CXCL8 gene was upregulated by approximately eightfold in HepG2 cells under IFI27 overexpression; other genes of interest, including CXCL2, TSPO, and CCDC85B, also exhibited varying degrees of upregulation. Considering that CXCL8 is the most prominently upregulated gene, it is worth investigating its role in the IFI27-mediated anti-HBV effect. We first tested whether IL-8 could inhibit HBV replication. As shown in Fig. 6D andE, neither intracellular overexpression of IL-8 nor addition of IL-8 to the cell culture supernatant inhibited HBV transcription in pHBV1.3-transfected cells.
It has been reported that the IL-8 receptors CXCR1 and CXCR2 are highly expressed in HepG2, Huh7, and other liver cancer cells (46). Moreover, IL-8 stimulation has been shown to activate a positive feedback loop, increasing the expression of IL-8 itself as well as CXCR1 and CXCR2, and to induce downstream factors such as N-cadherin, E-cadherin, and CD97 (47)(48)(49). To assess the functional responsiveness of HepG2 cells to IL-8, we treated the cells with IL-8 and examined the expression of IL-8, CXCR1, CXCR2, N-cad herin, E-cadherin, and CD97. As shown in Fig. S1, IL-8 treatment resulted in significant upregulation of all these genes, confirming that the IL-8 signaling pathway is active in HepG2 cells. Nonetheless, IL-8 treatment did not affect HBV transcription in HepG2 cells (Fig. 6D andE), supporting a conclusion that IL-8 does not mediate the antiviral effect of IFI27 against HBV.
## IFI27 inhibits HBV replication by promoting a decrease in C/EBPα expression
Given that IFI27 is predominantly localized in the cytoplasm and its overexpression does not cause a significant change in the host transcriptome of HepG2 cells, it is unlikely that IFI27 exerts its anti-HBV transcription effect directly in the nucleus or indirectly through modulating the transcription of TFs required by HBV. Hence, we hypothesized that IFI27 may affect the expression of certain TFs at the protein level. By analyzing the level of several TFs by Western blot, we found that IFI27 overexpression was able to decrease the protein level of C/EBPα. As shown in Fig. 7A, IFI27 overexpression resulted in a reduction of C/EBPα protein in HepG2 cells (lane 2 vs 1), and such effect became more pronounced when new protein synthesis was inhibited by cycloheximide (CHX) (lane 4 vs 3). The upper and lower bands represent the full-length C/EBPα protein of 42 kDa (p42) and the alternative translation product of 30 kDa (p30), respectively, both of which function as transcriptional activators (50). Furthermore, the induction of IFI27 in IFNα-treated HepG2 cells was accompanied by a decrease in C/EBPα, indicating that the IFNα-induced IFI27 is also able to reduce C/EBPα protein (Fig. 7B).
Since IFI27 is predominantly localized in the cytoplasm, we next conducted a cell fractionation experiment to determine the spatial correlation between IFI27 and C/EBPα. As shown in Fig. 7C, IFI27 induced the reduction of C/EBPα primarily in the cytoplasm (top panel, lanes 1 and 2), which consequently led to C/EBPα reduction in the nucleus (top panel, lanes 3 and 4). Interestingly, both the p42 and p30 isoforms displayed a doublet band pattern in the cell fractionation assay, although the underlying reason remains unknown. In addition, while p42 appears to be more abundant than p30 in whole-cell lysate and cytoplasmic fraction, the ratio of these two isoforms becomes reversed in nuclear fraction, indicating differential stability of the two isoforms in the nucleus. While FLAG-IFI27 was detected in the cytoplasmic fraction (bottom panel, lane 2), the detection of FLAG-IFI27 in the nuclear fraction (bottom panel, lane 4) was unexpected as the IF results did not show FLAG-IFI27 within the nucleus (Fig. 3A). We reasoned that the detected nuclear FLAG-IFI27 on Western blot was originally associated with the nuclear membrane at its cytoplasmic side. In this regard, a previous study has found that IFI27 is largely localized to the nuclear membrane in both 293 and HeLa cells (51).
It has been reported that C/EBPα binds to HBV Enhancer II, Core promoter, and S2 promoter, which contribute to the transcription of HBV (52)(53)(54). As observed in Fig. 8A, the chromatin immunoprecipitation (ChIP)-qPCR assay demonstrated that the binding of C/EBPα to the HBV genome was decreased under IFI27 overexpression, whereas the H3K27ac histone modification on the HBV genome, serving as an internal control, did not show any change. Moreover, using a luciferase assay, we confirmed that C/EBPα overexpression activates all four major HBV promoters (Fig. 8B). Consistently, overexpres sion of C/EBPα upregulated HBV mRNA levels in pHBV1.3-transfected HepG2 cells (Fig. 8C), and knockdown of endogenous C/EBPα expression in HepG2 cells downregulated the levels of HBV mRNA derived from the pHBV1.3 template (Fig. 8D), confirming the proviral effect of C/EBPα on HBV transcription level.
To assess the effect of IFI27-mediated C/EBPα downregulation in IFI27's antiviral activity against HBV transcription, we overexpressed C/EBPα in the context of pHBV1.3 and IFI27 co-transfection. As shown in Fig. 8E, IFI27 significantly reduced the HBV RNA level in the absence of C/EBPα (lane 2 vs 1); however, overexpression of C/EBPα significantly abolished the inhibitory effect of IFI27 on HBV RNA (lane 3), indicating that C/EBPα is a critical link between IFI27 and HBV suppression. Collectively, our results demonstrated that IFI27 can induce the reduction of C/EBPα protein and reduce its binding on the HBV genome to prevent it from activating HBV transcription.
## IFI27 recruits SKP2 as an E3 ubiquitin ligase to target C/EBPα for proteasomal degradation
Next, we investigated the mechanism underlying the IFI27-mediated reduction of C/EBPα. By using the proteasome inhibitor MG132, we observed that the IFI27-mediated reduction of C/EBPα was abrogated, indicating that IFI27 promotes C/EBPα degradation via the proteasomal pathway (Fig. 9A). Then, we asked whether IFI27 could induce the ubiquitination of C/EBPα. We conducted a co-immunoprecipitation (co-IP) assay by pulling down Myc-C/EBPα (Fig. 9B). The results revealed an increase in ubiquitination of Myc-C/EBPα when IFI27 was overexpressed (lane 2 vs 1). Moreover, when MG132 was added, the ubiquitination level of Myc-C/EBPα exhibited no difference between the control group and the IFI27 overexpression group (lane 4 vs 3), indicating that the inhibition of proteasome prevented the degradation of ubiquitinated Myc-C/EBPα. These results suggest that IFI27 enhances the ubiquitination of C/EBPα, leading to its degradation through the ubiquitin-proteasome pathway. It is worth noting that the co-IP assay did not reveal an interaction between IFI27 and C/EBPα. Ubiquitination specifies a stepwise enzymatic cascade that conjugates ubiquitin to target proteins covalently, and the final step requires direct interaction between an E3 ubiquitin ligase and its specific substrate (55). Although IFI27 does not directly interact with C/EBPα, the observed ubiquitination of C/EBPα induced by IFI27 indicates the involvement of an E3 ubiquitin ligase, which is likely recruited by IFI27 to facilitate C/EBPα ubiquitination. In this regard, our literature search identified SKP2, a known E3 ubiquitin ligase, as a potential candidate involved in IFI27-mediated C/EBPα degradation. A previous study demonstrated that IFI27 restricted HCV infection by inducing the ubiquitination and degradation of the viral protein NS5A through promoting SKP2-NS5A interaction (29). Other studies reported that cyclin-dependent kinase 2 triggers the degradation of C/EBPα through SKP2 in acute myeloid leukemia (AML) (56), and that SKP2 ubiquitinates C/EBPα through physically interacting with the C-terminal portion of C/EBPα in AML (57). Therefore, it is of interest to assess the potential role of SKP2 in IFI27-mediated HBV RNA reduction by comparing the inhibitory effect of IFI27 on C/EBPα expression between SKP2 KD and control KD groups. As shown in Fig. 9C, the overexpres sion of IFI27 consistently resulted in a decrease in the level of C/EBPα; however, when SKP2 was knocked down, the reduction effect was abolished, indicating that SKP2 is responsible for the IFI27-mediated C/EBPα degradation. We next assessed the antiviral activity of IFI27 in HBV-transfected HepG2 cells with and without knockdown of SKP2. As shown in Fig. 9D, depleting SKP2 partially abrogated the IFI27-mediated reduction of HBV RNA. Collectively, the above results revealed that IFI27 degrades C/EBPα through the ubiquitin-proteasome pathway, a process facilitated by the E3 ubiquitin ligase SKP2, thereby hindering C/EBPα binding to the HBV promoter and its activation of HBV transcription.
## DISCUSSION
The innate immune response serves as the first line of defense against numerous kinds of viral invaders, and IFNs play an essential role in the restriction of virus replication and propagation. More than 300 ISGs can be upregulated by interferons, functioning as innate immune effectors in antiviral responses, while some ISGs may also act as proviral factors for certain viruses (16,58). In terms of HBV infection, it is largely acknowledged that HBV encodes mechanisms to evade innate surveillance, at least in hepatocytes (59,60). Nonetheless, IFNα-based therapies have been used to treat chronic HBV infection for more than 3 decades, and a better understanding of ISGs' functions will help identify some essential points that may improve the outcomes of IFNα therapy and may even lead to the development of novel therapeutic approaches (20). In search of ISGs that inhibit HBV replication, we have previously obtained preliminary results showing that IFI27 overexpression inhibited HBV transcription and replication in pHBV1.3-transfected HepG2 cells (26). In the present study, we focused on validating the antiviral effect of IFI27 on HBV transcription and elucidating the underlying antiviral mechanism. We first demonstrated the high inducibility of IFI27 in human hepatocyte cells, including both hepatoma cell lines and PHHs, upon IFNα treatment, as well as the nonapoptotic antiviral activity of IFI27 (Fig. 1 to 3). Furthermore, IFI27 inhibits HBV transcription primarily by downregulating viral promoter activities (Fig. 2, 4, and5). This effect is at least partially attributed to a mechanism in which IFI27 promotes the degradation of a cellular transcription factor required by HBV, specifically C/EBPα, through the ubiquitin-protea some degradation pathway (Fig. 6 to 8). Lastly, an E3 ubiquitin ligase SKP2 is responsible for the IFI27-mediated C/EBPα degradation (Fig. 9). Our study thus provides further evidence suggesting that IFI27 is an antiviral ISG against HBV infection, and a model elucidating the IFI27-mediated inhibition of HBV infection by IFNα is proposed for the following discussion (Fig. 10).
Previous studies have reported that IFI27 exerts a broad-spectrum antiviral activity against multiple viruses in vitro and/or in vivo, including Sindbis virus, HBV, HCV, West Nile virus, Newcastle disease virus, and human immunodeficiency virus 1 (HIV-1), albeit through different mechanisms (26,(29)(30)(31)(32)61). In this study, we further investigated the antiviral effect and mechanism of IFI27 against HBV by utilizing various in vitro HBV replication systems, including transfection and infection models in multiple human hepatocyte-derived cell types. It has been previously reported that IFI27 is localized to mitochondria and possesses the pro-apoptotic activity to potentiate cell apoptosis by modulating mitochondrial membrane potential and releasing apoptosis-inducing factors in non-hepatic HT1080 and HEK293 cell lines (38,39). In our study, we observed that IFI27 is primarily expressed in the cytoplasm and partially co-localized with mitochondria (Fig. 3). However, we did not observe cell viability change upon IFI27 overexpression in HBV-transfected HepG2 cells, inferring a nonapoptotic antiviral activity of IFI27 in the context of HBV replication, although HBV has been reported to possess both pro-and anti-apoptotic activities (62)(63)(64). Nonetheless, in our study, IFI27 was demonstrated to be a novel anti-HBV ISG that inhibits HBV transcription via suppressing viral promoters' activities (Fig. 4).
HBV cccDNA serves as the bona fide transcription template for all viral mRNAs, and it resides in host cell nucleus as a stable episome, which is organized into a minichro mosome decorated by histones and various viral and cellular non-histone proteins (10). Consistent with the results obtained from HBV and IFI27 transfection systems, the antiviral effect of IFNα-induced endogenous IFI27 on cccDNA transcription was also confirmed in HBV infection system (Fig. 5). The transcription of cccDNA minichromo some is substantially regulated by epigenetic modifications of cccDNA-bound histones, and the HBV-encoded accessory protein HBx plays an indispensable role in support ing cccDNA transcription by maintaining a transcriptionally active epigenetic state of cccDNA (10,(65)(66)(67)(68)(69). Additionally, HBc has also been suggested to interact with the cccDNA, although whether it regulates cccDNA transcription remains debatable (65,(70)(71)(72)(73)(74)(75)(76)(77)(78)(79). In light of the above information, we tested the possibility that IFI27 may poten tially interact with viral factors and consequently influence HBV transcriptional activity. However, IFI27 remained active in inhibiting viral transcription in cells transfected with HBV with HBx-null or HBc-null mutants (data not shown), ruling out the involvement of HBx or HBc in the IFI27-mediated inhibition of HBV transcription. Since IFI27 is primarily localized in the cytoplasm (38,39) (Fig. 3), and a recent study suggests that it can regulate the AKT-β-catenin-PD-L1 nexus in NK cells (80), we thus speculated that IFI27 may alter certain signaling pathway(s) through initially interacting with the pathwayrelated factor(s) in the cytoplasm, which leads to a nuclear event that acts on HBV transcription. However, transcriptomic analyses revealed that IFI27 did not significantly alter the host transcriptome in HepG2 cells, including pathways and factors known to regulate HBV transcription (Fig. 6A andB). Moreover, the most highly induced factor, IL-8, showed no antiviral activity against HBV transcription in subsequent functional validation assays (Fig. 6C through E).
Although IFI27 overexpression does not alter the mRNA levels of a list of representa tive transcription factors known to regulate HBV transcription (Fig. 6A), previous studies demonstrated that IFI27 can directly bind to and promote the degradation of certain viral proteins (29,61). Inspired by these findings, we hypothesized that IFI27 might impact the protein stability of certain transcription factor(s) required by HBV rather than inhibiting their mRNA expression. Indeed, we found that overexpression of IFI27 resulted in a reduction of C/EBPα at the protein level, and that IFNα treatment also led to a reduction of C/EBPα protein accompanied by an induction of IFI27 (Fig. 7A andB), inferring that C/EBPα might be targeted by IFI27 for inhibition of HBV transcription. Moreover, we observed that IFI27 overexpression mainly reduces C/EBPα protein in the cytoplasm, resulting in a decrease in the total amount of C/EBPα and its binding with the HBV genome in the nucleus (Fig. 7C and8A). It has been reported that the p42/p30 ratio of C/EBPα plays a critical role in transcriptional regulation of specific target genes, such as DDIT3 in AML cells, with p30 serving as the primary isoform driving DDIT3 transcription (81). However, since IFI27 reduced the levels of both p42 and p30, it remains unclear which isoform is predominantly responsible for modulating HBV transcription, and this question warrants further investigation.
C/EBPα is the first member of a family of six transcription factors: C/EBPα, -β, -γ, -δ, -ε, and -ζ (82,83). C/EBPs play a crucial role in promoting gene expression by interacting with gene promoters. Once they are bound to the DNA, they can recruit co-activators, which subsequently modify chromatin structure to an open configuration and recruit basal transcription factors (84). It has been demonstrated that the binding of C/EBPα to HBV promoters and EnII plays a crucial role in the transcriptional activation of HBV and promotes its replication (54,(85)(86)(87). In line with previous findings, we also observed that C/EBPα overexpression markedly enhanced the activity of HBV promoters and viral pgRNA level, and knockdown of C/EBPα reduced HBV transcription (Fig. 8B through D). Furthermore, transcomplementation of C/EBPα expression abolished IFI27-mediated inhibition of HBV transcription (Fig. 8E), supporting the conclusion that IFI27 inhibits HBV transcription through downregulating C/EBPα. In the process of exploring the mechanisms of how IFI27 downregulates C/EBPα expression, we found that the inhibitory effect of IFI27 on C/EBPα expression could be almost completely reversed by proteasome inhibitor MG132 treatment, indicating that IFI27 promotes the proteasomal degradation of C/EBPα (Fig. 9A). Moreover, we demonstrated that IFI27 promotes the ubiquitination of C/EBPα, thereby directing it toward proteasomal degradation (Fig. 9B). Previous studies showed that IFI27 could interact with HIV Gag and HCV NS5A to promote protein degradation (29,61); however, IFI27 did not appear to bind to C/EBPα in the co-IP assay (Fig. 9B), which might be due to a possible transient and dynamic interaction between C/EBPα and IFI27 (a small 12 kDa hydrophobic protein). The ubiquitin-dependent proteasomal protein degrada tion mechanism involves the coordinated action of E1 ubiquitin-activating enzymes, E2 ubiquitin-conjugating enzymes, and E3 ubiquitin-protein ligases (88,89). Among these, E3 ubiquitin ligases play a crucial final role by determining substrate specificity and marking target proteins for proteasomal degradation. A previous study has reported that IFI27 recruits E3 ligase SKP2 to degrade HCV NS5A protein (29). In our study, SKP2 knockdown significantly abrogated the inhibitory effects of IFI27 on C/EBPα expression and HBV transcription (Fig. 9C andD), confirming that the antiviral function of IFI27 requires the activity of E3 ligase SKP2. It is worth noting that both HBV and HCV are hepatotropic viruses, and the IFI27-SKP2-mediated protein degradation mechanism may be utilized by host cells to inhibit both, particularly during HBV/HCV co-infection, as HCV is capable of inducing IFNα and IFI27 (59). Notably, while C/EBPα expression was almost fully restored by SKP2 knockdown under IFI27 overexpression, the IFI27-medi ated suppression of HBV transcription was only partially reversed (Fig. 9C andD). The underlying reason remains unclear but may be attributed to incomplete SKP2 depletion or the presence of additional SKP2-independent antiviral mechanisms mediated by IFI27. Regarding the latter possibility, a recent study reported that IFI27 binds to the Cp promoter region (nt 1715-1815) of the transfected pHBV1.3 plasmid in HepG2 cells; however, the presented ChIP-PCR data did not rigorously confirm a sequence-specific binding (90). Moreover, IFI27 is predominantly localized in the cytoplasm of HepG2 cells (Fig. 3), suggesting that it is unlikely to directly inhibit HBV transcription within the nucleus. To further elucidate the antiviral mechanisms of IFI27, a proteomic analysis of the IFI27 interactome is warranted and is currently underway in our lab.
It is anticipated that the degradation of C/EBPα by IFI27 might have broader effects on the cellular transcriptome. To further explore the downstream transcrip tional programs modulated by IFI27-mediated C/EBPα degradation, we performed gene set enrichment analysis (GSEA) of RNA-seq data from IFI27-overexpressing cells versus controls, using the Molecular Signatures Database (MSigDB). The GSEA results revealed that pathways downstream of C/EBPα, including PPAR signaling, fatty acid metabolism, biosynthesis of unsaturated fatty acids, cholesterol metabolism, comple ment and coagulation cascades, cholesterol biosynthesis (Medicus reference), and steroid biosynthesis (91)(92)(93)(94), were downregulated under IFI27 overexpression (Fig. S2). Interestingly, the TAVOR_CEBPA_TARGETS_DN set, which represents genes normally downregulated by C/EBPα, was upregulated by IFI27 (Fig. S2). Given the well-estab lished roles of C/EBPα and PPAR in regulating adipocyte differentiation and metabolic homeostasis (91,95), these bioinformatic data indicate that IFI27 suppresses C/EBPα expression and its downstream transcriptional network, particularly pathways related to hepatic and lipid metabolism. However, because most of these pathway-associated differentially expressed genes (DEGs) showed <2-fold changes, they were not highligh ted in our initial transcriptomic analysis (Fig. 6B). The relatively mild impact of C/EBPα loss on cellular target gene transcription, compared with its pronounced effect on HBV transcription, may be due to a higher demand of C/EBPα by viral genome to establish infection, at least in a transient period. The broader biological consequences of IFI27-C/EBPα-mediated cellular transcriptomic reprogramming on host functions and HBV infection remain to be determined and await further investigation.
Taken together, the phenotypic and mechanistic characterizations of IFI27-mediated antiviral effect on HBV replication presented in the current study not only shed light on IFI27 biology and virus-host interaction, but also provide insight into the development of novel antiviral strategies. Furthermore, emerging evidence suggests broader roles for IFI27 in inflammatory diseases, autoimmune disorders, and cancers (96). Therefore, it will be of interest to explore the potential involvement of IFI27 in the progression of HBV-associated chronic hepatitis, cirrhosis, and hepatocellular carcinoma in future studies.
## MATERIALS AND METHODS
## Cell cultures
HepG2 and Huh7 cells were maintained in Dulbecco's modified Eagle's medium/F-12 with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin supplemented. The HepG2-NTCP cell line supporting HBV infection was maintained in the same way as HepG2 and Huh7 cells plus additional 8 µg/mL blasticidin (97). Freshly isolated PHH cells were obtained from the Human Liver Tissue and Hepatocyte Research Resource (funded by NIDDK project no. R24DK139775) at The Pittsburgh Liver Research Center (funded by NIDDK grant no. P30DK120531), University of Pittsburgh, and cultured as previously described (98).
## Plasmids, siRNA, cytokines, and compounds
The HBV (genotype D) replication-competent plasmid pHBV1.3 containing a 1.3-mer replicon of viral genome, and pCMVHBV, transcribing HBV pgRNA under the control of a human CMV-IE promoter, were described previously (26,41,99). The HBV promoter firefly Luc reporter plasmids, including the EnII/Cp-Luc, S1p-Luc, S2p-Luc, and Xp-Luc, have been described in our previous publications (25,100). The CMV-IE promoter Renilla luciferase reporter plasmid pRL-CMV was purchased from Promega (Cat# E2261). Plasmid FLAG-IFI27 expresses the N-terminally FLAG-tagged IFI27 (26,101). The human IL-8 expression plasmid (Cat# RC202075) and Myc-tagged C/EBPα expression plasmid (Cat# RC218955) were purchased from OriGene. The transfection of plasmid DNA into cells was conducted by using Lipofectamine 3000 (Cat# L3000150, Thermo Fisher) according to the manufacturer's manual.
The siRNA for knocking down IFI27 (Cat# sc-105551), C/EBPα (Cat# sc-37047), SKP2 p45 (Cat# sc-74477), and control siRNA-A (Cat# sc-37007) was purchased from Santa Cruz Biotechnology. Lipofectamine RNAiMAX (Cat# 13778100, Thermo Fisher) was used to transfect siRNA into cell cultures.
The recombinant human IL-8/CXCL8 (Cat# 208-IL) and IFNα-2a (Cat# SRP4594) were purchased from R&D Systems and Sigma-Aldrich, respectively. The compound MTT (Cat# M2003), proteasome inhibitor MG132 (Cat# 474790), and protein synthesis inhibitor cycloheximide (Cat# C7698) were purchased from Sigma-Aldrich and dissolved in DMSO to prepare stock solutions per manufacturer's recommendations.
## Cytotoxicity assay
Cells were transfected with indicated plasmids for 4 days, then MTT compound was added to the culture supernatant at a final concentration of 0.5 mg/mL, and the plate was incubated at 37°C for 4 h. Next, the supernatant was gently removed by vacuum aspiration, and 150 µL DMSO was added to dissolve the reaction product at 37°C for 30 min. The colorimetric absorbance was measured at 570 nm with BioTek Synergy HTX multimode plate reader. The relative cell survival rate was determined by normalizing the OD 570 values to control groups.
## HBV infection
The infectious HBV particles were harvested from the supernatant of induced HepAD38 cells, and the quantification of virion genome equivalents was performed by following the previously established methods (97). HBV in vitro infection of HepG2-NTCP or PHH cells was conducted according to our previous publications (36,98,102).
## IFA
HepG2 cells were transfected with control vector or FLAG-IFI27 for 3 days. Cells were incubated with MitoTracker Deep Red FM (Cat# M22426, Invitrogen) for 20 min according to manufacturer's directions, followed by fixation with 4% paraformaldehyde for 20 min and permeabilization with 0.5% Triton X-100 in phosphate-buffered saline (PBS) for 1 h at room temperature. Cells were then blocked with IFA blocking buffer (10% FBS plus 2% bovine serum albumin in 1× PBS) for 1 h at room temperature and incuba ted with anti-FLAG M2 antibody (Cat# F3165, Sigma-Aldrich) at 4°C overnight. After washing with 1× PBS and a second blocking for 30 min, cells were stained with Alexa Fluor 488 dye-conjugated goat anti-mouse secondary antibody (Cat# A1101, Invitrogen), and the nuclei were counterstained with 4´,6-diamidino-2-phenylindole for 30 min at room temperature. Both the primary and the secondary antibodies were diluted in IFA blocking buffer. The cells were washed with 1× PBS and subjected to Olympus FV1000 MPE confocal microscopy analysis with the 60× or 20× objective. Images were analyzed using FV10-ASW 3.0 Viewer and ImageJ software.
## Cellular mRNA qPCR
Total cellular RNA was extracted by TRI Reagent (Cat# AM9738, Thermo Fisher) and subjected to reverse transcription to generate cDNA using SuperScript III reverse transcriptase (Cat# 18080093, Thermo Fisher). Real-time qPCR of cDNA was then performed using gene-specific primers of the indicated transcription factors (Table S1) and SYBR Green master mix (Cat# 04707516001, Roche) on the Roche LightCycler 96 system. Primers for E-cadherin (Cat# HP207683) and N-cadherin (Cat# HP205580) were purchased from OriGene. The relative mRNA expression levels of detected genes were normalized to the mRNA levels of β-actin from the same samples.
## HBV DNA and RNA analysis
Intracellular HBV RNA and core (capsid-associated) DNA were extracted and subjected to Northern blot and Southern blot, respectively, as described previously (25,103). Hybridization signals were exposed to a phosphorimager screen and scanned using the Typhoon FLA-7000 imager (GE Healthcare). HBV pC mRNA-specific qPCR was conducted according to our previous publications (36,65). HBV pC mRNA qPCR was normalized to cellular GAPDH mRNA qPCR as previously described (65).
## ChIP-qPCR
To assess the enrichment of C/EBPα on HBV genome, ChIP assay was performed using the ChIP-IT Express Chromatin Immunoprecipitation Kit (Cat# 53008, Active Motif ) as previously described (36,104). Briefly, after the prepared chromatin-containing cell lysate was divided into input and ChIP samples, the latter was subjected to immunoprecipita tion with protein G magnetic beads coated with nonimmune serum control IgG antibody (Cat# I8765, Sigma-Aldrich) or ChIP-grade specific antibodies, specifically anti-H3K27ac (Cat# ab4729, Abcam) and anti-C/EBPα (Cat# 8178, Cell Signaling). The ChIPed DNA was cleaned up by QIAquick PCR purification kit (Cat# 28106, Qiagen) and subjected to qPCR with HBV-specific primers as previously described (36,104). The occupancy of the protein of interest on the HBV genome was calculated and expressed as the percentage of input (% input).
## Western blot assay
Cells were rinsed with cold 1× PBS and lysed in 1× Laemmli buffer. Then, cell lysates were subjected to sonication at 20% amplitude with 10 s impulse and 2 s rest for six cycles using EpiShear Probe sonicator (Active Motif ), followed by resolution in either Novex 16% Tricine Gel (Cat# EC66952, Thermo Fisher) or Novex 12% Tris-Glycine Gel (Cat# XP00122, Thermo Fisher), depending on the size of the target protein. The proteins were transferred onto the Amersham Protran 0.45 NC nitrocellulose membrane (Cat# 160002, Cytiva). The membrane was blocked with WesternBreeze blocking buffer (Cat# WB7050, Thermo Fisher) and then probed with primary antibodies against the indicated proteins or epitope tags, including β-Actin (Cat# sc-47778, Santa Cruz), GAPDH (Cat# sc-47724, Santa Cruz), SKP2 p45 (Cat# sc-74477, Santa Cruz), Lamin A/C (Cat# sc-376248, Santa Cruz), ubiquitin (Cat# sc-8017, Santa Cruz), C/EBPα (Cat# 8178S, Cell Signaling), FLAG-tag M2 (Cat# F1804, Sigma-Aldrich), Myc-tag (Cat# 2278, Cell Signaling), and IFI27 (51). After the membrane was washed three times with 1× PBS, it was incubated with WesternSure Goat anti-mouse horseradish peroxidase (HRP) (Cat# 926-80010) or anti-Rabbit HRP (Cat# 926-80011) secondary antibody (Li-COR). Detection of the immunoblot signal was performed using enhanced chemiluminescence (ECL) with SuperSignal West Pico PLUS Luminol/Enhancer (Cat# 34580, Thermo Fisher) and visualized using the Li-COR Odyssey system.
## Coimmunoprecipitation assay
HepG2 cells in six-well plates were transfected with the indicated plasmids for 48 h. Cells were then washed with 1× PBS and lysed in 400 µL lysis buffer (1% NP-40, 10% glycerol, 2 mM EDTA, 50 mM Tris-HCl [pH 7.0], and 150 mM NaCl) supplemented with 1× Halt protease inhibitor cocktail (Cat# 87785, Thermo Fisher) and 100 µM phenylmethyl sulfonyl fluoride (PMSF) (Cat# 52332, Sigma-Aldrich) per well. The cell lysates were gently rotated at 4°C for 1 h, followed by centrifugation at 12,000 rpm in 4°C for 10 min. A total of 2.5% of the supernatant was saved as input for Western blot detection. The remaining supernatant was incubated with anti-C/EBPα antibody (Cat# 8178, Cell Signaling) by rotation at 4°C for 1 h. Then, 20 µL of protein A/G beads (Cat# sc-2003, Santa Cruz) were rinsed with the cell lysis buffer and added to the immunoprecipitation reaction. The mixture was rotated at 4°C overnight, and the beads were precipitated by spin-down and washed three times with cell lysis buffer. The supernatant was discarded, and the immunoprecipitated pellet was mixed with 1× Laemmli buffer and resolved by SDS-PAGE for Western blot assay as described above.
## Differential transcriptome analysis
Two groups of differential transcriptome analyses were performed in this study. One group was HepG2-NTCP cells with and without IFNα (1,000 IU/mL) treatment; the other group was HepG2 cells with and without FLAG-IFI27 overexpression. Total RNA was extracted from the cell samples and submitted in triplicate for mRNA sequencing at The Center for Medical Genomics at Indiana University School of Medicine or The Health Sciences Sequencing Core at UPMC Children's Hospital of Pittsburgh. The differential transcriptome analysis was performed following the protocols and bioinformatic pipeline described in our previous study (36).
## GSEA
RNA-seq data from IFI27-overexpressing and control cells were analyzed by GSEA (v4.4.0; Broad Institute, Cambridge, MA) using the C2 curated gene sets (C2-all) collection from the MSigDB (105)(106)(107). Genes were ranked according to differential expression, and enrichment statistics, including normalized enrichment score and false discovery rate, were calculated using default parameters.
## Statistical analysis
Data were analyzed using GraphPad Prism 10.0 and expressed as mean ± standard deviation. Student's t-test was used to determine statistical significance (P-value <0.05).
## References
1. Revill, Chisari, Block et al. (2019) "A global scientific strategy to cure hepatitis B"
2. Alter, Block, Brown et al. (2018) "A research agenda for curing chronic hepatitis B virus infection" *Hepatology*
3. Dusheiko, Agarwal, Maini (2023) "New approaches to chronic hepatitis B" *N Engl J Med*
4. Seeger, Mason (2000) "Hepatitis B virus biology" *Microbiol Mol Biol Rev*
5. Block, Guo, Guo (2007) "Molecular virology of hepatitis B virus for clinicians" *Clin Liver Dis*
6. Yan, Zhong, Xu et al. (2012) "Sodium taurocholate cotransporting polypeptide is a functional receptor for human hepatitis B and D virus" *Elife*
7. Seeger, Mason (2015) "Molecular biology of hepatitis B virus infection" *Virology (Auckl)*
8. Marchetti, Guo (2020) "New insights on molecular mechanism of hepatitis b virus covalently closed circular DNA formation" *Cells*
9. Xia, Guo (2020) "Hepatitis B virus cccDNA: formation, regulation and therapeutic potential" *Antiviral Res*
10. Hong, Kim, Guo (2017) "Epigenetic regulation of hepatitis B virus covalently closed circular DNA: implications for epigenetic therapy against chronic hepatitis B" *Hepatology*
12. Nassal (2008) "Hepatitis B viruses: reverse transcription a different way" *Virus Res*
13. Hu, Seeger (2015) "Hepadnavirus genome replication and persis tence" *Cold Spring Harb Perspect Med*
14. Jiang, Hildt (2020) "Intracellular trafficking of HBV particles" *Cells*
15. De Andrea, Ravera, Gioia et al. (2002) "The interferon system: an overview" *Eur J Paediatr Neurol*
16. Zoulim, Durantel (2015) "Antiviral therapies and prospects for a cure of chronic hepatitis B"
17. Sadler, Williams (2008) "Interferon-inducible antiviral effectors" *Nat Rev Immunol*
18. Der, Zhou, Williams et al. (1998) "Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays" *Proc Natl Acad Sci*
19. Robek, Boyd, Wieland et al. (2004) "Signal transduction pathways that inhibit hepatitis B virus replication" *Proc Natl Acad Sci*
20. Uprichard, Wieland, Althage et al. (2003) "Transcriptional and posttranscriptional control of hepatitis B virus gene expression" *Proc Natl Acad Sci*
21. Zhao, Liu, Tang et al. (2024) "Mechanism of interferon alpha therapy for chronic hepatitis B and potential approaches to improve its therapeutic efficacy" *Antiviral Res*
22. Lucifora, Xia, Reisinger et al. (2014) "Specific and nonhepatotoxic degradation of nuclear hepatitis B virus cccDNA" *Science*
23. Cheng, Zhao, Zhou et al. (2020) "Interferon alpha induces multiple cellular proteins that coordinately suppress hepadnaviral covalently closed circular DNA transcription" *J Virol*
24. Tan, Yi, Song et al. (2019) "Type-I-IFN-stimulated gene TRIM5γ inhibits HBV replication by promoting HBx degradation" *Cell Rep*
25. Liu, Nie, Mao et al. (2017) "Interferon-inducible ribonuclease ISG20 inhibits hepatitis B virus replication through directly binding to the epsilon stem-loop structure of viral RNA" *PLoS Pathog*
26. Mao, Nie, Cai et al. (2013) "Inhibition of Hepatitis B Virus Replication by the Host Zinc Finger Antiviral Protein" *PLoS Pathog*
27. Mao, Zhang, Jiang et al. (2011) "Indoleamine 2,3-dioxygenase mediates the antiviral effect of gamma interferon against hepatitis B virus in human hepatocytederived cells" *J Virol*
28. Yan, Zhao, Cai et al. (2015) "The interferon-inducible protein tetherin inhibits hepatitis B virus virion secretion" *J Virol*
29. Cheriyath, Leaman, Borden (2011) "Emerging roles of FAM14 family members (G1P3/ISG 6-16 and ISG12/IFI27) in innate immunity and cancer" *J Interferon Cytokine Res*
30. Xue, Yang, Wang et al. (2016) "ISG12a restricts hepatitis C virus infection through the ubiquitination-dependent degradation pathway" *J Virol*
31. (2025) *Full-Length Text Journal of Virology*
32. Yang, Meng, Xue et al. (2014) "MiR-942 mediates hepatitis C virus-induced apoptosis via regulation of ISG12a" *PLoS One*
33. Liu, Long, Liu et al. (2014) "ISG12a mediates cell response to Newcastle disease viral infection" *Virology (Auckl)*
34. Lucas, Richner, Diamond (2015) "The interferon-stimulated gene Ifi27l2a restricts west Nile virus infection and pathogenesis in a cell-type-and region-specific manner" *J Virol*
35. Gao, Zhu, Wu et al. (2021) "IFI27 may predict and evaluate the severity of respiratory syncytial virus infection in preterm infants" *Hereditas*
36. Shojaei, Shamshirian, Monkman et al. (2022) "IFI27 transcrip tion is an early predictor for COVID-19 outcomes, a multi-cohort observational study" *Front Immunol*
37. Mao, Dong, Shen et al. (2021) "RNA helicase DDX17 inhibits hepatitis B virus replication by blocking viral pregenomic RNA encapsidation"
38. Yu, Long, Shen et al. (2023) "Screening of an epigenetic compound library identifies BRD4 as a potential antiviral target for hepatitis B virus covalently closed circular DNA transcription" *Antiviral Res*
39. Marchetti, Zhang, Kim et al. (2022) "Proteomic analysis of nuclear hepatitis B virus relaxed circular DNAassociated proteins identifies UV-damaged DNA binding protein as a host factor involved in covalently closed circular DNA formation" *J Virol*
40. Rosebeck, Leaman (2008) "Mitochondrial localization and proapoptotic effects of the interferon-inducible protein ISG12a" *Apoptosis*
41. Gytz, Hansen, Skovbjerg et al. (2017) "Apoptotic properties of the type 1 interferon induced family of human mitochondrial membrane ISG12 proteins" *Biol Cell*
42. Chisari, Isogawa, Wieland (2009) "Pathogenesis of hepatitis B virus infection" *Pathol Biol*
43. Guo, Zhou, Jiang et al. (2007) "Regulation of hepatitis B virus replication by the phosphatidylinositol 3-Kinase-Akt signal transduction pathway" *J Virol*
44. Wang, Niklasch, Liu et al. (2020) "Interferon-inducible MX2 is a host restriction factor of hepatitis B virus replication" *J Hepatol*
45. Oropeza, Tarnow, Sridhar et al. (2020) "The Regulation of HBV Transcription and Replication" *Adv Exp Med Biol*
46. Moolla, Kew, Arbuthnot (2002) "Regulatory elements of hepatitis B virus transcription" *J Viral Hepat*
47. Mitra, Thapa, Guo et al. (2018) "Host functions used by hepatitis B virus to complete its life cycle: Implications for developing host-targeting agents to treat chronic hepatitis B" *Antiviral Res*
48. Bi, Zhang, Wang et al. (2019) "Interleukin-8 promotes cell migration via CXCR1 and CXCR2 in liver cancer" *Oncol Lett*
49. Meng, Zhang, Cai et al. (2023) "IL-8 is a novel prometastatic chemokine in intrahepatic cholangiocarcinoma that induces CXCR2-PI3K/AKT signaling upon CD97 activation" *Sci Rep*
50. Meier, Brieger (2025) "The role of IL-8 in cancer development and its impact on immunotherapy resistance" *Eur J Cancer*
51. Wen, Zhao, Huang et al. (2020) "IL-8 promotes cell migration through regulating EMT by activating the Wnt/βcatenin pathway in ovarian cancer" *J Cellular Molecular Medi*
52. Lin, Macdougald, Diehl et al. (1993) "A 30-kDa alternative translation product of the CCAAT/enhancer binding protein alpha message: transcriptional activator lacking antimitotic activity" *Proc Natl Acad Sci*
53. Martensen, Søgaard, Gjermandsen et al. (2001) "The interferon alpha induced protein ISG12 is localized to the nuclear membrane" *Eur J Biochem*
54. Quasdorff, Protzer (2010) "Control of hepatitis B virus at the level of transcription" *J Viral Hepat*
55. Unger, Shaul (1990) "The X protein of the hepatitis B virus acts as a transcription factor when targeted to its responsive element" *EMBO J*
56. López-Cabrera, Letovsky, Hu et al. (1991) "Transcriptional factor C/EBP binds to and transactivates the enhancer element II of the hepatitis B virus" *Virology (Auckl)*
57. Pickart (2001) "Mechanisms underlying ubiquitination" *Annu Rev Biochem*
58. Thacker, Mishra, Sharma et al. (2021) "CDK2-instigates C/EBPα degradation through SKP2 in Acute myeloid leukemia" *Med Oncol*
59. Thacker, Mishra, Sharma et al. (2020) "E3 ligase SCF SKP2 ubiquitinates and degrades tumor suppressor C/EBPα in acute myeloid leukemia" *Life Sci*
60. Schoggins, Rice (2011) "Interferon-stimulated genes and their antiviral effector functions" *Curr Opin Virol*
62. Wieland, Chisari (2005) "Stealth and cunning: hepatitis B and hepatitis C viruses" *J Virol*
63. Cheng, Xia, Serti et al. (2017) "Hepatitis B virus evades innate immunity of hepatocytes but activates cytokine production by macrophages" *Hepatology*
64. He, Pang, Li et al. (2024) "IFI27 inhibits HIV-1 replication by degrading Gag protein through the ubiquitin-proteasome pathway" *J Virol*
65. Assrir, Soussan, Kremsdorf et al. (2010) "Role of the hepatitis B virus proteins in pro-and anti-apoptotic processes" *Front Biosci (Landmark Ed)*
66. Lin, Zhang (2017) "Interference of apoptosis by hepatitis B virus" *Viruses*
67. Li, Ou (2023) "Regulation of mitochondrial metabolism by hepatitis B virus" *Viruses*
68. Kim, Zhou, Zhang et al. (2022) "Hepatitis B virus X protein counteracts high mobility group box 1 protein-mediated epigenetic silencing of covalently closed circular DNA" *PLoS Pathog*
69. Tropberger, Mercier, Robinson et al. (2015) "Mapping of histone modifications in episomal HBV cccDNA uncovers an unusual chromatin organization amenable to epigenetic manipulation" *Proc Natl Acad Sci*
70. Belloni, Pollicino, Nicola et al. (2009) "Nuclear HBx binds the HBV minichromo some and modifies the epigenetic regulation of cccDNA function" *Proc Natl Acad Sci*
71. Rivière, Gerossier, Ducroux et al. (2015) "HBx relieves chromatin-mediated transcriptional repression of hepatitis B viral cccDNA involving SETDB1 Full-Length Text Journal of Virology November"
72. *J Hepatol*
73. Dandri (2020) "Epigenetic modulation in chronic hepatitis B virus infection" *Semin Immunopathol*
74. Guo, Li, Zhao et al. (2011) "HBc binds to the CpG islands of HBV cccDNA and promotes an epigenetic permissive state" *Epigenetics*
75. Chong, Cheng, Tsoi et al. (2017) "Role of hepatitis B core protein in HBV transcription and recruitment of histone acetyltransferases to cccDNA minichromosome" *Antiviral Res*
76. Zhang, Mao, Guo et al. (2017) "Detection of HBV cccDNA methylation from clinical samples by bisulfite sequencing and methylation-specific PCR" *Methods Mol Biol*
77. Qi, Gao, Xu et al. (2016) "DNA polymerase κ is a key cellular factor for the formation of covalently closed circular DNA of hepatitis B virus" *PLoS Pathog*
78. Tu, Zehnder, Qu et al. (2021) "De novo synthesis of hepatitis B virus nucleocapsids is dispensable for the maintenance and transcrip tional regulation of cccDNA" *JHEP Reports*
79. Lucifora, Pastor, Charles et al. (2021) "Evidence for long-term association of virion-delivered HBV core protein with cccDNA independently of viral protein production" *JHEP Reports*
80. Zhong, Wu, Xu et al. (2022) "Hepatitis B virus core protein is not required for covalently closed circular DNA transcrip tional regulation" *J Virol*
81. Klumpp, Shimada, Allweiss et al. (2018) "Efficacy of NVR 3-778, alone and in combination with pegylated interferon, vs entecavir in uPA/SCID mice with humanized livers and HBV infection" *Gastroenterology*
82. Bock, Schwinn, Locarnini et al. (2001) "Structural organization of the hepatitis B virus minichromosome" *J Mol Biol*
83. Xia, Cheng, Nilsson et al. (2023) "Nucleolin binds to and regulates transcription of hepatitis B virus covalently closed circular DNA minichromosome" *Proc Natl Acad Sci*
84. Deng, Tian, Li et al. (2024) "ISG12a promotes immunotherapy of HBVassociated hepatocellular carcinoma through blocking TRIM21/AKT/βcatenin/PD-L1 axis"
85. Du, Wang, Liu et al. (2024) "C/EBPα-p30 confers AML cell susceptibility to the terminal unfolded protein response and resistance to Venetoclax by activating DDIT3 transcription" *J Exp Clin Cancer Res*
86. Ohlsson, Schuster, Hasemann et al. (2016) "The multifaceted functions of C/EBPα in normal and malignant haematopoiesis" *Leukemia*
87. Koschmieder, Halmos, Levantini et al. (2009) "Dysregulation of the C/EBPalpha differentiation pathway in human cancer" *J Clin Oncol*
88. Tolomeo, Grimaudo (2020) "The "Janus" role of C/EBPs family members in cancer progression" *Int J Mol Sci*
89. Spandau, Lee (1992) "Repression of the hepatitis B virus enhancer by a cellular factor" *J Gen Virol*
90. Choi, Park, Rho (1999) "Interaction of hepatitis B viral X protein and CCAAT/ enhancer-binding protein α synergistically activates the hepatitis B viral enhancer II/pregenomic promoter" *Journal of Biological Chemistry*
91. Ott, Ma, Li et al. (1999) "Regulation of hepatitis B virus expression in progenitor and differentiated cell types: evidence for negative transcriptional control in nonpermissive cells" *Gene Expr*
92. Dikic, Wakatsuki, Walters (2009) "Ubiquitin-binding domainsfrom structures to functions" *Nat Rev Mol Cell Biol*
93. Zheng, Shabek (2017) "Ubiquitin ligases: structure, function, and regulation" *Annu Rev Biochem*
94. Ullah, Sajid, Yan et al. (2021) "Antiviral activity of interferon alphainducible protein 27 against hepatitis B virus gene expression and replication" *Front Microbiol*
95. Rosen, Hsu, Wang et al. (2002) "C/EBPalpha induces adipogenesis through PPARgamma: a unified pathway" *Genes Dev*
96. Pedersen, Bereshchenko, Garcia-Silva et al. (2007) "Distinct C/EBPα motifs regulate lipogenic and gluconeogenic gene expression in vivo" *EMBO J*
97. Madsen, Siersbaek, Boergesen et al. (2014) "Peroxisome proliferator-activated re Proliferator-Activated Receptor γ and C/EBPα synergistically activate key metabolic adipocyte genes by assisted loading" *Mol Cell Biol*
98. Todisco, Santarsiero, Convertini et al. (2022) "PPAR alpha as a metabolic modulator of the liver: role in the pathogenesis of nonalcoholic steatohepatitis (NASH)" *Biology (Basel)*
99. Olofsson, Orho-Melander, William-Olsson et al. (2008) "CCAAT/enhancer binding protein α (C/EBPα) in adipose tissue regulates genes in lipid and glucose metabolism and a genetic variation in C/EBPα is associated with serum levels of triglycerides" *J Clin Endocrinol Metab*
100. Shojaei, Mclean (2025) "Interferon-stimulated gene IFI27 as a multifaceted candidate target in precision medicine" *Trends Immunol*
101. Yan, Zhang, Cai et al. (2015) "Spinoculation enhances HBV infection in NTCP-reconstituted hepatocytes"
102. Ibrahim, Liu, Zhang et al. (2024) "The loss of hepatitis B virus receptor NTCP/ SLC10A1 in human liver cancer cells is due to epigenetic silencing" *J Virol*
103. Guo, Jiang, Ma et al. (2009) "Activation of pattern recognition receptor-mediated innate immunity inhibits the replication of hepatitis B virus in human hepatocyte-derived cells" *J Virol*
104. Zhang, Liu, Liu et al. (2024) "The feasibility of establishing a hamster model for HBV infection: in vitro evidence" *mBio*
105. Jiang, Guo, Xu et al. (2008) "Identification of three interferon-inducible cellular enzymes that inhibit the replication of hepatitis C virus" *J Virol*
106. Mitra, Wang, Kim et al. (2019) "Hepatitis B virus precore protein p22 inhibits alpha interferon signaling by blocking STAT nuclear translocation" *J Virol*
107. Cai, Nie, Yan et al. (2013) "A southern blot assay for detection of hepatitis B virus covalently closed circular DNA from cell cultures" *Methods Mol Biol*
108. Kim, Guo (2024) "Hepatitis B virus covalently closed circular DNA chromatin immunoprecipitation assay" *Methods Mol Biol*
109. (2025) *Full-Length Text Journal of Virology*
110. Subramanian, Tamayo, Mootha et al. (2005) "Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles" *Proc Natl Acad Sci*
111. Liberzon, Subramanian, Pinchback et al. (2011) "Molecular signatures database (MSigDB) 3.0"
112. Liberzon, Birger, Thorvaldsdóttir et al. (2015) "The molecular signatures database (MSigDB) hallmark gene set collection" *Cell Syst*
113. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12710225&blobtype=pdf | # Associations of HIV status and the abundance of bla CTX-M and vanB resistance genes in stool samples of Ghanaian individuals
René Haugk, Holger Rohde, Fred Sarfo, Betty Norman, Albert Dompreh, Emmanuel Acheamfour-Akowuah, Shadrack Asibey, Richard Boateng, Edmund Kuffour, Veronica Cristanziano, Tafese Tufa, Torsten Feldt, Hagen Frickmann, Kirsten Eberhardt
## Abstract
Background: A cross-sectional study was performed to investigate associations of enteric colonization with resistant bacteria in Ghanaian individuals who were tested positive and negative for the human immunodeficiency virus (HIV). Methods: Abundance of the ESBL-(extended spectrum beta-lactamase-)type resistance-mediating gene bla CTX-M and the vancomycin resistant enterococci-(VRE-)associated genes vanA and vanB genes was associated with available clinical and epidemiological data on the study participants. Results: In terms of enteric carriage of ESBL-positive bacteria with CTX-M-type beta-lactam resistance genes, being HIV-positive (93.3% vs. 83.3%, P 5 0.003) and having low CD4þ T-lymphocyte counts <200 cells/μL (microliter) (96.8% vs. 91.2%, P 5 0.009) were identified as risk factors. Enteric carriage of ESBL-positive bacteria with CTX-M-type resistance genes was associated with poor immunological status in terms of lower CD4þ T-leukocyte counts, lower CD4þ/CD8þ ratios, higher viral replication, as well as with immune activation. For VRE, a non-significant trend for more VRE in control individuals without known HIV infection (6% vs. 2.5%, P 5 0.089) was observed. Conclusions: An association of ESBL colonization and immunological status was recorded. No such association was detected for VRE, suggesting different determinants of local VRE epidemiology.
## 1. INTRODUCTION
As repeatedly shown [1][2][3][4], colonization with resistant bacteria with the risk of difficult to treat infections is on the rise both in Sub-Saharan Africa in general [1,2] and in west African Ghana in particular [3,4]. While north Africa was confronted with high prevalence of carbapenem-resistant Enterobacterales as early as in this century's second decade [5,6], third generation cephalosporin resistance of the extended spectrum beta-lactamase (ESBL) type is a major resistance issue in sub-Saharan Africa [7][8][9][10][11][12]. Carbapenemases just recently started to emerge there [13]. As previous work has suggested that antimicrobial drug consumption is not the only risk factor for antimicrobial resistance [14], it seems appropriate to search for risk factors for the abundance of resistant bacterial colonization in African high endemicity settings.
One focus of recent epidemiological investigations has been on potential associations of acquired immunodeficiency syndrome (AIDS) caused by the human immunodeficiency virus (HIV) and the prevalence of colonizing resistant bacteria. In particular, several studies have assessed the prevalence of ESBL-expressing Enterobacterales in people-living-with-HIV (PLWH), also in west Africa. In the Senegal, high two-digit rates of ESBL-positivity among Enterobacterales causing blood-stream infections have been recorded [15]. In an assessment from Mozambique on bacteremia in HIV-positive febrile children, EBSL-positivity rates in the 50%-range were recorded for Klebsiella spp. (species) isolated from the children's blood stream [16]. In Nepal, high rates of ESBL-positive Enterobacterales were isolated from respiratory samples of PLWH with lower respiratory tract infections [17]. In Sub-Saharan Ethiopia, Tanzania, and Zimbabwe, but also in Germany and in the Netherlands, increased rates of enteric carriage of ESBL-positive Enterobacterales in PLWH have been described [18][19][20][21][22][23]. In sub-Saharan African studies on colonization with ESBL-positive Enterobacterales in PLWH which also assessed molecular mechanisms, bla CTX -M genes were most commonly identified as associated with local ESBL phenotypes, as described for Cameroon, Tanzania and Zimbabwe [20,21,23].
Recent application of cephalosporins and other antibiotic drugs, abundance of co-morbidities, and low CD4þ T-lymphocyte counts <350/μL were associated with increased carriage of ESBL-positive Enterobacterales in stool samples of Ethiopian and Tanzanian PLWH [18,21]. High enteric ESBL carriage in men-having-sex-with-men (MSM) and individuals with increased risk of acquiring sexually transmitted infections (STI) has generally been attributed to likely sexual transmission of ESBL-positive Enterobacterales [18,[24][25][26][27]. The effect was particularly pronounced in case of facilitating factors like anal-oral intercourse, frequent change of sexual partners, pre-exposure prophylaxis (PrEP) use, concomitant HIV infections, or the sharing of sex toys [24][25][26][27]. However, the sexual transmission route is unlikely to explain increased enteric carriage of ESBL-positive Enterobacterales as it was observed in HIV-positive children in Zimbabwe [20] who were too young to make consensual sexual activity likely. Instead, history of antibiotic consumption could be identified as a risk factor for the enteric carriage of ESBL-positive Enterobacterales in a cohort of Ethiopian HIV-positive children [22].
In Ethiopia, of note, several recent studies have also been conducted on colonization or infection caused by vancomycin-resistant enterococci (VRE) in PWLH [28][29][30][31], which hitherto is scarcely assessed in sub-Saharan African epidemiological research. Regionally increased VRE isolation rates in PWLH at various Ethiopian study sites [28][29][30][31] were associated with typical risk factors like selection under antibiotic pressure as well as history of treatment in local healthcare facilities [28,29]. In a recent epidemiological assessment on travelers returning from sub-Saharan Africa, VRE rates have been described as comparably low but associated with periodical shifts [7]. While enterococci are mostly isolated as colonizers rather than as causative agents of infections, systemic infections caused by Enterococcus spp. have been associated with increased severity and mortality rates in case of vancomycin resistance [32].
To contribute to available knowledge on associations of HIV infections and AIDS with enteric colonization with resistant bacteria in sub-Saharan Africa, a cross-sectional study was conducted with a population of Ghanaian PLWH as well as with a control population without known HIV infection. Molecular screening for ESBL-associated bla CTX-M genes, as common examples of ESBL-mediating genes, as well as for vancomycin resistance-associated vanA and vanB genes, as typical molecular resistance determinants in VRE colonizing the human enteric tract, was conducted and the results were associated with epidemiological, immunological, and socioeconomic information available for the assessed study population. By doing so, we aimed at screening for potentially underlying risk factors for resistant enteric colonization in Ghanaian individuals with and without HIV infections.
## 2. METHODS
## 2.1. Study design and sample materials
The investigation was conducted as a cross-sectional, hypothesis-forming, explorative assessment. The analyses were elements of a study on PLWH attending an HIV outpatient department at the Komfo Anokye Teaching Hospital (Kumasi, Ghana) as reported elsewhere [33,34]. Purpose of the study was to associate detections of gastrointestinal and other pathogens with immunological, clinical, and socio-demographic parameters in HIV-positive and HIV-negative Ghanaian adults. Approximately 50% of the assessed HIV-positive individuals were on antiretroviral combination therapy (cART) when the study was conducted. Accordingly, likely effects of cART could also be assessed. A smaller control population of HIV-negative Ghanaians was investigated within the same 12-months-study interval. Standardized questionnaires providing demographic, socio-economic, and clinical information were completed by the study participants supported with assistance by trained investigators.
## 2.2. Laboratory diagnostics
CD4þ T lymphocyte counts were measured from venous blood samples with a FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA, USA) at the Ghanaian study side. Quantification of the patient's HIV-1 viral load in the blood samples provided was done applying the Real-Time HIV-1 PCR (polymerase chain reaction) system (Abbott Diagnostics, Wiesbaden, Germany) according to the manufacturer's instructions. Collection of peripheral blood mononuclear cells (PBMCs) was further conducted from heparinized venous blood by centrifugation on a Ficoll/Hypaque (Biocoll Seperating Solution, Biochrom AG, Berlin, Germany) density gradient. Subsequently, the cells were washed with phosphate-buffered saline and then resuspended in Roswell Park Memorial Institute 1640 medium (Gibco Invitrogen, Carlsbad, CA, USA) supplemented with heatinactivated fetal calf serum (Biochrom AG, Berlin, Germany). After cryopreservation, the cells were shipped to Germany in liquid nitrogen. Upon arrival, staining of cell surface markers of immune activation was done as detailed elsewhere [34]. An LSRII flow cytometer (BD Biosciences, Heidelberg, Germany) allowed the recording of flow cytometric data and subsequently, the resulting information was analyzed with the software FlowJo (version 9.6.2, Tree Star, San Carlos, CA, USA).
The QIAamp stool DNA mini kit (Qiagen, Hilden, Germany) was used for desoxyribonucleic acid (DNA) extraction from the analyzed stool samples. The resulting eluates were afterwards stored at À80 8C. Using published oligonucleotides, a 3-plex real-time PCR assay was run on a LightCycler Pro device (Roche, Basel, Switzerland), comprising the ESBL gene bla CTX-M [7,35] and the vancomycin resistance-mediating genes vanA and vanB [7,36,37]. As detailed elsewhere, experimentally determined limits-of-detection ranged between 3.5 3 10 2 copies/μL and 1.6 3 10 3 copies/μL for the applied assays [7]. Details on the oligonucleotides are shown in Table 1.
The reaction mix comprised the HotStarTaq master mix (Qiagen, Hilden, Germany), a Mg 2þ concentration of 6.0 mM, primer and probe concentrations as indicated in Table 1, and 2 μL sample DNA eluate in total reaction volumes of 20 μL. The run conditions were as follows: initial denaturation at 95 8C for 15 min followed by 45 cycles of denaturation at 95 8C for 15 s, annealing at 60 8C for 60 s, elongation at 72 8C for 30 s, and subsequent hold at 40 8C for 20 s. Irrespective of the recorded cycle threshold (Ct) values, typical sigmoid-shaped amplification curves were accepted as positive real-time PCR signals. For quality control purposes, plasmid-based positive controls (sequence inserts in pEX-A128 vector plasmid backbones, produced by eurofins Genomics, Luxembourg, as provided in Table 1) and PCRgrade water-based negative controls were included in each run. Notably, PCR-inhibition within each sample was controlled with an in-house real-time PCR assay targeting a sequence fragment of Phocid Herpes Virus (PhHV) as detailed elsewhere [38].
Sole inclusion criterion for the here-presented study was the availability of residual stool sample material for the PCR assessments. PCR inhibition was the only exclusion criterion.
## 2.3. Statistics
All statistical analyses were conducted using R (version 4.4.3; R Foundation for Statistical Computing, Vienna, Austria). Categorical variables were compared with Fisher's exact test. Continuous variables were reported as median (IQR, interquartile range) and were compared using the Wilcoxon rank sum test. Multiple logistic regression was performed with the R package 'forestmodel'. Associations between continuous variables were assessed using Spearman's rank correlation coefficient (ρ). Statistical significance was defined as a twosided P-value < 0.05. No correction for multiple testing was applied due to the exploratory nature of the analyses [39].
## 2.4. Ethics
Compliance with the Declaration of Helsinki and all its amendments was ensured for the study. Sample collection and analysis followed protocols approved by the Committee on Human Research of the Kwame Nkrumah University of Science and Technology in Kumasi, Ghana: CHRPE/AP/12/11, and the ethics committee of the Medical Council in Hamburg, Germany: PV3771. In order to be included in the study, all participants or next-to-kin, if applicable, had to provide written informed consent.
## 3. RESULTS
## 3.1. Prevalence of resistance gene DNA within the stool samples of the study population
A total of 1,095 HIV-positive individuals and 107 HIV-negative controls were enrolled into the study. Residual stool samples suitable for resistance gene testing were available from 651 HIV-positive and 84 HIV-negative participants, who could be included in the here-presented assessment.
The overall prevalence of ESBL-associated bla CTX -M DNA in the tested cohort was 92.7% (number (n) 5 681). Stratification by HIV status revealed a significantly higher prevalence of bla CTX -M genes in HIV-positive individuals (93.9%, n 5 611) compared to HIV-negative controls (83.3%, n 5 70; P 5 0.003; Fig. 1A). Among HIV-positive participants, bla CTX -M detection rates further increased with decreasing CD4þ T lymphocyte counts: prevalence was 91.2% (n 5 477) in those with CD4þ counts ≥200 cells/μL and 96.8% (n 5 182) in those with CD4þ counts <200 cells/μL (P 5 0.009; Fig. 1B).
The vancomycin resistance gene vanA was not detected in any participant. The overall prevalence of vanB was 2.9%. Although not statistically significant, vanB prevalence was higher in HIV-negative individuals (6.0%, n 5 5) than in HIV-positive participants (2.5%, n 5 16; P 5 0.089). Focusing on CD4þ T-lymphocyte counts, vanB was detected in 3.4% (n 5 17) samples from individuals with CD4þ counts ≥200 cells/μL and in 2.2% (n 5 4) from those with CD4þ counts <200 cells/μL.
## 3.2. Comparison of demographic, socio-economic and clinical characteristics of the HIV cohort according to the presence or absence of resistance gene DNA in their stool samples
Table 2 summarizes the demographic, socio-economic, and clinical characteristics of HIV-positive participants stratified by the presence of bla CTX -M and vanB resistance genes. Among HIV-positive individuals, those with detectable bla CTX -M DNA had a median age of 40 years (IQR 33-47), compared to 42 years (IQR 36-50) in bla CTX -M negative participants (P 5 0.215). The proportion of females was similar comparing bla CTX -M positive (73.6%) and negative (74.4%) groups. Socioeconomic parameters, including access to tap water, electricity, and household refrigerator ownership, did not differ significantly between the groups. Regarding medical treatment, the proportion of participants Table 1. Target genes, calculated detection limits and oligonucleotides (including concentrations applied in the PCR mix) applied for the real-time PCR screening assays for bla CTX-M genes mediating CTX-M-type extended spectrum beta-lactamases (ESBL) expression and for vanA and vanB genes mediating vancomycin resistance. Hyphens in the oligonucleotide sequences have been inserted to increase the readability, not to delineate codon triplets
## PCR target CTX-M-type ESBL
Target gene bla CTX-M Detection limit 1.6 3 10 3 copies/μL Forward primer 1 (concentration) 5 0 -GCT-GGA-CTG-CCT-GCT-TCC-T-3 0 (0.32 pmol μL À1 ) Forward primer 2 (concentration) 5 0 -TGC-CGA-AAT-CAT-GGG-TAG-TG-3 0 (0.32 pmol μL À1 ) Forward primer 3 (concentration) 5 0 -CTA-CCC-ACA-TCG-TGG-GTT-GTC-3 0 (0.32 pmol μL À1 ) Forward primer 4 (concentration) 5 0 -ATT-CGG-GCC-GGC-TTA-CC-3 0 (0.32 pmol μL À1 ) Reverse primer 1 (concentration) 5 0 -CGT-TGG-TGG-TGC-CAT-AGY-CA-3 0 (0.32 pmol μL À1 ) Reverse primer 2 (concentration) 5 0 -TCG-TTG-GTG-GTG-CCA-TAA-TCT-3 0 (0.32 pmol μL À1 ) Reverse primer 3 (concentration) 5 0 -GAT-GTC-ATT-CGT-CGT-ACC-ATA-ATC-A-3 0 (0.32 pmol μL À1 ) Reverse primer 4 (concentration) 5 0 -ATC-ATT-GGT-GGT-GCC-GTA-GYC-3 0 (0.32 pmol μL À1 ) Reverse primer 5 (concentration) 5 0 -GCG-ATA-TCA-TTC-GTC-GTA-CCA-TAA-3 0 (0.32 pmol μL À1 ) Probe and modifications (concentration) European Journal of Microbiology and Immunology 15 (2025) 4, 184-194 receiving cART was significantly higher among bla CTX -M negative individuals (66.7%) compared to those who were bla CTX -M positive (39.8%; P 5 0.001). The use of trimethoprim/sulfamethoxazole (TMP/SMX) prophylaxis was comparable between the two groups. Notably, the median number of days since HIV diagnosis was substantially higher in bla CTX -M negative individuals (1,584 days, IQR 19-2,564) than in those with bla CTX -M detection (177 days, IQR 12-1,624; P 5 0.007). Clinical parameters such as body mass index (BMI), frequency of cough, diarrhea, fever, and weight loss did not show significant differences between bla CTX -M positive and negative groups. BMI was marginally and not significantly higher in bla CTX -M positive participants (median 24, IQR 21-26) compared to negatives (median 22, IQR 20-25; P 5 0.054). When stratified by vanB gene status, vanB positive participants were slightly younger (median 35 years, IQR 31-38) than vanB negatives (median 40 years, IQR 33-47; P 5 0.036). Other demographic and socio-economic characteristics, including gender distribution and household amenities, were similar between the groups. The proportion of participants on cART, use of TMP/SMX prophylaxis, and median days since HIV diagnosis did not differ significantly by vanB status. Likewise, clinical symptoms and BMI did not show significant associations with vanB detection.
## 3.3. Comparison of virological and immunological characteristics of cART naïve HIV-positive participants depending on the abundance or absence of bacterial resistance gene DNA in their stool samples
Virological and immunological parameters differed significantly according to resistance gene status among HIV-positive participants (Table 3). Individuals with detectable bla CTX -M DNA had a higher median viral load (4.4 log 10 copies/mL, IQR 1.6-5.4) compared to those without detectable bla CTX -M genes (1.6, IQR 1.6-3.9; P 5 0.009). The median CD4þ T cell count was significantly lower in bla CTX -M positive individuals (331/μL, IQR 140-564) than in bla CTX -M negative participants (510/μL, IQR 307-721; P 5 0.003), and the CD4þ/CD8þ ratio was also reduced (0.35 vs. 0.55; P 5 0.001). Markers of immune activation, including HLA-DRþ CD38þ CD4þ and CD8þ T cells, were significantly elevated in bla CTX -M positive individuals (both P 5 0.001). No significant differences were observed in CD8þ T cell counts, senescence, or exhaustion markers (CD57þ or PD-1þ T cells) between bla CTX -M positive and negative groups, although CD57þ CD4þ T cells trended higher among bla CTX -M positive participants (P 5 0.058).
In contrast, vanB positivity was not associated with significant differences in viral load, CD4þ or CD8þ T cell counts, CD4þ/CD8þ ratio, or most activation and exhaustion markers. However, the proportion of CD57þ CD4þ T cells was significantly higher in vanB positive individuals (26% vs. 14%; P 5 0.036).
## 3.4. Factors associated with co-colonization with resistant bacteria in the HIV-positive cohort
Logistic regression analyses were performed to identify factors independently associated with the detection of resistance genes among study participants (Fig. 2). In model A, intake of cART was significantly associated with a lower likelihood of bla CTX -M detection after adjustment for demographical parameters (odds ratio (OR) 0.34, 95% CI 0.16-0.66; P 5 0.002). In the alternative model (panel B), a higher CD4þ T cell count was associated with a reduced odds of bla CTX -M detection (OR 0.89 per 100 cells/μL, 95% CI 0.82-0.97; P 5 0.007).
For vanB detection, logistic regression showed that younger age was marginally associated with vanB positivity in panel C (OR 0.54 per 10 years, 95% CI 0.28-0.96; P 5 0.05), while neither sex nor cART intake were significant predictors. In panel D, only a marginal association with age was observed for vanB detection, and CD4þ T cell count was not identified as a predictor of vanB carriage.
## 3.5. Correlations of Ct values for specific resistance gene sequences with CD4þ T cell count, CD4þ/ CD8þ T cell ratio, and HIV viral load
The correlation analysis between Ct values from real-time PCR targeting bla CTX -M and immunological as well as virological parameters is summarized in Table 4. Ct values for bla CTX -M showed a modest but significant positive correlation with both CD4þ T cell count (Spearman's ρ 5 0.26, P < 0.001) and CD4þ/CD8þ T cell ratio (ρ 5 0.24, P < 0.001), and a significant negative correlation with viral load (ρ 5 À0.32, P < 0.001). In contrast, Ct values from PCR targeting vanB did not show significant correlations with any of these parameters (all P 5 1.000). Notably, median Ct-values (IQR) over the whole study population were 31 (28,35) for bla CTX-M genes and 34 (32,36) for vanB genes.
## 4. DISCUSSION
The study was conducted to characterize potential associations between the molecular screening results for ESBL-associated bla CTX-M genes as well as for vancomycin resistance-associated vanA and vanB genes and key epidemiological, immunological, and socio-economic indicators among Ghanaian PLWH and a control group without known HIV infection. The assessment led to several noteworthy observations which we discuss below. Focusing on the recorded bla CTX-M gene detections, the study once more confirmed very high enteric colonization rates with ESBL-positive Enterobacterales in Ghanaian individuals [40]. The very high bla CTX-M gene detection rates irrespective of the study participants' HIV status insinuate that among Ghanaians, ESBL gene-carrying bacteria could be considered as expectable enteric microbial flora. This finding aligns with historic reports on very high enteric colonization rates with ESBL-positive Enterobacterales even in travelers returning from west Africa [41,42]. Notably, bla CTX-M genes have not even been described as the quantitatively dominant molecular determinant of ESBL phenotypes in Ghanaian bacterial isolates. Instead, ESBL-associated bla SHV genes have recently been reported to be locally even more common [42]. In the present molecular assessment, we abstained from bla SHV assessment due to the challenging discrimination of ESBL-associated and non-ESBL-associated bla SHV genes just based on PCR.
When comparing the Ghanaian PLWH and the control individuals without known HIV infection, HIV-positive Ghanaians were more likely to harbor measurable quantities of ESBL-positive bacteria in their enteric tract. And even within the PLWH subpopulation, poor immunological status as characterized by low CD4þ T-lymphocyte counts and CD4þ/CD8þ T-lymphocyte ratios as well as by high viral loads associated with increased immune activation were demonstrated as determinants of increased likelihood of carrying ESBL-positive bacteria in the gut. The observation could be confirmed on the quantitative level as well: measured Ct values for bla CTX-M genes were positively correlated with CD4þ T-lymphocyte counts and CD4þ/ CD8þ T-lymphocyte ratios and negatively correlated with HIV viral load. As target sequence quantity within the samples and measured Ct-value are negative associated in real-time PCR, this means that quantitatively high abundance of bla CTX-M genes within the assessed stool samples of PLWH was associated with poor immune status as indicated by low CD4þ T-lymphocyte counts and CD4þ/CD8þ T-lymphocyte ratios as well as by high HIV viral loads.
The findings are consistent with previous reports from the international literature. In particular, low CD4þ T-lymphocyte counts <350/μL as indicators of poor immune status have previously been associated with increased enteric carriage of ESBL-positive Enterobacterales in PLWH from Ethiopia and Tanzania [18,21]. Further, high ESBL colonization rates in immunosuppressed organ transplant patients have been reported in a meta-analysis [42] and a recent review [43] mentioned immunosuppression and immunocompromising disease as risk factors for enteric colonization with ESBL-positive bacteria. However, the underlying studies [44,45] did not allow the calculation of effect strengths. The question on the reason for the observed associations is less easy to answer, because the abundance or nonabundance of resistance genes not necessarily interferes with a bacterial isolate's virulence and evolutionary fitness [46]. And without changes in virulence, the association of immunosuppression is most likely an indirect one. Regarding the higher likelihood of enteric carriage of an ESBL-positive bacterium in PLWH compared to the control group without known HIV infection, it seems reasonable to assume more contacts of HIV-positive individuals with medical care facilities, although respective information was not collected for this study. Recent stays at healthcare facilities are considered as well-established risk factors for colonization with ESBL-positive Enterobacterales [43]. Likewise, the number and intensity of hospital stays might also account for the observed increased risk of enteric colonization with ESBL-positive Enterobacterales for more severely immunosuppressed individuals within the PLWH subpopulation. This is because immunosuppression due to unknown and thus not adequately treated HIV infection might result in increased susceptibility to opportunistic infections and thus again to more and prolongated contacts with healthcare facilities. Next to transmission risks, antimicrobial therapies potentially associated with contacts to the healthcare system due to assumed or real infections can further facilitate the selection of resistant bacteria [43]. In the absence of collected data on the frequency and intensity of the study participants' recent contacts with medical facilities and on their consumption of antibiotics, this explanation remains speculative and should be addressed in future studies. However, uncontrolled intake of antibiotic drugs is indeed well-known to drive enteric carriage of ESBL-positive bacteria to substantially high proportions in Ghana [47]. Further, the hypothesis would also explain findings like decreased likelihood of ESBL-positivity in individuals on cART. Establishment of cART treatment and thus better immune status implies less susceptibility to opportunistic diseases and thus less contact to healthcare facilities as well as less non-targeted consumption of antimicrobial drugs, likely reducing the transmission and selection risk for ESBL-positive Enterobacterales. The same applies to the otherwise difficult to explain negative association between the time since HIV diagnosis and the abundance of ESBL-positive enteric colonization.
Notably, neither the socio-economic parameters assessed nor the prophylactic use of TMP/SMX had a measurable impact on the risk of ESBL-positive enteric colonization. The lack of association between prophylactic antibiotic exposure and ESBL-positive enteric colonization is quite surprising, because medical history of antibacterial therapy has been identified as an important risk factor for ESBL-positive colonization with large effect sizes [43,48]. A number of reasons may account for this observation. Hypothetically, baseline TMP/SMX resistance might have been so frequent in the study population that the additional drug intake could not show much additive effect. Alternatively, co-occurrence of TMP/SMX resistance and ESBL-type beta-lactam resistance might have been low, thus preventing intense TMP/SMX-induced selection of ESBL-expressing bacteria. Lastly, the lack of association between ESBL-positive enteric colonization and clinical symptoms was expected, as colonization is not associated with apparent disease.
Focusing on vanB as a molecular resistance determinant prevalent in west African VRE [49], a strikingly contrasting epidemiology was observed, suggesting other modes of transmission and enteric colonization in Ghana. In addition to a low overall detection rate within the assessed population, non-significant tendencies for a higher likelihood of positive test results in the control population without known HIV infection and for the immunologically less compromised individuals within the PLWH subpopulation at least did not provide any information pointing towards HIV infection-associated immunosuppression as a risk factor for enteric colonization with VRE in Ghana. While there was a signal for a potential association of VRE positivity and a higher proportion of CD57þ CD4þ T-lymphocytes as a sign of immune exhaustion with a low significance level, this single observation was considered as an incidental finding with uncertain relevance. A trend towards more VRE in younger Ghanaian individuals remained stable even in multi-variate analysis. This is in agreement with a European study showing a peak of VRE-associated bloodstream infections in young adults < 30 years-of-age next to another peak in elderly individuals [50]. Anyway, the here-described trend towards more VRE in young Ghanaians without known HIV-infection speaks in favor of transmission routes different from frequent or intense contacts with local healthcare facilities. However, the available epidemiological data from the present study are insufficient to provide evidence-guided hypotheses. Of note, vanA as another common molecular determinant of vancomycin resistance in enterococci isolated from human samples was not at all detected in the assessed population, although the gene has been described to be generally abundant in Africa [49].
The study has a number of limitations. First, this study had a retrospective design which precludes causal inferences. Second, a formal sample size was not calculated to inform effect size estimates between antimicrobial resistance mutations and dependent variables, hence the study can be considered as an exploratory, hypothesis-forming assessment only. Third, information on the patients' history of contacts to healthcare facilities or antibiotic treatments of opportunistic infections were not collected in the conception of the study. Fourth, the study was based on molecular assessments of resistance genes only, making statements on the identity and the vitality of the bacteria carrying those resistance genes not feasible. Fifth, the applied molecular screening assays showed limits-of-detection in the 10 2 to 10 3 copies/μL DNA eluate range even when testing positive control plasmids diluted in water. In complex sample matrices like stool, even higher detection limits need to be expected and accordingly, it is likely that low colonization densities may have gone undetected. Sixth, modern molecular approaches with a broader diagnostic spectrum like next-generation sequencing-based metagenomics could not be applied due to funding restrictions.
## 5. CONCLUSIONS
In spite of its limitations, the hypothesis-forming, explorative study indicated facilitating effects of HIV positivity and HIV-associated immunosuppression on the abundance of ESBL-positive bacteria in the enteric tract of Ghanaian PLWH. For VRE, in contrast, such an association was not detectable, suggesting other modes of transmission and spread within the Ghanaian population. As a side finding, the assessment once more confirmed very high enteric colonization rates with ESBL-positive bacteria compared to low colonization rates with VRE in the Ghanaian population irrespective of the HIV status. These very high ESBL rates need to be considered for antibiotic therapy decisions in case of suspected systemic bacterial infections with likely enteric source. This may be associated with increased therapy costs [51], prolongated antimicrobial therapy [52], especially if initial antimicrobial therapy did not cover a resistant causative agent of systemic infection, and high efforts for infection prevention and control procedures [53], because nosocomial transmission of resistant bacteria is frequent under resource-limited conditions [54].
## Conflicts of interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
## ABBREVIATIONS
## References
1. (2024) "Global burden of bacterial antimicrobial resistance 1990-2021: a systematic analysis with forecasts to 2050"
2. Toy, Pak, Duc et al. (2019) "Multicountry distribution and characterization of extended-spectrum β-lactamase-associated gram-negative bacteria from bloodstream infections in sub-Saharan Africa" *Clin Infect Dis*
3. Opintan, Newman (2017) "Prevalence of antimicrobial resistant pathogens from blood cultures: results from a laboratory based nationwide surveillance in Ghana" *Antimicrob Resist Infect Control*
4. Eibach, Campos, Krumkamp et al. (2016) "Extended spectrum beta-lactamase producing Enterobacteriaceae causing bloodstream infections in rural Ghana, 2007-2012" *Int J Med Microbiol*
5. Frickmann, Köller, Hagen et al. (2018) "Molecular epidemiology of multidrug-resistant bacteria isolated from Libyan and Syrian patients with war injuries in two Bundeswehr hospitals in Germany" *Eur J Microbiol Immunol (Bp)*
6. Kollenda, Frickmann, Helal et al. (2019) "Screening for carbapenemases in ertapenemresistant Enterobacteriaceae collected at a Tunisian hospital between 2014 and 2018" *Eur J Microbiol Immunol (Bp)*
7. Navabi, Wiemer, Halfter et al. (2024) "Spatial and temporal dynamics of the prevalence of resistance genes and gastrointestinal pathogens in stool samples of German deployment returnees" *Eur J Microbiol Immunol (Bp)*
8. Micheel, Hogan, Rakotoarivelo et al. (2015) "Identification of nasal colonization with β-lactamase-producing Enterobacteriaceae in patients, health care workers and students in Madagascar" *Eur J Microbiol Immunol (Bp)*
9. Hagen, Hinz, Frickmann (2015) "β-Lactamases encoded by bla CTX-M group I genes as determinants of resistance of ESBLpositive Enterobacteriaceae in European soldiers in tropical Mali" *Eur J Microbiol Immunol (Bp)*
10. Konaté, Dembélé, Guessennd et al. (2017) "Epidemiology and antibiotic resistance phenotypes of diarrheagenic Escherichia coli responsible for infantile gastroenteritis in Ouagadougou" *Eur J Microbiol Immunol (Bp)*
11. Ogbolu, Alli, Webber et al. (2018) "CTX-M-15 is established in most multidrug-resistant uropathogenic Enterobacteriaceae and Pseudomonaceae from hospitals in Nigeria" *Eur J Microbiol Immunol (Bp)*
12. Olowe, Choudhary, Schierack et al. (2013) "Pathotyping bla CTX-M Escherichia coli from Nigeria" *Eur J Microbiol Immunol (Bp)*
13. Odewale, Adefioye, Ojo et al. (2016) "Multidrug resistance of Acinetobacter baumannii in Ladoke Akintola University Teaching hospital" *Eur J Microbiol Immunol (Bp)*
14. Collignon, Beggs, Walsh et al. (2018) "Anthropological and socioeconomic factors contributing to global antimicrobial resistance: a univariate and multivariable analysis" *Lancet Planet Health*
15. Wembulua, Lakhe, Mbaye et al. (2013) "Antibacterial susceptibility patterns of bloodstream isolates in 74 HIV-infected patients hospitalized at the clinic of infectious and tropical diseases of Fann University Hospital, Dakar from 2013 to" *Med Trop Sante Int*
16. Kenga, Gebretsadik, Simbine et al. (2021) "Community-acquired bacteremia among HIVinfected and HIV-exposed uninfected children hospitalized with fever in Mozambique" *Int J Infect Dis*
17. Maharjan, Bastola, Adhikari et al. (2022) "Multidrug-resistant bacteria with ESBL genes: a growing threat among people living with HIV/AIDS in Nepal" *BMC Infect Dis*
18. Befikadu, Tamrat, Garedo et al. (2024) "Faecal carriage of extended-spectrum beta-lactamase and carbapenemase-producing enterobacterales among HIV patients at Jimma Medical Center" *BMC Microbiol*
19. Reinheimer, Keppler, Stephan et al. (2017) "Elevated prevalence of multidrug-resistant gram-negative organisms in HIV positive men" *BMC Infect Dis*
20. Wilmore, Kranzer, Williams et al. (2017) "Carriage of extended-spectrum beta-lactamaseproducing Enterobacteriaceae in HIV-infected children in Zimbabwe" *J Med Microbiol*
21. Manyahi, Moyo, Tellevik et al. (2020) "High prevalence of fecal carriage of extended spectrum β-lactamase-producing Enterobacteriaceae among newly HIV-diagnosed adults in a community setting in Tanzania" *Microb Drug Resist*
22. Bayleyegn, Fisaha, Kasew (2021) "Fecal carriage of extended spectrum beta-lactamase producing Enterobacteriaceae among HIV infected children at the University of Gondar Comprehensive Specialized Hospital Gondar" *AIDS Res Ther*
23. Dimani, Founou, Zemtsa et al. (2023) "Faecal carriage of multidrug-resistant and extended-spectrum β-lactamase-producing Enterobacterales in people living with HIV in Yaoundé" *Cameroon. J Glob Antimicrob Resist*
24. Boyd, Mathieu, Françoise et al. (2024) "Sexual behaviors and risk of extended-spectrum β-lactamaseproducing Enterobacterales carriage: a cross-sectional analysis" *Int J Infect Dis*
25. Danjean, Surgers, Royer et al. (2025) "Extensive dissemination of ESBL-producing Clonal Complex 14 Escherichia coli is likely spread through sexual transmission among men who have sex with men at risk of sexually transmitted infections" *J Infect*
26. Van Bilsen, Van Dulm, Matser et al. (2021) "High carriage of ESBLproducing Enterobacteriaceae associated with sexual activity among men who have sex with men" *Int J Antimicrob Agents*
27. Surgers, Chiarabini, Royer et al. (2022) "Evidence of sexual transmission of extended-spectrum β-lactamase-producing enterobacterales: a cross-sectional and prospective study" *Clin Infect Dis*
28. Abebe, Endris, Tiruneh et al. (2014) "Prevalence of vancomycin resistant Enterococci and associated risk factors among clients with and without HIV in Northwest Ethiopia: a crosssectional study"
29. Dadi, Solomon, Tesfaye (2021) "Vancomycin resistant Enterococci and its associated factors among HIV infected patients on anti-retroviral therapy in Ethiopia" *PLoS One*
30. Zike, Ahmed, Hailu et al. (2024) "Vancomycin resistant enterococci prevalence, antibiotic susceptibility patterns and colonization risk factors among HIV-positive patients in healthcare facilities in Debre Berhan Town" *Ethiopia. Infect Drug Resist*
31. Tilahun, Gedefie, Sahle (2023) "Asymptomatic carriage rate, multidrug resistance level, and associated risk factors of Enterococcus in clinical samples among HIV-positive patients attending at Debre Birhan comprehensive specialized hospital" *Biomed Res Int*
32. Bhavnani, Drake, Forrest et al. (2000) "A nationwide, multicenter, case-control study comparing risk factors, treatment, and outcome for vancomycinresistant and -susceptible enterococcal bacteremia" *Diagn Microbiol Infect Dis*
33. Sarfo, Eberhardt, Dompreh et al. (2015) "Helicobacter pylori infection is associated with higher CD4 T cell counts and lower HIV-1 viral loads in ART-Naïve HIV-positive patients in Ghana" *PLoS One*
34. Eberhardt, Sarfo, Dompreh et al. (2015) "Helicobacter pylori coinfection is associated with decreased markers of immune activation in ART-naive HIV-positive and in HIV-negative individuals in Ghana" *Clin Infect Dis*
35. Van Der Zee, Roorda, Bosman et al. (2014) "Multi-centre evaluation of real-time multiplex PCR for detection of carbapenemase genes OXA-48, VIM, IMP, NDM and KPC" *BMC Infect Dis*
36. Fang, Ohlsson, Jiang et al. (2012) "Screening for vancomycin-resistant enterococci: an efficient and economical laboratory-developed test" *Eur J Clin Microbiol Infect Dis*
37. Both, Berneking, Berinson et al. (2020) "Rapid identification of the vanA/vanB resistance determinant in Enterococcus sp. from blood cultures using the Cepheid Xpert vanA/vanB cartridge system" *Diagn Microbiol Infect Dis*
38. Niesters (2001) "Quantitation of viral load using real-time amplification techniques" *Methods*
39. Vanderweele, Mathur (2019) "Some desirable properties of the Bonferroni correction: is the Bonferroni correction really so bad?" *Am J Epidemiol*
40. Yamamoto, Hoang, Le et al. (2025) "Multinational comparison of the detection of extended-spectrum beta-lactamase genes in healthy resident feces" *Microbiol Spectr*
41. Lübbert, Straube, Stein et al. (2015) "Colonization with extended-spectrum beta-lactamase-producing and carbapenemase-producing Enterobacteriaceae in international travelers returning to Germany" *Int J Med Microbiol*
42. Alevizakos, Kallias, Flokas et al. (2017) "Colonization with extended-spectrum beta-lactamase-producing Enterobacteriaceae in solid organ transplantation: a meta-analysis and review" *Transpl Infect Dis*
43. Biehl, Schmidt-Hieber, Liss et al. (2016) "Colonization and infection with extended spectrum beta-lactamase producing Enterobacteriaceae in high-risk patients -review of the literature from a clinical perspective" *Crit Rev Microbiol*
44. Daoud, Moubareck, Hakime et al. (2006) "Extended spectrum beta-lactamase producing Enterobacteriaceae in Lebanese ICU patients: epidemiology and patterns of resistance" *J Gen Appl Microbiol*
45. Trecarichi, Tumbarello, Spanu et al. (2009) "Incidence and clinical impact of extended-spectrum-beta-lactamase (ESBL) production and fluoroquinolone resistance in bloodstream infections caused by Escherichia coli in patients with hematological malignancies" *J Infect*
46. Beceiro, Tomás, Bou (2013) "Antimicrobial resistance and virulence: a successful or deleterious association in the bacterial world?" *Clin Microbiol Rev*
47. Heinemann, Kleinjohann, Rolling et al. (2023) "Impact of antibiotic intake on the incidence of extended-spectrum β-lactamase-producing Enterobacterales in sub-Saharan Africa: results from a community-based longitudinal study" *Clin Microbiol Infect*
48. Arnan, Gudiol, Calatayud et al. (2011) "Risk factors for, and clinical relevance of, faecal extended-spectrum β-lactamase producing Escherichia coli (ESBL-EC) carriage in neutropenic patients with haematological malignancies" *Eur J Clin Microbiol Infect Dis*
49. Osei Sekyere, Mensah (2020) "Molecular epidemiology and mechanisms of antibiotic resistance in Enterococcus spp., Staphylococcus spp., and Streptococcus spp. in Africa: a systematic review from a One Health perspective" *Ann N Y Acad Sci*
50. Brinkwirth, Martins, Ayobami et al. (2022) "Germany's burden of disease of bloodstream infections due to vancomycin-resistant Enterococcus faecium between 2015-2020" *Microorganisms*
51. Marino, Maniaci, Lentini et al. (2025) "The global burden of multidrug-resistant bacteria" *Epidemiologia (Basel)*
52. Marino, Augello, Bellanca et al. (2025) "Antibiotic therapy duration for multidrug-resistant gram-negative bacterial infections: an evidence-based review" *Int J Mol Sci*
53. Münch, Hagen, Müller et al. (2017) "Colonization with multidrug-resistant bacteria -on the efficiency of local decolonization procedures" *Eur J Microbiol Immunol (Bp)*
54. Granzer, Hagen, Warnke et al. (2016) "Molecular epidemiology of carbapenem-resistant Acinetobacter baumannii complex isolates from patients that were injured during the Eastern Ukrainian conflict" *Eur J Microbiol Immunol (Bp)*
55. "), which permits unrestricted use, distribution, and reproduction in any medium for non-commercial purposes, provided the original author and source are credited, a link to the CC License is provided" |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12645909&blobtype=pdf | # MmuPV1 E7 promotes phenotypes associated with "high-risk" HPV infection in mouse keratinocytes
Kayla Duxbury, Liyan Zhang, Laura Muelhbauer, Mitchell Hayes, Joshua Coon, Megha Padi, James Romero- Masters
## Abstract
The E7 oncoprotein of mouse papillomavirus (MmuPV1) plays a pivotal role in both viral infection and cancer development. While earlier studies have identified key cellular targets of MmuPV1 E7, such as pRB and PTPN14, the broader impact of MmuPV1 E7 on keratinocyte homeostasis and shared activities with human papillomavi rus (HPV) E7 remains unclear. In this study, we employed proteomic and transcriptomic analyses using our established mouse keratinocyte model-previously instrumental in uncovering a novel function of MmuPV1 E6-to investigate the biological consequences of MmuPV1 E7 expression in mouse keratinocytes. Our findings reveal that MmuPV1 E7 induces cellular changes reminiscent of those driven by "high-risk" HPV infection implicated in cervical cancer. Notably, MmuPV1 E7 did not activate canonical E2F-respon sive gene expression or promote proliferation, reinforcing the idea that MmuPV1 E6 is the primary driver of cell cycle activation. However, MmuPV1 E7 expression led to a significant accumulation of stress keratin 17, a marker associated with immune evasion and elevated in both HPV16 transgenic models and MmuPV1 infections. Additionally, we observed enhanced PI3K-AKT-mTOR signaling, with increased levels of phosphorylated S6 kinase and heightened sensitivity to epidermal growth factor stimulation. Collectively, these results underscore the role of MmuPV1 E7 in promoting oncogenic phenotypes and highlight its relevance as a model for studying the molecular underpinnings of "high-risk" HPV-driven disease.IMPORTANCE In this study, we determined the ability of the MmuPV1 E7 oncoprotein in promoting disruption of keratinocyte homeostasis in mouse keratinocytes. Using a multiomics approach, we observed that MmuPV1 E7 promoted several phenotypes associated with "high-risk" human papillomavirus (HPV) infection. Specifically, we confirmed that MmuPV1 E7 does not increase E2F-responsive gene expression and proliferation of mouse keratinocytes. We did find that MmuPV1 E7 was able to increase the expression of stress keratin 17, which promotes immune evasion in papillomavi rus infections. Finally, MmuPV1 E7 showed increased expression of genes associated with PI3K-AKT-mTOR signaling. Consistent with this observation, MmuPV1 E7-express ing mouse keratinocytes had elevated phosphorylation of S6 kinase. We also found that MmuPV1 E7 potentiates this signaling through increased sensitivity to epidermal growth factor stimulation. Our collective data show that MmuPV1 E7 promotes several phenotypes associated with "high-risk" HPV infection and cancers.
HPVs ("high-risk" HPVs) are causative agents of cervical and head and neck cancers and cause 5% of the world's total cancer burden (3,4). In contrast, "low-risk" mucosal and cutaneous HPVs typically cause benign neoplastic disease, i.e., genital and skin warts, in patients (5)(6)(7). A small subset of cutaneous HPVs is associated with non-mela noma skin cancer in long-term immunosuppressed transplant patients and patients with the genetic disorder epidermodysplasia verruciformis (8)(9)(10).
Many decades of work have solidified the role of HPV in cancer development through activities of the virally encoded E6 and E7 oncogenes, in particular "high-risk" HPVs. The ability of "high-risk" HPV E6 and E7 to immortalize cells has been linked to inhibition of tumor suppressor signaling in cells (11)(12)(13). "High-risk" HPV E6's most well-charac terized activities are promoting degradation of p53 and increased telomerase reverse transcriptase (TERT) activity, which leads to prolonged survival of HPV-infected cells (14)(15)(16)(17)(18)(19). "High-risk" HPV E7's most well-characterized activities include the degrada tion of the retinoblastoma tumor suppressor (pRB) and the cellular protein tyrosine phosphatase non-receptor 14 (PTPN14) (20)(21)(22)(23)(24). The inactivation of pRB by HPV E7 promotes constitutive activation of E2F transcription factors and leads to dysregulation of cellular proliferation (25). The ability of HPV E7 to promote degradation of PTPN14 has been linked to resistance to anoikis and activation of oncogenic Yes-associated protein signaling (22)(23)(24). However, the significance of "high-risk" HPV E7's activities that contribute to disease development in vivo, including cancer, has been limited to the transgenic mouse model system (26)(27)(28)(29)(30)(31)(32)(33). In this model system, the most well-character ized activity of "high-risk" HPV E7 has been E7's interaction with pRB family members (33,34). In contrast, the role of PTPN14 in the transgenic animal model setting has never been studied. Cutaneous HPV E7 has been shown to interact with tumor suppressors pRB and PTPN14 (35)(36)(37)(38)(39). In contrast to "high-risk" HPV E7, cutaneous HPV E7 degradation of pRB is observed but not uniform among the various subtypes (36,37). Cutaneous HPV transgenic models have been used to study the role of cutaneous HPV E6 and E7 in vivo (40)(41)(42). However, cutaneous HPV E6 and E7 in general are understudied compared to their "high-risk" HPV counterparts.
The species-specific nature of papillomaviruses has inhibited the ability to study the contributions of HPV E6 and E7 to an infection in vivo. The murine papillomavirus, MmuPV1, discovered in 2011, enables us to study a papillomavirus infection within a genetically tractable preclinical animal model (43). MmuPV1 infects laboratory mice (Mus musculus), at all the same anatomical sites as HPV, and causes lesions that can progress to cancer, including skin, cervicovaginal tract, anal tract, and oral cavity (44)(45)(46)(47)(48)(49)(50)(51). These observations make MmuPV1 an attractive model for studying HPV-associated pathogen esis. MmuPV1 is a member of the Pipapillomavirus genus and is more genetically similar to cutaneous HPV genotypes. The MmuPV1 E6 and E7 oncoproteins are capable of binding to known cellular binding partners of HPV E6 and E7 (52)(53)(54). However, the biochemistry of these interactions more closely mimics the cutaneous HPV E6 and E7 oncoproteins (52)(53)(54)(55)(56)(57)(58). MmuPV1 E7 does interact with the tumor suppressor pRB and PTPN14 (53,54). MmuPV1 E7's interaction with pRB is not facilitated through an LXCXE motif but instead through amino acids in the C-terminus of MmuPV1 E7, which has been observed in HPV E7s that lack LXCXE motifs and canine papillomavirus (53,55). While the interaction between MmuPV1 E7 and pRB is distinctly different from the "high-risk" HPVs, previous work has shown that the interaction between MmuPV1 E7 and pRB plays a key role in pathogenesis (33). However, the impact of MmuPV1 E7 on pRB biology has remained elusive. MmuPV1 E7 interacts with the tumor suppressor PTPN14 using similar amino acids as "high-risk" HPV E7, but it remains unclear if MmuPV1 E7 promotes degradation of PTPN14 like the "high-risk" HPV E7s (54). Previous work has shown that MmuPV1 E7, like HPV E7, promotes inhibition of keratinocyte differentiation through its interaction with PTPN14 (22,23,35,54). While recent work has made significant progress in understanding the role of MmuPV1 E7 and the cellular interacting partners that play key roles in pathogenesis, our understanding of the impact of MmuPV1 E7 on keratinocyte biology remains limited to inhibition of differentiation.
In this study, we utilized a multiomics approach to identify the impact on cellu lar homeostasis in mouse keratinocytes (MKs) in the presence of MmuPV1 E7. Spe cifically, we performed quantitative mass spectrometry for proteomics analysis and RNA-seq for transcriptomic analysis. Both our RNA-seq and quantitative mass spec trometry analyses revealed that MmuPV1 E7 significantly alters cellular homeostasis and promotes phenotypes that have been associated with HPV infection and cancer, including "high-risk" HPV-associated disease. We observe a mild but significant increase in proliferation-associated genes. However, we do not see an increase in the proliferation rate of MmuPV1 E7-expressing MKs, nor did we see a significant increase in classic E2F-responsive gene expression. Interestingly, we observed a significant increase in the transcription of stress keratins, including keratin 17 (K17), which has been shown to reduce T cell recruitment to MmuPV1-infected warts (59). We observed a trend in increased protein steady-state levels of stress keratins 6a, 6b, and 16, whereas K17 showed a significant increase in the steady-state protein levels. Finally, we observed a significant increase in the activation of mTOR signaling in MmuPV1 E7-expressing MKs but not MmuPV1 E6-expressing MKs. We were able to determine that the increase in mTOR signaling is potentially due to increased sensitivity to epidermal growth factor receptor (EGFR) signaling as MmuPV1 E7-expressing MKs have an elevated response to EGF stimulation following EGF deprivation. Based upon these results, we conclude that MmuPV1 E7, like MmuPV1 E6, does significantly contribute to phenotypes associated with papillomavirus infection and pathogenesis, including "high-risk" HPV infection.
## RESULTS
## MmuPV1 E7 alters cellular homeostasis in mouse keratinocytes
Our previous work has found that MmuPV1 E7 interacts with known cellular targets of HPV E7, including pRB and PTPN14 (53,54). Through these studies, we found that both the interactions with pRB and PTPN14 contribute to viral pathogenesis (53,54). However, our understanding of the impact of MmuPV1 E7 on cellular homeostasis and pathogenesis is limited to MmuPV1 E7's interaction with PTPN14 restricting differentiation. To address this gap in knowledge, we performed two different unbiased omics analyses, RNA-seq (transcriptomics) and quantitative mass spectrometry (proteomics), to gain a broad understanding of the impact of MmuPV1 E7 on cellular homeostasis. For this analysis, we generated mouse keratinocyte cell strains (MKs) from neonate mouse skin (days 1-4 postpartum). Following the establishment of MKs, we transduced early-passage MKs (passage <5) with a retroviral vector that encoded the MmuPV1 E7 gene (pLXSP mE7) or an empty control vector, and cells were selected for MmuPV1 E7 expression using puromycin. MmuPV1 E7 expression was validated using reverse transcription-PCR (RT-PCR), as no antibody for immunoblot analysis is available for MmuPV1 E7. Following confirmation of MmuPV1 E7 expression, we subjected early-pas sage MmuPV1 E7-expressing and vector control MKs (passages 2 and 3) to transcriptomic and proteomic analysis. Our analyses found that MmuPV1 E7 significantly altered cellular homeostasis, with greater than 700 genes significantly altered transcriptionally (Table S1) and greater than 170 proteins (Table S2) significantly altered at the steady-state level (Table 1). One gene, Cth, and its corresponding protein, cystathionine gamma-lyase, were found to be conflicting between the two analyses (Table S3). However, MmuPV1 E7's impact on cellular homeostasis is less significant compared to MmuPV1 E6, both transcriptionally and on protein steady state (Table 1) (57). We generated a volcano plot to highlight the top 10 most significantly upregulated and top 10 most significantly downregulated proteins from our proteomics analysis (Fig. 1A). Several of these proteins are implicated in cervical cancer, including Ptgs2 (up), Hmox1 (up), Mkl1 (down), and Arhgap5 (down) (60)(61)(62)(63). We performed a similar analysis on our RNA-seq analysis and identified several genes that are also implicated in cervical cancer and HPV infection, including Krt16 (up), Krt6a (up), Hmox1 (up), and Krt15 (down) (Fig. 1B) (62,64,65). These observations suggest that MmuPV1 E7 promotes cellular changes that are observed in HPV+ cancers, including "high-risk" HPV+ cancers.
## MmuPV1 E7 promotes phenotypes associated with "high-risk" HPV infection
As the genes mentioned above were selected from our analysis, we wanted to determine, in an unbiased manner, the impact of MmuPV1 E7 on cellular signaling cascades and cellular biology. To determine the impact of MmuPV1 E7 on cellular protein-interacting networks, we performed STRING analysis on both our RNA-seq and proteomic analyses. Significantly upregulated [log2(FC) > 1 and P adj < 0.05] genes and downregulated [log2(FC) < -1 and P adj < 0.05] genes in both our RNA-seq and proteomic data were uploaded to the STRING database (https://string-db.org/). STRING analysis was performed individually on upregulated and downregulated gene sets. To generate interactive networks, STRING analysis settings were set to include the full STRING network, removed text mining as a source for active interaction, and high confidence to minimize less relevant interactions. STRING analysis of our RNA-seq data showed that MmuPV1 E7 increases the expression of CXCR2 ligands, tubulins, and stress keratins (Fig. 2A) and reduces expression of tumor growth factor beta (TGF-β) signaling proteins, collagens, WNT signaling proteins, and metabolism-related proteins (Fig. 2B). We performed an identical analysis on our proteomics data. MmuPV1 E7 increased the abundance of stress keratin and EGF-signaling proteins (Fig. 2C) and reduced the abundance of mitotic spindle, immunoproteasome, collagen, apoptosis, and WNT signaling proteins (Fig. 2D). Our analysis revealed synergism between our RNA-seq and proteomic analyses, particularly in stress keratins, WNT signaling, and collagen genes/proteins. Of particular interest, the increase in levels of stress keratins and CXCR2 ligands and the downregulation of immunoproteasome match observations made during "high-risk" HPV infection and have been linked to activities of "high-risk" HPV E5, E6, and E7 (63,(66)(67)(68)(69).
To determine MmuPV1 E7's impact on steady-state protein levels on the various signaling pathways in mouse keratinocytes, we utilized gene ontology (GO) analysis as an unbiased analysis to address this question. The GO analysis of our proteomic data showed that MmuPV1 E7 alters proteins associated with various cellular signal ing pathways (Fig. 3). MmuPV1 E7 expression correlated with an increase in proteins associated with "Keratinization" and "Establishment of Skin Barrier" (Fig. 3A), which was predominantly associated with the increase in the abundance of stress keratins K16 and K6a. This observation correlated with a decrease in protein abundance of genes associated with "Positive Regulation of Mesenchymal to Epithelial Transition" (Fig. 3B). These data suggest that MmuPV1 E7 appears to promote keratinocyte identity in our mouse keratinocyte culture model system.
To determine if the transcriptional changes we observed in our MmuPV1 E7-express ing cell strains could be tied to specific cellular signaling pathways, we performed gene set enrichment analysis (GSEA) on our RNA-seq analysis, as we had performed in our previous publication, and limited our analysis to the Hallmark gene sets, which are curated by the Broad Institute (57,70). We observed several signaling pathways that are up-and downregulated in the presence of MmuPV1 E7 compared to our vector control condition (Fig. 3). We observed significant positive enrichment of PI3K (PI3K-AKT-MTOR Signaling) and mTOR signaling (MTORC1 Signaling) in our GSEA (Fig. 3C). We also observed significant positive enrichment with cellular proliferation signaling pathways (G2M Checkpoint) (Fig. 3C). Additionally, we also observed positive enrichment of transcripts associated with TGF-β signaling (TGF Beta Signaling), which included a significant increase in the inhibitory SMAD SMAD7 [log2(FC) = 1.57 and adjusted P-value < 0.001) (Fig. 3C). We also observed significantly lower enrichment of immunerelated signaling pathways including type 1 and 2 interferon signaling (Interferon Alpha Response and Interferon Gamma Response) (Fig. 3D). We also observed nega tive enrichment of genes associated with WNT signaling (WNT Beta Catenin Signal ing) (Fig. 3D), which is similar to observations made with cutaneous HPV genotypes (71). Consistent with our proteomics analysis, we observed a negative enrichment of epithelial-to-mesenchymal transition genes (Epithelial Mesenchymal Transition) (Fig. 3D), suggesting that MmuPV1 E7 promotes expression of epithelial cell identity. Our collective unbiased analysis of our proteomic and transcriptomic analyses determined that MmuPV1 E7 significantly alters cellular homeostasis on its own and contributes to phenotypes observed during HPV infection.
## MmuPV1 E7 does not increase proliferation rate of mouse keratinocytes
A traditional hallmark of HPV E7 oncoproteins is increasing the expression of E2F transcription factor target genes by inhibiting the tumor suppressor pRB (25). We have previously shown that MmuPV1 E6 upregulates E2F target gene transcription in mouse keratinocytes (57). However, we observed a weak but significant positive enrichment of proliferation genes in MmuPV1 E7 expressing MKs (Fig. 4A through C). The strongest enrichment is in the G2M checkpoint gene set (G2M Checkpoint) (Fig. 4B). Interestingly, we did observe a weak but significant positive enrichment of mitotic spindle (Mitotic Spindle) genes transcriptionally, but our STRING analysis of our proteomics data shows a cluster of mitotic spindle proteins that are decreased in our proteomic analysis (Fig. 2D). To determine the impact of the increased enrichment of proliferation-associated genes in our RNA-seq analysis (Fig. 4A through C), we performed growth curve analysis of MmuPV1 E7-expressing MKs compared to vector control MKs (Fig. 4D andE) for cellular proliferation. We used our short-(4 days) and long-term (18 days) growth experiments that we previously published studying MmuPV1 E6-expressing MKs (57). We observed no significant difference in the growth rate between our MmuPV1 E7 and vector control MKs in either experiment (Fig. 4D andE). Using this data, we calculated the doubling time of MmuPV1 E7-expressing and vector control MKs and observed no difference (Fig. 4F). q-value < 0.05. NES is shown on the x-axis. Gene sets of note include "MYC Targets V2, " "PI3K-AKT-MTOR Signaling, " "MTORC1 Signaling, " "G2M Checkpoint, " and "Unfolded Protein Response. " (D) GSEA was performed on RNA-seq data specifically on downregulated transcripts with log2(FC) < -1 and P adj <0.05. Analysis was limited to the Hallmark gene sets, which are curated by the Broad Institute. The top 10 downregulated gene sets are shown, which have an NES < -1.5 and FDR q-value <0.05. NES is shown on the x-axis. Gene sets of note include "Interferon Alpha Response, " "Interferon Gamma Response, " "WNT Beta Catenin Signaling, " and "Epithelial Mesenchymal Transition. "
Using qRT-PCR, we confirmed expression of E7 in our MmuPV1 E7-expressing MKs with significantly lower signal in our vector control MKs (Fig. 4G). To validate the increased expression of proliferation-associated genes, we performed qRT-PCR using validated primers for E2F-responsive genes that are activated by HPV E7 (MCM2, MCM7, CCNE2, and PCNA) (Fig. 4H through K) (53,57). We did not observe a significant increase in the expression of the E2F-responsive genes (Fig. 4H through K). These results are consistent with previously published results (53). Therefore, our RNA-seq results may suggest a mild positive enrichment of proliferation-associated genes in our MmuPV1 E7-expressing MKs, but the alteration in proliferation genes did not correlate with changes in the proliferation and growth capacity of mouse keratinocytes.
## MmuPV1 E7 increases the expression of stress keratins in mouse keratino cytes
In our RNA-seq analysis, we observed a significant increase in the transcription of stress keratin genes (Krt6a, Krt6b, Krt16, and Krt17) (Table 2). Stress keratins have been previously shown to be increased during MmuPV1 infection in immunocompetent FVB/N mice, with Krt17 playing a key role in inhibiting T cell response (59). However, the mechanism by which MmuPV1 promotes expression of stress keratins, including Krt17, remains unknown. Additionally, we did detect two of the stress keratins in our proteo mics data (Krt6b and Krt16), and only Krt16 showed a significant increase in protein abundance (Table 2). We did not detect Krt6a and Krt17 in our proteomic analysis. To validate the increase in expression of the various stress keratins, we performed immunoblot analysis on vector control and MmuPV1 E7-expressing MKs (Fig. 5). We observed a significant increase in the steady-state levels of Krt17 in MmuPV1 E7-express ing MKs compared to the vector control MKs (Fig. 5). The other stress keratins were slightly but not significantly increased in the MmuPV1 E7-expressing MKs compared to vector control MKs (Fig. 5). These data suggest that MmuPV1 E7 increases the expression of Krt17 and may contribute to elevated expression of the other stress keratins during MmuPV1 infection. Thus, our data provide evidence that MmuPV1 E7 contributes to immune evasion through promoting K17 expression, which has previously been shown to decrease T cell recruitment to MmuPV1-infected warts (59).
## MmuPV1 E7 promotes mTOR signaling in mouse keratinocytes
Previous work has found that HPV and MmuPV1 both elevate EGFR and mTOR signaling in mice (66,67,(72)(73)(74). For "high-risk" HPVs, the literature has shown that E5 and E6 contribute to promotion of EGFR and mTOR signaling in human keratinocytes (66,67,(72)(73)(74). In our previously published MmuPV1 E6 RNA-seq analysis, we found that PI3K signaling (PI3K-AKT-MTOR Signaling) was not significantly positively enriched in our analysis (Fig. 6A) (57). We did see a significant positive enrichment of genes associated with mTOR signaling (MTORC1 Signaling) (Fig. 6B) (57). These results would suggest potential overlap between MmuPV1 E6 and HPV E6. However, we found that several of the genes that drive this positive enrichment are also associated with proliferation, consistent with our previous results (57). We did observe that genes associated with PI3K (PI3K-AKT-MTOR Signaling) and mTOR (MTORC1 Signaling) were both significantly positively enriched in our MmuPV1 E7-expressing MKs compared to vector control MKs (Fig. 6C andD). To determine whether MmuPV1 E6, MmuPV1 E7, or both acti vate the PI3K-AKT-mTOR signaling cascade, we performed immunoblot analysis for a downstream marker for activation of this signaling cascade, phosphorylated S6 kinase (P-S6), comparing MmuPV1 E6-expressing, MmuPV1 E7-expressing, and vector control MKs. Our immunoblot analysis showed that MmuPV1 E7 leads to a significant increase in the steady-state levels of P-S6 kinase relative to total S6 kinase (Fig. 6E andF). In contrast, MmuPV1 E6 expression led to a significant decrease in the steady-state levels of P-S6 kinase relative to total S6 kinase (Fig. 6E andF). These data would suggest that MmuPV1 E7, and not MmuPV1 E6, promotes activation of downstream targets of PI3K and mTOR signaling. Previous work has linked activation of S6 kinase and mTOR E7-expressing and vector control MKs using primers for MmuPV1 E7. Additionally, qRT-PCR was performed on cDNA using primers targeting MCM2 (H), MCM7 (I), CCNE2 (J), and PCNA (K). ΔΔCt was calculated with average and standard error shown.
The Wilcoxon rank-sum test was performed, and P-values are shown.
signaling through EGFR activation during HPV and MmuPV1 infection (67). To determine the impact of MmuPV1 E7 on EGFR signaling in mouse keratinocytes, we performed immunoblot analysis for phosphorylation of AKT (S473) and ERK1/2 (T202/Y204), which are known EGFR signaling phosphorylation sites. We observed a slight but non-significant increase in the steady-state ratio of phosphorylated AKT to total AKT in our MmuPV1 E7-expressing MKs (Fig. 6G andH). We did observe a significant decrease in the levels of phosphorylated ERK1/2 relative to total ERK1/2 in our MmuPV1 E7-expressing MKs (Fig. 6G andI). We observed no difference in the total levels of ERK1/2 (Fig. 6G). Our collective data suggest that MmuPV1 E7 promotes mTOR signaling (elevated P-S6), but upstream markers of EGFR signaling associated with mTOR signaling are not significantly increased.
## MmuPV1 E7 potentiates EGF signaling in mouse keratinocytes
As stated above, we did not observe a significant increase in the steady-state abun dance of markers of EGFR signaling with a trend in promoting phosphorylation of AKT. However, the mouse keratinocytes were grown and maintained in F-media containing ROCK inhibitor. F-media contains exogenous EGF and 5% serum (fetal bovine serum [FBS]). Therefore, the mouse keratinocytes are maintained in a tonic EGFR signaling environment, which could mask any phenotypic changes in EGFR signaling promoted by MmuPV1 E7. To address whether F-media is masking potential phenotypes, we grew mouse keratinocytes in EGF-low F-media by removing the exogenous EGF and lowering the FBS percentage from 5% to 1% (EGF-low). First, we performed a growth analysis of these cells in the EGF-low environment to determine if MmuPV1 E7 increases the proliferation of mouse keratinocytes in this environment. We plated cells and counted the keratinocytes 18 hours later to determine the number of cells plated. The media on the cells was changed to either F-media (without ROCK inhibitor) or EGF-low F-media. After 48 hours, we counted the cells and determined the fold change in growth. We observed no difference in the fold increase in the MmuPV1 E7-expressing MKs compared to the vector control MKs in the EGF-low F-media (Fig. 7A). As expected, we also observed no difference in the growth rate of MmuPV1 E7 mouse keratinocytes compared to vector control mouse keratinocytes in F-media (Fig. 7A). We did observe reduced growth in both MmuPV1 E7-expressing and vector control MKs when EGF levels were minimal (Fig. 7A). These observations suggest that MmuPV1 E7 does not promote the growth of mouse keratinocytes in an EGF-restrictive environment. While MmuPV1 E7 may not promote the growth of mouse keratinocytes when EGF is restricted, we were interested in determining whether MmuPV1 E7 potentiated EGFR signaling when EGF was added back to the cell culture media. To do this, we grew our mouse keratinocytes in F-media. When the cells were 24 hours prior to confluency, we washed them 3× with phosphate-buffered saline (PBS) to remove F-media and grew cells in EGF-low F-media for 18 hours to allow the EGFR signaling to stabilize in the MKs. Following stabilization in EGF-low F-media, we removed the feeders using trypsin the following morning. We stimulated vector control or MmuPV1 E7-expressing MKs with 10 ng/mL EGF for 5 minutes. After stimulation, cells were washed 2× with cold PBS to slow signaling events and collected in RIPA buffer. Lysates were subjected to immunoblot analysis using antibodies against AKT, P-AKT, and GAPDH. We observed an increase in P-AKT levels in our vector control cells following stimulation with EGF (Fig. 7B). Similarly, we also observed an increase in P-AKT in the MmuPV1 E7-expressing MKs with no consistent impact on total AKT levels (Fig. 7B; Fig. S1A). To determine the impact of MmuPV1 E7 on potentiating EGF signaling in mouse keratinocytes, we quantified the western blot by normalizing the levels of P-AKT to total AKT in the EGF-low and the EGF-stim groups. We observed that MmuPV1 E7 increased the ability of mouse keratinocytes to respond to EGF stimulation following 18 hours of growth in minimal EGF media (Fig. 7C). These data suggest that MmuPV1 E7 potentiates EGFR signaling, which likely contributes to increased mTOR signaling in mouse keratinocytes.
## Interactive network analysis shows connections between MmuPV1 E7, E7 interacting partners, and mTOR signaling
An interactome network analysis was performed to connect the differentially expressed genes (DEGs) in our RNA-seq analysis and the interactors of MmuPV1 E7 identified by the Munger lab (Tufts University) (53). The prize-collecting Steiner forest (PCSF) algorithm was used to find the pathways connecting the interactors and DEGs through the minimum number of protein-protein interactions and bridging nodes (also known as Steiner nodes). A key feature that we observed in our network analysis was an enrich ment of cell division and cell cycle proteins in the MmuPV1 E7 interactome data (yellow nodes) (Fig. 8). On the other hand, the DEGs were enriched for cell differentiation, cell motility, neurogenesis, and programmed cell death (pink nodes) (Fig. 8). Interestingly, the Gray Steiner (bridging) nodes in this analysis were enriched for genes associated with cell division, anoikis, mTORC1 signaling, T cell-mediated immunity, and chromatin remodel ing (green nodes) (Fig. 8). The enrichment in mTOR signaling among the bridging nodes suggests that MmuPV1 E7 may directly impact mTOR signaling in mouse keratinocytes through its interacting partners instead of an indirect mechanism (bolded nodes). The network map revealed several distinct clusters within the analysis, including clusters focused on mTOR signaling, anaphase-promoting complex (APC), and pRB (Fig. 8). The pRB cluster is particularly interesting, as there are several Steiner nodes that bridge E7 interacting partners, including CDK6, CDK9, CCNE1, and E2F1. The mTOR signaling cluster includes the Steiner nodes MTOR and RPTOR connected to the E7 interacting partners RPS6KB1 and MAPKAP1, providing evidence that MmuPV1 E7 may directly impact mTOR signaling through its interacting partners. Moreover, this connection between MmuPV1 E7 interacting partners and DEGs in MmuPV1 E7-expressing mouse keratinocytes also highlights the concordance among our transcriptomic, proteomic, and confirmation studies. The Wilcoxon rank-sum test was performed, with P-value shown.
## DISCUSSION
In this study, we complement our previous work on understanding the role of MmuPV1 E6 in promoting phenotypes associated with the hallmarks of cancers by performing multiomics analysis of MmuPV1 E7-expressing MKs (57). Our collective observations suggest that MmuPV1 E7 does alter cellular homeostasis and promotes phenotypes commonly observed in cancer-causing "high-risk" HPV infection. Previous studies of MmuPV1 E7 have identified at least two interacting partners, pRB and PTPN14, that are shared between HPV E7 and MmuPV1 E7 (53,54). However, the impact of these interactions on cellular homeostasis is understudied, and studies were limited to phenotypes associated with biological impacts of the "high-risk" HPV E7 oncogenes. To remedy this gap in knowledge, we performed a series of omics analyses examining the impact of MmuPV1 E7 on keratinocyte homeostasis. We utilized our ability to perform proteomic (quantitative mass spec) and transcriptomic (RNA-seq) analyses on mouse keratinocytes that stably express MmuPV1 E7 compared to vector control MKs. MmuPV1 E7 caused significant changes to the cellular transcriptome and proteome, including several changes that are implicated in HPV-associated cancers, including cervical and head and neck cancer (Fig. 1). Notable genes include Ptgs2 (up), Hmox1 (up), and Arhgap18 (down) (60-63, 75, 76). Importantly, little is known about the impact of HPV infection on these genes, and it remains unknown if the alterations of Ptgs2, Hmox1, and Arhgap18 are connected to activities of HPV E6 or E7. These observations raise new questions about how MmuPV1 E7 promotes cellular alterations that are related to "high-risk" HPV infections and could lead to the identification of new activities of MmuPV1 and HPV E7 that could be implicated in HPV-associated mucosal cancers and infections.
Overall, MmuPV1 E7 had a smaller phenotype compared to that of MmuPV1 E6 (Table 1). We also found that MmuPV1 E7 had a smaller impact on the differential expression of genes and proteins in our transcriptomic and proteomic analyses and in our GSEA or GO analysis (Fig. 6A through D). We did find that MmuPV1 E7 is expressed at a much lower level compared to MmuPV1 E6 in the RNA-seq analysis (data not shown). This observation suggests that high levels of MmuPV1 E7 could be toxic to cells, and the lower-level expression is what was selected for in our mouse keratinocyte system. However, we did find that our RNA-seq and proteomic analyses were consistent for genes that were detected in both analyses. We were able to determine that 35 genes/ proteins were differentially expressed in both of our analyses. Of those 35, 34 genes/ proteins show similar results in both the RNA-seq and proteomic analyses. The sole gene/protein that was divergent was Cth (cystathionine gamma-lyase), which showed increased gene expression but reduced protein levels. This would suggest that MmuPV1 E7 promotes reduced steady-state levels of the protein, and future studies could address this difference. Interestingly, two of our most differentially expressed genes/proteins-Hmox1 and Ptgs2-showed an increase in gene expression and protein abundance in our analysis, thus further strengthening the need for future studies to determine the impact of MmuPV1 E7 and HPV E6/E7 on their expression. In addition to these changes in transcription and protein abundance, we found convergence in our GO analysis and GSEA. Specifically, we observed similar enrichment in gene sets that share biological properties. Examples of this include the increases in "Cellular Response to Oxidative Stress" and "Desmosome" in GO analysis and "MTORC1 Signaling" and "Apical Junction" in GSEA, respectively. We also observed biologically similar pathways being negatively enriched in our analysis. Examples of shared negatively enriched pathways include "Carbohydrate Metabolism" and "Defense Response to Gram-Positive Bacterium" in GO analysis and "Oxidative Phosphorylation" and "Interferon Alpha/Gamma Response" in GSEA, respectively. These results suggest that while MmuPV1 E7 may have a smaller impact on cellular homeostasis, there is consistency in the alteration of biological processes both transcriptionally and at the protein level.
It is important to note that we are studying MmuPV1 E7 in isolation and that both MmuPV1 E6 and E7 are required for pathogenesis in vivo (52,53). The interplay between MmuPV1 E7 and E6 remains understudied. Future studies are needed to address the interplay between MmuPV1 E6 and E7 and how they promote the activities of one another. Additionally, there could be cellular alterations that are promoted only in the presence of both MmuPV1 E6 and E7, which have been missed in our studies on MmuPV1 E6 and E7 individually (57). Potential studies could perform transcriptomic and proteomic analyses on MKs that co-express MmuPV1 E6 and E7 to address this question. Additionally, future studies should include examination of the impact of MmuPV1 E6 and E7 co-expression in the context of the entire MmuPV1 genome in MKs to better reflect what is occurring in MmuPV1-infected mice.
Our studies further strengthened the argument that MmuPV1 E7 is not the major driver of proliferation in mouse keratinocytes (Fig. 6) (53). We did observe slight increases in expression of genes related to proliferation in our GSEA, including a positive enrichment of "E2F-Responsive Genes, " "G2M Checkpoint, " and "Mitotic Spindle" (Fig. 6). However, we did not observe a significant increase in the proliferation capacity of MmuPV1 E7-expressing MKs nor an increase in expression of classic E2F-responsive genes. These data suggest that MmuPV1 E7 may promote expression of proliferation genes using an alternative mechanism outside of E2F activation, and further studies are needed to address this observation. Interestingly, STRING analysis of our proteomics data revealed that there is a negative enrichment of several proteins in the mitotic spindle complex, which is in contrast to our RNA-seq analysis, which shows an enrichment of mitotic spindle-associated gene expression. This is the first case of a difference between our transcriptomic and proteomic analyses. However, MmuPV1 E7 does interact with several APC proteins, which were identified in the affinity purification/mass spectrometry analysis performed by the Munger lab (Tufts University) (Fig. 8) (53). This observation suggests that MmuPV1 E7 may impact the protein abundance of the APC complex and other cellular proteins that interact with this complex instead of acting on the transcrip tion of these proteins. The interaction between MmuPV1 E7 and APC complex proteins has yet to be confirmed and/or verified in the appropriate cell lines. Further studies are needed to determine the impact of these interactions in phenotypes associated with MmuPV1 E7 expression.
The impact of MmuPV1 E7's interaction with pRB still remains unclear, and our study further strengthens the argument that MmuPV1 E7's interaction with pRB does not promote traditional phenotypes. However, the interactome network analysis did reveal a distinct pRB cluster of DEGs, E7 interacting partners, and Steiner nodes with pRB appearing as the central node in this cluster (Fig. 8). Of particular interest are the Steiner ray nodes that bridge pRB and a subset of DEGs and other MmuPV1 E7 interacting partners. Steiner nodes of interest include PPP1CA (PP1 catalytic subunit), CDK6, CDK9, CCNE1 (cyclin E), CCND1 (cyclin D1), CDKN1A (p16), and E2F1. The E2F1 Steiner node does appear to connect pRB with TOPBP1 (E7 interacting protein) in our analysis. This is particularly interesting, as TOPBP1 has been shown to be required for genome replica tion and viral E2 activity (77)(78)(79). This observation suggests a potential role for MmuPV1 E7 in viral genome replication and maintenance potentially through its pRB interaction. In addition to this, we do not observe any other E2F transcription factor within this cluster, which potentially suggests that there is E2F1 specificity in the MmuPV1 E7-pRB interaction. The literature has shown that pRB does have a specific interaction with E2F1 outside of the other E2F transcription factors (80)(81)(82). The specific interaction between pRB and E2F1 has been shown to be a non-canonical activity of pRB and plays a role in EZH2 activity and DNA damage (80)(81)(82)(83). This result provides more evidence that MmuPV1 E7's interaction with pRB does appear to play a distinct role outside of the traditional pRB-E2F axis. Further studies are needed to better address these observations and characterize the role of MmuPV1 E7's interaction with pRB.
Our proteomic and transcriptomic analysis did reveal that MmuPV1 E7 promotes a number of phenotypes that are associated with modulation of the immune response, including repression of the type 1 and 2 interferon response (Fig. 3), increased expression of CXCR2 ligands (Fig. 2), and increased expression of stress keratins, in particular K17 (Fig. 5) (59,69,(84)(85)(86)(87)(88)(89)(90)(91). All of these phenotypes are well characterized for "high-risk" HPV infection and, to a lesser extent, cutaneous and "low-risk" HPV infections (84)(85)(86)(87)(88)(89)(90). The increase in CXCR2 ligand expression was observed in the HPV16 E6 and E7 transgenic animal model system and was linked to HPV16 E7 expression (69). Additionally, increased K17 expression has been observed in the HPV16 transgenic animal model system and during an MmuPV1 infection of immunocompetent mice (59,91). The increase in K17 expression is linked to repression of the T cell response and is a key mechanism by which papillomavirus infection promotes immune evasion in vivo (59). We were able to confirm that MmuPV1 E7 expression does increase the abundance of K17 in mouse keratinocytes transcriptionally (RNA-seq) (Table 2) and at the protein level (Fig. 5) (59). We still do not understand the mechanism by which MmuPV1 E7 may promote K17 expression, but previous studies in HPV16 transgenic mice revealed a connection between HPV E7's interaction with pRB and K17 expression (59,91). The increase in K17 by MmuPV1 E7 raises new questions about the impact of MmuPV1 E7 on pRB and the potential overlap in activities between HPV and MmuPV1 E7. Future studies comparing the MmuPV1 E7-pRB binding mutant and WT MmuPV1 E7 would determine the impact of MmuPV1 E7 on K17 expression.
We have observed that MmuPV1 E7 promotes mTOR signaling in the mouse keratinocyte model system at the transcriptional level and increases the abundance of markers of mTOR signaling, such as phosphorylation of S6 kinase (Fig. 6). Previous work has linked the increase in mTOR signaling to activities of HPV E6/E7 by increasing levels of IQGAP1, which potentiates EGFR signaling and leads to increased activation of PI3K-AKT-mTOR signaling (67). We did not observe an increase in the steady-state levels of phosphorylation sites in AKT and ERK, which are known targets of EGFR signaling (Fig. 6). However, MmuPV1 E7 expression does lead to increased sensitivity of EGFR to EGF stimulation following EGF deprivation (Fig. 7). The ability of MmuPV1 E7 to potentiate EGFR and mTOR signaling is promoted by "high-risk" HPV E6 and E5 oncogenes (66,67,(72)(73)(74). MmuPV1 infection has been found to elevate the protein abundance of IQGAP1 in mice (67). However, this study did not validate this observation that MmuPV1 E7 was responsible for promoting IQGAP1 expression. We do see increased levels of IQGAP1 in our proteomics of MmuPV1 E7-expressing MKs, but we did not confirm this observation in our analysis. Future studies would be needed to connect MmuPV1 E7 to elevated IQGAP1 levels. Importantly, our interactome network analysis did find an enrichment of mTORC1 signaling in the Steiner nodes, which are nodes that parsimoniously bridge between E7 cellular interacting proteins and differentially expressed genes (Fig. 8). Of note, MTOR and RPTOR genes are Steiner nodes that are connected to the MmuPV1 E7 interacting partners MAPKAP1 and RPS6KB1 (S6 kinase). MAPKAP1 is a subunit of mTOR complex 2 that acts as a scaffold protein and assists with substrate specificity of the mTOR catalytic subunit. It promotes metastasis, proliferation, and invasion in some cancers (68,92,93). RPS6KB1 is a ribosomal serine/threonine kinase that induces cell growth and proliferation and potentiates survival of some cancer subtypes (94)(95)(96)(97)(98). The interaction between MmuPV1 E7 and RPS6KB1 or MAPKAP1 would need to be validated in appropriate cell lines to determine if these are direct interactions. Further studies are needed to determine the impact of MmuPV1 E7 on MAPKAP1, RPS6KB1, and mTOR signaling.
It should be noted that our proteomics data found "Positive Regulation Of Macroau tophagy" to be positively enriched (Fig. 3A). This is an interesting observation, as mTOR is an inhibitor of autophagy. This enrichment of autophagy-related proteins is driven by a small subset of proteins that were detected in our proteomic analysis, including GPSM1, SESN2, and SNX18. This is not entirely unexpected, as other studies have shown that HPV also impacts autophagy. Studies on HPV16 E7 have found that E7 inhibits autophagy in head and neck cancer, which increases the sensitivity of these cells to radiation (99,100). This is in contrast to our own results, where we observe an enrichment of autophagy-related proteins. However, it remains unclear from our analysis if autophagy is activated or inhibited in the presence of MmuPV1 E7. Future studies are needed to address this question.
In conclusion, the data generated in this study have provided new and excit ing evidence that MmuPV1 E7 promotes phenotypes in MKs that are associated with "high-risk" HPV infection. While the biochemistry and molecular mechanisms of MmuPV1 E7 mostly mimic observations by cutaneous HPVs, our analysis challenges this hypothesis (53)(54)(55). The ability of MmuPV1 to promote infection and disease development, including cancer, at mucosal and cutaneous sites makes our observations not entirely unexpected (45)(46)(47)(48)(49). The categorization of MmuPV1 as solely a mimic of cutaneous HPVs may not entirely reflect the complex nature of the MmuPV1 model system. While interacting partners and biochemical activities of MmuPV1 E7 and E6 more reflect their cutaneous counterparts, the increasing evidence showing phenotypes commonly seen in mucosal disease and not cutaneous HPV disease suggests that MmuPV1 should not be categorized as one or the other (54,56). However, future studies further examining the role of MmuPV1 E7 in MmuPV1-associated mucosal disease are needed to shore up these observations, and comparative studies between the MmuPV1 and HPV model systems are needed to better understand the utility of the MmuPV1 model system as a model for HPV-associated disease, including mucosal disease.
## MATERIALS AND METHODS
## Cells
NIH 3T3 murine fibroblasts were a gift from Dr. Paul Lambert (obtained through ATCC) and grown in Dulbecco's modified Eagle medium (DMEM) containing 10% bovine calf serum containing Pen/Strep (Gibco). 293FT were a gift from Dr. Paul Lambert (obtained through ATCC) and grown in DMEM containing 10% FBS containing Pen/Strep antibiotic with 200 µg/mL G418. Mouse keratinocytes were isolated from the skin of four different neonate mice on the FVB/N background as previously described to create biological quadruplicate (53,54,57). Briefly, neonate mice were euthanized and soaked in betadine for 10 minutes. Following betadine treatment, euthanized pups were soaked in PBS containing 10% antibiotics twice before soaking in 70% ethanol. Skins were isolated and then washed in PBS containing 10% antibiotics. Skin tissue was then soaked in 0.25% trypsin-EDTA overnight at 4°C. The following morning, the epidermis and dermis were separated using sterilized forceps, minced with a single-edge razor blade, and then stirred for 1 hour at 37°C to generate a cell suspension in F-media. The cell suspension was strained using a 0.7 mm membrane (Corning) and cultured in F-media containing 10 mM ROCK inhibitor (Y-27632) in the presence of mitomycin C (MMC)-treated J2 3T3 fibroblasts (57). Early-passage mouse keratinocytes were transduced in keratinocyte serum-free medium (KSFM) (Gibco) with retroviruses encoding MmuPV1 E7 or vector control (pLXSP) (a gift from Dr. Karl Munger) (57). Transduced cells were put under selection 48 hours following transduction with 1 µg/mL of puromycin (Gibco). Follow ing selection, MKs were cultured in F-media containing ROCK inhibitor (Y-27632) with 1 µg/mL of puromycin.
## Plasmids
pLXSP MmuPV1 E7 (mE7) and pLXSP were gifts from Dr. Karl Munger (Tufts University). pCL-10A1 (Fisher Scientific) was used for packaging of pLXSP retroviral vectors and was co-transfected into 293FT cells with pLXSP retroviral vectors (57).
## Retrovirus production
293FT cells were plated into 60 mm tissue culture dishes 24 hours prior to transfection. Plated 293FT cells were changed to DMEM with 10% FBS without antibiotics. 293FT cells were transfected with pLXSP or pLXSP-MmuPV1 E7 and pCL-10A1 vectors using the Lipofectamine 2000 system (Invitrogen) following the manufacturer's instructions. Twenty-four hours post-transfection, 293FT cells were switched to 3 mL of KSFM, which was used for transductions. Forty-eight hours post-transfection, the supernatant was collected, centrifuged to remove cellular debris, and filtered using a 0.2 µm sterile filter (Thermo Fisher). The cleared supernatant was used for retroviral transductions.
## RNA isolation and RT-PCR
Following selection and establishment of MmuPV1 E7-expressing MKs, cells were collected and lysed in TRIzol Reagent (Invitrogen). RNA was isolated using the Direct-zol RNA MiniPrep Kit (Zymo Research) as we have previously published (57). cDNA was produced using RNA isolated from MKs using the QuantiTect Reverse Transcription Kit (Qiagen) using the manufacturer's protocol, as we have previously described (57). Briefly, we treated RNA with DNase to eliminate cellular DNA, and DNase-treated RNA was subjected to the RT reaction as described in the manufacturer's protocol. cDNA generated was used to perform PCR for MmuPV1 E7 and GAPDH or PGK1 (mouse-spe cific) as internal controls using the primers listed in Table 3. PCRs were run on a 2% agarose gel, which was stained with ethidium bromide, and imaged using the Chemi Doc Imaging System (Bio-Rad).
RNA isolated from MmuPV1 E7-expressing cells was used for RNA-seq analysis as described below. The remaining RNA was used to generate cDNA for qRT-PCRs for AKTand E2F-responsive genes, as previously described (57). cDNA was generated from RNA isolated from cells as described above. For qRT-PCR, SYBR Green (Bio-Rad) was used per the manufacturer's protocol. Primers used for qRT-PCR analysis are described in Table 3.
## Cell counting assays
We utilized our previously published methods for both the long-and short-term growth analyses (57). For the short-term growth analysis, we plated 0.25 × 10 5 cells into four wells of a six-well plate that contained MMC-treated J2 3T3s. Cells were then counted daily for 4 days. At each time point, feeder cells were removed using 0.05% trypsin-EDTA (Gibco). Following removal of feeders, keratinocytes were washed with Dulbecco's phosphate-buffered saline (DPBS) and then trypsinized. Cells were resuspended in a total volume of 1 mL and then counted using trypan blue staining to determine live cells on a hemocytometer. Average cell number was plotted. For long-term growth analysis, we plated 1 × 10 5 cells into a 60 mm dish containing MMC-treated J2 3T3s. Every 3 days, cells were counted. Briefly, plates were treated with 0.05% trypsin-EDTA to remove feeder cells. Following the removal of feeder cells, keratinocytes were washed with DPBS and then trypsinized. Cells were then resuspended in a total volume of 1.5 mL and counted using trypan blue staining to determine live cells on a hemocytometer. Total cell number was determined by multiplying the cell concentration by 1.5. At each time point, the total fold increase in cell number was determined, and the average cumulative fold increase was plotted. Cells were maintained in F-media without ROCK inhibitor for all the counting assays. All assays were performed in biological quadruplicate. For our MK model, a biological replicate is a population of MKs isolated from a single mouse and grown in isolation. Therefore, we generated MmuPV1 E7-expressing MKs from cells isolated from four individual mice, with vector control cells generated from the same population.
## Primer
Forward Reverse
Forward and reverse primers are shown for each qRT-PCR that was performed.
$$MmuPV1 E7 5´-GCGGGCAGACAAAGCTAAGA-3´5´-GCGACACTGTTCTCCGGTTC-3Ḿ CM2 5´-CGGAGTATGCGCAAGACTTT-3´5´-GCCACCAACTGCTTCAGTAT-3Ḿ CM7 5´-GAGGCCAGCAGATGTGATATT-3´5´-GGTGTGAAGCCACGAGATATG-3Ć CNE2 5´-ATTTGGCTTTGCTGAATGAAGT-3´5´-CAGTACTCTTTGGTGGTGTCATA-3Ṕ CNA 5´-GTTGTCACAAACAAGTAATGTGGAT-3´5´-CTCAGAAACGTTAGGTGAA-3Á KT1 5´-GGACTACTTGCACTCCGAGAAG-3´5´-CATAGTGGCACCGTCCTTGATC-3Ṕ GK1 5´-GATGCTTTCCGAGCCTCACTGT-3´5´-ACCAGCCTTCTGTGGCAGATTC-3Ǵ APDH 5´-GGAGAGTGTTTCCTCGTCCC-3´5´-ACTGTGCCGTTGAATTTGCC-3á$$
## Low-EGF growth assay
Cells were plated at 0.5 × 10 5 in three wells of a six-well plate containing feeders. After 18 hours, cells were counted to determine the initial number of cells plated at the time of media switch to low-EGF F-media. The other two wells were rinsed twice with 1 mL PBS. One well was treated with F-media, and the other had F-media with reduced FBS (1%) and without epidermal growth factor (low-EGF F-media). After 48 hours cells were counted as described above in 4 day cell counting. Fold increase in cell number was determined following counting, and average fold increase in cell number was plotted.
## EGF-stim
A total of 1 × 10 5 cells were plated in F-media on three 60 mm dishes and grown to near confluency. Once near confluency, two plates were washed three times with 1 mL PBS. Three milliliters of low-EGF F-media was added to each plate and allowed to grow overnight. Feeder cells were removed from the other plate using 0.05% trypsin-EDTA (Gibco) and washed twice with 1 mL PBS. One hundred fifty microliters of RIPA buffer containing 1:100 Phosphatase cocktail 2 and 3 (Millipore Sigma) and 1:7 cOmplete, Mini, EDTA-free Protease Inhibitor (Millipore Sigma), per the manufacturer's protocol, was added to each plate, and cells were scraped and collected. The following morning, feeder cells were removed from both plates and rinsed with 1 mL PBS. One plate was harvested with the RIPA mix. The other plate was treated with 3 mL low-EGF F-media spiked with 10 ng/mL EGF. Cells were incubated at 37°C for 5 minutes and immediately placed on ice. Cells were washed twice with 1 mL cold DPBS and lysed with 150 µL of RIPA mix. Lysates were subjected to immunoblot analysis as described below.
## Immunoblotting
Immunoblotting was performed as described previously (57). Briefly, confluent samples were collected and lysed in RIPA buffer containing phosphatase cocktails and protease inhibitors, as described above. Protein concentration was determined within the lysates using Bradford assay (Bio-Rad) per the manufacturer's protocol. Equivalent amounts of protein were separated in sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. Membranes were blocked in 5% milk or 5% bovine serum albumin (BSA) (for phosphorylated antibodies) for 1 hour and then incubated overnight with the appropriate primary antibodies diluted in 5% milk or 5% BSA (for phosphorylated antibodies) in 1× PBS and 0.1% Tween 20 (PBS-T). The primary antibodies used are described in Table 4. After being washed, the membranes were incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies (Rockland Immunology) in 5% milk in PBS-T for 1 hour at room temperature and then washed again. Bound antibodies were visualized using enhanced chemilumi nescent reagent (Thermo Fisher) according to the manufacturer's instructions.
## Proteomic analysis
Once the expression of MmuPV1 E7 was confirmed, frozen cell pellets were subjected to proteomic analysis, specifically quantitative mass spectrometry. Frozen keratinocyte cell pellets were resuspended in lysis buffer (6 M guanidine hydrochloride, 100 mM Tris, pH 8) and probe-sonicated (Misonix) until homogenized. Proteins were then precipitated by adding methanol to the solution to 90% and centrifuging at 10,000 × g for 5 minutes. The supernatant was discarded, and the protein pellets were resuspended in digestion buffer [8 M urea, 10 mM tris(2-carboxyethyl) phosphine (TCEP), 40 mM chloroacetic acid (CAA), 100 mM Tris] and sonicated in a bath sonicator for 5 minutes at 10°C with a program of 10 seconds off/20 seconds on (Qsonica). Endoproteinase Lys-C was added in a 100:1 protein:enzyme ratio, and the samples were incubated at room temperature for 4 hours. The samples were then diluted to 1.5 M urea with 100 mM Tris, and trypsin (Promega) was added in a 50:1 protein:enzyme ratio. After overnight incubation at room temperature, the resulting peptides were acidified to pH 2 with trifluoroacetic acid(TFA), desalted with Strata-X Polymeric solid-phase extraction cartridges (Phenomenex), and dried under vacuum. Each dried peptide sample was resuspended in 0.2% formic acid and loaded onto a 75 µm ID × 360 µm OD capillary column (New Objective) that was packed in-house with 1.7 µm BEH C18 particles (Waters). Chromatographic separations were performed with a Dionex UltiMate 3000 nano HPLC system (Thermo Scientific). The peptides were loaded in 100% A (0.2% formic acid in water) and eluted with increasing % B (0.2% formic acid in 80% acetonitrile [ACN]) over a 90 minute gradient. Mass spectrometric detection was performed with an Orbitrap Eclipse (Thermo Scientific), with MS1 scans taken in the Orbitrap (240,000 resolution, 300-1,350 m/z scan range, 50 ms maximum injection time, and 1 × 10 6 automatic gain control [AGC] target) and MS2 scans in the ion trap (turbo mode, 0.5 m/z isolation width, 150-1,350 m/z scan range, 14 ms maximum injection time, and AGC target of 3 × 10 4 ).
Mass spectrometry raw files were processed with MaxQuant (version 1.5.2.8) and searched against a database of reviewed mouse proteins plus isoforms (downloaded from UniProt on 12 September 2021). Default parameters were used. Cysteine carbami domethylation and methionine oxidation were set to fixed and variable modifications, respectively. The "proteinGroups.txt" file was processed by omitting reverse sequences, sequences only identified by site, and contaminants. The data were log2-transformed, and proteins were removed if they were not observed in at least 70% of the sam ples. Remaining missing values were imputed using Perseus (version 1.6.0.7). Statistical comparisons between groups were performed with two-tailed Student's t-tests. Volcano and enrichment plots were made using RStudio (101) with the ggplot2 (102) and ggrepel (103) packages.
## RNA-seq analysis
Following confirmation of expression of MmuPV1 E6 or E7, cells were lysed in TRIzol Reagent (Invitrogen) and total RNA isolated with the Direct-zol RNA MiniPrep Kit (Zymo Research). RNA quality was assessed with a 4200 TapeStation (Agilent) through the UW-Madison Biotechnology Center. Ribo-depletion, library preparation, and RNA-seq on a NovaSeq 6000 was performed by the Oklahoma Medical Research Foundation Clinical Genomics Center (Oklahoma City, OK). A hybrid index was created by the addition of MmuPV1 E7 sequences to GRCm39. Sequencing reads were aligned to the hybrid index using Docker on HTCondor through the UW-Madison Center for High Throughput Computing using STAR 2.7.6a (104,105). Read summarization was performed with featureCounts v.1.6.3 and Ensembl annotation release 105. Differential expression analysis was performed with DESeq2 v.1.34.0. Gene set enrichment analysis was performed with GSEA v.4.3.2 and the Hallmark Gene Set Collection from MSigDB (70,106,107). Volcano and enrichment plots were made using RStudio (101) with the ggplot2 (102) and ggrepel (103) packages.
## STRING analysis
From the RNA-seq and quantitative mass spectrometry analysis, the list of genes/proteins that met the requirement of log2(FC) > 1 or log2(FC) < -1 and P adj <0.05 was generated. Each list of genes was then separated into upregulated [log2(FC) > 1] or downregulated [log2(FC) < -1] and uploaded to the STRING database (string-db.org). Following the uploading of gene sets, STRING interaction networks were generated for each gene/ protein list with parameters set to highest confidence, no text mining, and hiding unconnected nodes. Nodes were pseudo-colored red for upregulated and blue for downregulated. Images of networks were downloaded for visualization.
## Interactome network analysis
DESeq2 differentially expressed genes with P adj < 0.05 were connected to MmuPV1 E7 interactors through the STRING protein-protein interaction network using the R package PCSF (108). The PCSF algorithm uses message passing to identify high-confidence subnetworks that contain the minimal set of edges connecting user-defined terminal nodes (i.e., the experimentally observed genes and proteins) through bridging nodes (also called Steiner nodes) that are not experimentally observed. PCSF was run using default parameters. Functional enrichment was run on the terminal nodes and Steiner nodes using g:Profiler (109).
## Statistics
All statistical tests were done using MStat software (https://oncology.wisc.edu/mstat/), GraphPad Prism (growth curves), or Microsoft Excel.
## References
1. Mcbride (2022) "Human papillomaviruses: diversity, infection and host interactions" *Nat Rev Microbiol*
2. Liu (2014) "Fields virology, 6th edition" *Clin Infect Dis*
3. Zur Hausen (2009) "Papillomaviruses in the causation of human cancers -a brief historical account" *Virology (Auckl)*
4. De Martel, Ferlay, Franceschi et al. (2012) "Global burden of cancers attributable to infections in 2008: a review and synthetic analysis" *Lancet Oncol*
5. Jablonska, Orth, Obalek et al. (1985) "Cutaneous warts. clinical, histologic, and virologic correlations" *Clin Dermatol*
6. Jabłońska, Majewski, Obalek et al. (1997) "Cutaneous warts" *Clin Dermatol*
7. (2025) *Full-Length Text Journal of Virology*
8. Tschandl, Rosendahl, Kittler (2014) "Cutaneous human papilloma virus infection: manifestations and diagnosis" *Curr Probl Dermatol*
9. Chahoud, Semaan, Chen et al. (2016) "Association between β-genus human papillomavirus and cutaneous squamous cell carcinoma in immunocompetent individuals-A meta-analysis" *JAMA Dermatol*
10. Arroyo Mühr, Hultin, Bzhalava et al. (2015) "Human papillomavirus type 197 is commonly present in skin tumors" *Int J Cancer*
11. Howley, Pfister (2015) "Beta genus papillomaviruses and skin cancer" *Virology (Auckl)*
12. Hawley-Nelson, Vousden, Hubbert et al. (1989) "HPV16 E6 and E7 proteins cooperate to immortalize human foreskin keratinocytes" *EMBO J*
13. Münger, Phelps, Bubb et al. (1989) "The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes" *J Virol*
14. Münger, Scheffner, Huibregtse et al. (1992) "Interactions of HPV E6 and E7 oncoproteins with tumour suppressor gene products" *Cancer Surv*
15. Scheffner, Huibregtse, Vierstra et al. (1993) "The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53" *Cell*
16. Huibregtse, Scheffner, Howley (1991) "A cellular protein mediates association of p53 with the E6 oncoprotein of human papillomavirus types 16 or 18" *EMBO J*
17. Huibregtse, Scheffner, Howley (1993) "Cloning and expression of the cDNA for E6-AP, a protein that mediates the interaction of the human papillomavirus E6 oncoprotein with p53" *Mol Cell Biol*
18. Scheffner, Werness, Huibregtse et al. (1990) "The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53" *Cell*
19. Werness, Levine, Howley (1990) "Association of human papillomavirus types 16 and 18 E6 proteins with p53" *Science*
20. Veldman, Horikawa, Barrett et al. (2001) "Transcriptional activation of the telomerase hTERT gene by human papillomavirus type 16 E6 oncoprotein" *J Virol*
21. Dyson, Howley, Münger et al. (1989) "The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product" *Science*
22. Münger, Werness, Dyson et al. (1989) "Complex formation of human papillomavirus E7 proteins with the retinoblastoma tumor suppressor gene product" *EMBO J*
23. White, Münger, Howley (2016) "High-risk human papillomavirus E7 proteins target PTPN14 for degradation" *mBio*
24. Hatterschide, Bohidar, Grace et al. (2019) "PTPN14 degradation by highrisk human papillomavirus E7 limits keratinocyte differentiation and contributes to HPV-mediated oncogenesis" *Proc Natl Acad Sci*
25. Hatterschide, Castagnino, Kim et al. (2022) "YAP1 activation by human papillomavirus E7 promotes basal cell identity in squamous epithelia" *Elife*
26. Chellappan, Kraus, Kroger et al. (1992) "Adenovirus E1A, simian virus 40 tumor antigen, and human papillomavirus E7 protein share the capacity to disrupt the interaction between transcription factor E2F and the retinoblastoma gene product" *Proc Natl Acad Sci*
27. Dong, Kloz, Accardi et al. (2005) "Skin hyperprolifera tion and susceptibility to chemical carcinogenesis in transgenic mice expressing E6 and E7 of human papillomavirus type 38" *J Virol*
28. Frazer, Leippe, Dunn et al. (1995) *Cancer Res*
29. Griep, Herber, Jeon et al. (1993) "Tumorigenicity by human papillomavirus type 16 E6 and E7 in transgenic mice correlates with alterations in epithelial cell growth and differentiation" *J Virol*
30. Gulliver, Herber, Liem et al. (1997) "Both conserved region 1 (CR1) and CR2 of the human papillomavirus type 16 E7 oncogene are required for induction of epidermal hyperplasia and tumor formation in transgenic mice" *J Virol*
31. Herber, Liem, Pitot et al. (1996) "Squamous epithelial hyperplasia and carcinoma in mice transgenic for the human papillomavirus type 16 E7 oncogene" *J Virol*
32. Lambert, Pan, Pitot et al. (1993) "Epidermal cancer associated with expression of human papillomavirus type 16 E6 and E7 oncogenes in the skin of transgenic mice" *Proc Natl Acad Sci*
33. Park, Shin, Lambert (2014) "High incidence of female reproductive tract cancers in FA-deficient HPV16-transgenic mice correlates with E7's induction of DNA damage response, an activity mediated by E7's inactivation of pocket proteins" *Oncogene*
34. Balsitis, Dick, Dyson et al. (2006) "Critical roles for non-pRb targets of human papillomavirus type 16 E7 in cervical carcinogenesis" *Cancer Res*
35. Shin, Sage, Lambert (2012) "Inactivating all three Rb family pocket proteins is insufficient to initiate cervical cancer" *Cancer Res*
36. Hatterschide, Brantly, Grace et al. (2020) "A conserved amino acid in the C terminus of human papillomavirus E7 mediates binding to PTPN14 and repression of epithelial differentiation" *J Virol*
37. Caldeira, Zehbe, Accardi et al. (2003) "The E6 and E7 proteins of the cutaneous human papillomavirus type 38 display transforming properties" *J Virol*
38. Vasiljević, Hazard, Eliasson et al. (2007) "Characterization of two novel cutaneous human papillomaviruses, HPV93 and HPV96" *J Gen Virol*
39. Cornet, Bouvard, Campo et al. (2012) "Comparative analysis of transforming properties of E6 and E7 from different beta human papillomavirus types" *J Virol*
40. Grace, Munger (2017) "Proteomic analysis of the gamma human papillomavirus type 197 E6 and E7 associated cellular proteins" *Virology (Auckl)*
41. Marcuzzi, Awerkiew, Hufbauer et al. (2014) "Tumor prevention in HPV8 transgenic mice by HPV8-E6 DNA vaccination" *Med Microbiol Immunol*
42. Viarisio, Müller-Decker, Accardi et al. (2018) "Beta HPV38 oncoproteins act with a hit-and-run mechanism in ultraviolet radiationinduced skin carcinogenesis in mice" *PLoS Pathog*
43. Viarisio, Mueller-Decker, Kloz et al. (2011) *Full-Length Text Journal of Virology*
44. "6 and E7 from beta HPV38 cooperate with ultraviolet light in the development of actinic keratosis-like lesions and squamous cell carcinoma in mice" *PLoS Pathog*
45. Ingle, Ghim, Joh et al. (2011) "Novel laboratory mouse papillomavirus (MusPV) infection" *Vet Pathol*
46. Cladel, Budgeon, Cooper et al. (2017) "Mouse papillomavirus infections spread to cutaneous sites with progression to malignancy" *J Gen Virol*
47. Cladel, Budgeon, Cooper et al. (2013) "Secondary infections, expanded tissue tropism, and evidence for malignant potential in immunocompromised mice infected with Mus musculus papillomavirus 1 DNA and virus" *J Virol*
48. Spurgeon, Uberoi, Mcgregor et al. (2019) "A novel in vivo infection model to study papillomavirus-mediated disease of the female reproductive tract" *mBio*
49. Wei, Buehler, Ward-Shaw et al. (2020) "An infection-based murine model for papillomavirus-associated head and neck cancer" *mBio*
50. Bilger, King, Schroeder et al. (2020) "A mouse model of oropharyngeal papillomavirus-induced neoplasia using novel tools for infection and nasal anesthesia" *Viruses*
51. Blaine-Sauer, Shin, Matkowskyj et al. (2021) "A novel model for papillomavirus-mediated anal disease and cancer using the mouse papillomavirus"
52. Uberoi, Yoshida, Lambert (2018) "Development of an in vivo infection model to study mouse papillomavirus-1 (MmuPV1)" *J Virol Methods*
53. Uberoi, Yoshida, Frazer et al. (2016) "Role of ultraviolet radiation in papillomavirus-induced disease" *PLoS Pathog*
54. Meyers, Uberoi, Grace et al. (2017) "Cutaneous HPV8 and MmuPV1 E6 proteins target the NOTCH and TGF-β tumor suppressors to inhibit differentiation and sustain keratinocyte proliferation" *PLoS Pathog*
55. Wei, Grace, Uberoi et al. (2021) "The mus musculus papillomavirus type 1 E7 protein binds to the retinoblastoma tumor suppressor: implications for viral pathogenesis"
56. Romero-Masters, Grace, Lee et al. "2023 MmuPV1 E7's interaction with PTPN14 delays epithelial differentiation and contributes to virus-induced skin disease" *PLoS Pathog*
57. Wang, Zhou, Prabhu et al. (2010) "The canine papillomavirus and gamma HPV E7 proteins use an alternative domain to bind and destabilize the retinoblastoma protein" *PLoS Pathog*
58. Romero-Masters, Lambert, Munger (2022) "Molecular mecha nisms of MmuPV1 E6 and E7 and implications for human disease" *Viruses*
59. Romero-Masters, Muehlbauer, Hayes et al. (2023) "MmuPV1 E6 induces cell proliferation and other hallmarks of cancer" *mBio*
60. Brimer, Drews (2017) "Association of papillomavirus E6 proteins with either MAML1 or E6AP clusters E6 proteins by structure, function, and evolutionary relatedness" *PLoS Pathog*
61. Wang, Uberoi, Spurgeon et al. (2020) "Stress keratin 17 enhances papillomavirus infection-induced disease by downregulating T cell recruitment" *PLoS Pathog*
62. Saldivar, Lopez, Feldman et al. (2007) "COX-2 overexpression as a biomarker of early cervical carcinogenesis: a pilot study" *Gynecol Oncol*
63. Kim, Oh, No et al. (2009) "Involvement of NF-B and AP-1 in COX-2 upregulation by human papillomavirus 16 E5 oncoprotein" *Carcinogenesis*
64. Lu, Abudukeyoumu, Zhang et al. (2021) "Heme oxygenase 1: a novel oncogene in multiple gynecological cancers" *Int J Biol Sci*
65. Thibodeau, Geddes, Fortier et al. (2015) "Gene Expression characterization of HPV positive head and neck cancer to predict response to chemoradia tion" *Head and Neck Pathol*
66. Nielsen, Hørding, Daugaard et al. (1991) "Cytokeratin intermediate filament pattern and human papillomavirus type in uterine cervical biopsies with different histological diagnosis" *Gynecol Obstet Invest*
67. Hobbs, Batazzi, Han et al. (2016) "Loss of keratin 17 induces tissue-specific cytokine polarization and cellular differentiation in HPV16-driven cervical tumorigenesis in vivo" *Oncogene*
68. Williams, Disbrow, Schlegel et al. (2005) "Requirement of epidermal growth factor receptor for hyperplasia induced by E5, a high-risk human papillomavirus oncogene" *Cancer Res*
70. Wei, Choi, Buehler et al. (2021) "Role of IQGAP1 in papillomavirus-associated head and neck tumorigenesis" *Cancers (Basel)*
71. Scarth, Wasson, Patterson et al. (2023) "Exploitation of ATP-sensitive potassium ion (K ATP ) channels by HPV promotes cervical cancer cell proliferation by contributing to MAPK/AP-1 signalling" *Oncogene*
72. Spurgeon, Den Boon, Horswill et al. (2017) "Human papillomavirus oncogenes reprogram the cervical cancer microenviron ment independently of and synergistically with estrogen" *Proc Natl Acad Sci*
73. Subramanian, Tamayo, Mootha et al. (2005) "Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles" *Proc Natl Acad Sci*
74. Wu, Grace, Munger (2023) "The HPV8 E6 protein targets the Hippo and Wnt signaling pathways as part of its arsenal to restrain keratinocyte differentiation"
75. Pim, Collins, Banks (1992) "Human papillomavirus type 16 E5 gene stimulates the transforming activity of the epidermal growth factor receptor" *Oncogene*
76. Crusius, Auvinen, Steuer et al. (1998) "The human papillomavirus type 16 E5-protein modulates ligand-dependent activation of the EGF receptor family in the human epithelial cell line HaCaT" *Exp Cell Res*
77. Wasson, Müller, Ross et al. (2017) "Human papillomavirus type 18 E5 oncogene supports cell cycle progression and impairs epithelial differentiation by modulating growth factor receptor signalling during the virus life cycle" *Oncotarget*
78. Munkarah, Ali-Fehmi (2005) "COX-2: a protein with an active role in gynecological cancers" *Curr Opin Obstet Gynecol*
79. Zou, Lyu, Jiang et al. (2020) "Use of peripheral blood transcriptomic biomarkers to distinguish high-grade cervical squamous intraepithelial Full-Length Text Journal of Virology November"
80. "lesions from low-grade lesions" *Oncol Lett*
81. Prabhakar, James, Youssef et al. (2024) "A human papillomavirus 16 E2-TopBP1 dependent SIRT1-p300 acetylation switch regulates mitotic viral and human protein levels and activates the DNA damage response" *MBio*
82. Prabhakar, James, Fontan et al. (2023) "Human papillomavirus 16 E2 interaction with TopBP1 is required for E2 and viral genome stability during the viral life cycle" *J Virol*
83. Prabhakar, Morgan (2024) "A new role for human papillomavirus 16 E2: mitotic activation of the DNA damage response to promote viral genome segregation" *Tumour Virus Res*
84. Dick, Goodrich, Sage et al. (2018) "Non-canonical functions of the RB protein in cancer" *Nat Rev Cancer*
85. Ishak, Marshall, Passos et al. (2016) "An RB-EZH2 complex mediates silencing of repetitive DNA sequences" *Mol Cell*
86. Ishak, Coschi, Roes et al. (2017) "Disruption of CDKresistant chromatin association by pRB causes DNA damage, mitotic errors, and reduces Condensin II recruitment" *Cell Cycle*
87. Dick, Dyson (2003) "pRB contains an E2F1-specific binding domain that allows E2F1-induced apoptosis to be regulated separately from other E2F activities" *Mol Cell*
88. Barnard, Mcmillan (1999) "The human papillomavirus E7 oncoprotein abrogates signaling mediated by interferon-alpha" *Virology (Auckl)*
89. Koromilas, Li, Matlashewski (2001) "Control of interferon signaling in human papillomavirus infection" *Cytokine Growth Factor Rev*
90. Byg, Vidlund, Vasiljevic et al. (2012) "NF-κB signalling is attenuated by the E7 protein from cutaneous human papillomaviruses" *Virus Res*
91. Lo Cigno, Calati, Albertini et al. (2020) "Subversion of host innate immunity by human papillomavirus oncoproteins" *Pathogens*
92. Luo, Donnelly, Gong et al. (2020) "HPV16 drives cancer immune escape via NLRX1-mediated degradation of STING" *J Clin Invest*
93. Lou, Huang, Zhou et al. (2023) "DNA virus oncoprotein HPV18 E7 selectively antagonizes cGAS-STING-triggered innate immune activation" *J Med Virol*
94. Miyauchi, Kim, Jones et al. (2023) "Human papillomavirus E5 suppresses immunity via inhibition of the immuno proteasome and STING pathway" *Cell Rep*
95. Zhussupbekova, Sinha, Kuo et al. (2016) "A mouse model of hyperproliferative human epithelium validated by keratin profiling shows an aberrant cytoskeletal response to injury" *EBioMedicine*
96. Jha, Alam, Kashyap et al. (2023) "Crosstalk between PD-L1 and Jak2-Stat3/ MAPK-AP1 signaling promotes oral cancer progression, invasion and therapy resistance" *Int Immunopharmacol*
97. Ezine, Lebbe, Dumaz (2023) "Unmasking the tumourigenic role of SIN1/MAPKAP1 in the mTOR complex 2" *Clinical & Translational Med*
98. Volarević, Thomas (2001) "Role of S6 phosphorylation and S6 kinase in cell growth" *Prog Nucleic Acid Res Mol Biol*
99. Yang, Guo, Li et al. (2023) "PPM1H is downregulated by ATF6 and dephosphorylates p-RPS6KB1 to inhibit progression of hepatocellular carcinoma" *Molecular Therapy -Nucleic Acids*
100. Amar-Schwartz, Hur, Jbara et al. (2022) "S6K1 phosphorylates Cdk1 and MSH6 to regulate DNA repair"
101. Li, Wang, Hong et al. (2025) "NLRP3 promotes the proliferation and metastasis of colorectal cancer by regulating the S6K1-GLI1 signaling pathway" *J Cancer*
102. Calderon-Aparicio, He, Simone (2024) "S6K1 controls DNA damage signaling modulated by the MRN complex to induce radioresistance in lung cancer" *Int J Mol Sci*
103. Kandathil, Akhondi, Kadletz-Wanke et al. (2024) "The dual role of autophagy in HPV-positive head and neck squamous cell carcinoma: a systematic review" *J Cancer Res Clin Oncol*
104. Dossou, Basu (2019) "The emerging roles of mTORC1 in macroma naging autophagy" *Cancers (Basel)*
105. (2025) "RStudio: Integrated Development Environment for R"
106. Hw (2009) "Ggplot2: elegant graphics for data analysis"
107. Slowilowskik (2024) "Ggrepel: Automatically Position Non-Overlapping Text Labels"
108. Tannenbaum (2006) "Center for High Throughput Computing"
109. Dobin, Davis, Schlesinger et al. (2013) "STAR: ultrafast universal RNA-seq aligner" *Bioinformatics*
110. Mootha, Lindgren, Eriksson et al. (2003) "PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes" *Nat Genet*
111. Liberzon, Birger, Thorvaldsdóttir et al. (2015) "The Molecular Signatures Database (MSigDB) hallmark gene set collection" *Cell Syst*
112. Akhmedov, Kedaigle, Chong et al. (2017) "PCSF: an R-package for network-based interpretation of high-throughput data" *PLoS Comput Biol*
113. Kolberg, Raudvere, Kuzmin et al. (2023) "G:Profiler-interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update)" *Nucleic Acids Res*
114. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12507738&blobtype=pdf | # Phenotypic characterization of two neuroinvasive Toscana virus strains clinically associated with self-limited and persistent infections in human neural cells and brain organoids
Stefania Vogiatzis, Gianni Gori Savellini, Chiara Terrosi, Marta Trevisan, Gabriele Anichini, Letizia Rizzo, Giulia Alessandri, Emanuela Dal, Camilla Lucca, Luisa Barzon, Maria Cusi, Chaitenya Verma, Santos Roney, Oswaldo Coimbra, Cruz, Fiocruz, Subhash Brazil, Mehto, Vogiatzis Gori, Anichini Rizzo
## Abstract
Background: Toscana virus (TOSV), a sandfly-borne phlebovirus, is a frequent cause of viral meningitis and meningoencephalitis in Mediterranean countries during summer months. Clinical outcomes vary from self-limited disease to prolonged persistent infection lasting over 3 months, but the mechanisms underlying these differences remain unclear. In this study, we compared the pathobiological features of two clinical TOSV strains, SI-1812 (associated with self-limited disease) and INMI (associated with persistent infection), using human neural models. Methods: Three human neural systems, DBTRG.05MG glioblastoma cells, embryonic stem cell-derived neurons, and human brain organoids (hBOs), were infected with the two TOSV strains. Viral replication, cytopathic effect, innate immune response, and viral protein expression were assessed. Furthermore, whole-genome sequencing was performed to identify strainspecific differences. Results: The INMI strain showed reduced replication and cytopathic effect compared to SI-1812, supporting persistent infection in hBOs for up to 21 days. SI-1812 infection triggered a strong interferon-b response even at early stages of infection and low viral titers, whereas INMI induced a modest innate immune response in the early stages of infection, likely supporting its persistence in hBOs. The timing of viral NSs protein expression differred between the two viral strains suggesting distinct mechanisms in RIG-I activation and inflammatory response modulation.Frontiers in Cellular and Infection Microbiology frontiersin.org 01
## Introduction
Toscana virus (TOSV) is a vector-borne virus, first isolated from Phlebotomus perniciosus and Phlebotomus perfiliewi in central Italy in 1971. This virus belongs to the Phlebovirus genus of the Phenuiviridae family, order Hareavirales (Verani et al., 1982;Charrel et al., 2012). It causes a variety of clinical syndromes ranging from a brief selflimiting febrile illness to retinitis, encephalitis, or meningoencephalitis. The viral particle has an envelope with two glycoproteins (G N and G C ) and a tripartite (L, M, and S) singlestranded RNA genome of negative polarity (Walter and Barr, 2011;Olschewski et al., 2020). TOSV, like other phleboviruses, uses envelope proteins for entry into target cells through interaction with host receptors and for the assembly of progeny particles in infected cells (Spiegel et al., 2016;Koch et al., 2023). Entry receptors and cofactors for TOSV in host cells are only partially known. TOSV subverts heparan sulfates (HS) glycosaminoglycan and C-type lectins, DC-SIGN and L-SIGN, to attach to the cell surface (Lozach et al., 2011;de Boer et al., 2012;Hofmann et al., 2013;Pietrantoni et al., 2015;Riblett et al., 2015;Leǵer et al., 2016;Tani et al., 2016). In the dermis, TOSV infects dendritic cells (DCs) through interaction with the C-type lectin DC-SIGN, which functions as an endocytic receptor (Lozach et al., 2011;Hofmann et al., 2013;Leǵer et al., 2016), while other C-type lectins (L-SIGN and LSECtin) may act as receptors on the surface of endothelial cells (Leǵer et al., 2016;Tani et al., 2016). The mechanism of TOSV invasion of the central nervous system (CNS) and the host receptors and factors involved in neuroinvasion remain unknown. In humans, symptoms of TOSV infection occur after an incubation period of 3-7 days and typically persist for about 7-12 days (Charrel et al., 2012;Ayhan and Charrel, 2020). Most TOSV infections in humans are asymptomatic or paucisymptomatic, presenting with flu-like symptoms such as headache, fever, nausea, or vomiting (Ayhan and Charrel, 2020). A high rate of asymptomatic or mild TOSV infections is suggested by studies in endemic areas reporting seroprevalences ranging from 5% to 30% (Valassina et al., 2003;Sanbonmatsu-Gaḿez et al., 2005;Calamusa et al., 2012;Cusi et al., 2013;Marchi et al., 2017;Gori Savellini et al., 2020). Patients with CNS involvement may present with aseptic acute meningitis, encephalitis, or meningoencephalitis (Varani et al., 2015;Ayhan and Charrel, 2020). The clinical picture of meningitis caused by TOSV is similar to that caused by other viral agents. However, levels of anti-inflammatory and antiviral mediators are significantly higher in the cerebrospinal fluid (CSF) of TOSV-infected patients than in those with other infectious or noninfectious neurological diseases (Verani et al., 1991;Braito et al., 1998;Hemmersbach-Miller et al., 2004;Varani et al., 2015). Other clinical manifestations are associated with peripheral neuropathy, including paraesthesia, hyporeflexia, hypotonia, imbalance, hyperaesthesia, and Guillain-Barrésyndrome (Sanbonmatsu-Gaḿez et al., 2009;Zanelli et al., 2013;Gori Savellini et al., 2015;Okar et al., 2021;Matusali et al., 2022). Rare atypical presentations, such as epididymo-orchitis during the acute phase, have been reported, with TOSV (INMI strain) i s o l a t e d f r o m t h e s e m i n a l fl u i d o f a p a t i e n t w i t h meningoencephalitis and persistent TOSV shedding in semen up to 59 days postsymptom onset (Zanelli et al., 2013;Matusali et al., 2022). Despite the clinical and epidemiological relevance of TOSV, strainspecific phenotypic differences-particularly those linked to persistent infections or diverse clinical outcomes-remain poorly characterized. Understanding these differences is crucial for improving patient management, guiding therapeutic strategies, and informing public health interventions. Characterizing strains such as TOSV INMI, which is associated with persistent infection, could provide novel insights into viral persistence mechanisms and host immune interactions that shape clinical outcomes. It is already known that the nonstructural NS protein of TOSV and other phleboviruses functions as an innate immune suppressor, delaying the host's response to infection. This viral protein, which acts as an E3 ubiquitin ligase to inhibit type I interferon beta (IFN-b) expression, may play a critical role in determining viral phenotype (Gori-Savellini et al., 2013;Gori Savellini et al., 2015;Gori Savellini et al., 2019). This study aims to phenotypically characterize the TOSV INMI strain in comparison with a reference strain (SI-1812) (Baldelli et al., 2004), using multiple neural cell models, including the human glioblastoma cell line DBTRG.05MG, human cortical neurons, and brain organoids derived from embryonic stem cells, to investigate infection and replication dynamics. Such detailed analysis will contribute to a deeper understanding of how strain-specific differences influence TOSV pathogenesis and persistence.
(DMEM) (EuroClone, Milan, Italy) supplemented with 100 U/ml penicillin/streptomycin (EuroClone) and 5% heat-inactivated fetal bovine serum (FBS) (EuroClone) at 37°C in a 5% CO 2supplemented incubator. TOSV strains SI-1812 and INMI were respectively isolated from the CSF and from the seminal fluid of patients with meningitis on Vero E6 cells at the Virology Laboratory of S. Maria delle Scotte Hospital in Siena, Italy. After isolation and characterization, viruses were propagated on Vero E6 cells, and stocks were prepared and stored at -80 °C. DBTRG.05MG human glioblastoma cells (ATCC CRL-2020) were cultured in RPMI medium (EuroClone) supplemented with 100 U/ml penicillin/ streptomycin and 10% FBS at 37 °C in a 5% CO 2supplemented incubator.
## Generation of neurons from human ESCs
To obtain human neurons, AAVS1-TRE3G-NGN2 human embryonic stem cells (hESC-NGN), kindly provided by J.W. Harper, were used and differentiated as indicated by Ordureau et al., with minor modifications (Ordureau et al., 2018). Briefly, H9-NGN2 cells were seeded on day 0 in a Geltrex ® -coated six-well plate (2 × 10 5 cells/well) in ND1 medium composed of DMEM/F12/NEAA (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA), human Brain-Derived Neurotrophic Factor (BDNF) (10 ng/ml; PeproTech, Thermo Fisher Scientific), human NT-3 (10 ng/ml, PeproTech, Thermo Fisher Scientific), human laminin (0.2 µg/ml, Invitrogen, Thermo Fisher Scientific), and doxycycline (2 µg/ml, Clontech, Takara Bio USA, San Jose, CA, USA). On day 1, ND1 medium was replaced with fresh ND1 medium; on day 2, ND1 medium was replaced with ND2 medium consisting of Neurobasal medium supplemented with B27/GlutaMAX (Invitrogen, Thermo Fisher Scientific), BDNF, NT3, and doxycycline (concentrations are the same as above). After day 2, 50% of the medium in each well was replaced every 2 days with fresh medium. On day 7, cells were detached by treatment with Accutase (Gibco, Thermo Fisher Scientific) and replated in Geltrex ® -coated six-well plates (4 × 10 5 cells/ well) in ND2 medium supplemented with Y27632 (10 µM, Thermo Fisher Scientific). On day 8, the medium was replaced with fresh ND2 Medium. After day 8, 50% of the medium in each well was replaced every 2 days until day 14, when cortical neurons reached the mature phase.
## Generation of human brain organoids from human ESCs
To obtain 3D human brain organoids (hBOs), the protocol developed by Lancaster and Knoblich was adopted with some modifications (Lancaster and Knoblich, 2014). Briefly, on day 0, human ES cells (H9, WiCell Institute, Madison, U.S.A.) were dissociated to a single-cell suspension and seeded in round-bottom, low-attachment 96-well plates (1 × 10 5 cells/well) in mTeSR complete medium (StemCell Technologies, Cologne, Germany) supplemented with Y27632 (10 µM, Thermo Fisher Scientific). On day 3, the medium was replaced with fresh mTeSR. On day 5, the medium was replaced with Neural Induction Medium, composed of DMEM/F12/NEAA (Gibco, Thermo Fisher Scientific) with GlutaMAX/B27 (Thermo Fisher Scientific), heparin (1 mg/ml, Merck Millipore, Burlington, MA, USA), and antibiotic-antimycotic (Anti-Anti, 1% v/v, Gibco, Thermo Fisher Scientific). The embryoid bodies (EBs) were fed every other day. On day 11, the EBs were embedded on Matrigel (Corning ™ , Merck Millipore) droplets and grown in a low-attachment six-well plate (eight to nine EBs/well) in differentiation medium (DM), composed of Neurobasal/DMEM/F12 (Thermo Fisher Scientific) supplemented with GlutaMAX/NEAA/B27/N2 (Thermo Fisher Scientific), with human insulin (0.00025%, v/v, Merck-Milliore, Milan, Italy) and 2mercaptoethanol (0.00035%, v/v, Gibco, Thermo Fisher Scientific) added. To prevent contamination, Anti-Anti (1%, v/v, Gibco, Thermo Fisher Scientific) was included. On day 13, DM was replaced with fresh medium. On day 15, the medium in each well was replaced with differentiation medium 2 (DM2), i.e., DM with B27 plus vitamin A (Thermo Fisher Scientific), and the plate was placed on an orbital shaker. From days 15 to 60, the hBOs were grown in an orbital shaker until the day of infection. The medium was changed twice per week.
## Viral infection and analysis of viral replication kinetics in neural cell cultures and organoids
DBTRG.05MG cells and human ESC-derived neurons were infected with SI-1812 or INMI TOSV strains at an MOI of 0.1. After 1 h of virus absorption, extensive washes were performed to remove the inoculum, and complete growth medium was added. Cell culture supernatants were collected at 1, 3, 5, 6, and 7 days postinfection (p.i.) and stored at -80 °C for further analysis. hBOs were infected with SI-1812 and INMI TOSV strains at an MOI of 0.2. For infection, hBOs were transferred into a tube containing the required amount of virus. After inoculation, hBOs were centrifuged at 1,100 rpm for 30 min. Subsequently, hBOs were incubated at 37 °C with 5% CO 2 for 60 min to allow the adsorption of the virus. The medium was then removed, the organoids were washed with PBS twice, and fresh DM2 was added. The organoids were incubated at 37 °C, with 5% CO 2 in an orbital shaker. Supernatants of infected organoids were collected at different time points (1,3,6,10,16,19, and 21 days postinfection) for further analysis. The organoids were also collected to perform immunofluorescence. The release of infectious viral particles by infected DBTRG.05MG cells, human ESC-derived neurons, and hBOs was determined by virus microtitration assay on Vero E6 cells cultured in 96-well plates, and the 50% Tissue Culture Infectious Dose (TCID 50 /ml) end point titer was calculated using the Reed and Muench method (Reed and Muench, 1938). Results were reported as mean values ± standard deviation (SD) from at least three independent experiments.
## Assessment of IFN-b production
Medium from DBTRG.05MG cells and human ESC-derived neurons, mock-or TOSV-infected, collected at 1, 3, 5, 6, and 7 days p.i as described above, and the medium from hBOs, mock-or TOSV-infected, collected at 1, 3, 6, 10, 16, 19, and21 days p.i., were processed for IFN-b quantification by enzyme-linked immunosorbent assay (ELISA) using a commercial kit (VeriKine-HS Human IFN Beta TCM ELISA Kit; PBL Assay Science, Piscataway, NJ, USA), following manufacturer's instructions. IFNb level was expressed as mean cytokine concentrations (pg/ml) ± SD from at least three independent experiments.
## Multiplex cytokine detection
Cell supernatant of mock-infected or TOSV-infected hBOs was centrifuged at 10,000 × g for 5 min, prior to performing the assay, and diluted 1:4 in Universal Assay Buffer. A multiplex biometric ELISA-based immunoassay, containing dyed microspheres conjugated with 10 different monoclonal antibodies specific for a target cytokine, was used according to the manufacturer's instructions (Human Luminex Discovery Assay, Bio-Techne, Minneapolis, MN, USA). Magnetic bead concentrations were measured using a Luminex ™ detection system (Luminex ™ 200 ™ Instrument System, Bio-Techne, Minneapolis, MN, USA). The following cytokines were analyzed: IFN-b, Interferon gamma (IFN-g), tumor necrosis factor (TNF-a), interleukin (IL)-1b, IL-2, IL-4, IL-6, IL-8, IL-10, and IL-12. The analyte concentration was calculated using a 7-point standard curve. Acquired data were analyzed using xPONENT v4.3.309.1, and cytokine concentrations were reported as picograms per milliliter.
## Immunofluorescence analysis
Neurons (10 4 cells/cover slip) were grown on Geltrex ® -treated coverslips. Cells were fixed in paraformaldehyde (4%, v/v, Thermo Fisher Scientific) for 20 min and permeabilized with Triton X-100 (0.1%, v/v) in Dulbecco's modified Phosphate Buffered Solution (DPBS) (Gibco, Thermo Fisher Scientific) for 5 min at room temperature. Cells were then incubated with the following primary antibodies diluted 1:100 in BSA (1%, w/v, in PBS): serum anti-TOSV, beta-3 Tubulin (Thermo Fisher Scientific). Nuclei were stained with the DRAQ5 ™ intercalating fluorescent probe (Thermo Fisher Scientific), and cells were observed with a confocal microscope (Leica, Buffalo Grove, IL, USA) at × 63 magnification and processed using Fiji-ImageJ software. Human brain organoids were fixed in paraformaldehyde (4%, v/v, Thermo Fisher Scientific) for 40 min and washed three times in DPBS. hBOs were placed in 10% w/v sucrose solution for 30 min, in 20% w/v sucrose solution for 30 min, and finally in 30% w/v sucrose solution at + 4 °C, until embedding with OCT solution (VWR International PBI, Milan, Italy) in a plastic mold. To solidify the solution rapidly, molds were put in contact with liquid nitrogen vapors. These organoids were prepared for cryosectioning and immunostaining. The 20-µm-thick sections generated were washed three times with DPBS and incubated with 4% BSA/DPBS containing 0.2% Triton X-100 for 1 h at room temperature. After washing twice in DPBS, primary antibodies diluted in 4% BSA/PBS were added and incubated overnight at 4 °C. Cells were incubated with the following primary antibodies: serum anti-TOSV, anti-TUJI1 (Thermo Fisher Scientific), and anti-cleaved caspase-3 (Asp175; Cell Signaling Technology, Leiden, The Netherlands). Nuclei were stained with DRAQ5 ™ interlayer fluorescent probe (Thermo Fisher Scientific), and cells were observed confocal microscope (Leica, Buffalo Grove, IL, USA) and processed by Fiji-ImageJ software. The experiment was repeated three times, and a total of 12 slices of processed brain tissue were analyzed. Vero E6 cells were infected with medium collected from hBOs at 6 days p.i. and fixed with cold acetone:methanol solution for 10 min. Immunofluorescence staining was then performed using an anti-N TOSV antibody, followed by a FITC-conjugated secondary antibody. Images were acquired with a fluorescence-equipped Eclipse Ts2 microscope (Nikon, Milan, Italy) and processed using Fiji-ImageJ software. Corrected total cell fluorescence (CTCF) was calculated as: CTCF = Integrated Density -(Area of selected cell × Mean fluorescence of background readings). Data are presented as mean CTCF ± SD from at least three independent fields.
## Western blot analysis
Endogenous Retinoic-acid Inducible Gene I (RIG-I) and TOSV SI-1812 or INMI NS expression was investigated in DBTRG.05MG cells infected as previously described and collected at 24, 48, and 72 h p.i. Whole-cell lysates were prepared in RIPA buffer (25 mM Tris-HCl [pH 7.5]; 150 mM NaCl; 1% Triton X-100) supplemented with COmplete ™ Protease Inhibitor (Merck-Millipore), and protein concentration was determined by BCA assay (Pierce, Milan, Italy). Fifty micrograms of total proteins, supplemented with Laemmli sample buffer and denatured for 5 min at 95 °C, was separated by SDS-PAGE and then transferred to a NitroBind nitrocellulose membrane (Santa Cruz Biotechnology, Heidelberg, Germany). After blocking with 5% nonfat dry milk, the filters were incubated O/N at room temperature with anti-RIG-I CARDs (1:5,000), anti-NSs (1:200), or antiactin (1:2,500; Invitrogen, Thermo Fisher Scientific) as a loading control. After being washed three times with PBS containing 0.05% Tween-20 (PBS-T), membranes were incubated with a horseradish peroxidase (HRP)-conjugated anti-mouse IgG secondary antibody (Merck-Millipore). After several washes, the target bands were visualized using an enhanced chemiluminescence kit (Pierce), and images were acquired with a ChemiDoc instrument (BioRad, Milan, Italy).
## Statistical analysis
Statistical significance was assessed with the two-tailed Chisquare test, paired or unpaired as appropriate. Results were considered statistically significant at p < 0.05. When appropriate, values are reported as mean ± SD. One-way ANOVA followed by a Bonferroni post hoc test was performed to compare the significance among different study groups. Data are presented as mean values ± SD of three independent experiments performed in triplicate, unless otherwise stated. All analyses were performed using GraphPad Prism software (v.8.0.1).
## Results
## TOSV growth kinetics in DBTRG.05MG human glioblastoma cells
To compare the infection and replication efficiency of the two TOSV strains (INMI and SI-1812) in CNS cells, we used different human cell systems, i.e., the DBTRG.05 MG glioblastoma cell line, cortical neurons, and brain organoids. The DBTRG.05 MG cell line, isolated from a female patient with glioblastoma, although highly aneuploid, responds to IFN-b stimulus with expression of proinflammatory genes, thus suggesting that it originated from astrocytes. Both TOSV INMI and SI-1812 strains were able to infect and grow in DBTRG.05 MG glioblastoma cells, although the INMI strain reached higher titers at 5 days (2.1 × 10 7 TCID 50 /ml ± 1.05 × 10 7 TCID 50 /ml, p = 0.03) and 6 days p.i. (7.5 × 10 5 TCID 50 /ml ± 3.8 × 10 5 TCID 50 /ml, p = 0.04) than the SI-1812 strain (1.5 × 10 6 ± 1.06 × 10 6 and 1.4 × 10 5 ± 5.5 × 10 4 TCID 50 /ml, respectively). At 7 days p.i., viral titer slightly decreased for both the strains (Figure 1A).
## TOSV growth kinetics in human ESCderived neurons
Differentiated cortical neurons were infected with TOSV INMI and SI-1812 at an MOI of 0.1 and assayed during the next 7 days. These cells were susceptible to TOSV infection and permissive to replication, as determined by titration of infectious viral particles and immunofluorescence. As shown in Figure 1B, in this neural cell system, TOSV INMI replication was slower than that of the SI-1812 strain in the first 3 days p.i., yielding significantly lower infectious viral titer in cell supernatant (day 3 p.i., SI-1812 titer 1.2 × 10 6 TCID 50 /ml ± 1.7 × 10 5 TCID 50 /ml vs. 6.6 × 10 3 TCID 50 /ml ± 2.5 × 10 3 TCID 50 /ml of the INMI strain, p = 0.028). However, the INMI strain appeared more capable of spreading in these cells than the SI-1812 strain. Indeed, although both strains reached a similar viral titer at 6 days p.i. (p = 0.074), immunofluorescence showed a higher number of positive cells in neurons infected with the INMI strain (p = 0.028) (Figure 2A). Furthermore, the fold-change fluorescence relative to the mock sample confirmed the enhanced capability of the INMI strain to spread in ESC-derived neurons (Figure 2B).
## TOSV growth kinetics in human brain organoids
hBOs were infected at day 60 of differentiation, when they presented complex cell-cell and cell-environment interactions similar to the in vivo condition. Briefly, hBOs were infected with TOSV INMI and SI-1812 strains at a 0.2 MOI. Experiments were conducted in biological triplicate, with five hBOs per experiment. Following infection, morphological sizing of hBOs was monitored Additionally, immunofluorescence staining was performed on hBOs, with a total of six slides per organoid (Figure 3). Also in this cell system, the TOSV INMI strain grew more slowly than the SI-1812 strain for the first 3 days p.i. (p = 0.045). At day 6 p.i., the viral titer (about 10 5 TCID 50 /ml) was similar for the two strains (p = 0.808). At 10 days p.i., however, the INMI strain showed a higher titer (p = 0.017) (8.1 × 10 5 TCID 50 /ml ± 3.1 × 10 5 TCID 50 /ml) than the SI-1812 strain (9.1 × 10 4 TCID 50 /ml ± 2.7 × 10 4 TCID 50 /ml) (Figure 3A). The difference further increased in the follow-up due to massive cell death and hBO disruption. Indeed, as shown in Figure 4, the size of SI-1812-infected hBOs was smaller than that of hBOs infected with INMI at day 16 p.i. (p < 0.0009).
Immunofluorescence performed on TOSV-infected hBOs at day 6 p.i. showed a similar distribution of infected cells for both strains (Figure 3B), but a higher level of apoptosis induction in INMIinfected hBOs compared to SI-1812-infected hBOs, as shown by coimmunostaining for cleaved caspase-3 (p < 0.0008, Figure 3C).
## Replication efficiency and cytopathic effect of INMI and SI-1812 TOSV strains in Vero cells
The two virus strains isolated from brain organoids at 6 days p.i., showing similar titer, were then inoculated in Vero E6 cells. At 3 days p.i., SI-1812 TOSV showed a CPE that was slightly more pronounced than the other virus. In contrast, at 4 days p.i., the CPE was markedly more extended in cells infected with the INMI strain than those infected with SI-1812, suggesting a greater spreading capability for this strain (Figures 5A-D). This result was further confirmed by immunofluorescence, which demonstrated the presence of the virus in a larger number of cells infected with the INMI strain at 4 days p.i. (Figures 5E-H), which, in turn, exhibited a reduced cytopathic effect and a decreased neural cell death. Indeed, the mean CTCF for infected INMI cells was 7.45 × 10 7 ± 1.87 × 10 7 , while for SI-1812 samples, it was 2.56 × 10 7 ± 1.27 × 10 7 (p = 0.002).
## Interferon-b induction in TOSV-infected cells
Interferon-b expression was evaluated in human DBTRG.05MG glioblastoma cells, ESC-derived human cortical neurons, and hBOs infected with the two strains of TOSV at different time points. Infection with TOSV INMI led to a significantly lower induction of IFN-b than the SI-1812 strain in the two cell types and hBOs (Figure 6). In infected DBTRG.05MG cells, the highest level of IFN-b was recorded at day 6 p.i. (478.46 pg/ml ± 59.51 pg/ml of the INMI strain vs. 1,427.57 pg/ml ± 80.19 pg/ml of the SI-1812 strain; p < 0.0001). It then began to drop only in cells infected with SI-1812, conceivably as a consequence of cell death due to cytopathic effect (Figure 6A). Similarly, along the 7-day time course in ESC-derived neurons, IFN-b was produced at lower levels in cells infected with the INMI strain concentration: 3.6 pg/ml ± 0.93 pg/ml) than in those infected with the SI-1812 strain (mean concentration: 7.8 pg/ml ± 3.7 pg/ml), which showed a significant and progressive increase in IFN-b secretion, doubling its level after the third day p.i. (6.1 pg/ml ± 0.92 pg/ml) until the seventh day p.i. (10.2 pg/ml ± 1.41 pg/ml) (p < 0.0001) (Figure 6B).
In hBOs, the difference of IFN-b induction by the two strains was more pronounced, with IFN-b levels about 10-fold lower in cells infected with the TOSV INMI strain (161 pg/ml ± 35.4 pg/ml) than in cells infected with the SI-1812 strain (1,148 pg/ml ± 161.3 pg/ml) at 3 days p.i. (p = 0.0005) (Figure 6C). In cells infected with the INMI strain, IFN-b increased at 6 days p.i. (398 pg/ml ± 55.4 pg/ ml) and showed no significant differences in the following days until day 21 p.i. (552 pg/ml ± 74.23 pg/ml) (Figure 6C). Multiplex cytokine analyses carried out on the growth medium of infected hBOs confirmed the higher production of IFN-b (day 10 p.i.: 1.9 × 10 4 pg/ml ± 1.5 × 10 3 pg/ml, p = 0.0001), as well as the proinflammatory cytokine IL-6 (day 10 p.i.: 3.4 × 10 3 pg/ml ± 2.1 × 10 2 pg/ml, p < 0.0001) and IL-8 (day 10 p.i.: 922 pg/ml ± 57.2 pg/ ml, p = 0.0035) in cells infected with SI-1812, compared to those infected with INMI (Figure 7). The other assayed interleukins were not detected. The high of IL-6 and IL-8 in the supernatant of SI-1812-infected cells demonstrated that this strain unveiled the neuropathogenesis of the brain by inducing a strong inflammation and destroying cells, in contrast to the INMI strain, which persisted longer in hBOs.
## RIG-I expression in DBTRG.05MG-infected cells
Previous data showed that the inhibitory effect of TOSV NSs on RIG-I was involved in the signaling cascade for type I IFN production, leading to degradation of RIG-I upon binding. In this study, we examined whether the interaction between RIG-I and NSs could have a functional consequence when DBTRG.05MG cells were infected with TOSV INMI or SI-1812 strain. For this purpose, we assessed the presence of RIG-I and NSs in these infected cells at 24, 48, and 72 h p.i. As shown in Figure 8, the RIG-I protein became detectable at 2 days p.i., while NSs were more prominent at 3 days p.i. in cells infected with the SI-1812 strain. In contrast, RIG-I was observed only at 3 days p.i., whereas NSs were strongly expressed as early as 1 day p.i. in cells infected with the INMI strain, suggesting that early and efficient expression of NSs in INMI-infected cells led to RIG-I degradation and hence impaired IFN-b response. This feature also confirmed that the different amount of IFN-b expressed by these cells depends on the TOSV strain used for infection.
## Discussion
It is well established that TOSV exhibits tropism for the CNS, causing aseptic acute meningitis, encephalitis, or meningoencephalitis (Charrel et al., 2012;Varani et al., 2015;Ayhan and Charrel, 2020). In this study, we used three human neural cell systems, i.e., DBTRG.05MG glioblastoma cells, ESC-derived cortical neurons, and brain organoids, to investigate in vitro the pathobiology and fitness of the TOSV SI-1812 and INMI strains (Cusi et al., 2005;Matusali et al., Analysis of pro-and anti-inflammatory cytokines in TOSV-infected human brain organoids. Representation of IFN-b, IL-6, and IL-8 concentrations in supernatants of hBOs infected with TOSV INMI and SI-1812 strains and collected at 1, 3, and 10 days p.i. Uninfected organoids served as negative controls (Ctr -). Among the analyzed cytokines, only IFN-b, IL-6, and IL-8 showed significant modulation due to viral infection. Data are presented as mean concentrations (pg/ml) ± SD from at least three independent experiments. Statistical analysis was performed using a two-tailed one-way ANOVA test (Dunnett) and reported as *p < 0.05, **p < 0.01, and ***p < 0.001.
## 2022
). The TOSV INMI strain was isolated from seminal fluid of a patient with an atypical clinical course, characterized by meningoencephalitis and persistent viral shedding in the seminal fluid for about 3 months. TOSV INMI infection efficiency, replication, CPE, and induction of the innate immune response were compared with the reference strain SI-1812 isolated from the CSF of a patient with acute meningoencephalitis. Both TOSV strains replicated and induced IFN-b production across all systems, but the INMI strain displayed slower growth kinetics (except in the glioblastoma cell line), leading to less CPE and less induction of IFN-b and other proinflammatory cytokines than the reference SI-1812 strain. In DBTRG.05MG cells, both TOSV strains exhibited similar replication kinetics over the first 4 days p.i., but INMI showed more pronounced growth by day 6 p.i. In ESC-derived neurons and hBOs, INMI displayed slower growth in the first 3 days but successively reached titers comparable to SI-1812. Notably, INMI persisted in organoids for up to 21 days, whereas infectious SI-1812 was no longer detectable in growth medium because of the massive disruption of the infected hBOs. This is in agreement with the enhanced capacity of the INMI strain to persist in the CNS in vivo (Matusali et al., 2022). Moreover, it is worth noting that SI-1812 infection induced the release of inflammatory factors such as IL-6 and IL-8 by glioblastoma cells, cortical neurons, and hBOs during injury, leading to brain damage (Cusi et al., 2016;Grebenciucova and VanHaerents, 2023).
Interestingly, INMI induced significantly lower levels of IFN-b across all time points and cell systems compared to SI-1812, with the most pronounced differences observed in hBOs. Our in vitro findings suggest potential implications for in vivo infection dynamics. The TOSV INMI strain was characterized by slower initial growth and elicited a weaker innate immune response. This attenuated immune activation might have facilitated the persistence of the virus within the human host. In contrast, the SI-1812 strain replicated faster, inducing a stronger immune response that limited its persistence in the host cells and its spreading in different body tissues. Based on these data, we also analyzed the expression of TOSV NS protein in infected cells. The role of TOSV NSs in counteracting innate immunity has been documented (Gori-Savellini et al., 2013;Gori Savellini et al., 2015;Gori Savellini et al., 2019), and the temporal kinetics of NSs expression should be considered. In cells infected with the SI-1812 strain, RIG-I expression was detectable at 24-48 h p.i., likely because NS accumulation became more evident only at 72 h p.i. (Figure 8). This suggests that NSs act as an antagonist mainly during the early phases of infection, after which Toll-like receptor 3 (TLR3) and other pattern recognition receptors contribute to IFN-b induction. Conversely, in cells infected with the INMI strain, NSs was expressed earlier, within 48h p.i., resulting in a delayed appearance of RIG-I, promoting its proteasomal degradation and, consequently, a delay in IFN-b expression at 72 h p.i. To identify the genetic basis of the different phenotypes of the two TOSV strains, we performed a whole-genome sequencing (GenBank Accession Nos. OR105998, MZ643217.1).
The NS proteins of the two virus strains differ by 9 amino acids distributed along the sequence. In the NS sequence of the INMI strain, a mutation (L 286 P) occurs within the TLQ motif (a.a. 285-287), which has previously been identified as critical for protein function and may significantly affect its activity (Gori Savellini et al., 2015). This change may result in the accumulation of NSs protein in the cytoplasm, which mediates RIG-I ubiquitination and degradation, thus suppressing early immune responses (Gori-Savellini et al., 2013;Gori Savellini et al., 2019). Further studies are ongoing to clarify the impact of these genetic differences. Finally, immunofluorescence analysis of brain organoids revealed that SI-1812 caused substantial cytopathic effects, reducing organoid size, whereas INMI induced apoptosis, as indicated by cleaved caspase-3 staining, due to the activation of interferonstimulated genes (ISGs), such as TNF-related apoptosis-inducing Ligand (TRAIL) and Fas ligand (FasL), which also promote programmed cell death (Chawla-Sarkar et al., 2003). While differential IFN-b activation may account for the observed behavioral differences between the strains, studies are ongoing to further elucidate the role of their genetic differences. RIG-I modulation by the two TOSV strains. Endogenous RIG-I protein expression was analyzed by immunoblotting in DBTRG.05MG cells, either mock-infected or infected with TOSV SI-1812 or INMI strains, at various time points p.i. Simultaneously, the expression of the viral nonstructural NS protein was evaluated using a specific antibody. To ensure proper comparison between samples, antiactin monoclonal antibody was utilized as a loading control.
## References
1. Ayhan, Charrel (2020) "An update on Toscana virus distribution, genetics, medical and diagnostic aspects" *Clin. Microbiol. Infect*
2. Baldelli, Ciufolini, Francisci et al. (2004) "Unusual presentation of life-threatening Toscana virus meningoencephalitis" *Clin. Infect. Dis*
3. Brai, Ciufolini, Pippi et al. (1998) "Phlebotomus-transmitted toscana virus infections of the central nervous system: a seven-year experience in Tuscany" *Scand. J. Infect. Dis*
4. Calamusa, Valenti, Vitale et al. (2012) "Seroprevalence of and risk factors for Toscana and Sicilian virus infection in a sample population of Sicily (Italy)" *J. Infect*
5. Charrel, Bichaud, De Lamballerie (2012) "Emergence of Toscana virus in the Mediterranean area" *World J. Virol*
6. Chawla-Sarkar, Lindner, Liu et al.
7. (2003) "Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis" *Apoptosis*
8. Cusi, Gandolfo, Terrosi et al. (2016) "Toscana virus infects dendritic and endothelial cells opening the way for the central nervous system" *J. Neurovirol*
9. Cusi, Gandolfo, Valentini et al. (2013) "Seroprevalence of antibodies to sandfly fever Sicilian virus in a sample population in Tuscany, Italy. Vector Borne Zoonotic Dis"
10. Cusi, Gori Savellini, Terrosi et al. (2005) "Development of a mouse model for the study of Toscana virus pathogenesis" *Virology*
11. De Boer, Kortekaas, De Haan et al. (2012) "Heparan sulfate facilitates Rift Valley fever virus entry into the cell" *J. Virol*
12. Gori Savellini, Anichini, Gandolfo et al. (2015) "Truncation of the Cterminal region of Toscana Virus NSs protein is critical for interferon-b antagonism and protein stability" *PLoS Pathog*
13. Gori Savellini, Gandolfo, Cusi (2020) "Epidemiology of Toscana virus in South Tuscany over the years 2011-2019" *J. Clin. Virol*
14. Gori-Savellini, Valentini, Cusi (2013) "Toscana virus NSs protein inhibits the induction of type I interferon by interacting with RIG-I" *J. Virol*
15. Grebenciucova, Vanhaerents (2023) "Interleukin 6: at the interface of human health and disease" *Front. Immunol*
16. Hemmersbach-Miller, Parola, Charrel et al. (2004) "Sandfly fever due to Toscana virus: an emerging infection in southern France" *Eur. J. Intern. Med*
17. Hofmann, Li, Zhang et al. (2013) "Severe fever with thrombocytopenia virus glycoproteins are targeted by neutralizing antibodies and can use DC-SIGN as a receptor for pH-dependent entry into human and animal cell lines" *J. Virol*
18. Koch, Xin, Obr et al. (2023) "The phenuivirus Toscana virus makes an atypical use of vacuolar acidity to enter host cells" *PLoS Pathog*
19. Lancaster, Knoblich (2014) "Generation of cerebral organoids from human pluripotent stem cells" *Nat. Protoc*
20. Leǵer, Tetard, Youness et al. (2016) "Differential Use of the C-Type Lectins L-SIGN and DC-SIGN for Phlebovirus Endocytosis" *Traffic*
21. Lozach, Kühbacher, Meier et al. (2011) "DC-SIGN as a receptor for phleboviruses" *Cell Host Microbe*
22. Marchi, Trombetta, Kistner et al. (2017) "Seroprevalence study of Toscana virus and viruses belonging to the Sandfly fever Naples antigenic complex in central and southern Italy" *J. Infect. Public Health*
23. Matusali, D'abramo, Terrosi et al. (2022) "Infectious Toscana Virus in Seminal Fluid of Young Man Returning from Elba Island" *Italy. Emerg. Infect. Dis*
24. Okar, Bekircan-Kurt, Hacıoglu et al. (2021) "Toscana virus associated with Guillain-Barrésyndrome: a casecontrol study" *Acta Neurol. Belg*
25. Olschewski, Cusack, Rosenthal (2020) "The Cap-Snatching Mechanism of Bunyaviruses" *Trends Microbiol*
26. Ordureau, Paulo, Zhang et al. (2018) "Dynamics of PARKIN-Dependent Mitochondrial Ubiquitylation in Induced Neurons and Model Systems Revealed by Digital Snapshot Proteomics" *Mol. Cell*
27. Pietrantoni, Fortuna, Remoli et al. (2015) "Bovine lactoferrin inhibits Toscana virus infection by binding to heparan sulphate" *Viruses*
28. Reed, Muench (1938) "A simple method of estimating fifty percent endpoints" *Am. J. Trop. Med. Hyg*
29. Riblett, Blomen, Jae et al. (2015) "A Haploid Genetic Screen Identifies Heparan Sulfate Proteoglycans Supporting Rift Valley Fever Virus Infection" *J. Virol*
30. Sanbonmatsu-Gaḿez, Peŕez-Ruiz, Collao et al. (2005) "Toscana virus in Spain" *Emerg. Infect. Dis*
31. Sanbonmatsu-Gaḿez, Peŕez-Ruiz, Palop-Borraś et al. (2009) "Unusual manifestation of toscana virus infection" *Spain. Emerg. Infect. Dis*
32. Spiegel, Plegge, Pöhlmann (2016) "The Role of Phlebovirus Glycoproteins in Viral Entry" *Assembly and Release. Viruses*
33. Tani, Shimojima, Fukushi et al. (2016) "Characterization of Glycoprotein-Mediated Entry of Severe Fever with Thrombocytopenia Syndrome Virus" *J. Virol*
34. Valassina, Valentini, Pugliese et al. (2003) "Serological survey of Toscana virus infections in a high-risk population in Italy" *Clin. Diagn. Lab. Immunol*
35. Varani, Gelsomino, Bartoletti et al. (2015) "Meningitis Caused by Toscana Virus Is Associated with Strong Antiviral Response in the CNS and Altered Frequency of Blood Antigen-Presenting Cells" *Viruses*
36. Verani, Ciufolini, Nicoletti et al. (1982) "Studi ecologici ed epidemiologici del virus Toscana, un arbovirus isolato da flebotomi [Ecological and epidemiological studies of Toscana virus, an arbovirus isolated from Phlebotomus" *Ann. Ist Super Sanita*
37. Verani, Nicoletti, Ciufolini et al. (1991) "Viruses transmitted by sandflies in Italy" *Parassitologia*
38. Walter, Barr (2011) "Recent advances in the molecular and cellular biology of bunyaviruses" *J. Gen. Virol*
39. Zanelli, Bianco, Cusi (2013) "Testicular involvement during Toscana virus infection: an unusual manifestation?" *Infection* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12505966&blobtype=pdf | # Soluble MFGE8 mediates cell entry of Crimean-Congo hemorrhagic fever virus
Xue Ma, Zhi-Sheng Xu, Yan Fu, Yanlong Ma, Wen-Tian Du, Qian Li, Ran Zhan, Sicheng Tian, Lulu Yang, Ziqiao Wang, Fei Feng, Zhichao Gao, Manli Wang, Sheng Cao, Yan-Yi Wang, Rong Zhang
## Abstract
Crimean Congo hemorrhagic fever virus (CCHFV) causes fatal tick-borne disease in humans and is a priority pathogen of the World Health Organization. No licensed vaccines or specific antiviral drugs are available. To understand the cell entry of CCHFV and identify potential antiviral targets to combat the disease, here, we perform the CRISPR knockout screen in wild-type cells, followed by a complementary CRISPR activation screen in cells deficient in common attachment factors (heparan sulfate, AXL, TIM-1). We identify the soluble milk fat globule-EGF factor 8 protein (MFGE8), also known as lactadherin, as a proviral factor for CCHFV infection. Overexpression of MFGE8 enhances the pseudotyped, tecVLP, and authentic CCHFV infection, while knockout decreases the infection. MFGE8 is found to promote the virus binding and internalization. Expression of MFGE8 with D48E mutation of the RGD motif and the use of pharmacological inhibitor and gene-editing suggests that MFGE8 mediates virus entry through integrin receptors on the cell surface. Further study demonstrates that soluble MFGE8 protein acts as a bridge to support the entry by binding to not only the reported phosphatidylserine (PtdSer) but also Gc protein on viral envelope and to integrins on cells. The finding of MFGE8 that can bind directly to Gc protein and the entry mode of CCHFV that employs a soluble protein may expand the tissue tropism and increase the pathogenicity of CCHFV. Our study also provides new insight into the underlying mechanisms of cell entry and development of countermeasures for CCHFV. IMPORTANCE CCHFV causes severe hemorrhagic fever outbreaks, with a mortality rate of up to 40%. Countries generally list CCHFV as one of the pathogens that requires the highest biosafety level 4 (BSL-4) of containment, which hinders the study of its cell biology and pathogenesis. LDLR was recently identified as a receptor for CCHFV, but other receptors or co-factors remain to be explored. We perform genome-wide CRISPR screens using a safe replication-competent CCHFV pseudovirus and identify a secretory MFGE8 protein that functions as an entry mediator by binding to both the Gc protein and PtdSer on the viral envelope and to the integrins on the cells. Cell entry mediated by a soluble protein may greatly expand the tissue tropism, and the strategies developed to disturb the interaction of MFGE8 with virions or with integrins may help to mitigate the fatal disease induced by CCHFV.
KEYWORDS Crimean-Congo hemorrhagic fever virus, virus entry, MFGE8C rimean-Congo hemorrhagic fever virus (CCHFV) is a tick-borne virus causing severe disease with a wide geographical distribution and a mortality rate of 30% or higher (1-3). Cases caused by CCHFV infection have been reported in Africa, the Middle East, Asia, as well as Southern and Eastern Europe (4). CCHFV can effectively infect various wild animals, such as small rodents, rabbits, ostriches, water buffaloes, etc. Importantly, humans can also be infected by contact with livestock that do not have obvious diseases
. The US Food and Drug Administration (FDA) has not yet approved any vaccines or treatments for CCHFV (6).
CCHFV is an enveloped negative sense RNA virus belonging to the Orthonairovirus genus in the Nairoviridae family of Bunyavirales order (7). CCHFV has three genome parts, including small segment (S), medium segment (M), and large segment (L). Fragments S, M, and L encode nucleoprotein (NP), glycoprotein precursor (GPC), and RNA-depend ent RNA polymerase (RdRp), respectively (8,9). The GPC is proteolytically processed to produce two main precursor proteins, 140 kDa PreGn and 85 kDa PreGc, which respec tively produce two mature structural proteins Gn (37 kDa) and Gc (75 kDa) (10). Gn and Gc participate in receptor binding and entry (11).
Identifying entry receptors or related co-factors of CCHFV is of significance to understand its tissue and species tropism, and pathogenesis, and help to develop strategies for disease control and prevention. Previous studies reported that nucleolin and DC-SIGN are entry factors for CCHFV, but further experiments are needed to confirm (12,13). The latest research has found that low-density lipoprotein receptor (LDLR) is an entry factor of CCHFV (14)(15)(16). The Gc protein of CCHFV is shown to bind directly to LDLR, which can mediate viral entry into various cell types and play roles in infection and pathogenesis in mice (14,15). In addition, the apolipoprotein E (ApoE) is found to be incorporated into CCHFV particles and enhances its infectivity by promoting LDLR dependent entry (15,16). However, the use of LDLR blocking antibodies or soluble LDLR protein could only partially block the CCHFV infection. Similarly, genetic knockout of LDLR in cell lines or mice does not completely prevent viral infection. These results suggest that CCHFV may have other receptors or co-factors that mediate the entry in certain cell types.
To comprehensively uncover host factors that are required for CCHFV entry, we performed genome-wide CRISPR screens using replication-competent vesicular stomatitis virus-based CCHFV pseudovirus (rVSV-CCHFV) carrying the CCHFV GPC (17). From our initial CRISPR knockout screen, we identified a suite of genes that are involved in biosynthesis of heparan sulfate (HS), and some other genes, e.g., AXL and HAVCR1 (TIM-1), which are common viral attachment factors. We then conducted CRISPR activation screen in cells for which those common attachment factors are depleted and identified the milk fat globule-EGF factor 8 protein (MFGE8) that, upon upregula tion, significantly promotes pseudotyped, transcription-and entry-competent virus-like particle (tecVLP), and authentic CCHFV infection. Mechanically, the secretory MFGE8 protein mediates the virus entry through integrin receptors on the cell membrane, and MFGE8 can directly bind to not only the reported phosphatidylserine (PtSer) but also Gc protein on virions. The identification of a soluble protein mediating cell entry may significantly expand the tissue tropism and increase the pathogenicity of CCHFV which is notorious for its fatal hemorrhagic systemic disease.
## RESULTS
## CRISPR knockout screen identifies common host factors required for rVSV-CCHFV pseudovirus infection
In previous work, we demonstrated that heparan sulfate plays a significant role in the adhesion of pseudotyped rVSV-CCHFV bearing the GPC to the plasma membrane of cells (17). To comprehensively uncover host factors required for CCHFV entry, we performed genome-wide CRISPR knockout screen using rVSV-CCHFV as model virus. The viral-resist ant cells were harvested for deep sequencing and data analysis (Table S1). As expected, a suite of genes related to heparan sulfate biosynthesis, such as B3GAT3, B4GALT7, EXT1, EXT2, XYLT2, were enriched (Fig. 1A). The genes associated with ER membrane protein complex (EMC1, EMC2, EMC3, etc.), members of V-ATPases (ATP60A1, ATP61A1, etc.), components of the conserved oligomeric Golgi (COG) complex (COG3, COG5, etc.), were also identified (Fig. 1A). In addition, the TAM receptor AXL and TIM receptor HAVCR1 (TIM-1), albeit with low score, showed up. Additionally, ITGAV and ITGB5, encoding the integrins that often form as αVβ5 heterodimer, were identified (Fig. 1A). KEGG and GO analysis also showed the obvious enrichment of genes in the HS proteoglycan biosynthe sis pathway (Fig. 1B), suggesting its critical role during rVSV-CCHFV infection.
The first step for viruses to invade cells is to adhere to the cell surface. Many adhesion molecules are involved in this process. Heparan sulfate has been identified as a common attachment factor for many viruses, such as flaviviruses (18), alphaviruses (19), filoviruses (3), and bunyaviruses like Rift Valley fever virus (RVFV) (20). AXL also plays an important role in Dengue virus (21), Zika virus (ZIKV) (22,23), and SARS-CoV-2 infections (24). As to TIM-1, it has been reported to mediate the infections of Ebola virus, DENV, West Nile virus (25), and hepatitis E virus (HEV) (26). Given the common function of these genes for viral adhesion and entry, we generated A549 triple-knockout cell line that B3GAT3, AXL, and TIM-1 were all depleted (A549-BAT), to examine the effect on rVSV-CCHFV infection. Surprisingly, as compared to wild-type A549 cells (A549-WT), A549-BAT showed a nearly complete resistance to viral infection at 12, 16, and 24 h (Fig. 1C). A549-WT was highly susceptible to rVSV-CCHFV, with the GFP-positive cell rate of up to 70% at 16 h postinfection, while A549-BAT remained largely uninfected at 12, 16, and 24 h, with only 2% of infected-cells at 24 h (Fig. 1C). The representative fluorescent images also showed the marked decrease of infection efficiency in A549-BAT cells for rVSV-CCHFV (Fig. 1D). These results indicated that the infection of rVSV-CCHFV is highly dependent on these common adhesion factors.
## CRISPR activation screen identifies MFGE8 as a proviral host factor for rVSV-CCHFV infection
To identify other potential entry factors of CCHFV independent of common adhesion molecules such as heparan sulfate as described above, we performed a complementary genome-wide CRISPR activation screen in A549-BAT cells. The cell library was infected with rVSV-CCHFV, and GFP-positive cells were sorted for subsequent extraction of genomic DNA, sgRNA sequencing, and data analysis (Table S1). As shown in Fig. 2A, the top hits from the screen were determined according to their MAGeCK scores and P values. After extensive validation with CRISPR activation sgRNAs in A549-BAT cells, we found that MFGE8 showed around 20-fold increase of infection for rVSV-CCHFV, but not rVSV, as determined by flow cytometry and microscopy (Fig. 2B through D). The upregulation of endogenous MFGE8 expression by CRISPR activation sgRNAs was also validated by Western blotting analysis (Fig. S1A).
MFGE8 is a lactadherin preproprotein involved in recognizing and engulfing apoptotic cells, wound healing, cancer, autoimmune disease, and inflammation resolution (27). Park et al. proposed that the blood levels of MFGE8 and the lymphoid organ sources of MFGE8 ensure that the majority of cell-free HIV-1 virus in infected individuals is exposed to MFGE8, and the HIV-1 envelope phosphatidylserine (PtdSer) provides a means for MFGE8 binding, which links viral particles to the αv integrin on host cells (28). Therefore, MFGE8 possibly also mediated the rVSV-CCHFV infection in a similar way.
We then performed Western blotting to detect the endogenous expression levels of intracellular and secreted MFGE8 in these cell lines. MFGE8 showed high expression in SW-13 cells, moderate expression in A549 and A549-BAT cells, and the lowest expression level in HeLa cells (Fig. S1B). To further verify the role of MFGE8 in promoting rVSV-CCHFV infection, we overexpressed MFGE8 cDNA not only in resistant A549-BAT cells, but also in wild-type A549, HeLa, and SW-13 cells. The expression levels of MFGE8 were validated by Western blotting analysis (Fig. S1C). As expected, the empty vector control showed infection efficiency of approximately 1%-3% in A549-BAT cells, while overexpression of MFGE8 increased the viral infection to around 50% (Fig. 2E). Viral growth kinetics were analyzed by measuring the titers in the supernatants. Starting from 12 h post-infection, the supernatants of MFGE8-overexpressing A549-BAT cells exhibited significantly higher viral titers compared to vector control cells (Fig. 2F). Additionally, overexpression of MFGE8 in wild-type A549, HeLa, and SW-13 cells significantly enhanced the rVSV-CCHFV infection (Fig. 2G through I). Next, we knockout the MFGE8 in wild-type A549 cells using CRISPR/Cas9 editing system and verified the knockout efficiency by Western blotting (Fig. S1D). It showed that editing of MFGE8 significantly decreased rVSV-CCHFV infection as compared to the control (Fig. 2J). Thus, these results suggested that MFGE8 is a proviral host factor for rVSV-CCHFV infection.
## MFGE8 promotes the infection of pseudotyped, tecVLP, and authentic CCHFV across multiple strains
To verify whether MFGE8 promotes the infection of multiple CCHFV strains, we subsequently packaged single-round VSV-based pseudoviruses bearing the GPCs from four different strains (Oman, Turkey, Afg2990, and YL16070). As expected, overexpression of MFGE8 in A549-BAT cells significantly enhanced the luminescence readings of all four pseudovirues (Fig. 3A). Moreover, in wild-type SW-13 cells, MFGE8 overexpression could increase the infection of four CCHFV pseudoviruses (Fig. 3B). Conversely, knockout of MFGE8 in A549-WT cells obviously inhibited the infection of these pseudoviruses (Fig. 3C).
Next, we attempted to confirm the proviral function of MFGE8 using authentic YL16070 strain of CCHFV available. Through quantifying viral RNA in infected cells at 72 h post-infection, overexpression of MFGE8 in A549-BAT cells increased CCHFV infection by around 20 folds (Fig. 3D). Likewise, MFGE8 could enhance the virus infection in overex pressing A549-WT cells (Fig. 3E). Similarly, knockout of MFGE8 reduced authentic CCHFV infection (Fig. 3F). We also investigated the effects of MFGE8 on tecVLP infectivity. Overexpression of MFGE8 in A549-BAT cells significantly enhanced the infection of tecVLPs of Oman and Turkey strains (Fig. 3G andH). Moreover, knockout of MFGE8 in A549-WT cells led to reduced infection of Turkey tecVLP (Fig. 3I). Therefore, based on the results of pseudotyped, tecVLP, and authentic viruses, it suggested that MFGE8 is a common proviral host factor for CCHFV across different strains.
## MFGE8 mediates rVSV-CCHFV entry into host cells through the integrins
Given the enhancement of MFGE8 for rVSV-CCHFV but not rVSV infection as described above in Fig. 2B andC, we speculated that MFGE8 affects the cell entry stage of rVSV-CCHFV that bears the GPC. The binding and internalization assays were performed, and indicated that more rVSV-CCHFV virions bind and enter MFGE8-overexpressing A549-BAT cells. Compared with vector control cells, overexpression of MFGE8 promoted the binding by 1.5 times and internalization by seven times (Fig. 4A). In MFGE8-editied A549-WT cells, the internalized rVSV-CCHFV was decreased by approximately 1.3 times as compared to the sgRNA control cells (Fig. 4B). The binding of virions in MFGE8-edited cells was not affected, probably due to the strong non-specific binding in A549-WT cells that express an abundance of attachment factors, such as heparan sulfate.
It has been shown that pre-incubation with recombinant MFGE8 protein can promote the transduction of lentiviruses packaged with alphavirus or baculovirus glycoproteins, via bridging the binding to PtdSer on the lentivirus envelope and to the integrins on cell surface (29). We investigated whether the recombinant MFGE8 protein can also increase the infection of rVSV-CCHFV. We pre-incubated A549-BAT cells with MFGE8 protein, starting at a concentration of 20 µg/mL with a serial of twofold dilutions, followed by infection with rVSV-CCHFV. It showed that the highest concentration of MFGE8 protein could increase the infection efficiency by about 30 times for rVSV-CCHFV but has no effect on rVSV infection (Fig. 4C). Similarly, we found that pre-incubation of MFGE8 protein with rVSV-CCHFV can markedly improve the infection efficiency (Fig. 4D). As negative control, the mpox virus B6 protein had no effect on rVSV-CCHFV infection (Fig. 4D). We also found that MFGE8 protein can promote the infection of IbAr10200 and Turkey tecVLPs in a dose-dependent manner (Fig. 4E andF).
MFGE8 has been reported to bind to integrin αVβ3 or αVβ5 receptor (27). Integrins recognize Arg-Gly-Asp (RGD) motif in their physiological ligands (30), and the epidermal growth factor (EGF) domain of MFGE8 displays an RGD motif (31,32). We attempted to explore whether MFGE8 promotes rVSV-CCHFV infection through integrins. Interestingly, in contrast to the wild-type MFGE8 that enhances the infection as presented above, overexpression of D48E mutant that disrupts the "RGD" motif in the EGF domain of MFGE8 resulted in a significant reduction of rVSV-CCHFV infection (Fig. 4G). The expres sion of MFGE8 mutant was validated by Western blotting analysis (Fig. S1E). We also used the Cilengitide, a cyclized RGD-containing pentapeptide that potently and selectively inhibits αvβ3 and αvβ5 integrins, to examine its effect on MFGE8-mediated infection. It showed that around 0.3 µM of Cilengitide could efficiently dampen the increased infection by MFGE8 overexpression, while 5 µM completely inhibited the proviral function of MFGE8 (Fig. 4H andI). The cytotoxicity of Cilengitide on cells was checked (Fig. S2).
To further investigate which integrin mediates the proviral function of MFGE8, we knocked out ITGB5 and ITGB3 in MFGE8-overexpressing A549-BAT cells, respectively. The results showed that ITGB5 knockout completely abolished the proviral effect of MFGE8, whereas ITGB3 knockout had subtle impact on infection efficiency compared to the sgControl group (Fig. 4J). This indicated that MFGE8 promotes CCHFV entry into host cells primarily through ITGB5, not ITGB3. Moreover, editing of ITGB5 with specific sgRNA significantly inhibited the rVSV-CCHFV infection in A549-WT cells (Fig. 4K). The above experiments indicated that the promotion of viral infection by MFGE8 is mediated by integrins.
## MFGE8 binds directly to the Gc protein of CCHFV
MFGE8 has a signal peptide located at the N-terminus, which plays a crucial role in guiding it into the secretion pathway. MFGE8 also possesses an EGF domain, followed by two coagulation domains (C1 and C2) (Fig. 5A). To identify critical regions of MFGE8 that are required for rVSV-CCHFV infection, we generated truncated variants lacking ΔEGF, ΔC1, ΔC2, or ΔC1C2 and validated the expression of ΔEGF and ΔC2 by Western blotting analysis but not ΔC1 and ΔC1C2 because the antibody only recognizes the C1 domain (Fig. 5B; Fig. S1F). Overexpression of truncated variants in A549-BAT cells showed that, compared to full-length MFGE8, ΔEGF completely lose their ability to promote rVSV-CCHFV infection. ΔC1 and ΔC1-C2 possibly also abolished the function, given the uncertainty of their expression. While ΔC2 showed less pronounced proviral effects compared to full-length MFGE8, it still increased infectivity two-fold (Fig. 5C). The deletion of EGF domain disrupts the interaction between MFGE8 and integrins, while the deletion of C1 and/or C2 domains may impair the binding to the virus. These results indicated that the three domains of MFGE8 are all possibly required for promoting rVSV-CCHFV infection.
To further assess the function of soluble MFGE8, we engineered chimeric MFGE8 proteins to render their expression on the plasma membrane (Fig. 5D). The soluble MFGE8 without the signal peptide was fused with a type II transmembrane protein, TMPRSS2, to replace the extracellular region. Similarly, the C1 and C2 domains of MFGE8 were fused with a type I transmembrane protein, MXRA8, to replace the extracellular region. TMPRSS2 and MXRA8 are two cell surface proteins required for coronavirus and alphavirus entry, respectively (33,34). The two engineered MFGE8 proteins were expressed and displayed on the cell surface similar to the soluble MFGE8 that is anchored to the surface through the integrins. Expression of engineered MFGE8 proteins in A549-BAT cells significantly enhanced rVSV-CCHFV infection, particularly the display of C1 and C2 domains on the cell surface in the backbone of MXRA8 (Fig. 5D). It has been shown that MFGE8 is a secreted protein that binds to the PtdSer through its C1 and C2 domains (32). Shao et al. further confirmed that, by mutating the "WGL" at amino acids 26-28 of MFGE8 to "AAA" or the "FG" at amino acids 81-82 to "AA, " the binding affinity between MFGE8 and membrane PtdSer decreased over 90% (35). To determine whether MFGE8 promotes rVSV-CCHFV infection by binding to the PtdSer on the evelople of viral particles, we constructed "WGL-AAA" and "FG-AA" mutants of MFGE8 for overexpression and validated by Western blot analysis (Fig. S1G). When compared to the full-length MFGE8, the mutants decreased the virus infection from 80% to 30% (Fig. 5F), suggesting that the binding of MFGE8 to PtdSer plays a significant role. However, the infection efficiency for mutants was still around 10-fold higher than the vector control (Fig. 5F).
The unexpected high infection efficiency for MFGE8 mutants that disrupt the binding to PtsSer implied that MFGE8 may also bind to other components of virions, such as the Gc protein that is exposed on the virion surface for receptor binding. To test this hypothesis, we expressed and purified Gc and Gn proteins of CCHFV (Fig. S3) and coated for ELISA-based binding assay. Interestingly, we found that MFGE8 is captured by the Gc monomer protein, with much higher binding ability than the Gn protein (Fig. 5G). An unrelated viral protein, mpox virus surface B6 protein, was used as control. Additionally, the MFGE8 or B6 control protein was coated as bait, and the increasing concentrations of Gc monomer or trimer protein was added for capture. The Gc trimer assembles in the endosome to facilitate membrane fusion. As expected, both Gc monomer and trimer could apparently bind to the MFGE8, but not B6 protein (Fig. 5H andI). Likewise, the rVSV-CCHFV particles could bind to the MFGE8 protein (Fig. 5J). What's more, biological layer interferometry showed that MFGE8 protein can efficiently bind to the Gc monomer of CCHFV, with the K d < 0.001 nM (Fig. 5K). In silico analysis revealed that both C1 and C2 regions of MFGE8 contribute to binding to CCHFV Gc trimer (Fig. S4).
In summary, we demonstrated that the enhancement of CCHFV infection by secretory MFGE8 protein may be the synergistic effect of binding to both PtdSer and Gc protein on the surface of CCHFV virions (Fig. 5L). The soluble MFGE8 acted as a bridge by binding to the virions and to the integrins on the cell plasma membrane to mediate the cell entry of CCHFV (Fig. 5L).
## DISCUSSION
CCHFV infects various animals, and humans suffer from serious diseases. Elucidating the cell entry process of CCHFV and identifying host factors that are required for entry will be of significance to understand the tissue tropism and pathogenesis and to develop antivirals targeting the invasion. Here, we performed genome-wide CRISPR knockout screen and identified some common host factors that are important for CCHFV entry. Additionally, we performed genome-wide CRISPR activation screen and identified the MFGE8 as a proviral host factor for CCHFV. Overexpression of MFGE8 could significantly promote the infection of pseudotyped, tecVLP, and authentic CCHFV, while knockout of MFGE8 decreased the infection. Pre-treatment of cells with recombinant MFGE8 protein or pre-incubation with virus could dose-dependently increase the rVSV-CCHFV and tecVLP infection. More importantly, we demonstrated that, in addition to the binding 1 h, then used to infect A549-BAT cells pre-transfected with plasmids expressing the L and N proteins. After 15 h of infection, NanoLuc activity was quantified.
## (G). Overexpression of MFGE8 with D48E mutation decreases the rVSV-CCHFV (MOI 3, 20 h) infection in A549-BAT cells. The percentage of GFP-positive cells were analyzed by flow cytometry. (H-I). The inhibitor of integrins decreases the rVSV-CCHFV infection. The MFGE8-overexpressing A549-BAT cells were pre-treated with integrin inhibitor cilengitide for 2 h at 37°C, and then infected with rVSV-CCHFV (MOI 3, 20 h) for flow cytometry analysis (H). The representative images were
taken with fluorescence microscope (I). Scale bar, 400 µm. (J). MFGE8-overexpressing A549-BAT cells were edited with control, ITGB5-, or ITGB3-specific sgRNAs, and infected with rVSV-CCHFV (MOI 3, 17 h), followed by flow cytometry analysis. (K). A549-WT cells edited with control or ITGB5 sgRNA, followed by infection with rVSV-CCHFV (MOI 0.5, 16 h) for flow cytometry analysis. Unpaired t-tests (A, B, K), One-way ANOVA with Dunnett's multiple comparisons test (E, F, G, H, J). ns, not significant; *, P < 0.1, **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.
## Research Article mBio
October ability with PtdSer as reported previously, MFGE8 also can bind directly to the CCHFV Gc protein. The binding to both molecules on the virion surface will apparently enhance the cell entry into cells.
Studies have shown that integrins play important roles in mediating the entry of a wide range of viruses, mostly via the direct interaction with a RGD motif exposed on the surface proteins of virions, such as SARS-CoV-2 (30), Zika (36), Foot and mouth disease virus (FMDV) (37), Varicella Zoster Virus (38), herpes simplex virus (HSV) (39), Hantaan (40), and other viruses (41). However, the integrins can also bind to host proteins, such as the soluble MFGE8 that possesses the RGD motif in the EGF domain, serving as a bridge between viruses and cells for entry. It has been shown that MFGE8 can enhance the transduction with lentiviruses pseudotyped with various envelope proteins, including Sindbis virus, Ross River virus, and baculovirus (gp64), through the binding to PtdSer on virus envelope, and to the integrins on the cell surface (29). As for CCHFV studied here, MFGE8 functions as an entry factor through similar mechanism of action. The exceptional finding in MFGE8 can also bind to the CCHFV Gc protein. It is noteworthy that both ITGAV and ITGB5 which form as the integrin αVβ5 heterodimer were identified from our initial CRISPR knockout screen, and proviral function of ITGB5 was validated. Whether the integrins can also directly interact with surface Gc protein to mediate the CCHFV entry needs to be further studied.
LDLR was recently identified as a receptor for CCHFV (14)(15)(16). Pre-treatment with anti-LDLR monoclonal antibodies can dose-dependently reduce CCHFV infection in primary HUVECs, with limited inhibitory effect on primary human PBMCs (14), suggest ing that there may be other receptors or co-factors present in specific types of cells for CCHFV infection. It also revealed that the exchangeable host ApoE incorporated into CCHFV particles mediates the interaction of virions with receptor LDLR or LPR8, to facilitate CCHFV entry into cells (15,16). In this study, the identification of soluble MFGE8 represents another mode of CCHFV entry, which is very similar to the Gas6 protein via the TAM receptors (Tyro3, Axl, and Mer) employed by many enveloped viruses (21,29,(42)(43)(44).
CCHFV replicates efficiently in cells such as epithelial cells, dendritic cells, and tissue resident macrophages. The productive infection of these cells promotes the spread of the virus and leads to early infection of local lymph nodes and peripheral blood monocytes, supporting the systemic transmission of the virus (45)(46)(47). MFGE8 is widely expressed in the liver, kidney, intestine, lung, mammary gland, brain, heart, spleen, and reproductive organs. After CCHFV infection, the liver is one of the main target organs. The highly expressed MFGE8 in the liver may directly interact with the CCHFV surface glycoprotein Gc and PtdSer, assisting in the internalization of the virus. Additionally, due to the wide distribution of cell types expressing the integrins, the viral particles captured by soluble MFGE8 in blood may significantly expand the tissue tropism of CCHFV, leading to the occurrence of systemic infection. Moreover, MFGE8 is a component of the milk (48). Although some studies indicated that breastfeeding transmission or virus in milk region of TMPRSS13 or MXRA8 was replaced. (E). Overexpression of chimeric MFGE8 proteins in A549-BAT cells, followed by infection with rVSV-CCHFV (MOI 3, 20 h) and flow cytometry analysis. (F) Overexpression of different mutants of MFGE8 in A549-BAT cells, followed by infection with rVSV-CCHFV (MOI 3, 20 h) and flow cytometry analysis. One-way ANOVA with Dunnett's multiple comparisons test (C, E, and F). ns, not significant; *P < 0.1; ***P < 0.001; ****P < 0.0001. (G) Binding of MFGE8 protein to CCHFV Gc monomer, Gn, and Mpox virus B6 control protein. These proteins were pre-coated onto ELISA plates at a concentration of 10 µg/mL, followed by incubation of different concentrations of MFGE8 protein. The binding was evaluated using anti-MFGE8 and HRP-conjugated secondary antibodies. (H-K) MFGE8 and B6 control proteins were coated onto ELISA plates at a concentration of 10 µg/mL. Then, different concentrations of CCHFV Gc monomer (H), Gc trimer (I), or rVSV-CCHFV particles (J) were incubated and detected using anti-CCHFV-Gc ADI-36121 antibody and HRP-conjugated secondary antibody. (K) The binding affinity of MFGE8 to CCHFV Gc domain. His-tagged MFGE8 protein was immobilized onto the Ni-NTA biosensors. Binding parameters of recombinant Gc monomer to MFGE8 protein were measured by biolayer interferometry (BLI). Using a 1/1 binding model to derive equilibrium dissociation constant (Kd) values. (L) The model of soluble MFGE8-mediated cell entry of CCHFV. The soluble MFGE8 protein binds directly to both Gc protein and phosphatidylserine (PtdSer) on the surface of CCHFV virions and acts as a bridge by binding to the integrins on the cell membrane to promote virus infection.
was not detected (49,50), whether MFGE8 plays a role in promoting the transmission via breast milk warrants further study.
The wide geographical distribution and large population of CCHFV infection urge the need to explore the host determinants of the tropism and pathogenesis. More rapid and reliable diagnosis, as well as effective vaccines and antiviral drugs, are needed to limit the burden of CCHFV on patients and public health. Our study identified a soluble host factor MFGE8 involved in the cell entry of CCHFV, and the strategies developed to disturb the interaction of MFGE8 with virions or with integrins may help to prevent and control the endemic of CCHFV in high-risk areas. Our study also provides new insight into the underlying mechanisms of cell entry of CCHFV.
## MATERIALS AND METHODS
## Cells and viruses
Vero E6 (Cell Bank of the Chinese Academy of Sciences, Shanghai, China), A549 (ATCC #CCL-185), HeLa (ATCC #CCL-2), SW13 (TCHu221), HEK 293T (ATCC # CRL-3216) all were cultured at 37°C in Dulbecco's modified Eagle's medium (Hyclone #SH30243.01) supplemented with 10% fetal bovine serum (FBS), 10 mM HEPES, 1 mM sodium pyruvate, 1 × non-essential amino acids, and 100 U/mL of penicillin-streptomycin. A549-BAT is a clonal cell line generated in our laboratory that B3GAT3, AXL, and TIM1 are deficient. All cell lines were routinely screened for mycoplasma contamination and found to be negative. The recombinant vesicular stomatitis virus expressing the GFP protein (rVSV-GFP, thereafter rVSV) and replication-competent VSV-based CCHFV pseudovirus (rVSV-CCHFV) carrying both CCHFV glycoprotein precursor (GPC) and GFP reporter were generated previously (17) and propagated and titrated in BHK-21 cells. CCHFV authentic virus (YL16070 strain), was propagated in Vero E6 cells and titrated in SW13 cells. All experiments involving CCHFV authentic virus were performed in the biosafety level 3 (BSL-3) facility of Wuhan Institute of Virology, Chinese Academy of Sciences, following the regulations.
## CRISPR screens
The human Brunello CRISPR knockout pooled library targeting 19,114 genes (Addgene #73178) or Calabrese activation pooled library targeting 18,885 genes (Addgene #92379) was a gift from David Root and John Doench (51) and packaged in 293 FT cells after co-transfection with psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259) using FugeneHD (Promega). At 48 h post transfection, supernatants were harvested, clarified by spinning at 3,000 rpm for 15 min, filtered, and aliquoted for storage at -80°C.
A549-Cas9 cells were generated by transduction of wild-type A549 with a packaged lentivirus lentiCas9-Blast (Addgene #52962). A549-BAT-dCas9 cells were generated by transduction of A549-BAT with a packaged lentivirus lenti dCAS-VP64_Blast (Addgene #61425). A549-Cas9 or A549-BAT-dCas9 cells were transduced with respective knockout or activation library at a multiplicity of infection (MOI) of ~0.3 by spinoculation at 1000 g and 32°C for 30 min in 12-well plates, followed by selection with puromycin for around 7 days. For knockout screen, cells were inoculated with rVSV-CCHFV pseudovirus (MOI 3) and incubated until nearly all cells were killed. The medium was changed, and remaining live cells grew to form colonies. The cells were then harvested and re-plated to the flasks. For activation screen, cells were inoculated with rVSV-CCHFV pseudovirus (MOI 3) for 24 h, and cells were harvested and sorted for the GFP positive population. Genomic DNA from surviving or sorted cells and uninfected cells was extracted for sgRNA amplification and next generation sequencing using an Illumina NovaSeq 6000 platform. The sgRNA sequences targeting specific genes were trimmed using the FASTX-Toolkit (http:// hannonlab.cshl.edu/fastx_toolkit/) and cutadapt 1.8.1 and further analyzed for sgRNA abundance and gene ranking by a published computational tool (MAGeCK) (see Table S1).
## Generation of overexpressing and knockout cells
To construct the overexpressing cells, human MFGE8 cDNA was purchased from SinoBiological (NM_005928.1) and cloned into pLV-EF1α-IRES-Puro vector (Addgene #85132). The full-length MFGE8 without signal peptide sequences were amplified and fused with TMPRSS2 gene to replace the extracellular region. Similarly, the C1 and C2 domains of MFGE8 were amplified and fused with MXRA8 gene to replace the extracellu lar region. The resulting lentivectors were co-transfected with helper plasmids psPAX2 (Addgene #12260) and pMD2.G (Addgene #12259) at a ratio of 2:2:1 into 293T cells to package the lentivirus. After 48 h, the cell supernatant was collected. Transduction of lentivirus into A549-BAT, A549, Hela, and SW13 cells was performed to overexpress the MFGE8 protein. After screening with puromycin for 7 days, the surviving cells were used for Western blotting validation of overexpression. To construct the knockout A549 cells, sgRNAs targeting the human MFGE8, ITGB3, and ITGB5 were synthesized and cloned into lentiCRISPR v2 (Addgene #52961) vector. The packaging of lentivirus, transduction, and puromycin selection is the same as the construction of overexpression cells above. The sgRNA sequences were listed in Table S2.
## Virus binding and internalization assay
For binding assay, A549-BAT or A549-WT cells were seeded in a 24-well plate. The next day, the plate with confluent cells was placed on ice for 15 min, washed twice with pre-cooled PBS, and then incubated with rVSV-CCHFV virus (MOI 5). After being placed on ice for 45 min, cells were washed five times with pre-cooled PBS and then lysed with RL lysis buffer (TIANGEN # DP430) for RNA extraction. For the internalization assay, after virus binding and wash as described above, cells were added with 2% FBS DMEM and placed in a 37°C incubator for 45 min. Uninternalized virions on the cell surface were removed by treating cells with 400 µg/mL protease K on ice for 45 min. After three washes, cells were lysed by RL lysis buffer for RNA extraction. The One Step PrimeScript RT-PCR Kit (Takara #RR064A) was used for RT-qPCR. The relative amount of bound or internalized virions was normalized to internal control GAPDH, followed by normalizing to the control cells. The primer sequences were listed in Table S2.
## Flow cytometry
The rVSV or rVSV-CCHFV infected cells were washed once with PBS, trypsinized, and then inactivated with 10% FBS DMEM. The cells were fixed with 2% PFA for 10 min and then centrifuged at 1,500 rpm for 5 min. The supernatant was discarded, and the cells were resuspended, washed with PBS, and subjected to flow cytometry (Thermo, Attune NxT). The percentage of GFP-positive cells was analyzed using the FlowJo v10.0.7.
## Packaging of single-round VSV pseudotyped with the GPCs from different CCHFV strains
The optimized GPC sequences of Oman (KR864901), Turkey (KR864902), Afg2990 (HM452306.1), and YL16070 (KY354082) were cloned into pCAGGS vector. 293T cells were transfected with plasmids for 24 h. After washing, cells were infected for 2 h with single-round VSV ΔG -Nluc-GFP virus (17) and then washed three times, followed by incubation with 2% FBS DMEM containing anti-VSV-G neutralizing antibody. After 24 h, the supernatants were collected and stored at -80°C for use.
## Packaging of transcription-and entry-competent virus-like particles of CCHFV
BHK-T7 cells were seeded in 6-well plates and cultured overnight. Cells were co-trans fected with the minigenome plasmid and helper plasmids (pCAGGS-CCHFV-N, pCAGGS-CCHFV-L, pCAGGS-CCHFV-GPC) using Fugene HD transfection reagent. At approximately 16-18 h post-transfection, the transfection medium was discarded and replaced with fresh complete medium. Three days post-transfection, tecVLP-containing supernatants were harvested and stored at -80°C.
## Plaque-forming assay
Briefly, Vero cell monolayers in 96-well plates were inoculated with serially diluted virus for 2 h and then overlaid with methylcellulose for 48 h. Cells were fixed with 2% PFA for 1 h, stained with crystal violet at room temperature for 15 min, and plaques were counted under microscopy.
## Protein blocking assay
Human recombinant MFGE8 protein with His tag at the C-terminus was purchased from SinoBiological (10853-H08B). The MFGE8 protein concentration was twofold diluted starting from 20 µg/mL. A549-BAT cells were incubated with different concentrations of MFGE8 protein at 37°C for 4 h and then washed once with PBS, followed by infection with rVSV-CCHFV virus (MOI 3). Cells were collected for flow cytometry analysis after 20 h. Alternatively, different concentrations of MFGE8 protein were pre-incubated with rVSV-CCHFV virus at 37°C for 1 h and then added to A549-BAT cells for infection for 20 h. Cells were then collected for flow cytometry analysis. For assays with tecVLP infection, A549-BAT cells in 6-well plates were transfected with pCAGGS-L and pCAGGS-N for 20 h. Cells were trypsinized and seeded into 96-well plates. TecVLPs were pre-incubated with serial dilutions of recombinant MFGE8 protein (0-10 μg/mL) for 1 h at 37°C prior to infection. At 15 h post-infection, cells were lysed with Nano-Glo Luciferase Lysis Buffer (Promega, Cat# N1120) for 10 min at room temperature. Luminescence was quantified using the FlexStation 3 (Molecular Devices).
## Cell viability assay
Cells were seeded in 96 well plates with a density of 5 × 10 4 . The next day, cells were treated with Cilengitide at the concentrations of 0, 0.625, 1.25, 2.5, 5, and 10 µM. After 24 h, 100 µL of the CellTiter-Lumi Luminescence-based Cell Vitality Detection Kit (Beyotime #C0065M) was added to each well. After incubation for 10 min at room temperature, luminescence was recorded by using a FlexStation 3 (Molecular Devices) with an integration time of 1 s per well.
## Integrin inhibition assay
A549-BAT were seeded in 96 well plates with a density of 5 × 10 4 . The next day, cells were treated with different concentrations of Cilengitide to inhibit the αvβ3 and αvβ5 for 2 h at 37°C, and DMSO was used as control. Cells were then infected with rVSV-CCHFV virus (MOI 3) in the presence of inhibitor. After 20 h, cells were collected for flow cytometry analysis.
## Authentic CCHFV infection
A549-BAT or A549-WT cells in 12-well plates were inoculated with authentic CCHFV (YL16070 strain, GenBank accession number: KY354082) at MOI of 0.01-0.25, and cells were lysed at 72 h post-infection. The virus replication was determined by analyzing the copies of S genome using RT-qPCR. The primer sequences were listed in Table S2.
## Western blotting
Cells were lysed in RIPA buffer (Beyotime #P0013B) containing protease inhibitors (Sigma-Aldrich #S8830). Cell lysates were clarified by centrifugation at 12,000 rpm for 10 min at 4°C. Samples were denatured at 95°C for 10 min, in reducing loading buffer (50 mM Tris, pH 6.8, 10% glycerol, 2% SDS, 0.02% [wt/vol] bromophenol blue, 100 mM DTT) and electrophoresed in 10% SDS polyacrylamide gels, and proteins were transfer red to PVDF membranes. Membranes were blocked with 5% non-fat dry powdered milk in TBST (100 mM NaCl, 10 mM Tris, pH7.6, 0.1% Tween 20) for 1 h at room temperature and then incubated with primary antibodies to detect MFGE8 (MFG-E8 monoclonal antibody, Proteintech, #67797-1-PBS, 1:3000) or actin (Beta Actin polyclo nal antibody, Proteintech, # 20536-1-AP, 1:3,000) at 4°C overnight. After three washes with TBST, the membrane was incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for 1 h. The membrane was washed again with TBST three times, each time for 10 min, and developed using SuperSignal West Pico or Femto chemiluminescent substrate according to the manufacturer's instructions (Thermo Fisher).
## Protein expression and purification
Expression plasmids based on pCAGGS vector were constructed by inserting CCHFV Gc10200 aa 1,041-1,579, Gn10200 aa 520-690, and mpox virus B6 aa 1-279 fused with the Strep-Tag II at the C-terminus. On the day before the experiment, FreeStyle 293 F cells were seeded at a density of 5 × 10 5 . The next day, expression plasmids were transfected using EZ-Trans transfection reagent (AC04L092). The supernatants were collected 4 days after transfection, centrifuged at 8,000 × g for 30 min, and used the Strep-TactinXT 4Flow resin for purification. The purified protein was dialyzed in PBS, filtered through a 0.2 µm filter, and stored at -80°C. Protein purity was confirmed through SDS-PAGE and Coomassie Brilliant Blue staining.
## ELISA-based protein-binding assay
96-well EIA/RIA high protein-binding affinity plate (Corning) was coated with purified CCHFV Gn, Gc monomer, or B6 protein (10 µg/mL per well) overnight at 4°C and blocked with 4% bovine serum albumin in PBS. Blocking buffer was removed by washing five times with PBST. MFGE8 protein was added at different concentrations and incubated at room temperature for 1 h. After five washes with PBST, primary antibody (MFG-E8 Monoclonal antibody, Proteintech #67797-1-PBS, 1:3,000) was incubated at room temperature for 1 h. After another five washes with PBST, HRP-conjugated secondary antibody was incubated at room temperature for 1 h. The 3,3′,5,5′-Tetramethylbenzidine (TMB) substrate (Invitrogen #34028) was added to each well at a volume of 100 µL and left for 15 min at room temperature to allow the colorimetric reaction to occur. The reaction was stopped by addition of 100 µL of 2 M H 2 SO 4 . Absorbances were read by an automated plate spectrometer at a wavelength of 450 nm and analyzed using SoftMax Pro software. Conversely, the plates were coated with MFGE8 or B6 control protein (10 µg/mL per well) overnight at 4°C and bound with different concentrations of Gc monomer, Gc trimer protein, or rVSV-CCHFV particles. Similar procedure was performed as described above.
## Bio-layer interferometry
The binding affinity between MFGE8 and Gc monomer of the CCHFV IbAr10200 strain was measured using the ForteBio Octet Red system (ForteBio, Inc). The His-tagged MFGE8 protein (15 µg/mL) was immobilized onto the NTA biosensors; the association and dissociation of the indicated concentrations of Gc protein to MFGE8 were monitored in 200 µL of PBS containing 0.02% Tween 20 and 0.1%BSA. The binding curves and kinetics of association and dissociation were analyzed using the ForteBio Data Analysis Software.
## Statistical analysis
GraphPad Prism version 10.3.1 software was used for graphical representation and statistical analysis. Two-tailed unpaired t-tests or two-way analysis of variance (ANOVA) with Sidak's multiple-comparison test or Dunnett's multiple-comparison test were used for data analysis. P values under 0.05 were considered statistically significant and the following denotations were used: ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05; ns (not significant), P > 0.05.
## References
1. Khan, Maupin, Rollin et al. (1997) "An outbreak of Crimean-Congo hemorrhagic fever in the United Arab Emirates, 1994-1995" *Am J Trop Med Hyg*
2. Kuehnert, Stefan, Badger et al. (2021) "Crimean-Congo hemorrhagic fever virus (CCHFV): a silent but widespread threat" *Curr Trop Med Rep*
3. Papa, Tsergouli, Tsioka et al. (2017) "Crimean-Congo hemorrhagic fever: tick-host-virus interactions" *Front Cell Infect Microbiol*
4. Hawman, Feldmann (2023) "Crimean-Congo haemorrhagic fever virus" *Nat Rev Microbiol*
5. Spengler, Bergeron, Rollin (2016) "Seroepidemiological studies of Crimean-Congo hemorrhagic fever virus in domestic and wild animals" *PLoS Negl Trop Dis*
6. Frank, Weaver, Raabe "State of the Clinical Science Working Group of the National Emerging Pathogens Training, Education Center's Special Pathogens Research Network2, State of the Clinical Science Working Group of the National Emerging Pathogens Training Education Center's Special Pathogens Research Network. 2024. Crimean-Congo hemorrhagic fever virus for clinicians-diagnosis, clinical management, and therapeutics" *Emerg Infect Dis*
7. Garrison, Alkhovsky, Sv et al. (2020) "ICTV virus taxonomy profile: Nairoviridae" *J Gen Virol*
8. Carter, Surtees, Walter et al. (2012) "Structure, function, and evolution of the Crimean-Congo hemorrhagic fever virus nucleocapsid protein" *J Virol*
9. Zivcec, Scholte, Spiropoulou et al. (2016) "Molecular insights into Crimean-Congo hemorrhagic fever virus" *Viruses*
10. Vincent, Sanchez, Erickson et al. (2003) "Crimean-Congo hemorrhagic fever virus glycoprotein proteolytic processing by subtilase SKI-1" *J Virol*
11. Hulswit, Paesen, Bowden et al. (2021) "Recent advances in bunyavirus glycoprotein research: precursor processing, receptor binding and structure" *Viruses*
12. Xiao, Feng, Zhu et al. (2011) "Identification of a putative Crimean-Congo hemorrhagic fever virus entry factor" *Biochem Biophys Res Commun*
13. Suda, Fukushi, Tani et al. (2016) "Analysis of the entry mechanism of Crimean-Congo hemorrha gic fever virus, using a vesicular stomatitis virus pseudotyping system" *Arch Virol*
14. Xu, Du, Wang et al. (2024) "LDLR is an entry receptor for Crimean-Congo hemorrhagic fever virus" *Cell Res*
15. Monteil, Wright, Dyczynski et al. (2024) "Crimean-Congo haemorrhagic fever virus uses LDLR to bind and enter host cells" *Nat Microbiol*
16. Ritter, Canus, Gautam et al. (2024) "The lowdensity lipoprotein receptor and apolipoprotein E associated with CCHFV particles mediate CCHFV entry into cells" *Nat Commun*
17. Ma, Ma, Wang et al. (2025) "The single amino acid change of R516K enables efficient generation of vesicular stomatitis virus-based Crimean-Congo hemorrhagic fever reporter virus" *J Med Virol*
18. Gao, Lin, He et al. (2019) "Role of heparan sulfate in the Zika virus entry, replication, and cell death" *Virology (Auckl)*
19. Gardner, Ebel, Ryman et al. (2011) "Heparan sulfate binding by natural eastern equine encephalitis viruses promotes neurovirulence" *Proc Natl Acad Sci*
20. Riblett, Blomen, Jae et al. (2016) "A haploid genetic screen identifies heparan sulfate proteoglycans supporting Rift Valley fever virus infection" *J Virol*
21. Meertens, Carnec, Lecoin et al. (2012) "The TIM and TAM families of phosphatidylserine receptors mediate dengue virus entry" *Cell Host Microbe*
22. Strange, Jiyarom, Zarandi et al. (2019) "Axl promotes Zika virus entry and modulates the antiviral state of human sertoli cells" *mBio*
23. Meertens, Labeau, Dejarnac et al. (2017) "Axl mediates ZIKA virus entry in human glial cells and modulates innate immune responses" *Cell Rep*
24. Wang, Qiu, Hou et al. (2021) "AXL is a candidate receptor for SARS-CoV-2 that promotes infection of pulmonary and bronchial epithelial cells" *Cell Res*
25. Richard, Zhang, Park et al. (2015) "Virionassociated phosphatidylethanolamine promotes TIM1-mediated infection by Ebola, dengue, and West Nile viruses" *Proc Natl Acad Sci*
26. Corneillie, Lemmens, Montpellier et al. (2023) "The phosphatidylserine receptor TIM1 promotes infection of enveloped hepatitis E virus" *Cell Mol Life Sci*
27. Liu, Zhang, Qi et al. (2023) "Targeting MFGE8 secreted by cancer-associated fibroblasts blocks angiogenesis and metastasis in esophageal squamous cell carcinoma" *Proc Natl Acad Sci*
28. Park, Kehrl (2019) "An integrin/MFG-E8 shuttle loads HIV-1 viral-like particles onto follicular dendritic cells in mouse lymph node" *Elife*
29. Morizono, Chen (2014) "Role of phosphatidylserine receptors in enveloped virus infection" *J Virol*
30. Zhang, Wang, Nguyen et al. (2023) "Integrin α 5 β 1 contributes to cell fusion and inflammation mediated by SARS-CoV-2 spike via RGD-independent interaction" *Proc Natl Acad Sci*
31. Taylor, Couto, Scallan et al. (1997) "Lactadherin (formerly BA46), a membrane-associated glycoprotein expressed in human milk and breast carcinomas, promotes Arg-Gly-Asp (RGD)-dependent cell adhesion" *DNA Cell Biol*
32. Andersen, Graversen, Fedosov et al. (2000) "Functional analyses of two cellular binding domains of bovine lactadherin" *Biochemistry*
33. Zhang, Kim, Fox et al. (2018) "Mxra8 is a receptor for multiple arthritogenic alphaviruses" *Nature*
34. 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*
35. Shao, Novakovic, Head et al. (2008) "Crystal structure of lactadherin C2 domain at 1.7A resolution with mutational and computational analyses of its membrane-binding motif" *J Biol Chem*
36. Wang, Zhang, Tiwari et al. (2020) "Integrin αvβ5 internalizes Zika virus during neural stem cells infection and provides a promising target for antiviral therapy" *Cell Rep*
37. Jackson, Mould, Sheppard et al. (2002) "Integrin αvβ1 is a receptor for foot-and-mouth disease virus" *J Virol*
38. Yang, Arvin, Oliver (2016) "Role for the αV integrin subunit in varicella-zoster virus-mediated fusion and infection" *J Virol*
39. Gianni, Leoni, Chesnokova et al. (2012) "Αvβ3-integrin is a major sensor and activator of innate immunity to herpes simplex virus-1" *Proc Natl Acad Sci*
40. Larson, Brown, Ye et al. (2005) "Peptide antagonists that inhibit Sin Nombre virus and hantaan virus entry through the β3integrin receptor" *J Virol*
41. Stewart, Nemerow (2007) "Cell integrins: commonly used receptors for diverse viral pathogens" *Trends Microbiol*
42. Bhattacharyya, Zagórska, Lew et al. (2013) "Enveloped viruses disable innate immune responses in dendritic cells by direct activation of TAM receptors" *Cell Host Microbe*
43. Hunt, Kolokoltsov, Davey (2011) "The Tyro3 receptor kinase Axl enhances macropinocytosis of Zaire ebolavirus" *J Virol*
44. Brindley, Hunt, Kondratowicz et al. (2011) "Tyrosine kinase receptor Axl enhances entry of Zaire ebolavirus without direct interactions with the viral glycoprotein" *Virology (Auckl)*
45. Burt, Swanepoel, Shieh et al. (1997) "Immunohistochemical and in situ localization of Crimean-Congo hemorrhagic fever (CCHF) virus in human tissues and implications for CCHF pathogenesis" *Arch Pathol Lab Med*
46. Connolly-Andersen, Douagi, Kraus et al. (2009) "Crimean Congo hemorrhagic fever virus infects human monocyte-derived dendritic cells" *Virology (Auckl)*
47. Akıncı, Bodur, Leblebicioglu (2013) "Pathogenesis of Crimean-Congo hemorrhagic fever" *Vector Borne Zoonotic Dis*
48. Hanayama, Nagata (2005) "Impaired involution of mammary glands in the absence of milk fat globule EGF factor 8" *Proc Natl Acad Sci*
49. Erbay, Cevik, Onguru et al. (2008) "Breastfeeding in Crimean-Congo haemorrhagic fever" *Scand J Infect Dis*
50. Özüpak, Albayrak (2020) "Molecular detection of Crimean-Congo hemorrhagic fever virus (CCHFV)in tick samples but not in blood and milk samples of domestic ruminant species (cattle, sheep and goat) in northern Turkey" *Pol J Vet Sci*
51. Doench, Fusi, Sullender et al. (2016) "Optimized sgRNA design to maximize activity and minimize offtarget effects of CRISPR-Cas9" *Nat Biotechnol* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12724217&blobtype=pdf | # Involvement of a tick-borne orthomyxovirus matrix protein in vRNP nuclear export
Vaille Swenson, Jack Hemsath, Iris Yousaf, Carla Weisend, Michael Barry, Hideki Ebihara, Satoko Yamaoka
## Abstract
Thogotoviruses are arthropod-borne viruses belonging to the genus Thogotovirus within the family Orthomyxoviridae. Like other orthomyxoviruses, such as the influenza A viruses (IAV), thogotoviruses replicate in the nucleus. As a result, progeny viral ribonucleoprotein complexes (vRNPs) must be exported to the cytoplasm prior to virion assembly and budding at the plasma membrane. In IAV, this export depends on binding of the viral nuclear export protein (NEP) to the cellular exportin chromosomal maintenance 1 (CRM1). In contrast, the mechanism of vRNP nuclear export, including identification of a protein with NEP functionality, has not been characterized for any thogotoviruses. Here, we characterized vRNP nuclear export in Dhori virus (DHOV), a prototypic member of the Thogotovirus genus. DHOV replication and nuclear export of the viral nucleoprotein were inhibited by the canonical CRM1 inhibitor leptomycin B (LMB), suggesting that DHOV vRNP export also utilizes CRM1. Interestingly, LMB treatment led to nuclear retention of the DHOV matrix (M) protein in both infected and transfected cells. Using a mammalian two-hybrid system, we found that DHOV M interacts with CRM1 through a nuclear export sequence (NES) located between amino acids 111 and 128. Mutation of hydrophobic residues within this NES reduced M-CRM1 interaction, abolished the NES phenotype when fused to a fluorescent protein, and impaired rescue of recombinant DHOV by reverse genetics. Together, our results reveal that DHOV vRNP nuclear export is CRM1-dependent and mediated by the M protein rather than a dedicated NEP-like protein, providing the first mechanistic insight into vRNP egress in the genus Thogotovirus.
IMPORTANCE Dhori virus (DHOV) is a pathogenic tick-borne virus in the genusThogotovirus, in the family Orthomyxoviridae. Despite evidence of DHOV exposure in various mammals, including humans, its basic biology is not well understood. We investigated how DHOV's progeny genome and protein complexes-viral ribonucleopro tein complexes (vRNPs)-are transported out of the nucleus. Our findings show that DHOV, like the influenza viruses, uses the cellular protein chromosomal maintenance protein 1 (CRM1) for vRNP export. We found that chemically inhibiting CRM1 completely blocked DHOV vRNP export, preventing the production of progeny viruses from infected cells. Screening of all known DHOV proteins revealed that the matrix protein, which forms the virus' scaffold, interacted with CRM1, suggesting it may link CRM1 to the vRNPs. These results advance our understanding of DHOV replication and suggest that chemically inhibiting vRNP export could be a way to treat thogotovirus infections.
KEYWORDSDhori virus, Thogotovirus, nuclear export protein, nuclear export, CRM1, vRNP, orthomyxovirus T hogotoviruses are arthropod-borne viruses belonging to the genus Thogotovirus within the family Orthomyxoviridae (1). This genus is divided into two distinct clades based on sequence homology-the Thogoto thogotovirus (THOV) and the Dhori virus December 2025 Volume 99 Issue 12 10.
## RESULTS
## CRM1 inhibition diminishes DHOV replication
First, the effect of inhibiting CRM1 with LMB, a canonical CRM1 inhibitor (31)(32)(33), on DHOV replication was investigated in the permissive cell line human hepatoma Huh7 (34). LMB treatment caused a dose-dependent reduction in DHOV titer, achieving approximately a 2-log decrease (over 90% reduction) compared to the vehicle control at a dose of 50 nM (Fig. 1A andB). Importantly, the DHOV titer at a dose of 50 nM was equivalent to the titer from culture supernatant samples harvested immediately after 1 hour viral adsorption, suggesting near-complete inhibition of the release of progeny viruses. Significant reductions in DHOV titer were observed at even lower LMB doses, with a calculated IC 50 of 2.96 nM. LMB-treated cells maintained a cell viability of 70% or greater relative to the vehicle control across all tested doses (Fig. 1B). These results indicate that LMB inhibits DHOV replication, suggesting that CRM1 plays a critical role in the DHOV life cycle.
## LMB treatment causes DHOV vRNPs to be retained in the nucleus
Next, we examined whether DHOV vRNPs are retained in the nucleus upon CRM1 inhibition by LMB, as is the case in IAV (33). In untreated and vehicle-treated cells, perinuclear dot-like signals of DHOV NP were detected outside the nucleus starting at 6 hours post-infection (hpi), indicating that vRNP export had initiated by this time point (Fig. 1C, arrowheads; see Fig. S1A and B at https://doi.org/10.5281/zenodo.17715494). In contrast, perinuclear localization of NP was not observed in LMB-treated cells at either 6 or 8 hpi (Fig. 1C; see Fig. S1B at https://doi.org/10.5281/zenodo.17715494). Analysis of these images revealed a statistically significant increase in the colocalization of NP with both the nucleus and CRM1 in LMB-treated cells at 6 and 8 hpi, compared to vehicle and untreated controls (Fig. 1D andE). These results strongly suggest that LMB inhibition of CRM1 induces DHOV vRNP retention in the nucleus, thereby inhibiting viral replication.
## The cellular distribution of DHOV M changes with LMB treatment, resulting in its nuclear retention
Previous studies have shown that singly expressed IAV NEP can be detected in both the nucleus and cytoplasm due to its ability to shuttle between these two compartments (35). To identify DHOV protein(s) with this same shuttling potential, six DHOV proteins (PB2, PB1, PA, GP, NP, and M) were individually expressed with a FLAG tag placed at either the N-or C-terminus of each ORF, and their cellular distribution was assessed. Regardless of FLAG tag placement, NP and the RdRp subunits, PB2, PB1, and PA, were predominantly nuclear-localized, while GP was predominantly cytoplasmic (Fig. 2A through E andG,H). In contrast, M displayed a diffuse localization pattern throughout the nucleus and cytoplasm (Fig. 2F though H), suggesting that DHOV M may be capable of shuttling between these compartments. Notably, upon LMB treatment, M became retained in the nucleus along with CRM1, in contrast with the diffuse localization observed in cells treated with the vehicle control (Fig. 3A). These findings suggest that CRM1 is necessary for M to shuttle from the nucleus to the cytoplasm.
The cellular distribution of DHOV M, with or without CRM1 inhibition by LMB, was further examined in DHOV-infected cells. In untreated and vehicle-treated cells, DHOV M localization mirrored the perinuclear signal of the DHOV NP during vRNP export beginning at 6 hpi (Fig. 3B, arrowheads; see Fig. S2A and B at https://doi.org/10.5281/ zenodo.17715494). In contrast, LMB treatment caused noticeable retention of both M and NP within the nucleus at both 6 and 8 hpi (Fig. 3B, arrowheads; see Fig. S2B at https://doi.org/10.5281/zenodo.17715494). Analysis of these images showed a statisti cally significant increase in the colocalization of M with the nucleus in LMB-treated cells, compared to both vehicle and untreated cells beginning at 6 hpi (Fig. 3C). These results strongly suggest that DHOV M is involved in the nuclear export of vRNPs.
## DHOV M interacts with CRM1 in a mammalian two-hybrid system
Given that the interaction between CRM1 and IAV NEP is critical for IAV vRNP export (21,36), the interaction between DHOV M and CRM1 was assessed using the Promega CheckMate System-a luciferase reporter-based mammalian two-hybrid system that quantitatively evaluates protein:protein interaction in mammalian cells via fusion to either the GAL4 DNA-binding domain or VP16 transcriptional activation domain (Fig. 4A) (37). Our results indicated that, similar to the positive control MyoD and ID (Fig. 4B andC, lanes 2-4), co-transfection of CRM1 and IAV NEP induced significant luciferase reporter activity compared to corresponding vector controls (Fig. 4B and C, lanes 5-7), consistent with previous findings (23). Co-transfection of DHOV M and CRM1 induced a statistically significant increase in firefly luciferase expression over corresponding vector controls, suggesting that these two proteins interact in vitro (Fig. 4B andC, lanes 5, 10-11). Importantly, this interaction was seen regardless of fusion protein orientation, with the opposite orientation (pACT/GAL4 DHOV M) also inducing a statistically significant Full-Length Text increase in firefly luciferase expression over vector controls (Fig. 4D andE, lanes 5-7), suggesting that DHOV M interacts with CRM1. Due to the presence of a CRM1-depend ent NES within IAV NP (36), DHOV NP was also tested for its potential interaction with CRM1. However, co-transfection of DHOV NP and CRM1 did not cause a statistically significant increase in firefly luciferase expression over both vector controls (Fig. 4B and C, lanes 5, 8-9), suggesting a lack of interaction between these two proteins.
## Serial truncations of DHOV M identify a region with nuclear export capability
The nuclear export capability of DHOV M was further examined by fusing M with the fluorescent protein mCherry (Fig. 5A) (36,38,39). While mCherry alone predominantly localized to the nucleus, the mCherry signal clearly shifted to the cytoplasm when fused to full-length M (Fig. 5B), indicating that DHOV M can mediate the export of this fusion protein from the nucleus to the cytoplasm. To identify the region within M responsible for nuclear export, we constructed a series of truncated M proteins fused to mCherry (Fig. 5A) and assessed their localization within transfected HEK-293 cells.
The M ORF was initially divided into three fragments, spanning amino acids 1-136, 69-204, and 137-271 (Fig. 5A). When these truncation fusion proteins were visualized, only fragments 69-204 localized to the cytoplasm, whereas 137-271 remained in the nucleus.
In contrast, fragments 1-136 displayed a diffuse distribution across both the nucleus and cytoplasm (Fig. 5B). Subdivision of amino acids 69-204 identified that fragments 69-136 predominantly localized to the cytoplasm, while fragments 103-170 and 137-204 showed minimal cytoplasmic localization. Further subdivision of amino acids 69-136 revealed that only 103-136 exhibited a clear cytoplasmic localization, like the pattern observed with full-length M, while 69-102 and 86-119 remained in the nucleus. Finally, division of 103-136 demonstrated that the 18 amino acid stretch spanning residues 111-128 retained the WT M localization pattern, whereas fragments 103-119 and 120-136 showed a distribution like WT mCherry (Fig. 5A andB). These results suggest that amino acids 111-128 within the DHOV M ORF contain a signal required for mediating nuclear export.
## DHOV M contains a putative LMB sensitive NES within amino acids 111-128
Given that full-length DHOV M was sensitive to LMB treatment in both transfected and infected cells (Fig. 3A through C; see Fig. S2B at https://doi.org/10.5281/ zenodo.17715494), we next investigated whether residues 111-128 alone were similarly sensitive to CRM1 inhibition. As previously demonstrated, both full-length M and the 111-128 fragment fused to mCherry localized to the cytoplasm of transfected cells in untreated and vehicle-treated controls (Fig. 6). However, treatment with LMB led to nuclear retention of both WT M and the 111-128 fragment, indicating their dependence on CRM1-mediated nuclear export (Fig. 6). These results suggest that amino acids 111-128 within the DHOV M ORF contain a CRM1-dependent NES.
The nuclear export of proteinaceous cargo through CRM1 is facilitated by the interaction between the NES within the cargo and the NES-binding groove of CRM1 (27). NES sequences can be identified by specific patterns of hydrophobic residues, generally following the consensus sequence Φ1-(X) 2-3 -Φ2-(X) 2-3 -Φ3-X-Φ4, where "X" denotes any amino acid and "Φ" denotes any hydrophobic amino acid (Leu, Ile, Val, Met, or Phe) (27). Although seven hydrophobic amino acids are present within DHOV M 111-128 (Fig. 7A), their distribution and spacing made it challenging to identify which residues could be essential for NES function. However, this region was highlighted as the highest-scoring NES candidate by the prediction tool LocNES (Fig. 7B) and was found to be highly conserved among thogotoviruses, particularly within the DHOV clade (Fig. 7C). Collec tively, these findings suggest that the hydrophobic residues within DHOV M 111-128 (hereafter referred to as NES1) may form a highly evolutionarily conserved NES (40).
## Mutation of the hydrophobic amino acids within DHOV M NES1 disrupts interaction with CRM1 and reverses NES phenotype
To investigate the role of DHOV M NES1 in mediating interaction with CRM1, each of the seven hydrophobic residues, V115, L116, M119, L120, I121, L122, and I127, were individu ally mutated to either another hydrophobic amino acid (alanine) or a hydrophilic amino acid (serine). The resulting DHOV M mutants were then analyzed using the Promega CheckMate System to assess their ability to induce firefly luciferase activity, suggesting CRM1 binding. Co-transfection of all NES1 mutants with CRM1 resulted in an increase in firefly luciferase expression over vector controls (Fig. 8A andB; see Fig. S3 at https:// doi.org/10.5281/zenodo.17715494). Among all mutants, only substitutions at I121 and L122 consistently resulted in a statistically significant decrease in firefly luciferase expression, regardless of whether alanine or serine was used (Fig. 8A andB; see Fig. S3 at https://doi.org/10.5281/zenodo.17715494). When these data were normalized to WT M, the signal from the I121 mutants was reduced by more than 10% relative to WT levels, while the L122 mutants showed an even greater reduction, with signals below 70% of those seen in WT M (Fig. 8D).
To further investigate the contribution of the hydrophobic residues with DHOV M NES1 to surrogate CRM1 interaction within the Promega CheckMate System, we generated double mutants targeting the residues whose mutations caused the greatest reduction in firefly luciferase activity (121/2A and 121/2S). Additionally, to evaluate the induction, reducing the activity to background levels regardless of the substituted amino acid (Fig. 8C andD; see Fig. S3 at https://doi.org/10.5281/zenodo.17715494). Together, these results indicate that the hydrophobic residues within DHOV M NES1-particularly I121 and L122-may be key contributors to CRM1 interaction. Building on our earlier observation that fusing DHOV M NES1 to mCherry reproduces an NES phenotype, we next evaluated how mutations within NES1 impact the subcellular distribution of mCherry-DHOV M NES1 fusion proteins. Double mutation of residues I121 and L122, as well as pan mutation of all hydrophobic residues to alanine, glycine, serine (All S), or arginine, was tested in our mCherry fusion protein system. Remarkably, both the double and pan mutants resulted in a complete reversal of the NES phenotype, resulting in fusion proteins that displayed a diffuse distribution across the nucleus and cytoplasm, like that of WT mCherry (Fig. 8E). Collectively, these results suggest a critical role of NES1 in mediating DHOV M interaction with CRM1 and NES functionality.
## Mutations in NES1 attenuate or abolish the rescue of DHOV from a reverse genetics system
Finally, the importance of the hydrophobic residues within NES1 was evaluated within the context of the viral life cycle using infectious DHOV. A human RNA polymerase I-driven reverse genetics system for DHOV was developed (Fig. 9A) and employed to rescue recombinant DHOV (rDHOV) as well as rDHOV mutants carrying either double (121/2A, 121/2S) or pan (All A, All G, All R) mutations within the hydrophobic residues of DHOV M NES1. Infectious rDHOV was successfully rescued only when all 10 plasmids were co-transfected in HEK-293T cells, but not when segment 1 (S1) or PB2 was absent (Fig. 9B). When the growth kinetics of WT DHOV was compared to rDHOV, no statisti cally significant differences in viral growth were observed at any time point (Fig. 9C), validating the biological functionality of rDHOV.
Although the double mutants 121/2A and 121/2S were successfully rescued, their rescue titers were consistently lower than that of rDHOV, with 121/2A yielding a half-log, and 121/2 S a nearly three-log decrease in rescue efficiency (Fig. 9B). In contrast, rDHOV carrying pan mutations could not be rescued, irrespective of the amino acid substitu tions introduced (Fig. 9B). To further quantify the attenuation of 121/2A and 121/2S, we measured both the plaque diameter and area of rDHOV on Vero-E6 cells. Both the plaque diameter and area of rDHOV 121/2A and 121/2S were significantly decreased compared to rDHOV, with 121/2S producing plaques only one-fourth the size of those formed by rDHOV (Fig. 9D andE). Collectively, these results suggest that mutating hydrophobic residues within DHOV M NES1-particularly I121 and L122-attenuates the resulting rDHOV, supporting a role for DHOV M NES1 in the viral life cycle through with CRM1.
## DISCUSSION
In this study, we present a CRM1-dependent mechanism for the nuclear export of DHOV vRNPs, mediated by the M protein. Since the identification of NEP in IAV (18,21), viral proteins with NEP functionality have been recognized across various orthomyxoviruses, including members of Betainfluenzavirus, Gammainfluenzavirus, and Deltainfluenzavirus genera (IBV, ICV, and IDV), as well as the Isavirus, Mykissvirus, and possibly Sardinovirus genera (41)(42)(43)(44). Our study is the first to characterize the molecular mechanism of vRNP export in the genus Thogotovirus, identifying DHOV M as an analog of IAV NEP. This finding proposes a novel concept that DHOV M has dual functionality, mediating both viral budding (45) and vRNP nuclear export within a single protein. This condensation of protein functionality could explain how members of the Thogotovirus genus compensate for the loss of two segments of coding capacity relative to IAV/IBV.
The nuclear export of newly formed vRNPs is a crucial step in the viral life cycle of the family Orthomyxoviridae (17). During the formation of IAV vRNP export complexes, NEP interacts with CRM1 through its N-terminal NES, while the C-terminus of NEP binds to M1 bound to the vRNP (18, 21, 23-26, 46, 47). The interaction between NEP and M1 is thought to require NEP interaction with the RdRp, which strengthens the binding of M1 to the vRNP (25,47). Interestingly, no direct interaction between IAV M1 and CRM1 has been reported (20); however, our findings indicate that DHOV M interacts with CRM1 (Fig. 4). Thus, DHOV M may combine the functionality of both NEP and M1 of IAV by bridging the gap between the vRNP complex and CRM1. Notably, we observed colocalization between M and NP during vRNP export in infected cells and retention of both DHOV NP and M within the nucleus upon LMB treatment (Fig. 3), suggesting that M plays a critical role in vRNP export through its interaction with CRM1. Successive truncations of the DHOV M ORF fused to mCherry identified a region spanning amino acids 111-128 that contains a cluster of seven hydrophobic residues and was sensitive to LMB treatment (Fig. 5 to 7). While most individual alanine and serine substitutions at each of these residues impaired the interaction between M and CRM1 in a mammalian two-hybrid system, mutations at residues 121 and 122 had the most pronounced negative impact (Fig. 8A through D). The negative impact of the NES1 mutation was further confirmed by pan mutation of all seven hydrophobic residues to alanine, glycine, or arginine, all of which abolished interaction with CRM1 in a mamma lian two-hybrid system and reversed the NES phenotype in fusion to mCherry (Fig. 8C through E). rDHOV carrying pan mutations of hydrophobic residues in NES1 could not be rescued using our reverse genetics system (Fig. 9B), whereas rDHOV 121/2A and 121/2S mutants were successfully rescued, but exhibited attenuation relative to WT rDHOV, with 121/2S showing the greatest level of attenuation.
The successful rescue of rDHOV 121/2A and 121/2S may be attributed to the presence of nearby hydrophobic residues, such as M119 and L120, which could compensate for the mutations of I121 and L122. Furthermore, the similarity between the side chains of alanine and the original methionine and leucine at residues 119 and 120 may enable DHOV M 121/2A to retain greater functionality, whereas substitution with the hydropho bic serine in 121/2S diminishes this activity. The sensitivity of amino acid selection in NES mutagenesis was demonstrated by Huang et al., who showed that IAV NEP NES2 point mutations to hydrophobic residues (F35A or L38A) still allowed mutant IAV to be rescued, whereas substitution with a hydrophilic amino acid (F35K or L38R) abolished the rescue (23). Further studies are required to identify additional potential NES within DHOV M and elucidate the full structure of the DHOV vRNP export complex.
Interestingly, although we demonstrated the presence of an NES between amino acids 111 and 128 within the M ORF, the M fragments 1-136 and 103-170 fused to mCherry showed some nuclear localization (Fig. 5B). One possible explanation for this discrepancy is the presence of an NLS within the N-terminus of DHOV M, as has been reported for IAV M1 (48). Screening of the DHOV M ORF for canonical NLS did not identify any obvious NLS sequence (data not shown). However, the nuclear localization of DHOV M upon LMB treatment (Fig. 3A through C) and homology to IAV M1 suggests that DHOV M may contain NLS. Aberrant folding of truncated DHOV M fragments 1-136 and 103-170 could partially sequester NES1, making it unavailable for interaction with CRM1 and thus diminishing its cytoplasmic localization. Different levels of cytoplasmic localization from the same region of DHOV M (e.g., fragments 103-170 vs 103-136, Fig. 5B) suggest that fusion protein localization is partially dependent on surrounding sequences.
In IAV, vRNP export is highly sensitive to LMB treatment, with doses as little as 5 nM resulting in complete nuclear retention of IAV H1N1 (A/WSN/33) vRNPs in A549 cells (33,49). This inhibitory concentration of LMB closely matches the LMB IC 50 of 2.96 nM that we determined for DHOV in Huh7 cells (Fig. 1B). The sensitivity of DHOV to LMB suggests that CRM1 inhibition may be a viable therapeutic strategy for thogotovirus infections; however, the use of LMB in vivo is limited by its cytotoxicity (50). To address this limitation, derivative CRM1 inhibitors known as selective inhibitors of nuclear export (SINE) were developed to minimize cytotoxicity (51). One such compound, verdinexor, was shown to potently and selectively inhibit vRNP export in A549 cells infected with various IAV or IBV strains, with IC 50 values ranging from 10 nM to 420 nM (52). Impor tantly, prophylactic and therapeutic treatment with verdinexor was also shown to reduce viral load and pathology of IAV infection while improving survival in a mouse model (52). These insights suggest that CRM1 inhibition by SINE compounds may be a potential therapy for DHOV and other thogotovirus infections, given the shared sensitivity of both DHOV and IAV to LMB treatment.
Overall, our study represents the first characterization of the nuclear export of progeny vRNPs of DHOV, a member of the Thogotovirus genus. We demonstrate that DHOV utilizes CRM1 as the primary exportin for vRNP egress, with M serving as an NEP by interacting with CRM1 via an NES. These findings enhance our understanding of thogotovirus biology and suggest that CRM1 inhibition may offer a potential therapeutic approach for treating thogotovirus infections.
## MATERIALS AND METHODS
## Alignment
To generate an alignment of multiple thogotovirus segment 6 ORFs, the amino acid sequence of the DHOV (strain I-611313; GenBank Accession: PQ469003-PQ469008), BRBV (strain Original; GenBank Accession: KP657750.3), OZV (strain EH8; GenBank Accession: NC_040734.1), Thailand tick thogotovirus (strain THOV/Boophilus sp./ Thailand; GenBank Accession: NC_078597.1), Sinu virus (strain CoB 38d; GenBank Accession: NC_078772.1), THOV (strain IIA; GenBank Accession: AF527529.1), Upolu virus (GenBank Accession: NC_078650.1), and Jos virus (GenBank Accession: HM627172.1) M proteins were all aligned using CLUSTAL W (53).
## Antibodies
A custom affinity purified rabbit polyclonal antibody was produced by Biomatik against peptide Cys-163 EDEQRDLWLEEVTRQLNTLTPVIRG 187 from BRBV M and confirmed to cross-react to DHOV M prior to being used as a primary antibody. Rabbit monoclo nal antibody against CRM1 was purchased from Cell Signaling Technology (46249S). Polyclonal mouse antibodies against DHOV NP were generated in-house. Briefly, BALB/c mice were immunized intramuscularly into the quadriceps with 25 µg of endotoxin-free pCAGGS-DHOV-NP combined with 25 µg of endotoxin-free plasmid expressing murine granulocyte-macrophage colony-stimulating factor (GMCSF) under the CMV promoter (pG-CMVi-GMCSF), as previously described (54)(55)(56). Mice were boosted with the same protocol twice at 3 week intervals. Four weeks after the final booster immunization, mice were anesthetized and exsanguinated, and whole blood samples were centrifuged at 13,000 × g for 10 minutes to separate out the serum. Whole serum was then uti lized in place of the purified antibody for all subsequent immunofluorescence experi ments. Mouse monoclonal antibody against the FLAG M2 epitope was purchased from Sigma-Aldrich (F3165). Goat anti-rabbit AlexaFluor 488 (ThermoFisher Scientific, A11008), goat anti-mouse AlexaFluor 594 (Invitrogen, A32740), or goat anti-mouse AlexaFluor 488 (Invitrogen, A11029) were used as secondary antibodies.
## Biosafety
All experiments using infectious DHOV were performed in biosafety level 2+ (BSL-2+) or biosafety level 3 (BSL-3) facilities at Mayo Clinic in accordance with approval and guidelines from the Mayo Clinic Institutional Biosafety Committee (IBC). Sample inactivation and removal from said facilities were performed in accordance with standard operating protocols approved by the IBC.
## Cell viability assay
Huh7 cells (1 × 10 4 cells/well) were seeded in 96-well plates 1 day before compound treatment. Briefly, existing medium on cells was replaced with 2% FBS 1% P/S DMEM supplemented with indicated doses of LMB or vehicle control and incubated at 37°C for 1 hour. Following incubation, the compound-containing media was removed and replaced with 2% FBS 1% P/S DMEM before being incubated at 37°C for 24 hours. Cell viability was then assessed using the CellTiter Glo Luminescent Cell Viability Assay (Promega) according to the manufacturer's instructions.
## Cells and virus
Huh7 (a kind gift from Yoshiharu Matsuura, Osaka University), HEK-293 (ATCC, CRL-1573), HEK-293T/17 (ATCC, CRL-11268), and Vero-E6 cells (ATCC, CRL-1586) were maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS; Thermo-Fisher Scientific) and 1% penicillin-streptomycin (P/S; Millipore Sigma) at 37°C in 5% CO 2 . An isolate of Dhori virus (DHOV) strain I-611313 was kindly provided by Brandy J. Russell of the Centers for Disease Control and Prevention (CDC). DHOV working stocks were prepared in Huh7 cells by passaging the original virus once and frozen at -80°C until use. DHOV stocks for sequencing were prepared in BHK-21 cells by passaging the original virus once.
The DHOV titer was determined by focus-forming assay. Briefly, Vero-E6 cells were seeded into 96-well plates (2 × 10 4 cells/well) 1 day before infection. Cells were infected with 10-fold serial dilutions of the virus. After 1 hour absorption with tilting every 15 minutes, cells were overlayed with a mixture of 1.2% carboxymethylcellulose and Temin's modified eagle medium (MEM). Two days post-infection, cells were fixed with 10% neutral buffered formalin before proceeding to immunofluorescence.
## Chemical compounds
Leptomycin B (L2913) was purchased from Millipore Sigma and stored at -20°C until use. The corresponding vehicle control was prepared by mixing molecular-grade methanol and water at a 7:3 ratio and stored at -20°C until use. Cycloheximide (01810) was purchased from Millipore Sigma, dissolved in sterile water, filter-sterilized, and stored -20°C until use.
## Colocalization analysis
Colocalization between indicated channels was quantified using the Just Another Colocalization plugin v2.1.4 (JaCoP) (57) in ImageJ (v1.54g) (58). If needed, thresholds were manually adjusted in JaCoP prior to quantification to appropriately cover the fluorescent signal.
## DHOV growth kinetics
Huh7 cells (3 × 10 6 cells/well) were seeded into 6-well plates 1 day before infection. Cells were infected with DHOV at a multiplicity of infection (MOI) of 0.1. After 1 hour adsorption with tilting every 15 minutes, cells were washed three times with serum-free DMEM, and 3.0 mL of DMEM supplemented with 2% FBS was added to the cells. Immediately after adding the medium, 0.5 mL of supernatants was harvested as a sample of 0 dpi. Up to 3 dpi, 0.5 mL of supernatants was harvested at the indicated time points and replaced with an equal volume of fresh medium supplemented with 2% FBS. All supernatant samples were stored at -80°C until use for titration.
## CRM1, DHOV NP, and DHOV M localization in infected cells
Huh7 cells (0.5 × 10 5 cells/well) were seeded in 24-well plates 1 day before infection and then infected with DHOV at an MOI of 100. Cells and virus were incubated for 1 hour at 37°C, with rocking every 15 minutes. The inoculum was then removed, and cells were washed three times with serum-free DMEM supplemented with 1% P/S. If undergoing compound treatment, cells were then treated with the indicated doses of compound in 2% FBS, 1% P/S DMEM for an additional hour at 37°C. Infected or compound-treated media was replaced with 2% FBS 1% P/S DMEM, and cells were incubated for a total of 2, 5, or 7 hours at 37°C following initial infection before proceeding to immunofluorescence. Rabbit anti-CRM1, mouse anti-DHOV NP, and rabbit anti-BRBV M were used as primary antibodies at a concentration of 1:500. Goat antirabbit AlexaFluor 488 and goat anti-mouse AlexaFluor 594 were used as secondary antibodies. Fluorescence images were obtained using a confocal microscope (ZEISS LSM 980 with Airyscan2) under 10 × magnification.
## Evaluation of LMB impact on DHOV titer
Huh7 cells (0.5 × 10 5 cells/well) were seeded in 24-well plates 1 day before infection and then infected with DHOV at an MOI of 0.1. Cells and viruses were incubated for 1 hour at 37°C, with rocking every 15 minutes. The inoculum was then removed, and cells were washed three times with serum-free DMEM supplemented with 1% P/S. Subsequently, 0.5 mL of 2% FBS 1% P/S DMEM with the indicated concentration of LMB or vehicle control was added. Following a 1 hour incubation at 37°C, compound-containing media was replaced with 2% FBS 1% P/S DMEM, and cells were incubated for 24 hours at 37°C. Supernatants were then harvested and stored at -80°C until use for titration.
## Identification of potential nuclear export sequences
Potential NES were identified by uploading the DHOV M ORF into LocNES (40).
## Immunofluorescence assays
Cells were fixed in 10% neutral buffered formalin, permeabilized using a 1:1 ratio of methanol:acetone for 10 minutes at room temperature and blocked in 1% bovine serum albumin (BSA) in PBS for at least 1 hour at room temperature. Primary antibodies were added at a concentration of 1:1,000 (unless otherwise noted) in PBS containing 1% BSA and incubated overnight at 4°C with rocking. Secondary antibodies were added at concentrations of 1:10,000 in PBS and incubated at room temperature for 1 hour with rocking. Nuclei were stained using Hoechst 33342 (ThermoFisher Scientific) according to the manufacturer's instructions.
## Mammalian two-hybrid assay
HEK-293 (0.8 × 10 5 cells/well) were seeded into 24-well plates 1 day before transfec tion. Cells were then co-transfected with 100 ng of the indicated pBIND and pACT fusion constructs and 100 ng pG5luc per well using TransIT-LTI (Mirus Bio) according to the manufacturer's instructions. After 96 hours of incubation post-transfection, cells were harvested and lysed with passive lysis buffer (Promega). Luciferase assays were performed using a dual-luciferase reporter assay system (Promega) according to the manufacturer's instructions. Renilla luciferase activity was used to normalize data to transfection efficiency using the following equation: (Firefly/Renilla). If needed, data were further normalized to corresponding vector controls: (sample/vector).
## Plasmid generation
Human RNA polymerase I (hPolI)-driven DHOV reverse genetics plasmids were generated for each segment. Briefly, DHOV segments 1-6 were amplified by RT-PCR from viral genomic RNA using segment-specific RT primers. cDNA was amplified using primers specific to the 3′ and 5′ NCR of each segment and cloned into a pPolI vector contain ing the hPolI promoter and terminator sequences as previously described (59,60). To generate expression vectors of each DHOV protein, the PB2, PB1, PA, GP, NP, and M ORFs were amplified, and the cDNA was cloned into a pCAGGS vector that contains the cytomegalovirus enhancer fused to the chicken beta-actin promoter, as previously described (60,61). FLAG-tagged expression vectors of each DHOV protein were created by amplifying each DHOV ORF with primers that inserted a FLAG tag epitope at either the N-terminus directly after the start codon or the C-terminus directly before the stop codon. These FLAG-tagged ORFs were then cloned back into the pCAGGS expression vector.
Plasmids pACT-Vector, pBIND-Vector, pACT-MyoD, pBIND-ID, and pG5luc were obtained from the CheckMate Mammalian Two-Hybrid System from Promega (E2440) (37). The human CRM1 ORF was amplified from pHAGE-XPO1-a gift from Gordon Mills & Kenneth Scott (Addgene plasmid #116804; http://n2t.net/addgene:116804; RRID:Addg ene_116804)-and then inserted into the pACT and pBIND vectors (62). The IAV NEP ORF was amplified in fragments to remove the NS1 coding region. Briefly, NEP nucleotides 24-221 were amplified from an expression vector of the IAV (A/WSN/1933(H1N1)) NS1 ORF (NS1 nucleotides 496-693) using a primer that also contained NEP nucleotides 1-23. NEP nucleotides 222-366 were ordered as two overlapping oligos. Overlapping extension PCR was then utilized to combine the three fragments into the NEP ORF before cDNA was inserted into the pACT and pBIND vectors. To generate pACT and pBIND constructs of DHOV NP and M, the NP and M ORFs were amplified from the untagged pCAGGS constructs and inserted into the vector backbones. Successive DHOV M fusions to mCherry were generated by amplifying the indicated section of DHOV M from pCAGGS-DHOV-M and then performing overlapping extension PCR to fuse the fragment to the C-terminus of mCherry directly before the stop codon. These fusion protein constructs were then inserted back into the pCAGGS vector. Point mutations in the DHOV M NES sequence were generated by PCR mutagenesis on the DHOV M ORF prior to it being reinserted into the pCAGGS, pACT, pBIND, and pPolI vectors.
## Plaque analysis
Images of 30 separate plaques from titration plates of the indicated viruses were taken using a confocal microscope (ZEISS LSM 980 with Airyscan2) under 10× magnification. Individual images were uploaded into ImageJ (v1.54g) (58), and a line was drawn border to border over the widest section of each individual plaque to calculate plaque diameter in pixels. The area of each plaque in pixels was determined by manually tracing its border using the freehand tool in ImageJ. Pixel measurements were subsequently converted to micrometers (µm) using the width of the scale bar included in the image as a reference.
## Rescue of rDHOV
HEK-293T cells (2 × 10 5 cells/well) were seeded into 12-well plates 1 day before transfec tion. Cells were co-transfected with 50 ng of each helper plasmid (pCAGGS-DHOV-PB2, pCAGGS-DHOV-PB1, pCAGGS-DHOV-PA, and pCAGGS-DHOV-NP) and 125 ng of each viral genome plasmid encoding segments 1-5 (pPolI-DHOV-S1, pPolI-DHOV-S2, pPolI-DHOV-S3, pPolI-DHOV-S4, and pPolI-DHOV-S5) and one version of segment 6 (pPolI-DHOV-S6-WT, -All A, -All G, -All R, -121/2 A or -121/2 S) using TransIT-LTI (Mirus Bio) according to the manufacturer's instructions. For the S1(-) and PB2(-) controls, pPolI-DHOV-S1 or pCAGGS-DHOV-PB2 were substituted for an equivalent amount of the empty vector control (pPolI-Empty or pCAGGS-Empty, respectively). After 24 hours of incubation, the cellular supernatant was removed and replaced with 1 mL of 2% FBS 1% P/S DMEM. Forty-eight hours post media change, cellular supernatant was harvested and stored at -80°C until use for titration.
## Single protein localization in transfected cells
All transfections were performed using TransIT-LTI (Mirus Bio) according to the manu facturer's instructions. To assess DHOV protein localization, Huh7 cells were seeded in 12-well plates 1 day before transfection with 1 ug of the indicated pCAGGS expression vector. After 24 hours, cells were directly fixed or treated with 100 nM LMB, vehicle control, or untreated media for 1 hour and then fixed. Protein localization was assessed using mouse anti-FLAG M2 or rabbit anti-BRBV M (1:500) as primary antibodies and goat anti-rabbit AlexaFluor 488 or goat anti-mouse AlexaFluor 488 as secondary antibodies. Fluorescence images were obtained using a ZOE Fluorescent Cell Imager (Bio-Rad) under 20 × magnification or a confocal microscope (ZEISS LSM 980 with Airyscan2) under 10 × magnification.
To assess mCherry fusion protein localization, HEK-293 cells (1 × 10 5 cells/well) were seeded into 12-well plates 1 day before transfection with 0.5 µg of the indicated pCAGGS expression vector. To evaluate localization following compound treatment, cells were prepared as described above and 24 hours post-transfection were treated for 2 hours with 2% FBS 1% P/S DMEM containing 100 µg/µL cycloheximide and 100 nM LMB, vehicle control, or untreated media before fixation. Fluorescence images were obtained using a Nikon Eclipse Ts2 microscope with an Excelis MPX-6 camera under 20 × magnification.
## Statistical analysis
Experiments were conducted with at least three independent replicates. Statistical analysis was performed using the t-test and one-way ANOVA with GraphPad Prism 10 version 10.3.0.
## Viral genome sequencing
To sequence the complete genome of DHOV-I-611313, viral RNA was extracted from DHOV stocks from infected BHK-21 cells and purified using the QIAamp Viral RNA Mini kit (QIAGEN). Full-length viral genome sequences were determined using 3′ and 5′ RACE and Sanger sequencing.
## References
1. Bendl, Fuchs, Kochs (2023) "Bourbon virus, a newly discovered zoonotic thogotovirus" *J Gen Virol*
2. Fuchs, Lamkiewicz, Kolesnikova et al. (2022) "Comparative study of ten thogotovirus isolates and their distinct in vivo characteristics" *J Virol*
3. Moore, Causey, Carey et al. (1975) "Arthropod-borne viral infections of man in Nigeria, 1964-1970" *Ann Trop Med Parasitol*
4. Butenko, Leshchinskaia, Semashko et al. (1987) "Dhori virus--a causative agent of human disease. 5 cases of laboratory infection" *Vopr Virusol*
5. Kosoy, Lambert, Hawkinson et al. (2014) "Novel thogotovirus associated with febrile illness and death" *Emerg Infect Dis*
6. (2023) "Japan: National Institute of Infectious Diseases"
7. Bricker, Shafiuddin, Gounder et al. (2019) "Therapeutic efficacy of favipiravir against Bourbon virus in mice" *PLoS Pathog*
8. Jackson, Gidlewski, Root et al. (2012) "Bourbon virus in wild and domestic animals" *Emerg Infect Dis*
9. Lledó, Giménez-Pardo, Gegúndez (2020) "Epidemiological study of thogoto and dhori virus infection in people bitten by ticks, and in sheep, in an area of Northern Spain" *Int J Environ Res Public Health*
10. Tran, Shimoda, Ishijima et al. (2022) "Zoonotic infection with Oz virus, a novel Thogotovirus" *Emerg Infect Dis*
11. Fuller, Freedman-Faulstich, Barnes (1987) "Complete nucleotide sequence of the tick-borne, orthomyxo-like Dhori/Indian/1313/61 virus nucleoprotein gene" *Virology*
12. Freedman-Faulstich, Fuller, Lin et al. (1990) "Evolutionary relatedness of the predicted gene product of RNA segment 2 of the tickborne Dhori virus and the PB1 polymerase gene of influenza viruses" *Virology*
13. Clay, Fuller (1992) "Nucleotide sequence of the tick-borne orthomyxo-like Dhori/India/1313/61 virus membrane protein gene" *J Gen Virol*
14. Clark, Karsch-Mizrachi, Lipman et al. (2016) *GenBank. Nucleic Acids Res*
15. Hagmaier, Jennings, Buse et al. (2003) "Novel gene product of Thogoto virus segment 6 codes for an interferon antagonist" *J Virol*
16. Dou, Revol, Östbye et al. (2018) "Influenza A virus cell entry, replication, virion assembly and movement" *Front Immunol*
17. O'neill, Talon, Palese (1998) "The influenza virus NEP (NS2 protein) mediates the nuclear export of viral ribonucleoproteins" *EMBO J*
18. Carter, Iqbal (2024) "The influenza A virus replication cycle: a comprehensive review" *Viruses*
19. Eisfeld, Neumann, Kawaoka (2015) "At the centre: influenza A virus ribonucleoproteins" *Nat Rev Microbiol*
20. Neumann, Hughes, Kawaoka (2000) "Influenza A virus NS2 protein mediates vRNP nuclear export through NES-independent interaction with hCRM1" *EMBO J*
21. Sakaguchi, Hirayama, Hiraki et al. (2003) "Nuclear export of influenza viral ribonucleoprotein is temperature-dependently inhibited by dissociation of viral matrix protein" *Virology*
22. Huang, Chen, Chen et al. (2013) "A second CRM1-dependent nuclear export signal in the influenza A virus NS2 protein contributes to the nuclear export of viral ribonucleoproteins" *J Virol*
23. Akarsu, Burmeister, Petosa et al. (2003) "Crystal structure of the M1 protein-binding domain of the influenza A virus nuclear export protein (NEP/NS2)" *EMBO J*
24. Brunotte, Flies, Bolte et al. (2014) "The nuclear export protein of H5N1 influenza A viruses recruits Matrix 1 (M1) protein to the viral ribonucleoprotein to mediate nuclear export" *J Biol Chem*
25. (2025) *Full-Length Text Journal of Virology*
26. Shimizu, Takizawa, Watanabe et al. (2011) "Crucial role of the influenza virus NS2 (NEP) C-terminal domain in M1 binding and nuclear export of vRNP" *FEBS Lett*
27. Lee, Pei, Baumhardt et al. (2019) "Structural prerequisites for CRM1-dependent nuclear export signaling peptides: accessibility, adapting conformation, and the stability at the binding site" *Sci Rep*
28. Fung, Fu, Chook (2017) "Nuclear export receptor CRM1 recognizes diverse conformations in nuclear export signals"
29. Monecke, Güttler, Neumann et al. (2009) "Crystal structure of the nuclear export receptor CRM1 in complex with Snurportin1 and RanGTP" *Science*
30. Yu, Liu, Cao et al. (2012) "Identification and characterization of three novel nuclear export signals in the influenza A virus nucleoprotein" *J Virol*
31. Kudo, Matsumori, Taoka et al. (1999) "Leptomycin B inactivates CRM1/exportin 1 by covalent modification at a cysteine residue in the central conserved region" *Proc Natl Acad Sci*
32. Rahmani, Dean (2017) "Leptomycin B alters the subcellular distribution of CRM1 (Exportin 1)" *Biochem Biophys Res Commun*
33. Watanabe, Takizawa, Katoh et al. (2001) "Inhibition of nuclear export of ribonucleoprotein complexes of influenza virus by leptomycin B" *Virus Res*
34. Fuchs, Straub, Seidl et al. (2019) "Essential role of interferon response in containing human pathogenic bourbon virus" *Emerg Infect Dis*
35. Gao, Wang, Cao et al. (2014) "Characteris tics of nucleocytoplasmic transport of H1N1 influenza A virus nuclear export protein" *J Virol*
36. Chutiwitoonchai, Kakisaka, Yamada et al. (2014) "Comparative analysis of seven viral nuclear export signals (NESs) reveals the crucial role of nuclear export mediated by the third NES consensus sequence of nucleoprotein (NP) in influenza A virus replication" *PLoS One*
37. Sadowski, Ma, Triezenberg et al. (1988) "GAL4-VP16 is an unusually potent transcriptional activator" *Nature*
38. Zhao, Xia, Huang et al. (2020) "Features of nuclear export signals of NS2 protein of influenza D virus" *Viruses*
39. Kosugi, Hasebe, Tomita et al. (2008) "Nuclear export signal consensus sequences defined using a localization-based yeast selection system" *Traffic*
40. Xu, Marquis, Pei et al. (2015) "LocNES: a computational tool for locating classical NESs in CRM1 cargo proteins" *Bioinformatics*
41. Skelton, Huber (2022) "Comparing influenza virus biology for understanding influenza D virus" *Viruses*
42. Biering, Falk, Hoel et al. (2002) "Segment 8 encodes a structural protein of infectious salmon anaemia virus (ISAV); the co-linear transcript from Segment 7 probably encodes a non-structural or minor structural protein" *Dis Aquat Org*
43. Batts, Lapatra, Katona et al. (2017) "Molecular characterization of a novel orthomyxovirus from rainbow and steelhead trout (Oncorhynchus mykiss)" *Virus Res*
44. Mohr, Crane, Hoad et al. (2020) "Pilchard orthomyxovirus (POMV). I. Characterisation of an emerging virus isolated from pilchards Sardinops sagax and Atlantic salmon Salmo salar" *Dis Aquat Organ*
45. Cler, Fuller, Bishop (1983) "Tick-borne viruses structurally similar to orthomyxoviruses" *Virology*
46. Iwatsuki-Horimoto, Horimoto, Fujii et al. (2004) "Generation of influenza A virus NS2 (NEP) mutants with an altered nuclear export signal sequence" *J Virol*
47. Hirohama, Yamashita, Asaka et al. (2023) "Intramolecular interaction of NEP regulated by CRM1 ensures the unidirectional transport of M1 for the nuclear export of influenza viral ribonucleoprotein" *Front Virol*
48. Ye, Robinson, Wagner (1995) "Nucleus-targeting domain of the matrix protein (M1) of influenza virus" *J Virol*
49. Schreiber, Boff, Anhlan et al. (2020) "Dissecting the mechanism of signaling-triggered nuclear export of newly synthesized influenza virus ribonucleoprotein complexes" *Proc Natl Acad Sci*
50. Newlands, Rustin, Brampton (1996) "Phase I trial of elactocin" *Br J Cancer*
51. Benkova, Mihalyova, Hajek et al. (2021) "Selinexor, selective inhibitor of nuclear export: Unselective bullet for blood cancers" *Blood Rev*
52. Perwitasari, Johnson, Yan et al. (2014) "Verdi nexor, a novel selective inhibitor of nuclear export, reduces influenza a virus replication in vitro and in vivo" *J Virol*
53. Thompson, Higgins, Gibson (1994) "CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice" *Nucleic Acids Res*
54. Barry, Barry, Johnston (1994) "Production of monoclonal antibodies by genetic immunization" *BioTechniques*
55. Barry, Johnston (1997) "Biological features of genetic immuniza tion" *Vaccine*
56. Xiang, Ertl (1995) "Manipulation of the immune response to a plasmid-encoded viral antigen by coinoculation with plasmids expressing cytokines" *Immunity*
57. Bolte, Cordelières (2006) "A guided tour into subcellular colocaliza tion analysis in light microscopy" *J Microsc*
58. Schneider, Rasband, Eliceiri (2012) "NIH Image to ImageJ: 25 years of image analysis" *Nat Methods*
59. Flick, Flick, Feldmann et al. (2003) "Reverse genetics for crimeancongo hemorrhagic fever virus" *J Virol*
60. Yamaoka, Weisend, Swenson et al. (2022) "Development of accelerated high-throughput antiviral screening systems for emerging orthomyxoviruses" *Antiviral Res*
61. Niwa, Yamamura, Miyazaki (1991) "Efficient selection for highexpression transfectants with a novel eukaryotic vector" *Gene*
62. Ng, Li, Jeong et al. (2018) "Systematic functional annotation of somatic mutations in cancer" *Cancer Cell*
63. (2025) *Full-Length Text Journal of Virology* |
biology | europe-pmc | https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC12652138&blobtype=pdf | # A Bibliometric Analysis of the HCV Drug-Resistant Majority and Minority Variants
Omega Mathew, Olaoluwa Fabiyi, Kuat Oshakbayev, Gulzhan Abuova, Aliya Konysbekova, Sreenu Vattipally, Syed Ali, Syed Abidi
## Abstract
Background: In recent decades, research on Hepatitis C Virus (HCV) drug-resistant variants has expanded; however, critical gaps remain in our understanding of global contributions, emerging trends, and future research directions. Here, we present a bibliometric analysis to understand the research themes and trends in research related to HCV drug-resistant variants published between 1999 and 2025. Methods: Publications related to HCV drugresistant variants published between 1999 and 2025 were searched on the Web of Science and Scopus databases. Publication metadata and content-based data were extracted and analyzed using Bibliometrix and VOSviewer for keyword co-occurrence plot and cluster analysis. Results: The analysis of 653 articles revealed a clear paradigm shift, driven by the introduction of direct-acting antivirals (DAAs), which led to a significant surge in annual publications, peaking between 2014 and 2018. This shift in focus led to an emphasis on DAA efficacy, resistance mechanisms, and advanced genotyping. The United States was the most productive country, with the highest number of publications (n = 134) and citations (n = 6458). The University of São Paulo was the most productive institution (n = 40), while Antimicrobial Agents and Chemotherapy published the highest number of articles in this field (n = 40). Susser S. was the most productive researcher. Collaboration networks were found to be predominantly centered in high-income countries. Analysis of studies on minority variants showed that most studies originated from Europe and the United States, identifying low-frequency resistance-associated substitutions (RASs) such as A156V, D168V, Y93H, and S282T, with prevalence ranging from <1% to 35%, which were frequently associated with viral breakthrough and reduced treatment response. Conclusions: The field successfully transitioned to the DAA era, but research output and collaboration networks were primarily driven by high-income countries, leaving a critical gap in data from Lowand Middle-Income Countries (LMICs). Closing this gap by integrating LMIC data is the next essential step to ensure global elimination strategies are effective for all countries from different income strata.
## 1. Introduction
Hepatitis C Virus (HCV) infection is a major global health challenge, responsible for 50 million infections and 24,200 deaths globally as of 2022 [1]. HCV is a bloodborne virus that primarily affects the liver, causing hepatitis, cirrhosis, and hepatocellular carcinoma [2]. The virus's high mutation rate, shaped by error-prone RNA-dependent RNA polymerase, high replication rate, and host-and drug-selection pressure, drives its genetic diversity, which leads to the emergence of drug-resistant variants/quasispecies that can negatively affect the treatment efficacy [3][4][5].
HCV research has undergone a temporal evolution, closely tied to the significant changes in HCV diagnostics and treatment strategies. Treatment has evolved from the interferon era to a revolution with first-generation DAAs, and currently to the pan-genotypic DAA era, which can achieve >95% cure rates across all major genotypes with a high barrier to resistance [6][7][8]. Interferon-based therapies were once the mainstay of treatment for HCV [9]. However, current HCV treatment relies on direct-acting antivirals (DAAs) targeting viral proteins essential for replication [10]. The first-generation DAAs, the NS3/4A protease inhibitors, were associated with a limited spectrum and efficacy, and significant resistance [6,7], leading to the introduction of second-generation DAAs, NS3/4A, NS5B polymerase, and NS5A inhibitors, which have broader genotype coverage and higher efficacy, but with substantial resistance [8,11]. Currently, the pan-genotypic third-generation DAAs, such as glecaprevir/pibrentasvir and sofosbuvir/velpatasvir, are used in the management of HCV because they are more effective (achieve a sustained virologic response rate of ≥95%) and have a high barrier to resistance [12,13]. Regimens are tailored based on genotype, prior treatment, cirrhosis status, and resistance profile [14,15]. Pan-genotypic DAAs have played a crucial role in reaching the WHO HCV global elimination target, which aims for a 90% diagnosis rate and an 80% treatment rate among infected people, resulting in an 80% decrease in new infections and a 65% decline in mortality [16].
HCV drug resistance arises from resistance-associated substitutions (RASs) occurring in the NS3/4A, NS5A, or NS5B regions of the viral genome, which can reduce drug binding or activity [5,13,[17][18][19]. RASs can exist naturally (baseline RASs) or emerge during treatment due to selective pressure from suboptimal treatment (e.g., monotherapy, poor adherence) [8]. Baseline RASs are detected in 5-20% of untreated patients, varying by genotype [20,21]. Baseline RASs are more common in genotypes 1a and 3, and treatmentemergent RASs, such as NS3 Q80K, complicate retreatment due to cross-resistance [22,23].
Drug resistance has significantly affected the global HCV elimination goal, especially in resource-limited settings with limited access to resistance testing [24,25]. Drug-resistant variants have significant public health implications, with both the dominant (majority) and low-frequency (minority) variants contributing to treatment failure, particularly in treatment-experienced or cirrhotic patients [12]. Therefore, resistance testing is essential to reduce the burden of resistant variants [26].
Advances in DAAs and molecular typing have improved outcomes; however, ongoing research is needed to improve access to HCV diagnostics and treatment, and address the issue of antiviral resistance [24]. A shift from Sanger sequencing, which detects dominant RASs but misses low-frequency variants (<20% of the viral population [27]), to next-generation sequencing (NGS) that detects low-frequency RASs (1-5%) with high sensitivity, has significantly improved resistance profiling [21,28]. Additionally, whole-genome deep sequencing technologies have also aided in identifying genotypes/subtypes for treatment selection, as well as in transmission and phylogenetic studies [27,29]. Although the transition from interferon therapy to pan-genotypic DAAs has reduced the risk of drug resistance to antiviral treatment [10], surveillance for viral quasispecies remains critical for predicting outbreaks and managing hard-to-treat populations, such as males (reportedly having lower cure rates than females), genotype 3 (more aggressive genotype) patients, and those with cirrhosis [14,30].
Bibliometrics has been applied in several fields of science [31], as it is an essential tool for assessing and analyzing the output of scientists, collaborations, and research trends [32]. This tool can be used to identify which countries and institutions have made the most significant contributions, which journals publish the most literature, and to determine the current state of research on HCV drug-resistant variants. While studies before 2010 primarily focused on HCV drug resistance mutations, very few have examined minority variants and their impact on treatment outcomes. Consequently, no bibliometric analysis thus far has reported the evolution of research themes, trends, and global contributions in this field. Therefore, this bibliometric study aimed to analyze articles on HCV drugresistant variant research to identify research trends and hotspots and connect the findings to the implementation of global elimination strategies.
## 2. Methodology
The Web of Science (WoS) database, which is widely used for bibliometric analysis [33], was sourced for this study. This database indexes over 12,000 high-impact journals across various disciplines, making it a reliable source for analyzing research output, collaborations, and trends, compared to the Scopus and PubMed databases [34,35]. To further enhance our search results, a complementary search was also performed in the Scopus database. We searched Scopus and WOS for literature published from 1999 to 2025, a period selected to capture the temporal dynamics of HCV drug resistance research, ranging from the use of interferon to modern DAAs. All results from the search query were exported in plain text format for screening and data analysis.
## 2.1. Search Strategy
This study utilized a combination of Medical Subject Headings (MeSH) and non-MESH terms to enhance article retrieval from Scopus and WoS using the following search string: (((ctTS = ("hepatitis C virus" OR "HCV" OR "hepacivirus") AND TS = ("drug resistan*" OR "antiviral resistan*" OR "resistance-associated substitution*" OR RAS)) AND TS = ("majority variant*" OR "minority variant*" OR "low-frequency variant*" OR "highfrequency variant*" OR quasispecies)) OR TS = ("low abundance" OR "high abundance" OR "low frequency" OR "high frequency")) OR TS = (variant* OR mutation* OR substitution*), and TITLE-ABS-KEY ((TS = ("hepatitis C virus" OR "HCV" OR "hepacivirus") AND ("drug resistant" OR "antiviral resistant" OR "resistance-associated substitution" OR RAS)) AND ("majority variant" OR "minority variant" OR "low-frequency variant*" OR "high-frequency variant*" OR quasispecies)), for WoS and Scopus, respectively.
## 2.2. Screening Protocol and Criteria
The initial search yielded 4712 records. Through a systematic screening process conducted independently by two researchers, 15 duplicate articles and 3913 items comprising review articles, conference proceedings, editorials, letters, and book chapters were excluded. The remaining 768 articles were manually screened based on titles, abstracts, and keywords, with a focus on relevance to "HCV drug-resistant majority and minority variants." Where necessary, we performed full-text evaluations to assess eligibility.
The inter-rater agreement, calculated using the kappa statistic, was found to be 0.90, suggesting excellent inter-rater agreement. Any disagreements between the reviewers were resolved through arbitration by the senior author, resulting in a final dataset comprising 653 articles for further analysis.
The literature retrieval and selection process is summarized in Figure 1, following the PRISMA guidelines [36].
## 2.3. Data Analysis
The 653 eligible articles were exported in plain text format. Two complementary bibliometric tools, Bibliometrix (v3.2.1) in R Studio 4.2.3 [37,38] and VOSviewer (v1.6.19) [39,40], were employed to, respectively, analyze annual publication trends, country and institutional contributions, journal distribution, and citation metrics; and keyword co-occurrence mapping, thematic cluster identification, and density visualization of research landscapes.
Additionally, co-authorship networks were generated from bibliographic data to identify patterns of collaboration. Normalized citation metrics were used to account for the varying citation windows of the publications.
## 2.4. In-Depth Analysis of Top-Cited Studies on HCV Minority Variant
To characterize fundamental research on HCV drug-resistant minority variants, a supplementary analysis was conducted on the most-cited articles on HCV minority variants from the final dataset. We systematically selected the top 10 articles with the highest total citation count (TC) and extracted information regarding sequencing method, minority variant detected, prevalence, and country of the corresponding author.
## 3. Results
## 3.1. Analysis of Publication Output
Analysis of 653 articles on HCV drug-resistant variants, from 1999 to 2025, suggested that the annual publication output rose steadily from 1999, peaking at 64 articles in 2018, followed by a gradual decline. The lowest output was recorded in 2002 (n = 4). The mean annual growth rate was 1.82% over the entire period (Figure 2). Notably, the period from 2014 to 2018 marked a surge phase, with an average output of 57.2 publications. However, by 2024, the publications had decreased to approximately one-fifth of the peak volume (Figure 2).
## 3.2. Analysis of Countries and Institutions
The top ten countries contributing to HCV drug-resistant variants research included a mix of developed nations, the USA, Japan, Italy, Spain, Germany, France, and Denmark, and newly industrialized/developing economies (Brazil, Iran, and China; Table 1). Among these, Germany stood out with the highest level of international collaboration (MCP ratio = 40.62).
Table 1. Publication output and international collaboration patterns by corresponding authors' countries. The table shows the top 10 countries contributing the most articles, their frequency (Freq), and the SCP (Single Country Publications), MCP (Multiple Country Publications), and MCP ratio (proportion of a country's publications produced through international collaboration). The Bibliometrix three-field plot, which examines the relationships among institutions, authors, and keywords [37], mapping the top 10 institutions, 15 authors, and 15 keywords, revealed that the research landscape is structured around specialized universities and private research institutions, with a key focus on NS5A, NS5B, and NS3 inhibitors (Figure 3). Analysis of publishing institutes revealed that most articles were from universities and public health organizations. Notably, the University of São Paulo (Brazil) produced the highest institutional output during the review period (n = 40; Figure 4).
## Country
## 3.3. Analysis of Journals
The 653 selected articles were published across 174 journals, highlighting the broad dissemination of research on HCV drug-resistant variants. Table 2 ranks the top ten most active journals in this field by publication volume. Among them, the journal Antimicrobial Agents and Chemotherapy published the most articles in the field (n = 46). However, the Journal of Virology dominated as the journal with the most cited articles (Total Citations = 1285, g-index = 35.85). Keyword: m-index: Annualized h-index; g-index: The number of top papers whose total citations equal at least the square of that number; TC = total citation.
## 3.4. Analysis of Citations
## 3.4.1. Top-Cited Papers on HCV Drug Resistance
Table 3 presents the ten most globally cited papers on HCV drug-resistant variants. The highest-cited study was "Characterization of resistance to the protease inhibitor boceprevir in hepatitis C virus-infected patients" by Susser S et al. [17], published in Hepatology in 2009, with 264 total citations, averaging 15.5 citations per year, while the article "NS5A resistance-associated substitutions in patients with genotype 1 hepatitis C virus: Prevalence and effect on treatment outcome" by Zeuzem S et al. [19] had the highest annual citation rate (19.3 citations/year). When adjusted for citation window differences using Normalized Total Citations (NTC), the article "Prevalence of resistance-associated substitutions in HCV NS5A, NS5B, or NS3 and outcomes of treatment with Ledipasvir and Sofosbuvir" by Sarrazin C et al. [6] ranked highest with NTC of 9.10. The U.S. dominated among the top 10 countries, accumulating 6458 citations and an average article citation of 45.48 (Table 4). The USA was followed by Japan, with 1884 citations, and Germany, with 1125 citations (all developed countries).
## 3.5. Analysis of Collaborative Networks
Analysis of co-authorship networks revealed distinct collaborative clusters representing both national and international research groups (Figure 5). The red cluster, one of the central and most influential networks, was primarily composed of European collaborators, especially centered in Germany, with Zeuzem acting as a key investigator linking researchers from clinical and virological backgrounds. The blue cluster represents a strong within-country network in Japan, indicating high national collaboration and consistent co-authorship among Japanese researchers. Similarly, the green cluster represents a collaborative effort between U.S. and European institutions, highlighting cross-continental research partnerships. The yellow cluster connects the U.S. and Japan; the pink cluster includes Italy and Spain; the orange cluster represents emerging collaborators from India and Denmark; and the brown cluster includes smaller European subgroups, such as Belgium and the Netherlands. The orange cluster depicts emerging or regional collaborations, while the pink cluster represents a relatively independent Southern European network that maintains limited links with the central European core. The brown cluster comprises peripheral European contributors, and the yellow cluster contains bridging authors who connect the green and blue clusters, thereby facilitating international collaboration between the U.S. and Japan. The dense interconnections suggest strong national collaboration within Germany and Japan, as well as consistent co-authorship across multiple publications.
## 3.6. Cluster Analysis and Temporal Trends in HCV Drug Resistance Research
The keyword cluster analysis results for terms with at least five occurrences, generated using VOSviewer, identified the following key terms reported in publications: viral evolution and resistance, deep sequencing, Hepatitis C virus, mutation, NS5B, RASs, viral quasispecies, protease, resistance mutation, viral quasispecies, deep sequencing, sofosbuvir, ledipasvir, daclatasvir, ombitasvir, ritonavir, simeprevir, interferon, alpha interferon, ribavirin, single-nucleotide polymorphism, and resistance, hepatitis C, HCV, molecular typing, genotyping, bioinformatics, pyrosequencing, next-generation sequencing, boceprevir, and telaprevir (Figure 6). An analysis of the temporal co-occurrence of keywords in HCV drug resistance research revealed a timeline of major themes from 1999 to 2025, displayed through a color gradient from purple to red. Early research (purple and blue) mainly focused on interferon therapy, ribavirin, and viral genotyping, representing the pre-DAA period. As the field progressed (green to yellow), attention shifted toward DAAs, such as protease and polymerase inhibitors. The latest phase (orange to red) highlights NGS, RASs, and pan-genotypic regi-mens, emphasizing the increasing use of molecular surveillance for monitoring antiviral resistance (Figure 7).
## 3.7. Analysis of Key Trends in Minority Variants Research
To understand the evolution, focus areas, and research dynamics within studies specifically addressing HCV minority variants, an in-depth analysis was conducted on the most highly cited papers in this field. The characteristics of the ten most-cited studies that explicitly focused on low-frequency variants are summarized in Table 5. The analysis reveals that the most influential studies predominantly originated from European countries (Belgium, Spain, France, Italy, England) and the United States, with Verbinnen, T being the most cited author (59 citations). Earlier studies (2002-2005) relied on clonal and Sanger sequencing, while later studies (2010-2018) used deep sequencing and ultra-deep sequencing platforms. The most commonly detected minority variants include F43S, A156G, A156V, D168V, and D168A in the NS3 protease gene; S282T in the NS5B polymerase gene; and T54S, V36A, V170A, Q80K, V36L, V55A, V36M, and Y93H in the NS5A region, with prevalence ranging from <1-35%. The analysis revealed that the detection of these RASs was consistently linked to reduced treatment efficacy, lower cure rates, and higher rates of treatment failure (Table 5).
## 4. Discussion
This bibliometric analysis examines research trends, publication and citation patterns, and key trends in the field of HCV drug-resistance research, with a focus on minority variants, from 1999 to 2025. By analyzing publications over this period, this study not only identifies temporal publication dynamics and key trends but also highlights persistent knowledge gaps within this field.
## 4.1. Major Publication Trends During the Pre-DAA and DAA Eras and Their Importance for Global Health
Pre-DAA era: From 1999 to 2009, research output was low but consistent (~10-30 papers/year), and most studies focused on interferon-ribavirin therapy, its (limited) efficacy, and significant side effects, diverting focus from resistance mechanisms [9]. Studies during this period mainly focused on resistance in non-responders, as viral escape variants were a secondary concern compared to broader treatment limitations [2].
DAA-era: Between 2010 and 2014, publication rates increased significantly, reaching ~60 papers/year by 2014, driven by the introduction of first-generation protease inhibitors including telaprevir, boceprevir, and danoprevir. Although these drugs improved sustained virological response (SVR rates), they were associated with high resistance rates, likely due to key mutations such as NS3/4A-R155K, NS3/4A-T54, and NS3/V36M, which posed significant clinical challenges such as immune evasion, high virologic relapse, and failure rates, etc. [47,58,59]. Publication output peaked at ~60-62 papers/year between 2015 and 2017, coinciding with the advent of second-generation DAAs, including NS5A inhibitors (e.g., ledipasvir, daclatasvir) and sofosbuvir-based regimens. These therapies improved treatment efficacy but introduced new resistance issues, particularly NS5A-associated mutation Y93H, prompting research into retreatment strategies [42,60,61]. From 2021 to 2025, publications related to drug resistance declined to ~30-40 papers/year, likely coinciding with the success of pan-genotypic regimens such as glecaprevir/pibrentasvir, which achieved SVR rates >95% with minimal resistance [13]. The use of pan-genomic regimens, for instance, sofosbuvir/velpatasvir/voxilaprevir, in patients who have failed prior NS5A-inhibitor treatment, helps overcome viral resistance, achieving cure rates of over 95% even in treatment-experienced patients who previously exhibited resistance to earlier-generation DAAs [13]. The pan-genomic regimens also prevent the emergence of resistant viral strains and reduce treatment failures [13].
The fall in research output (by ~80%) from 2018 to 2024-2025 may also reflect a shift in research focus from drug resistance, due to the success of pan-genotypic DAAs, to HCV elimination goals, point-of-care diagnostics, and addressing underserved populations [15,62]. This change in research focus suggests that highly effective pan-genotypic DAAs have rendered resistance a less critical clinical issue [8,12,13,15,23]. At the same time, the global push for elimination shifted attention and resources toward public health implementation, access, and equity [24][25][26]62]. Additionally, technological advancements such as next-generation sequencing (NGS) have fundamentally transformed the understanding and study of HCV drug resistance, viral diversity, and treatment outcomes [27,28].
The global distribution and accessibility of third-generation pan-genotypic DAAs represent a significant milestone in the worldwide effort to eliminate HCV, owing to their high efficacy (SVR > 95%) across all genotypes and patient subgroups, and their ability to markedly reduce the risk of treatment failure, viral breakthrough, and the emergence of new resistant strains [8,13,23]. Notably, the widespread rollout of sofosbuvir/velpatasvir and glecaprevir/pibrentasvir has already led to a sharp decline in resistance-related treatment failures in both clinical trials and real-world settings [13,14,23], underscoring their crucial role in controlling global HCV drug resistance. Despite this success, global inequities in access remain a substantial challenge. In LMICs, high drug prices, limited diagnostic infrastructure, and supply chain barriers restrict access to third-generation DAAs, threatening to sustain reservoirs of resistant HCV strains [24,25,63,64]. Thus, there is a need to systematically address these issues to ensure that HCV elimination goals can be successfully and equally achieved in all parts of the world [65,66].
## 4.2. Major Publishing Countries and Significant Challenges for Global HCV Research
The analysis of major countries contributing to HCV resistance research showed that the United States had the most research output (134 articles), followed by Japan (92 articles) and China (54 articles). This reflects robust research infrastructure in the U.S. and Japan [67] and China's growing focus on infectious diseases [68,69]. Notably, Germany demonstrated the highest level of international collaboration with an MCP ratio of 40.62. Universities and public health institutions, notably the University of São Paulo, the University of Copenhagen, and Hiroshima University, were key contributors, providing region-specific resistance data that may have played a key role in optimizing DAA treatments [70][71][72][73][74]. The three-field visualization mapping authors, keywords, and their affiliated institutions identified five distinct clusters, corresponding to established research teams from Germany, Denmark, Sweden, the U.S., and Japan, as well as emerging or peripheral collaborators alongside bridging authors from India and Italy. These research clusters focused mainly on "NS5A," "NS5B," and "NS3," suggesting that the most significant research activities were concentrated on the viral targets of DAAs, reflecting the field's focus on the mechanisms of antiviral resistance, as discussed in foundational papers on DAA efficacy and resistance [8,13,17]. Furthermore, the collaboration network also revealed that the research is primarily driven by collaborations among academic hospitals, universities, and pharmaceutical entities. This underscores the translational nature of the work, bridging basic virology with drug development.
While the collaboration network indicates that key contributions come from institutions in developed countries, it also reveals a critical gap in contributions from institutions in LMICs. This disparity threatens the equity of global elimination efforts, as highlighted in recent analyses [24,25]. For example, Alenzi [25] reported that in LMICs, factors such as out-of-pocket expenses, lower income, and lack of awareness contribute to reduced access to healthcare services and lower treatment uptake, especially among disadvantaged populations [25]. These inequities can be addressed through multifaceted interventions, such as providing free treatment (for example, through support from government or governmentprivate partnership), reducing the prices of DAAs, expanding healthcare coverage, and strategic collaboration between industry, academic, community, and non-profit researchers to support the production and distribution of generic medicines in LMICs [25,63,65,66]. These supports can also enhance screening, linkage to care, and treatment adherence, thereby narrowing the disparity gap and accelerating progress toward the global hepatitis C elimination goals.
Highly cited works include Susser et al. [17], with 264 citations for characterizing NS5A resistance-associated substitutions (RASs), and Zeuzem et al. [19], with a high annual citation rate (19.3 citations/year) for shaping treatment guidelines. Lin et al. [41,45] provided foundational insights into NS3/4A resistance mechanisms. The U.S. led in citations (6458), driven by NIH funding and pharmaceutical involvement, while Japan and Europe contributed region-specific data. China's increasing output may suggest its growing influence and contribution to the research. A gradual decline in annual citations (2.013 fewer citations per year) may signal a maturing field as DAA therapies stabilize. It is important to note that the environment of funding cuts could significantly impact HCV research, among other fields [75,76].
## 4.3. Key Trends in HCV Drug Resistance and Its Impact on Global HCV Research
Keyword analysis revealed thematic clusters reflecting the evolution of HCV research. The cluster of "interferon" and "ribavirin" reflects the pre-DAA era, characterized by low SVR rates, and "IL28B polymorphism" may indicate the influence of host genetics on viral control [77][78][79]. Similarly, keywords such as "sofosbuvir," "ledipasvir", and "daclatasvir" mark the DAA revolution, though resistance persisted across genotypes [80,81]. Additionally, keywords such as "mutation", "viral quasispecies", "NS5B", "protease", "viral evolution and resistance", "direct-acting antivirals (DAAs)", "molecular typing and genotyping", "bioinformatics", and "sequencing" highlight resistance under antiviral pressure, with deep sequencing enabling detection of low-frequency mutations [21,29].
Overall, the trend analysis shows a shift from interferon-based studies and early sequencing to DAAs and modern typing methods. HCV research has evolved from addressing interferon limitations (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009), to characterizing resistance with first-generation protease inhibitors (2010-2014), to NS5A resistance during the DAA peak (2015-2017), and finally to focusing on elimination and access with pan-genotypic regimens (2021-2025). The transition to NGS has enhanced resistance detection, treatment optimization, and elimination monitoring [82,83]. This advancement has significant public health implications for the elimination of HCV. By identifying low-frequency minority variants that often cause treatment failure, NGS enables clinicians to select the most effective regimens from the outset (for example, baseline mutations) [25,84]. This detailed genomic data also strengthens national surveillance, helping to track emerging resistance and transmission clusters, which is vital for effective elimination programs [85,86]. Ultimately, NGS facilitates a "precision medicine" approach that improves cure rates, even in complex cases [87,88]. Collectively, these gains make the fight against HCV more efficient and bring the WHO's 2030 elimination target within closer reach [25,89].
## 4.4. Research on HCV Minority Variants and Their Impact on Global HCV Elimination
The most cited research on minority variants was conducted in European countries (Belgium, Spain, France, and Italy) and the United States. This aligns with the broader bibliometric finding that high-income countries have driven the research agenda. Technologically, a clear evolution was seen from the early use of clonal and Sanger sequencing, which have limited sensitivity for low-frequency variants, to the adoption of more sensi-tive deep and ultra-deep sequencing methods in later studies. This shift was pivotal, as it unveiled a previously hidden layer of viral diversity (minor variants) including F43S, A156G, D168V/A in the NS3 region, S282T in NS5B, and key NS5A variants like Y93H, V36A/M/L, and T54S. These variants, which are known to confer resistance to various DAA classes, can pre-exist at low frequencies (<1% to 35%) and expand under selective drug pressure, leading to treatment failure [90][91][92].
The introduction of highly effective pan-genotypic DAAs has markedly reduced the risk that a single pre-existing minority variant will lead to treatment failure [13]. Despite this progress, concerns remain for specific, hard-to-treat populations. These include patients with advanced cirrhosis or prior treatment failures, who harbor more diverse and complex viral quasispecies [14,23]; individuals infected with genotype 3, where RASs like Y93H can still compromise the efficacy of certain regimens [18,23]; and male patients, who exhibit lower cure rates, potentially due to a complex interplay of factors [30].
In this context, NGS is a critical tool for ensuring therapeutic success. Critically, deploying NGS in LMICs is a strategic necessity for global elimination. It facilitates the characterization of region-specific viral genetics, ensuring treatment strategies are effective against local strains and optimizing the use of available therapies in resource-limited settings [24,25].
We anticipate certain limitations of this study. The main limitation relates to the database selection, as the analysis was limited to articles listed in WoS and Scopus. These are the two most trusted sources of bibliometric data [93]. However, WoS is widely used for bibliometric analysis [33], as a more reliable database compared to Scopus and PubMed databases [34,35]. Secondly, in this bibliometric analysis, all selected studies were published in English. It is important to mention that our search queries were not restricted by language; the search databases may have a bias towards English-language publications, as no non-English studies were identified [93]. However, for the current bibliometric analysis, which aims to provide key research trends in HCV drug resistance research, the impact of excluding relevant non-English studies may have been minimal [94,95]. This limitation can be addressed in a future systematic review on the same subject, where each included study carries substantial weight in the final evidence synthesis [96].
## 5. Conclusions
This bibliometric analysis provides a comprehensive overview of the evolving research landscape on HCV drug-resistant variants from 1999 to 2025. The findings show a clear shift from interferon failure to optimizing DAA efficacy through the NGS-based surveillance of minority variants. However, this progress was found to be geographically uneven, as the HCV research in this sphere, along with collaborations, was concentrated in highincome countries. This is concerning and highlights strategic vulnerability in the global elimination agenda. The most pressing gap seems no longer technical, but geographical and inequitable. Current and future research efforts should be focused on ensuring that pan-genotypic treatment is globally available and accessible to all patients. Furthermore, continued monitoring of minority variants, primarily through highly sensitive NGS-based assays, is required to ensure that the third-generation DAAs stay effective and can be effectively implemented as a strategy to eliminate HCV globally.
## Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijerph22111670/s1, File S1: Raw data used to generate figures and tables reported within the article.
## References
1. Pawlotsky (2006) "Virology of hepatitis B and C viruses and antiviral targets" *J. Hepatol*
2. Xu, Rong, Aranday-Cortes et al. (2022) "The Transmission Route and Selection Pressure in HCV Subtype 3a and 3b Chinese Infections: Evolutionary Kinetics and Selective Force Analysis" *Viruses*
3. Ikram, Obaid, Awan et al. (2017) "Identification of drug resistance and immune-driven variations in hepatitis C virus (HCV) NS3/4A, NS5A and NS5B regions reveals a new approach toward personalized medicine" *Antivir. Res*
4. Martinez, Franco (2020) "Therapy Implications of Hepatitis C Virus Genetic Diversity" *Viruses*
5. Sarrazin, Dvory-Sobol, Svarovskaia et al. (2016) "Prevalence of resistance-associated substitutions in HCV NS5A, NS5B, or NS3 and outcomes of treatment with ledipasvir and sofosbuvir" *Gastroenterology*
6. Welsch, Domingues, Susser et al. (2008) "Molecular basis of telaprevir resistance due to V36 and T54 mutations in the NS3-4A protease of the hepatitis C virus" *Genome Biol*
7. Cada, Leonard, Levien et al. (2015) *Hosp. Pharm*
8. Manns (2006) "Treating viral hepatitis C: Efficacy, side effects, and complications" *Gut*
9. Marco, Cannova, Ferrigno et al. (2025) "A Comprehensive Review of Antiviral Therapy for Hepatitis C: The Long Journey from Interferon to Pan-Genotypic Direct-Acting Antivirals (DAAs)" *Viruses*
10. Joharji, Alkortas, Ajlan et al. "Efficacy of generic sofosbuvir with daclatasvir compared to sofosbuvir/ledipasvir in genotype 4 hepatitis C virus: A prospective comparison with historical control" *Health Sci*
11. Yen, Chen, Lai et al. (1853) "Pan-Genotypic Direct-Acting Antiviral Agents for Undetermined or Mixed-Genotype Hepatitis C Infection: A Real-World Multi-Center Effectiveness Analysis" *J. Clin. Med*
12. Zarębska-Michaluk, Jaroszewicz, Parfieniuk-Kowerda et al. (2021) "Effectiveness and Safety of Pangenotypic Regimens in the Most Difficult to Treat Population of Genotype 3 HCV Infected Cirrhotics" *J. Clin. Med*
13. Pabjan, Brzdęk, Chrapek et al. (2022) "Are There Still Difficult-to-Treat Patients with Chronic Hepatitis C in the Era of Direct-Acting Antivirals? Viruses"
14. Flisiak, Zarębska-Michaluk, Berak et al. (2024) "Treatment with Sofosbuvir/Velpatasvir of the most difficult to cure HCV-infected population" *Pol. Arch. Intern. Med*
15. Hellard, Schroeder, Pedrana et al. (2020) "The Elimination of Hepatitis C as a Public Health Threat. Cold Spring Harb" *Perspect. Med*
16. Susser, Welsch, Wang et al. (2009) "Characterization of resistance to the protease inhibitor boceprevir in hepatitis C virus-infected patients" *Hepatology*
17. Kjellin, Kileng, Akaberi et al. (2019) "Effect of the baseline Y93H resistance-associated substitution in HCV genotype 3 for direct-acting antiviral treatment: Real-life experience from a multicenter study in Sweden and Norway" *Scand. J. Gastroenterol*
18. Zeuzem, Mizokami, Pianko et al. (2017) "NS5A resistance-associated substitutions in patients with genotype 1 hepatitis C virus: Prevalence and effect on treatment outcome" *J. Hepatol*
19. Bartels, Sullivan, Zhang et al. (2013) "Hepatitis C Virus Variants with Decreased Sensitivity to Direct-Acting Antivirals (DAAs) Were Rarely Observed in DAA-Naive Patients prior to Treatment" *J. Virol*
20. Chen, Perales, Soria et al. (2020) "Deep-sequencing reveals broad subtype-specific HCV resistance mutations associated with treatment failure" *Antivir. Res*
21. Sorbo, Cento, Di Maio et al. (2018) "Ceccherini-Silberstein, F. Hepatitis C virus drug resistance associated substitutions and their clinical relevance: Update" *Drug Resist. Updat*
22. Malandris, Kalopitas, Theocharidou et al. (2021) "The Role of RASs/RVs in the Current Management of HCV" *Viruses*
23. Su, Yang, Chang et al. (2023) "A new collaborative care approach toward hepatitis C elimination in marginalized populations" *J. Infect. Public Health*
24. Alenzi, Almeqdadi (2024) "Bridging the gap: Addressing disparities in hepatitis C screening, access to care, and treatment outcomes" *World J. Hepatol*
25. Biondi, Feld (2020) "Hepatitis C models of care: Approaches to elimination" *Can. Liver J*
26. Olmstead, Montoya, Chui et al. (2019) "A systematic, deep sequencing-based methodology for identification of mixed-genotype hepatitis C virus infections" *Infect. Genet. Evol*
27. Bradshaw, Bibby, Manso et al. (2022) "Clinical evaluation of a Hepatitis C Virus whole-genome sequencing pipeline for genotyping and resistance testing" *Clin. Microbiol. Infect*
28. Yamashita, Takeda, Takai et al. (2020) "Single-molecular real-time deep sequencing reveals the dynamics of multi-drug resistant haplotypes and structural variations in the hepatitis C virus genome" *Sci. Rep*
29. Tsukiyama-Kohara, Kohara (2017) "Hepatitis C Virus: Viral Quasispecies and Genotypes" *Int. J. Mol. Sci*
30. Mokhnacheva, Tsvetkova (2020) "Development of Bibliometrics as a Scientific Field" *Sci. Tech. Inf. Process*
31. Moral-Muñoz, Herrera-Viedma, Santisteban-Espejo et al. "Software tools for conducting bibliometric analysis in science: An up-to-date review"
32. Pranckut Ė "Web of Science (WoS) and Scopus: The Titans of Bibliographic Information in Today's Academic World. Publications 2021"
33. Martín-Martín, Thelwall, Orduna-Malea et al. (2021) "Dimensions, Web of Science, and OpenCitations' COCI: A multidisciplinary comparison of coverage via citations" *Scientometrics*
34. Visser, Van Eck, Waltman "Large-scale comparison of bibliographic data sources: Scopus, Web of Science" *Dimensions, Crossref, and Microsoft Academic. Quant. Sci. Stud*
35. Yan, Yu, Li et al. (2025) "A bibliometric analysis of HIV-1 drug-resistant minority variants from 1999 to 2024" *AIDS Res. Ther*
36. Zhao, Li (2023) "Worldwide trends in prediabetes from 1985 to 2022: A bibliometric analysis using bibliometrix R-tool" *Front. Public Health*
37. Markscheffel, Schröter (2021) "Comparison of two science mapping tools based on software technical evaluation and bibliometric case studies" *COLLNET J. Sci. Inf. Manag*
38. Ullah, Shen, Shah (2023) "Value co-creation in business-to-business context: A bibliometric analysis using HistCite and VOS viewer" *Front. Psychol*
39. Page, Mckenzie, Bossuyt et al. "The PRISMA 2020 statement: An updated guideline for reporting systematic reviews" *BMJ*
40. Lin, Lin, Luong et al. (2004) "In Vitro Resistance Studies of Hepatitis C Virus Serine Protease Inhibitors, VX-950 and BILN 2061" *J. Biol. Chem*
41. Fridell, Qiu, Wang et al. (2010) "Resistance Analysis of the Hepatitis C Virus NS5A Inhibitor BMS-790052 in an In Vitro Replicon System" *Antimicrob. Agents Chemother*
42. Lenz, Verbinnen, Lin et al. (2010) "In vitro resistance profile of the hepatitis C virus NS3/4A protease inhibitor TMC435" *Antimicrob. Agents Chemother*
43. Tong, Chase, Skelton et al. (2006) "Identification and analysis of fitness of resistance mutations against the HCV protease inhibitor SCH 503034" *Antivir. Res*
44. Lin, Gates, Rao et al. (2005) "In Vitro Studies of Cross-resistance Mutations against Two Hepatitis C Virus Serine Protease Inhibitors, VX-950 and BILN 2061" *J. Biol. Chem*
45. Svarovskaia, Dvory-Sobol, Parkin et al. (2014) "Infrequent development of resistance in genotype 1-6 hepatitis C virus-infected subjects treated with sofosbuvir in phase 2 and 3 clinical trials" *Clin. Infect. Dis*
46. Romano, Ali, Aydin et al. (2012) "The Molecular Basis of Drug Resistance against Hepatitis C Virus NS3/4A Protease Inhibitors" *PLoS Pathog*
47. Verbinnen, Van Marck, Vandenbroucke et al. (2010) "Tracking the evolution of multiple in vitro hepatitis C virus replicon variants under protease inhibitor selection pressure by 454 deep sequencing" *J. Virol*
48. Puig-Basagoiti, Forns, Furčić et al. (2005) "Dynamics of hepatitis C virus NS5A quasispecies during interferon and ribavirin therapy in responder and non-responder patients with genotype 1b chronic hepatitis C" *J. Gen. Virol*
49. Soler, Pellerin, Malnou et al. (2002) "Quasispecies heterogeneity and constraints on the evolution of the 5 ′ noncoding region of hepatitis C virus (HCV): Relationship with HCV resistance to interferon-alpha therapy" *Virology*
50. Di Maio, Cento, Mirabelli et al. (2014) "Ceccherini-Silberstein, F. Hepatitis C virus genetic variability and the presence of NS5B resistance-associated mutations as natural polymorphisms in selected genotypes could affect the response to NS5B inhibitors" *Antimicrob. Agents Chemother*
51. Gregori, Esteban, Cubero et al. (2013) "Ultra-deep pyrosequencing (UDPS) data treatment to study amplicon HCV minor variants" *PLoS ONE*
52. Soria, Gregori, Chen et al. (2018) "Pipeline for specific subtype amplification and drug resistance detection in hepatitis C virus" *BMC Infect. Dis*
53. Pogam, Yan, Chhabra et al. (2012) "Characterization of hepatitis C virus (HCV) quasispecies dynamics upon short-term dual therapy with the HCV NS5B nucleoside polymerase inhibitor mericitabine and the NS3/4 protease inhibitor danoprevir" *Antimicrob. Agents Chemother*
54. Akuta, Suzuki, Seko et al. (2013) "Emergence of telaprevir-resistant variants detected by ultra-deep sequencing after triple therapy in patients infected with HCV genotype 1" *J. Med. Virol*
55. Beloukas, King, Childs et al. (2015) "Detection of the NS3 Q80K polymorphism by Sanger and deep sequencing in hepatitis C virus genotype 1a strains in the UK" *Clin. Microbiol. Infect*
56. Murakami, Imamura, Hayes et al. (2014) "Ultradeep sequencing study of chronic hepatitis C virus genotype 1 infection in patients treated with daclatasvir, peginterferon, and ribavirin" *Antimicrob. Agents Chemother*
57. Salloum, Kluge, Kim et al. (2010) "The resistance mutation R155K in the NS3/4A protease of hepatitis C virus also leads the virus to escape from HLA-A*68-restricted CD8 T cells" *Antivir. Res*
58. Özen, Lin, Romano et al. (2018) "Resistance from Afar: Distal Mutation V36M Allosterically Modulates the Active Site to Accentuate Drug Resistance in HCV NS3/4A Protease"
59. Wang, Terrault, Reeves et al. (2018) "Prevalence and impact of baseline resistance-associated substitutions on the efficacy of ledipasvir/sofosbuvir or simeprevir/sofosbuvir against GT1 HCV infection" *Sci. Rep*
60. Fuentes, Abu-Dayyeh, De Salazar et al. (2023) "Molecular characterization of patients with chronic hepatitis C virus infection in Jordan: Implications on response to direct-acting antiviral therapy" *Int. J. Infect. Dis*
61. Ward, Simmons, Fraser et al. (2025) "Mc Pherson, S. Impact and cost-effectiveness of scaling up HCV testing and treatment strategies for achieving HCV elimination among people who inject drugs in England: A mathematical modelling study" *Lancet Reg. Health-Eur*
62. Nisingizwe, Tadrous, Janjua et al. (2025) "Global disparities in access to hepatitis C medicines before and during the early phase of the COVID-19 pandemic: An ARIMA-based interrupted time series analysis" *BMJ Public Health*
63. Venkatesh, Huang, Gurmessa et al. (2024) "Understanding Barriers to Hepatitis C Antiviral Treatment in Low-Middle-Income Countries" *Healthcare*
64. Reau, Lazarus, Solomon et al. "Closing the gap in HCV care: Strategic collaboration between industry, academic, community, and nonprofit researchers"
65. Hill, Simmons, Gotham et al. (2016) "Rapid reductions in prices for generic sofosbuvir and daclatasvir to treat hepatitis C" *J. Virus Erad*
66. Lee, Kang, Kim "Global Collaboration Research Strategies for Sustainability in the Post COVID-19 Era: Analyzing Virology-Related National-Funded Projects"
67. Lou (2020) "Infectious Disease Research in China" *ACS Infect. Dis*
68. Ma, Zhang, Li et al. (2023) "China's innovation and research contribution to combating neglected diseases: A secondary analysis of China's public research data" *Glob. Health Res. Policy*
69. Rodrigues, Campos, Bittar et al. (2022) "Selection dynamics of HCV genotype 3 resistance-associated substitutions under direct-acting antiviral therapy pressure" *Braz. J. Infect. Dis*
70. Torres, Silva, Mendes-Corrêa et al. (2021) "Prevalence and Pattern of Resistance in NS5A/NS5B in Hepatitis C Chronic Patients Genotype 3 Examined at a Public Health Laboratory in the State of São Paulo" *Brazil. Infect. Drug Resist*
71. Mejer, Fahnøe, Galli et al. (2020) "Mutations Identified in the Hepatitis C Virus (HCV) Polymerase of Patients with Chronic HCV Treated with Ribavirin Cause Resistance and Affect Viral Replication Fidelity" *Antimicrob. Agents Chemother*
72. Pham, Pedersen, Fahnøe et al. (2022) "HCV genome-wide analysis for development of efficient culture systems and unravelling of antiviral resistance in genotype 4" *Gut*
73. Teraoka, Uchida, Imamura et al. (2018) "Prevalence of NS5A resistance associated variants in NS5A inhibitor treatment failures and an effective treatment for NS5A-P32 deleted hepatitis C virus in humanized mice" *Biochem. Biophys. Res. Commun*
74. Mcgowan, Fried (2012) "Barriers to hepatitis C treatment" *Liver Int*
75. Faiman (2025) "The Impact of Federal Funding Cuts on Research, Practice, and Patient Care" *J. Adv. Pract. Oncol*
76. Ferreira, Ramos, Nunes et al. (2020) "Hepatitis C Virus: Evading the Intracellular Innate Immunity" *J. Clin. Med*
77. Lapa, Garbuglia, Capobianchi et al. (2019) "Hepatitis C Virus Genetic Variability, Human Immune Response, and Genome Polymorphisms: Which Is the Interplay? Cells"
78. Xu, Chen, Chen "Host Innate Immunity Against Hepatitis Viruses and Viral Immune Evasion" *Front. Microbiol. 2021*
79. Köklü, Köksal, Akarca et al. (2017) "Daclatasvir Plus Asunaprevir Dual Therapy for Chronic HCV Genotype 1b Infection: Results of Turkish Early Access Program" *Ann. Hepatol*
80. Menzo, Biagi, Nuzzo et al. (2018) "SVR 24 Achievement Two Weeks After a Tripled Dose of Daclatasvir in an HCV Genotype 3 Patient" *Ann. Hepatol*
81. Quiñones-Mateu, Avila, Reyes-Teran et al. (2014) "Deep sequencing: Becoming a critical tool in clinical virology" *J. Clin. Virol*
82. Satam, Joshi, Mangrolia et al. (2023) *Next-Generation Sequencing Technology: Current Trends and Advancements. Biology*
83. Lu, Muller, He (2020) "Applying next-generation sequencing to unravel the mutational landscape in viral quasispecies" *Virus Res*
84. Horsburgh, Walker, Kelleher et al. (2024) "Next-Generation Sequencing Methods for Near-Real-Time Molecular Epidemiology of HIV and HCV" *Rev. Med. Virol*
85. (2020) "Whole-genome sequencing as part of national and international surveillance programmes for antimicrobial resistance: A roadmap"
86. Yadav, Patil-Takbhate, Khandagale et al. (2023) "Next-Generation sequencing transforming clinical practice and precision medicine" *Clin. Chim. Acta*
87. Hendi, Mahdi, Alyafie (2025) "Advanced Hepatitis Management: Precision Medicine Integration"
88. Shivangi; Mishra, Gupta, Razdan et al. (2024) "Clinical diagnosis of viral hepatitis: Current status and future strategies" *Diagn. Microbiol. Infect. Dis*
89. Jardim, Bittar, Matos et al. (2013) "Analysis of HCV quasispecies dynamic under selective pressure of combined therapy" *BMC Infect. Dis*
90. Mushtaq, Hashmi, Khan et al. (2022) "Emergence and Persistence of Resistance-Associated Substitutions in HCV GT3 Patients Failing Direct-Acting Antivirals" *Front. Pharmacol*
91. Caputo, Diotti, Boeri et al. (2020) "Detection of low-level HCV variants in DAA treated patients: Comparison amongst three different NGS data analysis protocols" *Virol. J*
92. Tennant (2020) "Web of Science and Scopus are not global databases of knowledge" *Eur. Sci. Ed*
93. Dobrescu, Nussbaumer-Streit, Klerings et al. (2021) "Restricting evidence syntheses of interventions to English-language publications is a viable methodological shortcut for most medical topics: A systematic review" *J. Clin. Epidemiol*
94. Morrison, Polisena, Husereau et al. (2012) "The Effect of English-Language Restriction on Systematic Review-Based Meta-Analyses: A Systematic Review of Empirical Studies" *Int. J. Technol. Assess. Health Care*
95. Cumpston, Mckenzie, Welch et al. (2022) "Strengthening systematic reviews in public health: Guidance in the Cochrane Handbook for Systematic Reviews of Interventions, 2nd edition" *J. Public Health*
96. "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=PMC12512584&blobtype=pdf | # Striking tick-borne virus diversity and potential reservoirs documented during One-Health-based cross-sectional screening in Anatolia
Ender Dinçer, Mehmet Özkan Timurkan, Bekir Oğuz, Emre Ozan, Nüvit Coşkun, Şemistan Kızıltepe, Pelin Polat Dinçer, Adem Şahan, Deniz Yalçınkaya, Ömer Faruk Gökçeçik, Berat Nayir, Sercan Hüseyin Bayendur, Brian Bourke, Yvonne-Marie Linton, Koray Ergunay
## Abstract
Background An expansion of recently described human pathogenic tick-borne viruses from Central Asia toward Europe has been documented. Located on important bird migration routes, Anatolia is an intercontinental crossing hub with various climactic zones and with an abundance of endemic tick species. We sought to investigate tick-borne viruses utilizing a One Health approach encompassing host-removed ticks and host samples.
MethodsWe collected host-attached ticks and accompanying plasma in 2023-2024 at locations in 20 provinces representing the 7 distinct geographical regions in Anatolia. The hosts comprised cattle, sheep, dogs, goats, and tortoises. The ticks were morphologically identified, processed in pools, and these pools, along with plasma from cattle, sheep and goats, were subjected to nucleic acid purification and complementary DNA synthesis. Viruses were screened by generic (nairovirus) and specific (Jingmen tick virus, JMTV; Tacheng tick virus 1, TcTV-1; Tacheng tick virus 2, TcTV-2; and Tamdy virus, TAMV) amplification assays and characterized by sequencing.Results A total of 93 animal plasma samples and 1265 samples from 11 tick species were screened in 192 pools. Crimean-Congo hemorrhagic fever virus (CCHFV) was detected in five tick species in ten pools (5.2%). Three distinct virus lineages, including Europe 1 and 2, as well as Africa 1, were noted. TcTV-1 was identified in 6 tick species in 12 pools (6.3%) and in a cattle plasma sample. Analysis of the nucleoprotein-encoding sequences revealed two separate virus clades, distinct from those reported from Asia and Europe. TAMV was identified in two tick species (1%). We further detected JMTV in 7 pools (3.6%), with sequences forming a new clade phylogenetically closer to viruses of Asian origin than local strains. Finally, highly divergent sequences of a novel nairovirus, forming a distinct group sharing ancestors with TcTV-1, TAMV, and pangolin/tick-associated nairoviruses, was observed in four pools (2%), comprising Haemaphysalis parva ticks.
ConclusionsWe described a previously undocumented diversity of tick-borne viral pathogens, CCHFV, TcTV-1, and JMTV, in Anatolia. Possible animal reservoirs of TcTV-1 were identified. These pathogens and TAMV should be considered in the diagnostic workup of cases with symptoms associated with tick bites and in future surveillance efforts.
## Background
Ticks (order Ixodida) are obligate blood-feeding ectoparasites of mammals, birds, and reptiles, acting as primary biological vectors of viral pathogens affecting livestock, companion animals, and wildlife [1]. The prevalence and spectrum of tick-borne viruses has been on the rise for the several decades and has become a global public health threat, in addition to causing economic losses due to affected livestock and wildlife morbidity. Diseases due to tick-borne viruses are further impacted by climate, globalization, population movements, and modifications of landscapes and natural habitats [2]. Moreover, the capacity of ticks to transmit pathogens across multiple life stages generates complex epidemiological networks and opportunities for spillover to susceptible hosts and reservoirs [3].
A diverse group of viruses circulating between ticks and vertebrate hosts have been recognized as causative agents of human or animal infections. They are taxonomically classified into a single DNA (Asfarviridae) and several RNA virus families [1]. The majority of tick-borne viral pathogens with significant public health impact possess RNA genomes and are classified in the families Nairoviridae and Flaviviridae [4]. Members of the Nairoviridae family, such as Crimean-Congo hemorrhagic fever virus (CCHFV), typically have tri-segmented negative-sense RNA genomes, where open reading frames (ORFs) on each segment encode a nucleoprotein (segment S), a glycoprotein precursor (segment M), and an RNA-directed RNA polymerase (RdRP) (segment L) [5]. Flaviviruses, exemplified by tick-borne encephalitis virus (TBEV), possess a positive-sense RNA genome that contains a single long ORF, which is translated into a polyprotein that requires protease cleavage to generate mature viral proteins [6]. In addition to the observed expansion and increased incidence of these major tickborne viral agents of public concern, several newlydescribed viruses have been documented as causing mild-to-severe human infections following tick-borne transmission [1]. One such group includes Jingmen tick virus (JMTV) and Alongshan virus, which comprise a multicomponent genome composed of four segments, two of which encode viral proteins genetically and functionally related to flaviviruses [7]. Other recently described tick-borne pathogenic nairoviruses, such as Tacheng tick virus 1 (TcTV-1) and phenuiviruses such as Tacheng tick virus 2 (TcTV-2), have further been documented as expanding from their regions of initial detection [8]. Although severe symptoms of central nervous system or hemorrhagic disease might occur in affected humans or animals, many tick-borne viral infections are mild and only manifest as undifferentiated febrile disease, which may pass unnoticed or misdiagnosed [2]. Therefore, screening is crucial to reveal circulating and newly introduced pathogens and implement accurate diagnostics and mitigation strategies.
Located between Asia, Europe, and the Middle East, Türkiye occupies the Anatolian lands of Asia Minor and Eastern Thrace region of the Balkan Peninsula. With regions of diverse ecological features, Anatolia provides suitable habitats for many tick species and maintains a natural transmission zone for vector-borne infections between Asia, Africa, and Europe [9]. The primary tickborne virus circulating in Anatolia is CCHFV, which emerged in 2002 with symptomatic cases annually reported and prior evidence for human exposure [10]. Moreover, other viruses including JMTV have recently been identified in ticks from various parts of Anatolia, with unexplored human or animal health impact [10]. In this study, we sought to investigate tick-borne viruses using a One Health approach encompassing hostremoved ticks and host samples, utilizing a broad range of amplification assays and sequencing to characterize viral genomes.
## Methods
## Sample collection
The tick specimens were collected during May-October in 2023-2024 at locations in 20 provinces representing the 7 distinct geographical regions, comprising Istanbul (Marmara); Mugla (Aegean); Adana, Antalya, Kahramanmaras (Mediterranean); Kayseri, Kirsehir, Nevsehir, Yozgat (Central Anatolia); Ordu, Samsun, Sinop, Tokat (Black Sea); Agri, Ardahan, Erzurum, Igdir, Kars (Eastern Anatolia); and Sanliurfa and Sirnak (Southeastern Anatolia) (Fig. 1). Adult ticks attached to cattle (Bos taurus), sheep (Ovis aries), dogs (Canis familiaris), goats (Capra aegagrus hircus), and tortoises (Testudo graeca) were removed, kept in separate vials, transferred to the laboratory in dry ice, and identified to species morphologically, using available taxonomic keys [11,12]. The specimens were pooled according to collection site, sex, host, and species, up to a maximum of 18 individuals, and subsequently were ground by vortexing with tungsten carbide or steel beads (QIAgen, Hilden, Germany), in 200-500 µL of Dulbecco's phosphate-buffered saline, supplemented with 1% L-glutamine and 5% fetal bovine serum. Following centrifugation at 3000 rpm at 4°C for 6 min, the supernatant from each pool was collected and stored at -80°C. The plasma samples were collected in Antalya (Mediterranean) and Sanliurfa (Southeastern Anatolia) from cattle, sheep, and goats during routine veterinary examinations, and 200 µL of the supernatant or the plasma sample were used for nucleic acid purification using High Pure Viral Nucleic Acid Kit (Roche Diagnostics, Germany), with subsequent complementary DNA synthesis with random hexamers with RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Henningsdorf, Germany), as recommended by the manufacturers. Sample type and tick species distributions according to the region of collection are provided in Table S1.
## Generic amplification of nairoviruses
A singleplex polymerase chain reaction (PCR) amplification assay, targeting viral polymerase central motif A on the nairovirus L genomic segment, was used for screening [13]. Additional information on screening assays is available in Table S2. Amplification mix consisted of 2 mM MgCl 2 , 0.3 mM each dNTP, 0.5 µM forward and reverse primers, 2.5 units of Invitrogen Platinum TaqD-NAPolymerase (Thermo Fisher Scientific), and 3 µl of cDNA template in a 30 µl mix. Thermal cycling consisted of an initial denaturation step at 94°C (5 min), followed by 40 cycles of denaturation at 94°C (60 s), gradual increase of 0.2°C/cycle from 42°C for primer binding (60 s) and elongation at 72°C (60 s), with a final elongation at 72°C (3 min). Assay optimization was carried out using CCHFV strain Ank-2 (GenBank accession: MK309333) propagated in SW-13 cells.
## Targeted amplification of JMTV, TcTV1, TcTV2, and TAMV
Four species-specific PCRs were further performed to detect viruses in the samples. We set up nested reactions to amplify NS5-like protein/replicase on segment 1 (JMTV) and nucleoprotein on the S segment (TcTV-1 and TcTV-2), utilizing previously published primers sets [14][15][16] (Table S2). The reaction mix described for generic nairovirus amplification was used in each reaction. Thermal cycling for the first-round amplifications comprised an initial denaturation at 94°C (4 min), followed by 40 cycles at 94°C (45 s), annealing at 54°C (JMTV), 53°C (TcTVs) (60 s), and 72°C (60 s), with a final extension at 72°C (10 min). Second-round PCRs followed the same conditions except for primer annealing (53°C for JMTV and 52°C for TcTVs). For specific detection of Tamdy virus (TAMV), we used a singleplex PCR to amplify a different section of the RNA-dependent RNA polymerase coding region on the L segment [17] (Table S2), with an identical reaction mix as the generic nairovirus PCR and an annealing temperature of 60°C. TAMV isolate TT1 (NC078320-2), JMTV isolates T14-T15 (MN486261-2), and previously identified TcTV-1 and TcTV-2 positive tick samples were used for assay optimizations [18][19][20]. Nucleic acid extractions and preparation of amplification mixes were carried out in biosafety class II class cabinets. All standard precautions to prevent carry-over contamination were followed, with extraction, pre-, and post-PCR steps being strictly performed in spatially separated areas. All reaction batches were performed using several non-template controls.
## Amplicon visualization, sequencing, and phylogenetic analysis
Amplified products were visualized in a ChemiDoc XRS + imaging system (Bio-Rad Laboratories, Munich, Germany), following electrophoresis in 1-1.7% agarose gels and stained with SYBR Safe DNA gel stain (Thermo Fisher Scientific, Germany). PCR products of expected size were purified using a commercial kit (GeneJET; ThermoFisher Scientific) according to the manufacturer's instructions and sequenced in an ABI PRISM 3500xL Dx Genetic Analyzer (Thermo Fisher Scientific). Sequences were handled using Geneious Prime (v2025.0.3) (Biomatters Ltd., Auckland, New Zealand). Minimap2 and CLUSTALW were used for contig mapping, alignment, and pairwise comparisons [21,22]. BLASTN algorithm was used for similarity searches in the National Center for Biotechnology Information (NCBI) database [23]. Phylogenetic relationships between virus contigs and near relatives were explored using maximum likelihood analysis performed in MEGA v12.0.10 [24]. The optimal model for the phylogenetic and molecular evolutionary analyses in MEGA was determined using the built-in "Find Best DNA/protein-substitution model" tools.
## Results
A total of 1358 samples, comprising 1265 ticks in 192 pools and 93 animal plasma were screened. The ticks were identified as 11 species, including Haemaphysalis parva (21.1%), Rhipicephalus turanicus (20.2%), Hyalomma aegyptium (11.3%), and others (Table S3). Viruses were detected in 36 tick pools (18.7%) and a single plasma sample (1/93, 1.1%). Known tick-borne pathogens were identified in 33 samples (17.2%), comprising 32 ticks pools and a cattle plasma. No co-detection of viral targets or TcTV-2 was observed in any sample. Virus detection was not noted in pools with Hyalomma anatolicum, Ixodes ricinus, or Rhipicephalus sanguineus sensu lato ticks. Information on the samples with virus detection is provided in Table S4.
## Crimean-Congo hemorrhagic fever virus (CCHFV) diversity
We detected CCHFV sequences produced by generic nairovirus amplification in ten pools with Hyalomma marginatum (n = 5), Haemaphysalis parva (n = 2), Hae. punctata (n = 1), Dermacentor marginatus (n = 1), and Rh. turanicus (n = 1) samples (Table S4). Positive pools originated in Central Anatolian provinces (Kayseri, Kirsehir, Nevsehir, n = 4), the Aegean province of Mugla (n = 3), the Black Sea province of Samsun (n = 2), and the Eastern Anatolian province of Kars (n = 1). Detection prevalence was highest in H. marginatum ticks, with 41.6% (5/12) of the pools observed as positive. Alignment and pairwise comparisons revealed a maximum sequence divergence of 2% (Table S5). In the maximum likelihood tree, we observed that the CCHFV sequences were dispersed among three distinct virus clusters (Fig. 2). While all Central Anatolian sequences and one Black Sea sequence were grouped within the Europe 1 cluster, all sequences from Hy. marginatum pools collected in Mugla were placed within Africa 1. Moreover, two sequences from Samsun and Kars provinces were observed as closely related with the former CCHFV genotype Europe 2, currently classified as a separate species (Aigai virus, Orthonairovirus parahaemorrhagiae) within Nairoviridae [25]. Interestingly, two tick species from Samsun province were noted to harbor distinct genotypes of CCHFV, indicating local co-circulation of diverse virus genomes.
## Tacheng tick virus 1 (TcTV-1) clades
The most prevalent pathogen observed in the cohort was TcTV-1, detected in 12 tick pools (6.3%) and a cattle plasma (1.1%). Tick species observed to harbor TcTV-1 were Rhipicephalus bursa (n = 5), Hy. aegyptium (n = 3), D. marginatus (n = 1), Hae. parva (n = 1), Hy. excavatum (n = 1), and Hy. marginatum (n = 1). A total of 11 samples including the cattle plasma originating from Sanliurfa province were detected by virus specific amplification, whereas the generic nairovirus amplification yielded two positive samples with one Hy. excavatum pool (Antalya146) being reactive in both assays. Pairwise sequence comparisons revealed maximum divergences of 7.9% and 0.3% on the nucleocapsid (Table S5) and polymerase encoding amplicons, respectively. Maximum likelihood analysis based on the partial nucleocapsid sequences showed clustering of global virus genomes into four clades, three originating from Asia (Asia I-III) and one from Europe (I) (Fig. 3). Here, TcTV-1 sequences reported from China and Poland comprise two separate clades, while sequences previously identified from Anatolia including those generated in this study were observed to form two distinct clades, regardless of the infected tick species and location. We also observed a comparable pattern on the trees on the basis of the partial polymerase sequences (Fig. 2), despite having only a single clade from Anatolia being represented. TcTV-1-infected ticks were removed from all screened host species in the study (Table 1).
## Tamdy virus (TAMV) phylogeny
Two tick pools from Kayseri province (Central Anatolia) with Rh. turanicus and Hyalomma excavatum samples were positive for TAMV by specific and generic nairovirus amplification, respectively (Table S4). The maximum likelihood tree of the partial viral replicase amplified by generic nairovirus assay placed the Rh. turanicus pool within the TAMV among nairoviruses (Fig. 4). We built a separate tree using the sequence generated by specific amplification in the Hy. excavatum pool, which showed grouping with the previously reported TAMV Anatolian virus strain (TT1), among viruses from China and Russia (Fig. S1). Sequence divergence of 1.9% from this strain was noted in pairwise comparisons with both amplicons.
## Jingmen tick virus (JMTV) clustering
We generated JMTV amplicons in six Hae. parva pools from Eastern Anatolia (Agri and Kars provinces) and one D. marginatus pool from the Black Sea province of Samsun, with a total prevalence of 3.6%. Sequences in Hae. parva pools were identical, with a low overall divergence (0.4%) in all positive samples (Table S5). In the maximum likelihood analysis, the representative sequences grouped together and remained within the cluster comprising Asian JMTVs and closely related viruses (Fig. 5). Interestingly, complete or partial JMTV genomes previously documented in other regions of Asia Minor formed two distinct clusters, distant to those identified in this study, suggesting multiple JMTV genotypes present in various regions.
## Description of a novel nairovirus
We observed amplification by the generic nairovirus assay in four Hae. parva pools from Eastern Anatolia (Igdir and Kars provinces). Sequences generated in these pools showed 98.6-99.3% identities, producing no hits in MEGABLAST searches and outputs with 75% ≥ similarity to various nairoviruses in BLASTN. In the maximum likelihood tree, they formed a well-supported separate cluster, sharing ancestors with TcTV-1 clades within the TAMV genogroup [26] (Fig. 4). Pairwise comparisons with the TcTV-1 and TAMV sequences generated in this study displayed up to 31.6% divergence. We propose these sequences to represent a novel nairovirus belonging in the TAMV genogroup, distantly related to TcTV-1.
## Discussion
The purpose of this cross-sectional study was to investigate tick-borne viral pathogens and provide an update on recently documented viruses from across Anatolia, encompassing 19 provinces from 7 distinct geographical regions. Animal plasma and host-removed ticks, 1358 in total, were collected and screened using generic and specific amplification assays, followed by sequencing. We identified four viral pathogens comprising CCHFV, JMTV, TcTV-1, and TAMV, and a tentative novel nairovirus in 18.7% of the pooled tick and 1.1% of the plasma samples.
We detected CCHFV sequences in 5.2% of the pools spanning five tick genera, including the well-known vector Hy. marginatum. CCHFV is the endemic tick-borne virus and the major tick-associated public health threat in Anatolia, with more than 11,000 documented human infections with an average mortality of 4.8% between 2002 and 2018, following the initial emergence of clinical cases [10,27]. Infections due to CCHFV are widespread in areas of Africa, Eastern Europe, the Middle East, and Asia, mirroring the distribution of the principal Hyalomma spp. tick vector [28]. Moreover, CCHFV has recently become a public health concern in Europe, with more than ten countries reporting virus in ticks, vertebrate exposure, or clinical cases, and recent emergence noted in Bulgaria, Greece, Spain, and Portugal [28]. Although widely distributed across Anatolia, the majority of infections have been reported from central and eastern plateaus, especially around the Kelkit Valley [10,27]. Likewise, the majority of the CCHFV-positive pools in this study originated from central-eastern Anatolian provinces. Interestingly, we observed three distinct CCHFV lineages in the maximum likelihood analyses, including previously reported Europe 1 and 2, as well as Africa 1 [10]. CCHFV isolates have been observed to exhibit significant sequence diversity and are phylogenetically clustered into lineages or clades, mostly overlapping with continental virus distribution [29]. In Anatolia, Europe 1 has been the most frequently documented lineage in ticks and symptomatic humans [10,30]. Moreover, Europe 2 lineage, which has been reclassified as a separate species [25], has been identified in several tick species from Thrace and Anatolia, at times co-circulating with Europe 1 lineage viruses [18,19,31]. However, there are no records of CCHFV's African lineage documented in humans or ticks in Anatolia. Interestingly, all positive pools comprised Hy. marginatum ticks and Table 1 Virus detection rates according to tick pools and host species were collected at the Aegean province of Mugla, where symptomatic cases have been reported. These findings indicate that CCHFV genome diversity in Anatolia is expanding, presumably due to ongoing introductions of divergent virus clades, and potentially as a result of vector carryover by bird migration over the African-Western Palearctic flyway, as previously suggested for Spain [32,33]. Currently, the impact of newly introduced virus genotypes on virulence or local disease epidemiology is hard to assess. However, it is likely to contribute further to the existing virus diversity via recombination [30], for which in silico evidence was previously documented [18]. Hence, the expansion of new and emerging CCHFV lineages in Anatolia require close monitorization.
In the study, TcTV-1 was observed as the most frequently identified tick-borne pathogen, identified in 6 of the 11 species of tick species, with an overall prevalence of 6.3%. Virus genomes were further detected in a cattle plasma from Sanliurfa province (southeastern Anatolia). TcTV-1 is among the newly emerging tick-borne nairoviruses, producing tick-bite-associated febrile disease and skin rash in affected individuals, reported only from China thus far [14]. It was also documented as a co-infecting agent with Rickettsia in a case with febrile disease and meningitis, suggesting probable central nervous system (CNS) involvement during infection [34]. In symptomatic cases, virus genomes are present in throat swabs and urine, indicating possible transmission by direct contact with body fluids, as observed in CCHFV. Currently, TcTV-1 has only been reported to circulate in China, Poland, and Türkiye [8,14,20,35]. The main vector of TcTV-1 was initially considered as Dermacentor spp.; however, as more information accumulated from various regions, a wider range of susceptible ticks, including several Hyalomma spp., were noted [8,14,20,35]. In this study, we further documented TcTV-1 in Rh. bursa and Hae. parva, adding these tick species to the list of potential vectors. Particular domestic and wild animals were considered as susceptible, with exposure documented in sheep, cattle, and great gerbils (Rhombomys opimus) in China [36]. Here, we report virus sequences in cattle plasma, confirming exposure in this species and identifying cattle as a candidate virus reservoir for zoonotic infections. Our detection of infected ticks removed from sheep, goat, and dogs implicate a broader list of domestic animals possibly exposed and potentially participating in circulation. These preliminary findings require verification in larger cohorts to accurately describe TcTV-1 animal reservoirs. Of particular note is the marked TcTV-1 sequence diversity identified in the study. In the maximum likelihood analysis, we observed distinct virus clades corresponding to geographical location (Europe and Asia), as well as within Anatolia, regardless of tick species. These findings suggest local adaptation in virus genomes and possible multiple introductions, likely to impact transmission and virulence. Further investigations on complete TcTV-1 genomes are likely to provide better insights on sequence heterogeneity and impact on vertebrate infections.
Another nairovirus identified in the study is TAMV, with two pools of Rh. turanicus and Hy. excavatum samples with detectable virus genomes, clustering with TAMV Anatolian strain in the the maximum likelihood analyses. Initially isolated from Hy. asiaticum ticks parasitizing sheep in the Tamdinsky district of Uzbekistan in 1971 [37], TAMV was subsequently reported in Uzbekistan, Turkmenistan, Kyrgyzstan, Kazakhstan, Armenia, Azerbaijan, and recently from China [38,39]. TAMV infections in humans and animals remain understudied, where the virus has historically been associated with sporadic cases of febrile diseases in Kyrgyzstan [38], and human exposure in China and Pakistan [40,41]. Virus isolation from Bactrian camels and evidence for spillover from ticks to sheep, dogs, and rodents support possible animal reservoirs in nature [39][40][41][42]. Recently, various mouse infection models were established to facilitate investigations in pathogenicity [43]. In Anatolia, we initially reported TAMV in a Hyalomma spp. pool removed from Meriones tristrami (the rodent Tristram's jird), and subsequently generated the prototype virus complete genome from a Hy. aegyptium pool, both collected from Central Anatolia [18,44]. This region appears as a hotbed zone for TAMV circulation, and Rhipicephalus ticks could further be involved in maintenance. Virus replication across different tick species may contribute to previously identified recombinations in local TAMV genomes [18]. Deeper screening in tick populations and investigation of TAMV in cases with febrile disease will help to elucidate TAMV activity in the region and understand risks for further spillover in Asia Minor.
We further detected JMTV with an overall prevalence of 3.6%, mainly from the Eastern Anatolia and Black Sea regions. Although the partial sequences generated in the study lacked considerable diversity, they grouped as a separate clade within the JMTV and JMTV-like virus genomes reported from Asia, distant from JMTVs previously documented in various parts of Anatolia and Thrace [19,20,45]. They were detected in Rhipicephalus, Haemaphysalis, and Hyalomma spp. ticks, as well as in bat-collected Ixodes simplex. Complete and partial JMTV genomes were observed to form two distinct but related clades, further related with viruses reported from Balkans, with preliminary evidence of recombination [19,20,45]. Here, we demonstrate that the JMTV diversity is even more pronounced in Anatolia, with the reporting of this additional virus clade closer to viruses of Asian origin than local strains. JMTV is a globally distributed virus, documented from continental Asia, Africa, Europe Oceania, and North and South America in arthropods, reptiles, and mammals, including cattle, sheep, goats, and horses, in addition to cases with febrile disease [7]. Considerable genome diversity and potential for genomic exchange was noted, with several proposed virus lineages [46]. Interestingly, a recent analysis of JMTV phylogeography revealed a latitude-dependent evolutionary pattern and described three major virus lineages correlated with latitude, frequent intralineage recombination, and global migration events [47]. This might partially explain our findings, where the JMTV positive samples mainly originated from Eastern Anatolian provinces with considerably higher latitudes compared with Mediterranean or Aegean sites. Nevertheless, near-complete genome information will help to understand the extent of JMTV diversity in Anatolia. As JMTV was detected from CCHFV-infected individuals and a proposed impact on disease outcome [48], it is likely to identify similar cases in a CCHFV-endemic region such as Anatolia, for which efforts are ongoing by our group.
Finally, we generated a partial RNA-dependent RNA polymerase sequence of a cryptic nairovirus, highly divergent from its relatives in pairwise comparisons and forming a distinct group, sharing ancestors with TcTV-1, TAMV, and pangolin/tick-associated nairoviruses. The cryptic sequence was consistently present in Hae. parva pools from Eastern Anatolia and tentatively represents a novel nairovirus, distantly related to TcTV-1. We have previously characterized another novel nairovirus, named Meram virus, in Hy. aegyptium ticks from Central Anatolia [18]. Nevertheless, the current virus is only distantly-related to Meram virus among nairoviruses, and near-complete genome sequences are required for better characterization. Identification of replicative forms and RNA-based sequences of genomic segments would provide further evidence toward a replicating virus and exclude endogenous virus elements in tick genomes. Studies are underway for isolation and transcriptome sequencing.
Overall, our screening approach involving host-collected ticks and plasma from potential hosts can be considered a limitation to assessing temporal and spatial patterns virus circulation in locations with detection. Nevertheless, it produced usable information relevant for public health and may be utilized to direct further screening on selected target pathogens in larger cohort sizes. Follow-up studies with human samples and serological testing will enhance the One-Health-based investigations toward these viruses.
## Conclusions
We described a previously undocumented diversity of tick-borne viral pathogens, CCHFV, TcTV-1, and JMTV, in Anatolia. Evidence for probable animal reservoirs of TcTV-1 were further revealed, along with preliminary reporting of a novel nairovirus. These pathogens and TAMV should be considered in the diagnostic workup of cases with symptoms associated with tick bites and in future surveillance efforts.
## References
1. Mansfield, Jizhou, Phipps et al. (2017) "Emerging tick-borne viruses in the twenty-first century" *Front Cell Infect Microbiol*
2. Kazimírová, Thangamani, Bartíková et al. (2017) "Tick-borne viruses and biological processes at the tickhost-virus interface" *Front Cell Infect Microbiol*
3. Petit, Johnson, Mansfield (2024) "Vectorial dynamics underpinning current and future tick-borne virus emergence in Europe" *J Gen Virol*
4. Hubálek (2012) "Tick-borne viruses in Europe" *Parasitol Res*
5. Kuhn, Alkhovsky, Avšič-Županc et al. (2024) "ICTV virus taxonomy profile: nairoviridae 2024" *J Gen Virol*
6. Simmonds, Becher, Bukh et al. (2017) "ICTV virus taxonomy profile: Flaviviridae" *J Gen Virol*
7. Ogola, Roy, Wollenberg et al. (2025) "Strange relatives: the enigmatic arbojingmenviruses and orthoflaviviruses" *NPJ Viruses*
8. Ergunay, Bourke, Reinbold-Wasson et al. (2023) "The expanding range of emerging tickborne viruses in Eastern Europe and the Black Sea region" *Sci Rep*
9. Bursali, Keskin, Tekin (2012) "A review of the ticks (Acari: Ixodida) of Turkey: species diversity, hosts and geographical distribution" *Exp Appl Acarol*
10. Ergünay, Polat, Özkul (2020) "Vector-borne viruses in Turkey: a systematic review and bibliography" *Antivir Res*
11. Estrada-Pena, Mihalca, Petney (2017) "Ticks of Europe and North Africa: a guide to species identification"
12. Walker, Keirans, Horak (2020) "The Genus Rhipicephalus (Acari, Ixodidae): A Guide to the Brown Ticks of the World"
13. Honig, Osborne, St (2004) "The high genetic variation of viruses of the genus Nairovirus reflects the diversity of their predominant tick hosts" *Virology*
14. Liu, Zhang, Wang et al. (2020) "A tentative tamdy orthonairovirus related to febrile illness in northwestern China" *Clin Infect Dis*
15. Yu, Chen, Qin et al. (2020) "Identification and characterization of Jingmen tick virus in rodents from Xinjiang" *China Infect Genet Evol*
16. Dong, Yang, Wang et al. (2019) "Human Tacheng tick virus 2 infection" *Emerg Infect Dis*
17. Moming, Shen, Fang et al. (2021) "Evidence of human exposure to Tamdy virus, northwest China" *Emerg Infect Dis*
18. Ergünay, Dinçer, Kar et al. (2020) "Multiple orthonairoviruses including Crimean-Congo hemorrhagic fever virus, Tamdy virus and the novel Meram virus in Anatolia" *Ticks Tick Borne Dis*
19. Dinçer, Timurkan, Oğuz et al. (2022) "Several tick-borne pathogenic viruses in circulation in Anatolia" *Turk J Vet Anim Sci*
20. Dincer, Timurkan, Yalcınkaya et al. (2023) "Molecular detection of Tacheng Tick Virus-1 (TcTV-1) and Jingmen Tick Virus in ticks collected from wildlife and livestock in Turkey: first indication of TcTV-1 beyond China" *Vector Borne Zoonotic Dis*
21. Li (2018) "Minimap2: pairwise alignment for nucleotide sequences" *Bioinformatics*
22. Thompson, Higgins, Gibson (1994) "Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice" *Nucleic Acids Res*
23. Altschul, Gish, Miller et al. (1990) "Basic local alignment search tool" *J Mol Biol*
24. Kumar, Stecher, Suleski et al. (2024) "MEGA12: Molecular evolutionary genetic analysis version 12 for adaptive and green computing" *Mol Biol Evol*
25. Papa, Marklewitz, Paraskevopoulou et al. (2022) "History and classification of Aigai virus (formerly Crimean-Congo haemorrhagic fever virus genotype VI)" *J Gen Virol*
26. Walker, Widen, Wood et al. (2016) "A global genomic characterization of Nairoviruses identifies nine discrete genogroups with distinctive structural characteristics and host-vector associations" *Am J Trop Med Hyg*
27. Akyildiz, Bente, Keles et al. (2021) "High prevalence and different genotypes of Crimean-Congo hemorrhagic fever virus genome in questing unfed adult Hyalomma marginatum in Thrace" *Turkey. Ticks Tick Borne Dis*
28. Norman, Arce, Díaz-Menéndez et al. (2025) "Changes in the epidemiology of Crimean-Congo hemorrhagic fever: impact of travel and a One Health approach in the European region" *Travel Med Infect Dis*
29. Lukashev, Klimentov, Smirnova et al. (2016) "Phylogeography of Crimean Congo hemorrhagic fever virus" *PLoS ONE*
30. Ozkaya, Dincer, Carhan et al. (2010) "Molecular epidemiology of Crimean-Congo hemorrhagic fever virus in Turkey: occurrence of local topotype" *Virus Res*
31. Ergunay, Dincer, Justi et al. (2023) "Impact of nanopore-based metagenome sequencing on tick-borne virus detection" *Front Microbiol*
32. Portillo, Palomar, Santibáñez et al. (2021) "Epidemiological aspects of Crimean-Congo hemorrhagic fever in Western Europe: what about the future? Microorganisms"
33. Bacak, Ozsemir, Akyildiz et al. (2023) "Bidirectional tick transport by migratory birds of the African-Western Palearctic flyway over Turkish Thrace: observation of the current situation and future projection" *Parasitol Res*
34. Zhang, Jiang, Yang et al. (2021) "A tick bite patient with fever and meningitis co-infected with Rickettsia raoultii and Tacheng tick virus 1: a case report" *BMC Infect Dis*
35. Ergunay, Bourke, Reinbold-Wasson et al. (2024) "Novel clades of tick-borne pathogenic nairoviruses in Europe" *Infect Genet Evol*
36. Ji, Wang, Liu et al. (2023) "Tacheng tick virus 1 and Songling virus infection in great gerbils (Rhombomys opimus) in northwestern China" *J Wildl Dis*
37. Lvov, Sidorova, Gromashevsky et al. (1976) "from ticks Hyalomma asiaticum asiaticum Schulee et Schlottke, 1929, and Hyalomma plumbeum plumbeum Panzer 1796" *Arch Virol*
38. Lvov, Shchelkanov, Alkhovsky et al. (2015) "Zoonotic viruses of northern Eurasia: taxonomy and ecology"
39. Zhou, Ma, Hu et al. (2018) "Tamdy virus in ixodid ticks infesting bactrian camels" *Emerg Infect Dis*
40. Chen, Saqib, Khan et al. (2024) "Risk of infection with arboviruses in a healthy population in Pakistan based on seroprevalence" *Virol Sin*
41. Moming, Bai, Chen et al. (2024) "Epidemiological surveys revealed the risk of TAMV spill-over from ticks to hosts" *Infect Dis*
42. Cui, Bi, Guo et al. (2024) "Serological evidence of Bactrian camel infection with Tamdy virus" *Xinjiang. China Vector Borne Zoonotic Dis*
43. Cui, Zhu, Wang et al. (2025) "Tamdy virus pathogenesis in immunocompetent and immunocompromised mouse models" *Virulence*
44. Brinkmann, Dinçer, Polat et al. (2018) "A metagenomic survey identifies Tamdy orthonairovirus as well as divergent phlebo-, rhabdo-, chu-and flavi-like viruses in Anatolia" *Turkey. Ticks Tick-borne Dis*
45. Dinçer, Hacıoğlu, Kar et al. (2019) "Survey and characterization of Jingmen tick virus variants" *Viruses*
46. Li, Li, Tang et al. (2023) "Genomics evolution of Jingmen viruses associated with ticks and vertebrates" *Genomics*
47. Zhang, Wu, Wang et al. (2025) "Latitude-driven patterns and dynamics in Jingmen group viral lineages: spatial correlation, recombination, and phylogeography" *Infect Genet Evol*
48. Emmerich, Jakupi, Possel et al. (2018) "Viral metagenomics, genetic and evolutionary characteristics of Crimean-Congo hemorrhagic fever orthonairovirus in humans" *Kosovo Infect Genet Evol* |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.